[Bug c/105292] New: [sparc64] ICE in expand_expr_real_2 on sparc64 when compiling with -mcpu=niagara4

[Bug c/105292] New: [sparc64] ICE in expand_expr_real_2 on sparc64 when compiling with -mcpu=niagara4

[koachan+gccbugs at protonmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105292

            Bug ID: 105292
           Summary: [sparc64] ICE in expand_expr_real_2 on sparc64 when
                    compiling with -mcpu=niagara4
           Product: gcc
           Version: 11.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: koachan+gccbugs at protonmail dot com
  Target Milestone: ---

Created attachment 52820
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=52820&action=edit
Preprocessed source of ctf-open.c

Hello~

I'm getting this error on sparc64-linux-gnu when compiling gdb (commit
f0072f79e12 from git), in ctf-open.c:

gcc -O3 -mcpu=niagara4 -c libctf_la-ctf-open.c 
during RTL pass: expand
../../libctf/ctf-open.c: In function ‘ctf_bufopen_internal.part.0’:
../../libctf/ctf-open.c:1117:69: internal compiler error: in
expand_expr_real_2, at expr.c:9867
Please submit a full bug report,
with preprocessed source if appropriate.
See <https://gcc.gnu.org/bugs/> for instructions.

As far as I know, compiling with -mcpu=niagara3 and -mcpu=niagara2 also fails
with the same error.
I'm using "gcc version 11.2.0 (GCC)". Attached is the preprocessed source of
the offending file.

[Bug target/105292] [sparc64] ICE in expand_expr_real_2 on sparc64 when compiling with -mcpu=niagara4

[mikpelinux at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105292

Mikael Pettersson <mikpelinux at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mikpelinux at gmail dot com

--- Comment #1 from Mikael Pettersson <mikpelinux at gmail dot com> ---
I can reproduce with a gcc-11.2.0 cross to sparc64-linux-gnu. ICEs with -mcpu
selecting ultrasparc3 or niagara[2347]. Doesn't ice if -mcpu select plain
ultrasparc or niagara. Also doesn't ICE if -O3 is reduced to -O2.

Still ICEs with gcc-12-20220417.

[Bug target/105292] [sparc64] ICE in expand_expr_real_2 on sparc64 when compiling with -mcpu=niagara4

[mikpelinux at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105292

--- Comment #2 from Mikael Pettersson <mikpelinux at gmail dot com> ---
Works with gcc-9.4, ICEs with gcc-10.3 and above. Git bisect identified

# first bad commit: [9b75f56d4b7951c60a656396dddd4a65787b95bc] Apply maximum
nunits for BB SLP

as the cause of the ICEs, though it possibly only exposed a latent target
issue.

[Bug target/105292] [10/11/12/13 regression] ICE in expand_expr_real_2 when compiling with -mcpu=niagara4 -mvis2

[ebotcazou at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105292

Eric Botcazou <ebotcazou at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2022-05-06
                 CC|                            |ebotcazou at gcc dot gnu.org
            Summary|[sparc64] ICE in            |[10/11/12/13 regression]
                   |expand_expr_real_2 on       |ICE in expand_expr_real_2
                   |sparc64 when compiling with |when compiling with
                   |-mcpu=niagara4              |-mcpu=niagara4 -mvis2
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW
   Target Milestone|---                         |10.4

--- Comment #3 from Eric Botcazou <ebotcazou at gcc dot gnu.org> ---
It fails with -O3 -mcpu=niagara4 -mvis2 up to GCC 11 and s/-O3/-O2/ with GCC
12.

[Bug target/105292] [10/11/12/13 regression] ICE in expand_expr_real_2 when compiling with -mcpu=niagara4 -mvis2

[ebotcazou at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105292

--- Comment #4 from Eric Botcazou <ebotcazou at gcc dot gnu.org> ---
Created attachment 52931
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=52931&action=edit
Reduced testcase

[Bug target/105292] [10/11/12/13 regression] ICE in expand_expr_real_2 when compiling with -mcpu=niagara4 -mvis2

[ebotcazou at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105292

Eric Botcazou <ebotcazou at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Assignee|unassigned at gcc dot gnu.org      |ebotcazou at gcc dot gnu.org
             Status|NEW                         |ASSIGNED

--- Comment #5 from Eric Botcazou <ebotcazou at gcc dot gnu.org> ---
Fixing.

[Bug target/105292] [10/11/12/13 regression] ICE in expand_expr_real_2 when compiling with -mvis2

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105292

--- Comment #6 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Eric Botcazou <ebotcazou@gcc.gnu.org>:

https://gcc.gnu.org/g:60ab3d0f2d9f95045543fea361e976351e76b5a1

commit r13-223-g60ab3d0f2d9f95045543fea361e976351e76b5a1
Author: Eric Botcazou <ebotcazou@adacore.com>
Date:   Tue May 10 09:33:16 2022 +0200

    Fix internal error with vectorization on SPARC

    This is a regression present since the 10.x series, but the underlying
issue
    has been there since the TARGET_VEC_PERM_CONST hook was implemented, in the
    form of an ICE when expanding a constant VEC_PERM_EXPR in V4QI, while the
    back-end only supports V8QI constant VEC_PERM_EXPRs.

    gcc/
            PR target/105292
            * config/sparc/sparc.cc (sparc_vectorize_vec_perm_const): Return
            true only for 8-byte vector modes.

    gcc/testsuite/
            * gcc.target/sparc/20220510-1.c: New test.

