<airlied>
I'm not sure if something in llvm or translator changed
dliviu has quit []
<airlied>
not having much luck tracking down what is legal here for opencl
dliviu has joined #dri-devel
Lucretia has quit []
ngcortes has quit [Remote host closed the connection]
<jenatali>
airlied: I haven't re-sync'd dependencies in a long time. It definitely used to work
jessica_24 has quit [Quit: Connection closed for inactivity]
heat has quit [Ping timeout: 480 seconds]
sarnex has quit [Quit: Quit]
Company has quit [Quit: Leaving]
sarnex has joined #dri-devel
pnowack has joined #dri-devel
slattann has joined #dri-devel
sdutt has quit [Ping timeout: 480 seconds]
jewins has quit [Ping timeout: 480 seconds]
NiksDev has joined #dri-devel
gpoo has quit [Ping timeout: 480 seconds]
danvet has joined #dri-devel
pnowack has quit [Quit: pnowack]
Duke`` has joined #dri-devel
thellstrom has joined #dri-devel
<tomeu>
airlied: I was referring to how do you tell between expected and unexpected failures, as the fails list in mesa seems to be only about the ones in the subset that are run in CI
<airlied>
tomeu: i usually jsut diff the outputs
<tomeu>
airlied: what do you think about keeping in mesa the list of all failures?
<airlied>
since regressions should be easy to spot and therr shouldnt be too many flakes
frieder has joined #dri-devel
mlankhorst has joined #dri-devel
alanc has quit [Remote host closed the connection]
alanc has joined #dri-devel
<airlied>
well it depends on what cts version, but the list should be small
<tomeu>
in my machine, it would be at least 2 hours
<airlied>
tomeu: yeah I run it here, it takes usually an hour or so
<airlied>
make sure you use release builds of mesa/cts, but might be cool to keep result in CI and update using that change when CI CTS gets bumped I suppose
<tomeu>
yeah, the expectations files in mesa should relate to that mesa revision and whatever CTS version is being used
jkrzyszt has joined #dri-devel
Lucretia has joined #dri-devel
pochu has joined #dri-devel
aravind has joined #dri-devel
shfil has joined #dri-devel
iive has joined #dri-devel
rasterman has joined #dri-devel
iive has quit []
camus has joined #dri-devel
camus1 has quit [Read error: Connection reset by peer]
tursulin has joined #dri-devel
camus has quit [Remote host closed the connection]
camus has joined #dri-devel
lynxeye has joined #dri-devel
slattann has quit []
pcercuei has joined #dri-devel
slattann has joined #dri-devel
lemonzest has joined #dri-devel
Ahuj has joined #dri-devel
aravind has quit [Remote host closed the connection]
HayashiEsme[m] has joined #dri-devel
aravind has joined #dri-devel
pnowack has joined #dri-devel
pnowack has quit [Remote host closed the connection]
pnowack has joined #dri-devel
Surkow|laptop has quit [Quit: 418 I'm a teapot - NOP NOP NOP]
vivijim has joined #dri-devel
Surkow|laptop has joined #dri-devel
YuGiOhJCJ has joined #dri-devel
krh has quit []
krh has joined #dri-devel
jhli has quit [Remote host closed the connection]
frieder has quit [Remote host closed the connection]
rg3igalia has quit []
rg3igalia has joined #dri-devel
Danct12 has quit [Remote host closed the connection]
tfiga has quit []
tfiga has joined #dri-devel
frieder has joined #dri-devel
gpoo has joined #dri-devel
alatiera6 has joined #dri-devel
alatiera has quit [Ping timeout: 480 seconds]
adjtm has quit [Quit: Leaving]
slattann has quit []
Company has joined #dri-devel
Ahuj has quit [Ping timeout: 480 seconds]
sdutt has joined #dri-devel
pochu has quit [Ping timeout: 480 seconds]
JohnnyonFlame has joined #dri-devel
camus has quit [Remote host closed the connection]
camus has joined #dri-devel
robmur01 has joined #dri-devel
iive has joined #dri-devel
heat has joined #dri-devel
ppascher has quit [Quit: Gateway shutdown]
ppascher has joined #dri-devel
JohnnyonFlame has quit [Read error: Connection reset by peer]
aravind has quit []
JohnnyonFlame has joined #dri-devel
<Venemo>
hey guys
<Venemo>
why do nir_op_b2b1 and nir_op_i2b1 exist? there are very few places where these are emitted, and these opcodes are pretty much equivalent to ine with 0
<Venemo>
I already looked at that and it doesn't answer my question
<Venemo>
what is the difference between ine(a, 0) and b2b1(a) ?
