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: PowerPC validate_condition_mode (g77 regression fix)


David Edelsohn wrote:

>         I think this patch fixes the g77 regressions on PowerPC.
> 
> David

>     if (mode == CCFPmode
>         && ! flag_unsafe_math_optimizations
> +       && ! flag_finite_math_only
>         && (code == LE || code == GE

Ahhh, you're too fast for me (I forgot the --disable-multilib that's
absolutely necessary when doing anything with the rs6000 backend :-)

I tested it solely with ! flag_finite_math_only (and updated the comment to
say that we really wanted to test HONOR_NANS, but that it isn't possible
there - because the needed mode of the variables compared isn't passed
down).

However, this will do.

-- 
Toon Moene - mailto:toon@moene.indiv.nluug.nl - phoneto: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
Maintainer, GNU Fortran 77: http://gcc.gnu.org/onlinedocs/g77_news.html
Join GNU Fortran 95: http://g95.sourceforge.net/ (under construction)


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