[Bug target/36450] New: ICE in insert_restore with GET_MODE_ALIGNMENT mem

[Bug target/36450] New: ICE in insert_restore with GET_MODE_ALIGNMENT mem

[aldot at gcc dot gnu dot org]

/there.pentium4/build_i686/staging_dir/usr/bin/i686-linux-uclibc-gcc -c
libc/stdlib/strtod.c -o libc/stdlib/strtod.os -include ./include/libc-symbols.h
-Wall -Wstrict-prototypes -fno-strict-aliasing -Wnested-externs -Wshadow
-Wmissing-noreturn -Wmissing-format-attribute -Wformat=2 -Wmissing-prototypes
-Wmissing-declarations -Wnonnull -Wundef -ffunction-sections -fdata-sections
-m32 -fno-stack-protector -fno-builtin -nostdinc -I./include -I.
-I./libc/sysdeps/linux/i386 --sysroot=/there.pentium4/build_i686/staging_dir/
-isysroot /there.pentium4/build_i686/staging_dir -I=/usr/include -pipe -O3
-ftree-vect-loop-version -ftree-vectorize -fvariable-expansion-in-unroller
-funswitch-loops -funroll-all-loops -ftree-loop-distribution -ftree-loop-im
-fivopts -fvect-cost-model -fira
-I/there.pentium4/build_i686/staging_dir/usr/include -mtune=pentium4
-march=pentium4 -DUCLIBC_INTERNAL -std=gnu99 -Os -funit-at-a-time
-fno-tree-loop-optimize -fno-tree-dominator-opts -fno-strength-reduce
-fomit-frame-pointer -mpreferred-stack-boundary=2 -falign-functions=1
-falign-jumps=1 -falign-labels=1 -falign-loops=1
-I./libpthread/linuxthreads.old/sysdeps/unix/sysv/linux/i386
-I./libpthread/linuxthreads.old/sysdeps/i386
-I./libpthread/linuxthreads.old/sysdeps/unix/sysv/linux
-I./libpthread/linuxthreads.old/sysdeps/pthread -I./libpthread/linuxthreads.old
-I./libpthread -I/there.pentium4/toolchain_build_i686/linux/include/
-I/scratch/src/br.update.pentium4/build_i686/staging_dir/usr/bin/../lib/gcc/i686-linux-uclibc/4.4.0//include-fixed
-I/scratch/src/br.update.pentium4/build_i686/staging_dir/usr/bin/../lib/gcc/i686-linux-uclibc/4.4.0/include
In file included from libc/stdlib/strtod.c:8:
libc/stdlib/_strtod.c: In function 'strtod':
libc/stdlib/_strtod.c:580: internal compiler error: in insert_restore, at
caller-save.c:2332


-- 
           Summary: ICE in insert_restore with GET_MODE_ALIGNMENT mem
           Product: gcc
           Version: 4.4.0
            Status: UNCONFIRMED
          Keywords: ice-on-valid-code
          Severity: normal
          Priority: P3
         Component: target
        AssignedTo: vmakarov at redhat dot com
        ReportedBy: aldot at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug target/36450] [IRA] ICE in insert_restore with GET_MODE_ALIGNMENT mem

[aldot at gcc dot gnu dot org]

------- Comment #1 from aldot at gcc dot gnu dot org  2008-06-06 14:40 -------
Created an attachment (id=15726)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15726&action=view)
original testcase from uClibc

ira-branch revision 136341
Works with -fno-ira

works with -mpreferred-stack-boundary=4 (in i386)
fails with =2:

/there.pentium4/build_i686/staging_dir/usr/bin/i686-linux-uclibc-gcc -c
strtod.i -o bug.o -m32 -O2 -mpreferred-stack-boundary=2 -fira


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug target/36450] [IRA] ICE in insert_restore with GET_MODE_ALIGNMENT mem

[aldot at gcc dot gnu dot org]

------- Comment #2 from aldot at gcc dot gnu dot org  2008-06-09 17:08 -------
Vladimir,

Is the testcase ok as is or do you need it reduced?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug target/36450] [IRA] ICE in insert_restore with GET_MODE_ALIGNMENT mem

[vmakarov at redhat dot com]

------- Comment #3 from vmakarov at redhat dot com  2008-06-09 17:22 -------
The test is ok.  Thanks.

