#
✕ nick:marcan
#asahi · 2021-02-07 17:37
marcanmostly just the queue entry size is doubled IIRC
#asahi · 2021-02-07 17:37
marcanit shouldn't be hard
#asahi · 2021-02-07 17:37
marcanwould need to implement the same quirks
#asahi · 2021-02-07 17:37
marcannot sure if there are any other quirks, if there are they should be trivial to work around
#asahi · 2021-02-07 17:37
marcanjust not over PCI
#asahi · 2021-02-07 17:37
marcanwhich linux already has quirk workarounds for
#asahi · 2021-02-07 17:37
marcanit's the standard apple nvme
#asahi · 2021-02-07 17:36
marcansame as every other driver that has both versions, e.g. xHCI
#asahi · 2021-02-07 17:36
marcanthen we can instantiate it properly
#asahi · 2021-02-07 17:36
marcanI'll be adding a platform/DT binding for the NVMe driver once I get to that, shouldn't be hard
#asahi · 2021-02-07 17:35
marcan;-)
#asahi · 2021-02-07 17:35
marcanthen I debugged that ghettohci thing using that
#asahi · 2021-02-07 17:34
marcanas in turning it on and off
#asahi · 2021-02-07 17:34
marcanit bitbanged RS-232 over the sensor bar
#asahi · 2021-02-07 17:34
marcanso the first version was output only
#asahi · 2021-02-07 17:34
marcanno serial
#asahi · 2021-02-07 17:34
marcanbecause there was no good debug port
#asahi · 2021-02-07 17:34
marcanwe used that on the Wii U to get mini (yes the same python RPC protocol and codebase as m1n1) working
#asahi · 2021-02-07 17:33
marcan:D
#asahi · 2021-02-07 17:33
marcanwhich is completely insane
#asahi · 2021-02-07 17:33
marcanunless you're running a corellium kernel, then it's PCIe because they faked a PCIe bus for it
#asahi · 2021-02-07 17:33
marcannot at all
#asahi · 2021-02-07 17:33
marcanmodwizcode: it's not PCIe
#asahi · 2021-02-07 17:33
marcane.g. https://gist.github.com/marcan/4ce7e13321a03d85a73b10840f532892 OHCI + FT232 driver in <600 lines of code
#asahi · 2021-02-07 17:32
marcanmost "complex" devices are only complex to implement all features of
#asahi · 2021-02-07 17:32
marcanI actually want to try writing a minimal NVMe driver, I don't expect it to be that hard?
#asahi · 2021-02-07 17:32
marcanalso plenty of people have two M1s
#asahi · 2021-02-07 17:32
marcannever_released: I'll be fixing that next week
#asahi · 2021-02-07 17:31
marcane.g. m1n1 makes an excellent "boot and throw things at it via serial" loader, for almost literally anything because of the python rpc interface
#asahi · 2021-02-07 17:31
marcanwhich makes it easier to debug the next stage by loading at the previous one
#asahi · 2021-02-07 17:31
marcanwell, kind of, but the thing is it also makes it more modular
#asahi · 2021-02-07 17:30
marcanor something along those lines
#asahi · 2021-02-07 17:30
marcanbut I very much expect the production version of Asahi Linux to end up looking like m1n1 -> u-boot -> grub -> kernel, for end user installs
#asahi · 2021-02-07 17:30
marcanI said I don't want to have u-boot as a hard dependency in order to boot linux *at all* because being tied to piles of bootloader code makes it harder to do back to basics tests and I do like being able to throw kernels at m1n1 and play around in a small codebase
#asahi · 2021-02-07 17:28
marcannot at all
#asahi · 2021-02-07 17:28
marcanmodwizcode: me?
#asahi · 2021-02-07 17:28
marcanbut it should be *capable* of doing that :)
#asahi · 2021-02-07 17:28
marcanfor most users, probably not
#asahi · 2021-02-07 17:27
marcanand it gives us some flexibility while avoiding duplication of responsibilities
#asahi · 2021-02-07 17:26
marcanbut it's not like 3-bootloader chains are rare on arm
#asahi · 2021-02-07 17:26
marcanI mean if you want to throw all the code into SPL, that works too
#asahi · 2021-02-07 17:26
marcansure, but m1n1 already exists
#asahi · 2021-02-07 17:25
marcanI'm wondering if I can e.g. provide all the hardware-related init stuff in m1n1, then let u-boot provide the boot sequence and disk/FS services
#asahi · 2021-02-07 17:24
marcan(how much of UEFI does it do?)
#asahi · 2021-02-07 17:24
marcankettenis: u-boot provides UEFI, right? can it enhance an existing UEFI implementation, i.e. add services?
#asahi · 2021-02-07 17:22
marcan-n
#asahi · 2021-02-07 17:22
marcandoing all confing accesses over PSCI?
#asahi · 2021-02-07 17:21
marcanI'd be happy to have something standard-ish like that that m1n1 can provide, because then we can hide even more nonsense away from linux, like proper full suspend and dealing with initialization functions on resume
#asahi · 2021-02-07 17:20
marcanwell for SoCs with EL3 they already have it anyway
#asahi · 2021-02-07 17:19
marcanit's intended to be standard (on linux at least), AIUI

plus ›