<kisak>
oh no, the moment you hop off that MR treadmill you start to get rusty. Who knows, you might even go for a walk.
<karolherbst>
I'm already curious why I get some reviews from alyssa on my CL MRs 🤔🤔🤔
<kisak>
The rust is already showing itself.
Azirino has quit [Remote host closed the connection]
<alyssa>
jenatali: I thought I did r-b?
<alyssa>
well, lol'd-by
<alyssa>
close enough
<jenatali>
Oh is that what that means?
<alyssa>
:D
<alyssa>
karolherbst: Maybe Boston Dynamics has a vested interest in OpenCL ;-)
<karolherbst>
sounds plausible
<alyssa>
actually the new rumour was that Boston Dynamics is a cover for me working at John Deere on tractors
<alyssa>
I think that was also courtesy of zmike
<karolherbst>
I think I've heard something about that
<zmike>
no that's old news, I corrected the record already
<zmike>
you can't fool me
<karolherbst>
though I'm still catching up and are kinda stuck at you'll be going to apple
<karolherbst>
yes yes I know
<karolherbst>
I'm bad at catching up with rumours
<zmike>
she's going to dyson to work on vacuum tech
<karolherbst>
work on a dyson sphere? gotcha
<jenatali>
There's a joke about sucking to be made there but I'm not going to make it
<karolherbst>
you have to explain to me what you mean, I'm literally clueless
<alyssa>
zmike: are you saying my code sucks
<alyssa>
or will suck
<alyssa>
rather
kts has quit [Quit: Konversation terminated!]
<zmike>
sounds like you're saying it
<zmike>
you should have more confidence
<alyssa>
jenatali: too late i made it
<HdkR>
We already know John Deere is a red (green?) herring since you would never work at a company that is so anti-consumer :)
<alyssa>
HdkR: hey, quit it with your facts and logic here, this is a serious shitposting event
<alyssa>
:p
<zmike>
smh bringing facts to #dri-devel
<HdkR>
:D
<karolherbst>
what? serious facts in my chatroom?
<HdkR>
I just keep missing my opportunity to nab people for FEX. :P
<alyssa>
HdkR: Why would Dyson pay me to work on FEX?
<alyssa>
This is not skepticism, I'm asking you to come up with a reason, c'mon now =D
<HdkR>
For their new Gaming while you vacuum initiative. Gets people to clean more frequently
<zmike>
because FEX su...wait a minute
<alyssa>
zmike: ouch
<HdkR>
lol
<karolherbst>
there is one thing I'm curious about tho, why dyson?
<alyssa>
IDK ask zmike why should I know
<zmike>
need a custom mali-based graphics driver for the touchscreens on new vacuum line
<alyssa>
i literally just announced im done with mali
<HdkR>
Explains the OpenCL and AI accelerated cleaning
<alyssa>
as cover
<alyssa>
for my dyson job
<alyssa>
right?
<psykose>
dyson is getting acquired by apple and there's a new apple gpu for the touchscreens
<alyssa>
ooh now you're thinking
<HdkR>
A lifestyle company acquiring another lifestyle company
<HdkR>
Makes sense
<alyssa>
something something my alternative lifestyle
<zmike>
accurately describes working on apple hw
<kisak>
Too bad that iRobot went the way of OpenCV and camera mapping your house for Amazon instead of rendering on a ball. They're calling it the Dyson Sphere Project.
<karolherbst>
I've arrived at a conclusion: It's impossible that rusticl is wrong, hence radeonsi is broken, cased close
<alyssa>
no lies detected
<alyssa>
rusticl is written in rust and radeonsi is written in c
<alyssa>
pretty obvious who must be at fault
<karolherbst>
way a second, I have conclusive evidence
<karolherbst>
radeonsi still uses LLVM for compilation
<alyssa>
that wasn't conclusive?
<karolherbst>
let's say we have both conclusive evidence
<alyssa>
speaking of conclusive evidence
* alyssa
is playing with anholt's compare-perf.sh
<alyssa>
results are illuminating
<alyssa>
damn I sure waste a lot of time on optimizations that haven't actually helped.
