This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: Fix for PR fortran/3743

From: Jakub Jelinek <jakub at redhat dot com>
To: "Billinghurst, David (CRTS)" <David dot Billinghurst at riotinto dot com>
Cc: gcc-patches at gcc dot gnu dot org
Date: Wed, 30 Jan 2002 05:00:52 -0500
Subject: Re: Fix for PR fortran/3743
References: <8D00C32549556B4E977F81DBC24E985D4101ED@crtsmail1.technol_exch.corp.riotinto.org>
Reply-to: Jakub Jelinek <jakub at redhat dot com>

On Wed, Jan 30, 2002 at 09:13:50AM +0800, Billinghurst, David (CRTS) wrote:
> This completes Toon's fix for PR fortran/3743.  It rearranges the list of intrinsics in intrin.def so that they are sorted according to strcasecmp.
> 
> Tested on mips-sgi-irix6.5 (which failed previously) and i686-pc-cygwin (which passed previously)
> 
> 2002-01-30  David Billinghurst <David.Billinghurst@riotinto.com>
> 
> 	* intrin.def:  Swap BITTEST and BIT_SIZE in table of intrinsics
> 	so the are sorted when compared with strcasecmp.

Where do you get any guarantees that strcasecmp ("BITEST", "BIT_SIZE") > 0.
For POSIX locale this is required behaviour, but for other locales this is
undefined.
IMHO you'd much better just do the string comparison by hand in intrin.c,
doing first character comparison separately before cycling (that one is
highly likely not matching) and taking advantage that uc is already upper
case string.

	Jakub

References:
- Fix for PR fortran/3743
  - From: Billinghurst, David (CRTS)

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]