[PATCH v4 00/11] cxl: Add support to report region access coordinates to numa nodes

[PATCH v4 00/11] cxl: Add support to report region access coordinates to numa nodes

[Dave Jiang]

Hi Rafael,
Please review patches 1-4,10,11 and ack if they look ok to you. Thank you!

Hi Greg,
Please review patch 2 and 11 and ack the numa node bits if they look ok to you. Thank you!

v4:
- Introduce access class 0 and 1 for CXL access coordinates.
- See individual patches for detailed change log if applicable.

v3:
- Make attributes not visible if no data. (Jonathan)
- Fix documentation verbiage. (Jonathan)
- Check against read bandwidth instead of write bandwidth due to future RO devices. (Jonathan)
- Export node_set_perf_attrs() to all namespaces. (Jonathan)
- Remove setting of coordinate access level 1. (Jonathan)

v2:
- Move calculation function to core/cdat.c due to QTG series changes
- Make cxlr->coord static (Dan)
- Move calculation to cxl_region_attach to be under cxl_dpa_rwsem (Dan)
- Normalize perf latency numbers to nanoseconds (Brice)
- Update documentation with units and initiator details (Brice, Dan)
- Fix notifier return values (Dan)
- Use devm_add_action_or_reset() to unregister memory notifier (Dan)

This series adds support for computing the performance data of a CXL region
and also updates the performance data to the NUMA node. This series depends
on the CXL QOS class series that's pending 6.8 pull request.

CXL memory devices already attached before boot are enumerated by the BIOS.
The SRAT and HMAT tables are properly setup to including memory regions
enumerated from those CXL memory devices. For regions not programmed or a
hot-plugged CXL memory device, the BIOS does not have the relevant
information and the performance data has to be caluclated by the driver
post region assembly.

According to numaperf documentation [1] there are 2 access classes defined
for performance between an initiator node and a memory target node. Access
class "0" describes attributes between a memory target and the highest
performing initator local to the target. In this case the initiator can be
a CPU or an I/O initiator such as a GPU or NIC. Access class "1" describes
attributes between a memory target and the nearest CPU node. Both access
classes are calculated for the CXL memory target and updated for NUMA nodes
through HMAT_REPORTING code or directly depending on if the NUMA node is
described by the ACPI SRAT table.

Recall from [2] that the performance data for the ranges of a CXL memory device
is computed and cached. A CXL memory region can be backed by one or more
devices. Thus the performance data would be the aggregated bandwidth of all
devices that back a region and the worst latency out of all devices backing
the region.

See git branch [3] for convenience.

[1]: https://www.kernel.org/doc/Documentation/admin-guide/mm/numaperf.rst
[2]: https://lore.kernel.org/linux-cxl/170319606771.2212653.5435838660860735129.stgit@djiang5-mobl3/
[3]: https://git.kernel.org/pub/scm/linux/kernel/git/djiang/linux.git/log/?h=cxl-hmem-report

---

Dave Jiang (11):
      ACPI: HMAT: Remove register of memory node for generic target
      base/node / ACPI: Enumerate node access class for 'struct access_coordinate'
      ACPI: HMAT: Introduce 2 levels of generic port access class
      ACPI: HMAT / cxl: Add retrieval of generic port coordinates for both access classes
      cxl: Split out combine_coordinates() for common shared usage
      cxl: Split out host bridge access coordinates
      cxl: Set cxlmd->endpoint before adding port device
      cxl/region: Calculate performance data for a region
      cxl/region: Add sysfs attribute for locality attributes of CXL regions
      cxl/region: Add memory hotplug notifier for cxl region
      cxl/region: Deal with numa nodes not enumarated by SRAT


 Documentation/ABI/testing/sysfs-bus-cxl |  60 +++++++
 drivers/acpi/numa/hmat.c                |  86 +++++++---
 drivers/acpi/numa/srat.c                |  11 ++
 drivers/base/node.c                     |   7 +-
 drivers/cxl/acpi.c                      |   8 +-
 drivers/cxl/core/cdat.c                 | 142 ++++++++++++++--
 drivers/cxl/core/core.h                 |   4 +
 drivers/cxl/core/port.c                 |  44 +++--
 drivers/cxl/core/region.c               | 216 ++++++++++++++++++++++++
 drivers/cxl/cxl.h                       |  15 +-
 include/linux/acpi.h                    |  10 ++
 include/linux/memory.h                  |   1 +
 include/linux/node.h                    |  18 +-
 13 files changed, 565 insertions(+), 57 deletions(-)

--

[PATCH v4 01/11] ACPI: HMAT: Remove register of memory node for generic target

[Dave Jiang]

For generic targets, there's no reason to call
register_memory_node_under_compute_node() with the access levels that are
only visible to HMAT handling code. Only update the attributes and rename
hmat_register_generic_target_initiators() to hmat_update_generic_target().

Fixes: a3a3e341f169 ("acpi: numa: Add setting of generic port system locality attributes")
Cc: Rafael J. Wysocki <rafael@kernel.org>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
---
 drivers/acpi/numa/hmat.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/acpi/numa/hmat.c b/drivers/acpi/numa/hmat.c
index d6b85f0f6082..a26e7793ec4e 100644
--- a/drivers/acpi/numa/hmat.c
+++ b/drivers/acpi/numa/hmat.c
@@ -770,12 +770,12 @@ static void __hmat_register_target_initiators(struct memory_target *target,
 	}
 }
 
-static void hmat_register_generic_target_initiators(struct memory_target *target)
+static void hmat_update_generic_target(struct memory_target *target)
 {
 	static DECLARE_BITMAP(p_nodes, MAX_NUMNODES);
 
-	__hmat_register_target_initiators(target, p_nodes,
-					  NODE_ACCESS_CLASS_GENPORT_SINK);
+	hmat_update_target_attrs(target, p_nodes,
+				 NODE_ACCESS_CLASS_GENPORT_SINK);
 }
 
 static void hmat_register_target_initiators(struct memory_target *target)
@@ -835,7 +835,7 @@ static void hmat_register_target(struct memory_target *target)
 	 */
 	mutex_lock(&target_lock);
 	if (*(u16 *)target->gen_port_device_handle) {
-		hmat_register_generic_target_initiators(target);
+		hmat_update_generic_target(target);
 		target->registered = true;
 	}
 	mutex_unlock(&target_lock);

[PATCH v4 02/11] base/node / ACPI: Enumerate node access class for 'struct access_coordinate'

[Dave Jiang]

Both generic node and HMAT handling code have been using magic numbers to
indicate access classes for 'struct access_coordinate'. Introduce enums to
enumerate the access0 and access1 classes shared by the two subsystems.
Update the function parameters and callers as appropriate to utilize the
new enum.

Access0 is named to ACCESS_COORDINATE_LOCAL in order to indicate that the
access class is for 'struct access_coordinate' between a target node and
the nearest initiator node.

Access1 is named to ACCESS_COORDINATE_CPU in order to indicate that the
access class is for 'struct access_coordinate' between a target node and
the nearest CPU node.

Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Rafael J. Wysocki <rafael@kernel.org>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
---
 drivers/acpi/numa/hmat.c |   26 ++++++++++++++------------
 drivers/base/node.c      |    6 +++---
 include/linux/node.h     |   18 +++++++++++++++---
 3 files changed, 32 insertions(+), 18 deletions(-)

diff --git a/drivers/acpi/numa/hmat.c b/drivers/acpi/numa/hmat.c
index a26e7793ec4e..e0144cfbf1f3 100644
--- a/drivers/acpi/numa/hmat.c
+++ b/drivers/acpi/numa/hmat.c
@@ -59,9 +59,7 @@ struct target_cache {
 };
 
 enum {
-	NODE_ACCESS_CLASS_0 = 0,
-	NODE_ACCESS_CLASS_1,
-	NODE_ACCESS_CLASS_GENPORT_SINK,
+	NODE_ACCESS_CLASS_GENPORT_SINK = ACCESS_COORDINATE_MAX,
 	NODE_ACCESS_CLASS_MAX,
 };
 
@@ -374,11 +372,11 @@ static __init void hmat_update_target(unsigned int tgt_pxm, unsigned int init_px
 
 	if (target && target->processor_pxm == init_pxm) {
 		hmat_update_target_access(target, type, value,
-					  NODE_ACCESS_CLASS_0);
+					  ACCESS_COORDINATE_LOCAL);
 		/* If the node has a CPU, update access 1 */
 		if (node_state(pxm_to_node(init_pxm), N_CPU))
 			hmat_update_target_access(target, type, value,
-						  NODE_ACCESS_CLASS_1);
+						  ACCESS_COORDINATE_CPU);
 	}
 }
 
@@ -709,7 +707,8 @@ static void hmat_update_target_attrs(struct memory_target *target,
 	 */
 	if (target->processor_pxm != PXM_INVAL) {
 		cpu_nid = pxm_to_node(target->processor_pxm);
-		if (access == 0 || node_state(cpu_nid, N_CPU)) {
+		if (access == ACCESS_COORDINATE_LOCAL ||
+		    node_state(cpu_nid, N_CPU)) {
 			set_bit(target->processor_pxm, p_nodes);
 			return;
 		}
@@ -737,7 +736,8 @@ static void hmat_update_target_attrs(struct memory_target *target,
 		list_for_each_entry(initiator, &initiators, node) {
 			u32 value;
 
-			if (access == 1 && !initiator->has_cpu) {
+			if (access == ACCESS_COORDINATE_CPU &&
+			    !initiator->has_cpu) {
 				clear_bit(initiator->processor_pxm, p_nodes);
 				continue;
 			}
@@ -782,8 +782,10 @@ static void hmat_register_target_initiators(struct memory_target *target)
 {
 	static DECLARE_BITMAP(p_nodes, MAX_NUMNODES);
 
-	__hmat_register_target_initiators(target, p_nodes, 0);
-	__hmat_register_target_initiators(target, p_nodes, 1);
+	__hmat_register_target_initiators(target, p_nodes,
+					  ACCESS_COORDINATE_LOCAL);
+	__hmat_register_target_initiators(target, p_nodes,
+					  ACCESS_COORDINATE_CPU);
 }
 
 static void hmat_register_target_cache(struct memory_target *target)
@@ -854,8 +856,8 @@ static void hmat_register_target(struct memory_target *target)
 	if (!target->registered) {
 		hmat_register_target_initiators(target);
 		hmat_register_target_cache(target);
-		hmat_register_target_perf(target, NODE_ACCESS_CLASS_0);
-		hmat_register_target_perf(target, NODE_ACCESS_CLASS_1);
+		hmat_register_target_perf(target, ACCESS_COORDINATE_LOCAL);
+		hmat_register_target_perf(target, ACCESS_COORDINATE_CPU);
 		target->registered = true;
 	}
 	mutex_unlock(&target_lock);
@@ -927,7 +929,7 @@ static int hmat_calculate_adistance(struct notifier_block *self,
 		return NOTIFY_OK;
 
 	mutex_lock(&target_lock);
-	hmat_update_target_attrs(target, p_nodes, 1);
+	hmat_update_target_attrs(target, p_nodes, ACCESS_COORDINATE_CPU);
 	mutex_unlock(&target_lock);
 
 	perf = &target->coord[1];
diff --git a/drivers/base/node.c b/drivers/base/node.c
index cb2b6cc7f6e6..b4a449f07f2a 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -126,7 +126,7 @@ static void node_access_release(struct device *dev)
 }
 
 static struct node_access_nodes *node_init_node_access(struct node *node,
-						       unsigned int access)
+						       enum access_coordinate_class access)
 {
 	struct node_access_nodes *access_node;
 	struct device *dev;
@@ -191,7 +191,7 @@ static struct attribute *access_attrs[] = {
  * @access: The access class the for the given attributes
  */
 void node_set_perf_attrs(unsigned int nid, struct access_coordinate *coord,
-			 unsigned int access)
+			 enum access_coordinate_class access)
 {
 	struct node_access_nodes *c;
 	struct node *node;
@@ -689,7 +689,7 @@ int register_cpu_under_node(unsigned int cpu, unsigned int nid)
  */
 int register_memory_node_under_compute_node(unsigned int mem_nid,
 					    unsigned int cpu_nid,
-					    unsigned int access)
+					    enum access_coordinate_class access)
 {
 	struct node *init_node, *targ_node;
 	struct node_access_nodes *initiator, *target;
diff --git a/include/linux/node.h b/include/linux/node.h
index 25b66d705ee2..dfc004e4bee7 100644
--- a/include/linux/node.h
+++ b/include/linux/node.h
@@ -34,6 +34,18 @@ struct access_coordinate {
 	unsigned int write_latency;
 };
 
+/*
+ * ACCESS_COORDINATE_LOCAL correlates to ACCESS CLASS 0
+ *	- access_coordinate between target node and nearest initiator node
+ * ACCESS_COORDINATE_CPU correlates to ACCESS CLASS 1
+ *	- access_coordinate between target node and nearest CPU node
+ */
+enum access_coordinate_class {
+	ACCESS_COORDINATE_LOCAL,
+	ACCESS_COORDINATE_CPU,
+	ACCESS_COORDINATE_MAX
+};
+
 enum cache_indexing {
 	NODE_CACHE_DIRECT_MAP,
 	NODE_CACHE_INDEXED,
@@ -66,7 +78,7 @@ struct node_cache_attrs {
 #ifdef CONFIG_HMEM_REPORTING
 void node_add_cache(unsigned int nid, struct node_cache_attrs *cache_attrs);
 void node_set_perf_attrs(unsigned int nid, struct access_coordinate *coord,
-			 unsigned access);
+			 enum access_coordinate_class access);
 #else
 static inline void node_add_cache(unsigned int nid,
 				  struct node_cache_attrs *cache_attrs)
@@ -75,7 +87,7 @@ static inline void node_add_cache(unsigned int nid,
 
 static inline void node_set_perf_attrs(unsigned int nid,
 				       struct access_coordinate *coord,
-				       unsigned access)
+				       enum access_coordinate_class access)
 {
 }
 #endif
@@ -137,7 +149,7 @@ extern void unregister_memory_block_under_nodes(struct memory_block *mem_blk);
 
 extern int register_memory_node_under_compute_node(unsigned int mem_nid,
 						   unsigned int cpu_nid,
-						   unsigned access);
+						   enum access_coordinate_class access);
 #else
 static inline void node_dev_init(void)
 {

[PATCH v4 03/11] ACPI: HMAT: Introduce 2 levels of generic port access class

[Dave Jiang]

In order to compute access0 and access1 classes for CXL memory, 2 levels
of generic port information must be stored. Access0 will indicate the
generic port access coordinates to the closest initiator and access1
will indicate the generic port access coordinates to the cloest CPU.

