[omniORB] Re: Testing OmniORB4 for memory leaks on Linux RH9 w/
valgrind
Sean Parker
supinlick at yahoo.com
Thu Jun 9 17:40:56 BST 2005
Bruce -
> --- "Visscher, Bruce" <VISSCHB at RJRT.com> wrote:
<snip>>
> Sounds like a Linux pthreads issue to me. Do you loose
> 8K for each
> thread allocated?
I agree - exact same thing happens in my framework where I
use pthreads - it's just once per app run regardless of how
many threads are used (one of my tests starts hundreds of
threads)...
> If that isn't the case, then maybe
> it's just
> a one time allocation so doesn't really qualify as a
> leak.
>
Of course I'd preferably like to run valgrind "out of the
box" and have a 'clean' app result in 'no valgrind output
listings'
> > 2) several "use of uninitialized bytes"
>
> I'm not familiar with valgrind, but a couple of places
> where these
> are reported frequently are:
>
> (1) Complete red herring. E.g., instances of
> std::allocator which
> should really be empty but (depending on the
> implementation of both
> the compiler and the library) are not optimized away. Of
> course,
> omniORB doesn't utilize std::allocator so this has
> nothing to do
> with omniORB. But it could be something similar.
>
> (2) Part of the IOR used to be uninitialized memory.
> iirc, it was
> part of the "private" section of the stringified IOR so
> it can be
> anything the implementation desires.
>
In my code, I get this uninitialized-issue in lots of
ridiculous places, like:
while( x >> y )
{...}
valgrind complains that the while value is unitialized -
certainly a valgrind question... I was hoping someone would
have insight like - "Oh all you need to do to avoid that
is..." or "technically you should do ... before..."
> I am actually glad that I wasn't able to completely
> dissuade you from
> pursuing this.
"I come from a line of mentors that don't give up..." :-)
> Just thought you should know you're not
> the first
> to think about this.
I'm glad I'm not the first! so I can focus on ME being the
idiot! :-)
> Maybe you will find a bug no one
> else has.
>
> Bruce
>
If I do, I'd be 'the man' - but I'm not a coding God by any
means...
Sean
> -----------------------------------------
> CONFIDENTIALITY NOTE: This e-mail message, including any
> attachment(s),
> contains information that may be confidential, protected
> by the attorney-
> client or other legal privileges, and/or proprietary
> non-public
> information. If you are not an intended recipient of
> this message or an
> authorized assistant to an intended recipient, please
> notify the sender by
> replying to this message and then delete it from your
> system. Use,
> dissemination, distribution, or reproduction of this
> message and/or any of
> its attachments (if any) by unintended recipients is not
> authorized and
> may be unlawful.
>
>
God Bless
Sean Parker
__________________________________
Discover Yahoo!
Have fun online with music videos, cool games, IM and more. Check it out!
http://discover.yahoo.com/online.html
More information about the omniORB-list
mailing list