This is the mail archive of the
gdb-prs@sourceware.org
mailing list for the GDB project.
[Bug threads/13251] frequent multithreading slows down GDB to theextent of making it unusable
- From: "tromey at redhat dot com" <sourceware-bugzilla at sourceware dot org>
- To: gdb-prs at sourceware dot org
- Date: Wed, 09 Nov 2011 20:36:21 +0000
- Subject: [Bug threads/13251] frequent multithreading slows down GDB to theextent of making it unusable
- Auto-submitted: auto-generated
- References: <bug-13251-4717@http.sourceware.org/bugzilla/>
http://sourceware.org/bugzilla/show_bug.cgi?id=13251
Tom Tromey <tromey at redhat dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |tromey at redhat dot com
--- Comment #1 from Tom Tromey <tromey at redhat dot com> 2011-11-09 20:36:21 UTC ---
(In reply to comment #0)
> This is a follow-up on a bug that was submitted by me in Feb 2011 and never
> resolved fully. I waited for 7.3 and it is still there.
Which PR is this?
> The problem: the application program spawns a great many threads (in the
> thousands, altogether) which each lives for a few ms. The threads are started
> by boost::thread.
>
> Each thread seems to allocate some memory in gdb which is never freed fully.
> Further, the gdb process does something on a single CPU which grows longer in
> time the more threads have been started (and stopped). With the growing
> overhead
> on a single CPU, performance drops rapidly. In the end, the process grinds to
> a halt, with GDB using all of one CPUÂs capacity.
I wrote a simple program that has a loop creating and joining short-lived
(1 second) threads. I ran this for quite some time but didn't see any
performance degradation or memory increase.
Do you have, or can you construct, a test case we can use?
The fewer dependencies the better.
What platform are you on? I am using x86-64 Fedora 15.
--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.