The problem is very simple.  The assertion causing abort is simply wrong.  I
reproduced the same bug for the old register allocator.  I just used wetstone
for this.

The assertion is wrong because alignment can not be bigger the preferred stack
boundary (please see function.c::assign_stack_slot).  It is a very old code. 
Instead of 

gcc_assert (GET_MODE_ALIGNMENT (GET_MODE (mem) <= MEM_ALIGN (mem))

in caller_save.c::insert_save/insert_restore, the assertion should be

gcc_assert (MIN (PREFERRED_STACK_BOUNDARY, GET_MODE_ALIGNMENT (GET_MODE (mem))
<= MEM_ALIGN (mem))

I'll submit the patch to the branch today or tomorrow.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug target/36450] ICE in insert_restore/insert_save with GET_MODE_ALIGNMENT mem

[aldot at gcc dot gnu dot org]

------- Comment #4 from aldot at gcc dot gnu dot org  2008-06-10 08:18 -------
pristine trunk is also broken:

sort.i:10318: internal compiler error: in insert_save, at caller-save.c:787


-- 

aldot at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[IRA] ICE in insert_restore |ICE in
                   |with GET_MODE_ALIGNMENT mem |insert_restore/insert_save
                   |                            |with GET_MODE_ALIGNMENT mem
   Target Milestone|---                         |4.4.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug target/36450] ICE in insert_restore/insert_save with GET_MODE_ALIGNMENT mem

[vmakarov at gcc dot gnu dot org]

------- Comment #5 from vmakarov at gcc dot gnu dot org  2008-06-10 19:02 -------
Subject: Bug 36450

Author: vmakarov
Date: Tue Jun 10 19:01:37 2008
New Revision: 136638

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=136638
Log:
2008-06-10  Vladimir Makarov  <vmakarov@redhat.com>
        PR target/36450
        * caller-save.c (insert_restore, insert_save):  Don't require
        alignment more than PREFERRED_STACK_BOUNDARY in asserts.


Modified:
    branches/ira/gcc/ChangeLog
    branches/ira/gcc/caller-save.c
    branches/ira/gcc/ira-int.h


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug target/36450] ICE in insert_restore/insert_save with GET_MODE_ALIGNMENT mem

[aldot at gcc dot gnu dot org]

------- Comment #6 from aldot at gcc dot gnu dot org  2008-06-10 19:34 -------
Can you please also fix trunk?
Thanks alot


-- 

aldot at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2008-06-10 19:34:47
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug target/36450] ICE in insert_restore/insert_save with GET_MODE_ALIGNMENT mem

[hjl dot tools at gmail dot com]

------- Comment #7 from hjl dot tools at gmail dot com  2008-06-10 22:05 -------
Please note that some mode alignments are the hard requirement.
This problem will go away after stack branch is merged with
trunk.


-- 

hjl dot tools at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |hjl dot tools at gmail dot
                   |                            |com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug target/36450] ICE in insert_restore/insert_save with GET_MODE_ALIGNMENT mem

[aldot at gcc dot gnu dot org]

------- Comment #8 from aldot at gcc dot gnu dot org  2008-06-11 10:35 -------
Created an attachment (id=15747)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15747&action=view)
ICEs on trunk

This ICEs on trunk with -mpreferred-stack-boundary=2 but works wih =4

gcc-4.4-HEAD  -std=gnu99 -Wall -Wshadow -Wwrite-strings -Wundef
-Wstrict-prototypes -Wunused -Wunused-parameter -Wmissing-prototypes
-Wmissing-declarations -Wdeclaration-after-statement -Wold-style-definition -Os
-fno-builtin-strlen -finline-limit=0 -fomit-frame-pointer -ffunction-sections
-fdata-sections -fno-guess-branch-probability -funsigned-char -static-libgcc
-falign-functions=1 -falign-jumps=1 -falign-labels=1 -falign-loops=1
-march=i386 -mpreferred-stack-boundary=2 -O3 -ftree-vect-loop-version
-ftree-vectorize -fvariable-expansion-in-unroller -funswitch-loops
-funroll-all-loops -ftree-loop-distribution -ftree-loop-im -fivopts
-fvect-cost-model -mtune=pentium4 -march=pentium4 -Wfatal-errors -c -o sort.o
sort.5.i


