ChanServ changed the topic of #dri-devel to: <ajax> nothing involved with X should ever be unable to find a bar
heat has quit [Remote host closed the connection]
dviola has joined #dri-devel
iive has quit []
karolherbst_ has joined #dri-devel
karolherbst is now known as Guest1702
karolherbst_ is now known as karolherbst
Haaninjo has quit [Quit: Ex-Chat]
JohnnyonFlame has joined #dri-devel
soreau has quit [Read error: Connection reset by peer]
karolherbst_ has joined #dri-devel
Guest1702 has quit [Ping timeout: 480 seconds]
soreau has joined #dri-devel
karolherbst is now known as Guest1703
karolherbst_ is now known as karolherbst
Guest1703 has quit [Ping timeout: 480 seconds]
i-garrison has quit [Ping timeout: 480 seconds]
i-garrison has joined #dri-devel
rasterman has quit [Quit: Gettin' stinky!]
ybogdano has quit [Ping timeout: 480 seconds]
tursulin has quit [Read error: Connection reset by peer]
Guest1698 is now known as maxzor
<maxzor>
Do you know about good reads on DMA_BUF fences, for someone whom only knows old-school semaphores?
<maxzor>
are fences mature? there seem to be quite some complexity
co1umbarius has joined #dri-devel
columbarius has quit [Ping timeout: 480 seconds]
ngcortes has quit [Ping timeout: 480 seconds]
lcn has joined #dri-devel
ccr has joined #dri-devel
Terman has joined #dri-devel
rawoul has joined #dri-devel
nchery has quit [Quit: Leaving]
sarnex has quit [Quit: Quit]
sarnex has joined #dri-devel
haasn has joined #dri-devel
<cheako>
Hi, I'm trying to write a vklayer in rust... I got to the part where I call into a rust library and it calls my `PFN_vkGetInstanceProcAddr` looking for `vkGetInstanceProcAddr`. I'm unable to parse `VkLayerInstanceLink` for a usable value.
itoral has quit [Remote host closed the connection]
<javierm>
maxzor: ^
itoral has joined #dri-devel
<daniels>
javierm: yeah, there was that series, but then it misses out drm_syncobj
itoral has quit [Remote host closed the connection]
<maxzor>
what i took from Jason's talk and the following chat was that user-space fences are the "future" and backward compat is a mess ':)
<maxzor>
thank you javierm
itoral has joined #dri-devel
<maxzor>
how did early linux manage to never break userspace? where there not changes in cpu designs of magnitude similar to those of current gpu designs?
<daniels>
I think 'everything is bad' is a reasonable summary
<javierm>
maxzor: that's the takeaway from padovan's talk :)
<javierm>
I tried once to add implicit fences support to v4l2/media, it didn't end well
<daniels>
javierm: I hear modifiers are coming to V4L2 any decade now
<emersion>
… can't wait
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
<javierm>
daniels: yeah...
lemonzest has quit [Quit: WeeChat 3.4]
itoral has quit [Remote host closed the connection]
flacks has quit [Quit: Quitter]
itoral has joined #dri-devel
flacks has joined #dri-devel
<javierm>
daniels: IIRC the probelm with implicit fences and v4l2 was that user-space was unaware that a DQBUF was waiting for a fence and tried to QBUF again the same buffer causing a deadlock
<javierm>
so it wasn't possible to make it unaware of the implicit fences. With explicit fences you could signal user-space and only then queue the buffer again
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
hch12907 has joined #dri-devel
JoshuaAs- has quit [Server closed connection]
JoshuaAshton has joined #dri-devel
<daniels>
eugh
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
alatiera1 has joined #dri-devel
hch12907_ has quit [Ping timeout: 480 seconds]
hch12907 has quit [Read error: Connection reset by peer]
itoral has quit [Remote host closed the connection]
hch12907 has joined #dri-devel
itoral has joined #dri-devel
alatiera has quit [Ping timeout: 480 seconds]
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
devilhorns has joined #dri-devel
devilhorns has quit []
devilhorns has joined #dri-devel
itoral has quit [Remote host closed the connection]
tpalli has quit [Server closed connection]
tpalli has joined #dri-devel
itoral has joined #dri-devel
tjaalton has quit [Server closed connection]
tjaalton has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
Lucretia has quit [Server closed connection]
itoral has quit [Remote host closed the connection]
bnieuwenhuizen has quit [Server closed connection]
bnieuwenhuizen has joined #dri-devel
bnieuwenhuizen is now known as Guest1762
Lucretia has joined #dri-devel
itoral has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
kts has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
vvaryvdin has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
itoral has quit [Remote host closed the connection]
kts has quit [Quit: Konversation terminated!]
