- 02 Mar, 2015 16 commits
-
-
Masami Hiramatsu authored
Warn if given uprobe event accesses memory on older kernel. Until 3.14, uprobe event only supports accessing registers so this warns to upgrade kernel if uprobe-event returns -EINVAL and an argument of the event accesses memory ($stack, @+offset, and +|-offs() symtax). With this patch (on 3.10.0-123.13.2.el7.x86_64); ----- # ./perf probe -x ./perf warn_uprobe_event_compat stack=-0\(%sp\) Added new event: Failed to write event: Invalid argument Please upgrade your kernel to at least 3.14 to have access to feature -0(%sp) Error: Failed to add events. ----- Suggested-by:Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by:
Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20150228025329.32106.70581.stgit@localhost.localdomainSigned-off-by:
-
Ingo Molnar authored
On Debian-ish systems libbabeltrace-dev should be suggested as a package install as well. Signed-off-by:
Ingo Molnar <mingo@kernel.org> Cc: David Ahern <david.ahern@oracle.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20150228091849.GA28959@gmail.comSigned-off-by:
-
Ingo Molnar authored
Certain feature tests fail with link errors: triton:~/tip/tools/perf/config/feature-checks> make test-libbabeltrace.bin gcc -MD -o test-libbabeltrace.bin test-libbabeltrace.c # -lbabeltrace provided by /tmp/cc6dRSqd.o: In function `main': test-libbabeltrace.c:(.text+0xf): undefined reference to `bt_ctf_stream_class_get_packet_context_type' although they should already fail with a build error due to lack of a proper prototype for the function. Due to this I first tried to find which library was missing - while it was the whole feature that was missing from the .h file already. To solve this, propagate -Wall -Werror to all testcases and remove them from testcase Makefile rules that used them explicitly. A missing feature now outputs: triton:~/tip/tools/perf/config/feature-checks> make test-libbabeltrace.bin gcc -MD -Wall -Werror -o test-libbabeltrace.bin test-libbabeltrace.c # -lbabeltrace provided by test-libbabeltrace.c: In function ‘main’: test-libbabeltrace.c:6:2: error: implicit declaration of function ‘bt_ctf_stream_class_get_packet_context_type’ [-Werror=implicit-function-declaration] Signed-off-by:
-
Ingo Molnar authored
Before: No bfd.h/libbfd found, install binutils-dev[el]/zlib-static to gain symbol demangling After: No bfd.h/libbfd found, please install binutils-dev[el]/zlib-static/libiberty-dev to gain symbol demangling Change the message to the standard 'please install' language and also add libiberty-dev suggestion for Ubuntu systems. Signed-off-by:
-
Ingo Molnar authored
Before: Missing perl devel files. Disabling perl scripting support, consider installing perl-ExtUtils-Embed After: Missing perl devel files. Disabling perl scripting support, please install perl-ExtUtils-Embed/libperl-dev Change the message to the standard 'please install' language and adds Debian-ish package suggestion. Signed-off-by:
-
Ingo Molnar authored
Change the Python detection message from: config/Makefile:566: No python-config tool was found config/Makefile:566: Python support will not be built config/Makefile:565: No 'python-config' tool was found: disables Python support - please install python-devel/python-dev It's now a standard one-line message with a package install suggestion, and it also uses the standard language used by other feature detection messages. Signed-off-by:
-
Ingo Molnar authored
This message: Makefile:153: The path 'python-config' is not executable. Appears on every perf build that does not have a sufficient python environment installed. It's really just an internal detail of python configuration pass and users should not see it - and it's pretty meaningless to them in any case because the message is not very helpful. (So it's not executable. Why does that matter? What can the user do about it?) Remove the warning, the missing python feature warning is sufficient: config/Makefile:566: No python-config tool was found config/Makefile:566: Python support will not be built although even that one isn't very helpful to users: so no Python support will be built, what can the user do to fix that? Most other such warnings give package install suggestions. Signed-off-by:
