[00:00] <Prf_Jakob> Is Tim Gardner in here?
[12:13]  * rtg upgrades his laptop to quantal....
[13:20]  * henrix -> (late) lunch
[13:20] <Prf_Jakob> rtg: Hey
[13:21] <rtg> hmm?
[13:21]  * Prf_Jakob Bornecrantz
[13:21] <Prf_Jakob> vmwgfx guy
[13:21] <rtg> does it work for you ?
[13:21] <Prf_Jakob> I haven't test it actually 
[13:22] <Prf_Jakob> But I can do that.
[13:22] <Prf_Jakob> I'm wondering if you will cherry pick those to patch and maybe even the dumb_fb patch?
[13:23] <rtg> Prf_Jakob, the first patch is in Ubuntu-3.5.0-14.15, and is coming in Precise.
[13:23] <Prf_Jakob> Ah ok good
[13:23] <rtg> I'm reluctant to pick the FB patch since it does represent a more radical change
[13:24] <rtg> Prf_Jakob, the bug reporter did indicate that the minimal patch solved his issues
[13:24] <Prf_Jakob> the xf86-video-vmware packages install a modprobe rule that does the same.
[13:25] <rtg> Prf_Jakob, I'm fine with leaving it up to user space to do that
[13:25] <Prf_Jakob> Ok
[13:25] <rtg> Prf_Jakob, the first patch is in Precise Ubuntu-3.2.0-31.50 (likely still in -proposed)
[13:26] <Prf_Jakob> k thanks
[13:52] <hggdh> folks, I would really appreciate a look at bug 1049395, we have had 5 failures on 7 runs on it
[13:52] <ubot2> Launchpad bug 1049395 in linux "ext4: kernel BUG at /build/buildd/linux-2.6.38/fs/buffer.c:2900!" [Undecided,Confirmed] https://launchpad.net/bugs/1049395
[14:00] <bjf> hggdh, i'll look but as i was explaining yesterday, i've seen this occasionally myself and of the 13 commits that have gone into natty in the last two kernel versions, _none_ are against ext4. they are mostly networking or ecryptfs
[14:02] <bjf> hggdh: you could try 2.6.38-15.66 for comparison
[14:02] <doko> ogasawara, can somebody bump the ABI in linux-meta-lowlatency?
[14:02] <doko> quantal
[14:08] <ogasawara> doko: I don't see it needing an ABI bump yet for quantal, ie it's at -13.  am I missing something?  I do see it needs rebasing to -14 though.
[14:10] <doko> yeah, maybe rebasing. I see in NBS:
[14:10] <doko> linux-headers-3.5.0-13
[14:10] <doko>    	linux-headers-3.5.0-13-lowlatency	universe	amd64 i386
[14:12] <ogasawara> rtg: ^^ think you could rebase apw's lowlatency kernel and upload to universe today?  I've gotta bail here in a second.
[14:13] <rtg> ogasawara, can do
[14:13] <ogasawara> rtg: thanks
[14:14]  * ogasawara heads downtown
[14:14] <hggdh> bjf: OK. I will tag it qa-testing-passed, but I still do not like the number of failures
[14:14] <bjf> hggdh: can you test the previous kernel version?
[14:18] <hggdh> bjf: I can hack around it, yes, doing it manually
[14:23] <rtg> bjf, have you upgraded to Quantal on a machine that referenced the X packages in https://launchpad.net/~ubuntu-x-swat/+archive/q-lts-backport ?
[14:24] <hggdh> bjf: I will first re-run it to have a full repeat, then I will retest the current -updates kernel
[14:39] <bjf> rtg, no, all 3 of my test systems are servers and i've not tried that here at home, i could pretty easily though
[14:40] <rtg> bjf, If you've got a VM that would be handy. it seems some of the packages don't get updated. I'm suspecting some version issues.
[16:37] <jbicha> hey, bug 1049650 is annoying
[16:37] <ubot2> Launchpad bug 1049650 in linux "Regression: "Undefined video mode" with 3.5.0-14.16" [Medium,Incomplete] https://launchpad.net/bugs/1049650
[16:42] <rtg> sforshee, did your gmux crack do this ? ^^
[16:42] <sforshee> rtg, I seriously doubt it. It's not even an apple machine.
