#asahi-dev · 2022-06-12 11:03
kettenisjannau: it is always best of the original author submits the diffs so I was hoping marcan would send it directly#asahi-dev · 2022-06-11 21:12
kettenislooks good#asahi-dev · 2022-06-11 20:33
kettenisburying that in the big diff is defenitely worse#asahi-dev · 2022-06-11 20:32
kettenisand nice that you did the timeout increase as a separate diff#asahi-dev · 2022-06-11 20:29
kettenisas comments#asahi-dev · 2022-06-11 20:28
ketteniscool, a see a few (very minor) things, so I'll add those#asahi-dev · 2022-06-11 17:08
kettenisactually you're right; each timeout is fatal, so waiting 1s should be ok#asahi-dev · 2022-06-11 16:41
kettenisbut 100ms should be fine#asahi-dev · 2022-06-11 16:41
kettenis1s may be a bit excessive since there may be several timeouts and it may take rather long for things to fail#asahi-dev · 2022-06-11 13:44
kettenisanyway, if you have a version that you're happy with I can add some comments on github#asahi-dev · 2022-06-11 13:44
kettenisshould add a #define for it#asahi-dev · 2022-06-11 13:38
kettenisendpoint 10 isn't handled by the linux kernel rtkit code is it?#asahi-dev · 2022-06-11 13:37
kettenis(that would be somewhat silly but the Kconfig stuff is supposed to express the dependencies properly)#asahi-dev · 2022-06-11 13:36
kettenistrue, but rtkit.o is included even if you disable nvme#asahi-dev · 2022-06-11 13:36
kettenisso maybe apple_rtkit_init() should accept function pointers to map/unmap and an opaque pointer for those callbacks#asahi-dev · 2022-06-11 13:35
kettenisone issue is that rtkit.c probably should call the sart functions directly#asahi-dev · 2022-06-11 13:34
kettenisI don't, but others might#asahi-dev · 2022-06-11 13:33
kettenisdo we know what endpoint 10 is about?#asahi-dev · 2022-06-11 13:32
kettenisthen it is probably fine#asahi-dev · 2022-06-11 13:31
kettenisand I don't know what the policy is about embedding code with a different license into a file is#asahi-dev · 2022-06-11 13:30
kettenisyeah, I have the feeling the _t typedefs will cause some issue for example#asahi-dev · 2022-06-11 13:22
kettenisjannau: are you planning to do the cleanup and submit the u-boot diff upstream?#asahi-dev · 2022-06-11 13:00
kettenisthat is fairly straightforward, but I'll leave it up to you folks to decide what the most fool-proof way is ;)#asahi-dev · 2022-06-11 12:53
kettenisis it possible to prevent the smc driver from loading?#asahi-dev · 2022-06-11 12:52
kettenisah, you're talking about fixing a broken install#asahi-dev · 2022-06-11 12:49
kettenisthen you don't need to disable smc in the device tree#asahi-dev · 2022-06-11 12:49
kettenisso I think an updated kernel is needed first#asahi-dev · 2022-06-11 12:44
kettenisno smc means no wifi#asahi-dev · 2022-06-09 19:17
kettenisso that'll need some rewriting#asahi-dev · 2022-06-09 19:17
kettenisI don't think simply transplanting the m1n1 code will go down well with the u-boot maintainers#asahi-dev · 2022-06-09 19:17
ketteniscool#asahi-dev · 2022-06-09 19:14
kettenisdoes that mean it works now?#asahi-dev · 2022-06-09 15:06
kettenisand so does pinctrl#asahi-dev · 2022-06-09 15:05
kettenisso i2c goes back all the way to s5l8960x#asahi-dev · 2022-06-09 14:31
kettenisah#asahi-dev · 2022-06-09 14:29
kettenisso chainloading linux directly from m1n1 doesn't work either?#asahi-dev · 2022-06-09 14:26
kettenisbut I was indeed thinking of folding sart into nvme_apple.c code like you did#asahi-dev · 2022-06-09 14:24
kettenisjannau: thanks; I need to find some time to look into this a bit deeper#asahi-dev · 2022-06-08 08:28
kettenisspeaking about the dart, I don't think t6000 support has been picked up by mainline yet is it?#asahi-dev · 2022-06-08 08:08
kettenislooks like it (but I don't think we have support for the t8110 dart yet)#asahi-dev · 2022-06-08 07:58
kettenisapcie also has a new compatible (but your t6k comment applies here as well)#asahi-dev · 2022-06-07 20:14
kettenisI probably won't have time to work on that before thursday#asahi-dev · 2022-06-07 20:13
kettenisok, so that means adding sart support#asahi-dev · 2022-06-07 18:41
kettenisso no M2 ADTs yet?#asahi-dev · 2022-06-07 12:55
ketteniseh, do we know anything about the m2 DART yet?#asahi-dev · 2022-06-07 12:44
kettenisbut if we can get away with just increasing some timeouts I'm all for it...#asahi-dev · 2022-06-07 12:18
kettenisto be honest, the main reason to skip sart was that the dt bindings were still in flux#asahi-dev · 2022-06-07 09:54
kettenismaz: yes, I wonder if adding a current_effective_el() function would be a reasonable thing to do (which would return 1 in EL2 Host mode)#asahi-dev · 2022-06-07 06:45
kettenisuntil someone fixed the sign extension#asahi-dev · 2022-06-07 06:44
kettenismaz: u-boot was setting the wrong value in TCR_EL2, but thanks to accidental sign-extensions of (1 << 31) I didn't notice