15:00:24 #startmeeting Fedora ARM & AArch64 Status Meeting 15:00:24 Meeting started Tue Mar 3 15:00:24 2015 UTC. The chair is pwhalen. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:24 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:24 #chair pwhalen bconoboy pbrobinson dgilmore hrw jsmith kyle awafaa 15:00:24 Current chairs: awafaa bconoboy dgilmore hrw jsmith kyle pbrobinson pwhalen 15:00:41 good morning folks! happy Tuesday 15:01:11 .fasinfo kyle 15:01:12 kylem: User: kyle, Name: Kyle McMartin, email: kmcmartin@redhat.com, Creation: 2007-11-07, IRC Nick: kyle / kylem, Timezone: America/New_York, Locale: en, GPG key ID: 191fcd8a, Status: active 15:01:15 kylem: Approved Groups: fedorabugs cla_fedora cla_done packager cla_redhat provenpackager @gitkernel-arm64 15:02:17 just giving it a couple minutes for others to join 15:02:31 * pbrobinson is here 15:02:34 pwhalen: is there a specific agenda or just a free form? 15:02:40 .name is slimshady 15:02:51 howdy 15:03:02 awafaa: there is an agenda... sent to the mailing list 15:03:05 awafaa, we have a couple of items 15:03:22 ok, lets get started! 15:03:25 #topic 1) ==== Fedora 22 Testing ==== 15:03:38 #info Fedora 22 Alpha TC7 Testing Summary 15:03:38 #link https://fedoraproject.org/wiki/Test_Results:Fedora_22_Alpha_TC7_Summary 15:03:47 #info Blocker Bugs & Freeze Exceptions 15:03:47 #link https://qa.fedoraproject.org/blockerbugs/milestone/22/beta/buglist 15:04:00 we have one arm specific bug so far 15:04:09 #info ARM Specific Bugs 15:04:15 #info network spoke listed as not connected despite having assigned IP and hostname (BZ#1183807) 15:04:15 #link https://bugzilla.redhat.com/show_bug.cgi?id=1183807 15:05:20 .hellomynameis yselkowitz 15:05:22 yselkowitz: yselkowitz 'Yaakov Selkowitz' 15:05:38 other issue is desktop enabled images booting to initial-setup-text 15:05:50 some boots we get graphical, others text 15:06:19 it's a timing thing as there's no way to detect whether it;s text or graphical 15:06:25 but have the packages changed? 15:06:40 we use to only ship/enable the text one for minimal I thought 15:07:05 right, only graphical is enabled, but seems to fallback to text none the less 15:07:46 OK, maybe need to compare that to F-21 15:08:03 couple other small issues but not arm specific and being resolved. 15:08:08 see if the text package has crept into the dep chain 15:08:32 TC8 I think should be announced momentarily 15:08:34 will look at it some more 15:08:49 #info Fedup testing completed successfully on Calxeda Highbank (F21 -> F22_Alpha_TC7) 15:09:20 #info Fedora 22 now includes 'Server' specific image 15:09:52 #info F22_Alpha_TC8 should be announced shortly 15:10:07 #link https://dl.fedoraproject.org/pub/alt/stage/22_Alpha_TC8/ 15:10:32 aarch64 images? 15:11:00 hrw: will be coming shortly, just trying to get infra sorted out 15:11:01 Peters been threatening a TC soon 15:11:08 cool 15:11:25 * pbrobinson is trying to reinstall the PHX mustangs with F-21 15:11:27 any other testing results, etc? 15:11:29 * masta is here 15:12:41 I've fixed panda 15:12:50 will be working in today's kernel 15:13:03 nice 15:13:06 so unless a new kernel gets pulled in for TC/RC it'll be post alpha 15:13:23 but I wouldn't be surprised if we did get a new kernel for something 15:13:46 #topic 2) ==== kernel-4.0 Status ==== 15:13:54 while we're on the kernel 15:14:35 so I've tested the latest 4.0rc1 kernel on BBB, PB, CT, WB-quad 15:14:39 #info F22 Alpha TC7 includes 'kernel-4.0.0-0.rc1.git0.1.fc22.armv7hl' 15:14:47 possibly some others too 15:15:11 sounds similar 15:15:33 on aarch64, we have some issues on Seattle hardware 15:16:11 but i'm using older firmware, so i think i need to upgrade 15:16:21 we just need new firmware there, but there was some problems with the latest one, so we're waiting for it to get debugged. 15:17:03 quality firmware.... 15:17:13 #info new firmware expected to resolve boot issues on Seattle hardware 15:17:22 thanks kyle 15:17:39 anything else kernel related? 15:18:18 just test and report success/failure 15:18:58 righto 15:19:00 #topic 3) ==== Package Status & Issues ==== 15:19:32 pbrobinson, hrw anything causing grief? 