04:53 <gwolf> steev: As usual, you got it right ☺ Just rebuilt tqftpserv omitting the patch, and it works correctly. And, of course, my battery reporting is fine again. Yay! 😃

04:54 <gwolf> So I can finally use my laptop with a 100% upstream kernel! (as long, of course, as I don't expect working audio or external HDMI... but well, modified 5.19 and a USB audio dongle are still there for my eventual needs)

05:08 <steev> gwolf: mind filing a bug about it? :D

05:09 <gwolf> did so already :-)

05:09 <steev> oh nice

05:09 <gwolf> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1082675

05:19 <steev> thanks :) chimed in as well

06:11 <sibis> anonymix007[m]: yeah I was told like it would operate like any other external devices that qualcomm sells

06:11 <sibis> anonymix007[m]: i.e. the debug features are disabled, so in a sense it is a secure device

06:13 <macc24> kuruczgy: there might be an EC somewhere that has to have the lid switch stuff enabled first :/

06:13 <anonymix007[m]> sibis: Thanks! Well, that's unfortunate.

06:50 <macc24> wait

06:50 <macc24> is lenovo forcing high performance on cinebench?

06:50 <macc24> https://imgur.com/a/yYOnZSZ

06:50 <macc24> i was looking through oem142.inf

06:51 <macc24> and found /that/

06:56 <HdkR> Better than phone manufacturers pushing clocks higher than advertised clocks only in benchmarks

06:57 <HdkR> At least this is likely to give numbers that are consistent :P

06:58 <HdkR> Only so much performance can be pulled out of these Oryon cores when they are limited to 3.4Ghz

06:58 <macc24> arm overclocking >:3

06:59 <HdkR> Would be cool, but not common in ARM space

08:14 <macc24> how do acpi interrupts even work? why do i see GpioInt referencing gpios that don't even exist??

08:20 <JensGlathe[m]> I'm not the only one with this question

08:21 <Jasper[m]> <macc24> "how do acpi interrupts even work..." <- They are mapped in the windows driver

08:21 <Jasper[m]> pre xelite they can be calculated back, on xelite you need to do some different magic that Konrad wrote a script for

08:27 <macc24> do you have a link to that script?/

08:28 <Jasper[m]> macc24: https://github.com/linux-msm/acpi_tools/blob/master/GIO_CIPR_decoder.py this supposedly

08:28 <macc24> thanks

09:02 <lumag> steev, gwolf thanks for tracking the tqftpserv story!

09:11 <konradybcio> kuruczgy: do you have `pinctrl: qcom: x1e80100: Bypass PDC wakeup parent for now`?

09:13 <konradybcio> Jasper: macc24 fwiw this script is for building wakeup gpio maps in the pinctrl driver

09:14 <Jasper[m]> Ah yeah

09:15 <macc24> konradybcio: :|

09:15 <macc24> konradybcio: i have that commit and the power button doesn't work

09:16 <konradybcio> yoga?

09:16 <konradybcio> does it work without the commit?

09:16 <macc24> yes, yoga

09:16 <macc24> and lemme check

09:17 <macc24> wait s/power button/lid switch/

09:37 <macc24> konradybcio: without 'pinctrl: qcom: x1e80100: Bypass PDC wakeup parent for now' i see no difference in either /sys/kernel/debug/gpio output when comparing open/closed configurations and evtest doesn't see any events from the gpio-keys device :/

09:37 <konradybcio> did you configure the gpio properly with a pinctrl-0/pinctrl-names entry?

09:39 <macc24> i think so, this is the diff https://bpa.st/7LN2C

09:39 <konradybcio> ok you didn't configure it as output, that's good

09:40 <macc24> it's mostly stolen from x13s dts file, with gpio number changed to reflect what i saw in dsdt

09:40 <konradybcio> and the dsdt suggests no pull-up is disabled as well

09:40 <konradybcio> so.. it may be that there's some wild regulator required, or as someone mentioned, ec interaction..

09:41 <konradybcio> yeah, look up Debug = "RCVD EC INT" in your dsdt, there's a "ping lid0 object" line near it

09:42 <konradybcio> or hm no, that's unrelated

09:58 <macc24> there's also some code that talks to ec and in there i see a Debug = "EC enable/disable sent", can that be related?

