diff mbox series

[v7,08/11] cxl: Compute the entire CXL path latency and bandwidth data

Message ID 168695174591.3031571.6462488730394994238.stgit@djiang5-mobl3
State New, archived
Headers show
Series cxl: Add support for QTG ID retrieval for CXL subsystem | expand

Commit Message

Dave Jiang June 16, 2023, 9:42 p.m. UTC
CXL Memory Device SW Guide [1] rev1.0 2.11.2 provides instruction on how to
calculate latency and bandwidth for CXL memory device. Calculate minimum
bandwidth and total latency for the path from the CXL device to the root
port. The QTG id is retrieved by providing the performance data as input
and calling the root port callback ->get_qos_class(). The retrieved id is
stored with the cxl_port of the CXL device.

For example for a device that is directly attached to a host bus:
Total Latency = Device Latency (from CDAT) + Dev to Host Bus (HB) Link
		Latency + Generic Port Latency
Min Bandwidth = Min bandwidth for link bandwidth between HB
		and CXL device, device CDAT bandwidth, and Generic Port
		Bandwidth

For a device that has a switch in between host bus and CXL device:
Total Latency = Device (CDAT) Latency + Dev to Switch Link Latency +
		Switch (CDAT) Latency + Switch to HB Link Latency +
		Generic Port Latency
Min Bandwidth = Min bandwidth for link bandwidth between CXL device
		to CXL switch, CXL device CDAT bandwidth, CXL switch CDAT
		bandwidth, CXL switch to HB bandwidth, and Generic Port
		Bandwidth.

[1]: https://cdrdv2-public.intel.com/643805/643805_CXL%20Memory%20Device%20SW%20Guide_Rev1p0.pdf

Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>

---
v5:
- Use new API call cxl_endpoint_get_perf_coordinates().
- Use root_port->get_qos_class() (Dan)
- Add endieness handling to DSM input.
---
 drivers/cxl/core/cdat.c |    1 +
 drivers/cxl/cxl.h       |    1 +
 drivers/cxl/port.c      |   57 +++++++++++++++++++++++++++++++++++++++++++++--
 3 files changed, 56 insertions(+), 3 deletions(-)
diff mbox series

Patch

diff --git a/drivers/cxl/core/cdat.c b/drivers/cxl/core/cdat.c
index f233cebca37a..1a6b4aeacb0c 100644
--- a/drivers/cxl/core/cdat.c
+++ b/drivers/cxl/core/cdat.c
@@ -148,6 +148,7 @@  void cxl_cdat_dsmas_list_destroy(struct list_head *dsmas_list)
 
 	list_for_each_entry_safe(dentry, n, dsmas_list, list) {
 		list_del(&dentry->list);
+		kfree(dentry->qos_class);
 		kfree(dentry);
 	}
 }
diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h
index 580f44587ab0..5255edc3fe32 100644
--- a/drivers/cxl/cxl.h
+++ b/drivers/cxl/cxl.h
@@ -844,6 +844,7 @@  struct dsmas_entry {
 	struct range dpa_range;
 	u8 handle;
 	struct access_coordinate coord;
+	struct qos_class *qos_class;
 };
 
 #ifdef CONFIG_FIRMWARE_TABLE
diff --git a/drivers/cxl/port.c b/drivers/cxl/port.c
index 8b36a8633a15..4f2828caba7b 100644
--- a/drivers/cxl/port.c
+++ b/drivers/cxl/port.c
@@ -57,6 +57,53 @@  static int discover_region(struct device *dev, void *root)
 	return 0;
 }
 
+static int cxl_port_perf_data_calculate(struct cxl_port *port,
+					struct list_head *dsmas_list)
+{
+	struct qos_class *qos_class;
+	struct access_coordinate c;
+	struct qtg_dsm_input input;
+	struct cxl_port *root_port;
+	struct cxl_root *cxl_root;
+	struct dsmas_entry *dent;
+	int rc;
+
+	rc = cxl_endpoint_get_perf_coordinates(port, &c);
+	if (rc) {
+		dev_dbg(&port->dev, "Failed to retrieve perf coordinates.\n");
+		return rc;
+	}
+
+	root_port = find_cxl_root(port);
+	cxl_root = to_cxl_root(root_port);
+	if (!cxl_root->ops || !cxl_root->ops->get_qos_class)
+		return -EOPNOTSUPP;
+
+	list_for_each_entry(dent, dsmas_list, list) {
+		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);
+
+		input.rd_lat = cpu_to_le32(dent->coord.read_latency);
+		input.wr_lat = cpu_to_le32(dent->coord.write_latency);
+		input.rd_bw = cpu_to_le32(dent->coord.read_bandwidth);
+		input.wr_bw = cpu_to_le32(dent->coord.write_bandwidth);
+
+		qos_class = cxl_root->ops->get_qos_class(root_port, &input);
+		if (IS_ERR(qos_class))
+			continue;
+
+		dent->qos_class = qos_class;
+	}
+
+	return 0;
+}
+
 static int cxl_switch_port_probe(struct cxl_port *port)
 {
 	struct cxl_hdm *cxlhdm;
@@ -147,10 +194,14 @@  static int cxl_endpoint_port_probe(struct cxl_port *port)
 		LIST_HEAD(dsmas_list);
 
 		rc = cxl_cdat_endpoint_process(port, &dsmas_list);
-		if (rc < 0)
+		if (rc < 0) {
 			dev_dbg(&port->dev, "Failed to parse CDAT: %d\n", rc);
-
-		/* Performance data processing */
+		} else {
+			rc = cxl_port_perf_data_calculate(port, &dsmas_list);
+			if (rc)
+				dev_dbg(&port->dev,
+					"Failed to do perf coord calculations.\n");
+		}
 
 		cxl_cdat_dsmas_list_destroy(&dsmas_list);
 	}