#asahi-dev · 2022-07-20 09:21
kettenismeanwhile I did confirm that the crash is indeed caused by an SError during the SetVirtualAddressMap() EFI runtime call that happens early on in the OpenBSD kernel#asahi-dev · 2022-07-20 09:17
kettenisbut so far we've managed to avoid that#asahi-dev · 2022-07-20 09:17
ketteniswhen I named the u-boot defconfig file, I anticipated the need for a separate config at some point for a future SoC generation#asahi-dev · 2022-07-19 22:44
kettenispushed device tree updates to the u-boot asahi branch#asahi-dev · 2022-07-19 20:57
kettenisI *hate* Heisenbugs#asahi-dev · 2022-07-19 17:51
kettenisneed to dig into this#asahi-dev · 2022-07-19 17:49
kettenisjannau: unsure, kernel dies very early on, maybe at the point of the first EFI runtime services call#asahi-dev · 2022-07-19 11:31
kettenishmm, boot.bin from uefi-only-20220717-1.zip breaks OpenBSD :(#asahi-dev · 2022-07-19 09:51
kettenisthe m1n1 code suggests that is indeed the case#asahi-dev · 2022-07-19 09:48
kettenisthat code doesn't make a lot of sense to me; was the intention to somehow use the variable i as an index into a set of stream enable registers on t8110?#asahi-dev · 2022-07-19 09:47
kettenishttps://github.com/AsahiLinux/linux/blob/7f6b104b64539e3047b474c13dba17bdf5446d47/drivers/iommu/apple-dart.c#L470#asahi-dev · 2022-07-18 15:20
kettenisthe device implements function level reset?#asahi-dev · 2022-07-15 18:59
ketteniswell, there still is the pcie su3 controller issue ;)#asahi-dev · 2022-07-15 17:58
kettenisjannau: ah cool, I didn't realize that#asahi-dev · 2022-07-15 17:43
kettenisI don't think I'll implement support for bluetooth in u-boot ;)#asahi-dev · 2022-07-15 10:24
kettenisanyway, exposing this information in a GUI isn't something I particularly care about#asahi-dev · 2022-07-15 10:23
kettenisthink it is the same way on other BSDs#asahi-dev · 2022-07-15 10:23
kettenisfdt is exposed through ioctls in OpenBSD (the same way as the OpenFirmware device tree is exposed on sparc64 and the old Apple powerpc machines)#asahi-dev · 2022-07-15 10:16
kettenisfair enough; just wanted to make sure you were aware of this in case you wanted to go further down that path#asahi-dev · 2022-07-15 09:49
kettenissomebody who cares could probably implement the 64-bit interface for that and populate the tables with interesting stuff#asahi-dev · 2022-07-15 09:48
kettenisthat's why I disabled SMBIOS support in u-boot#asahi-dev · 2022-07-15 09:48
kettenisthis doesn't work on Apple hardware since there is no memory at addresses below 4G to store the SMBIOS tables#asahi-dev · 2022-07-15 09:47
ketteniswhere pointers to the relevant information blobs are 32-bit#asahi-dev · 2022-07-15 09:46
kettenisu-boot supports this, but only implements the traditional "32-bit" interface#asahi-dev · 2022-07-15 09:45
kettenischadmed: the consensus in the ARM/UEFI universe seems to be to provide information about firmware/hardware through SMBIOS#asahi-dev · 2022-07-12 12:51
kettenislooks good; thanks#asahi-dev · 2022-07-12 11:45
kettenishttps://gist.github.com/kettenis/ec17f79d50e36bddaa15cdb8c36ca5b9#asahi-dev · 2022-07-12 11:26
kettenismarcan: is there time to get some U-Boot/OpenBSD stuff in that report?#asahi-dev · 2022-07-12 11:25
kettenissomeone should do a cheap NVMe-oF enclosure ;)#asahi-dev · 2022-07-12 10:27
kettenisno, I think letting m1n1 do this is fine#asahi-dev · 2022-07-12 10:24
kettenisu-boot will still set the property if a serial# environment variable exists#asahi-dev · 2022-07-12 10:24
kettenischadmed: looks like that got list somehow with the move of some stuff into the device tree specification and dt-schema#asahi-dev · 2022-07-12 09:13
kettenismarcan, jannau: should we still reset ANS/ANS2 from U-Boot as well even if m1n1 is now fixed?#asahi-dev · 2022-07-12 08:23
kettenisbut I'm not sure I'm the one to improve it...#asahi-dev · 2022-07-12 08:22
kettenismarcan: the USB code in u-boot isn't great#asahi-dev · 2022-07-11 13:55
kettenisu-boot 2022.07 was released so I rebased the u-boot asahi branch once more to pick up the Makefile changes that drops -rc6#asahi-dev · 2022-07-11 13:05
kettenisah, ok#asahi-dev · 2022-07-11 12:58
kettenissince the boot picker uses the keyboard/touchpad, there must be a way to reinitialize#asahi-dev · 2022-07-11 12:37
kettenisso APPLE_RTKIT_HELPER is just a driver that brings up RTKit and processes log messages on the "standard" endpoints?#asahi-dev · 2022-07-11 12:36
kettenisthat reminds me; I did look at the mtp/dockchannel diff over the weekend#asahi-dev · 2022-07-11 12:14
kettenis2. jannau's patch to pick the paired ESP#asahi-dev · 2022-07-11 12:14
kettenis1. the nvme shutdown patch (you really need to send that one upstream)#asahi-dev · 2022-07-11 12:14
kettenisnot on that branch are:#asahi-dev · 2022-07-11 12:13
kettenisthe branch is up-to-date with u-boot master, which means it is pretty much u-boot 2022.07 which is supposed to be released today#asahi-dev · 2022-07-11 12:13
kettenismarcan: pushed the m2 patches onto the u-boot asahi branch in case that helps#asahi-dev · 2022-07-11 11:57
kettenisI should put the m2 patches on the u-boot asahi branch now that I reviewed them ;)#asahi-dev · 2022-07-09 14:40
kettenisthey'll probably switch to RISC-V before they run out just to save on licensing fees ;)#asahi-dev · 2022-07-09 14:37
kettenisrunning out of IDs is Apple's problem, not yours ;)#asahi-dev · 2022-07-09 14:37
kettenisheh#asahi-dev · 2022-07-09 14:37
kettenishe