This is the mail archive of the
systemtap@sourceware.org
mailing list for the systemtap project.
Re: Summary of nightly tests 20060208
- From: fche at redhat dot com (Frank Ch. Eigler)
- To: systemtap at sources dot redhat dot com
- Date: 08 Feb 2006 11:29:33 -0500
- Subject: Re: Summary of nightly tests 20060208
- References: <43EA0E13.2090701@redhat.com>
wcohen wrote:
> [...]
> There are some new failures when compared to the Feb 3, 2006 tests:
> FC5 i686, buildok-syscall.stp
> FC5 x86_64 buildok-syscall.stp
> [...]
> RHEL4-U3 x86_64: buildok-syscall.stp
I added "-u" to the buildok/syscall.stp test case, the effect of which
is to force each probe alias, function, and $target variable to be
sent right through to the compilation stage.
This was a shortcut. I know some $target variables don't resolve
sometimes, and the test should still probably tolerate this. My main
point with "-u" was to force all the auxiliary and embedded-C
functions to be compiled. (After all, that is specifically mentioned
in src/HACKING as a testing desideratum.) If the syscall.stp script
was expanded with just enough code to invoke said functions with
optimization (without -u), then -u could be taken back out.
- FChE