Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner AT cygwin DOT com Mail-Followup-To: cygwin AT cygwin DOT com Delivered-To: mailing list cygwin AT cygwin DOT com Message-ID: From: Don Bowman To: "'cygwin AT cygwin DOT com'" Subject: Re: GCC-3.1 Build Time Date: Sat, 11 May 2002 13:33:23 -0400 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Dr. Wayne Keen" wrote: >In follow up to some recent emails on gcc build time, I used the time >command this morning to monitor the time needed to build gcc-3.1 on a >2 GHz Dual Xeon machine at the office. Here is the breakdown: >configure took 3 minutes, 14 seconds >make bootstrap took 44 minutes, 28 seconds >make install took 1 minute, 58 seconds I've also been looking at this, the slow 'make' and file access times under cygwin. The simplest test is this: take a directory with files and subdirs, and run 'time ls -lR 1>/dev/null'. Do the same thing on e.g. BSD or Linux. Note the extreme difference (on mine its minutes under cygwin, seconds under BSD). Now, to prove its not a windows issue, run a PC NFS server, mount that same windows directory on the BSD box. Hmmm. Still faster even over NFS over NTFS. Conclusion: NTFS is fast enough, windows is fast enough, etc. The underlying performance issue here must be with 'stat' and/or 'readdir' I think, there's no sub-processes being spawned, environment passed, etc. Any suggestions anyone? --don -- 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/