http://bugs.winehq.org/show_bug.cgi?id=17273
Sebastian Lackner sebastian@fds-team.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |sebastian@fds-team.de
--- Comment #58 from Sebastian Lackner sebastian@fds-team.de --- For those that are affected by this problem, but haven't seen my announcement in bug 17195 yet: I have created a new patchset for named pipe message mode support, and need more feedback. The latest patchset passes all tests I've done so far - but I cannot test with all affected apps of course... ;)
You can find the patchset (and some more information) here: http://bugs.winehq.org/show_bug.cgi?id=17195#c157
Differences to Adams patchset: * Shouldn't modify the behaviour of byte-mode pipes at all - it still uses SOCK_STREAM for that. * Uses SOCK_SEQPACKET (like in Adams early implementations) instead of wineserver calls. * It also adds support for PIPE_NOWAIT on both byte-mode and message-mode sockets. * _Lots_ of wine tests to ensure, that it behaves like on Windows, even in (almost) all the corner cases, like reading zero byte messages, reading with a buffer of zero length, ...
If something doesn't work yet, please let me know. If someone has a rebased version of Adams patchset, also feel free to send it to me - I would like to know if his patchset also passes all the tests, so we have two working prototypes, and AJ can pick the one he likes more. ;)
@Anastasius: Could you please test especially with apps that didn't work with the old patchset?
Regards, Sebastian