[PATCH v14 23/34] cxl: Map CXL Endpoint Port and CXL Switch Port RAS registers

Terry Bowman posted 34 patches 3 weeks, 4 days ago
There is a newer version of this series
[PATCH v14 23/34] cxl: Map CXL Endpoint Port and CXL Switch Port RAS registers
Posted by Terry Bowman 3 weeks, 4 days ago
In preparation for CXL VH (Virtual Host) topology protocol error handling,
add RAS capability registered mapping for all ports in a CXL VH topology.
This includes the RAS capabilities of Switch Upstream Ports, Switch
Downstream Ports, Host Bridge Ports ("upstream"), and Root Ports
("downstream")

Update cxl_port_add_dport() to map the upstream RAS capability on first
'dport' attach, and downstream RAS capability on each 'dport' attach.
Arrange for dport mappings to be released at del_dport() time.

Signed-off-by: Terry Bowman <terry.bowman@amd.com>
Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
[djbw: reword changelog, fix devm handling]
Co-developed-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>

---

Changes in v13->v14:
- Correct message spelling (Terry)
---
 drivers/cxl/core/port.c       |  2 +-
 drivers/cxl/core/ras.c        | 11 +++++++++++
 drivers/cxl/cxl.h             |  2 ++
 drivers/cxl/cxlpci.h          |  4 ++++
 drivers/cxl/port.c            | 37 +++++++++++++++++++++++++++++++++++
 tools/testing/cxl/Kbuild      |  1 +
 tools/testing/cxl/test/mock.c | 12 ++++++++++++
 7 files changed, 68 insertions(+), 1 deletion(-)

diff --git a/drivers/cxl/core/port.c b/drivers/cxl/core/port.c
index 2184c20af011..2c4e28e7975c 100644
--- a/drivers/cxl/core/port.c
+++ b/drivers/cxl/core/port.c
@@ -1451,7 +1451,7 @@ static void del_dport(struct cxl_dport *dport)
 {
 	struct cxl_port *port = dport->port;
 
-	devm_release_action(&port->dev, unlink_dport, dport);
+	devres_release_group(&port->dev, dport);
 }
 
 static void del_dports(struct cxl_port *port)
diff --git a/drivers/cxl/core/ras.c b/drivers/cxl/core/ras.c
index 84abcf90fa99..76ac567724e3 100644
--- a/drivers/cxl/core/ras.c
+++ b/drivers/cxl/core/ras.c
@@ -162,6 +162,17 @@ void devm_cxl_dport_ras_setup(struct cxl_dport *dport)
 }
 EXPORT_SYMBOL_NS_GPL(devm_cxl_dport_ras_setup, "CXL");
 
+void devm_cxl_port_ras_setup(struct cxl_port *port)
+{
+	struct cxl_register_map *map = &port->reg_map;
+
+	map->host = &port->dev;
+	if (cxl_map_component_regs(map, &port->regs,
+				   BIT(CXL_CM_CAP_CAP_ID_RAS)))
+		dev_dbg(&port->dev, "Failed to map RAS capability\n");
+}
+EXPORT_SYMBOL_NS_GPL(devm_cxl_port_ras_setup, "CXL");
+
 void cxl_handle_cor_ras(struct device *dev, u64 serial, void __iomem *ras_base)
 {
 	void __iomem *addr;
diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
index 46491046f101..805923693707 100644
--- a/drivers/cxl/cxl.h
+++ b/drivers/cxl/cxl.h
@@ -607,6 +607,7 @@ struct cxl_dax_region {
  * @parent_dport: dport that points to this port in the parent
  * @decoder_ida: allocator for decoder ids
  * @reg_map: component and ras register mapping parameters
+ * @regs: mapped component registers
  * @nr_dports: number of entries in @dports
  * @hdm_end: track last allocated HDM decoder instance for allocation ordering
  * @commit_end: cursor to track highest committed decoder for commit ordering
@@ -628,6 +629,7 @@ struct cxl_port {
 	struct cxl_dport *parent_dport;
 	struct ida decoder_ida;
 	struct cxl_register_map reg_map;
+	struct cxl_component_regs regs;
 	int nr_dports;
 	int hdm_end;
 	int commit_end;
diff --git a/drivers/cxl/cxlpci.h b/drivers/cxl/cxlpci.h
index e41bb93d583a..ef4496b4e55e 100644
--- a/drivers/cxl/cxlpci.h
+++ b/drivers/cxl/cxlpci.h
@@ -82,6 +82,7 @@ void cxl_cor_error_detected(struct pci_dev *pdev);
 pci_ers_result_t cxl_error_detected(struct pci_dev *pdev,
 				    pci_channel_state_t state);
 void devm_cxl_dport_ras_setup(struct cxl_dport *dport);
+void devm_cxl_port_ras_setup(struct cxl_port *port);
 #else
 static inline void cxl_cor_error_detected(struct pci_dev *pdev) { }
 
