diff mbox series

[V6,XRT,Alveo,08/20] fpga: xrt: driver infrastructure

Message ID 20210512015339.5649-9-lizhi.hou@xilinx.com (mailing list archive)
State New
Headers show
Series XRT Alveo driver overview | expand

Commit Message

Lizhi Hou May 12, 2021, 1:53 a.m. UTC
Infrastructure code providing APIs for managing leaf driver instance
groups, facilitating inter-leaf driver calls and root calls.

Signed-off-by: Sonal Santan <sonal.santan@xilinx.com>
Signed-off-by: Max Zhen <max.zhen@xilinx.com>
Signed-off-by: Lizhi Hou <lizhi.hou@xilinx.com>
Reviewed-by: Tom Rix <trix@redhat.com>
---
 drivers/fpga/xrt/lib/subdev.c | 848 ++++++++++++++++++++++++++++++++++
 1 file changed, 848 insertions(+)
 create mode 100644 drivers/fpga/xrt/lib/subdev.c

Comments

Tom Rix May 13, 2021, 3:27 p.m. UTC | #1
On 5/11/21 6:53 PM, Lizhi Hou wrote:
> Infrastructure code providing APIs for managing leaf driver instance
> groups, facilitating inter-leaf driver calls and root calls.
>
> Signed-off-by: Sonal Santan <sonal.santan@xilinx.com>
> Signed-off-by: Max Zhen <max.zhen@xilinx.com>
> Signed-off-by: Lizhi Hou <lizhi.hou@xilinx.com>
> Reviewed-by: Tom Rix <trix@redhat.com>

> +
> +int xrt_subdev_root_request(struct xrt_device *self, u32 cmd, void *arg)
> +{
> +	struct device *dev = DEV(self);
> +	struct xrt_subdev_platdata *pdata = DEV_PDATA(self);
> +
> +	WARN_ON(!pdata->xsp_root_cb);
> +	return (*pdata->xsp_root_cb)(dev->parent, pdata->xsp_root_cb_arg, cmd, arg);

I am running v6 through clang's static analyzer.

It reports this problem

drivers/fpga/xrt/lib/subdev.c:72:9: warning: Called function pointer is 
null (null dereference)
         return (*pdata->xsp_root_cb)(dev->parent, 
pdata->xsp_root_cb_arg, cmd, arg);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Though there is a warn_on, can you also add a

return -EINVAL;

To cover this case?

Tom

> +}
Lizhi Hou May 20, 2021, 2:48 a.m. UTC | #2
On 05/13/2021 08:27 AM, Tom Rix wrote:
>
> On 5/11/21 6:53 PM, Lizhi Hou wrote:
>> Infrastructure code providing APIs for managing leaf driver instance
>> groups, facilitating inter-leaf driver calls and root calls.
>>
>> Signed-off-by: Sonal Santan <sonal.santan@xilinx.com>
>> Signed-off-by: Max Zhen <max.zhen@xilinx.com>
>> Signed-off-by: Lizhi Hou <lizhi.hou@xilinx.com>
>> Reviewed-by: Tom Rix <trix@redhat.com>
>
>> +
>> +int xrt_subdev_root_request(struct xrt_device *self, u32 cmd, void 
>> *arg)
>> +{
>> +     struct device *dev = DEV(self);
>> +     struct xrt_subdev_platdata *pdata = DEV_PDATA(self);
>> +
>> +     WARN_ON(!pdata->xsp_root_cb);
>> +     return (*pdata->xsp_root_cb)(dev->parent, 
>> pdata->xsp_root_cb_arg, cmd, arg);
>
> I am running v6 through clang's static analyzer.
>
> It reports this problem
>
> drivers/fpga/xrt/lib/subdev.c:72:9: warning: Called function pointer is
> null (null dereference)
>         return (*pdata->xsp_root_cb)(dev->parent,
> pdata->xsp_root_cb_arg, cmd, arg);
> ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> Though there is a warn_on, can you also add a
>
> return -EINVAL;
>
> To cover this case?
Sure. I will run through clang myself to check the fix.

