* [PATCH 03/12] VMCI: doorbell implementation.
From: George Zhang @ 2012-11-07 18:41 UTC (permalink / raw)
To: linux-kernel, georgezhang, virtualization; +Cc: pv-drivers, gregkh
In-Reply-To: <20121107183624.9658.78903.stgit@promb-2n-dhcp175.eng.vmware.com>
VMCI doorbell code allows for notifcations between host and guest.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
drivers/misc/vmw_vmci/vmci_doorbell.c | 605 +++++++++++++++++++++++++++++++++
drivers/misc/vmw_vmci/vmci_doorbell.h | 51 +++
2 files changed, 656 insertions(+), 0 deletions(-)
create mode 100644 drivers/misc/vmw_vmci/vmci_doorbell.c
create mode 100644 drivers/misc/vmw_vmci/vmci_doorbell.h
diff --git a/drivers/misc/vmw_vmci/vmci_doorbell.c b/drivers/misc/vmw_vmci/vmci_doorbell.c
new file mode 100644
index 0000000..eb87ca6
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_doorbell.c
@@ -0,0 +1,605 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#include <linux/vmw_vmci_defs.h>
+#include <linux/vmw_vmci_api.h>
+#include <linux/completion.h>
+#include <linux/hash.h>
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+
+#include "vmci_common_int.h"
+#include "vmci_datagram.h"
+#include "vmci_doorbell.h"
+#include "vmci_resource.h"
+#include "vmci_driver.h"
+#include "vmci_route.h"
+
+
+#define VMCI_DOORBELL_INDEX_BITS 6
+#define VMCI_DOORBELL_INDEX_TABLE_SIZE (1 << VMCI_DOORBELL_INDEX_BITS)
+#define VMCI_DOORBELL_HASH(_idx) hash_32(_idx, VMCI_DOORBELL_INDEX_BITS)
+
+/*
+ * DoorbellEntry describes the a doorbell notification handle allocated by the
+ * host.
+ */
+struct dbell_entry {
+ struct vmci_resource resource;
+ struct hlist_node node;
+ struct work_struct work;
+ vmci_callback notify_cb;
+ void *client_data;
+ u32 idx;
+ u32 priv_flags;
+ bool run_delayed;
+ atomic_t active; /* Only used by guest personality */
+};
+
+/* The VMCI index table keeps track of currently registered doorbells. */
+struct dbell_index_table {
+ spinlock_t lock; /* Index table lock */
+ struct hlist_head entries[VMCI_DOORBELL_INDEX_TABLE_SIZE];
+};
+
+static struct dbell_index_table vmci_doorbell_it = {
+ .lock = __SPIN_LOCK_UNLOCKED(vmci_doorbell_it.lock),
+};
+
+/*
+ * The max_notify_idx is one larger than the currently known bitmap index in
+ * use, and is used to determine how much of the bitmap needs to be scanned.
+ */
+static u32 max_notify_idx;
+
+/*
+ * The notify_idx_count is used for determining whether there are free entries
+ * within the bitmap (if notify_idx_count + 1 < max_notify_idx).
+ */
+static u32 notify_idx_count;
+
+/*
+ * The last_notify_idx_reserved is used to track the last index handed out - in
+ * the case where multiple handles share a notification index, we hand out
+ * indexes round robin based on last_notify_idx_reserved.
+ */
+static u32 last_notify_idx_reserved;
+
+/* This is a one entry cache used to by the index allocation. */
+static u32 last_notify_idx_released = PAGE_SIZE;
+
+
+/*
+ * Utility function that retrieves the privilege flags associated
+ * with a given doorbell handle. For guest endpoints, the
+ * privileges are determined by the context ID, but for host
+ * endpoints privileges are associated with the complete
+ * handle. Hypervisor endpoints are not yet supported.
+ */
+int vmci_dbell_get_priv_flags(struct vmci_handle handle, u32 *priv_flags)
+{
+ if (priv_flags == NULL || handle.context == VMCI_INVALID_ID)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ if (handle.context == VMCI_HOST_CONTEXT_ID) {
+ struct dbell_entry *entry;
+ struct vmci_resource *resource;
+
+ resource = vmci_resource_by_handle(handle,
+ VMCI_RESOURCE_TYPE_DOORBELL);
+ if (!resource)
+ return VMCI_ERROR_NOT_FOUND;
+
+ entry = container_of(resource, struct dbell_entry, resource);
+ *priv_flags = entry->priv_flags;
+ vmci_resource_put(resource);
+ } else if (handle.context == VMCI_HYPERVISOR_CONTEXT_ID) {
+ /*
+ * Hypervisor endpoints for notifications are not
+ * supported (yet).
+ */
+ return VMCI_ERROR_INVALID_ARGS;
+ } else {
+ *priv_flags = vmci_context_get_priv_flags(handle.context);
+ }
+
+ return VMCI_SUCCESS;
+}
+
+/*
+ * Find doorbell entry by bitmap index.
+ */
+static struct dbell_entry *dbell_index_table_find(u32 idx)
+{
+ u32 bucket = VMCI_DOORBELL_HASH(idx);
+ struct dbell_entry *dbell;
+ struct hlist_node *node;
+
+ hlist_for_each_entry(dbell, node, &vmci_doorbell_it.entries[bucket],
+ node) {
+ if (idx == dbell->idx)
+ return dbell;
+ }
+
+ return NULL;
+}
+
+/*
+ * Add the given entry to the index table. This willi take a reference to the
+ * entry's resource so that the entry is not deleted before it is removed from
+ * the * table.
+ */
+static void dbell_index_table_add(struct dbell_entry *entry)
+{
+ u32 bucket;
+ u32 new_notify_idx;
+
+ vmci_resource_get(&entry->resource);
+
+ spin_lock_bh(&vmci_doorbell_it.lock);
+
+ /*
+ * Below we try to allocate an index in the notification
+ * bitmap with "not too much" sharing between resources. If we
+ * use less that the full bitmap, we either add to the end if
+ * there are no unused flags within the currently used area,
+ * or we search for unused ones. If we use the full bitmap, we
+ * allocate the index round robin.
+ */
+ if (max_notify_idx < PAGE_SIZE || notify_idx_count < PAGE_SIZE) {
+ if (last_notify_idx_released < max_notify_idx &&
+ !dbell_index_table_find(last_notify_idx_released)) {
+ new_notify_idx = last_notify_idx_released;
+ last_notify_idx_released = PAGE_SIZE;
+ } else {
+ bool reused = false;
+ new_notify_idx = last_notify_idx_reserved;
+ if (notify_idx_count + 1 < max_notify_idx) {
+ do {
+ if (!dbell_index_table_find
+ (new_notify_idx)) {
+ reused = true;
+ break;
+ }
+ new_notify_idx = (new_notify_idx + 1) %
+ max_notify_idx;
+ } while (new_notify_idx !=
+ last_notify_idx_released);
+ }
+ if (!reused) {
+ new_notify_idx = max_notify_idx;
+ max_notify_idx++;
+ }
+ }
+ } else {
+ new_notify_idx = (last_notify_idx_reserved + 1) % PAGE_SIZE;
+ }
+
+ last_notify_idx_reserved = new_notify_idx;
+ notify_idx_count++;
+
+ entry->idx = new_notify_idx;
+ bucket = VMCI_DOORBELL_HASH(entry->idx);
+ hlist_add_head(&entry->node, &vmci_doorbell_it.entries[bucket]);
+
+ spin_unlock_bh(&vmci_doorbell_it.lock);
+}
+
+/*
+ * Remove the given entry from the index table. This will release() the
+ * entry's resource.
+ */
+static void dbell_index_table_remove(struct dbell_entry *entry)
+{
+ spin_lock_bh(&vmci_doorbell_it.lock);
+
+ hlist_del_init(&entry->node);
+
+ notify_idx_count--;
+ if (entry->idx == max_notify_idx - 1) {
+ /*
+ * If we delete an entry with the maximum known
+ * notification index, we take the opportunity to
+ * prune the current max. As there might be other
+ * unused indices immediately below, we lower the
+ * maximum until we hit an index in use.
+ */
+ while (max_notify_idx > 0 &&
+ !dbell_index_table_find(max_notify_idx - 1))
+ max_notify_idx--;
+ }
+
+ last_notify_idx_released = entry->idx;
+
+ spin_unlock_bh(&vmci_doorbell_it.lock);
+
+ vmci_resource_put(&entry->resource);
+}
+
+/*
+ * Creates a link between the given doorbell handle and the given
+ * index in the bitmap in the device backend. A notification state
+ * is created in hypervisor.
+ */
+static int dbell_link(struct vmci_handle handle, u32 notify_idx)
+{
+ struct vmci_doorbell_link_msg link_msg;
+
+ link_msg.hdr.dst = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID,
+ VMCI_DOORBELL_LINK);
+ link_msg.hdr.src = VMCI_ANON_SRC_HANDLE;
+ link_msg.hdr.payload_size = sizeof(link_msg) - VMCI_DG_HEADERSIZE;
+ link_msg.handle = handle;
+ link_msg.notify_idx = notify_idx;
+
+ return vmci_send_datagram(&link_msg.hdr);
+}
+
+/*
+ * Unlinks the given doorbell handle from an index in the bitmap in
+ * the device backend. The notification state is destroyed in hypervisor.
+ */
+static int dbell_unlink(struct vmci_handle handle)
+{
+ struct vmci_doorbell_unlink_msg unlink_msg;
+
+ unlink_msg.hdr.dst = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID,
+ VMCI_DOORBELL_UNLINK);
+ unlink_msg.hdr.src = VMCI_ANON_SRC_HANDLE;
+ unlink_msg.hdr.payload_size = sizeof(unlink_msg) - VMCI_DG_HEADERSIZE;
+ unlink_msg.handle = handle;
+
+ return vmci_send_datagram(&unlink_msg.hdr);
+}
+
+/*
+ * Notify another guest or the host. We send a datagram down to the
+ * host via the hypervisor with the notification info.
+ */
+static int dbell_notify_as_guest(struct vmci_handle handle, u32 priv_flags)
+{
+ struct vmci_doorbell_notify_msg notify_msg;
+
+ notify_msg.hdr.dst = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID,
+ VMCI_DOORBELL_NOTIFY);
+ notify_msg.hdr.src = VMCI_ANON_SRC_HANDLE;
+ notify_msg.hdr.payload_size = sizeof(notify_msg) - VMCI_DG_HEADERSIZE;
+ notify_msg.handle = handle;
+
+ return vmci_send_datagram(¬ify_msg.hdr);
+}
+
+/*
+ * Calls the specified callback in a delayed context.
+ */
+static void dbell_delayed_dispatch(struct work_struct *work)
+{
+ struct dbell_entry *entry = container_of(work,
+ struct dbell_entry, work);
+
+ entry->notify_cb(entry->client_data);
+ vmci_resource_put(&entry->resource);
+}
+
+/*
+ * Dispatches a doorbell notification to the host context.
+ */
+int vmci_dbell_host_context_notify(u32 src_cid, struct vmci_handle handle)
+{
+ struct dbell_entry *entry;
+ struct vmci_resource *resource;
+
+ if (VMCI_HANDLE_INVALID(handle)) {
+ pr_devel("Notifying an invalid doorbell (handle=0x%x:0x%x).\n",
+ handle.context, handle.resource);
+ return VMCI_ERROR_INVALID_ARGS;
+ }
+
+ resource = vmci_resource_by_handle(handle,
+ VMCI_RESOURCE_TYPE_DOORBELL);
+ if (!resource) {
+ pr_devel("Notifying an unknown doorbell (handle=0x%x:0x%x).\n",
+ handle.context, handle.resource);
+ return VMCI_ERROR_NOT_FOUND;
+ }
+
+ entry = container_of(resource, struct dbell_entry, resource);
+ if (entry->run_delayed) {
+ schedule_work(&entry->work);
+ } else {
+ entry->notify_cb(entry->client_data);
+ vmci_resource_put(resource);
+ }
+
+ return VMCI_SUCCESS;
+}
+
+/*
+ * Register the notification bitmap with the host.
+ */
+bool vmci_dbell_register_notification_bitmap(u32 bitmap_ppn)
+{
+ int result;
+ struct vmci_notify_bm_set_msg bitmap_set_msg;
+
+ bitmap_set_msg.hdr.dst = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID,
+ VMCI_SET_NOTIFY_BITMAP);
+ bitmap_set_msg.hdr.src = VMCI_ANON_SRC_HANDLE;
+ bitmap_set_msg.hdr.payload_size = sizeof(bitmap_set_msg) -
+ VMCI_DG_HEADERSIZE;
+ bitmap_set_msg.bitmap_ppn = bitmap_ppn;
+
+ result = vmci_send_datagram(&bitmap_set_msg.hdr);
+ if (result != VMCI_SUCCESS) {
+ pr_devel("Failed to register (PPN=%u) as notification bitmap (error=%d).\n",
+ bitmap_ppn, result);
+ return false;
+ }
+ return true;
+}
+
+/*
+ * Executes or schedules the handlers for a given notify index.
+ */
+static void dbell_fire_entries(u32 notify_idx)
+{
+ u32 bucket = VMCI_DOORBELL_HASH(notify_idx);
+ struct dbell_entry *dbell;
+ struct hlist_node *node;
+
+ spin_lock_bh(&vmci_doorbell_it.lock);
+
+ hlist_for_each_entry(dbell, node,
+ &vmci_doorbell_it.entries[bucket], node) {
+ if (dbell->idx == notify_idx &&
+ atomic_read(&dbell->active) == 1) {
+ BUG_ON(!dbell->notify_cb);
+ if (dbell->run_delayed) {
+ vmci_resource_get(&dbell->resource);
+ schedule_work(&dbell->work);
+ } else {
+ dbell->notify_cb(dbell->client_data);
+ }
+ }
+ }
+
+ spin_unlock_bh(&vmci_doorbell_it.lock);
+}
+
+/*
+ * Scans the notification bitmap, collects pending notifications,
+ * resets the bitmap and invokes appropriate callbacks.
+ */
+void vmci_dbell_scan_notification_entries(u8 *bitmap)
+{
+ u32 idx;
+
+ for (idx = 0; idx < max_notify_idx; idx++) {
+ if (bitmap[idx] & 0x1) {
+ bitmap[idx] &= ~1;
+ dbell_fire_entries(idx);
+ }
+ }
+}
+
+/*
+ * vmci_doorbell_create() - Creates a doorbell
+ * @handle: A handle used to track the resource. Can be invalid.
+ * @flags: Flag that determines context of callback.
+ * @priv_flags: Privileges flags.
+ * @notify_cb: The callback to be ivoked when the doorbell fires.
+ * @client_data: A parameter to be passed to the callback.
+ *
+ * Creates a doorbell with the given callback. If the handle is
+ * VMCI_INVALID_HANDLE, a free handle will be assigned, if
+ * possible. The callback can be run immediately (potentially with
+ * locks held - the default) or delayed (in a kernel thread) by
+ * specifying the flag VMCI_FLAG_DELAYED_CB. If delayed execution
+ * is selected, a given callback may not be run if the kernel is
+ * unable to allocate memory for the delayed execution (highly
+ * unlikely).
+ */
+int vmci_doorbell_create(struct vmci_handle *handle,
+ u32 flags,
+ u32 priv_flags,
+ vmci_callback notify_cb, void *client_data)
+{
+ struct dbell_entry *entry;
+ struct vmci_handle new_handle;
+ int result;
+
+ if (!handle || !notify_cb || flags & ~VMCI_FLAG_DELAYED_CB ||
+ priv_flags & ~VMCI_PRIVILEGE_ALL_FLAGS)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ entry = kmalloc(sizeof(*entry), GFP_KERNEL);
+ if (entry == NULL) {
+ pr_warn("Failed allocating memory for datagram entry.\n");
+ return VMCI_ERROR_NO_MEM;
+ }
+
+ if (VMCI_HANDLE_INVALID(*handle)) {
+ u32 context_id = vmci_get_context_id();
+
+ /* Let resource code allocate a free ID for us */
+ new_handle = vmci_make_handle(context_id, VMCI_INVALID_ID);
+ } else {
+ bool valid_context = false;
+
+ /*
+ * Validate the handle. We must do both of the checks below
+ * because we can be acting as both a host and a guest at the
+ * same time. We always allow the host context ID, since the
+ * host functionality is in practice always there with the
+ * unified driver.
+ */
+ if (handle->context == VMCI_HOST_CONTEXT_ID ||
+ (vmci_guest_code_active() &&
+ vmci_get_context_id() == handle->context)) {
+ valid_context = true;
+ }
+
+ if (!valid_context || handle->resource == VMCI_INVALID_ID) {
+ pr_devel("Invalid argument (handle=0x%x:0x%x).\n",
+ handle->context, handle->resource);
+ result = VMCI_ERROR_INVALID_ARGS;
+ goto free_mem;
+ }
+
+ new_handle = *handle;
+ }
+
+ entry->idx = 0;
+ INIT_HLIST_NODE(&entry->node);
+ entry->priv_flags = priv_flags;
+ INIT_WORK(&entry->work, dbell_delayed_dispatch);
+ entry->run_delayed = flags & VMCI_FLAG_DELAYED_CB;
+ entry->notify_cb = notify_cb;
+ entry->client_data = client_data;
+ atomic_set(&entry->active, 0);
+
+ result = vmci_resource_add(&entry->resource,
+ VMCI_RESOURCE_TYPE_DOORBELL,
+ new_handle);
+ if (result != VMCI_SUCCESS) {
+ pr_warn("Failed to add new resource (handle=0x%x:0x%x), error: %d\n",
+ new_handle.context, new_handle.resource, result);
+ goto free_mem;
+ }
+
+ if (vmci_guest_code_active()) {
+ dbell_index_table_add(entry);
+ result = dbell_link(new_handle, entry->idx);
+ if (VMCI_SUCCESS != result)
+ goto destroy_resource;
+
+ atomic_set(&entry->active, 1);
+ }
+
+ *handle = vmci_resource_handle(&entry->resource);
+
+ return result;
+
+ destroy_resource:
+ dbell_index_table_remove(entry);
+ vmci_resource_remove(&entry->resource);
+ free_mem:
+ kfree(entry);
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_doorbell_create);
+
+/*
+ * vmci_doorbell_destroy() - Destroy a doorbell.
+ * @handle: The handle tracking the resource.
+ *
+ * Destroys a doorbell previously created with vmcii_doorbell_create. This
+ * operation may block waiting for a callback to finish.
+ */
+int vmci_doorbell_destroy(struct vmci_handle handle)
+{
+ struct dbell_entry *entry;
+ struct vmci_resource *resource;
+
+ if (VMCI_HANDLE_INVALID(handle))
+ return VMCI_ERROR_INVALID_ARGS;
+
+ resource = vmci_resource_by_handle(handle,
+ VMCI_RESOURCE_TYPE_DOORBELL);
+ if (!resource) {
+ pr_devel("Failed to destroy doorbell (handle=0x%x:0x%x).\n",
+ handle.context, handle.resource);
+ return VMCI_ERROR_NOT_FOUND;
+ }
+
+ entry = container_of(resource, struct dbell_entry, resource);
+
+ if (vmci_guest_code_active()) {
+ int result;
+
+ dbell_index_table_remove(entry);
+
+ result = dbell_unlink(handle);
+ if (VMCI_SUCCESS != result) {
+
+ /*
+ * The only reason this should fail would be
+ * an inconsistency between guest and
+ * hypervisor state, where the guest believes
+ * it has an active registration whereas the
+ * hypervisor doesn't. One case where this may
+ * happen is if a doorbell is unregistered
+ * following a hibernation at a time where the
+ * doorbell state hasn't been restored on the
+ * hypervisor side yet. Since the handle has
+ * now been removed in the guest, we just
+ * print a warning and return success.
+ */
+ pr_devel("Unlink of doorbell (handle=0x%x:0x%x) unknown by hypervisor (error=%d).\n",
+ handle.context, handle.resource, result);
+ }
+ }
+
+ /*
+ * Now remove the resource from the table. It might still be in use
+ * after this, in a callback or still on the delayed work queue.
+ */
+ vmci_resource_put(&entry->resource);
+ vmci_resource_remove(&entry->resource);
+
+ kfree(entry);
+
+ return VMCI_SUCCESS;
+}
+EXPORT_SYMBOL_GPL(vmci_doorbell_destroy);
+
+/*
+ * vmci_doorbell_notify() - Ring the doorbell (and hide in the bushes).
+ * @dst: The handlle identifying the doorbell resource
+ * @priv_flags: Priviledge flags.
+ *
+ * Generates a notification on the doorbell identified by the
+ * handle. For host side generation of notifications, the caller
+ * can specify what the privilege of the calling side is.
+ */
+int vmci_doorbell_notify(struct vmci_handle dst, u32 priv_flags)
+{
+ int retval;
+ enum vmci_route route;
+ struct vmci_handle src;
+
+ if (VMCI_HANDLE_INVALID(dst) ||
+ (priv_flags & ~VMCI_PRIVILEGE_ALL_FLAGS))
+ return VMCI_ERROR_INVALID_ARGS;
+
+ src = VMCI_INVALID_HANDLE;
+ retval = vmci_route(&src, &dst, false, &route);
+ if (retval < VMCI_SUCCESS)
+ return retval;
+
+ if (VMCI_ROUTE_AS_HOST == route)
+ return vmci_ctx_notify_dbell(VMCI_HOST_CONTEXT_ID,
+ dst, priv_flags);
+
+ if (VMCI_ROUTE_AS_GUEST == route)
+ return dbell_notify_as_guest(dst, priv_flags);
+
+ pr_warn("Unknown route (%d) for doorbell.\n", route);
+ return VMCI_ERROR_DST_UNREACHABLE;
+}
+EXPORT_SYMBOL_GPL(vmci_doorbell_notify);
diff --git a/drivers/misc/vmw_vmci/vmci_doorbell.h b/drivers/misc/vmw_vmci/vmci_doorbell.h
new file mode 100644
index 0000000..e4c0b17
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_doorbell.h
@@ -0,0 +1,51 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#ifndef VMCI_DOORBELL_H
+#define VMCI_DOORBELL_H
+
+#include <linux/vmw_vmci_defs.h>
+#include <linux/types.h>
+
+#include "vmci_driver.h"
+
+/*
+ * VMCINotifyResourceInfo: Used to create and destroy doorbells, and
+ * generate a notification for a doorbell or queue pair.
+ */
+struct vmci_dbell_notify_resource_info {
+ struct vmci_handle handle;
+ u16 resource;
+ u16 action;
+ s32 result;
+};
+
+/*
+ * Structure used for checkpointing the doorbell mappings. It is
+ * written to the checkpoint as is, so changing this structure will
+ * break checkpoint compatibility.
+ */
+struct dbell_cpt_state {
+ struct vmci_handle handle;
+ u64 bitmap_idx;
+};
+
+int vmci_dbell_host_context_notify(u32 src_cid, struct vmci_handle handle);
+int vmci_dbell_get_priv_flags(struct vmci_handle handle, u32 *priv_flags);
+
+bool vmci_dbell_register_notification_bitmap(u32 bitmap_ppn);
+void vmci_dbell_scan_notification_entries(u8 *bitmap);
+
+#endif /* VMCI_DOORBELL_H */
^ permalink raw reply related
* [PATCH 04/12] VMCI: device driver implementaton.
From: George Zhang @ 2012-11-07 18:41 UTC (permalink / raw)
To: linux-kernel, georgezhang, virtualization; +Cc: pv-drivers, gregkh
In-Reply-To: <20121107183624.9658.78903.stgit@promb-2n-dhcp175.eng.vmware.com>
VMCI driver code implementes both the host and guest personalities of the VMCI driver.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
drivers/misc/vmw_vmci/vmci_driver.c | 116 +++++++++++++++++++++++++++++++++++
drivers/misc/vmw_vmci/vmci_driver.h | 50 +++++++++++++++
2 files changed, 166 insertions(+), 0 deletions(-)
create mode 100644 drivers/misc/vmw_vmci/vmci_driver.c
create mode 100644 drivers/misc/vmw_vmci/vmci_driver.h
diff --git a/drivers/misc/vmw_vmci/vmci_driver.c b/drivers/misc/vmw_vmci/vmci_driver.c
new file mode 100644
index 0000000..8c5db38
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_driver.c
@@ -0,0 +1,116 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#include <linux/vmw_vmci_defs.h>
+#include <linux/vmw_vmci_api.h>
+#include <linux/atomic.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+
+#include "vmci_driver.h"
+#include "vmci_event.h"
+
+static bool vmci_disable_host;
+module_param_named(disable_host, vmci_disable_host, bool, 0);
+MODULE_PARM_DESC(disable_host,
+ "Disable driver host personality (default=enabled)");
+
+static bool vmci_disable_guest;
+module_param_named(disable_guest, vmci_disable_guest, bool, 0);
+MODULE_PARM_DESC(disable_guest,
+ "Disable driver guest personality (default=enabled)");
+
+static bool vmci_guest_personality_initialized;
+static bool vmci_host_personality_initialized;
+
+/*
+ * vmci_get_context_id() - Gets the current context ID.
+ *
+ * Returns the current context ID. Note that since this is accessed only
+ * from code running in the host, this always returns the host context ID.
+ */
+u32 vmci_get_context_id(void)
+{
+ if (vmci_guest_code_active())
+ return vmci_get_vm_context_id();
+ else if (vmci_host_code_active())
+ return VMCI_HOST_CONTEXT_ID;
+
+ return VMCI_INVALID_ID;
+}
+EXPORT_SYMBOL_GPL(vmci_get_context_id);
+
+static int __init vmci_drv_init(void)
+{
+ int vmci_err;
+ int error;
+
+ vmci_err = vmci_event_init();
+ if (vmci_err < VMCI_SUCCESS) {
+ pr_err("Failed to initialize VMCIEvent (result=%d).\n", vmci_err);
+ return -EINVAL;
+ }
+
+ if (!vmci_disable_guest) {
+ error = vmci_guest_init();
+ if (error) {
+ pr_warn("Failed to initialize guest personality (err=%d).\n",
+ error);
+ } else {
+ vmci_guest_personality_initialized = true;
+ pr_info("Guest personality initialized and is %s\n",
+ vmci_guest_code_active() ?
+ "active" : "inactive");
+ }
+ }
+
+ if (!vmci_disable_host) {
+ error = vmci_host_init();
+ if (error) {
+ pr_warn("Unable to initialize host personality (err=%d).\n",
+ error);
+ } else {
+ vmci_host_personality_initialized = true;
+ pr_info("Initialized host personality\n");
+ }
+ }
+
+ if (!vmci_guest_personality_initialized &&
+ !vmci_host_personality_initialized) {
+ vmci_event_exit();
+ return -ENODEV;
+ }
+
+ return 0;
+}
+module_init(vmci_drv_init);
+
+static void __exit vmci_drv_exit(void)
+{
+ if (vmci_guest_personality_initialized)
+ vmci_guest_exit();
+
+ if (vmci_host_personality_initialized)
+ vmci_host_exit();
+
+ vmci_event_exit();
+}
+module_exit(vmci_drv_exit);
+
+MODULE_AUTHOR("VMware, Inc.");
+MODULE_DESCRIPTION("VMware Virtual Machine Communication Interface.");
+MODULE_VERSION(VMCI_DRIVER_VERSION_STRING);
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/misc/vmw_vmci/vmci_driver.h b/drivers/misc/vmw_vmci/vmci_driver.h
new file mode 100644
index 0000000..f69156a
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_driver.h
@@ -0,0 +1,50 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#ifndef _VMCI_DRIVER_H_
+#define _VMCI_DRIVER_H_
+
+#include <linux/vmw_vmci_defs.h>
+#include <linux/wait.h>
+
+#include "vmci_queue_pair.h"
+#include "vmci_context.h"
+
+enum vmci_obj_type {
+ VMCIOBJ_VMX_VM = 10,
+ VMCIOBJ_CONTEXT,
+ VMCIOBJ_SOCKET,
+ VMCIOBJ_NOT_SET,
+};
+
+/* For storing VMCI structures in file handles. */
+struct vmci_obj {
+ void *ptr;
+ enum vmci_obj_type type;
+};
+
+u32 vmci_get_context_id(void);
+int vmci_send_datagram(struct vmci_datagram *dg);
+
+int vmci_host_init(void);
+void vmci_host_exit(void);
+bool vmci_host_code_active(void);
+
+int vmci_guest_init(void);
+void vmci_guest_exit(void);
+bool vmci_guest_code_active(void);
+u32 vmci_get_vm_context_id(void);
+
+#endif /* _VMCI_DRIVER_H_ */
^ permalink raw reply related
* [PATCH 05/12] VMCI: event handling implementation.
From: George Zhang @ 2012-11-07 18:41 UTC (permalink / raw)
To: linux-kernel, georgezhang, virtualization; +Cc: pv-drivers, gregkh
In-Reply-To: <20121107183624.9658.78903.stgit@promb-2n-dhcp175.eng.vmware.com>
VMCI event code that manages event handlers and handles callbacks when specific
events fire.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
drivers/misc/vmw_vmci/vmci_event.c | 229 ++++++++++++++++++++++++++++++++++++
drivers/misc/vmw_vmci/vmci_event.h | 25 ++++
2 files changed, 254 insertions(+), 0 deletions(-)
create mode 100644 drivers/misc/vmw_vmci/vmci_event.c
create mode 100644 drivers/misc/vmw_vmci/vmci_event.h
diff --git a/drivers/misc/vmw_vmci/vmci_event.c b/drivers/misc/vmw_vmci/vmci_event.c
new file mode 100644
index 0000000..e956c18
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_event.c
@@ -0,0 +1,229 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#include <linux/vmw_vmci_defs.h>
+#include <linux/vmw_vmci_api.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+
+#include "vmci_driver.h"
+#include "vmci_event.h"
+
+#define EVENT_MAGIC 0xEABE0000
+#define VMCI_EVENT_MAX_ATTEMPTS 10
+
+struct vmci_subscription {
+ u32 id;
+ u32 event;
+ vmci_event_cb callback;
+ void *callback_data;
+ struct list_head node; /* on one of subscriber lists */
+};
+
+static struct list_head subscriber_array[VMCI_EVENT_MAX];
+static DEFINE_MUTEX(subscriber_mutex);
+
+int __init vmci_event_init(void)
+{
+ int i;
+
+ for (i = 0; i < VMCI_EVENT_MAX; i++)
+ INIT_LIST_HEAD(&subscriber_array[i]);
+
+ return VMCI_SUCCESS;
+}
+
+void vmci_event_exit(void)
+{
+ int e;
+
+ /* We free all memory at exit. */
+ for (e = 0; e < VMCI_EVENT_MAX; e++) {
+ struct vmci_subscription *cur, *p2;
+ list_for_each_entry_safe(cur, p2, &subscriber_array[e], node) {
+
+ /*
+ * We should never get here because all events
+ * should have been unregistered before we try
+ * to unload the driver module.
+ */
+ pr_warn("Unexpected free events occurring.\n");
+ list_del(&cur->node);
+ kfree(cur);
+ }
+ }
+}
+
+/*
+ * Find entry. Assumes subscriber_mutex is held.
+ */
+static struct vmci_subscription *event_find(u32 sub_id)
+{
+ int e;
+
+ for (e = 0; e < VMCI_EVENT_MAX; e++) {
+ struct vmci_subscription *cur;
+ list_for_each_entry(cur, &subscriber_array[e], node) {
+ if (cur->id == sub_id)
+ return cur;
+ }
+ }
+ return NULL;
+}
+
+/*
+ * Actually delivers the events to the subscribers.
+ * The callback function for each subscriber is invoked.
+ */
+static void event_deliver(struct vmci_event_msg *event_msg)
+{
+ struct vmci_subscription *cur;
+ struct list_head *subscriber_list;
+
+ rcu_read_lock();
+ subscriber_list = &subscriber_array[event_msg->event_data.event];
+ list_for_each_entry_rcu(cur, subscriber_list, node) {
+ BUG_ON(cur->event != event_msg->event_data.event);
+
+ cur->callback(cur->id, &event_msg->event_data,
+ cur->callback_data);
+ }
+ rcu_read_unlock();
+}
+
+/*
+ * Dispatcher for the VMCI_EVENT_RECEIVE datagrams. Calls all
+ * subscribers for given event.
+ */
+int vmci_event_dispatch(struct vmci_datagram *msg)
+{
+ struct vmci_event_msg *event_msg = (struct vmci_event_msg *)msg;
+
+ BUG_ON(msg->src.context != VMCI_HYPERVISOR_CONTEXT_ID ||
+ msg->dst.resource != VMCI_EVENT_HANDLER);
+
+ if (msg->payload_size < sizeof(u32) ||
+ msg->payload_size > sizeof(struct vmci_event_data_max))
+ return VMCI_ERROR_INVALID_ARGS;
+
+ if (!VMCI_EVENT_VALID(event_msg->event_data.event))
+ return VMCI_ERROR_EVENT_UNKNOWN;
+
+ event_deliver(event_msg);
+ return VMCI_SUCCESS;
+}
+
+/*
+ * vmci_event_subscribe() - Subscribe to a given event.
+ * @event: The event to subscribe to.
+ * @callback: The callback to invoke upon the event.
+ * @callback_data: Data to pass to the callback.
+ * @subscription_id: ID used to track subscription. Used with
+ * vmci_event_unsubscribe()
+ *
+ * Subscribes to the provided event. The callback specified will be
+ * fired from RCU critical section and therefore must not sleep.
+ */
+int vmci_event_subscribe(u32 event,
+ vmci_event_cb callback,
+ void *callback_data,
+ u32 *new_subscription_id)
+{
+ struct vmci_subscription *sub;
+ int attempts;
+ int retval;
+ bool have_new_id = false;
+
+ if (!new_subscription_id) {
+ pr_devel("%s: Invalid subscription (NULL).\n", __func__);
+ return VMCI_ERROR_INVALID_ARGS;
+ }
+
+ if (!VMCI_EVENT_VALID(event) || !callback) {
+ pr_devel("%s: Failed to subscribe to event (type=%d) (callback=%p) (data=%p).\n",
+ __func__, event, callback, callback_data);
+ return VMCI_ERROR_INVALID_ARGS;
+ }
+
+ sub = kzalloc(sizeof(*sub), GFP_KERNEL);
+ if (!sub)
+ return VMCI_ERROR_NO_MEM;
+
+ sub->id = VMCI_EVENT_MAX;
+ sub->event = event;
+ sub->callback = callback;
+ sub->callback_data = callback_data;
+ INIT_LIST_HEAD(&sub->node);
+
+ mutex_lock(&subscriber_mutex);
+
+ /* Creation of a new event is always allowed. */
+ for (attempts = 0; attempts < VMCI_EVENT_MAX_ATTEMPTS; attempts++) {
+ static u32 subscription_id;
+ /*
+ * We try to get an id a couple of time before
+ * claiming we are out of resources.
+ */
+
+ /* Test for duplicate id. */
+ if (!event_find(++subscription_id)) {
+ sub->id = subscription_id;
+ have_new_id = true;
+ break;
+ }
+ }
+
+ if (have_new_id) {
+ list_add_rcu(&sub->node, &subscriber_array[event]);
+ retval = VMCI_SUCCESS;
+ } else {
+ retval = VMCI_ERROR_NO_RESOURCES;
+ }
+
+ mutex_unlock(&subscriber_mutex);
+
+ *new_subscription_id = sub->id;
+ return retval;
+}
+EXPORT_SYMBOL_GPL(vmci_event_subscribe);
+
+/*
+ * vmci_event_unsubscribe() - unsubscribe from an event.
+ * @sub_id: A subscription ID as provided by vmci_event_subscribe()
+ *
+ * Unsubscribe from given event. Removes it from list and frees it.
+ * Will return callback_data if requested by caller.
+ */
+int vmci_event_unsubscribe(u32 sub_id)
+{
+ struct vmci_subscription *s;
+
+ mutex_lock(&subscriber_mutex);
+ s = event_find(sub_id);
+ if (s)
+ list_del_rcu(&s->node);
+ mutex_unlock(&subscriber_mutex);
+
+ if (!s)
+ return VMCI_ERROR_NOT_FOUND;
+
+ synchronize_rcu();
+ kfree(s);
+
+ return VMCI_SUCCESS;
+}
+EXPORT_SYMBOL_GPL(vmci_event_unsubscribe);
diff --git a/drivers/misc/vmw_vmci/vmci_event.h b/drivers/misc/vmw_vmci/vmci_event.h
new file mode 100644
index 0000000..7df9b1c
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_event.h
@@ -0,0 +1,25 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#ifndef __VMCI_EVENT_H__
+#define __VMCI_EVENT_H__
+
+#include <linux/vmw_vmci_api.h>
+
+int vmci_event_init(void);
+void vmci_event_exit(void);
+int vmci_event_dispatch(struct vmci_datagram *msg);
+
+#endif /*__VMCI_EVENT_H__ */
^ permalink raw reply related
* [PATCH 06/12] VMCI: handle array implementation.
From: George Zhang @ 2012-11-07 18:42 UTC (permalink / raw)
To: linux-kernel, georgezhang, virtualization; +Cc: pv-drivers, gregkh
In-Reply-To: <20121107183624.9658.78903.stgit@promb-2n-dhcp175.eng.vmware.com>
VMCI handle code adds support for dynamic arrays that will grow if they need to.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
drivers/misc/vmw_vmci/vmci_handle_array.c | 142 +++++++++++++++++++++++++++++
drivers/misc/vmw_vmci/vmci_handle_array.h | 52 +++++++++++
2 files changed, 194 insertions(+), 0 deletions(-)
create mode 100644 drivers/misc/vmw_vmci/vmci_handle_array.c
create mode 100644 drivers/misc/vmw_vmci/vmci_handle_array.h
diff --git a/drivers/misc/vmw_vmci/vmci_handle_array.c b/drivers/misc/vmw_vmci/vmci_handle_array.c
new file mode 100644
index 0000000..9122373
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_handle_array.c
@@ -0,0 +1,142 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#include <linux/slab.h>
+#include "vmci_handle_array.h"
+
+static size_t handle_arr_calc_size(size_t capacity)
+{
+ return sizeof(struct vmci_handle_arr) +
+ capacity * sizeof(struct vmci_handle);
+}
+
+struct vmci_handle_arr *vmci_handle_arr_create(size_t capacity)
+{
+ struct vmci_handle_arr *array;
+
+ if (capacity == 0)
+ capacity = VMCI_HANDLE_ARRAY_DEFAULT_SIZE;
+
+ array = kmalloc(handle_arr_calc_size(capacity), GFP_ATOMIC);
+ if (!array)
+ return NULL;
+
+ array->capacity = capacity;
+ array->size = 0;
+
+ return array;
+}
+
+void vmci_handle_arr_destroy(struct vmci_handle_arr *array)
+{
+ kfree(array);
+}
+
+void vmci_handle_arr_append_entry(struct vmci_handle_arr **array_ptr,
+ struct vmci_handle handle)
+{
+ struct vmci_handle_arr *array = *array_ptr;
+
+ if (unlikely(array->size >= array->capacity)) {
+ /* reallocate. */
+ struct vmci_handle_arr *new_array;
+ size_t new_capacity = array->capacity * VMCI_ARR_CAP_MULT;
+ size_t new_size = handle_arr_calc_size(new_capacity);
+
+ new_array = krealloc(array, new_size, GFP_ATOMIC);
+ if (!new_array)
+ return;
+
+ new_array->capacity = new_capacity;
+ *array_ptr = array = new_array;
+ }
+
+ array->entries[array->size] = handle;
+ array->size++;
+}
+
+/*
+ * Handle that was removed, VMCI_INVALID_HANDLE if entry not found.
+ */
+struct vmci_handle vmci_handle_arr_remove_entry(struct vmci_handle_arr *array,
+ struct vmci_handle entry_handle)
+{
+ struct vmci_handle handle = VMCI_INVALID_HANDLE;
+ size_t i;
+
+ for (i = 0; i < array->size; i++) {
+ if (VMCI_HANDLE_EQUAL(array->entries[i], entry_handle)) {
+ handle = array->entries[i];
+ array->size--;
+ array->entries[i] = array->entries[array->size];
+ array->entries[array->size] = VMCI_INVALID_HANDLE;
+ break;
+ }
+ }
+
+ return handle;
+}
+
+/*
+ * Handle that was removed, VMCI_INVALID_HANDLE if array was empty.
+ */
+struct vmci_handle vmci_handle_arr_remove_tail(struct vmci_handle_arr *array)
+{
+ struct vmci_handle handle = VMCI_INVALID_HANDLE;
+
+ if (array->size) {
+ array->size--;
+ handle = array->entries[array->size];
+ array->entries[array->size] = VMCI_INVALID_HANDLE;
+ }
+
+ return handle;
+}
+
+/*
+ * Handle at given index, VMCI_INVALID_HANDLE if invalid index.
+ */
+struct vmci_handle
+vmci_handle_arr_get_entry(const struct vmci_handle_arr *array, size_t index)
+{
+ if (unlikely(index >= array->size))
+ return VMCI_INVALID_HANDLE;
+
+ return array->entries[index];
+}
+
+bool vmci_handle_arr_has_entry(const struct vmci_handle_arr *array,
+ struct vmci_handle entry_handle)
+{
+ size_t i;
+
+ for (i = 0; i < array->size; i++)
+ if (VMCI_HANDLE_EQUAL(array->entries[i], entry_handle))
+ return true;
+
+ return false;
+}
+
+/*
+ * NULL if the array is empty. Otherwise, a pointer to the array
+ * of VMCI handles in the handle array.
+ */
+struct vmci_handle *vmci_handle_arr_get_handles(struct vmci_handle_arr *array)
+{
+ if (array->size)
+ return array->entries;
+
+ return NULL;
+}
diff --git a/drivers/misc/vmw_vmci/vmci_handle_array.h b/drivers/misc/vmw_vmci/vmci_handle_array.h
new file mode 100644
index 0000000..b5f3a7f
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_handle_array.h
@@ -0,0 +1,52 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#ifndef _VMCI_HANDLE_ARRAY_H_
+#define _VMCI_HANDLE_ARRAY_H_
+
+#include <linux/vmw_vmci_defs.h>
+#include <linux/types.h>
+
+#define VMCI_HANDLE_ARRAY_DEFAULT_SIZE 4
+#define VMCI_ARR_CAP_MULT 2 /* Array capacity multiplier */
+
+struct vmci_handle_arr {
+ size_t capacity;
+ size_t size;
+ struct vmci_handle entries[];
+};
+
+struct vmci_handle_arr *vmci_handle_arr_create(size_t capacity);
+void vmci_handle_arr_destroy(struct vmci_handle_arr *array);
+void vmci_handle_arr_append_entry(struct vmci_handle_arr **array_ptr,
+ struct vmci_handle handle);
+struct vmci_handle vmci_handle_arr_remove_entry(struct vmci_handle_arr *array,
+ struct vmci_handle
+ entry_handle);
+struct vmci_handle vmci_handle_arr_remove_tail(struct vmci_handle_arr *array);
+struct vmci_handle
+vmci_handle_arr_get_entry(const struct vmci_handle_arr *array, size_t index);
+bool vmci_handle_arr_has_entry(const struct vmci_handle_arr *array,
+ struct vmci_handle entry_handle);
+struct vmci_handle *vmci_handle_arr_get_handles(struct vmci_handle_arr *array);
+
+static inline size_t vmci_handle_arr_get_size(
+ const struct vmci_handle_arr *array)
+{
+ return array->size;
+}
+
+
+#endif /* _VMCI_HANDLE_ARRAY_H_ */
^ permalink raw reply related
* [PATCH 07/12] VMCI: queue pairs implementation.
From: George Zhang @ 2012-11-07 18:42 UTC (permalink / raw)
To: linux-kernel, georgezhang, virtualization; +Cc: pv-drivers, gregkh
In-Reply-To: <20121107183624.9658.78903.stgit@promb-2n-dhcp175.eng.vmware.com>
VMCI queue pairs allow for bi-directional ordered communication between host and guests.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
drivers/misc/vmw_vmci/vmci_queue_pair.c | 3506 +++++++++++++++++++++++++++++++
drivers/misc/vmw_vmci/vmci_queue_pair.h | 191 ++
2 files changed, 3697 insertions(+), 0 deletions(-)
create mode 100644 drivers/misc/vmw_vmci/vmci_queue_pair.c
create mode 100644 drivers/misc/vmw_vmci/vmci_queue_pair.h
diff --git a/drivers/misc/vmw_vmci/vmci_queue_pair.c b/drivers/misc/vmw_vmci/vmci_queue_pair.c
new file mode 100644
index 0000000..126f85e
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_queue_pair.c
@@ -0,0 +1,3506 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#include <linux/device-mapper.h>
+#include <linux/vmw_vmci_defs.h>
+#include <linux/vmw_vmci_api.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/socket.h>
+#include <linux/wait.h>
+
+#include "vmci_handle_array.h"
+#include "vmci_common_int.h"
+#include "vmci_queue_pair.h"
+#include "vmci_datagram.h"
+#include "vmci_resource.h"
+#include "vmci_context.h"
+#include "vmci_driver.h"
+#include "vmci_event.h"
+#include "vmci_route.h"
+
+/*
+ * In the following, we will distinguish between two kinds of VMX processes -
+ * the ones with versions lower than VMCI_VERSION_NOVMVM that use specialized
+ * VMCI page files in the VMX and supporting VM to VM communication and the
+ * newer ones that use the guest memory directly. We will in the following
+ * refer to the older VMX versions as old-style VMX'en, and the newer ones as
+ * new-style VMX'en.
+ *
+ * The state transition datagram is as follows (the VMCIQPB_ prefix has been
+ * removed for readability) - see below for more details on the transtions:
+ *
+ * -------------- NEW -------------
+ * | |
+ * \_/ \_/
+ * CREATED_NO_MEM <-----------------> CREATED_MEM
+ * | | |
+ * | o-----------------------o |
+ * | | |
+ * \_/ \_/ \_/
+ * ATTACHED_NO_MEM <----------------> ATTACHED_MEM
+ * | | |
+ * | o----------------------o |
+ * | | |
+ * \_/ \_/ \_/
+ * SHUTDOWN_NO_MEM <----------------> SHUTDOWN_MEM
+ * | |
+ * | |
+ * -------------> gone <-------------
+ *
+ * In more detail. When a VMCI queue pair is first created, it will be in the
+ * VMCIQPB_NEW state. It will then move into one of the following states:
+ *
+ * - VMCIQPB_CREATED_NO_MEM: this state indicates that either:
+ *
+ * - the created was performed by a host endpoint, in which case there is
+ * no backing memory yet.
+ *
+ * - the create was initiated by an old-style VMX, that uses
+ * vmci_qp_broker_set_page_store to specify the UVAs of the queue pair at
+ * a later point in time. This state can be distinguished from the one
+ * above by the context ID of the creator. A host side is not allowed to
+ * attach until the page store has been set.
+ *
+ * - VMCIQPB_CREATED_MEM: this state is the result when the queue pair
+ * is created by a VMX using the queue pair device backend that
+ * sets the UVAs of the queue pair immediately and stores the
+ * information for later attachers. At this point, it is ready for
+ * the host side to attach to it.
+ *
+ * Once the queue pair is in one of the created states (with the exception of
+ * the case mentioned for older VMX'en above), it is possible to attach to the
+ * queue pair. Again we have two new states possible:
+ *
+ * - VMCIQPB_ATTACHED_MEM: this state can be reached through the following
+ * paths:
+ *
+ * - from VMCIQPB_CREATED_NO_MEM when a new-style VMX allocates a queue
+ * pair, and attaches to a queue pair previously created by the host side.
+ *
+ * - from VMCIQPB_CREATED_MEM when the host side attaches to a queue pair
+ * already created by a guest.
+ *
+ * - from VMCIQPB_ATTACHED_NO_MEM, when an old-style VMX calls
+ * vmci_qp_broker_set_page_store (see below).
+ *
+ * - VMCIQPB_ATTACHED_NO_MEM: If the queue pair already was in the
+ * VMCIQPB_CREATED_NO_MEM due to a host side create, an old-style VMX will
+ * bring the queue pair into this state. Once vmci_qp_broker_set_page_store
+ * is called to register the user memory, the VMCIQPB_ATTACH_MEM state
+ * will be entered.
+ *
+ * From the attached queue pair, the queue pair can enter the shutdown states
+ * when either side of the queue pair detaches. If the guest side detaches
+ * first, the queue pair will enter the VMCIQPB_SHUTDOWN_NO_MEM state, where
+ * the content of the queue pair will no longer be available. If the host
+ * side detaches first, the queue pair will either enter the
+ * VMCIQPB_SHUTDOWN_MEM, if the guest memory is currently mapped, or
+ * VMCIQPB_SHUTDOWN_NO_MEM, if the guest memory is not mapped
+ * (e.g., the host detaches while a guest is stunned).
+ *
+ * New-style VMX'en will also unmap guest memory, if the guest is
+ * quiesced, e.g., during a snapshot operation. In that case, the guest
+ * memory will no longer be available, and the queue pair will transition from
+ * *_MEM state to a *_NO_MEM state. The VMX may later map the memory once more,
+ * in which case the queue pair will transition from the *_NO_MEM state at that
+ * point back to the *_MEM state. Note that the *_NO_MEM state may have changed,
+ * since the peer may have either attached or detached in the meantime. The
+ * values are laid out such that ++ on a state will move from a *_NO_MEM to a
+ * *_MEM state, and vice versa.
+ */
+
+/*
+ * VMCIMemcpy{To,From}QueueFunc() prototypes. Functions of these
+ * types are passed around to enqueue and dequeue routines. Note that
+ * often the functions passed are simply wrappers around memcpy
+ * itself.
+ *
+ * Note: In order for the memcpy typedefs to be compatible with the VMKernel,
+ * there's an unused last parameter for the hosted side. In
+ * ESX, that parameter holds a buffer type.
+ */
+typedef int vmci_memcpy_to_queue_func(struct vmci_queue *queue,
+ u64 queue_offset, const void *src,
+ size_t src_offset, size_t size);
+typedef int vmci_memcpy_from_queue_func(void *dest, size_t dest_offset,
+ const struct vmci_queue *queue,
+ u64 queue_offset, size_t size);
+
+/* The Kernel specific component of the struct vmci_queue structure. */
+struct vmci_queue_kern_if {
+ struct page **page;
+ struct page **header_page;
+ void *va;
+ struct mutex __mutex;
+ struct mutex *mutex;
+ bool host;
+ size_t num_pages;
+ bool mapped;
+};
+
+/*
+ * This structure is opaque to the clients.
+ */
+struct vmci_qp {
+ struct vmci_handle handle;
+ struct vmci_queue *produce_q;
+ struct vmci_queue *consume_q;
+ u64 produce_q_size;
+ u64 consume_q_size;
+ u32 peer;
+ u32 flags;
+ u32 priv_flags;
+ bool guest_endpoint;
+ unsigned int blocked;
+ unsigned int generation;
+ wait_queue_head_t event;
+};
+
+enum qp_broker_state {
+ VMCIQPB_NEW,
+ VMCIQPB_CREATED_NO_MEM,
+ VMCIQPB_CREATED_MEM,
+ VMCIQPB_ATTACHED_NO_MEM,
+ VMCIQPB_ATTACHED_MEM,
+ VMCIQPB_SHUTDOWN_NO_MEM,
+ VMCIQPB_SHUTDOWN_MEM,
+ VMCIQPB_GONE
+};
+
+#define QPBROKERSTATE_HAS_MEM(_qpb) (_qpb->state == VMCIQPB_CREATED_MEM || \
+ _qpb->state == VMCIQPB_ATTACHED_MEM || \
+ _qpb->state == VMCIQPB_SHUTDOWN_MEM)
+
+/*
+ * In the queue pair broker, we always use the guest point of view for
+ * the produce and consume queue values and references, e.g., the
+ * produce queue size stored is the guests produce queue size. The
+ * host endpoint will need to swap these around. The only exception is
+ * the local queue pairs on the host, in which case the host endpoint
+ * that creates the queue pair will have the right orientation, and
+ * the attaching host endpoint will need to swap.
+ */
+struct qp_entry {
+ struct list_head list_item;
+ struct vmci_handle handle;
+ u32 peer;
+ u32 flags;
+ u64 produce_size;
+ u64 consume_size;
+ u32 ref_count;
+};
+
+struct qp_broker_entry {
+ struct vmci_resource resource;
+ struct qp_entry qp;
+ u32 create_id;
+ u32 attach_id;
+ enum qp_broker_state state;
+ bool require_trusted_attach;
+ bool created_by_trusted;
+ bool vmci_page_files; /* Created by VMX using VMCI page files */
+ struct vmci_queue *produce_q;
+ struct vmci_queue *consume_q;
+ struct vmci_queue_header saved_produce_q;
+ struct vmci_queue_header saved_consume_q;
+ vmci_event_release_cb wakeup_cb;
+ void *client_data;
+ void *local_mem; /* Kernel memory for local queue pair */
+};
+
+struct qp_guest_endpoint {
+ struct vmci_resource resource;
+ struct qp_entry qp;
+ u64 num_ppns;
+ void *produce_q;
+ void *consume_q;
+ struct PPNSet ppn_set;
+};
+
+struct qp_list {
+ struct list_head head;
+ struct mutex mutex;
+};
+
+static struct qp_list qp_broker_list = {
+ .head = LIST_HEAD_INIT(qp_broker_list.head),
+ .mutex = __MUTEX_INITIALIZER(qp_broker_list.mutex),
+};
+
+static struct qp_list qp_guest_endpoints = {
+ .head = LIST_HEAD_INIT(qp_guest_endpoints.head),
+ .mutex = __MUTEX_INITIALIZER(qp_guest_endpoints.mutex),
+};
+
+#define INVALID_VMCI_GUEST_MEM_ID 0
+#define QPE_NUM_PAGES(_QPE) ((u32) \
+ (dm_div_up(_QPE.produce_size, PAGE_SIZE) + \
+ dm_div_up(_QPE.consume_size, PAGE_SIZE) + 2))
+
+
+/*
+ * Frees kernel VA space for a given queue and its queue header, and
+ * frees physical data pages.
+ */
+static void qp_free_queue(void *q, u64 size)
+{
+ struct vmci_queue *queue = q;
+
+ if (queue) {
+ u64 i = dm_div_up(size, PAGE_SIZE);
+
+ if (queue->kernel_if->mapped) {
+ vunmap(queue->kernel_if->va);
+ queue->kernel_if->va = NULL;
+ }
+
+ while (i)
+ __free_page(queue->kernel_if->page[--i]);
+
+ vfree(queue->q_header);
+ }
+}
+
+/*
+ * Allocates kernel VA space of specified size, plus space for the
+ * queue structure/kernel interface and the queue header. Allocates
+ * physical pages for the queue data pages.
