- 25 May, 2009 11 commits
-
-
Peter Zijlstra authored
remove the x86 specific interrupt throttle Signed-off-by:
Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: John Kacur <jkacur@redhat.com> LKML-Reference: <20090525153931.616671838@chello.nl> Signed-off-by:
Ingo Molnar <mingo@elte.hu>
-
Peter Zijlstra authored
Expose the INV and EDGE bits of the PMU to raw configs. Signed-off-by:
-
Peter Zijlstra authored
Ingo noticed that cpu counters had 0 context switches, even though there was plenty scheduling on the cpu. Signed-off-by:
-
Peter Zijlstra authored
Fail fork() when we fail inheritance for some reason (-ENOMEM most likely). Signed-off-by:
-
Peter Zijlstra authored
Paul noted that the new ptcrl() didn't work on child counters. Reported-by:
Paul Mackerras <paulus@samba.org> Signed-off-by:
-
Peter Zijlstra authored
extra_config_len isn't used for anything, remove it. Signed-off-by:
-
Peter Zijlstra authored
Fix: $ perf-top fatal: cannot handle -top internally Signed-off-by:
-
Ingo Molnar authored
This output: $ perf stat -e 0:1:k -e 0:1:u ./hello Performance counter stats for './hello': 140131 instructions (events) 1906968 instructions (events) Is quite confusing - as :k means "user instructions", :u means "kernel instructions". Flip them around - as the 'exclude' property is not intuitive in the flag naming. Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: John Kacur <jkacur@redhat.com> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <new-submission> Signed-off-by: -
Ingo Molnar authored
Initialize a task's perfcounters (inherit from parent, etc.) after the child task's scheduler fields have been initialized already. [ Impact: cleanup ] Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: John Kacur <jkacur@redhat.com> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <new-submission> Signed-off-by: -
Mike Galbraith authored
Iterate over the symbol table once per display interval, and copy/sort/tally/decay only those symbols which are active. Before: top - 10:14:53 up 4:08, 17 users, load average: 1.17, 1.53, 1.49 Tasks: 273 total, 5 running, 268 sleeping, 0 stopped, 0 zombie Cpu(s): 6.9%us, 38.2%sy, 0.0%ni, 19.9%id, 0.0%wa, 0.0%hi, 35.0%si, 0.0%st PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ P COMMAND 28504 root 20 0 1044 260 164 S 58 0.0 0:04.19 2 netserver 28499 root 20 0 1040 412 316 R 51 0.0 0:04.15 0 netperf 28500 root 20 0 1040 408 316 R 50 0.0 0:04.14 1 netperf 28503 root 20 0 1044 260 164 S 50 0.0 0:04.01 1 netserver 28501 root 20 0 1044 260 164 S 49 0.0 0:03.99 0 netserver 28502 root 20 0 1040 412 316 S 43 0.0 0:03.96 2 netperf 28468 root 20 0 1892m 325m 972 S 16 10.8 0:10.50 3 perf 28467 root 20 0 1892m 325m 972 R 2 10.8 0:00.72 3 perf After: top - 10:16:30 up 4:10, 17 users, load average: 2.27, 1.88, 1.62 Tasks: 273 total, 6 running, 267 sleeping, 0 stopped, 0 zombie Cpu(s): 2.5%us, 39.7%sy, 0.0%ni, 24.6%id, 0.0%wa, 0.0%hi, 33.3%si, 0.0%st PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ P COMMAND 28590 root 20 0 1040 412 316 S 54 0.0 0:07.85 2 netperf 28589 root 20 0 1044 260 164 R 54 0.0 0:07.84 0 netserver 28588 root 20 0 1040 412 316 R 50 0.0 0:07.89 1 netperf 28591 root 20 0 1044 256 164 S 50 0.0 0:07.82 1 netserver 28587 root 20 0 1040 408 316 R 47 0.0 0:07.61 0 netperf 28592 root 20 0 1044 260 164 R 47 0.0 0:07.85 2 netserver 28378 root 20 0 8732 1300 860 R 2 0.0 0:01.81 3 top 28577 root 20 0 1892m 165m 972 R 2 5.5 0:00.48 3 perf 28578 root 20 0 1892m 165m 972 S 2 5.5 0:00.04 3 perf [ Impact: optimization ] Signed-off-by:
Mike Galbraith <efault@gmx.de> Acked-by:
-
Ingo Molnar authored
NR_CPUS and NR_COUNTERS goes up quadratic ... 1024x4096 was far too ambitious upper limit - go for 256x256 which is still plenty. [ Impact: reduce perf tool memory consumption ] Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: John Kacur <jkacur@redhat.com> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <new-submission> Signed-off-by:
-
- 24 May, 2009 4 commits