[Bug target/105292] [10/11/12/13 regression] ICE in expand_expr_real_2 when compiling with -mvis2

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105292

--- Comment #7 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-12 branch has been updated by Eric Botcazou
<ebotcazou@gcc.gnu.org>:

https://gcc.gnu.org/g:b931d0cfd33fec6ffc259b1be5c832144b4e7bea

commit r12-8361-gb931d0cfd33fec6ffc259b1be5c832144b4e7bea
Author: Eric Botcazou <ebotcazou@adacore.com>
Date:   Tue May 10 09:33:16 2022 +0200

    Fix internal error with vectorization on SPARC

    This is a regression present since the 10.x series, but the underlying
issue
    has been there since the TARGET_VEC_PERM_CONST hook was implemented, in the
    form of an ICE when expanding a constant VEC_PERM_EXPR in V4QI, while the
    back-end only supports V8QI constant VEC_PERM_EXPRs.

    gcc/
            PR target/105292
            * config/sparc/sparc.cc (sparc_vectorize_vec_perm_const): Return
            true only for 8-byte vector modes.

    gcc/testsuite/
            * gcc.target/sparc/20220510-1.c: New test.

[Bug target/105292] [10/11/12/13 regression] ICE in expand_expr_real_2 when compiling with -mvis2

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105292

--- Comment #8 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-11 branch has been updated by Eric Botcazou
<ebotcazou@gcc.gnu.org>:

https://gcc.gnu.org/g:61dec0cb3398e35b67196941e1d7a99ad275482c

commit r11-9973-g61dec0cb3398e35b67196941e1d7a99ad275482c
Author: Eric Botcazou <ebotcazou@adacore.com>
Date:   Tue May 10 09:33:16 2022 +0200

    Fix internal error with vectorization on SPARC

    This is a regression present since the 10.x series, but the underlying
issue
    has been there since the TARGET_VEC_PERM_CONST hook was implemented, in the
    form of an ICE when expanding a constant VEC_PERM_EXPR in V4QI, while the
    back-end only supports V8QI constant VEC_PERM_EXPRs.

    gcc/
            PR target/105292
            * config/sparc/sparc.c (sparc_vectorize_vec_perm_const): Return
            true only for 8-byte vector modes.

    gcc/testsuite/
            * gcc.target/sparc/20220510-1.c: New test.

[Bug target/105292] [10/11/12/13 regression] ICE in expand_expr_real_2 when compiling with -mvis2

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105292

--- Comment #9 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-10 branch has been updated by Eric Botcazou
<ebotcazou@gcc.gnu.org>:

https://gcc.gnu.org/g:9cd6d49d43ff7a8e47ffe5358b2e8fe072ed0e69

commit r10-10607-g9cd6d49d43ff7a8e47ffe5358b2e8fe072ed0e69
Author: Eric Botcazou <ebotcazou@adacore.com>
Date:   Tue May 10 09:33:16 2022 +0200

    Fix internal error with vectorization on SPARC

    This is a regression present since the 10.x series, but the underlying
issue
    has been there since the TARGET_VEC_PERM_CONST hook was implemented, in the
    form of an ICE when expanding a constant VEC_PERM_EXPR in V4QI, while the
    back-end only supports V8QI constant VEC_PERM_EXPRs.

    gcc/
            PR target/105292
            * config/sparc/sparc.c (sparc_vectorize_vec_perm_const): Return
            true only for 8-byte vector modes.

    gcc/testsuite/
            * gcc.target/sparc/20220510-1.c: New test.

[Bug target/105292] [10/11/12/13 regression] ICE in expand_expr_real_2 when compiling with -mvis2

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105292

--- Comment #10 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-9 branch has been updated by Eric Botcazou
<ebotcazou@gcc.gnu.org>:

https://gcc.gnu.org/g:666d0884e1ebcbc53b62ee24b3847fff11ac134e

commit r9-10068-g666d0884e1ebcbc53b62ee24b3847fff11ac134e
Author: Eric Botcazou <ebotcazou@adacore.com>
Date:   Tue May 10 09:33:16 2022 +0200

    Fix internal error with vectorization on SPARC

    This is a regression present since the 10.x series, but the underlying
issue
    has been there since the TARGET_VEC_PERM_CONST hook was implemented, in the
    form of an ICE when expanding a constant VEC_PERM_EXPR in V4QI, while the
    back-end only supports V8QI constant VEC_PERM_EXPRs.

    gcc/
            PR target/105292
            * config/sparc/sparc.c (sparc_vectorize_vec_perm_const): Return
            true only for 8-byte vector modes.

    gcc/testsuite/
            * gcc.target/sparc/20220510-1.c: New test.

[Bug target/105292] [10/11/12/13 regression] ICE in expand_expr_real_2 when compiling with -mvis2

[ebotcazou at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105292

Eric Botcazou <ebotcazou at gcc dot gnu.org> changed:

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

--- Comment #11 from Eric Botcazou <ebotcazou at gcc dot gnu.org> ---
Thanks for reporting the problem.

[Bug target/105292] [10/11/12/13 regression] ICE in expand_expr_real_2 when compiling with -mvis2

[ebotcazou at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105292

Eric Botcazou <ebotcazou at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |sumbera at volny dot cz

--- Comment #12 from Eric Botcazou <ebotcazou at gcc dot gnu.org> ---
*** Bug 105792 has been marked as a duplicate of this bug. ***