#asahi-dev · 2021-11-01 14:53
svenj_ey: want me to send a patch or do you want to do it?#asahi-dev · 2021-11-01 14:53
sven*gets#asahi-dev · 2021-11-01 14:52
sventhen there's a compile time check in FIELD_PREP that get trigged because -EINVAL > 0b111#asahi-dev · 2021-11-01 14:52
svengcc figures out the range for apple_gpio_irq_set_type#asahi-dev · 2021-11-01 14:52
svenyup#asahi-dev · 2021-11-01 14:51
svenlet me fix that#asahi-dev · 2021-11-01 14:51
svenoh.. i think it's the EINVAL#asahi-dev · 2021-11-01 14:50
svenj_ey: ^-- huh#asahi-dev · 2021-11-01 14:50
sven././include/linux/compiler_types.h:335:38: error: call to ‘__compiletime_assert_289’ declared with attribute error: FIELD_PREP: value too large for the field#asahi-dev · 2021-11-01 14:50
svendrivers/pinctrl/pinctrl-apple-gpio.c: In function ‘apple_gpio_irq_set_type’:#asahi-dev · 2021-11-01 14:37
svenyup #asahi-dev · 2021-11-01 14:36
svenAnd for t6000 >> 14 and then << 10#asahi-dev · 2021-11-01 14:36
svenfor t8020 it will be >> 14 and then << 14#asahi-dev · 2021-11-01 14:35
svenfield_prep takes care of that#asahi-dev · 2021-11-01 14:35
svenDon’t think ao#asahi-dev · 2021-11-01 14:32
svenI guess they changed the PTE shift because they have phys addrs now that are larger than 40 bits now#asahi-dev · 2021-11-01 12:04
svenah, and usb just works once you use the new dart#asahi-dev · 2021-11-01 12:03
svenhrm, so different addr shift in the PTE and mandatory subpage protection#asahi-dev · 2021-11-01 11:59
svenbut i heard you made DART and USB work#asahi-dev · 2021-11-01 11:59
svenaw :(#asahi-dev · 2021-10-31 18:30
svenyeah#asahi-dev · 2021-10-31 18:30
svenwhen macos is running under it it has direct access to the gpu#asahi-dev · 2021-10-31 18:30
sventhe hypervisor is relatively dumb as well. it'll just pass-through any hardware and makes no attempt to protect itself#asahi-dev · 2021-10-30 20:59
svenhaha, posting that question on the official apple forum :D#asahi-dev · 2021-10-30 20:58
svenj_ey: hah! even better#asahi-dev · 2021-10-30 20:58
svenohh... looks like the mailbox driver will also be in linux-next next time the tree is built#asahi-dev · 2021-10-30 20:57
svenso i guess we "just" need to hook this all up to the common hid subsystem instead#asahi-dev · 2021-10-30 20:52
svenoh, those weird keyboard message are just hid?#asahi-dev · 2021-10-30 20:49
svensounds good to me#asahi-dev · 2021-10-30 20:46
svenand if the reset issue is still present on t6000 we can also just do "apple,t6000-dwc3", "apple,t8103-dwc", "snps,dwc3"#asahi-dev · 2021-10-30 20:45
sventrue. with that "snps,dwc3" we don't really need the "apple,dwc3" fallback#asahi-dev · 2021-10-30 20:38
svenfwiw, apple uses "usb-drd,t8103" for the M1 and "usb-drd,t6000", "usb-drd,t8103" for the M1 Pro/Max#asahi-dev · 2021-10-30 20:38
svenmarcan: kettenis: https://lore.kernel.org/linux-usb/96f0536e-6a45-4a44-85a7-1ce530a891cc@www.fastmail.com/T/#mde34c2b270cc36324c272a930ab017b4a69c3169 sounds like it's time to decide on the compatible for apple's dwc3 controller. any opinions on "apple,t8103-dwc3", "apple,dwc3", "snps,dwc3" vs. "apple,t8103-dwc3", snps,dwc3"?#asahi-dev · 2021-10-29 12:10
sveni wonder what those two weird bypass streams are now though#asahi-dev · 2021-10-29 12:08
svenbut at least the locked DART makes sense now. otherwise you could just switch the mapped firmware region to bypass-dapf i guess#asahi-dev · 2021-10-29 12:07
svenmarcan: huh, weird w.r.t DAPF. #asahi-dev · 2021-10-23 08:04
svenalyssa: https://github.com/AsahiLinux/linux/tree/mailbox/v4#asahi-dev · 2021-10-23 07:39
svenit's the first real vacation where i actually leave the country since ~18 months or so :>#asahi-dev · 2021-10-23 07:27
sveni should probably do that v4 this weekend before i go on vacation :D#asahi-dev · 2021-10-23 07:27
svenalyssa: uh. sure#asahi-dev · 2021-10-22 15:24
sven(region protection failure is BIT(7) in the error register)#asahi-dev · 2021-10-22 15:24
svenso i think it should work with those as well#asahi-dev · 2021-10-22 15:24
svenkettenis: i just tested it with a bit of a hack: i only cleared BIT(1) in the PTE and i got the "region protection" exception#asahi-dev · 2021-10-22 15:00
svenstill haven't tested for the pcie darts though#asahi-dev · 2021-10-22 15:00
svenkettenis: i think BIT(31) in DART_PARAMS1 is set when the DART can do region protection#asahi-dev · 2021-10-22 08:30
svenah, yes. but anyway, if the only complaint is about that attach-check-detach quirk in the end i can work around that. either i just introduce something like get_pgsize_for_dev to the iommu driver ops or i just limit the DART driver to 16k for now and kick the issue down the road if apple ever decides to bring back 4k DARTs#asahi-dev · 2021-10-22 08:22
svencould be. i know that the drivers for those are surprisingly complex#asahi-dev · 2021-10-22 08:15
svenimho this has to happen on the iommu level and not the driver level#asahi-dev · 2021-10-22 08:15
svenyeah, agreed