#asahi-dev · 2026-08-12 16:10
yukahas anyone seen this? "ieee80211 phy0: brcmf_inform_bss: BSS info version 116 unsupported"#asahi-dev · 2026-08-11 16:32
yukadice roll#asahi-dev · 2026-08-11 12:02
yuka*later#asahi-dev · 2026-08-11 12:02
yukaI'll send that idreg-override.c thing out late#asahi-dev · 2026-08-11 12:02
yukaconfirmed linux-next + https://termbin.com/8dyl + "idle=nop arm64.nowfxt" can boot on M4 smp#asahi-dev · 2026-08-11 11:50
yukaI'm not in a hurry, waiting for 7.5 is a fine workaround for me#asahi-dev · 2026-08-11 11:22
yukayes, you'll need to add t603x to require 3 power-domains#asahi-dev · 2026-08-11 11:21
yukaI guess apple,t6020-nvme-ans2 and apple,t6000-nvme-ans2 are only explicitly listed because one of them is in the old base compatible#asahi-dev · 2026-08-11 11:21
yukaor not?#asahi-dev · 2026-08-11 11:21
yukawhich you will add t603x to#asahi-dev · 2026-08-11 11:20
yukait changes the indentation of the if: block#asahi-dev · 2026-08-11 11:20
yukahttps://cyberchaos.dev/yuka/linux/-/commit/cfdd4d6bf48a9a2d0e2c02f992fa50c584145715#asahi-dev · 2026-08-11 11:20
yukathere will be conflicts#asahi-dev · 2026-08-11 11:15
yukasven: ah yes, then I'd have to wait until 7.5 to fully work around conflicts#asahi-dev · 2026-08-11 11:14
yukathe main point being that my change has actual driver changes, so it with its accompanying dt-bindings change would go through the nvme tree#asahi-dev · 2026-08-11 11:13
yukadepend on change means the nvme tree would pick up the M3 dt-bindings change too?#asahi-dev · 2026-08-11 10:59
yukajannau: if you send the M3 nvme stuff through the soc tree, then I can only do M4 nvme in 7.4, or it will result in merge conflicts. any idea how to deal with this?#asahi-dev · 2026-08-11 10:43
yukamaz: I prepared a new version of PCIe power control, but it's gotten very messy and I don't feel too good about sending it out like this. I can send it if you really want, but I don't expect it to go anywhere.#asahi-dev · 2026-08-11 10:38
yukajannau: sounds good, then I'll wait for that#asahi-dev · 2026-08-11 10:35
yukajannau: do you want to send that out eventually?#asahi-dev · 2026-08-11 10:34
yukait seems pretty trivial now with the 15.x fw figured out#asahi-dev · 2026-08-11 10:32
yukawhat is the plan for M3* nvme (new compatibles and device tree nodes) upstreaming?#asahi-dev · 2026-08-08 08:04
yukaWhich ioreg?#asahi-dev · 2026-08-08 08:04
yukaSo the question is if I want to try the wifi fw from 26.6, will it require additional changes and do I need to trace it#asahi-dev · 2026-08-08 08:03
yukaYou mean in the new devices?#asahi-dev · 2026-08-08 08:02
yukaI'm having issues connecting to a wifi network, so before I debug it I thought I would check macOS and sure enough macOS 14.8.3 also has issues connecting to it#asahi-dev · 2026-08-08 07:49
yukaI was going to try some newer macOS on my M3#asahi-dev · 2026-08-08 07:48
yukaOk, thanks for confirming#asahi-dev · 2026-08-08 07:47
yukasven: the code you used for macOS 27 on m1n1 hypervisor is the one someone wrote for some thesis and is not public yet?#asahi-dev · 2026-08-07 17:57
yukamakes sense, thanks for explaining#asahi-dev · 2026-08-07 16:56
yukawhat did I miss regarding the merge window? Wasn't it another week? But yeah, still missing some dt-bindings acks#asahi-dev · 2026-08-06 17:28
yukaidle=nop was merged into for-next/errata, but it only affects the idle loop and not the wfit in the delay function, so I will follow up with something for that soon#asahi-dev · 2026-08-06 11:11
yukaIt will#asahi-dev · 2026-08-06 11:10
yukaand then I have two additional small changes on m1n1 and kernel side ready for M4+ nvme support#asahi-dev · 2026-08-06 11:09
yukathey were complete bogus#asahi-dev · 2026-08-06 11:09
yukaI will close my m1n1 PR now since I assume you will mirror the kernel change to m1n1 correctly.#asahi-dev · 2026-08-06 11:08
yukaprobably that gist which claimed they both moved#asahi-dev · 2026-08-06 11:07
yukaAh nice, I just checked and LINEAR_SQ_CTRL actually can be written on M4 / A18 Pro (and probably also M3 with new FW). I do not know why I conflated the LINEAR_SQ_CTRL and UNKNOWN_CTRL.#asahi-dev · 2026-08-06 06:35
yukaThis is not helpful#asahi-dev · 2026-08-05 19:23
yukaThen it does not need always-on#asahi-dev · 2026-08-05 19:23
yukaps_pcie_sys_st is supposed to be in the nvme node's power-domains#asahi-dev · 2026-08-05 19:22
yukarhlvista[m] the nvme node's power-domains added in commit "M5 NVME nodes Device tree" also differ from the one on my branch#asahi-dev · 2026-08-05 18:53
yukaI see you marked ps_pcie_sys_st as always on#asahi-dev · 2026-08-05 18:51
yukarhlvista[m] the way your commits are structured makes it very difficult to see which additional changes are needed on top of my tree#asahi-dev · 2026-08-05 18:06
yukalike the pcie ones?#asahi-dev · 2026-08-02 21:23
yukaYep#asahi-dev · 2026-08-02 21:21
yuka*tipd#asahi-dev · 2026-08-02 21:21
yukaYes totally the linux driver then just sets the tips power state later on#asahi-dev · 2026-08-02 21:21
yukaSo yeah dockchannel uart in a way#asahi-dev · 2026-08-02 21:21
yukatipd/spmi in m1n1 is required for USE_DEBUGUSB on m3