On Fri, 2009-01-30 at 11:30 +0000, Richard Seguin wrote:
> More specifically,  in that link there will be another to obtain a
> valgrind log, please attach that as a separate attachment

Well, yeah.  There is definitely a memory leak in pidgin:

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND            
31463 brian     20   0  971m  41m  20m S    1  1.5   4:36.65 pidgin             

My experience with valgrind however is that it dramatically slows down
applications -- to the point of unusability.  Unfortunately, to that
effort, I use pidgin all day and I need it to be responsive.

I will give it a go, but I'm afraid I will not stick it out if it's
waaaay too slow.

Yeah.  Tried it.  Have been waiting literally minutes for the
application to even start but the windows have yet to even be painted
and I have a "memcheck" process consuming nearly 100% of a core here.  I
cannot function like this all day.

Oh wait.  OK.  It finally came up.  I let it run for a while.  It was
very slow.  One memcheck process ended up eating up 1GB of memory -- the
max I allow on my system here and got killed.  And then unfortunately,
the command was started again, overwriting the valgrind.log.

I'm afraid this is an unworkable situation for me to try to be
productive in.  Hopefully somebody else can oblige.

FWIW, my execution command line was:

$ G_SLICE=always-malloc G_DEBUG=gc-friendly  valgrind -v --tool=memcheck
--leak-check=full --num-callers=40 --log-file=valgrind.log pidgin

b.

-- 
segfault
https://bugs.launchpad.net/bugs/321258
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.

-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to