itoral has joined #dri-devel
aravind has quit [Ping timeout: 480 seconds]
itoral has quit [Remote host closed the connection]
ppascher has quit [Quit: Gateway shutdown]
itoral has joined #dri-devel
vvaryvdin has quit []
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
lemonzest has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
rellla has quit [Server closed connection]
JohnStultz[m] has quit [Server closed connection]
JohnStultz[m] has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
nchery has joined #dri-devel
rellla has joined #dri-devel
itoral has quit [Remote host closed the connection]
gawin has joined #dri-devel
JohnnyonF has joined #dri-devel
JohnnyonF has quit [Read error: Connection reset by peer]
JohnnyonFlame has quit [Ping timeout: 480 seconds]
macromorgan is now known as Guest1768
macromorgan has joined #dri-devel
Guest1768 has quit [Read error: Connection reset by peer]
cphealy_ has joined #dri-devel
devilhorns_ has joined #dri-devel
cphealy has quit [Read error: Connection reset by peer]
devilhorns has quit [Ping timeout: 480 seconds]
<javierm>
tzimmermann, sravn: hmm, I just noticed that patman (the tool I use to post patches) adds a "no changes since v1" if I don't specify the changes for individual commits...
dogukan has joined #dri-devel
<javierm>
tzimmermann, sravn: this is related to the ssd130x v2, the changelog is in the cover letter. I put it there because there were just too many changes and would add noise to the patches :)
kts has joined #dri-devel
pcercuei has quit [Ping timeout: 480 seconds]
kts_ has joined #dri-devel
kts_ has quit []
kts has quit [Ping timeout: 480 seconds]
kts has joined #dri-devel
rpigott has quit [Remote host closed the connection]
rpigott has joined #dri-devel
pcercuei has joined #dri-devel
jewins has joined #dri-devel
mattrope has joined #dri-devel
hch12907_ has joined #dri-devel
kts has quit [Quit: Konversation terminated!]
kts has joined #dri-devel
Arsen has joined #dri-devel
hch12907 has quit [Ping timeout: 480 seconds]
Danct12 has joined #dri-devel
<marex>
javierm: are you happy with patman ?
Daanct12 has quit [Ping timeout: 480 seconds]
cphealy_ has quit []
hch12907 has joined #dri-devel
sdutt has joined #dri-devel
hch12907_ has quit [Ping timeout: 480 seconds]
hch12907_ has joined #dri-devel
cphealy has joined #dri-devel
hch12907 has quit [Ping timeout: 480 seconds]
JohnnyonFlame has joined #dri-devel
hch12907 has joined #dri-devel
karolherbst is now known as karolerbst
karolerbst has quit []
karolerbst has joined #dri-devel
karolerbst has quit []
hch12907_ has quit [Ping timeout: 480 seconds]
karolerbst has joined #dri-devel
karolerbst has quit []
karolherbst has joined #dri-devel
hch12907_ has joined #dri-devel
hch12907 has quit [Ping timeout: 480 seconds]
alatiera18 has joined #dri-devel
maxzor_ has joined #dri-devel
alatiera1 has quit [Ping timeout: 480 seconds]
hch12907 has joined #dri-devel
ahajda has quit [Quit: Going offline, see ya! (www.adiirc.com)]
Major_Biscuit has quit [Ping timeout: 480 seconds]
hch12907_ has quit [Ping timeout: 480 seconds]
hch12907 has quit [Ping timeout: 480 seconds]
hch12907 has joined #dri-devel
mattrope has quit [Read error: Connection reset by peer]
hch12907_ has joined #dri-devel
cworth has joined #dri-devel
<anholt>
alyssa: you could just put up the MR adding it to the flakes and then someone would merge it, instead of waiting?
