00:02HdkR: Yay for then switching to RISC-v and making their own lives easier š
00:02HdkR: for them
00:03fdobridge_: <gāfxstrand> And it's not like NVIDIA is special here. No hardware company is going to spend millions of dollars worth of engineering and legal effort on 10-year-old hardware just because some Linux users are complaining.
00:04fdobridge_: <gāfxstrand> (Yes, I'm counting from the day they released the first GPU. At a hardware company, a product is over the moment the first unit ships.)
00:04fdobridge_: <pāavlo_it_115> I had this in mind: has anyone tried to make a nouveau-fw type from the reclocking unlock firmware (which was extracted from the NVIDIA binary)?
00:04fdobridge_: <pāavlo_it_115> https://cdn.discordapp.com/attachments/1034184951790305330/1200954539172450437/b88a28c2561d9324.png?ex=65c80f12&is=65b59a12&hm=882b8f2cb15d6a8906f339b1b892de3f026b4c36a13601eb16da0ae5f476ac3b&
00:04fdobridge_: <kāarolherbstš§š¦> though compared to the others NVidia has an impressive long support cycle for hardware
00:05fdobridge_: <kāarolherbstš§š¦> when did Kelper go EoL? last year?
00:05fdobridge_: <kāarolherbstš§š¦> that's like 10 years
00:05HdkR: Impressive amount of time for tech hardware
00:06fdobridge_: <kāarolherbstš§š¦> extracting the firmware isn't the problem
00:06fdobridge_: <rāedsheep> https://endoflife.date/nvidia-gpu "Ended 2 years and 4 months ago"
00:06fdobridge_: <kāarolherbstš§š¦> mh???
00:06fdobridge_: <pāavlo_it_115> Problem adapting nouveau to this firmware?
00:06fdobridge_: <rāedsheep> Going by previous pace Maxwell should already be dead
00:06fdobridge_: <kāarolherbstš§š¦> that's kinda bogus š
00:07fdobridge_: <rāedsheep> How so?
00:07HdkR: meanwhile AMD is cutting out Vega support from their software and they had hardware launch with it last year :(
00:07fdobridge_: <rāedsheep> Yeah...
00:07fdobridge_: <kāarolherbstš§š¦> https://www.nvidia.com/download/driverResults.aspx/215840/en-us/
00:08fdobridge_: <rāedsheep> Security support isn't over, but it's no longer considered active support
00:08fdobridge_: <kāarolherbstš§š¦> yeah well.. you get driver updates
00:08fdobridge_: <kāarolherbstš§š¦> and they do port to newer kernel versions afaik
00:09fdobridge_: <kāarolherbstš§š¦> but sure.. you won't get new features, but that's old hardware
00:10fdobridge_: <kāarolherbstš§š¦> from an end user pov only two things matter: 1. do I get updates and 2. does it run on the newest OS?
00:10fdobridge_: <pāavlo_it_115> it seems there are only security updates
00:10fdobridge_: <kāarolherbstš§š¦> so? it still runs on the newest kernel
00:10fdobridge_: <kāarolherbstš§š¦> that's a big deal
00:11fdobridge_: <rāedsheep> To be entirely clear, I do agree that having support for as long as they have provided it is commendable, and yes most new features just wouldn't work anyway.
00:11fdobridge_: <kāarolherbstš§š¦> they could have stopped at... whatever was released 2 years ago, but they didn't
00:12fdobridge_: <kāarolherbstš§š¦> if you get sec updates or perf opts for new games... meh
00:12fdobridge_: <kāarolherbstš§š¦> at this point the difference is pretty much not existing
00:13fdobridge_: <kāarolherbstš§š¦> yes
00:14fdobridge_: <aāirlied> yeah I think the thing kepler users are hitting down is wayland integration, egl etc
00:16fdobridge_: <aāirlied> yeah I think the thing kepler users are hitting now is wayland integration, egl etc (edited)
00:17fdobridge_: <kāarolherbstš§š¦> ahh yeah, fair
00:17fdobridge_: <kāarolherbstš§š¦> though not sure how usable that's on modern hw
00:17fdobridge_: <kāarolherbstš§š¦> I heard both ways
00:17fdobridge_: <gāfxstrand> They tend to care a bit more about security support because the potential legal fallout of security bugs is a lot higher.
00:19fdobridge_: <rāedsheep> For sure. I am also kind of playing devils advocate even bringing any of this up. In my day job I am annoyed if anybody is using something 6 years old, let alone 10.
00:19fdobridge_: <rāedsheep> If it were only up to me everybody would buy new hardware every 4 years, but not everybody had the money and I don't want landfills full of cards that some people might still tolerate.
00:20fdobridge_: <kāarolherbstš§š¦> thing is.. some of their GPUs are used in DCs for eternity
00:20fdobridge_: <kāarolherbstš§š¦> and for most parts, it doesn't matter if your GPU is a DC one or a desktop one. Though they also have their quadro line
00:23fdobridge_: <gāfxstrand> *Sandy Bridge has entered the chat.*
00:25fdobridge_: <rāedsheep> Just in the last week I have worked with people using Nahalem and earlyish AMD FX. Those old drivers can't render anything right anymore.
00:25fdobridge_: <kāarolherbstš§š¦> who is still using Sandy in the business world? š
00:25fdobridge_: <rāedsheep> Oh, it's a thing.
00:25fdobridge_: <kāarolherbstš§š¦> I mean.. sure
00:29fdobridge_: <gāfxstrand> IDK about the business world but I can't count the number of people who asked me for a Sandy Bridge Vulkan driver. š
00:30fdobridge_: <rāedsheep> Small businesses don't have the money to upgrade and don't have enough layers. The owner ends up directly making the decision to never upgrade, almost without fail.
00:30fdobridge_: <rāedsheep> Something has to explode before they will authorize anything.
00:32fdobridge_: <gāfxstrand> Yeah...
