**** BEGIN LOGGING AT Sat Aug 10 03:00:09 2019
Aug 10 05:34:20 <russell--> "05:45  ynezz: karlp: cleanup, it makes no sense to build it every time in the master, just because it might be used in few packages in packages feed" fwiw, scons takes 11 seconds to build
Aug 10 09:44:45 <ynezz> russell--: and another 60s to handle every bug report
Aug 10 10:27:46 <gch981213> mangix: Is that an ath79 builtin switch port? Try commenting out line 915 and line 981 of target/linux/generic/files/drivers/net/phy/ar8216.c (two lines with comments saying "disable port 5")
Aug 10 11:06:15 <DonkeyHotei> [Fri 2019-07-26 01:51:55 PM PDT]  <DonkeyHotei> gch981213: in commit a9360452, where you introduced target/linux/ath79/base-files/etc/init.d/bootcount, you used "boot()" instead of "start()" -- why?
Aug 10 11:06:16 <DonkeyHotei> [Fri 2019-07-26 05:27:59 PM PDT]  <gch981213> DonkeyHotei: It only needs to be called once when init is complete and users aren't supposed to call it.
Aug 10 11:06:16 <DonkeyHotei> [Fri 2019-07-26 11:42:38 PM PDT]  <DonkeyHotei> gch981213: so does that mean all the other target/linux/*/base-files/etc/init.d/bootcount files should be changed to match?
Aug 10 11:09:13 * gch981213 is scared of making tree-wide commits
Aug 10 11:09:47 <DonkeyHotei> well yes, that was why i asked about keeping it start() rather than boot()
Aug 10 11:11:49 <DonkeyHotei> there isn't a reason ath79 should be "special" in this regard
Aug 10 11:13:18 <gch981213> DonkeyHotei: I'll send a patch changing all bootcount to use boot() instead and if no one complains in a week I'll merge it.
Aug 10 11:13:57 <DonkeyHotei> gch981213: before you do that, make sure the sourcing of /lib/functions.sh matches in all the files too
Aug 10 11:14:47 <DonkeyHotei> and there is a third change i'd nominate for that patch too:
Aug 10 11:16:30 <DonkeyHotei> since target/linux/ath79/base-files/etc/init.d/bootcount did not yet exist when this was merged, i stuffed a bootcount into target/linux/ath79/base-files/etc/board.d/02_network
Aug 10 11:17:06 <DonkeyHotei> gch981213: it should be moved, but it requires /lib/functions.sh
Aug 10 11:20:28 <gch981213> DonkeyHotei: /etc/board.d/02_network is only executed on first boot isn't it?
Aug 10 11:21:04 <DonkeyHotei> iirc yes, but it wasn't where a bootcount should go
Aug 10 11:31:06 <gch981213> DonkeyHotei: How could a bootcount work if it only executes once during the first boot, then? bootloader increases bootcount on every booting attempts and if firmware doesn't reset it, bootloader tries to replace the firmware on next boot after bootcount exceeded a certain value.
Aug 10 11:32:15 <DonkeyHotei> gch981213: iirc, target/linux/ath79/base-files/etc/board.d/02_network is on every boot. in any case, it should be in target/linux/ath79/base-files/etc/init.d/bootcount instead
Aug 10 11:32:46 <DonkeyHotei> sorry i misread the question 10 min ago
Aug 10 12:08:06 <gch981213> DonkeyHotei: patch sent. and moving bootcount in 02_network is a separated commit in my staging tree: https://git.openwrt.org/?p=openwrt/staging/981213.git;a=commit;h=bde209504ca05cc8ebec1ff89d46fbe8af7953e6
Aug 10 18:28:06 <mangix> gch981213: will try when I can. this is a qca9531 device
Aug 10 18:30:34 <mangix> I use the gmac-config swap thing to get the ethernet working
Aug 10 21:00:02 <aparcar[m]> jow: could you please have a look at the patch we discussed at hamburg? https://patchwork.ozlabs.org/patch/1115478/
Aug 10 23:12:55 <mangix> gch981213: didn't work. the connection still dies.
Aug 10 23:17:05 * nickpower[m] sent a long message:  < https://matrix.org/_matrix/media/v1/download/matrix.org/AdgpWnELTBOKnChEvJtfxmxp >
**** ENDING LOGGING AT Sun Aug 11 02:59:57 2019