**** BEGIN LOGGING AT Wed Apr 14 02:59:56 2021
Apr 14 06:58:30 <rmilecki> hooray, bcm4908 built :) https://buildbot.openwrt.org/master/images/#/builders/71
Apr 14 07:42:42 <guidosarducci> rmilecki: any idea what happened to 21.02-SNAPSHOT build for ath79? There was a fix made for some build failures but no retries since then...
Apr 14 07:43:34 <rmilecki> guidosarducci: no idea about 21.02
Apr 14 07:44:04 <rmilecki> guidosarducci: mkmylofw regression and fix I did in master was never backported to the 21.02
Apr 14 07:44:36 <guidosarducci> rmilecki: oh, was that impacting 21.02 as well?
Apr 14 07:44:53 <rmilecki> guidosarducci: i couldn't affect 21.02 git branch
Apr 14 07:45:05 <rmilecki> guidosarducci: unless it was builders that run out of space and failed to build 21.02
Apr 14 07:46:03 <plntyk> did someone replicate the build errors of the bots with spidev_test or libmraa/libupm failing on all targets ? seems like those are somewhat related to setting cleanup flags or buildbot config itself
Apr 14 07:46:55 <guidosarducci> rmilecki: I don't recall any out-of-storage errors. There was a libnfnetlink problem that caused lots of package builds to fail. But after the fix, I haven't seen rebuilds since Sunday.
Apr 14 07:48:54 <rmilecki> guidosarducci: if you really need a proof: https://buildbot.openwrt.org/master/images/#builders/9/builds/49 https://buildbot.openwrt.org/master/images/#builders/16/builds/53 https://buildbot.openwrt.org/master/images/#builders/48/builds/51
Apr 14 07:52:03 <mangix> plntyk: libmraa/libupm are related to 32-bit host GCC missing
Apr 14 07:52:32 <mangix> spidev_test is some special kind of hell
Apr 14 07:53:00 <guidosarducci> rmilecki: not looking for "proof", just curious as I don't know the backstory. Links are handy, thanks. How to filter for particular branches being built?
Apr 14 09:14:59 <ynezz> plntyk: IIRC spidev_test probably fails as it can't be built with SDK due to some kernel dependency
Apr 14 09:15:55 <ynezz> which is not available in the SDK
Apr 14 09:18:42 <plntyk> probably - spidev_test is in tools/ dir of kernel - it was last shipped w. 18.06.x so SDK change probably correct
Apr 14 09:33:43 <owrt-2102-builds> Build [#27](https://buildbot.openwrt.org/openwrt-21.02/images/#builders/13/builds/27) of `ramips/rt305x` completed successfully.
Apr 14 13:14:37 <mangix> Oh nice. Kernel people removed single queue I/O schedulers.
Apr 14 13:46:40 <rsalvaterra> mangix: A long, long time ago… good riddance!
Apr 14 14:01:19 <mangix> Can't respond to that patch right now. I don't know if there's any point to it for OpenWrt.
Apr 14 14:04:46 <mangix> I remember I had a patch getting rid of all of them lea ing only noop. I didn't find a difference between it and cfq.
Apr 14 14:11:20 <rsalvaterra> mangix: What patch?
Apr 14 14:11:43 <mangix> The Linus one
Apr 14 14:11:50 <rsalvaterra> Anyway, you probably haven't noticed any difference on an SSD, right?
Apr 14 14:13:15 <mangix> I was testing /rom AFAIK
Apr 14 14:13:16 <mangix> .
Apr 14 14:13:22 <mangix> http://patchwork.ozlabs.org/project/openwrt/patch/20210414114230.2302716-1-linus.walleij@linaro.org/
Apr 14 14:13:55 <rsalvaterra> mangix: I see. Well, I'm not subscribed to the mailing list, but I'd ack it, fwiw.
Apr 14 14:15:31 <rsalvaterra> Oh, wait. Linus is building the bfq_sched in the kernel itself. Please don't. Make it a module.
Apr 14 14:17:00 <rsalvaterra> The udev rule I use is a bit different, though…
Apr 14 14:17:00 <mangix> That sounds good actually.
Apr 14 14:17:02 <rsalvaterra> ACTION=="add|change", SUBSYSTEM=="block", DRIVERS=="sd|sr", TEST=="queue/scheduler", ATTR{queue/scheduler}!="bfq", ATTR{queue/scheduler}="bfq"
Apr 14 14:19:50 <mangix> Hmmm wasn't there a kernel parameter to set the default?
Apr 14 14:20:35 <rsalvaterra> mangix: Nope. It's supposed to be set in sysfs, by design.
Apr 14 15:35:08 <owrt-snap-builds> Build [#55](https://buildbot.openwrt.org/master/images/#builders/16/builds/55) of `ath79/mikrotik` completed successfully.
Apr 14 15:43:14 <owrt-2102-builds> Build [#27](https://buildbot.openwrt.org/openwrt-21.02/images/#builders/19/builds/27) of `layerscape/armv7` failed.
Apr 14 15:50:52 <rmilecki> ./config.sub: line 1055: syntax error near unexpected token `newline'
Apr 14 15:50:56 <rmilecki> ./config.sub: line 1055: `	pentiumii-'
Apr 14 15:51:00 <rmilecki> configure: error: /usr/bin/env bash ./config.sub x86_64-pc-linux-gnu failed
Apr 14 15:51:07 <rmilecki> make[2]: *** [Makefile:35: /builder/shared-workdir/build/build_dir/host/gmp-6.2.1/.configured] Error 2
Apr 14 15:51:16 <rmilecki> i think that caused layerscape build to fail
Apr 14 16:41:34 <kristrev> what is the current preferred way to expose a GPIO that should be controlled by user? via board.d?
Apr 14 22:54:19 <owrt-snap-builds> Build [#49](https://buildbot.openwrt.org/master/images/#builders/64/builds/49) of `realtek/generic` completed successfully.
Apr 14 23:19:03 >KGB-0< https://tests.reproducible-builds.org/openwrt/openwrt_ath79.html has been updated. (99.1% images and 98.3% packages reproducible in our current test framework.)
Apr 15 00:37:56 <owrt-snap-builds> Build [#60](https://buildbot.openwrt.org/master/images/#builders/2/builds/60) of `layerscape/armv7` failed.
Apr 15 02:08:04 <owrt-snap-builds> Build [#61](https://buildbot.openwrt.org/master/images/#builders/2/builds/61) of `layerscape/armv7` completed successfully.
**** ENDING LOGGING AT Thu Apr 15 02:59:56 2021