-
Ingo Molnar authored
It's an auto-generated file. Signed-off-by:
-
Namhyung Kim authored
The 'perf record --group' option lacks documentation and confuses users. As -e/--event option already supports group spec, it should not be used anymore. Also add a short description of event group itself. Reported-by:
Stephane Eranian <eranian@google.com> Signed-off-by:
Namhyung Kim <namhyung@kernel.org> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1425266013-5034-1-git-send-email-namhyung@kernel.orgSigned-off-by:
-
Namhyung Kim authored
The perf record does not support -l option anymore, so nuke it. Signed-off-by:
-
Namhyung Kim authored
He Kuang reported that current perf tools failed to build when ARCH variable was given like above. It was because the name is different that internal directory name. I can see that David's sparc64 build has same problem. So fix it by applying the sed conversion script to the command line ARCH variable also, and fixing the converted name there (i.e. i386/x86_64 -> x86, sparc64 -> sparc). Reported-by:
He Kuang <hekuang@huawei.com> Signed-off-by:
-
Arnaldo Carvalho de Melo authored
In this commit: commit 363b785f Author: Don Zickus <dzickus@redhat.com> Date: Fri Mar 14 10:43:44 2014 -0400 perf tools: Speed up thread map generation We ended up emitting PERF_RECORD_FORK events after their corresponding PERF_RECORD_COMM, so the code below will remove the "existing thread" and then recreates it, unnecessarily: [root@ssdandy ~]# perf probe -x ~/bin/perf -L machine__process_fork_event <machine__process_fork_event@/home/acme/git/linux/tools/perf/util/machine.c:0> 0 int machine__process_fork_event(struct machine *machine, union perf_event *event, struct perf_sample *sample) 2 { 3 struct thread *thread = machine__find_thread(machine, event->fork.pid, event->fork.tid); 6 struct thread *parent = machine__findnew_thread(machine, event->fork.ppid, event->fork.ptid); /* if a thread currently exists for the thread id remove it */ if (thread != NULL) 12 machine__remove_thread(machine, thread); 14 thread = machine__findnew_thread(machine, event->fork.pid, event->fork.tid); 16 if (dump_trace) 17 perf_event__fprintf_task(event, stdout); 19 if (thread == NULL || parent == NULL || 20 thread__fork(thread, parent, sample->time) < 0) { 21 dump_printf("problem processing PERF_RECORD_FORK, skipping event.\n"); 22 return -1; } 25 return 0; 26 } [root@ssdandy ~]# perf probe -x ~/bin/perf fork_after_comm=machine__process_fork_event:12 Added new event: probe_perf:fork_after_comm (on machine__process_fork_event:12 in /home/acme/bin/perf) You can now use it in all perf tools, such as: perf record -e probe_perf:fork_after_comm -aR sleep 1 [root@ssdandy ~]# [root@ssdandy ~]# perf record -g -e probe_perf:* trace -o /tmp/bla ^C[ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.021 MB perf.data (30 samples) ] Terminated [root@ssdandy ~]# [root@ssdandy ~]# perf report --no-children --show-total-period --stdio # To display the perf.data header info, please use --header/--header-only options. # # Samples: 30 of event 'probe_perf:fork_after_comm' # Event count (approx.): 30 # # Overhead Period Command Shared Object Symbol # ........ ............ ....... ............. ............................... # 100.00% 30 trace trace [.] machine__process_fork_event | ---machine__process_fork_event __event__synthesize_thread.part.2 perf_event__synthesize_threads cmd_trace main __libc_start_main [root@ssdandy ~]# And Looking at 'perf report -D' output we see it: 0 0 0x8698 [0x30]: PERF_RECORD_COMM: auditd:703/707 0 0 0x86c8 [0x38]: PERF_RECORD_FORK(703:707):(703:703) Fix it by more closely mimicking how the kernel generates those records when a new fork happens, i.e. first a PERF_RECORD_FORK, then a PERF_RECORD_COMM. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: David Ahern <dsahern@gmail.com> Cc: Don Zickus <dzickus@redhat.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-h0emvymi2t3mw8dlqd6d6z73@git.kernel.orgSigned-off-by:
-
Suzuki K. Poulose authored
Commit 1971f59f (perf stat: Use read_counter in read_counter_aggr ) broke the perf stat output for unsupported counters. $ perf stat -v -a -C 0 -e CCI_400/config=24/ sleep 1 Warning: CCI_400/config=24/ event is not supported by the kernel. Performance counter stats for 'system wide': 0 CCI_400/config=24/ 1.080265400 seconds time elapsed Where it used to be : $ perf stat -v -a -C 0 -e CCI_400/config=24/ sleep 1 Warning: CCI_400/config=24/ event is not supported by the kernel. Performance counter stats for 'system wide': <not supported> CCI_400/config=24/ 1.083840675 seconds time elapsed This patch fixes the issues by checking if the counter is supported, before reading and logging the counter value. Signed-off-by:
Suzuki K. Poulose <suzuki.poulose@arm.com> Acked-by:
David Ahern <dsahern@gmail.com> Tested-by:
-
David Ahern authored
If JOBS is not by user perf tries to autodetect the number by grepping the number of CPUs from /proc/cpuinfo. 'grep -c' will always return an integer so after this command JOBS should be compared to 0, not "". Signed-off-by:
David Ahern <david.ahern@oracle.com> Link: http://lkml.kernel.org/r/1424303971-91904-1-git-send-email-david.ahern@oracle.comSigned-off-by:
-
David Ahern authored
The perf_time_to_tsc and tsc_to_perf_time functions are only used for x86. Make inclusion of tsc.c dependent on x86 as well. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
Needed to build perf/core buildable in some cases. Signed-off-by:
-
- 01 Mar, 2015 1 commit
-
-
Ingo Molnar authored
Merge tag 'perf-urgent-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent Pull perf/urgent fixes from Arnaldo Carvalho de Melo: - pthread_attr_setaffinity_np() feature detection build fixes (Adrian Hunter, Josh Boyer) - Fix probing for PERF_FLAG_FD_CLOEXEC flag (Adrian Hunter) - Fix order of arguments to memcpy_alloc_mem in 'perf bench' (Bruce Merry) - Sparc64 and Aarch64 build and segfault fixes (David Ahern) Signed-off-by:
-
- 28 Feb, 2015 1 commit
-
-
Ingo Molnar authored
Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo: User visible changes: - Fix SIGBUS failures due to misaligned accesses in Sparc64 (David Ahern) - Fix branch stack mode in 'perf report' (He Kuang) - Fix a 'perf probe' operator precedence bug (He Kuang) - Fix Support for different binaries with same name in 'perf diff' (Kan Liang) - Check kprobes blacklist when adding new events via 'perf probe' (Masami Hiramatsu) - Add --purge FILE to remove all caches of FILE in 'perf buildid-cache' (Masami Hiramatsu) - Show usage with some incorrect params (Masami Hiramatsu) - Add new buildid cache if update target is not cached in 'buildid-cache' (Masami Hiramatsu) - Allow listing events with 'tracepoint' prefix in 'perf list' (Yunlong Song) - Sort the output of 'perf list' (Yunlong Song) - Fix bash completion of 'perf --' (Yunlong Song) Developer Zone: - Handle strdup() failure path in 'perf probe' (Arnaldo Carvalho de Melo) - Fix get_real_path to free allocated memory in error path in 'perf probe' (Masami Hiramatsu) - Use pr_debug instead of verbose && pr_info perf buildid-cache (Masami Hiramatsu) - Fix building of 'perf data' with some gcc versions due to incorrect array struct entry (Yunlong Song) Signed-off-by:
-
- 27 Feb, 2015 16 commits
-
-
He Kuang authored
