2026-01-31
00:10chadmed has quit [Quit: Konversation terminated!]
00:15chadmed has joined #asahi-gpu
01:03CME has quit [Ping timeout: 480 seconds]
01:38CME has joined #asahi-gpu
02:13john-cabaj has joined #asahi-gpu
02:32cr1901__ has quit [Read error: Connection reset by peer]
02:34cr1901 has joined #asahi-gpu
02:41maettu102 has quit [Quit: The Lounge - https://thelounge.chat]
02:43maettu102 has joined #asahi-gpu
02:47olivial has quit [Ping timeout: 480 seconds]
02:47pb17 has quit [Ping timeout: 480 seconds]
02:49olivial has joined #asahi-gpu
03:03pb17 has joined #asahi-gpu
04:03michaelros has quit [Read error: Connection reset by peer]
04:04michaelros has joined #asahi-gpu
04:04michaelros has quit []
04:05michaelros has joined #asahi-gpu
04:15john-cabaj has quit [Ping timeout: 480 seconds]
05:57hamr8 has quit [Quit: The Lounge - https://thelounge.chat]
05:58hamr8 has joined #asahi-gpu
06:16rosalyntg has quit [Read error: Connection reset by peer]
06:16rosalyntg has joined #asahi-gpu
06:39pb17 has quit [Ping timeout: 480 seconds]
06:39pb17 has joined #asahi-gpu
07:45rainbyte has quit [Read error: Connection reset by peer]
07:45rainbyte has joined #asahi-gpu
08:06pb17 has quit [Ping timeout: 480 seconds]
08:14pb17 has joined #asahi-gpu
08:17integralpilot has joined #asahi-gpu
08:19kitlith has joined #asahi-gpu
08:19integralpilotokay looks like texture sampling has a new opcode, 0x8005 (was 0x31 in M1/M2), and it's 14 bytes long now (was 8-12 before)
08:28rainbyte has quit [Read error: Connection reset by peer]
08:28rainbyte has joined #asahi-gpu
08:59integralpilotokay so now i know that byte 6 is the dimention (0x09 = 2D, 0x80=1D, 0x39=3D, 0x13=Cube)
09:00integralpilotand byte 11 in the level of detail mode (0x00=implicit, 0x04=bias)
09:00integralpilotstill decoding the other bytes
09:06integralpilothave to fix up the hw testbed code (of applegpu) though as even in the M3 branch it doesn't seem to have been ported for M3 yet, doing that now.
09:13integralpilotokay seems to be working now but every 8th thread will only return 0s, finding out why now
09:28integralpiloti figured it out, now the hardware testbed works!
09:29integralpilotokay so instructions are documented (i.e. mov) but don't have a hardware testbed emulator yet so we can't test them (the testbed compares emulator vs real to make sure our expectations are right), so i'm going to work on expanding the emulator for M3
09:31integralpilot*only some instructions are documented for M3 by the existing work on M3 branch, but not many, and none have an emulator impl
09:31integralpilot(which i'm fixing now!)
09:35pb17 has quit [Ping timeout: 480 seconds]
09:40integralpilotOMG there's a 10-byte long bitop instruction (just one opcode for this) that has a 4-bit truth table encoding so that ANY 2-input boolean function will run with the same opcode
09:40integralpilotthat's like actually kinda cool love that Apple
09:57pb17 has joined #asahi-gpu
10:08nickchan has joined #asahi-gpu
11:24pb17 has quit [Ping timeout: 480 seconds]
11:29integralpilotthe existing applegpu testing method of running arbitary instructions on real hardware is broken, refactoring it to a new testing way i thought of that verifies our assumptions are right. hopefully it works.
11:30pb17 has joined #asahi-gpu
12:28integralpilot has quit [Quit: integralpilot]
14:05KongrooParadox has joined #asahi-gpu
14:07[KongrooParadox] has quit [Ping timeout: 480 seconds]
14:22yuyichao has quit [Read error: Connection reset by peer]
14:22yuyichao has joined #asahi-gpu
15:02yuyichao_ has joined #asahi-gpu
15:02yuyichao has quit [Read error: Connection reset by peer]
15:37karolherbstwondering if we should have a nir instruction for it, because writing opts in nir for it is simpler than in backends and every backend will need the same things... or maybe just have a util function, because in the end it's just three constants you apply the ops on and get a result which is your LUT
15:52karolherbstalyssa: funny.. nvidia uses the reverse order: 0xf0, 0xcc, 0xaa. Intel seems to be 0xaa, 0xcc, 0xf0
17:26pb17 has quit [Ping timeout: 480 seconds]
17:56pb17 has joined #asahi-gpu
18:09yuyichao_ has quit [Ping timeout: 480 seconds]
18:11yuyichao_ has joined #asahi-gpu
18:44[KongrooParadox] has joined #asahi-gpu
18:47KongrooParadox has quit [Ping timeout: 480 seconds]
19:32rainbyte has quit [Ping timeout: 480 seconds]
19:33olivial has quit [Ping timeout: 480 seconds]
19:41olivial has joined #asahi-gpu
20:13alicela1n has joined #asahi-gpu
20:41alicela1n has quit [Read error: Connection reset by peer]
21:06alicela1n has joined #asahi-gpu
21:15alessaba has joined #asahi-gpu
21:17alessaba has quit [Remote host closed the connection]
21:17alessaba has joined #asahi-gpu
21:18alessaba has quit [Remote host closed the connection]
21:20alessaba has joined #asahi-gpu
21:22alessaba has quit [Remote host closed the connection]
21:23alessaba has joined #asahi-gpu
21:23alessaba has quit [Remote host closed the connection]
21:28alessaba has joined #asahi-gpu
21:29alessaba has quit [Remote host closed the connection]
21:29clawd_bot has joined #asahi-gpu
21:49sven has quit [Remote host closed the connection]
21:52dhewg has quit [Remote host closed the connection]
21:52clawd_bot was kicked from #asahi-gpu by ChanServ [You are not permitted on this channel]
22:00alicela1n has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
22:00sven has joined #asahi-gpu
22:01dhewg has joined #asahi-gpu
22:02alicela1n has joined #asahi-gpu
22:20nightstrike has quit [Quit: Connection closed for inactivity]
22:29pb17 has quit [Ping timeout: 480 seconds]
22:38alicela1n has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
22:39pb17 has joined #asahi-gpu
22:41alicela1n has joined #asahi-gpu
22:55alicela1n has quit [Read error: Connection reset by peer]