10:05 <konradybcio> i wouldn't rule that out

12:54 <macc24> riiiight, the ec has an interrupt pin "0x0140" in acpi, which would be 320 and i don't see any mention of that gpio :/

12:54 <macc24> ^ in yoga slim 7x

12:56 <travmurav[m]> sounds like it's not at all gpio 320

12:59 <travmurav[m]> macc24: gpio66 I believe

13:01 <macc24> how did you-

13:02 <travmurav[m]> just need to translate the value 5 times through a bunch of lookup tables scattered across the world

13:03 <travmurav[m]> virtual gpio number -> raw interrupt number -> GIC SPI interrupt number -> PDC line number -> TLMM gpio number

13:03 <macc24> o_o thanks

14:25 <macc24> konradybcio: it probably wasn't it

14:25 <macc24> talked to ec the 'enable stuff' stuff and it didn't make any differernce :/

14:28 <abelvesa> somebody asked me not long ago about the USB MP on t14s. I don't remember who it was exactly, but if you are still around, checkout this branch:

14:28 <abelvesa> https://git.codelinaro.org/abel.vesa/linux/-/commits/x1e-next-20240924

14:30 <JensGlathe[m]> I guess that was me, thanks

14:31 <abelvesa> and if you see the following error by any chance, ignore it as it is unrelated to MP (it's the USB2 controller):

14:31 <abelvesa> [ 182.205989] usb usb5-port1: couldn't allocate usb_device

14:32 <JensGlathe[m]> Is it this EUSB2 odd thing I don't understand

14:32 <abelvesa> I enabled the usb_2 and it's phy as I thought there would be some device on it, but it doesn't work properly yet

14:34 <abelvesa> no, the eUSB2 is basically just the controller and PHY working at 1.2V and then needing a level shifter to interface with the connector

14:35 <abelvesa> that's where the repeater (smb2360) or the redriver (ptn2333) come into place

14:35 <abelvesa> starting with SM8550, eUSB2 is the only option, therefore the repeaters/redrivers are needed

14:36 <tobhe_> abelvesa: thx! will check it out

14:38 <tobhe> abelvesa: not sure if you had seen that, I had to revert the external DP dtb commits to get the 120hz OLED working with the last branch

14:40 <abelvesa> depending on your x1e device, it might have 3 or 4 smb2360 repeaters and then 2 or 3 ptn3222 redrivers

14:40 <abelvesa> tobhe: hm, would be interesting to find out why that is

14:42 <anonymix007[m]> abelvesa: should picking 1fbd18e7..7262a908 to jhovold's 6.11 branch be enough?

14:43 <abelvesa> anonymix007[m]: I think so

14:44 <JensGlathe[m]> In my case it was mdss_dp2 being activated but not physically wired up (?) The HP X14 has 2 type-c ports and 1 type-a (on the usb3-mp controller). Removing mdss_dp2 from the dts fixed the USB4 retimer issue (as in having no display at all). It now boots up with display again. Need to test dp altmode, but didn't have the time yet.

14:45 <anonymix007[m]> abelvesa:... (full message at <https://matrix.org/oftc/media/v1/media/download/ATfcctU_tSaOCUapvejVZu52WJRNgC7aWmW6xhi54U8Rew08ceK43GiDcFN1INoNjL5ThnktxqsfHna6n6ceBLdCeSK1zT4gAG1hdHJpeC5vcmcvYXFqU09rbmNWRGZXa3ZEc0RIbWdkS1Zn>)

14:46 <tobhe> abelvesa: I will try to do a little more research, just thought I'd bring it up :)

15:29 <abelvesa> tobhe: might be. Once I have that usb_2 working properly we should be able to detect it.

18:48 <MrCatfood[m]> i changed the dtb enabled gpu, and added zap shader to /lib/firmware/qcom/gen70500_zap.mbn but i getting this error: https://pastebin.com/ikjLUia9

18:48 <MrCatfood[m]> I'm not sure if i missed some step

18:52 <JensGlathe[m]> does the zap firmware match your device? Usually it is signed by device, and should be obtained from Windows. On windows the name is usually qcdxkmsuc8380.mbn.

18:52 <MrCatfood[m]> ah thats what i missed

18:56 <MrCatfood[m]> i have one in sys32 and one in driverstore, does it matter which one?

18:56 <JensGlathe[m]> driverstore

19:04 <kuruczgy[m]> konradybcio: Yes, `pinctrl: qcom: x1e80100: Bypass PDC wakeup parent for now` has beeen merged by Linus in 6.11-rc7

19:06 <kuruczgy> macc24: What do you mean by "talked to ec the 'enable stuff' stuff"? Is there a driver that can be used to talk to the EC?

19:44 <MrCatfood[m]> progress... (full message at <https://matrix.org/oftc/media/v1/media/download/AWyr73c1NzN6l9Jkbc0w-B2XcOZznsxbyi0PCrwOKzSHvx7dgZkTlBqKd3TgwowCUihnL-dMraYapNbD5CwmaQtCeSLG47SQAG1hdHJpeC5vcmcvamtia1RMQmVhekhsTlVNWFBOdXpyQUtX>)

19:46 <robclark> jfwiw you probably want to have mesa 24.2.x or ToT.. and then you can drop the hack kernel patch to hard-code the gpu-id as a740

20:37 <steev> does anyone else have a weird issue where the keyboard stops up arrowing randomly?

20:37 <steev> specifically the up arrow though