+ *
+ * PAGE m: struct vmci_queue_header (struct vmci_queue->q_header)
+ * PAGE m+1: struct vmci_queue
+ * PAGE m+1+q: struct vmci_queue_kern_if (struct vmci_queue->kernel_if)
+ * PAGE n-size: Data pages (struct vmci_queue->kernel_if->page[])
+ */
+static void *qp_alloc_queue(u64 size, u32 flags)
+{
+ u64 i;
+ struct vmci_queue *queue;
+ struct vmci_queue_header *q_header;
+ const u64 num_data_pages = dm_div_up(size, PAGE_SIZE);
+ const uint queue_size =
+ PAGE_SIZE +
+ sizeof(*queue) + sizeof(*(queue->kernel_if)) +
+ num_data_pages * sizeof(*(queue->kernel_if->page));
+
+ BUG_ON(size > VMCI_MAX_GUEST_QP_MEMORY);
+ BUG_ON(vmci_qp_pinned(flags) && size > VMCI_MAX_PINNED_QP_MEMORY);
+
+ q_header = vmalloc(queue_size);
+ if (!q_header)
+ return NULL;
+
+ queue = (void *)q_header + PAGE_SIZE;
+ queue->q_header = q_header;
+ queue->saved_header = NULL;
+ queue->kernel_if = (struct vmci_queue_kern_if *)(queue + 1);
+ queue->kernel_if->header_page = NULL; /* Unused in guest. */
+ queue->kernel_if->page = (struct page **)(queue->kernel_if + 1);
+ queue->kernel_if->host = false;
+ queue->kernel_if->va = NULL;
+ queue->kernel_if->mapped = false;
+
+ for (i = 0; i < num_data_pages; i++) {
+ queue->kernel_if->page[i] = alloc_pages(GFP_KERNEL, 0);
+ if (!queue->kernel_if->page[i])
+ goto fail;
+ }
+
+ if (vmci_qp_pinned(flags)) {
+ queue->kernel_if->va =
+ vmap(queue->kernel_if->page, num_data_pages, VM_MAP,
+ PAGE_KERNEL);
+ if (!queue->kernel_if->va)
+ goto fail;
+
+ queue->kernel_if->mapped = true;
+ }
+
+ return (void *)queue;
+
+ fail:
+ qp_free_queue(queue, i * PAGE_SIZE);
+ return NULL;
+}
+
+/*
+ * Copies from a given buffer or iovector to a VMCI Queue. Uses
+ * kmap()/kunmap() to dynamically map/unmap required portions of the queue
+ * by traversing the offset -> page translation structure for the queue.
+ * Assumes that offset + size does not wrap around in the queue.
+ */
+static int __qp_memcpy_to_queue(struct vmci_queue *queue,
+ u64 queue_offset,
+ const void *src,
+ size_t size,
+ bool is_iovec)
+{
+ struct vmci_queue_kern_if *kernel_if = queue->kernel_if;
+ size_t bytes_copied = 0;
+
+ while (bytes_copied < size) {
+ u64 page_index = (queue_offset + bytes_copied) / PAGE_SIZE;
+ size_t page_offset =
+ (queue_offset + bytes_copied) & (PAGE_SIZE - 1);
+ void *va;
+ size_t to_copy;
+
+ if (!kernel_if->mapped)
+ va = kmap(kernel_if->page[page_index]);
+ else
+ va = (void *)((u8 *) kernel_if->va +
+ (page_index * PAGE_SIZE));
+
+ if (size - bytes_copied > PAGE_SIZE - page_offset)
+ /* Enough payload to fill up from this page. */
+ to_copy = PAGE_SIZE - page_offset;
+ else
+ to_copy = size - bytes_copied;
+
+ if (is_iovec) {
+ struct iovec *iov = (struct iovec *)src;
+ int err;
+
+ /* The iovec will track bytes_copied internally. */
+ err = memcpy_fromiovec((u8 *) va + page_offset,
+ iov, to_copy);
+ if (err != 0) {
+ kunmap(kernel_if->page[page_index]);
+ return VMCI_ERROR_INVALID_ARGS;
+ }
+ } else {
+ memcpy((u8 *) va + page_offset,
+ (u8 *) src + bytes_copied, to_copy);
+ }
+
+ bytes_copied += to_copy;
+ if (!kernel_if->mapped)
+ kunmap(kernel_if->page[page_index]);
+ }
+
+ return VMCI_SUCCESS;
+}
+
+/*
+ * Copies to a given buffer or iovector from a VMCI Queue. Uses
+ * kmap()/kunmap() to dynamically map/unmap required portions of the queue
+ * by traversing the offset -> page translation structure for the queue.
+ * Assumes that offset + size does not wrap around in the queue.
+ */
+static int __qp_memcpy_from_queue(void *dest,
+ const struct vmci_queue *queue,
+ u64 queue_offset,
+ size_t size,
+ bool is_iovec)
+{
+ struct vmci_queue_kern_if *kernel_if = queue->kernel_if;
+ size_t bytes_copied = 0;
+
+ while (bytes_copied < size) {
+ u64 page_index = (queue_offset + bytes_copied) / PAGE_SIZE;
+ size_t page_offset =
+ (queue_offset + bytes_copied) & (PAGE_SIZE - 1);
+ void *va;
+ size_t to_copy;
+
+ if (!kernel_if->mapped)
+ va = kmap(kernel_if->page[page_index]);
+ else
+ va = (void *)((u8 *) kernel_if->va +
+ (page_index * PAGE_SIZE));
+
+ if (size - bytes_copied > PAGE_SIZE - page_offset)
+ /* Enough payload to fill up this page. */
+ to_copy = PAGE_SIZE - page_offset;
+ else
+ to_copy = size - bytes_copied;
+
+ if (is_iovec) {
+ struct iovec *iov = (struct iovec *)dest;
+ int err;
+
+ /* The iovec will track bytes_copied internally. */
+ err = memcpy_toiovec(iov, (u8 *) va + page_offset,
+ to_copy);
+ if (err != 0) {
+ kunmap(kernel_if->page[page_index]);
+ return VMCI_ERROR_INVALID_ARGS;
+ }
+ } else {
+ memcpy((u8 *) dest + bytes_copied,
+ (u8 *) va + page_offset, to_copy);
+ }
+
+ bytes_copied += to_copy;
+ if (!kernel_if->mapped)
+ kunmap(kernel_if->page[page_index]);
+ }
+
+ return VMCI_SUCCESS;
+}
+
+/*
+ * Allocates two list of PPNs --- one for the pages in the produce queue,
+ * and the other for the pages in the consume queue. Intializes the list
+ * of PPNs with the page frame numbers of the KVA for the two queues (and
+ * the queue headers).
+ */
+static int qp_alloc_ppn_set(void *prod_q,
+ u64 num_produce_pages,
+ void *cons_q,
+ u64 num_consume_pages, struct PPNSet *ppn_set)
+{
+ u32 *produce_ppns;
+ u32 *consume_ppns;
+ struct vmci_queue *produce_q = prod_q;
+ struct vmci_queue *consume_q = cons_q;
+ u64 i;
+
+ if (!produce_q || !num_produce_pages || !consume_q ||
+ !num_consume_pages || !ppn_set)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ if (ppn_set->initialized)
+ return VMCI_ERROR_ALREADY_EXISTS;
+
+ produce_ppns =
+ kmalloc(num_produce_pages * sizeof(*produce_ppns), GFP_KERNEL);
+ if (!produce_ppns)
+ return VMCI_ERROR_NO_MEM;
+
+ consume_ppns =
+ kmalloc(num_consume_pages * sizeof(*consume_ppns), GFP_KERNEL);
+ if (!consume_ppns) {
+ kfree(produce_ppns);
+ return VMCI_ERROR_NO_MEM;
+ }
+
+ produce_ppns[0] = page_to_pfn(vmalloc_to_page(produce_q->q_header));
+ for (i = 1; i < num_produce_pages; i++) {
+ unsigned long pfn;
+
+ produce_ppns[i] =
+ page_to_pfn(produce_q->kernel_if->page[i - 1]);
+ pfn = produce_ppns[i];
+
+ /* Fail allocation if PFN isn't supported by hypervisor. */
+ if (sizeof(pfn) > sizeof(*produce_ppns)
+ && pfn != produce_ppns[i])
+ goto ppn_error;
+ }
+
+ consume_ppns[0] = page_to_pfn(vmalloc_to_page(consume_q->q_header));
+ for (i = 1; i < num_consume_pages; i++) {
+ unsigned long pfn;
+
+ consume_ppns[i] =
+ page_to_pfn(consume_q->kernel_if->page[i - 1]);
+ pfn = consume_ppns[i];
+
+ /* Fail allocation if PFN isn't supported by hypervisor. */
+ if (sizeof(pfn) > sizeof(*consume_ppns)
+ && pfn != consume_ppns[i])
+ goto ppn_error;
+ }
+
+ ppn_set->num_produce_pages = num_produce_pages;
+ ppn_set->num_consume_pages = num_consume_pages;
+ ppn_set->produce_ppns = produce_ppns;
+ ppn_set->consume_ppns = consume_ppns;
+ ppn_set->initialized = true;
+ return VMCI_SUCCESS;
+
+ ppn_error:
+ kfree(produce_ppns);
+ kfree(consume_ppns);
+ return VMCI_ERROR_INVALID_ARGS;
+}
+
+/*
+ * Frees the two list of PPNs for a queue pair.
+ */
+static void qp_free_ppn_set(struct PPNSet *ppn_set)
+{
+ if (ppn_set->initialized) {
+ /* Do not call these functions on NULL inputs. */
+ kfree(ppn_set->produce_ppns);
+ kfree(ppn_set->consume_ppns);
+ }
+ memset(ppn_set, 0, sizeof(*ppn_set));
+}
+
+/*
+ * Populates the list of PPNs in the hypercall structure with the PPNS
+ * of the produce queue and the consume queue.
+ */
+static int qp_populate_ppn_set(u8 *call_buf, const struct PPNSet *ppn_set)
+{
+ memcpy(call_buf, ppn_set->produce_ppns,
+ ppn_set->num_produce_pages * sizeof(*ppn_set->produce_ppns));
+ memcpy(call_buf +
+ ppn_set->num_produce_pages * sizeof(*ppn_set->produce_ppns),
+ ppn_set->consume_ppns,
+ ppn_set->num_consume_pages * sizeof(*ppn_set->consume_ppns));
+
+ return VMCI_SUCCESS;
+}
+
+static int qp_memcpy_to_queue(struct vmci_queue *queue,
+ u64 queue_offset,
+ const void *src, size_t src_offset, size_t size)
+{
+ return __qp_memcpy_to_queue(queue, queue_offset,
+ (u8 *) src + src_offset, size, false);
+}
+
+static int qp_memcpy_from_queue(void *dest,
+ size_t dest_offset,
+ const struct vmci_queue *queue,
+ u64 queue_offset, size_t size)
+{
+ return __qp_memcpy_from_queue((u8 *) dest + dest_offset,
+ queue, queue_offset, size, false);
+}
+
+/*
+ * Copies from a given iovec from a VMCI Queue.
+ */
+static int qp_memcpy_to_queue_iov(struct vmci_queue *queue,
+ u64 queue_offset,
+ const void *src,
+ size_t src_offset, size_t size)
+{
+
+ /*
+ * We ignore src_offset because src is really a struct iovec * and will
+ * maintain offset internally.
+ */
+ return __qp_memcpy_to_queue(queue, queue_offset, src, size, true);
+}
+
+/*
+ * Copies to a given iovec from a VMCI Queue.
+ */
+static int qp_memcpy_from_queue_iov(void *dest,
+ size_t dest_offset,
+ const struct vmci_queue *queue,
+ u64 queue_offset, size_t size)
+{
+ /*
+ * We ignore dest_offset because dest is really a struct iovec * and
+ * will maintain offset internally.
+ */
+ return __qp_memcpy_from_queue(dest, queue, queue_offset, size, true);
+}
+
+/*
+ * Allocates kernel VA space of specified size plus space for the queue
+ * and kernel interface. This is different from the guest queue allocator,
+ * because we do not allocate our own queue header/data pages here but
+ * share those of the guest.
+ */
+static struct vmci_queue *qp_host_alloc_queue(u64 size)
+{
+ struct vmci_queue *queue;
+ const size_t num_pages = dm_div_up(size, PAGE_SIZE) + 1;
+ const size_t queue_size = sizeof(*queue) + sizeof(*(queue->kernel_if));
+ const size_t queue_page_size =
+ num_pages * sizeof(*queue->kernel_if->page);
+
+ queue = kzalloc(queue_size + queue_page_size, GFP_KERNEL);
+ if (queue) {
+ queue->q_header = NULL;
+ queue->saved_header = NULL;
+ queue->kernel_if =
+ (struct vmci_queue_kern_if *)((u8 *) queue +
+ sizeof(*queue));
+ queue->kernel_if->host = true;
+ queue->kernel_if->mutex = NULL;
+ queue->kernel_if->num_pages = num_pages;
+ queue->kernel_if->header_page =
+ (struct page **)((u8 *) queue + queue_size);
+ queue->kernel_if->page = &queue->kernel_if->header_page[1];
+ queue->kernel_if->va = NULL;
+ queue->kernel_if->mapped = false;
+ }
+
+ return queue;
+}
+
+/*
+ * Frees kernel memory for a given queue (header plus translation
+ * structure).
+ */
+static void qp_host_free_queue(struct vmci_queue *queue, u64 queue_size)
+{
+ kfree(queue);
+}
+
+/*
+ * Initialize the mutex for the pair of queues. This mutex is used to
+ * protect the q_header and the buffer from changing out from under any
+ * users of either queue. Of course, it's only any good if the mutexes
+ * are actually acquired. Queue structure must lie on non-paged memory
+ * or we cannot guarantee access to the mutex.
+ */
+static void qp_init_queue_mutex(struct vmci_queue *produce_q,
+ struct vmci_queue *consume_q)
+{
+ /*
+ * Only the host queue has shared state - the guest queues do not
+ * need to synchronize access using a queue mutex.
+ */
+
+ if (produce_q->kernel_if->host) {
+ produce_q->kernel_if->mutex = &produce_q->kernel_if->__mutex;
+ consume_q->kernel_if->mutex = &produce_q->kernel_if->__mutex;
+ mutex_init(produce_q->kernel_if->mutex);
+ }
+}
+
+/*
+ * Cleans up the mutex for the pair of queues.
+ */
+static void qp_cleanup_queue_mutex(struct vmci_queue *produce_q,
+ struct vmci_queue *consume_q)
+{
+ if (produce_q->kernel_if->host) {
+ produce_q->kernel_if->mutex = NULL;
+ consume_q->kernel_if->mutex = NULL;
+ }
+}
+
+/*
+ * Acquire the mutex for the queue. Note that the produce_q and
+ * the consume_q share a mutex. So, only one of the two need to
+ * be passed in to this routine. Either will work just fine.
+ */
+static void qp_acquire_queue_mutex(struct vmci_queue *queue)
+{
+ if (queue->kernel_if->host) {
+ mutex_lock(queue->kernel_if->mutex);
+ }
+}
+
+/*
+ * Release the mutex for the queue. Note that the produce_q and
+ * the consume_q share a mutex. So, only one of the two need to
+ * be passed in to this routine. Either will work just fine.
+ */
+static void qp_release_queue_mutex(struct vmci_queue *queue)
+{
+ if (queue->kernel_if->host) {
+ mutex_unlock(queue->kernel_if->mutex);
+ }
+}
+
+/*
+ * Helper function to release pages in the PageStoreAttachInfo
+ * previously obtained using get_user_pages.
+ */
+static void qp_release_pages(struct page **pages,
+ u64 num_pages, bool dirty)
+{
+ int i;
+
+ for (i = 0; i < num_pages; i++) {
+ if (dirty)
+ set_page_dirty(pages[i]);
+
+ page_cache_release(pages[i]);
+ pages[i] = NULL;
+ }
+}
+
+/*
+ * Lock the user pages referenced by the {produce,consume}Buffer
+ * struct into memory and populate the {produce,consume}Pages
+ * arrays in the attach structure with them.
+ */
+static int qp_host_get_user_memory(u64 produce_uva,
+ u64 consume_uva,
+ struct vmci_queue *produce_q,
+ struct vmci_queue *consume_q)
+{
+ int retval;
+ int err = VMCI_SUCCESS;
+
+ down_write(¤t->mm->mmap_sem);
+ retval = get_user_pages(current,
+ current->mm,
+ (uintptr_t) produce_uva,
+ produce_q->kernel_if->num_pages,
+ 1, 0, produce_q->kernel_if->header_page, NULL);
+ if (retval < produce_q->kernel_if->num_pages) {
+ pr_warn("get_user_pages(produce) failed (retval=%d)", retval);
+ qp_release_pages(produce_q->kernel_if->header_page, retval,
+ false);
+ err = VMCI_ERROR_NO_MEM;
+ goto out;
+ }
+
+ retval = get_user_pages(current,
+ current->mm,
+ (uintptr_t) consume_uva,
+ consume_q->kernel_if->num_pages,
+ 1, 0, consume_q->kernel_if->header_page, NULL);
+ if (retval < consume_q->kernel_if->num_pages) {
+ pr_warn("get_user_pages(consume) failed (retval=%d)", retval);
+ qp_release_pages(consume_q->kernel_if->header_page, retval,
+ false);
+ qp_release_pages(produce_q->kernel_if->header_page,
+ produce_q->kernel_if->num_pages, false);
+ err = VMCI_ERROR_NO_MEM;
+ }
+
+ out:
+ up_write(¤t->mm->mmap_sem);
+
+ return err;
+}
+
+/*
+ * Registers the specification of the user pages used for backing a queue
+ * pair. Enough information to map in pages is stored in the OS specific
+ * part of the struct vmci_queue structure.
+ */
+static int qp_host_register_user_memory(struct vmci_qp_page_store *page_store,
+ struct vmci_queue *produce_q,
+ struct vmci_queue *consume_q)
+{
+ u64 produce_uva;
+ u64 consume_uva;
+
+ /*
+ * The new style and the old style mapping only differs in
+ * that we either get a single or two UVAs, so we split the
+ * single UVA range at the appropriate spot.
+ */
+ produce_uva = page_store->pages;
+ consume_uva = page_store->pages +
+ produce_q->kernel_if->num_pages * PAGE_SIZE;
+ return qp_host_get_user_memory(produce_uva, consume_uva, produce_q,
+ consume_q);
+}
+
+/*
+ * Releases and removes the references to user pages stored in the attach
+ * struct. Pages are released from the page cache and may become
+ * swappable again.
+ */
+static void qp_host_unregister_user_memory(struct vmci_queue *produce_q,
+ struct vmci_queue *consume_q)
+{
+ qp_release_pages(produce_q->kernel_if->header_page,
+ produce_q->kernel_if->num_pages, true);
+ memset(produce_q->kernel_if->header_page, 0,
+ sizeof(*produce_q->kernel_if->header_page) *
+ produce_q->kernel_if->num_pages);
+ qp_release_pages(consume_q->kernel_if->header_page,
+ consume_q->kernel_if->num_pages, true);
+ memset(consume_q->kernel_if->header_page, 0,
+ sizeof(*consume_q->kernel_if->header_page) *
+ consume_q->kernel_if->num_pages);
+}
+
+/*
+ * Once qp_host_register_user_memory has been performed on a
+ * queue, the queue pair headers can be mapped into the
+ * kernel. Once mapped, they must be unmapped with
+ * qp_host_unmap_queues prior to calling
+ * qp_host_unregister_user_memory.
+ * Pages are pinned.
+ */
+static int qp_host_map_queues(struct vmci_queue *produce_q,
+ struct vmci_queue *consume_q)
+{
+ int result;
+
+ if (!produce_q->q_header || !consume_q->q_header) {
+ struct page *headers[2];
+
+ if (produce_q->q_header != consume_q->q_header)
+ return VMCI_ERROR_QUEUEPAIR_MISMATCH;
+
+ if (produce_q->kernel_if->header_page == NULL ||
+ *produce_q->kernel_if->header_page == NULL)
+ return VMCI_ERROR_UNAVAILABLE;
+
+ headers[0] = *produce_q->kernel_if->header_page;
+ headers[1] = *consume_q->kernel_if->header_page;
+
+ produce_q->q_header = vmap(headers, 2, VM_MAP, PAGE_KERNEL);
+ if (produce_q->q_header != NULL) {
+ consume_q->q_header =
+ (struct vmci_queue_header *)((u8 *)
+ produce_q->q_header +
+ PAGE_SIZE);
+ result = VMCI_SUCCESS;
+ } else {
+ pr_warn("vmap failed.");
+ result = VMCI_ERROR_NO_MEM;
+ }
+ } else {
+ result = VMCI_SUCCESS;
+ }
+
+ return result;
+}
+
+/*
+ * Unmaps previously mapped queue pair headers from the kernel.
+ * Pages are unpinned.
+ */
+static int qp_host_unmap_queues(u32 gid,
+ struct vmci_queue *produce_q,
+ struct vmci_queue *consume_q)
+{
+ if (produce_q->q_header) {
+ if (produce_q->q_header < consume_q->q_header)
+ vunmap(produce_q->q_header);
+ else
+ vunmap(consume_q->q_header);
+
+ produce_q->q_header = NULL;
+ consume_q->q_header = NULL;
+ }
+
+ return VMCI_SUCCESS;
+}
+
+/*
+ * Finds the entry in the list corresponding to a given handle. Assumes
+ * that the list is locked.
+ */
+static struct qp_entry *qp_list_find(struct qp_list *qp_list,
+ struct vmci_handle handle)
+{
+ struct qp_entry *entry;
+
+ if (VMCI_HANDLE_INVALID(handle))
+ return NULL;
+
+ list_for_each_entry(entry, &qp_list->head, list_item) {
+ if (VMCI_HANDLE_EQUAL(entry->handle, handle))
+ return entry;
+ }
+
+ return NULL;
+}
+
+/*
+ * Finds the entry in the list corresponding to a given handle.
+ */
+static struct qp_guest_endpoint *
+qp_guest_handle_to_entry(struct vmci_handle handle)
+{
+ struct qp_guest_endpoint *entry;
+ struct qp_entry *qp = qp_list_find(&qp_guest_endpoints, handle);
+
+ entry = qp ? container_of(
+ qp, struct qp_guest_endpoint, qp) : NULL;
+ return entry;
+}
+
+/*
+ * Finds the entry in the list corresponding to a given handle.
+ */
+static struct qp_broker_entry *
+qp_broker_handle_to_entry(struct vmci_handle handle)
+{
+ struct qp_broker_entry *entry;
+ struct qp_entry *qp = qp_list_find(&qp_broker_list, handle);
+
+ entry = qp ? container_of(
+ qp, struct qp_broker_entry, qp) : NULL;
+ return entry;
+}
+
+/*
+ * Dispatches a queue pair event message directly into the local event
+ * queue.
+ */
+static int qp_notify_peer_local(bool attach, struct vmci_handle handle)
+{
+ struct vmci_event_msg *e_msg;
+ struct vmci_event_payld_qp *e_payload;
+ /* buf is only 48 bytes. */
+ char buf[sizeof(*e_msg) + sizeof(*e_payload)];
+ u32 context_id;
+
+ context_id = vmci_get_context_id();
+
+ e_msg = (struct vmci_event_msg *)buf;
+ e_payload = vmci_event_data_payload(&e_msg->event_data);
+
+ e_msg->hdr.dst = vmci_make_handle(context_id, VMCI_EVENT_HANDLER);
+ e_msg->hdr.src = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID,
+ VMCI_CONTEXT_RESOURCE_ID);
+ e_msg->hdr.payload_size =
+ sizeof(*e_msg) + sizeof(*e_payload) - sizeof(e_msg->hdr);
+ e_msg->event_data.event =
+ attach ? VMCI_EVENT_QP_PEER_ATTACH : VMCI_EVENT_QP_PEER_DETACH;
+ e_payload->peer_id = context_id;
+ e_payload->handle = handle;
+
+ return vmci_event_dispatch(&e_msg->hdr);
+}
+
+/*
+ * Allocates and initializes a qp_guest_endpoint structure.
+ * Allocates a queue_pair rid (and handle) iff the given entry has
+ * an invalid handle. 0 through VMCI_RESERVED_RESOURCE_ID_MAX
+ * are reserved handles. Assumes that the QP list mutex is held
+ * by the caller.
+ */
+static struct qp_guest_endpoint *
+qp_guest_endpoint_create(struct vmci_handle handle,
+ u32 peer,
+ u32 flags,
+ u64 produce_size,
+ u64 consume_size,
+ void *produce_q,
+ void *consume_q)
+{
+ int result;
+ struct qp_guest_endpoint *entry;
+ /* One page each for the queue headers. */
+ const u64 num_ppns = dm_div_up(produce_size, PAGE_SIZE) +
+ dm_div_up(consume_size, PAGE_SIZE) + 2;
+
+ if (VMCI_HANDLE_INVALID(handle)) {
+ u32 context_id = vmci_get_context_id();
+
+ handle = vmci_make_handle(context_id, VMCI_INVALID_ID);
+ }
+
+ ASSERT(!VMCI_HANDLE_INVALID(handle) &&
+ qp_list_find(&qp_guest_endpoints, handle) == NULL);
+ entry = kzalloc(sizeof(*entry), GFP_KERNEL);
+ if (entry) {
+ entry->qp.peer = peer;
+ entry->qp.flags = flags;
+ entry->qp.produce_size = produce_size;
+ entry->qp.consume_size = consume_size;
+ entry->qp.ref_count = 0;
+ entry->num_ppns = num_ppns;
+ entry->produce_q = produce_q;
+ entry->consume_q = consume_q;
+ INIT_LIST_HEAD(&entry->qp.list_item);
+
+ /* Add resource obj */
+ result = vmci_resource_add(&entry->resource,
+ VMCI_RESOURCE_TYPE_QPAIR_GUEST,
+ handle);
+ entry->qp.handle = vmci_resource_handle(&entry->resource);
+ if ((result != VMCI_SUCCESS) ||
+ qp_list_find(&qp_guest_endpoints, entry->qp.handle)) {
+ pr_warn("Failed to add new resource (handle=0x%x:0x%x), error: %d",
+ handle.context, handle.resource, result);
+ kfree(entry);
+ entry = NULL;
+ }
+ }
+ return entry;
+}
+
+/*
+ * Frees a qp_guest_endpoint structure.
+ */
+static void qp_guest_endpoint_destroy(struct qp_guest_endpoint *entry)
+{
+ ASSERT(entry->qp.ref_count == 0);
+
+ qp_free_ppn_set(&entry->ppn_set);
+ qp_cleanup_queue_mutex(entry->produce_q, entry->consume_q);
+ qp_free_queue(entry->produce_q, entry->qp.produce_size);
+ qp_free_queue(entry->consume_q, entry->qp.consume_size);
+ /* Unlink from resource hash table and free callback */
+ vmci_resource_remove(&entry->resource);
+
+ kfree(entry);
+}
+
+/*
+ * Helper to make a queue_pairAlloc hypercall when the driver is
+ * supporting a guest device.
+ */
+static int qp_alloc_hypercall(const struct qp_guest_endpoint *entry)
+{
+ struct vmci_qp_alloc_msg *alloc_msg;
+ size_t msg_size;
+ int result;
+
+ if (!entry || entry->num_ppns <= 2)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ ASSERT(!(entry->qp.flags & VMCI_QPFLAG_LOCAL));
+
+ msg_size = sizeof(*alloc_msg) +
+ (size_t) entry->num_ppns * sizeof(u32);
+ alloc_msg = kmalloc(msg_size, GFP_KERNEL);
+ if (!alloc_msg)
+ return VMCI_ERROR_NO_MEM;
+
+ alloc_msg->hdr.dst = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID,
+ VMCI_QUEUEPAIR_ALLOC);
+ alloc_msg->hdr.src = VMCI_ANON_SRC_HANDLE;
+ alloc_msg->hdr.payload_size = msg_size - VMCI_DG_HEADERSIZE;
+ alloc_msg->handle = entry->qp.handle;
+ alloc_msg->peer = entry->qp.peer;
+ alloc_msg->flags = entry->qp.flags;
+ alloc_msg->produce_size = entry->qp.produce_size;
+ alloc_msg->consume_size = entry->qp.consume_size;
+ alloc_msg->num_ppns = entry->num_ppns;
+
+ result = qp_populate_ppn_set((u8 *) alloc_msg + sizeof(*alloc_msg),
+ &entry->ppn_set);
+ if (result == VMCI_SUCCESS)
+ result = vmci_send_datagram(&alloc_msg->hdr);
+
+ kfree(alloc_msg);
+
+ return result;
+}
+
+/*
+ * Helper to make a queue_pairDetach hypercall when the driver is
+ * supporting a guest device.
+ */
+static int qp_detatch_hypercall(struct vmci_handle handle)
+{
+ struct vmci_qp_detach_msg detach_msg;
+
+ detach_msg.hdr.dst = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID,
+ VMCI_QUEUEPAIR_DETACH);
+ detach_msg.hdr.src = VMCI_ANON_SRC_HANDLE;
+ detach_msg.hdr.payload_size = sizeof(handle);
+ detach_msg.handle = handle;
+
+ return vmci_send_datagram(&detach_msg.hdr);
+}
+
+/*
+ * Adds the given entry to the list. Assumes that the list is locked.
+ */
+static void qp_list_add_entry(struct qp_list *qp_list, struct qp_entry *entry)
+{
+ if (entry)
+ list_add(&entry->list_item, &qp_list->head);
+}
+
+/*
+ * Removes the given entry from the list. Assumes that the list is locked.
+ */
+static void qp_list_remove_entry(struct qp_list *qp_list,
+ struct qp_entry *entry)
+{
+ if (entry)
+ list_del(&entry->list_item);
+}
+
+/*
+ * Helper for VMCI queue_pair detach interface. Frees the physical
+ * pages for the queue pair.
+ */
+static int qp_detatch_guest_work(struct vmci_handle handle)
+{
+ int result;
+ struct qp_guest_endpoint *entry;
+ u32 ref_count = ~0; /* To avoid compiler warning below */
+
+ mutex_lock(&qp_guest_endpoints.mutex);
+
+ entry = qp_guest_handle_to_entry(handle);
+ if (!entry) {
+ mutex_unlock(&qp_guest_endpoints.mutex);
+ return VMCI_ERROR_NOT_FOUND;
+ }
+
+ ASSERT(entry->qp.ref_count >= 1);
+
+ if (entry->qp.flags & VMCI_QPFLAG_LOCAL) {
+ result = VMCI_SUCCESS;
+
+ if (entry->qp.ref_count > 1) {
+ result = qp_notify_peer_local(false, handle);
+ /*
+ * We can fail to notify a local queuepair
+ * because we can't allocate. We still want
+ * to release the entry if that happens, so
+ * don't bail out yet.
+ */
+ }
+ } else {
+ result = qp_detatch_hypercall(handle);
+ if (result < VMCI_SUCCESS) {
+ /*
+ * We failed to notify a non-local queuepair.
+ * That other queuepair might still be
+ * accessing the shared memory, so don't
+ * release the entry yet. It will get cleaned
+ * up by VMCIqueue_pair_Exit() if necessary
+ * (assuming we are going away, otherwise why
+ * did this fail?).
+ */
+
+ mutex_unlock(&qp_guest_endpoints.mutex);
+ return result;
+ }
+ }
+
+ /*
+ * If we get here then we either failed to notify a local queuepair, or
+ * we succeeded in all cases. Release the entry if required.
+ */
+
+ entry->qp.ref_count--;
+ if (entry->qp.ref_count == 0)
+ qp_list_remove_entry(&qp_guest_endpoints, &entry->qp);
+
+ /* If we didn't remove the entry, this could change once we unlock. */
+ if (entry)
+ ref_count = entry->qp.ref_count;
+
+ mutex_unlock(&qp_guest_endpoints.mutex);
+
+ if (ref_count == 0)
+ qp_guest_endpoint_destroy(entry);
+
+ return result;
+}
+
+/*
+ * This functions handles the actual allocation of a VMCI queue
+ * pair guest endpoint. Allocates physical pages for the queue
+ * pair. It makes OS dependent calls through generic wrappers.
+ */
+static int qp_alloc_guest_work(struct vmci_handle *handle,
+ struct vmci_queue **produce_q,
+ u64 produce_size,
+ struct vmci_queue **consume_q,
+ u64 consume_size,
+ u32 peer,
+ u32 flags,
+ u32 priv_flags)
+{
+ const u64 num_produce_pages =
+ dm_div_up(produce_size, PAGE_SIZE) + 1;
+ const u64 num_consume_pages =
+ dm_div_up(consume_size, PAGE_SIZE) + 1;
+ void *my_produce_q = NULL;
+ void *my_consume_q = NULL;
+ int result;
+ struct qp_guest_endpoint *queue_pair_entry = NULL;
+
+ if (priv_flags != VMCI_NO_PRIVILEGE_FLAGS)
+ return VMCI_ERROR_NO_ACCESS;
+
+ mutex_lock(&qp_guest_endpoints.mutex);
+
+ queue_pair_entry = qp_guest_handle_to_entry(*handle);
+ if (queue_pair_entry) {
+ if (queue_pair_entry->qp.flags & VMCI_QPFLAG_LOCAL) {
+ /* Local attach case. */
+ if (queue_pair_entry->qp.ref_count > 1) {
+ pr_devel("Error attempting to attach more "
+ "than once.");
+ result = VMCI_ERROR_UNAVAILABLE;
+ goto error_keep_entry;
+ }
+
+ if (queue_pair_entry->qp.produce_size != consume_size ||
+ queue_pair_entry->qp.consume_size !=
+ produce_size ||
+ queue_pair_entry->qp.flags !=
+ (flags & ~VMCI_QPFLAG_ATTACH_ONLY)) {
+ pr_devel("Error mismatched queue pair in "
+ "local attach.");
+ result = VMCI_ERROR_QUEUEPAIR_MISMATCH;
+ goto error_keep_entry;
+ }
+
+ /*
+ * Do a local attach. We swap the consume and
+ * produce queues for the attacher and deliver
+ * an attach event.
+ */
+ result = qp_notify_peer_local(true, *handle);
+ if (result < VMCI_SUCCESS)
+ goto error_keep_entry;
+
+ my_produce_q = queue_pair_entry->consume_q;
+ my_consume_q = queue_pair_entry->produce_q;
+ goto out;
+ }
+
+ result = VMCI_ERROR_ALREADY_EXISTS;
+ goto error_keep_entry;
+ }
+
+ my_produce_q = qp_alloc_queue(produce_size, flags);
+ if (!my_produce_q) {
+ pr_warn("Error allocating pages for produce queue.");
+ result = VMCI_ERROR_NO_MEM;
+ goto error;
+ }
+
+ my_consume_q = qp_alloc_queue(consume_size, flags);
+ if (!my_consume_q) {
+ pr_warn("Error allocating pages for consume queue.");
+ result = VMCI_ERROR_NO_MEM;
+ goto error;
+ }
+
+ queue_pair_entry = qp_guest_endpoint_create(*handle, peer, flags,
+ produce_size, consume_size,
+ my_produce_q, my_consume_q);
+ if (!queue_pair_entry) {
+ pr_warn("Error allocating memory in %s.", __func__);
+ result = VMCI_ERROR_NO_MEM;
+ goto error;
+ }
+
+ result = qp_alloc_ppn_set(my_produce_q, num_produce_pages, my_consume_q,
+ num_consume_pages,
+ &queue_pair_entry->ppn_set);
+ if (result < VMCI_SUCCESS) {
+ pr_warn("qp_alloc_ppn_set failed.");
+ goto error;
+ }
+
+ /*
+ * It's only necessary to notify the host if this queue pair will be
+ * attached to from another context.
+ */
+ if (queue_pair_entry->qp.flags & VMCI_QPFLAG_LOCAL) {
+ /* Local create case. */
+ u32 context_id = vmci_get_context_id();
+
+ /*
+ * Enforce similar checks on local queue pairs as we
+ * do for regular ones. The handle's context must
+ * match the creator or attacher context id (here they
+ * are both the current context id) and the
+ * attach-only flag cannot exist during create. We
+ * also ensure specified peer is this context or an
+ * invalid one.
+ */
+ if (queue_pair_entry->qp.handle.context != context_id ||
+ (queue_pair_entry->qp.peer != VMCI_INVALID_ID &&
+ queue_pair_entry->qp.peer != context_id)) {
+ result = VMCI_ERROR_NO_ACCESS;
+ goto error;
+ }
+
+ if (queue_pair_entry->qp.flags & VMCI_QPFLAG_ATTACH_ONLY) {
+ result = VMCI_ERROR_NOT_FOUND;
+ goto error;
+ }
+ } else {
+ result = qp_alloc_hypercall(queue_pair_entry);
+ if (result < VMCI_SUCCESS) {
+ pr_warn("qp_alloc_hypercall result = %d.", result);
+ goto error;
+ }
+ }
+
+ qp_init_queue_mutex((struct vmci_queue *)my_produce_q,
+ (struct vmci_queue *)my_consume_q);
+
+ qp_list_add_entry(&qp_guest_endpoints, &queue_pair_entry->qp);
+
+ out:
+ queue_pair_entry->qp.ref_count++;
+ *handle = queue_pair_entry->qp.handle;
+ *produce_q = (struct vmci_queue *)my_produce_q;
+ *consume_q = (struct vmci_queue *)my_consume_q;
+
+ /*
+ * We should initialize the queue pair header pages on a local
+ * queue pair create. For non-local queue pairs, the
+ * hypervisor initializes the header pages in the create step.
+ */
+ if ((queue_pair_entry->qp.flags & VMCI_QPFLAG_LOCAL) &&
+ queue_pair_entry->qp.ref_count == 1) {
+ vmci_q_header_init((*produce_q)->q_header, *handle);
+ vmci_q_header_init((*consume_q)->q_header, *handle);
+ }
+
+ mutex_unlock(&qp_guest_endpoints.mutex);
+
+ return VMCI_SUCCESS;
+
+ error:
+ mutex_unlock(&qp_guest_endpoints.mutex);
+ if (queue_pair_entry) {
+ /* The queues will be freed inside the destroy routine. */
+ qp_guest_endpoint_destroy(queue_pair_entry);
+ } else {
+ qp_free_queue(my_produce_q, produce_size);
+ qp_free_queue(my_consume_q, consume_size);
+ }
+ return result;
+
+ error_keep_entry:
+ /* This path should only be used when an existing entry was found. */
+ ASSERT(queue_pair_entry->qp.ref_count > 0);
+ mutex_unlock(&qp_guest_endpoints.mutex);
+ return result;
+}
+
+/*
+ * The first endpoint issuing a queue pair allocation will create the state
+ * of the queue pair in the queue pair broker.
+ *
+ * If the creator is a guest, it will associate a VMX virtual address range
+ * with the queue pair as specified by the page_store. For compatibility with
+ * older VMX'en, that would use a separate step to set the VMX virtual
+ * address range, the virtual address range can be registered later using
+ * vmci_qp_broker_set_page_store. In that case, a page_store of NULL should be
+ * used.
+ *
+ * If the creator is the host, a page_store of NULL should be used as well,
+ * since the host is not able to supply a page store for the queue pair.
+ *
+ * For older VMX and host callers, the queue pair will be created in the
+ * VMCIQPB_CREATED_NO_MEM state, and for current VMX callers, it will be
+ * created in VMCOQPB_CREATED_MEM state.
+ */
+static int qp_broker_create(struct vmci_handle handle,
+ u32 peer,
+ u32 flags,
+ u32 priv_flags,
+ u64 produce_size,
+ u64 consume_size,
+ struct vmci_qp_page_store *page_store,
+ struct vmci_ctx *context,
+ vmci_event_release_cb wakeup_cb,
+ void *client_data, struct qp_broker_entry **ent)
+{
+ struct qp_broker_entry *entry = NULL;
+ const u32 context_id = vmci_ctx_get_id(context);
+ bool is_local = flags & VMCI_QPFLAG_LOCAL;
+ int result;
+ u64 guest_produce_size;
+ u64 guest_consume_size;
+
+ /* Do not create if the caller asked not to. */
+ if (flags & VMCI_QPFLAG_ATTACH_ONLY)
+ return VMCI_ERROR_NOT_FOUND;
+
+ /*
+ * Creator's context ID should match handle's context ID or the creator
+ * must allow the context in handle's context ID as the "peer".
+ */
+ if (handle.context != context_id && handle.context != peer)
+ return VMCI_ERROR_NO_ACCESS;
+
+ if (VMCI_CONTEXT_IS_VM(context_id) && VMCI_CONTEXT_IS_VM(peer))
+ return VMCI_ERROR_DST_UNREACHABLE;
+
+ /*
+ * Creator's context ID for local queue pairs should match the
+ * peer, if a peer is specified.
+ */
+ if (is_local && peer != VMCI_INVALID_ID && context_id != peer)
+ return VMCI_ERROR_NO_ACCESS;
+
+ entry = kzalloc(sizeof(*entry), GFP_ATOMIC);
+ if (!entry)
+ return VMCI_ERROR_NO_MEM;
+
+ if (vmci_ctx_get_id(context) == VMCI_HOST_CONTEXT_ID && !is_local) {
+ /*
+ * The queue pair broker entry stores values from the guest
+ * point of view, so a creating host side endpoint should swap
+ * produce and consume values -- unless it is a local queue
+ * pair, in which case no swapping is necessary, since the local
+ * attacher will swap queues.
+ */
+
+ guest_produce_size = consume_size;
+ guest_consume_size = produce_size;
+ } else {
+ guest_produce_size = produce_size;
+ guest_consume_size = consume_size;
+ }
+
+ entry->qp.handle = handle;
+ entry->qp.peer = peer;
+ entry->qp.flags = flags;
+ entry->qp.produce_size = guest_produce_size;
+ entry->qp.consume_size = guest_consume_size;
+ entry->qp.ref_count = 1;
+ entry->create_id = context_id;
+ entry->attach_id = VMCI_INVALID_ID;
+ entry->state = VMCIQPB_NEW;
+ entry->require_trusted_attach =
+ !!(context->priv_flags & VMCI_PRIVILEGE_FLAG_RESTRICTED);
+ entry->created_by_trusted =
+ !!(priv_flags & VMCI_PRIVILEGE_FLAG_TRUSTED);
+ entry->vmci_page_files = false;
+ entry->wakeup_cb = wakeup_cb;
+ entry->client_data = client_data;
+ entry->produce_q = qp_host_alloc_queue(guest_produce_size);
+ if (entry->produce_q == NULL) {
+ result = VMCI_ERROR_NO_MEM;
+ goto error;
+ }
+ entry->consume_q = qp_host_alloc_queue(guest_consume_size);
+ if (entry->consume_q == NULL) {
+ result = VMCI_ERROR_NO_MEM;
+ goto error;
+ }
+
+ qp_init_queue_mutex(entry->produce_q, entry->consume_q);
+
+ INIT_LIST_HEAD(&entry->qp.list_item);
+
+ if (is_local) {
+ u8 *tmp;
+
+ entry->local_mem = kcalloc(QPE_NUM_PAGES(entry->qp),
+ PAGE_SIZE, GFP_KERNEL);
+ if (entry->local_mem == NULL) {
+ result = VMCI_ERROR_NO_MEM;
+ goto error;
+ }
+ entry->state = VMCIQPB_CREATED_MEM;
+ entry->produce_q->q_header = entry->local_mem;
+ tmp = (u8 *) entry->local_mem + PAGE_SIZE *
+ (dm_div_up(entry->qp.produce_size, PAGE_SIZE) + 1);
+ entry->consume_q->q_header = (struct vmci_queue_header *)tmp;
+ } else if (page_store) {
+ ASSERT(entry->create_id != VMCI_HOST_CONTEXT_ID || is_local);
+
+ /*
+ * The VMX already initialized the queue pair headers, so no
+ * need for the kernel side to do that.
+ */
+ result = qp_host_register_user_memory(page_store,
+ entry->produce_q,
+ entry->consume_q);
+ if (result < VMCI_SUCCESS)
+ goto error;
+
+ entry->state = VMCIQPB_CREATED_MEM;
+ } else {
+ /*
+ * A create without a page_store may be either a host
+ * side create (in which case we are waiting for the
+ * guest side to supply the memory) or an old style
+ * queue pair create (in which case we will expect a
+ * set page store call as the next step).
+ */
+ entry->state = VMCIQPB_CREATED_NO_MEM;
+ }
+
+ qp_list_add_entry(&qp_broker_list, &entry->qp);
+ if (ent != NULL)
+ *ent = entry;
+
+ /* Add to resource obj */
+ result = vmci_resource_add(&entry->resource,
+ VMCI_RESOURCE_TYPE_QPAIR_HOST,
+ handle);
+ if (result != VMCI_SUCCESS) {
+ pr_warn("Failed to add new resource (handle=0x%x:0x%x), error: %d",
+ handle.context, handle.resource, result);
+ goto error;
+ }
+
+ entry->qp.handle = vmci_resource_handle(&entry->resource);
+ if (is_local) {
+ vmci_q_header_init(entry->produce_q->q_header,
+ entry->qp.handle);
+ vmci_q_header_init(entry->consume_q->q_header,
+ entry->qp.handle);
+ }
+
+ vmci_ctx_qp_create(context, entry->qp.handle);
+
+ return VMCI_SUCCESS;
+
+ error:
+ if (entry != NULL) {
+ qp_host_free_queue(entry->produce_q, guest_produce_size);
+ qp_host_free_queue(entry->consume_q, guest_consume_size);
+ kfree(entry);
+ }
+
+ return result;
+}
+
+/*
+ * Enqueues an event datagram to notify the peer VM attached to
+ * the given queue pair handle about attach/detach event by the
+ * given VM. Returns Payload size of datagram enqueued on
+ * success, error code otherwise.
+ */
+static int qp_notify_peer(bool attach,
+ struct vmci_handle handle,
+ u32 my_id,
+ u32 peer_id)
+{
+ int rv;
+ struct vmci_event_msg *e_msg;
+ struct vmci_event_payld_qp *ev_payload;
+ char buf[sizeof(*e_msg) + sizeof(*ev_payload)];
+
+ if (VMCI_HANDLE_INVALID(handle) || my_id == VMCI_INVALID_ID ||
+ peer_id == VMCI_INVALID_ID)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ /*
+ * Notification message contains: queue pair handle and
+ * attaching/detaching VM's context id.
+ */
+ e_msg = (struct vmci_event_msg *)buf;
+
+ /*
+ * In vmci_ctx_enqueue_datagram() we enforce the upper limit on
+ * number of pending events from the hypervisor to a given VM
+ * otherwise a rogue VM could do an arbitrary number of attach
+ * and detach operations causing memory pressure in the host
+ * kernel.
+ */
+
+ /* Clear out any garbage. */
+ memset(e_msg, 0, sizeof(buf));
+
+ e_msg->hdr.dst = vmci_make_handle(peer_id, VMCI_EVENT_HANDLER);
+ e_msg->hdr.src = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID,
+ VMCI_CONTEXT_RESOURCE_ID);
+ e_msg->hdr.payload_size = sizeof(*e_msg) + sizeof(*ev_payload) -
+ sizeof(e_msg->hdr);
+ e_msg->event_data.event = attach ?
+ VMCI_EVENT_QP_PEER_ATTACH : VMCI_EVENT_QP_PEER_DETACH;
+ ev_payload = vmci_event_data_payload(&e_msg->event_data);
+ ev_payload->handle = handle;
+ ev_payload->peer_id = my_id;
+
+ rv = vmci_datagram_dispatch(VMCI_HYPERVISOR_CONTEXT_ID,
+ &e_msg->hdr, false);
+ if (rv < VMCI_SUCCESS)
+ pr_warn("Failed to enqueue queue_pair %s event datagram "
+ "for context (ID=0x%x).", attach ? "ATTACH" : "DETACH",
+ peer_id);
+
+ return rv;
+}
+
+/*
+ * The second endpoint issuing a queue pair allocation will attach to
+ * the queue pair registered with the queue pair broker.
+ *
+ * If the attacher is a guest, it will associate a VMX virtual address
+ * range with the queue pair as specified by the page_store. At this
+ * point, the already attach host endpoint may start using the queue
+ * pair, and an attach event is sent to it. For compatibility with
+ * older VMX'en, that used a separate step to set the VMX virtual
+ * address range, the virtual address range can be registered later
+ * using vmci_qp_broker_set_page_store. In that case, a page_store of
+ * NULL should be used, and the attach event will be generated once
+ * the actual page store has been set.
+ *
+ * If the attacher is the host, a page_store of NULL should be used as
+ * well, since the page store information is already set by the guest.
+ *
+ * For new VMX and host callers, the queue pair will be moved to the
+ * VMCIQPB_ATTACHED_MEM state, and for older VMX callers, it will be
+ * moved to the VMCOQPB_ATTACHED_NO_MEM state.
+ */
+static int qp_broker_attach(struct qp_broker_entry *entry,
+ u32 peer,
+ u32 flags,
+ u32 priv_flags,
+ u64 produce_size,
+ u64 consume_size,
+ struct vmci_qp_page_store *page_store,
+ struct vmci_ctx *context,
+ vmci_event_release_cb wakeup_cb,
+ void *client_data,
+ struct qp_broker_entry **ent)
+{
+ const u32 context_id = vmci_ctx_get_id(context);
+ bool is_local = flags & VMCI_QPFLAG_LOCAL;
+ int result;
+
+ if (entry->state != VMCIQPB_CREATED_NO_MEM &&
+ entry->state != VMCIQPB_CREATED_MEM)
+ return VMCI_ERROR_UNAVAILABLE;
+
+ if (is_local) {
+ if (!(entry->qp.flags & VMCI_QPFLAG_LOCAL) ||
+ context_id != entry->create_id) {
+ return VMCI_ERROR_INVALID_ARGS;
+ }
+ } else if (context_id == entry->create_id ||
+ context_id == entry->attach_id) {
+ return VMCI_ERROR_ALREADY_EXISTS;
+ }
+
+ ASSERT(entry->qp.ref_count < 2);
+ ASSERT(entry->attach_id == VMCI_INVALID_ID);
+
+ if (VMCI_CONTEXT_IS_VM(context_id) &&
+ VMCI_CONTEXT_IS_VM(entry->create_id))
+ return VMCI_ERROR_DST_UNREACHABLE;
+
+ /*
+ * If we are attaching from a restricted context then the queuepair
+ * must have been created by a trusted endpoint.
+ */
+ if ((context->priv_flags & VMCI_PRIVILEGE_FLAG_RESTRICTED) &&
+ !entry->created_by_trusted)
+ return VMCI_ERROR_NO_ACCESS;
+
+ /*
+ * If we are attaching to a queuepair that was created by a restricted
+ * context then we must be trusted.
+ */
+ if (entry->require_trusted_attach &&
+ (!(priv_flags & VMCI_PRIVILEGE_FLAG_TRUSTED)))
+ return VMCI_ERROR_NO_ACCESS;
+
+ /*
+ * If the creator specifies VMCI_INVALID_ID in "peer" field, access
+ * control check is not performed.
+ */
+ if (entry->qp.peer != VMCI_INVALID_ID && entry->qp.peer != context_id)
+ return VMCI_ERROR_NO_ACCESS;
+
+ if (entry->create_id == VMCI_HOST_CONTEXT_ID) {
+ /*
+ * Do not attach if the caller doesn't support Host Queue Pairs
+ * and a host created this queue pair.
+ */
+
+ if (!vmci_ctx_supports_host_qp(context))
+ return VMCI_ERROR_INVALID_RESOURCE;
+
+ } else if (context_id == VMCI_HOST_CONTEXT_ID) {
+ struct vmci_ctx *create_context;
+ bool supports_host_qp;
+
+ /*
+ * Do not attach a host to a user created queue pair if that
+ * user doesn't support host queue pair end points.
+ */
+
+ create_context = vmci_ctx_get(entry->create_id);
+ supports_host_qp = vmci_ctx_supports_host_qp(create_context);
+ vmci_ctx_put(create_context);
+
+ if (!supports_host_qp)
+ return VMCI_ERROR_INVALID_RESOURCE;
+ }
+
+ if ((entry->qp.flags & ~VMCI_QP_ASYMM) != (flags & ~VMCI_QP_ASYMM_PEER))
+ return VMCI_ERROR_QUEUEPAIR_MISMATCH;
+
+ if (context_id != VMCI_HOST_CONTEXT_ID) {
+ /*
+ * The queue pair broker entry stores values from the guest
+ * point of view, so an attaching guest should match the values
+ * stored in the entry.
+ */
+
+ if (entry->qp.produce_size != produce_size ||
+ entry->qp.consume_size != consume_size) {
+ return VMCI_ERROR_QUEUEPAIR_MISMATCH;
+ }
+ } else if (entry->qp.produce_size != consume_size ||
+ entry->qp.consume_size != produce_size) {
+ return VMCI_ERROR_QUEUEPAIR_MISMATCH;
+ }
+
+ if (context_id != VMCI_HOST_CONTEXT_ID) {
+ /*
+ * If a guest attached to a queue pair, it will supply
+ * the backing memory. If this is a pre NOVMVM vmx,
+ * the backing memory will be supplied by calling
+ * vmci_qp_broker_set_page_store() following the
+ * return of the vmci_qp_broker_alloc() call. If it is
+ * a vmx of version NOVMVM or later, the page store
+ * must be supplied as part of the
+ * vmci_qp_broker_alloc call. Under all circumstances
+ * must the initially created queue pair not have any
+ * memory associated with it already.
+ */
+
+ if (entry->state != VMCIQPB_CREATED_NO_MEM)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ if (page_store != NULL) {
+ /*
+ * Patch up host state to point to guest
+ * supplied memory. The VMX already
+ * initialized the queue pair headers, so no
+ * need for the kernel side to do that.
+ */
+
+ result = qp_host_register_user_memory(page_store,
+ entry->produce_q,
+ entry->consume_q);
+ if (result < VMCI_SUCCESS)
+ return result;
+
+ /*
+ * Preemptively load in the headers if non-blocking to
+ * prevent blocking later.
+ */
+ if (entry->qp.flags & VMCI_QPFLAG_NONBLOCK) {
+ result = qp_host_map_queues(entry->produce_q,
+ entry->consume_q);
+ if (result < VMCI_SUCCESS) {
+ qp_host_unregister_user_memory(
+ entry->produce_q,
+ entry->consume_q);
+ return result;
+ }
+ }
+
+ entry->state = VMCIQPB_ATTACHED_MEM;
+ } else {
+ entry->state = VMCIQPB_ATTACHED_NO_MEM;
+ }
+ } else if (entry->state == VMCIQPB_CREATED_NO_MEM) {
+ /*
+ * The host side is attempting to attach to a queue
+ * pair that doesn't have any memory associated with
+ * it. This must be a pre NOVMVM vmx that hasn't set
+ * the page store information yet, or a quiesced VM.
+ */
+
+ return VMCI_ERROR_UNAVAILABLE;
+ } else {
+ /*
+ * For non-blocking queue pairs, we cannot rely on
+ * enqueue/dequeue to map in the pages on the
+ * host-side, since it may block, so we make an
+ * attempt here.
+ */
+
+ if (flags & VMCI_QPFLAG_NONBLOCK) {
+ result =
+ qp_host_map_queues(entry->produce_q,
+ entry->consume_q);
+ if (result < VMCI_SUCCESS)
+ return result;
+
+ entry->qp.flags |= flags &
+ (VMCI_QPFLAG_NONBLOCK | VMCI_QPFLAG_PINNED);
+ }
+
+ /* The host side has successfully attached to a queue pair. */
+ entry->state = VMCIQPB_ATTACHED_MEM;
+ }
+
+ if (entry->state == VMCIQPB_ATTACHED_MEM) {
+ result =
+ qp_notify_peer(true, entry->qp.handle, context_id,
+ entry->create_id);
+ if (result < VMCI_SUCCESS)
+ pr_warn("Failed to notify peer (ID=0x%x) of "
+ "attach to queue pair (handle=0x%x:0x%x).",
+ entry->create_id, entry->qp.handle.context,
+ entry->qp.handle.resource);
+ }
+
+ entry->attach_id = context_id;
+ entry->qp.ref_count++;
+ if (wakeup_cb) {
+ entry->wakeup_cb = wakeup_cb;
+ entry->client_data = client_data;
+ }
+
+ /*
+ * When attaching to local queue pairs, the context already has
+ * an entry tracking the queue pair, so don't add another one.
+ */
+ if (!is_local)
+ vmci_ctx_qp_create(context, entry->qp.handle);
+ else
+ ASSERT(vmci_ctx_qp_exists(context, entry->qp.handle));
+
+ if (ent != NULL)
+ *ent = entry;
+
+ return VMCI_SUCCESS;
+}
+
+/*
+ * queue_pair_Alloc for use when setting up queue pair endpoints
+ * on the host.