Cc: Rafael J. Wysocki <rafael@kernel.org>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
---
 drivers/acpi/numa/hmat.c |   18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/drivers/acpi/numa/hmat.c b/drivers/acpi/numa/hmat.c
index e0144cfbf1f3..8dbb0e366059 100644
--- a/drivers/acpi/numa/hmat.c
+++ b/drivers/acpi/numa/hmat.c
@@ -59,7 +59,8 @@ struct target_cache {
 };
 
 enum {
-	NODE_ACCESS_CLASS_GENPORT_SINK = ACCESS_COORDINATE_MAX,
+	NODE_ACCESS_CLASS_GENPORT_SINK_LOCAL = ACCESS_COORDINATE_MAX,
+	NODE_ACCESS_CLASS_GENPORT_SINK_CPU,
 	NODE_ACCESS_CLASS_MAX,
 };
 
@@ -141,7 +142,7 @@ int acpi_get_genport_coordinates(u32 uid,
 	if (!target)
 		return -ENOENT;
 
-	*coord = target->coord[NODE_ACCESS_CLASS_GENPORT_SINK];
+	*coord = target->coord[NODE_ACCESS_CLASS_GENPORT_SINK_LOCAL];
 
 	return 0;
 }
@@ -695,7 +696,8 @@ static void hmat_update_target_attrs(struct memory_target *target,
 	int i;
 
 	/* Don't update for generic port if there's no device handle */
-	if (access == NODE_ACCESS_CLASS_GENPORT_SINK &&
+	if ((access == NODE_ACCESS_CLASS_GENPORT_SINK_LOCAL ||
+	     access == NODE_ACCESS_CLASS_GENPORT_SINK_CPU) &&
 	    !(*(u16 *)target->gen_port_device_handle))
 		return;
 
@@ -707,7 +709,8 @@ static void hmat_update_target_attrs(struct memory_target *target,
 	 */
 	if (target->processor_pxm != PXM_INVAL) {
 		cpu_nid = pxm_to_node(target->processor_pxm);
-		if (access == ACCESS_COORDINATE_LOCAL ||
+		if ((access == ACCESS_COORDINATE_LOCAL ||
+		     access == NODE_ACCESS_CLASS_GENPORT_SINK_LOCAL) &&
 		    node_state(cpu_nid, N_CPU)) {
 			set_bit(target->processor_pxm, p_nodes);
 			return;
@@ -736,7 +739,8 @@ static void hmat_update_target_attrs(struct memory_target *target,
 		list_for_each_entry(initiator, &initiators, node) {
 			u32 value;
 
-			if (access == ACCESS_COORDINATE_CPU &&
+			if ((access == ACCESS_COORDINATE_CPU &&
+			     access == NODE_ACCESS_CLASS_GENPORT_SINK_CPU) &&
 			    !initiator->has_cpu) {
 				clear_bit(initiator->processor_pxm, p_nodes);
 				continue;
@@ -775,7 +779,9 @@ static void hmat_update_generic_target(struct memory_target *target)
 	static DECLARE_BITMAP(p_nodes, MAX_NUMNODES);
 
 	hmat_update_target_attrs(target, p_nodes,
-				 NODE_ACCESS_CLASS_GENPORT_SINK);
+				 NODE_ACCESS_CLASS_GENPORT_SINK_LOCAL);
+	hmat_update_target_attrs(target, p_nodes,
+				 NODE_ACCESS_CLASS_GENPORT_SINK_CPU);
 }
 
 static void hmat_register_target_initiators(struct memory_target *target)

[PATCH v4 04/11] ACPI: HMAT / cxl: Add retrieval of generic port coordinates for both access classes

[Dave Jiang]

Update acpi_get_genport_coordinates() to allow retrieval of both access
classes of the 'struct access_coordinate' for a generic target. The update
will allow CXL code to compute access coordinates for both access class.

Cc: Rafael J. Wysocki <rafael@kernel.org>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
---
 drivers/acpi/numa/hmat.c |    8 ++++++--
 drivers/cxl/acpi.c       |    8 +++++---
 drivers/cxl/core/port.c  |    2 +-
 drivers/cxl/cxl.h        |    2 +-
 4 files changed, 13 insertions(+), 7 deletions(-)

diff --git a/drivers/acpi/numa/hmat.c b/drivers/acpi/numa/hmat.c
index 8dbb0e366059..5be79896bd08 100644
--- a/drivers/acpi/numa/hmat.c
+++ b/drivers/acpi/numa/hmat.c
@@ -126,7 +126,8 @@ static struct memory_target *acpi_find_genport_target(u32 uid)
 /**
  * acpi_get_genport_coordinates - Retrieve the access coordinates for a generic port
  * @uid: ACPI unique id
- * @coord: The access coordinates written back out for the generic port
+ * @coord: The access coordinates written back out for the generic port.
+ *	   Expect 2 levels array.
  *
  * Return: 0 on success. Errno on failure.
  *
@@ -142,7 +143,10 @@ int acpi_get_genport_coordinates(u32 uid,
 	if (!target)
 		return -ENOENT;
 
-	*coord = target->coord[NODE_ACCESS_CLASS_GENPORT_SINK_LOCAL];
+	coord[ACCESS_COORDINATE_LOCAL] =
+		target->coord[NODE_ACCESS_CLASS_GENPORT_SINK_LOCAL];
+	coord[ACCESS_COORDINATE_CPU] =
+		target->coord[NODE_ACCESS_CLASS_GENPORT_SINK_CPU];
 
 	return 0;
 }
diff --git a/drivers/cxl/acpi.c b/drivers/cxl/acpi.c
index afc712264d1c..7834fc208229 100644
--- a/drivers/cxl/acpi.c
+++ b/drivers/cxl/acpi.c
@@ -522,13 +522,15 @@ static int get_genport_coordinates(struct device *dev, struct cxl_dport *dport)
 	if (kstrtou32(acpi_device_uid(hb), 0, &uid))
 		return -EINVAL;
 
-	rc = acpi_get_genport_coordinates(uid, &dport->hb_coord);
+	rc = acpi_get_genport_coordinates(uid, dport->hb_coord);
 	if (rc < 0)
 		return rc;
 
 	/* Adjust back to picoseconds from nanoseconds */
-	dport->hb_coord.read_latency *= 1000;
-	dport->hb_coord.write_latency *= 1000;
+	for (int i = 0; i < ACCESS_COORDINATE_MAX; i++) {
+		dport->hb_coord[i].read_latency *= 1000;
+		dport->hb_coord[i].write_latency *= 1000;
+	}
 
 	return 0;
 }
diff --git a/drivers/cxl/core/port.c b/drivers/cxl/core/port.c
index 8c00fd6be730..7a5eff45e1e2 100644
--- a/drivers/cxl/core/port.c
+++ b/drivers/cxl/core/port.c
@@ -2151,7 +2151,7 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port,
 	}
 
 	/* Augment with the generic port (host bridge) perf data */
-	combine_coordinates(&c, &dport->hb_coord);
+	combine_coordinates(&c, &dport->hb_coord[ACCESS_COORDINATE_LOCAL]);
 
 	/* Get the calculated PCI paths bandwidth */
 	pdev = to_pci_dev(port->uport_dev->parent);
diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
index 492dbf63935f..0fc06455233d 100644
--- a/drivers/cxl/cxl.h
+++ b/drivers/cxl/cxl.h
@@ -673,7 +673,7 @@ struct cxl_dport {
 	struct cxl_port *port;
 	struct cxl_regs regs;
 	struct access_coordinate sw_coord;
-	struct access_coordinate hb_coord;
+	struct access_coordinate hb_coord[ACCESS_COORDINATE_MAX];
 	long link_latency;
 };
 

[PATCH v4 05/11] cxl: Split out combine_coordinates() for common shared usage

[Dave Jiang]

Refactor the common code of combining coordinates in order to reduce code.
Create a new function cxl_cooordinates_combine() it combine two 'struct
access_coordinate'.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
---
 drivers/cxl/core/cdat.c |   32 +++++++++++++++++++++++---------
 drivers/cxl/core/port.c |   18 ++----------------
 drivers/cxl/cxl.h       |    4 ++++
 3 files changed, 29 insertions(+), 25 deletions(-)

diff --git a/drivers/cxl/core/cdat.c b/drivers/cxl/core/cdat.c
index cd84d87f597a..4d542627d02d 100644
--- a/drivers/cxl/core/cdat.c
+++ b/drivers/cxl/core/cdat.c
@@ -183,15 +183,7 @@ static int cxl_port_perf_data_calculate(struct cxl_port *port,
 	xa_for_each(dsmas_xa, index, dent) {
 		int qos_class;
 
-		dent->coord.read_latency = dent->coord.read_latency +
-					   c.read_latency;
-		dent->coord.write_latency = dent->coord.write_latency +
-					    c.write_latency;
-		dent->coord.read_bandwidth = min_t(int, c.read_bandwidth,
-						   dent->coord.read_bandwidth);
-		dent->coord.write_bandwidth = min_t(int, c.write_bandwidth,
-						    dent->coord.write_bandwidth);
-
+		cxl_coordinates_combine(&dent->coord, &dent->coord, &c);
 		dent->entries = 1;
 		rc = cxl_root->ops->qos_class(root_port, &dent->coord, 1, &qos_class);
 		if (rc != 1)
@@ -514,4 +506,26 @@ void cxl_switch_parse_cdat(struct cxl_port *port)
 }
 EXPORT_SYMBOL_NS_GPL(cxl_switch_parse_cdat, CXL);
 
+/**
+ * cxl_coordinates_combine - Combine the two input coordinates into the first
+ *
+ * @c1: first coordinate, to be written to
+ * @c2: second coordinate
+ */
+void cxl_coordinates_combine(struct access_coordinate *out,
+			     struct access_coordinate *c1,
+			     struct access_coordinate *c2)
+{
+		if (c2->write_bandwidth)
+			out->write_bandwidth = min(c1->write_bandwidth,
+						   c2->write_bandwidth);
+		out->write_latency = c1->write_latency + c2->write_latency;
+
+		if (c2->read_bandwidth)
+			out->read_bandwidth = min(c1->read_bandwidth,
+						  c2->read_bandwidth);
+		out->read_latency = c1->read_latency + c2->read_latency;
+}
+EXPORT_SYMBOL_NS_GPL(cxl_coordinates_combine, CXL);
+
 MODULE_IMPORT_NS(CXL);
diff --git a/drivers/cxl/core/port.c b/drivers/cxl/core/port.c
index 7a5eff45e1e2..1d3a04ef8b4f 100644
--- a/drivers/cxl/core/port.c
+++ b/drivers/cxl/core/port.c
@@ -2095,20 +2095,6 @@ bool schedule_cxl_memdev_detach(struct cxl_memdev *cxlmd)
 }
 EXPORT_SYMBOL_NS_GPL(schedule_cxl_memdev_detach, CXL);
 
-static void combine_coordinates(struct access_coordinate *c1,
-				struct access_coordinate *c2)
-{
-		if (c2->write_bandwidth)
-			c1->write_bandwidth = min(c1->write_bandwidth,
-						  c2->write_bandwidth);
-		c1->write_latency += c2->write_latency;
-
-		if (c2->read_bandwidth)
-			c1->read_bandwidth = min(c1->read_bandwidth,
-						 c2->read_bandwidth);
-		c1->read_latency += c2->read_latency;
-}
-
 /**
  * cxl_endpoint_get_perf_coordinates - Retrieve performance numbers stored in dports
  *				   of CXL path
@@ -2142,7 +2128,7 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port,
 	 * nothing to gather.
 	 */
 	while (iter && !is_cxl_root(to_cxl_port(iter->dev.parent))) {
-		combine_coordinates(&c, &dport->sw_coord);
+		cxl_coordinates_combine(&c, &c, &dport->sw_coord);
 		c.write_latency += dport->link_latency;
 		c.read_latency += dport->link_latency;
 
@@ -2151,7 +2137,7 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port,
 	}
 
 	/* Augment with the generic port (host bridge) perf data */
-	combine_coordinates(&c, &dport->hb_coord[ACCESS_COORDINATE_LOCAL]);
+	cxl_coordinates_combine(&c, &c, &dport->hb_coord[ACCESS_COORDINATE_LOCAL]);
 
 	/* Get the calculated PCI paths bandwidth */
 	pdev = to_pci_dev(port->uport_dev->parent);
diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
index 0fc06455233d..0a0a121ee780 100644
--- a/drivers/cxl/cxl.h
+++ b/drivers/cxl/cxl.h
@@ -879,6 +879,10 @@ void cxl_switch_parse_cdat(struct cxl_port *port);
 int cxl_endpoint_get_perf_coordinates(struct cxl_port *port,
 				      struct access_coordinate *coord);
 
+void cxl_coordinates_combine(struct access_coordinate *out,
+			     struct access_coordinate *c1,
+			     struct access_coordinate *c2);
+
 /*
  * Unit test builds overrides this to __weak, find the 'strong' version
  * of these symbols in tools/testing/cxl/.

