#asahi-dev · 2022-03-12 09:43
jannauyes#asahi-dev · 2022-03-12 09:42
jannaumarcan: do you have a todo list for the release? if you're not working on it I can adapt the ESP uuid matching to passing the uuid in chosen#asahi-dev · 2022-03-12 09:39
jannauoh, dcp already blanks the mac book pro display, that was not working for the HDMI display on the mini#asahi-dev · 2022-03-11 23:10
jannaunot sure if we can as easily hide the notch with dcp#asahi-dev · 2022-03-11 23:09
jannauprobably not although a scaled display is nice on on the retina screen#asahi-dev · 2022-03-11 22:27
jannauand a working dcp framebuffer on t6001#asahi-dev · 2022-03-11 21:57
jannaualso the uint one is only used for a single property with a value of 0. the int one is used with properties which should be unsigned#asahi-dev · 2022-03-11 21:53
jannauargh apple, why? ServiceRelay::sr_set_uint_prop() and ServiceRelay::sr_setProperty_int()#asahi-dev · 2022-03-11 20:54
jannauyeah, t600x dcp crash fixed, constantly missed a mapped page for disp0#asahi-dev · 2022-03-11 18:02
jannauit was painless and reasonably fast for filtering devicetree-rebasing for m1n1#asahi-dev · 2022-03-11 18:01
jannauprobably not an option if you want to work on top of devicetree-rebasing#asahi-dev · 2022-03-11 18:00
jannaumarcan: try git-filter-repo https://github.com/newren/git-filter-repo#asahi-dev · 2022-03-11 17:35
jannauthat was the reason I deleted the the node#asahi-dev · 2022-03-11 17:34
jannauyes, we should not do anything with it as the port is missing from apple's devicetree as well#asahi-dev · 2022-03-11 16:48
jannaumaz: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/arm64/boot/dts/apple/t8103-j457.dts#n39 causes constant MSI interrupts for the ethernet card on port02#asahi-dev · 2022-03-11 16:13
jannaufixing it in the dts is probably the right thing, the port doesn't vanishes just because it's unused#asahi-dev · 2022-03-11 16:12
jannaumaz: one of the imac configs doesn't used pcie port 2. I deleted the port in the model specific dts file. That causes mismatched MSI interrupts. It's trivial to fix in the dts#asahi-dev · 2022-03-11 06:38
jannaumarcan: kettenis didn't want u-boot to become the canonical dtb provider https://github.com/AsahiLinux/m1n1/pull/123#asahi-dev · 2022-03-10 08:13
jannauchainload experiments/dcp.py problem on the t6001 solved, dcpep breaks if the shared memory is not zero initialized#asahi-dev · 2022-03-10 06:53
jannauchadmed: there are debian packages for all architectures so I suppose they should build on aarch64#asahi-dev · 2022-03-09 21:14
jannauyes that is unrelated. is that an error? does the display work?#asahi-dev · 2022-03-09 21:07
jannauwhy would allocating the shadow framebuffer (23M) break dcp after a chainload. the rest of m1n1's fb code is disabled#asahi-dev · 2022-03-09 09:07
jannaumy expectation was that we can support UEFI secure boot via u-boot and have not to do much else to satiesfy standard distro's secure boot needs#asahi-dev · 2022-03-09 08:50
jannaumarcan: alpha release users should update through 1tr once we add signature checks (unless you plan to implement that now as well)#asahi-dev · 2022-03-08 21:46
jannaulinux crash is a different problem#asahi-dev · 2022-03-08 21:32
jannausigh, chainloading m1n1 breaks dcp. it crashes immediately after starting it#asahi-dev · 2022-03-07 22:01
jannauwithout HV dcp crashes with "Message 1: MMU out of tables: 200,200,0"#asahi-dev · 2022-03-07 21:51
jannauthat said experiments/dcp.py works on t6001#asahi-dev · 2022-03-07 21:50
jannausigh, I should not use the virtual uart to upload retina display framebuffers#asahi-dev · 2022-03-06 09:50
jannauthe linux driver crashes dcp early on#asahi-dev · 2022-03-06 09:48
jannaumarcan: the DVA offset doesn't seem to matter. the linux driver doesn't set it at all and experiments/dcp.py uses 0x8_0000_0000 and still "works". Only problem so far are unimplemented callbacks#asahi-dev · 2022-03-06 09:21
jannauemergenz[m]: bluetooth is low priority for the apple silicon. as far as I'm aware nobody started looking into it#asahi-dev · 2022-03-06 09:16
jannaudoes the vm-size property in the dart nodes in the ADT describe any hardware properties or is it just macos policy?#asahi-dev · 2022-03-06 06:44
jannauback channel notifications for plug events look a little overkill for power measuremnet to me#asahi-dev · 2022-03-06 06:37
jannauyes, I assume dcp tries to talk to pmp#asahi-dev · 2022-03-06 06:36
jannauhttps://github.com/jannau/linux/commit/9b8e75c247231e22d18ed3a1b177e901a341638c#asahi-dev · 2022-03-06 06:35
jannaujust adding the power-domain in the dts is enough#asahi-dev · 2022-03-06 06:34
jannaudcp crashes on hdmi unplug if pmp is not powered on#asahi-dev · 2022-03-06 06:33
jannauon the topic of dcp power domains, I suspect that notify_pmp is an implicit power domain dependency. at least for dcp it is#asahi-dev · 2022-03-06 06:31
jannauit was not too bad. I tested without power management early on#asahi-dev · 2022-03-05 21:23
jannausome power management magic, it was missing from the t6001 ps_disp0_cpu0 power state#asahi-dev · 2022-03-05 21:19
jannausigh, it was the "apple,min-state = <4>;"#asahi-dev · 2022-03-05 19:19
jannaualyssa: still struggling with waking dcp up. even gone back to walking tables for dcp-dart#asahi-dev · 2022-03-05 17:34
jannaugood night#asahi-dev · 2022-03-05 17:30
jannaudcpep_cb_rt_bandwidth asks for the two missing registers and we map the disp0 mmio regs for dcp#asahi-dev · 2022-03-05 17:22
jannauat least the dart preserved memory setup works#asahi-dev · 2022-03-05 17:21
jannaualso disp0 has only 5 instead of 7 registers#asahi-dev · 2022-03-05 17:19
jannauit's currently hard-coded in the driver but it fails already to power on dcp/rtkit #asahi-dev · 2022-03-05 17:14
jannaumarcan: not sure what you mean#asahi-dev · 2022-03-05 16:35
jannauHV dcp tracing doesn't blow up on t600x. anyone willing to bet if the linux dcp driver will just work after adding the missing nodes to the device