#
…

2024-01-06

00:17nullspace has joined #asahi-gpu
00:27nullspace> The M1 lacks hardware instructions for image atomics, even though it has non-image atomics and non-atomic images
00:28nullspacei believe this isn't true for m2 and m3? hardware atomics are supported, right?
00:56nullspace has quit [Quit: Page closed]
01:23alyssaread a bit more carefully
01:39zawad has quit [Remote host closed the connection]
01:41zawad has joined #asahi-gpu
02:20Emantor has quit [Quit: ZNC - http://znc.in]
02:21Emantor has joined #asahi-gpu
02:34kidplayer666 has quit [Quit: Connection closed for inactivity]
03:01jeisom has joined #asahi-gpu
03:04skipwich has quit [Quit: DISCONNECT]
03:04skipwich has joined #asahi-gpu
03:35VeeBop has quit [Ping timeout: 480 seconds]
03:43capta1nt0ad has joined #asahi-gpu
03:49VeeBop has joined #asahi-gpu
03:50zawad has quit [Remote host closed the connection]
04:00jeisom has quit [Ping timeout: 480 seconds]
04:15marvin24_ has joined #asahi-gpu
04:15zawad has joined #asahi-gpu
04:18marvin24 has quit [Ping timeout: 480 seconds]
04:26capta1nt0ad has quit [Ping timeout: 480 seconds]
04:40capta1nt0ad has joined #asahi-gpu
04:57crabbedhaloablut has joined #asahi-gpu
05:02capta1nt0ad has quit [Ping timeout: 480 seconds]
05:06zawad has quit []
05:08loki_val has joined #asahi-gpu
05:12crabbedhaloablut has quit [Ping timeout: 480 seconds]
05:14Catyre has joined #asahi-gpu
05:14Catyre has quit [Remote host closed the connection]
05:18crabbedhaloablut has joined #asahi-gpu
05:19loki_val has quit [Ping timeout: 480 seconds]
05:20capta1nt0ad has joined #asahi-gpu
06:14capta1nt0ad has quit [Ping timeout: 480 seconds]
06:27possiblemeatball has quit [Quit: Quit]
06:52Carvajal has joined #asahi-gpu
07:22kidplayer666 has joined #asahi-gpu
09:37jeisom has joined #asahi-gpu
09:45jeisom has quit [Quit: Leaving]
09:47knedlik has joined #asahi-gpu
10:50knedlikSo I have successfully run the tests, and made a deduped text file of the non-passing ones: https://gist.github.com/KnedlikMCPE/4c4fffa220a668d0bb48851c77bb2062
10:50knedlikWhich of these would be a good thing to get started with?
10:52knedlikOh wait... forgot to include failed tests, one sec
10:56knedlikOkay, now it should be complete
11:57knedlik has quit [Quit: knedlik]
12:00john-cabaj has joined #asahi-gpu
12:08knedlik has joined #asahi-gpu
12:53knedlik has quit [Quit: knedlik]
12:53knedlik has joined #asahi-gpu
13:11john-cabaj has quit [Quit: john-cabaj]
13:12kidplayer666 has quit [Quit: Connection closed for inactivity]
13:51xcpy0 has quit [Quit: The Lounge - https://thelounge.chat]
13:58kidplayer666 has joined #asahi-gpu
15:01jeisom has joined #asahi-gpu
15:45zzywysm has joined #asahi-gpu
15:46zzywysm_ has quit [Ping timeout: 480 seconds]
16:50ameliabadelia has joined #asahi-gpu
16:56possiblemeatball has joined #asahi-gpu
18:29i509vcbknedlik: VK_EXT_global_priority(_query) the promoted KHR extension for that is implemented. If you did want to implement the older EXT extension I think you just need to enable those two extensions and test if anything changed during promotion?
