]> Git Repo - linux.git/commit
PCI: qcom: Add OPP support to scale performance
authorKrishna chaitanya chundru <[email protected]>
Wed, 19 Jun 2024 15:11:13 +0000 (20:41 +0530)
committerBjorn Helgaas <[email protected]>
Tue, 9 Jul 2024 21:56:43 +0000 (16:56 -0500)
commit5b6272e0efd53c487bfc5d2921d71ecae37b0446
tree303655e8eb8387aaf99e25024e2d60ab9d04b66c
parent100ae5d77f07f9f046106e228778c7aa1c6d3af3
PCI: qcom: Add OPP support to scale performance

QCOM Resource Power Manager-hardened (RPMh) is a hardware block which
maintains hardware state of a regulator by performing max aggregation of
the requests made by all of the clients.

PCIe controller can operate on different RPMh performance state of power
domain based on the speed of the link. And this performance state varies
from target to target, like some controllers support GEN3 in NOM (Nominal)
voltage corner, while some other supports GEN3 in low SVS (static voltage
scaling).

The SoC can be more power efficient if we scale the performance state
based on the aggregate PCIe link bandwidth.

Add Operating Performance Points (OPP) support to vote for RPMh state based
on the aggregate link bandwidth.

OPP can handle ICC bw voting also, so move ICC bw voting through OPP
framework if OPP entries are present.

As we are moving ICC voting as part of OPP, don't initialize ICC if OPP
is supported.

Before PCIe link is initialized vote for highest OPP in the OPP table,
so that we are voting for maximum voltage corner for the link to come up
in maximum supported speed.

Link: https://lore.kernel.org/linux-pci/[email protected]
Signed-off-by: Krishna chaitanya chundru <[email protected]>
Signed-off-by: Krzysztof WilczyƄski <[email protected]>
[bhelgaas: wrap comments to fit in 80 columns]
Signed-off-by: Bjorn Helgaas <[email protected]>
Reviewed-by: Manivannan Sadhasivam <[email protected]>
drivers/pci/controller/dwc/pcie-qcom.c
This page took 0.057005 seconds and 4 git commands to generate.