public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug middle-end/56964] New: ICE with -fno-sync-libcalls when target lacks atomic operations
@ 2013-04-15 10:45 mans at mansr dot com
  2014-06-17 19:34 ` [Bug middle-end/56964] " rwahl at gmx dot de
                   ` (2 more replies)
  0 siblings, 3 replies; 4+ messages in thread
From: mans at mansr dot com @ 2013-04-15 10:45 UTC (permalink / raw)
  To: gcc-bugs


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56964

             Bug #: 56964
           Summary: ICE with -fno-sync-libcalls when target lacks atomic
                    operations
    Classification: Unclassified
           Product: gcc
           Version: 4.8.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: middle-end
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: mans@mansr.com


Use of __atomic_* functions when compiling for a target without atomic
operations (e.g. ARMv6) while specifying -fno-sync-libcalls gives an ICE
instead of a proper error message.

$ cat foo.c
int *x;
void foo(void)
{
    __atomic_test_and_set(x, 0);
}

$ arm-unknown-linux-gnueabi-gcc-4.8.0 -c foo.c -march=armv6 -fno-sync-libcalls
foo.c: In function 'foo':
foo.c:4:26: internal compiler error: in emit_move_insn, at expr.c:3486
     __atomic_test_and_set(x, 0);
                          ^
Please submit a full bug report,
with preprocessed source if appropriate.

This happens for any target without atomic operations in hardware.


^ permalink raw reply	[flat|nested] 4+ messages in thread

* [Bug middle-end/56964] ICE with -fno-sync-libcalls when target lacks atomic operations
  2013-04-15 10:45 [Bug middle-end/56964] New: ICE with -fno-sync-libcalls when target lacks atomic operations mans at mansr dot com
@ 2014-06-17 19:34 ` rwahl at gmx dot de
  2014-06-17 21:31 ` rwahl at gmx dot de
  2021-11-29  7:32 ` pinskia at gcc dot gnu.org
  2 siblings, 0 replies; 4+ messages in thread
From: rwahl at gmx dot de @ 2014-06-17 19:34 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=56964

rwahl at gmx dot de changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |rwahl at gmx dot de

--- Comment #1 from rwahl at gmx dot de ---
I can reproduce this too when building multilib cross compiler:

cat
gcc-4.8.3/build-arm-pp-linux-uclibcgnueabi/arm-pp-linux-uclibcgnueabi/libatomic/config.log
...
configure:12468: checking for __atomic_test_and_set for size 1
configure:12487:
/home/nb/builds/toolchain-010/tmp_cross/build/gcc-4.8.3/build-arm-pp-linux-uclibcgnueabi/./gcc/xgcc
-B/home/nb/builds/toolchain-010/tmp_cross/build/gcc-4.8.3/build-arm-pp-linux-uclibcgnueabi/./gcc/
-B/home/nb/builds/toolchain-010/tmp_cross/build_root_2//bin/
-B/home/nb/builds/toolchain-010/tmp_cross/build_root_2//lib/ -isystem
/home/nb/builds/toolchain-010/tmp_cross/build_root_2//arm-pp-linux-uclibcgnueabi/include
--sysroot=/home/nb/builds/toolchain-010/tmp_cross/build_root_2//arm-pp-linux-uclibcgnueabi
  -o conftest -Os -fno-sync-libcalls    conftest.c  >&5
conftest.c: In function 'main':
conftest.c:40:27: internal compiler error: Segmentation fault
      __atomic_test_and_set(x, 0);
                           ^
Please submit a full bug report,
...

Under different build circumstances the build of conftest.c even just hangs
which was my original problem I was trying to solve and it took me half a day
to dig around and find that bug report.


^ permalink raw reply	[flat|nested] 4+ messages in thread

* [Bug middle-end/56964] ICE with -fno-sync-libcalls when target lacks atomic operations
  2013-04-15 10:45 [Bug middle-end/56964] New: ICE with -fno-sync-libcalls when target lacks atomic operations mans at mansr dot com
  2014-06-17 19:34 ` [Bug middle-end/56964] " rwahl at gmx dot de
@ 2014-06-17 21:31 ` rwahl at gmx dot de
  2021-11-29  7:32 ` pinskia at gcc dot gnu.org
  2 siblings, 0 replies; 4+ messages in thread
From: rwahl at gmx dot de @ 2014-06-17 21:31 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=56964

--- Comment #2 from rwahl at gmx dot de ---
Just found out that the mentioned hangs in my last comment were caused by
configuring gcc with --disable-checking so the asserts were removed.


^ permalink raw reply	[flat|nested] 4+ messages in thread

* [Bug middle-end/56964] ICE with -fno-sync-libcalls when target lacks atomic operations
  2013-04-15 10:45 [Bug middle-end/56964] New: ICE with -fno-sync-libcalls when target lacks atomic operations mans at mansr dot com
  2014-06-17 19:34 ` [Bug middle-end/56964] " rwahl at gmx dot de
  2014-06-17 21:31 ` rwahl at gmx dot de
@ 2021-11-29  7:32 ` pinskia at gcc dot gnu.org
  2 siblings, 0 replies; 4+ messages in thread
From: pinskia at gcc dot gnu.org @ 2021-11-29  7:32 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=56964

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

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

--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Fixed in GCC 4.9.x and GCC 5+. by the patch which fixed PR 61713. It is an
exact dup really.

*** This bug has been marked as a duplicate of bug 61713 ***

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2021-11-29  7:32 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-04-15 10:45 [Bug middle-end/56964] New: ICE with -fno-sync-libcalls when target lacks atomic operations mans at mansr dot com
2014-06-17 19:34 ` [Bug middle-end/56964] " rwahl at gmx dot de
2014-06-17 21:31 ` rwahl at gmx dot de
2021-11-29  7:32 ` pinskia at gcc dot gnu.org

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).