1 files changed, 159 insertions, 94 deletions

diff --git a/drivers/cxl/core/memdev.c b/drivers/cxl/core/memdev.c
index 14b547c07f54..d4e259f3a7e9 100644
--- a/drivers/cxl/core/memdev.c
+++ b/drivers/cxl/core/memdev.c
@@ -114,7 +114,7 @@ static DEVICE_ATTR_RO(serial);
 static ssize_t numa_node_show(struct device *dev, struct device_attribute *attr,
 			      char *buf)
 {
-	return sprintf(buf, "%d\n", dev_to_node(dev));
+	return sysfs_emit(buf, "%d\n", dev_to_node(dev));
 }
 static DEVICE_ATTR_RO(numa_node);
 
@@ -125,13 +125,16 @@ static ssize_t security_state_show(struct device *dev,
 	struct cxl_memdev *cxlmd = to_cxl_memdev(dev);
 	struct cxl_dev_state *cxlds = cxlmd->cxlds;
 	struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlds);
-	u64 reg = readq(cxlds->regs.mbox + CXLDEV_MBOX_BG_CMD_STATUS_OFFSET);
-	u32 pct = FIELD_GET(CXLDEV_MBOX_BG_CMD_COMMAND_PCT_MASK, reg);
-	u16 cmd = FIELD_GET(CXLDEV_MBOX_BG_CMD_COMMAND_OPCODE_MASK, reg);
 	unsigned long state = mds->security.state;
+	int rc = 0;
 
-	if (cmd == CXL_MBOX_OP_SANITIZE && pct != 100)
-		return sysfs_emit(buf, "sanitize\n");
+	/* sync with latest submission state */
+	mutex_lock(&mds->mbox_mutex);
+	if (mds->security.sanitize_active)
+		rc = sysfs_emit(buf, "sanitize\n");
+	mutex_unlock(&mds->mbox_mutex);
+	if (rc)
+		return rc;
 
 	if (!(state & CXL_PMEM_SEC_STATE_USER_PASS_SET))
 		return sysfs_emit(buf, "disabled\n");
@@ -152,24 +155,17 @@ static ssize_t security_sanitize_store(struct device *dev,
 				       const char *buf, size_t len)
 {
 	struct cxl_memdev *cxlmd = to_cxl_memdev(dev);
-	struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlmd->cxlds);
-	struct cxl_port *port = cxlmd->endpoint;
 	bool sanitize;
 	ssize_t rc;
 
 	if (kstrtobool(buf, &sanitize) || !sanitize)
 		return -EINVAL;
 
-	if (!port || !is_cxl_endpoint(port))
-		return -EINVAL;
-
-	/* ensure no regions are mapped to this memdev */
-	if (port->commit_end != -1)
-		return -EBUSY;
-
-	rc = cxl_mem_sanitize(mds, CXL_MBOX_OP_SANITIZE);
+	rc = cxl_mem_sanitize(cxlmd, CXL_MBOX_OP_SANITIZE);
+	if (rc)
+		return rc;
 
-	return rc ? rc : len;
+	return len;
 }
 static struct device_attribute dev_attr_security_sanitize =
 	__ATTR(sanitize, 0200, NULL, security_sanitize_store);
@@ -179,24 +175,17 @@ static ssize_t security_erase_store(struct device *dev,
 				    const char *buf, size_t len)
 {
 	struct cxl_memdev *cxlmd = to_cxl_memdev(dev);
-	struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlmd->cxlds);
-	struct cxl_port *port = cxlmd->endpoint;
 	ssize_t rc;
 	bool erase;
 
 	if (kstrtobool(buf, &erase) || !erase)
 		return -EINVAL;
 