18:30i509vcbYou can run individual tests with the -n argument on dEQP-vk to avoid running the entire suite every time
18:31AnuthaDev has joined #asahi-gpu
18:32i509vcbAnd the tests below testing for required format feature bits
18:33i509vcbI guess the problem you have there is that you know how many failed but not what specific test did
18:33i509vcbdeqp-runner gives you a results and failures.csv with every test name and whether it failed, crashed, flakes or passed
18:38i509vcb(That "Fail (Fail)" dedup is doing some heavy lifting)
18:46knedlikAlright, I was using the binary directly without deqp-runner, so I guess that was my issue
18:47knedlikI'll try taking a look at the 'vkGetPhysicalDeviceFormatProperties' from line 49 as you said
18:54knedlikHmm, I may be blind, but is the function not implemented yet at all? I'd assume "returning non-zero properties" doesn't happen in an unimplemented function
18:55i509vcbThat function is delegated to the FormatProperties2 function by mesa for us
18:56i509vcbVulkan has a few of these vkWhatever2 functions since 1.0 didn't consider adding pointer chains to some types or expected all data to be specified via a single fixed signature
18:58i509vcbSearch for vk_common_GetPhysicalDeviceFormatProperties to see where that forwarding is done
19:00knedlikSo I'm supposed to patch some function outside the asahi-specific tree and point it to some file in the asahi tree?
19:01i509vcbIt's all linked for you
19:01i509vcbYou just need to do stuff inside the agxv_ prefixed functions
19:02i509vcbPretty much look at agxv_GetPhysicalDeviceFormatProperties2
19:02knedlikThat makes sense now
19:03knedlikThere it is, I found it, thanks for the help
19:03i509vcbTLDR: Mesa makes a big function pointer lookup table for us
19:03knedlikIt makes sense for a project like this to be honest
19:04i509vcbMesa is also used for several other vulkan implementations so people do want to yeet some boilerplate
19:05knedlikI'd assume it should be as easy as adding a guard clause to check for the undefined format?
19:08i509vcbUhh I'd wonder what other drivers do here actually
19:08i509vcbI guess take a peek at nvk
19:09amelia has joined #asahi-gpu
19:09amelia is now known as Guest13092
19:13ameliabadelia has quit [Ping timeout: 480 seconds]
19:14AnuthaDev has quit []
19:15AnuthaDev has joined #asahi-gpu
19:22Guest13092 has quit [Ping timeout: 480 seconds]
19:31AnuthaDev has quit [Ping timeout: 480 seconds]
19:37AnuthaDev has joined #asahi-gpu
19:38amelia_ has joined #asahi-gpu
19:42knedlikI have a suspicion NVK does it in the image functions, which means we could just do a guard clause assuming it won't be needed elsewhere
19:49knedlikI need clarification here - the error is about returning non-zero properties, but the function is declared as void, where exactly is it returning the properties?
19:49knedlikNevermind, it's a pointer... :facepalm:
20:42i509vcbI guess we could just do the quick guard clause
20:42i509vcbThe properties are returned via one of the pointers
20:54knedlikI'm starting to get tired, but tommorow I'll do a quick test if the guard clause clears the test
20:55knedlikShould I make a pull request with the change to agxv/main once finished?
21:20maximbaz has quit [Quit: bye]
21:27john-cabaj has joined #asahi-gpu
21:36alyssaNVK is the reference drivr
21:50john-cabaj has quit [Ping timeout: 480 seconds]
21:58knedlik has quit [Quit: knedlik]
22:05AnuthaDev has quit []
22:10v9c6or has joined #asahi-gpu
22:15Stary has quit [Ping timeout: 480 seconds]
22:18amelia_ has quit [Ping timeout: 480 seconds]
22:20Stary has joined #asahi-gpu
22:24v9c6or has quit []
22:41xcpy0 has joined #asahi-gpu
23:16crabbedhaloablut has quit []
23:22seb91nihel has joined #asahi-gpu
23:22seb4nihel has quit [Ping timeout: 480 seconds]
23:41WindowPa- has quit [Ping timeout: 480 seconds]
23:52lawrence has joined #asahi-gpu
…