bug-coreutils
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

bug#10423: problem with "make check" coreutils-8.14 after build on AIX (


From: Michael Felt
Subject: bug#10423: problem with "make check" coreutils-8.14 after build on AIX (v5.3, xlc v7, ksh as default shell)
Date: Mon, 2 Jan 2012 22:43:05 +0100

Maybe showing less :p

export VERBOSE=yes
nohup make check

"head" nohup.out...

address@hidden:[/data/prj/coreutils-8.14/tests]tail -f nohup.out
        make  check-TESTS
FAIL: misc/help-version
FAIL: tail-2/inotify-race
PASS: misc/invalid-opt
FAIL: rm/ext3-perf
FAIL: rm/cycle
FAIL: cp/link-heap
FAIL: tail-2/inotify-hash-abuse
FAIL: tail-2/inotify-hash-abuse2
FAIL: tail-2/F-vs-missing
FAIL: tail-2/F-vs-rename
FAIL: tail-2/inotify-rotate
FAIL: chmod/no-x
FAIL: chgrp/basic
FAIL: rm/dangling-symlink
FAIL: misc/ls-time

Attached: the tests I ran before the first mail...


On Mon, Jan 2, 2012 at 10:33 PM, Jim Meyering <address@hidden> wrote:

> Michael Felt wrote:
>
> > Sorry, lost me there...
> >
> > So, hopefully to improve both our understanding of what is happening...
> >
> > 1. I am using ksh (using bash did not change anything, unfortunately).
> > 2. Running only init.sh returns the following (first not set, then set
> VERBOSE=
> > yes)
>
> Sorry I wasn't clear.
> I was explaining to Paul why VERBOSE=yes was having no effect.
> init.sh is used internally via most tests.  There's no point
> in running it stand-alone.
>
> Here's something you can try, though:
>
> edit init.sh and insert a "#" at the beginning of the set -x line.
> I.e., change this:
>
>    if $gl_set_x_corrupts_stderr_; then
>      warn_ "using SHELL=$SHELL with 'set -x' corrupts stderr"
>    else
>      set -x
>    fi
>
> to this:
>
>    if $gl_set_x_corrupts_stderr_; then
>      warn_ "using SHELL=$SHELL with 'set -x' corrupts stderr"
>    else
>       # set -x
>    fi
>
> Then run "make check VERBOSE=yes" and send us the tests/test-suite.log
> it produces.
>
> That will probably induce additional test failures,
> but at least it will allow us to see verbose output.
>

Attachment: check_nonroot.text
Description: Binary data

Attachment: check_root.text
Description: Binary data


reply via email to

[Prev in Thread] Current Thread [Next in Thread]