04 Jul, 2011
2 commits
-
…frederic/random-tracing into perf/core
03 Jul, 2011
6 commits
-
rbp is used in SAVE_ARGS_IRQ to save the old stack pointer
in order to restore it later in ret_from_intr.It is convenient because we save its value in the irq regs
and it's easily restored using the leave instruction.However this is a kind of abuse of the frame pointer which
role is to help unwinding the kernel by chaining frames
together, each node following the return address to the
previous frame.But although we are breaking the frame by changing the stack
pointer, there is no preceding return address before the new
frame. Hence using the frame pointer to link the two stacks
breaks the stack unwinders that find a random value instead of
a return address here.There is no workaround that can work in every case. We are using
the fixup_bp_irq_link() function to dereference that abused frame
pointer in the case of non nesting interrupt (which means stack
changed).
But that doesn't fix the case of interrupts that don't change the
stack (but we still have the unconditional frame link), which is
the case of hardirq interrupting softirq. We have no way to detect
this transition so the frame irq link is considered as a real frame
pointer and the return address is dereferenced but it is still a
spurious one.There are two possible results of this: either the spurious return
address, a random stack value, luckily belongs to the kernel text
and then the unwinding can continue and we just have a weird entry
in the stack trace. Or it doesn't belong to the kernel text and
unwinding stops there.This is the reason why stacktraces (including perf callchains) on
irqs that interrupted softirqs don't work very well.To solve this, we don't save the old stack pointer on rbp anymore
but we save it to a scratch register that we push on the new
stack and that we pop back later on irq return.This preserves the whole frame chain without spurious return addresses
in the middle and drops the need for the horrid fixup_bp_irq_link()
workaround.And finally irqs that interrupt softirq are sanely unwinded.
Before:
99.81% perf [kernel.kallsyms] [k] perf_pending_event
|
--- perf_pending_event
irq_work_run
smp_irq_work_interrupt
irq_work_interrupt
|
|--41.60%-- __read
| |
| |--99.90%-- create_worker
| | bench_sched_messaging
| | cmd_bench
| | run_builtin
| | main
| | __libc_start_main
| --0.10%-- [...]After:
1.64% swapper [kernel.kallsyms] [k] perf_pending_event
|
--- perf_pending_event
irq_work_run
smp_irq_work_interrupt
irq_work_interrupt
|
|--95.00%-- arch_irq_work_raise
| irq_work_queue
| __perf_event_overflow
| perf_swevent_overflow
| perf_swevent_event
| perf_tp_event
| perf_trace_softirq
| __do_softirq
| call_softirq
| do_softirq
| irq_exit
| |
| |--73.68%-- smp_apic_timer_interrupt
| | apic_timer_interrupt
| | |
| | |--96.43%-- amd_e400_idle
| | | cpu_idle
| | | start_secondarySigned-off-by: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Thomas Gleixner
Cc: H. Peter Anvin
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Jan Beulich -
The unwinder backlink in interrupt entry is very useless.