<pendingchaos>
b2b1/b2b32 convert between backend-specific 8/16/32-bit boolean representations
<pendingchaos>
so I think the only requirement is that "b2b1(b2b32(a)) == a"
<pendingchaos>
for example, b2b32() could convert true to either -1 or 1, and b2b1() is only expected to handle one of those
<pendingchaos>
though if b2b32() doesn't convert true to -1, I think you will want b2b1() to handle -1 anyways because that's what b2b32()'s constant folding does
<Venemo>
I think for ACO, a better solution would be to simply lower these to ine/ieq in NIR instead of handling them in the backend
<Venemo>
because that's what we emit anyway
<pendingchaos>
maybe, if there's some opts that would work with better ine/i2b1 and b2i32 instead of b2b1/b2b32
<Venemo>
then, they could potentially benefit from algebraic opts that target ine
<Venemo>
yeah
<Venemo>
I just saw a pattern like this in a CTS test this morning, though it's a good question if any real app uses this
JohnnyonF has joined #dri-devel
YuGiOhJCJ has quit [Quit: YuGiOhJCJ]
JohnnyonFlame has quit [Ping timeout: 480 seconds]
<dschuermann>
I wonder if we could migrate that idea to ballot() and inverse_ballot()
<dschuermann>
i.e. store shared bools as 1bit per lane :)
<Venemo>
are there any backends that still use 32-bit booleans?
jkrzyszt has quit [Remote host closed the connection]
adjtm has joined #dri-devel
adjtm has quit [Remote host closed the connection]
jkrzyszt has joined #dri-devel
Terman has quit [Remote host closed the connection]
lplc has quit [Ping timeout: 480 seconds]
lplc has joined #dri-devel
slattann has joined #dri-devel
lplc has quit [Ping timeout: 480 seconds]
lplc has joined #dri-devel
shfil has quit [Quit: Konversation terminated!]
lplc has quit [Ping timeout: 480 seconds]
lplc has joined #dri-devel
frieder has quit [Remote host closed the connection]
<graphitemaster>
This may be a dumb question, but does GLSL support swizzle into an out param
rcf has quit [Quit: WeeChat 3.3-dev]
<graphitemaster>
Like void foo(out vec3 x) { ... } foo(thing.yzw);
<graphitemaster>
I guess a better question is IF that should work
rcf has joined #dri-devel
lynxeye has quit []
lplc has quit [Ping timeout: 480 seconds]
slattann has quit []
<imirkin>
graphitemaster: what happens when you do foo(thing.xxx)?
lplc has joined #dri-devel
slattann has joined #dri-devel
shfil has joined #dri-devel
<bl4ckb0ne>
are VK_KHR_surface and VK_KHR_display always guaranteed to be available?
