- 29 Jun, 2009 3 commits
-
-
Paul Mackerras authored
Vince Weaver reported a 'perf stat' measurement overhead in the count of retired instructions, which can amount to a +6000 instructions inflated count in the reported count. At present, perf stat creates its counters on the perf process. Thus the counters count the fork and various other activity in both the parent and child, such as the resolver overhead for resolving PLT entries for any libc functions that haven't been called before, such as execvp. This reduces the overhead by creating the counters on the child process after the fork, using a couple of pipes to synchronize so that the child process waits until the parent has created the counters before doing the exec. To eliminate the PLT resolution overhead on calling execvp, this does a dummy execvp first which will always fail. With this, the overhead of executing a program goes down from over 4800 instructions to about 90 instructions on powerpc (32-bit). This was measured with a statically-linked program written in assembler which only does the 3 instructions needed to call _exit(0). Before: $ perf stat -e 0:1:u ./three Performance counter stats for './three': 4858 instructions 0.001274523 seconds time elapsed After: $ perf stat -e 0:1:u ./three Performance counter stats for './three': 92 instructions 0.000468153 seconds time elapsed Reported-by:Vince Weaver <vince@deater.net> Signed-off-by:
Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> LKML-Reference: <19016.41425.814043.870352@cargo.ozlabs.ibm.com> Signed-off-by:
Ingo Molnar <mingo@elte.hu>
-
Ingo Molnar authored
Peter expressed a strong preference for percentage based display of scaled values - so revert to that from the recently introduced multiplication-factor unit. Reported-by:
Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Jaswinder Singh Rajput <jaswinder@kernel.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> LKML-Reference: <new-submission> Signed-off-by:
-
Yinghai Lu authored
The print out should read the value before changing the value. Signed-off-by:
Yinghai Lu <yinghai@kernel.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> LKML-Reference: <4A487017.4090007@kernel.org> Signed-off-by:
-
- 28 Jun, 2009 1 commit
-
-
Jaswinder Singh Rajput authored
Set attrs and nr_counters if no event is selected and !null_run. Setting of attrs should depend on number of counters, so we need to memcpy only for sizeof(default_attrs) Also set nr_counters as ARRAY_SIZE(default_attrs) in place of hardcoded value. Signed-off-by:
Jaswinder Singh Rajput <jaswinderrajput@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> LKML-Reference: <1246126749.32198.16.camel@hpdv5.satnam> Signed-off-by:
-
- 27 Jun, 2009 4 commits
-
-
Jaswinder Singh Rajput authored
Increase size for event name to handle bigger names like 'L1-d$-prefetch-misses' Changed scaled counters from percentage to a multiplicative factor because the latter is more expressive. Also aligned the scaling factor, otherwise sometimes it looks like: 384 iTLB-load-misses (4.74x scaled) 452029 branch-loads (8.00x scaled) 5892 branch-load-misses (20.39x scaled) 972315 iTLB-loads (3.24x scaled) Before: 150708 L1-d$-stores (scaled from 23.57%) 428804 L1-d$-prefetches (scaled from 23.47%) 314446 L1-d$-prefetch-misses (scaled from 23.42%) 252626137 L1-i$-loads (scaled from 23.24%) 5297550 dTLB-load-misses (scaled from 23.96%) 106992392 branch-loads (scaled from 23.67%) 5239561 branch-load-misses (scaled from 23.43%) After: 1731713 L1-d$-loads ( 14.25x scaled) 44241 L1-d$-prefetches ( 3.88x scaled) 21076 L1-d$-prefetch-misses ( 3.40x scaled) 5789421 L1-i$-loads ( 3.78x scaled) 29645 dTLB-load-misses ( 2.95x scaled) 461474 branch-loads ( 6.52x scaled) 7493 branch-load-misses ( 26.57x scaled) Reported-by: -
Ingo Molnar authored
In multi-run (-r/--repeat) printouts, print out the noise of the wall-clock average as well. Also, fix a bug in printing out scaled counters: if it was not scaled then we should not update the average with -1. Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> LKML-Reference: <new-submission> Signed-off-by: -
Ingo Molnar authored
Allow a no-counters run. This can be useful to measure just elapsed wall-clock time - or to assess the raw overhead of perf stat itself, without running any counters. Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> LKML-Reference: <new-submission> Signed-off-by: -
Ingo Molnar authored
