From: Dave Jiang <dave.jiang@intel.com>
To: Richard Cheng <icheng@nvidia.com>,
dave@stgolabs.net, jic23@kernel.org, alison.schofield@intel.com,
vishal.l.verma@intel.com
Cc: iweiny@kernel.org, ming.li@zohomail.com,
linux-cxl@vger.kernel.org, linux-kernel@vger.kernel.org,
kaihengf@nvidia.com, kobak@nvidia.com, newtonl@nvidia.com,
kristinc@nvidia.com
Subject: Re: [PATCH v2] cxl/port: Fix endpoint RAS error unmasking
Date: Tue, 10 Nov 2026 13:07:22 -0700 [thread overview]
Message-ID: <24df53a1-57a9-4a43-808f-520e70a1e37d@intel.com> (raw)
In-Reply-To: <20261103063409.18275-1-icheng@nvidia.com>
On 11/2/26 11:34 PM, Richard Cheng wrote:
> Moving endpoint RAS handling from cxl_pci to cxl_port left
> cxl_ras_unmask() checking whether port->uport_dev is a PCI device.
> Endpoint ports always use a cxl_memdev as uport_dev, so this check
> returns early and leaves the CXL RAS error masks unchanged.
>
> Resolve the PCI device directly through the memdev's parent. Rename
> helper to cxl_endpoint_ras_unmask() and its param to endpoint to make
> its endpoint-only scope explicit.
>
> Fixes: dab7162d0ae7 ("cxl/port: Move endpoint component register management to cxl_port")
> Signed-off-by: Richard Cheng <icheng@nvidia.com>
Applied to cxl/next:
ed574daa568d cxl/port: Fix endpoint RAS error unmasking
> ---
> Changelog:
>
> v1 -> v2:
> - Clarify that endpoint ports always use a cxl_memdev as uport_dev, so
> the PCI device check always skips endpoint RAS unmasking
> - Simplify the PCI device lookup to use the memdev's parent directly
> - Rename cxl_ras_unmask() to cxl_endpoint_ras_unmask() and its param to
> endpoint to make the endpoint-only scope explicitly
> - Rephrase commit messages.
>
> Reproducer is a QEMU Type-3 guest
>
> 1. Ensure no CXL region is active and PCIe AER is enabled.
> 2. Restore the emulated RAS registers to their reset masks.
> 3. Reprobe cxl_pci and read the masks again.
>
> ```
> BDF=$(basename "$(dirname "$(readlink -f /sys/bus/cxl/devices/mem0)")")
> BAR0=$(awk 'NR == 1 { print $1 }' /sys/bus/pci/devices/$BDF/resource)
> UNC_MASK=$(printf '0x%x' $((BAR0 + 0x1084)))
> COR_MASK=$(printf '0x%x' $((BAR0 + 0x1090)))
>
> busybox devmem "$UNC_MASK" 32 0x1cfff
> busybox devmem "$COR_MASK" 32 0x7f
>
> echo "$BDF" > /sys/bus/pci/drivers/cxl_pci/unbind
> echo "$BDF" > /sys/bus/pci/drivers/cxl_pci/bind
> udevadm settle
>
> busybox devmem "$UNC_MASK" 32
> busybox devmem "$COR_MASK" 32
> ```
>
> Before the fix, the masks remain 0x1cfff and 0x7f. After the fix,
> both masks read as zero.
>
> Best regards,
> Richard Cheng.
> ---
> drivers/cxl/port.c | 16 +++++++++-------
> 1 file changed, 9 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/cxl/port.c b/drivers/cxl/port.c
> index c12fd0b89883..7b4726c7276d 100644
> --- a/drivers/cxl/port.c
> +++ b/drivers/cxl/port.c
> @@ -72,19 +72,21 @@ static int cxl_switch_port_probe(struct cxl_port *port)
> return 0;
> }
>
> -static int cxl_ras_unmask(struct cxl_port *port)
> +static int cxl_endpoint_ras_unmask(struct cxl_port *endpoint)
> {
> + struct cxl_memdev *cxlmd = to_cxl_memdev(endpoint->uport_dev);
> + struct device *parent = cxlmd->dev.parent;
> struct pci_dev *pdev;
> void __iomem *addr;
> u32 orig_val, val, mask;
> u16 cap;
> int rc;
>
> - if (!dev_is_pci(port->uport_dev))
> + if (!dev_is_pci(parent))
> return 0;
> - pdev = to_pci_dev(port->uport_dev);
> + pdev = to_pci_dev(parent);
>
> - if (!port->regs.ras) {
> + if (!endpoint->regs.ras) {
> pci_dbg(pdev, "No RAS registers.\n");
> return 0;
> }
> @@ -98,7 +100,7 @@ static int cxl_ras_unmask(struct cxl_port *port)
> return rc;
>
> if (cap & PCI_EXP_DEVCTL_URRE) {
> - addr = port->regs.ras + CXL_RAS_UNCORRECTABLE_MASK_OFFSET;
> + addr = endpoint->regs.ras + CXL_RAS_UNCORRECTABLE_MASK_OFFSET;
> orig_val = readl(addr);
>
> mask = CXL_RAS_UNCORRECTABLE_MASK_MASK |
> @@ -110,7 +112,7 @@ static int cxl_ras_unmask(struct cxl_port *port)
> }
>
> if (cap & PCI_EXP_DEVCTL_CERE) {
> - addr = port->regs.ras + CXL_RAS_CORRECTABLE_MASK_OFFSET;
> + addr = endpoint->regs.ras + CXL_RAS_CORRECTABLE_MASK_OFFSET;
> orig_val = readl(addr);
> val = orig_val & ~CXL_RAS_CORRECTABLE_MASK_MASK;
> writel(val, addr);
> @@ -152,7 +154,7 @@ static int cxl_endpoint_port_probe(struct cxl_port *port)
> devm_cxl_dport_rch_ras_setup(dport);
>
> devm_cxl_port_ras_setup(port);
> - if (cxl_ras_unmask(port))
> + if (cxl_endpoint_ras_unmask(port))
> dev_dbg(&port->dev, "failed to unmask RAS interrupts\n");
>
> /*
>
> base-commit: d135ab453a7400a34930bdda1ef4bdc8acfac80e
prev parent reply other threads:[~2026-11-10 20:07 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-03 6:34 Richard Cheng
2026-11-03 18:35 ` Cheatham, Benjamin
2026-11-03 23:15 ` Dave Jiang
2026-11-04 14:08 ` Cheatham, Benjamin
2026-11-10 20:07 ` Dave Jiang [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=24df53a1-57a9-4a43-808f-520e70a1e37d@intel.com \
--to=dave.jiang@intel.com \
--cc=alison.schofield@intel.com \
--cc=dave@stgolabs.net \
--cc=icheng@nvidia.com \
--cc=iweiny@kernel.org \
--cc=jic23@kernel.org \
--cc=kaihengf@nvidia.com \
--cc=kobak@nvidia.com \
--cc=kristinc@nvidia.com \
--cc=linux-cxl@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=ming.li@zohomail.com \
--cc=newtonl@nvidia.com \
--cc=vishal.l.verma@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®