Thanks,
Lizhi
>
> Tom
>
>> +}
>
diff mbox series

Patch

diff --git a/drivers/fpga/xrt/lib/subdev.c b/drivers/fpga/xrt/lib/subdev.c
new file mode 100644
index 000000000000..0713a8e7a40f
--- /dev/null
+++ b/drivers/fpga/xrt/lib/subdev.c
@@ -0,0 +1,848 @@ 
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (C) 2020-2021 Xilinx, Inc.
+ *
+ * Authors:
+ *	Cheng Zhen <maxz@xilinx.com>
+ */
+
+#include <linux/vmalloc.h>
+#include <linux/slab.h>
+#include "xleaf.h"
+#include "subdev_pool.h"
+#include "lib-drv.h"
+#include "metadata.h"
+
+extern struct bus_type xrt_bus_type;
+
+#define IS_ROOT_DEV(dev)	((dev)->bus != &xrt_bus_type)
+#define XRT_HOLDER_BUF_SZ	1024
+
+static inline struct device *find_root(struct xrt_device *xdev)
+{
+	struct device *d = DEV(xdev);
+
+	while (!IS_ROOT_DEV(d))
+		d = d->parent;
+	return d;
+}
+
+/*
+ * It represents a holder of a subdev. One holder can repeatedly hold a subdev
+ * as long as there is a unhold corresponding to a hold.
+ */
+struct xrt_subdev_holder {
+	struct list_head xsh_holder_list;
+	struct device *xsh_holder;
+	int xsh_count;
+	struct kref xsh_kref;
+};
+
+/*
+ * It represents a specific instance of platform driver for a subdev, which
+ * provides services to its clients (another subdev driver or root driver).
+ */
+struct xrt_subdev {
+	struct list_head xs_dev_list;
+	struct list_head xs_holder_list;
+	enum xrt_subdev_id xs_id;		/* type of subdev */
+	struct xrt_device *xs_xdev;
+	struct completion xs_holder_comp;
+};
+
+static struct xrt_subdev *xrt_subdev_alloc(void)
+{
+	struct xrt_subdev *sdev = kzalloc(sizeof(*sdev), GFP_KERNEL);
+
+	if (!sdev)
+		return NULL;
+
+	INIT_LIST_HEAD(&sdev->xs_dev_list);
+	INIT_LIST_HEAD(&sdev->xs_holder_list);
+	init_completion(&sdev->xs_holder_comp);
+	return sdev;
+}
+
+int xrt_subdev_root_request(struct xrt_device *self, u32 cmd, void *arg)
+{
+	struct device *dev = DEV(self);
+	struct xrt_subdev_platdata *pdata = DEV_PDATA(self);
+
+	WARN_ON(!pdata->xsp_root_cb);
+	return (*pdata->xsp_root_cb)(dev->parent, pdata->xsp_root_cb_arg, cmd, arg);
+}
+
+/*
+ * Subdev common sysfs nodes.
+ */
+static ssize_t holders_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	ssize_t len;
+	struct xrt_device *xdev = to_xrt_dev(dev);
+	struct xrt_root_get_holders holders = { xdev, buf, XRT_HOLDER_BUF_SZ };
+
+	len = xrt_subdev_root_request(xdev, XRT_ROOT_GET_LEAF_HOLDERS, &holders);
+	if (len >= holders.xpigh_holder_buf_len)
+		return len;
+	buf[len] = '\n';
+	return len + 1;
+}
+static DEVICE_ATTR_RO(holders);
+
+static struct attribute *xrt_subdev_attrs[] = {
+	&dev_attr_holders.attr,
+	NULL,
+};
+
+static ssize_t metadata_output(struct file *filp, struct kobject *kobj,
+			       struct bin_attribute *attr, char *buf, loff_t off, size_t count)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct xrt_device *xdev = to_xrt_dev(dev);
+	struct xrt_subdev_platdata *pdata = DEV_PDATA(xdev);
+	unsigned char *blob;
+	unsigned long  size;
+	ssize_t ret = 0;
+
+	blob = pdata->xsp_dtb;
+	size = xrt_md_size(dev, blob);
+	if (size == XRT_MD_INVALID_LENGTH) {
+		ret = -EINVAL;
+		goto failed;
+	}
+
+	if (off >= size) {
+		dev_dbg(dev, "offset (%lld) beyond total size: %ld\n", off, size);
+		goto failed;
+	}
+
+	if (off + count > size) {
+		dev_dbg(dev, "count (%ld) beyond left bytes: %lld\n",