It's actually not part of the stack frame chain and thus is
never used.Signed-off-by: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Thomas Gleixner
Cc: H. Peter Anvin
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Jan Beulich -
Just for clarity in the code. Have a first block that handles
the frame pointer and a separate one that handles pt_regs
pointer and its use.Signed-off-by: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Thomas Gleixner
Cc: H. Peter Anvin
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Jan Beulich -
The save_regs function that saves the regs on low level
irq entry is complicated because of the fact it changes
its stack in the middle and also because it manipulates
data allocated in the caller frame and accesses there
are directly calculated from callee rsp value with the
return address in the middle of the way.This complicates the static stack offsets calculation and
require more dynamic ones. It also needs a save/restore
of the function's return address.To simplify and optimize this, turn save_regs() into a
macro.Signed-off-by: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Thomas Gleixner
Cc: H. Peter Anvin
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Jan Beulich -
When regs are passed to dump_stack(), we fetch the frame
pointer from the regs but the stack pointer is taken from
the current frame.Thus the frame and stack pointers may not come from the same
context. For example this can result in the unwinder to
think the context is in irq, due to the current value of
the stack, but the frame pointer coming from the regs points
to a frame from another place. It then tries to fix up
the irq link but ends up dereferencing a random frame
pointer that doesn't belong to the irq stack:[ 9131.706906] ------------[ cut here ]------------
[ 9131.707003] WARNING: at arch/x86/kernel/dumpstack_64.c:129 dump_trace+0x2aa/0x330()
[ 9131.707003] Hardware name: AMD690VM-FMH
[ 9131.707003] Perf: bad frame pointer = 0000000000000005 in callchain
[ 9131.707003] Modules linked in:
[ 9131.707003] Pid: 1050, comm: perf Not tainted 3.0.0-rc3+ #181
[ 9131.707003] Call Trace:
[ 9131.707003] [] warn_slowpath_common+0x7a/0xb0
[ 9131.707003] [] warn_slowpath_fmt+0x41/0x50
[ 9131.707003] [] ? bad_to_user+0x6d/0x10be
[ 9131.707003] [] dump_trace+0x2aa/0x330
[ 9131.707003] [] ? native_sched_clock+0x13/0x50
[ 9131.707003] [] perf_callchain_kernel+0x54/0x70
[ 9131.707003] [] perf_prepare_sample+0x19f/0x2a0
[ 9131.707003] [] __perf_event_overflow+0x16c/0x290
[ 9131.707003] [] ? __perf_event_overflow+0x130/0x290
[ 9131.707003] [] ? native_sched_clock+0x13/0x50
[ 9131.707003] [] ? sched_clock+0x9/0x10
[ 9131.707003] [] ? T.375+0x15/0x90
[ 9131.707003] [] ? trace_hardirqs_on_caller+0x64/0x180
[ 9131.707003] [] ? trace_hardirqs_off+0xd/0x10
[ 9131.707003] [] perf_event_overflow+0x14/0x20
[ 9131.707003] [] perf_swevent_hrtimer+0x11c/0x130
[ 9131.707003] [] ? error_exit+0x51/0xb0
[ 9131.707003] [] __run_hrtimer+0x83/0x1e0
[ 9131.707003] [] ? perf_event_overflow+0x20/0x20
[ 9131.707003] [] hrtimer_interrupt+0x106/0x250
[ 9131.707003] [] ? trace_hardirqs_off_thunk+0x3a/0x3c
[ 9131.707003] [] smp_apic_timer_interrupt+0x53/0x90
[ 9131.707003] [] apic_timer_interrupt+0x13/0x20
[ 9131.707003] [] ? error_exit+0x51/0xb0
[ 9131.707003] [] ? error_exit+0x4c/0xb0
[ 9131.707003] ---[ end trace b2560d4876709347 ]---Fix this by simply taking the stack pointer from regs->sp
when regs are provided.Signed-off-by: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Thomas Gleixner
Cc: H. Peter Anvin
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo -
In order to prepare for fetching the stack pointer from the
regs when possible in dump_trace() instead of taking the
local one, save the current stack pointer in perf live regs saving.Signed-off-by: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Thomas Gleixner
Cc: H. Peter Anvin
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
01 Jul, 2011
13 commits
-
The patch a8b0ca17b80e ("perf: Remove the nmi parameter from the swevent
and overflow interface") missed a spot in the ppc hw_breakpoint code,
fix this up so things compile again.Reported-by: Ingo Molnar
Cc: Anton Blanchard
Cc: Eric B Munson
Signed-off-by: Peter Zijlstra
Link: http://lkml.kernel.org/n/tip-09pfip95g88s70iwkxu6nnbt@git.kernel.org
Signed-off-by: Ingo Molnar -
The v1 PMU does not have any fixed counters. Using the v2 constraints,
which do have fixed counters, causes an additional choice to be present
in the weight calculation, but not when actually scheduling the event,
leading to an event being not scheduled at all.Signed-off-by: Avi Kivity
Signed-off-by: Peter Zijlstra
Link: http://lkml.kernel.org/r/1309362157-6596-3-git-send-email-avi@redhat.com
Signed-off-by: Ingo Molnar -
The perf_event overflow handler does not receive any caller-derived
argument, so many callers need to resort to looking up the perf_event
in their local data structure. This is ugly and doesn't scale if a
single callback services many perf_events.Fix by adding a context parameter to perf_event_create_kernel_counter()
(and derived hardware breakpoints APIs) and storing it in the perf_event.
