Mailing-List: contact cygwin-developers-help AT sourceware DOT cygnus DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-developers-owner AT sources DOT redhat DOT com Delivered-To: mailing list cygwin-developers AT sources DOT redhat DOT com From: Chris Faylor Date: Thu, 12 Oct 2000 09:37:14 -0400 To: cygwin developers Subject: Re: strace of 100% CPU usage Message-ID: <20001012093714.C20025@cygnus.com> Reply-To: cygwin-developers AT sources DOT redhat DOT com Mail-Followup-To: cygwin developers References: <20001012132904 DOT 22636 DOT qmail AT web123 DOT yahoomail DOT com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.6i In-Reply-To: <20001012132904.22636.qmail@web123.yahoomail.com>; from earnie_boyd@yahoo.com on Thu, Oct 12, 2000 at 06:29:04AM -0700 On Thu, Oct 12, 2000 at 06:29:04AM -0700, Earnie Boyd wrote: >I've attached two straces one using a -f switch the other without it. I was >expecting to get something huge with this as the strace was of the make of >Cygwin. What I found was that an `sh' goes looping waiting on some >non-existant child and during this wait the CPU is 100% used with the waiting >`sh' job getting most of the time. The waiting `sh' job eventually times out >after ~5 minutes. Under strace we get: > ... sync_with_child: WaitForMultipleObjects timed out > Cannot fork: No more processes > Make: *** [all-apache] Error 2 >where ... is: ######### [main] sh ### > >Without the strace the make continues albeit abusivly and slowly. I actually duplicated this yesterday. I am still scratching my head over the cause. There was actually a bug in strace which causes the looping, though. If you rebuild from CVS the looping goes away and is replaced by a SIGSEGV. Thanks for the strace, though. cgf