Mailing-List: contact cygwin-help@cygwin.com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner@cygwin.com Mail-Followup-To: cygwin@cygwin.com Delivered-To: mailing list cygwin@cygwin.com Mime-Version: 1.0 (Apple Message framework v619) In-Reply-To: <20040830162046.GA919543@Worldnet> References: <41316E85.8010707@melbourne.sgi.com> <20040830155404.GQ8900@trixie.casa.cgf.cx> <20040830162046.GA919543@Worldnet> Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: <84A7B05C-FACF-11D8-A7AD-000A95701024@apple.com> Content-Transfer-Encoding: 7bit From: Kate Ebneter Subject: Re: cygwin1.dll problem with Hyperthreaded machines. Date: Mon, 30 Aug 2004 14:56:59 -0700 To: cygwin@cygwin.com X-IsSubscribed: yes On Aug 30, 2004, at 9:20 AM, Pierre A. Humblet wrote: > Huh? I don't even own a single dog, just a WinME. > It has crossed my mind that the fork bug has something to do with > the hyperthreaded machine problem, and even that it might well be > the last bug in Cygwin. > Processor affinity not helping is consistent with the fork bug. Interesting: The problem we see/saw was that when make attempted to run the compiler, the command string would not get passed along, causing the compiler to not unreasonably complain loudly about the lack of a command line. We verified that make was correctly generating the command line, and that the error occurred somewhere in the fork that launched the compiler. Unfortunately, of course, we could not consistently reproduce the problem. (Although, typically of such things, it always became much, much worse whenever we were pushing for a release.) FWIW, Kate Ebneter iPod Build Engineer Apple Computer, Inc. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/