#
✕ nick:sven
#asahi-dev · 2021-12-03 18:24
svenyes
#asahi-dev · 2021-12-03 18:23
sventhen the next stage needs to wake it up again with 0x60000000000020 or so
#asahi-dev · 2021-12-03 18:23
svenyou can still send 0x60000000000010 (and possibly even 0x60000000000002 for an even lower power state)
#asahi-dev · 2021-12-03 18:08
svennope, can't get it to reset after it was fully booted once :/
#asahi-dev · 2021-12-03 17:26
svenresetting APCIE_ST doesn't seem to do anything
#asahi-dev · 2021-12-03 17:23
svengoodnight!
#asahi-dev · 2021-12-03 17:23
svenyeah :/
#asahi-dev · 2021-12-03 17:22
sventhat would be annoying
#asahi-dev · 2021-12-03 17:22
svenhmmm..
#asahi-dev · 2021-12-03 17:16
svenp.write32(sart_base+0x40+6*4, 0x800000000>>12)
#asahi-dev · 2021-12-03 17:16
svenp.write32(sart_base+6*4, 0xffffffff)
#asahi-dev · 2021-12-03 17:16
svensart_base = 0x27bc50000
#asahi-dev · 2021-12-03 17:16
svenso i added this to the ans script and then i can only run it once as well
#asahi-dev · 2021-12-03 17:16
svenlet's see
#asahi-dev · 2021-12-03 17:16
svenmarcan: it doesn't work anymore if ANS has been fully booted once apparently
#asahi-dev · 2021-12-03 17:13
sven'night
#asahi-dev · 2021-12-03 17:12
sventhat's just the thunderbolt ones which have that slightly braindead secondary irq configuration. but let's see
#asahi-dev · 2021-12-03 17:12
sventhe ans mailbox shouldn't need any irq configuration
#asahi-dev · 2021-12-03 17:09
svensetting it to 0 before the reset doesn't seem to help. let me try tracing everything
#asahi-dev · 2021-12-03 17:05
svensee [cpu5][0xffff80001054cee4] MMIO: W.4 0x277400044 (ans[0], offset 0x44) = 0x10 in the HV trace :/
#asahi-dev · 2021-12-03 17:03
svenhrm, so i do that in linux and that apparently didn't work
#asahi-dev · 2021-12-03 17:03
svenlet's see
#asahi-dev · 2021-12-03 17:03
svenoh, urgh.
#asahi-dev · 2021-12-03 17:01
sventhis is clearly something very dumb
#asahi-dev · 2021-12-03 17:01
svenbut i manually poke that PMGR reg from shell it doesn't?!
#asahi-dev · 2021-12-03 17:01
sveni can run that thing twice and it seems to work
#asahi-dev · 2021-12-03 17:01
sveni'm just more confused WTF is going on :D
#asahi-dev · 2021-12-03 16:59
svenokay, why the hell does that work
#asahi-dev · 2021-12-03 16:58
svenoh.. SART probably
#asahi-dev · 2021-12-03 16:58
svenhrm, it hangs
#asahi-dev · 2021-12-03 16:56
svenlet me try that code
#asahi-dev · 2021-12-03 16:56
svenokay, wtf
#asahi-dev · 2021-12-03 16:49
svenit's a pity we can't see what iBoot does to reset ANS completely
#asahi-dev · 2021-12-03 16:47
svendoing it manually from shell.py doesn't seem to clear the running flag either
#asahi-dev · 2021-12-03 16:45
svenhm, strange
#asahi-dev · 2021-12-03 16:42
svenhttps://f.svpe.de/2ce0207905dbbf158ca8f5a858471adaa236bf6e075f650249a0e743af3b9d29_ans2_pmgr_reset2.txt and that's with the vanilla pmgr driver from the ML
#asahi-dev · 2021-12-03 16:38
sven(that also includes some additional debug reads)
#asahi-dev · 2021-12-03 16:38
svenmarcan: ^-- that should be the relevant part
#asahi-dev · 2021-12-03 16:38
svenhttps://f.svpe.de/d4d1ec8d667b665143e479b8341dd2336e113282e5185cb862d055a25a84b285_ans2_pmgr_reset.txt
#asahi-dev · 2021-12-03 16:33
svenah yes, it does
#asahi-dev · 2021-12-03 16:33
svenhrm, trace_range(range(0x23b700000, 0x23b700000+0x4000)) should work, right?
#asahi-dev · 2021-12-03 16:29
svensure
#asahi-dev · 2021-12-03 15:31
svenbecause afterwards ANS2 still has the CPU_RUN flag in that register set and i can't restart it
#asahi-dev · 2021-12-03 15:31
svenmarcan: ^-- that should mean the reset was successful, right?
#asahi-dev · 2021-12-03 15:31
sven[ 1.027570] apple-pmgr-pwrstate 23b700000.power-management:power-controller@3f0: PS 0x3f0: deassert reset
#asahi-dev · 2021-12-03 15:31
sven[ 1.023592] apple-pmgr-pwrstate 23b700000.power-management:power-controller@3f0: PS 0x3f0: assert reset
#asahi-dev · 2021-12-03 15:22
svenand it takes the email from the commit author (+ tags if available)
#asahi-dev · 2021-12-03 15:22
svenpretty sure someone added the github repo manually
#asahi-dev · 2021-12-03 14:31
svenit runs on all branches of that repo as well
#asahi-dev · 2021-12-03 14:31
svenyup, that thing is great :D

plus ›