<alyssa>
i mean i already knew that but
<karolherbst>
that reminds me, I have some "benchmarks" which are very good, where rusticl is 128x as fast as intel. Facts
<alyssa>
now i know that with numbers n=30
<karolherbst>
I wonder when I'll start doing actual performance optimizing
<karolherbst>
but seriously, this bugs just keeps bugging me and I have no idea what's wrong
<alyssa>
nod
<karolherbst>
I wonder if it's something really stupid
<alyssa>
anholt: btw, the "$i -gt 1" in compare-perf.sh should be "$1 -gt 2" since the 0th run didn't count, otherwise the first ministat invocation will complain
<psykose>
maybe the vacuum is making too much noise to focus? i hear dyson can help with that, we can get you a sponsorship..
<karolherbst>
mhhh
<HdkR>
karolherbst: If something is 100x faster, then something bad is happening on one end
Company has quit [Quit: Leaving]
<karolherbst>
HdkR: you don't want to know, because it's so stupid
jaganteki has quit [Remote host closed the connection]
<karolherbst>
all I can say is: rusticl isn't wrong
<karolherbst>
probably
<karolherbst>
turns out, there are some silly benchmarks trying to "hide" the fact, they are writing 128 times to the same memory by doing weirdo loops and shit
<HdkR>
I follow
<karolherbst>
apparently nir can look behind it
<karolherbst>
llvm can't
<karolherbst>
sooo....
<HdkR>
lol
<karolherbst>
fun fact
<karolherbst>
they report memory bandwidth and hope it's not getting optimized away
<karolherbst>
so apparently I get like TB/s of memory bandwidth on my laptop
<HdkR>
LLVM isn't a GPU compiler, reaffirmed
<HdkR>
TB/s of memory BW in the L1 cache maybe
<karolherbst>
I mean..
<karolherbst>
if you divide it by 128...
<karolherbst>
I think I need to find somebody at AMD to look into this for me 🙃
<kisak>
It's nice how radeonsi has been quietly (from the news side) preparing for not LLVM as an option.
mbrost has quit [Remote host closed the connection]
mbrost has joined #dri-devel
<karolherbst>
yeah uhm...
<karolherbst>
so here is this thing
<karolherbst>
I need proper function calls
<karolherbst>
*runs*
<HdkR>
OpenGL subroutines call to you
<karolherbst>
not the same thing
<HdkR>
Basically the same thing if the UBO is just a raw GPU pointer :P
<HdkR>
er, Uniform variable
<karolherbst>
kinda sad you don't have a driver in mesa, otherwise I could say things like "wanna throw a bug about not being able to compile a shader with 2.000.000 ssa values with your driver at you?" :P
<karolherbst>
*wanna me
<alyssa>
how does aco do with shaders that big OOI?
<HdkR>
It's an advantage that I have :)
<karolherbst>
probably the same as LLVM
<karolherbst>
OOM
<alyssa>
OOM.. right.
<alyssa>
we should fix that
<alyssa>
2 million isn't that big, there are gigabytes of RAM
<karolherbst>
so apparently some of those kernels compile on some amount of RAM, like 64GB for instance
<alyssa>
karolherbst: OOM in the backend or in NIR?
<karolherbst>
oh, nir is fine
<karolherbst>
it's RA which goes OOM
<karolherbst>
well
<alyssa>
oh
<karolherbst>
the nir shader is still huge
<karolherbst>
like 20GB or something
<alyssa>
we should definitely fix that, aco's RA should be able to cope with just a few million values
<karolherbst>
obviously
<karolherbst>
or make nir better in optimizing that stuff
<karolherbst>
I mean.. some glocal CSE won't hurt with a shader that big, will it?
<karolherbst>
but yeah anyway.. turns out inlining functions with thousand of instructions a couple of times does blow up some shaders
<karolherbst>
I'm still trying to find the person doing the work on the intel backend compiler for this 🙃
<alyssa>
i will not wire up images on asahi for opencl over holiday.
<alyssa>
i will not wire up images on asahi for opencl over holiday.
<alyssa>
i will not wire up images on asahi for opencl over holiday.
