#asahi-dev · 2026-10-03 03:51
nickchanjannau: please do not remove the MIT license, and all existing files probably need MIT AND GPL-2.0-or-later specified#asahi-dev · 2026-09-30 06:08
nickchanit would add boot args if sysregs are *NOT* locked#asahi-dev · 2026-09-30 06:08
nickchanhttps://github.com/AsahiLinux/m1n1/commit/3a69c73f59332b7b617835fd9a1c49dc89edcfd2 excuse me isn't this backwards#asahi-dev · 2026-09-27 14:46
nickchansven: mind picking these aarch32 rtkit crashlog section additions for 7.4: https://lore.kernel.org/asahi/20260919-rtkit-armv7-crashlog-v1-0-f4a671ea453c@gmail.com/T#asahi-dev · 2026-09-27 14:39
nickchanWhy is starting secondary always cursed for everybody#asahi-dev · 2026-09-27 14:39
nickchanspeaking of which maybe I should try to fix the cursed concurrency issue with start A7-A9 CPUs for the third time (after stopping them for 2 times) that makes the secondary die when it tries to eret from EL3 into EL1 that has been there since the beginning#asahi-dev · 2026-09-27 14:38
nickchanvalidated the pre-M1 cpufreq changes #asahi-dev · 2026-09-26 17:26
nickchanit probably doesn't work on m5ultra either though that's not confirmed#asahi-dev · 2026-09-26 16:09
nickchanand is actually how m1n1 check if a core is ecore#asahi-dev · 2026-09-26 16:08
nickchandespite how it's named in m1n1 MIDR_CORE_TYPE_P and MIDR_CORE_TYPE_M should be checked against MPIDR#asahi-dev · 2026-09-26 16:05
nickchanMPIDR?#asahi-dev · 2026-09-26 14:37
nickchanso what is fsm again#asahi-dev · 2026-09-26 12:44
nickchandelete model and set the simmutable flag from recovery#asahi-dev · 2026-09-26 08:05
nickchanplus ioreg is preinstalled software you are merely trying out the Mac experience#asahi-dev · 2026-09-26 07:54
nickchanioreg -p IODeviceTree -c0 -l | grep '"chip-id" = '#asahi-dev · 2026-09-26 07:54
nickchanjannau: ioreg does print chip-id#asahi-dev · 2026-09-24 18:19
nickchant6051 doesn't seem to exist#asahi-dev · 2026-09-24 18:18
nickchanm5 max and m5 ultra is chip id 0x6050#asahi-dev · 2026-09-24 13:00
nickchanif events are 16bit then it's quite likely the event table is different#asahi-dev · 2026-09-24 13:00
nickchanhttps://github.com/HoolockLinux/linux/commit/7463383e1275eb953f5294b6d3bf2542ca1092ec#asahi-dev · 2026-09-24 12:59
nickchanfor m3+ i think the only useful patch to you from my patches is the one that allows each implementation to declare their own event tables#asahi-dev · 2026-09-24 12:56
nickchanoss xnu does say m3 has 16bit event codes#asahi-dev · 2026-09-24 12:55
nickchanmyria: cpmu is in oss xnu#asahi-dev · 2026-09-24 12:54
nickchanthe rest of the changes is relatively small#asahi-dev · 2026-09-24 12:54
nickchanmost of the changes are the event list#asahi-dev · 2026-09-24 12:53
nickchanmyria: you can review the patches here: https://github.com/HoolockLinux/linux/tree/bits/020-pmu#asahi-dev · 2026-09-24 12:51
nickchanI may send it again if someone sends the m3 pmu patches#asahi-dev · 2026-09-24 12:51
nickchanIt is still in my downstream tree and covered by my tests#asahi-dev · 2026-09-24 12:51
nickchanthey are going nowhere so i stopped sending them#asahi-dev · 2026-09-24 12:47
nickchanmy rtkit crashlog patches are sitting on the list#asahi-dev · 2026-09-24 12:46
nickchanm6 clock speed < a20 clock speeds#asahi-dev · 2026-09-24 12:45
nickchanm6/a20 is another case of cores having different performance characistics #asahi-dev · 2026-09-24 12:44
nickchanfunctionally it can be seen from using the performance counters#asahi-dev · 2026-09-24 12:44
nickchanm3 max ecore has 3 units and m3 pro ecore has 2 units#asahi-dev · 2026-09-24 12:44
nickchanit is detailed in the apple silicon cpu optimization guide#asahi-dev · 2026-09-24 12:44
nickchanfl0_id_: number of FP/SIMD execution units#asahi-dev · 2026-09-24 12:43
nickchanI have not seen a shared midr on apple soc since a10/t2#asahi-dev · 2026-09-24 12:41
nickchanthe pro/max/ultra variant is a bit different from the PA space though#asahi-dev · 2026-09-24 12:41
nickchanm1 is close enough#asahi-dev · 2026-09-24 12:40
nickchanm3pro ecore != m3max ecore 🙃#asahi-dev · 2026-09-24 12:34
nickchanto the point m6/a20 cores are materially different from each other#asahi-dev · 2026-09-24 12:33
nickchana-series and m-series relations are a mess since the A16/A17/M3 #asahi-dev · 2026-09-24 12:23
nickchanmerely from being able to use the hardware mmu via shadow paging in a virtual machine could improve performance#asahi-dev · 2026-09-24 12:23
nickchanthe software MMU emulation is the perf killer of epic proportions#asahi-dev · 2026-09-24 12:22
nickchanfull system emulation is interesting because even without TSO#asahi-dev · 2026-09-22 16:15
nickchanI cannot believe there's zero installs on base M3 whatsoever#asahi-dev · 2026-09-22 16:15
nickchanthere's something not right about the stats server#asahi-dev · 2026-09-16 16:11
nickchanA20 Pro does have "Super Cores" in marketing now#asahi-dev · 2026-09-16 16:11
nickchananother thing that i have missed#asahi-dev · 2026-09-16 09:22
nickchannot enough to explode nvme cause adminq io buffer is only used for things like the smart data but still