#
✕ nick:kettenis1
#asahi · 2021-04-24 11:18
kettenis1It's just that I don't need them for u-boot ;)
#asahi · 2021-04-24 11:16
kettenis1Added https://github.com/AsahiLinux/docs/wiki/HW:GPIO as well
#asahi · 2021-04-24 08:45
kettenis1ams: https://github.com/AsahiLinux/docs/wiki/HW:APCIe
#asahi · 2021-04-23 21:20
kettenis1still has FM
#asahi · 2021-04-23 21:20
kettenis1actually: https://www.broadcom.com/products/wireless/wireless-lan-bluetooth/bcm43752
#asahi · 2021-04-23 21:19
kettenis1but I don't think the wifi 6 chips still have that
#asahi · 2021-04-23 21:18
kettenis1well, older broadcom wifi chips had an FM radio receiver on them
#asahi · 2021-04-23 21:13
kettenis1well, without power the module is definitely not going to emit any radio waves ;)
#asahi · 2021-04-23 21:06
kettenis1made a start with the json/yaml schema for those as well
#asahi · 2021-04-23 21:05
kettenis1I'll put an example for those on the wiki tomorrow
#asahi · 2021-04-23 21:05
kettenis1I also have looked at bindings for the pinctrl/gpio stuff
#asahi · 2021-04-23 21:04
kettenis1what's there now uses the bare minimum of properties that are not already part of the established bindings
#asahi · 2021-04-23 21:04
kettenis1ok, I updated the proposed pcie dt-bindings on the wiki
#asahi · 2021-04-03 22:33
kettenis1is using what I learned from reading the corellium code ok, or do I need to wait for someone to re-RE those bits?
#asahi · 2021-04-03 22:31
kettenis1at this point, the only reference I have for these is the correllium code
#asahi · 2021-04-03 22:30
kettenis12. the "fuse" tunables transplant some bits extracted from registers outside the pcie block (that are mentioned in the "reg" property of the pcie controller in the ADT) to some registers in the pcie block
#asahi · 2021-04-03 22:29
kettenis11. to process some of the phy tunables I need to enable some clocks that are controlled by registers in the pcie block
#asahi · 2021-04-03 22:27
kettenis1two issues there:
#asahi · 2021-04-03 22:27
kettenis1speaking about pcie: I'm looking at the pcie tunables in m1n1
#asahi · 2021-04-03 22:27
kettenis1related to pcie development
#asahi · 2021-04-03 22:24
kettenis1marcan: looks like that bot picked up a branch that it shouldn't
#asahi · 2021-04-02 09:55
kettenis1would be nice to have the clock support and tunables support integrated
#asahi · 2021-04-02 09:55
kettenis1maybe take a look at sven's various m1n1 diffs?
#asahi · 2021-04-01 11:03
kettenis1I'd say so
#asahi · 2021-04-01 10:58
kettenis1i'd say you only need to ducument "reg" and "compatible"
#asahi · 2021-04-01 10:32
kettenis1an "effective" model of the hardware is good enough
#asahi · 2021-04-01 10:31
kettenis1and in the end it won't really matter of course
#asahi · 2021-04-01 10:29
kettenis1it will be hard to figure out though what the real input clock is to the block
#asahi · 2021-04-01 10:27
kettenis1that suggests it is the same bus then
#asahi · 2021-04-01 10:27
kettenis1the bus clock that is
#asahi · 2021-04-01 10:27
kettenis1so with the right tools you might be able to actually measure the clock
#asahi · 2021-04-01 10:26
kettenis1i2c even
#asahi · 2021-04-01 10:26
kettenis1that suggests it is possible to bring the i2x signals out
#asahi · 2021-04-01 10:25
kettenis1https://github.com/AsahiLinux/docs/wiki/HW%3AUSB-PD
#asahi · 2021-04-01 10:19
kettenis1presumably you'd want to set the divisor based on a clock-frequency property
#asahi · 2021-04-01 09:49
kettenis1yeah, that makes sense
#asahi · 2021-04-01 09:49
kettenis1correllium has some pinmuxing for the i2c pins
#asahi · 2021-04-01 09:48
kettenis1sure, but it indicates the other clocks aren't important
#asahi · 2021-04-01 09:46
kettenis1correllium only has gating clocks for i2c
#asahi · 2021-03-29 15:37
kettenis1but the register for that is in the per-port register block so the comparison logic almost certainly lives there too
#asahi · 2021-03-29 15:36
kettenis1marcan: the address can be configured on a per-port basis
#asahi · 2021-03-29 13:46
kettenis1That is where I initialize the MSI config registers of the bridge
#asahi · 2021-03-29 13:46
kettenis1https://github.com/openbsd/src/blob/1f25fa5db5e5b3d240ad71b404053f099fcf5a13/sys/arch/arm64/dev/aplpcie.c#L171
#asahi · 2021-03-29 13:22
kettenis1maz: this is what the XHCI controller hardware reports
#asahi · 2021-03-29 13:19
kettenis1(this is output from the openbsd pcidump tool)
#asahi · 2021-03-29 13:18
kettenis1maz: it means the MSI-X table is at offset 0 in the 3rd BAR
#asahi · 2021-03-29 13:17
kettenis1And isn't quirked in the linux driver
#asahi · 2021-03-29 13:17
kettenis1This one has working MSI
#asahi · 2021-03-29 13:17
kettenis1so it looks as if XHCI only has 8 MSI-X vectors
#asahi · 2021-03-29 13:16
kettenis1 Enabled: no; table size 8 (BAR 2:0)

plus ›