00:04 <owrt-snap-builds> Build [#221](https://buildbot.openwrt.org/master/images/#builders/43/builds/221) of `oxnas/ox820` completed successfully.

00:09 <owrt-snap-builds> Build [#219](https://buildbot.openwrt.org/master/images/#builders/36/builds/219) of `mediatek/mt7629` completed successfully.

00:21 <owrt-snap-builds> Build [#217](https://buildbot.openwrt.org/master/images/#builders/55/builds/217) of `malta/be` completed successfully.

00:29 <owrt-snap-builds> Build [#221](https://buildbot.openwrt.org/master/images/#builders/33/builds/221) of `ipq806x/generic` completed successfully.

00:39 <Slimey> FATAL ERROR: Couldn't open "fsl/p1010si-pre.dtsi": No such file or directory

00:40 <owrt-snap-builds> Build [#221](https://buildbot.openwrt.org/master/images/#builders/60/builds/221) of `mpc85xx/p1010` completed successfully.

00:43 <Tusker> Is there a way in DTS to set a register on the phy gmac ? It seems that there is a register difference between when I ping in uboot vs not pinging in uboot, so wanted to try to set that register value in DT

00:46 <Tusker> https://pastebin.com/3WSpWY3x shows the difference, where 3800 != 7800 ... no clue what the register signifies

00:48 <Tusker> #define AR8035_GBCR 0x09

00:48 <Tusker> whatever GBCR is

00:50 <Tusker> ah, 1000base-t control register ... wonder what 7800 means that 3800 doesn't

00:58 <owrt-snap-builds> Build [#217](https://buildbot.openwrt.org/master/images/#builders/17/builds/217) of `ramips/rt305x` completed successfully.

00:58 <Tusker> nope, incorrect, it is 0x0A, which is 1000 BASE-T Status

01:07 <Tusker> difference between 3800 and 7800 looks like "remote receiver status"...

01:08 <Tusker> i think it's a red herring, eth1 has 3800 too

01:20 <owrt-snap-builds> Build [#233](https://buildbot.openwrt.org/master/images/#builders/65/builds/233) of `archs38/generic` completed successfully.

01:40 <owrt-snap-builds> Build [#216](https://buildbot.openwrt.org/master/images/#builders/42/builds/216) of `ramips/mt76x8` completed successfully.

02:00 <owrt-snap-builds> Build [#259](https://buildbot.openwrt.org/master/images/#builders/7/builds/259) of `armvirt/32` completed successfully.

02:02 <owrt-snap-builds> Build [#271](https://buildbot.openwrt.org/master/images/#builders/1/builds/271) of `ath79/generic` completed successfully.

02:06 <owrt-snap-builds> Build [#259](https://buildbot.openwrt.org/master/images/#builders/3/builds/259) of `at91/sam9x` completed successfully.

02:13 <owrt-snap-builds> Build [#215](https://buildbot.openwrt.org/master/images/#builders/62/builds/215) of `tegra/generic` completed successfully.

02:51 <owrt-snap-builds> Build [#217](https://buildbot.openwrt.org/master/images/#builders/29/builds/217) of `pistachio/generic` completed successfully.

02:56 <owrt-snap-builds> Build [#219](https://buildbot.openwrt.org/master/images/#builders/70/builds/219) of `bcm47xx/mips74k` completed successfully.

02:57 <digitalcircuit> I thought I'd be clever by trying an older SSD (Samsung 850 Pro) connected via a USB SATA case to my NBG6817 router - contrary to expectations, the SSD is actually more power hungry (in bursts) than the HDD, and within an hour of trying to backup, I've corrupted the ext4 filesystem on the SSD :D

02:57 <digitalcircuit> So.. I might try getting a powered USB hub after all with the USB 3.0 HDD, alongside looking at how to tinker with the DTS voltage levels. (The USB HDD is having trouble staying connected to my desktop/laptop directly, possibly from a fiddly microUSB 3.0 connector, so I feel like I either need to move it or replace it eventually anyways...)

