[Bug bootstrap/62304] New: [5 regression] ICE in follow_jumps, find_dead_or_set_registers

[Bug bootstrap/62304] New: [5 regression] ICE in follow_jumps, find_dead_or_set_registers

[ro at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62304

            Bug ID: 62304
           Summary: [5 regression] ICE in follow_jumps,
                    find_dead_or_set_registers
           Product: gcc
           Version: 5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: bootstrap
          Assignee: unassigned at gcc dot gnu.org
          Reporter: ro at gcc dot gnu.org
                CC: dmalcolm at gcc dot gnu.org, ebotcazou at gcc dot gnu.org
              Host: sparc*-sun-solaris2.*
            Target: sparc*-sun-solaris2.*
             Build: sparc*-sun-solaris2.*

David,

one of your recent patches (since 20140822, r214311) broke Solaris/SPARC
bootstrap: when compiling the stage1 libgcc, I get e.g.

/var/gcc/regression/trunk/11-gcc/build/./gcc/xgcc
-B/var/gcc/regression/trunk/11-gcc/build/./gcc/
-B/vol/gcc/sparc-sun-solaris2.11/bin/ -B/vol/gcc/sparc-sun-solaris2.11/lib/
-isystem /vol/gcc/sparc-sun-solaris2.11/include -isystem
/vol/gcc/sparc-sun-solaris2.11/sys-include    -g -O2 -m64 -O2  -g -O2 -DIN_GCC 
  -W -Wall -Wno-narrowing -Wwrite-strings -Wcast-qual -Wno-format
-Wstrict-prototypes -Wmissing-prototypes -Wold-style-definition  -isystem
./include   -fPIC -g -DIN_LIBGCC2 -fbuilding-libgcc -fno-stack-protector  
-fPIC -I. -I. -I../../.././gcc -I/vol/gcc/src/hg/trunk/local/libgcc
-I/vol/gcc/src/hg/trunk/local/libgcc/.
-I/vol/gcc/src/hg/trunk/local/libgcc/../gcc
-I/vol/gcc/src/hg/trunk/local/libgcc/../include  -DHAVE_CC_TLS  -o _absvsi2.o
-MT _absvsi2.o -MD -MP -MF _absvsi2.dep -DL_absvsi2 -c
/vol/gcc/src/hg/trunk/local/libgcc/libgcc2.c -fvisibility=hidden -DHIDE_EXPORTS
/vol/gcc/src/hg/trunk/local/libgcc/libgcc2.c: In function '__absvdi2':
/vol/gcc/src/hg/trunk/local/libgcc/libgcc2.c:232:1: internal compiler error: in
safe_as_a, at is-a.h:205
 }
 ^
0xae862f follow_jumps
        /vol/gcc/src/hg/trunk/local/gcc/reorg.c:2326
0xaeb4fb relax_delay_slots
        /vol/gcc/src/hg/trunk/local/gcc/reorg.c:3175
0xaedfbb dbr_schedule
        /vol/gcc/src/hg/trunk/local/gcc/reorg.c:3743
0xaeed83 rest_of_handle_delay_slots
        /vol/gcc/src/hg/trunk/local/gcc/reorg.c:3885
0xaeede7 execute
        /vol/gcc/src/hg/trunk/local/gcc/reorg.c:3916

resp.

