This is the mail archive of the
systemtap@sources.redhat.com
mailing list for the systemtap project.
Re: Exit probe bug
- From: Vara Prasad <prasadav at us dot ibm dot com>
- To: Karen Bennet <bennet at redhat dot com>
- Cc: Hien Nguyen <hien at us dot ibm dot com>, systemtap at sources dot redhat dot com
- Date: Wed, 09 Mar 2005 21:38:06 -0800
- Subject: Re: Exit probe bug
- References: <422E4C2F.3030803@us.ibm.com> <422FDA16.3040306@redhat.com>
Karen Bennet wrote:
Are you planning to submit this patch upstream for
acceptness and approval?
Yes, that is the plan. We need to come up with couple of good examples
of how exit probes are useful before we submit to LKML. I think Will's
page_fault example is a good one.
Hien Nguyen wrote:
I found a bug in the exit probe patch I sent out last week. The bug
is not effect the exit probe functionality. But it makes sense to
pass to the rprobe handler its counter part kprobe rather the
trampoline kprobe.
Hien.
------------------------------------------------------------------------
--- linux-2.6.10.works/arch/i386/kernel/kprobes.c.old 2005-03-08
16:56:34.113610333 -0800
+++ linux-2.6.10.works/arch/i386/kernel/kprobes.c 2005-03-08
16:13:30.000000000 -0800
@@ -197,7 +197,7 @@
}
if (ri->rp && !ri->rp->unregistering) {
- return ri->rp->handler(p, regs);
+ return ri->rp->handler(ri->rp->kprobe, regs);
}
return 0;
}