-
-
Ingo Molnar authored
In a default 'perf top' run the tool will create a counter for each online CPU. With enough CPUs this will eventually exhaust the default limit. So scale it up with the number of online CPUs. Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: John Kacur <jkacur@redhat.com> Cc: Mike Galbraith <efault@gmx.de> LKML-Reference: <new-submission> Signed-off-by: -
Mike Galbraith authored
c6eb13 increased stack usage such that perf-top now croaks on startup. Take event_array and mmap_array off the stack to prevent segfault on boxen with smallish ulimit -s setting. Signed-off-by:
-
Peter Zijlstra authored
now that pctrl() no longer disables other people's counters, remove the PMU cache code that deals with that. Signed-off-by:
-
Peter Zijlstra authored
Instead of en/dis-abling all counters acting on a particular task, en/dis- able all counters we created. [ v2: fix crash on first counter enable ] Signed-off-by:
-
- 23 May, 2009 5 commits
-
-
Peter Zijlstra authored
Use perf_counter_remove_from_context() to remove counters from the context. Signed-off-by:
-
Peter Zijlstra authored
recent userspace (F11) seems to already include the linux/unistd.h bits which means we cannot include the version in the kernel sources due to the header guards being the same. Ensure we include the kernel version first. Signed-off-by:
-
Peter Zijlstra authored
Ensure we're consistent with the context locks. context->mutex context->lock list_{add,del}_counter(); so that either lock is sufficient to stabilize the context. Signed-off-by: -
Peter Zijlstra authored
s/counter->mutex/counter->child_mutex/ and make sure its only used to protect child_list. The usage in __perf_counter_exit_task() doesn't appear to be problematic since ctx->mutex also covers anything related to fd tear-down. Signed-off-by:
-
Peter Zijlstra authored
We call perf_adjust_freq() from perf_counter_task_tick() which is is called under the rq->lock causing lock recursion. However, it's no longer required to be called under the rq->lock, so remove it from under it. Also, fix up some related comments. Signed-off-by:
-
- 22 May, 2009 4 commits
-
-
Ingo Molnar authored
I tried to run with 300 active counters and the tools bailed out because our limit was at 64. So increase the counter limit to 1024 and the CPU limit to 4096. Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> LKML-Reference: <new-submission> Signed-off-by: -
Ingo Molnar authored
Update the !CONFIG_PERF_COUNTERS prototype too, for perf_counter_task_sched_out(). [ Impact: build fix ] Signed-off-by:
-
Paul Mackerras authored
When monitoring a process and its descendants with a set of inherited counters, we can often get the situation in a context switch where both the old (outgoing) and new (incoming) process have the same set of counters, and their values are ultimately going to be added together. In that situation it doesn't matter which set of counters are used to count the activity for the new process, so there is really no need to go through the process of reading the hardware counters and updating the old task's counters and then setting up the PMU for the new task. This optimizes the context switch in this situation. Instead of scheduling out the perf_counter_context for the old task and scheduling in the new context, we simply transfer the old context to the new task and keep using it without interruption. The new context gets transferred to the old task. This means that both tasks still have a valid perf_counter_context, so no special case is introduced when the old task gets scheduled in again, either on this CPU or another CPU. The equivalence of contexts is detected by keeping a pointer in each cloned context pointing to the context it was cloned from. To cope with the situation where a context is changed by adding or removing counters after it has been cloned, we also keep a generation number on each context which is incremented every time a context is