[Bug target/94950] New: [8/9/10 regression] ICE in gcc.dg/pr94780.c on riscv64

[Bug target/94950] New: [8/9/10 regression] ICE in gcc.dg/pr94780.c on riscv64

[schwab@linux-m68k.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94950

            Bug ID: 94950
           Summary: [8/9/10 regression] ICE in gcc.dg/pr94780.c on riscv64
           Product: gcc
           Version: unknown
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: schwab@linux-m68k.org
                CC: jakub at gcc dot gnu.org, wilson at gcc dot gnu.org
            Blocks: 94780
  Target Milestone: ---
            Target: riscv64-*-*

The same as bug 94780 for the riscv backend:

/daten/riscv64/gcc/gcc-20200502/gcc/testsuite/gcc.dg/pr94780.c:8:1: internal
compiler error: Segmentation fault
0x74a3af crash_signal
        ../../gcc/toplev.c:328
0x7ea370 tree_check(tree_node*, char const*, int, char const*, tree_code)
        ../../gcc/tree.h:3293
0x7ea370 convert_nonlocal_reference_op
        ../../gcc/tree-nested.c:1064
0x9e3abb walk_tree_1(tree_node**, tree_node* (*)(tree_node**, int*, void*),
void*, hash_set<tree_node*, false, default_hash_traits<tree_node*> >*,
tree_node* (*)(tree_node**, int*, tree_node* (*)(tree_node**, int*, void*),
void*, hash_set<tree_node*, false, default_hash_traits<tree_node*> >*))
        ../../gcc/tree.c:12003
0x4be799 walk_gimple_op(gimple*, tree_node* (*)(tree_node**, int*, void*),
walk_stmt_info*)
        ../../gcc/gimple-walk.c:268
0x4be9e3 walk_gimple_stmt(gimple_stmt_iterator*, tree_node*
(*)(gimple_stmt_iterator*, bool*, walk_stmt_info*), tree_node* (*)(tree_node**,
int*, void*), walk_stmt_info*)
        ../../gcc/gimple-walk.c:596
0x4beb93 walk_gimple_seq_mod(gimple**, tree_node* (*)(gimple_stmt_iterator*,
bool*, walk_stmt_info*), tree_node* (*)(tree_node**, int*, void*),
walk_stmt_info*)
        ../../gcc/gimple-walk.c:51
0x4beadb walk_gimple_stmt(gimple_stmt_iterator*, tree_node*
(*)(gimple_stmt_iterator*, bool*, walk_stmt_info*), tree_node* (*)(tree_node**,
int*, void*), walk_stmt_info*)
        ../../gcc/gimple-walk.c:605
0x4beb93 walk_gimple_seq_mod(gimple**, tree_node* (*)(gimple_stmt_iterator*,
bool*, walk_stmt_info*), tree_node* (*)(tree_node**, int*, void*),
walk_stmt_info*)
        ../../gcc/gimple-walk.c:51
0x7e4801 walk_body
        ../../gcc/tree-nested.c:713
0x7e4801 walk_function
        ../../gcc/tree-nested.c:724
0x7e4801 walk_all_functions
        ../../gcc/tree-nested.c:789
0x7ecb2f lower_nested_functions(tree_node*)
        ../../gcc/tree-nested.c:3551
0x363201 cgraph_node::analyze()
        ../../gcc/cgraphunit.c:676
0x365df7 analyze_functions
        ../../gcc/cgraphunit.c:1227
0x366c45 symbol_table::finalize_compilation_unit()
        ../../gcc/cgraphunit.c:2974


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94780
[Bug 94780] [8/9 Regression] ICE in walk_body at gcc/tree-nested.c:713 since
r6-3632-gf6f69fb09c5f81df

[Bug target/94950] [8/9/10 regression] ICE in gcc.dg/pr94780.c on riscv64

[schwab@linux-m68k.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94950

--- Comment #1 from Andreas Schwab <schwab@linux-m68k.org> ---
Similar crash for pr94842.c:

/daten/riscv64/gcc/gcc-20200502/gcc/testsuite/gcc.dg/pr94842.c:8:1: internal
compiler error: Segmentation fault
0x74a3af crash_signal
        ../../gcc/toplev.c:328
0x7ea370 tree_check(tree_node*, char const*, int, char const*, tree_code)
        ../../gcc/tree.h:3293
0x7ea370 convert_nonlocal_reference_op
        ../../gcc/tree-nested.c:1064
0x9e3abb walk_tree_1(tree_node**, tree_node* (*)(tree_node**, int*, void*),
void*, hash_set<tree_node*, false, default_hash_traits<tree_node*> >*,
tree_node* (*)(tree_node**, int*, tree_node* (*)(tree_node**, int*, void*),
void*, hash_set<tree_node*, false, default_hash_traits<tree_node*> >*))
        ../../gcc/tree.c:12003
0x4be799 walk_gimple_op(gimple*, tree_node* (*)(tree_node**, int*, void*),
walk_stmt_info*)
        ../../gcc/gimple-walk.c:268
0x4be9e3 walk_gimple_stmt(gimple_stmt_iterator*, tree_node*
(*)(gimple_stmt_iterator*, bool*, walk_stmt_info*), tree_node* (*)(tree_node**,
int*, void*), walk_stmt_info*)
        ../../gcc/gimple-walk.c:596
0x4beb93 walk_gimple_seq_mod(gimple**, tree_node* (*)(gimple_stmt_iterator*,
bool*, walk_stmt_info*), tree_node* (*)(tree_node**, int*, void*),
walk_stmt_info*)
        ../../gcc/gimple-walk.c:51
0x4beadb walk_gimple_stmt(gimple_stmt_iterator*, tree_node*
(*)(gimple_stmt_iterator*, bool*, walk_stmt_info*), tree_node* (*)(tree_node**,
int*, void*), walk_stmt_info*)
        ../../gcc/gimple-walk.c:605
0x4beb93 walk_gimple_seq_mod(gimple**, tree_node* (*)(gimple_stmt_iterator*,
bool*, walk_stmt_info*), tree_node* (*)(tree_node**, int*, void*),
walk_stmt_info*)
        ../../gcc/gimple-walk.c:51
0x7e4801 walk_body
        ../../gcc/tree-nested.c:713
0x7e4801 walk_function
        ../../gcc/tree-nested.c:724
0x7e4801 walk_all_functions
        ../../gcc/tree-nested.c:789
0x7ecb2f lower_nested_functions(tree_node*)
        ../../gcc/tree-nested.c:3551
0x363201 cgraph_node::analyze()
        ../../gcc/cgraphunit.c:676
0x365df7 analyze_functions
        ../../gcc/cgraphunit.c:1227
0x366c45 symbol_table::finalize_compilation_unit()
        ../../gcc/cgraphunit.c:2974

[Bug target/94950] [8/9/10 regression] ICE in gcc.dg/pr94780.c on riscv64

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94950

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Assignee|unassigned at gcc dot gnu.org      |jakub at gcc dot gnu.org
   Last reconfirmed|                            |2020-05-05
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |ASSIGNED

--- Comment #2 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Created attachment 48452
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=48452&action=edit
gcc11-pr94950.patch

Untested obvious fix, but I have no way to actually test this myself except for
the testcase (which got fixed).

[Bug target/94950] [8/9/10/11 regression] ICE in gcc.dg/pr94780.c on riscv64

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94950

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[8/9/10 regression] ICE in  |[8/9/10/11 regression] ICE
                   |gcc.dg/pr94780.c on riscv64 |in gcc.dg/pr94780.c on
                   |                            |riscv64
   Target Milestone|---                         |8.5

[Bug target/94950] [8/9/10/11 regression] ICE in gcc.dg/pr94780.c on riscv64

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94950

--- Comment #3 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:b4ace720e004f736f1ee46b374c12f9826aad630

commit r11-128-gb4ace720e004f736f1ee46b374c12f9826aad630
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Wed May 6 09:40:33 2020 +0200

    riscv: Fix up riscv_atomic_assign_expand_fenv [PR94950]

    Similarly to the fixes on many other targets, riscv needs to use
TARGET_EXPR
    to avoid having the create_tmp_var_raw temporaries without proper
DECL_CONTEXT
    and not mentioned in local decls.

    2020-05-06  Jakub Jelinek  <jakub@redhat.com>

            PR target/94950
            * config/riscv/riscv-builtins.c (riscv_atomic_assign_expand_fenv):
Use
            TARGET_EXPR instead of MODIFY_EXPR for first assignment to
old_flags.

