summaryrefslogtreecommitdiffstats
path: root/sys/arch/sgi/hpc (follow)
Commit message (Collapse)AuthorAgeFilesLines
* spellingjsg2021-03-113-6/+6
|
* Change users of IFQ_SET_MAXLEN() and IFQ_IS_EMPTY() to use the "new" API.patrick2020-07-101-2/+2
| | | | ok dlg@ tobhe@
* Use <dev/clock_subr.h> in dsclock(4).visa2020-05-211-34/+43
| | | | Looks good to miod@
* Use <dev/clock_subr.h> in dpclock(4).visa2020-05-211-31/+42
| | | | Looks good to miod@
* Nuke unnecessary abstraction 'scsi_minphys()' which just callskrw2020-02-051-2/+2
| | | | | | 'minphys()'. Just use & check for NULL instead, since 'minphys()' is always called on the code path ([cd|sd|st]minphys) that calls physio().
* Use a consistant idiom/format when declaring scsi_adapter structureskrw2020-01-231-5/+2
| | | | | | | | | | | in drivers. Terse one liners, NULLs instead of 0's, explicitly specify all members, etc. Nuke #ifdef notyet blocks related to the scsi_adapter in aic. No intentional functional change. ok tedu@
* regen (missed this when I did the colemak update)abieber2019-05-131-2/+5
|
* prune files.* entries that refer to files not in treejsg2018-02-141-9/+1
| | | | ok krw@ mpi@
* Delete unnecessary <sys/file.h> includesguenther2017-12-301-2/+1
| | | | ok millert@ krw@
* Rename Debugger() into db_enter().mpi2017-04-301-2/+2
| | | | | | | Using a name with the 'db_' prefix makes it invisible from the dynamic profiler. ok deraadt@, kettenis@, visa@
* Unifdef KGDB.mpi2017-04-301-7/+5
| | | | | | It doesn't compile und hasn't been working during the last decade. ok kettenis@, deraadt@
* Fix logic in the driver preventing it to receive broadcast framesmpi2017-03-081-2/+2
| | | | | | | | before configuring an address. Make dhclient(8) work on sq(4). Problem reported by Frank Scheiner, diff from miod@
* move counting if_opackets next to counting if_obytes in if_enqueue.dlg2017-01-221-3/+1
| | | | | | | this means packets are consistently counted in one place, unlike the many and various ways that drivers thought they should do it. ok mpi@ deraadt@
* m_free() and m_freem() test for NULL. Simplify callers which had their ownjsg2016-11-291-3/+2
| | | | | | NULL tests. ok mpi@
* regenjca2016-08-311-2/+6
| | | | Reminded by miod
* Use the new input functions of wsmouse in mouse and touchscreen drivers.bru2016-06-051-2/+2
| | | | ok stsp@ kettenis@
* G/C IFQ_SET_READY().mpi2016-04-131-2/+1
|
* comment typommcc2015-12-101-2/+2
|
* No trailers has been the default and only option for 20 years, yet sometedu2015-12-081-2/+2
| | | | | | | drivers still set IFF_NOTRAILERS while others do not. Remove all usage of the flag from the drivers which in ancient times used it (and the modern drivers which blindly copied it from those drivers of yore). suggested by guenther. ok mpi
* replace IFF_OACTIVE manipulation with mpsafe operations.dlg2015-11-251-7/+8
| | | | | | | | | | | | | | | | | | | | | | | | there are two things shared between the network stack and drivers in the send path: the send queue and the IFF_OACTIVE flag. the send queue is now protected by a mutex. this diff makes the oactive functionality mpsafe too. IFF_OACTIVE is part of if_flags. there are two problems with that. firstly, if_flags is a short and we dont have any MI atomic operations to manipulate a short. secondly, while we could make the IFF_OACTIVE operates mpsafe, all changes to other flags would have to be made safe at the same time, otherwise a read-modify-write cycle on their updates could clobber the oactive change. instead, this moves the oactive mark into struct ifqueue and provides an API for changing it. there's ifq_set_oactive, ifq_clr_oactive, and ifq_is_oactive. these are modelled on ifsq_set_oactive, ifsq_clr_oactive, and ifsq_is_oactive in dragonflybsd. this diff includes changes to all the drivers manipulating IFF_OACTIVE to now use the ifsq_{set,clr_is}_oactive API too. ok kettenis@ mpi@ jmatthew@ deraadt@
* You only need <net/if_dl.h> if you're using LLADDR() or a sockaddr_dl.mpi2015-11-241-2/+1
|
* The only network driver needing <net/if_types.h> is upl(4) for IFT_OTHER.mpi2015-11-241-2/+1
|
* error: too many arguments to function 'ifq_deq_begin'dlg2015-11-211-2/+2
