From xemacs-m  Fri Jan 17 15:58:01 1997
Received: from cdc.noaa.gov (manager.Colorado.EDU [128.138.218.210])
          by xemacs.org (8.8.4/8.8.4) with SMTP
	  id PAA05083 for <xemacs-beta@xemacs.org>; Fri, 17 Jan 1997 15:57:59 -0600 (CST)
Received: from suomi by cdc.noaa.gov (SMI-8.6/SMI-SVR4)
	id OAA06273; Fri, 17 Jan 1997 14:57:57 -0700
Received: by suomi (SMI-8.6) id OAA26121; Fri, 17 Jan 1997 14:57:57 -0700
Sender: mdb@cdc.noaa.gov
To: Soren Dayton <csdayton@cs.uchicago.edu>
Cc: xemacs-beta@xemacs.org
Subject: Re: Xemacs taking up a lot of cycles while `idle'
References: <12215.853396330@alexandria11> <m2d8v4hoou.fsf@altair.xemacs.org> <xcdpvz4xd2x.fsf@gargoyle164.cs.uchicago.edu>
Organization: CIRES, University of Colorado
X-Attribution: mb
From: Mark Borges <mdb@cdc.noaa.gov>
Date: 17 Jan 1997 14:57:56 -0700
In-Reply-To: Soren Dayton's message of 17 Jan 1997 14:26:46 -0600
Message-ID: <vk20bknevv.fsf@cdc.noaa.gov>
Lines: 51
X-Mailer: Red Gnus v0.80/XEmacs 19.15

>> On 17 Jan 1997 14:26:46 -0600,
>> Soren Dayton(SD) wrote:
SD> Steven L Baur <steve@miranova.com> writes:
>> Soren Dayton writes:
>> 
>> > top says this about my xemacs process that has been idle for ten hours:
>> > 24652 csdayton  25    0   23M   16M cpu    26.8H 29.17% 25.78% xemacs
>> 
>> > Now, I just iconify it and let it sit and I get this (repeating
>> > indefinitely) from a truss.
>> 
>> > sigprocmask(SIG_UNBLOCK, 0xEFFFF0BC, 0x00000000) = 0
>> > unlink("/homes/csdayton/.saves-4859-gargoyle164.cs.uchicago.edu") Err#2 ENOENT
>> > ioctl(16, FIONREAD, 0xEFFFEFB4)                 = 0
>> > ioctl(16, FIONREAD, 0xEFFFEFB4)                 = 0

>> Isn't this due to defining BROKEN_SIGIO?

I thought it was BROKEN_SIGCHLD, according to this comment in
src/s/sol-2.5.h:

------------------------------------------------------------------------------
/* Apparently this is not necessary here, and it causes 10% CPU
   chewage. */
#undef BROKEN_SIGCHLD
------------------------------------------------------------------------------

It gets defined in sol-2.4.h, when using X11R6:

------------------------------------------------------------------------------
/* XEmacs addition: Raymond Toy says XEmacs completely misses SIGCHLD
   when compiled with GCC 2.7.0 (but not, apparently, with SunPro C?),
   X11R6, and Solaris 2.4.

   Someone else submitted a simple test program that duplicates this
   behavior, and says it has something to do with the fact that X11R6
   links with the threads library. */

#ifdef THIS_IS_X11R6
#define BROKEN_SIGCHLD
#endif
------------------------------------------------------------------------------

SD> I am not sure.  How was this treated differently in 19.14, where the
SD> problem does not occur?  (some of the problem has gone away by setting
SD> auto-save-timeout to 0, but not all)

Anyway, since you're using Solaris-2.5, it shouldn't be defined anyway.

-- 
  -mb-

