linux-kernel - [PATCH] LTTng 0.6.36 for 2.6.18 (now with markers)

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:	Fri, 24 Nov 2006 16:48:31 -0500
From:	Mathieu Desnoyers <mathieu.desnoyers@...ymtl.ca>
To:	linux-kernel@...r.kernel.org,
	Christoph Hellwig <hch@...radead.org>,
	Andrew Morton <akpm@...l.org>, Ingo Molnar <mingo@...hat.com>,
	Greg Kroah-Hartman <gregkh@...e.de>,
	Thomas Gleixner <tglx@...utronix.de>,
	Tom Zanussi <zanussi@...ibm.com>,
	Karim Yaghmour <karim@...rsys.com>,
	Paul Mundt <lethal@...ux-sh.org>, Jes Sorensen <jes@....com>,
	Richard J Moore <richardj_moore@...ibm.com>,
	"Martin J. Bligh" <mbligh@...igh.org>,
	Michel Dagenais <michel.dagenais@...ymtl.ca>,
	Douglas Niehaus <niehaus@...s.ku.edu>, ltt-dev@...fik.org,
	systemtap@...rces.redhat.com
Subject: [PATCH] LTTng 0.6.36 for 2.6.18 (now with markers)

Hi,

I have, since a few weeks, moved LTTng to the markers infrastructure. I
left the ltt-dev users and contributors test the markers and LTTng on various
architectures (i386, x86_64, PowerPC, -ppc, ARM, MIPS) before posting it on
LKML.

The most important new features since the my post :

- Use DebugFS
- Use the "Markers" infrastructure (updated since the last post on LKML).
- Dynamically loadable "probes" that connects to the markers.
- CPU Hotplug support (this piece seemed necessary for the Xen port I am
  currently working on)
- Use of per-CPU atomic operations even on SMP machines (no lock prefix, no
  memory barriers) to update the per-cpu counters. An explicit smp_wmb()
  is used at the one place where the subbuffers are tagged "full" and
  smp_rmb() is used in the buffer consumer just after it reads this counter
  indicating that the subbuffer is full.

I am not submitting the probes themselves, as they can be provided as separate
kernel modules.

Comments and constructive criticism are, as always, welcome.

The patches follow.

Mathieu


OpenPGP public key:              http://krystal.dyndns.org:8080/key/compudj.gpg
Key fingerprint:     8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68 
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/