| | | | found by deraadt@
* shuffle struct ifqueue so in flight mbufs are protected by a mutex.dlg2015-11-201-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | the code is refactored so the IFQ macros call newly implemented ifq functions. the ifq code is split so each discipline (priq and hfsc in our case) is an opaque set of operations that the common ifq code can call. the common code does the locking, accounting (ifq_len manipulation), and freeing of the mbuf if the disciplines enqueue function rejects it. theyre kind of like bufqs in the block layer with their fifo and nscan disciplines. the new api also supports atomic switching of disciplines at runtime. the hfsc setup in pf_ioctl.c has been tweaked to build a complete hfsc_if structure which it attaches to the send queue in a single operation, rather than attaching to the interface up front and building up a list of queues. the send queue is now mutexed, which raises the expectation that packets can be enqueued or purged on one cpu while another cpu is dequeueing them in a driver for transmission. a lot of drivers use IFQ_POLL to peek at an mbuf and attempt to fit it on the ring before committing to it with a later IFQ_DEQUEUE operation. if the mbuf gets freed in between the POLL and DEQUEUE operations, fireworks will ensue. to avoid this, the ifq api introduces ifq_deq_begin, ifq_deq_rollback, and ifq_deq_commit. ifq_deq_begin allows a driver to take the ifq mutex and get a reference to the mbuf they wish to try and tx. if there's space, they can ifq_deq_commit it to remove the mbuf and release the mutex. if there's no space, ifq_deq_rollback simply releases the mutex. this api was developed to make updating the drivers using IFQ_POLL easy, instead of having to do significant semantic changes to avoid POLL that we cannot test on all the hardware. the common code has been tested pretty hard, and all the driver modifications are straightforward except for de(4). if that breaks it can be dealt with later. ok mpi@ jmatthew@
* arp_ifinit() is no longer needed.mpi2015-10-251-4/+1
|
* Go back to the previous approach when managing individual HPC DMA descriptors:miod2015-09-184-124/+108
| | | | | | | | | | | | | | | | | | | provide again an optional storage for a copy of the descriptor in the `sync' (fetch) function, and use the returned address afterwards. On IP22 systems (in the broader sense of the term, thus IP20/IP22/IP24), descriptors will remain in uncached memory and no local copies need to be made. On IP28 systems, descriptors will remain in cached memory (so as to avoid switching to `slow mode'), but a local copy will be performed with the necessary cache eviction work, so that speculative code execution on R10000 will not touch the real descriptor. With this in place, all the explicit descriptor cache operations in if_sq, some of them being redundant or operating on the wrong number of descriptors, can be removed, with the HPC DMA wrappers taking care of doing the right thing. Tested on IP22 and IP28. IP26 still unhappy but no worse than before.
* Fix more ifmedia64 fallout in the kernel. It's hiding everywhere...stsp2015-09-141-2/+2
|
* Give up trying to map DMA descriptor in uncached memory on ECC flavours of themiod2015-09-055-91/+92
| | | | | | | | | | IP22 motherboard (IP26, IP28). Instead, do not ask for a BUS_DMA_COHERENT mapping, but perform explicit cache operations. This removes the need for the memory controller to switch between `fast' and `slow' mode every time a DMA descriptor is updated. Tested on IP22 and IP28.
* Increment if_ipackets in if_input().mpi2015-06-241-3/+1
| | | | | | | Note that pseudo-drivers not using if_input() are not affected by this conversion. ok mikeb@, kettenis@, claudio@, dlg@
* Follow the recent pckbc@isa changes and always establish all the necessarymiod2015-05-241-32/+8
| | | | | | | | | interrupts at pckbc attach time, and get rid of the `intr_establish' pckbc callback. Tested on hppa (gsckbc) and sgi (pckbc@hpc); not tested on sparc64 (pckbc@ebus) but this attachment was already behaving this way and its intr_establish callback was an empty function.
