#asahi-dev · 2025-02-23 04:44
nickchanCSSELR_EL1 and CCSIDR_EL1 but these registers are surely being weird on some SoCs they show that t8015 pcore having a whooping 16 bytes of L2 cache#asahi-dev · 2025-02-23 04:43
nickchanlinux uses some system registers to deal wtih architecturally visible cache sizes which may be different from what's stated in the dt#asahi-dev · 2025-02-23 04:40
nickchansven: yeah the dwi dt nodes is new even though the driver is on the list for months, and for the cache nodes, as far as I know it's only used to show information in sysfs#asahi-dev · 2025-02-22 23:54
leiofwiw, I spotted https://github.com/AsahiLinux/linux/blob/asahi/drivers/soc/apple/Kconfig#L89-L101 saying "persent" instead of "present"#asahi-dev · 2025-02-22 23:31
chadmedgiven that isp such an enormous patch set it might be worth waiting until the locked dart stuff lands in a finalised form. hopefully it won't be too much longer#asahi-dev · 2025-02-22 23:11
chaos_princessalyssa jannau: wdyt?#asahi-dev · 2025-02-22 23:10
chaos_princessim kinda not sure what to do w/ isp series v2. i rewrote the thing to use iommu-dma instead of drm mm, but that needs the apple,dma-range support, and it does not apply cleanly w/o the locked dart series that is currently in review#asahi-dev · 2025-02-22 20:57
mischa85 "name": "ATC0_USB_AON"#asahi-dev · 2025-02-22 20:57
mischa85 "id2": 289,#asahi-dev · 2025-02-22 20:57
mischa85 "unk2": 50331848,#asahi-dev · 2025-02-22 20:57
mischa85 "unk1": 512,#asahi-dev · 2025-02-22 20:57
mischa85 "index": 0,#asahi-dev · 2025-02-22 20:57
mischa85and t6041:#asahi-dev · 2025-02-22 20:57
mischa85 "id2": 166,#asahi-dev · 2025-02-22 20:57
mischa85 "name": "ATC0_USB_AON"#asahi-dev · 2025-02-22 20:57
mischa85 "map": 7,#asahi-dev · 2025-02-22 20:57
mischa85about the missing ps-regs on t6041: could it be the offsets moved to the devices node? for example -> t6031: "index": 26,#asahi-dev · 2025-02-22 20:19
nicolas17https://alx.sh/fs still redirects to the old wiki#asahi-dev · 2025-02-22 20:18
nicolas17should the wiki pages be replaced with links to the new docs site?#asahi-dev · 2025-02-22 20:07
jannauthe docs site is based on the wiki git repo: https://github.com/AsahiLinux/docs/graphs/contributors#asahi-dev · 2025-02-22 20:02
nicolas17were all wiki pages moved to the new mkdocs site?#asahi-dev · 2025-02-22 19:53
nicolas17was that supposed to be "iBoot1, iBoot2"?#asahi-dev · 2025-02-22 19:53
nicolas17"iBoot: Apple's bootloader. Can refer to iBoot1, iBoot1, or any of iBSS, iBEC, or even the SecureROM itself"#asahi-dev · 2025-02-22 17:16
svenhttps://lore.kernel.org/asahi/20250220-caches-v1-0-2c7011097768@gmail.com/ was supposed to be the second one#asahi-dev · 2025-02-22 17:16
leiothey were also the same link twice (accidentally?) btw#asahi-dev · 2025-02-22 17:09
svenoh, they are only two days old. i somehow thought there were already on the list a while longer#asahi-dev · 2025-02-22 17:09
svenit would also be helpful to reviewed-bys or tested-bys on https://lore.kernel.org/asahi/20250220-dwi-dt-v1-0-d742431c7373@gmail.com/ and https://lore.kernel.org/asahi/20250220-dwi-dt-v1-0-d742431c7373@gmail.com/ before I can merge them#asahi-dev · 2025-02-22 17:02
sveni'll let them sit on the ML for a week or two and wait for some reviews and then just pull them into asahi-soc/* for the next PR#asahi-dev · 2025-02-22 17:01
jannauthanks#asahi-dev · 2025-02-22 17:01
svenokay, send out those rtkit fixes#asahi-dev · 2025-02-22 16:53
chaos_princesssven: in my case it is for the sep tracer#asahi-dev · 2025-02-22 16:52
jannausven: ack. it was simplest way to it working after rebasing on the the renesas change#asahi-dev · 2025-02-22 16:51
svennot trying to stop you, just confused why we need all that complexity inside m1n1 :)#asahi-dev · 2025-02-22 16:50
fl0_idsven I don't know enough details. I just thought it could be a nice project to work on either for me as I only have an m3 or for others who want to do sth but do not have any hardware#asahi-dev · 2025-02-22 16:50
jannauI think the idea for requiring sep/xart in m1n1 was authenticated 2nd stage chainloading but I would defer that until sep is working#asahi-dev · 2025-02-22 16:49
svenjannau: ack, so that commit likely needs a do-over now that the renesas part is a proper module#asahi-dev · 2025-02-22 16:48
sven*on#asahi-dev · 2025-02-22 16:48
svenxart/gigalocker is continuous in the disk fwiw, so all we need to do is find its offset#asahi-dev · 2025-02-22 16:48
svenright, but why do we need that in m1n1?#asahi-dev · 2025-02-22 16:47
fl0_idsven afaik, as chaos_princess said, not full apfs, but just find some file, write some stuff to it. I'm still learning about it, but as they mentioned xart/gigalocker, it involves managing access to the xart (extended anti replay technology), the storage part of the SEP#asahi-dev · 2025-02-22 16:46
jannausven: see v6.11..7fc439afc9282 for the orginal changes#asahi-dev · 2025-02-22 16:46
svenjannau: ok, so that xhci firmware handling needs looking at before we try to submit it#asahi-dev · 2025-02-22 16:45
leiojannau: thanks, I've never had that as we apparently had vulkan p.use.masked still accidentally on system gtk and such, so best if you check it out :)#asahi-dev · 2025-02-22 16:44
jannauxhci-pci.c was renamed due to the previous handling for the renesas firmware loading. That was changed in by 25f51b76f90f ("xhci-pci: Make xhci-pci-renesas a proper modular driver") in 6.12#asahi-dev · 2025-02-22 16:41
jannauI'm currently working on the gpu branch get closer to upstream / patches on rust-for-linux#asahi-dev · 2025-02-22 16:40
jannausven: if you have time please do. "soc: apple: rtkit: Add devm_apple_rtkit_free()" is not needed#asahi-dev · 2025-02-22 16:37
svenalso, does anyone know why xhci-pci.c was renamed to xhci-pci-core.c in 150-xhci-firmware? #asahi-dev · 2025-02-22 16:36
svenok, so we can drop 1 or 2 and could submit the rest? wanna do that or should I take a look?#asahi-dev · 2025-02-22 16:36
jannauleio: I'll take a look