Attention is currently required from: Bora Guvendik, Selma Bensaid, Tim Wawrzynczak. Cliff Huang has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/61783 )
Change subject: mb/google/brya: reduce the time for WWAN _ON method delay. ......................................................................
Patch Set 1:
(3 comments)
Commit Message:
https://review.coreboot.org/c/coreboot/+/61783/comment/0be99cf8_a972865e PS1, Line 9: During booting to OS, ACPI calls _ON and immediately calls _OFF method.
I think the logs are just indicating the status of each PowerResource after the kernel has decided w […]
I think you are correct!
https://review.coreboot.org/c/coreboot/+/61783/comment/657d301d_476917e4 PS1, Line 16: INFO kernel: [ 0.189801] ACPI: Power Resource [RTD3] (on) : INFO kernel: [ 0.206510] ACPI: Power Resource [RTD3] (off)
Actually, I think this is just the status of the two different RTD3 PowerResource objects (one for W […]
I think you are correct. Thanks for pointing out. However, the log does indicate the time spent for the 1st power resource. This is the boot log with preivous delay: 1000 msec [ 0.089461] ACPI: Power Resource [RTD3] (on) [ 1.125167] ACPI: Power Resource [RTD3] (off) <- there 1+ second due to the delay [ 1.125577] ACPI: Power Resource [PR00] (on) [ 1.437048] ACPI: Power Resource [PRIC] (off) [ 1.437112] ACPI: Power Resource [PRIC] (off) [ 1.437804] ACPI: Power Resource [TBT0] (on) [ 1.437820] ACPI: Power Resource [TBT1] (on) [ 1.437837] ACPI: Power Resource [D3C] (on)
File src/mainboard/google/brya/variants/baseboard/brya/devicetree.cb:
https://review.coreboot.org/c/coreboot/+/61783/comment/d60412b7_342062e1 PS1, Line 158: register "reset_delay_ms" = "10"
What is the minimum time required by the datasheet?
The datasheet only shows there is 20 seconds HW initialization after de-assertion of PERST#. However, drivers needs to ensure that. not in ACPI. The datasheet does not provide data on what time duration needs to meet for the next assertion, though. The purpose of delay is just ensure that it is not causing a glitch by _OFF and immediately followed _ON. Note that we also, turn off/on the L23, modPHY clock, an src clock in the methods. If this will never be the case, can I drop this delay entirely?