#
✕ nick:kettenis
#asahi-dev · 2021-11-28 23:18
kettenisnot unlikely that the applespi driver needs some adjustments
#asahi-dev · 2021-11-28 22:05
kettenisu-boot supports btrfs though
#asahi-dev · 2021-11-28 18:22
kettenisI'm focussing on upstreaming the missing drivers first
#asahi-dev · 2021-11-28 18:20
kettenisdon't think that'll help
#asahi-dev · 2021-11-28 18:20
kettenisyup
#asahi-dev · 2021-11-28 18:19
kettenisI think those are only used for the legacy boot path
#asahi-dev · 2021-11-28 18:19
kettenisthere are also some hardcoded memory addresses in the u-boot environment
#asahi-dev · 2021-11-28 18:18
kettenisand we might actually be able to build one from the DT
#asahi-dev · 2021-11-28 18:17
kettenisbut we can have multiple memory maps
#asahi-dev · 2021-11-28 18:17
kettenisyeah
#asahi-dev · 2021-11-28 18:17
kettenismarcan: I need to add a t6000-specific memory map
#asahi-dev · 2021-11-28 18:15
kettenisj_ey: I think I came to the conclusion that it would be pointless
#asahi-dev · 2021-11-28 18:10
kettenisi'm considering to simply avoid turning off power domains in u-boot
#asahi-dev · 2021-11-28 18:09
kettenisit is unaware of multiple devices depending on a power domain
#asahi-dev · 2021-11-28 18:09
kettenisspeaking about u-boot, its power domains framework is a bit dumb
#asahi-dev · 2021-11-28 18:07
kettenisOpenBSD won't run on the M1 pro/max until I have a working u-boot and a driver for the new AIC version
#asahi-dev · 2021-11-28 18:02
kettenismy understanding of how it works is still incomplete, but heck of a lot better than whoever write the corellium driver ;)
#asahi-dev · 2021-11-28 18:01
kettenisFWIW, I spent quite some time poking at the hardware writing that driver
#asahi-dev · 2021-11-28 17:54
ketteniscorrect
#asahi-dev · 2021-11-28 17:48
kettenisit reads the fifo level register and the rx level should be 0 at the start of a transfer
#asahi-dev · 2021-11-28 15:10
kettenisbut if you don't clear it, it will still be set at the start of the next transfer
#asahi-dev · 2021-11-28 15:09
kettenisI've defenitely seen that bit get set when the transfer is complete
#asahi-dev · 2021-11-26 12:23
kettenis"simple-audio" and "audio-graph"
#asahi-dev · 2021-11-26 12:22
kettenisand it isn't clear to me what the preferred way of doing that is
#asahi-dev · 2021-11-26 12:21
kettenisthere are two schemes for specifying how the various audio components are related
#asahi-dev · 2021-11-26 12:13
kettenisand then use the "reg" property in the child nodes to provide offsets within the block of registers
#asahi-dev · 2021-11-26 12:13
kettenisyou can add a "syscon" compatible to the MCA block node to make it provide a regmap
#asahi-dev · 2021-11-26 12:12
kettenissure; but maybe check the existing bindings to see if there is an established pattern for such things
#asahi-dev · 2021-11-26 12:09
kettenis"apple,cluster"
#asahi-dev · 2021-11-26 12:08
kettenisthen I'd suggest using a property for that
#asahi-dev · 2021-11-26 12:06
ketteniswhat do you need the cluster index for?
#asahi-dev · 2021-11-26 12:05
kettenisthat sounds wrong
#asahi-dev · 2021-11-26 11:49
kettenisand Apple will maybe source different RAM modules at some point that need different parameters?
#asahi-dev · 2021-11-26 11:48
kettenisbut maybe the actual values depend on the RAM modules integrated in the SoC?
#asahi-dev · 2021-11-26 11:46
kettenisif you choose to put the values in the driver that is
#asahi-dev · 2021-11-26 11:46
kettenisso then you'd need a "generic" compatible and you'd only apply the values if a more specific compatible matches as well
#asahi-dev · 2021-11-26 11:12
kettenisI mean, things will still work if for some future SoC the driver isn't there
#asahi-dev · 2021-11-26 11:10
kettenisthe MCC driver isn't really essential is it?
#asahi-dev · 2021-11-24 16:16
ketteniswould be interesting to see if Apple uses a custom firmware on not
#asahi-dev · 2021-11-24 16:14
kettenissome of the google results suggested it is just a firmware upgrade ;)
#asahi-dev · 2021-11-24 16:00
kettenisgotta save those 2 cents for an EEPROM!
#asahi-dev · 2021-11-23 19:56
kettenisbeing explicit probably is a good thing though, since it allows automated checking of the actual device trees
#asahi-dev · 2021-11-23 16:17
kettenismarcan: still reads ok
#asahi-dev · 2021-11-23 16:10
kettenismaz: I'll push the u-boot change to the branch once I've seen that hit the mailing list
#asahi-dev · 2021-11-23 16:03
kettenisso that'll need to be sneaked into the DT
#asahi-dev · 2021-11-23 16:03
kettenismeanwhile pali noted that PERST really should be active low
#asahi-dev · 2021-11-23 15:50
kettenisfun
#asahi-dev · 2021-11-23 15:50
kettenissending ANS2 the 0x600...220 message doesn't seem to hurt
#asahi-dev · 2021-11-23 08:11
kettenisfor example on the various rasberry pi boards or the Turris MOX
#asahi-dev · 2021-11-23 08:10
kettenisenabling/disabling hardware in the DT based on what hardware is actually configured/present is standard practice

plus ›