diff options
| author |   Linus Torvalds <torvalds@ppc970.osdl.org> | 2005-04-16 15:20:36 -0700 |
|---|---|---|
| committer | Linus Torvalds <torvalds@ppc970.osdl.org> | 2005-04-16 15:20:36 -0700 |
| commit | 1da177e4c3f41524e886b7f1b8a0c1fc7321cac2 (patch) | |
| tree | 0bba044c4ce775e45a88a51686b5d9f90697ea9d /arch/um | |
| download | linux-dev-1da177e4c3f41524e886b7f1b8a0c1fc7321cac2.tar.xz linux-dev-1da177e4c3f41524e886b7f1b8a0c1fc7321cac2.zip | |
Linux-2.6.12-rc2
Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.
Let it rip!
Diffstat (limited to 'arch/um')
295 files changed, 35908 insertions, 0 deletions
diff --git a/arch/um/Kconfig b/arch/um/Kconfig new file mode 100644 index 000000000000..9a23df182123 --- /dev/null +++ b/arch/um/Kconfig @@ -0,0 +1,320 @@ +# UML uses the generic IRQ sugsystem +config GENERIC_HARDIRQS + bool + default y + +config UML + bool + default y + +# XXX: does UM have a mmu/swap? +config MMU + bool + default y + +mainmenu "Linux/Usermode Kernel Configuration" + +config ISA + bool + +config SBUS + bool + +config PCI + bool + +config UID16 + bool + default y + +config RWSEM_GENERIC_SPINLOCK + bool + default y + +config GENERIC_CALIBRATE_DELAY + bool + default y + +menu "UML-specific options" + +config MODE_TT + bool "Tracing thread support" + default y + help + This option controls whether tracing thread support is compiled + into UML. Normally, this should be set to Y. If you intend to + use only skas mode (and the host has the skas patch applied to it), + then it is OK to say N here. + +config STATIC_LINK + bool "Force a static link" + default n + depends on !MODE_TT + help + If CONFIG_MODE_TT is disabled, then this option gives you the ability + to force a static link of UML. Normally, if only skas mode is built + in to UML, it will be linked as a shared binary. This is inconvenient + for use in a chroot jail. So, if you intend to run UML inside a + chroot, and you disable CONFIG_MODE_TT, you probably want to say Y + here. + +config MODE_SKAS + bool "Separate Kernel Address Space support" + default y + help + This option controls whether skas (separate kernel address space) + support is compiled in. If you have applied the skas patch to the + host, then you certainly want to say Y here (and consider saying N + to CONFIG_MODE_TT). Otherwise, it is safe to say Y. Disabling this + option will shrink the UML binary slightly. + +source "arch/um/Kconfig_arch" + +config LD_SCRIPT_STATIC + bool + default y + depends on MODE_TT || STATIC_LINK + +config LD_SCRIPT_DYN + bool + default y + depends on !LD_SCRIPT_STATIC + +config NET + bool "Networking support" + help + Unless you really know what you are doing, you should say Y here. + The reason is that some programs need kernel networking support even + when running on a stand-alone machine that isn't connected to any + other computer. If you are upgrading from an older kernel, you + should consider updating your networking tools too because changes + in the kernel and the tools often go hand in hand. The tools are + contained in the package net-tools, the location and version number + of which are given in <file:Documentation/Changes>. + + For a general introduction to Linux networking, it is highly + recommended to read the NET-HOWTO, available from + <http://www.tldp.org/docs.html#howto>. + + +source "fs/Kconfig.binfmt" + +config HOSTFS + tristate "Host filesystem" + help + While the User-Mode Linux port uses its own root file system for + booting and normal file access, this module lets the UML user + access files stored on the host. It does not require any + network connection between the Host and UML. An example use of + this might be: + + mount none /tmp/fromhost -t hostfs -o /tmp/umlshare + + where /tmp/fromhost is an empty directory inside UML and + /tmp/umlshare is a directory on the host with files the UML user + wishes to access. + + For more information, see + <http://user-mode-linux.sourceforge.net/hostfs.html>. + + If you'd like to be able to work with files stored on the host, + say Y or M here; otherwise say N. + +config HPPFS + tristate "HoneyPot ProcFS (EXPERIMENTAL)" + depends on BROKEN + help + hppfs (HoneyPot ProcFS) is a filesystem which allows UML /proc + entries to be overridden, removed, or fabricated from the host. + Its purpose is to allow a UML to appear to be a physical machine + by removing or changing anything in /proc which gives away the + identity of a UML. + + See <http://user-mode-linux.sf.net/hppfs.html> for more information. + + You only need this if you are setting up a UML honeypot. Otherwise, + it is safe to say 'N' here. + + If you are actively using it, please ask for it to be fixed. In this + moment, it does not work on 2.6 (it works somehow on 2.4). + +config MCONSOLE + bool "Management console" + default y + help + The user mode linux management console is a low-level interface to + the kernel, somewhat like the i386 SysRq interface. Since there is + a full-blown operating system running under every user mode linux + instance, there is much greater flexibility possible than with the + SysRq mechanism. + + If you answer 'Y' to this option, to use this feature, you need the + mconsole client (called uml_mconsole) which is present in CVS in + 2.4.5-9um and later (path /tools/mconsole), and is also in the + distribution RPM package in 2.4.6 and later. + + It is safe to say 'Y' here. + +config MAGIC_SYSRQ + bool "Magic SysRq key" + depends on MCONSOLE + ---help--- + If you say Y here, you will have some control over the system even + if the system crashes for example during kernel debugging (e.g., you + will be able to flush the buffer cache to disk, reboot the system + immediately or dump some status information). A key for each of the + possible requests is provided. + + This is the feature normally accomplished by pressing a key + while holding SysRq (Alt+PrintScreen). + + On UML, this is accomplished by sending a "sysrq" command with + mconsole, followed by the letter for the requested command. + + The keys are documented in <file:Documentation/sysrq.txt>. Don't say Y + unless you really know what this hack does. + +config HOST_2G_2G + bool "2G/2G host address space split" + default n + help + This is needed when the host on which you run has a 2G/2G memory + split, instead of the customary 3G/1G. + + Note that to enable such a host + configuration, which makes sense only in some cases, you need special + host patches. + + So, if you do not know what to do here, say 'N'. + +config SMP + bool "Symmetric multi-processing support (EXPERIMENTAL)" + default n + depends on MODE_TT && EXPERIMENTAL + help + This option enables UML SMP support. + It is NOT related to having a real SMP box. Not directly, at least. + + UML implements virtual SMP by allowing as many processes to run + simultaneously on the host as there are virtual processors configured. + + Obviously, if the host is a uniprocessor, those processes will + timeshare, but, inside UML, will appear to be running simultaneously. + If the host is a multiprocessor, then UML processes may run + simultaneously, depending on the host scheduler. + + This, however, is supported only in TT mode. So, if you use the SKAS + patch on your host, switching to TT mode and enabling SMP usually gives + you worse performances. + Also, since the support for SMP has been under-developed, there could + be some bugs being exposed by enabling SMP. + + If you don't know what to do, say N. + +config NR_CPUS + int "Maximum number of CPUs (2-32)" + range 2 32 + depends on SMP + default "32" + +config NEST_LEVEL + int "Nesting level" + default "0" + help + This is set to the number of layers of UMLs that this UML will be run + in. Normally, this is zero, meaning that it will run directly on the + host. Setting it to one will build a UML that can run inside a UML + that is running on the host. Generally, if you intend this UML to run + inside another UML, set CONFIG_NEST_LEVEL to one more than the host + UML. + + Note that if the hosting UML has its CONFIG_KERNEL_HALF_GIGS set to + greater than one, then the guest UML should have its CONFIG_NEST_LEVEL + set to the host's CONFIG_NEST_LEVEL + CONFIG_KERNEL_HALF_GIGS. + Only change this if you are running nested UMLs. + +config KERNEL_HALF_GIGS + int "Kernel address space size (in .5G units)" + default "1" + help + This determines the amount of address space that UML will allocate for + its own, measured in half Gigabyte units. The default is 1. + Change this only if you need to boot UML with an unusually large amount + of physical memory. + +config HIGHMEM + bool "Highmem support" + +config KERNEL_STACK_ORDER + int "Kernel stack size order" + default 2 + help + This option determines the size of UML kernel stacks. They will + be 1 << order pages. The default is OK unless you're running Valgrind + on UML, in which case, set this to 3. + +config UML_REAL_TIME_CLOCK + bool "Real-time Clock" + default y + help + This option makes UML time deltas match wall clock deltas. This should + normally be enabled. The exception would be if you are debugging with + UML and spend long times with UML stopped at a breakpoint. In this + case, when UML is restarted, it will call the timer enough times to make + up for the time spent at the breakpoint. This could result in a + noticable lag. If this is a problem, then disable this option. + +endmenu + +source "init/Kconfig" + +source "drivers/base/Kconfig" + +source "arch/um/Kconfig_char" + +source "drivers/block/Kconfig" + +config NETDEVICES + bool + default NET + +source "arch/um/Kconfig_net" + +source "net/Kconfig" + +source "fs/Kconfig" + +source "security/Kconfig" + +source "crypto/Kconfig" + +source "lib/Kconfig" + +menu "SCSI support" +depends on BROKEN + +config SCSI + tristate "SCSI support" + +# This gives us free_dma, which scsi.c wants. +config GENERIC_ISA_DMA + bool + depends on SCSI + default y + +source "arch/um/Kconfig_scsi" + +endmenu + +source "drivers/md/Kconfig" + +if BROKEN + source "drivers/mtd/Kconfig" +endif + +#This is just to shut up some Kconfig warnings, so no prompt. +config INPUT + bool + default n + +source "arch/um/Kconfig.debug" diff --git a/arch/um/Kconfig.debug b/arch/um/Kconfig.debug new file mode 100644 index 000000000000..b89989de364d --- /dev/null +++ b/arch/um/Kconfig.debug @@ -0,0 +1,53 @@ +menu "Kernel hacking" + +source "lib/Kconfig.debug" + +config FRAME_POINTER + bool + default y if DEBUG_INFO + +config PT_PROXY + bool "Enable ptrace proxy" + depends on XTERM_CHAN && DEBUG_INFO && MODE_TT + help + This option enables a debugging interface which allows gdb to debug + the kernel without needing to actually attach to kernel threads. + If you want to do kernel debugging, say Y here; otherwise say N. + +config GPROF + bool "Enable gprof support" + depends on DEBUG_INFO && MODE_SKAS && !MODE_TT + help + This allows profiling of a User-Mode Linux kernel with the gprof + utility. + + See <http://user-mode-linux.sourceforge.net/gprof.html> for more + details. + + If you're involved in UML kernel development and want to use gprof, + say Y. If you're unsure, say N. + +config GCOV + bool "Enable gcov support" + depends on DEBUG_INFO && MODE_SKAS + help + This option allows developers to retrieve coverage data from a UML + session. + + See <http://user-mode-linux.sourceforge.net/gprof.html> for more + details. + + If you're involved in UML kernel development and want to use gcov, + say Y. If you're unsure, say N. + +config SYSCALL_DEBUG + bool "Enable system call debugging" + default N + depends on DEBUG_INFO + help + This adds some system debugging to UML, including keeping a ring buffer + with recent system calls and some global and per-task statistics. + + If unsure, say N + +endmenu diff --git a/arch/um/Kconfig_char b/arch/um/Kconfig_char new file mode 100644 index 000000000000..3e50fdb67626 --- /dev/null +++ b/arch/um/Kconfig_char @@ -0,0 +1,208 @@ + +menu "Character Devices" + +config STDERR_CONSOLE + bool "stderr console" + default y + help + console driver which dumps all printk messages to stderr. + +config STDIO_CONSOLE + bool + default y + +config SSL + bool "Virtual serial line" + help + The User-Mode Linux environment allows you to create virtual serial + lines on the UML that are usually made to show up on the host as + ttys or ptys. + + See <http://user-mode-linux.sourceforge.net/input.html> for more + information and command line examples of how to use this facility. + + Unless you have a specific reason for disabling this, say Y. + +config NULL_CHAN + bool "null channel support" + help + This option enables support for attaching UML consoles and serial + lines to a device similar to /dev/null. Data written to it disappears + and there is never any data to be read. + +config PORT_CHAN + bool "port channel support" + help + This option enables support for attaching UML consoles and serial + lines to host portals. They may be accessed with 'telnet <host> + <port number>'. Any number of consoles and serial lines may be + attached to a single portal, although what UML device you get when + you telnet to that portal will be unpredictable. + It is safe to say 'Y' here. + +config PTY_CHAN + bool "pty channel support" + help + This option enables support for attaching UML consoles and serial + lines to host pseudo-terminals. Access to both traditional + pseudo-terminals (/dev/pty*) and pts pseudo-terminals are controlled + with this option. The assignment of UML devices to host devices + will be announced in the kernel message log. + It is safe to say 'Y' here. + +config TTY_CHAN + bool "tty channel support" + help + This option enables support for attaching UML consoles and serial + lines to host terminals. Access to both virtual consoles + (/dev/tty*) and the slave side of pseudo-terminals (/dev/ttyp* and + /dev/pts/*) are controlled by this option. + It is safe to say 'Y' here. + +config XTERM_CHAN + bool "xterm channel support" + help + This option enables support for attaching UML consoles and serial + lines to xterms. Each UML device so assigned will be brought up in + its own xterm. + If you disable this option, then CONFIG_PT_PROXY will be disabled as + well, since UML's gdb currently requires an xterm. + It is safe to say 'Y' here. + +config NOCONFIG_CHAN + bool + default !(XTERM_CHAN && TTY_CHAN && PTY_CHAN && PORT_CHAN && NULL_CHAN) + +config CON_ZERO_CHAN + string "Default main console channel initialization" + default "fd:0,fd:1" + help + This is the string describing the channel to which the main console + will be attached by default. This value can be overridden from the + command line. The default value is "fd:0,fd:1", which attaches the + main console to stdin and stdout. + It is safe to leave this unchanged. + +config CON_CHAN + string "Default console channel initialization" + default "xterm" + help + This is the string describing the channel to which all consoles + except the main console will be attached by default. This value can + be overridden from the command line. The default value is "xterm", + which brings them up in xterms. + It is safe to leave this unchanged, although you may wish to change + this if you expect the UML that you build to be run in environments + which don't have X or xterm available. + +config SSL_CHAN + string "Default serial line channel initialization" + default "pty" + help + This is the string describing the channel to which the serial lines + will be attached by default. This value can be overridden from the + command line. The default value is "pty", which attaches them to + traditional pseudo-terminals. + It is safe to leave this unchanged, although you may wish to change + this if you expect the UML that you build to be run in environments + which don't have a set of /dev/pty* devices. + +config UNIX98_PTYS + bool "Unix98 PTY support" + ---help--- + A pseudo terminal (PTY) is a software device consisting of two + halves: a master and a slave. The slave device behaves identical to + a physical terminal; the master device is used by a process to + read data from and write data to the slave, thereby emulating a + terminal. Typical programs for the master side are telnet servers + and xterms. + + Linux has traditionally used the BSD-like names /dev/ptyxx for + masters and /dev/ttyxx for slaves of pseudo terminals. This scheme + has a number of problems. The GNU C library glibc 2.1 and later, + however, supports the Unix98 naming standard: in order to acquire a + pseudo terminal, a process opens /dev/ptmx; the number of the pseudo + terminal is then made available to the process and the pseudo + terminal slave can be accessed as /dev/pts/<number>. What was + traditionally /dev/ttyp2 will then be /dev/pts/2, for example. + + All modern Linux systems use the Unix98 ptys. Say Y unless + you're on an embedded system and want to conserve memory. + +config LEGACY_PTYS + bool "Legacy (BSD) PTY support" + default y + ---help--- + A pseudo terminal (PTY) is a software device consisting of two + halves: a master and a slave. The slave device behaves identical to + a physical terminal; the master device is used by a process to + read data from and write data to the slave, thereby emulating a + terminal. Typical programs for the master side are telnet servers + and xterms. + + Linux has traditionally used the BSD-like names /dev/ptyxx + for masters and /dev/ttyxx for slaves of pseudo + terminals. This scheme has a number of problems, including + security. This option enables these legacy devices; on most + systems, it is safe to say N. + + +config LEGACY_PTY_COUNT + int "Maximum number of legacy PTY in use" + depends on LEGACY_PTYS + default "256" + ---help--- + The maximum number of legacy PTYs that can be used at any one time. + The default is 256, and should be more than enough. Embedded + systems may want to reduce this to save memory. + + When not in use, each legacy PTY occupies 12 bytes on 32-bit + architectures and 24 bytes on 64-bit architectures. + +config WATCHDOG + bool "Watchdog Timer Support" + +config WATCHDOG_NOWAYOUT + bool "Disable watchdog shutdown on close" + depends on WATCHDOG + +config SOFT_WATCHDOG + tristate "Software Watchdog" + depends on WATCHDOG + +config UML_WATCHDOG + tristate "UML watchdog" + depends on WATCHDOG + +config UML_SOUND + tristate "Sound support" + help + This option enables UML sound support. If enabled, it will pull in + soundcore and the UML hostaudio relay, which acts as a intermediary + between the host's dsp and mixer devices and the UML sound system. + It is safe to say 'Y' here. + +config SOUND + tristate + default UML_SOUND + +config HOSTAUDIO + tristate + default UML_SOUND + +config UML_RANDOM + tristate "Hardware random number generator" + help + This option enables UML's "hardware" random number generator. It + attaches itself to the host's /dev/random, supplying as much entropy + as the host has, rather than the small amount the UML gets from its + own drivers. It registers itself as a standard hardware random number + generator, major 10, minor 183, and the canonical device name is + /dev/hwrng. + The way to make use of this is to install the rng-tools package + (check your distro, or download from + http://sourceforge.net/projects/gkernel/). rngd periodically reads + /dev/hwrng and injects the entropy into /dev/random. + +endmenu + diff --git a/arch/um/Kconfig_i386 b/arch/um/Kconfig_i386 new file mode 100644 index 000000000000..203c242201b6 --- /dev/null +++ b/arch/um/Kconfig_i386 @@ -0,0 +1,24 @@ +config 64_BIT + bool + default n + +config TOP_ADDR + hex + default 0xc0000000 if !HOST_2G_2G + default 0x80000000 if HOST_2G_2G + +config 3_LEVEL_PGTABLES + bool "Three-level pagetables" + default n + help + Three-level pagetables will let UML have more than 4G of physical + memory. All the memory that can't be mapped directly will be treated + as high memory. + +config ARCH_HAS_SC_SIGNALS + bool + default y + +config ARCH_REUSE_HOST_VSYSCALL_AREA + bool + default y diff --git a/arch/um/Kconfig_net b/arch/um/Kconfig_net new file mode 100644 index 000000000000..1c2f9a70d91d --- /dev/null +++ b/arch/um/Kconfig_net @@ -0,0 +1,180 @@ + +menu "UML Network Devices" + depends on NET + +# UML virtual driver +config UML_NET + bool "Virtual network device" + help + While the User-Mode port cannot directly talk to any physical + hardware devices, this choice and the following transport options + provide one or more virtual network devices through which the UML + kernels can talk to each other, the host, and with the host's help, + machines on the outside world. + + For more information, including explanations of the networking and + sample configurations, see + <http://user-mode-linux.sourceforge.net/networking.html>. + + If you'd like to be able to enable networking in the User-Mode + linux environment, say Y; otherwise say N. Note that you must + enable at least one of the following transport options to actually + make use of UML networking. + +config UML_NET_ETHERTAP + bool "Ethertap transport" + depends on UML_NET + help + The Ethertap User-Mode Linux network transport allows a single + running UML to exchange packets with its host over one of the + host's Ethertap devices, such as /dev/tap0. Additional running + UMLs can use additional Ethertap devices, one per running UML. + While the UML believes it's on a (multi-device, broadcast) virtual + Ethernet network, it's in fact communicating over a point-to-point + link with the host. + + To use this, your host kernel must have support for Ethertap + devices. Also, if your host kernel is 2.4.x, it must have + CONFIG_NETLINK_DEV configured as Y or M. + + For more information, see + <http://user-mode-linux.sourceforge.net/networking.html> That site + has examples of the UML command line to use to enable Ethertap + networking. + + If you'd like to set up an IP network with the host and/or the + outside world, say Y to this, the Daemon Transport and/or the + Slip Transport. You'll need at least one of them, but may choose + more than one without conflict. If you don't need UML networking, + say N. + +config UML_NET_TUNTAP + bool "TUN/TAP transport" + depends on UML_NET + help + The UML TUN/TAP network transport allows a UML instance to exchange + packets with the host over a TUN/TAP device. This option will only + work with a 2.4 host, unless you've applied the TUN/TAP patch to + your 2.2 host kernel. + + To use this transport, your host kernel must have support for TUN/TAP + devices, either built-in or as a module. + +config UML_NET_SLIP + bool "SLIP transport" + depends on UML_NET + help + The slip User-Mode Linux network transport allows a running UML to + network with its host over a point-to-point link. Unlike Ethertap, + which can carry any Ethernet frame (and hence even non-IP packets), + the slip transport can only carry IP packets. + + To use this, your host must support slip devices. + + For more information, see + <http://user-mode-linux.sourceforge.net/networking.html>. That site + has examples of the UML command line to use to enable slip + networking, and details of a few quirks with it. + + The Ethertap Transport is preferred over slip because of its + limitations. If you prefer slip, however, say Y here. Otherwise + choose the Multicast transport (to network multiple UMLs on + multiple hosts), Ethertap (to network with the host and the + outside world), and/or the Daemon transport (to network multiple + UMLs on a single host). You may choose more than one without + conflict. If you don't need UML networking, say N. + +config UML_NET_DAEMON + bool "Daemon transport" + depends on UML_NET + help + This User-Mode Linux network transport allows one or more running + UMLs on a single host to communicate with each other, but not to + the host. + + To use this form of networking, you'll need to run the UML + networking daemon on the host. + + For more information, see + <http://user-mode-linux.sourceforge.net/networking.html> That site + has examples of the UML command line to use to enable Daemon + networking. + + If you'd like to set up a network with other UMLs on a single host, + say Y. If you need a network between UMLs on multiple physical + hosts, choose the Multicast Transport. To set up a network with + the host and/or other IP machines, say Y to the Ethertap or Slip + transports. You'll need at least one of them, but may choose + more than one without conflict. If you don't need UML networking, + say N. + +config UML_NET_MCAST + bool "Multicast transport" + depends on UML_NET + help + This Multicast User-Mode Linux network transport allows multiple + UMLs (even ones running on different host machines!) to talk to + each other over a virtual ethernet network. However, it requires + at least one UML with one of the other transports to act as a + bridge if any of them need to be able to talk to their hosts or any + other IP machines. + + To use this, your host kernel(s) must support IP Multicasting. + + For more information, see + <http://user-mode-linux.sourceforge.net/networking.html> That site + has examples of the UML command line to use to enable Multicast + networking, and notes about the security of this approach. + + If you need UMLs on multiple physical hosts to communicate as if + they shared an Ethernet network, say Y. If you need to communicate + with other IP machines, make sure you select one of the other + transports (possibly in addition to Multicast; they're not + exclusive). If you don't need to network UMLs say N to each of + the transports. + +config UML_NET_PCAP + bool "pcap transport" + depends on UML_NET && BROKEN + help + The pcap transport makes a pcap packet stream on the host look + like an ethernet device inside UML. This is useful for making + UML act as a network monitor for the host. You must have libcap + installed in order to build the pcap transport into UML. + + For more information, see + <http://user-mode-linux.sourceforge.net/networking.html> That site + has examples of the UML command line to use to enable this option. + + If you intend to use UML as a network monitor for the host, say + Y here. Otherwise, say N. + +config UML_NET_SLIRP + bool "SLiRP transport" + depends on UML_NET + help + The SLiRP User-Mode Linux network transport allows a running UML + to network by invoking a program that can handle SLIP encapsulated + packets. This is commonly (but not limited to) the application + known as SLiRP, a program that can re-socket IP packets back onto + the host on which it is run. Only IP packets are supported, + unlike other network transports that can handle all Ethernet + frames. In general, slirp allows the UML the same IP connectivity + to the outside world that the host user is permitted, and unlike + other transports, SLiRP works without the need of root level + privleges, setuid binaries, or SLIP devices on the host. This + also means not every type of connection is possible, but most + situations can be accomodated with carefully crafted slirp + commands that can be passed along as part of the network device's + setup string. The effect of this transport on the UML is similar + that of a host behind a firewall that masquerades all network + connections passing through it (but is less secure). + + To use this you should first have slirp compiled somewhere + accessible on the host, and have read its documentation. If you + don't need UML networking, say N. + + Startup example: "eth0=slirp,FE:FD:01:02:03:04,/usr/local/bin/slirp" + +endmenu + diff --git a/arch/um/Kconfig_scsi b/arch/um/Kconfig_scsi new file mode 100644 index 000000000000..c291c942b1a8 --- /dev/null +++ b/arch/um/Kconfig_scsi @@ -0,0 +1,58 @@ +comment "SCSI support type (disk, tape, CD-ROM)" + depends on SCSI + +config BLK_DEV_SD + tristate "SCSI disk support" + depends on SCSI + +config SD_EXTRA_DEVS + int "Maximum number of SCSI disks that can be loaded as modules" + depends on BLK_DEV_SD + default "40" + +config CHR_DEV_ST + tristate "SCSI tape support" + depends on SCSI + +config BLK_DEV_SR + tristate "SCSI CD-ROM support" + depends on SCSI + +config BLK_DEV_SR_VENDOR + bool "Enable vendor-specific extensions (for SCSI CDROM)" + depends on BLK_DEV_SR + +config SR_EXTRA_DEVS + int "Maximum number of CDROM devices that can be loaded as modules" + depends on BLK_DEV_SR + default "2" + +config CHR_DEV_SG + tristate "SCSI generic support" + depends on SCSI + +comment "Some SCSI devices (e.g. CD jukebox) support multiple LUNs" + depends on SCSI + +#if [ "$CONFIG_EXPERIMENTAL" = "y" ]; then +config SCSI_DEBUG_QUEUES + bool "Enable extra checks in new queueing code" + depends on SCSI + +#fi +config SCSI_MULTI_LUN + bool "Probe all LUNs on each SCSI device" + depends on SCSI + +config SCSI_CONSTANTS + bool "Verbose SCSI error reporting (kernel size +=12K)" + depends on SCSI + +config SCSI_LOGGING + bool "SCSI logging facility" + depends on SCSI + +config SCSI_DEBUG + tristate "SCSI debugging host simulator (EXPERIMENTAL)" + depends on SCSI + diff --git a/arch/um/Kconfig_x86_64 b/arch/um/Kconfig_x86_64 new file mode 100644 index 000000000000..768dc6626a8d --- /dev/null +++ b/arch/um/Kconfig_x86_64 @@ -0,0 +1,15 @@ +config 64_BIT + bool + default y + +config 3_LEVEL_PGTABLES + bool + default y + +config ARCH_HAS_SC_SIGNALS + bool + default n + +config ARCH_REUSE_HOST_VSYSCALL_AREA + bool + default n diff --git a/arch/um/Makefile b/arch/um/Makefile new file mode 100644 index 000000000000..97bca6b5ca95 --- /dev/null +++ b/arch/um/Makefile @@ -0,0 +1,210 @@ +# +# Copyright (C) 2002 Jeff Dike (jdike@karaya.com) +# Licensed under the GPL +# + +ARCH_DIR := arch/um +OS := $(shell uname -s) +# We require bash because the vmlinux link and loader script cpp use bash +# features. +SHELL := /bin/bash + +filechk_gen_header = $< + +core-y += $(ARCH_DIR)/kernel/ \ + $(ARCH_DIR)/drivers/ \ + $(ARCH_DIR)/os-$(OS)/ + +# Have to precede the include because the included Makefiles reference them. +SYMLINK_HEADERS := archparam.h system.h sigcontext.h processor.h ptrace.h \ + arch-signal.h module.h vm-flags.h +SYMLINK_HEADERS := $(foreach header,$(SYMLINK_HEADERS),include/asm-um/$(header)) + +# XXX: The "os" symlink is only used by arch/um/include/os.h, which includes +# ../os/include/file.h +# +# These are cleaned up during mrproper. Please DO NOT fix it again, this is +# the Correct Thing(tm) to do! +ARCH_SYMLINKS = include/asm-um/arch $(ARCH_DIR)/include/sysdep $(ARCH_DIR)/os \ + $(SYMLINK_HEADERS) $(ARCH_DIR)/include/uml-config.h + +GEN_HEADERS += $(ARCH_DIR)/include/task.h $(ARCH_DIR)/include/kern_constants.h + +um-modes-$(CONFIG_MODE_TT) += tt +um-modes-$(CONFIG_MODE_SKAS) += skas + +MODE_INCLUDE += $(foreach mode,$(um-modes-y),\ + -I$(srctree)/$(ARCH_DIR)/kernel/$(mode)/include) + +MAKEFILES-INCL += $(foreach mode,$(um-modes-y),\ + $(srctree)/$(ARCH_DIR)/Makefile-$(mode)) + +ifneq ($(MAKEFILES-INCL),) + include $(MAKEFILES-INCL) +endif + +ARCH_INCLUDE := -I$(ARCH_DIR)/include +SYS_DIR := $(ARCH_DIR)/include/sysdep-$(SUBARCH) + +include $(srctree)/$(ARCH_DIR)/Makefile-$(SUBARCH) + +core-y += $(SUBARCH_CORE) +libs-y += $(SUBARCH_LIBS) + +# -Derrno=kernel_errno - This turns all kernel references to errno into +# kernel_errno to separate them from the libc errno. This allows -fno-common +# in CFLAGS. Otherwise, it would cause ld to complain about the two different +# errnos. + +CFLAGS += $(CFLAGS-y) -D__arch_um__ -DSUBARCH=\"$(SUBARCH)\" \ + $(ARCH_INCLUDE) $(MODE_INCLUDE) + +USER_CFLAGS := $(patsubst -I%,,$(CFLAGS)) +USER_CFLAGS := $(patsubst -D__KERNEL__,,$(USER_CFLAGS)) $(ARCH_INCLUDE) \ + $(MODE_INCLUDE) $(ARCH_USER_CFLAGS) +CFLAGS += -Derrno=kernel_errno -Dsigprocmask=kernel_sigprocmask +CFLAGS += $(call cc-option,-fno-unit-at-a-time,) + +#This will adjust *FLAGS accordingly to the platform. +include $(srctree)/$(ARCH_DIR)/Makefile-os-$(OS) + +# These are needed for clean and mrproper, since in that case .config is not +# included; the values here are meaningless + +CONFIG_NEST_LEVEL ?= 0 +CONFIG_KERNEL_HALF_GIGS ?= 0 + +SIZE = (($(CONFIG_NEST_LEVEL) + $(CONFIG_KERNEL_HALF_GIGS)) * 0x20000000) + +ifeq ($(CONFIG_MODE_SKAS), y) +$(SYS_HEADERS) : $(ARCH_DIR)/include/skas_ptregs.h +endif + +.PHONY: linux + +all: linux + +linux: vmlinux + ln -f $< $@ + +define archhelp + echo '* linux - Binary kernel image (./linux) - for backward' + echo ' compatibility only, this creates a hard link to the' + echo ' real kernel binary, the the "vmlinux" binary you' + echo ' find in the kernel root.' +endef + +$(shell cd $(ARCH_DIR) && ln -sf Kconfig_$(SUBARCH) Kconfig_arch) + +prepare: $(ARCH_SYMLINKS) $(SYS_HEADERS) $(GEN_HEADERS) \ + $(ARCH_DIR)/kernel/vmlinux.lds.S + +LINK-$(CONFIG_LD_SCRIPT_STATIC) += -static +LINK-$(CONFIG_LD_SCRIPT_DYN) += -Wl,-rpath,/lib + +LD_SCRIPT-$(CONFIG_LD_SCRIPT_STATIC) := uml.lds.S +LD_SCRIPT-$(CONFIG_LD_SCRIPT_DYN) := dyn.lds.S + +CPP_MODE-$(CONFIG_MODE_TT) := -DMODE_TT +CONFIG_KERNEL_STACK_ORDER ?= 2 +STACK_SIZE := $(shell echo $$[ 4096 * (1 << $(CONFIG_KERNEL_STACK_ORDER)) ] ) + +ifndef START + START = $$(($(TOP_ADDR) - $(SIZE))) +endif + +CPPFLAGS_vmlinux.lds = $(shell echo -U$(SUBARCH) \ + -DSTART=$(START) -DELF_ARCH=$(ELF_ARCH) \ + -DELF_FORMAT=\"$(ELF_FORMAT)\" $(CPP_MODE-y) \ + -DKERNEL_STACK_SIZE=$(STACK_SIZE)) + +#The wrappers will select whether using "malloc" or the kernel allocator. +LINK_WRAPS = -Wl,--wrap,malloc -Wl,--wrap,free -Wl,--wrap,calloc + +CFLAGS_vmlinux = $(LINK-y) $(LINK_WRAPS) +define cmd_vmlinux__ + $(CC) $(CFLAGS_vmlinux) -o $@ \ + -Wl,-T,$(vmlinux-lds) $(vmlinux-init) \ + -Wl,--start-group $(vmlinux-main) -Wl,--end-group \ + -L/usr/lib -lutil \ + $(filter-out $(vmlinux-lds) $(vmlinux-init) $(vmlinux-main) \ + FORCE ,$^) ; rm -f linux +endef + +#When cleaning we don't include .config, so we don't include +#TT or skas makefiles and don't clean skas_ptregs.h. +CLEAN_FILES += linux x.i gmon.out $(ARCH_DIR)/include/uml-config.h \ + $(GEN_HEADERS) $(ARCH_DIR)/include/skas_ptregs.h + +MRPROPER_FILES += $(SYMLINK_HEADERS) $(ARCH_SYMLINKS) \ + $(addprefix $(ARCH_DIR)/kernel/,$(KERN_SYMLINKS)) $(ARCH_DIR)/os \ + $(ARCH_DIR)/Kconfig_arch + +archclean: + $(Q)$(MAKE) $(clean)=$(ARCH_DIR)/util + @find . \( -name '*.bb' -o -name '*.bbg' -o -name '*.da' \ + -o -name '*.gcov' \) -type f -print | xargs rm -f + +#We need to re-preprocess this when the symlink dest changes. +#So we touch it when needed. +$(ARCH_DIR)/kernel/vmlinux.lds.S: FORCE + $(Q)if [ "$(shell readlink $@)" != "$(LD_SCRIPT-y)" ]; then \ + echo ' SYMLINK $@'; \ + ln -sf $(LD_SCRIPT-y) $@; \ + touch $@; \ + fi; + +$(SYMLINK_HEADERS): + @echo ' SYMLINK $@' + $(Q)cd $(TOPDIR)/$(dir $@) ; \ + ln -sf $(basename $(notdir $@))-$(SUBARCH)$(suffix $@) $(notdir $@) + +include/asm-um/arch: + @echo ' SYMLINK $@' + $(Q)cd $(TOPDIR)/include/asm-um && ln -sf ../asm-$(SUBARCH) arch + +$(ARCH_DIR)/include/sysdep: + @echo ' SYMLINK $@' + $(Q)cd $(ARCH_DIR)/include && ln -sf sysdep-$(SUBARCH) sysdep + +$(ARCH_DIR)/os: + @echo ' SYMLINK $@' + $(Q)cd $(ARCH_DIR) && ln -sf os-$(OS) os + +# Generated files +define filechk_umlconfig + sed 's/ CONFIG/ UML_CONFIG/' +endef + +$(ARCH_DIR)/include/uml-config.h : include/linux/autoconf.h + $(call filechk,umlconfig) + +$(ARCH_DIR)/include/task.h: $(ARCH_DIR)/util/mk_task + $(call filechk,gen_header) + +$(ARCH_DIR)/include/user_constants.h: $(ARCH_DIR)/os/util/mk_user_constants + $(call filechk,gen_header) + +$(ARCH_DIR)/include/kern_constants.h: $(ARCH_DIR)/util/mk_constants + $(call filechk,gen_header) + +$(ARCH_DIR)/include/skas_ptregs.h: $(ARCH_DIR)/kernel/skas/util/mk_ptregs + $(call filechk,gen_header) + +$(ARCH_DIR)/os/util/mk_user_constants: $(ARCH_DIR)/os/util FORCE ; + +$(ARCH_DIR)/util/mk_task $(ARCH_DIR)/util/mk_constants: $(ARCH_DIR)/include/user_constants.h $(ARCH_DIR)/util \ + FORCE ; + +$(ARCH_DIR)/kernel/skas/util/mk_ptregs: $(ARCH_DIR)/kernel/skas/util FORCE ; + +$(ARCH_DIR)/util: scripts_basic $(SYS_DIR)/sc.h FORCE + $(Q)$(MAKE) $(build)=$@ + +$(ARCH_DIR)/kernel/skas/util: scripts_basic FORCE + $(Q)$(MAKE) $(build)=$@ + +$(ARCH_DIR)/os/util: scripts_basic FORCE + $(Q)$(MAKE) $(build)=$@ + +export SUBARCH USER_CFLAGS OS diff --git a/arch/um/Makefile-i386 b/arch/um/Makefile-i386 new file mode 100644 index 000000000000..97b223bfa78e --- /dev/null +++ b/arch/um/Makefile-i386 @@ -0,0 +1,44 @@ +SUBARCH_CORE := arch/um/sys-i386/ + +TOP_ADDR := $(CONFIG_TOP_ADDR) + +ifeq ($(CONFIG_MODE_SKAS),y) + ifneq ($(CONFIG_MODE_TT),y) + START := 0x8048000 + endif +endif + +CFLAGS += -U__$(SUBARCH)__ -U$(SUBARCH) +ARCH_USER_CFLAGS := + +ifneq ($(CONFIG_GPROF),y) +ARCH_CFLAGS += -DUM_FASTCALL +endif + +ELF_ARCH := $(SUBARCH) +ELF_FORMAT := elf32-$(SUBARCH) + +OBJCOPYFLAGS := -O binary -R .note -R .comment -S + +SYS_UTIL_DIR := $(ARCH_DIR)/sys-i386/util + +SYS_HEADERS := $(SYS_DIR)/sc.h $(SYS_DIR)/thread.h + +prepare: $(SYS_HEADERS) + +$(SYS_DIR)/sc.h: $(SYS_UTIL_DIR)/mk_sc + $(call filechk,gen_header) + +$(SYS_DIR)/thread.h: $(SYS_UTIL_DIR)/mk_thread + $(call filechk,gen_header) + +$(SYS_UTIL_DIR)/mk_sc: scripts_basic FORCE + $(Q)$(MAKE) $(build)=$(SYS_UTIL_DIR) $@ + +$(SYS_UTIL_DIR)/mk_thread: scripts_basic $(ARCH_SYMLINKS) $(GEN_HEADERS) FORCE + $(Q)$(MAKE) $(build)=$(SYS_UTIL_DIR) $@ + +$(SYS_UTIL_DIR): scripts_basic include/asm FORCE + $(Q)$(MAKE) $(build)=$(SYS_UTIL_DIR) + +CLEAN_FILES += $(SYS_HEADERS) diff --git a/arch/um/Makefile-ia64 b/arch/um/Makefile-ia64 new file mode 100644 index 000000000000..f84dc23b0f6e --- /dev/null +++ b/arch/um/Makefile-ia64 @@ -0,0 +1 @@ +START_ADDR = 0x1000000000000000 diff --git a/arch/um/Makefile-os-Linux b/arch/um/Makefile-os-Linux new file mode 100644 index 000000000000..0c0f9a1cbbad --- /dev/null +++ b/arch/um/Makefile-os-Linux @@ -0,0 +1,8 @@ +# +# Copyright (C) 2000 Jeff Dike (jdike@karaya.com) +# Licensed under the GPL +# + +# To get a definition of F_SETSIG +USER_CFLAGS += -D_GNU_SOURCE -D_LARGEFILE64_SOURCE +CFLAGS += -D_LARGEFILE64_SOURCE diff --git a/arch/um/Makefile-ppc b/arch/um/Makefile-ppc new file mode 100644 index 000000000000..66fd2003e165 --- /dev/null +++ b/arch/um/Makefile-ppc @@ -0,0 +1,9 @@ +ifeq ($(CONFIG_HOST_2G_2G), y) +START_ADDR = 0x80000000 +else +START_ADDR = 0xc0000000 +endif +ARCH_CFLAGS = -U__powerpc__ -D__UM_PPC__ + +# The arch is ppc, but the elf32 name is powerpc +ELF_SUBARCH = powerpc diff --git a/arch/um/Makefile-skas b/arch/um/Makefile-skas new file mode 100644 index 000000000000..fd18ec572271 --- /dev/null +++ b/arch/um/Makefile-skas @@ -0,0 +1,14 @@ +# +# Copyright (C) 2002 Jeff Dike (jdike@karaya.com) +# Licensed under the GPL +# + +GPROF_OPT += -pg +GCOV_OPT += -fprofile-arcs -ftest-coverage + +CFLAGS-$(CONFIG_GCOV) += $(GCOV_OPT) +CFLAGS-$(CONFIG_GPROF) += $(GPROF_OPT) +LINK-$(CONFIG_GCOV) += $(GCOV_OPT) +LINK-$(CONFIG_GPROF) += $(GPROF_OPT) + +GEN_HEADERS += $(ARCH_DIR)/include/skas_ptregs.h diff --git a/arch/um/Makefile-tt b/arch/um/Makefile-tt new file mode 100644 index 000000000000..03f7b10cfd0b --- /dev/null +++ b/arch/um/Makefile-tt @@ -0,0 +1,5 @@ +# +# Copyright (C) 2002 Jeff Dike (jdike@karaya.com) +# Licensed under the GPL +# + diff --git a/arch/um/Makefile-x86_64 b/arch/um/Makefile-x86_64 new file mode 100644 index 000000000000..a77971133e91 --- /dev/null +++ b/arch/um/Makefile-x86_64 @@ -0,0 +1,32 @@ +# Copyright 2003 - 2004 Pathscale, Inc +# Released under the GPL + +SUBARCH_LIBS := arch/um/sys-x86_64/ +START := 0x60000000 + +CFLAGS += -U__$(SUBARCH)__ -fno-builtin +ARCH_USER_CFLAGS := -D__x86_64__ + +ELF_ARCH := i386:x86-64 +ELF_FORMAT := elf64-x86-64 + +SYS_UTIL_DIR := $(ARCH_DIR)/sys-x86_64/util +SYS_DIR := $(ARCH_DIR)/include/sysdep-x86_64 + +SYS_HEADERS = $(SYS_DIR)/sc.h $(SYS_DIR)/thread.h + +prepare: $(SYS_HEADERS) + +$(SYS_DIR)/sc.h: $(SYS_UTIL_DIR)/mk_sc + $(call filechk,gen_header) + +$(SYS_DIR)/thread.h: $(SYS_UTIL_DIR)/mk_thread + $(call filechk,gen_header) + +$(SYS_UTIL_DIR)/mk_sc: scripts_basic FORCE + $(Q)$(MAKE) $(build)=$(SYS_UTIL_DIR) $@ + +$(SYS_UTIL_DIR)/mk_thread: scripts_basic $(ARCH_SYMLINKS) $(GEN_HEADERS) FORCE + $(Q)$(MAKE) $(build)=$(SYS_UTIL_DIR) $@ + +CLEAN_FILES += $(SYS_HEADERS) diff --git a/arch/um/config.release b/arch/um/config.release new file mode 100644 index 000000000000..fc68bcb9294e --- /dev/null +++ b/arch/um/config.release @@ -0,0 +1,333 @@ +# +# Automatically generated make config: don't edit +# +CONFIG_USERMODE=y +# CONFIG_ISA is not set +# CONFIG_SBUS is not set +# CONFIG_PCI is not set +CONFIG_UID16=y +CONFIG_RWSEM_XCHGADD_ALGORITHM=y + +# +# Code maturity level options +# +CONFIG_EXPERIMENTAL=y + +# +# General Setup +# +CONFIG_STDIO_CONSOLE=y +CONFIG_NET=y +CONFIG_SYSVIPC=y +CONFIG_BSD_PROCESS_ACCT=y +CONFIG_SYSCTL=y +CONFIG_BINFMT_AOUT=y +CONFIG_BINFMT_ELF=y +CONFIG_BINFMT_MISC=y +CONFIG_UNIX98_PTYS=y +CONFIG_UNIX98_PTY_COUNT=256 +CONFIG_SSL=y +CONFIG_HOSTFS=y +CONFIG_MCONSOLE=y +CONFIG_MAGIC_SYSRQ=y +# CONFIG_HOST_2G_2G is not set +# CONFIG_UML_SMP is not set +# CONFIG_SMP is not set +CONFIG_CON_ZERO_CHAN="fd:0,fd:1" +CONFIG_CON_CHAN="xterm" +CONFIG_SSL_CHAN="pty" +CONFIG_NEST_LEVEL=0 +CONFIG_KERNEL_HALF_GIGS=1 + +# +# Loadable module support +# +CONFIG_MODULES=y +CONFIG_KMOD=y + +# +# Devices +# +CONFIG_BLK_DEV_UBD=y +# CONFIG_BLK_DEV_UBD_SYNC is not set +CONFIG_BLK_DEV_LOOP=y +CONFIG_BLK_DEV_NBD=y +CONFIG_BLK_DEV_RAM=y +CONFIG_BLK_DEV_RAM_SIZE=4096 +CONFIG_BLK_DEV_INITRD=y +# CONFIG_MMAPPER is not set +CONFIG_UML_SOUND=y +CONFIG_SOUND=y +CONFIG_HOSTAUDIO=y +# CONFIG_UML_WATCHDOG is not set +# CONFIG_TTY_LOG is not set +CONFIG_FD_CHAN=y +# CONFIG_NULL_CHAN is not set +CONFIG_PORT_CHAN=y +CONFIG_PTY_CHAN=y +CONFIG_TTY_CHAN=y +CONFIG_XTERM_CHAN=y + +# +# Networking options +# +CONFIG_PACKET=y +CONFIG_PACKET_MMAP=y +# CONFIG_NETLINK_DEV is not set +# CONFIG_NETFILTER is not set +# CONFIG_FILTER is not set +CONFIG_UNIX=y +CONFIG_INET=y +# CONFIG_IP_MULTICAST is not set +# CONFIG_IP_ADVANCED_ROUTER is not set +# CONFIG_IP_PNP is not set +# CONFIG_NET_IPIP is not set +# CONFIG_NET_IPGRE is not set +# CONFIG_ARPD is not set +# CONFIG_INET_ECN is not set +# CONFIG_SYN_COOKIES is not set +# CONFIG_IPV6 is not set +# CONFIG_KHTTPD is not set +# CONFIG_ATM is not set +# CONFIG_VLAN_8021Q is not set + +# +# +# +# CONFIG_IPX is not set +# CONFIG_ATALK is not set +# CONFIG_DECNET is not set +# CONFIG_BRIDGE is not set +# CONFIG_X25 is not set +# CONFIG_LAPB is not set +# CONFIG_LLC is not set +# CONFIG_NET_DIVERT is not set +# CONFIG_ECONET is not set +# CONFIG_WAN_ROUTER is not set +# CONFIG_NET_HW_FLOWCONTROL is not set + +# +# QoS and/or fair queueing +# +# CONFIG_NET_SCHED is not set + +# +# Network device support +# +CONFIG_UML_NET=y +CONFIG_UML_NET_ETHERTAP=y +CONFIG_UML_NET_TUNTAP=y +CONFIG_UML_NET_SLIP=y +CONFIG_UML_NET_DAEMON=y +CONFIG_UML_NET_MCAST=y +CONFIG_NETDEVICES=y + +# +# ARCnet devices +# +# CONFIG_ARCNET is not set +CONFIG_DUMMY=y +CONFIG_BONDING=m +CONFIG_EQUALIZER=m +CONFIG_TUN=y +# CONFIG_ETHERTAP is not set + +# +# Ethernet (10 or 100Mbit) +# +# CONFIG_NET_ETHERNET is not set + +# +# Ethernet (1000 Mbit) +# +# CONFIG_ACENIC is not set +# CONFIG_DL2K is not set +# CONFIG_MYRI_SBUS is not set +# CONFIG_NS83820 is not set +# CONFIG_HAMACHI is not set +# CONFIG_YELLOWFIN is not set +# CONFIG_SK98LIN is not set +# CONFIG_FDDI is not set +# CONFIG_HIPPI is not set +CONFIG_PLIP=m +CONFIG_PPP=m +CONFIG_PPP_MULTILINK=y +# CONFIG_PPP_FILTER is not set +# CONFIG_PPP_ASYNC is not set +CONFIG_PPP_SYNC_TTY=m +CONFIG_PPP_DEFLATE=m +CONFIG_PPP_BSDCOMP=m +CONFIG_PPPOE=m +CONFIG_SLIP=m +CONFIG_SLIP_COMPRESSED=y +CONFIG_SLIP_SMART=y +# CONFIG_SLIP_MODE_SLIP6 is not set + +# +# Wireless LAN (non-hamradio) +# +# CONFIG_NET_RADIO is not set + +# +# Token Ring devices +# +# CONFIG_TR is not set +# CONFIG_NET_FC is not set +# CONFIG_RCPCI is not set +CONFIG_SHAPER=m + +# +# Wan interfaces +# +# CONFIG_WAN is not set + +# +# File systems +# +CONFIG_QUOTA=y +CONFIG_AUTOFS_FS=m +CONFIG_AUTOFS4_FS=m +CONFIG_REISERFS_FS=m +# CONFIG_REISERFS_CHECK is not set +# CONFIG_REISERFS_PROC_INFO is not set +CONFIG_ADFS_FS=m +# CONFIG_ADFS_FS_RW is not set +CONFIG_AFFS_FS=m +CONFIG_HFS_FS=m +CONFIG_BFS_FS=m +CONFIG_EXT3_FS=y +CONFIG_JBD=y +# CONFIG_JBD_DEBUG is not set +CONFIG_FAT_FS=y +CONFIG_MSDOS_FS=y +CONFIG_UMSDOS_FS=y +CONFIG_VFAT_FS=y +CONFIG_EFS_FS=m +# CONFIG_JFFS_FS is not set +# CONFIG_JFFS2_FS is not set +CONFIG_CRAMFS=m +CONFIG_TMPFS=y +CONFIG_RAMFS=m +CONFIG_ISO9660_FS=y +# CONFIG_JOLIET is not set +# CONFIG_ZISOFS is not set +CONFIG_MINIX_FS=m +CONFIG_VXFS_FS=m +# CONFIG_NTFS_FS is not set +# CONFIG_NTFS_RW is not set +CONFIG_HPFS_FS=m +CONFIG_PROC_FS=y +CONFIG_DEVFS_FS=y +CONFIG_DEVFS_MOUNT=y +# CONFIG_DEVFS_DEBUG is not set +CONFIG_DEVPTS_FS=y +CONFIG_QNX4FS_FS=m +# CONFIG_QNX4FS_RW is not set +CONFIG_ROMFS_FS=m +CONFIG_EXT2_FS=y +CONFIG_SYSV_FS=m +CONFIG_UDF_FS=m +CONFIG_UFS_FS=m +# CONFIG_UFS_FS_WRITE is not set + +# +# Network File Systems +# +# CONFIG_CODA_FS is not set +# CONFIG_INTERMEZZO_FS is not set +CONFIG_NFS_FS=y +CONFIG_NFS_V3=y +# CONFIG_ROOT_NFS is not set +CONFIG_NFSD=y +CONFIG_NFSD_V3=y +CONFIG_SUNRPC=y +CONFIG_LOCKD=y +CONFIG_LOCKD_V4=y +# CONFIG_SMB_FS is not set +# CONFIG_NCP_FS is not set +# CONFIG_NCPFS_PACKET_SIGNING is not set +# CONFIG_NCPFS_IOCTL_LOCKING is not set +# CONFIG_NCPFS_STRONG is not set +# CONFIG_NCPFS_NFS_NS is not set +# CONFIG_NCPFS_OS2_NS is not set +# CONFIG_NCPFS_SMALLDOS is not set +# CONFIG_NCPFS_NLS is not set +# CONFIG_NCPFS_EXTRAS is not set +# CONFIG_ZISOFS_FS is not set +CONFIG_ZLIB_FS_INFLATE=m + +# +# Partition Types +# +# CONFIG_PARTITION_ADVANCED is not set +CONFIG_MSDOS_PARTITION=y +# CONFIG_SMB_NLS is not set +CONFIG_NLS=y + +# +# Native Language Support +# +CONFIG_NLS_DEFAULT="iso8859-1" +# CONFIG_NLS_CODEPAGE_437 is not set +# CONFIG_NLS_CODEPAGE_737 is not set +# CONFIG_NLS_CODEPAGE_775 is not set +# CONFIG_NLS_CODEPAGE_850 is not set +# CONFIG_NLS_CODEPAGE_852 is not set +# CONFIG_NLS_CODEPAGE_855 is not set +# CONFIG_NLS_CODEPAGE_857 is not set +# CONFIG_NLS_CODEPAGE_860 is not set +# CONFIG_NLS_CODEPAGE_861 is not set +# CONFIG_NLS_CODEPAGE_862 is not set +# CONFIG_NLS_CODEPAGE_863 is not set +# CONFIG_NLS_CODEPAGE_864 is not set +# CONFIG_NLS_CODEPAGE_865 is not set +# CONFIG_NLS_CODEPAGE_866 is not set +# CONFIG_NLS_CODEPAGE_869 is not set +# CONFIG_NLS_CODEPAGE_936 is not set +# CONFIG_NLS_CODEPAGE_950 is not set +# CONFIG_NLS_CODEPAGE_932 is not set +# CONFIG_NLS_CODEPAGE_949 is not set +# CONFIG_NLS_CODEPAGE_874 is not set +# CONFIG_NLS_ISO8859_8 is not set +# CONFIG_NLS_CODEPAGE_1250 is not set +# CONFIG_NLS_CODEPAGE_1251 is not set +# CONFIG_NLS_ISO8859_1 is not set +# CONFIG_NLS_ISO8859_2 is not set +# CONFIG_NLS_ISO8859_3 is not set +# CONFIG_NLS_ISO8859_4 is not set +# CONFIG_NLS_ISO8859_5 is not set +# CONFIG_NLS_ISO8859_6 is not set +# CONFIG_NLS_ISO8859_7 is not set +# CONFIG_NLS_ISO8859_9 is not set +# CONFIG_NLS_ISO8859_13 is not set +# CONFIG_NLS_ISO8859_14 is not set +# CONFIG_NLS_ISO8859_15 is not set +# CONFIG_NLS_KOI8_R is not set +# CONFIG_NLS_KOI8_U is not set +# CONFIG_NLS_UTF8 is not set + +# +# Multi-device support (RAID and LVM) +# +# CONFIG_MD is not set +# CONFIG_BLK_DEV_MD is not set +# CONFIG_MD_LINEAR is not set +# CONFIG_MD_RAID0 is not set +# CONFIG_MD_RAID1 is not set +# CONFIG_MD_RAID5 is not set +# CONFIG_MD_MULTIPATH is not set +# CONFIG_BLK_DEV_LVM is not set + +# +# Memory Technology Devices (MTD) +# +# CONFIG_MTD is not set + +# +# Kernel hacking +# +# CONFIG_DEBUG_SLAB is not set +# CONFIG_DEBUG_INFO is not set +# CONFIG_PT_PROXY is not set +# CONFIG_GPROF is not set +# CONFIG_GCOV is not set diff --git a/arch/um/defconfig b/arch/um/defconfig new file mode 100644 index 000000000000..fc3075c589d8 --- /dev/null +++ b/arch/um/defconfig @@ -0,0 +1,417 @@ +# +# Automatically generated make config: don't edit +# Linux kernel version: 2.6.12-rc1-bk1 +# Sun Mar 20 16:53:00 2005 +# +CONFIG_GENERIC_HARDIRQS=y +CONFIG_UML=y +CONFIG_MMU=y +CONFIG_UID16=y +CONFIG_RWSEM_GENERIC_SPINLOCK=y +CONFIG_GENERIC_CALIBRATE_DELAY=y + +# +# UML-specific options +# +CONFIG_MODE_TT=y +CONFIG_MODE_SKAS=y +# CONFIG_64_BIT is not set +CONFIG_TOP_ADDR=0xc0000000 +# CONFIG_3_LEVEL_PGTABLES is not set +CONFIG_ARCH_HAS_SC_SIGNALS=y +CONFIG_ARCH_REUSE_HOST_VSYSCALL_AREA=y +CONFIG_LD_SCRIPT_STATIC=y +CONFIG_NET=y +CONFIG_BINFMT_ELF=y +CONFIG_BINFMT_MISC=m +CONFIG_HOSTFS=y +CONFIG_MCONSOLE=y +# CONFIG_MAGIC_SYSRQ is not set +# CONFIG_HOST_2G_2G is not set +# CONFIG_SMP is not set +CONFIG_NEST_LEVEL=0 +CONFIG_KERNEL_HALF_GIGS=1 +# CONFIG_HIGHMEM is not set +CONFIG_KERNEL_STACK_ORDER=2 +CONFIG_UML_REAL_TIME_CLOCK=y + +# +# Code maturity level options +# +CONFIG_EXPERIMENTAL=y +CONFIG_CLEAN_COMPILE=y +CONFIG_BROKEN_ON_SMP=y + +# +# General setup +# +CONFIG_LOCALVERSION="" +CONFIG_SWAP=y +CONFIG_SYSVIPC=y +CONFIG_POSIX_MQUEUE=y +CONFIG_BSD_PROCESS_ACCT=y +# CONFIG_BSD_PROCESS_ACCT_V3 is not set +CONFIG_SYSCTL=y +# CONFIG_AUDIT is not set +# CONFIG_HOTPLUG is not set +CONFIG_KOBJECT_UEVENT=y +CONFIG_IKCONFIG=y +CONFIG_IKCONFIG_PROC=y +# CONFIG_EMBEDDED is not set +CONFIG_KALLSYMS=y +# CONFIG_KALLSYMS_ALL is not set +CONFIG_KALLSYMS_EXTRA_PASS=y +CONFIG_BASE_FULL=y +CONFIG_FUTEX=y +CONFIG_EPOLL=y +CONFIG_SHMEM=y +CONFIG_CC_ALIGN_FUNCTIONS=0 +CONFIG_CC_ALIGN_LABELS=0 +CONFIG_CC_ALIGN_LOOPS=0 +CONFIG_CC_ALIGN_JUMPS=0 +# CONFIG_TINY_SHMEM is not set +CONFIG_BASE_SMALL=0 + +# +# Loadable module support +# +CONFIG_MODULES=y +CONFIG_MODULE_UNLOAD=y +# CONFIG_MODULE_FORCE_UNLOAD is not set +CONFIG_OBSOLETE_MODPARM=y +# CONFIG_MODULE_SRCVERSION_ALL is not set +CONFIG_KMOD=y + +# +# Generic Driver Options +# +CONFIG_STANDALONE=y +CONFIG_PREVENT_FIRMWARE_BUILD=y +# CONFIG_FW_LOADER is not set +# CONFIG_DEBUG_DRIVER is not set + +# +# Character Devices +# +CONFIG_STDERR_CONSOLE=y +CONFIG_STDIO_CONSOLE=y +CONFIG_SSL=y +CONFIG_NULL_CHAN=y +CONFIG_PORT_CHAN=y +CONFIG_PTY_CHAN=y +CONFIG_TTY_CHAN=y +CONFIG_XTERM_CHAN=y +# CONFIG_NOCONFIG_CHAN is not set +CONFIG_CON_ZERO_CHAN="fd:0,fd:1" +CONFIG_CON_CHAN="xterm" +CONFIG_SSL_CHAN="pty" +CONFIG_UNIX98_PTYS=y +CONFIG_LEGACY_PTYS=y +CONFIG_LEGACY_PTY_COUNT=256 +# CONFIG_WATCHDOG is not set +CONFIG_UML_SOUND=m +CONFIG_SOUND=m +CONFIG_HOSTAUDIO=m +CONFIG_UML_RANDOM=y + +# +# Block devices +# +CONFIG_BLK_DEV_UBD=y +CONFIG_BLK_DEV_UBD_SYNC=y +CONFIG_BLK_DEV_COW_COMMON=y +CONFIG_BLK_DEV_LOOP=m +# CONFIG_BLK_DEV_CRYPTOLOOP is not set +CONFIG_BLK_DEV_NBD=m +# CONFIG_BLK_DEV_RAM is not set +CONFIG_BLK_DEV_RAM_COUNT=16 +CONFIG_INITRAMFS_SOURCE="" +# CONFIG_LBD is not set + +# +# IO Schedulers +# +CONFIG_IOSCHED_NOOP=y +CONFIG_IOSCHED_AS=y +CONFIG_IOSCHED_DEADLINE=y +CONFIG_IOSCHED_CFQ=y +# CONFIG_ATA_OVER_ETH is not set +CONFIG_NETDEVICES=y + +# +# UML Network Devices +# +CONFIG_UML_NET=y +CONFIG_UML_NET_ETHERTAP=y +CONFIG_UML_NET_TUNTAP=y +CONFIG_UML_NET_SLIP=y +CONFIG_UML_NET_DAEMON=y +CONFIG_UML_NET_MCAST=y +CONFIG_UML_NET_SLIRP=y + +# +# Networking support +# + +# +# Networking options +# +CONFIG_PACKET=y +CONFIG_PACKET_MMAP=y +# CONFIG_NETLINK_DEV is not set +CONFIG_UNIX=y +# CONFIG_NET_KEY is not set +CONFIG_INET=y +# CONFIG_IP_MULTICAST is not set +# CONFIG_IP_ADVANCED_ROUTER is not set +# CONFIG_IP_PNP is not set +# CONFIG_NET_IPIP is not set +# CONFIG_NET_IPGRE is not set +# CONFIG_ARPD is not set +# CONFIG_SYN_COOKIES is not set +# CONFIG_INET_AH is not set +# CONFIG_INET_ESP is not set +# CONFIG_INET_IPCOMP is not set +# CONFIG_INET_TUNNEL is not set +CONFIG_IP_TCPDIAG=y +# CONFIG_IP_TCPDIAG_IPV6 is not set +# CONFIG_IPV6 is not set +# CONFIG_NETFILTER is not set + +# +# SCTP Configuration (EXPERIMENTAL) +# +# CONFIG_IP_SCTP is not set +# CONFIG_ATM is not set +# CONFIG_BRIDGE is not set +# CONFIG_VLAN_8021Q is not set +# CONFIG_DECNET is not set +# CONFIG_LLC2 is not set +# CONFIG_IPX is not set +# CONFIG_ATALK is not set +# CONFIG_X25 is not set +# CONFIG_LAPB is not set +# CONFIG_NET_DIVERT is not set +# CONFIG_ECONET is not set +# CONFIG_WAN_ROUTER is not set + +# +# QoS and/or fair queueing +# +# CONFIG_NET_SCHED is not set +# CONFIG_NET_CLS_ROUTE is not set + +# +# Network testing +# +# CONFIG_NET_PKTGEN is not set +# CONFIG_NETPOLL is not set +# CONFIG_NET_POLL_CONTROLLER is not set +# CONFIG_HAMRADIO is not set +# CONFIG_IRDA is not set +# CONFIG_BT is not set +CONFIG_DUMMY=m +# CONFIG_BONDING is not set +# CONFIG_EQUALIZER is not set +CONFIG_TUN=m + +# +# Wan interfaces +# +# CONFIG_WAN is not set +CONFIG_PPP=m +# CONFIG_PPP_MULTILINK is not set +# CONFIG_PPP_FILTER is not set +# CONFIG_PPP_ASYNC is not set +# CONFIG_PPP_SYNC_TTY is not set +# CONFIG_PPP_DEFLATE is not set +# CONFIG_PPP_BSDCOMP is not set +# CONFIG_PPPOE is not set +CONFIG_SLIP=m +# CONFIG_SLIP_COMPRESSED is not set +# CONFIG_SLIP_SMART is not set +# CONFIG_SLIP_MODE_SLIP6 is not set +# CONFIG_SHAPER is not set +# CONFIG_NETCONSOLE is not set + +# +# File systems +# +CONFIG_EXT2_FS=y +# CONFIG_EXT2_FS_XATTR is not set +CONFIG_EXT3_FS=y +# CONFIG_EXT3_FS_XATTR is not set +CONFIG_JBD=y +# CONFIG_JBD_DEBUG is not set +CONFIG_REISERFS_FS=y +# CONFIG_REISERFS_CHECK is not set +# CONFIG_REISERFS_PROC_INFO is not set +# CONFIG_REISERFS_FS_XATTR is not set +# CONFIG_JFS_FS is not set + +# +# XFS support +# +# CONFIG_XFS_FS is not set +# CONFIG_MINIX_FS is not set +# CONFIG_ROMFS_FS is not set +CONFIG_QUOTA=y +# CONFIG_QFMT_V1 is not set +# CONFIG_QFMT_V2 is not set +CONFIG_QUOTACTL=y +CONFIG_DNOTIFY=y +CONFIG_AUTOFS_FS=m +CONFIG_AUTOFS4_FS=m + +# +# CD-ROM/DVD Filesystems +# +CONFIG_ISO9660_FS=m +CONFIG_JOLIET=y +# CONFIG_ZISOFS is not set +# CONFIG_UDF_FS is not set + +# +# DOS/FAT/NT Filesystems +# +# CONFIG_MSDOS_FS is not set +# CONFIG_VFAT_FS is not set +# CONFIG_NTFS_FS is not set + +# +# Pseudo filesystems +# +CONFIG_PROC_FS=y +CONFIG_PROC_KCORE=y +CONFIG_SYSFS=y +# CONFIG_DEVFS_FS is not set +# CONFIG_DEVPTS_FS_XATTR is not set +CONFIG_TMPFS=y +# CONFIG_TMPFS_XATTR is not set +# CONFIG_HUGETLB_PAGE is not set +CONFIG_RAMFS=y + +# +# Miscellaneous filesystems +# +# CONFIG_ADFS_FS is not set +# CONFIG_AFFS_FS is not set +# CONFIG_HFS_FS is not set +# CONFIG_HFSPLUS_FS is not set +# CONFIG_BEFS_FS is not set +# CONFIG_BFS_FS is not set +# CONFIG_EFS_FS is not set +# CONFIG_CRAMFS is not set +# CONFIG_VXFS_FS is not set +# CONFIG_HPFS_FS is not set +# CONFIG_QNX4FS_FS is not set +# CONFIG_SYSV_FS is not set +# CONFIG_UFS_FS is not set + +# +# Network File Systems +# +# CONFIG_NFS_FS is not set +# CONFIG_NFSD is not set +# CONFIG_SMB_FS is not set +# CONFIG_CIFS is not set +# CONFIG_NCP_FS is not set +# CONFIG_CODA_FS is not set +# CONFIG_AFS_FS is not set + +# +# Partition Types +# +# CONFIG_PARTITION_ADVANCED is not set +CONFIG_MSDOS_PARTITION=y + +# +# Native Language Support +# +CONFIG_NLS=y +CONFIG_NLS_DEFAULT="iso8859-1" +# CONFIG_NLS_CODEPAGE_437 is not set +# CONFIG_NLS_CODEPAGE_737 is not set +# CONFIG_NLS_CODEPAGE_775 is not set +# CONFIG_NLS_CODEPAGE_850 is not set +# CONFIG_NLS_CODEPAGE_852 is not set +# CONFIG_NLS_CODEPAGE_855 is not set +# CONFIG_NLS_CODEPAGE_857 is not set +# CONFIG_NLS_CODEPAGE_860 is not set +# CONFIG_NLS_CODEPAGE_861 is not set +# CONFIG_NLS_CODEPAGE_862 is not set +# CONFIG_NLS_CODEPAGE_863 is not set +# CONFIG_NLS_CODEPAGE_864 is not set +# CONFIG_NLS_CODEPAGE_865 is not set +# CONFIG_NLS_CODEPAGE_866 is not set +# CONFIG_NLS_CODEPAGE_869 is not set +# CONFIG_NLS_CODEPAGE_936 is not set +# CONFIG_NLS_CODEPAGE_950 is not set +# CONFIG_NLS_CODEPAGE_932 is not set +# CONFIG_NLS_CODEPAGE_949 is not set +# CONFIG_NLS_CODEPAGE_874 is not set +# CONFIG_NLS_ISO8859_8 is not set +# CONFIG_NLS_CODEPAGE_1250 is not set +# CONFIG_NLS_CODEPAGE_1251 is not set +# CONFIG_NLS_ASCII is not set +# CONFIG_NLS_ISO8859_1 is not set +# CONFIG_NLS_ISO8859_2 is not set +# CONFIG_NLS_ISO8859_3 is not set +# CONFIG_NLS_ISO8859_4 is not set +# CONFIG_NLS_ISO8859_5 is not set +# CONFIG_NLS_ISO8859_6 is not set +# CONFIG_NLS_ISO8859_7 is not set +# CONFIG_NLS_ISO8859_9 is not set +# CONFIG_NLS_ISO8859_13 is not set +# CONFIG_NLS_ISO8859_14 is not set +# CONFIG_NLS_ISO8859_15 is not set +# CONFIG_NLS_KOI8_R is not set +# CONFIG_NLS_KOI8_U is not set +# CONFIG_NLS_UTF8 is not set + +# +# Security options +# +# CONFIG_KEYS is not set +# CONFIG_SECURITY is not set + +# +# Cryptographic options +# +# CONFIG_CRYPTO is not set + +# +# Hardware crypto devices +# + +# +# Library routines +# +# CONFIG_CRC_CCITT is not set +CONFIG_CRC32=m +# CONFIG_LIBCRC32C is not set + +# +# Multi-device support (RAID and LVM) +# +# CONFIG_MD is not set +# CONFIG_INPUT is not set + +# +# Kernel hacking +# +# CONFIG_PRINTK_TIME is not set +CONFIG_DEBUG_KERNEL=y +CONFIG_LOG_BUF_SHIFT=14 +# CONFIG_SCHEDSTATS is not set +# CONFIG_DEBUG_SLAB is not set +# CONFIG_DEBUG_SPINLOCK is not set +# CONFIG_DEBUG_SPINLOCK_SLEEP is not set +# CONFIG_DEBUG_KOBJECT is not set +CONFIG_DEBUG_INFO=y +# CONFIG_DEBUG_FS is not set +CONFIG_FRAME_POINTER=y +CONFIG_PT_PROXY=y +# CONFIG_GPROF is not set +# CONFIG_GCOV is not set +# CONFIG_SYSCALL_DEBUG is not set diff --git a/arch/um/drivers/Makefile b/arch/um/drivers/Makefile new file mode 100644 index 000000000000..323f72c64cd2 --- /dev/null +++ b/arch/um/drivers/Makefile @@ -0,0 +1,46 @@ +# +# Copyright (C) 2000, 2002, 2003 Jeff Dike (jdike@karaya.com) +# Licensed under the GPL +# + +# pcap is broken in 2.5 because kbuild doesn't allow pcap.a to be linked +# in to pcap.o + +slip-objs := slip_kern.o slip_user.o +slirp-objs := slirp_kern.o slirp_user.o +daemon-objs := daemon_kern.o daemon_user.o +mcast-objs := mcast_kern.o mcast_user.o +#pcap-objs := pcap_kern.o pcap_user.o $(PCAP) +net-objs := net_kern.o net_user.o +mconsole-objs := mconsole_kern.o mconsole_user.o +hostaudio-objs := hostaudio_kern.o +ubd-objs := ubd_kern.o ubd_user.o +port-objs := port_kern.o port_user.o +harddog-objs := harddog_kern.o harddog_user.o + +obj-y := stdio_console.o fd.o chan_kern.o chan_user.o line.o +obj-$(CONFIG_SSL) += ssl.o +obj-$(CONFIG_STDERR_CONSOLE) += stderr_console.o + +obj-$(CONFIG_UML_NET_SLIP) += slip.o +obj-$(CONFIG_UML_NET_SLIRP) += slirp.o +obj-$(CONFIG_UML_NET_DAEMON) += daemon.o +obj-$(CONFIG_UML_NET_MCAST) += mcast.o +#obj-$(CONFIG_UML_NET_PCAP) += pcap.o $(PCAP) +obj-$(CONFIG_UML_NET) += net.o +obj-$(CONFIG_MCONSOLE) += mconsole.o +obj-$(CONFIG_MMAPPER) += mmapper_kern.o +obj-$(CONFIG_BLK_DEV_UBD) += ubd.o +obj-$(CONFIG_HOSTAUDIO) += hostaudio.o +obj-$(CONFIG_NULL_CHAN) += null.o +obj-$(CONFIG_PORT_CHAN) += port.o +obj-$(CONFIG_PTY_CHAN) += pty.o +obj-$(CONFIG_TTY_CHAN) += tty.o +obj-$(CONFIG_XTERM_CHAN) += xterm.o xterm_kern.o +obj-$(CONFIG_UML_WATCHDOG) += harddog.o +obj-$(CONFIG_BLK_DEV_COW_COMMON) += cow_user.o +obj-$(CONFIG_UML_RANDOM) += random.o + +USER_OBJS := fd.o null.o pty.o tty.o xterm.o + +include arch/um/scripts/Makefile.rules diff --git a/arch/um/drivers/chan_kern.c b/arch/um/drivers/chan_kern.c new file mode 100644 index 000000000000..1f77deb3fd23 --- /dev/null +++ b/arch/um/drivers/chan_kern.c @@ -0,0 +1,577 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <linux/stddef.h> +#include <linux/kernel.h> +#include <linux/list.h> +#include <linux/slab.h> +#include <linux/tty.h> +#include <linux/string.h> +#include <linux/tty_flip.h> +#include <asm/irq.h> +#include "chan_kern.h" +#include "user_util.h" +#include "kern.h" +#include "irq_user.h" +#include "sigio.h" +#include "line.h" +#include "os.h" + +#ifdef CONFIG_NOCONFIG_CHAN +static void *not_configged_init(char *str, int device, struct chan_opts *opts) +{ + printk(KERN_ERR "Using a channel type which is configured out of " + "UML\n"); + return(NULL); +} + +static int not_configged_open(int input, int output, int primary, void *data, + char **dev_out) +{ + printk(KERN_ERR "Using a channel type which is configured out of " + "UML\n"); + return(-ENODEV); +} + +static void not_configged_close(int fd, void *data) +{ + printk(KERN_ERR "Using a channel type which is configured out of " + "UML\n"); +} + +static int not_configged_read(int fd, char *c_out, void *data) +{ + printk(KERN_ERR "Using a channel type which is configured out of " + "UML\n"); + return(-EIO); +} + +static int not_configged_write(int fd, const char *buf, int len, void *data) +{ + printk(KERN_ERR "Using a channel type which is configured out of " + "UML\n"); + return(-EIO); +} + +static int not_configged_console_write(int fd, const char *buf, int len, + void *data) +{ + printk(KERN_ERR "Using a channel type which is configured out of " + "UML\n"); + return(-EIO); +} + +static int not_configged_window_size(int fd, void *data, unsigned short *rows, + unsigned short *cols) +{ + printk(KERN_ERR "Using a channel type which is configured out of " + "UML\n"); + return(-ENODEV); +} + +static void not_configged_free(void *data) +{ + printk(KERN_ERR "Using a channel type which is configured out of " + "UML\n"); +} + +static struct chan_ops not_configged_ops = { + .init = not_configged_init, + .open = not_configged_open, + .close = not_configged_close, + .read = not_configged_read, + .write = not_configged_write, + .console_write = not_configged_console_write, + .window_size = not_configged_window_size, + .free = not_configged_free, + .winch = 0, +}; +#endif /* CONFIG_NOCONFIG_CHAN */ + +void generic_close(int fd, void *unused) +{ + os_close_file(fd); +} + +int generic_read(int fd, char *c_out, void *unused) +{ + int n; + + n = os_read_file(fd, c_out, sizeof(*c_out)); + + if(n == -EAGAIN) + return(0); + else if(n == 0) + return(-EIO); + return(n); +} + +/* XXX Trivial wrapper around os_write_file */ + +int generic_write(int fd, const char *buf, int n, void *unused) +{ + return(os_write_file(fd, buf, n)); +} + +int generic_window_size(int fd, void *unused, unsigned short *rows_out, + unsigned short *cols_out) +{ + int rows, cols; + int ret; + + ret = os_window_size(fd, &rows, &cols); + if(ret < 0) + return(ret); + + ret = ((*rows_out != rows) || (*cols_out != cols)); + + *rows_out = rows; + *cols_out = cols; + + return(ret); +} + +void generic_free(void *data) +{ + kfree(data); +} + +static void tty_receive_char(struct tty_struct *tty, char ch) +{ + if(tty == NULL) return; + + if(I_IXON(tty) && !I_IXOFF(tty) && !tty->raw) { + if(ch == STOP_CHAR(tty)){ + stop_tty(tty); + return; + } + else if(ch == START_CHAR(tty)){ + start_tty(tty); + return; + } + } + + if((tty->flip.flag_buf_ptr == NULL) || + (tty->flip.char_buf_ptr == NULL)) + return; + tty_insert_flip_char(tty, ch, TTY_NORMAL); +} + +static int open_one_chan(struct chan *chan, int input, int output, int primary) +{ + int fd; + + if(chan->opened) return(0); + if(chan->ops->open == NULL) fd = 0; + else fd = (*chan->ops->open)(input, output, primary, chan->data, + &chan->dev); + if(fd < 0) return(fd); + chan->fd = fd; + + chan->opened = 1; + return(0); +} + +int open_chan(struct list_head *chans) +{ + struct list_head *ele; + struct chan *chan; + int ret, err = 0; + + list_for_each(ele, chans){ + chan = list_entry(ele, struct chan, list); + ret = open_one_chan(chan, chan->input, chan->output, + chan->primary); + if(chan->primary) err = ret; + } + return(err); +} + +void chan_enable_winch(struct list_head *chans, struct tty_struct *tty) +{ + struct list_head *ele; + struct chan *chan; + + list_for_each(ele, chans){ + chan = list_entry(ele, struct chan, list); + if(chan->primary && chan->output && chan->ops->winch){ + register_winch(chan->fd, tty); + return; + } + } +} + +void enable_chan(struct list_head *chans, struct tty_struct *tty) +{ + struct list_head *ele; + struct chan *chan; + + list_for_each(ele, chans){ + chan = list_entry(ele, struct chan, list); + if(!chan->opened) continue; + + line_setup_irq(chan->fd, chan->input, chan->output, tty); + } +} + +void close_chan(struct list_head *chans) +{ + struct chan *chan; + + /* Close in reverse order as open in case more than one of them + * refers to the same device and they save and restore that device's + * state. Then, the first one opened will have the original state, + * so it must be the last closed. + */ + list_for_each_entry_reverse(chan, chans, list) { + if(!chan->opened) continue; + if(chan->ops->close != NULL) + (*chan->ops->close)(chan->fd, chan->data); + chan->opened = 0; + chan->fd = -1; + } +} + +int write_chan(struct list_head *chans, const char *buf, int len, + int write_irq) +{ + struct list_head *ele; + struct chan *chan = NULL; + int n, ret = 0; + + list_for_each(ele, chans) { + chan = list_entry(ele, struct chan, list); + if (!chan->output || (chan->ops->write == NULL)) + continue; + n = chan->ops->write(chan->fd, buf, len, chan->data); + if (chan->primary) { + ret = n; + if ((ret == -EAGAIN) || ((ret >= 0) && (ret < len))) + reactivate_fd(chan->fd, write_irq); + } + } + return(ret); +} + +int console_write_chan(struct list_head *chans, const char *buf, int len) +{ + struct list_head *ele; + struct chan *chan; + int n, ret = 0; + + list_for_each(ele, chans){ + chan = list_entry(ele, struct chan, list); + if(!chan->output || (chan->ops->console_write == NULL)) + continue; + n = chan->ops->console_write(chan->fd, buf, len, chan->data); + if(chan->primary) ret = n; + } + return(ret); +} + +int console_open_chan(struct line *line, struct console *co, struct chan_opts *opts) +{ + if (!list_empty(&line->chan_list)) + return 0; + + if (0 != parse_chan_pair(line->init_str, &line->chan_list, + line->init_pri, co->index, opts)) + return -1; + if (0 != open_chan(&line->chan_list)) + return -1; + printk("Console initialized on /dev/%s%d\n",co->name,co->index); + return 0; +} + +int chan_window_size(struct list_head *chans, unsigned short *rows_out, + unsigned short *cols_out) +{ + struct list_head *ele; + struct chan *chan; + + list_for_each(ele, chans){ + chan = list_entry(ele, struct chan, list); + if(chan->primary){ + if(chan->ops->window_size == NULL) return(0); + return(chan->ops->window_size(chan->fd, chan->data, + rows_out, cols_out)); + } + } + return(0); +} + +void free_one_chan(struct chan *chan) +{ + list_del(&chan->list); + if(chan->ops->free != NULL) + (*chan->ops->free)(chan->data); + free_irq_by_fd(chan->fd); + if(chan->primary && chan->output) ignore_sigio_fd(chan->fd); + kfree(chan); +} + +void free_chan(struct list_head *chans) +{ + struct list_head *ele, *next; + struct chan *chan; + + list_for_each_safe(ele, next, chans){ + chan = list_entry(ele, struct chan, list); + free_one_chan(chan); + } +} + +static int one_chan_config_string(struct chan *chan, char *str, int size, + char **error_out) +{ + int n = 0; + + if(chan == NULL){ + CONFIG_CHUNK(str, size, n, "none", 1); + return(n); + } + + CONFIG_CHUNK(str, size, n, chan->ops->type, 0); + + if(chan->dev == NULL){ + CONFIG_CHUNK(str, size, n, "", 1); + return(n); + } + + CONFIG_CHUNK(str, size, n, ":", 0); + CONFIG_CHUNK(str, size, n, chan->dev, 0); + + return(n); +} + +static int chan_pair_config_string(struct chan *in, struct chan *out, + char *str, int size, char **error_out) +{ + int n; + + n = one_chan_config_string(in, str, size, error_out); + str += n; + size -= n; + + if(in == out){ + CONFIG_CHUNK(str, size, n, "", 1); + return(n); + } + + CONFIG_CHUNK(str, size, n, ",", 1); + n = one_chan_config_string(out, str, size, error_out); + str += n; + size -= n; + CONFIG_CHUNK(str, size, n, "", 1); + + return(n); +} + +int chan_config_string(struct list_head *chans, char *str, int size, + char **error_out) +{ + struct list_head *ele; + struct chan *chan, *in = NULL, *out = NULL; + + list_for_each(ele, chans){ + chan = list_entry(ele, struct chan, list); + if(!chan->primary) + continue; + if(chan->input) + in = chan; + if(chan->output) + out = chan; + } + + return(chan_pair_config_string(in, out, str, size, error_out)); +} + +struct chan_type { + char *key; + struct chan_ops *ops; +}; + +struct chan_type chan_table[] = { + { "fd", &fd_ops }, + +#ifdef CONFIG_NULL_CHAN + { "null", &null_ops }, +#else + { "null", ¬_configged_ops }, +#endif + +#ifdef CONFIG_PORT_CHAN + { "port", &port_ops }, +#else + { "port", ¬_configged_ops }, +#endif + +#ifdef CONFIG_PTY_CHAN + { "pty", &pty_ops }, + { "pts", &pts_ops }, +#else + { "pty", ¬_configged_ops }, + { "pts", ¬_configged_ops }, +#endif + +#ifdef CONFIG_TTY_CHAN + { "tty", &tty_ops }, +#else + { "tty", ¬_configged_ops }, +#endif + +#ifdef CONFIG_XTERM_CHAN + { "xterm", &xterm_ops }, +#else + { "xterm", ¬_configged_ops }, +#endif +}; + +static struct chan *parse_chan(char *str, int pri, int device, + struct chan_opts *opts) +{ + struct chan_type *entry; + struct chan_ops *ops; + struct chan *chan; + void *data; + int i; + + ops = NULL; + data = NULL; + for(i = 0; i < sizeof(chan_table)/sizeof(chan_table[0]); i++){ + entry = &chan_table[i]; + if(!strncmp(str, entry->key, strlen(entry->key))){ + ops = entry->ops; + str += strlen(entry->key); + break; + } + } + if(ops == NULL){ + printk(KERN_ERR "parse_chan couldn't parse \"%s\"\n", + str); + return(NULL); + } + if(ops->init == NULL) return(NULL); + data = (*ops->init)(str, device, opts); + if(data == NULL) return(NULL); + + chan = kmalloc(sizeof(*chan), GFP_KERNEL); + if(chan == NULL) return(NULL); + *chan = ((struct chan) { .list = LIST_HEAD_INIT(chan->list), + .primary = 1, + .input = 0, + .output = 0, + .opened = 0, + .fd = -1, + .pri = pri, + .ops = ops, + .data = data }); + return(chan); +} + +int parse_chan_pair(char *str, struct list_head *chans, int pri, int device, + struct chan_opts *opts) +{ + struct chan *new, *chan; + char *in, *out; + + if(!list_empty(chans)){ + chan = list_entry(chans->next, struct chan, list); + if(chan->pri >= pri) return(0); + free_chan(chans); + INIT_LIST_HEAD(chans); + } + + out = strchr(str, ','); + if(out != NULL){ + in = str; + *out = '\0'; + out++; + new = parse_chan(in, pri, device, opts); + if(new == NULL) return(-1); + new->input = 1; + list_add(&new->list, chans); + + new = parse_chan(out, pri, device, opts); + if(new == NULL) return(-1); + list_add(&new->list, chans); + new->output = 1; + } + else { + new = parse_chan(str, pri, device, opts); + if(new == NULL) return(-1); + list_add(&new->list, chans); + new->input = 1; + new->output = 1; + } + return(0); +} + +int chan_out_fd(struct list_head *chans) +{ + struct list_head *ele; + struct chan *chan; + + list_for_each(ele, chans){ + chan = list_entry(ele, struct chan, list); + if(chan->primary && chan->output) + return(chan->fd); + } + return(-1); +} + +void chan_interrupt(struct list_head *chans, struct work_struct *task, + struct tty_struct *tty, int irq) +{ + struct list_head *ele, *next; + struct chan *chan; + int err; + char c; + + list_for_each_safe(ele, next, chans){ + chan = list_entry(ele, struct chan, list); + if(!chan->input || (chan->ops->read == NULL)) continue; + do { + if((tty != NULL) && + (tty->flip.count >= TTY_FLIPBUF_SIZE)){ + schedule_work(task); + goto out; + } + err = chan->ops->read(chan->fd, &c, chan->data); + if(err > 0) + tty_receive_char(tty, c); + } while(err > 0); + + if(err == 0) reactivate_fd(chan->fd, irq); + if(err == -EIO){ + if(chan->primary){ + if(tty != NULL) + tty_hangup(tty); + line_disable(tty, irq); + close_chan(chans); + free_chan(chans); + return; + } + else { + if(chan->ops->close != NULL) + chan->ops->close(chan->fd, chan->data); + free_one_chan(chan); + } + } + } + out: + if(tty) tty_flip_buffer_push(tty); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/chan_user.c b/arch/um/drivers/chan_user.c new file mode 100644 index 000000000000..583b8e137c33 --- /dev/null +++ b/arch/um/drivers/chan_user.c @@ -0,0 +1,210 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include <unistd.h> +#include <stdlib.h> +#include <errno.h> +#include <termios.h> +#include <string.h> +#include <signal.h> +#include <sys/stat.h> +#include <sys/ioctl.h> +#include <sys/socket.h> +#include "kern_util.h" +#include "user_util.h" +#include "chan_user.h" +#include "user.h" +#include "helper.h" +#include "os.h" +#include "choose-mode.h" +#include "mode.h" + +int generic_console_write(int fd, const char *buf, int n, void *unused) +{ + struct termios save, new; + int err; + + if(isatty(fd)){ + CATCH_EINTR(err = tcgetattr(fd, &save)); + if (err) + goto error; + new = save; + /* The terminal becomes a bit less raw, to handle \n also as + * "Carriage Return", not only as "New Line". Otherwise, the new + * line won't start at the first column.*/ + new.c_oflag |= OPOST; + CATCH_EINTR(err = tcsetattr(fd, TCSAFLUSH, &new)); + if (err) + goto error; + } + err = generic_write(fd, buf, n, NULL); + /* Restore raw mode, in any case; we *must* ignore any error apart + * EINTR, except for debug.*/ + if(isatty(fd)) + CATCH_EINTR(tcsetattr(fd, TCSAFLUSH, &save)); + return(err); +error: + return(-errno); +} + +/* + * UML SIGWINCH handling + * + * The point of this is to handle SIGWINCH on consoles which have host ttys and + * relay them inside UML to whatever might be running on the console and cares + * about the window size (since SIGWINCH notifies about terminal size changes). + * + * So, we have a separate thread for each host tty attached to a UML device + * (side-issue - I'm annoyed that one thread can't have multiple controlling + * ttys for purposed of handling SIGWINCH, but I imagine there are other reasons + * that doesn't make any sense). + * + * SIGWINCH can't be received synchronously, so you have to set up to receive it + * as a signal. That being the case, if you are going to wait for it, it is + * convenient to sit in a pause() and wait for the signal to bounce you out of + * it (see below for how we make sure to exit only on SIGWINCH). + */ + +static void winch_handler(int sig) +{ +} + +struct winch_data { + int pty_fd; + int pipe_fd; + int close_me; +}; + +static int winch_thread(void *arg) +{ + struct winch_data *data = arg; + sigset_t sigs; + int pty_fd, pipe_fd; + int count, err; + char c = 1; + + os_close_file(data->close_me); + pty_fd = data->pty_fd; + pipe_fd = data->pipe_fd; + count = os_write_file(pipe_fd, &c, sizeof(c)); + if(count != sizeof(c)) + printk("winch_thread : failed to write synchronization " + "byte, err = %d\n", -count); + + /* We are not using SIG_IGN on purpose, so don't fix it as I thought to + * do! If using SIG_IGN, the pause() call below would not stop on + * SIGWINCH. */ + + signal(SIGWINCH, winch_handler); + sigfillset(&sigs); + sigdelset(&sigs, SIGWINCH); + /* Block anything else than SIGWINCH. */ + if(sigprocmask(SIG_SETMASK, &sigs, NULL) < 0){ + printk("winch_thread : sigprocmask failed, errno = %d\n", + errno); + exit(1); + } + + if(setsid() < 0){ + printk("winch_thread : setsid failed, errno = %d\n", errno); + exit(1); + } + + err = os_new_tty_pgrp(pty_fd, os_getpid()); + if(err < 0){ + printk("winch_thread : new_tty_pgrp failed, err = %d\n", -err); + exit(1); + } + + /* These are synchronization calls between various UML threads on the + * host - since they are not different kernel threads, we cannot use + * kernel semaphores. We don't use SysV semaphores because they are + * persistant. */ + count = os_read_file(pipe_fd, &c, sizeof(c)); + if(count != sizeof(c)) + printk("winch_thread : failed to read synchronization byte, " + "err = %d\n", -count); + + while(1){ + /* This will be interrupted by SIGWINCH only, since other signals + * are blocked.*/ + pause(); + + count = os_write_file(pipe_fd, &c, sizeof(c)); + if(count != sizeof(c)) + printk("winch_thread : write failed, err = %d\n", + -count); + } +} + +static int winch_tramp(int fd, struct tty_struct *tty, int *fd_out) +{ + struct winch_data data; + unsigned long stack; + int fds[2], pid, n, err; + char c; + + err = os_pipe(fds, 1, 1); + if(err < 0){ + printk("winch_tramp : os_pipe failed, err = %d\n", -err); + return(err); + } + + data = ((struct winch_data) { .pty_fd = fd, + .pipe_fd = fds[1], + .close_me = fds[0] } ); + pid = run_helper_thread(winch_thread, &data, 0, &stack, 0); + if(pid < 0){ + printk("fork of winch_thread failed - errno = %d\n", errno); + return(pid); + } + + os_close_file(fds[1]); + *fd_out = fds[0]; + n = os_read_file(fds[0], &c, sizeof(c)); + if(n != sizeof(c)){ + printk("winch_tramp : failed to read synchronization byte\n"); + printk("read failed, err = %d\n", -n); + printk("fd %d will not support SIGWINCH\n", fd); + *fd_out = -1; + } + return(pid); +} + +void register_winch(int fd, struct tty_struct *tty) +{ + int pid, thread, thread_fd; + int count; + char c = 1; + + if(!isatty(fd)) + return; + + pid = tcgetpgrp(fd); + if(!CHOOSE_MODE_PROC(is_tracer_winch, is_skas_winch, pid, fd, + tty) && (pid == -1)){ + thread = winch_tramp(fd, tty, &thread_fd); + if(fd != -1){ + register_winch_irq(thread_fd, fd, thread, tty); + + count = os_write_file(thread_fd, &c, sizeof(c)); + if(count != sizeof(c)) + printk("register_winch : failed to write " + "synchronization byte, err = %d\n", + -count); + } + } +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/cow.h b/arch/um/drivers/cow.h new file mode 100644 index 000000000000..4fcbe8b1b77e --- /dev/null +++ b/arch/um/drivers/cow.h @@ -0,0 +1,42 @@ +#ifndef __COW_H__ +#define __COW_H__ + +#include <asm/types.h> + +#if __BYTE_ORDER == __BIG_ENDIAN +# define ntohll(x) (x) +# define htonll(x) (x) +#elif __BYTE_ORDER == __LITTLE_ENDIAN +# define ntohll(x) bswap_64(x) +# define htonll(x) bswap_64(x) +#else +#error "__BYTE_ORDER not defined" +#endif + +extern int init_cow_file(int fd, char *cow_file, char *backing_file, + int sectorsize, int alignment, int *bitmap_offset_out, + unsigned long *bitmap_len_out, int *data_offset_out); + +extern int file_reader(__u64 offset, char *buf, int len, void *arg); +extern int read_cow_header(int (*reader)(__u64, char *, int, void *), + void *arg, __u32 *version_out, + char **backing_file_out, time_t *mtime_out, + unsigned long long *size_out, int *sectorsize_out, + __u32 *align_out, int *bitmap_offset_out); + +extern int write_cow_header(char *cow_file, int fd, char *backing_file, + int sectorsize, int alignment, + unsigned long long *size); + +extern void cow_sizes(int version, __u64 size, int sectorsize, int align, + int bitmap_offset, unsigned long *bitmap_len_out, + int *data_offset_out); + +#endif + +/* + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/cow_sys.h b/arch/um/drivers/cow_sys.h new file mode 100644 index 000000000000..c83fc5d68936 --- /dev/null +++ b/arch/um/drivers/cow_sys.h @@ -0,0 +1,48 @@ +#ifndef __COW_SYS_H__ +#define __COW_SYS_H__ + +#include "kern_util.h" +#include "user_util.h" +#include "os.h" +#include "user.h" + +static inline void *cow_malloc(int size) +{ + return(um_kmalloc(size)); +} + +static inline void cow_free(void *ptr) +{ + kfree(ptr); +} + +#define cow_printf printk + +static inline char *cow_strdup(char *str) +{ + return(uml_strdup(str)); +} + +static inline int cow_seek_file(int fd, unsigned long long offset) +{ + return(os_seek_file(fd, offset)); +} + +static inline int cow_file_size(char *file, unsigned long long *size_out) +{ + return(os_file_size(file, size_out)); +} + +static inline int cow_write_file(int fd, char *buf, int size) +{ + return(os_write_file(fd, buf, size)); +} + +#endif + +/* + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/cow_user.c b/arch/um/drivers/cow_user.c new file mode 100644 index 000000000000..a8ce6fc3ef26 --- /dev/null +++ b/arch/um/drivers/cow_user.c @@ -0,0 +1,378 @@ +#include <stddef.h> +#include <string.h> +#include <errno.h> +/* _XOPEN_SOURCE is needed for pread, but we define _GNU_SOURCE, which defines + * that. + */ +#include <unistd.h> +#include <byteswap.h> +#include <sys/time.h> +#include <sys/param.h> +#include <sys/user.h> +#include <netinet/in.h> + +#include "os.h" + +#include "cow.h" +#include "cow_sys.h" + +#define PATH_LEN_V1 256 + +struct cow_header_v1 { + int magic; + int version; + char backing_file[PATH_LEN_V1]; + time_t mtime; + __u64 size; + int sectorsize; +}; + +#define PATH_LEN_V2 MAXPATHLEN + +struct cow_header_v2 { + __u32 magic; + __u32 version; + char backing_file[PATH_LEN_V2]; + time_t mtime; + __u64 size; + int sectorsize; +}; + +/* Define PATH_LEN_V3 as the usual value of MAXPATHLEN, just hard-code it in + * case other systems have different values for MAXPATHLEN + */ +#define PATH_LEN_V3 4096 + +/* Changes from V2 - + * PATH_LEN_V3 as described above + * Explicitly specify field bit lengths for systems with different + * lengths for the usual C types. Not sure whether char or + * time_t should be changed, this can be changed later without + * breaking compatibility + * Add alignment field so that different alignments can be used for the + * bitmap and data + * Add cow_format field to allow for the possibility of different ways + * of specifying the COW blocks. For now, the only value is 0, + * for the traditional COW bitmap. + * Move the backing_file field to the end of the header. This allows + * for the possibility of expanding it into the padding required + * by the bitmap alignment. + * The bitmap and data portions of the file will be aligned as specified + * by the alignment field. This is to allow COW files to be + * put on devices with restrictions on access alignments, such as + * /dev/raw, with a 512 byte alignment restriction. This also + * allows the data to be more aligned more strictly than on + * sector boundaries. This is needed for ubd-mmap, which needs + * the data to be page aligned. + * Fixed (finally!) the rounding bug + */ + +struct cow_header_v3 { + __u32 magic; + __u32 version; + __u32 mtime; + __u64 size; + __u32 sectorsize; + __u32 alignment; + __u32 cow_format; + char backing_file[PATH_LEN_V3]; +}; + +/* COW format definitions - for now, we have only the usual COW bitmap */ +#define COW_BITMAP 0 + +union cow_header { + struct cow_header_v1 v1; + struct cow_header_v2 v2; + struct cow_header_v3 v3; +}; + +#define COW_MAGIC 0x4f4f4f4d /* MOOO */ +#define COW_VERSION 3 + +#define DIV_ROUND(x, len) (((x) + (len) - 1) / (len)) +#define ROUND_UP(x, align) DIV_ROUND(x, align) * (align) + +void cow_sizes(int version, __u64 size, int sectorsize, int align, + int bitmap_offset, unsigned long *bitmap_len_out, + int *data_offset_out) +{ + if(version < 3){ + *bitmap_len_out = (size + sectorsize - 1) / (8 * sectorsize); + + *data_offset_out = bitmap_offset + *bitmap_len_out; + *data_offset_out = (*data_offset_out + sectorsize - 1) / + sectorsize; + *data_offset_out *= sectorsize; + } + else { + *bitmap_len_out = DIV_ROUND(size, sectorsize); + *bitmap_len_out = DIV_ROUND(*bitmap_len_out, 8); + + *data_offset_out = bitmap_offset + *bitmap_len_out; + *data_offset_out = ROUND_UP(*data_offset_out, align); + } +} + +static int absolutize(char *to, int size, char *from) +{ + char save_cwd[256], *slash; + int remaining; + + if(getcwd(save_cwd, sizeof(save_cwd)) == NULL) { + cow_printf("absolutize : unable to get cwd - errno = %d\n", + errno); + return(-1); + } + slash = strrchr(from, '/'); + if(slash != NULL){ + *slash = '\0'; + if(chdir(from)){ + *slash = '/'; + cow_printf("absolutize : Can't cd to '%s' - " + "errno = %d\n", from, errno); + return(-1); + } + *slash = '/'; + if(getcwd(to, size) == NULL){ + cow_printf("absolutize : unable to get cwd of '%s' - " + "errno = %d\n", from, errno); + return(-1); + } + remaining = size - strlen(to); + if(strlen(slash) + 1 > remaining){ + cow_printf("absolutize : unable to fit '%s' into %d " + "chars\n", from, size); + return(-1); + } + strcat(to, slash); + } + else { + if(strlen(save_cwd) + 1 + strlen(from) + 1 > size){ + cow_printf("absolutize : unable to fit '%s' into %d " + "chars\n", from, size); + return(-1); + } + strcpy(to, save_cwd); + strcat(to, "/"); + strcat(to, from); + } + chdir(save_cwd); + return(0); +} + +int write_cow_header(char *cow_file, int fd, char *backing_file, + int sectorsize, int alignment, unsigned long long *size) +{ + struct cow_header_v3 *header; + unsigned long modtime; + int err; + + err = cow_seek_file(fd, 0); + if(err < 0){ + cow_printf("write_cow_header - lseek failed, err = %d\n", -err); + goto out; + } + + err = -ENOMEM; + header = cow_malloc(sizeof(*header)); + if(header == NULL){ + cow_printf("Failed to allocate COW V3 header\n"); + goto out; + } + header->magic = htonl(COW_MAGIC); + header->version = htonl(COW_VERSION); + + err = -EINVAL; + if(strlen(backing_file) > sizeof(header->backing_file) - 1){ + cow_printf("Backing file name \"%s\" is too long - names are " + "limited to %d characters\n", backing_file, + sizeof(header->backing_file) - 1); + goto out_free; + } + + if(absolutize(header->backing_file, sizeof(header->backing_file), + backing_file)) + goto out_free; + + err = os_file_modtime(header->backing_file, &modtime); + if(err < 0){ + cow_printf("Backing file '%s' mtime request failed, " + "err = %d\n", header->backing_file, -err); + goto out_free; + } + + err = cow_file_size(header->backing_file, size); + if(err < 0){ + cow_printf("Couldn't get size of backing file '%s', " + "err = %d\n", header->backing_file, -err); + goto out_free; + } + + header->mtime = htonl(modtime); + header->size = htonll(*size); + header->sectorsize = htonl(sectorsize); + header->alignment = htonl(alignment); + header->cow_format = COW_BITMAP; + + err = os_write_file(fd, header, sizeof(*header)); + if(err != sizeof(*header)){ + cow_printf("Write of header to new COW file '%s' failed, " + "err = %d\n", cow_file, -err); + goto out_free; + } + err = 0; + out_free: + cow_free(header); + out: + return(err); +} + +int file_reader(__u64 offset, char *buf, int len, void *arg) +{ + int fd = *((int *) arg); + + return(pread(fd, buf, len, offset)); +} + +/* XXX Need to sanity-check the values read from the header */ + +int read_cow_header(int (*reader)(__u64, char *, int, void *), void *arg, + __u32 *version_out, char **backing_file_out, + time_t *mtime_out, unsigned long long *size_out, + int *sectorsize_out, __u32 *align_out, + int *bitmap_offset_out) +{ + union cow_header *header; + char *file; + int err, n; + unsigned long version, magic; + + header = cow_malloc(sizeof(*header)); + if(header == NULL){ + cow_printf("read_cow_header - Failed to allocate header\n"); + return(-ENOMEM); + } + err = -EINVAL; + n = (*reader)(0, (char *) header, sizeof(*header), arg); + if(n < offsetof(typeof(header->v1), backing_file)){ + cow_printf("read_cow_header - short header\n"); + goto out; + } + + magic = header->v1.magic; + if(magic == COW_MAGIC) { + version = header->v1.version; + } + else if(magic == ntohl(COW_MAGIC)){ + version = ntohl(header->v1.version); + } + /* No error printed because the non-COW case comes through here */ + else goto out; + + *version_out = version; + + if(version == 1){ + if(n < sizeof(header->v1)){ + cow_printf("read_cow_header - failed to read V1 " + "header\n"); + goto out; + } + *mtime_out = header->v1.mtime; + *size_out = header->v1.size; + *sectorsize_out = header->v1.sectorsize; + *bitmap_offset_out = sizeof(header->v1); + *align_out = *sectorsize_out; + file = header->v1.backing_file; + } + else if(version == 2){ + if(n < sizeof(header->v2)){ + cow_printf("read_cow_header - failed to read V2 " + "header\n"); + goto out; + } + *mtime_out = ntohl(header->v2.mtime); + *size_out = ntohll(header->v2.size); + *sectorsize_out = ntohl(header->v2.sectorsize); + *bitmap_offset_out = sizeof(header->v2); + *align_out = *sectorsize_out; + file = header->v2.backing_file; + } + else if(version == 3){ + if(n < sizeof(header->v3)){ + cow_printf("read_cow_header - failed to read V2 " + "header\n"); + goto out; + } + *mtime_out = ntohl(header->v3.mtime); + *size_out = ntohll(header->v3.size); + *sectorsize_out = ntohl(header->v3.sectorsize); + *align_out = ntohl(header->v3.alignment); + *bitmap_offset_out = ROUND_UP(sizeof(header->v3), *align_out); + file = header->v3.backing_file; + } + else { + cow_printf("read_cow_header - invalid COW version\n"); + goto out; + } + err = -ENOMEM; + *backing_file_out = cow_strdup(file); + if(*backing_file_out == NULL){ + cow_printf("read_cow_header - failed to allocate backing " + "file\n"); + goto out; + } + err = 0; + out: + cow_free(header); + return(err); +} + +int init_cow_file(int fd, char *cow_file, char *backing_file, int sectorsize, + int alignment, int *bitmap_offset_out, + unsigned long *bitmap_len_out, int *data_offset_out) +{ + unsigned long long size, offset; + char zero = 0; + int err; + + err = write_cow_header(cow_file, fd, backing_file, sectorsize, + alignment, &size); + if(err) + goto out; + + *bitmap_offset_out = ROUND_UP(sizeof(struct cow_header_v3), alignment); + cow_sizes(COW_VERSION, size, sectorsize, alignment, *bitmap_offset_out, + bitmap_len_out, data_offset_out); + + offset = *data_offset_out + size - sizeof(zero); + err = cow_seek_file(fd, offset); + if(err < 0){ + cow_printf("cow bitmap lseek failed : err = %d\n", -err); + goto out; + } + + /* does not really matter how much we write it is just to set EOF + * this also sets the entire COW bitmap + * to zero without having to allocate it + */ + err = cow_write_file(fd, &zero, sizeof(zero)); + if(err != sizeof(zero)){ + cow_printf("Write of bitmap to new COW file '%s' failed, " + "err = %d\n", cow_file, -err); + err = -EINVAL; + goto out; + } + + return(0); + + out: + return(err); +} + +/* + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/daemon.h b/arch/um/drivers/daemon.h new file mode 100644 index 000000000000..7326c42f7ef9 --- /dev/null +++ b/arch/um/drivers/daemon.h @@ -0,0 +1,35 @@ +/* + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "net_user.h" + +#define SWITCH_VERSION 3 + +struct daemon_data { + char *sock_type; + char *ctl_sock; + void *ctl_addr; + void *data_addr; + void *local_addr; + int fd; + int control; + void *dev; +}; + +extern struct net_user_info daemon_user_info; + +extern int daemon_user_write(int fd, void *buf, int len, + struct daemon_data *pri); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/daemon_kern.c b/arch/um/drivers/daemon_kern.c new file mode 100644 index 000000000000..30d285b266af --- /dev/null +++ b/arch/um/drivers/daemon_kern.c @@ -0,0 +1,108 @@ +/* + * Copyright (C) 2001 Lennert Buytenhek (buytenh@gnu.org) and + * James Leu (jleu@mindspring.net). + * Copyright (C) 2001 by various other people who didn't put their name here. + * Licensed under the GPL. + */ + +#include "linux/kernel.h" +#include "linux/init.h" +#include "linux/netdevice.h" +#include "linux/etherdevice.h" +#include "net_kern.h" +#include "net_user.h" +#include "daemon.h" + +struct daemon_init { + char *sock_type; + char *ctl_sock; +}; + +void daemon_init(struct net_device *dev, void *data) +{ + struct uml_net_private *pri; + struct daemon_data *dpri; + struct daemon_init *init = data; + + pri = dev->priv; + dpri = (struct daemon_data *) pri->user; + dpri->sock_type = init->sock_type; + dpri->ctl_sock = init->ctl_sock; + dpri->fd = -1; + dpri->control = -1; + dpri->dev = dev; + + printk("daemon backend (uml_switch version %d) - %s:%s", + SWITCH_VERSION, dpri->sock_type, dpri->ctl_sock); + printk("\n"); +} + +static int daemon_read(int fd, struct sk_buff **skb, + struct uml_net_private *lp) +{ + *skb = ether_adjust_skb(*skb, ETH_HEADER_OTHER); + if(*skb == NULL) return(-ENOMEM); + return(net_recvfrom(fd, (*skb)->mac.raw, + (*skb)->dev->mtu + ETH_HEADER_OTHER)); +} + +static int daemon_write(int fd, struct sk_buff **skb, + struct uml_net_private *lp) +{ + return(daemon_user_write(fd, (*skb)->data, (*skb)->len, + (struct daemon_data *) &lp->user)); +} + +static struct net_kern_info daemon_kern_info = { + .init = daemon_init, + .protocol = eth_protocol, + .read = daemon_read, + .write = daemon_write, +}; + +int daemon_setup(char *str, char **mac_out, void *data) +{ + struct daemon_init *init = data; + char *remain; + + *init = ((struct daemon_init) + { .sock_type = "unix", + .ctl_sock = "/tmp/uml.ctl" }); + + remain = split_if_spec(str, mac_out, &init->sock_type, &init->ctl_sock, + NULL); + if(remain != NULL) + printk(KERN_WARNING "daemon_setup : Ignoring data socket " + "specification\n"); + + return(1); +} + +static struct transport daemon_transport = { + .list = LIST_HEAD_INIT(daemon_transport.list), + .name = "daemon", + .setup = daemon_setup, + .user = &daemon_user_info, + .kern = &daemon_kern_info, + .private_size = sizeof(struct daemon_data), + .setup_size = sizeof(struct daemon_init), +}; + +static int register_daemon(void) +{ + register_transport(&daemon_transport); + return(1); +} + +__initcall(register_daemon); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/daemon_user.c b/arch/um/drivers/daemon_user.c new file mode 100644 index 000000000000..cf15b4a8b517 --- /dev/null +++ b/arch/um/drivers/daemon_user.c @@ -0,0 +1,197 @@ +/* + * Copyright (C) 2001 Lennert Buytenhek (buytenh@gnu.org) and + * James Leu (jleu@mindspring.net). + * Copyright (C) 2001 by various other people who didn't put their name here. + * Licensed under the GPL. + */ + +#include <errno.h> +#include <unistd.h> +#include <stdint.h> +#include <sys/socket.h> +#include <sys/un.h> +#include <sys/time.h> +#include "net_user.h" +#include "daemon.h" +#include "kern_util.h" +#include "user_util.h" +#include "user.h" +#include "os.h" + +#define MAX_PACKET (ETH_MAX_PACKET + ETH_HEADER_OTHER) + +enum request_type { REQ_NEW_CONTROL }; + +#define SWITCH_MAGIC 0xfeedface + +struct request_v3 { + uint32_t magic; + uint32_t version; + enum request_type type; + struct sockaddr_un sock; +}; + +static struct sockaddr_un *new_addr(void *name, int len) +{ + struct sockaddr_un *sun; + + sun = um_kmalloc(sizeof(struct sockaddr_un)); + if(sun == NULL){ + printk("new_addr: allocation of sockaddr_un failed\n"); + return(NULL); + } + sun->sun_family = AF_UNIX; + memcpy(sun->sun_path, name, len); + return(sun); +} + +static int connect_to_switch(struct daemon_data *pri) +{ + struct sockaddr_un *ctl_addr = pri->ctl_addr; + struct sockaddr_un *local_addr = pri->local_addr; + struct sockaddr_un *sun; + struct request_v3 req; + int fd, n, err; + + pri->control = socket(AF_UNIX, SOCK_STREAM, 0); + if(pri->control < 0){ + printk("daemon_open : control socket failed, errno = %d\n", + errno); + return(-errno); + } + + if(connect(pri->control, (struct sockaddr *) ctl_addr, + sizeof(*ctl_addr)) < 0){ + printk("daemon_open : control connect failed, errno = %d\n", + errno); + err = -errno; + goto out; + } + + fd = socket(AF_UNIX, SOCK_DGRAM, 0); + if(fd < 0){ + printk("daemon_open : data socket failed, errno = %d\n", + errno); + err = -errno; + goto out; + } + if(bind(fd, (struct sockaddr *) local_addr, sizeof(*local_addr)) < 0){ + printk("daemon_open : data bind failed, errno = %d\n", + errno); + err = -errno; + goto out_close; + } + + sun = um_kmalloc(sizeof(struct sockaddr_un)); + if(sun == NULL){ + printk("new_addr: allocation of sockaddr_un failed\n"); + err = -ENOMEM; + goto out_close; + } + + req.magic = SWITCH_MAGIC; + req.version = SWITCH_VERSION; + req.type = REQ_NEW_CONTROL; + req.sock = *local_addr; + n = os_write_file(pri->control, &req, sizeof(req)); + if(n != sizeof(req)){ + printk("daemon_open : control setup request failed, err = %d\n", + -n); + err = -ENOTCONN; + goto out; + } + + n = os_read_file(pri->control, sun, sizeof(*sun)); + if(n != sizeof(*sun)){ + printk("daemon_open : read of data socket failed, err = %d\n", + -n); + err = -ENOTCONN; + goto out_close; + } + + pri->data_addr = sun; + return(fd); + + out_close: + os_close_file(fd); + out: + os_close_file(pri->control); + return(err); +} + +static void daemon_user_init(void *data, void *dev) +{ + struct daemon_data *pri = data; + struct timeval tv; + struct { + char zero; + int pid; + int usecs; + } name; + + if(!strcmp(pri->sock_type, "unix")) + pri->ctl_addr = new_addr(pri->ctl_sock, + strlen(pri->ctl_sock) + 1); + name.zero = 0; + name.pid = os_getpid(); + gettimeofday(&tv, NULL); + name.usecs = tv.tv_usec; + pri->local_addr = new_addr(&name, sizeof(name)); + pri->dev = dev; + pri->fd = connect_to_switch(pri); + if(pri->fd < 0){ + kfree(pri->local_addr); + pri->local_addr = NULL; + } +} + +static int daemon_open(void *data) +{ + struct daemon_data *pri = data; + return(pri->fd); +} + +static void daemon_remove(void *data) +{ + struct daemon_data *pri = data; + + os_close_file(pri->fd); + os_close_file(pri->control); + if(pri->data_addr != NULL) kfree(pri->data_addr); + if(pri->ctl_addr != NULL) kfree(pri->ctl_addr); + if(pri->local_addr != NULL) kfree(pri->local_addr); +} + +int daemon_user_write(int fd, void *buf, int len, struct daemon_data *pri) +{ + struct sockaddr_un *data_addr = pri->data_addr; + + return(net_sendto(fd, buf, len, data_addr, sizeof(*data_addr))); +} + +static int daemon_set_mtu(int mtu, void *data) +{ + return(mtu); +} + +struct net_user_info daemon_user_info = { + .init = daemon_user_init, + .open = daemon_open, + .close = NULL, + .remove = daemon_remove, + .set_mtu = daemon_set_mtu, + .add_address = NULL, + .delete_address = NULL, + .max_packet = MAX_PACKET - ETH_HEADER_OTHER +}; + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/fd.c b/arch/um/drivers/fd.c new file mode 100644 index 000000000000..f0b888f66e05 --- /dev/null +++ b/arch/um/drivers/fd.c @@ -0,0 +1,108 @@ +/* + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <termios.h> +#include <errno.h> +#include "user.h" +#include "user_util.h" +#include "chan_user.h" + +struct fd_chan { + int fd; + int raw; + struct termios tt; + char str[sizeof("1234567890\0")]; +}; + +static void *fd_init(char *str, int device, struct chan_opts *opts) +{ + struct fd_chan *data; + char *end; + int n; + + if(*str != ':'){ + printk("fd_init : channel type 'fd' must specify a file " + "descriptor\n"); + return(NULL); + } + str++; + n = strtoul(str, &end, 0); + if((*end != '\0') || (end == str)){ + printk("fd_init : couldn't parse file descriptor '%s'\n", str); + return(NULL); + } + data = um_kmalloc(sizeof(*data)); + if(data == NULL) return(NULL); + *data = ((struct fd_chan) { .fd = n, + .raw = opts->raw }); + return(data); +} + +static int fd_open(int input, int output, int primary, void *d, char **dev_out) +{ + struct fd_chan *data = d; + int err; + + if(data->raw && isatty(data->fd)){ + CATCH_EINTR(err = tcgetattr(data->fd, &data->tt)); + if(err) + return(err); + + err = raw(data->fd); + if(err) + return(err); + } + sprintf(data->str, "%d", data->fd); + *dev_out = data->str; + return(data->fd); +} + +static void fd_close(int fd, void *d) +{ + struct fd_chan *data = d; + int err; + + if(data->raw && isatty(fd)){ + CATCH_EINTR(err = tcsetattr(fd, TCSAFLUSH, &data->tt)); + if(err) + printk("Failed to restore terminal state - " + "errno = %d\n", -err); + data->raw = 0; + } +} + +static int fd_console_write(int fd, const char *buf, int n, void *d) +{ + struct fd_chan *data = d; + + return(generic_console_write(fd, buf, n, &data->tt)); +} + +struct chan_ops fd_ops = { + .type = "fd", + .init = fd_init, + .open = fd_open, + .close = fd_close, + .read = generic_read, + .write = generic_write, + .console_write = fd_console_write, + .window_size = generic_window_size, + .free = generic_free, + .winch = 1, +}; + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/harddog_kern.c b/arch/um/drivers/harddog_kern.c new file mode 100644 index 000000000000..147ec19f6bb9 --- /dev/null +++ b/arch/um/drivers/harddog_kern.c @@ -0,0 +1,190 @@ +/* UML hardware watchdog, shamelessly stolen from: + * + * SoftDog 0.05: A Software Watchdog Device + * + * (c) Copyright 1996 Alan Cox <alan@redhat.com>, All Rights Reserved. + * http://www.redhat.com + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + * Neither Alan Cox nor CymruNet Ltd. admit liability nor provide + * warranty for any of this software. This material is provided + * "AS-IS" and at no charge. + * + * (c) Copyright 1995 Alan Cox <alan@lxorguk.ukuu.org.uk> + * + * Software only watchdog driver. Unlike its big brother the WDT501P + * driver this won't always recover a failed machine. + * + * 03/96: Angelo Haritsis <ah@doc.ic.ac.uk> : + * Modularised. + * Added soft_margin; use upon insmod to change the timer delay. + * NB: uses same minor as wdt (WATCHDOG_MINOR); we could use separate + * minors. + * + * 19980911 Alan Cox + * Made SMP safe for 2.3.x + * + * 20011127 Joel Becker (jlbec@evilplan.org> + * Added soft_noboot; Allows testing the softdog trigger without + * requiring a recompile. + * Added WDIOC_GETTIMEOUT and WDIOC_SETTIMOUT. + */ + +#include <linux/module.h> +#include <linux/config.h> +#include <linux/types.h> +#include <linux/kernel.h> +#include <linux/fs.h> +#include <linux/mm.h> +#include <linux/miscdevice.h> +#include <linux/watchdog.h> +#include <linux/reboot.h> +#include <linux/smp_lock.h> +#include <linux/init.h> +#include <asm/uaccess.h> +#include "helper.h" +#include "mconsole.h" + +MODULE_LICENSE("GPL"); + +/* Locked by the BKL in harddog_open and harddog_release */ +static int timer_alive; +static int harddog_in_fd = -1; +static int harddog_out_fd = -1; + +/* + * Allow only one person to hold it open + */ + +extern int start_watchdog(int *in_fd_ret, int *out_fd_ret, char *sock); + +static int harddog_open(struct inode *inode, struct file *file) +{ + int err; + char *sock = NULL; + + lock_kernel(); + if(timer_alive) + return -EBUSY; +#ifdef CONFIG_HARDDOG_NOWAYOUT + __module_get(THIS_MODULE); +#endif + +#ifdef CONFIG_MCONSOLE + sock = mconsole_notify_socket(); +#endif + err = start_watchdog(&harddog_in_fd, &harddog_out_fd, sock); + if(err) return(err); + + timer_alive = 1; + unlock_kernel(); + return nonseekable_open(inode, file); +} + +extern void stop_watchdog(int in_fd, int out_fd); + +static int harddog_release(struct inode *inode, struct file *file) +{ + /* + * Shut off the timer. + */ + lock_kernel(); + + stop_watchdog(harddog_in_fd, harddog_out_fd); + harddog_in_fd = -1; + harddog_out_fd = -1; + + timer_alive=0; + unlock_kernel(); + return 0; +} + +extern int ping_watchdog(int fd); + +static ssize_t harddog_write(struct file *file, const char *data, size_t len, + loff_t *ppos) +{ + /* + * Refresh the timer. + */ + if(len) + return(ping_watchdog(harddog_out_fd)); + return 0; +} + +static int harddog_ioctl(struct inode *inode, struct file *file, + unsigned int cmd, unsigned long arg) +{ + static struct watchdog_info ident = { + WDIOC_SETTIMEOUT, + 0, + "UML Hardware Watchdog" + }; + switch (cmd) { + default: + return -ENOTTY; + case WDIOC_GETSUPPORT: + if(copy_to_user((struct harddog_info *)arg, &ident, + sizeof(ident))) + return -EFAULT; + return 0; + case WDIOC_GETSTATUS: + case WDIOC_GETBOOTSTATUS: + return put_user(0,(int *)arg); + case WDIOC_KEEPALIVE: + return(ping_watchdog(harddog_out_fd)); + } +} + +static struct file_operations harddog_fops = { + .owner = THIS_MODULE, + .write = harddog_write, + .ioctl = harddog_ioctl, + .open = harddog_open, + .release = harddog_release, +}; + +static struct miscdevice harddog_miscdev = { + .minor = WATCHDOG_MINOR, + .name = "watchdog", + .fops = &harddog_fops, +}; + +static char banner[] __initdata = KERN_INFO "UML Watchdog Timer\n"; + +static int __init harddog_init(void) +{ + int ret; + + ret = misc_register(&harddog_miscdev); + + if (ret) + return ret; + + printk(banner); + + return(0); +} + +static void __exit harddog_exit(void) +{ + misc_deregister(&harddog_miscdev); +} + +module_init(harddog_init); +module_exit(harddog_exit); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/harddog_user.c b/arch/um/drivers/harddog_user.c new file mode 100644 index 000000000000..d934181b8d4c --- /dev/null +++ b/arch/um/drivers/harddog_user.c @@ -0,0 +1,143 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <unistd.h> +#include <errno.h> +#include "user_util.h" +#include "user.h" +#include "helper.h" +#include "mconsole.h" +#include "os.h" +#include "choose-mode.h" +#include "mode.h" + +struct dog_data { + int stdin; + int stdout; + int close_me[2]; +}; + +static void pre_exec(void *d) +{ + struct dog_data *data = d; + + dup2(data->stdin, 0); + dup2(data->stdout, 1); + dup2(data->stdout, 2); + os_close_file(data->stdin); + os_close_file(data->stdout); + os_close_file(data->close_me[0]); + os_close_file(data->close_me[1]); +} + +int start_watchdog(int *in_fd_ret, int *out_fd_ret, char *sock) +{ + struct dog_data data; + int in_fds[2], out_fds[2], pid, n, err; + char pid_buf[sizeof("nnnnn\0")], c; + char *pid_args[] = { "/usr/bin/uml_watchdog", "-pid", pid_buf, NULL }; + char *mconsole_args[] = { "/usr/bin/uml_watchdog", "-mconsole", NULL, + NULL }; + char **args = NULL; + + err = os_pipe(in_fds, 1, 0); + if(err < 0){ + printk("harddog_open - os_pipe failed, err = %d\n", -err); + goto out; + } + + err = os_pipe(out_fds, 1, 0); + if(err < 0){ + printk("harddog_open - os_pipe failed, err = %d\n", -err); + goto out_close_in; + } + + data.stdin = out_fds[0]; + data.stdout = in_fds[1]; + data.close_me[0] = out_fds[1]; + data.close_me[1] = in_fds[0]; + + if(sock != NULL){ + mconsole_args[2] = sock; + args = mconsole_args; + } + else { + /* XXX The os_getpid() is not SMP correct */ + sprintf(pid_buf, "%d", CHOOSE_MODE(tracing_pid, os_getpid())); + args = pid_args; + } + + pid = run_helper(pre_exec, &data, args, NULL); + + os_close_file(out_fds[0]); + os_close_file(in_fds[1]); + + if(pid < 0){ + err = -pid; + printk("harddog_open - run_helper failed, errno = %d\n", -err); + goto out_close_out; + } + + n = os_read_file(in_fds[0], &c, sizeof(c)); + if(n == 0){ + printk("harddog_open - EOF on watchdog pipe\n"); + helper_wait(pid); + err = -EIO; + goto out_close_out; + } + else if(n < 0){ + printk("harddog_open - read of watchdog pipe failed, " + "err = %d\n", -n); + helper_wait(pid); + err = n; + goto out_close_out; + } + *in_fd_ret = in_fds[0]; + *out_fd_ret = out_fds[1]; + return(0); + + out_close_in: + os_close_file(in_fds[0]); + os_close_file(in_fds[1]); + out_close_out: + os_close_file(out_fds[0]); + os_close_file(out_fds[1]); + out: + return(err); +} + +void stop_watchdog(int in_fd, int out_fd) +{ + os_close_file(in_fd); + os_close_file(out_fd); +} + +int ping_watchdog(int fd) +{ + int n; + char c = '\n'; + + n = os_write_file(fd, &c, sizeof(c)); + if(n != sizeof(c)){ + printk("ping_watchdog - write failed, err = %d\n", -n); + if(n < 0) + return(n); + return(-EIO); + } + return 1; + +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/hostaudio_kern.c b/arch/um/drivers/hostaudio_kern.c new file mode 100644 index 000000000000..d5742783e19d --- /dev/null +++ b/arch/um/drivers/hostaudio_kern.c @@ -0,0 +1,352 @@ +/* + * Copyright (C) 2002 Steve Schmidtke + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/module.h" +#include "linux/init.h" +#include "linux/slab.h" +#include "linux/fs.h" +#include "linux/sound.h" +#include "linux/soundcard.h" +#include "asm/uaccess.h" +#include "kern_util.h" +#include "init.h" +#include "os.h" + +struct hostaudio_state { + int fd; +}; + +struct hostmixer_state { + int fd; +}; + +#define HOSTAUDIO_DEV_DSP "/dev/sound/dsp" +#define HOSTAUDIO_DEV_MIXER "/dev/sound/mixer" + +/* Only changed from linux_main at boot time */ +char *dsp = HOSTAUDIO_DEV_DSP; +char *mixer = HOSTAUDIO_DEV_MIXER; + +#define DSP_HELP \ +" This is used to specify the host dsp device to the hostaudio driver.\n" \ +" The default is \"" HOSTAUDIO_DEV_DSP "\".\n\n" + +#define MIXER_HELP \ +" This is used to specify the host mixer device to the hostaudio driver.\n"\ +" The default is \"" HOSTAUDIO_DEV_MIXER "\".