delorie.com/archives/browse.cgi | search |
Mailing-List: | contact cygwin-help AT cygwin DOT com; run by ezmlm |
List-Subscribe: | <mailto:cygwin-subscribe AT cygwin DOT com> |
List-Archive: | <http://sourceware.org/ml/cygwin/> |
List-Post: | <mailto:cygwin AT cygwin DOT com> |
List-Help: | <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/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 |
From: | "Chris January" <chris AT atomice DOT net> |
To: | <cygwin AT cygwin DOT com> |
Subject: | RE: Bug fix to /proc/cpuinfo implementation |
Date: | Wed, 23 Jun 2004 18:07:16 +0100 |
Message-ID: | <010c01c45944$8940ccb0$0207a8c0@avocado> |
MIME-Version: | 1.0 |
In-Reply-To: | <1087996722.18400.69.camel@whiteice.nop.iki.fi> |
X-IsSubscribed: | yes |
> Hi > > I hope I email to correct mailing list... > > It seems that '/proc/cpuinfo' doesn't report 3dnow and > 3dnowext support correctly. > > Because I had been using it for recognizing processor > features I looked briefly into fhandler_proc.cc (taken > freshly from cvs) and I think I fixed the problem. The > changed file compiles and *should* correctly detect presence > of 3dnow instruction support (by using AMD's 0x8000..1 > extended function). I haven't have time to test it (= figure out how > build+install process actually works) because I decided to > write my own > cpuid detection routines from scratch instead. Actually nearly all of the processor feature flags are wrong for AMD processors. This is a known bug. Your patch looks to be going about the right way to fix it. I'll try to add the rest of the AMD flags in. Chris -- 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/
webmaster | delorie software privacy |
Copyright © 2019 by DJ Delorie | Updated Jul 2019 |