<alyssa>
*bart simpson*
<karolherbst>
well.. you could do it for gl instead :P
<alyssa>
or vulkan
<alyssa>
maybe dyson will pay for agxv
<HdkR>
You could do it for Vulkan and let Zink handle OpenGL for you :P
<alyssa>
for vacuum dxvk
<karolherbst>
but yeah...
<karolherbst>
I should boot up my m2 again...
<karolherbst>
for uhm...
<karolherbst>
reasons
<alyssa>
I'll probably do images in May
<karolherbst>
how is the compuer shader work going?
<karolherbst>
noice
<alyssa>
Maybe I'll have funding from the vacuum company by then
<karolherbst>
passing the CTS in may should be feasible
<karolherbst>
last time I ran the CTS on agx it was like 50% pass rate
<karolherbst>
that was in december
<karolherbst>
though a lot of stuff just passed because it doesn't report "not supported" and if images are disabled...
<karolherbst>
I think I'll do radeonsi first, because it's like 25 tests to fix left
<alyssa>
karolherbst: please don't work on rusticl+asahi until I have gles/vk level compute finished off / give the green light, it will end up being stressful for me otherwise i expect
<alyssa>
in the mean time, delete clover? :fire:
<karolherbst>
:D
<karolherbst>
well..
<karolherbst>
svm is assigned to marge
krushia has joined #dri-devel
<karolherbst>
I wonder if I have enough to make HIP/SyCL run now as well...
alyssa has quit [Quit: leaving]
crabbedhaloablut has quit [Ping timeout: 480 seconds]
mbrost has quit [Read error: Connection reset by peer]
godvino has joined #dri-devel
crabbedhaloablut has joined #dri-devel
<robclark>
gfxstrand: ir3 can really indirect into register file (or const file)
<robclark>
large arrays are useful to lower to scratch (we do) since register pressure is a thing.. but small arrays are useful to keep in gpr since we can fold indirect into alu instructions
<robclark>
ahh, I guess emma covered that
* robclark
catching up on scrollback
crabbedhaloablut has quit []
<HdkR>
RF indexing is still such a cool feature. I wish more things supported that
JohnnyonFlame has quit [Read error: No route to host]
crabbedhaloablut has joined #dri-devel
JohnnyonFlame has joined #dri-devel
kts has joined #dri-devel
agd5f_ has joined #dri-devel
kts has quit [Remote host closed the connection]
agd5f has quit [Remote host closed the connection]
anarsoul has joined #dri-devel
anarsoul has quit [Remote host closed the connection]
krushia has quit [Remote host closed the connection]
krushia has joined #dri-devel
JohnnyonF has joined #dri-devel
kts has joined #dri-devel
Duke`` has joined #dri-devel
kzd has quit [Ping timeout: 480 seconds]
JohnnyonFlame has quit [Ping timeout: 480 seconds]
<mdnavare>
danvet, daniels , robclark : I am running through my dim set up and when it tries to pull the drm-tip, i see this issue:./dim setup
<mdnavare>
fatal: Could not read from remote repository.
<mdnavare>
Please make sure you have the correct access rights
<mdnavare>
and the repository exists.
<mdnavare>
dim: Failed to fetch drm-tip
aravind has joined #dri-devel
<dolphin>
mdnavare: does ssh work to people.freedesktop.org?
<ishitatsuyuki>
HdkR, do you know some examples of code that can be lowered to RF indexing in practice? AMD ISA also has some support for it, so maybe we should have it in ACO too
yuq825 has joined #dri-devel
bmodem has joined #dri-devel
<HdkR>
ishitatsuyuki: I used to have a good example like half a decade ago but I forget what it was now
<ishitatsuyuki>
oof
<mupuf>
anholt: yes, we do automatically reboot before trying again. This is useful for the navi10, but maybe we can drop that on navi21
<mupuf>
anholt: set B2C_SESSION_REBOOT_REGEX to an empty string if you want to disable the behaviour
<HdkR>
ishitatsuyuki: I think it was something in an uber shader that generated a bunch of state in to RF and then later stages would be more optimal because it could then just index the various bits that were necessary?
