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 Date: Sat, 8 Sep 2001 13:03:09 -0400 From: Christopher Faylor To: cygwin-developers AT cygwin DOT com Subject: Re: cygwin slowdown in current cvs version -- need a volunteer Message-ID: <20010908130309.A12339@redhat.com> Reply-To: cygwin-developers AT cygwin DOT com Mail-Followup-To: cygwin-developers AT cygwin DOT com References: <130160175780 DOT 20010908204017 AT logos-m DOT ru> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <130160175780.20010908204017@logos-m.ru> User-Agent: Mutt/1.3.21i On Sat, Sep 08, 2001 at 08:40:17PM +0400, egor duda wrote: >did anybody noticed that cygwin runs slower than before? i've just >tried to perform full rebuild of newlib and received the following >results: > >current cvs: 241.13user 397.30system 11:10.49elapsed 95%CPU >(0avgtext+0avgdata 0maxresident)k 0inputs+0outputs >(0major+0minor)pagefaults 0swaps > >cvs from 2001.08.07: 150.72user 126.65system 5:00.59elapsed 92%CPU >(0avgtext+0avgdata 0maxresident)k 0inputs+0outputs >(0major+0minor)pagefaults 0swaps > >Chris if you want to make release now, please delay it yet one more >time: i think 100% slowdown is not the thing we want to make public. >i'm doing binary search now to determine which check-in caused this. Ok. This is the third time that I've sent 1.3.3 to sources.redhat.com and killed the announcement message. There has to be a better way of doing this. Can someone volunteer to run regressions like this when I announce that 1.3.3 is about to be released? I did run the test suite this time but I would like someone else to be doing this. We have *72* people on this mailing list and just about all of you vowed that you were going to be contributing something to cygwin when I asked you why you wanted to join. I suspect that some of you haven't even sent email to the list. I'd like to see someone running cygwin regression tests on a regular basis. Will anyone volunteer? This essentially entails running "make check" in the winsup/testsuite directory and will involve coming up with some kind of method of timing cygwin. Maybe just a build of cygwin and/or newlib is adequate. I'd prefer not to have to do too much handholding here but maybe Corinna or Egor would be interested in helping someone figure out the test suite if it proves to be troublesome. cgf