When perf.data file is obtained using 'perf record -b', perf report should use branch stack mode to generate output. But this function is broken by improper comparison between boolean and constant -1. before this patch: $ perf report -b -i perf.data Samples: 16 of event 'cycles', Event count (approx.): 3171896 Overhead Command Shared Object Symbol 13.59% ls [kernel.kallsyms] [k] prio_tree_remove 13.16% ls [kernel.kallsyms] [k] change_pte_range 12.09% ls [kernel.kallsyms] [k] page_fault 12.02% ls [kernel.kallsyms] [k] zap_pte_range ... after this patch: $ perf report -b -i perf.data Samples: 256 of event 'cycles', Event count (approx.): 256 Overhead Command Source Shared Object Source Symbol Target Shared Object Target Symbol 9.38% ls [unknown] [k] 0000000000000000 [unknown] [k] 0000000000000000 6.25% ls libc-2.19.so [.] _dl_addr libc-2.19.so [.] _dl_addr 6.25% ls [kernel.kallsyms] [k] zap_pte_range [kernel.kallsyms] [k] zap_pte_range 6.25% ls [kernel.kallsyms] [k] change_pte_range [kernel.kallsyms] [k] change_pte_range 0.39% ls [kernel.kallsyms] [k] prio_tree_remove [kernel.kallsyms] [k] prio_tree_remove ... Signed-off-by: -
Masami Hiramatsu authored
Show usage if no action is specified or unexpected parameter is given. In other words, be more user friendly. Signed-off-by:
-
Masami Hiramatsu authored
Use pr_debug instead of the combination of verbose and pr_info. "if (verbose) pr_info(...)" is same as "pr_debug(...)", replace it. Signed-off-by:
-
Masami Hiramatsu authored
Add --purge FILE to remove all caches of FILE. Since the current --remove FILE removes a cache which has same build-id of given FILE. Since the command takes a FILE path, it can confuse user who tries to remove cache about FILE path. ----- # ./perf buildid-cache -v --add ./perf Adding 133b7b5486d987a5ab5c3ebf4ea14941f45d4d4f ./perf: Ok # (update the ./perf binary) # ./perf buildid-cache -v --remove ./perf Removing 305bbd1be68f66eca7e2d78db294653031edfa79 ./perf: FAIL ./perf wasn't in the cache ----- Actually, the --remove's FAIL is not shown, it just silently fails. So, this patch adds --purge FILE action for such usecase. perf buildid-cache --purge FILE removes all caches which has same FILE path. In other words, it removes all caches including old binaries. ----- # ./perf buildid-cache -v --add ./perf Adding 133b7b5486d987a5ab5c3ebf4ea14941f45d4d4f ./perf: Ok # (update the ./perf binary) # ./perf buildid-cache -v --purge ./perf Removing 133b7b5486d987a5ab5c3ebf4ea14941f45d4d4f ./perf: Ok ----- BTW, if you want to purge all the caches, remove ~/.debug/* . Signed-off-by:
-
Yunlong Song authored
The perf-completion.sh uses a predefined string '--help --version --exec-path --html-path --paginate --no-pager --perf-dir --work-tree --debugfs-dir' for the bash completion of 'perf --*', which has two problems: Problem 1: If the options of perf are changed (see handle_options() in perf.c), the perf-completion.sh has to be changed at the same time. If not, the bash completion of 'perf --*' and the options which perf really supports will be inconsistent. Problem 2: When typing another single character after 'perf --', e.g. 'h', and hit TAB key to get the bash completion of 'perf --h', the character 'h' disappears at once. This is not what we want, we wish the bash completion can return '--help --html-path' and then we can continue to choose one. To solve this problem, we add '--list-opts' to perf, which now supports 'perf --list-opts' directly, and its result can be used in bash completion now. Example: Before this patch: $ perf --h <-- hit TAB key after character 'h' $ perf -- <-- 'h' disappears and no required result After this patch: $ perf --h <-- hit TAB key after character 'h' --help --html-path <-- the required result Signed-off-by:
Yunlong Song <yunlong.song@huawei.com> Tested-by:
-
Yunlong Song authored