+			(unsigned long)count, size - off);
+		count = size - off;
+	}
+	memcpy(buf, blob + off, count);
+
+	ret = count;
+failed:
+	return ret;
+}
+
+static struct bin_attribute meta_data_attr = {
+	.attr = {
+		.name = "metadata",
+		.mode = 0400
+	},
+	.read = metadata_output,
+	.size = 0
+};
+
+static struct bin_attribute  *xrt_subdev_bin_attrs[] = {
+	&meta_data_attr,
+	NULL,
+};
+
+static const struct attribute_group xrt_subdev_attrgroup = {
+	.attrs = xrt_subdev_attrs,
+	.bin_attrs = xrt_subdev_bin_attrs,
+};
+
+/*
+ * Given the device metadata, parse it to get IO ranges and construct
+ * resource array.
+ */
+static int
+xrt_subdev_getres(struct device *parent, enum xrt_subdev_id id,
+		  char *dtb, struct resource **res, int *res_num)
+{
+	struct xrt_subdev_platdata *pdata;
+	struct resource *pci_res = NULL;
+	const __be64 *bar_range;
+	const __be32 *bar_idx;
+	char *ep_name = NULL, *compat = NULL;
+	uint bar;
+	int count1 = 0, count2 = 0, ret;
+
+	if (!dtb)
+		return -EINVAL;
+
+	pdata = DEV_PDATA(to_xrt_dev(parent));
+
+	/* go through metadata and count endpoints in it */
+	xrt_md_get_next_endpoint(parent, dtb, NULL, NULL, &ep_name, &compat);
+	while (ep_name) {
+		ret = xrt_md_get_prop(parent, dtb, ep_name, compat,
+				      XRT_MD_PROP_IO_OFFSET, (const void **)&bar_range, NULL);
+		if (!ret)
+			count1++;
+		xrt_md_get_next_endpoint(parent, dtb, ep_name, compat, &ep_name, &compat);
+	}
+	if (!count1)
+		return 0;
+
+	/* allocate resource array for all endpoints been found in metadata */
+	*res = vzalloc(sizeof(**res) * count1);
+
+	/* go through all endpoints again and get IO range for each endpoint */
+	ep_name = NULL;
+	xrt_md_get_next_endpoint(parent, dtb, NULL, NULL, &ep_name, &compat);
+	while (ep_name) {
+		ret = xrt_md_get_prop(parent, dtb, ep_name, compat,
+				      XRT_MD_PROP_IO_OFFSET, (const void **)&bar_range, NULL);
+		if (ret)
+			continue;
+		xrt_md_get_prop(parent, dtb, ep_name, compat,
+				XRT_MD_PROP_BAR_IDX, (const void **)&bar_idx, NULL);
+		bar = bar_idx ? be32_to_cpu(*bar_idx) : 0;
+		xleaf_get_root_res(to_xrt_dev(parent), bar, &pci_res);
+		(*res)[count2].start = pci_res->start + be64_to_cpu(bar_range[0]);
+		(*res)[count2].end = pci_res->start + be64_to_cpu(bar_range[0]) +
+			be64_to_cpu(bar_range[1]) - 1;
+		(*res)[count2].flags = IORESOURCE_MEM;
+		/* check if there is conflicted resource */
+		ret = request_resource(pci_res, *res + count2);
+		if (ret) {
+			dev_err(parent, "Conflict resource %pR\n", *res + count2);
+			vfree(*res);
+			*res_num = 0;
+			*res = NULL;
+			return ret;
+		}
+		release_resource(*res + count2);
+
+		(*res)[count2].parent = pci_res;
+
+		xrt_md_find_endpoint(parent, pdata->xsp_dtb, ep_name,
+				     compat, &(*res)[count2].name);
+
+		count2++;
+		xrt_md_get_next_endpoint(parent, dtb, ep_name, compat, &ep_name, &compat);
+	}
+
+	WARN_ON(count1 != count2);
+	*res_num = count2;
+
+	return 0;
+}
+
+static inline enum xrt_dev_file_mode
+xleaf_devnode_mode(struct xrt_device *xdev)
+{
+	return DEV_FILE_OPS(xdev)->xsf_mode;
+}
+
+static bool xrt_subdev_cdev_auto_creation(struct xrt_device *xdev)
+{
+	enum xrt_dev_file_mode mode = xleaf_devnode_mode(xdev);
+
+	if (!xleaf_devnode_enabled(xdev))
+		return false;
+
+	return (mode == XRT_DEV_FILE_DEFAULT || mode == XRT_DEV_FILE_MULTI_INST);
+}
+
+static struct xrt_subdev *
