=== amitk is now known as amitk-afk [07:57] moin === smb` is now known as smb === amitk-afk is now known as amitk === jevin__ is now known as jevin [09:24] * apw struggles on with pants internet [09:24] * apw struggles on with pants internet [09:25] * smb struggles with pant sleepiness [09:26] Better than pantsless internet! [09:27] RAOF, As long as I don't have to see that. :-) [09:28] smb 96433f6ee49032d7a8bda76de2b05cfde2914354 [09:40] * cking offers apw some of his spare internet connection capacity [09:40] RAOF, i would settle for using it pantless if it went faster === neunon is now known as snoonan [09:50] brb [10:26] * henrix reboots [10:41] Hey, can someone tell me if linux-image-lts-quantal / so stops being updated for precise after the next LTS is released? [10:43] jpds, i believe that is the plan, people there will be updated to the TT kernel on PP and can stay there [10:43] apw: So there'll be a linux-image-lts-t kernel for precise? [10:44] jpds, that is my understanding indeed, that we offer an 14.04 kernel as the hwe kernel on 12.04, and the other lts kernels will roll to that one [10:45] apw: What about people on precise, that want to enable hardware that's enabled... after T? [10:57] jpds, they need to be on T [12:14] rtg, apw: You're going to want commit 829a337c48b835c322e625e390b0bb07769e913b from my ppc tree [12:14] UBUNTU: SAUCE: 8250: Wrap ACPI calls with ifdefs for non-ACPI systems [12:15] All the commits you cherry picked trashed that driver for anything non-x86 [12:31] infinity: Fixed linux-ppc is waiting to build [12:48] BenC, I had some trouble getting that to build on armhf (so I turned off 8250 for armhf IIRC). Have you sent that patch upstream yet ? [12:48] rtg: No, I'm hoping the person that keeps cherry picking for stable kernels will do it for me eventually :) [12:49] I guess it fell off my todo list. [13:05] BenC, see 053fac36b1d9f76adde96a2f731965aaab3c632b in Linus repo. I'll see about getting that fixed in Raring === yofel_ is now known as yofel [13:13] * henrix -> lunch [13:37] rtg, thanks [13:37] apw, uh, for what ? [13:38] handling BenC's issue [13:38] ah, no problem [13:38] rtg: that fix does it…I'll disappear my commit from ppc branch next time I sync to ubuntu/master [13:38] build testing armhf right now [14:01] rtg, just pushed a bit more fixing for the linaro -> nexus7 rename [14:01] apw, oh, did I push that without testing ? I got distracted I think before it was finished [14:02] hum, ideas why i915.ko built from a dkms package is not loaded on haswell, but i915_hsw.ko [14:02] rtg no i think what you pushed would build fine, it is reconfiguration and the like which needed more tweaks [14:02] apw, cool, thanks. [14:02] tjaalton, PCI IDs ? [14:02] tjaalton, well i think it would load both and whichever binds get sit ? [14:02] gets the device [14:03] hmm right [14:03] tjaalton, also where is i915_hsw.ko, is that in updates/ or just in the normal place [14:03] normal place [14:03] and the dkms where is that, updates ? [14:03] yeah I see in dmesg that it's probably trying to load both [14:03] yep [14:04] as i think i expect updates to be loaded first, but probe is async i think [14:04] lspci shows i915 being used, but lsmod shows i915_hsw [14:04] so I think lspci lies [14:04] so ... perhaps the dkms package should include an /etc/modprobe.d/stop-haswell.conf which blacklists i915_hsw.ko [14:04] yep [14:05] probably the cleanest way out of this [14:05] but that is a bit pants [14:05] :) [14:06] rtg: the dkms version supports all, _hsw only haswell. so the dkms package works for !HSW. anyway, blacklist it is then [14:07] tjaalton, ack. Is this going to be your preferred solution for Haswell platforms ? [14:08] rtg: until i915_hsw is rebased, but the dkms is needed right