\n\n" + +#ifndef MODULE +static int set_dsp(char *name, int *add) +{ + dsp = name; + return(0); +} + +__uml_setup("dsp=", set_dsp, "dsp=<dsp device>\n" DSP_HELP); + +static int set_mixer(char *name, int *add) +{ + mixer = name; + return(0); +} + +__uml_setup("mixer=", set_mixer, "mixer=<mixer device>\n" MIXER_HELP); + +#else /*MODULE*/ + +MODULE_PARM(dsp, "s"); +MODULE_PARM_DESC(dsp, DSP_HELP); + +MODULE_PARM(mixer, "s"); +MODULE_PARM_DESC(mixer, MIXER_HELP); + +#endif + +/* /dev/dsp file operations */ + +static ssize_t hostaudio_read(struct file *file, char *buffer, size_t count, + loff_t *ppos) +{ + struct hostaudio_state *state = file->private_data; + void *kbuf; + int err; + +#ifdef DEBUG + printk("hostaudio: read called, count = %d\n", count); +#endif + + kbuf = kmalloc(count, GFP_KERNEL); + if(kbuf == NULL) + return(-ENOMEM); + + err = os_read_file(state->fd, kbuf, count); + if(err < 0) + goto out; + + if(copy_to_user(buffer, kbuf, err)) + err = -EFAULT; + + out: + kfree(kbuf); + return(err); +} + +static ssize_t hostaudio_write(struct file *file, const char *buffer, + size_t count, loff_t *ppos) +{ + struct hostaudio_state *state = file->private_data; + void *kbuf; + int err; + +#ifdef DEBUG + printk("hostaudio: write called, count = %d\n", count); +#endif + + kbuf = kmalloc(count, GFP_KERNEL); + if(kbuf == NULL) + return(-ENOMEM); + + err = -EFAULT; + if(copy_from_user(kbuf, buffer, count)) + goto out; + + err = os_write_file(state->fd, kbuf, count); + if(err < 0) + goto out; + *ppos += err; + + out: + kfree(kbuf); + return(err); +} + +static unsigned int hostaudio_poll(struct file *file, + struct poll_table_struct *wait) +{ + unsigned int mask = 0; + +#ifdef DEBUG + printk("hostaudio: poll called (unimplemented)\n"); +#endif + + return(mask); +} + +static int hostaudio_ioctl(struct inode *inode, struct file *file, + unsigned int cmd, unsigned long arg) +{ + struct hostaudio_state *state = file->private_data; + unsigned long data = 0; + int err; + +#ifdef DEBUG + printk("hostaudio: ioctl called, cmd = %u\n", cmd); +#endif + switch(cmd){ + case SNDCTL_DSP_SPEED: + case SNDCTL_DSP_STEREO: + case SNDCTL_DSP_GETBLKSIZE: + case SNDCTL_DSP_CHANNELS: + case SNDCTL_DSP_SUBDIVIDE: + case SNDCTL_DSP_SETFRAGMENT: + if(get_user(data, (int *) arg)) + return(-EFAULT); + break; + default: + break; + } + + err = os_ioctl_generic(state->fd, cmd, (unsigned long) &data); + + switch(cmd){ + case SNDCTL_DSP_SPEED: + case SNDCTL_DSP_STEREO: + case SNDCTL_DSP_GETBLKSIZE: + case SNDCTL_DSP_CHANNELS: + case SNDCTL_DSP_SUBDIVIDE: + case SNDCTL_DSP_SETFRAGMENT: + if(put_user(data, (int *) arg)) + return(-EFAULT); + break; + default: + break; + } + + return(err); +} + +static int hostaudio_open(struct inode *inode, struct file *file) +{ + struct hostaudio_state *state; + int r = 0, w = 0; + int ret; + +#ifdef DEBUG + printk("hostaudio: open called (host: %s)\n", dsp); +#endif + + state = kmalloc(sizeof(struct hostaudio_state), GFP_KERNEL); + if(state == NULL) + return(-ENOMEM); + + if(file->f_mode & FMODE_READ) r = 1; + if(file->f_mode & FMODE_WRITE) w = 1; + + ret = os_open_file(dsp, of_set_rw(OPENFLAGS(), r, w), 0); + if(ret < 0){ + kfree(state); + return(ret); + } + + state->fd = ret; + file->private_data = state; + return(0); +} + +static int hostaudio_release(struct inode *inode, struct file *file) +{ + struct hostaudio_state *state = file->private_data; + +#ifdef DEBUG + printk("hostaudio: release called\n"); +#endif + + os_close_file(state->fd); + kfree(state); + + return(0); +} + +/* /dev/mixer file operations */ + +static int hostmixer_ioctl_mixdev(struct inode *inode, struct file *file, + unsigned int cmd, unsigned long arg) +{ + struct hostmixer_state *state = file->private_data; + +#ifdef DEBUG + printk("hostmixer: ioctl called\n"); +#endif + + return(os_ioctl_generic(state->fd, cmd, arg)); +} + +static int hostmixer_open_mixdev(struct inode *inode, struct file *file) +{ + struct hostmixer_state *state; + int r = 0, w = 0; + int ret; + +#ifdef DEBUG + printk("hostmixer: open called (host: %s)\n", mixer); +#endif + + state = kmalloc(sizeof(struct hostmixer_state), GFP_KERNEL); + if(state == NULL) return(-ENOMEM); + + if(file->f_mode & FMODE_READ) r = 1; + if(file->f_mode & FMODE_WRITE) w = 1; + + ret = os_open_file(mixer, of_set_rw(OPENFLAGS(), r, w), 0); + + if(ret < 0){ + printk("hostaudio_open_mixdev failed to open '%s', err = %d\n", + dsp, -ret); + kfree(state); + return(ret); + } + + file->private_data = state; + return(0); +} + +static int hostmixer_release(struct inode *inode, struct file *file) +{ + struct hostmixer_state *state = file->private_data; + +#ifdef DEBUG + printk("hostmixer: release called\n"); +#endif + + os_close_file(state->fd); + kfree(state); + + return(0); +} + + +/* kernel module operations */ + +static struct file_operations hostaudio_fops = { + .owner = THIS_MODULE, + .llseek = no_llseek, + .read = hostaudio_read, + .write = hostaudio_write, + .poll = hostaudio_poll, + .ioctl = hostaudio_ioctl, + .mmap = NULL, + .open = hostaudio_open, + .release = hostaudio_release, +}; + +static struct file_operations hostmixer_fops = { + .owner = THIS_MODULE, + .llseek = no_llseek, + .ioctl = hostmixer_ioctl_mixdev, + .open = hostmixer_open_mixdev, + .release = hostmixer_release, +}; + +struct { + int dev_audio; + int dev_mixer; +} module_data; + +MODULE_AUTHOR("Steve Schmidtke"); +MODULE_DESCRIPTION("UML Audio Relay"); +MODULE_LICENSE("GPL"); + +static int __init hostaudio_init_module(void) +{ + printk(KERN_INFO "UML Audio Relay (host dsp = %s, host mixer = %s)\n", + dsp, mixer); + + module_data.dev_audio = register_sound_dsp(&hostaudio_fops, -1); + if(module_data.dev_audio < 0){ + printk(KERN_ERR "hostaudio: couldn't register DSP device!\n"); + return -ENODEV; + } + + module_data.dev_mixer = register_sound_mixer(&hostmixer_fops, -1); + if(module_data.dev_mixer < 0){ + printk(KERN_ERR "hostmixer: couldn't register mixer " + "device!\n"); + unregister_sound_dsp(module_data.dev_audio); + return -ENODEV; + } + + return 0; +} + +static void __exit hostaudio_cleanup_module (void) +{ + unregister_sound_mixer(module_data.dev_mixer); + unregister_sound_dsp(module_data.dev_audio); +} + +module_init(hostaudio_init_module); +module_exit(hostaudio_cleanup_module); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c new file mode 100644 index 000000000000..6924f273ced9 --- /dev/null +++ b/arch/um/drivers/line.c @@ -0,0 +1,681 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/slab.h" +#include "linux/list.h" +#include "linux/kd.h" +#include "linux/interrupt.h" +#include "linux/devfs_fs_kernel.h" +#include "asm/uaccess.h" +#include "chan_kern.h" +#include "irq_user.h" +#include "line.h" +#include "kern.h" +#include "user_util.h" +#include "kern_util.h" +#include "os.h" +#include "irq_kern.h" + +#define LINE_BUFSIZE 4096 + +static irqreturn_t line_interrupt(int irq, void *data, struct pt_regs *unused) +{ + struct tty_struct *tty = data; + struct line *line = tty->driver_data; + + if (line) + chan_interrupt(&line->chan_list, &line->task, tty, irq); + return IRQ_HANDLED; +} + +static void line_timer_cb(void *arg) +{ + struct tty_struct *tty = arg; + struct line *line = tty->driver_data; + + line_interrupt(line->driver->read_irq, arg, NULL); +} + +static int write_room(struct line *dev) +{ + int n; + + if (dev->buffer == NULL) + return (LINE_BUFSIZE - 1); + + n = dev->head - dev->tail; + if (n <= 0) + n = LINE_BUFSIZE + n; + return (n - 1); +} + +static int buffer_data(struct line *line, const char *buf, int len) +{ + int end, room; + + if(line->buffer == NULL){ + line->buffer = kmalloc(LINE_BUFSIZE, GFP_ATOMIC); + if (line->buffer == NULL) { + printk("buffer_data - atomic allocation failed\n"); + return(0); + } + line->head = line->buffer; + line->tail = line->buffer; + } + + room = write_room(line); + len = (len > room) ? room : len; + + end = line->buffer + LINE_BUFSIZE - line->tail; + if(len < end){ + memcpy(line->tail, buf, len); + line->tail += len; + } + else { + memcpy(line->tail, buf, end); + buf += end; + memcpy(line->buffer, buf, len - end); + line->tail = line->buffer + len - end; + } + + return(len); +} + +static int flush_buffer(struct line *line) +{ + int n, count; + + if ((line->buffer == NULL) || (line->head == line->tail)) + return(1); + + if (line->tail < line->head) { + count = line->buffer + LINE_BUFSIZE - line->head; + n = write_chan(&line->chan_list, line->head, count, + line->driver->write_irq); + if (n < 0) + return(n); + if (n == count) + line->head = line->buffer; + else { + line->head += n; + return(0); + } + } + + count = line->tail - line->head; + n = write_chan(&line->chan_list, line->head, count, + line->driver->write_irq); + if(n < 0) return(n); + + line->head += n; + return(line->head == line->tail); +} + +int line_write(struct tty_struct *tty, const unsigned char *buf, int len) +{ + struct line *line = tty->driver_data; + unsigned long flags; + int n, err, ret = 0; + + if(tty->stopped) return 0; + + down(&line->sem); + if(line->head != line->tail){ + local_irq_save(flags); + ret = buffer_data(line, buf, len); + err = flush_buffer(line); + local_irq_restore(flags); + if(err <= 0 && (err != -EAGAIN || !ret)) + ret = err; + } + else { + n = write_chan(&line->chan_list, buf, len, + line->driver->write_irq); + if(n < 0){ + ret = n; + goto out_up; + } + + len -= n; + ret += n; + if(len > 0) + ret += buffer_data(line, buf + n, len); + } + out_up: + up(&line->sem); + return(ret); +} + +void line_put_char(struct tty_struct *tty, unsigned char ch) +{ + line_write(tty, &ch, sizeof(ch)); +} + +void line_set_termios(struct tty_struct *tty, struct termios * old) +{ + /* nothing */ +} + +int line_chars_in_buffer(struct tty_struct *tty) +{ + return 0; +} + +static struct { + int cmd; + char *level; + char *name; +} tty_ioctls[] = { + /* don't print these, they flood the log ... */ + { TCGETS, NULL, "TCGETS" }, + { TCSETS, NULL, "TCSETS" }, + { TCSETSW, NULL, "TCSETSW" }, + { TCFLSH, NULL, "TCFLSH" }, + { TCSBRK, NULL, "TCSBRK" }, + + /* general tty stuff */ + { TCSETSF, KERN_DEBUG, "TCSETSF" }, + { TCGETA, KERN_DEBUG, "TCGETA" }, + { TIOCMGET, KERN_DEBUG, "TIOCMGET" }, + { TCSBRKP, KERN_DEBUG, "TCSBRKP" }, + { TIOCMSET, KERN_DEBUG, "TIOCMSET" }, + + /* linux-specific ones */ + { TIOCLINUX, KERN_INFO, "TIOCLINUX" }, + { KDGKBMODE, KERN_INFO, "KDGKBMODE" }, + { KDGKBTYPE, KERN_INFO, "KDGKBTYPE" }, + { KDSIGACCEPT, KERN_INFO, "KDSIGACCEPT" }, +}; + +int line_ioctl(struct tty_struct *tty, struct file * file, + unsigned int cmd, unsigned long arg) +{ + int ret; + int i; + + ret = 0; + switch(cmd) { +#ifdef TIOCGETP + case TIOCGETP: + case TIOCSETP: + case TIOCSETN: +#endif +#ifdef TIOCGETC + case TIOCGETC: + case TIOCSETC: +#endif +#ifdef TIOCGLTC + case TIOCGLTC: + case TIOCSLTC: +#endif + case TCGETS: + case TCSETSF: + case TCSETSW: + case TCSETS: + case TCGETA: + case TCSETAF: + case TCSETAW: + case TCSETA: + case TCXONC: + case TCFLSH: + case TIOCOUTQ: + case TIOCINQ: + case TIOCGLCKTRMIOS: + case TIOCSLCKTRMIOS: + case TIOCPKT: + case TIOCGSOFTCAR: + case TIOCSSOFTCAR: + return -ENOIOCTLCMD; +#if 0 + case TCwhatever: + /* do something */ + break; +#endif + default: + for (i = 0; i < ARRAY_SIZE(tty_ioctls); i++) + if (cmd == tty_ioctls[i].cmd) + break; + if (i < ARRAY_SIZE(tty_ioctls)) { + if (NULL != tty_ioctls[i].level) + printk("%s%s: %s: ioctl %s called\n", + tty_ioctls[i].level, __FUNCTION__, + tty->name, tty_ioctls[i].name); + } else { + printk(KERN_ERR "%s: %s: unknown ioctl: 0x%x\n", + __FUNCTION__, tty->name, cmd); + } + ret = -ENOIOCTLCMD; + break; + } + return(ret); +} + +static irqreturn_t line_write_interrupt(int irq, void *data, + struct pt_regs *unused) +{ + struct tty_struct *tty = data; + struct line *line = tty->driver_data; + int err; + + err = flush_buffer(line); + if(err == 0) + return(IRQ_NONE); + else if(err < 0){ + line->head = line->buffer; + line->tail = line->buffer; + } + + if(tty == NULL) + return(IRQ_NONE); + + if(test_bit(TTY_DO_WRITE_WAKEUP, &tty->flags) && + (tty->ldisc.write_wakeup != NULL)) + (tty->ldisc.write_wakeup)(tty); + + /* BLOCKING mode + * In blocking mode, everything sleeps on tty->write_wait. + * Sleeping in the console driver would break non-blocking + * writes. + */ + + if(waitqueue_active(&tty->write_wait)) + wake_up_interruptible(&tty->write_wait); + return(IRQ_HANDLED); +} + +int line_setup_irq(int fd, int input, int output, struct tty_struct *tty) +{ + struct line *line = tty->driver_data; + struct line_driver *driver = line->driver; + int err = 0, flags = SA_INTERRUPT | SA_SHIRQ | SA_SAMPLE_RANDOM; + + if(input) err = um_request_irq(driver->read_irq, fd, IRQ_READ, + line_interrupt, flags, + driver->read_irq_name, tty); + if(err) return(err); + if(output) err = um_request_irq(driver->write_irq, fd, IRQ_WRITE, + line_write_interrupt, flags, + driver->write_irq_name, tty); + line->have_irq = 1; + return(err); +} + +void line_disable(struct tty_struct *tty, int current_irq) +{ + struct line *line = tty->driver_data; + + if(!line->have_irq) + return; + + if(line->driver->read_irq == current_irq) + free_irq_later(line->driver->read_irq, tty); + else { + free_irq_by_irq_and_dev(line->driver->read_irq, tty); + free_irq(line->driver->read_irq, tty); + } + + if(line->driver->write_irq == current_irq) + free_irq_later(line->driver->write_irq, tty); + else { + free_irq_by_irq_and_dev(line->driver->write_irq, tty); + free_irq(line->driver->write_irq, tty); + } + + line->have_irq = 0; +} + +int line_open(struct line *lines, struct tty_struct *tty, + struct chan_opts *opts) +{ + struct line *line; + int err = 0; + + line = &lines[tty->index]; + tty->driver_data = line; + + down(&line->sem); + if (tty->count == 1) { + if (!line->valid) { + err = -ENODEV; + goto out; + } + if (list_empty(&line->chan_list)) { + err = parse_chan_pair(line->init_str, &line->chan_list, + line->init_pri, tty->index, opts); + if(err) goto out; + err = open_chan(&line->chan_list); + if(err) goto out; + } + enable_chan(&line->chan_list, tty); + INIT_WORK(&line->task, line_timer_cb, tty); + } + + if(!line->sigio){ + chan_enable_winch(&line->chan_list, tty); + line->sigio = 1; + } + chan_window_size(&line->chan_list, &tty->winsize.ws_row, + &tty->winsize.ws_col); + line->count++; + +out: + up(&line->sem); + return(err); +} + +void line_close(struct tty_struct *tty, struct file * filp) +{ + struct line *line = tty->driver_data; + + down(&line->sem); + line->count--; + if (tty->count == 1) { + line_disable(tty, -1); + tty->driver_data = NULL; + } + up(&line->sem); +} + +void close_lines(struct line *lines, int nlines) +{ + int i; + + for(i = 0; i < nlines; i++) + close_chan(&lines[i].chan_list); +} + +int line_setup(struct line *lines, int num, char *init, int all_allowed) +{ + int i, n; + char *end; + + if(*init == '=') n = -1; + else { + n = simple_strtoul(init, &end, 0); + if(*end != '='){ + printk(KERN_ERR "line_setup failed to parse \"%s\"\n", + init); + return(0); + } + init = end; + } + init++; + if((n >= 0) && (n >= num)){ + printk("line_setup - %d out of range ((0 ... %d) allowed)\n", + n, num - 1); + return(0); + } + else if (n >= 0){ + if (lines[n].count > 0) { + printk("line_setup - device %d is open\n", n); + return(0); + } + if (lines[n].init_pri <= INIT_ONE){ + lines[n].init_pri = INIT_ONE; + if (!strcmp(init, "none")) + lines[n].valid = 0; + else { + lines[n].init_str = init; + lines[n].valid = 1; + } + } + } + else if(!all_allowed){ + printk("line_setup - can't configure all devices from " + "mconsole\n"); + return(0); + } + else { + for(i = 0; i < num; i++){ + if(lines[i].init_pri <= INIT_ALL){ + lines[i].init_pri = INIT_ALL; + if(!strcmp(init, "none")) lines[i].valid = 0; + else { + lines[i].init_str = init; + lines[i].valid = 1; + } + } + } + } + return(1); +} + +int line_config(struct line *lines, int num, char *str) +{ + char *new = uml_strdup(str); + + if(new == NULL){ + printk("line_config - uml_strdup failed\n"); + return(-ENOMEM); + } + return(!line_setup(lines, num, new, 0)); +} + +int line_get_config(char *name, struct line *lines, int num, char *str, + int size, char **error_out) +{ + struct line *line; + char *end; + int dev, n = 0; + + dev = simple_strtoul(name, &end, 0); + if((*end != '\0') || (end == name)){ + *error_out = "line_get_config failed to parse device number"; + return(0); + } + + if((dev < 0) || (dev >= num)){ + *error_out = "device number of of range"; + return(0); + } + + line = &lines[dev]; + + down(&line->sem); + if(!line->valid) + CONFIG_CHUNK(str, size, n, "none", 1); + else if(line->count == 0) + CONFIG_CHUNK(str, size, n, line->init_str, 1); + else n = chan_config_string(&line->chan_list, str, size, error_out); + up(&line->sem); + + return(n); +} + +int line_remove(struct line *lines, int num, char *str) +{ + char config[sizeof("conxxxx=none\0")]; + + sprintf(config, "%s=none", str); + return(!line_setup(lines, num, config, 0)); +} + +int line_write_room(struct tty_struct *tty) +{ + struct line *dev = tty->driver_data; + int room; + + if (tty->stopped) + return 0; + room = write_room(dev); + if (0 == room) + printk(KERN_DEBUG "%s: %s: no room left in buffer\n", + __FUNCTION__,tty->name); + return room; +} + +struct tty_driver *line_register_devfs(struct lines *set, + struct line_driver *line_driver, + struct tty_operations *ops, struct line *lines, + int nlines) +{ + int i; + struct tty_driver *driver = alloc_tty_driver(nlines); + + if (!driver) + return NULL; + + driver->driver_name = line_driver->name; + driver->name = line_driver->device_name; + driver->devfs_name = line_driver->devfs_name; + driver->major = line_driver->major; + driver->minor_start = line_driver->minor_start; + driver->type = line_driver->type; + driver->subtype = line_driver->subtype; + driver->flags = TTY_DRIVER_REAL_RAW; + driver->init_termios = tty_std_termios; + tty_set_operations(driver, ops); + + if (tty_register_driver(driver)) { + printk("%s: can't register %s driver\n", + __FUNCTION__,line_driver->name); + put_tty_driver(driver); + return NULL; + } + + for(i = 0; i < nlines; i++){ + if(!lines[i].valid) + tty_unregister_device(driver, i); + } + + mconsole_register_dev(&line_driver->mc); + return driver; +} + +void lines_init(struct line *lines, int nlines) +{ + struct line *line; + int i; + + for(i = 0; i < nlines; i++){ + line = &lines[i]; + INIT_LIST_HEAD(&line->chan_list); + sema_init(&line->sem, 1); + if(line->init_str != NULL){ + line->init_str = uml_strdup(line->init_str); + if(line->init_str == NULL) + printk("lines_init - uml_strdup returned " + "NULL\n"); + } + } +} + +struct winch { + struct list_head list; + int fd; + int tty_fd; + int pid; + struct tty_struct *tty; +}; + +irqreturn_t winch_interrupt(int irq, void *data, struct pt_regs *unused) +{ + struct winch *winch = data; + struct tty_struct *tty; + struct line *line; + int err; + char c; + + if(winch->fd != -1){ + err = generic_read(winch->fd, &c, NULL); + if(err < 0){ + if(err != -EAGAIN){ + printk("winch_interrupt : read failed, " + "errno = %d\n", -err); + printk("fd %d is losing SIGWINCH support\n", + winch->tty_fd); + return(IRQ_HANDLED); + } + goto out; + } + } + tty = winch->tty; + if (tty != NULL) { + line = tty->driver_data; + chan_window_size(&line->chan_list, + &tty->winsize.ws_row, + &tty->winsize.ws_col); + kill_pg(tty->pgrp, SIGWINCH, 1); + } + out: + if(winch->fd != -1) + reactivate_fd(winch->fd, WINCH_IRQ); + return(IRQ_HANDLED); +} + +DECLARE_MUTEX(winch_handler_sem); +LIST_HEAD(winch_handlers); + +void register_winch_irq(int fd, int tty_fd, int pid, struct tty_struct *tty) +{ + struct winch *winch; + + down(&winch_handler_sem); + winch = kmalloc(sizeof(*winch), GFP_KERNEL); + if (winch == NULL) { + printk("register_winch_irq - kmalloc failed\n"); + goto out; + } + *winch = ((struct winch) { .list = LIST_HEAD_INIT(winch->list), + .fd = fd, + .tty_fd = tty_fd, + .pid = pid, + .tty = tty }); + list_add(&winch->list, &winch_handlers); + if(um_request_irq(WINCH_IRQ, fd, IRQ_READ, winch_interrupt, + SA_INTERRUPT | SA_SHIRQ | SA_SAMPLE_RANDOM, + "winch", winch) < 0) + printk("register_winch_irq - failed to register IRQ\n"); + out: + up(&winch_handler_sem); +} + +static void winch_cleanup(void) +{ + struct list_head *ele; + struct winch *winch; + + list_for_each(ele, &winch_handlers){ + winch = list_entry(ele, struct winch, list); + if(winch->fd != -1){ + deactivate_fd(winch->fd, WINCH_IRQ); + os_close_file(winch->fd); + } + if(winch->pid != -1) + os_kill_process(winch->pid, 1); + } +} +__uml_exitcall(winch_cleanup); + +char *add_xterm_umid(char *base) +{ + char *umid, *title; + int len; + + umid = get_umid(1); + if(umid == NULL) return(base); + + len = strlen(base) + strlen(" ()") + strlen(umid) + 1; + title = kmalloc(len, GFP_KERNEL); + if(title == NULL){ + printk("Failed to allocate buffer for xterm title\n"); + return(base); + } + + snprintf(title, len, "%s (%s)", base, umid); + return(title); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/mcast.h b/arch/um/drivers/mcast.h new file mode 100644 index 000000000000..a2c6db243458 --- /dev/null +++ b/arch/um/drivers/mcast.h @@ -0,0 +1,30 @@ +/* + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "net_user.h" + +struct mcast_data { + char *addr; + unsigned short port; + void *mcast_addr; + int ttl; + void *dev; +}; + +extern struct net_user_info mcast_user_info; + +extern int mcast_user_write(int fd, void *buf, int len, + struct mcast_data *pri); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/mcast_kern.c b/arch/um/drivers/mcast_kern.c new file mode 100644 index 000000000000..faf714e87b5b --- /dev/null +++ b/arch/um/drivers/mcast_kern.c @@ -0,0 +1,143 @@ +/* + * user-mode-linux networking multicast transport + * Copyright (C) 2001 by Harald Welte <laforge@gnumonks.org> + * + * based on the existing uml-networking code, which is + * Copyright (C) 2001 Lennert Buytenhek (buytenh@gnu.org) and + * James Leu (jleu@mindspring.net). + * Copyright (C) 2001 by various other people who didn't put their name here. + * + * Licensed under the GPL. + */ + +#include "linux/kernel.h" +#include "linux/init.h" +#include "linux/netdevice.h" +#include "linux/etherdevice.h" +#include "linux/in.h" +#include "linux/inet.h" +#include "net_kern.h" +#include "net_user.h" +#include "mcast.h" + +struct mcast_init { + char *addr; + int port; + int ttl; +}; + +void mcast_init(struct net_device *dev, void *data) +{ + struct uml_net_private *pri; + struct mcast_data *dpri; + struct mcast_init *init = data; + + pri = dev->priv; + dpri = (struct mcast_data *) pri->user; + dpri->addr = init->addr; + dpri->port = init->port; + dpri->ttl = init->ttl; + dpri->dev = dev; + + printk("mcast backend "); + printk("multicast adddress: %s:%u, TTL:%u ", + dpri->addr, dpri->port, dpri->ttl); + + printk("\n"); +} + +static int mcast_read(int fd, struct sk_buff **skb, struct uml_net_private *lp) +{ + *skb = ether_adjust_skb(*skb, ETH_HEADER_OTHER); + if(*skb == NULL) return(-ENOMEM); + return(net_recvfrom(fd, (*skb)->mac.raw, + (*skb)->dev->mtu + ETH_HEADER_OTHER)); +} + +static int mcast_write(int fd, struct sk_buff **skb, + struct uml_net_private *lp) +{ + return mcast_user_write(fd, (*skb)->data, (*skb)->len, + (struct mcast_data *) &lp->user); +} + +static struct net_kern_info mcast_kern_info = { + .init = mcast_init, + .protocol = eth_protocol, + .read = mcast_read, + .write = mcast_write, +}; + +int mcast_setup(char *str, char **mac_out, void *data) +{ + struct mcast_init *init = data; + char *port_str = NULL, *ttl_str = NULL, *remain; + char *last; + int n; + + *init = ((struct mcast_init) + { .addr = "239.192.168.1", + .port = 1102, + .ttl = 1 }); + + remain = split_if_spec(str, mac_out, &init->addr, &port_str, &ttl_str, + NULL); + if(remain != NULL){ + printk(KERN_ERR "mcast_setup - Extra garbage on " + "specification : '%s'\n", remain); + return(0); + } + + if(port_str != NULL){ + n = simple_strtoul(port_str, &last, 10); + if((*last != '\0') || (last == port_str)){ + printk(KERN_ERR "mcast_setup - Bad port : '%s'\n", + port_str); + return(0); + } + init->port = htons(n); + } + + if(ttl_str != NULL){ + init->ttl = simple_strtoul(ttl_str, &last, 10); + if((*last != '\0') || (last == ttl_str)){ + printk(KERN_ERR "mcast_setup - Bad ttl : '%s'\n", + ttl_str); + return(0); + } + } + + printk(KERN_INFO "Configured mcast device: %s:%u-%u\n", init->addr, + init->port, init->ttl); + + return(1); +} + +static struct transport mcast_transport = { + .list = LIST_HEAD_INIT(mcast_transport.list), + .name = "mcast", + .setup = mcast_setup, + .user = &mcast_user_info, + .kern = &mcast_kern_info, + .private_size = sizeof(struct mcast_data), + .setup_size = sizeof(struct mcast_init), +}; + +static int register_mcast(void) +{ + register_transport(&mcast_transport); + return(1); +} + +__initcall(register_mcast); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/mcast_user.c b/arch/um/drivers/mcast_user.c new file mode 100644 index 000000000000..0fe1d9fa9139 --- /dev/null +++ b/arch/um/drivers/mcast_user.c @@ -0,0 +1,177 @@ +/* + * user-mode-linux networking multicast transport + * Copyright (C) 2001 by Harald Welte <laforge@gnumonks.org> + * + * based on the existing uml-networking code, which is + * Copyright (C) 2001 Lennert Buytenhek (buytenh@gnu.org) and + * James Leu (jleu@mindspring.net). + * Copyright (C) 2001 by various other people who didn't put their name here. + * + * Licensed under the GPL. + * + */ + +#include <errno.h> +#include <unistd.h> +#include <linux/inet.h> +#include <sys/socket.h> +#include <sys/un.h> +#include <sys/time.h> +#include <netinet/in.h> +#include "net_user.h" +#include "mcast.h" +#include "kern_util.h" +#include "user_util.h" +#include "user.h" +#include "os.h" + +#define MAX_PACKET (ETH_MAX_PACKET + ETH_HEADER_OTHER) + +static struct sockaddr_in *new_addr(char *addr, unsigned short port) +{ + struct sockaddr_in *sin; + + sin = um_kmalloc(sizeof(struct sockaddr_in)); + if(sin == NULL){ + printk("new_addr: allocation of sockaddr_in failed\n"); + return(NULL); + } + sin->sin_family = AF_INET; + sin->sin_addr.s_addr = in_aton(addr); + sin->sin_port = port; + return(sin); +} + +static void mcast_user_init(void *data, void *dev) +{ + struct mcast_data *pri = data; + + pri->mcast_addr = new_addr(pri->addr, pri->port); + pri->dev = dev; +} + +static int mcast_open(void *data) +{ + struct mcast_data *pri = data; + struct sockaddr_in *sin = pri->mcast_addr; + struct ip_mreq mreq; + int fd, yes = 1; + + + if ((sin->sin_addr.s_addr == 0) || (sin->sin_port == 0)) { + fd = -EINVAL; + goto out; + } + + fd = socket(AF_INET, SOCK_DGRAM, 0); + if (fd < 0){ + printk("mcast_open : data socket failed, errno = %d\n", + errno); + fd = -ENOMEM; + goto out; + } + + if (setsockopt(fd, SOL_SOCKET, SO_REUSEADDR, &yes, sizeof(yes)) < 0) { + printk("mcast_open: SO_REUSEADDR failed, errno = %d\n", + errno); + os_close_file(fd); + fd = -EINVAL; + goto out; + } + + /* set ttl according to config */ + if (setsockopt(fd, SOL_IP, IP_MULTICAST_TTL, &pri->ttl, + sizeof(pri->ttl)) < 0) { + printk("mcast_open: IP_MULTICAST_TTL failed, error = %d\n", + errno); + os_close_file(fd); + fd = -EINVAL; + goto out; + } + + /* set LOOP, so data does get fed back to local sockets */ + if (setsockopt(fd, SOL_IP, IP_MULTICAST_LOOP, &yes, sizeof(yes)) < 0) { + printk("mcast_open: IP_MULTICAST_LOOP failed, error = %d\n", + errno); + os_close_file(fd); + fd = -EINVAL; + goto out; + } + + /* bind socket to mcast address */ + if (bind(fd, (struct sockaddr *) sin, sizeof(*sin)) < 0) { + printk("mcast_open : data bind failed, errno = %d\n", errno); + os_close_file(fd); + fd = -EINVAL; + goto out; + } + + /* subscribe to the multicast group */ + mreq.imr_multiaddr.s_addr = sin->sin_addr.s_addr; + mreq.imr_interface.s_addr = 0; + if (setsockopt(fd, SOL_IP, IP_ADD_MEMBERSHIP, + &mreq, sizeof(mreq)) < 0) { + printk("mcast_open: IP_ADD_MEMBERSHIP failed, error = %d\n", + errno); + printk("There appears not to be a multicast-capable network " + "interface on the host.\n"); + printk("eth0 should be configured in order to use the " + "multicast transport.\n"); + os_close_file(fd); + fd = -EINVAL; + } + + out: + return(fd); +} + +static void mcast_close(int fd, void *data) +{ + struct ip_mreq mreq; + struct mcast_data *pri = data; + struct sockaddr_in *sin = pri->mcast_addr; + + mreq.imr_multiaddr.s_addr = sin->sin_addr.s_addr; + mreq.imr_interface.s_addr = 0; + if (setsockopt(fd, SOL_IP, IP_DROP_MEMBERSHIP, + &mreq, sizeof(mreq)) < 0) { + printk("mcast_open: IP_DROP_MEMBERSHIP failed, error = %d\n", + errno); + } + + os_close_file(fd); +} + +int mcast_user_write(int fd, void *buf, int len, struct mcast_data *pri) +{ + struct sockaddr_in *data_addr = pri->mcast_addr; + + return(net_sendto(fd, buf, len, data_addr, sizeof(*data_addr))); +} + +static int mcast_set_mtu(int mtu, void *data) +{ + return(mtu); +} + +struct net_user_info mcast_user_info = { + .init = mcast_user_init, + .open = mcast_open, + .close = mcast_close, + .remove = NULL, + .set_mtu = mcast_set_mtu, + .add_address = NULL, + .delete_address = NULL, + .max_packet = MAX_PACKET - ETH_HEADER_OTHER +}; + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c new file mode 100644 index 000000000000..d7c7adcc0a67 --- /dev/null +++ b/arch/um/drivers/mconsole_kern.c @@ -0,0 +1,619 @@ +/* + * Copyright (C) 2001 Lennert Buytenhek (buytenh@gnu.org) + * Copyright (C) 2001 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/kernel.h" +#include "linux/slab.h" +#include "linux/init.h" +#include "linux/notifier.h" +#include "linux/reboot.h" +#include "linux/utsname.h" +#include "linux/ctype.h" +#include "linux/interrupt.h" +#include "linux/sysrq.h" +#include "linux/workqueue.h" +#include "linux/module.h" +#include "linux/file.h" +#include "linux/fs.h" +#include "linux/namei.h" +#include "linux/proc_fs.h" +#include "linux/syscalls.h" +#include "asm/irq.h" +#include "asm/uaccess.h" +#include "user_util.h" +#include "kern_util.h" +#include "kern.h" +#include "mconsole.h" +#include "mconsole_kern.h" +#include "irq_user.h" +#include "init.h" +#include "os.h" +#include "umid.h" +#include "irq_kern.h" + +static int do_unlink_socket(struct notifier_block *notifier, + unsigned long what, void *data) +{ + return(mconsole_unlink_socket()); +} + + +static struct notifier_block reboot_notifier = { + .notifier_call = do_unlink_socket, + .priority = 0, +}; + +/* Safe without explicit locking for now. Tasklets provide their own + * locking, and the interrupt handler is safe because it can't interrupt + * itself and it can only happen on CPU 0. + */ + +LIST_HEAD(mc_requests); + +static void mc_work_proc(void *unused) +{ + struct mconsole_entry *req; + unsigned long flags; + + while(!list_empty(&mc_requests)){ + local_save_flags(flags); + req = list_entry(mc_requests.next, struct mconsole_entry, + list); + list_del(&req->list); + local_irq_restore(flags); + req->request.cmd->handler(&req->request); + kfree(req); + } +} + +DECLARE_WORK(mconsole_work, mc_work_proc, NULL); + +static irqreturn_t mconsole_interrupt(int irq, void *dev_id, + struct pt_regs *regs) +{ + /* long to avoid size mismatch warnings from gcc */ + long fd; + struct mconsole_entry *new; + struct mc_request req; + + fd = (long) dev_id; + while (mconsole_get_request(fd, &req)){ + if(req.cmd->context == MCONSOLE_INTR) + (*req.cmd->handler)(&req); + else { + new = kmalloc(sizeof(*new), GFP_ATOMIC); + if(new == NULL) + mconsole_reply(&req, "Out of memory", 1, 0); + else { + new->request = req; + list_add(&new->list, &mc_requests); + } + } + } + if(!list_empty(&mc_requests)) + schedule_work(&mconsole_work); + reactivate_fd(fd, MCONSOLE_IRQ); + return(IRQ_HANDLED); +} + +void mconsole_version(struct mc_request *req) +{ + char version[256]; + + sprintf(version, "%s %s %s %s %s", system_utsname.sysname, + system_utsname.nodename, system_utsname.release, + system_utsname.version, system_utsname.machine); + mconsole_reply(req, version, 0, 0); +} + +void mconsole_log(struct mc_request *req) +{ + int len; + char *ptr = req->request.data; + + ptr += strlen("log "); + + len = req->len - (ptr - req->request.data); + printk("%.*s", len, ptr); + mconsole_reply(req, "", 0, 0); +} + +/* This is a more convoluted version of mconsole_proc, which has some stability + * problems; however, we need it fixed, because it is expected that UML users + * mount HPPFS instead of procfs on /proc. And we want mconsole_proc to still + * show the real procfs content, not the ones from hppfs.*/ +#if 0 +void mconsole_proc(struct mc_request *req) +{ + struct nameidata nd; + struct file_system_type *proc; + struct super_block *super; + struct file *file; + int n, err; + char *ptr = req->request.data, *buf; + + ptr += strlen("proc"); + while(isspace(*ptr)) ptr++; + + proc = get_fs_type("proc"); + if(proc == NULL){ + mconsole_reply(req, "procfs not registered", 1, 0); + goto out; + } + + super = (*proc->get_sb)(proc, 0, NULL, NULL); + put_filesystem(proc); + if(super == NULL){ + mconsole_reply(req, "Failed to get procfs superblock", 1, 0); + goto out; + } + up_write(&super->s_umount); + + nd.dentry = super->s_root; + nd.mnt = NULL; + nd.flags = O_RDONLY + 1; + nd.last_type = LAST_ROOT; + + /* START: it was experienced that the stability problems are closed + * if commenting out these two calls + the below read cycle. To + * make UML crash again, it was enough to readd either one.*/ + err = link_path_walk(ptr, &nd); + if(err){ + mconsole_reply(req, "Failed to look up file", 1, 0); + goto out_kill; + } + + file = dentry_open(nd.dentry, nd.mnt, O_RDONLY); + if(IS_ERR(file)){ + mconsole_reply(req, "Failed to open file", 1, 0); + goto out_kill; + } + /*END*/ + + buf = kmalloc(PAGE_SIZE, GFP_KERNEL); + if(buf == NULL){ + mconsole_reply(req, "Failed to allocate buffer", 1, 0); + goto out_fput; + } + + if((file->f_op != NULL) && (file->f_op->read != NULL)){ + do { + n = (*file->f_op->read)(file, buf, PAGE_SIZE - 1, + &file->f_pos); + if(n >= 0){ + buf[n] = '\0'; + mconsole_reply(req, buf, 0, (n > 0)); + } + else { + mconsole_reply(req, "Read of file failed", + 1, 0); + goto out_free; + } + } while(n > 0); + } + else mconsole_reply(req, "", 0, 0); + + out_free: + kfree(buf); + out_fput: + fput(file); + out_kill: + deactivate_super(super); + out: ; +} +#endif + +void mconsole_proc(struct mc_request *req) +{ + char path[64]; + char *buf; + int len; + int fd; + int first_chunk = 1; + char *ptr = req->request.data; + + ptr += strlen("proc"); + while(isspace(*ptr)) ptr++; + snprintf(path, sizeof(path), "/proc/%s", ptr); + + fd = sys_open(path, 0, 0); + if (fd < 0) { + mconsole_reply(req, "Failed to open file", 1, 0); + printk("open %s: %d\n",path,fd); + goto out; + } + + buf = kmalloc(PAGE_SIZE, GFP_KERNEL); + if(buf == NULL){ + mconsole_reply(req, "Failed to allocate buffer", 1, 0); + goto out_close; + } + + for (;;) { + len = sys_read(fd, buf, PAGE_SIZE-1); + if (len < 0) { + mconsole_reply(req, "Read of file failed", 1, 0); + goto out_free; + } + /*Begin the file content on his own line.*/ + if (first_chunk) { + mconsole_reply(req, "\n", 0, 1); + first_chunk = 0; + } + if (len == PAGE_SIZE-1) { + buf[len] = '\0'; + mconsole_reply(req, buf, 0, 1); + } else { + buf[len] = '\0'; + mconsole_reply(req, buf, 0, 0); + break; + } + } + + out_free: + kfree(buf); + out_close: + sys_close(fd); + out: + /* nothing */; +} + +#define UML_MCONSOLE_HELPTEXT \ +"Commands: \n\ + version - Get kernel version \n\ + help - Print this message \n\ + halt - Halt UML \n\ + reboot - Reboot UML \n\ + config <dev>=<config> - Add a new device to UML; \n\ + same syntax as command line \n\ + config <dev> - Query the configuration of a device \n\ + remove <dev> - Remove a device from UML \n\ + sysrq <letter> - Performs the SysRq action controlled by the letter \n\ + cad - invoke the Ctl-Alt-Del handler \n\ + stop - pause the UML; it will do nothing until it receives a 'go' \n\ + go - continue the UML after a 'stop' \n\ + log <string> - make UML enter <string> into the kernel log\n\ + proc <file> - returns the contents of the UML's /proc/<file>\n\ +" + +void mconsole_help(struct mc_request *req) +{ + mconsole_reply(req, UML_MCONSOLE_HELPTEXT, 0, 0); +} + +void mconsole_halt(struct mc_request *req) +{ + mconsole_reply(req, "", 0, 0); + machine_halt(); +} + +void mconsole_reboot(struct mc_request *req) +{ + mconsole_reply(req, "", 0, 0); + machine_restart(NULL); +} + +extern void ctrl_alt_del(void); + +void mconsole_cad(struct mc_request *req) +{ + mconsole_reply(req, "", 0, 0); + ctrl_alt_del(); +} + +void mconsole_go(struct mc_request *req) +{ + mconsole_reply(req, "Not stopped", 1, 0); +} + +void mconsole_stop(struct mc_request *req) +{ + deactivate_fd(req->originating_fd, MCONSOLE_IRQ); + os_set_fd_block(req->originating_fd, 1); + mconsole_reply(req, "", 0, 0); + while(mconsole_get_request(req->originating_fd, req)){ + if(req->cmd->handler == mconsole_go) break; + (*req->cmd->handler)(req); + } + os_set_fd_block(req->originating_fd, 0); + reactivate_fd(req->originating_fd, MCONSOLE_IRQ); + mconsole_reply(req, "", 0, 0); +} + +/* This list is populated by __initcall routines. */ + +LIST_HEAD(mconsole_devices); + +void mconsole_register_dev(struct mc_device *new) +{ + list_add(&new->list, &mconsole_devices); +} + +static struct mc_device *mconsole_find_dev(char *name) +{ + struct list_head *ele; + struct mc_device *dev; + + list_for_each(ele, &mconsole_devices){ + dev = list_entry(ele, struct mc_device, list); + if(!strncmp(name, dev->name, strlen(dev->name))) + return(dev); + } + return(NULL); +} + +#define CONFIG_BUF_SIZE 64 + +static void mconsole_get_config(int (*get_config)(char *, char *, int, + char **), + struct mc_request *req, char *name) +{ + char default_buf[CONFIG_BUF_SIZE], *error, *buf; + int n, size; + + if(get_config == NULL){ + mconsole_reply(req, "No get_config routine defined", 1, 0); + return; + } + + error = NULL; + size = sizeof(default_buf)/sizeof(default_buf[0]); + buf = default_buf; + + while(1){ + n = (*get_config)(name, buf, size, &error); + if(error != NULL){ + mconsole_reply(req, error, 1, 0); + goto out; + } + + if(n <= size){ + mconsole_reply(req, buf, 0, 0); + goto out; + } + + if(buf != default_buf) + kfree(buf); + + size = n; + buf = kmalloc(size, GFP_KERNEL); + if(buf == NULL){ + mconsole_reply(req, "Failed to allocate buffer", 1, 0); + return; + } + } + out: + if(buf != default_buf) + kfree(buf); + +} + +void mconsole_config(struct mc_request *req) +{ + struct mc_device *dev; + char *ptr = req->request.data, *name; + int err; + + ptr += strlen("config"); + while(isspace(*ptr)) ptr++; + dev = mconsole_find_dev(ptr); + if(dev == NULL){ + mconsole_reply(req, "Bad configuration option", 1, 0); + return; + } + + name = &ptr[strlen(dev->name)]; + ptr = name; + while((*ptr != '=') && (*ptr != '\0')) + ptr++; + + if(*ptr == '='){ + err = (*dev->config)(name); + mconsole_reply(req, "", err, 0); + } + else mconsole_get_config(dev->get_config, req, name); +} + +void mconsole_remove(struct mc_request *req) +{ + struct mc_device *dev; + char *ptr = req->request.data; + int err; + + ptr += strlen("remove"); + while(isspace(*ptr)) ptr++; + dev = mconsole_find_dev(ptr); + if(dev == NULL){ + mconsole_reply(req, "Bad remove option", 1, 0); + return; + } + err = (*dev->remove)(&ptr[strlen(dev->name)]); + mconsole_reply(req, "", err, 0); +} + +#ifdef CONFIG_MAGIC_SYSRQ +void mconsole_sysrq(struct mc_request *req) +{ + char *ptr = req->request.data; + + ptr += strlen("sysrq"); + while(isspace(*ptr)) ptr++; + + mconsole_reply(req, "", 0, 0); + handle_sysrq(*ptr, ¤t->thread.regs, NULL); +} +#else +void mconsole_sysrq(struct mc_request *req) +{ + mconsole_reply(req, "Sysrq not compiled in", 1, 0); +} +#endif + +/* Changed by mconsole_setup, which is __setup, and called before SMP is + * active. + */ +static char *notify_socket = NULL; + +int mconsole_init(void) +{ + /* long to avoid size mismatch warnings from gcc */ + long sock; + int err; + char file[256]; + + if(umid_file_name("mconsole", file, sizeof(file))) return(-1); + snprintf(mconsole_socket_name, sizeof(file), "%s", file); + + sock = os_create_unix_socket(file, sizeof(file), 1); + if (sock < 0){ + printk("Failed to initialize management console\n"); + return(1); + } + + register_reboot_notifier(&reboot_notifier); + + err = um_request_irq(MCONSOLE_IRQ, sock, IRQ_READ, mconsole_interrupt, + SA_INTERRUPT | SA_SHIRQ | SA_SAMPLE_RANDOM, + "mconsole", (void *)sock); + if (err){ + printk("Failed to get IRQ for management console\n"); + return(1); + } + + if(notify_socket != NULL){ + notify_socket = uml_strdup(notify_socket); + if(notify_socket != NULL) + mconsole_notify(notify_socket, MCONSOLE_SOCKET, + mconsole_socket_name, + strlen(mconsole_socket_name) + 1); + else printk(KERN_ERR "mconsole_setup failed to strdup " + "string\n"); + } + + printk("mconsole (version %d) initialized on %s\n", + MCONSOLE_VERSION, mconsole_socket_name); + return(0); +} + +__initcall(mconsole_init); + +static int write_proc_mconsole(struct file *file, const char __user *buffer, + unsigned long count, void *data) +{ + char *buf; + + buf = kmalloc(count + 1, GFP_KERNEL); + if(buf == NULL) + return(-ENOMEM); + + if(copy_from_user(buf, buffer, count)){ + count = -EFAULT; + goto out; + } + + buf[count] = '\0'; + + mconsole_notify(notify_socket, MCONSOLE_USER_NOTIFY, buf, count); + out: + kfree(buf); + return(count); +} + +static int create_proc_mconsole(void) +{ + struct proc_dir_entry *ent; + + if(notify_socket == NULL) return(0); + + ent = create_proc_entry("mconsole", S_IFREG | 0200, NULL); + if(ent == NULL){ + printk("create_proc_mconsole : create_proc_entry failed\n"); + return(0); + } + + ent->read_proc = NULL; + ent->write_proc = write_proc_mconsole; + return(0); +} + +static DEFINE_SPINLOCK(notify_spinlock); + +void lock_notify(void) +{ + spin_lock(¬ify_spinlock); +} + +void unlock_notify(void) +{ + spin_unlock(¬ify_spinlock); +} + +__initcall(create_proc_mconsole); + +#define NOTIFY "=notify:" + +static int mconsole_setup(char *str) +{ + if(!strncmp(str, NOTIFY, strlen(NOTIFY))){ + str += strlen(NOTIFY); + notify_socket = str; + } + else printk(KERN_ERR "mconsole_setup : Unknown option - '%s'\n", str); + return(1); +} + +__setup("mconsole", mconsole_setup); + +__uml_help(mconsole_setup, +"mconsole=notify:<socket>\n" +" Requests that the mconsole driver send a message to the named Unix\n" +" socket containing the name of the mconsole socket. This also serves\n" +" to notify outside processes when UML has booted far enough to respond\n" +" to mconsole requests.\n\n" +); + +static int notify_panic(struct notifier_block *self, unsigned long unused1, + void *ptr) +{ + char *message = ptr; + + if(notify_socket == NULL) return(0); + + mconsole_notify(notify_socket, MCONSOLE_PANIC, message, + strlen(message) + 1); + return(0); +} + +static struct notifier_block panic_exit_notifier = { + .notifier_call = notify_panic, + .next = NULL, + .priority = 1 +}; + +static int add_notifier(void) +{ + notifier_chain_register(&panic_notifier_list, &panic_exit_notifier); + return(0); +} + +__initcall(add_notifier); + +char *mconsole_notify_socket(void) +{ + return(notify_socket); +} + +EXPORT_SYMBOL(mconsole_notify_socket); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/mconsole_user.c b/arch/um/drivers/mconsole_user.c new file mode 100644 index 000000000000..fe5afb13252c --- /dev/null +++ b/arch/um/drivers/mconsole_user.c @@ -0,0 +1,215 @@ +/* + * Copyright (C) 2001 Lennert Buytenhek (buytenh@gnu.org) + * Copyright (C) 2001 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <errno.h> +#include <signal.h> +#include <sys/socket.h> +#include <sys/types.h> +#include <sys/uio.h> +#include <sys/un.h> +#include <unistd.h> +#include "user.h" +#include "mconsole.h" +#include "umid.h" + +static struct mconsole_command commands[] = { + { "version", mconsole_version, MCONSOLE_INTR }, + { "halt", mconsole_halt, MCONSOLE_PROC }, + { "reboot", mconsole_reboot, MCONSOLE_PROC }, + { "config", mconsole_config, MCONSOLE_PROC }, + { "remove", mconsole_remove, MCONSOLE_PROC }, + { "sysrq", mconsole_sysrq, MCONSOLE_INTR }, + { "help", mconsole_help, MCONSOLE_INTR }, + { "cad", mconsole_cad, MCONSOLE_INTR }, + { "stop", mconsole_stop, MCONSOLE_PROC }, + { "go", mconsole_go, MCONSOLE_INTR }, + { "log", mconsole_log, MCONSOLE_INTR }, + { "proc", mconsole_proc, MCONSOLE_PROC }, +}; + +/* Initialized in mconsole_init, which is an initcall */ +char mconsole_socket_name[256]; + +int mconsole_reply_v0(struct mc_request *req, char *reply) +{ + struct iovec iov; + struct msghdr msg; + + iov.iov_base = reply; + iov.iov_len = strlen(reply); + + msg.msg_name = &(req->origin); + msg.msg_namelen = req->originlen; + msg.msg_iov = &iov; + msg.msg_iovlen = 1; + msg.msg_control = NULL; + msg.msg_controllen = 0; + msg.msg_flags = 0; + + return sendmsg(req->originating_fd, &msg, 0); +} + +static struct mconsole_command *mconsole_parse(struct mc_request *req) +{ + struct mconsole_command *cmd; + int i; + + for(i=0;i<sizeof(commands)/sizeof(commands[0]);i++){ + cmd = &commands[i]; + if(!strncmp(req->request.data, cmd->command, + strlen(cmd->command))){ + return(cmd); + } + } + return(NULL); +} + +#define MIN(a,b) ((a)<(b) ? (a):(b)) + +#define STRINGX(x) #x +#define STRING(x) STRINGX(x) + +int mconsole_get_request(int fd, struct mc_request *req) +{ + int len; + + req->originlen = sizeof(req->origin); + req->len = recvfrom(fd, &req->request, sizeof(req->request), 0, + (struct sockaddr *) req->origin, &req->originlen); + if (req->len < 0) + return 0; + + req->originating_fd = fd; + + if(req->request.magic != MCONSOLE_MAGIC){ + /* Unversioned request */ + len = MIN(sizeof(req->request.data) - 1, + strlen((char *) &req->request)); + memmove(req->request.data, &req->request, len); + req->request.data[len] = '\0'; + + req->request.magic = MCONSOLE_MAGIC; + req->request.version = 0; + req->request.len = len; + + mconsole_reply_v0(req, "ERR Version 0 mconsole clients are " + "not supported by this driver"); + return(0); + } + + if(req->request.len >= MCONSOLE_MAX_DATA){ + mconsole_reply(req, "Request too large", 1, 0); + return(0); + } + if(req->request.version != MCONSOLE_VERSION){ + mconsole_reply(req, "This driver only supports version " + STRING(MCONSOLE_VERSION) " clients", 1, 0); + } + + req->request.data[req->request.len] = '\0'; + req->cmd = mconsole_parse(req); + if(req->cmd == NULL){ + mconsole_reply(req, "Unknown command", 1, 0); + return(0); + } + + return(1); +} + +int mconsole_reply(struct mc_request *req, char *str, int err, int more) +{ + struct mconsole_reply reply; + int total, len, n; + + total = strlen(str); + do { + reply.err = err; + + /* err can only be true on the first packet */ + err = 0; + + len = MIN(total, MCONSOLE_MAX_DATA - 1); + + if(len == total) reply.more = more; + else reply.more = 1; + + memcpy(reply.data, str, len); + reply.data[len] = '\0'; + total -= len; + str += len; + reply.len = len + 1; + + len = sizeof(reply) + reply.len - sizeof(reply.data); + + n = sendto(req->originating_fd, &reply, len, 0, + (struct sockaddr *) req->origin, req->originlen); + + if(n < 0) return(-errno); + } while(total > 0); + return(0); +} + +int mconsole_unlink_socket(void) +{ + unlink(mconsole_socket_name); + return 0; +} + +static int notify_sock = -1; + +int mconsole_notify(char *sock_name, int type, const void *data, int len) +{ + struct sockaddr_un target; + struct mconsole_notify packet; + int n, err = 0; + + lock_notify(); + if(notify_sock < 0){ + notify_sock = socket(PF_UNIX, SOCK_DGRAM, 0); + if(notify_sock < 0){ + printk("mconsole_notify - socket failed, errno = %d\n", + errno); + err = -errno; + } + } + unlock_notify(); + + if(err) + return(err); + + target.sun_family = AF_UNIX; + strcpy(target.sun_path, sock_name); + + packet.magic = MCONSOLE_MAGIC; + packet.version = MCONSOLE_VERSION; + packet.type = type; + len = (len > sizeof(packet.data)) ? sizeof(packet.data) : len; + packet.len = len; + memcpy(packet.data, data, len); + + err = 0; + len = sizeof(packet) + packet.len - sizeof(packet.data); + n = sendto(notify_sock, &packet, len, 0, (struct sockaddr *) &target, + sizeof(target)); + if(n < 0){ + printk("mconsole_notify - sendto failed, errno = %d\n", errno); + err = -errno; + } + return(err); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/mmapper_kern.c b/arch/um/drivers/mmapper_kern.c new file mode 100644 index 000000000000..a63231dffe05 --- /dev/null +++ b/arch/um/drivers/mmapper_kern.c @@ -0,0 +1,150 @@ +/* + * arch/um/drivers/mmapper_kern.c + * + * BRIEF MODULE DESCRIPTION + * + * Copyright (C) 2000 RidgeRun, Inc. + * Author: RidgeRun, Inc. + * Greg Lonnon glonnon@ridgerun.com or info@ridgerun.com + * + */ + +#include <linux/types.h> +#include <linux/kdev_t.h> +#include <linux/time.h> +#include <linux/devfs_fs_kernel.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/slab.h> +#include <linux/init.h> +#include <linux/smp_lock.h> +#include <asm/uaccess.h> +#include <asm/irq.h> +#include <asm/pgtable.h> +#include "mem_user.h" +#include "user_util.h" + +/* These are set in mmapper_init, which is called at boot time */ +static unsigned long mmapper_size; +static unsigned long p_buf = 0; +static char *v_buf = NULL; + +static ssize_t +mmapper_read(struct file *file, char *buf, size_t count, loff_t *ppos) +{ + if(*ppos > mmapper_size) + return -EINVAL; + + if(count + *ppos > mmapper_size) + count = count + *ppos - mmapper_size; + + if(count < 0) + return -EINVAL; + + copy_to_user(buf,&v_buf[*ppos],count); + + return count; +} + +static ssize_t +mmapper_write(struct file *file, const char *buf, size_t count, loff_t *ppos) +{ + if(*ppos > mmapper_size) + return -EINVAL; + + if(count + *ppos > mmapper_size) + count = count + *ppos - mmapper_size; + + if(count < 0) + return -EINVAL; + + copy_from_user(&v_buf[*ppos],buf,count); + + return count; +} + +static int +mmapper_ioctl(struct inode *inode, struct file *file, unsigned int cmd, + unsigned long arg) +{ + return(-ENOIOCTLCMD); +} + +static int +mmapper_mmap(struct file *file, struct vm_area_struct * vma) +{ + int ret = -EINVAL; + int size; + + lock_kernel(); + if (vma->vm_pgoff != 0) + goto out; + + size = vma->vm_end - vma->vm_start; + if(size > mmapper_size) return(-EFAULT); + + /* XXX A comment above remap_pfn_range says it should only be + * called when the mm semaphore is held + */ + if (remap_pfn_range(vma, vma->vm_start, p_buf >> PAGE_SHIFT, size, + vma->vm_page_prot)) + goto out; + ret = 0; +out: + unlock_kernel(); + return ret; +} + +static int +mmapper_open(struct inode *inode, struct file *file) +{ + return 0; +} + +static int +mmapper_release(struct inode *inode, struct file *file) +{ + return 0; +} + +static struct file_operations mmapper_fops = { + .owner = THIS_MODULE, + .read = mmapper_read, + .write = mmapper_write, + .ioctl = mmapper_ioctl, + .mmap = mmapper_mmap, + .open = mmapper_open, + .release = mmapper_release, +}; + +static int __init mmapper_init(void) +{ + printk(KERN_INFO "Mapper v0.1\n"); + + v_buf = (char *) find_iomem("mmapper", &mmapper_size); + if(mmapper_size == 0){ + printk(KERN_ERR "mmapper_init - find_iomem failed\n"); + return(0); + } + + p_buf = __pa(v_buf); + + devfs_mk_cdev(MKDEV(30, 0), S_IFCHR|S_IRUGO|S_IWUGO, "mmapper"); + return(0); +} + +static void mmapper_exit(void) +{ +} + +module_init(mmapper_init); +module_exit(mmapper_exit); + +MODULE_AUTHOR("Greg Lonnon <glonnon@ridgerun.com>"); +MODULE_DESCRIPTION("DSPLinux simulator mmapper driver"); +/* + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c new file mode 100644 index 000000000000..4eeaf88c1e97 --- /dev/null +++ b/arch/um/drivers/net_kern.c @@ -0,0 +1,896 @@ +/* + * Copyright (C) 2001 Lennert Buytenhek (buytenh@gnu.org) and + * James Leu (jleu@mindspring.net). + * Copyright (C) 2001 by various other people who didn't put their name here. + * Licensed under the GPL. + */ + +#include "linux/config.h" +#include "linux/kernel.h" +#include "linux/netdevice.h" +#include "linux/rtnetlink.h" +#include "linux/skbuff.h" +#include "linux/socket.h" +#include "linux/spinlock.h" +#include "linux/module.h" +#include "linux/init.h" +#include "linux/etherdevice.h" +#include "linux/list.h" +#include "linux/inetdevice.h" +#include "linux/ctype.h" +#include "linux/bootmem.h" +#include "linux/ethtool.h" +#include "asm/uaccess.h" +#include "user_util.h" +#include "kern_util.h" +#include "net_kern.h" +#include "net_user.h" +#include "mconsole_kern.h" +#include "init.h" +#include "irq_user.h" +#include "irq_kern.h" + +#define DRIVER_NAME "uml-netdev" + +static DEFINE_SPINLOCK(opened_lock); +LIST_HEAD(opened); + +static int uml_net_rx(struct net_device *dev) +{ + struct uml_net_private *lp = dev->priv; + int pkt_len; + struct sk_buff *skb; + + /* If we can't allocate memory, try again next round. */ + skb = dev_alloc_skb(dev->mtu); + if (skb == NULL) { + lp->stats.rx_dropped++; + return 0; + } + + skb->dev = dev; + skb_put(skb, dev->mtu); + skb->mac.raw = skb->data; + pkt_len = (*lp->read)(lp->fd, &skb, lp); + + if (pkt_len > 0) { + skb_trim(skb, pkt_len); + skb->protocol = (*lp->protocol)(skb); + netif_rx(skb); + + lp->stats.rx_bytes += skb->len; + lp->stats.rx_packets++; + return pkt_len; + } + + kfree_skb(skb); + return pkt_len; +} + +irqreturn_t uml_net_interrupt(int irq, void *dev_id, struct pt_regs *regs) +{ + struct net_device *dev = dev_id; + struct uml_net_private *lp = dev->priv; + int err; + + if(!netif_running(dev)) + return(IRQ_NONE); + + spin_lock(&lp->lock); + while((err = uml_net_rx(dev)) > 0) ; + if(err < 0) { + printk(KERN_ERR + "Device '%s' read returned %d, shutting it down\n", + dev->name, err); + dev_close(dev); + goto out; + } + reactivate_fd(lp->fd, UM_ETH_IRQ); + + out: + spin_unlock(&lp->lock); + return(IRQ_HANDLED); +} + +static int uml_net_open(struct net_device *dev) +{ + struct uml_net_private *lp = dev->priv; + char addr[sizeof("255.255.255.255\0")]; + int err; + + spin_lock(&lp->lock); + + if(lp->fd >= 0){ + err = -ENXIO; + goto out; + } + + if(!lp->have_mac){ + dev_ip_addr(dev, addr, &lp->mac[2]); + set_ether_mac(dev, lp->mac); + } + + lp->fd = (*lp->open)(&lp->user); + if(lp->fd < 0){ + err = lp->fd; + goto out; + } + + err = um_request_irq(dev->irq, lp->fd, IRQ_READ, uml_net_interrupt, + SA_INTERRUPT | SA_SHIRQ, dev->name, dev); + if(err != 0){ + printk(KERN_ERR "uml_net_open: failed to get irq(%d)\n", err); + if(lp->close != NULL) (*lp->close)(lp->fd, &lp->user); + lp->fd = -1; + err = -ENETUNREACH; + } + + lp->tl.data = (unsigned long) &lp->user; + netif_start_queue(dev); + + /* clear buffer - it can happen that the host side of the interface + * is full when we get here. In this case, new data is never queued, + * SIGIOs never arrive, and the net never works. + */ + while((err = uml_net_rx(dev)) > 0) ; + + out: + spin_unlock(&lp->lock); + return(err); +} + +static int uml_net_close(struct net_device *dev) +{ + struct uml_net_private *lp = dev->priv; + + netif_stop_queue(dev); + spin_lock(&lp->lock); + + free_irq_by_irq_and_dev(dev->irq, dev); + free_irq(dev->irq, dev); + if(lp->close != NULL) + (*lp->close)(lp->fd, &lp->user); + lp->fd = -1; + + spin_unlock(&lp->lock); + return 0; +} + +static int uml_net_start_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct uml_net_private *lp = dev->priv; + unsigned long flags; + int len; + + netif_stop_queue(dev); + + spin_lock_irqsave(&lp->lock, flags); + + len = (*lp->write)(lp->fd, &skb, lp); + + if(len == skb->len) { + lp->stats.tx_packets++; + lp->stats.tx_bytes += skb->len; + dev->trans_start = jiffies; + netif_start_queue(dev); + + /* this is normally done in the interrupt when tx finishes */ + netif_wake_queue(dev); + } + else if(len == 0){ + netif_start_queue(dev); + lp->stats.tx_dropped++; + } + else { + netif_start_queue(dev); + printk(KERN_ERR "uml_net_start_xmit: failed(%d)\n", len); + } + + spin_unlock_irqrestore(&lp->lock, flags); + + dev_kfree_skb(skb); + + return 0; +} + +static struct net_device_stats *uml_net_get_stats(struct net_device *dev) +{ + struct uml_net_private *lp = dev->priv; + return &lp->stats; +} + +static void uml_net_set_multicast_list(struct net_device *dev) +{ + if (dev->flags & IFF_PROMISC) return; + else if (dev->mc_count) dev->flags |= IFF_ALLMULTI; + else dev->flags &= ~IFF_ALLMULTI; +} + +static void uml_net_tx_timeout(struct net_device *dev) +{ + dev->trans_start = jiffies; + netif_wake_queue(dev); +} + +static int uml_net_set_mac(struct net_device *dev, void *addr) +{ + struct uml_net_private *lp = dev->priv; + struct sockaddr *hwaddr = addr; + + spin_lock(&lp->lock); + memcpy(dev->dev_addr, hwaddr->sa_data, ETH_ALEN); + spin_unlock(&lp->lock); + + return(0); +} + +static int uml_net_change_mtu(struct net_device *dev, int new_mtu) +{ + struct uml_net_private *lp = dev->priv; + int err = 0; + + spin_lock(&lp->lock); + + new_mtu = (*lp->set_mtu)(new_mtu, &lp->user); + if(new_mtu < 0){ + err = new_mtu; + goto out; + } + + dev->mtu = new_mtu; + + out: + spin_unlock(&lp->lock); + return err; +} + +static int uml_net_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) +{ + static const struct ethtool_drvinfo info = { + .cmd = ETHTOOL_GDRVINFO, + .driver = DRIVER_NAME, + .version = "42", + }; + void *useraddr; + u32 ethcmd; + + switch (cmd) { + case SIOCETHTOOL: + useraddr = ifr->ifr_data; + if (copy_from_user(ðcmd, useraddr, sizeof(ethcmd))) + return -EFAULT; + switch (ethcmd) { + case ETHTOOL_GDRVINFO: + if (copy_to_user(useraddr, &info, sizeof(info))) + return -EFAULT; + return 0; + default: + return -EOPNOTSUPP; + } + default: + return -EINVAL; + } +} + +void uml_net_user_timer_expire(unsigned long _conn) +{ +#ifdef undef + struct connection *conn = (struct connection *)_conn; + + dprintk(KERN_INFO "uml_net_user_timer_expire [%p]\n", conn); + do_connect(conn); +#endif +} + +static DEFINE_SPINLOCK(devices_lock); +static struct list_head devices = LIST_HEAD_INIT(devices); + +static struct device_driver uml_net_driver = { + .name = DRIVER_NAME, + .bus = &platform_bus_type, +}; +static int driver_registered; + +static int eth_configure(int n, void *init, char *mac, + struct transport *transport) +{ + struct uml_net *device; + struct net_device *dev; + struct uml_net_private *lp; + int save, err, size; + + size = transport->private_size + sizeof(struct uml_net_private) + + sizeof(((struct uml_net_private *) 0)->user); + + device = kmalloc(sizeof(*device), GFP_KERNEL); + if (device == NULL) { + printk(KERN_ERR "eth_configure failed to allocate uml_net\n"); + return(1); + } + + memset(device, 0, sizeof(*device)); + INIT_LIST_HEAD(&device->list); + device->index = n; + + spin_lock(&devices_lock); + list_add(&device->list, &devices); + spin_unlock(&devices_lock); + + if (setup_etheraddr(mac, device->mac)) + device->have_mac = 1; + + printk(KERN_INFO "Netdevice %d ", n); + if (device->have_mac) + printk("(%02x:%02x:%02x:%02x:%02x:%02x) ", + device->mac[0], device->mac[1], + device->mac[2], device->mac[3], + device->mac[4], device->mac[5]); + printk(": "); + dev = alloc_etherdev(size); + if (dev == NULL) { + printk(KERN_ERR "eth_configure: failed to allocate device\n"); + return 1; + } + + /* sysfs register */ + if (!driver_registered) { + driver_register(¨_net_driver); + driver_registered = 1; + } + device->pdev.id = n; + device->pdev.name = DRIVER_NAME; + platform_device_register(&device->pdev); + SET_NETDEV_DEV(dev,&device->pdev.dev); + + /* If this name ends up conflicting with an existing registered + * netdevice, that is OK, register_netdev{,ice}() will notice this + * and fail. + */ + snprintf(dev->name, sizeof(dev->name), "eth%d", n); + device->dev = dev; + + (*transport->kern->init)(dev, init); + + dev->mtu = transport->user->max_packet; + dev->open = uml_net_open; + dev->hard_start_xmit = uml_net_start_xmit; + dev->stop = uml_net_close; + dev->get_stats = uml_net_get_stats; + dev->set_multicast_list = uml_net_set_multicast_list; + dev->tx_timeout = uml_net_tx_timeout; + dev->set_mac_address = uml_net_set_mac; + dev->change_mtu = uml_net_change_mtu; + dev->do_ioctl = uml_net_ioctl; + dev->watchdog_timeo = (HZ >> 1); + dev->irq = UM_ETH_IRQ; + + rtnl_lock(); + err = register_netdevice(dev); + rtnl_unlock(); + if (err) { + device->dev = NULL; + /* XXX: should we call ->remove() here? */ + free_netdev(dev); + return 1; + } + lp = dev->priv; + + /* lp.user is the first four bytes of the transport data, which + * has already been initialized. This structure assignment will + * overwrite that, so we make sure that .user gets overwritten with + * what it already has. + */ + save = lp->user[0]; + *lp = ((struct uml_net_private) + { .list = LIST_HEAD_INIT(lp->list), + .dev = dev, + .fd = -1, + .mac = { 0xfe, 0xfd, 0x0, 0x0, 0x0, 0x0}, + .have_mac = device->have_mac, + .protocol = transport->kern->protocol, + .open = transport->user->open, + .close = transport->user->close, + .remove = transport->user->remove, + .read = transport->kern->read, + .write = transport->kern->write, + .add_address = transport->user->add_address, + .delete_address = transport->user->delete_address, + .set_mtu = transport->user->set_mtu, + .user = { save } }); + + init_timer(&lp->tl); + spin_lock_init(&lp->lock); + lp->tl.function = uml_net_user_timer_expire; + if (lp->have_mac) + memcpy(lp->mac, device->mac, sizeof(lp->mac)); + + if (transport->user->init) + (*transport->user->init)(&lp->user, dev); + + if (device->have_mac) + set_ether_mac(dev, device->mac); + + spin_lock(&opened_lock); + list_add(&lp->list, &opened); + spin_unlock(&opened_lock); + + return(0); +} + +static struct uml_net *find_device(int n) +{ + struct uml_net *device; + struct list_head *ele; + + spin_lock(&devices_lock); + list_for_each(ele, &devices){ + device = list_entry(ele, struct uml_net, list); + if(device->index == n) + goto out; + } + device = NULL; + out: + spin_unlock(&devices_lock); + return(device); +} + +static int eth_parse(char *str, int *index_out, char **str_out) +{ + char *end; + int n; + + n = simple_strtoul(str, &end, 0); + if(end == str){ + printk(KERN_ERR "eth_setup: Failed to parse '%s'\n", str); + return(1); + } + if(n < 0){ + printk(KERN_ERR "eth_setup: device %d is negative\n", n); + return(1); + } + str = end; + if(*str != '='){ + printk(KERN_ERR + "eth_setup: expected '=' after device number\n"); + return(1); + } + str++; + if(find_device(n)){ + printk(KERN_ERR "eth_setup: Device %d already configured\n", + n); + return(1); + } + if(index_out) *index_out = n; + *str_out = str; + return(0); +} + +struct eth_init { + struct list_head list; + char *init; + int index; +}; + +/* Filled in at boot time. Will need locking if the transports become + * modular. + */ +struct list_head transports = LIST_HEAD_INIT(transports); + +/* Filled in during early boot */ +struct list_head eth_cmd_line = LIST_HEAD_INIT(eth_cmd_line); + +static int check_transport(struct transport *transport, char *eth, int n, + void **init_out, char **mac_out) +{ + int len; + + len = strlen(transport->name); + if(strncmp(eth, transport->name, len)) + return(0); + + eth += len; + if(*eth == ',') + eth++; + else if(*eth != '\0') + return(0); + + *init_out = kmalloc(transport->setup_size, GFP_KERNEL); + if(*init_out == NULL) + return(1); + + if(!transport->setup(eth, mac_out, *init_out)){ + kfree(*init_out); + *init_out = NULL; + } + return(1); +} + +void register_transport(struct transport *new) +{ + struct list_head *ele, *next; + struct eth_init *eth; + void *init; + char *mac = NULL; + int match; + + list_add(&new->list, &transports); + + list_for_each_safe(ele, next, ð_cmd_line){ + eth = list_entry(ele, struct eth_init, list); + match = check_transport(new, eth->init, eth->index, &init, + &mac); + if(!match) + continue; + else if(init != NULL){ + eth_configure(eth->index, init, mac, new); + kfree(init); + } + list_del(ð->list); + } +} + +static int eth_setup_common(char *str, int index) +{ + struct list_head *ele; + struct transport *transport; + void *init; + char *mac = NULL; + + list_for_each(ele, &transports){ + transport = list_entry(ele, struct transport, list); + if(!check_transport(transport, str, index, &init, &mac)) + continue; + if(init != NULL){ + eth_configure(index, init, mac, transport); + kfree(init); + } + return(1); + } + return(0); +} + +static int eth_setup(char *str) +{ + struct eth_init *new; + int n, err; + + err = eth_parse(str, &n, &str); + if(err) return(1); + + new = alloc_bootmem(sizeof(new)); + if (new == NULL){ + printk("eth_init : alloc_bootmem failed\n"); + return(1); + } + + INIT_LIST_HEAD(&new->list); + new->index = n; + new->init = str; + + list_add_tail(&new->list, ð_cmd_line); + return(1); +} + +__setup("eth", eth_setup); +__uml_help(eth_setup, +"eth[0-9]+=<transport>,<options>\n" +" Configure a network device.\n\n" +); + +#if 0 +static int eth_init(void) +{ + struct list_head *ele, *next; + struct eth_init *eth; + + list_for_each_safe(ele, next, ð_cmd_line){ + eth = list_entry(ele, struct eth_init, list); + + if(eth_setup_common(eth->init, eth->index)) + list_del(ð->list); + } + + return(1); +} +__initcall(eth_init); +#endif + +static int net_config(char *str) +{ + int n, err; + + err = eth_parse(str, &n, &str); + if(err) return(err); + + str = uml_strdup(str); + if(str == NULL){ + printk(KERN_ERR "net_config failed to strdup string\n"); + return(-1); + } + err = !eth_setup_common(str, n); + if(err) + kfree(str); + return(err); +} + +static int net_remove(char *str) +{ + struct uml_net *device; + struct net_device *dev; + struct uml_net_private *lp; + char *end; + int n; + + n = simple_strtoul(str, &end, 0); + if((*end != '\0') || (end == str)) + return(-1); + + device = find_device(n); + if(device == NULL) + return(0); + + dev = device->dev; + lp = dev->priv; + if(lp->fd > 0) return(-1); + if(lp->remove != NULL) (*lp->remove)(&lp->user); + unregister_netdev(dev); + platform_device_unregister(&device->pdev); + + list_del(&device->list); + kfree(device); + free_netdev(dev); + return(0); +} + +static struct mc_device net_mc = { + .name = "eth", + .config = net_config, + .get_config = NULL, + .remove = net_remove, +}; + +static int uml_inetaddr_event(struct notifier_block *this, unsigned long event, + void *ptr) +{ + struct in_ifaddr *ifa = ptr; + u32 addr = ifa->ifa_address; + u32 netmask = ifa->ifa_mask; + struct net_device *dev = ifa->ifa_dev->dev; + struct uml_net_private *lp; + void (*proc)(unsigned char *, unsigned char *, void *); + unsigned char addr_buf[4], netmask_buf[4]; + + if(dev->open != uml_net_open) return(NOTIFY_DONE); + + lp = dev->priv; + + proc = NULL; + switch (event){ + case NETDEV_UP: + proc = lp->add_address; + break; + case NETDEV_DOWN: + proc = lp->delete_address; + break; + } + if(proc != NULL){ + addr_buf[0] = addr & 0xff; + addr_buf[1] = (addr >> 8) & 0xff; + addr_buf[2] = (addr >> 16) & 0xff; + addr_buf[3] = addr >> 24; + netmask_buf[0] = netmask & 0xff; + netmask_buf[1] = (netmask >> 8) & 0xff; + netmask_buf[2] = (netmask >> 16) & 0xff; + netmask_buf[3] = netmask >> 24; + (*proc)(addr_buf, netmask_buf, &lp->user); + } + return(NOTIFY_DONE); +} + +struct notifier_block uml_inetaddr_notifier = { + .notifier_call = uml_inetaddr_event, +}; + +static int uml_net_init(void) +{ + struct list_head *ele; + struct uml_net_private *lp; + struct in_device *ip; + struct in_ifaddr *in; + + mconsole_register_dev(&net_mc); + register_inetaddr_notifier(¨_inetaddr_notifier); + + /* Devices may have been opened already, so the uml_inetaddr_notifier + * didn't get a chance to run for them. This fakes it so that + * addresses which have already been set up get handled properly. + */ + list_for_each(ele, &opened){ + lp = list_entry(ele, struct uml_net_private, list); + ip = lp->dev->ip_ptr; + if(ip == NULL) continue; + in = ip->ifa_list; + while(in != NULL){ + uml_inetaddr_event(NULL, NETDEV_UP, in); + in = in->ifa_next; + } + } + + return(0); +} + +__initcall(uml_net_init); + +static void close_devices(void) +{ + struct list_head *ele; + struct uml_net_private *lp; + + list_for_each(ele, &opened){ + lp = list_entry(ele, struct uml_net_private, list); + if((lp->close != NULL) && (lp->fd >= 0)) + (*lp->close)(lp->fd, &lp->user); + if(lp->remove != NULL) (*lp->remove)(&lp->user); + } +} + +__uml_exitcall(close_devices); + +int setup_etheraddr(char *str, unsigned char *addr) +{ + char *end; + int i; + + if(str == NULL) + return(0); + for(i=0;i<6;i++){ + addr[i] = simple_strtoul(str, &end, 16); + if((end == str) || + ((*end != ':') && (*end != ',') && (*end != '\0'))){ + printk(KERN_ERR + "setup_etheraddr: failed to parse '%s' " + "as an ethernet address\n", str); + return(0); + } + str = end + 1; + } + if(addr[0] & 1){ + printk(KERN_ERR + "Attempt to assign a broadcast ethernet address to a " + "device disallowed\n"); + return(0); + } + return(1); +} + +void dev_ip_addr(void *d, char *buf, char *bin_buf) +{ + struct net_device *dev = d; + struct in_device *ip = dev->ip_ptr; + struct in_ifaddr *in; + u32 addr; + + if((ip == NULL) || ((in = ip->ifa_list) == NULL)){ + printk(KERN_WARNING "dev_ip_addr - device not assigned an " + "IP address\n"); + return; + } + addr = in->ifa_address; + sprintf(buf, "%d.%d.%d.%d", addr & 0xff, (addr >> 8) & 0xff, + (addr >> 16) & 0xff, addr >> 24); + if(bin_buf){ + bin_buf[0] = addr & 0xff; + bin_buf[1] = (addr >> 8) & 0xff; + bin_buf[2] = (addr >> 16) & 0xff; + bin_buf[3] = addr >> 24; + } +} + +void set_ether_mac(void *d, unsigned char *addr) +{ + struct net_device *dev = d; + + memcpy(dev->dev_addr, addr, ETH_ALEN); +} + +struct sk_buff *ether_adjust_skb(struct sk_buff *skb, int extra) +{ + if((skb != NULL) && (skb_tailroom(skb) < extra)){ + struct sk_buff *skb2; + + skb2 = skb_copy_expand(skb, 0, extra, GFP_ATOMIC); + dev_kfree_skb(skb); + skb = skb2; + } + if(skb != NULL) skb_put(skb, extra); + return(skb); +} + +void iter_addresses(void *d, void (*cb)(unsigned char *, unsigned char *, + void *), + void *arg) +{ + struct net_device *dev = d; + struct in_device *ip = dev->ip_ptr; + struct in_ifaddr *in; + unsigned char address[4], netmask[4]; + + if(ip == NULL) return; + in = ip->ifa_list; + while(in != NULL){ + address[0] = in->ifa_address & 0xff; + address[1] = (in->ifa_address >> 8) & 0xff; + address[2] = (in->ifa_address >> 16) & 0xff; + address[3] = in->ifa_address >> 24; + netmask[0] = in->ifa_mask & 0xff; + netmask[1] = (in->ifa_mask >> 8) & 0xff; + netmask[2] = (in->ifa_mask >> 16) & 0xff; + netmask[3] = in->ifa_mask >> 24; + (*cb)(address, netmask, arg); + in = in->ifa_next; + } +} + +int dev_netmask(void *d, void *m) +{ + struct net_device *dev = d; + struct in_device *ip = dev->ip_ptr; + struct in_ifaddr *in; + __u32 *mask_out = m; + + if(ip == NULL) + return(1); + + in = ip->ifa_list; + if(in == NULL) + return(1); + + *mask_out = in->ifa_mask; + return(0); +} + +void *get_output_buffer(int *len_out) +{ + void *ret; + + ret = (void *) __get_free_pages(GFP_KERNEL, 0); + if(ret) *len_out = PAGE_SIZE; + else *len_out = 0; + return(ret); +} + +void free_output_buffer(void *buffer) +{ + free_pages((unsigned long) buffer, 0); +} + +int tap_setup_common(char *str, char *type, char **dev_name, char **mac_out, + char **gate_addr) +{ + char *remain; + + remain = split_if_spec(str, dev_name, mac_out, gate_addr, NULL); + if(remain != NULL){ + printk("tap_setup_common - Extra garbage on specification : " + "'%s'\n", remain); + return(1); + } + + return(0); +} + +unsigned short eth_protocol(struct sk_buff *skb) +{ + return(eth_type_trans(skb, skb->dev)); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/net_user.c b/arch/um/drivers/net_user.c new file mode 100644 index 000000000000..47229fe4a813 --- /dev/null +++ b/arch/um/drivers/net_user.c @@ -0,0 +1,255 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stddef.h> +#include <stdarg.h> +#include <unistd.h> +#include <stdio.h> +#include <errno.h> +#include <stdlib.h> +#include <string.h> +#include <sys/socket.h> +#include <sys/wait.h> +#include "user.h" +#include "user_util.h" +#include "kern_util.h" +#include "net_user.h" +#include "helper.h" +#include "os.h" + +int tap_open_common(void *dev, char *gate_addr) +{ + int tap_addr[4]; + + if(gate_addr == NULL) return(0); + if(sscanf(gate_addr, "%d.%d.%d.%d", &tap_addr[0], + &tap_addr[1], &tap_addr[2], &tap_addr[3]) != 4){ + printk("Invalid tap IP address - '%s'\n", gate_addr); + return(-EINVAL); + } + return(0); +} + +void tap_check_ips(char *gate_addr, char *eth_addr) +{ + int tap_addr[4]; + + if((gate_addr != NULL) && + (sscanf(gate_addr, "%d.%d.%d.%d", &tap_addr[0], + &tap_addr[1], &tap_addr[2], &tap_addr[3]) == 4) && + (eth_addr[0] == tap_addr[0]) && + (eth_addr[1] == tap_addr[1]) && + (eth_addr[2] == tap_addr[2]) && + (eth_addr[3] == tap_addr[3])){ + printk("The tap IP address and the UML eth IP address" + " must be different\n"); + } +} + +void read_output(int fd, char *output, int len) +{ + int remain, n, actual; + char c; + + if(output == NULL){ + output = &c; + len = sizeof(c); + } + + *output = '\0'; + n = os_read_file(fd, &remain, sizeof(remain)); + if(n != sizeof(remain)){ + printk("read_output - read of length failed, err = %d\n", -n); + return; + } + + while(remain != 0){ + n = (remain < len) ? remain : len; + actual = os_read_file(fd, output, n); + if(actual != n){ + printk("read_output - read of data failed, " + "err = %d\n", -actual); + return; + } + remain -= actual; + } + return; +} + +int net_read(int fd, void *buf, int len) +{ + int n; + + n = os_read_file(fd, buf, len); + + if(n == -EAGAIN) + return(0); + else if(n == 0) + return(-ENOTCONN); + return(n); +} + +int net_recvfrom(int fd, void *buf, int len) +{ + int n; + + while(((n = recvfrom(fd, buf, len, 0, NULL, NULL)) < 0) && + (errno == EINTR)) ; + + if(n < 0){ + if(errno == EAGAIN) return(0); + return(-errno); + } + else if(n == 0) return(-ENOTCONN); + return(n); +} + +int net_write(int fd, void *buf, int len) +{ + int n; + + n = os_write_file(fd, buf, len); + + if(n == -EAGAIN) + return(0); + else if(n == 0) + return(-ENOTCONN); + return(n); +} + +int net_send(int fd, void *buf, int len) +{ + int n; + + while(((n = send(fd, buf, len, 0)) < 0) && (errno == EINTR)) ; + if(n < 0){ + if(errno == EAGAIN) return(0); + return(-errno); + } + else if(n == 0) return(-ENOTCONN); + return(n); +} + +int net_sendto(int fd, void *buf, int len, void *to, int sock_len) +{ + int n; + + while(((n = sendto(fd, buf, len, 0, (struct sockaddr *) to, + sock_len)) < 0) && (errno == EINTR)) ; + if(n < 0){ + if(errno == EAGAIN) return(0); + return(-errno); + } + else if(n == 0) return(-ENOTCONN); + return(n); +} + +struct change_pre_exec_data { + int close_me; + int stdout; +}; + +static void change_pre_exec(void *arg) +{ + struct change_pre_exec_data *data = arg; + + os_close_file(data->close_me); + dup2(data->stdout, 1); +} + +static int change_tramp(char **argv, char *output, int output_len) +{ + int pid, fds[2], err; + struct change_pre_exec_data pe_data; + + err = os_pipe(fds, 1, 0); + if(err < 0){ + printk("change_tramp - pipe failed, err = %d\n", -err); + return(err); + } + pe_data.close_me = fds[0]; + pe_data.stdout = fds[1]; + pid = run_helper(change_pre_exec, &pe_data, argv, NULL); + + read_output(fds[0], output, output_len); + os_close_file(fds[0]); + os_close_file(fds[1]); + + if (pid > 0) + CATCH_EINTR(err = waitpid(pid, NULL, 0)); + return(pid); +} + +static void change(char *dev, char *what, unsigned char *addr, + unsigned char *netmask) +{ + char addr_buf[sizeof("255.255.255.255\0")]; + char netmask_buf[sizeof("255.255.255.255\0")]; + char version[sizeof("nnnnn\0")]; + char *argv[] = { "uml_net", version, what, dev, addr_buf, + netmask_buf, NULL }; + char *output; + int output_len, pid; + + sprintf(version, "%d", UML_NET_VERSION); + sprintf(addr_buf, "%d.%d.%d.%d", addr[0], addr[1], addr[2], addr[3]); + sprintf(netmask_buf, "%d.%d.%d.%d", netmask[0], netmask[1], + netmask[2], netmask[3]); + + output_len = page_size(); + output = um_kmalloc(output_len); + if(output == NULL) + printk("change : failed to allocate output buffer\n"); + + pid = change_tramp(argv, output, output_len); + if(pid < 0) return; + + if(output != NULL){ + printk("%s", output); + kfree(output); + } +} + +void open_addr(unsigned char *addr, unsigned char *netmask, void *arg) +{ + change(arg, "add", addr, netmask); +} + +void close_addr(unsigned char *addr, unsigned char *netmask, void *arg) +{ + change(arg, "del", addr, netmask); +} + +char *split_if_spec(char *str, ...) +{ + char **arg, *end; + va_list ap; + + va_start(ap, str); + while((arg = va_arg(ap, char **)) != NULL){ + if(*str == '\0') + return(NULL); + end = strchr(str, ','); + if(end != str) + *arg = str; + if(end == NULL) + return(NULL); + *end++ = '\0'; + str = end; + } + va_end(ap); + return(str); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/null.c b/arch/um/drivers/null.c new file mode 100644 index 000000000000..14cc5f78398a --- /dev/null +++ b/arch/um/drivers/null.c @@ -0,0 +1,56 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <errno.h> +#include "chan_user.h" +#include "os.h" + +static int null_chan; + +static void *null_init(char *str, int device, struct chan_opts *opts) +{ + return(&null_chan); +} + +static int null_open(int input, int output, int primary, void *d, + char **dev_out) +{ + *dev_out = NULL; + return(os_open_file(DEV_NULL, of_rdwr(OPENFLAGS()), 0)); +} + +static int null_read(int fd, char *c_out, void *unused) +{ + return(-ENODEV); +} + +static void null_free(void *data) +{ +} + +struct chan_ops null_ops = { + .type = "null", + .init = null_init, + .open = null_open, + .close = generic_close, + .read = null_read, + .write = generic_write, + .console_write = generic_console_write, + .window_size = generic_window_size, + .free = null_free, + .winch = 0, +}; + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/pcap_kern.c b/arch/um/drivers/pcap_kern.c new file mode 100644 index 000000000000..07c80f2156ef --- /dev/null +++ b/arch/um/drivers/pcap_kern.c @@ -0,0 +1,123 @@ +/* + * Copyright (C) 2002 Jeff Dike <jdike@karaya.com> + * Licensed under the GPL. + */ + +#include "linux/init.h" +#include "linux/netdevice.h" +#include "linux/etherdevice.h" +#include "net_kern.h" +#include "net_user.h" +#include "pcap_user.h" + +struct pcap_init { + char *host_if; + int promisc; + int optimize; + char *filter; +}; + +void pcap_init(struct net_device *dev, void *data) +{ + struct uml_net_private *pri; + struct pcap_data *ppri; + struct pcap_init *init = data; + + pri = dev->priv; + ppri = (struct pcap_data *) pri->user; + ppri->host_if = init->host_if; + ppri->promisc = init->promisc; + ppri->optimize = init->optimize; + ppri->filter = init->filter; +} + +static int pcap_read(int fd, struct sk_buff **skb, + struct uml_net_private *lp) +{ + *skb = ether_adjust_skb(*skb, ETH_HEADER_OTHER); + if(*skb == NULL) return(-ENOMEM); + return(pcap_user_read(fd, (*skb)->mac.raw, + (*skb)->dev->mtu + ETH_HEADER_OTHER, + (struct pcap_data *) &lp->user)); +} + +static int pcap_write(int fd, struct sk_buff **skb, struct uml_net_private *lp) +{ + return(-EPERM); +} + +static struct net_kern_info pcap_kern_info = { + .init = pcap_init, + .protocol = eth_protocol, + .read = pcap_read, + .write = pcap_write, +}; + +int pcap_setup(char *str, char **mac_out, void *data) +{ + struct pcap_init *init = data; + char *remain, *host_if = NULL, *options[2] = { NULL, NULL }; + int i; + + *init = ((struct pcap_init) + { .host_if = "eth0", + .promisc = 1, + .optimize = 0, + .filter = NULL }); + + remain = split_if_spec(str, &host_if, &init->filter, + &options[0], &options[1], NULL); + if(remain != NULL){ + printk(KERN_ERR "pcap_setup - Extra garbage on " + "specification : '%s'\n", remain); + return(0); + } + + if(host_if != NULL) + init->host_if = host_if; + + for(i = 0; i < sizeof(options)/sizeof(options[0]); i++){ + if(options[i] == NULL) + continue; + if(!strcmp(options[i], "promisc")) + init->promisc = 1; + else if(!strcmp(options[i], "nopromisc")) + init->promisc = 0; + else if(!strcmp(options[i], "optimize")) + init->optimize = 1; + else if(!strcmp(options[i], "nooptimize")) + init->optimize = 0; + else printk("pcap_setup : bad option - '%s'\n", options[i]); + } + + return(1); +} + +static struct transport pcap_transport = { + .list = LIST_HEAD_INIT(pcap_transport.list), + .name = "pcap", + .setup = pcap_setup, + .user = &pcap_user_info, + .kern = &pcap_kern_info, + .private_size = sizeof(struct pcap_data), + .setup_size = sizeof(struct pcap_init), +}; + +static int register_pcap(void) +{ + register_transport(&pcap_transport); + return(1); +} + +__initcall(register_pcap); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/pcap_user.c b/arch/um/drivers/pcap_user.c new file mode 100644 index 000000000000..edfcb29273e1 --- /dev/null +++ b/arch/um/drivers/pcap_user.c @@ -0,0 +1,143 @@ +/* + * Copyright (C) 2002 Jeff Dike <jdike@karaya.com> + * Licensed under the GPL. + */ + +#include <unistd.h> +#include <stdlib.h> +#include <string.h> +#include <errno.h> +#include <pcap.h> +#include <asm/types.h> +#include "net_user.h" +#include "pcap_user.h" +#include "user.h" + +#define MAX_PACKET (ETH_MAX_PACKET + ETH_HEADER_OTHER) + +#define PCAP_FD(p) (*(int *)(p)) + +static void pcap_user_init(void *data, void *dev) +{ + struct pcap_data *pri = data; + pcap_t *p; + char errors[PCAP_ERRBUF_SIZE]; + + p = pcap_open_live(pri->host_if, MAX_PACKET, pri->promisc, 0, errors); + if(p == NULL){ + printk("pcap_user_init : pcap_open_live failed - '%s'\n", + errors); + return; + } + + pri->dev = dev; + pri->pcap = p; +} + +static int pcap_open(void *data) +{ + struct pcap_data *pri = data; + __u32 netmask; + int err; + + if(pri->pcap == NULL) + return(-ENODEV); + + if(pri->filter != NULL){ + err = dev_netmask(pri->dev, &netmask); + if(err < 0){ + printk("pcap_open : dev_netmask failed\n"); + return(-EIO); + } + + pri->compiled = um_kmalloc(sizeof(struct bpf_program)); + if(pri->compiled == NULL){ + printk("pcap_open : kmalloc failed\n"); + return(-ENOMEM); + } + + err = pcap_compile(pri->pcap, + (struct bpf_program *) pri->compiled, + pri->filter, pri->optimize, netmask); + if(err < 0){ + printk("pcap_open : pcap_compile failed - '%s'\n", + pcap_geterr(pri->pcap)); + return(-EIO); + } + + err = pcap_setfilter(pri->pcap, pri->compiled); + if(err < 0){ + printk("pcap_open : pcap_setfilter failed - '%s'\n", + pcap_geterr(pri->pcap)); + return(-EIO); + } + } + + return(PCAP_FD(pri->pcap)); +} + +static void pcap_remove(void *data) +{ + struct pcap_data *pri = data; + + if(pri->compiled != NULL) + pcap_freecode(pri->compiled); + + pcap_close(pri->pcap); +} + +struct pcap_handler_data { + char *buffer; + int len; +}; + +static void handler(u_char *data, const struct pcap_pkthdr *header, + const u_char *packet) +{ + int len; + + struct pcap_handler_data *hdata = (struct pcap_handler_data *) data; + + len = hdata->len < header->caplen ? hdata->len : header->caplen; + memcpy(hdata->buffer, packet, len); + hdata->len = len; +} + +int pcap_user_read(int fd, void *buffer, int len, struct pcap_data *pri) +{ + struct pcap_handler_data hdata = ((struct pcap_handler_data) + { .buffer = buffer, + .len = len }); + int n; + + n = pcap_dispatch(pri->pcap, 1, handler, (u_char *) &hdata); + if(n < 0){ + printk("pcap_dispatch failed - %s\n", pcap_geterr(pri->pcap)); + return(-EIO); + } + else if(n == 0) + return(0); + return(hdata.len); +} + +struct net_user_info pcap_user_info = { + .init = pcap_user_init, + .open = pcap_open, + .close = NULL, + .remove = pcap_remove, + .set_mtu = NULL, + .add_address = NULL, + .delete_address = NULL, + .max_packet = MAX_PACKET - ETH_HEADER_OTHER +}; + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/pcap_user.h b/arch/um/drivers/pcap_user.h new file mode 100644 index 000000000000..58f9f6a1420f --- /dev/null +++ b/arch/um/drivers/pcap_user.h @@ -0,0 +1,31 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "net_user.h" + +struct pcap_data { + char *host_if; + int promisc; + int optimize; + char *filter; + void *compiled; + void *pcap; + void *dev; +}; + +extern struct net_user_info pcap_user_info; + +extern int pcap_user_read(int fd, void *buf, int len, struct pcap_data *pri); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/port.h b/arch/um/drivers/port.h new file mode 100644 index 000000000000..9117609a575d --- /dev/null +++ b/arch/um/drivers/port.h @@ -0,0 +1,30 @@ +/* + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __PORT_H__ +#define __PORT_H__ + +extern void *port_data(int port); +extern int port_wait(void *data); +extern void port_kern_close(void *d); +extern int port_connection(int fd, int *socket_out, int *pid_out); +extern int port_listen_fd(int port); +extern void port_read(int fd, void *data); +extern void port_kern_free(void *d); +extern int port_rcv_fd(int fd); +extern void port_remove_dev(void *d); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/port_kern.c b/arch/um/drivers/port_kern.c new file mode 100644 index 000000000000..b5ee07472f79 --- /dev/null +++ b/arch/um/drivers/port_kern.c @@ -0,0 +1,309 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/list.h" +#include "linux/sched.h" +#include "linux/slab.h" +#include "linux/interrupt.h" +#include "linux/irq.h" +#include "linux/spinlock.h" +#include "linux/errno.h" +#include "asm/atomic.h" +#include "asm/semaphore.h" +#include "asm/errno.h" +#include "kern_util.h" +#include "kern.h" +#include "irq_user.h" +#include "irq_kern.h" +#include "port.h" +#include "init.h" +#include "os.h" + +struct port_list { + struct list_head list; + atomic_t wait_count; + int has_connection; + struct completion done; + int port; + int fd; + spinlock_t lock; + struct list_head pending; + struct list_head connections; +}; + +struct port_dev { + struct port_list *port; + int helper_pid; + int telnetd_pid; +}; + +struct connection { + struct list_head list; + int fd; + int helper_pid; + int socket[2]; + int telnetd_pid; + struct port_list *port; +}; + +static irqreturn_t pipe_interrupt(int irq, void *data, struct pt_regs *regs) +{ + struct connection *conn = data; + int fd; + + fd = os_rcv_fd(conn->socket[0], &conn->helper_pid); + if(fd < 0){ + if(fd == -EAGAIN) + return(IRQ_NONE); + + printk(KERN_ERR "pipe_interrupt : os_rcv_fd returned %d\n", + -fd); + os_close_file(conn->fd); + } + + list_del(&conn->list); + + conn->fd = fd; + list_add(&conn->list, &conn->port->connections); + + complete(&conn->port->done); + return(IRQ_HANDLED); +} + +#define NO_WAITER_MSG \ + "****\n" \ + "There are currently no UML consoles waiting for port connections.\n" \ + "Either disconnect from one to make it available or activate some more\n" \ + "by enabling more consoles in the UML /etc/inittab.\n" \ + "****\n" + +static int port_accept(struct port_list *port) +{ + struct connection *conn; + int fd, socket[2], pid, ret = 0; + + fd = port_connection(port->fd, socket, &pid); + if(fd < 0){ + if(fd != -EAGAIN) + printk(KERN_ERR "port_accept : port_connection " + "returned %d\n", -fd); + goto out; + } + + conn = kmalloc(sizeof(*conn), GFP_ATOMIC); + if(conn == NULL){ + printk(KERN_ERR "port_accept : failed to allocate " + "connection\n"); + goto out_close; + } + *conn = ((struct connection) + { .list = LIST_HEAD_INIT(conn->list), + .fd = fd, + .socket = { socket[0], socket[1] }, + .telnetd_pid = pid, + .port = port }); + + if(um_request_irq(TELNETD_IRQ, socket[0], IRQ_READ, pipe_interrupt, + SA_INTERRUPT | SA_SHIRQ | SA_SAMPLE_RANDOM, + "telnetd", conn)){ + printk(KERN_ERR "port_accept : failed to get IRQ for " + "telnetd\n"); + goto out_free; + } + + if(atomic_read(&port->wait_count) == 0){ + os_write_file(fd, NO_WAITER_MSG, sizeof(NO_WAITER_MSG)); + printk("No one waiting for port\n"); + } + list_add(&conn->list, &port->pending); + return(1); + + out_free: + kfree(conn); + out_close: + os_close_file(fd); + if(pid != -1) + os_kill_process(pid, 1); + out: + return(ret); +} + +DECLARE_MUTEX(ports_sem); +struct list_head ports = LIST_HEAD_INIT(ports); + +void port_work_proc(void *unused) +{ + struct port_list *port; + struct list_head *ele; + unsigned long flags; + + local_irq_save(flags); + list_for_each(ele, &ports){ + port = list_entry(ele, struct port_list, list); + if(!port->has_connection) + continue; + reactivate_fd(port->fd, ACCEPT_IRQ); + while(port_accept(port)) ; + port->has_connection = 0; + } + local_irq_restore(flags); +} + +DECLARE_WORK(port_work, port_work_proc, NULL); + +static irqreturn_t port_interrupt(int irq, void *data, struct pt_regs *regs) +{ + struct port_list *port = data; + + port->has_connection = 1; + schedule_work(&port_work); + return(IRQ_HANDLED); +} + +void *port_data(int port_num) +{ + struct list_head *ele; + struct port_list *port; + struct port_dev *dev = NULL; + int fd; + + down(&ports_sem); + list_for_each(ele, &ports){ + port = list_entry(ele, struct port_list, list); + if(port->port == port_num) goto found; + } + port = kmalloc(sizeof(struct port_list), GFP_KERNEL); + if(port == NULL){ + printk(KERN_ERR "Allocation of port list failed\n"); + goto out; + } + + fd = port_listen_fd(port_num); + if(fd < 0){ + printk(KERN_ERR "binding to port %d failed, errno = %d\n", + port_num, -fd); + goto out_free; + } + if(um_request_irq(ACCEPT_IRQ, fd, IRQ_READ, port_interrupt, + SA_INTERRUPT | SA_SHIRQ | SA_SAMPLE_RANDOM, "port", + port)){ + printk(KERN_ERR "Failed to get IRQ for port %d\n", port_num); + goto out_close; + } + + *port = ((struct port_list) + { .list = LIST_HEAD_INIT(port->list), + .wait_count = ATOMIC_INIT(0), + .has_connection = 0, + .port = port_num, + .fd = fd, + .pending = LIST_HEAD_INIT(port->pending), + .connections = LIST_HEAD_INIT(port->connections) }); + spin_lock_init(&port->lock); + init_completion(&port->done); + list_add(&port->list, &ports); + + found: + dev = kmalloc(sizeof(struct port_dev), GFP_KERNEL); + if(dev == NULL){ + printk(KERN_ERR "Allocation of port device entry failed\n"); + goto out; + } + + *dev = ((struct port_dev) { .port = port, + .helper_pid = -1, + .telnetd_pid = -1 }); + goto out; + + out_free: + kfree(port); + out_close: + os_close_file(fd); + out: + up(&ports_sem); + return(dev); +} + +int port_wait(void *data) +{ + struct port_dev *dev = data; + struct connection *conn; + struct port_list *port = dev->port; + int fd; + + atomic_inc(&port->wait_count); + while(1){ + fd = -ERESTARTSYS; + if(wait_for_completion_interruptible(&port->done)) + goto out; + + spin_lock(&port->lock); + + conn = list_entry(port->connections.next, struct connection, + list); + list_del(&conn->list); + spin_unlock(&port->lock); + + os_shutdown_socket(conn->socket[0], 1, 1); + os_close_file(conn->socket[0]); + os_shutdown_socket(conn->socket[1], 1, 1); + os_close_file(conn->socket[1]); + + /* This is done here because freeing an IRQ can't be done + * within the IRQ handler. So, pipe_interrupt always ups + * the semaphore regardless of whether it got a successful + * connection. Then we loop here throwing out failed + * connections until a good one is found. + */ + free_irq_by_irq_and_dev(TELNETD_IRQ, conn); + free_irq(TELNETD_IRQ, conn); + + if(conn->fd >= 0) break; + os_close_file(conn->fd); + kfree(conn); + } + + fd = conn->fd; + dev->helper_pid = conn->helper_pid; + dev->telnetd_pid = conn->telnetd_pid; + kfree(conn); + out: + atomic_dec(&port->wait_count); + return fd; +} + +void port_remove_dev(void *d) +{ + struct port_dev *dev = d; + + if(dev->helper_pid != -1) + os_kill_process(dev->helper_pid, 0); + if(dev->telnetd_pid != -1) + os_kill_process(dev->telnetd_pid, 1); + dev->helper_pid = -1; + dev->telnetd_pid = -1; +} + +void port_kern_free(void *d) +{ + struct port_dev *dev = d; + + port_remove_dev(dev); + kfree(dev); +} + +static void free_port(void) +{ + struct list_head *ele; + struct port_list *port; + + list_for_each(ele, &ports){ + port = list_entry(ele, struct port_list, list); + free_irq_by_fd(port->fd); + os_close_file(port->fd); + } +} + +__uml_exitcall(free_port); diff --git a/arch/um/drivers/port_user.c b/arch/um/drivers/port_user.c new file mode 100644 index 000000000000..14dd2002d2da --- /dev/null +++ b/arch/um/drivers/port_user.c @@ -0,0 +1,225 @@ +/* + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stddef.h> +#include <stdlib.h> +#include <string.h> +#include <errno.h> +#include <unistd.h> +#include <termios.h> +#include <sys/socket.h> +#include <sys/un.h> +#include <netinet/in.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "chan_user.h" +#include "port.h" +#include "helper.h" +#include "os.h" + +struct port_chan { + int raw; + struct termios tt; + void *kernel_data; + char dev[sizeof("32768\0")]; +}; + +static void *port_init(char *str, int device, struct chan_opts *opts) +{ + struct port_chan *data; + void *kern_data; + char *end; + int port; + + if(*str != ':'){ + printk("port_init : channel type 'port' must specify a " + "port number\n"); + return(NULL); + } + str++; + port = strtoul(str, &end, 0); + if((*end != '\0') || (end == str)){ + printk("port_init : couldn't parse port '%s'\n", str); + return(NULL); + } + + kern_data = port_data(port); + if(kern_data == NULL) + return(NULL); + + data = um_kmalloc(sizeof(*data)); + if(data == NULL) + goto err; + + *data = ((struct port_chan) { .raw = opts->raw, + .kernel_data = kern_data }); + sprintf(data->dev, "%d", port); + + return(data); + err: + port_kern_free(kern_data); + return(NULL); +} + +static void port_free(void *d) +{ + struct port_chan *data = d; + + port_kern_free(data->kernel_data); + kfree(data); +} + +static int port_open(int input, int output, int primary, void *d, + char **dev_out) +{ + struct port_chan *data = d; + int fd, err; + + fd = port_wait(data->kernel_data); + if((fd >= 0) && data->raw){ + CATCH_EINTR(err = tcgetattr(fd, &data->tt)); + if(err) + return(err); + + err = raw(fd); + if(err) + return(err); + } + *dev_out = data->dev; + return(fd); +} + +static void port_close(int fd, void *d) +{ + struct port_chan *data = d; + + port_remove_dev(data->kernel_data); + os_close_file(fd); +} + +static int port_console_write(int fd, const char *buf, int n, void *d) +{ + struct port_chan *data = d; + + return(generic_console_write(fd, buf, n, &data->tt)); +} + +struct chan_ops port_ops = { + .type = "port", + .init = port_init, + .open = port_open, + .close = port_close, + .read = generic_read, + .write = generic_write, + .console_write = port_console_write, + .window_size = generic_window_size, + .free = port_free, + .winch = 1, +}; + +int port_listen_fd(int port) +{ + struct sockaddr_in addr; + int fd, err, arg; + + fd = socket(PF_INET, SOCK_STREAM, 0); + if(fd == -1) + return(-errno); + + arg = 1; + if(setsockopt(fd, SOL_SOCKET, SO_REUSEADDR, &arg, sizeof(arg)) < 0){ + err = -errno; + goto out; + } + + addr.sin_family = AF_INET; + addr.sin_port = htons(port); + addr.sin_addr.s_addr = htonl(INADDR_ANY); + if(bind(fd, (struct sockaddr *) &addr, sizeof(addr)) < 0){ + err = -errno; + goto out; + } + + if(listen(fd, 1) < 0){ + err = -errno; + goto out; + } + + err = os_set_fd_block(fd, 0); + if(err < 0) + goto out; + + return(fd); + out: + os_close_file(fd); + return(err); +} + +struct port_pre_exec_data { + int sock_fd; + int pipe_fd; +}; + +void port_pre_exec(void *arg) +{ + struct port_pre_exec_data *data = arg; + + dup2(data->sock_fd, 0); + dup2(data->sock_fd, 1); + dup2(data->sock_fd, 2); + os_close_file(data->sock_fd); + dup2(data->pipe_fd, 3); + os_shutdown_socket(3, 1, 0); + os_close_file(data->pipe_fd); +} + +int port_connection(int fd, int *socket, int *pid_out) +{ + int new, err; + char *argv[] = { "/usr/sbin/in.telnetd", "-L", + "/usr/lib/uml/port-helper", NULL }; + struct port_pre_exec_data data; + + new = os_accept_connection(fd); + if(new < 0) + return(new); + + err = os_pipe(socket, 0, 0); + if(err < 0) + goto out_close; + + data = ((struct port_pre_exec_data) + { .sock_fd = new, + .pipe_fd = socket[1] }); + + err = run_helper(port_pre_exec, &data, argv, NULL); + if(err < 0) + goto out_shutdown; + + *pid_out = err; + return(new); + + out_shutdown: + os_shutdown_socket(socket[0], 1, 1); + os_close_file(socket[0]); + os_shutdown_socket(socket[1], 1, 1); + os_close_file(socket[1]); + out_close: + os_close_file(new); + return(err); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/pty.c b/arch/um/drivers/pty.c new file mode 100644 index 000000000000..ed84d01df6cc --- /dev/null +++ b/arch/um/drivers/pty.c @@ -0,0 +1,162 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <unistd.h> +#include <string.h> +#include <errno.h> +#include <termios.h> +#include "chan_user.h" +#include "user.h" +#include "user_util.h" +#include "kern_util.h" +#include "os.h" + +struct pty_chan { + void (*announce)(char *dev_name, int dev); + int dev; + int raw; + struct termios tt; + char dev_name[sizeof("/dev/pts/0123456\0")]; +}; + +static void *pty_chan_init(char *str, int device, struct chan_opts *opts) +{ + struct pty_chan *data; + + data = um_kmalloc(sizeof(*data)); + if(data == NULL) return(NULL); + *data = ((struct pty_chan) { .announce = opts->announce, + .dev = device, + .raw = opts->raw }); + return(data); +} + +static int pts_open(int input, int output, int primary, void *d, + char **dev_out) +{ + struct pty_chan *data = d; + char *dev; + int fd, err; + + fd = get_pty(); + if(fd < 0){ + printk("open_pts : Failed to open pts\n"); + return(-errno); + } + if(data->raw){ + CATCH_EINTR(err = tcgetattr(fd, &data->tt)); + if(err) + return(err); + + err = raw(fd); + if(err) + return(err); + } + + dev = ptsname(fd); + sprintf(data->dev_name, "%s", dev); + *dev_out = data->dev_name; + if (data->announce) + (*data->announce)(dev, data->dev); + return(fd); +} + +static int getmaster(char *line) +{ + char *pty, *bank, *cp; + int master, err; + + pty = &line[strlen("/dev/ptyp")]; + for (bank = "pqrs"; *bank; bank++) { + line[strlen("/dev/pty")] = *bank; + *pty = '0'; + if (os_stat_file(line, NULL) < 0) + break; + for (cp = "0123456789abcdef"; *cp; cp++) { + *pty = *cp; + master = os_open_file(line, of_rdwr(OPENFLAGS()), 0); + if (master >= 0) { + char *tp = &line[strlen("/dev/")]; + + /* verify slave side is usable */ + *tp = 't'; + err = os_access(line, OS_ACC_RW_OK); + *tp = 'p'; + if(err == 0) return(master); + (void) os_close_file(master); + } + } + } + return(-1); +} + +static int pty_open(int input, int output, int primary, void *d, + char **dev_out) +{ + struct pty_chan *data = d; + int fd, err; + char dev[sizeof("/dev/ptyxx\0")] = "/dev/ptyxx"; + + fd = getmaster(dev); + if(fd < 0) + return(-errno); + + if(data->raw){ + err = raw(fd); + if(err) + return(err); + } + + if(data->announce) (*data->announce)(dev, data->dev); + + sprintf(data->dev_name, "%s", dev); + *dev_out = data->dev_name; + return(fd); +} + +static int pty_console_write(int fd, const char *buf, int n, void *d) +{ + struct pty_chan *data = d; + + return(generic_console_write(fd, buf, n, &data->tt)); +} + +struct chan_ops pty_ops = { + .type = "pty", + .init = pty_chan_init, + .open = pty_open, + .close = generic_close, + .read = generic_read, + .write = generic_write, + .console_write = pty_console_write, + .window_size = generic_window_size, + .free = generic_free, + .winch = 0, +}; + +struct chan_ops pts_ops = { + .type = "pts", + .init = pty_chan_init, + .open = pts_open, + .close = generic_close, + .read = generic_read, + .write = generic_write, + .console_write = pty_console_write, + .window_size = generic_window_size, + .free = generic_free, + .winch = 0, +}; + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/random.c b/arch/um/drivers/random.c new file mode 100644 index 000000000000..d43e9fab05a7 --- /dev/null +++ b/arch/um/drivers/random.c @@ -0,0 +1,122 @@ +/* Much of this ripped from hw_random.c */ + +#include <linux/module.h> +#include <linux/fs.h> +#include <linux/miscdevice.h> +#include <linux/delay.h> +#include <asm/uaccess.h> +#include "os.h" + +/* + * core module and version information + */ +#define RNG_VERSION "1.0.0" +#define RNG_MODULE_NAME "random" +#define RNG_DRIVER_NAME RNG_MODULE_NAME " virtual driver " RNG_VERSION +#define PFX RNG_MODULE_NAME ": " + +#define RNG_MISCDEV_MINOR 183 /* official */ + +static int random_fd = -1; + +static int rng_dev_open (struct inode *inode, struct file *filp) +{ + /* enforce read-only access to this chrdev */ + if ((filp->f_mode & FMODE_READ) == 0) + return -EINVAL; + if (filp->f_mode & FMODE_WRITE) + return -EINVAL; + + return 0; +} + +static ssize_t rng_dev_read (struct file *filp, char __user *buf, size_t size, + loff_t * offp) +{ + u32 data; + int n, ret = 0, have_data; + + while(size){ + n = os_read_file(random_fd, &data, sizeof(data)); + if(n > 0){ + have_data = n; + while (have_data && size) { + if (put_user((u8)data, buf++)) { + ret = ret ? : -EFAULT; + break; + } + size--; + ret++; + have_data--; + data>>=8; + } + } + else if(n == -EAGAIN){ + if (filp->f_flags & O_NONBLOCK) + return ret ? : -EAGAIN; + + if(need_resched()){ + current->state = TASK_INTERRUPTIBLE; + schedule_timeout(1); + } + } + else return n; + if (signal_pending (current)) + return ret ? : -ERESTARTSYS; + } + return ret; +} + +static struct file_operations rng_chrdev_ops = { + .owner = THIS_MODULE, + .open = rng_dev_open, + .read = rng_dev_read, +}; + +static struct miscdevice rng_miscdev = { + RNG_MISCDEV_MINOR, + RNG_MODULE_NAME, + &rng_chrdev_ops, +}; + +/* + * rng_init - initialize RNG module + */ +static int __init rng_init (void) +{ + int err; + + err = os_open_file("/dev/random", of_read(OPENFLAGS()), 0); + if(err < 0) + goto out; + + random_fd = err; + + err = os_set_fd_block(random_fd, 0); + if(err) + goto err_out_cleanup_hw; + + err = misc_register (&rng_miscdev); + if (err) { + printk (KERN_ERR PFX "misc device register failed\n"); + goto err_out_cleanup_hw; + } + + out: + return err; + + err_out_cleanup_hw: + random_fd = -1; + goto out; +} + +/* + * rng_cleanup - shutdown RNG module + */ +static void __exit rng_cleanup (void) +{ + misc_deregister (&rng_miscdev); +} + +module_init (rng_init); +module_exit (rng_cleanup); diff --git a/arch/um/drivers/slip.h b/arch/um/drivers/slip.h new file mode 100644 index 000000000000..495f2f1b1420 --- /dev/null +++ b/arch/um/drivers/slip.h @@ -0,0 +1,39 @@ +#ifndef __UM_SLIP_H +#define __UM_SLIP_H + +#define BUF_SIZE 1500 + /* two bytes each for a (pathological) max packet of escaped chars + * + * terminating END char + initial END char */ +#define ENC_BUF_SIZE (2 * BUF_SIZE + 2) + +struct slip_data { + void *dev; + char name[sizeof("slnnnnn\0")]; + char *addr; + char *gate_addr; + int slave; + char ibuf[ENC_BUF_SIZE]; + char obuf[ENC_BUF_SIZE]; + int more; /* more data: do not read fd until ibuf has been drained */ + int pos; + int esc; +}; + +extern struct net_user_info slip_user_info; + +extern int set_umn_addr(int fd, char *addr, char *ptp_addr); +extern int slip_user_read(int fd, void *buf, int len, struct slip_data *pri); +extern int slip_user_write(int fd, void *buf, int len, struct slip_data *pri); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/slip_kern.c b/arch/um/drivers/slip_kern.c new file mode 100644 index 000000000000..0886eedba213 --- /dev/null +++ b/arch/um/drivers/slip_kern.c @@ -0,0 +1,109 @@ +#include "linux/config.h" +#include "linux/kernel.h" +#include "linux/stddef.h" +#include "linux/init.h" +#include "linux/netdevice.h" +#include "linux/if_arp.h" +#include "net_kern.h" +#include "net_user.h" +#include "kern.h" +#include "slip.h" + +struct slip_init { + char *gate_addr; +}; + +void slip_init(struct net_device *dev, void *data) +{ + struct uml_net_private *private; + struct slip_data *spri; + struct slip_init *init = data; + + private = dev->priv; + spri = (struct slip_data *) private->user; + *spri = ((struct slip_data) + { .name = { '\0' }, + .addr = NULL, + .gate_addr = init->gate_addr, + .slave = -1, + .ibuf = { '\0' }, + .obuf = { '\0' }, + .pos = 0, + .esc = 0, + .dev = dev }); + + dev->init = NULL; + dev->hard_header_len = 0; + dev->addr_len = 4; + dev->type = ARPHRD_ETHER; + dev->tx_queue_len = 256; + dev->flags = IFF_NOARP; + printk("SLIP backend - SLIP IP = %s\n", spri->gate_addr); +} + +static unsigned short slip_protocol(struct sk_buff *skbuff) +{ + return(htons(ETH_P_IP)); +} + +static int slip_read(int fd, struct sk_buff **skb, + struct uml_net_private *lp) +{ + return(slip_user_read(fd, (*skb)->mac.raw, (*skb)->dev->mtu, + (struct slip_data *) &lp->user)); +} + +static int slip_write(int fd, struct sk_buff **skb, + struct uml_net_private *lp) +{ + return(slip_user_write(fd, (*skb)->data, (*skb)->len, + (struct slip_data *) &lp->user)); +} + +struct net_kern_info slip_kern_info = { + .init = slip_init, + .protocol = slip_protocol, + .read = slip_read, + .write = slip_write, +}; + +static int slip_setup(char *str, char **mac_out, void *data) +{ + struct slip_init *init = data; + + *init = ((struct slip_init) + { .gate_addr = NULL }); + + if(str[0] != '\0') + init->gate_addr = str; + return(1); +} + +static struct transport slip_transport = { + .list = LIST_HEAD_INIT(slip_transport.list), + .name = "slip", + .setup = slip_setup, + .user = &slip_user_info, + .kern = &slip_kern_info, + .private_size = sizeof(struct slip_data), + .setup_size = sizeof(struct slip_init), +}; + +static int register_slip(void) +{ + register_transport(&slip_transport); + return(1); +} + +__initcall(register_slip); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/slip_proto.h b/arch/um/drivers/slip_proto.h new file mode 100644 index 000000000000..7206361ace45 --- /dev/null +++ b/arch/um/drivers/slip_proto.h @@ -0,0 +1,93 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __UM_SLIP_PROTO_H__ +#define __UM_SLIP_PROTO_H__ + +/* SLIP protocol characters. */ +#define SLIP_END 0300 /* indicates end of frame */ +#define SLIP_ESC 0333 /* indicates byte stuffing */ +#define SLIP_ESC_END 0334 /* ESC ESC_END means END 'data' */ +#define SLIP_ESC_ESC 0335 /* ESC ESC_ESC means ESC 'data' */ + +static inline int slip_unesc(unsigned char c,char *buf,int *pos, int *esc) +{ + int ret; + + switch(c){ + case SLIP_END: + *esc = 0; + ret=*pos; + *pos=0; + return(ret); + case SLIP_ESC: + *esc = 1; + return(0); + case SLIP_ESC_ESC: + if(*esc){ + *esc = 0; + c = SLIP_ESC; + } + break; + case SLIP_ESC_END: + if(*esc){ + *esc = 0; + c = SLIP_END; + } + break; + } + buf[(*pos)++] = c; + return(0); +} + +static inline int slip_esc(unsigned char *s, unsigned char *d, int len) +{ + unsigned char *ptr = d; + unsigned char c; + + /* + * Send an initial END character to flush out any + * data that may have accumulated in the receiver + * due to line noise. + */ + + *ptr++ = SLIP_END; + + /* + * For each byte in the packet, send the appropriate + * character sequence, according to the SLIP protocol. + */ + + while (len-- > 0) { + switch(c = *s++) { + case SLIP_END: + *ptr++ = SLIP_ESC; + *ptr++ = SLIP_ESC_END; + break; + case SLIP_ESC: + *ptr++ = SLIP_ESC; + *ptr++ = SLIP_ESC_ESC; + break; + default: + *ptr++ = c; + break; + } + } + *ptr++ = SLIP_END; + return (ptr - d); +} + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/slip_user.c b/arch/um/drivers/slip_user.c new file mode 100644 index 000000000000..d94846b1b4cf --- /dev/null +++ b/arch/um/drivers/slip_user.c @@ -0,0 +1,280 @@ +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <stddef.h> +#include <sched.h> +#include <string.h> +#include <errno.h> +#include <sys/termios.h> +#include <sys/wait.h> +#include <sys/signal.