+xrt_subdev_create(struct device *parent, enum xrt_subdev_id id,
+		  xrt_subdev_root_cb_t pcb, void *pcb_arg, char *dtb)
+{
+	struct xrt_subdev_platdata *pdata = NULL;
+	struct xrt_subdev *sdev = NULL;
+	struct xrt_device *xdev = NULL;
+	struct resource *res = NULL;
+	unsigned long dtb_len = 0;
+	bool dtb_alloced = false;
+	int res_num = 0;
+	size_t pdata_sz;
+	int ret;
+
+	sdev = xrt_subdev_alloc();
+	if (!sdev) {
+		dev_err(parent, "failed to alloc subdev for ID %d", id);
+		return NULL;
+	}
+	sdev->xs_id = id;
+
+	if (!dtb) {
+		ret = xrt_md_create(parent, &dtb);
+		if (ret) {
+			dev_err(parent, "can't create empty dtb: %d", ret);
+			goto fail;
+		}
+		dtb_alloced = true;
+	}
+	xrt_md_pack(parent, dtb);
+	dtb_len = xrt_md_size(parent, dtb);
+	if (dtb_len == XRT_MD_INVALID_LENGTH) {
+		dev_err(parent, "invalid metadata len %ld", dtb_len);
+		goto fail1;
+	}
+	pdata_sz = sizeof(struct xrt_subdev_platdata) + dtb_len;
+
+	/* Prepare platform data passed to subdev. */
+	pdata = vzalloc(pdata_sz);
+	if (!pdata)
+		goto fail1;
+
+	pdata->xsp_root_cb = pcb;
+	pdata->xsp_root_cb_arg = pcb_arg;
+	memcpy(pdata->xsp_dtb, dtb, dtb_len);
+	if (id == XRT_SUBDEV_GRP) {
+		/* Group can only be created by root driver. */
+		pdata->xsp_root_name = dev_name(parent);
+	} else {
+		struct xrt_device *grp = to_xrt_dev(parent);
+
+		/* Leaf can only be created by group driver. */
+		WARN_ON(to_xrt_drv(parent->driver)->subdev_id != XRT_SUBDEV_GRP);
+		pdata->xsp_root_name = DEV_PDATA(grp)->xsp_root_name;
+	}
+
+	/* Create subdev. */
+	if (id != XRT_SUBDEV_GRP) {
+		int rc = xrt_subdev_getres(parent, id, dtb, &res, &res_num);
+
+		if (rc) {
+			dev_err(parent, "failed to get resource for %s: %d",
+				xrt_drv_name(id), rc);
+			goto fail2;
+		}
+	}
+	xdev = xrt_device_register(parent, id, res, res_num, pdata, pdata_sz);
+	vfree(res);
+	if (!xdev) {
+		dev_err(parent, "failed to create subdev for %s", xrt_drv_name(id));
+		goto fail2;
+	}
+	sdev->xs_xdev = xdev;
+
+	if (device_attach(DEV(xdev)) != 1) {
+		xrt_err(xdev, "failed to attach");
+		goto fail3;
+	}
+
+	if (sysfs_create_group(&DEV(xdev)->kobj, &xrt_subdev_attrgroup))
+		xrt_err(xdev, "failed to create sysfs group");
+
+	/*
+	 * Create sysfs sym link under root for leaves
+	 * under random groups for easy access to them.
+	 */
+	if (id != XRT_SUBDEV_GRP) {
+		if (sysfs_create_link(&find_root(xdev)->kobj,
+				      &DEV(xdev)->kobj, dev_name(DEV(xdev)))) {
+			xrt_err(xdev, "failed to create sysfs link");
+		}
+	}
+
+	/* All done, ready to handle req thru cdev. */
+	if (xrt_subdev_cdev_auto_creation(xdev))
+		xleaf_devnode_create(xdev, DEV_FILE_OPS(xdev)->xsf_dev_name, NULL);
+
+	vfree(pdata);
+	return sdev;
+
+fail3:
+	xrt_device_unregister(sdev->xs_xdev);
+fail2:
+	vfree(pdata);
+fail1:
+	if (dtb_alloced)
+		vfree(dtb);
+fail:
+	kfree(sdev);
+	return NULL;
+}
+
+static void xrt_subdev_destroy(struct xrt_subdev *sdev)
+{
+	struct xrt_device *xdev = sdev->xs_xdev;
+	struct device *dev = DEV(xdev);
+
+	/* Take down the device node */
+	if (xrt_subdev_cdev_auto_creation(xdev))
+		xleaf_devnode_destroy(xdev);
+	if (sdev->xs_id != XRT_SUBDEV_GRP)
+		sysfs_remove_link(&find_root(xdev)->kobj, dev_name(dev));
+	sysfs_remove_group(&dev->kobj, &xrt_subdev_attrgroup);
+	xrt_device_unregister(xdev);
+	kfree(sdev);
+}
+