@@ -93,6 +94,9 @@ static inline pci_ers_result_t cxl_error_detected(struct pci_dev *pdev,
 static inline void devm_cxl_dport_ras_setup(struct cxl_dport *dport)
 {
 }
+static inline void devm_cxl_port_ras_setup(struct cxl_port *port)
+{
+}
 #endif
 
 #endif /* __CXL_PCI_H__ */
diff --git a/drivers/cxl/port.c b/drivers/cxl/port.c
index 8f8fc98c1428..0d6e010e21ca 100644
--- a/drivers/cxl/port.c
+++ b/drivers/cxl/port.c
@@ -176,11 +176,29 @@ static struct cxl_port *cxl_port_devres_group(struct cxl_port *port)
 DEFINE_FREE(cxl_port_group_free, struct cxl_port *,
 	    if (!IS_ERR_OR_NULL(_T)) devres_release_group(&(_T)->dev, _T))
 
+static struct cxl_dport *cxl_dport_devres_group(struct cxl_dport *dport)
+{
+	if (!devres_open_group(&dport->port->dev, dport, GFP_KERNEL))
+		return ERR_PTR(-ENOMEM);
+	return dport;
+}
+DEFINE_FREE(cxl_dport_group_free, struct cxl_dport *,
+	if (!IS_ERR_OR_NULL(_T)) devres_release_group(&(_T)->port->dev, _T))
+
 static void cxl_port_group_close(struct cxl_port *port)
 {
 	devres_remove_group(&port->dev, port);
 }
 
+/*
+ * Unlike the port group, that just facilitates unwind of setup failures, the
+ * dport group needs to stay live for del_dport() to reference.
+ */
+static void cxl_dport_group_close(struct cxl_dport *dport)
+{
+	devres_close_group(&dport->port->dev, dport);
+}
+
 static struct cxl_dport *cxl_port_add_dport(struct cxl_port *port,
 					    struct device *dport_dev)
 {
@@ -209,6 +227,13 @@ static struct cxl_dport *cxl_port_add_dport(struct cxl_port *port,
 		rc = devm_cxl_switch_port_decoders_setup(port);
 		if (rc)
 			return ERR_PTR(rc);
+
+		/*
+		 * RAS setup is optional, either driver operation can continue
+		 * on failure, or the device does not implement RAS registers.
+		 */
+		devm_cxl_port_ras_setup(port);
+
 		/*
 		 * Note, when nr_dports returns to zero the port is unregistered
 		 * and triggers cleanup. I.e. no need for open-coded release
@@ -220,12 +245,24 @@ static struct cxl_dport *cxl_port_add_dport(struct cxl_port *port,
 	if (IS_ERR(new_dport))
 		return new_dport;
 
+	/*
+	 * Establish a group for all dport resources that need to be released
+	 * when the dport is deleted.
+	 */
+	struct cxl_dport *dport_group __free(cxl_dport_group_free) =
+		cxl_dport_devres_group(new_dport);
+	if (IS_ERR(dport_group))
+		return ERR_CAST(dport_group);
+
 	rc = cxl_dport_autoremove(new_dport);
 	if (rc)
 		return ERR_PTR(rc);
 
+	devm_cxl_dport_ras_setup(new_dport);
+
 	cxl_switch_parse_cdat(new_dport);
 
+	cxl_dport_group_close(no_free_ptr(dport_group));
 	cxl_port_group_close(no_free_ptr(port_group));
 
 	dev_dbg(&port->dev, "dport[%d] id:%d dport_dev: %s added\n",
diff --git a/tools/testing/cxl/Kbuild b/tools/testing/cxl/Kbuild
index 7250bedf0448..6c516019600e 100644
--- a/tools/testing/cxl/Kbuild
+++ b/tools/testing/cxl/Kbuild
@@ -13,6 +13,7 @@ ldflags-y += --wrap=devm_cxl_endpoint_decoders_setup
 ldflags-y += --wrap=hmat_get_extended_linear_cache_size
 ldflags-y += --wrap=cxl_add_dport_by_dev
 ldflags-y += --wrap=devm_cxl_switch_port_decoders_setup
+ldflags-y += --wrap=devm_cxl_port_ras_setup
 
