Dave Airlie

Okay I have llama.cpp coopmat2 paths running on radv for the first time, they are 10 times slower that coopmat but gotta start somewhere

there comes a time around 100,000 SSA indexes where inline the shit out of it all starts to become questionable.

I have a lot of laptops, one of the turing ones, Lenovo p53 the display stopped turning on yesterday, I decided to open it up and battery was swelling bad, just dropped the battery off at a disposal place, but the laptop is pretty toast. Can't see a dislodged cable, but I probably need an ifixit kit to dig deeper.

took a wrong turn and now I'm in the mesa vulkan meta blit/resolve paths and wondering if I should just back away slowly.

@never_released@mastodon.social do I want to ask where you dug that up?

@never_released@mastodon.social it was the best of times, it was the worst of times.

@hrw@society.oftrolls.com thankfully this all hidden in fw

HDMI FRL support on nouveau, please test if you have HDMI2.1 and high refresh monitor https://lore.kernel.org/nouveau/20260423004552.3289884-1-airlied@gmail.com/T/#t

Reached the point of development cycle with nouveau on the spark where I want to throw it through the window, so I should be pretty close to getting display lit up

deviceName = NVK GB20B

though CTS is a failfest due to some sync or coherency issue, and display isn't working yet.

sometimes it's easier to just add new printks than to work out how to switch on the pre-existing ones that are buried in driver macros and dynamic debug.

like I'm co-running a 300 person organization, like wtf

Ran some numbers for upcoming recording
git log --format='%aN' v-1..v drivers/gpu/drm/ include/uapi/drm/ include/drm/ | sort -u | wc -l

v6.15 326
v6.16 322
v6.17 300
v6.18 334
v6.19 332
v7.0-rc6 346

I was actually surprised it was this many, my mental model has always been around 100 contributors a cycle.

once again I find myself debugging code written by a version of me who seems to have a superior intellect.

https://airlied.blogspot.com/2026/02/drm-subsystem-ai-patch-review.html vibe coding an AI patch reviewer

@davidculley@hachyderm.io @benaaasaaas@mastodon.social @fabio@manganiello.eu I've swapped GPUs 1000s of times with no strap and on carpet, never seen a problem. I think modern electronics are just less fryable. I've worn out a PCIe slot level of swapping. Maybe Ive taken out the ethernet on board, but I think I knocked off a capacitor.

@josefbacik@hachyderm.io I also think in open source community we are more open to people with a wider range of mental health statuses than your average US-owned employers, so that filter of hiring/HR weeds out some people. I think this benefits open source projects in a lot of ways, and often allows voices to be heard that wouldn't normally. But there is also the effect of allowing the "technically correct" assholes to find an initial platform for their work, that they might not get in corporate spaces.