00:07 fdobridge: <j​ekstrand> PSA: I'm rebasing the nvk/main branch on main to pick up the new Vulkan header
04:32 fdobridge: <j​ekstrand> Rebase completed a few hours ago
21:42 rmckeever: skeggsb_ I'm having some trouble running Vulkan CTS tests, maybe you can help. I'm using deqp-runner to run all the tests. After running `sudo service lightdm stop`, the program is not freezing up any more, but several minutes into the tests deqp-runner just starts showing this:
21:42 rmckeever: Pass: 39797, Fail: 207, Crash: 64, Warn: 1, Skip: 236888, Timeout: 7, Missing: 6007, Flake: 29, Duration: 12:18, Remaining: 58:54
21:42 rmckeever: ERROR - dEQP error: FATAL ERROR: vk.enumeratePhysicalDevices(instance, &numDevices, DE_NULL): VK_ERROR_INITIALIZATION_FAILED at vkQueryUtil.cpp:81
21:42 rmckeever: ERROR - Failure getting run results: No results parsed. Is your caselist out of sync with your deqp binary? (See "tests2/c559.r3.log")
21:42 rmckeever: Pass: 39798, Fail: 207, Crash: 64, Warn: 1, Skip: 236892, Timeout: 8, Missing: 6501, Flake: 29, Duration: 12:20, Remaining: 58:56
21:42 rmckeever: ERROR - dEQP error: FATAL ERROR: vk.enumeratePhysicalDevices(instance, &numDevices, DE_NULL): VK_ERROR_INITIALIZATION_FAILED at vkQueryUtil.cpp:81
21:42 rmckeever: ERROR - Failure getting run results: No results parsed. Is your caselist out of sync with your deqp binary? (See "tests2/c567.r1.log")
21:42 rmckeever: Pass: 39798, Fail: 207, Crash: 64, Warn: 1, Skip: 236893, Timeout: 8, Missing: 7000, Flake: 29, Duration: 12:22, Remaining: 58:58
21:42 rmckeever: ERROR - dEQP error: FATAL ERROR: vk.enumeratePhysicalDevices(instance, &numDevices, DE_NULL): VK_ERROR_INITIALIZATION_FAILED at vkQueryUtil.cpp:81
21:42 rmckeever: ERROR - Failure getting run results: No results parsed. Is your caselist out of sync with your deqp binary? (See "tests2/c555.r3.log")
21:42 rmckeever: Pass: 39837, Fail: 207, Crash: 64, Warn: 1, Skip: 236974, Timeout: 9, Missing: 7379, Flake: 29, Duration: 12:24, Remaining: 59:00
21:49 airlied: rmckeever: anything in dmesg?
21:52 rmckeever: I'll check
21:57 rmckeever: [ 1706.775666] nouveau 0000:01:00.0: fifo: PBDMA0: 80000000 [] ch 14 00003474 f294240c
21:57 rmckeever: [ 1706.775917] nouveau 0000:01:00.0: fifo: PBDMA0: 80000000 [] ch 14 00003474 f294240c
21:57 rmckeever: [ 1706.775985] nouveau 0000:01:00.0: fifo: PBDMA0: 80000000 [] ch 14 00003474 f294240c
22:02 airlied: ah yeah are you running the 06.00 kernel?
22:05 rmckeever: It's a custom kernel from https://gitlab.freedesktop.org/skeggsb/nouveau/-/tree/00.06-gr-ampere
22:06 airlied: ah cool
22:06 airlied: yeah I think the fifo PBDMA0 is the death knell we don't alwasy recover from
22:10 rmckeever: Maybe it's the power manager. I'll try turning it off.
22:26 rmckeever: Turning off power manager didn't help.