[Bug c/34720] New: ICE in real_to_decimal, at real.c:1656

[Bug c/34720] New: ICE in real_to_decimal, at real.c:1656

[tgall dot foo at gmail dot com]

overview:

building gcc 4.3 with gcc 4.3 fails on a 64 bit ppc64 box. (yes, running as
elf64)

gcc -v
Using built-in specs.
Target: powerpc64-unknown-linux-gnu
Configured with:
/var/tmp/portage/sys-devel/gcc-4.3.0_alpha20080104/work/gcc-4.3-20080104/configure
--prefix=/usr
--bindir=/usr/powerpc64-unknown-linux-gnu/gcc-bin/4.3.0-alpha20080104
--includedir=/usr/lib/gcc/powerpc64-unknown-linux-gnu/4.3.0-alpha20080104/include
--datadir=/usr/share/gcc-data/powerpc64-unknown-linux-gnu/4.3.0-alpha20080104
--mandir=/usr/share/gcc-data/powerpc64-unknown-linux-gnu/4.3.0-alpha20080104/man
--infodir=/usr/share/gcc-data/powerpc64-unknown-linux-gnu/4.3.0-alpha20080104/info
--with-gxx-include-dir=/usr/lib/gcc/powerpc64-unknown-linux-gnu/4.3.0-alpha20080104/include/g++-v4
--host=powerpc64-unknown-linux-gnu --build=powerpc64-unknown-linux-gnu
--disable-altivec --disable-nls --with-system-zlib --disable-checking
--disable-werror --enable-secureplt --disable-libunwind-exceptions
--disable-multilib --disable-libmudflap --disable-libssp --disable-libgcj
--enable-languages=c --enable-shared --enable-threads=posix
--enable-__cxa_atexit --enable-clocale=gnu
Thread model: posix
gcc version 4.3.0-alpha20080104  (experimental) (GCC)


./configure options:
        --prefix=/usr           
        --bindir=/usr/powerpc64-unknown-linux-gnu/gcc-bin/4.3.0-alpha20080104   
       
--includedir=/usr/lib/gcc/powerpc64-unknown-linux-gnu/4.3.0-alpha20080104/include 
       
--datadir=/usr/share/gcc-data/powerpc64-unknown-linux-gnu/4.3.0-alpha20080104   
       
--mandir=/usr/share/gcc-data/powerpc64-unknown-linux-gnu/4.3.0-alpha20080104/man 
       
--infodir=/usr/share/gcc-data/powerpc64-unknown-linux-gnu/4.3.0-alpha20080104/info 
       
--with-gxx-include-dir=/usr/lib/gcc/powerpc64-unknown-linux-gnu/4.3.0-alpha20080104/include/g++-v4 
        --host=powerpc64-unknown-linux-gnu 
        --build=powerpc64-unknown-linux-gnu 
        --disable-altivec 
        --enable-nls 
        --without-included-gettext              
        --with-system-zlib              
        --disable-checking              
        --disable-werror                
        --enable-secureplt              
        --disable-libunwind-exceptions 
        --disable-multilib 
        --disable-libmudflap 
        --disable-libssp 
        --disable-libgcj 
        --enable-languages=c,c++,treelang 
        --enable-shared 
        --enable-threads=posix 
        --enable-__cxa_atexit 
        --enable-clocale=gnu


output + cmd issued that died

sed -e s/TARGET_GETGROUPS_T/gid_t/ \
         
/var/tmp/portage/sys-devel/gcc-4.3.0_alpha20080104/work/gcc-4.3-20080104/gcc/sys-types.h
/var/tmp/portage/sys-devel/gcc-4.3.0_alpha20080104/work/gcc-4.3-20080104/gcc/sys-protos.h
> SYSCALLS.c
/var/tmp/portage/sys-devel/gcc-4.3.0_alpha20080104/work/build/./gcc/xgcc
-B/var/tmp/portage/sys-devel/gcc-4.3.0_alpha20080104/work/build/./gcc/
-B/usr/powerpc64-unknown-linux-gnu/bin/ -B/usr/powerpc64-unknown-linux-gnu/lib/
-isystem /usr/powerpc64-unknown-linux-gnu/include -isystem
/usr/powerpc64-unknown-linux-gnu/sys-include -O2 -g -O2 -pipe   -DIN_GCC    -W
-Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes
-Wold-style-definition -Wno-strict-prototypes -Wno-error -isystem ./include 
-I. -I.
-I/var/tmp/portage/sys-devel/gcc-4.3.0_alpha20080104/work/gcc-4.3-20080104/gcc
-I/var/tmp/portage/sys-devel/gcc-4.3.0_alpha20080104/work/gcc-4.3-20080104/gcc/.
-I/var/tmp/portage/sys-devel/gcc-4.3.0_alpha20080104/work/gcc-4.3-20080104/gcc/../include
-I/var/tmp/portage/sys-devel/gcc-4.3.0_alpha20080104/work/gcc-4.3-20080104/gcc/../libcpp/include