-	if (!port || !is_cxl_endpoint(port))
-		return -EINVAL;
-
-	/* ensure no regions are mapped to this memdev */
-	if (port->commit_end != -1)
-		return -EBUSY;
-
-	rc = cxl_mem_sanitize(mds, CXL_MBOX_OP_SECURE_ERASE);
+	rc = cxl_mem_sanitize(cxlmd, CXL_MBOX_OP_SECURE_ERASE);
+	if (rc)
+		return rc;
 
-	return rc ? rc : len;
+	return len;
 }
 static struct device_attribute dev_attr_security_erase =
 	__ATTR(erase, 0200, NULL, security_erase_store);
@@ -238,11 +227,17 @@ int cxl_trigger_poison_list(struct cxl_memdev *cxlmd)
 	if (!port || !is_cxl_endpoint(port))
 		return -EINVAL;
 
-	rc = down_read_interruptible(&cxl_dpa_rwsem);
+	rc = down_read_interruptible(&cxl_region_rwsem);
 	if (rc)
 		return rc;
 
-	if (port->commit_end == -1) {
+	rc = down_read_interruptible(&cxl_dpa_rwsem);
+	if (rc) {
+		up_read(&cxl_region_rwsem);
+		return rc;
+	}
+
+	if (cxl_num_decoders_committed(port) == 0) {
 		/* No regions mapped to this memdev */
 		rc = cxl_get_poison_by_memdev(cxlmd);
 	} else {
@@ -250,6 +245,7 @@ int cxl_trigger_poison_list(struct cxl_memdev *cxlmd)
 		rc =  cxl_get_poison_by_endpoint(port);
 	}
 	up_read(&cxl_dpa_rwsem);
+	up_read(&cxl_region_rwsem);
 
 	return rc;
 }
@@ -293,7 +289,7 @@ static struct cxl_region *cxl_dpa_to_region(struct cxl_memdev *cxlmd, u64 dpa)
 		.dpa = dpa,
 	};
 	port = cxlmd->endpoint;
-	if (port && is_cxl_endpoint(port) && port->commit_end != -1)
+	if (port && is_cxl_endpoint(port) && cxl_num_decoders_committed(port))
 		device_for_each_child(&port->dev, &ctx, __cxl_dpa_to_region);
 
 	return ctx.cxlr;
@@ -335,10 +331,16 @@ int cxl_inject_poison(struct cxl_memdev *cxlmd, u64 dpa)
 	if (!IS_ENABLED(CONFIG_DEBUG_FS))
 		return 0;
 
-	rc = down_read_interruptible(&cxl_dpa_rwsem);
+	rc = down_read_interruptible(&cxl_region_rwsem);
 	if (rc)
 		return rc;
 
+	rc = down_read_interruptible(&cxl_dpa_rwsem);
+	if (rc) {
+		up_read(&cxl_region_rwsem);
+		return rc;
+	}
+
 	rc = cxl_validate_poison_dpa(cxlmd, dpa);
 	if (rc)
 		goto out;
@@ -366,6 +368,7 @@ int cxl_inject_poison(struct cxl_memdev *cxlmd, u64 dpa)
 	trace_cxl_poison(cxlmd, cxlr, &record, 0, 0, CXL_POISON_TRACE_INJECT);
 out:
 	up_read(&cxl_dpa_rwsem);
+	up_read(&cxl_region_rwsem);
 
 	return rc;
 }
@@ -383,10 +386,16 @@ int cxl_clear_poison(struct cxl_memdev *cxlmd, u64 dpa)
 	if (!IS_ENABLED(CONFIG_DEBUG_FS))
 		return 0;
 
-	rc = down_read_interruptible(&cxl_dpa_rwsem);
+	rc = down_read_interruptible(&cxl_region_rwsem);
 	if (rc)
 		return rc;
 
