Change in coreboot[master]: soc/intel/common/block/pcie/rtd3: Add PM methods to the device.

[Cliff Huang (Code Review)]

Attention is currently required from: Bora Guvendik, Lance Zhao, Anil Kumar K, Subrata Banik, Tim Wawrzynczak, Paul Menzel, Thejaswani Putta, Patrick Rudolph.
Cliff Huang has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/61352 )

Change subject: soc/intel/common/block/pcie/rtd3: Add PM methods to the device.
......................................................................

Patch Set 6:

(1 comment)

File src/soc/intel/common/block/pcie/rtd3/chip.h:

https://review.coreboot.org/c/coreboot/+/61352/comment/d8a26f54_7a33cef8 
PS5, Line 9: ACPI_PCIE_RP_EMIT_NONE      = 0x00,   /* None            */
           : 	ACPI_PCIE_RP_EMIT_L23       = 0x01,   /* L23             */
           : 	ACPI_PCIE_RP_EMIT_PSD0      = 0x02,   /* PSD0            */
           : 	ACPI_PCIE_RP_EMIT_SRCK      = 0x04,   /* SRCK            */
           : 	ACPI_PCIE_RP_EMIT_L23_PSD0  = 0x03,   /* L23, PSD0       */
           : 	ACPI_PCIE_RP_EMIT_L23_SRCK  = 0x05,   /* L23, SRCK       */
           : 	ACPI_PCIE_RP_EMIT_PSD0_SRCK = 0x06,   /* PSD0, SRCK      */
           : 	ACPI_PCIE_RP_EMIT_ALL       = 0x07    /* L23, PSD0, SRCK */
           : };
> Since we don't want the default behavior of this driver to change, ACPI_PCIE_RP_EMIT_ALL should be ` […]

Currently, each bit is corresponding method (i.e. bit 0: L23, bit 1: PSD0, bit 2: SRCK) so that it will be easier to determine which particular method is ported using mask in the device code. 
Should we only port those methods when the device needs them? At this time, only 5G device needs it. If a device needs it, the corresponding 'ext_pm_support' in rtd3 should be set to what it is needed and available. how about this?