aboutsummaryrefslogtreecommitdiff
path: root/tools/perf
AgeCommit message (Collapse)AuthorFilesLines
2010-01-13perf tools: Enhance glob string matchingMasami Hiramatsu1-4/+61
Enhance strglobmatch() for supporting character classes([CHARS], complementation and ranges are also supported) and escaped special characters (\*, \? etc). Signed-off-by: Masami Hiramatsu <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: systemtap <[email protected]> Cc: DLE <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Mike Galbraith <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2010-01-13perf tools: Support tracepoint glob matchingMasami Hiramatsu1-3/+8
Support glob wildcard when selecting tracepoint events by -e option. Without this patch, perf-tools supports 'GROUP:*:record' syntax for selecting all tracepoints under GROUP group. With this patch, user can choose tracepoints more flexibly by using partial wildcards, e.g. 'block:*bio*:record'. Signed-off-by: Masami Hiramatsu <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: systemtap <[email protected]> Cc: DLE <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Mike Galbraith <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2010-01-13perf probe: Show probe list in pagerMasami Hiramatsu1-0/+3
Show probe list in pager, because the list can be longer than a page. Signed-off-by: Masami Hiramatsu <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: systemtap <[email protected]> Cc: DLE <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Mike Galbraith <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2010-01-13perf probe: Remove newline from die()Masami Hiramatsu1-6/+6
Remove newline from die(), because it is automatically added. Signed-off-by: Masami Hiramatsu <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: systemtap <[email protected]> Cc: DLE <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Mike Galbraith <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2010-01-13perf tools: Handle relocatable kernelsArnaldo Carvalho de Melo6-5/+133
DSOs don't have this problem because the kernel emits a PERF_MMAP for each new executable mapping it performs on monitored threads. To fix the kernel case we simulate the same behaviour, by having 'perf record' to synthesize a PERF_MMAP for the kernel, encoded like this: [root@doppio ~]# perf record -a -f sleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.344 MB perf.data (~15038 samples) ] [root@doppio ~]# perf report -D | head -10 0xd0 [0x40]: event: 1 . . ... raw event: size 64 bytes . 0000: 01 00 00 00 00 00 40 00 00 00 00 00 00 00 00 00 ......@........ . 0010: 00 00 00 81 ff ff ff ff 00 00 00 00 00 00 00 00 ............... . 0020: 00 00 00 00 00 00 00 00 5b 6b 65 72 6e 65 6c 2e ........ [kernel . 0030: 6b 61 6c 6c 73 79 6d 73 2e 5f 74 65 78 74 5d 00 kallsyms._text] . 0xd0 [0x40]: PERF_RECORD_MMAP 0/0: [0xffffffff81000000((nil)) @ (nil)]: [kernel.kallsyms._text] I.e. we identify such event as having: .pid = 0 .filename = [kernel.kallsyms.REFNAME] .start = REFNAME addr in /proc/kallsyms at 'perf record' time and use now a hardcoded value of '.text' for REFNAME. Then, later, in 'perf report', if there are any kernel hits and thus we need to resolve kernel symbols, we search for REFNAME and if its address changed, relocation happened and we thus must change the kernel mapping routines to one that uses .pgoff as the relocation to apply. This way we use the same mechanism used for the other DSOs and don't have to do a two pass in all the kernel symbols. Reported-by: Xiao Guangrong <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: "H. Peter Anvin" <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Xiao Guangrong <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2010-01-13perf report: Fix --no-call-chain option handlingArnaldo Carvalho de Melo1-2/+12
To avoid the funny: [root@doppio ~]# perf record -a -f sleep 2s [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.334 MB perf.data (~14572 samples) ] [root@doppio ~]# perf report --no-call-graph selected -g but no callchain data. Did you call perf record without -g? And fix the bug reported by peterz when we do indeed record with callchains and then ask for a report without: [root@doppio ~]# perf record -a -g -f sleep 2s [root@doppio ~]# perf report --no-call-graph Segmentation fault [root@doppio ~]# Reported-by: Peter Zijlstra <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2010-01-13perf session: Keep pointers to the vmlinux mapsArnaldo Carvalho de Melo3-19/+15