The field can be accessed from the callback as event->overflow_handler_context.
All callers are updated.Signed-off-by: Avi Kivity
Signed-off-by: Peter Zijlstra
Link: http://lkml.kernel.org/r/1309362157-6596-2-git-send-email-avi@redhat.com
Signed-off-by: Ingo Molnar -
Add a NODE level to the generic cache events which is used to measure
local vs remote memory accesses. Like all other cache events, an
ACCESS is HIT+MISS, if there is no way to distinguish between reads
and writes do reads only etc..The below needs filling out for !x86 (which I filled out with
unsupported events).I'm fairly sure ARM can leave it like that since it doesn't strike me as
an architecture that even has NUMA support. SH might have something since
it does appear to have some NUMA bits.Sparc64, PowerPC and MIPS certainly want a good look there since they
clearly are NUMA capable.Signed-off-by: Peter Zijlstra
Cc: David Miller
Cc: Anton Blanchard
Cc: David Daney
Cc: Deng-Cheng Zhu
Cc: Paul Mundt
Cc: Will Deacon
Cc: Robert Richter
Cc: Stephane Eranian
Link: http://lkml.kernel.org/r/1303508226.4865.8.camel@laptop
Signed-off-by: Ingo Molnar -
Since the OFFCORE registers are fully symmetric, try the other one
when the specified one is already in use.Signed-off-by: Peter Zijlstra
Link: http://lkml.kernel.org/r/1306141897.18455.8.camel@twins
Signed-off-by: Ingo Molnar -
This patch adds Intel Sandy Bridge offcore_response support by
providing the low-level constraint table for those events.On Sandy Bridge, there are two offcore_response events. Each uses
its own dedictated extra register. But those registers are NOT shared
between sibling CPUs when HT is on unlike Nehalem/Westmere. They are
always private to each CPU. But they still need to be controlled within
an event group. All events within an event group must use the same
value for the extra MSR. That's not controlled by the second patch in
this series.Furthermore on Sandy Bridge, the offcore_response events have NO
counter constraints contrary to what the official documentation
indicates, so drop the events from the contraint table.Signed-off-by: Stephane Eranian
Signed-off-by: Peter Zijlstra
Link: http://lkml.kernel.org/r/20110606145712.GA7304@quad
Signed-off-by: Ingo Molnar -
The validate_group() function needs to validate events with
extra shared regs. Within an event group, only events with
the same value for the extra reg can co-exist. This was not
checked by validate_group() because it was missing the
shared_regs logic.This patch changes the allocation of the fake cpuc used for
validation to also point to a fake shared_regs structure such
that group events be properly testing.It modifies __intel_shared_reg_get_constraints() to use
spin_lock_irqsave() to avoid lockdep issues.Signed-off-by: Stephane Eranian
Signed-off-by: Peter Zijlstra
Link: http://lkml.kernel.org/r/20110606145708.GA7279@quad
Signed-off-by: Ingo Molnar -
This patch improves the code managing the extra shared registers
used for offcore_response events on Intel Nehalem/Westmere. The
idea is to use static allocation instead of dynamic allocation.
