Hi Everyone, On Wed, 2008-05-14 at 16:56 -0700, Dan Cauchy wrote: > Linux Foundation CGL Conf Call > > May 14, 2008 [...] > o Joe to send out latest consolidated comments and updates from > the Collab Summit to the mailing list I debated quite a bit on what was the best way to present the notes I collected together from everyone after the Austin meeting, specifically regarding the SCOPE gaps document and eventually decided to scribble directly on the PDF. I hope this works for everyone. Let me know if something isn't clear or if it ends up looking like I just grabbed something out of /dev/urandom. The other part of the notes come directly from Dan's notes in the open session (that'd be Day 2 of the collab summit, as opposed to the working session, which was on Day 3). I hope I haven't mangled his notes too much: CGL Gaps * AVL.3.2 ?? force unmount * need to split into 3 ?whys? * send to LKML with the details on the ?why? * this has been discussed before so it should be doable * AVL.4.2 ?? replaceable OOM killer * configurable OOM killer probably will meet the requirement, this req came years ago when there was no OOM around * AVL.7.1.2 multi-path access to storage * Multi path is there * AVL.14.0 ? excessive CPU cycle usage detection * Arjan Comment: I don?t like the term ?CPU cycle? * This is really ?CPU usage time?, not cycle * This needs to be reworded * AVL.15.0 fast application restart * Arjan: I?m concerned about this one * AVL.28.0? ? Support of Mlocked Page Limits * Arjan: this is not big deal, only concern is that it?s a global, implementation will be tricky in scalability, but it makes total sense * SMM.7.8 ?? Support for User Locked Page Reporting * Arjan: I think we have something for this * SFA.5.0 ? live app dump trigger * Arjan: I think that exists, issue with the term ?live?, should be ?non fatal? * SFA.6.0 ? system snapshot * Corbet: this is there if you trust hibernation code * SFA.9.0 ? kernel sampling for profiling * Arjan: this is there, see oProfile for #3 * SFA.14.0 ? Per-Thread CPU Time Limits and Signalling * Arjan: I don?t think we have this, we need to check * SFA.13.0 ? live application dump * Arjan: this is live snapshot, can we specify a limit, need an indicator/guideline to how long that is (e.g. is it milliseconds?) * PRF.1.2 ?? Configurable Scheduler Quantum For Round Robin Scheduling Policy * Arjan: no more quantum, this is old, doesn?t make sense any more * Remove * PRF1.11 ? priority inheritance inside API * Arjan: this is there, I think they mean processes * This needs to be put in the deprecated pile and see if anyone screams * PRF.4.1 - Forwarding tables fast access * There is now an O(log(n)) algorithm in 2.6.25, this is done * Move to satisfied * PRF.4.4? ? Low Software Overhead For Message Latency * Arjan: I think this is there * PRF.11.1 ? app pre loading non-boot * Arjan: what do you mean by pre-loading, cause prelink exists today * Satisfied * STD.11.2- Diameter * OpenDiameter and OpenRadius exist and should meet this * Verify and move to satisfied Joe MacDonald, Member of Technical Staff, Wind River direct 613.270.5750 mobile 613.291.7421 fax 613.592.2283 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://lists.linux-foundation.org/pipermail/lf_carrier/attachments/20080520/d345baae/attachment-0001.htm -------------- next part -------------- A non-text attachment was scrubbed... Name: SCOPE_CGOS_GAPS_PROFILE_v2-CGL-wg-notes-from-Austin.pdf Type: application/pdf Size: 215324 bytes Desc: not available Url : http://lists.linux-foundation.org/pipermail/lf_carrier/attachments/20080520/d345baae/attachment-0001.pdf -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: This is a digitally signed message part Url : http://lists.linux-foundation.org/pipermail/lf_carrier/attachments/20080520/d345baae/attachment-0001.pgp