#
✕ nick:povik
#asahi-dev · 2022-03-07 15:12
povikmaybe we should settle that too before we start caring about backwards compatibility
#asahi-dev · 2022-03-07 15:12
poviknot exntesively
#asahi-dev · 2022-03-07 15:12
poviki worry that there is a divider-by-half in the way
#asahi-dev · 2022-03-07 15:12
povikon t8103 it's the clock_frequencies value exactly
#asahi-dev · 2022-03-07 15:12
povikmarcan: i had to set the NCO's input clock (in DT) to be half of what's in clock_frequencies in ADT
#asahi-dev · 2022-03-07 15:03
povikgood
#asahi-dev · 2022-03-07 15:01
povikscript appears to be okay
#asahi-dev · 2022-03-07 14:53
povik(with iboot values in muxes on t8103)
#asahi-dev · 2022-03-07 14:52
povikwhich is how it indeed is
#asahi-dev · 2022-03-07 14:52
povikah, i shared the 5th channel between mca4 and mca5
#asahi-dev · 2022-03-07 14:52
povikhmm, don't we have 6th NCO channel? what i did in the DT for t8103, then? :)
#asahi-dev · 2022-03-07 14:51
povikshouldn't matter in the end
#asahi-dev · 2022-03-07 14:51
poviki would clip the presets at the fourth or fifth (as i posted it)
#asahi-dev · 2022-03-07 14:47
povikAOP_MCA is on t8103 also, and there the 5th and 6th clock is for vanilla MCA
#asahi-dev · 2022-03-07 14:47
povikit's possible but would be very surprising
#asahi-dev · 2022-03-07 14:47
povikdon't think so
#asahi-dev · 2022-03-07 14:44
povikmaybe we should remove that last 0x09
#asahi-dev · 2022-03-07 14:44
povikweird thing is on t6000 the reg[2] has 6 clusters worth of clocks, but it only has 4 clusters
#asahi-dev · 2022-03-07 14:43
povikyes, with N=2
#asahi-dev · 2022-03-07 14:43
povikwill test it in a minute on both
#asahi-dev · 2022-03-07 14:43
povikon t8103 it should be a no-op and on t600x it should do the right thing
#asahi-dev · 2022-03-07 14:43
povikmarcan: something like that: https://tpaste.us/Yrnr
#asahi-dev · 2022-03-07 14:34
poviklet me send you a python script with what i think we should do in m1n1
#asahi-dev · 2022-03-07 14:33
poviklet's start with that
#asahi-dev · 2022-03-07 14:32
povikdepends on that
#asahi-dev · 2022-03-07 14:32
povikwhen do you plan that release?
#asahi-dev · 2022-03-07 14:31
povikno capture on t600x
#asahi-dev · 2022-03-07 14:31
povikDcow[m]1: as you said, only two speakers alive
#asahi-dev · 2022-03-07 14:14
povikalso, it's a one-time thing
#asahi-dev · 2022-03-07 14:14
povikthat tells you the addresses, the values cannot be derived from ADT as far as i can tell
#asahi-dev · 2022-03-07 14:14
povikit is that narrow reg range in mca-switch node, if you remember
#asahi-dev · 2022-03-07 14:13
povikmarcan: it's hardcoded
#asahi-dev · 2022-03-07 14:02
povikin the baseline variant of left/right only, needs to be said
#asahi-dev · 2022-03-07 14:02
povik(well, almost)
#asahi-dev · 2022-03-07 14:02
povikaaand with that, speakers are up on t6000 with no outside help
#asahi-dev · 2022-03-07 14:01
povikof course, we were missing ^, duh :)
#asahi-dev · 2022-03-07 14:01
povikCLK_SET_RATE_NO_REPARENT | CLK_SET_RATE_PARENT
#asahi-dev · 2022-03-07 13:41
povikan alternative is having m1n1 put the values there
#asahi-dev · 2022-03-07 13:41
povik(untested, i am about to test it)
#asahi-dev · 2022-03-07 13:41
poviki hacked together a trivial driver: https://github.com/povik/linux/commit/6e36354e8aaf00be1e03e2c27e19c0aed90478b5
#asahi-dev · 2022-03-07 13:41
povikon t6000 iboot doesn't leave the right values in clock muxes on input of MCA
#asahi-dev · 2022-03-06 19:07
povikexactly, you can feel the lack of robh and his like
#asahi-dev · 2022-03-06 18:59
povikwriting bits of DT using ADT as reference is like an "insane to sane" translation
#asahi-dev · 2022-03-06 18:59
povikour bindings are so sexier than apple's, i must chuckle a bit
#asahi-dev · 2022-03-06 10:52
povikdo i need to take care of something?
#asahi-dev · 2022-03-06 10:52
poviki am adding a device link somewhere other than in a probe function of my driver
#asahi-dev · 2022-03-06 10:51
poviki am not sure what synchronization with runtime PM means here
#asahi-dev · 2022-03-06 10:51
povik * with runtime PM.
#asahi-dev · 2022-03-06 10:51
povik * The caller is responsible for the proper synchronization of the link creation
#asahi-dev · 2022-03-06 10:51
povikhttps://github.com/torvalds/linux/blob/master/drivers/base/core.c#L629

plus ›