[Bug rtl-optimization/56912] New: scheduler change breaks linux kernel LTO build with 4.8

[Bug rtl-optimization/56912] New: scheduler change breaks linux kernel LTO build with 4.8

[andi-gcc at firstfloor dot org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56912

             Bug #: 56912
           Summary: scheduler change breaks linux kernel LTO build with
                    4.8
    Classification: Unclassified
           Product: gcc
           Version: unknown
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: rtl-optimization
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: andi-gcc@firstfloor.org


For the Linux kernel LTO build I get ICEs during LTO (segfaults) with the
recent 4.8 branch. I bisected it down to this patch and reverting fixes it.

No simple test case unfortunately as it is LTO 


            Backport from mainline
            2013-02-25  Andrey Belevantsev  <abel@ispras.ru>
            Alexander Monakov  <amonakov@ispras.ru>

            PR middle-end/56077
            * sched-deps.c (sched_analyze_insn): When reg_pending_barrier,
            flush pending lists also on non-jumps.  Adjust comment.



Typical crash:



#7  <signal handler called>
#8  sched_analyze_1 (deps=0x7fff550a5c00, x=0x7fa0311e2ed0,
insn=0x7fa0311f40d8)
    at ../../gcc/gcc/sched-deps.c:2479
#9  0x0000000000b668d5 in sched_analyze_insn (deps=deps@entry=0x7fff550a5c00, 
    x=0x7fa0311e2e70, insn=insn@entry=0x7fa0311f40d8) at
../../gcc/gcc/sched-deps.c:2859
#10 0x0000000000b6859b in deps_analyze_insn (deps=deps@entry=0x7fff550a5c00, 
    insn=insn@entry=0x7fa0311f40d8) at ../../gcc/gcc/sched-deps.c:3505
#11 0x0000000000b689c3 in sched_analyze (deps=0x7fff550a5c00, head=<optimized
out>, 
    tail=0x7fa0311f8c18) at ../../gcc/gcc/sched-deps.c:3653
#12 0x000000000070b635 in compute_block_dependences (bb=0) at
../../gcc/gcc/sched-rgn.c:2702
#13 sched_rgn_compute_dependencies (rgn=rgn@entry=5) at
../../gcc/gcc/sched-rgn.c:3140
#14 0x000000000070df84 in schedule_region (rgn=5) at
../../gcc/gcc/sched-rgn.c:2915
#15 schedule_insns () at ../../gcc/gcc/sched-rgn.c:3299
#16 schedule_insns () at ../../gcc/gcc/sched-rgn.c:3278
#17 0x000000000070e3b1 in rest_of_handle_sched2 () at
../../gcc/gcc/sched-rgn.c:3523
#18 0x00000000006b534e in execute_one_pass (pass=pass@entry=0x112e240
<pass_sched2>)
    at ../../gcc/gcc/passes.c:2084
#19 0x00000000006b56bd in execute_pass_list (pass=0x112e240 <pass_sched2>)
    at ../../gcc/gcc/passes.c:2139
#20 0x00000000006b56cf in execute_pass_list (pass=0x112d840 <pass_postreload>)
    at ../../gcc/gcc/passes.c:2140
#21 0x00000000006b56cf in execute_pass_list (pass=0x112d8a0
<pass_rest_of_compilation>)
    at ../../gcc/gcc/passes.c:2140
#22 0x0000000000792043 in tree_rest_of_compilation (fndecl=0x7fa03f899700)
    at ../../gcc/gcc/tree-optimize.c:422
#23 0x0000000000536f7b in cgraph_expand_function (node=0x7fa03c49b5a0)

[Bug rtl-optimization/56912] scheduler change breaks linux kernel LTO build with 4.8

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56912

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> 2013-04-11 00:39:29 UTC ---
>No simple test case unfortunately as it is LTO 

Actually it should not be hard to get a testcase if you use mutli-delta.  I
have reduced some LTO testcases already using mutli-delta and I had a customer
use it too to reduce the testcases for us when they reported the bugs due to
LTO.

[Bug rtl-optimization/56912] scheduler change breaks linux kernel LTO build with 4.8

[abel at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56912

Andrey Belevantsev <abel at gcc dot gnu.org> changed:

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

--- Comment #2 from Andrey Belevantsev <abel at gcc dot gnu.org> 2013-04-11 06:22:51 UTC ---
I will take a look given the test case.  It doesn't have to be small, any steps
to reproduce the issue would be fine (hopefully).

[Bug rtl-optimization/56912] [4.8/4.9 Regression] scheduler change breaks linux kernel LTO build with 4.8

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56912

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Blocks|                            |56077
   Target Milestone|---                         |4.8.1
            Summary|scheduler change breaks     |[4.8/4.9 Regression]
                   |linux kernel LTO build with |scheduler change breaks
                   |4.8                         |linux kernel LTO build with
                   |                            |4.8

[Bug rtl-optimization/56912] [4.8/4.9 Regression] scheduler change breaks linux kernel LTO build with 4.8

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56912

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|4.8.1                       |4.8.2

--- Comment #3 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
GCC 4.8.1 has been released.

[Bug rtl-optimization/56912] [4.8/4.9 Regression] scheduler change breaks linux kernel LTO build with 4.8

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56912

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|4.8.2                       |4.8.3

--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
GCC 4.8.2 has been released.

[Bug rtl-optimization/56912] [4.8/4.9 Regression] scheduler change breaks linux kernel LTO build with 4.8

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56912

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING
   Last reconfirmed|                            |2013-10-25
     Ever confirmed|0                           |1

--- Comment #5 from Richard Biener <rguenth at gcc dot gnu.org> ---
Does this issue still prevail?

[Bug rtl-optimization/56912] [4.8/4.9 Regression] scheduler change breaks linux kernel LTO build with 4.8

[andi-gcc at firstfloor dot org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56912

--- Comment #6 from Andi Kleen <andi-gcc at firstfloor dot org> ---
Seems to be fixed.

[Bug rtl-optimization/56912] [4.8/4.9 Regression] scheduler change breaks linux kernel LTO build with 4.8

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56912

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

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

--- Comment #7 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed.