**** BEGIN LOGGING AT Thu Feb 06 02:59:57 2020 Feb 06 09:43:26 Hauke: Thanks. merged Feb 06 10:09:35 fyi, kernel bumps pushed to staging. including 5.4 Feb 06 10:16:36 anyone remember off the top of their head when we got 64 bit counters for network interface bytes? Feb 06 10:17:15 not important, just got a query about some confusing interface counters on something with 200 days uptime, and I noticed that 32bit bytes is only like 4gig Feb 06 10:39:13 Hauke: the SUBDIRS patch fixed a lot of build issues already on 5.4. Thanks a lot for that! Feb 06 10:42:27 Hauke: now this one is still left: https://pastebin.com/raw/uQ3FDimA Feb 06 11:00:09 fixed that bug in my staging ^^ Feb 06 11:43:43 jow: KanjiMonster: ping Feb 06 11:52:59 xback: pong Feb 06 11:53:31 jow: quick question regarding: https://git.openwrt.org/?p=openwrt/staging/xback.git;a=blobdiff;f=package/kernel/ath10k-ct/Makefile;h=b63b68d32181ff7fed335203fd8fe42d7e02cbd6;hp=e1f92875c1c1a817ea9f739ef351154994298569;hb=e508e863f02d38733012e43fd077455a528fda82;hpb=319c47d9fd1de7be839b602fa4a645bd2bb0f259 Feb 06 11:53:56 is there any syntax possible for this var to choose a version based on used kernel version? Feb 06 11:58:00 I'd try CT_KVER="-$(if $(LINUX_5_4),5.4,4.19)" Feb 06 11:59:23 i'll give it spin :) thank you Feb 06 11:59:44 maybe it must be $(CONFIG_LINUX_5_4) Feb 06 11:59:50 try both, can't test it myself right now Feb 06 12:00:18 no worries, thanks for the info! Feb 06 13:02:51 xback: ping Feb 06 13:03:01 ldir: pong Feb 06 13:03:39 xback: in your 5.4 'fun' have you had issues with missing crypto modules? e.g. crypto/sha256-ssse3.ko Feb 06 13:04:05 ldir: I know about the splitup Feb 06 13:04:56 but didnt encounter it yet Feb 06 13:05:52 Ill probably reach that point any time soon as I'm working my way through building targets currently with all features enabled Feb 06 13:06:54 ok, what's the splitup ? Feb 06 13:07:17 oh, this is x86_64 - so for the apu2. Feb 06 13:25:42 ldir: I mean that crypto stuff has been split in several pieces Feb 06 13:26:10 so the separate pieces need to be included now, not only a generic one saying "crypto modules" :-) Feb 06 13:27:36 hurray! another target that builds with all kmods enabled (imx6) Feb 06 13:31:00 ldir: which version of the WIP patch are you trying currently? Feb 06 13:31:43 I merged a bunch of fixes from Hauke this morning and also fixed some missing x86 config files Feb 06 13:57:29 jow: CT_KVER="-$(if $(CONFIG_LINUX_5_4),5.4,4.19)" works nicely! Thanks again. (tested both ways) Feb 06 14:13:11 xback: Trying to build your 5.4 tree for ath79 results in a patch failure for "target/linux/ath79/patches-5.4/403-mtd_fix_cfi_cmdset_0002_status_check.patch". Feb 06 14:15:24 which exact device? Feb 06 14:16:23 xback: Multiple devices, TARGET_DEVICE_ath79_generic_DEVICE_ubnt_unifi and TARGET_DEVICE_ath79_generic_DEVICE_ubnt_unifiac-pro Feb 06 14:17:01 mamarley: checking now .. Feb 06 14:18:10 yes, I see it. Feb 06 14:29:02 hi all, any ideas how to build some package without rebuilding again and again of deps in container? Feb 06 14:37:19 without including all .o .la files Feb 06 14:38:58 after building app i committed changes in container and result was 3gb image, which looks to me too much Feb 06 14:56:45 xback: sorry - it all got rather busy. Feb 06 16:33:25 mamarley: I pushed a fix which is worth testing. Could you try it and provide feedback please? Feb 06 16:38:47 xback: "drivers/leds/leds-reset.c:75:14: error: too few arguments to function '__of_reset_control_get'" Feb 06 16:38:47 "drivers/leds/leds-reset.c:83:9: error: implicit declaration of function 'devm_of_led_classdev_register'; did you mean 'devm_led_classdev_register'? [-Werror=implicit-function-declaration]" Feb 06 16:39:23 aaah. I didn't compile test the whole thing yet Feb 06 16:46:19 mamarley: i already have a fix for that in my tree Feb 06 16:46:29 i'll get round to publishing it on the weekend Feb 06 16:46:39 mamarley: you can just use NULL for that new parameter Feb 06 16:46:49 it was the 2nd or 3rd Feb 06 16:47:08 and i should send that driver upstream really Feb 06 17:38:47 mamarley: I have ath79 working with 5.4 Feb 06 17:38:53 but not cleaned up Feb 06 17:39:46 xback: I would like to fix some more bugs on the weekend Feb 06 18:03:26 I actually ended up dropping that patch because my device doesn't use the driver in question. It ended up compiling, but bricked my device. Sadly I don't have serial on this device, so I won't be able to tell what happened. Sorry. Feb 06 19:27:28 xback: comment above says "match as closely as possible to whatever mac80211 backports version is being used", so selected kernel shouldn't matter? Feb 06 19:35:30 I think this PR is ready to push: https://github.com/openwrt/openwrt/pull/2756 Feb 06 19:35:59 Pleas check it over first I am know expert Feb 06 22:05:54 Hauke: Is that in the "for_koen" tree? **** ENDING LOGGING AT Fri Feb 07 02:59:57 2020