00:34fdobridge_: <kāarolherbstš§š¦> cursed rust stuff I'm working on: https://godbolt.org/z/Mv1droof9
00:34fdobridge_: <kāarolherbstš§š¦> yeah.. I mean.. Linux users be like that
00:34fdobridge_: <kāarolherbstš§š¦> I still fixed nv30 bugs in 2023 š
00:34fdobridge_: <gāfxstrand> Yup
00:35fdobridge_: <kāarolherbstš§š¦> or was that 2022?
00:35fdobridge_: <kāarolherbstš§š¦> I don't even remember
00:35fdobridge_: <kāarolherbstš§š¦> I need that for... things... but that cursed pattern matching is cursed
00:36fdobridge_: <kāarolherbstš§š¦> how does that even work... š
00:37fdobridge_: <gāfxstrand> Uh oh... Do I want to know?
00:37fdobridge_: <kāarolherbstš§š¦> š
00:38fdobridge_: <kāarolherbstš§š¦> it's going to be awesome, but uhhh...
00:38fdobridge_: <kāarolherbstš§š¦> a lot of research on how I want things to be
00:38fdobridge_: <kāarolherbstš§š¦> apparently `ref base` matches on the actual field name...
00:38fdobridge_: <pāavlo_it_115> I have a gt 630 kepler video card here now. And now I'm thinking about how to do automatic reclocking (just for the sake of the experiment). I'm thinking about how drivers determine when to increase the frequency. It's banal to think if something happens on the monitor. But I want to go the other way (because the option with tracking... I have no idea how it will work.) You can make a variable file type. And when starting the 3D program
00:39fdobridge_: <kāarolherbstš§š¦> arbitrary names: https://godbolt.org/z/eb77exqof
00:39fdobridge_: <kāarolherbstš§š¦> @gfxstrand I was wondering if you need something like that as well š Or well.. needed š
00:41fdobridge_: <gāfxstrand> Yeah, those matches are a little crazy
00:41fdobridge_: <gāfxstrand> I've seen far worse
00:41fdobridge_: <kāarolherbstš§š¦> heh
00:43fdobridge_: <kāarolherbstš§š¦> anyway.. I kinda want to rework my `cl_mem` thing in rusticl, and I need to figure out how to do inheritance in a sane way I won't have to rewrite everything else as this also has to match with that pointer -> rust type thing going on... and pulling the base struct like that actually helps for stuff
00:43fdobridge_: <gāfxstrand> Yeah...
00:45fdobridge_: <kāarolherbstš§š¦> and it's also going to be useful for CL pipes, as those are also just `cl_mem` objects :ferrisUpsideDown: honestly.. who designed this API
00:45RSpliet: Lee Howes!
00:45RSpliet: Well, among many others
00:50fdobridge_: <gāfxstrand> Rust-based GL state tracker when? š
00:51fdobridge_: <kāarolherbstš§š¦> pain
01:03fdobridge_: <gāfxstrand> I'm kidding, of course. No one should ever reimplement GL again.
01:03fdobridge_: <kāarolherbstš§š¦> tell that to google
01:04fdobridge_: <gāfxstrand> I once had notions of cleaning up the Mesa state tracker but no... Best leave a working monstrosity alone... š¬
01:04fdobridge_: <gāfxstrand> Google just likes reinventing things. At least Microsoft knew better.
01:17fdobridge_: <gāfxstrand> Also, Google has only ever reimplemented GLES, to my knowledge. Even ES 3.1 + AEP is massively easier to get right than GL 4.6 compat profile
01:18HdkR: On Android you can now just ship Zink and get GL :P
01:21HdkR: Android even has the new-ish EGL extension to get frontbuffer rendering, which was a major concern five years ago or whatever
01:51HdkR: Oh wait, I wonder if they have an equivalent Vulkan extension for that EGL extension. I was reaching that there would be one
02:01fdobridge_: <gāfxstrand> There is a Vulkan extension for front buffer rendering. It was driven by a guy at Samsung so I assume it's shipped on some mobile somewhere.
02:12HdkR: Nice. So at least some device has it, even if it is only like Exynos+RDNA which is super small percentage
02:15HdkR: Is that `SHARED_CONTINUOUS_REFRESH_KHR`?
02:15HdkR: Or something more magical?
02:38HdkR: `This allows the application to use the image while the presention engine is accessing it, in order to reduce the latency between rendering and presentation.` Sounds like it
04:16TimurTabi: karolherbst: yes, I have an idea on how to clean it up.
05:04fdobridge_: <gāfxstrand> Shared presentable image, I think. IDK it's on AMD or some Qualcomm device. It's been a while so it was probably happening in tandem with or before the AMD negotiations.
05:46fdobridge_: <māhenning> @gfxstrand Oh, I just realized that your mmu fault could be related to a NAK bug I started fixing in September and then promptly forgot about. Maybe try this patch and see if it changes anything: https://gitlab.freedesktop.org/mhenning/mesa/-/commit/ee983530d51f2d4c383e10cb16224c4d2b1a6da9
05:46fdobridge_: <māhenning> that's obviously not a complete fix and it isn't well-tested, but I could write up the rest of it pretty quickly if it helps
05:47fdobridge_: <māhenning> (or if the shader in question doesn't use ldg, stg, or atomg, then it's something different)
05:55fdobridge_: <gāfxstrand> Yeah, the shader doesn't touch memory. In fact, it crashes mid-MME, so... Nothing to do with shaders. š
05:55fdobridge_: <gāfxstrand> Actually... I should look closer at `NVK_DEBUG=vm`... š¤
05:56fdobridge_: <gāfxstrand> It could be that the CTS is deleting the command buffer before it's done.
05:56fdobridge_: <gāfxstrand> It wouldn't have been noticed on most drivers because they reference count somewhere. We just tear down the VA on the spot.
07:08fdobridge_: <!āDodoNVK (she) š±š¹> Is it technically possible (just like Fermi or TeraScale Vulkan)? :cursedgears:
08:40fdobridge_: <aāirlied> @gfxstrand that is possible problem, some CTS object being cleaned before idle, would match the pain
08:43fdobridge_: <aāirlied> also https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/27255
08:51fdobridge_: <tāom3026> yay i feel useful beyond my endless nonsense in here, i actually helped find something!
