#
#asahi-dev · 2026-08-15 18:23
yukaBut this makes the difference of reaching >4GHz on my machines
#asahi-dev · 2026-08-15 18:22
yukaMaybe there is an even higher one?
#asahi-dev · 2026-08-15 18:22
svenbut good to know
#asahi-dev · 2026-08-15 18:22
svenhuh, i was somehow convinced that we couldn't boost to the highest pstate yet
#asahi-dev · 2026-08-15 18:21
yuka'sleep' = deep wfi again
#asahi-dev · 2026-08-15 18:21
yukabecause that does smp_start_secondaries and then the other cores sleep while a measurement is being done on one core of the cluster
#asahi-dev · 2026-08-15 18:20
yukaIf the core needed to be powered off, then the cpu pstate experiments in m1n1 could never measure the higher frequencies
#asahi-dev · 2026-08-15 18:20
svenah, i thought that wasn't enough
#asahi-dev · 2026-08-15 18:19
yukacorresponds to wfi losing state and also the last core which is not in 'deep wfi' can boost high
#asahi-dev · 2026-08-15 18:19
sveni thought for boosting to the highest state the other cores had to be actually powered off. but i could be wrong, never looked into that in details
#asahi-dev · 2026-08-15 18:19
yukaThe thing we do using the 'debug' sysreg in cpuidle-apple
#asahi-dev · 2026-08-15 18:18
yukaI don't think so?
#asahi-dev · 2026-08-15 18:18
svenand for boosting you need more, don't you?
#asahi-dev · 2026-08-15 18:18
yukaSorry with deep sleep I meant the wfi mode which loses state
#asahi-dev · 2026-08-15 18:17
svenboth chaos and me tried and while we can get the cores to go into deep sleep / poweroff we can't wake from the state again if we put all cores into it
#asahi-dev · 2026-08-15 18:16
svenonce we actually get that working on m1-m3 we can try filing a radar again and see what happens
#asahi-dev · 2026-08-15 18:16
sven*modify
#asahi-dev · 2026-08-15 18:15
svenso we still don't really know how deep sleep works, but it's more than just that CYC_OVRD sysreg, see the m1n1 code for deep_sleep. and xnu is still allowed to modified the other stuff so it wouldn't surprise if you can somehow still force them to sleep
#asahi-dev · 2026-08-15 18:11
yukaAs in, only the first core of each cluster can boost high
#asahi-dev · 2026-08-15 18:11
yukaAlso I wonder if xnu somehow accounts for the thing where the first core of each cluster does not enter deep sleep, and thus inhibits any other core in the cluster from boosting to a higher frequency
#asahi-dev · 2026-08-15 18:09
yuka*not allowed or don't know how
#asahi-dev · 2026-08-15 18:08
yukat8140/t6040 the same
#asahi-dev · 2026-08-15 18:08
yuka... frequency so I guess that's something.
#asahi-dev · 2026-08-15 18:08
yukaMy conclusion from looking into cpufreq on t8132 is: whatever is done in cpufreq_init() does not work as-is there. The APSC_DIS and APSC_BUSY do not behave as previously and I also couldn't find any new bit like that. Regarding the other cluster features, even if we figure out what xnu does there exactly, we're not allowed to modify them on the boot cluster. But disregarding all that, simply setting the pstate works as before and does increase the...
#asahi-dev · 2026-08-15 17:34
roxfan2looks like 4 int32 or floats
#asahi-dev · 2026-08-15 17:05
svenvery strange
#asahi-dev · 2026-08-15 17:05
yukano pointers, I must have looked at the wrong area last time
#asahi-dev · 2026-08-15 17:04
yuka"7f 20 03 d5 7f 20 03 d5 7f 20 03 d5 7f 20 03 d5"
#asahi-dev · 2026-08-15 17:04
yukait's empty besides this:
#asahi-dev · 2026-08-15 17:01
yukadidn't immediately look like code to me
#asahi-dev · 2026-08-15 17:01
svenweird
#asahi-dev · 2026-08-15 17:01
yukait was some pointers
#asahi-dev · 2026-08-15 17:01
svenusually it’s stuff like coprocessor firmware etc
#asahi-dev · 2026-08-15 17:00
svendoesn’t matter for the PR but I’m curious what’s in there
#asahi-dev · 2026-08-15 17:00
svenright, have you looked at those contents? it’s weird they just lock down a random region
#asahi-dev · 2026-08-15 16:59
yukaI can dump the contents ofc
#asahi-dev · 2026-08-15 16:59
yukaAs I wrote in my PR: It is unclear if this region serves a special purpose
#asahi-dev · 2026-08-15 16:55
svendo you know what’s in that memory region?
#asahi-dev · 2026-08-15 16:55
yukaok, let me update the branch
#asahi-dev · 2026-08-15 16:55
yukalmao
#asahi-dev · 2026-08-15 16:55
svenknowing Apple the solution for that radar will be „let’s also lock down that region on the boot core“ :D
#asahi-dev · 2026-08-15 16:55
svenI’d add the workaround
#asahi-dev · 2026-08-15 16:55
yukasven: do you think we should add a reserved-memory workaround for the ctrr(?) ro issue or write a feedback assistant or both?
#asahi-dev · 2026-08-15 16:17
chaos_princesshow does one figure out which memory carveouts are for dcp?
#asahi-dev · 2026-08-15 15:02
yukahttps://termbin.com/eiitlb
#asahi-dev · 2026-08-15 14:56
yukathey are structured very differently
#asahi-dev · 2026-08-15 14:56
yukaI can not decipher the tz_regs for t8140/t6041
#asahi-dev · 2026-08-15 14:42
yukabut m1n1 does not care about the carveout-memory-map in mmu_add_default_mappings, which is why it needs special handling for the mcc carveouts
#asahi-dev · 2026-08-15 14:41
yukaAh, they are covered by the carveout-memory-map, so they are reserved to the kernel
#asahi-dev · 2026-08-15 14:39
yukabut I guess this is not a problem since the kernel does not blanket map everything?

plus ›