+	rc = down_read_interruptible(&cxl_dpa_rwsem);
+	if (rc) {
+		up_read(&cxl_region_rwsem);
+		return rc;
+	}
+
 	rc = cxl_validate_poison_dpa(cxlmd, dpa);
 	if (rc)
 		goto out;
@@ -423,6 +432,7 @@ int cxl_clear_poison(struct cxl_memdev *cxlmd, u64 dpa)
 	trace_cxl_poison(cxlmd, cxlr, &record, 0, 0, CXL_POISON_TRACE_CLEAR);
 out:
 	up_read(&cxl_dpa_rwsem);
+	up_read(&cxl_region_rwsem);
 
 	return rc;
 }
@@ -437,13 +447,41 @@ static struct attribute *cxl_memdev_attributes[] = {
 	NULL,
 };
 
+static ssize_t pmem_qos_class_show(struct device *dev,
+				   struct device_attribute *attr, char *buf)
+{
+	struct cxl_memdev *cxlmd = to_cxl_memdev(dev);
+	struct cxl_dev_state *cxlds = cxlmd->cxlds;
+	struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlds);
+
+	return sysfs_emit(buf, "%d\n", mds->pmem_perf.qos_class);
+}
+
+static struct device_attribute dev_attr_pmem_qos_class =
+	__ATTR(qos_class, 0444, pmem_qos_class_show, NULL);
+
 static struct attribute *cxl_memdev_pmem_attributes[] = {
 	&dev_attr_pmem_size.attr,
+	&dev_attr_pmem_qos_class.attr,
 	NULL,
 };
 
+static ssize_t ram_qos_class_show(struct device *dev,
+				  struct device_attribute *attr, char *buf)
+{
+	struct cxl_memdev *cxlmd = to_cxl_memdev(dev);
+	struct cxl_dev_state *cxlds = cxlmd->cxlds;
+	struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlds);
+
+	return sysfs_emit(buf, "%d\n", mds->ram_perf.qos_class);
+}
+
+static struct device_attribute dev_attr_ram_qos_class =
+	__ATTR(qos_class, 0444, ram_qos_class_show, NULL);
+
 static struct attribute *cxl_memdev_ram_attributes[] = {
 	&dev_attr_ram_size.attr,
+	&dev_attr_ram_qos_class.attr,
 	NULL,
 };
 
@@ -467,14 +505,42 @@ static struct attribute_group cxl_memdev_attribute_group = {
 	.is_visible = cxl_memdev_visible,
 };
 
+static umode_t cxl_ram_visible(struct kobject *kobj, struct attribute *a, int n)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct cxl_memdev *cxlmd = to_cxl_memdev(dev);
+	struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlmd->cxlds);
+
+	if (a == &dev_attr_ram_qos_class.attr)
+		if (mds->ram_perf.qos_class == CXL_QOS_CLASS_INVALID)
+			return 0;
+
+	return a->mode;
+}
+
 static struct attribute_group cxl_memdev_ram_attribute_group = {
 	.name = "ram",
 	.attrs = cxl_memdev_ram_attributes,
+	.is_visible = cxl_ram_visible,
 };
 
+static umode_t cxl_pmem_visible(struct kobject *kobj, struct attribute *a, int n)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct cxl_memdev *cxlmd = to_cxl_memdev(dev);
+	struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlmd->cxlds);
+
+	if (a == &dev_attr_pmem_qos_class.attr)
+		if (mds->pmem_perf.qos_class == CXL_QOS_CLASS_INVALID)
+			return 0;
+
+	return a->mode;
+}
+
 static struct attribute_group cxl_memdev_pmem_attribute_group = {
 	.name = "pmem",
 	.attrs = cxl_memdev_pmem_attributes,
+	.is_visible = cxl_pmem_visible,
 };
 
 static umode_t cxl_memdev_security_visible(struct kobject *kobj,
@@ -509,6 +575,13 @@ static const struct attribute_group *cxl_memdev_attribute_groups[] = {
 	NULL,
 };
 
+void cxl_memdev_update_perf(struct cxl_memdev *cxlmd)
+{
+	sysfs_update_group(&cxlmd->dev.kobj, &cxl_memdev_ram_attribute_group);
+	sysfs_update_group(&cxlmd->dev.kobj, &cxl_memdev_pmem_attribute_group);
+}
+EXPORT_SYMBOL_NS_GPL(cxl_memdev_update_perf, CXL);
+
 static const struct device_type cxl_memdev_type = {
 	.name = "cxl_memdev",
 	.release = cxl_memdev_release,
@@ -556,21 +629,11 @@ void clear_exclusive_cxl_commands(struct cxl_memdev_state *mds,
 }
 EXPORT_SYMBOL_NS_GPL(clear_exclusive_cxl_commands, CXL);
 
-static void cxl_memdev_security_shutdown(struct device *dev)
-{
-	struct cxl_memdev *cxlmd = to_cxl_memdev(dev);
-	struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlmd->cxlds);
-
-	if (mds->security.poll)
-		cancel_delayed_work_sync(&mds->security.poll_dwork);
-}
-
 static void cxl_memdev_shutdown(struct device *dev)
 {
 	struct cxl_memdev *cxlmd = to_cxl_memdev(dev);
 
 	down_write(&cxl_memdev_rwsem);
-	cxl_memdev_security_shutdown(dev);
 	cxlmd->cxlds = NULL;
 	up_write(&cxl_memdev_rwsem);
 }
@@ -580,8 +643,8 @@ static void cxl_memdev_unregister(void *_cxlmd)
 	struct cxl_memdev *cxlmd = _cxlmd;
 	struct device *dev = &cxlmd->dev;
 
-	cxl_memdev_shutdown(dev);
 	cdev_device_del(&cxlmd->cdev, dev);
+	cxl_memdev_shutdown(dev);
 	put_device(dev);
 }
 
@@ -961,17 +1024,16 @@ static const struct fw_upload_ops cxl_memdev_fw_ops = {
         .cleanup = cxl_fw_cleanup,
 };
 
-static void devm_cxl_remove_fw_upload(void *fwl)
+static void cxl_remove_fw_upload(void *fwl)
 {
 	firmware_upload_unregister(fwl);
 }
 
-int cxl_memdev_setup_fw_upload(struct cxl_memdev_state *mds)
+int devm_cxl_setup_fw_upload(struct device *host, struct cxl_memdev_state *mds)
 {
 	struct cxl_dev_state *cxlds = &mds->cxlds;
 	struct device *dev = &cxlds->cxlmd->dev;
 	struct fw_upload *fwl;
-	int rc;
 
 	if (!test_bit(CXL_MEM_COMMAND_ID_GET_FW_INFO, mds->enabled_cmds))
 		return 0;
@@ -979,19 +1041,10 @@ int cxl_memdev_setup_fw_upload(struct cxl_memdev_state *mds)
 	fwl = firmware_upload_register(THIS_MODULE, dev, dev_name(dev),
 				       &cxl_memdev_fw_ops, mds);
 	if (IS_ERR(fwl))
-		return dev_err_probe(dev, PTR_ERR(fwl),
-				     "Failed to register firmware loader\n");
-
-	rc = devm_add_action_or_reset(cxlds->dev, devm_cxl_remove_fw_upload,
-				      fwl);
-	if (rc)
-		dev_err(dev,
-			"Failed to add firmware loader remove action: %d\n",
-			rc);
-
-	return rc;
+		return PTR_ERR(fwl);
+	return devm_add_action_or_reset(host, cxl_remove_fw_upload, fwl);
 }
-EXPORT_SYMBOL_NS_GPL(cxl_memdev_setup_fw_upload, CXL);
+EXPORT_SYMBOL_NS_GPL(devm_cxl_setup_fw_upload, CXL);
 
 static const struct file_operations cxl_memdev_fops = {
 	.owner = THIS_MODULE,
@@ -1002,36 +1055,8 @@ static const struct file_operations cxl_memdev_fops = {
 	.llseek = noop_llseek,
 };
 
