This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
[Bug translator/10788] -t statistics report on derived statement probes instead of mark probes
- From: "fche at redhat dot com" <sourceware-bugzilla at sourceware dot org>
- To: systemtap at sources dot redhat dot com
- Date: 17 Oct 2009 12:16:42 -0000
- Subject: [Bug translator/10788] -t statistics report on derived statement probes instead of mark probes
- References: <20091016124037.10788.mjw@redhat.com>
- Reply-to: sourceware-bugzilla at sourceware dot org
------- Additional Comments From fche at redhat dot com 2009-10-17 12:16 -------
(In reply to comment #3)
> Yes, if that was the case, then having more granular statistics would be nice to
> have. But in this case you can see the variance even with really simple probes
> like: stap -t -e 'probe syscall.* { printf(".") }' -c 'sleep 5'
> [... dots ...]
> probe syscall.* (<input>:1:1), hits: 1236852, cycles: 440min/533avg/20108max
> Maybe this deserves its own bug report if it doesn't exist yet.
I don't think so -- the average here is close to the minimum, over a million+
hits. The maximum translates to a ~hundred cache misses, which is plausible
considering the amount of code & data involved in printf().
--
http://sourceware.org/bugzilla/show_bug.cgi?id=10788
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.