#asahi-dev · 2026-05-17 13:27
jannauthis isn't much further#asahi-dev · 2026-05-17 13:08
jannauthe latter with macos as partner, the former with tbt devices#asahi-dev · 2026-05-17 13:07
jannausven: what was your last broken tbt state? I get either up to "lane_init.c:343: Entering Phase 2" or "lane_init.c:396: Entering Phase 3" with "lane_init.c:936: Timeout reading AT register"#asahi-dev · 2026-05-17 08:09
jannauyuka: no, not sure whether it's that valueable. it's based on your+lisa's splitting. the command and output changes are intended for easier CI integration#asahi-dev · 2026-05-17 08:08
jannauyes, I dropped the the other two changes from v2#asahi-dev · 2026-05-17 08:01
jannausven: yes, no known/comments issues. be careful about which dt-bindings patches to pick#asahi-dev · 2026-05-15 19:57
jannauI've tagged and pushed asahi-7.0.8-1 to github#asahi-dev · 2026-05-15 19:37
jannaufl0_id: I haven't looked at t6031 PCIe#asahi-dev · 2026-05-15 12:50
jannaunickchan: dev_pm_domain_attach_list() https://elixir.bootlin.com/linux/v7.0.1/source/drivers/base/power/common.c#L174#asahi-dev · 2026-05-15 12:45
jannauthere is a utility function though if there are no ordering requirements#asahi-dev · 2026-05-15 10:30
jannauStanfordAppliedCyber[m]: yuka: chip-revision 17 which matches apple's custom revision encoding scheme in MIDR (extending Revision with bits from Variant) #asahi-dev · 2026-05-14 12:41
jannauthe larger mess is "a17 pro == t8130" vs. "a18 pro == t8140" vs. "m4 == T8132"#asahi-dev · 2026-05-14 12:37
jannaudifference between Pro and Max is the GPU die#asahi-dev · 2026-05-14 12:37
jannauI think there is only a single M5 Pro/Max/Ultra CPU die type #asahi-dev · 2026-05-14 12:35
jannauwhile everest / sawtooth, ... are CPU core code names#asahi-dev · 2026-05-14 12:34
jannauapple uses islands as code names for the SoC since a while: https://asahilinux.org/docs/hw/soc/soc-codenames/#socs#asahi-dev · 2026-05-12 10:11
jannausmc firmware is system global and updated via macOS updates#asahi-dev · 2026-05-12 10:09
jannaucy8aer: probably macos 26.5 smc firmware. I'm quite certain with 26.4 it's still orange at the charge limit#asahi-dev · 2026-05-12 08:43
jannauStanfordAppliedCyber[m]: I think splitting WRITEMEM (possible READMEM as well) request in 16 or 32k chunks would help. as far as I can see the proxyclient is running into timeouts because debugusb buffers too much data and processing on the target is not fast enough#asahi-dev · 2026-05-10 16:01
jannauset 2 is sbu#asahi-dev · 2026-05-10 15:59
jannauand 0x01840306 is SBU1/SBU2 as well#asahi-dev · 2026-05-10 15:58
jannauyuka: cs-sw does not send 0x01800306, it mixes in the variable serial pins in. see `vdm[1] |= 1 << (cxt->serial_pin_set + 16);`#asahi-dev · 2026-05-10 12:56
jannaufaster than uart#asahi-dev · 2026-05-10 12:56
jannauit's even slower on the neo, 350 ms. not sure anything changed or if I measured something else. I think I did memreads for the benchmark but fetching the ADT was clearly faster as well#asahi-dev · 2026-05-10 12:33
jannaudockchannel_uart seems to have to wait ~175ms every 2k of data, no wonder it feels slows. does not match what I measured initally on the a18 pro#asahi-dev · 2026-05-10 12:11
jannauthe arm-io compatible should be per soc family and not model#asahi-dev · 2026-05-10 12:06
jannauand in that case don't check the offset but just the reg directly#asahi-dev · 2026-05-10 12:05
jannauI've discussed with yuka yester to simply add `if (adt_is_compatibe("arm-io,t8140")) { ... }` for every device which needs it#asahi-dev · 2026-05-10 11:56
jannaufor each 16k block dockchannel_uart_getbyte() spins 8 times on DATA_RX_COUNT for over 100ms#asahi-dev · 2026-05-10 11:44
jannaulooks like we're missing data#asahi-dev · 2026-05-10 11:44
jannauno, not always and waiting between the reads in the form of debug printf to the fb seems to help#asahi-dev · 2026-05-10 11:28
jannauchainload via debugusb works if I chunk iodev_read() into SZ_16K chunks#asahi-dev · 2026-05-09 11:26
jannauyuka: have you tried debugusb?#asahi-dev · 2026-05-09 11:26
jannauarryn: if you have a second apple silicon mac iboot will log to debugusb. output is not very informative (file name hashes and line numbers) but it logs when it is finished#asahi-dev · 2026-05-09 11:21
jannauarryn: I'm hearing reports that m1n1 works to some extend on a m5 max#asahi-dev · 2026-05-09 10:26
jannauno more bits/170-atcphy but it looks like one of the fixes hasn't been picked up#asahi-dev · 2026-05-09 10:23
jannauI've pushed asahi-wip rebased onto v7.0.5 to github, please test#asahi-dev · 2026-05-09 07:51
jannaumy question was more if anyone besides povik used it#asahi-dev · 2026-05-09 07:50
jannau just the card->dapm dynamic allocation change or anything else?#asahi-dev · 2026-05-09 07:47
jannauchadmed: is "ASoC: dapm: Export new 'graph.dot' file in debugfs" useful? caused minor issues in the v7.0 rebase (already fixed) but I was wondering if we want to keep it#asahi-dev · 2026-05-08 13:58
jannauthere is one remaining bug. mtp keyboard and trackpad aren't fully probed when u-boot initialized the mtp keyboard#asahi-dev · 2026-05-08 13:57
jannauall pushed to https://github.com/AsahiLinux/u-boot/tree/asahi-releng and released as far copr package#asahi-dev · 2026-05-08 13:56
jannaualso needed to add the new t8103 based fallback compatibles#asahi-dev · 2026-05-08 13:55
jannausiirko[m]: I'm not sure what power-domains have to do with this but supporting M3 in u-boot was straightforward: https://lore.kernel.org/u-boot/20260507-apple-m3-support-v2-1-ce3db1c9e3ec@jannau.net/T/#u#asahi-dev · 2026-05-07 21:10
jannauI tagged and pushed asahi-6.19.14-2 with dirty frag fixes#asahi-dev · 2026-05-07 11:44
jannauI wouldn't expect that to be active at fuOS entry and I also don't see how that breaks `b .`#asahi-dev · 2026-05-07 11:43
jannauthe key feature for A19/M5 is EMTE big part of Apple's MIE https://security.apple.com/blog/memory-integrity-enforcement/#asahi-dev · 2026-05-06 19:52
jannauyuka: fixed#asahi-dev · 2026-05-06 18:59
jannauI think that is by definition true (unless some modified it). I think you want to compare against `rvbar` and please keep the `cpu.state == "running"` skip#asahi-dev · 2026-05-06 18:54
jannauit is. see `new_base = u.base` at the beginning of the file