Minutes of 6/23/05 meeting

[Jim Keniston <jkenisto at us dot ibm dot com>]

Minutes of SystemTap Conference Call, June 23, 2005

Participants:
IBM:
	Vara Prasad (varap@us.ibm.com -- Beaverton)
	Jim Keniston (jkenisto@us.ibm.com -- Beaverton)
	Hien Nguyen (nguyhien@us.ibm.com -- Beaverton)
	Kevin Stafford (krstaffo@us.ibm.com - Beaverton)
	Tom Zanussi (trz@us.ibm.com -- Chicago)
	Ananth Mavinakayanahalli (amavin@redhat.com, ananth@in.ibm.com --
Boston)
	Prasanna Panchamukhi (prasanna@in.ibm.com -- Bangalore)
	Maneesh Soni (maneesh@in.ibm.com -- Bangalore)
Red Hat:
	Will Cohen (wcohen@redhat.com -- Raleigh)
	Martin Hunt (hunt@redhat.com -- Seattle)
	Graydon Hoare (graydon@redhat.com)
Intel:
	Brad Chen (brad.chen@intel.com -- Santa Clara)
	Rusty Lynch (rusty.lynch@intel.com -- Hillsboro, OR)
	Anil KeshavaMurthy (anil.s.keshavamurthy@intel.com -- Hillsboro)
	Charles Spirakis (charles.spirakis@intel.com)
Hitachi:
	Satoshi Oshima

Actions Required (ARs):
>From 5/19/05:
19. Vara: Collect sample kprobes modules and post them to the
SystemTap web site.

21. Rusty, Will: Look into adding kprobes tests to LDP and the
Red Hat suite, respectively.

22. Rusty: Look into x86_64 limitation re: temporary disarming of
reentrant probes.

23. Jim, Martin: Ensure that the runtime's stack-trace feature works
appropriately when there are return probes in the call stack.
[Update: Martin is working on this.]

>From 6/9/05:
26. Rusty/Brad: Forward email address of Intel test person to Jay
Turner.

>From 6/16/05:
28. All: Review Charles's documents on profiling and timestamps:
src/tapsets/{profile,timestamp} in CVS.

29. Jim: Post *probes performance measurements before and after
Rusty's recent return-probes enhancements.  [DONE 6/24/05]

>From 6/23/05:
30. Rusty: Review Will's rendition of Rusty's test that exercises
probes on a wide variety of instruction types.

31. Anil: Publish description of problem and solution regarding ia64
probes set on slot 1 of a byndle.

32. Martin/Tom: Benchmark relayfs performance vs. netlink in a
kprobes/SystemTap context.

33. Vara: Publish time slot for tracing/SystemTap BOF at OLS.

34. Vara/Satoshi: Discuss possibility of LKST demo at BOF.

35. Prasanna: Contact Andrew M to request that all approved kprobes
patches get pushed to Linus.

36. Jim: Create and submit Documentation/kprobes.txt writeup.
[Update: Doc by Prasanna & Jim posted to SystemTap list 6/28/05]

Elena and Frank are at the gcc summit.

Satoshi introduced himself.  He's a contributor to LKST.

Will reported that Ananth has backported all relevant kprobes patches
to RHEL4 U2.  Will has enhanced the test suite to test these features.
See the tests directory in CVS.  These tests are designed to be
runnable outside the Red Hat test harness as well.  Rusty agreed to
review Will's rendition of Rusty's test that exercises probes on a
wide variety of instruction types.

SystemTap will need to be an RPM; we need to start giving this
serious thought.  For example, will kernel-module code such as the
runtime and tapsets be included in the SystemTap RPM?

Vara asked about prospects for an automatic, periodic build and test
run.  The Red Hat test system requires the software to be packaged
as an RPM.  Alternative build/test systems were recommended: Mozilla
Tinderbox and the gcc system.  Will will investigate these.

Graydon has been working on the translator -- specifically on
mapping probe specifications to dwarf information.  This currently
relies on a dwarf library that needs to be added.  It should be
available soon on Rawhide (rawhide.redhat.com?).

Anil continues to work on atomic probe insertion for ia64 kprobes.
To get around a potential deadlock, register_kprobe() will reject
probes set in slot 1 of an instruction bundle.  This workaround
won't be needed on newer ia64 chips.  Anil will publish a writeup
about this on the SystemTap list.

The need for a netlink-vs.-relayfs performance comparison has arisen
again, in response to Tom's request for Andrew M to push relayfs
to Linus.  Martin will be able to do this soon; Tom will help.

Brad & co. are making progress on performance tapsets.  There's
a dependency on perfmon.

Vara is leading a BOF at OLS on tracing and SystemTap.  LKST
developers will attend; Satoshi offered to provide an LKST demo.
Vara will publish the time slot for his BOF as soon as he knows.
Most of our Intel contingent will attend OLS.

Andrew M has pushed Hien's return-probes patch to Linus.  No word yet
on Rusty's return-probes enhancements or Tom's 4 relayfs patches.
Prasanna will contact Andrew to request that all approved kprobes
patches get pushed to Linus.

We need a writeup on kprobes in the kernel's Documentation directory.
Jim will write this; Prasanna will push it.