changed. When a context is cloned we take a copy of the parent's generation number, and two cloned contexts are equivalent only if they have the same parent and the same generation number. In order that the parent context pointer remains valid (and is not reused), we increment the parent context's reference count for each context cloned from it. Since we don't have individual fds for the counters in a cloned context, the only thing that can make two clones of a given parent different after they have been cloned is enabling or disabling all counters with prctl. To account for this, we keep a count of the number of enabled counters in each context. Two contexts must have the same number of enabled counters to be considered equivalent. Here are some measurements of the context switch time as measured with the lat_ctx benchmark from lmbench, comparing the times obtained with and without this patch series: -----Unmodified----- With this patch series Counters: none 2 HW 4H+4S none 2 HW 4H+4S 2 processes: Average 3.44 6.45 11.24 3.12 3.39 3.60 St dev 0.04 0.04 0.13 0.05 0.17 0.19 8 processes: Average 6.45 8.79 14.00 5.57 6.23 7.57 St dev 1.27 1.04 0.88 1.42 1.46 1.42 32 processes: Average 5.56 8.43 13.78 5.28 5.55 7.15 St dev 0.41 0.47 0.53 0.54 0.57 0.81 The numbers are the mean and standard deviation of 20 runs of lat_ctx. The "none" columns are lat_ctx run directly without any counters. The "2 HW" columns are with lat_ctx run under perfstat, counting cycles and instructions. The "4H+4S" columns are lat_ctx run under perfstat with 4 hardware counters and 4 software counters (cycles, instructions, cache references, cache misses, task clock, context switch, cpu migrations, and page faults). [ Impact: performance optimization of counter context-switches ] Signed-off-by:
-
Paul Mackerras authored
This replaces the struct perf_counter_context in the task_struct with a pointer to a dynamically allocated perf_counter_context struct. The main reason for doing is this is to allow us to transfer a perf_counter_context from one task to another when we do lazy PMU switching in a later patch. This has a few side-benefits: the task_struct becomes a little smaller, we save some memory because only tasks that have perf_counters attached get a perf_counter_context allocated for them, and we can remove the inclusion of <linux/perf_counter.h> in sched.h, meaning that we don't end up recompiling nearly everything whenever perf_counter.h changes. The perf_counter_context structures are reference-counted and freed when the last reference is dropped. A context can have references from its task and the counters on its task. Counters can outlive the task so it is possible that a context will be freed well after its task has exited. Contexts are allocated on fork if the parent had a context, or otherwise the first time that a per-task counter is created on a task. In the latter case, we set the context pointer in the task struct locklessly using an atomic compare-and-exchange operation in case we raced with some other task in creating a context for the subject task. This also removes the task pointer from the perf_counter struct. The task pointer was not used anywhere and would make it harder to move a context from one task to another. Anything that needed to know which task a counter was attached to was already using counter->ctx->task. The __perf_counter_init_context function moves up in perf_counter.c so that it can be called from find_get_context, and now initializes the refcount, but is otherwise unchanged. We were potentially calling list_del_counter twice: once from __perf_counter_exit_task when the task exits and once from __perf_counter_remove_from_context when the counter's fd gets closed. This adds a check in list_del_counter so it doesn't do anything if the counter has already been removed from the lists. Since perf_counter_task_sched_in doesn't do anything if the task doesn't have a context, and leaves cpuctx->task_ctx = NULL, this adds code to __perf_install_in_context to set cpuctx->task_ctx if necessary, i.e. in the case where the current task adds the first counter to itself and thus creates a context for itself. This also adds similar code to __perf_counter_enable to handle a similar situation which can arise when the counters have been disabled using prctl; that also leaves cpuctx->task_ctx = NULL. [ Impact: refactor counter context management to prepare for new feature ] Signed-off-by:
-
- 20 May, 2009 5 commits
-
-
Ingo Molnar authored
Disable the PMU globally before removing a counter from a context. This fixes the following lockup: [22081.741922] ------------[ cut here ]------------ [22081.746668] WARNING: at arch/x86/kernel/cpu/perf_counter.c:803 intel_pmu_handle_irq+0x9b/0x24e() [22081.755624] Hardware name: X8DTN [22081.758903] perfcounters: irq loop stuck! [22081.762985] Modules linked in: [22081.766136] Pid: 11082, comm: perf Not tainted 2.6.30-rc6-tip #226 [22081.772432] Call Trace: [22081.774940] <NMI> [<ffffffff81019aed>] ? intel_pmu_handle_irq+0x9b/0x24e [22081.781993] [<ffffffff81019aed>] ? intel_pmu_handle_irq+0x9b/0x24e [22081.788368] [<ffffffff8104505c>] ? warn_slowpath_common+0x77/0xa3 [22081.794649] [<ffffffff810450d3>] ? warn_slowpath_fmt+0x40/0x45 [22081.800696] [<ffffffff81019aed>] ? intel_pmu_handle_irq+0x9b/0x24e [22081.807080] [<ffffffff814d1a72>] ? perf_counter_nmi_handler+0x3f/0x4a [22081.813751] [<ffffffff814d2d09>] ? notifier_call_chain+0x58/0x86 [22081.819951] [<ffffffff8105b250>] ? notify_die+0x2d/0x32 [22081.825392] [<ffffffff814d1414>] ? do_nmi+0x8e/0x242 [22081.830538] [<ffffffff814d0f0a>] ? nmi+0x1a/0x20 [22081.835342] [<ffffffff8117e102>] ? selinux_file_free_security+0x0/0x1a [22081.842105] [<ffffffff81018793>] ? x86_pmu_disable_counter+0x15/0x41 [22081.848673] <<EOE>> [<ffffffff81018f3d>] ? x86_pmu_disable+0x86/0x103 [22081.855512] [<ffffffff8108fedd>] ? __perf_counter_remove_from_context+0x0/0xfe [22081.862926] [<ffffffff8108fcbc>] ? counter_sched_out+0x30/0xce [22081.868909] [<ffffffff8108ff36>] ? __perf_counter_remove_from_context+0x59/0xfe [22081.876382] [<ffffffff8106808a>] ? smp_call_function_single+0x6c/0xe6 [22081.882955] [<ffffffff81091b96>] ? perf_release+0x86/0x14c [22081.888600] [<ffffffff810c4c84>] ? __fput+0xe7/0x195 [22081.893718] [<ffffffff810c213e>] ? filp_close+0x5b/0x62 [22081.899107] [<ffffffff81046a70>] ? put_files_struct+0x64/0xc2 [22081.905031] [<ffffffff8104841a>] ? do_exit+0x1e2/0x6ef [22081.910360] [<ffffffff814d0a60>] ? _spin_lock_irqsave+0x9/0xe [22081.916292] [<ffffffff8104898e>] ? do_group_exit+0x67/0x93 [22081.921953] [<ffffffff810489cc>] ? sys_exit_group+0x12/0x16 [22081.927759] [<ffffffff8100baab>] ? system_call_fastpath+0x16/0x1b [22081.934076] ---[ end trace 3a3936ce3e1b4505 ]--- And could potentially also fix the lockup reported by Marcelo Tosatti. Also, print more debug info in case of a detected lockup. [ Impact: fix lockup ] Reported-by:
Marcelo Tosatti <mtosatti@redhat.com> Acked-by:
-
Peter Zijlstra authored
Avoid a function call for !group counters by directly calling the counter function. [ Impact: micro-optimize the code ] Signed-off-by:
-
Peter Zijlstra authored
Currently we call hrtimer_cancel() unconditionally on disable of time based software counters. Avoid when possible. [ Impact: micro-optimize the code ] Signed-off-by:
-
Peter Zijlstra authored
For the dynamic irq_period code, log whenever we change the period so that analyzing code can normalize the event flow. [ Impact: add new feature to allow more precise profiling ] Signed-off-by:
-
Peter Zijlstra authored
Instead of disabling RR scheduling of the counters, use a different list that does not get rotated to iterate the counters on inheritance. [ Impact: cleanup, optimization ] Signed-off-by:
-
- 19 May, 2009 2 commits
-
-
Ingo Molnar authored
Context rotation should not occur when we are in the middle of walking the counter list when inheriting counters ... [ Impact: fix occasionally incorrect perf stat results ] Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Signed-off-by: -
