#
✕ nick:sven
#asahi-dev · 2025-05-10 11:40
sventhe NVMe maintainers didn’t like that very much though
#asahi-dev · 2025-05-10 11:40
sveniirc my original version re-used those quirks when I added platform support to pci.c
#asahi-dev · 2025-05-10 11:39
svenyup
#asahi-dev · 2025-05-10 11:18
svenbut assume it's a pcie device ofc
#asahi-dev · 2025-05-10 11:18
sventhe pcie driver has quirks for that other mode I think
#asahi-dev · 2025-05-10 07:52
svenalright, nvmem for atcphy sent out. time to address the SMC review comments now
#asahi-dev · 2025-05-08 17:35
sveni'm just very happy it works right now and don't want to look at it funny and break it all again
#asahi-dev · 2025-05-08 17:35
svenand i'm not in the mood for bike shedding :P
#asahi-dev · 2025-05-08 17:34
sven:P
#asahi-dev · 2025-05-08 17:34
sven<sven> feel free to send a patch
#asahi-dev · 2025-05-08 17:34
sveni'm not touching that magic sequence since it works
#asahi-dev · 2025-05-08 17:33
svenfeel free to send a patch
#asahi-dev · 2025-05-08 17:31
sventhere's a pci hotplug interrupt in the spec but that only works 25% of the time correrctly, 25% with a 5-30 seconds delay and 50% just not at all
#asahi-dev · 2025-05-08 17:30
svenessentially NHI needs to setup a tunnel for each pcie switch/device and i have to match that to the correct thing in the pcie topology to be able to inject hotplug events
#asahi-dev · 2025-05-08 17:29
svenbecause ofc the normal pcie hotplug thing just doesn't work correctly
#asahi-dev · 2025-05-08 17:29
sventhe absolute worst part will be that i have to match up the pcie tree with the tbt tree to make hotplug work properly
#asahi-dev · 2025-05-08 17:29
svenso it's a platform device there with a bunch of quirks and a needs a bunch of out-of-band notifications
#asahi-dev · 2025-05-08 17:29
svenand the spec says NHI is a pci device but apple doesn't care about the spec
#asahi-dev · 2025-05-08 17:28
svendrivers/thunderbolt
#asahi-dev · 2025-05-08 17:28
svenNHI is the thing that sets up the thunderbolt link and what's tunneled inside
#asahi-dev · 2025-05-08 17:27
svenapple calls that acio iirc
#asahi-dev · 2025-05-08 17:27
sventhunderbolt native host interface
#asahi-dev · 2025-05-08 17:27
svenaka dcp and tbt-nhi
#asahi-dev · 2025-05-08 17:27
sven"the rest of the owl"
#asahi-dev · 2025-05-08 17:25
sven*for
#asahi-dev · 2025-05-08 17:25
sventechnically that series also contains 99% of what's required fro dp-altmode and tbt inside atcphy
#asahi-dev · 2025-05-08 17:24
svendefinitely keeping that other description that starts with "As mad as it sounds, " though :D
#asahi-dev · 2025-05-08 17:23
svenyeah, maybe
#asahi-dev · 2025-05-08 17:23
svenlol :D
#asahi-dev · 2025-05-08 17:23
svenyeah, i'm going to be very happy once that's in -next
#asahi-dev · 2025-05-08 17:20
svenstill needs more work though
#asahi-dev · 2025-05-08 17:20
svenhttps://github.com/AsahiLinux/linux/tree/sven/atcphy this also doesn't look horrible anymore
#asahi-dev · 2025-05-08 17:17
svenalyssa: yeah, platform/apple or soc/apple. i think it makes more sense in mfd but i don't have a strong opinion
#asahi-dev · 2025-05-08 17:15
svenlol
#asahi-dev · 2025-05-08 17:15
svenwasn't me, i'm just including that commit :D
#asahi-dev · 2025-05-08 17:11
svenreally depends on if he's happy with stuffing all that smc protocol stuff in there or if we wants us to move that somewhere else
#asahi-dev · 2025-05-08 17:11
svensmc mostly needs input from the mfd maintainer anyway
#asahi-dev · 2025-05-08 17:11
svenlol :D
#asahi-dev · 2025-05-08 17:10
svenuh, that happens automatically
#asahi-dev · 2025-05-08 16:17
svenit's essentially the nvmem core changes to support out fuse configuration + efuses device tree nodes
#asahi-dev · 2025-05-08 16:17
svenhttps://github.com/AsahiLinux/linux/tree/sven/nvmem will probably send that out this weekend since it's kinda independent from the rest of the atcphy mess
#asahi-dev · 2025-05-07 20:42
svenwhen soldering the pico make sure to use pin headers. I didn’t read the README and soldered it down directly and then couldn’t connect two usb cables anymore and had to desolder it again 🙃
#asahi-dev · 2025-05-06 11:48
svenlol
#asahi-dev · 2025-05-06 11:35
sveniirc povik wrote some SIO driver and tracer but I'm not sure if he bothered to figure out how it's connected to the SPI controller
#asahi-dev · 2025-05-05 20:08
sventhanks for checking!
#asahi-dev · 2025-05-04 18:14
svenyeah, and also makes adding new things less scary because it's obvious to see what's going on
#asahi-dev · 2025-05-04 18:12
sven*<
#asahi-dev · 2025-05-04 18:12
svenespecially because with M4 we'd need && part <= T8132_DONAN all over the place
#asahi-dev · 2025-05-04 18:11
svensure, i just wanted to get rid of all these if (part >= WHATEVER && part != SOMETHING_ELSE)
#asahi-dev · 2025-05-04 18:01
svenhttps://github.com/AsahiLinux/m1n1/pull/460 which cleans up chickens.c could also use some testing as stage1 on various devices if anyone is bored

plus ›