#
…

2026-08-22

00:05linuxgemini has quit [Quit: getting (hopefully fresh) air o/]
00:05maettu102 has quit [Quit: brb]
00:05maettu102 has joined #asahi-re
00:06Mary has quit [Quit: .]
00:07Mary has joined #asahi-re
00:08linuxgemini has joined #asahi-re
00:14rzhikharevich has quit [Ping timeout: 480 seconds]
00:23rzhikharevich has joined #asahi-re
00:31rzhikharevich has quit [Ping timeout: 480 seconds]
01:04bennofs3 has quit [Ping timeout: 480 seconds]
01:48bennofs3 has joined #asahi-re
02:00yurix has quit [Remote host closed the connection]
02:00yurix has joined #asahi-re
03:10rzhikharevich has joined #asahi-re
03:12bennofs3 has quit [Ping timeout: 480 seconds]
03:18rzhikharevich has quit [Ping timeout: 480 seconds]
03:19bennofs3 has joined #asahi-re
03:45rzhikharevich has joined #asahi-re
03:54rzhikharevich has quit [Ping timeout: 480 seconds]
04:09rzhikharevich has joined #asahi-re
04:17rzhikharevich has quit [Ping timeout: 480 seconds]
04:23rzhikharevich has joined #asahi-re
04:23pb17 has quit [Remote host closed the connection]
04:23pb17 has joined #asahi-re
04:31rzhikharevich has quit [Ping timeout: 480 seconds]
04:39rzhikharevich has joined #asahi-re
04:47rzhikharevich has quit [Ping timeout: 480 seconds]
04:55rzhikharevich has joined #asahi-re
05:03rzhikharevich has quit [Ping timeout: 480 seconds]
05:10rzhikharevich has joined #asahi-re
05:19rzhikharevich has quit [Ping timeout: 480 seconds]
05:24nela has quit [Ping timeout: 480 seconds]
05:24mrkajetanp has joined #asahi-re
05:26mrkajetanp_ has quit [Read error: Connection reset by peer]
05:31nela has joined #asahi-re
05:52scottjg4 has joined #asahi-re
06:02nela has quit [Ping timeout: 480 seconds]
06:02rzhikharevich has joined #asahi-re
06:21rzhikharevich has quit [Ping timeout: 480 seconds]
06:32rzhikharevich has joined #asahi-re
06:38sven has quit [Quit: The Lounge - https://thelounge.chat]
06:39sven has joined #asahi-re
06:41rzhikharevich has quit [Ping timeout: 480 seconds]
06:45maettu102 has quit [Quit: brb]
06:49maettu102 has joined #asahi-re
06:55nnym_ has joined #asahi-re
06:55nnym has quit [Read error: Connection reset by peer]
07:00weirdtreething has quit [Remote host closed the connection]
07:00weirdtreething has joined #asahi-re
07:05rzhikharevich has joined #asahi-re
07:20rzhikharevich has quit [Ping timeout: 480 seconds]
07:25pb17 has quit [Ping timeout: 480 seconds]
07:34pb17 has joined #asahi-re
08:13rzhikharevich has joined #asahi-re
08:21rzhikharevich has quit [Ping timeout: 480 seconds]
08:47ralim8 has quit [Quit: The Lounge - https://thelounge.chat]
08:47ralim8 has joined #asahi-re
09:50bennofs3 has quit [Quit: Ping timeout (120 seconds)]
09:51bennofs3 has joined #asahi-re
10:26pb17 has quit [Ping timeout: 480 seconds]
10:28pb17 has joined #asahi-re
12:17rzhikharevich has joined #asahi-re
12:26rzhikharevich has quit [Ping timeout: 480 seconds]
12:26rzhikharevich has joined #asahi-re
12:29pb17 has quit [Ping timeout: 480 seconds]
12:33pb17 has joined #asahi-re
12:42rzhikharevich has quit [Ping timeout: 480 seconds]
13:01rzhikharevich has joined #asahi-re
13:18rzhikharevich has quit [Remote host closed the connection]
13:18rzhikharevich has joined #asahi-re
13:27rzhikharevich has quit [Ping timeout: 480 seconds]
13:51TheLink has quit [Quit: Bye!]
13:53TheLink has joined #asahi-re
14:26rzhikharevich has joined #asahi-re
14:34rzhikharevich has quit [Ping timeout: 480 seconds]
14:37chadmed has quit [Quit: Konversation terminated!]
14:37chadmed has joined #asahi-re
15:31chaos_princessanyone with a m3 vanilla/pro and a tracing setup on 14.8.3 mind helping out find the tb reset address?
15:33sveni guess it also isn't hardcoded in the kext anymore but goes through some virtual(?) reg range that doesn't exist in the ADT now?
