01:10 karolherbst: imirkin: if you want I could test that PIPE_CAP_MAX_TEXTURE_UPLOAD_MEMORY_BUDGET thing with my 770m
01:11 karolherbst: one of the slowest Kepler GDDR5 GPUs, but I guess better than nothing
01:11 karolherbst: uhm, slowest memory
01:12 karolherbst: mhh, actually there are slower ones, but those are also all mobilec chips
01:12 imirkin: DDR3 is probably slower still ;)
01:12 karolherbst: yeah
01:13 imirkin: i suspect the fact that i have PCIe 2.0 and not 3.0 also plays a factor
01:13 karolherbst: that one has around 100GB/s
01:13 karolherbst: ohh evil, a GDDR5 one with 40 GM/s
01:13 karolherbst: *GB
01:13 karolherbst: 64 bus width... calling for it
01:13 imirkin: anyways, i tested with nv34, G92 with GDDR2 (400mhz out of a total of 800mhz), and a GM206 at lowest clocks
01:13 karolherbst: mhh
01:14 imirkin: so ... not exactly the PERFECT performance test scenarios...
01:14 karolherbst: ohh I can test on my newer GPU as well
01:14 karolherbst: 970m :D
01:14 karolherbst: 120 GB/s memory
01:14 karolherbst: that's more like it
01:14 imirkin: feel free to play around with it
01:14 karolherbst: yeah
01:14 imirkin: i saw almost no difference between 4M and 64M
01:15 imirkin: and very little difference between 1M and 4M
01:15 karolherbst: what's the most relevant factor here? PCIe bandwidth?
01:15 imirkin: no clue
01:15 karolherbst: k
01:15 imirkin: no-more-oom seemed like a nice little benefit
01:15 karolherbst: I think I actually have the reclocking patches on that gm204 active, so I should probably test there
01:15 karolherbst: true
01:45 pmoreau: karolherbst: How do you specify for a Meson build of Mesa which LLVM version to use? I tried setting PATH to include my custom build of LLVM first, but it still ends up linking against my system LLVM install rather than my custom build.
01:45 karolherbst: pmoreau: either export PATH whenever you use meson/ninja
01:46 karolherbst: or write a cross file and specify llvm-config
01:47 pmoreau: I’ll try with exporting or writing a cross file, but setting PATH on the command line whenever I run a meson or ninja command was not enough.
01:48 karolherbst: pmoreau: weird, works for me
01:50 pmoreau: I’ll try again next week once I’m back home. I could try on my old laptop that I have with me, but not sure if 1) your NIR work would let me run OpenCL on NV50 and 2) compiling LLVM on this Intel Core 2 Duo is going to take years! :-D
01:50 karolherbst: should work and why compiling llvm?
01:50 karolherbst: ohh right, I use my system llvm :D
01:50 karolherbst: but yeah, I used to use my local one
01:51 pmoreau: Because you need LLVM 8 for the latest SPIRV-LLVM-Translator, which isn’t released yet.
01:52 karolherbst: and?
01:53 karolherbst: get a distribution where you can install such a llvm system wide :p (if you don't fear mesa master)
01:53 pmoreau: And Arch does not provide packages for unreleased versions usually. Maybe there is a different repo that has them?
01:53 karolherbst: probably
01:53 karolherbst: there always is
01:54 pmoreau: There is an AUR package for LLVM SVN, but AUR packages are built locally so it’s not going to help with build time.
01:55 karolherbst: :/
01:55 karolherbst: let it compile over night :D
01:55 HdkR: Does it only need LLVM or does it need LLVM and Clang?
