ChanServ changed the topic of #dri-devel to: <ajax> nothing involved with X should ever be unable to find a bar
columbarius has joined #dri-devel
sukbeom has quit [Ping timeout: 480 seconds]
co1umbarius has quit [Ping timeout: 480 seconds]
JohnnyonF has joined #dri-devel
zackr has quit [Remote host closed the connection]
JohnnyonFlame has quit [Ping timeout: 480 seconds]
nchery has quit [Quit: Leaving]
<anarsoul>
is it OK to introduce my own intrinsics in nir that will be used only in my backend?
<anarsoul>
I'm adding native support for texture2DProj to lima and the trick is that coords and projector need to be merged into a single source
<alyssa>
anarsoul: yes, provided it's suffixed _lima and there's a good reason to do so
<alyssa>
see the existing _v3d intrinsics etc
<anarsoul>
alyssa: well, there's load_varying instruction that divides x and y by w or z
<anarsoul>
it's output is fed directly into sampler instruction
<anarsoul>
it's also used if we need to load coords from a register
<anarsoul>
so my idea is to add a lowering pass that merges coords and projector into ssa = load_input_projected() if they come from the same varying, otherwise it converts them into ssa1 = vec3(coords, projector) ssa = load_reg_projected(ssa1)
<anarsoul>
lower_texture pass with lower_txp is just not good for lima, because it downgrades coords to fp16 even if coords and projector are in the same varying, see https://gitlab.freedesktop.org/mesa/mesa/-/issues/5385
<alyssa>
Yeah, that sounds reasonable.
pnowack has quit [Quit: pnowack]
idr has quit [Quit: Leaving]
fxkamd has quit []
rgallaispou1 has joined #dri-devel
rgallaispou has quit [Ping timeout: 480 seconds]
rgallaispou has joined #dri-devel
rgallaispou1 has quit [Read error: Connection reset by peer]
<dcbaker>
karolherbst: I've been tracking down meson + rust bugs this week since I've been working odd bursts of time. There's still an issue that IIRC has to do with rlib1 depends on rlib2, and should be rebuilt if rlib2 is rebuilt, is that correct?
orbea has quit [Remote host closed the connection]
orbea has joined #dri-devel
sdutt has quit [Ping timeout: 480 seconds]
thellstrom1 has joined #dri-devel
thellstrom has quit [Read error: Connection reset by peer]
rgallaispou has quit [Read error: Connection reset by peer]
rgallaispou1 has quit [Read error: Connection reset by peer]
rgallaispou has joined #dri-devel
Luc has joined #dri-devel
JohnnyonFlame has joined #dri-devel
JohnnyonF has quit [Ping timeout: 480 seconds]
jernej_ has joined #dri-devel
jernej has quit [Ping timeout: 480 seconds]
JohnnyonFlame has quit [Read error: Connection reset by peer]
Duke`` has joined #dri-devel
lemonzest has joined #dri-devel
rgallaispou1 has joined #dri-devel
rgallaispou1 has quit [Read error: Connection reset by peer]
rgallaispou has quit [Ping timeout: 480 seconds]
rgallaispou has joined #dri-devel
mlankhorst has joined #dri-devel
mattrope has quit [Read error: Connection reset by peer]
YuGiOhJCJ has joined #dri-devel
Luc has quit [Remote host closed the connection]
YuGiOhJCJ has quit []
Danct12 has quit [Quit: Quitting]
ella-0_ has joined #dri-devel
ella-0_ has quit []
ella-0_ has joined #dri-devel
ella-0_ has quit []
ella-0 has joined #dri-devel
rbrune has joined #dri-devel
X-Scale` has joined #dri-devel
X-Scale has quit [Ping timeout: 480 seconds]
gouchi has joined #dri-devel
nctcf^ has quit [Remote host closed the connection]
Luc has joined #dri-devel
Luc has quit [Remote host closed the connection]
danvet has joined #dri-devel
Ahuj has quit [Remote host closed the connection]
tobiasjakobi has joined #dri-devel
tobiasjakobi has quit [Remote host closed the connection]
DottorLeo has joined #dri-devel
<DottorLeo>
hi! I'm testing DOOM (2016) on my kaby lake laptop (yes i know it's very underpowered for that game ^^", i keep it to test Mesa). It has an openGL 4.5 and Vulkan 1.0 render. Shouldn't be the Vulkan render a bit faster than openGL? I don't expext a miracle but shouldn't be faster?
