#asahi-re · 2026-07-21 06:34
yukafor example the pmgr disp*cpu pstates#asahi-re · 2026-07-21 06:33
yukaon M4/M5, I've seen a lot of SErrors on very particular mmio reads/writes#asahi-re · 2026-07-21 06:33
yukaI'm not so sure of that#asahi-re · 2026-07-20 17:49
yuka(probably t6041 too, don't have one to test)#asahi-re · 2026-07-20 17:48
yukanow someone go write code with that please :)#asahi-re · 2026-07-20 17:47
yukat8132/t6040 pcie moves APCIE_PHY_CTRL_RESET from bit 7 to bit 4#asahi-re · 2026-07-18 20:17
yukaHaving tooling in m1n1 for dumping mmio areas which partially cause SErrors would be very useful#asahi-re · 2026-07-18 18:59
yukaso the instruction which caused it already passed#asahi-re · 2026-07-18 18:59
yukait's asynchronous#asahi-re · 2026-07-18 18:59
yukaSError does not usually occur at the instruction that caused it#asahi-re · 2026-07-18 18:53
yukaI haven't seen this issue before#asahi-re · 2026-07-18 18:52
yukaI'm interested in what was the problem#asahi-re · 2026-07-18 17:18
yuka(the proxyclient side is enough to fix your issue, no need to re-install m1n1)#asahi-re · 2026-07-18 17:17
yukasorry, I forgot I always used to comment this out when working with M5#asahi-re · 2026-07-18 17:17
yukayou need https://github.com/AsahiLinux/m1n1/pull/626#asahi-re · 2026-07-17 19:03
yukabooting and proxy worked out of the box iirc, but the usb cdc-acm proxy needed this: https://github.com/AsahiLinux/m1n1/pull/578#asahi-re · 2026-07-17 19:01
yukaAnd on M5 Pro for sure#asahi-re · 2026-07-17 19:01
yukaapparently, because someone just ran my m1n1 experiment on it#asahi-re · 2026-06-30 22:31
yukaI'm all ears#asahi-re · 2026-06-30 22:28
yukait's definitely possible to use the nvme from raw boot mode, since the hypervised mac os of StanfordAppliedCyber already does that#asahi-re · 2026-06-30 22:27
yukamischa85: wdym?#asahi-re · 2026-06-30 22:23
yukais the most likely explanation#asahi-re · 2026-06-30 22:23
yukathey moved on m4#asahi-re · 2026-06-29 22:23
yukaAnd the MLX stuff probably distributes the work on cpu/gpu/ane depending on what's best suited or able to accelerate the workload?#asahi-re · 2026-06-29 22:18
yukaah, fair#asahi-re · 2026-06-29 22:15
yukathis doesn't seem useless?#asahi-re · 2026-06-29 22:15
yukahttps://github.com/ggml-org/whisper.cpp/pull/1021#asahi-re · 2026-06-29 22:05
yukais more understanding necessary to use more features or improve the performance?#asahi-re · 2026-06-29 22:05
yukahow relevant is this anyways, given there's already an ane driver?#asahi-re · 2026-06-18 06:39
yukaMiddle-Core#asahi-re · 2026-06-18 06:39
yukalet's call it exactly that#asahi-re · 2026-06-18 06:39
yukaand M-Core is added in the middle#asahi-re · 2026-06-18 06:39
yukaP-Core continues to be the biggest core#asahi-re · 2026-06-18 06:34
yukaYes I also saw it in M5 device trees already#asahi-re · 2026-06-18 06:33
yukaso we shouldn't adopt them#asahi-re · 2026-06-18 06:33
yukayou are saying apple's marketing terms are not actually consistent with the hardware#asahi-re · 2026-06-18 06:31
yukawhat is the problem with that?#asahi-re · 2026-06-18 06:31
yukacan you elaborate?#asahi-re · 2026-06-16 11:21
yukaSorry if I missed it but what did you / are you planning to reverse engineer?#asahi-re · 2026-06-03 22:57
yukaOops wrong channel#asahi-re · 2026-06-03 22:55
yukaso the one which can boost to 4GHz because the other pcore goes to deep sleep#asahi-re · 2026-06-03 22:54
yukaGot latency measured on the boot core of the Neo#asahi-re · 2026-05-04 12:51
yukaAlso I don't think the manual-availability thing can be a valid indicator, since dockchannel-hid will not work without initializing dart-mtp dapf#asahi-re · 2026-05-04 12:48
yukayou also have some pmgr stuff in there, can you confirm this is actually working to toggle pmgr nodes, and how you did this (no code, just explain the properties from the adt and how they are used)?#asahi-re · 2026-05-04 12:46
yukaI can init the dart-pmp dapf just fine#asahi-re · 2026-05-04 10:43
yukaThe existing variant worked just fine#asahi-re · 2026-05-04 10:43
yukaHmm, I did not need a new dapf_init_t8140#asahi-re · 2026-05-01 07:27
yukahow did you figure it out? looking at mac os binaries?#asahi-re · 2026-05-01 07:27
yukachewyeclaire: very cool! don't have access to my M4 device currently but will try this next week then#asahi-re · 2026-02-17 18:10
yukasimilar