This is the mail archive of the systemtap@sourceware.org mailing list for the systemtap project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] itrace spin lock fix


Dave Nomura wrote:
> Back in August Frank reported a problem running the itrace test on an
> x86-64 machine.  This was caused by an uninitialized spin_lock and
> possibly by some utrace callouts that didn't need in the spin_lock
> protected regions.  This patch fixes that problem.

Your patch's ChangeLog is wrong.  You should put the entry in
src/runtime/ChangeLog, not src/ChangeLog.

I ran this patch on a RHEL5 x86_64 (2.6.18-92.1.13.el5debug) machine.
Systemtap couldn't compile modules without the following small patch:

diff --git a/tapsets.cxx b/tapsets.cxx

index bed2796..057a554 100644

--- a/tapsets.cxx

+++ b/tapsets.cxx

@@ -5778,6 +5778,7 @@ itrace_derived_probe_group::emit_module_decls
(systemtap_\
session& s)



   s.op->newline();

   s.op->newline() << "/* ---- itrace probes ---- */";

+  s.op->newline() << "#include \"task_finder.c\"";

   s.op->newline() << "struct stap_itrace_probe {";

   s.op->indent(1);

   s.op->newline() << "struct stap_task_finder_target tgt;";

With the above patch, systemtap could compile modules again.  The
itrace.exp test script ran (with the early exit deleted), but failed
because the itrace probes never got hit.

-- 
David Smith
dsmith@redhat.com
Red Hat
http://www.redhat.com
256.217.0141 (direct)
256.837.0057 (fax)


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]