+struct xrt_device *
+xleaf_get_leaf(struct xrt_device *xdev, xrt_subdev_match_t match_cb, void *match_arg)
+{
+	int rc;
+	struct xrt_root_get_leaf get_leaf = {
+		xdev, match_cb, match_arg, };
+
+	rc = xrt_subdev_root_request(xdev, XRT_ROOT_GET_LEAF, &get_leaf);
+	if (rc)
+		return NULL;
+	return get_leaf.xpigl_tgt_xdev;
+}
+EXPORT_SYMBOL_GPL(xleaf_get_leaf);
+
+bool xleaf_has_endpoint(struct xrt_device *xdev, const char *endpoint_name)
+{
+	struct resource	*res;
+	int i = 0;
+
+	do {
+		res = xrt_get_resource(xdev, IORESOURCE_MEM, i);
+		if (res && !strncmp(res->name, endpoint_name, strlen(res->name) + 1))
+			return true;
+		++i;
+	} while (res);
+
+	return false;
+}
+EXPORT_SYMBOL_GPL(xleaf_has_endpoint);
+
+int xleaf_put_leaf(struct xrt_device *xdev, struct xrt_device *leaf)
+{
+	struct xrt_root_put_leaf put_leaf = { xdev, leaf };
+
+	return xrt_subdev_root_request(xdev, XRT_ROOT_PUT_LEAF, &put_leaf);
+}
+EXPORT_SYMBOL_GPL(xleaf_put_leaf);
+
+int xleaf_create_group(struct xrt_device *xdev, char *dtb)
+{
+	return xrt_subdev_root_request(xdev, XRT_ROOT_CREATE_GROUP, dtb);
+}
+EXPORT_SYMBOL_GPL(xleaf_create_group);
+
+int xleaf_destroy_group(struct xrt_device *xdev, int instance)
+{
+	return xrt_subdev_root_request(xdev, XRT_ROOT_REMOVE_GROUP, (void *)(uintptr_t)instance);
+}
+EXPORT_SYMBOL_GPL(xleaf_destroy_group);
+
+int xleaf_wait_for_group_bringup(struct xrt_device *xdev)
+{
+	return xrt_subdev_root_request(xdev, XRT_ROOT_WAIT_GROUP_BRINGUP, NULL);
+}
+EXPORT_SYMBOL_GPL(xleaf_wait_for_group_bringup);
+
+static ssize_t
+xrt_subdev_get_holders(struct xrt_subdev *sdev, char *buf, size_t len)
+{
+	const struct list_head *ptr;
+	struct xrt_subdev_holder *h;
+	ssize_t n = 0;
+
+	list_for_each(ptr, &sdev->xs_holder_list) {
+		h = list_entry(ptr, struct xrt_subdev_holder, xsh_holder_list);
+		n += snprintf(buf + n, len - n, "%s:%d ",
+			      dev_name(h->xsh_holder), kref_read(&h->xsh_kref));
+		/* Truncation is fine here. Buffer content is only for debugging. */
+		if (n >= (len - 1))
+			break;
+	}
+	return n;
+}
+
+void xrt_subdev_pool_init(struct device *dev, struct xrt_subdev_pool *spool)
+{
+	INIT_LIST_HEAD(&spool->xsp_dev_list);
+	spool->xsp_owner = dev;
+	mutex_init(&spool->xsp_lock);
+	spool->xsp_closing = false;
+}
+
+static void xrt_subdev_free_holder(struct xrt_subdev_holder *holder)
+{
+	list_del(&holder->xsh_holder_list);
+	vfree(holder);
+}
+
+static void xrt_subdev_pool_wait_for_holders(struct xrt_subdev_pool *spool, struct xrt_subdev *sdev)
+{
+	const struct list_head *ptr, *next;
+	char holders[128];
+	struct xrt_subdev_holder *holder;
+	struct mutex *lk = &spool->xsp_lock;
+
+	while (!list_empty(&sdev->xs_holder_list)) {
+		int rc;
+
+		/* It's most likely a bug if we ever enters this loop. */
+		xrt_subdev_get_holders(sdev, holders, sizeof(holders));
+		xrt_err(sdev->xs_xdev, "awaits holders: %s", holders);
+		mutex_unlock(lk);
+		rc = wait_for_completion_killable(&sdev->xs_holder_comp);
+		mutex_lock(lk);
+		if (rc == -ERESTARTSYS) {
+			xrt_err(sdev->xs_xdev, "give up on waiting for holders, clean up now");
+			list_for_each_safe(ptr, next, &sdev->xs_holder_list) {
+				holder = list_entry(ptr, struct xrt_subdev_holder, xsh_holder_list);
+				xrt_subdev_free_holder(holder);
+			}
+		}
+	}
+}
+
+void xrt_subdev_pool_fini(struct xrt_subdev_pool *spool)