So that tools such as 'perf probe' don't have to lookup '[kernel.kallsyms]' but instead access them directly after perf_session__create_kernel_maps or map_groups__create_kernel_maps. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Masami Hiramatsu <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2010-01-13perf tools: Create write_padded routine out of __dsos__write_buildid_tableArnaldo Carvalho de Melo1-6/+16
Will be used by other options where padding is needed. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2010-01-13perf symbols: Export symbol_type__is_aArnaldo Carvalho de Melo2-1/+3
Will be needed by the new HEADER_DSO_INFO feature that will be a HEADER_BUILD_ID superset, replacing it. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2010-01-13perf symbols: Generalise the kallsyms parsing routineArnaldo Carvalho de Melo2-26/+50
Will be used to find an specific symbol by name on 'perf record' to support relocation reference symbols to support relocatable kernels. Still have to conver the perf trace tools to use it instead of their current reimplementation. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2010-01-13perf tools: Fix --pid option for statLiming Wang1-45/+61
current pid option doesn't work for perf stat. Change it to what perf record --pid acts as. Signed-off-by: Liming Wang <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2010-01-13Merge branch 'perf/urgent' into perf/coreIngo Molnar3-22/+34
Merge reason: queue up dependent patch, update to -rc4 Signed-off-by: Ingo Molnar <[email protected]>
2010-01-13perf tools: Check if /dev/null can be used as the -o gcc argumentArnaldo Carvalho de Melo1-11/+22
At least on Debian PARISC64, using: acme@parisc:~/git/linux-2.6-tip$ gcc -v Using built-in specs. Target: hppa-linux-gnu Configured with: ../src/configure -v --with-pkgversion='Debian 4.3.4-6' --with-bugurl=file:///usr/share/doc/gcc-4.3/README.Bugs --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --enable-shared --enable-multiarch --enable-linker-build-id --with-system-zlib --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --enable-nls --with-gxx-include-dir=/usr/include/c++/4.3 --program-suffix=-4.3 --enable-clocale=gnu --enable-libstdcxx-debug --enable-objc-gc --enable-mpfr --disable-libssp --enable-checking=release --build=hppa-linux-gnu --host=hppa-linux-gnu --target=hppa-linux-gnu Thread model: posix gcc version 4.3.4 (Debian 4.3.4-6) there are issues about using 'gcc -o /dev/null': /usr/bin/ld: final link failed: File truncated collect2: ld returned 1 exit status So we test that and use /dev/null in environments where it works, while using an .INTERMEDIATE file on those where it can't be used, so that the .perf.dev.null file can be used instead and then deleted when make exits. Researched-with: Kyle McMartin <[email protected]> Researched-with: Mauro Carvalho Chehab <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2010-01-13perf tools: Move QUIET_STDERR def to before first useArnaldo Carvalho de Melo1-5/+6
QUIET_STDERR is used when detecting if -fstack-protector-all can be used. Noticed while building the perf tools on a Debian PARISC64 machine. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-30perf diff: Fix support for all --sort combinationsArnaldo Carvalho de Melo1-31/+21
When we finish creating the hist_entries we _already_ have them sorted "by name", in fact by what is in --sort, that is exactly how we can find the pairs in perf_session__match_hists as 'comm', 'dso' & 'symbol' all are strings we need to find the matches in the baseline session. So only do the sort by hits followed by a resort by --sort if we need to find the position for shwowing the --displacement of hist entries. Now all these modes work correctly: Example is a simple 'perf record -f find / > /dev/null' ran twice then followed by the following commands: $ perf diff -f --sort comm # Baseline Delta Command # ........ .......... ....... # 0.00% +100.00% find $ perf diff -f --sort dso # Baseline Delta Shared Object # ........ .......... .................. # 59.97% -0.44% [kernel] 21.17% +0.28% libc-2.5.so 18.49% +0.16% [ext3] 0.37% find $ perf diff -f --sort symbol | head -8 # Baseline Delta Symbol # ........ .......... ...... # 6.21% +0.36% [k] ext3fs_dirhash 3.43% +0.41% [.] __GI_strlen 3.53% +0.16% [k] __kmalloc 3.17% +0.49% [k] system_call 3.06% +0.37% [k] ext3_htree_store_dirent $ perf diff -f --sort dso,symbol | head -8 # Baseline Delta Shared Object Symbol # ........ .......... .................. ...... # 6.21% +0.36% [ext3] [k] ext3fs_dirhash 3.43% +0.41% libc-2.5.so [.] __GI_strlen 3.53% +0.16% [kernel] [k] __kmalloc 3.17% +0.49% [kernel] [k] system_call 3.06% +0.37% [ext3] [k] ext3_htree_store_dirent $ And we don't have to do two expensive resorts in the common, non --displacement case. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-30perf diff: Don't add the period for unresolved symbolsArnaldo Carvalho de Melo1-2/+2