This simplifies greatly the get and put constraint routines for
those events.The patch also renames per_core to shared_regs because the same
data structure gets used whether or not HT is on. When HT is
off, those events still need to coordination because they use
a extra MSR that has to be shared within an event group.Signed-off-by: Stephane Eranian
Signed-off-by: Peter Zijlstra
Link: http://lkml.kernel.org/r/20110606145703.GA7258@quad
Signed-off-by: Ingo Molnar -
Since only samples call perf_output_sample() its much saner (and more
correct) to put the sample logic in there than in the
perf_output_begin()/perf_output_end() pair.Saves a useless argument, reduces conditionals and shrinks
struct perf_output_handle, win!Signed-off-by: Peter Zijlstra
Link: http://lkml.kernel.org/n/tip-2crpvsx3cqu67q3zqjbnlpsc@git.kernel.org
Signed-off-by: Ingo Molnar -
The nmi parameter indicated if we could do wakeups from the current
context, if not, we would set some state and self-IPI and let the
resulting interrupt do the wakeup.For the various event classes:
- hardware: nmi=0; PMI is in fact an NMI or we run irq_work_run from
the PMI-tail (ARM etc.)
- tracepoint: nmi=0; since tracepoint could be from NMI context.
- software: nmi=[0,1]; some, like the schedule thing cannot
perform wakeups, and hence need 0.As one can see, there is very little nmi=1 usage, and the down-side of
not using it is that on some platforms some software events can have a
jiffy delay in wakeup (when arch_irq_work_raise isn't implemented).The up-side however is that we can remove the nmi parameter and save a
bunch of conditionals in fast paths.Signed-off-by: Peter Zijlstra
Cc: Michael Cree
Cc: Will Deacon
Cc: Deng-Cheng Zhu
Cc: Anton Blanchard
Cc: Eric B Munson
Cc: Heiko Carstens
Cc: Paul Mundt
Cc: David S. Miller
Cc: Frederic Weisbecker
Cc: Jason Wessel
Cc: Don Zickus
Link: http://lkml.kernel.org/n/tip-agjev8eu666tvknpb3iaj0fg@git.kernel.org
Signed-off-by: Ingo Molnar -
Due to restriction and specifics of Netburst PMU we need a separated
event for NMI watchdog. In particular every Netburst event
consumes not just a counter and a config register, but also an
additional ESCR register.Since ESCR registers are grouped upon counters (i.e. if ESCR is occupied
for some event there is no room for another event to enter until its
released) we need to pick up the "least" used ESCR (or the most available
one) for nmi-watchdog purposes -- so MSR_P4_CRU_ESCR2/3 was chosen.With this patch nmi-watchdog and perf top should be able to run simultaneously.
Signed-off-by: Cyrill Gorcunov
CC: Lin Ming
CC: Arnaldo Carvalho de Melo
CC: Frederic Weisbecker
Tested-and-reviewed-by: Don Zickus
Tested-and-reviewed-by: Stephane Eranian
Signed-off-by: Peter Zijlstra
Link: http://lkml.kernel.org/r/20110623124918.GC13050@sun
Signed-off-by: Ingo Molnar -
Commit e360adbe29 ("irq_work: Add generic hardirq context
callbacks") fouled up the Alpha bit, not properly naming the
arch specific function that raises the 'self-IPI'.Signed-off-by: Peter Zijlstra
Cc: Michael Cree
Cc: stable@kernel.org # 37+
Link: http://lkml.kernel.org/n/tip-gukh0txmql2l4thgrekzzbfy@git.kernel.org
Signed-off-by: Ingo Molnar -
Commit e360adbe29 ("irq_work: Add generic hardirq context
callbacks") fouled up the ppc bit, not properly naming the
arch specific function that raises the 'self-IPI'.Cc: Huang Ying
Cc: Benjamin Herrenschmidt
Cc: Anton Blanchard
Cc: Eric B Munson
Cc: stable@kernel.org # 37+
Signed-off-by: Peter Zijlstra
Link: http://lkml.kernel.org/n/tip-eg0aqien8p1aqvzu9dft6dtv@git.kernel.org
Signed-off-by: Ingo Molnar
28 Jun, 2011
2 commits
-
To make SLUB work on UML we need this_cpu_cmpxchg from
asm-generic/percpu.h.Signed-off-by: Richard Weinberger
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
commit 21a3c96 uses node_start/end_pfn(nid) for detection start/end
of nodes. But, it's not defined in linux/mmzone.h but defined in
/arch/???/include/mmzone.h which is included only under
CONFIG_NEED_MULTIPLE_NODES=y.Then, we see
mm/page_cgroup.c: In function 'page_cgroup_init':
mm/page_cgroup.c:308: error: implicit declaration of function 'node_start_pfn'
mm/page_cgroup.c:309: error: implicit declaration of function 'node_end_pfn'So, fixiing page_cgroup.c is an idea...