 DRIVERS := ../../../drivers
 CXL_SRC := $(DRIVERS)/cxl
diff --git a/tools/testing/cxl/test/mock.c b/tools/testing/cxl/test/mock.c
index 8883357ee50d..a0b87bbb2f75 100644
--- a/tools/testing/cxl/test/mock.c
+++ b/tools/testing/cxl/test/mock.c
@@ -246,6 +246,18 @@ void __wrap_devm_cxl_dport_ras_setup(struct cxl_dport *dport)
 }
 EXPORT_SYMBOL_NS_GPL(__wrap_devm_cxl_dport_ras_setup, "CXL");
 
+void __wrap_devm_cxl_port_ras_setup(struct cxl_port *port)
+{
+	int index;
+	struct cxl_mock_ops *ops = get_cxl_mock_ops(&index);
+
+	if (!ops || !ops->is_mock_port(port->uport_dev))
+		devm_cxl_port_ras_setup(port);
+
+	put_cxl_mock_ops(index);
+}
+EXPORT_SYMBOL_NS_GPL(__wrap_devm_cxl_port_ras_setup, "CXL");
+
 struct cxl_dport *__wrap_cxl_add_dport_by_dev(struct cxl_port *port,
 					      struct device *dport_dev)
 {
-- 
2.34.1
Re: [PATCH v14 23/34] cxl: Map CXL Endpoint Port and CXL Switch Port RAS registers
Posted by Jonathan Cameron 3 weeks, 3 days ago
On Wed, 14 Jan 2026 12:20:44 -0600
Terry Bowman <terry.bowman@amd.com> wrote:

> In preparation for CXL VH (Virtual Host) topology protocol error handling,
> add RAS capability registered mapping for all ports in a CXL VH topology.
> This includes the RAS capabilities of Switch Upstream Ports, Switch
> Downstream Ports, Host Bridge Ports ("upstream"), and Root Ports
> ("downstream")
> 
> Update cxl_port_add_dport() to map the upstream RAS capability on first
> 'dport' attach, and downstream RAS capability on each 'dport' attach.
> Arrange for dport mappings to be released at del_dport() time.
> 
> Signed-off-by: Terry Bowman <terry.bowman@amd.com>
> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com>
> Reviewed-by: Dave Jiang <dave.jiang@intel.com>
> [djbw: reword changelog, fix devm handling]
> Co-developed-by: Dan Williams <dan.j.williams@intel.com>
> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
> 
One comment inline on which level we handle failures in ras setup at but
that's already true so things aren't made worse by this.

Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com>

I'm not particularly keen on failing to pass errors up to
callers of devm_cxl_dport_ras_setup() which could then cleanly
ignore them with comments saying why. However, that predates this
anyway so a question for another day.
Re: [PATCH v14 23/34] cxl: Map CXL Endpoint Port and CXL Switch Port RAS registers
Posted by Dave Jiang 3 weeks, 4 days ago

On 1/14/26 11:20 AM, Terry Bowman wrote:
> In preparation for CXL VH (Virtual Host) topology protocol error handling,
> add RAS capability registered mapping for all ports in a CXL VH topology.
> This includes the RAS capabilities of Switch Upstream Ports, Switch
> Downstream Ports, Host Bridge Ports ("upstream"), and Root Ports
> ("downstream")
> 
> Update cxl_port_add_dport() to map the upstream RAS capability on first
> 'dport' attach, and downstream RAS capability on each 'dport' attach.
> Arrange for dport mappings to be released at del_dport() time.
> 
> Signed-off-by: Terry Bowman <terry.bowman@amd.com>
> Reviewed-by: Jonathan Cameron <jonathan.cameron@huawei.com>
> Reviewed-by: Dave Jiang <dave.jiang@intel.com>
> [djbw: reword changelog, fix devm handling]