13:20fdobridge_: <!āDodoNVK (she) š±š¹> https://mstdn.games/@ReverseModule/111833704210505779
14:12fdobridge_: <gāfxstrand> No. No compute shaders.
14:52fdobridge_: <gāfxstrand> I thought about that but then I remembered I'm running with `NVK_DEBUG=push_sync` so it can't be unbinding while it's running but then I remembered that threaded submit exists but then I remembered that I'm able to successfully dump so my CPU maps must still exist and those get torn down before the GPU unmap. I also just ran w/ valgrind and didn't get any use-after-free so all my CPU data structures still exist all the way through.
14:52fdobridge_: <gāfxstrand> At this point, I really think we're looking at either a GSP bug or a VM_BIND bug.
14:58fdobridge_: <!āDodoNVK (she) š±š¹> GSP firmware or GSP implementation? :nouveau:
15:00fdobridge_: <kāarolherbstš§š¦> could be both
15:38fdobridge_: <gāfxstrand> https://tenor.com/view/why-not-both-why-not-take-both-gif-11478682
15:43fdobridge_: <gāfxstrand> That's possible but we've already had bugs with not properly sizing some fixed size array somewhere. This feels a lot like that.
15:52fdobridge_: <kāarolherbstš§š¦> ~~just wait until we have our new driver then all those bugs will be fixed~~
15:52fdobridge_: <kāarolherbstš§š¦> which in all seriousness it probably indeed the quickest way to get rid of those bugs :ferrisUpsideDown:
15:57fdobridge_: <Sāid> ~~nova?~~
15:59fdobridge_: <!āDodoNVK (she) š±š¹> I doubt that will be upstreamed though
16:07fdobridge_: <sāamantas5855> I think tesla 1 and 2 has those
16:07fdobridge_: <sāamantas5855> What if, TesVK
16:09fdobridge_: <gāfxstrand> If the bug is that we did a calculation wrong, a new driver won't fix that if it uses the same calculation.
16:09fdobridge_: <!āDodoNVK (she) š±š¹> I might have to get a 9500 GT again if that becomes a thing šŗšŗšŗ
16:11fdobridge_: <sāamantas5855> I have a tesla
16:11fdobridge_: <sāamantas5855> I think 9600
16:12fdobridge_: <!āDodoNVK (she) š±š¹> Is it a Model GT or Model GTX? š
16:15fdobridge_: <sāamantas5855> https://cdn.discordapp.com/attachments/1034184951790305330/1201198773750218802/signal-2023-01-04-15-24-10-756.jpg?ex=65c8f288&is=65b67d88&hm=750e7377278baee668f940236142c07a5f0cd815559d929591ebac74fd27188c&
16:15fdobridge_: <sāamantas5855> here
16:15fdobridge_: <sāamantas5855> tesla support would unironically be useful since gt210 and gt220 gpus are still sold as display adapters
16:15fdobridge_: <sāamantas5855> at least here
16:18fdobridge_: <!āDodoNVK (she) š±š¹> This doesn't seem to be a 9600 GT according to images on Google
16:19fdobridge_: <gāfxstrand> Tesla probably could do very basic Vulkan but that's going to be a different driver.
16:19fdobridge_: <gāfxstrand> Yeah, Tesla is the new GeForce 2.
16:20fdobridge_: <gāfxstrand> Well, it would require a very creative kernel driver....
16:20fdobridge_: <sāamantas5855> if it is enough to run a compositor it would be useful
16:21fdobridge_: <sāamantas5855> what about fermi?
16:21fdobridge_: <sāamantas5855> https://cdn.discordapp.com/attachments/1034184951790305330/1201200316155187390/Screenshot_20240128-181835.png?ex=65c8f3f8&is=65b67ef8&hm=25421c048cb1aece87f6d0ae3273e99773d16092f288dadafb96153b4ba6a221&
16:21fdobridge_: <sāamantas5855> if I go to gpus and sort by price tesla comes first
16:21fdobridge_: <gāfxstrand> Fermi is possible. It's all about how much work you want to do.
16:21fdobridge_: <sāamantas5855> is fermi that much different from kepler?
16:22fdobridge_: <gāfxstrand> Like, as long as you have compute, vertex, and geometry shaders, it's probably possible.
16:22fdobridge_: <gāfxstrand> It's additional compiler work and the copy engine is different.
16:23fdobridge_: <gāfxstrand> Worst case, we have to implement copies with shaders like other hardware.
16:23fdobridge_: <!āDodoNVK (she) š±š¹> Fermi is probably more useful as a radiator thanks to its high power consumption /s
16:23fdobridge_: <gāfxstrand> Note that I didn't say "I". š
16:24fdobridge_: <sāamantas5855> https://tenor.com/view/thermi-nvidia-gpu-580-480-gif-19880982
16:25fdobridge_: <kāarolherbstš§š¦> Ben said that we _could_ implement the copy engine via firmware (which we'd have to write ourselves)....
16:25fdobridge_: <kāarolherbstš§š¦> or something
16:25fdobridge_: <māhenning> The gl driver should already be enough to run a compositor on tesla, but it doesn't get much attention any more and might be a little buggy
16:25fdobridge_: <kāarolherbstš§š¦> but apparently nvidia has firmware for that as well...
16:25fdobridge_: <kāarolherbstš§š¦> we could ask for that (if we don't forget)
16:25fdobridge_: <kāarolherbstš§š¦> but...
16:25fdobridge_: <kāarolherbstš§š¦> but...
16:25fdobridge_: <sāamantas5855> but they won't give it?
16:25fdobridge_: <kāarolherbstš§š¦> nvk on fermi is pointless
16:26fdobridge_: <kāarolherbstš§š¦> nvidia decide that fermi is the gen where they want to boot with the lowest clocks possible, and without reclocking you have GPUs being slower than CPU rendering
16:26fdobridge_: <kāarolherbstš§š¦> literally
16:27fdobridge_: <sāamantas5855> I thought we could reclock fermi
16:27fdobridge_: <kāarolherbstš§š¦> well.. maybe not if you use a CPU from the same time :ferrisUpsideDown:
16:27fdobridge_: <kāarolherbstš§š¦> nah..
