ChanServ changed the topic of #freedesktop to: https://www.freedesktop.org infrastructure and online services || for questions about freedesktop.org projects, please see each project's contact || for discussions about specifications, please use https://gitlab.freedesktop.org/xdg or xdg@lists.freedesktop.org
alanc has quit [Remote host closed the connection]
alanc has joined #freedesktop
Haaninjo has quit [Quit: Ex-Chat]
columbarius has joined #freedesktop
co1umbarius has quit [Ping timeout: 480 seconds]
Leopold__ has joined #freedesktop
Leopold_ has quit [Ping timeout: 480 seconds]
nektro has joined #freedesktop
<nektro>
is there an env var that can be set to change the directories fontconfig checks for fonts?
abrotman has quit [Remote host closed the connection]
abrotman has joined #freedesktop
alpernebbi has quit [Ping timeout: 480 seconds]
alpernebbi has joined #freedesktop
lack has quit [Read error: Connection reset by peer]
<daniels>
Venemo: just to double-check, it was you who was saying a few weeks ago that you wanted some particular person or an automated tool to just always keep Vulkan CTS up to date with the most recent version rather than developers doing it, right?
<Venemo>
daniels: I expressed some desire to be able to use a newer Vulkan CTS for one driver without needing to deal with upgrading all of the drivers at once
<Venemo>
I don't really see how this MR accomplishes that though
<daniels>
Venemo: ah right, no, it doesn't
Trevinho[m] has joined #freedesktop
Wallbraker has joined #freedesktop
aenderboy[m] has joined #freedesktop
alatiera_afk[m] has joined #freedesktop
anomalous_creator[m] has joined #freedesktop
bendlas[m] has joined #freedesktop
BLumia[m] has joined #freedesktop
bstrie[m] has joined #freedesktop
cassidy[m] has joined #freedesktop
chrysn[m]1 has joined #freedesktop
cmeissl[m] has joined #freedesktop
dabrain34[m]1 has joined #freedesktop
dcbaker has joined #freedesktop
DemiMarieObenour[m] has joined #freedesktop
dylan-m[m] has joined #freedesktop
ewlsh[m] has joined #freedesktop
gallo[m] has joined #freedesktop
gkiagia has joined #freedesktop
gnfzdz[m] has joined #freedesktop
Hazematman has joined #freedesktop
heftig has joined #freedesktop
jenatali has joined #freedesktop
jjardon[m] has joined #freedesktop
kashirin[m] has joined #freedesktop
kusma has joined #freedesktop
mairacanal[m] has joined #freedesktop
marcel203s[m] has joined #freedesktop
Mark[m] has joined #freedesktop
msizanoen[m] has joined #freedesktop
muhlinux[m] has joined #freedesktop
nazarewk[m] has joined #freedesktop
nielsdg has joined #freedesktop
nirbheek_ has joined #freedesktop
DavidHeidelberg[m] has joined #freedesktop
pitsch[m] has joined #freedesktop
pv has joined #freedesktop
razze[m] has joined #freedesktop
dabrain34[m] has joined #freedesktop
scorpion2185[m]1 has joined #freedesktop
sewn has joined #freedesktop
siddh has joined #freedesktop
sergi has joined #freedesktop
Sumera[m] has joined #freedesktop
swick[m] has joined #freedesktop
Nova[m] has joined #freedesktop
tintou has joined #freedesktop
underpantsgnome[m] has joined #freedesktop
<daniels>
what it does do is find newer versions of Vulkan CTS, upgrade them, schedule a pipeline run, check the results, and send an MR to update it optionally accompanied by changes to expectations files
<Venemo>
daniels: sounds like a step in the right direction. though I have to admit that I don't really have enough knowledge of how the CI works, so can't really judge or review that MR.
<daniels>
alright
<daniels>
well that may just happen at some point in future then
<bentiss>
thaller: technically, after that date, you will have to go to the pipeline tab, and manually click "run pipeline"
<thaller>
bentiss, I see. I didn't understand how this prevents evil-guy from just adding the snippet in their gitlab-ci.yaml.
<thaller>
(that's why it will stop working, is that so?)