[Bug target/94950] [8/9/10 regression] ICE in gcc.dg/pr94780.c on riscv64

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94950

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[8/9/10/11 regression] ICE  |[8/9/10 regression] ICE in
                   |in gcc.dg/pr94780.c on      |gcc.dg/pr94780.c on riscv64
                   |riscv64                     |

--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Should be fixed on the trunk so far.

[Bug target/94950] [8/9/10 regression] ICE in gcc.dg/pr94780.c on riscv64

[wilson at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94950

--- Comment #5 from Jim Wilson <wilson at gcc dot gnu.org> ---
I tested it with an rv64gc-linux cross compiler.  The patch fixes these
failures:
FAIL: gcc.dg/pr94780.c (internal compiler error)                                
FAIL: gcc.dg/pr94780.c (test for excess errors)                                 
FAIL: gcc.dg/pr94842.c (internal compiler error)                                
FAIL: gcc.dg/pr94842.c (test for excess errors)
There are no regressions.

I think it should be backported to the gcc-10 release branch.

[Bug target/94950] [8/9/10 regression] ICE in gcc.dg/pr94780.c on riscv64

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94950

--- Comment #6 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
(In reply to Jim Wilson from comment #5)
> I tested it with an rv64gc-linux cross compiler.  The patch fixes these

Thanks.

> I think it should be backported to the gcc-10 release branch.

Sure, but at this point I'd prefer after 10.1.0 is released tomorrow.  This
isn't a recent regression and so shouldn't block the release.

[Bug target/94950] [8/9/10 regression] ICE in gcc.dg/pr94780.c on riscv64

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94950

--- Comment #7 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-10 branch has been updated by Jakub Jelinek
<jakub@gcc.gnu.org>:

https://gcc.gnu.org/g:5454a13add37fa6a8eedbf9d2f6bdc63a7825e2c

commit r10-8113-g5454a13add37fa6a8eedbf9d2f6bdc63a7825e2c
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Wed May 6 09:40:33 2020 +0200

    riscv: Fix up riscv_atomic_assign_expand_fenv [PR94950]

    Similarly to the fixes on many other targets, riscv needs to use
TARGET_EXPR
    to avoid having the create_tmp_var_raw temporaries without proper
DECL_CONTEXT
    and not mentioned in local decls.

    2020-05-06  Jakub Jelinek  <jakub@redhat.com>

            PR target/94950
            * config/riscv/riscv-builtins.c (riscv_atomic_assign_expand_fenv):
Use
            TARGET_EXPR instead of MODIFY_EXPR for first assignment to
old_flags.

[Bug target/94950] [8/9 regression] ICE in gcc.dg/pr94780.c on riscv64

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94950

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[8/9/10 regression] ICE in  |[8/9 regression] ICE in
                   |gcc.dg/pr94780.c on riscv64 |gcc.dg/pr94780.c on riscv64

--- Comment #8 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Fixed for 10.2+ too.

[Bug target/94950] [8/9 regression] ICE in gcc.dg/pr94780.c on riscv64

[wilson at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94950

Jim Wilson <wilson at gcc dot gnu.org> changed:

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

--- Comment #9 from Jim Wilson <wilson at gcc dot gnu.org> ---
Fixed on mainline and gcc-10 branch.

[Bug target/94950] [8/9 regression] ICE in gcc.dg/pr94780.c on riscv64

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94950

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

https://gcc.gnu.org/g:79002374e81588fa1052c21f3d397cf9ad61f97d

commit r9-8896-g79002374e81588fa1052c21f3d397cf9ad61f97d
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Wed May 6 09:40:33 2020 +0200

    riscv: Fix up riscv_atomic_assign_expand_fenv [PR94950]

    Similarly to the fixes on many other targets, riscv needs to use
TARGET_EXPR
    to avoid having the create_tmp_var_raw temporaries without proper
DECL_CONTEXT
    and not mentioned in local decls.

    2020-05-06  Jakub Jelinek  <jakub@redhat.com>

            PR target/94950
            * config/riscv/riscv-builtins.c (riscv_atomic_assign_expand_fenv):
Use
            TARGET_EXPR instead of MODIFY_EXPR for first assignment to
old_flags.

    (cherry picked from commit 5454a13add37fa6a8eedbf9d2f6bdc63a7825e2c)