[PATCH v4 06/11] cxl: Split out host bridge access coordinates

[Dave Jiang]

The difference between access class 0 and access class 1 for 'struct
access_coordinate', if any, is that class 0 is for the distance from
the target to the closest initiator and that class 1 is for the distance
from the target to the cloest CPU. For CXL memory, the nearest initiator
may not necessarily be a CPU node. The performance path from the CXL
endpoint to the host bridge should remain the same. However, the numbers
extracted and stored from HMAT is the difference for the two access
classes. Split out the performance numbers for the host bridge (generic
target) from the calculation of the entire path in order to allow
calculation of both access classes for a CXL region.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
---
 drivers/cxl/core/cdat.c |   27 ++++++++++++++++++++++-----
 drivers/cxl/core/port.c |   34 +++++++++++++++++++++++++++++++---
 drivers/cxl/cxl.h       |    2 ++
 3 files changed, 55 insertions(+), 8 deletions(-)

diff --git a/drivers/cxl/core/cdat.c b/drivers/cxl/core/cdat.c
index 4d542627d02d..6e3998723aaa 100644
--- a/drivers/cxl/core/cdat.c
+++ b/drivers/cxl/core/cdat.c
@@ -161,7 +161,8 @@ static int cxl_cdat_endpoint_process(struct cxl_port *port,
 static int cxl_port_perf_data_calculate(struct cxl_port *port,
 					struct xarray *dsmas_xa)
 {
-	struct access_coordinate c;
+	struct access_coordinate ep_c;
+	struct access_coordinate coord[ACCESS_COORDINATE_MAX];
 	struct cxl_port *root_port;
 	struct cxl_root *cxl_root;
 	struct dsmas_entry *dent;
@@ -169,9 +170,15 @@ static int cxl_port_perf_data_calculate(struct cxl_port *port,
 	unsigned long index;
 	int rc;
 
-	rc = cxl_endpoint_get_perf_coordinates(port, &c);
+	rc = cxl_endpoint_get_perf_coordinates(port, &ep_c);
 	if (rc) {
-		dev_dbg(&port->dev, "Failed to retrieve perf coordinates.\n");
+		dev_dbg(&port->dev, "Failed to retrieve ep perf coordinates.\n");
+		return rc;
+	}
+
+	rc = cxl_hb_get_perf_coordinates(port, coord);
+	if (rc)  {
+		dev_dbg(&port->dev, "Failed to retrieve hb perf coordinates.\n");
 		return rc;
 	}
 
@@ -183,9 +190,19 @@ static int cxl_port_perf_data_calculate(struct cxl_port *port,
 	xa_for_each(dsmas_xa, index, dent) {
 		int qos_class;
 
-		cxl_coordinates_combine(&dent->coord, &dent->coord, &c);
+		cxl_coordinates_combine(&dent->coord, &dent->coord, &ep_c);
+		/*
+		 * Keeping the host bridge coordinates separate from the dsmas
+		 * coordinates in order to allow calculation of access class
+		 * 0 and 1 for region later.
+		 */
+		cxl_coordinates_combine(&coord[ACCESS_COORDINATE_LOCAL],
+					&coord[ACCESS_COORDINATE_LOCAL],
+					&dent->coord);
 		dent->entries = 1;
-		rc = cxl_root->ops->qos_class(root_port, &dent->coord, 1, &qos_class);
+		rc = cxl_root->ops->qos_class(root_port,
+					      &coord[ACCESS_COORDINATE_LOCAL],
+					      1, &qos_class);
 		if (rc != 1)
 			continue;
 
diff --git a/drivers/cxl/core/port.c b/drivers/cxl/core/port.c
index 1d3a04ef8b4f..0e4e65cd3806 100644
--- a/drivers/cxl/core/port.c
+++ b/drivers/cxl/core/port.c
@@ -2095,6 +2095,37 @@ bool schedule_cxl_memdev_detach(struct cxl_memdev *cxlmd)
 }
 EXPORT_SYMBOL_NS_GPL(schedule_cxl_memdev_detach, CXL);
 
+/**
+ * cxl_hb_get_perf_coordinates - Retrieve performance numbers from host bridge
+ *
+ * @port: endpoint cxl_port
+ * @coord: output access coordinates
+ *
+ * Return: errno on failure, 0 on success.
+ */
+int cxl_hb_get_perf_coordinates(struct cxl_port *port,
+				struct access_coordinate *coord)
+{
+	struct cxl_port *iter = port;
+	struct cxl_dport *dport;
+
+	if (!is_cxl_endpoint(port))
+		return -EINVAL;
+
+	dport = iter->parent_dport;
+	while (iter && !is_cxl_root(to_cxl_port(iter->dev.parent))) {
+		iter = to_cxl_port(iter->dev.parent);
+		dport = iter->parent_dport;
+	}
+
+	coord[ACCESS_COORDINATE_LOCAL] =
+		dport->hb_coord[ACCESS_COORDINATE_LOCAL];
+	coord[ACCESS_COORDINATE_CPU] =
+		dport->hb_coord[ACCESS_COORDINATE_CPU];
+
+	return 0;
+}
+
 /**
  * cxl_endpoint_get_perf_coordinates - Retrieve performance numbers stored in dports
  *				   of CXL path
@@ -2136,9 +2167,6 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port,
 		dport = iter->parent_dport;
 	}
 
-	/* Augment with the generic port (host bridge) perf data */
-	cxl_coordinates_combine(&c, &c, &dport->hb_coord[ACCESS_COORDINATE_LOCAL]);
-
 	/* Get the calculated PCI paths bandwidth */
 	pdev = to_pci_dev(port->uport_dev->parent);
 	bw = pcie_bandwidth_available(pdev, NULL, NULL, NULL);
diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
index 0a0a121ee780..80e6bd294e18 100644
--- a/drivers/cxl/cxl.h
+++ b/drivers/cxl/cxl.h
@@ -878,6 +878,8 @@ void cxl_switch_parse_cdat(struct cxl_port *port);
 
 int cxl_endpoint_get_perf_coordinates(struct cxl_port *port,
 				      struct access_coordinate *coord);
+int cxl_hb_get_perf_coordinates(struct cxl_port *port,
+				struct access_coordinate *coord);
 
 void cxl_coordinates_combine(struct access_coordinate *out,
 			     struct access_coordinate *c1,

[PATCH v4 07/11] cxl: Set cxlmd->endpoint before adding port device

[Dave Jiang]

Move setting of cxlmd->endpoint to before calling add_device() on the port
device. Otherwise when referencing cxlmd->endpoint in region discovery code
that is triggered by the port driver probe function, the endpoint port
pointer is not valid.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
---
 drivers/cxl/core/port.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/cxl/core/port.c b/drivers/cxl/core/port.c
index 0e4e65cd3806..f63cc5230269 100644
--- a/drivers/cxl/core/port.c
+++ b/drivers/cxl/core/port.c
@@ -826,6 +826,7 @@ static struct cxl_port *__devm_cxl_add_port(struct device *host,
 		 */
 		port->reg_map = cxlds->reg_map;
 		port->reg_map.host = &port->dev;
+		cxlmd->endpoint = port;
 	} else if (parent_dport) {
 		rc = dev_set_name(dev, "port%d", port->id);
 		if (rc)
@@ -1369,7 +1370,6 @@ int cxl_endpoint_autoremove(struct cxl_memdev *cxlmd, struct cxl_port *endpoint)
 
 	get_device(host);
 	get_device(&endpoint->dev);
-	cxlmd->endpoint = endpoint;
 	cxlmd->depth = endpoint->depth;
 	return devm_add_action_or_reset(dev, delete_endpoint, cxlmd);
 }

[PATCH v4 08/11] cxl/region: Calculate performance data for a region

[Dave Jiang]

Calculate and store the performance data for a CXL region. Find the worst
read and write latency for all the included ranges from each of the devices
that attributes to the region and designate that as the latency data. Sum
all the read and write bandwidth data for each of the device region and
that is the total bandwidth for the region.

The perf list is expected to be constructed before the endpoint decoders
are registered and thus there should be no early reading of the entries
from the region assemble action. The calling of the region qos calculate
function is under the protection of cxl_dpa_rwsem and will ensure that
all DPA associated work has completed.

Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
---
v4:
- Calculate access classes 0 and 1 by retrieving host bridge coords
- Add lockdep assert for cxl_dpa_rwsem (Dan)
- Clarify that HMAT code is HMEM_REPORTING code. (Dan)
---
 drivers/cxl/core/cdat.c   |   74 +++++++++++++++++++++++++++++++++++++++++++++
 drivers/cxl/core/region.c |    2 +
 drivers/cxl/cxl.h         |    4 ++
 3 files changed, 80 insertions(+)

diff --git a/drivers/cxl/core/cdat.c b/drivers/cxl/core/cdat.c
index 6e3998723aaa..7acb5837afad 100644
--- a/drivers/cxl/core/cdat.c
+++ b/drivers/cxl/core/cdat.c
@@ -8,6 +8,7 @@
 #include "cxlpci.h"
 #include "cxlmem.h"
 #include "cxl.h"
+#include "core.h"
 
 struct dsmas_entry {
 	struct range dpa_range;
@@ -546,3 +547,76 @@ void cxl_coordinates_combine(struct access_coordinate *out,
 EXPORT_SYMBOL_NS_GPL(cxl_coordinates_combine, CXL);
 
 MODULE_IMPORT_NS(CXL);
+
+void cxl_region_perf_data_calculate(struct cxl_region *cxlr,
+				    struct cxl_endpoint_decoder *cxled)
+{
+	struct cxl_memdev *cxlmd = cxled_to_memdev(cxled);
+	struct cxl_port *port = cxlmd->endpoint;
+	struct cxl_dev_state *cxlds = cxlmd->cxlds;
+	struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlds);
+	struct access_coordinate hb_coord[ACCESS_COORDINATE_MAX];
+	struct access_coordinate coord;
+	struct range dpa = {
+			.start = cxled->dpa_res->start,
+			.end = cxled->dpa_res->end,
+	};
+	struct list_head *perf_list;
+	struct cxl_dpa_perf *perf;
+	bool found = false;
+	int rc;
+
+	switch (cxlr->mode) {
+	case CXL_DECODER_RAM:
+		perf_list = &mds->ram_perf_list;
+		break;
+	case CXL_DECODER_PMEM:
+		perf_list = &mds->pmem_perf_list;
+		break;
+	default:
+		return;
+	}
+
+	lockdep_assert_held(&cxl_dpa_rwsem);
+
+	list_for_each_entry(perf, perf_list, list) {
+		if (range_contains(&perf->dpa_range, &dpa)) {
+			found = true;
+			break;
+		}
+	}
+
+	if (!found)
+		return;
+
+	rc = cxl_hb_get_perf_coordinates(port, hb_coord);
+	if (rc)  {
+		dev_dbg(&port->dev, "Failed to retrieve hb perf coordinates.\n");
+		return;
+	}
+
+	for (int i = 0; i < ACCESS_COORDINATE_MAX; i++) {
+		/* Pickup the host bridge coords */
+		cxl_coordinates_combine(&coord, &hb_coord[i], &perf->coord);
+
+		/* Get total bandwidth and the worst latency for the cxl region */
+		cxlr->coord[i].read_latency = max_t(unsigned int,
+						    cxlr->coord[i].read_latency,
+						    coord.read_latency);
+		cxlr->coord[i].write_latency = max_t(unsigned int,
+						     cxlr->coord[i].write_latency,
+						     coord.write_latency);
+		cxlr->coord[i].read_bandwidth += coord.read_bandwidth;
+		cxlr->coord[i].write_bandwidth += coord.write_bandwidth;
+
+		/*
+		 * Convert latency to nanosec from picosec to be consistent
+		 * with the resulting latency coordinates computed by the
+		 * HMAT_REPORTING code.
+		 */
+		cxlr->coord[i].read_latency =
+			DIV_ROUND_UP(cxlr->coord[i].read_latency, 1000);
+		cxlr->coord[i].write_latency =
+			DIV_ROUND_UP(cxlr->coord[i].write_latency, 1000);
+	}
+}
diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
index 57a5901d5a60..7f19b533c5ae 100644
--- a/drivers/cxl/core/region.c
+++ b/drivers/cxl/core/region.c
@@ -1722,6 +1722,8 @@ static int cxl_region_attach(struct cxl_region *cxlr,
 		return -EINVAL;
 	}
 
+	cxl_region_perf_data_calculate(cxlr, cxled);
+
 	if (test_bit(CXL_REGION_F_AUTO, &cxlr->flags)) {
 		int i;
 
diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
index 80e6bd294e18..f6637fa33113 100644
--- a/drivers/cxl/cxl.h
+++ b/drivers/cxl/cxl.h
@@ -519,6 +519,7 @@ struct cxl_region_params {
  * @cxlr_pmem: (for pmem regions) cached copy of the nvdimm bridge
  * @flags: Region state flags
  * @params: active + config params for the region
+ * @coord: QoS access coordinates for the region
  */
 struct cxl_region {
 	struct device dev;
@@ -529,6 +530,7 @@ struct cxl_region {
 	struct cxl_pmem_region *cxlr_pmem;
 	unsigned long flags;
 	struct cxl_region_params params;
+	struct access_coordinate coord[ACCESS_COORDINATE_MAX];
 };
 
 struct cxl_nvdimm_bridge {
@@ -880,6 +882,8 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port,
 				      struct access_coordinate *coord);
 int cxl_hb_get_perf_coordinates(struct cxl_port *port,
 				struct access_coordinate *coord);
+void cxl_region_perf_data_calculate(struct cxl_region *cxlr,
+				    struct cxl_endpoint_decoder *cxled);
 
 void cxl_coordinates_combine(struct access_coordinate *out,
 			     struct access_coordinate *c1,

[PATCH v4 09/11] cxl/region: Add sysfs attribute for locality attributes of CXL regions

[Dave Jiang]

Add read/write latencies and bandwidth sysfs attributes for the enabled CXL
region. The bandwidth is the aggregated bandwidth of all devices that
contribute to the CXL region. The latency is the worst latency of the
device amongst all the devices that contribute to the CXL region.

Signed-off-by: Dave Jiang <dave.jiang@intel.com>
---
v4:
- Added support for access0 and access1 attributes
- Updated documentation for access levels
- Update verbiage of explanation. (Dan)
---
 Documentation/ABI/testing/sysfs-bus-cxl |   60 ++++++++++++++
 drivers/cxl/core/region.c               |  134 +++++++++++++++++++++++++++++++
 2 files changed, 194 insertions(+)

diff --git a/Documentation/ABI/testing/sysfs-bus-cxl b/Documentation/ABI/testing/sysfs-bus-cxl
index fff2581b8033..5f8c26815399 100644
--- a/Documentation/ABI/testing/sysfs-bus-cxl
+++ b/Documentation/ABI/testing/sysfs-bus-cxl
@@ -552,3 +552,63 @@ Description:
 		attribute is only visible for devices supporting the
 		capability. The retrieved errors are logged as kernel
 		events when cxl_poison event tracing is enabled.
+
+
+What:		/sys/bus/cxl/devices/regionZ/accessY/read_bandwidth
+Date:		Jan, 2024
+KernelVersion:	v6.9
+Contact:	linux-cxl@vger.kernel.org
+Description:
+		(RO) The aggregated read bandwidth of the region. The number is
+		the accumulated read bandwidth of all CXL memory devices that
+		contributes to the region in MB/s. It is identical data that
+		should appear in
+		/sys/devices/system/node/nodeX/accessY/initiators/read_bandwidth.
+		See Documentation/ABI/stable/sysfs-devices-node. access0 provides
+		the number to the closest initiator and access1 provides the
+		number to the closest CPU.
+
+
+What:		/sys/bus/cxl/devices/regionZ/accessY/write_bandwidth
+Date:		Jan, 2024
+KernelVersion:	v6.9
+Contact:	linux-cxl@vger.kernel.org
+Description:
+		(RO) The aggregated write bandwidth of the region. The number is
+		the accumulated write bandwidth of all CXL memory devices that
+		contributes to the region in MB/s. It is identical data that
+		should appear in
+		/sys/devices/system/node/nodeX/accessY/initiators/write_bandwidth.
+		See Documentation/ABI/stable/sysfs-devices-node. access0 provides
+		the number to the closest initiator and access1 provides the
+		number to the closest CPU.
+
+
+What:		/sys/bus/cxl/devices/regionZ/accessY/read_latency
+Date:		Jan, 2024
+KernelVersion:	v6.9
+Contact:	linux-cxl@vger.kernel.org
+Description:
+		(RO) The read latency of the region. The number is
+		the worst read latency of all CXL memory devices that
+		contributes to the region in nanoseconds. It is identical data
+		that should appear in
+		/sys/devices/system/node/nodeX/accessY/initiators/read_latency.
+		See Documentation/ABI/stable/sysfs-devices-node. access0 provides
+		the number to the closest initiator and access1 provides the
+		number to the closest CPU.
+
+
+What:		/sys/bus/cxl/devices/regionZ/accessY/write_latency
+Date:		Jan, 2024
+KernelVersion:	v6.9
+Contact:	linux-cxl@vger.kernel.org
+Description:
+		(RO) The write latency of the region. The number is
+		the worst write latency of all CXL memory devices that
+		contributes to the region in nanoseconds. It is identical data
+		that should appear in
+		/sys/devices/system/node/nodeX/accessY/initiators/write_latency.
+		See Documentation/ABI/stable/sysfs-devices-node. access0 provides
+		the number to the closest initiator and access1 provides the
+		number to the closest CPU.
diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
index 7f19b533c5ae..978d1707964a 100644
--- a/drivers/cxl/core/region.c
+++ b/drivers/cxl/core/region.c
@@ -30,6 +30,138 @@
 
 static struct cxl_region *to_cxl_region(struct device *dev);
 
+#define __ACCESS0_ATTR_RO(_name) {				\
+	.attr	= { .name = __stringify(_name), .mode = 0444 },	\
+	.show	= _name##_access0_show,				\
+}
+
+#define ACCESS0_DEVICE_ATTR_RO(_name) \
+	struct device_attribute dev_attr_access0_##_name = __ACCESS0_ATTR_RO(_name)
+
+#define ACCESS0_ATTR(attrib)					\
+static ssize_t attrib##_access0_show(struct device *dev,	\
+			   struct device_attribute *attr,	\
+			   char *buf)				\
+{								\
+	struct cxl_region *cxlr = to_cxl_region(dev);		\
+								\
+	if (cxlr->coord[0].attrib == 0)				\
+		return -ENOENT;					\
+								\
+	return sysfs_emit(buf, "%u\n", cxlr->coord[0].attrib);	\
+}								\
+static ACCESS0_DEVICE_ATTR_RO(attrib)
+
+ACCESS0_ATTR(read_bandwidth);
+ACCESS0_ATTR(read_latency);
+ACCESS0_ATTR(write_bandwidth);
+ACCESS0_ATTR(write_latency);
+
+static struct attribute *access0_coordinate_attrs[] = {
+	&dev_attr_access0_read_bandwidth.attr,
+	&dev_attr_access0_write_bandwidth.attr,
+	&dev_attr_access0_read_latency.attr,
+	&dev_attr_access0_write_latency.attr,
+	NULL,
+};
+
+static umode_t cxl_region_access0_coordinate_visible(struct kobject *kobj,
+						     struct attribute *a, int n)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct cxl_region *cxlr = to_cxl_region(dev);
+
+	if (a == &dev_attr_access0_read_latency.attr &&
+	    cxlr->coord[ACCESS_COORDINATE_LOCAL].read_latency == 0)
+		return 0;
+
+	if (a == &dev_attr_access0_write_latency.attr &&
+	    cxlr->coord[ACCESS_COORDINATE_LOCAL].write_latency == 0)
+		return 0;
+
+	if (a == &dev_attr_access0_read_bandwidth.attr &&
+	    cxlr->coord[ACCESS_COORDINATE_LOCAL].read_bandwidth == 0)
+		return 0;
+
+	if (a == &dev_attr_access0_write_bandwidth.attr &&
+	    cxlr->coord[ACCESS_COORDINATE_LOCAL].write_bandwidth == 0)
+		return 0;
+
+	return a->mode;
+}
+
+#define __ACCESS1_ATTR_RO(_name) {				\
+	.attr	= { .name = __stringify(_name), .mode = 0444 },	\
+	.show	= _name##_access1_show,				\
+}
+
+#define ACCESS1_DEVICE_ATTR_RO(_name) \
+	struct device_attribute dev_attr_access1_##_name = __ACCESS1_ATTR_RO(_name)
+
+#define ACCESS1_ATTR(attrib)					\
+static ssize_t attrib##_access1_show(struct device *dev,	\
+			   struct device_attribute *attr,	\
+			   char *buf)				\
+{								\
+	struct cxl_region *cxlr = to_cxl_region(dev);		\
+								\
+	if (cxlr->coord[1].attrib == 0)				\
+		return -ENOENT;					\
+								\
+	return sysfs_emit(buf, "%u\n", cxlr->coord[1].attrib);	\
+}								\
+static ACCESS1_DEVICE_ATTR_RO(attrib)
+
+ACCESS1_ATTR(read_bandwidth);
+ACCESS1_ATTR(read_latency);
+ACCESS1_ATTR(write_bandwidth);
+ACCESS1_ATTR(write_latency);
+
+static struct attribute *access1_coordinate_attrs[] = {
+	&dev_attr_access1_read_bandwidth.attr,
+	&dev_attr_access1_write_bandwidth.attr,
+	&dev_attr_access1_read_latency.attr,
+	&dev_attr_access1_write_latency.attr,
+	NULL,
+};
+
+static umode_t cxl_region_access1_coordinate_visible(struct kobject *kobj,
+						     struct attribute *a, int n)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct cxl_region *cxlr = to_cxl_region(dev);
+
+	if (a == &dev_attr_access1_read_latency.attr &&
+	    cxlr->coord[ACCESS_COORDINATE_CPU].read_latency == 0)
+		return 0;
+
+	if (a == &dev_attr_access1_write_latency.attr &&
+	    cxlr->coord[ACCESS_COORDINATE_CPU].write_latency == 0)
+		return 0;
+
+	if (a == &dev_attr_access1_read_bandwidth.attr &&
+	    cxlr->coord[ACCESS_COORDINATE_CPU].read_bandwidth == 0)
+		return 0;
+
+	if (a == &dev_attr_access1_write_bandwidth.attr &&
+	    cxlr->coord[ACCESS_COORDINATE_CPU].write_bandwidth == 0)
+		return 0;
+
+	return a->mode;
+}
+
+static const struct attribute_group cxl_region_access0_coordinate_group = {
+	.name = "access0",
+	.attrs = access0_coordinate_attrs,
+	.is_visible = cxl_region_access0_coordinate_visible,
+};
+
+static const struct attribute_group cxl_region_access1_coordinate_group = {
+	.name = "access1",
+	.attrs = access1_coordinate_attrs,
+	.is_visible = cxl_region_access1_coordinate_visible,
+};
+
 static ssize_t uuid_show(struct device *dev, struct device_attribute *attr,
 			 char *buf)
 {
@@ -2039,6 +2171,8 @@ static const struct attribute_group *region_groups[] = {
 	&cxl_base_attribute_group,
 	&cxl_region_group,
 	&cxl_region_target_group,
+	&cxl_region_access0_coordinate_group,
+	&cxl_region_access1_coordinate_group,
 	NULL,
 };
 

[PATCH v4 10/11] cxl: Add memory hotplug notifier for cxl region

[Dave Jiang]

When the CXL region is formed, the driver would computed the performance
data for the region. However this data is not available at the node data
collection that has been populated by the HMAT during kernel
initialization. Add a memory hotplug notifier to update the access
coordinates to the 'struct memory_target' context kept by the
HMAT_REPORTING code.

Add CXL_CALLBACK_PRI for a memory hotplug callback priority. Set the
priority number to be called before HMAT_CALLBACK_PRI. The CXL update must
happen before hmat_callback().

A new HMAT_REPORING helper hmat_update_target_coordinates() is added in
order to allow CXL to update the memory_target access coordinates.

A new ext_updated member is added to the memory_target to indicate that
the access coordinates within the memory_target has been updated by an
external agent such as CXL. This prevents data being overwritten by the
hmat_update_target_attrs() triggered by hmat_callback().

Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Rafael J. Wysocki <rafael@kernel.org>
Reviewed-by: "Huang, Ying" <ying.huang@intel.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
---
v4:
- Update access_coordinates via HMAT_REPORING code and let hmat_callback()
  notifier to expose the new numbers to sysfs.
- Introduce CXL_CALLBACK_PRI and remove HMAT language (Dan)
---
 drivers/acpi/numa/hmat.c  |   36 ++++++++++++++++++++++
 drivers/cxl/core/cdat.c   |    6 ++++
 drivers/cxl/core/core.h   |    3 ++
 drivers/cxl/core/region.c |   75 +++++++++++++++++++++++++++++++++++++++++++++
 drivers/cxl/cxl.h         |    3 ++
 include/linux/acpi.h      |    9 +++++
 include/linux/memory.h    |    1 +
 7 files changed, 133 insertions(+)

diff --git a/drivers/acpi/numa/hmat.c b/drivers/acpi/numa/hmat.c
index 5be79896bd08..dc75a3355164 100644
--- a/drivers/acpi/numa/hmat.c
+++ b/drivers/acpi/numa/hmat.c
@@ -74,6 +74,7 @@ struct memory_target {
 	struct node_cache_attrs cache_attrs;
 	u8 gen_port_device_handle[ACPI_SRAT_DEVICE_HANDLE_SIZE];
 	bool registered;
+	bool ext_updated;	/* externally updated */
 };
 
 struct memory_initiator {
@@ -328,6 +329,35 @@ static void hmat_update_target_access(struct memory_target *target,
 	}
 }
 
+int hmat_update_target_coordinates(int nid, struct access_coordinate *coord,
+				   enum access_coordinate_class access)
+{
+	struct memory_target *target;
+	int pxm;
+
+	if (nid == NUMA_NO_NODE)
+		return -EINVAL;
+
+	pxm = node_to_pxm(nid);
+	guard(mutex)(&target_lock);
+	target = find_mem_target(pxm);
+	if (!target)
+		return -ENODEV;
+
+	hmat_update_target_access(target, ACPI_HMAT_READ_LATENCY,
+				  coord->read_latency, access);
+	hmat_update_target_access(target, ACPI_HMAT_WRITE_LATENCY,
+				  coord->write_latency, access);
+	hmat_update_target_access(target, ACPI_HMAT_READ_BANDWIDTH,
+				  coord->read_bandwidth, access);
+	hmat_update_target_access(target, ACPI_HMAT_WRITE_BANDWIDTH,
+				  coord->write_bandwidth, access);
+	target->ext_updated = true;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(hmat_update_target_coordinates);
+
 static __init void hmat_add_locality(struct acpi_hmat_locality *hmat_loc)
 {
 	struct memory_locality *loc;
@@ -699,6 +729,12 @@ static void hmat_update_target_attrs(struct memory_target *target,
 	u32 best = 0;
 	int i;
 
+	/*
+	 * Don't update if an external agent has changed the data.
+	 */
+	if (target->ext_updated)
+		return;
+
 	/* Don't update for generic port if there's no device handle */
 	if ((access == NODE_ACCESS_CLASS_GENPORT_SINK_LOCAL ||
 	     access == NODE_ACCESS_CLASS_GENPORT_SINK_CPU) &&
diff --git a/drivers/cxl/core/cdat.c b/drivers/cxl/core/cdat.c
index 7acb5837afad..3556c897ece4 100644
--- a/drivers/cxl/core/cdat.c
+++ b/drivers/cxl/core/cdat.c
@@ -620,3 +620,9 @@ void cxl_region_perf_data_calculate(struct cxl_region *cxlr,
 			DIV_ROUND_UP(cxlr->coord[i].write_latency, 1000);
 	}
 }
+
+int cxl_update_hmat_access_coordinates(int nid, struct cxl_region *cxlr,
+				       enum access_coordinate_class access)
+{
+	return hmat_update_target_coordinates(nid, &cxlr->coord[access], access);
+}
diff --git a/drivers/cxl/core/core.h b/drivers/cxl/core/core.h
index 3b64fb1b9ed0..e19800a7ce06 100644
--- a/drivers/cxl/core/core.h
+++ b/drivers/cxl/core/core.h
@@ -90,4 +90,7 @@ enum cxl_poison_trace_type {
 
 long cxl_pci_get_latency(struct pci_dev *pdev);
 
+int cxl_update_hmat_access_coordinates(int nid, struct cxl_region *cxlr,
+				       enum access_coordinate_class access);
+
 #endif /* __CXL_CORE_H__ */
diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
index 978d1707964a..ae1f34e1cd05 100644
--- a/drivers/cxl/core/region.c
+++ b/drivers/cxl/core/region.c
@@ -4,6 +4,7 @@
 #include <linux/genalloc.h>
 #include <linux/device.h>
 #include <linux/module.h>
+#include <linux/memory.h>
 #include <linux/slab.h>
 #include <linux/uuid.h>
 #include <linux/sort.h>
@@ -156,12 +157,22 @@ static const struct attribute_group cxl_region_access0_coordinate_group = {
 	.is_visible = cxl_region_access0_coordinate_visible,
 };
 
+static const struct attribute_group *get_cxl_region_access0_group(void)
+{
+	return &cxl_region_access0_coordinate_group;
+}
+
 static const struct attribute_group cxl_region_access1_coordinate_group = {
 	.name = "access1",
 	.attrs = access1_coordinate_attrs,
 	.is_visible = cxl_region_access1_coordinate_visible,
 };
 
+static const struct attribute_group *get_cxl_region_access1_group(void)
+{
+	return &cxl_region_access1_coordinate_group;
+}
+
 static ssize_t uuid_show(struct device *dev, struct device_attribute *attr,
 			 char *buf)
 {
@@ -3066,6 +3077,65 @@ static int is_system_ram(struct resource *res, void *arg)
 	return 1;
 }
 
+static bool cxl_region_update_coordinates(struct cxl_region *cxlr, int nid)
+{
+	int cset = 0;
+	int rc;
+
+	for (int i = 0; i < ACCESS_COORDINATE_MAX; i++) {
+		if (cxlr->coord[i].read_bandwidth) {
+			rc = cxl_update_hmat_access_coordinates(nid, cxlr, i);
+			if (rc == 0)
+				cset++;
+		}
+	}
+
+	if (!cset)
+		return false;
+
+	rc = sysfs_update_group(&cxlr->dev.kobj, get_cxl_region_access0_group());
+	if (rc)
+		dev_dbg(&cxlr->dev, "Failed to update access0 group\n");
+
+	rc = sysfs_update_group(&cxlr->dev.kobj, get_cxl_region_access1_group());
+	if (rc)
+		dev_dbg(&cxlr->dev, "Failed to update access1 group\n");
+
+	return true;
+}
+
+static int cxl_region_perf_attrs_callback(struct notifier_block *nb,
+					  unsigned long action, void *arg)
+{
+	struct cxl_region *cxlr = container_of(nb, struct cxl_region,
+					       memory_notifier);
+	struct cxl_region_params *p = &cxlr->params;
+	struct cxl_endpoint_decoder *cxled = p->targets[0];
+	struct cxl_decoder *cxld = &cxled->cxld;
+	struct memory_notify *mnb = arg;
+	int nid = mnb->status_change_nid;
+	int region_nid;
+
+	if (nid == NUMA_NO_NODE || action != MEM_ONLINE)
+		return NOTIFY_DONE;
+
+	region_nid = phys_to_target_node(cxld->hpa_range.start);
+	if (nid != region_nid)
+		return NOTIFY_DONE;
+
+	if (!cxl_region_update_coordinates(cxlr, nid))
+		return NOTIFY_DONE;
+
+	return NOTIFY_OK;
+}
+
+static void remove_coord_notifier(void *data)
+{
+	struct cxl_region *cxlr = data;
+
+	unregister_memory_notifier(&cxlr->memory_notifier);
+}
+
 static int cxl_region_probe(struct device *dev)
 {
 	struct cxl_region *cxlr = to_cxl_region(dev);
@@ -3091,6 +3161,11 @@ static int cxl_region_probe(struct device *dev)
 		goto out;
 	}
 
+	cxlr->memory_notifier.notifier_call = cxl_region_perf_attrs_callback;
+	cxlr->memory_notifier.priority = CXL_CALLBACK_PRI;
+	register_memory_notifier(&cxlr->memory_notifier);
+	rc = devm_add_action_or_reset(&cxlr->dev, remove_coord_notifier, cxlr);
+
 	/*
 	 * From this point on any path that changes the region's state away from
 	 * CXL_CONFIG_COMMIT is also responsible for releasing the driver.
diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
index f6637fa33113..556348c8835e 100644
--- a/drivers/cxl/cxl.h
+++ b/drivers/cxl/cxl.h
@@ -6,6 +6,7 @@
 
 #include <linux/libnvdimm.h>
 #include <linux/bitfield.h>
+#include <linux/notifier.h>
 #include <linux/bitops.h>
 #include <linux/log2.h>
 #include <linux/node.h>
@@ -520,6 +521,7 @@ struct cxl_region_params {
  * @flags: Region state flags
  * @params: active + config params for the region
  * @coord: QoS access coordinates for the region
+ * @memory_notifier: notifier for setting the access coordinates to node
  */
 struct cxl_region {
 	struct device dev;
@@ -531,6 +533,7 @@ struct cxl_region {
 	unsigned long flags;
 	struct cxl_region_params params;
 	struct access_coordinate coord[ACCESS_COORDINATE_MAX];
+	struct notifier_block memory_notifier;
 };
 
 struct cxl_nvdimm_bridge {
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index 8b0761c682f9..1c664948b2ae 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -427,12 +427,21 @@ extern bool acpi_osi_is_win8(void);
 
 #ifdef CONFIG_ACPI_HMAT
 int acpi_get_genport_coordinates(u32 uid, struct access_coordinate *coord);
+int hmat_update_target_coordinates(int nid, struct access_coordinate *coord,
+				   enum access_coordinate_class access);
 #else
 static inline int acpi_get_genport_coordinates(u32 uid,
 					       struct access_coordinate *coord)
 {
 	return -EOPNOTSUPP;
 }
+
+static inline int hmat_update_target_coordinates(int nid,
+						 struct access_coordinate *coord,
+						 enum access_coordinate_class access)
+{
+	return -EOPNOTSUPP;
+}
 #endif
 
 #ifdef CONFIG_ACPI_NUMA
diff --git a/include/linux/memory.h b/include/linux/memory.h
index f53cfdaaaa41..d8588256578a 100644
--- a/include/linux/memory.h
+++ b/include/linux/memory.h
@@ -114,6 +114,7 @@ struct mem_section;
 #define DEFAULT_CALLBACK_PRI	0
 #define SLAB_CALLBACK_PRI	1
 #define HMAT_CALLBACK_PRI	2
+#define CXL_CALLBACK_PRI	5
 #define MM_COMPUTE_BATCH_PRI	10
 #define CPUSET_CALLBACK_PRI	10
 #define MEMTIER_HOTPLUG_PRI	100

[PATCH v4 11/11] cxl: Deal with numa nodes not enumarated by SRAT

[Dave Jiang]

For the numa nodes that are not created by SRAT, no memory_target is
allocated and is not managed by the HMAT_REPORTING code. Therefore
hmat_callback() memory hotplug notifier will exit early on those NUMA
nodes. The CXL memory hotplug notifier will need to call
node_set_perf_attrs() directly in order to setup the access sysfs
attributes.

In acpi_numa_init(), the last proximity domain (pxm) id created by SRAT is
stored. Add a helper function acpi_node_backed_by_real_pxm() in order to
check if a NUMA node id is defined by SRAT or created by CFMWS or some
other methods.

node_set_perf_attrs() symbol is exported to allow update of perf attribs
for a node. The sysfs path of
/sys/devices/system/node/nodeX/access0/initiators/* is created by
ndoe_set_perf_attrs() for the various attributes where nodeX is matched
to the NUMA node of the CXL region.

Cc: Rafael J. Wysocki <rafael@kernel.org>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
---
 drivers/acpi/numa/srat.c  |   11 +++++++++++
 drivers/base/node.c       |    1 +
 drivers/cxl/core/cdat.c   |    5 +++++
 drivers/cxl/core/core.h   |    1 +
 drivers/cxl/core/region.c |    7 ++++++-
 include/linux/acpi.h      |    1 +
 6 files changed, 25 insertions(+), 1 deletion(-)

diff --git a/drivers/acpi/numa/srat.c b/drivers/acpi/numa/srat.c
index 12f330b0eac0..2f6f15b3891d 100644
--- a/drivers/acpi/numa/srat.c
+++ b/drivers/acpi/numa/srat.c
@@ -29,6 +29,8 @@ static int node_to_pxm_map[MAX_NUMNODES]
 unsigned char acpi_srat_revision __initdata;
 static int acpi_numa __initdata;
 
+static int last_real_pxm;
+
 void __init disable_srat(void)
 {
 	acpi_numa = -1;
@@ -536,6 +538,7 @@ int __init acpi_numa_init(void)
 		if (node_to_pxm_map[i] > fake_pxm)
 			fake_pxm = node_to_pxm_map[i];
 	}
+	last_real_pxm = fake_pxm;
 	fake_pxm++;
 	acpi_table_parse_cedt(ACPI_CEDT_TYPE_CFMWS, acpi_parse_cfmws,
 			      &fake_pxm);
@@ -547,6 +550,14 @@ int __init acpi_numa_init(void)
 	return 0;
 }
 
+bool acpi_node_backed_by_real_pxm(int nid)
+{
+	int pxm = node_to_pxm(nid);
+
+	return pxm <= last_real_pxm;
+}
+EXPORT_SYMBOL_GPL(acpi_node_backed_by_real_pxm);
+
 static int acpi_get_pxm(acpi_handle h)
 {
 	unsigned long long pxm;
diff --git a/drivers/base/node.c b/drivers/base/node.c
index b4a449f07f2a..8d0b09769b77 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -215,6 +215,7 @@ void node_set_perf_attrs(unsigned int nid, struct access_coordinate *coord,
 		}
 	}
 }
+EXPORT_SYMBOL_GPL(node_set_perf_attrs);
 
