* kernel crash when probe context_switch function
@ 2005-11-23 9:09 Mao, Bibo
2005-11-23 9:31 ` Martin Hunt
0 siblings, 1 reply; 2+ messages in thread
From: Mao, Bibo @ 2005-11-23 9:09 UTC (permalink / raw)
To: systemtap
Hi,
When I run ./testsuite/buildok/six.stp by "stap -gv" command, it
crashed, the script is like this
probe kernel.inline("context_switch") {
log ("found an inline function")
}
Does there exist some critical code area in kernel code where probe
point can not be put? And such as context_switch, it can not be probed
by trap method of kprobe.
Regards
Bibo,mao
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: kernel crash when probe context_switch function
2005-11-23 9:09 kernel crash when probe context_switch function Mao, Bibo
@ 2005-11-23 9:31 ` Martin Hunt
0 siblings, 0 replies; 2+ messages in thread
From: Martin Hunt @ 2005-11-23 9:31 UTC (permalink / raw)
To: Mao, Bibo; +Cc: systemtap
See http://sources.redhat.com/bugzilla/show_bug.cgi?id=1564
To be safe in ALL possible places, I'm changing the IO code to simply
write the data to a buffer and set a flag. (This is what oprofile does.)
Once I finish with some buffering changes, I'll get that checked in.
Until then, avoid IO in scheduler functions.
Martin
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2005-11-23 9:31 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2005-11-23 9:09 kernel crash when probe context_switch function Mao, Bibo
2005-11-23 9:31 ` Martin Hunt
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).