<bentiss>
yeah
<Venemo>
daniels: I'm sorry to disappoint
<bentiss>
so we get some review before running the pipeline
<bentiss>
thaller: the problem is that I miscalculate the timing, and I'll be on vacation next week, so I might just postpone that cutoff by a week
<Venemo>
but there is very little info on how the ci system actually works
<daniels>
Venemo: there are some docs around, but much like the entire rest of Mesa, some of them are buried in random comments or wildly out of date or ...
<bentiss>
thaller: so you'll need the snippet first, and sometime in the future, you (or anyone with NM access) will need to manually run the pipline on MRs from occasional contributors
<bentiss>
thaller: yep
<thaller>
bentiss, got it. and the "run pipeline" button, is that the "two arrows in a circle" icon?
<bentiss>
thaller: there is an actual "Run Pipline" button, on top of the list in the pipeline tab
<Venemo>
daniels: so looking at the MR, it seems to just rename a few patch files, which is I guess fine, and edits build-deqp.sh which also makes some sense. what is not clear to me is, where and when does build-deqp.sh run? which ci job is it part of? is the cts built for every ci pipeline or is it downloaded from somewhere? how does it work?
<daniels>
bentiss: oooh, les vacances, nice! hope it's nice and relaxing :)
<bentiss>
thaller: but you might be able to just click the 2 circular arrows and it *might* just work
<daniels>
Venemo: build-deqp.sh is part of the container build job, so it gets rebuilt whenever the containers get rebuilt, and those get rebuilt whenever the dependencies change
<bentiss>
daniels: I hope too (and that's why I probably won't so the change in the CI)
<bentiss>
won't do
<daniels>
Venemo: the trigger is https://gitlab.freedesktop.org/gfx-ci/ci-uprev which is an external tool that runs as a cronjob (effectively) to check if there are newer versions of any applicable dependencies, and if so, tries to see if updating that dependency works; that's the thing which generated that update-Piglit-in-Mesa MR which I Just linked you to
<Venemo>
daniels: how does this MR change that process?
<Venemo>
will this build a new container for a new vk cts version?
<thaller>
I see it... hm... then I need the snippet first for that button?
<Venemo>
how is that then picked up by the ci jobs that drivers use?
<thaller>
ok, will merge the patch. I think I understood it. Thank you!!!
<bentiss>
thaller: no worries
<daniels>
Venemo: currently the only dependencies which get polled are piglit and virglrenderer; this adds support for checking for newer Vulkan CTS versions and doing an automate upgrade of the version of the Vulkan CTS which is executed by driver builds
<daniels>
when it finds one, it triggers an update and measures the results, which will result in an MR to update the version and, if required, any changes to driver expectation files
<eric_engestrom>
Venemo: I don't know the uprev tool, but my guess is that the two important parts in this MR are putting the version in a `VULKANCTS_REV` variable that can be modified using a script, and the contentious last commit that avoids manual action when checking if the update works
<Venemo>
it all sounds good to me
<daniels>
eric_engestrom: I agree with you that the succeed-on-patch-fail bit needs to be dropped
<eric_engestrom>
or reworked, eg. by gating it on the user running the pipeline
<daniels>
eric_engestrom: eh, if Piglit uprevs e.g. introduce a ton of crashes, it files an MR in draft status noting that the upgrade breaks the world and someone needs to do something manual with it; I'd imagine it would be the same for failing to rebase the Android CTS patches
<eric_engestrom>
that works too
<Venemo>
thanks for explaining it, it does seem like a good thing
<dabrain34[m]>
it seems to work fine at the beginning of the build for the gstreamer repo fetch but then it fails on the next one "orc" and also for the crates fetching
<dabrain34[m]>
can it be a FDO limitation ?
<bentiss>
dabrain34[m]: I don't think so, it seems that the DNS forgot about the IP of github.com, and I doubt this is FDO specific. I really don't see why this would happen tough
<bentiss>
so.... just retry?
<dabrain34[m]>
I retried and it fails each time the same
<daniels>
you'd have to ask the GNOME people what their images require
<dabrain34[m]>
thanks I will
ximion has quit [Quit: Detached from the Matrix]
AbleBacon has joined #freedesktop
siddh has quit [Quit: Reconnecting]
siddh has joined #freedesktop
siddh has quit []
siddh has joined #freedesktop
<eric_engestrom>
who's managing @mesa-ci-reports? (might be good to write that in its bio)
<eric_engestrom>
asking because its "pending job alert" message has a typo that bothers me every time I get an email: "more then X minutes"
<eric_engestrom>
🤓
<daniels>
hahaha
<daniels>
it's us - sergi, gallo[m], DavidHeidelberg[m], koike
<daniels>
it looks like the bio feature got disabled? but I've put the URL at least
<eric_engestrom>
daniels: thanks!
