[Bug testsuite/112729] New: gcc.target/i386/apx-interrupt-1.c etc. FAIL

[Bug testsuite/112729] New: gcc.target/i386/apx-interrupt-1.c etc. FAIL

[ro at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112729

            Bug ID: 112729
           Summary: gcc.target/i386/apx-interrupt-1.c etc. FAIL
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: testsuite
          Assignee: unassigned at gcc dot gnu.org
          Reporter: ro at gcc dot gnu.org
                CC: wwwhhhyyy333 at gmail dot com, zewei.mo at intel dot com
  Target Milestone: ---
            Target: i386-pc-solaris2.11, x86_64-unknown-freebsd12.4

Several of the gcc.target/i386/apx-*.c tests FAIL on different targets:

* On Darwin/x86 they are skipped because the assembler doesn't support the cfi
  directives.

* On Solaris/x86 with the native assembler, part of the failures is for the
  same reason: lack of cfi support.  Instead of hardcoding more and more
targets,
  I have a patch that introduces a new cfi effective-target keyword for the
  testsuite to handle that properly instead.

* However, on Solaris/x86 with gas (which does support the cfi directives),
  there are still quite a number of failures.  I'm attaching the assembler
  output for comparison.

FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 130, -136 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 131, -128 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 132, -120 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 133, -112 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 134, -104 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 135, -96 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 136, -88 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 137, -80 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 138, -72 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 139, -64 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 140, -56 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 141, -48 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 142, -40 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 143, -32 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 144, -24 1
FAIL: gcc.target/i386/apx-interrupt-1.c scan-assembler-times
\\\\t\\\\.cfi_offset 145, -16 1
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times
.cfi_def_cfa_offset 16 2
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times
.cfi_def_cfa_offset 32 2
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times
.cfi_def_cfa_offset 48 2
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times
.cfi_def_cfa_offset 56 2
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times
.cfi_def_cfa_offset 8(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times .cfi_offset 12,
-40(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times .cfi_offset 13,
-32(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times .cfi_offset 14,
-24(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times .cfi_offset 15,
-16(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times .cfi_offset 6,
-48(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times .cfi_restore
6(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times pop2[\\\\t
]*[^\\n\\r]*%r12[^\\n\\r]*%rbp[^\\n\\r]*(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times pop2[\\\\t
]*[^\\n\\r]*%r14[^\\n\\r]*%r13[^\\n\\r]*(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times
popq[^\\n\\r]*%r15(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times push2[\\\\t
]*[^\\n\\r]*%r13[^\\n\\r]*%r14[^\\n\\r]*(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times push2[\\\\t
]*[^\\n\\r]*%rbp[^\\n\\r]*%r12[^\\n\\r]*(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2-1.c scan-assembler-times
pushq[^\\n\\r]*%r15(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
.cfi_def_cfa_offset 16 2
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
.cfi_def_cfa_offset 32 2
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
.cfi_def_cfa_offset 48 2
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
.cfi_def_cfa_offset 56 2
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
.cfi_def_cfa_offset 8(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
.cfi_offset 12, -40(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
.cfi_offset 13, -32(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
.cfi_offset 14, -24(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
.cfi_offset 15, -16(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
.cfi_offset 6, -48(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
.cfi_restore 6(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
pop2[\\\\t ]*[^\\n\\r]*%r12[^\\n\\r]*%rbp[^\\n\\r]*(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
pop2[\\\\t ]*[^\\n\\r]*%r14[^\\n\\r]*%r13[^\\n\\r]*(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
popq[^\\n\\r]*%r15(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
push2[\\\\t ]*[^\\n\\r]*%r13[^\\n\\r]*%r14[^\\n\\r]*(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
push2[\\\\t ]*[^\\n\\r]*%rbp[^\\n\\r]*%r12[^\\n\\r]*(?:\\n|[ \\\\t]+#) 1
FAIL: gcc.target/i386/apx-push2pop2_force_drap-1.c scan-assembler-times
pushq[^\\n\\r]*%r15(?:\\n|[ \\\\t]+#) 1

* I strongly suspect that the FreeBSD 12.4/x86_64 failures are the same,
although
  that's difficult to tell from the gcc-testresults mails.

It seems something is amiss here: maybe the tests are Linux-specific in some
way.  In that case, the tests should be adjusted accordingly.