Since we don't add histograms buckets for them, this way the sum of baselines should be 100%. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-30perf report: Add --hide-unresolved/-U command line optionArnaldo Carvalho de Melo1-1/+4
Useful to match the 'overhead' column in 'perf report' with the 'baseline' one in 'perf diff'. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-30perf header: perf_header__push_event() shouldn't dieArnaldo Carvalho de Melo3-14/+22
Just propagate eventual errors. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-30perf header: Do_read shouldn't dieArnaldo Carvalho de Melo1-14/+21
Propagate the errors instead, its callers already propagate other errors. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-30perf probe: Change CONFIG_KPROBE_TRACER to CONFIG_KPROBE_EVENTLiming Wang1-1/+1
make the config name consistent Signed-off-by: Liming Wang <[email protected]> Acked-by: Masami Hiramatsu <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-30perf tools: Unify event type descriptionLiming Wang1-9/+10
make event type description to a unified array and the array index consistent to perf_type_id. Signed-off-by: Liming Wang <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Masami Hiramatsu <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf trace: Fix forgotten close of file/dirXiao Guangrong1-0/+5
Signed-off-by: Xiao Guangrong <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Clark Williams <[email protected]> Cc: John Kacur <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf trace: Clean up find_debugfs()Xiao Guangrong2-24/+6
Remove redundant code for 'perf trace' Signed-off-by: Xiao Guangrong <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Clark Williams <[email protected]> Cc: John Kacur <[email protected]> LKML-Reference: <[email protected]> [ v2: resolved conflicts with recent changes ] Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf tools: Mount debugfs automaticallyXiao Guangrong3-11/+9
Mount debugfs filesystem under '/sys/kernel/debug', if it's not mounted. Signed-off-by: Xiao Guangrong <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Clark Williams <[email protected]> Cc: John Kacur <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf kmem: Fix statistics typoWenji Huang1-2/+2
Replace bytes_req with bytes_alloc. Signed-off-by: Wenji Huang <[email protected]> Reviewed-by: Li Zefan <[email protected]> Cc: [email protected] LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf tools: Do a few more directory handling optimizationsUlrich Drepper2-38/+29
A few more optimizations for perf when dealing with directories. Some of them significantly cut down the work which has to be done. d_type should always be set; otherwise fix the kernel code. And there are functions available to parse fstab-like files, so use them. Signed-off-by: Ulrich Drepper <[email protected]> Acked-by: Pekka Enberg <[email protected]> Cc: [email protected] Cc: [email protected] Cc: [email protected] Cc: [email protected] Cc: [email protected] Cc: [email protected] Cc: [email protected] LKML-Reference: <[email protected]> [ v2: two small stylistic fixlets ] Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf: Rename perf_event_hw_event in design documentTim Blechmann1-4/+4
perf_event_hw_event has been renamed to perf_event_attr. The design document was still using the old name, though. Signed-off-by: Tim Blechmann <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Frederic Weisbecker <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf record: Introduce a symtab cacheArnaldo Carvalho de Melo6-9/+165