-I/var/tmp/portage/sys-devel/gcc-4.3.0_alpha20080104/work/gcc-4.3-20080104/gcc/../libdecnumber
-I/var/tmp/portage/sys-devel/gcc-4.3.0_alpha20080104/work/gcc-4.3-20080104/gcc/../libdecnumber/dpd
-I../libdecnumber    \
          -aux-info SYSCALLS.c.X -S -o tmp-SYSCALLS.s SYSCALLS.c
<built-in>:0: internal compiler error: in real_to_decimal, at real.c:1656
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.

I'm still digging into this .. I'll attach the -save-temps output in a bit


-- 
           Summary: ICE in real_to_decimal, at real.c:1656
           Product: gcc
           Version: 4.3.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: tgall dot foo at gmail dot com
 GCC build triplet: powerpc64-unknown-linux-gnu
  GCC host triplet: powerpc64-unknown-linux-gnu
GCC target triplet: powerpc64-unknown-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34720

[Bug c/34720] ICE in real_to_decimal, at real.c:1656

[tgall dot foo at gmail dot com]

------- Comment #1 from tgall dot foo at gmail dot com  2008-01-09 05:50 -------
Two things to be clear on. This is a build of the Jan 4 2008 snapshot. No
patches are applied.

SYSCALLS.i
====
# 1 "SYSCALLS.c"
# 1 "/var/tmp/portage/sys-devel/gcc-4.3.0_alpha20080104/work/build/gcc//"
# 1 "<built-in>"

====




-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34720

[Bug c/34720] ICE in real_to_decimal, at real.c:1656

[tgall dot foo at gmail dot com]

------- Comment #2 from tgall dot foo at gmail dot com  2008-01-09 07:06 -------
on a lark I tried turning off decimal float support

no change.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34720

[Bug c/34720] ICE in real_to_decimal, at real.c:1656

[bje at gcc dot gnu dot org]

------- Comment #3 from bje at gcc dot gnu dot org  2008-01-17 05:27 -------
I cannot reproduce this using either r131302 (a revision close to 2008-01-04)
or today's head revision.  Can you please check again using an updated source
tree?

>From the paths listed in your bug report, it appears that you are building from
the Portage package system.  That source tree may be different to the FSF tree
if it has patches applied by the package.

Thanks, Ben


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34720

[Bug c/34720] ICE in real_to_decimal, at real.c:1656

[tgall dot foo at gmail dot com]

------- Comment #4 from tgall dot foo at gmail dot com  2008-01-23 15:28 -------
Hi Ben,

Yesterday I picked this up again using the 0118 snapshot. Same failure. 

Yes this is being built inside of portage. No patches are being applied. (I've
disabled that)

Following the direction of Peter Bergner I've also modified the params to
configure to be more inline with what he says is being used for gcc testing.

These values are:

        --prefix=/usr           
        --bindir=/usr/powerpc64-unknown-linux-gnu/gcc-bin/4.3.0-alpha20080118   
       
--includedir=/usr/lib/gcc/powerpc64-unknown-linux-gnu/4.3.0-alpha20080118/include 
       
--datadir=/usr/share/gcc-data/powerpc64-unknown-linux-gnu/4.3.0-alpha20080118   
       
--mandir=/usr/share/gcc-data/powerpc64-unknown-linux-gnu/4.3.0-alpha20080118/man 
       
--infodir=/usr/share/gcc-data/powerpc64-unknown-linux-gnu/4.3.0-alpha20080118/info 
       