/var/gcc/regression/trunk/11-gcc/build/./gcc/xgcc
-B/var/gcc/regression/trunk/11-gcc/build/./gcc/
-B/vol/gcc/sparc-sun-solaris2.11/bin/ -B/vol/gcc/sparc-sun-solaris2.11/lib/
-isystem /vol/gcc/sparc-sun-solaris2.11/include -isystem
/vol/gcc/sparc-sun-solaris2.11/sys-include    -g -O2 -m64 -O2  -g -O2 -DIN_GCC 
  -W -Wall -Wno-narrowing -Wwrite-strings -Wcast-qual -Wno-format
-Wstrict-prototypes -Wmissing-prototypes -Wold-style-definition  -isystem
./include   -fPIC -g -DIN_LIBGCC2 -fbuilding-libgcc -fno-stack-protector  
-fPIC -I. -I. -I../../.././gcc -I/vol/gcc/src/hg/trunk/local/libgcc
-I/vol/gcc/src/hg/trunk/local/libgcc/.
-I/vol/gcc/src/hg/trunk/local/libgcc/../gcc
-I/vol/gcc/src/hg/trunk/local/libgcc/../include  -DHAVE_CC_TLS  -o _addvdi3.o
-MT _addvdi3.o -MD -MP -MF _addvdi3.dep -DL_addvdi3 -c
/vol/gcc/src/hg/trunk/local/libgcc/libgcc2.c -fvisibility=hidden -DHIDE_EXPORTS
/vol/gcc/src/hg/trunk/local/libgcc/libgcc2.c: In function '__addvti3':
/vol/gcc/src/hg/trunk/local/libgcc/libgcc2.c:109:1: internal compiler error: in
safe_as_a, at is-a.h:205
 }
 ^
0xaf0bbb find_dead_or_set_registers
        /vol/gcc/src/hg/trunk/local/gcc/resource.c:500
0xaf1683 find_dead_or_set_registers
        /vol/gcc/src/hg/trunk/local/gcc/resource.c:577
0xaf3e33 mark_target_live_regs(rtx_insn*, rtx_insn*, resources*)
        /vol/gcc/src/hg/trunk/local/gcc/resource.c:1115
0xae8b0b fill_slots_from_thread
        /vol/gcc/src/hg/trunk/local/gcc/reorg.c:2404
0xaea88b fill_eager_delay_slots
        /vol/gcc/src/hg/trunk/local/gcc/reorg.c:2906
0xaedfaf dbr_schedule
        /vol/gcc/src/hg/trunk/local/gcc/reorg.c:3742
0xaeed83 rest_of_handle_delay_slots
        /vol/gcc/src/hg/trunk/local/gcc/reorg.c:3885
0xaeede7 execute
        /vol/gcc/src/hg/trunk/local/gcc/reorg.c:3916

Need to determine which exact patch caused this and if it reproduces in a cross
compiler.

  Rainer

[Bug bootstrap/62304] [5 regression] ICE in follow_jumps, find_dead_or_set_registers

[ro at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62304

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

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

[Bug bootstrap/62304] [5 regression] ICE in follow_jumps, find_dead_or_set_registers

[dmalcolm at redhat dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62304

Dave Malcolm <dmalcolm at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dmalcolm at redhat dot com

--- Comment #1 from Dave Malcolm <dmalcolm at redhat dot com> ---
The crash in find_dead_or_set_registers is the one discussed in:
https://gcc.gnu.org/ml/gcc-patches/2014-08/msg02619.html
and I introduced it in r214693 (aka patch #225)

The crash in follow_jumps is similar to it, and occurs here:
    new_thread = follow_jumps (JUMP_LABEL_AS_INSN (new_thread), insn,
                   &crossing);
where JUMP_LABEL_AS_INSN erroneously tries to coerce JUMP_LABEL (new_thread) to
be an insn, but it's a RETURN.  I introduced this one in r214684 (aka patch
#218)

Sorry; am working on a fix.

