00:23 fdobridge: <g​fxstrand> Does GL work?
00:30 fdobridge: <a​zkali> Yup
00:48 fdobridge: <g​fxstrand> Okay. IDK why NVK isn't working then.
00:48 fdobridge: <g​fxstrand> I could probably figure it out if I got myself some Linux on my switch.
00:49 fdobridge: <C​onan Kudo (ニール・ゴンパ)> has the GSP licensing stuff been worked out yet?
00:49 fdobridge: <g​fxstrand> I believe GSP is redistributable but I'm not sure on the details.
00:50 fdobridge: <C​onan Kudo (ニール・ゴンパ)> this is the latest I'm aware of: https://github.com/NVIDIA/open-gpu-kernel-modules/issues/62#issuecomment-1338149520
00:50 fdobridge: <g​fxstrand> If it isn't yet, that sounds like something Red Hat can sort out.
00:52 fdobridge: <g​fxstrand> Yeah, that sounds like something for Red Hat to push on.
00:54 fdobridge: <C​onan Kudo (ニール・ゴンパ)> I guess I'll ask @Mr Fall🐧 once they're online then
00:55 fdobridge: <k​arolherbst🐧🦀> ohh I think I might have forgotten to ask 🙃
00:55 fdobridge: <C​onan Kudo (ニール・ゴンパ)> uhh how are you still online?
00:55 fdobridge: <C​onan Kudo (ニール・ゴンパ)> isn't it like brain-dead o-clock in the morning there?
01:03 fdobridge: <M​ohamexiety> flexible sleep schedule 😆
02:29 fdobridge: <C​onan Kudo (ニール・ゴンパ)> can you fire off a request asap-ish then, please? 🙂
02:30 fdobridge: <k​arolherbst🐧🦀> yeah, will do
02:36 fdobridge: <g​fxstrand> @Mr Fall🐧 doesn't sleep. Or if he does, he sleeps like a computer with WoL attached to Discord.
02:36 fdobridge: <k​arolherbst🐧🦀> wait, I do sleep
02:36 fdobridge: <k​arolherbst🐧🦀> I'm a normal human being
02:37 fdobridge: <C​onan Kudo (ニール・ゴンパ)> people say the same thing about me 🤣
02:40 fdobridge: <k​arolherbst🐧🦀> anyway, i'ts _only_ 4:40 am here 😄
02:51 fdobridge: <C​onan Kudo (ニール・ゴンパ)> people freak out when I say stuff and they know it's like 2~4 am my time
02:52 fdobridge: <C​onan Kudo (ニール・ゴンパ)> but unfortunately, it's the hazard of being an American OSS developer in community open source projects
02:53 fdobridge: <C​onan Kudo (ニール・ゴンパ)> there's increasingly a dearth of folks in my timezones in community open source projects... most are in Europe, with a sprinkle of them in Japan and Australia
03:15 fdobridge: <!​[NVK Whacker] Echo (she) 🇱🇹> How can I attach WoL to Discord? 😅
10:27 fdobridge: <🌺​ ¿butterflies? 🌸> imagine a world where nvgpu has fully stable abis
11:54 RSpliet: nvuser: those are rather bland control registers from the looks of it. the coefficients are more interesting. Are those in 137004?
11:55 RSpliet: can't remember. Anyway, peeking in that region doesn't harm. Just don't poke :-P
16:25 fdobridge: <D​adSchoorse> will nak fuse ffma in the backend or will it just use the late algebraic option?
17:02 fdobridge: <g​fxstrand> Probably in NIR somehow. TBD exactly what that looks like, though. The late algebraic option is kinda meh. There's an intel-specific pass that's better. Might steal that.
17:42 nvuser: RSpliet: aha, for sure I won't try to write anything there.. well I just did ./nvapeek on each PLL listed in gf100 xml document, I'm not sure if those are where I should have looked
17:43 nvuser: Changing the nvidia control panel perf values changed something with those values, so I guess that means they do something for kepler
17:44 nvuser: the ones I posted are with max perf (1006 MHz core clock)
17:45 fdobridge: <p​homes> I reworked the conditional rendering MR again. It precalculates if condtional rendering is active in a separate mme. This makes it only use a single temporary register in draw, clear, and dispatch calls. Maybe it is good enough for pre-turing?
17:50 nvuser: RSpliet: 137004 was not listed there that I could tell, where I pasted the 0x137<etc> values are the ones I found, the bottom values are the output for each respectively
17:55 nvuser: oh hm, the xml does seem to suggest 'coef' offsets, maybe I should append those and get that output as well
18:00 nvuser: okay I'll give it another go heh
18:01 fdobridge: <B​yLaws> @phomes any reason for not using the actual conditional rendering regs?