+ */
+static int qp_broker_alloc(struct vmci_handle handle,
+ u32 peer,
+ u32 flags,
+ u32 priv_flags,
+ u64 produce_size,
+ u64 consume_size,
+ struct vmci_qp_page_store *page_store,
+ struct vmci_ctx *context,
+ vmci_event_release_cb wakeup_cb,
+ void *client_data,
+ struct qp_broker_entry **ent,
+ bool *swap)
+{
+ const u32 context_id = vmci_ctx_get_id(context);
+ bool create;
+ struct qp_broker_entry *entry = NULL;
+ bool is_local = flags & VMCI_QPFLAG_LOCAL;
+ int result;
+
+ if (VMCI_HANDLE_INVALID(handle) ||
+ (flags & ~VMCI_QP_ALL_FLAGS) || is_local ||
+ !(produce_size || consume_size) ||
+ !context || context_id == VMCI_INVALID_ID ||
+ handle.context == VMCI_INVALID_ID) {
+ return VMCI_ERROR_INVALID_ARGS;
+ }
+
+ if (page_store && !VMCI_QP_PAGESTORE_IS_WELLFORMED(page_store))
+ return VMCI_ERROR_INVALID_ARGS;
+
+ /*
+ * In the initial argument check, we ensure that non-vmkernel hosts
+ * are not allowed to create local queue pairs.
+ */
+
+ ASSERT(!is_local);
+
+ mutex_lock(&qp_broker_list.mutex);
+
+ if (!is_local && vmci_ctx_qp_exists(context, handle)) {
+ pr_devel("Context (ID=0x%x) already attached to queue "
+ "pair (handle=0x%x:0x%x).", context_id,
+ handle.context, handle.resource);
+ mutex_unlock(&qp_broker_list.mutex);
+ return VMCI_ERROR_ALREADY_EXISTS;
+ }
+
+ if (handle.resource != VMCI_INVALID_ID)
+ entry = qp_broker_handle_to_entry(handle);
+
+ if (!entry) {
+ create = true;
+ result =
+ qp_broker_create(handle, peer, flags, priv_flags,
+ produce_size, consume_size, page_store,
+ context, wakeup_cb, client_data, ent);
+ } else {
+ create = false;
+ result =
+ qp_broker_attach(entry, peer, flags, priv_flags,
+ produce_size, consume_size, page_store,
+ context, wakeup_cb, client_data, ent);
+ }
+
+ mutex_unlock(&qp_broker_list.mutex);
+
+ if (swap)
+ *swap = (context_id == VMCI_HOST_CONTEXT_ID) &&
+ !(create && is_local);
+
+ return result;
+}
+
+/*
+ * This function implements the kernel API for allocating a queue
+ * pair.
+ */
+static int qp_alloc_host_work(struct vmci_handle *handle,
+ struct vmci_queue **produce_q,
+ u64 produce_size,
+ struct vmci_queue **consume_q,
+ u64 consume_size,
+ u32 peer,
+ u32 flags,
+ u32 priv_flags,
+ vmci_event_release_cb wakeup_cb,
+ void *client_data)
+{
+ struct vmci_ctx *context;
+ struct qp_broker_entry *entry;
+ int result;
+ bool swap;
+
+ if (VMCI_HANDLE_INVALID(*handle)) {
+ handle->context = VMCI_HOST_CONTEXT_ID;
+ *handle = vmci_make_handle(handle->context, VMCI_INVALID_ID);
+ }
+
+ context = vmci_ctx_get(VMCI_HOST_CONTEXT_ID);
+ entry = NULL;
+ result =
+ qp_broker_alloc(*handle, peer, flags, priv_flags,
+ produce_size, consume_size, NULL, context,
+ wakeup_cb, client_data, &entry, &swap);
+ if (result == VMCI_SUCCESS) {
+ if (swap) {
+ /*
+ * If this is a local queue pair, the attacher
+ * will swap around produce and consume
+ * queues.
+ */
+
+ *produce_q = entry->consume_q;
+ *consume_q = entry->produce_q;
+ } else {
+ *produce_q = entry->produce_q;
+ *consume_q = entry->consume_q;
+ }
+ } else {
+ *handle = VMCI_INVALID_HANDLE;
+ pr_devel("queue pair broker failed to alloc (result=%d).",
+ result);
+ }
+ vmci_ctx_put(context);
+ return result;
+}
+
+/*
+ * Allocates a VMCI queue_pair. Only checks validity of input
+ * arguments. The real work is done in the host or guest
+ * specific function.
+ */
+int vmci_qp_alloc(struct vmci_handle *handle,
+ struct vmci_queue **produce_q,
+ u64 produce_size,
+ struct vmci_queue **consume_q,
+ u64 consume_size,
+ u32 peer,
+ u32 flags,
+ u32 priv_flags,
+ bool guest_endpoint,
+ vmci_event_release_cb wakeup_cb,
+ void *client_data)
+{
+ if (!handle || !produce_q || !consume_q ||
+ (!produce_size && !consume_size) || (flags & ~VMCI_QP_ALL_FLAGS))
+ return VMCI_ERROR_INVALID_ARGS;
+
+ if (guest_endpoint) {
+ return qp_alloc_guest_work(handle, produce_q,
+ produce_size, consume_q,
+ consume_size, peer,
+ flags, priv_flags);
+ } else {
+ return qp_alloc_host_work(handle, produce_q,
+ produce_size, consume_q,
+ consume_size, peer, flags,
+ priv_flags, wakeup_cb, client_data);
+ }
+}
+
+/*
+ * This function implements the host kernel API for detaching from
+ * a queue pair.
+ */
+static int qp_detatch_host_work(struct vmci_handle handle)
+{
+ int result;
+ struct vmci_ctx *context;
+
+ context = vmci_ctx_get(VMCI_HOST_CONTEXT_ID);
+
+ result = vmci_qp_broker_detach(handle, context);
+
+ vmci_ctx_put(context);
+ return result;
+}
+
+/*
+ * Detaches from a VMCI queue_pair. Only checks validity of input argument.
+ * Real work is done in the host or guest specific function.
+ */
+static int qp_detatch(struct vmci_handle handle, bool guest_endpoint)
+{
+ if (VMCI_HANDLE_INVALID(handle))
+ return VMCI_ERROR_INVALID_ARGS;
+
+ if (guest_endpoint)
+ return qp_detatch_guest_work(handle);
+ else
+ return qp_detatch_host_work(handle);
+}
+
+/*
+ * Returns the entry from the head of the list. Assumes that the list is
+ * locked.
+ */
+static struct qp_entry *qp_list_get_head(struct qp_list *qp_list)
+{
+ if (!list_empty(&qp_list->head)) {
+ struct qp_entry *entry =
+ list_first_entry(&qp_list->head, struct qp_entry,
+ list_item);
+ return entry;
+ }
+
+ return NULL;
+}
+
+void vmci_qp_broker_exit(void)
+{
+ struct qp_entry *entry;
+ struct qp_broker_entry *be;
+
+ mutex_lock(&qp_broker_list.mutex);
+
+ while ((entry = qp_list_get_head(&qp_broker_list))) {
+ be = (struct qp_broker_entry *)entry;
+
+ qp_list_remove_entry(&qp_broker_list, entry);
+ kfree(be);
+ }
+
+ mutex_unlock(&qp_broker_list.mutex);
+}
+
+/*
+ * Requests that a queue pair be allocated with the VMCI queue
+ * pair broker. Allocates a queue pair entry if one does not
+ * exist. Attaches to one if it exists, and retrieves the page
+ * files backing that queue_pair. Assumes that the queue pair
+ * broker lock is held.
+ */
+int vmci_qp_broker_alloc(struct vmci_handle handle,
+ u32 peer,
+ u32 flags,
+ u32 priv_flags,
+ u64 produce_size,
+ u64 consume_size,
+ struct vmci_qp_page_store *page_store,
+ struct vmci_ctx *context)
+{
+ return qp_broker_alloc(handle, peer, flags, priv_flags,
+ produce_size, consume_size,
+ page_store, context, NULL, NULL, NULL, NULL);
+}
+
+/*
+ * VMX'en with versions lower than VMCI_VERSION_NOVMVM use a separate
+ * step to add the UVAs of the VMX mapping of the queue pair. This function
+ * provides backwards compatibility with such VMX'en, and takes care of
+ * registering the page store for a queue pair previously allocated by the
+ * VMX during create or attach. This function will move the queue pair state
+ * to either from VMCIQBP_CREATED_NO_MEM to VMCIQBP_CREATED_MEM or
+ * VMCIQBP_ATTACHED_NO_MEM to VMCIQBP_ATTACHED_MEM. If moving to the
+ * attached state with memory, the queue pair is ready to be used by the
+ * host peer, and an attached event will be generated.
+ *
+ * Assumes that the queue pair broker lock is held.
+ *
+ * This function is only used by the hosted platform, since there is no
+ * issue with backwards compatibility for vmkernel.
+ */
+int vmci_qp_broker_set_page_store(struct vmci_handle handle,
+ u64 produce_uva,
+ u64 consume_uva,
+ struct vmci_ctx *context)
+{
+ struct qp_broker_entry *entry;
+ int result;
+ const u32 context_id = vmci_ctx_get_id(context);
+
+ if (VMCI_HANDLE_INVALID(handle) || !context ||
+ context_id == VMCI_INVALID_ID)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ /*
+ * We only support guest to host queue pairs, so the VMX must
+ * supply UVAs for the mapped page files.
+ */
+
+ if (produce_uva == 0 || consume_uva == 0)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ mutex_lock(&qp_broker_list.mutex);
+
+ if (!vmci_ctx_qp_exists(context, handle)) {
+ pr_warn("Context (ID=0x%x) not attached to queue pair "
+ "(handle=0x%x:0x%x).", context_id, handle.context,
+ handle.resource);
+ result = VMCI_ERROR_NOT_FOUND;
+ goto out;
+ }
+
+ entry = qp_broker_handle_to_entry(handle);
+ if (!entry) {
+ result = VMCI_ERROR_NOT_FOUND;
+ goto out;
+ }
+
+ /*
+ * If I'm the owner then I can set the page store.
+ *
+ * Or, if a host created the queue_pair and I'm the attached peer
+ * then I can set the page store.
+ */
+ if (entry->create_id != context_id &&
+ (entry->create_id != VMCI_HOST_CONTEXT_ID ||
+ entry->attach_id != context_id)) {
+ result = VMCI_ERROR_QUEUEPAIR_NOTOWNER;
+ goto out;
+ }
+
+ if (entry->state != VMCIQPB_CREATED_NO_MEM &&
+ entry->state != VMCIQPB_ATTACHED_NO_MEM) {
+ result = VMCI_ERROR_UNAVAILABLE;
+ goto out;
+ }
+
+ result = qp_host_get_user_memory(produce_uva, consume_uva,
+ entry->produce_q, entry->consume_q);
+ if (result < VMCI_SUCCESS)
+ goto out;
+
+ result = qp_host_map_queues(entry->produce_q, entry->consume_q);
+ if (result < VMCI_SUCCESS) {
+ qp_host_unregister_user_memory(entry->produce_q,
+ entry->consume_q);
+ goto out;
+ }
+
+ if (entry->state == VMCIQPB_CREATED_NO_MEM) {
+ entry->state = VMCIQPB_CREATED_MEM;
+ } else {
+ ASSERT(entry->state == VMCIQPB_ATTACHED_NO_MEM);
+ entry->state = VMCIQPB_ATTACHED_MEM;
+ }
+ entry->vmci_page_files = true;
+
+ if (entry->state == VMCIQPB_ATTACHED_MEM) {
+ result =
+ qp_notify_peer(true, handle, context_id, entry->create_id);
+ if (result < VMCI_SUCCESS) {
+ pr_warn("Failed to notify peer (ID=0x%x) of "
+ "attach to queue pair (handle=0x%x:0x%x).",
+ entry->create_id, entry->qp.handle.context,
+ entry->qp.handle.resource);
+ }
+ }
+
+ result = VMCI_SUCCESS;
+ out:
+ mutex_unlock(&qp_broker_list.mutex);
+ return result;
+}
+
+/*
+ * Resets saved queue headers for the given QP broker
+ * entry. Should be used when guest memory becomes available
+ * again, or the guest detaches.
+ */
+static void qp_reset_saved_headers(struct qp_broker_entry *entry)
+{
+ entry->produce_q->saved_header = NULL;
+ entry->consume_q->saved_header = NULL;
+}
+
+/*
+ * The main entry point for detaching from a queue pair registered with the
+ * queue pair broker. If more than one endpoint is attached to the queue
+ * pair, the first endpoint will mainly decrement a reference count and
+ * generate a notification to its peer. The last endpoint will clean up
+ * the queue pair state registered with the broker.
+ *
+ * When a guest endpoint detaches, it will unmap and unregister the guest
+ * memory backing the queue pair. If the host is still attached, it will
+ * no longer be able to access the queue pair content.
+ *
+ * If the queue pair is already in a state where there is no memory
+ * registered for the queue pair (any *_NO_MEM state), it will transition to
+ * the VMCIQPB_SHUTDOWN_NO_MEM state. This will also happen, if a guest
+ * endpoint is the first of two endpoints to detach. If the host endpoint is
+ * the first out of two to detach, the queue pair will move to the
+ * VMCIQPB_SHUTDOWN_MEM state.
+ */
+int vmci_qp_broker_detach(struct vmci_handle handle, struct vmci_ctx *context)
+{
+ struct qp_broker_entry *entry;
+ const u32 context_id = vmci_ctx_get_id(context);
+ u32 peer_id;
+ bool is_local = false;
+ int result;
+
+ if (VMCI_HANDLE_INVALID(handle) || !context ||
+ context_id == VMCI_INVALID_ID) {
+ return VMCI_ERROR_INVALID_ARGS;
+ }
+
+ mutex_lock(&qp_broker_list.mutex);
+
+ if (!vmci_ctx_qp_exists(context, handle)) {
+ pr_devel("Context (ID=0x%x) not attached to queue pair "
+ "(handle=0x%x:0x%x).", context_id, handle.context,
+ handle.resource);
+ result = VMCI_ERROR_NOT_FOUND;
+ goto out;
+ }
+
+ entry = qp_broker_handle_to_entry(handle);
+ if (!entry) {
+ pr_devel("Context (ID=0x%x) reports being attached to "
+ "queue pair(handle=0x%x:0x%x) that isn't present "
+ "in broker.", context_id, handle.context,
+ handle.resource);
+ result = VMCI_ERROR_NOT_FOUND;
+ goto out;
+ }
+
+ if (context_id != entry->create_id && context_id != entry->attach_id) {
+ result = VMCI_ERROR_QUEUEPAIR_NOTATTACHED;
+ goto out;
+ }
+
+ if (context_id == entry->create_id) {
+ peer_id = entry->attach_id;
+ entry->create_id = VMCI_INVALID_ID;
+ } else {
+ peer_id = entry->create_id;
+ entry->attach_id = VMCI_INVALID_ID;
+ }
+ entry->qp.ref_count--;
+
+ is_local = entry->qp.flags & VMCI_QPFLAG_LOCAL;
+
+ if (context_id != VMCI_HOST_CONTEXT_ID) {
+ bool headers_mapped;
+
+ ASSERT(!is_local);
+
+ /*
+ * Pre NOVMVM vmx'en may detach from a queue pair
+ * before setting the page store, and in that case
+ * there is no user memory to detach from. Also, more
+ * recent VMX'en may detach from a queue pair in the
+ * quiesced state.
+ */
+
+ qp_acquire_queue_mutex(entry->produce_q);
+ headers_mapped = entry->produce_q->q_header ||
+ entry->consume_q->q_header;
+ if (QPBROKERSTATE_HAS_MEM(entry)) {
+ result =
+ qp_host_unmap_queues(INVALID_VMCI_GUEST_MEM_ID,
+ entry->produce_q,
+ entry->consume_q);
+ if (result < VMCI_SUCCESS)
+ pr_warn("Failed to unmap queue headers "
+ "for queue pair "
+ "(handle=0x%x:0x%x,result=%d).",
+ handle.context, handle.resource,
+ result);
+
+ if (entry->vmci_page_files)
+ qp_host_unregister_user_memory(entry->produce_q,
+ entry->
+ consume_q);
+ else
+ qp_host_unregister_user_memory(entry->produce_q,
+ entry->
+ consume_q);
+
+ }
+
+ if (!headers_mapped)
+ qp_reset_saved_headers(entry);
+
+ qp_release_queue_mutex(entry->produce_q);
+
+ if (!headers_mapped && entry->wakeup_cb)
+ entry->wakeup_cb(entry->client_data);
+
+ } else {
+ if (entry->wakeup_cb) {
+ entry->wakeup_cb = NULL;
+ entry->client_data = NULL;
+ }
+ }
+
+ if (entry->qp.ref_count == 0) {
+ qp_list_remove_entry(&qp_broker_list, &entry->qp);
+
+ if (is_local)
+ kfree(entry->local_mem);
+
+ qp_cleanup_queue_mutex(entry->produce_q, entry->consume_q);
+ qp_host_free_queue(entry->produce_q, entry->qp.produce_size);
+ qp_host_free_queue(entry->consume_q, entry->qp.consume_size);
+ /* Unlink from resource hash table and free callback */
+ vmci_resource_remove(&entry->resource);
+
+ kfree(entry);
+
+ vmci_ctx_qp_destroy(context, handle);
+ } else {
+ ASSERT(peer_id != VMCI_INVALID_ID);
+ qp_notify_peer(false, handle, context_id, peer_id);
+ if (context_id == VMCI_HOST_CONTEXT_ID &&
+ QPBROKERSTATE_HAS_MEM(entry)) {
+ entry->state = VMCIQPB_SHUTDOWN_MEM;
+ } else {
+ entry->state = VMCIQPB_SHUTDOWN_NO_MEM;
+ }
+
+ if (!is_local)
+ vmci_ctx_qp_destroy(context, handle);
+
+ }
+ result = VMCI_SUCCESS;
+ out:
+ mutex_unlock(&qp_broker_list.mutex);
+ return result;
+}
+
+/*
+ * Establishes the necessary mappings for a queue pair given a
+ * reference to the queue pair guest memory. This is usually
+ * called when a guest is unquiesced and the VMX is allowed to
+ * map guest memory once again.
+ */
+int vmci_qp_broker_map(struct vmci_handle handle,
+ struct vmci_ctx *context,
+ u64 guest_mem)
+{
+ struct qp_broker_entry *entry;
+ const u32 context_id = vmci_ctx_get_id(context);
+ bool is_local = false;
+ int result;
+
+ if (VMCI_HANDLE_INVALID(handle) || !context ||
+ context_id == VMCI_INVALID_ID)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ mutex_lock(&qp_broker_list.mutex);
+
+ if (!vmci_ctx_qp_exists(context, handle)) {
+ pr_devel("Context (ID=0x%x) not attached to queue pair "
+ "(handle=0x%x:0x%x).", context_id, handle.context,
+ handle.resource);
+ result = VMCI_ERROR_NOT_FOUND;
+ goto out;
+ }
+
+ entry = qp_broker_handle_to_entry(handle);
+ if (!entry) {
+ pr_devel("Context (ID=0x%x) reports being attached to "
+ "queue pair (handle=0x%x:0x%x) that isn't present "
+ "in broker.", context_id, handle.context,
+ handle.resource);
+ result = VMCI_ERROR_NOT_FOUND;
+ goto out;
+ }
+
+ if (context_id != entry->create_id && context_id != entry->attach_id) {
+ result = VMCI_ERROR_QUEUEPAIR_NOTATTACHED;
+ goto out;
+ }
+
+ is_local = entry->qp.flags & VMCI_QPFLAG_LOCAL;
+ result = VMCI_SUCCESS;
+
+ if (context_id != VMCI_HOST_CONTEXT_ID) {
+ struct vmci_qp_page_store page_store;
+
+ ASSERT(entry->state == VMCIQPB_CREATED_NO_MEM ||
+ entry->state == VMCIQPB_SHUTDOWN_NO_MEM ||
+ entry->state == VMCIQPB_ATTACHED_NO_MEM);
+ ASSERT(!is_local);
+
+ page_store.pages = guest_mem;
+ page_store.len = QPE_NUM_PAGES(entry->qp);
+
+ qp_acquire_queue_mutex(entry->produce_q);
+ qp_reset_saved_headers(entry);
+ result =
+ qp_host_register_user_memory(&page_store,
+ entry->produce_q,
+ entry->consume_q);
+ qp_release_queue_mutex(entry->produce_q);
+ if (result == VMCI_SUCCESS) {
+ /* Move state from *_NO_MEM to *_MEM */
+
+ entry->state++;
+
+ ASSERT(entry->state == VMCIQPB_CREATED_MEM ||
+ entry->state == VMCIQPB_SHUTDOWN_MEM ||
+ entry->state == VMCIQPB_ATTACHED_MEM);
+
+ if (entry->wakeup_cb)
+ entry->wakeup_cb(entry->client_data);
+ }
+ }
+
+ out:
+ mutex_unlock(&qp_broker_list.mutex);
+ return result;
+}
+
+/*
+ * Saves a snapshot of the queue headers for the given QP broker
+ * entry. Should be used when guest memory is unmapped.
+ * Results:
+ * VMCI_SUCCESS on success, appropriate error code if guest memory
+ * can't be accessed..
+ */
+static int qp_save_headers(struct qp_broker_entry *entry)
+{
+ int result;
+
+ if (entry->produce_q->saved_header != NULL &&
+ entry->consume_q->saved_header != NULL) {
+ /*
+ * If the headers have already been saved, we don't need to do
+ * it again, and we don't want to map in the headers
+ * unnecessarily.
+ */
+
+ return VMCI_SUCCESS;
+ }
+
+ if (NULL == entry->produce_q->q_header ||
+ NULL == entry->consume_q->q_header) {
+ result = qp_host_map_queues(entry->produce_q, entry->consume_q);
+ if (result < VMCI_SUCCESS)
+ return result;
+ }
+
+ memcpy(&entry->saved_produce_q, entry->produce_q->q_header,
+ sizeof(entry->saved_produce_q));
+ entry->produce_q->saved_header = &entry->saved_produce_q;
+ memcpy(&entry->saved_consume_q, entry->consume_q->q_header,
+ sizeof(entry->saved_consume_q));
+ entry->consume_q->saved_header = &entry->saved_consume_q;
+
+ return VMCI_SUCCESS;
+}
+
+/*
+ * Removes all references to the guest memory of a given queue pair, and
+ * will move the queue pair from state *_MEM to *_NO_MEM. It is usually
+ * called when a VM is being quiesced where access to guest memory should
+ * avoided.
+ */
+int vmci_qp_broker_unmap(struct vmci_handle handle,
+ struct vmci_ctx *context,
+ u32 gid)
+{
+ struct qp_broker_entry *entry;
+ const u32 context_id = vmci_ctx_get_id(context);
+ bool is_local = false;
+ int result;
+
+ if (VMCI_HANDLE_INVALID(handle) || !context ||
+ context_id == VMCI_INVALID_ID)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ mutex_lock(&qp_broker_list.mutex);
+
+ if (!vmci_ctx_qp_exists(context, handle)) {
+ pr_devel("Context (ID=0x%x) not attached to queue pair "
+ "(handle=0x%x:0x%x).", context_id,
+ handle.context, handle.resource);
+ result = VMCI_ERROR_NOT_FOUND;
+ goto out;
+ }
+
+ entry = qp_broker_handle_to_entry(handle);
+ if (!entry) {
+ pr_devel("Context (ID=0x%x) reports being attached to "
+ "queue pair (handle=0x%x:0x%x) that isn't present "
+ "in broker.", context_id, handle.context,
+ handle.resource);
+ result = VMCI_ERROR_NOT_FOUND;
+ goto out;
+ }
+
+ if (context_id != entry->create_id && context_id != entry->attach_id) {
+ result = VMCI_ERROR_QUEUEPAIR_NOTATTACHED;
+ goto out;
+ }
+
+ is_local = entry->qp.flags & VMCI_QPFLAG_LOCAL;
+
+ if (context_id != VMCI_HOST_CONTEXT_ID) {
+ ASSERT(entry->state != VMCIQPB_CREATED_NO_MEM &&
+ entry->state != VMCIQPB_SHUTDOWN_NO_MEM &&
+ entry->state != VMCIQPB_ATTACHED_NO_MEM);
+ ASSERT(!is_local);
+
+ qp_acquire_queue_mutex(entry->produce_q);
+ result = qp_save_headers(entry);
+ if (result < VMCI_SUCCESS)
+ pr_warn("Failed to save queue headers for "
+ "queue pair (handle=0x%x:0x%x,result=%d).",
+ handle.context, handle.resource, result);
+
+ qp_host_unmap_queues(gid, entry->produce_q, entry->consume_q);
+
+ /*
+ * On hosted, when we unmap queue pairs, the VMX will also
+ * unmap the guest memory, so we invalidate the previously
+ * registered memory. If the queue pair is mapped again at a
+ * later point in time, we will need to reregister the user
+ * memory with a possibly new user VA.
+ */
+ qp_host_unregister_user_memory(entry->produce_q,
+ entry->consume_q);
+
+ /*
+ * Move state from *_MEM to *_NO_MEM.
+ */
+ entry->state--;
+
+ qp_release_queue_mutex(entry->produce_q);
+ }
+
+ result = VMCI_SUCCESS;
+
+ out:
+ mutex_unlock(&qp_broker_list.mutex);
+ return result;
+}
+
+/*
+ * Destroys all guest queue pair endpoints. If active guest queue
+ * pairs still exist, hypercalls to attempt detach from these
+ * queue pairs will be made. Any failure to detach is silently
+ * ignored.
+ */
+void vmci_qp_guest_endpoints_exit(void)
+{
+ struct qp_entry *entry;
+ struct qp_guest_endpoint *ep;
+
+ mutex_lock(&qp_guest_endpoints.mutex);
+
+ while ((entry = qp_list_get_head(&qp_guest_endpoints))) {
+ ep = (struct qp_guest_endpoint *)entry;
+
+ /* Don't make a hypercall for local queue_pairs. */
+ if (!(entry->flags & VMCI_QPFLAG_LOCAL))
+ qp_detatch_hypercall(entry->handle);
+
+ /* We cannot fail the exit, so let's reset ref_count. */
+ entry->ref_count = 0;
+ qp_list_remove_entry(&qp_guest_endpoints, entry);
+
+ qp_guest_endpoint_destroy(ep);
+ }
+
+ mutex_unlock(&qp_guest_endpoints.mutex);
+}
+
+/*
+ * Helper routine that will lock the queue pair before subsequent
+ * operations.
+ * Note: Non-blocking on the host side is currently only implemented in ESX.
+ * Since non-blocking isn't yet implemented on the host personality we
+ * have no reason to acquire a spin lock. So to avoid the use of an
+ * unnecessary lock only acquire the mutex if we can block.
+ * Note: It is assumed that QPFLAG_PINNED implies QPFLAG_NONBLOCK. Therefore
+ * we can use the same locking function for access to both the queue
+ * and the queue headers as it is the same logic. Assert this behvior.
+ */
+static void qp_lock(const struct vmci_qp *qpair)
+{
+ ASSERT(!vmci_qp_pinned(qpair->flags) ||
+ (vmci_qp_pinned(qpair->flags) && !vmci_can_block(qpair->flags)));
+
+ if (vmci_can_block(qpair->flags))
+ qp_acquire_queue_mutex(qpair->produce_q);
+}
+
+/*
+ * Helper routine that unlocks the queue pair after calling
+ * qp_lock. Respects non-blocking and pinning flags.
+ */
+static void qp_unlock(const struct vmci_qp *qpair)
+{
+ if (vmci_can_block(qpair->flags))
+ qp_release_queue_mutex(qpair->produce_q);
+}
+
+/*
+ * The queue headers may not be mapped at all times. If a queue is
+ * currently not mapped, it will be attempted to do so.
+ */
+static int qp_map_queue_headers(struct vmci_queue *produce_q,
+ struct vmci_queue *consume_q,
+ bool can_block)
+{
+ int result;
+
+ if (NULL == produce_q->q_header || NULL == consume_q->q_header) {
+ if (can_block)
+ result = qp_host_map_queues(produce_q, consume_q);
+ else
+ result = VMCI_ERROR_QUEUEPAIR_NOT_READY;
+
+ if (result < VMCI_SUCCESS)
+ return (produce_q->saved_header &&
+ consume_q->saved_header) ?
+ VMCI_ERROR_QUEUEPAIR_NOT_READY :
+ VMCI_ERROR_QUEUEPAIR_NOTATTACHED;
+ }
+
+ return VMCI_SUCCESS;
+}
+
+/*
+ * Helper routine that will retrieve the produce and consume
+ * headers of a given queue pair. If the guest memory of the
+ * queue pair is currently not available, the saved queue headers
+ * will be returned, if these are available.
+ */
+static int qp_get_queue_headers(const struct vmci_qp *qpair,
+ struct vmci_queue_header **produce_q_header,
+ struct vmci_queue_header **consume_q_header)
+{
+ int result;
+
+ result = qp_map_queue_headers(qpair->produce_q, qpair->consume_q,
+ vmci_can_block(qpair->flags));
+ if (result == VMCI_SUCCESS) {
+ *produce_q_header = qpair->produce_q->q_header;
+ *consume_q_header = qpair->consume_q->q_header;
+ } else if (qpair->produce_q->saved_header &&
+ qpair->consume_q->saved_header) {
+ ASSERT(!qpair->guest_endpoint);
+ *produce_q_header = qpair->produce_q->saved_header;
+ *consume_q_header = qpair->consume_q->saved_header;
+ result = VMCI_SUCCESS;
+ }
+
+ return result;
+}
+
+/*
+ * Callback from VMCI queue pair broker indicating that a queue
+ * pair that was previously not ready, now either is ready or
+ * gone forever.
+ */
+static int qp_wakeup_cb(void *client_data)
+{
+ struct vmci_qp *qpair = (struct vmci_qp *)client_data;
+
+ qp_lock(qpair);
+ while (qpair->blocked > 0) {
+ qpair->blocked--;
+ qpair->generation++;
+ wake_up(&qpair->event);
+ }
+ qp_unlock(qpair);
+
+ return VMCI_SUCCESS;
+}
+
+/*
+ * Makes the calling thread wait for the queue pair to become
+ * ready for host side access. Returns true when thread is
+ * woken up after queue pair state change, false otherwise.
+ */
+static bool qp_wait_for_ready_queue(struct vmci_qp *qpair)
+{
+ unsigned int generation;
+
+ if (unlikely(qpair->guest_endpoint))
+ ASSERT(false);
+
+ if (qpair->flags & VMCI_QPFLAG_NONBLOCK)
+ return false;
+
+ qpair->blocked++;
+ generation = qpair->generation;
+ qp_unlock(qpair);
+ wait_event(qpair->event, generation != qpair->generation);
+ qp_lock(qpair);
+
+ return true;
+}
+
+/*
+ * Enqueues a given buffer to the produce queue using the provided
+ * function. As many bytes as possible (space available in the queue)
+ * are enqueued. Assumes the queue->mutex has been acquired. Returns
+ * VMCI_ERROR_QUEUEPAIR_NOSPACE if no space was available to enqueue
+ * data, VMCI_ERROR_INVALID_SIZE, if any queue pointer is outside the
+ * queue (as defined by the queue size), VMCI_ERROR_INVALID_ARGS, if
+ * an error occured when accessing the buffer,
+ * VMCI_ERROR_QUEUEPAIR_NOTATTACHED, if the queue pair pages aren't
+ * available. Otherwise, the number of bytes written to the queue is
+ * returned. Updates the tail pointer of the produce queue.
+ */
+static ssize_t qp_enqueue_locked(struct vmci_queue *produce_q,
+ struct vmci_queue *consume_q,
+ const u64 produce_q_size,
+ const void *buf,
+ size_t buf_size,
+ vmci_memcpy_to_queue_func memcpy_to_queue,
+ bool can_block)
+{
+ s64 free_space;
+ u64 tail;
+ size_t written;
+ ssize_t result;
+
+ result = qp_map_queue_headers(produce_q, consume_q, can_block);
+ if (unlikely(result != VMCI_SUCCESS))
+ return result;
+
+ free_space = vmci_q_header_free_space(produce_q->q_header,
+ consume_q->q_header,
+ produce_q_size);
+ if (free_space == 0)
+ return VMCI_ERROR_QUEUEPAIR_NOSPACE;
+
+ if (free_space < VMCI_SUCCESS)
+ return (ssize_t) free_space;
+
+ written = (size_t) (free_space > buf_size ? buf_size : free_space);
+ tail = vmci_q_header_producer_tail(produce_q->q_header);
+ if (likely(tail + written < produce_q_size)) {
+ result = memcpy_to_queue(produce_q, tail, buf, 0, written);
+ } else {
+ /* Tail pointer wraps around. */
+
+ const size_t tmp = (size_t) (produce_q_size - tail);
+
+ result = memcpy_to_queue(produce_q, tail, buf, 0, tmp);
+ if (result >= VMCI_SUCCESS)
+ result = memcpy_to_queue(produce_q, 0, buf, tmp,
+ written - tmp);
+ }
+
+ if (result < VMCI_SUCCESS)
+ return result;
+
+ vmci_q_header_add_producer_tail(produce_q->q_header, written,
+ produce_q_size);
+ return written;
+}
+
+/*
+ * Dequeues data (if available) from the given consume queue. Writes data
+ * to the user provided buffer using the provided function.
+ * Assumes the queue->mutex has been acquired.
+ * Results:
+ * VMCI_ERROR_QUEUEPAIR_NODATA if no data was available to dequeue.
+ * VMCI_ERROR_INVALID_SIZE, if any queue pointer is outside the queue
+ * (as defined by the queue size).
+ * VMCI_ERROR_INVALID_ARGS, if an error occured when accessing the buffer.
+ * Otherwise the number of bytes dequeued is returned.
+ * Side effects:
+ * Updates the head pointer of the consume queue.
+ */
+static ssize_t qp_dequeue_locked(struct vmci_queue *produce_q,
+ struct vmci_queue *consume_q,
+ const u64 consume_q_size,
+ void *buf,
+ size_t buf_size,
+ vmci_memcpy_from_queue_func memcpy_from_queue,
+ bool update_consumer,
+ bool can_block)
+{
+ s64 buf_ready;
+ u64 head;
+ size_t read;
+ ssize_t result;
+
+ result = qp_map_queue_headers(produce_q, consume_q, can_block);
+ if (unlikely(result != VMCI_SUCCESS))
+ return result;
+
+ buf_ready = vmci_q_header_buf_ready(consume_q->q_header,
+ produce_q->q_header,
+ consume_q_size);
+ if (buf_ready == 0)
+ return VMCI_ERROR_QUEUEPAIR_NODATA;
+
+ if (buf_ready < VMCI_SUCCESS)
+ return (ssize_t) buf_ready;
+
+ read = (size_t) (buf_ready > buf_size ? buf_size : buf_ready);
+ head = vmci_q_header_consumer_head(produce_q->q_header);
+ if (likely(head + read < consume_q_size)) {
+ result = memcpy_from_queue(buf, 0, consume_q, head, read);
+ } else {
+ /* Head pointer wraps around. */
+
+ const size_t tmp = (size_t) (consume_q_size - head);
+
+ result = memcpy_from_queue(buf, 0, consume_q, head, tmp);
+ if (result >= VMCI_SUCCESS)
+ result = memcpy_from_queue(buf, tmp, consume_q, 0,
+ read - tmp);
+
+ }
+
+ if (result < VMCI_SUCCESS)
+ return result;
+
+ if (update_consumer)
+ vmci_q_header_add_consumer_head(produce_q->q_header,
+ read, consume_q_size);
+
+ return read;
+}
+
+/*
+ * vmci_qpair_alloc() - Allocates a queue pair.
+ * @qpair: Pointer for the new vmci_qp struct.
+ * @handle: Handle to track the resource.
+ * @produce_qsize: Desired size of the producer queue.
+ * @consume_qsize: Desired size of the consumer queue.
+ * @peer: ContextID of the peer.
+ * @flags: VMCI flags.
+ * @priv_flags: VMCI priviledge flags.
+ *
+ * This is the client interface for allocating the memory for a
+ * vmci_qp structure and then attaching to the underlying
+ * queue. If an error occurs allocating the memory for the
+ * vmci_qp structure no attempt is made to attach. If an
+ * error occurs attaching, then the structure is freed.
+ */
+int vmci_qpair_alloc(struct vmci_qp **qpair,
+ struct vmci_handle *handle,
+ u64 produce_qsize,
+ u64 consume_qsize,
+ u32 peer,
+ u32 flags,
+ u32 priv_flags)
+{
+ struct vmci_qp *my_qpair;
+ int retval;
+ struct vmci_handle src = VMCI_INVALID_HANDLE;
+ struct vmci_handle dst = vmci_make_handle(peer, VMCI_INVALID_ID);
+ enum vmci_route route;
+ vmci_event_release_cb wakeup_cb;
+ void *client_data;
+
+ /*
+ * Restrict the size of a queuepair. The device already
+ * enforces a limit on the total amount of memory that can be
+ * allocated to queuepairs for a guest. However, we try to
+ * allocate this memory before we make the queuepair
+ * allocation hypercall. On Linux, we allocate each page
+ * separately, which means rather than fail, the guest will
+ * thrash while it tries to allocate, and will become
+ * increasingly unresponsive to the point where it appears to
+ * be hung. So we place a limit on the size of an individual
+ * queuepair here, and leave the device to enforce the
+ * restriction on total queuepair memory. (Note that this
+ * doesn't prevent all cases; a user with only this much
+ * physical memory could still get into trouble.) The error
+ * used by the device is NO_RESOURCES, so use that here too.
+ */
+
+ if (produce_qsize + consume_qsize < max(produce_qsize, consume_qsize) ||
+ produce_qsize + consume_qsize > VMCI_MAX_GUEST_QP_MEMORY)
+ return VMCI_ERROR_NO_RESOURCES;
+
+ retval = vmci_route(&src, &dst, false, &route);
+ if (retval < VMCI_SUCCESS)
+ route = vmci_guest_code_active() ?
+ VMCI_ROUTE_AS_GUEST : VMCI_ROUTE_AS_HOST;
+
+ /* If NONBLOCK or PINNED is set, we better be the guest personality. */
+ if ((!vmci_can_block(flags) || vmci_qp_pinned(flags)) &&
+ VMCI_ROUTE_AS_GUEST != route) {
+ pr_devel("Not guest personality w/ NONBLOCK OR PINNED set");
+ return VMCI_ERROR_INVALID_ARGS;
+ }
+
+ /*
+ * Limit the size of pinned QPs and check sanity.
+ *
+ * Pinned pages implies non-blocking mode. Mutexes aren't acquired
+ * when the NONBLOCK flag is set in qpair code; and also should not be
+ * acquired when the PINNED flagged is set. Since pinning pages
+ * implies we want speed, it makes no sense not to have NONBLOCK
+ * set if PINNED is set. Hence enforce this implication.
+ */
+ if (vmci_qp_pinned(flags)) {
+ if (vmci_can_block(flags)) {
+ pr_err("Attempted to enable pinning w/o non-blocking");
+ return VMCI_ERROR_INVALID_ARGS;
+ }
+
+ if (produce_qsize + consume_qsize > VMCI_MAX_PINNED_QP_MEMORY)
+ return VMCI_ERROR_NO_RESOURCES;
+ }
+
+ my_qpair = kzalloc(sizeof(*my_qpair), GFP_KERNEL);
+ if (!my_qpair)
+ return VMCI_ERROR_NO_MEM;
+
+ my_qpair->produce_q_size = produce_qsize;
+ my_qpair->consume_q_size = consume_qsize;
+ my_qpair->peer = peer;
+ my_qpair->flags = flags;
+ my_qpair->priv_flags = priv_flags;
+
+ wakeup_cb = NULL;
+ client_data = NULL;
+
+ if (VMCI_ROUTE_AS_HOST == route) {
+ my_qpair->guest_endpoint = false;
+ if (!(flags & VMCI_QPFLAG_LOCAL)) {
+ my_qpair->blocked = 0;
+ my_qpair->generation = 0;
+ init_waitqueue_head(&my_qpair->event);
+ wakeup_cb = qp_wakeup_cb;
+ client_data = (void *)my_qpair;
+ }
+ } else {
+ my_qpair->guest_endpoint = true;
+ }
+
+ retval = vmci_qp_alloc(handle,
+ &my_qpair->produce_q,
+ my_qpair->produce_q_size,
+ &my_qpair->consume_q,
+ my_qpair->consume_q_size,
+ my_qpair->peer,
+ my_qpair->flags,
+ my_qpair->priv_flags,
+ my_qpair->guest_endpoint,
+ wakeup_cb, client_data);
+
+ if (retval < VMCI_SUCCESS) {
+ kfree(my_qpair);
+ return retval;
+ }
+
+ *qpair = my_qpair;
+ my_qpair->handle = *handle;
+
+ return retval;
+}
+EXPORT_SYMBOL_GPL(vmci_qpair_alloc);
+
+/*
+ * vmci_qpair_detach() - Detatches the client from a queue pair.
+ * @qpair: Reference of a pointer to the qpair struct.
+ *
+ * This is the client interface for detaching from a VMCIQPair.
+ * Note that this routine will free the memory allocated for the
+ * vmci_qp structure too.
+ */
+int vmci_qpair_detach(struct vmci_qp **qpair)
+{
+ int result;
+ struct vmci_qp *old_qpair;
+
+ if (!qpair || !(*qpair))
+ return VMCI_ERROR_INVALID_ARGS;
+
+ old_qpair = *qpair;
+ result = qp_detatch(old_qpair->handle, old_qpair->guest_endpoint);
+
+ /*
+ * The guest can fail to detach for a number of reasons, and
+ * if it does so, it will cleanup the entry (if there is one).
+ * The host can fail too, but it won't cleanup the entry
+ * immediately, it will do that later when the context is
+ * freed. Either way, we need to release the qpair struct
+ * here; there isn't much the caller can do, and we don't want
+ * to leak.
+ */
+
+ memset(old_qpair, 0, sizeof(*old_qpair));
+ old_qpair->handle = VMCI_INVALID_HANDLE;
+ old_qpair->peer = VMCI_INVALID_ID;
+ kfree(old_qpair);
+ *qpair = NULL;
+
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_qpair_detach);
+
+/*
+ * vmci_qpair_get_produce_indexes() - Retrieves the indexes of the producer.
+ * @qpair: Pointer to the queue pair struct.
+ * @producer_tail: Reference used for storing producer tail index.
+ * @consumer_head: Reference used for storing the consumer head index.
+ *
+ * This is the client interface for getting the current indexes of the
+ * QPair from the point of the view of the caller as the producer.
+ */
+int vmci_qpair_get_produce_indexes(const struct vmci_qp *qpair,
+ u64 *producer_tail,
+ u64 *consumer_head)
+{
+ struct vmci_queue_header *produce_q_header;
+ struct vmci_queue_header *consume_q_header;
+ int result;
+
+ if (!qpair)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ qp_lock(qpair);
+ result =
+ qp_get_queue_headers(qpair, &produce_q_header, &consume_q_header);
+ if (result == VMCI_SUCCESS)
+ vmci_q_header_get_pointers(produce_q_header, consume_q_header,
+ producer_tail, consumer_head);
+ qp_unlock(qpair);
+
+ if (result == VMCI_SUCCESS &&
+ ((producer_tail && *producer_tail >= qpair->produce_q_size) ||
+ (consumer_head && *consumer_head >= qpair->produce_q_size)))
+ return VMCI_ERROR_INVALID_SIZE;
+
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_qpair_get_produce_indexes);
+
+/*
+ * vmci_qpair_get_consume_indexes() - Retrieves the indexes of the comsumer.
+ * @qpair: Pointer to the queue pair struct.
+ * @consumer_tail: Reference used for storing consumer tail index.
+ * @producer_head: Reference used for storing the producer head index.
+ *
+ * This is the client interface for getting the current indexes of the
+ * QPair from the point of the view of the caller as the consumer.
+ */
+int vmci_qpair_get_consume_indexes(const struct vmci_qp *qpair,
+ u64 *consumer_tail,
+ u64 *producer_head)
+{
+ struct vmci_queue_header *produce_q_header;
+ struct vmci_queue_header *consume_q_header;
+ int result;
+
+ if (!qpair)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ qp_lock(qpair);
+ result =
+ qp_get_queue_headers(qpair, &produce_q_header, &consume_q_header);
+ if (result == VMCI_SUCCESS)
+ vmci_q_header_get_pointers(consume_q_header, produce_q_header,
+ consumer_tail, producer_head);
+ qp_unlock(qpair);
+
+ if (result == VMCI_SUCCESS &&
+ ((consumer_tail && *consumer_tail >= qpair->consume_q_size) ||
+ (producer_head && *producer_head >= qpair->consume_q_size)))
+ return VMCI_ERROR_INVALID_SIZE;
+
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_qpair_get_consume_indexes);
+
+/*
+ * vmci_qpair_produce_free_space() - Retrieves free space in producer queue.
+ * @qpair: Pointer to the queue pair struct.
+ *
+ * This is the client interface for getting the amount of free
+ * space in the QPair from the point of the view of the caller as
+ * the producer which is the common case. Returns < 0 if err, else
+ * available bytes into which data can be enqueued if > 0.
+ */
+s64 vmci_qpair_produce_free_space(const struct vmci_qp *qpair)
+{
+ struct vmci_queue_header *produce_q_header;
+ struct vmci_queue_header *consume_q_header;
+ s64 result;
+
+ if (!qpair)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ qp_lock(qpair);
+ result =
+ qp_get_queue_headers(qpair, &produce_q_header, &consume_q_header);
+ if (result == VMCI_SUCCESS)
+ result = vmci_q_header_free_space(produce_q_header,
+ consume_q_header,
+ qpair->produce_q_size);
+ else
+ result = 0;
+
+ qp_unlock(qpair);
+
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_qpair_produce_free_space);
+
+/*
+ * vmci_qpair_consume_free_space() - Retrieves free space in consumer queue.
+ * @qpair: Pointer to the queue pair struct.
+ *
+ * This is the client interface for getting the amount of free
+ * space in the QPair from the point of the view of the caller as
+ * the consumer which is not the common case. Returns < 0 if err, else
+ * available bytes into which data can be enqueued if > 0.
+ */
+s64 vmci_qpair_consume_free_space(const struct vmci_qp *qpair)
+{
+ struct vmci_queue_header *produce_q_header;
+ struct vmci_queue_header *consume_q_header;
+ s64 result;
+
+ if (!qpair)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ qp_lock(qpair);
+ result =
+ qp_get_queue_headers(qpair, &produce_q_header, &consume_q_header);
+ if (result == VMCI_SUCCESS)
+ result = vmci_q_header_free_space(consume_q_header,
+ produce_q_header,
+ qpair->consume_q_size);
+ else
+ result = 0;
+
+ qp_unlock(qpair);
+
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_qpair_consume_free_space);
+
+/*
+ * vmci_qpair_produce_buf_ready() - Gets bytes ready to read from
+ * producer queue.
+ * @qpair: Pointer to the queue pair struct.
+ *
+ * This is the client interface for getting the amount of
+ * enqueued data in the QPair from the point of the view of the
+ * caller as the producer which is not the common case. Returns < 0 if err,
+ * else available bytes that may be read.
+ */
+s64 vmci_qpair_produce_buf_ready(const struct vmci_qp *qpair)
+{
+ struct vmci_queue_header *produce_q_header;
+ struct vmci_queue_header *consume_q_header;
+ s64 result;
+
+ if (!qpair)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ qp_lock(qpair);
+ result =
+ qp_get_queue_headers(qpair, &produce_q_header, &consume_q_header);
+ if (result == VMCI_SUCCESS)
+ result = vmci_q_header_buf_ready(produce_q_header,
+ consume_q_header,
+ qpair->produce_q_size);
+ else
+ result = 0;
+
+ qp_unlock(qpair);
+
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_qpair_produce_buf_ready);
+
+/*
+ * vmci_qpair_consume_buf_ready() - Gets bytes ready to read from
+ * consumer queue.
+ * @qpair: Pointer to the queue pair struct.
+ *
+ * This is the client interface for getting the amount of
+ * enqueued data in the QPair from the point of the view of the
+ * caller as the consumer which is the normal case. Returns < 0 if err,
+ * else available bytes that may be read.
+ */
+s64 vmci_qpair_consume_buf_ready(const struct vmci_qp *qpair)
+{
+ struct vmci_queue_header *produce_q_header;
+ struct vmci_queue_header *consume_q_header;
+ s64 result;
+
+ if (!qpair)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ qp_lock(qpair);
+ result =
+ qp_get_queue_headers(qpair, &produce_q_header, &consume_q_header);
+ if (result == VMCI_SUCCESS)
+ result = vmci_q_header_buf_ready(consume_q_header,
+ produce_q_header,
+ qpair->consume_q_size);
+ else
+ result = 0;
+
+ qp_unlock(qpair);
+
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_qpair_consume_buf_ready);
+
+/*
+ * vmci_qpair_enqueue() - Throw data on the queue.
+ * @qpair: Pointer to the queue pair struct.
+ * @buf: Pointer to buffer containing data
+ * @buf_size: Length of buffer.
+ * @buf_type: Buffer type (Unused).
+ *
+ * This is the client interface for enqueueing data into the queue.
+ * Returns number of bytes enqueued or < 0 on error.
+ */
+ssize_t vmci_qpair_enqueue(struct vmci_qp *qpair,
+ const void *buf,
+ size_t buf_size,
+ int buf_type)
+{
+ ssize_t result;
+
+ if (!qpair || !buf)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ qp_lock(qpair);
+
+ do {
+ result = qp_enqueue_locked(qpair->produce_q,
+ qpair->consume_q,
+ qpair->produce_q_size,
+ buf, buf_size,
+ qp_memcpy_to_queue,
+ vmci_can_block(qpair->flags));
+
+ if (result == VMCI_ERROR_QUEUEPAIR_NOT_READY &&
+ !qp_wait_for_ready_queue(qpair))
+ result = VMCI_ERROR_WOULD_BLOCK;
+
+ } while (result == VMCI_ERROR_QUEUEPAIR_NOT_READY);
+
+ qp_unlock(qpair);
+
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_qpair_enqueue);
+
+/*
+ * vmci_qpair_dequeue() - Get data from the queue.
+ * @qpair: Pointer to the queue pair struct.
+ * @buf: Pointer to buffer for the data
+ * @buf_size: Length of buffer.
+ * @buf_type: Buffer type (Unused).
+ *
+ * This is the client interface for dequeueing data from the queue.
+ * Returns number of bytes dequeued or < 0 on error.
+ */
+ssize_t vmci_qpair_dequeue(struct vmci_qp *qpair,
+ void *buf,
+ size_t buf_size,
+ int buf_type)
+{
+ ssize_t result;
+
+ if (!qpair || !buf)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ qp_lock(qpair);
+
+ do {
+ result = qp_dequeue_locked(qpair->produce_q,
+ qpair->consume_q,
+ qpair->consume_q_size,
+ buf, buf_size,
+ qp_memcpy_from_queue, true,
+ vmci_can_block(qpair->flags));
+
+ if (result == VMCI_ERROR_QUEUEPAIR_NOT_READY &&
+ !qp_wait_for_ready_queue(qpair))
+ result = VMCI_ERROR_WOULD_BLOCK;
+
+ } while (result == VMCI_ERROR_QUEUEPAIR_NOT_READY);
+
+ qp_unlock(qpair);
+
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_qpair_dequeue);
+
+/*
+ * vmci_qpair_peek() - Peek at the data in the queue.
+ * @qpair: Pointer to the queue pair struct.
+ * @buf: Pointer to buffer for the data
+ * @buf_size: Length of buffer.
+ * @buf_type: Buffer type (Unused on Linux).
+ *
+ * This is the client interface for peeking into a queue. (I.e.,
+ * copy data from the queue without updating the head pointer.)
+ * Returns number of bytes dequeued or < 0 on error.
+ */
+ssize_t vmci_qpair_peek(struct vmci_qp *qpair,
+ void *buf,
+ size_t buf_size,
+ int buf_type)
+{
+ ssize_t result;
+
+ if (!qpair || !buf)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ qp_lock(qpair);
+
+ do {
+ result = qp_dequeue_locked(qpair->produce_q,
+ qpair->consume_q,
+ qpair->consume_q_size,
+ buf, buf_size,
+ qp_memcpy_from_queue, false,
+ vmci_can_block(qpair->flags));
+
+ if (result == VMCI_ERROR_QUEUEPAIR_NOT_READY &&
+ !qp_wait_for_ready_queue(qpair))
+ result = VMCI_ERROR_WOULD_BLOCK;
+
+ } while (result == VMCI_ERROR_QUEUEPAIR_NOT_READY);
+
+ qp_unlock(qpair);
+
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_qpair_peek);
+
+/*
+ * vmci_qpair_enquev() - Throw data on the queue using iov.
+ * @qpair: Pointer to the queue pair struct.
+ * @iov: Pointer to buffer containing data
+ * @iov_size: Length of buffer.
+ * @buf_type: Buffer type (Unused).
+ *
+ * This is the client interface for enqueueing data into the queue.
+ * This function uses IO vectors to handle the work. Returns number
+ * of bytes enqueued or < 0 on error.
+ */
+ssize_t vmci_qpair_enquev(struct vmci_qp *qpair,
+ void *iov,
+ size_t iov_size,
+ int buf_type)
+{
+ ssize_t result;
+
+ if (!qpair || !iov)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ qp_lock(qpair);
+
+ do {
+ result = qp_enqueue_locked(qpair->produce_q,
+ qpair->consume_q,
+ qpair->produce_q_size,
+ iov, iov_size,
+ qp_memcpy_to_queue_iov,
+ vmci_can_block(qpair->flags));
+
+ if (result == VMCI_ERROR_QUEUEPAIR_NOT_READY &&
+ !qp_wait_for_ready_queue(qpair))
+ result = VMCI_ERROR_WOULD_BLOCK;
+
+ } while (result == VMCI_ERROR_QUEUEPAIR_NOT_READY);
+
+ qp_unlock(qpair);
+
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_qpair_enquev);
+
+/*
+ * vmci_qpair_dequev() - Get data from the queue using iov.
+ * @qpair: Pointer to the queue pair struct.
+ * @iov: Pointer to buffer for the data
+ * @iov_size: Length of buffer.
+ * @buf_type: Buffer type (Unused).
+ *
+ * This is the client interface for dequeueing data from the queue.
+ * This function uses IO vectors to handle the work. Returns number
+ * of bytes dequeued or < 0 on error.
+ */
+ssize_t vmci_qpair_dequev(struct vmci_qp *qpair,
+ void *iov,
+ size_t iov_size,
+ int buf_type)
+{
+ ssize_t result;
+
+ qp_lock(qpair);
+
+ if (!qpair || !iov)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ do {
+ result = qp_dequeue_locked(qpair->produce_q,
+ qpair->consume_q,
+ qpair->consume_q_size,
+ iov, iov_size,
+ qp_memcpy_from_queue_iov,
+ true, vmci_can_block(qpair->flags));
+
+ if (result == VMCI_ERROR_QUEUEPAIR_NOT_READY &&
+ !qp_wait_for_ready_queue(qpair))
+ result = VMCI_ERROR_WOULD_BLOCK;
+
+ } while (result == VMCI_ERROR_QUEUEPAIR_NOT_READY);
+
+ qp_unlock(qpair);
+
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_qpair_dequev);
+
+/*
+ * vmci_qpair_peekv() - Peek at the data in the queue using iov.
+ * @qpair: Pointer to the queue pair struct.
+ * @iov: Pointer to buffer for the data
+ * @iov_size: Length of buffer.
+ * @buf_type: Buffer type (Unused on Linux).
+ *
+ * This is the client interface for peeking into a queue. (I.e.,
+ * copy data from the queue without updating the head pointer.)
+ * This function uses IO vectors to handle the work. Returns number
+ * of bytes peeked or < 0 on error.