 /**
  * struct node_cache_info - Internal tracking for memory node caches
diff --git a/drivers/cxl/core/cdat.c b/drivers/cxl/core/cdat.c
index 3556c897ece4..7d7163f999e8 100644
--- a/drivers/cxl/core/cdat.c
+++ b/drivers/cxl/core/cdat.c
@@ -626,3 +626,8 @@ int cxl_update_hmat_access_coordinates(int nid, struct cxl_region *cxlr,
 {
 	return hmat_update_target_coordinates(nid, &cxlr->coord[access], access);
 }
+
+bool cxl_need_node_perf_attrs_update(int nid)
+{
+	return !acpi_node_backed_by_real_pxm(nid);
+}
diff --git a/drivers/cxl/core/core.h b/drivers/cxl/core/core.h
index e19800a7ce06..bc5a95665aa0 100644
--- a/drivers/cxl/core/core.h
+++ b/drivers/cxl/core/core.h
@@ -92,5 +92,6 @@ long cxl_pci_get_latency(struct pci_dev *pdev);
 
 int cxl_update_hmat_access_coordinates(int nid, struct cxl_region *cxlr,
 				       enum access_coordinate_class access);
+bool cxl_need_node_perf_attrs_update(int nid);
 
 #endif /* __CXL_CORE_H__ */
diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
index ae1f34e1cd05..66f126067bda 100644
--- a/drivers/cxl/core/region.c
+++ b/drivers/cxl/core/region.c
@@ -3084,7 +3084,12 @@ static bool cxl_region_update_coordinates(struct cxl_region *cxlr, int nid)
 
 	for (int i = 0; i < ACCESS_COORDINATE_MAX; i++) {
 		if (cxlr->coord[i].read_bandwidth) {
-			rc = cxl_update_hmat_access_coordinates(nid, cxlr, i);
+			rc = 0;
+			if (cxl_need_node_perf_attrs_update(nid))
+				node_set_perf_attrs(nid, &cxlr->coord[i], i);
+			else
+				rc = cxl_update_hmat_access_coordinates(nid, cxlr, i);
+
 			if (rc == 0)
 				cset++;
 		}
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index 1c664948b2ae..3067c6aad431 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -447,6 +447,7 @@ static inline int hmat_update_target_coordinates(int nid,
 #ifdef CONFIG_ACPI_NUMA
 int acpi_map_pxm_to_node(int pxm);
 int acpi_get_node(acpi_handle handle);
+bool acpi_node_backed_by_real_pxm(int nid);
 
 /**
  * pxm_to_online_node - Map proximity ID to online node

RE: [PATCH v4 05/11] cxl: Split out combine_coordinates() for common shared usage

[Dan Williams]

Dave Jiang wrote:
> Refactor the common code of combining coordinates in order to reduce code.
> Create a new function cxl_cooordinates_combine() it combine two 'struct
> access_coordinate'.
> 
> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
> ---
>  drivers/cxl/core/cdat.c |   32 +++++++++++++++++++++++---------
>  drivers/cxl/core/port.c |   18 ++----------------
>  drivers/cxl/cxl.h       |    4 ++++
>  3 files changed, 29 insertions(+), 25 deletions(-)
> 
> diff --git a/drivers/cxl/core/cdat.c b/drivers/cxl/core/cdat.c
> index cd84d87f597a..4d542627d02d 100644
> --- a/drivers/cxl/core/cdat.c
> +++ b/drivers/cxl/core/cdat.c
> @@ -183,15 +183,7 @@ static int cxl_port_perf_data_calculate(struct cxl_port *port,
>  	xa_for_each(dsmas_xa, index, dent) {
>  		int qos_class;
>  
> -		dent->coord.read_latency = dent->coord.read_latency +
> -					   c.read_latency;
> -		dent->coord.write_latency = dent->coord.write_latency +
> -					    c.write_latency;
> -		dent->coord.read_bandwidth = min_t(int, c.read_bandwidth,
> -						   dent->coord.read_bandwidth);
> -		dent->coord.write_bandwidth = min_t(int, c.write_bandwidth,
> -						    dent->coord.write_bandwidth);
> -
> +		cxl_coordinates_combine(&dent->coord, &dent->coord, &c);
>  		dent->entries = 1;
>  		rc = cxl_root->ops->qos_class(root_port, &dent->coord, 1, &qos_class);
>  		if (rc != 1)
> @@ -514,4 +506,26 @@ void cxl_switch_parse_cdat(struct cxl_port *port)
>  }
>  EXPORT_SYMBOL_NS_GPL(cxl_switch_parse_cdat, CXL);
>  
> +/**
> + * cxl_coordinates_combine - Combine the two input coordinates into the first
> + *
> + * @c1: first coordinate, to be written to
> + * @c2: second coordinate
> + */
> +void cxl_coordinates_combine(struct access_coordinate *out,
> +			     struct access_coordinate *c1,
> +			     struct access_coordinate *c2)
> +{
> +		if (c2->write_bandwidth)
> +			out->write_bandwidth = min(c1->write_bandwidth,
> +						   c2->write_bandwidth);
> +		out->write_latency = c1->write_latency + c2->write_latency;
> +
> +		if (c2->read_bandwidth)
> +			out->read_bandwidth = min(c1->read_bandwidth,
> +						  c2->read_bandwidth);
> +		out->read_latency = c1->read_latency + c2->read_latency;
> +}
> +EXPORT_SYMBOL_NS_GPL(cxl_coordinates_combine, CXL);

There is no need for EXPORT_SYMBOL() when the definition and the only
caller exist within the same compilation unit, cxl_core.o.

However, given there is nothing "CXL" about this function it likely wants
to move out of cxl_core.o if another caller ever arrives.

Re: [PATCH v4 11/11] cxl: Deal with numa nodes not enumarated by SRAT

[Alison Schofield]

On Fri, Jan 19, 2024 at 10:24:11AM -0700, Dave Jiang wrote:
> For the numa nodes that are not created by SRAT, no memory_target is
> allocated and is not managed by the HMAT_REPORTING code. Therefore
> hmat_callback() memory hotplug notifier will exit early on those NUMA
> nodes. The CXL memory hotplug notifier will need to call
> node_set_perf_attrs() directly in order to setup the access sysfs
> attributes.
> 
> In acpi_numa_init(), the last proximity domain (pxm) id created by SRAT is
> stored. Add a helper function acpi_node_backed_by_real_pxm() in order to
> check if a NUMA node id is defined by SRAT or created by CFMWS or some
> other methods.

I'm thinking the 'or some other methods' can be dropped. In chat,
we mentioned emulated nodes, but they don't make PXM assignments.
Maybe I misunderstand, but I thought NUMA emulation can only be
enabled when there is no physical NUMA architecture.

Aside from clearing up the emulated or other nodes story...LGTM.

Reviewed-by: Alison Schofield <alison.schofield@intel.com>

> 
> node_set_perf_attrs() symbol is exported to allow update of perf attribs
> for a node. The sysfs path of
> /sys/devices/system/node/nodeX/access0/initiators/* is created by
> ndoe_set_perf_attrs() for the various attributes where nodeX is matched
> to the NUMA node of the CXL region.
> 
> Cc: Rafael J. Wysocki <rafael@kernel.org>
> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
> ---
>  drivers/acpi/numa/srat.c  |   11 +++++++++++
>  drivers/base/node.c       |    1 +
>  drivers/cxl/core/cdat.c   |    5 +++++
>  drivers/cxl/core/core.h   |    1 +
>  drivers/cxl/core/region.c |    7 ++++++-
>  include/linux/acpi.h      |    1 +
>  6 files changed, 25 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/acpi/numa/srat.c b/drivers/acpi/numa/srat.c
> index 12f330b0eac0..2f6f15b3891d 100644
> --- a/drivers/acpi/numa/srat.c
> +++ b/drivers/acpi/numa/srat.c
> @@ -29,6 +29,8 @@ static int node_to_pxm_map[MAX_NUMNODES]
>  unsigned char acpi_srat_revision __initdata;
>  static int acpi_numa __initdata;
>  
> +static int last_real_pxm;
> +
>  void __init disable_srat(void)
>  {
>  	acpi_numa = -1;
> @@ -536,6 +538,7 @@ int __init acpi_numa_init(void)
>  		if (node_to_pxm_map[i] > fake_pxm)
>  			fake_pxm = node_to_pxm_map[i];
>  	}
> +	last_real_pxm = fake_pxm;
>  	fake_pxm++;
>  	acpi_table_parse_cedt(ACPI_CEDT_TYPE_CFMWS, acpi_parse_cfmws,
>  			      &fake_pxm);
> @@ -547,6 +550,14 @@ int __init acpi_numa_init(void)
>  	return 0;
>  }
>  
> +bool acpi_node_backed_by_real_pxm(int nid)
> +{
> +	int pxm = node_to_pxm(nid);
> +
> +	return pxm <= last_real_pxm;
> +}
> +EXPORT_SYMBOL_GPL(acpi_node_backed_by_real_pxm);
> +
>  static int acpi_get_pxm(acpi_handle h)
>  {
>  	unsigned long long pxm;
> diff --git a/drivers/base/node.c b/drivers/base/node.c
> index b4a449f07f2a..8d0b09769b77 100644
> --- a/drivers/base/node.c
> +++ b/drivers/base/node.c
> @@ -215,6 +215,7 @@ void node_set_perf_attrs(unsigned int nid, struct access_coordinate *coord,
>  		}
>  	}
>  }
> +EXPORT_SYMBOL_GPL(node_set_perf_attrs);
>  
>  /**
>   * struct node_cache_info - Internal tracking for memory node caches
> diff --git a/drivers/cxl/core/cdat.c b/drivers/cxl/core/cdat.c
> index 3556c897ece4..7d7163f999e8 100644
> --- a/drivers/cxl/core/cdat.c
> +++ b/drivers/cxl/core/cdat.c
> @@ -626,3 +626,8 @@ int cxl_update_hmat_access_coordinates(int nid, struct cxl_region *cxlr,
>  {
>  	return hmat_update_target_coordinates(nid, &cxlr->coord[access], access);
>  }
> +
> +bool cxl_need_node_perf_attrs_update(int nid)
> +{
> +	return !acpi_node_backed_by_real_pxm(nid);
> +}
> diff --git a/drivers/cxl/core/core.h b/drivers/cxl/core/core.h
> index e19800a7ce06..bc5a95665aa0 100644
> --- a/drivers/cxl/core/core.h
> +++ b/drivers/cxl/core/core.h
> @@ -92,5 +92,6 @@ long cxl_pci_get_latency(struct pci_dev *pdev);
>  
>  int cxl_update_hmat_access_coordinates(int nid, struct cxl_region *cxlr,
>  				       enum access_coordinate_class access);
> +bool cxl_need_node_perf_attrs_update(int nid);
>  
>  #endif /* __CXL_CORE_H__ */
> diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
> index ae1f34e1cd05..66f126067bda 100644
> --- a/drivers/cxl/core/region.c
> +++ b/drivers/cxl/core/region.c
> @@ -3084,7 +3084,12 @@ static bool cxl_region_update_coordinates(struct cxl_region *cxlr, int nid)
>  
>  	for (int i = 0; i < ACCESS_COORDINATE_MAX; i++) {
>  		if (cxlr->coord[i].read_bandwidth) {
> -			rc = cxl_update_hmat_access_coordinates(nid, cxlr, i);
> +			rc = 0;
> +			if (cxl_need_node_perf_attrs_update(nid))
> +				node_set_perf_attrs(nid, &cxlr->coord[i], i);
> +			else
> +				rc = cxl_update_hmat_access_coordinates(nid, cxlr, i);
> +
>  			if (rc == 0)
>  				cset++;
>  		}
> diff --git a/include/linux/acpi.h b/include/linux/acpi.h
> index 1c664948b2ae..3067c6aad431 100644
> --- a/include/linux/acpi.h
> +++ b/include/linux/acpi.h
> @@ -447,6 +447,7 @@ static inline int hmat_update_target_coordinates(int nid,
>  #ifdef CONFIG_ACPI_NUMA
>  int acpi_map_pxm_to_node(int pxm);
>  int acpi_get_node(acpi_handle handle);
> +bool acpi_node_backed_by_real_pxm(int nid);
>  
>  /**
>   * pxm_to_online_node - Map proximity ID to online node
> 
> 

Re: [PATCH v4 05/11] cxl: Split out combine_coordinates() for common shared usage

[Dave Jiang]

On 1/19/24 17:35, Dan Williams wrote:
> Dave Jiang wrote:
>> Refactor the common code of combining coordinates in order to reduce code.
>> Create a new function cxl_cooordinates_combine() it combine two 'struct
>> access_coordinate'.
>>
>> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
>> ---
>>  drivers/cxl/core/cdat.c |   32 +++++++++++++++++++++++---------
>>  drivers/cxl/core/port.c |   18 ++----------------
>>  drivers/cxl/cxl.h       |    4 ++++
>>  3 files changed, 29 insertions(+), 25 deletions(-)
>>
>> diff --git a/drivers/cxl/core/cdat.c b/drivers/cxl/core/cdat.c
>> index cd84d87f597a..4d542627d02d 100644
>> --- a/drivers/cxl/core/cdat.c
>> +++ b/drivers/cxl/core/cdat.c
>> @@ -183,15 +183,7 @@ static int cxl_port_perf_data_calculate(struct cxl_port *port,
>>  	xa_for_each(dsmas_xa, index, dent) {
>>  		int qos_class;
>>  
>> -		dent->coord.read_latency = dent->coord.read_latency +
>> -					   c.read_latency;
>> -		dent->coord.write_latency = dent->coord.write_latency +
>> -					    c.write_latency;
>> -		dent->coord.read_bandwidth = min_t(int, c.read_bandwidth,
>> -						   dent->coord.read_bandwidth);
>> -		dent->coord.write_bandwidth = min_t(int, c.write_bandwidth,
>> -						    dent->coord.write_bandwidth);
>> -
>> +		cxl_coordinates_combine(&dent->coord, &dent->coord, &c);
>>  		dent->entries = 1;
>>  		rc = cxl_root->ops->qos_class(root_port, &dent->coord, 1, &qos_class);
>>  		if (rc != 1)
>> @@ -514,4 +506,26 @@ void cxl_switch_parse_cdat(struct cxl_port *port)
>>  }
>>  EXPORT_SYMBOL_NS_GPL(cxl_switch_parse_cdat, CXL);
>>  
>> +/**
>> + * cxl_coordinates_combine - Combine the two input coordinates into the first
>> + *
>> + * @c1: first coordinate, to be written to
>> + * @c2: second coordinate
>> + */
>> +void cxl_coordinates_combine(struct access_coordinate *out,
>> +			     struct access_coordinate *c1,
>> +			     struct access_coordinate *c2)
>> +{
>> +		if (c2->write_bandwidth)
>> +			out->write_bandwidth = min(c1->write_bandwidth,
>> +						   c2->write_bandwidth);
>> +		out->write_latency = c1->write_latency + c2->write_latency;
>> +
>> +		if (c2->read_bandwidth)
>> +			out->read_bandwidth = min(c1->read_bandwidth,
>> +						  c2->read_bandwidth);
>> +		out->read_latency = c1->read_latency + c2->read_latency;
>> +}
>> +EXPORT_SYMBOL_NS_GPL(cxl_coordinates_combine, CXL);
> 
> There is no need for EXPORT_SYMBOL() when the definition and the only
> caller exist within the same compilation unit, cxl_core.o.
> 
> However, given there is nothing "CXL" about this function it likely wants
> to move out of cxl_core.o if another caller ever arrives.

