<robmur01>
FWIW I did (eventually) manage to run a 32-bit compat mesa on my RK3399 too (5.3.1 kernel), so I'm all out of ideas as to what's up with mrfixit2001's configuration :/
_whitelogger has joined #panfrost
davidlt has joined #panfrost
megi has joined #panfrost
davidlt has quit [Ping timeout: 240 seconds]
davidlt has joined #panfrost
jailbox has quit [Remote host closed the connection]
* alyssa
is now running mainline+panfrost on all her machines
<alyssa>
Dogfoodin' 24/7
<alyssa>
Feels Good (TM)
<tomeu>
robmur01: with no code changes?
<tomeu>
alyssa: cool!
<robmur01>
tomeu: all I did was rebuild the stock Arch Linux generic v8 kernel package with CONFIG_COMPAT enabled, then just chrooted into a v7 filesystem, built mesa and kmscube there, and ran
<robmur01>
other than spending most of the weekend debugging the secure memory issue that kept killing the board at the end of the ~3-hour rebuild, no drama :)
<tomeu>
that's quite cool (that it worked)
<tomeu>
wonder about something else than kmscube
<tomeu>
have been considering running nightlies on armhf just in case
<robmur01>
I can try other stuff tonight now that I have the PoC nailed down
<tomeu>
nice
EmilKarlson has quit [Read error: Connection reset by peer]
thefloweringash has quit [Write error: Connection reset by peer]
TheCycoONE has quit [Read error: Connection reset by peer]
CrystalGamma[m] has quit [Read error: Connection reset by peer]
eballetbo[m] has quit [Remote host closed the connection]
<tomeu>
bbrezillon: looks like the clear fixes break on the veyron :/
<tomeu>
ah no, you just need to remove the tests from the failures list
<bbrezillon>
tomeu: yep
<bbrezillon>
that's what I'm doing
<bbrezillon>
but it looks like the diff is not the same for t760
<tomeu>
awesome, it will be less of a hassle once we stop having failurs :p
<tomeu>
guess some bugs remain in t760 in some tests?
<bbrezillon>
oh, and we should try to add more t760 boards to the farm if possible
<bbrezillon>
since now every MR triggers a test on those boards
<bbrezillon>
tomeu: BTW, is there any way to start only panfrost jobs?
<bbrezillon>
right now I manually cancel everything and start the jobs I care about
<tomeu>
bbrezillon: does it make things any faster?
<tomeu>
I just let them run
<tomeu>
we could change the ci run policy to only run certain jobs based on the branch name
<tomeu>
could be confusing
<bbrezillon>
tomeu: does it make things faster? => a little bit
<bbrezillon>
but it takes longer than it used too when we were only testing panfrost stuff
<bbrezillon>
*used to
<tomeu>
ah yes, maybe you can try configuring your repo to use our gitlab-ci file as before? don't know if that would work
<bbrezillon>
tomeu: that means keeping a separate .gitlab-ci.yml in the tree?
<bbrezillon>
tomeu: can we make jobs optional based on variables?
<tomeu>
well, we already have a lava-gitlab-ci.yml that is included into the main one
<tomeu>
I think we should look into that, yeah
<bbrezillon>
yes, I tried pointing to lava-gitlab-ci directly, but it doesn't work
davidlt has quit [Ping timeout: 240 seconds]
yann has quit [Ping timeout: 240 seconds]
marvs has quit [Ping timeout: 252 seconds]
<tomeu>
bbrezillon: how is it looking, have you found a good way of running only panfrost jobs?
<tomeu>
maybe we should have a src/panfrost/gitlab-ci.yml file that includes lava-gitlab-ci.yml
<tomeu>
with a different run-policy
<tomeu>
(maybe)
TheCycoONE has joined #panfrost
<bbrezillon>
tomeu: nope
<bbrezillon>
sorry, end of the day in my timezone :)
<bbrezillon>
I'll try to look at it later this week
<bbrezillon>
tomeu: but having our own CI file with a different run-policy could do the trick, indeed