drop the line above

DJ

> Co-developed-by: Dan Williams <dan.j.williams@intel.com>
> Signed-off-by: Dan Williams <dan.j.williams@intel.com>
> 
> ---
> 
> Changes in v13->v14:
> - Correct message spelling (Terry)
> ---
>  drivers/cxl/core/port.c       |  2 +-
>  drivers/cxl/core/ras.c        | 11 +++++++++++
>  drivers/cxl/cxl.h             |  2 ++
>  drivers/cxl/cxlpci.h          |  4 ++++
>  drivers/cxl/port.c            | 37 +++++++++++++++++++++++++++++++++++
>  tools/testing/cxl/Kbuild      |  1 +
>  tools/testing/cxl/test/mock.c | 12 ++++++++++++
>  7 files changed, 68 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/cxl/core/port.c b/drivers/cxl/core/port.c
> index 2184c20af011..2c4e28e7975c 100644
> --- a/drivers/cxl/core/port.c
> +++ b/drivers/cxl/core/port.c
> @@ -1451,7 +1451,7 @@ static void del_dport(struct cxl_dport *dport)
>  {
>  	struct cxl_port *port = dport->port;
>  
> -	devm_release_action(&port->dev, unlink_dport, dport);
> +	devres_release_group(&port->dev, dport);
>  }
>  
>  static void del_dports(struct cxl_port *port)
> diff --git a/drivers/cxl/core/ras.c b/drivers/cxl/core/ras.c
> index 84abcf90fa99..76ac567724e3 100644
> --- a/drivers/cxl/core/ras.c
> +++ b/drivers/cxl/core/ras.c
> @@ -162,6 +162,17 @@ void devm_cxl_dport_ras_setup(struct cxl_dport *dport)
>  }
>  EXPORT_SYMBOL_NS_GPL(devm_cxl_dport_ras_setup, "CXL");
>  
> +void devm_cxl_port_ras_setup(struct cxl_port *port)
> +{
> +	struct cxl_register_map *map = &port->reg_map;
> +
> +	map->host = &port->dev;
> +	if (cxl_map_component_regs(map, &port->regs,
> +				   BIT(CXL_CM_CAP_CAP_ID_RAS)))
> +		dev_dbg(&port->dev, "Failed to map RAS capability\n");
> +}
> +EXPORT_SYMBOL_NS_GPL(devm_cxl_port_ras_setup, "CXL");
> +
>  void cxl_handle_cor_ras(struct device *dev, u64 serial, void __iomem *ras_base)
>  {
>  	void __iomem *addr;
> diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
> index 46491046f101..805923693707 100644
> --- a/drivers/cxl/cxl.h
> +++ b/drivers/cxl/cxl.h
> @@ -607,6 +607,7 @@ struct cxl_dax_region {
>   * @parent_dport: dport that points to this port in the parent
>   * @decoder_ida: allocator for decoder ids
>   * @reg_map: component and ras register mapping parameters
> + * @regs: mapped component registers
>   * @nr_dports: number of entries in @dports
>   * @hdm_end: track last allocated HDM decoder instance for allocation ordering
>   * @commit_end: cursor to track highest committed decoder for commit ordering
> @@ -628,6 +629,7 @@ struct cxl_port {
>  	struct cxl_dport *parent_dport;
>  	struct ida decoder_ida;
>  	struct cxl_register_map reg_map;
> +	struct cxl_component_regs regs;
>  	int nr_dports;
>  	int hdm_end;
>  	int commit_end;
> diff --git a/drivers/cxl/cxlpci.h b/drivers/cxl/cxlpci.h
> index e41bb93d583a..ef4496b4e55e 100644
> --- a/drivers/cxl/cxlpci.h
> +++ b/drivers/cxl/cxlpci.h
> @@ -82,6 +82,7 @@ void cxl_cor_error_detected(struct pci_dev *pdev);
>  pci_ers_result_t cxl_error_detected(struct pci_dev *pdev,
>  				    pci_channel_state_t state);
>  void devm_cxl_dport_ras_setup(struct cxl_dport *dport);
> +void devm_cxl_port_ras_setup(struct cxl_port *port);
>  #else
>  static inline void cxl_cor_error_detected(struct pci_dev *pdev) { }
>  
> @@ -93,6 +94,9 @@ static inline pci_ers_result_t cxl_error_detected(struct pci_dev *pdev,
>  static inline void devm_cxl_dport_ras_setup(struct cxl_dport *dport)
>  {
>  }
> +static inline void devm_cxl_port_ras_setup(struct cxl_port *port)
> +{
> +}
>  #endif
>  
>  #endif /* __CXL_PCI_H__ */
> diff --git a/drivers/cxl/port.c b/drivers/cxl/port.c
> index 8f8fc98c1428..0d6e010e21ca 100644
> --- a/drivers/cxl/port.c
> +++ b/drivers/cxl/port.c