+{
+	struct list_head *dl = &spool->xsp_dev_list;
+	struct mutex *lk = &spool->xsp_lock;
+
+	mutex_lock(lk);
+	if (spool->xsp_closing) {
+		mutex_unlock(lk);
+		return;
+	}
+	spool->xsp_closing = true;
+	mutex_unlock(lk);
+
+	/* Remove subdev in the reverse order of added. */
+	while (!list_empty(dl)) {
+		struct xrt_subdev *sdev = list_first_entry(dl, struct xrt_subdev, xs_dev_list);
+
+		xrt_subdev_pool_wait_for_holders(spool, sdev);
+		list_del(&sdev->xs_dev_list);
+		xrt_subdev_destroy(sdev);
+	}
+}
+
+static struct xrt_subdev_holder *xrt_subdev_find_holder(struct xrt_subdev *sdev,
+							struct device *holder_dev)
+{
+	struct list_head *hl = &sdev->xs_holder_list;
+	struct xrt_subdev_holder *holder;
+	const struct list_head *ptr;
+
+	list_for_each(ptr, hl) {
+		holder = list_entry(ptr, struct xrt_subdev_holder, xsh_holder_list);
+		if (holder->xsh_holder == holder_dev)
+			return holder;
+	}
+	return NULL;
+}
+
+static int xrt_subdev_hold(struct xrt_subdev *sdev, struct device *holder_dev)
+{
+	struct xrt_subdev_holder *holder = xrt_subdev_find_holder(sdev, holder_dev);
+	struct list_head *hl = &sdev->xs_holder_list;
+
+	if (!holder) {
+		holder = vzalloc(sizeof(*holder));
+		if (!holder)
+			return -ENOMEM;
+		holder->xsh_holder = holder_dev;
+		kref_init(&holder->xsh_kref);
+		list_add_tail(&holder->xsh_holder_list, hl);
+	} else {
+		kref_get(&holder->xsh_kref);
+	}
+
+	return 0;
+}
+
+static void xrt_subdev_free_holder_kref(struct kref *kref)
+{
+	struct xrt_subdev_holder *holder = container_of(kref, struct xrt_subdev_holder, xsh_kref);
+
+	xrt_subdev_free_holder(holder);
+}
+
+static int
+xrt_subdev_release(struct xrt_subdev *sdev, struct device *holder_dev)
+{
+	struct xrt_subdev_holder *holder = xrt_subdev_find_holder(sdev, holder_dev);
+	struct list_head *hl = &sdev->xs_holder_list;
+
+	if (!holder) {
+		dev_err(holder_dev, "can't release, %s did not hold %s",
+			dev_name(holder_dev), dev_name(DEV(sdev->xs_xdev)));
+		return -EINVAL;
+	}
+	kref_put(&holder->xsh_kref, xrt_subdev_free_holder_kref);
+
+	/* kref_put above may remove holder from list. */
+	if (list_empty(hl))
+		complete(&sdev->xs_holder_comp);
+	return 0;
+}
+
+int xrt_subdev_pool_add(struct xrt_subdev_pool *spool, enum xrt_subdev_id id,
+			xrt_subdev_root_cb_t pcb, void *pcb_arg, char *dtb)
+{
+	struct mutex *lk = &spool->xsp_lock;
+	struct list_head *dl = &spool->xsp_dev_list;
+	struct xrt_subdev *sdev;
+	int ret = 0;
+
+	sdev = xrt_subdev_create(spool->xsp_owner, id, pcb, pcb_arg, dtb);
+	if (sdev) {
+		mutex_lock(lk);
+		if (spool->xsp_closing) {
+			/* No new subdev when pool is going away. */
+			xrt_err(sdev->xs_xdev, "pool is closing");
+			ret = -ENODEV;
+		} else {
+			list_add(&sdev->xs_dev_list, dl);
+		}
+		mutex_unlock(lk);
+		if (ret)
+			xrt_subdev_destroy(sdev);
+	} else {
+		ret = -EINVAL;
+	}
+
+	ret = ret ? ret : sdev->xs_xdev->instance;
+	return ret;
+}
+
+int xrt_subdev_pool_del(struct xrt_subdev_pool *spool, enum xrt_subdev_id id, int instance)
+{
+	const struct list_head *ptr;
+	struct mutex *lk = &spool->xsp_lock;
+	struct list_head *dl = &spool->xsp_dev_list;
+	struct xrt_subdev *sdev;
+	int ret = -ENOENT;
+
+	mutex_lock(lk);
+	if (spool->xsp_closing) {
+		/* Pool is going away, all subdevs will be gone. */
+		mutex_unlock(lk);
+		return 0;
+	}
