aboutsummaryrefslogtreecommitdiffstats
path: root/tools/perf/scripts/python/check-perf-trace.py (unfollow)
AgeCommit message (Collapse)AuthorFilesLines
2011-03-18MN10300: Allow KGDB to use the MN10300 serial portsDavid Howells1-0/+75
Allow KGDB to use the MN10300 serial ports through the polled I/O interface provided via the TTY/serial layer and the kgdboc driver. This allows the kernel to be started with something like: kgdboc=ttySM0,115200 added to the command line. Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Emulate single stepping in KGDB on MN10300David Howells3-12/+425
Emulate single stepping in KGDB on MN10300 by way of temporary breakpoint insertion. These breakpoints are never actually seen by KGDB, and will overlay KGDB's own breakpoints. The breakpoints are removed by switch_to() and reinstalled on switching back so that if preemption occurs, the preempting task doesn't hit them (though it will still hit KGDB's regular breakpoints). If KGDB is reentered for any reason, then the single step breakpoint is completely erased and must be set again by the debugger. We take advantage of the fact that KGDB will effectively halt all other CPUs whilst this CPU is single-stepping to avoid SMP problems. If the single-stepping task is preempted and killed without KGDB being reinvoked, then the breakpoint(s) will be cleared and KGDB will be jumped back into. Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Generalise kernel debugger kernel halt, reboot or power off hookDavid Howells1-3/+3
Generalise the kernel debugger hook for notification of halt, reboot or power off. This is used by gdbstub to tell the debugger it is exiting. This will be useful for KGDB too. Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18KGDB: Notify GDB of machine halt, reboot or power offDavid Howells2-0/+31
Notify GDB of the machine halting, rebooting or powering off by sending it an exited command (remote protocol command 'W'). This is done by calling: void gdbstub_exit(int status) from the arch's machine_{halt,restart,power_off}() functions with an appropriate exit status to be reported to GDB. Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Use KGDBDavid Howells6-2/+289
2011-03-18MN10300: Create generic kernel debugger hooksDavid Howells14-248/+317
Create generic kernel debugger hooks in the MN10300 arch and make gdbstub use them. This is a preparation for KGDB support. Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Create general kernel debugger cache flushingDavid Howells17-158/+592
Create general kernel debugger cache flushing for MN10300 and get rid of the old stuff that gdbstub was using. Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Introduce a general config option for kernel debugger hooksDavid Howells3-1/+25
Introduce a general config option for kernel debugger hooks so that both gdbstub and kgdb can use it and add a header file for both debuggers to use. Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: The icache invalidate functions should disable the icache firstDavid Howells3-83/+140
The icache invalidate functions should disable the icache on AM33 and wait for it to quiesce before attempting to invalidate it, and should then wait for it to quiesce again before reenabling it, but on AM34 they should invalidate directly. The same goes for the dcache invalidation, but this isn't used much. Whilst we're at it, this can be wrapped in assembler macros to remove duplicate code. The AM33 manual states that: An operation that invalidates the cache, switches the writing mode, or changes the way mode must be performed after disabling the cache, checking the busy bit, and confirming that the cache is not in operation. for the dcache [sec 2.8.3.2.1]. This is not stated so for the icache [sec 2.8.3.1.1] but the example code there suggests that it is. Whilst the AM34 manual states that the cache must be disabled for both the icache [sec 1.8.3.2.1] and the dcache [sec 1.8.3.2.1], the Panasonic hardware engineers say the manual is wrong and that disabling the caches for invalidation is wrong. Furthermore, they say that disabling the caches on the AM34 whilst running an SMP kernel can lead to incoherency between the various CPU caches and should thus be avoided. Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: gdbstub: Restrict single-stepping to non-preemptable non-SMP configsDavid Howells2-2/+18