hch12907 has quit [Ping timeout: 480 seconds]
hch12907 has joined #dri-devel
mszyprow has joined #dri-devel
mattrope has joined #dri-devel
hch12907_ has quit [Ping timeout: 480 seconds]
hch12907_ has joined #dri-devel
<daniels>
^ yeah, unless an iris person is going to bottom it out and land a fix in the next couple of hours, adding it to flakes is going to win
william has quit [Ping timeout: 480 seconds]
hch12907 has quit [Ping timeout: 480 seconds]
hch12907 has joined #dri-devel
nchery has quit [Ping timeout: 480 seconds]
nchery has joined #dri-devel
hch12907_ has quit [Ping timeout: 480 seconds]
mszyprow has quit [Ping timeout: 480 seconds]
hch12907_ has joined #dri-devel
hch12907 has quit [Ping timeout: 480 seconds]
hch12907 has joined #dri-devel
* alyssa
went off to do other things
hch12907_ has quit [Ping timeout: 480 seconds]
nchery has quit [Ping timeout: 480 seconds]
nchery has joined #dri-devel
hch12907_ has joined #dri-devel
robert_mader has joined #dri-devel
hch12907 has quit [Ping timeout: 480 seconds]
hch12907 has joined #dri-devel
ngcortes has joined #dri-devel
hch12907_ has quit [Ping timeout: 480 seconds]
hch12907_ has joined #dri-devel
gouchi has joined #dri-devel
sdutt has quit []
sdutt has joined #dri-devel
hch12907 has quit [Ping timeout: 480 seconds]
hch12907 has joined #dri-devel
hch12907_ has quit [Ping timeout: 480 seconds]
hch12907_ has joined #dri-devel
alatiera18 is now known as alatiera
mszyprow has joined #dri-devel
hch12907 has quit [Ping timeout: 480 seconds]
hch12907 has joined #dri-devel
kts_ has joined #dri-devel
hch12907_ has quit [Ping timeout: 480 seconds]
kts has quit [Ping timeout: 480 seconds]
<graphitemaster>
Friday's are great because when the unit test fails you can just comment them out and push the code, forget about it come Monday after a weekend long binge of malt whiskey and no one is the wiser. My motto has always been "what good is a unit test unless it actually affects someone". Has Mesa tried this with DEQP / CTS?
hch12907_ has joined #dri-devel
<imirkin>
graphitemaster: yes. that's why we have the fails files.
<imirkin>
just put a test in there, and no one will ever know.
<jenatali>
There we go, GL4.2 :)
<imirkin>
woohoo!
<imirkin>
what's missing for 4.3? something like stencil texturing?
<imirkin>
4.4 has qbo, which will be a happy fun time...
<graphitemaster>
imirkin, It'll start getting suspicious when that directory gets larger and the success one approaches empty.
<graphitemaster>
That dude's shorts are way too short. Also who wears socks and work boots in shorts. Someone teach this man a thing about drip. That crocheted sweater is from like the 70s. This just looks like depression ngl XD
hch12907_ has quit [Ping timeout: 480 seconds]
<imirkin>
i see you latched onto the most important bit of that poster...
hch12907_ has joined #dri-devel
devilhorns_ has quit []
<jenatali>
imirkin: Texture format casting
robert_mader has left #dri-devel [#dri-devel]
<jenatali>
I think we've talked about it, we can only cast by class, but GL requires casting by size
<imirkin>
jenatali: i thought texture views were GL 4.2...
<jenatali>
Nope, 4.3
<jenatali>
Well at least according to Mesa's versioning logic :)
<imirkin>
yeah, that's right. my memory was off.
<imirkin>
i thought they came at the same time as images. but they did not.
<graphitemaster>
This is the problem with posters, if you want your messaging to land you have to nail the photograph too, especially if you're going to violate simple photographic principles like the rule of thirds, white balance, and the distinctive lack of three point lighting on your subject (natural light is a meme)
<jenatali>
No, but they did come with compute
<imirkin>
graphitemaster: it's not the best photo. they have some other posters which really nail the image part.
<graphitemaster>
Aside; what is a qbo? Query buffer object?