-static void put_sanitize(void *data)
-{
-	struct cxl_memdev_state *mds = data;
-
-	sysfs_put(mds->security.sanitize_node);
-}
-
-static int cxl_memdev_security_init(struct cxl_memdev *cxlmd)
-{
-	struct cxl_dev_state *cxlds = cxlmd->cxlds;
-	struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlds);
-	struct device *dev = &cxlmd->dev;
-	struct kernfs_node *sec;
-
-	sec = sysfs_get_dirent(dev->kobj.sd, "security");
-	if (!sec) {
-		dev_err(dev, "sysfs_get_dirent 'security' failed\n");
-		return -ENODEV;
-	}
-	mds->security.sanitize_node = sysfs_get_dirent(sec, "state");
-	sysfs_put(sec);
-	if (!mds->security.sanitize_node) {
-		dev_err(dev, "sysfs_get_dirent 'state' failed\n");
-		return -ENODEV;
-	}
-
-	return devm_add_action_or_reset(cxlds->dev, put_sanitize, mds);
- }
-
-struct cxl_memdev *devm_cxl_add_memdev(struct cxl_dev_state *cxlds)
+struct cxl_memdev *devm_cxl_add_memdev(struct device *host,
+				       struct cxl_dev_state *cxlds)
 {
 	struct cxl_memdev *cxlmd;
 	struct device *dev;
@@ -1059,11 +1084,7 @@ struct cxl_memdev *devm_cxl_add_memdev(struct cxl_dev_state *cxlds)
 	if (rc)
 		goto err;
 
-	rc = cxl_memdev_security_init(cxlmd);
-	if (rc)
-		goto err;
-
-	rc = devm_add_action_or_reset(cxlds->dev, cxl_memdev_unregister, cxlmd);
+	rc = devm_add_action_or_reset(host, cxl_memdev_unregister, cxlmd);
 	if (rc)
 		return ERR_PTR(rc);
 	return cxlmd;
@@ -1079,6 +1100,50 @@ err:
 }
 EXPORT_SYMBOL_NS_GPL(devm_cxl_add_memdev, CXL);
 
+static void sanitize_teardown_notifier(void *data)
+{
+	struct cxl_memdev_state *mds = data;
+	struct kernfs_node *state;
+
+	/*
+	 * Prevent new irq triggered invocations of the workqueue and
+	 * flush inflight invocations.
+	 */
+	mutex_lock(&mds->mbox_mutex);
+	state = mds->security.sanitize_node;
+	mds->security.sanitize_node = NULL;
+	mutex_unlock(&mds->mbox_mutex);
+
+	cancel_delayed_work_sync(&mds->security.poll_dwork);
+	sysfs_put(state);
+}
+
+int devm_cxl_sanitize_setup_notifier(struct device *host,
+				     struct cxl_memdev *cxlmd)
+{
+	struct cxl_dev_state *cxlds = cxlmd->cxlds;
+	struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlds);
+	struct kernfs_node *sec;
+
+	if (!test_bit(CXL_SEC_ENABLED_SANITIZE, mds->security.enabled_cmds))
+		return 0;
+
+	/*
+	 * Note, the expectation is that @cxlmd would have failed to be
+	 * created if these sysfs_get_dirent calls fail.
+	 */
+	sec = sysfs_get_dirent(cxlmd->dev.kobj.sd, "security");
+	if (!sec)
+		return -ENOENT;
+	mds->security.sanitize_node = sysfs_get_dirent(sec, "state");
+	sysfs_put(sec);
+	if (!mds->security.sanitize_node)
+		return -ENOENT;
+
+	return devm_add_action_or_reset(host, sanitize_teardown_notifier, mds);
+}
+EXPORT_SYMBOL_NS_GPL(devm_cxl_sanitize_setup_notifier, CXL);
+
 __init int cxl_memdev_init(void)
 {
 	dev_t devt;