[16:43] <rtg> sforshee, its about the only think even vaguely related to video
[16:43] <sforshee> rtg, apple-gmux shouldn't load unless the right hid appears in the ACPI tables, and it should only be there on apple machines
[16:44] <rtg> sforshee, right
[16:46] <jstultz> Curious if anyone in Ubuntu is looking at the pstore infrastructure (using it to capture oops logs, etc)?
[16:47] <rtg> jstultz, I think cking had some interest in it
[16:47] <jstultz> rtg:  cool
[16:48] <cking> well, my UEFI boxes with Quantal do save kernel log messages to the UEFI variable space, so that's a debugging win for me
[16:48] <cking> not exercised pstore yet though
[16:49] <jstultz> cking:  cool :)  part of why I ask, is that Anton Vorontsov at linaro has been working on getting some of the android features merged into pstore
[16:49] <jstultz> cking:  so for instance on arm, it can use persistent-ram (doesn't clear on warm reset) instead of bios provided flash to store dmesg logs..
[16:50] <jstultz> cking:  also there's been recent work to get ftrace output captured by pstore
[16:50] <cking> jstultz, that's a nice debugging feature. could have done with that several years ago working on ARM devices.
[16:51] <jstultz> cking:  but i just wanted to make sure these features were promoted enough so folks who might be able to integrate them know they're there.
[16:52] <cking> jstultz, when are these changes going to land in mainline?
[16:52] <jstultz> cking:  3.6
[16:53] <jstultz> cking:  they should all be there already
[16:53] <jstultz> (the ftrace bits may still be pending for 3.7..  )
[16:53] <cking> ok, so Quantal+1 now
[16:53] <jstultz> cking:  yep... 
[16:54] <cking> jstultz, are these just ARM only features? the ftrace in pstore sounds like a cool idea, but probably uses up lots of memory
[16:54] <jstultz> cking:  the hope is that pstore will abstract some of the uefi vs arm differences
[16:54] <jstultz> cking:  not necessarily..  i think you can size the memory space..  
[16:55] <jstultz> cking:  it just helps to get exactly what functions were called right before an oops
[16:55] <cking> jstultz, indeed, that would be a big win 
[16:55] <jstultz> cking:  the arm only aspect of the persistent-ram part of pstore is really only hardware based..
[16:56] <jstultz> most x86 hardware clears ram state over warm reset... 
[16:56] <cking> ok
[16:56] <jstultz> if an architecture doesn't clear ram, then it should be usable.
[16:56] <cking> indeed, pity about x86 PCs clearing memory really.
[16:57] <jstultz> and we can still try to log to ram regardless, but we just won't find anything on reset
[16:57] <jstultz> (although that could be wasteful.. but atleast provides a way to avoid configuration twiddling)
[16:57]  * cking adds it to his list of things to examine and test on suitable kit
[16:58] <jstultz> cking:  anyway, just wanted to raise awareness... thanks for listening! :)
[16:58] <jstultz> cking:  feel free to ping me or cbou on #linaro-kernel if you have further questions
[16:58] <cking> jstultz, I appreciate you letting me know, much appreciated :-)
[16:58] <cking> ack
[17:35]  * rtg -> lunch
[17:57]  * henrix -> EOD
[19:17] <rtg> arges, flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid amd_dcm aperfmperf pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 popcnt aes xsave avx f16c lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt lwp fma4 tce nodeid
[19:17] <rtg> _msr tbm topoext perfctr_core arat cpb hw_pstate npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold bmi1
[19:17] <arges> rtg, yup it has avx and fma4
[19:19] <arges> rtg, https://bugs.launchpad.net/ubuntu/+source/eglibc/+bug/956051
[19:19] <ubot2> Launchpad bug 956051 in eglibc "libc6 crash while running 'xm'" [High,Confirmed]
[20:23]  * rtg -> EOD