#asahi-dev · 2022-04-11 05:42
kevans91hmm, am I reading these pci nodes correctly? https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/arm64/boot/dts/apple/t8103.dtsi#n446 -> my really limited understanding is that reg should encode the b:d:f (which I think we (FreeBSD) are doing wrong), but these reg values also don't seem to line up with the iommu-map if I'm reading that right#asahi-dev · 2022-04-11 02:46
kevans91big hairy lock?#asahi-dev · 2022-04-09 19:17
kevans91kettenis: ah, makes sense- I'll resist the urge to rip out our equivalent, then =-)#asahi-dev · 2022-04-09 16:35
kevans91kettenis: it looks like you can kill sc_cpuremap over in aplintc, it's write-only as of fast ipi support #asahi-dev · 2022-04-09 14:50
kevans91kettenis: no problem, thanks for the good reference drivers :-)#asahi-dev · 2022-04-09 05:01
kevans91kettenis: I suspect this isn't quite right: https://github.com/openbsd/src/blob/master/sys/arch/arm64/dev/aplintc.c#L37 (noting that APL_IPI_LOCAL_RR_EL1 and APL_IPI_GLOBAL_RR_EL1 probably aren't the same)#asahi-dev · 2022-04-06 15:43
kevans91(broken in the sense that it didn't work by default because tty_dev is optional)#asahi-dev · 2022-04-06 15:32
kevans91this seems to fix the u-boot support in the proxyclient: https://github.com/AsahiLinux/m1n1/pull/189#asahi-dev · 2022-04-06 15:25
kevans91jannau: perfect, thanks!#asahi-dev · 2022-04-06 03:25
kevans91er, presumably that's the kernel payload#asahi-dev · 2022-04-06 03:25
kevans91anyone have a built Image.gz they'd be willing to toss somewhere? I have an m1n1 change to linux.py that I'd like to test, but not really any linux build setup (and no real use for setting one up otherwise)#asahi-dev · 2022-04-05 16:12
kevans91it'd be most convenient if you could drive the proxyclient on the guest, but that seems nontrivial#asahi-dev · 2022-04-05 16:08
kevans91in my experience, at least#asahi-dev · 2022-04-05 16:08
kevans91it needs to hit the proxy stuff after not detecting an executable payload#asahi-dev · 2022-04-03 19:02
kevans91the downside is we don't get any of our loader scripts, but having to manually 'load' then 'boot' is still a huge improvement in workflow#asahi-dev · 2022-04-03 19:01
kevans91on the plus side, my freebsd abomination works with a somewhat small u-boot change#asahi-dev · 2022-04-03 18:57
kevans91it kind of looks like you can just omit the fdt address from booti and it'll use the modified one, anyways#asahi-dev · 2022-04-03 18:50
kevans91maybe all of the important stuff is already covered by bootargs#asahi-dev · 2022-04-03 18:49
kevans91the prepared dt's not at that address, not sure it's necessarily intended to be passing on the unadulterated input fdt#asahi-dev · 2022-04-03 18:48
kevans91woot, it's alive. the only concerning part of linux.py's uboot stuff might be the use of dtb_addr#asahi-dev · 2022-04-03 18:25
kevans91maybe they'll accept a patch to let it work again#asahi-dev · 2022-04-03 18:25
kevans91it's interesting because there's hints that it used to work. efi_run_image() will explicitly use a memory device path if one's not set, under the assumption that it, e.g., may be loaded directly into memory via JTAG#asahi-dev · 2022-04-03 17:34
kevans91nope, apparently not. :-( it checks the address you feed it against a prior efi_set_bootdev()#asahi-dev · 2022-04-03 17:31
kevans91it also sets the efi bootdev, which is the side effect that I need to make this work#asahi-dev · 2022-04-03 17:30
kevans91my image is currently in a well-known place, but there's no way to point u-boot at said place and fake loaded image information, I guess#asahi-dev · 2022-04-03 17:29
kevans91the alternative seems to be FIT + bootm, which looks like it will recognize the preloaded FIT image and do the right thing#asahi-dev · 2022-04-03 17:26
kevans91or i guess custom u-boot isn't the end of the world#asahi-dev · 2022-04-03 17:26
kevans91ah, that's unfortunate. u-boot really wants you to `load` it before you can `bootefi` it, so I might have to switch loaders just for testing#asahi-dev · 2022-04-03 16:02
kevans91final setup will use standard u-boot efi, mostly just hacking together something for quicker development iteration#asahi-dev · 2022-04-03 16:02
kevans91nice, thanks for considering it upstreamable. :-) I'm trying to do minimal hacking to m1n1/u-boot to minimize the patches I need to send the other developer working on freebsd support (or anyone else I can get to engage)#asahi-dev · 2022-04-03 15:32
kevans91sweet, with a proper freebsd.py I should able t stop preprocessing our kernel to fake cpio#asahi-dev · 2022-04-03 15:07
kevans91i might be able to hack this into submission#asahi-dev · 2022-04-03 15:06
kevans91ooh, that's interesting#asahi-dev · 2022-04-03 15:04
kevans91freebsd's loader#asahi-dev · 2022-04-03 14:57
kevans91so it'd be quite neat if I could further eliminate that step and pass both loader (uefi payload) and kernel (elf) along#asahi-dev · 2022-04-03 14:57
kevans91i'm hacking our loader to take an "initramfs", chop off the cpio magic then execute that, but there's problems there and I still have to pass a flash drive back and forth to hack on our loader#asahi-dev · 2022-04-03 14:55
kevans91no, I want m1n1 to execute u-boot and pass it a UEFI payload to execute as the next stage#asahi-dev · 2022-04-03 14:53
kevans91anyone know off-hand if u-boot can be convinced to run a UEFI m1n1-attached payload? I've already repurposed 'initramfs' for passing our kernel#asahi-dev · 2022-03-28 03:05
kevans91nods#asahi-dev · 2022-03-28 03:04
kevans91to be fair, it does seem a pretty natural expression when ~ is viewed as 'bitwise NOT'#asahi-dev · 2022-03-28 03:02
kevans91hey now, lua isn't that bad#asahi-dev · 2022-03-25 04:23
kevans91I had glanced through the wiki and didn't quite see anything describing what you'd already done there#asahi-dev · 2022-03-25 04:21
kevans91a bit stalled until my second m1 comes in to do any of that#asahi-dev · 2022-03-25 04:21
kevans91going to see if i can hit it with a debug-break over uart first, though#asahi-dev · 2022-03-25 04:21
kevans91the underlying bits (you called it your protocol) landed, though? I might be interested in picking that up to figure out what I've got wrong#asahi-dev · 2022-03-25 03:35
kevans91marcan: yo, on 2021-05-04~ish, you floated the idea of implementing gdbstub in python. did you end up starting on that at all?#asahi-dev · 2022-03-21 23:18
kevans91oh, you can do gdb over uart#asahi-dev · 2022-03-21 23:16
kevans91hmm, not sure off-hand how communication to that would look#asahi-dev · 2022-03-21 23:15
kevans91contemplates the merits of gluing gdbstub or similar into m1n1