Re: [Libevent-users] Another possible iocp issue.

[Nick Mathewson <nickm@xxxxxxxxxxxxx>]

On Mon, Nov 22, 2010 at 3:11 PM, Kelly Brock <Kerby@xxxxxxxxxxx> wrote:
> Hi Folks,
>
>        I'm still working up an example testbed for this but figured I'd
> give you a heads up.  Basically I am getting a memory corruption in the
> following (rather extreme) case:
>
> 1.      Create a standard listener.
> 2.      Setup 1000 buffer event items.  Thread safe, deferred and close on
> free.
> 3.      Issue connect requests on all 1000.  (NOTE: have not started calling
> eventbase_loop so many of these fail as part of the test.  Purely a single
> threaded test.)
> 4.      Go into a standard eventbase loop.
>
>        Currently I can repeatedly crash this test case and it seems fairly
> consistently to be failing in evbuffer_reserve_space on a completely empty
> buffer chain.  Specifically the windows debug allocator is complaining that
> something wrote into a previously free'd memory block.
>
>        I'm still double checking my code for this so I may still be the
> culprit.  But it seems unlikely at this time since the code when run against
> the select loop seems to be fine, it also works under Linux and Os X without
> any issues.
>
>        I'll post more info and a testbed as soon as possible.

Hi, Kelly!  Some test code would probably indeed be necessary to fix
any bugs here.  Also useful for the time being would be a
cut-and-paste of the info _exactly_ from your debugger, not just a
summary.  iow, *where* in evbuffer_reserve_space is the crash?  *what*
is the complaint from the windows debug allocator?  Those would help
too, especially if the test code is going to be more than a day or so,
since I'd guess you already have them now.

(Also, what version of libevent?)


yrs,
-- 
Nick
***********************************************************************
To unsubscribe, send an e-mail to majordomo@xxxxxxxxxxxxx with
unsubscribe libevent-users    in the body.