#asahi-gpu · 2023-02-15 16:09
linaMost of what I did today is not UAPI-specific, it's all the batch tracking stuff to make explicit sync work#asahi-gpu · 2023-02-15 16:01
alyssapossibly some of the agx_batch stuff could too? IDK#asahi-gpu · 2023-02-15 16:00
alyssasince there's nothing downstream specific about drmSyncobjWait#asahi-gpu · 2023-02-15 16:00
alyssaand agx_fence.h#asahi-gpu · 2023-02-15 16:00
alyssaagx_fence.c can probably go upstream#asahi-gpu · 2023-02-15 15:58
alyssathose 2/3 patches can probably go in now#asahi-gpu · 2023-02-15 15:58
alyssaditto for shader key clone overread#asahi-gpu · 2023-02-15 15:58
alyssaDefinitely the memctx bug fix is, and should be split off into its own MR (this can happen ahead of the macOS stuff)#asahi-gpu · 2023-02-15 15:57
alyssaOnly question then is whether any of the code from agx/linux-explicit-sync is *not* specific to the UAPI (and should be upstreamed ahead of the new UAPI)#asahi-gpu · 2023-02-15 15:57
alyssaOnly question then is whether any of the code from agx/linux-explicit-sync is *not* specific to the UAPI (and should be upstreamed ahead of the new UAPI)#asahi-gpu · 2023-02-15 15:57
alyssa5. I use agx/linux-explicit-sync as the new base for my feature work and pick on top.#asahi-gpu · 2023-02-15 15:56
alyssa4. You rebase explicit sync on top of agx/linux and squash everything down to a new agx/linux-explicit-sync branch, that is just the new UAPI and not the old.#asahi-gpu · 2023-02-15 15:56
alyssa3. Rebase my agx/next upstream. It now consists of Linux UAPI patches + some feature work that's not yet ready. I drop the feature work and push an agx/linux branch.#asahi-gpu · 2023-02-15 15:55
alyssa2. Land my outstanding asahi MRs upstream#asahi-gpu · 2023-02-15 15:55
alyssa1. Remove macOS upstream#asahi-gpu · 2023-02-15 15:55
alyssalina: As for how to land this... what's easiest for me is going to be#asahi-gpu · 2023-02-15 15:54
alyssabuilds new kernel#asahi-gpu · 2023-02-15 15:53
linaI'll be back in a bit, I'm going to get some conbini dinner or something#asahi-gpu · 2023-02-15 15:52
alyssasweats#asahi-gpu · 2023-02-15 15:52
alyssayeah I get the warning too#asahi-gpu · 2023-02-15 15:51
linaIt wasn't actually breaking anything, just warning#asahi-gpu · 2023-02-15 15:51
linaI think a compiler upgrade started complaining about that one#asahi-gpu · 2023-02-15 15:51
lina^^;;#asahi-gpu · 2023-02-15 15:51
alyssaI'm regretting the unsafe union#asahi-gpu · 2023-02-15 15:51
alyssalina: "asahi: Fix shader key cloning overreads"#asahi-gpu · 2023-02-15 15:49
linaI couldn't figure out how to get real-time traces, so I don't know how much of the overhead is context switching, mailbox stuff, etc.#asahi-gpu · 2023-02-15 15:49
alyssaThanks#asahi-gpu · 2023-02-15 15:48
linaalyssa: Okay ^^#asahi-gpu · 2023-02-15 15:48
alyssaand rebase your branch on top of that#asahi-gpu · 2023-02-15 15:48
alyssalina: Please send the "drop macOS" MR, I'll review and merge#asahi-gpu · 2023-02-15 15:48
linaAt this point I think the next biggest submission overhead thing is just memory allocs, so I probably want to add some general mechanism to pool together random unrelated allocs for a single submission. But that can probably wait a bit.#asahi-gpu · 2023-02-15 15:48
alyssanice#asahi-gpu · 2023-02-15 15:47
linaalyssa: I also did some perf opt on the kernel side, there was some easy low-hanging CPU fruit with array initialization that gave me an extra 10% or so on glmark (for 4000+ FPS workloads, won't do as much for things that are more GPU-limited)#asahi-gpu · 2023-02-15 15:46
jannauyes. apple, x-gene and thunderx2#asahi-gpu · 2023-02-15 15:45
linajannau: Oh, I thought it was more general. So basically just Apple then?#asahi-gpu · 2023-02-15 15:44
linaAny ideas on how to land this? Most of the changes are in general code, but then end up talking to syncobjs of course. At this point we probably want to drop macOS first?#asahi-gpu · 2023-02-15 15:44
jannauno, just on all arm without the standard arm PMUv3#asahi-gpu · 2023-02-15 15:43
alyssaa few more fps is good ;0#asahi-gpu · 2023-02-15 15:43
alyssayeah that's a nice bump#asahi-gpu · 2023-02-15 15:42
linaI'm regularly hitting 200+ FPS on xonotic now at least ^^#asahi-gpu · 2023-02-15 15:42
alyssawill this make my supertuxkart go brr#asahi-gpu · 2023-02-15 15:42
alyssasounds like I got a kernel upgrade to do#asahi-gpu · 2023-02-15 15:42
alyssaahahahah got it nice#asahi-gpu · 2023-02-15 15:41
linaAlso perf works again. It was an upstream regression (on all ARM!), jannau found it while I was busy with mesa ^^#asahi-gpu · 2023-02-15 15:40
alyssaheh#asahi-gpu · 2023-02-15 15:40
linaI'm still not sure if the firmware stuff is really 100% robust yet, we'll find out when I turn on timestamps again since that made it really obvious in the past.#asahi-gpu · 2023-02-15 15:39
alyssaas long as the fault handling is nice and robut that sounds fine :)#asahi-gpu · 2023-02-15 15:39
alyssathanks for the heads up about exit faults#asahi-gpu · 2023-02-15 15:38
alyssalina: niiiiice :)