From 0d37aa34f8806bb443dd3c8621fd9bdbb50c58bb Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Thu, 19 Jan 2012 14:08:15 -0200 Subject: perf tools: Introduce per user view The new --uid command line option will show only the tasks for a given user, using the proc interface to figure out the existing tasks. Kernel work is needed to close races at startup, but this should already be useful in many use cases. Cc: David Ahern Cc: Frederic Weisbecker Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/n/tip-bdnspm000gw2l984a2t53o8z@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/perf.h | 1 + 1 file changed, 1 insertion(+) (limited to 'tools/perf/perf.h') diff --git a/tools/perf/perf.h b/tools/perf/perf.h index 64f8bee31ce..92af1688bae 100644 --- a/tools/perf/perf.h +++ b/tools/perf/perf.h @@ -188,6 +188,7 @@ void pthread__unblock_sigwinch(void); struct perf_record_opts { pid_t target_pid; pid_t target_tid; + uid_t uid; bool call_graph; bool group; bool inherit_stat; -- cgit v1.2.3-18-g5258 From b52956c961be3a04182ae7b776623531601e0fb7 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 8 Feb 2012 09:32:52 -0700 Subject: perf tools: Allow multiple threads or processes in record, stat, top Allow a user to collect events for multiple threads or processes using a comma separated list. e.g., collect data on a VM and its vhost thread: perf top -p 21483,21485 perf stat -p 21483,21485 -ddd perf record -p 21483,21485 or monitoring vcpu threads perf top -t 21488,21489 perf stat -t 21488,21489 -ddd perf record -t 21488,21489 Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1328718772-16688-1-git-send-email-dsahern@gmail.com Signed-off-by: David Ahern Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/perf.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'tools/perf/perf.h') diff --git a/tools/perf/perf.h b/tools/perf/perf.h index 92af1688bae..deb17dba4a5 100644 --- a/tools/perf/perf.h +++ b/tools/perf/perf.h @@ -186,8 +186,8 @@ extern const char perf_version_string[]; void pthread__unblock_sigwinch(void); struct perf_record_opts { - pid_t target_pid; - pid_t target_tid; + const char *target_pid; + const char *target_tid; uid_t uid; bool call_graph; bool group; -- cgit v1.2.3-18-g5258 From 7e1ccd3804281fc0755eb726b654469c40a96d89 Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Thu, 9 Feb 2012 16:12:38 +0100 Subject: perf tools: cleanup initialization of attr->size The perf_event_attr size needs to be initialized in all cases because it captures the ABI version. This patch moves the initialization of the field from the perf_event_open() syscall stub to its proper location in the event_attr_init(). Cc: Ingo Molnar Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20120209151238.GA10272@quad Signed-off-by: Stephane Eranian Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/perf.h | 1 - 1 file changed, 1 deletion(-) (limited to 'tools/perf/perf.h') diff --git a/tools/perf/perf.h b/tools/perf/perf.h index deb17dba4a5..03a04561814 100644 --- a/tools/perf/perf.h +++ b/tools/perf/perf.h @@ -167,7 +167,6 @@ sys_perf_event_open(struct perf_event_attr *attr, pid_t pid, int cpu, int group_fd, unsigned long flags) { - attr->size = sizeof(*attr); return syscall(__NR_perf_event_open, attr, pid, cpu, group_fd, flags); } -- cgit v1.2.3-18-g5258 From 0c9781280fb672ca09c997df3f14ba506bbdb977 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 14 Feb 2012 14:05:30 -0200 Subject: perf tools: Handle kernels that don't support attr.exclude_{guest,host} Just fall back to resetting those fields, if set, warning the user that that feature is not available. If guest samples appear they will just be discarded because no struct machine will be found and thus the event will be accounted as not handled and dropped, see 0c09571. Reported-by: Namhyung Kim Tested-by: Joerg Roedel Cc: David Ahern Cc: Frederic Weisbecker Cc: Joerg Roedel Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/n/tip-vuwxig36mzprl5n7nzvnxxsh@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/perf.h | 1 + 1 file changed, 1 insertion(+) (limited to 'tools/perf/perf.h') diff --git a/tools/perf/perf.h b/tools/perf/perf.h index 03a04561814..8b9c43635f4 100644 --- a/tools/perf/perf.h +++ b/tools/perf/perf.h @@ -199,6 +199,7 @@ struct perf_record_opts { bool sample_address; bool sample_time; bool sample_id_all_avail; + bool exclude_guest_missing; bool system_wide; bool