#asahi-gpu · 2022-10-02 14:25
karolherbstprobably makes sense to some degree being more flexible here as you know input/outputs are just memory anyway#asahi-gpu · 2022-10-02 14:25
karolherbstyeah.. I've heard#asahi-gpu · 2022-10-02 14:24
karolherbstthough my hope is to get to know more about other hardware with my CL work#asahi-gpu · 2022-10-02 14:22
karolherbstdid he#asahi-gpu · 2022-10-02 14:22
karolherbst:D#asahi-gpu · 2022-10-02 14:22
karolherbstyeah.. probably#asahi-gpu · 2022-10-02 14:21
karolherbstat least for vulkan and compute#asahi-gpu · 2022-10-02 14:21
karolherbstnvidia does all of its submission in userspace even#asahi-gpu · 2022-10-02 14:21
karolherbstright, not saying they are fundamentally different, just some hardware doens't need anything to be configured on the kernel side for things to run#asahi-gpu · 2022-10-02 14:19
karolherbstbut yes, we have to make sure in userspace to not cause the context to crash, but that's something the kernel isn't involved with#asahi-gpu · 2022-10-02 14:19
karolherbstit's literally just memory#asahi-gpu · 2022-10-02 14:19
karolherbstthere is no metadata attached when doing submissions#asahi-gpu · 2022-10-02 14:19
karolherbstthe only reason the kernel rejects is if the context crashes or we run out of memory#asahi-gpu · 2022-10-02 14:18
karolherbstwe don't have that in nvidia#asahi-gpu · 2022-10-02 14:18
karolherbstI just never had to deal with those problems myself :D#asahi-gpu · 2022-10-02 14:18
karolherbstyeah#asahi-gpu · 2022-10-02 14:18
karolherbstand make sure each batch is correct while recording#asahi-gpu · 2022-10-02 14:17
karolherbstyeah... just means that you will have to do multiple submissions when submitting a vulkan cmd buffer#asahi-gpu · 2022-10-02 14:16
karolherbstright.. hardware limitations exist and those are usually exposed in vulkan as limits#asahi-gpu · 2022-10-02 14:16
karolherbstwe do have more virtual limits, like the amount of bos allowed to use in a submission, but that's pretty much it#asahi-gpu · 2022-10-02 14:16
karolherbstswitching render targets is just a command#asahi-gpu · 2022-10-02 14:15
karolherbstnvidia doesn't because it's all done in the command buffer#asahi-gpu · 2022-10-02 14:15
karolherbstwell... depends#asahi-gpu · 2022-10-02 14:15
karolherbstright#asahi-gpu · 2022-10-02 14:15
karolherbsteither you split it up or something else#asahi-gpu · 2022-10-02 14:15
karolherbstso you have ot make sure the kernel accepts all of them#asahi-gpu · 2022-10-02 14:15
karolherbstand submit them explicitly#asahi-gpu · 2022-10-02 14:14
karolherbstyeah.. it's not special, but vulkan allows application to construct cmd buffers directly#asahi-gpu · 2022-10-02 14:14
karolherbstattachment lists#asahi-gpu · 2022-10-02 14:14
karolherbstyou will have to workaround that limit#asahi-gpu · 2022-10-02 14:14
karolherbstfor vulkan.... it's super annoying#asahi-gpu · 2022-10-02 14:14
karolherbstat least for GL that's easy to solve#asahi-gpu · 2022-10-02 14:13
karolherbstthough usually command buffers get that huge that flushing is a must anyway#asahi-gpu · 2022-10-02 14:13
karolherbstso as long things stay simple it's fine, but if games switch the render targets often it's bad#asahi-gpu · 2022-10-02 14:12
karolherbstokay#asahi-gpu · 2022-10-02 14:12
karolherbstlike "draw this shader pipeline"#asahi-gpu · 2022-10-02 14:12
karolherbstlina: one shader invocation #asahi-gpu · 2022-10-02 14:11
karolherbstright.. I meant it more like error checking and filling out proper structs#asahi-gpu · 2022-10-02 14:11
karolherbstso you can batch multiple render jobs into one submission and the kernel just sanitizes, submits and fence it#asahi-gpu · 2022-10-02 14:10
karolherbstyeah, that's not as terrible as I initially thought then#asahi-gpu · 2022-10-02 14:10
karolherbstokay#asahi-gpu · 2022-10-02 14:10
karolherbstthe hardware I know is more flexible. E.g. on nvidia you can assign "classes" to context which have priviliged access, which the kernel only allows for itself, and userspace can only allocated classes without special priviliges#asahi-gpu · 2022-10-02 14:09
karolherbstyeah okay, that's.. problematic#asahi-gpu · 2022-10-02 14:09
karolherbstahhh#asahi-gpu · 2022-10-02 14:09
karolherbstand it's more like filling actual structs with data do describe jobs, is that correct?#asahi-gpu · 2022-10-02 14:08
karolherbstmhhh, normally I'd expect that userpace just fills command buffers, sends that to the kernel driver and that's it. But it kind of sounds like that doesn't really exists here?#asahi-gpu · 2022-10-02 13:58
karolherbstwhat are those registers GL needs btw? Do you have some other ways of accessing that besides adding ioctls? Or is that mostly in the "context initialization" area?#asahi-gpu · 2022-10-02 12:18
karolherbstand with completely I mean: you can't do anything except looking at the screen#asahi-gpu · 2022-10-02 12:17
karolherbstalyssa: well.. try to come up with _good_ reasons why rendering wrong colors makes the UI break like completely :P#asahi-gpu · 2022-10-02 12:17
karolherbstyeah.. maybe it's fixed now