15:00:29 <pwhalen> #startmeeting Fedora ARM and AArch64 Status Meeting 15:00:29 <zodbot> Meeting started Tue Sep 3 15:00:29 2019 UTC. 15:00:29 <zodbot> This meeting is logged and archived in a public location. 15:00:29 <zodbot> The chair is pwhalen. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:29 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:29 <zodbot> The meeting name has been set to 'fedora_arm_and_aarch64_status_meeting' 15:00:29 <pwhalen> #chair pwhalen pbrobinson jlinton 15:00:29 <pwhalen> #topic roll call 15:00:29 <zodbot> Current chairs: jlinton pbrobinson pwhalen 15:00:45 <pwhalen> good morning folks, who's here today? 15:00:48 * jlinton waves 15:01:03 * pbrobinson is notionally here 15:03:16 <pwhalen> ok, lets get started 15:03:25 <pwhalen> #topic 1) ==== Userspace Status ==== 15:03:31 <pwhalen> any new user space issues? 15:03:35 <pbrobinson> yes 15:03:55 <pbrobinson> there's been some reports on devel about regression in gcc/c++ 15:04:00 <pbrobinson> on aarch64 15:04:20 <pbrobinson> I suspect something has happened in gcc 9.2.1 release 15:04:36 <pbrobinson> it would be great if someone could take a lead on that 15:05:25 <pwhalen> any pointer to the mailing list? 15:05:59 <pwhalen> I see it now 15:07:51 <pwhalen> #info Possible regression in toolchain on aarch64 15:07:58 <pwhalen> #link https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/thread/GUU3OJTCOE2HKGQ5HLO424EOWFO2TPLX/ 15:10:15 <pbrobinson> there's been 2-3, just search aarch64 15:11:03 <pwhalen> right, looks like a couple are linked in that mailing list item. 15:11:53 <pwhalen> any other user space issues? 15:12:10 <pbrobinson> jlinton: any chance you could take a look? 15:12:28 <jlinton> Let me see if I can pawn it off on one of our toolchain guys 15:12:37 <pbrobinson> jlinton: even better :-D 15:12:50 <jlinton> (faster, and I'm deep in a NVDIMM debug) 15:12:50 <pwhalen> thanks jlinton 15:13:56 <pwhalen> #topic 2) ==== Kernel Status ==== 15:14:10 <pwhalen> #info F31: kernel-5.3.0-0.rc7.git0.1.fc31 15:14:10 <pwhalen> #link https://koji.fedoraproject.org/koji/buildinfo?buildID=1370569 15:15:04 <pwhalen> thats building now, did anyone see any regressions in rc6? 15:17:37 <pwhalen> only issue that comes to mind is with older aarch64 firmware requiring 'arm-smmu.disable_bypass=n' to boot with device tree. acpi seems ok. 15:17:47 <pwhalen> that was with 5.2 as well 15:18:28 <pwhalen> eg seattle - https://bugzilla.redhat.com/show_bug.cgi?id=1724276 15:19:34 <pwhalen> #info F30: kernel-5.2.11-200.fc30 15:19:34 <pwhalen> #link https://koji.fedoraproject.org/koji/buildinfo?buildID=1366771 15:19:41 <pwhalen> #info Please test and report any issues to the list or #fedora-arm. 15:19:49 <pwhalen> any other kernel issues/news? 15:20:24 <pwhalen> #topic 3) ==== Bootloader Status ==== 15:20:57 <pwhalen> #info uboot-tools-2019.10-0.3.rc3.fc31 15:20:57 <pwhalen> #link https://koji.fedoraproject.org/koji/buildinfo?buildID=1365901 15:21:39 <pwhalen> Dont recall any issues when testing last week. Anyone else? 15:22:28 <pwhalen> wandboard doesnt boot from usb but that isnt new with rc3 15:24:04 <pwhalen> #topic 4) ==== Fedora 31 ==== 15:24:18 <pwhalen> #info Fedora 31 Beta Blockers 15:24:18 <pwhalen> #link https://qa.fedoraproject.org/blockerbugs/milestone/31/beta/buglist 15:25:02 <pwhalen> looks like a bunch of new proposed blockers 15:25:41 <pwhalen> there is also a new nominated nightly for testing - https://fedoraproject.org/wiki/Test_Results:Current_Installation_Test 15:26:54 <pwhalen> #info 2019-09-17 - Beta Release (Preferred Target) 15:27:14 <pwhalen> #info 2019-09-24 - Beta Release (Target #1) 15:28:42 <pwhalen> anything else for f31? 15:29:17 <pwhalen> #topic 5) == Open Floor == 15:31:24 <jlinton> There is also some chatter about -fpic with g++ on rawhide not working 15:32:38 <jlinton> which appear to be related to the two linked reports above 15:32:40 <pwhalen> jlinton, fun. was that on one of the mailing lists? 15:32:45 <pwhalen> ok 15:33:59 <jlinton> Yes, fedora_dev "Please fix the aarch64 g++ pci problems in f32 rawhide" 15:34:12 <jlinton> looks like autoconf is turning off pic for C++ 15:34:25 <jlinton> which is maybe the root of the problems in at least one of those other reports 15:36:52 <pwhalen> thanks 15:37:03 <pwhalen> anything else for open floor? 15:37:09 <pwhalen> will leave it open for a couple minutes 15:38:03 <jlinton> As a late comment for F31, I tested the server compose from last week (thurs?) on the hikey/mcbin and they boot/instal 15:38:46 <jlinton> I didn't get too far into detailed testing (various components/etc) but the basic image can create a bootable system 15:39:36 <pwhalen> nice, thanks jlinton. we're getting down to it for beta so all feedback is good to hear 15:40:47 <pwhalen> #endmeeting