This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: Release announcement suggestions for 0.7
Hi Frank,
On Wed, 2008-06-18 at 23:02 -0400, Frank Ch. Eigler wrote:
> Mark Wielaard <mwielaard@redhat.com> writes:
> > There are lists of test results (make installcheck) on various systems I
> > had access to. The 2.6.25.6-55.fc9.i686 results don't look that good
> > though and are from a qemu-kvm instance. On real hardware I have not
> > been able to get a full make installcheck run because the machine will
> > randomly hang. Till now I have been unable to get to the bottom of this
> > issue (and strangely in a virtual environment everything seems fine).
>
> IMO we should consider this mysteriosity as a "blocker" for the
> release. Could you try a 2.6.24 kernel on the same hardware &
> systemtap version?
yes, 2.6.24 on this machine seems to produce test results similar to the
x86_64/2.6.24 results I posted (without the 64-bit failures of course).
> > For a full overview see the stapex(5) manual page.
>
> (I believe that this man page does not contain any information about
> the separate example script files. It probably should.
You are right. It does contain:
SEE ALSO
/usr/local/doc/systemtap*/examples
That should be updated to the location we decide that the examples will
be installed to (assuming they won't stay in
testsuite/systemtap.examples). I'll update it to say:
"For more examples see the stapex(5) manual page".
> > - Stack backtraces for x86 and x86-64 are generated by a dwarf
> > debuginfo-based unwinder [...]
>
> (This part is under maintenance and may be premature to announce
> until e.g. it works for userspace.)
OK, I'll remove it.
> > - Inclusion of Compile server (stap-server) and Compile server client
> > (stap-client).
>
> These parts are brand new prototypes.
OK. Dave, do you want to have them mentioned so people check them out
and give feedback, or is it too early for that?
> > = Know issues with this release:
> > <FIXME>
>
> The hangs on 2.6.25 would be one. :-(
Yeah, trying to get a better look idea now. dsmith pointed me at the
netconsole kernel support (this machine doesn't have a physical serial
port) to see if I can capture any output/oops/kprint with that. I'll try
to get that to work since qemu clearly does something to make the issue
disappear.
Cheers,
Mark