This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: Fortran, committed: segfault with allocate and stat for derived types with default initialization (pr 68078)
- From: Christophe Lyon <christophe dot lyon at linaro dot org>
- To: Louis Krupp <louis dot krupp at zoho dot com>
- Cc: Dominique d'Humières <dominiq at lps dot ens dot fr>, "fortran at gcc dot gnu dot org" <fortran at gcc dot gnu dot org>, gcc-patches <gcc-patches at gcc dot gnu dot org>
- Date: Mon, 19 Sep 2016 11:35:21 +0200
- Subject: Re: Fortran, committed: segfault with allocate and stat for derived types with default initialization (pr 68078)
- Authentication-results: sourceware.org; auth=none
- References: <3313458C-C159-4E73-846A-04EBD216B6D5@lps.ens.fr> <1573ebd5ded.b19fdc3849927.1415137124436549445@zoho.com>
Hi,
On 18 September 2016 at 21:19, Louis Krupp <louis.krupp@zoho.com> wrote:
> Two possibilities:
>
> 1. malloc() doesn't silently return NULL on Darwin when it runs out of memory; it always generates an error message.
>
> 2. setrlimit() doesn't work the same on Darwin as it does on Linux, and the test program is hitting a system limit. It so happens that when I first looked at this bug, I ignored the "ulimit -v 1000000" instruction (I have a bad habit of not paying attention), and I let the program do memory allocation forever. It eventually got a segfault deep within malloc(). I was able to reproduce this in C as well as in Fortran. Perhaps something similar happens with Darwin, although at least on Darwin malloc() issues an error, however mysterious, instead of crashing.
>
> Two ways to go from here:
>
> 1. If anyone can let me ssh into a Darwin system, I can see if I can get this to work.
>
> 2. If that's not possible, add a dg-skip-if so that we don't try to run this test on Darwin, since it requires some OS support, and it doesn't just involve Fortran.
>
> Louis
On my side, I've noticed that this new test fails on arm*linux and
aarch64*linux, using QEMU:
set_vm_limit: Operation not permitted
Christophe
>
> ---- On Sun, 18 Sep 2016 03:45:38 -0700 Dominique d'Humières<dominiq@lps.ens.fr> wrote ----
> > > Fixed in revision 240219.
> >
> > The test fails on x86_64-apple-darwin15 (at least):
> >
> > FAIL: gfortran.dg/pr68078.f90 -O0 output pattern test
> > FAIL: gfortran.dg/pr68078.f90 -O1 output pattern test
> > FAIL: gfortran.dg/pr68078.f90 -O2 output pattern test
> > FAIL: gfortran.dg/pr68078.f90 -O3 -fomit-frame-pointer -funroll-loops -fpeel-loops -ftracer -finline-functions output pattern test
> > FAIL: gfortran.dg/pr68078.f90 -O3 -g output pattern test
> > FAIL: gfortran.dg/pr68078.f90 -Os output pattern test
> > FAIL: gfortran.dg/pr68078.f90 -g -flto output pattern test
> >
> > The failures are
> >
> > Output was:
> > pr68078.exe(37410,0xa381f000) malloc: *** mach_vm_map(size=8388608) failed (error code=3)
> > *** error: can't allocate region securely
> > *** set a breakpoint in malloc_error_break to debug
> > foo_ptr allocation failed
> > pr68078.exe(37410,0xa381f000) malloc: *** mach_vm_map(size=8388608) failed (error code=3)
> > *** error: can't allocate region securely
> > *** set a breakpoint in malloc_error_break to debug
> > foo_obj allocation failed
> >
> > Should match:
> > *foo_ptr allocation failed(
> > |^M
> > |^M) *foo_obj allocation failed(
> > |^M
> > |^M) *foo_array allocation failed(
> > |^M
> > |^M)
> >
> > with -O0
> >
> > and
> >
> > Output was:
> > pr68078.exe(39345,0xa381f000) malloc: *** mach_vm_map(size=8388608) failed (error code=3)
> > *** error: can't allocate region securely
> > *** set a breakpoint in malloc_error_break to debug
> > foo_ptr allocation failed
> >
> > Should match:
> > *foo_ptr allocation failed(
> > |^M
> > |^M) *foo_obj allocation failed(
> > |^M
> > |^M) *foo_array allocation failed(
> > |^M
> > |^M)
> >
> > for the other options.
> >
> > Dominique
> >
> >
> >
>
>
>