16:27fdobridge_: <kāarolherbstš§š¦> there are patches, but it's never been really tested
16:27fdobridge_: <kāarolherbstš§š¦> and probably quite a bit of work
16:27fdobridge_: <sāamantas5855> so kepler and maxwell can be reclocked but fermi not?
16:28fdobridge_: <kāarolherbstš§š¦> correct
16:28fdobridge_: <sāamantas5855> that includes maxwell 1 and 2 right?
16:29fdobridge_: <kāarolherbstš§š¦> yes and no
16:29fdobridge_: <kāarolherbstš§š¦> maxwell 2 is uhm.. needs signed firmware for fan controls
16:29fdobridge_: <!āDodoNVK (she) š±š¹> At least work is being freed up from the CPU
16:29fdobridge_: <kāarolherbstš§š¦> but if the GPU fans are controlled by something else, it's fine
16:29fdobridge_: <kāarolherbstš§š¦> we don't support it out of the box tho
16:30fdobridge_: <sāamantas5855> Any progress on pascal and volta?
16:30fdobridge_: <kāarolherbstš§š¦> yeah.. for rendering a desktop it's good enough, well.. if you use a FHD display
16:30fdobridge_: <kāarolherbstš§š¦> will never happen
16:30fdobridge_: <kāarolherbstš§š¦> Kepler and maxwell1 will probably be the only gens we will reclock before GSP (Turing+)
16:31fdobridge_: <kāarolherbstš§š¦> NVK does kinda support it, but it's just equally pointless
16:31fdobridge_: <!āDodoNVK (she) š±š¹> Can't Tesla be reclocked too?
16:31fdobridge_: <kāarolherbstš§š¦> yeah.. to some degree
16:31fdobridge_: <kāarolherbstš§š¦> some can, some can't
16:31fdobridge_: <kāarolherbstš§š¦> it's just that if you support Kepler and Turing+, supporting all the boards in the middle is more or less for free š
16:32fdobridge_: <kāarolherbstš§š¦> took me a day to fix all volta issues and never has ran nvk on volta before :ferrisUpsideDown:
16:32fdobridge_: <kāarolherbstš§š¦> took me a day to fix all volta issues and nobody has ran nvk on volta before :ferrisUpsideDown: (edited)
16:32fdobridge_: <māarysaka> I have two of those cuties around still
16:32fdobridge_: <māarysaka> https://cdn.discordapp.com/attachments/1034184951790305330/1201203165517860904/IMG_2111.jpg?ex=65c8f69f&is=65b6819f&hm=e2b24a0bd50472991ecccacdf85694fd408c1da5b0792ae9b0a0e321ad63e030&
16:32fdobridge_: <kāarolherbstš§š¦> I have the same
16:32fdobridge_: <!āDodoNVK (she) š±š¹> Volta is much more niche than Pascal
16:33fdobridge_: <kāarolherbstš§š¦> sure, but also cursed
16:33fdobridge_: <kāarolherbstš§š¦> š
16:33fdobridge_: <kāarolherbstš§š¦> it's like pascal with a new ISA
16:34fdobridge_: <sāamantas5855> 9500gt?
16:35fdobridge_: <māarysaka> 6600LE
16:35fdobridge_: <sāamantas5855> also damn is that svideo
16:35fdobridge_: <māarysaka> I still have the motherboard and SLI module too
16:36fdobridge_: <sāamantas5855> I have never used sli
16:36fdobridge_: <sāamantas5855> or crossfire
16:36fdobridge_: <sāamantas5855> but I remember using these cards on agp 8x
16:44fdobridge_: <māarysaka> one of mine is not working anymore but I have a feeling that changing one of those capacitors could help, might be dry considering how old this thing is
16:45fdobridge_: <māarysaka> (and looking at the image again, one on that card might be close to dead too)
16:48fdobridge_: <kāarolherbstš§š¦> yeah.. same
16:48fdobridge_: <kāarolherbstš§š¦> though my Teslas are mostly fine
16:49fdobridge_: <kāarolherbstš§š¦> what worries me more are the pre Tesla ones š
16:49fdobridge_: <kāarolherbstš§š¦> I think half of them are dead
16:49fdobridge_: <kāarolherbstš§š¦> maybe it's nouveau being broken
16:49fdobridge_: <kāarolherbstš§š¦> but....
16:49fdobridge_: <kāarolherbstš§š¦> it works flawlessly one some and random rendering corruption and other funky things on others, where they hit the same path in the driver
16:49fdobridge_: <kāarolherbstš§š¦> it's weird š
16:49fdobridge_: <māarysaka> I think I archived some drivers for Windows for those on the webarchive :P
16:50fdobridge_: <māarysaka> (I should really revive my archiver again)
16:50fdobridge_: <kāarolherbstš§š¦> uhhh
16:50fdobridge_: <kāarolherbstš§š¦> what would I need? Windows 2000?
16:50fdobridge_: <kāarolherbstš§š¦> though I guess XP might work š
16:50fdobridge_: <māarysaka> https://archive.org/details/nvidia-drivers
16:51fdobridge_: <māarysaka> Should have everything down to 18x I think
16:51fdobridge_: <māarysaka> (and until 2021 as I haven't started the scanner again)
16:51fdobridge_: <sāamantas5855> I have some REALLY old nvidia drivers on some CDs
16:51fdobridge_: <sāamantas5855> they are for windows tho
16:51fdobridge_: <kāarolherbstš§š¦> `Linux x64 (AMD64/EM64T) Display Driver NVIDIA Certified 304.137` on linux š
16:51fdobridge_: <kāarolherbstš§š¦> that's not _that_ old
16:51fdobridge_: <māarysaka> that's "recent"
16:52fdobridge_: <kāarolherbstš§š¦> though mhh
16:52fdobridge_: <kāarolherbstš§š¦> maybe that's a tesla one?