[Bug testsuite/112729] gcc.target/i386/apx-interrupt-1.c etc. FAIL

[ro at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112729

Rainer Orth <ro at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |14.0

[Bug testsuite/112729] gcc.target/i386/apx-interrupt-1.c etc. FAIL

[ro at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112729

--- Comment #1 from Rainer Orth <ro at gcc dot gnu.org> ---
Created attachment 56695
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=56695&action=edit
64-bit i386-pc-solaris2.11 assembler output

[Bug testsuite/112729] gcc.target/i386/apx-interrupt-1.c etc. FAIL

[wwwhhhyyy333 at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112729

--- Comment #2 from Hongyu Wang <wwwhhhyyy333 at gmail dot com> ---
The cfi scan fails was caused by -fno-omit-frame-pointer which force push the
frame pointer first and the cfi info become different. By default we have
-fomit-frame-pointer on linux, but not other targets. I'd just add
-fomit-frame-pointer to these tests.

[Bug testsuite/112729] gcc.target/i386/apx-interrupt-1.c etc. FAIL

[wwwhhhyyy333 at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112729

--- Comment #3 from Hongyu Wang <wwwhhhyyy333 at gmail dot com> ---
Created attachment 56703
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=56703&action=edit
A patch

Hi Rainer, can you help verify if the change make these test pass on
solaris/FreeBSD?

[Bug testsuite/112729] gcc.target/i386/apx-interrupt-1.c etc. FAIL

[ro at CeBiTec dot Uni-Bielefeld.DE]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112729

--- Comment #4 from ro at CeBiTec dot Uni-Bielefeld.DE <ro at CeBiTec dot Uni-Bielefeld.DE> ---
> --- Comment #2 from Hongyu Wang <wwwhhhyyy333 at gmail dot com> ---
> The cfi scan fails was caused by -fno-omit-frame-pointer which force push the
> frame pointer first and the cfi info become different. By default we have
> -fomit-frame-pointer on linux, but not other targets. I'd just add
> -fomit-frame-pointer to these tests.

That's one of the first differences between Linux and other x86 targets
that come to mind.  I'd have sworn that's what I tried before reporting
the bug.  Apparently not ;-(

[Bug testsuite/112729] gcc.target/i386/apx-interrupt-1.c etc. FAIL

[ro at CeBiTec dot Uni-Bielefeld.DE]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112729

--- Comment #5 from ro at CeBiTec dot Uni-Bielefeld.DE <ro at CeBiTec dot Uni-Bielefeld.DE> ---
> --- Comment #3 from Hongyu Wang <wwwhhhyyy333 at gmail dot com> ---
[...]
> Hi Rainer, can you help verify if the change make these test pass on
> solaris/FreeBSD?

They do on Solaris/x86.  Thanks.

FreeBSD was more involved (I'd only noticed the failure there on
gcc-testresults).  Unfortunately, there's no FreeBSD/x86_64 system in
the cfarm.  However, I had an old FreeBSD 13 VM around which could be
coaxed into a gcc build with some effort.  With your patch, the failures
are gone there, too.

One nit:

diff --git a/gcc/testsuite/gcc.target/i386/apx-push2pop2-1.c
b/gcc/testsuite/gcc.target/i386/apx-push2pop2-1.c
index d78c96d36a3..1d73aab1128 100644
--- a/gcc/testsuite/gcc.target/i386/apx-push2pop2-1.c
+++ b/gcc/testsuite/gcc.target/i386/apx-push2pop2-1.c
@@ -1,5 +1,5 @@
 /* { dg-do compile { target { ! ia32 } } } */
-/* { dg-options "-O2 -mapx-features=push2pop2" } */
+/* { dg-options "-O2 -fomit-frame-pointer -mapx-features=push2pop2" } */
 /* { dg-skip-if "does not emit .cfi_xxx" "*-*-darwin*" } */

 extern int bar (int);
diff --git a/gcc/testsuite/gcc.target/i386/apx-push2pop2_force_drap-1.c
b/gcc/testsuite/gcc.target/i386/apx-push2pop2_force_drap-1.c
index 3cac7b10769..4e2259f0c99 100644
--- a/gcc/testsuite/gcc.target/i386/apx-push2pop2_force_drap-1.c
+++ b/gcc/testsuite/gcc.target/i386/apx-push2pop2_force_drap-1.c
@@ -1,5 +1,5 @@
 /* { dg-do compile { target { ! ia32 } } } */
-/* { dg-options "-O2 -mapx-features=push2pop2 -mforce-drap" } */
+/* { dg-options "-O2 -mapx-features=push2pop2 -fomit-frame-pointer
-mforce-drap" } */
 /* { dg-skip-if "does not emit .cfi_xxx" "*-*-darwin*" } */

 #include "apx-push2pop2-1.c"

Is there a reason to have -fomit-frame-pointer once before and once
after -mapx-features=push2pop2?

