perf tools: New comm infrastructure
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest4 patches 15 repliesIn review
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/4 | perf tools: Use an accessor to read thread commtools/ tools/ tools/ +10 more | +37 | 0 replies | |
| 2/4 | perf tools: Add time argument on comm settingtools/ tools/ tools/ +7 more | +60 | 0 replies | |
| 3/4 | perf tools: Add new comm infrastructuretools/ tools/ tools/ +2 more | +199 | 0 replies | |
| 4/4 | perf tools: Compare hists comm by addressestools/ | +2 | 3 replies |
Reviews trailers per patch
No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.
Cover letter [PATCH 0/4] perf tools: New comm infrastructure
The way we handle hists sorted by comm is to first gather them by tid then in the end merge/collapse hists that end up with the same comm.
But merging hists has shown some performances issues, especially with callchain where the operation can be very heavy.
So this new comm infrastructure aims at removing comm collapses. It brings two features:
1) Keep track of comms lifecycle by storing timestamps when the comms are set. This way we can map the precise comm to any thread:time couple. This only works if the PERF_SAMPLE_ID comes along comm and fork events, otherwise we only track the latest comm set for a thread.
This can provide us more precise comm sorted hists by distinguishing pre and post exec timeframes into seperate hists for a single thread.
Note that although the comm infrastructure is ready to do this, I haven't yet made the perf tools support that. It's a TODO entry.
2) Allocate comms only once instead of duplicating them for all threads sharing a same one. Two threads having the same comm should now point to the same string. As a result we can compare hists thread comm by address.
The big upside is that we can now live sort comm hists instead of collapsing them in the end of the processing.
I've seen very nice performance results on perf report. Roughly a 1.5x to 2x on perf report default stdio output with callchains.
You can try this branch:
git://git.kernel.org/pub/scm/linux/kernel/git/frederic/linux-dynticks.git
perf/comm
May be merging that with Namhyung callchains patches could provide some cumulative nice results.
thanks.
---
Frederic Weisbecker (4):
perf tools: Use an accessor to read thread comm perf tools: Add time argument on comm setting perf tools: Add new comm infrastructure perf tools: Compare hists comm by addresses
tools/perf/Makefile | 2 + tools/perf/builtin-kmem.c | 2 +- tools/perf/builtin-lock.c | 2 +- tools/perf/builtin-sched.c | 16 ++-- tools/perf/builtin-script.c | 6 +- tools/perf/builtin-top.c | 2 +- tools/perf/builtin-trace.c | 14 ++-- tools/perf/tests/code-reading.c | 2 +- tools/perf/tests/hists_link.c | 6 +- tools/perf/ui/browsers/hists.c | 10 +- tools/perf/util/comm.c | 107 ++++++++++++++++++++ tools/perf/util/comm.h | 20 ++++ tools/perf/util/event.c | 28 +++--- tools/perf/util/machine.c | 34 ++++--- tools/perf/util/machine.h | 18 ++- .../perf/util/scripting-engines/trace-event-perl.c | 2 +- .../util/scripting-engines/trace-event-python.c | 4 +- tools/perf/util/session.c | 2 +- tools/perf/util/sort.c | 13 ++- tools/perf/util/thread.c | 95 +++++++++++++---- tools/perf/util/thread.h | 8 +- 21 files changed, 293 insertions(+), 100 deletions(-)