Extend 'perf list --raw-dump' to 'perf list --raw-dump [hw|sw|cache |tracepoint|pmu|event_glob]' in order to show the raw-dump of a certain kind of events rather than all of the events. Example: Before this patch: $ perf list --raw-dump hw branch-instructions branch-misses bus-cycles cache-misses cache-references cpu-cycles instructions stalled-cycles-backend stalled-cycles-frontend alignment-faults context-switches cpu-clock cpu-migrations emulation-faults major-faults minor-faults page-faults task-clock ... ... writeback:writeback_thread_start writeback:writeback_thread_stop writeback:writeback_wait_iff_congested writeback:writeback_wake_background writeback:writeback_wake_thread As shown above, all of the events are printed. After this patch: $ perf list --raw-dump hw branch-instructions branch-misses bus-cycles cache-misses cache-references cpu-cycles instructions stalled-cycles-backend stalled-cycles-frontend As shown above, only the hw events are printed. Signed-off-by:
-
Yunlong Song authored
Do not need print_events_type or __print_events_type for listing hw/sw events, let print_symbol_events do its job instead. Moreover, print_symbol_events can also handle event_glob and name_only. Signed-off-by:
-
Yunlong Song authored
If the long_name of a 'struct option' is defined as NULL, --list-opts will incorrectly print '--(null)' in its output. As a result, '--(null)' will finally appear in the case of bash completion, e.g. 'perf record --'. Example: Before this patch: $ perf record --list-opts --event --filter --pid --tid --realtime --no-buffering --raw-samples --all-cpus --cpu --count --output --no-inherit --freq --mmap-pages --group --(null) --call-graph --verbose --quiet --stat --data --timestamp --period --no-samples --no-buildid-cache --no-buildid --cgroup --delay --uid --branch-any --branch-filter --weight --transaction --per-thread --intr-regs After this patch: $ perf record --list-opts --event --filter --pid --tid --realtime --no-buffering --raw-samples --all-cpus --cpu --count --output --no-inherit --freq --mmap-pages --group --call-graph --verbose --quiet --stat --data --timestamp --period --no-samples --no-buildid-cache --no-buildid --cgroup --delay --uid --branch-any --branch-filter --weight --transaction --per-thread --intr-regs Signed-off-by:
-
Yunlong Song authored
Distinguish the output of 'perf list --list-opts' or 'perf --list-cmds' with the next command prompt, which also happens in other cases (e.g. record, report ...). Example: Before this patch: $perf list --list-opts --raw-dump $ <-- the output and the next command prompt are at the same line After this patch: $perf list --list-opts --raw-dump $ <-- the new line Signed-off-by: -
Yunlong Song authored
If somebody happens to name an event with the beginning of 'tracepoint' (e.g. tracepoint_foo), then it will never be showed with perf list event_glob, thus we parse the argument 'tracepoint' more carefully for accuracy. Example: Before this patch: $ perf list tracepoint_foo:* jbd2:jbd2_start_commit [Tracepoint event] jbd2:jbd2_commit_locking [Tracepoint event] jbd2:jbd2_run_stats [Tracepoint event] block:block_rq_issue [Tracepoint event] block:block_bio_complete [Tracepoint event] block:block_bio_backmerge [Tracepoint event] block:block_getrq [Tracepoint event] ... ... As shown above, all of the tracepoint events are printed. In fact, the command's real intention is to print the events of tracepoint_foo. After this patch: $ perf list tracepoint_foo:* tracepoint_foo:tp_foo_enter [Tracepoint event] tracepoint_foo:tp_foo_exit [Tracepoint event] As shown above, only the events of tracepoint_foo are printed. Signed-off-by:
-
Yunlong Song authored