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "net_user.h" +#include "slip.h" +#include "slip_proto.h" +#include "helper.h" +#include "os.h" + +void slip_user_init(void *data, void *dev) +{ + struct slip_data *pri = data; + + pri->dev = dev; +} + +static int set_up_tty(int fd) +{ + int i; + struct termios tios; + + if (tcgetattr(fd, &tios) < 0) { + printk("could not get initial terminal attributes\n"); + return(-1); + } + + tios.c_cflag = CS8 | CREAD | HUPCL | CLOCAL; + tios.c_iflag = IGNBRK | IGNPAR; + tios.c_oflag = 0; + tios.c_lflag = 0; + for (i = 0; i < NCCS; i++) + tios.c_cc[i] = 0; + tios.c_cc[VMIN] = 1; + tios.c_cc[VTIME] = 0; + + cfsetospeed(&tios, B38400); + cfsetispeed(&tios, B38400); + + if (tcsetattr(fd, TCSAFLUSH, &tios) < 0) { + printk("failed to set terminal attributes\n"); + return(-1); + } + return(0); +} + +struct slip_pre_exec_data { + int stdin; + int stdout; + int close_me; +}; + +static void slip_pre_exec(void *arg) +{ + struct slip_pre_exec_data *data = arg; + + if(data->stdin >= 0) dup2(data->stdin, 0); + dup2(data->stdout, 1); + if(data->close_me >= 0) os_close_file(data->close_me); +} + +static int slip_tramp(char **argv, int fd) +{ + struct slip_pre_exec_data pe_data; + char *output; + int status, pid, fds[2], err, output_len; + + err = os_pipe(fds, 1, 0); + if(err < 0){ + printk("slip_tramp : pipe failed, err = %d\n", -err); + return(err); + } + + err = 0; + pe_data.stdin = fd; + pe_data.stdout = fds[1]; + pe_data.close_me = fds[0]; + pid = run_helper(slip_pre_exec, &pe_data, argv, NULL); + + if(pid < 0) err = pid; + else { + output_len = page_size(); + output = um_kmalloc(output_len); + if(output == NULL) + printk("slip_tramp : failed to allocate output " + "buffer\n"); + + os_close_file(fds[1]); + read_output(fds[0], output, output_len); + if(output != NULL){ + printk("%s", output); + kfree(output); + } + CATCH_EINTR(err = waitpid(pid, &status, 0)); + if(err < 0) + err = errno; + else if(!WIFEXITED(status) || (WEXITSTATUS(status) != 0)){ + printk("'%s' didn't exit with status 0\n", argv[0]); + err = -EINVAL; + } + } + + os_close_file(fds[0]); + + return(err); +} + +static int slip_open(void *data) +{ + struct slip_data *pri = data; + char version_buf[sizeof("nnnnn\0")]; + char gate_buf[sizeof("nnn.nnn.nnn.nnn\0")]; + char *argv[] = { "uml_net", version_buf, "slip", "up", gate_buf, + NULL }; + int sfd, mfd, err; + + mfd = get_pty(); + if(mfd < 0){ + printk("umn : Failed to open pty, err = %d\n", -mfd); + return(mfd); + } + sfd = os_open_file(ptsname(mfd), of_rdwr(OPENFLAGS()), 0); + if(sfd < 0){ + printk("Couldn't open tty for slip line, err = %d\n", -sfd); + os_close_file(mfd); + return(sfd); + } + if(set_up_tty(sfd)) return(-1); + pri->slave = sfd; + pri->pos = 0; + pri->esc = 0; + if(pri->gate_addr != NULL){ + sprintf(version_buf, "%d", UML_NET_VERSION); + strcpy(gate_buf, pri->gate_addr); + + err = slip_tramp(argv, sfd); + + if(err < 0){ + printk("slip_tramp failed - err = %d\n", -err); + return(err); + } + err = os_get_ifname(pri->slave, pri->name); + if(err < 0){ + printk("get_ifname failed, err = %d\n", -err); + return(err); + } + iter_addresses(pri->dev, open_addr, pri->name); + } + else { + err = os_set_slip(sfd); + if(err < 0){ + printk("Failed to set slip discipline encapsulation - " + "err = %d\n", -err); + return(err); + } + } + return(mfd); +} + +static void slip_close(int fd, void *data) +{ + struct slip_data *pri = data; + char version_buf[sizeof("nnnnn\0")]; + char *argv[] = { "uml_net", version_buf, "slip", "down", pri->name, + NULL }; + int err; + + if(pri->gate_addr != NULL) + iter_addresses(pri->dev, close_addr, pri->name); + + sprintf(version_buf, "%d", UML_NET_VERSION); + + err = slip_tramp(argv, pri->slave); + + if(err != 0) + printk("slip_tramp failed - errno = %d\n", -err); + os_close_file(fd); + os_close_file(pri->slave); + pri->slave = -1; +} + +int slip_user_read(int fd, void *buf, int len, struct slip_data *pri) +{ + int i, n, size, start; + + if(pri->more>0) { + i = 0; + while(i < pri->more) { + size = slip_unesc(pri->ibuf[i++], + pri->ibuf, &pri->pos, &pri->esc); + if(size){ + memcpy(buf, pri->ibuf, size); + memmove(pri->ibuf, &pri->ibuf[i], pri->more-i); + pri->more=pri->more-i; + return(size); + } + } + pri->more=0; + } + + n = net_read(fd, &pri->ibuf[pri->pos], sizeof(pri->ibuf) - pri->pos); + if(n <= 0) return(n); + + start = pri->pos; + for(i = 0; i < n; i++){ + size = slip_unesc(pri->ibuf[start + i], + pri->ibuf, &pri->pos, &pri->esc); + if(size){ + memcpy(buf, pri->ibuf, size); + memmove(pri->ibuf, &pri->ibuf[start+i+1], n-(i+1)); + pri->more=n-(i+1); + return(size); + } + } + return(0); +} + +int slip_user_write(int fd, void *buf, int len, struct slip_data *pri) +{ + int actual, n; + + actual = slip_esc(buf, pri->obuf, len); + n = net_write(fd, pri->obuf, actual); + if(n < 0) return(n); + else return(len); +} + +static int slip_set_mtu(int mtu, void *data) +{ + return(mtu); +} + +static void slip_add_addr(unsigned char *addr, unsigned char *netmask, + void *data) +{ + struct slip_data *pri = data; + + if(pri->slave < 0) return; + open_addr(addr, netmask, pri->name); +} + +static void slip_del_addr(unsigned char *addr, unsigned char *netmask, + void *data) +{ + struct slip_data *pri = data; + + if(pri->slave < 0) return; + close_addr(addr, netmask, pri->name); +} + +struct net_user_info slip_user_info = { + .init = slip_user_init, + .open = slip_open, + .close = slip_close, + .remove = NULL, + .set_mtu = slip_set_mtu, + .add_address = slip_add_addr, + .delete_address = slip_del_addr, + .max_packet = BUF_SIZE +}; + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/slirp.h b/arch/um/drivers/slirp.h new file mode 100644 index 000000000000..04e407d1e44a --- /dev/null +++ b/arch/um/drivers/slirp.h @@ -0,0 +1,51 @@ +#ifndef __UM_SLIRP_H +#define __UM_SLIRP_H + +#define BUF_SIZE 1500 + /* two bytes each for a (pathological) max packet of escaped chars + * + * terminating END char + initial END char */ +#define ENC_BUF_SIZE (2 * BUF_SIZE + 2) + +#define SLIRP_MAX_ARGS 100 +/* + * XXX this next definition is here because I don't understand why this + * initializer doesn't work in slirp_kern.c: + * + * argv : { init->argv[ 0 ... SLIRP_MAX_ARGS-1 ] }, + * + * or why I can't typecast like this: + * + * argv : (char* [SLIRP_MAX_ARGS])(init->argv), + */ +struct arg_list_dummy_wrapper { char *argv[SLIRP_MAX_ARGS]; }; + +struct slirp_data { + void *dev; + struct arg_list_dummy_wrapper argw; + int pid; + int slave; + char ibuf[ENC_BUF_SIZE]; + char obuf[ENC_BUF_SIZE]; + int more; /* more data: do not read fd until ibuf has been drained */ + int pos; + int esc; +}; + +extern struct net_user_info slirp_user_info; + +extern int set_umn_addr(int fd, char *addr, char *ptp_addr); +extern int slirp_user_read(int fd, void *buf, int len, struct slirp_data *pri); +extern int slirp_user_write(int fd, void *buf, int len, struct slirp_data *pri); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/slirp_kern.c b/arch/um/drivers/slirp_kern.c new file mode 100644 index 000000000000..c9d6b52a831d --- /dev/null +++ b/arch/um/drivers/slirp_kern.c @@ -0,0 +1,135 @@ +#include "linux/kernel.h" +#include "linux/stddef.h" +#include "linux/init.h" +#include "linux/netdevice.h" +#include "linux/if_arp.h" +#include "net_kern.h" +#include "net_user.h" +#include "kern.h" +#include "slirp.h" + +struct slirp_init { + struct arg_list_dummy_wrapper argw; /* XXX should be simpler... */ +}; + +void slirp_init(struct net_device *dev, void *data) +{ + struct uml_net_private *private; + struct slirp_data *spri; + struct slirp_init *init = data; + int i; + + private = dev->priv; + spri = (struct slirp_data *) private->user; + *spri = ((struct slirp_data) + { .argw = init->argw, + .pid = -1, + .slave = -1, + .ibuf = { '\0' }, + .obuf = { '\0' }, + .pos = 0, + .esc = 0, + .dev = dev }); + + dev->init = NULL; + dev->hard_header_len = 0; + dev->header_cache_update = NULL; + dev->hard_header_cache = NULL; + dev->hard_header = NULL; + dev->addr_len = 0; + dev->type = ARPHRD_SLIP; + dev->tx_queue_len = 256; + dev->flags = IFF_NOARP; + printk("SLIRP backend - command line:"); + for(i=0;spri->argw.argv[i]!=NULL;i++) { + printk(" '%s'",spri->argw.argv[i]); + } + printk("\n"); +} + +static unsigned short slirp_protocol(struct sk_buff *skbuff) +{ + return(htons(ETH_P_IP)); +} + +static int slirp_read(int fd, struct sk_buff **skb, + struct uml_net_private *lp) +{ + return(slirp_user_read(fd, (*skb)->mac.raw, (*skb)->dev->mtu, + (struct slirp_data *) &lp->user)); +} + +static int slirp_write(int fd, struct sk_buff **skb, + struct uml_net_private *lp) +{ + return(slirp_user_write(fd, (*skb)->data, (*skb)->len, + (struct slirp_data *) &lp->user)); +} + +struct net_kern_info slirp_kern_info = { + .init = slirp_init, + .protocol = slirp_protocol, + .read = slirp_read, + .write = slirp_write, +}; + +static int slirp_setup(char *str, char **mac_out, void *data) +{ + struct slirp_init *init = data; + int i=0; + + *init = ((struct slirp_init) + { argw : { { "slirp", NULL } } }); + + str = split_if_spec(str, mac_out, NULL); + + if(str == NULL) { /* no command line given after MAC addr */ + return(1); + } + + do { + if(i>=SLIRP_MAX_ARGS-1) { + printk("slirp_setup: truncating slirp arguments\n"); + break; + } + init->argw.argv[i++] = str; + while(*str && *str!=',') { + if(*str=='_') *str=' '; + str++; + } + if(*str!=',') + break; + *str++='\0'; + } while(1); + init->argw.argv[i]=NULL; + return(1); +} + +static struct transport slirp_transport = { + .list = LIST_HEAD_INIT(slirp_transport.list), + .name = "slirp", + .setup = slirp_setup, + .user = &slirp_user_info, + .kern = &slirp_kern_info, + .private_size = sizeof(struct slirp_data), + .setup_size = sizeof(struct slirp_init), +}; + +static int register_slirp(void) +{ + register_transport(&slirp_transport); + return(1); +} + +__initcall(register_slirp); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/slirp_user.c b/arch/um/drivers/slirp_user.c new file mode 100644 index 000000000000..c322515c71cc --- /dev/null +++ b/arch/um/drivers/slirp_user.c @@ -0,0 +1,201 @@ +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <stddef.h> +#include <sched.h> +#include <string.h> +#include <errno.h> +#include <sys/wait.h> +#include <sys/signal.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "net_user.h" +#include "slirp.h" +#include "slip_proto.h" +#include "helper.h" +#include "os.h" + +void slirp_user_init(void *data, void *dev) +{ + struct slirp_data *pri = data; + + pri->dev = dev; +} + +struct slirp_pre_exec_data { + int stdin; + int stdout; +}; + +static void slirp_pre_exec(void *arg) +{ + struct slirp_pre_exec_data *data = arg; + + if(data->stdin != -1) dup2(data->stdin, 0); + if(data->stdout != -1) dup2(data->stdout, 1); +} + +static int slirp_tramp(char **argv, int fd) +{ + struct slirp_pre_exec_data pe_data; + int pid; + + pe_data.stdin = fd; + pe_data.stdout = fd; + pid = run_helper(slirp_pre_exec, &pe_data, argv, NULL); + + return(pid); +} + +/* XXX This is just a trivial wrapper around os_pipe */ +static int slirp_datachan(int *mfd, int *sfd) +{ + int fds[2], err; + + err = os_pipe(fds, 1, 1); + if(err < 0){ + printk("slirp_datachan: Failed to open pipe, err = %d\n", -err); + return(err); + } + + *mfd = fds[0]; + *sfd = fds[1]; + return(0); +} + +static int slirp_open(void *data) +{ + struct slirp_data *pri = data; + int sfd, mfd, pid, err; + + err = slirp_datachan(&mfd, &sfd); + if(err) + return(err); + + pid = slirp_tramp(pri->argw.argv, sfd); + + if(pid < 0){ + printk("slirp_tramp failed - errno = %d\n", -pid); + os_close_file(sfd); + os_close_file(mfd); + return(pid); + } + + pri->slave = sfd; + pri->pos = 0; + pri->esc = 0; + + pri->pid = pid; + + return(mfd); +} + +static void slirp_close(int fd, void *data) +{ + struct slirp_data *pri = data; + int status,err; + + os_close_file(fd); + os_close_file(pri->slave); + + pri->slave = -1; + + if(pri->pid<1) { + printk("slirp_close: no child process to shut down\n"); + return; + } + +#if 0 + if(kill(pri->pid, SIGHUP)<0) { + printk("slirp_close: sending hangup to %d failed (%d)\n", + pri->pid, errno); + } +#endif + + CATCH_EINTR(err = waitpid(pri->pid, &status, WNOHANG)); + if(err < 0) { + printk("slirp_close: waitpid returned %d\n", errno); + return; + } + + if(err == 0) { + printk("slirp_close: process %d has not exited\n"); + return; + } + + pri->pid = -1; +} + +int slirp_user_read(int fd, void *buf, int len, struct slirp_data *pri) +{ + int i, n, size, start; + + if(pri->more>0) { + i = 0; + while(i < pri->more) { + size = slip_unesc(pri->ibuf[i++], + pri->ibuf,&pri->pos,&pri->esc); + if(size){ + memcpy(buf, pri->ibuf, size); + memmove(pri->ibuf, &pri->ibuf[i], pri->more-i); + pri->more=pri->more-i; + return(size); + } + } + pri->more=0; + } + + n = net_read(fd, &pri->ibuf[pri->pos], sizeof(pri->ibuf) - pri->pos); + if(n <= 0) return(n); + + start = pri->pos; + for(i = 0; i < n; i++){ + size = slip_unesc(pri->ibuf[start + i], + pri->ibuf,&pri->pos,&pri->esc); + if(size){ + memcpy(buf, pri->ibuf, size); + memmove(pri->ibuf, &pri->ibuf[start+i+1], n-(i+1)); + pri->more=n-(i+1); + return(size); + } + } + return(0); +} + +int slirp_user_write(int fd, void *buf, int len, struct slirp_data *pri) +{ + int actual, n; + + actual = slip_esc(buf, pri->obuf, len); + n = net_write(fd, pri->obuf, actual); + if(n < 0) return(n); + else return(len); +} + +static int slirp_set_mtu(int mtu, void *data) +{ + return(mtu); +} + +struct net_user_info slirp_user_info = { + .init = slirp_user_init, + .open = slirp_open, + .close = slirp_close, + .remove = NULL, + .set_mtu = slirp_set_mtu, + .add_address = NULL, + .delete_address = NULL, + .max_packet = BUF_SIZE +}; + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/ssl.c b/arch/um/drivers/ssl.c new file mode 100644 index 000000000000..c5839c3141f8 --- /dev/null +++ b/arch/um/drivers/ssl.c @@ -0,0 +1,251 @@ +/* + * Copyright (C) 2000, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/fs.h" +#include "linux/tty.h" +#include "linux/tty_driver.h" +#include "linux/major.h" +#include "linux/mm.h" +#include "linux/init.h" +#include "linux/console.h" +#include "asm/termbits.h" +#include "asm/irq.h" +#include "line.h" +#include "ssl.h" +#include "chan_kern.h" +#include "user_util.h" +#include "kern_util.h" +#include "kern.h" +#include "init.h" +#include "irq_user.h" +#include "mconsole_kern.h" +#include "2_5compat.h" + +static int ssl_version = 1; + +/* Referenced only by tty_driver below - presumably it's locked correctly + * by the tty driver. + */ + +static struct tty_driver *ssl_driver; + +#define NR_PORTS 64 + +void ssl_announce(char *dev_name, int dev) +{ + printk(KERN_INFO "Serial line %d assigned device '%s'\n", dev, + dev_name); +} + +static struct chan_opts opts = { + .announce = ssl_announce, + .xterm_title = "Serial Line #%d", + .raw = 1, + .tramp_stack = 0, + .in_kernel = 1, +}; + +static int ssl_config(char *str); +static int ssl_get_config(char *dev, char *str, int size, char **error_out); +static int ssl_remove(char *str); + +static struct line_driver driver = { + .name = "UML serial line", + .device_name = "ttyS", + .devfs_name = "tts/", + .major = TTY_MAJOR, + .minor_start = 64, + .type = TTY_DRIVER_TYPE_SERIAL, + .subtype = 0, + .read_irq = SSL_IRQ, + .read_irq_name = "ssl", + .write_irq = SSL_WRITE_IRQ, + .write_irq_name = "ssl-write", + .symlink_from = "serial", + .symlink_to = "tts", + .mc = { + .name = "ssl", + .config = ssl_config, + .get_config = ssl_get_config, + .remove = ssl_remove, + }, +}; + +/* The array is initialized by line_init, which is an initcall. The + * individual elements are protected by individual semaphores. + */ +static struct line serial_lines[NR_PORTS] = + { [0 ... NR_PORTS - 1] = LINE_INIT(CONFIG_SSL_CHAN, &driver) }; + +static struct lines lines = LINES_INIT(NR_PORTS); + +static int ssl_config(char *str) +{ + return(line_config(serial_lines, + sizeof(serial_lines)/sizeof(serial_lines[0]), str)); +} + +static int ssl_get_config(char *dev, char *str, int size, char **error_out) +{ + return(line_get_config(dev, serial_lines, + sizeof(serial_lines)/sizeof(serial_lines[0]), + str, size, error_out)); +} + +static int ssl_remove(char *str) +{ + return(line_remove(serial_lines, + sizeof(serial_lines)/sizeof(serial_lines[0]), str)); +} + +int ssl_open(struct tty_struct *tty, struct file *filp) +{ + return line_open(serial_lines, tty, &opts); +} + +#if 0 +static int ssl_chars_in_buffer(struct tty_struct *tty) +{ + return(0); +} + +static void ssl_flush_buffer(struct tty_struct *tty) +{ + return; +} + +static void ssl_throttle(struct tty_struct * tty) +{ + printk(KERN_ERR "Someone should implement ssl_throttle\n"); +} + +static void ssl_unthrottle(struct tty_struct * tty) +{ + printk(KERN_ERR "Someone should implement ssl_unthrottle\n"); +} + +static void ssl_stop(struct tty_struct *tty) +{ + printk(KERN_ERR "Someone should implement ssl_stop\n"); +} + +static void ssl_start(struct tty_struct *tty) +{ + printk(KERN_ERR "Someone should implement ssl_start\n"); +} + +void ssl_hangup(struct tty_struct *tty) +{ +} +#endif + +static struct tty_operations ssl_ops = { + .open = ssl_open, + .close = line_close, + .write = line_write, + .put_char = line_put_char, + .write_room = line_write_room, + .chars_in_buffer = line_chars_in_buffer, + .set_termios = line_set_termios, + .ioctl = line_ioctl, +#if 0 + .flush_chars = ssl_flush_chars, + .flush_buffer = ssl_flush_buffer, + .throttle = ssl_throttle, + .unthrottle = ssl_unthrottle, + .stop = ssl_stop, + .start = ssl_start, + .hangup = ssl_hangup, +#endif +}; + +/* Changed by ssl_init and referenced by ssl_exit, which are both serialized + * by being an initcall and exitcall, respectively. + */ +static int ssl_init_done = 0; + +static void ssl_console_write(struct console *c, const char *string, + unsigned len) +{ + struct line *line = &serial_lines[c->index]; + + down(&line->sem); + console_write_chan(&line->chan_list, string, len); + up(&line->sem); +} + +static struct tty_driver *ssl_console_device(struct console *c, int *index) +{ + *index = c->index; + return ssl_driver; +} + +static int ssl_console_setup(struct console *co, char *options) +{ + struct line *line = &serial_lines[co->index]; + + return console_open_chan(line,co,&opts); +} + +static struct console ssl_cons = { + .name = "ttyS", + .write = ssl_console_write, + .device = ssl_console_device, + .setup = ssl_console_setup, + .flags = CON_PRINTBUFFER, + .index = -1, +}; + +int ssl_init(void) +{ + char *new_title; + + printk(KERN_INFO "Initializing software serial port version %d\n", + ssl_version); + ssl_driver = line_register_devfs(&lines, &driver, &ssl_ops, + serial_lines, ARRAY_SIZE(serial_lines)); + + lines_init(serial_lines, sizeof(serial_lines)/sizeof(serial_lines[0])); + + new_title = add_xterm_umid(opts.xterm_title); + if (new_title != NULL) + opts.xterm_title = new_title; + + ssl_init_done = 1; + register_console(&ssl_cons); + return(0); +} +late_initcall(ssl_init); + +static void ssl_exit(void) +{ + if (!ssl_init_done) + return; + close_lines(serial_lines, + sizeof(serial_lines)/sizeof(serial_lines[0])); +} +__uml_exitcall(ssl_exit); + +static int ssl_chan_setup(char *str) +{ + return(line_setup(serial_lines, + sizeof(serial_lines)/sizeof(serial_lines[0]), + str, 1)); +} + +__setup("ssl", ssl_chan_setup); +__channel_help(ssl_chan_setup, "ssl"); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/ssl.h b/arch/um/drivers/ssl.h new file mode 100644 index 000000000000..98412aa66607 --- /dev/null +++ b/arch/um/drivers/ssl.h @@ -0,0 +1,23 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SSL_H__ +#define __SSL_H__ + +extern int ssl_read(int fd, int line); +extern void ssl_receive_char(int line, char ch); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/stderr_console.c b/arch/um/drivers/stderr_console.c new file mode 100644 index 000000000000..98565b53d170 --- /dev/null +++ b/arch/um/drivers/stderr_console.c @@ -0,0 +1,45 @@ +#include <linux/init.h> +#include <linux/console.h> + +#include "chan_user.h" + +/* ----------------------------------------------------------------------------- */ +/* trivial console driver -- simply dump everything to stderr */ + +/* + * Don't register by default -- as this registeres very early in the + * boot process it becomes the default console. And as this isn't a + * real tty driver init isn't able to open /dev/console then. + * + * In most cases this isn't what you want ... + */ +static int use_stderr_console = 0; + +static void stderr_console_write(struct console *console, const char *string, + unsigned len) +{ + generic_write(2 /* stderr */, string, len, NULL); +} + +static struct console stderr_console = { + .name "stderr", + .write stderr_console_write, + .flags CON_PRINTBUFFER, +}; + +static int __init stderr_console_init(void) +{ + if (use_stderr_console) + register_console(&stderr_console); + return 0; +} +console_initcall(stderr_console_init); + +static int stderr_setup(char *str) +{ + if (!str) + return 0; + use_stderr_console = simple_strtoul(str,&str,0); + return 1; +} +__setup("stderr=", stderr_setup); diff --git a/arch/um/drivers/stdio_console.c b/arch/um/drivers/stdio_console.c new file mode 100644 index 000000000000..e604d7c87695 --- /dev/null +++ b/arch/um/drivers/stdio_console.c @@ -0,0 +1,205 @@ +/* + * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/posix_types.h" +#include "linux/tty.h" +#include "linux/tty_flip.h" +#include "linux/types.h" +#include "linux/major.h" +#include "linux/kdev_t.h" +#include "linux/console.h" +#include "linux/string.h" +#include "linux/sched.h" +#include "linux/list.h" +#include "linux/init.h" +#include "linux/interrupt.h" +#include "linux/slab.h" +#include "linux/hardirq.h" +#include "asm/current.h" +#include "asm/irq.h" +#include "stdio_console.h" +#include "line.h" +#include "chan_kern.h" +#include "user_util.h" +#include "kern_util.h" +#include "irq_user.h" +#include "mconsole_kern.h" +#include "init.h" +#include "2_5compat.h" + +#define MAX_TTYS (16) + +/* ----------------------------------------------------------------------------- */ + +/* Referenced only by tty_driver below - presumably it's locked correctly + * by the tty driver. + */ + +static struct tty_driver *console_driver; + +void stdio_announce(char *dev_name, int dev) +{ + printk(KERN_INFO "Virtual console %d assigned device '%s'\n", dev, + dev_name); +} + +static struct chan_opts opts = { + .announce = stdio_announce, + .xterm_title = "Virtual Console #%d", + .raw = 1, + .tramp_stack = 0, + .in_kernel = 1, +}; + +static int con_config(char *str); +static int con_get_config(char *dev, char *str, int size, char **error_out); +static int con_remove(char *str); + +static struct line_driver driver = { + .name = "UML console", + .device_name = "tty", + .devfs_name = "vc/", + .major = TTY_MAJOR, + .minor_start = 0, + .type = TTY_DRIVER_TYPE_CONSOLE, + .subtype = SYSTEM_TYPE_CONSOLE, + .read_irq = CONSOLE_IRQ, + .read_irq_name = "console", + .write_irq = CONSOLE_WRITE_IRQ, + .write_irq_name = "console-write", + .symlink_from = "ttys", + .symlink_to = "vc", + .mc = { + .name = "con", + .config = con_config, + .get_config = con_get_config, + .remove = con_remove, + }, +}; + +static struct lines console_lines = LINES_INIT(MAX_TTYS); + +/* The array is initialized by line_init, which is an initcall. The + * individual elements are protected by individual semaphores. + */ +struct line vts[MAX_TTYS] = { LINE_INIT(CONFIG_CON_ZERO_CHAN, &driver), + [ 1 ... MAX_TTYS - 1 ] = + LINE_INIT(CONFIG_CON_CHAN, &driver) }; + +static int con_config(char *str) +{ + return(line_config(vts, sizeof(vts)/sizeof(vts[0]), str)); +} + +static int con_get_config(char *dev, char *str, int size, char **error_out) +{ + return(line_get_config(dev, vts, sizeof(vts)/sizeof(vts[0]), str, + size, error_out)); +} + +static int con_remove(char *str) +{ + return(line_remove(vts, sizeof(vts)/sizeof(vts[0]), str)); +} + +static int con_open(struct tty_struct *tty, struct file *filp) +{ + return line_open(vts, tty, &opts); +} + +static int con_init_done = 0; + +static struct tty_operations console_ops = { + .open = con_open, + .close = line_close, + .write = line_write, + .write_room = line_write_room, + .chars_in_buffer = line_chars_in_buffer, + .set_termios = line_set_termios, + .ioctl = line_ioctl, +}; + +static void uml_console_write(struct console *console, const char *string, + unsigned len) +{ + struct line *line = &vts[console->index]; + + down(&line->sem); + console_write_chan(&line->chan_list, string, len); + up(&line->sem); +} + +static struct tty_driver *uml_console_device(struct console *c, int *index) +{ + *index = c->index; + return console_driver; +} + +static int uml_console_setup(struct console *co, char *options) +{ + struct line *line = &vts[co->index]; + + return console_open_chan(line,co,&opts); +} + +static struct console stdiocons = { + .name = "tty", + .write = uml_console_write, + .device = uml_console_device, + .setup = uml_console_setup, + .flags = CON_PRINTBUFFER, + .index = -1, + .data = &vts, +}; + +int stdio_init(void) +{ + char *new_title; + + console_driver = line_register_devfs(&console_lines, &driver, + &console_ops, vts, + ARRAY_SIZE(vts)); + if (NULL == console_driver) + return -1; + printk(KERN_INFO "Initialized stdio console driver\n"); + + lines_init(vts, sizeof(vts)/sizeof(vts[0])); + + new_title = add_xterm_umid(opts.xterm_title); + if(new_title != NULL) + opts.xterm_title = new_title; + + con_init_done = 1; + register_console(&stdiocons); + return(0); +} +late_initcall(stdio_init); + +static void console_exit(void) +{ + if (!con_init_done) + return; + close_lines(vts, sizeof(vts)/sizeof(vts[0])); +} +__uml_exitcall(console_exit); + +static int console_chan_setup(char *str) +{ + return(line_setup(vts, sizeof(vts)/sizeof(vts[0]), str, 1)); +} +__setup("con", console_chan_setup); +__channel_help(console_chan_setup, "con"); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/stdio_console.h b/arch/um/drivers/stdio_console.h new file mode 100644 index 000000000000..505a3d5bea5e --- /dev/null +++ b/arch/um/drivers/stdio_console.h @@ -0,0 +1,21 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __STDIO_CONSOLE_H +#define __STDIO_CONSOLE_H + +extern void save_console_flags(void); +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/tty.c b/arch/um/drivers/tty.c new file mode 100644 index 000000000000..6fbb670ee274 --- /dev/null +++ b/arch/um/drivers/tty.c @@ -0,0 +1,92 @@ +/* + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <termios.h> +#include <errno.h> +#include <unistd.h> +#include "chan_user.h" +#include "user_util.h" +#include "user.h" +#include "os.h" + +struct tty_chan { + char *dev; + int raw; + struct termios tt; +}; + +static void *tty_chan_init(char *str, int device, struct chan_opts *opts) +{ + struct tty_chan *data; + + if(*str != ':'){ + printk("tty_init : channel type 'tty' must specify " + "a device\n"); + return(NULL); + } + str++; + + data = um_kmalloc(sizeof(*data)); + if(data == NULL) + return(NULL); + *data = ((struct tty_chan) { .dev = str, + .raw = opts->raw }); + + return(data); +} + +static int tty_open(int input, int output, int primary, void *d, + char **dev_out) +{ + struct tty_chan *data = d; + int fd, err; + + fd = os_open_file(data->dev, of_set_rw(OPENFLAGS(), input, output), 0); + if(fd < 0) return(fd); + if(data->raw){ + CATCH_EINTR(err = tcgetattr(fd, &data->tt)); + if(err) + return(err); + + err = raw(fd); + if(err) + return(err); + } + + *dev_out = data->dev; + return(fd); +} + +static int tty_console_write(int fd, const char *buf, int n, void *d) +{ + struct tty_chan *data = d; + + return(generic_console_write(fd, buf, n, &data->tt)); +} + +struct chan_ops tty_ops = { + .type = "tty", + .init = tty_chan_init, + .open = tty_open, + .close = generic_close, + .read = generic_read, + .write = generic_write, + .console_write = tty_console_write, + .window_size = generic_window_size, + .free = generic_free, + .winch = 0, +}; + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c new file mode 100644 index 000000000000..4d8b165bfa48 --- /dev/null +++ b/arch/um/drivers/ubd_kern.c @@ -0,0 +1,1669 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +/* 2001-09-28...2002-04-17 + * Partition stuff by James_McMechan@hotmail.com + * old style ubd by setting UBD_SHIFT to 0 + * 2002-09-27...2002-10-18 massive tinkering for 2.5 + * partitions have changed in 2.5 + * 2003-01-29 more tinkering for 2.5.59-1 + * This should now address the sysfs problems and has + * the symlink for devfs to allow for booting with + * the common /dev/ubd/discX/... names rather than + * only /dev/ubdN/discN this version also has lots of + * clean ups preparing for ubd-many. + * James McMechan + */ + +#define MAJOR_NR UBD_MAJOR +#define UBD_SHIFT 4 + +#include "linux/config.h" +#include "linux/module.h" +#include "linux/blkdev.h" +#include "linux/hdreg.h" +#include "linux/init.h" +#include "linux/devfs_fs_kernel.h" +#include "linux/cdrom.h" +#include "linux/proc_fs.h" +#include "linux/ctype.h" +#include "linux/capability.h" +#include "linux/mm.h" +#include "linux/vmalloc.h" +#include "linux/blkpg.h" +#include "linux/genhd.h" +#include "linux/spinlock.h" +#include "asm/segment.h" +#include "asm/uaccess.h" +#include "asm/irq.h" +#include "asm/types.h" +#include "asm/tlbflush.h" +#include "user_util.h" +#include "mem_user.h" +#include "kern_util.h" +#include "kern.h" +#include "mconsole_kern.h" +#include "init.h" +#include "irq_user.h" +#include "irq_kern.h" +#include "ubd_user.h" +#include "2_5compat.h" +#include "os.h" +#include "mem.h" +#include "mem_kern.h" +#include "cow.h" + +enum ubd_req { UBD_READ, UBD_WRITE, UBD_MMAP }; + +struct io_thread_req { + enum ubd_req op; + int fds[2]; + unsigned long offsets[2]; + unsigned long long offset; + unsigned long length; + char *buffer; + int sectorsize; + unsigned long sector_mask; + unsigned long long cow_offset; + unsigned long bitmap_words[2]; + int map_fd; + unsigned long long map_offset; + int error; +}; + +extern int open_ubd_file(char *file, struct openflags *openflags, + char **backing_file_out, int *bitmap_offset_out, + unsigned long *bitmap_len_out, int *data_offset_out, + int *create_cow_out); +extern int create_cow_file(char *cow_file, char *backing_file, + struct openflags flags, int sectorsize, + int alignment, int *bitmap_offset_out, + unsigned long *bitmap_len_out, + int *data_offset_out); +extern int read_cow_bitmap(int fd, void *buf, int offset, int len); +extern void do_io(struct io_thread_req *req); + +static inline int ubd_test_bit(__u64 bit, unsigned char *data) +{ + __u64 n; + int bits, off; + + bits = sizeof(data[0]) * 8; + n = bit / bits; + off = bit % bits; + return((data[n] & (1 << off)) != 0); +} + +static inline void ubd_set_bit(__u64 bit, unsigned char *data) +{ + __u64 n; + int bits, off; + + bits = sizeof(data[0]) * 8; + n = bit / bits; + off = bit % bits; + data[n] |= (1 << off); +} +/*End stuff from ubd_user.h*/ + +#define DRIVER_NAME "uml-blkdev" + +static DEFINE_SPINLOCK(ubd_io_lock); +static DEFINE_SPINLOCK(ubd_lock); + +static void (*do_ubd)(void); + +static int ubd_open(struct inode * inode, struct file * filp); +static int ubd_release(struct inode * inode, struct file * file); +static int ubd_ioctl(struct inode * inode, struct file * file, + unsigned int cmd, unsigned long arg); + +#define MAX_DEV (8) + +/* Changed in early boot */ +static int ubd_do_mmap = 0; +#define UBD_MMAP_BLOCK_SIZE PAGE_SIZE + +static struct block_device_operations ubd_blops = { + .owner = THIS_MODULE, + .open = ubd_open, + .release = ubd_release, + .ioctl = ubd_ioctl, +}; + +/* Protected by the queue_lock */ +static request_queue_t *ubd_queue; + +/* Protected by ubd_lock */ +static int fake_major = MAJOR_NR; + +static struct gendisk *ubd_gendisk[MAX_DEV]; +static struct gendisk *fake_gendisk[MAX_DEV]; + +#ifdef CONFIG_BLK_DEV_UBD_SYNC +#define OPEN_FLAGS ((struct openflags) { .r = 1, .w = 1, .s = 1, .c = 0, \ + .cl = 1 }) +#else +#define OPEN_FLAGS ((struct openflags) { .r = 1, .w = 1, .s = 0, .c = 0, \ + .cl = 1 }) +#endif + +/* Not protected - changed only in ubd_setup_common and then only to + * to enable O_SYNC. + */ +static struct openflags global_openflags = OPEN_FLAGS; + +struct cow { + char *file; + int fd; + unsigned long *bitmap; + unsigned long bitmap_len; + int bitmap_offset; + int data_offset; +}; + +struct ubd { + char *file; + int count; + int fd; + __u64 size; + struct openflags boot_openflags; + struct openflags openflags; + int no_cow; + struct cow cow; + struct platform_device pdev; + + int map_writes; + int map_reads; + int nomap_writes; + int nomap_reads; + int write_maps; +}; + +#define DEFAULT_COW { \ + .file = NULL, \ + .fd = -1, \ + .bitmap = NULL, \ + .bitmap_offset = 0, \ + .data_offset = 0, \ +} + +#define DEFAULT_UBD { \ + .file = NULL, \ + .count = 0, \ + .fd = -1, \ + .size = -1, \ + .boot_openflags = OPEN_FLAGS, \ + .openflags = OPEN_FLAGS, \ + .no_cow = 0, \ + .cow = DEFAULT_COW, \ + .map_writes = 0, \ + .map_reads = 0, \ + .nomap_writes = 0, \ + .nomap_reads = 0, \ + .write_maps = 0, \ +} + +struct ubd ubd_dev[MAX_DEV] = { [ 0 ... MAX_DEV - 1 ] = DEFAULT_UBD }; + +static int ubd0_init(void) +{ + struct ubd *dev = &ubd_dev[0]; + + if(dev->file == NULL) + dev->file = "root_fs"; + return(0); +} + +__initcall(ubd0_init); + +/* Only changed by fake_ide_setup which is a setup */ +static int fake_ide = 0; +static struct proc_dir_entry *proc_ide_root = NULL; +static struct proc_dir_entry *proc_ide = NULL; + +static void make_proc_ide(void) +{ + proc_ide_root = proc_mkdir("ide", NULL); + proc_ide = proc_mkdir("ide0", proc_ide_root); +} + +static int proc_ide_read_media(char *page, char **start, off_t off, int count, + int *eof, void *data) +{ + int len; + + strcpy(page, "disk\n"); + len = strlen("disk\n"); + len -= off; + if (len < count){ + *eof = 1; + if (len <= 0) return 0; + } + else len = count; + *start = page + off; + return len; +} + +static void make_ide_entries(char *dev_name) +{ + struct proc_dir_entry *dir, *ent; + char name[64]; + + if(proc_ide_root == NULL) make_proc_ide(); + + dir = proc_mkdir(dev_name, proc_ide); + if(!dir) return; + + ent = create_proc_entry("media", S_IFREG|S_IRUGO, dir); + if(!ent) return; + ent->nlink = 1; + ent->data = NULL; + ent->read_proc = proc_ide_read_media; + ent->write_proc = NULL; + sprintf(name,"ide0/%s", dev_name); + proc_symlink(dev_name, proc_ide_root, name); +} + +static int fake_ide_setup(char *str) +{ + fake_ide = 1; + return(1); +} + +__setup("fake_ide", fake_ide_setup); + +__uml_help(fake_ide_setup, +"fake_ide\n" +" Create ide0 entries that map onto ubd devices.\n\n" +); + +static int parse_unit(char **ptr) +{ + char *str = *ptr, *end; + int n = -1; + + if(isdigit(*str)) { + n = simple_strtoul(str, &end, 0); + if(end == str) + return(-1); + *ptr = end; + } + else if (('a' <= *str) && (*str <= 'h')) { + n = *str - 'a'; + str++; + *ptr = str; + } + return(n); +} + +static int ubd_setup_common(char *str, int *index_out) +{ + struct ubd *dev; + struct openflags flags = global_openflags; + char *backing_file; + int n, err, i; + + if(index_out) *index_out = -1; + n = *str; + if(n == '='){ + char *end; + int major; + + str++; + if(!strcmp(str, "mmap")){ + CHOOSE_MODE(printk("mmap not supported by the ubd " + "driver in tt mode\n"), + ubd_do_mmap = 1); + return(0); + } + + if(!strcmp(str, "sync")){ + global_openflags = of_sync(global_openflags); + return(0); + } + major = simple_strtoul(str, &end, 0); + if((*end != '\0') || (end == str)){ + printk(KERN_ERR + "ubd_setup : didn't parse major number\n"); + return(1); + } + + err = 1; + spin_lock(&ubd_lock); + if(fake_major != MAJOR_NR){ + printk(KERN_ERR "Can't assign a fake major twice\n"); + goto out1; + } + + fake_major = major; + + printk(KERN_INFO "Setting extra ubd major number to %d\n", + major); + err = 0; + out1: + spin_unlock(&ubd_lock); + return(err); + } + + n = parse_unit(&str); + if(n < 0){ + printk(KERN_ERR "ubd_setup : couldn't parse unit number " + "'%s'\n", str); + return(1); + } + if(n >= MAX_DEV){ + printk(KERN_ERR "ubd_setup : index %d out of range " + "(%d devices, from 0 to %d)\n", n, MAX_DEV, MAX_DEV - 1); + return(1); + } + + err = 1; + spin_lock(&ubd_lock); + + dev = &ubd_dev[n]; + if(dev->file != NULL){ + printk(KERN_ERR "ubd_setup : device already configured\n"); + goto out; + } + + if (index_out) + *index_out = n; + + for (i = 0; i < 4; i++) { + switch (*str) { + case 'r': + flags.w = 0; + break; + case 's': + flags.s = 1; + break; + case 'd': + dev->no_cow = 1; + break; + case '=': + str++; + goto break_loop; + default: + printk(KERN_ERR "ubd_setup : Expected '=' or flag letter (r,s or d)\n"); + goto out; + } + str++; + } + + if (*str == '=') + printk(KERN_ERR "ubd_setup : Too many flags specified\n"); + else + printk(KERN_ERR "ubd_setup : Expected '='\n"); + goto out; + +break_loop: + err = 0; + backing_file = strchr(str, ','); + + if (!backing_file) { + backing_file = strchr(str, ':'); + } + + if(backing_file){ + if(dev->no_cow) + printk(KERN_ERR "Can't specify both 'd' and a " + "cow file\n"); + else { + *backing_file = '\0'; + backing_file++; + } + } + dev->file = str; + dev->cow.file = backing_file; + dev->boot_openflags = flags; +out: + spin_unlock(&ubd_lock); + return(err); +} + +static int ubd_setup(char *str) +{ + ubd_setup_common(str, NULL); + return(1); +} + +__setup("ubd", ubd_setup); +__uml_help(ubd_setup, +"ubd<n><flags>=<filename>[(:|,)<filename2>]\n" +" This is used to associate a device with a file in the underlying\n" +" filesystem. When specifying two filenames, the first one is the\n" +" COW name and the second is the backing file name. As separator you can\n" +" use either a ':' or a ',': the first one allows writing things like;\n" +" ubd0=~/Uml/root_cow:~/Uml/root_backing_file\n" +" while with a ',' the shell would not expand the 2nd '~'.\n" +" When using only one filename, UML will detect whether to thread it like\n" +" a COW file or a backing file. To override this detection, add the 'd'\n" +" flag:\n" +" ubd0d=BackingFile\n" +" Usually, there is a filesystem in the file, but \n" +" that's not required. Swap devices containing swap files can be\n" +" specified like this. Also, a file which doesn't contain a\n" +" filesystem can have its contents read in the virtual \n" +" machine by running 'dd' on the device. <n> must be in the range\n" +" 0 to 7. Appending an 'r' to the number will cause that device\n" +" to be mounted read-only. For example ubd1r=./ext_fs. Appending\n" +" an 's' will cause data to be written to disk on the host immediately.\n\n" +); + +static int udb_setup(char *str) +{ + printk("udb%s specified on command line is almost certainly a ubd -> " + "udb TYPO\n", str); + return(1); +} + +__setup("udb", udb_setup); +__uml_help(udb_setup, +"udb\n" +" This option is here solely to catch ubd -> udb typos, which can be\n\n" +" to impossible to catch visually unless you specifically look for\n\n" +" them. The only result of any option starting with 'udb' is an error\n\n" +" in the boot output.\n\n" +); + +static int fakehd_set = 0; +static int fakehd(char *str) +{ + printk(KERN_INFO "fakehd : Changing ubd name to \"hd\".\n"); + fakehd_set = 1; + return 1; +} + +__setup("fakehd", fakehd); +__uml_help(fakehd, +"fakehd\n" +" Change the ubd device name to \"hd\".\n\n" +); + +static void do_ubd_request(request_queue_t * q); + +/* Only changed by ubd_init, which is an initcall. */ +int thread_fd = -1; + +/* Changed by ubd_handler, which is serialized because interrupts only + * happen on CPU 0. + */ +int intr_count = 0; + +/* call ubd_finish if you need to serialize */ +static void __ubd_finish(struct request *req, int error) +{ + int nsect; + + if(error){ + end_request(req, 0); + return; + } + nsect = req->current_nr_sectors; + req->sector += nsect; + req->buffer += nsect << 9; + req->errors = 0; + req->nr_sectors -= nsect; + req->current_nr_sectors = 0; + end_request(req, 1); +} + +static inline void ubd_finish(struct request *req, int error) +{ + spin_lock(&ubd_io_lock); + __ubd_finish(req, error); + spin_unlock(&ubd_io_lock); +} + +/* Called without ubd_io_lock held */ +static void ubd_handler(void) +{ + struct io_thread_req req; + struct request *rq = elv_next_request(ubd_queue); + int n, err; + + do_ubd = NULL; + intr_count++; + n = os_read_file(thread_fd, &req, sizeof(req)); + if(n != sizeof(req)){ + printk(KERN_ERR "Pid %d - spurious interrupt in ubd_handler, " + "err = %d\n", os_getpid(), -n); + spin_lock(&ubd_io_lock); + end_request(rq, 0); + spin_unlock(&ubd_io_lock); + return; + } + + if((req.op != UBD_MMAP) && + ((req.offset != ((__u64) (rq->sector)) << 9) || + (req.length != (rq->current_nr_sectors) << 9))) + panic("I/O op mismatch"); + + if(req.map_fd != -1){ + err = physmem_subst_mapping(req.buffer, req.map_fd, + req.map_offset, 1); + if(err) + printk("ubd_handler - physmem_subst_mapping failed, " + "err = %d\n", -err); + } + + ubd_finish(rq, req.error); + reactivate_fd(thread_fd, UBD_IRQ); + do_ubd_request(ubd_queue); +} + +static irqreturn_t ubd_intr(int irq, void *dev, struct pt_regs *unused) +{ + ubd_handler(); + return(IRQ_HANDLED); +} + +/* Only changed by ubd_init, which is an initcall. */ +static int io_pid = -1; + +void kill_io_thread(void) +{ + if(io_pid != -1) + os_kill_process(io_pid, 1); +} + +__uml_exitcall(kill_io_thread); + +static int ubd_file_size(struct ubd *dev, __u64 *size_out) +{ + char *file; + + file = dev->cow.file ? dev->cow.file : dev->file; + return(os_file_size(file, size_out)); +} + +static void ubd_close(struct ubd *dev) +{ + if(ubd_do_mmap) + physmem_forget_descriptor(dev->fd); + os_close_file(dev->fd); + if(dev->cow.file == NULL) + return; + + if(ubd_do_mmap) + physmem_forget_descriptor(dev->cow.fd); + os_close_file(dev->cow.fd); + vfree(dev->cow.bitmap); + dev->cow.bitmap = NULL; +} + +static int ubd_open_dev(struct ubd *dev) +{ + struct openflags flags; + char **back_ptr; + int err, create_cow, *create_ptr; + + dev->openflags = dev->boot_openflags; + create_cow = 0; + create_ptr = (dev->cow.file != NULL) ? &create_cow : NULL; + back_ptr = dev->no_cow ? NULL : &dev->cow.file; + dev->fd = open_ubd_file(dev->file, &dev->openflags, back_ptr, + &dev->cow.bitmap_offset, &dev->cow.bitmap_len, + &dev->cow.data_offset, create_ptr); + + if((dev->fd == -ENOENT) && create_cow){ + dev->fd = create_cow_file(dev->file, dev->cow.file, + dev->openflags, 1 << 9, PAGE_SIZE, + &dev->cow.bitmap_offset, + &dev->cow.bitmap_len, + &dev->cow.data_offset); + if(dev->fd >= 0){ + printk(KERN_INFO "Creating \"%s\" as COW file for " + "\"%s\"\n", dev->file, dev->cow.file); + } + } + + if(dev->fd < 0){ + printk("Failed to open '%s', errno = %d\n", dev->file, + -dev->fd); + return(dev->fd); + } + + if(dev->cow.file != NULL){ + err = -ENOMEM; + dev->cow.bitmap = (void *) vmalloc(dev->cow.bitmap_len); + if(dev->cow.bitmap == NULL){ + printk(KERN_ERR "Failed to vmalloc COW bitmap\n"); + goto error; + } + flush_tlb_kernel_vm(); + + err = read_cow_bitmap(dev->fd, dev->cow.bitmap, + dev->cow.bitmap_offset, + dev->cow.bitmap_len); + if(err < 0) + goto error; + + flags = dev->openflags; + flags.w = 0; + err = open_ubd_file(dev->cow.file, &flags, NULL, NULL, NULL, + NULL, NULL); + if(err < 0) goto error; + dev->cow.fd = err; + } + return(0); + error: + os_close_file(dev->fd); + return(err); +} + +static int ubd_new_disk(int major, u64 size, int unit, + struct gendisk **disk_out) + +{ + struct gendisk *disk; + char from[sizeof("ubd/nnnnn\0")], to[sizeof("discnnnnn/disc\0")]; + int err; + + disk = alloc_disk(1 << UBD_SHIFT); + if(disk == NULL) + return(-ENOMEM); + + disk->major = major; + disk->first_minor = unit << UBD_SHIFT; + disk->fops = &ubd_blops; + set_capacity(disk, size / 512); + if(major == MAJOR_NR){ + sprintf(disk->disk_name, "ubd%c", 'a' + unit); + sprintf(disk->devfs_name, "ubd/disc%d", unit); + sprintf(from, "ubd/%d", unit); + sprintf(to, "disc%d/disc", unit); + err = devfs_mk_symlink(from, to); + if(err) + printk("ubd_new_disk failed to make link from %s to " + "%s, error = %d\n", from, to, err); + } + else { + sprintf(disk->disk_name, "ubd_fake%d", unit); + sprintf(disk->devfs_name, "ubd_fake/disc%d", unit); + } + + /* sysfs register (not for ide fake devices) */ + if (major == MAJOR_NR) { + ubd_dev[unit].pdev.id = unit; + ubd_dev[unit].pdev.name = DRIVER_NAME; + platform_device_register(&ubd_dev[unit].pdev); + disk->driverfs_dev = &ubd_dev[unit].pdev.dev; + } + + disk->private_data = &ubd_dev[unit]; + disk->queue = ubd_queue; + add_disk(disk); + + *disk_out = disk; + return 0; +} + +#define ROUND_BLOCK(n) ((n + ((1 << 9) - 1)) & (-1 << 9)) + +static int ubd_add(int n) +{ + struct ubd *dev = &ubd_dev[n]; + int err; + + if(dev->file == NULL) + return(-ENODEV); + + if (ubd_open_dev(dev)) + return(-ENODEV); + + err = ubd_file_size(dev, &dev->size); + if(err < 0) + return(err); + + dev->size = ROUND_BLOCK(dev->size); + + err = ubd_new_disk(MAJOR_NR, dev->size, n, &ubd_gendisk[n]); + if(err) + return(err); + + if(fake_major != MAJOR_NR) + ubd_new_disk(fake_major, dev->size, n, + &fake_gendisk[n]); + + /* perhaps this should also be under the "if (fake_major)" above */ + /* using the fake_disk->disk_name and also the fakehd_set name */ + if (fake_ide) + make_ide_entries(ubd_gendisk[n]->disk_name); + + ubd_close(dev); + return 0; +} + +static int ubd_config(char *str) +{ + int n, err; + + str = uml_strdup(str); + if(str == NULL){ + printk(KERN_ERR "ubd_config failed to strdup string\n"); + return(1); + } + err = ubd_setup_common(str, &n); + if(err){ + kfree(str); + return(-1); + } + if(n == -1) return(0); + + spin_lock(&ubd_lock); + err = ubd_add(n); + if(err) + ubd_dev[n].file = NULL; + spin_unlock(&ubd_lock); + + return(err); +} + +static int ubd_get_config(char *name, char *str, int size, char **error_out) +{ + struct ubd *dev; + int n, len = 0; + + n = parse_unit(&name); + if((n >= MAX_DEV) || (n < 0)){ + *error_out = "ubd_get_config : device number out of range"; + return(-1); + } + + dev = &ubd_dev[n]; + spin_lock(&ubd_lock); + + if(dev->file == NULL){ + CONFIG_CHUNK(str, size, len, "", 1); + goto out; + } + + CONFIG_CHUNK(str, size, len, dev->file, 0); + + if(dev->cow.file != NULL){ + CONFIG_CHUNK(str, size, len, ",", 0); + CONFIG_CHUNK(str, size, len, dev->cow.file, 1); + } + else CONFIG_CHUNK(str, size, len, "", 1); + + out: + spin_unlock(&ubd_lock); + return(len); +} + +static int ubd_remove(char *str) +{ + struct ubd *dev; + int n, err = -ENODEV; + + n = parse_unit(&str); + + if((n < 0) || (n >= MAX_DEV)) + return(err); + + dev = &ubd_dev[n]; + if(dev->count > 0) + return(-EBUSY); /* you cannot remove a open disk */ + + err = 0; + spin_lock(&ubd_lock); + + if(ubd_gendisk[n] == NULL) + goto out; + + del_gendisk(ubd_gendisk[n]); + put_disk(ubd_gendisk[n]); + ubd_gendisk[n] = NULL; + + if(fake_gendisk[n] != NULL){ + del_gendisk(fake_gendisk[n]); + put_disk(fake_gendisk[n]); + fake_gendisk[n] = NULL; + } + + platform_device_unregister(&dev->pdev); + *dev = ((struct ubd) DEFAULT_UBD); + err = 0; + out: + spin_unlock(&ubd_lock); + return(err); +} + +static struct mc_device ubd_mc = { + .name = "ubd", + .config = ubd_config, + .get_config = ubd_get_config, + .remove = ubd_remove, +}; + +static int ubd_mc_init(void) +{ + mconsole_register_dev(&ubd_mc); + return 0; +} + +__initcall(ubd_mc_init); + +static struct device_driver ubd_driver = { + .name = DRIVER_NAME, + .bus = &platform_bus_type, +}; + +int ubd_init(void) +{ + int i; + + devfs_mk_dir("ubd"); + if (register_blkdev(MAJOR_NR, "ubd")) + return -1; + + ubd_queue = blk_init_queue(do_ubd_request, &ubd_io_lock); + if (!ubd_queue) { + unregister_blkdev(MAJOR_NR, "ubd"); + return -1; + } + + if (fake_major != MAJOR_NR) { + char name[sizeof("ubd_nnn\0")]; + + snprintf(name, sizeof(name), "ubd_%d", fake_major); + devfs_mk_dir(name); + if (register_blkdev(fake_major, "ubd")) + return -1; + } + driver_register(&ubd_driver); + for (i = 0; i < MAX_DEV; i++) + ubd_add(i); + return 0; +} + +late_initcall(ubd_init); + +int ubd_driver_init(void){ + unsigned long stack; + int err; + + /* Set by CONFIG_BLK_DEV_UBD_SYNC or ubd=sync.*/ + if(global_openflags.s){ + printk(KERN_INFO "ubd: Synchronous mode\n"); + /* Letting ubd=sync be like using ubd#s= instead of ubd#= is + * enough. So use anyway the io thread. */ + } + stack = alloc_stack(0, 0); + io_pid = start_io_thread(stack + PAGE_SIZE - sizeof(void *), + &thread_fd); + if(io_pid < 0){ + printk(KERN_ERR + "ubd : Failed to start I/O thread (errno = %d) - " + "falling back to synchronous I/O\n", -io_pid); + io_pid = -1; + return(0); + } + err = um_request_irq(UBD_IRQ, thread_fd, IRQ_READ, ubd_intr, + SA_INTERRUPT, "ubd", ubd_dev); + if(err != 0) + printk(KERN_ERR "um_request_irq failed - errno = %d\n", -err); + return(err); +} + +device_initcall(ubd_driver_init); + +static int ubd_open(struct inode *inode, struct file *filp) +{ + struct gendisk *disk = inode->i_bdev->bd_disk; + struct ubd *dev = disk->private_data; + int err = 0; + + if(dev->count == 0){ + err = ubd_open_dev(dev); + if(err){ + printk(KERN_ERR "%s: Can't open \"%s\": errno = %d\n", + disk->disk_name, dev->file, -err); + goto out; + } + } + dev->count++; + if((filp->f_mode & FMODE_WRITE) && !dev->openflags.w){ + if(--dev->count == 0) ubd_close(dev); + err = -EROFS; + } + out: + return(err); +} + +static int ubd_release(struct inode * inode, struct file * file) +{ + struct gendisk *disk = inode->i_bdev->bd_disk; + struct ubd *dev = disk->private_data; + + if(--dev->count == 0) + ubd_close(dev); + return(0); +} + +static void cowify_bitmap(__u64 io_offset, int length, unsigned long *cow_mask, + __u64 *cow_offset, unsigned long *bitmap, + __u64 bitmap_offset, unsigned long *bitmap_words, + __u64 bitmap_len) +{ + __u64 sector = io_offset >> 9; + int i, update_bitmap = 0; + + for(i = 0; i < length >> 9; i++){ + if(cow_mask != NULL) + ubd_set_bit(i, (unsigned char *) cow_mask); + if(ubd_test_bit(sector + i, (unsigned char *) bitmap)) + continue; + + update_bitmap = 1; + ubd_set_bit(sector + i, (unsigned char *) bitmap); + } + + if(!update_bitmap) + return; + + *cow_offset = sector / (sizeof(unsigned long) * 8); + + /* This takes care of the case where we're exactly at the end of the + * device, and *cow_offset + 1 is off the end. So, just back it up + * by one word. Thanks to Lynn Kerby for the fix and James McMechan + * for the original diagnosis. + */ + if(*cow_offset == ((bitmap_len + sizeof(unsigned long) - 1) / + sizeof(unsigned long) - 1)) + (*cow_offset)--; + + bitmap_words[0] = bitmap[*cow_offset]; + bitmap_words[1] = bitmap[*cow_offset + 1]; + + *cow_offset *= sizeof(unsigned long); + *cow_offset += bitmap_offset; +} + +static void cowify_req(struct io_thread_req *req, unsigned long *bitmap, + __u64 bitmap_offset, __u64 bitmap_len) +{ + __u64 sector = req->offset >> 9; + int i; + + if(req->length > (sizeof(req->sector_mask) * 8) << 9) + panic("Operation too long"); + + if(req->op == UBD_READ) { + for(i = 0; i < req->length >> 9; i++){ + if(ubd_test_bit(sector + i, (unsigned char *) bitmap)) + ubd_set_bit(i, (unsigned char *) + &req->sector_mask); + } + } + else cowify_bitmap(req->offset, req->length, &req->sector_mask, + &req->cow_offset, bitmap, bitmap_offset, + req->bitmap_words, bitmap_len); +} + +static int mmap_fd(struct request *req, struct ubd *dev, __u64 offset) +{ + __u64 sector; + unsigned char *bitmap; + int bit, i; + + /* mmap must have been requested on the command line */ + if(!ubd_do_mmap) + return(-1); + + /* The buffer must be page aligned */ + if(((unsigned long) req->buffer % UBD_MMAP_BLOCK_SIZE) != 0) + return(-1); + + /* The request must be a page long */ + if((req->current_nr_sectors << 9) != PAGE_SIZE) + return(-1); + + if(dev->cow.file == NULL) + return(dev->fd); + + sector = offset >> 9; + bitmap = (unsigned char *) dev->cow.bitmap; + bit = ubd_test_bit(sector, bitmap); + + for(i = 1; i < req->current_nr_sectors; i++){ + if(ubd_test_bit(sector + i, bitmap) != bit) + return(-1); + } + + if(bit || (rq_data_dir(req) == WRITE)) + offset += dev->cow.data_offset; + + /* The data on disk must be page aligned */ + if((offset % UBD_MMAP_BLOCK_SIZE) != 0) + return(-1); + + return(bit ? dev->fd : dev->cow.fd); +} + +static int prepare_mmap_request(struct ubd *dev, int fd, __u64 offset, + struct request *req, + struct io_thread_req *io_req) +{ + int err; + + if(rq_data_dir(req) == WRITE){ + /* Writes are almost no-ops since the new data is already in the + * host page cache + */ + dev->map_writes++; + if(dev->cow.file != NULL) + cowify_bitmap(io_req->offset, io_req->length, + &io_req->sector_mask, &io_req->cow_offset, + dev->cow.bitmap, dev->cow.bitmap_offset, + io_req->bitmap_words, + dev->cow.bitmap_len); + } + else { + int w; + + if((dev->cow.file != NULL) && (fd == dev->cow.fd)) + w = 0; + else w = dev->openflags.w; + + if((dev->cow.file != NULL) && (fd == dev->fd)) + offset += dev->cow.data_offset; + + err = physmem_subst_mapping(req->buffer, fd, offset, w); + if(err){ + printk("physmem_subst_mapping failed, err = %d\n", + -err); + return(1); + } + dev->map_reads++; + } + io_req->op = UBD_MMAP; + io_req->buffer = req->buffer; + return(0); +} + +/* Called with ubd_io_lock held */ +static int prepare_request(struct request *req, struct io_thread_req *io_req) +{ + struct gendisk *disk = req->rq_disk; + struct ubd *dev = disk->private_data; + __u64 offset; + int len, fd; + + if(req->rq_status == RQ_INACTIVE) return(1); + + if((rq_data_dir(req) == WRITE) && !dev->openflags.w){ + printk("Write attempted on readonly ubd device %s\n", + disk->disk_name); + end_request(req, 0); + return(1); + } + + offset = ((__u64) req->sector) << 9; + len = req->current_nr_sectors << 9; + + io_req->fds[0] = (dev->cow.file != NULL) ? dev->cow.fd : dev->fd; + io_req->fds[1] = dev->fd; + io_req->map_fd = -1; + io_req->cow_offset = -1; + io_req->offset = offset; + io_req->length = len; + io_req->error = 0; + io_req->sector_mask = 0; + + fd = mmap_fd(req, dev, io_req->offset); + if(fd > 0){ + /* If mmapping is otherwise OK, but the first access to the + * page is a write, then it's not mapped in yet. So we have + * to write the data to disk first, then we can map the disk + * page in and continue normally from there. + */ + if((rq_data_dir(req) == WRITE) && !is_remapped(req->buffer)){ + io_req->map_fd = dev->fd; + io_req->map_offset = io_req->offset + + dev->cow.data_offset; + dev->write_maps++; + } + else return(prepare_mmap_request(dev, fd, io_req->offset, req, + io_req)); + } + + if(rq_data_dir(req) == READ) + dev->nomap_reads++; + else dev->nomap_writes++; + + io_req->op = (rq_data_dir(req) == READ) ? UBD_READ : UBD_WRITE; + io_req->offsets[0] = 0; + io_req->offsets[1] = dev->cow.data_offset; + io_req->buffer = req->buffer; + io_req->sectorsize = 1 << 9; + + if(dev->cow.file != NULL) + cowify_req(io_req, dev->cow.bitmap, dev->cow.bitmap_offset, + dev->cow.bitmap_len); + + return(0); +} + +/* Called with ubd_io_lock held */ +static void do_ubd_request(request_queue_t *q) +{ + struct io_thread_req io_req; + struct request *req; + int err, n; + + if(thread_fd == -1){ + while((req = elv_next_request(q)) != NULL){ + err = prepare_request(req, &io_req); + if(!err){ + do_io(&io_req); + __ubd_finish(req, io_req.error); + } + } + } + else { + if(do_ubd || (req = elv_next_request(q)) == NULL) + return; + err = prepare_request(req, &io_req); + if(!err){ + do_ubd = ubd_handler; + n = os_write_file(thread_fd, (char *) &io_req, + sizeof(io_req)); + if(n != sizeof(io_req)) + printk("write to io thread failed, " + "errno = %d\n", -n); + } + } +} + +static int ubd_ioctl(struct inode * inode, struct file * file, + unsigned int cmd, unsigned long arg) +{ + struct hd_geometry __user *loc = (struct hd_geometry __user *) arg; + struct ubd *dev = inode->i_bdev->bd_disk->private_data; + struct hd_driveid ubd_id = { + .cyls = 0, + .heads = 128, + .sectors = 32, + }; + + switch (cmd) { + struct hd_geometry g; + struct cdrom_volctrl volume; + case HDIO_GETGEO: + if(!loc) return(-EINVAL); + g.heads = 128; + g.sectors = 32; + g.cylinders = dev->size / (128 * 32 * 512); + g.start = get_start_sect(inode->i_bdev); + return(copy_to_user(loc, &g, sizeof(g)) ? -EFAULT : 0); + + case HDIO_GET_IDENTITY: + ubd_id.cyls = dev->size / (128 * 32 * 512); + if(copy_to_user((char __user *) arg, (char *) &ubd_id, + sizeof(ubd_id))) + return(-EFAULT); + return(0); + + case CDROMVOLREAD: + if(copy_from_user(&volume, (char __user *) arg, sizeof(volume))) + return(-EFAULT); + volume.channel0 = 255; + volume.channel1 = 255; + volume.channel2 = 255; + volume.channel3 = 255; + if(copy_to_user((char __user *) arg, &volume, sizeof(volume))) + return(-EFAULT); + return(0); + } + return(-EINVAL); +} + +static int ubd_check_remapped(int fd, unsigned long address, int is_write, + __u64 offset) +{ + __u64 bitmap_offset; + unsigned long new_bitmap[2]; + int i, err, n; + + /* If it's not a write access, we can't do anything about it */ + if(!is_write) + return(0); + + /* We have a write */ + for(i = 0; i < sizeof(ubd_dev) / sizeof(ubd_dev[0]); i++){ + struct ubd *dev = &ubd_dev[i]; + + if((dev->fd != fd) && (dev->cow.fd != fd)) + continue; + + /* It's a write to a ubd device */ + + if(!dev->openflags.w){ + /* It's a write access on a read-only device - probably + * shouldn't happen. If the kernel is trying to change + * something with no intention of writing it back out, + * then this message will clue us in that this needs + * fixing + */ + printk("Write access to mapped page from readonly ubd " + "device %d\n", i); + return(0); + } + + /* It's a write to a writeable ubd device - it must be COWed + * because, otherwise, the page would have been mapped in + * writeable + */ + + if(!dev->cow.file) + panic("Write fault on writeable non-COW ubd device %d", + i); + + /* It should also be an access to the backing file since the + * COW pages should be mapped in read-write + */ + + if(fd == dev->fd) + panic("Write fault on a backing page of ubd " + "device %d\n", i); + + /* So, we do the write, copying the backing data to the COW + * file... + */ + + err = os_seek_file(dev->fd, offset + dev->cow.data_offset); + if(err < 0) + panic("Couldn't seek to %lld in COW file of ubd " + "device %d, err = %d", + offset + dev->cow.data_offset, i, -err); + + n = os_write_file(dev->fd, (void *) address, PAGE_SIZE); + if(n != PAGE_SIZE) + panic("Couldn't copy data to COW file of ubd " + "device %d, err = %d", i, -n); + + /* ... updating the COW bitmap... */ + + cowify_bitmap(offset, PAGE_SIZE, NULL, &bitmap_offset, + dev->cow.bitmap, dev->cow.bitmap_offset, + new_bitmap, dev->cow.bitmap_len); + + err = os_seek_file(dev->fd, bitmap_offset); + if(err < 0) + panic("Couldn't seek to %lld in COW file of ubd " + "device %d, err = %d", bitmap_offset, i, -err); + + n = os_write_file(dev->fd, new_bitmap, sizeof(new_bitmap)); + if(n != sizeof(new_bitmap)) + panic("Couldn't update bitmap of ubd device %d, " + "err = %d", i, -n); + + /* Maybe we can map the COW page in, and maybe we can't. If + * it is a pre-V3 COW file, we can't, since the alignment will + * be wrong. If it is a V3 or later COW file which has been + * moved to a system with a larger page size, then maybe we + * can't, depending on the exact location of the page. + */ + + offset += dev->cow.data_offset; + + /* Remove the remapping, putting the original anonymous page + * back. If the COW file can be mapped in, that is done. + * Otherwise, the COW page is read in. + */ + + if(!physmem_remove_mapping((void *) address)) + panic("Address 0x%lx not remapped by ubd device %d", + address, i); + if((offset % UBD_MMAP_BLOCK_SIZE) == 0) + physmem_subst_mapping((void *) address, dev->fd, + offset, 1); + else { + err = os_seek_file(dev->fd, offset); + if(err < 0) + panic("Couldn't seek to %lld in COW file of " + "ubd device %d, err = %d", offset, i, + -err); + + n = os_read_file(dev->fd, (void *) address, PAGE_SIZE); + if(n != PAGE_SIZE) + panic("Failed to read page from offset %llx of " + "COW file of ubd device %d, err = %d", + offset, i, -n); + } + + return(1); + } + + /* It's not a write on a ubd device */ + return(0); +} + +static struct remapper ubd_remapper = { + .list = LIST_HEAD_INIT(ubd_remapper.list), + .proc = ubd_check_remapped, +}; + +static int ubd_remapper_setup(void) +{ + if(ubd_do_mmap) + register_remapper(&ubd_remapper); + + return(0); +} + +__initcall(ubd_remapper_setup); + +static int same_backing_files(char *from_cmdline, char *from_cow, char *cow) +{ + struct uml_stat buf1, buf2; + int err; + + if(from_cmdline == NULL) return(1); + if(!strcmp(from_cmdline, from_cow)) return(1); + + err = os_stat_file(from_cmdline, &buf1); + if(err < 0){ + printk("Couldn't stat '%s', err = %d\n", from_cmdline, -err); + return(1); + } + err = os_stat_file(from_cow, &buf2); + if(err < 0){ + printk("Couldn't stat '%s', err = %d\n", from_cow, -err); + return(1); + } + if((buf1.ust_dev == buf2.ust_dev) && (buf1.ust_ino == buf2.ust_ino)) + return(1); + + printk("Backing file mismatch - \"%s\" requested,\n" + "\"%s\" specified in COW header of \"%s\"\n", + from_cmdline, from_cow, cow); + return(0); +} + +static int backing_file_mismatch(char *file, __u64 size, time_t mtime) +{ + unsigned long modtime; + long long actual; + int err; + + err = os_file_modtime(file, &modtime); + if(err < 0){ + printk("Failed to get modification time of backing file " + "\"%s\", err = %d\n", file, -err); + return(err); + } + + err = os_file_size(file, &actual); + if(err < 0){ + printk("Failed to get size of backing file \"%s\", " + "err = %d\n", file, -err); + return(err); + } + + if(actual != size){ + /*__u64 can be a long on AMD64 and with %lu GCC complains; so + * the typecast.*/ + printk("Size mismatch (%llu vs %llu) of COW header vs backing " + "file\n", (unsigned long long) size, actual); + return(-EINVAL); + } + if(modtime != mtime){ + printk("mtime mismatch (%ld vs %ld) of COW header vs backing " + "file\n", mtime, modtime); + return(-EINVAL); + } + return(0); +} + +int read_cow_bitmap(int fd, void *buf, int offset, int len) +{ + int err; + + err = os_seek_file(fd, offset); + if(err < 0) + return(err); + + err = os_read_file(fd, buf, len); + if(err < 0) + return(err); + + return(0); +} + +int open_ubd_file(char *file, struct openflags *openflags, + char **backing_file_out, int *bitmap_offset_out, + unsigned long *bitmap_len_out, int *data_offset_out, + int *create_cow_out) +{ + time_t mtime; + unsigned long long size; + __u32 version, align; + char *backing_file; + int fd, err, sectorsize, same, mode = 0644; + + fd = os_open_file(file, *openflags, mode); + if(fd < 0){ + if((fd == -ENOENT) && (create_cow_out != NULL)) + *create_cow_out = 1; + if(!openflags->w || + ((fd != -EROFS) && (fd != -EACCES))) return(fd); + openflags->w = 0; + fd = os_open_file(file, *openflags, mode); + if(fd < 0) + return(fd); + } + + err = os_lock_file(fd, openflags->w); + if(err < 0){ + printk("Failed to lock '%s', err = %d\n", file, -err); + goto out_close; + } + + if(backing_file_out == NULL) return(fd); + + err = read_cow_header(file_reader, &fd, &version, &backing_file, &mtime, + &size, §orsize, &align, bitmap_offset_out); + if(err && (*backing_file_out != NULL)){ + printk("Failed to read COW header from COW file \"%s\", " + "errno = %d\n", file, -err); + goto out_close; + } + if(err) return(fd); + + if(backing_file_out == NULL) return(fd); + + same = same_backing_files(*backing_file_out, backing_file, file); + + if(!same && !backing_file_mismatch(*backing_file_out, size, mtime)){ + printk("Switching backing file to '%s'\n", *backing_file_out); + err = write_cow_header(file, fd, *backing_file_out, + sectorsize, align, &size); + if(err){ + printk("Switch failed, errno = %d\n", -err); + return(err); + } + } + else { + *backing_file_out = backing_file; + err = backing_file_mismatch(*backing_file_out, size, mtime); + if(err) goto out_close; + } + + cow_sizes(version, size, sectorsize, align, *bitmap_offset_out, + bitmap_len_out, data_offset_out); + + return(fd); + out_close: + os_close_file(fd); + return(err); +} + +int create_cow_file(char *cow_file, char *backing_file, struct openflags flags, + int sectorsize, int alignment, int *bitmap_offset_out, + unsigned long *bitmap_len_out, int *data_offset_out) +{ + int err, fd; + + flags.c = 1; + fd = open_ubd_file(cow_file, &flags, NULL, NULL, NULL, NULL, NULL); + if(fd < 0){ + err = fd; + printk("Open of COW file '%s' failed, errno = %d\n", cow_file, + -err); + goto out; + } + + err = init_cow_file(fd, cow_file, backing_file, sectorsize, alignment, + bitmap_offset_out, bitmap_len_out, + data_offset_out); + if(!err) + return(fd); + os_close_file(fd); + out: + return(err); +} + +static int update_bitmap(struct io_thread_req *req) +{ + int n; + + if(req->cow_offset == -1) + return(0); + + n = os_seek_file(req->fds[1], req->cow_offset); + if(n < 0){ + printk("do_io - bitmap lseek failed : err = %d\n", -n); + return(1); + } + + n = os_write_file(req->fds[1], &req->bitmap_words, + sizeof(req->bitmap_words)); + if(n != sizeof(req->bitmap_words)){ + printk("do_io - bitmap update failed, err = %d fd = %d\n", -n, + req->fds[1]); + return(1); + } + + return(0); +} + +void do_io(struct io_thread_req *req) +{ + char *buf; + unsigned long len; + int n, nsectors, start, end, bit; + int err; + __u64 off; + + if(req->op == UBD_MMAP){ + /* Touch the page to force the host to do any necessary IO to + * get it into memory + */ + n = *((volatile int *) req->buffer); + req->error = update_bitmap(req); + return; + } + + nsectors = req->length / req->sectorsize; + start = 0; + do { + bit = ubd_test_bit(start, (unsigned char *) &req->sector_mask); + end = start; + while((end < nsectors) && + (ubd_test_bit(end, (unsigned char *) + &req->sector_mask) == bit)) + end++; + + off = req->offset + req->offsets[bit] + + start * req->sectorsize; + len = (end - start) * req->sectorsize; + buf = &req->buffer[start * req->sectorsize]; + + err = os_seek_file(req->fds[bit], off); + if(err < 0){ + printk("do_io - lseek failed : err = %d\n", -err); + req->error = 1; + return; + } + if(req->op == UBD_READ){ + n = 0; + do { + buf = &buf[n]; + len -= n; + n = os_read_file(req->fds[bit], buf, len); + if (n < 0) { + printk("do_io - read failed, err = %d " + "fd = %d\n", -n, req->fds[bit]); + req->error = 1; + return; + } + } while((n < len) && (n != 0)); + if (n < len) memset(&buf[n], 0, len - n); + } + else { + n = os_write_file(req->fds[bit], buf, len); + if(n != len){ + printk("do_io - write failed err = %d " + "fd = %d\n", -n, req->fds[bit]); + req->error = 1; + return; + } + } + + start = end; + } while(start < nsectors); + + req->error = update_bitmap(req); +} + +/* Changed in start_io_thread, which is serialized by being called only + * from ubd_init, which is an initcall. + */ +int kernel_fd = -1; + +/* Only changed by the io thread */ +int io_count = 0; + +int io_thread(void *arg) +{ + struct io_thread_req req; + int n; + + ignore_sigwinch_sig(); + while(1){ + n = os_read_file(kernel_fd, &req, sizeof(req)); + if(n != sizeof(req)){ + if(n < 0) + printk("io_thread - read failed, fd = %d, " + "err = %d\n", kernel_fd, -n); + else { + printk("io_thread - short read, fd = %d, " + "length = %d\n", kernel_fd, n); + } + continue; + } + io_count++; + do_io(&req); + n = os_write_file(kernel_fd, &req, sizeof(req)); + if(n != sizeof(req)) + printk("io_thread - write failed, fd = %d, err = %d\n", + kernel_fd, -n); + } +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/ubd_user.c b/arch/um/drivers/ubd_user.c new file mode 100644 index 000000000000..b94d2bc4fe06 --- /dev/null +++ b/arch/um/drivers/ubd_user.c @@ -0,0 +1,75 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2001 Ridgerun,Inc (glonnon@ridgerun.com) + * Licensed under the GPL + */ + +#include <stddef.h> +#include <unistd.h> +#include <errno.h> +#include <sched.h> +#include <signal.h> +#include <string.h> +#include <netinet/in.h> +#include <sys/time.h> +#include <sys/socket.h> +#include <sys/mman.h> +#include <sys/param.h> +#include "asm/types.h" +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "ubd_user.h" +#include "os.h" +#include "cow.h" + +#include <endian.h> +#include <byteswap.h> + +void ignore_sigwinch_sig(void) +{ + signal(SIGWINCH, SIG_IGN); +} + +int start_io_thread(unsigned long sp, int *fd_out) +{ + int pid, fds[2], err; + + err = os_pipe(fds, 1, 1); + if(err < 0){ + printk("start_io_thread - os_pipe failed, err = %d\n", -err); + goto out; + } + + kernel_fd = fds[0]; + *fd_out = fds[1]; + + pid = clone(io_thread, (void *) sp, CLONE_FILES | CLONE_VM | SIGCHLD, + NULL); + if(pid < 0){ + printk("start_io_thread - clone failed : errno = %d\n", errno); + err = -errno; + goto out_close; + } + + return(pid); + + out_close: + os_close_file(fds[0]); + os_close_file(fds[1]); + kernel_fd = -1; + *fd_out = -1; + out: + return(err); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/xterm.c b/arch/um/drivers/xterm.c new file mode 100644 index 000000000000..93dc1911363f --- /dev/null +++ b/arch/um/drivers/xterm.c @@ -0,0 +1,225 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <string.h> +#include <errno.h> +#include <termios.h> +#include <signal.h> +#include <sched.h> +#include <sys/socket.h> +#include "kern_util.h" +#include "chan_user.h" +#include "helper.h" +#include "user_util.h" +#include "user.h" +#include "os.h" +#include "xterm.h" + +struct xterm_chan { + int pid; + int helper_pid; + char *title; + int device; + int raw; + struct termios tt; + unsigned long stack; + int direct_rcv; +}; + +/* Not static because it's called directly by the tt mode gdb code */ +void *xterm_init(char *str, int device, struct chan_opts *opts) +{ + struct xterm_chan *data; + + data = malloc(sizeof(*data)); + if(data == NULL) return(NULL); + *data = ((struct xterm_chan) { .pid = -1, + .helper_pid = -1, + .device = device, + .title = opts->xterm_title, + .raw = opts->raw, + .stack = opts->tramp_stack, + .direct_rcv = !opts->in_kernel } ); + return(data); +} + +/* Only changed by xterm_setup, which is a setup */ +static char *terminal_emulator = "xterm"; +static char *title_switch = "-T"; +static char *exec_switch = "-e"; + +static int __init xterm_setup(char *line, int *add) +{ + *add = 0; + terminal_emulator = line; + + line = strchr(line, ','); + if(line == NULL) return(0); + *line++ = '\0'; + if(*line) title_switch = line; + + line = strchr(line, ','); + if(line == NULL) return(0); + *line++ = '\0'; + if(*line) exec_switch = line; + + return(0); +} + +__uml_setup("xterm=", xterm_setup, +"xterm=<terminal emulator>,<title switch>,<exec switch>\n" +" Specifies an alternate terminal emulator to use for the debugger,\n" +" consoles, and serial lines when they are attached to the xterm channel.\n" +" The values are the terminal emulator binary, the switch it uses to set\n" +" its title, and the switch it uses to execute a subprocess,\n" +" respectively. The title switch must have the form '<switch> title',\n" +" not '<switch>=title'. Similarly, the exec switch must have the form\n" +" '<switch> command arg1 arg2 ...'.\n" +" The default values are 'xterm=xterm,-T,-e'. Values for gnome-terminal\n" +" are 'xterm=gnome-terminal,-t,-x'.\n\n" +); + +/* XXX This badly needs some cleaning up in the error paths + * Not static because it's called directly by the tt mode gdb code + */ +int xterm_open(int input, int output, int primary, void *d, + char **dev_out) +{ + struct xterm_chan *data = d; + unsigned long stack; + int pid, fd, new, err; + char title[256], file[] = "/tmp/xterm-pipeXXXXXX"; + char *argv[] = { terminal_emulator, title_switch, title, exec_switch, + "/usr/lib/uml/port-helper", "-uml-socket", + file, NULL }; + + if(os_access(argv[4], OS_ACC_X_OK) < 0) + argv[4] = "port-helper"; + + /* Check that DISPLAY is set, this doesn't guarantee the xterm + * will work but w/o it we can be pretty sure it won't. */ + if (!getenv("DISPLAY")) { + printk("xterm_open: $DISPLAY not set.\n"); + return -ENODEV; + } + + fd = mkstemp(file); + if(fd < 0){ + printk("xterm_open : mkstemp failed, errno = %d\n", errno); + return(-errno); + } + + if(unlink(file)){ + printk("xterm_open : unlink failed, errno = %d\n", errno); + return(-errno); + } + os_close_file(fd); + + fd = os_create_unix_socket(file, sizeof(file), 1); + if(fd < 0){ + printk("xterm_open : create_unix_socket failed, errno = %d\n", + -fd); + return(fd); + } + + sprintf(title, data->title, data->device); + stack = data->stack; + pid = run_helper(NULL, NULL, argv, &stack); + if(pid < 0){ + printk("xterm_open : run_helper failed, errno = %d\n", -pid); + return(pid); + } + + if(data->stack == 0) free_stack(stack, 0); + + if (data->direct_rcv) { + new = os_rcv_fd(fd, &data->helper_pid); + } else { + err = os_set_fd_block(fd, 0); + if(err < 0){ + printk("xterm_open : failed to set descriptor " + "non-blocking, err = %d\n", -err); + return(err); + } + new = xterm_fd(fd, &data->helper_pid); + } + if(new < 0){ + printk("xterm_open : os_rcv_fd failed, err = %d\n", -new); + goto out; + } + + CATCH_EINTR(err = tcgetattr(new, &data->tt)); + if(err){ + new = err; + goto out; + } + + if(data->raw){ + err = raw(new); + if(err){ + new = err; + goto out; + } + } + + data->pid = pid; + *dev_out = NULL; + out: + unlink(file); + return(new); +} + +/* Not static because it's called directly by the tt mode gdb code */ +void xterm_close(int fd, void *d) +{ + struct xterm_chan *data = d; + + if(data->pid != -1) + os_kill_process(data->pid, 1); + data->pid = -1; + if(data->helper_pid != -1) + os_kill_process(data->helper_pid, 0); + data->helper_pid = -1; + os_close_file(fd); +} + +static void xterm_free(void *d) +{ + free(d); +} + +static int xterm_console_write(int fd, const char *buf, int n, void *d) +{ + struct xterm_chan *data = d; + + return(generic_console_write(fd, buf, n, &data->tt)); +} + +struct chan_ops xterm_ops = { + .type = "xterm", + .init = xterm_init, + .open = xterm_open, + .close = xterm_close, + .read = generic_read, + .write = generic_write, + .console_write = xterm_console_write, + .window_size = generic_window_size, + .free = xterm_free, + .winch = 1, +}; + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/xterm.h b/arch/um/drivers/xterm.h new file mode 100644 index 000000000000..f33a6e77b186 --- /dev/null +++ b/arch/um/drivers/xterm.h @@ -0,0 +1,22 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __XTERM_H__ +#define __XTERM_H__ + +extern int xterm_fd(int socket, int *pid_out); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/drivers/xterm_kern.c b/arch/um/drivers/xterm_kern.c new file mode 100644 index 000000000000..7917b9d1cec8 --- /dev/null +++ b/arch/um/drivers/xterm_kern.c @@ -0,0 +1,93 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/errno.h" +#include "linux/slab.h" +#include "linux/signal.h" +#include "linux/interrupt.h" +#include "asm/semaphore.h" +#include "asm/irq.h" +#include "irq_user.h" +#include "irq_kern.h" +#include "kern_util.h" +#include "os.h" +#include "xterm.h" + +struct xterm_wait { + struct completion ready; + int fd; + int pid; + int new_fd; +}; + +static irqreturn_t xterm_interrupt(int irq, void *data, struct pt_regs *regs) +{ + struct xterm_wait *xterm = data; + int fd; + + fd = os_rcv_fd(xterm->fd, &xterm->pid); + if(fd == -EAGAIN) + return(IRQ_NONE); + + xterm->new_fd = fd; + complete(&xterm->ready); + return(IRQ_HANDLED); +} + +int xterm_fd(int socket, int *pid_out) +{ + struct xterm_wait *data; + int err, ret; + + data = kmalloc(sizeof(*data), GFP_KERNEL); + if(data == NULL){ + printk(KERN_ERR "xterm_fd : failed to allocate xterm_wait\n"); + return(-ENOMEM); + } + + /* This is a locked semaphore... */ + *data = ((struct xterm_wait) + { .fd = socket, + .pid = -1, + .new_fd = -1 }); + init_completion(&data->ready); + + err = um_request_irq(XTERM_IRQ, socket, IRQ_READ, xterm_interrupt, + SA_INTERRUPT | SA_SHIRQ | SA_SAMPLE_RANDOM, + "xterm", data); + if (err){ + printk(KERN_ERR "xterm_fd : failed to get IRQ for xterm, " + "err = %d\n", err); + ret = err; + goto out; + } + + /* ... so here we wait for an xterm interrupt. + * + * XXX Note, if the xterm doesn't work for some reason (eg. DISPLAY + * isn't set) this will hang... */ + wait_for_completion(&data->ready); + + free_irq_by_irq_and_dev(XTERM_IRQ, data); + free_irq(XTERM_IRQ, data); + + ret = data->new_fd; + *pid_out = data->pid; + out: + kfree(data); + + return(ret); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/2_5compat.h b/arch/um/include/2_5compat.h new file mode 100644 index 000000000000..abdb015a4d71 --- /dev/null +++ b/arch/um/include/2_5compat.h @@ -0,0 +1,24 @@ +/* + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __2_5_COMPAT_H__ +#define __2_5_COMPAT_H__ + +#define INIT_HARDSECT(arr, maj, sizes) + +#define SET_PRI(task) do ; while(0) + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/chan_kern.h b/arch/um/include/chan_kern.h new file mode 100644 index 000000000000..da9a6717e7a4 --- /dev/null +++ b/arch/um/include/chan_kern.h @@ -0,0 +1,60 @@ +/* + * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __CHAN_KERN_H__ +#define __CHAN_KERN_H__ + +#include "linux/tty.h" +#include "linux/list.h" +#include "linux/console.h" +#include "chan_user.h" +#include "line.h" + +struct chan { + struct list_head list; + char *dev; + unsigned int primary:1; + unsigned int input:1; + unsigned int output:1; + unsigned int opened:1; + int fd; + enum chan_init_pri pri; + struct chan_ops *ops; + void *data; +}; + +extern void chan_interrupt(struct list_head *chans, struct work_struct *task, + struct tty_struct *tty, int irq); +extern int parse_chan_pair(char *str, struct list_head *chans, int pri, + int device, struct chan_opts *opts); +extern int open_chan(struct list_head *chans); +extern int write_chan(struct list_head *chans, const char *buf, int len, + int write_irq); +extern int console_write_chan(struct list_head *chans, const char *buf, + int len); +extern int console_open_chan(struct line *line, struct console *co, + struct chan_opts *opts); +extern void close_chan(struct list_head *chans); +extern void chan_enable_winch(struct list_head *chans, struct tty_struct *tty); +extern void enable_chan(struct list_head *chans, struct tty_struct *tty); +extern int chan_window_size(struct list_head *chans, + unsigned short *rows_out, + unsigned short *cols_out); +extern int chan_out_fd(struct list_head *chans); +extern int chan_config_string(struct list_head *chans, char *str, int size, + char **error_out); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/chan_user.h b/arch/um/include/chan_user.h new file mode 100644 index 000000000000..f77d9aa4c164 --- /dev/null +++ b/arch/um/include/chan_user.h @@ -0,0 +1,67 @@ +/* + * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __CHAN_USER_H__ +#define __CHAN_USER_H__ + +#include "init.h" + +struct chan_opts { + void (*announce)(char *dev_name, int dev); + char *xterm_title; + int raw; + unsigned long tramp_stack; + int in_kernel; +}; + +enum chan_init_pri { INIT_STATIC, INIT_ALL, INIT_ONE }; + +struct chan_ops { + char *type; + void *(*init)(char *, int, struct chan_opts *); + int (*open)(int, int, int, void *, char **); + void (*close)(int, void *); + int (*read)(int, char *, void *); + int (*write)(int, const char *, int, void *); + int (*console_write)(int, const char *, int, void *); + int (*window_size)(int, void *, unsigned short *, unsigned short *); + void (*free)(void *); + int winch; +}; + +extern struct chan_ops fd_ops, null_ops, port_ops, pts_ops, pty_ops, tty_ops, + xterm_ops; + +extern void generic_close(int fd, void *unused); +extern int generic_read(int fd, char *c_out, void *unused); +extern int generic_write(int fd, const char *buf, int n, void *unused); +extern int generic_console_write(int fd, const char *buf, int n, void *state); +extern int generic_window_size(int fd, void *unused, unsigned short *rows_out, + unsigned short *cols_out); +extern void generic_free(void *data); + +struct tty_struct; +extern void register_winch(int fd, struct tty_struct *tty); +extern void register_winch_irq(int fd, int tty_fd, int pid, struct tty_struct *tty); + +#define __channel_help(fn, prefix) \ +__uml_help(fn, prefix "[0-9]*=<channel description>\n" \ +" Attach a console or serial line to a host channel. See\n" \ +" http://user-mode-linux.sourceforge.net/input.html for a complete\n" \ +" description of this switch.\n\n" \ +); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/choose-mode.h b/arch/um/include/choose-mode.h new file mode 100644 index 000000000000..8e6b62f5e9ac --- /dev/null +++ b/arch/um/include/choose-mode.h @@ -0,0 +1,42 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __CHOOSE_MODE_H__ +#define __CHOOSE_MODE_H__ + +#include "uml-config.h" + +#if defined(UML_CONFIG_MODE_TT) && defined(UML_CONFIG_MODE_SKAS) +#define CHOOSE_MODE(tt, skas) (mode_tt ? (tt) : (skas)) + +#elif defined(UML_CONFIG_MODE_SKAS) +#define CHOOSE_MODE(tt, skas) (skas) + +#elif defined(UML_CONFIG_MODE_TT) +#define CHOOSE_MODE(tt, skas) (tt) +#endif + +#define CHOOSE_MODE_PROC(tt, skas, args...) \ + CHOOSE_MODE(tt(args), skas(args)) + +extern int mode_tt; +static inline void *__choose_mode(void *tt, void *skas) { + return mode_tt ? tt : skas; +} + +#define __CHOOSE_MODE(tt, skas) (*( (typeof(tt) *) __choose_mode(&(tt), &(skas)))) + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/elf_user.h b/arch/um/include/elf_user.h new file mode 100644 index 000000000000..53516b637272 --- /dev/null +++ b/arch/um/include/elf_user.h @@ -0,0 +1,19 @@ +/* + * Copyright (C) 2004 Fujitsu Siemens Computers GmbH + * Author: Bodo Stroesser <bstroesser@fujitsu-siemens.com> + * Licensed under the GPL + */ + +#ifndef __ELF_USER_H__ +#define __ELF_USER_H__ + +/* For compilation on a host that doesn't support AT_SYSINFO (Linux 2.4) */ + +#ifndef AT_SYSINFO +#define AT_SYSINFO 32 +#endif +#ifndef AT_SYSINFO_EHDR +#define AT_SYSINFO_EHDR 33 +#endif + +#endif diff --git a/arch/um/include/frame_kern.h b/arch/um/include/frame_kern.h new file mode 100644 index 000000000000..ce9514f57211 --- /dev/null +++ b/arch/um/include/frame_kern.h @@ -0,0 +1,32 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __FRAME_KERN_H_ +#define __FRAME_KERN_H_ + +#define _S(nr) (1<<((nr)-1)) +#define _BLOCKABLE (~(_S(SIGKILL) | _S(SIGSTOP))) + +extern int setup_signal_stack_sc(unsigned long stack_top, int sig, + struct k_sigaction *ka, + struct pt_regs *regs, + sigset_t *mask); +extern int setup_signal_stack_si(unsigned long stack_top, int sig, + struct k_sigaction *ka, + struct pt_regs *regs, siginfo_t *info, + sigset_t *mask); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/helper.h b/arch/um/include/helper.h new file mode 100644 index 000000000000..162ac31192fd --- /dev/null +++ b/arch/um/include/helper.h @@ -0,0 +1,27 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __HELPER_H__ +#define __HELPER_H__ + +extern int run_helper(void (*pre_exec)(void *), void *pre_data, char **argv, + unsigned long *stack_out); +extern int run_helper_thread(int (*proc)(void *), void *arg, + unsigned int flags, unsigned long *stack_out, + int stack_order); +extern int helper_wait(int pid); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/init.h b/arch/um/include/init.h new file mode 100644 index 000000000000..55c2693f8778 --- /dev/null +++ b/arch/um/include/init.h @@ -0,0 +1,132 @@ +#ifndef _LINUX_UML_INIT_H +#define _LINUX_UML_INIT_H + +/* These macros are used to mark some functions or + * initialized data (doesn't apply to uninitialized data) + * as `initialization' functions. The kernel can take this + * as hint that the function is used only during the initialization + * phase and free up used memory resources after + * + * Usage: + * For functions: + * + * You should add __init immediately before the function name, like: + * + * static void __init initme(int x, int y) + * { + * extern int z; z = x * y; + * } + * + * If the function has a prototype somewhere, you can also add + * __init between closing brace of the prototype and semicolon: + * + * extern int initialize_foobar_device(int, int, int) __init; + * + * For initialized data: + * You should insert __initdata between the variable name and equal + * sign followed by value, e.g.: + * + * static int init_variable __initdata = 0; + * static char linux_logo[] __initdata = { 0x32, 0x36, ... }; + * + * Don't forget to initialize data not at file scope, i.e. within a function, + * as gcc otherwise puts the data into the bss section and not into the init + * section. + * + * Also note, that this data cannot be "const". + */ + +#ifndef _LINUX_INIT_H +typedef int (*initcall_t)(void); +typedef void (*exitcall_t)(void); + +/* These are for everybody (although not all archs will actually + discard it in modules) */ +#define __init __attribute__ ((__section__ (".init.text"))) +#define __initdata __attribute__ ((__section__ (".init.data"))) +#define __exitdata __attribute__ ((__section__(".exit.data"))) +#define __exit_call __attribute_used__ __attribute__ ((__section__ (".exitcall.exit"))) + +#ifdef MODULE +#define __exit __attribute__ ((__section__(".exit.text"))) +#else +#define __exit __attribute_used__ __attribute__ ((__section__(".exit.text"))) +#endif + +#endif + +#ifndef MODULE +struct uml_param { + const char *str; + int (*setup_func)(char *, int *); +}; + +extern initcall_t __uml_initcall_start, __uml_initcall_end; +extern initcall_t __uml_postsetup_start, __uml_postsetup_end; +extern const char *__uml_help_start, *__uml_help_end; +#endif + +#define __uml_initcall(fn) \ + static initcall_t __uml_initcall_##fn __uml_init_call = fn + +#define __uml_exitcall(fn) \ + static exitcall_t __uml_exitcall_##fn __uml_exit_call = fn + +extern struct uml_param __uml_setup_start, __uml_setup_end; + +#define __uml_postsetup(fn) \ + static initcall_t __uml_postsetup_##fn __uml_postsetup_call = fn + +#define __non_empty_string(dummyname,string) \ + struct __uml_non_empty_string_struct_##dummyname \ + { \ + char _string[sizeof(string)-2]; \ + } + +#ifndef MODULE +#define __uml_setup(str, fn, help...) \ + __non_empty_string(fn ##_setup, str); \ + __uml_help(fn, help); \ + static char __uml_setup_str_##fn[] __initdata = str; \ + static struct uml_param __uml_setup_##fn __uml_init_setup = { __uml_setup_str_##fn, fn } +#else +#define __uml_setup(str, fn, help...) \ + +#endif + +#define __uml_help(fn, help...) \ + __non_empty_string(fn ##__help, help); \ + static char __uml_help_str_##fn[] __initdata = help; \ + static const char *__uml_help_##fn __uml_setup_help = __uml_help_str_##fn + +/* + * Mark functions and data as being only used at initialization + * or exit time. + */ +#define __uml_init_setup __attribute_used__ __attribute__ ((__section__ (".uml.setup.init"))) +#define __uml_setup_help __attribute_used__ __attribute__ ((__section__ (".uml.help.init"))) +#define __uml_init_call __attribute_used__ __attribute__ ((__section__ (".uml.initcall.init"))) +#define __uml_postsetup_call __attribute_used__ __attribute__ ((__section__ (".uml.postsetup.init"))) +#define __uml_exit_call __attribute_used__ __attribute__ ((__section__ (".uml.exitcall.exit"))) + +#ifndef __KERNEL__ + +#define __initcall(fn) static initcall_t __initcall_##fn __init_call = fn +#define __exitcall(fn) static exitcall_t __exitcall_##fn __exit_call = fn + +#define __init_call __attribute__ ((unused,__section__ (".initcall.init"))) + +#endif + +#endif /* _LINUX_UML_INIT_H */ + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/initrd.h b/arch/um/include/initrd.h new file mode 100644 index 000000000000..439b9a814985 --- /dev/null +++ b/arch/um/include/initrd.h @@ -0,0 +1,22 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __INITRD_USER_H__ +#define __INITRD_USER_H__ + +extern int load_initrd(char *filename, void *buf, int size); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/irq_kern.h b/arch/um/include/irq_kern.h new file mode 100644 index 000000000000..3af52a634c4c --- /dev/null +++ b/arch/um/include/irq_kern.h @@ -0,0 +1,28 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __IRQ_KERN_H__ +#define __IRQ_KERN_H__ + +#include "linux/interrupt.h" + +extern int um_request_irq(unsigned int irq, int fd, int type, + irqreturn_t (*handler)(int, void *, + struct pt_regs *), + unsigned long irqflags, const char * devname, + void *dev_id); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/irq_user.h b/arch/um/include/irq_user.h new file mode 100644 index 000000000000..f724b717213f --- /dev/null +++ b/arch/um/include/irq_user.h @@ -0,0 +1,36 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __IRQ_USER_H__ +#define __IRQ_USER_H__ + +enum { IRQ_READ, IRQ_WRITE }; + +extern void sigio_handler(int sig, union uml_pt_regs *regs); +extern int activate_fd(int irq, int fd, int type, void *dev_id); +extern void free_irq_by_irq_and_dev(unsigned int irq, void *dev_id); +extern void free_irq_by_fd(int fd); +extern void reactivate_fd(int fd, int irqnum); +extern void deactivate_fd(int fd, int irqnum); +extern int deactivate_all_fds(void); +extern void forward_interrupts(int pid); +extern void init_irq_signals(int on_sigstack); +extern void forward_ipi(int fd, int pid); +extern void free_irq_later(int irq, void *dev_id); +extern int activate_ipi(int fd, int pid); +extern unsigned long irq_lock(void); +extern void irq_unlock(unsigned long flags); +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/kern.h b/arch/um/include/kern.h new file mode 100644 index 000000000000..1e3170768b5c --- /dev/null +++ b/arch/um/include/kern.h @@ -0,0 +1,49 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __KERN_H__ +#define __KERN_H__ + +/* These are all user-mode things which are convenient to call directly + * from kernel code and for which writing a wrapper is too much of a pain. + * The regular include files can't be included because this file is included + * only into kernel code, and user-space includes conflict with kernel + * includes. + */ + +extern int errno; + +extern int clone(int (*proc)(void *), void *sp, int flags, void *data); +extern int sleep(int); +extern int printf(char *fmt, ...); +extern char *strerror(int errnum); +extern char *ptsname(int __fd); +extern int munmap(void *, int); +extern void *sbrk(int increment); +extern void *malloc(int size); +extern void perror(char *err); +extern int kill(int pid, int sig); +extern int getuid(void); +extern int getgid(void); +extern int pause(void); +extern int write(int, const void *, int); +extern int exit(int); +extern int close(int); +extern int read(unsigned int, char *, int); +extern int pipe(int *); +extern int sched_yield(void); +extern int ptrace(int op, int pid, long addr, long data); +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/kern_util.h b/arch/um/include/kern_util.h new file mode 100644 index 000000000000..15389c886b41 --- /dev/null +++ b/arch/um/include/kern_util.h @@ -0,0 +1,125 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __KERN_UTIL_H__ +#define __KERN_UTIL_H__ + +#include "linux/threads.h" +#include "sysdep/ptrace.h" + +extern int ncpus; +extern char *linux_prog; +extern char *gdb_init; +extern int kmalloc_ok; +extern int timer_irq_inited; +extern int jail; +extern int nsyscalls; + +extern struct task_struct *idle_threads[NR_CPUS]; + +#define UML_ROUND_DOWN(addr) ((void *)(((unsigned long) addr) & PAGE_MASK)) +#define UML_ROUND_UP(addr) \ + UML_ROUND_DOWN(((unsigned long) addr) + PAGE_SIZE - 1) + +extern int kernel_fork(unsigned long flags, int (*fn)(void *), void * arg); +extern unsigned long stack_sp(unsigned long page); +extern int kernel_thread_proc(void *data); +extern void syscall_segv(int sig); +extern int current_pid(void); +extern unsigned long alloc_stack(int order, int atomic); +extern int do_signal(void); +extern int is_stack_fault(unsigned long sp); +extern unsigned long segv(unsigned long address, unsigned long ip, + int is_write, int is_user, void *sc); +extern int handle_page_fault(unsigned long address, unsigned long ip, + int is_write, int is_user, int *code_out); +extern void syscall_ready(void); +extern void set_tracing(void *t, int tracing); +extern int is_tracing(void *task); +extern int segv_syscall(void); +extern void kern_finish_exec(void *task, int new_pid, unsigned long stack); +extern int page_size(void); +extern unsigned long page_mask(void); +extern int need_finish_fork(void); +extern void free_stack(unsigned long stack, int order); +extern void add_input_request(int op, void (*proc)(int), void *arg); +extern char *current_cmd(void); +extern void timer_handler(int sig, union uml_pt_regs *regs); +extern int set_signals(int enable); +extern void force_sigbus(void); +extern int pid_to_processor_id(int pid); +extern void block_signals(void); +extern void unblock_signals(void); +extern void deliver_signals(void *t); +extern int next_syscall_index(int max); +extern int next_trap_index(int max); +extern void default_idle(void); +extern void finish_fork(void); +extern void paging_init(void); +extern void init_flush_vm(void); +extern void *syscall_sp(void *t); +extern void syscall_trace(union uml_pt_regs *regs, int entryexit); +extern int hz(void); +extern void uml_idle_timer(void); +extern unsigned int do_IRQ(int irq, union uml_pt_regs *regs); +extern int external_pid(void *t); +extern void boot_timer_handler(int sig); +extern void interrupt_end(void); +extern void initial_thread_cb(void (*proc)(void *), void *arg); +extern int debugger_signal(int status, int pid); +extern void debugger_parent_signal(int status, int pid); +extern void child_signal(int pid, int status); +extern int init_ptrace_proxy(int idle_pid, int startup, int stop); +extern int init_parent_proxy(int pid); +extern int singlestepping(void *t); +extern void check_stack_overflow(void *ptr); +extern void relay_signal(int sig, union uml_pt_regs *regs); +extern void not_implemented(void); +extern int user_context(unsigned long sp); +extern void timer_irq(union uml_pt_regs *regs); +extern void unprotect_stack(unsigned long stack); +extern void do_uml_exitcalls(void); +extern int attach_debugger(int idle_pid, int pid, int stop); +extern void bad_segv(unsigned long address, unsigned long ip, int is_write); +extern int config_gdb(char *str); +extern int remove_gdb(void); +extern char *uml_strdup(char *string); +extern void unprotect_kernel_mem(void); +extern void protect_kernel_mem(void); +extern void uml_cleanup(void); +extern void set_current(void *t); +extern void lock_signalled_task(void *t); +extern void IPI_handler(int cpu); +extern int jail_setup(char *line, int *add); +extern void *get_init_task(void); +extern int clear_user_proc(void *buf, int size); +extern int copy_to_user_proc(void *to, void *from, int size); +extern int copy_from_user_proc(void *to, void *from, int size); +extern int strlen_user_proc(char *str); +extern void bus_handler(int sig, union uml_pt_regs *regs); +extern void winch(int sig, union uml_pt_regs *regs); +extern long execute_syscall(void *r); +extern int smp_sigio_handler(void); +extern void *get_current(void); +extern struct task_struct *get_task(int pid, int require); +extern void machine_halt(void); +extern int is_syscall(unsigned long addr); +extern void arch_switch(void); +extern void free_irq(unsigned int, void *); +extern int um_in_interrupt(void); +extern int cpu(void); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/line.h b/arch/um/include/line.h new file mode 100644 index 000000000000..6d81ecc17be5 --- /dev/null +++ b/arch/um/include/line.h @@ -0,0 +1,108 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __LINE_H__ +#define __LINE_H__ + +#include "linux/list.h" +#include "linux/workqueue.h" +#include "linux/tty.h" +#include "linux/interrupt.h" +#include "asm/semaphore.h" +#include "chan_user.h" +#include "mconsole_kern.h" + +struct line_driver { + char *name; + char *device_name; + char *devfs_name; + short major; + short minor_start; + short type; + short subtype; + int read_irq; + char *read_irq_name; + int write_irq; + char *write_irq_name; + char *symlink_from; + char *symlink_to; + struct mc_device mc; +}; + +struct line { + char *init_str; + int init_pri; + struct list_head chan_list; + int valid; + int count; + struct semaphore sem; + char *buffer; + char *head; + char *tail; + int sigio; + struct work_struct task; + struct line_driver *driver; + int have_irq; +}; + +#define LINE_INIT(str, d) \ + { init_str : str, \ + init_pri : INIT_STATIC, \ + chan_list : { }, \ + valid : 1, \ + sem : { }, \ + buffer : NULL, \ + head : NULL, \ + tail : NULL, \ + sigio : 0, \ + driver : d, \ + have_irq : 0 } + +struct lines { + int num; +}; + +#define LINES_INIT(n) { num : n } + +extern void line_close(struct tty_struct *tty, struct file * filp); +extern int line_open(struct line *lines, struct tty_struct *tty, + struct chan_opts *opts); +extern int line_setup(struct line *lines, int num, char *init, + int all_allowed); +extern int line_write(struct tty_struct *tty, const unsigned char *buf, int len); +extern void line_put_char(struct tty_struct *tty, unsigned char ch); +extern void line_set_termios(struct tty_struct *tty, struct termios * old); +extern int line_chars_in_buffer(struct tty_struct *tty); +extern int line_write_room(struct tty_struct *tty); +extern int line_ioctl(struct tty_struct *tty, struct file * file, + unsigned int cmd, unsigned long arg); +extern char *add_xterm_umid(char *base); +extern int line_setup_irq(int fd, int input, int output, struct tty_struct *tty); +extern void line_close_chan(struct line *line); +extern void line_disable(struct tty_struct *tty, int current_irq); +extern struct tty_driver * line_register_devfs(struct lines *set, + struct line_driver *line_driver, + struct tty_operations *driver, + struct line *lines, + int nlines); +extern void lines_init(struct line *lines, int nlines); +extern void close_lines(struct line *lines, int nlines); +extern int line_config(struct line *lines, int num, char *str); +extern int line_remove(struct line *lines, int num, char *str); +extern int line_get_config(char *dev, struct line *lines, int num, char *str, + int size, char **error_out); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/mconsole.h b/arch/um/include/mconsole.h new file mode 100644 index 000000000000..9fbe3083fdd8 --- /dev/null +++ b/arch/um/include/mconsole.h @@ -0,0 +1,103 @@ +/* + * Copyright (C) 2001 Lennert Buytenhek (buytenh@gnu.org) + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __MCONSOLE_H__ +#define __MCONSOLE_H__ + +#ifndef __KERNEL__ +#include <stdint.h> +#define u32 uint32_t +#endif + +#define MCONSOLE_MAGIC (0xcafebabe) +#define MCONSOLE_MAX_DATA (512) +#define MCONSOLE_VERSION 2 + +struct mconsole_request { + u32 magic; + u32 version; + u32 len; + char data[MCONSOLE_MAX_DATA]; +}; + +struct mconsole_reply { + u32 err; + u32 more; + u32 len; + char data[MCONSOLE_MAX_DATA]; +}; + +struct mconsole_notify { + u32 magic; + u32 version; + enum { MCONSOLE_SOCKET, MCONSOLE_PANIC, MCONSOLE_HANG, + MCONSOLE_USER_NOTIFY } type; + u32 len; + char data[MCONSOLE_MAX_DATA]; +}; + +struct mc_request; + +enum mc_context { MCONSOLE_INTR, MCONSOLE_PROC }; + +struct mconsole_command +{ + char *command; + void (*handler)(struct mc_request *req); + enum mc_context context; +}; + +struct mc_request +{ + int len; + int as_interrupt; + + int originating_fd; + int originlen; + unsigned char origin[128]; /* sockaddr_un */ + + struct mconsole_request request; + struct mconsole_command *cmd; +}; + +extern char mconsole_socket_name[]; + +extern int mconsole_unlink_socket(void); +extern int mconsole_reply(struct mc_request *req, char *reply, int err, + int more); + +extern void mconsole_version(struct mc_request *req); +extern void mconsole_help(struct mc_request *req); +extern void mconsole_halt(struct mc_request *req); +extern void mconsole_reboot(struct mc_request *req); +extern void mconsole_config(struct mc_request *req); +extern void mconsole_remove(struct mc_request *req); +extern void mconsole_sysrq(struct mc_request *req); +extern void mconsole_cad(struct mc_request *req); +extern void mconsole_stop(struct mc_request *req); +extern void mconsole_go(struct mc_request *req); +extern void mconsole_log(struct mc_request *req); +extern void mconsole_proc(struct mc_request *req); + +extern int mconsole_get_request(int fd, struct mc_request *req); +extern int mconsole_notify(char *sock_name, int type, const void *data, + int len); +extern char *mconsole_notify_socket(void); +extern void lock_notify(void); +extern void unlock_notify(void); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/mconsole_kern.h b/arch/um/include/mconsole_kern.h new file mode 100644 index 000000000000..61c274fcee5d --- /dev/null +++ b/arch/um/include/mconsole_kern.h @@ -0,0 +1,62 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __MCONSOLE_KERN_H__ +#define __MCONSOLE_KERN_H__ + +#include "linux/config.h" +#include "linux/list.h" +#include "mconsole.h" + +struct mconsole_entry { + struct list_head list; + struct mc_request request; +}; + +struct mc_device { + struct list_head list; + char *name; + int (*config)(char *); + int (*get_config)(char *, char *, int, char **); + int (*remove)(char *); +}; + +#define CONFIG_CHUNK(str, size, current, chunk, end) \ +do { \ + current += strlen(chunk); \ + if(current >= size) \ + str = NULL; \ + if(str != NULL){ \ + strcpy(str, chunk); \ + str += strlen(chunk); \ + } \ + if(end) \ + current++; \ +} while(0) + +#ifdef CONFIG_MCONSOLE + +extern void mconsole_register_dev(struct mc_device *new); + +#else + +static inline void mconsole_register_dev(struct mc_device *new) +{ +} + +#endif + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/mem.h b/arch/um/include/mem.h new file mode 100644 index 000000000000..10c46c38949a --- /dev/null +++ b/arch/um/include/mem.h @@ -0,0 +1,28 @@ +/* + * Copyright (C) 2002, 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#ifndef __MEM_H__ +#define __MEM_H__ + +#include "linux/types.h" + +extern int phys_mapping(unsigned long phys, __u64 *offset_out); +extern int physmem_subst_mapping(void *virt, int fd, __u64 offset, int w); +extern int is_remapped(void *virt); +extern int physmem_remove_mapping(void *virt); +extern void physmem_forget_descriptor(int fd); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/mem_kern.h b/arch/um/include/mem_kern.h new file mode 100644 index 000000000000..cb7e196d366b --- /dev/null +++ b/arch/um/include/mem_kern.h @@ -0,0 +1,30 @@ +/* + * Copyright (C) 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#ifndef __MEM_KERN_H__ +#define __MEM_KERN_H__ + +#include "linux/list.h" +#include "linux/types.h" + +struct remapper { + struct list_head list; + int (*proc)(int, unsigned long, int, __u64); +}; + +extern void register_remapper(struct remapper *info); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/mem_user.h b/arch/um/include/mem_user.h new file mode 100644 index 000000000000..d6404bb64662 --- /dev/null +++ b/arch/um/include/mem_user.h @@ -0,0 +1,83 @@ +/* + * arch/um/include/mem_user.h + * + * BRIEF MODULE DESCRIPTION + * user side memory interface for support IO memory inside user mode linux + * + * Copyright (C) 2001 RidgeRun, Inc. + * Author: RidgeRun, Inc. + * Greg Lonnon glonnon@ridgerun.com or info@ridgerun.com + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN + * NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, + * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT + * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF + * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON + * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF + * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write to the Free Software Foundation, Inc., + * 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +#ifndef _MEM_USER_H +#define _MEM_USER_H + +struct iomem_region { + struct iomem_region *next; + char *driver; + int fd; + int size; + unsigned long phys; + unsigned long virt; +}; + +extern struct iomem_region *iomem_regions; +extern int iomem_size; + +#define ROUND_4M(n) ((((unsigned long) (n)) + (1 << 22)) & ~((1 << 22) - 1)) + +extern unsigned long host_task_size; +extern unsigned long task_size; + +extern void check_devanon(void); +extern int init_mem_user(void); +extern int create_mem_file(unsigned long len); +extern void setup_memory(void *entry); +extern unsigned long find_iomem(char *driver, unsigned long *len_out); +extern int init_maps(unsigned long physmem, unsigned long iomem, + unsigned long highmem); +extern unsigned long get_vm(unsigned long len); +extern void setup_physmem(unsigned long start, unsigned long usable, + unsigned long len, unsigned long highmem); +extern void add_iomem(char *name, int fd, unsigned long size); +extern unsigned long phys_offset(unsigned long phys); +extern void unmap_physmem(void); +extern void map_memory(unsigned long virt, unsigned long phys, + unsigned long len, int r, int w, int x); +extern int protect_memory(unsigned long addr, unsigned long len, + int r, int w, int x, int must_succeed); +extern unsigned long get_kmem_end(void); +extern void check_tmpexec(void); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/mode.h b/arch/um/include/mode.h new file mode 100644 index 000000000000..786cf563eb05 --- /dev/null +++ b/arch/um/include/mode.h @@ -0,0 +1,30 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __MODE_H__ +#define __MODE_H__ + +#include "uml-config.h" + +#ifdef UML_CONFIG_MODE_TT +#include "mode-tt.h" +#endif + +#ifdef UML_CONFIG_MODE_SKAS +#include "mode-skas.h" +#endif + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/mode_kern.h b/arch/um/include/mode_kern.h new file mode 100644 index 000000000000..2d88afd0cf16 --- /dev/null +++ b/arch/um/include/mode_kern.h @@ -0,0 +1,30 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __MODE_KERN_H__ +#define __MODE_KERN_H__ + +#include "linux/config.h" + +#ifdef CONFIG_MODE_TT +#include "mode_kern-tt.h" +#endif + +#ifdef CONFIG_MODE_SKAS +#include "mode_kern-skas.h" +#endif + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/net_kern.h b/arch/um/include/net_kern.h new file mode 100644 index 000000000000..1c07949a13d6 --- /dev/null +++ b/arch/um/include/net_kern.h @@ -0,0 +1,82 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __UM_NET_KERN_H +#define __UM_NET_KERN_H + +#include "linux/netdevice.h" +#include "linux/skbuff.h" +#include "linux/socket.h" +#include "linux/list.h" + +struct uml_net { + struct list_head list; + struct net_device *dev; + struct platform_device pdev; + int index; + unsigned char mac[ETH_ALEN]; + int have_mac; +}; + +struct uml_net_private { + struct list_head list; + spinlock_t lock; + struct net_device *dev; + struct timer_list tl; + struct net_device_stats stats; + int fd; + unsigned char mac[ETH_ALEN]; + int have_mac; + unsigned short (*protocol)(struct sk_buff *); + int (*open)(void *); + void (*close)(int, void *); + void (*remove)(void *); + int (*read)(int, struct sk_buff **skb, struct uml_net_private *); + int (*write)(int, struct sk_buff **skb, struct uml_net_private *); + + void (*add_address)(unsigned char *, unsigned char *, void *); + void (*delete_address)(unsigned char *, unsigned char *, void *); + int (*set_mtu)(int mtu, void *); + int user[1]; +}; + +struct net_kern_info { + void (*init)(struct net_device *, void *); + unsigned short (*protocol)(struct sk_buff *); + int (*read)(int, struct sk_buff **skb, struct uml_net_private *); + int (*write)(int, struct sk_buff **skb, struct uml_net_private *); +}; + +struct transport { + struct list_head list; + char *name; + int (*setup)(char *, char **, void *); + struct net_user_info *user; + struct net_kern_info *kern; + int private_size; + int setup_size; +}; + +extern struct net_device *ether_init(int); +extern unsigned short ether_protocol(struct sk_buff *); +extern int setup_etheraddr(char *str, unsigned char *addr); +extern struct sk_buff *ether_adjust_skb(struct sk_buff *skb, int extra); +extern int tap_setup_common(char *str, char *type, char **dev_name, + char **mac_out, char **gate_addr); +extern void register_transport(struct transport *new); +extern unsigned short eth_protocol(struct sk_buff *skb); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/net_user.h b/arch/um/include/net_user.h new file mode 100644 index 000000000000..36807b796e9f --- /dev/null +++ b/arch/um/include/net_user.h @@ -0,0 +1,66 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __UM_NET_USER_H__ +#define __UM_NET_USER_H__ + +#define ETH_ADDR_LEN (6) +#define ETH_HEADER_ETHERTAP (16) +#define ETH_HEADER_OTHER (14) +#define ETH_MAX_PACKET (1500) + +#define UML_NET_VERSION (4) + +struct net_user_info { + void (*init)(void *, void *); + int (*open)(void *); + void (*close)(int, void *); + void (*remove)(void *); + int (*set_mtu)(int mtu, void *); + void (*add_address)(unsigned char *, unsigned char *, void *); + void (*delete_address)(unsigned char *, unsigned char *, void *); + int max_packet; +}; + +extern void ether_user_init(void *data, void *dev); +extern void dev_ip_addr(void *d, char *buf, char *bin_buf); +extern void set_ether_mac(void *d, unsigned char *addr); +extern void iter_addresses(void *d, void (*cb)(unsigned char *, + unsigned char *, void *), + void *arg); + +extern void *get_output_buffer(int *len_out); +extern void free_output_buffer(void *buffer); + +extern int tap_open_common(void *dev, char *gate_addr); +extern void tap_check_ips(char *gate_addr, char *eth_addr); + +extern void read_output(int fd, char *output_out, int len); + +extern int net_read(int fd, void *buf, int len); +extern int net_recvfrom(int fd, void *buf, int len); +extern int net_write(int fd, void *buf, int len); +extern int net_send(int fd, void *buf, int len); +extern int net_sendto(int fd, void *buf, int len, void *to, int sock_len); + +extern void open_addr(unsigned char *addr, unsigned char *netmask, void *arg); +extern void close_addr(unsigned char *addr, unsigned char *netmask, void *arg); + +extern char *split_if_spec(char *str, ...); + +extern int dev_netmask(void *d, void *m); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/os.h b/arch/um/include/os.h new file mode 100644 index 000000000000..07340c8cf203 --- /dev/null +++ b/arch/um/include/os.h @@ -0,0 +1,183 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __OS_H__ +#define __OS_H__ + +#include "asm/types.h" +#include "../os/include/file.h" + +#define OS_TYPE_FILE 1 +#define OS_TYPE_DIR 2 +#define OS_TYPE_SYMLINK 3 +#define OS_TYPE_CHARDEV 4 +#define OS_TYPE_BLOCKDEV 5 +#define OS_TYPE_FIFO 6 +#define OS_TYPE_SOCK 7 + +/* os_access() flags */ +#define OS_ACC_F_OK 0 /* Test for existence. */ +#define OS_ACC_X_OK 1 /* Test for execute permission. */ +#define OS_ACC_W_OK 2 /* Test for write permission. */ +#define OS_ACC_R_OK 4 /* Test for read permission. */ +#define OS_ACC_RW_OK (OS_ACC_W_OK | OS_ACC_R_OK) /* Test for RW permission */ + +/* + * types taken from stat_file() in hostfs_user.c + * (if they are wrong here, they are wrong there...). + */ +struct uml_stat { + int ust_dev; /* device */ + unsigned long long ust_ino; /* inode */ + int ust_mode; /* protection */ + int ust_nlink; /* number of hard links */ + int ust_uid; /* user ID of owner */ + int ust_gid; /* group ID of owner */ + unsigned long long ust_size; /* total size, in bytes */ + int ust_blksize; /* blocksize for filesystem I/O */ + unsigned long long ust_blocks; /* number of blocks allocated */ + unsigned long ust_atime; /* time of last access */ + unsigned long ust_mtime; /* time of last modification */ + unsigned long ust_ctime; /* time of last change */ +}; + +struct openflags { + unsigned int r : 1; + unsigned int w : 1; + unsigned int s : 1; /* O_SYNC */ + unsigned int c : 1; /* O_CREAT */ + unsigned int t : 1; /* O_TRUNC */ + unsigned int a : 1; /* O_APPEND */ + unsigned int e : 1; /* O_EXCL */ + unsigned int cl : 1; /* FD_CLOEXEC */ +}; + +#define OPENFLAGS() ((struct openflags) { .r = 0, .w = 0, .s = 0, .c = 0, \ + .t = 0, .a = 0, .e = 0, .cl = 0 }) + +static inline struct openflags of_read(struct openflags flags) +{ + flags.r = 1; + return(flags); +} + +static inline struct openflags of_write(struct openflags flags) +{ + flags.w = 1; + return(flags); +} + +static inline struct openflags of_rdwr(struct openflags flags) +{ + return(of_read(of_write(flags))); +} + +static inline struct openflags of_set_rw(struct openflags flags, int r, int w) +{ + flags.r = r; + flags.w = w; + return(flags); +} + +static inline struct openflags of_sync(struct openflags flags) +{ + flags.s = 1; + return(flags); +} + +static inline struct openflags of_create(struct openflags flags) +{ + flags.c = 1; + return(flags); +} + +static inline struct openflags of_trunc(struct openflags flags) +{ + flags.t = 1; + return(flags); +} + +static inline struct openflags of_append(struct openflags flags) +{ + flags.a = 1; + return(flags); +} + +static inline struct openflags of_excl(struct openflags flags) +{ + flags.e = 1; + return(flags); +} + +static inline struct openflags of_cloexec(struct openflags flags) +{ + flags.cl = 1; + return(flags); +} + +extern int os_stat_file(const char *file_name, struct uml_stat *buf); +extern int os_stat_fd(const int fd, struct uml_stat *buf); +extern int os_access(const char *file, int mode); +extern void os_print_error(int error, const char* str); +extern int os_get_exec_close(int fd, int *close_on_exec); +extern int os_set_exec_close(int fd, int close_on_exec); +extern int os_ioctl_generic(int fd, unsigned int cmd, unsigned long arg); +extern int os_window_size(int fd, int *rows, int *cols); +extern int os_new_tty_pgrp(int fd, int pid); +extern int os_get_ifname(int fd, char *namebuf); +extern int os_set_slip(int fd); +extern int os_set_owner(int fd, int pid); +extern int os_sigio_async(int master, int slave); +extern int os_mode_fd(int fd, int mode); + +extern int os_seek_file(int fd, __u64 offset); +extern int os_open_file(char *file, struct openflags flags, int mode); +extern int os_read_file(int fd, void *buf, int len); +extern int os_write_file(int fd, const void *buf, int count); +extern int os_file_size(char *file, long long *size_out); +extern int os_file_modtime(char *file, unsigned long *modtime); +extern int os_pipe(int *fd, int stream, int close_on_exec); +extern int os_set_fd_async(int fd, int owner); +extern int os_clear_fd_async(int fd); +extern int os_set_fd_block(int fd, int blocking); +extern int os_accept_connection(int fd); +extern int os_create_unix_socket(char *file, int len, int close_on_exec); +extern int os_shutdown_socket(int fd, int r, int w); +extern void os_close_file(int fd); +extern int os_rcv_fd(int fd, int *helper_pid_out); +extern int create_unix_socket(char *file, int len, int close_on_exec); +extern int os_connect_socket(char *name); +extern int os_file_type(char *file); +extern int os_file_mode(char *file, struct openflags *mode_out); +extern int os_lock_file(int fd, int excl); + +extern unsigned long os_process_pc(int pid); +extern int os_process_parent(int pid); +extern void os_stop_process(int pid); +extern void os_kill_process(int pid, int reap_child); +extern void os_kill_ptraced_process(int pid, int reap_child); +extern void os_usr1_process(int pid); +extern int os_getpid(void); + +extern int os_map_memory(void *virt, int fd, unsigned long long off, + unsigned long len, int r, int w, int x); +extern int os_protect_memory(void *addr, unsigned long len, + int r, int w, int x); +extern int os_unmap_memory(void *addr, int len); +extern void os_flush_stdout(void); +extern unsigned long long os_usecs(void); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/process.h b/arch/um/include/process.h new file mode 100644 index 000000000000..5af9157ff54f --- /dev/null +++ b/arch/um/include/process.h @@ -0,0 +1,25 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __PROCESS_H__ +#define __PROCESS_H__ + +#include <signal.h> + +extern void sig_handler(int sig, struct sigcontext sc); +extern void alarm_handler(int sig, struct sigcontext sc); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/ptrace_user.h b/arch/um/include/ptrace_user.h new file mode 100644 index 000000000000..f3450e6bc18d --- /dev/null +++ b/arch/um/include/ptrace_user.h @@ -0,0 +1,60 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __PTRACE_USER_H__ +#define __PTRACE_USER_H__ + +#include "sysdep/ptrace_user.h" + +extern int ptrace_getregs(long pid, unsigned long *regs_out); +extern int ptrace_setregs(long pid, unsigned long *regs_in); +extern int ptrace_getfpregs(long pid, unsigned long *regs_out); +extern int ptrace_setfpregs(long pid, unsigned long *regs); +extern void arch_enter_kernel(void *task, int pid); +extern void arch_leave_kernel(void *task, int pid); +extern void ptrace_pokeuser(unsigned long addr, unsigned long data); + + +/* syscall emulation path in ptrace */ + +#ifndef PTRACE_SYSEMU +#define PTRACE_SYSEMU 31 +#endif +#ifndef PTRACE_SYSEMU_SINGLESTEP +#define PTRACE_SYSEMU_SINGLESTEP 32 +#endif + +/* On architectures, that started to support PTRACE_O_TRACESYSGOOD + * in linux 2.4, there are two different definitions of + * PTRACE_SETOPTIONS: linux 2.4 uses 21 while linux 2.6 uses 0x4200. + * For binary compatibility, 2.6 also supports the old "21", named + * PTRACE_OLDSETOPTION. On these architectures, UML always must use + * "21", to ensure the kernel runs on 2.4 and 2.6 host without + * recompilation. So, we use PTRACE_OLDSETOPTIONS in UML. + * We also want to be able to build the kernel on 2.4, which doesn't + * have PTRACE_OLDSETOPTIONS. So, if it is missing, we declare + * PTRACE_OLDSETOPTIONS to to be the same as PTRACE_SETOPTIONS. + * + * On architectures, that start to support PTRACE_O_TRACESYSGOOD on + * linux 2.6, PTRACE_OLDSETOPTIONS never is defined, and also isn't + * supported by the host kernel. In that case, our trick lets us use + * the new 0x4200 with the name PTRACE_OLDSETOPTIONS. + */ +#ifndef PTRACE_OLDSETOPTIONS +#define PTRACE_OLDSETOPTIONS PTRACE_SETOPTIONS +#endif + +void set_using_sysemu(int value); +int get_using_sysemu(void); +extern int sysemu_supported; + +#define SELECT_PTRACE_OPERATION(sysemu_mode, singlestep_mode) \ + (((int[3][3] ) { \ + { PTRACE_SYSCALL, PTRACE_SYSCALL, PTRACE_SINGLESTEP }, \ + { PTRACE_SYSEMU, PTRACE_SYSEMU, PTRACE_SINGLESTEP }, \ + { PTRACE_SYSEMU, PTRACE_SYSEMU_SINGLESTEP, PTRACE_SYSEMU_SINGLESTEP }}) \ + [sysemu_mode][singlestep_mode]) + +#endif diff --git a/arch/um/include/registers.h b/arch/um/include/registers.h new file mode 100644 index 000000000000..8744abb5224f --- /dev/null +++ b/arch/um/include/registers.h @@ -0,0 +1,29 @@ +/* + * Copyright (C) 2004 PathScale, Inc + * Licensed under the GPL + */ + +#ifndef __REGISTERS_H +#define __REGISTERS_H + +#include "sysdep/ptrace.h" + +extern void init_thread_registers(union uml_pt_regs *to); +extern int save_fp_registers(int pid, unsigned long *fp_regs); +extern int restore_fp_registers(int pid, unsigned long *fp_regs); +extern void save_registers(int pid, union uml_pt_regs *regs); +extern void restore_registers(int pid, union uml_pt_regs *regs); +extern void init_registers(int pid); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sigcontext.h b/arch/um/include/sigcontext.h new file mode 100644 index 000000000000..59816ca7a8df --- /dev/null +++ b/arch/um/include/sigcontext.h @@ -0,0 +1,25 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __UML_SIGCONTEXT_H__ +#define __UML_SIGCONTEXT_H__ + +#include "sysdep/sigcontext.h" + +extern int sc_size(void *data); +extern void sc_to_sc(void *to_ptr, void *from_ptr); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sigio.h b/arch/um/include/sigio.h new file mode 100644 index 000000000000..37d76e29a147 --- /dev/null +++ b/arch/um/include/sigio.h @@ -0,0 +1,28 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SIGIO_H__ +#define __SIGIO_H__ + +extern int write_sigio_irq(int fd); +extern int register_sigio_fd(int fd); +extern int read_sigio_fd(int fd); +extern int add_sigio_fd(int fd, int read); +extern int ignore_sigio_fd(int fd); +extern void sigio_lock(void); +extern void sigio_unlock(void); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/signal_kern.h b/arch/um/include/signal_kern.h new file mode 100644 index 000000000000..aeb5d5ab1dfd --- /dev/null +++ b/arch/um/include/signal_kern.h @@ -0,0 +1,22 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SIGNAL_KERN_H__ +#define __SIGNAL_KERN_H__ + +extern int have_signals(void *t); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/signal_user.h b/arch/um/include/signal_user.h new file mode 100644 index 000000000000..b075e543d864 --- /dev/null +++ b/arch/um/include/signal_user.h @@ -0,0 +1,28 @@ +/* + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SIGNAL_USER_H__ +#define __SIGNAL_USER_H__ + +extern int signal_stack_size; + +extern int change_sig(int signal, int on); +extern void set_sigstack(void *stack, int size); +extern void set_handler(int sig, void (*handler)(int), int flags, ...); +extern int set_signals(int enable); +extern int get_signals(void); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/skas_ptrace.h b/arch/um/include/skas_ptrace.h new file mode 100644 index 000000000000..cfb5fb4f5b91 --- /dev/null +++ b/arch/um/include/skas_ptrace.h @@ -0,0 +1,36 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_PTRACE_H +#define __SKAS_PTRACE_H + +struct ptrace_faultinfo { + int is_write; + unsigned long addr; +}; + +struct ptrace_ldt { + int func; + void *ptr; + unsigned long bytecount; +}; + +#define PTRACE_FAULTINFO 52 +#define PTRACE_SIGPENDING 53 +#define PTRACE_LDT 54 +#define PTRACE_SWITCH_MM 55 + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/syscall_user.h b/arch/um/include/syscall_user.h new file mode 100644 index 000000000000..811d0ec2445e --- /dev/null +++ b/arch/um/include/syscall_user.h @@ -0,0 +1,23 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SYSCALL_USER_H +#define __SYSCALL_USER_H + +extern int record_syscall_start(int syscall); +extern void record_syscall_end(int index, long result); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-i386/checksum.h b/arch/um/include/sysdep-i386/checksum.h new file mode 100644 index 000000000000..3a2a45811aa3 --- /dev/null +++ b/arch/um/include/sysdep-i386/checksum.h @@ -0,0 +1,219 @@ +/* + * Licensed under the GPL + */ + +#ifndef __UM_SYSDEP_CHECKSUM_H +#define __UM_SYSDEP_CHECKSUM_H + +#include "linux/in6.h" +#include "linux/string.h" + +/* + * computes the checksum of a memory block at buff, length len, + * and adds in "sum" (32-bit) + * + * returns a 32-bit number suitable for feeding into itself + * or csum_tcpudp_magic + * + * this function must be called with even lengths, except + * for the last fragment, which may be odd + * + * it's best to have buff aligned on a 32-bit boundary + */ +unsigned int csum_partial(const unsigned char * buff, int len, + unsigned int sum); + +/* + * the same as csum_partial, but copies from src while it + * checksums, and handles user-space pointer exceptions correctly, when needed. + * + * here even more important to align src and dst on a 32-bit (or even + * better 64-bit) boundary + */ + +unsigned int csum_partial_copy_to(const unsigned char *src, unsigned char *dst, + int len, int sum, int *err_ptr); +unsigned int csum_partial_copy_from(const unsigned char *src, unsigned char *dst, + int len, int sum, int *err_ptr); + +/* + * Note: when you get a NULL pointer exception here this means someone + * passed in an incorrect kernel address to one of these functions. + * + * If you use these functions directly please don't forget the + * access_ok(). + */ + +static __inline__ +unsigned int csum_partial_copy_nocheck(const unsigned char *src, unsigned char *dst, + int len, int sum) +{ + memcpy(dst, src, len); + return(csum_partial(dst, len, sum)); +} + +static __inline__ +unsigned int csum_partial_copy_from_user(const unsigned char *src, unsigned char *dst, + int len, int sum, int *err_ptr) +{ + return csum_partial_copy_from(src, dst, len, sum, err_ptr); +} + +/* + * These are the old (and unsafe) way of doing checksums, a warning message + * will be printed if they are used and an exception occurs. + * + * these functions should go away after some time. + */ + +#define csum_partial_copy_fromuser csum_partial_copy_from_user +unsigned int csum_partial_copy(const unsigned char *src, unsigned char *dst, int len, int sum); + +/* + * This is a version of ip_compute_csum() optimized for IP headers, + * which always checksum on 4 octet boundaries. + * + * By Jorge Cwik <jorge@laser.satlink.net>, adapted for linux by + * Arnt Gulbrandsen. + */ +static inline unsigned short ip_fast_csum(unsigned char * iph, + unsigned int ihl) +{ + unsigned int sum; + + __asm__ __volatile__( + "movl (%1), %0 ;\n" + "subl $4, %2 ;\n" + "jbe 2f ;\n" + "addl 4(%1), %0 ;\n" + "adcl 8(%1), %0 ;\n" + "adcl 12(%1), %0 ;\n" +"1: adcl 16(%1), %0 ;\n" + "lea 4(%1), %1 ;\n" + "decl %2 ;\n" + "jne 1b ;\n" + "adcl $0, %0 ;\n" + "movl %0, %2 ;\n" + "shrl $16, %0 ;\n" + "addw %w2, %w0 ;\n" + "adcl $0, %0 ;\n" + "notl %0 ;\n" +"2: ;\n" + /* Since the input registers which are loaded with iph and ipl + are modified, we must also specify them as outputs, or gcc + will assume they contain their original values. */ + : "=r" (sum), "=r" (iph), "=r" (ihl) + : "1" (iph), "2" (ihl) + : "memory"); + return(sum); +} + +/* + * Fold a partial checksum + */ + +static inline unsigned int csum_fold(unsigned int sum) +{ + __asm__( + "addl %1, %0 ;\n" + "adcl $0xffff, %0 ;\n" + : "=r" (sum) + : "r" (sum << 16), "0" (sum & 0xffff0000) + ); + return (~sum) >> 16; +} + +static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, + unsigned long daddr, + unsigned short len, + unsigned short proto, + unsigned int sum) +{ + __asm__( + "addl %1, %0 ;\n" + "adcl %2, %0 ;\n" + "adcl %3, %0 ;\n" + "adcl $0, %0 ;\n" + : "=r" (sum) + : "g" (daddr), "g"(saddr), "g"((ntohs(len)<<16)+proto*256), "0"(sum)); + return sum; +} + +/* + * computes the checksum of the TCP/UDP pseudo-header + * returns a 16-bit checksum, already complemented + */ +static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, + unsigned long daddr, + unsigned short len, + unsigned short proto, + unsigned int sum) +{ + return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); +} + +/* + * this routine is used for miscellaneous IP-like checksums, mainly + * in icmp.c + */ + +static inline unsigned short ip_compute_csum(unsigned char * buff, int len) +{ + return csum_fold (csum_partial(buff, len, 0)); +} + +#define _HAVE_ARCH_IPV6_CSUM +static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr, + struct in6_addr *daddr, + __u32 len, + unsigned short proto, + unsigned int sum) +{ + __asm__( + "addl 0(%1), %0 ;\n" + "adcl 4(%1), %0 ;\n" + "adcl 8(%1), %0 ;\n" + "adcl 12(%1), %0 ;\n" + "adcl 0(%2), %0 ;\n" + "adcl 4(%2), %0 ;\n" + "adcl 8(%2), %0 ;\n" + "adcl 12(%2), %0 ;\n" + "adcl %3, %0 ;\n" + "adcl %4, %0 ;\n" + "adcl $0, %0 ;\n" + : "=&r" (sum) + : "r" (saddr), "r" (daddr), + "r"(htonl(len)), "r"(htonl(proto)), "0"(sum)); + + return csum_fold(sum); +} + +/* + * Copy and checksum to user + */ +#define HAVE_CSUM_COPY_USER +static __inline__ unsigned int csum_and_copy_to_user(const unsigned char *src, + unsigned char *dst, + int len, int sum, int *err_ptr) +{ + if (access_ok(VERIFY_WRITE, dst, len)) + return(csum_partial_copy_to(src, dst, len, sum, err_ptr)); + + if (len) + *err_ptr = -EFAULT; + + return -1; /* invalid checksum */ +} + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-i386/ptrace.h b/arch/um/include/sysdep-i386/ptrace.h new file mode 100644 index 000000000000..661d495e2044 --- /dev/null +++ b/arch/um/include/sysdep-i386/ptrace.h @@ -0,0 +1,241 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SYSDEP_I386_PTRACE_H +#define __SYSDEP_I386_PTRACE_H + +#include "uml-config.h" +#include "user_constants.h" + +#define MAX_REG_NR (UM_FRAME_SIZE / sizeof(unsigned long)) +#define MAX_REG_OFFSET (UM_FRAME_SIZE) + +extern void update_debugregs(int seq); + +/* syscall emulation path in ptrace */ + +#ifndef PTRACE_SYSEMU +#define PTRACE_SYSEMU 31 +#endif + +void set_using_sysemu(int value); +int get_using_sysemu(void); +extern int sysemu_supported; + +#ifdef UML_CONFIG_MODE_TT +#include "sysdep/sc.h" +#endif + +#ifdef UML_CONFIG_MODE_SKAS + +#include "skas_ptregs.h" + +#define REGS_IP(r) ((r)[HOST_IP]) +#define REGS_SP(r) ((r)[HOST_SP]) +#define REGS_EFLAGS(r) ((r)[HOST_EFLAGS]) +#define REGS_EAX(r) ((r)[HOST_EAX]) +#define REGS_EBX(r) ((r)[HOST_EBX]) +#define REGS_ECX(r) ((r)[HOST_ECX]) +#define REGS_EDX(r) ((r)[HOST_EDX]) +#define REGS_ESI(r) ((r)[HOST_ESI]) +#define REGS_EDI(r) ((r)[HOST_EDI]) +#define REGS_EBP(r) ((r)[HOST_EBP]) +#define REGS_CS(r) ((r)[HOST_CS]) +#define REGS_SS(r) ((r)[HOST_SS]) +#define REGS_DS(r) ((r)[HOST_DS]) +#define REGS_ES(r) ((r)[HOST_ES]) +#define REGS_FS(r) ((r)[HOST_FS]) +#define REGS_GS(r) ((r)[HOST_GS]) + +#define REGS_SET_SYSCALL_RETURN(r, res) REGS_EAX(r) = (res) + +#define REGS_RESTART_SYSCALL(r) IP_RESTART_SYSCALL(REGS_IP(r)) + +#define REGS_SEGV_IS_FIXABLE(r) SEGV_IS_FIXABLE((r)->trap_type) + +#define REGS_FAULT_ADDR(r) ((r)->fault_addr) + +#define REGS_FAULT_WRITE(r) FAULT_WRITE((r)->fault_type) + +#endif +#ifndef PTRACE_SYSEMU_SINGLESTEP +#define PTRACE_SYSEMU_SINGLESTEP 32 +#endif + +#include "choose-mode.h" + +union uml_pt_regs { +#ifdef UML_CONFIG_MODE_TT + struct tt_regs { + long syscall; + void *sc; + } tt; +#endif +#ifdef UML_CONFIG_MODE_SKAS + struct skas_regs { + unsigned long regs[HOST_FRAME_SIZE]; + unsigned long fp[HOST_FP_SIZE]; + unsigned long xfp[HOST_XFP_SIZE]; + unsigned long fault_addr; + unsigned long fault_type; + unsigned long trap_type; + long syscall; + int is_user; + } skas; +#endif +}; + +#define EMPTY_UML_PT_REGS { } + +extern int mode_tt; + +#define UPT_SC(r) ((r)->tt.sc) +#define UPT_IP(r) \ + __CHOOSE_MODE(SC_IP(UPT_SC(r)), REGS_IP((r)->skas.regs)) +#define UPT_SP(r) \ + __CHOOSE_MODE(SC_SP(UPT_SC(r)), REGS_SP((r)->skas.regs)) +#define UPT_EFLAGS(r) \ + __CHOOSE_MODE(SC_EFLAGS(UPT_SC(r)), REGS_EFLAGS((r)->skas.regs)) +#define UPT_EAX(r) \ + __CHOOSE_MODE(SC_EAX(UPT_SC(r)), REGS_EAX((r)->skas.regs)) +#define UPT_EBX(r) \ + __CHOOSE_MODE(SC_EBX(UPT_SC(r)), REGS_EBX((r)->skas.regs)) +#define UPT_ECX(r) \ + __CHOOSE_MODE(SC_ECX(UPT_SC(r)), REGS_ECX((r)->skas.regs)) +#define UPT_EDX(r) \ + __CHOOSE_MODE(SC_EDX(UPT_SC(r)), REGS_EDX((r)->skas.regs)) +#define UPT_ESI(r) \ + __CHOOSE_MODE(SC_ESI(UPT_SC(r)), REGS_ESI((r)->skas.regs)) +#define UPT_EDI(r) \ + __CHOOSE_MODE(SC_EDI(UPT_SC(r)), REGS_EDI((r)->skas.regs)) +#define UPT_EBP(r) \ + __CHOOSE_MODE(SC_EBP(UPT_SC(r)), REGS_EBP((r)->skas.regs)) +#define UPT_ORIG_EAX(r) \ + __CHOOSE_MODE((r)->tt.syscall, (r)->skas.syscall) +#define UPT_CS(r) \ + __CHOOSE_MODE(SC_CS(UPT_SC(r)), REGS_CS((r)->skas.regs)) +#define UPT_SS(r) \ + __CHOOSE_MODE(SC_SS(UPT_SC(r)), REGS_SS((r)->skas.regs)) +#define UPT_DS(r) \ + __CHOOSE_MODE(SC_DS(UPT_SC(r)), REGS_DS((r)->skas.regs)) +#define UPT_ES(r) \ + __CHOOSE_MODE(SC_ES(UPT_SC(r)), REGS_ES((r)->skas.regs)) +#define UPT_FS(r) \ + __CHOOSE_MODE(SC_FS(UPT_SC(r)), REGS_FS((r)->skas.regs)) +#define UPT_GS(r) \ + __CHOOSE_MODE(SC_GS(UPT_SC(r)), REGS_GS((r)->skas.regs)) + +#define UPT_SYSCALL_ARG1(r) UPT_EBX(r) +#define UPT_SYSCALL_ARG2(r) UPT_ECX(r) +#define UPT_SYSCALL_ARG3(r) UPT_EDX(r) +#define UPT_SYSCALL_ARG4(r) UPT_ESI(r) +#define UPT_SYSCALL_ARG5(r) UPT_EDI(r) +#define UPT_SYSCALL_ARG6(r) UPT_EBP(r) + +extern int user_context(unsigned long sp); + +#define UPT_IS_USER(r) \ + CHOOSE_MODE(user_context(UPT_SP(r)), (r)->skas.is_user) + +struct syscall_args { + unsigned long args[6]; +}; + +#define SYSCALL_ARGS(r) ((struct syscall_args) \ + { .args = { UPT_SYSCALL_ARG1(r), \ + UPT_SYSCALL_ARG2(r), \ + UPT_SYSCALL_ARG3(r), \ + UPT_SYSCALL_ARG4(r), \ + UPT_SYSCALL_ARG5(r), \ + UPT_SYSCALL_ARG6(r) } } ) + +#define UPT_REG(regs, reg) \ + ({ unsigned long val; \ + switch(reg){ \ + case EIP: val = UPT_IP(regs); break; \ + case UESP: val = UPT_SP(regs); break; \ + case EAX: val = UPT_EAX(regs); break; \ + case EBX: val = UPT_EBX(regs); break; \ + case ECX: val = UPT_ECX(regs); break; \ + case EDX: val = UPT_EDX(regs); break; \ + case ESI: val = UPT_ESI(regs); break; \ + case EDI: val = UPT_EDI(regs); break; \ + case EBP: val = UPT_EBP(regs); break; \ + case ORIG_EAX: val = UPT_ORIG_EAX(regs); break; \ + case CS: val = UPT_CS(regs); break; \ + case SS: val = UPT_SS(regs); break; \ + case DS: val = UPT_DS(regs); break; \ + case ES: val = UPT_ES(regs); break; \ + case FS: val = UPT_FS(regs); break; \ + case GS: val = UPT_GS(regs); break; \ + case EFL: val = UPT_EFLAGS(regs); break; \ + default : \ + panic("Bad register in UPT_REG : %d\n", reg); \ + val = -1; \ + } \ + val; \ + }) + + +#define UPT_SET(regs, reg, val) \ + do { \ + switch(reg){ \ + case EIP: UPT_IP(regs) = val; break; \ + case UESP: UPT_SP(regs) = val; break; \ + case EAX: UPT_EAX(regs) = val; break; \ + case EBX: UPT_EBX(regs) = val; break; \ + case ECX: UPT_ECX(regs) = val; break; \ + case EDX: UPT_EDX(regs) = val; break; \ + case ESI: UPT_ESI(regs) = val; break; \ + case EDI: UPT_EDI(regs) = val; break; \ + case EBP: UPT_EBP(regs) = val; break; \ + case ORIG_EAX: UPT_ORIG_EAX(regs) = val; break; \ + case CS: UPT_CS(regs) = val; break; \ + case SS: UPT_SS(regs) = val; break; \ + case DS: UPT_DS(regs) = val; break; \ + case ES: UPT_ES(regs) = val; break; \ + case FS: UPT_FS(regs) = val; break; \ + case GS: UPT_GS(regs) = val; break; \ + case EFL: UPT_EFLAGS(regs) = val; break; \ + default : \ + panic("Bad register in UPT_SET : %d\n", reg); \ + break; \ + } \ + } while (0) + +#define UPT_SET_SYSCALL_RETURN(r, res) \ + CHOOSE_MODE(SC_SET_SYSCALL_RETURN(UPT_SC(r), (res)), \ + REGS_SET_SYSCALL_RETURN((r)->skas.regs, (res))) + +#define UPT_RESTART_SYSCALL(r) \ + CHOOSE_MODE(SC_RESTART_SYSCALL(UPT_SC(r)), \ + REGS_RESTART_SYSCALL((r)->skas.regs)) + +#define UPT_ORIG_SYSCALL(r) UPT_EAX(r) +#define UPT_SYSCALL_NR(r) UPT_ORIG_EAX(r) +#define UPT_SYSCALL_RET(r) UPT_EAX(r) + +#define UPT_SEGV_IS_FIXABLE(r) \ + CHOOSE_MODE(SC_SEGV_IS_FIXABLE(UPT_SC(r)), \ + REGS_SEGV_IS_FIXABLE(&r->skas)) + +#define UPT_FAULT_ADDR(r) \ + __CHOOSE_MODE(SC_FAULT_ADDR(UPT_SC(r)), REGS_FAULT_ADDR(&r->skas)) + +#define UPT_FAULT_WRITE(r) \ + CHOOSE_MODE(SC_FAULT_WRITE(UPT_SC(r)), REGS_FAULT_WRITE(&r->skas)) + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-i386/ptrace_user.h b/arch/um/include/sysdep-i386/ptrace_user.h new file mode 100644 index 000000000000..eca8066e7a43 --- /dev/null +++ b/arch/um/include/sysdep-i386/ptrace_user.h @@ -0,0 +1,62 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SYSDEP_I386_PTRACE_USER_H__ +#define __SYSDEP_I386_PTRACE_USER_H__ + +#include <sys/ptrace.h> +#include <linux/ptrace.h> +#include <asm/ptrace.h> + +#define PT_OFFSET(r) ((r) * sizeof(long)) + +#define PT_SYSCALL_NR(regs) ((regs)[ORIG_EAX]) +#define PT_SYSCALL_NR_OFFSET PT_OFFSET(ORIG_EAX) + +#define PT_SYSCALL_ARG1_OFFSET PT_OFFSET(EBX) +#define PT_SYSCALL_ARG2_OFFSET PT_OFFSET(ECX) +#define PT_SYSCALL_ARG3_OFFSET PT_OFFSET(EDX) +#define PT_SYSCALL_ARG4_OFFSET PT_OFFSET(ESI) +#define PT_SYSCALL_ARG5_OFFSET PT_OFFSET(EDI) + +#define PT_SYSCALL_RET_OFFSET PT_OFFSET(EAX) + +#define PT_IP_OFFSET PT_OFFSET(EIP) +#define PT_IP(regs) ((regs)[EIP]) +#define PT_SP(regs) ((regs)[UESP]) + +#ifndef FRAME_SIZE +#define FRAME_SIZE (17) +#endif +#define FRAME_SIZE_OFFSET (FRAME_SIZE * sizeof(unsigned long)) + +#define FP_FRAME_SIZE (27) +#define FPX_FRAME_SIZE (128) + +#ifdef PTRACE_GETREGS +#define UM_HAVE_GETREGS +#endif + +#ifdef PTRACE_SETREGS +#define UM_HAVE_SETREGS +#endif + +#ifdef PTRACE_GETFPREGS +#define UM_HAVE_GETFPREGS +#endif + +#ifdef PTRACE_SETFPREGS +#define UM_HAVE_SETFPREGS +#endif + +#ifdef PTRACE_GETFPXREGS +#define UM_HAVE_GETFPXREGS +#endif + +#ifdef PTRACE_SETFPXREGS +#define UM_HAVE_SETFPXREGS +#endif + +#endif diff --git a/arch/um/include/sysdep-i386/sigcontext.h b/arch/um/include/sysdep-i386/sigcontext.h new file mode 100644 index 000000000000..dfee589de360 --- /dev/null +++ b/arch/um/include/sysdep-i386/sigcontext.h @@ -0,0 +1,49 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SYS_SIGCONTEXT_I386_H +#define __SYS_SIGCONTEXT_I386_H + +#include <sysdep/sc.h> + +#define IP_RESTART_SYSCALL(ip) ((ip) -= 2) + +#define SC_RESTART_SYSCALL(sc) IP_RESTART_SYSCALL(SC_IP(sc)) +#define SC_SET_SYSCALL_RETURN(sc, result) SC_EAX(sc) = (result) + +#define SC_FAULT_ADDR(sc) SC_CR2(sc) +#define SC_FAULT_TYPE(sc) SC_ERR(sc) + +#define FAULT_WRITE(err) (err & 2) +#define TO_SC_ERR(is_write) ((is_write) ? 2 : 0) + +#define SC_FAULT_WRITE(sc) (FAULT_WRITE(SC_ERR(sc))) + +#define SC_TRAP_TYPE(sc) SC_TRAPNO(sc) + +/* ptrace expects that, at the start of a system call, %eax contains + * -ENOSYS, so this makes it so. + */ +#define SC_START_SYSCALL(sc) do SC_EAX(sc) = -ENOSYS; while(0) + +/* This is Page Fault */ +#define SEGV_IS_FIXABLE(trap) (trap == 14) + +#define SC_SEGV_IS_FIXABLE(sc) (SEGV_IS_FIXABLE(SC_TRAPNO(sc))) + +extern unsigned long *sc_sigmask(void *sc_ptr); +extern int sc_get_fpregs(unsigned long buf, void *sc_ptr); + +#endif +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-i386/signal.h b/arch/um/include/sysdep-i386/signal.h new file mode 100644 index 000000000000..b1e1f7a77499 --- /dev/null +++ b/arch/um/include/sysdep-i386/signal.h @@ -0,0 +1,25 @@ +/* + * Copyright (C) 2004 PathScale, Inc + * Licensed under the GPL + */ + +#ifndef __I386_SIGNAL_H_ +#define __I386_SIGNAL_H_ + +#include <signal.h> + +#define ARCH_GET_SIGCONTEXT(sc, sig) \ + do sc = (struct sigcontext *) (&sig + 1); while(0) + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-i386/syscalls.h b/arch/um/include/sysdep-i386/syscalls.h new file mode 100644 index 000000000000..5db81ec9087d --- /dev/null +++ b/arch/um/include/sysdep-i386/syscalls.h @@ -0,0 +1,123 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "asm/unistd.h" +#include "sysdep/ptrace.h" + +typedef long syscall_handler_t(struct pt_regs); + +/* Not declared on x86, incompatible declarations on x86_64, so these have + * to go here rather than in sys_call_table.c + */ +extern syscall_handler_t sys_ptrace; +extern syscall_handler_t sys_rt_sigaction; + +extern syscall_handler_t old_mmap_i386; + +#define EXECUTE_SYSCALL(syscall, regs) \ + ((long (*)(struct syscall_args)) (*sys_call_table[syscall]))(SYSCALL_ARGS(®s->regs)) + +extern long sys_mmap2(unsigned long addr, unsigned long len, + unsigned long prot, unsigned long flags, + unsigned long fd, unsigned long pgoff); + +/* On i386 they choose a meaningless naming.*/ +#define __NR_kexec_load __NR_sys_kexec_load + +#define ARCH_SYSCALLS \ + [ __NR_waitpid ] = (syscall_handler_t *) sys_waitpid, \ + [ __NR_break ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_oldstat ] = (syscall_handler_t *) sys_stat, \ + [ __NR_umount ] = (syscall_handler_t *) sys_oldumount, \ + [ __NR_stime ] = um_stime, \ + [ __NR_oldfstat ] = (syscall_handler_t *) sys_fstat, \ + [ __NR_stty ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_gtty ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_nice ] = (syscall_handler_t *) sys_nice, \ + [ __NR_ftime ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_prof ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_signal ] = (syscall_handler_t *) sys_signal, \ + [ __NR_lock ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_mpx ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_ulimit ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_oldolduname ] = (syscall_handler_t *) sys_olduname, \ + [ __NR_sigaction ] = (syscall_handler_t *) sys_sigaction, \ + [ __NR_sgetmask ] = (syscall_handler_t *) sys_sgetmask, \ + [ __NR_ssetmask ] = (syscall_handler_t *) sys_ssetmask, \ + [ __NR_sigsuspend ] = (syscall_handler_t *) sys_sigsuspend, \ + [ __NR_sigpending ] = (syscall_handler_t *) sys_sigpending, \ + [ __NR_oldlstat ] = (syscall_handler_t *) sys_lstat, \ + [ __NR_readdir ] = old_readdir, \ + [ __NR_profil ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_socketcall ] = (syscall_handler_t *) sys_socketcall, \ + [ __NR_olduname ] = (syscall_handler_t *) sys_uname, \ + [ __NR_iopl ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_idle ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_ipc ] = (syscall_handler_t *) sys_ipc, \ + [ __NR_sigreturn ] = (syscall_handler_t *) sys_sigreturn, \ + [ __NR_sigprocmask ] = (syscall_handler_t *) sys_sigprocmask, \ + [ __NR_bdflush ] = (syscall_handler_t *) sys_bdflush, \ + [ __NR__llseek ] = (syscall_handler_t *) sys_llseek, \ + [ __NR__newselect ] = (syscall_handler_t *) sys_select, \ + [ __NR_vm86 ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_mmap ] = (syscall_handler_t *) old_mmap_i386, \ + [ __NR_ugetrlimit ] = (syscall_handler_t *) sys_getrlimit, \ + [ __NR_mmap2 ] = (syscall_handler_t *) sys_mmap2, \ + [ __NR_truncate64 ] = (syscall_handler_t *) sys_truncate64, \ + [ __NR_ftruncate64 ] = (syscall_handler_t *) sys_ftruncate64, \ + [ __NR_stat64 ] = (syscall_handler_t *) sys_stat64, \ + [ __NR_lstat64 ] = (syscall_handler_t *) sys_lstat64, \ + [ __NR_fstat64 ] = (syscall_handler_t *) sys_fstat64, \ + [ __NR_fcntl64 ] = (syscall_handler_t *) sys_fcntl64, \ + [ __NR_sendfile64 ] = (syscall_handler_t *) sys_sendfile64, \ + [ __NR_statfs64 ] = (syscall_handler_t *) sys_statfs64, \ + [ __NR_fstatfs64 ] = (syscall_handler_t *) sys_fstatfs64, \ + [ __NR_fadvise64_64 ] = (syscall_handler_t *) sys_fadvise64_64, \ + [ __NR_select ] = (syscall_handler_t *) old_select, \ + [ __NR_vm86old ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_modify_ldt ] = (syscall_handler_t *) sys_modify_ldt, \ + [ __NR_lchown32 ] = (syscall_handler_t *) sys_lchown, \ + [ __NR_getuid32 ] = (syscall_handler_t *) sys_getuid, \ + [ __NR_getgid32 ] = (syscall_handler_t *) sys_getgid, \ + [ __NR_geteuid32 ] = (syscall_handler_t *) sys_geteuid, \ + [ __NR_getegid32 ] = (syscall_handler_t *) sys_getegid, \ + [ __NR_setreuid32 ] = (syscall_handler_t *) sys_setreuid, \ + [ __NR_setregid32 ] = (syscall_handler_t *) sys_setregid, \ + [ __NR_getgroups32 ] = (syscall_handler_t *) sys_getgroups, \ + [ __NR_setgroups32 ] = (syscall_handler_t *) sys_setgroups, \ + [ __NR_fchown32 ] = (syscall_handler_t *) sys_fchown, \ + [ __NR_setresuid32 ] = (syscall_handler_t *) sys_setresuid, \ + [ __NR_getresuid32 ] = (syscall_handler_t *) sys_getresuid, \ + [ __NR_setresgid32 ] = (syscall_handler_t *) sys_setresgid, \ + [ __NR_getresgid32 ] = (syscall_handler_t *) sys_getresgid, \ + [ __NR_chown32 ] = (syscall_handler_t *) sys_chown, \ + [ __NR_setuid32 ] = (syscall_handler_t *) sys_setuid, \ + [ __NR_setgid32 ] = (syscall_handler_t *) sys_setgid, \ + [ __NR_setfsuid32 ] = (syscall_handler_t *) sys_setfsuid, \ + [ __NR_setfsgid32 ] = (syscall_handler_t *) sys_setfsgid, \ + [ __NR_pivot_root ] = (syscall_handler_t *) sys_pivot_root, \ + [ __NR_mincore ] = (syscall_handler_t *) sys_mincore, \ + [ __NR_madvise ] = (syscall_handler_t *) sys_madvise, \ + [ 222 ] = (syscall_handler_t *) sys_ni_syscall, \ + [ 223 ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_set_thread_area ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_get_thread_area ] = (syscall_handler_t *) sys_ni_syscall, \ + [ 251 ] = (syscall_handler_t *) sys_ni_syscall, \ + [ 285 ] = (syscall_handler_t *) sys_ni_syscall, + +/* 222 doesn't yet have a name in include/asm-i386/unistd.h */ + +#define LAST_ARCH_SYSCALL 285 + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-ia64/ptrace.h b/arch/um/include/sysdep-ia64/ptrace.h new file mode 100644 index 000000000000..42dd8fb6f2f9 --- /dev/null +++ b/arch/um/include/sysdep-ia64/ptrace.h @@ -0,0 +1,26 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SYSDEP_IA64_PTRACE_H +#define __SYSDEP_IA64_PTRACE_H + +struct sys_pt_regs { + int foo; +}; + +#define EMPTY_REGS { 0 } + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-ia64/sigcontext.h b/arch/um/include/sysdep-ia64/sigcontext.h new file mode 100644 index 000000000000..f15fb25260ba --- /dev/null +++ b/arch/um/include/sysdep-ia64/sigcontext.h @@ -0,0 +1,20 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SYSDEP_IA64_SIGCONTEXT_H +#define __SYSDEP_IA64_SIGCONTEXT_H + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-ia64/syscalls.h b/arch/um/include/sysdep-ia64/syscalls.h new file mode 100644 index 000000000000..4a1f46ef1ebc --- /dev/null +++ b/arch/um/include/sysdep-ia64/syscalls.h @@ -0,0 +1,20 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SYSDEP_IA64_SYSCALLS_H +#define __SYSDEP_IA64_SYSCALLS_H + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-ppc/ptrace.h b/arch/um/include/sysdep-ppc/ptrace.h new file mode 100644 index 000000000000..8a27353733a9 --- /dev/null +++ b/arch/um/include/sysdep-ppc/ptrace.h @@ -0,0 +1,104 @@ +/* + * Licensed under the GPL + */ + +#ifndef __SYS_PTRACE_PPC_H +#define __SYS_PTRACE_PPC_H + +#include "linux/config.h" +#include "linux/types.h" + +/* the following taken from <asm-ppc/ptrace.h> */ + +#ifdef CONFIG_PPC64 +#define PPC_REG unsigned long /*long*/ +#else +#define PPC_REG unsigned long +#endif +struct sys_pt_regs_s { + PPC_REG gpr[32]; + PPC_REG nip; + PPC_REG msr; + PPC_REG orig_gpr3; /* Used for restarting system calls */ + PPC_REG ctr; + PPC_REG link; + PPC_REG xer; + PPC_REG ccr; + PPC_REG mq; /* 601 only (not used at present) */ + /* Used on APUS to hold IPL value. */ + PPC_REG trap; /* Reason for being here */ + PPC_REG dar; /* Fault registers */ + PPC_REG dsisr; + PPC_REG result; /* Result of a system call */ +}; + +#define NUM_REGS (sizeof(struct sys_pt_regs_s) / sizeof(PPC_REG)) + +struct sys_pt_regs { + PPC_REG regs[sizeof(struct sys_pt_regs_s) / sizeof(PPC_REG)]; +}; + +#define UM_MAX_REG (PT_FPR0) +#define UM_MAX_REG_OFFSET (UM_MAX_REG * sizeof(PPC_REG)) + +#define EMPTY_REGS { { [ 0 ... NUM_REGS - 1] = 0 } } + +#define UM_REG(r, n) ((r)->regs[n]) + +#define UM_SYSCALL_RET(r) UM_REG(r, PT_R3) +#define UM_SP(r) UM_REG(r, PT_R1) +#define UM_IP(r) UM_REG(r, PT_NIP) +#define UM_ELF_ZERO(r) UM_REG(r, PT_FPSCR) +#define UM_SYSCALL_NR(r) UM_REG(r, PT_R0) +#define UM_SYSCALL_ARG1(r) UM_REG(r, PT_ORIG_R3) +#define UM_SYSCALL_ARG2(r) UM_REG(r, PT_R4) +#define UM_SYSCALL_ARG3(r) UM_REG(r, PT_R5) +#define UM_SYSCALL_ARG4(r) UM_REG(r, PT_R6) +#define UM_SYSCALL_ARG5(r) UM_REG(r, PT_R7) +#define UM_SYSCALL_ARG6(r) UM_REG(r, PT_R8) + +#define UM_SYSCALL_NR_OFFSET (PT_R0 * sizeof(PPC_REG)) +#define UM_SYSCALL_RET_OFFSET (PT_R3 * sizeof(PPC_REG)) +#define UM_SYSCALL_ARG1_OFFSET (PT_R3 * sizeof(PPC_REG)) +#define UM_SYSCALL_ARG2_OFFSET (PT_R4 * sizeof(PPC_REG)) +#define UM_SYSCALL_ARG3_OFFSET (PT_R5 * sizeof(PPC_REG)) +#define UM_SYSCALL_ARG4_OFFSET (PT_R6 * sizeof(PPC_REG)) +#define UM_SYSCALL_ARG5_OFFSET (PT_R7 * sizeof(PPC_REG)) +#define UM_SYSCALL_ARG6_OFFSET (PT_R8 * sizeof(PPC_REG)) +#define UM_SP_OFFSET (PT_R1 * sizeof(PPC_REG)) +#define UM_IP_OFFSET (PT_NIP * sizeof(PPC_REG)) +#define UM_ELF_ZERO_OFFSET (PT_R3 * sizeof(PPC_REG)) + +#define UM_SET_SYSCALL_RETURN(_regs, result) \ +do { \ + if (result < 0) { \ + (_regs)->regs[PT_CCR] |= 0x10000000; \ + UM_SYSCALL_RET((_regs)) = -result; \ + } else { \ + UM_SYSCALL_RET((_regs)) = result; \ + } \ +} while(0) + +extern void shove_aux_table(unsigned long sp); +#define UM_FIX_EXEC_STACK(sp) shove_aux_table(sp); + +/* These aren't actually defined. The undefs are just to make sure + * everyone's clear on the concept. + */ +#undef UML_HAVE_GETREGS +#undef UML_HAVE_GETFPREGS +#undef UML_HAVE_SETREGS +#undef UML_HAVE_SETFPREGS + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-ppc/sigcontext.h b/arch/um/include/sysdep-ppc/sigcontext.h new file mode 100644 index 000000000000..f20d965de9c7 --- /dev/null +++ b/arch/um/include/sysdep-ppc/sigcontext.h @@ -0,0 +1,62 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SYS_SIGCONTEXT_PPC_H +#define __SYS_SIGCONTEXT_PPC_H + +#define DSISR_WRITE 0x02000000 + +#define SC_FAULT_ADDR(sc) ({ \ + struct sigcontext *_sc = (sc); \ + long retval = -1; \ + switch (_sc->regs->trap) { \ + case 0x300: \ + /* data exception */ \ + retval = _sc->regs->dar; \ + break; \ + case 0x400: \ + /* instruction exception */ \ + retval = _sc->regs->nip; \ + break; \ + default: \ + panic("SC_FAULT_ADDR: unhandled trap type\n"); \ + } \ + retval; \ + }) + +#define SC_FAULT_WRITE(sc) ({ \ + struct sigcontext *_sc = (sc); \ + long retval = -1; \ + switch (_sc->regs->trap) { \ + case 0x300: \ + /* data exception */ \ + retval = !!(_sc->regs->dsisr & DSISR_WRITE); \ + break; \ + case 0x400: \ + /* instruction exception: not a write */ \ + retval = 0; \ + break; \ + default: \ + panic("SC_FAULT_ADDR: unhandled trap type\n"); \ + } \ + retval; \ + }) + +#define SC_IP(sc) ((sc)->regs->nip) +#define SC_SP(sc) ((sc)->regs->gpr[1]) +#define SEGV_IS_FIXABLE(sc) (1) + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-ppc/syscalls.h b/arch/um/include/sysdep-ppc/syscalls.h new file mode 100644 index 000000000000..679df351e19b --- /dev/null +++ b/arch/um/include/sysdep-ppc/syscalls.h @@ -0,0 +1,53 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +typedef long syscall_handler_t(unsigned long arg1, unsigned long arg2, + unsigned long arg3, unsigned long arg4, + unsigned long arg5, unsigned long arg6); + +#define EXECUTE_SYSCALL(syscall, regs) \ + (*sys_call_table[syscall])(UM_SYSCALL_ARG1(®s), \ + UM_SYSCALL_ARG2(®s), \ + UM_SYSCALL_ARG3(®s), \ + UM_SYSCALL_ARG4(®s), \ + UM_SYSCALL_ARG5(®s), \ + UM_SYSCALL_ARG6(®s)) + +extern syscall_handler_t sys_mincore; +extern syscall_handler_t sys_madvise; + +/* old_mmap needs the correct prototype since syscall_kern.c includes + * this file. + */ +int old_mmap(unsigned long addr, unsigned long len, + unsigned long prot, unsigned long flags, + unsigned long fd, unsigned long offset); + +#define ARCH_SYSCALLS \ + [ __NR_modify_ldt ] = sys_ni_syscall, \ + [ __NR_pciconfig_read ] = sys_ni_syscall, \ + [ __NR_pciconfig_write ] = sys_ni_syscall, \ + [ __NR_pciconfig_iobase ] = sys_ni_syscall, \ + [ __NR_pivot_root ] = sys_ni_syscall, \ + [ __NR_multiplexer ] = sys_ni_syscall, \ + [ __NR_mmap ] = old_mmap, \ + [ __NR_madvise ] = sys_madvise, \ + [ __NR_mincore ] = sys_mincore, \ + [ __NR_iopl ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_utimes ] = (syscall_handler_t *) sys_utimes, \ + [ __NR_fadvise64 ] = (syscall_handler_t *) sys_fadvise64, + +#define LAST_ARCH_SYSCALL __NR_fadvise64 + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-x86_64/checksum.h b/arch/um/include/sysdep-x86_64/checksum.h new file mode 100644 index 000000000000..572c6c19be33 --- /dev/null +++ b/arch/um/include/sysdep-x86_64/checksum.h @@ -0,0 +1,151 @@ +/* + * Licensed under the GPL + */ + +#ifndef __UM_SYSDEP_CHECKSUM_H +#define __UM_SYSDEP_CHECKSUM_H + +#include "linux/string.h" +#include "linux/in6.h" +#include "asm/uaccess.h" + +extern unsigned int csum_partial_copy_from(const unsigned char *src, unsigned char *dst, int len, + int sum, int *err_ptr); +extern unsigned csum_partial(const unsigned char *buff, unsigned len, + unsigned sum); + +/* + * Note: when you get a NULL pointer exception here this means someone + * passed in an incorrect kernel address to one of these functions. + * + * If you use these functions directly please don't forget the + * access_ok(). + */ + +static __inline__ +unsigned int csum_partial_copy_nocheck(const unsigned char *src, unsigned char *dst, + int len, int sum) +{ + memcpy(dst, src, len); + return(csum_partial(dst, len, sum)); +} + +static __inline__ +unsigned int csum_partial_copy_from_user(const unsigned char *src, unsigned char *dst, + int len, int sum, int *err_ptr) +{ + return csum_partial_copy_from(src, dst, len, sum, err_ptr); +} + +/** + * csum_fold - Fold and invert a 32bit checksum. + * sum: 32bit unfolded sum + * + * Fold a 32bit running checksum to 16bit and invert it. This is usually + * the last step before putting a checksum into a packet. + * Make sure not to mix with 64bit checksums. + */ +static inline unsigned int csum_fold(unsigned int sum) +{ + __asm__( + " addl %1,%0\n" + " adcl $0xffff,%0" + : "=r" (sum) + : "r" (sum << 16), "0" (sum & 0xffff0000) + ); + return (~sum) >> 16; +} + +/** + * csum_tcpup_nofold - Compute an IPv4 pseudo header checksum. + * @saddr: source address + * @daddr: destination address + * @len: length of packet + * @proto: ip protocol of packet + * @sum: initial sum to be added in (32bit unfolded) + * + * Returns the pseudo header checksum the input data. Result is + * 32bit unfolded. + */ +static inline unsigned long +csum_tcpudp_nofold(unsigned saddr, unsigned daddr, unsigned short len, + unsigned short proto, unsigned int sum) +{ + asm(" addl %1, %0\n" + " adcl %2, %0\n" + " adcl %3, %0\n" + " adcl $0, %0\n" + : "=r" (sum) + : "g" (daddr), "g" (saddr), "g" ((ntohs(len)<<16)+proto*256), "0" (sum)); + return sum; +} + +/* + * computes the checksum of the TCP/UDP pseudo-header + * returns a 16-bit checksum, already complemented + */ +static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, + unsigned long daddr, + unsigned short len, + unsigned short proto, + unsigned int sum) +{ + return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); +} + +/** + * ip_fast_csum - Compute the IPv4 header checksum efficiently. + * iph: ipv4 header + * ihl: length of header / 4 + */ +static inline unsigned short ip_fast_csum(unsigned char *iph, unsigned int ihl) +{ + unsigned int sum; + + asm( " movl (%1), %0\n" + " subl $4, %2\n" + " jbe 2f\n" + " addl 4(%1), %0\n" + " adcl 8(%1), %0\n" + " adcl 12(%1), %0\n" + "1: adcl 16(%1), %0\n" + " lea 4(%1), %1\n" + " decl %2\n" + " jne 1b\n" + " adcl $0, %0\n" + " movl %0, %2\n" + " shrl $16, %0\n" + " addw %w2, %w0\n" + " adcl $0, %0\n" + " notl %0\n" + "2:" + /* Since the input registers which are loaded with iph and ipl + are modified, we must also specify them as outputs, or gcc + will assume they contain their original values. */ + : "=r" (sum), "=r" (iph), "=r" (ihl) + : "1" (iph), "2" (ihl) + : "memory"); + return(sum); +} + +static inline unsigned add32_with_carry(unsigned a, unsigned b) +{ + asm("addl %2,%0\n\t" + "adcl $0,%0" + : "=r" (a) + : "0" (a), "r" (b)); + return a; +} + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-x86_64/ptrace.h b/arch/um/include/sysdep-x86_64/ptrace.h new file mode 100644 index 000000000000..915c82daffbd --- /dev/null +++ b/arch/um/include/sysdep-x86_64/ptrace.h @@ -0,0 +1,264 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ + +#ifndef __SYSDEP_X86_64_PTRACE_H +#define __SYSDEP_X86_64_PTRACE_H + +#include "uml-config.h" +#include "user_constants.h" + +#define MAX_REG_OFFSET (UM_FRAME_SIZE) +#define MAX_REG_NR ((MAX_REG_OFFSET) / sizeof(unsigned long)) + +#ifdef UML_CONFIG_MODE_TT +#include "sysdep/sc.h" +#endif + +#ifdef UML_CONFIG_MODE_SKAS +#include "skas_ptregs.h" + +#define REGS_IP(r) ((r)[HOST_IP]) +#define REGS_SP(r) ((r)[HOST_SP]) + +#define REGS_RBX(r) ((r)[HOST_RBX]) +#define REGS_RCX(r) ((r)[HOST_RCX]) +#define REGS_RDX(r) ((r)[HOST_RDX]) +#define REGS_RSI(r) ((r)[HOST_RSI]) +#define REGS_RDI(r) ((r)[HOST_RDI]) +#define REGS_RBP(r) ((r)[HOST_RBP]) +#define REGS_RAX(r) ((r)[HOST_RAX]) +#define REGS_R8(r) ((r)[HOST_R8]) +#define REGS_R9(r) ((r)[HOST_R9]) +#define REGS_R10(r) ((r)[HOST_R10]) +#define REGS_R11(r) ((r)[HOST_R11]) +#define REGS_R12(r) ((r)[HOST_R12]) +#define REGS_R13(r) ((r)[HOST_R13]) +#define REGS_R14(r) ((r)[HOST_R14]) +#define REGS_R15(r) ((r)[HOST_R15]) +#define REGS_CS(r) ((r)[HOST_CS]) +#define REGS_EFLAGS(r) ((r)[HOST_EFLAGS]) +#define REGS_SS(r) ((r)[HOST_SS]) + +#define HOST_FS_BASE 21 +#define HOST_GS_BASE 22 +#define HOST_DS 23 +#define HOST_ES 24 +#define HOST_FS 25 +#define HOST_GS 26 + +#define REGS_FS_BASE(r) ((r)[HOST_FS_BASE]) +#define REGS_GS_BASE(r) ((r)[HOST_GS_BASE]) +#define REGS_DS(r) ((r)[HOST_DS]) +#define REGS_ES(r) ((r)[HOST_ES]) +#define REGS_FS(r) ((r)[HOST_FS]) +#define REGS_GS(r) ((r)[HOST_GS]) + +#define REGS_ORIG_RAX(r) ((r)[HOST_ORIG_RAX]) + +#define REGS_SET_SYSCALL_RETURN(r, res) REGS_RAX(r) = (res) + +#define REGS_RESTART_SYSCALL(r) IP_RESTART_SYSCALL(REGS_IP(r)) + +#define REGS_SEGV_IS_FIXABLE(r) SEGV_IS_FIXABLE((r)->trap_type) + +#define REGS_FAULT_ADDR(r) ((r)->fault_addr) + +#define REGS_FAULT_WRITE(r) FAULT_WRITE((r)->fault_type) + +#define REGS_TRAP(r) ((r)->trap_type) + +#define REGS_ERR(r) ((r)->fault_type) + +#endif + +#include "choose-mode.h" + +/* XXX */ +union uml_pt_regs { +#ifdef UML_CONFIG_MODE_TT + struct tt_regs { + long syscall; + unsigned long orig_rax; + void *sc; + } tt; +#endif +#ifdef UML_CONFIG_MODE_SKAS + struct skas_regs { + /* XXX */ + unsigned long regs[27]; + unsigned long fp[65]; + unsigned long fault_addr; + unsigned long fault_type; + unsigned long trap_type; + long syscall; + int is_user; + } skas; +#endif +}; + +#define EMPTY_UML_PT_REGS { } + +/* XXX */ +extern int mode_tt; + +#define UPT_RBX(r) __CHOOSE_MODE(SC_RBX(UPT_SC(r)), REGS_RBX((r)->skas.regs)) +#define UPT_RCX(r) __CHOOSE_MODE(SC_RCX(UPT_SC(r)), REGS_RCX((r)->skas.regs)) +#define UPT_RDX(r) __CHOOSE_MODE(SC_RDX(UPT_SC(r)), REGS_RDX((r)->skas.regs)) +#define UPT_RSI(r) __CHOOSE_MODE(SC_RSI(UPT_SC(r)), REGS_RSI((r)->skas.regs)) +#define UPT_RDI(r) __CHOOSE_MODE(SC_RDI(UPT_SC(r)), REGS_RDI((r)->skas.regs)) +#define UPT_RBP(r) __CHOOSE_MODE(SC_RBP(UPT_SC(r)), REGS_RBP((r)->skas.regs)) +#define UPT_RAX(r) __CHOOSE_MODE(SC_RAX(UPT_SC(r)), REGS_RAX((r)->skas.regs)) +#define UPT_R8(r) __CHOOSE_MODE(SC_R8(UPT_SC(r)), REGS_R8((r)->skas.regs)) +#define UPT_R9(r) __CHOOSE_MODE(SC_R9(UPT_SC(r)), REGS_R9((r)->skas.regs)) +#define UPT_R10(r) __CHOOSE_MODE(SC_R10(UPT_SC(r)), REGS_R10((r)->skas.regs)) +#define UPT_R11(r) __CHOOSE_MODE(SC_R11(UPT_SC(r)), REGS_R11((r)->skas.regs)) +#define UPT_R12(r) __CHOOSE_MODE(SC_R12(UPT_SC(r)), REGS_R12((r)->skas.regs)) +#define UPT_R13(r) __CHOOSE_MODE(SC_R13(UPT_SC(r)), REGS_R13((r)->skas.regs)) +#define UPT_R14(r) __CHOOSE_MODE(SC_R14(UPT_SC(r)), REGS_R14((r)->skas.regs)) +#define UPT_R15(r) __CHOOSE_MODE(SC_R15(UPT_SC(r)), REGS_R15((r)->skas.regs)) +#define UPT_CS(r) __CHOOSE_MODE(SC_CS(UPT_SC(r)), REGS_CS((r)->skas.regs)) +#define UPT_FS(r) __CHOOSE_MODE(SC_FS(UPT_SC(r)), REGS_FS((r)->skas.regs)) +#define UPT_GS(r) __CHOOSE_MODE(SC_GS(UPT_SC(r)), REGS_GS((r)->skas.regs)) +#define UPT_DS(r) __CHOOSE_MODE(SC_DS(UPT_SC(r)), REGS_DS((r)->skas.regs)) +#define UPT_ES(r) __CHOOSE_MODE(SC_ES(UPT_SC(r)), REGS_ES((r)->skas.regs)) +#define UPT_CS(r) __CHOOSE_MODE(SC_CS(UPT_SC(r)), REGS_CS((r)->skas.regs)) +#define UPT_ORIG_RAX(r) \ + __CHOOSE_MODE((r)->tt.orig_rax, REGS_ORIG_RAX((r)->skas.regs)) + +#define UPT_IP(r) __CHOOSE_MODE(SC_IP(UPT_SC(r)), REGS_IP((r)->skas.regs)) +#define UPT_SP(r) __CHOOSE_MODE(SC_SP(UPT_SC(r)), REGS_SP((r)->skas.regs)) + +#define UPT_EFLAGS(r) \ + __CHOOSE_MODE(SC_EFLAGS(UPT_SC(r)), REGS_EFLAGS((r)->skas.regs)) +#define UPT_SC(r) ((r)->tt.sc) +#define UPT_SYSCALL_NR(r) __CHOOSE_MODE((r)->tt.syscall, (r)->skas.syscall) + +extern int user_context(unsigned long sp); + +#define UPT_IS_USER(r) \ + CHOOSE_MODE(user_context(UPT_SP(r)), (r)->skas.is_user) + +#define UPT_SYSCALL_ARG1(r) UPT_RDI(r) +#define UPT_SYSCALL_ARG2(r) UPT_RSI(r) +#define UPT_SYSCALL_ARG3(r) UPT_RDX(r) +#define UPT_SYSCALL_ARG4(r) UPT_R10(r) +#define UPT_SYSCALL_ARG5(r) UPT_R8(r) +#define UPT_SYSCALL_ARG6(r) UPT_R9(r) + +struct syscall_args { + unsigned long args[6]; +}; + +#define SYSCALL_ARGS(r) ((struct syscall_args) \ + { .args = { UPT_SYSCALL_ARG1(r), \ + UPT_SYSCALL_ARG2(r), \ + UPT_SYSCALL_ARG3(r), \ + UPT_SYSCALL_ARG4(r), \ + UPT_SYSCALL_ARG5(r), \ + UPT_SYSCALL_ARG6(r) } } ) + +#define UPT_REG(regs, reg) \ + ({ unsigned long val; \ + switch(reg){ \ + case R8: val = UPT_R8(regs); break; \ + case R9: val = UPT_R9(regs); break; \ + case R10: val = UPT_R10(regs); break; \ + case R11: val = UPT_R11(regs); break; \ + case R12: val = UPT_R12(regs); break; \ + case R13: val = UPT_R13(regs); break; \ + case R14: val = UPT_R14(regs); break; \ + case R15: val = UPT_R15(regs); break; \ + case RIP: val = UPT_IP(regs); break; \ + case RSP: val = UPT_SP(regs); break; \ + case RAX: val = UPT_RAX(regs); break; \ + case RBX: val = UPT_RBX(regs); break; \ + case RCX: val = UPT_RCX(regs); break; \ + case RDX: val = UPT_RDX(regs); break; \ + case RSI: val = UPT_RSI(regs); break; \ + case RDI: val = UPT_RDI(regs); break; \ + case RBP: val = UPT_RBP(regs); break; \ + case ORIG_RAX: val = UPT_ORIG_RAX(regs); break; \ + case CS: val = UPT_CS(regs); break; \ + case DS: val = UPT_DS(regs); break; \ + case ES: val = UPT_ES(regs); break; \ + case FS: val = UPT_FS(regs); break; \ + case GS: val = UPT_GS(regs); break; \ + case EFLAGS: val = UPT_EFLAGS(regs); break; \ + default : \ + panic("Bad register in UPT_REG : %d\n", reg); \ + val = -1; \ + } \ + val; \ + }) + + +#define UPT_SET(regs, reg, val) \ + ({ unsigned long val; \ + switch(reg){ \ + case R8: UPT_R8(regs) = val; break; \ + case R9: UPT_R9(regs) = val; break; \ + case R10: UPT_R10(regs) = val; break; \ + case R11: UPT_R11(regs) = val; break; \ + case R12: UPT_R12(regs) = val; break; \ + case R13: UPT_R13(regs) = val; break; \ + case R14: UPT_R14(regs) = val; break; \ + case R15: UPT_R15(regs) = val; break; \ + case RIP: UPT_IP(regs) = val; break; \ + case RSP: UPT_SP(regs) = val; break; \ + case RAX: UPT_RAX(regs) = val; break; \ + case RBX: UPT_RBX(regs) = val; break; \ + case RCX: UPT_RCX(regs) = val; break; \ + case RDX: UPT_RDX(regs) = val; break; \ + case RSI: UPT_RSI(regs) = val; break; \ + case RDI: UPT_RDI(regs) = val; break; \ + case RBP: UPT_RBP(regs) = val; break; \ + case ORIG_RAX: UPT_ORIG_RAX(regs) = val; break; \ + case CS: UPT_CS(regs) = val; break; \ + case DS: UPT_DS(regs) = val; break; \ + case ES: UPT_ES(regs) = val; break; \ + case FS: UPT_FS(regs) = val; break; \ + case GS: UPT_GS(regs) = val; break; \ + case EFLAGS: UPT_EFLAGS(regs) = val; break; \ + default : \ + panic("Bad register in UPT_SET : %d\n", reg); \ + break; \ + } \ + val; \ + }) + +#define UPT_SET_SYSCALL_RETURN(r, res) \ + CHOOSE_MODE(SC_SET_SYSCALL_RETURN(UPT_SC(r), (res)), \ + REGS_SET_SYSCALL_RETURN((r)->skas.regs, (res))) + +#define UPT_RESTART_SYSCALL(r) \ + CHOOSE_MODE(SC_RESTART_SYSCALL(UPT_SC(r)), \ + REGS_RESTART_SYSCALL((r)->skas.regs)) + +#define UPT_SEGV_IS_FIXABLE(r) \ + CHOOSE_MODE(SC_SEGV_IS_FIXABLE(UPT_SC(r)), \ + REGS_SEGV_IS_FIXABLE(&r->skas)) + +#define UPT_FAULT_ADDR(r) \ + __CHOOSE_MODE(SC_FAULT_ADDR(UPT_SC(r)), REGS_FAULT_ADDR(&r->skas)) + +#define UPT_FAULT_WRITE(r) \ + CHOOSE_MODE(SC_FAULT_WRITE(UPT_SC(r)), REGS_FAULT_WRITE(&r->skas)) + +#define UPT_TRAP(r) __CHOOSE_MODE(SC_TRAP_TYPE(UPT_SC(r)), REGS_TRAP(&r->skas)) +#define UPT_ERR(r) __CHOOSE_MODE(SC_FAULT_TYPE(UPT_SC(r)), REGS_ERR(&r->skas)) + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-x86_64/ptrace_user.h b/arch/um/include/sysdep-x86_64/ptrace_user.h new file mode 100644 index 000000000000..31729973fb14 --- /dev/null +++ b/arch/um/include/sysdep-x86_64/ptrace_user.h @@ -0,0 +1,69 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ + +#ifndef __SYSDEP_X86_64_PTRACE_USER_H__ +#define __SYSDEP_X86_64_PTRACE_USER_H__ + +#define __FRAME_OFFSETS +#include <sys/ptrace.h> +#include <linux/ptrace.h> +#include <asm/ptrace.h> +#undef __FRAME_OFFSETS + +#define PT_INDEX(off) ((off) / sizeof(unsigned long)) + +#define PT_SYSCALL_NR(regs) ((regs)[PT_INDEX(ORIG_RAX)]) +#define PT_SYSCALL_NR_OFFSET (ORIG_RAX) + +#define PT_SYSCALL_ARG1(regs) (((unsigned long *) (regs))[PT_INDEX(RDI)]) +#define PT_SYSCALL_ARG1_OFFSET (RDI) + +#define PT_SYSCALL_ARG2(regs) (((unsigned long *) (regs))[PT_INDEX(RSI)]) +#define PT_SYSCALL_ARG2_OFFSET (RSI) + +#define PT_SYSCALL_ARG3(regs) (((unsigned long *) (regs))[PT_INDEX(RDX)]) +#define PT_SYSCALL_ARG3_OFFSET (RDX) + +#define PT_SYSCALL_ARG4(regs) (((unsigned long *) (regs))[PT_INDEX(RCX)]) +#define PT_SYSCALL_ARG4_OFFSET (RCX) + +#define PT_SYSCALL_ARG5(regs) (((unsigned long *) (regs))[PT_INDEX(R8)]) +#define PT_SYSCALL_ARG5_OFFSET (R8) + +#define PT_SYSCALL_ARG6(regs) (((unsigned long *) (regs))[PT_INDEX(R9)]) +#define PT_SYSCALL_ARG6_OFFSET (R9) + +#define PT_SYSCALL_RET_OFFSET (RAX) + +#define PT_IP_OFFSET (RIP) +#define PT_IP(regs) ((regs)[PT_INDEX(RIP)]) + +#define PT_SP_OFFSET (RSP) +#define PT_SP(regs) ((regs)[PT_INDEX(RSP)]) + +#define PT_ORIG_RAX_OFFSET (ORIG_RAX) +#define PT_ORIG_RAX(regs) ((regs)[PT_INDEX(ORIG_RAX)]) + +/* x86_64 FC3 doesn't define this in /usr/include/linux/ptrace.h even though + * it's defined in the kernel's include/linux/ptrace.h. Additionally, use the + * 2.4 name and value for 2.4 host compatibility. + */ +#ifndef PTRACE_OLDSETOPTIONS +#define PTRACE_OLDSETOPTIONS 21 +#endif + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-x86_64/sigcontext.h b/arch/um/include/sysdep-x86_64/sigcontext.h new file mode 100644 index 000000000000..1e38a54ff4cf --- /dev/null +++ b/arch/um/include/sysdep-x86_64/sigcontext.h @@ -0,0 +1,49 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ + +#ifndef __SYSDEP_X86_64_SIGCONTEXT_H +#define __SYSDEP_X86_64_SIGCONTEXT_H + +#include <sysdep/sc.h> + +#define IP_RESTART_SYSCALL(ip) ((ip) -= 2) + +#define SC_RESTART_SYSCALL(sc) IP_RESTART_SYSCALL(SC_IP(sc)) +#define SC_SET_SYSCALL_RETURN(sc, result) SC_RAX(sc) = (result) + +#define SC_FAULT_ADDR(sc) SC_CR2(sc) +#define SC_FAULT_TYPE(sc) SC_ERR(sc) + +#define FAULT_WRITE(err) ((err) & 2) + +#define SC_FAULT_WRITE(sc) FAULT_WRITE(SC_FAULT_TYPE(sc)) + +#define SC_TRAP_TYPE(sc) SC_TRAPNO(sc) + +/* ptrace expects that, at the start of a system call, %eax contains + * -ENOSYS, so this makes it so. + */ + +#define SC_START_SYSCALL(sc) do SC_RAX(sc) = -ENOSYS; while(0) + +#define SEGV_IS_FIXABLE(trap) ((trap) == 14) +#define SC_SEGV_IS_FIXABLE(sc) SEGV_IS_FIXABLE(SC_TRAP_TYPE(sc)) + +extern unsigned long *sc_sigmask(void *sc_ptr); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ + diff --git a/arch/um/include/sysdep-x86_64/signal.h b/arch/um/include/sysdep-x86_64/signal.h new file mode 100644 index 000000000000..e5e52756fab4 --- /dev/null +++ b/arch/um/include/sysdep-x86_64/signal.h @@ -0,0 +1,27 @@ +/* + * Copyright (C) 2004 PathScale, Inc + * Licensed under the GPL + */ + +#ifndef __X86_64_SIGNAL_H_ +#define __X86_64_SIGNAL_H_ + +#define ARCH_GET_SIGCONTEXT(sc, sig_addr) \ + do { \ + struct ucontext *__uc; \ + asm("movq %%rdx, %0" : "=r" (__uc)); \ + sc = (struct sigcontext *) &__uc->uc_mcontext; \ + } while(0) + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysdep-x86_64/syscalls.h b/arch/um/include/sysdep-x86_64/syscalls.h new file mode 100644 index 000000000000..b187a4157ff3 --- /dev/null +++ b/arch/um/include/sysdep-x86_64/syscalls.h @@ -0,0 +1,91 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ + +#ifndef __SYSDEP_X86_64_SYSCALLS_H__ +#define __SYSDEP_X86_64_SYSCALLS_H__ + +#include <linux/msg.h> +#include <linux/shm.h> + +typedef long syscall_handler_t(void); + +extern syscall_handler_t *ia32_sys_call_table[]; + +#define EXECUTE_SYSCALL(syscall, regs) \ + (((long (*)(long, long, long, long, long, long)) \ + (*sys_call_table[syscall]))(UPT_SYSCALL_ARG1(®s->regs), \ + UPT_SYSCALL_ARG2(®s->regs), \ + UPT_SYSCALL_ARG3(®s->regs), \ + UPT_SYSCALL_ARG4(®s->regs), \ + UPT_SYSCALL_ARG5(®s->regs), \ + UPT_SYSCALL_ARG6(®s->regs))) + +extern long old_mmap(unsigned long addr, unsigned long len, + unsigned long prot, unsigned long flags, + unsigned long fd, unsigned long pgoff); +extern syscall_handler_t wrap_sys_shmat; +extern syscall_handler_t sys_modify_ldt; +extern syscall_handler_t sys_arch_prctl; + +#define ARCH_SYSCALLS \ + [ __NR_mmap ] = (syscall_handler_t *) old_mmap, \ + [ __NR_select ] = (syscall_handler_t *) sys_select, \ + [ __NR_mincore ] = (syscall_handler_t *) sys_mincore, \ + [ __NR_madvise ] = (syscall_handler_t *) sys_madvise, \ + [ __NR_shmget ] = (syscall_handler_t *) sys_shmget, \ + [ __NR_shmat ] = (syscall_handler_t *) wrap_sys_shmat, \ + [ __NR_shmctl ] = (syscall_handler_t *) sys_shmctl, \ + [ __NR_semop ] = (syscall_handler_t *) sys_semop, \ + [ __NR_semget ] = (syscall_handler_t *) sys_semget, \ + [ __NR_semctl ] = (syscall_handler_t *) sys_semctl, \ + [ __NR_shmdt ] = (syscall_handler_t *) sys_shmdt, \ + [ __NR_msgget ] = (syscall_handler_t *) sys_msgget, \ + [ __NR_msgsnd ] = (syscall_handler_t *) sys_msgsnd, \ + [ __NR_msgrcv ] = (syscall_handler_t *) sys_msgrcv, \ + [ __NR_msgctl ] = (syscall_handler_t *) sys_msgctl, \ + [ __NR_pivot_root ] = (syscall_handler_t *) sys_pivot_root, \ + [ __NR_tuxcall ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_security ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_epoll_ctl_old ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_epoll_wait_old ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_modify_ldt ] = (syscall_handler_t *) sys_modify_ldt, \ + [ __NR_arch_prctl ] = (syscall_handler_t *) sys_arch_prctl, \ + [ __NR_socket ] = (syscall_handler_t *) sys_socket, \ + [ __NR_connect ] = (syscall_handler_t *) sys_connect, \ + [ __NR_accept ] = (syscall_handler_t *) sys_accept, \ + [ __NR_recvfrom ] = (syscall_handler_t *) sys_recvfrom, \ + [ __NR_recvmsg ] = (syscall_handler_t *) sys_recvmsg, \ + [ __NR_sendmsg ] = (syscall_handler_t *) sys_sendmsg, \ + [ __NR_bind ] = (syscall_handler_t *) sys_bind, \ + [ __NR_listen ] = (syscall_handler_t *) sys_listen, \ + [ __NR_getsockname ] = (syscall_handler_t *) sys_getsockname, \ + [ __NR_getpeername ] = (syscall_handler_t *) sys_getpeername, \ + [ __NR_socketpair ] = (syscall_handler_t *) sys_socketpair, \ + [ __NR_sendto ] = (syscall_handler_t *) sys_sendto, \ + [ __NR_shutdown ] = (syscall_handler_t *) sys_shutdown, \ + [ __NR_setsockopt ] = (syscall_handler_t *) sys_setsockopt, \ + [ __NR_getsockopt ] = (syscall_handler_t *) sys_getsockopt, \ + [ __NR_iopl ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_set_thread_area ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_get_thread_area ] = (syscall_handler_t *) sys_ni_syscall, \ + [ __NR_semtimedop ] = (syscall_handler_t *) sys_semtimedop, \ + [ 251 ] = (syscall_handler_t *) sys_ni_syscall, + +#define LAST_ARCH_SYSCALL 251 +#define NR_syscalls 1024 + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/sysrq.h b/arch/um/include/sysrq.h new file mode 100644 index 000000000000..2ce9423460b3 --- /dev/null +++ b/arch/um/include/sysrq.h @@ -0,0 +1,6 @@ +#ifndef __UM_SYSRQ_H +#define __UM_SYSRQ_H + +extern void show_trace(unsigned long *stack); + +#endif diff --git a/arch/um/include/tempfile.h b/arch/um/include/tempfile.h new file mode 100644 index 000000000000..e36d9e0f5105 --- /dev/null +++ b/arch/um/include/tempfile.h @@ -0,0 +1,21 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __TEMPFILE_H__ +#define __TEMPFILE_H__ + +extern int make_tempfile(const char *template, char **tempname, int do_unlink); + +#endif +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/time_user.h b/arch/um/include/time_user.h new file mode 100644 index 000000000000..6793a2fcd0ae --- /dev/null +++ b/arch/um/include/time_user.h @@ -0,0 +1,18 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __TIME_USER_H__ +#define __TIME_USER_H__ + +extern void timer(void); +extern void switch_timers(int to_real); +extern void set_interval(int timer_type); +extern void idle_sleep(int secs); +extern void enable_timer(void); +extern void disable_timer(void); +extern unsigned long time_lock(void); +extern void time_unlock(unsigned long); + +#endif diff --git a/arch/um/include/tlb.h b/arch/um/include/tlb.h new file mode 100644 index 000000000000..da1097285b8c --- /dev/null +++ b/arch/um/include/tlb.h @@ -0,0 +1,67 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __TLB_H__ +#define __TLB_H__ + +#include "um_mmu.h" + +struct host_vm_op { + enum { MMAP, MUNMAP, MPROTECT } type; + union { + struct { + unsigned long addr; + unsigned long len; + unsigned int r:1; + unsigned int w:1; + unsigned int x:1; + int fd; + __u64 offset; + } mmap; + struct { + unsigned long addr; + unsigned long len; + } munmap; + struct { + unsigned long addr; + unsigned long len; + unsigned int r:1; + unsigned int w:1; + unsigned int x:1; + } mprotect; + } u; +}; + +extern void mprotect_kernel_vm(int w); +extern void force_flush_all(void); +extern void fix_range_common(struct mm_struct *mm, unsigned long start_addr, + unsigned long end_addr, int force, int data, + void (*do_ops)(int, struct host_vm_op *, int)); +extern int flush_tlb_kernel_range_common(unsigned long start, + unsigned long end); + +extern int add_mmap(unsigned long virt, unsigned long phys, unsigned long len, + int r, int w, int x, struct host_vm_op *ops, int index, + int last_filled, int data, + void (*do_ops)(int, struct host_vm_op *, int)); +extern int add_munmap(unsigned long addr, unsigned long len, + struct host_vm_op *ops, int index, int last_filled, + int data, void (*do_ops)(int, struct host_vm_op *, int)); +extern int add_mprotect(unsigned long addr, unsigned long len, int r, int w, + int x, struct host_vm_op *ops, int index, + int last_filled, int data, + void (*do_ops)(int, struct host_vm_op *, int)); +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/ubd_user.h b/arch/um/include/ubd_user.h new file mode 100644 index 000000000000..bb66517f0739 --- /dev/null +++ b/arch/um/include/ubd_user.h @@ -0,0 +1,26 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2001 RidgeRun, Inc (glonnon@ridgerun.com) + * Licensed under the GPL + */ + +#ifndef __UM_UBD_USER_H +#define __UM_UBD_USER_H + +extern void ignore_sigwinch_sig(void); +extern int start_io_thread(unsigned long sp, int *fds_out); +extern int io_thread(void *arg); +extern int kernel_fd; + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/um_mmu.h b/arch/um/include/um_mmu.h new file mode 100644 index 000000000000..0fa643238300 --- /dev/null +++ b/arch/um/include/um_mmu.h @@ -0,0 +1,40 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __ARCH_UM_MMU_H +#define __ARCH_UM_MMU_H + +#include "uml-config.h" +#include "choose-mode.h" + +#ifdef UML_CONFIG_MODE_TT +#include "mmu-tt.h" +#endif + +#ifdef UML_CONFIG_MODE_SKAS +#include "mmu-skas.h" +#endif + +typedef union mm_context { +#ifdef UML_CONFIG_MODE_TT + struct mmu_context_tt tt; +#endif +#ifdef UML_CONFIG_MODE_SKAS + struct mmu_context_skas skas; +#endif +} mm_context_t; + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/um_uaccess.h b/arch/um/include/um_uaccess.h new file mode 100644 index 000000000000..6e348cb6de24 --- /dev/null +++ b/arch/um/include/um_uaccess.h @@ -0,0 +1,125 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __ARCH_UM_UACCESS_H +#define __ARCH_UM_UACCESS_H + +#include "linux/config.h" +#include "choose-mode.h" + +#ifdef CONFIG_MODE_TT +#include "uaccess-tt.h" +#endif + +#ifdef CONFIG_MODE_SKAS +#include "uaccess-skas.h" +#endif + +#define access_ok(type, addr, size) \ + CHOOSE_MODE_PROC(access_ok_tt, access_ok_skas, type, addr, size) + +/* this function will go away soon - use access_ok() instead */ +static inline int __deprecated verify_area(int type, const void __user *addr, unsigned long size) +{ + return (CHOOSE_MODE_PROC(verify_area_tt, verify_area_skas, type, addr, + size)); +} + +static inline int copy_from_user(void *to, const void __user *from, int n) +{ + return(CHOOSE_MODE_PROC(copy_from_user_tt, copy_from_user_skas, to, + from, n)); +} + +static inline int copy_to_user(void __user *to, const void *from, int n) +{ + return(CHOOSE_MODE_PROC(copy_to_user_tt, copy_to_user_skas, to, + from, n)); +} + +/* + * strncpy_from_user: - Copy a NUL terminated string from userspace. + * @dst: Destination address, in kernel space. This buffer must be at + * least @count bytes long. + * @src: Source address, in user space. + * @count: Maximum number of bytes to copy, including the trailing NUL. + * + * Copies a NUL-terminated string from userspace to kernel space. + * + * On success, returns the length of the string (not including the trailing + * NUL). + * + * If access to userspace fails, returns -EFAULT (some data may have been + * copied). + * + * If @count is smaller than the length of the string, copies @count bytes + * and returns @count. + */ + +static inline int strncpy_from_user(char *dst, const char __user *src, int count) +{ + return(CHOOSE_MODE_PROC(strncpy_from_user_tt, strncpy_from_user_skas, + dst, src, count)); +} + +/* + * __clear_user: - Zero a block of memory in user space, with less checking. + * @to: Destination address, in user space. + * @n: Number of bytes to zero. + * + * Zero a block of memory in user space. Caller must check + * the specified block with access_ok() before calling this function. + * + * Returns number of bytes that could not be cleared. + * On success, this will be zero. + */ +static inline int __clear_user(void *mem, int len) +{ + return(CHOOSE_MODE_PROC(__clear_user_tt, __clear_user_skas, mem, len)); +} + +/* + * clear_user: - Zero a block of memory in user space. + * @to: Destination address, in user space. + * @n: Number of bytes to zero. + * + * Zero a block of memory in user space. + * + * Returns number of bytes that could not be cleared. + * On success, this will be zero. + */ +static inline int clear_user(void __user *mem, int len) +{ + return(CHOOSE_MODE_PROC(clear_user_tt, clear_user_skas, mem, len)); +} + +/* + * strlen_user: - Get the size of a string in user space. + * @str: The string to measure. + * @n: The maximum valid length + * + * Get the size of a NUL-terminated string in user space. + * + * Returns the size of the string INCLUDING the terminating NUL. + * On exception, returns 0. + * If the string is too long, returns a value greater than @n. + */ +static inline int strnlen_user(const void __user *str, long len) +{ + return(CHOOSE_MODE_PROC(strnlen_user_tt, strnlen_user_skas, str, len)); +} + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/umid.h b/arch/um/include/umid.h new file mode 100644 index 000000000000..11373c851f15 --- /dev/null +++ b/arch/um/include/umid.h @@ -0,0 +1,22 @@ +/* + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __UMID_H__ +#define __UMID_H__ + +extern int umid_file_name(char *name, char *buf, int len); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/uml_uaccess.h b/arch/um/include/uml_uaccess.h new file mode 100644 index 000000000000..f77eb6428453 --- /dev/null +++ b/arch/um/include/uml_uaccess.h @@ -0,0 +1,28 @@ +/* + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __UML_UACCESS_H__ +#define __UML_UACCESS_H__ + +extern int __do_copy_to_user(void *to, const void *from, int n, + void **fault_addr, void **fault_catcher); +extern unsigned long __do_user_copy(void *to, const void *from, int n, + void **fault_addr, void **fault_catcher, + void (*op)(void *to, const void *from, + int n), int *faulted_out); +void __do_copy(void *to, const void *from, int n); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/user.h b/arch/um/include/user.h new file mode 100644 index 000000000000..57ee9e261228 --- /dev/null +++ b/arch/um/include/user.h @@ -0,0 +1,32 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __USER_H__ +#define __USER_H__ + +extern void panic(const char *fmt, ...); +extern int printk(const char *fmt, ...); +extern void schedule(void); +extern void *um_kmalloc(int size); +extern void *um_kmalloc_atomic(int size); +extern void kfree(void *ptr); +extern int in_aton(char *str); +extern int open_gdb_chan(void); +extern int strlcpy(char *, const char *, int); +extern void *um_vmalloc(int size); +extern void vfree(void *ptr); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/include/user_util.h b/arch/um/include/user_util.h new file mode 100644 index 000000000000..103cd320386c --- /dev/null +++ b/arch/um/include/user_util.h @@ -0,0 +1,106 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __USER_UTIL_H__ +#define __USER_UTIL_H__ + +#include "sysdep/ptrace.h" + +#define CATCH_EINTR(expr) while ((errno = 0, ((expr) < 0)) && (errno == EINTR)) + +extern int mode_tt; + +extern int grantpt(int __fd); +extern int unlockpt(int __fd); +extern char *ptsname(int __fd); + +struct cpu_task { + int pid; + void *task; +}; + +extern struct cpu_task cpu_tasks[]; + +struct signal_info { + void (*handler)(int, union uml_pt_regs *); + int is_irq; +}; + +extern struct signal_info sig_info[]; + +extern unsigned long low_physmem; +extern unsigned long high_physmem; +extern unsigned long uml_physmem; +extern unsigned long uml_reserved; +extern unsigned long end_vm; +extern unsigned long start_vm; +extern unsigned long highmem; + +extern char host_info[]; + +extern char saved_command_line[]; +extern char command_line[]; + +extern char *tempdir; + +extern unsigned long _stext, _etext, _sdata, _edata, __bss_start, _end; +extern unsigned long _unprotected_end; +extern unsigned long brk_start; + +extern int pty_output_sigio; +extern int pty_close_sigio; + +extern void stop(void); +extern void stack_protections(unsigned long address); +extern void task_protections(unsigned long address); +extern int wait_for_stop(int pid, int sig, int cont_type, void *relay); +extern void *add_signal_handler(int sig, void (*handler)(int)); +extern int start_fork_tramp(void *arg, unsigned long temp_stack, + int clone_flags, int (*tramp)(void *)); +extern int linux_main(int argc, char **argv); +extern void set_cmdline(char *cmd); +extern void input_cb(void (*proc)(void *), void *arg, int arg_len); +extern int get_pty(void); +extern void *um_kmalloc(int size); +extern int switcheroo(int fd, int prot, void *from, void *to, int size); +extern void setup_machinename(char *machine_out); +extern void setup_hostinfo(void); +extern void add_arg(char *arg); +extern void init_new_thread_stack(void *sig_stack, void (*usr1_handler)(int)); +extern void init_new_thread_signals(int altstack); +extern void do_exec(int old_pid, int new_pid); +extern void tracer_panic(char *msg, ...); +extern char *get_umid(int only_if_set); +extern void do_longjmp(void *p, int val); +extern int detach(int pid, int sig); +extern int attach(int pid); +extern void kill_child_dead(int pid); +extern int cont(int pid); +extern void check_ptrace(void); +extern void check_sigio(void); +extern int run_kernel_thread(int (*fn)(void *), void *arg, void **jmp_ptr); +extern void write_sigio_workaround(void); +extern void arch_check_bugs(void); +extern int cpu_feature(char *what, char *buf, int len); +extern int arch_handle_signal(int sig, union uml_pt_regs *regs); +extern int arch_fixup(unsigned long address, void *sc_ptr); +extern void forward_pending_sigio(int target); +extern int can_do_skas(void); +extern void arch_init_thread(void); +extern int setjmp_wrapper(void (*proc)(void *, void *), ...); +extern int raw(int fd); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile new file mode 100644 index 000000000000..dc796c1bf39e --- /dev/null +++ b/arch/um/kernel/Makefile @@ -0,0 +1,58 @@ +# +# Copyright (C) 2002 Jeff Dike (jdike@karaya.com) +# Licensed under the GPL +# + +extra-y := vmlinux.lds +clean-files := vmlinux.lds.S config.tmp + +obj-y = checksum.o config.o exec_kern.o exitcode.o \ + helper.o init_task.o irq.o irq_user.o ksyms.o main.o mem.o mem_user.o \ + physmem.o process.o process_kern.o ptrace.o reboot.o resource.o \ + sigio_user.o sigio_kern.o signal_kern.o signal_user.o smp.o \ + syscall_kern.o sysrq.o sys_call_table.o tempfile.o time.o time_kern.o \ + tlb.o trap_kern.o trap_user.o uaccess_user.o um_arch.o umid.o \ + user_util.o + +obj-$(CONFIG_BLK_DEV_INITRD) += initrd_kern.o initrd_user.o +obj-$(CONFIG_GPROF) += gprof_syms.o +obj-$(CONFIG_GCOV) += gmon_syms.o +obj-$(CONFIG_TTY_LOG) += tty_log.o +obj-$(CONFIG_SYSCALL_DEBUG) += syscall_user.o + +obj-$(CONFIG_MODE_TT) += tt/ +obj-$(CONFIG_MODE_SKAS) += skas/ + +# This needs be compiled with frame pointers regardless of how the rest of the +# kernel is built. +CFLAGS_frame.o := -fno-omit-frame-pointer + +user-objs-$(CONFIG_TTY_LOG) += tty_log.o + +USER_OBJS := $(user-objs-y) config.o helper.o main.o process.o tempfile.o \ + time.o tty_log.o umid.o user_util.o frame.o + +include arch/um/scripts/Makefile.rules + +targets += config.c + +# Be careful with the below Sed code - sed is pitfall-rich! +# We use sed to lower build requirements, for "embedded" builders for instance. + +$(obj)/config.tmp: $(objtree)/.config FORCE + $(call if_changed,quote1) + +quiet_cmd_quote1 = QUOTE $@ + cmd_quote1 = sed -e 's/"/\\"/g' -e 's/^/"/' -e 's/$$/\\n"/' \ + $< > $@ + +$(obj)/config.c: $(src)/config.c.in $(obj)/config.tmp FORCE + $(call if_changed,quote2) + +quiet_cmd_quote2 = QUOTE $@ + cmd_quote2 = sed -e '/CONFIG/{' \ + -e 's/"CONFIG"\;/""/' \ + -e 'r $(obj)/config.tmp' \ + -e 'a""\;' \ + -e '}' \ + $< > $@ diff --git a/arch/um/kernel/checksum.c b/arch/um/kernel/checksum.c new file mode 100644 index 000000000000..e69b2be951d1 --- /dev/null +++ b/arch/um/kernel/checksum.c @@ -0,0 +1,36 @@ +#include "asm/uaccess.h" +#include "linux/errno.h" +#include "linux/module.h" + +unsigned int arch_csum_partial(const unsigned char *buff, int len, int sum); + +unsigned int csum_partial(unsigned char *buff, int len, int sum) +{ + return arch_csum_partial(buff, len, sum); +} + +EXPORT_SYMBOL(csum_partial); + +unsigned int csum_partial_copy_to(const unsigned char *src, + unsigned char __user *dst, int len, int sum, + int *err_ptr) +{ + if(copy_to_user(dst, src, len)){ + *err_ptr = -EFAULT; + return(-1); + } + + return(arch_csum_partial(src, len, sum)); +} + +unsigned int csum_partial_copy_from(const unsigned char __user *src, + unsigned char *dst, int len, int sum, + int *err_ptr) +{ + if(copy_from_user(dst, src, len)){ + *err_ptr = -EFAULT; + return(-1); + } + + return arch_csum_partial(dst, len, sum); +} diff --git a/arch/um/kernel/config.c.in b/arch/um/kernel/config.c.in new file mode 100644 index 000000000000..c062cbfe386e --- /dev/null +++ b/arch/um/kernel/config.c.in @@ -0,0 +1,32 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include "init.h" + +static __initdata char *config = "CONFIG"; + +static int __init print_config(char *line, int *add) +{ + printf("%s", config); + exit(0); +} + +__uml_setup("--showconfig", print_config, +"--showconfig\n" +" Prints the config file that this UML binary was generated from.\n\n" +); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/dyn.lds.S b/arch/um/kernel/dyn.lds.S new file mode 100644 index 000000000000..715b0838a68c --- /dev/null +++ b/arch/um/kernel/dyn.lds.S @@ -0,0 +1,176 @@ +#include <asm-generic/vmlinux.lds.h> + +OUTPUT_FORMAT(ELF_FORMAT) +OUTPUT_ARCH(ELF_ARCH) +ENTRY(_start) +jiffies = jiffies_64; + +SECTIONS +{ + PROVIDE (__executable_start = START); + . = START + SIZEOF_HEADERS; + .interp : { *(.interp) } + /* Used in arch/um/kernel/mem.c. Any memory between START and __binary_start + * is remapped.*/ + __binary_start = .; + . = ALIGN(4096); /* Init code and data */ + _stext = .; + __init_begin = .; + .init.text : { + _sinittext = .; + *(.init.text) + _einittext = .; + } + + . = ALIGN(4096); + + /* Read-only sections, merged into text segment: */ + .hash : { *(.hash) } + .dynsym : { *(.dynsym) } + .dynstr : { *(.dynstr) } + .gnu.version : { *(.gnu.version) } + .gnu.version_d : { *(.gnu.version_d) } + .gnu.version_r : { *(.gnu.version_r) } + .rel.init : { *(.rel.init) } + .rela.init : { *(.rela.init) } + .rel.text : { *(.rel.text .rel.text.* .rel.gnu.linkonce.t.*) } + .rela.text : { *(.rela.text .rela.text.* .rela.gnu.linkonce.t.*) } + .rel.fini : { *(.rel.fini) } + .rela.fini : { *(.rela.fini) } + .rel.rodata : { *(.rel.rodata .rel.rodata.* .rel.gnu.linkonce.r.*) } + .rela.rodata : { *(.rela.rodata .rela.rodata.* .rela.gnu.linkonce.r.*) } + .rel.data : { *(.rel.data .rel.data.* .rel.gnu.linkonce.d.*) } + .rela.data : { *(.rela.data .rela.data.* .rela.gnu.linkonce.d.*) } + .rel.tdata : { *(.rel.tdata .rel.tdata.* .rel.gnu.linkonce.td.*) } + .rela.tdata : { *(.rela.tdata .rela.tdata.* .rela.gnu.linkonce.td.*) } + .rel.tbss : { *(.rel.tbss .rel.tbss.* .rel.gnu.linkonce.tb.*) } + .rela.tbss : { *(.rela.tbss .rela.tbss.* .rela.gnu.linkonce.tb.*) } + .rel.ctors : { *(.rel.ctors) } + .rela.ctors : { *(.rela.ctors) } + .rel.dtors : { *(.rel.dtors) } + .rela.dtors : { *(.rela.dtors) } + .rel.got : { *(.rel.got) } + .rela.got : { *(.rela.got) } + .rel.bss : { *(.rel.bss .rel.bss.* .rel.gnu.linkonce.b.*) } + .rela.bss : { *(.rela.bss .rela.bss.* .rela.gnu.linkonce.b.*) } + .rel.plt : { *(.rel.plt) } + .rela.plt : { *(.rela.plt) } + .init : { + KEEP (*(.init)) + } =0x90909090 + .plt : { *(.plt) } + .text : { + *(.text) + SCHED_TEXT + LOCK_TEXT + *(.fixup) + *(.stub .text.* .gnu.linkonce.t.*) + /* .gnu.warning sections are handled specially by elf32.em. */ + *(.gnu.warning) + } =0x90909090 + .fini : { + KEEP (*(.fini)) + } =0x90909090 + + .kstrtab : { *(.kstrtab) } + + #include "asm/common.lds.S" + + init.data : { *(.init.data) } + + /* Ensure the __preinit_array_start label is properly aligned. We + could instead move the label definition inside the section, but + the linker would then create the section even if it turns out to + be empty, which isn't pretty. */ + . = ALIGN(32 / 8); + .preinit_array : { *(.preinit_array) } + .init_array : { *(.init_array) } + .fini_array : { *(.fini_array) } + .data : { + . = ALIGN(KERNEL_STACK_SIZE); /* init_task */ + *(.data.init_task) + *(.data .data.* .gnu.linkonce.d.*) + SORT(CONSTRUCTORS) + } + .data1 : { *(.data1) } + .tdata : { *(.tdata .tdata.* .gnu.linkonce.td.*) } + .tbss : { *(.tbss .tbss.* .gnu.linkonce.tb.*) *(.tcommon) } + .eh_frame : { KEEP (*(.eh_frame)) } + .gcc_except_table : { *(.gcc_except_table) } + .dynamic : { *(.dynamic) } + .ctors : { + /* gcc uses crtbegin.o to find the start of + the constructors, so we make sure it is + first. Because this is a wildcard, it + doesn't matter if the user does not + actually link against crtbegin.o; the + linker won't look for a file to match a + wildcard. The wildcard also means that it + doesn't matter which directory crtbegin.o + is in. */ + KEEP (*crtbegin.o(.ctors)) + /* We don't want to include the .ctor section from + from the crtend.o file until after the sorted ctors. + The .ctor section from the crtend file contains the + end of ctors marker and it must be last */ + KEEP (*(EXCLUDE_FILE (*crtend.o ) .ctors)) + KEEP (*(SORT(.ctors.*))) + KEEP (*(.ctors)) + } + .dtors : { + KEEP (*crtbegin.o(.dtors)) + KEEP (*(EXCLUDE_FILE (*crtend.o ) .dtors)) + KEEP (*(SORT(.dtors.*))) + KEEP (*(.dtors)) + } + .jcr : { KEEP (*(.jcr)) } + .got : { *(.got.plt) *(.got) } + _edata = .; + PROVIDE (edata = .); + __bss_start = .; + .bss : { + *(.dynbss) + *(.bss .bss.* .gnu.linkonce.b.*) + *(COMMON) + /* Align here to ensure that the .bss section occupies space up to + _end. Align after .bss to ensure correct alignment even if the + .bss section disappears because there are no input sections. */ + . = ALIGN(32 / 8); + . = ALIGN(32 / 8); + } + _end = .; + PROVIDE (end = .); + /* Stabs debugging sections. */ + .stab 0 : { *(.stab) } + .stabstr 0 : { *(.stabstr) } + .stab.excl 0 : { *(.stab.excl) } + .stab.exclstr 0 : { *(.stab.exclstr) } + .stab.index 0 : { *(.stab.index) } + .stab.indexstr 0 : { *(.stab.indexstr) } + .comment 0 : { *(.comment) } + /* DWARF debug sections. + Symbols in the DWARF debugging sections are relative to the beginning + of the section so we begin them at 0. */ + /* DWARF 1 */ + .debug 0 : { *(.debug) } + .line 0 : { *(.line) } + /* GNU DWARF 1 extensions */ + .debug_srcinfo 0 : { *(.debug_srcinfo) } + .debug_sfnames 0 : { *(.debug_sfnames) } + /* DWARF 1.1 and DWARF 2 */ + .debug_aranges 0 : { *(.debug_aranges) } + .debug_pubnames 0 : { *(.debug_pubnames) } + /* DWARF 2 */ + .debug_info 0 : { *(.debug_info .gnu.linkonce.wi.*) } + .debug_abbrev 0 : { *(.debug_abbrev) } + .debug_line 0 : { *(.debug_line) } + .debug_frame 0 : { *(.debug_frame) } + .debug_str 0 : { *(.debug_str) } + .debug_loc 0 : { *(.debug_loc) } + .debug_macinfo 0 : { *(.debug_macinfo) } + /* SGI/MIPS DWARF 2 extensions */ + .debug_weaknames 0 : { *(.debug_weaknames) } + .debug_funcnames 0 : { *(.debug_funcnames) } + .debug_typenames 0 : { *(.debug_typenames) } + .debug_varnames 0 : { *(.debug_varnames) } +} diff --git a/arch/um/kernel/exec_kern.c b/arch/um/kernel/exec_kern.c new file mode 100644 index 000000000000..49ddabe69be7 --- /dev/null +++ b/arch/um/kernel/exec_kern.c @@ -0,0 +1,91 @@ +/* + * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/slab.h" +#include "linux/smp_lock.h" +#include "linux/ptrace.h" +#include "asm/ptrace.h" +#include "asm/pgtable.h" +#include "asm/tlbflush.h" +#include "asm/uaccess.h" +#include "user_util.h" +#include "kern_util.h" +#include "mem_user.h" +#include "kern.h" +#include "irq_user.h" +#include "tlb.h" +#include "2_5compat.h" +#include "os.h" +#include "time_user.h" +#include "choose-mode.h" +#include "mode_kern.h" + +void flush_thread(void) +{ + CHOOSE_MODE(flush_thread_tt(), flush_thread_skas()); +} + +void start_thread(struct pt_regs *regs, unsigned long eip, unsigned long esp) +{ + CHOOSE_MODE_PROC(start_thread_tt, start_thread_skas, regs, eip, esp); +} + +extern void log_exec(char **argv, void *tty); + +static long execve1(char *file, char __user * __user *argv, + char *__user __user *env) +{ + long error; + +#ifdef CONFIG_TTY_LOG + log_exec(argv, current->tty); +#endif + error = do_execve(file, argv, env, ¤t->thread.regs); + if (error == 0){ + task_lock(current); + current->ptrace &= ~PT_DTRACE; + task_unlock(current); + set_cmdline(current_cmd()); + } + return(error); +} + +long um_execve(char *file, char __user *__user *argv, char __user *__user *env) +{ + long err; + + err = execve1(file, argv, env); + if(!err) + do_longjmp(current->thread.exec_buf, 1); + return(err); +} + +long sys_execve(char *file, char __user *__user *argv, + char __user *__user *env) +{ + long error; + char *filename; + + lock_kernel(); + filename = getname((char __user *) file); + error = PTR_ERR(filename); + if (IS_ERR(filename)) goto out; + error = execve1(filename, argv, env); + putname(filename); + out: + unlock_kernel(); + return(error); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/exitcode.c b/arch/um/kernel/exitcode.c new file mode 100644 index 000000000000..0ea87f24b36f --- /dev/null +++ b/arch/um/kernel/exitcode.c @@ -0,0 +1,73 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/init.h" +#include "linux/ctype.h" +#include "linux/proc_fs.h" +#include "asm/uaccess.h" + +/* If read and write race, the read will still atomically read a valid + * value. + */ +int uml_exitcode = 0; + +static int read_proc_exitcode(char *page, char **start, off_t off, + int count, int *eof, void *data) +{ + int len; + + len = sprintf(page, "%d\n", uml_exitcode); + len -= off; + if(len <= off+count) *eof = 1; + *start = page + off; + if(len > count) len = count; + if(len < 0) len = 0; + return(len); +} + +static int write_proc_exitcode(struct file *file, const char __user *buffer, + unsigned long count, void *data) +{ + char *end, buf[sizeof("nnnnn\0")]; + int tmp; + + if(copy_from_user(buf, buffer, count)) + return(-EFAULT); + tmp = simple_strtol(buf, &end, 0); + if((*end != '\0') && !isspace(*end)) + return(-EINVAL); + uml_exitcode = tmp; + return(count); +} + +static int make_proc_exitcode(void) +{ + struct proc_dir_entry *ent; + + ent = create_proc_entry("exitcode", 0600, &proc_root); + if(ent == NULL){ + printk("make_proc_exitcode : Failed to register " + "/proc/exitcode\n"); + return(0); + } + + ent->read_proc = read_proc_exitcode; + ent->write_proc = write_proc_exitcode; + + return(0); +} + +__initcall(make_proc_exitcode); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/gmon_syms.c b/arch/um/kernel/gmon_syms.c new file mode 100644 index 000000000000..2c86e7fdb014 --- /dev/null +++ b/arch/um/kernel/gmon_syms.c @@ -0,0 +1,34 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/module.h" + +extern void __bb_init_func(void *); +EXPORT_SYMBOL(__bb_init_func); + +/* This is defined (and referred to in profiling stub code) only by some GCC + * versions in libgcov. + * + * Since SuSE backported the fix, we cannot handle it depending on GCC version. + * So, unconditinally export it. But also give it a weak declaration, which will + * be overriden by any other one. + */ + +extern void __gcov_init(void *) __attribute__((weak)); +EXPORT_SYMBOL(__gcov_init); + +extern void __gcov_merge_add(void *) __attribute__((weak)); +EXPORT_SYMBOL(__gcov_merge_add); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/gprof_syms.c b/arch/um/kernel/gprof_syms.c new file mode 100644 index 000000000000..9244f018d44c --- /dev/null +++ b/arch/um/kernel/gprof_syms.c @@ -0,0 +1,20 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/module.h" + +extern void mcount(void); +EXPORT_SYMBOL(mcount); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/helper.c b/arch/um/kernel/helper.c new file mode 100644 index 000000000000..13b1f5c2f7ee --- /dev/null +++ b/arch/um/kernel/helper.c @@ -0,0 +1,173 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <errno.h> +#include <sched.h> +#include <sys/signal.h> +#include <sys/wait.h> +#include "user.h" +#include "kern_util.h" +#include "user_util.h" +#include "os.h" + +struct helper_data { + void (*pre_exec)(void*); + void *pre_data; + char **argv; + int fd; +}; + +/* Debugging aid, changed only from gdb */ +int helper_pause = 0; + +static void helper_hup(int sig) +{ +} + +static int helper_child(void *arg) +{ + struct helper_data *data = arg; + char **argv = data->argv; + int errval; + + if(helper_pause){ + signal(SIGHUP, helper_hup); + pause(); + } + if(data->pre_exec != NULL) + (*data->pre_exec)(data->pre_data); + execvp(argv[0], argv); + errval = errno; + printk("execvp of '%s' failed - errno = %d\n", argv[0], errno); + os_write_file(data->fd, &errval, sizeof(errval)); + os_kill_process(os_getpid(), 0); + return(0); +} + +/* Returns either the pid of the child process we run or -E* on failure. + * XXX The alloc_stack here breaks if this is called in the tracing thread */ +int run_helper(void (*pre_exec)(void *), void *pre_data, char **argv, + unsigned long *stack_out) +{ + struct helper_data data; + unsigned long stack, sp; + int pid, fds[2], ret, n; + + if((stack_out != NULL) && (*stack_out != 0)) + stack = *stack_out; + else stack = alloc_stack(0, um_in_interrupt()); + if(stack == 0) + return(-ENOMEM); + + ret = os_pipe(fds, 1, 0); + if(ret < 0){ + printk("run_helper : pipe failed, ret = %d\n", -ret); + goto out_free; + } + + ret = os_set_exec_close(fds[1], 1); + if(ret < 0){ + printk("run_helper : setting FD_CLOEXEC failed, ret = %d\n", + -ret); + goto out_close; + } + + sp = stack + page_size() - sizeof(void *); + data.pre_exec = pre_exec; + data.pre_data = pre_data; + data.argv = argv; + data.fd = fds[1]; + pid = clone(helper_child, (void *) sp, CLONE_VM | SIGCHLD, &data); + if(pid < 0){ + printk("run_helper : clone failed, errno = %d\n", errno); + ret = -errno; + goto out_close; + } + + os_close_file(fds[1]); + fds[1] = -1; + + /*Read the errno value from the child.*/ + n = os_read_file(fds[0], &ret, sizeof(ret)); + if(n < 0){ + printk("run_helper : read on pipe failed, ret = %d\n", -n); + ret = n; + os_kill_process(pid, 1); + } + else if(n != 0){ + CATCH_EINTR(n = waitpid(pid, NULL, 0)); + ret = -errno; + } else { + ret = pid; + } + +out_close: + if (fds[1] != -1) + os_close_file(fds[1]); + os_close_file(fds[0]); +out_free: + if(stack_out == NULL) + free_stack(stack, 0); + else *stack_out = stack; + return(ret); +} + +int run_helper_thread(int (*proc)(void *), void *arg, unsigned int flags, + unsigned long *stack_out, int stack_order) +{ + unsigned long stack, sp; + int pid, status; + + stack = alloc_stack(stack_order, um_in_interrupt()); + if(stack == 0) return(-ENOMEM); + + sp = stack + (page_size() << stack_order) - sizeof(void *); + pid = clone(proc, (void *) sp, flags | SIGCHLD, arg); + if(pid < 0){ + printk("run_helper_thread : clone failed, errno = %d\n", + errno); + return(-errno); + } + if(stack_out == NULL){ + CATCH_EINTR(pid = waitpid(pid, &status, 0)); + if(pid < 0){ + printk("run_helper_thread - wait failed, errno = %d\n", + errno); + pid = -errno; + } + if(!WIFEXITED(status) || (WEXITSTATUS(status) != 0)) + printk("run_helper_thread - thread returned status " + "0x%x\n", status); + free_stack(stack, stack_order); + } + else *stack_out = stack; + return(pid); +} + +int helper_wait(int pid, int block) +{ + int ret; + + CATCH_EINTR(ret = waitpid(pid, NULL, WNOHANG)); + if(ret < 0){ + printk("helper_wait : waitpid failed, errno = %d\n", errno); + return(-errno); + } + return(ret); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/init_task.c b/arch/um/kernel/init_task.c new file mode 100644 index 000000000000..cd7c85be0a1b --- /dev/null +++ b/arch/um/kernel/init_task.c @@ -0,0 +1,61 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/mm.h" +#include "linux/module.h" +#include "linux/sched.h" +#include "linux/init_task.h" +#include "linux/mqueue.h" +#include "asm/uaccess.h" +#include "asm/pgtable.h" +#include "user_util.h" +#include "mem_user.h" + +static struct fs_struct init_fs = INIT_FS; +struct mm_struct init_mm = INIT_MM(init_mm); +static struct files_struct init_files = INIT_FILES; +static struct signal_struct init_signals = INIT_SIGNALS(init_signals); +static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand); +EXPORT_SYMBOL(init_mm); + +/* + * Initial task structure. + * + * All other task structs will be allocated on slabs in fork.c + */ + +struct task_struct init_task = INIT_TASK(init_task); + +EXPORT_SYMBOL(init_task); + +/* + * Initial thread structure. + * + * We need to make sure that this is 16384-byte aligned due to the + * way process stacks are handled. This is done by having a special + * "init_task" linker map entry.. + */ + +union thread_union init_thread_union +__attribute__((__section__(".data.init_task"))) = +{ INIT_THREAD_INFO(init_task) }; + +void unprotect_stack(unsigned long stack) +{ + protect_memory(stack, (1 << CONFIG_KERNEL_STACK_ORDER) * PAGE_SIZE, + 1, 1, 0, 1); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/initrd_kern.c b/arch/um/kernel/initrd_kern.c new file mode 100644 index 000000000000..fc568af468b9 --- /dev/null +++ b/arch/um/kernel/initrd_kern.c @@ -0,0 +1,59 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/init.h" +#include "linux/bootmem.h" +#include "linux/initrd.h" +#include "asm/types.h" +#include "user_util.h" +#include "kern_util.h" +#include "initrd.h" +#include "init.h" +#include "os.h" + +/* Changed by uml_initrd_setup, which is a setup */ +static char *initrd __initdata = NULL; + +static int __init read_initrd(void) +{ + void *area; + long long size; + int err; + + if(initrd == NULL) return 0; + err = os_file_size(initrd, &size); + if(err) return 0; + area = alloc_bootmem(size); + if(area == NULL) return 0; + if(load_initrd(initrd, area, size) == -1) return 0; + initrd_start = (unsigned long) area; + initrd_end = initrd_start + size; + return 0; +} + +__uml_postsetup(read_initrd); + +static int __init uml_initrd_setup(char *line, int *add) +{ + initrd = line; + return 0; +} + +__uml_setup("initrd=", uml_initrd_setup, +"initrd=<initrd image>\n" +" This is used to boot UML from an initrd image. The argument is the\n" +" name of the file containing the image.\n\n" +); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/initrd_user.c b/arch/um/kernel/initrd_user.c new file mode 100644 index 000000000000..cb90681e151c --- /dev/null +++ b/arch/um/kernel/initrd_user.c @@ -0,0 +1,46 @@ +/* + * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <unistd.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <errno.h> + +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "initrd.h" +#include "os.h" + +int load_initrd(char *filename, void *buf, int size) +{ + int fd, n; + + fd = os_open_file(filename, of_read(OPENFLAGS()), 0); + if(fd < 0){ + printk("Opening '%s' failed - err = %d\n", filename, -fd); + return(-1); + } + n = os_read_file(fd, buf, size); + if(n != size){ + printk("Read of %d bytes from '%s' failed, err = %d\n", size, + filename, -n); + return(-1); + } + + os_close_file(fd); + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c new file mode 100644 index 000000000000..d71e8f00810f --- /dev/null +++ b/arch/um/kernel/irq.c @@ -0,0 +1,178 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + * Derived (i.e. mostly copied) from arch/i386/kernel/irq.c: + * Copyright (C) 1992, 1998 Linus Torvalds, Ingo Molnar + */ + +#include "linux/config.h" +#include "linux/kernel.h" +#include "linux/module.h" +#include "linux/smp.h" +#include "linux/irq.h" +#include "linux/kernel_stat.h" +#include "linux/interrupt.h" +#include "linux/random.h" +#include "linux/slab.h" +#include "linux/file.h" +#include "linux/proc_fs.h" +#include "linux/init.h" +#include "linux/seq_file.h" +#include "linux/profile.h" +#include "linux/hardirq.h" +#include "asm/irq.h" +#include "asm/hw_irq.h" +#include "asm/atomic.h" +#include "asm/signal.h" +#include "asm/system.h" +#include "asm/errno.h" +#include "asm/uaccess.h" +#include "user_util.h" +#include "kern_util.h" +#include "irq_user.h" +#include "irq_kern.h" + + +/* + * Generic, controller-independent functions: + */ + +int show_interrupts(struct seq_file *p, void *v) +{ + int i = *(loff_t *) v, j; + struct irqaction * action; + unsigned long flags; + + if (i == 0) { + seq_printf(p, " "); + for_each_online_cpu(j) + seq_printf(p, "CPU%d ",j); + seq_putc(p, '\n'); + } + + if (i < NR_IRQS) { + spin_lock_irqsave(&irq_desc[i].lock, flags); + action = irq_desc[i].action; + if (!action) + goto skip; + seq_printf(p, "%3d: ",i); +#ifndef CONFIG_SMP + seq_printf(p, "%10u ", kstat_irqs(i)); +#else + for_each_online_cpu(j) + seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]); +#endif + seq_printf(p, " %14s", irq_desc[i].handler->typename); + seq_printf(p, " %s", action->name); + + for (action=action->next; action; action = action->next) + seq_printf(p, ", %s", action->name); + + seq_putc(p, '\n'); +skip: + spin_unlock_irqrestore(&irq_desc[i].lock, flags); + } else if (i == NR_IRQS) { + seq_putc(p, '\n'); + } + + return 0; +} + +/* + * do_IRQ handles all normal device IRQ's (the special + * SMP cross-CPU interrupts have their own specific + * handlers). + */ +unsigned int do_IRQ(int irq, union uml_pt_regs *regs) +{ + irq_enter(); + __do_IRQ(irq, (struct pt_regs *) regs); + irq_exit(); + return 1; +} + +int um_request_irq(unsigned int irq, int fd, int type, + irqreturn_t (*handler)(int, void *, struct pt_regs *), + unsigned long irqflags, const char * devname, + void *dev_id) +{ + int err; + + err = request_irq(irq, handler, irqflags, devname, dev_id); + if(err) + return(err); + + if(fd != -1) + err = activate_fd(irq, fd, type, dev_id); + return(err); +} +EXPORT_SYMBOL(um_request_irq); +EXPORT_SYMBOL(reactivate_fd); + +static DEFINE_SPINLOCK(irq_spinlock); + +unsigned long irq_lock(void) +{ + unsigned long flags; + + spin_lock_irqsave(&irq_spinlock, flags); + return(flags); +} + +void irq_unlock(unsigned long flags) +{ + spin_unlock_irqrestore(&irq_spinlock, flags); +} + +/* presently hw_interrupt_type must define (startup || enable) && + * disable && end */ +static void dummy(unsigned int irq) +{ +} + +static struct hw_interrupt_type SIGIO_irq_type = { + .typename = "SIGIO", + .disable = dummy, + .enable = dummy, + .ack = dummy, + .end = dummy +}; + +static struct hw_interrupt_type SIGVTALRM_irq_type = { + .typename = "SIGVTALRM", + .shutdown = dummy, /* never called */ + .disable = dummy, + .enable = dummy, + .ack = dummy, + .end = dummy +}; + +void __init init_IRQ(void) +{ + int i; + + irq_desc[TIMER_IRQ].status = IRQ_DISABLED; + irq_desc[TIMER_IRQ].action = NULL; + irq_desc[TIMER_IRQ].depth = 1; + irq_desc[TIMER_IRQ].handler = &SIGVTALRM_irq_type; + enable_irq(TIMER_IRQ); + for(i=1;i<NR_IRQS;i++){ + irq_desc[i].status = IRQ_DISABLED; + irq_desc[i].action = NULL; + irq_desc[i].depth = 1; + irq_desc[i].handler = &SIGIO_irq_type; + enable_irq(i); + } + init_irq_signals(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/irq_user.c b/arch/um/kernel/irq_user.c new file mode 100644 index 000000000000..6d6f9484b884 --- /dev/null +++ b/arch/um/kernel/irq_user.c @@ -0,0 +1,443 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <unistd.h> +#include <errno.h> +#include <signal.h> +#include <string.h> +#include <sys/poll.h> +#include <sys/types.h> +#include <sys/time.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "process.h" +#include "signal_user.h" +#include "sigio.h" +#include "irq_user.h" +#include "os.h" + +struct irq_fd { + struct irq_fd *next; + void *id; + int fd; + int type; + int irq; + int pid; + int events; + int current_events; + int freed; +}; + +static struct irq_fd *active_fds = NULL; +static struct irq_fd **last_irq_ptr = &active_fds; + +static struct pollfd *pollfds = NULL; +static int pollfds_num = 0; +static int pollfds_size = 0; + +extern int io_count, intr_count; + +void sigio_handler(int sig, union uml_pt_regs *regs) +{ + struct irq_fd *irq_fd, *next; + int i, n; + + if(smp_sigio_handler()) return; + while(1){ + n = poll(pollfds, pollfds_num, 0); + if(n < 0){ + if(errno == EINTR) continue; + printk("sigio_handler : poll returned %d, " + "errno = %d\n", n, errno); + break; + } + if(n == 0) break; + + irq_fd = active_fds; + for(i = 0; i < pollfds_num; i++){ + if(pollfds[i].revents != 0){ + irq_fd->current_events = pollfds[i].revents; + pollfds[i].fd = -1; + } + irq_fd = irq_fd->next; + } + + for(irq_fd = active_fds; irq_fd != NULL; irq_fd = next){ + next = irq_fd->next; + if(irq_fd->current_events != 0){ + irq_fd->current_events = 0; + do_IRQ(irq_fd->irq, regs); + + /* This is here because the next irq may be + * freed in the handler. If a console goes + * away, both the read and write irqs will be + * freed. After do_IRQ, ->next will point to + * a good IRQ. + * Irqs can't be freed inside their handlers, + * so the next best thing is to have them + * marked as needing freeing, so that they + * can be freed here. + */ + next = irq_fd->next; + if(irq_fd->freed){ + free_irq(irq_fd->irq, irq_fd->id); + free_irq_by_irq_and_dev(irq_fd->irq, + irq_fd->id); + } + } + } + } +} + +int activate_ipi(int fd, int pid) +{ + return(os_set_fd_async(fd, pid)); +} + +static void maybe_sigio_broken(int fd, int type) +{ + if(isatty(fd)){ + if((type == IRQ_WRITE) && !pty_output_sigio){ + write_sigio_workaround(); + add_sigio_fd(fd, 0); + } + else if((type == IRQ_READ) && !pty_close_sigio){ + write_sigio_workaround(); + add_sigio_fd(fd, 1); + } + } +} + +int activate_fd(int irq, int fd, int type, void *dev_id) +{ + struct pollfd *tmp_pfd; + struct irq_fd *new_fd, *irq_fd; + unsigned long flags; + int pid, events, err, n, size; + + pid = os_getpid(); + err = os_set_fd_async(fd, pid); + if(err < 0) + goto out; + + new_fd = um_kmalloc(sizeof(*new_fd)); + err = -ENOMEM; + if(new_fd == NULL) + goto out; + + if(type == IRQ_READ) events = POLLIN | POLLPRI; + else events = POLLOUT; + *new_fd = ((struct irq_fd) { .next = NULL, + .id = dev_id, + .fd = fd, + .type = type, + .irq = irq, + .pid = pid, + .events = events, + .current_events = 0, + .freed = 0 } ); + + /* Critical section - locked by a spinlock because this stuff can + * be changed from interrupt handlers. The stuff above is done + * outside the lock because it allocates memory. + */ + + /* Actually, it only looks like it can be called from interrupt + * context. The culprit is reactivate_fd, which calls + * maybe_sigio_broken, which calls write_sigio_workaround, + * which calls activate_fd. However, write_sigio_workaround should + * only be called once, at boot time. That would make it clear that + * this is called only from process context, and can be locked with + * a semaphore. + */ + flags = irq_lock(); + for(irq_fd = active_fds; irq_fd != NULL; irq_fd = irq_fd->next){ + if((irq_fd->fd == fd) && (irq_fd->type == type)){ + printk("Registering fd %d twice\n", fd); + printk("Irqs : %d, %d\n", irq_fd->irq, irq); + printk("Ids : 0x%x, 0x%x\n", irq_fd->id, dev_id); + goto out_unlock; + } + } + + n = pollfds_num; + if(n == pollfds_size){ + while(1){ + /* Here we have to drop the lock in order to call + * kmalloc, which might sleep. If something else + * came in and changed the pollfds array, we free + * the buffer and try again. + */ + irq_unlock(flags); + size = (pollfds_num + 1) * sizeof(pollfds[0]); + tmp_pfd = um_kmalloc(size); + flags = irq_lock(); + if(tmp_pfd == NULL) + goto out_unlock; + if(n == pollfds_size) + break; + kfree(tmp_pfd); + } + if(pollfds != NULL){ + memcpy(tmp_pfd, pollfds, + sizeof(pollfds[0]) * pollfds_size); + kfree(pollfds); + } + pollfds = tmp_pfd; + pollfds_size++; + } + + if(type == IRQ_WRITE) + fd = -1; + + pollfds[pollfds_num] = ((struct pollfd) { .fd = fd, + .events = events, + .revents = 0 }); + pollfds_num++; + + *last_irq_ptr = new_fd; + last_irq_ptr = &new_fd->next; + + irq_unlock(flags); + + /* This calls activate_fd, so it has to be outside the critical + * section. + */ + maybe_sigio_broken(fd, type); + + return(0); + + out_unlock: + irq_unlock(flags); + kfree(new_fd); + out: + return(err); +} + +static void free_irq_by_cb(int (*test)(struct irq_fd *, void *), void *arg) +{ + struct irq_fd **prev; + unsigned long flags; + int i = 0; + + flags = irq_lock(); + prev = &active_fds; + while(*prev != NULL){ + if((*test)(*prev, arg)){ + struct irq_fd *old_fd = *prev; + if((pollfds[i].fd != -1) && + (pollfds[i].fd != (*prev)->fd)){ + printk("free_irq_by_cb - mismatch between " + "active_fds and pollfds, fd %d vs %d\n", + (*prev)->fd, pollfds[i].fd); + goto out; + } + memcpy(&pollfds[i], &pollfds[i + 1], + (pollfds_num - i - 1) * sizeof(pollfds[0])); + pollfds_num--; + if(last_irq_ptr == &old_fd->next) + last_irq_ptr = prev; + *prev = (*prev)->next; + if(old_fd->type == IRQ_WRITE) + ignore_sigio_fd(old_fd->fd); + kfree(old_fd); + continue; + } + prev = &(*prev)->next; + i++; + } + out: + irq_unlock(flags); +} + +struct irq_and_dev { + int irq; + void *dev; +}; + +static int same_irq_and_dev(struct irq_fd *irq, void *d) +{ + struct irq_and_dev *data = d; + + return((irq->irq == data->irq) && (irq->id == data->dev)); +} + +void free_irq_by_irq_and_dev(unsigned int irq, void *dev) +{ + struct irq_and_dev data = ((struct irq_and_dev) { .irq = irq, + .dev = dev }); + + free_irq_by_cb(same_irq_and_dev, &data); +} + +static int same_fd(struct irq_fd *irq, void *fd) +{ + return(irq->fd == *((int *) fd)); +} + +void free_irq_by_fd(int fd) +{ + free_irq_by_cb(same_fd, &fd); +} + +static struct irq_fd *find_irq_by_fd(int fd, int irqnum, int *index_out) +{ + struct irq_fd *irq; + int i = 0; + + for(irq=active_fds; irq != NULL; irq = irq->next){ + if((irq->fd == fd) && (irq->irq == irqnum)) break; + i++; + } + if(irq == NULL){ + printk("find_irq_by_fd doesn't have descriptor %d\n", fd); + goto out; + } + if((pollfds[i].fd != -1) && (pollfds[i].fd != fd)){ + printk("find_irq_by_fd - mismatch between active_fds and " + "pollfds, fd %d vs %d, need %d\n", irq->fd, + pollfds[i].fd, fd); + irq = NULL; + goto out; + } + *index_out = i; + out: + return(irq); +} + +void free_irq_later(int irq, void *dev_id) +{ + struct irq_fd *irq_fd; + unsigned long flags; + + flags = irq_lock(); + for(irq_fd = active_fds; irq_fd != NULL; irq_fd = irq_fd->next){ + if((irq_fd->irq == irq) && (irq_fd->id == dev_id)) + break; + } + if(irq_fd == NULL){ + printk("free_irq_later found no irq, irq = %d, " + "dev_id = 0x%p\n", irq, dev_id); + goto out; + } + irq_fd->freed = 1; + out: + irq_unlock(flags); +} + +void reactivate_fd(int fd, int irqnum) +{ + struct irq_fd *irq; + unsigned long flags; + int i; + + flags = irq_lock(); + irq = find_irq_by_fd(fd, irqnum, &i); + if(irq == NULL){ + irq_unlock(flags); + return; + } + + pollfds[i].fd = irq->fd; + + irq_unlock(flags); + + /* This calls activate_fd, so it has to be outside the critical + * section. + */ + maybe_sigio_broken(fd, irq->type); +} + +void deactivate_fd(int fd, int irqnum) +{ + struct irq_fd *irq; + unsigned long flags; + int i; + + flags = irq_lock(); + irq = find_irq_by_fd(fd, irqnum, &i); + if(irq == NULL) + goto out; + pollfds[i].fd = -1; + out: + irq_unlock(flags); +} + +int deactivate_all_fds(void) +{ + struct irq_fd *irq; + int err; + + for(irq=active_fds;irq != NULL;irq = irq->next){ + err = os_clear_fd_async(irq->fd); + if(err) + return(err); + } + /* If there is a signal already queued, after unblocking ignore it */ + set_handler(SIGIO, SIG_IGN, 0, -1); + + return(0); +} + +void forward_ipi(int fd, int pid) +{ + int err; + + err = os_set_owner(fd, pid); + if(err < 0) + printk("forward_ipi: set_owner failed, fd = %d, me = %d, " + "target = %d, err = %d\n", fd, os_getpid(), pid, -err); +} + +void forward_interrupts(int pid) +{ + struct irq_fd *irq; + unsigned long flags; + int err; + + flags = irq_lock(); + for(irq=active_fds;irq != NULL;irq = irq->next){ + err = os_set_owner(irq->fd, pid); + if(err < 0){ + /* XXX Just remove the irq rather than + * print out an infinite stream of these + */ + printk("Failed to forward %d to pid %d, err = %d\n", + irq->fd, pid, -err); + } + + irq->pid = pid; + } + irq_unlock(flags); +} + +void init_irq_signals(int on_sigstack) +{ + __sighandler_t h; + int flags; + + flags = on_sigstack ? SA_ONSTACK : 0; + if(timer_irq_inited) h = (__sighandler_t) alarm_handler; + else h = boot_timer_handler; + + set_handler(SIGVTALRM, h, flags | SA_RESTART, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, -1); + set_handler(SIGIO, (__sighandler_t) sig_handler, flags | SA_RESTART, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + signal(SIGWINCH, SIG_IGN); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/ksyms.c b/arch/um/kernel/ksyms.c new file mode 100644 index 000000000000..b41d3397d07b --- /dev/null +++ b/arch/um/kernel/ksyms.c @@ -0,0 +1,137 @@ +/* + * Copyright (C) 2001 - 2004 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/module.h" +#include "linux/string.h" +#include "linux/smp_lock.h" +#include "linux/spinlock.h" +#include "linux/highmem.h" +#include "asm/current.h" +#include "asm/delay.h" +#include "asm/processor.h" +#include "asm/unistd.h" +#include "asm/pgalloc.h" +#include "asm/pgtable.h" +#include "asm/page.h" +#include "asm/tlbflush.h" +#include "kern_util.h" +#include "user_util.h" +#include "mem_user.h" +#include "os.h" +#include "helper.h" + +EXPORT_SYMBOL(stop); +EXPORT_SYMBOL(uml_physmem); +EXPORT_SYMBOL(set_signals); +EXPORT_SYMBOL(get_signals); +EXPORT_SYMBOL(kernel_thread); +EXPORT_SYMBOL(__const_udelay); +EXPORT_SYMBOL(__udelay); +EXPORT_SYMBOL(sys_waitpid); +EXPORT_SYMBOL(task_size); +EXPORT_SYMBOL(flush_tlb_range); +EXPORT_SYMBOL(host_task_size); +EXPORT_SYMBOL(arch_validate); +EXPORT_SYMBOL(get_kmem_end); + +EXPORT_SYMBOL(page_to_phys); +EXPORT_SYMBOL(phys_to_page); +EXPORT_SYMBOL(high_physmem); +EXPORT_SYMBOL(empty_zero_page); +EXPORT_SYMBOL(um_virt_to_phys); +EXPORT_SYMBOL(__virt_to_page); +EXPORT_SYMBOL(to_phys); +EXPORT_SYMBOL(to_virt); +EXPORT_SYMBOL(mode_tt); +EXPORT_SYMBOL(handle_page_fault); +EXPORT_SYMBOL(find_iomem); +EXPORT_SYMBOL(end_iomem); + +#ifdef CONFIG_MODE_TT +EXPORT_SYMBOL(strncpy_from_user_tt); +EXPORT_SYMBOL(copy_from_user_tt); +EXPORT_SYMBOL(copy_to_user_tt); +#endif + +#ifdef CONFIG_MODE_SKAS +EXPORT_SYMBOL(strncpy_from_user_skas); +EXPORT_SYMBOL(copy_to_user_skas); +EXPORT_SYMBOL(copy_from_user_skas); +#endif +EXPORT_SYMBOL(uml_strdup); + +EXPORT_SYMBOL(os_stat_fd); +EXPORT_SYMBOL(os_stat_file); +EXPORT_SYMBOL(os_access); +EXPORT_SYMBOL(os_print_error); +EXPORT_SYMBOL(os_get_exec_close); +EXPORT_SYMBOL(os_set_exec_close); +EXPORT_SYMBOL(os_getpid); +EXPORT_SYMBOL(os_open_file); +EXPORT_SYMBOL(os_read_file); +EXPORT_SYMBOL(os_write_file); +EXPORT_SYMBOL(os_seek_file); +EXPORT_SYMBOL(os_lock_file); +EXPORT_SYMBOL(os_ioctl_generic); +EXPORT_SYMBOL(os_pipe); +EXPORT_SYMBOL(os_file_type); +EXPORT_SYMBOL(os_file_mode); +EXPORT_SYMBOL(os_file_size); +EXPORT_SYMBOL(os_flush_stdout); +EXPORT_SYMBOL(os_close_file); +EXPORT_SYMBOL(os_set_fd_async); +EXPORT_SYMBOL(os_set_fd_block); +EXPORT_SYMBOL(helper_wait); +EXPORT_SYMBOL(os_shutdown_socket); +EXPORT_SYMBOL(os_create_unix_socket); +EXPORT_SYMBOL(os_connect_socket); +EXPORT_SYMBOL(os_accept_connection); +EXPORT_SYMBOL(os_rcv_fd); +EXPORT_SYMBOL(run_helper); +EXPORT_SYMBOL(start_thread); +EXPORT_SYMBOL(dump_thread); + +EXPORT_SYMBOL(do_gettimeofday); +EXPORT_SYMBOL(do_settimeofday); + +/* This is here because UML expands open to sys_open, not to a system + * call instruction. + */ +EXPORT_SYMBOL(sys_open); +EXPORT_SYMBOL(sys_lseek); +EXPORT_SYMBOL(sys_read); +EXPORT_SYMBOL(sys_wait4); + +#ifdef CONFIG_SMP + +/* required for SMP */ + +extern void FASTCALL( __write_lock_failed(rwlock_t *rw)); +EXPORT_SYMBOL(__write_lock_failed); + +extern void FASTCALL( __read_lock_failed(rwlock_t *rw)); +EXPORT_SYMBOL(__read_lock_failed); + +#endif + +#ifdef CONFIG_HIGHMEM +EXPORT_SYMBOL(kmap); +EXPORT_SYMBOL(kunmap); +EXPORT_SYMBOL(kmap_atomic); +EXPORT_SYMBOL(kunmap_atomic); +EXPORT_SYMBOL(kmap_atomic_to_page); +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/main.c b/arch/um/kernel/main.c new file mode 100644 index 000000000000..a17c49703f9b --- /dev/null +++ b/arch/um/kernel/main.c @@ -0,0 +1,271 @@ +/* + * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <unistd.