kernel.git/tools/perf/builtin-trace.c, branch linux-3.13.y

perf trace: Decode architecture-specific signal numbers

2014-03-24T04:44:05Z

commit 02c5bb4a352a4cca56e9b5d3a2a57d61062eb2e1 upstream. SIGSTKFLT is not defined on alpha, mips or sparc. SIGEMT and SIGSWI are defined on some architectures and should be decoded here if so. Signed-off-by: Ben Hutchings Fixes: 8bad5b0abfdb ('perf trace: Beautify signal number arg in several syscalls') Cc: Ingo Molnar Cc: Paul Mackerras Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1391648441.3003.101.camel@deadeye.wl.decadent.org.uk Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Greg Kroah-Hartman

perf trace: Add fallback definition of EFD_SEMAPHORE

2014-03-07T06:06:27Z

commit 79d26a6a19ace19faabf8d8d27d3430be2e26d34 upstream. glibc 2.17 is missing this on sparc, despite the fact that it's not architecture-specific. Signed-off-by: Ben Hutchings Fixes: 49af9e93adfa ('perf trace: Beautify eventfd2 'flags' arg') Cc: Cc: Ingo Molnar Cc: Paul Mackerras Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1391648435.3003.100.camel@deadeye.wl.decadent.org.uk Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Greg Kroah-Hartman

perf trace: Fix ioctl 'request' beautifier build problems on !(i386 || x86_64) arches

2014-03-07T06:06:21Z

commit 844ae5b46c08dbc7ba695b543c023f9cf3bbf9ff upstream. Supporting decoding the ioctl 'request' parameter needs more work to properly support more architectures, the current approach doesn't work on at least powerpc and sparc, as reported by Ben Hutchings in http://lkml.kernel.org/r/1391593985.3003.48.camel@deadeye.wl.decadent.org.uk . Work around that by making it to be ifdefed for the architectures known to work with the current, limited approach, i386 and x86_64 till better code is written. Reported-by: Ben Hutchings Acked-by: Ben Hutchings Cc: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/n/tip-ss04k11insqlu329xh5g02q0@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Greg Kroah-Hartman

perf trace: Tweak summary output

2013-11-14T18:59:20Z

Tweak the summary output as suggested by Ingo Molnar: [penberg@localhost ~]$ perf trace -a --duration 10000 --summary -- sleep 1 ^C Summary of events: Xorg (817), 148 events, 0.0%, 0.000 msec syscall calls min avg max stddev (msec) (msec) (msec) (%) --------------- -------- --------- --------- --------- ------ read 7 0.002 0.004 0.011 32.00% rt_sigprocmask 40 0.001 0.001 0.002 1.31% ioctl 6 0.002 0.003 0.005 19.45% writev 7 0.004 0.018 0.059 43.76% select 9 0.000 74.513 507.869 74.61% setitimer 4 0.001 0.002 0.002 10.08% Suggested-by: Ingo Molnar Signed-off-by: Pekka Enberg Acked-by: Ingo Molnar Cc: David Ahern Cc: Ingo Molnar Link: http://lkml.kernel.org/r/1384345308-24404-1-git-send-email-penberg@kernel.org Signed-off-by: Arnaldo Carvalho de Melo

perf target: Shorten perf_target to target

2013-11-12T19:51:03Z

Getting unwieldly long, for this app domain should be descriptive enough and the use of __ to separate the class from the method names should help with avoiding clashes with other code bases. Reported-by: David Ahern Suggested-by: Ingo Molnar Cc: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/20131112113427.GA4053@ghostprotocols.net Signed-off-by: Arnaldo Carvalho de Melo

perf trace: Add summary only option

2013-11-12T19:24:38Z

Per request from Pekka make --summary a summary only option meaning do not show the individual system calls. Add another option to see all syscalls along with the summary. In addition use 's' and 'S' as shortcuts for the options. Requested-by: Pekka Enberg Signed-off-by: David Ahern Tested-by: Pekka Enberg Cc: Adrian Hunter Cc: Ingo Molnar Cc: Jiri Olsa Cc: Namhyung Kim Cc: Pekka Enberg Link: http://lkml.kernel.org/r/1384273875-3751-1-git-send-email-dsahern@gmail.com Signed-off-by: Arnaldo Carvalho de Melo

perf trace: Simplify '--summary' output

2013-11-12T16:00:38Z

The output of 'perf trace --summary' tries to be too cute with formatting and makes it very hard to read. Simplify it in the spirit of "strace -c": [penberg@localhost libtrading]$ perf trace -a --duration 10000 --summary -- sleep 1 ^C Summary of events: dbus-daemon (555), 10 events, 0.0%, 0.000 msec msec/call syscall calls min avg max stddev --------------- -------- -------- -------- -------- ------ sendmsg 2 0.002 0.005 0.008 55.00 recvmsg 2 0.002 0.003 0.005 44.00 epoll_wait 1 0.000 0.000 0.000 0.00 NetworkManager (667), 56 events, 0.0%, 0.000 msec msec/call syscall calls min avg max stddev --------------- -------- -------- -------- -------- ------ poll 2 0.000 0.002 0.003 100.00 sendmsg 10 0.004 0.007 0.016 15.41 recvmsg 16 0.002 0.003 0.005 8.24 zfs-fuse (669), 4 events, 0.0%, 0.000 msec msec/call syscall calls min avg max stddev --------------- -------- -------- -------- -------- ------ futex 2 0.000 0.001 0.002 100.00 Signed-off-by: Pekka Enberg Cc: David Ahern Cc: Ingo Molnar Link: http://lkml.kernel.org/r/1384267334-18953-1-git-send-email-penberg@kernel.org Signed-off-by: Arnaldo Carvalho de Melo

perf trace: Change syscall summary duration order

2013-11-12T16:00:38Z

Switch duration order to minimum, average, maximum for the '--summary' command line option because it's more natural to read. Signed-off-by: Pekka Enberg Acked-by: David Ahern Cc: David Ahern Cc: Ingo Molnar Link: http://lkml.kernel.org/r/1384265410-12344-1-git-send-email-penberg@kernel.org Signed-off-by: Arnaldo Carvalho de Melo

perf trace: Fix segfault on perf trace -i perf.data

2013-11-12T16:00:37Z

When replaying a previous record session, it'll get a segfault since it doesn't initialize raw_syscalls enter/exit tracepoint's evsel->priv for caching the format fields. So fix it by properly initializing sys_enter/exit evsels that comes from reading the perf.data file header. Signed-off-by: Namhyung Kim Cc: David Ahern Cc: Ingo Molnar Cc: Paul Mackerras Cc: Pekka Enberg Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1384237500-22991-2-git-send-email-namhyung@kernel.org [ Split the syscall tp field caching part in the previous patch ] Signed-off-by: Arnaldo Carvalho de Melo

perf trace: Separate tp syscall field caching into init routine to be reused

2013-11-12T16:00:36Z

We need to set this in evsels coming out of a perf.data file header, not just for new ones created for live sessions. So separate the code that caches the syscall entry/exit tracepoint format fields into a new function that will be used in the next changeset. Signed-off-by: Namhyung Kim Cc: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Jiri Olsa Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/20131112115700.GC4053@ghostprotocols.net Signed-off-by: Arnaldo Carvalho de Melo