15:33svenyeah, that one also made me get the M2 one wrong at first :(
15:33chaos_princessthe reg range is in the adt on t6031, but i can't figure out how to map kext-internal numbers to adt numbers
15:33sveni couldn't figure out where that weird reg range comes from because ugh... c++ vtables
15:34svenin the M2 kext the range for the address was some offset that doesn't exist in the adt
15:34svenand once i had the right addr it didn't line up with the delta they add to anything in the ADT
15:34chaos_princessSo, my approach is - set a hw breakpoint in kext, read the address
15:35sveni think they somehow initialize that table with the adt entries first and then add a few more 'special' ones
15:35svenbut i didn't manage to track that down
15:35chaos_princessyea, ghidra outright choked on pmgr init, so, meh, not interested in going that way
15:35svenyeah, exactly
16:00bgb has quit [Remote host closed the connection]
16:00bgb has joined #asahi-re
16:27fstyle_funk has joined #asahi-re
16:29fstyle_funk has quit []
16:30fstyle_funk has joined #asahi-re
16:34bin01 has joined #asahi-re
16:40fstyle_funk has quit [Quit: Leaving]
16:41fstyle_funk has joined #asahi-re
17:02fstyle_funk has quit [Remote host closed the connection]
17:22rzhikharevich has joined #asahi-re
17:32Ariadne has quit [Ping timeout: 480 seconds]
17:38rzhikharevich has quit [Ping timeout: 480 seconds]
17:40bin01 has quit [Quit: Leaving]
17:47rzhikharevich has joined #asahi-re
17:55rzhikharevich has quit [Ping timeout: 480 seconds]
18:23rzhikharevich has joined #asahi-re
18:32rzhikharevich has quit [Ping timeout: 480 seconds]
18:50jannauchaos_princess: t8122/acio1 cio-reset 0x2d0780008
18:51chaos_princessm3 should also be using the pro/max style reset so acio0 should be the same
18:52jannauit writes 0x2 and polls until it is zero
18:53svensounds like the t600x style if that was port 2
18:53sventhat address also looks like t600x style
18:56rzhikharevich has joined #asahi-re
19:04jannauand 0x350780008 for t6030/acio1, same 0x2 write and poll for 0x0
19:05rzhikharevich has quit [Ping timeout: 480 seconds]
19:06jannauand just to make sure, acio0 on t8122 writes 0x1 to 0x2d0780008
19:07aurora has quit [Ping timeout: 480 seconds]
19:19garrettgr has joined #asahi-re
19:24rzhikharevich has joined #asahi-re
19:31chaos_princesswe should probably do something about dart remap
19:37rzhikharevich has quit [Remote host closed the connection]
19:45WindowPain6 has joined #asahi-re
19:51WindowPain has quit [Ping timeout: 480 seconds]
19:51WindowPain6 is now known as WindowPain
19:52svenlol, wtf
19:53svenso there are 12 tx rings and 12 rx rings and 24 dart stream ids. maybe one per id now?
19:54chaos_princessthat is my suspicion too
19:54svendoes that even work with the current dart driver? doesn't it support at most 3(?) darts?
19:54svenor did we fix that at some point?
19:54chaos_princessexcept there is 25 stream ids
19:54chaos_princessidk, worked on a m3max
19:55svenguess that fixed at some point then
19:55svenit's gonna program all of them to the same pagetable then
19:55chaos_princessmacos also appears to smash all of it down to stream 0 and only program that
19:56svenah, it's 3 dart instances but there's no limit on the streams within each dart
19:56sveni don't think we need remap for this, only difference to macos will be that it will invalidate all 24 streams at once instead of just setting a single stream id
19:57sventhe dt will just look a bit ugly, but oh well
19:57chaos_princessis it 25 identical copies of the page table, or one table programmed 25 times?
19:57svenone table programmed 25 times
19:58chaos_princessah, in that case yes, we don't care about remap
19:58chaos_princessi assumed it was the former
19:58svenit stuffs all those 25 dart instances into a single iommu domain and each iommu domain has a pagetable
19:58svenor was it iommu group?
19:58svenone of those :D
19:59svenif you trace it you should see the same pagetable in all TTBRs
20:00svenand if we wanted to use remap for this because the invalidation is faster we should be able to set that up entirely in apple-dart.c I think
20:01chaos_princessi was mostly concerned about the extra memory use since i incorrectly assumed it was duplicating the page table
20:02svenyeah, that would be bad
20:20jannauI have a hacked-up remap implementation in dart_probe(). just checks if remap exists and then programs it for t60[02]x acio
20:23jannaubut I don't want to write proper dt-bindings for it and since invalidation is not faster with remap on t6020 the ugly dts is a weak reason for it
20:25svenwe already have a sidmask per dart in some struct in there, the hack i had in mind was to just setup the first SID and enable remap to that one for all the others
20:26svenright now it does a foreach_set_bit over the sidmask, but i bet we could make that use remap instead and just program a single sid
20:27svenno dt changes required
20:30jannauI haven't seen 27 remapped sids on t8122 or t6030
20:30jannaubut that would still have multiple darts in iommus
20:33jannauit was t8122
20:58jannau1-11 are remapped to 0, 12 and 13 are translated with separate ttbr, 14 and 15 are bypass and 16-27 are remapped to 0
20:59chaos_princessi suspect 1-11 and 16-27 are per-ring
20:59chaos_princessas removing 16 made it not able to communicate
21:01svenwonder what 12 and 13 are then
21:02jannaunothing mapped there
21:03svenweird, i'm not aware of anything in the NHI that does DMA outside the rx/tx rings. and rtkit on t8103/t6000 uses sram mapped to the SoC bus
21:03jannauhas anyone experimented with FLUSH_VA_RANGE on hw_rev 2 t8110 darts?
21:05jannau12, 13 don't have valid ttbr so they are probably usnused
21:12svenand we've seen weird bypass SIDs since t8103 and don't set them up on linux anyway
21:14svenso maybe really just one SID for each rx/tx ring then
21:37dj has joined #asahi-re
21:38djanyone tried if smp issues are still there on t6050 on latest betas? there's been some iboot revs. not sure if apple would randomly drop a fix but one can hope
21:43dj has quit [Quit: bye]
22:18rzhikharevich has joined #asahi-re
22:26rzhikharevich has quit [Ping timeout: 480 seconds]
22:27rzhikharevich has joined #asahi-re
22:27pb17 has quit [Ping timeout: 480 seconds]
22:27leah has quit [Ping timeout: 480 seconds]
22:29pb17 has joined #asahi-re
22:36rzhikharevich has quit [Ping timeout: 480 seconds]
22:39rzhikharevich has joined #asahi-re
22:47rzhikharevich has quit [Ping timeout: 480 seconds]
…