Re: 23.0.60; Occasional hangs in flyspell-mode and ispell-word

[Stefan Monnier]

>> I personally use an Emacs patched as follows:
>> 
>> Index: src/process.c
>> ===================================================================
>> RCS file: /sources/emacs/emacs/src/process.c,v
>> retrieving revision 1.538
>> diff -u -r1.538 process.c
>> --- src/process.c    27 Mar 2008 20:52:24 -0000      1.538
>> +++ src/process.c    27 Mar 2008 20:52:32 -0000
>> @@ -4272,6 +4272,9 @@
>> FD_ZERO (&Connecting);
>> #endif
>> +  if (time_limit == 0 && wait_proc && !NILP (Vinhibit_quit))
>> +    error ("Blocking call to accept-process-output with quit inhibited!!");
>> +
>> /* If wait_proc is a process to watch, set wait_channel accordingly.  */
>> if (wait_proc != NULL)
>> wait_channel = wait_proc->infd;
>> 
>> so the accept-process-output refuses to do its job if it's run with
>> quit-inhibited, which is the typical cause of problems such as the one
>> you're describing.  But your backtrace indicates that inhibit-quit is
>> (most likely) not set, so your problem seems to be different.

> It is a bit off topic but why do you not apply this to the CVS sources?

It breaks some Elisp packages.  Those packages arguably have a bug, but
the bug may very rarely show up (if ever).


        Stefan