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]

PR43610, try reversing floating-point comparisons even if !FLOAT_LIB_COMPARE_RETURNS_BOOL

From: Paolo Bonzini <bonzini at gnu dot org>
To: gcc-patches at gcc dot gnu dot org
Date: Sun, 09 May 2010 16:13:52 +0200
Subject: PR43610, try reversing floating-point comparisons even if !FLOAT_LIB_COMPARE_RETURNS_BOOL
Connect(): No such file or directory

The patch looks big, but it is mostly moving code around and splitting an if statement into two. The bulk is

       if (code_to_optab[reversed]
-         && (libfunc = optab_libfunc (code_to_optab[reversed], mode))
-         && FLOAT_LIB_COMPARE_RETURNS_BOOL (mode, reversed))
+         && (libfunc = optab_libfunc (code_to_optab[reversed], mode)))

...

if (comparison == UNORDERED - || FLOAT_LIB_COMPARE_RETURNS_BOOL (mode, comparison)) - comparison = reversed_p ? EQ : NE; + || FLOAT_LIB_COMPARE_RETURNS_BOOL (mode, comparison) + || reversed_p) + *ptest = gen_rtx_fmt_ee (reversed_p ? EQ : NE, VOIDmode, target, false_rtx); + else + *ptest = gen_rtx_fmt_ee (comparison, VOIDmode, target, const0_rtx);

- *ptest = gen_rtx_fmt_ee (comparison, VOIDmode, target, const0_rtx);

Tested by the reporter. Ok for 4.5.1 and 4.6?

Thanks,

Paolo

Attachment: pr43610.patch
Description: Text document

Follow-Ups:
- Re: PR43610, try reversing floating-point comparisons even if !FLOAT_LIB_COMPARE_RETURNS_BOOL
  - From: Richard Guenther

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