Re: [PATCH v3] usb: xhci-pci: Add AMD 600 series to xhci-pci-prom21

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] usb: xhci-pci: missing PM quirk for new Prom21 chips and incomplete refactoring
--

commit 88cfe4d5f35bfe8fa1398a1c64bdb16fc99db18c
Author: Stephen Horvath <linux@stevetech.au>

usb: xhci-pci: Add AMD 600 series to xhci-pci-prom21

This commit adds the device IDs for the AMD 600 series chipsets (which feature
Promontory 21 chips) to the xhci-pci-prom21 driver and documentation. It
ensures they are recognized alongside the 800 series chipsets.

> diff --git a/drivers/usb/host/xhci-pci.h b/drivers/usb/host/xhci-pci.h
> index 11f435f94322a..533efc7db8012 100644
> --- a/drivers/usb/host/xhci-pci.h
> +++ b/drivers/usb/host/xhci-pci.h
> @@ -4,6 +4,9 @@
>  #ifndef XHCI_PCI_H
>  #define XHCI_PCI_H
>  
> +#define PCI_DEVICE_ID_AMD_PROM21_XHCI_43F7	0x43f7
> +#define PCI_DEVICE_ID_AMD_PROM21_XHCI_43F9	0x43f9
> +#define PCI_DEVICE_ID_AMD_PROM21_XHCI_43FA	0x43fa

[Severity: Medium]
In drivers/usb/host/xhci-pci.c:xhci_pci_quirks(), there is an existing quirk
check that applies to one of the Promontory 21 chips:

    if (pdev->vendor == PCI_VENDOR_ID_AMD && pdev->device == 0x43f7)
        xhci->quirks |= XHCI_DEFAULT_PM_RUNTIME_ALLOW;

Should this hardcoded value be refactored to use the newly defined
PCI_DEVICE_ID_AMD_PROM21_XHCI_43F7 macro?

Additionally, since the new Promontory 21 chips (0x43f9, 0x43fa) are identical
silicon to 0x43f7, do they also need to be added to this check?

Without this quirk, runtime PM will not be enabled by default for motherboards
with B665 or A620 chipsets, causing a power regression when initializing the
xHCI controller.

>  #define PCI_DEVICE_ID_AMD_PROM21_XHCI_43FC	0x43fc
>  #define PCI_DEVICE_ID_AMD_PROM21_XHCI_43FD	0x43fd
>

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261004002731.49927-1-mirko.jechow@gmx.de?part=1