[Bug rtl-optimization/63823] New: [5.0 Regression] MIPS will not build due to LRA ICE with 64 bit ABI

[Bug rtl-optimization/63823] New: [5.0 Regression] MIPS will not build due to LRA ICE with 64 bit ABI

[sje at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63823

            Bug ID: 63823
           Summary: [5.0 Regression] MIPS will not build due to LRA ICE
                    with 64 bit ABI
           Product: gcc
           Version: 5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: rtl-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: sje at gcc dot gnu.org

My x86 to MIPS build is failing when the new compiler cannot compile a file
from glibc.  A cut-down preprocessed test case is attached.  It only fails with
the 64 bit ABI, with LRA, and with -O2 -g.

% install-mips-linux-gnu/bin/mips-linux-gnu-gcc -mips64r2 -mabi=64 -O2 -g -c
x.c
x.c: In function '__ieee754_jnl':
x.c:73:1: internal compiler error: in alter_subregs, at lra-spills.c:633
 }
 ^
0x8ddafa alter_subregs
    /scratch/sellcey/repos/ice/src/gcc/gcc/lra-spills.c:633
0x8df1fd lra_final_code_change()
    /scratch/sellcey/repos/ice/src/gcc/gcc/lra-spills.c:739
0x8c48fb lra(_IO_FILE*)
    /scratch/sellcey/repos/ice/src/gcc/gcc/lra.c:2358
0x880729 do_reload
    /scratch/sellcey/repos/ice/src/gcc/gcc/ira.c:5380
0x880729 execute
    /scratch/sellcey/repos/ice/src/gcc/gcc/ira.c:5550
Please submit a full bug report,
with preprocessed source if appropriate.
Please include the complete backtrace with any bug report.
See <http://gcc.gnu.org/bugs.html> for instructions.

[Bug rtl-optimization/63823] [5.0 Regression] MIPS will not build due to LRA ICE with 64 bit ABI

[sje at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63823

--- Comment #1 from Steve Ellcey <sje at gcc dot gnu.org> ---
Created attachment 33941
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=33941&action=edit
Preprocessed test case

[Bug rtl-optimization/63823] [5.0 Regression] MIPS will not build due to LRA ICE with 64 bit ABI

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63823

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
This was working as of revision 217256 for me.

[Bug rtl-optimization/63823] [5.0 Regression] MIPS will not build due to LRA ICE with 64 bit ABI

[sje at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63823

--- Comment #3 from Steve Ellcey <sje at gcc dot gnu.org> ---
I think the failure may have started with r217265, but I have not confirmed
that for sure.

r217265 | vmakarov | 2014-11-09 08:45:15 -0800 (Sun, 09 Nov 2014) | 34 lines

2014-11-09  Vladimir Makarov  <vmakarov@redhat.com>

        PR rtl-optimization/63620
        * lra-constraints.c (substitute_pseudo): Add prefix lra_ to the
        name.  Move to lra.c.  Make it external.
        (substitute_pseudo_within_insn): Ditto.
        (inherit_reload_reg, split_reg, remove_inheritance_pseudos): Use
        the new names.
        (undo_optional_reloads): Ditto.
        * lra-int.h (lra_dump_bitmap_with_title, lra_substitute_pseudo):
        New prototypes.
.
.
.

[Bug rtl-optimization/63823] [5.0 Regression] MIPS will not build due to LRA ICE with 64 bit ABI

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63823

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |5.0

[Bug rtl-optimization/63823] [5.0 Regression] MIPS will not build due to LRA ICE with 64 bit ABI

[vmakarov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63823

--- Comment #6 from Vladimir Makarov <vmakarov at gcc dot gnu.org> ---
(In reply to Robert Suchanek from comment #5)
> It appears that enabling the debug info can trigger the ICE. In the
> testcase, after the patch, an instruction 1136 gets deleted and all
> references to pseudo 704 meant to be updated.
> 
> The following change in process_bb_lives () triggers the assertion later in
> the pass.
> 
> +             EXECUTE_IF_SET_IN_BITMAP
> +               (&lra_reg_info[dst_regno].insn_bitmap, 0, uid, bi)
> +               {
> +                 insn = lra_insn_recog_data[uid]->insn;
> +                 lra_substitute_pseudo_within_insn (insn, dst_regno,
> +                                                    SET_SRC (set));
> +                 lra_update_insn_regno_info (insn);
> +               }
> 
> On the first iteration, the list of insns using pseudo 704 is correct:
> 
> (gdb) call debug_bitmap(&lra_reg_info[704].insn_bitmap)
> 
> first = 0x188b720 current = 0x188b770 indx = 3
>         0x188b720 next = 0x188b770 prev = (nil) indx = 2
>                 bits = { 343 }
>         0x188b770 next = (nil) prev = 0x188b720 indx = 3
>                 bits = { 384 }
> 
> Insn 343 gets updated, pseudo 704 is replaced with original pseudo 234.
> However, the call to lra_update_insn_regno_info destroys the remaining
> references to pseudo 704 via invalidate_insn_data_regno_info so we end up
> with the following:
> 
> (gdb) call debug_bitmap(&lra_reg_info[704].insn_bitmap)
>                                                        
> first = 0x1855770 current = 0x1855770 indx = 3         
>         0x1855770 next = (nil) prev = (nil) indx = 3   
>                 bits = { 384 }
> 
> The next iteration(s) in EXECUTE_IF_SET_IN_BITMAP will not happen as
> bmp_iter_set () returns false. Insn 384 is left unchanged and the compiler
> fails later on assertion.

Thanks for analyzing this, Robert.  I've been working on some LRA
rematerialization issues and found this too.  I'll post a patch fixing this
issue soon.  If it works for you, I'll commit it on this week.

[Bug rtl-optimization/63823] [5.0 Regression] MIPS will not build due to LRA ICE with 64 bit ABI

[vmakarov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63823

--- Comment #7 from Vladimir Makarov <vmakarov at gcc dot gnu.org> ---
Created attachment 33956
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=33956&action=edit
The proposed patch

Here is the proposed patch.

[Bug rtl-optimization/63823] [5.0 Regression] MIPS will not build due to LRA ICE with 64 bit ABI

[robert.suchanek at imgtec dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63823

--- Comment #8 from Robert Suchanek <robert.suchanek at imgtec dot com> ---
Yes, the patch works. Glibc built fine on mips64-linux-gnu target. Thanks Vlad.

[Bug rtl-optimization/63823] [5.0 Regression] MIPS will not build due to LRA ICE with 64 bit ABI

[vmakarov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63823

--- Comment #9 from Vladimir Makarov <vmakarov at gcc dot gnu.org> ---
The patch was committed as a part of latest rematerialization patches.  So the
trunk has a fix for this.  It would be nice to confirm the fix on the trunk and
close the PR in the case of confirmation.

[Bug rtl-optimization/63823] [5.0 Regression] MIPS will not build due to LRA ICE with 64 bit ABI

[sje at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63823

Steve Ellcey <sje at gcc dot gnu.org> changed:

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

--- Comment #10 from Steve Ellcey <sje at gcc dot gnu.org> ---
My nightly build and test of the trunk on MIPS worked so I will close this out
as resolved/fixed.