This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
Re: [rfa/testsuite/mi] More timeout fixes
- To: Andrew Cagney <ac131313 at cygnus dot com>
- Subject: Re: [rfa/testsuite/mi] More timeout fixes
- From: Daniel Jacobowitz <drow at mvista dot com>
- Date: Thu, 1 Nov 2001 09:45:28 -0500
- Cc: gdb-patches at sources dot redhat dot com
- References: <20011027020350.A3935@nevyn.them.org> <3BE0C1F4.4050403@cygnus.com>
On Wed, Oct 31, 2001 at 10:31:00PM -0500, Andrew Cagney wrote:
> >ARM/Linux has an entirely different set of failures than my other test
> >targets did, and found a whole new set of timeout bugs in the MI testsuite.
> >
> >Rather than add the fearsome number of regular expressions, I added some
> >utility functions to lib/mi-support.exp for stepping, continuing, etc.
> >I've
> >tested these on both ARM/Linux (which fails many of them) and i386/Linux
> >(which passes them all), so I'm pretty confident in them; none the less, I
> >wouldn't mind if a TCL guru cared to look them over.
>
> How many failures are you seeing and of what form? I see 22 on a native
> and rougly similar on simulator. SID gets nearer 60 but still runs
> smoothly.
The bulk of them seem to be bad-debug-info related; I'm currently
testing with GCC 2.95.3 and will move up to 3.x soonish (next time I
have time to run tests, in fact).
# of expected passes 7362
# of unexpected failures 300
# of unexpected successes 34
# of expected failures 186
# of unresolved testcases 58
# of untested testcases 6
--
Daniel Jacobowitz Carnegie Mellon University
MontaVista Software Debian GNU/Linux Developer