This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: outdated dejagnu information on http://gcc.gnu.org/install/test.html
- To: jsm28 at cam dot ac dot uk
- Subject: Re: outdated dejagnu information on http://gcc.gnu.org/install/test.html
- From: Laurent Guerby <guerby at acm dot org>
- Date: Sat, 24 Feb 2001 20:42:29 +0100
- CC: pfeifer at dbai dot tuwien dot ac dot at, kraai at alumni dot carnegiemellon dot edu, gcc at gcc dot gnu dot org
- References: <Pine.LNX.4.32.0102241921030.21363-100000@kern.srcf.societies.cam.ac.uk>
- Reply-to: guerby at acm dot org
Joseph S. Myers:
> You can see from grepping gcc-testresults that these bogus failures occur
> regularly for some people (e.g. build@netwinder.org on i386), sometimes
> for others and not at all for others.
By "grepping gcc-testresults", I assume you mean doing a grep on the
archive of the mailing list.
> If any of the people who have got such bogus failures (towards the
> end of gcc.dg/format/ tests, from output truncation) understand
> dejagnu internals (who here does?), perhaps you could look at the
> problem?
Well, I don't know dejagnu at all, and my results didn't have any
format failure (grep format gcc.sum.sent|grep -v PASS is empty).
A few RPM queries say that I have on my system:
tcl-8.0.5-30
tclx-8.0.5-30
expect-5.28-30
tk-8.0.5-30
tkinter-1.5.2-7
tksysv-1.0-6
dejagnu-1.4-0
I believe all but dejagnu are shipped standard with RH 6.1.
May be the build@netwinder.org people seeing failures could look at
what they have installed? Or may be better have "make check" output
version info for dejagnu and its various components in the standard
report?
--
Laurent Guerby <guerby@acm.org>