Now a cache will be created in a ~/.debug debuginfo like hierarchy, so that at the end of a 'perf record' session all the binaries (with build-ids) involved get collected and indexed by their build-ids, so that perf report can find them. This is interesting when developing software where you want to do a 'perf diff' with the previous build and opens avenues for lots more interesting tools, like a 'perf diff --graph' that takes more than two binaries into account. Tunables for collecting just the symtabs can be added if one doesn't want to have the full binary, but having the full binary allows things like 'perf rerecord' or other tools that can re-run the tests by having access to the exact binary in some perf.data file, so it may well be interesting to keep the full binary there. Space consumption is minimised by trying to use hard links, a 'perf cache' tool to manage the space used, a la ccache is required to purge older entries. With this in place it will be possible also to introduce new commands, 'perf archive' and 'perf restore' (or some more suitable and future proof names) to create a cpio/tar file with the perf data and the files in the cache that _had_ perf hits of interest. There are more aspects to polish, like finding the right vmlinux file to cache, etc, but this is enough for a first step. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf session: Remove redundant prefix & suffix from perf_event_opsArnaldo Carvalho de Melo9-64/+64
Since now all that we have are perf event handlers, leave just the name of the event. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf session: Move full_paths config to symbol_confArnaldo Carvalho de Melo5-5/+5
Now perf_event_ops has just that, event handlers. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf session: Move total_unknown to perf_session->unknown eventsArnaldo Carvalho de Melo2-2/+3
As this is a session property, not belonging to perf_event_ops, that can be shared by many perf_session instances. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf session: Remove sample_type_check from event_opsArnaldo Carvalho de Melo7-27/+31
This is really something tools need to do before asking for the events to be processed, leaving perf_session__process_events to do just that, process events. Also add a msg parameter to perf_session__has_traces() so that the right message can be printed, fixing a regression added by me in the previous cset (right timechart message) and also fixing 'perf kmem', that was not asking if 'perf kmem record' was ran. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf session: Share the common trace sample_check routine as ↵Arnaldo Carvalho de Melo6-51/+17
perf_session__has_traces Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf tools: Move the map class definition to a separate headerArnaldo Carvalho de Melo5-62/+80
And this resulted in the need for adding some missing includes in some places that were getting the definitions needed out of sheer luck. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf session: Move the event processing routines to session.cArnaldo Carvalho de Melo3-253/+245
No need for an extra "data_map" file since the routines there operate mainly on a perf_session instance. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf: Use format string of printf to align stringsAmerigo Wang1-2/+1
Instead of filling whitespaces to do alignment, use printf's format string. This simplifies the code a bit. Signed-off-by: WANG Cong <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Frederic Weisbecker <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf tools: Add missing header files to LIB_H Makefile variableArnaldo Carvalho de Melo1-0/+8
So that changes in them trigger rebuilds, like when we're doing bisects. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-28perf record: We should fork only if a program was specified to runArnaldo Carvalho de Melo1-6/+8
IOW: Now 'perf record -a' works, this was a bug introduced in: 856e96608a72412d319e498a3a7c557571f811bd "perf record: Properly synchronize child creation" Also fix the -C usage, i.e. allow for profiling all the tasks in one CPU. Reported-by: Pekka Enberg <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-18perf diff: Fix usage array, it must end with a NULL entryArnaldo Carvalho de Melo4-6/+6
Fixing this: [acme@doppio linux-2.6-tip]$ perf diff --hell Error: unknown option `hell' usage: perf diff [<options>] [old_file] [new_file] Segmentation fault [acme@doppio linux-2.6-tip]$ Also go over the other such arrays to check if they all were OK, they are, but there were some minor changes to do like making one static and renaming another to match the command it refers to. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-18perf session: Make events_stats u64 to avoid overflow on 32-bit archesArnaldo Carvalho de Melo2-3/+3