now [14:09] tjaalton, so you're still gonna send a pull request for quantal eventually ? [14:09] yeah I hope so :) [14:09] k [14:09] been fighting fires so didn't get to it yet [14:11] tjaalton, my kid fights fires in the summer. it pays pretty well. [14:12] hehe [14:37] guys [14:38] centrino 6300 on raring has a big issue with disconnecting then becoming unavailable [14:39] * ogasawara back in 20 [14:48] bjf, fyi, bug bots are broken again on cranberry due to an upgrade :-/ [14:48] jsalisbury, that's special [15:12] ppisati, so the plan is to leave the panda desktops as is and dont touch thzem at all [15:30] ogra_: then i'll sync up R/omap4 wrt to Q/omap4 [15:30] rtg: ^ [15:30] ppisati, ack [15:30] yep [15:30] and then let it rot :) [15:31] ppisati, ogra_: if omap4 is now only for the builders, then why bother with an omap4 in Raring ? [15:33] rtg, we want the community to still have builder options ... and (worse) still want to have them for dogfooding desktop stuff on arm [15:33] i would have gone with server images from mainline kernels ... [15:33] ogra_, can we still do that with the device tree kernel in raring ? I'd still like to rename it to something more generic. [15:34] ogra_: +1 [15:34] rtg: doing it now [15:34] rtg: s/omap/multiarm/g [15:34] rtg, i dont care what you guys do as long as pvr still works for desktop [15:35] well, s/i/they/ (whoever) :) === kentb-out is now known as kentb-mtg [15:42] ogra_, can you find the bit of build magic for the N4 kernel ? there has to be a bit more to it then just using arch/arm/configs/mako_defconfig [15:43] i have never built any official phablet image ... only ported to my own devices and they definitely use a cm_*_defconfig [15:43] but i dont see a cm_mako_defconfig in the tree [15:44] cm_x2xx_defconfig cm_x300_defconfig [15:44] yep i see them [15:45] kernel/lge/mako/AndroidKernel.mk ... [15:45] ogra_, looking at the git repo implies that mako_defconfig is in use 'cause that where some config options have been changed for the phablet image [15:47] did you try running a fresh repo sync ? [15:49] * ogra_ doesnt really see where KERNEL_DEFCONFIG gets defined [15:51] aha device/lge/mako/BoardConfig.mk [15:51] yeah its mako_defconfig, you are right [15:53] rtg, so i suspect you have to mimic whatever kernel/lge/mako/AndroidKernel.mk does to get a proper config [15:54] there seems to be a bunch of perl touching it [15:55] ogra_, that lge directory does not exist in git://phablet.ubuntu.com/CyanogenMod/lge-kernel-mako.git [15:58] weird, must be something repo does merge then [15:59] likely from CyanogenMod/android_device_lge_mako [15:59] thats whats defined in the top level manifest.xml here [16:00] http://paste.ubuntu.com/5631532/ [16:00] thats the AndroidKernel.mk in my tree [16:01] ogra_, cool, thanks [16:01] line 14 and 15 look like they touch it [17:13] apw, git://kernel.ubuntu.com/ubuntu/ubuntu-nexus4.git - can you give me a hand and tell me why config-check wants to run debian.master/config/enforce ? The N7 script is identical, yet I don't have that failure. [17:16] apw, oh, well actually now I _do_ have the same failure. its part of your patch 'UBUNTU: [packaging] Rename from linaro to nexus7 -- part 2' [17:29] * rtg -> lunch [17:31] rtg: could you refrain from touching R/master-next abi today? [17:34] * ppisati -> workout [17:34] later [18:02] bah, ESTA and visas. what a pain [18:03] and visas ? [18:03] do you need both in the UK ? [18:11] ppisati, why just the abi ? [18:57] * henrix -> EOD === JanC_ is now known as JanC [20:06] * rtg -> EOD [20:35] q: i'm trying to build a kernel out of the git tree, and have noticed that a number of the firmware files are dropped - [20:36] is there an easy way to figure out what I need to pull in from linux-firmware?