ChanServ changed the topic of #dri-devel to: <ajax> nothing involved with X should ever be unable to find a bar
nchery is now known as Guest2588
nchery has joined #dri-devel
Guest2588 has quit [Ping timeout: 480 seconds]
anholt__ has joined #dri-devel
iive has quit []
anholt_ has quit [Ping timeout: 480 seconds]
co1umbarius has joined #dri-devel
columbarius has quit [Ping timeout: 480 seconds]
ngcortes has quit [Remote host closed the connection]
ngcortes has joined #dri-devel
JohnnyonFlame has joined #dri-devel
<karolherbst> ehhh
<karolherbst> clc can't parse spirvs...
<karolherbst> I mean.. non llvm spirvs
ybogdano has quit [Ping timeout: 480 seconds]
Emmy_ has quit [Remote host closed the connection]
Daanct12 has joined #dri-devel
Emmy_ has joined #dri-devel
ngcortes has quit [Remote host closed the connection]
fxkamd has quit []
Emantor has quit [Quit: ZNC - http://znc.in]
Emantor has joined #dri-devel
anarsoul has quit [Remote host closed the connection]
anarsoul has joined #dri-devel
camus has quit []
sdutt_ has joined #dri-devel
camus has joined #dri-devel
jhli_ has joined #dri-devel
aswar002_ has joined #dri-devel
mdnavare_ has joined #dri-devel
rsripada_ has joined #dri-devel
aknautiy has quit [Remote host closed the connection]
aknautiy has joined #dri-devel
Ryback_ has quit [Remote host closed the connection]
unerlige has quit [Remote host closed the connection]
Ryback_ has joined #dri-devel
unerlige has joined #dri-devel
jhli has quit [Ping timeout: 480 seconds]
aswar002 has quit [Ping timeout: 480 seconds]
mdnavare has quit [Ping timeout: 480 seconds]
rsripada has quit [Ping timeout: 480 seconds]
sdutt has quit [Ping timeout: 480 seconds]
heat has joined #dri-devel
LexSfX has quit [Ping timeout: 480 seconds]
LexSfX has joined #dri-devel
jewins1 has joined #dri-devel
ramaling has quit [Remote host closed the connection]
ramaling has joined #dri-devel
jewins has quit [Ping timeout: 480 seconds]
HankB has quit [Remote host closed the connection]
HankB has joined #dri-devel
anujp has quit [Ping timeout: 480 seconds]
Lyude has quit [Quit: WeeChat 3.4]
Lyude has joined #dri-devel
mhenning has joined #dri-devel
oneforall2 has quit [Remote host closed the connection]
oneforall2 has joined #dri-devel
heat has quit [Ping timeout: 480 seconds]
ppascher has joined #dri-devel
Company has quit [Quit: Leaving]
HankB has quit [Remote host closed the connection]
HankB has joined #dri-devel
morphis has quit []
mhenning has quit [Quit: mhenning]
cheako has quit [Quit: Connection closed for inactivity]
Daanct12 has quit [Remote host closed the connection]
Kayden is now known as Guest2607
Duke`` has joined #dri-devel
jewins1 has quit [Read error: Connection reset by peer]
Guest2607 is now known as Kayden
sdutt_ has quit [Read error: Connection reset by peer]
bond has joined #dri-devel
bond is now known as Guest2609
camus has quit [Ping timeout: 480 seconds]
rpigott has quit [Remote host closed the connection]
rpigott has joined #dri-devel
cleverca22[m] has joined #dri-devel
Guest2609 has quit []
camus has joined #dri-devel
lplc has quit [Ping timeout: 480 seconds]
rasterman has joined #dri-devel
YuGiOhJCJ has joined #dri-devel
lplc has joined #dri-devel
enunes has quit [Remote host closed the connection]
gouchi has joined #dri-devel
enunes has joined #dri-devel
bond has joined #dri-devel
bond is now known as Guest2620
Haaninjo has joined #dri-devel
digetx is now known as Guest2623
Guest2623 has quit [Remote host closed the connection]
digetx has joined #dri-devel
Guest2620 has quit []
maxzor has quit [Ping timeout: 480 seconds]
YuGiOhJCJ has quit [Quit: YuGiOhJCJ]
pcercuei has joined #dri-devel
HankB has quit [Remote host closed the connection]
HankB has joined #dri-devel
tobiasjakobi has joined #dri-devel
HankB has quit [Remote host closed the connection]
HankB has joined #dri-devel
flacks has quit [Quit: Quitter]
flacks has joined #dri-devel
tobiasjakobi has quit []
pcercuei has quit [Quit: brb]
sagar__ has quit [Remote host closed the connection]
sagar__ has joined #dri-devel
HankB has quit [Remote host closed the connection]
HankB has joined #dri-devel
ella-0_ has joined #dri-devel
ella-0 has quit [Read error: Connection reset by peer]
sravn has quit [Quit: WeeChat 3.3]
pcercuei has joined #dri-devel
Lucretia has quit [Read error: Connection reset by peer]
rkanwal has joined #dri-devel
Lucretia has joined #dri-devel
icecream95 has quit [Ping timeout: 480 seconds]
rkanwal has quit [Quit: rkanwal]
cheako has joined #dri-devel
nchery has quit [Read error: Connection reset by peer]
maxzor has joined #dri-devel
Duke`` has quit []
Duke`` has joined #dri-devel
sravn has joined #dri-devel
Company has joined #dri-devel
rasterman has quit [Quit: Gettin' stinky!]
<karolherbst> jekstrand: ohh no... I just noticed today that I am leaking cl_kernels :D
<karolherbst> release did a retain
<karolherbst> and piglit assumes support for denormals :(
<karolherbst> heh
<karolherbst> and now the one fail I was looking into fails the same on Intels stack
<karolherbst> :D
<karolherbst> yeah well..
<karolherbst> guess I only fix the API bugs then
Duke`` has quit [Ping timeout: 480 seconds]
Duke`` has joined #dri-devel
heat has joined #dri-devel
<karolherbst> jenatali: ... I think we don't implement CL_KERNEL_ATTRIBUTES, do we? :D
<karolherbst> (the CTS test doesn't fail if an impl doesn't(
<jenatali> Correct
<jenatali> I filled a CTS bug on that missing
<karolherbst> ahh
<karolherbst> I was trying to implement it, but clc doesn't contain the info that
<karolherbst> *for
<jenatali> Doesn't it?
<karolherbst> I don't think so
<jenatali> What's missing?
<karolherbst> would expect it on clc_kernel_info, but there is nothing
<karolherbst> or is the vec_hint the only possible attribute?
<jenatali> Yeah it is
<karolherbst> ahh
<karolherbst> so I just need to construct the string based on that
<jenatali> At least the only one that spir-v has
<karolherbst> right
<karolherbst> but with that we can't implement it correctly
heat has quit [Read error: Connection reset by peer]
<karolherbst> ohh .. I should have finished reading the sentence
<karolherbst> surrounding whitespace and embedded newlines should be removed
heat has joined #dri-devel
<karolherbst> what about double spaces?
<karolherbst> jenatali: btw.. there are more
<karolherbst> vec_type_hint, work_group_size_hint and reqd_work_group_size
<jenatali> Oh right but those you need anyway
<karolherbst> yeah..
<karolherbst> I mean.. I think the API is stupid to tell us to "return the string as they were declared" but...
<jenatali> Yup
<karolherbst> technically we are not able to reconstruct what we've got there :( oh well
<karolherbst> *gotten
<karolherbst> ah well
<karolherbst> as long as the CTS is happy I don't really care
<karolherbst> ohh the CTS only checks for vec_type_hint
<karolherbst> seriously.. this CTS
<jenatali> Yep
<jenatali> And only on spir or spir-v kernels
<karolherbst> it does test it on real OpenCL C ones
<karolherbst> OpenCL-CTS/test_conformance/api/test_kernel_attributes.cpp
aswar002_ has quit [Ping timeout: 480 seconds]
<jenatali> Oh that must be new
<karolherbst> could be
<karolherbst> "Date: Thu May 13 09:20:45 2021 +0100"
<karolherbst> jenatali: uhh.. it actually tests all three
<karolherbst> mhh but the translator doesn't return it as a string anyway
<karolherbst> guess I'll just return it without any spaces and stuff
<jenatali> Yeah it gets lost in translation pretty early
<karolherbst> I am a bit surprised that vec_type_hint is not deprecated as all GPU have scalar ISAs anyway :D
<karolherbst> I am actually wondering where setting it does indeed make a difference
camus has quit []
lichun has joined #dri-devel
enunes has quit [Quit: ZNC - https://znc.in]
lichun has quit []
iive has joined #dri-devel
jernej has quit [Ping timeout: 480 seconds]
mimi89999 has joined #dri-devel
<mimi89999> Hello
<mimi89999> I'm getting `fatal error: cannot open file '/usr/lib/clc/polaris10-amdgcn-mesa-mesa3d.bc': No such file or directory` when attempting to use OpenCL on my GPU.
maxzor has quit [Ping timeout: 480 seconds]
jernej has joined #dri-devel
sdutt has joined #dri-devel
<HdkR> mimi89999: Looks like your libclc package didn't provide that file for whatever reason
<mimi89999> Is it normally part of libclc?
<karolherbst> "It is registered with the number #1" :D
<karolherbst> Guess I am just very early this year
<HdkR> mimi89999: Yes
<FLHerne> Arch package has /usr/share/clc/polaris10-amdgcn-mesa-mesa3d.bc
<karolherbst> ...
<FLHerne> not sure if /usr/share vs /usr/lib matters when searching for it
<karolherbst> it does
* FLHerne doesn't have an AMD gpu handy to play with
<karolherbst> FLHerne: what's the content of /usr/lib64/pkgconfig/libclc.pc ?
<karolherbst> I think we parse the location out of that file
<karolherbst> but not 100% sure
<karolherbst> libexecdir has to match obviously
<karolherbst> ahh yes, we parse the pkgconfig file
<FLHerne> pkgconfig file isn't at that path, it's /usr/share/pkgconfig/libclc.pc , but yes `/usr/share/pkgconfig/libclc.pc`
<FLHerne> oops
<FLHerne> `libexecdir=/usr/share/clc`
<karolherbst> yeah, then stuff should just work (tm)
<FLHerne> mimi89999: what distro/package (or other source of libclc) are you using?
RSpliet has quit [Quit: Bye bye man, bye bye]
RSpliet has joined #dri-devel
alanc has quit [Remote host closed the connection]
alanc has joined #dri-devel
<mimi89999> I think I was just missing `libclc-amdgcn`
<mimi89999> Sorry
mimi89999 has left #dri-devel [#dri-devel]
heat has quit [Remote host closed the connection]
heat has joined #dri-devel
<karolherbst> jenatali: .... uchar != char :(
<karolherbst> I think we have to expand clc_vec_hint_type a little as well
<jenatali> IIRC we don't have that info
<karolherbst> annoying
<karolherbst> but seems like you are right
<karolherbst> jenatali: mhhh the CTS only tests for unsigned...
<jenatali> That works
<karolherbst> :D
<karolherbst> I hate my for doing that :D
<karolherbst> *me
<jenatali> The whole point is just the size and count so you can vectorize, the signedness doesn't affect how you'd vectorize
<karolherbst> I know
<karolherbst> but the spec is the spec is the spec
<karolherbst> but..
<karolherbst> given that like literally 99.99% of the hw used with compute is scalar anyway
<karolherbst> can't really think of a real optimization we could do based on that information
<karolherbst> just something with memory banks and layouting, maybe even making use of cache lines... but...
bgs_ has quit [Remote host closed the connection]
bgs has joined #dri-devel
<karolherbst> btw, I am already a bit nervous since michael last article :D
<karolherbst> ehh.. uhm... wrong channel, but still
gouchi has quit [Remote host closed the connection]
Duke`` has quit [Ping timeout: 480 seconds]
anujp has joined #dri-devel
Haaninjo has quit [Quit: Ex-Chat]
kchibisov has quit [Read error: No route to host]
kchibisov has joined #dri-devel
sdutt has quit [Ping timeout: 480 seconds]
pcercuei has quit [Quit: dodo]
heat_ has joined #dri-devel
heat has quit [Read error: Connection reset by peer]