Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sources.redhat.com/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com
Date: Fri, 22 Nov 2002 02:51:34 +0100
From: thomas <tom AT huno DOT net>
Reply-To: thomas <tom AT huno DOT net>
X-Priority: 3 (Normal)
Message-ID: <1253138828.20021122025134@huno.net>
To: Christopher Faylor <cygwin AT cygwin DOT com>
Subject: Re: old cygwin distributions
In-Reply-To: <20021121233432.GF26759@redhat.com>
References: <022101c291aa$97665890$78d96f83 AT pomello>
 <7916445640 DOT 20021121213235 AT huno DOT net> <01c901c291a1$476f16a0$78d96f83 AT pomello>
 <1119233000 DOT 20021121221902 AT huno DOT net> <022101c291aa$97665890$78d96f83 AT pomello>
 <5 DOT 1 DOT 0 DOT 14 DOT 2 DOT 20021121143225 DOT 02ac95f0 AT pop3 DOT cris DOT com>
 <7825298296 DOT 20021122000007 AT huno DOT net> <20021121233432 DOT GF26759 AT redhat DOT com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Christopher Faylor <cgf AT redhat DOT com> wrote:

> And, that's part of the problem.  The words "cygwin scheduling bug" make
> very little sense.  Cygwin is not an OS which schedules things.  There
> was also a mention of fifos, which doesn't make sense in a cygwin
> context.

> The bottom line is that I tend to ignore problems where the description
> makes no sense.  It's a simple fact of life that we have to pick and
> choose what we do if we want to sleep 7 hours a night.

i'm sorry, you are right. i somehow expected the developers to magically
know whats going on :) anyway i narrowed the problem down when i ran the
whole thing through strace and it seems to be a pipe performance
problem. it seems that newer cygwin versions transfer the data in much
bigger blocks (or is it chunks?) and this causes a problem here.

> Out of curiousity, have you tried the most recent cygwin snapshot?  I'm
> wondering if "scheduling bug" translates into "emacs cygwin is using
> 100% of my CPU".  That bug is supposed to be fixed in cygwin snapshots.

unfortunately no.

and yes fifo is just the name cdrecord uses for buffers. nothing to do
with mkfifo and the like.

i will post a new detailed bug report in a few minutes, since the old
one didn't really describe the problem (and has a wrong subject).

thomas


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/