03:07 <slh> good luck! (I'm just having fun with an ASRock G10, which also has a number of issues (just arrived a few hours ago, so I don't even have an idea about its stability yet), with the DSA PR and maybe also kernel 5.10 and its NAND, well, need to test v5.4 before claiming that, but there's something fishy, or the second hand device is on the brink of dying)

03:08 <owrt-snap-builds> Build [#218](https://buildbot.openwrt.org/master/images/#builders/14/builds/218) of `bcm63xx/generic` completed successfully.

03:30 <owrt-snap-builds> Build [#225](https://buildbot.openwrt.org/master/images/#builders/13/builds/225) of `lantiq/xway_legacy` completed successfully.

04:39 <wulfy23> digitalcircuit: have you ran these tests with wifi off?

04:39 <wulfy23> the fact that OEM makes such a prominent 'disable-usb3' tunable is ominous

04:43 <digitalcircuit> wulfy23: Not yet, good point. My desktop is situated somewhere I can't easily run a cable to it, but I can move my laptop around.

04:43 <digitalcircuit> My laptop has generally connected to a second access point (ASUS RT-AC68U) that's connected to the NBG6817 via Ethernet, so it at least doesn't seem to be direct. But I'll still keep in mind trying the SFTP-to-USB drive test with disabling the WiFi radios (presumably freeing up power and/or stopping other devices connecting).

04:43 <digitalcircuit> Trying using the USB 2.0 port instead is another option, too.

07:28 <fda> my avm7320 does no longer boot with a new checkout due to kernel panic https://pastebin.com/9PVfKtCX

07:29 <fda> trace shows BUG kmem_cache_free_bulk

07:54 <owrt-snap-builds> Build [#225](https://buildbot.openwrt.org/master/images/#builders/22/builds/225) of `ipq40xx/generic` failed.

07:58 <fda> edb6bc199049769be76bb3966debcc2b31511e4b works

08:21 <fda> blocktrron_: ^

10:08 <owrt-snap-builds> Build [#221](https://buildbot.openwrt.org/master/images/#builders/56/builds/221) of `kirkwood/generic` completed successfully.

10:27 <Tusker> I'm a bit stuck... how do I figure out what the factory firmware does to activate the gmac0 on the ipq8068 ? The gpio all look good now between working state and non-working state, and the registers only look slightly different in the 1gbs link status (3800 vs 7800). Am a bit lost :) any ideas ?

10:33 <Tusker> there is a bit of oddness around mii-tool output vs ethtool output. mii-tool eth0 shows 10baseT-FD no link, when I turn autoneg off with ethtool... but shows 1000baseT-FD flow-control, link ok if I turn autoneg on

11:59 <Tusker> looking at the factory firmware, the only thing it does weird is set the PLL RATE manually in the /etc/rc.d/network script --- echo 733000000 > /proc/sys/dev/nss/clock/current_freq --- which seems to relate to NSS_PLL_RATE(733000000, 58, 16, 25, 0x014b5625), but that might only be related to NSS, not without NSS...

12:04 <Tusker> anyway, time to shutdown the PC for the weekend :) any ideas would be appreciated, and I'll check the irc logs :) have a good weekend everyone

14:09 <ttocsr> blocktrron I'm having issues with your recent commit to ethtool

14:11 <ttocsr> ethtool-full has a recursive dependency

14:12 <owrt-snap-builds> Build [#226](https://buildbot.openwrt.org/master/images/#builders/13/builds/226) of `lantiq/xway_legacy` failed.

18:45 <owrt-snap-builds> Build [#226](https://buildbot.openwrt.org/master/images/#builders/22/builds/226) of `ipq40xx/generic` completed successfully.

19:18 <stintel> sooo ... ARM-based QorIQ devices are already supported in the Layerscape target, the mpc85xx is not really appropriate for the ppc64-based QorIQ devices, so what about a new target named qoriq

19:18 <stintel> I think I will go with that, and send out a first RFC this weekend

21:15 <agb[m]> +1 for a qoriq target

21:25 <stintel> agb[m]: alright :) shall I CC you when sending my series ?

21:56 <blocktrron_> fda: context?

23:10 <fda> blocktrron_: its serial console output 2 seconds after kernel start