=== bryceh is now known as bryce === henrix_` is now known as henrix === doko_ is now known as doko [10:01] kengyu, i have a query re the verisoning of fwts, as this is not yet in debian should the version not actually be like 0.26.00-0ubuntu1 [10:35] apw, I think this is a good idea (though fwts is not in Debian yet). BTW we are the upstream too, for now we have not made any distro-specific patch, so the suffix is always the lowest. But the suffix like -0ubuntu1 is helpful to avoid any potential confusion IMO. Thanks for spotting this. I will use the -0ubuntu1 suffix from the next upload. === _jmp__ is now known as _jmp_ [12:55] * henrix -> lunch === amitk is now known as amitk-afk [13:59] apw, check that out. 527897ccd968c86ad3265d62962c8beccdb94e47 is the fastest I've ever gotten a patch merged. [14:01] rtg i saw :) === amitk-afk is now known as amitk [14:16] where can I find latest 3.7 for quantal? Latest build for Quantal in kernel-ppa is 3.7rc2 . I want to check if I reproduce bug 1088433 [14:16] Launchpad bug 1088433 in linux (Ubuntu) "PERCPU: allocation failed when loading module kvm" [Undecided,Confirmed] https://launchpad.net/bugs/1088433 [14:17] jibel, that one goes back at least as far as 3.2 on certain machines. [14:18] jibel, https://launchpad.net/~ubuntu-x-swat/+archive/r-lts-backport [14:26] rtg, on this machine, I don't find trace of this message in logs before to 3.5.0-19, is it something I should worry about? it doesn't seem to impact the behavior of the system [14:34] jibel, it appears to be benign, but then I have not actually tested the kvm module for functionality. [14:36] * rtg back on in a bit [15:09] cking, did you see this on LKML ? might have some impact on fwts: "[PATCH] ACPI: add documentation about ACPI 5 enumeration" [15:10] * cking has a peek [15:11] nice [15:13] I so love ACPI 5 [15:14] cking, thought you might :) [15:14] :-/ [15:28] sforshee: I heard you were working on getting everything working on the Macbook Pro [15:28] sforshee: if you have stuff to test in raring, feel free to send them my way :) [15:29] shadeslayer, which model do you have? Most of the work I'm doing now with the macs is centered on Broadcom wireless [15:29] I have the 8,2 [15:29] yeah, the wireless on that one is not very good [15:29] *nod* [15:30] if by not very good you mean doesn't work out of the box, then yeah, it's not very good :P [15:30] I've got a different model with the same wireless that I plan to work with, but that wireless chipset is difficult [15:30] we have a very limited understanding of the phy [15:31] I see [15:31] sforshee: I can get wireless to work with hacks, what's more problematic for me is video drivers [15:31] that requires kernel patching and what not [15:32] shadeslayer, I assume then you're trying to do native EFI boot? I think graphics works okay under CSM on that machine. [15:33] yeah, native EFI :) [15:33] * shadeslayer has to use fbdev right now [15:34] so 3.8 should have patches that let you the radeon with modesetting under EFI boot [15:34] switching to the integrated card still isn't going to work though === bdmurray_ is now known as bdmurray [16:17] brendand, do you have a Dell Latitude E6410 in your lab? [16:20] sforshee: oh awesome, still no 3.8 debs on http://kernel.ubuntu.com/~kernel-ppa/mainline/ though [16:20] so I'll have to wait [16:20] bjf - indeed. with nvidia graphics [16:20] shadeslayer, there's no 3.8 yet. 3.7 is still in progress ;-) [16:20] the patches are queued up to go into 3.8 right now though [16:21] ah [16:21] brendand, ah, i want intel graphics [16:21] brendand, how about a lenovo edge with intel graphics? [16:21] sforshee: so basically in 3.8 I can boot with i915.modeset=0 and ATi will work? [16:22] the value if i915.modeset=0 should have no impact on your radeon graphics [16:22] *of [16:22] bjf - which Edge? [16:23] brendand: 0301GXG ? does that make sense? [16:24] bjf - looks like an SKU [16:24] brendand, i'm using the dmi info in a bug and that's about as good as i'm getting right now [16:25] brendand, thanks anyway. if i get more info i'll ping you again. [16:25] bjf - i think i found it. if you point me to the bug i can probably narrow it down [16:25] bjf - google says that an Edge 15, but we might have a few [16:26] brendand, bug #1067622 [16:26] sforshee: I see, it's mostly that if I boot with KMS the EFI VGA and Intel drivers conflict [16:26] and it just hangs [16:26] sforshee: any chance we can see a kernel in raring with said patches? [16:27] shadeslayer, I