Restrict single-stepping through the kernel using gdbstub to non-preemptable non-SMP configs as gdbstub has to do software single-stepping by means of temporary breakpoints. Hardware single-stepping is unavailable as Panasonic have not sufficiently documented the interface to it. Software single-stepping through preemptable or SMP kernels runs into problems as it makes it much more likely that the wrong thread will hit the temporary breakpoints. It seems impractical to work around the problem for the most part. It could be possible to make a UP preemptable kernel switch temporary breakpoints in and out in switch_to(). Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Clear ASB2364 peripheral interrupt masks before enabling interruptsDavid Howells2-0/+55
Clear the interrupt mask registers of ASB2364 peripherals before enabling interrupts so that any peripherals that weren't dealt with by the bootloader after a reboot (if there was one) won't cause an interrupt storm when interrupts are first enabled before the drivers are initialised. Also, attempt to reset the peripherals attached to the FPGA. Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Fix the ASB2364 gdbport UART register defsDavid Howells1-12/+12
Fix the ASB2364 gdbport UART register definitions. These registers are actually 2 bytes apart, not 4 (which the ASB2303 and ASB2305 are). Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Fix ASB2364 FPGA register defsDavid Howells1-1/+1
Fix the definition of the ASB2364 FPGA IRQ detect registers. They accidentally got defined to be the same as the mask registers when the patches were being ported to the upstream kernel. Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Select GENERIC_HARDIRQS_NO_DEPRECATEDThomas Gleixner3-5/+6
All chips converted. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Select HAVE_GENERIC_HARDIRQS rather than GENERIC_HARDIRQSDavid Howells1-1/+1
Select HAVE_GENERIC_HARDIRQS rather than GENERIC_HARDIRQS in MN10300's main Kconfig file to avoid this warning: warning: (MN10300) selects GENERIC_HARDIRQS which has unmet direct dependencies (HAVE_GENERIC_HARDIRQS) Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Convert ASB2364 FPGA irq_chip to new functionsThomas Gleixner1-13/+13
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Convert ipi irq_chip to new functionsThomas Gleixner1-8/+22
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Convert serial irq_chip to new functionsThomas Gleixner1-5/+10
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Convert cpu irq_chips to new functionsThomas Gleixner1-29/+33
Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Remove unused mn10300_intc_* functionsThomas Gleixner2-30/+0
No users. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Remove stale irq_chip.end - V2Thomas Gleixner1-1/+0
irq_chip.end is obsolete with the removal of __do_IRQ(). Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Use clockevents_calc_mult_shift()Thomas Gleixner3-26/+3
Use clockevents_calc_mult_shift() instead of the homebrewn function in mn10300/kernel/time.c. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Acked-by: John Stultz <johnstul@us.ibm.com> Cc: Koichi Yasutake <yasutake.koichi@jp.panasonic.com> Cc: Mark Salter <msalter@redhat.com> Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Use clocksource_register_hz()Thomas Gleixner3-21/+1
clocksource_register_hz() calculates the shift/mult pair for the clocksource. Remove the mn10300 duplicate implementation. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Acked-by: John Stultz <johnstul@us.ibm.com> Cc: Koichi Yasutake <yasutake.koichi@jp.panasonic.com> Cc: Mark Salter <msalter@redhat.com> Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-18MN10300: Remove stale codeThomas Gleixner3-67/+2
mn10300 implements clocksource and clockevents and selects them unconditionally in Kconfig. Remove the stale code which seems to be a leftover of the conversion. Cleanup the configuration switches as well. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Acked-by: John Stultz <johnstul@us.ibm.com> Cc: Koichi Yasutake <yasutake.koichi@jp.panasonic.com> Cc: Mark Salter <msalter@redhat.com> Signed-off-by: David Howells <dhowells@redhat.com>
2011-03-17smp_call_function_interrupt: use typedef and %pfMilton Miller1-8/+8