+ */
+ssize_t vmci_qpair_peekv(struct vmci_qp *qpair,
+ void *iov,
+ size_t iov_size,
+ int buf_type)
+{
+ ssize_t result;
+
+ if (!qpair || !iov)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ qp_lock(qpair);
+
+ do {
+ result = qp_dequeue_locked(qpair->produce_q,
+ qpair->consume_q,
+ qpair->consume_q_size,
+ iov, iov_size,
+ qp_memcpy_from_queue_iov,
+ false, vmci_can_block(qpair->flags));
+
+ if (result == VMCI_ERROR_QUEUEPAIR_NOT_READY &&
+ !qp_wait_for_ready_queue(qpair))
+ result = VMCI_ERROR_WOULD_BLOCK;
+
+ } while (result == VMCI_ERROR_QUEUEPAIR_NOT_READY);
+
+ qp_unlock(qpair);
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_qpair_peekv);
diff --git a/drivers/misc/vmw_vmci/vmci_queue_pair.h b/drivers/misc/vmw_vmci/vmci_queue_pair.h
new file mode 100644
index 0000000..8d8d6a1
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_queue_pair.h
@@ -0,0 +1,191 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#ifndef _VMCI_QUEUE_PAIR_H_
+#define _VMCI_QUEUE_PAIR_H_
+
+#include <linux/vmw_vmci_defs.h>
+#include <linux/types.h>
+
+#include "vmci_context.h"
+
+/* Callback needed for correctly waiting on events. */
+typedef int (*vmci_event_release_cb) (void *client_data);
+
+/* Guest device port I/O. */
+struct PPNSet {
+ u64 num_produce_pages;
+ u64 num_consume_pages;
+ u32 *produce_ppns;
+ u32 *consume_ppns;
+ bool initialized;
+};
+
+/* VMCIqueue_pairAllocInfo */
+struct vmci_qp_alloc_info {
+ struct vmci_handle handle;
+ u32 peer;
+ u32 flags;
+ u64 produce_size;
+ u64 consume_size;
+ u64 ppn_va; /* Start VA of queue pair PPNs. */
+ u64 num_ppns;
+ s32 result;
+ u32 version;
+};
+
+/* VMCIqueue_pairSetVAInfo */
+struct vmci_qp_set_va_info {
+ struct vmci_handle handle;
+ u64 va; /* Start VA of queue pair PPNs. */
+ u64 num_ppns;
+ u32 version;
+ s32 result;
+};
+
+/*
+ * For backwards compatibility, here is a version of the
+ * VMCIqueue_pairPageFileInfo before host support end-points was added.
+ * Note that the current version of that structure requires VMX to
+ * pass down the VA of the mapped file. Before host support was added
+ * there was nothing of the sort. So, when the driver sees the ioctl
+ * with a parameter that is the sizeof
+ * VMCIqueue_pairPageFileInfo_NoHostQP then it can infer that the version
+ * of VMX running can't attach to host end points because it doesn't
+ * provide the VA of the mapped files.
+ *
+ * The Linux driver doesn't get an indication of the size of the
+ * structure passed down from user space. So, to fix a long standing
+ * but unfiled bug, the _pad field has been renamed to version.
+ * Existing versions of VMX always initialize the PageFileInfo
+ * structure so that _pad, er, version is set to 0.
+ *
+ * A version value of 1 indicates that the size of the structure has
+ * been increased to include two UVA's: produce_uva and consume_uva.
+ * These UVA's are of the mmap()'d queue contents backing files.
+ *
+ * In addition, if when VMX is sending down the
+ * VMCIqueue_pairPageFileInfo structure it gets an error then it will
+ * try again with the _NoHostQP version of the file to see if an older
+ * VMCI kernel module is running.
+ */
+
+/* VMCIqueue_pairPageFileInfo */
+struct vmci_qp_page_file_info {
+ struct vmci_handle handle;
+ u64 produce_page_file; /* User VA. */
+ u64 consume_page_file; /* User VA. */
+ u64 produce_page_file_size; /* Size of the file name array. */
+ u64 consume_page_file_size; /* Size of the file name array. */
+ s32 result;
+ u32 version; /* Was _pad. */
+ u64 produce_va; /* User VA of the mapped file. */
+ u64 consume_va; /* User VA of the mapped file. */
+};
+
+/* vmci queuepair detach info */
+struct vmci_qp_dtch_info {
+ struct vmci_handle handle;
+ s32 result;
+ u32 _pad;
+};
+
+/*
+ * struct vmci_qp_page_store describes how the memory of a given queue pair
+ * is backed. When the queue pair is between the host and a guest, the
+ * page store consists of references to the guest pages. On vmkernel,
+ * this is a list of PPNs, and on hosted, it is a user VA where the
+ * queue pair is mapped into the VMX address space.
+ */
+struct vmci_qp_page_store {
+ /* Reference to pages backing the queue pair. */
+ u64 pages;
+ /* Length of pageList/virtual addres range (in pages). */
+ u32 len;
+};
+
+/*
+ * This data type contains the information about a queue.
+ * There are two queues (hence, queue pairs) per transaction model between a
+ * pair of end points, A & B. One queue is used by end point A to transmit
+ * commands and responses to B. The other queue is used by B to transmit
+ * commands and responses.
+ *
+ * struct vmci_queue_kern_if is a per-OS defined Queue structure. It contains
+ * either a direct pointer to the linear address of the buffer contents or a
+ * pointer to structures which help the OS locate those data pages. See
+ * vmciKernelIf.c for each platform for its definition.
+ */
+struct vmci_queue {
+ struct vmci_queue_header *q_header;
+ struct vmci_queue_header *saved_header;
+ struct vmci_queue_kern_if *kernel_if;
+};
+
+/*
+ * Utility function that checks whether the fields of the page
+ * store contain valid values.
+ * Result:
+ * true if the page store is wellformed. false otherwise.
+ */
+static inline bool
+VMCI_QP_PAGESTORE_IS_WELLFORMED(struct vmci_qp_page_store *page_store)
+{
+ return page_store->len >= 2;
+}
+
+/*
+ * Helper function to check if the non-blocking flag
+ * is set for a given queue pair.
+ */
+static inline bool vmci_can_block(u32 flags)
+{
+ return !(flags & VMCI_QPFLAG_NONBLOCK);
+}
+
+/*
+ * Helper function to check if the queue pair is pinned
+ * into memory.
+ */
+static inline bool vmci_qp_pinned(u32 flags)
+{
+ return flags & VMCI_QPFLAG_PINNED;
+}
+
+void vmci_qp_broker_exit(void);
+int vmci_qp_broker_alloc(struct vmci_handle handle, u32 peer,
+ u32 flags, u32 priv_flags,
+ u64 produce_size, u64 consume_size,
+ struct vmci_qp_page_store *page_store,
+ struct vmci_ctx *context);
+int vmci_qp_broker_set_page_store(struct vmci_handle handle,
+ u64 produce_uva, u64 consume_uva,
+ struct vmci_ctx *context);
+int vmci_qp_broker_detach(struct vmci_handle handle, struct vmci_ctx *context);
+
+void vmci_qp_guest_endpoints_exit(void);
+
+int vmci_qp_alloc(struct vmci_handle *handle,
+ struct vmci_queue **produce_q, u64 produce_size,
+ struct vmci_queue **consume_q, u64 consume_size,
+ u32 peer, u32 flags, u32 priv_flags,
+ bool guest_endpoint, vmci_event_release_cb wakeup_cb,
+ void *client_data);
+int vmci_qp_broker_map(struct vmci_handle handle,
+ struct vmci_ctx *context, u64 guest_mem);
+int vmci_qp_broker_unmap(struct vmci_handle handle,
+ struct vmci_ctx *context, u32 gid);
+
+#endif /* _VMCI_QUEUE_PAIR_H_ */
^ permalink raw reply related
* [PATCH 08/12] VMCI: resource object implementation.
From: George Zhang @ 2012-11-07 18:42 UTC (permalink / raw)
To: linux-kernel, georgezhang, virtualization; +Cc: pv-drivers, gregkh
In-Reply-To: <20121107183624.9658.78903.stgit@promb-2n-dhcp175.eng.vmware.com>
VMCI resource tracks all used resources within the vmci code.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
drivers/misc/vmw_vmci/vmci_resource.c | 232 +++++++++++++++++++++++++++++++++
drivers/misc/vmw_vmci/vmci_resource.h | 59 ++++++++
2 files changed, 291 insertions(+), 0 deletions(-)
create mode 100644 drivers/misc/vmw_vmci/vmci_resource.c
create mode 100644 drivers/misc/vmw_vmci/vmci_resource.h
diff --git a/drivers/misc/vmw_vmci/vmci_resource.c b/drivers/misc/vmw_vmci/vmci_resource.c
new file mode 100644
index 0000000..b241dde
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_resource.c
@@ -0,0 +1,232 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#include <linux/vmw_vmci_defs.h>
+#include <linux/hash.h>
+#include <linux/types.h>
+#include <linux/rculist.h>
+
+#include "vmci_common_int.h"
+#include "vmci_resource.h"
+#include "vmci_driver.h"
+
+
+#define VMCI_RESOURCE_HASH_BITS 7
+#define VMCI_RESOURCE_HASH_BUCKETS (1 << VMCI_RESOURCE_HASH_BITS)
+
+struct vmci_hash_table {
+ spinlock_t lock;
+ struct hlist_head entries[VMCI_RESOURCE_HASH_BUCKETS];
+};
+
+static struct vmci_hash_table vmci_resource_table = {
+ .lock = __SPIN_LOCK_UNLOCKED(vmci_resource_table.lock),
+};
+
+static unsigned int vmci_resource_hash(struct vmci_handle handle)
+{
+ return hash_32(VMCI_HANDLE_TO_RESOURCE_ID(handle),
+ VMCI_RESOURCE_HASH_BITS);
+}
+
+/*
+ * Gets a resource (if one exists) matching given handle from the hash table.
+ */
+static struct vmci_resource *vmci_resource_lookup(struct vmci_handle handle)
+{
+ struct vmci_resource *r, *resource = NULL;
+ struct hlist_node *node;
+ unsigned int idx = vmci_resource_hash(handle);
+
+ rcu_read_lock();
+ hlist_for_each_entry_rcu(r, node,
+ &vmci_resource_table.entries[idx], node) {
+ u32 rid = VMCI_HANDLE_TO_RESOURCE_ID(r->handle);
+ u32 cid = VMCI_HANDLE_TO_CONTEXT_ID(r->handle);
+
+ if (rid == VMCI_HANDLE_TO_RESOURCE_ID(handle) &&
+ (cid == VMCI_HANDLE_TO_CONTEXT_ID(handle) ||
+ cid == VMCI_INVALID_ID)) {
+ resource = r;
+ break;
+ }
+ }
+ rcu_read_unlock();
+
+ return resource;
+}
+
+/*
+ * Find an unused resource ID and return it. The first
+ * VMCI_RESERVED_RESOURCE_ID_MAX are reserved so we start from
+ * its value + 1.
+ * Returns VMCI resource id on success, VMCI_INVALID_ID on failure.
+ */
+static u32 vmci_resource_find_id(u32 context_id)
+{
+ static u32 resource_id = VMCI_RESERVED_RESOURCE_ID_MAX + 1;
+ u32 old_rid = resource_id;
+ u32 current_rid;
+
+ /*
+ * Generate a unique resource ID. Keep on trying until we wrap around
+ * in the RID space.
+ */
+ BUG_ON(old_rid <= VMCI_RESERVED_RESOURCE_ID_MAX);
+
+ do {
+ struct vmci_handle handle;
+
+ current_rid = resource_id;
+ resource_id++;
+ if (unlikely(resource_id == VMCI_INVALID_ID)) {
+ /* Skip the reserved rids. */
+ resource_id = VMCI_RESERVED_RESOURCE_ID_MAX + 1;
+ }
+
+ handle = vmci_make_handle(context_id, current_rid);
+ if (!vmci_resource_lookup(handle))
+ return current_rid;
+ } while (resource_id != old_rid);
+
+ return VMCI_INVALID_ID;
+}
+
+
+int vmci_resource_add(struct vmci_resource *resource,
+ enum vmci_resource_type resource_type,
+ struct vmci_handle handle)
+
+{
+ unsigned int idx;
+ int result;
+
+ spin_lock(&vmci_resource_table.lock);
+
+ if (handle.resource == VMCI_INVALID_ID) {
+ handle.resource = vmci_resource_find_id(handle.context);
+ if (handle.resource == VMCI_INVALID_ID) {
+ result = VMCI_ERROR_NO_HANDLE;
+ goto out;
+ }
+ } else if (vmci_resource_lookup(handle)) {
+ result = VMCI_ERROR_ALREADY_EXISTS;
+ goto out;
+ }
+
+ resource->handle = handle;
+ resource->type = resource_type;
+ INIT_HLIST_NODE(&resource->node);
+ kref_init(&resource->kref);
+ init_completion(&resource->done);
+
+ idx = vmci_resource_hash(resource->handle);
+ BUG_ON(idx >= VMCI_RESOURCE_HASH_BUCKETS);
+ hlist_add_head_rcu(&resource->node, &vmci_resource_table.entries[idx]);
+
+ result = VMCI_SUCCESS;
+
+out:
+ spin_unlock(&vmci_resource_table.lock);
+ return result;
+}
+
+void vmci_resource_remove(struct vmci_resource *resource)
+{
+ struct vmci_handle handle = resource->handle;
+ unsigned int idx = vmci_resource_hash(handle);
+ struct vmci_resource *r;
+ struct hlist_node *node;
+
+ /* Remove resource from hash table. */
+ spin_lock(&vmci_resource_table.lock);
+
+ hlist_for_each_entry(r, node, &vmci_resource_table.entries[idx], node) {
+ if (VMCI_HANDLE_EQUAL(r->handle, resource->handle)) {
+ BUG_ON(r != resource);
+ hlist_del_init_rcu(&r->node);
+ break;
+ }
+ }
+
+ spin_unlock(&vmci_resource_table.lock);
+ synchronize_rcu();
+
+ vmci_resource_put(resource);
+ wait_for_completion(&resource->done);
+}
+
+struct vmci_resource *
+vmci_resource_by_handle(struct vmci_handle resource_handle,
+ enum vmci_resource_type resource_type)
+{
+ struct vmci_resource *r, *resource = NULL;
+
+ rcu_read_lock();
+
+ r = vmci_resource_lookup(resource_handle);
+ if (r &&
+ (resource_type == r->type ||
+ resource_type == VMCI_RESOURCE_TYPE_ANY)) {
+ resource = vmci_resource_get(r);
+ }
+
+ rcu_read_unlock();
+
+ return resource;
+}
+
+/*
+ * Get a reference to given resource.
+ */
+struct vmci_resource *vmci_resource_get(struct vmci_resource *resource)
+{
+ kref_get(&resource->kref);
+
+ return resource;
+}
+
+static void vmci_release_resource(struct kref *kref)
+{
+ struct vmci_resource *resource =
+ container_of(kref, struct vmci_resource, kref);
+
+ /* Verify the resource has been unlinked from hash table */
+ WARN_ON(!hlist_unhashed(&resource->node));
+
+ /* Signal that container of this resource can now be destroyed */
+ complete(&resource->done);
+}
+
+/*
+ * Resource's release function will get called if last reference.
+ * If it is the last reference, then we are sure that nobody else
+ * can increment the count again (it's gone from the resource hash
+ * table), so there's no need for locking here.
+ */
+int vmci_resource_put(struct vmci_resource *resource)
+{
+ /*
+ * We propagate the information back to caller in case it wants to know
+ * whether entry was freed.
+ */
+ return kref_put(&resource->kref, vmci_release_resource) ?
+ VMCI_SUCCESS_ENTRY_DEAD : VMCI_SUCCESS;
+}
+
+struct vmci_handle vmci_resource_handle(struct vmci_resource *resource)
+{
+ return resource->handle;
+}
diff --git a/drivers/misc/vmw_vmci/vmci_resource.h b/drivers/misc/vmw_vmci/vmci_resource.h
new file mode 100644
index 0000000..9190cd2
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_resource.h
@@ -0,0 +1,59 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#ifndef _VMCI_RESOURCE_H_
+#define _VMCI_RESOURCE_H_
+
+#include <linux/vmw_vmci_defs.h>
+#include <linux/types.h>
+
+#include "vmci_context.h"
+
+
+enum vmci_resource_type {
+ VMCI_RESOURCE_TYPE_ANY,
+ VMCI_RESOURCE_TYPE_API,
+ VMCI_RESOURCE_TYPE_GROUP,
+ VMCI_RESOURCE_TYPE_DATAGRAM,
+ VMCI_RESOURCE_TYPE_DOORBELL,
+ VMCI_RESOURCE_TYPE_QPAIR_GUEST,
+ VMCI_RESOURCE_TYPE_QPAIR_HOST
+};
+
+struct vmci_resource {
+ struct vmci_handle handle;
+ enum vmci_resource_type type;
+ struct hlist_node node;
+ struct kref kref;
+ struct completion done;
+};
+
+
+int vmci_resource_add(struct vmci_resource *resource,
+ enum vmci_resource_type resource_type,
+ struct vmci_handle handle);
+
+void vmci_resource_remove(struct vmci_resource *resource);
+
+struct vmci_resource *
+vmci_resource_by_handle(struct vmci_handle resource_handle,
+ enum vmci_resource_type resource_type);
+
+struct vmci_resource *vmci_resource_get(struct vmci_resource *resource);
+int vmci_resource_put(struct vmci_resource *resource);
+
+struct vmci_handle vmci_resource_handle(struct vmci_resource *resource);
+
+#endif /* _VMCI_RESOURCE_H_ */
^ permalink raw reply related
* [PATCH 09/12] VMCI: routing implementation.
From: George Zhang @ 2012-11-07 18:42 UTC (permalink / raw)
To: linux-kernel, georgezhang, virtualization; +Cc: pv-drivers, gregkh
In-Reply-To: <20121107183624.9658.78903.stgit@promb-2n-dhcp175.eng.vmware.com>
VMCI routing code is responsible for routing between various hosts/guests as well as
routing in nested scenarios.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
drivers/misc/vmw_vmci/vmci_route.c | 229 ++++++++++++++++++++++++++++++++++++
drivers/misc/vmw_vmci/vmci_route.h | 30 +++++
2 files changed, 259 insertions(+), 0 deletions(-)
create mode 100644 drivers/misc/vmw_vmci/vmci_route.c
create mode 100644 drivers/misc/vmw_vmci/vmci_route.h
diff --git a/drivers/misc/vmw_vmci/vmci_route.c b/drivers/misc/vmw_vmci/vmci_route.c
new file mode 100644
index 0000000..adf2f9f
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_route.c
@@ -0,0 +1,229 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#include <linux/vmw_vmci_defs.h>
+#include <linux/vmw_vmci_api.h>
+
+#include "vmci_common_int.h"
+#include "vmci_context.h"
+#include "vmci_driver.h"
+#include "vmci_route.h"
+
+/*
+ * Make a routing decision for the given source and destination handles.
+ * This will try to determine the route using the handles and the available
+ * devices. Will set the source context if it is invalid.
+ */
+int vmci_route(struct vmci_handle *src,
+ const struct vmci_handle *dst,
+ bool from_guest,
+ enum vmci_route *route)
+{
+ bool has_host_device = vmci_host_code_active();
+ bool has_guest_device = vmci_guest_code_active();
+
+ *route = VMCI_ROUTE_NONE;
+
+ /*
+ * "from_guest" is only ever set to true by
+ * IOCTL_VMCI_DATAGRAM_SEND (or by the vmkernel equivalent),
+ * which comes from the VMX, so we know it is coming from a
+ * guest.
+ *
+ * To avoid inconsistencies, test these once. We will test
+ * them again when we do the actual send to ensure that we do
+ * not touch a non-existent device.
+ */
+
+ /* Must have a valid destination context. */
+ if (VMCI_INVALID_ID == dst->context)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ /* Anywhere to hypervisor. */
+ if (VMCI_HYPERVISOR_CONTEXT_ID == dst->context) {
+
+ /*
+ * If this message already came from a guest then we
+ * cannot send it to the hypervisor. It must come
+ * from a local client.
+ */
+ if (from_guest)
+ return VMCI_ERROR_DST_UNREACHABLE;
+
+ /*
+ * We must be acting as a guest in order to send to
+ * the hypervisor.
+ */
+ if (!has_guest_device)
+ return VMCI_ERROR_DEVICE_NOT_FOUND;
+
+ /* And we cannot send if the source is the host context. */
+ if (VMCI_HOST_CONTEXT_ID == src->context)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ /*
+ * If the client passed the ANON source handle then
+ * respect it (both context and resource are invalid).
+ * However, if they passed only an invalid context,
+ * then they probably mean ANY, in which case we
+ * should set the real context here before passing it
+ * down.
+ */
+ if (VMCI_INVALID_ID == src->context &&
+ VMCI_INVALID_ID != src->resource)
+ src->context = vmci_get_context_id();
+
+ /* Send from local client down to the hypervisor. */
+ *route = VMCI_ROUTE_AS_GUEST;
+ return VMCI_SUCCESS;
+ }
+
+ /* Anywhere to local client on host. */
+ if (VMCI_HOST_CONTEXT_ID == dst->context) {
+ /*
+ * If it is not from a guest but we are acting as a
+ * guest, then we need to send it down to the host.
+ * Note that if we are also acting as a host then this
+ * will prevent us from sending from local client to
+ * local client, but we accept that restriction as a
+ * way to remove any ambiguity from the host context.
+ */
+ if (src->context == VMCI_HYPERVISOR_CONTEXT_ID) {
+ /*
+ * If the hypervisor is the source, this is
+ * host local communication. The hypervisor
+ * may send vmci event datagrams to the host
+ * itself, but it will never send datagrams to
+ * an "outer host" through the guest device.
+ */
+
+ if (has_host_device) {
+ *route = VMCI_ROUTE_AS_HOST;
+ return VMCI_SUCCESS;
+ } else {
+ return VMCI_ERROR_DEVICE_NOT_FOUND;
+ }
+ }
+
+ if (!from_guest && has_guest_device) {
+ /* If no source context then use the current. */
+ if (VMCI_INVALID_ID == src->context)
+ src->context = vmci_get_context_id();
+
+ /* Send it from local client down to the host. */
+ *route = VMCI_ROUTE_AS_GUEST;
+ return VMCI_SUCCESS;
+ }
+
+ /*
+ * Otherwise we already received it from a guest and
+ * it is destined for a local client on this host, or
+ * it is from another local client on this host. We
+ * must be acting as a host to service it.
+ */
+ if (!has_host_device)
+ return VMCI_ERROR_DEVICE_NOT_FOUND;
+
+ if (VMCI_INVALID_ID == src->context) {
+ /*
+ * If it came from a guest then it must have a
+ * valid context. Otherwise we can use the
+ * host context.
+ */
+ if (from_guest)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ src->context = VMCI_HOST_CONTEXT_ID;
+ }
+
+ /* Route to local client. */
+ *route = VMCI_ROUTE_AS_HOST;
+ return VMCI_SUCCESS;
+ }
+
+ /*
+ * If we are acting as a host then this might be destined for
+ * a guest.
+ */
+ if (has_host_device) {
+ /* It will have a context if it is meant for a guest. */
+ if (vmci_ctx_exists(dst->context)) {
+ if (VMCI_INVALID_ID == src->context) {
+ /*
+ * If it came from a guest then it
+ * must have a valid context.
+ * Otherwise we can use the host
+ * context.
+ */
+
+ if (from_guest)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ src->context = VMCI_HOST_CONTEXT_ID;
+ } else if (VMCI_CONTEXT_IS_VM(src->context) &&
+ src->context != dst->context) {
+ /*
+ * VM to VM communication is not
+ * allowed. Since we catch all
+ * communication destined for the host
+ * above, this must be destined for a
+ * VM since there is a valid context.
+ */
+
+ ASSERT(VMCI_CONTEXT_IS_VM(dst->context));
+
+ return VMCI_ERROR_DST_UNREACHABLE;
+ }
+
+ /* Pass it up to the guest. */
+ *route = VMCI_ROUTE_AS_HOST;
+ return VMCI_SUCCESS;
+ } else if (!has_guest_device) {
+ /*
+ * The host is attempting to reach a CID
+ * without an active context, and we can't
+ * send it down, since we have no guest
+ * device.
+ */
+
+ return VMCI_ERROR_DST_UNREACHABLE;
+ }
+ }
+
+ /*
+ * We must be a guest trying to send to another guest, which means
+ * we need to send it down to the host. We do not filter out VM to
+ * VM communication here, since we want to be able to use the guest
+ * driver on older versions that do support VM to VM communication.
+ */
+ if (!has_guest_device) {
+ /*
+ * Ending up here means we have neither guest nor host
+ * device.
+ */
+ return VMCI_ERROR_DEVICE_NOT_FOUND;
+ }
+
+ /* If no source context then use the current context. */
+ if (VMCI_INVALID_ID == src->context)
+ src->context = vmci_get_context_id();
+
+ /*
+ * Send it from local client down to the host, which will
+ * route it to the other guest for us.
+ */
+ *route = VMCI_ROUTE_AS_GUEST;
+ return VMCI_SUCCESS;
+}
diff --git a/drivers/misc/vmw_vmci/vmci_route.h b/drivers/misc/vmw_vmci/vmci_route.h
new file mode 100644
index 0000000..3b30e82
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_route.h
@@ -0,0 +1,30 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#ifndef _VMCI_ROUTE_H_
+#define _VMCI_ROUTE_H_
+
+#include <linux/vmw_vmci_defs.h>
+
+enum vmci_route {
+ VMCI_ROUTE_NONE,
+ VMCI_ROUTE_AS_HOST,
+ VMCI_ROUTE_AS_GUEST,
+};
+
+int vmci_route(struct vmci_handle *src, const struct vmci_handle *dst,
+ bool from_guest, enum vmci_route *route);
+
+#endif /* _VMCI_ROUTE_H_ */
^ permalink raw reply related
* [PATCH 10/12] VMCI: guest side driver implementation.
From: George Zhang @ 2012-11-07 18:42 UTC (permalink / raw)
To: linux-kernel, georgezhang, virtualization; +Cc: pv-drivers, gregkh
In-Reply-To: <20121107183624.9658.78903.stgit@promb-2n-dhcp175.eng.vmware.com>
VMCI guest side driver code implementation.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
drivers/misc/vmw_vmci/vmci_guest.c | 762 ++++++++++++++++++++++++++++++++++++
1 files changed, 762 insertions(+), 0 deletions(-)
create mode 100644 drivers/misc/vmw_vmci/vmci_guest.c
diff --git a/drivers/misc/vmw_vmci/vmci_guest.c b/drivers/misc/vmw_vmci/vmci_guest.c
new file mode 100644
index 0000000..eedbe4d
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_guest.c
@@ -0,0 +1,762 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#include <linux/vmw_vmci_defs.h>
+#include <linux/vmw_vmci_api.h>
+#include <linux/moduleparam.h>
+#include <linux/interrupt.h>
+#include <linux/highmem.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/smp.h>
+#include <linux/io.h>
+
+#include "vmci_common_int.h"
+#include "vmci_datagram.h"
+#include "vmci_doorbell.h"
+#include "vmci_context.h"
+#include "vmci_driver.h"
+#include "vmci_event.h"
+
+#define VMCI_UTIL_NUM_RESOURCES 1
+
+static bool vmci_disable_msi;
+module_param_named(disable_msi, vmci_disable_msi, bool, 0);
+MODULE_PARM_DESC(disable_msi, "Disable MSI use in driver - (default=0)");
+
+static bool vmci_disable_msix;
+module_param_named(disable_msix, vmci_disable_msix, bool, 0);
+MODULE_PARM_DESC(disable_msix, "Disable MSI-X use in driver - (default=0)");
+
+static u32 ctx_update_sub_id = VMCI_INVALID_ID;
+static u32 vm_context_id = VMCI_INVALID_ID;
+
+struct vmci_guest_device {
+ struct device *dev; /* PCI device we are attached to */
+ void __iomem *iobase;
+
+ unsigned int irq;
+ unsigned int intr_type;
+ bool exclusive_vectors;
+ struct msix_entry msix_entries[VMCI_MAX_INTRS];
+
+ struct tasklet_struct datagram_tasklet;
+ struct tasklet_struct bm_tasklet;
+
+ void *data_buffer;
+ void *notification_bitmap;
+};
+
+/* vmci_dev singleton device and supporting data*/
+static struct vmci_guest_device *vmci_dev_g;
+static DEFINE_SPINLOCK(vmci_dev_spinlock);
+
+static atomic_t vmci_num_guest_devices = ATOMIC_INIT(0);
+
+bool vmci_guest_code_active(void)
+{
+ return atomic_read(&vmci_num_guest_devices) != 0;
+}
+
+u32 vmci_get_vm_context_id(void)
+{
+ if (vm_context_id == VMCI_INVALID_ID) {
+ u32 result;
+ struct vmci_datagram get_cid_msg;
+ get_cid_msg.dst =
+ vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID,
+ VMCI_GET_CONTEXT_ID);
+ get_cid_msg.src = VMCI_ANON_SRC_HANDLE;
+ get_cid_msg.payload_size = 0;
+ result = vmci_send_datagram(&get_cid_msg);
+ if (result >= 0)
+ vm_context_id = result;
+ }
+ return vm_context_id;
+}
+
+/*
+ * VM to hypervisor call mechanism. We use the standard VMware naming
+ * convention since shared code is calling this function as well.
+ */
+int vmci_send_datagram(struct vmci_datagram *dg)
+{
+ unsigned long flags;
+ int result;
+
+ /* Check args. */
+ if (dg == NULL)
+ return VMCI_ERROR_INVALID_ARGS;
+
+ /*
+ * Need to acquire spinlock on the device because the datagram
+ * data may be spread over multiple pages and the monitor may
+ * interleave device user rpc calls from multiple
+ * VCPUs. Acquiring the spinlock precludes that
+ * possibility. Disabling interrupts to avoid incoming
+ * datagrams during a "rep out" and possibly landing up in
+ * this function.
+ */
+ spin_lock_irqsave(&vmci_dev_spinlock, flags);
+
+ if (vmci_dev_g) {
+ iowrite8_rep(vmci_dev_g->iobase + VMCI_DATA_OUT_ADDR,
+ dg, VMCI_DG_SIZE(dg));
+ result = ioread32(vmci_dev_g->iobase + VMCI_RESULT_LOW_ADDR);
+ } else {
+ result = VMCI_ERROR_UNAVAILABLE;
+ }
+
+ spin_unlock_irqrestore(&vmci_dev_spinlock, flags);
+
+ return result;
+}
+EXPORT_SYMBOL_GPL(vmci_send_datagram);
+
+/*
+ * Gets called with the new context id if updated or resumed.
+ * Context id.
+ */
+static void vmci_guest_cid_update(u32 sub_id,
+ const struct vmci_event_data *event_data,
+ void *client_data)
+{
+ const struct vmci_event_payld_ctx *ev_payload =
+ vmci_event_data_const_payload(event_data);
+
+ if (sub_id != ctx_update_sub_id) {
+ pr_devel("Invalid subscriber (ID=0x%x).\n", sub_id);
+ return;
+ }
+
+ if (!event_data || ev_payload->context_id == VMCI_INVALID_ID) {
+ pr_devel("Invalid event data.\n");
+ return;
+ }
+
+ pr_devel("Updating context from (ID=0x%x) to (ID=0x%x) on event (type=%d).\n",
+ vm_context_id, ev_payload->context_id, event_data->event);
+
+ vm_context_id = ev_payload->context_id;
+}
+
+/*
+ * Verify that the host supports the hypercalls we need. If it does not,
+ * try to find fallback hypercalls and use those instead. Returns
+ * true if required hypercalls (or fallback hypercalls) are
+ * supported by the host, false otherwise.
+ */
+static bool vmci_check_host_caps(struct pci_dev *pdev)
+{
+ bool result;
+ struct vmci_resource_query_msg *msg;
+ u32 msg_size = sizeof(struct vmci_resource_query_hdr) +
+ VMCI_UTIL_NUM_RESOURCES * sizeof(u32);
+ struct vmci_datagram *check_msg;
+
+ check_msg = kmalloc(msg_size, GFP_KERNEL);
+ if (!check_msg) {
+ dev_err(&pdev->dev, "%s: Insufficient memory.\n", __func__);
+ return false;
+ }
+
+ check_msg->dst = vmci_make_handle(VMCI_HYPERVISOR_CONTEXT_ID,
+ VMCI_RESOURCES_QUERY);
+ check_msg->src = VMCI_ANON_SRC_HANDLE;
+ check_msg->payload_size = msg_size - VMCI_DG_HEADERSIZE;
+ msg = (struct vmci_resource_query_msg *)VMCI_DG_PAYLOAD(check_msg);
+
+ msg->num_resources = VMCI_UTIL_NUM_RESOURCES;
+ msg->resources[0] = VMCI_GET_CONTEXT_ID;
+
+ /* Checks that hyper calls are supported */
+ result = vmci_send_datagram(check_msg) == 0x01;
+ kfree(check_msg);
+
+ dev_dbg(&pdev->dev, "%s: Host capability check: %s.\n",
+ __func__, result ? "PASSED" : "FAILED");
+
+ /* We need the vector. There are no fallbacks. */
+ return result;
+}
+
+/*
+ * Reads datagrams from the data in port and dispatches them. We
+ * always start reading datagrams into only the first page of the
+ * datagram buffer. If the datagrams don't fit into one page, we
+ * use the maximum datagram buffer size for the remainder of the
+ * invocation. This is a simple heuristic for not penalizing
+ * small datagrams.
+ *
+ * This function assumes that it has exclusive access to the data
+ * in port for the duration of the call.
+ */
+static void vmci_dispatch_dgs(unsigned long data)
+{
+ struct vmci_guest_device *vmci_dev = (struct vmci_guest_device *)data;
+ u8 *dg_in_buffer = vmci_dev->data_buffer;
+ struct vmci_datagram *dg;
+ size_t dg_in_buffer_size = VMCI_MAX_DG_SIZE;
+ size_t current_dg_in_buffer_size = PAGE_SIZE;
+ size_t remaining_bytes;
+
+ BUILD_BUG_ON(VMCI_MAX_DG_SIZE < PAGE_SIZE);
+
+ ioread8_rep(vmci_dev->iobase + VMCI_DATA_IN_ADDR,
+ vmci_dev->data_buffer, current_dg_in_buffer_size);
+ dg = (struct vmci_datagram *)dg_in_buffer;
+ remaining_bytes = current_dg_in_buffer_size;
+
+ while (dg->dst.resource != VMCI_INVALID_ID ||
+ remaining_bytes > PAGE_SIZE) {
+ unsigned dg_in_size;
+
+ /*
+ * When the input buffer spans multiple pages, a datagram can
+ * start on any page boundary in the buffer.
+ */
+ if (dg->dst.resource == VMCI_INVALID_ID) {
+ ASSERT(remaining_bytes > PAGE_SIZE);
+ dg = (struct vmci_datagram *)roundup((uintptr_t)
+ dg + 1, PAGE_SIZE);
+ ASSERT((u8 *) dg <
+ dg_in_buffer + current_dg_in_buffer_size);
+ remaining_bytes =
+ (size_t) (dg_in_buffer + current_dg_in_buffer_size -
+ (u8 *) dg);
+ continue;
+ }
+
+ dg_in_size = VMCI_DG_SIZE_ALIGNED(dg);
+
+ if (dg_in_size <= dg_in_buffer_size) {
+ int result;
+
+ /*
+ * If the remaining bytes in the datagram
+ * buffer doesn't contain the complete
+ * datagram, we first make sure we have enough
+ * room for it and then we read the reminder
+ * of the datagram and possibly any following
+ * datagrams.
+ */
+ if (dg_in_size > remaining_bytes) {
+ if (remaining_bytes !=
+ current_dg_in_buffer_size) {
+
+ /*
+ * We move the partial
+ * datagram to the front and
+ * read the reminder of the
+ * datagram and possibly
+ * following calls into the
+ * following bytes.
+ */
+ memmove(dg_in_buffer, dg_in_buffer +
+ current_dg_in_buffer_size -
+ remaining_bytes,
+ remaining_bytes);
+ dg = (struct vmci_datagram *)
+ dg_in_buffer;
+ }
+
+ if (current_dg_in_buffer_size !=
+ dg_in_buffer_size)
+ current_dg_in_buffer_size =
+ dg_in_buffer_size;
+
+ ioread8_rep(vmci_dev->iobase +
+ VMCI_DATA_IN_ADDR,
+ vmci_dev->data_buffer +
+ remaining_bytes,
+ current_dg_in_buffer_size -
+ remaining_bytes);
+ }
+
+ /*
+ * We special case event datagrams from the
+ * hypervisor.
+ */
+ if (dg->src.context == VMCI_HYPERVISOR_CONTEXT_ID &&
+ dg->dst.resource == VMCI_EVENT_HANDLER) {
+ result = vmci_event_dispatch(dg);
+ } else {
+ result = vmci_datagram_invoke_guest_handler(dg);
+ }
+ if (result < VMCI_SUCCESS)
+ dev_dbg(vmci_dev->dev,
+ "Datagram with resource (ID=0x%x) failed (err=%d).\n",
+ dg->dst.resource, result);
+
+ /* On to the next datagram. */
+ dg = (struct vmci_datagram *)((u8 *) dg +
+ dg_in_size);
+ } else {
+ size_t bytes_to_skip;
+
+ /*
+ * Datagram doesn't fit in datagram buffer of maximal
+ * size. We drop it.
+ */
+ dev_dbg(vmci_dev->dev,
+ "Failed to receive datagram (size=%u bytes).\n",
+ dg_in_size);
+
+ bytes_to_skip = dg_in_size - remaining_bytes;
+ if (current_dg_in_buffer_size != dg_in_buffer_size)
+ current_dg_in_buffer_size = dg_in_buffer_size;
+
+ for (;;) {
+ ioread8_rep(vmci_dev->iobase +
+ VMCI_DATA_IN_ADDR,
+ vmci_dev->data_buffer,
+ current_dg_in_buffer_size);
+ if (bytes_to_skip <= current_dg_in_buffer_size)
+ break;
+
+ bytes_to_skip -= current_dg_in_buffer_size;
+ }
+ dg = (struct vmci_datagram *)(dg_in_buffer +
+ bytes_to_skip);
+ }
+
+ remaining_bytes =
+ (size_t) (dg_in_buffer + current_dg_in_buffer_size -
+ (u8 *) dg);
+
+ if (remaining_bytes < VMCI_DG_HEADERSIZE) {
+ /* Get the next batch of datagrams. */
+
+ ioread8_rep(vmci_dev->iobase + VMCI_DATA_IN_ADDR,
+ vmci_dev->data_buffer,
+ current_dg_in_buffer_size);
+ dg = (struct vmci_datagram *)dg_in_buffer;
+ remaining_bytes = current_dg_in_buffer_size;
+ }
+ }
+}
+
+/*
+ * Scans the notification bitmap for raised flags, clears them
+ * and handles the notifications.
+ */
+static void vmci_process_bitmap(unsigned long data)
+{
+ struct vmci_guest_device *dev = (struct vmci_guest_device *)data;
+
+ if (!dev->notification_bitmap) {
+ dev_dbg(dev->dev, "No bitmap present in %s.\n", __func__);
+ return;
+ }
+
+ vmci_dbell_scan_notification_entries(dev->notification_bitmap);
+}
+
+/*
+ * Enable MSI-X. Try exclusive vectors first, then shared vectors.
+ */
+static int vmci_enable_msix(struct pci_dev *pdev,
+ struct vmci_guest_device *vmci_dev)
+{
+ int i;
+ int result;
+
+ for (i = 0; i < VMCI_MAX_INTRS; ++i) {
+ vmci_dev->msix_entries[i].entry = i;
+ vmci_dev->msix_entries[i].vector = i;
+ }
+
+ result = pci_enable_msix(pdev, vmci_dev->msix_entries, VMCI_MAX_INTRS);
+ if (result == 0)
+ vmci_dev->exclusive_vectors = true;
+ else if (result > 0)
+ result = pci_enable_msix(pdev, vmci_dev->msix_entries, 1);
+
+ return result;
+}
+
+/*
+ * Interrupt handler for legacy or MSI interrupt, or for first MSI-X
+ * interrupt (vector VMCI_INTR_DATAGRAM).
+ */
+static irqreturn_t vmci_interrupt(int irq, void *_dev)
+{
+ struct vmci_guest_device *dev = _dev;
+
+ /*
+ * If we are using MSI-X with exclusive vectors then we simply schedule
+ * the datagram tasklet, since we know the interrupt was meant for us.
+ * Otherwise we must read the ICR to determine what to do.
+ */
+
+ if (dev->intr_type == VMCI_INTR_TYPE_MSIX && dev->exclusive_vectors) {
+ tasklet_schedule(&dev->datagram_tasklet);
+ } else {
+ unsigned int icr;
+
+ ASSERT(dev->intr_type == VMCI_INTR_TYPE_INTX ||
+ dev->intr_type == VMCI_INTR_TYPE_MSI);
+
+ /* Acknowledge interrupt and determine what needs doing. */
+ icr = ioread32(dev->iobase + VMCI_ICR_ADDR);
+ if (icr == 0 || icr == ~0)
+ return IRQ_NONE;
+
+ if (icr & VMCI_ICR_DATAGRAM) {
+ tasklet_schedule(&dev->datagram_tasklet);
+ icr &= ~VMCI_ICR_DATAGRAM;
+ }
+
+ if (icr & VMCI_ICR_NOTIFICATION) {
+ tasklet_schedule(&dev->bm_tasklet);
+ icr &= ~VMCI_ICR_NOTIFICATION;
+ }
+
+ if (icr != 0)
+ dev_warn(dev->dev,
+ "Ignoring unknown interrupt cause (%d).\n", icr);
+ }
+
+ return IRQ_HANDLED;
+}
+
+/*
+ * Interrupt handler for MSI-X interrupt vector VMCI_INTR_NOTIFICATION,
+ * which is for the notification bitmap. Will only get called if we are
+ * using MSI-X with exclusive vectors.
+ */
+static irqreturn_t vmci_interrupt_bm(int irq, void *_dev)
+{
+ struct vmci_guest_device *dev = _dev;
+
+ /* For MSI-X we can just assume it was meant for us. */
+ ASSERT(dev->intr_type == VMCI_INTR_TYPE_MSIX && dev->exclusive_vectors);
+ tasklet_schedule(&dev->bm_tasklet);
+
+ return IRQ_HANDLED;
+}
+
+/*
+ * Most of the initialization at module load time is done here.
+ */
+static int __devinit vmci_guest_probe_device(struct pci_dev *pdev,
+ const struct pci_device_id *id)
+{
+ struct vmci_guest_device *vmci_dev;
+ void __iomem *iobase;
+ unsigned int capabilities;
+ unsigned long cmd;
+ int vmci_err;
+ int error;
+
+ dev_dbg(&pdev->dev, "Probing for vmci/PCI guest device.\n");
+
+ error = pcim_enable_device(pdev);
+ if (error) {
+ dev_err(&pdev->dev,
+ "Failed to enable VMCI device: %d\n", error);
+ return error;
+ }
+
+ error = pcim_iomap_regions(pdev, 1 << 0, MODULE_NAME);
+ if (error) {
+ dev_err(&pdev->dev, "Failed to reserve/map IO regions.\n");
+ return error;
+ }
+
+ iobase = pcim_iomap_table(pdev)[0];
+
+ dev_info(&pdev->dev, "Found VMCI PCI device at %#lx, irq %u.\n",
+ (unsigned long)iobase, pdev->irq);
+
+ vmci_dev = devm_kzalloc(&pdev->dev, sizeof(*vmci_dev), GFP_KERNEL);
+ if (!vmci_dev) {
+ dev_err(&pdev->dev,
+ "Can't allocate memory for VMCI device.\n");
+ return -ENOMEM;
+ }
+
+ vmci_dev->dev = &pdev->dev;
+ vmci_dev->intr_type = VMCI_INTR_TYPE_INTX;
+ vmci_dev->exclusive_vectors = false;
+ vmci_dev->iobase = iobase;
+
+ tasklet_init(&vmci_dev->datagram_tasklet,
+ vmci_dispatch_dgs, (unsigned long)vmci_dev);
+ tasklet_init(&vmci_dev->bm_tasklet,
+ vmci_process_bitmap, (unsigned long)vmci_dev);
+
+ vmci_dev->data_buffer = vmalloc(VMCI_MAX_DG_SIZE);
+ if (!vmci_dev->data_buffer) {
+ dev_err(&pdev->dev,
+ "Can't allocate memory for datagram buffer.\n");
+ return -ENOMEM;
+ }
+
+ pci_set_master(pdev); /* To enable queue_pair functionality. */
+
+ /*
+ * Verify that the VMCI Device supports the capabilities that
+ * we need. If the device is missing capabilities that we would
+ * like to use, check for fallback capabilities and use those
+ * instead (so we can run a new VM on old hosts). Fail the load if
+ * a required capability is missing and there is no fallback.
+ *
+ * Right now, we need datagrams. There are no fallbacks.
+ */
+ capabilities = ioread32(vmci_dev->iobase + VMCI_CAPS_ADDR);
+ if (!(capabilities & VMCI_CAPS_DATAGRAM)) {
+ dev_err(&pdev->dev, "Device does not support datagrams.\n");
+ error = -ENXIO;
+ goto err_free_data_buffer;
+ }
+
+ /*
+ * If the hardware supports notifications, we will use that as
+ * well.
+ */
+ if (capabilities & VMCI_CAPS_NOTIFICATIONS) {
+ vmci_dev->notification_bitmap = vmalloc(PAGE_SIZE);
+ if (!vmci_dev->notification_bitmap) {
+ dev_warn(&pdev->dev,
+ "Unable to allocate notification bitmap.\n");
+ } else {
+ memset(vmci_dev->notification_bitmap, 0, PAGE_SIZE);
+ capabilities |= VMCI_CAPS_NOTIFICATIONS;
+ }
+ }
+
+ dev_info(&pdev->dev, "Using capabilities 0x%x.\n", capabilities);
+
+ /* Let the host know which capabilities we intend to use. */
+ iowrite32(capabilities, vmci_dev->iobase + VMCI_CAPS_ADDR);
+
+ /* Set up global device so that we can start sending datagrams */
+ spin_lock_irq(&vmci_dev_spinlock);
+ vmci_dev_g = vmci_dev;
+ spin_unlock_irq(&vmci_dev_spinlock);
+
+ /*
+ * Register notification bitmap with device if that capability is
+ * used.
+ */
+ if (capabilities & VMCI_CAPS_NOTIFICATIONS) {
+ struct page *page =
+ vmalloc_to_page(vmci_dev->notification_bitmap);
+ unsigned long bitmap_ppn = page_to_pfn(page);
+ if (!vmci_dbell_register_notification_bitmap(bitmap_ppn)) {
+ dev_warn(&pdev->dev,
+ "VMCI device unable to register notification bitmap with PPN 0x%x.\n",
+ (u32) bitmap_ppn);
+ goto err_remove_vmci_dev_g;
+ }
+ }
+
+ /* Check host capabilities. */
+ if (!vmci_check_host_caps(pdev))
+ goto err_remove_bitmap;
+
+ /* Enable device. */
+
+ /*
+ * We subscribe to the VMCI_EVENT_CTX_ID_UPDATE here so we can
+ * update the internal context id when needed.
+ */
+ vmci_err = vmci_event_subscribe(VMCI_EVENT_CTX_ID_UPDATE,
+ vmci_guest_cid_update, NULL,
+ &ctx_update_sub_id);
+ if (vmci_err < VMCI_SUCCESS)
+ dev_warn(&pdev->dev,
+ "Failed to subscribe to event (type=%d): %d\n",
+ VMCI_EVENT_CTX_ID_UPDATE, vmci_err);
+
+ /*
+ * Enable interrupts. Try MSI-X first, then MSI, and then fallback on
+ * legacy interrupts.
+ */
+ if (!vmci_disable_msix && !vmci_enable_msix(pdev, vmci_dev)) {
+ vmci_dev->intr_type = VMCI_INTR_TYPE_MSIX;
+ vmci_dev->irq = vmci_dev->msix_entries[0].vector;
+ } else if (!vmci_disable_msi && !pci_enable_msi(pdev)) {
+ vmci_dev->intr_type = VMCI_INTR_TYPE_MSI;
+ vmci_dev->irq = pdev->irq;
+ } else {
+ vmci_dev->intr_type = VMCI_INTR_TYPE_INTX;
+ vmci_dev->irq = pdev->irq;
+ }
+
+ /*
+ * Request IRQ for legacy or MSI interrupts, or for first
+ * MSI-X vector.
+ */
+ error = request_irq(vmci_dev->irq, vmci_interrupt, IRQF_SHARED,
+ MODULE_NAME, vmci_dev);
+ if (error) {
+ dev_err(&pdev->dev, "Irq %u in use: %d\n", vmci_dev->irq, error);
+ goto err_disable_msi;
+ }
+
+ /*
+ * For MSI-X with exclusive vectors we need to request an
+ * interrupt for each vector so that we get a separate
+ * interrupt handler routine. This allows us to distinguish
+ * between the vectors.
+ */
+ if (vmci_dev->exclusive_vectors) {
+ ASSERT(vmci_dev->intr_type == VMCI_INTR_TYPE_MSIX);
+ error = request_irq(vmci_dev->msix_entries[1].vector,
+ vmci_interrupt_bm, 0, MODULE_NAME,
+ vmci_dev);
+ if (error) {
+ dev_err(&pdev->dev,
+ "Failed to allocate irq %u: %d\n",
+ vmci_dev->msix_entries[1].vector, error);
+ goto err_free_irq;
+ }
+ }
+
+ dev_dbg(&pdev->dev, "Registered device.\n");
+
+ atomic_inc(&vmci_num_guest_devices);
+
+ /* Enable specific interrupt bits. */
+ cmd = VMCI_IMR_DATAGRAM;
+ if (capabilities & VMCI_CAPS_NOTIFICATIONS)
+ cmd |= VMCI_IMR_NOTIFICATION;
+ iowrite32(cmd, vmci_dev->iobase + VMCI_IMR_ADDR);
+
+ /* Enable interrupts. */
+ iowrite32(VMCI_CONTROL_INT_ENABLE,
+ vmci_dev->iobase + VMCI_CONTROL_ADDR);
+
+ pci_set_drvdata(pdev, vmci_dev);
+ return 0;
+
+err_free_irq:
+ free_irq(vmci_dev->irq, &vmci_dev);
+ tasklet_kill(&vmci_dev->datagram_tasklet);
+ tasklet_kill(&vmci_dev->bm_tasklet);
+
+err_disable_msi:
+ if (vmci_dev->intr_type == VMCI_INTR_TYPE_MSIX)
+ pci_disable_msix(pdev);
+ else if (vmci_dev->intr_type == VMCI_INTR_TYPE_MSI)
+ pci_disable_msi(pdev);
+
+ vmci_err = vmci_event_unsubscribe(ctx_update_sub_id);
+ if (vmci_err < VMCI_SUCCESS)
+ dev_warn(&pdev->dev,
+ "Failed to unsubscribe from event (type=%d) with subscriber (ID=0x%x): %d\n",
+ VMCI_EVENT_CTX_ID_UPDATE, ctx_update_sub_id, vmci_err);
+
+err_remove_bitmap:
+ if (vmci_dev->notification_bitmap) {
+ iowrite32(VMCI_CONTROL_RESET,
+ vmci_dev->iobase + VMCI_CONTROL_ADDR);
+ vfree(vmci_dev->notification_bitmap);
+ }
+
+err_remove_vmci_dev_g:
+ spin_lock_irq(&vmci_dev_spinlock);
+ vmci_dev_g = NULL;
+ spin_unlock_irq(&vmci_dev_spinlock);
+
+err_free_data_buffer:
+ vfree(vmci_dev->data_buffer);
+
+ /* The rest are managed resources and will be freed by PCI core */
+ return error;
+}
+
+static void __devexit vmci_guest_remove_device(struct pci_dev *pdev)
+{
+ struct vmci_guest_device *vmci_dev = pci_get_drvdata(pdev);
+ int vmci_err;
+
+ dev_dbg(&pdev->dev, "Removing device\n");
+
+ atomic_dec(&vmci_num_guest_devices);
+
+ vmci_qp_guest_endpoints_exit();
+
+ vmci_err = vmci_event_unsubscribe(ctx_update_sub_id);
+ if (vmci_err < VMCI_SUCCESS)
+ dev_warn(&pdev->dev,
+ "Failed to unsubscribe from event (type=%d) with subscriber (ID=0x%x): %d\n",
+ VMCI_EVENT_CTX_ID_UPDATE, ctx_update_sub_id, vmci_err);
+
+ spin_lock_irq(&vmci_dev_spinlock);
+ vmci_dev_g = NULL;
+ spin_unlock_irq(&vmci_dev_spinlock);
+
+ dev_dbg(&pdev->dev, "Resetting vmci device\n");
+ iowrite32(VMCI_CONTROL_RESET, vmci_dev->iobase + VMCI_CONTROL_ADDR);
+
+ /*
+ * Free IRQ and then disable MSI/MSI-X as appropriate. For
+ * MSI-X, we might have multiple vectors, each with their own
+ * IRQ, which we must free too.
+ */
+ free_irq(vmci_dev->irq, vmci_dev);
+ if (vmci_dev->intr_type == VMCI_INTR_TYPE_MSIX) {
+ if (vmci_dev->exclusive_vectors)
+ free_irq(vmci_dev->msix_entries[1].vector, vmci_dev);
+ pci_disable_msix(pdev);
+ } else if (vmci_dev->intr_type == VMCI_INTR_TYPE_MSI) {
+ pci_disable_msi(pdev);
+ }
+
+ tasklet_kill(&vmci_dev->datagram_tasklet);
+ tasklet_kill(&vmci_dev->bm_tasklet);
+
+ if (vmci_dev->notification_bitmap) {
+ /*
+ * The device reset above cleared the bitmap state of the
+ * device, so we can safely free it here.
+ */
+
+ vfree(vmci_dev->notification_bitmap);
+ }
+
+ vfree(vmci_dev->data_buffer);
+
+ /* The rest are managed resources and will be freed by PCI core */
+}
+
+static DEFINE_PCI_DEVICE_TABLE(vmci_ids) = {
+ { PCI_DEVICE(PCI_VENDOR_ID_VMWARE, PCI_DEVICE_ID_VMWARE_VMCI), },
+ { 0 },
+};
+MODULE_DEVICE_TABLE(pci, vmci_ids);
+
+static struct pci_driver vmci_guest_driver = {
+ .name = MODULE_NAME,
+ .id_table = vmci_ids,
+ .probe = vmci_guest_probe_device,
+ .remove = __devexit_p(vmci_guest_remove_device),
+};
+
+int __init vmci_guest_init(void)
+{
+ return pci_register_driver(&vmci_guest_driver);
+}
+
+void __exit vmci_guest_exit(void)
+{
+ pci_unregister_driver(&vmci_guest_driver);
+}
^ permalink raw reply related
* [PATCH 11/12] VMCI: host side driver implementation.
