#asahi · 2022-03-26 13:54
jannauI just crashed u-boot via usb, moved the keyboard (same thinkpad compact kbd with a yanky micro usb connector)#asahi · 2022-03-26 13:48
jannauu-boot repeats the last command on enter#asahi · 2022-03-26 13:46
jannaudoesnt happen with with an usb keyboard, with hw uart following "nknown command ' - try 'help'"#asahi · 2022-03-26 10:12
jannaumain use was benchmarking as it was the only easily available armv8 SoC. faster testing was helpful too despite the annoyance of having to deal with ios. later solved by jailbreaking#asahi · 2022-03-26 10:06
jannauno, to deploy ios apps to test on a ipad mini 2 #asahi · 2022-03-26 10:02
jannauI'm just here for fast armv8 chips. My only x86 mac was a mac mini to test armv8 neon code on an A7#asahi · 2022-03-26 09:42
jannauhas someone recently (with macos 12.3 firmware) tried turning the the hdmi connected display off and on? or switched to a different display? It works on the mac studio#asahi · 2022-03-25 07:58
jannauhttps://fedoraproject.org/wiki/Changes/ReplaceFbdevDrivers#asahi · 2022-03-25 07:58
jannauchadmed: fedora has or at least planned to switch to simpledrm so I would expect that plymouth should work or has been fixed#asahi · 2022-03-24 20:07
jannaugeerlingguy: https://github.com/AsahiLinux/m1n1/pull/178#asahi · 2022-03-24 20:04
jannauyour package index is probably outdated#asahi · 2022-03-24 20:03
jannaugeerlingguy: 'pacman -Syu', current version is containerd-1.6.2-1#asahi · 2022-03-24 20:02
jannaugeerlingguy: do you fancy a reported-by: annotation in the commit message?#asahi · 2022-03-24 20:01
jannaugeerlingguy: yes, I think so. I have a fix which should apply to the mini as well as far as I can tell from Apple's template devicetree#asahi · 2022-03-24 19:45
jannauyes, apple seems to changed how they announce the link speed for the 10gb nic in 12.3 (or earlier)#asahi · 2022-03-24 19:38
jannauthat seems to to be the way apple use to describe the port for 1gb and 10gb#asahi · 2022-03-24 19:36
jannauthere is also a "target-link-speed = 4"#asahi · 2022-03-24 19:35
jannauwe initialize the pcie port as PCIe 1 because apple describes it as "maximum-link-speed = 1" which is clearly wrong#asahi · 2022-03-24 19:33
jannauthat looks better: 'pci 0000:03:00.0: 15.753 Gb/s available PCIe bandwidth, limited by 16.0 GT/s PCIe x1 link at 0000:00:02.0 (capable of 63.012 Gb/s with 16.0 GT/s PCIe x4 link)'#asahi · 2022-03-24 19:26
jannauprobably a problem with pcie initialization#asahi · 2022-03-24 19:23
jannauthe studio uses the same nic#asahi · 2022-03-24 19:23
jannauthat line is from dmesg and sounds if the pcie link is just running with 2.000 Gb/s which would explain poor performance#asahi · 2022-03-24 19:22
jannaubut I see following on the mac studio: "pci 0000:03:00.0: 2.000 Gb/s available PCIe bandwidth, limited by 2.5 GT/s PCIe x1 link at 0000:00:02.0 (capable of 63.012 Gb/s with 16.0 GT/s PCIe x4 link)"#asahi · 2022-03-24 19:21
jannaugeerlingguy: not that I know of. we all bought the mini before the 10gb option was available#asahi · 2022-03-24 19:15
jannauI'm very certain this is following firefox upstream bug https://bugzilla.mozilla.org/show_bug.cgi?id=1757571#asahi · 2022-03-24 18:49
jannauhttps://github.com/AsahiLinux/alarm-PKGBUILDs/commit/9ae781529ba348330c7ca9fe6a7012c28fe400d9#asahi · 2022-03-24 18:48
jannauI don't think it is related to the page size#asahi · 2022-03-24 18:48
jannauit sounds like the same issue we saw last week with the arch linux arm firefox 98#asahi · 2022-03-24 07:20
jannaushouldn't firefox nowadays default to wnative wayland#asahi · 2022-03-24 07:20
jannauscaling with kwin_wayland works well with dcp#asahi · 2022-03-23 21:56
jannaupimox seems to be targeted for raspberry pi 4 so arm64 should not be a problem#asahi · 2022-03-23 21:41
jannauthe fdt in uboot is in ${fdtcontroladdr}, fdt_addr is used to load a dtb#asahi · 2022-03-22 08:23
jannauI think most packagers do nothing and get the default page size of the compile host which is usually 4k#asahi · 2022-03-21 07:33
jannauGlanzmann: USB is also missing upstream, we haven''t added it yet to the devicetree since the reconnect fix is missing#asahi · 2022-03-21 07:30
jannauGlanzmann: also missing from your list of not upstreamed things is keyboard/trackpad. can't be deduced from SPI without HW specific knowledge#asahi · 2022-03-21 07:28
jannauGlanzmann: I haven't checked the status but I believe at most the misc fixes from the wifi changes are merged#asahi · 2022-03-20 17:16
jannauI guess I should try to crash xnu/dcp by modifying the regions in the ADT#asahi · 2022-03-20 17:13
jannausven: pt-region-0 / l2-tt-0#asahi · 2022-03-20 17:12
jannauit is in the ADT#asahi · 2022-03-20 17:11
jannaunot sure what xnu does, it could simply not touch the disp/dcp DARTS#asahi · 2022-03-20 17:10
jannauit seemed slightly less bad than scanning the darts for mapped regions#asahi · 2022-03-20 17:09
jannauthey seem to be stable over macos 11 and 12 and for t8103 and t600x#asahi · 2022-03-20 17:06
jannauwith dcp? do you still have a dcp branch with walk_pages?#asahi · 2022-03-20 17:05
jannaualyssa: m1n1 from this branch https://github.com/jannau/m1n1/tree/dcp#asahi · 2022-03-20 17:03
jannaualyssa: dcp has become a little complicated due the reserved memory handling. It needs a modified m1n1 (chainloading is sufficient) and is incompatible with u-boot#asahi · 2022-03-20 16:47
jannauI can try that#asahi · 2022-03-20 16:46
jannausven: I think dcpext bringup is just replacing dcp/disp0 with dcpext0/dispext0. the powerup issue is then hopefully just the atc phy config#asahi · 2022-03-20 16:40
jannauargh, yes. it's on atc3-phy. how did I convince myself it isn't? Did I forgot to check and remembered the results from the mac studio?#asahi · 2022-03-20 16:32
jannauit's physical hdmi, looking exactly like the mini hdmi#asahi · 2022-03-20 16:31
jannaumacbook pro 14/16 has hdmi as well und looks identical to dcp/disp0 except that iboot doesn't powers it on