01:55 karolherbst: clang as well of course
01:55 HdkR: ick, clang takes forever
01:56 airlied: or fix the spirv-llvm-translator to build against llvm7:-P
01:56 karolherbst: clang takes less than llvm though
01:56 karolherbst: there is a branch actually
01:56 karolherbst: never tried it
01:56 HdkR: Weird, clang takes longer than llvm for me :P
01:56 airlied: yeah but its old
01:56 karolherbst: yeah
01:56 airlied: it does build from the branch
01:56 karolherbst: but it also doesn't contain the additions we need for mesa anyway
01:57 pmoreau: There’s a MR to merge in the fixes from the master branch into the release-70 branch
01:57 karolherbst: sadly that llvm testing thing makes it impossible to have the same branch working for multiple releases
01:58 pmoreau: :-/
01:58 karolherbst: the tests are stupid anyway
01:58 karolherbst: I don't see the point in testing for order of instructions really
01:58 karolherbst: more of a hazzle than beneficial
01:59 karolherbst: of final bytecodes that is
02:00 karolherbst: or well, it would be usefull if the same ll file would compile to the same spir-v, no matter what version of llvm you have
02:44 blipk: any suggestions on getting 4K display to work on nouvue drivers? keep getting 'configure crtc 0 failed' when trying to change output to any higher res with xrandr
02:44 blipk: seems my monitor is sending bad EDID data, also only get 'default' output in xrandr rather than specifics, if that has anything to do with it
02:44 airlied: sounds like you have nomodeset and are using vesa
02:46 blipk: hmm
02:46 blipk: so I should try disabling nomodeset?
02:46 blipk: I just installed the nvidia driver package and works fine
02:47 blipk: I'd rather use free software though
02:47 airlied: well then you need to uninstall nvidia, and remove nomodeset
02:47 imirkin: or return the nvidia board and buy amd
02:47 blipk: right, it wasnt working before I installed nvidia though and I didnt turn on nomodeset
02:47 blipk: would installing them have turned that off?
02:48 airlied: yes
02:49 blipk: okay
02:49 blipk: well I broke my distro in other ways too
02:49 blipk: so I'll give that a shot when I reinstall
02:49 blipk: whats an alternative to vesa?
02:49 blipk: if its just not the nomodeset issue
02:50 airlied: it's just nomodeset
02:50 imirkin: could be a million things, would have to debug it.
02:50 imirkin: in general nouveau's pretty good at the basics of modesetting though
02:51 imirkin: [until i touched it with my grimey hands... waiting for reports of issues to flow in for 4.20...]
11:16 blipk: so I'm trying to get my GTX 1060 to work with nouveau and a 4K display over display port, seems I might need video acceleration, is that just impossible because of the signed proprietary nvidia blobs that I need?
11:29 HdkR: Video being video decoder acceleration or something else that isn't related to the 3D engine?
11:29 HdkR: Because you at least need the signed blobs for the 3D engine, which are available for pascal
14:37 joepublic: Curious, PhyX ("Including the GPU accelleration code" according to Phoronix) being released under BSD license -- is that helpful to any extent, or is it still too much of the wrong thing to help nouveau?
14:38 joepublic: if I could spell which I obviously cannot
14:40 gnarface: i wonder if it means that someone can fix physX support in Borderlands2 and the pre-seqel for Linux
14:50 loonycyborg: PhysX probably is just overkill for games, so they're opensourcing it so at least scientists could make use of it
14:54 karolherbst: joepublic: physx is cuda
14:54 karolherbst: well, makes use of cuda
14:54 karolherbst: without a cuda runtime, no physx
14:57 karolherbst: or it was at least. maybe it can use compute shaders now. dunno
14:58 joepublic: I am just hoping that Nvidia edges ever closer to providing helpful needful things.
20:49 Lyude: karolherbst: poke; do you know if the PCI driver on linux changes the power state of PCI devices before their respective drivers are actually loaded?
20:49 Lyude: like-if I had a nvidia gpu that the bios put into D3 mode then I boot up, but nouveau isn't loaded at any point, will it stay in D3?
21:11 karolherbst: Lyude: I am sure the pci subsystem would put it into D0 mode
21:12 karolherbst: but... maybe it doesn't?
21:12 karolherbst: I think it's done somewhere in enable_device
21:12 Lyude: karolherbst: alright
21:13 karolherbst: yep
21:13 karolherbst: do_pci_enable_device
21:13 karolherbst: calls pci_set_power_state(dev, PCI_D0)
21:13 Lyude: and devices are just enabled by default?
21:13 karolherbst: you have to aaik
21:13 karolherbst: *afaik
21:14 karolherbst: either by the driver or by the subsystem