never had to use i915.modeset=0, fwiw [16:28] sforshee: do you have hybrid graphics or just the intel chip? [16:28] if the patches are in 3.8 then raring will get them when we rebase the kernel onto 3.8 [16:28] I see [16:29] shadeslayer, what work I've done with the macbook hybrid graphics was done on a macbook pro 8,2. I don't have that machine anymore however. [16:29] ah [16:29] bjf - sure that's the right bug? [16:30] sforshee: right now I actually boot with radeon.modeset=0 , without it I get : "fb: conflicting fb hw usage radeondrmfb vs EFI VGA - removing generic driver" [16:30] brendand, https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1087622 [16:30] Launchpad bug 1087622 in linux (Ubuntu) "Linux kernel 3.5.0-20 won't boot [unable to handle kernel paging request at f91fe4fc in trace_event_raw_init+0xb/0x20]" [Undecided,Confirmed] [16:30] brendand, sorry about that [16:32] bjf - bingo! [16:32] henrix, ^ we might have the HW to repro the problem [16:33] bjf - well, i've got here an Edge 15 with Intel graphics, PCI ID matches [16:33] brendand, can we get someone to install the quantal on that and then try the latest -proposed kernel? [16:34] bjf - ah right, it's a matter of not booting. no problem [16:34] brendand, we're trying to get this fixed up for the 12.04.2 release [16:36] shadeslayer, that message isn't anything to be concerned about *if* the modesetting driver is working (but it's not in your case). It's basically just saying that the radeon driver is trying to use the same memory resources as the EFI fb driver and that radeon is being given precedence. [16:36] oh [16:36] interestingly, the boot just stops after that [16:36] can't even switch to a TTY [16:37] shadeslayer, it's not that the boot stops, it's just that your video is completely hosed [16:37] oh ... I wouldn't know, since no sound is played ( because I don't have autologin ) [16:38] sforshee: any tips on how to avoid that? [16:38] shadeslayer, either use bios-compatible boot or wait for 3.8 [16:38] alright, will wait for 3.8 === asac` is now known as asac [17:16] brb [17:47] * ppisati -> gy [17:47] m [18:10] * rtg -> lunch [18:37] bjf / henrix: Mayday. I may have just released the lucid SRU to -updates when I meant to release the hardy one. [18:38] ouch! [18:38] infinity, meh, probably not a big deal [18:39] It only fixes a single bug, and is only lacking cert-testing on the workflow. [18:39] Regression and the rest were done. [18:40] infinity: i believe its an CVE, so we may want to ping jj...? [18:40] He already did his signoff too, but yeah. [18:40] oh, ok [18:40] The only unfinished task was Brendan's. [18:41] So, probably not a big deal. [18:41] But since I can't unwind this... [18:41] apw: Want to ring up Brendan and ask him to join in on this conversation? [18:43] As it is, I'm just going to mark it released, cause there's not a whole bunch else I can do, but if Brendan can fast-track the hwcert stuff and make sure it's not somehow weirdly broken, that would be lovely. [18:44] infinity, that patch is for ipv6. i doubt they do any ipv6 testing [18:45] Yeah, you're likely right. [18:45] And if there were miscompiled binaries, the regression-testing would have picked it up. [18:45] infinity, i think all is fine [18:45] I'm not higely concerned. [18:45] Just annoyed with myself for fat-fingering. [18:45] Well, fat-braining? [18:46] ok, so... we'll leave it as it is and wait for brendand to finish the cert task. [18:47] correct? [18:47] henrix, infinity can mark his task as fix released [18:47] I already did. [18:47] Since it is. [18:47] yup [18:49] I look forward to a few weeks without working on Mondays... === henrix is now known as henrix_ === henrix_ is now known as henrix [19:11] apw, so how did you debug this async populate_rootfs stuff ? even using early_printk I can't seem to get any console output before it just locks up after "Loading initial ramdisk" === kees_ is now known as kees [22:12] * rtg -> EOD === chiluk is now known as chiluk_away [23:07] does the 12.10 have this patch: https://bugzilla.kernel.org/show_bug.cgi?id=48721 [23:07] bugzilla.kernel.org bug 48721 in Power-Processor "i915: acpi_cpufreq can't change CPU clock anymore - Clevo W150HRM w/ i7 2720QM CPU" [Normal,New] [23:07] well it is on i3 also [23:11] freedomrun, yes, we have 6af2d180f82151cf3d58952e35a4f96e45bc453a in our tree [23:12] freedomrun, if that's what you are talking about [23:12] bjf oh thank God and you .. thnx man u saved my day === henrix is now known as henrix_