From: George Zhang @ 2012-11-07 18:42 UTC (permalink / raw)
To: linux-kernel, georgezhang, virtualization; +Cc: pv-drivers, gregkh
In-Reply-To: <20121107183624.9658.78903.stgit@promb-2n-dhcp175.eng.vmware.com>
VMCI host side driver code implementation.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
drivers/misc/vmw_vmci/vmci_host.c | 1033 +++++++++++++++++++++++++++++++++++++
1 files changed, 1033 insertions(+), 0 deletions(-)
create mode 100644 drivers/misc/vmw_vmci/vmci_host.c
diff --git a/drivers/misc/vmw_vmci/vmci_host.c b/drivers/misc/vmw_vmci/vmci_host.c
new file mode 100644
index 0000000..7580979
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_host.c
@@ -0,0 +1,1033 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#include <linux/vmw_vmci_defs.h>
+#include <linux/vmw_vmci_api.h>
+#include <linux/moduleparam.h>
+#include <linux/miscdevice.h>
+#include <linux/interrupt.h>
+#include <linux/highmem.h>
+#include <linux/atomic.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/sched.h>
+#include <linux/file.h>
+#include <linux/init.h>
+#include <linux/poll.h>
+#include <linux/pci.h>
+#include <linux/smp.h>
+#include <linux/fs.h>
+#include <linux/io.h>
+
+#include "vmci_handle_array.h"
+#include "vmci_common_int.h"
+#include "vmci_queue_pair.h"
+#include "vmci_datagram.h"
+#include "vmci_doorbell.h"
+#include "vmci_resource.h"
+#include "vmci_context.h"
+#include "vmci_driver.h"
+#include "vmci_event.h"
+
+#define VMCI_UTIL_NUM_RESOURCES 1
+
+enum {
+ VMCI_NOTIFY_RESOURCE_QUEUE_PAIR = 0,
+ VMCI_NOTIFY_RESOURCE_DOOR_BELL = 1,
+};
+
+enum {
+ VMCI_NOTIFY_RESOURCE_ACTION_NOTIFY = 0,
+ VMCI_NOTIFY_RESOURCE_ACTION_CREATE = 1,
+ VMCI_NOTIFY_RESOURCE_ACTION_DESTROY = 2,
+};
+
+/*
+ * VMCI driver initialization. This block can also be used to
+ * pass initial group membership etc.
+ */
+struct vmci_init_blk {
+ u32 cid;
+ u32 flags;
+};
+
+/* VMCIqueue_pairAllocInfo_VMToVM */
+struct vmci_qp_alloc_info_vmvm {
+ struct vmci_handle handle;
+ u32 peer;
+ u32 flags;
+ u64 produce_size;
+ u64 consume_size;
+ u64 produce_page_file; /* User VA. */
+ u64 consume_page_file; /* User VA. */
+ u64 produce_page_file_size; /* Size of the file name array. */
+ u64 consume_page_file_size; /* Size of the file name array. */
+ s32 result;
+ u32 _pad;
+};
+
+/* VMCISetNotifyInfo: Used to pass notify flag's address to the host driver. */
+struct vmci_set_notify_info {
+ u64 notify_uva;
+ s32 result;
+ u32 _pad;
+};
+
+/*
+ * Per-instance host state
+ */
+struct vmci_host_dev {
+ struct vmci_ctx *context;
+ int user_version;
+ enum vmci_obj_type ct_type;
+ struct mutex lock; /* Mutex lock for vmci context access */
+};
+
+static struct vmci_ctx *host_context;
+static bool vmci_host_device_initialized;
+static atomic_t vmci_host_active_users = ATOMIC_INIT(0);
+
+/*
+ * Determines whether the VMCI host personality is
+ * available. Since the core functionality of the host driver is
+ * always present, all guests could possibly use the host
+ * personality. However, to minimize the deviation from the
+ * pre-unified driver state of affairs, we only consider the host
+ * device active if there is no active guest device or if there
+ * are VMX'en with active VMCI contexts using the host device.
+ */
+bool vmci_host_code_active(void)
+{
+ return vmci_host_device_initialized &&
+ (!vmci_guest_code_active() ||
+ atomic_read(&vmci_host_active_users) > 0);
+}
+
+/*
+ * Called on open of /dev/vmci.
+ */
+static int vmci_host_open(struct inode *inode, struct file *filp)
+{
+ struct vmci_host_dev *vmci_host_dev;
+
+ vmci_host_dev = kzalloc(sizeof(struct vmci_host_dev), GFP_KERNEL);
+ if (vmci_host_dev == NULL)
+ return -ENOMEM;
+
+ vmci_host_dev->ct_type = VMCIOBJ_NOT_SET;
+ mutex_init(&vmci_host_dev->lock);
+ filp->private_data = vmci_host_dev;
+
+ return 0;
+}
+
+/*
+ * Called on close of /dev/vmci, most often when the process
+ * exits.
+ */
+static int vmci_host_close(struct inode *inode, struct file *filp)
+{
+ struct vmci_host_dev *vmci_host_dev = filp->private_data;
+
+ if (vmci_host_dev->ct_type == VMCIOBJ_CONTEXT) {
+ vmci_ctx_destroy(vmci_host_dev->context);
+ vmci_host_dev->context = NULL;
+
+ /*
+ * The number of active contexts is used to track whether any
+ * VMX'en are using the host personality. It is incremented when
+ * a context is created through the IOCTL_VMCI_INIT_CONTEXT
+ * ioctl.
+ */
+ atomic_dec(&vmci_host_active_users);
+ }
+ vmci_host_dev->ct_type = VMCIOBJ_NOT_SET;
+
+ kfree(vmci_host_dev);
+ filp->private_data = NULL;
+ return 0;
+}
+
+/*
+ * This is used to wake up the VMX when a VMCI call arrives, or
+ * to wake up select() or poll() at the next clock tick.
+ */
+static unsigned int vmci_host_poll(struct file *filp, poll_table *wait)
+{
+ struct vmci_host_dev *vmci_host_dev = filp->private_data;
+ struct vmci_ctx *context = vmci_host_dev->context;
+ unsigned int mask = 0;
+
+ if (vmci_host_dev->ct_type == VMCIOBJ_CONTEXT) {
+ /* Check for VMCI calls to this VM context. */
+ if (wait)
+ poll_wait(filp, &context->host_context.wait_queue,
+ wait);
+
+ spin_lock(&context->lock);
+ if (context->pending_datagrams > 0 ||
+ vmci_handle_arr_get_size(
+ context->pending_doorbell_array) > 0) {
+ mask = POLLIN;
+ }
+ spin_unlock(&context->lock);
+ }
+ return mask;
+}
+
+/*
+ * Copies the handles of a handle array into a user buffer, and
+ * returns the new length in userBufferSize. If the copy to the
+ * user buffer fails, the functions still returns VMCI_SUCCESS,
+ * but retval != 0.
+ */
+static int drv_cp_harray_to_user(void __user *user_buf_uva,
+ u64 *user_buf_size,
+ struct vmci_handle_arr *handle_array,
+ int *retval)
+{
+ u32 array_size = 0;
+ struct vmci_handle *handles;
+
+ if (handle_array)
+ array_size = vmci_handle_arr_get_size(handle_array);
+
+ if (array_size * sizeof(*handles) > *user_buf_size)
+ return VMCI_ERROR_MORE_DATA;
+
+ *user_buf_size = array_size * sizeof(*handles);
+ if (*user_buf_size)
+ *retval = copy_to_user(user_buf_uva,
+ vmci_handle_arr_get_handles
+ (handle_array), *user_buf_size);
+
+ return VMCI_SUCCESS;
+}
+
+/*
+ * Sets up a given context for notify to work. Calls drv_map_bool_ptr()
+ * which maps the notify boolean in user VA in kernel space.
+ */
+static int vmci_host_setup_notify(struct vmci_ctx *context,
+ unsigned long uva)
+{
+ struct page *page;
+ int retval;
+
+ if (context->notify_page) {
+ pr_devel("%s: Notify mechanism is already set up.\n", __func__);
+ return VMCI_ERROR_DUPLICATE_ENTRY;
+ }
+
+ if (!access_ok(VERIFY_WRITE, (void __user *)uva, sizeof(bool)))
+ return VMCI_ERROR_GENERIC;
+
+ /*
+ * Lock physical page backing a given user VA.
+ */
+ down_read(¤t->mm->mmap_sem);
+ retval = get_user_pages(current, current->mm,
+ PAGE_ALIGN(uva),
+ 1, 1, 0, &page, NULL);
+ up_read(¤t->mm->mmap_sem);
+ if (retval != 1)
+ return VMCI_ERROR_GENERIC;
+
+ /*
+ * Map the locked page and set up notify pointer.
+ */
+ context->notify = kmap(page) + (uva & (PAGE_SIZE - 1));
+ vmci_ctx_check_signal_notify(context);
+
+ return VMCI_SUCCESS;
+}
+
+static int vmci_host_get_version(struct vmci_host_dev *vmci_host_dev,
+ unsigned int cmd, void __user *uptr)
+{
+ if (cmd == IOCTL_VMCI_VERSION2) {
+ int __user *vptr = uptr;
+ if (get_user(vmci_host_dev->user_version, vptr))
+ return -EFAULT;
+ }
+
+ /*
+ * The basic logic here is:
+ *
+ * If the user sends in a version of 0 tell it our version.
+ * If the user didn't send in a version, tell it our version.
+ * If the user sent in an old version, tell it -its- version.
+ * If the user sent in an newer version, tell it our version.
+ *
+ * The rationale behind telling the caller its version is that
+ * Workstation 6.5 required that VMX and VMCI kernel module were
+ * version sync'd. All new VMX users will be programmed to
+ * handle the VMCI kernel module version.
+ */
+
+ if (vmci_host_dev->user_version > 0 &&
+ vmci_host_dev->user_version < VMCI_VERSION_HOSTQP) {
+ return vmci_host_dev->user_version;
+ }
+
+ return VMCI_VERSION;
+}
+
+#define vmci_ioctl_err(fmt, ...) \
+ pr_devel("%s: " fmt, ioctl_name, ##__VA_ARGS__)
+
+static int vmci_host_do_init_context(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ struct vmci_init_blk init_block;
+ const struct cred *cred;
+ int retval;
+
+ if (copy_from_user(&init_block, uptr, sizeof(init_block))) {
+ vmci_ioctl_err("error reading init block.\n");
+ return -EFAULT;
+ }
+
+ mutex_lock(&vmci_host_dev->lock);
+
+ if (vmci_host_dev->ct_type != VMCIOBJ_NOT_SET) {
+ vmci_ioctl_err("received VMCI init on initialized handle.\n");
+ retval = -EINVAL;
+ goto out;
+ }
+
+ if (init_block.flags & ~VMCI_PRIVILEGE_FLAG_RESTRICTED) {
+ vmci_ioctl_err("unsupported VMCI restriction flag.\n");
+ retval = -EINVAL;
+ goto out;
+ }
+
+ cred = get_current_cred();
+ vmci_host_dev->context = vmci_ctx_create(init_block.cid,
+ init_block.flags, 0,
+ vmci_host_dev->user_version,
+ cred);
+ put_cred(cred);
+ if (IS_ERR(vmci_host_dev->context)) {
+ retval = PTR_ERR(vmci_host_dev->context);
+ vmci_ioctl_err("error initializing context.\n");
+ goto out;
+ }
+
+ /*
+ * Copy cid to userlevel, we do this to allow the VMX
+ * to enforce its policy on cid generation.
+ */
+ init_block.cid = vmci_ctx_get_id(vmci_host_dev->context);
+ if (copy_to_user(uptr, &init_block, sizeof(init_block))) {
+ vmci_ctx_destroy(vmci_host_dev->context);
+ vmci_host_dev->context = NULL;
+ vmci_ioctl_err("error writing init block.\n");
+ retval = -EFAULT;
+ goto out;
+ }
+
+ ASSERT(init_block.cid != VMCI_INVALID_ID);
+ vmci_host_dev->ct_type = VMCIOBJ_CONTEXT;
+ atomic_inc(&vmci_host_active_users);
+
+ retval = 0;
+
+out:
+ mutex_unlock(&vmci_host_dev->lock);
+ return retval;
+}
+
+static int vmci_host_do_send_datagram(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ struct vmci_datagram_snd_rcv_info send_info;
+ struct vmci_datagram *dg = NULL;
+ u32 cid;
+
+ if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) {
+ vmci_ioctl_err("only valid for contexts.\n");
+ return -EINVAL;
+ }
+
+ if (copy_from_user(&send_info, uptr, sizeof(send_info)))
+ return -EFAULT;
+
+ if (send_info.len > VMCI_MAX_DG_SIZE) {
+ vmci_ioctl_err("datagram is too big (size=%d).\n",
+ send_info.len);
+ return -EINVAL;
+ }
+
+ if (send_info.len < sizeof(*dg)) {
+ vmci_ioctl_err("datagram is too small (size=%d).\n",
+ send_info.len);
+ return -EINVAL;
+ }
+
+ dg = kmalloc(send_info.len, GFP_KERNEL);
+ if (!dg) {
+ vmci_ioctl_err("cannot allocate memory to dispatch datagram.\n");
+ return -ENOMEM;
+ }
+
+ if (copy_from_user(dg, (void __user *)(uintptr_t)send_info.addr,
+ send_info.len)) {
+ vmci_ioctl_err("error getting datagram.\n");
+ kfree(dg);
+ return -EFAULT;
+ }
+
+ pr_devel("Datagram dst (handle=0x%x:0x%x) src (handle=0x%x:0x%x), payload (size=%llu bytes).\n",
+ dg->dst.context, dg->dst.resource,
+ dg->src.context, dg->src.resource,
+ (unsigned long long)dg->payload_size);
+
+ /* Get source context id. */
+ cid = vmci_ctx_get_id(vmci_host_dev->context);
+ ASSERT(cid != VMCI_INVALID_ID);
+ send_info.result = vmci_datagram_dispatch(cid, dg, true);
+ kfree(dg);
+
+ return copy_to_user(uptr, &send_info, sizeof(send_info)) ? -EFAULT : 0;
+}
+
+static int vmci_host_do_receive_datagram(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ struct vmci_datagram_snd_rcv_info recv_info;
+ struct vmci_datagram *dg = NULL;
+ int retval;
+ size_t size;
+
+ if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) {
+ vmci_ioctl_err("only valid for contexts.\n");
+ return -EINVAL;
+ }
+
+ if (copy_from_user(&recv_info, uptr, sizeof(recv_info)))
+ return -EFAULT;
+
+ ASSERT(vmci_host_dev->ct_type == VMCIOBJ_CONTEXT);
+ size = recv_info.len;
+ recv_info.result = vmci_ctx_dequeue_datagram(vmci_host_dev->context,
+ &size, &dg);
+
+ if (recv_info.result >= VMCI_SUCCESS) {
+ void __user *ubuf = (void __user *)(uintptr_t)recv_info.addr;
+ retval = copy_to_user(ubuf, dg, VMCI_DG_SIZE(dg));
+ kfree(dg);
+ if (retval != 0)
+ return -EFAULT;
+ }
+
+ return copy_to_user(uptr, &recv_info, sizeof(recv_info)) ? -EFAULT : 0;
+}
+
+static int vmci_host_do_alloc_queuepair(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ struct vmci_handle handle;
+ int vmci_status;
+ int __user *retptr;
+ u32 cid;
+
+ if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) {
+ vmci_ioctl_err("only valid for contexts.\n");
+ return -EINVAL;
+ }
+
+ cid = vmci_ctx_get_id(vmci_host_dev->context);
+
+ if (vmci_host_dev->user_version < VMCI_VERSION_NOVMVM) {
+ struct vmci_qp_alloc_info_vmvm alloc_info;
+ struct vmci_qp_alloc_info_vmvm __user *info = uptr;
+
+ if (copy_from_user(&alloc_info, uptr, sizeof(alloc_info)))
+ return -EFAULT;
+
+ handle = alloc_info.handle;
+ retptr = &info->result;
+
+ vmci_status = vmci_qp_broker_alloc(alloc_info.handle,
+ alloc_info.peer,
+ alloc_info.flags,
+ VMCI_NO_PRIVILEGE_FLAGS,
+ alloc_info.produce_size,
+ alloc_info.consume_size,
+ NULL,
+ vmci_host_dev->context);
+
+ if (vmci_status == VMCI_SUCCESS)
+ vmci_status = VMCI_SUCCESS_QUEUEPAIR_CREATE;
+ } else {
+ struct vmci_qp_alloc_info alloc_info;
+ struct vmci_qp_alloc_info __user *info = uptr;
+ struct vmci_qp_page_store page_store;
+
+ if (copy_from_user(&alloc_info, uptr, sizeof(alloc_info)))
+ return -EFAULT;
+
+ handle = alloc_info.handle;
+ retptr = &info->result;
+
+ page_store.pages = alloc_info.ppn_va;
+ page_store.len = alloc_info.num_ppns;
+
+ vmci_status = vmci_qp_broker_alloc(alloc_info.handle,
+ alloc_info.peer,
+ alloc_info.flags,
+ VMCI_NO_PRIVILEGE_FLAGS,
+ alloc_info.produce_size,
+ alloc_info.consume_size,
+ &page_store,
+ vmci_host_dev->context);
+ }
+
+ if (put_user(vmci_status, retptr)) {
+ if (vmci_status >= VMCI_SUCCESS) {
+ vmci_status = vmci_qp_broker_detach(handle,
+ vmci_host_dev->context);
+ BUG_ON(vmci_status < VMCI_SUCCESS);
+ }
+ return -EFAULT;
+ }
+
+ return 0;
+}
+
+static int vmci_host_do_queuepair_setva(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ struct vmci_qp_set_va_info set_va_info;
+ struct vmci_qp_set_va_info __user *info = uptr;
+ s32 result;
+
+ if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) {
+ vmci_ioctl_err("only valid for contexts.\n");
+ return -EINVAL;
+ }
+
+ if (vmci_host_dev->user_version < VMCI_VERSION_NOVMVM) {
+ vmci_ioctl_err("is not allowed.\n");
+ return -EINVAL;
+ }
+
+ if (copy_from_user(&set_va_info, uptr, sizeof(set_va_info)))
+ return -EFAULT;
+
+ if (set_va_info.va) {
+ /*
+ * VMX is passing down a new VA for the queue
+ * pair mapping.
+ */
+ result = vmci_qp_broker_map(set_va_info.handle,
+ vmci_host_dev->context,
+ set_va_info.va);
+ } else {
+ /*
+ * The queue pair is about to be unmapped by
+ * the VMX.
+ */
+ result = vmci_qp_broker_unmap(set_va_info.handle,
+ vmci_host_dev->context, 0);
+ }
+
+ return put_user(result, &info->result) ? -EFAULT : 0;
+}
+
+static int vmci_host_do_queuepair_setpf(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ struct vmci_qp_page_file_info page_file_info;
+ struct vmci_qp_page_file_info __user *info = uptr;
+ s32 result;
+
+ if (vmci_host_dev->user_version < VMCI_VERSION_HOSTQP ||
+ vmci_host_dev->user_version >= VMCI_VERSION_NOVMVM) {
+ vmci_ioctl_err("not supported on this VMX (version=%d).\n",
+ vmci_host_dev->user_version);
+ return -EINVAL;
+ }
+
+ if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) {
+ vmci_ioctl_err("only valid for contexts.\n");
+ return -EINVAL;
+ }
+
+ if (copy_from_user(&page_file_info, uptr, sizeof(*info)))
+ return -EFAULT;
+
+ /*
+ * Communicate success pre-emptively to the caller. Note that the
+ * basic premise is that it is incumbent upon the caller not to look at
+ * the info.result field until after the ioctl() returns. And then,
+ * only if the ioctl() result indicates no error. We send up the
+ * SUCCESS status before calling SetPageStore() store because failing
+ * to copy up the result code means unwinding the SetPageStore().
+ *
+ * It turns out the logic to unwind a SetPageStore() opens a can of
+ * worms. For example, if a host had created the queue_pair and a
+ * guest attaches and SetPageStore() is successful but writing success
+ * fails, then ... the host has to be stopped from writing (anymore)
+ * data into the queue_pair. That means an additional test in the
+ * VMCI_Enqueue() code path. Ugh.
+ */
+
+ if (put_user(VMCI_SUCCESS, &info->result)) {
+ /*
+ * In this case, we can't write a result field of the
+ * caller's info block. So, we don't even try to
+ * SetPageStore().
+ */
+ return -EFAULT;
+ }
+
+ result = vmci_qp_broker_set_page_store(page_file_info.handle,
+ page_file_info.produce_va,
+ page_file_info.consume_va,
+ vmci_host_dev->context);
+ if (result < VMCI_SUCCESS) {
+ if (put_user(result, &info->result)) {
+ /*
+ * Note that in this case the SetPageStore()
+ * call failed but we were unable to
+ * communicate that to the caller (because the
+ * copy_to_user() call failed). So, if we
+ * simply return an error (in this case
+ * -EFAULT) then the caller will know that the
+ * SetPageStore failed even though we couldn't
+ * put the result code in the result field and
+ * indicate exactly why it failed.
+ *
+ * That says nothing about the issue where we
+ * were once able to write to the caller's info
+ * memory and now can't. Something more
+ * serious is probably going on than the fact
+ * that SetPageStore() didn't work.
+ */
+ return -EFAULT;
+ }
+ }
+
+ return 0;
+}
+
+static int vmci_host_do_qp_detach(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ struct vmci_qp_dtch_info detach_info;
+ struct vmci_qp_dtch_info __user *info = uptr;
+ s32 result;
+
+ if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) {
+ vmci_ioctl_err("only valid for contexts.\n");
+ return -EINVAL;
+ }
+
+ if (copy_from_user(&detach_info, uptr, sizeof(detach_info)))
+ return -EFAULT;
+
+ result = vmci_qp_broker_detach(detach_info.handle,
+ vmci_host_dev->context);
+ if (result == VMCI_SUCCESS &&
+ vmci_host_dev->user_version < VMCI_VERSION_NOVMVM) {
+ result = VMCI_SUCCESS_LAST_DETACH;
+ }
+
+ return put_user(result, &info->result) ? -EFAULT : 0;
+}
+
+static int vmci_host_do_ctx_add_notify(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ struct vmci_ctx_info ar_info;
+ struct vmci_ctx_info __user *info = uptr;
+ s32 result;
+ u32 cid;
+
+ if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) {
+ vmci_ioctl_err("only valid for contexts.\n");
+ return -EINVAL;
+ }
+
+ if (copy_from_user(&ar_info, uptr, sizeof(ar_info)))
+ return -EFAULT;
+
+ cid = vmci_ctx_get_id(vmci_host_dev->context);
+ result = vmci_ctx_add_notification(cid, ar_info.remote_cid);
+
+ return put_user(result, &info->result) ? -EFAULT : 0;
+}
+
+static int vmci_host_do_ctx_remove_notify(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ struct vmci_ctx_info ar_info;
+ struct vmci_ctx_info __user *info = uptr;
+ u32 cid;
+ int result;
+
+ if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) {
+ vmci_ioctl_err("only valid for contexts.\n");
+ return -EINVAL;
+ }
+
+ if (copy_from_user(&ar_info, uptr, sizeof(ar_info)))
+ return -EFAULT;
+
+ cid = vmci_ctx_get_id(vmci_host_dev->context);
+ result = vmci_ctx_remove_notification(cid,
+ ar_info.remote_cid);
+
+ return put_user(result, &info->result) ? -EFAULT : 0;
+}
+
+static int vmci_host_do_ctx_get_cpt_state(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ struct vmci_ctx_chkpt_buf_info get_info;
+ u32 cid;
+ void *cpt_buf;
+ int retval;
+
+ if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) {
+ vmci_ioctl_err("only valid for contexts.\n");
+ return -EINVAL;
+ }
+
+ if (copy_from_user(&get_info, uptr, sizeof(get_info)))
+ return -EFAULT;
+
+ cid = vmci_ctx_get_id(vmci_host_dev->context);
+ get_info.result = vmci_ctx_get_chkpt_state(cid, get_info.cpt_type,
+ &get_info.buf_size, &cpt_buf);
+ if (get_info.result == VMCI_SUCCESS && get_info.buf_size) {
+ void __user *ubuf = (void __user *)(uintptr_t)get_info.cpt_buf;
+ retval = copy_to_user(ubuf, cpt_buf, get_info.buf_size);
+ kfree(cpt_buf);
+
+ if (retval)
+ return -EFAULT;
+ }
+
+ return copy_to_user(uptr, &get_info, sizeof(get_info)) ? -EFAULT : 0;
+}
+
+static int vmci_host_do_ctx_set_cpt_state(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ struct vmci_ctx_chkpt_buf_info set_info;
+ u32 cid;
+ void *cpt_buf;
+ int retval;
+
+ if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) {
+ vmci_ioctl_err("only valid for contexts.\n");
+ return -EINVAL;
+ }
+
+ if (copy_from_user(&set_info, uptr, sizeof(set_info)))
+ return -EFAULT;
+
+ cpt_buf = kmalloc(set_info.buf_size, GFP_KERNEL);
+ if (!cpt_buf) {
+ vmci_ioctl_err("cannot allocate memory to set cpt state (type=%d).\n",
+ set_info.cpt_type);
+ return -ENOMEM;
+ }
+
+ if (copy_from_user(cpt_buf, (void __user *)(uintptr_t)set_info.cpt_buf,
+ set_info.buf_size)) {
+ retval = -EFAULT;
+ goto out;
+ }
+
+ cid = vmci_ctx_get_id(vmci_host_dev->context);
+ set_info.result = vmci_ctx_set_chkpt_state(cid, set_info.cpt_type,
+ set_info.buf_size, cpt_buf);
+
+ retval = copy_to_user(uptr, &set_info, sizeof(set_info)) ? -EFAULT : 0;
+
+out:
+ kfree(cpt_buf);
+ return retval;
+}
+
+static int vmci_host_do_get_context_id(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ u32 __user *u32ptr = uptr;
+
+ return put_user(VMCI_HOST_CONTEXT_ID, u32ptr) ? -EFAULT : 0;
+}
+
+static int vmci_host_do_set_notify(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ struct vmci_set_notify_info notify_info;
+
+ if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) {
+ vmci_ioctl_err("only valid for contexts.\n");
+ return -EINVAL;
+ }
+
+ if (copy_from_user(¬ify_info, uptr, sizeof(notify_info)))
+ return -EFAULT;
+
+ if (notify_info.notify_uva) {
+ notify_info.result =
+ vmci_host_setup_notify(vmci_host_dev->context,
+ notify_info.notify_uva);
+ } else {
+ vmci_ctx_unset_notify(vmci_host_dev->context);
+ notify_info.result = VMCI_SUCCESS;
+ }
+
+ return copy_to_user(uptr, ¬ify_info, sizeof(notify_info)) ?
+ -EFAULT : 0;
+}
+
+static int vmci_host_do_notify_resource(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ struct vmci_dbell_notify_resource_info info;
+ u32 cid;
+
+ if (vmci_host_dev->user_version < VMCI_VERSION_NOTIFY) {
+ vmci_ioctl_err("invalid for current VMX versions.\n");
+ return -EINVAL;
+ }
+
+ if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) {
+ vmci_ioctl_err("only valid for contexts.\n");
+ return -EINVAL;
+ }
+
+ if (copy_from_user(&info, uptr, sizeof(info)))
+ return -EFAULT;
+
+ cid = vmci_ctx_get_id(vmci_host_dev->context);
+
+ switch (info.action) {
+ case VMCI_NOTIFY_RESOURCE_ACTION_NOTIFY:
+ if (info.resource == VMCI_NOTIFY_RESOURCE_DOOR_BELL) {
+ u32 flags = VMCI_NO_PRIVILEGE_FLAGS;
+ info.result = vmci_ctx_notify_dbell(cid, info.handle,
+ flags);
+ } else {
+ info.result = VMCI_ERROR_UNAVAILABLE;
+ }
+ break;
+
+ case VMCI_NOTIFY_RESOURCE_ACTION_CREATE:
+ info.result = vmci_ctx_dbell_create(cid, info.handle);
+ break;
+
+ case VMCI_NOTIFY_RESOURCE_ACTION_DESTROY:
+ info.result = vmci_ctx_dbell_destroy(cid, info.handle);
+ break;
+
+ default:
+ vmci_ioctl_err("got unknown action (action=%d).\n",
+ info.action);
+ info.result = VMCI_ERROR_INVALID_ARGS;
+ }
+
+ return copy_to_user(uptr, &info, sizeof(info)) ? -EFAULT : 0;
+}
+
+static int vmci_host_do_recv_notifications(struct vmci_host_dev *vmci_host_dev,
+ const char *ioctl_name,
+ void __user *uptr)
+{
+ struct vmci_ctx_notify_recv_info info;
+ struct vmci_handle_arr *db_handle_array;
+ struct vmci_handle_arr *qp_handle_array;
+ void __user *ubuf;
+ u32 cid;
+ int retval = 0;
+
+ if (vmci_host_dev->ct_type != VMCIOBJ_CONTEXT) {
+ vmci_ioctl_err("only valid for contexts.\n");
+ return -EINVAL;
+ }
+
+ if (vmci_host_dev->user_version < VMCI_VERSION_NOTIFY) {
+ vmci_ioctl_err("not supported for the current vmx version.\n");
+ return -EINVAL;
+ }
+
+ if (copy_from_user(&info, uptr, sizeof(info)))
+ return -EFAULT;
+
+ if ((info.db_handle_buf_size && !info.db_handle_buf_uva) ||
+ (info.qp_handle_buf_size && !info.qp_handle_buf_uva)) {
+ return -EINVAL;
+ }
+
+ cid = vmci_ctx_get_id(vmci_host_dev->context);
+
+ info.result = vmci_ctx_rcv_notifications_get(cid,
+ &db_handle_array, &qp_handle_array);
+ if (info.result != VMCI_SUCCESS)
+ return copy_to_user(uptr, &info, sizeof(info)) ? -EFAULT : 0;
+
+ ubuf = (void __user *)(uintptr_t)info.db_handle_buf_uva;
+ info.result = drv_cp_harray_to_user(ubuf, &info.db_handle_buf_size,
+ db_handle_array, &retval);
+ if (info.result == VMCI_SUCCESS && !retval) {
+ ubuf = (void __user *)(uintptr_t)info.qp_handle_buf_uva;
+ info.result = drv_cp_harray_to_user(ubuf,
+ &info.qp_handle_buf_size,
+ qp_handle_array, &retval);
+ }
+
+ if (!retval && copy_to_user(uptr, &info, sizeof(info)))
+ retval = -EFAULT;
+
+ vmci_ctx_rcv_notifications_release(cid,
+ db_handle_array, qp_handle_array,
+ info.result == VMCI_SUCCESS && !retval);
+
+ return retval;
+}
+
+
+static long vmci_host_unlocked_ioctl(struct file *filp,
+ unsigned int iocmd, unsigned long ioarg)
+{
+#define VMCI_DO_IOCTL(ioctl_name, ioctl_fn) \
+ case ioctl_name: \
+ return ioctl_fn(vmci_host_dev, \
+ __stringify(ioctl_name), \
+ uptr)
+
+ struct vmci_host_dev *vmci_host_dev = filp->private_data;
+ void __user *uptr = (void __user *)ioarg;
+
+ switch (iocmd) {
+ VMCI_DO_IOCTL(IOCTL_VMCI_INIT_CONTEXT, vmci_host_do_init_context);
+ VMCI_DO_IOCTL(IOCTL_VMCI_DATAGRAM_SEND, vmci_host_do_send_datagram);
+ VMCI_DO_IOCTL(IOCTL_VMCI_DATAGRAM_RECEIVE,
+ vmci_host_do_receive_datagram);
+ VMCI_DO_IOCTL(IOCTL_VMCI_QUEUEPAIR_ALLOC, vmci_host_do_alloc_queuepair);
+ VMCI_DO_IOCTL(IOCTL_VMCI_QUEUEPAIR_SETVA, vmci_host_do_queuepair_setva);
+ VMCI_DO_IOCTL(IOCTL_VMCI_QUEUEPAIR_SETPAGEFILE,
+ vmci_host_do_queuepair_setpf);
+ VMCI_DO_IOCTL(IOCTL_VMCI_QUEUEPAIR_DETACH, vmci_host_do_qp_detach);
+ VMCI_DO_IOCTL(IOCTL_VMCI_CTX_ADD_NOTIFICATION,
+ vmci_host_do_ctx_add_notify);
+ VMCI_DO_IOCTL(IOCTL_VMCI_CTX_REMOVE_NOTIFICATION,
+ vmci_host_do_ctx_remove_notify);
+ VMCI_DO_IOCTL(IOCTL_VMCI_CTX_GET_CPT_STATE,
+ vmci_host_do_ctx_get_cpt_state);
+ VMCI_DO_IOCTL(IOCTL_VMCI_CTX_SET_CPT_STATE,
+ vmci_host_do_ctx_set_cpt_state);
+ VMCI_DO_IOCTL(IOCTL_VMCI_GET_CONTEXT_ID, vmci_host_do_get_context_id);
+ VMCI_DO_IOCTL(IOCTL_VMCI_SET_NOTIFY, vmci_host_do_set_notify);
+ VMCI_DO_IOCTL(IOCTL_VMCI_NOTIFY_RESOURCE, vmci_host_do_notify_resource);
+ VMCI_DO_IOCTL(IOCTL_VMCI_NOTIFICATIONS_RECEIVE,
+ vmci_host_do_recv_notifications);
+
+ case IOCTL_VMCI_VERSION:
+ case IOCTL_VMCI_VERSION2:
+ return vmci_host_get_version(vmci_host_dev, iocmd, uptr);
+
+ default:
+ pr_devel("%s: Unknown ioctl (iocmd=%d).\n", __func__, iocmd);
+ return -EINVAL;
+ }
+
+#undef VMCI_DO_IOCTL
+}
+
+static const struct file_operations vmuser_fops = {
+ .owner = THIS_MODULE,
+ .open = vmci_host_open,
+ .release = vmci_host_close,
+ .poll = vmci_host_poll,
+ .unlocked_ioctl = vmci_host_unlocked_ioctl,
+ .compat_ioctl = vmci_host_unlocked_ioctl,
+};
+
+static struct miscdevice vmci_host_miscdev = {
+ .name = MODULE_NAME,
+ .minor = MISC_DYNAMIC_MINOR,
+ .fops = &vmuser_fops,
+};
+
+int __init vmci_host_init(void)
+{
+ int error;
+
+ host_context = vmci_ctx_create(VMCI_HOST_CONTEXT_ID,
+ VMCI_DEFAULT_PROC_PRIVILEGE_FLAGS,
+ -1, VMCI_VERSION, NULL);
+ if (IS_ERR(host_context)) {
+ error = PTR_ERR(host_context);
+ pr_warn("Failed to initialize VMCIContext (error%d).\n",
+ error);
+ return error;
+ }
+
+ error = misc_register(&vmci_host_miscdev);
+ if (error) {
+ pr_warn("Module registration error (name=%s, major=%d, minor=%d, err=%d).\n",
+ vmci_host_miscdev.name,
+ MISC_MAJOR, vmci_host_miscdev.minor,
+ error);
+ pr_warn("Unable to initialize host personality\n");
+ vmci_ctx_destroy(host_context);
+ return error;
+ }
+
+ pr_info("VMCI host device registered (name=%s, major=%d, minor=%d).\n",
+ vmci_host_miscdev.name, MISC_MAJOR, vmci_host_miscdev.minor);
+
+ vmci_host_device_initialized = true;
+ return 0;
+}
+
+void __exit vmci_host_exit(void)
+{
+ int error;
+
+ vmci_host_device_initialized = false;
+
+ error = misc_deregister(&vmci_host_miscdev);
+ if (error)
+ pr_warn("Error unregistering character device: %d\n", error);
+
+ vmci_ctx_destroy(host_context);
+ vmci_qp_broker_exit();
+
+ pr_debug("VMCI host driver module unloaded\n");
+}
^ permalink raw reply related
* [PATCH 12/12] VMCI: Some header and config files.
From: George Zhang @ 2012-11-07 18:43 UTC (permalink / raw)
To: linux-kernel, georgezhang, virtualization; +Cc: pv-drivers, gregkh
In-Reply-To: <20121107183624.9658.78903.stgit@promb-2n-dhcp175.eng.vmware.com>
VMCI head config patch Adds all the necessary files to enable building of the VMCI
module with the Linux Makefiles and Kconfig systems. Also adds the header files used
for building modules against the driver.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
drivers/misc/Kconfig | 1
drivers/misc/Makefile | 2
drivers/misc/vmw_vmci/Kconfig | 16 +
drivers/misc/vmw_vmci/Makefile | 4
drivers/misc/vmw_vmci/vmci_common_int.h | 34 +
include/linux/vmw_vmci_api.h | 82 +++
include/linux/vmw_vmci_defs.h | 973 +++++++++++++++++++++++++++++++
7 files changed, 1112 insertions(+), 0 deletions(-)
create mode 100644 drivers/misc/vmw_vmci/Kconfig
create mode 100644 drivers/misc/vmw_vmci/Makefile
create mode 100644 drivers/misc/vmw_vmci/vmci_common_int.h
create mode 100644 include/linux/vmw_vmci_api.h
create mode 100644 include/linux/vmw_vmci_defs.h
diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
index 2661f6e..fe38c7a 100644
--- a/drivers/misc/Kconfig
+++ b/drivers/misc/Kconfig
@@ -517,4 +517,5 @@ source "drivers/misc/lis3lv02d/Kconfig"
source "drivers/misc/carma/Kconfig"
source "drivers/misc/altera-stapl/Kconfig"
source "drivers/misc/mei/Kconfig"
+source "drivers/misc/vmw_vmci/Kconfig"
endmenu
diff --git a/drivers/misc/Makefile b/drivers/misc/Makefile
index 456972f..21ed953 100644
--- a/drivers/misc/Makefile
+++ b/drivers/misc/Makefile
@@ -51,3 +51,5 @@ obj-y += carma/
obj-$(CONFIG_USB_SWITCH_FSA9480) += fsa9480.o
obj-$(CONFIG_ALTERA_STAPL) +=altera-stapl/
obj-$(CONFIG_INTEL_MEI) += mei/
+obj-$(CONFIG_MAX8997_MUIC) += max8997-muic.o
+obj-$(CONFIG_VMWARE_VMCI) += vmw_vmci/
diff --git a/drivers/misc/vmw_vmci/Kconfig b/drivers/misc/vmw_vmci/Kconfig
new file mode 100644
index 0000000..55015e7
--- /dev/null
+++ b/drivers/misc/vmw_vmci/Kconfig
@@ -0,0 +1,16 @@
+#
+# VMware VMCI device
+#
+
+config VMWARE_VMCI
+ tristate "VMware VMCI Driver"
+ depends on X86
+ help
+ This is VMware's Virtual Machine Communication Interface. It enables
+ high-speed communication between host and guest in a virtual
+ environment via the VMCI virtual device.
+
+ If unsure, say N.
+
+ To compile this driver as a module, choose M here: the
+ module will be called vmw_vmci.
diff --git a/drivers/misc/vmw_vmci/Makefile b/drivers/misc/vmw_vmci/Makefile
new file mode 100644
index 0000000..4da9893
--- /dev/null
+++ b/drivers/misc/vmw_vmci/Makefile
@@ -0,0 +1,4 @@
+obj-$(CONFIG_VMWARE_VMCI) += vmw_vmci.o
+vmw_vmci-y += vmci_context.o vmci_datagram.o vmci_doorbell.o \
+ vmci_driver.o vmci_event.o vmci_guest.o vmci_handle_array.o \
+ vmci_host.o vmci_queue_pair.o vmci_resource.o vmci_route.o
diff --git a/drivers/misc/vmw_vmci/vmci_common_int.h b/drivers/misc/vmw_vmci/vmci_common_int.h
new file mode 100644
index 0000000..77667ec
--- /dev/null
+++ b/drivers/misc/vmw_vmci/vmci_common_int.h
@@ -0,0 +1,34 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#ifndef _VMCI_COMMONINT_H_
+#define _VMCI_COMMONINT_H_
+
+#include <linux/printk.h>
+
+#define ASSERT(cond) BUG_ON(!(cond))
+
+#define PCI_VENDOR_ID_VMWARE 0x15AD
+#define PCI_DEVICE_ID_VMWARE_VMCI 0x0740
+#define VMCI_DRIVER_VERSION_STRING "1.0.0.0-k"
+#define MODULE_NAME "vmw_vmci"
+
+/* Print magic... whee! */
+#ifdef pr_fmt
+#undef pr_fmt
+#define pr_fmt(fmt) MODULE_NAME ": " fmt
+#endif
+
+#endif /* _VMCI_COMMONINT_H_ */
diff --git a/include/linux/vmw_vmci_api.h b/include/linux/vmw_vmci_api.h
new file mode 100644
index 0000000..193129d
--- /dev/null
+++ b/include/linux/vmw_vmci_api.h
@@ -0,0 +1,82 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#ifndef __VMW_VMCI_API_H__
+#define __VMW_VMCI_API_H__
+
+#include <linux/uidgid.h>
+#include <linux/vmw_vmci_defs.h>
+
+#undef VMCI_KERNEL_API_VERSION
+#define VMCI_KERNEL_API_VERSION_1 1
+#define VMCI_KERNEL_API_VERSION_2 2
+#define VMCI_KERNEL_API_VERSION VMCI_KERNEL_API_VERSION_2
+
+typedef void (vmci_device_shutdown_fn) (void *device_registration,
+ void *user_data);
+
+int vmci_datagram_create_handle(u32 resource_id, u32 flags,
+ vmci_datagram_recv_cb recv_cb,
+ void *client_data,
+ struct vmci_handle *out_handle);
+int vmci_datagram_create_handle_priv(u32 resource_id, u32 flags, u32 priv_flags,
+ vmci_datagram_recv_cb recv_cb,
+ void *client_data,
+ struct vmci_handle *out_handle);
+int vmci_datagram_destroy_handle(struct vmci_handle handle);
+int vmci_datagram_send(struct vmci_datagram *msg);
+int vmci_doorbell_create(struct vmci_handle *handle, u32 flags,
+ u32 priv_flags,
+ vmci_callback notify_cb, void *client_data);
+int vmci_doorbell_destroy(struct vmci_handle handle);
+int vmci_doorbell_notify(struct vmci_handle handle, u32 priv_flags);
+u32 vmci_get_contextid(void);
+bool vmci_is_context_owner(u32 context_id, kuid_t uid);
+
+int vmci_event_subscribe(u32 event,
+ vmci_event_cb callback, void *callback_data,
+ u32 *subid);
+int vmci_event_unsubscribe(u32 subid);
+u32 vmci_context_get_priv_flags(u32 context_id);
+int vmci_qpair_alloc(struct vmci_qp **qpair,
+ struct vmci_handle *handle,
+ u64 produce_qsize,
+ u64 consume_qsize,
+ u32 peer, u32 flags, u32 priv_flags);
+int vmci_qpair_detach(struct vmci_qp **qpair);
+int vmci_qpair_get_produce_indexes(const struct vmci_qp *qpair,
+ u64 *producer_tail,
+ u64 *consumer_head);
+int vmci_qpair_get_consume_indexes(const struct vmci_qp *qpair,
+ u64 *consumer_tail,
+ u64 *producer_head);
+s64 vmci_qpair_produce_free_space(const struct vmci_qp *qpair);
+s64 vmci_qpair_produce_buf_ready(const struct vmci_qp *qpair);
+s64 vmci_qpair_consume_free_space(const struct vmci_qp *qpair);
+s64 vmci_qpair_consume_buf_ready(const struct vmci_qp *qpair);
+ssize_t vmci_qpair_enqueue(struct vmci_qp *qpair,
+ const void *buf, size_t buf_size, int mode);
+ssize_t vmci_qpair_dequeue(struct vmci_qp *qpair,
+ void *buf, size_t buf_size, int mode);
+ssize_t vmci_qpair_peek(struct vmci_qp *qpair, void *buf, size_t buf_size,
+ int mode);
+ssize_t vmci_qpair_enquev(struct vmci_qp *qpair,
+ void *iov, size_t iov_size, int mode);
+ssize_t vmci_qpair_dequev(struct vmci_qp *qpair,
+ void *iov, size_t iov_size, int mode);
+ssize_t vmci_qpair_peekv(struct vmci_qp *qpair, void *iov, size_t iov_size,
+ int mode);
+
+#endif /* !__VMW_VMCI_API_H__ */
diff --git a/include/linux/vmw_vmci_defs.h b/include/linux/vmw_vmci_defs.h
new file mode 100644
index 0000000..40526e8
--- /dev/null
+++ b/include/linux/vmw_vmci_defs.h
@@ -0,0 +1,973 @@
+/*
+ * VMware VMCI Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+ * or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+ * for more details.
+ */
+
+#ifndef _VMW_VMCI_DEF_H_
+#define _VMW_VMCI_DEF_H_
+
+#include <linux/atomic.h>
+
+/* Register offsets. */
+#define VMCI_STATUS_ADDR 0x00
+#define VMCI_CONTROL_ADDR 0x04
+#define VMCI_ICR_ADDR 0x08
+#define VMCI_IMR_ADDR 0x0c
+#define VMCI_DATA_OUT_ADDR 0x10
+#define VMCI_DATA_IN_ADDR 0x14
+#define VMCI_CAPS_ADDR 0x18
+#define VMCI_RESULT_LOW_ADDR 0x1c
+#define VMCI_RESULT_HIGH_ADDR 0x20
+
+/* Max number of devices. */
+#define VMCI_MAX_DEVICES 1
+
+/* Status register bits. */
+#define VMCI_STATUS_INT_ON 0x1
+
+/* Control register bits. */
+#define VMCI_CONTROL_RESET 0x1
+#define VMCI_CONTROL_INT_ENABLE 0x2
+#define VMCI_CONTROL_INT_DISABLE 0x4
+
+/* Capabilities register bits. */
+#define VMCI_CAPS_HYPERCALL 0x1
+#define VMCI_CAPS_GUESTCALL 0x2
+#define VMCI_CAPS_DATAGRAM 0x4
+#define VMCI_CAPS_NOTIFICATIONS 0x8
+
+/* Interrupt Cause register bits. */
+#define VMCI_ICR_DATAGRAM 0x1
+#define VMCI_ICR_NOTIFICATION 0x2
+
+/* Interrupt Mask register bits. */
+#define VMCI_IMR_DATAGRAM 0x1
+#define VMCI_IMR_NOTIFICATION 0x2
+
+/* Interrupt type. */
+enum {
+ VMCI_INTR_TYPE_INTX = 0,
+ VMCI_INTR_TYPE_MSI = 1,
+ VMCI_INTR_TYPE_MSIX = 2,
+};
+
+/* Maximum MSI/MSI-X interrupt vectors in the device. */
+#define VMCI_MAX_INTRS 2
+
+/*
+ * Supported interrupt vectors. There is one for each ICR value above,
+ * but here they indicate the position in the vector array/message ID.
+ */
+enum {
+ VMCI_INTR_DATAGRAM = 0,
+ VMCI_INTR_NOTIFICATION = 1,
+};
+
+/*
+ * A single VMCI device has an upper limit of 128MB on the amount of
+ * memory that can be used for queue pairs.
+ */
+#define VMCI_MAX_GUEST_QP_MEMORY (128 * 1024 * 1024)
+
+/*
+ * Queues with pre-mapped data pages must be small, so that we don't pin
+ * too much kernel memory (especially on vmkernel). We limit a queuepair to
+ * 32 KB, or 16 KB per queue for symmetrical pairs.
+ */
+#define VMCI_MAX_PINNED_QP_MEMORY (32 * 1024)
+
+/*
+ * We have a fixed set of resource IDs available in the VMX.
+ * This allows us to have a very simple implementation since we statically
+ * know how many will create datagram handles. If a new caller arrives and
+ * we have run out of slots we can manually increment the maximum size of
+ * available resource IDs.
+ *
+ * VMCI reserved hypervisor datagram resource IDs.
+ */
+enum {
+ VMCI_RESOURCES_QUERY = 0,
+ VMCI_GET_CONTEXT_ID = 1,
+ VMCI_SET_NOTIFY_BITMAP = 2,
+ VMCI_DOORBELL_LINK = 3,
+ VMCI_DOORBELL_UNLINK = 4,
+ VMCI_DOORBELL_NOTIFY = 5,
+ /*
+ * VMCI_DATAGRAM_REQUEST_MAP and VMCI_DATAGRAM_REMOVE_MAP are
+ * obsoleted by the removal of VM to VM communication.
+ */
+ VMCI_DATAGRAM_REQUEST_MAP = 6,
+ VMCI_DATAGRAM_REMOVE_MAP = 7,
+ VMCI_EVENT_SUBSCRIBE = 8,
+ VMCI_EVENT_UNSUBSCRIBE = 9,
+ VMCI_QUEUEPAIR_ALLOC = 10,
+ VMCI_QUEUEPAIR_DETACH = 11,
+
+ /*
+ * VMCI_VSOCK_VMX_LOOKUP was assigned to 12 for Fusion 3.0/3.1,
+ * WS 7.0/7.1 and ESX 4.1
+ */
+ VMCI_HGFS_TRANSPORT = 13,
+ VMCI_UNITY_PBRPC_REGISTER = 14,
+ VMCI_RPC_PRIVILEGED = 15,
+ VMCI_RPC_UNPRIVILEGED = 16,
+ VMCI_RESOURCE_MAX = 17,
+};
+
+/*
+ * struct vmci_handle - Ownership information structure
+ * @context: The VMX context ID.
+ * @resource: The resource ID (used for locating in resource hash).
+ *
+ * The vmci_handle structure is used to track resources used within
+ * vmw_vmci.
+ */
+struct vmci_handle {
+ u32 context;
+ u32 resource;
+};
+
+typedef u32 vmci_id;
+typedef u32 vmci_privilege_flags;
+typedef u32 vmci_event;
+
+static inline struct vmci_handle VMCI_MAKE_HANDLE(vmci_id cid, vmci_id rid)
+{
+ struct vmci_handle h;
+ h.context = cid;
+ h.resource = rid;
+ return h;
+}
+
+#define VMCI_HANDLE_TO_CONTEXT_ID(_handle) ((_handle).context)
+#define VMCI_HANDLE_TO_RESOURCE_ID(_handle) ((_handle).resource)
+#define VMCI_HANDLE_EQUAL(_h1, _h2) ((_h1).context == (_h2).context && \
+ (_h1).resource == (_h2).resource)
+
+#define VMCI_INVALID_ID ~0
+static const struct vmci_handle VMCI_INVALID_HANDLE = { VMCI_INVALID_ID,
+ VMCI_INVALID_ID
+};
+
+#define VMCI_HANDLE_INVALID(_handle) \
+ VMCI_HANDLE_EQUAL((_handle), VMCI_INVALID_HANDLE)
+
+/*
+ * The below defines can be used to send anonymous requests.
+ * This also indicates that no response is expected.
+ */
+#define VMCI_ANON_SRC_CONTEXT_ID VMCI_INVALID_ID
+#define VMCI_ANON_SRC_RESOURCE_ID VMCI_INVALID_ID
+#define VMCI_ANON_SRC_HANDLE vmci_make_handle(VMCI_ANON_SRC_CONTEXT_ID, \
+ VMCI_ANON_SRC_RESOURCE_ID)
+
+/* The lowest 16 context ids are reserved for internal use. */
+#define VMCI_RESERVED_CID_LIMIT ((u32) 16)
+
+/*
+ * Hypervisor context id, used for calling into hypervisor
+ * supplied services from the VM.
+ */
+#define VMCI_HYPERVISOR_CONTEXT_ID 0
+
+/*
+ * Well-known context id, a logical context that contains a set of
+ * well-known services. This context ID is now obsolete.
+ */
+#define VMCI_WELL_KNOWN_CONTEXT_ID 1
+
+/*
+ * Context ID used by host endpoints.
+ */
+#define VMCI_HOST_CONTEXT_ID 2
+
+#define VMCI_CONTEXT_IS_VM(_cid) (VMCI_INVALID_ID != (_cid) && \
+ (_cid) > VMCI_HOST_CONTEXT_ID)
+
+/*
+ * The VMCI_CONTEXT_RESOURCE_ID is used together with vmci_make_handle to make
+ * handles that refer to a specific context.
+ */
+#define VMCI_CONTEXT_RESOURCE_ID 0
+
+/*
+ * VMCI error codes.
+ */
+enum {
+ VMCI_SUCCESS_QUEUEPAIR_ATTACH = 5,
+ VMCI_SUCCESS_QUEUEPAIR_CREATE = 4,
+ VMCI_SUCCESS_LAST_DETACH = 3,
+ VMCI_SUCCESS_ACCESS_GRANTED = 2,
+ VMCI_SUCCESS_ENTRY_DEAD = 1,
+ VMCI_SUCCESS = 0,
+ VMCI_ERROR_INVALID_RESOURCE = (-1),
+ VMCI_ERROR_INVALID_ARGS = (-2),
+ VMCI_ERROR_NO_MEM = (-3),
+ VMCI_ERROR_DATAGRAM_FAILED = (-4),
+ VMCI_ERROR_MORE_DATA = (-5),
+ VMCI_ERROR_NO_MORE_DATAGRAMS = (-6),
+ VMCI_ERROR_NO_ACCESS = (-7),
+ VMCI_ERROR_NO_HANDLE = (-8),
+ VMCI_ERROR_DUPLICATE_ENTRY = (-9),
+ VMCI_ERROR_DST_UNREACHABLE = (-10),
+ VMCI_ERROR_PAYLOAD_TOO_LARGE = (-11),
+ VMCI_ERROR_INVALID_PRIV = (-12),
+ VMCI_ERROR_GENERIC = (-13),
+ VMCI_ERROR_PAGE_ALREADY_SHARED = (-14),
+ VMCI_ERROR_CANNOT_SHARE_PAGE = (-15),
+ VMCI_ERROR_CANNOT_UNSHARE_PAGE = (-16),
+ VMCI_ERROR_NO_PROCESS = (-17),
+ VMCI_ERROR_NO_DATAGRAM = (-18),
+ VMCI_ERROR_NO_RESOURCES = (-19),
+ VMCI_ERROR_UNAVAILABLE = (-20),
+ VMCI_ERROR_NOT_FOUND = (-21),
+ VMCI_ERROR_ALREADY_EXISTS = (-22),
+ VMCI_ERROR_NOT_PAGE_ALIGNED = (-23),
+ VMCI_ERROR_INVALID_SIZE = (-24),
+ VMCI_ERROR_REGION_ALREADY_SHARED = (-25),
+ VMCI_ERROR_TIMEOUT = (-26),
+ VMCI_ERROR_DATAGRAM_INCOMPLETE = (-27),
+ VMCI_ERROR_INCORRECT_IRQL = (-28),
+ VMCI_ERROR_EVENT_UNKNOWN = (-29),
+ VMCI_ERROR_OBSOLETE = (-30),
+ VMCI_ERROR_QUEUEPAIR_MISMATCH = (-31),
+ VMCI_ERROR_QUEUEPAIR_NOTSET = (-32),
+ VMCI_ERROR_QUEUEPAIR_NOTOWNER = (-33),
+ VMCI_ERROR_QUEUEPAIR_NOTATTACHED = (-34),
+ VMCI_ERROR_QUEUEPAIR_NOSPACE = (-35),
+ VMCI_ERROR_QUEUEPAIR_NODATA = (-36),
+ VMCI_ERROR_BUSMEM_INVALIDATION = (-37),
+ VMCI_ERROR_MODULE_NOT_LOADED = (-38),
+ VMCI_ERROR_DEVICE_NOT_FOUND = (-39),
+ VMCI_ERROR_QUEUEPAIR_NOT_READY = (-40),
+ VMCI_ERROR_WOULD_BLOCK = (-41),
+
+ /* VMCI clients should return error code within this range */
+ VMCI_ERROR_CLIENT_MIN = (-500),
+ VMCI_ERROR_CLIENT_MAX = (-550),
+
+ /* Internal error codes. */
+ VMCI_SHAREDMEM_ERROR_BAD_CONTEXT = (-1000),
+};
+
+/* VMCI reserved events. */
+enum {
+ /* Only applicable to guest endpoints */
+ VMCI_EVENT_CTX_ID_UPDATE = 0,
+
+ /* Applicable to guest and host */
+ VMCI_EVENT_CTX_REMOVED = 1,
+
+ /* Only applicable to guest endpoints */
+ VMCI_EVENT_QP_RESUMED = 2,
+
+ /* Applicable to guest and host */
+ VMCI_EVENT_QP_PEER_ATTACH = 3,
+
+ /* Applicable to guest and host */
+ VMCI_EVENT_QP_PEER_DETACH = 4,
+
+ /*
+ * Applicable to VMX and vmk. On vmk,
+ * this event has the Context payload type.
