Bug 11576 - per-process perf events
Summary: per-process perf events
Status: RESOLVED FIXED
Alias: None
Product: systemtap
Classification: Unclassified
Component: translator (show other bugs)
Version: unspecified
: P2 normal
Target Milestone: ---
Assignee: Stan Cox
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-05-06 19:01 UTC by Frank Ch. Eigler
Modified: 2013-01-28 19:45 UTC (History)
0 users

See Also:
Host:
Target:
Build:
Last reconfirmed:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Frank Ch. Eigler 2010-05-06 19:01:49 UTC 
In runtime/perf.c, it should be a straightforward thing to
detect that we're running in target-process mode (stap -c/-x),
and interpret that as a request to bind the perf.* events
only to that process.  (One would pass a -1 cpu and a proper
pid number to perf_event_create_kernel_counter.)

OTOH, this could be an option at the probe point level, as in
probe process(NNN).perf.* ?
Comment 1 Frank Ch. Eigler 2012-11-05 16:36:57 UTC 
In order to enjoy the tapset/perf.* aliasing for perfctr names,
instead of process(...).perf*, the new probe points could be named
perf.*.process("..."), relying on bug #14313's suffix-passthrough
magic.
Comment 2 Stan Cox 2013-01-28 19:45:16 UTC 
Add process("NAME") to perf probe.
commit: 6a8fe809a8c