#asahi-dev · 2022-02-06 15:15
kettenisdoes nvme work?#asahi-dev · 2022-02-06 15:10
kettenism1n1 should do that now, but maybe it is subtly broken#asahi-dev · 2022-02-06 15:10
kettenismaybe the u-boot,dm-pre-reloc properties aren't added?#asahi-dev · 2022-02-06 15:09
kettenisah, that means the debug uart does work!#asahi-dev · 2022-02-06 15:04
kettenis?#asahi-dev · 2022-02-06 15:04
kettenisdid you adjust the various CONFIG_DEBUG_UART options in apple_m1_defconfig/#asahi-dev · 2022-02-06 15:03
kettenis:(#asahi-dev · 2022-02-06 15:03
kettenisI think with a recent m1n1 a Linux device tree may also just work#asahi-dev · 2022-02-06 15:01
kettenisnvme might actually work; usb will need some changes to the dart driver#asahi-dev · 2022-02-06 14:59
kettenisjannau: if you can test that I'll add your tested-by tag#asahi-dev · 2022-02-06 14:56
kettenisok, https://github.com/kettenis/u-boot/tree/t6000 is now rebased on upstream master and has nvme and spi-hid support included on top of that#asahi-dev · 2022-02-06 14:00
kettenisand see whether I can push something that includes nvme and keyboard stuff#asahi-dev · 2022-02-06 14:00
kettenisok, let me cook a diff for that#asahi-dev · 2022-02-06 13:58
kettenisshould probably also round down the framebuffer base and do the same for the normal DRAM mappings#asahi-dev · 2022-02-06 13:56
kettenisjannau: I suppose it also works when you use SZ_4K instead SZ_16K?#asahi-dev · 2022-02-06 13:56
kettenistrue#asahi-dev · 2022-02-06 13:54
kettenisI was more thinking of whether it is safe to map memory that hasn't been reserved in the device tree by m1n1#asahi-dev · 2022-02-06 13:52
kettenisalthough, maybe that is something that should be fixed in m1n1?#asahi-dev · 2022-02-06 13:51
kettenisoh, so close...#asahi-dev · 2022-02-06 13:49
kettenisso something is still wrong with my logic to switch the memory map based on the SoC compatible?#asahi-dev · 2022-02-06 13:47
kettenisbuilt a version of the asahi-installer that installed m1n1 with a u-boot payload and had two OpenBSD developers install OpenBSD that way#asahi-dev · 2022-02-06 13:46
kettenisI've started polishing things on my end already#asahi-dev · 2022-02-06 13:43
kettenishaving anything change in those areas that make drivers fail would be painful for upgrading#asahi-dev · 2022-02-06 13:43
kettenisI think the important bits are nvme, spi, spi-hid and smc#asahi-dev · 2022-02-06 13:12
kettenismarcan: I noticed a different reg address in the device tree, but it seems the smc registers are visible at both 0x23e400000 and 0x23e050000#asahi-dev · 2022-02-06 13:11
kettenisyou could make it a soft-fail in the pcie driver#asahi-dev · 2022-02-06 13:03
kettenis(and the cache; you can't enable the cache without enabling the MMU on ARMv8)#asahi-dev · 2022-02-06 13:01
kettenisdcache_enable() is what enables the MMU#asahi-dev · 2022-02-06 12:59
kettenisso I think it is all fine#asahi-dev · 2022-02-06 12:59
kettenis(the pcie driver needs redoing before I can submit it upstream)#asahi-dev · 2022-02-06 12:58
kettenisso I don't think it is a big loss#asahi-dev · 2022-02-06 12:58
kettenisalthough my plan is to switch over to a different branch that doesn't have the smc and pcie drivers anyway#asahi-dev · 2022-02-06 12:57
kettenisI can obviously fix that#asahi-dev · 2022-02-06 12:57
kettenisthe pcie driver can't find the pwren gpio and therefore returns an error which makes the driver probe fail#asahi-dev · 2022-02-06 12:54
kettenisso I think that is acceptable if you really think it is better to do it this way#asahi-dev · 2022-02-06 12:54
kettenismarcan: making the gpio a subnode of the smc node breaks pcie in u-boot, but I can still boot from nvme#asahi-dev · 2022-02-06 12:12
kettenisyeah, hacking on new stuff is much more fun than dealing with the linux review bureacracy ;#asahi-dev · 2022-02-06 11:59
kettenisbtw, is there any progress on upstreaming the various spi bits?#asahi-dev · 2022-02-06 11:52
kettenisand "disp_dart: iommu@231304000" isn't locked but resetting it breaks the framebuffer?#asahi-dev · 2022-02-06 11:52
kettenisso do I understand correctly that "dcp_dart: iommu@23130c000" is the locked one?#asahi-dev · 2022-02-06 11:50
kettenissuch that adding the new nodes doesn't break stuff#asahi-dev · 2022-02-06 11:50
kettenislooking at your stuff right now because I may need to take some pre-emptive action in U-Boot and OpenBSD for this#asahi-dev · 2022-02-06 11:49
kettenisI think that fear is justified#asahi-dev · 2022-02-06 00:54
kettenisbut nice work#asahi-dev · 2022-02-06 00:54
kettenisI'll take a closer look at the code tomorrow#asahi-dev · 2022-02-06 00:54
kettenisanyway, zzz time here#asahi-dev · 2022-02-06 00:52
kettenisso maybe take another look at my u-boot device trees#asahi-dev · 2022-02-06 00:52
kettenisjudging from the ADTs#asahi-dev · 2022-02-06 00:51
kettenisbut the rtc bit seems to work in the same way#asahi-dev · 2022-02-06 00:51
kettenisthose have a differently named pmu