* Remove all audio format conversion code from the kernel (btw holdingratchov2015-05-111-2/+2
| | | | | | | | | the kernel_lock), as we already do better conversions in user-mode. Yet, no need for every single driver to fiddle with the conversion code as they are done transparently by common MI code. With help from armani and miod, support from mpi ok armani@
* Print irq informations in pckbc_set_inputhandler().mpi2015-05-041-3/+2
| | | | ok miod@
* Convert to if_input(), tested by miod@ and sebastia@.mpi2015-03-291-8/+4
|
* unifdef INETtedu2014-12-221-3/+1
|
* Avoid extra space in dmesgmiod2014-12-071-2/+2
|
* Probe the keyboard for its dip switches also if it is the console keyboard,miod2014-12-071-76/+128
| | | | | so that we can select the proper layout automagically and have it apply even at UKC time.
* Check for the `oscillator failure' condition, and do not register as a reliablemiod2014-10-111-10/+26
| | | | time source in that case.
* Remove redundant setting of if_mtu which ether_ifattach() alreadybrad2014-08-271-2/+1
| | | | | | takes care of. ok dlg@
* <netinet/in_systm.h> is no longer needed.mpi2014-07-221-7/+2
|
* It's init as a process that's special, not init's original thread.guenther2014-07-111-2/+2
| | | | | | Remember initprocess instead of initproc. ok matthew@ blambert@
* Standardize xfer byte counts to ssize_t rather than a mix of size_t and intmiod2014-06-271-6/+8
| | | | | | | | | (we need to make them signed to spot controller overruns), and fix format strings accordingly. While there, make sure every runtime printf is prefixed by either the complete target information, if available, or at least the driver name with the proper instance number - supported systems with > 1 wdsc are quite common.
* Regenmiod2014-05-221-119/+553
|
* Yet another evil awk script (not perl, I'm not in LibreSSL mode at the moment)miod2014-05-222-0/+347
| | | | | | | | to convert the PS/2 keyboard layouts to SGI serial keyboard layouts. Only the subset of layouts known to have existed (as listed in IRIX's <sys/kbd.h> header) get converted, and specific flavours (iopener, dec LK) are ignored as well. `nodead' flavours, when existing, are preserved.
* Extend the keyboard communication routines to be able to work in polling mode;miod2014-05-221-39/+150
| | | | | | | | | | | | use this to read the DIP switches from the keyboard at attach time. Change the state machine to allow for a `DIP switch prefix' scan code to be received while we are not attempting to read the DIP switches, for the `international' key (not found in regular us layouts, documented is the `GERlessthan' key in sgi's keyboard(7) manual page) will return the aforementioned scancode, instead of the one documented in the manual. Thanks to sebastia@ for lending me his german layout keyboard.
* Format string fixes and removal of -Wno-format for sgi. Based upon anmiod2014-05-191-2/+2
| | | | initial diff from jasper@
* rename wd33c93 to wd33c93ctrl (but keep the filenames as is) so we canjasper2014-04-181-2/+2
| | | | | | add attributes to it later; as wd33c93 is not a valid device name. ok miod@
* Attempt to make user changes of keyboard layout a bit more `sticky' on wsmuxmiod2014-01-261-2/+2
| | | | | | | | | | | | | | | | | kernels: - keyboard drivers will now tell wskbd if the keyboard layout they ask for is a default value, or a value they are 100% sure of (either because your kernel has a XXXKBD_LAYOUT option, or because the driver can tell the keyboard layout, e.g. by the country code on USB keyboards which provide it, such as Sun's) - when attaching a keyboard with a non-default layout, the layout will become the default layout of the mux for new keyboard attachments if the mux doesn't have a layout set already. - when changing the keyboard layout of a particular keyboard with an ioctl (i.e. using kbd(8) or wsconsctl(8)), the layout will become the default layout of the mux for new keyboard attachments. ok mpi@
* Convert wskbd_set_mixervolume() to use a task internally instead ofmpi2013-11-041-5/+3
| | | | | | being called in a workq. ok kettenis@
* Fix lies in comments, and apply some KNF and unused or duplicate prototypemiod2013-10-211-2/+2
| | | | removals.
* Correctly probe for the Ethernet chip on HPC 1.5 expansion boards. Gets rid ofmiod2013-09-281-5/+9
| | | | | false sq positives ("sq not configured" since rejected by the sq driver) in dmesg, for SCSI HPC boards.
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.