This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH, i386]: Remove dead x87 cbranch helpers
- From: Uros Bizjak <ubizjak at gmail dot com>
- To: Rainer Orth <ro at cebitec dot uni-bielefeld dot de>
- Cc: "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>
- Date: Tue, 24 Oct 2017 18:13:30 +0200
- Subject: Re: [PATCH, i386]: Remove dead x87 cbranch helpers
- Authentication-results: sourceware.org; auth=none
- References: <CAFULd4Zt5tZs84PMzZ7xG2gWqae8BW=_1ifGCVdsK-dMSJxy4Q@mail.gmail.com> <yddo9ow4ksf.fsf@CeBiTec.Uni-Bielefeld.DE>
On Tue, Oct 24, 2017 at 5:39 PM, Rainer Orth
<ro@cebitec.uni-bielefeld.de> wrote:
> Hi Uros,
>
>> Remove dead code, obsoleted by cbranch rewrite years ago.
>>
>> 2017-10-11 Uros Bizjak <ubizjak@gmail.com>
>>
>> * config/i386/i386.md (*cmp<X87MODEF:mode>_<SWI24:mode>_i387):
>> Do not use float_operator operator predicate.
>> (*cmp<X87MODEF:mode>_<SWI24:mode>_cc_i387): Ditto.
>> * config/i386/predicates.md (float_operator): Remove predicate.
>>
>> 2017-10-11 Uros Bizjak <ubizjak@gmail.com>
>>
>> * config/i386/i386.md (*jcc<mode>_0_i387): Remove insn pattern.
>> (*jccxf_i387): Ditto.
>> (*jcc<mode>_i387): Ditto.
>> (*jccu<mode>_i387): Ditto.
>> (*jcc<X87MODEF:mode>_<SWI24:mode>_i387): Ditto.
>> (*jcc_*_i387 splitters): Remove.
>> * config/i386/i386-protos.h (ix86_split_fp_branch): Remove prototype.
>> * config/i386/i386.c (ix86_split_fp_branch): Remove.
>> * config/i386/predicates.md (ix86_swapped_fp_comparison_operator):
>> Remove predicate.
>>
>> testsuite/ChangeLog:
>>
>> 2017-10-11 Uros Bizjak <ubizjak@gmail.com>
>>
>> * gcc.target/i386/387-ficom-1.c: New test.
>>
>> Bootstrapped and regression tested on x86_64-linux-gnu {,-m32}.
>>
>> Committed to mainline SVN.
>
> the new testcase and its 387-ficom-2.c companion FAIL on Solaris/x86
> with /bin/as:
>
> +FAIL: gcc.target/i386/387-ficom-1.c scan-assembler-times ficomps 3 (found 0 times)
> +FAIL: gcc.target/i386/387-ficom-2.c scan-assembler-times ficomps 3 (found 0 times)
>
> With gas, it's
>
> ficomps s
> ficomps s
> ficomps 4(%esp)
>
> while for Solaris as this is used:
>
> ficomp s
> ficomp s
> ficomp 4(%esp)
>
> The following patch accounts for the difference. Tested with the
> appropriate runtest invocation on i386-pc-solaris2.11 and
> x86_64-pc-linux-gnu. Ok for mainline?
>
> Rainer
>
> --
> -----------------------------------------------------------------------------
> Rainer Orth, Center for Biotechnology, Bielefeld University
>
>
> 2017-10-24 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
>
> * gcc.target/i386/387-ficom-1.c: Allow for ficomp without s
> suffix.
> * gcc.target/i386/387-ficom-2.c: Likewise.
OK.
Thanks,
Uros.