mmpong version 0.9.1-1 failed to build on amd64 with GCC-4.6/oneiric

[Matthias Klose]

mmpong version 0.9.1-1 failed to build on amd64 with GCC-4.6/oneiric
Link to failed build: https://launchpad.net/ubuntu/+archive/test-rebuild-20110413/+buildjob/2462263

Details about the rebuild:
https://lists.ubuntu.com/archives/ubuntu-devel/2011-April/033042.html

Direct link to the build log: https://launchpad.net/ubuntu/+archive/test-rebuild-20110413/+buildjob/2462263/+files/buildlog_ubuntu-natty-amd64.mmpong_0.9.1-1_FAILEDTOBUILD.txt.gz

This log snippet might be of interest, since it triggered the matcher 'Purging chroot-autobuild'.
Excerpt 1121 lines into the build log:

/build/buildd/mmpong-0.9.1/lib/message.c:65:2: error: (near initialization for 'byte_order[9].offset')
/build/buildd/mmpong-0.9.1/lib/message.c:67:2: error: initializer element is not computable at load time
/build/buildd/mmpong-0.9.1/lib/message.c:67:2: error: (near initialization for 'byte_order[11].offset')
/build/buildd/mmpong-0.9.1/lib/message.c:90:2: error: initializer element is not computable at load time
/build/buildd/mmpong-0.9.1/lib/message.c:90:2: error: (near initialization for 'byte_order[32].offset')
/build/buildd/mmpong-0.9.1/lib/message.c:92:2: error: initializer element is not computable at load time
/build/buildd/mmpong-0.9.1/lib/message.c:92:2: error: (near initialization for 'byte_order[34].offset')
make[3]: *** [lib/CMakeFiles/libmmpong.dir/message.o] Error 1
make[3]: Leaving directory `/build/buildd/mmpong-0.9.1/obj-x86_64-linux-gnu'
make[2]: *** [lib/CMakeFiles/libmmpong.dir/all] Error 2
make[2]: Leaving directory `/build/buildd/mmpong-0.9.1/obj-x86_64-linux-gnu'
make[1]: *** [all] Error 2
make[1]: Leaving directory `/build/buildd/mmpong-0.9.1/obj-x86_64-linux-gnu'
make: *** [debian/stamp-makefile-build] Error 2
dpkg-buildpackage: error: debian/rules build gave error exit status 2
******************************************************************************
Build finished at 20110424-1640
FAILED [dpkg-buildpackage died]
Purging chroot-autobuild/build/buildd/mmpong-0.9.1

[Matthias Klose (doko)]

still ftbfs in Debian
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=642144

[Doko-v (doko-v)]

Created attachment 25379
preprocessed source

[forwarded from http://bugs.debian.org/642144]

this may be a duplicate of PR38354, compiles on 32bit targets, on 64bit targets it succeeds with 4.4, but fails with 4.5, 4.6 and trunk:

$ gcc -c message.i
message.i:154:2: error: initializer element is not computable at load time
message.i:154:2: error: (near initialization for 'byte_order[1].offset')

[Launchpad Janitor (janitor)]

This bug was fixed in the package mmpong - 0.9.1-1ubuntu1

---------------
mmpong (0.9.1-1ubuntu1) oneiric; urgency=low

  * Build using g++-4.4 on amd64. LP: #771087.
 -- Matthias Klose <email address hidden> Thu, 29 Sep 2011 12:24:31 +0200

[Pinskia (pinskia)]

Anyreason Why "((int)( ((void *)&(nmsgbuf.payload.part.ball.pos[0])) - ((void *)&(nmsgbuf)) ))" is being used instead of offsetof ?

