[Bug middle-end/62018] FAIL: gcc.dg/torture/ftrapv-1.c * execution test on x86_64-apple-darwin13
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62018
Francois-Xavier Coudert changed:
What|Removed |Added
Status|WAITING |NEW
CC||fxcoudert at gcc dot gnu.org
--- Comment #7 from Francois-Xavier Coudert ---
Here's a minimal reproducer:
int iaddv (int a, int b) { return a + b; }
int main (void)
{ volatile int x = iaddv (__INT_MAX__, 1); }
Compiled with gcc trunk (or 4.8 or 4.9) with option -ftrapv, it executes
without fault. If I do the same on a linux box, it traps (as it should).
However, on both the tree dumps are the same: i.e. I cannot see, from the dump
tree, where the trap checks and instructions should be. Richard, I am willing
to debug this, but don't know where to go after the above analysis.
[Bug middle-end/62018] FAIL: gcc.dg/torture/ftrapv-1.c * execution test on x86_64-apple-darwin13
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62018
--- Comment #8 from Francois-Xavier Coudert ---
(In reply to Francois-Xavier Coudert from comment #7)
> Richard, I am willing to debug this, but don't know where to go after the
> above analysis.
I think it's a libgcc miscompilation somehow. Directly calling __addvsi3 as
follows:
int __addvsi3 (int, int);
int main (void)
{ volatile int x = __addvsi3 (__INT_MAX__, 1); }
does not abort, even though (in my understanding) it should.
[Bug middle-end/62018] FAIL: gcc.dg/torture/ftrapv-1.c * execution test on x86_64-apple-darwin13
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62018 Richard Biener changed: What|Removed |Added Status|UNCONFIRMED |WAITING Last reconfirmed||2014-08-05 Ever confirmed|0 |1 --- Comment #1 from Richard Biener --- Can you please investigate a bit yourself? The test relies on fork()/wait() to work and properly return the exit status of the forked process.
[Bug middle-end/62018] FAIL: gcc.dg/torture/ftrapv-1.c * execution test on x86_64-apple-darwin13
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62018 --- Comment #2 from Dominique d'Humieres --- > Can you please investigate a bit yourself? The test relies on fork()/wait() > to work and properly return the exit status of the forked process. Where should I start?
[Bug middle-end/62018] FAIL: gcc.dg/torture/ftrapv-1.c * execution test on x86_64-apple-darwin13
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62018 --- Comment #4 from Dominique d'Humieres --- > With a debugger? Yerk!-(
[Bug middle-end/62018] FAIL: gcc.dg/torture/ftrapv-1.c * execution test on x86_64-apple-darwin13
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62018 --- Comment #3 from rguenther at suse dot de --- On Tue, 5 Aug 2014, dominiq at lps dot ens.fr wrote: > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62018 > > --- Comment #2 from Dominique d'Humieres --- > > Can you please investigate a bit yourself? The test relies on fork()/wait() > > to work and properly return the exit status of the forked process. > > Where should I start? With a debugger?
[Bug middle-end/62018] FAIL: gcc.dg/torture/ftrapv-1.c * execution test on x86_64-apple-darwin13
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62018 --- Comment #5 from Dominique d'Humieres --- If I remove the attribute 'no clone', the test succeeds with clang, but fails with at least gcc 4.8.3, 4.9.1, and trunk.
[Bug middle-end/62018] FAIL: gcc.dg/torture/ftrapv-1.c * execution test on x86_64-apple-darwin13
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62018 --- Comment #6 from Dominique d'Humieres --- Created attachment 33367 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=33367&action=edit Preprocessed and assembly files for gcc and clang Preprocessed and assembly files for gcc (ftrapv-1_gcc.(i|s)) and clang (ftrapv-1_clang.(i|s)).