+ */
+ VMCI_EVENT_MEM_ACCESS_ON = 5,
+
+ /*
+ * Applicable to VMX and vmk. Same as
+ * above for the payload type.
+ */
+ VMCI_EVENT_MEM_ACCESS_OFF = 6,
+ VMCI_EVENT_MAX = 7,
+};
+
+/*
+ * Of the above events, a few are reserved for use in the VMX, and
+ * other endpoints (guest and host kernel) should not use them. For
+ * the rest of the events, we allow both host and guest endpoints to
+ * subscribe to them, to maintain the same API for host and guest
+ * endpoints.
+ */
+#define VMCI_EVENT_VALID_VMX(_event) ((_event) == VMCI_EVENT_MEM_ACCESS_ON || \
+ (_event) == VMCI_EVENT_MEM_ACCESS_OFF)
+
+#define VMCI_EVENT_VALID(_event) ((_event) < VMCI_EVENT_MAX && \
+ !VMCI_EVENT_VALID_VMX(_event))
+
+/* Reserved guest datagram resource ids. */
+#define VMCI_EVENT_HANDLER 0
+
+/*
+ * VMCI coarse-grained privileges (per context or host
+ * process/endpoint. An entity with the restricted flag is only
+ * allowed to interact with the hypervisor and trusted entities.
+ */
+enum {
+ VMCI_NO_PRIVILEGE_FLAGS = 0,
+ VMCI_PRIVILEGE_FLAG_RESTRICTED = 1,
+ VMCI_PRIVILEGE_FLAG_TRUSTED = 2,
+ VMCI_PRIVILEGE_ALL_FLAGS = (VMCI_PRIVILEGE_FLAG_RESTRICTED |
+ VMCI_PRIVILEGE_FLAG_TRUSTED),
+ VMCI_DEFAULT_PROC_PRIVILEGE_FLAGS = VMCI_NO_PRIVILEGE_FLAGS,
+ VMCI_LEAST_PRIVILEGE_FLAGS = VMCI_PRIVILEGE_FLAG_RESTRICTED,
+ VMCI_MAX_PRIVILEGE_FLAGS = VMCI_PRIVILEGE_FLAG_TRUSTED,
+};
+
+/* 0 through VMCI_RESERVED_RESOURCE_ID_MAX are reserved. */
+#define VMCI_RESERVED_RESOURCE_ID_MAX 1023
+
+/*
+ * Driver version.
+ *
+ * Increment major version when you make an incompatible change.
+ * Compatibility goes both ways (old driver with new executable
+ * as well as new driver with old executable).
+ */
+
+/* Never change VMCI_VERSION_SHIFT_WIDTH */
+#define VMCI_VERSION_SHIFT_WIDTH 16
+#define VMCI_MAKE_VERSION(_major, _minor) \
+ ((_major) << VMCI_VERSION_SHIFT_WIDTH | (u16) (_minor))
+
+#define VMCI_VERSION_MAJOR(v) ((u32) (v) >> VMCI_VERSION_SHIFT_WIDTH)
+#define VMCI_VERSION_MINOR(v) ((u16) (v))
+
+/*
+ * VMCI_VERSION is always the current version. Subsequently listed
+ * versions are ways of detecting previous versions of the connecting
+ * application (i.e., VMX).
+ *
+ * VMCI_VERSION_NOVMVM: This version removed support for VM to VM
+ * communication.
+ *
+ * VMCI_VERSION_NOTIFY: This version introduced doorbell notification
+ * support.
+ *
+ * VMCI_VERSION_HOSTQP: This version introduced host end point support
+ * for hosted products.
+ *
+ * VMCI_VERSION_PREHOSTQP: This is the version prior to the adoption of
+ * support for host end-points.
+ *
+ * VMCI_VERSION_PREVERS2: This fictional version number is intended to
+ * represent the version of a VMX which doesn't call into the driver
+ * with ioctl VERSION2 and thus doesn't establish its version with the
+ * driver.
+ */
+
+#define VMCI_VERSION VMCI_VERSION_NOVMVM
+#define VMCI_VERSION_NOVMVM VMCI_MAKE_VERSION(11, 0)
+#define VMCI_VERSION_NOTIFY VMCI_MAKE_VERSION(10, 0)
+#define VMCI_VERSION_HOSTQP VMCI_MAKE_VERSION(9, 0)
+#define VMCI_VERSION_PREHOSTQP VMCI_MAKE_VERSION(8, 0)
+#define VMCI_VERSION_PREVERS2 VMCI_MAKE_VERSION(1, 0)
+
+#define VMCI_SOCKETS_MAKE_VERSION(_p) \
+ ((((_p)[0] & 0xFF) << 24) | (((_p)[1] & 0xFF) << 16) | ((_p)[2]))
+
+/*
+ * Linux defines _IO* macros, but the core kernel code ignore the encoded
+ * ioctl value. It is up to individual drivers to decode the value (for
+ * example to look at the size of a structure to determine which version
+ * of a specific command should be used) or not (which is what we
+ * currently do, so right now the ioctl value for a given command is the
+ * command itself).
+ *
+ * Hence, we just define the IOCTL_VMCI_foo values directly, with no
+ * intermediate IOCTLCMD_ representation.
+ */
+#define IOCTLCMD(_cmd) IOCTL_VMCI_ ## _cmd
+
+enum {
+ /*
+ * We need to bracket the range of values used for ioctls,
+ * because x86_64 Linux forces us to explicitly register ioctl
+ * handlers by value for handling 32 bit ioctl syscalls.
+ * Hence FIRST and LAST. Pick something for FIRST that
+ * doesn't collide with vmmon (2001+).
+ */
+ IOCTLCMD(FIRST) = 1951,
+ IOCTLCMD(VERSION) = IOCTLCMD(FIRST),
+
+ /* BEGIN VMCI */
+ IOCTLCMD(INIT_CONTEXT),
+
+ /*
+ * The following two were used for process and datagram
+ * process creation. They are not used anymore and reserved
+ * for future use. They will fail if issued.
+ */
+ IOCTLCMD(RESERVED1),
+ IOCTLCMD(RESERVED2),
+
+ /*
+ * The following used to be for shared memory. It is now
+ * unused and and is reserved for future use. It will fail if
+ * issued.
+ */
+ IOCTLCMD(RESERVED3),
+
+ /*
+ * The follwoing three were also used to be for shared
+ * memory. An old WS6 user-mode client might try to use them
+ * with the new driver, but since we ensure that only contexts
+ * created by VMX'en of the appropriate version
+ * (VMCI_VERSION_NOTIFY or VMCI_VERSION_NEWQP) or higher use
+ * these ioctl, everything is fine.
+ */
+ IOCTLCMD(QUEUEPAIR_SETVA),
+ IOCTLCMD(NOTIFY_RESOURCE),
+ IOCTLCMD(NOTIFICATIONS_RECEIVE),
+ IOCTLCMD(VERSION2),
+ IOCTLCMD(QUEUEPAIR_ALLOC),
+ IOCTLCMD(QUEUEPAIR_SETPAGEFILE),
+ IOCTLCMD(QUEUEPAIR_DETACH),
+ IOCTLCMD(DATAGRAM_SEND),
+ IOCTLCMD(DATAGRAM_RECEIVE),
+ IOCTLCMD(DATAGRAM_REQUEST_MAP),
+ IOCTLCMD(DATAGRAM_REMOVE_MAP),
+ IOCTLCMD(CTX_ADD_NOTIFICATION),
+ IOCTLCMD(CTX_REMOVE_NOTIFICATION),
+ IOCTLCMD(CTX_GET_CPT_STATE),
+ IOCTLCMD(CTX_SET_CPT_STATE),
+ IOCTLCMD(GET_CONTEXT_ID),
+ IOCTLCMD(LAST),
+ /* END VMCI */
+
+ /*
+ * VMCI Socket IOCTLS are defined next and go from
+ * IOCTLCMD(LAST) (1972) to 1990. VMware reserves a range of
+ * 4 ioctls for VMCI Sockets to grow. We cannot reserve many
+ * ioctls here since we are close to overlapping with vmmon
+ * ioctls (2001+). Define a meta-ioctl if running out of this
+ * binary space.
+ */
+ IOCTLCMD(SOCKETS_FIRST) = IOCTLCMD(LAST),
+ IOCTLCMD(SOCKETS_VERSION) = IOCTLCMD(SOCKETS_FIRST),
+ IOCTLCMD(SOCKETS_BIND),
+ IOCTLCMD(SOCKETS_SET_SYMBOLS),
+ IOCTLCMD(SOCKETS_CONNECT),
+ /*
+ * The next two values are public (vmci_sockets.h) and cannot be
+ * changed. That means the number of values above these cannot be
+ * changed either unless the base index (specified below) is updated
+ * accordingly.
+ */
+ IOCTLCMD(SOCKETS_GET_AF_VALUE),
+ IOCTLCMD(SOCKETS_GET_LOCAL_CID),
+ IOCTLCMD(SOCKETS_GET_SOCK_NAME),
+ IOCTLCMD(SOCKETS_GET_SOCK_OPT),
+ IOCTLCMD(SOCKETS_GET_VM_BY_NAME),
+ IOCTLCMD(SOCKETS_IOCTL),
+ IOCTLCMD(SOCKETS_LISTEN),
+ IOCTLCMD(SOCKETS_RECV),
+ IOCTLCMD(SOCKETS_RECV_FROM),
+ IOCTLCMD(SOCKETS_SELECT),
+ IOCTLCMD(SOCKETS_SEND),
+ IOCTLCMD(SOCKETS_SEND_TO),
+ IOCTLCMD(SOCKETS_SET_SOCK_OPT),
+ IOCTLCMD(SOCKETS_SHUTDOWN),
+ IOCTLCMD(SOCKETS_SOCKET), /* 1990 on Linux. */
+
+ IOCTLCMD(SOCKETS_LAST) = 1994, /* 1994 on Linux. */
+
+ /*
+ * The VSockets ioctls occupy the block above. We define a
+ * new range of VMCI ioctls to maintain binary compatibility
+ * between the user land and the kernel driver. Careful,
+ * vmmon ioctls start from 2001, so this means we can add only
+ * 4 new VMCI ioctls. Define a meta-ioctl if running out of
+ * this binary space.
+ */
+ IOCTLCMD(FIRST2),
+ IOCTLCMD(SET_NOTIFY) = IOCTLCMD(FIRST2), /* 1995 on Linux. */
+ IOCTLCMD(LAST2),
+};
+
+/* Clean up helper macros */
+#undef IOCTLCMD
+
+/*
+ * struct vmci_queue_header - VMCI Queue Header information.
+ *
+ * A Queue cannot stand by itself as designed. Each Queue's header
+ * contains a pointer into itself (the producer_tail) and into its peer
+ * (consumer_head). The reason for the separation is one of
+ * accessibility: Each end-point can modify two things: where the next
+ * location to enqueue is within its produce_q (producer_tail); and
+ * where the next dequeue location is in its consume_q (consumer_head).
+ *
+ * An end-point cannot modify the pointers of its peer (guest to
+ * guest; NOTE that in the host both queue headers are mapped r/w).
+ * But, each end-point needs read access to both Queue header
+ * structures in order to determine how much space is used (or left)
+ * in the Queue. This is because for an end-point to know how full
+ * its produce_q is, it needs to use the consumer_head that points into
+ * the produce_q but -that- consumer_head is in the Queue header for
+ * that end-points consume_q.
+ *
+ * Thoroughly confused? Sorry.
+ *
+ * producer_tail: the point to enqueue new entrants. When you approach
+ * a line in a store, for example, you walk up to the tail.
+ *
+ * consumer_head: the point in the queue from which the next element is
+ * dequeued. In other words, who is next in line is he who is at the
+ * head of the line.
+ *
+ * Also, producer_tail points to an empty byte in the Queue, whereas
+ * consumer_head points to a valid byte of data (unless producer_tail ==
+ * consumer_head in which case consumer_head does not point to a valid
+ * byte of data).
+ *
+ * For a queue of buffer 'size' bytes, the tail and head pointers will be in
+ * the range [0, size-1].
+ *
+ * If produce_q_header->producer_tail == consume_q_header->consumer_head
+ * then the produce_q is empty.
+ */
+struct vmci_queue_header {
+ /* All fields are 64bit and aligned. */
+ struct vmci_handle handle; /* Identifier. */
+ atomic64_t producer_tail; /* Offset in this queue. */
+ atomic64_t consumer_head; /* Offset in peer queue. */
+};
+
+/*
+ * struct vmci_datagram - Base struct for vmci datagrams.
+ * @dst: A vmci_handle that tracks the destination of the datagram.
+ * @src: A vmci_handle that tracks the source of the datagram.
+ * @payload_size: The size of the payload.
+ *
+ * vmci_datagram structs are used when sending vmci datagrams. They include
+ * the necessary source and destination information to properly route
+ * the information along with the size of the package.
+ */
+struct vmci_datagram {
+ struct vmci_handle dst;
+ struct vmci_handle src;
+ u64 payload_size;
+};
+
+/*
+ * Second flag is for creating a well-known handle instead of a per context
+ * handle. Next flag is for deferring datagram delivery, so that the
+ * datagram callback is invoked in a delayed context (not interrupt context).
+ */
+#define VMCI_FLAG_DG_NONE 0
+#define VMCI_FLAG_WELLKNOWN_DG_HND 0x1
+#define VMCI_FLAG_ANYCID_DG_HND 0x2
+#define VMCI_FLAG_DG_DELAYED_CB 0x4
+
+/*
+ * Maximum supported size of a VMCI datagram for routable datagrams.
+ * Datagrams going to the hypervisor are allowed to be larger.
+ */
+#define VMCI_MAX_DG_SIZE (17 * 4096)
+#define VMCI_MAX_DG_PAYLOAD_SIZE (VMCI_MAX_DG_SIZE - \
+ sizeof(struct vmci_datagram))
+#define VMCI_DG_PAYLOAD(_dg) (void *)((char *)(_dg) + \
+ sizeof(struct vmci_datagram))
+#define VMCI_DG_HEADERSIZE sizeof(struct vmci_datagram)
+#define VMCI_DG_SIZE(_dg) (VMCI_DG_HEADERSIZE + (size_t)(_dg)->payload_size)
+#define VMCI_DG_SIZE_ALIGNED(_dg) ((VMCI_DG_SIZE(_dg) + 7) & (~((size_t) 0x7)))
+#define VMCI_MAX_DATAGRAM_QUEUE_SIZE (VMCI_MAX_DG_SIZE * 2)
+
+struct vmci_event_payload_qp {
+ struct vmci_handle handle; /* queue_pair handle. */
+ u32 peer_id; /* Context id of attaching/detaching VM. */
+ u32 _pad;
+};
+
+/* Flags for VMCI queue_pair API. */
+enum {
+ /* Fail alloc if QP not created by peer. */
+ VMCI_QPFLAG_ATTACH_ONLY = 1 << 0,
+
+ /* Only allow attaches from local context. */
+ VMCI_QPFLAG_LOCAL = 1 << 1,
+
+ /* Host won't block when guest is quiesced. */
+ VMCI_QPFLAG_NONBLOCK = 1 << 2,
+
+ /* Pin data pages in ESX. Used with NONBLOCK */
+ VMCI_QPFLAG_PINNED = 1 << 3,
+
+ /* Update the following flag when adding new flags. */
+ VMCI_QP_ALL_FLAGS = (VMCI_QPFLAG_ATTACH_ONLY | VMCI_QPFLAG_LOCAL |
+ VMCI_QPFLAG_NONBLOCK | VMCI_QPFLAG_PINNED),
+
+ /* Convenience flags */
+ VMCI_QP_ASYMM = (VMCI_QPFLAG_NONBLOCK | VMCI_QPFLAG_PINNED),
+ VMCI_QP_ASYMM_PEER = (VMCI_QPFLAG_ATTACH_ONLY | VMCI_QP_ASYMM),
+};
+
+/*
+ * We allow at least 1024 more event datagrams from the hypervisor past the
+ * normally allowed datagrams pending for a given context. We define this
+ * limit on event datagrams from the hypervisor to guard against DoS attack
+ * from a malicious VM which could repeatedly attach to and detach from a queue
+ * pair, causing events to be queued at the destination VM. However, the rate
+ * at which such events can be generated is small since it requires a VM exit
+ * and handling of queue pair attach/detach call at the hypervisor. Event
+ * datagrams may be queued up at the destination VM if it has interrupts
+ * disabled or if it is not draining events for some other reason. 1024
+ * datagrams is a grossly conservative estimate of the time for which
+ * interrupts may be disabled in the destination VM, but at the same time does
+ * not exacerbate the memory pressure problem on the host by much (size of each
+ * event datagram is small).
+ */
+#define VMCI_MAX_DATAGRAM_AND_EVENT_QUEUE_SIZE \
+ (VMCI_MAX_DATAGRAM_QUEUE_SIZE + \
+ 1024 * (sizeof(struct vmci_datagram) + \
+ sizeof(struct vmci_event_data_max)))
+
+/*
+ * Struct used for querying, via VMCI_RESOURCES_QUERY, the availability of
+ * hypervisor resources. Struct size is 16 bytes. All fields in struct are
+ * aligned to their natural alignment.
+ */
+struct vmci_resource_query_hdr {
+ struct vmci_datagram hdr;
+ u32 num_resources;
+ u32 _padding;
+};
+
+/*
+ * Convenience struct for negotiating vectors. Must match layout of
+ * VMCIResourceQueryHdr minus the struct vmci_datagram header.
+ */
+struct vmci_resource_query_msg {
+ u32 num_resources;
+ u32 _padding;
+ u32 resources[1];
+};
+
+/*
+ * The maximum number of resources that can be queried using
+ * VMCI_RESOURCE_QUERY is 31, as the result is encoded in the lower 31
+ * bits of a positive return value. Negative values are reserved for
+ * errors.
+ */
+#define VMCI_RESOURCE_QUERY_MAX_NUM 31
+
+/* Maximum size for the VMCI_RESOURCE_QUERY request. */
+#define VMCI_RESOURCE_QUERY_MAX_SIZE \
+ (sizeof(struct vmci_resource_query_hdr) + \
+ sizeof(u32) * VMCI_RESOURCE_QUERY_MAX_NUM)
+
+/*
+ * Struct used for setting the notification bitmap. All fields in
+ * struct are aligned to their natural alignment.
+ */
+struct vmci_notify_bm_set_msg {
+ struct vmci_datagram hdr;
+ u32 bitmap_ppn;
+ u32 _pad;
+};
+
+/*
+ * Struct used for linking a doorbell handle with an index in the
+ * notify bitmap. All fields in struct are aligned to their natural
+ * alignment.
+ */
+struct vmci_doorbell_link_msg {
+ struct vmci_datagram hdr;
+ struct vmci_handle handle;
+ u64 notify_idx;
+};
+
+/*
+ * Struct used for unlinking a doorbell handle from an index in the
+ * notify bitmap. All fields in struct are aligned to their natural
+ * alignment.
+ */
+struct vmci_doorbell_unlink_msg {
+ struct vmci_datagram hdr;
+ struct vmci_handle handle;
+};
+
+/*
+ * Struct used for generating a notification on a doorbell handle. All
+ * fields in struct are aligned to their natural alignment.
+ */
+struct vmci_doorbell_notify_msg {
+ struct vmci_datagram hdr;
+ struct vmci_handle handle;
+};
+
+/*
+ * This struct is used to contain data for events. Size of this struct is a
+ * multiple of 8 bytes, and all fields are aligned to their natural alignment.
+ */
+struct vmci_event_data {
+ u32 event; /* 4 bytes. */
+ u32 _pad;
+ /* Event payload is put here. */
+};
+
+/*
+ * Define the different VMCI_EVENT payload data types here. All structs must
+ * be a multiple of 8 bytes, and fields must be aligned to their natural
+ * alignment.
+ */
+struct vmci_event_payld_ctx {
+ u32 context_id; /* 4 bytes. */
+ u32 _pad;
+};
+
+struct vmci_event_payld_qp {
+ struct vmci_handle handle; /* queue_pair handle. */
+ u32 peer_id; /* Context id of attaching/detaching VM. */
+ u32 _pad;
+};
+
+/*
+ * We define the following struct to get the size of the maximum event
+ * data the hypervisor may send to the guest. If adding a new event
+ * payload type above, add it to the following struct too (inside the
+ * union).
+ */
+struct vmci_event_data_max {
+ struct vmci_event_data event_data;
+ union {
+ struct vmci_event_payld_ctx context_payload;
+ struct vmci_event_payld_qp qp_payload;
+ } ev_data_payload;
+};
+
+/*
+ * Struct used for VMCI_EVENT_SUBSCRIBE/UNSUBSCRIBE and
+ * VMCI_EVENT_HANDLER messages. Struct size is 32 bytes. All fields
+ * in struct are aligned to their natural alignment.
+ */
+struct vmci_event_msg {
+ struct vmci_datagram hdr;
+
+ /* Has event type and payload. */
+ struct vmci_event_data event_data;
+
+ /* Payload gets put here. */
+};
+
+/*
+ * Structs used for queue_pair alloc and detach messages. We align fields of
+ * these structs to 64bit boundaries.
+ */
+struct vmci_qp_alloc_msg {
+ struct vmci_datagram hdr;
+ struct vmci_handle handle;
+ u32 peer;
+ u32 flags;
+ u64 produce_size;
+ u64 consume_size;
+ u64 num_ppns;
+
+ /* List of PPNs placed here. */
+};
+
+struct vmci_qp_detach_msg {
+ struct vmci_datagram hdr;
+ struct vmci_handle handle;
+};
+
+/* VMCI Doorbell API. */
+#define VMCI_FLAG_DELAYED_CB 0x01
+
+typedef void (*vmci_callback) (void *client_data);
+
+/*
+ * struct vmci_qp - A vmw_vmci queue pair handle.
+ *
+ * This structure is used as a handle to a queue pair created by
+ * VMCI. It is intentionally left opaque to clients.
+ */
+struct vmci_qp;
+
+/* Callback needed for correctly waiting on events. */
+typedef int (*vmci_datagram_recv_cb) (void *client_data,
+ struct vmci_datagram *msg);
+
+/* VMCI Event API. */
+typedef void (*vmci_event_cb) (u32 sub_id, const struct vmci_event_data *ed,
+ void *client_data);
+
+/*
+ * We use the following inline function to access the payload data
+ * associated with an event data.
+ */
+static inline const void *
+vmci_event_data_const_payload(const struct vmci_event_data *ev_data)
+{
+ return (const char *)ev_data + sizeof(*ev_data);
+}
+
+static inline void *vmci_event_data_payload(struct vmci_event_data *ev_data)
+{
+ return (void *)vmci_event_data_const_payload(ev_data);
+}
+
+/*
+ * Helper to add a given offset to a head or tail pointer. Wraps the
+ * value of the pointer around the max size of the queue.
+ */
+static inline void vmci_qp_add_pointer(atomic64_t *var,
+ size_t add,
+ u64 size)
+{
+ u64 new_val = atomic64_read(var);
+
+ if (new_val >= size - add)
+ new_val -= size;
+
+ new_val += add;
+
+ atomic64_set(var, new_val);
+}
+
+/*
+ * Helper routine to get the Producer Tail from the supplied queue.
+ */
+static inline u64
+vmci_q_header_producer_tail(const struct vmci_queue_header *q_header)
+{
+ struct vmci_queue_header *qh = (struct vmci_queue_header *)q_header;
+ return atomic64_read(&qh->producer_tail);
+}
+
+/*
+ * Helper routine to get the Consumer Head from the supplied queue.
+ */
+static inline u64
+vmci_q_header_consumer_head(const struct vmci_queue_header *q_header)
+{
+ struct vmci_queue_header *qh = (struct vmci_queue_header *)q_header;
+ return atomic64_read(&qh->consumer_head);
+}
+
+/*
+ * Helper routine to increment the Producer Tail. Fundamentally,
+ * vmci_qp_add_pointer() is used to manipulate the tail itself.
+ */
+static inline void
+vmci_q_header_add_producer_tail(struct vmci_queue_header *q_header,
+ size_t add,
+ u64 queue_size)
+{
+ vmci_qp_add_pointer(&q_header->producer_tail, add, queue_size);
+}
+
+/*
+ * Helper routine to increment the Consumer Head. Fundamentally,
+ * vmci_qp_add_pointer() is used to manipulate the head itself.
+ */
+static inline void
+vmci_q_header_add_consumer_head(struct vmci_queue_header *q_header,
+ size_t add,
+ u64 queue_size)
+{
+ vmci_qp_add_pointer(&q_header->consumer_head, add, queue_size);
+}
+
+/*
+ * Helper routine for getting the head and the tail pointer for a queue.
+ * Both the VMCIQueues are needed to get both the pointers for one queue.
+ */
+static inline void
+vmci_q_header_get_pointers(const struct vmci_queue_header *produce_q_header,
+ const struct vmci_queue_header *consume_q_header,
+ u64 *producer_tail,
+ u64 *consumer_head)
+{
+ if (producer_tail)
+ *producer_tail = vmci_q_header_producer_tail(produce_q_header);
+
+ if (consumer_head)
+ *consumer_head = vmci_q_header_consumer_head(consume_q_header);
+}
+
+static inline void vmci_q_header_init(struct vmci_queue_header *q_header,
+ const struct vmci_handle handle)
+{
+ q_header->handle = handle;
+ atomic64_set(&q_header->producer_tail, 0);
+ atomic64_set(&q_header->consumer_head, 0);
+}
+
+/*
+ * Finds available free space in a produce queue to enqueue more
+ * data or reports an error if queue pair corruption is detected.
+ */
+static s64
+vmci_q_header_free_space(const struct vmci_queue_header *produce_q_header,
+ const struct vmci_queue_header *consume_q_header,
+ const u64 produce_q_size)
+{
+ u64 tail;
+ u64 head;
+ u64 free_space;
+
+ tail = vmci_q_header_producer_tail(produce_q_header);
+ head = vmci_q_header_consumer_head(consume_q_header);
+
+ if (tail >= produce_q_size || head >= produce_q_size)
+ return VMCI_ERROR_INVALID_SIZE;
+
+ /*
+ * Deduct 1 to avoid tail becoming equal to head which causes
+ * ambiguity. If head and tail are equal it means that the
+ * queue is empty.
+ */
+ if (tail >= head)
+ free_space = produce_q_size - (tail - head) - 1;
+ else
+ free_space = head - tail - 1;
+
+ return free_space;
+}
+
+/*
+ * vmci_q_header_free_space() does all the heavy lifting of
+ * determing the number of free bytes in a Queue. This routine,
+ * then subtracts that size from the full size of the Queue so
+ * the caller knows how many bytes are ready to be dequeued.
+ * Results:
+ * On success, available data size in bytes (up to MAX_INT64).
+ * On failure, appropriate error code.
+ */
+static inline s64
+vmci_q_header_buf_ready(const struct vmci_queue_header *consume_q_header,
+ const struct vmci_queue_header *produce_q_header,
+ const u64 consume_q_size)
+{
+ s64 free_space;
+
+ free_space = vmci_q_header_free_space(consume_q_header,
+ produce_q_header, consume_q_size);
+ if (free_space < VMCI_SUCCESS)
+ return free_space;
+
+ return consume_q_size - free_space - 1;
+}
+
+static inline struct vmci_handle vmci_make_handle(u32 cid, u32 rid)
+{
+ struct vmci_handle h;
+
+ h.context = cid;
+ h.resource = rid;
+
+ return h;
+}
+
+#endif /* _VMW_VMCI_DEF_H_ */
^ permalink raw reply related
* [PATCH 0/6] VSOCK for Linux upstreaming
From: George Zhang @ 2012-11-07 18:45 UTC (permalink / raw)
To: netdev, linux-kernel, georgezhang, virtualization
Cc: pv-drivers, gregkh, davem
* * *
This series of VSOCK linux upstreaming patches include latest udpate from
VMware.
Summary of changes:
- Fix some new sparse issues.
- Remove some unneeded casts for VMCI.
- add more __user annotations for VMCI.
- Remove kernel version-specific bits from vSockets.
* * *
In an effort to improve the out-of-the-box experience with Linux
kernels for VMware users, VMware is working on readying the Virtual
Machine Communication Interface (vmw_vmci) and VMCI Sockets (VSOCK)
(vmw_vsock) kernel modules for inclusion in the Linux kernel. The
purpose of this post is to acquire feedback on the vmw_vsock kernel
module. The vmw_vmci kernel module has been presented in an early post.
* * *
VMCI Sockets allows virtual machines to communicate with host kernel
modules and the VMware hypervisors. VMCI Sockets kernel module has
dependency on VMCI kernel module. User level applications both in
a virtual machine and on the host can use vmw_vmci through VMCI
Sockets API which facilitates fast and efficient communication
between guest virtual machines and their host. A socket
address family designed to be compatible with UDP and TCP at the
interface level. Today, VMCI and VMCI Sockets are used by the VMware
shared folders (HGFS) and various VMware Tools components inside the
guest for zero-config, network-less access to VMware host services. In
addition to this, VMware's users are using VMCI Sockets for various
applications, where network access of the virtual machine is
restricted or non-existent. Examples of this are VMs communicating
with device proxies for proprietary hardware running as host
applications and automated testing of applications running within
virtual machines.
The VMware VMCI Sockets are similar to other socket types, like
Berkeley UNIX socket interface. The VMCI sockets module supports
both connection-oriented stream sockets like TCP, and connectionless
datagram sockets like UDP. The VSOCK protocol family is defined as
"AF_VSOCK" and the socket operations split for SOCK_DGRAM and
SOCK_STREAM.
For additional information about the use of VMCI and in particular
VMCI Sockets, please refer to the VMCI Socket Programming Guide
available at https://www.vmware.com/support/developer/vmci-sdk/.
---
George Zhang (6):
VSOCK: vsock protocol implementation.
VSOCK: vsock address implementaion.
VSOCK: notification implementation.
VSOCK: statistics implementation.
VSOCK: utility functions.
VSOCK: header and config files.
include/linux/socket.h | 4
net/Kconfig | 1
net/Makefile | 1
net/vmw_vsock/Kconfig | 14
net/vmw_vsock/Makefile | 4
net/vmw_vsock/af_vsock.c | 4135 +++++++++++++++++++++++++++++++++++
net/vmw_vsock/af_vsock.h | 179 ++
net/vmw_vsock/notify.c | 984 ++++++++
net/vmw_vsock/notify.h | 130 +
net/vmw_vsock/notify_qstate.c | 627 +++++
net/vmw_vsock/stats.c | 37
net/vmw_vsock/stats.h | 219 ++
net/vmw_vsock/util.c | 626 +++++
net/vmw_vsock/util.h | 312 +++
net/vmw_vsock/vmci_sockets.h | 517 ++++
net/vmw_vsock/vmci_sockets_packet.h | 106 +
net/vmw_vsock/vsock_addr.c | 264 ++
net/vmw_vsock/vsock_addr.h | 40
net/vmw_vsock/vsock_common.h | 130 +
net/vmw_vsock/vsock_packet.h | 131 +
net/vmw_vsock/vsock_version.h | 29
21 files changed, 8489 insertions(+), 1 deletions(-)
create mode 100644 net/vmw_vsock/Kconfig
create mode 100644 net/vmw_vsock/Makefile
create mode 100644 net/vmw_vsock/af_vsock.c
create mode 100644 net/vmw_vsock/af_vsock.h
create mode 100644 net/vmw_vsock/notify.c
create mode 100644 net/vmw_vsock/notify.h
create mode 100644 net/vmw_vsock/notify_qstate.c
create mode 100644 net/vmw_vsock/stats.c
create mode 100644 net/vmw_vsock/stats.h
create mode 100644 net/vmw_vsock/util.c
create mode 100644 net/vmw_vsock/util.h
create mode 100644 net/vmw_vsock/vmci_sockets.h
create mode 100644 net/vmw_vsock/vmci_sockets_packet.h
create mode 100644 net/vmw_vsock/vsock_addr.c
create mode 100644 net/vmw_vsock/vsock_addr.h
create mode 100644 net/vmw_vsock/vsock_common.h
create mode 100644 net/vmw_vsock/vsock_packet.h
create mode 100644 net/vmw_vsock/vsock_version.h
--
Signature
^ permalink raw reply
* [PATCH 1/6] VSOCK: vsock protocol implementation.
From: George Zhang @ 2012-11-07 18:45 UTC (permalink / raw)
To: netdev, linux-kernel, georgezhang, virtualization
Cc: pv-drivers, gregkh, davem
In-Reply-To: <20121107184340.10612.3372.stgit@promb-2n-dhcp175.eng.vmware.com>
VSOCK linux socket module for VMCI Sockets protocol family.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
net/vmw_vsock/af_vsock.c | 4135 ++++++++++++++++++++++++++++++++++++++++++++++
net/vmw_vsock/af_vsock.h | 179 ++
2 files changed, 4314 insertions(+), 0 deletions(-)
create mode 100644 net/vmw_vsock/af_vsock.c
create mode 100644 net/vmw_vsock/af_vsock.h
diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
new file mode 100644
index 0000000..4d31c39
--- /dev/null
+++ b/net/vmw_vsock/af_vsock.c
@@ -0,0 +1,4135 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2007-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * af_vsock.c --
+ *
+ * Linux socket module for the VMCI Sockets protocol family.
+ */
+
+/*
+ * Implementation notes:
+ *
+ * - There are two kinds of sockets: those created by user action (such as
+ * calling socket(2)) and those created by incoming connection request packets.
+ *
+ * - There are two "global" tables, one for bound sockets (sockets that have
+ * specified an address that they are responsible for) and one for connected
+ * sockets (sockets that have established a connection with another socket).
+ * These tables are "global" in that all sockets on the system are placed
+ * within them. - Note, though, that the bound table contains an extra entry
+ * for a list of unbound sockets and SOCK_DGRAM sockets will always remain in
+ * that list. The bound table is used solely for lookup of sockets when packets
+ * are received and that's not necessary for SOCK_DGRAM sockets since we create
+ * a datagram handle for each and need not perform a lookup. Keeping SOCK_DGRAM
+ * sockets out of the bound hash buckets will reduce the chance of collisions
+ * when looking for SOCK_STREAM sockets and prevents us from having to check the
+ * socket type in the hash table lookups.
+ *
+ * - Sockets created by user action will either be "client" sockets that
+ * initiate a connection or "server" sockets that listen for connections; we do
+ * not support simultaneous connects (two "client" sockets connecting).
+ *
+ * - "Server" sockets are referred to as listener sockets throughout this
+ * implementation because they are in the SS_LISTEN state. When a connection
+ * request is received (the second kind of socket mentioned above), we create a
+ * new socket and refer to it as a pending socket. These pending sockets are
+ * placed on the pending connection list of the listener socket. When future
+ * packets are received for the address the listener socket is bound to, we
+ * check if the source of the packet is from one that has an existing pending
+ * connection. If it does, we process the packet for the pending socket. When
+ * that socket reaches the connected state, it is removed from the listener
+ * socket's pending list and enqueued in the listener socket's accept queue.
+ * Callers of accept(2) will accept connected sockets from the listener socket's
+ * accept queue. If the socket cannot be accepted for some reason then it is
+ * marked rejected. Once the connection is accepted, it is owned by the user
+ * process and the responsibility for cleanup falls with that user process.
+ *
+ * - It is possible that these pending sockets will never reach the connected
+ * state; in fact, we may never receive another packet after the connection
+ * request. Because of this, we must schedule a cleanup function to run in the
+ * future, after some amount of time passes where a connection should have been
+ * established. This function ensures that the socket is off all lists so it
+ * cannot be retrieved, then drops all references to the socket so it is cleaned
+ * up (sock_put() -> sk_free() -> our sk_destruct implementation). Note this
+ * function will also cleanup rejected sockets, those that reach the connected
+ * state but leave it before they have been accepted.
+ *
+ * - Sockets created by user action will be cleaned up when the user process
+ * calls close(2), causing our release implementation to be called. Our release
+ * implementation will perform some cleanup then drop the last reference so our
+ * sk_destruct implementation is invoked. Our sk_destruct implementation will
+ * perform additional cleanup that's common for both types of sockets.
+ *
+ * - A socket's reference count is what ensures that the structure won't be
+ * freed. Each entry in a list (such as the "global" bound and connected tables
+ * and the listener socket's pending list and connected queue) ensures a
+ * reference. When we defer work until process context and pass a socket as our
+ * argument, we must ensure the reference count is increased to ensure the
+ * socket isn't freed before the function is run; the deferred function will
+ * then drop the reference.
+ */
+
+#include <linux/types.h>
+
+#define EXPORT_SYMTAB
+#include <linux/kmod.h>
+#include <linux/socket.h>
+#include <linux/net.h>
+#include <linux/skbuff.h>
+#include <linux/miscdevice.h>
+#include <linux/poll.h>
+#include <linux/smp.h>
+#include <linux/bitops.h>
+#include <linux/list.h>
+#include <linux/wait.h>
+#include <linux/init.h>
+#include <linux/io.h>
+
+#include <linux/module.h>
+#include <linux/unistd.h>
+#include <linux/stddef.h> /* for NULL */
+#include <net/sock.h>
+#include <linux/kernel.h>
+#include <linux/workqueue.h>
+#include <linux/mutex.h>
+/*
+ * Include linux/cred.h via linux/sched.h - it is not nice, but as cpp does not
+ * have #ifexist...
+ */
+#include <linux/sched.h>
+
+#include "af_vsock.h"
+#include "stats.h"
+#include "util.h"
+#include "vsock_version.h"
+
+/*
+ * Prototypes
+ */
+
+/* Internal functions. */
+static bool vsock_vmci_proto_to_notify_struct(struct sock *sk,
+ vsock_proto_version * proto,
+ bool old_pkt_proto);
+static int vsock_vmci_recv_dgram_cb(void *data, struct vmci_datagram *dg);
+static int vsock_vmci_recv_stream_cb(void *data, struct vmci_datagram *dg);
+static void vsock_vmci_peer_attach_cb(vmci_id sub_id,
+ const struct vmci_event_data *ed,
+ void *client_data);
+static void vsock_vmci_peer_detach_cb(vmci_id sub_id,
+ const struct vmci_event_data *ed,
+ void *client_data);
+static void vsock_vmci_recv_pkt_work(struct work_struct *work);
+static int vsock_vmci_recv_listen(struct sock *sk, vsock_packet *pkt);
+static int vsock_vmci_recv_connecting_server(struct sock *sk,
+ struct sock *pending,
+ vsock_packet *pkt);
+static int vsock_vmci_recv_connecting_client(struct sock *sk,
+ vsock_packet *pkt);
+static int vsock_vmci_recv_connecting_client_negotiate(struct sock *sk,
+ vsock_packet *pkt);
+static int vsock_vmci_recv_connecting_client_invalid(struct sock *sk,
+ vsock_packet *pkt);
+static int vsock_vmci_recv_connected(struct sock *sk, vsock_packet *pkt);
+static int __vsock_vmci_bind(struct sock *sk, struct sockaddr_vm *addr);
+static struct sock *__vsock_vmci_create(struct net *net,
+ struct socket *sock,
+ struct sock *parent, gfp_t priority,
+ unsigned short type);
+static int vsock_vmci_register_with_vmci(void);
+static void vsock_vmci_unregister_with_vmci(void);
+
+/* Socket operations. */
+static void vsock_vmci_sk_destruct(struct sock *sk);
+static int vsock_vmci_queue_rcv_skb(struct sock *sk, struct sk_buff *skb);
+static int vsock_vmci_release(struct socket *sock);
+static int vsock_vmci_bind(struct socket *sock,
+ struct sockaddr *addr, int addr_len);
+static int vsock_vmci_dgram_connect(struct socket *sock,
+ struct sockaddr *addr, int addr_len,
+ int flags);
+static int vsock_vmci_stream_connect(struct socket *sock, struct sockaddr *addr,
+ int addr_len, int flags);
+static int vsock_vmci_accept(struct socket *sock, struct socket *newsock,
+ int flags);
+static int vsock_vmci_getname(struct socket *sock, struct sockaddr *addr,
+ int *addr_len, int peer);
+static unsigned int vsock_vmci_poll(struct file *file, struct socket *sock,
+ poll_table *wait);
+static int vsock_vmci_listen(struct socket *sock, int backlog);
+static int vsock_vmci_shutdown(struct socket *sock, int mode);
+
+typedef unsigned int vsock_setsockopt_len_type;
+static int vsock_vmci_stream_setsockopt(struct socket *sock, int level,
+ int optname, char __user *optval,
+ vsock_setsockopt_len_type optlen);
+
+static int vsock_vmci_stream_getsockopt(struct socket *sock, int level,
+ int optname, char __user *optval,
+ int __user *optlen);
+
+static int vsock_vmci_dgram_sendmsg(struct kiocb *kiocb,
+ struct socket *sock, struct msghdr *msg,
+ size_t len);
+static int vsock_vmci_dgram_recvmsg(struct kiocb *kiocb, struct socket *sock,
+ struct msghdr *msg, size_t len, int flags);
+static int vsock_vmci_stream_sendmsg(struct kiocb *kiocb, struct socket *sock,
+ struct msghdr *msg, size_t len);
+static int vsock_vmci_stream_recvmsg(struct kiocb *kiocb, struct socket *sock,
+ struct msghdr *msg, size_t len, int flags);
+
+static int vsock_vmci_create(struct net *net,
+ struct socket *sock, int protocol, int kern);
+
+/*
+ * Variables.
+ */
+
+/* Protocol family. */
+static struct proto vsock_vmci_proto = {
+ .name = "AF_VMCI",
+ .owner = THIS_MODULE,
+ .obj_size = sizeof(vsock_vmci_sock),
+};
+
+static const struct net_proto_family vsock_vmci_family_ops = {
+ .family = AF_VSOCK,
+ .create = vsock_vmci_create,
+ .owner = THIS_MODULE,
+};
+
+/* Socket operations, split for DGRAM and STREAM sockets. */
+static const struct proto_ops vsock_vmci_dgram_ops = {
+ .family = PF_VSOCK,
+ .owner = THIS_MODULE,
+ .release = vsock_vmci_release,
+ .bind = vsock_vmci_bind,
+ .connect = vsock_vmci_dgram_connect,
+ .socketpair = sock_no_socketpair,
+ .accept = sock_no_accept,
+ .getname = vsock_vmci_getname,
+ .poll = vsock_vmci_poll,
+ .ioctl = sock_no_ioctl,
+ .listen = sock_no_listen,
+ .shutdown = vsock_vmci_shutdown,
+ .setsockopt = sock_no_setsockopt,
+ .getsockopt = sock_no_getsockopt,
+ .sendmsg = vsock_vmci_dgram_sendmsg,
+ .recvmsg = vsock_vmci_dgram_recvmsg,
+ .mmap = sock_no_mmap,
+ .sendpage = sock_no_sendpage,
+};
+
+static const struct proto_ops vsock_vmci_stream_ops = {
+ .family = PF_VSOCK,
+ .owner = THIS_MODULE,
+ .release = vsock_vmci_release,
+ .bind = vsock_vmci_bind,
+ .connect = vsock_vmci_stream_connect,
+ .socketpair = sock_no_socketpair,
+ .accept = vsock_vmci_accept,
+ .getname = vsock_vmci_getname,
+ .poll = vsock_vmci_poll,
+ .ioctl = sock_no_ioctl,
+ .listen = vsock_vmci_listen,
+ .shutdown = vsock_vmci_shutdown,
+ .setsockopt = vsock_vmci_stream_setsockopt,
+ .getsockopt = vsock_vmci_stream_getsockopt,
+ .sendmsg = vsock_vmci_stream_sendmsg,
+ .recvmsg = vsock_vmci_stream_recvmsg,
+ .mmap = sock_no_mmap,
+ .sendpage = sock_no_sendpage,
+};
+
+typedef struct vsock_recv_pkt_info {
+ struct work_struct work;
+ struct sock *sk;
+ vsock_packet pkt;
+} vsock_recv_pkt_info;
+
+static struct vmci_handle vmci_stream_handle = { VMCI_INVALID_ID,
+ VMCI_INVALID_ID };
+
+static vmci_id qp_resumed_sub_id = VMCI_INVALID_ID;
+
+static int PROTOCOL_OVERRIDE = -1;
+
+/*
+ * Netperf benchmarks have shown significant throughput improvements when the
+ * QP size is bumped from 64k to 256k. These measurements were taken during the
+ * K/L.next timeframe. Give users better performance by default.
+ */
+#define VSOCK_DEFAULT_QP_SIZE_MIN 128
+#define VSOCK_DEFAULT_QP_SIZE 262144
+#define VSOCK_DEFAULT_QP_SIZE_MAX 262144
+
+/*
+ * The default peer timeout indicates how long we will wait for a peer response
+ * to a control message.
+ */
+#define VSOCK_DEFAULT_CONNECT_TIMEOUT (2 * HZ)
+
+#define LOG_PACKET(_pkt)
+
+/**
+ * vsock_vmci_old_proto_override --
+ *
+ * Check to see if the user has asked us to override all sockets to use the
+ * vsock notify protocol.
+ *
+ * Results: true if there is a protocol override in effect. - old_pkt_proto is
+ * true the original protocol should be used. False if there is no override in
+ * effect.
+ *
+ * Side effects: None.
+ */
+
+static bool vsock_vmci_old_proto_override(bool *old_pkt_proto)
+{
+ if (PROTOCOL_OVERRIDE != -1) {
+ if (PROTOCOL_OVERRIDE == 0)
+ *old_pkt_proto = true;
+ else
+ *old_pkt_proto = false;
+
+ pr_info("Proto override in use.\n");
+ return true;
+ }
+
+ return false;
+}
+
+/*
+ * vsock_vmci_proto_to_notify_struct --
+ *
+ * Given a particular notify protocol version, setup the socket's notify struct
+ * correctly.
+ *
+ * Results: true on success, false otherwise.
+ *
+ * Side effects: None.
+ */
+
+static bool
+vsock_vmci_proto_to_notify_struct(struct sock *sk,
+ vsock_proto_version *proto,
+ bool old_pkt_proto)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ if (old_pkt_proto) {
+ if (*proto != VSOCK_PROTO_INVALID) {
+ pr_err("Can't set both an old and new protocol\n");
+ return false;
+ }
+ vsk->notify_ops = &vsock_vmci_notify_pkt_ops;
+ goto exit;
+ }
+
+ switch (*proto) {
+ case VSOCK_PROTO_PKT_ON_NOTIFY:
+ vsk->notify_ops = &vsock_vmci_notify_pkt_q_state_ops;
+ break;
+ default:
+ pr_err("Unknown notify protocol version\n");
+ return false;
+ }
+
+exit:
+ NOTIFYCALL(vsk, socket_init, sk);
+ return true;
+}
+
+/*
+ * vsock_vmci_new_proto_supported_versions
+ *
+ * Gets the supported REQUEST2/NEGOTIATE2 vsock protocol versions.
+ *
+ * Results: Either 1 specific protocol version (override mode) or
+ * VSOCK_PROTO_ALL_SUPPORTED.
+ *
+ * Side effects: None.
+ */
+
+static vsock_proto_version vsock_vmci_new_proto_supported_versions(void)
+{
+ if (PROTOCOL_OVERRIDE != -1)
+ return PROTOCOL_OVERRIDE;
+
+ return VSOCK_PROTO_ALL_SUPPORTED;
+}
+
+/*
+ * VSockSocket_Trusted --
+ *
+ * We allow two kinds of sockets to communicate with a restricted VM: 1)
+ * trusted sockets 2) sockets from applications running as the same user as the
+ * VM (this is only true for the host side and only when using hosted products)
+ *
+ * Results: true if trusted communication is allowed to peer_cid, false
+ * otherwise.
+ *
+ * Side effects: None.
+ */
+
+static bool vsock_vmci_trusted(vsock_vmci_sock *vsock, vmci_id peer_cid)
+{
+ return vsock->trusted ||
+ vmci_is_context_owner(peer_cid, vsock->owner->uid);
+}
+
+/*
+ * VSockSocket_AllowDgram --
+ *
+ * We allow sending datagrams to and receiving datagrams from a restricted VM
+ * only if it is trusted as described in vsock_vmci_trusted.
+ *
+ * Results: true if datagram communication is allowed to peer_cid, false
+ * otherwise.
+ *
+ * Side effects: None.
+ */
+
+static bool vsock_vmci_allow_dgram(vsock_vmci_sock *vsock, vmci_id peer_cid)
+{
+ if (vsock->cached_peer != peer_cid) {
+ vsock->cached_peer = peer_cid;
+ if (!vsock_vmci_trusted(vsock, peer_cid) &&
+ (vmci_context_get_priv_flags(peer_cid) &
+ VMCI_PRIVILEGE_FLAG_RESTRICTED)) {
+ vsock->cached_peer_allow_dgram = false;
+ } else {
+ vsock->cached_peer_allow_dgram = true;
+ }
+ }
+
+ return vsock->cached_peer_allow_dgram;
+}
+
+/*
+ * vmci_sock_get_local_c_id --
+ *
+ * Kernel interface that allows external kernel modules to get the current VMCI
+ * context id. This version of the function is exported to kernel clients and
+ * should not change.
+ *
+ * Results: The context id on success, a negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+int vmci_sock_get_local_c_id(void)
+{
+ /* FIXME: needed? */
+ return vmci_get_context_id();
+}
+EXPORT_SYMBOL_GPL(vmci_sock_get_local_c_id);
+
+/*
+ * Helper functions.
+ */
+
+/*
+ * vsock_vmci_queue_pair_alloc --
+ *
+ * Allocates or attaches to a queue pair. Tries to register with trusted status
+ * if requested but does not fail if the queuepair could not be allocate as
+ * trusted (running in the guest)
+ *
+ * Results: 0 on success. A VSock error on error.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_queue_pair_alloc(struct vmci_qp **qpair,
+ struct vmci_handle *handle,
+ u64 produce_size,
+ u64 consume_size,
+ vmci_id peer, u32 flags, bool trusted)
+{
+ int err = 0;
+
+ if (trusted) {
+ /*
+ * Try to allocate our queue pair as trusted. This will only
+ * work if vsock is running in the host.
+ */
+
+ err = vmci_qpair_alloc(qpair, handle, produce_size,
+ consume_size,
+ peer, flags,
+ VMCI_PRIVILEGE_FLAG_TRUSTED);
+ if (err != VMCI_ERROR_NO_ACCESS)
+ goto out;
+
+ }
+
+ err = vmci_qpair_alloc(qpair, handle, produce_size, consume_size,
+ peer, flags, VMCI_NO_PRIVILEGE_FLAGS);
+out:
+ if (err < 0) {
+ pr_err("Could not attach to queue pair with %d\n",
+ err);
+ err = vsock_vmci_error_to_vsock_error(err);
+ }
+
+ return err;
+}
+
+/*
+ * vsock_vmci_datagram_create_hnd --
+ *
+ * Creates a datagram handle. Tries to register with trusted status but does
+ * not fail if the handler could not be allocated as trusted (running in the
+ * guest).
+ *
+ * Results: 0 on success. A VMCI error on error.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_datagram_create_hnd(vmci_id resource_id,
+ u32 flags,
+ vmci_datagram_recv_cb recv_cb,
+ void *client_data,
+ struct vmci_handle *out_handle)
+{
+ int err = 0;
+
+ /*
+ * Try to allocate our datagram handler as trusted. This will only work
+ * if vsock is running in the host.
+ */
+
+ err = vmci_datagram_create_handle_priv(resource_id, flags,
+ VMCI_PRIVILEGE_FLAG_TRUSTED,
+ recv_cb,
+ client_data, out_handle);
+
+ if (err == VMCI_ERROR_NO_ACCESS)
+ err = vmci_datagram_create_handle(resource_id, flags,
+ recv_cb, client_data,
+ out_handle);
+
+ return err;
+}
+
+/*
+ * vsock_vmci_recv_dgram_cb --
+ *
+ * VMCI Datagram receive callback. This function is used specifically for
+ * SOCK_DGRAM sockets.
+ *
+ * This is invoked as part of a tasklet that's scheduled when the VMCI
+ * interrupt fires. This is run in bottom-half context and if it ever needs to
+ * sleep it should defer that work to a work queue.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: An sk_buff is created and queued with this socket.
+ */
+
+static int vsock_vmci_recv_dgram_cb(void *data, struct vmci_datagram *dg)
+{
+ struct sock *sk;
+ size_t size;
+ struct sk_buff *skb;
+ vsock_vmci_sock *vsk;
+
+ ASSERT(dg->payload_size <= VMCI_MAX_DG_PAYLOAD_SIZE);
+
+ sk = (struct sock *)data;
+
+ /*
+ * This handler is privileged when this module is running on the host.
+ * We will get datagrams from all endpoints (even VMs that are in a
+ * restricted context). If we get one from a restricted context then
+ * the destination socket must be trusted.
+ *
+ * NOTE: We access the socket struct without holding the lock here.
+ * This is ok because the field we are interested is never modified
+ * outside of the create and destruct socket functions.
+ */
+ vsk = vsock_sk(sk);
+ if (!vsock_vmci_allow_dgram(vsk, VMCI_HANDLE_TO_CONTEXT_ID(dg->src)))
+ return VMCI_ERROR_NO_ACCESS;
+
+ size = VMCI_DG_SIZE(dg);
+
+ /*
+ * Attach the packet to the socket's receive queue as an sk_buff.
+ */
+ skb = alloc_skb(size, GFP_ATOMIC);
+ if (skb) {
+ /* sk_receive_skb() will do a sock_put(), so hold here. */
+ sock_hold(sk);
+ skb_put(skb, size);
+ memcpy(skb->data, dg, size);
+ sk_receive_skb(sk, skb, 0);
+ }
+
+ return VMCI_SUCCESS;
+}
+
+/*
+ * vsock_vmci_recv_stream_cb --
+ *
+ * VMCI stream receive callback for control datagrams. This function is used
+ * specifically for SOCK_STREAM sockets.
+ *
+ * This is invoked as part of a tasklet that's scheduled when the VMCI
+ * interrupt fires. This is run in bottom-half context but it defers most of
+ * its work to the packet handling work queue.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int vsock_vmci_recv_stream_cb(void *data, struct vmci_datagram *dg)
+{
+ struct sock *sk;
+ struct sockaddr_vm dst;
+ struct sockaddr_vm src;
+ vsock_packet *pkt;
+ vsock_vmci_sock *vsk;
+ bool bh_process_pkt;
+ int err;
+
+ ASSERT(dg->payload_size <= VMCI_MAX_DG_PAYLOAD_SIZE);
+
+ sk = NULL;
+ err = VMCI_SUCCESS;
+ bh_process_pkt = false;
+
+ /*
+ * Ignore incoming packets from contexts without sockets, or resources
+ * that aren't vsock implementations.
+ */
+
+ if (!vsock_addr_socket_context_stream
+ (VMCI_HANDLE_TO_CONTEXT_ID(dg->src))
+ || VSOCK_PACKET_RID != VMCI_HANDLE_TO_RESOURCE_ID(dg->src)) {
+ return VMCI_ERROR_NO_ACCESS;
+ }
+
+ if (VMCI_DG_SIZE(dg) < sizeof *pkt)
+ /* Drop datagrams that do not contain full VSock packets. */
+ return VMCI_ERROR_INVALID_ARGS;
+
+ pkt = (vsock_packet *) dg;
+
+ LOG_PACKET(pkt);
+
+ /*
+ * Find the socket that should handle this packet. First we look for a
+ * connected socket and if there is none we look for a socket bound to
+ * the destintation address.
+ */
+ vsock_addr_init(&src, VMCI_HANDLE_TO_CONTEXT_ID(pkt->dg.src),
+ pkt->src_port);
+
+ vsock_addr_init(&dst, VMCI_HANDLE_TO_CONTEXT_ID(pkt->dg.dst),
+ pkt->dst_port);
+
+ sk = vsock_vmci_find_connected_socket(&src, &dst);
+ if (!sk) {
+ sk = vsock_vmci_find_bound_socket(&dst);
+ if (!sk) {
+ /*
+ * We could not find a socket for this specified
+ * address. If this packet is a RST, we just drop it.