16:52fdobridge_: <kāarolherbstš§š¦> ahh no
16:52fdobridge_: <kāarolherbstš§š¦> FX is even older
16:52fdobridge_: <kāarolherbstš§š¦> that's nv30, right?
16:52fdobridge_: <kāarolherbstš§š¦> 5 FX I mean
16:52fdobridge_: <māarysaka> I *think*
16:53fdobridge_: <kāarolherbstš§š¦> "177.67" on "2008.8.19 " š
16:53fdobridge_: <māarysaka> the oldest driver I have around https://archive.org/details/174-16-geforce-winxp-32bit-english-whql
16:53fdobridge_: <kāarolherbstš§š¦> I don't even want to know...
16:53fdobridge_: <kāarolherbstš§š¦> but yeah..
16:53fdobridge_: <māarysaka> pre 2008 doesn't follow the proper CDN format that NVIDIA uses
16:53fdobridge_: <sāamantas5855> I have older ones
16:53fdobridge_: <kāarolherbstš§š¦> latest driver for nv40 is "304.137 September 19, 2017", which is impressive
16:53fdobridge_: <sāamantas5855> windows 98, me and stuff
16:53fdobridge_: <kāarolherbstš§š¦> that's on linux
16:54fdobridge_: <māarysaka> nice
16:54fdobridge_: <kāarolherbstš§š¦> 4.x something I guess?
16:54fdobridge_: <kāarolherbstš§š¦> yeah.. 4.13 or so
16:54fdobridge_: <kāarolherbstš§š¦> maybe 4.12
16:54fdobridge_: <sāamantas5855> I should archive them
16:54fdobridge_: <kāarolherbstš§š¦> depending on luck
16:55fdobridge_: <kāarolherbstš§š¦> though I guess 4.4 would be the better choice here because longterm
16:55fdobridge_: <kāarolherbstš§š¦> wait...
16:55fdobridge_: <kāarolherbstš§š¦> that's still supported :ferrisUpsideDown:
16:55fdobridge_: <kāarolherbstš§š¦> ehh wait
16:55fdobridge_: <kāarolherbstš§š¦> that got dropped
16:56fdobridge_: <!āDodoNVK (she) š±š¹> I have at least 2 phones with older stock kernel versions than that
16:56fdobridge_: <kāarolherbstš§š¦> last 4.4 release is in 2022...
16:57fdobridge_: <māarysaka> If I ever find the energy to rewrite my scanner, I will try to catch those too I guess
16:57fdobridge_: <sāamantas5855> https://cdn.discordapp.com/attachments/1034184951790305330/1201209375956680785/signal-2021-11-24-02-30-12-718.png?ex=65c8fc68&is=65b68768&hm=79dfc94ab47b1e235a67ec2c36ccf262ef7f5abb1159c936a7785afac7db726d&
16:57fdobridge_: <sāamantas5855> https://cdn.discordapp.com/attachments/1034184951790305330/1201209430000279633/signal-2021-11-24-00-33-01-360.png?ex=65c8fc75&is=65b68775&hm=50608993c53d45da08bf35465806d34a4e3c205d23824286934a3219b78aaeb9&
16:57fdobridge_: <sāamantas5855> matrox man
16:57fdobridge_: <sāamantas5855> https://cdn.discordapp.com/attachments/1034184951790305330/1201209549848326144/signal-2021-11-24-20-10-53-847.png?ex=65c8fc92&is=65b68792&hm=54efec39f5d58043f249c91fc6623b1495e052c25efb6a018949d70cadf5a5f1&
16:58fdobridge_: <!āDodoNVK (she) š±š¹> I wonder if it will end up like Bibliogram
16:58fdobridge_: <kāarolherbstš§š¦> the time where all GUI was horrible
16:58fdobridge_: <sāamantas5855> yeah
16:58fdobridge_: <kāarolherbstš§š¦> without exceptions. I don't take any questions
16:59fdobridge_: <kāarolherbstš§š¦> I mean.. XP was also horrible, but in a way the 2000s deserved
17:00fdobridge_: <sāamantas5855> xp was color pukw
17:00fdobridge_: <sāamantas5855> xp was color puke (edited)
17:00fdobridge_: <māarysaka> I just hope I will not crash the archiver pool again... Last time I may have uploaded 2k installers and that was representing 50% of the driver collection on the archive.org, so I understand that it might be a bit too much for tapes to catch up š
17:00fdobridge_: <sāamantas5855> have you donated to archive too.
17:00fdobridge_: <!āDodoNVK (she) š±š¹> And also nostalgic
17:00fdobridge_: <sāamantas5855> have you donated to archive too? (edited)
17:00fdobridge_: <kāarolherbstš§š¦> I mean.. the style was unique in its own
17:00fdobridge_: <kāarolherbstš§š¦> so it's fine
17:00fdobridge_: <kāarolherbstš§š¦> but GUI from the 90s is all terrible
17:00fdobridge_: <kāarolherbstš§š¦> š
17:01fdobridge_: <sāamantas5855> I never used windows so I'm thankful I'm not nostalgic for xp
17:01fdobridge_: <sāamantas5855> we did have 95 at school tho
17:01fdobridge_: <kāarolherbstš§š¦> I think what I'm trying to say is, I understand people liking the XP style
17:01fdobridge_: <sāamantas5855> I'm nostalgic for ubuntu with gnome 2
17:01fdobridge_: <kāarolherbstš§š¦> mhhh
17:02fdobridge_: <kāarolherbstš§š¦> I mean.. unity the X11 compositor was kinda cool
17:02KitsuWhooa: I still use unity :p
17:02KitsuWhooa: (also technically compiz is the compositor and unity is a plugin for it)
17:04fdobridge_: <sāamantas5855> not really
17:04fdobridge_: <sāamantas5855> everyone hated it back then
17:04fdobridge_: <sāamantas5855> also perf was bad
17:04fdobridge_: <sāamantas5855> it even had amazon ads
17:05fdobridge_: <sāamantas5855> When unity came out I think I had a pentium 2 or 3
17:12waterisnice: so the main routine 524 624 724=1872 624+624+512=1760 so is 724+724+512=1960 and 12+12=24 that 288+88+24 so what another buffer expects is 224+88+488 in other words 724+724-1024=424, so if you generate 288 instead of 224 you get 512 to eliminate, if 224 its sums as 448, so three times that value 672-448 comes back as 224, 224+500 is 724, for an example 724 400 300 would use the same logic, after eliminating the stuff 400 and 300 that combined
17:12waterisnice: to 1024, 724+724=1448/2 is 724. Such routines are for access and are generated by compiler, queue jumping is with indexes which i talk about next time after evading the ban again.