<imirkin>
graphitemaster: like "incompetence" or "disloyalty"
<imirkin>
yes, qbo = query buffer object
<graphitemaster>
Those are GL 4.3? things
<imirkin>
4.4
hch12907 has quit [Ping timeout: 480 seconds]
<imirkin>
(well, core in 4.4)
<jenatali>
I think qbos will be easy enough, but I haven't looked too closely
<graphitemaster>
We should commission a disloyalty poster for everyone who non-ironically prefers Vulkan to Modern OpenGL (DSA)
<imirkin>
iirc radeon had to resort to compute shaders to implement qbo
<imirkin>
and for nvidia i had to write some ... uh ... creative macros
hch12907 has joined #dri-devel
<graphitemaster>
I feel like QBOs could just be emulated with an SSBO and some shader injection
<graphitemaster>
Even something like samples passed is just an increment in the fragment shader right :P
<imirkin>
graphitemaster: well, you can emulate them by ignoring the spirit of the qbo
<imirkin>
but the spirit of the qbo is that it's async
cworth has quit [Ping timeout: 480 seconds]
<graphitemaster>
Anything can be made async with some frames of latency
<imirkin>
heheh
kts_ has quit []
hch12907_ has quit [Ping timeout: 480 seconds]
hch12907_ has joined #dri-devel
cworth has joined #dri-devel
pnowack has quit [Quit: pnowack]
Guest1762 has quit []
hch12907 has quit [Ping timeout: 480 seconds]
bnieuwenhuizen has joined #dri-devel
hch12907_ has quit [Ping timeout: 480 seconds]
hch12907 has joined #dri-devel
<graphitemaster>
Are there any Objective C toolchains for Linux that have the Metal API headers for cross compiling
<graphitemaster>
Also, how come there are no memory barriers needed for indirect buffers in OpenGL
<graphitemaster>
Like, there's no GL_DRAW_INDIRECT_BARRIER_BIT, or GL_DISPATCH_INDIRECT_BARRIER_BIT
<graphitemaster>
Do draws and dispatches implicitly do memory barriers for those?
<graphitemaster>
Oh there is a GL_COMMAND_BARRIER_BIT
tzimmermann has quit [Quit: Leaving]
<graphitemaster>
I was missing so many barriers lol
pnowack has joined #dri-devel
<graphitemaster>
Mind you GL_COMMAND_BARRIER_BIT says only for Draw*Indirect, there's no comment about DISPATCH_INDIRECT or glDispatchIndirect
<imirkin>
jenatali: what's up with this? spec/!opengl 4.2/required-sized-texture-formats: fail
<jenatali>
It's a D3D limitation. We've since relaxed it and it passes for me locally, but the CI machine's using an old OS/D3D
<imirkin>
ah cool
<jenatali>
D3D required top mips of block-compressed textures to be 4-pixel aligned, for no good reason since lower mips didn't have to be aligned
<imirkin>
ha ok
<imirkin>
so that means you're just not allowed to use compression for 18-px-wide images?
<jenatali>
You'd have to pad it and put a manual gutter on it, yeah
<imirkin>
or you have to have all kinds of fixup internally to remember to ignore the last 2 pixels?
<jenatali>
Like I said, no good reason
<imirkin>
and this is at at the D3D level? not in WARP or whatever?
<jenatali>
Yep
<jenatali>
The API disallows it
<imirkin>
harsh.
<imirkin>
i guess most compressed textures i've seen are POT, so *4 isn't such a big deal
<jenatali>
Yeah, exactly. But GL needed it. That was the one feature we added to D3D to support GLOn12 :)
<linkmauve>
Oh, so this is why this Unity game’s data uses BGRA8888 instead of BC7 for non-multiple-of-4 textures!
<imirkin>
ahaha
<linkmauve>
TIL.
<imirkin>
"well, i know how to solve this problem..."
<jenatali>
Damn, DOOM still doesn't work even at 4.2. Looks like Mesa's missing a WGL extension it requires
<ajax>
please don't say OML_swap_method please don't say OML_swap_method
<imirkin>
doom 3, presumably?
