#
✕ nick:marcan
#asahi-dev · 2023-04-10 16:55
marcanrobher: what's the best practice for DMA addresses in the DT? use the device cell sizes, or hardcode it to 64-bit always, or something else? (so kettenis_ doesn't hate me when we change the bindings when upstreaming)
#asahi-dev · 2023-04-10 16:53
marcanwill test DCP tomorrow
#asahi-dev · 2023-04-10 16:53
marcanMTP works as intended now with the high DVAs
#asahi-dev · 2023-04-10 16:53
marcanI really need to get some dinner and sleep now :)
#asahi-dev · 2023-04-10 16:52
marcanjannau: pushed and added an apple,dma-range thing, hopefully it does what you want for DCP and we can get rid of the DRAM mask story.
#asahi-dev · 2023-04-10 15:53
marcanyeah
#asahi-dev · 2023-04-10 15:49
marcanlet's do that
#asahi-dev · 2023-04-10 15:49
marcanyeah, we could also use dma-ranges but that only works on buses and it feels like very much the wrong abstraction
#asahi-dev · 2023-04-10 15:47
marcanyeah that's... not ideal :/
#asahi-dev · 2023-04-10 15:45
marcanlooks like this is only settable at the iommu, so just like macOS, it has to go in the DART...
#asahi-dev · 2023-04-10 15:40
marcan(both at its input and in the page tables of course)
#asahi-dev · 2023-04-10 15:40
marcanjannau: I get the feeling vm-base and vm-size are that (the VM range you're supposed to use for iovas/dvas), except when the dart IAS doesn't fit it the top bits just get truncated off
#asahi-dev · 2023-04-10 15:38
marcanI know there's a thing in the dart, but is there one for the device? or should this be in the dart?
#asahi-dev · 2023-04-10 15:36
marcanjannau: how do we deal with that? is there a way to specify the DMA address range?
#asahi-dev · 2023-04-10 15:35
marcanyeah ok
#asahi-dev · 2023-04-10 15:35
marcanI read the dart dump wrong
#asahi-dev · 2023-04-10 15:35
marcanah, I'm dumb, macOS does map it there
#asahi-dev · 2023-04-10 15:05
marcansomething else is going on
#asahi-dev · 2023-04-10 15:05
marcanthis is interesting, with the DART ias limit patch using that high bit in MTP addresses works,but not without it, but macOS does use ias=42 for this
#asahi-dev · 2023-04-10 15:02
marcanfixing
#asahi-dev · 2023-04-10 15:01
marcanyes, that's from chadmed :p
#asahi-dev · 2023-04-10 14:47
marcanack, I'll just pull them directly into my branch
#asahi-dev · 2023-04-10 14:43
marcanhttps://mrcn.st/p/6pL2kvYH https://mrcn.st/p/FtM1BCFv
#asahi-dev · 2023-04-10 14:41
marcanI'm going to get some dinner first though
#asahi-dev · 2023-04-10 14:41
marcanjust pcie left on my side I think?
#asahi-dev · 2023-04-10 14:41
marcanpushed what I have so far
#asahi-dev · 2023-04-10 14:39
marcanand so does the trackpad after getting the firmware in order
#asahi-dev · 2023-04-10 14:37
marcanI think the keyboard works now :)
#asahi-dev · 2023-04-10 14:35
marcanit's kind of funny how I'm running into all these outright bugs that we just somehow never caught on M2
#asahi-dev · 2023-04-10 14:35
marcanwait how did this ever work, I was looking at the wrong OF node?
#asahi-dev · 2023-04-10 14:15
marcannow having GPIO issues...
#asahi-dev · 2023-04-10 14:10
marcanI guess somehow Apple forgot to check for the nGnRnE bit on M2? heh
#asahi-dev · 2023-04-10 14:09
marcanwe only check the parent bus so...
#asahi-dev · 2023-04-10 14:08
marcanwait how did this even work on M2 then?
#asahi-dev · 2023-04-10 14:06
marcanwhen I stopped tracing with the HV
#asahi-dev · 2023-04-10 14:06
marcanlol, having nGnRnE issues with MTP, I thought we'd sorted that mess out a couple years ago...
#asahi-dev · 2023-04-10 13:48
marcanfixed.
#asahi-dev · 2023-04-10 13:48
marcanN: Name="Apple MTP keyboard"
#asahi-dev · 2023-04-10 13:48
marcanI: Bus=0019 Vendor=05ac Product=0353 Version=0270
#asahi-dev · 2023-04-10 13:48
marcanjust the tag in the report, it's some random maximum
#asahi-dev · 2023-04-10 13:41
marcan"let's just have a random maximum for no reason"
#asahi-dev · 2023-04-10 13:41
marcanlol 71f6fa90a3
#asahi-dev · 2023-04-10 13:37
marcanoh no, the HID reports really are wacky
#asahi-dev · 2023-04-10 13:34
marcanjannau: for MTP, macOS is using an offset but that doesn't work for me (it seems to put the full address through the DART). I think the DAPF config may be wrong in that case.
#asahi-dev · 2023-04-10 13:17
marcanthe HID descriptors look fine in the Python dump so I wonder what happened here, maybe just a bug
#asahi-dev · 2023-04-10 11:41
marcanpushed the kernel
#asahi-dev · 2023-04-10 11:41
marcanwill look at the rest later
#asahi-dev · 2023-04-10 11:41
marcan4-level DART works and MTP starts up, though the HID descriptors are borked (possibly some MTP firmware ABI change)
#asahi-dev · 2023-04-10 11:18
marcanfor the laptops I added ~everything, didn't look at the mini at all though
#asahi-dev · 2023-04-10 11:18
marcanthat device isn't done

plus ›