> @@ -176,11 +176,29 @@ static struct cxl_port *cxl_port_devres_group(struct cxl_port *port)
>  DEFINE_FREE(cxl_port_group_free, struct cxl_port *,
>  	    if (!IS_ERR_OR_NULL(_T)) devres_release_group(&(_T)->dev, _T))
>  
> +static struct cxl_dport *cxl_dport_devres_group(struct cxl_dport *dport)
> +{
> +	if (!devres_open_group(&dport->port->dev, dport, GFP_KERNEL))
> +		return ERR_PTR(-ENOMEM);
> +	return dport;
> +}
> +DEFINE_FREE(cxl_dport_group_free, struct cxl_dport *,
> +	if (!IS_ERR_OR_NULL(_T)) devres_release_group(&(_T)->port->dev, _T))
> +
>  static void cxl_port_group_close(struct cxl_port *port)
>  {
>  	devres_remove_group(&port->dev, port);
>  }
>  
> +/*
> + * Unlike the port group, that just facilitates unwind of setup failures, the
> + * dport group needs to stay live for del_dport() to reference.
> + */
> +static void cxl_dport_group_close(struct cxl_dport *dport)
> +{
> +	devres_close_group(&dport->port->dev, dport);
> +}
> +
>  static struct cxl_dport *cxl_port_add_dport(struct cxl_port *port,
>  					    struct device *dport_dev)
>  {
> @@ -209,6 +227,13 @@ static struct cxl_dport *cxl_port_add_dport(struct cxl_port *port,
>  		rc = devm_cxl_switch_port_decoders_setup(port);
>  		if (rc)
>  			return ERR_PTR(rc);
> +
> +		/*
> +		 * RAS setup is optional, either driver operation can continue
> +		 * on failure, or the device does not implement RAS registers.
> +		 */
> +		devm_cxl_port_ras_setup(port);
> +
>  		/*
>  		 * Note, when nr_dports returns to zero the port is unregistered
>  		 * and triggers cleanup. I.e. no need for open-coded release
> @@ -220,12 +245,24 @@ static struct cxl_dport *cxl_port_add_dport(struct cxl_port *port,
>  	if (IS_ERR(new_dport))
>  		return new_dport;
>  
> +	/*
> +	 * Establish a group for all dport resources that need to be released
> +	 * when the dport is deleted.
> +	 */
> +	struct cxl_dport *dport_group __free(cxl_dport_group_free) =
> +		cxl_dport_devres_group(new_dport);
> +	if (IS_ERR(dport_group))
> +		return ERR_CAST(dport_group);
> +
>  	rc = cxl_dport_autoremove(new_dport);
>  	if (rc)
>  		return ERR_PTR(rc);
>  
> +	devm_cxl_dport_ras_setup(new_dport);
> +
>  	cxl_switch_parse_cdat(new_dport);
>  
> +	cxl_dport_group_close(no_free_ptr(dport_group));
>  	cxl_port_group_close(no_free_ptr(port_group));
>  
>  	dev_dbg(&port->dev, "dport[%d] id:%d dport_dev: %s added\n",
> diff --git a/tools/testing/cxl/Kbuild b/tools/testing/cxl/Kbuild
> index 7250bedf0448..6c516019600e 100644
> --- a/tools/testing/cxl/Kbuild
> +++ b/tools/testing/cxl/Kbuild
> @@ -13,6 +13,7 @@ ldflags-y += --wrap=devm_cxl_endpoint_decoders_setup
>  ldflags-y += --wrap=hmat_get_extended_linear_cache_size
>  ldflags-y += --wrap=cxl_add_dport_by_dev
>  ldflags-y += --wrap=devm_cxl_switch_port_decoders_setup
> +ldflags-y += --wrap=devm_cxl_port_ras_setup
>  
>  DRIVERS := ../../../drivers
>  CXL_SRC := $(DRIVERS)/cxl
> diff --git a/tools/testing/cxl/test/mock.c b/tools/testing/cxl/test/mock.c
> index 8883357ee50d..a0b87bbb2f75 100644
> --- a/tools/testing/cxl/test/mock.c
> +++ b/tools/testing/cxl/test/mock.c
> @@ -246,6 +246,18 @@ void __wrap_devm_cxl_dport_ras_setup(struct cxl_dport *dport)
>  }
>  EXPORT_SYMBOL_NS_GPL(__wrap_devm_cxl_dport_ras_setup, "CXL");
>  
> +void __wrap_devm_cxl_port_ras_setup(struct cxl_port *port)
> +{
> +	int index;
> +	struct cxl_mock_ops *ops = get_cxl_mock_ops(&index);
> +
> +	if (!ops || !ops->is_mock_port(port->uport_dev))
> +		devm_cxl_port_ras_setup(port);
> +
> +	put_cxl_mock_ops(index);
> +}
> +EXPORT_SYMBOL_NS_GPL(__wrap_devm_cxl_port_ras_setup, "CXL");
> +
>  struct cxl_dport *__wrap_cxl_add_dport_by_dev(struct cxl_port *port,
>  					      struct device *dport_dev)
>  {