[Bug testsuite/112729] gcc.target/i386/apx-interrupt-1.c etc. FAIL

[ro at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112729

Rainer Orth <ro at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                URL|                            |https://gcc.gnu.org/piperma
                   |                            |il/gcc-patches/2023-Novembe
                   |                            |r/638447.html

--- Comment #6 from Rainer Orth <ro at gcc dot gnu.org> ---
Patch for the cfi directive issue posted.

[Bug testsuite/112729] gcc.target/i386/apx-interrupt-1.c etc. FAIL

[wwwhhhyyy333 at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112729

--- Comment #7 from Hongyu Wang <wwwhhhyyy333 at gmail dot com> ---
(In reply to ro@CeBiTec.Uni-Bielefeld.DE from comment #5)
> 
> Is there a reason to have -fomit-frame-pointer once before and once
> after -mapx-features=push2pop2?

Ah, thanks for pointing that out. Will adjust the order to keep them after
-mapx-features.

[Bug testsuite/112729] gcc.target/i386/apx-interrupt-1.c etc. FAIL

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112729

--- Comment #8 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Hongyu Wang <hongyuw@gcc.gnu.org>:

https://gcc.gnu.org/g:99fa0bfd63d97825c4221dcd3123940f1d0e6291

commit r14-5943-g99fa0bfd63d97825c4221dcd3123940f1d0e6291
Author: Hongyu Wang <hongyu.wang@intel.com>
Date:   Tue Nov 28 11:24:01 2023 +0800

    [i386] Fix push2pop2 test fail on non-linux target [PR112729]

    On linux x86-64, -fomit-frame-pointer was by default enabled so the
    push2pop2 tests cfi scans are based on it. On other target with
    -fno-omit-frame-pointer the cfi scan will be wrong as the frame pointer
    is pushed at first. Add -fomit-frame-pointer to these tests that related
    to cfi scan.

    gcc/testsuite/ChangeLog:

            PR target/112729
            * gcc.target/i386/apx-interrupt-1.c: Add -fomit-frame-pointer.
            * gcc.target/i386/apx-push2pop2-1.c: Likewise.
            * gcc.target/i386/apx-push2pop2_force_drap-1.c: Likewise.

[Bug testsuite/112729] gcc.target/i386/apx-interrupt-1.c etc. FAIL

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112729

--- Comment #9 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Rainer Orth <ro@gcc.gnu.org>:

https://gcc.gnu.org/g:8ee480441eaec87ca6e0b0963fe9aade7edd1e8a

commit r14-5968-g8ee480441eaec87ca6e0b0963fe9aade7edd1e8a
Author: Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
Date:   Wed Nov 29 14:52:04 2023 +0100

    testsuite, i386: Only check for cfi directives if supported [PR112729]

    gcc.target/i386/apx-interrupt-1.c and two more tests FAIL on Solaris/x86
    with the native assembler.  Like Darwin as, it doesn't support cfi
    directives.  Instead of adding more and more targets in every affected
    test, this patch introduces a cfi effective-target keyword to check for
    the prerequisite.

    Tested on i386-pc-solaris2.11 (as and gas), x86_64-pc-linux-gnu, and
    x86_64-apple-darwin23.1.0.

    2023-11-24  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>

            gcc/testsuite:
            PR testsuite/112729
            * lib/target-supports.exp (check_effective_target_cfi): New proc.
            * gcc.target/i386/apx-interrupt-1.c: Require cfi instead of
            skipping on *-*-darwin*.
            * gcc.target/i386/apx-push2pop2_force_drap-1.c: Likewise.
            * gcc.target/i386/apx-push2pop2-1.c: Likewise.

            gcc:
            PR testsuite/112729
            * doc/sourcebuild.texi (Effective-Target Keywords, Environment
            attributes): Document cfi.

[Bug testsuite/112729] gcc.target/i386/apx-interrupt-1.c etc. FAIL

[ro at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112729

Rainer Orth <ro at gcc dot gnu.org> changed:

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

--- Comment #10 from Rainer Orth <ro at gcc dot gnu.org> ---
AFAICS, all aspects of this PR have been dealt with now.