<FLHerne>
It is on most hardware
<DottorLeo>
i've tried it but seems a bit worse than OpenGL :/ù
pcercuei has joined #dri-devel
Luc has joined #dri-devel
DottorLeo has quit [Quit: Leaving]
alexeymin has quit [Remote host closed the connection]
Luc has quit [Remote host closed the connection]
alatiera has joined #dri-devel
alatiera is now known as Guest937
ella-0 has quit []
ella-0 has joined #dri-devel
sarnex has quit [Remote host closed the connection]
<mareko>
is it Kaby Lake G?
Hi-Angel has joined #dri-devel
gawin has joined #dri-devel
Hi-Angel has quit [Ping timeout: 480 seconds]
NiksDev has joined #dri-devel
gawin has quit [Ping timeout: 480 seconds]
Ahuj has joined #dri-devel
hikiko_bsd has joined #dri-devel
NiksDev has quit [Ping timeout: 480 seconds]
sukbeom has joined #dri-devel
alanc has quit [Remote host closed the connection]
alanc has joined #dri-devel
Company has joined #dri-devel
mlankhorst has quit [Ping timeout: 480 seconds]
sarnex has joined #dri-devel
gawin has joined #dri-devel
X-Scale has joined #dri-devel
X-Scale` has quit [Ping timeout: 480 seconds]
mattrope has joined #dri-devel
iive has joined #dri-devel
illwieckz has quit []
illwieckz has joined #dri-devel
hikiko_bsd has quit [Ping timeout: 480 seconds]
<gawin>
mareko: c standard is vague on this topic, maybe it'd be better to convert code to use offsetof?
agd5f has quit [Ping timeout: 480 seconds]
agx has quit [Read error: Connection reset by peer]
agx has joined #dri-devel
sukbeom has quit [Ping timeout: 480 seconds]
gawin has quit [Ping timeout: 480 seconds]
gouchi has quit [Quit: Quitte]
camus has quit []
sdutt has joined #dri-devel
<mareko>
I don't understand
<mareko>
&p->Base is like a c++ upcast equal to p
<mareko>
whatever is reporting the error should not be reporting the error
gawin has joined #dri-devel
rasterman has joined #dri-devel
<gawin>
mareko: I'm gonna try to explain one by one:
<gawin>
you're right probably fix in current form isn't helping, but c doesn't like calling access operator "->" on NULL
<gawin>
pointer, (even though there's not overwriting there's access to memory)
<gawin>
errors were reported by ubsan
sdutt has quit [Remote host closed the connection]
<mareko>
it's not -> because it's not dereferenced
<mareko>
it's most likely a bug in ubsan
<gawin>
I mean the issue is if "->" is implemented like (*(old_tcp)).Base
<mareko>
that's impossible due to &
<gawin>
in c isn't "->" faster evaluated than "&"?
<FLHerne>
Yes
<FLHerne>
&foo->bar is &(foo->bar), and NULL->bar is undefined behaviour even if you don't use the result
<DPA>
Would it make sense to add UDL to kmsro?
<mareko>
gawin: what? &-> is the same as "+"
<gawin>
where can I find something on this topic?
JohnnyonFlame has joined #dri-devel
sdutt has joined #dri-devel
<airlied>
mareko: unfortunately C isnt clear and at least gcc and clang used to disagree
sdutt has quit [Read error: Connection reset by peer]
Duke`` has quit [Ping timeout: 480 seconds]
<cwabbott>
lol, "light reading"
<cwabbott>
mareko: it's important to keep in mind that oftentimes even when a thing maps eventually to a particular hardware instruction, like &foo->bar mapping to +, there are many layers in between and so it doesn't necessarily have to "mean" the same thing, due to compiler developers wanting to optimize things
<cwabbott>
every language (e.g. C, SPIR-V, NIR, your backend, etc etc) defines its own meaning of the operations it contains, with an eye towards what it eventually gets mapped to but *also* what optimizations compilers want to do in the meantime
<cwabbott>
and sometimes it doesn't match what you'd naively expect, but there are usually good reasons for that
shashanks has quit [Ping timeout: 480 seconds]
flacks has joined #dri-devel
tobiasjakobi has joined #dri-devel
tobiasjakobi has quit [Remote host closed the connection]
gouchi has joined #dri-devel
gouchi has quit []
danvet has quit [Ping timeout: 480 seconds]
kts has joined #dri-devel
kts has quit [Quit: Konversation terminated!]
<DrNick>
but then you run headlong into the reality that #define offsetof(st, m) ((size_t)&(((st *)0)->m)) is the historical implementation of the offsetof macro and any compiler that doesn't compile that correctly is trash
gawin has quit [Ping timeout: 480 seconds]
rbrune has quit [Ping timeout: 480 seconds]
mattrope has quit [Read error: Connection reset by peer]