>>>> "Jerry" == Jerry James
> Ar an t-ochtú lá déag de mí Iúil, scríobh Stephen J. Turnbull:
>> I would guess it's in the lstream code, which Jerry and
>> Bennett have been playing with over the last year or so. I
>> suspect their patches are responsible for the "Write failed
>> flushing stdout buffer.\nwrite stdout: broken pipe" warnings
>> I've been seeing using CVS recently.
Jerry> Really? As far as I know, this is the only nontrivial
Jerry> change I made to the lstream code:
2004-11-12 Jerry James <james(a)xemacs.org>
* lstream.c (Lstream_really_write): Don't call the flusher if the
writer returned an error.
That's the one I have in mind. I didn't realize it was that old,
though, the flusher warnings are more recent. In both cases I'm just
guessing, though---Aidan asked for "any ideas", if I really thought I
knew what the problem was, I would have said so before.
Jerry> I remember talking with Tony about a bug he was chasing,
Jerry> but I can't seem to find a corresponding ChangeLog entry,
Jerry> nor does "cvs log" show anything I can identify as
Jerry> originating with Tony.
I think that's the one; it had a long thread with patches from both
you and Tony in it.
Tony also had a more recent patch, where he lowered the wait time in a
loop by a couple orders of magnitude. But I'm wrong about that being
in lstreams, that's the process-unix patch:
2006-03-23 Tony Bennett <tbennett(a)nvidia.com>
* process-unix.c (unix_send_process): Reduce timeout for
Faccept_process_output to 10ms. This dramatically speeds up many
functions that read from processes, eg, inferior interpreter
processes as in ilisp or python-mode.
I'm hazy on gnuserv internals, but couldn't it be related through
Jerry> The fact that it only manifests on Linux suggests that it
Jerry> is a Linux bug, but I'll see if I can track it down. It's
Jerry> been bothering me, too.
Do you mean XEmacs is running on Linux, or that the TTY is on Linux?
I can't reproduce with XEmacs running on Linux but displaying to a Mac
OS X Terminal.
School of Systems and Information Engineering http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Ask not how you can "do" free software business;
ask what your business can "do for" free software.