This is the mail archive of the
ecos-discuss@sources.redhat.com
mailing list for the eCos project.
Re: uipc_socket.c (and cyg_tcp_maxidle)
Hi Hans,
No problem :-). I'm forwarding this on to the list so others know it
helps.
BTW - any comments from people? In uipc_socket.c soalloc() the
zalloci() is assigned to "so", so it looks right to do the zfreei() on
"so".
Can my patch be applied to CVS?
Thanks,
Will
On Thu, 2005-06-16 at 18:40 +0200, Hans Hübner wrote:
> Hi Will,
>
> thanks for the patch, really! I only started with eCos 2 weeks ago and I
> stumbled across this very problem soon, as the default number of sockets is so
> low and the bug seems to be triggered by socket shortage. With the patch
> applied, my application seems to run much more stable and also the TCP
> counters now look plausible. You really made my day, thanks!
>
> -Hans
>
> On Thu, 16 Jun 2005, Will Lentz wrote:
>
> > The patch does help me when I run out of sockets via the socket() call,
> > but I'm wondering if the zfreei() in sodealloc() is getting the correct
> > second argument. I'll dig into this deeper...
> >
> > Cheers,
> > Will
> >
> > On Thu, 2005-06-16 at 17:46 +0200, Hans Hübner wrote:
> >> On Thu, 16 Jun 2005, Will Lentz wrote:
> >>
> >>> zfreei() is a function, and doesn't directly modify its first argument.
> >>> But - it passes "so" as its second argument, and it modifies the value
> >>> of so->so_zone through that. Looking at zfreei:
> >>
> >> Oops, it helps being awake while posting! Thanks for the patch, I'll apply it
> >> to my source tree immediately :)
> >>
> >> Cheers,
> >> Hans
> >
> >
> >
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss