<tzimmermann>
javierm, do you want to take another look at the num_planes patchset? i only added the refactoring in gem-vram that you mentioned
<MrCooper>
feaneron: can you run the app with the environment variables LIBGL_DEBUG=verbose EGL_LOG_LEVEL=debug and pastebin the resulting stderr output?
<javierm>
tzimmermann: sure, let me look
<tzimmermann>
javierm, i forgot about that subtil typo :)
<tzimmermann>
i'll fix it before landing
frankbinns1 has quit []
<javierm>
tzimmermann: hehe, no worries. Old english ftw :)
frankbinns has joined #dri-devel
Daanct12 has joined #dri-devel
hch12907 has quit [Ping timeout: 480 seconds]
<javierm>
tzimmermann: the patches look good to me. Also provided my r-b to the one that didn't have already
JohnnyonFlame has quit [Read error: Connection reset by peer]
<tzimmermann>
javierm, thanks
hch12907 has joined #dri-devel
whald has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
jkrzyszt has joined #dri-devel
co1umbarius has joined #dri-devel
itoral has quit [Remote host closed the connection]
columbarius has quit [Ping timeout: 480 seconds]
itoral has joined #dri-devel
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
pcercuei has joined #dri-devel
nchery has quit [Remote host closed the connection]
nchery has joined #dri-devel
itoral has quit [Remote host closed the connection]
MajorBiscuit has quit [Remote host closed the connection]
itoral has joined #dri-devel
MajorBiscuit has joined #dri-devel
ceyusa has joined #dri-devel
nchery has quit [Ping timeout: 480 seconds]
sdutt has quit [Ping timeout: 480 seconds]
bmodem has quit [Ping timeout: 480 seconds]
lumag_ has quit [Remote host closed the connection]
lumag_ has joined #dri-devel
rasterman- has joined #dri-devel
rasterman has quit [Read error: No route to host]
itoral has quit [Remote host closed the connection]
bmodem has joined #dri-devel
itoral has joined #dri-devel
bmodem has quit [Ping timeout: 480 seconds]
nchery has joined #dri-devel
columbarius has joined #dri-devel
co1umbarius 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
icecream95 has quit [Ping timeout: 480 seconds]
itoral has quit [Remote host closed the connection]
itoral has joined #dri-devel
MajorBiscuit has quit [Ping timeout: 480 seconds]
itoral has quit [Remote host closed the connection]
devilhorns has joined #dri-devel
itoral has joined #dri-devel
flacks has quit [Quit: Quitter]
MajorBiscuit has joined #dri-devel
flacks has joined #dri-devel
robertfoss[m] has joined #dri-devel
bmodem has joined #dri-devel
dllud has joined #dri-devel
dllud_ has quit [Read error: Connection reset by peer]
hch12907 has quit [Ping timeout: 480 seconds]
shadeslayer is now known as Guest796
shadeslayer has joined #dri-devel
rasterman- has quit [Read error: No route to host]
<feaneron>
the problematic path happens in a thread, and afaics there's no debug information coming from it
<feaneron>
yesterday i thought it was using the wrong EGLConfig, with a mismatching format, but that doesn't seem to be the case - it seems to be matching GBM_FORMAT_ARGB8888, so we're good
<feaneron>
though it's probably the case of using the EGL_KHR_no_config_context extension
gouchi has joined #dri-devel
gouchi has quit []
Company has joined #dri-devel
ahajda has joined #dri-devel
Surkow|laptop has quit [Ping timeout: 480 seconds]
Surkow|laptop has joined #dri-devel
aravind has quit [Ping timeout: 480 seconds]
maxzor has quit [Ping timeout: 480 seconds]
itoral has quit []
maxzor has joined #dri-devel
<MrCooper>
feaneron: looks like there are two EGL contexts, one created by OBS itself which uses llvmpipe (not clear why from the output), one from the pipewire capture plugin which uses iris
<feaneron>
that's correct - the graphics thread has a dedicated context
<feaneron>
the other context is probably created by qt
<MrCooper>
curiously, one of the backtraces you pasted on #gnome-shell showed llvmpipe symbols in iris_dri.so, I wonder if there's some kind of bad interaction between the two *_dri.so objects
gouchi has joined #dri-devel
gouchi has quit [Remote host closed the connection]
<MrCooper>
scratch that, should be one and the same file
<feaneron>
hm, even with MESA_LOADER_DRIVER_OVERRIDE=iris, one of the egl contexts is llvmpipe
kts has joined #dri-devel
<feaneron>
is it too problematic that there are 2 threads interacting with egl?
<feaneron>
oddly enough, this crash doesn't happen when using egl+x11 and egl+wayland. it's specific to egl+gbm
<MrCooper>
shouldn't be an issue in general, so long as they don't try to make the same EGL context current at the same time
sdutt has joined #dri-devel
<feaneron>
yeah, i'm pretty sure that doesn't happen
gouchi has joined #dri-devel
gouchi has quit []
<MrCooper>
feaneron: ah, that could explain it — the EGL GBM platform requires DRM master for HW acceleration, which isn't possible in a graphical session (the display server is already DRM master, so OBS cannot become it)
<feaneron>
oh no :(
<feaneron>
so it's not possible to use gbm in clients, only compositors?
<MrCooper>
I mean, that could explain why it uses llvmpipe, it probably shouldn't crash anyway
<shadeslayer>
^^
<feaneron>
i guess that's a dead end for me then :/
<MrCooper>
feaneron: GBM directly yes, the EGL GBM platform no AFAIK; I think you can use the surfaceless platform instead
<MrCooper>
the GBM platform is for display servers
<feaneron>
hm, i'll check how to use gbm with surfaceless then
<feaneron>
using eglExportDMABUFImageQueryMESA() is not an option apparently, so it needs to use gbm at some point
<pq>
you allocate gbm_bo with GBM, and then *independently* of GBM you initialize EGL to use the surfaceless platform.
<MrCooper>
feaneron: then again, Xwayland seems to use the GBM platform, which contradicts what I wrote above
<feaneron>
right. then i guess i'd have to import the gbm_bo into EGL through DMA-BUF, and use fbos to render on it?
sarnex_ has quit []
sarnex has joined #dri-devel
gouchi has joined #dri-devel
gouchi has quit [Remote host closed the connection]
<pq>
feaneron, yes
<pq>
MrCooper, maybe llvmpipe on EGL GBM platform tries to allocate dumb buffers? Maybe that is a privileged operation? While a hardware driver on EGL GBM platform can allocate on a render node.
ella-0_ has joined #dri-devel
ella-0 has quit [Read error: Connection reset by peer]
<MrCooper>
pq: that would still leave the question why it uses llvmpipe in the first place though
<feaneron>
the graphics thread, which is the one crashing, is using iris though
sdutt has quit [Ping timeout: 480 seconds]
<feaneron>
i'm trying pq's gbm + egl image + dma-buf importing suggestion, let's see how it goes
<MrCooper>
feaneron: the crash happens in llvmpipe code
<feaneron>
even the one in iris_dri.so? is that shared code?
<MrCooper>
yes, all *_dri.so are hardlinks to the same file
heat has joined #dri-devel
<feaneron>
welp, pq's suggestion gets us the exact same crash
<feaneron>
here i'm creating a gbm_bo, then an EGLImage from it (through DMA-BUF), then i bind it to a GL texture, and finally this GL texture is used by an fbo
<feaneron>
hmm
<feaneron>
if i load the gbm_device from /dev/dri/renderD128 instead of "/dev/dri/card0, it fails to create the gbm_bo!
sdutt has joined #dri-devel
sdutt has quit []
<zmike>
jekstrand: does this look right for an algebraic rule for rewriting unpack_64_2x32( bcsel( a, pack_64_2x32(b), pack_64_2x32(c) ) ) -> bcsel(a, b, c) :
<zmike>
(('unpack_64_2x32', ('bcsel', a, ('pack_64_2x32', b), ('pack_64_2x32', c))), ('bcsel', a, b, c)),
Danct12 has quit [Remote host closed the connection]
kts has joined #dri-devel
sarnex has quit [Ping timeout: 480 seconds]
Danct12 has joined #dri-devel
sarnex has joined #dri-devel
kts has quit [Quit: Konversation terminated!]
sdutt has joined #dri-devel
<zmike>
hm I guess that's not quite what I want either since the bcsels can be nested infinitely
Danct12 has quit [Remote host closed the connection]
Daanct12 has quit [Ping timeout: 480 seconds]
tobiasjakobi has joined #dri-devel
tobiasjakobi has quit []
mbrost has joined #dri-devel
aravind has joined #dri-devel
stuart has joined #dri-devel
MajorBiscuit has quit [Ping timeout: 480 seconds]
rgallaispou has quit [Ping timeout: 480 seconds]
tobiasjakobi has joined #dri-devel
tobiasjakobi has quit []
lumag_ has quit [Remote host closed the connection]
lumag_ has joined #dri-devel
qyliss has quit [Quit: bye]
lumag_ has quit [Remote host closed the connection]
lumag_ has joined #dri-devel
qyliss has joined #dri-devel
alyssa has joined #dri-devel
<alyssa>
../src/panfrost/bifrost/bifrost_compile.c:165:54: error: static_assert with no message is a C++17 extension [-Werror,-Wc++17-extensions]
<alyssa>
...Eh?
<glehmann>
zmike: Maybe you want something like these two? (('unpack_64_2x32', ('bcsel', a, b, ('pack_64_2x32', c))), ('bcsel', a, ('unpack_64_2x32', b), c)), (('unpack_64_2x32', ('bcsel', a, ('pack_64_2x32', b), c)), ('bcsel', a, b, ('unpack_64_2x32', c))),
<alyssa>
apparently it's supposed to take a message.
<zmike>
glehmann: I think maybe I just need to run lower_int64 repeatedly
<jannau>
error message looks strange but static_assert without message is only C23, would have expected to be a gnu extension though
<alyssa>
zmike: that shouldn't be possible
<zmike>
zink: making the impossible possible
<alyssa>
jannau: it's working on my gnu extension hell compiler (no warning, just not on the debian-android build)
<alyssa>
s/possible/necessary/
<alyssa>
jannau: so I wonder what warning flag I need for conformant C11 behaviour of the compiler then
tzimmermann has quit [Quit: Leaving]
<jannau>
'-std=c11' could be enough
frieder has quit [Remote host closed the connection]
EasyIP2023 has joined #dri-devel
<alyssa>
apparently not.. I just added a message :p
EasyIP2023 has left #dri-devel [#dri-devel]
Danct12 has joined #dri-devel
hch12907 has quit [Ping timeout: 480 seconds]
heat has quit [Read error: Connection reset by peer]
heat_ has joined #dri-devel
jkrzyszt has quit [Ping timeout: 480 seconds]
jeeeun84 has quit []
eukara_ has quit []
aravind has quit [Ping timeout: 480 seconds]
sdutt has quit [Ping timeout: 480 seconds]
jeeeun84 has joined #dri-devel
<jekstrand>
dcbaker: How do I tell meson where to find llvm-config?
devilhorns has quit []
<dcbaker>
jekstrand: with a native config add `[binaries]\nllvm-config = <path to llvm config>` then run meson with `meson setup --native-file <native file> builddir`
eukara has joined #dri-devel
sdutt has joined #dri-devel
sdutt has quit [Read error: Connection reset by peer]
Duke`` has joined #dri-devel
<feaneron>
MrCooper: hm, i just learned about drmGetMagic() and drmAuthenticateMagic(), is it required for clients to call this pair of functions?
<feaneron>
actually, i guess it's only useful for drm master...
ppascher has quit [Ping timeout: 480 seconds]
<jekstrand>
dcbaker: Thanks! Git it building... mostly. Now I need to figure out why LLVM is missing symbols. :-;/
<dcbaker>
is it missing LLVMPasses by chance?
<dcbaker>
I got a report in meson about mesa failing to build with very recent llvm because we're not linking that lib, but I haven't had a chance to figure out if that's our fault or llvm-configs
<feaneron>
AH wait!!! if i pass an fd to /dev/dri/card0 to gbm_create_device(), it tries to create a primary node; if i pass /dev/dri/renderD128 instead, it tries to create a render node
<feaneron>
okay
<feaneron>
and egl fails to create a DRI screen
<feaneron>
aha
<feaneron>
it tries to use iris; dri2 fails to create the dri screen; it falls back to llvmpipe
mszyprow has quit [Ping timeout: 480 seconds]
<airlied>
jekstrand: if its coroutines, there is an mr
lynxeye has quit [Quit: Leaving.]
sdutt has joined #dri-devel
co1umbarius has joined #dri-devel
columbarius has quit [Ping timeout: 480 seconds]
maxzor has quit [Remote host closed the connection]
maxzor has joined #dri-devel
Haaninjo has joined #dri-devel
ppascher has joined #dri-devel
Surkow|laptop is now known as Surkow
mszyprow has joined #dri-devel
tobiasjakobi has joined #dri-devel
tobiasjakobi has quit []
danvet has quit [Ping timeout: 480 seconds]
rasterman has quit [Quit: Gettin' stinky!]
mszyprow has quit [Ping timeout: 480 seconds]
heat_ has quit [Read error: Connection reset by peer]
heat_ has joined #dri-devel
soreau has quit [Quit: Leaving]
soreau has joined #dri-devel
maxzor has quit []
mvlad has quit [Remote host closed the connection]
ahajda has quit [Remote host closed the connection]
mclasen has quit []
mclasen has joined #dri-devel
Duke`` has quit [Ping timeout: 480 seconds]
mbrost_ has joined #dri-devel
mclasen has quit []
mclasen has joined #dri-devel
<zmike>
dcbaker: \o/
mbrost has quit [Ping timeout: 480 seconds]
mszyprow has joined #dri-devel
<dcbaker>
zmike: thanks for all the backports
<zmike>
what a release!
<dcbaker>
it's the biggest release we've had in a while for sure
<zmike>
monumental
<dcbaker>
You're just saying that because it has kopper in it :D
<zmike>
I am
janesma has joined #dri-devel
mclasen has quit []
mclasen has joined #dri-devel
AndrewR has quit [Ping timeout: 480 seconds]
mszyprow has quit [Ping timeout: 480 seconds]
ngcortes has joined #dri-devel
mclasen has quit []
mclasen has joined #dri-devel
AndrewR has joined #dri-devel
gawin has joined #dri-devel
alanc has quit [Remote host closed the connection]
alanc has joined #dri-devel
LexSfX has quit []
LexSfX has joined #dri-devel
sdutt_ has joined #dri-devel
sdutt has quit [Read error: Connection reset by peer]
heat_ has quit [Remote host closed the connection]
heat_ has joined #dri-devel
tursulin has quit [Remote host closed the connection]
Haaninjo has quit [Quit: Ex-Chat]
lemonzest has quit [Quit: WeeChat 3.4]
heat_ has quit [Read error: No route to host]
heat has joined #dri-devel
gawin has quit [Ping timeout: 480 seconds]
morphis has quit [Ping timeout: 480 seconds]
morphis has joined #dri-devel
guru_ has quit [Remote host closed the connection]
oneforall2 has joined #dri-devel
lumag_ has quit [Remote host closed the connection]