#asahi-gpu · 2021-03-23 21:13
chrisfi havent looked in detail at the calling convention, if you want to figure it out, great#asahi-gpu · 2021-03-23 21:13
chrisfi havent looked in detail at the calling convention, if you want to figure it out, great#asahi-gpu · 2021-03-23 21:13
DarkShadow44that does make sense. Maybe the mantle calling convention just starts function parameters from r20 upwards...#asahi-gpu · 2021-03-23 21:13
DarkShadow44that does make sense. Maybe the mantle calling convention just starts function parameters from r20 upwards...#asahi-gpu · 2021-03-23 21:12
chrisfDarkShadow44: some other rando function may have different values in every lane#asahi-gpu · 2021-03-23 21:12
chrisfDarkShadow44: some other rando function may have different values in every lane#asahi-gpu · 2021-03-23 21:11
chrisfDarkShadow44: when the buffer is a parameter to the entrypoint function you know it's uniform across all lanes#asahi-gpu · 2021-03-23 21:11
chrisfDarkShadow44: when the buffer is a parameter to the entrypoint function you know it's uniform across all lanes#asahi-gpu · 2021-03-23 21:11
DarkShadow44makes sense, thanks#asahi-gpu · 2021-03-23 21:11
DarkShadow44makes sense, thanks#asahi-gpu · 2021-03-23 21:10
DarkShadow44though I'm not entirely sure why the main GPU program expects them starting from u0_u1 but the .metallib expects them from r20_r21#asahi-gpu · 2021-03-23 21:10
DarkShadow44though I'm not entirely sure why the main GPU program expects them starting from u0_u1 but the .metallib expects them from r20_r21#asahi-gpu · 2021-03-23 21:09
chrisfDarkShadow44: some of the details of that pseudocode are wrong [what happens with a hole in the mask, etc], i had pretty incomplete information at the time#asahi-gpu · 2021-03-23 21:09
chrisfDarkShadow44: some of the details of that pseudocode are wrong [what happens with a hole in the mask, etc], i had pretty incomplete information at the time#asahi-gpu · 2021-03-23 21:08
DarkShadow44Yeah, I expected something like that#asahi-gpu · 2021-03-23 21:08
DarkShadow44Yeah, I expected something like that#asahi-gpu · 2021-03-23 21:05
chrisfdougall has since figured out a lot more of the encoding but we're short a description of exactly what happens#asahi-gpu · 2021-03-23 21:05
chrisfdougall has since figured out a lot more of the encoding but we're short a description of exactly what happens#asahi-gpu · 2021-03-23 21:03
chrisfDarkShadow44: there's a better description of the "simple" loads and stores in the older docs at https://hthh.github.io/m1-gpu-re/#asahi-gpu · 2021-03-23 21:03
chrisfDarkShadow44: there's a better description of the "simple" loads and stores in the older docs at https://hthh.github.io/m1-gpu-re/#asahi-gpu · 2021-03-23 21:02
chrisfDarkShadow44: should be the base address of the buffer#asahi-gpu · 2021-03-23 21:02
chrisfDarkShadow44: should be the base address of the buffer#asahi-gpu · 2021-03-23 20:48
DarkShadow44hmm maybe the adress of the buffer on the gpu#asahi-gpu · 2021-03-23 20:48
DarkShadow44hmm maybe the adress of the buffer on the gpu#asahi-gpu · 2021-03-23 20:37
DarkShadow44hmm maybe the adress of the buffer on the gpu#asahi-gpu · 2021-03-23 20:37
DarkShadow44hmm maybe the adress of the buffer on the gpu#asahi-gpu · 2021-03-23 20:02
DarkShadow44I think I understand most of the others, but that currently trips me up. The GPU just softfaults, and I guess my lack of understanding of this part is to blame#asahi-gpu · 2021-03-23 20:02
DarkShadow44I think I understand most of the others, but that currently trips me up. The GPU just softfaults, and I guess my lack of understanding of this part is to blame#asahi-gpu · 2021-03-23 20:02
DarkShadow44do you know what exactly u0_u1 do here?#asahi-gpu · 2021-03-23 20:02
DarkShadow44do you know what exactly u0_u1 do here?#asahi-gpu · 2021-03-23 20:02
DarkShadow44device_load 1, 0, 0, 4, 0, i32, quad, r12_r13_r14_r15, u0_u1, r32, unsigned, lsl 2#asahi-gpu · 2021-03-23 20:02
DarkShadow44device_load 1, 0, 0, 4, 0, i32, quad, r12_r13_r14_r15, u0_u1, r32, unsigned, lsl 2#asahi-gpu · 2021-03-23 20:02
DarkShadow44Assuming the following disassembly from applegpu/disassembly.py:#asahi-gpu · 2021-03-23 20:02
DarkShadow44Assuming the following disassembly from applegpu/disassembly.py:#asahi-gpu · 2021-03-21 17:32
DarkShadow44on your github it's still maked TODO(), that's why I ask#asahi-gpu · 2021-03-21 17:32
DarkShadow44on your github it's still maked TODO(), that's why I ask#asahi-gpu · 2021-03-21 17:31
DarkShadow44if you don't have anything, maybe I could write something up?#asahi-gpu · 2021-03-21 17:31
DarkShadow44if you don't have anything, maybe I could write something up?#asahi-gpu · 2021-03-21 17:31
DarkShadow44Because for simplicity reason, I need those first to start with proper testing#asahi-gpu · 2021-03-21 17:31
DarkShadow44Because for simplicity reason, I need those first to start with proper testing#asahi-gpu · 2021-03-21 17:31
DarkShadow44dougall: Do you already have documentation prepared to device_store/device_load?#asahi-gpu · 2021-03-21 17:31
DarkShadow44dougall: Do you already have documentation prepared to device_store/device_load?#asahi-gpu · 2021-03-21 11:56
dougallheh - yeah, iirc they just use a couple of (hopefully) uninteresting instructions i haven't named yet, so it might be a good beginner instruction to re#asahi-gpu · 2021-03-21 11:56
dougallheh - yeah, iirc they just use a couple of (hopefully) uninteresting instructions i haven't named yet, so it might be a good beginner instruction to re#asahi-gpu · 2021-03-21 11:29
bloomMesa doesn't even ave asin/acos/atan acceleration wired up, it's lowered to a (iirc polynomial..?) approximation unconditionally :-P#asahi-gpu · 2021-03-21 11:29
bloomMesa doesn't even ave asin/acos/atan acceleration wired up, it's lowered to a (iirc polynomial..?) approximation unconditionally :-P#asahi-gpu · 2021-03-21 11:24
bloomasin/acos/atan even more so#asahi-gpu · 2021-03-21 11:24
bloomasin/acos/atan even more so#asahi-gpu · 2021-03-21 11:24
bloom(nowadays)#asahi-gpu · 2021-03-21 11:24
bloom(nowadays)