#
#asahi-gpu · 2023-07-17 20:32
alyssalina: silly question but, if CPU and GPU time are synched, why do we need anything asahi specific at all?
#asahi-gpu · 2023-07-17 15:25
karolherbstand then I can also just plump it through.. just not having a great plan of how to do that yet
#asahi-gpu · 2023-07-17 15:25
karolherbstmy current prototype is pushing all those enqueued system values into cb0 and rely on drivers to have the current system values, but I think I might have to start doing driver dependent lowering of system values soonish
#asahi-gpu · 2023-07-17 15:25
alyssakarolherbst: shrug.
#asahi-gpu · 2023-07-17 15:24
karolherbstalyssa: are you doing anything special with cb0 btw? Or would using the system value for dispatched group size still be beneficail over pulling data from cb0?
#asahi-gpu · 2023-07-17 15:22
alyssayep
#asahi-gpu · 2023-07-17 15:14
karolherbstdispatched block size is a bit pointless, because super trivial to emulate, but yeah
#asahi-gpu · 2023-07-17 15:13
karolherbstawesome :)
#asahi-gpu · 2023-07-17 15:13
karolherbstohh.. next message
#asahi-gpu · 2023-07-17 15:13
karolherbstalyssa: noice, also including sys values like amount of threads in the _current_ block?
#asahi-gpu · 2023-07-16 12:31
linaNot syscall, ioctl
#asahi-gpu · 2023-07-16 12:30
linarosefromthedead, j`ey: Looks like the timekeeping stuff is pretty deep kernel internals and I don't think it's worth fighting to figure out how to precisely expose the mapping to userspace... so I think a syscall that gets the GPU time and the current CLOCK_* time (not synchronized, so with a small error offset) is probably the least painful way to go for now
#asahi-gpu · 2023-07-16 12:14
j`eyI was just confirming the 'I think that directly reads the counter from userspace outright'
#asahi-gpu · 2023-07-16 12:13
linaj`ey: Problem is we can't use any internal APIs here, we need to go through the official entrypoints...
#asahi-gpu · 2023-07-16 12:12
rosefromthedeadokie
#asahi-gpu · 2023-07-16 12:11
linaThey *can* be thunked for 64bit in FEX. 32bit is always emulated right now. I haven't set up thunks so all my tests to date have been emulated.
#asahi-gpu · 2023-07-16 12:11
j`ey__arch_get_hw_counter in arch/arm64/include/asm/vdso/gettimeofday.h is what the vdso uses (I added the CNTVCTSS support a while back)
#asahi-gpu · 2023-07-16 12:11
rosefromthedeaderr, maybe i don't, i thought gl drivers were also native in fex and box
#asahi-gpu · 2023-07-16 12:09
rosefromthedeadi see
#asahi-gpu · 2023-07-16 12:09
rosefromthedeadahhh yeah
#asahi-gpu · 2023-07-16 12:08
linaWhat won't work is directly reading the ARM64 register from Mesa
#asahi-gpu · 2023-07-16 12:08
linaThey should, so that should work
#asahi-gpu · 2023-07-16 12:08
rosefromthedeaddo vdso calls not break out of x86 and call the native vdso?
#asahi-gpu · 2023-07-16 12:08
rosefromthedeadwait you've lost me
#asahi-gpu · 2023-07-16 12:07
linaSince vDSO should be plumbed in properly for emulation
#asahi-gpu · 2023-07-16 12:07
rosefromthedeadright
#asahi-gpu · 2023-07-16 12:07
linaThe problem is that won't work from an emulated x86 environment, we need an ioctl or gettimeofday for that
#asahi-gpu · 2023-07-16 12:07
linaWhich... I guess we could just do anyway? Though I need to check whether Linux guarantees that access is always enabled
#asahi-gpu · 2023-07-16 12:06
linaYes, that goes through vdso and I think that directly reads the counter from userspace outright
#asahi-gpu · 2023-07-16 12:04
rosefromthedeadwhat's the advantage of that? is it faster to get CLOCK_MONOTONIC?
#asahi-gpu · 2023-07-16 12:03
linaThere might also be some conversion factor we can provide instead to let userspace know how to convert from CLOCK_MONOTONIC or similar, but I need to investigate the kernel time tracking code to see how that works
#asahi-gpu · 2023-07-16 12:03
rosefromthedeadi guess that makes life easier
#asahi-gpu · 2023-07-16 12:03
rosefromthedeadoh wow
#asahi-gpu · 2023-07-16 12:03
lina(the raw ARM timer)
#asahi-gpu · 2023-07-16 12:02
linarosefromthedead: It's the same as the CPU time, I've been meaning to add an ioctl to grab that
#asahi-gpu · 2023-07-16 11:09
rosefromthedeadi've got the job finish timestamp wired up but gl also needs a way to read the time without submitting a job
#asahi-gpu · 2023-07-16 11:07
rosefromthedeadlina: do you know a way to read the current gpu time?
#asahi-gpu · 2023-07-16 10:36
linaThose messages are harmless, it's just the lockups were related to that so I wanted to see...
#asahi-gpu · 2023-07-16 10:35
linaYeah, that would be expected then!
#asahi-gpu · 2023-07-16 10:35
j`eyah ok, on #asahi-alt ry_darcy says the FF hangs are gone, but he's seeing those messages a lot
#asahi-gpu · 2023-07-16 10:35
linaI deliberately made that a bit noisy to see what the behavior is if someone runs into the issue again...
#asahi-gpu · 2023-07-16 10:34
linaj`ey: That was for debugging, if the lockups are gone I'll revert them ^^
#asahi-gpu · 2023-07-16 10:01
j`ey*Blocking, not locking
#asahi-gpu · 2023-07-16 10:01
j`eylina: in eb7a0ead749d269fc03596f4a6e2d789eab1fd4e 'locking due to fragment queue full' (and others) were changed from dev_dbg to dev_info, was that intentional?
#asahi-gpu · 2023-07-15 03:06
linaThis was a while ago though... I could be totally off
#asahi-gpu · 2023-07-15 03:06
linaBut I think I saw something where one of the dispatch modes ended up with some table provided that I assumed had something to do with workgroups
#asahi-gpu · 2023-07-15 03:06
linaMaybe? I might be misremembering too
#asahi-gpu · 2023-07-15 03:06
alyssaI think we might be talking about differnet things
#asahi-gpu · 2023-07-15 03:05
alyssalina: hm?
#asahi-gpu · 2023-07-15 03:02
linaalyssa: I'm pretty sure I saw that some point, when you do non-uniform sizes there's this big structure/table thing with the sizes I think?

plus ›