<HdkR>
But that's so vague at this point that the idea of storing 16 vec4s in registers or whatever doesn't really help
<ishitatsuyuki>
it does feel a bit vague indeed
<ishitatsuyuki>
but let me keep some of that in mind
bgs has joined #dri-devel
<HdkR>
If you find any shaders generating some large thread local array, gimme a link so I can refresh my mind on it at some point :P
orbea has quit [Remote host closed the connection]
orbea has joined #dri-devel
adavy has joined #dri-devel
kusma has quit []
fxkamd has quit []
bmodem1 has joined #dri-devel
bmodem has quit [Ping timeout: 480 seconds]
<mdnavare>
dolphin: Should i try to ssh to prople.freedesktop.org and check?
<mdnavare>
people.freedesktop.org
<dolphin>
yeah, the basic SSH setup needs to work first, seems like it doesn't
<mdnavare>
Nope it doesnt I get the same denied error: Permanently added 'people.freedesktop.org' (ED25519) to the list of known hosts.
<dolphin>
Host *.freedesktop.org\n User dolphin\n IdentityFile ~/.ssh/id_freedesktop
<mdnavare>
so basically i had the set up working and also commit right with my keys i had on my intel linux laptop, now that i moved to Google I am trying to set up the dim tools and it shows denied
<dolphin>
does your SSH config have something like that? of course with your freedesktop user login :)
<dolphin>
I think you may have to supply a new ssh public key to FDO admins to update to your account
danvet has quit [Remote host closed the connection]
<mdnavare>
Hmm I dont have that entry in my ssh config either
<mdnavare>
Let me try requesting FDO admins to update my account first, how do i do that? I dont remember top of my head
danvet has joined #dri-devel
<mdnavare>
dolphin: Should i just file a bug at fdo and ask the admins to update my account with new ssh public key?
<zmike>
eric_engestrom: how is the branching and is it yet time to start backporting to 23.1
<eric_engestrom>
zmike: the branchpoint happened yesterday morning, but then I had issues merging !22260 that needed to go in before the first -rc, which got merged this morning
<eric_engestrom>
I'm a bit busy so I won't post the -rc1 and email until after work today, but the staging/23.1 branch is now at the state that will be -rc1, so you can start doing things with it already (testing, backport MRs, etc.)
<zmike>
oh cool
<eric_engestrom>
for backport MRs I won't merge them until I've posted -rc1 though
<zmike>
hm I have a lot of patches that I was just going to backport directly
<zmike>
or is that not allowed anymore
<eric_engestrom>
no, please make an MR so that the CI verifies we don't backport too many bugs :)
<zmike>
gasp
<zmike>
how dare you
<zmike>
I ONLY backport bugs
<eric_engestrom>
but also, if you `cc: mesa-stable` or `fixes: abcdef123` it will automatically be handled
<zmike>
yea, I have a bunch of patches without that
<eric_engestrom>
I didn't wanna say it :P
<zmike>
:P
underpantsgnome[m] has quit []
guru_ has joined #dri-devel
oneforall2 has quit [Ping timeout: 480 seconds]
gawin has joined #dri-devel
<emersion>
ishitatsuyuki: thanks, it turned out to be an issue in my code
<ishitatsuyuki>
good to hear it's resolved
<emersion>
a bit sad that the validation layers didn't complain about my timeline sema value being 0
<gawin>
DavidHeidelberg[m]: fixes is better than cc:stable? I was always slapping second one.
Soroush has quit []
Soroush has joined #dri-devel
<DavidHeidelberg[m]>
gawin: it should be, someone told me that Fixes automatically pick for backports
<emersion>
fixes is better, it allows the fix to be propagated only to the trees that need it
gpoo has joined #dri-devel
<kisak>
Fixes: <commitid> gets read as, if release branch has <commitid>, then pick this up too.
Soroush has quit []
<kisak>
particularly useful near branchpoints
Soroush has joined #dri-devel
<jenatali>
My rule of thumb is if it was regressed sometime in the last year, use fixes, otherwise don't bother trying to search for the right commit
<zmike>
smh not having every commit memorized by the release it shipped in so you know whether to use fixes
kxkamil has joined #dri-devel
<emersion>
there are only a few hundred thousand commits after all
camus1 has quit [Remote host closed the connection]
<haasn>
in 2023, is there any way to get actual, working VRR in GL/Vulkan/whatever apps on Linux?