<imirkin>
(that said, i think you can do thing.xyz = foo, but what happens when you do thing.xxx = foo? dunno)
lplc has quit [Ping timeout: 480 seconds]
lplc has joined #dri-devel
slattann has quit []
slattann has joined #dri-devel
tobiasjakobi has joined #dri-devel
lplc has quit [Ping timeout: 480 seconds]
lplc has joined #dri-devel
JohnnyonFlame has joined #dri-devel
jhli has joined #dri-devel
<airlied>
most dsts are writemasks notbswizzles
JohnnyonF has quit [Ping timeout: 480 seconds]
tobiasjakobi has quit [Remote host closed the connection]
mbrost has joined #dri-devel
tzimmermann has joined #dri-devel
<tzimmermann>
airlied, danvet, drm-misc-fixes is empty. so no PR this week
<imirkin>
airlied: in reality, yea ... but i dunno what glsl specifies
<imirkin>
iirc glsl compiler has special logic to peer through swizzles (which are ir_rvalue's)
<graphitemaster>
I mean it appears to work
<graphitemaster>
Is it UB though
<imirkin>
graphitemaster: it's the equivalent question of whether "a.yzw = b.xyz" is legal
<imirkin>
if that's legal, then your thing is legal
<imirkin>
i don't know the answer to the first thing though
<ajax>
airlied: how'd you test? my fedora 34 Xwayland doesn't have shm pixmaps
<ajax>
Xvfb would i suppose but i didn't think it had DRI3
hch12907 has quit [Ping timeout: 480 seconds]
hch12907 has joined #dri-devel
mbrost has quit [Ping timeout: 480 seconds]
rasterman has quit [Quit: Gettin' stinky!]
Duke`` has quit [Ping timeout: 480 seconds]
mbrost has joined #dri-devel
<airlied>
ajax: logged into xorg session
jessica_24 has joined #dri-devel
ngcortes has joined #dri-devel
adjtm has quit [Ping timeout: 480 seconds]
rasterman has joined #dri-devel
danvet has quit [Ping timeout: 480 seconds]
mlankhorst has quit [Ping timeout: 480 seconds]
i-garrison has quit []
thellstrom has quit [Read error: Connection reset by peer]
thellstrom has joined #dri-devel
pnowack has quit [Quit: pnowack]
V has joined #dri-devel
alatiera65 has joined #dri-devel
mbrost has quit [Ping timeout: 480 seconds]
alatiera6 has quit [Ping timeout: 480 seconds]
mbrost has joined #dri-devel
rasterman has quit [Quit: Gettin' stinky!]
co1umbarius has quit [Ping timeout: 480 seconds]
Bennett has joined #dri-devel
NiksDev has quit [Ping timeout: 480 seconds]
co1umbarius has joined #dri-devel
vivijim has quit [Ping timeout: 480 seconds]
pcercuei has quit [Quit: dodo]
shfil has quit [Ping timeout: 480 seconds]
iive has quit []
heat has quit [Read error: Connection reset by peer]
jkrzyszt has quit [Ping timeout: 480 seconds]
tursulin has quit [Read error: Connection reset by peer]
<imirkin>
mattst88: i built the gentoo 21.2.1 ebuild with +intel +i965 -iris, and ended up with something that still tries to load iris by default (for SKL)
<imirkin>
i think there was some fix to the mesa build to avoid setting prefer-iris to true if iris isn't being built or something?
<mattst88>
yeah, probably
<mattst88>
I also need to wire up crocus
<imirkin>
anyways, it leads to a slightly confusing situation if you keep your existing VIDEO_CARDS setting around
<mattst88>
yeah
<mattst88>
I'm eagerly awaiting the day where we don't have multiple drivers for the same hardware
<imirkin>
well, i was also surprised it wasn't enabled by +intel
<mattst88>
that what wasn't enabled by +intel?
<imirkin>
iris
<mattst88>
ah.
<mattst88>
yeah, that's definitely an oversight
<mattst88>
in general VIDEO_CARDS=... is a nightmare, since there are a bunch of similarly named things
<mattst88>
amdgpu and radeonsi
<mattst88>
which do you need? who knows! probably both!
<imirkin>
yeah. i had intel and i965 enabled :)
<mattst88>
I thought you used ironlake? did someone get a new machine? :)
<imirkin>
at work i have skl
<mattst88>
ahh
<imirkin>
at home i have i7-920 without a (intel) gpu, and a snb laptop