On Wed, Aug 14, 2013 at 8:26 AM, Daniel Stenberg <daniel_at_haxx.se> wrote:
> On Sat, 10 Aug 2013, Daniel Stenberg wrote:
> I've now been able to reproduce the problem using test case 107 with my
> brand new event-based curl_easy_perform() replacement.
>
> I'll soon A) post my event-based thing as a patch for review and comments
> and B) see if I can use this new setup to fix test 107 when run event-based!

Excellent, glad to hear that I'm not crazy. :) Thanks for all your
help with this issue, I would have had no idea where to start. In the
meantime, will forcing it to register for read and write on all
sockets cause any problems? If not, I'll just leave that work-around
in place for the moment. If so, I'll probably set up a slightly more
specific one that tries to target only the proper socket.

Thanks again for your help.
Bill
-------------------------------------------------------------------
List admin: http://cool.haxx.se/list/listinfo/curl-library
Etiquette: http://curl.haxx.se/mail/etiquette.html
Received on 2013-08-14