This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
Re: [PATCH] Fix broken interp.exp testcase
- From: "M.M. Kettenis" <m dot m dot kettenis at alumnus dot utwente dot nl>
- To: Nick Roberts <nickrob at snap dot net dot nz>, Mark Kettenis <kettenis at gnu dot org>, gdb-patches at sources dot redhat dot com
- Date: Mon, 07 Feb 2005 14:40:01 +0000
- Subject: Re: [PATCH] Fix broken interp.exp testcase
Nick Roberts <nickrob@snap.net.nz> wrote:
>
> > The new testcase is failing for me:
> >
> > Running ../../../src/gdb/testsuite/gdb.base/interp.exp ...
> > PASS: gdb.base/interp.exp: interpreter-exec mi "-var-update *"
> > FAIL: gdb.base/interp.exp: interpreter-exec console "show version"
> >
> > I committed the attached patch to fix it.
>
> Well it was passing for me, as it does with your patch. I must have something
> wrong with my setup.
Well, the behaviour of expect can be really odd, especially when doing
regexp matches. Nothing to really worry about; just keep it in mind.
> Running the whole testsuite, I get:
>
> === gdb Summary ===
>
> # of expected passes 10077
> # of unexpected failures 361
> # of unexpected successes 1
> # of expected failures 46
> # of known failures 60
> # of unsupported tests 11
> /home/nick/src/gdb/testsuite/../../gdb/gdb version 6.3.50.20050203-cvs -nx
>
>
> which seems an unacceptable number of fails and makes running the testsuite
> rather pointless. Have you any ideas of what I might be doing wrong?
What system is this? It seems unusually high if you're running gdb on
an up-to-date i386 GNU/Linux system, but on other systems (including
older i386 GNU/Linux) the number of failures can be fairly high because
of:
* Bugs in gdb.
* Bugs in the operating system kernel & (thread) libraries.
* Bugs in the compiler you're using.
Mark