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <signal.h> +#include <errno.h> +#include <sys/resource.h> +#include <sys/mman.h> +#include <sys/user.h> +#include <asm/page.h> +#include "user_util.h" +#include "kern_util.h" +#include "mem_user.h" +#include "signal_user.h" +#include "time_user.h" +#include "irq_user.h" +#include "user.h" +#include "init.h" +#include "mode.h" +#include "choose-mode.h" +#include "uml-config.h" +#include "irq_user.h" +#include "time_user.h" +#include "os.h" + +/* Set in set_stklim, which is called from main and __wrap_malloc. + * __wrap_malloc only calls it if main hasn't started. + */ +unsigned long stacksizelim; + +/* Set in main */ +char *linux_prog; + +#define PGD_BOUND (4 * 1024 * 1024) +#define STACKSIZE (8 * 1024 * 1024) +#define THREAD_NAME_LEN (256) + +static void set_stklim(void) +{ + struct rlimit lim; + + if(getrlimit(RLIMIT_STACK, &lim) < 0){ + perror("getrlimit"); + exit(1); + } + if((lim.rlim_cur == RLIM_INFINITY) || (lim.rlim_cur > STACKSIZE)){ + lim.rlim_cur = STACKSIZE; + if(setrlimit(RLIMIT_STACK, &lim) < 0){ + perror("setrlimit"); + exit(1); + } + } + stacksizelim = (lim.rlim_cur + PGD_BOUND - 1) & ~(PGD_BOUND - 1); +} + +static __init void do_uml_initcalls(void) +{ + initcall_t *call; + + call = &__uml_initcall_start; + while (call < &__uml_initcall_end){; + (*call)(); + call++; + } +} + +static void last_ditch_exit(int sig) +{ + CHOOSE_MODE(kmalloc_ok = 0, (void) 0); + signal(SIGINT, SIG_DFL); + signal(SIGTERM, SIG_DFL); + signal(SIGHUP, SIG_DFL); + uml_cleanup(); + exit(1); +} + +extern int uml_exitcode; + +extern void scan_elf_aux( char **envp); + +int main(int argc, char **argv, char **envp) +{ + char **new_argv; + sigset_t mask; + int ret, i; + + /* Enable all signals except SIGIO - in some environments, we can + * enter with some signals blocked + */ + + sigemptyset(&mask); + sigaddset(&mask, SIGIO); + if(sigprocmask(SIG_SETMASK, &mask, NULL) < 0){ + perror("sigprocmask"); + exit(1); + } + +#ifdef UML_CONFIG_MODE_TT + /* Allocate memory for thread command lines */ + if(argc < 2 || strlen(argv[1]) < THREAD_NAME_LEN - 1){ + + char padding[THREAD_NAME_LEN] = { + [ 0 ... THREAD_NAME_LEN - 2] = ' ', '\0' + }; + + new_argv = malloc((argc + 2) * sizeof(char*)); + if(!new_argv) { + perror("Allocating extended argv"); + exit(1); + } + + new_argv[0] = argv[0]; + new_argv[1] = padding; + + for(i = 2; i <= argc; i++) + new_argv[i] = argv[i - 1]; + new_argv[argc + 1] = NULL; + + execvp(new_argv[0], new_argv); + perror("execing with extended args"); + exit(1); + } +#endif + + linux_prog = argv[0]; + + set_stklim(); + + new_argv = malloc((argc + 1) * sizeof(char *)); + if(new_argv == NULL){ + perror("Mallocing argv"); + exit(1); + } + for(i=0;i<argc;i++){ + new_argv[i] = strdup(argv[i]); + if(new_argv[i] == NULL){ + perror("Mallocing an arg"); + exit(1); + } + } + new_argv[argc] = NULL; + + set_handler(SIGINT, last_ditch_exit, SA_ONESHOT | SA_NODEFER, -1); + set_handler(SIGTERM, last_ditch_exit, SA_ONESHOT | SA_NODEFER, -1); + set_handler(SIGHUP, last_ditch_exit, SA_ONESHOT | SA_NODEFER, -1); + + scan_elf_aux( envp); + + do_uml_initcalls(); + ret = linux_main(argc, argv); + + /* Disable SIGPROF - I have no idea why libc doesn't do this or turn + * off the profiling time, but UML dies with a SIGPROF just before + * exiting when profiling is active. + */ + change_sig(SIGPROF, 0); + + /* Reboot */ + if(ret){ + int err; + + printf("\n"); + + /* stop timers and set SIG*ALRM to be ignored */ + disable_timer(); + + /* disable SIGIO for the fds and set SIGIO to be ignored */ + err = deactivate_all_fds(); + if(err) + printf("deactivate_all_fds failed, errno = %d\n", + -err); + + /* Let any pending signals fire now. This ensures + * that they won't be delivered after the exec, when + * they are definitely not expected. + */ + unblock_signals(); + + execvp(new_argv[0], new_argv); + perror("Failed to exec kernel"); + ret = 1; + } + printf("\n"); + return(uml_exitcode); +} + +#define CAN_KMALLOC() \ + (kmalloc_ok && CHOOSE_MODE((os_getpid() != tracing_pid), 1)) + +extern void *__real_malloc(int); + +void *__wrap_malloc(int size) +{ + void *ret; + + if(!CAN_KMALLOC()) + return(__real_malloc(size)); + else if(size <= PAGE_SIZE) /* finding contiguos pages can be hard*/ + ret = um_kmalloc(size); + else ret = um_vmalloc(size); + + /* glibc people insist that if malloc fails, errno should be + * set by malloc as well. So we do. + */ + if(ret == NULL) + errno = ENOMEM; + + return(ret); +} + +void *__wrap_calloc(int n, int size) +{ + void *ptr = __wrap_malloc(n * size); + + if(ptr == NULL) return(NULL); + memset(ptr, 0, n * size); + return(ptr); +} + +extern void __real_free(void *); + +extern unsigned long high_physmem; + +void __wrap_free(void *ptr) +{ + unsigned long addr = (unsigned long) ptr; + + /* We need to know how the allocation happened, so it can be correctly + * freed. This is done by seeing what region of memory the pointer is + * in - + * physical memory - kmalloc/kfree + * kernel virtual memory - vmalloc/vfree + * anywhere else - malloc/free + * If kmalloc is not yet possible, then either high_physmem and/or + * end_vm are still 0 (as at startup), in which case we call free, or + * we have set them, but anyway addr has not been allocated from those + * areas. So, in both cases __real_free is called. + * + * CAN_KMALLOC is checked because it would be bad to free a buffer + * with kmalloc/vmalloc after they have been turned off during + * shutdown. + * XXX: However, we sometimes shutdown CAN_KMALLOC temporarily, so + * there is a possibility for memory leaks. + */ + + if((addr >= uml_physmem) && (addr < high_physmem)){ + if(CAN_KMALLOC()) + kfree(ptr); + } + else if((addr >= start_vm) && (addr < end_vm)){ + if(CAN_KMALLOC()) + vfree(ptr); + } + else __real_free(ptr); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c new file mode 100644 index 000000000000..f156661781cb --- /dev/null +++ b/arch/um/kernel/mem.c @@ -0,0 +1,359 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/stddef.h" +#include "linux/kernel.h" +#include "linux/mm.h" +#include "linux/bootmem.h" +#include "linux/swap.h" +#include "linux/highmem.h" +#include "linux/gfp.h" +#include "asm/page.h" +#include "asm/fixmap.h" +#include "asm/pgalloc.h" +#include "user_util.h" +#include "kern_util.h" +#include "kern.h" +#include "mem_user.h" +#include "uml_uaccess.h" +#include "os.h" + +extern char __binary_start; + +/* Changed during early boot */ +unsigned long *empty_zero_page = NULL; +unsigned long *empty_bad_page = NULL; +pgd_t swapper_pg_dir[PTRS_PER_PGD]; +unsigned long highmem; +int kmalloc_ok = 0; + +static unsigned long brk_end; + +void unmap_physmem(void) +{ + os_unmap_memory((void *) brk_end, uml_reserved - brk_end); +} + +static void map_cb(void *unused) +{ + map_memory(brk_end, __pa(brk_end), uml_reserved - brk_end, 1, 1, 0); +} + +#ifdef CONFIG_HIGHMEM +static void setup_highmem(unsigned long highmem_start, + unsigned long highmem_len) +{ + struct page *page; + unsigned long highmem_pfn; + int i; + + highmem_pfn = __pa(highmem_start) >> PAGE_SHIFT; + for(i = 0; i < highmem_len >> PAGE_SHIFT; i++){ + page = &mem_map[highmem_pfn + i]; + ClearPageReserved(page); + set_bit(PG_highmem, &page->flags); + set_page_count(page, 1); + __free_page(page); + } +} +#endif + +void mem_init(void) +{ + unsigned long start; + + max_low_pfn = (high_physmem - uml_physmem) >> PAGE_SHIFT; + + /* clear the zero-page */ + memset((void *) empty_zero_page, 0, PAGE_SIZE); + + /* Map in the area just after the brk now that kmalloc is about + * to be turned on. + */ + brk_end = (unsigned long) UML_ROUND_UP(sbrk(0)); + map_cb(NULL); + initial_thread_cb(map_cb, NULL); + free_bootmem(__pa(brk_end), uml_reserved - brk_end); + uml_reserved = brk_end; + + /* Fill in any hole at the start of the binary */ + start = (unsigned long) &__binary_start & PAGE_MASK; + if(uml_physmem != start){ + map_memory(uml_physmem, __pa(uml_physmem), start - uml_physmem, + 1, 1, 0); + } + + /* this will put all low memory onto the freelists */ + totalram_pages = free_all_bootmem(); + totalhigh_pages = highmem >> PAGE_SHIFT; + totalram_pages += totalhigh_pages; + num_physpages = totalram_pages; + max_pfn = totalram_pages; + printk(KERN_INFO "Memory: %luk available\n", + (unsigned long) nr_free_pages() << (PAGE_SHIFT-10)); + kmalloc_ok = 1; + +#ifdef CONFIG_HIGHMEM + setup_highmem(end_iomem, highmem); +#endif +} + +static void __init fixrange_init(unsigned long start, unsigned long end, + pgd_t *pgd_base) +{ + pgd_t *pgd; + pmd_t *pmd; + pte_t *pte; + int i, j; + unsigned long vaddr; + + vaddr = start; + i = pgd_index(vaddr); + j = pmd_index(vaddr); + pgd = pgd_base + i; + + for ( ; (i < PTRS_PER_PGD) && (vaddr < end); pgd++, i++) { + pmd = (pmd_t *)pgd; + for (; (j < PTRS_PER_PMD) && (vaddr != end); pmd++, j++) { + if (pmd_none(*pmd)) { + pte = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE); + set_pmd(pmd, __pmd(_KERNPG_TABLE + + (unsigned long) __pa(pte))); + if (pte != pte_offset_kernel(pmd, 0)) + BUG(); + } + vaddr += PMD_SIZE; + } + j = 0; + } +} + +#ifdef CONFIG_HIGHMEM +pte_t *kmap_pte; +pgprot_t kmap_prot; + +#define kmap_get_fixmap_pte(vaddr) \ + pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k(vaddr), (vaddr)),\ + (vaddr)), (vaddr)) + +static void __init kmap_init(void) +{ + unsigned long kmap_vstart; + + /* cache the first kmap pte */ + kmap_vstart = __fix_to_virt(FIX_KMAP_BEGIN); + kmap_pte = kmap_get_fixmap_pte(kmap_vstart); + + kmap_prot = PAGE_KERNEL; +} + +static void init_highmem(void) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + unsigned long vaddr; + + /* + * Permanent kmaps: + */ + vaddr = PKMAP_BASE; + fixrange_init(vaddr, vaddr + PAGE_SIZE*LAST_PKMAP, swapper_pg_dir); + + pgd = swapper_pg_dir + pgd_index(vaddr); + pud = pud_offset(pgd, vaddr); + pmd = pmd_offset(pud, vaddr); + pte = pte_offset_kernel(pmd, vaddr); + pkmap_page_table = pte; + + kmap_init(); +} +#endif /* CONFIG_HIGHMEM */ + +static void __init fixaddr_user_init( void) +{ +#if CONFIG_ARCH_REUSE_HOST_VSYSCALL_AREA + long size = FIXADDR_USER_END - FIXADDR_USER_START; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + unsigned long paddr, vaddr = FIXADDR_USER_START; + + if ( ! size ) + return; + + fixrange_init( FIXADDR_USER_START, FIXADDR_USER_END, swapper_pg_dir); + paddr = (unsigned long)alloc_bootmem_low_pages( size); + memcpy( (void *)paddr, (void *)FIXADDR_USER_START, size); + paddr = __pa(paddr); + for ( ; size > 0; size-=PAGE_SIZE, vaddr+=PAGE_SIZE, paddr+=PAGE_SIZE){ + pgd = swapper_pg_dir + pgd_index(vaddr); + pud = pud_offset(pgd, vaddr); + pmd = pmd_offset(pud, vaddr); + pte = pte_offset_kernel(pmd, vaddr); + pte_set_val( (*pte), paddr, PAGE_READONLY); + } +#endif +} + +void paging_init(void) +{ + unsigned long zones_size[MAX_NR_ZONES], vaddr; + int i; + + empty_zero_page = (unsigned long *) alloc_bootmem_low_pages(PAGE_SIZE); + empty_bad_page = (unsigned long *) alloc_bootmem_low_pages(PAGE_SIZE); + for(i=0;i<sizeof(zones_size)/sizeof(zones_size[0]);i++) + zones_size[i] = 0; + zones_size[0] = (end_iomem >> PAGE_SHIFT) - (uml_physmem >> PAGE_SHIFT); + zones_size[2] = highmem >> PAGE_SHIFT; + free_area_init(zones_size); + + /* + * Fixed mappings, only the page table structure has to be + * created - mappings will be set by set_fixmap(): + */ + vaddr = __fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK; + fixrange_init(vaddr, FIXADDR_TOP, swapper_pg_dir); + + fixaddr_user_init(); + +#ifdef CONFIG_HIGHMEM + init_highmem(); +#endif +} + +struct page *arch_validate(struct page *page, int mask, int order) +{ + unsigned long addr, zero = 0; + int i; + + again: + if(page == NULL) return(page); + if(PageHighMem(page)) return(page); + + addr = (unsigned long) page_address(page); + for(i = 0; i < (1 << order); i++){ + current->thread.fault_addr = (void *) addr; + if(__do_copy_to_user((void __user *) addr, &zero, + sizeof(zero), + ¤t->thread.fault_addr, + ¤t->thread.fault_catcher)){ + if(!(mask & __GFP_WAIT)) return(NULL); + else break; + } + addr += PAGE_SIZE; + } + + if(i == (1 << order)) return(page); + page = alloc_pages(mask, order); + goto again; +} + +/* This can't do anything because nothing in the kernel image can be freed + * since it's not in kernel physical memory. + */ + +void free_initmem(void) +{ +} + +#ifdef CONFIG_BLK_DEV_INITRD + +void free_initrd_mem(unsigned long start, unsigned long end) +{ + if (start < end) + printk ("Freeing initrd memory: %ldk freed\n", + (end - start) >> 10); + for (; start < end; start += PAGE_SIZE) { + ClearPageReserved(virt_to_page(start)); + set_page_count(virt_to_page(start), 1); + free_page(start); + totalram_pages++; + } +} + +#endif + +void show_mem(void) +{ + int pfn, total = 0, reserved = 0; + int shared = 0, cached = 0; + int highmem = 0; + struct page *page; + + printk("Mem-info:\n"); + show_free_areas(); + printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); + pfn = max_mapnr; + while(pfn-- > 0) { + page = pfn_to_page(pfn); + total++; + if(PageHighMem(page)) + highmem++; + if(PageReserved(page)) + reserved++; + else if(PageSwapCache(page)) + cached++; + else if(page_count(page)) + shared += page_count(page) - 1; + } + printk("%d pages of RAM\n", total); + printk("%d pages of HIGHMEM\n", highmem); + printk("%d reserved pages\n", reserved); + printk("%d pages shared\n", shared); + printk("%d pages swap cached\n", cached); +} + +/* + * Allocate and free page tables. + */ + +pgd_t *pgd_alloc(struct mm_struct *mm) +{ + pgd_t *pgd = (pgd_t *)__get_free_page(GFP_KERNEL); + + if (pgd) { + memset(pgd, 0, USER_PTRS_PER_PGD * sizeof(pgd_t)); + memcpy(pgd + USER_PTRS_PER_PGD, + swapper_pg_dir + USER_PTRS_PER_PGD, + (PTRS_PER_PGD - USER_PTRS_PER_PGD) * sizeof(pgd_t)); + } + return pgd; +} + +void pgd_free(pgd_t *pgd) +{ + free_page((unsigned long) pgd); +} + +pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) +{ + pte_t *pte; + + pte = (pte_t *)__get_free_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO); + return pte; +} + +struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address) +{ + struct page *pte; + + pte = alloc_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO); + return pte; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/mem_user.c b/arch/um/kernel/mem_user.c new file mode 100644 index 000000000000..4a663fd434bb --- /dev/null +++ b/arch/um/kernel/mem_user.c @@ -0,0 +1,273 @@ +/* + * arch/um/kernel/mem_user.c + * + * BRIEF MODULE DESCRIPTION + * user side memory routines for supporting IO memory inside user mode linux + * + * Copyright (C) 2001 RidgeRun, Inc. + * Author: RidgeRun, Inc. + * Greg Lonnon glonnon@ridgerun.com or info@ridgerun.com + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN + * NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, + * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT + * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF + * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON + * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF + * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write to the Free Software Foundation, Inc., + * 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +#include <stdio.h> +#include <stdlib.h> +#include <stddef.h> +#include <stdarg.h> +#include <unistd.h> +#include <errno.h> +#include <string.h> +#include <fcntl.h> +#include <sys/types.h> +#include <sys/mman.h> +#include "kern_util.h" +#include "user.h" +#include "user_util.h" +#include "mem_user.h" +#include "init.h" +#include "os.h" +#include "tempfile.h" +#include "kern_constants.h" + +#define TEMPNAME_TEMPLATE "vm_file-XXXXXX" + +static int create_tmp_file(unsigned long len) +{ + int fd, err; + char zero; + + fd = make_tempfile(TEMPNAME_TEMPLATE, NULL, 1); + if(fd < 0) { + os_print_error(fd, "make_tempfile"); + exit(1); + } + + err = os_mode_fd(fd, 0777); + if(err < 0){ + os_print_error(err, "os_mode_fd"); + exit(1); + } + err = os_seek_file(fd, len); + if(err < 0){ + os_print_error(err, "os_seek_file"); + exit(1); + } + zero = 0; + err = os_write_file(fd, &zero, 1); + if(err != 1){ + os_print_error(err, "os_write_file"); + exit(1); + } + + return(fd); +} + +void check_tmpexec(void) +{ + void *addr; + int err, fd = create_tmp_file(UM_KERN_PAGE_SIZE); + + addr = mmap(NULL, UM_KERN_PAGE_SIZE, + PROT_READ | PROT_WRITE | PROT_EXEC, MAP_PRIVATE, fd, 0); + printf("Checking PROT_EXEC mmap in /tmp..."); + fflush(stdout); + if(addr == MAP_FAILED){ + err = errno; + perror("failed"); + if(err == EPERM) + printf("/tmp must be not mounted noexec\n"); + exit(1); + } + printf("OK\n"); + munmap(addr, UM_KERN_PAGE_SIZE); + + os_close_file(fd); +} + +static int have_devanon = 0; + +void check_devanon(void) +{ + int fd; + + printk("Checking for /dev/anon on the host..."); + fd = open("/dev/anon", O_RDWR); + if(fd < 0){ + printk("Not available (open failed with errno %d)\n", errno); + return; + } + + printk("OK\n"); + have_devanon = 1; +} + +static int create_anon_file(unsigned long len) +{ + void *addr; + int fd; + + fd = open("/dev/anon", O_RDWR); + if(fd < 0) { + os_print_error(fd, "opening /dev/anon"); + exit(1); + } + + addr = mmap(NULL, len, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0); + if(addr == MAP_FAILED){ + perror("mapping physmem file"); + exit(1); + } + munmap(addr, len); + + return(fd); +} + +int create_mem_file(unsigned long len) +{ + int err, fd; + + if(have_devanon) + fd = create_anon_file(len); + else fd = create_tmp_file(len); + + err = os_set_exec_close(fd, 1); + if(err < 0) + os_print_error(err, "exec_close"); + return(fd); +} + +struct iomem_region *iomem_regions = NULL; +int iomem_size = 0; + +static int __init parse_iomem(char *str, int *add) +{ + struct iomem_region *new; + struct uml_stat buf; + char *file, *driver; + int fd, err, size; + + driver = str; + file = strchr(str,','); + if(file == NULL){ + printf("parse_iomem : failed to parse iomem\n"); + goto out; + } + *file = '\0'; + file++; + fd = os_open_file(file, of_rdwr(OPENFLAGS()), 0); + if(fd < 0){ + os_print_error(fd, "parse_iomem - Couldn't open io file"); + goto out; + } + + err = os_stat_fd(fd, &buf); + if(err < 0){ + os_print_error(err, "parse_iomem - cannot stat_fd file"); + goto out_close; + } + + new = malloc(sizeof(*new)); + if(new == NULL){ + perror("Couldn't allocate iomem_region struct"); + goto out_close; + } + + size = (buf.ust_size + UM_KERN_PAGE_SIZE) & ~(UM_KERN_PAGE_SIZE - 1); + + *new = ((struct iomem_region) { .next = iomem_regions, + .driver = driver, + .fd = fd, + .size = size, + .phys = 0, + .virt = 0 }); + iomem_regions = new; + iomem_size += new->size + UM_KERN_PAGE_SIZE; + + return(0); + out_close: + os_close_file(fd); + out: + return(1); +} + +__uml_setup("iomem=", parse_iomem, +"iomem=<name>,<file>\n" +" Configure <file> as an IO memory region named <name>.\n\n" +); + +int protect_memory(unsigned long addr, unsigned long len, int r, int w, int x, + int must_succeed) +{ + int err; + + err = os_protect_memory((void *) addr, len, r, w, x); + if(err < 0){ + if(must_succeed) + panic("protect failed, err = %d", -err); + else return(err); + } + return(0); +} + +#if 0 +/* Debugging facility for dumping stuff out to the host, avoiding the timing + * problems that come with printf and breakpoints. + * Enable in case of emergency. + */ + +int logging = 1; +int logging_fd = -1; + +int logging_line = 0; +char logging_buf[512]; + +void log(char *fmt, ...) +{ + va_list ap; + struct timeval tv; + struct openflags flags; + + if(logging == 0) return; + if(logging_fd < 0){ + flags = of_create(of_trunc(of_rdwr(OPENFLAGS()))); + logging_fd = os_open_file("log", flags, 0644); + } + gettimeofday(&tv, NULL); + sprintf(logging_buf, "%d\t %u.%u ", logging_line++, tv.tv_sec, + tv.tv_usec); + va_start(ap, fmt); + vsprintf(&logging_buf[strlen(logging_buf)], fmt, ap); + va_end(ap); + write(logging_fd, logging_buf, strlen(logging_buf)); +} +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/physmem.c b/arch/um/kernel/physmem.c new file mode 100644 index 000000000000..420e6d51fa0f --- /dev/null +++ b/arch/um/kernel/physmem.c @@ -0,0 +1,478 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/mm.h" +#include "linux/rbtree.h" +#include "linux/slab.h" +#include "linux/vmalloc.h" +#include "linux/bootmem.h" +#include "linux/module.h" +#include "asm/types.h" +#include "asm/pgtable.h" +#include "kern_util.h" +#include "user_util.h" +#include "mode_kern.h" +#include "mem.h" +#include "mem_user.h" +#include "os.h" +#include "kern.h" +#include "init.h" + +struct phys_desc { + struct rb_node rb; + int fd; + __u64 offset; + void *virt; + unsigned long phys; + struct list_head list; +}; + +static struct rb_root phys_mappings = RB_ROOT; + +static struct rb_node **find_rb(void *virt) +{ + struct rb_node **n = &phys_mappings.rb_node; + struct phys_desc *d; + + while(*n != NULL){ + d = rb_entry(*n, struct phys_desc, rb); + if(d->virt == virt) + return(n); + + if(d->virt > virt) + n = &(*n)->rb_left; + else + n = &(*n)->rb_right; + } + + return(n); +} + +static struct phys_desc *find_phys_mapping(void *virt) +{ + struct rb_node **n = find_rb(virt); + + if(*n == NULL) + return(NULL); + + return(rb_entry(*n, struct phys_desc, rb)); +} + +static void insert_phys_mapping(struct phys_desc *desc) +{ + struct rb_node **n = find_rb(desc->virt); + + if(*n != NULL) + panic("Physical remapping for %p already present", + desc->virt); + + rb_link_node(&desc->rb, (*n)->rb_parent, n); + rb_insert_color(&desc->rb, &phys_mappings); +} + +LIST_HEAD(descriptor_mappings); + +struct desc_mapping { + int fd; + struct list_head list; + struct list_head pages; +}; + +static struct desc_mapping *find_mapping(int fd) +{ + struct desc_mapping *desc; + struct list_head *ele; + + list_for_each(ele, &descriptor_mappings){ + desc = list_entry(ele, struct desc_mapping, list); + if(desc->fd == fd) + return(desc); + } + + return(NULL); +} + +static struct desc_mapping *descriptor_mapping(int fd) +{ + struct desc_mapping *desc; + + desc = find_mapping(fd); + if(desc != NULL) + return(desc); + + desc = kmalloc(sizeof(*desc), GFP_ATOMIC); + if(desc == NULL) + return(NULL); + + *desc = ((struct desc_mapping) + { .fd = fd, + .list = LIST_HEAD_INIT(desc->list), + .pages = LIST_HEAD_INIT(desc->pages) }); + list_add(&desc->list, &descriptor_mappings); + + return(desc); +} + +int physmem_subst_mapping(void *virt, int fd, __u64 offset, int w) +{ + struct desc_mapping *fd_maps; + struct phys_desc *desc; + unsigned long phys; + int err; + + fd_maps = descriptor_mapping(fd); + if(fd_maps == NULL) + return(-ENOMEM); + + phys = __pa(virt); + desc = find_phys_mapping(virt); + if(desc != NULL) + panic("Address 0x%p is already substituted\n", virt); + + err = -ENOMEM; + desc = kmalloc(sizeof(*desc), GFP_ATOMIC); + if(desc == NULL) + goto out; + + *desc = ((struct phys_desc) + { .fd = fd, + .offset = offset, + .virt = virt, + .phys = __pa(virt), + .list = LIST_HEAD_INIT(desc->list) }); + insert_phys_mapping(desc); + + list_add(&desc->list, &fd_maps->pages); + + virt = (void *) ((unsigned long) virt & PAGE_MASK); + err = os_map_memory(virt, fd, offset, PAGE_SIZE, 1, w, 0); + if(!err) + goto out; + + rb_erase(&desc->rb, &phys_mappings); + kfree(desc); + out: + return(err); +} + +static int physmem_fd = -1; + +static void remove_mapping(struct phys_desc *desc) +{ + void *virt = desc->virt; + int err; + + rb_erase(&desc->rb, &phys_mappings); + list_del(&desc->list); + kfree(desc); + + err = os_map_memory(virt, physmem_fd, __pa(virt), PAGE_SIZE, 1, 1, 0); + if(err) + panic("Failed to unmap block device page from physical memory, " + "errno = %d", -err); +} + +int physmem_remove_mapping(void *virt) +{ + struct phys_desc *desc; + + virt = (void *) ((unsigned long) virt & PAGE_MASK); + desc = find_phys_mapping(virt); + if(desc == NULL) + return(0); + + remove_mapping(desc); + return(1); +} + +void physmem_forget_descriptor(int fd) +{ + struct desc_mapping *desc; + struct phys_desc *page; + struct list_head *ele, *next; + __u64 offset; + void *addr; + int err; + + desc = find_mapping(fd); + if(desc == NULL) + return; + + list_for_each_safe(ele, next, &desc->pages){ + page = list_entry(ele, struct phys_desc, list); + offset = page->offset; + addr = page->virt; + remove_mapping(page); + err = os_seek_file(fd, offset); + if(err) + panic("physmem_forget_descriptor - failed to seek " + "to %lld in fd %d, error = %d\n", + offset, fd, -err); + err = os_read_file(fd, addr, PAGE_SIZE); + if(err < 0) + panic("physmem_forget_descriptor - failed to read " + "from fd %d to 0x%p, error = %d\n", + fd, addr, -err); + } + + list_del(&desc->list); + kfree(desc); +} + +EXPORT_SYMBOL(physmem_forget_descriptor); +EXPORT_SYMBOL(physmem_remove_mapping); +EXPORT_SYMBOL(physmem_subst_mapping); + +void arch_free_page(struct page *page, int order) +{ + void *virt; + int i; + + for(i = 0; i < (1 << order); i++){ + virt = __va(page_to_phys(page + i)); + physmem_remove_mapping(virt); + } +} + +int is_remapped(void *virt) +{ + struct phys_desc *desc = find_phys_mapping(virt); + + return(desc != NULL); +} + +/* Changed during early boot */ +unsigned long high_physmem; + +extern unsigned long physmem_size; + +void *to_virt(unsigned long phys) +{ + return((void *) uml_physmem + phys); +} + +unsigned long to_phys(void *virt) +{ + return(((unsigned long) virt) - uml_physmem); +} + +int init_maps(unsigned long physmem, unsigned long iomem, unsigned long highmem) +{ + struct page *p, *map; + unsigned long phys_len, phys_pages, highmem_len, highmem_pages; + unsigned long iomem_len, iomem_pages, total_len, total_pages; + int i; + + phys_pages = physmem >> PAGE_SHIFT; + phys_len = phys_pages * sizeof(struct page); + + iomem_pages = iomem >> PAGE_SHIFT; + iomem_len = iomem_pages * sizeof(struct page); + + highmem_pages = highmem >> PAGE_SHIFT; + highmem_len = highmem_pages * sizeof(struct page); + + total_pages = phys_pages + iomem_pages + highmem_pages; + total_len = phys_len + iomem_pages + highmem_len; + + if(kmalloc_ok){ + map = kmalloc(total_len, GFP_KERNEL); + if(map == NULL) + map = vmalloc(total_len); + } + else map = alloc_bootmem_low_pages(total_len); + + if(map == NULL) + return(-ENOMEM); + + for(i = 0; i < total_pages; i++){ + p = &map[i]; + set_page_count(p, 0); + SetPageReserved(p); + INIT_LIST_HEAD(&p->lru); + } + + max_mapnr = total_pages; + return(0); +} + +struct page *phys_to_page(const unsigned long phys) +{ + return(&mem_map[phys >> PAGE_SHIFT]); +} + +struct page *__virt_to_page(const unsigned long virt) +{ + return(&mem_map[__pa(virt) >> PAGE_SHIFT]); +} + +phys_t page_to_phys(struct page *page) +{ + return((page - mem_map) << PAGE_SHIFT); +} + +pte_t mk_pte(struct page *page, pgprot_t pgprot) +{ + pte_t pte; + + pte_set_val(pte, page_to_phys(page), pgprot); + if(pte_present(pte)) + pte_mknewprot(pte_mknewpage(pte)); + return(pte); +} + +/* Changed during early boot */ +static unsigned long kmem_top = 0; + +unsigned long get_kmem_end(void) +{ + if(kmem_top == 0) + kmem_top = CHOOSE_MODE(kmem_end_tt, kmem_end_skas); + return(kmem_top); +} + +void map_memory(unsigned long virt, unsigned long phys, unsigned long len, + int r, int w, int x) +{ + __u64 offset; + int fd, err; + + fd = phys_mapping(phys, &offset); + err = os_map_memory((void *) virt, fd, offset, len, r, w, x); + if(err) { + if(err == -ENOMEM) + printk("try increasing the host's " + "/proc/sys/vm/max_map_count to <physical " + "memory size>/4096\n"); + panic("map_memory(0x%lx, %d, 0x%llx, %ld, %d, %d, %d) failed, " + "err = %d\n", virt, fd, offset, len, r, w, x, err); + } +} + +#define PFN_UP(x) (((x) + PAGE_SIZE-1) >> PAGE_SHIFT) + +void setup_physmem(unsigned long start, unsigned long reserve_end, + unsigned long len, unsigned long highmem) +{ + unsigned long reserve = reserve_end - start; + int pfn = PFN_UP(__pa(reserve_end)); + int delta = (len - reserve) >> PAGE_SHIFT; + int err, offset, bootmap_size; + + physmem_fd = create_mem_file(len + highmem); + + offset = uml_reserved - uml_physmem; + err = os_map_memory((void *) uml_reserved, physmem_fd, offset, + len - offset, 1, 1, 0); + if(err < 0){ + os_print_error(err, "Mapping memory"); + exit(1); + } + + bootmap_size = init_bootmem(pfn, pfn + delta); + free_bootmem(__pa(reserve_end) + bootmap_size, + len - bootmap_size - reserve); +} + +int phys_mapping(unsigned long phys, __u64 *offset_out) +{ + struct phys_desc *desc = find_phys_mapping(__va(phys & PAGE_MASK)); + int fd = -1; + + if(desc != NULL){ + fd = desc->fd; + *offset_out = desc->offset; + } + else if(phys < physmem_size){ + fd = physmem_fd; + *offset_out = phys; + } + else if(phys < __pa(end_iomem)){ + struct iomem_region *region = iomem_regions; + + while(region != NULL){ + if((phys >= region->phys) && + (phys < region->phys + region->size)){ + fd = region->fd; + *offset_out = phys - region->phys; + break; + } + region = region->next; + } + } + else if(phys < __pa(end_iomem) + highmem){ + fd = physmem_fd; + *offset_out = phys - iomem_size; + } + + return(fd); +} + +static int __init uml_mem_setup(char *line, int *add) +{ + char *retptr; + physmem_size = memparse(line,&retptr); + return 0; +} +__uml_setup("mem=", uml_mem_setup, +"mem=<Amount of desired ram>\n" +" This controls how much \"physical\" memory the kernel allocates\n" +" for the system. The size is specified as a number followed by\n" +" one of 'k', 'K', 'm', 'M', which have the obvious meanings.\n" +" This is not related to the amount of memory in the host. It can\n" +" be more, and the excess, if it's ever used, will just be swapped out.\n" +" Example: mem=64M\n\n" +); + +unsigned long find_iomem(char *driver, unsigned long *len_out) +{ + struct iomem_region *region = iomem_regions; + + while(region != NULL){ + if(!strcmp(region->driver, driver)){ + *len_out = region->size; + return(region->virt); + } + } + + return(0); +} + +int setup_iomem(void) +{ + struct iomem_region *region = iomem_regions; + unsigned long iomem_start = high_physmem + PAGE_SIZE; + int err; + + while(region != NULL){ + err = os_map_memory((void *) iomem_start, region->fd, 0, + region->size, 1, 1, 0); + if(err) + printk("Mapping iomem region for driver '%s' failed, " + "errno = %d\n", region->driver, -err); + else { + region->virt = iomem_start; + region->phys = __pa(region->virt); + } + + iomem_start += region->size + PAGE_SIZE; + region = region->next; + } + + return(0); +} + +__initcall(setup_iomem); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c new file mode 100644 index 000000000000..f76a2692adca --- /dev/null +++ b/arch/um/kernel/process.c @@ -0,0 +1,423 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <unistd.h> +#include <signal.h> +#include <sched.h> +#include <errno.h> +#include <stdarg.h> +#include <stdlib.h> +#include <setjmp.h> +#include <sys/time.h> +#include <sys/wait.h> +#include <sys/mman.h> +#include <asm/unistd.h> +#include <asm/page.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "process.h" +#include "signal_kern.h" +#include "signal_user.h" +#include "sysdep/ptrace.h" +#include "sysdep/sigcontext.h" +#include "irq_user.h" +#include "ptrace_user.h" +#include "time_user.h" +#include "init.h" +#include "os.h" +#include "uml-config.h" +#include "ptrace_user.h" +#include "choose-mode.h" +#include "mode.h" +#ifdef UML_CONFIG_MODE_SKAS +#include "skas.h" +#include "skas_ptrace.h" +#include "registers.h" +#endif + +void init_new_thread_stack(void *sig_stack, void (*usr1_handler)(int)) +{ + int flags = 0, pages; + + if(sig_stack != NULL){ + pages = (1 << UML_CONFIG_KERNEL_STACK_ORDER); + set_sigstack(sig_stack, pages * page_size()); + flags = SA_ONSTACK; + } + if(usr1_handler) set_handler(SIGUSR1, usr1_handler, flags, -1); +} + +void init_new_thread_signals(int altstack) +{ + int flags = altstack ? SA_ONSTACK : 0; + + set_handler(SIGSEGV, (__sighandler_t) sig_handler, flags, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + set_handler(SIGTRAP, (__sighandler_t) sig_handler, flags, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + set_handler(SIGFPE, (__sighandler_t) sig_handler, flags, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + set_handler(SIGILL, (__sighandler_t) sig_handler, flags, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + set_handler(SIGBUS, (__sighandler_t) sig_handler, flags, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + set_handler(SIGWINCH, (__sighandler_t) sig_handler, flags, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + set_handler(SIGUSR2, (__sighandler_t) sig_handler, + flags, SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + signal(SIGHUP, SIG_IGN); + + init_irq_signals(altstack); +} + +struct tramp { + int (*tramp)(void *); + void *tramp_data; + unsigned long temp_stack; + int flags; + int pid; +}; + +/* See above for why sigkill is here */ + +int sigkill = SIGKILL; + +int outer_tramp(void *arg) +{ + struct tramp *t; + int sig = sigkill; + + t = arg; + t->pid = clone(t->tramp, (void *) t->temp_stack + page_size()/2, + t->flags, t->tramp_data); + if(t->pid > 0) wait_for_stop(t->pid, SIGSTOP, PTRACE_CONT, NULL); + kill(os_getpid(), sig); + _exit(0); +} + +int start_fork_tramp(void *thread_arg, unsigned long temp_stack, + int clone_flags, int (*tramp)(void *)) +{ + struct tramp arg; + unsigned long sp; + int new_pid, status, err; + + /* The trampoline will run on the temporary stack */ + sp = stack_sp(temp_stack); + + clone_flags |= CLONE_FILES | SIGCHLD; + + arg.tramp = tramp; + arg.tramp_data = thread_arg; + arg.temp_stack = temp_stack; + arg.flags = clone_flags; + + /* Start the process and wait for it to kill itself */ + new_pid = clone(outer_tramp, (void *) sp, clone_flags, &arg); + if(new_pid < 0) + return(new_pid); + + CATCH_EINTR(err = waitpid(new_pid, &status, 0)); + if(err < 0) + panic("Waiting for outer trampoline failed - errno = %d", + errno); + + if(!WIFSIGNALED(status) || (WTERMSIG(status) != SIGKILL)) + panic("outer trampoline didn't exit with SIGKILL, " + "status = %d", status); + + return(arg.pid); +} + +static int ptrace_child(void *arg) +{ + int ret; + int pid = os_getpid(), ppid = getppid(); + int sc_result; + + if(ptrace(PTRACE_TRACEME, 0, 0, 0) < 0){ + perror("ptrace"); + os_kill_process(pid, 0); + } + os_stop_process(pid); + + /*This syscall will be intercepted by the parent. Don't call more than + * once, please.*/ + sc_result = os_getpid(); + + if (sc_result == pid) + ret = 1; /*Nothing modified by the parent, we are running + normally.*/ + else if (sc_result == ppid) + ret = 0; /*Expected in check_ptrace and check_sysemu when they + succeed in modifying the stack frame*/ + else + ret = 2; /*Serious trouble! This could be caused by a bug in + host 2.6 SKAS3/2.6 patch before release -V6, together + with a bug in the UML code itself.*/ + _exit(ret); +} + +static int start_ptraced_child(void **stack_out) +{ + void *stack; + unsigned long sp; + int pid, n, status; + + stack = mmap(NULL, PAGE_SIZE, PROT_READ | PROT_WRITE | PROT_EXEC, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + if(stack == MAP_FAILED) + panic("check_ptrace : mmap failed, errno = %d", errno); + sp = (unsigned long) stack + PAGE_SIZE - sizeof(void *); + pid = clone(ptrace_child, (void *) sp, SIGCHLD, NULL); + if(pid < 0) + panic("check_ptrace : clone failed, errno = %d", errno); + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + if(n < 0) + panic("check_ptrace : wait failed, errno = %d", errno); + if(!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGSTOP)) + panic("check_ptrace : expected SIGSTOP, got status = %d", + status); + + *stack_out = stack; + return(pid); +} + +/* When testing for SYSEMU support, if it is one of the broken versions, we must + * just avoid using sysemu, not panic, but only if SYSEMU features are broken. + * So only for SYSEMU features we test mustpanic, while normal host features + * must work anyway!*/ +static int stop_ptraced_child(int pid, void *stack, int exitcode, int mustpanic) +{ + int status, n, ret = 0; + + if(ptrace(PTRACE_CONT, pid, 0, 0) < 0) + panic("check_ptrace : ptrace failed, errno = %d", errno); + CATCH_EINTR(n = waitpid(pid, &status, 0)); + if(!WIFEXITED(status) || (WEXITSTATUS(status) != exitcode)) { + int exit_with = WEXITSTATUS(status); + if (exit_with == 2) + printk("check_ptrace : child exited with status 2. " + "Serious trouble happening! Try updating your " + "host skas patch!\nDisabling SYSEMU support."); + printk("check_ptrace : child exited with exitcode %d, while " + "expecting %d; status 0x%x", exit_with, + exitcode, status); + if (mustpanic) + panic("\n"); + else + printk("\n"); + ret = -1; + } + + if(munmap(stack, PAGE_SIZE) < 0) + panic("check_ptrace : munmap failed, errno = %d", errno); + return ret; +} + +static int force_sysemu_disabled = 0; + +static int __init nosysemu_cmd_param(char *str, int* add) +{ + force_sysemu_disabled = 1; + return 0; +} + +__uml_setup("nosysemu", nosysemu_cmd_param, + "nosysemu\n" + " Turns off syscall emulation patch for ptrace (SYSEMU) on.\n" + " SYSEMU is a performance-patch introduced by Laurent Vivier. It changes\n" + " behaviour of ptrace() and helps reducing host context switch rate.\n" + " To make it working, you need a kernel patch for your host, too.\n" + " See http://perso.wanadoo.fr/laurent.vivier/UML/ for further information.\n\n"); + +static void __init check_sysemu(void) +{ + void *stack; + int pid, syscall, n, status, count=0; + + printk("Checking syscall emulation patch for ptrace..."); + sysemu_supported = 0; + pid = start_ptraced_child(&stack); + + if(ptrace(PTRACE_SYSEMU, pid, 0, 0) < 0) + goto fail; + + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + if (n < 0) + panic("check_sysemu : wait failed, errno = %d", errno); + if(!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGTRAP)) + panic("check_sysemu : expected SIGTRAP, " + "got status = %d", status); + + n = ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_RET_OFFSET, + os_getpid()); + if(n < 0) + panic("check_sysemu : failed to modify system " + "call return, errno = %d", errno); + + if (stop_ptraced_child(pid, stack, 0, 0) < 0) + goto fail_stopped; + + sysemu_supported = 1; + printk("OK\n"); + set_using_sysemu(!force_sysemu_disabled); + + printk("Checking advanced syscall emulation patch for ptrace..."); + pid = start_ptraced_child(&stack); + while(1){ + count++; + if(ptrace(PTRACE_SYSEMU_SINGLESTEP, pid, 0, 0) < 0) + goto fail; + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + if(n < 0) + panic("check_ptrace : wait failed, errno = %d", errno); + if(!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGTRAP)) + panic("check_ptrace : expected (SIGTRAP|SYSCALL_TRAP), " + "got status = %d", status); + + syscall = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_NR_OFFSET, + 0); + if(syscall == __NR_getpid){ + if (!count) + panic("check_ptrace : SYSEMU_SINGLESTEP doesn't singlestep"); + n = ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_RET_OFFSET, + os_getpid()); + if(n < 0) + panic("check_sysemu : failed to modify system " + "call return, errno = %d", errno); + break; + } + } + if (stop_ptraced_child(pid, stack, 0, 0) < 0) + goto fail_stopped; + + sysemu_supported = 2; + printk("OK\n"); + + if ( !force_sysemu_disabled ) + set_using_sysemu(sysemu_supported); + return; + +fail: + stop_ptraced_child(pid, stack, 1, 0); +fail_stopped: + printk("missing\n"); +} + +void __init check_ptrace(void) +{ + void *stack; + int pid, syscall, n, status; + + printk("Checking that ptrace can change system call numbers..."); + pid = start_ptraced_child(&stack); + + if (ptrace(PTRACE_OLDSETOPTIONS, pid, 0, (void *)PTRACE_O_TRACESYSGOOD) < 0) + panic("check_ptrace: PTRACE_SETOPTIONS failed, errno = %d", errno); + + while(1){ + if(ptrace(PTRACE_SYSCALL, pid, 0, 0) < 0) + panic("check_ptrace : ptrace failed, errno = %d", + errno); + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + if(n < 0) + panic("check_ptrace : wait failed, errno = %d", errno); + if(!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGTRAP + 0x80)) + panic("check_ptrace : expected SIGTRAP + 0x80, " + "got status = %d", status); + + syscall = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_NR_OFFSET, + 0); + if(syscall == __NR_getpid){ + n = ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET, + __NR_getppid); + if(n < 0) + panic("check_ptrace : failed to modify system " + "call, errno = %d", errno); + break; + } + } + stop_ptraced_child(pid, stack, 0, 1); + printk("OK\n"); + check_sysemu(); +} + +int run_kernel_thread(int (*fn)(void *), void *arg, void **jmp_ptr) +{ + sigjmp_buf buf; + int n; + + *jmp_ptr = &buf; + n = sigsetjmp(buf, 1); + if(n != 0) + return(n); + (*fn)(arg); + return(0); +} + +void forward_pending_sigio(int target) +{ + sigset_t sigs; + + if(sigpending(&sigs)) + panic("forward_pending_sigio : sigpending failed"); + if(sigismember(&sigs, SIGIO)) + kill(target, SIGIO); +} + +#ifdef UML_CONFIG_MODE_SKAS +static inline int check_skas3_ptrace_support(void) +{ + struct ptrace_faultinfo fi; + void *stack; + int pid, n, ret = 1; + + printf("Checking for the skas3 patch in the host..."); + pid = start_ptraced_child(&stack); + + n = ptrace(PTRACE_FAULTINFO, pid, 0, &fi); + if (n < 0) { + if(errno == EIO) + printf("not found\n"); + else { + perror("not found"); + } + ret = 0; + } else { + printf("found\n"); + } + + init_registers(pid); + stop_ptraced_child(pid, stack, 1, 1); + + return(ret); +} + +int can_do_skas(void) +{ + int ret = 1; + + printf("Checking for /proc/mm..."); + if (os_access("/proc/mm", OS_ACC_W_OK) < 0) { + printf("not found\n"); + ret = 0; + goto out; + } else { + printf("found\n"); + } + + ret = check_skas3_ptrace_support(); +out: + return ret; +} +#else +int can_do_skas(void) +{ + return(0); +} +#endif diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c new file mode 100644 index 000000000000..1d719d5b4bb9 --- /dev/null +++ b/arch/um/kernel/process_kern.c @@ -0,0 +1,500 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Copyright 2003 PathScale, Inc. + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/kernel.h" +#include "linux/sched.h" +#include "linux/interrupt.h" +#include "linux/mm.h" +#include "linux/slab.h" +#include "linux/utsname.h" +#include "linux/fs.h" +#include "linux/utime.h" +#include "linux/smp_lock.h" +#include "linux/module.h" +#include "linux/init.h" +#include "linux/capability.h" +#include "linux/vmalloc.h" +#include "linux/spinlock.h" +#include "linux/proc_fs.h" +#include "linux/ptrace.h" +#include "linux/random.h" +#include "asm/unistd.h" +#include "asm/mman.h" +#include "asm/segment.h" +#include "asm/stat.h" +#include "asm/pgtable.h" +#include "asm/processor.h" +#include "asm/tlbflush.h" +#include "asm/uaccess.h" +#include "asm/user.h" +#include "user_util.h" +#include "kern_util.h" +#include "kern.h" +#include "signal_kern.h" +#include "signal_user.h" +#include "init.h" +#include "irq_user.h" +#include "mem_user.h" +#include "time_user.h" +#include "tlb.h" +#include "frame_kern.h" +#include "sigcontext.h" +#include "2_5compat.h" +#include "os.h" +#include "mode.h" +#include "mode_kern.h" +#include "choose-mode.h" + +/* This is a per-cpu array. A processor only modifies its entry and it only + * cares about its entry, so it's OK if another processor is modifying its + * entry. + */ +struct cpu_task cpu_tasks[NR_CPUS] = { [0 ... NR_CPUS - 1] = { -1, NULL } }; + +struct task_struct *get_task(int pid, int require) +{ + struct task_struct *ret; + + read_lock(&tasklist_lock); + ret = find_task_by_pid(pid); + read_unlock(&tasklist_lock); + + if(require && (ret == NULL)) panic("get_task couldn't find a task\n"); + return(ret); +} + +int external_pid(void *t) +{ + struct task_struct *task = t ? t : current; + + return(CHOOSE_MODE_PROC(external_pid_tt, external_pid_skas, task)); +} + +int pid_to_processor_id(int pid) +{ + int i; + + for(i = 0; i < ncpus; i++){ + if(cpu_tasks[i].pid == pid) return(i); + } + return(-1); +} + +void free_stack(unsigned long stack, int order) +{ + free_pages(stack, order); +} + +unsigned long alloc_stack(int order, int atomic) +{ + unsigned long page; + int flags = GFP_KERNEL; + + if(atomic) flags |= GFP_ATOMIC; + page = __get_free_pages(flags, order); + if(page == 0) + return(0); + stack_protections(page); + return(page); +} + +int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags) +{ + int pid; + + current->thread.request.u.thread.proc = fn; + current->thread.request.u.thread.arg = arg; + pid = do_fork(CLONE_VM | CLONE_UNTRACED | flags, 0, NULL, 0, NULL, + NULL); + if(pid < 0) + panic("do_fork failed in kernel_thread, errno = %d", pid); + return(pid); +} + +void switch_mm(struct mm_struct *prev, struct mm_struct *next, + struct task_struct *tsk) +{ + int cpu = smp_processor_id(); + + if (prev != next) + cpu_clear(cpu, prev->cpu_vm_mask); + cpu_set(cpu, next->cpu_vm_mask); +} + +void set_current(void *t) +{ + struct task_struct *task = t; + + cpu_tasks[task->thread_info->cpu] = ((struct cpu_task) + { external_pid(task), task }); +} + +void *_switch_to(void *prev, void *next, void *last) +{ + return(CHOOSE_MODE(switch_to_tt(prev, next), + switch_to_skas(prev, next))); +} + +void interrupt_end(void) +{ + if(need_resched()) schedule(); + if(test_tsk_thread_flag(current, TIF_SIGPENDING)) do_signal(); +} + +void release_thread(struct task_struct *task) +{ + CHOOSE_MODE(release_thread_tt(task), release_thread_skas(task)); +} + +void exit_thread(void) +{ + CHOOSE_MODE(exit_thread_tt(), exit_thread_skas()); + unprotect_stack((unsigned long) current_thread); +} + +void *get_current(void) +{ + return(current); +} + +void prepare_to_copy(struct task_struct *tsk) +{ +} + +int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, + unsigned long stack_top, struct task_struct * p, + struct pt_regs *regs) +{ + p->thread = (struct thread_struct) INIT_THREAD; + return(CHOOSE_MODE_PROC(copy_thread_tt, copy_thread_skas, nr, + clone_flags, sp, stack_top, p, regs)); +} + +void initial_thread_cb(void (*proc)(void *), void *arg) +{ + int save_kmalloc_ok = kmalloc_ok; + + kmalloc_ok = 0; + CHOOSE_MODE_PROC(initial_thread_cb_tt, initial_thread_cb_skas, proc, + arg); + kmalloc_ok = save_kmalloc_ok; +} + +unsigned long stack_sp(unsigned long page) +{ + return(page + PAGE_SIZE - sizeof(void *)); +} + +int current_pid(void) +{ + return(current->pid); +} + +void default_idle(void) +{ + uml_idle_timer(); + + atomic_inc(&init_mm.mm_count); + current->mm = &init_mm; + current->active_mm = &init_mm; + + while(1){ + /* endless idle loop with no priority at all */ + SET_PRI(current); + + /* + * although we are an idle CPU, we do not want to + * get into the scheduler unnecessarily. + */ + if(need_resched()) + schedule(); + + idle_sleep(10); + } +} + +void cpu_idle(void) +{ + CHOOSE_MODE(init_idle_tt(), init_idle_skas()); +} + +int page_size(void) +{ + return(PAGE_SIZE); +} + +unsigned long page_mask(void) +{ + return(PAGE_MASK); +} + +void *um_virt_to_phys(struct task_struct *task, unsigned long addr, + pte_t *pte_out) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + + if(task->mm == NULL) + return(ERR_PTR(-EINVAL)); + pgd = pgd_offset(task->mm, addr); + if(!pgd_present(*pgd)) + return(ERR_PTR(-EINVAL)); + + pud = pud_offset(pgd, addr); + if(!pud_present(*pud)) + return(ERR_PTR(-EINVAL)); + + pmd = pmd_offset(pud, addr); + if(!pmd_present(*pmd)) + return(ERR_PTR(-EINVAL)); + + pte = pte_offset_kernel(pmd, addr); + if(!pte_present(*pte)) + return(ERR_PTR(-EINVAL)); + + if(pte_out != NULL) + *pte_out = *pte; + return((void *) (pte_val(*pte) & PAGE_MASK) + (addr & ~PAGE_MASK)); +} + +char *current_cmd(void) +{ +#if defined(CONFIG_SMP) || defined(CONFIG_HIGHMEM) + return("(Unknown)"); +#else + void *addr = um_virt_to_phys(current, current->mm->arg_start, NULL); + return IS_ERR(addr) ? "(Unknown)": __va((unsigned long) addr); +#endif +} + +void force_sigbus(void) +{ + printk(KERN_ERR "Killing pid %d because of a lack of memory\n", + current->pid); + lock_kernel(); + sigaddset(¤t->pending.signal, SIGBUS); + recalc_sigpending(); + current->flags |= PF_SIGNALED; + do_exit(SIGBUS | 0x80); +} + +void dump_thread(struct pt_regs *regs, struct user *u) +{ +} + +void enable_hlt(void) +{ + panic("enable_hlt"); +} + +EXPORT_SYMBOL(enable_hlt); + +void disable_hlt(void) +{ + panic("disable_hlt"); +} + +EXPORT_SYMBOL(disable_hlt); + +void *um_kmalloc(int size) +{ + return(kmalloc(size, GFP_KERNEL)); +} + +void *um_kmalloc_atomic(int size) +{ + return(kmalloc(size, GFP_ATOMIC)); +} + +void *um_vmalloc(int size) +{ + return(vmalloc(size)); +} + +unsigned long get_fault_addr(void) +{ + return((unsigned long) current->thread.fault_addr); +} + +EXPORT_SYMBOL(get_fault_addr); + +void not_implemented(void) +{ + printk(KERN_DEBUG "Something isn't implemented in here\n"); +} + +EXPORT_SYMBOL(not_implemented); + +int user_context(unsigned long sp) +{ + unsigned long stack; + + stack = sp & (PAGE_MASK << CONFIG_KERNEL_STACK_ORDER); + return(stack != (unsigned long) current_thread); +} + +extern void remove_umid_dir(void); + +__uml_exitcall(remove_umid_dir); + +extern exitcall_t __uml_exitcall_begin, __uml_exitcall_end; + +void do_uml_exitcalls(void) +{ + exitcall_t *call; + + call = &__uml_exitcall_end; + while (--call >= &__uml_exitcall_begin) + (*call)(); +} + +char *uml_strdup(char *string) +{ + char *new; + + new = kmalloc(strlen(string) + 1, GFP_KERNEL); + if(new == NULL) return(NULL); + strcpy(new, string); + return(new); +} + +void *get_init_task(void) +{ + return(&init_thread_union.thread_info.task); +} + +int copy_to_user_proc(void __user *to, void *from, int size) +{ + return(copy_to_user(to, from, size)); +} + +int copy_from_user_proc(void *to, void __user *from, int size) +{ + return(copy_from_user(to, from, size)); +} + +int clear_user_proc(void __user *buf, int size) +{ + return(clear_user(buf, size)); +} + +int strlen_user_proc(char __user *str) +{ + return(strlen_user(str)); +} + +int smp_sigio_handler(void) +{ +#ifdef CONFIG_SMP + int cpu = current_thread->cpu; + IPI_handler(cpu); + if(cpu != 0) + return(1); +#endif + return(0); +} + +int um_in_interrupt(void) +{ + return(in_interrupt()); +} + +int cpu(void) +{ + return(current_thread->cpu); +} + +static atomic_t using_sysemu = ATOMIC_INIT(0); +int sysemu_supported; + +void set_using_sysemu(int value) +{ + if (value > sysemu_supported) + return; + atomic_set(&using_sysemu, value); +} + +int get_using_sysemu(void) +{ + return atomic_read(&using_sysemu); +} + +static int proc_read_sysemu(char *buf, char **start, off_t offset, int size,int *eof, void *data) +{ + if (snprintf(buf, size, "%d\n", get_using_sysemu()) < size) /*No overflow*/ + *eof = 1; + + return strlen(buf); +} + +static int proc_write_sysemu(struct file *file,const char *buf, unsigned long count,void *data) +{ + char tmp[2]; + + if (copy_from_user(tmp, buf, 1)) + return -EFAULT; + + if (tmp[0] >= '0' && tmp[0] <= '2') + set_using_sysemu(tmp[0] - '0'); + return count; /*We use the first char, but pretend to write everything*/ +} + +int __init make_proc_sysemu(void) +{ + struct proc_dir_entry *ent; + if (!sysemu_supported) + return 0; + + ent = create_proc_entry("sysemu", 0600, &proc_root); + + if (ent == NULL) + { + printk("Failed to register /proc/sysemu\n"); + return(0); + } + + ent->read_proc = proc_read_sysemu; + ent->write_proc = proc_write_sysemu; + + return 0; +} + +late_initcall(make_proc_sysemu); + +int singlestepping(void * t) +{ + struct task_struct *task = t ? t : current; + + if ( ! (task->ptrace & PT_DTRACE) ) + return(0); + + if (task->thread.singlestep_syscall) + return(1); + + return 2; +} + +unsigned long arch_align_stack(unsigned long sp) +{ + if (randomize_va_space) + sp -= get_random_int() % 8192; + return sp & ~0xf; +} + + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/ptrace.c b/arch/um/kernel/ptrace.c new file mode 100644 index 000000000000..3a99ee6d94eb --- /dev/null +++ b/arch/um/kernel/ptrace.c @@ -0,0 +1,388 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/mm.h" +#include "linux/errno.h" +#include "linux/smp_lock.h" +#include "linux/security.h" +#include "linux/ptrace.h" +#include "linux/audit.h" +#ifdef CONFIG_PROC_MM +#include "linux/proc_mm.h" +#endif +#include "asm/ptrace.h" +#include "asm/uaccess.h" +#include "kern_util.h" +#include "skas_ptrace.h" +#include "sysdep/ptrace.h" + +/* + * Called by kernel/ptrace.c when detaching.. + */ +void ptrace_disable(struct task_struct *child) +{ + child->ptrace &= ~PT_DTRACE; + child->thread.singlestep_syscall = 0; +} + +long sys_ptrace(long request, long pid, long addr, long data) +{ + struct task_struct *child; + int i, ret; + + lock_kernel(); + ret = -EPERM; + if (request == PTRACE_TRACEME) { + /* are we already being traced? */ + if (current->ptrace & PT_PTRACED) + goto out; + + ret = security_ptrace(current->parent, current); + if (ret) + goto out; + + /* set the ptrace bit in the process flags. */ + current->ptrace |= PT_PTRACED; + ret = 0; + goto out; + } + ret = -ESRCH; + read_lock(&tasklist_lock); + child = find_task_by_pid(pid); + if (child) + get_task_struct(child); + read_unlock(&tasklist_lock); + if (!child) + goto out; + + ret = -EPERM; + if (pid == 1) /* you may not mess with init */ + goto out_tsk; + + if (request == PTRACE_ATTACH) { + ret = ptrace_attach(child); + goto out_tsk; + } + + ret = ptrace_check_attach(child, request == PTRACE_KILL); + if (ret < 0) + goto out_tsk; + + switch (request) { + /* when I and D space are separate, these will need to be fixed. */ + case PTRACE_PEEKTEXT: /* read word at location addr. */ + case PTRACE_PEEKDATA: { + unsigned long tmp; + int copied; + + ret = -EIO; + copied = access_process_vm(child, addr, &tmp, sizeof(tmp), 0); + if (copied != sizeof(tmp)) + break; + ret = put_user(tmp, (unsigned long __user *) data); + break; + } + + /* read the word at location addr in the USER area. */ + case PTRACE_PEEKUSR: { + unsigned long tmp; + + ret = -EIO; + if ((addr & 3) || addr < 0) + break; + + tmp = 0; /* Default return condition */ + if(addr < MAX_REG_OFFSET){ + tmp = getreg(child, addr); + } + else if((addr >= offsetof(struct user, u_debugreg[0])) && + (addr <= offsetof(struct user, u_debugreg[7]))){ + addr -= offsetof(struct user, u_debugreg[0]); + addr = addr >> 2; + tmp = child->thread.arch.debugregs[addr]; + } + ret = put_user(tmp, (unsigned long __user *) data); + break; + } + + /* when I and D space are separate, this will have to be fixed. */ + case PTRACE_POKETEXT: /* write the word at location addr. */ + case PTRACE_POKEDATA: + ret = -EIO; + if (access_process_vm(child, addr, &data, sizeof(data), + 1) != sizeof(data)) + break; + ret = 0; + break; + + case PTRACE_POKEUSR: /* write the word at location addr in the USER area */ + ret = -EIO; + if ((addr & 3) || addr < 0) + break; + + if (addr < MAX_REG_OFFSET) { + ret = putreg(child, addr, data); + break; + } +#if 0 /* XXX x86_64 */ + else if((addr >= offsetof(struct user, u_debugreg[0])) && + (addr <= offsetof(struct user, u_debugreg[7]))){ + addr -= offsetof(struct user, u_debugreg[0]); + addr = addr >> 2; + if((addr == 4) || (addr == 5)) break; + child->thread.arch.debugregs[addr] = data; + ret = 0; + } +#endif + + break; + + case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ + case PTRACE_CONT: { /* restart after signal. */ + ret = -EIO; + if ((unsigned long) data > _NSIG) + break; + + child->ptrace &= ~PT_DTRACE; + child->thread.singlestep_syscall = 0; + if (request == PTRACE_SYSCALL) { + set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); + } + else { + clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); + } + child->exit_code = data; + wake_up_process(child); + ret = 0; + break; + } + +/* + * make the child exit. Best I can do is send it a sigkill. + * perhaps it should be put in the status that it wants to + * exit. + */ + case PTRACE_KILL: { + ret = 0; + if (child->exit_state == EXIT_ZOMBIE) /* already dead */ + break; + + child->ptrace &= ~PT_DTRACE; + child->thread.singlestep_syscall = 0; + child->exit_code = SIGKILL; + wake_up_process(child); + break; + } + + case PTRACE_SINGLESTEP: { /* set the trap flag. */ + ret = -EIO; + if ((unsigned long) data > _NSIG) + break; + clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); + child->ptrace |= PT_DTRACE; + child->thread.singlestep_syscall = 0; + child->exit_code = data; + /* give it a chance to run. */ + wake_up_process(child); + ret = 0; + break; + } + + case PTRACE_DETACH: + /* detach a process that was attached. */ + ret = ptrace_detach(child, data); + break; + +#ifdef PTRACE_GETREGS + case PTRACE_GETREGS: { /* Get all gp regs from the child. */ + if (!access_ok(VERIFY_WRITE, (unsigned long *)data, + MAX_REG_OFFSET)) { + ret = -EIO; + break; + } + for ( i = 0; i < MAX_REG_OFFSET; i += sizeof(long) ) { + __put_user(getreg(child, i), + (unsigned long __user *) data); + data += sizeof(long); + } + ret = 0; + break; + } +#endif +#ifdef PTRACE_SETREGS + case PTRACE_SETREGS: { /* Set all gp regs in the child. */ + unsigned long tmp = 0; + if (!access_ok(VERIFY_READ, (unsigned *)data, + MAX_REG_OFFSET)) { + ret = -EIO; + break; + } + for ( i = 0; i < MAX_REG_OFFSET; i += sizeof(long) ) { + __get_user(tmp, (unsigned long __user *) data); + putreg(child, i, tmp); + data += sizeof(long); + } + ret = 0; + break; + } +#endif +#ifdef PTRACE_GETFPREGS + case PTRACE_GETFPREGS: /* Get the child FPU state. */ + ret = get_fpregs(data, child); + break; +#endif +#ifdef PTRACE_SETFPREGS + case PTRACE_SETFPREGS: /* Set the child FPU state. */ + ret = set_fpregs(data, child); + break; +#endif +#ifdef PTRACE_GETFPXREGS + case PTRACE_GETFPXREGS: /* Get the child FPU state. */ + ret = get_fpxregs(data, child); + break; +#endif +#ifdef PTRACE_SETFPXREGS + case PTRACE_SETFPXREGS: /* Set the child FPU state. */ + ret = set_fpxregs(data, child); + break; +#endif + case PTRACE_FAULTINFO: { + struct ptrace_faultinfo fault; + + fault = ((struct ptrace_faultinfo) + { .is_write = child->thread.err, + .addr = child->thread.cr2 }); + ret = copy_to_user((unsigned long __user *) data, &fault, + sizeof(fault)); + if(ret) + break; + break; + } + case PTRACE_SIGPENDING: + ret = copy_to_user((unsigned long __user *) data, + &child->pending.signal, + sizeof(child->pending.signal)); + break; + + case PTRACE_LDT: { + struct ptrace_ldt ldt; + + if(copy_from_user(&ldt, (unsigned long __user *) data, + sizeof(ldt))){ + ret = -EIO; + break; + } + + /* This one is confusing, so just punt and return -EIO for + * now + */ + ret = -EIO; + break; + } +#ifdef CONFIG_PROC_MM + case PTRACE_SWITCH_MM: { + struct mm_struct *old = child->mm; + struct mm_struct *new = proc_mm_get_mm(data); + + if(IS_ERR(new)){ + ret = PTR_ERR(new); + break; + } + + atomic_inc(&new->mm_users); + child->mm = new; + child->active_mm = new; + mmput(old); + ret = 0; + break; + } +#endif + default: + ret = ptrace_request(child, request, addr, data); + break; + } + out_tsk: + put_task_struct(child); + out: + unlock_kernel(); + return ret; +} + +void send_sigtrap(struct task_struct *tsk, union uml_pt_regs *regs, + int error_code) +{ + struct siginfo info; + + memset(&info, 0, sizeof(info)); + info.si_signo = SIGTRAP; + info.si_code = TRAP_BRKPT; + + /* User-mode eip? */ + info.si_addr = UPT_IS_USER(regs) ? (void __user *) UPT_IP(regs) : NULL; + + /* Send us the fakey SIGTRAP */ + force_sig_info(SIGTRAP, &info, tsk); +} + +/* XXX Check PT_DTRACE vs TIF_SINGLESTEP for singlestepping check and + * PT_PTRACED vs TIF_SYSCALL_TRACE for syscall tracing check + */ +void syscall_trace(union uml_pt_regs *regs, int entryexit) +{ + int is_singlestep = (current->ptrace & PT_DTRACE) && entryexit; + int tracesysgood; + + if (unlikely(current->audit_context)) { + if (!entryexit) + audit_syscall_entry(current, + UPT_SYSCALL_NR(®s->regs), + UPT_SYSCALL_ARG1(®s->regs), + UPT_SYSCALL_ARG2(®s->regs), + UPT_SYSCALL_ARG3(®s->regs), + UPT_SYSCALL_ARG4(®s->regs)); + else + audit_syscall_exit(current, + UPT_SYSCALL_RET(®s->regs)); + } + + /* Fake a debug trap */ + if (is_singlestep) + send_sigtrap(current, regs, 0); + + if (!test_thread_flag(TIF_SYSCALL_TRACE)) + return; + + if (!(current->ptrace & PT_PTRACED)) + return; + + /* the 0x80 provides a way for the tracing parent to distinguish + between a syscall stop and SIGTRAP delivery */ + tracesysgood = (current->ptrace & PT_TRACESYSGOOD); + ptrace_notify(SIGTRAP | (tracesysgood ? 0x80 : 0)); + + if (entryexit) /* force do_signal() --> is_syscall() */ + set_thread_flag(TIF_SIGPENDING); + + /* this isn't the same as continuing with a signal, but it will do + * for normal use. strace only continues with a signal if the + * stopping signal is not SIGTRAP. -brl + */ + if (current->exit_code) { + send_sig(current->exit_code, current, 1); + current->exit_code = 0; + } +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/reboot.c b/arch/um/kernel/reboot.c new file mode 100644 index 000000000000..207f89d74908 --- /dev/null +++ b/arch/um/kernel/reboot.c @@ -0,0 +1,79 @@ +/* + * Copyright (C) 2000, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/module.h" +#include "linux/sched.h" +#include "user_util.h" +#include "kern_util.h" +#include "kern.h" +#include "os.h" +#include "mode.h" +#include "choose-mode.h" + +#ifdef CONFIG_SMP +static void kill_idlers(int me) +{ +#ifdef CONFIG_MODE_TT + struct task_struct *p; + int i; + + for(i = 0; i < sizeof(idle_threads)/sizeof(idle_threads[0]); i++){ + p = idle_threads[i]; + if((p != NULL) && (p->thread.mode.tt.extern_pid != me)) + os_kill_process(p->thread.mode.tt.extern_pid, 0); + } +#endif +} +#endif + +static void kill_off_processes(void) +{ + CHOOSE_MODE(kill_off_processes_tt(), kill_off_processes_skas()); +#ifdef CONFIG_SMP + kill_idlers(os_getpid()); +#endif +} + +void uml_cleanup(void) +{ + kill_off_processes(); + do_uml_exitcalls(); +} + +void machine_restart(char * __unused) +{ + do_uml_exitcalls(); + kill_off_processes(); + CHOOSE_MODE(reboot_tt(), reboot_skas()); +} + +EXPORT_SYMBOL(machine_restart); + +void machine_power_off(void) +{ + do_uml_exitcalls(); + kill_off_processes(); + CHOOSE_MODE(halt_tt(), halt_skas()); +} + +EXPORT_SYMBOL(machine_power_off); + +void machine_halt(void) +{ + machine_power_off(); +} + +EXPORT_SYMBOL(machine_halt); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/resource.c b/arch/um/kernel/resource.c new file mode 100644 index 000000000000..32188e12e8af --- /dev/null +++ b/arch/um/kernel/resource.c @@ -0,0 +1,23 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/pci.h" + +unsigned long resource_fixup(struct pci_dev * dev, struct resource * res, + unsigned long start, unsigned long size) +{ + return start; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/sigio_kern.c b/arch/um/kernel/sigio_kern.c new file mode 100644 index 000000000000..229988463c4c --- /dev/null +++ b/arch/um/kernel/sigio_kern.c @@ -0,0 +1,63 @@ +/* + * Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/kernel.h" +#include "linux/list.h" +#include "linux/slab.h" +#include "linux/signal.h" +#include "linux/interrupt.h" +#include "init.h" +#include "sigio.h" +#include "irq_user.h" +#include "irq_kern.h" + +/* Protected by sigio_lock() called from write_sigio_workaround */ +static int sigio_irq_fd = -1; + +static irqreturn_t sigio_interrupt(int irq, void *data, struct pt_regs *unused) +{ + read_sigio_fd(sigio_irq_fd); + reactivate_fd(sigio_irq_fd, SIGIO_WRITE_IRQ); + return(IRQ_HANDLED); +} + +int write_sigio_irq(int fd) +{ + int err; + + err = um_request_irq(SIGIO_WRITE_IRQ, fd, IRQ_READ, sigio_interrupt, + SA_INTERRUPT | SA_SAMPLE_RANDOM, "write sigio", + NULL); + if(err){ + printk("write_sigio_irq : um_request_irq failed, err = %d\n", + err); + return(-1); + } + sigio_irq_fd = fd; + return(0); +} + +static DEFINE_SPINLOCK(sigio_spinlock); + +void sigio_lock(void) +{ + spin_lock(&sigio_spinlock); +} + +void sigio_unlock(void) +{ + spin_unlock(&sigio_spinlock); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/sigio_user.c b/arch/um/kernel/sigio_user.c new file mode 100644 index 000000000000..668df13d8c9d --- /dev/null +++ b/arch/um/kernel/sigio_user.c @@ -0,0 +1,431 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <unistd.h> +#include <stdlib.h> +#include <termios.h> +#include <pty.h> +#include <signal.h> +#include <errno.h> +#include <string.h> +#include <sched.h> +#include <sys/socket.h> +#include <sys/poll.h> +#include "init.h" +#include "user.h" +#include "kern_util.h" +#include "user_util.h" +#include "sigio.h" +#include "helper.h" +#include "os.h" + +/* Changed during early boot */ +int pty_output_sigio = 0; +int pty_close_sigio = 0; + +/* Used as a flag during SIGIO testing early in boot */ +static volatile int got_sigio = 0; + +void __init handler(int sig) +{ + got_sigio = 1; +} + +struct openpty_arg { + int master; + int slave; + int err; +}; + +static void openpty_cb(void *arg) +{ + struct openpty_arg *info = arg; + + info->err = 0; + if(openpty(&info->master, &info->slave, NULL, NULL, NULL)) + info->err = -errno; +} + +void __init check_one_sigio(void (*proc)(int, int)) +{ + struct sigaction old, new; + struct openpty_arg pty = { .master = -1, .slave = -1 }; + int master, slave, err; + + initial_thread_cb(openpty_cb, &pty); + if(pty.err){ + printk("openpty failed, errno = %d\n", -pty.err); + return; + } + + master = pty.master; + slave = pty.slave; + + if((master == -1) || (slave == -1)){ + printk("openpty failed to allocate a pty\n"); + return; + } + + /* Not now, but complain so we now where we failed. */ + err = raw(master); + if (err < 0) + panic("check_sigio : __raw failed, errno = %d\n", -err); + + err = os_sigio_async(master, slave); + if(err < 0) + panic("tty_fds : sigio_async failed, err = %d\n", -err); + + if(sigaction(SIGIO, NULL, &old) < 0) + panic("check_sigio : sigaction 1 failed, errno = %d\n", errno); + new = old; + new.sa_handler = handler; + if(sigaction(SIGIO, &new, NULL) < 0) + panic("check_sigio : sigaction 2 failed, errno = %d\n", errno); + + got_sigio = 0; + (*proc)(master, slave); + + os_close_file(master); + os_close_file(slave); + + if(sigaction(SIGIO, &old, NULL) < 0) + panic("check_sigio : sigaction 3 failed, errno = %d\n", errno); +} + +static void tty_output(int master, int slave) +{ + int n; + char buf[512]; + + printk("Checking that host ptys support output SIGIO..."); + + memset(buf, 0, sizeof(buf)); + + while(os_write_file(master, buf, sizeof(buf)) > 0) ; + if(errno != EAGAIN) + panic("check_sigio : write failed, errno = %d\n", errno); + while(((n = os_read_file(slave, buf, sizeof(buf))) > 0) && !got_sigio) ; + + if (got_sigio) { + printk("Yes\n"); + pty_output_sigio = 1; + } else if (n == -EAGAIN) { + printk("No, enabling workaround\n"); + } else { + panic("check_sigio : read failed, err = %d\n", n); + } +} + +static void tty_close(int master, int slave) +{ + printk("Checking that host ptys support SIGIO on close..."); + + os_close_file(slave); + if(got_sigio){ + printk("Yes\n"); + pty_close_sigio = 1; + } + else printk("No, enabling workaround\n"); +} + +void __init check_sigio(void) +{ + if((os_access("/dev/ptmx", OS_ACC_R_OK) < 0) && + (os_access("/dev/ptyp0", OS_ACC_R_OK) < 0)){ + printk("No pseudo-terminals available - skipping pty SIGIO " + "check\n"); + return; + } + check_one_sigio(tty_output); + check_one_sigio(tty_close); +} + +/* Protected by sigio_lock(), also used by sigio_cleanup, which is an + * exitcall. + */ +static int write_sigio_pid = -1; + +/* These arrays are initialized before the sigio thread is started, and + * the descriptors closed after it is killed. So, it can't see them change. + * On the UML side, they are changed under the sigio_lock. + */ +static int write_sigio_fds[2] = { -1, -1 }; +static int sigio_private[2] = { -1, -1 }; + +struct pollfds { + struct pollfd *poll; + int size; + int used; +}; + +/* Protected by sigio_lock(). Used by the sigio thread, but the UML thread + * synchronizes with it. + */ +struct pollfds current_poll = { + .poll = NULL, + .size = 0, + .used = 0 +}; + +struct pollfds next_poll = { + .poll = NULL, + .size = 0, + .used = 0 +}; + +static int write_sigio_thread(void *unused) +{ + struct pollfds *fds, tmp; + struct pollfd *p; + int i, n, respond_fd; + char c; + + fds = ¤t_poll; + while(1){ + n = poll(fds->poll, fds->used, -1); + if(n < 0){ + if(errno == EINTR) continue; + printk("write_sigio_thread : poll returned %d, " + "errno = %d\n", n, errno); + } + for(i = 0; i < fds->used; i++){ + p = &fds->poll[i]; + if(p->revents == 0) continue; + if(p->fd == sigio_private[1]){ + n = os_read_file(sigio_private[1], &c, sizeof(c)); + if(n != sizeof(c)) + printk("write_sigio_thread : " + "read failed, err = %d\n", -n); + tmp = current_poll; + current_poll = next_poll; + next_poll = tmp; + respond_fd = sigio_private[1]; + } + else { + respond_fd = write_sigio_fds[1]; + fds->used--; + memmove(&fds->poll[i], &fds->poll[i + 1], + (fds->used - i) * sizeof(*fds->poll)); + } + + n = os_write_file(respond_fd, &c, sizeof(c)); + if(n != sizeof(c)) + printk("write_sigio_thread : write failed, " + "err = %d\n", -n); + } + } +} + +static int need_poll(int n) +{ + if(n <= next_poll.size){ + next_poll.used = n; + return(0); + } + if(next_poll.poll != NULL) kfree(next_poll.poll); + next_poll.poll = um_kmalloc_atomic(n * sizeof(struct pollfd)); + if(next_poll.poll == NULL){ + printk("need_poll : failed to allocate new pollfds\n"); + next_poll.size = 0; + next_poll.used = 0; + return(-1); + } + next_poll.size = n; + next_poll.used = n; + return(0); +} + +/* Must be called with sigio_lock held, because it's needed by the marked + * critical section. */ +static void update_thread(void) +{ + unsigned long flags; + int n; + char c; + + flags = set_signals(0); + n = os_write_file(sigio_private[0], &c, sizeof(c)); + if(n != sizeof(c)){ + printk("update_thread : write failed, err = %d\n", -n); + goto fail; + } + + n = os_read_file(sigio_private[0], &c, sizeof(c)); + if(n != sizeof(c)){ + printk("update_thread : read failed, err = %d\n", -n); + goto fail; + } + + set_signals(flags); + return; + fail: + /* Critical section start */ + if(write_sigio_pid != -1) + os_kill_process(write_sigio_pid, 1); + write_sigio_pid = -1; + os_close_file(sigio_private[0]); + os_close_file(sigio_private[1]); + os_close_file(write_sigio_fds[0]); + os_close_file(write_sigio_fds[1]); + /* Critical section end */ + set_signals(flags); +} + +int add_sigio_fd(int fd, int read) +{ + int err = 0, i, n, events; + + sigio_lock(); + for(i = 0; i < current_poll.used; i++){ + if(current_poll.poll[i].fd == fd) + goto out; + } + + n = current_poll.used + 1; + err = need_poll(n); + if(err) + goto out; + + for(i = 0; i < current_poll.used; i++) + next_poll.poll[i] = current_poll.poll[i]; + + if(read) events = POLLIN; + else events = POLLOUT; + + next_poll.poll[n - 1] = ((struct pollfd) { .fd = fd, + .events = events, + .revents = 0 }); + update_thread(); + out: + sigio_unlock(); + return(err); +} + +int ignore_sigio_fd(int fd) +{ + struct pollfd *p; + int err = 0, i, n = 0; + + sigio_lock(); + for(i = 0; i < current_poll.used; i++){ + if(current_poll.poll[i].fd == fd) break; + } + if(i == current_poll.used) + goto out; + + err = need_poll(current_poll.used - 1); + if(err) + goto out; + + for(i = 0; i < current_poll.used; i++){ + p = ¤t_poll.poll[i]; + if(p->fd != fd) next_poll.poll[n++] = current_poll.poll[i]; + } + if(n == i){ + printk("ignore_sigio_fd : fd %d not found\n", fd); + err = -1; + goto out; + } + + update_thread(); + out: + sigio_unlock(); + return(err); +} + +static int setup_initial_poll(int fd) +{ + struct pollfd *p; + + p = um_kmalloc(sizeof(struct pollfd)); + if(p == NULL){ + printk("setup_initial_poll : failed to allocate poll\n"); + return(-1); + } + *p = ((struct pollfd) { .fd = fd, + .events = POLLIN, + .revents = 0 }); + current_poll = ((struct pollfds) { .poll = p, + .used = 1, + .size = 1 }); + return(0); +} + +void write_sigio_workaround(void) +{ + unsigned long stack; + int err; + + sigio_lock(); + if(write_sigio_pid != -1) + goto out; + + err = os_pipe(write_sigio_fds, 1, 1); + if(err < 0){ + printk("write_sigio_workaround - os_pipe 1 failed, " + "err = %d\n", -err); + goto out; + } + err = os_pipe(sigio_private, 1, 1); + if(err < 0){ + printk("write_sigio_workaround - os_pipe 2 failed, " + "err = %d\n", -err); + goto out_close1; + } + if(setup_initial_poll(sigio_private[1])) + goto out_close2; + + write_sigio_pid = run_helper_thread(write_sigio_thread, NULL, + CLONE_FILES | CLONE_VM, &stack, 0); + + if(write_sigio_pid < 0) goto out_close2; + + if(write_sigio_irq(write_sigio_fds[0])) + goto out_kill; + + out: + sigio_unlock(); + return; + + out_kill: + os_kill_process(write_sigio_pid, 1); + write_sigio_pid = -1; + out_close2: + os_close_file(sigio_private[0]); + os_close_file(sigio_private[1]); + out_close1: + os_close_file(write_sigio_fds[0]); + os_close_file(write_sigio_fds[1]); + sigio_unlock(); +} + +int read_sigio_fd(int fd) +{ + int n; + char c; + + n = os_read_file(fd, &c, sizeof(c)); + if(n != sizeof(c)){ + if(n < 0) { + printk("read_sigio_fd - read failed, err = %d\n", -n); + return(n); + } + else { + printk("read_sigio_fd - short read, bytes = %d\n", n); + return(-EIO); + } + } + return(n); +} + +static void sigio_cleanup(void) +{ + if (write_sigio_pid != -1) { + os_kill_process(write_sigio_pid, 1); + write_sigio_pid = -1; + } +} + +__uml_exitcall(sigio_cleanup); diff --git a/arch/um/kernel/signal_kern.c b/arch/um/kernel/signal_kern.c new file mode 100644 index 000000000000..7807a3e8c426 --- /dev/null +++ b/arch/um/kernel/signal_kern.c @@ -0,0 +1,213 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/stddef.h" +#include "linux/sys.h" +#include "linux/sched.h" +#include "linux/wait.h" +#include "linux/kernel.h" +#include "linux/smp_lock.h" +#include "linux/module.h" +#include "linux/slab.h" +#include "linux/tty.h" +#include "linux/binfmts.h" +#include "linux/ptrace.h" +#include "asm/signal.h" +#include "asm/uaccess.h" +#include "asm/unistd.h" +#include "user_util.h" +#include "asm/ucontext.h" +#include "kern_util.h" +#include "signal_kern.h" +#include "signal_user.h" +#include "kern.h" +#include "frame_kern.h" +#include "sigcontext.h" +#include "mode.h" + +EXPORT_SYMBOL(block_signals); +EXPORT_SYMBOL(unblock_signals); + +#define _S(nr) (1<<((nr)-1)) + +#define _BLOCKABLE (~(_S(SIGKILL) | _S(SIGSTOP))) + +/* + * OK, we're invoking a handler + */ +static int handle_signal(struct pt_regs *regs, unsigned long signr, + struct k_sigaction *ka, siginfo_t *info, + sigset_t *oldset) +{ + unsigned long sp; + int err; + + /* Always make any pending restarted system calls return -EINTR */ + current_thread_info()->restart_block.fn = do_no_restart_syscall; + + /* Did we come from a system call? */ + if(PT_REGS_SYSCALL_NR(regs) >= 0){ + /* If so, check system call restarting.. */ + switch(PT_REGS_SYSCALL_RET(regs)){ + case -ERESTART_RESTARTBLOCK: + case -ERESTARTNOHAND: + PT_REGS_SYSCALL_RET(regs) = -EINTR; + break; + + case -ERESTARTSYS: + if (!(ka->sa.sa_flags & SA_RESTART)) { + PT_REGS_SYSCALL_RET(regs) = -EINTR; + break; + } + /* fallthrough */ + case -ERESTARTNOINTR: + PT_REGS_RESTART_SYSCALL(regs); + PT_REGS_ORIG_SYSCALL(regs) = PT_REGS_SYSCALL_NR(regs); + break; + } + } + + sp = PT_REGS_SP(regs); + if((ka->sa.sa_flags & SA_ONSTACK) && (sas_ss_flags(sp) == 0)) + sp = current->sas_ss_sp + current->sas_ss_size; + +#ifdef CONFIG_ARCH_HAS_SC_SIGNALS + if(!(ka->sa.sa_flags & SA_SIGINFO)) + err = setup_signal_stack_sc(sp, signr, ka, regs, oldset); + else +#endif + err = setup_signal_stack_si(sp, signr, ka, regs, info, oldset); + + if(err){ + spin_lock_irq(¤t->sighand->siglock); + current->blocked = *oldset; + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + force_sigsegv(signr, current); + } + else if(!(ka->sa.sa_flags & SA_NODEFER)){ + spin_lock_irq(¤t->sighand->siglock); + sigorsets(¤t->blocked, ¤t->blocked, + &ka->sa.sa_mask); + sigaddset(¤t->blocked, signr); + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + } + + return err; +} + +static int kern_do_signal(struct pt_regs *regs, sigset_t *oldset) +{ + struct k_sigaction ka_copy; + siginfo_t info; + int sig, handled_sig = 0; + + while((sig = get_signal_to_deliver(&info, &ka_copy, regs, NULL)) > 0){ + handled_sig = 1; + /* Whee! Actually deliver the signal. */ + if(!handle_signal(regs, sig, &ka_copy, &info, oldset)) + break; + } + + /* Did we come from a system call? */ + if(!handled_sig && (PT_REGS_SYSCALL_NR(regs) >= 0)){ + /* Restart the system call - no handlers present */ + if(PT_REGS_SYSCALL_RET(regs) == -ERESTARTNOHAND || + PT_REGS_SYSCALL_RET(regs) == -ERESTARTSYS || + PT_REGS_SYSCALL_RET(regs) == -ERESTARTNOINTR){ + PT_REGS_ORIG_SYSCALL(regs) = PT_REGS_SYSCALL_NR(regs); + PT_REGS_RESTART_SYSCALL(regs); + } + else if(PT_REGS_SYSCALL_RET(regs) == -ERESTART_RESTARTBLOCK){ + PT_REGS_SYSCALL_RET(regs) = __NR_restart_syscall; + PT_REGS_RESTART_SYSCALL(regs); + } + } + + /* This closes a way to execute a system call on the host. If + * you set a breakpoint on a system call instruction and singlestep + * from it, the tracing thread used to PTRACE_SINGLESTEP the process + * rather than PTRACE_SYSCALL it, allowing the system call to execute + * on the host. The tracing thread will check this flag and + * PTRACE_SYSCALL if necessary. + */ + if(current->ptrace & PT_DTRACE) + current->thread.singlestep_syscall = + is_syscall(PT_REGS_IP(¤t->thread.regs)); + return(handled_sig); +} + +int do_signal(void) +{ + return(kern_do_signal(¤t->thread.regs, ¤t->blocked)); +} + +/* + * Atomically swap in the new signal mask, and wait for a signal. + */ +long sys_sigsuspend(int history0, int history1, old_sigset_t mask) +{ + sigset_t saveset; + + mask &= _BLOCKABLE; + spin_lock_irq(¤t->sighand->siglock); + saveset = current->blocked; + siginitset(¤t->blocked, mask); + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + + PT_REGS_SYSCALL_RET(¤t->thread.regs) = -EINTR; + while (1) { + current->state = TASK_INTERRUPTIBLE; + schedule(); + if(kern_do_signal(¤t->thread.regs, &saveset)) + return(-EINTR); + } +} + +long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize) +{ + sigset_t saveset, newset; + + /* XXX: Don't preclude handling different sized sigset_t's. */ + if (sigsetsize != sizeof(sigset_t)) + return -EINVAL; + + if (copy_from_user(&newset, unewset, sizeof(newset))) + return -EFAULT; + sigdelsetmask(&newset, ~_BLOCKABLE); + + spin_lock_irq(¤t->sighand->siglock); + saveset = current->blocked; + current->blocked = newset; + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + + PT_REGS_SYSCALL_RET(¤t->thread.regs) = -EINTR; + while (1) { + current->state = TASK_INTERRUPTIBLE; + schedule(); + if (kern_do_signal(¤t->thread.regs, &saveset)) + return(-EINTR); + } +} + +long sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss) +{ + return(do_sigaltstack(uss, uoss, PT_REGS_SP(¤t->thread.regs))); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/signal_user.c b/arch/um/kernel/signal_user.c new file mode 100644 index 000000000000..62f457835fb1 --- /dev/null +++ b/arch/um/kernel/signal_user.c @@ -0,0 +1,157 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <unistd.h> +#include <stdlib.h> +#include <signal.h> +#include <errno.h> +#include <stdarg.h> +#include <string.h> +#include <sys/mman.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "signal_user.h" +#include "signal_kern.h" +#include "sysdep/sigcontext.h" +#include "sigcontext.h" + +void set_sigstack(void *sig_stack, int size) +{ + stack_t stack = ((stack_t) { .ss_flags = 0, + .ss_sp = (__ptr_t) sig_stack, + .ss_size = size - sizeof(void *) }); + + if(sigaltstack(&stack, NULL) != 0) + panic("enabling signal stack failed, errno = %d\n", errno); +} + +void set_handler(int sig, void (*handler)(int), int flags, ...) +{ + struct sigaction action; + va_list ap; + int mask; + + va_start(ap, flags); + action.sa_handler = handler; + sigemptyset(&action.sa_mask); + while((mask = va_arg(ap, int)) != -1){ + sigaddset(&action.sa_mask, mask); + } + va_end(ap); + action.sa_flags = flags; + action.sa_restorer = NULL; + if(sigaction(sig, &action, NULL) < 0) + panic("sigaction failed"); +} + +int change_sig(int signal, int on) +{ + sigset_t sigset, old; + + sigemptyset(&sigset); + sigaddset(&sigset, signal); + sigprocmask(on ? SIG_UNBLOCK : SIG_BLOCK, &sigset, &old); + return(!sigismember(&old, signal)); +} + +/* Both here and in set/get_signal we don't touch SIGPROF, because we must not + * disable profiling; it's safe because the profiling code does not interact + * with the kernel code at all.*/ + +static void change_signals(int type) +{ + sigset_t mask; + + sigemptyset(&mask); + sigaddset(&mask, SIGVTALRM); + sigaddset(&mask, SIGALRM); + sigaddset(&mask, SIGIO); + if(sigprocmask(type, &mask, NULL) < 0) + panic("Failed to change signal mask - errno = %d", errno); +} + +void block_signals(void) +{ + change_signals(SIG_BLOCK); +} + +void unblock_signals(void) +{ + change_signals(SIG_UNBLOCK); +} + +/* These are the asynchronous signals. SIGVTALRM and SIGARLM are handled + * together under SIGVTALRM_BIT. SIGPROF is excluded because we want to + * be able to profile all of UML, not just the non-critical sections. If + * profiling is not thread-safe, then that is not my problem. We can disable + * profiling when SMP is enabled in that case. + */ +#define SIGIO_BIT 0 +#define SIGVTALRM_BIT 1 + +static int enable_mask(sigset_t *mask) +{ + int sigs; + + sigs = sigismember(mask, SIGIO) ? 0 : 1 << SIGIO_BIT; + sigs |= sigismember(mask, SIGVTALRM) ? 0 : 1 << SIGVTALRM_BIT; + sigs |= sigismember(mask, SIGALRM) ? 0 : 1 << SIGVTALRM_BIT; + return(sigs); +} + +int get_signals(void) +{ + sigset_t mask; + + if(sigprocmask(SIG_SETMASK, NULL, &mask) < 0) + panic("Failed to get signal mask"); + return(enable_mask(&mask)); +} + +int set_signals(int enable) +{ + sigset_t mask; + int ret; + + sigemptyset(&mask); + if(enable & (1 << SIGIO_BIT)) + sigaddset(&mask, SIGIO); + if(enable & (1 << SIGVTALRM_BIT)){ + sigaddset(&mask, SIGVTALRM); + sigaddset(&mask, SIGALRM); + } + + /* This is safe - sigprocmask is guaranteed to copy locally the + * value of new_set, do his work and then, at the end, write to + * old_set. + */ + if(sigprocmask(SIG_UNBLOCK, &mask, &mask) < 0) + panic("Failed to enable signals"); + ret = enable_mask(&mask); + sigemptyset(&mask); + if((enable & (1 << SIGIO_BIT)) == 0) + sigaddset(&mask, SIGIO); + if((enable & (1 << SIGVTALRM_BIT)) == 0){ + sigaddset(&mask, SIGVTALRM); + sigaddset(&mask, SIGALRM); + } + if(sigprocmask(SIG_BLOCK, &mask, NULL) < 0) + panic("Failed to block signals"); + + return(ret); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/Makefile b/arch/um/kernel/skas/Makefile new file mode 100644 index 000000000000..d37d1bfcd6f7 --- /dev/null +++ b/arch/um/kernel/skas/Makefile @@ -0,0 +1,13 @@ +# +# Copyright (C) 2002 - 2004 Jeff Dike (jdike@addtoit.com) +# Licensed under the GPL +# + +obj-y := exec_kern.o mem.o mem_user.o mmu.o process.o process_kern.o \ + syscall_kern.o syscall_user.o time.o tlb.o trap_user.o uaccess.o \ + +subdir- := util + +USER_OBJS := process.o time.o + +include arch/um/scripts/Makefile.rules diff --git a/arch/um/kernel/skas/exec_kern.c b/arch/um/kernel/skas/exec_kern.c new file mode 100644 index 000000000000..c6b4d5dba789 --- /dev/null +++ b/arch/um/kernel/skas/exec_kern.c @@ -0,0 +1,41 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/kernel.h" +#include "asm/current.h" +#include "asm/page.h" +#include "asm/signal.h" +#include "asm/ptrace.h" +#include "asm/uaccess.h" +#include "asm/mmu_context.h" +#include "tlb.h" +#include "skas.h" +#include "um_mmu.h" +#include "os.h" + +void flush_thread_skas(void) +{ + force_flush_all(); + switch_mm_skas(current->mm->context.skas.mm_fd); +} + +void start_thread_skas(struct pt_regs *regs, unsigned long eip, + unsigned long esp) +{ + set_fs(USER_DS); + PT_REGS_IP(regs) = eip; + PT_REGS_SP(regs) = esp; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/include/mmu-skas.h b/arch/um/kernel/skas/include/mmu-skas.h new file mode 100644 index 000000000000..4cd60d7213f3 --- /dev/null +++ b/arch/um/kernel/skas/include/mmu-skas.h @@ -0,0 +1,24 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_MMU_H +#define __SKAS_MMU_H + +struct mmu_context_skas { + int mm_fd; +}; + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/include/mode-skas.h b/arch/um/kernel/skas/include/mode-skas.h new file mode 100644 index 000000000000..c1e33bd788db --- /dev/null +++ b/arch/um/kernel/skas/include/mode-skas.h @@ -0,0 +1,34 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __MODE_SKAS_H__ +#define __MODE_SKAS_H__ + +#include <sysdep/ptrace.h> + +extern unsigned long exec_regs[]; +extern unsigned long exec_fp_regs[]; +extern unsigned long exec_fpx_regs[]; +extern int have_fpx_regs; + +extern void user_time_init_skas(void); +extern void sig_handler_common_skas(int sig, void *sc_ptr); +extern void halt_skas(void); +extern void reboot_skas(void); +extern void kill_off_processes_skas(void); +extern int is_skas_winch(int pid, int fd, void *data); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/include/mode_kern-skas.h b/arch/um/kernel/skas/include/mode_kern-skas.h new file mode 100644 index 000000000000..94c564962378 --- /dev/null +++ b/arch/um/kernel/skas/include/mode_kern-skas.h @@ -0,0 +1,53 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_MODE_KERN_H__ +#define __SKAS_MODE_KERN_H__ + +#include "linux/sched.h" +#include "asm/page.h" +#include "asm/ptrace.h" + +extern void flush_thread_skas(void); +extern void *switch_to_skas(void *prev, void *next); +extern void start_thread_skas(struct pt_regs *regs, unsigned long eip, + unsigned long esp); +extern int copy_thread_skas(int nr, unsigned long clone_flags, + unsigned long sp, unsigned long stack_top, + struct task_struct *p, struct pt_regs *regs); +extern void release_thread_skas(struct task_struct *task); +extern void exit_thread_skas(void); +extern void initial_thread_cb_skas(void (*proc)(void *), void *arg); +extern void init_idle_skas(void); +extern void flush_tlb_kernel_range_skas(unsigned long start, + unsigned long end); +extern void flush_tlb_kernel_vm_skas(void); +extern void __flush_tlb_one_skas(unsigned long addr); +extern void flush_tlb_range_skas(struct vm_area_struct *vma, + unsigned long start, unsigned long end); +extern void flush_tlb_mm_skas(struct mm_struct *mm); +extern void force_flush_all_skas(void); +extern long execute_syscall_skas(void *r); +extern void before_mem_skas(unsigned long unused); +extern unsigned long set_task_sizes_skas(int arg, unsigned long *host_size_out, + unsigned long *task_size_out); +extern int start_uml_skas(void); +extern int external_pid_skas(struct task_struct *task); +extern int thread_pid_skas(struct task_struct *task); + +#define kmem_end_skas (host_task_size - 1024 * 1024) + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/include/proc_mm.h b/arch/um/kernel/skas/include/proc_mm.h new file mode 100644 index 000000000000..cce61a679052 --- /dev/null +++ b/arch/um/kernel/skas/include/proc_mm.h @@ -0,0 +1,55 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_PROC_MM_H +#define __SKAS_PROC_MM_H + +#define MM_MMAP 54 +#define MM_MUNMAP 55 +#define MM_MPROTECT 56 +#define MM_COPY_SEGMENTS 57 + +struct mm_mmap { + unsigned long addr; + unsigned long len; + unsigned long prot; + unsigned long flags; + unsigned long fd; + unsigned long offset; +}; + +struct mm_munmap { + unsigned long addr; + unsigned long len; +}; + +struct mm_mprotect { + unsigned long addr; + unsigned long len; + unsigned int prot; +}; + +struct proc_mm_op { + int op; + union { + struct mm_mmap mmap; + struct mm_munmap munmap; + struct mm_mprotect mprotect; + int copy_segments; + } u; +}; + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/include/skas.h b/arch/um/kernel/skas/include/skas.h new file mode 100644 index 000000000000..f0702c2c7204 --- /dev/null +++ b/arch/um/kernel/skas/include/skas.h @@ -0,0 +1,46 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_H +#define __SKAS_H + +#include "sysdep/ptrace.h" + +extern int userspace_pid[]; + +extern void switch_threads(void *me, void *next); +extern void thread_wait(void *sw, void *fb); +extern void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr, + void (*handler)(int)); +extern int start_idle_thread(void *stack, void *switch_buf_ptr, + void **fork_buf_ptr); +extern int user_thread(unsigned long stack, int flags); +extern void userspace(union uml_pt_regs *regs); +extern void new_thread_proc(void *stack, void (*handler)(int sig)); +extern void remove_sigstack(void); +extern void new_thread_handler(int sig); +extern void handle_syscall(union uml_pt_regs *regs); +extern void map(int fd, unsigned long virt, unsigned long len, int r, int w, + int x, int phys_fd, unsigned long long offset); +extern int unmap(int fd, void *addr, unsigned long len); +extern int protect(int fd, unsigned long addr, unsigned long len, + int r, int w, int x); +extern void user_signal(int sig, union uml_pt_regs *regs); +extern int new_mm(int from); +extern void start_userspace(int cpu); +extern long execute_syscall_skas(void *r); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/include/uaccess-skas.h b/arch/um/kernel/skas/include/uaccess-skas.h new file mode 100644 index 000000000000..11986c9b9ddf --- /dev/null +++ b/arch/um/kernel/skas/include/uaccess-skas.h @@ -0,0 +1,45 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_UACCESS_H +#define __SKAS_UACCESS_H + +#include "asm/errno.h" +#include "asm/fixmap.h" + +#define access_ok_skas(type, addr, size) \ + ((segment_eq(get_fs(), KERNEL_DS)) || \ + (((unsigned long) (addr) < TASK_SIZE) && \ + ((unsigned long) (addr) + (size) <= TASK_SIZE)) || \ + ((type == VERIFY_READ ) && \ + ((unsigned long) (addr) >= FIXADDR_USER_START) && \ + ((unsigned long) (addr) + (size) <= FIXADDR_USER_END) && \ + ((unsigned long) (addr) + (size) >= (unsigned long)(addr)))) + +static inline int verify_area_skas(int type, const void * addr, + unsigned long size) +{ + return(access_ok_skas(type, addr, size) ? 0 : -EFAULT); +} + +extern int copy_from_user_skas(void *to, const void *from, int n); +extern int copy_to_user_skas(void *to, const void *from, int n); +extern int strncpy_from_user_skas(char *dst, const char *src, int count); +extern int __clear_user_skas(void *mem, int len); +extern int clear_user_skas(void *mem, int len); +extern int strnlen_user_skas(const void *str, int len); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/mem.c b/arch/um/kernel/skas/mem.c new file mode 100644 index 000000000000..438db2f43456 --- /dev/null +++ b/arch/um/kernel/skas/mem.c @@ -0,0 +1,35 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/mm.h" +#include "mem_user.h" + +unsigned long set_task_sizes_skas(int arg, unsigned long *host_size_out, + unsigned long *task_size_out) +{ + /* Round up to the nearest 4M */ + unsigned long top = ROUND_4M((unsigned long) &arg); + +#ifdef CONFIG_HOST_TASK_SIZE + *host_size_out = CONFIG_HOST_TASK_SIZE; + *task_size_out = CONFIG_HOST_TASK_SIZE; +#else + *host_size_out = top; + *task_size_out = top; +#endif + return(((unsigned long) set_task_sizes_skas) & ~0xffffff); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/mem_user.c b/arch/um/kernel/skas/mem_user.c new file mode 100644 index 000000000000..1310bf1e88d1 --- /dev/null +++ b/arch/um/kernel/skas/mem_user.c @@ -0,0 +1,102 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <errno.h> +#include <sys/mman.h> +#include "mem_user.h" +#include "mem.h" +#include "user.h" +#include "os.h" +#include "proc_mm.h" + +void map(int fd, unsigned long virt, unsigned long len, int r, int w, + int x, int phys_fd, unsigned long long offset) +{ + struct proc_mm_op map; + int prot, n; + + prot = (r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) | + (x ? PROT_EXEC : 0); + + map = ((struct proc_mm_op) { .op = MM_MMAP, + .u = + { .mmap = + { .addr = virt, + .len = len, + .prot = prot, + .flags = MAP_SHARED | + MAP_FIXED, + .fd = phys_fd, + .offset = offset + } } } ); + n = os_write_file(fd, &map, sizeof(map)); + if(n != sizeof(map)) + printk("map : /proc/mm map failed, err = %d\n", -n); +} + +int unmap(int fd, void *addr, unsigned long len) +{ + struct proc_mm_op unmap; + int n; + + unmap = ((struct proc_mm_op) { .op = MM_MUNMAP, + .u = + { .munmap = + { .addr = (unsigned long) addr, + .len = len } } } ); + n = os_write_file(fd, &unmap, sizeof(unmap)); + if(n != sizeof(unmap)) { + if(n < 0) + return(n); + else if(n > 0) + return(-EIO); + } + + return(0); +} + +int protect(int fd, unsigned long addr, unsigned long len, int r, int w, + int x, int must_succeed) +{ + struct proc_mm_op protect; + int prot, n; + + prot = (r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) | + (x ? PROT_EXEC : 0); + + protect = ((struct proc_mm_op) { .op = MM_MPROTECT, + .u = + { .mprotect = + { .addr = (unsigned long) addr, + .len = len, + .prot = prot } } } ); + + n = os_write_file(fd, &protect, sizeof(protect)); + if(n != sizeof(protect)) { + if(n == 0) return(0); + + if(must_succeed) + panic("protect failed, err = %d", -n); + + return(-EIO); + } + + return(0); +} + +void before_mem_skas(unsigned long unused) +{ +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/mmu.c b/arch/um/kernel/skas/mmu.c new file mode 100644 index 000000000000..6cb9a6d028a9 --- /dev/null +++ b/arch/um/kernel/skas/mmu.c @@ -0,0 +1,48 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/list.h" +#include "linux/spinlock.h" +#include "linux/slab.h" +#include "asm/current.h" +#include "asm/segment.h" +#include "asm/mmu.h" +#include "os.h" +#include "skas.h" + +int init_new_context_skas(struct task_struct *task, struct mm_struct *mm) +{ + int from; + + if((current->mm != NULL) && (current->mm != &init_mm)) + from = current->mm->context.skas.mm_fd; + else from = -1; + + mm->context.skas.mm_fd = new_mm(from); + if(mm->context.skas.mm_fd < 0){ + printk("init_new_context_skas - new_mm failed, errno = %d\n", + mm->context.skas.mm_fd); + return(mm->context.skas.mm_fd); + } + + return(0); +} + +void destroy_context_skas(struct mm_struct *mm) +{ + os_close_file(mm->context.skas.mm_fd); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/process.c b/arch/um/kernel/skas/process.c new file mode 100644 index 000000000000..b4ffaaa81241 --- /dev/null +++ b/arch/um/kernel/skas/process.c @@ -0,0 +1,339 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <unistd.h> +#include <errno.h> +#include <signal.h> +#include <setjmp.h> +#include <sched.h> +#include <sys/wait.h> +#include <sys/mman.h> +#include <sys/user.h> +#include <asm/unistd.h> +#include "user.h" +#include "ptrace_user.h" +#include "time_user.h" +#include "sysdep/ptrace.h" +#include "user_util.h" +#include "kern_util.h" +#include "skas.h" +#include "sysdep/sigcontext.h" +#include "os.h" +#include "proc_mm.h" +#include "skas_ptrace.h" +#include "chan_user.h" +#include "signal_user.h" +#include "registers.h" + +int is_skas_winch(int pid, int fd, void *data) +{ + if(pid != os_getpid()) + return(0); + + register_winch_irq(-1, fd, -1, data); + return(1); +} + +static void handle_segv(int pid) +{ + struct ptrace_faultinfo fault; + int err; + + err = ptrace(PTRACE_FAULTINFO, pid, 0, &fault); + if(err) + panic("handle_segv - PTRACE_FAULTINFO failed, errno = %d\n", + errno); + + segv(fault.addr, 0, FAULT_WRITE(fault.is_write), 1, NULL); +} + +/*To use the same value of using_sysemu as the caller, ask it that value (in local_using_sysemu)*/ +static void handle_trap(int pid, union uml_pt_regs *regs, int local_using_sysemu) +{ + int err, status; + + /* Mark this as a syscall */ + UPT_SYSCALL_NR(regs) = PT_SYSCALL_NR(regs->skas.regs); + + if (!local_using_sysemu) + { + err = ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET, __NR_getpid); + if(err < 0) + panic("handle_trap - nullifying syscall failed errno = %d\n", + errno); + + err = ptrace(PTRACE_SYSCALL, pid, 0, 0); + if(err < 0) + panic("handle_trap - continuing to end of syscall failed, " + "errno = %d\n", errno); + + CATCH_EINTR(err = waitpid(pid, &status, WUNTRACED)); + if((err < 0) || !WIFSTOPPED(status) || + (WSTOPSIG(status) != SIGTRAP + 0x80)) + panic("handle_trap - failed to wait at end of syscall, " + "errno = %d, status = %d\n", errno, status); + } + + handle_syscall(regs); +} + +static int userspace_tramp(void *arg) +{ + init_new_thread_signals(0); + enable_timer(); + ptrace(PTRACE_TRACEME, 0, 0, 0); + os_stop_process(os_getpid()); + return(0); +} + +/* Each element set once, and only accessed by a single processor anyway */ +#undef NR_CPUS +#define NR_CPUS 1 +int userspace_pid[NR_CPUS]; + +void start_userspace(int cpu) +{ + void *stack; + unsigned long sp; + int pid, status, n; + + stack = mmap(NULL, PAGE_SIZE, PROT_READ | PROT_WRITE | PROT_EXEC, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + if(stack == MAP_FAILED) + panic("start_userspace : mmap failed, errno = %d", errno); + sp = (unsigned long) stack + PAGE_SIZE - sizeof(void *); + + pid = clone(userspace_tramp, (void *) sp, + CLONE_FILES | CLONE_VM | SIGCHLD, NULL); + if(pid < 0) + panic("start_userspace : clone failed, errno = %d", errno); + + do { + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + if(n < 0) + panic("start_userspace : wait failed, errno = %d", + errno); + } while(WIFSTOPPED(status) && (WSTOPSIG(status) == SIGVTALRM)); + + if(!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGSTOP)) + panic("start_userspace : expected SIGSTOP, got status = %d", + status); + + if (ptrace(PTRACE_OLDSETOPTIONS, pid, NULL, (void *)PTRACE_O_TRACESYSGOOD) < 0) + panic("start_userspace : PTRACE_SETOPTIONS failed, errno=%d\n", + errno); + + if(munmap(stack, PAGE_SIZE) < 0) + panic("start_userspace : munmap failed, errno = %d\n", errno); + + userspace_pid[cpu] = pid; +} + +void userspace(union uml_pt_regs *regs) +{ + int err, status, op, pid = userspace_pid[0]; + int local_using_sysemu; /*To prevent races if using_sysemu changes under us.