This is the mail archive of the
mailing list for the GCC project.
Re: [RFA:] PR35293 and testsuite: don't run tests using truncation on targets without support
- From: Hans-Peter Nilsson <hans-peter dot nilsson at axis dot com>
- To: fxcoudert at gmail dot com
- Cc: hans-peter dot nilsson at axis dot com, gcc-patches at gcc dot gnu dot org, fortran at gcc dot gnu dot org, janis187 at us dot ibm dot com
- Date: Wed, 5 Mar 2008 02:04:55 +0100
- Subject: Re: [RFA:] PR35293 and testsuite: don't run tests using truncation on targets without support
Thanks for the review!
> From: FX Coudert <email@example.com>
> Date: Wed, 5 Mar 2008 00:09:24 +0000
> Indeed, approval by a testsuite maintainer is needed.
I got it (thanks Janis) and as the testsuite changes are useful
on their own (i.e. without a "noisy truncation failure"), I've
> > +# of unexpected failures 57
> Great! Some of them seem to be related to REWIND, is that a problem
> with the lack of lseek()?
Uh, I dunno, I haven't looked closely. There is lseek support
(if it had failed, I wouldn't have seen the "noisy error"
message ;) so it's not *that* trivial. But maybe there are bugs
in that support. FWIW, some are calls to abort:
program stopped with signal 6.
Fortran runtime error: End of file
and some are link errors for stuff required but not present:
chmod_1.f90:(.text+0x174): undefined reference to `__gfortran_chmod_i4_sub'
chmod_1.f90:(.text+0x192): undefined reference to `__gfortran_access_func'
chmod_1.f90:(.text+0x1e0): undefined reference to `__gfortran_getuid'
(etc.) No whole-sale cleanup at this time.