15:00:06 #startmeeting Fedora ARM & AArch64 Status Meeting 15:00:06 Meeting started Tue Oct 6 15:00:06 2015 UTC. The chair is pwhalen. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:06 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:06 #chair pwhalen bconoboy pbrobinson dgilmore hrw jsmith dmarlin ahs3 yselkowitz zsun 15:00:06 Current chairs: ahs3 bconoboy dgilmore dmarlin hrw jsmith pbrobinson pwhalen yselkowitz zsun 15:00:14 o/ 15:00:14 * pbrobinson is here 15:00:17 good morning folks 15:00:55 hey all 15:01:04 I am here but in another meeting also 15:02:48 ok, lets gets started 15:02:51 #topic 1) ==== Package Status & Issues ==== 15:03:02 how do we look pbrobinson, hrw ? 15:03:19 good, I would say 15:03:24 nothing major, libipt is blocking some stuff, I just need to work out the optimal shadow config for that 15:03:36 that's just for rawhide 15:03:42 f23 is looking really good 15:03:46 we got binutils fixed in f23, libint builds with it so it is a matter of getting it back into queue 15:04:17 for rawhide I plan to replace gccxml bdeps with castxml to get few more packages built on aarch64 15:04:37 * jsmith joins late 15:04:43 do you happen to have some stats handy? 15:04:52 from shadow 15:05:10 statistics: {'older': 128, 'local_only': 0, 'remote_only': 214, 'same': 16740, 'newer': 0, 'total_missing_builds': 303} 15:05:24 f23 ^^ 15:05:37 #info binutils now fixed in f23, libint needs to be requeued 15:05:57 #info F23 Koji-Shadow stats - {'older': 128, 'local_only': 0, 'remote_only': 214, 'same': 16740, 'newer': 0, 'total_missing_builds': 303} 15:06:26 f22 GA for comparison: statistics: {'older': 150, 'local_only': 0, 'remote_only': 245, 'same': 16169, 'newer': 1, 'total_missing_builds': 314} 15:06:47 #info Hrw to replace gccxml bdeps with castxml to get few more packages built on aarch64 in rawhide 15:06:50 about 600 more packages, delta is lower etc 15:07:02 #info f22 GA for comparison: statistics: {'older': 150, 'local_only': 0, 'remote_only': 245, 'same': 16169, 'newer': 1, 'total_missing_builds': 314} 15:07:25 nice 15:07:51 pbrobinson: can you export list of 'total_missing_builds' ones? 15:08:03 hrw: not easily 15:08:06 ok 15:09:13 and that stat could mean builds of the same package, iirc.. ie different nvr's? 15:09:42 good point pwhalen 15:09:44 pwhalen: yes 15:10:10 cool, well looks good. 15:10:28 anything else for packages? 15:10:42 nope 15:10:45 #topic 2) ==== Kernel Status ==== 15:11:56 nothing major from my PoV, anyone know of any issues 15:12:06 * jsmith isn't aware of anything new 15:12:57 nothing from me 15:13:10 i was testing 4.3.0-0.rc3.git4.1, looked good on the bbb 15:13:30 bpi - include/linux/rcupdate.h:578 Illegal context switch in RCU read-side critical section! 15:13:35 it booted though 15:13:39 http://paste.fedoraproject.org/275351/42318144/ 15:14:10 wandboard blew up - http://paste.fedoraproject.org/275346/14208614/ 15:15:02 for aarch64, the latest i looked at was- 4.3.0-0.rc1.git3.1.fc24 15:15:28 which gave the first couple of lines of output then nothing on console, will look at something newer 15:15:42 4.3.0-0.rc2.git0.1.fc24.aarch64 is what I run on pinkiepie for over 8 days 15:16:14 but thats looks like own build... hmm, have to check fedora one 15:16:24 hrw: please specify what HW your hostnames are when you make a comment ;-) 15:16:29 yea.. looks like we have newer now 15:16:30 so i will 15:16:40 pbrobinson: right ;D 15:16:41 i know pinkiepie :) 15:16:46 pinkiepie is apm mustang 15:16:46 heh 15:16:58 should be a rc4 as of today 15:17:00 father, my little pony reference :) 15:17:05 i got it 15:17:32 we have kernel-4.3.0-0.rc4.git0.1.fc24 15:17:36 http://arm.koji.fedoraproject.org/koji/buildinfo?buildID=319039 15:17:48 * pwhalen installs 15:17:55 anything else for our kernel? 15:17:58 s 15:18:17 #topic 3) ==== F23 TC1 for ARM ==== 15:18:29 #info Testing so far: https://fedoraproject.org/wiki/Test_Results:Current_Summary 15:18:29 #info F23 Blockers/FE's - https://qa.fedoraproject.org/blockerbugs/milestone/23/final/buglist 15:18:44 ive not hit anything arm specific. 