+	list_for_each(ptr, dl) {
+		sdev = list_entry(ptr, struct xrt_subdev, xs_dev_list);
+		if (sdev->xs_id != id || sdev->xs_xdev->instance != instance)
+			continue;
+		xrt_subdev_pool_wait_for_holders(spool, sdev);
+		list_del(&sdev->xs_dev_list);
+		ret = 0;
+		break;
+	}
+	mutex_unlock(lk);
+	if (ret)
+		return ret;
+
+	xrt_subdev_destroy(sdev);
+	return 0;
+}
+
+static int xrt_subdev_pool_get_impl(struct xrt_subdev_pool *spool, xrt_subdev_match_t match,
+				    void *arg, struct device *holder_dev, struct xrt_subdev **sdevp)
+{
+	struct xrt_device *xdev = (struct xrt_device *)arg;
+	struct list_head *dl = &spool->xsp_dev_list;
+	struct mutex *lk = &spool->xsp_lock;
+	struct xrt_subdev *sdev = NULL;
+	const struct list_head *ptr;
+	struct xrt_subdev *d = NULL;
+	int ret = -ENOENT;
+
+	mutex_lock(lk);
+
+	if (!xdev) {
+		if (match == XRT_SUBDEV_MATCH_PREV) {
+			sdev = list_empty(dl) ? NULL :
+				list_last_entry(dl, struct xrt_subdev, xs_dev_list);
+		} else if (match == XRT_SUBDEV_MATCH_NEXT) {
+			sdev = list_first_entry_or_null(dl, struct xrt_subdev, xs_dev_list);
+		}
+	}
+
+	list_for_each(ptr, dl) {
+		d = list_entry(ptr, struct xrt_subdev, xs_dev_list);
+		if (match == XRT_SUBDEV_MATCH_PREV || match == XRT_SUBDEV_MATCH_NEXT) {
+			if (d->xs_xdev != xdev)
+				continue;
+		} else {
+			if (!match(d->xs_id, d->xs_xdev, arg))
+				continue;
+		}
+
+		if (match == XRT_SUBDEV_MATCH_PREV)
+			sdev = !list_is_first(ptr, dl) ? list_prev_entry(d, xs_dev_list) : NULL;
+		else if (match == XRT_SUBDEV_MATCH_NEXT)
+			sdev = !list_is_last(ptr, dl) ? list_next_entry(d, xs_dev_list) : NULL;
+		else
+			sdev = d;
+	}
+
+	if (sdev)
+		ret = xrt_subdev_hold(sdev, holder_dev);
+
+	mutex_unlock(lk);
+
+	if (!ret)
+		*sdevp = sdev;
+	return ret;
+}
+
+int xrt_subdev_pool_get(struct xrt_subdev_pool *spool, xrt_subdev_match_t match, void *arg,
+			struct device *holder_dev, struct xrt_device **xdevp)
+{
+	int rc;
+	struct xrt_subdev *sdev;
+
+	rc = xrt_subdev_pool_get_impl(spool, match, arg, holder_dev, &sdev);
+	if (rc) {
+		if (rc != -ENOENT)
+			dev_err(holder_dev, "failed to hold device: %d", rc);
+		return rc;
+	}
+
+	if (!IS_ROOT_DEV(holder_dev)) {
+		xrt_dbg(to_xrt_dev(holder_dev), "%s <<==== %s",
+			dev_name(holder_dev), dev_name(DEV(sdev->xs_xdev)));
+	}
+
+	*xdevp = sdev->xs_xdev;
+	return 0;
+}
+
+static int xrt_subdev_pool_put_impl(struct xrt_subdev_pool *spool, struct xrt_device *xdev,
+				    struct device *holder_dev)
+{
+	const struct list_head *ptr;
+	struct mutex *lk = &spool->xsp_lock;
+	struct list_head *dl = &spool->xsp_dev_list;
+	struct xrt_subdev *sdev;
+	int ret = -ENOENT;
+
+	mutex_lock(lk);
+	list_for_each(ptr, dl) {
+		sdev = list_entry(ptr, struct xrt_subdev, xs_dev_list);
+		if (sdev->xs_xdev != xdev)
+			continue;
+		ret = xrt_subdev_release(sdev, holder_dev);
+		break;
+	}
+	mutex_unlock(lk);
+
+	return ret;
+}
+
+int xrt_subdev_pool_put(struct xrt_subdev_pool *spool, struct xrt_device *xdev,
+			struct device *holder_dev)
+{
+	int ret = xrt_subdev_pool_put_impl(spool, xdev, holder_dev);
+
+	if (ret)
+		return ret;
+
+	if (!IS_ROOT_DEV(holder_dev)) {
+		xrt_dbg(to_xrt_dev(holder_dev), "%s <<==X== %s",
+			dev_name(holder_dev), dev_name(DEV(xdev)));
+	}
+	return 0;
+}
+
+void xrt_subdev_pool_trigger_event(struct xrt_subdev_pool *spool, enum xrt_events e)
