#
✕ nick:kettenis
#asahi-dev · 2023-05-01 14:35
kettenisyes, but I think I can pull all that off
#asahi-dev · 2023-05-01 14:25
kettenisloading the firmware directly from the ESP would make updating u-boot easier and I suspect it would result in less resistence upstream
#asahi-dev · 2023-05-01 14:13
kettenishmm, u-boot actually has some infrastructure to load firmware from a filesystem
#asahi-dev · 2023-05-01 11:37
kettenismaybe I should introduce a tiny abstraction layer in u-boot that prepends vendorfw/ such that the firmware name in the driver matches what Linux has
#asahi-dev · 2023-05-01 11:30
ketteniss/pauths/paths (I've been playing around with pointer authentication last week)
#asahi-dev · 2023-05-01 11:28
kettenismarcan: when we change the installer to produce a .cpio with just the firmwares that u-boot cares about, it will still have pauths that start with vendorfw/ ?
#asahi-dev · 2023-05-01 10:01
kettenisyeah, that'd make sense
#asahi-dev · 2023-05-01 09:50
ketteniswe probably wouldn't want u-boot to do this by default anyway
#asahi-dev · 2023-05-01 09:48
kettenisdoing this in u-boot would involve writing an spmi driver
#asahi-dev · 2023-04-28 15:35
kettenisah, ok
#asahi-dev · 2023-04-28 15:34
kettenisyou're going to change the linux driver to do that?
#asahi-dev · 2023-04-28 15:33
kettenisthat makes more sense
#asahi-dev · 2023-04-28 15:31
kettenisI should have some time to get u-boot in releasable shape next week
#asahi-dev · 2023-04-27 12:56
kettenisat least that doesn't depend on the room ;)
#asahi-dev · 2023-04-27 12:55
kettenisjust use headphones ;)
#asahi-dev · 2023-04-27 11:18
kettenisyes
#asahi-dev · 2023-04-27 11:17
kettenisno, but it will do resets
#asahi-dev · 2023-04-27 10:50
kettenisthe android bcm4389 driver has different code there handling two different cases and none if that matches what the asahi linux brcmfmac driver does
#asahi-dev · 2023-04-27 10:48
kettenisdriver needs some changes; on openbsd I end up in the equivalent of brcmf_chip_sysmem_ramsize() but the code in that function is wrong
#asahi-dev · 2023-04-27 10:45
kettenisit shows up for on openbsd
#asahi-dev · 2023-04-27 10:41
kettenisargh, why did it create a merge commit for that...
#asahi-dev · 2023-04-27 10:38
kettenisthanks; I'll merge it and tweak the commit message before sending it upstream
#asahi-dev · 2023-04-27 10:20
kettenisis it ok if I do that and keep kaazoo's s-o-b?
#asahi-dev · 2023-04-27 10:20
ketteniskaazoo, marcan: that commit message probably needs a bit of tweaking to be acceptable upstream
#asahi-dev · 2023-04-27 10:17
kettenis_jannau_: got conflicting reactions on that one
#asahi-dev · 2023-04-27 10:12
kettenisI was planning to merge the M2 Pro/Max stuff and the asmedia firmware loading stuff into the asahi branch soon
#asahi-dev · 2023-04-27 10:10
ketteniskaazoo even
#asahi-dev · 2023-04-27 10:09
ketteniskazoo: I thik that needs a s-o-b with a valid e-mail address
#asahi-dev · 2023-04-27 09:59
kettenisthose commits do have marcan's signed-of-by
#asahi-dev · 2023-04-27 09:59
kettenisI think I should try to upstream at least marcan's rtkit diffs, but I read conflicting things about what tags should be added
#asahi-dev · 2023-04-27 09:55
ketteniscan someone explain me the rules about upstreaming other people's diffs in this strange world of magic git commit message tags?
#asahi-dev · 2023-04-17 19:29
kettenisI like the sound of that
#asahi-dev · 2023-04-17 13:16
kettenisbut maybe the linux kernel already saves/restores those
#asahi-dev · 2023-04-17 13:15
kettenisthings like the pointer authentication keys
#asahi-dev · 2023-04-17 13:15
kettenismarcan: I believe I read somewhere that "deep wfi" dropped a bit more state than the GPRs you save/restore
#asahi-dev · 2023-04-17 13:01
kettenisfor cpu0 that will require assistence from m1n1
#asahi-dev · 2023-04-17 13:00
kettenisand you need to bring them back up starting from rvbar
#asahi-dev · 2023-04-17 13:00
kettenisyes, there should be a state where the cores lose all state
#asahi-dev · 2023-04-17 12:58
kettenisThis "deep" wfi state is probably designed to allow boosting of one of the other cores in the cluster
#asahi-dev · 2023-04-17 12:57
kettenisThis can't power off the cluster completely since the cores still maintain some state
#asahi-dev · 2023-04-17 12:57
kettenisregarding the power savings for deep wfi; I didn't get much savings with that when I played with it under OpenBSD
#asahi-dev · 2023-04-17 12:53
kettenismarcan: sorry, I started writing things down, but got distracted by other stuff again
#asahi-dev · 2023-04-16 16:35
ketteniswhich doesn't make sense as a memory address either
#asahi-dev · 2023-04-16 16:34
kettenisbut 0xf0025f << 12 = 0x25f000
#asahi-dev · 2023-04-16 16:16
kettenis(t8112)
#asahi-dev · 2023-04-16 16:16
kettenisthis is on m2?
#asahi-dev · 2023-04-16 16:15
kettenisso it is sending a message with a pre-assigned address?
#asahi-dev · 2023-04-16 15:06
kettenisI don't see any oslog messages on my machines (all with 13.2 or even older firmware)
#asahi-dev · 2023-04-16 12:56
kettenisbut that is best done once we have bindings
#asahi-dev · 2023-04-16 12:56
ketteniswe still have to get the mtp stuff in u-boot upstream at some point...

plus ›