Use the newly added smp_call_func_t in smp_call_function_interrupt for the func variable, and make the comment above the WARN more assertive and explicit. Also, func is a function pointer and does not need an offset, so use %pf not %pS. Signed-off-by: Milton Miller <miltonm@bga.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-17smp_call_function_many: handle concurrent clearing of maskMilton Miller1-3/+10
Mike Galbraith reported finding a lockup ("perma-spin bug") where the cpumask passed to smp_call_function_many was cleared by other cpu(s) while a cpu was preparing its call_data block, resulting in no cpu to clear the last ref and unlock the block. Having cpus clear their bit asynchronously could be useful on a mask of cpus that might have a translation context, or cpus that need a push to complete an rcu window. Instead of adding a BUG_ON and requiring yet another cpumask copy, just detect the race and handle it. Note: arch_send_call_function_ipi_mask must still handle an empty cpumask because the data block is globally visible before the that arch callback is made. And (obviously) there are no guarantees to which cpus are notified if the mask is changed during the call; only cpus that were online and had their mask bit set during the whole call are guaranteed to be called. Reported-by: Mike Galbraith <efault@gmx.de> Reported-by: Jan Beulich <JBeulich@novell.com> Acked-by: Jan Beulich <jbeulich@novell.com> Cc: stable@kernel.org Signed-off-by: Milton Miller <miltonm@bga.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-17call_function_many: add missing orderingMilton Miller1-13/+33
Paul McKenney's review pointed out two problems with the barriers in the 2.6.38 update to the smp call function many code. First, a barrier that would force the func and info members of data to be visible before their consumption in the interrupt handler was missing. This can be solved by adding a smp_wmb between setting the func and info members and setting setting the cpumask; this will pair with the existing and required smp_rmb ordering the cpumask read before the read of refs. This placement avoids the need a second smp_rmb in the interrupt handler which would be executed on each of the N cpus executing the call request. (I was thinking this barrier was present but was not). Second, the previous write to refs (establishing the zero that we the interrupt handler was testing from all cpus) was performed by a third party cpu. This would invoke transitivity which, as a recient or concurrent addition to memory-barriers.txt now explicitly states, would require a full smp_mb(). However, we know the cpumask will only be set by one cpu (the data owner) and any preivous iteration of the mask would have cleared by the reading cpu. By redundantly writing refs to 0 on the owning cpu before the smp_wmb, the write to refs will follow the same path as the writes that set the cpumask, which in turn allows us to keep the barrier in the interrupt handler a smp_rmb instead of promoting it to a smp_mb (which will be be executed by N cpus for each of the possible M elements on the list). I moved and expanded the comment about our (ab)use of the rcu list primitives for the concurrent walk earlier into this function. I considered moving the first two paragraphs to the queue list head and lock, but felt it would have been too disconected from the code. Cc: Paul McKinney <paulmck@linux.vnet.ibm.com> Cc: stable@kernel.org (2.6.32 and later) Signed-off-by: Milton Miller <miltonm@bga.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-17call_function_many: fix list delete vs add raceMilton Miller1-7/+13
Peter pointed out there was nothing preventing the list_del_rcu in smp_call_function_interrupt from running before the list_add_rcu in smp_call_function_many. Fix this by not setting refs until we have gotten the lock for the list. Take advantage of the wmb in list_add_rcu to save an explicit additional one. I tried to force this race with a udelay before the lock & list_add and by mixing all 64 online cpus with just 3 random cpus in the mask, but was unsuccessful. Still, inspection shows a valid race, and the fix is a extension of the existing protection window in the current code. Cc: stable@kernel.org (v2.6.32 and later) Reported-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Milton Miller <miltonm@bga.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-17mm: PageBuddy and mapcount robustnessAndrea Arcangeli2-4/+11