<imirkin>
(doom 2 came out ... late 90's, so can't need too high a GL version... and doom was pre-GL)
<jenatali>
ajax: Swap_tear_control
<jenatali>
imirkin: DOOM 2016
<imirkin>
right
<jenatali>
We've been able to run DOOM 3 for a while now, that's only 3.3 I think
<imirkin>
oh wait, doom 3 != doom 2016
<imirkin>
i got my dooms off-by-one
<imirkin>
i forgot about doom 3, i guess. and assumed the recent one *was* doom 3
<imirkin>
i had already stopped playing video games and hadn't started hacking on graphics :)
<jenatali>
The recent one is Eternal :P
<cheako>
This is more a comment on vulkan, but create_device has no direct access to the instance.
<jenatali>
And Eternal is Vulkan-only, so none of my GL work is going to make that work :D
<imirkin>
sigh. i guess 2016 isn't recent anymore.
<imirkin>
feels recent.
<imirkin>
wasn't that like last year?
<mattst88>
imirkin: all those prepandemic years basically run together at this point
Guest1812 has quit []
nchery has quit [Ping timeout: 480 seconds]
maxzor_ has joined #dri-devel
<graphitemaster>
Doesn't the driver already pad textures anyways for GOBs
<graphitemaster>
I never understood the limitations on POT textures actually, from a hardware perspective.
maxzor has quit [Remote host closed the connection]
<imirkin>
well, e.g. when you hit the border
maxzor has joined #dri-devel
<graphitemaster>
Because padding of the texture data is needed regardless of if it's POT or not since everything works in what ever tile size is.
<imirkin>
depends on the width of the texture
<imirkin>
not the GOB alignment
<graphitemaster>
If there's a good resource on that I'd like to read up on it.
<imirkin>
as for POT, earlier GPUs only supported POT textures to make minification simpler, i guess
Kayden has joined #dri-devel
<graphitemaster>
I always assumed it had something to do with border color
<graphitemaster>
Or wrapping behavior
<graphitemaster>
And maybe the interpolator being 24.8 bit, which it still is _shrug_
<imirkin>
just makes minifcation simpler
<graphitemaster>
Aside, is there any hardware that has more a precise texel interpolator, maybe exposed with an extension or something. I could really use something like that over the absolute mess which is my custom bilinear filtering
<graphitemaster>
a/has more a/has a more
ppascher has joined #dri-devel
<imirkin>
graphitemaster: the nvidia sampler descriptor has a lot of bits about this stuff
<graphitemaster>
> Given the improvements in global memory atomic performance, many atomics can be performed on devices of compute capability 3.x nearly as quickly as memory loads
<imirkin>
but we don't really expose them, at least in nouveau
<graphitemaster>
:|
<sravn>
marex: Took a closer look as the patch and sent my r-b so go ahead
anujp has joined #dri-devel
ngcortes has quit [Ping timeout: 480 seconds]
thellstrom1 has quit [Remote host closed the connection]
thellstrom has joined #dri-devel
maxzor has quit [Remote host closed the connection]
maxzor has joined #dri-devel
<imirkin>
graphitemaster: the other thing is that doing "swizzled" addressing is a lot simpler with POT
thellstrom has quit [Read error: Connection reset by peer]
<alyssa>
midgard is so cute
thellstrom has joined #dri-devel
mszyprow has joined #dri-devel
ds` has joined #dri-devel
gawin has quit [Ping timeout: 480 seconds]
Duke`` has quit [Ping timeout: 480 seconds]
<cheako>
I got my layer working!
alyssa has left #dri-devel [#dri-devel]
mvlad has quit [Remote host closed the connection]
gawin has joined #dri-devel
dreda has quit [Server closed connection]
dreda has joined #dri-devel
dreda is now known as Guest1825
Haaninjo has quit [Quit: Ex-Chat]
danvet has quit [Ping timeout: 480 seconds]
gouchi has quit [Remote host closed the connection]
dogukan has quit [Quit: Konversation terminated!]
maxzor has quit [Remote host closed the connection]
maxzor has joined #dri-devel
lcn_ has joined #dri-devel
lcn has quit [Remote host closed the connection]
lcn_ is now known as lcn
fxkamd has quit []
ngcortes has joined #dri-devel
ngcortes has quit []
ngcortes has joined #dri-devel
mlankhorst has quit [Ping timeout: 480 seconds]
<emersion>
daniels: damn, i should try haircuts too
maxzor has quit [Remote host closed the connection]