#asahi-dev · 2022-02-06 00:51
kettenissomethought needs to be put into adding support for t6000/t6001#asahi-dev · 2022-02-06 00:50
kettenisdefenitely @f#asahi-dev · 2022-02-06 00:50
kettenispmu-sera@0 should probably be pmu@f#asahi-dev · 2022-02-06 00:49
kettenishaving the rtc subnode is fine with me, but naming it apple,sera-pmu-rtc would make more sense#asahi-dev · 2022-02-06 00:48
kettenisthe spmi node should also have a apple,t8103-spmi compatible#asahi-dev · 2022-02-06 00:48
kettenisso unless you have a good reason to change it, please use that#asahi-dev · 2022-02-06 00:47
kettenisalso, I'm using apple,sera-pmu as the compatible#asahi-dev · 2022-02-06 00:46
kettenisbut the pmu should probably be in t8103-jxxx.dtsi since it is a separate chip#asahi-dev · 2022-02-06 00:46
kettenisjeffmiw: I think the spmi node should be in t8103.dtsi#asahi-dev · 2022-02-05 23:00
kettenisprobably in a different power domain#asahi-dev · 2022-02-05 22:02
kettenisand I noticed that i2c controller driver doesn't actually wait until the write transaction on the i2c bus completes#asahi-dev · 2022-02-05 22:01
kettenisso one thing I noticed is that if I do a write and immediately do a read it sometimes fails unless I add a 1ms delay#asahi-dev · 2022-02-04 22:48
kettenisright#asahi-dev · 2022-02-04 22:39
ketteniswell, that isn't an AIC register#asahi-dev · 2022-02-04 22:02
kettenisso you'll need another property to describe which register to look at#asahi-dev · 2022-02-04 22:02
ketteniswell, I think the interrupts property can only describe interrupts that are actually connected to the interrupt controller#asahi-dev · 2022-02-04 21:51
kettenisoh wait, audio is on admac-sio and admac-aop-audio is used for something else...#asahi-dev · 2022-02-04 21:49
ketteniswait, index 1 maps to 0x0038 and index 2 maps to 0x0034?#asahi-dev · 2022-02-04 21:46
kettenisso you mean for the other ADMAC you have to look at the register at offset 0x0030 to get the mask of channels that have interrupts pending?#asahi-dev · 2022-02-04 21:40
kettenisI don't understand, there is only a single interrupt listed for those admac controllers#asahi-dev · 2022-02-04 13:43
ketteniswhat is a "GART in the traditional sense"?#asahi-dev · 2022-02-04 09:21
kettenismarcan: we still don't really have a clue what the 0x800000 on the wifi gpios is about is it?#asahi-dev · 2022-02-03 20:14
kettenis(just realized that adding the display DARTs to the device tree killed the framebuffer on OpenBSD, probably because the code was putting it in bypass mode)#asahi-dev · 2022-02-03 20:12
kettenisthat seems undesirable#asahi-dev · 2022-02-03 20:12
kettenisah, still, resetting that dart will kill the framebuffer isn't it?#asahi-dev · 2022-02-03 19:35
kettenisjust the locked ones#asahi-dev · 2022-02-03 19:23
kettenismy preferred approach would be to not reset the DART in any way and have the DT list the iova space that should not be touched as reserved somehow#asahi-dev · 2022-02-03 16:42
kettenisah, the asahi installer doesn't copy the wifi firmware yet?#asahi-dev · 2022-02-03 09:32
kettenisreally feels like you guys are trying to run while we can barely walk...#asahi-dev · 2022-02-02 09:36
kettenismarcan: any progress on the smc mfd stuff?#asahi-dev · 2022-02-01 17:57
kettenissignify uses sha256 as the hash, but these days you can probably do better#asahi-dev · 2022-02-01 17:54
kettenisto verify you verify that /m1n1.sig was signed with the appropriate key and then you check that the hash for the loaded /m1n1.bin matches the signed hash#asahi-dev · 2022-02-01 17:53
kettenisso that's what /m1n1.sig would be#asahi-dev · 2022-02-01 17:52
kettenisyou typically sign the hash#asahi-dev · 2022-02-01 17:49
kettenissignify uses its own implementation that is derived from djb's public domain implementation#asahi-dev · 2022-02-01 17:48
kettenis(ed25519 I mean, not the rust implementation)#asahi-dev · 2022-02-01 17:48
ketteniswhich happens to be what OpenBSD's signify tool uses as well#asahi-dev · 2022-02-01 17:47
kettenised25519 was mentioned#asahi-dev · 2022-02-01 09:09
kettenisbut just reading the cs42l83 revision register seems to work reliably for me#asahi-dev · 2022-02-01 09:07
kettenismarcan: I have seen some issues with reading back registers from the tas5770 codec immediately after writing to them (in OpenBSD)#asahi-dev · 2022-01-31 09:06
kettenisI see too many open source software projects that end up with insane build-dependencies#asahi-dev · 2022-01-31 09:06
kettenisanyway, don't just use tools because they're cool#asahi-dev · 2022-01-31 09:05
ketteniswell, if you are using alloc...#asahi-dev · 2022-01-31 08:42
kettenis(full disclosure: I'm a rust skeptic)#asahi-dev · 2022-01-31 08:42
ketteniscan we please avoid pulling in the world?#asahi-dev · 2022-01-30 21:01
ketteniswe don't have a mailing list (yet) so a branch on github is probably te next best thing#asahi-dev · 2022-01-29 16:13
ketteniswith 0 meaning disabled probably#asahi-dev · 2022-01-29 16:04
kettenisnot all of the register definitions in mca.c on the asahi branch make sense to me#asahi-dev · 2022-01-29 16:00
kettenispovik: that would probably be helpful#asahi-dev · 2022-01-29 15:51
kettenisyup