Vince Weaver reported that there's a handful of #ifdef __MINGW32__ sections in the code. Remove them as they are in essence dead code - as unlike upstream Git, the perf tool is unlikely to be ported to Windows. Reported-by:
-
- 26 Jun, 2009 3 commits
-
-
Peter Zijlstra authored
Complete the counter swap by indeed switching the times too and updating the userpage after modifying the counter values. Signed-off-by:
-
Frederic Weisbecker authored
Use the newly created callchains radix tree to gather the chains stats from the recorded events and then print the callchains for all of them, sorted by hits, using the "-c" parameter with perf report. Example: 66.15% [k] atm_clip_exit 63.08% 0xffffffffffffff80 0xffffffff810196a8 0xffffffff810c14c8 0xffffffff8101a79c 0xffffffff810194f3 0xffffffff8106ab7f 0xffffffff8106abe5 0xffffffff8106acde 0xffffffff8100d94b 0xffffffff8153e7ea [...] 1.54% 0xffffffffffffff80 0xffffffff810196a8 0xffffffff810c14c8 0xffffffff8101a79c [...] Symbols are not yet resolved. Signed-off-by:Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> LKML-Reference: <1246026481-8314-3-git-send-email-fweisbec@gmail.com> Signed-off-by:
-
Frederic Weisbecker authored
We plan to display the callchains depending on some user-configurable parameters. To gather the callchains stats from the recorded stream in a fast way, this patch introduces an ad hoc radix tree adapted for callchains and also a rbtree to sort these callchains once we have gathered every events from the stream. Signed-off-by:
-
- 25 Jun, 2009 14 commits
-
-
Frederic Weisbecker authored
Building latest perfcounter fails on the following error: builtin-record.c: In function ‘create_counter’: builtin-record.c:451: erreur: ignoring return value of ‘read’, declared with attribute warn_unused_result make: *** [builtin-record.o] Erreur 1 Just check if we successfully read the perf file descriptor. Signed-off-by:
-
Jaswinder Singh Rajput authored
Add 'l1d' and 'l1i' aliases again as shortcuts - just dont make them the primary display alias. Signed-off-by:
-
Peter Zijlstra authored
Provide the basic infrastructure to provide per task stats. Signed-off-by:
-
Peter Zijlstra authored
Now that we can collect per task statistics, add modes that make use of that facility. Signed-off-by:
-
Peter Zijlstra authored
The PERF_EVENT_READ implementation made me realize we don't actually need the sample_type int the output sample, since we already have that in the perf_counter_attr information. Therefore, remove the PERF_EVENT_MISC_OVERFLOW bit and the event->type overloading, and imply put counter overflow samples in a PERF_EVENT_SAMPLE type. This also fixes the issue that event->type was only 32-bit and sample_type had 64 usable bits. Signed-off-by:
-
Peter Zijlstra authored
With the introduction of PERF_EVENT_READ we have the possibility to provide accurate counter values for individual tasks in a task hierarchy. However, due to the lazy context switching used for similar counter contexts our current per task counts are way off. In order to maintain some of the lazy switch benefits we don't disable it out-right, but simply iterate the active counters and flip the values between the contexts. This only reads the counters but does not need to reprogram the full PMU. Signed-off-by:
-
Peter Zijlstra authored
Provide a read() like event which can be used to log the counter value at specific sites such as child->parent folding on exit. In order to be useful, we log the counter parent ID, not the actual counter ID, since userspace can only relate parent IDs to perf_counter_attr constructs. Signed-off-by:
-
Peter Zijlstra authored
Update the mmap control page with the needed information to use the userspace RDPMC instruction for self monitoring. Signed-off-by:
-
Peter Zijlstra authored
Add the needed time scale to the self-profile mmap information. Signed-off-by:
-
Peter Zijlstra authored
Since there are two distinct sections to the control page, move them apart so that possible extentions don't overlap. Signed-off-by:
-
Peter Zijlstra authored
Create a structured file format that includes the full perf_counter_attr and all its relevant counter IDs so that the reporting program has full information. Signed-off-by:
-
Jaswinder Singh Rajput authored
