#
✕ nick:kettenis
#asahi-dev · 2023-06-07 21:41
kettenisthe scripting stuff is a bit of a hack and possible going away in the future
#asahi-dev · 2023-06-07 21:32
kettenisturns out it isn't
#asahi-dev · 2023-06-07 21:31
kettenisbut I thought the current state would be a good halfway house
#asahi-dev · 2023-06-07 21:31
kettenisEnabling the EFI boot menu would probably solve the issue since then you can simply pick the USB drive as a boot option from the menu
#asahi-dev · 2023-06-07 21:28
kettenisbut I didn't think about testing the case where the designated nvme partition does have a bootaa64.efi
#asahi-dev · 2023-06-07 21:27
kettenisI tested the case where nvme doesn't have a bootaa64.efi and it boots from usb just fine then
#asahi-dev · 2023-06-07 21:26
kettenisThe diff that breaks "run bootcmd_usb0" is probably https://github.com/AsahiLinux/u-boot/commit/b1488ddcc7261845e404e915c4db06ec0ee59ecd
#asahi-dev · 2023-06-06 07:36
kettenis(yes I should actually set up a Linux machine)
#asahi-dev · 2023-06-06 07:35
kettenisI looked at it a bit, but the lack of rustup support for OpenBSD makes it a bit hard for me to work on the installer
#asahi-dev · 2023-06-06 07:31
kettenisnot had it fail yet on the m2 pro mini
#asahi-dev · 2023-06-06 07:31
kettenishah
#asahi-dev · 2023-06-06 07:29
kettenismarcan: glad the PCIe issue was found
#asahi-dev · 2023-06-01 14:21
kettenisfortunately lina's diff applied without issues
#asahi-dev · 2023-06-01 14:20
kettenisj`ey: I didn't realize it either; until I ran into the floating-point literals ;)
#asahi-dev · 2023-06-01 13:03
kettenisbut it needs to go through whatever the upstream dtc repo is to end up in upstream u-boot
#asahi-dev · 2023-06-01 13:03
kettenisthat dtc patch is on the asahi-related branches of the Asahi u-boot repo
#asahi-dev · 2023-05-30 13:37
ketteniserh, how is that possible?
#asahi-dev · 2023-05-30 13:35
kettenisthe m2 pro mini doesn't need a gpio toggle though
#asahi-dev · 2023-05-30 13:35
kettenisso if the studio needs a gpio toggle to enable xhci, u-boot won't even see the xhci
#asahi-dev · 2023-05-30 13:35
kettenisu-boot doesn't know about the smc gpio's yet
#asahi-dev · 2023-05-30 13:34
kettenisyup
#asahi-dev · 2023-05-30 13:34
kettenisit bails and ignores the device
#asahi-dev · 2023-05-30 13:30
kettenisah, ok
#asahi-dev · 2023-05-30 13:27
kettenisso maybe I should hide the start value on OpenBSD as well...
#asahi-dev · 2023-05-30 13:13
kettenisbut there doesn't seem to be an in-tree driver for those laptops
#asahi-dev · 2023-05-30 13:12
kettenisI think some of the "consumer" laptops from lenovo have a similar mechanism to what apple does (as opposed to the mechanism on the thinkpads)
#asahi-dev · 2023-05-30 13:08
kettenisI suspect changing the start value was a later addition and whoever documented it didn't repeat the statement about the rounding?
#asahi-dev · 2023-05-30 13:05
kettenis(this is what I do on OpenBSD)
#asahi-dev · 2023-05-30 13:05
kettenismarcan: I'd say using the stop value would be more intuitive since the idea is to limit the charging to a certain value
#asahi-dev · 2023-05-12 10:13
kettenisI'm not familliar with that code ;)
#asahi-dev · 2023-05-12 10:12
kettenisotherwise the CPU might be interrupted and if some other code executes a WFI instruction the CPU will lose state
#asahi-dev · 2023-05-12 10:11
kettenisI think apple_cpu_deep_wfi() should only be called with interrupts disabled; is that guaranteed?
#asahi-dev · 2023-05-11 07:39
kettenisits temperature rises from ~34 degC to ~48 degC
#asahi-dev · 2023-05-11 07:39
kettenisthe only temperature sensors I've got hooked up on OpenBSD is the "wireless" one
#asahi-dev · 2023-05-11 07:38
kettenisthe fan speed reported by the SMC doesn't really change
#asahi-dev · 2023-05-10 21:20
kettenismaz: so you're seeing it as well?
#asahi-dev · 2023-05-09 14:40
kettenisthe issue has been on my radar since forever but 1) it never makes it to the top of the list of things for me to work on and 2) I don't seem to have USB devices that hit the issue
#asahi-dev · 2023-05-09 14:37
kettenisone issue is that some of the callers expect that it always returns a valid pointer, whereas there is a "return NULL" in that function
#asahi-dev · 2023-05-09 14:37
kettenismarcan: I think all the issues we care about are in xhci_wait_for_event()
#asahi-dev · 2023-05-09 12:15
kettenisultimately those that have the hardware that triggers the issue need to invest time, either by continuing nagging about this on the u-boot lists or fix the issue
#asahi-dev · 2023-05-09 11:58
kettenisespecially if you can reproduce the issue with u-boot on some other board
#asahi-dev · 2023-05-09 11:58
ketteniskeep complaining about this on the u-boot mailing lists ;)
#asahi-dev · 2023-05-09 09:36
kettenisI suppose we could call usb_init() from the board code based on whether we already have a keyboard or not
#asahi-dev · 2023-05-09 09:29
kettenismaybe time for me to finally come up with a diff to address at least the obvious problems with aborted usb transfers in the xhci host code
#asahi-dev · 2023-05-09 09:28
kettenisunfortunately your last attempt to get marex to look into the usb induced crashes doesn't seem to have worked
#asahi-dev · 2023-05-09 09:27
kettenis_jannau_: but problematic for the desktop machines
#asahi-dev · 2023-05-09 08:26
kettenismarcan: if you don't start the usb controllers, you lose the usb keyboard
#asahi-dev · 2023-05-08 14:41
kettenisbtw, I think I have a way now to provide OpenBSD support in the Asahi installer
#asahi-dev · 2023-05-08 14:11
kettenisyou'll need usb mass storage class support as well
#asahi-dev · 2023-05-05 15:54
kettenisbetter; thanks

plus ›