But node_start_pfn()/node_end_pfn() is a very generic macro and
should be implemented in the same manner for all archs.
(m32r has different implementation...)This patch removes definitions of node_start/end_pfn() in each archs
and defines a unified one in linux/mmzone.h. It's not under
CONFIG_NEED_MULTIPLE_NODES, now.A result of macro expansion is here (mm/page_cgroup.c)
for !NUMA
start_pfn = ((&contig_page_data)->node_start_pfn);
end_pfn = ({ pg_data_t *__pgdat = (&contig_page_data); __pgdat->node_start_pfn + __pgdat->node_spanned_pages;});for NUMA (x86-64)
start_pfn = ((node_data[nid])->node_start_pfn);
end_pfn = ({ pg_data_t *__pgdat = (node_data[nid]); __pgdat->node_start_pfn + __pgdat->node_spanned_pages;});Changelog:
- fixed to avoid using "nid" twice in node_end_pfn() macro.Reported-and-acked-by: Randy Dunlap
Reported-and-tested-by: Ingo Molnar
Acked-by: Mel Gorman
Signed-off-by: KAMEZAWA Hiroyuki
Signed-off-by: Linus Torvalds
27 Jun, 2011
2 commits
-
* 'fixes' of master.kernel.org:/home/rmk/linux-2.6-arm:
ARM: pm: ensure ARMv7 CPUs save and restore the TLS register
ARM: pm: proc-v7: fix missing struct processor pointers for suspend code
ARM: 6969/1: plat-iop: fix build error
ARM: 6961/1: zImage: Add build-time check for correctly-sized proc_type entries
ARM: SMP: wait for CPU to be marked active
ARM: 6963/1: Thumb-2: Relax relocation requirements for non-function symbols
ARM: 6962/1: mach-h720x: fix build error
ARM: 6959/1: SMP build fix for entry-macro-multi.S -
* 'for-linus' of git://git390.marist.edu/pub/scm/linux-2.6:
[S390] allow setting of upper 32 bit in smp_ctl_set_bit
[S390] hwsampler: Set a sane default sampling rate
[S390] s390: enforce HW limits for the initial sampling rate
[S390] kvm-s390: fix kconfig dependencies
24 Jun, 2011
3 commits
-
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6:
x86/PCI/ACPI: fix type mismatch
PCI: fix new kernel-doc warning
PCI: Fix warning in drivers/pci/probe.c on sparc64 -
Ensure that the TLS register is saved and restored over a suspend
cycle, so that userspace programs don't see a corrupted TLS value.Tested-by: Kevin Hilman
Acked-by: Jean Pihet
Signed-off-by: Russell King -
Add the missing suspend/resume pointers for the suspend code. This
is needed when building for multiple CPUs.Tested-by: Kevin Hilman
Acked-by: Jean Pihet
Signed-off-by: Russell King
22 Jun, 2011
5 commits
-
The bit shift operation in smp_ctl_set_bit does not specify the type
of the shifted bit so integer is used as default. Therefore it is not
possible to set bits in the upper 32 bit of the control register if
the kernel runs in 64 bit mode. Fix this by specifying the type as
unsigned long.Signed-off-by: Jan Glauber
Signed-off-by: Martin Schwidefsky -
The sampling interval for the hardware sampler is specified in cycles.
