00:34 <KGB-2> https://tests.reproducible-builds.org/openwrt/openwrt_omap.html has been updated. (11.1% images and 100.0% packages reproducible in our current test framework.)

06:14 <owrt-images-builds> Build [#40](https://buildbot.openwrt.org/images/#/builders/151/builds/40) of `master_layerscape/armv7` completed successfully.

09:36 <rmilecki> after month with no complains / regressions reports I'm backporting base-files changes for postinst to 23.05

09:36 <rmilecki> it's needed to fix uhttpd invalid behaviour & regression

13:18 <schmars[m]> x86/64 snapshot imagebuilder broken at the moment, would have needed a packages rebuild first (libubox update)

16:04 <hauke> Ansuel: did you activate some setting to make dependabot create pull requests

16:05 <Ansuel> hauke yes it's just for the actions... having hard time making the bot regenerate with the correct commit title prefix tho...

17:06 <mrkiko> Ansuel: robimarko: in my opinion the 10g port works, and the wan surely does - don't remember where aqr is

17:16 <robimarko> mrkiko: I am not following, what device?

17:19 <mrkiko> robimarko: nbg7815

17:20 <rmilecki> nbd: do you know why Mediatek has "WiFi7 Non-MLO SDK Release" and "Beta Release" for Filogic 880?

17:20 <rmilecki> is MLO support still not ready?

17:21 <mrkiko> robimarko: and I am pretty sure I've seen the "Programm phy......" thing in u-boot

17:21 <mrkiko> robimarko: just not sure if it does so even when I didn't interrupt u-boot. But i can assure that 10g port worked fine

17:22 <mrkiko> robimarko: and following the related forum thread I seen no complaints - just people testing it and seeing it doesn't reach 10G kinda

17:22 <mrkiko> so - maybe I used it when it wasn't broken and then got broken, but I see no posts

17:36 <robimarko> mrkiko: OK, I dont have that device

17:41 <mrkiko> robimarko: I have one and have UART access - if needed i can set up a remote access for you to test things out.

17:42 <robimarko> No, need Ansuel has one as well

17:44 <mrkiko> robimarko: infact i was wondering if there are different u-boot versions doing things differently even tough I know this is really improbabl3

17:44 <mrkiko> *improbable*

17:51 <Ansuel> mrkiko no the fw is loaded by uboot only if boot is interrupted. (but i have my nbg with a modified uboot so my router is not the 100% common thing)

17:51 <Ansuel> no idea if the zyxel 4th bootloader load the aquantia firmware (but from what i remember it didn't)

18:24 <mrkiko> Ansuel: oh, so you got rid of the horrible zloadr thing?

18:24 <Ansuel> yep

18:24 <Ansuel> i need the thing to develop... having to regenerate and unlock uboot everytime was too much

18:25 <Ansuel> the thing have spi so could experiment a bit with custom uboot... i plan to make it even better just to have some fun with it

18:39 <mrkiko> Ansuel: well - if you can build something usable I might also flash one of mine with it to get rid of zloadr. When I was trying out the fan thing I actually unlocked it several times, yes

18:40 <mrkiko> Ansuel: I mean - I had to enter pw several time

18:47 <Ansuel> problem is that i assume the recovery feature doesn't work anymore (as i think it was handled by the zyloader)

20:26 <xdarklight> Ansuel: robimarko: I think you two are the most likely developers to know about this - I have an ipq40xx Fritz Box 7520 (hardware is identical to 7530) and with Linux 6.1 I have a strange issue where traffic within LAN is limited to ~10Mbit/s. I have one IoT device which only has a 10Mbit/s port, if I unplug that then my 3 other devices (all with Gbit/s ports) can RX/TX at ~700-800Mbit/s (tested using iperf3). with Linux 5.15 the Gbit/s

20:26 <xdarklight> devices can RX/TX at ~950Mbit/s regardless of the IoT device being plugged in or not. is this a known issue and if not: do you have a hint where I should start debugging?

20:27 <Ansuel> disable pause frame and report back

20:27 <xdarklight> on all ports or just the IoT device one?

20:43 <xdarklight> Ansuel: https://pastebin.com/RA9qQrMg - according to ethtool it's off on all ports for both, RX and TX

20:45 <xdarklight> Ansuel: also "ethtool -S lan1 | grep -i pause" (and repeating that for lan2, lan3, lan4) shows zero RxPause and TxPause frames after an iperf3 run