*/ + + while(1){ + restore_registers(pid, regs); + + /* Now we set local_using_sysemu to be used for one loop */ + local_using_sysemu = get_using_sysemu(); + + op = SELECT_PTRACE_OPERATION(local_using_sysemu, singlestepping(NULL)); + + err = ptrace(op, pid, 0, 0); + if(err) + panic("userspace - could not resume userspace process, " + "pid=%d, ptrace operation = %d, errno = %d\n", + op, errno); + + CATCH_EINTR(err = waitpid(pid, &status, WUNTRACED)); + if(err < 0) + panic("userspace - waitpid failed, errno = %d\n", + errno); + + regs->skas.is_user = 1; + save_registers(pid, regs); + UPT_SYSCALL_NR(regs) = -1; /* Assume: It's not a syscall */ + + if(WIFSTOPPED(status)){ + switch(WSTOPSIG(status)){ + case SIGSEGV: + handle_segv(pid); + break; + case SIGTRAP + 0x80: + handle_trap(pid, regs, local_using_sysemu); + break; + case SIGTRAP: + relay_signal(SIGTRAP, regs); + break; + case SIGIO: + case SIGVTALRM: + case SIGILL: + case SIGBUS: + case SIGFPE: + case SIGWINCH: + user_signal(WSTOPSIG(status), regs); + break; + default: + printk("userspace - child stopped with signal " + "%d\n", WSTOPSIG(status)); + } + interrupt_end(); + + /* Avoid -ERESTARTSYS handling in host */ + PT_SYSCALL_NR(regs->skas.regs) = -1; + } + } +} + +void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr, + void (*handler)(int)) +{ + unsigned long flags; + sigjmp_buf switch_buf, fork_buf; + + *switch_buf_ptr = &switch_buf; + *fork_buf_ptr = &fork_buf; + + /* Somewhat subtle - siglongjmp restores the signal mask before doing + * the longjmp. This means that when jumping from one stack to another + * when the target stack has interrupts enabled, an interrupt may occur + * on the source stack. This is bad when starting up a process because + * it's not supposed to get timer ticks until it has been scheduled. + * So, we disable interrupts around the sigsetjmp to ensure that + * they can't happen until we get back here where they are safe. + */ + flags = get_signals(); + block_signals(); + if(sigsetjmp(fork_buf, 1) == 0) + new_thread_proc(stack, handler); + + remove_sigstack(); + + set_signals(flags); +} + +void thread_wait(void *sw, void *fb) +{ + sigjmp_buf buf, **switch_buf = sw, *fork_buf; + + *switch_buf = &buf; + fork_buf = fb; + if(sigsetjmp(buf, 1) == 0) + siglongjmp(*fork_buf, 1); +} + +void switch_threads(void *me, void *next) +{ + sigjmp_buf my_buf, **me_ptr = me, *next_buf = next; + + *me_ptr = &my_buf; + if(sigsetjmp(my_buf, 1) == 0) + siglongjmp(*next_buf, 1); +} + +static sigjmp_buf initial_jmpbuf; + +/* XXX Make these percpu */ +static void (*cb_proc)(void *arg); +static void *cb_arg; +static sigjmp_buf *cb_back; + +int start_idle_thread(void *stack, void *switch_buf_ptr, void **fork_buf_ptr) +{ + sigjmp_buf **switch_buf = switch_buf_ptr; + int n; + + *fork_buf_ptr = &initial_jmpbuf; + n = sigsetjmp(initial_jmpbuf, 1); + if(n == 0) + new_thread_proc((void *) stack, new_thread_handler); + else if(n == 1) + remove_sigstack(); + else if(n == 2){ + (*cb_proc)(cb_arg); + siglongjmp(*cb_back, 1); + } + else if(n == 3){ + kmalloc_ok = 0; + return(0); + } + else if(n == 4){ + kmalloc_ok = 0; + return(1); + } + siglongjmp(**switch_buf, 1); +} + +void remove_sigstack(void) +{ + stack_t stack = ((stack_t) { .ss_flags = SS_DISABLE, + .ss_sp = NULL, + .ss_size = 0 }); + + if(sigaltstack(&stack, NULL) != 0) + panic("disabling signal stack failed, errno = %d\n", errno); +} + +void initial_thread_cb_skas(void (*proc)(void *), void *arg) +{ + sigjmp_buf here; + + cb_proc = proc; + cb_arg = arg; + cb_back = &here; + + block_signals(); + if(sigsetjmp(here, 1) == 0) + siglongjmp(initial_jmpbuf, 2); + unblock_signals(); + + cb_proc = NULL; + cb_arg = NULL; + cb_back = NULL; +} + +void halt_skas(void) +{ + block_signals(); + siglongjmp(initial_jmpbuf, 3); +} + +void reboot_skas(void) +{ + block_signals(); + siglongjmp(initial_jmpbuf, 4); +} + +void switch_mm_skas(int mm_fd) +{ + int err; + +#warning need cpu pid in switch_mm_skas + err = ptrace(PTRACE_SWITCH_MM, userspace_pid[0], 0, mm_fd); + if(err) + panic("switch_mm_skas - PTRACE_SWITCH_MM failed, errno = %d\n", + errno); +} + +void kill_off_processes_skas(void) +{ +#warning need to loop over userspace_pids in kill_off_processes_skas + os_kill_ptraced_process(userspace_pid[0], 1); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/process_kern.c b/arch/um/kernel/skas/process_kern.c new file mode 100644 index 000000000000..5d096ea63b97 --- /dev/null +++ b/arch/um/kernel/skas/process_kern.c @@ -0,0 +1,213 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/slab.h" +#include "linux/ptrace.h" +#include "linux/proc_fs.h" +#include "linux/file.h" +#include "linux/errno.h" +#include "linux/init.h" +#include "asm/uaccess.h" +#include "asm/atomic.h" +#include "kern_util.h" +#include "time_user.h" +#include "signal_user.h" +#include "skas.h" +#include "os.h" +#include "user_util.h" +#include "tlb.h" +#include "kern.h" +#include "mode.h" +#include "proc_mm.h" +#include "registers.h" + +void *switch_to_skas(void *prev, void *next) +{ + struct task_struct *from, *to; + + from = prev; + to = next; + + /* XXX need to check runqueues[cpu].idle */ + if(current->pid == 0) + switch_timers(0); + + to->thread.prev_sched = from; + set_current(to); + + switch_threads(&from->thread.mode.skas.switch_buf, + to->thread.mode.skas.switch_buf); + + if(current->pid == 0) + switch_timers(1); + + return(current->thread.prev_sched); +} + +extern void schedule_tail(struct task_struct *prev); + +void new_thread_handler(int sig) +{ + int (*fn)(void *), n; + void *arg; + + fn = current->thread.request.u.thread.proc; + arg = current->thread.request.u.thread.arg; + change_sig(SIGUSR1, 1); + thread_wait(¤t->thread.mode.skas.switch_buf, + current->thread.mode.skas.fork_buf); + + if(current->thread.prev_sched != NULL) + schedule_tail(current->thread.prev_sched); + current->thread.prev_sched = NULL; + + /* The return value is 1 if the kernel thread execs a process, + * 0 if it just exits + */ + n = run_kernel_thread(fn, arg, ¤t->thread.exec_buf); + if(n == 1) + userspace(¤t->thread.regs.regs); + else do_exit(0); +} + +void new_thread_proc(void *stack, void (*handler)(int sig)) +{ + init_new_thread_stack(stack, handler); + os_usr1_process(os_getpid()); +} + +void release_thread_skas(struct task_struct *task) +{ +} + +void exit_thread_skas(void) +{ +} + +void fork_handler(int sig) +{ + change_sig(SIGUSR1, 1); + thread_wait(¤t->thread.mode.skas.switch_buf, + current->thread.mode.skas.fork_buf); + + force_flush_all(); + if(current->thread.prev_sched == NULL) + panic("blech"); + + schedule_tail(current->thread.prev_sched); + current->thread.prev_sched = NULL; + + userspace(¤t->thread.regs.regs); +} + +int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp, + unsigned long stack_top, struct task_struct * p, + struct pt_regs *regs) +{ + void (*handler)(int); + + if(current->thread.forking){ + memcpy(&p->thread.regs.regs.skas, + ¤t->thread.regs.regs.skas, + sizeof(p->thread.regs.regs.skas)); + REGS_SET_SYSCALL_RETURN(p->thread.regs.regs.skas.regs, 0); + if(sp != 0) REGS_SP(p->thread.regs.regs.skas.regs) = sp; + + handler = fork_handler; + } + else { + init_thread_registers(&p->thread.regs.regs); + p->thread.request.u.thread = current->thread.request.u.thread; + handler = new_thread_handler; + } + + new_thread(p->thread_info, &p->thread.mode.skas.switch_buf, + &p->thread.mode.skas.fork_buf, handler); + return(0); +} + +int new_mm(int from) +{ + struct proc_mm_op copy; + int n, fd; + + fd = os_open_file("/proc/mm", of_cloexec(of_write(OPENFLAGS())), 0); + if(fd < 0) + return(fd); + + if(from != -1){ + copy = ((struct proc_mm_op) { .op = MM_COPY_SEGMENTS, + .u = + { .copy_segments = from } } ); + n = os_write_file(fd, ©, sizeof(copy)); + if(n != sizeof(copy)) + printk("new_mm : /proc/mm copy_segments failed, " + "err = %d\n", -n); + } + + return(fd); +} + +void init_idle_skas(void) +{ + cpu_tasks[current_thread->cpu].pid = os_getpid(); + default_idle(); +} + +extern void start_kernel(void); + +static int start_kernel_proc(void *unused) +{ + int pid; + + block_signals(); + pid = os_getpid(); + + cpu_tasks[0].pid = pid; + cpu_tasks[0].task = current; +#ifdef CONFIG_SMP + cpu_online_map = cpumask_of_cpu(0); +#endif + start_kernel(); + return(0); +} + +int start_uml_skas(void) +{ + start_userspace(0); + + init_new_thread_signals(1); + uml_idle_timer(); + + init_task.thread.request.u.thread.proc = start_kernel_proc; + init_task.thread.request.u.thread.arg = NULL; + return(start_idle_thread(init_task.thread_info, + &init_task.thread.mode.skas.switch_buf, + &init_task.thread.mode.skas.fork_buf)); +} + +int external_pid_skas(struct task_struct *task) +{ +#warning Need to look up userspace_pid by cpu + return(userspace_pid[0]); +} + +int thread_pid_skas(struct task_struct *task) +{ +#warning Need to look up userspace_pid by cpu + return(userspace_pid[0]); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/syscall_kern.c b/arch/um/kernel/skas/syscall_kern.c new file mode 100644 index 000000000000..bdf040ce5b8e --- /dev/null +++ b/arch/um/kernel/skas/syscall_kern.c @@ -0,0 +1,43 @@ +/* + * Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/sys.h" +#include "linux/ptrace.h" +#include "asm/errno.h" +#include "asm/unistd.h" +#include "asm/ptrace.h" +#include "asm/current.h" +#include "sysdep/syscalls.h" +#include "kern_util.h" + +extern syscall_handler_t *sys_call_table[]; + +long execute_syscall_skas(void *r) +{ + struct pt_regs *regs = r; + long res; + int syscall; + + current->thread.nsyscalls++; + nsyscalls++; + syscall = UPT_SYSCALL_NR(®s->regs); + + if((syscall >= NR_syscalls) || (syscall < 0)) + res = -ENOSYS; + else res = EXECUTE_SYSCALL(syscall, regs); + + return(res); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/syscall_user.c b/arch/um/kernel/skas/syscall_user.c new file mode 100644 index 000000000000..2828e6e37721 --- /dev/null +++ b/arch/um/kernel/skas/syscall_user.c @@ -0,0 +1,44 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <signal.h> +#include "kern_util.h" +#include "uml-config.h" +#include "syscall_user.h" +#include "sysdep/ptrace.h" +#include "sysdep/sigcontext.h" +#include "skas.h" + +void handle_syscall(union uml_pt_regs *regs) +{ + long result; +#if UML_CONFIG_SYSCALL_DEBUG + int index; + + index = record_syscall_start(UPT_SYSCALL_NR(regs)); +#endif + + syscall_trace(regs, 0); + result = execute_syscall_skas(regs); + + REGS_SET_SYSCALL_RETURN(regs->skas.regs, result); + + syscall_trace(regs, 1); +#if UML_CONFIG_SYSCALL_DEBUG + record_syscall_end(index, result); +#endif +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/time.c b/arch/um/kernel/skas/time.c new file mode 100644 index 000000000000..98091494b897 --- /dev/null +++ b/arch/um/kernel/skas/time.c @@ -0,0 +1,30 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <sys/signal.h> +#include <sys/time.h> +#include "time_user.h" +#include "process.h" +#include "user.h" + +void user_time_init_skas(void) +{ + if(signal(SIGALRM, (__sighandler_t) alarm_handler) == SIG_ERR) + panic("Couldn't set SIGALRM handler"); + if(signal(SIGVTALRM, (__sighandler_t) alarm_handler) == SIG_ERR) + panic("Couldn't set SIGVTALRM handler"); + set_interval(ITIMER_VIRTUAL); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/tlb.c b/arch/um/kernel/skas/tlb.c new file mode 100644 index 000000000000..b8c5e71763d1 --- /dev/null +++ b/arch/um/kernel/skas/tlb.c @@ -0,0 +1,85 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Copyright 2003 PathScale, Inc. + * Licensed under the GPL + */ + +#include "linux/stddef.h" +#include "linux/sched.h" +#include "linux/mm.h" +#include "asm/page.h" +#include "asm/pgtable.h" +#include "asm/mmu.h" +#include "user_util.h" +#include "mem_user.h" +#include "mem.h" +#include "skas.h" +#include "os.h" +#include "tlb.h" + +static void do_ops(int fd, struct host_vm_op *ops, int last) +{ + struct host_vm_op *op; + int i; + + for(i = 0; i <= last; i++){ + op = &ops[i]; + switch(op->type){ + case MMAP: + map(fd, op->u.mmap.addr, op->u.mmap.len, + op->u.mmap.r, op->u.mmap.w, op->u.mmap.x, + op->u.mmap.fd, op->u.mmap.offset); + break; + case MUNMAP: + unmap(fd, (void *) op->u.munmap.addr, + op->u.munmap.len); + break; + case MPROTECT: + protect(fd, op->u.mprotect.addr, op->u.mprotect.len, + op->u.mprotect.r, op->u.mprotect.w, + op->u.mprotect.x); + break; + default: + printk("Unknown op type %d in do_ops\n", op->type); + break; + } + } +} + +static void fix_range(struct mm_struct *mm, unsigned long start_addr, + unsigned long end_addr, int force) +{ + int fd = mm->context.skas.mm_fd; + + fix_range_common(mm, start_addr, end_addr, force, fd, do_ops); +} + +void __flush_tlb_one_skas(unsigned long addr) +{ + flush_tlb_kernel_range_common(addr, addr + PAGE_SIZE); +} + +void flush_tlb_range_skas(struct vm_area_struct *vma, unsigned long start, + unsigned long end) +{ + if(vma->vm_mm == NULL) + flush_tlb_kernel_range_common(start, end); + else fix_range(vma->vm_mm, start, end, 0); +} + +void flush_tlb_mm_skas(struct mm_struct *mm) +{ + /* Don't bother flushing if this address space is about to be + * destroyed. + */ + if(atomic_read(&mm->mm_users) == 0) + return; + + fix_range(mm, 0, host_task_size, 0); + flush_tlb_kernel_range_common(start_vm, end_vm); +} + +void force_flush_all_skas(void) +{ + fix_range(current->mm, 0, host_task_size, 1); +} diff --git a/arch/um/kernel/skas/trap_user.c b/arch/um/kernel/skas/trap_user.c new file mode 100644 index 000000000000..8e9b46d4702e --- /dev/null +++ b/arch/um/kernel/skas/trap_user.c @@ -0,0 +1,71 @@ +/* + * Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include <signal.h> +#include <errno.h> +#include "sysdep/ptrace.h" +#include "signal_user.h" +#include "user_util.h" +#include "kern_util.h" +#include "task.h" +#include "sigcontext.h" + +void sig_handler_common_skas(int sig, void *sc_ptr) +{ + struct sigcontext *sc = sc_ptr; + struct skas_regs *r; + struct signal_info *info; + int save_errno = errno; + int save_user; + + /* This is done because to allow SIGSEGV to be delivered inside a SEGV + * handler. This can happen in copy_user, and if SEGV is disabled, + * the process will die. + * XXX Figure out why this is better than SA_NODEFER + */ + if(sig == SIGSEGV) + change_sig(SIGSEGV, 1); + + r = &TASK_REGS(get_current())->skas; + save_user = r->is_user; + r->is_user = 0; + r->fault_addr = SC_FAULT_ADDR(sc); + r->fault_type = SC_FAULT_TYPE(sc); + r->trap_type = SC_TRAP_TYPE(sc); + + change_sig(SIGUSR1, 1); + info = &sig_info[sig]; + if(!info->is_irq) unblock_signals(); + + (*info->handler)(sig, (union uml_pt_regs *) r); + + errno = save_errno; + r->is_user = save_user; +} + +void user_signal(int sig, union uml_pt_regs *regs) +{ + struct signal_info *info; + + regs->skas.is_user = 1; + regs->skas.fault_addr = 0; + regs->skas.fault_type = 0; + regs->skas.trap_type = 0; + info = &sig_info[sig]; + (*info->handler)(sig, regs); + + unblock_signals(); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/uaccess.c b/arch/um/kernel/skas/uaccess.c new file mode 100644 index 000000000000..7575ec489b63 --- /dev/null +++ b/arch/um/kernel/skas/uaccess.c @@ -0,0 +1,259 @@ +/* + * Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/stddef.h" +#include "linux/kernel.h" +#include "linux/string.h" +#include "linux/fs.h" +#include "linux/highmem.h" +#include "asm/page.h" +#include "asm/pgtable.h" +#include "asm/uaccess.h" +#include "kern_util.h" +#include "user_util.h" + +extern void *um_virt_to_phys(struct task_struct *task, unsigned long addr, + pte_t *pte_out); + +static unsigned long maybe_map(unsigned long virt, int is_write) +{ + pte_t pte; + int err; + + void *phys = um_virt_to_phys(current, virt, &pte); + int dummy_code; + + if(IS_ERR(phys) || (is_write && !pte_write(pte))){ + err = handle_page_fault(virt, 0, is_write, 1, &dummy_code); + if(err) + return(0); + phys = um_virt_to_phys(current, virt, NULL); + } + return((unsigned long) phys); +} + +static int do_op(unsigned long addr, int len, int is_write, + int (*op)(unsigned long addr, int len, void *arg), void *arg) +{ + struct page *page; + int n; + + addr = maybe_map(addr, is_write); + if(addr == -1) + return(-1); + + page = phys_to_page(addr); + addr = (unsigned long) kmap(page) + (addr & ~PAGE_MASK); + n = (*op)(addr, len, arg); + kunmap(page); + + return(n); +} + +static void do_buffer_op(void *jmpbuf, void *arg_ptr) +{ + va_list args; + unsigned long addr; + int len, is_write, size, remain, n; + int (*op)(unsigned long, int, void *); + void *arg; + int *res; + + /* Some old gccs recognize __va_copy, but not va_copy */ + __va_copy(args, *(va_list *)arg_ptr); + addr = va_arg(args, unsigned long); + len = va_arg(args, int); + is_write = va_arg(args, int); + op = va_arg(args, void *); + arg = va_arg(args, void *); + res = va_arg(args, int *); + va_end(args); + size = min(PAGE_ALIGN(addr) - addr, (unsigned long) len); + remain = len; + + current->thread.fault_catcher = jmpbuf; + n = do_op(addr, size, is_write, op, arg); + if(n != 0){ + *res = (n < 0 ? remain : 0); + goto out; + } + + addr += size; + remain -= size; + if(remain == 0){ + *res = 0; + goto out; + } + + while(addr < ((addr + remain) & PAGE_MASK)){ + n = do_op(addr, PAGE_SIZE, is_write, op, arg); + if(n != 0){ + *res = (n < 0 ? remain : 0); + goto out; + } + + addr += PAGE_SIZE; + remain -= PAGE_SIZE; + } + if(remain == 0){ + *res = 0; + goto out; + } + + n = do_op(addr, remain, is_write, op, arg); + if(n != 0) + *res = (n < 0 ? remain : 0); + else *res = 0; + out: + current->thread.fault_catcher = NULL; +} + +static int buffer_op(unsigned long addr, int len, int is_write, + int (*op)(unsigned long addr, int len, void *arg), + void *arg) +{ + int faulted, res; + + faulted = setjmp_wrapper(do_buffer_op, addr, len, is_write, op, arg, + &res); + if(!faulted) + return(res); + + return(addr + len - (unsigned long) current->thread.fault_addr); +} + +static int copy_chunk_from_user(unsigned long from, int len, void *arg) +{ + unsigned long *to_ptr = arg, to = *to_ptr; + + memcpy((void *) to, (void *) from, len); + *to_ptr += len; + return(0); +} + +int copy_from_user_skas(void *to, const void __user *from, int n) +{ + if(segment_eq(get_fs(), KERNEL_DS)){ + memcpy(to, (__force void*)from, n); + return(0); + } + + return(access_ok_skas(VERIFY_READ, from, n) ? + buffer_op((unsigned long) from, n, 0, copy_chunk_from_user, &to): + n); +} + +static int copy_chunk_to_user(unsigned long to, int len, void *arg) +{ + unsigned long *from_ptr = arg, from = *from_ptr; + + memcpy((void *) to, (void *) from, len); + *from_ptr += len; + return(0); +} + +int copy_to_user_skas(void __user *to, const void *from, int n) +{ + if(segment_eq(get_fs(), KERNEL_DS)){ + memcpy((__force void*)to, from, n); + return(0); + } + + return(access_ok_skas(VERIFY_WRITE, to, n) ? + buffer_op((unsigned long) to, n, 1, copy_chunk_to_user, &from) : + n); +} + +static int strncpy_chunk_from_user(unsigned long from, int len, void *arg) +{ + char **to_ptr = arg, *to = *to_ptr; + int n; + + strncpy(to, (void *) from, len); + n = strnlen(to, len); + *to_ptr += n; + + if(n < len) + return(1); + return(0); +} + +int strncpy_from_user_skas(char *dst, const char __user *src, int count) +{ + int n; + char *ptr = dst; + + if(segment_eq(get_fs(), KERNEL_DS)){ + strncpy(dst, (__force void*)src, count); + return(strnlen(dst, count)); + } + + if(!access_ok_skas(VERIFY_READ, src, 1)) + return(-EFAULT); + + n = buffer_op((unsigned long) src, count, 0, strncpy_chunk_from_user, + &ptr); + if(n != 0) + return(-EFAULT); + return(strnlen(dst, count)); +} + +static int clear_chunk(unsigned long addr, int len, void *unused) +{ + memset((void *) addr, 0, len); + return(0); +} + +int __clear_user_skas(void __user *mem, int len) +{ + return(buffer_op((unsigned long) mem, len, 1, clear_chunk, NULL)); +} + +int clear_user_skas(void __user *mem, int len) +{ + if(segment_eq(get_fs(), KERNEL_DS)){ + memset((__force void*)mem, 0, len); + return(0); + } + + return(access_ok_skas(VERIFY_WRITE, mem, len) ? + buffer_op((unsigned long) mem, len, 1, clear_chunk, NULL) : len); +} + +static int strnlen_chunk(unsigned long str, int len, void *arg) +{ + int *len_ptr = arg, n; + + n = strnlen((void *) str, len); + *len_ptr += n; + + if(n < len) + return(1); + return(0); +} + +int strnlen_user_skas(const void __user *str, int len) +{ + int count = 0, n; + + if(segment_eq(get_fs(), KERNEL_DS)) + return(strnlen((__force char*)str, len) + 1); + + n = buffer_op((unsigned long) str, len, 0, strnlen_chunk, &count); + if(n == 0) + return(count + 1); + return(-EFAULT); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/util/Makefile b/arch/um/kernel/skas/util/Makefile new file mode 100644 index 000000000000..17f5909d60f7 --- /dev/null +++ b/arch/um/kernel/skas/util/Makefile @@ -0,0 +1,4 @@ +hostprogs-y := mk_ptregs +always := $(hostprogs-y) + +mk_ptregs-objs := mk_ptregs-$(SUBARCH).o diff --git a/arch/um/kernel/skas/util/mk_ptregs-i386.c b/arch/um/kernel/skas/util/mk_ptregs-i386.c new file mode 100644 index 000000000000..0788dd05bcac --- /dev/null +++ b/arch/um/kernel/skas/util/mk_ptregs-i386.c @@ -0,0 +1,51 @@ +#include <stdio.h> +#include <asm/ptrace.h> +#include <asm/user.h> + +#define PRINT_REG(name, val) printf("#define HOST_%s %d\n", (name), (val)) + +int main(int argc, char **argv) +{ + printf("/* Automatically generated by " + "arch/um/kernel/skas/util/mk_ptregs */\n"); + printf("\n"); + printf("#ifndef __SKAS_PT_REGS_\n"); + printf("#define __SKAS_PT_REGS_\n"); + printf("\n"); + printf("#define HOST_FRAME_SIZE %d\n", FRAME_SIZE); + printf("#define HOST_FP_SIZE %d\n", + sizeof(struct user_i387_struct) / sizeof(unsigned long)); + printf("#define HOST_XFP_SIZE %d\n", + sizeof(struct user_fxsr_struct) / sizeof(unsigned long)); + + PRINT_REG("IP", EIP); + PRINT_REG("SP", UESP); + PRINT_REG("EFLAGS", EFL); + PRINT_REG("EAX", EAX); + PRINT_REG("EBX", EBX); + PRINT_REG("ECX", ECX); + PRINT_REG("EDX", EDX); + PRINT_REG("ESI", ESI); + PRINT_REG("EDI", EDI); + PRINT_REG("EBP", EBP); + PRINT_REG("CS", CS); + PRINT_REG("SS", SS); + PRINT_REG("DS", DS); + PRINT_REG("FS", FS); + PRINT_REG("ES", ES); + PRINT_REG("GS", GS); + printf("\n"); + printf("#endif\n"); + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/util/mk_ptregs-x86_64.c b/arch/um/kernel/skas/util/mk_ptregs-x86_64.c new file mode 100644 index 000000000000..67aee92a70ef --- /dev/null +++ b/arch/um/kernel/skas/util/mk_ptregs-x86_64.c @@ -0,0 +1,68 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ + +#include <stdio.h> +#define __FRAME_OFFSETS +#include <asm/ptrace.h> + +#define PRINT_REG(name, val) \ + printf("#define HOST_%s (%d / sizeof(unsigned long))\n", (name), (val)) + +int main(int argc, char **argv) +{ + printf("/* Automatically generated by " + "arch/um/kernel/skas/util/mk_ptregs */\n"); + printf("\n"); + printf("#ifndef __SKAS_PT_REGS_\n"); + printf("#define __SKAS_PT_REGS_\n"); + printf("#define HOST_FRAME_SIZE (%d / sizeof(unsigned long))\n", + FRAME_SIZE); + PRINT_REG("RBX", RBX); + PRINT_REG("RCX", RCX); + PRINT_REG("RDI", RDI); + PRINT_REG("RSI", RSI); + PRINT_REG("RDX", RDX); + PRINT_REG("RBP", RBP); + PRINT_REG("RAX", RAX); + PRINT_REG("R8", R8); + PRINT_REG("R9", R9); + PRINT_REG("R10", R10); + PRINT_REG("R11", R11); + PRINT_REG("R12", R12); + PRINT_REG("R13", R13); + PRINT_REG("R14", R14); + PRINT_REG("R15", R15); + PRINT_REG("ORIG_RAX", ORIG_RAX); + PRINT_REG("CS", CS); + PRINT_REG("SS", SS); + PRINT_REG("EFLAGS", EFLAGS); +#if 0 + PRINT_REG("FS", FS); + PRINT_REG("GS", GS); + PRINT_REG("DS", DS); + PRINT_REG("ES", ES); +#endif + + PRINT_REG("IP", RIP); + PRINT_REG("SP", RSP); + printf("#define HOST_FP_SIZE 0\n"); + printf("#define HOST_XFP_SIZE 0\n"); + printf("\n"); + printf("\n"); + printf("#endif\n"); + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/smp.c b/arch/um/kernel/smp.c new file mode 100644 index 000000000000..72113b0a96e7 --- /dev/null +++ b/arch/um/kernel/smp.c @@ -0,0 +1,269 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/percpu.h" +#include "asm/pgalloc.h" +#include "asm/tlb.h" + +/* For some reason, mmu_gathers are referenced when CONFIG_SMP is off. */ +DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); + +#ifdef CONFIG_SMP + +#include "linux/sched.h" +#include "linux/module.h" +#include "linux/threads.h" +#include "linux/interrupt.h" +#include "linux/err.h" +#include "linux/hardirq.h" +#include "asm/smp.h" +#include "asm/processor.h" +#include "asm/spinlock.h" +#include "user_util.h" +#include "kern_util.h" +#include "kern.h" +#include "irq_user.h" +#include "os.h" + +/* CPU online map, set by smp_boot_cpus */ +cpumask_t cpu_online_map = CPU_MASK_NONE; +cpumask_t cpu_possible_map = CPU_MASK_NONE; + +EXPORT_SYMBOL(cpu_online_map); +EXPORT_SYMBOL(cpu_possible_map); + +/* Per CPU bogomips and other parameters + * The only piece used here is the ipi pipe, which is set before SMP is + * started and never changed. + */ +struct cpuinfo_um cpu_data[NR_CPUS]; + +/* A statistic, can be a little off */ +int num_reschedules_sent = 0; + +/* Not changed after boot */ +struct task_struct *idle_threads[NR_CPUS]; + +void smp_send_reschedule(int cpu) +{ + os_write_file(cpu_data[cpu].ipi_pipe[1], "R", 1); + num_reschedules_sent++; +} + +void smp_send_stop(void) +{ + int i; + + printk(KERN_INFO "Stopping all CPUs..."); + for(i = 0; i < num_online_cpus(); i++){ + if(i == current_thread->cpu) + continue; + os_write_file(cpu_data[i].ipi_pipe[1], "S", 1); + } + printk("done\n"); +} + +static cpumask_t smp_commenced_mask = CPU_MASK_NONE; +static cpumask_t cpu_callin_map = CPU_MASK_NONE; + +static int idle_proc(void *cpup) +{ + int cpu = (int) cpup, err; + + err = os_pipe(cpu_data[cpu].ipi_pipe, 1, 1); + if(err < 0) + panic("CPU#%d failed to create IPI pipe, err = %d", cpu, -err); + + activate_ipi(cpu_data[cpu].ipi_pipe[0], + current->thread.mode.tt.extern_pid); + + wmb(); + if (cpu_test_and_set(cpu, cpu_callin_map)) { + printk("huh, CPU#%d already present??\n", cpu); + BUG(); + } + + while (!cpu_isset(cpu, smp_commenced_mask)) + cpu_relax(); + + cpu_set(cpu, cpu_online_map); + default_idle(); + return(0); +} + +static struct task_struct *idle_thread(int cpu) +{ + struct task_struct *new_task; + unsigned char c; + + current->thread.request.u.thread.proc = idle_proc; + current->thread.request.u.thread.arg = (void *) cpu; + new_task = fork_idle(cpu); + if(IS_ERR(new_task)) + panic("copy_process failed in idle_thread, error = %ld", + PTR_ERR(new_task)); + + cpu_tasks[cpu] = ((struct cpu_task) + { .pid = new_task->thread.mode.tt.extern_pid, + .task = new_task } ); + idle_threads[cpu] = new_task; + CHOOSE_MODE(os_write_file(new_task->thread.mode.tt.switch_pipe[1], &c, + sizeof(c)), + ({ panic("skas mode doesn't support SMP"); })); + return(new_task); +} + +void smp_prepare_cpus(unsigned int maxcpus) +{ + struct task_struct *idle; + unsigned long waittime; + int err, cpu, me = smp_processor_id(); + int i; + + for (i = 0; i < ncpus; ++i) + cpu_set(i, cpu_possible_map); + + cpu_clear(me, cpu_online_map); + cpu_set(me, cpu_online_map); + cpu_set(me, cpu_callin_map); + + err = os_pipe(cpu_data[me].ipi_pipe, 1, 1); + if(err < 0) + panic("CPU#0 failed to create IPI pipe, errno = %d", -err); + + activate_ipi(cpu_data[me].ipi_pipe[0], + current->thread.mode.tt.extern_pid); + + for(cpu = 1; cpu < ncpus; cpu++){ + printk("Booting processor %d...\n", cpu); + + idle = idle_thread(cpu); + + init_idle(idle, cpu); + unhash_process(idle); + + waittime = 200000000; + while (waittime-- && !cpu_isset(cpu, cpu_callin_map)) + cpu_relax(); + + if (cpu_isset(cpu, cpu_callin_map)) + printk("done\n"); + else printk("failed\n"); + } +} + +void smp_prepare_boot_cpu(void) +{ + cpu_set(smp_processor_id(), cpu_online_map); +} + +int __cpu_up(unsigned int cpu) +{ + cpu_set(cpu, smp_commenced_mask); + while (!cpu_isset(cpu, cpu_online_map)) + mb(); + return(0); +} + +int setup_profiling_timer(unsigned int multiplier) +{ + printk(KERN_INFO "setup_profiling_timer\n"); + return(0); +} + +void smp_call_function_slave(int cpu); + +void IPI_handler(int cpu) +{ + unsigned char c; + int fd; + + fd = cpu_data[cpu].ipi_pipe[0]; + while (os_read_file(fd, &c, 1) == 1) { + switch (c) { + case 'C': + smp_call_function_slave(cpu); + break; + + case 'R': + set_tsk_need_resched(current); + break; + + case 'S': + printk("CPU#%d stopping\n", cpu); + while(1) + pause(); + break; + + default: + printk("CPU#%d received unknown IPI [%c]!\n", cpu, c); + break; + } + } +} + +int hard_smp_processor_id(void) +{ + return(pid_to_processor_id(os_getpid())); +} + +static DEFINE_SPINLOCK(call_lock); +static atomic_t scf_started; +static atomic_t scf_finished; +static void (*func)(void *info); +static void *info; + +void smp_call_function_slave(int cpu) +{ + atomic_inc(&scf_started); + (*func)(info); + atomic_inc(&scf_finished); +} + +int smp_call_function(void (*_func)(void *info), void *_info, int nonatomic, + int wait) +{ + int cpus = num_online_cpus() - 1; + int i; + + if (!cpus) + return 0; + + /* Can deadlock when called with interrupts disabled */ + WARN_ON(irqs_disabled()); + + spin_lock_bh(&call_lock); + atomic_set(&scf_started, 0); + atomic_set(&scf_finished, 0); + func = _func; + info = _info; + + for_each_online_cpu(i) + os_write_file(cpu_data[i].ipi_pipe[1], "C", 1); + + while (atomic_read(&scf_started) != cpus) + barrier(); + + if (wait) + while (atomic_read(&scf_finished) != cpus) + barrier(); + + spin_unlock_bh(&call_lock); + return 0; +} + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/sys_call_table.c b/arch/um/kernel/sys_call_table.c new file mode 100644 index 000000000000..7fc06c85b29d --- /dev/null +++ b/arch/um/kernel/sys_call_table.c @@ -0,0 +1,276 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Copyright 2003 PathScale, Inc. + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/unistd.h" +#include "linux/sys.h" +#include "linux/swap.h" +#include "linux/syscalls.h" +#include "linux/sysctl.h" +#include "asm/signal.h" +#include "sysdep/syscalls.h" +#include "kern_util.h" + +#ifdef CONFIG_NFSD +#define NFSSERVCTL sys_nfsservctl +#else +#define NFSSERVCTL sys_ni_syscall +#endif + +#define LAST_GENERIC_SYSCALL __NR_keyctl + +#if LAST_GENERIC_SYSCALL > LAST_ARCH_SYSCALL +#define LAST_SYSCALL LAST_GENERIC_SYSCALL +#else +#define LAST_SYSCALL LAST_ARCH_SYSCALL +#endif + +extern syscall_handler_t sys_fork; +extern syscall_handler_t sys_execve; +extern syscall_handler_t um_time; +extern syscall_handler_t um_stime; +extern syscall_handler_t sys_pipe; +extern syscall_handler_t sys_olduname; +extern syscall_handler_t sys_sigaction; +extern syscall_handler_t sys_sigsuspend; +extern syscall_handler_t old_readdir; +extern syscall_handler_t sys_uname; +extern syscall_handler_t sys_ipc; +extern syscall_handler_t sys_sigreturn; +extern syscall_handler_t sys_clone; +extern syscall_handler_t sys_rt_sigreturn; +extern syscall_handler_t sys_sigaltstack; +extern syscall_handler_t sys_vfork; +extern syscall_handler_t old_select; +extern syscall_handler_t sys_modify_ldt; +extern syscall_handler_t sys_rt_sigsuspend; +extern syscall_handler_t sys_mbind; +extern syscall_handler_t sys_get_mempolicy; +extern syscall_handler_t sys_set_mempolicy; +extern syscall_handler_t sys_sys_setaltroot; + +syscall_handler_t *sys_call_table[] = { + [ __NR_restart_syscall ] = (syscall_handler_t *) sys_restart_syscall, + [ __NR_exit ] = (syscall_handler_t *) sys_exit, + [ __NR_fork ] = (syscall_handler_t *) sys_fork, + [ __NR_read ] = (syscall_handler_t *) sys_read, + [ __NR_write ] = (syscall_handler_t *) sys_write, + + /* These three are declared differently in asm/unistd.h */ + [ __NR_open ] = (syscall_handler_t *) sys_open, + [ __NR_close ] = (syscall_handler_t *) sys_close, + [ __NR_creat ] = (syscall_handler_t *) sys_creat, + [ __NR_link ] = (syscall_handler_t *) sys_link, + [ __NR_unlink ] = (syscall_handler_t *) sys_unlink, + [ __NR_execve ] = (syscall_handler_t *) sys_execve, + + /* declared differently in kern_util.h */ + [ __NR_chdir ] = (syscall_handler_t *) sys_chdir, + [ __NR_time ] = um_time, + [ __NR_mknod ] = (syscall_handler_t *) sys_mknod, + [ __NR_chmod ] = (syscall_handler_t *) sys_chmod, + [ __NR_lchown ] = (syscall_handler_t *) sys_lchown16, + [ __NR_lseek ] = (syscall_handler_t *) sys_lseek, + [ __NR_getpid ] = (syscall_handler_t *) sys_getpid, + [ __NR_mount ] = (syscall_handler_t *) sys_mount, + [ __NR_setuid ] = (syscall_handler_t *) sys_setuid16, + [ __NR_getuid ] = (syscall_handler_t *) sys_getuid16, + [ __NR_ptrace ] = (syscall_handler_t *) sys_ptrace, + [ __NR_alarm ] = (syscall_handler_t *) sys_alarm, + [ __NR_pause ] = (syscall_handler_t *) sys_pause, + [ __NR_utime ] = (syscall_handler_t *) sys_utime, + [ __NR_access ] = (syscall_handler_t *) sys_access, + [ __NR_sync ] = (syscall_handler_t *) sys_sync, + [ __NR_kill ] = (syscall_handler_t *) sys_kill, + [ __NR_rename ] = (syscall_handler_t *) sys_rename, + [ __NR_mkdir ] = (syscall_handler_t *) sys_mkdir, + [ __NR_rmdir ] = (syscall_handler_t *) sys_rmdir, + + /* Declared differently in asm/unistd.h */ + [ __NR_dup ] = (syscall_handler_t *) sys_dup, + [ __NR_pipe ] = (syscall_handler_t *) sys_pipe, + [ __NR_times ] = (syscall_handler_t *) sys_times, + [ __NR_brk ] = (syscall_handler_t *) sys_brk, + [ __NR_setgid ] = (syscall_handler_t *) sys_setgid16, + [ __NR_getgid ] = (syscall_handler_t *) sys_getgid16, + [ __NR_geteuid ] = (syscall_handler_t *) sys_geteuid16, + [ __NR_getegid ] = (syscall_handler_t *) sys_getegid16, + [ __NR_acct ] = (syscall_handler_t *) sys_acct, + [ __NR_umount2 ] = (syscall_handler_t *) sys_umount, + [ __NR_ioctl ] = (syscall_handler_t *) sys_ioctl, + [ __NR_fcntl ] = (syscall_handler_t *) sys_fcntl, + [ __NR_setpgid ] = (syscall_handler_t *) sys_setpgid, + [ __NR_umask ] = (syscall_handler_t *) sys_umask, + [ __NR_chroot ] = (syscall_handler_t *) sys_chroot, + [ __NR_ustat ] = (syscall_handler_t *) sys_ustat, + [ __NR_dup2 ] = (syscall_handler_t *) sys_dup2, + [ __NR_getppid ] = (syscall_handler_t *) sys_getppid, + [ __NR_getpgrp ] = (syscall_handler_t *) sys_getpgrp, + [ __NR_setsid ] = (syscall_handler_t *) sys_setsid, + [ __NR_setreuid ] = (syscall_handler_t *) sys_setreuid16, + [ __NR_setregid ] = (syscall_handler_t *) sys_setregid16, + [ __NR_sethostname ] = (syscall_handler_t *) sys_sethostname, + [ __NR_setrlimit ] = (syscall_handler_t *) sys_setrlimit, + [ __NR_getrlimit ] = (syscall_handler_t *) sys_old_getrlimit, + [ __NR_getrusage ] = (syscall_handler_t *) sys_getrusage, + [ __NR_gettimeofday ] = (syscall_handler_t *) sys_gettimeofday, + [ __NR_settimeofday ] = (syscall_handler_t *) sys_settimeofday, + [ __NR_getgroups ] = (syscall_handler_t *) sys_getgroups16, + [ __NR_setgroups ] = (syscall_handler_t *) sys_setgroups16, + [ __NR_symlink ] = (syscall_handler_t *) sys_symlink, + [ __NR_readlink ] = (syscall_handler_t *) sys_readlink, + [ __NR_uselib ] = (syscall_handler_t *) sys_uselib, + [ __NR_swapon ] = (syscall_handler_t *) sys_swapon, + [ __NR_reboot ] = (syscall_handler_t *) sys_reboot, + [ __NR_munmap ] = (syscall_handler_t *) sys_munmap, + [ __NR_truncate ] = (syscall_handler_t *) sys_truncate, + [ __NR_ftruncate ] = (syscall_handler_t *) sys_ftruncate, + [ __NR_fchmod ] = (syscall_handler_t *) sys_fchmod, + [ __NR_fchown ] = (syscall_handler_t *) sys_fchown16, + [ __NR_getpriority ] = (syscall_handler_t *) sys_getpriority, + [ __NR_setpriority ] = (syscall_handler_t *) sys_setpriority, + [ __NR_statfs ] = (syscall_handler_t *) sys_statfs, + [ __NR_fstatfs ] = (syscall_handler_t *) sys_fstatfs, + [ __NR_ioperm ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_syslog ] = (syscall_handler_t *) sys_syslog, + [ __NR_setitimer ] = (syscall_handler_t *) sys_setitimer, + [ __NR_getitimer ] = (syscall_handler_t *) sys_getitimer, + [ __NR_stat ] = (syscall_handler_t *) sys_newstat, + [ __NR_lstat ] = (syscall_handler_t *) sys_newlstat, + [ __NR_fstat ] = (syscall_handler_t *) sys_newfstat, + [ __NR_vhangup ] = (syscall_handler_t *) sys_vhangup, + [ __NR_wait4 ] = (syscall_handler_t *) sys_wait4, + [ __NR_swapoff ] = (syscall_handler_t *) sys_swapoff, + [ __NR_sysinfo ] = (syscall_handler_t *) sys_sysinfo, + [ __NR_fsync ] = (syscall_handler_t *) sys_fsync, + [ __NR_clone ] = (syscall_handler_t *) sys_clone, + [ __NR_setdomainname ] = (syscall_handler_t *) sys_setdomainname, + [ __NR_uname ] = (syscall_handler_t *) sys_newuname, + [ __NR_adjtimex ] = (syscall_handler_t *) sys_adjtimex, + [ __NR_mprotect ] = (syscall_handler_t *) sys_mprotect, + [ __NR_create_module ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_init_module ] = (syscall_handler_t *) sys_init_module, + [ __NR_delete_module ] = (syscall_handler_t *) sys_delete_module, + [ __NR_get_kernel_syms ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_quotactl ] = (syscall_handler_t *) sys_quotactl, + [ __NR_getpgid ] = (syscall_handler_t *) sys_getpgid, + [ __NR_fchdir ] = (syscall_handler_t *) sys_fchdir, + [ __NR_sysfs ] = (syscall_handler_t *) sys_sysfs, + [ __NR_personality ] = (syscall_handler_t *) sys_personality, + [ __NR_afs_syscall ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_setfsuid ] = (syscall_handler_t *) sys_setfsuid16, + [ __NR_setfsgid ] = (syscall_handler_t *) sys_setfsgid16, + [ __NR_getdents ] = (syscall_handler_t *) sys_getdents, + [ __NR_flock ] = (syscall_handler_t *) sys_flock, + [ __NR_msync ] = (syscall_handler_t *) sys_msync, + [ __NR_readv ] = (syscall_handler_t *) sys_readv, + [ __NR_writev ] = (syscall_handler_t *) sys_writev, + [ __NR_getsid ] = (syscall_handler_t *) sys_getsid, + [ __NR_fdatasync ] = (syscall_handler_t *) sys_fdatasync, + [ __NR__sysctl ] = (syscall_handler_t *) sys_sysctl, + [ __NR_mlock ] = (syscall_handler_t *) sys_mlock, + [ __NR_munlock ] = (syscall_handler_t *) sys_munlock, + [ __NR_mlockall ] = (syscall_handler_t *) sys_mlockall, + [ __NR_munlockall ] = (syscall_handler_t *) sys_munlockall, + [ __NR_sched_setparam ] = (syscall_handler_t *) sys_sched_setparam, + [ __NR_sched_getparam ] = (syscall_handler_t *) sys_sched_getparam, + [ __NR_sched_setscheduler ] = (syscall_handler_t *) sys_sched_setscheduler, + [ __NR_sched_getscheduler ] = (syscall_handler_t *) sys_sched_getscheduler, + [ __NR_sched_yield ] = (syscall_handler_t *) yield, + [ __NR_sched_get_priority_max ] = (syscall_handler_t *) sys_sched_get_priority_max, + [ __NR_sched_get_priority_min ] = (syscall_handler_t *) sys_sched_get_priority_min, + [ __NR_sched_rr_get_interval ] = (syscall_handler_t *) sys_sched_rr_get_interval, + [ __NR_nanosleep ] = (syscall_handler_t *) sys_nanosleep, + [ __NR_mremap ] = (syscall_handler_t *) sys_mremap, + [ __NR_setresuid ] = (syscall_handler_t *) sys_setresuid16, + [ __NR_getresuid ] = (syscall_handler_t *) sys_getresuid16, + [ __NR_query_module ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_poll ] = (syscall_handler_t *) sys_poll, + [ __NR_nfsservctl ] = (syscall_handler_t *) NFSSERVCTL, + [ __NR_setresgid ] = (syscall_handler_t *) sys_setresgid16, + [ __NR_getresgid ] = (syscall_handler_t *) sys_getresgid16, + [ __NR_prctl ] = (syscall_handler_t *) sys_prctl, + [ __NR_rt_sigreturn ] = (syscall_handler_t *) sys_rt_sigreturn, + [ __NR_rt_sigaction ] = (syscall_handler_t *) sys_rt_sigaction, + [ __NR_rt_sigprocmask ] = (syscall_handler_t *) sys_rt_sigprocmask, + [ __NR_rt_sigpending ] = (syscall_handler_t *) sys_rt_sigpending, + [ __NR_rt_sigtimedwait ] = (syscall_handler_t *) sys_rt_sigtimedwait, + [ __NR_rt_sigqueueinfo ] = (syscall_handler_t *) sys_rt_sigqueueinfo, + [ __NR_rt_sigsuspend ] = (syscall_handler_t *) sys_rt_sigsuspend, + [ __NR_pread64 ] = (syscall_handler_t *) sys_pread64, + [ __NR_pwrite64 ] = (syscall_handler_t *) sys_pwrite64, + [ __NR_chown ] = (syscall_handler_t *) sys_chown16, + [ __NR_getcwd ] = (syscall_handler_t *) sys_getcwd, + [ __NR_capget ] = (syscall_handler_t *) sys_capget, + [ __NR_capset ] = (syscall_handler_t *) sys_capset, + [ __NR_sigaltstack ] = (syscall_handler_t *) sys_sigaltstack, + [ __NR_sendfile ] = (syscall_handler_t *) sys_sendfile, + [ __NR_getpmsg ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_putpmsg ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_vfork ] = (syscall_handler_t *) sys_vfork, + [ __NR_getdents64 ] = (syscall_handler_t *) sys_getdents64, + [ __NR_gettid ] = (syscall_handler_t *) sys_gettid, + [ __NR_readahead ] = (syscall_handler_t *) sys_readahead, + [ __NR_setxattr ] = (syscall_handler_t *) sys_setxattr, + [ __NR_lsetxattr ] = (syscall_handler_t *) sys_lsetxattr, + [ __NR_fsetxattr ] = (syscall_handler_t *) sys_fsetxattr, + [ __NR_getxattr ] = (syscall_handler_t *) sys_getxattr, + [ __NR_lgetxattr ] = (syscall_handler_t *) sys_lgetxattr, + [ __NR_fgetxattr ] = (syscall_handler_t *) sys_fgetxattr, + [ __NR_listxattr ] = (syscall_handler_t *) sys_listxattr, + [ __NR_llistxattr ] = (syscall_handler_t *) sys_llistxattr, + [ __NR_flistxattr ] = (syscall_handler_t *) sys_flistxattr, + [ __NR_removexattr ] = (syscall_handler_t *) sys_removexattr, + [ __NR_lremovexattr ] = (syscall_handler_t *) sys_lremovexattr, + [ __NR_fremovexattr ] = (syscall_handler_t *) sys_fremovexattr, + [ __NR_tkill ] = (syscall_handler_t *) sys_tkill, + [ __NR_futex ] = (syscall_handler_t *) sys_futex, + [ __NR_sched_setaffinity ] = (syscall_handler_t *) sys_sched_setaffinity, + [ __NR_sched_getaffinity ] = (syscall_handler_t *) sys_sched_getaffinity, + [ __NR_io_setup ] = (syscall_handler_t *) sys_io_setup, + [ __NR_io_destroy ] = (syscall_handler_t *) sys_io_destroy, + [ __NR_io_getevents ] = (syscall_handler_t *) sys_io_getevents, + [ __NR_io_submit ] = (syscall_handler_t *) sys_io_submit, + [ __NR_io_cancel ] = (syscall_handler_t *) sys_io_cancel, + [ __NR_exit_group ] = (syscall_handler_t *) sys_exit_group, + [ __NR_lookup_dcookie ] = (syscall_handler_t *) sys_lookup_dcookie, + [ __NR_epoll_create ] = (syscall_handler_t *) sys_epoll_create, + [ __NR_epoll_ctl ] = (syscall_handler_t *) sys_epoll_ctl, + [ __NR_epoll_wait ] = (syscall_handler_t *) sys_epoll_wait, + [ __NR_remap_file_pages ] = (syscall_handler_t *) sys_remap_file_pages, + [ __NR_set_tid_address ] = (syscall_handler_t *) sys_set_tid_address, + [ __NR_timer_create ] = (syscall_handler_t *) sys_timer_create, + [ __NR_timer_settime ] = (syscall_handler_t *) sys_timer_settime, + [ __NR_timer_gettime ] = (syscall_handler_t *) sys_timer_gettime, + [ __NR_timer_getoverrun ] = (syscall_handler_t *) sys_timer_getoverrun, + [ __NR_timer_delete ] = (syscall_handler_t *) sys_timer_delete, + [ __NR_clock_settime ] = (syscall_handler_t *) sys_clock_settime, + [ __NR_clock_gettime ] = (syscall_handler_t *) sys_clock_gettime, + [ __NR_clock_getres ] = (syscall_handler_t *) sys_clock_getres, + [ __NR_clock_nanosleep ] = (syscall_handler_t *) sys_clock_nanosleep, + [ __NR_tgkill ] = (syscall_handler_t *) sys_tgkill, + [ __NR_utimes ] = (syscall_handler_t *) sys_utimes, + [ __NR_fadvise64 ] = (syscall_handler_t *) sys_fadvise64, + [ __NR_vserver ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_mbind ] = (syscall_handler_t *) sys_mbind, + [ __NR_get_mempolicy ] = (syscall_handler_t *) sys_get_mempolicy, + [ __NR_set_mempolicy ] = (syscall_handler_t *) sys_set_mempolicy, + [ __NR_mq_open ] = (syscall_handler_t *) sys_mq_open, + [ __NR_mq_unlink ] = (syscall_handler_t *) sys_mq_unlink, + [ __NR_mq_timedsend ] = (syscall_handler_t *) sys_mq_timedsend, + [ __NR_mq_timedreceive ] = (syscall_handler_t *) sys_mq_timedreceive, + [ __NR_mq_notify ] = (syscall_handler_t *) sys_mq_notify, + [ __NR_mq_getsetattr ] = (syscall_handler_t *) sys_mq_getsetattr, + [ __NR_kexec_load ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_waitid ] = (syscall_handler_t *) sys_waitid, + [ __NR_add_key ] = (syscall_handler_t *) sys_add_key, + [ __NR_request_key ] = (syscall_handler_t *) sys_request_key, + [ __NR_keyctl ] = (syscall_handler_t *) sys_keyctl, + + ARCH_SYSCALLS + [ LAST_SYSCALL + 1 ... NR_syscalls ] = + (syscall_handler_t *) sys_ni_syscall +}; diff --git a/arch/um/kernel/syscall_kern.c b/arch/um/kernel/syscall_kern.c new file mode 100644 index 000000000000..42731e04f50f --- /dev/null +++ b/arch/um/kernel/syscall_kern.c @@ -0,0 +1,176 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/file.h" +#include "linux/smp_lock.h" +#include "linux/mm.h" +#include "linux/utsname.h" +#include "linux/msg.h" +#include "linux/shm.h" +#include "linux/sys.h" +#include "linux/syscalls.h" +#include "linux/unistd.h" +#include "linux/slab.h" +#include "linux/utime.h" +#include "asm/mman.h" +#include "asm/uaccess.h" +#include "asm/ipc.h" +#include "kern_util.h" +#include "user_util.h" +#include "sysdep/syscalls.h" +#include "mode_kern.h" +#include "choose-mode.h" + +/* Unlocked, I don't care if this is a bit off */ +int nsyscalls = 0; + +long sys_fork(void) +{ + long ret; + + current->thread.forking = 1; + ret = do_fork(SIGCHLD, 0, NULL, 0, NULL, NULL); + current->thread.forking = 0; + return(ret); +} + +long sys_vfork(void) +{ + long ret; + + current->thread.forking = 1; + ret = do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, 0, NULL, 0, NULL, + NULL); + current->thread.forking = 0; + return(ret); +} + +/* common code for old and new mmaps */ +long sys_mmap2(unsigned long addr, unsigned long len, + unsigned long prot, unsigned long flags, + unsigned long fd, unsigned long pgoff) +{ + long error = -EBADF; + struct file * file = NULL; + + flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); + if (!(flags & MAP_ANONYMOUS)) { + file = fget(fd); + if (!file) + goto out; + } + + down_write(¤t->mm->mmap_sem); + error = do_mmap_pgoff(file, addr, len, prot, flags, pgoff); + up_write(¤t->mm->mmap_sem); + + if (file) + fput(file); + out: + return error; +} + +long old_mmap(unsigned long addr, unsigned long len, + unsigned long prot, unsigned long flags, + unsigned long fd, unsigned long offset) +{ + long err = -EINVAL; + if (offset & ~PAGE_MASK) + goto out; + + err = sys_mmap2(addr, len, prot, flags, fd, offset >> PAGE_SHIFT); + out: + return err; +} +/* + * sys_pipe() is the normal C calling standard for creating + * a pipe. It's not the way unix traditionally does this, though. + */ +long sys_pipe(unsigned long __user * fildes) +{ + int fd[2]; + long error; + + error = do_pipe(fd); + if (!error) { + if (copy_to_user(fildes, fd, sizeof(fd))) + error = -EFAULT; + } + return error; +} + + +long sys_uname(struct old_utsname * name) +{ + long err; + if (!name) + return -EFAULT; + down_read(&uts_sem); + err=copy_to_user(name, &system_utsname, sizeof (*name)); + up_read(&uts_sem); + return err?-EFAULT:0; +} + +long sys_olduname(struct oldold_utsname * name) +{ + long error; + + if (!name) + return -EFAULT; + if (!access_ok(VERIFY_WRITE,name,sizeof(struct oldold_utsname))) + return -EFAULT; + + down_read(&uts_sem); + + error = __copy_to_user(&name->sysname,&system_utsname.sysname, + __OLD_UTS_LEN); + error |= __put_user(0,name->sysname+__OLD_UTS_LEN); + error |= __copy_to_user(&name->nodename,&system_utsname.nodename, + __OLD_UTS_LEN); + error |= __put_user(0,name->nodename+__OLD_UTS_LEN); + error |= __copy_to_user(&name->release,&system_utsname.release, + __OLD_UTS_LEN); + error |= __put_user(0,name->release+__OLD_UTS_LEN); + error |= __copy_to_user(&name->version,&system_utsname.version, + __OLD_UTS_LEN); + error |= __put_user(0,name->version+__OLD_UTS_LEN); + error |= __copy_to_user(&name->machine,&system_utsname.machine, + __OLD_UTS_LEN); + error |= __put_user(0,name->machine+__OLD_UTS_LEN); + + up_read(&uts_sem); + + error = error ? -EFAULT : 0; + + return error; +} + +DEFINE_SPINLOCK(syscall_lock); + +static int syscall_index = 0; + +int next_syscall_index(int limit) +{ + int ret; + + spin_lock(&syscall_lock); + ret = syscall_index; + if(++syscall_index == limit) + syscall_index = 0; + spin_unlock(&syscall_lock); + return(ret); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/syscall_user.c b/arch/um/kernel/syscall_user.c new file mode 100644 index 000000000000..01b711e00a85 --- /dev/null +++ b/arch/um/kernel/syscall_user.c @@ -0,0 +1,48 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <sys/time.h> +#include "kern_util.h" +#include "syscall_user.h" + +struct { + int syscall; + int pid; + long result; + struct timeval start; + struct timeval end; +} syscall_record[1024]; + +int record_syscall_start(int syscall) +{ + int max, index; + + max = sizeof(syscall_record)/sizeof(syscall_record[0]); + index = next_syscall_index(max); + + syscall_record[index].syscall = syscall; + syscall_record[index].pid = current_pid(); + syscall_record[index].result = 0xdeadbeef; + gettimeofday(&syscall_record[index].start, NULL); + return(index); +} + +void record_syscall_end(int index, long result) +{ + syscall_record[index].result = result; + gettimeofday(&syscall_record[index].end, NULL); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/sysrq.c b/arch/um/kernel/sysrq.c new file mode 100644 index 000000000000..e630438f9e73 --- /dev/null +++ b/arch/um/kernel/sysrq.c @@ -0,0 +1,81 @@ +/* + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/kernel.h" +#include "linux/module.h" +#include "linux/kallsyms.h" +#include "asm/page.h" +#include "asm/processor.h" +#include "sysrq.h" +#include "user_util.h" + +void show_trace(unsigned long * stack) +{ + /* XXX: Copy the CONFIG_FRAME_POINTER stack-walking backtrace from + * arch/i386/kernel/traps.c, and then move this to sys-i386/sysrq.c.*/ + unsigned long addr; + + if (!stack) { + stack = (unsigned long*) &stack; + WARN_ON(1); + } + + printk("Call Trace: \n"); + while (((long) stack & (THREAD_SIZE-1)) != 0) { + addr = *stack; + if (__kernel_text_address(addr)) { + printk("%08lx: [<%08lx>]", (unsigned long) stack, addr); + print_symbol(" %s", addr); + printk("\n"); + } + stack++; + } + printk("\n"); +} + +/* + * stack dumps generator - this is used by arch-independent code. + * And this is identical to i386 currently. + */ +void dump_stack(void) +{ + unsigned long stack; + + show_trace(&stack); +} +EXPORT_SYMBOL(dump_stack); + +/*Stolen from arch/i386/kernel/traps.c */ +static int kstack_depth_to_print = 24; + +/* This recently started being used in arch-independent code too, as in + * kernel/sched.c.*/ +void show_stack(struct task_struct *task, unsigned long *esp) +{ + unsigned long *stack; + int i; + + if (esp == NULL) { + if (task != current) { + esp = (unsigned long *) KSTK_ESP(task); + /* Which one? No actual difference - just coding style.*/ + //esp = (unsigned long *) PT_REGS_IP(&task->thread.regs); + } else { + esp = (unsigned long *) &esp; + } + } + + stack = esp; + for(i = 0; i < kstack_depth_to_print; i++) { + if (kstack_end(stack)) + break; + if (i && ((i % 8) == 0)) + printk("\n "); + printk("%08lx ", *stack++); + } + + show_trace(esp); +} diff --git a/arch/um/kernel/tempfile.c b/arch/um/kernel/tempfile.c new file mode 100644 index 000000000000..b1674bc1395d --- /dev/null +++ b/arch/um/kernel/tempfile.c @@ -0,0 +1,82 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <string.h> +#include <errno.h> +#include <sys/param.h> +#include "init.h" + +/* Modified from create_mem_file and start_debugger */ +static char *tempdir = NULL; + +static void __init find_tempdir(void) +{ + char *dirs[] = { "TMP", "TEMP", "TMPDIR", NULL }; + int i; + char *dir = NULL; + + if(tempdir != NULL) return; /* We've already been called */ + for(i = 0; dirs[i]; i++){ + dir = getenv(dirs[i]); + if((dir != NULL) && (*dir != '\0')) + break; + } + if((dir == NULL) || (*dir == '\0')) + dir = "/tmp"; + + tempdir = malloc(strlen(dir) + 2); + if(tempdir == NULL){ + fprintf(stderr, "Failed to malloc tempdir, " + "errno = %d\n", errno); + return; + } + strcpy(tempdir, dir); + strcat(tempdir, "/"); +} + +int make_tempfile(const char *template, char **out_tempname, int do_unlink) +{ + char tempname[MAXPATHLEN]; + int fd; + + find_tempdir(); + if (*template != '/') + strcpy(tempname, tempdir); + else + *tempname = 0; + strcat(tempname, template); + fd = mkstemp(tempname); + if(fd < 0){ + fprintf(stderr, "open - cannot create %s: %s\n", tempname, + strerror(errno)); + return -1; + } + if(do_unlink && (unlink(tempname) < 0)){ + perror("unlink"); + return -1; + } + if(out_tempname){ + *out_tempname = strdup(tempname); + if(*out_tempname == NULL){ + perror("strdup"); + return -1; + } + } + return(fd); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c new file mode 100644 index 000000000000..c40c86a3f918 --- /dev/null +++ b/arch/um/kernel/time.c @@ -0,0 +1,167 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <time.h> +#include <sys/time.h> +#include <signal.h> +#include <errno.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "process.h" +#include "signal_user.h" +#include "time_user.h" +#include "kern_constants.h" + +/* XXX This really needs to be declared and initialized in a kernel file since + * it's in <linux/time.h> + */ +extern struct timespec wall_to_monotonic; + +extern struct timeval xtime; + +struct timeval local_offset = { 0, 0 }; + +void timer(void) +{ + gettimeofday(&xtime, NULL); + timeradd(&xtime, &local_offset, &xtime); +} + +void set_interval(int timer_type) +{ + int usec = 1000000/hz(); + struct itimerval interval = ((struct itimerval) { { 0, usec }, + { 0, usec } }); + + if(setitimer(timer_type, &interval, NULL) == -1) + panic("setitimer failed - errno = %d\n", errno); +} + +void enable_timer(void) +{ + int usec = 1000000/hz(); + struct itimerval enable = ((struct itimerval) { { 0, usec }, + { 0, usec }}); + if(setitimer(ITIMER_VIRTUAL, &enable, NULL)) + printk("enable_timer - setitimer failed, errno = %d\n", + errno); +} + +void disable_timer(void) +{ + struct itimerval disable = ((struct itimerval) { { 0, 0 }, { 0, 0 }}); + if((setitimer(ITIMER_VIRTUAL, &disable, NULL) < 0) || + (setitimer(ITIMER_REAL, &disable, NULL) < 0)) + printk("disnable_timer - setitimer failed, errno = %d\n", + errno); + /* If there are signals already queued, after unblocking ignore them */ + set_handler(SIGALRM, SIG_IGN, 0, -1); + set_handler(SIGVTALRM, SIG_IGN, 0, -1); +} + +void switch_timers(int to_real) +{ + struct itimerval disable = ((struct itimerval) { { 0, 0 }, { 0, 0 }}); + struct itimerval enable = ((struct itimerval) { { 0, 1000000/hz() }, + { 0, 1000000/hz() }}); + int old, new; + + if(to_real){ + old = ITIMER_VIRTUAL; + new = ITIMER_REAL; + } + else { + old = ITIMER_REAL; + new = ITIMER_VIRTUAL; + } + + if((setitimer(old, &disable, NULL) < 0) || + (setitimer(new, &enable, NULL))) + printk("switch_timers - setitimer failed, errno = %d\n", + errno); +} + +void uml_idle_timer(void) +{ + if(signal(SIGVTALRM, SIG_IGN) == SIG_ERR) + panic("Couldn't unset SIGVTALRM handler"); + + set_handler(SIGALRM, (__sighandler_t) alarm_handler, + SA_RESTART, SIGUSR1, SIGIO, SIGWINCH, SIGVTALRM, -1); + set_interval(ITIMER_REAL); +} + +extern int do_posix_clock_monotonic_gettime(struct timespec *tp); + +void time_init(void) +{ + struct timespec now; + + if(signal(SIGVTALRM, boot_timer_handler) == SIG_ERR) + panic("Couldn't set SIGVTALRM handler"); + set_interval(ITIMER_VIRTUAL); + + do_posix_clock_monotonic_gettime(&now); + wall_to_monotonic.tv_sec = -now.tv_sec; + wall_to_monotonic.tv_nsec = -now.tv_nsec; +} + +/* Declared in linux/time.h, which can't be included here */ +extern void clock_was_set(void); + +void do_gettimeofday(struct timeval *tv) +{ + unsigned long flags; + + flags = time_lock(); + gettimeofday(tv, NULL); + timeradd(tv, &local_offset, tv); + time_unlock(flags); + clock_was_set(); +} + +int do_settimeofday(struct timespec *tv) +{ + struct timeval now; + unsigned long flags; + struct timeval tv_in; + + if ((unsigned long) tv->tv_nsec >= UM_NSEC_PER_SEC) + return -EINVAL; + + tv_in.tv_sec = tv->tv_sec; + tv_in.tv_usec = tv->tv_nsec / 1000; + + flags = time_lock(); + gettimeofday(&now, NULL); + timersub(&tv_in, &now, &local_offset); + time_unlock(flags); + + return(0); +} + +void idle_sleep(int secs) +{ + struct timespec ts; + + ts.tv_sec = secs; + ts.tv_nsec = 0; + nanosleep(&ts, NULL); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/time_kern.c b/arch/um/kernel/time_kern.c new file mode 100644 index 000000000000..2461cd73ca87 --- /dev/null +++ b/arch/um/kernel/time_kern.c @@ -0,0 +1,203 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/kernel.h" +#include "linux/module.h" +#include "linux/unistd.h" +#include "linux/stddef.h" +#include "linux/spinlock.h" +#include "linux/time.h" +#include "linux/sched.h" +#include "linux/interrupt.h" +#include "linux/init.h" +#include "linux/delay.h" +#include "asm/irq.h" +#include "asm/param.h" +#include "asm/current.h" +#include "kern_util.h" +#include "user_util.h" +#include "time_user.h" +#include "mode.h" +#include "os.h" + +u64 jiffies_64 = INITIAL_JIFFIES; + +EXPORT_SYMBOL(jiffies_64); + +int hz(void) +{ + return(HZ); +} + +/* + * Scheduler clock - returns current time in nanosec units. + */ +unsigned long long sched_clock(void) +{ + return (unsigned long long)jiffies_64 * (1000000000 / HZ); +} + +/* Changed at early boot */ +int timer_irq_inited = 0; + +static int first_tick; +static unsigned long long prev_usecs; +#ifdef CONFIG_UML_REAL_TIME_CLOCK +static long long delta; /* Deviation per interval */ +#endif + +#define MILLION 1000000 + +void timer_irq(union uml_pt_regs *regs) +{ + unsigned long long ticks = 0; + + if(!timer_irq_inited){ + /* This is to ensure that ticks don't pile up when + * the timer handler is suspended */ + first_tick = 0; + return; + } + + if(first_tick){ +#ifdef CONFIG_UML_REAL_TIME_CLOCK + /* We've had 1 tick */ + unsigned long long usecs = os_usecs(); + + delta += usecs - prev_usecs; + prev_usecs = usecs; + + /* Protect against the host clock being set backwards */ + if(delta < 0) + delta = 0; + + ticks += (delta * HZ) / MILLION; + delta -= (ticks * MILLION) / HZ; +#else + ticks = 1; +#endif + } + else { + prev_usecs = os_usecs(); + first_tick = 1; + } + + while(ticks > 0){ + do_IRQ(TIMER_IRQ, regs); + ticks--; + } +} + +void boot_timer_handler(int sig) +{ + struct pt_regs regs; + + CHOOSE_MODE((void) + (UPT_SC(®s.regs) = (struct sigcontext *) (&sig + 1)), + (void) (regs.regs.skas.is_user = 0)); + do_timer(®s); +} + +irqreturn_t um_timer(int irq, void *dev, struct pt_regs *regs) +{ + unsigned long flags; + + do_timer(regs); + write_seqlock_irqsave(&xtime_lock, flags); + timer(); + write_sequnlock_irqrestore(&xtime_lock, flags); + return(IRQ_HANDLED); +} + +long um_time(int __user *tloc) +{ + struct timeval now; + + do_gettimeofday(&now); + if (tloc) { + if (put_user(now.tv_sec, tloc)) + now.tv_sec = -EFAULT; + } + return now.tv_sec; +} + +long um_stime(int __user *tptr) +{ + int value; + struct timespec new; + + if (get_user(value, tptr)) + return -EFAULT; + new.tv_sec = value; + new.tv_nsec = 0; + do_settimeofday(&new); + return 0; +} + +void __udelay(unsigned long usecs) +{ + int i, n; + + n = (loops_per_jiffy * HZ * usecs) / MILLION; + for(i=0;i<n;i++) ; +} + +void __const_udelay(unsigned long usecs) +{ + int i, n; + + n = (loops_per_jiffy * HZ * usecs) / MILLION; + for(i=0;i<n;i++) ; +} + +void timer_handler(int sig, union uml_pt_regs *regs) +{ + local_irq_disable(); + update_process_times(CHOOSE_MODE(user_context(UPT_SP(regs)), (regs)->skas.is_user)); + local_irq_enable(); + if(current_thread->cpu == 0) + timer_irq(regs); +} + +static DEFINE_SPINLOCK(timer_spinlock); + +unsigned long time_lock(void) +{ + unsigned long flags; + + spin_lock_irqsave(&timer_spinlock, flags); + return(flags); +} + +void time_unlock(unsigned long flags) +{ + spin_unlock_irqrestore(&timer_spinlock, flags); +} + +int __init timer_init(void) +{ + int err; + + CHOOSE_MODE(user_time_init_tt(), user_time_init_skas()); + err = request_irq(TIMER_IRQ, um_timer, SA_INTERRUPT, "timer", NULL); + if(err != 0) + printk(KERN_ERR "timer_init : request_irq failed - " + "errno = %d\n", -err); + timer_irq_inited = 1; + return(0); +} + +__initcall(timer_init); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tlb.c b/arch/um/kernel/tlb.c new file mode 100644 index 000000000000..eda477edfdf5 --- /dev/null +++ b/arch/um/kernel/tlb.c @@ -0,0 +1,369 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/mm.h" +#include "asm/page.h" +#include "asm/pgalloc.h" +#include "asm/tlbflush.h" +#include "choose-mode.h" +#include "mode_kern.h" +#include "user_util.h" +#include "tlb.h" +#include "mem.h" +#include "mem_user.h" +#include "os.h" + +#define ADD_ROUND(n, inc) (((n) + (inc)) & ~((inc) - 1)) + +void fix_range_common(struct mm_struct *mm, unsigned long start_addr, + unsigned long end_addr, int force, int data, + void (*do_ops)(int, struct host_vm_op *, int)) +{ + pgd_t *npgd; + pud_t *npud; + pmd_t *npmd; + pte_t *npte; + unsigned long addr, end; + int r, w, x; + struct host_vm_op ops[16]; + int op_index = -1, last_op = sizeof(ops) / sizeof(ops[0]) - 1; + + if(mm == NULL) return; + + for(addr = start_addr; addr < end_addr;){ + npgd = pgd_offset(mm, addr); + if(!pgd_present(*npgd)){ + end = ADD_ROUND(addr, PGDIR_SIZE); + if(end > end_addr) + end = end_addr; + if(force || pgd_newpage(*npgd)){ + op_index = add_munmap(addr, end - addr, ops, + op_index, last_op, data, + do_ops); + pgd_mkuptodate(*npgd); + } + addr = end; + continue; + } + + npud = pud_offset(npgd, addr); + if(!pud_present(*npud)){ + end = ADD_ROUND(addr, PUD_SIZE); + if(end > end_addr) + end = end_addr; + if(force || pud_newpage(*npud)){ + op_index = add_munmap(addr, end - addr, ops, + op_index, last_op, data, + do_ops); + pud_mkuptodate(*npud); + } + addr = end; + continue; + } + + npmd = pmd_offset(npud, addr); + if(!pmd_present(*npmd)){ + end = ADD_ROUND(addr, PMD_SIZE); + if(end > end_addr) + end = end_addr; + if(force || pmd_newpage(*npmd)){ + op_index = add_munmap(addr, end - addr, ops, + op_index, last_op, data, + do_ops); + pmd_mkuptodate(*npmd); + } + addr = end; + continue; + } + + npte = pte_offset_kernel(npmd, addr); + r = pte_read(*npte); + w = pte_write(*npte); + x = pte_exec(*npte); + if(!pte_dirty(*npte)) + w = 0; + if(!pte_young(*npte)){ + r = 0; + w = 0; + } + if(force || pte_newpage(*npte)){ + if(pte_present(*npte)) + op_index = add_mmap(addr, + pte_val(*npte) & PAGE_MASK, + PAGE_SIZE, r, w, x, ops, + op_index, last_op, data, + do_ops); + else op_index = add_munmap(addr, PAGE_SIZE, ops, + op_index, last_op, data, + do_ops); + } + else if(pte_newprot(*npte)) + op_index = add_mprotect(addr, PAGE_SIZE, r, w, x, ops, + op_index, last_op, data, + do_ops); + + *npte = pte_mkuptodate(*npte); + addr += PAGE_SIZE; + } + (*do_ops)(data, ops, op_index); +} + +int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) +{ + struct mm_struct *mm; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + unsigned long addr, last; + int updated = 0, err; + + mm = &init_mm; + for(addr = start; addr < end;){ + pgd = pgd_offset(mm, addr); + if(!pgd_present(*pgd)){ + last = ADD_ROUND(addr, PGDIR_SIZE); + if(last > end) + last = end; + if(pgd_newpage(*pgd)){ + updated = 1; + err = os_unmap_memory((void *) addr, + last - addr); + if(err < 0) + panic("munmap failed, errno = %d\n", + -err); + } + addr = last; + continue; + } + + pud = pud_offset(pgd, addr); + if(!pud_present(*pud)){ + last = ADD_ROUND(addr, PUD_SIZE); + if(last > end) + last = end; + if(pud_newpage(*pud)){ + updated = 1; + err = os_unmap_memory((void *) addr, + last - addr); + if(err < 0) + panic("munmap failed, errno = %d\n", + -err); + } + addr = last; + continue; + } + + pmd = pmd_offset(pud, addr); + if(!pmd_present(*pmd)){ + last = ADD_ROUND(addr, PMD_SIZE); + if(last > end) + last = end; + if(pmd_newpage(*pmd)){ + updated = 1; + err = os_unmap_memory((void *) addr, + last - addr); + if(err < 0) + panic("munmap failed, errno = %d\n", + -err); + } + addr = last; + continue; + } + + pte = pte_offset_kernel(pmd, addr); + if(!pte_present(*pte) || pte_newpage(*pte)){ + updated = 1; + err = os_unmap_memory((void *) addr, + PAGE_SIZE); + if(err < 0) + panic("munmap failed, errno = %d\n", + -err); + if(pte_present(*pte)) + map_memory(addr, + pte_val(*pte) & PAGE_MASK, + PAGE_SIZE, 1, 1, 1); + } + else if(pte_newprot(*pte)){ + updated = 1; + protect_memory(addr, PAGE_SIZE, 1, 1, 1, 1); + } + addr += PAGE_SIZE; + } + return(updated); +} + +void flush_tlb_page(struct vm_area_struct *vma, unsigned long address) +{ + address &= PAGE_MASK; + flush_tlb_range(vma, address, address + PAGE_SIZE); +} + +void flush_tlb_all(void) +{ + flush_tlb_mm(current->mm); +} + +void flush_tlb_kernel_range(unsigned long start, unsigned long end) +{ + CHOOSE_MODE_PROC(flush_tlb_kernel_range_tt, + flush_tlb_kernel_range_common, start, end); +} + +void flush_tlb_kernel_vm(void) +{ + CHOOSE_MODE(flush_tlb_kernel_vm_tt(), + flush_tlb_kernel_range_common(start_vm, end_vm)); +} + +void __flush_tlb_one(unsigned long addr) +{ + CHOOSE_MODE_PROC(__flush_tlb_one_tt, __flush_tlb_one_skas, addr); +} + +void flush_tlb_range(struct vm_area_struct *vma, unsigned long start, + unsigned long end) +{ + CHOOSE_MODE_PROC(flush_tlb_range_tt, flush_tlb_range_skas, vma, start, + end); +} + +void flush_tlb_mm(struct mm_struct *mm) +{ + CHOOSE_MODE_PROC(flush_tlb_mm_tt, flush_tlb_mm_skas, mm); +} + +void force_flush_all(void) +{ + CHOOSE_MODE(force_flush_all_tt(), force_flush_all_skas()); +} + +pgd_t *pgd_offset_proc(struct mm_struct *mm, unsigned long address) +{ + return(pgd_offset(mm, address)); +} + +pud_t *pud_offset_proc(pgd_t *pgd, unsigned long address) +{ + return(pud_offset(pgd, address)); +} + +pmd_t *pmd_offset_proc(pud_t *pud, unsigned long address) +{ + return(pmd_offset(pud, address)); +} + +pte_t *pte_offset_proc(pmd_t *pmd, unsigned long address) +{ + return(pte_offset_kernel(pmd, address)); +} + +pte_t *addr_pte(struct task_struct *task, unsigned long addr) +{ + pgd_t *pgd = pgd_offset(task->mm, addr); + pud_t *pud = pud_offset(pgd, addr); + pmd_t *pmd = pmd_offset(pud, addr); + + return(pte_offset_map(pmd, addr)); +} + +int add_mmap(unsigned long virt, unsigned long phys, unsigned long len, + int r, int w, int x, struct host_vm_op *ops, int index, + int last_filled, int data, + void (*do_ops)(int, struct host_vm_op *, int)) +{ + __u64 offset; + struct host_vm_op *last; + int fd; + + fd = phys_mapping(phys, &offset); + if(index != -1){ + last = &ops[index]; + if((last->type == MMAP) && + (last->u.mmap.addr + last->u.mmap.len == virt) && + (last->u.mmap.r == r) && (last->u.mmap.w == w) && + (last->u.mmap.x == x) && (last->u.mmap.fd == fd) && + (last->u.mmap.offset + last->u.mmap.len == offset)){ + last->u.mmap.len += len; + return(index); + } + } + + if(index == last_filled){ + (*do_ops)(data, ops, last_filled); + index = -1; + } + + ops[++index] = ((struct host_vm_op) { .type = MMAP, + .u = { .mmap = { + .addr = virt, + .len = len, + .r = r, + .w = w, + .x = x, + .fd = fd, + .offset = offset } + } }); + return(index); +} + +int add_munmap(unsigned long addr, unsigned long len, struct host_vm_op *ops, + int index, int last_filled, int data, + void (*do_ops)(int, struct host_vm_op *, int)) +{ + struct host_vm_op *last; + + if(index != -1){ + last = &ops[index]; + if((last->type == MUNMAP) && + (last->u.munmap.addr + last->u.mmap.len == addr)){ + last->u.munmap.len += len; + return(index); + } + } + + if(index == last_filled){ + (*do_ops)(data, ops, last_filled); + index = -1; + } + + ops[++index] = ((struct host_vm_op) { .type = MUNMAP, + .u = { .munmap = { + .addr = addr, + .len = len } } }); + return(index); +} + +int add_mprotect(unsigned long addr, unsigned long len, int r, int w, int x, + struct host_vm_op *ops, int index, int last_filled, int data, + void (*do_ops)(int, struct host_vm_op *, int)) +{ + struct host_vm_op *last; + + if(index != -1){ + last = &ops[index]; + if((last->type == MPROTECT) && + (last->u.mprotect.addr + last->u.mprotect.len == addr) && + (last->u.mprotect.r == r) && (last->u.mprotect.w == w) && + (last->u.mprotect.x == x)){ + last->u.mprotect.len += len; + return(index); + } + } + + if(index == last_filled){ + (*do_ops)(data, ops, last_filled); + index = -1; + } + + ops[++index] = ((struct host_vm_op) { .type = MPROTECT, + .u = { .mprotect = { + .addr = addr, + .len = len, + .r = r, + .w = w, + .x = x } } }); + return(index); +} diff --git a/arch/um/kernel/trap_kern.c b/arch/um/kernel/trap_kern.c new file mode 100644 index 000000000000..47e766e6ba10 --- /dev/null +++ b/arch/um/kernel/trap_kern.c @@ -0,0 +1,251 @@ +/* + * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/kernel.h" +#include "asm/errno.h" +#include "linux/sched.h" +#include "linux/mm.h" +#include "linux/spinlock.h" +#include "linux/config.h" +#include "linux/init.h" +#include "linux/ptrace.h" +#include "asm/semaphore.h" +#include "asm/pgtable.h" +#include "asm/pgalloc.h" +#include "asm/tlbflush.h" +#include "asm/a.out.h" +#include "asm/current.h" +#include "asm/irq.h" +#include "user_util.h" +#include "kern_util.h" +#include "kern.h" +#include "chan_kern.h" +#include "mconsole_kern.h" +#include "2_5compat.h" +#include "mem.h" +#include "mem_kern.h" + +int handle_page_fault(unsigned long address, unsigned long ip, + int is_write, int is_user, int *code_out) +{ + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + unsigned long page; + int err = -EFAULT; + + *code_out = SEGV_MAPERR; + down_read(&mm->mmap_sem); + vma = find_vma(mm, address); + if(!vma) + goto out; + else if(vma->vm_start <= address) + goto good_area; + else if(!(vma->vm_flags & VM_GROWSDOWN)) + goto out; + else if(!ARCH_IS_STACKGROW(address)) + goto out; + else if(expand_stack(vma, address)) + goto out; + + good_area: + *code_out = SEGV_ACCERR; + if(is_write && !(vma->vm_flags & VM_WRITE)) + goto out; + page = address & PAGE_MASK; + pgd = pgd_offset(mm, page); + pud = pud_offset(pgd, page); + pmd = pmd_offset(pud, page); + do { + survive: + switch (handle_mm_fault(mm, vma, address, is_write)){ + case VM_FAULT_MINOR: + current->min_flt++; + break; + case VM_FAULT_MAJOR: + current->maj_flt++; + break; + case VM_FAULT_SIGBUS: + err = -EACCES; + goto out; + case VM_FAULT_OOM: + err = -ENOMEM; + goto out_of_memory; + default: + BUG(); + } + pgd = pgd_offset(mm, page); + pud = pud_offset(pgd, page); + pmd = pmd_offset(pud, page); + pte = pte_offset_kernel(pmd, page); + } while(!pte_present(*pte)); + err = 0; + *pte = pte_mkyoung(*pte); + if(pte_write(*pte)) *pte = pte_mkdirty(*pte); + flush_tlb_page(vma, page); + out: + up_read(&mm->mmap_sem); + return(err); + +/* + * We ran out of memory, or some other thing happened to us that made + * us unable to handle the page fault gracefully. + */ +out_of_memory: + if (current->pid == 1) { + up_read(&mm->mmap_sem); + yield(); + down_read(&mm->mmap_sem); + goto survive; + } + goto out; +} + +LIST_HEAD(physmem_remappers); + +void register_remapper(struct remapper *info) +{ + list_add(&info->list, &physmem_remappers); +} + +static int check_remapped_addr(unsigned long address, int is_write) +{ + struct remapper *remapper; + struct list_head *ele; + __u64 offset; + int fd; + + fd = phys_mapping(__pa(address), &offset); + if(fd == -1) + return(0); + + list_for_each(ele, &physmem_remappers){ + remapper = list_entry(ele, struct remapper, list); + if((*remapper->proc)(fd, address, is_write, offset)) + return(1); + } + + return(0); +} + +unsigned long segv(unsigned long address, unsigned long ip, int is_write, + int is_user, void *sc) +{ + struct siginfo si; + void *catcher; + int err; + + if(!is_user && (address >= start_vm) && (address < end_vm)){ + flush_tlb_kernel_vm(); + return(0); + } + else if(check_remapped_addr(address & PAGE_MASK, is_write)) + return(0); + else if(current->mm == NULL) + panic("Segfault with no mm"); + err = handle_page_fault(address, ip, is_write, is_user, &si.si_code); + + catcher = current->thread.fault_catcher; + if(!err) + return(0); + else if(catcher != NULL){ + current->thread.fault_addr = (void *) address; + do_longjmp(catcher, 1); + } + else if(current->thread.fault_addr != NULL) + panic("fault_addr set but no fault catcher"); + else if(arch_fixup(ip, sc)) + return(0); + + if(!is_user) + panic("Kernel mode fault at addr 0x%lx, ip 0x%lx", + address, ip); + + if(err == -EACCES){ + si.si_signo = SIGBUS; + si.si_errno = 0; + si.si_code = BUS_ADRERR; + si.si_addr = (void *)address; + force_sig_info(SIGBUS, &si, current); + } + else if(err == -ENOMEM){ + printk("VM: killing process %s\n", current->comm); + do_exit(SIGKILL); + } + else { + si.si_signo = SIGSEGV; + si.si_addr = (void *) address; + current->thread.cr2 = address; + current->thread.err = is_write; + force_sig_info(SIGSEGV, &si, current); + } + return(0); +} + +void bad_segv(unsigned long address, unsigned long ip, int is_write) +{ + struct siginfo si; + + si.si_signo = SIGSEGV; + si.si_code = SEGV_ACCERR; + si.si_addr = (void *) address; + current->thread.cr2 = address; + current->thread.err = is_write; + force_sig_info(SIGSEGV, &si, current); +} + +void relay_signal(int sig, union uml_pt_regs *regs) +{ + if(arch_handle_signal(sig, regs)) return; + if(!UPT_IS_USER(regs)) + panic("Kernel mode signal %d", sig); + force_sig(sig, current); +} + +void bus_handler(int sig, union uml_pt_regs *regs) +{ + if(current->thread.fault_catcher != NULL) + do_longjmp(current->thread.fault_catcher, 1); + else relay_signal(sig, regs); +} + +void winch(int sig, union uml_pt_regs *regs) +{ + do_IRQ(WINCH_IRQ, regs); +} + +void trap_init(void) +{ +} + +DEFINE_SPINLOCK(trap_lock); + +static int trap_index = 0; + +int next_trap_index(int limit) +{ + int ret; + + spin_lock(&trap_lock); + ret = trap_index; + if(++trap_index == limit) + trap_index = 0; + spin_unlock(&trap_lock); + return(ret); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/trap_user.c b/arch/um/kernel/trap_user.c new file mode 100644 index 000000000000..50a4042a509f --- /dev/null +++ b/arch/um/kernel/trap_user.c @@ -0,0 +1,120 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <errno.h> +#include <setjmp.h> +#include <signal.h> +#include <sys/time.h> +#include <sys/wait.h> +#include <asm/page.h> +#include <asm/unistd.h> +#include <asm/ptrace.h> +#include "init.h" +#include "sysdep/ptrace.h" +#include "sigcontext.h" +#include "sysdep/sigcontext.h" +#include "irq_user.h" +#include "signal_user.h" +#include "time_user.h" +#include "task.h" +#include "mode.h" +#include "choose-mode.h" +#include "kern_util.h" +#include "user_util.h" +#include "os.h" + +void kill_child_dead(int pid) +{ + kill(pid, SIGKILL); + kill(pid, SIGCONT); + do { + int n; + CATCH_EINTR(n = waitpid(pid, NULL, 0)); + if (n > 0) + kill(pid, SIGCONT); + else + break; + } while(1); +} + +/* Unlocked - don't care if this is a bit off */ +int nsegfaults = 0; + +struct { + unsigned long address; + int is_write; + int pid; + unsigned long sp; + int is_user; +} segfault_record[1024]; + +void segv_handler(int sig, union uml_pt_regs *regs) +{ + int index, max; + + if(UPT_IS_USER(regs) && !UPT_SEGV_IS_FIXABLE(regs)){ + bad_segv(UPT_FAULT_ADDR(regs), UPT_IP(regs), + UPT_FAULT_WRITE(regs)); + return; + } + max = sizeof(segfault_record)/sizeof(segfault_record[0]); + index = next_trap_index(max); + + nsegfaults++; + segfault_record[index].address = UPT_FAULT_ADDR(regs); + segfault_record[index].pid = os_getpid(); + segfault_record[index].is_write = UPT_FAULT_WRITE(regs); + segfault_record[index].sp = UPT_SP(regs); + segfault_record[index].is_user = UPT_IS_USER(regs); + segv(UPT_FAULT_ADDR(regs), UPT_IP(regs), UPT_FAULT_WRITE(regs), + UPT_IS_USER(regs), regs); +} + +void usr2_handler(int sig, union uml_pt_regs *regs) +{ + CHOOSE_MODE(syscall_handler_tt(sig, regs), (void) 0); +} + +struct signal_info sig_info[] = { + [ SIGTRAP ] { .handler = relay_signal, + .is_irq = 0 }, + [ SIGFPE ] { .handler = relay_signal, + .is_irq = 0 }, + [ SIGILL ] { .handler = relay_signal, + .is_irq = 0 }, + [ SIGWINCH ] { .handler = winch, + .is_irq = 1 }, + [ SIGBUS ] { .handler = bus_handler, + .is_irq = 0 }, + [ SIGSEGV] { .handler = segv_handler, + .is_irq = 0 }, + [ SIGIO ] { .handler = sigio_handler, + .is_irq = 1 }, + [ SIGVTALRM ] { .handler = timer_handler, + .is_irq = 1 }, + [ SIGALRM ] { .handler = timer_handler, + .is_irq = 1 }, + [ SIGUSR2 ] { .handler = usr2_handler, + .is_irq = 0 }, +}; + +void do_longjmp(void *b, int val) +{ + sigjmp_buf *buf = b; + + siglongjmp(*buf, val); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/Makefile b/arch/um/kernel/tt/Makefile new file mode 100644 index 000000000000..3d5177df3504 --- /dev/null +++ b/arch/um/kernel/tt/Makefile @@ -0,0 +1,28 @@ +# +# Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com) +# Licensed under the GPL +# + +extra-y := unmap_fin.o +clean-files := unmap_tmp.o + +obj-y = exec_kern.o exec_user.o gdb.o ksyms.o mem.o mem_user.o process_kern.o \ + syscall_kern.o syscall_user.o time.o tlb.o tracer.o trap_user.o \ + uaccess.o uaccess_user.o + +obj-$(CONFIG_PT_PROXY) += gdb_kern.o ptproxy/ + +USER_OBJS := gdb.o time.o tracer.o + +include arch/um/scripts/Makefile.rules + +UNMAP_CFLAGS := $(patsubst -pg -DPROFILING,,$(USER_CFLAGS)) +UNMAP_CFLAGS := $(patsubst -fprofile-arcs -ftest-coverage,,$(UNMAP_CFLAGS)) + +#XXX: partially copied from arch/um/scripts/Makefile.rules +$(obj)/unmap.o: c_flags = -Wp,-MD,$(depfile) $(UNMAP_CFLAGS) + +$(obj)/unmap_fin.o : $(obj)/unmap.o + $(LD) -r -o $(obj)/unmap_tmp.o $< $(shell $(CC) -print-file-name=libc.a) + $(OBJCOPY) $(obj)/unmap_tmp.o $@ -G switcheroo + diff --git a/arch/um/kernel/tt/exec_kern.c b/arch/um/kernel/tt/exec_kern.c new file mode 100644 index 000000000000..065b504a653b --- /dev/null +++ b/arch/um/kernel/tt/exec_kern.c @@ -0,0 +1,87 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/kernel.h" +#include "linux/mm.h" +#include "asm/signal.h" +#include "asm/ptrace.h" +#include "asm/uaccess.h" +#include "asm/pgalloc.h" +#include "asm/tlbflush.h" +#include "user_util.h" +#include "kern_util.h" +#include "irq_user.h" +#include "time_user.h" +#include "signal_user.h" +#include "mem_user.h" +#include "os.h" +#include "tlb.h" +#include "mode.h" + +static int exec_tramp(void *sig_stack) +{ + init_new_thread_stack(sig_stack, NULL); + init_new_thread_signals(1); + os_stop_process(os_getpid()); + return(0); +} + +void flush_thread_tt(void) +{ + unsigned long stack; + int new_pid; + + stack = alloc_stack(0, 0); + if(stack == 0){ + printk(KERN_ERR + "flush_thread : failed to allocate temporary stack\n"); + do_exit(SIGKILL); + } + + new_pid = start_fork_tramp(current->thread_info, stack, 0, exec_tramp); + if(new_pid < 0){ + printk(KERN_ERR + "flush_thread : new thread failed, errno = %d\n", + -new_pid); + do_exit(SIGKILL); + } + + if(current_thread->cpu == 0) + forward_interrupts(new_pid); + current->thread.request.op = OP_EXEC; + current->thread.request.u.exec.pid = new_pid; + unprotect_stack((unsigned long) current_thread); + os_usr1_process(os_getpid()); + change_sig(SIGUSR1, 1); + + change_sig(SIGUSR1, 0); + enable_timer(); + free_page(stack); + protect_memory(uml_reserved, high_physmem - uml_reserved, 1, 1, 0, 1); + task_protections((unsigned long) current_thread); + force_flush_all(); + unblock_signals(); +} + +void start_thread_tt(struct pt_regs *regs, unsigned long eip, + unsigned long esp) +{ + set_fs(USER_DS); + flush_tlb_mm(current->mm); + PT_REGS_IP(regs) = eip; + PT_REGS_SP(regs) = esp; + PT_FIX_EXEC_STACK(esp); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/exec_user.c b/arch/um/kernel/tt/exec_user.c new file mode 100644 index 000000000000..a92c02ff2ce3 --- /dev/null +++ b/arch/um/kernel/tt/exec_user.c @@ -0,0 +1,57 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <unistd.h> +#include <stdlib.h> +#include <sched.h> +#include <errno.h> +#include <sys/wait.h> +#include <signal.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "ptrace_user.h" +#include "os.h" + +void do_exec(int old_pid, int new_pid) +{ + unsigned long regs[FRAME_SIZE]; + int err; + + if((ptrace(PTRACE_ATTACH, new_pid, 0, 0) < 0) || + (ptrace(PTRACE_CONT, new_pid, 0, 0) < 0)) + tracer_panic("do_exec failed to attach proc - errno = %d", + errno); + + CATCH_EINTR(err = waitpid(new_pid, 0, WUNTRACED)); + if (err < 0) + tracer_panic("do_exec failed to attach proc in waitpid - errno = %d", + errno); + + if(ptrace_getregs(old_pid, regs) < 0) + tracer_panic("do_exec failed to get registers - errno = %d", + errno); + + os_kill_ptraced_process(old_pid, 0); + + if (ptrace(PTRACE_OLDSETOPTIONS, new_pid, 0, (void *)PTRACE_O_TRACESYSGOOD) < 0) + tracer_panic("do_exec: PTRACE_SETOPTIONS failed, errno = %d", errno); + + if(ptrace_setregs(new_pid, regs) < 0) + tracer_panic("do_exec failed to start new proc - errno = %d", + errno); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/gdb.c b/arch/um/kernel/tt/gdb.c new file mode 100644 index 000000000000..19a0ad7b35b3 --- /dev/null +++ b/arch/um/kernel/tt/gdb.c @@ -0,0 +1,278 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <errno.h> +#include <string.h> +#include <signal.h> +#include <sys/types.h> +#include "ptrace_user.h" +#include "uml-config.h" +#include "kern_constants.h" +#include "chan_user.h" +#include "init.h" +#include "user.h" +#include "debug.h" +#include "kern_util.h" +#include "user_util.h" +#include "tt.h" +#include "sysdep/thread.h" + +extern int debugger_pid; +extern int debugger_fd; +extern int debugger_parent; + +int detach(int pid, int sig) +{ + return(ptrace(PTRACE_DETACH, pid, 0, sig)); +} + +int attach(int pid) +{ + int err; + + err = ptrace(PTRACE_ATTACH, pid, 0, 0); + if(err < 0) return(-errno); + else return(err); +} + +int cont(int pid) +{ + return(ptrace(PTRACE_CONT, pid, 0, 0)); +} + +#ifdef UML_CONFIG_PT_PROXY + +int debugger_signal(int status, pid_t pid) +{ + return(debugger_proxy(status, pid)); +} + +void child_signal(pid_t pid, int status) +{ + child_proxy(pid, status); +} + +static void gdb_announce(char *dev_name, int dev) +{ + printf("gdb assigned device '%s'\n", dev_name); +} + +static struct chan_opts opts = { + .announce = gdb_announce, + .xterm_title = "UML kernel debugger", + .raw = 0, + .tramp_stack = 0, + .in_kernel = 0, +}; + +/* Accessed by the tracing thread, which automatically serializes access */ +static void *xterm_data; +static int xterm_fd; + +extern void *xterm_init(char *, int, struct chan_opts *); +extern int xterm_open(int, int, int, void *, char **); +extern void xterm_close(int, void *); + +int open_gdb_chan(void) +{ + char stack[UM_KERN_PAGE_SIZE], *dummy; + + opts.tramp_stack = (unsigned long) stack; + xterm_data = xterm_init("", 0, &opts); + xterm_fd = xterm_open(1, 1, 1, xterm_data, &dummy); + return(xterm_fd); +} + +static void exit_debugger_cb(void *unused) +{ + if(debugger_pid != -1){ + if(gdb_pid != -1){ + fake_child_exit(); + gdb_pid = -1; + } + else kill_child_dead(debugger_pid); + debugger_pid = -1; + if(debugger_parent != -1) + detach(debugger_parent, SIGINT); + } + if(xterm_data != NULL) xterm_close(xterm_fd, xterm_data); +} + +static void exit_debugger(void) +{ + initial_thread_cb(exit_debugger_cb, NULL); +} + +__uml_exitcall(exit_debugger); + +struct gdb_data { + char *str; + int err; +}; + +static void config_gdb_cb(void *arg) +{ + struct gdb_data *data = arg; + void *task; + int pid; + + data->err = -1; + if(debugger_pid != -1) exit_debugger_cb(NULL); + if(!strncmp(data->str, "pid,", strlen("pid,"))){ + data->str += strlen("pid,"); + pid = strtoul(data->str, NULL, 0); + task = cpu_tasks[0].task; + debugger_pid = attach_debugger(TASK_EXTERN_PID(task), pid, 0); + if(debugger_pid != -1){ + data->err = 0; + gdb_pid = pid; + } + return; + } + data->err = 0; + debugger_pid = start_debugger(linux_prog, 0, 0, &debugger_fd); + init_proxy(debugger_pid, 0, 0); +} + +int gdb_config(char *str) +{ + struct gdb_data data; + + if(*str++ != '=') return(-1); + data.str = str; + initial_thread_cb(config_gdb_cb, &data); + return(data.err); +} + +void remove_gdb_cb(void *unused) +{ + exit_debugger_cb(NULL); +} + +int gdb_remove(char *unused) +{ + initial_thread_cb(remove_gdb_cb, NULL); + return(0); +} + +void signal_usr1(int sig) +{ + if(debugger_pid != -1){ + printf("The debugger is already running\n"); + return; + } + debugger_pid = start_debugger(linux_prog, 0, 0, &debugger_fd); + init_proxy(debugger_pid, 0, 0); +} + +int init_ptrace_proxy(int idle_pid, int startup, int stop) +{ + int pid, status; + + pid = start_debugger(linux_prog, startup, stop, &debugger_fd); + status = wait_for_stop(idle_pid, SIGSTOP, PTRACE_CONT, NULL); + if(pid < 0){ + cont(idle_pid); + return(-1); + } + init_proxy(pid, 1, status); + return(pid); +} + +int attach_debugger(int idle_pid, int pid, int stop) +{ + int status = 0, err; + + err = attach(pid); + if(err < 0){ + printf("Failed to attach pid %d, errno = %d\n", pid, -err); + return(-1); + } + if(stop) status = wait_for_stop(idle_pid, SIGSTOP, PTRACE_CONT, NULL); + init_proxy(pid, 1, status); + return(pid); +} + +#ifdef notdef /* Put this back in when it does something useful */ +static int __init uml_gdb_init_setup(char *line, int *add) +{ + gdb_init = uml_strdup(line); + return 0; +} + +__uml_setup("gdb=", uml_gdb_init_setup, +"gdb=<channel description>\n\n" +); +#endif + +static int __init uml_gdb_pid_setup(char *line, int *add) +{ + gdb_pid = strtoul(line, NULL, 0); + *add = 0; + return 0; +} + +__uml_setup("gdb-pid=", uml_gdb_pid_setup, +"gdb-pid=<pid>\n" +" gdb-pid is used to attach an external debugger to UML. This may be\n" +" an already-running gdb or a debugger-like process like strace.\n\n" +); + +#else + +int debugger_signal(int status, pid_t pid){ return(0); } +void child_signal(pid_t pid, int status){ } +int init_ptrace_proxy(int idle_pid, int startup, int stop) +{ + printf("debug requested when CONFIG_PT_PROXY is off\n"); + kill_child_dead(idle_pid); + exit(1); +} + +void signal_usr1(int sig) +{ + printf("debug requested when CONFIG_PT_PROXY is off\n"); +} + +int attach_debugger(int idle_pid, int pid, int stop) +{ + printf("attach_debugger called when CONFIG_PT_PROXY " + "is off\n"); + return(-1); +} + +int config_gdb(char *str) +{ + return(-1); +} + +int remove_gdb(void) +{ + return(-1); +} + +int init_parent_proxy(int pid) +{ + return(-1); +} + +void debugger_parent_signal(int status, int pid) +{ +} + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/gdb_kern.c b/arch/um/kernel/tt/gdb_kern.c new file mode 100644 index 000000000000..93fb121f86af --- /dev/null +++ b/arch/um/kernel/tt/gdb_kern.c @@ -0,0 +1,40 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/init.h" +#include "linux/config.h" +#include "mconsole_kern.h" + +#ifdef CONFIG_MCONSOLE + +extern int gdb_config(char *str); +extern int gdb_remove(char *unused); + +static struct mc_device gdb_mc = { + .name = "gdb", + .config = gdb_config, + .remove = gdb_remove, +}; + +int gdb_mc_init(void) +{ + mconsole_register_dev(&gdb_mc); + return(0); +} + +__initcall(gdb_mc_init); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/include/debug.h b/arch/um/kernel/tt/include/debug.h new file mode 100644 index 000000000000..8eff674107ca --- /dev/null +++ b/arch/um/kernel/tt/include/debug.h @@ -0,0 +1,29 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) and + * Lars Brinkhoff. + * Licensed under the GPL + */ + +#ifndef __DEBUG_H +#define __DEBUG_H + +extern int debugger_proxy(int status, pid_t pid); +extern void child_proxy(pid_t pid, int status); +extern void init_proxy (pid_t pid, int waiting, int status); +extern int start_debugger(char *prog, int startup, int stop, int *debugger_fd); +extern void fake_child_exit(void); +extern int gdb_config(char *str); +extern int gdb_remove(char *unused); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/include/mmu-tt.h b/arch/um/kernel/tt/include/mmu-tt.h new file mode 100644 index 000000000000..0440510ab3fe --- /dev/null +++ b/arch/um/kernel/tt/include/mmu-tt.h @@ -0,0 +1,23 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __TT_MMU_H +#define __TT_MMU_H + +struct mmu_context_tt { +}; + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/include/mode-tt.h b/arch/um/kernel/tt/include/mode-tt.h new file mode 100644 index 000000000000..efe462019069 --- /dev/null +++ b/arch/um/kernel/tt/include/mode-tt.h @@ -0,0 +1,35 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __MODE_TT_H__ +#define __MODE_TT_H__ + +#include "sysdep/ptrace.h" + +enum { OP_NONE, OP_EXEC, OP_FORK, OP_TRACE_ON, OP_REBOOT, OP_HALT, OP_CB }; + +extern int tracing_pid; + +extern int tracer(int (*init_proc)(void *), void *sp); +extern void user_time_init_tt(void); +extern void sig_handler_common_tt(int sig, void *sc); +extern void syscall_handler_tt(int sig, union uml_pt_regs *regs); +extern void reboot_tt(void); +extern void halt_tt(void); +extern int is_tracer_winch(int pid, int fd, void *data); +extern void kill_off_processes_tt(void); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/include/mode_kern-tt.h b/arch/um/kernel/tt/include/mode_kern-tt.h new file mode 100644 index 000000000000..28aaab3448fa --- /dev/null +++ b/arch/um/kernel/tt/include/mode_kern-tt.h @@ -0,0 +1,53 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __TT_MODE_KERN_H__ +#define __TT_MODE_KERN_H__ + +#include "linux/sched.h" +#include "asm/page.h" +#include "asm/ptrace.h" +#include "asm/uaccess.h" + +extern void *switch_to_tt(void *prev, void *next); +extern void flush_thread_tt(void); +extern void start_thread_tt(struct pt_regs *regs, unsigned long eip, + unsigned long esp); +extern int copy_thread_tt(int nr, unsigned long clone_flags, unsigned long sp, + unsigned long stack_top, struct task_struct *p, + struct pt_regs *regs); +extern void release_thread_tt(struct task_struct *task); +extern void exit_thread_tt(void); +extern void initial_thread_cb_tt(void (*proc)(void *), void *arg); +extern void init_idle_tt(void); +extern void flush_tlb_kernel_range_tt(unsigned long start, unsigned long end); +extern void flush_tlb_kernel_vm_tt(void); +extern void __flush_tlb_one_tt(unsigned long addr); +extern void flush_tlb_range_tt(struct vm_area_struct *vma, + unsigned long start, unsigned long end); +extern void flush_tlb_mm_tt(struct mm_struct *mm); +extern void force_flush_all_tt(void); +extern long execute_syscall_tt(void *r); +extern void before_mem_tt(unsigned long brk_start); +extern unsigned long set_task_sizes_tt(int arg, unsigned long *host_size_out, + unsigned long *task_size_out); +extern int start_uml_tt(void); +extern int external_pid_tt(struct task_struct *task); +extern int thread_pid_tt(struct task_struct *task); + +#define kmem_end_tt (host_task_size - ABOVE_KMEM) + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/include/tt.h b/arch/um/kernel/tt/include/tt.h new file mode 100644 index 000000000000..c667b67af405 --- /dev/null +++ b/arch/um/kernel/tt/include/tt.h @@ -0,0 +1,46 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __TT_H__ +#define __TT_H__ + +#include "sysdep/ptrace.h" + +extern int gdb_pid; +extern int debug; +extern int debug_stop; +extern int debug_trace; + +extern int honeypot; + +extern int fork_tramp(void *sig_stack); +extern int do_proc_op(void *t, int proc_id); +extern int tracer(int (*init_proc)(void *), void *sp); +extern void attach_process(int pid); +extern void tracer_panic(char *format, ...); +extern void set_init_pid(int pid); +extern int set_user_mode(void *task); +extern void set_tracing(void *t, int tracing); +extern int is_tracing(void *task); +extern void syscall_handler(int sig, union uml_pt_regs *regs); +extern void exit_kernel(int pid, void *task); +extern void do_syscall(void *task, int pid, int local_using_sysemu); +extern void do_sigtrap(void *task); +extern int is_valid_pid(int pid); +extern void remap_data(void *segment_start, void *segment_end, int w); +extern long execute_syscall_tt(void *r); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/include/uaccess-tt.h b/arch/um/kernel/tt/include/uaccess-tt.h new file mode 100644 index 000000000000..f0bad010cebd --- /dev/null +++ b/arch/um/kernel/tt/include/uaccess-tt.h @@ -0,0 +1,71 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#ifndef __TT_UACCESS_H +#define __TT_UACCESS_H + +#include "linux/string.h" +#include "linux/sched.h" +#include "asm/processor.h" +#include "asm/errno.h" +#include "asm/current.h" +#include "asm/a.out.h" +#include "uml_uaccess.h" + +#define ABOVE_KMEM (16 * 1024 * 1024) + +extern unsigned long end_vm; +extern unsigned long uml_physmem; + +#define under_task_size(addr, size) \ + (((unsigned long) (addr) < TASK_SIZE) && \ + (((unsigned long) (addr) + (size)) < TASK_SIZE)) + +#define is_stack(addr, size) \ + (((unsigned long) (addr) < STACK_TOP) && \ + ((unsigned long) (addr) >= STACK_TOP - ABOVE_KMEM) && \ + (((unsigned long) (addr) + (size)) <= STACK_TOP)) + +#define access_ok_tt(type, addr, size) \ + ((type == VERIFY_READ) || (segment_eq(get_fs(), KERNEL_DS)) || \ + (((unsigned long) (addr) <= ((unsigned long) (addr) + (size))) && \ + (under_task_size(addr, size) || is_stack(addr, size)))) + +static inline int verify_area_tt(int type, const void * addr, + unsigned long size) +{ + return(access_ok_tt(type, addr, size) ? 0 : -EFAULT); +} + +extern unsigned long get_fault_addr(void); + +extern int __do_copy_from_user(void *to, const void *from, int n, + void **fault_addr, void **fault_catcher); +extern int __do_strncpy_from_user(char *dst, const char *src, size_t n, + void **fault_addr, void **fault_catcher); +extern int __do_clear_user(void *mem, size_t len, void **fault_addr, + void **fault_catcher); +extern int __do_strnlen_user(const char *str, unsigned long n, + void **fault_addr, void **fault_catcher); + +extern int copy_from_user_tt(void *to, const void *from, int n); +extern int copy_to_user_tt(void *to, const void *from, int n); +extern int strncpy_from_user_tt(char *dst, const char *src, int count); +extern int __clear_user_tt(void *mem, int len); +extern int clear_user_tt(void *mem, int len); +extern int strnlen_user_tt(const void *str, int len); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ksyms.c b/arch/um/kernel/tt/ksyms.c new file mode 100644 index 000000000000..92ec85d67c7c --- /dev/null +++ b/arch/um/kernel/tt/ksyms.c @@ -0,0 +1,28 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/module.h" +#include "asm/uaccess.h" +#include "mode.h" + +EXPORT_SYMBOL(__do_copy_from_user); +EXPORT_SYMBOL(__do_copy_to_user); +EXPORT_SYMBOL(__do_strncpy_from_user); +EXPORT_SYMBOL(__do_strnlen_user); +EXPORT_SYMBOL(__do_clear_user); + +EXPORT_SYMBOL(tracing_pid); +EXPORT_SYMBOL(honeypot); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/mem.c b/arch/um/kernel/tt/mem.c new file mode 100644 index 000000000000..74346a04a2b2 --- /dev/null +++ b/arch/um/kernel/tt/mem.c @@ -0,0 +1,51 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/stddef.h" +#include "linux/config.h" +#include "linux/mm.h" +#include "asm/uaccess.h" +#include "mem_user.h" +#include "kern_util.h" +#include "user_util.h" +#include "kern.h" +#include "tt.h" + +void before_mem_tt(unsigned long brk_start) +{ + if(debug) + remap_data(UML_ROUND_DOWN(&_stext), UML_ROUND_UP(&_etext), 1); + remap_data(UML_ROUND_DOWN(&_sdata), UML_ROUND_UP(&_edata), 1); + remap_data(UML_ROUND_DOWN(&__bss_start), UML_ROUND_UP(&_end), 1); +} + +#ifdef CONFIG_HOST_2G_2G +#define TOP 0x80000000 +#else +#define TOP 0xc0000000 +#endif + +#define SIZE ((CONFIG_NEST_LEVEL + CONFIG_KERNEL_HALF_GIGS) * 0x20000000) +#define START (TOP - SIZE) + +unsigned long set_task_sizes_tt(int arg, unsigned long *host_size_out, + unsigned long *task_size_out) +{ + /* Round up to the nearest 4M */ + *host_size_out = ROUND_4M((unsigned long) &arg); + *task_size_out = START; + return(START); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/mem_user.c b/arch/um/kernel/tt/mem_user.c new file mode 100644 index 000000000000..3085267459b1 --- /dev/null +++ b/arch/um/kernel/tt/mem_user.c @@ -0,0 +1,49 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <stdio.h> +#include <unistd.h> +#include <string.h> +#include <errno.h> +#include <sys/mman.h> +#include "tt.h" +#include "mem_user.h" +#include "user_util.h" + +void remap_data(void *segment_start, void *segment_end, int w) +{ + void *addr; + unsigned long size; + int data, prot; + + if(w) prot = PROT_WRITE; + else prot = 0; + prot |= PROT_READ | PROT_EXEC; + size = (unsigned long) segment_end - + (unsigned long) segment_start; + data = create_mem_file(size); + addr = mmap(NULL, size, PROT_WRITE | PROT_READ, MAP_SHARED, data, 0); + if(addr == MAP_FAILED){ + perror("mapping new data segment"); + exit(1); + } + memcpy(addr, segment_start, size); + if(switcheroo(data, prot, addr, segment_start, size) < 0){ + printf("switcheroo failed\n"); + exit(1); + } +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/process_kern.c b/arch/um/kernel/tt/process_kern.c new file mode 100644 index 000000000000..f19f7c18febe --- /dev/null +++ b/arch/um/kernel/tt/process_kern.c @@ -0,0 +1,476 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/signal.h" +#include "linux/kernel.h" +#include "linux/interrupt.h" +#include "linux/ptrace.h" +#include "asm/system.h" +#include "asm/pgalloc.h" +#include "asm/ptrace.h" +#include "asm/tlbflush.h" +#include "irq_user.h" +#include "signal_user.h" +#include "kern_util.h" +#include "user_util.h" +#include "os.h" +#include "kern.h" +#include "sigcontext.h" +#include "time_user.h" +#include "mem_user.h" +#include "tlb.h" +#include "mode.h" +#include "init.h" +#include "tt.h" + +void *switch_to_tt(void *prev, void *next, void *last) +{ + struct task_struct *from, *to, *prev_sched; + unsigned long flags; + int err, vtalrm, alrm, prof, cpu; + char c; + /* jailing and SMP are incompatible, so this doesn't need to be + * made per-cpu + */ + static int reading; + + from = prev; + to = next; + + to->thread.prev_sched = from; + + cpu = from->thread_info->cpu; + if(cpu == 0) + forward_interrupts(to->thread.mode.tt.extern_pid); +#ifdef CONFIG_SMP + forward_ipi(cpu_data[cpu].ipi_pipe[0], to->thread.mode.tt.extern_pid); +#endif + local_irq_save(flags); + + vtalrm = change_sig(SIGVTALRM, 0); + alrm = change_sig(SIGALRM, 0); + prof = change_sig(SIGPROF, 0); + + forward_pending_sigio(to->thread.mode.tt.extern_pid); + + c = 0; + set_current(to); + + reading = 0; + err = os_write_file(to->thread.mode.tt.switch_pipe[1], &c, sizeof(c)); + if(err != sizeof(c)) + panic("write of switch_pipe failed, err = %d", -err); + + reading = 1; + if((from->exit_state == EXIT_ZOMBIE) || + (from->exit_state == EXIT_DEAD)) + os_kill_process(os_getpid(), 0); + + err = os_read_file(from->thread.mode.tt.switch_pipe[0], &c, sizeof(c)); + if(err != sizeof(c)) + panic("read of switch_pipe failed, errno = %d", -err); + + /* If the process that we have just scheduled away from has exited, + * then it needs to be killed here. The reason is that, even though + * it will kill itself when it next runs, that may be too late. Its + * stack will be freed, possibly before then, and if that happens, + * we have a use-after-free situation. So, it gets killed here + * in case it has not already killed itself. + */ + prev_sched = current->thread.prev_sched; + if((prev_sched->exit_state == EXIT_ZOMBIE) || + (prev_sched->exit_state == EXIT_DEAD)) + os_kill_process(prev_sched->thread.mode.tt.extern_pid, 1); + + change_sig(SIGVTALRM, vtalrm); + change_sig(SIGALRM, alrm); + change_sig(SIGPROF, prof); + + arch_switch(); + + flush_tlb_all(); + local_irq_restore(flags); + + return(current->thread.prev_sched); +} + +void release_thread_tt(struct task_struct *task) +{ + int pid = task->thread.mode.tt.extern_pid; + + if(os_getpid() != pid) + os_kill_process(pid, 0); +} + +void exit_thread_tt(void) +{ + os_close_file(current->thread.mode.tt.switch_pipe[0]); + os_close_file(current->thread.mode.tt.switch_pipe[1]); +} + +void suspend_new_thread(int fd) +{ + int err; + char c; + + os_stop_process(os_getpid()); + err = os_read_file(fd, &c, sizeof(c)); + if(err != sizeof(c)) + panic("read failed in suspend_new_thread, err = %d", -err); +} + +void schedule_tail(task_t *prev); + +static void new_thread_handler(int sig) +{ + unsigned long disable; + int (*fn)(void *); + void *arg; + + fn = current->thread.request.u.thread.proc; + arg = current->thread.request.u.thread.arg; + + UPT_SC(¤t->thread.regs.regs) = (void *) (&sig + 1); + disable = (1 << (SIGVTALRM - 1)) | (1 << (SIGALRM - 1)) | + (1 << (SIGIO - 1)) | (1 << (SIGPROF - 1)); + SC_SIGMASK(UPT_SC(¤t->thread.regs.regs)) &= ~disable; + + suspend_new_thread(current->thread.mode.tt.switch_pipe[0]); + + force_flush_all(); + if(current->thread.prev_sched != NULL) + schedule_tail(current->thread.prev_sched); + current->thread.prev_sched = NULL; + + init_new_thread_signals(1); + enable_timer(); + free_page(current->thread.temp_stack); + set_cmdline("(kernel thread)"); + + change_sig(SIGUSR1, 1); + change_sig(SIGVTALRM, 1); + change_sig(SIGPROF, 1); + local_irq_enable(); + if(!run_kernel_thread(fn, arg, ¤t->thread.exec_buf)) + do_exit(0); + + /* XXX No set_user_mode here because a newly execed process will + * immediately segfault on its non-existent IP, coming straight back + * to the signal handler, which will call set_user_mode on its way + * out. This should probably change since it's confusing. + */ +} + +static int new_thread_proc(void *stack) +{ + /* local_irq_disable is needed to block out signals until this thread is + * properly scheduled. Otherwise, the tracing thread will get mighty + * upset about any signals that arrive before that. + * This has the complication that it sets the saved signal mask in + * the sigcontext to block signals. This gets restored when this + * thread (or a descendant, since they get a copy of this sigcontext) + * returns to userspace. + * So, this is compensated for elsewhere. + * XXX There is still a small window until local_irq_disable() actually + * finishes where signals are possible - shouldn't be a problem in + * practice since SIGIO hasn't been forwarded here yet, and the + * local_irq_disable should finish before a SIGVTALRM has time to be + * delivered. + */ + + local_irq_disable(); + init_new_thread_stack(stack, new_thread_handler); + os_usr1_process(os_getpid()); + change_sig(SIGUSR1, 1); + return(0); +} + +/* Signal masking - signals are blocked at the start of fork_tramp. They + * are re-enabled when finish_fork_handler is entered by fork_tramp hitting + * itself with a SIGUSR1. set_user_mode has to be run with SIGUSR1 off, + * so it is blocked before it's called. They are re-enabled on sigreturn + * despite the fact that they were blocked when the SIGUSR1 was issued because + * copy_thread copies the parent's sigcontext, including the signal mask + * onto the signal frame. + */ + +void finish_fork_handler(int sig) +{ + UPT_SC(¤t->thread.regs.regs) = (void *) (&sig + 1); + suspend_new_thread(current->thread.mode.tt.switch_pipe[0]); + + force_flush_all(); + if(current->thread.prev_sched != NULL) + schedule_tail(current->thread.prev_sched); + current->thread.prev_sched = NULL; + + enable_timer(); + change_sig(SIGVTALRM, 1); + local_irq_enable(); + if(current->mm != current->parent->mm) + protect_memory(uml_reserved, high_physmem - uml_reserved, 1, + 1, 0, 1); + task_protections((unsigned long) current_thread); + + free_page(current->thread.temp_stack); + local_irq_disable(); + change_sig(SIGUSR1, 0); + set_user_mode(current); +} + +int fork_tramp(void *stack) +{ + local_irq_disable(); + arch_init_thread(); + init_new_thread_stack(stack, finish_fork_handler); + + os_usr1_process(os_getpid()); + change_sig(SIGUSR1, 1); + return(0); +} + +int copy_thread_tt(int nr, unsigned long clone_flags, unsigned long sp, + unsigned long stack_top, struct task_struct * p, + struct pt_regs *regs) +{ + int (*tramp)(void *); + int new_pid, err; + unsigned long stack; + + if(current->thread.forking) + tramp = fork_tramp; + else { + tramp = new_thread_proc; + p->thread.request.u.thread = current->thread.request.u.thread; + } + + err = os_pipe(p->thread.mode.tt.switch_pipe, 1, 1); + if(err < 0){ + printk("copy_thread : pipe failed, err = %d\n", -err); + return(err); + } + + stack = alloc_stack(0, 0); + if(stack == 0){ + printk(KERN_ERR "copy_thread : failed to allocate " + "temporary stack\n"); + return(-ENOMEM); + } + + clone_flags &= CLONE_VM; + p->thread.temp_stack = stack; + new_pid = start_fork_tramp(p->thread_info, stack, clone_flags, tramp); + if(new_pid < 0){ + printk(KERN_ERR "copy_thread : clone failed - errno = %d\n", + -new_pid); + return(new_pid); + } + + if(current->thread.forking){ + sc_to_sc(UPT_SC(&p->thread.regs.regs), + UPT_SC(¤t->thread.regs.regs)); + SC_SET_SYSCALL_RETURN(UPT_SC(&p->thread.regs.regs), 0); + if(sp != 0) SC_SP(UPT_SC(&p->thread.regs.regs)) = sp; + } + p->thread.mode.tt.extern_pid = new_pid; + + current->thread.request.op = OP_FORK; + current->thread.request.u.fork.pid = new_pid; + os_usr1_process(os_getpid()); + + /* Enable the signal and then disable it to ensure that it is handled + * here, and nowhere else. + */ + change_sig(SIGUSR1, 1); + + change_sig(SIGUSR1, 0); + err = 0; + return(err); +} + +void reboot_tt(void) +{ + current->thread.request.op = OP_REBOOT; + os_usr1_process(os_getpid()); + change_sig(SIGUSR1, 1); +} + +void halt_tt(void) +{ + current->thread.request.op = OP_HALT; + os_usr1_process(os_getpid()); + change_sig(SIGUSR1, 1); +} + +void kill_off_processes_tt(void) +{ + struct task_struct *p; + int me; + + me = os_getpid(); + for_each_process(p){ + if(p->thread.mode.tt.extern_pid != me) + os_kill_process(p->thread.mode.tt.extern_pid, 0); + } + if(init_task.thread.mode.tt.extern_pid != me) + os_kill_process(init_task.thread.mode.tt.extern_pid, 0); +} + +void initial_thread_cb_tt(void (*proc)(void *), void *arg) +{ + if(os_getpid() == tracing_pid){ + (*proc)(arg); + } + else { + current->thread.request.op = OP_CB; + current->thread.request.u.cb.proc = proc; + current->thread.request.u.cb.arg = arg; + os_usr1_process(os_getpid()); + change_sig(SIGUSR1, 1); + + change_sig(SIGUSR1, 0); + } +} + +int do_proc_op(void *t, int proc_id) +{ + struct task_struct *task; + struct thread_struct *thread; + int op, pid; + + task = t; + thread = &task->thread; + op = thread->request.op; + switch(op){ + case OP_NONE: + case OP_TRACE_ON: + break; + case OP_EXEC: + pid = thread->request.u.exec.pid; + do_exec(thread->mode.tt.extern_pid, pid); + thread->mode.tt.extern_pid = pid; + cpu_tasks[task->thread_info->cpu].pid = pid; + break; + case OP_FORK: + attach_process(thread->request.u.fork.pid); + break; + case OP_CB: + (*thread->request.u.cb.proc)(thread->request.u.cb.arg); + break; + case OP_REBOOT: + case OP_HALT: + break; + default: + tracer_panic("Bad op in do_proc_op"); + break; + } + thread->request.op = OP_NONE; + return(op); +} + +void init_idle_tt(void) +{ + default_idle(); +} + +extern void start_kernel(void); + +static int start_kernel_proc(void *unused) +{ + int pid; + + block_signals(); + pid = os_getpid(); + + cpu_tasks[0].pid = pid; + cpu_tasks[0].task = current; +#ifdef CONFIG_SMP + cpu_online_map = cpumask_of_cpu(0); +#endif + if(debug) os_stop_process(pid); + start_kernel(); + return(0); +} + +void set_tracing(void *task, int tracing) +{ + ((struct task_struct *) task)->thread.mode.tt.tracing = tracing; +} + +int is_tracing(void *t) +{ + return (((struct task_struct *) t)->thread.mode.tt.tracing); +} + +int set_user_mode(void *t) +{ + struct task_struct *task; + + task = t ? t : current; + if(task->thread.mode.tt.tracing) + return(1); + task->thread.request.op = OP_TRACE_ON; + os_usr1_process(os_getpid()); + return(0); +} + +void set_init_pid(int pid) +{ + int err; + + init_task.thread.mode.tt.extern_pid = pid; + err = os_pipe(init_task.thread.mode.tt.switch_pipe, 1, 1); + if(err) + panic("Can't create switch pipe for init_task, errno = %d", + -err); +} + +int start_uml_tt(void) +{ + void *sp; + int pages; + + pages = (1 << CONFIG_KERNEL_STACK_ORDER); + sp = (void *) ((unsigned long) init_task.thread_info) + + pages * PAGE_SIZE - sizeof(unsigned long); + return(tracer(start_kernel_proc, sp)); +} + +int external_pid_tt(struct task_struct *task) +{ + return(task->thread.mode.tt.extern_pid); +} + +int thread_pid_tt(struct task_struct *task) +{ + return(task->thread.mode.tt.extern_pid); +} + +int is_valid_pid(int pid) +{ + struct task_struct *task; + + read_lock(&tasklist_lock); + for_each_process(task){ + if(task->thread.mode.tt.extern_pid == pid){ + read_unlock(&tasklist_lock); + return(1); + } + } + read_unlock(&tasklist_lock); + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ptproxy/Makefile b/arch/um/kernel/tt/ptproxy/Makefile new file mode 100644 index 000000000000..3ad5b774de59 --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/Makefile @@ -0,0 +1,10 @@ +# +# Copyright (C) 2002 Jeff Dike (jdike@karaya.com) +# Licensed under the GPL +# + +obj-y = proxy.o ptrace.o sysdep.o wait.o + +USER_OBJS := $(obj-y) + +include arch/um/scripts/Makefile.rules diff --git a/arch/um/kernel/tt/ptproxy/proxy.c b/arch/um/kernel/tt/ptproxy/proxy.c new file mode 100644 index 000000000000..58800c50b10e --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/proxy.c @@ -0,0 +1,377 @@ +/********************************************************************** +proxy.c + +Copyright (C) 1999 Lars Brinkhoff. See the file COPYING for licensing +terms and conditions. + +Jeff Dike (jdike@karaya.com) : Modified for integration into uml +**********************************************************************/ + +/* XXX This file shouldn't refer to CONFIG_* */ + +#include <errno.h> +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <signal.h> +#include <string.h> +#include <termios.h> +#include <sys/wait.h> +#include <sys/types.h> +#include <sys/ioctl.h> +#include <asm/unistd.h> +#include "ptrace_user.h" + +#include "ptproxy.h" +#include "sysdep.h" +#include "wait.h" + +#include "user_util.h" +#include "user.h" +#include "os.h" +#include "tempfile.h" + +static int debugger_wait(debugger_state *debugger, int *status, int options, + int (*syscall)(debugger_state *debugger, pid_t child), + int (*normal_return)(debugger_state *debugger, + pid_t unused), + int (*wait_return)(debugger_state *debugger, + pid_t unused)) +{ + if(debugger->real_wait){ + debugger->handle_trace = normal_return; + syscall_continue(debugger->pid); + debugger->real_wait = 0; + return(1); + } + debugger->wait_status_ptr = status; + debugger->wait_options = options; + if((debugger->debugee != NULL) && debugger->debugee->event){ + syscall_continue(debugger->pid); + wait_for_stop(debugger->pid, SIGTRAP, PTRACE_SYSCALL, + NULL); + (*wait_return)(debugger, -1); + return(0); + } + else if(debugger->wait_options & WNOHANG){ + syscall_cancel(debugger->pid, 0); + debugger->handle_trace = syscall; + return(0); + } + else { + syscall_pause(debugger->pid); + debugger->handle_trace = wait_return; + debugger->waiting = 1; + } + return(1); +} + +/* + * Handle debugger trap, i.e. syscall. + */ + +int debugger_syscall(debugger_state *debugger, pid_t child) +{ + long arg1, arg2, arg3, arg4, arg5, result; + int syscall, ret = 0; + + syscall = get_syscall(debugger->pid, &arg1, &arg2, &arg3, &arg4, + &arg5); + + switch(syscall){ + case __NR_execve: + /* execve never returns */ + debugger->handle_trace = debugger_syscall; + break; + + case __NR_ptrace: + if(debugger->debugee->pid != 0) arg2 = debugger->debugee->pid; + if(!debugger->debugee->in_context) + child = debugger->debugee->pid; + result = proxy_ptrace(debugger, arg1, arg2, arg3, arg4, child, + &ret); + syscall_cancel(debugger->pid, result); + debugger->handle_trace = debugger_syscall; + return(ret); + +#ifdef __NR_waitpid + case __NR_waitpid: +#endif + case __NR_wait4: + if(!debugger_wait(debugger, (int *) arg2, arg3, + debugger_syscall, debugger_normal_return, + proxy_wait_return)) + return(0); + break; + + case __NR_kill: + if(!debugger->debugee->in_context) + child = debugger->debugee->pid; + if(arg1 == debugger->debugee->pid){ + result = kill(child, arg2); + syscall_cancel(debugger->pid, result); + debugger->handle_trace = debugger_syscall; + return(0); + } + else debugger->handle_trace = debugger_normal_return; + break; + + default: + debugger->handle_trace = debugger_normal_return; + } + + syscall_continue(debugger->pid); + return(0); +} + +/* Used by the tracing thread */ +static debugger_state parent; +static int parent_syscall(debugger_state *debugger, int pid); + +int init_parent_proxy(int pid) +{ + parent = ((debugger_state) { .pid = pid, + .wait_options = 0, + .wait_status_ptr = NULL, + .waiting = 0, + .real_wait = 0, + .expecting_child = 0, + .handle_trace = parent_syscall, + .debugee = NULL } ); + return(0); +} + +int parent_normal_return(debugger_state *debugger, pid_t unused) +{ + debugger->handle_trace = parent_syscall; + syscall_continue(debugger->pid); + return(0); +} + +static int parent_syscall(debugger_state *debugger, int pid) +{ + long arg1, arg2, arg3, arg4, arg5; + int syscall; + + syscall = get_syscall(pid, &arg1, &arg2, &arg3, &arg4, &arg5); + + if((syscall == __NR_wait4) +#ifdef __NR_waitpid + || (syscall == __NR_waitpid) +#endif + ){ + debugger_wait(&parent, (int *) arg2, arg3, parent_syscall, + parent_normal_return, parent_wait_return); + } + else ptrace(PTRACE_SYSCALL, pid, 0, 0); + return(0); +} + +int debugger_normal_return(debugger_state *debugger, pid_t unused) +{ + debugger->handle_trace = debugger_syscall; + syscall_continue(debugger->pid); + return(0); +} + +void debugger_cancelled_return(debugger_state *debugger, int result) +{ + debugger->handle_trace = debugger_syscall; + syscall_set_result(debugger->pid, result); + syscall_continue(debugger->pid); +} + +/* Used by the tracing thread */ +static debugger_state debugger; +static debugee_state debugee; + +void init_proxy (pid_t debugger_pid, int stopped, int status) +{ + debugger.pid = debugger_pid; + debugger.handle_trace = debugger_syscall; + debugger.debugee = &debugee; + debugger.waiting = 0; + debugger.real_wait = 0; + debugger.expecting_child = 0; + + debugee.pid = 0; + debugee.traced = 0; + debugee.stopped = stopped; + debugee.event = 0; + debugee.zombie = 0; + debugee.died = 0; + debugee.wait_status = status; + debugee.in_context = 1; +} + +int debugger_proxy(int status, int pid) +{ + int ret = 0, sig; + + if(WIFSTOPPED(status)){ + sig = WSTOPSIG(status); + if (sig == SIGTRAP) + ret = (*debugger.handle_trace)(&debugger, pid); + + else if(sig == SIGCHLD){ + if(debugger.expecting_child){ + ptrace(PTRACE_SYSCALL, debugger.pid, 0, sig); + debugger.expecting_child = 0; + } + else if(debugger.waiting) + real_wait_return(&debugger); + else { + ptrace(PTRACE_SYSCALL, debugger.pid, 0, sig); + debugger.real_wait = 1; + } + } + else ptrace(PTRACE_SYSCALL, debugger.pid, 0, sig); + } + else if(WIFEXITED(status)){ + tracer_panic("debugger (pid %d) exited with status %d", + debugger.pid, WEXITSTATUS(status)); + } + else if(WIFSIGNALED(status)){ + tracer_panic("debugger (pid %d) exited with signal %d", + debugger.pid, WTERMSIG(status)); + } + else { + tracer_panic("proxy got unknown status (0x%x) on debugger " + "(pid %d)", status, debugger.pid); + } + return(ret); +} + +void child_proxy(pid_t pid, int status) +{ + debugee.event = 1; + debugee.wait_status = status; + + if(WIFSTOPPED(status)){ + debugee.stopped = 1; + debugger.expecting_child = 1; + kill(debugger.pid, SIGCHLD); + } + else if(WIFEXITED(status) || WIFSIGNALED(status)){ + debugee.zombie = 1; + debugger.expecting_child = 1; + kill(debugger.pid, SIGCHLD); + } + else panic("proxy got unknown status (0x%x) on child (pid %d)", + status, pid); +} + +void debugger_parent_signal(int status, int pid) +{ + int sig; + + if(WIFSTOPPED(status)){ + sig = WSTOPSIG(status); + if(sig == SIGTRAP) (*parent.handle_trace)(&parent, pid); + else ptrace(PTRACE_SYSCALL, pid, 0, sig); + } +} + +void fake_child_exit(void) +{ + int status, pid; + + child_proxy(1, W_EXITCODE(0, 0)); + while(debugger.waiting == 1){ + CATCH_EINTR(pid = waitpid(debugger.pid, &status, WUNTRACED)); + if(pid != debugger.pid){ + printk("fake_child_exit - waitpid failed, " + "errno = %d\n", errno); + return; + } + debugger_proxy(status, debugger.pid); + } + CATCH_EINTR(pid = waitpid(debugger.pid, &status, WUNTRACED)); + if(pid != debugger.pid){ + printk("fake_child_exit - waitpid failed, " + "errno = %d\n", errno); + return; + } + if(ptrace(PTRACE_DETACH, debugger.pid, 0, SIGCONT) < 0) + printk("fake_child_exit - PTRACE_DETACH failed, errno = %d\n", + errno); +} + +char gdb_init_string[] = +"att 1 \n\ +b panic \n\ +b stop \n\ +handle SIGWINCH nostop noprint pass \n\ +"; + +int start_debugger(char *prog, int startup, int stop, int *fd_out) +{ + int slave, child; + + slave = open_gdb_chan(); + child = fork(); + if(child == 0){ + char *tempname = NULL; + int fd; + + if(setsid() < 0) perror("setsid"); + if((dup2(slave, 0) < 0) || (dup2(slave, 1) < 0) || + (dup2(slave, 2) < 0)){ + printk("start_debugger : dup2 failed, errno = %d\n", + errno); + exit(1); + } + if(ioctl(0, TIOCSCTTY, 0) < 0){ + printk("start_debugger : TIOCSCTTY failed, " + "errno = %d\n", errno); + exit(1); + } + if(tcsetpgrp (1, os_getpid()) < 0){ + printk("start_debugger : tcsetpgrp failed, " + "errno = %d\n", errno); +#ifdef notdef + exit(1); +#endif + } + fd = make_tempfile("/tmp/gdb_init-XXXXXX", &tempname, 0); + if(fd < 0){ + printk("start_debugger : make_tempfile failed," + "err = %d\n", -fd); + exit(1); + } + os_write_file(fd, gdb_init_string, sizeof(gdb_init_string) - 1); + if(startup){ + if(stop){ + os_write_file(fd, "b start_kernel\n", + strlen("b start_kernel\n")); + } + os_write_file(fd, "c\n", strlen("c\n")); + } + if(ptrace(PTRACE_TRACEME, 0, 0, 0) < 0){ + printk("start_debugger : PTRACE_TRACEME failed, " + "errno = %d\n", errno); + exit(1); + } + execlp("gdb", "gdb", "--command", tempname, prog, NULL); + printk("start_debugger : exec of gdb failed, errno = %d\n", + errno); + } + if(child < 0){ + printk("start_debugger : fork for gdb failed, errno = %d\n", + errno); + return(-1); + } + *fd_out = slave; + return(child); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ptproxy/ptproxy.h b/arch/um/kernel/tt/ptproxy/ptproxy.h new file mode 100644 index 000000000000..5eb0285b1968 --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/ptproxy.h @@ -0,0 +1,61 @@ +/********************************************************************** +ptproxy.h + +Copyright (C) 1999 Lars Brinkhoff. See the file COPYING for licensing +terms and conditions. +**********************************************************************/ + +#ifndef __PTPROXY_H +#define __PTPROXY_H + +#include <sys/types.h> + +typedef struct debugger debugger_state; +typedef struct debugee debugee_state; + +struct debugger +{ + pid_t pid; + int wait_options; + int *wait_status_ptr; + unsigned int waiting : 1; + unsigned int real_wait : 1; + unsigned int expecting_child : 1; + int (*handle_trace) (debugger_state *, pid_t); + + debugee_state *debugee; +}; + +struct debugee +{ + pid_t pid; + int wait_status; + unsigned int died : 1; + unsigned int event : 1; + unsigned int stopped : 1; + unsigned int trace_singlestep : 1; + unsigned int trace_syscall : 1; + unsigned int traced : 1; + unsigned int zombie : 1; + unsigned int in_context : 1; +}; + +extern int debugger_syscall(debugger_state *debugger, pid_t pid); +extern int debugger_normal_return (debugger_state *debugger, pid_t unused); + +extern long proxy_ptrace (struct debugger *, int, pid_t, long, long, pid_t, + int *strace_out); +extern void debugger_cancelled_return(debugger_state *debugger, int result); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ptproxy/ptrace.c b/arch/um/kernel/tt/ptproxy/ptrace.c new file mode 100644 index 000000000000..528a5fc8d887 --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/ptrace.c @@ -0,0 +1,237 @@ +/********************************************************************** +ptrace.c + +Copyright (C) 1999 Lars Brinkhoff. See the file COPYING for licensing +terms and conditions. + +Jeff Dike (jdike@karaya.com) : Modified for integration into uml +**********************************************************************/ + +#include <errno.h> +#include <unistd.h> +#include <signal.h> +#include <sys/types.h> +#include <sys/time.h> +#include <sys/wait.h> + +#include "ptproxy.h" +#include "debug.h" +#include "user_util.h" +#include "kern_util.h" +#include "ptrace_user.h" +#include "tt.h" + +long proxy_ptrace(struct debugger *debugger, int arg1, pid_t arg2, + long arg3, long arg4, pid_t child, int *ret) +{ + sigset_t relay; + long result; + int status; + + *ret = 0; + if(debugger->debugee->died) return(-ESRCH); + + switch(arg1){ + case PTRACE_ATTACH: + if(debugger->debugee->traced) return(-EPERM); + + debugger->debugee->pid = arg2; + debugger->debugee->traced = 1; + + if(is_valid_pid(arg2) && (arg2 != child)){ + debugger->debugee->in_context = 0; + kill(arg2, SIGSTOP); + debugger->debugee->event = 1; + debugger->debugee->wait_status = W_STOPCODE(SIGSTOP); + } + else { + debugger->debugee->in_context = 1; + if(debugger->debugee->stopped) + child_proxy(child, W_STOPCODE(SIGSTOP)); + else kill(child, SIGSTOP); + } + + return(0); + + case PTRACE_DETACH: + if(!debugger->debugee->traced) return(-EPERM); + + debugger->debugee->traced = 0; + debugger->debugee->pid = 0; + if(!debugger->debugee->in_context) + kill(child, SIGCONT); + + return(0); + + case PTRACE_CONT: + if(!debugger->debugee->in_context) return(-EPERM); + *ret = PTRACE_CONT; + return(ptrace(PTRACE_CONT, child, arg3, arg4)); + +#ifdef UM_HAVE_GETFPREGS + case PTRACE_GETFPREGS: + { + long regs[FP_FRAME_SIZE]; + int i, result; + + result = ptrace(PTRACE_GETFPREGS, child, 0, regs); + if(result == -1) return(-errno); + + for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++) + ptrace(PTRACE_POKEDATA, debugger->pid, arg4 + 4 * i, + regs[i]); + return(result); + } +#endif + +#ifdef UM_HAVE_GETFPXREGS + case PTRACE_GETFPXREGS: + { + long regs[FPX_FRAME_SIZE]; + int i, result; + + result = ptrace(PTRACE_GETFPXREGS, child, 0, regs); + if(result == -1) return(-errno); + + for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++) + ptrace(PTRACE_POKEDATA, debugger->pid, arg4 + 4 * i, + regs[i]); + return(result); + } +#endif + +#ifdef UM_HAVE_GETREGS + case PTRACE_GETREGS: + { + long regs[FRAME_SIZE]; + int i, result; + + result = ptrace(PTRACE_GETREGS, child, 0, regs); + if(result == -1) return(-errno); + + for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++) + ptrace (PTRACE_POKEDATA, debugger->pid, + arg4 + 4 * i, regs[i]); + return(result); + } + break; +#endif + + case PTRACE_KILL: + result = ptrace(PTRACE_KILL, child, arg3, arg4); + if(result == -1) return(-errno); + + return(result); + + case PTRACE_PEEKDATA: + case PTRACE_PEEKTEXT: + case PTRACE_PEEKUSR: + /* The value being read out could be -1, so we have to + * check errno to see if there's an error, and zero it + * beforehand so we're not faked out by an old error + */ + + errno = 0; + result = ptrace(arg1, child, arg3, 0); + if((result == -1) && (errno != 0)) return(-errno); + + result = ptrace(PTRACE_POKEDATA, debugger->pid, arg4, result); + if(result == -1) return(-errno); + + return(result); + + case PTRACE_POKEDATA: + case PTRACE_POKETEXT: + case PTRACE_POKEUSR: + result = ptrace(arg1, child, arg3, arg4); + if(result == -1) return(-errno); + + if(arg1 == PTRACE_POKEUSR) ptrace_pokeuser(arg3, arg4); + return(result); + +#ifdef UM_HAVE_SETFPREGS + case PTRACE_SETFPREGS: + { + long regs[FP_FRAME_SIZE]; + int i; + + for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++) + regs[i] = ptrace (PTRACE_PEEKDATA, debugger->pid, + arg4 + 4 * i, 0); + result = ptrace(PTRACE_SETFPREGS, child, 0, regs); + if(result == -1) return(-errno); + + return(result); + } +#endif + +#ifdef UM_HAVE_SETFPXREGS + case PTRACE_SETFPXREGS: + { + long regs[FPX_FRAME_SIZE]; + int i; + + for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++) + regs[i] = ptrace (PTRACE_PEEKDATA, debugger->pid, + arg4 + 4 * i, 0); + result = ptrace(PTRACE_SETFPXREGS, child, 0, regs); + if(result == -1) return(-errno); + + return(result); + } +#endif + +#ifdef UM_HAVE_SETREGS + case PTRACE_SETREGS: + { + long regs[FRAME_SIZE]; + int i; + + for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++) + regs[i] = ptrace(PTRACE_PEEKDATA, debugger->pid, + arg4 + 4 * i, 0); + result = ptrace(PTRACE_SETREGS, child, 0, regs); + if(result == -1) return(-errno); + + return(result); + } +#endif + + case PTRACE_SINGLESTEP: + if(!debugger->debugee->in_context) return(-EPERM); + sigemptyset(&relay); + sigaddset(&relay, SIGSEGV); + sigaddset(&relay, SIGILL); + sigaddset(&relay, SIGBUS); + result = ptrace(PTRACE_SINGLESTEP, child, arg3, arg4); + if(result == -1) return(-errno); + + status = wait_for_stop(child, SIGTRAP, PTRACE_SINGLESTEP, + &relay); + child_proxy(child, status); + return(result); + + case PTRACE_SYSCALL: + if(!debugger->debugee->in_context) return(-EPERM); + result = ptrace(PTRACE_SYSCALL, child, arg3, arg4); + if(result == -1) return(-errno); + + *ret = PTRACE_SYSCALL; + return(result); + + case PTRACE_TRACEME: + default: + return(-EINVAL); + } +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ptproxy/sysdep.c b/arch/um/kernel/tt/ptproxy/sysdep.c new file mode 100644 index 000000000000..a5f0e01e214e --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/sysdep.c @@ -0,0 +1,70 @@ +/********************************************************************** +sysdep.c + +Copyright (C) 1999 Lars Brinkhoff. See the file COPYING for licensing +terms and conditions. +**********************************************************************/ + +#include <stdio.h> +#include <string.h> +#include <stdlib.h> +#include <signal.h> +#include <errno.h> +#include <sys/types.h> +#include <linux/unistd.h> +#include "ptrace_user.h" +#include "user_util.h" +#include "user.h" + +int get_syscall(pid_t pid, long *arg1, long *arg2, long *arg3, long *arg4, + long *arg5) +{ + *arg1 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG1_OFFSET, 0); + *arg2 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG2_OFFSET, 0); + *arg3 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG3_OFFSET, 0); + *arg4 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG4_OFFSET, 0); + *arg5 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG5_OFFSET, 0); + return(ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_NR_OFFSET, 0)); +} + +void syscall_cancel(pid_t pid, int result) +{ + if((ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET, + __NR_getpid) < 0) || + (ptrace(PTRACE_SYSCALL, pid, 0, 0) < 0) || + (wait_for_stop(pid, SIGTRAP, PTRACE_SYSCALL, NULL) < 0) || + (ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_RET_OFFSET, result) < 0) || + (ptrace(PTRACE_SYSCALL, pid, 0, 0) < 0)) + printk("ptproxy: couldn't cancel syscall: errno = %d\n", + errno); +} + +void syscall_set_result(pid_t pid, long result) +{ + ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_RET_OFFSET, result); +} + +void syscall_continue(pid_t pid) +{ + ptrace(PTRACE_SYSCALL, pid, 0, 0); +} + +int syscall_pause(pid_t pid) +{ + if(ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET, __NR_pause) < 0){ + printk("syscall_change - ptrace failed, errno = %d\n", errno); + return(-1); + } + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ptproxy/sysdep.h b/arch/um/kernel/tt/ptproxy/sysdep.h new file mode 100644 index 000000000000..735f488049aa --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/sysdep.h @@ -0,0 +1,25 @@ +/********************************************************************** +sysdep.h + +Copyright (C) 1999 Lars Brinkhoff. +Copyright (C) 2001 Jeff Dike (jdike@karaya.com) +See the file COPYING for licensing terms and conditions. +**********************************************************************/ + +extern int get_syscall(pid_t pid, long *arg1, long *arg2, long *arg3, + long *arg4, long *arg5); +extern void syscall_cancel (pid_t pid, long result); +extern void syscall_set_result (pid_t pid, long result); +extern void syscall_continue (pid_t pid); +extern int syscall_pause(pid_t pid); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ptproxy/wait.c b/arch/um/kernel/tt/ptproxy/wait.c new file mode 100644 index 000000000000..12f6319d8d76 --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/wait.c @@ -0,0 +1,86 @@ +/********************************************************************** +wait.c + +Copyright (C) 1999 Lars Brinkhoff. See the file COPYING for licensing +terms and conditions. + +**********************************************************************/ + +#include <errno.h> +#include <signal.h> +#include <sys/wait.h> + +#include "ptproxy.h" +#include "sysdep.h" +#include "wait.h" +#include "user_util.h" +#include "ptrace_user.h" +#include "sysdep/ptrace.h" +#include "sysdep/sigcontext.h" + +int proxy_wait_return(struct debugger *debugger, pid_t unused) +{ + debugger->waiting = 0; + + if(debugger->debugee->died || (debugger->wait_options & __WCLONE)){ + debugger_cancelled_return(debugger, -ECHILD); + return(0); + } + + if(debugger->debugee->zombie && debugger->debugee->event) + debugger->debugee->died = 1; + + if(debugger->debugee->event){ + debugger->debugee->event = 0; + ptrace(PTRACE_POKEDATA, debugger->pid, + debugger->wait_status_ptr, + debugger->debugee->wait_status); + /* if (wait4) + ptrace (PTRACE_POKEDATA, pid, rusage_ptr, ...); */ + debugger_cancelled_return(debugger, debugger->debugee->pid); + return(0); + } + + /* pause will return -EINTR, which happens to be right for wait */ + debugger_normal_return(debugger, -1); + return(0); +} + +int parent_wait_return(struct debugger *debugger, pid_t unused) +{ + return(debugger_normal_return(debugger, -1)); +} + +int real_wait_return(struct debugger *debugger) +{ + unsigned long ip; + int pid; + + pid = debugger->pid; + + ip = ptrace(PTRACE_PEEKUSR, pid, PT_IP_OFFSET, 0); + IP_RESTART_SYSCALL(ip); + + if(ptrace(PTRACE_POKEUSR, pid, PT_IP_OFFSET, ip) < 0) + tracer_panic("real_wait_return : Failed to restart system " + "call, errno = %d\n", errno); + + if((ptrace(PTRACE_SYSCALL, debugger->pid, 0, SIGCHLD) < 0) || + (ptrace(PTRACE_SYSCALL, debugger->pid, 0, 0) < 0) || + (ptrace(PTRACE_SYSCALL, debugger->pid, 0, 0) < 0) || + debugger_normal_return(debugger, -1)) + tracer_panic("real_wait_return : gdb failed to wait, " + "errno = %d\n", errno); + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ptproxy/wait.h b/arch/um/kernel/tt/ptproxy/wait.h new file mode 100644 index 000000000000..542e73ee2cee --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/wait.h @@ -0,0 +1,15 @@ +/********************************************************************** +wait.h + +Copyright (C) 1999 Lars Brinkhoff. See the file COPYING for licensing +terms and conditions. +**********************************************************************/ + +#ifndef __PTPROXY_WAIT_H +#define __PTPROXY_WAIT_H + +extern int proxy_wait_return(struct debugger *debugger, pid_t unused); +extern int real_wait_return(struct debugger *debugger); +extern int parent_wait_return(struct debugger *debugger, pid_t unused); + +#endif diff --git a/arch/um/kernel/tt/syscall_kern.c b/arch/um/kernel/tt/syscall_kern.c new file mode 100644 index 000000000000..2650a628719e --- /dev/null +++ b/arch/um/kernel/tt/syscall_kern.c @@ -0,0 +1,47 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/types.h" +#include "linux/utime.h" +#include "linux/sys.h" +#include "linux/ptrace.h" +#include "asm/unistd.h" +#include "asm/ptrace.h" +#include "asm/uaccess.h" +#include "asm/stat.h" +#include "sysdep/syscalls.h" +#include "kern_util.h" + +extern syscall_handler_t *sys_call_table[]; + +long execute_syscall_tt(void *r) +{ + struct pt_regs *regs = r; + long res; + int syscall; + +#ifdef CONFIG_SYSCALL_DEBUG + current->thread.nsyscalls++; + nsyscalls++; +#endif + syscall = UPT_SYSCALL_NR(®s->regs); + + if((syscall >= NR_syscalls) || (syscall < 0)) + res = -ENOSYS; + else res = EXECUTE_SYSCALL(syscall, regs); + + return(res); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/syscall_user.c b/arch/um/kernel/tt/syscall_user.c new file mode 100644 index 000000000000..e4e7e9c2224c --- /dev/null +++ b/arch/um/kernel/tt/syscall_user.c @@ -0,0 +1,90 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <unistd.h> +#include <signal.h> +#include <errno.h> +#include <asm/unistd.h> +#include "sysdep/ptrace.h" +#include "sigcontext.h" +#include "ptrace_user.h" +#include "task.h" +#include "user_util.h" +#include "kern_util.h" +#include "syscall_user.h" +#include "tt.h" + + +void syscall_handler_tt(int sig, union uml_pt_regs *regs) +{ + void *sc; + long result; + int syscall; +#ifdef UML_CONFIG_DEBUG_SYSCALL + int index; +#endif + + syscall = UPT_SYSCALL_NR(regs); + sc = UPT_SC(regs); + SC_START_SYSCALL(sc); + +#ifdef UML_CONFIG_DEBUG_SYSCALL + index = record_syscall_start(syscall); +#endif + syscall_trace(regs, 0); + result = execute_syscall_tt(regs); + + /* regs->sc may have changed while the system call ran (there may + * have been an interrupt or segfault), so it needs to be refreshed. + */ + UPT_SC(regs) = sc; + + SC_SET_SYSCALL_RETURN(sc, result); + + syscall_trace(regs, 1); +#ifdef UML_CONFIG_DEBUG_SYSCALL + record_syscall_end(index, result); +#endif +} + +void do_sigtrap(void *task) +{ + UPT_SYSCALL_NR(TASK_REGS(task)) = -1; +} + +void do_syscall(void *task, int pid, int local_using_sysemu) +{ + unsigned long proc_regs[FRAME_SIZE]; + + if(ptrace_getregs(pid, proc_regs) < 0) + tracer_panic("Couldn't read registers"); + + UPT_SYSCALL_NR(TASK_REGS(task)) = PT_SYSCALL_NR(proc_regs); + + if(((unsigned long *) PT_IP(proc_regs) >= &_stext) && + ((unsigned long *) PT_IP(proc_regs) <= &_etext)) + tracer_panic("I'm tracing myself and I can't get out"); + + /* advanced sysemu mode set syscall number to -1 automatically */ + if (local_using_sysemu==2) + return; + + /* syscall number -1 in sysemu skips syscall restarting in host */ + if(ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET, + local_using_sysemu ? -1 : __NR_getpid) < 0) + tracer_panic("do_syscall : Nullifying syscall failed, " + "errno = %d", errno); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/time.c b/arch/um/kernel/tt/time.c new file mode 100644 index 000000000000..8565b71b07cd --- /dev/null +++ b/arch/um/kernel/tt/time.c @@ -0,0 +1,28 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <signal.h> +#include <sys/time.h> +#include <time_user.h> +#include "process.h" +#include "user.h" + +void user_time_init_tt(void) +{ + if(signal(SIGVTALRM, (__sighandler_t) alarm_handler) == SIG_ERR) + panic("Couldn't set SIGVTALRM handler"); + set_interval(ITIMER_VIRTUAL); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/tlb.c b/arch/um/kernel/tt/tlb.c new file mode 100644 index 000000000000..203216ad86f1 --- /dev/null +++ b/arch/um/kernel/tt/tlb.c @@ -0,0 +1,149 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Copyright 2003 PathScale, Inc. + * Licensed under the GPL + */ + +#include "linux/stddef.h" +#include "linux/kernel.h" +#include "linux/sched.h" +#include "linux/mm.h" +#include "asm/page.h" +#include "asm/pgtable.h" +#include "asm/uaccess.h" +#include "asm/tlbflush.h" +#include "user_util.h" +#include "mem_user.h" +#include "os.h" +#include "tlb.h" + +static void do_ops(int unused, struct host_vm_op *ops, int last) +{ + struct host_vm_op *op; + int i; + + for(i = 0; i <= last; i++){ + op = &ops[i]; + switch(op->type){ + case MMAP: + os_map_memory((void *) op->u.mmap.addr, op->u.mmap.fd, + op->u.mmap.offset, op->u.mmap.len, + op->u.mmap.r, op->u.mmap.w, + op->u.mmap.x); + break; + case MUNMAP: + os_unmap_memory((void *) op->u.munmap.addr, + op->u.munmap.len); + break; + case MPROTECT: + protect_memory(op->u.mprotect.addr, op->u.munmap.len, + op->u.mprotect.r, op->u.mprotect.w, + op->u.mprotect.x, 1); + break; + default: + printk("Unknown op type %d in do_ops\n", op->type); + break; + } + } +} + +static void fix_range(struct mm_struct *mm, unsigned long start_addr, + unsigned long end_addr, int force) +{ + if((current->thread.mode.tt.extern_pid != -1) && + (current->thread.mode.tt.extern_pid != os_getpid())) + panic("fix_range fixing wrong address space, current = 0x%p", + current); + + fix_range_common(mm, start_addr, end_addr, force, 0, do_ops); +} + +atomic_t vmchange_seq = ATOMIC_INIT(1); + +void flush_tlb_kernel_range_tt(unsigned long start, unsigned long end) +{ + if(flush_tlb_kernel_range_common(start, end)) + atomic_inc(&vmchange_seq); +} + +static void protect_vm_page(unsigned long addr, int w, int must_succeed) +{ + int err; + + err = protect_memory(addr, PAGE_SIZE, 1, w, 1, must_succeed); + if(err == 0) return; + else if((err == -EFAULT) || (err == -ENOMEM)){ + flush_tlb_kernel_range(addr, addr + PAGE_SIZE); + protect_vm_page(addr, w, 1); + } + else panic("protect_vm_page : protect failed, errno = %d\n", err); +} + +void mprotect_kernel_vm(int w) +{ + struct mm_struct *mm; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + unsigned long addr; + + mm = &init_mm; + for(addr = start_vm; addr < end_vm;){ + pgd = pgd_offset(mm, addr); + pud = pud_offset(pgd, addr); + pmd = pmd_offset(pud, addr); + if(pmd_present(*pmd)){ + pte = pte_offset_kernel(pmd, addr); + if(pte_present(*pte)) protect_vm_page(addr, w, 0); + addr += PAGE_SIZE; + } + else addr += PMD_SIZE; + } +} + +void flush_tlb_kernel_vm_tt(void) +{ + flush_tlb_kernel_range(start_vm, end_vm); +} + +void __flush_tlb_one_tt(unsigned long addr) +{ + flush_tlb_kernel_range(addr, addr + PAGE_SIZE); +} + +void flush_tlb_range_tt(struct vm_area_struct *vma, unsigned long start, + unsigned long end) +{ + if(vma->vm_mm != current->mm) return; + + /* Assumes that the range start ... end is entirely within + * either process memory or kernel vm + */ + if((start >= start_vm) && (start < end_vm)){ + if(flush_tlb_kernel_range_common(start, end)) + atomic_inc(&vmchange_seq); + } + else fix_range(vma->vm_mm, start, end, 0); +} + +void flush_tlb_mm_tt(struct mm_struct *mm) +{ + unsigned long seq; + + if(mm != current->mm) return; + + fix_range(mm, 0, STACK_TOP, 0); + + seq = atomic_read(&vmchange_seq); + if(current->thread.mode.tt.vm_seq == seq) + return; + current->thread.mode.tt.vm_seq = seq; + flush_tlb_kernel_range_common(start_vm, end_vm); +} + +void force_flush_all_tt(void) +{ + fix_range(current->mm, 0, STACK_TOP, 1); + flush_tlb_kernel_range_common(start_vm, end_vm); +} diff --git a/arch/um/kernel/tt/tracer.c b/arch/um/kernel/tt/tracer.c new file mode 100644 index 000000000000..7b5d937e5955 --- /dev/null +++ b/arch/um/kernel/tt/tracer.c @@ -0,0 +1,480 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <stdarg.h> +#include <unistd.h> +#include <signal.h> +#include <errno.h> +#include <sched.h> +#include <string.h> +#include <sys/mman.h> +#include <sys/time.h> +#include <sys/wait.h> +#include "user.h" +#include "sysdep/ptrace.h" +#include "sigcontext.h" +#include "sysdep/sigcontext.h" +#include "os.h" +#include "signal_user.h" +#include "user_util.h" +#include "mem_user.h" +#include "process.h" +#include "kern_util.h" +#include "chan_user.h" +#include "ptrace_user.h" +#include "mode.h" +#include "tt.h" + +static int tracer_winch[2]; + +int is_tracer_winch(int pid, int fd, void *data) +{ + if(pid != tracing_pid) + return(0); + + register_winch_irq(tracer_winch[0], fd, -1, data); + return(1); +} + +static void tracer_winch_handler(int sig) +{ + int n; + char c = 1; + + n = os_write_file(tracer_winch[1], &c, sizeof(c)); + if(n != sizeof(c)) + printk("tracer_winch_handler - write failed, err = %d\n", -n); +} + +/* Called only by the tracing thread during initialization */ + +static void setup_tracer_winch(void) +{ + int err; + + err = os_pipe(tracer_winch, 1, 1); + if(err < 0){ + printk("setup_tracer_winch : os_pipe failed, err = %d\n", -err); + return; + } + signal(SIGWINCH, tracer_winch_handler); +} + +void attach_process(int pid) +{ + if((ptrace(PTRACE_ATTACH, pid, 0, 0) < 0) || + (ptrace(PTRACE_CONT, pid, 0, 0) < 0)) + tracer_panic("OP_FORK failed to attach pid"); + wait_for_stop(pid, SIGSTOP, PTRACE_CONT, NULL); + if (ptrace(PTRACE_OLDSETOPTIONS, pid, 0, (void *)PTRACE_O_TRACESYSGOOD) < 0) + tracer_panic("OP_FORK: PTRACE_SETOPTIONS failed, errno = %d", errno); + if(ptrace(PTRACE_CONT, pid, 0, 0) < 0) + tracer_panic("OP_FORK failed to continue process"); +} + +void tracer_panic(char *format, ...) +{ + va_list ap; + + va_start(ap, format); + vprintf(format, ap); + va_end(ap); + printf("\n"); + while(1) pause(); +} + +static void tracer_segv(int sig, struct sigcontext sc) +{ + printf("Tracing thread segfault at address 0x%lx, ip 0x%lx\n", + SC_FAULT_ADDR(&sc), SC_IP(&sc)); + while(1) + pause(); +} + +/* Changed early in boot, and then only read */ +int debug = 0; +int debug_stop = 1; +int debug_parent = 0; +int honeypot = 0; + +static int signal_tramp(void *arg) +{ + int (*proc)(void *); + + if(honeypot && munmap((void *) (host_task_size - 0x10000000), + 0x10000000)) + panic("Unmapping stack failed"); + if(ptrace(PTRACE_TRACEME, 0, 0, 0) < 0) + panic("ptrace PTRACE_TRACEME failed"); + os_stop_process(os_getpid()); + change_sig(SIGWINCH, 0); + signal(SIGUSR1, SIG_IGN); + change_sig(SIGCHLD, 0); + signal(SIGSEGV, (__sighandler_t) sig_handler); + set_cmdline("(idle thread)"); + set_init_pid(os_getpid()); + proc = arg; + return((*proc)(NULL)); +} + +static void sleeping_process_signal(int pid, int sig) +{ + switch(sig){ + /* These two result from UML being ^Z-ed and bg-ed. PTRACE_CONT is + * right because the process must be in the kernel already. + */ + case SIGCONT: + case SIGTSTP: + if(ptrace(PTRACE_CONT, pid, 0, sig) < 0) + tracer_panic("sleeping_process_signal : Failed to " + "continue pid %d, signal = %d, " + "errno = %d\n", pid, sig, errno); + break; + + /* This happens when the debugger (e.g. strace) is doing system call + * tracing on the kernel. During a context switch, the current task + * will be set to the incoming process and the outgoing process will + * hop into write and then read. Since it's not the current process + * any more, the trace of those will land here. So, we need to just + * PTRACE_SYSCALL it. + */ + case (SIGTRAP + 0x80): + if(ptrace(PTRACE_SYSCALL, pid, 0, 0) < 0) + tracer_panic("sleeping_process_signal : Failed to " + "PTRACE_SYSCALL pid %d, errno = %d\n", + pid, errno); + break; + case SIGSTOP: + break; + default: + tracer_panic("sleeping process %d got unexpected " + "signal : %d\n", pid, sig); + break; + } +} + +/* Accessed only by the tracing thread */ +int debugger_pid = -1; +int debugger_parent = -1; +int debugger_fd = -1; +int gdb_pid = -1; + +struct { + int pid; + int signal; + unsigned long addr; + struct timeval time; +} signal_record[1024][32]; + +int signal_index[32]; +int nsignals = 0; +int debug_trace = 0; +extern int io_nsignals, io_count, intr_count; + +extern void signal_usr1(int sig); + +int tracing_pid = -1; + +int tracer(int (*init_proc)(void *), void *sp) +{ + void *task = NULL; + int status, pid = 0, sig = 0, cont_type, tracing = 0, op = 0; + int proc_id = 0, n, err, old_tracing = 0, strace = 0; + int local_using_sysemu = 0; +#ifdef UML_CONFIG_SYSCALL_DEBUG + unsigned long eip = 0; + int last_index; +#endif + signal(SIGPIPE, SIG_IGN); + setup_tracer_winch(); + tracing_pid = os_getpid(); + printf("tracing thread pid = %d\n", tracing_pid); + + pid = clone(signal_tramp, sp, CLONE_FILES | SIGCHLD, init_proc); + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + if(n < 0){ + printf("waitpid on idle thread failed, errno = %d\n", errno); + exit(1); + } + if (ptrace(PTRACE_OLDSETOPTIONS, pid, 0, (void *)PTRACE_O_TRACESYSGOOD) < 0) { + printf("Failed to PTRACE_SETOPTIONS for idle thread, errno = %d\n", errno); + exit(1); + } + if((ptrace(PTRACE_CONT, pid, 0, 0) < 0)){ + printf("Failed to continue idle thread, errno = %d\n", errno); + exit(1); + } + + signal(SIGSEGV, (sighandler_t) tracer_segv); + signal(SIGUSR1, signal_usr1); + if(debug_trace){ + printf("Tracing thread pausing to be attached\n"); + stop(); + } + if(debug){ + if(gdb_pid != -1) + debugger_pid = attach_debugger(pid, gdb_pid, 1); + else debugger_pid = init_ptrace_proxy(pid, 1, debug_stop); + if(debug_parent){ + debugger_parent = os_process_parent(debugger_pid); + init_parent_proxy(debugger_parent); + err = attach(debugger_parent); + if(err){ + printf("Failed to attach debugger parent %d, " + "errno = %d\n", debugger_parent, -err); + debugger_parent = -1; + } + else { + if(ptrace(PTRACE_SYSCALL, debugger_parent, + 0, 0) < 0){ + printf("Failed to continue debugger " + "parent, errno = %d\n", errno); + debugger_parent = -1; + } + } + } + } + set_cmdline("(tracing thread)"); + while(1){ + CATCH_EINTR(pid = waitpid(-1, &status, WUNTRACED)); + if(pid <= 0){ + if(errno != ECHILD){ + printf("wait failed - errno = %d\n", errno); + } + continue; + } + if(pid == debugger_pid){ + int cont = 0; + + if(WIFEXITED(status) || WIFSIGNALED(status)) + debugger_pid = -1; + /* XXX Figure out how to deal with gdb and SMP */ + else cont = debugger_signal(status, cpu_tasks[0].pid); + if(cont == PTRACE_SYSCALL) strace = 1; + continue; + } + else if(pid == debugger_parent){ + debugger_parent_signal(status, pid); + continue; + } + nsignals++; + if(WIFEXITED(status)) ; +#ifdef notdef + { + printf("Child %d exited with status %d\n", pid, + WEXITSTATUS(status)); + } +#endif + else if(WIFSIGNALED(status)){ + sig = WTERMSIG(status); + if(sig != 9){ + printf("Child %d exited with signal %d\n", pid, + sig); + } + } + else if(WIFSTOPPED(status)){ + proc_id = pid_to_processor_id(pid); + sig = WSTOPSIG(status); +#ifdef UML_CONFIG_SYSCALL_DEBUG + if(signal_index[proc_id] == 1024){ + signal_index[proc_id] = 0; + last_index = 1023; + } + else last_index = signal_index[proc_id] - 1; + if(((sig == SIGPROF) || (sig == SIGVTALRM) || + (sig == SIGALRM)) && + (signal_record[proc_id][last_index].signal == sig)&& + (signal_record[proc_id][last_index].pid == pid)) + signal_index[proc_id] = last_index; + signal_record[proc_id][signal_index[proc_id]].pid = pid; + gettimeofday(&signal_record[proc_id][signal_index[proc_id]].time, NULL); + eip = ptrace(PTRACE_PEEKUSR, pid, PT_IP_OFFSET, 0); + signal_record[proc_id][signal_index[proc_id]].addr = eip; + signal_record[proc_id][signal_index[proc_id]++].signal = sig; +#endif + if(proc_id == -1){ + sleeping_process_signal(pid, sig); + continue; + } + + task = cpu_tasks[proc_id].task; + tracing = is_tracing(task); + old_tracing = tracing; + + /* Assume: no syscall, when coming from user */ + if ( tracing ) + do_sigtrap(task); + + switch(sig){ + case SIGUSR1: + sig = 0; + op = do_proc_op(task, proc_id); + switch(op){ + /* + * This is called when entering user mode; after + * this, we start intercepting syscalls. + * + * In fact, a process is started in kernel mode, + * so with is_tracing() == 0 (and that is reset + * when executing syscalls, since UML kernel has + * the right to do syscalls); + */ + case OP_TRACE_ON: + arch_leave_kernel(task, pid); + tracing = 1; + break; + case OP_REBOOT: + case OP_HALT: + unmap_physmem(); + kmalloc_ok = 0; + os_kill_ptraced_process(pid, 0); + /* Now let's reap remaining zombies */ + errno = 0; + do { + waitpid(-1, &status, + WUNTRACED); + } while (errno != ECHILD); + return(op == OP_REBOOT); + case OP_NONE: + printf("Detaching pid %d\n", pid); + detach(pid, SIGSTOP); + continue; + default: + break; + } + /* OP_EXEC switches host processes on us, + * we want to continue the new one. + */ + pid = cpu_tasks[proc_id].pid; + break; + case (SIGTRAP + 0x80): + if(!tracing && (debugger_pid != -1)){ + child_signal(pid, status & 0x7fff); + continue; + } + tracing = 0; + /* local_using_sysemu has been already set + * below, since if we are here, is_tracing() on + * the traced task was 1, i.e. the process had + * already run through one iteration of the + * loop which executed a OP_TRACE_ON request.*/ + do_syscall(task, pid, local_using_sysemu); + sig = SIGUSR2; + break; + case SIGTRAP: + if(!tracing && (debugger_pid != -1)){ + child_signal(pid, status); + continue; + } + tracing = 0; + break; + case SIGPROF: + if(tracing) sig = 0; + break; + case SIGCHLD: + case SIGHUP: + sig = 0; + break; + case SIGSEGV: + case SIGIO: + case SIGALRM: + case SIGVTALRM: + case SIGFPE: + case SIGBUS: + case SIGILL: + case SIGWINCH: + + default: + tracing = 0; + break; + } + set_tracing(task, tracing); + + if(!tracing && old_tracing) + arch_enter_kernel(task, pid); + + if(!tracing && (debugger_pid != -1) && (sig != 0) && + (sig != SIGALRM) && (sig != SIGVTALRM) && + (sig != SIGSEGV) && (sig != SIGTRAP) && + (sig != SIGUSR2) && (sig != SIGIO) && + (sig != SIGFPE)){ + child_signal(pid, status); + continue; + } + + local_using_sysemu = get_using_sysemu(); + + if(tracing) + cont_type = SELECT_PTRACE_OPERATION(local_using_sysemu, + singlestepping(task)); + else if((debugger_pid != -1) && strace) + cont_type = PTRACE_SYSCALL; + else + cont_type = PTRACE_CONT; + + if(ptrace(cont_type, pid, 0, sig) != 0){ + tracer_panic("ptrace failed to continue " + "process - errno = %d\n", + errno); + } + } + } + return(0); +} + +static int __init uml_debug_setup(char *line, int *add) +{ + char *next; + + debug = 1; + *add = 0; + if(*line != '=') return(0); + line++; + + while(line != NULL){ + next = strchr(line, ','); + if(next) *next++ = '\0'; + + if(!strcmp(line, "go")) debug_stop = 0; + else if(!strcmp(line, "parent")) debug_parent = 1; + else printf("Unknown debug option : '%s'\n", line); + + line = next; + } + return(0); +} + +__uml_setup("debug", uml_debug_setup, +"debug\n" +" Starts up the kernel under the control of gdb. See the \n" +" kernel debugging tutorial and the debugging session pages\n" +" at http://user-mode-linux.sourceforge.net/ for more information.\n\n" +); + +static int __init uml_debugtrace_setup(char *line, int *add) +{ + debug_trace = 1; + return 0; +} +__uml_setup("debugtrace", uml_debugtrace_setup, +"debugtrace\n" +" Causes the tracing thread to pause until it is attached by a\n" +" debugger and continued. This is mostly for debugging crashes\n" +" early during boot, and should be pretty much obsoleted by\n" +" the debug switch.\n\n" +); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/trap_user.c b/arch/um/kernel/tt/trap_user.c new file mode 100644 index 000000000000..92a3820ca543 --- /dev/null +++ b/arch/um/kernel/tt/trap_user.c @@ -0,0 +1,60 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <errno.h> +#include <signal.h> +#include "sysdep/ptrace.h" +#include "signal_user.h" +#include "user_util.h" +#include "kern_util.h" +#include "task.h" +#include "tt.h" + +void sig_handler_common_tt(int sig, void *sc_ptr) +{ + struct sigcontext *sc = sc_ptr; + struct tt_regs save_regs, *r; + struct signal_info *info; + int save_errno = errno, is_user; + + /* This is done because to allow SIGSEGV to be delivered inside a SEGV + * handler. This can happen in copy_user, and if SEGV is disabled, + * the process will die. + */ + if(sig == SIGSEGV) + change_sig(SIGSEGV, 1); + + r = &TASK_REGS(get_current())->tt; + save_regs = *r; + is_user = user_context(SC_SP(sc)); + r->sc = sc; + if(sig != SIGUSR2) + r->syscall = -1; + + info = &sig_info[sig]; + if(!info->is_irq) unblock_signals(); + + (*info->handler)(sig, (union uml_pt_regs *) r); + + if(is_user){ + interrupt_end(); + block_signals(); + set_user_mode(NULL); + } + *r = save_regs; + errno = save_errno; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/uaccess.c b/arch/um/kernel/tt/uaccess.c new file mode 100644 index 000000000000..a72aa632972f --- /dev/null +++ b/arch/um/kernel/tt/uaccess.c @@ -0,0 +1,73 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "asm/uaccess.h" + +int copy_from_user_tt(void *to, const void __user *from, int n) +{ + if(!access_ok_tt(VERIFY_READ, from, n)) + return(n); + + return(__do_copy_from_user(to, from, n, ¤t->thread.fault_addr, + ¤t->thread.fault_catcher)); +} + +int copy_to_user_tt(void __user *to, const void *from, int n) +{ + if(!access_ok_tt(VERIFY_WRITE, to, n)) + return(n); + + return(__do_copy_to_user(to, from, n, ¤t->thread.fault_addr, + ¤t->thread.fault_catcher)); +} + +int strncpy_from_user_tt(char *dst, const char __user *src, int count) +{ + int n; + + if(!access_ok_tt(VERIFY_READ, src, 1)) + return(-EFAULT); + + n = __do_strncpy_from_user(dst, src, count, + ¤t->thread.fault_addr, + ¤t->thread.fault_catcher); + if(n < 0) return(-EFAULT); + return(n); +} + +int __clear_user_tt(void __user *mem, int len) +{ + return(__do_clear_user(mem, len, + ¤t->thread.fault_addr, + ¤t->thread.fault_catcher)); +} + +int clear_user_tt(void __user *mem, int len) +{ + if(!access_ok_tt(VERIFY_WRITE, mem, len)) + return(len); + + return(__do_clear_user(mem, len, ¤t->thread.fault_addr, + ¤t->thread.fault_catcher)); +} + +int strnlen_user_tt(const void __user *str, int len) +{ + return(__do_strnlen_user(str, len, + ¤t->thread.fault_addr, + ¤t->thread.fault_catcher)); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/uaccess_user.c b/arch/um/kernel/tt/uaccess_user.c new file mode 100644 index 000000000000..f01475512ecb --- /dev/null +++ b/arch/um/kernel/tt/uaccess_user.c @@ -0,0 +1,98 @@ +/* + * Copyright (C) 2001 Chris Emerson (cemerson@chiark.greenend.org.uk) + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <setjmp.h> +#include <string.h> +#include "user_util.h" +#include "uml_uaccess.h" +#include "task.h" +#include "kern_util.h" + +int __do_copy_from_user(void *to, const void *from, int n, + void **fault_addr, void **fault_catcher) +{ + struct tt_regs save = TASK_REGS(get_current())->tt; + unsigned long fault; + int faulted; + + fault = __do_user_copy(to, from, n, fault_addr, fault_catcher, + __do_copy, &faulted); + TASK_REGS(get_current())->tt = save; + + if(!faulted) return(0); + else return(n - (fault - (unsigned long) from)); +} + +static void __do_strncpy(void *dst, const void *src, int count) +{ + strncpy(dst, src, count); +} + +int __do_strncpy_from_user(char *dst, const char *src, unsigned long count, + void **fault_addr, void **fault_catcher) +{ + struct tt_regs save = TASK_REGS(get_current())->tt; + unsigned long fault; + int faulted; + + fault = __do_user_copy(dst, src, count, fault_addr, fault_catcher, + __do_strncpy, &faulted); + TASK_REGS(get_current())->tt = save; + + if(!faulted) return(strlen(dst)); + else return(-1); +} + +static void __do_clear(void *to, const void *from, int n) +{ + memset(to, 0, n); +} + +int __do_clear_user(void *mem, unsigned long len, + void **fault_addr, void **fault_catcher) +{ + struct tt_regs save = TASK_REGS(get_current())->tt; + unsigned long fault; + int faulted; + + fault = __do_user_copy(mem, NULL, len, fault_addr, fault_catcher, + __do_clear, &faulted); + TASK_REGS(get_current())->tt = save; + + if(!faulted) return(0); + else return(len - (fault - (unsigned long) mem)); +} + +int __do_strnlen_user(const char *str, unsigned long n, + void **fault_addr, void **fault_catcher) +{ + struct tt_regs save = TASK_REGS(get_current())->tt; + int ret; + unsigned long *faddrp = (unsigned long *)fault_addr; + sigjmp_buf jbuf; + + *fault_catcher = &jbuf; + if(sigsetjmp(jbuf, 1) == 0) + ret = strlen(str) + 1; + else ret = *faddrp - (unsigned long) str; + + *fault_addr = NULL; + *fault_catcher = NULL; + + TASK_REGS(get_current())->tt = save; + return ret; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/unmap.c b/arch/um/kernel/tt/unmap.c new file mode 100644 index 000000000000..3f7aecdbe532 --- /dev/null +++ b/arch/um/kernel/tt/unmap.c @@ -0,0 +1,31 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <sys/mman.h> + +int switcheroo(int fd, int prot, void *from, void *to, int size) +{ + if(munmap(to, size) < 0){ + return(-1); + } + if(mmap(to, size, prot, MAP_SHARED | MAP_FIXED, fd, 0) != to){ + return(-1); + } + if(munmap(from, size) < 0){ + return(-1); + } + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tty_log.c b/arch/um/kernel/tty_log.c new file mode 100644 index 000000000000..9ada656f68ce --- /dev/null +++ b/arch/um/kernel/tty_log.c @@ -0,0 +1,230 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) and + * geoffrey hing <ghing@net.ohio-state.edu> + * Licensed under the GPL + */ + +#include <errno.h> +#include <string.h> +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <sys/time.h> +#include "init.h" +#include "user.h" +#include "kern_util.h" +#include "os.h" + +#define TTY_LOG_DIR "./" + +/* Set early in boot and then unchanged */ +static char *tty_log_dir = TTY_LOG_DIR; +static int tty_log_fd = -1; + +#define TTY_LOG_OPEN 1 +#define TTY_LOG_CLOSE 2 +#define TTY_LOG_WRITE 3 +#define TTY_LOG_EXEC 4 + +#define TTY_READ 1 +#define TTY_WRITE 2 + +struct tty_log_buf { + int what; + unsigned long tty; + int len; + int direction; + unsigned long sec; + unsigned long usec; +}; + +int open_tty_log(void *tty, void *current_tty) +{ + struct timeval tv; + struct tty_log_buf data; + char buf[strlen(tty_log_dir) + sizeof("01234567890-01234567\0")]; + int fd; + + gettimeofday(&tv, NULL); + if(tty_log_fd != -1){ + data = ((struct tty_log_buf) { .what = TTY_LOG_OPEN, + .tty = (unsigned long) tty, + .len = sizeof(current_tty), + .direction = 0, + .sec = tv.tv_sec, + .usec = tv.tv_usec } ); + os_write_file(tty_log_fd, &data, sizeof(data)); + os_write_file(tty_log_fd, ¤t_tty, data.len); + return(tty_log_fd); + } + + sprintf(buf, "%s/%0u-%0u", tty_log_dir, (unsigned int) tv.tv_sec, + (unsigned int) tv.tv_usec); + + fd = os_open_file(buf, of_append(of_create(of_rdwr(OPENFLAGS()))), + 0644); + if(fd < 0){ + printk("open_tty_log : couldn't open '%s', errno = %d\n", + buf, -fd); + } + return(fd); +} + +void close_tty_log(int fd, void *tty) +{ + struct tty_log_buf data; + struct timeval tv; + + if(tty_log_fd != -1){ + gettimeofday(&tv, NULL); + data = ((struct tty_log_buf) { .what = TTY_LOG_CLOSE, + .tty = (unsigned long) tty, + .len = 0, + .direction = 0, + .sec = tv.tv_sec, + .usec = tv.tv_usec } ); + os_write_file(tty_log_fd, &data, sizeof(data)); + return; + } + os_close_file(fd); +} + +static int log_chunk(int fd, const char *buf, int len) +{ + int total = 0, try, missed, n; + char chunk[64]; + + while(len > 0){ + try = (len > sizeof(chunk)) ? sizeof(chunk) : len; + missed = copy_from_user_proc(chunk, (char *) buf, try); + try -= missed; + n = os_write_file(fd, chunk, try); + if(n != try) { + if(n < 0) + return(n); + return(-EIO); + } + if(missed != 0) + return(-EFAULT); + + len -= try; + total += try; + buf += try; + } + + return(total); +} + +int write_tty_log(int fd, const char *buf, int len, void *tty, int is_read) +{ + struct timeval tv; + struct tty_log_buf data; + int direction; + + if(fd == tty_log_fd){ + gettimeofday(&tv, NULL); + direction = is_read ? TTY_READ : TTY_WRITE; + data = ((struct tty_log_buf) { .what = TTY_LOG_WRITE, + .tty = (unsigned long) tty, + .len = len, + .direction = direction, + .sec = tv.tv_sec, + .usec = tv.tv_usec } ); + os_write_file(tty_log_fd, &data, sizeof(data)); + } + + return(log_chunk(fd, buf, len)); +} + +void log_exec(char **argv, void *tty) +{ + struct timeval tv; + struct tty_log_buf data; + char **ptr,*arg; + int len; + + if(tty_log_fd == -1) return; + + gettimeofday(&tv, NULL); + + len = 0; + for(ptr = argv; ; ptr++){ + if(copy_from_user_proc(&arg, ptr, sizeof(arg))) + return; + if(arg == NULL) break; + len += strlen_user_proc(arg); + } + + data = ((struct tty_log_buf) { .what = TTY_LOG_EXEC, + .tty = (unsigned long) tty, + .len = len, + .direction = 0, + .sec = tv.tv_sec, + .usec = tv.tv_usec } ); + os_write_file(tty_log_fd, &data, sizeof(data)); + + for(ptr = argv; ; ptr++){ + if(copy_from_user_proc(&arg, ptr, sizeof(arg))) + return; + if(arg == NULL) break; + log_chunk(tty_log_fd, arg, strlen_user_proc(arg)); + } +} + +extern void register_tty_logger(int (*opener)(void *, void *), + int (*writer)(int, const char *, int, + void *, int), + void (*closer)(int, void *)); + +static int register_logger(void) +{ + register_tty_logger(open_tty_log, write_tty_log, close_tty_log); + return(0); +} + +__uml_initcall(register_logger); + +static int __init set_tty_log_dir(char *name, int *add) +{ + tty_log_dir = name; + return 0; +} + +__uml_setup("tty_log_dir=", set_tty_log_dir, +"tty_log_dir=<directory>\n" +" This is used to specify the directory where the logs of all pty\n" +" data from this UML machine will be written.\n\n" +); + +static int __init set_tty_log_fd(char *name, int *add) +{ + char *end; + + tty_log_fd = strtoul(name, &end, 0); + if((*end != '\0') || (end == name)){ + printf("set_tty_log_fd - strtoul failed on '%s'\n", name); + tty_log_fd = -1; + } + + *add = 0; + return 0; +} + +__uml_setup("tty_log_fd=", set_tty_log_fd, +"tty_log_fd=<fd>\n" +" This is used to specify a preconfigured file descriptor to which all\n" +" tty data will be written. Preconfigure the descriptor with something\n" +" like '10>tty_log tty_log_fd=10'.\n\n" +); + + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/uaccess_user.c b/arch/um/kernel/uaccess_user.c new file mode 100644 index 000000000000..d035257ed0af --- /dev/null +++ b/arch/um/kernel/uaccess_user.c @@ -0,0 +1,64 @@ +/* + * Copyright (C) 2001 Chris Emerson (cemerson@chiark.greenend.org.uk) + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <setjmp.h> +#include <string.h> + +/* These are here rather than tt/uaccess.c because skas mode needs them in + * order to do SIGBUS recovery when a tmpfs mount runs out of room. + */ + +unsigned long __do_user_copy(void *to, const void *from, int n, + void **fault_addr, void **fault_catcher, + void (*op)(void *to, const void *from, + int n), int *faulted_out) +{ + unsigned long *faddrp = (unsigned long *) fault_addr, ret; + + sigjmp_buf jbuf; + *fault_catcher = &jbuf; + if(sigsetjmp(jbuf, 1) == 0){ + (*op)(to, from, n); + ret = 0; + *faulted_out = 0; + } + else { + ret = *faddrp; + *faulted_out = 1; + } + *fault_addr = NULL; + *fault_catcher = NULL; + return ret; +} + +void __do_copy(void *to, const void *from, int n) +{ + memcpy(to, from, n); +} + + +int __do_copy_to_user(void *to, const void *from, int n, + void **fault_addr, void **fault_catcher) +{ + unsigned long fault; + int faulted; + + fault = __do_user_copy(to, from, n, fault_addr, fault_catcher, + __do_copy, &faulted); + if(!faulted) return(0); + else return(n - (fault - (unsigned long) to)); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c new file mode 100644 index 000000000000..5c49d88eed3d --- /dev/null +++ b/arch/um/kernel/um_arch.c @@ -0,0 +1,467 @@ +/* + * Copyright (C) 2000, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/kernel.h" +#include "linux/sched.h" +#include "linux/notifier.h" +#include "linux/mm.h" +#include "linux/types.h" +#include "linux/tty.h" +#include "linux/init.h" +#include "linux/bootmem.h" +#include "linux/spinlock.h" +#include "linux/utsname.h" +#include "linux/sysrq.h" +#include "linux/seq_file.h" +#include "linux/delay.h" +#include "linux/module.h" +#include "asm/page.h" +#include "asm/pgtable.h" +#include "asm/ptrace.h" +#include "asm/elf.h" +#include "asm/user.h" +#include "ubd_user.h" +#include "asm/current.h" +#include "asm/setup.h" +#include "user_util.h" +#include "kern_util.h" +#include "kern.h" +#include "mem_user.h" +#include "mem.h" +#include "umid.h" +#include "initrd.h" +#include "init.h" +#include "os.h" +#include "choose-mode.h" +#include "mode_kern.h" +#include "mode.h" + +#define DEFAULT_COMMAND_LINE "root=98:0" + +/* Changed in linux_main and setup_arch, which run before SMP is started */ +char command_line[COMMAND_LINE_SIZE] = { 0 }; + +void add_arg(char *arg) +{ + if (strlen(command_line) + strlen(arg) + 1 > COMMAND_LINE_SIZE) { + printf("add_arg: Too many command line arguments!\n"); + exit(1); + } + if(strlen(command_line) > 0) + strcat(command_line, " "); + strcat(command_line, arg); +} + +struct cpuinfo_um boot_cpu_data = { + .loops_per_jiffy = 0, + .ipi_pipe = { -1, -1 } +}; + +unsigned long thread_saved_pc(struct task_struct *task) +{ + return(os_process_pc(CHOOSE_MODE_PROC(thread_pid_tt, thread_pid_skas, + task))); +} + +static int show_cpuinfo(struct seq_file *m, void *v) +{ + int index = 0; + +#ifdef CONFIG_SMP + index = (struct cpuinfo_um *) v - cpu_data; + if (!cpu_online(index)) + return 0; +#endif + + seq_printf(m, "processor\t: %d\n", index); + seq_printf(m, "vendor_id\t: User Mode Linux\n"); + seq_printf(m, "model name\t: UML\n"); + seq_printf(m, "mode\t\t: %s\n", CHOOSE_MODE("tt", "skas")); + seq_printf(m, "host\t\t: %s\n", host_info); + seq_printf(m, "bogomips\t: %lu.%02lu\n\n", + loops_per_jiffy/(500000/HZ), + (loops_per_jiffy/(5000/HZ)) % 100); + + return(0); +} + +static void *c_start(struct seq_file *m, loff_t *pos) +{ + return *pos < NR_CPUS ? cpu_data + *pos : NULL; +} + +static void *c_next(struct seq_file *m, void *v, loff_t *pos) +{ + ++*pos; + return c_start(m, pos); +} + +static void c_stop(struct seq_file *m, void *v) +{ +} + +struct seq_operations cpuinfo_op = { + .start = c_start, + .next = c_next, + .stop = c_stop, + .show = show_cpuinfo, +}; + +pte_t * __bad_pagetable(void) +{ + panic("Someone should implement __bad_pagetable"); + return(NULL); +} + +/* Set in linux_main */ +unsigned long host_task_size; +unsigned long task_size; + +unsigned long uml_start; + +/* Set in early boot */ +unsigned long uml_physmem; +unsigned long uml_reserved; +unsigned long start_vm; +unsigned long end_vm; +int ncpus = 1; + +#ifdef CONFIG_MODE_TT +/* Pointer set in linux_main, the array itself is private to each thread, + * and changed at address space creation time so this poses no concurrency + * problems. + */ +static char *argv1_begin = NULL; +static char *argv1_end = NULL; +#endif + +/* Set in early boot */ +static int have_root __initdata = 0; +long physmem_size = 32 * 1024 * 1024; + +void set_cmdline(char *cmd) +{ +#ifdef CONFIG_MODE_TT + char *umid, *ptr; + + if(CHOOSE_MODE(honeypot, 0)) return; + + umid = get_umid(1); + if(umid != NULL){ + snprintf(argv1_begin, + (argv1_end - argv1_begin) * sizeof(*ptr), + "(%s) ", umid); + ptr = &argv1_begin[strlen(argv1_begin)]; + } + else ptr = argv1_begin; + + snprintf(ptr, (argv1_end - ptr) * sizeof(*ptr), "[%s]", cmd); + memset(argv1_begin + strlen(argv1_begin), '\0', + argv1_end - argv1_begin - strlen(argv1_begin)); +#endif +} + +static char *usage_string = +"User Mode Linux v%s\n" +" available at http://user-mode-linux.sourceforge.net/\n\n"; + +static int __init uml_version_setup(char *line, int *add) +{ + printf("%s\n", system_utsname.release); + exit(0); + + return 0; +} + +__uml_setup("--version", uml_version_setup, +"--version\n" +" Prints the version number of the kernel.\n\n" +); + +static int __init uml_root_setup(char *line, int *add) +{ + have_root = 1; + return 0; +} + +__uml_setup("root=", uml_root_setup, +"root=<file containing the root fs>\n" +" This is actually used by the generic kernel in exactly the same\n" +" way as in any other kernel. If you configure a number of block\n" +" devices and want to boot off something other than ubd0, you \n" +" would use something like:\n" +" root=/dev/ubd5\n\n" +); + +#ifdef CONFIG_SMP +static int __init uml_ncpus_setup(char *line, int *add) +{ + if (!sscanf(line, "%d", &ncpus)) { + printf("Couldn't parse [%s]\n", line); + return -1; + } + + return 0; +} + +__uml_setup("ncpus=", uml_ncpus_setup, +"ncpus=<# of desired CPUs>\n" +" This tells an SMP kernel how many virtual processors to start.\n\n" +); +#endif + +static int force_tt = 0; + +#if defined(CONFIG_MODE_TT) && defined(CONFIG_MODE_SKAS) +#define DEFAULT_TT 0 + +static int __init mode_tt_setup(char *line, int *add) +{ + force_tt = 1; + return(0); +} + +#else +#ifdef CONFIG_MODE_SKAS + +#define DEFAULT_TT 0 + +static int __init mode_tt_setup(char *line, int *add) +{ + printf("CONFIG_MODE_TT disabled - 'mode=tt' ignored\n"); + return(0); +} + +#else +#ifdef CONFIG_MODE_TT + +#define DEFAULT_TT 1 + +static int __init mode_tt_setup(char *line, int *add) +{ + printf("CONFIG_MODE_SKAS disabled - 'mode=tt' redundant\n"); + return(0); +} + +#else + +#error Either CONFIG_MODE_TT or CONFIG_MODE_SKAS must be enabled + +#endif +#endif +#endif + +__uml_setup("mode=tt", mode_tt_setup, +"mode=tt\n" +" When both CONFIG_MODE_TT and CONFIG_MODE_SKAS are enabled, this option\n" +" forces UML to run in tt (tracing thread) mode. It is not the default\n" +" because it's slower and less secure than skas mode.\n\n" +); + +int mode_tt = DEFAULT_TT; + +static int __init Usage(char *line, int *add) +{ + const char **p; + + printf(usage_string, system_utsname.release); + p = &__uml_help_start; + while (p < &__uml_help_end) { + printf("%s", *p); + p++; + } + exit(0); + + return 0; +} + +__uml_setup("--help", Usage, +"--help\n" +" Prints this message.\n\n" +); + +static int __init uml_checksetup(char *line, int *add) +{ + struct uml_param *p; + + p = &__uml_setup_start; + while(p < &__uml_setup_end) { + int n; + + n = strlen(p->str); + if(!strncmp(line, p->str, n)){ + if (p->setup_func(line + n, add)) return 1; + } + p++; + } + return 0; +} + +static void __init uml_postsetup(void) +{ + initcall_t *p; + + p = &__uml_postsetup_start; + while(p < &__uml_postsetup_end){ + (*p)(); + p++; + } + return; +} + +/* Set during early boot */ +unsigned long brk_start; +unsigned long end_iomem; +EXPORT_SYMBOL(end_iomem); + +#define MIN_VMALLOC (32 * 1024 * 1024) + +int linux_main(int argc, char **argv) +{ + unsigned long avail, diff; + unsigned long virtmem_size, max_physmem; + unsigned int i, add; + + for (i = 1; i < argc; i++){ + if((i == 1) && (argv[i][0] == ' ')) continue; + add = 1; + uml_checksetup(argv[i], &add); + if (add) + add_arg(argv[i]); + } + if(have_root == 0) + add_arg(DEFAULT_COMMAND_LINE); + + mode_tt = force_tt ? 1 : !can_do_skas(); +#ifndef CONFIG_MODE_TT + if (mode_tt) { + /*Since CONFIG_MODE_TT is #undef'ed, force_tt cannot be 1. So, + * can_do_skas() returned 0, and the message is correct. */ + printf("Support for TT mode is disabled, and no SKAS support is present on the host.\n"); + exit(1); + } +#endif + uml_start = CHOOSE_MODE_PROC(set_task_sizes_tt, set_task_sizes_skas, 0, + &host_task_size, &task_size); + + /* Need to check this early because mmapping happens before the + * kernel is running. + */ + check_tmpexec(); + + brk_start = (unsigned long) sbrk(0); + CHOOSE_MODE_PROC(before_mem_tt, before_mem_skas, brk_start); + /* Increase physical memory size for exec-shield users + so they actually get what they asked for. This should + add zero for non-exec shield users */ + + diff = UML_ROUND_UP(brk_start) - UML_ROUND_UP(&_end); + if(diff > 1024 * 1024){ + printf("Adding %ld bytes to physical memory to account for " + "exec-shield gap\n", diff); + physmem_size += UML_ROUND_UP(brk_start) - UML_ROUND_UP(&_end); + } + + uml_physmem = uml_start; + + /* Reserve up to 4M after the current brk */ + uml_reserved = ROUND_4M(brk_start) + (1 << 22); + + setup_machinename(system_utsname.machine); + +#ifdef CONFIG_MODE_TT + argv1_begin = argv[1]; + argv1_end = &argv[1][strlen(argv[1])]; +#endif + + highmem = 0; + iomem_size = (iomem_size + PAGE_SIZE - 1) & PAGE_MASK; + max_physmem = get_kmem_end() - uml_physmem - iomem_size - MIN_VMALLOC; + + /* Zones have to begin on a 1 << MAX_ORDER page boundary, + * so this makes sure that's true for highmem + */ + max_physmem &= ~((1 << (PAGE_SHIFT + MAX_ORDER)) - 1); + if(physmem_size + iomem_size > max_physmem){ + highmem = physmem_size + iomem_size - max_physmem; + physmem_size -= highmem; +#ifndef CONFIG_HIGHMEM + highmem = 0; + printf("CONFIG_HIGHMEM not enabled - physical memory shrunk " + "to %ld bytes\n", physmem_size); +#endif + } + + high_physmem = uml_physmem + physmem_size; + end_iomem = high_physmem + iomem_size; + high_memory = (void *) end_iomem; + + start_vm = VMALLOC_START; + + setup_physmem(uml_physmem, uml_reserved, physmem_size, highmem); + if(init_maps(physmem_size, iomem_size, highmem)){ + printf("Failed to allocate mem_map for %ld bytes of physical " + "memory and %ld bytes of highmem\n", physmem_size, + highmem); + exit(1); + } + + virtmem_size = physmem_size; + avail = get_kmem_end() - start_vm; + if(physmem_size > avail) virtmem_size = avail; + end_vm = start_vm + virtmem_size; + + if(virtmem_size < physmem_size) + printf("Kernel virtual memory size shrunk to %ld bytes\n", + virtmem_size); + + uml_postsetup(); + + task_protections((unsigned long) &init_thread_info); + os_flush_stdout(); + + return(CHOOSE_MODE(start_uml_tt(), start_uml_skas())); +} + +extern int uml_exitcode; + +static int panic_exit(struct notifier_block *self, unsigned long unused1, + void *unused2) +{ + bust_spinlocks(1); + show_regs(&(current->thread.regs)); + bust_spinlocks(0); + uml_exitcode = 1; + machine_halt(); + return(0); +} + +static struct notifier_block panic_exit_notifier = { + .notifier_call = panic_exit, + .next = NULL, + .priority = 0 +}; + +void __init setup_arch(char **cmdline_p) +{ + notifier_chain_register(&panic_notifier_list, &panic_exit_notifier); + paging_init(); + strlcpy(saved_command_line, command_line, COMMAND_LINE_SIZE); + *cmdline_p = command_line; + setup_hostinfo(); +} + +void __init check_bugs(void) +{ + arch_check_bugs(); + check_ptrace(); + check_sigio(); + check_devanon(); +} + +void apply_alternatives(void *start, void *end) +{ +} diff --git a/arch/um/kernel/umid.c b/arch/um/kernel/umid.c new file mode 100644 index 000000000000..186c28885016 --- /dev/null +++ b/arch/um/kernel/umid.c @@ -0,0 +1,325 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <unistd.h> +#include <errno.h> +#include <string.h> +#include <stdlib.h> +#include <dirent.h> +#include <signal.h> +#include <sys/stat.h> +#include <sys/param.h> +#include "user.h" +#include "umid.h" +#include "init.h" +#include "os.h" +#include "user_util.h" +#include "choose-mode.h" + +#define UMID_LEN 64 +#define UML_DIR "~/.uml/" + +/* Changed by set_umid and make_umid, which are run early in boot */ +static char umid[UMID_LEN] = { 0 }; + +/* Changed by set_uml_dir and make_uml_dir, which are run early in boot */ +static char *uml_dir = UML_DIR; + +/* Changed by set_umid */ +static int umid_is_random = 1; +static int umid_inited = 0; + +static int make_umid(int (*printer)(const char *fmt, ...)); + +static int __init set_umid(char *name, int is_random, + int (*printer)(const char *fmt, ...)) +{ + if(umid_inited){ + (*printer)("Unique machine name can't be set twice\n"); + return(-1); + } + + if(strlen(name) > UMID_LEN - 1) + (*printer)("Unique machine name is being truncated to %d " + "characters\n", UMID_LEN); + strlcpy(umid, name, sizeof(umid)); + + umid_is_random = is_random; + umid_inited = 1; + return 0; +} + +static int __init set_umid_arg(char *name, int *add) +{ + *add = 0; + return(set_umid(name, 0, printf)); +} + +__uml_setup("umid=", set_umid_arg, +"umid=<name>\n" +" This is used to assign a unique identity to this UML machine and\n" +" is used for naming the pid file and management console socket.\n\n" +); + +int __init umid_file_name(char *name, char *buf, int len) +{ + int n; + + if(!umid_inited && make_umid(printk)) return(-1); + + n = strlen(uml_dir) + strlen(umid) + strlen(name) + 1; + if(n > len){ + printk("umid_file_name : buffer too short\n"); + return(-1); + } + + sprintf(buf, "%s%s/%s", uml_dir, umid, name); + return(0); +} + +extern int tracing_pid; + +static int __init create_pid_file(void) +{ + char file[strlen(uml_dir) + UMID_LEN + sizeof("/pid\0")]; + char pid[sizeof("nnnnn\0")]; + int fd, n; + + if(umid_file_name("pid", file, sizeof(file))) return 0; + + fd = os_open_file(file, of_create(of_excl(of_rdwr(OPENFLAGS()))), + 0644); + if(fd < 0){ + printf("Open of machine pid file \"%s\" failed: %s\n", + file, strerror(-fd)); + return 0; + } + + sprintf(pid, "%d\n", os_getpid()); + n = os_write_file(fd, pid, strlen(pid)); + if(n != strlen(pid)) + printf("Write of pid file failed - err = %d\n", -n); + os_close_file(fd); + return 0; +} + +static int actually_do_remove(char *dir) +{ + DIR *directory; + struct dirent *ent; + int len; + char file[256]; + + directory = opendir(dir); + if(directory == NULL){ + printk("actually_do_remove : couldn't open directory '%s', " + "errno = %d\n", dir, errno); + return(1); + } + while((ent = readdir(directory)) != NULL){ + if(!strcmp(ent->d_name, ".") || !strcmp(ent->d_name, "..")) + continue; + len = strlen(dir) + sizeof("/") + strlen(ent->d_name) + 1; + if(len > sizeof(file)){ + printk("Not deleting '%s' from '%s' - name too long\n", + ent->d_name, dir); + continue; + } + sprintf(file, "%s/%s", dir, ent->d_name); + if(unlink(file) < 0){ + printk("actually_do_remove : couldn't remove '%s' " + "from '%s', errno = %d\n", ent->d_name, dir, + errno); + return(1); + } + } + if(rmdir(dir) < 0){ + printk("actually_do_remove : couldn't rmdir '%s', " + "errno = %d\n", dir, errno); + return(1); + } + return(0); +} + +void remove_umid_dir(void) +{ + char dir[strlen(uml_dir) + UMID_LEN + 1]; + if(!umid_inited) return; + + sprintf(dir, "%s%s", uml_dir, umid); + actually_do_remove(dir); +} + +char *get_umid(int only_if_set) +{ + if(only_if_set && umid_is_random) return(NULL); + return(umid); +} + +int not_dead_yet(char *dir) +{ + char file[strlen(uml_dir) + UMID_LEN + sizeof("/pid\0")]; + char pid[sizeof("nnnnn\0")], *end; + int dead, fd, p, n; + + sprintf(file, "%s/pid", dir); + dead = 0; + fd = os_open_file(file, of_read(OPENFLAGS()), 0); + if(fd < 0){ + if(fd != -ENOENT){ + printk("not_dead_yet : couldn't open pid file '%s', " + "err = %d\n", file, -fd); + return(1); + } + dead = 1; + } + if(fd > 0){ + n = os_read_file(fd, pid, sizeof(pid)); + if(n < 0){ + printk("not_dead_yet : couldn't read pid file '%s', " + "err = %d\n", file, -n); + return(1); + } + p = strtoul(pid, &end, 0); + if(end == pid){ + printk("not_dead_yet : couldn't parse pid file '%s', " + "errno = %d\n", file, errno); + dead = 1; + } + if(((kill(p, 0) < 0) && (errno == ESRCH)) || + (p == CHOOSE_MODE(tracing_pid, os_getpid()))) + dead = 1; + } + if(!dead) return(1); + return(actually_do_remove(dir)); +} + +static int __init set_uml_dir(char *name, int *add) +{ + if((strlen(name) > 0) && (name[strlen(name) - 1] != '/')){ + uml_dir = malloc(strlen(name) + 2); + if(uml_dir == NULL){ + printf("Failed to malloc uml_dir - error = %d\n", + errno); + uml_dir = name; + /* Return 0 here because do_initcalls doesn't look at + * the return value. + */ + return(0); + } + sprintf(uml_dir, "%s/", name); + } + else uml_dir = name; + return(0); +} + +static int __init make_uml_dir(void) +{ + char dir[MAXPATHLEN + 1] = { '\0' }; + int len; + + if(*uml_dir == '~'){ + char *home = getenv("HOME"); + + if(home == NULL){ + printf("make_uml_dir : no value in environment for " + "$HOME\n"); + exit(1); + } + strlcpy(dir, home, sizeof(dir)); + uml_dir++; + } + len = strlen(dir); + strncat(dir, uml_dir, sizeof(dir) - len); + len = strlen(dir); + if((len > 0) && (len < sizeof(dir) - 1) && (dir[len - 1] != '/')){ + dir[len] = '/'; + dir[len + 1] = '\0'; + } + + uml_dir = malloc(strlen(dir) + 1); + if(uml_dir == NULL){ + printf("make_uml_dir : malloc failed, errno = %d\n", errno); + exit(1); + } + strcpy(uml_dir, dir); + + if((mkdir(uml_dir, 0777) < 0) && (errno != EEXIST)){ + printf("Failed to mkdir %s: %s\n", uml_dir, strerror(errno)); + return(-1); + } + return 0; +} + +static int __init make_umid(int (*printer)(const char *fmt, ...)) +{ + int fd, err; + char tmp[strlen(uml_dir) + UMID_LEN + 1]; + + strlcpy(tmp, uml_dir, sizeof(tmp)); + + if(!umid_inited){ + strcat(tmp, "XXXXXX"); + fd = mkstemp(tmp); + if(fd < 0){ + (*printer)("make_umid - mkstemp(%s) failed: %s\n", + tmp,strerror(errno)); + return(1); + } + + os_close_file(fd); + /* There's a nice tiny little race between this unlink and + * the mkdir below. It'd be nice if there were a mkstemp + * for directories. + */ + unlink(tmp); + set_umid(&tmp[strlen(uml_dir)], 1, printer); + } + + sprintf(tmp, "%s%s", uml_dir, umid); + + err = mkdir(tmp, 0777); + if(err < 0){ + if(errno == EEXIST){ + if(not_dead_yet(tmp)){ + (*printer)("umid '%s' is in use\n", umid); + return(-1); + } + err = mkdir(tmp, 0777); + } + } + if(err < 0){ + (*printer)("Failed to create %s - errno = %d\n", umid, errno); + return(-1); + } + + return(0); +} + +__uml_setup("uml_dir=", set_uml_dir, +"uml_dir=<directory>\n" +" The location to place the pid and umid files.\n\n" +); + +static int __init make_umid_setup(void) +{ + /* one function with the ordering we need ... */ + make_uml_dir(); + make_umid(printf); + return create_pid_file(); +} +__uml_postsetup(make_umid_setup); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/uml.lds.S b/arch/um/kernel/uml.lds.S new file mode 100644 index 000000000000..76eadb309189 --- /dev/null +++ b/arch/um/kernel/uml.lds.S @@ -0,0 +1,106 @@ +#include <asm-generic/vmlinux.lds.h> + +OUTPUT_FORMAT(ELF_FORMAT) +OUTPUT_ARCH(ELF_ARCH) +ENTRY(_start) +jiffies = jiffies_64; + +SECTIONS +{ + /*This must contain the right address - not quite the default ELF one.*/ + PROVIDE (__executable_start = START); + . = START + SIZEOF_HEADERS; + + /* Used in arch/um/kernel/mem.c. Any memory between START and __binary_start + * is remapped.*/ + __binary_start = .; +#ifdef MODE_TT + .thread_private : { + __start_thread_private = .; + errno = .; + . += 4; + arch/um/kernel/tt/unmap_fin.o (.data) + __end_thread_private = .; + } + . = ALIGN(4096); + .remap : { arch/um/kernel/tt/unmap_fin.o (.text) } + + /* We want it only if we are in MODE_TT. In both cases, however, when MODE_TT + * is off the resulting binary segfaults.*/ + + . = ALIGN(4096); /* Init code and data */ +#endif + + _stext = .; + __init_begin = .; + .init.text : { + _sinittext = .; + *(.init.text) + _einittext = .; + } + . = ALIGN(4096); + .text : + { + *(.text) + SCHED_TEXT + LOCK_TEXT + *(.fixup) + /* .gnu.warning sections are handled specially by elf32.em. */ + *(.gnu.warning) + *(.gnu.linkonce.t*) + } + + #include "asm/common.lds.S" + + init.data : { *(init.data) } + .data : + { + . = ALIGN(KERNEL_STACK_SIZE); /* init_task */ + *(.data.init_task) + *(.data) + *(.gnu.linkonce.d*) + CONSTRUCTORS + } + .data1 : { *(.data1) } + .ctors : + { + *(.ctors) + } + .dtors : + { + *(.dtors) + } + + .got : { *(.got.plt) *(.got) } + .dynamic : { *(.dynamic) } + /* We want the small data sections together, so single-instruction offsets + can access them all, and initialized data all before uninitialized, so + we can shorten the on-disk segment size. */ + .sdata : { *(.sdata) } + _edata = .; + PROVIDE (edata = .); + . = ALIGN(0x1000); + .sbss : + { + __bss_start = .; + PROVIDE(_bss_start = .); + *(.sbss) + *(.scommon) + } + .bss : + { + *(.dynbss) + *(.bss) + *(COMMON) + } + _end = . ; + PROVIDE (end = .); + /* Stabs debugging sections. */ + .stab 0 : { *(.stab) } + .stabstr 0 : { *(.stabstr) } + .stab.excl 0 : { *(.stab.excl) } + .stab.exclstr 0 : { *(.stab.exclstr) } + .stab.index 0 : { *(.stab.index) } + .stab.indexstr 0 : { *(.stab.indexstr) } + .comment 0 : { *(.comment) } +} diff --git a/arch/um/kernel/user_util.c b/arch/um/kernel/user_util.c new file mode 100644 index 000000000000..954ff67cc8b3 --- /dev/null +++ b/arch/um/kernel/user_util.c @@ -0,0 +1,173 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <limits.h> +#include <setjmp.h> +#include <sys/mman.h> +#include <sys/stat.h> +#include <sys/utsname.h> +#include <sys/param.h> +#include <sys/time.h> +#include "asm/types.h" +#include <ctype.h> +#include <signal.h> +#include <wait.h> +#include <errno.h> +#include <stdarg.h> +#include <sched.h> +#include <termios.h> +#include <string.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "mem_user.h" +#include "init.h" +#include "helper.h" +#include "ptrace_user.h" +#include "uml-config.h" + +void stop(void) +{ + while(1) sleep(1000000); +} + +void stack_protections(unsigned long address) +{ + int prot = PROT_READ | PROT_WRITE | PROT_EXEC; + + if(mprotect((void *) address, page_size(), prot) < 0) + panic("protecting stack failed, errno = %d", errno); +} + +void task_protections(unsigned long address) +{ + unsigned long guard = address + page_size(); + unsigned long stack = guard + page_size(); + int prot = 0, pages; + +#ifdef notdef + if(mprotect((void *) stack, page_size(), prot) < 0) + panic("protecting guard page failed, errno = %d", errno); +#endif + pages = (1 << UML_CONFIG_KERNEL_STACK_ORDER) - 2; + prot = PROT_READ | PROT_WRITE | PROT_EXEC; + if(mprotect((void *) stack, pages * page_size(), prot) < 0) + panic("protecting stack failed, errno = %d", errno); +} + +int wait_for_stop(int pid, int sig, int cont_type, void *relay) +{ + sigset_t *relay_signals = relay; + int status, ret; + + while(1){ + CATCH_EINTR(ret = waitpid(pid, &status, WUNTRACED)); + if((ret < 0) || + !WIFSTOPPED(status) || (WSTOPSIG(status) != sig)){ + if(ret < 0){ + printk("wait failed, errno = %d\n", + errno); + } + else if(WIFEXITED(status)) + printk("process %d exited with status %d\n", + pid, WEXITSTATUS(status)); + else if(WIFSIGNALED(status)) + printk("process %d exited with signal %d\n", + pid, WTERMSIG(status)); + else if((WSTOPSIG(status) == SIGVTALRM) || + (WSTOPSIG(status) == SIGALRM) || + (WSTOPSIG(status) == SIGIO) || + (WSTOPSIG(status) == SIGPROF) || + (WSTOPSIG(status) == SIGCHLD) || + (WSTOPSIG(status) == SIGWINCH) || + (WSTOPSIG(status) == SIGINT)){ + ptrace(cont_type, pid, 0, WSTOPSIG(status)); + continue; + } + else if((relay_signals != NULL) && + sigismember(relay_signals, WSTOPSIG(status))){ + ptrace(cont_type, pid, 0, WSTOPSIG(status)); + continue; + } + else printk("process %d stopped with signal %d\n", + pid, WSTOPSIG(status)); + panic("wait_for_stop failed to wait for %d to stop " + "with %d\n", pid, sig); + } + return(status); + } +} + +int raw(int fd) +{ + struct termios tt; + int err; + + CATCH_EINTR(err = tcgetattr(fd, &tt)); + if (err < 0) { + printk("tcgetattr failed, errno = %d\n", errno); + return(-errno); + } + + cfmakeraw(&tt); + + CATCH_EINTR(err = tcsetattr(fd, TCSADRAIN, &tt)); + if (err < 0) { + printk("tcsetattr failed, errno = %d\n", errno); + return(-errno); + } + + /* XXX tcsetattr could have applied only some changes + * (and cfmakeraw() is a set of changes) */ + return(0); +} + +void setup_machinename(char *machine_out) +{ + struct utsname host; + + uname(&host); + strcpy(machine_out, host.machine); +} + +char host_info[(_UTSNAME_LENGTH + 1) * 4 + _UTSNAME_NODENAME_LENGTH + 1]; + +void setup_hostinfo(void) +{ + struct utsname host; + + uname(&host); + sprintf(host_info, "%s %s %s %s %s", host.sysname, host.nodename, + host.release, host.version, host.machine); +} + +int setjmp_wrapper(void (*proc)(void *, void *), ...) +{ + va_list args; + sigjmp_buf buf; + int n; + + n = sigsetjmp(buf, 1); + if(n == 0){ + va_start(args, proc); + (*proc)(&buf, &args); + } + va_end(args); + return(n); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/Makefile b/arch/um/os-Linux/Makefile new file mode 100644 index 000000000000..4ddf540284ce --- /dev/null +++ b/arch/um/os-Linux/Makefile @@ -0,0 +1,13 @@ +# +# Copyright (C) 2000 Jeff Dike (jdike@karaya.com) +# Licensed under the GPL +# + +obj-y = elf_aux.o file.o process.o signal.o time.o tty.o user_syms.o drivers/ \ + sys-$(SUBARCH)/ + +USER_OBJS := elf_aux.o file.o process.o signal.o time.o tty.o + +CFLAGS_user_syms.o += -DSUBARCH_$(SUBARCH) + +include arch/um/scripts/Makefile.rules diff --git a/arch/um/os-Linux/drivers/Makefile b/arch/um/os-Linux/drivers/Makefile new file mode 100644 index 000000000000..6c546dc9222b --- /dev/null +++ b/arch/um/os-Linux/drivers/Makefile @@ -0,0 +1,13 @@ +# +# Copyright (C) 2000, 2002 Jeff Dike (jdike@karaya.com) +# Licensed under the GPL +# + +ethertap-objs := ethertap_kern.o ethertap_user.o +tuntap-objs := tuntap_kern.o tuntap_user.o + +obj-y = +obj-$(CONFIG_UML_NET_ETHERTAP) += ethertap.o +obj-$(CONFIG_UML_NET_TUNTAP) += tuntap.o + +include arch/um/scripts/Makefile.rules diff --git a/arch/um/os-Linux/drivers/etap.h b/arch/um/os-Linux/drivers/etap.h new file mode 100644 index 000000000000..b84f6c4740f7 --- /dev/null +++ b/arch/um/os-Linux/drivers/etap.h @@ -0,0 +1,27 @@ +/* + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "net_user.h" + +struct ethertap_data { + char *dev_name; + char *gate_addr; + int data_fd; + int control_fd; + void *dev; +}; + +extern struct net_user_info ethertap_user_info; + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/drivers/ethertap_kern.c b/arch/um/os-Linux/drivers/ethertap_kern.c new file mode 100644 index 000000000000..6ae4b19d9f50 --- /dev/null +++ b/arch/um/os-Linux/drivers/ethertap_kern.c @@ -0,0 +1,119 @@ +/* + * Copyright (C) 2001 Lennert Buytenhek (buytenh@gnu.org) and + * James Leu (jleu@mindspring.net). + * Copyright (C) 2001 by various other people who didn't put their name here. + * Licensed under the GPL. + */ + +#include "linux/init.h" +#include "linux/netdevice.h" +#include "linux/etherdevice.h" +#include "net_kern.h" +#include "net_user.h" +#include "etap.h" + +struct ethertap_init { + char *dev_name; + char *gate_addr; +}; + +static void etap_init(struct net_device *dev, void *data) +{ + struct uml_net_private *pri; + struct ethertap_data *epri; + struct ethertap_init *init = data; + + pri = dev->priv; + epri = (struct ethertap_data *) pri->user; + epri->dev_name = init->dev_name; + epri->gate_addr = init->gate_addr; + epri->data_fd = -1; + epri->control_fd = -1; + epri->dev = dev; + + printk("ethertap backend - %s", epri->dev_name); + if (epri->gate_addr != NULL) + printk(", IP = %s", epri->gate_addr); + printk("\n"); +} + +static int etap_read(int fd, struct sk_buff **skb, struct uml_net_private *lp) +{ + int len; + + *skb = ether_adjust_skb(*skb, ETH_HEADER_ETHERTAP); + if(*skb == NULL) return(-ENOMEM); + len = net_recvfrom(fd, (*skb)->mac.raw, + (*skb)->dev->mtu + 2 * ETH_HEADER_ETHERTAP); + if(len <= 0) return(len); + skb_pull(*skb, 2); + len -= 2; + return(len); +} + +static int etap_write(int fd, struct sk_buff **skb, struct uml_net_private *lp) +{ + if(skb_headroom(*skb) < 2){ + struct sk_buff *skb2; + + skb2 = skb_realloc_headroom(*skb, 2); + dev_kfree_skb(*skb); + if (skb2 == NULL) return(-ENOMEM); + *skb = skb2; + } + skb_push(*skb, 2); + return(net_send(fd, (*skb)->data, (*skb)->len)); +} + +struct net_kern_info ethertap_kern_info = { + .init = etap_init, + .protocol = eth_protocol, + .read = etap_read, + .write = etap_write, +}; + +int ethertap_setup(char *str, char **mac_out, void *data) +{ + struct ethertap_init *init = data; + + *init = ((struct ethertap_init) + { .dev_name = NULL, + .gate_addr = NULL }); + if(tap_setup_common(str, "ethertap", &init->dev_name, mac_out, + &init->gate_addr)) + return(0); + if(init->dev_name == NULL){ + printk("ethertap_setup : Missing tap device name\n"); + return(0); + } + + return(1); +} + +static struct transport ethertap_transport = { + .list = LIST_HEAD_INIT(ethertap_transport.list), + .name = "ethertap", + .setup = ethertap_setup, + .user = ðertap_user_info, + .kern = ðertap_kern_info, + .private_size = sizeof(struct ethertap_data), +}; + +static int register_ethertap(void) +{ + register_transport(ðertap_transport); + return(1); +} + +__initcall(register_ethertap); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/drivers/ethertap_user.c b/arch/um/os-Linux/drivers/ethertap_user.c new file mode 100644 index 000000000000..cd4d6544da71 --- /dev/null +++ b/arch/um/os-Linux/drivers/ethertap_user.c @@ -0,0 +1,240 @@ +/* + * Copyright (C) 2001 Lennert Buytenhek (buytenh@gnu.org) and + * James Leu (jleu@mindspring.net). + * Copyright (C) 2001 by various other people who didn't put their name here. + * Licensed under the GPL. + */ + +#include <stdio.h> +#include <unistd.h> +#include <stddef.h> +#include <stdlib.h> +#include <sys/errno.h> +#include <sys/socket.h> +#include <sys/wait.h> +#include <sys/un.h> +#include <net/if.h> +#include "user.h" +#include "kern_util.h" +#include "user_util.h" +#include "net_user.h" +#include "etap.h" +#include "helper.h" +#include "os.h" + +#define MAX_PACKET ETH_MAX_PACKET + +void etap_user_init(void *data, void *dev) +{ + struct ethertap_data *pri = data; + + pri->dev = dev; +} + +struct addr_change { + enum { ADD_ADDR, DEL_ADDR } what; + unsigned char addr[4]; + unsigned char netmask[4]; +}; + +static void etap_change(int op, unsigned char *addr, unsigned char *netmask, + int fd) +{ + struct addr_change change; + void *output; + int n; + + change.what = op; + memcpy(change.addr, addr, sizeof(change.addr)); + memcpy(change.netmask, netmask, sizeof(change.netmask)); + n = os_write_file(fd, &change, sizeof(change)); + if(n != sizeof(change)) + printk("etap_change - request failed, err = %d\n", -n); + output = um_kmalloc(page_size()); + if(output == NULL) + printk("etap_change : Failed to allocate output buffer\n"); + read_output(fd, output, page_size()); + if(output != NULL){ + printk("%s", output); + kfree(output); + } +} + +static void etap_open_addr(unsigned char *addr, unsigned char *netmask, + void *arg) +{ + etap_change(ADD_ADDR, addr, netmask, *((int *) arg)); +} + +static void etap_close_addr(unsigned char *addr, unsigned char *netmask, + void *arg) +{ + etap_change(DEL_ADDR, addr, netmask, *((int *) arg)); +} + +struct etap_pre_exec_data { + int control_remote; + int control_me; + int data_me; +}; + +static void etap_pre_exec(void *arg) +{ + struct etap_pre_exec_data *data = arg; + + dup2(data->control_remote, 1); + os_close_file(data->data_me); + os_close_file(data->control_me); +} + +static int etap_tramp(char *dev, char *gate, int control_me, + int control_remote, int data_me, int data_remote) +{ + struct etap_pre_exec_data pe_data; + int pid, status, err, n; + char version_buf[sizeof("nnnnn\0")]; + char data_fd_buf[sizeof("nnnnnn\0")]; + char gate_buf[sizeof("nnn.nnn.nnn.nnn\0")]; + char *setup_args[] = { "uml_net", version_buf, "ethertap", dev, + data_fd_buf, gate_buf, NULL }; + char *nosetup_args[] = { "uml_net", version_buf, "ethertap", + dev, data_fd_buf, NULL }; + char **args, c; + + sprintf(data_fd_buf, "%d", data_remote); + sprintf(version_buf, "%d", UML_NET_VERSION); + if(gate != NULL){ + strcpy(gate_buf, gate); + args = setup_args; + } + else args = nosetup_args; + + err = 0; + pe_data.control_remote = control_remote; + pe_data.control_me = control_me; + pe_data.data_me = data_me; + pid = run_helper(etap_pre_exec, &pe_data, args, NULL); + + if(pid < 0) err = pid; + os_close_file(data_remote); + os_close_file(control_remote); + n = os_read_file(control_me, &c, sizeof(c)); + if(n != sizeof(c)){ + printk("etap_tramp : read of status failed, err = %d\n", -n); + return(-EINVAL); + } + if(c != 1){ + printk("etap_tramp : uml_net failed\n"); + err = -EINVAL; + CATCH_EINTR(n = waitpid(pid, &status, 0)); + if(n < 0) + err = -errno; + else if(!WIFEXITED(status) || (WEXITSTATUS(status) != 1)) + printk("uml_net didn't exit with status 1\n"); + } + return(err); +} + +static int etap_open(void *data) +{ + struct ethertap_data *pri = data; + char *output; + int data_fds[2], control_fds[2], err, output_len; + + err = tap_open_common(pri->dev, pri->gate_addr); + if(err) return(err); + + err = os_pipe(data_fds, 0, 0); + if(err < 0){ + printk("data os_pipe failed - err = %d\n", -err); + return(err); + } + + err = os_pipe(control_fds, 1, 0); + if(err < 0){ + printk("control os_pipe failed - err = %d\n", -err); + return(err); + } + + err = etap_tramp(pri->dev_name, pri->gate_addr, control_fds[0], + control_fds[1], data_fds[0], data_fds[1]); + output_len = page_size(); + output = um_kmalloc(output_len); + read_output(control_fds[0], output, output_len); + + if(output == NULL) + printk("etap_open : failed to allocate output buffer\n"); + else { + printk("%s", output); + kfree(output); + } + + if(err < 0){ + printk("etap_tramp failed - err = %d\n", -err); + return(err); + } + + pri->data_fd = data_fds[0]; + pri->control_fd = control_fds[0]; + iter_addresses(pri->dev, etap_open_addr, &pri->control_fd); + return(data_fds[0]); +} + +static void etap_close(int fd, void *data) +{ + struct ethertap_data *pri = data; + + iter_addresses(pri->dev, etap_close_addr, &pri->control_fd); + os_close_file(fd); + os_shutdown_socket(pri->data_fd, 1, 1); + os_close_file(pri->data_fd); + pri->data_fd = -1; + os_close_file(pri->control_fd); + pri->control_fd = -1; +} + +static int etap_set_mtu(int mtu, void *data) +{ + return(mtu); +} + +static void etap_add_addr(unsigned char *addr, unsigned char *netmask, + void *data) +{ + struct ethertap_data *pri = data; + + tap_check_ips(pri->gate_addr, addr); + if(pri->control_fd == -1) return; + etap_open_addr(addr, netmask, &pri->control_fd); +} + +static void etap_del_addr(unsigned char *addr, unsigned char *netmask, + void *data) +{ + struct ethertap_data *pri = data; + + if(pri->control_fd == -1) return; + etap_close_addr(addr, netmask, &pri->control_fd); +} + +struct net_user_info ethertap_user_info = { + .init = etap_user_init, + .open = etap_open, + .close = etap_close, + .remove = NULL, + .set_mtu = etap_set_mtu, + .add_address = etap_add_addr, + .delete_address = etap_del_addr, + .max_packet = MAX_PACKET - ETH_HEADER_ETHERTAP +}; + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/drivers/tuntap.h b/arch/um/os-Linux/drivers/tuntap.h new file mode 100644 index 000000000000..25d4a2868814 --- /dev/null +++ b/arch/um/os-Linux/drivers/tuntap.h @@ -0,0 +1,32 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __UM_TUNTAP_H +#define __UM_TUNTAP_H + +#include "net_user.h" + +struct tuntap_data { + char *dev_name; + int fixed_config; + char *gate_addr; + int fd; + void *dev; +}; + +extern struct net_user_info tuntap_user_info; + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/drivers/tuntap_kern.c b/arch/um/os-Linux/drivers/tuntap_kern.c new file mode 100644 index 000000000000..4202b9ebad4c --- /dev/null +++ b/arch/um/os-Linux/drivers/tuntap_kern.c @@ -0,0 +1,104 @@ +/* + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/stddef.h" +#include "linux/netdevice.h" +#include "linux/etherdevice.h" +#include "linux/skbuff.h" +#include "linux/init.h" +#include "asm/errno.h" +#include "net_kern.h" +#include "net_user.h" +#include "tuntap.h" + +struct tuntap_init { + char *dev_name; + char *gate_addr; +}; + +static void tuntap_init(struct net_device *dev, void *data) +{ + struct uml_net_private *pri; + struct tuntap_data *tpri; + struct tuntap_init *init = data; + + pri = dev->priv; + tpri = (struct tuntap_data *) pri->user; + tpri->dev_name = init->dev_name; + tpri->fixed_config = (init->dev_name != NULL); + tpri->gate_addr = init->gate_addr; + tpri->fd = -1; + tpri->dev = dev; + + printk("TUN/TAP backend - "); + if (tpri->gate_addr != NULL) + printk("IP = %s", tpri->gate_addr); + printk("\n"); +} + +static int tuntap_read(int fd, struct sk_buff **skb, + struct uml_net_private *lp) +{ + *skb = ether_adjust_skb(*skb, ETH_HEADER_OTHER); + if(*skb == NULL) return(-ENOMEM); + return(net_read(fd, (*skb)->mac.raw, + (*skb)->dev->mtu + ETH_HEADER_OTHER)); +} + +static int tuntap_write(int fd, struct sk_buff **skb, + struct uml_net_private *lp) +{ + return(net_write(fd, (*skb)->data, (*skb)->len)); +} + +struct net_kern_info tuntap_kern_info = { + .init = tuntap_init, + .protocol = eth_protocol, + .read = tuntap_read, + .write = tuntap_write, +}; + +int tuntap_setup(char *str, char **mac_out, void *data) +{ + struct tuntap_init *init = data; + + *init = ((struct tuntap_init) + { .dev_name = NULL, + .gate_addr = NULL }); + if(tap_setup_common(str, "tuntap", &init->dev_name, mac_out, + &init->gate_addr)) + return(0); + + return(1); +} + +static struct transport tuntap_transport = { + .list = LIST_HEAD_INIT(tuntap_transport.list), + .name = "tuntap", + .setup = tuntap_setup, + .user = &tuntap_user_info, + .kern = &tuntap_kern_info, + .private_size = sizeof(struct tuntap_data), + .setup_size = sizeof(struct tuntap_init), +}; + +static int register_tuntap(void) +{ + register_transport(&tuntap_transport); + return(1); +} + +__initcall(register_tuntap); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/drivers/tuntap_user.c b/arch/um/os-Linux/drivers/tuntap_user.c new file mode 100644 index 000000000000..4b83c6c3f48d --- /dev/null +++ b/arch/um/os-Linux/drivers/tuntap_user.c @@ -0,0 +1,225 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stddef.h> +#include <stdlib.h> +#include <unistd.h> +#include <errno.h> +#include <sys/wait.h> +#include <sys/socket.h> +#include <sys/un.h> +#include <sys/uio.h> +#include <sys/ioctl.h> +#include <net/if.h> +#include <linux/if_tun.h> +#include "net_user.h" +#include "tuntap.h" +#include "kern_util.h" +#include "user_util.h" +#include "user.h" +#include "helper.h" +#include "os.h" + +#define MAX_PACKET ETH_MAX_PACKET + +void tuntap_user_init(void *data, void *dev) +{ + struct tuntap_data *pri = data; + + pri->dev = dev; +} + +static void tuntap_add_addr(unsigned char *addr, unsigned char *netmask, + void *data) +{ + struct tuntap_data *pri = data; + + tap_check_ips(pri->gate_addr, addr); + if((pri->fd == -1) || pri->fixed_config) return; + open_addr(addr, netmask, pri->dev_name); +} + +static void tuntap_del_addr(unsigned char *addr, unsigned char *netmask, + void *data) +{ + struct tuntap_data *pri = data; + + if((pri->fd == -1) || pri->fixed_config) return; + close_addr(addr, netmask, pri->dev_name); +} + +struct tuntap_pre_exec_data { + int stdout; + int close_me; +}; + +static void tuntap_pre_exec(void *arg) +{ + struct tuntap_pre_exec_data *data = arg; + + dup2(data->stdout, 1); + os_close_file(data->close_me); +} + +static int tuntap_open_tramp(char *gate, int *fd_out, int me, int remote, + char *buffer, int buffer_len, int *used_out) +{ + struct tuntap_pre_exec_data data; + char version_buf[sizeof("nnnnn\0")]; + char *argv[] = { "uml_net", version_buf, "tuntap", "up", gate, + NULL }; + char buf[CMSG_SPACE(sizeof(*fd_out))]; + struct msghdr msg; + struct cmsghdr *cmsg; + struct iovec iov; + int pid, n; + + sprintf(version_buf, "%d", UML_NET_VERSION); + + data.stdout = remote; + data.close_me = me; + + pid = run_helper(tuntap_pre_exec, &data, argv, NULL); + + if(pid < 0) return(-pid); + + os_close_file(remote); + + msg.msg_name = NULL; + msg.msg_namelen = 0; + if(buffer != NULL){ + iov = ((struct iovec) { buffer, buffer_len }); + msg.msg_iov = &iov; + msg.msg_iovlen = 1; + } + else { + msg.msg_iov = NULL; + msg.msg_iovlen = 0; + } + msg.msg_control = buf; + msg.msg_controllen = sizeof(buf); + msg.msg_flags = 0; + n = recvmsg(me, &msg, 0); + *used_out = n; + if(n < 0){ + printk("tuntap_open_tramp : recvmsg failed - errno = %d\n", + errno); + return(-errno); + } + CATCH_EINTR(waitpid(pid, NULL, 0)); + + cmsg = CMSG_FIRSTHDR(&msg); + if(cmsg == NULL){ + printk("tuntap_open_tramp : didn't receive a message\n"); + return(-EINVAL); + } + if((cmsg->cmsg_level != SOL_SOCKET) || + (cmsg->cmsg_type != SCM_RIGHTS)){ + printk("tuntap_open_tramp : didn't receive a descriptor\n"); + return(-EINVAL); + } + *fd_out = ((int *) CMSG_DATA(cmsg))[0]; + return(0); +} + +static int tuntap_open(void *data) +{ + struct ifreq ifr; + struct tuntap_data *pri = data; + char *output, *buffer; + int err, fds[2], len, used; + + err = tap_open_common(pri->dev, pri->gate_addr); + if(err < 0) + return(err); + + if(pri->fixed_config){ + pri->fd = os_open_file("/dev/net/tun", of_rdwr(OPENFLAGS()), 0); + if(pri->fd < 0){ + printk("Failed to open /dev/net/tun, err = %d\n", + -pri->fd); + return(pri->fd); + } + memset(&ifr, 0, sizeof(ifr)); + ifr.ifr_flags = IFF_TAP | IFF_NO_PI; + strlcpy(ifr.ifr_name, pri->dev_name, sizeof(ifr.ifr_name)); + if(ioctl(pri->fd, TUNSETIFF, (void *) &ifr) < 0){ + printk("TUNSETIFF failed, errno = %d\n", errno); + os_close_file(pri->fd); + return(-errno); + } + } + else { + err = os_pipe(fds, 0, 0); + if(err < 0){ + printk("tuntap_open : os_pipe failed - err = %d\n", + -err); + return(err); + } + + buffer = get_output_buffer(&len); + if(buffer != NULL) len--; + used = 0; + + err = tuntap_open_tramp(pri->gate_addr, &pri->fd, fds[0], + fds[1], buffer, len, &used); + + output = buffer; + if(err < 0) { + printk("%s", output); + free_output_buffer(buffer); + printk("tuntap_open_tramp failed - err = %d\n", -err); + return(err); + } + + pri->dev_name = uml_strdup(buffer); + output += IFNAMSIZ; + printk("%s", output); + free_output_buffer(buffer); + + os_close_file(fds[0]); + iter_addresses(pri->dev, open_addr, pri->dev_name); + } + + return(pri->fd); +} + +static void tuntap_close(int fd, void *data) +{ + struct tuntap_data *pri = data; + + if(!pri->fixed_config) + iter_addresses(pri->dev, close_addr, pri->dev_name); + os_close_file(fd); + pri->fd = -1; +} + +static int tuntap_set_mtu(int mtu, void *data) +{ + return(mtu); +} + +struct net_user_info tuntap_user_info = { + .init = tuntap_user_init, + .open = tuntap_open, + .close = tuntap_close, + .remove = NULL, + .set_mtu = tuntap_set_mtu, + .add_address = tuntap_add_addr, + .delete_address = tuntap_del_addr, + .max_packet = MAX_PACKET +}; + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/elf_aux.c b/arch/um/os-Linux/elf_aux.c new file mode 100644 index 000000000000..9aee0b62ebca --- /dev/null +++ b/arch/um/os-Linux/elf_aux.c @@ -0,0 +1,66 @@ +/* + * arch/um/kernel/elf_aux.c + * + * Scan the Elf auxiliary vector provided by the host to extract + * information about vsyscall-page, etc. + * + * Copyright (C) 2004 Fujitsu Siemens Computers GmbH + * Author: Bodo Stroesser (bodo.stroesser@fujitsu-siemens.com) + */ +#include <elf.h> +#include <stddef.h> +#include "init.h" +#include "elf_user.h" + +#if ELF_CLASS == ELFCLASS32 +typedef Elf32_auxv_t elf_auxv_t; +#else +typedef Elf64_auxv_t elf_auxv_t; +#endif + +char * elf_aux_platform; +long elf_aux_hwcap; + +unsigned long vsyscall_ehdr; +unsigned long vsyscall_end; + +unsigned long __kernel_vsyscall; + +__init void scan_elf_aux( char **envp) +{ + long page_size = 0; + elf_auxv_t * auxv; + + while ( *envp++ != NULL) ; + + for ( auxv = (elf_auxv_t *)envp; auxv->a_type != AT_NULL; auxv++) { + switch ( auxv->a_type ) { + case AT_SYSINFO: + __kernel_vsyscall = auxv->a_un.a_val; + break; + case AT_SYSINFO_EHDR: + vsyscall_ehdr = auxv->a_un.a_val; + break; + case AT_HWCAP: + elf_aux_hwcap = auxv->a_un.a_val; + break; + case AT_PLATFORM: + elf_aux_platform = auxv->a_un.a_ptr; + break; + case AT_PAGESZ: + page_size = auxv->a_un.a_val; + break; + } + } + if ( ! __kernel_vsyscall || ! vsyscall_ehdr || + ! elf_aux_hwcap || ! elf_aux_platform || + ! page_size || (vsyscall_ehdr % page_size) ) { + __kernel_vsyscall = 0; + vsyscall_ehdr = 0; + elf_aux_hwcap = 0; + elf_aux_platform = "i586"; + } + else { + vsyscall_end = vsyscall_ehdr + page_size; + } +} diff --git a/arch/um/os-Linux/file.c b/arch/um/os-Linux/file.c new file mode 100644 index 000000000000..77d4066d1af8 --- /dev/null +++ b/arch/um/os-Linux/file.c @@ -0,0 +1,680 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <unistd.h> +#include <errno.h> +#include <fcntl.h> +#include <signal.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <sys/socket.h> +#include <sys/un.h> +#include <sys/ioctl.h> +#include <sys/mount.h> +#include <sys/uio.h> +#include "os.h" +#include "user.h" +#include "kern_util.h" + +static void copy_stat(struct uml_stat *dst, struct stat64 *src) +{ + *dst = ((struct uml_stat) { + .ust_dev = src->st_dev, /* device */ + .ust_ino = src->st_ino, /* inode */ + .ust_mode = src->st_mode, /* protection */ + .ust_nlink = src->st_nlink, /* number of hard links */ + .ust_uid = src->st_uid, /* user ID of owner */ + .ust_gid = src->st_gid, /* group ID of owner */ + .ust_size = src->st_size, /* total size, in bytes */ + .ust_blksize = src->st_blksize, /* blocksize for filesys I/O */ + .ust_blocks = src->st_blocks, /* number of blocks allocated */ + .ust_atime = src->st_atime, /* time of last access */ + .ust_mtime = src->st_mtime, /* time of last modification */ + .ust_ctime = src->st_ctime, /* time of last change */ + }); +} + +int os_stat_fd(const int fd, struct uml_stat *ubuf) +{ + struct stat64 sbuf; + int err; + + do { + err = fstat64(fd, &sbuf); + } while((err < 0) && (errno == EINTR)) ; + + if(err < 0) + return(-errno); + + if(ubuf != NULL) + copy_stat(ubuf, &sbuf); + return(err); +} + +int os_stat_file(const char *file_name, struct uml_stat *ubuf) +{ + struct stat64 sbuf; + int err; + + do { + err = stat64(file_name, &sbuf); + } while((err < 0) && (errno == EINTR)) ; + + if(err < 0) + return(-errno); + + if(ubuf != NULL) + copy_stat(ubuf, &sbuf); + return(err); +} + +int os_access(const char* file, int mode) +{ + int amode, err; + + amode=(mode&OS_ACC_R_OK ? R_OK : 0) | (mode&OS_ACC_W_OK ? W_OK : 0) | + (mode&OS_ACC_X_OK ? X_OK : 0) | (mode&OS_ACC_F_OK ? F_OK : 0) ; + + err = access(file, amode); + if(err < 0) + return(-errno); + + return(0); +} + +void os_print_error(int error, const char* str) +{ + errno = error < 0 ? -error : error; + + perror(str); +} + +/* FIXME? required only by hostaudio (because it passes ioctls verbatim) */ +int os_ioctl_generic(int fd, unsigned int cmd, unsigned long arg) +{ + int err; + + err = ioctl(fd, cmd, arg); + if(err < 0) + return(-errno); + + return(err); +} + +int os_window_size(int fd, int *rows, int *cols) +{ + struct winsize size; + + if(ioctl(fd, TIOCGWINSZ, &size) < 0) + return(-errno); + + *rows = size.ws_row; + *cols = size.ws_col; + + return(0); +} + +int os_new_tty_pgrp(int fd, int pid) +{ + if(ioctl(fd, TIOCSCTTY, 0) < 0){ + printk("TIOCSCTTY failed, errno = %d\n", errno); + return(-errno); + } + + if(tcsetpgrp(fd, pid) < 0){ + printk("tcsetpgrp failed, errno = %d\n", errno); + return(-errno); + } + + return(0); +} + +/* FIXME: ensure namebuf in os_get_if_name is big enough */ +int os_get_ifname(int fd, char* namebuf) +{ + if(ioctl(fd, SIOCGIFNAME, namebuf) < 0) + return(-errno); + + return(0); +} + +int os_set_slip(int fd) +{ + int disc, sencap; + + disc = N_SLIP; + if(ioctl(fd, TIOCSETD, &disc) < 0){ + printk("Failed to set slip line discipline - " + "errno = %d\n", errno); + return(-errno); + } + + sencap = 0; + if(ioctl(fd, SIOCSIFENCAP, &sencap) < 0){ + printk("Failed to set slip encapsulation - " + "errno = %d\n", errno); + return(-errno); + } + + return(0); +} + +int os_set_owner(int fd, int pid) +{ + if(fcntl(fd, F_SETOWN, pid) < 0){ + int save_errno = errno; + + if(fcntl(fd, F_GETOWN, 0) != pid) + return(-save_errno); + } + + return(0); +} + +/* FIXME? moved wholesale from sigio_user.c to get fcntls out of that file */ +int os_sigio_async(int master, int slave) +{ + int flags; + + flags = fcntl(master, F_GETFL); + if(flags < 0) { + printk("fcntl F_GETFL failed, errno = %d\n", errno); + return(-errno); + } + + if((fcntl(master, F_SETFL, flags | O_NONBLOCK | O_ASYNC) < 0) || + (fcntl(master, F_SETOWN, os_getpid()) < 0)){ + printk("fcntl F_SETFL or F_SETOWN failed, errno = %d\n", + errno); + return(-errno); + } + + if((fcntl(slave, F_SETFL, flags | O_NONBLOCK) < 0)){ + printk("fcntl F_SETFL failed, errno = %d\n", errno); + return(-errno); + } + + return(0); +} + +int os_mode_fd(int fd, int mode) +{ + int err; + + do { + err = fchmod(fd, mode); + } while((err < 0) && (errno==EINTR)) ; + + if(err < 0) + return(-errno); + + return(0); +} + +int os_file_type(char *file) +{ + struct uml_stat buf; + int err; + + err = os_stat_file(file, &buf); + if(err < 0) + return(err); + + if(S_ISDIR(buf.ust_mode)) return(OS_TYPE_DIR); + else if(S_ISLNK(buf.ust_mode)) return(OS_TYPE_SYMLINK); + else if(S_ISCHR(buf.ust_mode)) return(OS_TYPE_CHARDEV); + else if(S_ISBLK(buf.ust_mode)) return(OS_TYPE_BLOCKDEV); + else if(S_ISFIFO(buf.ust_mode)) return(OS_TYPE_FIFO); + else if(S_ISSOCK(buf.ust_mode)) return(OS_TYPE_SOCK); + else return(OS_TYPE_FILE); +} + +int os_file_mode(char *file, struct openflags *mode_out) +{ + int err; + + *mode_out = OPENFLAGS(); + + err = os_access(file, OS_ACC_W_OK); + if((err < 0) && (err != -EACCES)) + return(err); + + *mode_out = of_write(*mode_out); + + err = os_access(file, OS_ACC_R_OK); + if((err < 0) && (err != -EACCES)) + return(err); + + *mode_out = of_read(*mode_out); + + return(0); +} + +int os_open_file(char *file, struct openflags flags, int mode) +{ + int fd, f = 0; + + if(flags.r && flags.w) f = O_RDWR; + else if(flags.r) f = O_RDONLY; + else if(flags.w) f = O_WRONLY; + else f = 0; + + if(flags.s) f |= O_SYNC; + if(flags.c) f |= O_CREAT; + if(flags.t) f |= O_TRUNC; + if(flags.e) f |= O_EXCL; + + fd = open64(file, f, mode); + if(fd < 0) + return(-errno); + + if(flags.cl && fcntl(fd, F_SETFD, 1)){ + os_close_file(fd); + return(-errno); + } + + return(fd); +} + +int os_connect_socket(char *name) +{ + struct sockaddr_un sock; + int fd, err; + + sock.sun_family = AF_UNIX; + snprintf(sock.sun_path, sizeof(sock.sun_path), "%s", name); + + fd = socket(AF_UNIX, SOCK_STREAM, 0); + if(fd < 0) + return(fd); + + err = connect(fd, (struct sockaddr *) &sock, sizeof(sock)); + if(err) + return(-errno); + + return(fd); +} + +void os_close_file(int fd) +{ + close(fd); +} + +int os_seek_file(int fd, __u64 offset) +{ + __u64 actual; + + actual = lseek64(fd, offset, SEEK_SET); + if(actual != offset) + return(-errno); + return(0); +} + +static int fault_buffer(void *start, int len, + int (*copy_proc)(void *addr, void *buf, int len)) +{ + int page = getpagesize(), i; + char c; + + for(i = 0; i < len; i += page){ + if((*copy_proc)(start + i, &c, sizeof(c))) + return(-EFAULT); + } + if((len % page) != 0){ + if((*copy_proc)(start + len - 1, &c, sizeof(c))) + return(-EFAULT); + } + return(0); +} + +static int file_io(int fd, void *buf, int len, + int (*io_proc)(int fd, void *buf, int len), + int (*copy_user_proc)(void *addr, void *buf, int len)) +{ + int n, err; + + do { + n = (*io_proc)(fd, buf, len); + if((n < 0) && (errno == EFAULT)){ + err = fault_buffer(buf, len, copy_user_proc); + if(err) + return(err); + n = (*io_proc)(fd, buf, len); + } + } while((n < 0) && (errno == EINTR)); + + if(n < 0) + return(-errno); + return(n); +} + +int os_read_file(int fd, void *buf, int len) +{ + return(file_io(fd, buf, len, (int (*)(int, void *, int)) read, + copy_from_user_proc)); +} + +int os_write_file(int fd, const void *buf, int len) +{ + return(file_io(fd, (void *) buf, len, + (int (*)(int, void *, int)) write, copy_to_user_proc)); +} + +int os_file_size(char *file, long long *size_out) +{ + struct uml_stat buf; + int err; + + err = os_stat_file(file, &buf); + if(err < 0){ + printk("Couldn't stat \"%s\" : err = %d\n", file, -err); + return(err); + } + + if(S_ISBLK(buf.ust_mode)){ + int fd, blocks; + + fd = os_open_file(file, of_read(OPENFLAGS()), 0); + if(fd < 0){ + printk("Couldn't open \"%s\", errno = %d\n", file, -fd); + return(fd); + } + if(ioctl(fd, BLKGETSIZE, &blocks) < 0){ + printk("Couldn't get the block size of \"%s\", " + "errno = %d\n", file, errno); + err = -errno; + os_close_file(fd); + return(err); + } + *size_out = ((long long) blocks) * 512; + os_close_file(fd); + return(0); + } + *size_out = buf.ust_size; + return(0); +} + +int os_file_modtime(char *file, unsigned long *modtime) +{ + struct uml_stat buf; + int err; + + err = os_stat_file(file, &buf); + if(err < 0){ + printk("Couldn't stat \"%s\" : err = %d\n", file, -err); + return(err); + } + + *modtime = buf.ust_mtime; + return(0); +} + +int os_get_exec_close(int fd, int* close_on_exec) +{ + int ret; + + do { + ret = fcntl(fd, F_GETFD); + } while((ret < 0) && (errno == EINTR)) ; + + if(ret < 0) + return(-errno); + + *close_on_exec = (ret&FD_CLOEXEC) ? 1 : 0; + return(ret); +} + +int os_set_exec_close(int fd, int close_on_exec) +{ + int flag, err; + + if(close_on_exec) flag = FD_CLOEXEC; + else flag = 0; + + do { + err = fcntl(fd, F_SETFD, flag); + } while((err < 0) && (errno == EINTR)) ; + + if(err < 0) + return(-errno); + return(err); +} + +int os_pipe(int *fds, int stream, int close_on_exec) +{ + int err, type = stream ? SOCK_STREAM : SOCK_DGRAM; + + err = socketpair(AF_UNIX, type, 0, fds); + if(err < 0) + return(-errno); + + if(!close_on_exec) + return(0); + + err = os_set_exec_close(fds[0], 1); + if(err < 0) + goto error; + + err = os_set_exec_close(fds[1], 1); + if(err < 0) + goto error; + + return(0); + + error: + printk("os_pipe : Setting FD_CLOEXEC failed, err = %d\n", -err); + os_close_file(fds[1]); + os_close_file(fds[0]); + return(err); +} + +int os_set_fd_async(int fd, int owner) +{ + /* XXX This should do F_GETFL first */ + if(fcntl(fd, F_SETFL, O_ASYNC | O_NONBLOCK) < 0){ + printk("os_set_fd_async : failed to set O_ASYNC and " + "O_NONBLOCK on fd # %d, errno = %d\n", fd, errno); + return(-errno); + } +#ifdef notdef + if(fcntl(fd, F_SETFD, 1) < 0){ + printk("os_set_fd_async : Setting FD_CLOEXEC failed, " + "errno = %d\n", errno); + } +#endif + + if((fcntl(fd, F_SETSIG, SIGIO) < 0) || + (fcntl(fd, F_SETOWN, owner) < 0)){ + printk("os_set_fd_async : Failed to fcntl F_SETOWN " + "(or F_SETSIG) fd %d to pid %d, errno = %d\n", fd, + owner, errno); + return(-errno); + } + + return(0); +} + +int os_clear_fd_async(int fd) +{ + int flags = fcntl(fd, F_GETFL); + + flags &= ~(O_ASYNC | O_NONBLOCK); + if(fcntl(fd, F_SETFL, flags) < 0) + return(-errno); + return(0); +} + +int os_set_fd_block(int fd, int blocking) +{ + int flags; + + flags = fcntl(fd, F_GETFL); + + if(blocking) flags &= ~O_NONBLOCK; + else flags |= O_NONBLOCK; + + if(fcntl(fd, F_SETFL, flags) < 0){ + printk("Failed to change blocking on fd # %d, errno = %d\n", + fd, errno); + return(-errno); + } + return(0); +} + +int os_accept_connection(int fd) +{ + int new; + + new = accept(fd, NULL, 0); + if(new < 0) + return(-errno); + return(new); +} + +#ifndef SHUT_RD +#define SHUT_RD 0 +#endif + +#ifndef SHUT_WR +#define SHUT_WR 1 +#endif + +#ifndef SHUT_RDWR +#define SHUT_RDWR 2 +#endif + +int os_shutdown_socket(int fd, int r, int w) +{ + int what, err; + + if(r && w) what = SHUT_RDWR; + else if(r) what = SHUT_RD; + else if(w) what = SHUT_WR; + else { + printk("os_shutdown_socket : neither r or w was set\n"); + return(-EINVAL); + } + err = shutdown(fd, what); + if(err < 0) + return(-errno); + return(0); +} + +int os_rcv_fd(int fd, int *helper_pid_out) +{ + int new, n; + char buf[CMSG_SPACE(sizeof(new))]; + struct msghdr msg; + struct cmsghdr *cmsg; + struct iovec iov; + + msg.msg_name = NULL; + msg.msg_namelen = 0; + iov = ((struct iovec) { .iov_base = helper_pid_out, + .iov_len = sizeof(*helper_pid_out) }); + msg.msg_iov = &iov; + msg.msg_iovlen = 1; + msg.msg_control = buf; + msg.msg_controllen = sizeof(buf); + msg.msg_flags = 0; + + n = recvmsg(fd, &msg, 0); + if(n < 0) + return(-errno); + + else if(n != sizeof(iov.iov_len)) + *helper_pid_out = -1; + + cmsg = CMSG_FIRSTHDR(&msg); + if(cmsg == NULL){ + printk("rcv_fd didn't receive anything, error = %d\n", errno); + return(-1); + } + if((cmsg->cmsg_level != SOL_SOCKET) || + (cmsg->cmsg_type != SCM_RIGHTS)){ + printk("rcv_fd didn't receive a descriptor\n"); + return(-1); + } + + new = ((int *) CMSG_DATA(cmsg))[0]; + return(new); +} + +int os_create_unix_socket(char *file, int len, int close_on_exec) +{ + struct sockaddr_un addr; + int sock, err; + + sock = socket(PF_UNIX, SOCK_DGRAM, 0); + if (sock < 0){ + printk("create_unix_socket - socket failed, errno = %d\n", + errno); + return(-errno); + } + + if(close_on_exec) { + err = os_set_exec_close(sock, 1); + if(err < 0) + printk("create_unix_socket : close_on_exec failed, " + "err = %d", -err); + } + + addr.sun_family = AF_UNIX; + + /* XXX Be more careful about overflow */ + snprintf(addr.sun_path, len, "%s", file); + + err = bind(sock, (struct sockaddr *) &addr, sizeof(addr)); + if (err < 0){ + printk("create_listening_socket at '%s' - bind failed, " + "errno = %d\n", file, errno); + return(-errno); + } + + return(sock); +} + +void os_flush_stdout(void) +{ + fflush(stdout); +} + +int os_lock_file(int fd, int excl) +{ + int type = excl ? F_WRLCK : F_RDLCK; + struct flock lock = ((struct flock) { .l_type = type, + .l_whence = SEEK_SET, + .l_start = 0, + .l_len = 0 } ); + int err, save; + + err = fcntl(fd, F_SETLK, &lock); + if(!err) + goto out; + + save = -errno; + err = fcntl(fd, F_GETLK, &lock); + if(err){ + err = -errno; + goto out; + } + + printk("F_SETLK failed, file already locked by pid %d\n", lock.l_pid); + err = save; + out: + return(err); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/include/file.h b/arch/um/os-Linux/include/file.h new file mode 100644 index 000000000000..d82711efacfa --- /dev/null +++ b/arch/um/os-Linux/include/file.h @@ -0,0 +1,22 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __OS_FILE_H__ +#define __OS_FILE_H__ + +#define DEV_NULL "/dev/null" + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/process.c b/arch/um/os-Linux/process.c new file mode 100644 index 000000000000..ba9ca1cc790a --- /dev/null +++ b/arch/um/os-Linux/process.c @@ -0,0 +1,171 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include <unistd.h> +#include <stdio.h> +#include <errno.h> +#include <signal.h> +#include <linux/unistd.h> +#include <sys/mman.h> +#include <sys/wait.h> +#include "ptrace_user.h" +#include "os.h" +#include "user.h" +#include "user_util.h" + +#define ARBITRARY_ADDR -1 +#define FAILURE_PID -1 + +#define STAT_PATH_LEN sizeof("/proc/#######/stat\0") +#define COMM_SCANF "%*[^)])" + +unsigned long os_process_pc(int pid) +{ + char proc_stat[STAT_PATH_LEN], buf[256]; + unsigned long pc; + int fd, err; + + sprintf(proc_stat, "/proc/%d/stat", pid); + fd = os_open_file(proc_stat, of_read(OPENFLAGS()), 0); + if(fd < 0){ + printk("os_process_pc - couldn't open '%s', err = %d\n", + proc_stat, -fd); + return(ARBITRARY_ADDR); + } + err = os_read_file(fd, buf, sizeof(buf)); + if(err < 0){ + printk("os_process_pc - couldn't read '%s', err = %d\n", + proc_stat, -err); + os_close_file(fd); + return(ARBITRARY_ADDR); + } + os_close_file(fd); + pc = ARBITRARY_ADDR; + if(sscanf(buf, "%*d " COMM_SCANF " %*c %*d %*d %*d %*d %*d %*d %*d " + "%*d %*d %*d %*d %*d %*d %*d %*d %*d %*d %*d %*d %*d %*d " + "%*d %*d %*d %*d %*d %lu", &pc) != 1){ + printk("os_process_pc - couldn't find pc in '%s'\n", buf); + } + return(pc); +} + +int os_process_parent(int pid) +{ + char stat[STAT_PATH_LEN]; + char data[256]; + int parent, n, fd; + + if(pid == -1) return(-1); + + snprintf(stat, sizeof(stat), "/proc/%d/stat", pid); + fd = os_open_file(stat, of_read(OPENFLAGS()), 0); + if(fd < 0){ + printk("Couldn't open '%s', err = %d\n", stat, -fd); + return(FAILURE_PID); + } + + n = os_read_file(fd, data, sizeof(data)); + os_close_file(fd); + + if(n < 0){ + printk("Couldn't read '%s', err = %d\n", stat, -n); + return(FAILURE_PID); + } + + parent = FAILURE_PID; + n = sscanf(data, "%*d " COMM_SCANF " %*c %d", &parent); + if(n != 1) + printk("Failed to scan '%s'\n", data); + + return(parent); +} + +void os_stop_process(int pid) +{ + kill(pid, SIGSTOP); +} + +void os_kill_process(int pid, int reap_child) +{ + kill(pid, SIGKILL); + if(reap_child) + CATCH_EINTR(waitpid(pid, NULL, 0)); + +} + +/* Kill off a ptraced child by all means available. kill it normally first, + * then PTRACE_KILL it, then PTRACE_CONT it in case it's in a run state from + * which it can't exit directly. + */ + +void os_kill_ptraced_process(int pid, int reap_child) +{ + kill(pid, SIGKILL); + ptrace(PTRACE_KILL, pid); + ptrace(PTRACE_CONT, pid); + if(reap_child) + CATCH_EINTR(waitpid(pid, NULL, 0)); +} + +void os_usr1_process(int pid) +{ + kill(pid, SIGUSR1); +} + +/*Don't use the glibc version, which caches the result in TLS. It misses some + * syscalls, and also breaks with clone(), which does not unshare the TLS.