17:21fdobridge_: <gāfxstrand> Ah... Compiz...
17:30fdobridge_: <!āDodoNVK (she) š±š¹> That Estonian is at least nice in this message (I wonder when that person started doing weird stuff)
17:38fdobridge_: <pāavlo_it_115> Hope dies last)
17:45fdobridge_: <pāavlo_it_115> Wtf
17:45fdobridge_: <pāavlo_it_115> https://cdn.discordapp.com/attachments/1034184951790305330/1201221635181641858/IMG_20240128_194408.jpg?ex=65c907d3&is=65b692d3&hm=a33525c93b78048f1d369cf182a7236ba60060c489603e1f7bf868e964bcb5a5&
17:47fdobridge_: <pāavlo_it_115> Wtf) (edited)
17:47fdobridge_: <pāavlo_it_115> https://cdn.discordapp.com/attachments/1034184951790305330/1201221635181641858/IMG_20240128_194408.jpg?ex=65c907d3&is=65b692d3&hm=a33525c93b78048f1d369cf182a7236ba60060c489603e1f7bf868e964bcb5a5&
17:47fdobridge_: <pāavlo_it_115> Maybe xf86-video-modesettings crash this
17:49soreau: wayfire is probably the best bet for 'compiz on wayland'
17:50fdobridge_: <pāavlo_it_115> Maybe xf86-video-modesettings crash this
17:50fdobridge_: <pāavlo_it_115> Yes! (edited)
17:50fdobridge_: <kāarolherbstš§š¦> somebody needs to do this, and it won't be any of the current nouveau developers š
17:51fdobridge_: <pāavlo_it_115> Were there any attempts to bypass the firmware?
17:52fdobridge_: <pāavlo_it_115> I don't remember very well, but someone here did that. and added "nvidiaboost = 0-2" or something like that to the kernel parameters
17:57fdobridge_: <kāarolherbstš§š¦> that has nothing to do with the firmware
17:57fdobridge_: <kāarolherbstš§š¦> that's just how power management works
17:58fdobridge_: <kāarolherbstš§š¦> 0 (the default) just means safe clocks you probably will never hit thermal limits with
17:58fdobridge_: <kāarolherbstš§š¦> 1 is generally something GPUs are marketed with as their block and 2 is "until thermal limits"
18:28fdobridge_: <tāom3026> If thats the same person as "marrt" or similiar nick it was going on in 2015 or so
18:31fdobridge_: <gāfxstrand> Yeah, he's been around a long time
18:32fdobridge_: <pāavlo_it_115> Do you have a list of VK_EXT/VK_KHR extensions that are in the works/already done for specific chip generations?
18:35fdobridge_: <pāavlo_it_115> Do you have a list of VK_EXT/VK_KHR extensions that are in the works/already done for specific chip generations? If there are, look for them on gitlab? (edited)
18:41fdobridge_: <gāfxstrand> nvk_physical_device.c is pretty easy to read. Otherwise, there's the GitLab issue tracker.
18:47fdobridge_: <gāfxstrand> And of course features.txt and mesamatrix if you're looking for a quick 1000 mile view.
18:48liquidben: so first let's elaborate how those numbers are generated, easy -- 224 comes as 624-400 424 724-300 24 as 524-500 288 comes as 512-224 etc, so all is self contained, but the alpha state compiler is still months away since i rarely work on it, and having some other issues still, but i do work on it whenever i get time, hope to make a prototype this year. The talks about order vs queue jumping conditionals etc and index generation are
18:48liquidben: the next things up tomorrow, still testing. but same note goes for avoiding the division on 724 line, and same logic means 724-512 is 212 times to from times three gets single without division, i had success on queue jumping too but i test a bit more how to describe this on another day, cheers. that too and off to bed now.
18:52fdobridge_: <gāfxstrand> In terms of plans, the closest thing we have is the issue tracker. I think we've got most of the NVK and Zink features ticked off but there's still some test failures with Zink that need to be triaged. Then VKD3D is next.
18:58fdobridge_: <pāavlo_it_115> thx
19:02fdobridge_: <pāhomes_> wrt plans I am thinking about trying VK_KHR_performance_query unless that conflicts with anyone else's plans?
19:23fdobridge_: <aāirlied> Zcull is out there if someone wants to make a real difference š
21:01fdobridge_: <kāarolherbstš§š¦> I wonder if we want allow yuzu to send NVK the plain GPU binaries š though that's going to be pain in regards to binding all the resources properly
21:23liquidben: Short rant! People tend to get insulted that some courageous bloke comes to talk who never lift any technological weights , so i comment i actually did not intend to fire anyone who is not assaulting me, and is the system bad, absolutely no, it's incredibly correct, cause essentially the same back frame or so to speak exactly the all invented methods work in flow but they tick inside compressed, so i can not be able to fire anyone either even
21:23liquidben: if i wanted to, cause essentially i am using their code, but i stash it with a trick, i mean really the science has been amazing in computer tech. The compiler and hw is all very correct. So linux is a wonder too or unixes, so that is the short thing, IT's exactly the same code that runs in just compressed format. I AM nobody to tell you what to do, by law i have human and etiquette honoring rights, for an example having gotten assaulted behind
21:23liquidben: the back and all the abuse is not something i can like, but systems is another thing, this i approve and state is very correctly done, just modern ways none wants to share as of yet. So this was short rant, back to sleep. world is perhaps at war but this is not a war to me, i do not disrespect the computer science at all, i approve all of that being above my capabilities that has been achieved. back to sleep.