Ingo Molnar authored
Fix counter lifetime bugs which explain the crashes reported by Marcelo Tosatti and Arnaldo Carvalho de Melo. The new rule is: flushing + freeing is only done for a task's own counters, never for other tasks. [ Impact: fix crashes/lockups with inherited counters ] Reported-by:
Arnaldo Carvalho de Melo <acme@redhat.com> Reported-by:
-
- 18 May, 2009 3 commits
-
-
Ingo Molnar authored
We have to set up the LVT entry only at counter init time, not at every switch-in time. There's friction between NMI and non-NMI use here - we'll probably remove the per counter configurability of it - but until then, dont slow down things ... [ Impact: micro-optimization ] Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Srivatsa Vaddagiri <vatsa@in.ibm.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> LKML-Reference: <new-submission> Signed-off-by: -
Paul Mackerras authored
Commit 9e35ad38 ("perf_counter: Rework the perf counter disable/enable") added code to the powerpc hw_perf_enable (renamed from hw_perf_restore) to test cpuhw->disabled and return immediately if it is not set (i.e. if the PMU is already enabled). Unfortunately the test got added before cpuhw was initialized, resulting in an oops the first time hw_perf_enable got called. This fixes it by moving the initialization of cpuhw to before cpuhw->disabled is tested. [ Impact: fix oops-causing bug on powerpc ] Signed-off-by:
-
Ingo Molnar authored
Merge reason: this branch was on an -rc4 base, merge it up to -rc6 to get the latest upstream fixes. Signed-off-by:
-
- 17 May, 2009 4 commits
-
-
Ingo Molnar authored
The quirk to irq_period unearthed an unrobustness we had in the hw_counter initialization sequence: we left irq_period at 0, which was then quirked up to 2 ... which then generated a _lot_ of interrupts during 'perf stat' runs, slowed them down and skewed the counter results in general. Initialize irq_period to the maximum instead. [ Impact: fix perf stat results ] Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Srivatsa Vaddagiri <vatsa@in.ibm.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Signed-off-by: -
Ingo Molnar authored
Flushing counters in __exit_signal() with irqs disabled is not a good idea as perf_counter_exit_task() acquires mutexes. So flush it before acquiring the tasklist lock. (Note, we still need a fix for when the PID has been unhashed.) [ Impact: fix crash with inherited counters ] Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Srivatsa Vaddagiri <vatsa@in.ibm.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Marcelo Tosatti <mtosatti@redhat.com> Signed-off-by: -
Peter Zijlstra authored
Srivatsa Vaddagiri reported that a Java workload triggers this warning in kernel/exit.c: WARN_ON_ONCE(!list_empty(&tsk->perf_counter_ctx.counter_list)); Add the inherited counter propagation on self-detach, this could cause counter leaks and incomplete stats in threaded code like the below: #include <pthread.h> #include <unistd.h> void *thread(void *arg) { sleep(5); return NULL; } void main(void) { pthread_t thr; pthread_create(&thr, NULL, thread, NULL); } Reported-by:Srivatsa Vaddagiri <vatsa@in.ibm.com> Signed-off-by:
-
Peter Zijlstra authored
Clean up code that open-coded the list_{add,del}_counter() code in __perf_counter_exit_task() which consequently diverged. This could lead to software counter crashes. Also, fold the ctx->nr_counter inc/dec into those functions and clean up some of the related code. [ Impact: fix potential sw counter crash, cleanup ] Signed-off-by:
-
- 16 May, 2009 1 commit
-
-
Linus Torvalds authored
-
- 15 May, 2009 1 commit
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6Linus Torvalds authored
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci-2.6: PCI MSI: Fix MSI-X with NIU cards PCI: Fix pci-e port driver slot_reset bad default return value
-
