linux-kernel - Re: [PATCH] xhci: workaround for S3 issue on AMD SNPS 3.0 xHC

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <897a7917-5ea6-025b-c516-482188759e0a@intel.com>
Date:   Wed, 16 Sep 2020 15:30:11 +0300
From:   Mathias Nyman <mathias.nyman@...el.com>
To:     Nehal Bakulchandra Shah <Nehal-bakulchandra.Shah@....com>,
        gregkh@...uxfoundation.org, linux-usb@...r.kernel.org,
        linux-kernel@...r.kernel.org, Sandeep.Singh@....com,
        yuanmei@...ovo.com
Subject: Re: [PATCH] xhci: workaround for S3 issue on AMD SNPS 3.0 xHC

On 31.8.2020 9.52, Nehal Bakulchandra Shah wrote:
> From: Nehal Bakulchandra Shah <Nehal-Bakulchandra.shah@....com>
> 
> On some platform of AMD, S3 fails with HCE and SRE errors.To fix this,
> sparse controller enable bit has to be disabled.
> 
> Signed-off-by: Nehal Bakulchandra Shah <Nehal-Bakulchandra.shah@....com>
> ---
>  drivers/usb/host/xhci-pci.c | 12 ++++++++++++
>  drivers/usb/host/xhci.h     |  1 +
>  2 files changed, 13 insertions(+)
> 
> diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c
> index 3feaafebfe58..865a16e6c1ed 100644
> --- a/drivers/usb/host/xhci-pci.c
> +++ b/drivers/usb/host/xhci-pci.c
> @@ -160,6 +160,9 @@ static void xhci_pci_quirks(struct device *dev, struct xhci_hcd *xhci)
>  	    (pdev->device == 0x15e0 || pdev->device == 0x15e1))
>  		xhci->quirks |= XHCI_SNPS_BROKEN_SUSPEND;
>  
> +	if (pdev->vendor == PCI_VENDOR_ID_AMD && pdev->device == 0x15e5)
> +		xhci->quirks |= XHCI_DISABLE_SPARSE;
> +
>  	if (pdev->vendor == PCI_VENDOR_ID_AMD)
>  		xhci->quirks |= XHCI_TRUST_TX_LENGTH;
>  
> @@ -371,6 +374,15 @@ static int xhci_pci_probe(struct pci_dev *dev, const struct pci_device_id *id)
>  	/* USB 2.0 roothub is stored in the PCI device now. */
>  	hcd = dev_get_drvdata(&dev->dev);
>  	xhci = hcd_to_xhci(hcd);
> +
> +	if (xhci->quirks & XHCI_DISABLE_SPARSE) {
> +		u32 reg;
> +
> +		reg = readl(hcd->regs + 0xC12C);
> +		reg &=  ~BIT(17);
> +		writel(reg, hcd->regs + 0xC12C);
> +	}
> +

Is disabling the bit once in probe enough?
xHC will be reset after hibernation as well, does this bit need to be cleared after that?

Also consider turning this into a separate function with a proper description,
see xhci_pme_quirk() for example. Avoids cluttering probe.
Actually if this bit only needs to be cleared once then that function could be called from xhci_pci_setup()

-Mathias