<haasn>
ideally not requiring VkDisplayKHR?
<haasn>
on mesa drivers, I mean
<emersion>
that's up to the compositor
<emersion>
many compositors do support it
<haasn>
sway?
<haasn>
it works in sway just moving the mouse cursor but doesn't work inside GL apps
<emersion>
if the GL app is never late to submit a frame, it'll just use the full framerate
<haasn>
Seems to all work fine under KWin/Plasma also
<haasn>
I think it works better in actual games?
<haasn>
Or at least it’s harder to spot duplicated frames there
OftenTimeConsuming has quit [Remote host closed the connection]
dsrt^ has quit [Remote host closed the connection]
OftenTimeConsuming has joined #dri-devel
rasterman has quit [Quit: Gettin' stinky!]
<haasn>
Oh nice, after switching from Plasma Wayland back to X11 I get flickering garbage at the bottom
<haasn>
Seems X11 doesn't want to do freesync at all anymore though
gawin has quit [Ping timeout: 480 seconds]
kts has joined #dri-devel
gpoo has quit [Ping timeout: 480 seconds]
kts has quit []
kzd has joined #dri-devel
gpoo has joined #dri-devel
smilessh has quit [Remote host closed the connection]
yuq825 has quit []
smiles_1111 has joined #dri-devel
gpoo has quit [Ping timeout: 480 seconds]
gpoo has joined #dri-devel
stuarts has joined #dri-devel
camus has joined #dri-devel
camus has quit [Remote host closed the connection]
camus has joined #dri-devel
<tleydxdy>
follow-up on the drm-file pid issue
Kwiboo has quit [Quit: .]
<tleydxdy>
if the game allocates vram, would be only be added to the game's vm list? or would X also have access to all
Kwiboo has joined #dri-devel
<MrCooper>
karolherbst: looks like rusticl's fill_image/create_box is setting pipe_box::depth = 0 for 2D regions, should be = 1
tursulin has quit [Ping timeout: 480 seconds]
<karolherbst>
ohh really...
<karolherbst>
yeah, that's not good :D
<karolherbst>
mhhh
<karolherbst>
yeah.. I think I need to wrap creation of pipe_box and require a target for those fix ups
<karolherbst>
yeah.. there are some other places like that where it could be a potential problem
<jenatali>
zmike: That failure is from st_TexImage for a multisampled texture. Is that supposed to work?
<zmike>
yes
<zmike>
it seems to on every other driver
<jenatali>
🤔
<zmike>
indeed
<jenatali>
That's... not something I can easily make work
<zmike>
then how is the test not failing at present?
<jenatali>
'cause it's not trying to CPU upload to an MSAA texture. The null texture that was created was non-MSAA
<zmike>
I see
DemiMarieObenour[m] is now known as DemiMarie
<jenatali>
I haven't looked to see why that worked in the first place, it might've been working by pure accident
<jenatali>
But this CPU upload of an MSAA texture is not going to work, we'd need to blit it. At which point we'd really prefer to just get a clear rather than a texture upload
<zmike>
tbf I haven't looked at the test either
<zmike>
I just saw that you were the only one failing
swalker__ has quit [Remote host closed the connection]
<jenatali>
I think it's a coincidence, the test binds a program and does a clear and a readback, and it's just that during the readback Mesa sets up the state for the program and creates this null texture
<zmike>
ah ok
<jenatali>
And while trying to init the null texture we hit an error condition
<jenatali>
Like... I could skip the copy and the test would probably work
<zmike>
hm so it might be the case that I need a special condition for the texsubdata there with ms
<zmike>
I'll look into it
<jenatali>
Yeah, I think so
<zmike>
thx for your expert debugging
<jenatali>
Heh I wouldn't go that far. Run test under debugger, see debug message, report debug message to you :P
<zmike>
E X P E R T
stuarts has quit [Remote host closed the connection]
kts has joined #dri-devel
kts has quit [Quit: Konversation terminated!]
rasterman has joined #dri-devel
djbw__ has joined #dri-devel
ChaosPrincess has joined #dri-devel
dcz_ has joined #dri-devel
<dcz_>
what hardware does rusticl support?