+{
+	struct xrt_device *tgt = NULL;
+	struct xrt_subdev *sdev = NULL;
+	struct xrt_event evt;
+
+	while (!xrt_subdev_pool_get_impl(spool, XRT_SUBDEV_MATCH_NEXT,
+					 tgt, spool->xsp_owner, &sdev)) {
+		tgt = sdev->xs_xdev;
+		evt.xe_evt = e;
+		evt.xe_subdev.xevt_subdev_id = sdev->xs_id;
+		evt.xe_subdev.xevt_subdev_instance = tgt->instance;
+		xrt_subdev_root_request(tgt, XRT_ROOT_EVENT_SYNC, &evt);
+		xrt_subdev_pool_put_impl(spool, tgt, spool->xsp_owner);
+	}
+}
+
+void xrt_subdev_pool_handle_event(struct xrt_subdev_pool *spool, struct xrt_event *evt)
+{
+	struct xrt_device *tgt = NULL;
+	struct xrt_subdev *sdev = NULL;
+
+	while (!xrt_subdev_pool_get_impl(spool, XRT_SUBDEV_MATCH_NEXT,
+					 tgt, spool->xsp_owner, &sdev)) {
+		tgt = sdev->xs_xdev;
+		xleaf_call(tgt, XRT_XLEAF_EVENT, evt);
+		xrt_subdev_pool_put_impl(spool, tgt, spool->xsp_owner);
+	}
+}
+
+ssize_t xrt_subdev_pool_get_holders(struct xrt_subdev_pool *spool,
+				    struct xrt_device *xdev, char *buf, size_t len)
+{
+	const struct list_head *ptr;
+	struct mutex *lk = &spool->xsp_lock;
+	struct list_head *dl = &spool->xsp_dev_list;
+	struct xrt_subdev *sdev;
+	ssize_t ret = 0;
+
+	mutex_lock(lk);
+	list_for_each(ptr, dl) {
+		sdev = list_entry(ptr, struct xrt_subdev, xs_dev_list);
+		if (sdev->xs_xdev != xdev)
+			continue;
+		ret = xrt_subdev_get_holders(sdev, buf, len);
+		break;
+	}
+	mutex_unlock(lk);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(xrt_subdev_pool_get_holders);
+
+int xleaf_broadcast_event(struct xrt_device *xdev, enum xrt_events evt, bool async)
+{
+	struct xrt_event e = { evt, };
+	enum xrt_root_cmd cmd = async ? XRT_ROOT_EVENT_ASYNC : XRT_ROOT_EVENT_SYNC;
+
+	WARN_ON(evt == XRT_EVENT_POST_CREATION || evt == XRT_EVENT_PRE_REMOVAL);
+	return xrt_subdev_root_request(xdev, cmd, &e);
+}
+EXPORT_SYMBOL_GPL(xleaf_broadcast_event);
+
+void xleaf_hot_reset(struct xrt_device *xdev)
+{
+	xrt_subdev_root_request(xdev, XRT_ROOT_HOT_RESET, NULL);
+}
+EXPORT_SYMBOL_GPL(xleaf_hot_reset);
+
+void xleaf_get_root_res(struct xrt_device *xdev, u32 region_id, struct resource **res)
+{
+	struct xrt_root_get_res arg = { 0 };
+
+	arg.xpigr_region_id = region_id;
+	xrt_subdev_root_request(xdev, XRT_ROOT_GET_RESOURCE, &arg);
+	*res = arg.xpigr_res;
+}
+
+void xleaf_get_root_id(struct xrt_device *xdev, unsigned short *vendor, unsigned short *device,
+		       unsigned short *subvendor, unsigned short *subdevice)
+{
+	struct xrt_root_get_id id = { 0 };
+
+	WARN_ON(!vendor && !device && !subvendor && !subdevice);
+
+	xrt_subdev_root_request(xdev, XRT_ROOT_GET_ID, (void *)&id);
+	if (vendor)
+		*vendor = id.xpigi_vendor_id;
+	if (device)
+		*device = id.xpigi_device_id;
+	if (subvendor)
+		*subvendor = id.xpigi_sub_vendor_id;
+	if (subdevice)
+		*subdevice = id.xpigi_sub_device_id;
+}
+
+struct device *xleaf_register_hwmon(struct xrt_device *xdev, const char *name, void *drvdata,
+				    const struct attribute_group **grps)
+{
+	struct xrt_root_hwmon hm = { true, name, drvdata, grps, };
+
+	xrt_subdev_root_request(xdev, XRT_ROOT_HWMON, (void *)&hm);
+	return hm.xpih_hwmon_dev;
+}
+
+void xleaf_unregister_hwmon(struct xrt_device *xdev, struct device *hwmon)
+{
+	struct xrt_root_hwmon hm = { false, };
+
+	hm.xpih_hwmon_dev = hwmon;
+	xrt_subdev_root_request(xdev, XRT_ROOT_HWMON, (void *)&hm);
+}