[Bug bootstrap/62304] [5 regression] ICE in follow_jumps, find_dead_or_set_registers

[ro at CeBiTec dot Uni-Bielefeld.DE]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62304

--- Comment #2 from ro at CeBiTec dot Uni-Bielefeld.DE <ro at CeBiTec dot Uni-Bielefeld.DE> ---
> --- Comment #1 from Dave Malcolm <dmalcolm at redhat dot com> ---
> The crash in find_dead_or_set_registers is the one discussed in:
> https://gcc.gnu.org/ml/gcc-patches/2014-08/msg02619.html
> and I introduced it in r214693 (aka patch #225)
>
> The crash in follow_jumps is similar to it, and occurs here:
>     new_thread = follow_jumps (JUMP_LABEL_AS_INSN (new_thread), insn,
>                    &crossing);
> where JUMP_LABEL_AS_INSN erroneously tries to coerce JUMP_LABEL (new_thread) to
> be an insn, but it's a RETURN.  I introduced this one in r214684 (aka patch
> #218)

I see, thanks.  I completely lost track what's happening in this patch
series...

> Sorry; am working on a fix.

Great, thanks.

    Rainer

[Bug bootstrap/62304] [5 regression] ICE in follow_jumps, find_dead_or_set_registers

[dmalcolm at redhat dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62304

--- Comment #3 from Dave Malcolm <dmalcolm at redhat dot com> ---
Do you have preprocessed source handy for the reorg.c crash?

[Bug bootstrap/62304] [5 regression] ICE in follow_jumps, find_dead_or_set_registers

[dmalcolm at redhat dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62304

--- Comment #4 from Dave Malcolm <dmalcolm at redhat dot com> ---
(In reply to Dave Malcolm from comment #3)
> Do you have preprocessed source handy for the reorg.c crash?

No need; I've reproduced it locally now.

[Bug bootstrap/62304] [5 regression] ICE in follow_jumps, find_dead_or_set_registers

[dmalcolm at redhat dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62304

--- Comment #5 from Dave Malcolm <dmalcolm at redhat dot com> ---
Created attachment 33416
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=33416&action=edit
Candidate patch.  Bootstrapped on x86_64; seems to work on stage1 cris and
sparc

[Bug bootstrap/62304] [5 regression] ICE in follow_jumps, find_dead_or_set_registers

[sje at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62304

--- Comment #6 from Steve Ellcey <sje at gcc dot gnu.org> ---
(In reply to Dave Malcolm from comment #5)
> Created attachment 33416 [details]
> Candidate patch.  Bootstrapped on x86_64; seems to work on stage1 cris and
> sparc

This patch fixed my MIPS build.  I am running the testsuite now and it 
looks good so far.

[Bug bootstrap/62304] [5 regression] ICE in follow_jumps, find_dead_or_set_registers

[ro at CeBiTec dot Uni-Bielefeld.DE]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62304

--- Comment #7 from ro at CeBiTec dot Uni-Bielefeld.DE <ro at CeBiTec dot Uni-Bielefeld.DE> ---
> --- Comment #5 from Dave Malcolm <dmalcolm at redhat dot com> ---
> Created attachment 33416
>   --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=33416&action=edit
> Candidate patch.  Bootstrapped on x86_64; seems to work on stage1 cris and
> sparc

With this patch, sparc-sun-solaris2.11 bootstrap completed
successfully.  Compared to the last one of a week ago, there are two
testsuite regressions, but I assume they are unrelated:

FAIL: gcc.dg/20111227-2.c scan-rtl-dump ree "Elimination opportunities = 3
realized = 3"
FAIL: gcc.dg/vect/vect-33.c scan-tree-dump-times vect "Alignment of access
forced using versioning" 1

    Rainer

[Bug bootstrap/62304] [5 regression] ICE in follow_jumps, find_dead_or_set_registers

[dmalcolm at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62304

--- Comment #8 from dmalcolm at gcc dot gnu.org ---
Author: dmalcolm
Date: Sat Aug 30 14:34:02 2014
New Revision: 214752

URL: https://gcc.gnu.org/viewcvs?rev=214752&root=gcc&view=rev
Log:
    PR bootstrap/62304