22:26fdobridge_: <gāfxstrand> You're welcome to try! That would actually help a lot of we can get some perf counters.
22:29fdobridge_: <gāfxstrand> Meh. Most desktop GPUs are going to be so much more powerful than a switch that a little more ALU won't matter much. I have thought about it for the sake of swapping out NAK shaders for blob shaders
22:29fdobridge_: <gāfxstrand> Yeah, I should give that a crack once I've got 1.3 conformance
22:30fdobridge_: <kāarolherbstš§š¦> yeah.. not saying it's useful š
22:30fdobridge_: <kāarolherbstš§š¦> but it can reduce the CPU overhead a lot
22:30fdobridge_: <kāarolherbstš§š¦> and might allow to run it on tegra based devices...
22:31fdobridge_: <kāarolherbstš§š¦> nintendo is goint to ~~kill~~ sue us
22:37fdobridge_: <gāfxstrand> Yeah. I'm curious to know how close we can get to the original shaders with just NIR optimizing things. Same thought with consuming PTX.
22:38fdobridge_: <kāarolherbstš§š¦> I'm more concerned about CPU overhead here
22:39fdobridge_: <kāarolherbstš§š¦> like... the dolphin folks were able to do a massive workaround to get rid of mid-game stuttering, because they had a full hardware emulation shader they could run in place.. the yuzu folks probably don't have that luxury ever...
22:40fdobridge_: <kāarolherbstš§š¦> so instead of parsing/translating/compiling the entire shader, you just run it
22:44fdobridge_: <gāfxstrand> Yeah...
22:45fdobridge_: <aāirlied> Won't they need a recompile on most NVIDIA hw though?
22:45fdobridge_: <kāarolherbstš§š¦> sure
22:45fdobridge_: <kāarolherbstš§š¦> but not on one generation š
22:45fdobridge_: <kāarolherbstš§š¦> I'm not saying it's very useful
22:45fdobridge_: <kāarolherbstš§š¦> however, it has its niche use case š
22:45fdobridge_: <gāfxstrand> If we could disassemble to NAK and reassemble, it might be possible...
22:46fdobridge_: <aāirlied> Like emulating a switch on a switch is hardly interesting, well maybe at little,
22:46fdobridge_: <kāarolherbstš§š¦> could buy a jetson
22:46fdobridge_: <aāirlied> And if it targets Maxwell our reclocking sucks
22:46fdobridge_: <kāarolherbstš§š¦> ~~people could sell knock-off switches for half the price and get sued into oblivion~~
22:46fdobridge_: <kāarolherbstš§š¦> ahh, so here is the thing
22:47fdobridge_: <kāarolherbstš§š¦> on tegra that kinda works
22:47fdobridge_: <kāarolherbstš§š¦> well..
22:47fdobridge_: <kāarolherbstš§š¦> not dynamically
22:47fdobridge_: <kāarolherbstš§š¦> however
22:47fdobridge_: <kāarolherbstš§š¦> one could fix the stuff
22:47fdobridge_: <aāirlied> Will anyone though? It's been fixable for years
22:47fdobridge_: <kāarolherbstš§š¦> in my fantasies somebody will
22:47linkmauve: I haven’t followed much, what is the state of NVK on the current 6.7 on the Switch?
22:48fdobridge_: <aāirlied> When they fix it the can write the Vulkan extensik
22:48linkmauve: Running Switch games without rebooting and without wasting CPU cycles to disassemble/recompile would be useful I think.
22:48fdobridge_: <aāirlied> When they fix it the can write the Vulkan extension (edited)
22:48fdobridge_: <kāarolherbstš§š¦> look.. if somebody wants to sell knock-off switches, they might even bother to fix it š
22:49fdobridge_: <dāadschoorse> why would you use maxwell tegra for that though
22:49fdobridge_: <aāirlied> I don't think there is much value in us fixing it without a clear use case, just because we could, writing Vulkan exts it pretty easy
22:53fdobridge_: <gāfxstrand> As an interesting compiler project, I think you can translate most of Maxwell to Turing without redoing RA. There's a couple instructions that might be tricky.
22:53fdobridge_: <kāarolherbstš§š¦> mhhhh
22:53fdobridge_: <kāarolherbstš§š¦> bitfield stuff is different
22:53fdobridge_: <kāarolherbstš§š¦> like that insert/extract stuff
22:53fdobridge_: <gāfxstrand> Yeah. That's the tricky bit
22:54fdobridge_: <kāarolherbstš§š¦> and that texgrad lowering is going to be a pain...
22:54fdobridge_: <kāarolherbstš§š¦> and apparently some games hit that
22:54fdobridge_: <gāfxstrand> Rearranging texture sources would be tricky
22:55fdobridge_: <kāarolherbstš§š¦> nah, that's the trivial part :ferrisUpsideDown:
22:55fdobridge_: <gāfxstrand> Well yeah.
22:55fdobridge_: <kāarolherbstš§š¦> texgrad lowering is that funky "put shader into quad more and to quad ops" thingy
22:55fdobridge_: <kāarolherbstš§š¦> *mode
22:55fdobridge_: <kāarolherbstš§š¦> and it uses the barrier stuff on volta+
22:56fdobridge_: <kāarolherbstš§š¦> mhh.. though might be possible
22:56fdobridge_: <kāarolherbstš§š¦> dunno
22:56fdobridge_: <pāac85> This recompiling stuff is something Nintendo themselves will have to do to run switch games on switch2 isn't it?
22:57fdobridge_: <gāfxstrand> Yes if they let you plug in the same cartridge.
22:57fdobridge_: <kāarolherbstš§š¦> they could also just ship two SoCs š
22:57fdobridge_: <gāfxstrand> But if they are in the loop, they can recompile or transpile offline.
22:57fdobridge_: <kāarolherbstš§š¦> yeah...
22:57fdobridge_: <gāfxstrand> That's been the way in the past...