Change the _mapcount value indicating PageBuddy from -2 to -128 for more robusteness against page_mapcount() undeflows. Use reset_page_mapcount instead of __ClearPageBuddy in bad_page to ignore the previous retval of PageBuddy(). Signed-off-by: Andrea Arcangeli <aarcange@redhat.com> Reported-by: Hugh Dickins <hughd@google.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-17arch/tile: support newer binutils assembler shift semanticsChris Metcalf3-9/+29
This change supports building the kernel with newer binutils where a shift of greater than the word size is no longer interpreted silently as modulo the word size, but instead generates a warning. Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>
2011-03-17RPC: killing RPC tasks races fixedStanislav Kinsbursky1-1/+3
RPC task RPC_TASK_QUEUED bit is set must be checked before trying to wake up task rpc_killall_tasks() because task->tk_waitqueue can not be set (equal to NULL). Also, as Trond Myklebust mentioned, such approach (instead of checking tk_waitqueue to NULL) allows us to "optimise away the call to rpc_wake_up_queued_task() altogether for those tasks that aren't queued". Here is an example of dereferencing of tk_waitqueue equal to NULL: CPU 0 CPU 1 CPU 2 -------------------- --------------------- -------------------------- nfs4_run_open_task rpc_run_task rpc_execute rpc_set_active rpc_make_runnable (waiting) rpc_async_schedule nfs4_open_prepare nfs_wait_on_sequence nfs_umount_begin rpc_killall_tasks rpc_wake_up_task rpc_wake_up_queued_task spin_lock(tk_waitqueue == NULL) BUG() rpc_sleep_on spin_lock(&q->lock) __rpc_sleep_on task->tk_waitqueue = q Signed-off-by: Stanislav Kinsbursky <skinsbursky@openvz.org> Cc: stable@kernel.org Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2011-03-17xprt: remove redundant checkj223yang@asset.uwaterloo.ca1-1/+1
remove redundant check. Signed-off-by: Jinqiu Yang <crindy646@gmail.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2011-03-17SUNRPC: Convert struct rpc_xprt to use atomic_t countersTrond Myklebust2-10/+9
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
2011-03-17SUNRPC: Ensure we always run the tk_callback before tk_actionTrond Myklebust1-8/+6
This fixes a race in which the task->tk_callback() puts the rpc_task to sleep, setting a new callback. Under certain circumstances, the current code may end up executing the task->tk_action before it gets round to the callback. Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com> Cc: stable@kernel.org
2011-03-17KVM: unbreak userspace that does not sets tss addressGleb Natapov1-0/+13
Commit 6440e5967bc broke old userspaces that do not set tss address before entering vcpu. Unbreak it by setting tss address to a safe value on the first vcpu entry. New userspaces should set tss address, so print warning in case it doesn't. Signed-off-by: Gleb Natapov <gleb@redhat.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2011-03-17KVM: MMU: cleanup pte write pathXiao Guangrong3-56/+32
This patch does: - call vcpu->arch.mmu.update_pte directly - use gfn_to_pfn_atomic in update_pte path The suggestion is from Avi. Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Signed-off-by: Avi Kivity <avi@redhat.com>
2011-03-17KVM: MMU: introduce a common function to get no-dirty-logged slotXiao Guangrong1-20/+17
Cleanup the code of pte_prefetch_gfn_to_memslot and mapping_level_dirty_bitmap Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Signed-off-by: Avi Kivity <avi@redhat.com>
2011-03-17KVM: fix rcu usage in init_rmode_* functionsXiao Guangrong1-4/+8
fix: [ 3494.671786] stack backtrace: [ 3494.671789] Pid: 10527, comm: qemu-system-x86 Not tainted 2.6.38-rc6+ #23 [ 3494.671790] Call Trace: [ 3494.671796] [] ? lockdep_rcu_dereference+0x9d/0xa5 [ 3494.671826] [] ? kvm_memslots+0x6b/0x73 [kvm] [ 3494.671834] [] ? gfn_to_memslot+0x16/0x4f [kvm] [ 3494.671843] [] ? gfn_to_hva+0x16/0x27 [kvm] [ 3494.671851] [] ? kvm_write_guest_page+0x31/0x83 [kvm] [ 3494.671861] [] ? kvm_clear_guest_page+0x1a/0x1c [kvm] [ 3494.671867] [] ? vmx_set_tss_addr+0x83/0x122 [kvm_intel] and: [ 8328.789599] stack backtrace: [ 8328.789601] Pid: 18736, comm: qemu-system-x86 Not tainted 2.6.38-rc6+ #23 [ 8328.789603] Call Trace: [ 8328.789609] [] ? lockdep_rcu_dereference+0x9d/0xa5 [ 8328.789621] [] ? kvm_memslots+0x6b/0x73 [kvm] [ 8328.789628] [] ? gfn_to_memslot+0x16/0x4f [kvm] [ 8328.789635] [] ? gfn_to_hva+0x16/0x27 [kvm] [ 8328.789643] [] ? kvm_write_guest_page+0x31/0x83 [kvm] [ 8328.789699] [] ? kvm_clear_guest_page+0x1a/0x1c [kvm] [ 8328.789713] [] ? vmx_create_vcpu+0x316/0x3c8 [kvm_intel] Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Signed-off-by: Avi Kivity <avi@redhat.com>