    * gcc/reorg.c (skip_consecutive_labels): Convert return type and
    param back from rtx_insn * to rtx.  Rename param from "label" to
    "label_or_return", reintroducing "label" as an rtx_insn * after
    we've ensured it's not a RETURN.
    (first_active_target_insn): Likewise for return type and param;
    add a checked cast to rtx_insn * once we've ensured "insn" is not
    a RETURN.
    (steal_delay_list_from_target): Convert param "pnew_thread" back
    from rtx_insn ** to rtx *.  Replace use of JUMP_LABEL_AS_INSN
    with JUMP_LABEL.
    (own_thread_p): Convert param "thread" back from an rtx_insn * to
    an rtx.  Introduce local rtx_insn * "thread_insn" with a checked
    cast once we've established we're not dealing with a RETURN,
    renaming subsequent uses of "thread" to "thread_insn".
    (fill_simple_delay_slots): Convert uses of JUMP_LABEL_AS_INSN back
    to JUMP_LABEL.
    (follow_jumps): Convert return type and param "label" from
    rtx_insn * back to rtx.  Move initialization of "value" to after
    the handling for ANY_RETURN_P, adding a checked cast there to
    rtx_insn *.  Convert local rtx_insn * "this_label" to an rtx and
    rename to "this_label_or_return", reintroducing "this_label" as
    an rtx_insn * once we've handled the case where it could be an
    ANY_RETURN_P.
    (fill_slots_from_thread): Rename param "thread" to
    "thread_or_return", converting from an rtx_insn * back to an rtx.
    Reintroduce name "thread" as an rtx_insn * local with a checked
    cast once we've handled the case of it being an ANY_RETURN_P.
    Convert local "new_thread" from an rtx_insn * back to an rtx.
    Add a checked cast when assigning to "trial" from "new_thread".
    Convert use of JUMP_LABEL_AS_INSN back to JUMP_LABEL.  Add a
    checked cast to rtx_insn * from "new_thread" when invoking
    get_label_before.
    (fill_eager_delay_slots): Convert locals "target_label",
    "insn_at_target" from rtx_insn * back to rtx.
    Convert uses of JUMP_LABEL_AS_INSN back to JUMP_LABEL.
    (relax_delay_slots): Convert locals "trial", "target_label" from
    rtx_insn * back to rtx.  Convert uses of JUMP_LABEL_AS_INSN back
    to JUMP_LABEL.  Add a checked cast to rtx_insn * on "trial" when
    invoking update_block.
    (dbr_schedule): Convert use of JUMP_LABEL_AS_INSN back to
    JUMP_LABEL; this removes all JUMP_LABEL_AS_INSN from reorg.c.

    * resource.h (mark_target_live_regs): Undo erroneous conversion
    of second param of r214693, converting it back from rtx_insn * to
    rtx, since it could be a RETURN.

    * resource.c (find_dead_or_set_registers): Similarly, convert
    param "jump_target" back from an rtx_insn ** to an rtx *, as we
    could be writing back a RETURN.  Rename local rtx_insn * "next" to
    "next_insn", and introduce "lab_or_return" as a local rtx,
    handling the case where JUMP_LABEL (this_jump_insn) is a RETURN.
    (mark_target_live_regs): Undo erroneous conversion
    of second param of r214693, converting it back from rtx_insn * to
    rtx, since it could be a RETURN.  Rename it from "target" to
    "target_maybe_return", reintroducing the name "target" as a local
    rtx_insn * with a checked cast, after we've handled the case of
    ANY_RETURN_P.


Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/reorg.c
    trunk/gcc/resource.c
    trunk/gcc/resource.h

[Bug bootstrap/62304] [5 regression] ICE in follow_jumps, find_dead_or_set_registers

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62304

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at gcc dot gnu.org

--- Comment #9 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Is this now fixed?

[Bug bootstrap/62304] [5 regression] ICE in follow_jumps, find_dead_or_set_registers

[ebotcazou at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62304

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

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

--- Comment #10 from Eric Botcazou <ebotcazou at gcc dot gnu.org> ---
Sure.