Pekka Enberg reported weird percentages in perf report. It turns out we are overflowing a 32-bit variables in struct events_stats on 32-bit architectures. Before: [acme@ana linux-2.6-tip]$ perf report -i pekka.perf.data 2> /dev/null | head -10 281.96% Xorg b710a561 [.] 0x000000b710a561 140.15% Xorg [kernel] [k] __initramfs_end 51.56% metacity libgobject-2.0.so.0.2000.1 [.] 0x00000000026e46 35.12% evolution libcairo.so.2.10800.6 [.] 0x000000000203bd 33.84% metacity libpthread-2.9.so [.] 0x00000000007a3d After: [acme@ana linux-2.6-tip]$ perf report -i pekka.perf.data 2> /dev/null | head -10 30.04% Xorg b710a561 [.] 0x000000b710a561 14.93% Xorg [kernel] [k] __initramfs_end 5.49% metacity libgobject-2.0.so.0.2000.1 [.] 0x00000000026e46 3.74% evolution libcairo.so.2.10800.6 [.] 0x000000000203bd 3.61% metacity libpthread-2.9.so [.] 0x00000000007a3d Reported-by: Pekka Enberg <[email protected]> Tested-by: Pekka Enberg <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-17perf probe: Check new event nameMasami Hiramatsu1-0/+15
Check new event name is same syntax as a C symbol in perf command. In other words, checking the name is as like as other tracepoint events. This can prevent user to create an event with useless name (e.g. foo|bar, foo*bar). Signed-off-by: Masami Hiramatsu <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Steven Rostedt <[email protected]> Cc: Jim Keniston <[email protected]> Cc: Ananth N Mavinakayanahalli <[email protected]> Cc: Christoph Hellwig <[email protected]> Cc: Jason Baron <[email protected]> Cc: K.Prasad <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Srikar Dronamraju <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: systemtap <[email protected]> Cc: DLE <[email protected]> LKML-Reference: <[email protected]> [ v2: minor cleanups ] Signed-off-by: Ingo Molnar <[email protected]>
2009-12-17perf probe: Check whether debugfs path is correctMasami Hiramatsu1-0/+4
Check whether the debugfs path is correct before executing a command, because perf-probe depends on debugfs. Signed-off-by: Masami Hiramatsu <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Steven Rostedt <[email protected]> Cc: Jim Keniston <[email protected]> Cc: Ananth N Mavinakayanahalli <[email protected]> Cc: Christoph Hellwig <[email protected]> Cc: Jason Baron <[email protected]> Cc: K.Prasad <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Srikar Dronamraju <[email protected]> Cc: systemtap <[email protected]> Cc: DLE <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-17perf probe: Fix libdwarf include path for DebianMasami Hiramatsu2-28/+34
Fix libdwarf include path to fit debian-like systems too. Borislav Petkov reported: > even after installing libdwarf-dev on my debian box here, > make in tools/perf/ still complains that it cannot find libdwarf: > > Makefile:491: No libdwarf.h found or old libdwarf.h found, disables dwarf > support. Please install libdwarf-dev/libdwarf-devel >= 20081231 > > The problem is that the include path on debian is not > /usr/include/libdwarf/ but simply /usr/include because the debian > package libdwarf-dev puts the headers straight into > /usr/include. This patch adds -I/usr/include/libdwarf to BASIC_CFLAGS and fix probe-finder.h to include just libdwarf.h/dwarf.h. This patch also adds a workaround for the undefined _MIPS_SZLONG bug in libdwarf.h. Reported-by: Borislav Petkov <[email protected]> Signed-off-by: Masami Hiramatsu <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Srikar Dronamraju <[email protected]> Cc: Gabor Gombas <[email protected]> Cc: systemtap <[email protected]> Cc: DLE <[email protected]> LKML-Reference: <[email protected]> [ v2: small stylistic fixlets to probe-finder.h ] Signed-off-by: Ingo Molnar <[email protected]>
2009-12-16perf record: Use per-task-per-cpu events for inherited eventsPeter Zijlstra1-2/+2
Create events with a pid and cpu contraint for inherited events so that we get a stream per cpu, instead of all cpus contending on a single stream. Signed-off-by: Peter Zijlstra <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: [email protected] Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-16perf record: Properly synchronize child creationPeter Zijlstra1-38/+71