(see SA23-2260-01 The Load-Program-Parameter and the CPU-Measurement
Facilities)
The current default value will therefore result in millions of samples.
This patch changes the default sampling interval to 4M, which will
result in ~1500 samples per second on a z196 reducing the overhead
of sampling.Signed-off-by: Christian Borntraeger
Signed-off-by: Martin Schwidefsky -
On specific configurations with hwsampler opcontrol --start returns an
error on "echo 1 >/dev/oprofile/enable". Turns out that the hw sampling
interval is not checked against the hardware limits.Signed-off-by: Christian Borntraeger
Signed-off-by: Martin Schwidefsky -
A user can create the Kconfig combination !VIRTUALIZATION, S390_GUEST
which results in the following warnings:warning: (S390_GUEST) selects VIRTIO which has unmet direct dependencies (VIRTUALIZATION)
warning: (S390_GUEST && VIRTIO_PCI && VIRTIO_BALLOON) selects VIRTIO_RING which has unmet direct dependencies (VIRTUALIZATION && VIRTIO)
warning: (S390_GUEST) selects VIRTIO which has unmet direct dependencies (VIRTUALIZATION)
warning: (S390_GUEST && VIRTIO_PCI && VIRTIO_BALLOON) selects VIRTIO_RING which has unmet direct dependencies (VIRTUALIZATION && VIRTIO)S390_GUEST has to select VIRTUALIZATION before selecting VIRTIO and
friends.Reported-by: Jan Glauber
Signed-off-by: Christian Borntraeger
Signed-off-by: Martin Schwidefsky -
MN10300's asm/uaccess.h needs to #include linux/kernel.h to get might_sleep()
otherwise it fails to build on MN10300 allyesconfig. This fails in a few
places with messages like the following:In file included from security/keys/trusted.c:14:
include/linux/uaccess.h: In function '__copy_from_user_nocache':
include/linux/uaccess.h:52: error: implicit declaration of function 'might_sleep'Signed-off-by: David Howells
Signed-off-by: Linus Torvalds
21 Jun, 2011
6 commits
-
The iop13xx_defconfig didn't build since the platform code uses
defines from . Simply add the include so it
compiles.Signed-off-by: Linus Walleij
Signed-off-by: Russell King -
It is easy to mis-maintain the proc_types table such that the
entries become wrongly-sized and misaligned when the kernel is
built in Thumb-2.This patch adds an assembly-time check which will turn most common
size/alignment mistakes in this table into build failures, to avoid
having to debug the boot-time kernel hang which would happen if the
resulting kernel were actually booted.Signed-off-by: Dave Martin
Acked-by: Nicolas Pitre
Signed-off-by: Russell King -
When we bring a CPU online, we should wait for it to become active
before entering the idle thread, so we know that the scheduler and
thread migration is going to work.Signed-off-by: Russell King
-
* 'msm-fix' of git://codeaurora.org/quic/kernel/davidb/linux-msm:
msm: timer: Fix DGT rate on 8960 and 8660
msm: timer: compensate for timer shift in msm_read_timer_count
msm: timer: Fix SMP build error -
The DGT runs at 27 MHz divided by 4 on 8660 and 8960.
Signed-off-by: Stephen Boyd
Signed-off-by: David Brown -
* 'stable/bug.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:
xen/setup: Fix for incorrect xen_extra_mem_start.
xen: When calling power_off, don't call the halt function.
xen: Fix compile warning when CONFIG_SMP is not defined.
xen: support CONFIG_MAXSMP
xen: partially revert "xen: set max_pfn_mapped to the last pfn mapped"
20 Jun, 2011
1 commit
-
* 'kvm-updates/3.0' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: Fix register corruption in pvclock_scale_delta
KVM: MMU: fix opposite condition in mapping_level_dirty_bitmap
KVM: VMX: do not overwrite uptodate vcpu->arch.cr3 on KVM_SET_SREGS
KVM: MMU: Fix build warnings in walk_addr_generic()