2011-03-17KVM: fix kvmclock regression due to missing clock updateNikola Ciprich1-1/+1
commit 387b9f97750444728962b236987fbe8ee8cc4f8c moved kvm_request_guest_time_update(vcpu), breaking 32bit SMP guests using kvm-clock. Fix this by moving (new) clock update function to proper place. Signed-off-by: Nikola Ciprich <nikola.ciprich@linuxbox.cz> Acked-by: Zachary Amsden <zamsden@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
2011-03-17KVM: emulator: Fix permission checking in io permission bitmapGleb Natapov1-3/+2
Currently if io port + len crosses 8bit boundary in io permission bitmap the check may allow IO that otherwise should not be allowed. The patch fixes that. Signed-off-by: Gleb Natapov <gleb@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
2011-03-17KVM: emulator: Fix io permission checking for 64bit guestGleb Natapov3-21/+35
Current implementation truncates upper 32bit of TR base address during IO permission bitmap check. The patch fixes this. Reported-and-tested-by: Francis Moreau <francis.moro@gmail.com> Signed-off-by: Gleb Natapov <gleb@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
2011-03-17KVM: SVM: Load %gs earlier if CONFIG_X86_32_LAZY_GS=nAvi Kivity1-0/+5
With CONFIG_CC_STACKPROTECTOR, we need a valid %gs at all times, so disable lazy reload and do an eager reload immediately after the vmexit. Reported-by: IVAN ANGELOV <ivangotoy@gmail.com> Acked-By: Joerg Roedel <joerg.roedel@amd.com> Signed-off-by: Avi Kivity <avi@redhat.com>
2011-03-17KVM: x86: Remove useless regs_page pointer from kvm_lapicTakuya Yoshikawa3-7/+5
Access to this page is mostly done through the regs member which holds the address to this page. The exceptions are in vmx_vcpu_reset() and kvm_free_lapic() and these both can easily be converted to using regs. Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp> Signed-off-by: Avi Kivity <avi@redhat.com>
2011-03-17KVM: improve comment on rcu use in irqfd_deassignMichael S. Tsirkin1-2/+3
The RCU use in kvm_irqfd_deassign is tricky: we have rcu_assign_pointer but no synchronize_rcu: synchronize_rcu is done by kvm_irq_routing_update which we share a spinlock with. Fix up a comment in an attempt to make this clearer. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
2011-03-17KVM: MMU: remove unused macrosXiao Guangrong2-8/+0
These macros are not used, so removed Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Signed-off-by: Avi Kivity <avi@redhat.com>
2011-03-17KVM: MMU: cleanup page alloc and freeXiao Guangrong1-5/+5
Using __get_free_page instead of alloc_page and page_address, using free_page instead of __free_page and virt_to_page Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Signed-off-by: Avi Kivity <avi@redhat.com>
2011-03-17KVM: MMU: do not record gfn in kvm_mmu_pte_writeXiao Guangrong3-7/+2
No need to record the gfn to verifier the pte has the same mode as current vcpu, it's because we only speculatively update the pte only if the pte and vcpu have the same mode Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Signed-off-by: Avi Kivity <avi@redhat.com>
2011-03-17KVM: MMU: move mmu pages calculated out of mmu lockXiao Guangrong1-5/+5
kvm_mmu_calculate_mmu_pages need to walk all memslots and it's protected by kvm->slots_lock, so move it out of mmu spinlock Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Signed-off-by: Avi Kivity <avi@redhat.com>
2011-03-17KVM: MMU: set spte accessed bit properlyXiao Guangrong1-1/+1
Set spte accessed bit only if guest_initiated == 1 that means the really accessed Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Signed-off-by: Avi Kivity <avi@redhat.com>
2011-03-17KVM: MMU: fix kvm_mmu_slot_remove_write_access dropping intermediate W bitsXiao Guangrong1-2/+7
Only remove write access in the last sptes. Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.