#asahi-gpu · 2021-04-04 02:35
DarkShadow44bloom: For my shader emulator that I use for testing: https://github.com/DarkShadow44/AsahiLinux-gpu/blob/asmtool/asmtool/asmtool/emulator.c#asahi-gpu · 2021-04-04 02:32
DarkShadow44ideally, I'd want exactly the same behavior as the GPU#asahi-gpu · 2021-04-04 02:32
DarkShadow44bloom: load/store allow the data format "SRGBA8", and I have no idea how to convert that properly#asahi-gpu · 2021-04-04 02:31
DarkShadow44dougall: You use a table, did you only get that through brute force testing?#asahi-gpu · 2021-04-04 02:31
DarkShadow44Still working on load/store emulator... Anyone knows a proper RGB<->sRGB implementation?#asahi-gpu · 2021-04-01 17:02
DarkShadow44basically, I'm here to help out and learn in the progress ^^#asahi-gpu · 2021-04-01 17:01
DarkShadow44I mean, I've written my own small compiler back in the days, but that's that#asahi-gpu · 2021-04-01 17:01
DarkShadow44compiler theory is not really my thing though#asahi-gpu · 2021-04-01 17:01
DarkShadow44but I wanted to get into understanding GPU hardware and drivers as well, that's why I started here#asahi-gpu · 2021-04-01 17:00
DarkShadow44I'm pretty interested in re, that's why I did a bunch of work for wine#asahi-gpu · 2021-04-01 17:00
DarkShadow44hm, I'm interested in a lot of things#asahi-gpu · 2021-04-01 16:57
DarkShadow44I've still got most instructions to implement and test as well#asahi-gpu · 2021-04-01 16:57
DarkShadow44no rush, I'm thankful for all the help I can get ^^#asahi-gpu · 2021-04-01 16:43
DarkShadow44On my own, I might not be the best for architecture/codestyle#asahi-gpu · 2021-04-01 16:42
DarkShadow44would you be interested in helping me turn that into something useable for mesa?#asahi-gpu · 2021-04-01 16:42
DarkShadow44I should be finished with my asmtool architecture soon(ish), btw#asahi-gpu · 2021-04-01 16:40
DarkShadow44same here, C++ seems just so... messy#asahi-gpu · 2021-04-01 16:35
DarkShadow44bloom: Ah, only really ever saw the picture inline in the Gallium3D article :D#asahi-gpu · 2021-04-01 16:33
DarkShadow44oh, sorry then#asahi-gpu · 2021-04-01 16:31
DarkShadow44not sure how accurate that diagram still is though#asahi-gpu · 2021-04-01 16:31
DarkShadow44https://en.wikipedia.org/wiki/File:Mesa_layers_of_crap_2016.svg#asahi-gpu · 2021-04-01 16:30
DarkShadow44of course#asahi-gpu · 2021-04-01 16:28
DarkShadow44alright#asahi-gpu · 2021-04-01 16:26
DarkShadow44chrisf: What do you plan to use?#asahi-gpu · 2021-04-01 16:24
DarkShadow44regarding both mesa drivers, should they be in C or C++?#asahi-gpu · 2021-04-01 16:23
DarkShadow44but that seems to not be the case#asahi-gpu · 2021-04-01 16:23
DarkShadow44hm, I see. I thought maybe gallium also supports vulkan#asahi-gpu · 2021-04-01 16:05
DarkShadow44(is there even a difference?)#asahi-gpu · 2021-04-01 16:05
DarkShadow44when making a mesa driver, do we make a gallium3d driver or a vulkan driver?#asahi-gpu · 2021-03-27 01:05
DarkShadow44What do you mean?#asahi-gpu · 2021-03-27 00:51
DarkShadow44I know that feeling#asahi-gpu · 2021-03-27 00:48
DarkShadow44heh, should make things easier#asahi-gpu · 2021-03-27 00:47
DarkShadow44mh, I see#asahi-gpu · 2021-03-27 00:43
DarkShadow44what exactly is that "fp16vec2" thing?#asahi-gpu · 2021-03-27 00:43
DarkShadow44huh, interesting#asahi-gpu · 2021-03-27 00:22
DarkShadow44need to make tests for that#asahi-gpu · 2021-03-27 00:22
DarkShadow44mh, it's a bit odd that the mask is one of the parts that are cut off#asahi-gpu · 2021-03-27 00:19
DarkShadow44yeah, easier on the cache#asahi-gpu · 2021-03-27 00:17
DarkShadow44I thought an advantage of, for example, ARM was easier decoding due to fix lengths#asahi-gpu · 2021-03-27 00:16
DarkShadow44although I gotta admit, I don't understand in having the benefit of having variable length instructions like that#asahi-gpu · 2021-03-27 00:16
DarkShadow44no unknown bits is best#asahi-gpu · 2021-03-27 00:16
DarkShadow44I mean, I wouldn't necessarily use it, but it'd be good to know, IMHO#asahi-gpu · 2021-03-27 00:15
DarkShadow44Sure thing, but that's where tests come in, no?#asahi-gpu · 2021-03-27 00:13
DarkShadow44most instructions support such a cut off#asahi-gpu · 2021-03-27 00:13
DarkShadow44neither have I, but maybe they planned it into the hardware#asahi-gpu · 2021-03-27 00:11
DarkShadow44shoul cut off the last two bytes#asahi-gpu · 2021-03-27 00:11
DarkShadow44that doesn't mean we can't manually generate bytecode like that >:D#asahi-gpu · 2021-03-27 00:09
DarkShadow44just a theory#asahi-gpu · 2021-03-27 00:09
DarkShadow44although, it raises a question: Since registers are specified as Rx:R, maybe bit 31 is for another "cut off"?#asahi-gpu · 2021-03-27 00:09
DarkShadow44I assume it's split into Fx:D because of the different instruction length, since all bytes that are cut off are assumed 0