Remove that ugly usleep and provide proper serialization between parent and child just like perf-stat does. Signed-off-by: Peter Zijlstra <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: [email protected] Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-16perf diff: Percent calcs should use double valuesArnaldo Carvalho de Melo1-4/+5
Otherwise we do integer math and the delta values round up to multiples of 1.0%. Also, calculate absolute values. Things look precise now: $ perf report -i perf.data.old --sort dso,symbol | head -13 9.02% libc-2.10.1.so [.] _IO_vfprintf_internal 4.88% find [.] 0x00000000014af0 2.91% [kernel] [k] __kmalloc 2.85% [kernel] [k] ext4_htree_store_dirent 2.50% libc-2.10.1.so [.] __GI_memmove 2.44% [kernel] [k] half_md4_transform 2.43% [kernel] [k] _spin_lock 2.33% [kernel] [k] system_call $ perf report -i perf.data --sort dso,symbol | head -13 8.55% libc-2.10.1.so [.] _IO_vfprintf_internal 3.11% [kernel] [k] __kmalloc 3.07% [kernel] [k] ext4_htree_store_dirent 2.66% find [.] 0x00000000016bcf 2.61% [kernel] [k] _atomic_dec_and_lock 2.46% [kernel] [k] half_md4_transform 2.41% libc-2.10.1.so [.] __GI_memmove 2.30% find [.] 0x00000000009219 $ perf diff | head -13 9.02% -0.47% libc-2.10.1.so [.] _IO_vfprintf_internal 2.91% +0.20% [kernel] [k] __kmalloc 2.85% +0.23% [kernel] [k] ext4_htree_store_dirent 1.99% +0.62% [kernel] [k] _atomic_dec_and_lock 2.44% +0.02% [kernel] [k] half_md4_transform 2.50% -0.09% libc-2.10.1.so [.] __GI_memmove 1.88% +0.01% [kernel] [k] __d_lookup 2.43% -0.75% [kernel] [k] _spin_lock 0.97% +0.62% [kernel] [k] path_get 1.99% -0.42% libc-2.10.1.so [.] _int_malloc $ Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-16perf diff: Change the default sort order to "dso,symbol"Arnaldo Carvalho de Melo1-0/+2
This is a more intuitive / more meaningful default: $ perf diff | head -8 9.02% +1.00% libc-2.10.1.so [.] _IO_vfprintf_internal 2.91% -1.00% [kernel] [k] __kmalloc 2.85% -1.00% [kernel] [k] ext4_htree_store_dirent 1.99% -1.00% [kernel] [k] _atomic_dec_and_lock 2.44% [kernel] $ Suggested-by: Ingo Molnar <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-16perf diff: Use perf_session__fprintf_hists just like 'perf record'Arnaldo Carvalho de Melo8-110/+169
That means that almost everything you can do with 'perf report' can be done with 'perf diff', for instance: $ perf record -f find / > /dev/null [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.062 MB perf.data (~2699 samples) ] $ perf record -f find / > /dev/null [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.062 MB perf.data (~2687 samples) ] perf diff | head -8 9.02% +1.00% find libc-2.10.1.so [.] _IO_vfprintf_internal 2.91% -1.00% find [kernel] [k] __kmalloc 2.85% -1.00% find [kernel] [k] ext4_htree_store_dirent 1.99% -1.00% find [kernel] [k] _atomic_dec_and_lock 2.44% find [kernel] [k] half_md4_transform $ So if you want to zoom into libc: $ perf diff --dsos libc-2.10.1.so | head -8 37.34% find [.] _IO_vfprintf_internal 10.34% find [.] __GI_memmove 8.25% +2.00% find [.] _int_malloc 5.07% -1.00% find [.] __GI_mempcpy 7.62% +2.00% find [.] _int_free $ And if there were multiple commands using libc, it is also possible to aggregate them all by using --sort symbol: $ perf diff --dsos libc-2.10.1.so --sort symbol | head -8 37.34% [.] _IO_vfprintf_internal 10.34% [.] __GI_memmove 8.25% +2.00% [.] _int_malloc 5.07% -1.00% [.] __GI_mempcpy 7.62% +2.00% [.] _int_free $ The displacement column now is off by default, to use it: perf diff -m --dsos libc-2.10.1.so --sort symbol | head -8 37.34% [.] _IO_vfprintf_internal 10.34% [.] __GI_memmove 8.25% +2.00% [.] _int_malloc 5.07% -1.00% +2 [.] __GI_mempcpy 7.62% +2.00% -1 [.] _int_free $ Using -t/--field-separator can be used for scripting: $ perf diff -t, -m --dsos libc-2.10.1.so --sort symbol | head -8 37.34, , ,[.] _IO_vfprintf_internal 10.34, , ,[.] __GI_memmove 8.25,+2.00%, ,[.] _int_malloc 5.07,-1.00%, +2,[.] __GI_mempcpy 7.62,+2.00%, -1,[.] _int_free 6.99,+1.00%, -1,[.] _IO_new_file_xsputn 1.89,-2.00%, +4,[.] __readdir64 $ Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-16perf report: Fix cut'n'paste error recently introducedArnaldo Carvalho de Melo1-1/+1
Introduced in: d599db3fc5dd4f1e8432fdbc6d899584b25f4dff "perf report: Generalize perf_session__fprintf_hists()" Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-12-16perf session: Move perf report specific hits out of perf_session__fprintf_histsArnaldo Carvalho de Melo2-11/+5
Those don't make sense for tools such as 'perf diff'. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.