[PATCH v18 11/13] PCI: Cache PCI DSN into pci_dev->dsn during probe

[Terry Bowman]

Subsequent CXL error-reporting code paths need to log the PCI Device
Serial Number (DSN) as part of trace events emitted from interrupt or
panic context. Computing the DSN there via pci_get_dsn() requires PCI
configuration space reads, which are slow, can fail when the link is
down or frozen, and may not be safe in some contexts.

Add a u64 dsn field to struct pci_dev and populate it from pci_get_dsn()
during pci_init_capabilities() at probe time via pci_dsn_init(). Only
write dev->dsn when the read succeeds. The zero initial value from
pci_dev allocation already represents 'no DSN available.'

pci_get_dsn() is not modified because it remains a pure config-space read
with no side effects on pci_dev. The cache is written exclusively by
pci_dsn_init() at probe time.

Signed-off-by: Terry Bowman <terry.bowman@xxxxxxx>

---

Changes in v17->v18:
- New commit.
---
 drivers/cxl/core/ras.c     | 11 ++++++-----
 drivers/cxl/core/ras_rch.c |  4 ++--
 drivers/cxl/pci.c          |  2 +-
 drivers/pci/probe.c        | 14 ++++++++++++++
 include/linux/pci.h        |  1 +
 5 files changed, 24 insertions(+), 8 deletions(-)

diff --git a/drivers/cxl/core/ras.c b/drivers/cxl/core/ras.c
index acf40b2396c3b..69b320c74469c 100644
--- a/drivers/cxl/core/ras.c
+++ b/drivers/cxl/core/ras.c
@@ -100,10 +100,10 @@ void cxl_cper_handle_prot_err(struct cxl_cper_prot_err_work_data *data)
 	dport = cxl_find_dport_by_dev(port, &pdev->dev);
 
 	if (data->severity == AER_CORRECTABLE)
-		cxl_cper_trace_corr_prot_err(port, dport, pci_get_dsn(pdev),
+		cxl_cper_trace_corr_prot_err(port, dport, pdev->dsn,
 					     &data->ras_cap);
 	else
-		cxl_cper_trace_uncorr_prot_err(port, dport, pci_get_dsn(pdev),
+		cxl_cper_trace_uncorr_prot_err(port, dport, pdev->dsn,
 					       &data->ras_cap);
 }
 EXPORT_SYMBOL_GPL(cxl_cper_handle_prot_err);
@@ -195,7 +195,7 @@ void cxl_do_recovery(struct pci_dev *pdev, struct cxl_port *port, struct cxl_dpo
 		return;
 	}
 
-	if (cxl_handle_ras(port, dport, ras_base, pci_get_dsn(pdev)))
+	if (cxl_handle_ras(port, dport, ras_base, pdev->dsn))
 		panic("CXL cachemem error");
 
 	dev_dbg(&pdev->dev,
@@ -307,7 +307,7 @@ pci_ers_result_t cxl_pci_error_detected(struct pci_dev *pdev,
 		 * CXL.mem traffic.
 		 */
 		ue = cxl_handle_ras(port, NULL, to_ras_base(port, NULL),
-				    pci_get_dsn(pdev));
+				    pdev->dsn);
 	}
 
 	/*
@@ -339,7 +339,8 @@ static void cxl_handle_proto_error(struct pci_dev *pdev, struct cxl_port *port,
 				   struct cxl_dport *dport, int severity)
 {
 	if (severity == AER_CORRECTABLE)
-		cxl_handle_cor_ras(port, dport, to_ras_base(port, dport), pci_get_dsn(pdev));
+		cxl_handle_cor_ras(port, dport, to_ras_base(port, dport),
+				   pdev->dsn);
 	else
 		cxl_do_recovery(pdev, port, dport);
 }
diff --git a/drivers/cxl/core/ras_rch.c b/drivers/cxl/core/ras_rch.c
index 0385d2f4a2f66..14bb3bdb2d092 100644
--- a/drivers/cxl/core/ras_rch.c
+++ b/drivers/cxl/core/ras_rch.c
@@ -118,8 +118,8 @@ void cxl_handle_rdport_errors(struct pci_dev *pdev)
 
 	pci_print_aer(pdev, severity, &aer_regs);
 	if (severity == AER_CORRECTABLE)
-		cxl_handle_cor_ras(dport->port, dport, to_ras_base(port, dport),
-				   pci_get_dsn(pdev));
+		cxl_handle_cor_ras(dport->port, dport,
+				   to_ras_base(port, dport), pdev->dsn);
 	else
 		cxl_do_recovery(pdev, dport->port, dport);
 }
diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c
index 6cf1db7b85020..45a994cc782a9 100644
--- a/drivers/cxl/pci.c
+++ b/drivers/cxl/pci.c
@@ -807,7 +807,7 @@ static int cxl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	if (!dvsec)
 		pci_warn(pdev, "Device DVSEC not present, skip CXL.mem init\n");
 
-	mds = cxl_memdev_state_create(&pdev->dev, pci_get_dsn(pdev), dvsec);
+	mds = cxl_memdev_state_create(&pdev->dev, pdev->dsn, dvsec);
 	if (IS_ERR(mds))
 		return PTR_ERR(mds);
 	cxlds = &mds->cxlds;
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index dd0abbc63e18d..92ece5ec211f8 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -2638,6 +2638,19 @@ void pcie_report_downtraining(struct pci_dev *dev)
 	__pcie_print_link_status(dev, false);
 }
 
+/*
+ * Cache the Device Serial Number for use in contexts where config-space
+ * reads are unsafe (interrupt, panic).  Process-context callers that
+ * need a fresh value (e.g. hotplug device replacement) use pci_get_dsn().
+ */
+static void pci_dsn_init(struct pci_dev *dev)
+{
+	u64 dsn = pci_get_dsn(dev);
+
+	if (dsn)
+		dev->dsn = dsn;
+}
+
 static void pci_imm_ready_init(struct pci_dev *dev)
 {
 	u16 status;
@@ -2674,6 +2687,7 @@ static void pci_init_capabilities(struct pci_dev *dev)
 	pci_rebar_init(dev);		/* Resizable BAR */
 	pci_dev3_init(dev);		/* Device 3 capabilities */
 	pci_ide_init(dev);		/* Link Integrity and Data Encryption */
+	pci_dsn_init(dev);		/* Serial number */
 
 	pcie_report_downtraining(dev);
 	pci_init_reset_methods(dev);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 64b308b6e61c1..48a1622639190 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -386,6 +386,7 @@ struct pci_dev {
 	unsigned long	*dma_alias_mask;/* Mask of enabled devfn aliases */
 
 	struct pci_driver *driver;	/* Driver bound to this device */
+	u64		dsn;		/* PCI Device Serial Number */
 	u64		dma_mask;	/* Mask of the bits of bus address this
 					   device implements.  Normally this is
 					   0xffffffff.  You only need to change
-- 
2.34.1