+ * If it is another packet, we send a RST. Note that
+ * we do not send a RST reply to RSTs so that we do not
+ * continually send RSTs between two endpoints.
+ *
+ * Note that since this is a reply, dst is src and src
+ * is dst.
+ */
+ if (VSOCK_SEND_RESET_BH(&dst, &src, pkt) < 0)
+ pr_err("unable to send reset.\n");
+
+ err = VMCI_ERROR_NOT_FOUND;
+ goto out;
+ }
+ }
+
+ /*
+ * If the received packet type is beyond all types known to this
+ * implementation, reply with an invalid message. Hopefully this will
+ * help when implementing backwards compatibility in the future.
+ */
+ if (pkt->type >= VSOCK_PACKET_TYPE_MAX) {
+ VSOCK_SEND_INVALID_BH(&dst, &src);
+ err = VMCI_ERROR_INVALID_ARGS;
+ goto out;
+ }
+
+ /*
+ * This handler is privileged when this module is running on the host.
+ * We will get datagram connect requests from all endpoints (even VMs
+ * that are in a restricted context). If we get one from a restricted
+ * context then the destination socket must be trusted.
+ *
+ * NOTE: We access the socket struct without holding the lock here.
+ * This is ok because the field we are interested is never modified
+ * outside of the create and destruct socket functions.
+ */
+ vsk = vsock_sk(sk);
+ if (!vsock_vmci_allow_dgram
+ (vsk, VMCI_HANDLE_TO_CONTEXT_ID(pkt->dg.src))) {
+ err = VMCI_ERROR_NO_ACCESS;
+ goto out;
+ }
+
+ /*
+ * We do most everything in a work queue, but let's fast path the
+ * notification of reads and writes to help data transfer performance.
+ * We can only do this if there is no process context code executing
+ * for this socket since that may change the state.
+ */
+ bh_lock_sock(sk);
+
+ if (!sock_owned_by_user(sk) && sk->sk_state == SS_CONNECTED)
+ NOTIFYCALL(vsk, handle_notify_pkt, sk, pkt, true, &dst, &src,
+ &bh_process_pkt);
+
+ bh_unlock_sock(sk);
+
+ if (!bh_process_pkt) {
+ vsock_recv_pkt_info *recv_pkt_info;
+
+ recv_pkt_info = kmalloc(sizeof *recv_pkt_info, GFP_ATOMIC);
+ if (!recv_pkt_info) {
+ if (VSOCK_SEND_RESET_BH(&dst, &src, pkt) < 0)
+ pr_err("unable to send reset\n");
+
+ err = VMCI_ERROR_NO_MEM;
+ goto out;
+ }
+
+ recv_pkt_info->sk = sk;
+ memcpy(&recv_pkt_info->pkt, pkt, sizeof recv_pkt_info->pkt);
+ INIT_WORK(&recv_pkt_info->work, vsock_vmci_recv_pkt_work);
+
+ schedule_work(&recv_pkt_info->work);
+ /*
+ * Clear sk so that the reference count incremented by one of
+ * the Find functions above is not decremented below. We need
+ * that reference count for the packet handler we've scheduled
+ * to run.
+ */
+ sk = NULL;
+ }
+
+out:
+ if (sk)
+ sock_put(sk);
+
+ return err;
+}
+
+/*
+ * vsock_vmci_peer_attach_cb --
+ *
+ * Invoked when a peer attaches to a queue pair.
+ *
+ * Right now this does not do anything.
+ *
+ * Results: None.
+ *
+ * Side effects: May modify socket state and signal socket.
+ */
+
+static void vsock_vmci_peer_attach_cb(vmci_id sub_id,
+ const struct vmci_event_data *e_data,
+ void *client_data)
+{
+ struct sock *sk = client_data;
+ const struct vmci_event_payload_qp *e_payload;
+ vsock_vmci_sock *vsk;
+
+ e_payload = vmci_event_data_const_payload(e_data);
+
+ vsk = vsock_sk(sk);
+
+ /*
+ * We don't ask for delayed CBs when we subscribe to this event (we
+ * pass 0 as flags to vmci_event_subscribe()). VMCI makes no
+ * guarantees in that case about what context we might be running in,
+ * so it could be BH or process, blockable or non-blockable. So we
+ * need to account for all possible contexts here.
+ */
+ local_bh_disable();
+ bh_lock_sock(sk);
+
+ /*
+ * XXX This is lame, we should provide a way to lookup sockets by
+ * qp_handle.
+ */
+ if (VMCI_HANDLE_EQUAL(vsk->qp_handle, e_payload->handle)) {
+ /*
+ * XXX This doesn't do anything, but in the future we may want
+ * to set a flag here to verify the attach really did occur and
+ * we weren't just sent a datagram claiming it was.
+ */
+ goto out;
+ }
+
+out:
+ bh_unlock_sock(sk);
+ local_bh_enable();
+}
+
+/*
+ *
+ * vsock_vmci_handle_detach --
+ *
+ * Perform the work necessary when the peer has detached.
+ *
+ * Note that this assumes the socket lock is held.
+ *
+ * Results: None.
+ *
+ * Side effects: The socket's and its peer's shutdown mask will be set
+ * appropriately, and any callers waiting on this socket will be awoken.
+ */
+
+static void vsock_vmci_handle_detach(struct sock *sk)
+{
+ vsock_vmci_sock *vsk;
+
+ vsk = vsock_sk(sk);
+ if (!VMCI_HANDLE_INVALID(vsk->qp_handle)) {
+ sock_set_flag(sk, SOCK_DONE);
+
+ /*
+ * On a detach the peer will not be sending or receiving
+ * anymore.
+ */
+ vsk->peer_shutdown = SHUTDOWN_MASK;
+
+ /*
+ * We should not be sending anymore since the peer won't be
+ * there to receive, but we can still receive if there is data
+ * left in our consume queue.
+ */
+ if (vsock_vmci_stream_has_data(vsk) <= 0) {
+ if (sk->sk_state == SS_CONNECTING) {
+ /*
+ * The peer may detach from a queue pair while
+ * we are still in the connecting state, i.e.,
+ * if the peer VM is killed after attaching to
+ * a queue pair, but before we complete the
+ * handshake. In that case, we treat the detach
+ * event like a reset.
+ */
+
+ sk->sk_state = SS_UNCONNECTED;
+ sk->sk_err = ECONNRESET;
+ sk->sk_error_report(sk);
+ return;
+ }
+ sk->sk_state = SS_UNCONNECTED;
+ }
+ sk->sk_state_change(sk);
+ }
+}
+
+/*
+ * vsock_vmci_peer_detach_cb --
+ *
+ * Invoked when a peer detaches from a queue pair.
+ *
+ * Results: None.
+ *
+ * Side effects: May modify socket state and signal socket.
+ */
+
+static void vsock_vmci_peer_detach_cb(vmci_id sub_id,
+ const struct vmci_event_data *e_data,
+ void *client_data)
+{
+ struct sock *sk = client_data;
+ const struct vmci_event_payload_qp *e_payload;
+ vsock_vmci_sock *vsk;
+
+ e_payload = vmci_event_data_const_payload(e_data);
+ vsk = vsock_sk(sk);
+ if (VMCI_HANDLE_INVALID(e_payload->handle))
+ return;
+
+ /* Same rules for locking as for peer_attach_cb(). */
+ local_bh_disable();
+ bh_lock_sock(sk);
+
+ /*
+ * XXX This is lame, we should provide a way to lookup sockets by
+ * qp_handle.
+ */
+ if (VMCI_HANDLE_EQUAL(vsk->qp_handle, e_payload->handle))
+ vsock_vmci_handle_detach(sk);
+
+ bh_unlock_sock(sk);
+ local_bh_enable();
+}
+
+/*
+ * vsock_vmci_qp_resumed_cb --
+ *
+ * Invoked when a VM is resumed. We must mark all connected stream sockets as
+ * detached.
+ *
+ * Results: None.
+ *
+ * Side effects: May modify socket state and signal socket.
+ */
+
+static void vsock_vmci_qp_resumed_cb(vmci_id sub_id,
+ const struct vmci_event_data *e_data,
+ void *client_data)
+{
+ int i;
+
+ spin_lock_bh(&vsock_table_lock);
+
+ /*
+ * XXX This loop should probably be provided by util.{h,c}, but that's
+ * for another day.
+ */
+ for (i = 0; i < ARRAY_SIZE(vsock_connected_table); i++) {
+ vsock_vmci_sock *vsk;
+
+ list_for_each_entry(vsk, &vsock_connected_table[i],
+ connected_table) {
+ struct sock *sk = sk_vsock(vsk);
+
+ /*
+ * XXX Technically this is racy but the resulting
+ * outcome from such a race is relatively harmless. My
+ * next change will be a fix to this.
+ */
+ vsock_vmci_handle_detach(sk);
+ }
+ }
+
+ spin_unlock_bh(&vsock_table_lock);
+}
+
+/*
+ * vsock_vmci_pending_work --
+ *
+ * Releases the resources for a pending socket if it has not reached the
+ * connected state and been accepted by a user process.
+ *
+ * Results: None.
+ *
+ * Side effects: The socket may be removed from the connected list and all its
+ * resources freed.
+ */
+
+static void vsock_vmci_pending_work(struct work_struct *work)
+{
+ struct sock *sk;
+ struct sock *listener;
+ vsock_vmci_sock *vsk;
+ bool cleanup;
+
+ vsk = container_of(work, vsock_vmci_sock, dwork.work);
+ sk = sk_vsock(vsk);
+ listener = vsk->listener;
+ cleanup = true;
+
+ lock_sock(listener);
+ lock_sock(sk);
+
+ /*
+ * The socket should be on the pending list or the accept queue, but
+ * not both. It's also possible that the socket isn't on either.
+ */
+ ASSERT((vsock_vmci_is_pending(sk) && !vsock_vmci_in_accept_queue(sk))
+ || (!vsock_vmci_is_pending(sk) && vsock_vmci_in_accept_queue(sk))
+ || (!vsock_vmci_is_pending(sk)
+ && !vsock_vmci_in_accept_queue(sk)));
+
+ if (vsock_vmci_is_pending(sk)) {
+ vsock_vmci_remove_pending(listener, sk);
+ } else if (!vsk->rejected) {
+ /*
+ * We are not on the pending list and accept() did not reject
+ * us, so we must have been accepted by our user process. We
+ * just need to drop our references to the sockets and be on
+ * our way.
+ */
+ cleanup = false;
+ goto out;
+ }
+
+ listener->sk_ack_backlog--;
+
+ /*
+ * We need to remove ourself from the global connected sockets list so
+ * incoming packets can't find this socket, and to reduce the reference
+ * count.
+ */
+ if (vsock_vmci_in_connected_table(sk))
+ vsock_vmci_remove_connected(sk);
+
+ sk->sk_state = SS_FREE;
+
+out:
+ release_sock(sk);
+ release_sock(listener);
+ if (cleanup)
+ sock_put(sk);
+
+ sock_put(sk);
+ sock_put(listener);
+}
+
+/*
+ * vsock_vmci_recv_pkt_work --
+ *
+ * Handles an incoming control packet for the provided socket. This is the
+ * state machine for our stream sockets.
+ *
+ * Results: None.
+ *
+ * Side effects: May set state and wakeup threads waiting for socket state to
+ * change.
+ */
+
+static void vsock_vmci_recv_pkt_work(struct work_struct *work)
+{
+ vsock_recv_pkt_info *recv_pkt_info;
+ vsock_packet *pkt;
+ struct sock *sk;
+
+ recv_pkt_info = container_of(work, vsock_recv_pkt_info, work);
+ sk = recv_pkt_info->sk;
+ pkt = &recv_pkt_info->pkt;
+
+ ASSERT(pkt->type < VSOCK_PACKET_TYPE_MAX);
+
+ lock_sock(sk);
+
+ switch (sk->sk_state) {
+ case SS_LISTEN:
+ vsock_vmci_recv_listen(sk, pkt);
+ break;
+ case SS_CONNECTING:
+ /*
+ * Processing of pending connections for servers goes through
+ * the listening socket, so see vsock_vmci_recv_listen() for
+ * that path.
+ */
+ vsock_vmci_recv_connecting_client(sk, pkt);
+ break;
+ case SS_CONNECTED:
+ vsock_vmci_recv_connected(sk, pkt);
+ break;
+ default:
+ /*
+ * Because this function does not run in the same context as
+ * vsock_vmci_recv_stream_cb it is possible that the socket has
+ * closed. We need to let the other side know or it could be
+ * sitting in a connect and hang forever. Send a reset to
+ * prevent that.
+ */
+ VSOCK_SEND_RESET(sk, pkt);
+ goto out;
+ }
+
+out:
+ release_sock(sk);
+ kfree(recv_pkt_info);
+ /*
+ * Release reference obtained in the stream callback when we fetched
+ * this socket out of the bound or connected list.
+ */
+ sock_put(sk);
+}
+
+/*
+ * vsock_vmci_recv_listen --
+ *
+ * Receives packets for sockets in the listen state.
+ *
+ * Note that this assumes the socket lock is held.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: A new socket may be created and a negotiate control packet is
+ * sent.
+ */
+
+static int vsock_vmci_recv_listen(struct sock *sk, vsock_packet *pkt)
+{
+ struct sock *pending;
+ vsock_vmci_sock *vpending;
+ int err;
+ u64 qp_size;
+ bool old_request = false;
+ bool old_pkt_proto = false;
+
+ ASSERT(sk->sk_state == SS_LISTEN);
+
+ err = 0;
+
+ /*
+ * Because we are in the listen state, we could be receiving a packet
+ * for ourself or any previous connection requests that we received.
+ * If it's the latter, we try to find a socket in our list of pending
+ * connections and, if we do, call the appropriate handler for the
+ * state that that socket is in. Otherwise we try to service the
+ * connection request.
+ */
+ pending = vsock_vmci_get_pending(sk, pkt);
+ if (pending) {
+ lock_sock(pending);
+ switch (pending->sk_state) {
+ case SS_CONNECTING:
+ err =
+ vsock_vmci_recv_connecting_server(sk, pending, pkt);
+ break;
+ default:
+ VSOCK_SEND_RESET(pending, pkt);
+ err = -EINVAL;
+ }
+
+ if (err < 0)
+ vsock_vmci_remove_pending(sk, pending);
+
+ release_sock(pending);
+ vsock_vmci_release_pending(pending);
+
+ return err;
+ }
+
+ /*
+ * The listen state only accepts connection requests. Reply with a
+ * reset unless we received a reset.
+ */
+
+ if (!(pkt->type == VSOCK_PACKET_TYPE_REQUEST ||
+ pkt->type == VSOCK_PACKET_TYPE_REQUEST2)) {
+ VSOCK_REPLY_RESET(pkt);
+ return -EINVAL;
+ }
+
+ if (pkt->u.size == 0) {
+ VSOCK_REPLY_RESET(pkt);
+ return -EINVAL;
+ }
+
+ /*
+ * If this socket can't accommodate this connection request, we send a
+ * reset. Otherwise we create and initialize a child socket and reply
+ * with a connection negotiation.
+ */
+ if (sk->sk_ack_backlog >= sk->sk_max_ack_backlog) {
+ VSOCK_REPLY_RESET(pkt);
+ return -ECONNREFUSED;
+ }
+
+ pending = __vsock_vmci_create(sock_net(sk), NULL, sk, GFP_KERNEL,
+ sk->sk_type);
+ if (!pending) {
+ VSOCK_SEND_RESET(sk, pkt);
+ return -ENOMEM;
+ }
+
+ vpending = vsock_sk(pending);
+ ASSERT(vsock_sk(sk)->local_addr.svm_port == pkt->dst_port);
+
+ vsock_addr_init(&vpending->local_addr,
+ VMCI_HANDLE_TO_CONTEXT_ID(pkt->dg.dst), pkt->dst_port);
+ vsock_addr_init(&vpending->remote_addr,
+ VMCI_HANDLE_TO_CONTEXT_ID(pkt->dg.src), pkt->src_port);
+
+ /*
+ * If the proposed size fits within our min/max, accept it. Otherwise
+ * propose our own size.
+ */
+ if (pkt->u.size >= vpending->queue_pair_min_size &&
+ pkt->u.size <= vpending->queue_pair_max_size) {
+ qp_size = pkt->u.size;
+ } else {
+ qp_size = vpending->queue_pair_size;
+ }
+
+ /*
+ * Figure out if we are using old or new requests based on the
+ * overrides pkt types sent by our peer.
+ */
+ if (vsock_vmci_old_proto_override(&old_pkt_proto)) {
+ old_request = old_pkt_proto;
+ } else {
+ if (pkt->type == VSOCK_PACKET_TYPE_REQUEST)
+ old_request = true;
+ else if (pkt->type == VSOCK_PACKET_TYPE_REQUEST2)
+ old_request = false;
+
+ }
+
+ if (old_request) {
+ /* Handle a REQUEST (or override) */
+ vsock_proto_version version = VSOCK_PROTO_INVALID;
+ if (vsock_vmci_proto_to_notify_struct(pending, &version, true))
+ err = VSOCK_SEND_NEGOTIATE(pending, qp_size);
+ else
+ err = -EINVAL;
+
+ } else {
+ /* Handle a REQUEST2 (or override) */
+ int proto_int = pkt->proto;
+ int pos;
+ u16 active_proto_version = 0;
+
+ /*
+ * The list of possible protocols is the intersection of all
+ * protocols the client supports ... plus all the protocols we
+ * support.
+ */
+ proto_int &= vsock_vmci_new_proto_supported_versions();
+
+ /* We choose the highest possible protocol version and use that
+ * one. */
+ pos = fls(proto_int);
+ if (pos) {
+ active_proto_version = (1 << (pos - 1));
+ if (vsock_vmci_proto_to_notify_struct
+ (pending, &active_proto_version, false))
+ err =
+ VSOCK_SEND_NEGOTIATE2(pending, qp_size,
+ active_proto_version);
+ else
+ err = -EINVAL;
+
+ } else {
+ err = -EINVAL;
+ }
+ }
+
+ if (err < 0) {
+ VSOCK_SEND_RESET(sk, pkt);
+ sock_put(pending);
+ err = vsock_vmci_error_to_vsock_error(err);
+ goto out;
+ }
+
+ vsock_vmci_add_pending(sk, pending);
+ sk->sk_ack_backlog++;
+
+ pending->sk_state = SS_CONNECTING;
+ vpending->produce_size = vpending->consume_size = qp_size;
+ vpending->queue_pair_size = qp_size;
+
+ NOTIFYCALL(vpending, process_request, pending);
+
+ /*
+ * We might never receive another message for this socket and it's not
+ * connected to any process, so we have to ensure it gets cleaned up
+ * ourself. Our delayed work function will take care of that. Note
+ * that we do not ever cancel this function since we have few
+ * guarantees about its state when calling cancel_delayed_work().
+ * Instead we hold a reference on the socket for that function and make
+ * it capable of handling cases where it needs to do nothing but
+ * release that reference.
+ */
+ vpending->listener = sk;
+ sock_hold(sk);
+ sock_hold(pending);
+ INIT_DELAYED_WORK(&vpending->dwork, vsock_vmci_pending_work);
+ schedule_delayed_work(&vpending->dwork, HZ);
+
+out:
+ return err;
+}
+
+/*
+ * vsock_vmci_recv_connecting_server --
+ *
+ * Receives packets for sockets in the connecting state on the server side.
+ *
+ * Connecting sockets on the server side can only receive queue pair offer
+ * packets. All others should be treated as cause for closing the connection.
+ *
+ * Note that this assumes the socket lock is held for both sk and pending.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: A queue pair may be created, an attach control packet may be
+ * sent, the socket may transition to the connected state, and a pending caller
+ * in accept() may be woken up.
+ */
+
+static int
+vsock_vmci_recv_connecting_server(struct sock *listener,
+ struct sock *pending, vsock_packet *pkt)
+{
+ vsock_vmci_sock *vpending;
+ struct vmci_handle handle;
+ struct vmci_qp *qpair;
+ bool is_local;
+ u32 flags;
+ vmci_id detach_sub_id;
+ int err;
+ int skerr;
+
+ ASSERT(listener->sk_state == SS_LISTEN);
+ ASSERT(pending->sk_state == SS_CONNECTING);
+
+ vpending = vsock_sk(pending);
+ detach_sub_id = VMCI_INVALID_ID;
+
+ switch (pkt->type) {
+ case VSOCK_PACKET_TYPE_OFFER:
+ if (VMCI_HANDLE_INVALID(pkt->u.handle)) {
+ VSOCK_SEND_RESET(pending, pkt);
+ skerr = EPROTO;
+ err = -EINVAL;
+ goto destroy;
+ }
+ break;
+ default:
+ /* Close and cleanup the connection. */
+ VSOCK_SEND_RESET(pending, pkt);
+ skerr = EPROTO;
+ err = pkt->type == VSOCK_PACKET_TYPE_RST ? 0 : -EINVAL;
+ goto destroy;
+ }
+
+ ASSERT(pkt->type == VSOCK_PACKET_TYPE_OFFER);
+
+ /*
+ * In order to complete the connection we need to attach to the offered
+ * queue pair and send an attach notification. We also subscribe to the
+ * detach event so we know when our peer goes away, and we do that
+ * before attaching so we don't miss an event. If all this succeeds,
+ * we update our state and wakeup anything waiting in accept() for a
+ * connection.
+ */
+
+ /*
+ * We don't care about attach since we ensure the other side has
+ * attached by specifying the ATTACH_ONLY flag below.
+ */
+ err = vmci_event_subscribe(VMCI_EVENT_QP_PEER_DETACH,
+ vsock_vmci_peer_detach_cb,
+ pending, &detach_sub_id);
+ if (err < VMCI_SUCCESS) {
+ VSOCK_SEND_RESET(pending, pkt);
+ err = vsock_vmci_error_to_vsock_error(err);
+ skerr = -err;
+ goto destroy;
+ }
+
+ vpending->detach_sub_id = detach_sub_id;
+
+ /* Now attach to the queue pair the client created. */
+ handle = pkt->u.handle;
+
+ /*
+ * vpending->local_addr always has a context id so we do not need to
+ * worry about VMADDR_CID_ANY in this case.
+ */
+ is_local =
+ vpending->remote_addr.svm_cid == vpending->local_addr.svm_cid;
+ flags = VMCI_QPFLAG_ATTACH_ONLY;
+ flags |= is_local ? VMCI_QPFLAG_LOCAL : 0;
+
+ err = vsock_vmci_queue_pair_alloc(&qpair,
+ &handle,
+ vpending->produce_size,
+ vpending->consume_size,
+ VMCI_HANDLE_TO_CONTEXT_ID(pkt->
+ dg.src),
+ flags,
+ vsock_vmci_trusted(
+ vpending,
+ vpending->remote_addr.svm_cid));
+ if (err < 0) {
+ VSOCK_SEND_RESET(pending, pkt);
+ skerr = -err;
+ goto destroy;
+ }
+
+ ASSERT(VMCI_HANDLE_EQUAL(handle, pkt->u.handle));
+ vpending->qp_handle = handle;
+ vpending->qpair = qpair;
+
+ /*
+ * When we send the attach message, we must be ready to handle incoming
+ * control messages on the newly connected socket. So we move the
+ * pending socket to the connected state before sending the attach
+ * message. Otherwise, an incoming packet triggered by the attach being
+ * received by the peer may be processed concurrently with what happens
+ * below after sending the attach message, and that incoming packet
+ * will find the listening socket instead of the (currently) pending
+ * socket. Note that enqueueing the socket increments the reference
+ * count, so even if a reset comes before the connection is accepted,
+ * the socket will be valid until it is removed from the queue.
+ *
+ * If we fail sending the attach below, we remove the socket from the
+ * connected list and move the socket to SS_UNCONNECTED before
+ * releasing the lock, so a pending slow path processing of an incoming
+ * packet will not see the socket in the connected state in that case.
+ */
+ pending->sk_state = SS_CONNECTED;
+
+ vsock_vmci_insert_connected(vsock_connected_sockets_vsk(vpending),
+ pending);
+
+ /* Notify our peer of our attach. */
+ err = VSOCK_SEND_ATTACH(pending, handle);
+ if (err < 0) {
+ vsock_vmci_remove_connected(pending);
+ pr_err("Could not send attach\n");
+ VSOCK_SEND_RESET(pending, pkt);
+ err = vsock_vmci_error_to_vsock_error(err);
+ skerr = -err;
+ goto destroy;
+ }
+
+ /*
+ * We have a connection. Move the now connected socket from the
+ * listener's pending list to the accept queue so callers of accept()
+ * can find it.
+ */
+ vsock_vmci_remove_pending(listener, pending);
+ vsock_vmci_enqueue_accept(listener, pending);
+
+ /*
+ * Callers of accept() will be be waiting on the listening socket, not
+ * the pending socket.
+ */
+ listener->sk_state_change(listener);
+
+ return 0;
+
+destroy:
+ pending->sk_err = skerr;
+ pending->sk_state = SS_UNCONNECTED;
+ /*
+ * As long as we drop our reference, all necessary cleanup will handle
+ * when the cleanup function drops its reference and our destruct
+ * implementation is called. Note that since the listen handler will
+ * remove pending from the pending list upon our failure, the cleanup
+ * function won't drop the additional reference, which is why we do it
+ * here.
+ */
+ sock_put(pending);
+
+ return err;
+}
+
+/*
+ * vsock_vmci_recv_connecting_client --
+ *
+ * Receives packets for sockets in the connecting state on the client side.
+ *
+ * Connecting sockets on the client side should only receive attach packets.
+ * All others should be treated as cause for closing the connection.
+ *
+ * Note that this assumes the socket lock is held for both sk and pending.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: The socket may transition to the connected state and wakeup
+ * the pending caller of connect().
+ */
+
+static int
+vsock_vmci_recv_connecting_client(struct sock *sk, vsock_packet *pkt)
+{
+ vsock_vmci_sock *vsk;
+ int err;
+ int skerr;
+
+ ASSERT(sk->sk_state == SS_CONNECTING);
+
+ vsk = vsock_sk(sk);
+
+ switch (pkt->type) {
+ case VSOCK_PACKET_TYPE_ATTACH:
+ if (VMCI_HANDLE_INVALID(pkt->u.handle) ||
+ !VMCI_HANDLE_EQUAL(pkt->u.handle, vsk->qp_handle)) {
+ skerr = EPROTO;
+ err = -EINVAL;
+ goto destroy;
+ }
+
+ /*
+ * Signify the socket is connected and wakeup the waiter in
+ * connect(). Also place the socket in the connected table for
+ * accounting (it can already be found since it's in the bound
+ * table).
+ */
+ sk->sk_state = SS_CONNECTED;
+ sk->sk_socket->state = SS_CONNECTED;
+ vsock_vmci_insert_connected(vsock_connected_sockets_vsk(vsk),
+ sk);
+ sk->sk_state_change(sk);
+
+ break;
+ case VSOCK_PACKET_TYPE_NEGOTIATE:
+ case VSOCK_PACKET_TYPE_NEGOTIATE2:
+ if (pkt->u.size == 0 ||
+ VMCI_HANDLE_TO_CONTEXT_ID(pkt->dg.src) !=
+ vsk->remote_addr.svm_cid
+ || pkt->src_port != vsk->remote_addr.svm_port
+ || !VMCI_HANDLE_INVALID(vsk->qp_handle) || vsk->qpair
+ || vsk->produce_size != 0 || vsk->consume_size != 0
+ || vsk->attach_sub_id != VMCI_INVALID_ID
+ || vsk->detach_sub_id != VMCI_INVALID_ID) {
+ skerr = EPROTO;
+ err = -EINVAL;
+
+ goto destroy;
+ }
+
+ err = vsock_vmci_recv_connecting_client_negotiate(sk, pkt);
+ if (err) {
+ skerr = -err;
+ goto destroy;
+ }
+
+ break;
+ case VSOCK_PACKET_TYPE_INVALID:
+ err = vsock_vmci_recv_connecting_client_invalid(sk, pkt);
+ if (err) {
+ skerr = -err;
+ goto destroy;
+ }
+
+ break;
+ case VSOCK_PACKET_TYPE_RST:
+ /*
+ * Older versions of the linux code (WS 6.5 / ESX 4.0) used to
+ * continue processing here after they sent an INVALID packet.
+ * This meant that we got a RST after the INVALID. We ignore a
+ * RST after an INVALID. The common code doesn't send the RST
+ * ... so we can hang if an old version of the common code
+ * fails between getting a REQUEST and sending an OFFER back.
+ * Not much we can do about it... except hope that it doesn't
+ * happen.
+ */
+ if (vsk->ignore_connecting_rst) {
+ vsk->ignore_connecting_rst = false;
+ } else {
+ skerr = ECONNRESET;
+ err = 0;
+ goto destroy;
+ }
+
+ break;
+ default:
+ /* Close and cleanup the connection. */
+ skerr = EPROTO;
+ err = -EINVAL;
+ goto destroy;
+ }
+
+ ASSERT(pkt->type == VSOCK_PACKET_TYPE_ATTACH ||
+ pkt->type == VSOCK_PACKET_TYPE_NEGOTIATE ||
+ pkt->type == VSOCK_PACKET_TYPE_NEGOTIATE2 ||
+ pkt->type == VSOCK_PACKET_TYPE_INVALID ||
+ pkt->type == VSOCK_PACKET_TYPE_RST);
+
+ return 0;
+
+destroy:
+ VSOCK_SEND_RESET(sk, pkt);
+
+ sk->sk_state = SS_UNCONNECTED;
+ sk->sk_err = skerr;
+ sk->sk_error_report(sk);
+ return err;
+}
+
+/*
+ * vsock_vmci_recv_connecting_client_negotiate --
+ *
+ * Handles a negotiate packet for a client in the connecting state.
+ *
+ * Note that this assumes the socket lock is held for both sk and pending.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: The socket may transition to the connected state and wakeup
+ * the pending caller of connect().
+ */
+
+static int
+vsock_vmci_recv_connecting_client_negotiate(struct sock *sk, vsock_packet *pkt)
+{
+ int err;
+ vsock_vmci_sock *vsk;
+ struct vmci_handle handle;
+ struct vmci_qp *qpair;
+ vmci_id attach_sub_id;
+ vmci_id detach_sub_id;
+ bool is_local;
+ u32 flags;
+ bool old_proto = true;
+ bool old_pkt_proto;
+ vsock_proto_version version;
+
+ vsk = vsock_sk(sk);
+ handle = VMCI_INVALID_HANDLE;
+ attach_sub_id = VMCI_INVALID_ID;
+ detach_sub_id = VMCI_INVALID_ID;
+
+ ASSERT(pkt->u.size > 0);
+ ASSERT(vsk->remote_addr.svm_cid ==
+ VMCI_HANDLE_TO_CONTEXT_ID(pkt->dg.src));
+ ASSERT(vsk->remote_addr.svm_port == pkt->src_port);
+ ASSERT(VMCI_HANDLE_INVALID(vsk->qp_handle));
+ ASSERT(vsk->qpair == NULL);
+ ASSERT(vsk->produce_size == 0);
+ ASSERT(vsk->consume_size == 0);
+ ASSERT(vsk->attach_sub_id == VMCI_INVALID_ID);
+ ASSERT(vsk->detach_sub_id == VMCI_INVALID_ID);
+
+ /*
+ * If we have gotten here then we should be past the point where old
+ * linux vsock could have sent the bogus rst.
+ */
+ vsk->sent_request = false;
+ vsk->ignore_connecting_rst = false;
+
+ /* Verify that we're OK with the proposed queue pair size */
+ if (pkt->u.size < vsk->queue_pair_min_size ||
+ pkt->u.size > vsk->queue_pair_max_size) {
+ err = -EINVAL;
+ goto destroy;
+ }
+
+ /*
+ * At this point we know the CID the peer is using to talk to us.
+ */
+
+ if (vsk->local_addr.svm_cid == VMADDR_CID_ANY)
+ vsk->local_addr.svm_cid =
+ VMCI_HANDLE_TO_CONTEXT_ID(pkt->dg.dst);
+
+ /*
+ * Setup the notify ops to be the highest supported version that both
+ * the server and the client support.
+ */
+
+ if (vsock_vmci_old_proto_override(&old_pkt_proto)) {
+ old_proto = old_pkt_proto;
+ } else {
+ if (pkt->type == VSOCK_PACKET_TYPE_NEGOTIATE)
+ old_proto = true;
+ else if (pkt->type == VSOCK_PACKET_TYPE_NEGOTIATE2)
+ old_proto = false;
+
+ }
+
+ if (old_proto)
+ version = VSOCK_PROTO_INVALID;
+ else
+ version = pkt->proto;
+
+ if (!vsock_vmci_proto_to_notify_struct(sk, &version, old_proto)) {
+ err = -EINVAL;
+ goto destroy;
+ }
+
+ /*
+ * Subscribe to attach and detach events first.
+ *
+ * XXX We attach once for each queue pair created for now so it is easy
+ * to find the socket (it's provided), but later we should only
+ * subscribe once and add a way to lookup sockets by queue pair handle.
+ */
+ err = vmci_event_subscribe(VMCI_EVENT_QP_PEER_ATTACH,
+ vsock_vmci_peer_attach_cb,
+ sk, &attach_sub_id);
+ if (err < VMCI_SUCCESS) {
+ err = vsock_vmci_error_to_vsock_error(err);
+ goto destroy;
+ }
+
+ err = vmci_event_subscribe(VMCI_EVENT_QP_PEER_DETACH,
+ vsock_vmci_peer_detach_cb,
+ sk, &detach_sub_id);
+ if (err < VMCI_SUCCESS) {
+ err = vsock_vmci_error_to_vsock_error(err);
+ goto destroy;
+ }
+
+ /* Make VMCI select the handle for us. */
+ handle = VMCI_INVALID_HANDLE;
+ is_local = vsk->remote_addr.svm_cid == vsk->local_addr.svm_cid;
+ flags = is_local ? VMCI_QPFLAG_LOCAL : 0;
+
+ err = vsock_vmci_queue_pair_alloc(&qpair,
+ &handle,
+ pkt->u.size,
+ pkt->u.size,
+ vsk->remote_addr.svm_cid,
+ flags,
+ vsock_vmci_trusted(
+ vsk,
+ vsk->
+ remote_addr.svm_cid));
+ if (err < 0)
+ goto destroy;
+
+ err = VSOCK_SEND_QP_OFFER(sk, handle);
+ if (err < 0) {
+ err = vsock_vmci_error_to_vsock_error(err);
+ goto destroy;
+ }
+
+ vsk->qp_handle = handle;
+ vsk->qpair = qpair;
+
+ vsk->produce_size = vsk->consume_size = pkt->u.size;
+
+ vsk->attach_sub_id = attach_sub_id;
+ vsk->detach_sub_id = detach_sub_id;
+
+ NOTIFYCALL(vsk, process_negotiate, sk);
+
+ return 0;
+
+destroy:
+ if (attach_sub_id != VMCI_INVALID_ID) {
+ vmci_event_unsubscribe(attach_sub_id);
+ ASSERT(vsk->attach_sub_id == VMCI_INVALID_ID);
+ }
+
+ if (detach_sub_id != VMCI_INVALID_ID) {
+ vmci_event_unsubscribe(detach_sub_id);
+ ASSERT(vsk->detach_sub_id == VMCI_INVALID_ID);
+ }
+
+ if (!VMCI_HANDLE_INVALID(handle)) {
+ vmci_qpair_detach(&qpair);
+ ASSERT(VMCI_HANDLE_INVALID(vsk->qp_handle));
+ }
+
+ return err;
+}
+
+/*
+ * vsock_vmci_recv_connecting_client_invalid --
+ *
+ * Handles an invalid packet for a client in the connecting state.
+ *
+ * Note that this assumes the socket lock is held for both sk and pending.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_recv_connecting_client_invalid(struct sock *sk, vsock_packet *pkt)
+{
+ int err = 0;
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ if (vsk->sent_request) {
+ vsk->sent_request = false;
+ vsk->ignore_connecting_rst = true;
+
+ err = VSOCK_SEND_CONN_REQUEST(sk, vsk->queue_pair_size);
+ if (err < 0)
+ err = vsock_vmci_error_to_vsock_error(err);
+ else
+ err = 0;
+
+ }
+
+ return err;
+}
+
+/*
+ * vsock_vmci_recv_connected --
+ *
+ * Receives packets for sockets in the connected state.
+ *
+ * Connected sockets should only ever receive detach, wrote, read, or reset
+ * control messages. Others are treated as errors that are ignored.
+ *
+ * Wrote and read signify that the peer has produced or consumed, respectively.
+ *
+ * Detach messages signify that the connection is being closed cleanly and
+ * reset messages signify that the connection is being closed in error.
+ *
+ * Note that this assumes the socket lock is held.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: A queue pair may be created, an offer control packet sent, and
+ * the socket may transition to the connecting state.
+ *
+ */
+
+static int vsock_vmci_recv_connected(struct sock *sk, vsock_packet *pkt)
+{
+ vsock_vmci_sock *vsk;
+ bool pkt_processed = false;
+
+ ASSERT(sk->sk_state == SS_CONNECTED);
+
+ /*
+ * In cases where we are closing the connection, it's sufficient to
+ * mark the state change (and maybe error) and wake up any waiting
+ * threads. Since this is a connected socket, it's owned by a user
+ * process and will be cleaned up when the failure is passed back on
+ * the current or next system call. Our system call implementations
+ * must therefore check for error and state changes on entry and when
+ * being awoken.
+ */
+ switch (pkt->type) {
+ case VSOCK_PACKET_TYPE_SHUTDOWN:
+ if (pkt->u.mode) {
+ vsk = vsock_sk(sk);
+
+ vsk->peer_shutdown |= pkt->u.mode;
+ sk->sk_state_change(sk);
+ }
+ break;
+
+ case VSOCK_PACKET_TYPE_RST:
+ vsk = vsock_sk(sk);
+ /*
+ * It is possible that we sent our peer a message (e.g a
+ * WAITING_READ) right before we got notified that the peer had
+ * detached. If that happens then we can get a RST pkt back
+ * from our peer even though there is data available for us to
+ * read. In that case, don't shutdown the socket completely but
+ * instead allow the local client to finish reading data off
+ * the queuepair. Always treat a RST pkt in connected mode like
+ * a clean shutdown.
+ */
+ sock_set_flag(sk, SOCK_DONE);
+ vsk->peer_shutdown = SHUTDOWN_MASK;
+ if (vsock_vmci_stream_has_data(vsk) <= 0)
+ sk->sk_state = SS_DISCONNECTING;
+
+ sk->sk_state_change(sk);
+ break;
+
+ default:
+ vsk = vsock_sk(sk);
+ NOTIFYCALL(vsk, handle_notify_pkt, sk, pkt, false, NULL, NULL,
+ &pkt_processed);
+ if (!pkt_processed)
+ return -EINVAL;
+
+ break;
+ }
+
+ return 0;
+}
+
+/*
+ * __vsock_vmci_send_control_pkt --
+ *
+ * Common code to send a control packet.
+ *
+ * Results: Size of datagram sent on success, negative error code otherwise. If
+ * convert_error is true, error code is a vsock error, otherwise, result is a
+ * VMCI error code.
+ *
+ * Side effects: None.
+ */
+
+static int
+__vsock_vmci_send_control_pkt(vsock_packet *pkt,
+ struct sockaddr_vm *src,
+ struct sockaddr_vm *dst,
+ vsock_packet_type type,
+ u64 size,
+ u64 mode,
+ vsock_waiting_info *wait,
+ vsock_proto_version proto,
+ struct vmci_handle handle, bool convert_error)
+{
+ int err;
+
+ BUG_ON(vsock_addr_validate(src) != 0);
+ BUG_ON(vsock_addr_validate(dst) != 0);
+
+ vsock_packet_init(pkt, src, dst, type, size, mode, wait, proto, handle);
+ LOG_PACKET(pkt);
+ VSOCK_STATS_CTLPKT_LOG(pkt->type);
+ err = vmci_datagram_send(&pkt->dg);
+ if (convert_error && (err < 0))
+ return vsock_vmci_error_to_vsock_error(err);
+
+ return err;
+}
+
+/*
+ * vsock_vmci_reply_control_pkt_fast --
+ *
+ * Sends a control packet back to the source of an incoming packet. The control
+ * packet is allocated in the stack.
+ *
+ * Results: Size of datagram sent on success, negative error code otherwise.
+ *
+ * Side effects: None.
+ */
+
+int
+vsock_vmci_reply_control_pkt_fast(vsock_packet *pkt,
+ vsock_packet_type type,
+ u64 size,
+ u64 mode,
+ vsock_waiting_info *wait,
+ struct vmci_handle handle)
+{
+ vsock_packet reply;
+ struct sockaddr_vm src, dst;
+
+ if (pkt->type == VSOCK_PACKET_TYPE_RST) {
+ return 0;
+ } else {
+ vsock_packet_get_addresses(pkt, &src, &dst);
+ return __vsock_vmci_send_control_pkt(&reply, &src, &dst, type,
+ size, mode, wait,
+ VSOCK_PROTO_INVALID,
+ handle, true);
+ }
+}
+
+/*
+ * vsock_vmci_send_control_pkt_bh --
+ *
+ * Sends a control packet from bottom-half context. The control packet is
+ * static data to minimize the resource cost.
+ *
+ * Results: Size of datagram sent on success, negative error code otherwise.
+ * Note that we return a VMCI error message since that's what callers will need
+ * to provide.
+ *
+ * Side effects: None.
+ */
+
+int
+vsock_vmci_send_control_pkt_bh(struct sockaddr_vm *src,
+ struct sockaddr_vm *dst,
+ vsock_packet_type type,
+ u64 size,
+ u64 mode,
+ vsock_waiting_info *wait,
+ struct vmci_handle handle)
+{
+ /*
+ * Note that it is safe to use a single packet across all CPUs since
+ * two tasklets of the same type are guaranteed to not ever run
+ * simultaneously. If that ever changes, or VMCI stops using tasklets,
+ * we can use per-cpu packets.
+ */
+ static vsock_packet pkt;
+
+ return __vsock_vmci_send_control_pkt(&pkt, src, dst, type,
+ size, mode, wait,
+ VSOCK_PROTO_INVALID, handle,
+ false);
+}
+
+/*
+ * vsock_vmci_send_control_pkt --
+ *
+ * Sends a control packet.
+ *
+ * Results: Size of datagram sent on success, negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+int
+vsock_vmci_send_control_pkt(struct sock *sk,
+ vsock_packet_type type,
+ u64 size,
+ u64 mode,
+ vsock_waiting_info *wait,
+ vsock_proto_version proto,
+ struct vmci_handle handle)
+{
+ vsock_packet *pkt;
+ vsock_vmci_sock *vsk;
+ int err;
+
+ vsk = vsock_sk(sk);
+
+ if (!vsock_addr_bound(&vsk->local_addr))
+ return -EINVAL;
+
+ if (!vsock_addr_bound(&vsk->remote_addr))
+ return -EINVAL;
+
+ pkt = kmalloc(sizeof *pkt, GFP_KERNEL);
+ if (!pkt)
+ return -ENOMEM;
+
+ err =
+ __vsock_vmci_send_control_pkt(pkt, &vsk->local_addr,
+ &vsk->remote_addr, type, size, mode,
+ wait, proto, handle, true);
+ kfree(pkt);
+
+ return err;
+}
+
+/*
+ * __vsock_vmci_bind --
+ *
+ * Common functionality needed to bind the specified address to the VSocket.
+ * If VMADDR_CID_ANY or VMADDR_PORT_ANY are specified, the context ID or port
+ * are selected automatically.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: On success, a new datagram handle is created.
+ */
+
+static int __vsock_vmci_bind(struct sock *sk, struct sockaddr_vm *addr)
+{
+ static unsigned int port = LAST_RESERVED_PORT + 1;
+ struct sockaddr_vm new_addr;
+ vsock_vmci_sock *vsk;
+ vmci_id cid;
+ int err;
+
+ vsk = vsock_sk(sk);
+
+ /* First ensure this socket isn't already bound. */
+ if (vsock_addr_bound(&vsk->local_addr))
+ return -EINVAL;
+
+ /*
+ * Now bind to the provided address or select appropriate values if
+ * none are provided (VMADDR_CID_ANY and VMADDR_PORT_ANY). Note that
+ * like AF_INET prevents binding to a non-local IP address (in most
+ * cases), we only allow binding to the local CID.
+ */
+ vsock_addr_init(&new_addr, VMADDR_CID_ANY, VMADDR_PORT_ANY);
+
+ cid = vmci_get_context_id();
+ if (addr->svm_cid != cid && addr->svm_cid != VMADDR_CID_ANY)
+ return -EADDRNOTAVAIL;
+
+ new_addr.svm_cid = addr->svm_cid;
+
+ switch (sk->sk_socket->type) {
+ case SOCK_STREAM:
+ spin_lock_bh(&vsock_table_lock);
+
+ if (addr->svm_port == VMADDR_PORT_ANY) {
+ bool found = false;
+ unsigned int i;
+
+ for (i = 0; i < MAX_PORT_RETRIES; i++) {
+ if (port <= LAST_RESERVED_PORT)
+ port = LAST_RESERVED_PORT + 1;
+
+ new_addr.svm_port = port++;
+
+ if (!__vsock_vmci_find_bound_socket
+ (&new_addr)) {
+ found = true;
+ break;
+ }
+ }
+
+ if (!found) {
+ err = -EADDRNOTAVAIL;
+ goto out;
+ }
+ } else {
+ /*
+ * If port is in reserved range, ensure caller
+ * has necessary privileges.
+ */
+ if (addr->svm_port <= LAST_RESERVED_PORT &&
+ !capable(CAP_NET_BIND_SERVICE)) {
+ err = -EACCES;
+ goto out;
+ }
+
+ new_addr.svm_port = addr->svm_port;
+ if (__vsock_vmci_find_bound_socket(&new_addr)) {
+ err = -EADDRINUSE;
+ goto out;
+ }
+
+ }
+ break;
+ case SOCK_DGRAM: {
+ u32 flags = 0;
+
+ /*
+ * VMCI will select a resource ID for us if we provide
+ * VMCI_INVALID_ID.i
+ */
+ new_addr.svm_port = addr->svm_port == VMADDR_PORT_ANY ?
+ VMCI_INVALID_ID : addr->svm_port;
+
+ if (new_addr.svm_port <= LAST_RESERVED_PORT &&
+ !capable(CAP_NET_BIND_SERVICE)) {
+ err = -EACCES;
+ goto out;
+ }
+
+ if (new_addr.svm_cid == VMADDR_CID_ANY)
+ flags = VMCI_FLAG_ANYCID_DG_HND;
+
+ err = vsock_vmci_datagram_create_hnd(new_addr.svm_port,
+ flags,
+ vsock_vmci_recv_dgram_cb,
+ sk, &vsk->dg_handle);
+ if (err < VMCI_SUCCESS) {
+ err = vsock_vmci_error_to_vsock_error(err);
+ goto out;
+ }
+
+ new_addr.svm_port = VMCI_HANDLE_TO_RESOURCE_ID(vsk->dg_handle);
+ break;
+ }
+ default:
+ err = -EINVAL;
+ goto out;
+ }
+
+ vsock_addr_init(&vsk->local_addr, new_addr.svm_cid, new_addr.svm_port);
+
+ /*
+ * Remove stream sockets from the unbound list and add them to the hash
+ * table for easy lookup by its address. The unbound list is simply an
+ * extra entry at the end of the hash table, a trick used by AF_UNIX.
+ */
+ if (sk->sk_socket->type == SOCK_STREAM) {
+ __vsock_vmci_remove_bound(sk);
+ __vsock_vmci_insert_bound(vsock_bound_sockets(&vsk->local_addr),
+ sk);
+ spin_unlock_bh(&vsock_table_lock);
+ }
+
+ BUG_ON(vsock_addr_validate(&vsk->local_addr) != 0);
+ return 0;
+
+out:
+ if (sk->sk_socket->type == SOCK_STREAM)
+ spin_unlock_bh(&vsock_table_lock);
+
+ return err;
+}
+
+/*
+ * __vsock_vmci_create --
+ *
+ * Does the work to create the sock structure. Note: If sock is NULL then the
+ * type field must be non-zero. Otherwise, sock is non-NULL and the type of
+ * sock is used in the newly created socket.
+ *
+ * Results: sock structure on success, NULL on failure.
+ *
+ * Side effects: Allocated sk is added to the unbound sockets list iff it is
+ * owned by a struct socket.
+ */
+
+static struct sock *__vsock_vmci_create(struct net *net,
+ struct socket *sock,
+ struct sock *parent,
+ gfp_t priority, unsigned short type)
+{
+ struct sock *sk;
+ vsock_vmci_sock *psk;
+ vsock_vmci_sock *vsk;
+
+ ASSERT((sock && !type) || (!sock && type));
+
+ vsk = NULL;
+
+ sk = sk_alloc(net, vsock_vmci_family_ops.family, priority,
+ &vsock_vmci_proto);
+ if (!sk)
+ return NULL;
+
+ sock_init_data(sock, sk);
+
+ /*
+ * sk->sk_type is normally set in sock_init_data, but only if sock is
+ * non-NULL. We make sure that our sockets always have a type by
+ * setting it here if needed.
+ */
+ if (!sock)
+ sk->sk_type = type;
+
+ vsk = vsock_sk(sk);
+ vsock_addr_init(&vsk->local_addr, VMADDR_CID_ANY, VMADDR_PORT_ANY);
+ vsock_addr_init(&vsk->remote_addr, VMADDR_CID_ANY, VMADDR_PORT_ANY);
+
+ sk->sk_destruct = vsock_vmci_sk_destruct;
+ sk->sk_backlog_rcv = vsock_vmci_queue_rcv_skb;
+ sk->sk_state = 0;
+ sock_reset_flag(sk, SOCK_DONE);
+
+ INIT_LIST_HEAD(&vsk->bound_table);
+ INIT_LIST_HEAD(&vsk->connected_table);
+ vsk->dg_handle = VMCI_INVALID_HANDLE;
+ vsk->qp_handle = VMCI_INVALID_HANDLE;
+ vsk->qpair = NULL;
+ vsk->produce_size = vsk->consume_size = 0;
+ vsk->listener = NULL;
+ INIT_LIST_HEAD(&vsk->pending_links);
+ INIT_LIST_HEAD(&vsk->accept_queue);
+ vsk->rejected = false;
+ vsk->sent_request = false;
+ vsk->ignore_connecting_rst = false;
+ vsk->attach_sub_id = vsk->detach_sub_id = VMCI_INVALID_ID;
+ vsk->peer_shutdown = 0;
+
+ if (parent) {
+ psk = vsock_sk(parent);
+ vsk->trusted = psk->trusted;
+ vsk->owner = get_cred(psk->owner);
+ vsk->queue_pair_size = psk->queue_pair_size;
+ vsk->queue_pair_min_size = psk->queue_pair_min_size;
+ vsk->queue_pair_max_size = psk->queue_pair_max_size;
+ vsk->connect_timeout = psk->connect_timeout;
+ } else {
+ vsk->trusted = capable(CAP_NET_ADMIN);
+ vsk->owner = get_current_cred();
+ vsk->queue_pair_size = VSOCK_DEFAULT_QP_SIZE;
+ vsk->queue_pair_min_size = VSOCK_DEFAULT_QP_SIZE_MIN;
+ vsk->queue_pair_max_size = VSOCK_DEFAULT_QP_SIZE_MAX;
+ vsk->connect_timeout = VSOCK_DEFAULT_CONNECT_TIMEOUT;
+ }
+
+ vsk->notify_ops = NULL;
+
+ if (sock)
+ vsock_vmci_insert_bound(vsock_unbound_sockets, sk);
+
+ return sk;
+}
+
+/*
+ * __vsock_vmci_release --
+ *
+ * Releases the provided socket.
+ *
+ * Results: None.
+ *
+ * Side effects: Any pending sockets are also released.
+ */
+
+static void __vsock_vmci_release(struct sock *sk)
+{
+ if (sk) {
+ struct sk_buff *skb;
+ struct sock *pending;
+ struct vsock_vmci_sock *vsk;
+
+ vsk = vsock_sk(sk);
+ pending = NULL; /* Compiler warning. */
+
+ if (vsock_vmci_in_bound_table(sk))
+ vsock_vmci_remove_bound(sk);
+
+ if (vsock_vmci_in_connected_table(sk))
+ vsock_vmci_remove_connected(sk);
+
+ if (!VMCI_HANDLE_INVALID(vsk->dg_handle)) {
+ vmci_datagram_destroy_handle(vsk->dg_handle);
+ vsk->dg_handle = VMCI_INVALID_HANDLE;
+ }
+
+ lock_sock(sk);
+ sock_orphan(sk);
+ sk->sk_shutdown = SHUTDOWN_MASK;
+
+ while ((skb = skb_dequeue(&sk->sk_receive_queue)))
+ kfree_skb(skb);
+
+ /* Clean up any sockets that never were accepted. */
+ while ((pending = vsock_vmci_dequeue_accept(sk)) != NULL) {
+ __vsock_vmci_release(pending);
+ sock_put(pending);
+ }
+
+ release_sock(sk);
+ sock_put(sk);
+ }
+}
+
+/*
+ * Sock operations.
+ */
+
+/*
+ * vsock_vmci_sk_destruct --
+ *
+ * Destroys the provided socket. This is called by sk_free(), which is invoke
+ * when the reference count of the socket drops to zero.
+ *
+ * Results: None.
+ *
+ * Side effects: Socket count is decremented.
+ */
+
+static void vsock_vmci_sk_destruct(struct sock *sk)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ if (vsk->attach_sub_id != VMCI_INVALID_ID) {
+ vmci_event_unsubscribe(vsk->attach_sub_id);
+ vsk->attach_sub_id = VMCI_INVALID_ID;
+ }
+
+ if (vsk->detach_sub_id != VMCI_INVALID_ID) {
+ vmci_event_unsubscribe(vsk->detach_sub_id);
+ vsk->detach_sub_id = VMCI_INVALID_ID;
+ }
+
+ if (!VMCI_HANDLE_INVALID(vsk->qp_handle)) {
+ vmci_qpair_detach(&vsk->qpair);
+ vsk->qp_handle = VMCI_INVALID_HANDLE;
+ vsk->produce_size = vsk->consume_size = 0;
+ }
+
+ /*
+ * Each list entry holds a reference on the socket, so we should not
+ * even be here if the socket is in one of our lists. If we are we
+ * have a stray sock_put() that needs to go away.
+ */
+ ASSERT(!vsock_vmci_in_bound_table(sk));
+ ASSERT(!vsock_vmci_in_connected_table(sk));
+ ASSERT(!vsock_vmci_is_pending(sk));
+ ASSERT(!vsock_vmci_in_accept_queue(sk));
+
+ /*
+ * When clearing these addresses, there's no need to set the family and
+ * possibly register the address family with the kernel.
+ */
+ vsock_addr_init(&vsk->local_addr, VMADDR_CID_ANY, VMADDR_PORT_ANY);
+ vsock_addr_init(&vsk->remote_addr, VMADDR_CID_ANY, VMADDR_PORT_ANY);
+
+ NOTIFYCALL(vsk, socket_destruct, sk);
+
+ put_cred(vsk->owner);
+
+ VSOCK_STATS_CTLPKT_DUMP_ALL();
+ VSOCK_STATS_HIST_DUMP_ALL();
+ VSOCK_STATS_TOTALS_DUMP_ALL();
+}
+
+/*
+ * vsock_vmci_queue_rcv_skb --
+ *
+ * Receives skb on the socket's receive queue.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int vsock_vmci_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
+{
+ int err;
+
+ err = sock_queue_rcv_skb(sk, skb);
+ if (err)
+ kfree_skb(skb);
+
+ return err;
+}
+
+/*
+ * vsock_vmci_register_with_vmci --
+ *
+ * Registers with the VMCI device, and creates control message and event
+ * handlers.