It's mostly used by core/cdat.c but eventually also used by core/port.c. So it's all within the core. 
> 

RE: [PATCH v4 08/11] cxl/region: Calculate performance data for a region

[Wonjae Lee]

On Fri, Jan 19, 2024 at 10:23:52AM -0700, Dave Jiang wrote:
> Calculate and store the performance data for a CXL region. Find the worst
> read and write latency for all the included ranges from each of the devices
> that attributes to the region and designate that as the latency data. Sum
> all the read and write bandwidth data for each of the device region and
> that is the total bandwidth for the region.
>
> The perf list is expected to be constructed before the endpoint decoders
> are registered and thus there should be no early reading of the entries
> from the region assemble action. The calling of the region qos calculate
> function is under the protection of cxl_dpa_rwsem and will ensure that
> all DPA associated work has completed.
>
> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
> ---
> v4:
> - Calculate access classes 0 and 1 by retrieving host bridge coords
> - Add lockdep assert for cxl_dpa_rwsem (Dan)
> - Clarify that HMAT code is HMEM_REPORTING code. (Dan)
> ---
>  drivers/cxl/core/cdat.c    74 +++++++++++++++++++++++++++++++++++++++++++++
>  drivers/cxl/core/region.c     2 +
>  drivers/cxl/cxl.h            4 ++
>  3 files changed, 80 insertions(+)
>
> diff --git a/drivers/cxl/core/cdat.c b/drivers/cxl/core/cdat.c
> index 6e3998723aaa..7acb5837afad 100644
> --- a/drivers/cxl/core/cdat.c
> +++ b/drivers/cxl/core/cdat.c
> @@ -8,6 +8,7 @@
>  #include "cxlpci.h"
>  #include "cxlmem.h"
>  #include "cxl.h"
> +#include "core.h"
>
>  struct dsmas_entry {
>  struct range dpa_range;
> @@ -546,3 +547,76 @@ void cxl_coordinates_combine(struct access_coordinate *out,
>  EXPORT_SYMBOL_NS_GPL(cxl_coordinates_combine, CXL);
>
>  MODULE_IMPORT_NS(CXL);
> +
> +void cxl_region_perf_data_calculate(struct cxl_region *cxlr,
> +                struct cxl_endpoint_decoder *cxled)
> +{
> + struct cxl_memdev *cxlmd = cxled_to_memdev(cxled);
> + struct cxl_port *port = cxlmd->endpoint;
> + struct cxl_dev_state *cxlds = cxlmd->cxlds;
> + struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlds);
> + struct access_coordinate hb_coord[ACCESS_COORDINATE_MAX];
> + struct access_coordinate coord;
> + struct range dpa = {
> +        .start = cxled->dpa_res->start,
> +        .end = cxled->dpa_res->end,
> + };
> + struct list_head *perf_list;
> + struct cxl_dpa_perf *perf;
> + bool found = false;
> + int rc;
> +
> + switch (cxlr->mode) {
> + case CXL_DECODER_RAM:
> +    perf_list = &mds->ram_perf_list;
> +    break;
> + case CXL_DECODER_PMEM:
> +    perf_list = &mds->pmem_perf_list;
> +    break;
> + default:
> +    return;
> + }
> +
> + lockdep_assert_held(&cxl_dpa_rwsem);
> +
> + list_for_each_entry(perf, perf_list, list) {
> +    if (range_contains(&perf->dpa_range, &dpa)) {
> +        found = true;
> +        break;
> +    }
> + }
> +
> + if (!found)
> +    return;
> +
> + rc = cxl_hb_get_perf_coordinates(port, hb_coord);
> + if (rc)  {
> +    dev_dbg(&port->dev, "Failed to retrieve hb perf coordinates.\n");
> +    return;
> + }
> +
> + for (int i = 0; i < ACCESS_COORDINATE_MAX; i++) {
> +    /* Pickup the host bridge coords */
> +    cxl_coordinates_combine(&coord, &hb_coord[i], &perf->coord);
> +
> +    /* Get total bandwidth and the worst latency for the cxl region */
> +    cxlr->coord[i].read_latency = max_t(unsigned int,
> +                        cxlr->coord[i].read_latency,
> +                        coord.read_latency);
> +    cxlr->coord[i].write_latency = max_t(unsigned int,
> +                          cxlr->coord[i].write_latency,
> +                          coord.write_latency);
> +    cxlr->coord[i].read_bandwidth += coord.read_bandwidth;
> +    cxlr->coord[i].write_bandwidth += coord.write_bandwidth;
> +
> +    /*
> +      * Convert latency to nanosec from picosec to be consistent
> +      * with the resulting latency coordinates computed by the
> +      * HMAT_REPORTING code.
> +      */
> +    cxlr->coord[i].read_latency =
> +        DIV_ROUND_UP(cxlr->coord[i].read_latency, 1000);
> +    cxlr->coord[i].write_latency =
> +        DIV_ROUND_UP(cxlr->coord[i].write_latency, 1000);

Hello,

I ran into a bit of confusion and have a question while validating CDAT
behaviour with physical CXL devices. (I'm not sure if this is the right
thread to ask this question, sorry if it isn't.)

IIUC, the raw data of latency is in picosec, but the comments on the
struct access_coordinate say that the latency units are in nanosec:
 * @read_latency:   Read latency in nanoseconds
 * @write_latency:  Write latency in nanoseconds

This was a bit confusing at first, as the raw data of latency are in
ps, and the structure that stores the latency expects units of ns.

I saw that you have already had a discussion with Brice about the
pico/nanosecond unit conversion. My question is, are there any plans to
store latency number of cxl port in nanoseconds or change the comments
of coords structure?

Thanks,
Wonjae

> + }
> +}
> diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
> index 57a5901d5a60..7f19b533c5ae 100644
> --- a/drivers/cxl/core/region.c
> +++ b/drivers/cxl/core/region.c
> @@ -1722,6 +1722,8 @@ static int cxl_region_attach(struct cxl_region *cxlr,
>      return -EINVAL;
>  }
>
> + cxl_region_perf_data_calculate(cxlr, cxled);
> +
>  if (test_bit(CXL_REGION_F_AUTO, &cxlr->flags)) {
>      int i;
>
> diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
> index 80e6bd294e18..f6637fa33113 100644
> --- a/drivers/cxl/cxl.h
> +++ b/drivers/cxl/cxl.h
> @@ -519,6 +519,7 @@ struct cxl_region_params {
>  * @cxlr_pmem: (for pmem regions) cached copy of the nvdimm bridge
>  * @flags: Region state flags
>  * @params: active + config params for the region
> + * @coord: QoS access coordinates for the region
>  */
>  struct cxl_region {
>  struct device dev;
> @@ -529,6 +530,7 @@ struct cxl_region {
>  struct cxl_pmem_region *cxlr_pmem;
>  unsigned long flags;
>  struct cxl_region_params params;
> + struct access_coordinate coord[ACCESS_COORDINATE_MAX];
>  };
>
>  struct cxl_nvdimm_bridge {
> @@ -880,6 +882,8 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port,
>                    struct access_coordinate *coord);
>  int cxl_hb_get_perf_coordinates(struct cxl_port *port,
>              struct access_coordinate *coord);
> +void cxl_region_perf_data_calculate(struct cxl_region *cxlr,
> +                struct cxl_endpoint_decoder *cxled);
>
>  void cxl_coordinates_combine(struct access_coordinate *out,
>                struct access_coordinate *c1,
>
>
>

Re: [PATCH v4 08/11] cxl/region: Calculate performance data for a region

[Dave Jiang]

On 1/30/24 19:22, Wonjae Lee wrote:
> On Fri, Jan 19, 2024 at 10:23:52AM -0700, Dave Jiang wrote:
>> Calculate and store the performance data for a CXL region. Find the worst
>> read and write latency for all the included ranges from each of the devices
>> that attributes to the region and designate that as the latency data. Sum
>> all the read and write bandwidth data for each of the device region and
>> that is the total bandwidth for the region.
>>
>> The perf list is expected to be constructed before the endpoint decoders
>> are registered and thus there should be no early reading of the entries
>> from the region assemble action. The calling of the region qos calculate
>> function is under the protection of cxl_dpa_rwsem and will ensure that
>> all DPA associated work has completed.
>>
>> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
>> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
>> ---
>> v4:
>> - Calculate access classes 0 and 1 by retrieving host bridge coords
>> - Add lockdep assert for cxl_dpa_rwsem (Dan)
>> - Clarify that HMAT code is HMEM_REPORTING code. (Dan)
>> ---
>>   drivers/cxl/core/cdat.c    74 +++++++++++++++++++++++++++++++++++++++++++++
>>   drivers/cxl/core/region.c     2 +
>>   drivers/cxl/cxl.h            4 ++
>>   3 files changed, 80 insertions(+)
>>
>> diff --git a/drivers/cxl/core/cdat.c b/drivers/cxl/core/cdat.c
>> index 6e3998723aaa..7acb5837afad 100644
>> --- a/drivers/cxl/core/cdat.c
>> +++ b/drivers/cxl/core/cdat.c
>> @@ -8,6 +8,7 @@
>>   #include "cxlpci.h"
>>   #include "cxlmem.h"
>>   #include "cxl.h"
>> +#include "core.h"
>>
>>   struct dsmas_entry {
>>   struct range dpa_range;
>> @@ -546,3 +547,76 @@ void cxl_coordinates_combine(struct access_coordinate *out,
>>   EXPORT_SYMBOL_NS_GPL(cxl_coordinates_combine, CXL);
>>
>>   MODULE_IMPORT_NS(CXL);
>> +
>> +void cxl_region_perf_data_calculate(struct cxl_region *cxlr,
>> +                struct cxl_endpoint_decoder *cxled)
>> +{
>> + struct cxl_memdev *cxlmd = cxled_to_memdev(cxled);
>> + struct cxl_port *port = cxlmd->endpoint;
>> + struct cxl_dev_state *cxlds = cxlmd->cxlds;
>> + struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlds);
>> + struct access_coordinate hb_coord[ACCESS_COORDINATE_MAX];
>> + struct access_coordinate coord;
>> + struct range dpa = {
>> +        .start = cxled->dpa_res->start,
>> +        .end = cxled->dpa_res->end,
>> + };
>> + struct list_head *perf_list;
>> + struct cxl_dpa_perf *perf;
>> + bool found = false;
>> + int rc;
>> +
>> + switch (cxlr->mode) {
>> + case CXL_DECODER_RAM:
>> +    perf_list = &mds->ram_perf_list;
>> +    break;
>> + case CXL_DECODER_PMEM:
>> +    perf_list = &mds->pmem_perf_list;
>> +    break;
>> + default:
>> +    return;
>> + }
>> +
>> + lockdep_assert_held(&cxl_dpa_rwsem);
>> +
>> + list_for_each_entry(perf, perf_list, list) {
>> +    if (range_contains(&perf->dpa_range, &dpa)) {
>> +        found = true;
>> +        break;
>> +    }
>> + }
>> +
>> + if (!found)
>> +    return;
>> +
>> + rc = cxl_hb_get_perf_coordinates(port, hb_coord);
>> + if (rc)  {
>> +    dev_dbg(&port->dev, "Failed to retrieve hb perf coordinates.\n");
>> +    return;
>> + }
>> +
>> + for (int i = 0; i < ACCESS_COORDINATE_MAX; i++) {
>> +    /* Pickup the host bridge coords */
>> +    cxl_coordinates_combine(&coord, &hb_coord[i], &perf->coord);
>> +
>> +    /* Get total bandwidth and the worst latency for the cxl region */
>> +    cxlr->coord[i].read_latency = max_t(unsigned int,
>> +                        cxlr->coord[i].read_latency,
>> +                        coord.read_latency);
>> +    cxlr->coord[i].write_latency = max_t(unsigned int,
>> +                          cxlr->coord[i].write_latency,
>> +                          coord.write_latency);
>> +    cxlr->coord[i].read_bandwidth += coord.read_bandwidth;
>> +    cxlr->coord[i].write_bandwidth += coord.write_bandwidth;
>> +
>> +    /*
>> +      * Convert latency to nanosec from picosec to be consistent
>> +      * with the resulting latency coordinates computed by the
>> +      * HMAT_REPORTING code.
>> +      */
>> +    cxlr->coord[i].read_latency =
>> +        DIV_ROUND_UP(cxlr->coord[i].read_latency, 1000);
>> +    cxlr->coord[i].write_latency =
>> +        DIV_ROUND_UP(cxlr->coord[i].write_latency, 1000);
> 
> Hello,
> 
> I ran into a bit of confusion and have a question while validating CDAT
> behaviour with physical CXL devices. (I'm not sure if this is the right
> thread to ask this question, sorry if it isn't.)
> 
> IIUC, the raw data of latency is in picosec, but the comments on the
> struct access_coordinate say that the latency units are in nanosec:
>  * @read_latency:   Read latency in nanoseconds
>  * @write_latency:  Write latency in nanoseconds
> 
> This was a bit confusing at first, as the raw data of latency are in
> ps, and the structure that stores the latency expects units of ns.