15:19:46 just ghc 15:19:53 all the rest of the core bits are built 15:20:10 is that an issue with ghci? 15:20:21 and the rest of the failures are the usual minor packages with regressions and pretty standard failures 15:20:26 #info ghc continues to be an issue on aarch64 15:20:31 #link https://arm.koji.fedoraproject.org/koji/packageinfo?packageID=2229 15:20:48 awafaa: upstream is terrible and doesn't test their crap? 15:20:55 hah. 15:21:06 stepping away for a sec, hilite if you need me 15:21:18 pbrobinson: heh, ok - so not the fact that something is missing an aarch64 port 15:21:18 #info the ghc issue is being tracked in https://bugzilla.redhat.com/show_bug.cgi?id=1195231 15:21:50 awafaa: we've had basic support for some time, it's an issue with the 7.8.4 rebase 15:22:05 moarvm would be nice to get update - https://bugzilla.redhat.com/show_bug.cgi?id=1186358 15:22:25 awafaa: but then it seems to do things like change ABI depending on the number of cores it's built with so anyone's guess 15:22:40 oh wow 15:22:47 impressive 15:23:07 hrw: have you pinged the maintainer or had any comms outside the BZ? 15:23:51 pbrobinson: will mail him directly 15:24:09 hrw: let me know if you have any issues 15:24:15 ok 15:24:45 any other problem packages? 15:24:54 not from me 15:25:04 csnappy? 15:25:05 as a heads up for those that like node 15:25:38 as joyent refuse to update their stable release with a modern v8 engine 15:25:55 io.js are now working on fixing their build infra to support aarch64 15:26:18 hrw: yes, that's just a minor one if you could look at it, not really a core one for the meeting 15:26:24 sure 15:26:30 should be available soon, they have access to hardware and are moving forward 15:26:34 awafaa: you're incorrect 15:26:47 awafaa: nodejs 0.12 supports a v8 that supports aarch64 15:26:58 when did that get pushed out the door? 15:27:14 awafaa: a while back, sometime early feb? 15:27:49 pbrobinson: thanks for the correction 15:28:14 awafaa: I've pinged the Fedora maintainer about his upgrade intentions but not got a response 15:29:17 #topic 4) ==== F22 Documentation ==== 15:29:28 awafaa: FYI http://blog.nodejs.org/2015/02/06/node-v0-12-0-stable/ 15:29:35 #info Initial armhfp F22 Installation page created 15:29:35 #link https://fedoraproject.org/wiki/Architectures/ARM/F22/Installation 15:30:04 thats followup from the last meeting. copied and pasted from f21, changing release 15:30:12 pwhalen: I've not yet had time to look at it 15:30:38 sure, i did last week but forgot to email out. we'll talk about it next week, edit will be needed of course 15:31:02 i dont imagine jsmith had time to do anything with it either 15:31:24 jsmith was travelling last week so probably not 15:31:43 i'll do the aarch64 page shortly, that needs some more work this time around 15:32:13 anything else for docs? ideas? 15:32:58 not for this meeting 15:33:01 #topic 4) == Open Floor == 15:33:13 meeting time change for daylight savings time? 15:33:37 when is DST? 15:33:45 nope... it should stay on UTC time :-D 15:33:46 in north america its this weekend, which as i understand it most Fedora meetings follow 15:34:06 +1 for UTC time 15:34:09 pwhalen: maybe for Americans like you ;-) 15:34:19 how dare you 15:34:24 :-D 15:34:24 DST hits europe on the last sunday of march 15:34:28 proud Canadian 15:34:46 Canada is America B? 15:34:46 yeah, UTC please 15:34:58 ok, im fine keeping it .. hopefully doesnt conflict 15:35:05 honestly doesn't particularly bother me 15:35:20 i have a question about armv8l 15:35:22 it's just too easy to wind up Canadians :-P 15:35:39 pbrobinson: just like kiwis like you? ;) 15:35:42 awafaa: as in 32 bit on a64 HW? 15:35:47 oh Peter if it was, youd drive me to drink long ago 15:35:50 pbrobinson: yup 15:36:02 awafaa: what's the question 15:37:31 * kylem sits with baited breath. 15:38:06 a good number of packages need to get updated if they want to compile on armv8l rather than armv7l 15:38:23 what are your proposals for the fixes? 15:38:33 awafaa: we're not planning, at this point in time, doing armv8l packages 15:38:33 and is it something that fedora even cares about? 15:38:46 we'll probably just end up lying about it in uname, by adding a personality for it. 15:39:02 awafaa: basically what kylem said 15:39:03 since we can fix up the SIGILL in kernel space on the unimplemented instructions. 15:39:18 fair enough, thanks 15:39:37 * kylem has been looking at it, but way on the back burner. 15:41:22 anything else for open floor? 15:42:21 someone got usb on mustang with 4.0-rc? 15:43:00 dmarlin was looking at it 15:43:29 I believe it requires a firmware fix (device tree setup) 15:43:41 ah, right 15:44:13 This meeting will self destruct in five seconds.. 5...4...3.. 15:44:30 woo! more firmware updates! 15:44:40 :) 15:44:51 #endmeeting