<dcz_>
is it all models covered by radeonsi?
stuarts has joined #dri-devel
<mdnavare>
daniels:
<mdnavare>
daniels: Could you help me update the ssh public keys for my fdo account so I can re setup my dim
khfeng has quit [Ping timeout: 480 seconds]
vliaskov_ has quit [Ping timeout: 480 seconds]
Duke`` has joined #dri-devel
jkrzyszt has quit [Ping timeout: 480 seconds]
Guest10844 is now known as go4godvin
<daniels>
mdnavare: yep I did see that and I’ll change it later
mbrost has joined #dri-devel
i509vcb has joined #dri-devel
iive has joined #dri-devel
lynxeye has quit [Quit: Leaving.]
bgs has quit [Remote host closed the connection]
<mdnavare>
Okay thank you so much daniels
jfalempe has quit [Quit: Leaving]
bmodem1 has quit [Ping timeout: 480 seconds]
gouchi has joined #dri-devel
mbrost has quit [Ping timeout: 480 seconds]
smiles_1111 has quit [Ping timeout: 480 seconds]
<ChaosPrincess>
hi all. I am writing a display driver, and my hardware and linux has differing ideas on which order the pixels should be in memory, with mine scanning out from the bottom left corner and then going up, and then right at the end of the row. as a result, the image ends up being turned 90 degrees anticlockwise (kinda). Is there a way to make drm rotate everything?
ngcortes has joined #dri-devel
Cyrinux9 has quit []
Cyrinux9 has joined #dri-devel
ngcortes_ has joined #dri-devel
godvino has joined #dri-devel
ngcortes has quit [Ping timeout: 480 seconds]
alyssa has joined #dri-devel
<alyssa>
jenatali: look into u_transfer_helper's msaa_map?
alyssa has left #dri-devel [#dri-devel]
<jenatali>
alyssa: I don't think D3D allows copies between MSAA resources and buffers, which is how upload/download is done
<jenatali>
Any kind of workaround would involve a second hop through a single-sampled resource, either a blit or a resolve
<jenatali>
Which, if we have to do it because GL says so, then fine, but considering we're not doing it yet, uploading a single black texel doesn't seem like the reason to implement that kind of thing
ngcortes_ has quit [Ping timeout: 480 seconds]
godvino1 has joined #dri-devel
dcz_ has quit [Ping timeout: 480 seconds]
heat has quit [Remote host closed the connection]
lemonzest has quit [Quit: WeeChat 3.6]
heat has joined #dri-devel
godvino has quit [Ping timeout: 480 seconds]
gpoo has left #dri-devel [Leaving]
rasterman has quit [Quit: Gettin' stinky!]
alyssa has joined #dri-devel
Haaninjo has quit [Read error: Connection reset by peer]
Haaninjo_ has joined #dri-devel
godvino1 has quit [Read error: Connection reset by peer]
godvino has joined #dri-devel
the_sea_peoples has joined #dri-devel
YuGiOhJCJ has joined #dri-devel
ngcortes has joined #dri-devel
danvet has quit [Ping timeout: 480 seconds]
kzd has quit [Quit: kzd]
godvino has quit [Quit: WeeChat 3.6]
alyssa has quit [Quit: leaving]
kzd has joined #dri-devel
Duke`` has quit [Ping timeout: 480 seconds]
CME has quit [Ping timeout: 480 seconds]
CME has joined #dri-devel
Duke`` has joined #dri-devel
pcercuei has quit [Quit: Lost terminal]
iive has quit [Quit: They came for me...]
gouchi has quit [Remote host closed the connection]
<karolherbst>
MrCooper: actually... it doesn't because the API validation makes sure it's 1 not 0, or at lesat should be
<karolherbst>
but mh.. maybe fill_image/fill_buffer doesn't care? let me see...
OftenTimeConsuming is now known as Guest11045
OftenTimeConsuming has joined #dri-devel
Guest11045 has quit [Remote host closed the connection]