**** BEGIN LOGGING AT Wed Apr 07 02:59:57 2021 Apr 07 07:38:48 aparcar[m]: did you have a chance to read what I wrote to you on #openwrt-adm ? Apr 07 07:39:54 Yea what point did I miss to react on? Apr 07 07:40:20 aparcar[m]: so I'll try again here one more time, you've started from the end, you've added new build dependency to the buildbots (haven't seen it discussed anywhere), forget to add it to README.md and perhaps to the build prereq check as well Apr 07 07:41:48 This PR adds the prereq: https://github.com/openwrt/openwrt/pull/3855/files#diff-a72f6d1945236d66072717ee1c0a8c155f51cbb77e9127b5e011dc9f1e95be3eR57 You're right that it's missing in the README, thanks Apr 07 07:42:18 Starting them from the end results in the advantage of not breaking all builds by requiring a lib which isn't installed on any of the buildbots Apr 07 07:44:23 and we don't have control over all buildworkers, so we're going to disable few of them anyway Apr 07 07:45:50 a few != break all builds Apr 07 07:46:36 anyway I'm happy to revert it and hand it over to you, guidosarducci already offered to have it as a host tool instead of host dep Apr 07 07:46:50 no time, sorry Apr 07 07:50:18 BTW stintel is one of sysadmins so he has access to the ansible repo and can update the buildworkers under control when needed Apr 07 07:54:11 he wants a second person to back it, since you were against it a did not encourage to proceed Apr 07 08:23:08 >KGB-0< https://tests.reproducible-builds.org/openwrt/openwrt_kirkwood.html has been updated. (100.0% images and 98.3% packages reproducible in our current test framework.) Apr 07 08:24:30 Grommish: zram could help, depending on the required RAM and available CPU time. It's a textbook space-time tradeoff. Apr 07 13:20:21 nbd: I was looking at the recent mt76 commits, as I was curious about the commit which fixed the broadcast issue, but don't see anything obvious. Which one was it? :) Apr 07 13:26:45 rsalvaterra: i assume it's https://github.com/openwrt/mt76/commit/26658849a4653b69e594c4831cd219d54f1d5830 Apr 07 13:29:09 Oh, it does seem like it, thanks, blocktrron. Apr 07 13:29:30 I thought it was more recent than that. :) Apr 07 13:52:17 jow: can you take a look at the [PATCH] firmware-utils: bcm4908img: convert into bcm63xximg package Apr 07 13:52:20 please? https://patchwork.ozlabs.org/project/openwrt/patch/20210401150647.15974-1-zajec5@gmail.com/ Apr 07 13:52:43 i need that image format tool to be available for host as well as target Apr 07 13:53:08 so even though it's a firmware-util, I decided to make it a package Apr 07 13:53:10 is that OK? Apr 07 14:34:08 rsalvaterra: that commit is an important part, yes. i think this one is important too, at least if you occasionally scan on the ap interface: https://github.com/openwrt/mt76/commit/bf6d9ee4acd1aad05509190b3f67ab4eec972cff Apr 07 16:13:23 could someone please take a look at https://github.com/openwrt/openwrt/pull/3934 ? it fixes external kernel builds. Apr 07 18:09:14 Can someone check my syntax? Can I use !@(SMALL_FLASH || LOW_MEMORY_FOOTPRINT) for a package DEPENDS Apr 07 18:10:17 That would be a no :D Apr 07 18:13:15 @!SMALL_FLASH @!LOW_MEMORY_FOOTPRINT seems to work.. Anyone have a cleaner way? Apr 07 19:08:12 Hauke: I take from the people on the netdev list that GSWIP_MII_CFG_RGMII_IBS is something that needs testing first. are you fine with me removing it? Apr 07 19:08:47 Hauke: (I am happy with it, I'd keep the definition but remove the usage of GSWIP_MII_CFG_RGMII_IBS) Apr 07 19:18:39 xdarklight: yes I am fine with removing it Apr 07 19:18:51 I just adde it becasue I saw it in the spec Apr 07 19:32:02 Hauke: ok, sure. I'll ask one more question on the list and then re-spin tomorrow Apr 07 20:08:04 cd /2 **** ENDING LOGGING AT Thu Apr 08 02:59:57 2021