[00:47] Hm. who here looks after the panda kernel tree for ubuntu? [00:48] ojn: You want ppisati (or to file a bug). [00:59] infinity, thanks [01:05] christ, I forgot just how horrible launchpad is to use. [01:08] ojn: the BTS part? [01:10] twb: well, all of it. But in general it's hard to find how to do whatever you came to it to do. In this case, how to file a bug against the panda kernel. I had to find the package name by doing some random searches that happened to hit on bugs for it, go to the package, and there open the bug [01:10] btw, it's open at: https://bugs.launchpad.net/ubuntu/+source/linux-ti-omap4/+bug/965840 [01:10] I use https://help.launchpad.net/Bugs/EmailInterface when I have to deal with Ubuntu-specific problems; everything else I reproduce and deal with upstream so as to avoid launchpad [01:10] Launchpad bug 965840 in linux-ti-omap4 "Need to apply patch: "ARM: Do not call flush_cache_user_range with mmap_sem held"" [Undecided,New] [01:10] (Sorry, it took a while to find the URL) [01:10] ah there it is. :) [01:11] As you can see from http://cyber.com.au/~twb/.reportbugrc I was not in a good mood [01:14] I'm not sure pasting that was entirely necessary, but "thanks". [01:14] * twb looks briefly embarrased; goes back to lurking [01:15] ojn: Thanks for the report, I added a task for linux-ac100 as well, the only other ARM SMP kernel we currently ship. [01:15] why do i need an rc file to report a bug? If i'm not native on the system it's completely useless to have native tools for bug reporting [01:15] infinity, you're basing that on the chromeos branch though, right? I have added the patch there already [01:15] ojn: I assume this has no effect on non-smp (your description implies as much), and doesn't need applying to our omap3 and mx53 kernels? [01:16] infinity, given that it's a semaphore it can still happen (if you go to sleep across the call, then another thread comes in), but the scope is much much smaller [01:16] ojn: The ac100 task is just to make sure we already have the patch but, yeah, we may already. [01:16] ojn: Well, if it's theoretically triggerable on UP machines, I'll add tasks across the board. [01:25] infinity, thanks. it'd be awesome if that could go out as an update to current release instead of just into next release as well. not sure how to flag that on the report (if i even can do that) [01:26] ojn: We can do SRU tasks for it, though, with an LTS coming up, I'm not sure if it's worth the effort. [01:33] infinity, i guess you assume everybody always updates. far from always the case [01:34] ojn: No, I make no such assumption. But we do sort of assume that people might prefer to move to an LTS (and armhf) at some point soon, rather than one of the less-than-stellar releases leading up to it. [01:35] ojn: We do backport changes to old release kernels, though. But not every change is as critical as the reporter thinks, etc. [01:36] ojn: Which release in particular were you interested in? [01:36] ojn: maverick is EOL in a month, but I can add natty and oneiric tasks and see about getting it pulled into the next SRU round. === michaelh1 is now known as michaelh1|away === michaelh1|away is now known as michaelh1 === doko_ is now known as doko === zyga is now known as zyga-afk === zyga-afk is now known as zyga === zyga is now known as zyga-afk === prpplague^2 is now known as prpplague === zyga-afk is now known as zyga [15:38] ogra: do you have a 4460? [15:39] ppisati, yes [15:39] hmm. libo ftbfs on armel but not on armhf [15:39] I wonder if a give-back 'solves' it [15:40] ogra_: http://people.canonical.com/~ppisati/linux-image-3.2.0-1411-omap4_3.2.0-1411.14_armhf.deb [15:40] ogra_: tell if hdmi works there (btw should be) [15:40] *tell me [15:41] i dont have any working install on my panda atm [15:41] GrueMaster: are you around? [15:41] ppisati: Yep, pulling the kernel now. [15:41] GrueMaster: thanks [15:42] where would we be without him ! [15:43] ogra_: Ask that question again in a couple of weeks. [15:43] * ogra_ will definitely do that [15:43] that didn't sound so good... [15:45] ppisati: Woot! Video is back. [15:45] cool [15:45] (splash screen so far). [15:49] Ok, oem-config is coming up. But no usb input (keyboard/mouse). [15:50] let me rip apart boot.scr and add console output. [15:55] Probably need to do a proper kernel install (which I can't do on a preinstalled-desktop). Let me do some hackery to get the modules loaded. [16:02] ppisati: I think we'll just have to roll with this for now. I have no way of updating the kernel on the image easily. I could hack an image from another panda with usb drive support, but I can't hack the image on an SD (the panda bug of removing the SD even when it isn't mounted is still there). [16:19] ppisati: Ok, after some wack hackery, I have the kernel installed properly, keyboard/mouse work, life seems better. I'll let you know if I see anything in dmesg. [16:20] GrueMaster: cool, thanks [16:22] it takes so little to make life better for him :) [16:24] I retract part of that last statement. I have mouse but no keyboard. :( [16:26] ppisati: ^^^ [16:26] GrueMaster: if you have mouse, you should have keyboard too [16:27] One would think so. [16:27] Especially since this is an integrated keyboard. [16:27] out of battery ? [16:28] Wired. Not remote. [16:30] * GrueMaster checks earlyprintk output for obvious issues. [16:34] sigh. [16:34] * GrueMaster hates bugs that seemingly vanish when in debug mode. [16:36] btw, we will want to disable the built-in camera before kernel freeze. Most people don't have one, and those that do can easily rebuild the kernel to include it. [17:26] infinity, thanks. sorry, had to go offline for the rest of the night. For me, current live release is what matters most since that's where my user is at. [17:29] ojn: As in oneiric? [17:29] (And soon precise) [17:30] yeah, that's it. i forget the names of the releases these days. === yofel_ is now known as yofel [19:04] infinity: ping [19:05] morphis: Sup? [19:05] infinity: can I ask you something in private? [19:05] morphis: Asking publicly if you can ask something in private is a bit odd. ;) [19:06] morphis: But go nuts. === rsalveti` is now known as rsalveti