#asahi-dev · 2026-09-18 07:44
svenyay, tb dt-bindings have been reviewed 🎉#asahi-dev · 2026-09-15 10:14
svenlol#asahi-dev · 2026-09-10 10:46
sventhough i don't think i looked for an MMIO reg that contains the protocol adapter index#asahi-dev · 2026-09-10 10:46
svenhttps://f.svpe.de/d09f1fa677f4a843e9c9eb42840f8992861131d2474ec1c200760a5c7fd9c13f_dpin.h found it#asahi-dev · 2026-09-10 09:04
svenyeah, I remember it being very simple #asahi-dev · 2026-09-10 07:49
svenjust noticed as well :D#asahi-dev · 2026-09-10 07:49
svenlol#asahi-dev · 2026-09-10 07:48
sveni knew it had to be somewhere#asahi-dev · 2026-09-10 07:48
svenhah, thanks!#asahi-dev · 2026-09-10 07:27
svenit might’ve been paste because I looked at old branches and didn’t see it either #asahi-dev · 2026-09-10 07:26
svenno idea unfortunately #asahi-dev · 2026-09-10 07:22
svenmainly curious if the protocol adapter id is in some MMIO or if we need it in the dt instead #asahi-dev · 2026-09-10 07:21
svenjannau: do you still have the dpin code I wrote ages ago for dp tunneling somewhere by chance? I don’t remember if it worked but I do remember reverse engineering it #asahi-dev · 2026-09-10 06:00
svenhuh, I thought it was hv.framework only #asahi-dev · 2026-09-10 06:00
sventcg?#asahi-dev · 2026-09-09 19:40
svenwonder if they actually program them differently now though#asahi-dev · 2026-09-09 19:39
svensounds like confirmation for the "one stream for each rx/tx ring" theory we had#asahi-dev · 2026-09-09 19:28
sveniirc it boots to a single user shell and only runs under apple's hv.framework so you can't pass through hardware#asahi-dev · 2026-09-09 19:27
svenit's great work but probably not all that useful for hw tracing#asahi-dev · 2026-09-09 19:27
svenhe's been talking about this for a while in an iphone hacker discord (hack different)#asahi-dev · 2026-09-09 19:26
svensee https://github.com/jprx/qemu-sptm#is-this-ai-slop#asahi-dev · 2026-09-09 10:02
sveni've said this before, but at ccc on my m1 dcp+llvmpipe was fine. took me more than a day to notice i didn't have the gpu driver#asahi-dev · 2026-09-08 17:52
svenmyria: ^--#asahi-dev · 2026-09-08 17:52
svenhttps://lore.kernel.org/asahi/dbf17fa6-1af6-467b-8b3d-dca8476dc785@marcan.st/#asahi-dev · 2026-09-07 13:13
svenI think it’s just 7.3-rc1#asahi-dev · 2026-09-06 18:38
svenhttps://lore.kernel.org/asahi/20260906-b4-apple-soc-tbt-v2-0-1f80085f93fb@kernel.org/#asahi-dev · 2026-09-06 18:22
svenfixed the rest#asahi-dev · 2026-09-06 18:17
svens/the/end/ for the port unlock stuff and then it should make sense#asahi-dev · 2026-09-06 17:26
svenguess we need to do the immutable branch dance then for the first two typec commits#asahi-dev · 2026-09-06 17:26
svenso that means the tipd commit definitely have to go through the usb tree#asahi-dev · 2026-09-06 17:22
svenugh, true#asahi-dev · 2026-09-06 17:21
sveni don't mind feedback like that if it's combined with other useful feedback ;)#asahi-dev · 2026-09-06 17:11
svenhttps://github.com/AsahiLinux/linux/tree/b4/apple-soc-tbt probably read to go out like this#asahi-dev · 2026-09-06 17:09
sven 37 files changed, 4529 insertions(+), 132 deletions(-) 🫠#asahi-dev · 2026-09-06 17:08
svenhah :D#asahi-dev · 2026-09-06 15:40
sven22 patches for tbt v2 now :/#asahi-dev · 2026-09-06 14:52
svendunno how the cq even works then because we should definitely hit its size#asahi-dev · 2026-09-06 14:51
svenyeah, the submission "queue" is a vector/array here#asahi-dev · 2026-09-06 14:27
sventhat's the whole idea behind doing stuff like "M3-nvme", "M1-nvme". that essentially says "This device works with M1-nvme already but may have additional, optional features with M3-nvme"#asahi-dev · 2026-09-06 14:27
svenand you can just match on "M3-nvme" and only enabled 252 queue size at a later point then#asahi-dev · 2026-09-06 14:26
sventhe compatible will be something like "M3-nvme", "M1-nvme" or whatever anyway#asahi-dev · 2026-09-06 14:24
svenmight increase performance but that can always be a follow-up#asahi-dev · 2026-09-06 14:14
svenfeel free, the names are all made up anyway and if there's a name that fits better we should do it#asahi-dev · 2026-09-06 13:30
svensince that's an actual queue and whatever is running on the SoC and ANS have to agree about that value#asahi-dev · 2026-09-06 13:30
sveni would expect it to refer to the CQ though#asahi-dev · 2026-09-06 13:29
svenor i just typoed it once and copied that from m1n1 to linux then 😬#asahi-dev · 2026-09-06 13:24
sventhe io submission "queue" isn't really a queue, more like a vector, isn't it? so you could just get away with using less entries than available#asahi-dev · 2026-09-05 14:43
svencan’t hurt to fix the reasonable issues#asahi-dev · 2026-09-05 14:27
svenno strong opinion