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug target/36450] [4.4 Regression] ICE in insert_restore/insert_save with GET_MODE_ALIGNMENT mem

[aldot at gcc dot gnu dot org]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2555 bytes --]



------- Comment #9 from aldot at gcc dot gnu dot org  2008-06-11 18:12 -------
(In reply to comment #8)
> Created an attachment (id=15747)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15747&action=view) [edit]
> ICEs on trunk
> 
> This ICEs on trunk with -mpreferred-stack-boundary=2 but works wih =4

s/wih =4/with anything but 2/

and of course works with e.g. 4.3.0 20080218 or trunk from a few weeks ago.

gcc-4.4-HEAD -S -o /dev/null -fno-guess-branch-probability
-mpreferred-stack-boundary=2 -Os sort.5.i -v
Using built-in specs.
Target: i686-linux-gnu
Configured with: ../../src/gcc-4.4/configure -v --enable-languages=c,fortran
--prefix=/opt/i686/gcc-4.4/ --enable-shared --with-system-zlib
--libexecdir=/opt/i686/gcc-4.4/lib --enable-nls --without-included-gettext
--enable-threads=posix --program-suffix=-4.4-HEAD --enable-__cxa_atexit
--enable-libstdcxx-allocator=mt --enable-clocale=gnu --enable-libstdcxx-debug
--enable-mpfr --disable-werror --enable-checking=yes --enable-debug
--with-gmp=/opt/i686/gmp-4.2.2/ --with-mpfr=/opt/i686/mpfr-2.3.0
--disable-multilib --disable-intermodule i686-linux-gnu
Thread model: posix
gcc version 4.4.0 20080611 (experimental) (GCC) 
COLLECT_GCC_OPTIONS='-S' '-o' '/dev/null' '-fno-guess-branch-probability'
'-mpreferred-stack-boundary=2' '-Os' '-v' '-mtune=generic'
 /opt/i686/gcc-4.4/bin/../../gcc-4.4/lib/gcc/i686-linux-gnu/4.4.0/cc1
-fpreprocessed sort.5.i -quiet -dumpbase sort.5.i -mpreferred-stack-boundary=2
-mtune=generic -auxbase-strip /dev/null -Os -version
-fno-guess-branch-probability -o /dev/null
GNU C (GCC) version 4.4.0 20080611 (experimental) (i686-linux-gnu)
        compiled by GNU C version 4.4.0 20080611 (experimental), GMP version
4.2.1, MPFR version 2.3.0-p4.
GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096
Compiler executable checksum: 0d5bbb5fdfd81762cf963cab59cb1e94
sort.5.i:13: warning: useless storage class specifier in empty declaration
sort.5.i: In function ‘sort_main’:
sort.5.i:75: internal compiler error: in insert_save, at caller-save.c:787
Please submit a full bug report,


-- 

aldot at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|ICE in                      |[4.4 Regression] ICE in
                   |insert_restore/insert_save  |insert_restore/insert_save
                   |with GET_MODE_ALIGNMENT mem |with GET_MODE_ALIGNMENT mem


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug target/36450] [4.4 Regression] ICE in insert_restore/insert_save with GET_MODE_ALIGNMENT mem

[aldot at gcc dot gnu dot org]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1866 bytes --]



------- Comment #10 from aldot at gcc dot gnu dot org  2008-06-17 13:54 -------
boostrapping pristine trunk still fails with:

