Re: [PATCH] PCI: Add ATS-disable quirk for AMD Stoney GPUs

[David Woodhouse]

On Fri, 2017-04-07 at 14:40 +0200, Joerg Roedel wrote:
> 
> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
> index 6736836..3bc9856 100644
> --- a/drivers/pci/quirks.c
> +++ b/drivers/pci/quirks.c
> @@ -4634,3 +4634,22 @@ static void quirk_no_aersid(struct pci_dev *pdev)
> ÂDECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x2031, quirk_no_aersid);
> ÂDECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x2032, quirk_no_aersid);
> ÂDECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_INTEL, 0x2033, quirk_no_aersid);
> +
> +/*
> + * Some devices have a broken ATS implementation causing IOMMU stalls.
> + * Don't use ATS for those devices.
> + */
> +static void quirk_disable_ats(struct pci_dev *pdev)
> +{
> +#ifdef CONFIG_PCI_ATS
> +ÂÂÂÂÂÂÂ/*
> +ÂÂÂÂÂÂÂ * Set pdev->ats_cap = 0 to make pci_enable_ats() bail out
> +ÂÂÂÂÂÂÂ * early.
> +ÂÂÂÂÂÂÂ */
> +ÂÂÂÂÂÂÂdev_info(&pdev->dev, "QUIRK: Disabling ATS");
> +ÂÂÂÂÂÂÂpdev->ats_cap = 0;
> +#endif
> +}
> +
> +/* AMD Stoney platform GPU */
> +DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, 0x98e4, quirk_disable_ats);

Why not put the #ifdef around *all* of the above?

Attachment: smime.p7s
Description: S/MIME cryptographic signature