22:57fdobridge_: <kāarolherbstš§š¦> I know
22:58fdobridge_: <kāarolherbstš§š¦> but the switch actually has working game updates in place
22:58fdobridge_: <kāarolherbstš§š¦> sooo....
22:58fdobridge_: <kāarolherbstš§š¦> but I guess nvidia could also provide a transpiler.. I at least don't think games actually contain command buffers and it all goes through some API, but not quite sure
22:58fdobridge_: <gāfxstrand> Yup
22:59fdobridge_: <kāarolherbstš§š¦> not sure if yuzu needs to emulate that low-level...
22:59fdobridge_: <kāarolherbstš§š¦> but I think they have to...
23:00fdobridge_: <pāac85> Mmm, but it's one thing to put a tiny gameboy arm CPU in the Ds, it's another thing to have to tegras
23:00fdobridge_: <pāac85> Mmm, but it's one thing to put a tiny gameboy arm CPU in the Ds, it's another thing to have two tegras (edited)
23:00fdobridge_: <kāarolherbstš§š¦> you can turn the other off
23:00fdobridge_: <kāarolherbstš§š¦> it's not really a space issue, just a cooling one
23:00fdobridge_: <pāac85> The user space driver is statically linked with the game I think
23:00fdobridge_: <pāac85> I'd guess yuzu does interpret command buffers
23:00fdobridge_: <kāarolherbstš§š¦> yeah...
23:00fdobridge_: <kāarolherbstš§š¦> probably
23:01fdobridge_: <pāac85> That's what all emulators do, emulating at an api level has never been a thing AFAIK
23:01fdobridge_: <kāarolherbstš§š¦> might make more sense to just target the UAPI directly š
23:01fdobridge_: <pāac85> Yeah exactly
23:02fdobridge_: <pāac85> You emulate at the kernel level and just run all of the userspace stuff
23:24fdobridge_: <gāfxstrand> Not really. If you take the 20 nm X1 design and fab it at 5nm, it'll sip power.
23:25fdobridge_: <kāarolherbstš§š¦> tho they already shrinked it to 16nm
23:25fdobridge_: <gāfxstrand> And you can just power it off when playing switch 2 games and run the bigger one low power for switch 1 games.
23:25fdobridge_: <pāac85> I suppose you really only need the gpu
23:26fdobridge_: <pāac85> I don't think games have low level access to anything else
23:26fdobridge_: <gāfxstrand> Or they could potentially make a GPU that supports both instruction sets.
23:26fdobridge_: <kāarolherbstš§š¦> cursed
23:26fdobridge_: <kāarolherbstš§š¦> but...
23:26fdobridge_: <kāarolherbstš§š¦> the sub-channels have the API for that :ferrisUpsideDown:
23:26fdobridge_: <pāac85> Yeah cursed
23:27fdobridge_: <kāarolherbstš§š¦> I wonder if they added it because of stuff like that
23:27fdobridge_: <kāarolherbstš§š¦> though I have _no_ idea if it works at all
23:27fdobridge_: <gāfxstrand> Or put it in the shader header or something
23:27fdobridge_: <kāarolherbstš§š¦> yeah...
23:27fdobridge_: <gāfxstrand> It depends on how much custom chip design they're willing to pay NVIDIA for
23:27fdobridge_: <pāac85> Have two decoders in the shaders?
23:28fdobridge_: <pāac85> Sounds messy
23:28fdobridge_: <kāarolherbstš§š¦> they'll probably just put two SoCs on that thing...
23:28fdobridge_: <gāfxstrand> Probably
23:28fdobridge_: <pāac85> I think the most likely thing is they use sw emulation
23:29fdobridge_: <kāarolherbstš§š¦> you just mux the peripherals and boot it up on demand or something...
23:29fdobridge_: <gāfxstrand> Or that
23:29fdobridge_: <kāarolherbstš§š¦> yeah...
23:29fdobridge_: <kāarolherbstš§š¦> tho that's kinda expensive
23:29fdobridge_: <kāarolherbstš§š¦> I mean...
23:29HdkR: What's some dark silicon hanging off the AXI between friends? :P
23:29fdobridge_: <kāarolherbstš§š¦> they do ship emulators š
23:29fdobridge_: <gāfxstrand> Now that we're running real GPUs in consoles, emulation is a lot more practical.
23:29fdobridge_: <kāarolherbstš§š¦> not sure they have a GC emulator already on the switch?
23:30fdobridge_: <pāac85> Yes
23:30fdobridge_: <kāarolherbstš§š¦> I know there is the N64 one
23:30HdkR: NGC emulator was used for the Mario 3D collection
23:30fdobridge_: <kāarolherbstš§š¦> ahhh
23:30fdobridge_: <pāac85> The super Mario galaxy ports run in sort of emulation
23:30fdobridge_: <kāarolherbstš§š¦> but that's like...
23:30fdobridge_: <kāarolherbstš§š¦> not pure emulation, is it
23:30fdobridge_: <pāac85> Not sure about the details
23:31HdkR: For Mario Sunshine it's theoretically pure emulation, Mario Galaxy has been reported to just be a GPU emulator with the rest recompiled or something
23:31fdobridge_: <pāac85> I think they had an n64 emulator on the Wii btw
23:31fdobridge_: <kāarolherbstš§š¦> I even own the 3D collection :ferrisUpsideDown:
23:31fdobridge_: <pāac85> Actually I know they did
23:31fdobridge_: <kāarolherbstš§š¦> yeah.. recompiling is way more sane
23:32fdobridge_: <pāac85> I guess the flipper is peculiar enough that porting it made little sense
23:32HdkR: Emulating a 485Mhz PowerPC CPU isn't too hard :P
23:32fdobridge_: <pāac85> I guess the flipper is peculiar enough that porting thay part made little sense (edited)
23:32fdobridge_: <kāarolherbstš§š¦> well..
23:33HdkR: Especially with games that don't push the CPU and GPU to the maximum, which is nicer. Dolphin has had "idle skip" forever
23:34fdobridge_: <kāarolherbstš§š¦> right...