This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
[Bug kprobes/2183] kretprobe on schedule() leaks kretprobe_instances
- From: "jkenisto at us dot ibm dot com" <sourceware-bugzilla at sourceware dot org>
- To: systemtap at sources dot redhat dot com
- Date: 20 Jan 2006 00:14:04 -0000
- Subject: [Bug kprobes/2183] kretprobe on schedule() leaks kretprobe_instances
- References: <20060119233133.2183.jkenisto@us.ibm.com>
- Reply-to: sourceware-bugzilla at sourceware dot org
------- Additional Comments From jkenisto at us dot ibm dot com 2006-01-20 00:14 -------
Created an attachment (id=844)
--> (http://sourceware.org/bugzilla/attachment.cgi?id=844&action=view)
kretprobes test program
Attached is a module that sets return probes on all the functions named in the
probes[] array, and on exit, reports the number of calls, returns, and
kretprobe misses for each function.
On i386, I do indeed see a lot of misses for schedule() with maxactive set to
5, as in rp.c, but when I crank maxactive up to 5000, I see no misses for
schedule(). We would expect a lot of tasks to be stopped in schedule() at any
one time. If that number exceeds maxactive, we start seeing misses.
--
http://sourceware.org/bugzilla/show_bug.cgi?id=2183
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.