[00:15] infinity, too early to tell, am keeping an eye on it [00:15] infinity, at least it's just quantal [03:06] is git://kernel.ubuntu.com/ubuntu/ubuntu-raring.git meant to be usable right now? [03:06] (i assume not, there is no debian/ or ubuntu/, only a debian.master) [04:51] any SSD experts care to chime in? http://askubuntu.com/questions/206088/is-ssd-trim-support-still-automatic-in-12-10 [05:21] jcastro_, trim isn't a queuing command in sata, which means everything else stops until it completes [05:22] jcastro_, which makes it a pessimization on most ssd's made in the last 5 years [05:22] next, trim is just a hint to the device; it's not required to actually do anything in response === smb` is now known as smb [09:05] apw: hd on taipei.tyler is broken for 4 days, I have no chance to meet rtg, could you help to reboot the machine or try to see if there is any way to fix it, thanks. [09:08] AceLan, i can try, i thought that ike had access to that one [09:08] apw: ike is not in office this week [09:28] brb [09:32] AceLan, it does look broken, is there anything on /home or is it just scratch space like gomeisa et al [09:37] apw: acelan@tyler:~$ cat ~/.bashrc [09:37] cat: /home/acelan/.bashrc: Input/output error [09:37] apw: and there are many sdb error log in dmesg [09:37] apw: sorry, i have to leave now to pick up my son and bring him home [09:38] AceLan, np, i'll talk to is [09:38] apw: thanks [10:27] * ppisati -> out for dinner [11:20] ogasawara, fyi i have seeded the linux-raring-meta.git and linux-raring-signed.git [11:43] apw: what's the diff between drm-intel-nightly and drm-intel-experimental mainline builds? [11:43] exper [11:43] tjaalton, experimental is now dead and replaced by nightly ... something bryce asked for [11:44] ok, they both have a build for raring from 23rd [11:44] tjaalton, yeah thats when he asked me to switch it i think, so thats not supprising [11:44] heh, ok [13:27] guys, have you ever noticed that loop devices don't show partitions? [13:28] e.g. [13:28] http://paste.ubuntu.com/1304882/ [13:28] do any of you know why? [13:29] ppisati, I would say they tell so in gendisk [13:29] smb: your sentence doesn't computer for me [13:29] ppisati, You should be able to use kpartx and device-mapper to get paritions mapped [13:29] ah k [13:30] ppisati, I think there is a flag in the gendisk struct to say whether a blk device can have partitions [13:31] smb: indeed kpartx has an example showing exactly how to mount an img [13:31] smb: i mean, a partition inside an img [13:31] ppisati, yep. [13:32] ppisati, Problem is minor numbers and space between devices normally used for partitions [13:37] ppisati, I normally use runes like: http://smackerelofopinion.blogspot.co.uk/2010/08/loop-devices-device-mapper-and-kpartx.html [13:40] cking: nice [13:59] * rtg will _not_ be running 3.7-rc2 on a Lenovo X120e until an mm/NUMA patch is merged that corrects a serious performance regression on a whimpy AMD E350 CPU.. [14:00] rtg sounds slightly (more?) grumpy [14:01] smb, I forgot I'd left that kernel on my machine and was totally frustrated at how long things were taking. this is my travel machine I'm getting ready for UDS [14:03] rtg, Oh and I thought it was because of the reason that justifies some belated HB wishes 3:-) [14:04] HB? [14:04] cking, my question as well [14:04] rtg, Could be a calendar fault, though I thought it was your birthday yesterday [14:05] smb, oh, that. yes it was [14:05] cking, in that context one can assume HB is short for "happy birthday" ? [14:05] it was supposed to be [14:07] Might also mean "HofBräuhaus". Which would be good for wishful thinking, too [14:07] beer-tastic [14:57] hello, forgive me if that has already been discussed before: isn't Q-proposed kernel update 3.5.0-18 introducing the ext4 corruption bug that has recently been found in stable kernels (changelog entry: " jbd2: don't write superblock when if its empty") ? [14:58] herton, ^^ [15:00] pstolowski, investigation upstream isn't concluded yet, and to reproduce it seems you need at least nobarrier used as options, which isn't the default [15:00] pstolowski, https://plus.google.com/117091380454742934025 [15:00] pstolowski, yes, i haven't checked this a.m. but it is still being investigated [15:01] herton: ok, I'm not following lkml, this commit was just mentioned on phoronix.net (quoting lkml) as a likely culprit [15:02] bjf, herton: ok, thanks for clarifitcation :) [16:59] protip: phoronix is not a source of quality analysis [17:00] Is it possible to prevent the ahci driver from initializing certain controllers (by pci id, or bus id, etc.)? Or perhaps to explicitly define the ones to initialize? [17:40] is there any kernel generic way of knowing from where the kernel loads device firmware bins> [17:40] ? [17:43] rtg: is git://kernel.ubuntu.com/ubuntu/ubuntu-raring.git meant to be usable right now? [17:43] (i assume not, there is no debian/ or ubuntu/, only a debian.master) [17:45] hallyn, everything is in master-next. it builds and boots but is a little slow due to an mm/NUMA bug on small CPUs [17:45] rtg: ah! thanks! [17:45] I'm gonna leave master as vanilla upstream until we upload 3.7 [17:47] * rtg -> lunch === yofel_ is now known as yofel [18:43] * rtg goes to pack for UDS