Added new alias for events. On AMD box: $ ./perf stat -e l1d -e l1d-misses -e l1d-write -e l1d-prefetch -e l1d-prefetch-miss -e l1i -e l1i-misses -e l1i-prefetch -e l2 -e l2-misses -e l2-write -e dtlb -e dtlb-misses -e itlb -e itlb-misses -e bpu -e bpu-misses -- ls -lR /usr/include/ > /dev/null Before : Performance counter stats for 'ls -lR /usr/include/': 248064467 L1-data-Cache-Load-Referencees (scaled from 23.27%) 1001433 L1-data-Cache-Load-Misses (scaled from 23.34%) 153691 L1-data-Cache-Store-Referencees (scaled from 23.34%) 423248 L1-data-Cache-Prefetch-Referencees (scaled from 23.33%) 302138 L1-data-Cache-Prefetch-Misses (scaled from 23.25%) 251217546 L1-instruction-Cache-Load-Referencees (scaled from 23.25%) 5757005 L1-instruction-Cache-Load-Misses (scaled from 23.23%) 93435 L1-instruction-Cache-Prefetch-Referencees (scaled from 23.24%) 6496073 L2-Cache-Load-Referencees (scaled from 23.32%) 609485 L2-Cache-Load-Misses (scaled from 23.45%) 6876991 L2-Cache-Store-Referencees (scaled from 23.71%) 248922840 Data-TLB-Cache-Load-Referencees (scaled from 23.94%) 5828386 Data-TLB-Cache-Load-Misses (scaled from 24.17%) 257613506 Instruction-TLB-Cache-Load-Referencees (scaled from 24.20%) 6833 Instruction-TLB-Cache-Load-Misses (scaled from 23.88%) 109043606 Branch-Cache-Load-Referencees (scaled from 23.64%) 5552296 Branch-Cache-Load-Misses (scaled from 23.42%) 0.413702461 seconds time elapsed. After : Peformance counter stats for 'ls -lR /usr/include/': 266590464 L1-d$-loads (scaled from 23.03%) 1222273 L1-d$-load-misses (scaled from 23.58%) 146204 L1-d$-stores (scaled from 23.83%) 406344 L1-d$-prefetches (scaled from 24.09%) 283748 L1-d$-prefetch-misses (scaled from 24.10%) 249650965 L1-i$-loads (scaled from 23.80%) 3353961 L1-i$-load-misses (scaled from 23.82%) 104599 L1-i$-prefetches (scaled from 23.68%) 4836405 LLC-loads (scaled from 23.67%) 498214 LLC-load-misses (scaled from 23.66%) 4953994 LLC-stores (scaled from 23.64%) 243354097 dTLB-loads (scaled from 23.77%) 6468584 dTLB-load-misses (scaled from 23.74%) 249719549 iTLB-loads (scaled from 23.25%) 5060 iTLB-load-misses (scaled from 23.00%) 112343016 branch-loads (scaled from 22.76%) 5528876 branch-load-misses (scaled from 22.54%) 0.427154051 seconds time elapsed. Reported-by : Ingo Molnar <mingo@elte.hu> Signed-off-by: -
Jaswinder Singh Rajput authored
Made new table for cache operartion stat 'hw_cache_stat' as: L1I : Read and prefetch only ITLB and BPU : Read-only introduce is_cache_op_valid() for cache operation validity And checks for valid cache operations. Reported-by : Ingo Molnar <mingo@elte.hu> Signed-off-by:
-
Johannes Weiner authored
Looking backward for the first space from the end of a line in /proc/pid/maps does not find the start of the pathname of the mapped file if it contains a space. Since the only slashes we have in this file occur in the (absolute!) pathname column of file mappings, looking for the first slash in a line is a safe method to find the name. Signed-off-by:
Johannes Weiner <hannes@cmpxchg.org> Cc: Stefani Seibold <stefani@seibold.net> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Alexey Dobriyan <adobriyan@gmail.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> LKML-Reference: <20090624190835.GA25548@cmpxchg.org> Signed-off-by:
-
- 24 Jun, 2009 4 commits
-
-
Ingo Molnar authored
Much of perf's libraries comes from the Git project. I noticed that the files (in tools/perf/util/*.[ch] and elsewhere) are quite spartan wrt. credits, so lets add a CREDITS file that includes an (incomplete!) list of main contributors. Thanks guys, these libraries are really useful. Special thanks go to Johannes Schindelin and Junio C Hamano for coming up with this list. List-Composed-By:
Johannes Schindelin <Johannes.Schindelin@gmx.de> Cc: Junio C Hamano <gitster@pobox.com> Signed-off-by:
-
Jaswinder Singh Rajput authored
Remove dead code and do some code alignment. Signed-off-by:
-
Yong Wang authored
Previous code made an assumption that the power on value of global control MSR has enabled all fixed and general purpose counters properly. However, this is not the case for certain Intel processors, such as Atom - and it might also be firmware dependent. Each enable bit in IA32_PERF_GLOBAL_CTRL is AND'ed with the enable bits for all privilege levels in the respective IA32_PERFEVTSELx or IA32_PERF_FIXED_CTR_CTRL MSRs to start/stop the counting of respective counters. Counting is enabled if the AND'ed results is true; counting is disabled when the result is false. The end result is that all fixed counters are always disabled on Atom processors because the assumption is just invalid. Fix this by not initializing the ctrl-mask out of the global MSR, but setting it to perf_counter_mask. Reported-by:
