Re: [V] Re: [AC21.5R21.4] Re: [R21.4] Dealing with I/O in the child process on Unix

SY> VETO SY> |--==> "SY" == Steve Youngs <youngs(a)xemacs.org&gt; writes: SY> APPROVE COMMIT 21.5 RECOMMEND 21.4 SY> Vetoing my own approval... my head is spinning :-P SY> |--==> "MS" == Michael Sperber <sperber(a)informatik.uni-tuebingen.de&gt; writes: MS> 2002-07-20 Mike Sperber <mike(a)xemacs.org&gt; MS> * process-unix.c (child_setup): Don't try to close file MS> descriptors for chid process once again---it's already being done MS> in close_process_descs. MS> (unix_create_process): Call begin_dont_check_for_quit to inhibit MS> unwanted interaction (and thus breaking of X event synchronicity) MS> in the child. SY> Thanks for the patch, I'll commit this shortly. SY> Hi Mike! SY> Unfortunately, this patch causes problems for me with external SY> processes. SY> If I load up Dired, mark a couple of files and then do 'E' to grep for SY> something, grep just sits there (presumably waiting for input). SY> Manually kill the grep process and there doesn't seem to be any error SY> raised. XEmacs is inoperable until grep is killed and then returns to SY> normal operation. SY> Mike, I'm not exactly sure what is causing it, but I'd suspect the SY> change to 'unix_create_process'. I'll revert the patch *unless* I SY> hear back from you in the next 24 hours. [1] And if there is anything SY> I can do to help debug this, just yell. Erh, well, sure, if you can track down your problem to the patch, then back it out, by all means. If the patch causes it, however, that means there's something seriously screwed up somewhere down below. Also note that there are *other* subprocess-related bugs in the system beside the one this patch is fixing---you have one pending yourself. -- Cheers =8-} Mike Friede, Völkerverständigung und überhaupt blabla