/scratch/obj.i686/gcc-4.4.orig/./gcc/xgcc
-B/scratch/obj.i686/gcc-4.4.orig/./gcc/
-B/opt/i686/gcc-4.4.orig//i686-linux-gnu/bin/
-B/opt/i686/gcc-4.4.orig//i686-linux-gnu/lib/ -isystem
/opt/i686/gcc-4.4.orig//i686-linux-gnu/include -isystem
/opt/i686/gcc-4.4.orig//i686-linux-gnu/sys-include -O2 -march=pentium4
-mtune=pentium4 -pipe -DIN_GCC -mpreferred-stack-boundary=2 -W -Wall
-Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -Wold-style-definition
-isystem ./include -fPIC -g -DHAVE_GTHR_DEFAULT -DIN_LIBGCC2
-D__GCC_FLOAT_NOT_NEEDED -I. -I. -I../.././gcc
-I../../../../src/gcc-4.4.orig/libgcc -I../../../../src/gcc-4.4.orig/libgcc/.
-I../../../../src/gcc-4.4.orig/libgcc/../gcc
-I../../../../src/gcc-4.4.orig/libgcc/../include
-I../../../../src/gcc-4.4.orig/libgcc/config/libbid -DENABLE_DECIMAL_BID_FORMAT
-DHAVE_CC_TLS -DUSE_TLS -o bid64_div.o -MT bid64_div.o -MD -MP -MF
bid64_div.dep -c ../../../../src/gcc-4.4.orig/libgcc/config/libbid/bid64_div.c
In file included from
../../../../src/gcc-4.4.orig/libgcc/config/libbid/bid64_div.c:62:
../../../../src/gcc-4.4.orig/libgcc/config/libbid/bid_div_macros.h: In function
‘__div_256_by_128’:
../../../../src/gcc-4.4.orig/libgcc/config/libbid/bid_div_macros.h:542:
internal compiler error: in insert_save, at caller-save.c:787
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.
make[3]: *** [bid64_div.o] Error 1


-- 

aldot at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |build


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug target/36450] [4.4 Regression] ICE in insert_restore/insert_save with GET_MODE_ALIGNMENT mem

[rguenth at gcc dot gnu dot org]

-- 

rguenth at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P1


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug target/36450] [4.4 Regression] ICE in insert_restore/insert_save with GET_MODE_ALIGNMENT mem

[hjl dot tools at gmail dot com]

------- Comment #11 from hjl dot tools at gmail dot com  2008-07-19 23:58 -------
*** Bug 36875 has been marked as a duplicate of this bug. ***


-- 

hjl dot tools at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |sebastian dot bw at freenet
                   |                            |dot de


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug target/36450] [4.4 Regression] ICE in insert_restore/insert_save with GET_MODE_ALIGNMENT mem

[hjl dot tools at gmail dot com]

------- Comment #12 from hjl dot tools at gmail dot com  2008-07-22 13:44 -------
*** Bug 36896 has been marked as a duplicate of this bug. ***


-- 

hjl dot tools at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |linuxl4 at sohu dot com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug target/36450] [4.4 Regression] ICE in insert_restore/insert_save with GET_MODE_ALIGNMENT mem

[hjl dot tools at gmail dot com]

------- Comment #13 from hjl dot tools at gmail dot com  2008-07-30 21:53 -------
Do you still see it after revision 38335? If you really want, you
can check MAX_SUPPORTED_STACK_ALIGNMENT, but not PREFERRED_STACK_BOUNDARY.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug middle-end/36450] [4.4 Regression] ICE in insert_restore/insert_save with GET_MODE_ALIGNMENT mem

[hjl dot tools at gmail dot com]

------- Comment #14 from hjl dot tools at gmail dot com  2008-07-30 21:58 -------
A patch is posted at

http://gcc.gnu.org/ml/gcc-patches/2008-07/msg02378.html


-- 

hjl dot tools at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                URL|                            |http://gcc.gnu.org/ml/gcc-
                   |                            |patches/2008-
                   |                            |07/msg02378.html
          Component|target                      |middle-end


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug middle-end/36450] [4.4 Regression] ICE in insert_restore/insert_save with GET_MODE_ALIGNMENT mem

[vmakarov at gcc dot gnu dot org]

------- Comment #15 from vmakarov at gcc dot gnu dot org  2008-08-06 20:46 -------
Subject: Bug 36450

Author: vmakarov
Date: Wed Aug  6 20:44:53 2008
New Revision: 138826

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=138826
Log:
2008-08-04  H.J. Lu  <hongjiu.lu@intel.com>

        PR middle-end/36450
        * caller-save.c (insert_restore): Replace
        PREFERRED_STACK_BOUNDARY with MAX_SUPPORTED_STACK_ALIGNMENT.
        (insert_save): Likewise.


Modified:
    branches/ira/gcc/ChangeLog
    branches/ira/gcc/caller-save.c


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450

[Bug middle-end/36450] [4.4 Regression] ICE in insert_restore/insert_save with GET_MODE_ALIGNMENT mem

[hjl dot tools at gmail dot com]

------- Comment #16 from hjl dot tools at gmail dot com  2008-08-06 20:50 -------
Should be fixed now.


-- 

hjl dot tools at gmail dot com changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36450