18:06 fdobridge: <B​yLaws> I guess you need a temp buffer for inverted
18:06 fdobridge: <B​yLaws> I guess youd need a temp buffer for inverted with them (edited)
18:14 fdobridge: <p​homes> @ByLaws no reason other than me not knowing about them 🙂 I am still learning new things every time I decide spend a little time on mesa
18:14 fdobridge: <B​yLaws> Ah
18:15 fdobridge: <B​yLaws> SET_RENDER_ENABLE_B_OFFSET
18:15 fdobridge: <B​yLaws> With SET_RENDER_ENABLE_C_MODE_CONDITIONAL, if data at that address is zero or won't draw
18:15 fdobridge: <B​yLaws> Or clear
18:16 fdobridge: <B​yLaws> I.e. draw methods just won't do anything
18:16 fdobridge: <B​yLaws> Even if they are called
18:16 fdobridge: <B​yLaws> Idk if you'd wanna look into since your current approach makes inverted easier
18:17 nvuser: okay, I'll put that down to pebkac.. I think I have the right values now, will just double check before posting
18:18 fdobridge: <B​yLaws> Otherwise you'd have to use I2m with a fifo ref to upload the data + a zero constant
18:18 fdobridge: <B​yLaws> Then use SET_RENDER_ENABLE_C_MODE_RENDER_IF_EQUAL
18:19 fdobridge: <B​yLaws> Idk if that helps at all
18:19 fdobridge: <p​homes> @ByLaws thanks 🙂 I will see if I can get that to work. Maybe they can fix the remaining fails. My method would only work for things that are called in mme's
18:19 fdobridge: <B​yLaws> With SET_RENDER_ENABLE_C_MODE_CONDITIONAL, if data at that address is zero it won't draw (edited)
18:20 fdobridge: <B​yLaws> Also wanna ensure NVB197_SET_RENDER_ENABLE_OVERRIDE is 0
18:20 fdobridge: <B​yLaws> And SET_RENDER_ENABLE_CONTROL_CONDITIONAL_LOAD_CONSTANT_BUFFER_FALSE is set I imagine
18:24 nvuser: RSpliet: I guess you mean the coef offsets are appended to the PLL source, so doing that for each clock this is what I have: https://dpaste.com/CR2CJRWBP
18:25 nvuser: replacing last hex digit with 4 is how I got coef for each
18:27 nvuser: Is CLK0 the core clock aka 0f?
18:54 fdobridge: <g​fxstrand> PSA for anyone playing with NAK: You'll need a magic meson branch: https://github.com/dcbaker/meson/tree/submit/rust-cargo-subproject
18:56 fdobridge: <!​[NVK Whacker] Echo (she) 🇱🇹> I was able to compile it with regular arch-meson
18:56 fdobridge: <!​[NVK Whacker] Echo (she) 🇱🇹>
18:56 fdobridge: <!​[NVK Whacker] Echo (she) 🇱🇹> But vkcube doesn't work yet so I think NAK is not worth messing around right now
18:56 fdobridge: <g​fxstrand> Really? Interesting...
18:58 fdobridge: <!​[NVK Whacker] Echo (she) 🇱🇹> Yes
20:00 nvuser: let me know if that's everything needed
20:00 nvuser: cheers
20:47 fdobridge: <g​fxstrand> @ASDQueerFromEU Did you get any steam games running stably with GSP?
20:48 fdobridge: <!​[NVK Whacker] Echo (she) 🇱🇹> Even something as simple as SuperTuxKart (with Vulkan renderer) causes the `fail ttm_validate` error 🐸
20:53 fdobridge: <g​fxstrand> kk
20:53 fdobridge: <g​fxstrand> @Mr Fall🐧 power management should be good on a 980M, no?
21:00 fdobridge: <g​fxstrand> @ASDQueerFromEU Was there a game or two that you thought ran pretty well? Looking for something that would make a good demo.
21:08 fdobridge: <M​ohamexiety> airlied and Esdras usually used Talos iirc
21:09 fdobridge: <E​sdras Tarsis> I tested Talos without gsp
21:10 fdobridge: <E​sdras Tarsis> Does it need to be a 3D game?
21:11 fdobridge: <!​[NVK Whacker] Echo (she) 🇱🇹> I don't have Talos
21:13 fdobridge: <k​arolherbst🐧🦀> uhm... well... kinda
21:13 fdobridge: <k​arolherbst🐧🦀> if you patch the kernel that is
21:14 fdobridge: <k​arolherbst🐧🦀> could just enable it, but we don't because there is no reliable way to identify if the GPU has a proper fan or we mess up
21:14 fdobridge: <k​arolherbst🐧🦀> so it's currently not enabled
21:18 fdobridge: <g​fxstrand> oh
21:18 fdobridge: <g​fxstrand> Let's assume this is a $400 laptop that I'm not going to cry about if it burns. 🙂
21:19 fdobridge: <g​fxstrand> Also, we're looking at Alienware machines. They should have good fans.
21:20 fdobridge: <E​sdras Tarsis> Epic gave it away for free
21:20 fdobridge: <!​[NVK Whacker] Echo (she) 🇱🇹> Let's double-check if I have it
21:24 fdobridge: <k​arolherbst🐧🦀> yeah, it's not about the fans really, it's just that power management in nouveau is quite yolo
21:24 fdobridge: <k​arolherbst🐧🦀> we don't even have protection against overheating in palce
21:24 fdobridge: <k​arolherbst🐧🦀> but I do have some patches
21:24 fdobridge: <k​arolherbst🐧🦀> let me dig them up tomorrow
21:30 fdobridge: <g​fxstrand> ok
21:33 fdobridge: <g​fxstrand> I'm fine with a bit of YOLO. Like, if I just have to find a frequency that works and lock it there, that's fine.
21:50 fdobridge: <C​onan Kudo (ニール・ゴンパ)> did you do this? 🙂
21:51 fdobridge: <k​arolherbst🐧🦀> we'll talk about it tomorrow with Nvidia
21:54 fdobridge: <C​onan Kudo (ニール・ゴンパ)> Sweet!