[Bug target/104104] New: Assembler messages: Error: number of operands mismatch for `vxorps'

[Bug target/104104] New: Assembler messages: Error: number of operands mismatch for `vxorps'

[asolokha at gmx dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104104

            Bug ID: 104104
           Summary: Assembler messages: Error: number of operands mismatch
                    for `vxorps'
           Product: gcc
           Version: 12.0
            Status: UNCONFIRMED
          Keywords: assemble-failure
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: asolokha at gmx dot com
  Target Milestone: ---
            Target: x86_64-unknown-linux-gnu

GNU as 2.37 fails to assemble the output of gcc 12.0.0 20220116 snapshot
(g:90045c5df5b3c8853e7740fb72a11aead1c489bb), compiled w/ -march=alderlake
-masm=intel -O1 -fallow-store-data-races -funroll-all-loops, w/ the following
diagnostics:

% x86_64-unknown-linux-gnu-gcc-12.0.0 -march=alderlake -masm=intel -O1
-fallow-store-data-races -funroll-all-loops -c fo56tlbm.c
/tmp/ccSYPU7W.s: Assembler messages:
/tmp/ccSYPU7W.s:160: Error: number of operands mismatch for `vxorps'
/tmp/ccSYPU7W.s:173: Error: number of operands mismatch for `vxorps'

__attribute__ ((simd)) short int
foo (void)
{
  return 0;
}

Indeed, w/ -masm=intel gcc emits plain vxorps w/o any operands.

[Bug target/104104] [12 Regression] Assembler messages: Error: number of operands mismatch for `vxorps'

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104104

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
            Summary|Assembler messages: Error:  |[12 Regression] Assembler
                   |number of operands mismatch |messages: Error: number of
                   |for `vxorps'                |operands mismatch for
                   |                            |`vxorps'
             Status|UNCONFIRMED                 |ASSIGNED
                 CC|                            |jakub at gcc dot gnu.org
           Assignee|unassigned at gcc dot gnu.org      |jakub at gcc dot gnu.org
   Target Milestone|---                         |12.0
   Last reconfirmed|                            |2022-01-18

[Bug target/104104] [12 Regression] Assembler messages: Error: number of operands mismatch for `vxorps'

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104104

--- Comment #1 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Created attachment 52219
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=52219&action=edit
gcc12-pr104104.patch

Untested fix.

[Bug target/104104] [12 Regression] Assembler messages: Error: number of operands mismatch for `vxorps'

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104104

--- Comment #2 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Jakub Jelinek <jakub@gcc.gnu.org>:

https://gcc.gnu.org/g:ddce00dba244d889d688490517fb106169a72f01

commit r12-6715-gddce00dba244d889d688490517fb106169a72f01
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Wed Jan 19 02:24:06 2022 +0100

    i386: Fix GLC tuning with -masm=intel [PR104104]

    > > On Sat, Jan 15, 2022 at 5:39 PM Hongyu Wang <wwwhhhyyy333@gmail.com>
wrote:
    > > > Thanks for the suggestion, here is the updated patch that survived
    > > > bootstrap/regtest.

    Unfortunately the patch results in assembler failures with -masm=intel.

    > > > > +  if (TARGET_DEST_FALSE_DEPENDENCY
    > > > > +      && get_attr_dest_false_dep (insn) ==
    > > > > +        DEST_FALSE_DEP_TRUE)
    > > > > +    output_asm_insn ("vxorps\t{%x0, %x0, %x0}", operands);

    All the vxorps insns were emitted like the above, which means for
-masm=sysv
    it looks like
            vxorps  %xmm3, %xmm3, %xmm3
    but for -masm=intel like:
            vxorps
    We want obviously
            vxorps  xmm3, xmm3, xmm3
    so the following patch just drops the errorneous {}s.

    2022-01-19  Jakub Jelinek  <jakub@redhat.com>

            PR target/104104
            * config/i386/sse.md
            (<avx512>_<complexopname>_<mode><maskc_name><round_name>,
           
avx512fp16_<complexopname>sh_v8hf<mask_scalarc_name><round_scalarcz_name>,
            avx512dq_mul<mode>3<mask_name>,
<avx2_avx512>_permvar<mode><mask_name>,
            avx2_perm<mode>_1<mask_name>, avx512f_perm<mode>_1<mask_name>,
            avx512dq_rangep<mode><mask_name><round_saeonly_name>,
            avx512dq_ranges<mode><mask_scalar_name><round_saeonly_scalar_name>,
            <avx512>_getmant<mode><mask_name><round_saeonly_name>,
           
avx512f_vgetmant<mode><mask_scalar_name><round_saeonly_scalar_name>):
            Use vxorps\t%x0, %x0, %x0 instead of vxorps\t{%x0, %x0, %x0}.

            * gcc.target/i386/pr104104.c: New test.

[Bug target/104104] [12 Regression] Assembler messages: Error: number of operands mismatch for `vxorps'

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104104

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|ASSIGNED                    |RESOLVED

--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Fixed.