kernel/linux.git/tools/perf/builtin-diff.c, branch v2.6.38

perf symbols: Add symfs option for off-box analysis using specified tree

2010-12-21T22:17:51+00:00

The symfs argument allows analysis of perf.data file using a locally accessible filesystem tree with debug symbols - e.g., tree created during image builds, sshfs mount, loop mounted KVM disk images, USB keys, initrds, etc. Anything with an OS tree can be analyzed from anywhere without the need to populate a local data store with build-ids. Commiter notes: o Fixed up symfs="/" variants handling. o prefixed DSO__ORIG_GUEST_KMODULE case with symfs too, avoiding use of files outside the symfs directory. LKML-Reference: <1291926427-28846-1-git-send-email-daahern@cisco.com> Signed-off-by: David Ahern Signed-off-by: Arnaldo Carvalho de Melo

perf record,report,annotate,diff: Process events in order

2010-12-21T22:17:51+00:00

This patch changes perf report to ask for the ID info on all events be default if recording from multiple CPUs. Perf report, annotate and diff will now process the events in order if the kernel is able to provide timestamps on all events. This ensures that events such as COMM and MMAP which are necessary to correctly interpret samples are processed prior to those samples so that they are attributed correctly. Before: # perf record ./cachetest # perf report # Events: 6K cycles # # Overhead Command Shared Object Symbol # ........ ....... ................. ............................... # 74.11% :3259 [unknown] [k] 0x4a6c 1.50% cachetest ld-2.11.2.so [.] 0x1777c 1.46% :3259 [kernel.kallsyms] [k] .perf_event_mmap_ctx 1.25% :3259 [kernel.kallsyms] [k] restore 0.74% :3259 [kernel.kallsyms] [k] ._raw_spin_lock 0.71% :3259 [kernel.kallsyms] [k] .filemap_fault 0.66% :3259 [kernel.kallsyms] [k] .memset 0.54% cachetest [kernel.kallsyms] [k] .sha_transform 0.54% :3259 [kernel.kallsyms] [k] .copy_4K_page 0.54% :3259 [kernel.kallsyms] [k] .find_get_page 0.52% :3259 [kernel.kallsyms] [k] .trace_hardirqs_off 0.50% :3259 [kernel.kallsyms] [k] .__do_fault After: # perf report # Events: 6K cycles # # Overhead Command Shared Object Symbol # ........ ....... ................. ............................... # 44.28% cachetest cachetest [.] sumArrayNaive 22.53% cachetest cachetest [.] sumArrayOptimal 6.59% cachetest ld-2.11.2.so [.] 0x1777c 2.13% cachetest [unknown] [k] 0x340 1.46% cachetest [kernel.kallsyms] [k] .perf_event_mmap_ctx 1.25% cachetest [kernel.kallsyms] [k] restore 0.74% cachetest [kernel.kallsyms] [k] ._raw_spin_lock 0.71% cachetest [kernel.kallsyms] [k] .filemap_fault 0.66% cachetest [kernel.kallsyms] [k] .memset 0.54% cachetest [kernel.kallsyms] [k] .copy_4K_page 0.54% cachetest [kernel.kallsyms] [k] .find_get_page 0.54% cachetest [kernel.kallsyms] [k] .sha_transform 0.52% cachetest [kernel.kallsyms] [k] .trace_hardirqs_off 0.50% cachetest [kernel.kallsyms] [k] .__do_fault Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Thomas Gleixner LKML-Reference: <1291872833-839-1-git-send-email-imunsie@au1.ibm.com> Signed-off-by: Ian Munsie Signed-off-by: Arnaldo Carvalho de Melo

perf session: Fallback to unordered processing if no sample_id_all

2010-12-21T22:17:51+00:00

If we are running the new perf on an old kernel without support for sample_id_all, we should fall back to the old unordered processing of events. If we didn't than we would *always* process events without timestamps out of order, whether or not we hit a reordering race. In other words, instead of there being a chance of not attributing samples correctly, we would guarantee that samples would not be attributed. While processing all events without timestamps before events with timestamps may seem like an intuitive solution, it falls down as PERF_RECORD_EXIT events would also be processed before any samples. Even with a workaround for that case, samples before/after an exec would not be attributed correctly. This patch allows commands to indicate whether they need to fall back to unordered processing, so that commands that do not care about timestamps on every event will not be affected. If we do fallback, this will print out a warning if report -D was invoked. This patch adds the test in perf_session__new so that we only need to test once per session. Commands that do not use an event_ops (such as record and top) can simply pass NULL in it's place. Acked-by: Thomas Gleixner Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Peter Zijlstra Cc: Thomas Gleixner LKML-Reference: <1291951882-sup-6069@au1.ibm.com> Signed-off-by: Ian Munsie Signed-off-by: Arnaldo Carvalho de Melo