Sort the output according to ASCII character list (using strcmp), which supports both number sequence and alphabet sequence. Example: Before this patch: $ perf list 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] ... ... jbd2:jbd2_start_commit [Tracepoint event] jbd2:jbd2_commit_locking [Tracepoint event] jbd2:jbd2_run_stats [Tracepoint event] block:block_rq_issue [Tracepoint event] block:block_bio_complete [Tracepoint event] block:block_bio_backmerge [Tracepoint event] block:block_getrq [Tracepoint event] ... ... After this patch: $ perf list List of pre-defined events (to be used in -e): branch-instructions OR branches [Hardware event] branch-misses [Hardware event] bus-cycles [Hardware event] cache-misses [Hardware event] cache-references [Hardware event] cpu-cycles OR cycles [Hardware event] instructions [Hardware event] ... ... block:block_bio_backmerge [Tracepoint event] block:block_bio_complete [Tracepoint event] block:block_getrq [Tracepoint event] block:block_rq_issue [Tracepoint event] jbd2:jbd2_commit_locking [Tracepoint event] jbd2:jbd2_run_stats [Tracepoint event] jbd2:jbd2_start_commit [Tracepoint event] ... ... Signed-off-by:
-
Yunlong Song authored
The recent new patch "perf tools: Add new 'perf data' command" (commit 2245bf14 in acme's git repo perf/core) has caused a building error when compiling the source code of perf: cc1: warnings being treated as errors builtin-data.c:89: error: missing initializer builtin-data.c:89: error: (near initialization for ‘data_cmds[1].summary’) make[2]: *** [builtin-data.o] Error 1 make[2]: *** Waiting for unfinished jobs.... LD bench/perf-in.o LD tests/perf-in.o make[1]: *** [perf-in.o] Error 2 make: *** [all] Error 2 This patch fixes the building error above. Signed-off-by:
-
He Kuang authored
The minus operator has higher precedence than ?: Add parentheses around ?: fix this. Before this patch: $ echo 'p:myprobe do_sys_open' > /sys/kernel/debug/tracing/kprobe_events $ perf probe -l -k ../vmlinux kprobes:myprobe (on do_sys_open) After this patch: $ echo 'p:myprobe do_sys_open' > /sys/kernel/debug/tracing/kprobe_events $ perf probe -l -k ../vmlinux kprobes:myprobe (on do_sys_open@linux.git/fs/open.c) Signed-off-by: -
Kan Liang authored
Currently, the perf diff only works with same binaries. That's because it compares the symbol start address. It doesn't work if the perf.data comes from different binaries. This patch matches the symbol names. Actually, perf diff once intended to compare the symbol names. The commit as below can look for a pair by name. 604c5c92 (perf diff: Change the default sort order to "dso,symbol") However, at that time, perf diff used a global list of dsos. That means the binaries which has same name can only be loaded once. That's a problem for comparing different binaries. For example, we have an old binary and an updated binary. They very likely have same name and most of the functions, so only dsos from old binary will be loaded. When processing the data from updated binary, perf still use the symbol information from old binary. That's wrong. Then the commit as below used IP to replace symbol name. 9c443dfd ("perf diff: Fix support for all --sort combinations") >From that time, perf diff starts to compare the symbol address. The global dsos is discarded from a patch in 2010. a1645ce1 ("perf: 'perf kvm' tool for monitoring guest performance from host") However, at that time, perf diff already compared by address. So perf diff cannot work for different binaries as well. This patch actually rolls back the perf diff to original design. The document is also changed, so everybody knows the original design is to compare the symbol names. Here are some examples: The only difference between example_v1.c and example_v2.c is the location of f2 and f3. There is no change in behavior, but the previous perf diff display the wrong differential profile. example_v1.c noinline void f3(void) { volatile int i; for (i = 0; i < 10000;) { if(i%2) i++; else i++; } } noinline void f2(void) { volatile int a = 100, b, c; for (b = 0; b < 10000; b++) c = a * b; } noinline void f1(void) { f2(); f3(); } int main() { int i; for (i = 0; i < 100000; i++) f1(); } example_v2.c noinline void f2(void) { volatile int a = 100, b, c; for (b = 0; b < 10000; b++) c = a * b; } noinline void f3(void) { volatile int i; for (i = 0; i < 10000;) { if(i%2) i++; else i++; } } noinline void f1(void) { f2(); f3(); } int main() { int i; for (i = 0; i < 100000; i++) f1(); } [lk@localhost perf_diff]$ gcc example_v1.c -o example [lk@localhost perf_diff]$ perf record -o example_v1.data ./example [ perf record: Woken up 4 times to write data ] [ perf record: Captured and wrote 0.813 MB example_v1.data (~35522 samples) ] [lk@localhost perf_diff]$ gcc example_v2.c -o example [lk@localhost perf_diff]$ perf record -o example_v2.data ./example [ perf record: Woken up 4 times to write data ] [ perf record: Captured and wrote 0.824 MB example_v2.data (~36015 samples) ] Old perf diff result: [lk@localhost perf_diff]$ perf diff example_v1.data example_v2.data Event 'cycles' Baseline Delta Shared Object Symbol ........ ....... ................ ............................... [kernel.vmlinux] [k] __perf_event_task_sched_out 0.00% [kernel.vmlinux] [k] apic_timer_interrupt [kernel.vmlinux] [k] idle_cpu [kernel.vmlinux] [k] intel_pstate_timer_func [kernel.vmlinux] [k] native_read_msr_safe 0.00% [kernel.vmlinux] [k] native_read_tsc 0.00% [kernel.vmlinux] [k] native_write_msr_safe [kernel.vmlinux] [k] ntp_tick_length 0.00% [kernel.vmlinux] [k] rb_erase 0.00% [kernel.vmlinux] [k] tick_sched_timer 0.00% [kernel.vmlinux] [k] unmap_single_vma 0.00% [kernel.vmlinux] [k] update_wall_time 0.00% example [.] f1 46.24% example [.] f2 53.71% -7.55% example [.] f3 +53.81% example [.] f3 0.02% example [.] main New perf diff result: [lk@localhost perf_diff]$ perf diff example_v1.data example_v2.data [kernel.vmlinux] [k] __perf_event_task_sched_out 0.00% [kernel.vmlinux] [k] apic_timer_interrupt [kernel.vmlinux] [k] idle_cpu [kernel.vmlinux] [k] intel_pstate_timer_func [kernel.vmlinux] [k] native_read_msr_safe 0.00% [kernel.vmlinux] [k] native_read_tsc 0.00% [kernel.vmlinux] [k] native_write_msr_safe [kernel.vmlinux] [k] ntp_tick_length 0.00% [kernel.vmlinux] [k] rb_erase 0.00% [kernel.vmlinux] [k] tick_sched_timer 0.00% [kernel.vmlinux] [k] unmap_single_vma 0.00% [kernel.vmlinux] [k] update_wall_time 0.00% example [.] f1 46.24% -0.08% example [.] f2 53.71% +0.11% example [.] f3 0.02% example [.] main Signed-off-by:
Kan Liang <kan.liang@intel.com> Acked-by:
Jiri Olsa <jolsa@kernel.org> Acked-by:
-
Masami Hiramatsu authored
Add new buildid cache if the update target file is not cached. This can happen when an old binary is replaced by new one after caching the old one. In this case, user sees his operation just failed. But it does not look straight, since user just pass the binary "path", not "build-id". ---- # ./perf buildid-cache --add ./perf (update ./perf to new binary) # ./perf buildid-cache --update ./perf ./perf wasn't in the cache # ---- This patch adds given new binary to cache if the new binary is not cached. So we'll not see the above error. ---- # ./perf buildid-cache --add ./perf (update ./perf to new binary) # ./perf buildid-cache --update ./perf # ---- Signed-off-by:
-
Arnaldo Carvalho de Melo authored
We could end up returning 0 (Ok) with a NULL raw_path. Fix it. Acked-by:
-
- 26 Feb, 2015 5 commits
-
-
Masami Hiramatsu authored
Fix get_real_path to free allocated memory when comp_dir is used for complementing path and getting an error. Signed-off-by:
-
Masami Hiramatsu authored
Recent linux kernel provides a blacklist of the functions which can not be probed. perf probe can now check this blacklist before setting new events and indicate better error message for users. Without this patch, ---- # perf probe --add vmalloc_fault Added new event: Failed to write event: Invalid argument Error: Failed to add events. ---- With this patch ---- # perf probe --add vmalloc_fault Added new event: Warning: Skipped probing on blacklisted function: vmalloc_fault ---- Reported-by:Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20150219143113.14434.5387.stgit@localhost.localdomainSigned-off-by:
-
David Ahern authored
On Sparc64 perf-trace is failing in many spots due to extended load instructions being used on misaligned accesses. (gdb) run trace ls Starting program: /tmp/perf/perf trace ls [Thread debugging using libthread_db enabled] Detaching after fork from child process 169460. <ls output removed> Program received signal SIGBUS, Bus error. 0x000000000014f4dc in tp_field__u64 (field=0x4cc700, sample=0x7feffffa098) at builtin-trace.c:61 warning: Source file is more recent than executable. 61 TP_UINT_FIELD(64); (gdb) bt 0 0x000000000014f4dc in tp_field__u64 (field=0x4cc700, sample=0x7feffffa098) at builtin-trace.c:61 1 0x0000000000156ad4 in trace__sys_exit (trace=0x7feffffc268, evsel=0x4cc580, event=0xfffffc0104912000, sample=0x7feffffa098) at builtin-trace.c:1701 2 0x0000000000158c14 in trace__run (trace=0x7feffffc268, argc=1, argv=0x7fefffff360) at builtin-trace.c:2160 3 0x000000000015b78c in cmd_trace (argc=1, argv=0x7fefffff360, prefix=0x0) at builtin-trace.c:2609 4 0x0000000000107d94 in run_builtin (p=0x4549c8, argc=2, argv=0x7fefffff360) at perf.c:341 5 0x0000000000108140 in handle_internal_command (argc=2, argv=0x7fefffff360) at perf.c:400 6 0x0000000000108308 in run_argv (argcp=0x7feffffef2c, argv=0x7feffffef20) at perf.c:444 7 0x0000000000108728 in main (argc=2, argv=0x7fefffff360) at perf.c:559 (gdb) p *sample $1 = {ip = 4391276, pid = 169472, tid = 169472, time = 6303014583281250, addr = 0, id = 72082, stream_id = 18446744073709551615, period = 1, weight = 0, transaction = 0, cpu = 73, raw_size = 36, data_src = 84410401, flags = 0, insn_len = 0, raw_data = 0xfffffc010491203c, callchain = 0x0, branch_stack = 0x0, user_regs = {abi = 0, mask = 0, regs = 0x0, cache_regs = 0x7feffffa098, cache_mask = 0}, intr_regs = {abi = 0, mask = 0, regs = 0x0, cache_regs = 0x7feffffa098, cache_mask = 0}, user_stack = { offset = 0, size = 0, data = 0x0}, read = {time_enabled = 0, time_running = 0, {group = {nr = 0, values = 0x0}, one = {value = 0, id = 0}}}} (gdb) p *field $2 = {offset = 16, {integer = 0x14f4a8 <tp_field__u64>, pointer = 0x14f4a8 <tp_field__u64>}} sample->raw_data is guaranteed to not be 8-byte aligned because it is preceded by the size as a u3. So accessing raw data with an extended load instruction causes the SIGBUS. Resolve by using memcpy to a temporary variable of appropriate size. Signed-off-by: -
Ingo Molnar authored
Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo: New user selectable features: - Support recording running/enabled time in 'perf record' (Andi Kleen) - New tool: 'perf data' for converting perf.data to other formats, initially for the CTF (Common Trace Format) from LTTng (Jiri Olsa, Sebastian Siewior) User visible changes: - Only insert blank duration bracket when tracing syscalls in 'perf trace' (Arnaldo Carvalho de Melo) - Filter out the trace pid when no threads are specified in 'perf trace' (Arnaldo Carvalho de Melo) - Add 'perf trace' man page entry for --event (Arnaldo Carvalho de Melo) - Dump stack on segfaults in 'perf trace' (Arnaldo Carvalho de Melo) Infrastructure changes: - Introduce set_filter_pid and set_filter_pids methods in the evlist class (Arnaldo Carvalho de Melo) - Some perf_session untanglement patches, removing the need to pass a perf_session instance for things that are related to evlists, so that tools that don't deal with perf.data files like trace in live mode can make use of the ordered_events class (Arnaldo Carvalho de Melo) Signed-off-by:
-
Ingo Molnar authored
Signed-off-by:
-
- 25 Feb, 2015 1 commit
-
-
David Ahern authored
The recent build changes cause perf to not compile for sparc64 since the arch/sparc64/Build file does not exist: /home/dahern/kernels/linux.git/tools/build/Makefile.build:40: arch/sparc64/Build: No such file or directory Fix by converting the sparc64 RAW_ARCH to sparc ARCH -- similar to what is done for x86_64. Signed-off-by:
-
