I noticed the last few times I logged in that the desktop environment took forever to load. I wasn’t timing it, but I would say 60 seconds or more.
I had not encountered this problem before, but now it was every time, so I looked at dmesg to see what the heck it was doing.
This block was repeated dozens of times, to the point that I couldn’t scroll back farther to see anything else in the output to the terminal:
[ 134.436332] panfrost ff9a0000.gpu: js fault, js=0, status=JOB_BUS_FAULT, head=0x128fe6c0, tail=0x128fe6c0
[ 134.940112] panfrost ff9a0000.gpu: gpu sched timeout, js=0, config=0x3b01, status=0x8, head=0x129226c0, tail=0x129226c0, sched_job=00000000f67a8357
[ 134.952054] panfrost ff9a0000.gpu: Unhandled Page fault in AS1 at VA 0x00000000134172B8
Reason: TODO
raw fault status: 0x61002C3
decoded fault status: SLAVE FAULT
exception type 0xC3: TRANSLATION_FAULT_3
access type 0x2: READ
source id 0x610
The last thing in dmesg is this:
[ 134.952126] panfrost ff9a0000.gpu: GPU Fault 0x00000080 (DELAYED_BUS_FAULT) at 0x00000000145fb580
[ 134.952147] panfrost ff9a0000.gpu: There were multiple GPU faults - some have not been reported
[ 134.952218] panfrost ff9a0000.gpu: js fault, js=1, status=JOB_BUS_FAULT, head=0x10360dc0, tail=0x10360dc0
The only other message visible in the output is this, but I had seen this set of messages before I was getting all the panfrost faults:
[ 129.340385] brcmfmac mmc0:0001:1: loading /lib/firmware/6.16.0-1-MANJARO-ARM/brcm/brcmfmac43456-sdio.clm_blob failed with error -20
[ 129.460368] brcmfmac: brcmf_c_process_txcap_blob: no txcap_blob available (err=-2)
[ 129.460928] brcmfmac: brcmf_c_preinit_dcmds: Firmware: BCM4345/9 wl0: Feb 11 2020 11:54:51 version 7.45.96.61 (be7af2d@shgit) (r745790) FWID 01-a41d86bd es7.c5.n4.a3
I am on branch arm-unstable, and I am up to date, except for the wpa_supplicant package, which is v2.10. At one point I reverted that package and added to IgnorePkg because my WiFi stopped working after an update.