<eric_engestrom>
also, feels funny to see the ö on my name; I've dropped it virtually everywhere because there's so many systems that don't handle anything outside of US-ASCII, to the point I'm not used to seeing it like that anymore, outside of legal documents ^^
<daniels>
np!
<daniels>
oh oops - want me to ASCIIise it?
<eric_engestrom>
nah I don't care
<eric_engestrom>
it just got some weird "this is formal" feeling out of me ^^
<MrCooper>
"so many systems that don't handle anything outside of US-ASCII" in 2023, übersad
<daniels>
hey, at least I didn't use your middle name(s) :P
<eric_engestrom>
haha yeah, when you hear your full name with your middle name usually you're in trouble - either from your parent or from the law
MajorBiscuit has joined #freedesktop
MajorBiscuit has quit []
MajorBiscuit has joined #freedesktop
Leopold has joined #freedesktop
MajorBiscuit has quit [Ping timeout: 480 seconds]
Leopold__ has quit [Ping timeout: 480 seconds]
ximion has joined #freedesktop
MajorBiscuit has joined #freedesktop
woo2 has joined #freedesktop
<woo2>
Hey all - is anyone here familiar with libqmi or the em7455? Trying to debug a new batch of LTE modules and getting an undocumented qdl-message error (134) while using qmi-firmware-update
jarthur has joined #freedesktop
MajorBiscuit has quit [Quit: WeeChat 3.6]
<eric_engestrom>
woo2: I doubt you'll have much luck on this channel :)
<eric_engestrom>
bentiss: is gitlab down right now? both web and ssh are timing out
<eric_engestrom>
no wait, ssh is being incredibly slow but then fails with "git@gitlab.freedesktop.org: Permission denied (publickey)."
<ndufresne>
not responding on the web for me
<bentiss>
all webservices are to the roof right now (though arguably, it's only 2 CPUs)
<ndufresne>
sporadic would be more accurate
<ndufresne>
which matches the conditions ;-P
blatant has quit [Quit: WeeChat 3.8]
<ndufresne>
spammers got pissed that we have spam feature, and now they are running DDoS ?
blatant has joined #freedesktop
<bentiss>
the traffic is rather normal
<bentiss>
it went down
<bentiss>
and it's reactive now
<bentiss>
I wonder if some bot did not fetched some big tar gz of repos
a-l-e has quit [Ping timeout: 480 seconds]
<eric_engestrom>
and ssh is working again :]
<bentiss>
it looks like the only limit we got for webservice is the number of puma threads. Nothing in k8s is limiting the pod. So I guess I should just spawn more webservice pods???
* bentiss
tries to double the number of webservice pods
blatant has quit [Quit: WeeChat 3.8]
blatant has joined #freedesktop
vkareh has quit [Quit: WeeChat 3.6]
<ndufresne>
maybe it comes by waves ...
blatant has quit [Quit: WeeChat 3.8]
anomalous_creator[m] has quit []
MatrixTravelerbot[m]1 has quit []
blatant has joined #freedesktop
jjardon[m] has quit [Quit: Client limit exceeded: 20000]
tintou has quit []
Trevinho[m] has quit []
cmeissl[m] has quit []
M0xMRTT[m] has quit [Quit: Client limit exceeded: 20000]
danvet has quit [Ping timeout: 480 seconds]
siqueira has joined #freedesktop
blatant has quit [Quit: WeeChat 3.8]
blatant has joined #freedesktop
blatant has quit [Quit: WeeChat 3.8]
Leopold_ has quit []
Leopold_ has joined #freedesktop
blatant has joined #freedesktop
blatant has quit []
pzanoni` has joined #freedesktop
pzanoni has quit [Ping timeout: 480 seconds]
pzanoni has joined #freedesktop
pzanoni` has quit [Ping timeout: 480 seconds]
Haaninjo has quit [Quit: Ex-Chat]
damian_ has quit [Read error: Connection reset by peer]
damian has joined #freedesktop
amurray has left #freedesktop [ERC 5.5 (IRC client for GNU Emacs 29.0.90)]