period; unsigned int freq; -- cgit v1.2.3-18-g5258 From 808e122630d45a7f036d25582474d70548a87e2c Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 14 Feb 2012 14:18:57 -0200 Subject: perf tools: Invert the sample_id_all logic Instead of requiring that users of perf_record_opts set .sample_id_all_avail to true, just invert the logic, using .sample_id_all_missing, that doesn't need to be explicitely initialized since gcc will zero members ommitted in a struct initialization. Just like the newly introduced .exclude_{guest,host} feature test. Cc: David Ahern Cc: Frederic Weisbecker Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/n/tip-ab772uzk78cwybihf0vt7kxw@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/perf.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'tools/perf/perf.h') diff --git a/tools/perf/perf.h b/tools/perf/perf.h index 8b9c43635f4..f0227e93665 100644 --- a/tools/perf/perf.h +++ b/tools/perf/perf.h @@ -198,7 +198,7 @@ struct perf_record_opts { bool raw_samples; bool sample_address; bool sample_time; - bool sample_id_all_avail; + bool sample_id_all_missing; bool exclude_guest_missing; bool system_wide; bool period; -- cgit v1.2.3-18-g5258 From b5387528f31d98acedf06e930554b563d87e2383 Mon Sep 17 00:00:00 2001 From: Roberto Agostino Vitillo Date: Thu, 9 Feb 2012 23:21:01 +0100 Subject: perf tools: Add code to support PERF_SAMPLE_BRANCH_STACK This patch adds: - ability to parse samples with PERF_SAMPLE_BRANCH_STACK - sort on branches (dso_from, symbol_from, dso_to, symbol_to, mispredict) - build histograms on branches Signed-off-by: Roberto Agostino Vitillo Signed-off-by: Stephane Eranian Cc: peterz@infradead.org Cc: acme@redhat.com Cc: robert.richter@amd.com Cc: ming.m.lin@intel.com Cc: andi@firstfloor.org Cc: asharma@fb.com Cc: vweaver1@eecs.utk.edu Cc: khandual@linux.vnet.ibm.com Cc: dsahern@gmail.com Link: http://lkml.kernel.org/r/1328826068-11713-12-git-send-email-eranian@google.com Signed-off-by: Ingo Molnar --- tools/perf/perf.h | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) (limited to 'tools/perf/perf.h') diff --git a/tools/perf/perf.h b/tools/perf/perf.h index f0227e93665..358f40135c4 100644 --- a/tools/perf/perf.h +++ b/tools/perf/perf.h @@ -179,6 +179,23 @@ struct ip_callchain { u64 ips[0]; }; +struct branch_flags { + u64 mispred:1; + u64 predicted:1; + u64 reserved:62; +}; + +struct branch_entry { + u64 from; + u64 to; + struct branch_flags flags; +}; + +struct branch_stack { + u64 nr; + struct branch_entry entries[0]; +}; + extern bool perf_host, perf_guest; extern const char perf_version_string[]; -- cgit v1.2.3-18-g5258 From bdfebd848f2a14e639031a0b0e61d7c7ee5e5fd2 Mon Sep 17 00:00:00 2001 From: Roberto Agostino Vitillo Date: Thu, 9 Feb 2012 23:21:02 +0100 Subject: perf record: Add support for sampling taken branch This patch adds a new option to enable taken branch stack sampling, i.e., leverage the PERF_SAMPLE_BRANCH_STACK feature of perf_events. There is a new option to active this mode: -b. It is possible to pass a set of filters to select the type of branches to sample. The following filters are available: - any : any type of branches - any_call : any function call or system call - any_ret : any function return or system call return - any_ind : any indirect branch - u: only when the branch target is at the user level - k: only when the branch target is in the kernel - hv: only when the branch target is in the hypervisor Filters can be combined by passing a comma separated list to the option: $ perf record -b any_call,u -e cycles:u branchy Signed-off-by: Roberto Agostino Vitillo Signed-off-by: Stephane Eranian Cc: peterz@infradead.org Cc: acme@redhat.com Cc: robert.richter@amd.com Cc: ming.m.lin@intel.com Cc: andi@firstfloor.org Cc: asharma@fb.com Cc: vweaver1@eecs.utk.edu Cc: khandual@linux.vnet.ibm.com Cc: dsahern@gmail.com Link: http://lkml.kernel.org/r/1328826068-11713-13-git-send-email-eranian@google.com Signed-off-by: Ingo Molnar --- tools/perf/perf.h | 1 + 1 file changed, 1 insertion(+) (limited to 'tools/perf/perf.h') diff --git a/tools/perf/perf.h b/tools/perf/perf.h index 358f40135c4..eec392e4806 100644 --- a/tools/perf/perf.h +++ b/tools/perf/perf.h @@ -222,6 +222,7 @@ struct perf_record_opts { unsigned int freq; unsigned int mmap_pages; unsigned int user_freq; + int branch_stack; u64 default_interval; u64 user_interval; const char *cpu_list; -- cgit v1.2.3-18-g5258