Right. The numbers stored with the HMAT_REPORTING code and eventually NUMA nodes are normalized to nanoseconds, even though the raw data is in picoseconds. For CXL, I left the CDAT and computed numbers as raw numbers (picoseconds) until the final step when I calculate the latency for the entire region. And then it gets converted to nanoseconds in order to write back to the memory_target for HMAT_REPORTING. The numbers we retrieve from HMAT_REPORTING for the generic target is already in nanoseconds.


> 
> I saw that you have already had a discussion with Brice about the
> pico/nanosecond unit conversion. My question is, are there any plans to
> store latency number of cxl port in nanoseconds or change the comments
> of coords structure?

The numbers for the coords structure will remain in nanoseconds as it always have been. 

> 
> Thanks,
> Wonjae
> 
>> + }
>> +}
>> diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
>> index 57a5901d5a60..7f19b533c5ae 100644
>> --- a/drivers/cxl/core/region.c
>> +++ b/drivers/cxl/core/region.c
>> @@ -1722,6 +1722,8 @@ static int cxl_region_attach(struct cxl_region *cxlr,
>>       return -EINVAL;
>>   }
>>
>> + cxl_region_perf_data_calculate(cxlr, cxled);
>> +
>>   if (test_bit(CXL_REGION_F_AUTO, &cxlr->flags)) {
>>       int i;
>>
>> diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
>> index 80e6bd294e18..f6637fa33113 100644
>> --- a/drivers/cxl/cxl.h
>> +++ b/drivers/cxl/cxl.h
>> @@ -519,6 +519,7 @@ struct cxl_region_params {
>>   * @cxlr_pmem: (for pmem regions) cached copy of the nvdimm bridge
>>   * @flags: Region state flags
>>   * @params: active + config params for the region
>> + * @coord: QoS access coordinates for the region
>>   */
>>   struct cxl_region {
>>   struct device dev;
>> @@ -529,6 +530,7 @@ struct cxl_region {
>>   struct cxl_pmem_region *cxlr_pmem;
>>   unsigned long flags;
>>   struct cxl_region_params params;
>> + struct access_coordinate coord[ACCESS_COORDINATE_MAX];
>>   };
>>
>>   struct cxl_nvdimm_bridge {
>> @@ -880,6 +882,8 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port,
>>                     struct access_coordinate *coord);
>>   int cxl_hb_get_perf_coordinates(struct cxl_port *port,
>>               struct access_coordinate *coord);
>> +void cxl_region_perf_data_calculate(struct cxl_region *cxlr,
>> +                struct cxl_endpoint_decoder *cxled);
>>
>>   void cxl_coordinates_combine(struct access_coordinate *out,
>>                 struct access_coordinate *c1,
>>
>>
>>

Re: [PATCH v4 08/11] cxl/region: Calculate performance data for a region

[Jonathan Cameron]

On Wed, 31 Jan 2024 08:56:41 -0700
Dave Jiang <dave.jiang@intel.com> wrote:

> On 1/30/24 19:22, Wonjae Lee wrote:
> > On Fri, Jan 19, 2024 at 10:23:52AM -0700, Dave Jiang wrote:  
> >> Calculate and store the performance data for a CXL region. Find the worst
> >> read and write latency for all the included ranges from each of the devices
> >> that attributes to the region and designate that as the latency data. Sum
> >> all the read and write bandwidth data for each of the device region and
> >> that is the total bandwidth for the region.
> >>
> >> The perf list is expected to be constructed before the endpoint decoders
> >> are registered and thus there should be no early reading of the entries
> >> from the region assemble action. The calling of the region qos calculate
> >> function is under the protection of cxl_dpa_rwsem and will ensure that
> >> all DPA associated work has completed.
> >>
> >> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> >> Signed-off-by: Dave Jiang <dave.jiang@intel.com>
> >> ---
> >> v4:
> >> - Calculate access classes 0 and 1 by retrieving host bridge coords
> >> - Add lockdep assert for cxl_dpa_rwsem (Dan)
> >> - Clarify that HMAT code is HMEM_REPORTING code. (Dan)
> >> ---
> >>   drivers/cxl/core/cdat.c    74 +++++++++++++++++++++++++++++++++++++++++++++
> >>   drivers/cxl/core/region.c     2 +
> >>   drivers/cxl/cxl.h            4 ++
> >>   3 files changed, 80 insertions(+)
> >>
> >> diff --git a/drivers/cxl/core/cdat.c b/drivers/cxl/core/cdat.c
> >> index 6e3998723aaa..7acb5837afad 100644
> >> --- a/drivers/cxl/core/cdat.c
> >> +++ b/drivers/cxl/core/cdat.c
> >> @@ -8,6 +8,7 @@
> >>   #include "cxlpci.h"
> >>   #include "cxlmem.h"
> >>   #include "cxl.h"
> >> +#include "core.h"
> >>
> >>   struct dsmas_entry {
> >>   struct range dpa_range;
> >> @@ -546,3 +547,76 @@ void cxl_coordinates_combine(struct access_coordinate *out,
> >>   EXPORT_SYMBOL_NS_GPL(cxl_coordinates_combine, CXL);
> >>
> >>   MODULE_IMPORT_NS(CXL);
> >> +
> >> +void cxl_region_perf_data_calculate(struct cxl_region *cxlr,
> >> +                struct cxl_endpoint_decoder *cxled)
> >> +{
> >> + struct cxl_memdev *cxlmd = cxled_to_memdev(cxled);
> >> + struct cxl_port *port = cxlmd->endpoint;
> >> + struct cxl_dev_state *cxlds = cxlmd->cxlds;
> >> + struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlds);
> >> + struct access_coordinate hb_coord[ACCESS_COORDINATE_MAX];
> >> + struct access_coordinate coord;
> >> + struct range dpa = {
> >> +        .start = cxled->dpa_res->start,
> >> +        .end = cxled->dpa_res->end,
> >> + };
> >> + struct list_head *perf_list;
> >> + struct cxl_dpa_perf *perf;
> >> + bool found = false;
> >> + int rc;
> >> +
> >> + switch (cxlr->mode) {
> >> + case CXL_DECODER_RAM:
> >> +    perf_list = &mds->ram_perf_list;
> >> +    break;
> >> + case CXL_DECODER_PMEM:
> >> +    perf_list = &mds->pmem_perf_list;
> >> +    break;
> >> + default:
> >> +    return;
> >> + }
> >> +
> >> + lockdep_assert_held(&cxl_dpa_rwsem);
> >> +
> >> + list_for_each_entry(perf, perf_list, list) {
> >> +    if (range_contains(&perf->dpa_range, &dpa)) {
> >> +        found = true;
> >> +        break;
> >> +    }
> >> + }
> >> +
> >> + if (!found)
> >> +    return;
> >> +
> >> + rc = cxl_hb_get_perf_coordinates(port, hb_coord);
> >> + if (rc)  {
> >> +    dev_dbg(&port->dev, "Failed to retrieve hb perf coordinates.\n");
> >> +    return;
> >> + }
> >> +
> >> + for (int i = 0; i < ACCESS_COORDINATE_MAX; i++) {
> >> +    /* Pickup the host bridge coords */
> >> +    cxl_coordinates_combine(&coord, &hb_coord[i], &perf->coord);
> >> +
> >> +    /* Get total bandwidth and the worst latency for the cxl region */
> >> +    cxlr->coord[i].read_latency = max_t(unsigned int,
> >> +                        cxlr->coord[i].read_latency,
> >> +                        coord.read_latency);
> >> +    cxlr->coord[i].write_latency = max_t(unsigned int,
> >> +                          cxlr->coord[i].write_latency,
> >> +                          coord.write_latency);
> >> +    cxlr->coord[i].read_bandwidth += coord.read_bandwidth;
> >> +    cxlr->coord[i].write_bandwidth += coord.write_bandwidth;
> >> +
> >> +    /*
> >> +      * Convert latency to nanosec from picosec to be consistent
> >> +      * with the resulting latency coordinates computed by the
> >> +      * HMAT_REPORTING code.
> >> +      */
> >> +    cxlr->coord[i].read_latency =
> >> +        DIV_ROUND_UP(cxlr->coord[i].read_latency, 1000);
> >> +    cxlr->coord[i].write_latency =
> >> +        DIV_ROUND_UP(cxlr->coord[i].write_latency, 1000);  
> > 
> > Hello,
> > 
> > I ran into a bit of confusion and have a question while validating CDAT
> > behaviour with physical CXL devices. (I'm not sure if this is the right
> > thread to ask this question, sorry if it isn't.)
> > 
> > IIUC, the raw data of latency is in picosec, but the comments on the
> > struct access_coordinate say that the latency units are in nanosec:
> >  * @read_latency:   Read latency in nanoseconds
> >  * @write_latency:  Write latency in nanoseconds
> > 
> > This was a bit confusing at first, as the raw data of latency are in
> > ps, and the structure that stores the latency expects units of ns.  
> 
> Right. The numbers stored with the HMAT_REPORTING code and eventually NUMA nodes are normalized to nanoseconds, even though the raw data is in picoseconds. For CXL, I left the CDAT and computed numbers as raw numbers (picoseconds) until the final step when I calculate the latency for the entire region. And then it gets converted to nanoseconds in order to write back to the memory_target for HMAT_REPORTING. The numbers we retrieve from HMAT_REPORTING for the generic target is already in nanoseconds.

HMAT had a non backwards compatible change in units between the first
and second revs of that table. See ACPI 6.2 for revision 1.

With another hat on I'm a bit embarrassed about that :( .

> 
> 
> > 
> > I saw that you have already had a discussion with Brice about the
> > pico/nanosecond unit conversion. My question is, are there any plans to
> > store latency number of cxl port in nanoseconds or change the comments
> > of coords structure?  
> 
> The numbers for the coords structure will remain in nanoseconds as it always have been. 
> 
> > 
> > Thanks,
> > Wonjae
> >   
> >> + }
> >> +}
> >> diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c
> >> index 57a5901d5a60..7f19b533c5ae 100644
> >> --- a/drivers/cxl/core/region.c
> >> +++ b/drivers/cxl/core/region.c
> >> @@ -1722,6 +1722,8 @@ static int cxl_region_attach(struct cxl_region *cxlr,
> >>       return -EINVAL;
> >>   }
> >>
> >> + cxl_region_perf_data_calculate(cxlr, cxled);
> >> +
> >>   if (test_bit(CXL_REGION_F_AUTO, &cxlr->flags)) {
> >>       int i;
> >>
> >> diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
> >> index 80e6bd294e18..f6637fa33113 100644
> >> --- a/drivers/cxl/cxl.h
> >> +++ b/drivers/cxl/cxl.h
> >> @@ -519,6 +519,7 @@ struct cxl_region_params {
> >>   * @cxlr_pmem: (for pmem regions) cached copy of the nvdimm bridge
> >>   * @flags: Region state flags
> >>   * @params: active + config params for the region
> >> + * @coord: QoS access coordinates for the region
> >>   */
> >>   struct cxl_region {
> >>   struct device dev;
> >> @@ -529,6 +530,7 @@ struct cxl_region {
> >>   struct cxl_pmem_region *cxlr_pmem;
> >>   unsigned long flags;
> >>   struct cxl_region_params params;
> >> + struct access_coordinate coord[ACCESS_COORDINATE_MAX];
> >>   };
> >>
> >>   struct cxl_nvdimm_bridge {
> >> @@ -880,6 +882,8 @@ int cxl_endpoint_get_perf_coordinates(struct cxl_port *port,
> >>                     struct access_coordinate *coord);
> >>   int cxl_hb_get_perf_coordinates(struct cxl_port *port,
> >>               struct access_coordinate *coord);
> >> +void cxl_region_perf_data_calculate(struct cxl_region *cxlr,
> >> +                struct cxl_endpoint_decoder *cxled);
> >>
> >>   void cxl_coordinates_combine(struct access_coordinate *out,
> >>                 struct access_coordinate *c1,
> >>
> >>
> >>  
>