*/ +inline _syscall0(pid_t, getpid) + +int os_getpid(void) +{ + return(getpid()); +} + +int os_map_memory(void *virt, int fd, unsigned long long off, unsigned long len, + int r, int w, int x) +{ + void *loc; + int prot; + + prot = (r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) | + (x ? PROT_EXEC : 0); + + loc = mmap64((void *) virt, len, prot, MAP_SHARED | MAP_FIXED, + fd, off); + if(loc == MAP_FAILED) + return(-errno); + return(0); +} + +int os_protect_memory(void *addr, unsigned long len, int r, int w, int x) +{ + int prot = ((r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) | + (x ? PROT_EXEC : 0)); + + if(mprotect(addr, len, prot) < 0) + return(-errno); + return(0); +} + +int os_unmap_memory(void *addr, int len) +{ + int err; + + err = munmap(addr, len); + if(err < 0) + return(-errno); + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/signal.c b/arch/um/os-Linux/signal.c new file mode 100644 index 000000000000..7eac1baf5975 --- /dev/null +++ b/arch/um/os-Linux/signal.c @@ -0,0 +1,48 @@ +/* + * Copyright (C) 2004 PathScale, Inc + * Licensed under the GPL + */ + +#include <signal.h> +#include "time_user.h" +#include "mode.h" +#include "sysdep/signal.h" + +void sig_handler(int sig) +{ + struct sigcontext *sc; + + ARCH_GET_SIGCONTEXT(sc, sig); + CHOOSE_MODE_PROC(sig_handler_common_tt, sig_handler_common_skas, + sig, sc); +} + +extern int timer_irq_inited; + +void alarm_handler(int sig) +{ + struct sigcontext *sc; + + ARCH_GET_SIGCONTEXT(sc, sig); + if(!timer_irq_inited) return; + + if(sig == SIGALRM) + switch_timers(0); + + CHOOSE_MODE_PROC(sig_handler_common_tt, sig_handler_common_skas, + sig, sc); + + if(sig == SIGALRM) + switch_timers(1); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/sys-i386/Makefile b/arch/um/os-Linux/sys-i386/Makefile new file mode 100644 index 000000000000..340ef26f5944 --- /dev/null +++ b/arch/um/os-Linux/sys-i386/Makefile @@ -0,0 +1,10 @@ +# +# Copyright (C) 2000 Jeff Dike (jdike@karaya.com) +# Licensed under the GPL +# + +obj-$(CONFIG_MODE_SKAS) = registers.o + +USER_OBJS := $(obj-y) + +include arch/um/scripts/Makefile.rules diff --git a/arch/um/os-Linux/sys-i386/registers.c b/arch/um/os-Linux/sys-i386/registers.c new file mode 100644 index 000000000000..148645b14480 --- /dev/null +++ b/arch/um/os-Linux/sys-i386/registers.c @@ -0,0 +1,132 @@ +/* + * Copyright (C) 2004 PathScale, Inc + * Licensed under the GPL + */ + +#include <errno.h> +#include <string.h> +#include "sysdep/ptrace_user.h" +#include "sysdep/ptrace.h" +#include "uml-config.h" +#include "skas_ptregs.h" +#include "registers.h" +#include "user.h" + +/* These are set once at boot time and not changed thereafter */ + +static unsigned long exec_regs[HOST_FRAME_SIZE]; +static unsigned long exec_fp_regs[HOST_FP_SIZE]; +static unsigned long exec_fpx_regs[HOST_XFP_SIZE]; +static int have_fpx_regs = 1; + +void init_thread_registers(union uml_pt_regs *to) +{ + memcpy(to->skas.regs, exec_regs, sizeof(to->skas.regs)); + memcpy(to->skas.fp, exec_fp_regs, sizeof(to->skas.fp)); + if(have_fpx_regs) + memcpy(to->skas.xfp, exec_fpx_regs, sizeof(to->skas.xfp)); +} + +/* XXX These need to use [GS]ETFPXREGS and copy_sc_{to,from}_user_skas needs + * to pass in a sufficiently large buffer + */ +int save_fp_registers(int pid, unsigned long *fp_regs) +{ + if(ptrace(PTRACE_GETFPREGS, pid, 0, fp_regs) < 0) + return(-errno); + return(0); +} + +int restore_fp_registers(int pid, unsigned long *fp_regs) +{ + if(ptrace(PTRACE_SETFPREGS, pid, 0, fp_regs) < 0) + return(-errno); + return(0); +} + +static int move_registers(int pid, int int_op, union uml_pt_regs *regs, + int fp_op, unsigned long *fp_regs) +{ + if(ptrace(int_op, pid, 0, regs->skas.regs) < 0) + return(-errno); + + if(ptrace(fp_op, pid, 0, fp_regs) < 0) + return(-errno); + + return(0); +} + +void save_registers(int pid, union uml_pt_regs *regs) +{ + unsigned long *fp_regs; + int err, fp_op; + + if(have_fpx_regs){ + fp_op = PTRACE_GETFPXREGS; + fp_regs = regs->skas.xfp; + } + else { + fp_op = PTRACE_GETFPREGS; + fp_regs = regs->skas.fp; + } + + err = move_registers(pid, PTRACE_GETREGS, regs, fp_op, fp_regs); + if(err) + panic("save_registers - saving registers failed, errno = %d\n", + -err); +} + +void restore_registers(int pid, union uml_pt_regs *regs) +{ + unsigned long *fp_regs; + int err, fp_op; + + if(have_fpx_regs){ + fp_op = PTRACE_SETFPXREGS; + fp_regs = regs->skas.xfp; + } + else { + fp_op = PTRACE_SETFPREGS; + fp_regs = regs->skas.fp; + } + + err = move_registers(pid, PTRACE_SETREGS, regs, fp_op, fp_regs); + if(err) + panic("restore_registers - saving registers failed, " + "errno = %d\n", -err); +} + +void init_registers(int pid) +{ + int err; + + err = ptrace(PTRACE_GETREGS, pid, 0, exec_regs); + if(err) + panic("check_ptrace : PTRACE_GETREGS failed, errno = %d", + err); + + err = ptrace(PTRACE_GETFPXREGS, pid, 0, exec_fpx_regs); + if(!err) + return; + + have_fpx_regs = 0; + if(err != EIO) + panic("check_ptrace : PTRACE_GETFPXREGS failed, errno = %d", + err); + + err = ptrace(PTRACE_GETFPREGS, pid, 0, exec_fp_regs); + if(err) + panic("check_ptrace : PTRACE_GETFPREGS failed, errno = %d", + err); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/sys-x86_64/Makefile b/arch/um/os-Linux/sys-x86_64/Makefile new file mode 100644 index 000000000000..340ef26f5944 --- /dev/null +++ b/arch/um/os-Linux/sys-x86_64/Makefile @@ -0,0 +1,10 @@ +# +# Copyright (C) 2000 Jeff Dike (jdike@karaya.com) +# Licensed under the GPL +# + +obj-$(CONFIG_MODE_SKAS) = registers.o + +USER_OBJS := $(obj-y) + +include arch/um/scripts/Makefile.rules diff --git a/arch/um/os-Linux/sys-x86_64/registers.c b/arch/um/os-Linux/sys-x86_64/registers.c new file mode 100644 index 000000000000..6286c974bbeb --- /dev/null +++ b/arch/um/os-Linux/sys-x86_64/registers.c @@ -0,0 +1,81 @@ +/* + * Copyright (C) 2004 PathScale, Inc + * Licensed under the GPL + */ + +#include <errno.h> +#include <string.h> +#include "ptrace_user.h" +#include "uml-config.h" +#include "skas_ptregs.h" +#include "registers.h" +#include "user.h" + +/* These are set once at boot time and not changed thereafter */ + +static unsigned long exec_regs[HOST_FRAME_SIZE]; +static unsigned long exec_fp_regs[HOST_FP_SIZE]; + +void init_thread_registers(union uml_pt_regs *to) +{ + memcpy(to->skas.regs, exec_regs, sizeof(to->skas.regs)); + memcpy(to->skas.fp, exec_fp_regs, sizeof(to->skas.fp)); +} + +static int move_registers(int pid, int int_op, int fp_op, + union uml_pt_regs *regs) +{ + if(ptrace(int_op, pid, 0, regs->skas.regs) < 0) + return(-errno); + + if(ptrace(fp_op, pid, 0, regs->skas.fp) < 0) + return(-errno); + + return(0); +} + +void save_registers(int pid, union uml_pt_regs *regs) +{ + int err; + + err = move_registers(pid, PTRACE_GETREGS, PTRACE_GETFPREGS, regs); + if(err) + panic("save_registers - saving registers failed, errno = %d\n", + -err); +} + +void restore_registers(int pid, union uml_pt_regs *regs) +{ + int err; + + err = move_registers(pid, PTRACE_SETREGS, PTRACE_SETFPREGS, regs); + if(err) + panic("restore_registers - saving registers failed, " + "errno = %d\n", -err); +} + +void init_registers(int pid) +{ + int err; + + err = ptrace(PTRACE_GETREGS, pid, 0, exec_regs); + if(err) + panic("check_ptrace : PTRACE_GETREGS failed, errno = %d", + err); + + err = ptrace(PTRACE_GETFPREGS, pid, 0, exec_fp_regs); + if(err) + panic("check_ptrace : PTRACE_GETFPREGS failed, errno = %d", + err); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/time.c b/arch/um/os-Linux/time.c new file mode 100644 index 000000000000..cf30a39bc484 --- /dev/null +++ b/arch/um/os-Linux/time.c @@ -0,0 +1,21 @@ +#include <stdlib.h> +#include <sys/time.h> + +unsigned long long os_usecs(void) +{ + struct timeval tv; + + gettimeofday(&tv, NULL); + return((unsigned long long) tv.tv_sec * 1000000 + tv.tv_usec); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/tty.c b/arch/um/os-Linux/tty.c new file mode 100644 index 000000000000..4cfdd18ea1ef --- /dev/null +++ b/arch/um/os-Linux/tty.c @@ -0,0 +1,61 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <errno.h> +#include "os.h" +#include "user.h" +#include "kern_util.h" + +struct grantpt_info { + int fd; + int res; + int err; +}; + +static void grantpt_cb(void *arg) +{ + struct grantpt_info *info = arg; + + info->res = grantpt(info->fd); + info->err = errno; +} + +int get_pty(void) +{ + struct grantpt_info info; + int fd; + + fd = os_open_file("/dev/ptmx", of_rdwr(OPENFLAGS()), 0); + if(fd < 0){ + printk("get_pty : Couldn't open /dev/ptmx - err = %d\n", -fd); + return(fd); + } + + info.fd = fd; + initial_thread_cb(grantpt_cb, &info); + + if(info.res < 0){ + printk("get_pty : Couldn't grant pty - errno = %d\n", + -info.err); + return(-1); + } + if(unlockpt(fd) < 0){ + printk("get_pty : Couldn't unlock pty - errno = %d\n", errno); + return(-1); + } + return(fd); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/user_syms.c b/arch/um/os-Linux/user_syms.c new file mode 100644 index 000000000000..75d7af9ae1d2 --- /dev/null +++ b/arch/um/os-Linux/user_syms.c @@ -0,0 +1,95 @@ +#include "linux/types.h" +#include "linux/module.h" + +/* Some of this are builtin function (some are not but could in the future), + * so I *must* declare good prototypes for them and then EXPORT them. + * The kernel code uses the macro defined by include/linux/string.h, + * so I undef macros; the userspace code does not include that and I + * add an EXPORT for the glibc one.*/ + +#undef strlen +#undef strstr +#undef memcpy +#undef memset + +extern size_t strlen(const char *); +extern void *memcpy(void *, const void *, size_t); +extern void *memmove(void *, const void *, size_t); +extern void *memset(void *, int, size_t); +extern int printf(const char *, ...); + +EXPORT_SYMBOL(strlen); +EXPORT_SYMBOL(memcpy); +EXPORT_SYMBOL(memmove); +EXPORT_SYMBOL(memset); +EXPORT_SYMBOL(printf); + +EXPORT_SYMBOL(strstr); + +/* Here, instead, I can provide a fake prototype. Yes, someone cares: genksyms. + * However, the modules will use the CRC defined *here*, no matter if it is + * good; so the versions of these symbols will always match + */ +#define EXPORT_SYMBOL_PROTO(sym) \ + int sym(void); \ + EXPORT_SYMBOL(sym); + +#ifdef SUBARCH_i386 +EXPORT_SYMBOL(vsyscall_ehdr); +EXPORT_SYMBOL(vsyscall_end); +#endif + +EXPORT_SYMBOL_PROTO(__errno_location); + +EXPORT_SYMBOL_PROTO(access); +EXPORT_SYMBOL_PROTO(open); +EXPORT_SYMBOL_PROTO(open64); +EXPORT_SYMBOL_PROTO(close); +EXPORT_SYMBOL_PROTO(read); +EXPORT_SYMBOL_PROTO(write); +EXPORT_SYMBOL_PROTO(dup2); +EXPORT_SYMBOL_PROTO(__xstat); +EXPORT_SYMBOL_PROTO(__lxstat); +EXPORT_SYMBOL_PROTO(__lxstat64); +EXPORT_SYMBOL_PROTO(lseek); +EXPORT_SYMBOL_PROTO(lseek64); +EXPORT_SYMBOL_PROTO(chown); +EXPORT_SYMBOL_PROTO(truncate); +EXPORT_SYMBOL_PROTO(utime); +EXPORT_SYMBOL_PROTO(chmod); +EXPORT_SYMBOL_PROTO(rename); +EXPORT_SYMBOL_PROTO(__xmknod); + +EXPORT_SYMBOL_PROTO(symlink); +EXPORT_SYMBOL_PROTO(link); +EXPORT_SYMBOL_PROTO(unlink); +EXPORT_SYMBOL_PROTO(readlink); + +EXPORT_SYMBOL_PROTO(mkdir); +EXPORT_SYMBOL_PROTO(rmdir); +EXPORT_SYMBOL_PROTO(opendir); +EXPORT_SYMBOL_PROTO(readdir); +EXPORT_SYMBOL_PROTO(closedir); +EXPORT_SYMBOL_PROTO(seekdir); +EXPORT_SYMBOL_PROTO(telldir); + +EXPORT_SYMBOL_PROTO(ioctl); + +EXPORT_SYMBOL_PROTO(pread64); +EXPORT_SYMBOL_PROTO(pwrite64); + +EXPORT_SYMBOL_PROTO(statfs); +EXPORT_SYMBOL_PROTO(statfs64); + +EXPORT_SYMBOL_PROTO(getuid); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/os-Linux/util/Makefile b/arch/um/os-Linux/util/Makefile new file mode 100644 index 000000000000..fb00ddf969bd --- /dev/null +++ b/arch/um/os-Linux/util/Makefile @@ -0,0 +1,4 @@ +hostprogs-y := mk_user_constants +always := $(hostprogs-y) + +mk_user_constants-objs := mk_user_constants.o diff --git a/arch/um/os-Linux/util/mk_user_constants.c b/arch/um/os-Linux/util/mk_user_constants.c new file mode 100644 index 000000000000..0933518aa8bd --- /dev/null +++ b/arch/um/os-Linux/util/mk_user_constants.c @@ -0,0 +1,29 @@ +#include <stdio.h> +#include <asm/types.h> +/* For some reason, x86_64 nowhere defines u64 and u32, even though they're + * used throughout the headers. + */ +typedef __u64 u64; +typedef __u32 u32; +#include <asm/user.h> + +int main(int argc, char **argv) +{ + printf("/*\n"); + printf(" * Generated by mk_user_constants\n"); + printf(" */\n"); + printf("\n"); + printf("#ifndef __UM_USER_CONSTANTS_H\n"); + printf("#define __UM_USER_CONSTANTS_H\n"); + printf("\n"); + /* I'd like to use FRAME_SIZE from ptrace.h here, but that's wrong on + * x86_64 (216 vs 168 bytes). user_regs_struct is the correct size on + * both x86_64 and i386. + */ + printf("#define UM_FRAME_SIZE %d\n", (int) sizeof(struct user_regs_struct)); + + printf("\n"); + printf("#endif\n"); + + return(0); +} diff --git a/arch/um/scripts/Makefile.rules b/arch/um/scripts/Makefile.rules new file mode 100644 index 000000000000..143f6fea0763 --- /dev/null +++ b/arch/um/scripts/Makefile.rules @@ -0,0 +1,13 @@ +# =========================================================================== +# arch/um: Generic definitions +# =========================================================================== + +USER_SINGLE_OBJS = $(foreach f,$(patsubst %.o,%,$(obj-y) $(obj-m)),$($(f)-objs)) +USER_OBJS += $(filter %_user.o,$(obj-y) $(obj-m) $(USER_SINGLE_OBJS)) + +USER_OBJS := $(foreach file,$(USER_OBJS),$(obj)/$(file)) + +$(USER_OBJS): c_flags = -Wp,-MD,$(depfile) $(USER_CFLAGS) $(CFLAGS_$(notdir $@)) + +quiet_cmd_make_link = SYMLINK $@ +cmd_make_link = rm -f $@; ln -sf $(srctree)/arch/$(SUBARCH)/$($(notdir $@)-dir)/$(notdir $@) $@ diff --git a/arch/um/sys-i386/Makefile b/arch/um/sys-i386/Makefile new file mode 100644 index 000000000000..71b47e618605 --- /dev/null +++ b/arch/um/sys-i386/Makefile @@ -0,0 +1,29 @@ +obj-y = bitops.o bugs.o checksum.o delay.o fault.o ksyms.o ldt.o ptrace.o \ + ptrace_user.o semaphore.o signal.o sigcontext.o syscalls.o sysrq.o + +obj-$(CONFIG_HIGHMEM) += highmem.o +obj-$(CONFIG_MODULES) += module.o + +USER_OBJS := bugs.o ptrace_user.o sigcontext.o fault.o + +include arch/um/scripts/Makefile.rules + +SYMLINKS = bitops.c semaphore.c highmem.c module.c + +# this needs to be before the foreach, because clean-files does not accept +# complete paths like $(src)/$f. +clean-files := $(SYMLINKS) + +targets += $(SYMLINKS) + +SYMLINKS := $(foreach f,$(SYMLINKS),$(obj)/$f) + +bitops.c-dir = lib +semaphore.c-dir = kernel +highmem.c-dir = mm +module.c-dir = kernel + +$(SYMLINKS): FORCE + $(call if_changed,make_link) + +subdir- := util diff --git a/arch/um/sys-i386/bugs.c b/arch/um/sys-i386/bugs.c new file mode 100644 index 000000000000..41b0ab2fe830 --- /dev/null +++ b/arch/um/sys-i386/bugs.c @@ -0,0 +1,222 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <unistd.h> +#include <errno.h> +#include <string.h> +#include <sys/signal.h> +#include <asm/ldt.h> +#include "kern_util.h" +#include "user.h" +#include "sysdep/ptrace.h" +#include "task.h" +#include "os.h" + +#define MAXTOKEN 64 + +/* Set during early boot */ +int host_has_cmov = 1; +int host_has_xmm = 0; + +static char token(int fd, char *buf, int len, char stop) +{ + int n; + char *ptr, *end, c; + + ptr = buf; + end = &buf[len]; + do { + n = os_read_file(fd, ptr, sizeof(*ptr)); + c = *ptr++; + if(n != sizeof(*ptr)){ + if(n == 0) return(0); + printk("Reading /proc/cpuinfo failed, err = %d\n", -n); + if(n < 0) + return(n); + else + return(-EIO); + } + } while((c != '\n') && (c != stop) && (ptr < end)); + + if(ptr == end){ + printk("Failed to find '%c' in /proc/cpuinfo\n", stop); + return(-1); + } + *(ptr - 1) = '\0'; + return(c); +} + +static int find_cpuinfo_line(int fd, char *key, char *scratch, int len) +{ + int n; + char c; + + scratch[len - 1] = '\0'; + while(1){ + c = token(fd, scratch, len - 1, ':'); + if(c <= 0) + return(0); + else if(c != ':'){ + printk("Failed to find ':' in /proc/cpuinfo\n"); + return(0); + } + + if(!strncmp(scratch, key, strlen(key))) + return(1); + + do { + n = os_read_file(fd, &c, sizeof(c)); + if(n != sizeof(c)){ + printk("Failed to find newline in " + "/proc/cpuinfo, err = %d\n", -n); + return(0); + } + } while(c != '\n'); + } + return(0); +} + +int cpu_feature(char *what, char *buf, int len) +{ + int fd, ret = 0; + + fd = os_open_file("/proc/cpuinfo", of_read(OPENFLAGS()), 0); + if(fd < 0){ + printk("Couldn't open /proc/cpuinfo, err = %d\n", -fd); + return(0); + } + + if(!find_cpuinfo_line(fd, what, buf, len)){ + printk("Couldn't find '%s' line in /proc/cpuinfo\n", what); + goto out_close; + } + + token(fd, buf, len, '\n'); + ret = 1; + + out_close: + os_close_file(fd); + return(ret); +} + +static int check_cpu_flag(char *feature, int *have_it) +{ + char buf[MAXTOKEN], c; + int fd, len = sizeof(buf)/sizeof(buf[0]); + + printk("Checking for host processor %s support...", feature); + fd = os_open_file("/proc/cpuinfo", of_read(OPENFLAGS()), 0); + if(fd < 0){ + printk("Couldn't open /proc/cpuinfo, err = %d\n", -fd); + return(0); + } + + *have_it = 0; + if(!find_cpuinfo_line(fd, "flags", buf, sizeof(buf) / sizeof(buf[0]))) + goto out; + + c = token(fd, buf, len - 1, ' '); + if(c < 0) goto out; + else if(c != ' '){ + printk("Failed to find ' ' in /proc/cpuinfo\n"); + goto out; + } + + while(1){ + c = token(fd, buf, len - 1, ' '); + if(c < 0) goto out; + else if(c == '\n') break; + + if(!strcmp(buf, feature)){ + *have_it = 1; + goto out; + } + } + out: + if(*have_it == 0) printk("No\n"); + else if(*have_it == 1) printk("Yes\n"); + os_close_file(fd); + return(1); +} + +#if 0 /* This doesn't work in tt mode, plus it's causing compilation problems + * for some people. + */ +static void disable_lcall(void) +{ + struct modify_ldt_ldt_s ldt; + int err; + + bzero(&ldt, sizeof(ldt)); + ldt.entry_number = 7; + ldt.base_addr = 0; + ldt.limit = 0; + err = modify_ldt(1, &ldt, sizeof(ldt)); + if(err) + printk("Failed to disable lcall7 - errno = %d\n", errno); +} +#endif + +void arch_init_thread(void) +{ +#if 0 + disable_lcall(); +#endif +} + +void arch_check_bugs(void) +{ + int have_it; + + if(os_access("/proc/cpuinfo", OS_ACC_R_OK) < 0){ + printk("/proc/cpuinfo not available - skipping CPU capability " + "checks\n"); + return; + } + if(check_cpu_flag("cmov", &have_it)) + host_has_cmov = have_it; + if(check_cpu_flag("xmm", &have_it)) + host_has_xmm = have_it; +} + +int arch_handle_signal(int sig, union uml_pt_regs *regs) +{ + unsigned char tmp[2]; + + /* This is testing for a cmov (0x0f 0x4x) instruction causing a + * SIGILL in init. + */ + if((sig != SIGILL) || (TASK_PID(get_current()) != 1)) return(0); + + if (copy_from_user_proc(tmp, (void *) UPT_IP(regs), 2)) + panic("SIGILL in init, could not read instructions!\n"); + if((tmp[0] != 0x0f) || ((tmp[1] & 0xf0) != 0x40)) + return(0); + + if(host_has_cmov == 0) + panic("SIGILL caused by cmov, which this processor doesn't " + "implement, boot a filesystem compiled for older " + "processors"); + else if(host_has_cmov == 1) + panic("SIGILL caused by cmov, which this processor claims to " + "implement"); + else if(host_has_cmov == -1) + panic("SIGILL caused by cmov, couldn't tell if this processor " + "implements it, boot a filesystem compiled for older " + "processors"); + else panic("Bad value for host_has_cmov (%d)", host_has_cmov); + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-i386/checksum.S b/arch/um/sys-i386/checksum.S new file mode 100644 index 000000000000..a11171fb6223 --- /dev/null +++ b/arch/um/sys-i386/checksum.S @@ -0,0 +1,460 @@ +/* + * INET An implementation of the TCP/IP protocol suite for the LINUX + * operating system. INET is implemented using the BSD Socket + * interface as the means of communication with the user level. + * + * IP/TCP/UDP checksumming routines + * + * Authors: Jorge Cwik, <jorge@laser.satlink.net> + * Arnt Gulbrandsen, <agulbra@nvg.unit.no> + * Tom May, <ftom@netcom.com> + * Pentium Pro/II routines: + * Alexander Kjeldaas <astor@guardian.no> + * Finn Arne Gangstad <finnag@guardian.no> + * Lots of code moved from tcp.c and ip.c; see those files + * for more names. + * + * Changes: Ingo Molnar, converted csum_partial_copy() to 2.1 exception + * handling. + * Andi Kleen, add zeroing on error + * converted to pure assembler + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include <linux/config.h> +#include <asm/errno.h> + +/* + * computes a partial checksum, e.g. for TCP/UDP fragments + */ + +/* +unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum) + */ + +.text +.align 4 +.globl arch_csum_partial + +#ifndef CONFIG_X86_USE_PPRO_CHECKSUM + + /* + * Experiments with Ethernet and SLIP connections show that buff + * is aligned on either a 2-byte or 4-byte boundary. We get at + * least a twofold speedup on 486 and Pentium if it is 4-byte aligned. + * Fortunately, it is easy to convert 2-byte alignment to 4-byte + * alignment for the unrolled loop. + */ +arch_csum_partial: + pushl %esi + pushl %ebx + movl 20(%esp),%eax # Function arg: unsigned int sum + movl 16(%esp),%ecx # Function arg: int len + movl 12(%esp),%esi # Function arg: unsigned char *buff + testl $2, %esi # Check alignment. + jz 2f # Jump if alignment is ok. + subl $2, %ecx # Alignment uses up two bytes. + jae 1f # Jump if we had at least two bytes. + addl $2, %ecx # ecx was < 2. Deal with it. + jmp 4f +1: movw (%esi), %bx + addl $2, %esi + addw %bx, %ax + adcl $0, %eax +2: + movl %ecx, %edx + shrl $5, %ecx + jz 2f + testl %esi, %esi +1: movl (%esi), %ebx + adcl %ebx, %eax + movl 4(%esi), %ebx + adcl %ebx, %eax + movl 8(%esi), %ebx + adcl %ebx, %eax + movl 12(%esi), %ebx + adcl %ebx, %eax + movl 16(%esi), %ebx + adcl %ebx, %eax + movl 20(%esi), %ebx + adcl %ebx, %eax + movl 24(%esi), %ebx + adcl %ebx, %eax + movl 28(%esi), %ebx + adcl %ebx, %eax + lea 32(%esi), %esi + dec %ecx + jne 1b + adcl $0, %eax +2: movl %edx, %ecx + andl $0x1c, %edx + je 4f + shrl $2, %edx # This clears CF +3: adcl (%esi), %eax + lea 4(%esi), %esi + dec %edx + jne 3b + adcl $0, %eax +4: andl $3, %ecx + jz 7f + cmpl $2, %ecx + jb 5f + movw (%esi),%cx + leal 2(%esi),%esi + je 6f + shll $16,%ecx +5: movb (%esi),%cl +6: addl %ecx,%eax + adcl $0, %eax +7: + popl %ebx + popl %esi + ret + +#else + +/* Version for PentiumII/PPro */ + +arch_csum_partial: + pushl %esi + pushl %ebx + movl 20(%esp),%eax # Function arg: unsigned int sum + movl 16(%esp),%ecx # Function arg: int len + movl 12(%esp),%esi # Function arg: const unsigned char *buf + + testl $2, %esi + jnz 30f +10: + movl %ecx, %edx + movl %ecx, %ebx + andl $0x7c, %ebx + shrl $7, %ecx + addl %ebx,%esi + shrl $2, %ebx + negl %ebx + lea 45f(%ebx,%ebx,2), %ebx + testl %esi, %esi + jmp *%ebx + + # Handle 2-byte-aligned regions +20: addw (%esi), %ax + lea 2(%esi), %esi + adcl $0, %eax + jmp 10b + +30: subl $2, %ecx + ja 20b + je 32f + movzbl (%esi),%ebx # csumming 1 byte, 2-aligned + addl %ebx, %eax + adcl $0, %eax + jmp 80f +32: + addw (%esi), %ax # csumming 2 bytes, 2-aligned + adcl $0, %eax + jmp 80f + +40: + addl -128(%esi), %eax + adcl -124(%esi), %eax + adcl -120(%esi), %eax + adcl -116(%esi), %eax + adcl -112(%esi), %eax + adcl -108(%esi), %eax + adcl -104(%esi), %eax + adcl -100(%esi), %eax + adcl -96(%esi), %eax + adcl -92(%esi), %eax + adcl -88(%esi), %eax + adcl -84(%esi), %eax + adcl -80(%esi), %eax + adcl -76(%esi), %eax + adcl -72(%esi), %eax + adcl -68(%esi), %eax + adcl -64(%esi), %eax + adcl -60(%esi), %eax + adcl -56(%esi), %eax + adcl -52(%esi), %eax + adcl -48(%esi), %eax + adcl -44(%esi), %eax + adcl -40(%esi), %eax + adcl -36(%esi), %eax + adcl -32(%esi), %eax + adcl -28(%esi), %eax + adcl -24(%esi), %eax + adcl -20(%esi), %eax + adcl -16(%esi), %eax + adcl -12(%esi), %eax + adcl -8(%esi), %eax + adcl -4(%esi), %eax +45: + lea 128(%esi), %esi + adcl $0, %eax + dec %ecx + jge 40b + movl %edx, %ecx +50: andl $3, %ecx + jz 80f + + # Handle the last 1-3 bytes without jumping + notl %ecx # 1->2, 2->1, 3->0, higher bits are masked + movl $0xffffff,%ebx # by the shll and shrl instructions + shll $3,%ecx + shrl %cl,%ebx + andl -128(%esi),%ebx # esi is 4-aligned so should be ok + addl %ebx,%eax + adcl $0,%eax +80: + popl %ebx + popl %esi + ret + +#endif + +/* +unsigned int csum_partial_copy_generic (const char *src, char *dst, + int len, int sum, int *src_err_ptr, int *dst_err_ptr) + */ + +/* + * Copy from ds while checksumming, otherwise like csum_partial + * + * The macros SRC and DST specify the type of access for the instruction. + * thus we can call a custom exception handler for all access types. + * + * FIXME: could someone double-check whether I haven't mixed up some SRC and + * DST definitions? It's damn hard to trigger all cases. I hope I got + * them all but there's no guarantee. + */ + +#define SRC(y...) \ + 9999: y; \ + .section __ex_table, "a"; \ + .long 9999b, 6001f ; \ + .previous + +#define DST(y...) \ + 9999: y; \ + .section __ex_table, "a"; \ + .long 9999b, 6002f ; \ + .previous + +.align 4 +.globl csum_partial_copy_generic_i386 + +#ifndef CONFIG_X86_USE_PPRO_CHECKSUM + +#define ARGBASE 16 +#define FP 12 + +csum_partial_copy_generic_i386: + subl $4,%esp + pushl %edi + pushl %esi + pushl %ebx + movl ARGBASE+16(%esp),%eax # sum + movl ARGBASE+12(%esp),%ecx # len + movl ARGBASE+4(%esp),%esi # src + movl ARGBASE+8(%esp),%edi # dst + + testl $2, %edi # Check alignment. + jz 2f # Jump if alignment is ok. + subl $2, %ecx # Alignment uses up two bytes. + jae 1f # Jump if we had at least two bytes. + addl $2, %ecx # ecx was < 2. Deal with it. + jmp 4f +SRC(1: movw (%esi), %bx ) + addl $2, %esi +DST( movw %bx, (%edi) ) + addl $2, %edi + addw %bx, %ax + adcl $0, %eax +2: + movl %ecx, FP(%esp) + shrl $5, %ecx + jz 2f + testl %esi, %esi +SRC(1: movl (%esi), %ebx ) +SRC( movl 4(%esi), %edx ) + adcl %ebx, %eax +DST( movl %ebx, (%edi) ) + adcl %edx, %eax +DST( movl %edx, 4(%edi) ) + +SRC( movl 8(%esi), %ebx ) +SRC( movl 12(%esi), %edx ) + adcl %ebx, %eax +DST( movl %ebx, 8(%edi) ) + adcl %edx, %eax +DST( movl %edx, 12(%edi) ) + +SRC( movl 16(%esi), %ebx ) +SRC( movl 20(%esi), %edx ) + adcl %ebx, %eax +DST( movl %ebx, 16(%edi) ) + adcl %edx, %eax +DST( movl %edx, 20(%edi) ) + +SRC( movl 24(%esi), %ebx ) +SRC( movl 28(%esi), %edx ) + adcl %ebx, %eax +DST( movl %ebx, 24(%edi) ) + adcl %edx, %eax +DST( movl %edx, 28(%edi) ) + + lea 32(%esi), %esi + lea 32(%edi), %edi + dec %ecx + jne 1b + adcl $0, %eax +2: movl FP(%esp), %edx + movl %edx, %ecx + andl $0x1c, %edx + je 4f + shrl $2, %edx # This clears CF +SRC(3: movl (%esi), %ebx ) + adcl %ebx, %eax +DST( movl %ebx, (%edi) ) + lea 4(%esi), %esi + lea 4(%edi), %edi + dec %edx + jne 3b + adcl $0, %eax +4: andl $3, %ecx + jz 7f + cmpl $2, %ecx + jb 5f +SRC( movw (%esi), %cx ) + leal 2(%esi), %esi +DST( movw %cx, (%edi) ) + leal 2(%edi), %edi + je 6f + shll $16,%ecx +SRC(5: movb (%esi), %cl ) +DST( movb %cl, (%edi) ) +6: addl %ecx, %eax + adcl $0, %eax +7: +5000: + +# Exception handler: +.section .fixup, "ax" + +6001: + movl ARGBASE+20(%esp), %ebx # src_err_ptr + movl $-EFAULT, (%ebx) + + # zero the complete destination - computing the rest + # is too much work + movl ARGBASE+8(%esp), %edi # dst + movl ARGBASE+12(%esp), %ecx # len + xorl %eax,%eax + rep ; stosb + + jmp 5000b + +6002: + movl ARGBASE+24(%esp), %ebx # dst_err_ptr + movl $-EFAULT,(%ebx) + jmp 5000b + +.previous + + popl %ebx + popl %esi + popl %edi + popl %ecx # equivalent to addl $4,%esp + ret + +#else + +/* Version for PentiumII/PPro */ + +#define ROUND1(x) \ + SRC(movl x(%esi), %ebx ) ; \ + addl %ebx, %eax ; \ + DST(movl %ebx, x(%edi) ) ; + +#define ROUND(x) \ + SRC(movl x(%esi), %ebx ) ; \ + adcl %ebx, %eax ; \ + DST(movl %ebx, x(%edi) ) ; + +#define ARGBASE 12 + +csum_partial_copy_generic_i386: + pushl %ebx + pushl %edi + pushl %esi + movl ARGBASE+4(%esp),%esi #src + movl ARGBASE+8(%esp),%edi #dst + movl ARGBASE+12(%esp),%ecx #len + movl ARGBASE+16(%esp),%eax #sum +# movl %ecx, %edx + movl %ecx, %ebx + movl %esi, %edx + shrl $6, %ecx + andl $0x3c, %ebx + negl %ebx + subl %ebx, %esi + subl %ebx, %edi + lea -1(%esi),%edx + andl $-32,%edx + lea 3f(%ebx,%ebx), %ebx + testl %esi, %esi + jmp *%ebx +1: addl $64,%esi + addl $64,%edi + SRC(movb -32(%edx),%bl) ; SRC(movb (%edx),%bl) + ROUND1(-64) ROUND(-60) ROUND(-56) ROUND(-52) + ROUND (-48) ROUND(-44) ROUND(-40) ROUND(-36) + ROUND (-32) ROUND(-28) ROUND(-24) ROUND(-20) + ROUND (-16) ROUND(-12) ROUND(-8) ROUND(-4) +3: adcl $0,%eax + addl $64, %edx + dec %ecx + jge 1b +4: movl ARGBASE+12(%esp),%edx #len + andl $3, %edx + jz 7f + cmpl $2, %edx + jb 5f +SRC( movw (%esi), %dx ) + leal 2(%esi), %esi +DST( movw %dx, (%edi) ) + leal 2(%edi), %edi + je 6f + shll $16,%edx +5: +SRC( movb (%esi), %dl ) +DST( movb %dl, (%edi) ) +6: addl %edx, %eax + adcl $0, %eax +7: +.section .fixup, "ax" +6001: movl ARGBASE+20(%esp), %ebx # src_err_ptr + movl $-EFAULT, (%ebx) + # zero the complete destination (computing the rest is too much work) + movl ARGBASE+8(%esp),%edi # dst + movl ARGBASE+12(%esp),%ecx # len + xorl %eax,%eax + rep; stosb + jmp 7b +6002: movl ARGBASE+24(%esp), %ebx # dst_err_ptr + movl $-EFAULT, (%ebx) + jmp 7b +.previous + + popl %esi + popl %edi + popl %ebx + ret + +#undef ROUND +#undef ROUND1 + +#endif diff --git a/arch/um/sys-i386/delay.c b/arch/um/sys-i386/delay.c new file mode 100644 index 000000000000..20d37dbbaf08 --- /dev/null +++ b/arch/um/sys-i386/delay.c @@ -0,0 +1,14 @@ +void __delay(unsigned long time) +{ + /* Stolen from the i386 __loop_delay */ + int d0; + __asm__ __volatile__( + "\tjmp 1f\n" + ".align 16\n" + "1:\tjmp 2f\n" + ".align 16\n" + "2:\tdecl %0\n\tjns 2b" + :"=&a" (d0) + :"0" (time)); +} + diff --git a/arch/um/sys-i386/fault.c b/arch/um/sys-i386/fault.c new file mode 100644 index 000000000000..d0bbcdfdb53f --- /dev/null +++ b/arch/um/sys-i386/fault.c @@ -0,0 +1,42 @@ +/* + * Copyright (C) 2002 - 2004 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include <signal.h> +#include "sysdep/ptrace.h" +#include "sysdep/sigcontext.h" + +/* These two are from asm-um/uaccess.h and linux/module.h, check them. */ +struct exception_table_entry +{ + unsigned long insn; + unsigned long fixup; +}; + +const struct exception_table_entry *search_exception_tables(unsigned long add); + +/* Compare this to arch/i386/mm/extable.c:fixup_exception() */ +int arch_fixup(unsigned long address, void *sc_ptr) +{ + struct sigcontext *sc = sc_ptr; + const struct exception_table_entry *fixup; + + fixup = search_exception_tables(address); + if(fixup != 0){ + sc->eip = fixup->fixup; + return(1); + } + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-i386/ksyms.c b/arch/um/sys-i386/ksyms.c new file mode 100644 index 000000000000..74f70a120458 --- /dev/null +++ b/arch/um/sys-i386/ksyms.c @@ -0,0 +1,17 @@ +#include "linux/module.h" +#include "linux/in6.h" +#include "linux/rwsem.h" +#include "asm/byteorder.h" +#include "asm/semaphore.h" +#include "asm/uaccess.h" +#include "asm/checksum.h" +#include "asm/errno.h" + +EXPORT_SYMBOL(__down_failed); +EXPORT_SYMBOL(__down_failed_interruptible); +EXPORT_SYMBOL(__down_failed_trylock); +EXPORT_SYMBOL(__up_wakeup); + +/* Networking helper routines. */ +EXPORT_SYMBOL(csum_partial_copy_from); +EXPORT_SYMBOL(csum_partial_copy_to); diff --git a/arch/um/sys-i386/ldt.c b/arch/um/sys-i386/ldt.c new file mode 100644 index 000000000000..31bcb2f997d4 --- /dev/null +++ b/arch/um/sys-i386/ldt.c @@ -0,0 +1,98 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/slab.h" +#include "asm/uaccess.h" +#include "asm/ptrace.h" +#include "choose-mode.h" +#include "kern.h" + +#ifdef CONFIG_MODE_TT +extern int modify_ldt(int func, void *ptr, unsigned long bytecount); + +/* XXX this needs copy_to_user and copy_from_user */ + +int sys_modify_ldt_tt(int func, void __user *ptr, unsigned long bytecount) +{ + if (!access_ok(VERIFY_READ, ptr, bytecount)) + return -EFAULT; + + return modify_ldt(func, ptr, bytecount); +} +#endif + +#ifdef CONFIG_MODE_SKAS +extern int userspace_pid; + +#include "skas_ptrace.h" + +int sys_modify_ldt_skas(int func, void __user *ptr, unsigned long bytecount) +{ + struct ptrace_ldt ldt; + void *buf; + int res, n; + + buf = kmalloc(bytecount, GFP_KERNEL); + if(buf == NULL) + return(-ENOMEM); + + res = 0; + + switch(func){ + case 1: + case 0x11: + res = copy_from_user(buf, ptr, bytecount); + break; + } + + if(res != 0){ + res = -EFAULT; + goto out; + } + + ldt = ((struct ptrace_ldt) { .func = func, + .ptr = buf, + .bytecount = bytecount }); + res = ptrace(PTRACE_LDT, userspace_pid, 0, (unsigned long) &ldt); + if(res < 0) + goto out; + + switch(func){ + case 0: + case 2: + n = res; + res = copy_to_user(ptr, buf, n); + if(res != 0) + res = -EFAULT; + else + res = n; + break; + } + + out: + kfree(buf); + return(res); +} +#endif + +int sys_modify_ldt(int func, void __user *ptr, unsigned long bytecount) +{ + return(CHOOSE_MODE_PROC(sys_modify_ldt_tt, sys_modify_ldt_skas, func, + ptr, bytecount)); +} + + + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-i386/ptrace.c b/arch/um/sys-i386/ptrace.c new file mode 100644 index 000000000000..e470d28cdf84 --- /dev/null +++ b/arch/um/sys-i386/ptrace.c @@ -0,0 +1,369 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <linux/config.h> +#include <linux/compiler.h> +#include "linux/sched.h" +#include "asm/elf.h" +#include "asm/ptrace.h" +#include "asm/uaccess.h" +#include "asm/unistd.h" +#include "sysdep/ptrace.h" +#include "sysdep/sigcontext.h" +#include "sysdep/sc.h" + +void arch_switch(void) +{ + update_debugregs(current->thread.arch.debugregs_seq); +} + +int is_syscall(unsigned long addr) +{ + unsigned short instr; + int n; + + n = copy_from_user(&instr, (void __user *) addr, sizeof(instr)); + if(n){ + printk("is_syscall : failed to read instruction from 0x%lx\n", + addr); + return(0); + } + /* int 0x80 or sysenter */ + return((instr == 0x80cd) || (instr == 0x340f)); +} + +/* determines which flags the user has access to. */ +/* 1 = access 0 = no access */ +#define FLAG_MASK 0x00044dd5 + +int putreg(struct task_struct *child, int regno, unsigned long value) +{ + regno >>= 2; + switch (regno) { + case FS: + if (value && (value & 3) != 3) + return -EIO; + PT_REGS_FS(&child->thread.regs) = value; + return 0; + case GS: + if (value && (value & 3) != 3) + return -EIO; + PT_REGS_GS(&child->thread.regs) = value; + return 0; + case DS: + case ES: + if (value && (value & 3) != 3) + return -EIO; + value &= 0xffff; + break; + case SS: + case CS: + if ((value & 3) != 3) + return -EIO; + value &= 0xffff; + break; + case EFL: + value &= FLAG_MASK; + value |= PT_REGS_EFLAGS(&child->thread.regs); + break; + } + PT_REGS_SET(&child->thread.regs, regno, value); + return 0; +} + +unsigned long getreg(struct task_struct *child, int regno) +{ + unsigned long retval = ~0UL; + + regno >>= 2; + switch (regno) { + case FS: + case GS: + case DS: + case ES: + case SS: + case CS: + retval = 0xffff; + /* fall through */ + default: + retval &= PT_REG(&child->thread.regs, regno); + } + return retval; +} + +struct i387_fxsave_struct { + unsigned short cwd; + unsigned short swd; + unsigned short twd; + unsigned short fop; + long fip; + long fcs; + long foo; + long fos; + long mxcsr; + long reserved; + long st_space[32]; /* 8*16 bytes for each FP-reg = 128 bytes */ + long xmm_space[32]; /* 8*16 bytes for each XMM-reg = 128 bytes */ + long padding[56]; +}; + +/* + * FPU tag word conversions. + */ + +static inline unsigned short twd_i387_to_fxsr( unsigned short twd ) +{ + unsigned int tmp; /* to avoid 16 bit prefixes in the code */ + + /* Transform each pair of bits into 01 (valid) or 00 (empty) */ + tmp = ~twd; + tmp = (tmp | (tmp>>1)) & 0x5555; /* 0V0V0V0V0V0V0V0V */ + /* and move the valid bits to the lower byte. */ + tmp = (tmp | (tmp >> 1)) & 0x3333; /* 00VV00VV00VV00VV */ + tmp = (tmp | (tmp >> 2)) & 0x0f0f; /* 0000VVVV0000VVVV */ + tmp = (tmp | (tmp >> 4)) & 0x00ff; /* 00000000VVVVVVVV */ + return tmp; +} + +static inline unsigned long twd_fxsr_to_i387( struct i387_fxsave_struct *fxsave ) +{ + struct _fpxreg *st = NULL; + unsigned long twd = (unsigned long) fxsave->twd; + unsigned long tag; + unsigned long ret = 0xffff0000; + int i; + +#define FPREG_ADDR(f, n) ((char *)&(f)->st_space + (n) * 16); + + for ( i = 0 ; i < 8 ; i++ ) { + if ( twd & 0x1 ) { + st = (struct _fpxreg *) FPREG_ADDR( fxsave, i ); + + switch ( st->exponent & 0x7fff ) { + case 0x7fff: + tag = 2; /* Special */ + break; + case 0x0000: + if ( !st->significand[0] && + !st->significand[1] && + !st->significand[2] && + !st->significand[3] ) { + tag = 1; /* Zero */ + } else { + tag = 2; /* Special */ + } + break; + default: + if ( st->significand[3] & 0x8000 ) { + tag = 0; /* Valid */ + } else { + tag = 2; /* Special */ + } + break; + } + } else { + tag = 3; /* Empty */ + } + ret |= (tag << (2 * i)); + twd = twd >> 1; + } + return ret; +} + +/* + * FXSR floating point environment conversions. + */ + +#ifdef CONFIG_MODE_TT +static inline int convert_fxsr_to_user_tt(struct _fpstate __user *buf, + struct pt_regs *regs) +{ + struct i387_fxsave_struct *fxsave = SC_FXSR_ENV(PT_REGS_SC(regs)); + unsigned long env[7]; + struct _fpreg __user *to; + struct _fpxreg *from; + int i; + + env[0] = (unsigned long)fxsave->cwd | 0xffff0000; + env[1] = (unsigned long)fxsave->swd | 0xffff0000; + env[2] = twd_fxsr_to_i387(fxsave); + env[3] = fxsave->fip; + env[4] = fxsave->fcs | ((unsigned long)fxsave->fop << 16); + env[5] = fxsave->foo; + env[6] = fxsave->fos; + + if ( __copy_to_user( buf, env, 7 * sizeof(unsigned long) ) ) + return 1; + + to = &buf->_st[0]; + from = (struct _fpxreg *) &fxsave->st_space[0]; + for ( i = 0 ; i < 8 ; i++, to++, from++ ) { + if ( __copy_to_user( to, from, sizeof(*to) ) ) + return 1; + } + return 0; +} +#endif + +static inline int convert_fxsr_to_user(struct _fpstate __user *buf, + struct pt_regs *regs) +{ + return(CHOOSE_MODE(convert_fxsr_to_user_tt(buf, regs), 0)); +} + +#ifdef CONFIG_MODE_TT +static inline int convert_fxsr_from_user_tt(struct pt_regs *regs, + struct _fpstate __user *buf) +{ + struct i387_fxsave_struct *fxsave = SC_FXSR_ENV(PT_REGS_SC(regs)); + unsigned long env[7]; + struct _fpxreg *to; + struct _fpreg __user *from; + int i; + + if ( __copy_from_user( env, buf, 7 * sizeof(long) ) ) + return 1; + + fxsave->cwd = (unsigned short)(env[0] & 0xffff); + fxsave->swd = (unsigned short)(env[1] & 0xffff); + fxsave->twd = twd_i387_to_fxsr((unsigned short)(env[2] & 0xffff)); + fxsave->fip = env[3]; + fxsave->fop = (unsigned short)((env[4] & 0xffff0000) >> 16); + fxsave->fcs = (env[4] & 0xffff); + fxsave->foo = env[5]; + fxsave->fos = env[6]; + + to = (struct _fpxreg *) &fxsave->st_space[0]; + from = &buf->_st[0]; + for ( i = 0 ; i < 8 ; i++, to++, from++ ) { + if ( __copy_from_user( to, from, sizeof(*from) ) ) + return 1; + } + return 0; +} +#endif + +static inline int convert_fxsr_from_user(struct pt_regs *regs, + struct _fpstate __user *buf) +{ + return(CHOOSE_MODE(convert_fxsr_from_user_tt(regs, buf), 0)); +} + +int get_fpregs(unsigned long buf, struct task_struct *child) +{ + int err; + + err = convert_fxsr_to_user((struct _fpstate __user *) buf, + &child->thread.regs); + if(err) return(-EFAULT); + else return(0); +} + +int set_fpregs(unsigned long buf, struct task_struct *child) +{ + int err; + + err = convert_fxsr_from_user(&child->thread.regs, + (struct _fpstate __user *) buf); + if(err) return(-EFAULT); + else return(0); +} + +#ifdef CONFIG_MODE_TT +int get_fpxregs_tt(unsigned long buf, struct task_struct *tsk) +{ + struct pt_regs *regs = &tsk->thread.regs; + struct i387_fxsave_struct *fxsave = SC_FXSR_ENV(PT_REGS_SC(regs)); + int err; + + err = __copy_to_user((void __user *) buf, fxsave, + sizeof(struct user_fxsr_struct)); + if(err) return -EFAULT; + else return 0; +} +#endif + +int get_fpxregs(unsigned long buf, struct task_struct *tsk) +{ + return(CHOOSE_MODE(get_fpxregs_tt(buf, tsk), 0)); +} + +#ifdef CONFIG_MODE_TT +int set_fpxregs_tt(unsigned long buf, struct task_struct *tsk) +{ + struct pt_regs *regs = &tsk->thread.regs; + struct i387_fxsave_struct *fxsave = SC_FXSR_ENV(PT_REGS_SC(regs)); + int err; + + err = __copy_from_user(fxsave, (void __user *) buf, + sizeof(struct user_fxsr_struct) ); + if(err) return -EFAULT; + else return 0; +} +#endif + +int set_fpxregs(unsigned long buf, struct task_struct *tsk) +{ + return(CHOOSE_MODE(set_fpxregs_tt(buf, tsk), 0)); +} + +#ifdef notdef +int dump_fpu(struct pt_regs *regs, elf_fpregset_t *fpu) +{ + fpu->cwd = (((SC_FP_CW(PT_REGS_SC(regs)) & 0xffff) << 16) | + (SC_FP_SW(PT_REGS_SC(regs)) & 0xffff)); + fpu->swd = SC_FP_CSSEL(PT_REGS_SC(regs)) & 0xffff; + fpu->twd = SC_FP_IPOFF(PT_REGS_SC(regs)); + fpu->fip = SC_FP_CSSEL(PT_REGS_SC(regs)) & 0xffff; + fpu->fcs = SC_FP_DATAOFF(PT_REGS_SC(regs)); + fpu->foo = SC_FP_DATASEL(PT_REGS_SC(regs)); + fpu->fos = 0; + memcpy(fpu->st_space, (void *) SC_FP_ST(PT_REGS_SC(regs)), + sizeof(fpu->st_space)); + return(1); +} +#endif + +#ifdef CONFIG_MODE_TT +static inline void copy_fpu_fxsave_tt(struct pt_regs *regs, + struct user_i387_struct *buf) +{ + struct i387_fxsave_struct *fpu = SC_FXSR_ENV(PT_REGS_SC(regs)); + unsigned short *to; + unsigned short *from; + int i; + + memcpy( buf, fpu, 7 * sizeof(long) ); + + to = (unsigned short *) &buf->st_space[0]; + from = (unsigned short *) &fpu->st_space[0]; + for ( i = 0 ; i < 8 ; i++, to += 5, from += 8 ) { + memcpy( to, from, 5 * sizeof(unsigned short) ); + } +} +#endif + +static inline void copy_fpu_fxsave(struct pt_regs *regs, + struct user_i387_struct *buf) +{ + (void) CHOOSE_MODE(copy_fpu_fxsave_tt(regs, buf), 0); +} + +int dump_fpu(struct pt_regs *regs, elf_fpregset_t *fpu ) +{ + copy_fpu_fxsave(regs, (struct user_i387_struct *) fpu); + return(1); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-i386/ptrace_user.c b/arch/um/sys-i386/ptrace_user.c new file mode 100644 index 000000000000..7c376c95de50 --- /dev/null +++ b/arch/um/sys-i386/ptrace_user.c @@ -0,0 +1,131 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <errno.h> +#include <unistd.h> +#include <linux/stddef.h> +#include "ptrace_user.h" +/* Grr, asm/user.h includes asm/ptrace.h, so has to follow ptrace_user.h */ +#include <asm/user.h> +#include "kern_util.h" +#include "sysdep/thread.h" +#include "user.h" +#include "os.h" + +int ptrace_getregs(long pid, unsigned long *regs_out) +{ + if (ptrace(PTRACE_GETREGS, pid, 0, regs_out) < 0) + return -errno; + return 0; +} + +int ptrace_setregs(long pid, unsigned long *regs) +{ + if (ptrace(PTRACE_SETREGS, pid, 0, regs) < 0) + return -errno; + return 0; +} + +int ptrace_getfpregs(long pid, unsigned long *regs) +{ + if (ptrace(PTRACE_GETFPREGS, pid, 0, regs) < 0) + return -errno; + return 0; +} + +int ptrace_setfpregs(long pid, unsigned long *regs) +{ + if (ptrace(PTRACE_SETFPREGS, pid, 0, regs) < 0) + return -errno; + return 0; +} + +static void write_debugregs(int pid, unsigned long *regs) +{ + struct user *dummy; + int nregs, i; + + dummy = NULL; + nregs = sizeof(dummy->u_debugreg)/sizeof(dummy->u_debugreg[0]); + for(i = 0; i < nregs; i++){ + if((i == 4) || (i == 5)) continue; + if(ptrace(PTRACE_POKEUSR, pid, &dummy->u_debugreg[i], + regs[i]) < 0) + printk("write_debugregs - ptrace failed on " + "register %d, value = 0x%x, errno = %d\n", i, + regs[i], errno); + } +} + +static void read_debugregs(int pid, unsigned long *regs) +{ + struct user *dummy; + int nregs, i; + + dummy = NULL; + nregs = sizeof(dummy->u_debugreg)/sizeof(dummy->u_debugreg[0]); + for(i = 0; i < nregs; i++){ + regs[i] = ptrace(PTRACE_PEEKUSR, pid, + &dummy->u_debugreg[i], 0); + } +} + +/* Accessed only by the tracing thread */ +static unsigned long kernel_debugregs[8] = { [ 0 ... 7 ] = 0 }; +static int debugregs_seq = 0; + +void arch_enter_kernel(void *task, int pid) +{ + read_debugregs(pid, TASK_DEBUGREGS(task)); + write_debugregs(pid, kernel_debugregs); +} + +void arch_leave_kernel(void *task, int pid) +{ + read_debugregs(pid, kernel_debugregs); + write_debugregs(pid, TASK_DEBUGREGS(task)); +} + +void ptrace_pokeuser(unsigned long addr, unsigned long data) +{ + if((addr < offsetof(struct user, u_debugreg[0])) || + (addr > offsetof(struct user, u_debugreg[7]))) + return; + addr -= offsetof(struct user, u_debugreg[0]); + addr = addr >> 2; + if(kernel_debugregs[addr] == data) return; + + kernel_debugregs[addr] = data; + debugregs_seq++; +} + +static void update_debugregs_cb(void *arg) +{ + int pid = *((int *) arg); + + write_debugregs(pid, kernel_debugregs); +} + +void update_debugregs(int seq) +{ + int me; + + if(seq == debugregs_seq) return; + + me = os_getpid(); + initial_thread_cb(update_debugregs_cb, &me); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-i386/sigcontext.c b/arch/um/sys-i386/sigcontext.c new file mode 100644 index 000000000000..467d489c31cd --- /dev/null +++ b/arch/um/sys-i386/sigcontext.c @@ -0,0 +1,71 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stddef.h> +#include <string.h> +#include <asm/ptrace.h> +#include <asm/sigcontext.h> +#include "sysdep/ptrace.h" +#include "kern_util.h" + +void sc_to_sc(void *to_ptr, void *from_ptr) +{ + struct sigcontext *to = to_ptr, *from = from_ptr; + + memcpy(to, from, sizeof(*to) + sizeof(struct _fpstate)); + if(from->fpstate != NULL) + to->fpstate = (struct _fpstate *) (to + 1); +} + +unsigned long *sc_sigmask(void *sc_ptr) +{ + struct sigcontext *sc = sc_ptr; + return &sc->oldmask; +} + +int sc_get_fpregs(unsigned long buf, void *sc_ptr) +{ + struct sigcontext *sc = sc_ptr; + struct _fpstate *from = sc->fpstate, *to = (struct _fpstate *) buf; + int err = 0; + + if(from == NULL){ + err |= clear_user_proc(&to->cw, sizeof(to->cw)); + err |= clear_user_proc(&to->sw, sizeof(to->sw)); + err |= clear_user_proc(&to->tag, sizeof(to->tag)); + err |= clear_user_proc(&to->ipoff, sizeof(to->ipoff)); + err |= clear_user_proc(&to->cssel, sizeof(to->cssel)); + err |= clear_user_proc(&to->dataoff, sizeof(to->dataoff)); + err |= clear_user_proc(&to->datasel, sizeof(to->datasel)); + err |= clear_user_proc(&to->_st, sizeof(to->_st)); + } + else { + err |= copy_to_user_proc(&to->cw, &from->cw, sizeof(to->cw)); + err |= copy_to_user_proc(&to->sw, &from->sw, sizeof(to->sw)); + err |= copy_to_user_proc(&to->tag, &from->tag, + sizeof(to->tag)); + err |= copy_to_user_proc(&to->ipoff, &from->ipoff, + sizeof(to->ipoff)); + err |= copy_to_user_proc(&to->cssel,& from->cssel, + sizeof(to->cssel)); + err |= copy_to_user_proc(&to->dataoff, &from->dataoff, + sizeof(to->dataoff)); + err |= copy_to_user_proc(&to->datasel, &from->datasel, + sizeof(to->datasel)); + err |= copy_to_user_proc(to->_st, from->_st, sizeof(to->_st)); + } + return(err); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-i386/signal.c b/arch/um/sys-i386/signal.c new file mode 100644 index 000000000000..76ba87254b25 --- /dev/null +++ b/arch/um/sys-i386/signal.c @@ -0,0 +1,382 @@ +/* + * Copyright (C) 2004 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/signal.h" +#include "linux/ptrace.h" +#include "asm/current.h" +#include "asm/ucontext.h" +#include "asm/uaccess.h" +#include "asm/unistd.h" +#include "frame_kern.h" +#include "signal_user.h" +#include "sigcontext.h" +#include "registers.h" +#include "mode.h" + +#ifdef CONFIG_MODE_SKAS + +#include "skas.h" + +static int copy_sc_from_user_skas(struct pt_regs *regs, + struct sigcontext *from) +{ + struct sigcontext sc; + unsigned long fpregs[HOST_FP_SIZE]; + int err; + + err = copy_from_user(&sc, from, sizeof(sc)); + err |= copy_from_user(fpregs, sc.fpstate, sizeof(fpregs)); + if(err) + return(err); + + REGS_GS(regs->regs.skas.regs) = sc.gs; + REGS_FS(regs->regs.skas.regs) = sc.fs; + REGS_ES(regs->regs.skas.regs) = sc.es; + REGS_DS(regs->regs.skas.regs) = sc.ds; + REGS_EDI(regs->regs.skas.regs) = sc.edi; + REGS_ESI(regs->regs.skas.regs) = sc.esi; + REGS_EBP(regs->regs.skas.regs) = sc.ebp; + REGS_SP(regs->regs.skas.regs) = sc.esp; + REGS_EBX(regs->regs.skas.regs) = sc.ebx; + REGS_EDX(regs->regs.skas.regs) = sc.edx; + REGS_ECX(regs->regs.skas.regs) = sc.ecx; + REGS_EAX(regs->regs.skas.regs) = sc.eax; + REGS_IP(regs->regs.skas.regs) = sc.eip; + REGS_CS(regs->regs.skas.regs) = sc.cs; + REGS_EFLAGS(regs->regs.skas.regs) = sc.eflags; + REGS_SS(regs->regs.skas.regs) = sc.ss; + regs->regs.skas.fault_addr = sc.cr2; + regs->regs.skas.fault_type = FAULT_WRITE(sc.err); + regs->regs.skas.trap_type = sc.trapno; + + err = restore_fp_registers(userspace_pid[0], fpregs); + if(err < 0){ + printk("copy_sc_from_user_skas - PTRACE_SETFPREGS failed, " + "errno = %d\n", err); + return(1); + } + + return(0); +} + +int copy_sc_to_user_skas(struct sigcontext *to, struct _fpstate *to_fp, + struct pt_regs *regs, unsigned long fault_addr, + int fault_type) +{ + struct sigcontext sc; + unsigned long fpregs[HOST_FP_SIZE]; + int err; + + sc.gs = REGS_GS(regs->regs.skas.regs); + sc.fs = REGS_FS(regs->regs.skas.regs); + sc.es = REGS_ES(regs->regs.skas.regs); + sc.ds = REGS_DS(regs->regs.skas.regs); + sc.edi = REGS_EDI(regs->regs.skas.regs); + sc.esi = REGS_ESI(regs->regs.skas.regs); + sc.ebp = REGS_EBP(regs->regs.skas.regs); + sc.esp = REGS_SP(regs->regs.skas.regs); + sc.ebx = REGS_EBX(regs->regs.skas.regs); + sc.edx = REGS_EDX(regs->regs.skas.regs); + sc.ecx = REGS_ECX(regs->regs.skas.regs); + sc.eax = REGS_EAX(regs->regs.skas.regs); + sc.eip = REGS_IP(regs->regs.skas.regs); + sc.cs = REGS_CS(regs->regs.skas.regs); + sc.eflags = REGS_EFLAGS(regs->regs.skas.regs); + sc.esp_at_signal = regs->regs.skas.regs[UESP]; + sc.ss = regs->regs.skas.regs[SS]; + sc.cr2 = fault_addr; + sc.err = TO_SC_ERR(fault_type); + sc.trapno = regs->regs.skas.trap_type; + + err = save_fp_registers(userspace_pid[0], fpregs); + if(err < 0){ + printk("copy_sc_to_user_skas - PTRACE_GETFPREGS failed, " + "errno = %d\n", err); + return(1); + } + to_fp = (to_fp ? to_fp : (struct _fpstate *) (to + 1)); + sc.fpstate = to_fp; + + if(err) + return(err); + + return(copy_to_user(to, &sc, sizeof(sc)) || + copy_to_user(to_fp, fpregs, sizeof(fpregs))); +} +#endif + +#ifdef CONFIG_MODE_TT + +/* These copy a sigcontext to/from userspace. They copy the fpstate pointer, + * blowing away the old, good one. So, that value is saved, and then restored + * after the sigcontext copy. In copy_from, the variable holding the saved + * fpstate pointer, and the sigcontext that it should be restored to are both + * in the kernel, so we can just restore using an assignment. In copy_to, the + * saved pointer is in the kernel, but the sigcontext is in userspace, so we + * copy_to_user it. + */ +int copy_sc_from_user_tt(struct sigcontext *to, struct sigcontext *from, + int fpsize) +{ + struct _fpstate *to_fp, *from_fp; + unsigned long sigs; + int err; + + to_fp = to->fpstate; + from_fp = from->fpstate; + sigs = to->oldmask; + err = copy_from_user(to, from, sizeof(*to)); + to->oldmask = sigs; + to->fpstate = to_fp; + if(to_fp != NULL) + err |= copy_from_user(to_fp, from_fp, fpsize); + return(err); +} + +int copy_sc_to_user_tt(struct sigcontext *to, struct _fpstate *fp, + struct sigcontext *from, int fpsize) +{ + struct _fpstate *to_fp, *from_fp; + int err; + + to_fp = (fp ? fp : (struct _fpstate *) (to + 1)); + from_fp = from->fpstate; + err = copy_to_user(to, from, sizeof(*to)); + if(from_fp != NULL){ + err |= copy_to_user(&to->fpstate, &to_fp, sizeof(to->fpstate)); + err |= copy_to_user(to_fp, from_fp, fpsize); + } + return(err); +} +#endif + +static int copy_sc_from_user(struct pt_regs *to, void __user *from) +{ + int ret; + + ret = CHOOSE_MODE(copy_sc_from_user_tt(UPT_SC(&to->regs), from, + sizeof(struct _fpstate)), + copy_sc_from_user_skas(to, from)); + return(ret); +} + +static int copy_sc_to_user(struct sigcontext *to, struct _fpstate *fp, + struct pt_regs *from) +{ + return(CHOOSE_MODE(copy_sc_to_user_tt(to, fp, UPT_SC(&from->regs), + sizeof(*fp)), + copy_sc_to_user_skas(to, fp, from, + current->thread.cr2, + current->thread.err))); +} + +static int copy_ucontext_to_user(struct ucontext *uc, struct _fpstate *fp, + sigset_t *set, unsigned long sp) +{ + int err = 0; + + err |= put_user(current->sas_ss_sp, &uc->uc_stack.ss_sp); + err |= put_user(sas_ss_flags(sp), &uc->uc_stack.ss_flags); + err |= put_user(current->sas_ss_size, &uc->uc_stack.ss_size); + err |= copy_sc_to_user(&uc->uc_mcontext, fp, ¤t->thread.regs); + err |= copy_to_user(&uc->uc_sigmask, set, sizeof(*set)); + return(err); +} + +struct sigframe +{ + char *pretcode; + int sig; + struct sigcontext sc; + struct _fpstate fpstate; + unsigned long extramask[_NSIG_WORDS-1]; + char retcode[8]; +}; + +struct rt_sigframe +{ + char *pretcode; + int sig; + struct siginfo *pinfo; + void *puc; + struct siginfo info; + struct ucontext uc; + struct _fpstate fpstate; + char retcode[8]; +}; + +int setup_signal_stack_sc(unsigned long stack_top, int sig, + struct k_sigaction *ka, struct pt_regs *regs, + sigset_t *mask) +{ + struct sigframe __user *frame; + void *restorer; + int err = 0; + + stack_top &= -8UL; + frame = (struct sigframe *) stack_top - 1; + if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) + return 1; + + restorer = (void *) frame->retcode; + if(ka->sa.sa_flags & SA_RESTORER) + restorer = ka->sa.sa_restorer; + + err |= __put_user(restorer, &frame->pretcode); + err |= __put_user(sig, &frame->sig); + err |= copy_sc_to_user(&frame->sc, NULL, regs); + err |= __put_user(mask->sig[0], &frame->sc.oldmask); + if (_NSIG_WORDS > 1) + err |= __copy_to_user(&frame->extramask, &mask->sig[1], + sizeof(frame->extramask)); + + /* + * This is popl %eax ; movl $,%eax ; int $0x80 + * + * WE DO NOT USE IT ANY MORE! It's only left here for historical + * reasons and because gdb uses it as a signature to notice + * signal handler stack frames. + */ + err |= __put_user(0xb858, (short __user *)(frame->retcode+0)); + err |= __put_user(__NR_sigreturn, (int __user *)(frame->retcode+2)); + err |= __put_user(0x80cd, (short __user *)(frame->retcode+6)); + + if(err) + return(err); + + PT_REGS_SP(regs) = (unsigned long) frame; + PT_REGS_IP(regs) = (unsigned long) ka->sa.sa_handler; + PT_REGS_EAX(regs) = (unsigned long) sig; + PT_REGS_EDX(regs) = (unsigned long) 0; + PT_REGS_ECX(regs) = (unsigned long) 0; + + if ((current->ptrace & PT_DTRACE) && (current->ptrace & PT_PTRACED)) + ptrace_notify(SIGTRAP); + return(0); +} + +int setup_signal_stack_si(unsigned long stack_top, int sig, + struct k_sigaction *ka, struct pt_regs *regs, + siginfo_t *info, sigset_t *mask) +{ + struct rt_sigframe __user *frame; + void *restorer; + int err = 0; + + stack_top &= -8UL; + frame = (struct rt_sigframe *) stack_top - 1; + if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) + return 1; + + restorer = (void *) frame->retcode; + if(ka->sa.sa_flags & SA_RESTORER) + restorer = ka->sa.sa_restorer; + + err |= __put_user(restorer, &frame->pretcode); + err |= __put_user(sig, &frame->sig); + err |= __put_user(&frame->info, &frame->pinfo); + err |= __put_user(&frame->uc, &frame->puc); + err |= copy_siginfo_to_user(&frame->info, info); + err |= copy_ucontext_to_user(&frame->uc, &frame->fpstate, mask, + PT_REGS_SP(regs)); + + /* + * This is movl $,%eax ; int $0x80 + * + * WE DO NOT USE IT ANY MORE! It's only left here for historical + * reasons and because gdb uses it as a signature to notice + * signal handler stack frames. + */ + err |= __put_user(0xb8, (char __user *)(frame->retcode+0)); + err |= __put_user(__NR_rt_sigreturn, (int __user *)(frame->retcode+1)); + err |= __put_user(0x80cd, (short __user *)(frame->retcode+5)); + + if(err) + return(err); + + PT_REGS_SP(regs) = (unsigned long) frame; + PT_REGS_IP(regs) = (unsigned long) ka->sa.sa_handler; + PT_REGS_EAX(regs) = (unsigned long) sig; + PT_REGS_EDX(regs) = (unsigned long) &frame->info; + PT_REGS_ECX(regs) = (unsigned long) &frame->uc; + + if ((current->ptrace & PT_DTRACE) && (current->ptrace & PT_PTRACED)) + ptrace_notify(SIGTRAP); + return(0); +} + +long sys_sigreturn(struct pt_regs regs) +{ + unsigned long sp = PT_REGS_SP(¤t->thread.regs); + struct sigframe __user *frame = (struct sigframe *)(sp - 8); + sigset_t set; + struct sigcontext __user *sc = &frame->sc; + unsigned long __user *oldmask = &sc->oldmask; + unsigned long __user *extramask = frame->extramask; + int sig_size = (_NSIG_WORDS - 1) * sizeof(unsigned long); + + if(copy_from_user(&set.sig[0], oldmask, sizeof(&set.sig[0])) || + copy_from_user(&set.sig[1], extramask, sig_size)) + goto segfault; + + sigdelsetmask(&set, ~_BLOCKABLE); + + spin_lock_irq(¤t->sighand->siglock); + current->blocked = set; + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + + if(copy_sc_from_user(¤t->thread.regs, sc)) + goto segfault; + + /* Avoid ERESTART handling */ + PT_REGS_SYSCALL_NR(¤t->thread.regs) = -1; + return(PT_REGS_SYSCALL_RET(¤t->thread.regs)); + + segfault: + force_sig(SIGSEGV, current); + return 0; +} + +long sys_rt_sigreturn(struct pt_regs regs) +{ + unsigned long __user sp = PT_REGS_SP(¤t->thread.regs); + struct rt_sigframe __user *frame = (struct rt_sigframe *) (sp - 4); + sigset_t set; + struct ucontext __user *uc = &frame->uc; + int sig_size = _NSIG_WORDS * sizeof(unsigned long); + + if(copy_from_user(&set, &uc->uc_sigmask, sig_size)) + goto segfault; + + sigdelsetmask(&set, ~_BLOCKABLE); + + spin_lock_irq(¤t->sighand->siglock); + current->blocked = set; + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + + if(copy_sc_from_user(¤t->thread.regs, &uc->uc_mcontext)) + goto segfault; + + /* Avoid ERESTART handling */ + PT_REGS_SYSCALL_NR(¤t->thread.regs) = -1; + return(PT_REGS_SYSCALL_RET(¤t->thread.regs)); + + segfault: + force_sig(SIGSEGV, current); + return 0; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-i386/syscalls.c b/arch/um/sys-i386/syscalls.c new file mode 100644 index 000000000000..335e2d89504d --- /dev/null +++ b/arch/um/sys-i386/syscalls.c @@ -0,0 +1,210 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/shm.h" +#include "asm/ipc.h" +#include "asm/mman.h" +#include "asm/uaccess.h" +#include "asm/unistd.h" + +/* + * Perform the select(nd, in, out, ex, tv) and mmap() system + * calls. Linux/i386 didn't use to be able to handle more than + * 4 system call parameters, so these system calls used a memory + * block for parameter passing.. + */ + +struct mmap_arg_struct { + unsigned long addr; + unsigned long len; + unsigned long prot; + unsigned long flags; + unsigned long fd; + unsigned long offset; +}; + +extern int old_mmap(unsigned long addr, unsigned long len, + unsigned long prot, unsigned long flags, + unsigned long fd, unsigned long offset); + +long old_mmap_i386(struct mmap_arg_struct __user *arg) +{ + struct mmap_arg_struct a; + int err = -EFAULT; + + if (copy_from_user(&a, arg, sizeof(a))) + goto out; + + err = old_mmap(a.addr, a.len, a.prot, a.flags, a.fd, a.offset); + out: + return err; +} + +struct sel_arg_struct { + unsigned long n; + fd_set __user *inp; + fd_set __user *outp; + fd_set __user *exp; + struct timeval __user *tvp; +}; + +long old_select(struct sel_arg_struct __user *arg) +{ + struct sel_arg_struct a; + + if (copy_from_user(&a, arg, sizeof(a))) + return -EFAULT; + /* sys_select() does the appropriate kernel locking */ + return sys_select(a.n, a.inp, a.outp, a.exp, a.tvp); +} + +/* The i386 version skips reading from %esi, the fourth argument. So we must do + * this, too. + */ +long sys_clone(unsigned long clone_flags, unsigned long newsp, + int __user *parent_tid, int unused, int __user *child_tid) +{ + long ret; + + /* XXX: normal arch do here this pass, and also pass the regs to + * do_fork, instead of NULL. Currently the arch-independent code + * ignores these values, while the UML code (actually it's + * copy_thread) does the right thing. But this should change, + probably. */ + /*if (!newsp) + newsp = UPT_SP(current->thread.regs);*/ + current->thread.forking = 1; + ret = do_fork(clone_flags, newsp, NULL, 0, parent_tid, child_tid); + current->thread.forking = 0; + return(ret); +} + +/* + * sys_ipc() is the de-multiplexer for the SysV IPC calls.. + * + * This is really horribly ugly. + */ +long sys_ipc (uint call, int first, int second, + int third, void __user *ptr, long fifth) +{ + int version, ret; + + version = call >> 16; /* hack for backward compatibility */ + call &= 0xffff; + + switch (call) { + case SEMOP: + return sys_semtimedop(first, (struct sembuf *) ptr, second, + NULL); + case SEMTIMEDOP: + return sys_semtimedop(first, (struct sembuf *) ptr, second, + (const struct timespec *) fifth); + case SEMGET: + return sys_semget (first, second, third); + case SEMCTL: { + union semun fourth; + if (!ptr) + return -EINVAL; + if (get_user(fourth.__pad, (void **) ptr)) + return -EFAULT; + return sys_semctl (first, second, third, fourth); + } + + case MSGSND: + return sys_msgsnd (first, (struct msgbuf *) ptr, + second, third); + case MSGRCV: + switch (version) { + case 0: { + struct ipc_kludge tmp; + if (!ptr) + return -EINVAL; + + if (copy_from_user(&tmp, + (struct ipc_kludge *) ptr, + sizeof (tmp))) + return -EFAULT; + return sys_msgrcv (first, tmp.msgp, second, + tmp.msgtyp, third); + } + default: + panic("msgrcv with version != 0"); + return sys_msgrcv (first, + (struct msgbuf *) ptr, + second, fifth, third); + } + case MSGGET: + return sys_msgget ((key_t) first, second); + case MSGCTL: + return sys_msgctl (first, second, (struct msqid_ds *) ptr); + + case SHMAT: + switch (version) { + default: { + ulong raddr; + ret = do_shmat (first, (char *) ptr, second, &raddr); + if (ret) + return ret; + return put_user (raddr, (ulong *) third); + } + case 1: /* iBCS2 emulator entry point */ + if (!segment_eq(get_fs(), get_ds())) + return -EINVAL; + return do_shmat (first, (char *) ptr, second, (ulong *) third); + } + case SHMDT: + return sys_shmdt ((char *)ptr); + case SHMGET: + return sys_shmget (first, second, third); + case SHMCTL: + return sys_shmctl (first, second, + (struct shmid_ds *) ptr); + default: + return -ENOSYS; + } +} + +long sys_sigaction(int sig, const struct old_sigaction __user *act, + struct old_sigaction __user *oact) +{ + struct k_sigaction new_ka, old_ka; + int ret; + + if (act) { + old_sigset_t mask; + if (!access_ok(VERIFY_READ, act, sizeof(*act)) || + __get_user(new_ka.sa.sa_handler, &act->sa_handler) || + __get_user(new_ka.sa.sa_restorer, &act->sa_restorer)) + return -EFAULT; + __get_user(new_ka.sa.sa_flags, &act->sa_flags); + __get_user(mask, &act->sa_mask); + siginitset(&new_ka.sa.sa_mask, mask); + } + + ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL); + + if (!ret && oact) { + if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) || + __put_user(old_ka.sa.sa_handler, &oact->sa_handler) || + __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer)) + return -EFAULT; + __put_user(old_ka.sa.sa_flags, &oact->sa_flags); + __put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask); + } + + return ret; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-i386/sysrq.c b/arch/um/sys-i386/sysrq.c new file mode 100644 index 000000000000..281fc7b8ca00 --- /dev/null +++ b/arch/um/sys-i386/sysrq.c @@ -0,0 +1,35 @@ +/* + * Copyright (C) 2001 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/kernel.h" +#include "linux/smp.h" +#include "linux/sched.h" +#include "asm/ptrace.h" +#include "sysrq.h" + +void show_regs(struct pt_regs *regs) +{ + printk("\n"); + printk("EIP: %04lx:[<%08lx>] CPU: %d %s", + 0xffff & PT_REGS_CS(regs), PT_REGS_IP(regs), + smp_processor_id(), print_tainted()); + if (PT_REGS_CS(regs) & 3) + printk(" ESP: %04lx:%08lx", 0xffff & PT_REGS_SS(regs), + PT_REGS_SP(regs)); + printk(" EFLAGS: %08lx\n %s\n", PT_REGS_EFLAGS(regs), + print_tainted()); + printk("EAX: %08lx EBX: %08lx ECX: %08lx EDX: %08lx\n", + PT_REGS_EAX(regs), PT_REGS_EBX(regs), + PT_REGS_ECX(regs), + PT_REGS_EDX(regs)); + printk("ESI: %08lx EDI: %08lx EBP: %08lx", + PT_REGS_ESI(regs), PT_REGS_EDI(regs), + PT_REGS_EBP(regs)); + printk(" DS: %04lx ES: %04lx\n", + 0xffff & PT_REGS_DS(regs), + 0xffff & PT_REGS_ES(regs)); + + show_trace((unsigned long *) ®s); +} diff --git a/arch/um/sys-i386/util/Makefile b/arch/um/sys-i386/util/Makefile new file mode 100644 index 000000000000..34860f9ca7b0 --- /dev/null +++ b/arch/um/sys-i386/util/Makefile @@ -0,0 +1,8 @@ + +hostprogs-y := mk_sc mk_thread +always := $(hostprogs-y) + +mk_thread-objs := mk_thread_kern.o mk_thread_user.o + +HOSTCFLAGS_mk_thread_kern.o := $(CFLAGS) $(CPPFLAGS) +HOSTCFLAGS_mk_thread_user.o := $(USER_CFLAGS) diff --git a/arch/um/sys-i386/util/mk_sc.c b/arch/um/sys-i386/util/mk_sc.c new file mode 100644 index 000000000000..85cbd30396f7 --- /dev/null +++ b/arch/um/sys-i386/util/mk_sc.c @@ -0,0 +1,52 @@ +#include <stdio.h> +#include <signal.h> +#include <linux/stddef.h> + +#define SC_OFFSET(name, field) \ + printf("#define " name "(sc) *((unsigned long *) &(((char *) (sc))[%d]))\n",\ + offsetof(struct sigcontext, field)) + +#define SC_FP_OFFSET(name, field) \ + printf("#define " name \ + "(sc) *((unsigned long *) &(((char *) (SC_FPSTATE(sc)))[%d]))\n",\ + offsetof(struct _fpstate, field)) + +#define SC_FP_OFFSET_PTR(name, field, type) \ + printf("#define " name \ + "(sc) ((" type " *) &(((char *) (SC_FPSTATE(sc)))[%d]))\n",\ + offsetof(struct _fpstate, field)) + +int main(int argc, char **argv) +{ + SC_OFFSET("SC_IP", eip); + SC_OFFSET("SC_SP", esp); + SC_OFFSET("SC_FS", fs); + SC_OFFSET("SC_GS", gs); + SC_OFFSET("SC_DS", ds); + SC_OFFSET("SC_ES", es); + SC_OFFSET("SC_SS", ss); + SC_OFFSET("SC_CS", cs); + SC_OFFSET("SC_EFLAGS", eflags); + SC_OFFSET("SC_EAX", eax); + SC_OFFSET("SC_EBX", ebx); + SC_OFFSET("SC_ECX", ecx); + SC_OFFSET("SC_EDX", edx); + SC_OFFSET("SC_EDI", edi); + SC_OFFSET("SC_ESI", esi); + SC_OFFSET("SC_EBP", ebp); + SC_OFFSET("SC_TRAPNO", trapno); + SC_OFFSET("SC_ERR", err); + SC_OFFSET("SC_CR2", cr2); + SC_OFFSET("SC_FPSTATE", fpstate); + SC_OFFSET("SC_SIGMASK", oldmask); + SC_FP_OFFSET("SC_FP_CW", cw); + SC_FP_OFFSET("SC_FP_SW", sw); + SC_FP_OFFSET("SC_FP_TAG", tag); + SC_FP_OFFSET("SC_FP_IPOFF", ipoff); + SC_FP_OFFSET("SC_FP_CSSEL", cssel); + SC_FP_OFFSET("SC_FP_DATAOFF", dataoff); + SC_FP_OFFSET("SC_FP_DATASEL", datasel); + SC_FP_OFFSET_PTR("SC_FP_ST", _st, "struct _fpstate"); + SC_FP_OFFSET_PTR("SC_FXSR_ENV", _fxsr_env, "void"); + return(0); +} diff --git a/arch/um/sys-i386/util/mk_thread_kern.c b/arch/um/sys-i386/util/mk_thread_kern.c new file mode 100644 index 000000000000..948b1ce85230 --- /dev/null +++ b/arch/um/sys-i386/util/mk_thread_kern.c @@ -0,0 +1,22 @@ +#include "linux/config.h" +#include "linux/stddef.h" +#include "linux/sched.h" + +extern void print_head(void); +extern void print_constant_ptr(char *name, int value); +extern void print_constant(char *name, char *type, int value); +extern void print_tail(void); + +#define THREAD_OFFSET(field) offsetof(struct task_struct, thread.field) + +int main(int argc, char **argv) +{ + print_head(); + print_constant_ptr("TASK_DEBUGREGS", THREAD_OFFSET(arch.debugregs)); +#ifdef CONFIG_MODE_TT + print_constant("TASK_EXTERN_PID", "int", THREAD_OFFSET(mode.tt.extern_pid)); +#endif + print_tail(); + return(0); +} + diff --git a/arch/um/sys-i386/util/mk_thread_user.c b/arch/um/sys-i386/util/mk_thread_user.c new file mode 100644 index 000000000000..2620cd6aa1f1 --- /dev/null +++ b/arch/um/sys-i386/util/mk_thread_user.c @@ -0,0 +1,30 @@ +#include <stdio.h> + +void print_head(void) +{ + printf("/*\n"); + printf(" * Generated by mk_thread\n"); + printf(" */\n"); + printf("\n"); + printf("#ifndef __UM_THREAD_H\n"); + printf("#define __UM_THREAD_H\n"); + printf("\n"); +} + +void print_constant_ptr(char *name, int value) +{ + printf("#define %s(task) ((unsigned long *) " + "&(((char *) (task))[%d]))\n", name, value); +} + +void print_constant(char *name, char *type, int value) +{ + printf("#define %s(task) *((%s *) &(((char *) (task))[%d]))\n", name, type, + value); +} + +void print_tail(void) +{ + printf("\n"); + printf("#endif\n"); +} diff --git a/arch/um/sys-ia64/Makefile b/arch/um/sys-ia64/Makefile new file mode 100644 index 000000000000..d02f4c265232 --- /dev/null +++ b/arch/um/sys-ia64/Makefile @@ -0,0 +1,11 @@ +OBJ = built-in.o + +OBJS = + +all: $(OBJ) + +$(OBJ): $(OBJS) + rm -f $@ + $(LD) $(LINKFLAGS) --start-group $^ --end-group -o $@ + +clean-files := $(OBJS) link.ld diff --git a/arch/um/sys-ppc/Makefile b/arch/um/sys-ppc/Makefile new file mode 100644 index 000000000000..af200268fddb --- /dev/null +++ b/arch/um/sys-ppc/Makefile @@ -0,0 +1,69 @@ +OBJ = built-in.o + +.S.o: + $(CC) $(AFLAGS) -D__ASSEMBLY__ -D__UM_PPC__ -c $< -o $*.o + +OBJS = ptrace.o sigcontext.o semaphore.o checksum.o miscthings.o misc.o \ + ptrace_user.o sysrq.o + +EXTRA_AFLAGS := -DCONFIG_PPC32 -I. -I$(TOPDIR)/arch/ppc/kernel + +all: $(OBJ) + +$(OBJ): $(OBJS) + rm -f $@ + $(LD) $(LINKFLAGS) --start-group $^ --end-group -o $@ + +ptrace_user.o: ptrace_user.c + $(CC) -D__KERNEL__ $(USER_CFLAGS) $(EXTRA_CFLAGS) -c -o $@ $< + +sigcontext.o: sigcontext.c + $(CC) $(USER_CFLAGS) $(EXTRA_CFLAGS) -c -o $@ $< + +semaphore.c: + rm -f $@ + ln -s $(TOPDIR)/arch/ppc/kernel/$@ $@ + +checksum.S: + rm -f $@ + ln -s $(TOPDIR)/arch/ppc/lib/$@ $@ + +mk_defs.c: + rm -f $@ + ln -s $(TOPDIR)/arch/ppc/kernel/$@ $@ + +ppc_defs.head: + rm -f $@ + ln -s $(TOPDIR)/arch/ppc/kernel/$@ $@ + +ppc_defs.h: mk_defs.c ppc_defs.head \ + $(TOPDIR)/include/asm-ppc/mmu.h \ + $(TOPDIR)/include/asm-ppc/processor.h \ + $(TOPDIR)/include/asm-ppc/pgtable.h \ + $(TOPDIR)/include/asm-ppc/ptrace.h +# $(CC) $(CFLAGS) -S mk_defs.c + cp ppc_defs.head ppc_defs.h +# for bk, this way we can write to the file even if it's not checked out + echo '#define THREAD 608' >> ppc_defs.h + echo '#define PT_REGS 8' >> ppc_defs.h + echo '#define CLONE_VM 256' >> ppc_defs.h +# chmod u+w ppc_defs.h +# grep '^#define' mk_defs.s >> ppc_defs.h +# rm mk_defs.s + +# the asm link is horrible, and breaks the other targets. This is also +# not going to work with parallel makes. + +checksum.o: checksum.S + rm -f asm + ln -s $(TOPDIR)/include/asm-ppc asm + $(CC) $(EXTRA_AFLAGS) $(AFLAGS) -D__ASSEMBLY__ -D__UM_PPC__ -c $< -o $*.o + rm -f asm + +misc.o: misc.S ppc_defs.h + rm -f asm + ln -s $(TOPDIR)/include/asm-ppc asm + $(CC) $(EXTRA_AFLAGS) $(AFLAGS) -D__ASSEMBLY__ -D__UM_PPC__ -c $< -o $*.o + rm -f asm + +clean-files := $(OBJS) ppc_defs.h checksum.S semaphore.c mk_defs.c diff --git a/arch/um/sys-ppc/misc.S b/arch/um/sys-ppc/misc.S new file mode 100644 index 000000000000..11b7bd768cfd --- /dev/null +++ b/arch/um/sys-ppc/misc.S @@ -0,0 +1,116 @@ +/* + * This file contains miscellaneous low-level functions. + * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org) + * + * Largely rewritten by Cort Dougan (cort@cs.nmt.edu) + * and Paul Mackerras. + * + * A couple of functions stolen from arch/ppc/kernel/misc.S for UML + * by Chris Emerson. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + */ + +#include <linux/config.h> +#include <asm/processor.h> +#include "ppc_asm.h" + +#if defined(CONFIG_4xx) || defined(CONFIG_8xx) +#define CACHE_LINE_SIZE 16 +#define LG_CACHE_LINE_SIZE 4 +#define MAX_COPY_PREFETCH 1 +#elif !defined(CONFIG_PPC64BRIDGE) +#define CACHE_LINE_SIZE 32 +#define LG_CACHE_LINE_SIZE 5 +#define MAX_COPY_PREFETCH 4 +#else +#define CACHE_LINE_SIZE 128 +#define LG_CACHE_LINE_SIZE 7 +#define MAX_COPY_PREFETCH 1 +#endif /* CONFIG_4xx || CONFIG_8xx */ + + .text + +/* + * Clear a page using the dcbz instruction, which doesn't cause any + * memory traffic (except to write out any cache lines which get + * displaced). This only works on cacheable memory. + */ +_GLOBAL(clear_page) + li r0,4096/CACHE_LINE_SIZE + mtctr r0 +#ifdef CONFIG_8xx + li r4, 0 +1: stw r4, 0(r3) + stw r4, 4(r3) + stw r4, 8(r3) + stw r4, 12(r3) +#else +1: dcbz 0,r3 +#endif + addi r3,r3,CACHE_LINE_SIZE + bdnz 1b + blr + +/* + * Copy a whole page. We use the dcbz instruction on the destination + * to reduce memory traffic (it eliminates the unnecessary reads of + * the destination into cache). This requires that the destination + * is cacheable. + */ +#define COPY_16_BYTES \ + lwz r6,4(r4); \ + lwz r7,8(r4); \ + lwz r8,12(r4); \ + lwzu r9,16(r4); \ + stw r6,4(r3); \ + stw r7,8(r3); \ + stw r8,12(r3); \ + stwu r9,16(r3) + +_GLOBAL(copy_page) + addi r3,r3,-4 + addi r4,r4,-4 + li r5,4 + +#ifndef CONFIG_8xx +#if MAX_COPY_PREFETCH > 1 + li r0,MAX_COPY_PREFETCH + li r11,4 + mtctr r0 +11: dcbt r11,r4 + addi r11,r11,CACHE_LINE_SIZE + bdnz 11b +#else /* MAX_COPY_PREFETCH == 1 */ + dcbt r5,r4 + li r11,CACHE_LINE_SIZE+4 +#endif /* MAX_COPY_PREFETCH */ +#endif /* CONFIG_8xx */ + + li r0,4096/CACHE_LINE_SIZE + mtctr r0 +1: +#ifndef CONFIG_8xx + dcbt r11,r4 + dcbz r5,r3 +#endif + COPY_16_BYTES +#if CACHE_LINE_SIZE >= 32 + COPY_16_BYTES +#if CACHE_LINE_SIZE >= 64 + COPY_16_BYTES + COPY_16_BYTES +#if CACHE_LINE_SIZE >= 128 + COPY_16_BYTES + COPY_16_BYTES + COPY_16_BYTES + COPY_16_BYTES +#endif +#endif +#endif + bdnz 1b + blr diff --git a/arch/um/sys-ppc/miscthings.c b/arch/um/sys-ppc/miscthings.c new file mode 100644 index 000000000000..373061c50129 --- /dev/null +++ b/arch/um/sys-ppc/miscthings.c @@ -0,0 +1,53 @@ +#include "linux/threads.h" +#include "linux/stddef.h" // for NULL +#include "linux/elf.h" // for AT_NULL + +/* The following function nicked from arch/ppc/kernel/process.c and + * adapted slightly */ +/* + * XXX ld.so expects the auxiliary table to start on + * a 16-byte boundary, so we have to find it and + * move it up. :-( + */ +void shove_aux_table(unsigned long sp) +{ + int argc; + char *p; + unsigned long e; + unsigned long aux_start, offset; + + argc = *(int *)sp; + sp += sizeof(int) + (argc + 1) * sizeof(char *); + /* skip over the environment pointers */ + do { + p = *(char **)sp; + sp += sizeof(char *); + } while (p != NULL); + aux_start = sp; + /* skip to the end of the auxiliary table */ + do { + e = *(unsigned long *)sp; + sp += 2 * sizeof(unsigned long); + } while (e != AT_NULL); + offset = ((aux_start + 15) & ~15) - aux_start; + if (offset != 0) { + do { + sp -= sizeof(unsigned long); + e = *(unsigned long *)sp; + *(unsigned long *)(sp + offset) = e; + } while (sp > aux_start); + } +} +/* END stuff taken from arch/ppc/kernel/process.c */ + + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-ppc/ptrace.c b/arch/um/sys-ppc/ptrace.c new file mode 100644 index 000000000000..a971366d3277 --- /dev/null +++ b/arch/um/sys-ppc/ptrace.c @@ -0,0 +1,28 @@ +#include "linux/sched.h" +#include "asm/ptrace.h" + +int putreg(struct task_struct *child, unsigned long regno, + unsigned long value) +{ + child->thread.process_regs.regs[regno >> 2] = value; + return 0; +} + +unsigned long getreg(struct task_struct *child, unsigned long regno) +{ + unsigned long retval = ~0UL; + + retval &= child->thread.process_regs.regs[regno >> 2]; + return retval; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-ppc/ptrace_user.c b/arch/um/sys-ppc/ptrace_user.c new file mode 100644 index 000000000000..ff0b9c077a13 --- /dev/null +++ b/arch/um/sys-ppc/ptrace_user.c @@ -0,0 +1,39 @@ +#include <errno.h> +#include <asm/ptrace.h> +#include "sysdep/ptrace.h" + +int ptrace_getregs(long pid, unsigned long *regs_out) +{ + int i; + for (i=0; i < sizeof(struct sys_pt_regs)/sizeof(PPC_REG); ++i) { + errno = 0; + regs_out->regs[i] = ptrace(PTRACE_PEEKUSR, pid, i*4, 0); + if (errno) { + return -errno; + } + } + return 0; +} + +int ptrace_setregs(long pid, unsigned long *regs_in) +{ + int i; + for (i=0; i < sizeof(struct sys_pt_regs)/sizeof(PPC_REG); ++i) { + if (i != 34 /* FIXME: PT_ORIG_R3 */ && i <= PT_MQ) { + if (ptrace(PTRACE_POKEUSR, pid, i*4, regs_in->regs[i]) < 0) { + return -errno; + } + } + } + return 0; +} +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-ppc/sigcontext.c b/arch/um/sys-ppc/sigcontext.c new file mode 100644 index 000000000000..5d430fc994af --- /dev/null +++ b/arch/um/sys-ppc/sigcontext.c @@ -0,0 +1,15 @@ +#include "asm/ptrace.h" +#include "asm/sigcontext.h" +#include "sysdep/ptrace.h" +#include "user_util.h" + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-ppc/sysrq.c b/arch/um/sys-ppc/sysrq.c new file mode 100644 index 000000000000..82d6e9335bb6 --- /dev/null +++ b/arch/um/sys-ppc/sysrq.c @@ -0,0 +1,43 @@ +/* + * Copyright (C) 2001 Chris Emerson (cemerson@chiark.greenend.org.uk) + * Licensed under the GPL + */ + +#include "linux/kernel.h" +#include "linux/smp.h" +#include "asm/ptrace.h" +#include "sysrq.h" + +void show_regs(struct pt_regs_subarch *regs) +{ + printk("\n"); + printk("show_regs(): insert regs here.\n"); +#if 0 + printk("\n"); + printk("EIP: %04x:[<%08lx>] CPU: %d",0xffff & regs->xcs, regs->eip, + smp_processor_id()); + if (regs->xcs & 3) + printk(" ESP: %04x:%08lx",0xffff & regs->xss, regs->esp); + printk(" EFLAGS: %08lx\n", regs->eflags); + printk("EAX: %08lx EBX: %08lx ECX: %08lx EDX: %08lx\n", + regs->eax, regs->ebx, regs->ecx, regs->edx); + printk("ESI: %08lx EDI: %08lx EBP: %08lx", + regs->esi, regs->edi, regs->ebp); + printk(" DS: %04x ES: %04x\n", + 0xffff & regs->xds, 0xffff & regs->xes); +#endif + + show_trace(®s->gpr[1]); +} + + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-x86_64/Makefile b/arch/um/sys-x86_64/Makefile new file mode 100644 index 000000000000..2129e3143559 --- /dev/null +++ b/arch/um/sys-x86_64/Makefile @@ -0,0 +1,37 @@ +# +# Copyright 2003 PathScale, Inc. +# +# Licensed under the GPL +# + +lib-y = bitops.o bugs.o csum-partial.o delay.o fault.o mem.o memcpy.o \ + ptrace.o ptrace_user.o semaphore.o sigcontext.o signal.o \ + syscalls.o sysrq.o thunk.o + +USER_OBJS := ptrace_user.o sigcontext.o + +include arch/um/scripts/Makefile.rules + +SYMLINKS = bitops.c csum-copy.S csum-partial.c csum-wrappers.c memcpy.S \ + semaphore.c thunk.S + +# this needs to be before the foreach, because clean-files does not accept +# complete paths like $(src)/$f. +clean-files := $(SYMLINKS) + +targets += $(SYMLINKS) + +SYMLINKS := $(foreach f,$(SYMLINKS),$(obj)/$f) + +bitops.c-dir = lib +csum-copy.S-dir = lib +csum-partial.c-dir = lib +csum-wrappers.c-dir = lib +memcpy.S-dir = lib +semaphore.c-dir = kernel +thunk.S-dir = lib + +$(SYMLINKS): FORCE + $(call if_changed,make_link) + +CFLAGS_csum-partial.o := -Dcsum_partial=arch_csum_partial diff --git a/arch/um/sys-x86_64/bugs.c b/arch/um/sys-x86_64/bugs.c new file mode 100644 index 000000000000..fdce7ea98ca7 --- /dev/null +++ b/arch/um/sys-x86_64/bugs.c @@ -0,0 +1,122 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/errno.h" +#include "asm/system.h" +#include "asm/pda.h" +#include "sysdep/ptrace.h" +#include "os.h" + +void arch_init_thread(void) +{ +} + +void arch_check_bugs(void) +{ +} + +int arch_handle_signal(int sig, union uml_pt_regs *regs) +{ + return(0); +} + +#define MAXTOKEN 64 + +/* Set during early boot */ +int host_has_cmov = 1; +int host_has_xmm = 0; + +static char token(int fd, char *buf, int len, char stop) +{ + int n; + char *ptr, *end, c; + + ptr = buf; + end = &buf[len]; + do { + n = os_read_file(fd, ptr, sizeof(*ptr)); + c = *ptr++; + if(n != sizeof(*ptr)){ + if(n == 0) return(0); + printk("Reading /proc/cpuinfo failed, err = %d\n", -n); + if(n < 0) + return(n); + else + return(-EIO); + } + } while((c != '\n') && (c != stop) && (ptr < end)); + + if(ptr == end){ + printk("Failed to find '%c' in /proc/cpuinfo\n", stop); + return(-1); + } + *(ptr - 1) = '\0'; + return(c); +} + +static int find_cpuinfo_line(int fd, char *key, char *scratch, int len) +{ + int n; + char c; + + scratch[len - 1] = '\0'; + while(1){ + c = token(fd, scratch, len - 1, ':'); + if(c <= 0) + return(0); + else if(c != ':'){ + printk("Failed to find ':' in /proc/cpuinfo\n"); + return(0); + } + + if(!strncmp(scratch, key, strlen(key))) + return(1); + + do { + n = os_read_file(fd, &c, sizeof(c)); + if(n != sizeof(c)){ + printk("Failed to find newline in " + "/proc/cpuinfo, err = %d\n", -n); + return(0); + } + } while(c != '\n'); + } + return(0); +} + +int cpu_feature(char *what, char *buf, int len) +{ + int fd, ret = 0; + + fd = os_open_file("/proc/cpuinfo", of_read(OPENFLAGS()), 0); + if(fd < 0){ + printk("Couldn't open /proc/cpuinfo, err = %d\n", -fd); + return(0); + } + + if(!find_cpuinfo_line(fd, what, buf, len)){ + printk("Couldn't find '%s' line in /proc/cpuinfo\n", what); + goto out_close; + } + + token(fd, buf, len, '\n'); + ret = 1; + + out_close: + os_close_file(fd); + return(ret); +} + +/* Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-x86_64/delay.c b/arch/um/sys-x86_64/delay.c new file mode 100644 index 000000000000..f3b5187942b4 --- /dev/null +++ b/arch/um/sys-x86_64/delay.c @@ -0,0 +1,26 @@ +/* + * Copyright 2003 PathScale, Inc. + * Copied from arch/x86_64 + * + * Licensed under the GPL + */ + +#include "asm/processor.h" + +void __delay(unsigned long loops) +{ + unsigned long i; + + for(i = 0; i < loops; i++) ; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-x86_64/fault.c b/arch/um/sys-x86_64/fault.c new file mode 100644 index 000000000000..cee1513c5c31 --- /dev/null +++ b/arch/um/sys-x86_64/fault.c @@ -0,0 +1,23 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ + +#include "user.h" + +int arch_fixup(unsigned long address, void *sc_ptr) +{ + /* XXX search_exception_tables() */ + return(0); +} + +/* Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-x86_64/mem.c b/arch/um/sys-x86_64/mem.c new file mode 100644 index 000000000000..3f59a0a4f156 --- /dev/null +++ b/arch/um/sys-x86_64/mem.c @@ -0,0 +1,25 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ + +#include "linux/mm.h" +#include "asm/page.h" +#include "asm/mman.h" + +unsigned long vm_stack_flags = __VM_STACK_FLAGS; +unsigned long vm_stack_flags32 = __VM_STACK_FLAGS; +unsigned long vm_data_default_flags = __VM_DATA_DEFAULT_FLAGS; +unsigned long vm_data_default_flags32 = __VM_DATA_DEFAULT_FLAGS; +unsigned long vm_force_exec32 = PROT_EXEC; + +/* Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-x86_64/ptrace.c b/arch/um/sys-x86_64/ptrace.c new file mode 100644 index 000000000000..8c146b2a1e00 --- /dev/null +++ b/arch/um/sys-x86_64/ptrace.c @@ -0,0 +1,138 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ + +#define __FRAME_OFFSETS +#include "asm/ptrace.h" +#include "linux/sched.h" +#include "linux/errno.h" +#include "asm/elf.h" + +/* XXX x86_64 */ +unsigned long not_ss; +unsigned long not_ds; +unsigned long not_es; + +#define SC_SS(r) (not_ss) +#define SC_DS(r) (not_ds) +#define SC_ES(r) (not_es) + +/* determines which flags the user has access to. */ +/* 1 = access 0 = no access */ +#define FLAG_MASK 0x44dd5UL + +int putreg(struct task_struct *child, int regno, unsigned long value) +{ + unsigned long tmp; + +#ifdef TIF_IA32 + /* Some code in the 64bit emulation may not be 64bit clean. + Don't take any chances. */ + if (test_tsk_thread_flag(child, TIF_IA32)) + value &= 0xffffffff; +#endif + switch (regno){ + case FS: + case GS: + case DS: + case ES: + case SS: + case CS: + if (value && (value & 3) != 3) + return -EIO; + value &= 0xffff; + break; + + case FS_BASE: + case GS_BASE: + if (!((value >> 48) == 0 || (value >> 48) == 0xffff)) + return -EIO; + break; + + case EFLAGS: + value &= FLAG_MASK; + tmp = PT_REGS_EFLAGS(&child->thread.regs) & ~FLAG_MASK; + value |= tmp; + break; + } + + PT_REGS_SET(&child->thread.regs, regno, value); + return 0; +} + +unsigned long getreg(struct task_struct *child, int regno) +{ + unsigned long retval = ~0UL; + switch (regno) { + case FS: + case GS: + case DS: + case ES: + case SS: + case CS: + retval = 0xffff; + /* fall through */ + default: + retval &= PT_REG(&child->thread.regs, regno); +#ifdef TIF_IA32 + if (test_tsk_thread_flag(child, TIF_IA32)) + retval &= 0xffffffff; +#endif + } + return retval; +} + +void arch_switch(void) +{ +/* XXX + printk("arch_switch\n"); +*/ +} + +int is_syscall(unsigned long addr) +{ + panic("is_syscall"); +} + +int dump_fpu(struct pt_regs *regs, elf_fpregset_t *fpu ) +{ + panic("dump_fpu"); + return(1); +} + +int get_fpregs(unsigned long buf, struct task_struct *child) +{ + panic("get_fpregs"); + return(0); +} + +int set_fpregs(unsigned long buf, struct task_struct *child) +{ + panic("set_fpregs"); + return(0); +} + +int get_fpxregs(unsigned long buf, struct task_struct *tsk) +{ + panic("get_fpxregs"); + return(0); +} + +int set_fpxregs(unsigned long buf, struct task_struct *tsk) +{ + panic("set_fxpregs"); + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-x86_64/ptrace_user.c b/arch/um/sys-x86_64/ptrace_user.c new file mode 100644 index 000000000000..12e404c6fa46 --- /dev/null +++ b/arch/um/sys-x86_64/ptrace_user.c @@ -0,0 +1,51 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ + +#include <stddef.h> +#include <errno.h> +#include "ptrace_user.h" +#include "user.h" +#include "kern_constants.h" + +int ptrace_getregs(long pid, unsigned long *regs_out) +{ + if(ptrace(PTRACE_GETREGS, pid, 0, regs_out) < 0) + return(-errno); + return(0); +} + +int ptrace_setregs(long pid, unsigned long *regs) +{ + if(ptrace(PTRACE_SETREGS, pid, 0, regs) < 0) + return(-errno); + return(0); +} + +void ptrace_pokeuser(unsigned long addr, unsigned long data) +{ + panic("ptrace_pokeuser"); +} + +#define DS 184 +#define ES 192 +#define __USER_DS 0x2b + +void arch_enter_kernel(void *task, int pid) +{ +} + +void arch_leave_kernel(void *task, int pid) +{ +#ifdef UM_USER_CS + if(ptrace(PTRACE_POKEUSR, pid, CS, UM_USER_CS) < 0) + printk("POKEUSR CS failed"); +#endif + + if(ptrace(PTRACE_POKEUSR, pid, DS, __USER_DS) < 0) + printk("POKEUSR DS failed"); + if(ptrace(PTRACE_POKEUSR, pid, ES, __USER_DS) < 0) + printk("POKEUSR ES failed"); +} diff --git a/arch/um/sys-x86_64/sigcontext.c b/arch/um/sys-x86_64/sigcontext.c new file mode 100644 index 000000000000..c88e64def6f2 --- /dev/null +++ b/arch/um/sys-x86_64/sigcontext.c @@ -0,0 +1,39 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ + +#include <stdio.h> +#include <string.h> +#include <signal.h> +#include "user.h" + +void sc_to_sc(void *to_ptr, void *from_ptr) +{ + struct sigcontext *to = to_ptr, *from = from_ptr; + int size = sizeof(*to); /* + sizeof(struct _fpstate); */ + + memcpy(to, from, size); + if(from->fpstate != NULL) + to->fpstate = (struct _fpstate *) (to + 1); + + to->fpstate = NULL; +} + +unsigned long *sc_sigmask(void *sc_ptr) +{ + struct sigcontext *sc = sc_ptr; + + return(&sc->oldmask); +} + +/* Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-x86_64/signal.c b/arch/um/sys-x86_64/signal.c new file mode 100644 index 000000000000..5bc5a0d796e5 --- /dev/null +++ b/arch/um/sys-x86_64/signal.c @@ -0,0 +1,276 @@ +/* + * Copyright (C) 2003 PathScale, Inc. + * Licensed under the GPL + */ + +#include "linux/stddef.h" +#include "linux/errno.h" +#include "linux/personality.h" +#include "linux/ptrace.h" +#include "asm/current.h" +#include "asm/uaccess.h" +#include "asm/sigcontext.h" +#include "asm/ptrace.h" +#include "asm/arch/ucontext.h" +#include "choose-mode.h" +#include "sysdep/ptrace.h" +#include "frame_kern.h" + +#ifdef CONFIG_MODE_SKAS + +#include "skas.h" + +static int copy_sc_from_user_skas(struct pt_regs *regs, + struct sigcontext *from) +{ + int err = 0; + +#define GETREG(regs, regno, sc, regname) \ + __get_user((regs)->regs.skas.regs[(regno) / sizeof(unsigned long)], \ + &(sc)->regname) + + err |= GETREG(regs, R8, from, r8); + err |= GETREG(regs, R9, from, r9); + err |= GETREG(regs, R10, from, r10); + err |= GETREG(regs, R11, from, r11); + err |= GETREG(regs, R12, from, r12); + err |= GETREG(regs, R13, from, r13); + err |= GETREG(regs, R14, from, r14); + err |= GETREG(regs, R15, from, r15); + err |= GETREG(regs, RDI, from, rdi); + err |= GETREG(regs, RSI, from, rsi); + err |= GETREG(regs, RBP, from, rbp); + err |= GETREG(regs, RBX, from, rbx); + err |= GETREG(regs, RDX, from, rdx); + err |= GETREG(regs, RAX, from, rax); + err |= GETREG(regs, RCX, from, rcx); + err |= GETREG(regs, RSP, from, rsp); + err |= GETREG(regs, RIP, from, rip); + err |= GETREG(regs, EFLAGS, from, eflags); + err |= GETREG(regs, CS, from, cs); + +#undef GETREG + + return(err); +} + +int copy_sc_to_user_skas(struct sigcontext *to, struct _fpstate *to_fp, + struct pt_regs *regs, unsigned long mask) +{ + unsigned long eflags; + int err = 0; + + err |= __put_user(0, &to->gs); + err |= __put_user(0, &to->fs); + +#define PUTREG(regs, regno, sc, regname) \ + __put_user((regs)->regs.skas.regs[(regno) / sizeof(unsigned long)], \ + &(sc)->regname) + + err |= PUTREG(regs, RDI, to, rdi); + err |= PUTREG(regs, RSI, to, rsi); + err |= PUTREG(regs, RBP, to, rbp); + err |= PUTREG(regs, RSP, to, rsp); + err |= PUTREG(regs, RBX, to, rbx); + err |= PUTREG(regs, RDX, to, rdx); + err |= PUTREG(regs, RCX, to, rcx); + err |= PUTREG(regs, RAX, to, rax); + err |= PUTREG(regs, R8, to, r8); + err |= PUTREG(regs, R9, to, r9); + err |= PUTREG(regs, R10, to, r10); + err |= PUTREG(regs, R11, to, r11); + err |= PUTREG(regs, R12, to, r12); + err |= PUTREG(regs, R13, to, r13); + err |= PUTREG(regs, R14, to, r14); + err |= PUTREG(regs, R15, to, r15); + err |= PUTREG(regs, CS, to, cs); /* XXX x86_64 doesn't do this */ + err |= __put_user(current->thread.err, &to->err); + err |= __put_user(current->thread.trap_no, &to->trapno); + err |= PUTREG(regs, RIP, to, rip); + err |= PUTREG(regs, EFLAGS, to, eflags); +#undef PUTREG + + err |= __put_user(mask, &to->oldmask); + err |= __put_user(current->thread.cr2, &to->cr2); + + return(err); +} + +#endif + +#ifdef CONFIG_MODE_TT +int copy_sc_from_user_tt(struct sigcontext *to, struct sigcontext *from, + int fpsize) +{ + struct _fpstate *to_fp, *from_fp; + unsigned long sigs; + int err; + + to_fp = to->fpstate; + from_fp = from->fpstate; + sigs = to->oldmask; + err = copy_from_user(to, from, sizeof(*to)); + to->oldmask = sigs; + return(err); +} + +int copy_sc_to_user_tt(struct sigcontext *to, struct _fpstate *fp, + struct sigcontext *from, int fpsize) +{ + struct _fpstate *to_fp, *from_fp; + int err; + + to_fp = (fp ? fp : (struct _fpstate *) (to + 1)); + from_fp = from->fpstate; + err = copy_to_user(to, from, sizeof(*to)); + return(err); +} + +#endif + +static int copy_sc_from_user(struct pt_regs *to, void __user *from) +{ + int ret; + + ret = CHOOSE_MODE(copy_sc_from_user_tt(UPT_SC(&to->regs), from, + sizeof(struct _fpstate)), + copy_sc_from_user_skas(to, from)); + return(ret); +} + +static int copy_sc_to_user(struct sigcontext *to, struct _fpstate *fp, + struct pt_regs *from, unsigned long mask) +{ + return(CHOOSE_MODE(copy_sc_to_user_tt(to, fp, UPT_SC(&from->regs), + sizeof(*fp)), + copy_sc_to_user_skas(to, fp, from, mask))); +} + +struct rt_sigframe +{ + char *pretcode; + struct ucontext uc; + struct siginfo info; +}; + +#define round_down(m, n) (((m) / (n)) * (n)) + +int setup_signal_stack_si(unsigned long stack_top, int sig, + struct k_sigaction *ka, struct pt_regs * regs, + siginfo_t *info, sigset_t *set) +{ + struct rt_sigframe __user *frame; + struct _fpstate __user *fp = NULL; + int err = 0; + struct task_struct *me = current; + + frame = (struct rt_sigframe __user *) + round_down(stack_top - sizeof(struct rt_sigframe), 16) - 8; + frame -= 128; + + if (!access_ok(VERIFY_WRITE, fp, sizeof(struct _fpstate))) + goto out; + +#if 0 /* XXX */ + if (save_i387(fp) < 0) + err |= -1; +#endif + if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame))) + goto out; + + if (ka->sa.sa_flags & SA_SIGINFO) { + err |= copy_siginfo_to_user(&frame->info, info); + if (err) + goto out; + } + + /* Create the ucontext. */ + err |= __put_user(0, &frame->uc.uc_flags); + err |= __put_user(0, &frame->uc.uc_link); + err |= __put_user(me->sas_ss_sp, &frame->uc.uc_stack.ss_sp); + err |= __put_user(sas_ss_flags(PT_REGS_SP(regs)), + &frame->uc.uc_stack.ss_flags); + err |= __put_user(me->sas_ss_size, &frame->uc.uc_stack.ss_size); + err |= copy_sc_to_user(&frame->uc.uc_mcontext, fp, regs, set->sig[0]); + err |= __put_user(fp, &frame->uc.uc_mcontext.fpstate); + if (sizeof(*set) == 16) { + __put_user(set->sig[0], &frame->uc.uc_sigmask.sig[0]); + __put_user(set->sig[1], &frame->uc.uc_sigmask.sig[1]); + } + else + err |= __copy_to_user(&frame->uc.uc_sigmask, set, + sizeof(*set)); + + /* Set up to return from userspace. If provided, use a stub + already in userspace. */ + /* x86-64 should always use SA_RESTORER. */ + if (ka->sa.sa_flags & SA_RESTORER) + err |= __put_user(ka->sa.sa_restorer, &frame->pretcode); + else + /* could use a vstub here */ + goto out; + + if (err) + goto out; + + /* Set up registers for signal handler */ + { + struct exec_domain *ed = current_thread_info()->exec_domain; + if (unlikely(ed && ed->signal_invmap && sig < 32)) + sig = ed->signal_invmap[sig]; + } + + PT_REGS_RDI(regs) = sig; + /* In case the signal handler was declared without prototypes */ + PT_REGS_RAX(regs) = 0; + + /* This also works for non SA_SIGINFO handlers because they expect the + next argument after the signal number on the stack. */ + PT_REGS_RSI(regs) = (unsigned long) &frame->info; + PT_REGS_RDX(regs) = (unsigned long) &frame->uc; + PT_REGS_RIP(regs) = (unsigned long) ka->sa.sa_handler; + + PT_REGS_RSP(regs) = (unsigned long) frame; + out: + return(err); +} + +long sys_rt_sigreturn(struct pt_regs *regs) +{ + unsigned long sp = PT_REGS_SP(¤t->thread.regs); + struct rt_sigframe __user *frame = + (struct rt_sigframe __user *)(sp - 8); + struct ucontext __user *uc = &frame->uc; + sigset_t set; + + if(copy_from_user(&set, &uc->uc_sigmask, sizeof(set))) + goto segfault; + + sigdelsetmask(&set, ~_BLOCKABLE); + + spin_lock_irq(¤t->sighand->siglock); + current->blocked = set; + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + + if(copy_sc_from_user(¤t->thread.regs, &uc->uc_mcontext)) + goto segfault; + + /* Avoid ERESTART handling */ + PT_REGS_SYSCALL_NR(¤t->thread.regs) = -1; + return(PT_REGS_SYSCALL_RET(¤t->thread.regs)); + + segfault: + force_sig(SIGSEGV, current); + return 0; +} +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-x86_64/syscalls.c b/arch/um/sys-x86_64/syscalls.c new file mode 100644 index 000000000000..68205a03364c --- /dev/null +++ b/arch/um/sys-x86_64/syscalls.c @@ -0,0 +1,186 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ + +#include "linux/linkage.h" +#include "linux/slab.h" +#include "linux/shm.h" +#include "asm/uaccess.h" +#define __FRAME_OFFSETS +#include "asm/ptrace.h" +#include "asm/unistd.h" +#include "asm/prctl.h" /* XXX This should get the constants from libc */ +#include "choose-mode.h" + +asmlinkage long wrap_sys_shmat(int shmid, char __user *shmaddr, int shmflg) +{ + unsigned long raddr; + + return do_shmat(shmid, shmaddr, shmflg, &raddr) ?: (long) raddr; +} + +#ifdef CONFIG_MODE_TT +extern int modify_ldt(int func, void *ptr, unsigned long bytecount); + +long sys_modify_ldt_tt(int func, void *ptr, unsigned long bytecount) +{ + /* XXX This should check VERIFY_WRITE depending on func, check this + * in i386 as well. + */ + if (!access_ok(VERIFY_READ, ptr, bytecount)) + return -EFAULT; + return(modify_ldt(func, ptr, bytecount)); +} +#endif + +#ifdef CONFIG_MODE_SKAS +extern int userspace_pid[]; + +long sys_modify_ldt_skas(int func, void *ptr, unsigned long bytecount) +{ + struct ptrace_ldt ldt; + void *buf; + int res, n; + + buf = kmalloc(bytecount, GFP_KERNEL); + if(buf == NULL) + return(-ENOMEM); + + res = 0; + + switch(func){ + case 1: + case 0x11: + res = copy_from_user(buf, ptr, bytecount); + break; + } + + if(res != 0){ + res = -EFAULT; + goto out; + } + + ldt = ((struct ptrace_ldt) { .func = func, + .ptr = buf, + .bytecount = bytecount }); +#warning Need to look up userspace_pid by cpu + res = ptrace(PTRACE_LDT, userspace_pid[0], 0, (unsigned long) &ldt); + if(res < 0) + goto out; + + switch(func){ + case 0: + case 2: + n = res; + res = copy_to_user(ptr, buf, n); + if(res != 0) + res = -EFAULT; + else + res = n; + break; + } + + out: + kfree(buf); + return(res); +} +#endif + +long sys_modify_ldt(int func, void *ptr, unsigned long bytecount) +{ + return(CHOOSE_MODE_PROC(sys_modify_ldt_tt, sys_modify_ldt_skas, func, + ptr, bytecount)); +} + +#ifdef CONFIG_MODE_TT +extern long arch_prctl(int code, unsigned long addr); + +static long arch_prctl_tt(int code, unsigned long addr) +{ + unsigned long tmp; + long ret; + + switch(code){ + case ARCH_SET_GS: + case ARCH_SET_FS: + ret = arch_prctl(code, addr); + break; + case ARCH_GET_FS: + case ARCH_GET_GS: + ret = arch_prctl(code, (unsigned long) &tmp); + if(!ret) + ret = put_user(tmp, &addr); + break; + default: + ret = -EINVAL; + break; + } + + return(ret); +} +#endif + +#ifdef CONFIG_MODE_SKAS + +static long arch_prctl_skas(int code, unsigned long addr) +{ + long ret = 0; + + switch(code){ + case ARCH_SET_GS: + current->thread.regs.regs.skas.regs[GS_BASE / sizeof(unsigned long)] = addr; + break; + case ARCH_SET_FS: + current->thread.regs.regs.skas.regs[FS_BASE / sizeof(unsigned long)] = addr; + break; + case ARCH_GET_FS: + ret = put_user(current->thread.regs.regs.skas.regs[GS / sizeof(unsigned long)], &addr); + break; + case ARCH_GET_GS: + ret = put_user(current->thread.regs.regs.skas.regs[FS / sizeof(unsigned \ +long)], &addr); + break; + default: + ret = -EINVAL; + break; + } + + return(ret); +} +#endif + +long sys_arch_prctl(int code, unsigned long addr) +{ + return(CHOOSE_MODE_PROC(arch_prctl_tt, arch_prctl_skas, code, addr)); +} + +long sys_clone(unsigned long clone_flags, unsigned long newsp, + void __user *parent_tid, void __user *child_tid) +{ + long ret; + + /* XXX: normal arch do here this pass, and also pass the regs to + * do_fork, instead of NULL. Currently the arch-independent code + * ignores these values, while the UML code (actually it's + * copy_thread) does the right thing. But this should change, + probably. */ + /*if (!newsp) + newsp = UPT_SP(current->thread.regs);*/ + current->thread.forking = 1; + ret = do_fork(clone_flags, newsp, NULL, 0, parent_tid, child_tid); + current->thread.forking = 0; + return(ret); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-x86_64/sysrq.c b/arch/um/sys-x86_64/sysrq.c new file mode 100644 index 000000000000..ddf74691a610 --- /dev/null +++ b/arch/um/sys-x86_64/sysrq.c @@ -0,0 +1,49 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ + +#include "linux/kernel.h" +#include "linux/utsname.h" +#include "linux/module.h" +#include "asm/current.h" +#include "asm/ptrace.h" +#include "sysrq.h" + +void __show_regs(struct pt_regs * regs) +{ + printk("\n"); + print_modules(); + printk("Pid: %d, comm: %.20s %s %s\n", + current->pid, current->comm, print_tainted(), system_utsname.release); + printk("RIP: %04lx:[<%016lx>] ", PT_REGS_CS(regs) & 0xffff, + PT_REGS_RIP(regs)); + printk("\nRSP: %016lx EFLAGS: %08lx\n", PT_REGS_RSP(regs), + PT_REGS_EFLAGS(regs)); + printk("RAX: %016lx RBX: %016lx RCX: %016lx\n", + PT_REGS_RAX(regs), PT_REGS_RBX(regs), PT_REGS_RCX(regs)); + printk("RDX: %016lx RSI: %016lx RDI: %016lx\n", + PT_REGS_RDX(regs), PT_REGS_RSI(regs), PT_REGS_RDI(regs)); + printk("RBP: %016lx R08: %016lx R09: %016lx\n", + PT_REGS_RBP(regs), PT_REGS_R8(regs), PT_REGS_R9(regs)); + printk("R10: %016lx R11: %016lx R12: %016lx\n", + PT_REGS_R10(regs), PT_REGS_R11(regs), PT_REGS_R12(regs)); + printk("R13: %016lx R14: %016lx R15: %016lx\n", + PT_REGS_R13(regs), PT_REGS_R14(regs), PT_REGS_R15(regs)); +} + +void show_regs(struct pt_regs *regs) +{ + __show_regs(regs); + show_trace((unsigned long *) ®s); +} + +/* Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/sys-x86_64/util/Makefile b/arch/um/sys-x86_64/util/Makefile new file mode 100644 index 000000000000..002607980864 --- /dev/null +++ b/arch/um/sys-x86_64/util/Makefile @@ -0,0 +1,10 @@ +# Copyright 2003 - 2004 Pathscale, Inc +# Released under the GPL + +hostprogs-y := mk_sc mk_thread +always := $(hostprogs-y) + +mk_thread-objs := mk_thread_kern.o mk_thread_user.o + +HOSTCFLAGS_mk_thread_kern.o := $(CFLAGS) $(CPPFLAGS) +HOSTCFLAGS_mk_thread_user.o := $(USER_CFLAGS) diff --git a/arch/um/sys-x86_64/util/mk_sc.c b/arch/um/sys-x86_64/util/mk_sc.c new file mode 100644 index 000000000000..c236e213918d --- /dev/null +++ b/arch/um/sys-x86_64/util/mk_sc.c @@ -0,0 +1,58 @@ +/* Copyright (C) 2003 - 2004 PathScale, Inc + * Released under the GPL + */ + +#include <stdio.h> +#include <signal.h> +#include <linux/stddef.h> + +#define SC_OFFSET(name, field) \ + printf("#define " name \ + "(sc) *((unsigned long *) &(((char *) (sc))[%ld]))\n",\ + offsetof(struct sigcontext, field)) + +#define SC_FP_OFFSET(name, field) \ + printf("#define " name \ + "(sc) *((unsigned long *) &(((char *) (SC_FPSTATE(sc)))[%ld]))\n",\ + offsetof(struct _fpstate, field)) + +#define SC_FP_OFFSET_PTR(name, field, type) \ + printf("#define " name \ + "(sc) ((" type " *) &(((char *) (SC_FPSTATE(sc)))[%d]))\n",\ + offsetof(struct _fpstate, field)) + +int main(int argc, char **argv) +{ + SC_OFFSET("SC_RBX", rbx); + SC_OFFSET("SC_RCX", rcx); + SC_OFFSET("SC_RDX", rdx); + SC_OFFSET("SC_RSI", rsi); + SC_OFFSET("SC_RDI", rdi); + SC_OFFSET("SC_RBP", rbp); + SC_OFFSET("SC_RAX", rax); + SC_OFFSET("SC_R8", r8); + SC_OFFSET("SC_R9", r9); + SC_OFFSET("SC_R10", r10); + SC_OFFSET("SC_R11", r11); + SC_OFFSET("SC_R12", r12); + SC_OFFSET("SC_R13", r13); + SC_OFFSET("SC_R14", r14); + SC_OFFSET("SC_R15", r15); + SC_OFFSET("SC_IP", rip); + SC_OFFSET("SC_SP", rsp); + SC_OFFSET("SC_CR2", cr2); + SC_OFFSET("SC_ERR", err); + SC_OFFSET("SC_TRAPNO", trapno); + SC_OFFSET("SC_CS", cs); + SC_OFFSET("SC_FS", fs); + SC_OFFSET("SC_GS", gs); + SC_OFFSET("SC_EFLAGS", eflags); + SC_OFFSET("SC_SIGMASK", oldmask); +#if 0 + SC_OFFSET("SC_ORIG_RAX", orig_rax); + SC_OFFSET("SC_DS", ds); + SC_OFFSET("SC_ES", es); + SC_OFFSET("SC_SS", ss); +#endif + return(0); +} diff --git a/arch/um/sys-x86_64/util/mk_thread_kern.c b/arch/um/sys-x86_64/util/mk_thread_kern.c new file mode 100644 index 000000000000..a281673f02b2 --- /dev/null +++ b/arch/um/sys-x86_64/util/mk_thread_kern.c @@ -0,0 +1,21 @@ +#include "linux/config.h" +#include "linux/stddef.h" +#include "linux/sched.h" + +extern void print_head(void); +extern void print_constant_ptr(char *name, int value); +extern void print_constant(char *name, char *type, int value); +extern void print_tail(void); + +#define THREAD_OFFSET(field) offsetof(struct task_struct, thread.field) + +int main(int argc, char **argv) +{ + print_head(); +#ifdef CONFIG_MODE_TT + print_constant("TASK_EXTERN_PID", "int", THREAD_OFFSET(mode.tt.extern_pid)); +#endif + print_tail(); + return(0); +} + diff --git a/arch/um/sys-x86_64/util/mk_thread_user.c b/arch/um/sys-x86_64/util/mk_thread_user.c new file mode 100644 index 000000000000..7989725568b8 --- /dev/null +++ b/arch/um/sys-x86_64/util/mk_thread_user.c @@ -0,0 +1,30 @@ +#include <stdio.h> + +void print_head(void) +{ + printf("/*\n"); + printf(" * Generated by mk_thread\n"); + printf(" */\n"); + printf("\n"); + printf("#ifndef __UM_THREAD_H\n"); + printf("#define __UM_THREAD_H\n"); + printf("\n"); +} + +void print_constant_ptr(char *name, int value) +{ + printf("#define %s(task) ((unsigned long *) " + "&(((char *) (task))[%d]))\n", name, value); +} + +void print_constant(char *name, char *type, int value) +{ + printf("#define %s(task) *((%s *) &(((char *) (task))[%d]))\n", name, type, + value); +} + +void print_tail(void) +{ + printf("\n"); + printf("#endif\n"); +} diff --git a/arch/um/util/Makefile b/arch/um/util/Makefile new file mode 100644 index 000000000000..e2ab71209f3f --- /dev/null +++ b/arch/um/util/Makefile @@ -0,0 +1,8 @@ +hostprogs-y := mk_task mk_constants +always := $(hostprogs-y) + +mk_task-objs := mk_task_user.o mk_task_kern.o +mk_constants-objs := mk_constants_user.o mk_constants_kern.o + +HOSTCFLAGS_mk_task_kern.o := $(CFLAGS) $(CPPFLAGS) +HOSTCFLAGS_mk_constants_kern.o := $(CFLAGS) $(CPPFLAGS) diff --git a/arch/um/util/mk_constants_kern.c b/arch/um/util/mk_constants_kern.c new file mode 100644 index 000000000000..cdcb1232a1ea --- /dev/null +++ b/arch/um/util/mk_constants_kern.c @@ -0,0 +1,28 @@ +#include "linux/kernel.h" +#include "linux/stringify.h" +#include "linux/time.h" +#include "asm/page.h" + +extern void print_head(void); +extern void print_constant_str(char *name, char *value); +extern void print_constant_int(char *name, int value); +extern void print_tail(void); + +int main(int argc, char **argv) +{ + print_head(); + print_constant_int("UM_KERN_PAGE_SIZE", PAGE_SIZE); + + print_constant_str("UM_KERN_EMERG", KERN_EMERG); + print_constant_str("UM_KERN_ALERT", KERN_ALERT); + print_constant_str("UM_KERN_CRIT", KERN_CRIT); + print_constant_str("UM_KERN_ERR", KERN_ERR); + print_constant_str("UM_KERN_WARNING", KERN_WARNING); + print_constant_str("UM_KERN_NOTICE", KERN_NOTICE); + print_constant_str("UM_KERN_INFO", KERN_INFO); + print_constant_str("UM_KERN_DEBUG", KERN_DEBUG); + + print_constant_int("UM_NSEC_PER_SEC", NSEC_PER_SEC); + print_tail(); + return(0); +} diff --git a/arch/um/util/mk_constants_user.c b/arch/um/util/mk_constants_user.c new file mode 100644 index 000000000000..8f4d7e50be7c --- /dev/null +++ b/arch/um/util/mk_constants_user.c @@ -0,0 +1,28 @@ +#include <stdio.h> + +void print_head(void) +{ + printf("/*\n"); + printf(" * Generated by mk_constants\n"); + printf(" */\n"); + printf("\n"); + printf("#ifndef __UM_CONSTANTS_H\n"); + printf("#define __UM_CONSTANTS_H\n"); + printf("\n"); +} + +void print_constant_str(char *name, char *value) +{ + printf("#define %s \"%s\"\n", name, value); +} + +void print_constant_int(char *name, int value) +{ + printf("#define %s %d\n", name, value); +} + +void print_tail(void) +{ + printf("\n"); + printf("#endif\n"); +} diff --git a/arch/um/util/mk_task_kern.c b/arch/um/util/mk_task_kern.c new file mode 100644 index 000000000000..c218103315ed --- /dev/null +++ b/arch/um/util/mk_task_kern.c @@ -0,0 +1,17 @@ +#include "linux/sched.h" +#include "linux/stddef.h" + +extern void print(char *name, char *type, int offset); +extern void print_ptr(char *name, char *type, int offset); +extern void print_head(void); +extern void print_tail(void); + +int main(int argc, char **argv) +{ + print_head(); + print_ptr("TASK_REGS", "union uml_pt_regs", + offsetof(struct task_struct, thread.regs)); + print("TASK_PID", "int", offsetof(struct task_struct, pid)); + print_tail(); + return(0); +} diff --git a/arch/um/util/mk_task_user.c b/arch/um/util/mk_task_user.c new file mode 100644 index 000000000000..9db849f3f3ac --- /dev/null +++ b/arch/um/util/mk_task_user.c @@ -0,0 +1,30 @@ +#include <stdio.h> + +void print(char *name, char *type, int offset) +{ + printf("#define %s(task) *((%s *) &(((char *) (task))[%d]))\n", name, type, + offset); +} + +void print_ptr(char *name, char *type, int offset) +{ + printf("#define %s(task) ((%s *) &(((char *) (task))[%d]))\n", name, type, + offset); +} + +void print_head(void) +{ + printf("/*\n"); + printf(" * Generated by mk_task\n"); + printf(" */\n"); + printf("\n"); + printf("#ifndef __TASK_H\n"); + printf("#define __TASK_H\n"); + printf("\n"); +} + +void print_tail(void) +{ + printf("\n"); + printf("#endif\n"); +} |