15:18:59 looking pretty good for us. KDE and Workstation 'working' to some degree 15:19:20 is that new (with llvm 3.7 update ?) 15:19:55 I report workstation is working fine with lastest update (on jetson) 15:20:41 nchauvet: nice 15:21:00 nchauvet: any custom kernels or patches anywhere there? 15:21:29 workstation also worked on the the cubietruck, kde i tested on the bpi 15:21:53 pbrobinson, 4.3 kernel, anything else isn't wired yet to accelerate using nouveau 15:22:34 that's why I was guessing about llvm 3.7 update 15:23:03 working with kernel-4.2.2-300.fc23.armv7hl, (f23 TC1) 15:23:25 any other testing of TC1? 15:25:03 I've got as far as downloading it, will be trying it tomorrow when I get a moment 15:26:04 i imagine we'll be seeing an aarch64 f23 TC soon? or what are our plans there? 15:27:04 none as yet, soon, trying to get a few other bits out of the way this week first 15:27:36 I still need to unpack hardware 15:28:30 sounds good. 15:28:53 #topic 4) == Open Floor == 15:29:31 someone plans to buy pixel-c and get fedora on it? 15:29:45 I pushed out aarch64 updates for all current releases this morning 15:30:11 ah, right.. masta did you post a howto for the chromebook? id like to get that up and running 15:30:19 hrw: I have not heard of anyone doing so 15:30:30 pwhalen: indeed, me too 15:31:10 dgilmore, I think uboot-2015.10 could make it 15:31:23 it has the patch I've previously mentionned 15:31:25 nchauvet: make what? 15:31:44 chainload uboot with the chromebook 15:32:06 I suspect now the pixel-c is out there will be tegra210 chromebooks before long 15:32:37 Pixel-C is Ryu board which was originally a chromebook 15:32:59 Smaug chromebook seems to be a manufacturing state according to commits 15:33:03 it would be nice to see masta document his process for the booting of the various chromebooks 15:33:45 any other comment on rhbz#1259833 ? host triplet ? 15:34:37 pwhalen, still no answear back to this issue, are you abble to reproduce ? 15:35:54 rpm -E %{_host} on any armhfp installed system, compare with infrastrucuture using a scratch build -> it doesn't match 15:36:00 nchauvet: I'm still not sure it's needed and I'm not sure what impact it'll have on other non llvm builds 15:36:31 dgilmore, thoughts? 15:37:04 pbrobinson, why are you saying about llvm ? the report doesn't tell about llvm , can you just read the bug ? 15:37:23 nchauvet: I've read it a number of times 15:37:33 not this week 15:37:46 pbrobinson, good, so don't mention llvm it's totally unrelated 15:38:14 nchauvet: I don't see what's broken and what it's fixing 15:38:39 nchauvet: and when I spoke to you about it at flock you did mention llvm 15:38:43 in the same context 15:38:53 pwhalen: I filed a rpm bug for it being incorrect 15:39:48 dgilmore, thx, can you link to the bug 15:39:51 dgilmore, bz#? 15:41:14 https://bugzilla.redhat.com/show_bug.cgi?id=1265700 15:41:25 thanks! 15:41:32 there has been no comment from rpm folks 15:42:57 dgilmore, thx I really apprecite, someone going to push this issue forward, I don't have enought history and background to deal with that issue on my level 15:43:39 cool. 15:43:46 anything else for today? 15:44:10 * pwhalen googled hathor 15:44:47 pbrobinson & pwhalen: I'm still working on chrome books. I've got a process that sucks, and won't publish until it's not my hacks 15:44:58 I got many of them working, some more than others. 15:45:12 looking to get the rockchip variants 15:45:17 masta, ok, but we the people, like hacks 15:45:19 masta: concentrate on tegra ;-) 15:45:37 * nchauvet subscribe on any tegra chromebook donation 15:45:40 pbrobinson: tegra, ok =( that one needs help 15:46:19 masta, also if we can see what youve done, we can also try to 'help' 15:46:26 masta: feel free to take any/all CB discussion to arm channel, many hands light work and all that 15:46:35 masta: post early, post often. :) 15:46:35 ok ok 15:47:06 thank you sir 15:47:08 I'm hoping the exynos variants can be done with straight fedora 15:47:49 that means u-boot and kernel 15:48:03 * masta is hopeful 15:48:19 eif 15:48:22 eof even 15:49:00 last call.. 15:50:00 #endmeeting