From owner-freebsd-questions Thu May 16 21:38:02 1996 Return-Path: owner-questions Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id VAA10088 for questions-outgoing; Thu, 16 May 1996 21:38:02 -0700 (PDT) Received: from rocky.sri.MT.net (rocky.sri.MT.net [204.182.243.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id VAA10083 for ; Thu, 16 May 1996 21:37:59 -0700 (PDT) Received: (from nate@localhost) by rocky.sri.MT.net (8.6.12/8.6.12) id WAA25870; Thu, 16 May 1996 22:37:54 -0600 Date: Thu, 16 May 1996 22:37:54 -0600 From: Nate Williams Message-Id: <199605170437.WAA25870@rocky.sri.MT.net> To: Peter Mutsaers Cc: freebsd-questions@FreeBSD.org Subject: Re: make world with gcc version 2.7.2 or pgcc 2.7.2.9? In-Reply-To: <87n338b78i.fsf@plm.simplex.nl> References: <87n338b78i.fsf@plm.simplex.nl> Sender: owner-questions@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > Did anyone try to rebuild everything, including the kernel, with gcc > version 2.7.2 or pgcc, and with -O2 or higher? I believe Kaleb Keithley rebuilt X with one of them and it didn't work. This implies to me it's a compiler bug, as there are quite a few known optimizer bugs in gcc 2.7.2 for the x86, and at least one generic optimizer bug that was posted on gnu.gcc.bug. > Using a newer compiler with more aggressive optimizing can speed up > the entire system quite a bit I hope; it is a shame not to use that > possibility. Which would you rather have, 'known good' code that is slower or 'probably bogus' and faster code. Since there are known bugs, given a system the size of FreeBSD the possibility of the bug being tickled *somewhere* is high. And, given that the speedups for stock gcc aren't stupendous (unlike pgcc which can be quite high, but the code generation is suspect in many cases) it's not worth it. But, according to Warner Losh, gcc 2.7.3 is due out Real Soon Now, so you folks with lots of free time on your hands can build everything you can get your hands on it and see if anything breaks. :) Nate