perf session: Parse sample earlier

2010-12-05T01:05:19+00:00

At perf_session__process_event, so that we reduce the number of lines in eache tool sample processing routine that now receives a sample_data pointer already parsed. This will also be useful in the next patch, where we'll allow sample the identity fields in MMAP, FORK, EXIT, etc, when it will be possible to see (cpu, timestamp) just after before every event. Also validate callchains in perf_session__process_event, i.e. as early as possible, and keep a counter of the number of events discarded due to invalid callchains, warning the user about it if it happens. There is an assumption that was kept that all events have the same sample_type, that will be dealt with in the future, when this preexisting limitation will be removed. Tested-by: Thomas Gleixner Reviewed-by: Thomas Gleixner Acked-by: Ian Munsie Acked-by: Thomas Gleixner Cc: Frédéric Weisbecker Cc: Ian Munsie Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Paul Mackerras Cc: Stephane Eranian LKML-Reference: <1291318772-30880-4-git-send-email-acme@infradead.org> Signed-off-by: Arnaldo Carvalho de Melo

perf diff: Fix displacement and modules options short flag

2010-12-01T20:22:45+00:00

The --displacement and --modules options to perf diff both use -m as a short flag. Change --displacement to use -M since other perf commands use -m, --modules. Cc: Ingo Molnar Cc: Paul Mackerras Cc: Peter Zijlstra LKML-Reference: <1291168642-11402-4-git-send-email-shawn.bohrer@gmail.com> Signed-off-by: Shawn Bohrer Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Remove unneeded code for tracking the cwd in perf sessions

2010-07-27T14:46:12+00:00

Tidy-up patch to remove some code and struct perf_session data members which are no longer needed due to the previous patch: "perf tools: Don't abbreviate file paths relative to the cwd". LKML-Reference: Signed-off-by: Dave Martin Signed-off-by: Arnaldo Carvalho de Melo

perf tools: Make event__preprocess_sample parse the sample

2010-06-05T12:35:19+00:00

Simplifying the tools that were using both in sequence and allowing upcoming simplifications, such as Arun's patch to sort by cpus. Cc: David S. Miller Cc: Frédéric Weisbecker Cc: Ingo Molnar Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Tom Zanussi LKML-Reference: Signed-off-by: Arnaldo Carvalho de Melo

perf report: Report number of events, not samples

2010-05-14T17:19:35+00:00

Number of samples is meaningless after we switched to auto-freq, so report the number of events, i.e. not the sum of the different periods, but the number PERF_RECORD_SAMPLE emitted by the kernel. While doing this I noticed that naming "count" to the sum of all the event periods can be confusing, so rename it to .period, just like in struct sample.data, so that we become more consistent. This helps with the next step, that was to record in struct hist_entry the number of sample events for each instance, we need that because we use it to generate the number of events when applying filters to the tree of hist entries like it is being done in the TUI report browser. Suggested-by: Ingo Molnar Cc: Frédéric Weisbecker Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Tom Zanussi LKML-Reference: Signed-off-by: Arnaldo Carvalho de Melo

perf hist: Clarify events_stats fields usage

2010-05-14T16:16:55+00:00

The events_stats.total field is too generic, rename it to .total_period, and also add a comment explaining that it is the sum of all the .period fields in samples, that is needed because we use auto-freq to avoid sampling artifacts. Ditto for events_stats.lost, that is the sum of all lost_event.lost fields, i.e. the number of events the kernel dropped. Looking at the users, builtin-sched.c can make use of these fields and stop doing it again. Cc: Frédéric Weisbecker Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Tom Zanussi LKML-Reference: Signed-off-by: Arnaldo Carvalho de Melo

perf hist: Introduce hists class and move lots of methods to it

2010-05-10T16:13:49+00:00

In cbbc79a we introduced support for multiple events by introducing a new "event_stat_id" struct and then made several perf_session methods receive a point to it instead of a pointer to perf_session, and kept the event_stats and hists rb_tree in perf_session. While working on the new newt based browser, I realised that it would be better to introduce a new class, "hists" (short for "histograms"), renaming the "event_stat_id" struct and the perf_session methods that were really "hists" methods, as they manipulate only struct hists members, not touching anything in the other perf_session members. Other optimizations, such as calculating the maximum lenght of a symbol name present in an hists instance will be possible as we add them, avoiding a re-traversal just for finding that information. The rationale for the name "hists" to replace "event_stat_id" is that we may have multiple sets of hists for the same event_stat id, as, for instance, the 'perf diff' tool has, so event stat id is not what characterizes what this struct and the functions that manipulate it do. Cc: Eric B Munson Cc: Frédéric Weisbecker Cc: Mike Galbraith Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Tom Zanussi LKML-Reference: Signed-off-by: Arnaldo Carvalho de Melo