alyssa changed the topic of #panfrost to: Panfrost - FLOSS Mali Midgard & Bifrost - - Logs - Transientification is terminating. Memory reductions in progress.
stikonas has quit [Remote host closed the connection]
<hanetzer> senzu bean!
* hanetzer pelts robher
* HdkR wonders how many weebs catch the reference
* hanetzer pelts HdkR
<alyssa> robher: "// SPDX-License-Identifier:GPL-2.0"
<alyssa> This needs to be dual licensed with MIT so we can use it in userspace..?
<HdkR> hanetzer: Speaking of which the ending of the TFS Abridged version was great
<robher> alyssa: yes, good catch. It can be dual or 'GPL-2.0 WITH Linux-syscall-note'
<alyssa> I didn't realise there was a syscall note applying here, but that makes sense?
Elpaulo has quit [Ping timeout: 250 seconds]
Elpaulo has joined #panfrost
<Lyude> (can I vote gpl-2 with Linux syscall note?)
<alyssa> Yeah, blending is bork on here
<alyssa> Blending is fine on T760 in MFBD mode
<alyssa> And um, I know blending was fine on junior-year Panfrost, so..
<alyssa> Blending fixed
<alyssa> 90+, 32-
<alyssa> Hope I can clean this up
<alyssa> Anyways, I'm handling runtime-able hw selection atm
<alyssa> It's still kind of ugly code, but this is the right time to be doing that kind of cleanup
<alyssa> It feels really good to do this cleanup anyway
<alyssa> Like imagine, a single binary that can work on any Midgard system
<HdkR> Mentions Panfrost :)
<alyssa> So it does :)
<alyssa> <--- doesn't touch Biforst
<alyssa> But I'm excited for you guys!
<alyssa> ...Technically a good 40% of my work helps Bifrost but still
<alyssa> I should probably get yuq's EGL_BUFFER_AGE code back in review, it's needed for us too (if we don't want wallpapering fees)
<anarsoul> alyssa: I think daniels revived yuq's code
<HdkR> alyssa: Lyude: cwabbott: If you want N2 boards then I'll need addresses to ship them to. Can get them now. Took a bit longer than expected :P
<alyssa> anarsoul: That's a little different, I'm specifically talking about disabling BUFFER_AGE
<alyssa> Not just weston, also applies to e..g sway
<anarsoul> oh, OK
<alyssa> (It's also helpful, but slightly unrelated? I dunno.)
<alyssa> ^^ Just C201 things :)
<alyssa> Hey, nice, Panfrost and audio working on the same machine, must be nice :P
<alyssa> Wonder how long until Kodi 18 hits Debian
<alyssa> ---Debian is frozen right now isn't it--
<alyssa> Soft freeze begins today. Lovely.
<alyssa> 349+, 270-
<alyssa> ...wrong direction but it's worth it I guess :P
<tomeu> robher: are you planning to add MMU-related commits to the MR you opened?
ckeepax has quit [Quit: WeeChat 0.4.3]
<tomeu> alyssa: have fixed gpu version detection in the new driver, and I'm wondering how are you detecting that with kbase
pH5 has quit [Disconnected by services]
pH5 has joined #panfrost
<tomeu> robher: by the pantest code you pushed, I guess you haven't looked at allocations bigger than 0x29528?
<tomeu> actually, depends on what previous allocations were made
raster has joined #panfrost
mmind00 has quit [Remote host closed the connection]
mmind00 has joined #panfrost
jernej has quit [Quit: Free ZNC ~ Powered by LunarBNC:]
raster has quit [Remote host closed the connection]
ckeepax has joined #panfrost
_whitelogger has joined #panfrost
<tomeu> robher: I'm starting to implement job submission in the drm backend, based on a guessed up ioctl
<robher> tomeu: that boils down to code in mali_kbase_jm_hw.c. Which then got me looking at feature regs that tell how many cores and coherent clusters.
<tomeu> oh, i saw that passing by today
<tomeu> have you found it?
<tomeu> robher: ^
<robher> tomeu: there's only one cluster and we don't need jobslot 2 at least for what we care about for now.
<tomeu> robher: so you have pushed everything you had to push already?
<robher> tomeu: All I have is dumping some feature registers.
<tomeu> ah, thought you had some fixes for the MMU
<tomeu> robher: so the SUBMIT ioctl would basically contain an array of kbase_jd_atom?
<robher> tomeu: wait, I do need to push the gpu scheduler skeleton.
<tomeu> robher: is pantest running successively for you without errors?
<tomeu> the one you pushed is oopsing after I run it for 3 or 4 times
<robher> Just run it 2 times then. ;)
<tomeu> yeah well, the backtraces look very similar to when I try to allocate more than a few thousand bytes :)
<tomeu> robher: have you started moving pantest into igt? I could do that once I finish this
<robher> tomeu: nope. And I need to go work on DT for a few days...
<tomeu> ok, I prefer to not invest much time on the MMU, as you seem to have gone quite deep there already
<tomeu> so I can do igt today and next week will check if you have made progress on the MMU issues
afaerber has quit [Quit: Leaving]
afaerber has joined #panfrost
<tomeu> robher: do you have a tree in kernelci? we are already running igt on a few machines with midgard there, so it could be quite easy to run these tests there automatically
<robher> tomeu: no, I don't.
<tomeu> ok, maybe we'll think later of a good way of doing it
<robher> tomeu: I figured out the mapping issue. The io-pgtable lib expects everything aligned. Testing a fix now.
<tomeu> just in time, I'm debugging the igt tests now :)
<tomeu> and that's all from me for today
<tomeu> alyssa: if you push your T-760 stuff, I will try to find time to get the runtime switching working with the new driver
afaerber has quit [Ping timeout: 268 seconds]
afaerber has joined #panfrost
BenG83 has joined #panfrost
<alyssa> The WIP code is pushed to my personal branch, but there's some more cleanup needed before it can hit upstream
<robher> tomeu: pushed fix and few other pieces to panfrost/linux.
NeuroScr has quit [Ping timeout: 250 seconds]
pH5 has quit [Quit: bye]
stikonas has joined #panfrost
jernej has joined #panfrost
NeuroScr has joined #panfrost
BenG83 has quit [Quit: Leaving]
afaerber has quit [Quit: Leaving]
<alyssa> robher: "Don't need 3rd one until we have compute support"
<alyssa> FWIW, compute support is really simple (at least for simple stuff, ES3.1 style compute... OpenCL obviously makes things a lot more complicated), so it'll probably be implemented at least a little bit sooner than later, for debug purposes if nothing else
<alyssa> (Simple since vertex shaders are just compute jobs in the hw)
<alyssa> It's true that the raw values of GPU feature detection registers are passed through, but they're _also_ processed in kernel space (at least partially)
<alyssa> See mali_kbase_gpuprops.c for that in action
<alyssa> Not that we have to worry about that :)
<HdkR> alyssa: But what about OpenCL embedded profile? :P
<alyssa> HdkR: I don't know much about CL to answer that :P
<HdkR> basically drops the 64bit feature
<alyssa> That makes me happy :p
<HdkR> A few other things as well