02:27 fdobridge_: <S​id> possibly, since all my testing before the recent patchset just ended up in system-wide freezes
02:45 wz: hello everyone
02:46 wz: I had a problem。 I try to trace cuda "cuLaunchKernel" MMIO, but there is no log in mmiotrace.log when invoke cuLaunchKernel, why?
02:46 wz: can everybody hear me?
02:49 fdobridge_: <S​id> messages are coming through, yes
02:50 wz: thanks. Who can help me?
02:50 fdobridge_: <S​id> however I have very little knowledge about what you're looking for ^^'
02:52 wz: thanks
02:55 wz: Or, it said Mmiotracer can not trace Low ISA range tracing, apply 0001-ioremap-do-not-handle-the-low-ISA-range-specially.patch to kernel will resolve this. But I can't open the address supported by MmioTracerDeveloper page. Who can give me one? please
17:13 fdobridge_: <S​id> tried mesa from 14th december, no DEVICE_LOST, but also game didn't load into main menu
17:14 fdobridge_: <S​id> time to bisect to find out when we get our device_lost
17:14 fdobridge_: <S​id> (by "from 14th dec" I mean at the last commit made on that date)
17:20 fdobridge_: <S​id> oh btw whenever game fails to load main menu there's a `job killed` in the dmesg
17:25 fdobridge_: <S​id> dec 19th: device_lost
17:49 fdobridge_: <S​id> ...nvm I think I goofed something up because tired and sleepy
18:04 fdobridge_: <S​id> basically I was testing with dxvk 2.3 for newer versions and went back to dxvk 1.10.3 for the older versions of mesa
18:05 fdobridge_: <S​id> now I'm not sure if older versions of dxvk have this logline
18:05 fdobridge_: <S​id> `err: DxvkSubmissionQueue: Command submission failed: VK_ERROR_DEVICE_LOST`
18:06 fdobridge_: <S​id> ah, they do not
18:06 fdobridge_: <S​id> so I was probably running into device_lost even on the 14th dec build
18:07 fdobridge_: <S​id> meaning I just wasted an hour :D
18:08 fdobridge_: <S​id> however someone I know pointed out how this is likely a kernel error, since, and I quote, `job timed out, channel killed, get ENODEV`
18:08 fdobridge_: <S​id> `you can't trace further without an understanding of hardware`
18:10 fdobridge_: <S​id> anyway, I think I'm gonna head to bed.. head hurts atm
18:10 fdobridge_: <S​id> happy holidays frendos
18:11 fdobridge_: <!​DodoNVK (she) 🇱🇹> 🍾
18:11 fdobridge_: <S​id> oh, new loglines with the patchset on 6.7-rc7 btw
18:11 fdobridge_: <S​id> ```
18:12 fdobridge_: <S​id> [ 4279.669507] nouveau 0000:01:00.0: Metal.exe[67432]: job timeout, channel 24 killed!
18:12 fdobridge_: <S​id> [ 4288.652786] nouveau 0000:01:00.0: gsp: rc engn:00000001 chid:24 type:13 scope:1 part:233
18:12 fdobridge_: <S​id> [ 4288.652795] nouveau 0000:01:00.0: fifo:001001:0003:0018:[Metal.exe[67432]] errored - disabling channel
18:12 fdobridge_: <S​id> ```
18:12 fdobridge_: <S​id> ```
18:12 fdobridge_: <S​id> [ 1826.876187] nouveau 0000:01:00.0: Metal.exe[21405]: job timeout, channel 24 killed!
18:12 fdobridge_: <S​id> [ 1833.570140] nouveau 0000:01:00.0: gsp: mmu fault queued
18:12 fdobridge_: <S​id> [ 1833.570166] nouveau 0000:01:00.0: gsp: rc engn:00000001 chid:24 type:31 scope:1 part:233
18:12 fdobridge_: <S​id> [ 1833.570175] nouveau 0000:01:00.0: fifo:001001:0003:0018:[Metal.exe[21405]] errored - disabling channel
18:12 fdobridge_: <S​id> ``` (edited)
22:36 xilynx: hi
22:37 xilynx: hey guys! Recently I have tryed nouveau driver with a corebooted hp 8300 cmt (seabios) board, and the thing is that proprietary nvidia drivers work, but noveau doesn't. Does someone have a knowledge about what could be the problem? Does coreboot/nouveau conflict in some way? I have gtx780 (kepler)
23:20 fdobridge_: <a​zkali> Little heads up regarding mainline on NX (Switch) I have a full KDE build ready in my pipeline. The port is in very good shape. I've been adding support for v2 switch (Mariko) good news is kernel boots and doesn't break the device lol, bad news is that it hangs very early in kernel. I am currently starting to do some light test of NVK but as it's been said here, it's still early days for Maxwell support.
23:20 fdobridge_: <a​zkali> I've recently rebased on v6.6. looking forward to future progress
23:21 fdobridge_: <a​zkali> https://cdn.discordapp.com/attachments/1034184951790305330/1188984900263620668/rn_image_picker_lib_temp_33e973c2-b2ab-4739-acce-85aa8d538ea7.jpg?ex=659c837a&is=658a0e7a&hm=c37a2220950afc5f8cc0c730d2589b1b5ea19dbd2540f1bc411737a508ef6184&
23:22 fdobridge_: <a​zkali> Beside the whole audio stack all the hw is supported (on v1) some bugs remain though
23:24 fdobridge_: <a​zkali> I'd need to debug the super high power consumption, but basically that means fixing DVFS in mainline. Idk if those changes will be accepted soon or not. My current plan is to boot v2 fully and then start upstreaming my patches
23:25 fdobridge_: <a​zkali> [source](https://gitlab.azka.li/l4t-community/kernel/mainline/linux)
23:25 fdobridge_: <a​zkali> [kernel source](https://gitlab.azka.li/l4t-community/kernel/mainline/linux) (edited)