Marcel Loose <loose <at> astron.nl> writes:

> 
> sh-3.00$ uname -a
> Linux dop131 2.6.13-15.18-smp #1 SMP Tue Oct 2 17:36:20 UTC 2007 x86_64 
x86_64 x86_64 GNU/Linux

Thanks for the report.

Test 151 plays with /dev/full on systems that have it.  From your log:

./torture.at:289: ./config.status --header=-:input </dev/null >/dev/full
stderr:
awk: ./confx32011/defines.awk:37: (FILENAME=input FNR=1) warning: error writing 
standard output (No space left on device)
stdout:
./torture.at:289: exit code was 0, expected 1

Hmm.  Did we miss a spot where config.status does not properly catch failures 
of subsidiary processes in writing to a full disk?

Test 208 looks like a configuration setup issue.  Maybe it is possible to teach 
the testsuite how to skip this test if it is run on systems like yours where 
the fortran compiler is broken?

> configure:2106: checking whether we are using the GNU Fortran 77 compiler
> configure:2125: g77 -c  conftest.F >&5
> g77: installation problem, cannot exec `cc1': No such file or directory
> configure:2132: $? = 1
> configure: failed program was:
> |       program main
> | #ifndef __GNUC__
> |        choke me
> | #endif
> | 
> |       end
> configure:2149: result: no

-- 
Eric Blake




Reply via email to