linux-kernel - Re: [PATCH 3/5] PCI: brcmstb: enable CRS

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

Message-ID: <20200430203252.GA62266@bjorn-Precision-5520>
Date:   Thu, 30 Apr 2020 15:32:52 -0500
From:   Bjorn Helgaas <helgaas@...nel.org>
To:     Jim Quinlan <james.quinlan@...adcom.com>
Cc:     Nicolas Saenz Julienne <nsaenzjulienne@...e.de>,
        Lorenzo Pieralisi <lorenzo.pieralisi@....com>,
        Rob Herring <robh@...nel.org>,
        Bjorn Helgaas <bhelgaas@...gle.com>,
        Florian Fainelli <f.fainelli@...il.com>,
        "maintainer:BROADCOM BCM7XXX ARM ARCHITECTURE" 
        <bcm-kernel-feedback-list@...adcom.com>,
        "moderated list:BROADCOM BCM2711/BCM2835 ARM ARCHITECTURE" 
        <linux-rpi-kernel@...ts.infradead.org>,
        "moderated list:BROADCOM BCM2711/BCM2835 ARM ARCHITECTURE" 
        <linux-arm-kernel@...ts.infradead.org>,
        "open list:PCI NATIVE HOST BRIDGE AND ENDPOINT DRIVERS" 
        <linux-pci@...r.kernel.org>,
        open list <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 3/5] PCI: brcmstb: enable CRS

On Thu, Apr 30, 2020 at 02:55:20PM -0400, Jim Quinlan wrote:
> From: Jim Quinlan <jquinlan@...adcom.com>
> 
> Configuration Retry Request Status is off by default on this
> PCIe controller.  Turn it on.

Are you talking about CRS itself, i.e., the ability of a Root Port to
deal with Completions with Configuration Retry Request Status?  That
really shouldn't be switchable in the hardware since it's a required
feature for all PCIe devices.

Or are you talking about CRS Software Visibility, which is controlled
by a bit in the PCIe Root Control register?  That *should* be managed
by the PCI core in pci_enable_crs().  Does that generic method of
controlling it not work for this device?

It looks like maybe the latter, since the generic:

  #define  PCI_EXP_RTCTL_CRSSVE   0x0010

matches your new PCIE_RC_CFG_PCIE_ROOT_CAP_CONTROL_RC_CRS_EN_MASK.

If pci_enable_crs() doesn't work on this device, it sounds like a
hardware defect that we need to work around, but I'm not sure that
just enabling it unconditionally here is the right thing.

> Signed-off-by: Jim Quinlan <jquinlan@...adcom.com>
> ---
>  drivers/pci/controller/pcie-brcmstb.c | 9 +++++++++
>  1 file changed, 9 insertions(+)
> 
> diff --git a/drivers/pci/controller/pcie-brcmstb.c b/drivers/pci/controller/pcie-brcmstb.c
> index 5b0dec5971b8..2bc913c0262c 100644
> --- a/drivers/pci/controller/pcie-brcmstb.c
> +++ b/drivers/pci/controller/pcie-brcmstb.c
> @@ -34,6 +34,9 @@
>  #define BRCM_PCIE_CAP_REGS				0x00ac
>  
>  /* Broadcom STB PCIe Register Offsets */
> +#define PCIE_RC_CFG_PCIE_ROOT_CAP_CONTROL			0x00c8
> +#define  PCIE_RC_CFG_PCIE_ROOT_CAP_CONTROL_RC_CRS_EN_MASK	0x10
> +
>  #define PCIE_RC_CFG_VENDOR_VENDOR_SPECIFIC_REG1				0x0188
>  #define  PCIE_RC_CFG_VENDOR_VENDOR_SPECIFIC_REG1_ENDIAN_MODE_BAR2_MASK	0xc
>  #define  PCIE_RC_CFG_VENDOR_SPCIFIC_REG1_LITTLE_ENDIAN			0x0
> @@ -827,6 +830,12 @@ static int brcm_pcie_setup(struct brcm_pcie *pcie)
>  		 pci_speed_string(pcie_link_speed[cls]), nlw,
>  		 ssc_good ? "(SSC)" : "(!SSC)");
>  
> +	/* Enable configuration request retry (CRS) */
> +	tmp = readl(base + PCIE_RC_CFG_PCIE_ROOT_CAP_CONTROL);
> +	u32p_replace_bits(&tmp, 1,
> +			  PCIE_RC_CFG_PCIE_ROOT_CAP_CONTROL_RC_CRS_EN_MASK);
> +	writel(tmp, base + PCIE_RC_CFG_PCIE_ROOT_CAP_CONTROL);
> +
>  	/* PCIe->SCB endian mode for BAR */
>  	tmp = readl(base + PCIE_RC_CFG_VENDOR_VENDOR_SPECIFIC_REG1);
>  	u32p_replace_bits(&tmp, PCIE_RC_CFG_VENDOR_SPCIFIC_REG1_LITTLE_ENDIAN,
> -- 
> 2.17.1
>