Stephane Eranian <eranian@googlemail.com> Signed-off-by:
Yong Wang <yong.y.wang@intel.com> Cc: Arjan van de Ven <arjan@infradead.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> LKML-Reference: <20090624021324.GA2788@ywang-moblin2.bj.intel.com> Signed-off-by:
-
Roel Kluin authored
size_t res cannot be less than 0 - fread returns 0 on error. [ Updated by: René Scharfe <rene.scharfe@lsrfire.ath.cx> ] Reported-by:
Roel Kluin <roel.kluin@gmail.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Junio C Hamano <gitster@pobox.com> LKML-Reference: <4A3FB479.2090902@lsrfire.ath.cx> Signed-off-by:
-
- 23 Jun, 2009 7 commits
-
-
Jaswinder Singh Rajput authored
Error message should use stderr for verbose (-v), otherwise message will be lost for: $ ./perf stat -v <cmd> > /dev/null For example on AMD bus-cycles event is not available so now it looks like: $ ./perf stat -v -e bus-cycles ls > /dev/null Error: counter 0, sys_perf_counter_open() syscall returned with -1 (Invalid argument) Performance counter stats for 'ls': <not counted> bus-cycles 0.006765877 seconds time elapsed. Signed-off-by: -
Ingo Molnar authored
Reported-by:
Brice Goglin <Brice.Goglin@inria.fr> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> LKML-Reference: <new-submission> Signed-off-by:
-
Peter Zijlstra authored
We don't need to add usage counts for swcounter and attr usage models for inherited counters since the parent counter will always have one, which suffices to generate the needed output. This avoids up to 3 global atomic increments per inherited counter. LKML-Reference: <new-submission> Signed-off-by:
-
Peter Zijlstra authored
Teach perf_counter_alloc() about inheritance so that we can optimize the inherit path in the next patch. Remove the child_counter->atrr.inherit = 1 line because the only way to get there is if parent_counter->attr.inherit == 1 and we copy the attrs. Signed-off-by:
-
Peter Zijlstra authored
Similar to tracepoints, use an enable variable to reduce overhead when unused. Only look for a counter of a particular event type when we know there is at least one in the system. Signed-off-by:
-
Peter Zijlstra authored
Martin Schwidefsky reported "perf report" symbol resolution problems on S390. Since we only report MMAP, not MUNMAP, we have to deal with overlapping maps. We used to simply throw out the old map on the assumption whole maps got unmapped. This obviously doesn't deal with partial unmaps. However it appears some dynamic linkers do fancy partial unmaps (s390), so do something more elaborate and truncate the old maps, only removing them when they've been fully covered. This resolves (part of) the S390 symbol resolution problems. Reported-by:
Martin Schwidefsky <schwidefsky@de.ibm.com> Tested-by:
-
Jaswinder Singh Rajput authored
-l is not supported, it should be -S for scale. Signed-off-by:
-
- 22 Jun, 2009 4 commits
-
-
Jaswinder Singh Rajput authored
"faults" should be alias for "page-faults" Also fixed alignment and 80 characters issue Signed-off-by:
-
Peter Zijlstra authored
Print more symbol relocation related info under -vv. Signed-off-by:
Peter Zijlstra <peterz@infradead.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Paul Mackerras <paulus@samba.org> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> LKML-Reference: <new-submission> Signed-off-by:
-
Jaswinder Singh Rajput authored
By introducing alias member in event_symbol : 1. duplicate lines are removed, like: cpu-cycles and cycles branch-instructions and branches context-switches and cs cpu-migrations and migrations 2. We can also add alias for another events. Now ./perf list looks like : List of pre-defined events (to be used in -e): cpu-cycles OR cycles [Hardware event] instructions [Hardware event] cache-references [Hardware event] cache-misses [Hardware event] branch-instructions OR branches [Hardware event] branch-misses [Hardware event] bus-cycles [Hardware event] cpu-clock [Software event] task-clock [Software event] page-faults [Software event] faults [Software event] minor-faults [Software event] major-faults [Software event] context-switches OR cs [Software event] cpu-migrations OR migrations [Software event] rNNN [raw hardware event descriptor] Signed-off-by:
-
Jaswinder Singh Rajput authored
Define separate declarations for H/W and S/W events to: 1. Shorten name to save some space so that we can add more members 2. Fix alignment 3. Avoid declaring HARDWARE/SOFTWARE again and again. Removed unused CR(x, y) Signed-off-by:
-