--with-gxx-include-dir=/usr/lib/gcc/powerpc64-unknown-linux-gnu/4.3.0-alpha20080118/include/g++-v4 
        --host=powerpc64-unknown-linux-gnu 
        --build=powerpc64-unknown-linux-gnu 
        --disable-altivec 
        --enable-nls 
        --without-included-gettext              
        --with-system-zlib              
        --disable-checking              
        --disable-werror                
        --enable-secureplt              
        --disable-libunwind-exceptions 
        --disable-multilib 
        --disable-libmudflap 
        --disable-libssp 
        --disable-libgcj 
        --enable-decimal-float=yes 
        --with-cpu=default64 
        --with-long-double-128 
        --disable-alsa 
        --with-gmp=/usr/lib 
        --with-mpfr=/usr/lib 
        --enable-languages=c,c++,treelang 
        --enable-shared 
        --enable-threads=posix 
        --enable-__cxa_atexit 
        --enable-clocale=gnu  

See how much further I get today looking into this.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34720

[Bug c/34720] ICE in real_to_decimal, at real.c:1656

[tgall dot foo at gmail dot com]

------- Comment #5 from tgall dot foo at gmail dot com  2008-01-23 21:31 -------
so the call path involved here is from gcc/c-cppbuiltin.c 
builtin_define_with_hex_fp_value

which is calling real_to_decimal and its in that function we're running into
problems.

The string rep of the real is : 0x0.ffffffp128 aka  __FLT_MAX_10_EXP__

I am able to do a real_to_hexadecimal successfully and get : 0x0.ffp+128

Inside of real_to_decimal once we get down to digit=rtd_divmod(&r, &pten) we're
in obvious trouble as digit comes back with a value of:

7378697629483820646  which in hex comes out to be all 6's or a pattern of
repeating nibbles of "0110". 

This is making my head hurt.




-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34720

[Bug c/34720] ICE in real_to_decimal, at real.c:1656

[bje at gcc dot gnu dot org]

------- Comment #6 from bje at gcc dot gnu dot org  2008-01-24 02:33 -------
Thanks; I will keep investigating.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34720

[Bug c/34720] ICE in real_to_decimal, at real.c:1656

[roger at eyesopen dot com]

------- Comment #7 from roger at eyesopen dot com  2008-01-29 01:12 -------
I'm also seeing this same failure with "make profiledbootstrap" on
x86_64-unknown-linux-gnu.  A "make bootstrap" on the same machine completes and
regression tests fine (14 unexpected failures in gcc).  I suspect that the
miscompilation is either non-deterministic or is caused by an optimization that
only triggers on some targets and/or with additional profile information.

Perhaps we should regression hunt for the change that broke things.  It might
not be anything to do with real.c or decimal floating point.


-- 

roger at eyesopen dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |roger at eyesopen dot com
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
  GCC build triplet|powerpc64-unknown-linux-gnu |*64-unknown-linux-gnu
   GCC host triplet|powerpc64-unknown-linux-gnu |*64-unknown-linux-gnu
 GCC target triplet|powerpc64-unknown-linux-gnu |*64-unknown-linux-gnu
   Last reconfirmed|0000-00-00 00:00:00         |2008-01-29 01:12:50
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34720

[Bug c/34720] ICE in real_to_decimal, at real.c:1656

[tgall dot foo at gmail dot com]

------- Comment #8 from tgall dot foo at gmail dot com  2008-02-05 16:24 -------
this bug might in some ways be related bug an older bug in gcc 4.1 where make
profiledbootstrap on power3 was busted.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28326

When I have been running into this problem it has been on power5. I'm backing
off with a make bootstrap build now.

>From last night's test I KNOW you can't do the following:

1) using a gcc 4.1.2 toolchain
   build gcc 4.3 with make profiledbootstrap
   (this will succeed)
2) using gcc 4.3 tooclain
   build gcc 4.3 with make bootstrap
   this will fail.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34720

[Bug c/34720] ICE in real_to_decimal, at real.c:1656

[ubizjak at gmail dot com]

------- Comment #9 from ubizjak at gmail dot com  2008-02-07 20:52 -------
Related to PR33992 ?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34720

[Bug c/34720] ICE in real_to_decimal, at real.c:1656

[ubizjak at gmail dot com]

------- Comment #10 from ubizjak at gmail dot com  2008-02-15 13:39 -------
Duplicate of PR33992.

*** This bug has been marked as a duplicate of 33992 ***


-- 

ubizjak at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  BugsThisDependsOn|33992                       |
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=34720