[Mikpe (mikpe)]

This test case started being rejected with r145254:
http://gcc.gnu.org/ml/gcc-cvs/2009-03/msg00761.html

[Mikpe (mikpe)]

Created attachment 25393
reduced test case

> cat pr50565.c
/* pr50565.c */
struct netmessage {
    unsigned char pos[2];
};
static struct netmessage nmsgbuf;
const int offset0 = (int)((void*)&nmsgbuf.pos[0] - (void*)&nmsgbuf) + 0*(1U);
const int offset1 = (int)((void*)&nmsgbuf.pos[0] - (void*)&nmsgbuf) + 1*(1U);
> objdir/gcc/xgcc -Bobjdir/gcc/ -O2 -S pr50565.c
pr50565.c:7: error: initializer element is not computable at load time

That's the offset1 initializer it's complaining about. Note that the offset0 initializer is accepted. If we comment out the offset1 definition we get:

> perl -pi -e 's,^const int offset1,//const int offset1,g' pr50565.c
> objdir/gcc/xgcc -Bobjdir/gcc/ -O2 -S pr50565.c
> cat pr50565.s
        .file "pr50565.c"
.globl offset0
        .section .rodata
        .align 4
        .type offset0, @object
        .size offset0, 4
offset0:
        .long 0
        .local nmsgbuf
        .comm nmsgbuf,2,1
        .ident "GCC: (GNU) 4.5.0 20090329 (experimental)"
        .section .note.GNU-stack,"",@progbits

The common offsetof emulation is clearly computable at compile-time, so one would expect the same + 1 to also be computable at compile-time, but gcc doesn't seem to think so.

[Jsm28 (jsm28)]

Patch (to convert.c) submitted: http://gcc.gnu.org/ml/gcc-patches/2011-10/msg00894.html

[Jsm28 (jsm28)]

Author: jsm28
Date: Wed Oct 12 11:56:03 2011
New Revision: 179845

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=179845
Log:
 PR c/50565
 * convert.c (convert_to_integer): Do not narrow operands of
 pointer subtraction.

testsuite:
 * gcc.c-torture/compile/pr50565-1.c,
 gcc.c-torture/compile/pr50565-2.c: New tests.

Added:
    trunk/gcc/testsuite/gcc.c-torture/compile/pr50565-1.c
    trunk/gcc/testsuite/gcc.c-torture/compile/pr50565-2.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/convert.c
    trunk/gcc/testsuite/ChangeLog

[Jsm28 (jsm28)]

Author: jsm28
Date: Wed Oct 12 11:57:36 2011
New Revision: 179846

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=179846
Log:
 PR c/50565
 * convert.c (convert_to_integer): Do not narrow operands of
 pointer subtraction.

testsuite:
 * gcc.c-torture/compile/pr50565-1.c,
 gcc.c-torture/compile/pr50565-2.c: New tests.

Added:
    branches/gcc-4_6-branch/gcc/testsuite/gcc.c-torture/compile/pr50565-1.c
    branches/gcc-4_6-branch/gcc/testsuite/gcc.c-torture/compile/pr50565-2.c
Modified:
    branches/gcc-4_6-branch/gcc/ChangeLog
    branches/gcc-4_6-branch/gcc/convert.c
    branches/gcc-4_6-branch/gcc/testsuite/ChangeLog

[Jsm28 (jsm28)]

Author: jsm28
Date: Wed Oct 12 11:58:47 2011
New Revision: 179847

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=179847
Log:
 PR c/50565
 * convert.c (convert_to_integer): Do not narrow operands of
 pointer subtraction.

testsuite:
 * gcc.c-torture/compile/pr50565-1.c,
 gcc.c-torture/compile/pr50565-2.c: New tests.

Added:
    branches/gcc-4_5-branch/gcc/testsuite/gcc.c-torture/compile/pr50565-1.c
    branches/gcc-4_5-branch/gcc/testsuite/gcc.c-torture/compile/pr50565-2.c
Modified:
    branches/gcc-4_5-branch/gcc/ChangeLog
    branches/gcc-4_5-branch/gcc/convert.c
    branches/gcc-4_5-branch/gcc/testsuite/ChangeLog

[Jsm28 (jsm28)]

Fixed for 4.5.4, 4.6.2, 4.7.

[Matthias Klose (doko)]

fixed in precise