+ *
+ * Results: Zero on success, error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int vsock_vmci_register_with_vmci(void)
+{
+ int err = 0;
+
+ /*
+ * Create the datagram handle that we will use to send and receive all
+ * VSocket control messages for this context.
+ */
+ err = vsock_vmci_datagram_create_hnd(VSOCK_PACKET_RID,
+ VMCI_FLAG_ANYCID_DG_HND,
+ vsock_vmci_recv_stream_cb, NULL,
+ &vmci_stream_handle);
+ if (err < VMCI_SUCCESS) {
+ pr_err("Unable to create datagram handle. (%d)\n",
+ err);
+ err = vsock_vmci_error_to_vsock_error(err);
+ goto out;
+ }
+
+ err = vmci_event_subscribe(VMCI_EVENT_QP_RESUMED,
+ vsock_vmci_qp_resumed_cb,
+ NULL, &qp_resumed_sub_id);
+ if (err < VMCI_SUCCESS) {
+ pr_err("Unable to subscribe to resumed event. (%d)\n",
+ err);
+ err = vsock_vmci_error_to_vsock_error(err);
+ qp_resumed_sub_id = VMCI_INVALID_ID;
+ goto out;
+ }
+
+out:
+ if (err != 0)
+ vsock_vmci_unregister_with_vmci();
+
+ return err;
+}
+
+/*
+ * vsock_vmci_unregister_with_vmci --
+ *
+ * Destroys control message and event handlers, and unregisters with the VMCI
+ * device
+ *
+ * Results: None.
+ *
+ * Side effects: Our socket implementation is no longer accessible.
+ */
+
+static void vsock_vmci_unregister_with_vmci(void)
+{
+ if (!VMCI_HANDLE_INVALID(vmci_stream_handle)) {
+ if (vmci_datagram_destroy_handle(vmci_stream_handle) !=
+ VMCI_SUCCESS)
+ pr_err("Couldn't destroy datagram handle.\n");
+
+ vmci_stream_handle = VMCI_INVALID_HANDLE;
+ }
+
+ if (qp_resumed_sub_id != VMCI_INVALID_ID) {
+ vmci_event_unsubscribe(qp_resumed_sub_id);
+ qp_resumed_sub_id = VMCI_INVALID_ID;
+ }
+}
+
+/*
+ * vsock_vmci_stream_has_data --
+ *
+ * Gets the amount of data available for a given stream socket's consume queue.
+ *
+ * Note that this assumes the socket lock is held.
+ *
+ * Results: The amount of data available or a VMCI error code on failure.
+ *
+ * Side effects: None.
+ */
+
+s64 vsock_vmci_stream_has_data(vsock_vmci_sock *vsk)
+{
+ return vmci_qpair_consume_buf_ready(vsk->qpair);
+}
+
+/*
+ * vsock_vmci_stream_has_space --
+ *
+ * Gets the amount of space available for a give stream socket's produce queue.
+ *
+ * Note that this assumes the socket lock is held.
+ *
+ * Results: The amount of space available or a VMCI error code on failure.
+ *
+ * Side effects: None.
+ */
+
+s64 vsock_vmci_stream_has_space(vsock_vmci_sock *vsk)
+{
+ return vmci_qpair_produce_free_space(vsk->qpair);
+}
+
+/*
+ * Socket operations.
+ */
+
+/*
+ *
+ * vsock_vmci_release --
+ *
+ * Releases the provided socket by freeing the contents of its queue. This is
+ * called when a user process calls close(2) on the socket.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int vsock_vmci_release(struct socket *sock)
+{
+ __vsock_vmci_release(sock->sk);
+ sock->sk = NULL;
+ sock->state = SS_FREE;
+
+ return 0;
+}
+
+/*
+ * vsock_vmci_bind --
+ *
+ * Binds the provided address to the provided socket.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_bind(struct socket *sock, struct sockaddr *addr, int addr_len)
+{
+ int err;
+ struct sock *sk;
+ struct sockaddr_vm *vmci_addr;
+
+ sk = sock->sk;
+
+ if (vsock_addr_cast(addr, addr_len, &vmci_addr) != 0)
+ return -EINVAL;
+
+ lock_sock(sk);
+ err = __vsock_vmci_bind(sk, vmci_addr);
+ release_sock(sk);
+
+ return err;
+}
+
+/*
+ * vsock_vmci_dgram_connect --
+ *
+ * Connects a datagram socket. This can be called multiple times to change the
+ * socket's association and can be called with a sockaddr whose family is set
+ * to AF_UNSPEC to dissolve any existing association.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_dgram_connect(struct socket *sock,
+ struct sockaddr *addr, int addr_len, int flags)
+{
+ int err;
+ struct sock *sk;
+ vsock_vmci_sock *vsk;
+ struct sockaddr_vm *remote_addr;
+
+ sk = sock->sk;
+ vsk = vsock_sk(sk);
+
+ err = vsock_addr_cast(addr, addr_len, &remote_addr);
+ if (err == -EAFNOSUPPORT && remote_addr->svm_family == AF_UNSPEC) {
+ lock_sock(sk);
+ vsock_addr_init(&vsk->remote_addr, VMADDR_CID_ANY,
+ VMADDR_PORT_ANY);
+ sock->state = SS_UNCONNECTED;
+ release_sock(sk);
+ return 0;
+ } else if (err != 0)
+ return -EINVAL;
+
+ lock_sock(sk);
+
+ if (!vsock_addr_bound(&vsk->local_addr)) {
+ struct sockaddr_vm local_addr;
+
+ vsock_addr_init(&local_addr, VMADDR_CID_ANY, VMADDR_PORT_ANY);
+ err = __vsock_vmci_bind(sk, &local_addr);
+ if (err != 0)
+ goto out;
+
+ }
+
+ if (!vsock_addr_socket_context_dgram(remote_addr->svm_cid,
+ remote_addr->svm_port)) {
+ err = -EINVAL;
+ goto out;
+ }
+
+ memcpy(&vsk->remote_addr, remote_addr, sizeof vsk->remote_addr);
+ sock->state = SS_CONNECTED;
+
+out:
+ release_sock(sk);
+ return err;
+}
+
+/*
+ * vsock_vmci_connect_timeout --
+ *
+ * Asynchronous connection attempts schedule this timeout function to notify
+ * the connector of an unsuccessfull connection attempt. If the socket is still
+ * in the connecting state and hasn't been closed, we mark the socket as timed
+ * out. Otherwise, we do nothing.
+ *
+ * Results: None.
+ *
+ * Side effects: May destroy the socket.
+ */
+
+static void vsock_vmci_connect_timeout(struct work_struct *work)
+{
+ struct sock *sk;
+ vsock_vmci_sock *vsk;
+
+ vsk = container_of(work, vsock_vmci_sock, dwork.work);
+ sk = sk_vsock(vsk);
+
+ lock_sock(sk);
+ if (sk->sk_state == SS_CONNECTING &&
+ (sk->sk_shutdown != SHUTDOWN_MASK)) {
+ sk->sk_state = SS_UNCONNECTED;
+ sk->sk_err = ETIMEDOUT;
+ sk->sk_error_report(sk);
+ }
+ release_sock(sk);
+
+ sock_put(sk);
+}
+
+/*
+ * vsock_vmci_stream_connect --
+ *
+ * Connects a stream socket.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_stream_connect(struct socket *sock,
+ struct sockaddr *addr, int addr_len, int flags)
+{
+ int err;
+ struct sock *sk;
+ vsock_vmci_sock *vsk;
+ struct sockaddr_vm *remote_addr;
+ long timeout;
+ bool old_pkt_proto = false;
+ DEFINE_WAIT(wait);
+
+ err = 0;
+ sk = sock->sk;
+ vsk = vsock_sk(sk);
+
+ lock_sock(sk);
+
+ /* XXX AF_UNSPEC should make us disconnect like AF_INET. */
+ switch (sock->state) {
+ case SS_CONNECTED:
+ err = -EISCONN;
+ goto out;
+ case SS_DISCONNECTING:
+ err = -EINVAL;
+ goto out;
+ case SS_CONNECTING:
+ /*
+ * This continues on so we can move sock into the SS_CONNECTED
+ * state once the connection has completed (at which point err
+ * will be set to zero also). Otherwise, we will either wait
+ * for the connection or return -EALREADY should this be a
+ * non-blocking call.
+ */
+ err = -EALREADY;
+ break;
+ default:
+ ASSERT(sk->sk_state == SS_FREE ||
+ sk->sk_state == SS_UNCONNECTED ||
+ sk->sk_state == SS_LISTEN);
+ if ((sk->sk_state == SS_LISTEN) ||
+ vsock_addr_cast(addr, addr_len, &remote_addr) != 0) {
+ err = -EINVAL;
+ goto out;
+ }
+
+ /*
+ * The hypervisor and well-known contexts do not have socket
+ * endpoints.
+ */
+ if (!vsock_addr_socket_context_stream(remote_addr->svm_cid)) {
+ err = -ENETUNREACH;
+ goto out;
+ }
+
+ /* Set the remote address that we are connecting to. */
+ memcpy(&vsk->remote_addr, remote_addr, sizeof vsk->remote_addr);
+
+ /* Autobind this socket to the local address if necessary. */
+ if (!vsock_addr_bound(&vsk->local_addr)) {
+ struct sockaddr_vm local_addr;
+
+ vsock_addr_init(&local_addr, VMADDR_CID_ANY,
+ VMADDR_PORT_ANY);
+ err = __vsock_vmci_bind(sk, &local_addr);
+ if (err != 0)
+ goto out;
+
+ }
+
+ sk->sk_state = SS_CONNECTING;
+
+ if (vsock_vmci_old_proto_override(&old_pkt_proto)
+ && old_pkt_proto) {
+ err = VSOCK_SEND_CONN_REQUEST(sk, vsk->queue_pair_size);
+ if (err < 0) {
+ sk->sk_state = SS_UNCONNECTED;
+ goto out;
+ }
+ } else {
+ int supported_proto_versions =
+ vsock_vmci_new_proto_supported_versions();
+ err =
+ VSOCK_SEND_CONN_REQUEST2(sk, vsk->queue_pair_size,
+ supported_proto_versions);
+ if (err < 0) {
+ sk->sk_state = SS_UNCONNECTED;
+ goto out;
+ }
+
+ vsk->sent_request = true;
+ }
+
+ /*
+ * Mark sock as connecting and set the error code to in
+ * progress in case this is a non-blocking connect.
+ */
+ sock->state = SS_CONNECTING;
+ err = -EINPROGRESS;
+ }
+
+ /*
+ * The receive path will handle all communication until we are able to
+ * enter the connected state. Here we wait for the connection to be
+ * completed or a notification of an error.
+ */
+ timeout = vsk->connect_timeout;
+ prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
+
+ while (sk->sk_state != SS_CONNECTED && sk->sk_err == 0) {
+ if (flags & O_NONBLOCK) {
+ /*
+ * If we're not going to block, we schedule a timeout
+ * function to generate a timeout on the connection
+ * attempt, in case the peer doesn't respond in a
+ * timely manner. We hold on to the socket until the
+ * timeout fires.
+ */
+ sock_hold(sk);
+ INIT_DELAYED_WORK(&vsk->dwork,
+ vsock_vmci_connect_timeout);
+ schedule_delayed_work(&vsk->dwork, timeout);
+
+ /* Skip ahead to preserve error code set above. */
+ goto out_wait;
+ }
+
+ release_sock(sk);
+ timeout = schedule_timeout(timeout);
+ lock_sock(sk);
+
+ if (signal_pending(current)) {
+ err = sock_intr_errno(timeout);
+ goto out_wait_error;
+ } else if (timeout == 0) {
+ err = -ETIMEDOUT;
+ goto out_wait_error;
+ }
+
+ prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
+ }
+
+ if (sk->sk_err) {
+ err = -sk->sk_err;
+ goto out_wait_error;
+ } else {
+ ASSERT(sk->sk_state == SS_CONNECTED);
+ err = 0;
+ }
+
+out_wait:
+ finish_wait(sk_sleep(sk), &wait);
+out:
+ release_sock(sk);
+ return err;
+
+out_wait_error:
+ sk->sk_state = SS_UNCONNECTED;
+ sock->state = SS_UNCONNECTED;
+ goto out_wait;
+}
+
+/*
+ * vsock_vmci_accept --
+ *
+ * Accepts next available connection request for this socket.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_accept(struct socket *sock, struct socket *newsock, int flags)
+{
+ struct sock *listener;
+ int err;
+ struct sock *connected;
+ vsock_vmci_sock *vconnected;
+ long timeout;
+ DEFINE_WAIT(wait);
+
+ err = 0;
+ listener = sock->sk;
+
+ lock_sock(listener);
+
+ if (sock->type != SOCK_STREAM) {
+ err = -EOPNOTSUPP;
+ goto out;
+ }
+
+ if (listener->sk_state != SS_LISTEN) {
+ err = -EINVAL;
+ goto out;
+ }
+
+ /*
+ * Wait for children sockets to appear; these are the new sockets
+ * created upon connection establishment.
+ */
+ timeout = sock_sndtimeo(listener, flags & O_NONBLOCK);
+ prepare_to_wait(sk_sleep(listener), &wait, TASK_INTERRUPTIBLE);
+
+ while ((connected = vsock_vmci_dequeue_accept(listener)) == NULL &&
+ listener->sk_err == 0) {
+ release_sock(listener);
+ timeout = schedule_timeout(timeout);
+ lock_sock(listener);
+
+ if (signal_pending(current)) {
+ err = sock_intr_errno(timeout);
+ goto out_wait;
+ } else if (timeout == 0) {
+ err = -EAGAIN;
+ goto out_wait;
+ }
+
+ prepare_to_wait(sk_sleep(listener), &wait, TASK_INTERRUPTIBLE);
+ }
+
+ if (listener->sk_err)
+ err = -listener->sk_err;
+
+ if (connected) {
+ listener->sk_ack_backlog--;
+
+ lock_sock(connected);
+ vconnected = vsock_sk(connected);
+
+ /*
+ * If the listener socket has received an error, then we should
+ * reject this socket and return. Note that we simply mark the
+ * socket rejected, drop our reference, and let the cleanup
+ * function handle the cleanup; the fact that we found it in
+ * the listener's accept queue guarantees that the cleanup
+ * function hasn't run yet.
+ */
+ if (err) {
+ vconnected->rejected = true;
+ release_sock(connected);
+ sock_put(connected);
+ goto out_wait;
+ }
+
+ newsock->state = SS_CONNECTED;
+ sock_graft(connected, newsock);
+ release_sock(connected);
+ sock_put(connected);
+ }
+
+out_wait:
+ finish_wait(sk_sleep(listener), &wait);
+out:
+ release_sock(listener);
+ return err;
+}
+
+/*
+ * vsock_vmci_getname --
+ *
+ * Provides the local or remote address for the socket.
+ *
+ * Results: Zero on success, negative error code otherwise.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_getname(struct socket *sock,
+ struct sockaddr *addr, int *addr_len, int peer)
+{
+ int err;
+ struct sock *sk;
+ vsock_vmci_sock *vsk;
+ struct sockaddr_vm *vmci_addr;
+
+ sk = sock->sk;
+ vsk = vsock_sk(sk);
+ err = 0;
+
+ lock_sock(sk);
+
+ if (peer) {
+ if (sock->state != SS_CONNECTED) {
+ err = -ENOTCONN;
+ goto out;
+ }
+ vmci_addr = &vsk->remote_addr;
+ } else {
+ vmci_addr = &vsk->local_addr;
+ }
+
+ if (!vmci_addr) {
+ err = -EINVAL;
+ goto out;
+ }
+
+ /*
+ * sys_getsockname() and sys_getpeername() pass us a
+ * MAX_SOCK_ADDR-sized buffer and don't set addr_len. Unfortunately
+ * that macro is defined in socket.c instead of .h, so we hardcode its
+ * value here.
+ */
+ ASSERT_ON_COMPILE(sizeof *vmci_addr <= 128);
+ memcpy(addr, vmci_addr, sizeof *vmci_addr);
+ *addr_len = sizeof *vmci_addr;
+
+out:
+ release_sock(sk);
+ return err;
+}
+
+/*
+ * vsock_vmci_poll --
+ *
+ * Waits on file for activity then provides mask indicating state of socket.
+ *
+ * Results: Mask of flags containing socket state.
+ *
+ * Side effects: None.
+ */
+
+static unsigned int
+vsock_vmci_poll(struct file *file, struct socket *sock, poll_table *wait)
+{
+ struct sock *sk;
+ unsigned int mask;
+ vsock_vmci_sock *vsk;
+
+ sk = sock->sk;
+ vsk = vsock_sk(sk);
+
+ poll_wait(file, sk_sleep(sk), wait);
+ mask = 0;
+
+ if (sk->sk_err)
+ /* Signify that there has been an error on this socket. */
+ mask |= POLLERR;
+
+ /*
+ * INET sockets treat local write shutdown and peer write shutdown as a
+ * case of POLLHUP set.
+ */
+ if ((sk->sk_shutdown == SHUTDOWN_MASK) ||
+ ((sk->sk_shutdown & SEND_SHUTDOWN) &&
+ (vsk->peer_shutdown & SEND_SHUTDOWN))) {
+ mask |= POLLHUP;
+ }
+
+ if (sk->sk_shutdown & RCV_SHUTDOWN ||
+ vsk->peer_shutdown & SEND_SHUTDOWN) {
+ mask |= POLLRDHUP;
+ }
+
+ if (sock->type == SOCK_DGRAM) {
+ /*
+ * For datagram sockets we can read if there is something in
+ * the queue and write as long as the socket isn't shutdown for
+ * sending.
+ */
+ if (!skb_queue_empty(&sk->sk_receive_queue) ||
+ (sk->sk_shutdown & RCV_SHUTDOWN)) {
+ mask |= POLLIN | POLLRDNORM;
+ }
+
+ if (!(sk->sk_shutdown & SEND_SHUTDOWN))
+ mask |= POLLOUT | POLLWRNORM | POLLWRBAND;
+
+ } else if (sock->type == SOCK_STREAM) {
+ lock_sock(sk);
+
+ /*
+ * Listening sockets that have connections in their accept
+ * queue can be read.
+ */
+ if (sk->sk_state == SS_LISTEN
+ && !vsock_vmci_is_accept_queue_empty(sk))
+ mask |= POLLIN | POLLRDNORM;
+
+ /*
+ * If there is something in the queue then we can read.
+ */
+ if (!VMCI_HANDLE_INVALID(vsk->qp_handle) &&
+ !(sk->sk_shutdown & RCV_SHUTDOWN)) {
+ bool data_ready_now = false;
+ int ret = 0;
+ NOTIFYCALLRET(vsk, ret, poll_in, sk, 1,
+ &data_ready_now);
+ if (ret < 0) {
+ mask |= POLLERR;
+ } else {
+ if (data_ready_now)
+ mask |= POLLIN | POLLRDNORM;
+
+ }
+ }
+
+ /*
+ * Sockets whose connections have been closed, reset, or
+ * terminated should also be considered read, and we check the
+ * shutdown flag for that.
+ */
+ if (sk->sk_shutdown & RCV_SHUTDOWN ||
+ vsk->peer_shutdown & SEND_SHUTDOWN) {
+ mask |= POLLIN | POLLRDNORM;
+ }
+
+ /*
+ * Connected sockets that can produce data can be written.
+ */
+ if (sk->sk_state == SS_CONNECTED) {
+ if (!(sk->sk_shutdown & SEND_SHUTDOWN)) {
+ bool space_avail_now = false;
+ int ret = 0;
+
+ NOTIFYCALLRET(vsk, ret, poll_out, sk, 1,
+ &space_avail_now);
+ if (ret < 0) {
+ mask |= POLLERR;
+ } else {
+ if (space_avail_now)
+ /*
+ * Remove POLLWRBAND since INET
+ * sockets are not setting it.
+ */
+ mask |= POLLOUT | POLLWRNORM;
+
+ }
+ }
+ }
+
+ /*
+ * Simulate INET socket poll behaviors, which sets
+ * POLLOUT|POLLWRNORM when peer is closed and nothing to read,
+ * but local send is not shutdown.
+ */
+ if (sk->sk_state == SS_UNCONNECTED) {
+ if (!(sk->sk_shutdown & SEND_SHUTDOWN))
+ mask |= POLLOUT | POLLWRNORM;
+
+ }
+
+ release_sock(sk);
+ }
+
+ return mask;
+}
+
+/*
+ * vsock_vmci_listen --
+ *
+ * Signify that this socket is listening for connection requests.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int vsock_vmci_listen(struct socket *sock, int backlog)
+{
+ int err;
+ struct sock *sk;
+ vsock_vmci_sock *vsk;
+
+ sk = sock->sk;
+
+ lock_sock(sk);
+
+ if (sock->type != SOCK_STREAM) {
+ err = -EOPNOTSUPP;
+ goto out;
+ }
+
+ if (sock->state != SS_UNCONNECTED) {
+ err = -EINVAL;
+ goto out;
+ }
+
+ vsk = vsock_sk(sk);
+
+ if (!vsock_addr_bound(&vsk->local_addr)) {
+ err = -EINVAL;
+ goto out;
+ }
+
+ sk->sk_max_ack_backlog = backlog;
+ sk->sk_state = SS_LISTEN;
+
+ err = 0;
+
+out:
+ release_sock(sk);
+ return err;
+}
+
+/*
+ * vsock_vmci_shutdown --
+ *
+ * Shuts down the provided socket in the provided method.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int vsock_vmci_shutdown(struct socket *sock, int mode)
+{
+ int err;
+ struct sock *sk;
+
+ /*
+ * User level uses SHUT_RD (0) and SHUT_WR (1), but the kernel uses
+ * RCV_SHUTDOWN (1) and SEND_SHUTDOWN (2), so we must increment mode
+ * here like the other address families do. Note also that the
+ * increment makes SHUT_RDWR (2) into RCV_SHUTDOWN | SEND_SHUTDOWN (3),
+ * which is what we want.
+ */
+ mode++;
+
+ if ((mode & ~SHUTDOWN_MASK) || !mode)
+ return -EINVAL;
+
+ /*
+ * If this is a STREAM socket and it is not connected then bail out
+ * immediately. If it is a DGRAM socket then we must first kick the socket
+ * so that it wakes up from any sleeping calls, for example recv(), and then
+ * afterwards return the error.
+ */
+
+ sk = sock->sk;
+ if (sock->state == SS_UNCONNECTED) {
+ err = -ENOTCONN;
+ if (sk->sk_type == SOCK_STREAM)
+ return err;
+ } else {
+ sock->state = SS_DISCONNECTING;
+ err = 0;
+ }
+
+ /* Receive and send shutdowns are treated alike. */
+ mode = mode & (RCV_SHUTDOWN | SEND_SHUTDOWN);
+ if (mode) {
+ lock_sock(sk);
+ sk->sk_shutdown |= mode;
+ sk->sk_state_change(sk);
+ release_sock(sk);
+
+ if (sk->sk_type == SOCK_STREAM) {
+ sock_reset_flag(sk, SOCK_DONE);
+ VSOCK_SEND_SHUTDOWN(sk, mode);
+ }
+ }
+
+ return err;
+}
+
+/*
+ * vsock_vmci_dgram_sendmsg --
+ *
+ * Sends a datagram.
+ *
+ * Results: Number of bytes sent on success, negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_dgram_sendmsg(struct kiocb *kiocb,
+ struct socket *sock, struct msghdr *msg, size_t len)
+{
+ int err;
+ struct sock *sk;
+ vsock_vmci_sock *vsk;
+ struct sockaddr_vm *remote_addr;
+ struct vmci_datagram *dg;
+
+ if (msg->msg_flags & MSG_OOB)
+ return -EOPNOTSUPP;
+
+ if (len > VMCI_MAX_DG_PAYLOAD_SIZE)
+ return -EMSGSIZE;
+
+ /* For now, MSG_DONTWAIT is always assumed... */
+ err = 0;
+ sk = sock->sk;
+ vsk = vsock_sk(sk);
+
+ lock_sock(sk);
+
+ if (!vsock_addr_bound(&vsk->local_addr)) {
+ struct sockaddr_vm local_addr;
+
+ vsock_addr_init(&local_addr, VMADDR_CID_ANY, VMADDR_PORT_ANY);
+ err = __vsock_vmci_bind(sk, &local_addr);
+ if (err != 0)
+ goto out;
+
+ }
+
+ /*
+ * If the provided message contains an address, use that. Otherwise
+ * fall back on the socket's remote handle (if it has been connected).
+ */
+ if (msg->msg_name &&
+ vsock_addr_cast(msg->msg_name, msg->msg_namelen,
+ &remote_addr) == 0) {
+ /* Ensure this address is of the right type and is a valid
+ * destination. */
+
+ if (remote_addr->svm_cid == VMADDR_CID_ANY)
+ remote_addr->svm_cid = vmci_get_context_id();
+
+ if (!vsock_addr_bound(remote_addr)) {
+ err = -EINVAL;
+ goto out;
+ }
+ } else if (sock->state == SS_CONNECTED) {
+ remote_addr = &vsk->remote_addr;
+
+ if (remote_addr->svm_cid == VMADDR_CID_ANY)
+ remote_addr->svm_cid = vmci_get_context_id();
+
+ /* XXX Should connect() or this function ensure remote_addr is
+ * bound? */
+ if (!vsock_addr_bound(&vsk->remote_addr)) {
+ err = -EINVAL;
+ goto out;
+ }
+ } else {
+ err = -EINVAL;
+ goto out;
+ }
+
+ /*
+ * Make sure that we don't allow a userlevel app to send datagrams to
+ * the hypervisor that modify VMCI device state.
+ */
+ if (!vsock_addr_socket_context_dgram(remote_addr->svm_cid,
+ remote_addr->svm_port)) {
+ err = -EINVAL;
+ goto out;
+ }
+
+ if (!vsock_vmci_allow_dgram(vsk, remote_addr->svm_cid)) {
+ err = -EPERM;
+ goto out;
+ }
+
+ /*
+ * Allocate a buffer for the user's message and our packet header.
+ */
+ dg = kmalloc(len + sizeof *dg, GFP_KERNEL);
+ if (!dg) {
+ err = -ENOMEM;
+ goto out;
+ }
+
+ memcpy_fromiovec(VMCI_DG_PAYLOAD(dg), msg->msg_iov, len);
+
+ dg->dst = VMCI_MAKE_HANDLE(remote_addr->svm_cid, remote_addr->svm_port);
+ dg->src =
+ VMCI_MAKE_HANDLE(vsk->local_addr.svm_cid, vsk->local_addr.svm_port);
+
+ dg->payload_size = len;
+
+ err = vmci_datagram_send(dg);
+ kfree(dg);
+ if (err < 0) {
+ err = vsock_vmci_error_to_vsock_error(err);
+ goto out;
+ }
+
+ err -= sizeof *dg;
+
+out:
+ release_sock(sk);
+ return err;
+}
+
+/*
+ * vsock_vmci_stream_setsockopt --
+ *
+ * Set a socket option on a stream socket
+ *
+ * Results: 0 on success, negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int vsock_vmci_stream_setsockopt(struct socket *sock,
+ int level,
+ int optname,
+ char __user *optval,
+ vsock_setsockopt_len_type optlen)
+{
+ int err;
+ struct sock *sk;
+ vsock_vmci_sock *vsk;
+ u64 val;
+
+ if (level != AF_VSOCK)
+ return -ENOPROTOOPT;
+
+#define COPY_IN(_v) \
+ do { \
+ if (optlen < sizeof _v) { \
+ err = -EINVAL; \
+ goto exit; \
+ } \
+ if (copy_from_user(&_v, optval, sizeof _v) != 0) { \
+ err = -EFAULT; \
+ goto exit; \
+ } \
+ } while (0)
+
+ err = 0;
+ sk = sock->sk;
+ vsk = vsock_sk(sk);
+
+ ASSERT(vsk->queue_pair_min_size <= vsk->queue_pair_size &&
+ vsk->queue_pair_size <= vsk->queue_pair_max_size);
+
+ lock_sock(sk);
+
+ switch (optname) {
+ case SO_VMCI_BUFFER_SIZE:
+ COPY_IN(val);
+ if (val < vsk->queue_pair_min_size)
+ vsk->queue_pair_min_size = val;
+
+ if (val > vsk->queue_pair_max_size)
+ vsk->queue_pair_max_size = val;
+
+ vsk->queue_pair_size = val;
+ break;
+
+ case SO_VMCI_BUFFER_MAX_SIZE:
+ COPY_IN(val);
+ if (val < vsk->queue_pair_size)
+ vsk->queue_pair_size = val;
+
+ vsk->queue_pair_max_size = val;
+ break;
+
+ case SO_VMCI_BUFFER_MIN_SIZE:
+ COPY_IN(val);
+ if (val > vsk->queue_pair_size)
+ vsk->queue_pair_size = val;
+
+ vsk->queue_pair_min_size = val;
+ break;
+
+ case SO_VMCI_CONNECT_TIMEOUT: {
+ struct timeval tv;
+ COPY_IN(tv);
+ if (tv.tv_sec >= 0 && tv.tv_usec < USEC_PER_SEC &&
+ tv.tv_sec < (MAX_SCHEDULE_TIMEOUT / HZ - 1)) {
+ vsk->connect_timeout = tv.tv_sec * HZ +
+ DIV_ROUND_UP(tv.tv_usec, (1000000 / HZ));
+ if (vsk->connect_timeout == 0)
+ vsk->connect_timeout =
+ VSOCK_DEFAULT_CONNECT_TIMEOUT;
+
+ } else {
+ err = -ERANGE;
+ }
+ break;
+ }
+
+ default:
+ err = -ENOPROTOOPT;
+ break;
+ }
+
+#undef COPY_IN
+
+ ASSERT(vsk->queue_pair_min_size <= vsk->queue_pair_size &&
+ vsk->queue_pair_size <= vsk->queue_pair_max_size);
+exit:
+ release_sock(sk);
+ return err;
+}
+
+/*
+ * vsock_vmci_stream_getsockopt --
+ *
+ * Get a socket option for a stream socket
+ *
+ * Results: 0 on success, negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int vsock_vmci_stream_getsockopt(struct socket *sock,
+ int level, int optname,
+ char __user *optval,
+ int __user *optlen)
+{
+ int err;
+ int len;
+ struct sock *sk;
+ vsock_vmci_sock *vsk;
+
+ if (level != AF_VSOCK)
+ return -ENOPROTOOPT;
+
+ err = get_user(len, optlen);
+ if (err != 0)
+ return err;
+
+#define COPY_OUT(_v) \
+ do { \
+ if (len < sizeof _v) \
+ return -EINVAL; \
+ \
+ len = sizeof _v; \
+ if (copy_to_user(optval, &_v, len) != 0) \
+ return -EFAULT; \
+ \
+ } while (0)
+
+ err = 0;
+ sk = sock->sk;
+ vsk = vsock_sk(sk);
+
+ switch (optname) {
+ case SO_VMCI_BUFFER_SIZE:
+ COPY_OUT(vsk->queue_pair_size);
+ break;
+
+ case SO_VMCI_BUFFER_MAX_SIZE:
+ COPY_OUT(vsk->queue_pair_max_size);
+ break;
+
+ case SO_VMCI_BUFFER_MIN_SIZE:
+ COPY_OUT(vsk->queue_pair_min_size);
+ break;
+
+ case SO_VMCI_CONNECT_TIMEOUT: {
+ struct timeval tv;
+ tv.tv_sec = vsk->connect_timeout / HZ;
+ tv.tv_usec =
+ (vsk->connect_timeout -
+ tv.tv_sec * HZ) * (1000000 / HZ);
+ COPY_OUT(tv);
+ break;
+ }
+ default:
+ return -ENOPROTOOPT;
+ }
+
+ err = put_user(len, optlen);
+ if (err != 0)
+ return -EFAULT;
+
+#undef COPY_OUT
+
+ return 0;
+}
+
+/*
+ * vsock_vmci_stream_sendmsg --
+ *
+ * Sends a message on the socket.
+ *
+ * Results: Number of bytes sent on success, negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_stream_sendmsg(struct kiocb *kiocb,
+ struct socket *sock, struct msghdr *msg, size_t len)
+{
+ struct sock *sk;
+ vsock_vmci_sock *vsk;
+ ssize_t total_written;
+ long timeout;
+ int err;
+ vsock_vmci_send_notify_data send_data;
+
+ DEFINE_WAIT(wait);
+
+ sk = sock->sk;
+ vsk = vsock_sk(sk);
+ total_written = 0;
+ err = 0;
+
+ if (msg->msg_flags & MSG_OOB)
+ return -EOPNOTSUPP;
+
+ lock_sock(sk);
+
+ /* Callers should not provide a destination with stream sockets. */
+ if (msg->msg_namelen) {
+ err = sk->sk_state == SS_CONNECTED ? -EISCONN : -EOPNOTSUPP;
+ goto out;
+ }
+
+ /* Send data only if both sides are not shutdown in the direction. */
+ if (sk->sk_shutdown & SEND_SHUTDOWN ||
+ vsk->peer_shutdown & RCV_SHUTDOWN) {
+ err = -EPIPE;
+ goto out;
+ }
+
+ if (sk->sk_state != SS_CONNECTED ||
+ !vsock_addr_bound(&vsk->local_addr)) {
+ err = -ENOTCONN;
+ goto out;
+ }
+
+ if (!vsock_addr_bound(&vsk->remote_addr)) {
+ err = -EDESTADDRREQ;
+ goto out;
+ }
+
+ /*
+ * Wait for room in the produce queue to enqueue our user's data.
+ */
+ timeout = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT);
+
+ NOTIFYCALLRET(vsk, err, send_init, sk, &send_data);
+ if (err < 0)
+ goto out;
+
+ prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
+
+ while (total_written < len) {
+ ssize_t written;
+
+ while (vsock_vmci_stream_has_space(vsk) == 0 &&
+ sk->sk_err == 0 &&
+ !(sk->sk_shutdown & SEND_SHUTDOWN) &&
+ !(vsk->peer_shutdown & RCV_SHUTDOWN)) {
+
+ /* Don't wait for non-blocking sockets. */
+ if (timeout == 0) {
+ err = -EAGAIN;
+ goto out_wait;
+ }
+
+ NOTIFYCALLRET(vsk, err, send_pre_block, sk, &send_data);
+
+ if (err < 0)
+ goto out_wait;
+
+ release_sock(sk);
+ timeout = schedule_timeout(timeout);
+ lock_sock(sk);
+ if (signal_pending(current)) {
+ err = sock_intr_errno(timeout);
+ goto out_wait;
+ } else if (timeout == 0) {
+ err = -EAGAIN;
+ goto out_wait;
+ }
+
+ prepare_to_wait(sk_sleep(sk), &wait,
+ TASK_INTERRUPTIBLE);
+ }
+
+ /*
+ * These checks occur both as part of and after the loop
+ * conditional since we need to check before and after
+ * sleeping.
+ */
+ if (sk->sk_err) {
+ err = -sk->sk_err;
+ goto out_wait;
+ } else if ((sk->sk_shutdown & SEND_SHUTDOWN) ||
+ (vsk->peer_shutdown & RCV_SHUTDOWN)) {
+ err = -EPIPE;
+ goto out_wait;
+ }
+
+ VSOCK_STATS_STREAM_PRODUCE_HIST(vsk);
+
+ NOTIFYCALLRET(vsk, err, send_pre_enqueue, sk, &send_data);
+ if (err < 0)
+ goto out_wait;
+
+ /*
+ * Note that enqueue will only write as many bytes as are free
+ * in the produce queue, so we don't need to ensure len is
+ * smaller than the queue size. It is the caller's
+ * responsibility to check how many bytes we were able to send.
+ */
+
+ written = vmci_qpair_enquev(vsk->qpair, msg->msg_iov,
+ len - total_written, 0);
+ if (written < 0) {
+ err = -ENOMEM;
+ goto out_wait;
+ }
+
+ total_written += written;
+
+ NOTIFYCALLRET(vsk, err, send_post_enqueue, sk, written,
+ &send_data);
+ if (err < 0)
+ goto out_wait;
+
+ }
+
+ ASSERT(total_written <= INT_MAX);
+
+out_wait:
+ if (total_written > 0) {
+ VSOCK_STATS_STREAM_PRODUCE(total_written);
+ err = total_written;
+ }
+ finish_wait(sk_sleep(sk), &wait);
+out:
+ release_sock(sk);
+ return err;
+}
+
+/*
+ * vsock_vmci_dgram_recvmsg --
+ *
+ * Receives a datagram and places it in the caller's msg.
+ *
+ * Results: The size of the payload on success, negative value on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_dgram_recvmsg(struct kiocb *kiocb,
+ struct socket *sock,
+ struct msghdr *msg, size_t len, int flags)
+{
+ int err;
+ int noblock;
+ struct sock *sk;
+ struct vmci_datagram *dg;
+ size_t payload_len;
+ struct sk_buff *skb;
+
+ sk = sock->sk;
+ noblock = flags & MSG_DONTWAIT;
+
+ if (flags & MSG_OOB || flags & MSG_ERRQUEUE)
+ return -EOPNOTSUPP;
+
+ /* Retrieve the head sk_buff from the socket's receive queue. */
+ err = 0;
+ skb = skb_recv_datagram(sk, flags, noblock, &err);
+ if (err)
+ return err;
+
+ if (!skb)
+ return -EAGAIN;
+
+ dg = (struct vmci_datagram *) skb->data;
+ if (!dg)
+ /* err is 0, meaning we read zero bytes. */
+ goto out;
+
+ payload_len = dg->payload_size;
+ /* Ensure the sk_buff matches the payload size claimed in the packet. */
+ if (payload_len != skb->len - sizeof *dg) {
+ err = -EINVAL;
+ goto out;
+ }
+
+ if (payload_len > len) {
+ payload_len = len;
+ msg->msg_flags |= MSG_TRUNC;
+ }
+
+ /* Place the datagram payload in the user's iovec. */
+ err =
+ skb_copy_datagram_iovec(skb, sizeof *dg, msg->msg_iov, payload_len);
+ if (err)
+ goto out;
+
+ msg->msg_namelen = 0;
+ if (msg->msg_name) {
+ struct sockaddr_vm *vmci_addr;
+
+ /* Provide the address of the sender. */
+ vmci_addr = (struct sockaddr_vm *)msg->msg_name;
+ vsock_addr_init(vmci_addr,
+ VMCI_HANDLE_TO_CONTEXT_ID(dg->src),
+ VMCI_HANDLE_TO_RESOURCE_ID(dg->src));
+ msg->msg_namelen = sizeof *vmci_addr;
+ }
+ err = payload_len;
+
+out:
+ skb_free_datagram(sk, skb);
+ return err;
+}
+
+/*
+ * vsock_vmci_stream_recvmsg --
+ *
+ * Receives a datagram and places it in the caller's msg.
+ *
+ * Results: The size of the payload on success, negative value on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_stream_recvmsg(struct kiocb *kiocb,
+ struct socket *sock,
+ struct msghdr *msg, size_t len, int flags)
+{
+ struct sock *sk;
+ vsock_vmci_sock *vsk;
+ int err;
+ size_t target;
+ ssize_t copied;
+ long timeout;
+
+ vsock_vmci_recv_notify_data recv_data;
+
+ DEFINE_WAIT(wait);
+
+ sk = sock->sk;
+ vsk = vsock_sk(sk);
+ err = 0;
+
+ lock_sock(sk);
+
+ if (sk->sk_state != SS_CONNECTED) {
+ /*
+ * Recvmsg is supposed to return 0 if a peer performs an
+ * orderly shutdown. Differentiate between that case and when a
+ * peer has not connected or a local shutdown occured with the
+ * SOCK_DONE flag.
+ */
+ if (sock_flag(sk, SOCK_DONE))
+ err = 0;
+ else
+ err = -ENOTCONN;
+
+ goto out;
+ }
+
+ if (flags & MSG_OOB) {
+ err = -EOPNOTSUPP;
+ goto out;
+ }
+
+ /*
+ * We don't check peer_shutdown flag here since peer may actually shut
+ * down, but there can be data in the VMCI queue that local socket can
+ * receive.
+ */
+ if (sk->sk_shutdown & RCV_SHUTDOWN) {
+ err = 0;
+ goto out;
+ }
+
+ /*
+ * It is valid on Linux to pass in a zero-length receive buffer. This
+ * is not an error. We may as well bail out now. Note that if we
+ * don't, we will fail "ASSERT(copied >= target)" after we dequeue,
+ * because the minimum target is always 1 byte.
+ */
+ if (!len) {
+ err = 0;
+ goto out;
+ }
+
+ /*
+ * We must not copy less than target bytes into the user's buffer
+ * before returning successfully, so we wait for the consume queue to
+ * have that much data to consume before dequeueing. Note that this
+ * makes it impossible to handle cases where target is greater than the
+ * queue size.
+ */
+ target = sock_rcvlowat(sk, flags & MSG_WAITALL, len);
+ if (target >= vsk->consume_size) {
+ err = -ENOMEM;
+ goto out;
+ }
+ timeout = sock_rcvtimeo(sk, flags & MSG_DONTWAIT);
+ copied = 0;
+
+ NOTIFYCALLRET(vsk, err, recv_init, sk, target, &recv_data);
+ if (err < 0)
+ goto out;
+
+ prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
+
+ while (1) {
+ s64 ready = vsock_vmci_stream_has_data(vsk);
+
+ if (ready < 0) {
+ /*
+ * Invalid queue pair content. XXX This should be
+ * changed to a connection reset in a later change.
+ */
+
+ err = -ENOMEM;
+ goto out_wait;
+ } else if (ready > 0) {
+ ssize_t read;
+
+ VSOCK_STATS_STREAM_CONSUME_HIST(vsk);
+
+ NOTIFYCALLRET(vsk, err, recv_pre_dequeue, sk, target,
+ &recv_data);
+ if (err < 0)
+ break;
+
+ if (flags & MSG_PEEK)
+ read =
+ vmci_qpair_peekv(vsk->qpair, msg->msg_iov,
+ len - copied, 0);
+ else
+ read =
+ vmci_qpair_dequev(vsk->qpair, msg->msg_iov,
+ len - copied, 0);
+
+ if (read < 0) {
+ err = -ENOMEM;
+ break;
+ }
+
+ ASSERT(read <= INT_MAX);
+ copied += read;
+
+ NOTIFYCALLRET(vsk, err, recv_post_dequeue, sk, target,
+ read, !(flags & MSG_PEEK), &recv_data);
+ if (err < 0)
+ goto out_wait;
+
+ if (read >= target || flags & MSG_PEEK)
+ break;
+
+ target -= read;
+ } else {
+ if (sk->sk_err != 0 || (sk->sk_shutdown & RCV_SHUTDOWN)
+ || (vsk->peer_shutdown & SEND_SHUTDOWN)) {
+ break;
+ }
+ /* Don't wait for non-blocking sockets. */
+ if (timeout == 0) {
+ err = -EAGAIN;
+ break;
+ }
+
+ NOTIFYCALLRET(vsk, err, recv_pre_block, sk, target,
+ &recv_data);
+ if (err < 0)
+ break;
+
+ release_sock(sk);
+ timeout = schedule_timeout(timeout);
+ lock_sock(sk);
+
+ if (signal_pending(current)) {
+ err = sock_intr_errno(timeout);
+ break;
+ } else if (timeout == 0) {
+ err = -EAGAIN;
+ break;
+ }
+
+ prepare_to_wait(sk_sleep(sk), &wait,
+ TASK_INTERRUPTIBLE);
+ }
+ }
+
+ if (sk->sk_err)
+ err = -sk->sk_err;
+ else if (sk->sk_shutdown & RCV_SHUTDOWN)
+ err = 0;
+
+ if (copied > 0) {
+ /*
+ * We only do these additional bookkeeping/notification steps
+ * if we actually copied something out of the queue pair
+ * instead of just peeking ahead.
+ */
+
+ if (!(flags & MSG_PEEK)) {
+ VSOCK_STATS_STREAM_CONSUME(copied);
+
+ /*
+ * If the other side has shutdown for sending and there
+ * is nothing more to read, then modify the socket
+ * state.
+ */
+ if (vsk->peer_shutdown & SEND_SHUTDOWN) {
+ if (vsock_vmci_stream_has_data(vsk) <= 0) {
+ sk->sk_state = SS_UNCONNECTED;
+ sock_set_flag(sk, SOCK_DONE);
+ sk->sk_state_change(sk);
+ }
+ }
+ }
+ err = copied;
+ }
+
+out_wait:
+ finish_wait(sk_sleep(sk), &wait);
+out:
+ release_sock(sk);
+ return err;
+}
+
+/*
+ * Protocol operation.
+ */
+
+/*
+ * vsock_vmci_create --
+ *
+ * Creates a VSocket socket.
+ *
+ * Results: Zero on success, negative error code on failure.
+ *
+ * Side effects: Socket count is incremented.
+ */
+
+static int
+vsock_vmci_create(struct net *net, struct socket *sock, int protocol, int kern)
+{
+ if (!sock)
+ return -EINVAL;
+
+ if (protocol)
+ return -EPROTONOSUPPORT;
+
+ switch (sock->type) {
+ case SOCK_DGRAM:
+ sock->ops = &vsock_vmci_dgram_ops;
+ break;
+ case SOCK_STREAM:
+ sock->ops = &vsock_vmci_stream_ops;
+ break;
+ default:
+ return -ESOCKTNOSUPPORT;
+ }
+
+ sock->state = SS_UNCONNECTED;
+
+ return __vsock_vmci_create(net, sock, NULL, GFP_KERNEL,
+ 0) ? 0 : -ENOMEM;
+}
+
+/*
+ * Device operations.
+ */
+
+static long vsock_vmci_dev_do_ioctl(struct file *filp,
+ unsigned int cmd, void __user *ptr)
+{
+ static const u16 parts[4] = { VSOCK_DRIVER_VERSION_COMMAS };
+ u32 __user *p = ptr;
+ int retval = 0;
+ u32 version;
+
+ switch (cmd) {
+ case IOCTL_VMCI_SOCKETS_VERSION:
+ version = VMCI_SOCKETS_MAKE_VERSION(parts);
+ if (put_user(version, p) != 0)
+ retval = -EFAULT;
+ break;
+
+ case IOCTL_VMCI_SOCKETS_GET_AF_VALUE:
+ if (put_user(AF_VSOCK, p) != 0)
+ retval = -EFAULT;
+
+ break;
+
+ case IOCTL_VMCI_SOCKETS_GET_LOCAL_CID:
+ if (put_user(vmci_get_context_id(), p) != 0)
+ retval = -EFAULT;
+
+ break;
+
+ default:
+ pr_err("Unknown ioctl %d\n", cmd);
+ retval = -EINVAL;
+ }
+
+ return retval;
+}
+
+static long vsock_vmci_dev_ioctl(struct file *filp,
+ unsigned int cmd, unsigned long arg)
+{
+ return vsock_vmci_dev_do_ioctl(filp, cmd, (void __user *)arg);
+}
+
+#ifdef CONFIG_COMPAT
+static long vsock_vmci_dev_compat_ioctl(struct file *filp,
+ unsigned int cmd, unsigned long arg)
+{
+ return vsock_vmci_dev_do_ioctl(filp, cmd, compat_ptr(arg));
+}
+#endif
+
+static const struct file_operations vsock_vmci_device_ops = {
+ .owner = THIS_MODULE,
+ .unlocked_ioctl = vsock_vmci_dev_ioctl,
+#ifdef CONFIG_COMPAT
+ .compat_ioctl = vsock_vmci_dev_compat_ioctl,
+#endif
+ .open = nonseekable_open,
+};
+
+static struct miscdevice vsock_vmci_device = {
+ .name = "vsock",
+ .minor = MISC_DYNAMIC_MINOR,
+ .fops = &vsock_vmci_device_ops,
+};
+
+
+/*
+ * Module operations.
+ */
+
+/*
+ * vsock_vmci_init --
+ *
+ * Initialization routine for the VSockets module.
+ *
+ * Results: Zero on success, error code on failure.
+ *
+ * Side effects: The VSocket protocol family and socket operations are
+ * registered.
+ */
+
+static int __init vsock_vmci_init(void)
+{
+ int err;
+
+ request_module("vmci");
+
+ err = misc_register(&vsock_vmci_device);
+ if (err) {
+ pr_err("Failed to register misc device\n");
+ return -ENOENT;
+ }
+
+ err = vsock_vmci_register_with_vmci();
+ if (err) {
+ pr_err("Cannot register with VMCI device.\n");
+ goto err_misc_deregister;
+ }
+
+ err = proto_register(&vsock_vmci_proto, 1); /* we want our slab */
+ if (err) {
+ pr_err("Cannot register vsock protocol.\n");
+ goto err_unregister_with_vmci;
+ }
+
+ err = sock_register(&vsock_vmci_family_ops);
+ if (err) {
+ pr_err("could not register af_vsock (%d) address family: %d\n",
+ AF_VSOCK, err);
+ goto err_unregister_proto;
+ }
+
+ vsock_vmci_init_tables();
+ return 0;
+
+err_unregister_proto:
+ proto_unregister(&vsock_vmci_proto);
+err_unregister_with_vmci:
+ vsock_vmci_unregister_with_vmci();
+err_misc_deregister:
+ misc_deregister(&vsock_vmci_device);
+ return err;
+}
+
+/*
+ * VSocketVmciExit --
+ *
+ * VSockets module exit routine.
+ *
+ * Results: None.
+ *
+ * Side effects: Unregisters VSocket protocol family and socket operations.
+ */
+
+static void __exit vsock_vmci_exit(void)
+{
+ misc_deregister(&vsock_vmci_device);
+ sock_unregister(AF_VSOCK);
+ proto_unregister(&vsock_vmci_proto);
+ /* Need reset ? */
+ VSOCK_STATS_RESET();
+ vsock_vmci_unregister_with_vmci();
+}
+
+module_init(vsock_vmci_init);
+module_exit(vsock_vmci_exit);
+
+MODULE_AUTHOR("VMware, Inc.");
+MODULE_DESCRIPTION("VMware Virtual Socket Family");
+MODULE_VERSION(VSOCK_DRIVER_VERSION_STRING);
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS("vmware_vsock");
diff --git a/net/vmw_vsock/af_vsock.h b/net/vmw_vsock/af_vsock.h
new file mode 100644
index 0000000..c434afc
--- /dev/null
+++ b/net/vmw_vsock/af_vsock.h
@@ -0,0 +1,179 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2007-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * af_vsock.h --
+ *
+ * Definitions for Linux VSockets module.
+ */
+
+#ifndef __AF_VSOCK_H__
+#define __AF_VSOCK_H__
+
+#include <linux/kernel.h>
+#include <linux/workqueue.h>
+#include <linux/vmw_vmci_defs.h>
+#include <linux/vmw_vmci_api.h>
+
+#include "vsock_common.h"
+#include "vsock_packet.h"
+#include "notify.h"
+
+#define vsock_sk(__sk) ((vsock_vmci_sock *)__sk)
+#define sk_vsock(__vsk) (&(__vsk)->sk)
+
+typedef struct vsock_vmci_sock {
+ /* sk must be the first member. */
+ struct sock sk;
+ struct sockaddr_vm local_addr;
+ struct sockaddr_vm remote_addr;
+ /* Links for the global tables of bound and connected sockets. */
+ struct list_head bound_table;
+ struct list_head connected_table;
+ /*
+ * Accessed without the socket lock held. This means it can never be
+ * modified outsided of socket create or destruct.
+ */
+ bool trusted;
+ bool cached_peer_allow_dgram; /* Dgram communication allowed to
+ * cached peer? */
+ vmci_id cached_peer; /* Context ID of last dgram destination check. */
+ const struct cred *owner;
+ struct vmci_handle dg_handle; /* For SOCK_DGRAM only. */
+ /* Rest are SOCK_STREAM only. */
+ struct vmci_handle qp_handle;
+ struct vmci_qp *qpair;
+ u64 produce_size;
+ u64 consume_size;
+ u64 queue_pair_size;
+ u64 queue_pair_min_size;
+ u64 queue_pair_max_size;
+ long connect_timeout;
+ vsock_vmci_notify notify;
+ vsock_vmci_notify_ops *notify_ops;
+ vmci_id attach_sub_id;
+ vmci_id detach_sub_id;
+ /* Listening socket that this came from. */
+ struct sock *listener;
+ /*
+ * Used for pending list and accept queue during connection handshake.
+ * The listening socket is the head for both lists. Sockets created
+ * for connection requests are placed in the pending list until they
+ * are connected, at which point they are put in the accept queue list
+ * so they can be accepted in accept(). If accept() cannot accept the
+ * connection, it is marked as rejected so the cleanup function knows
+ * to clean up the socket.
+ */
+ struct list_head pending_links;
+ struct list_head accept_queue;
+ bool rejected;
+ struct delayed_work dwork;
+ u32 peer_shutdown;
+ bool sent_request;
+ bool ignore_connecting_rst;
+} vsock_vmci_sock;
+
+int vsock_vmci_send_control_pkt_bh(struct sockaddr_vm *src,
+ struct sockaddr_vm *dst,
+ vsock_packet_type type,
+ u64 size,
+ u64 mode,
+ vsock_waiting_info *wait,
+ struct vmci_handle handle);
+int vsock_vmci_reply_control_pkt_fast(vsock_packet *pkt,
+ vsock_packet_type type, u64 size,
+ u64 mode, vsock_waiting_info *wait,
+ struct vmci_handle handle);
+int vsock_vmci_send_control_pkt(struct sock *sk, vsock_packet_type type,
+ u64 size, u64 mode,
+ vsock_waiting_info *wait,
+ vsock_proto_version version,
+ struct vmci_handle handle);
+
+s64 vsock_vmci_stream_has_data(vsock_vmci_sock *vsk);
+s64 vsock_vmci_stream_has_space(vsock_vmci_sock *vsk);
+
+#define VSOCK_SEND_RESET_BH(_dst, _src, _pkt) \
+ ((_pkt)->type == VSOCK_PACKET_TYPE_RST) ? \
+ 0 : \
+ vsock_vmci_send_control_pkt_bh( \
+ _dst, _src, \
+ VSOCK_PACKET_TYPE_RST, 0, \
+ 0, NULL, VMCI_INVALID_HANDLE)
+#define VSOCK_SEND_INVALID_BH(_dst, _src) \
+ vsock_vmci_send_control_pkt_bh(_dst, _src, \
+ VSOCK_PACKET_TYPE_INVALID, 0, \
+ 0, NULL, VMCI_INVALID_HANDLE)
+#define VSOCK_SEND_WROTE_BH(_dst, _src) \
+ vsock_vmci_send_control_pkt_bh(_dst, _src, VSOCK_PACKET_TYPE_WROTE, 0, \
+ 0, NULL, VMCI_INVALID_HANDLE)
+#define VSOCK_SEND_READ_BH(_dst, _src) \
+ vsock_vmci_send_control_pkt_bh((_dst), (_src), \
+ VSOCK_PACKET_TYPE_READ, 0, \
+ 0, NULL, VMCI_INVALID_HANDLE)
+#define VSOCK_SEND_RESET(_sk, _pkt) \
+ ((_pkt)->type == VSOCK_PACKET_TYPE_RST) ? \
+ 0 : \
+ vsock_vmci_send_control_pkt( \
+ _sk, VSOCK_PACKET_TYPE_RST, \
+ 0, 0, NULL, VSOCK_PROTO_INVALID, \
+ VMCI_INVALID_HANDLE)
+#define VSOCK_SEND_NEGOTIATE(_sk, _size) \
+ vsock_vmci_send_control_pkt(_sk, VSOCK_PACKET_TYPE_NEGOTIATE, \
+ _size, 0, NULL, VSOCK_PROTO_INVALID, \
+ VMCI_INVALID_HANDLE)
+#define VSOCK_SEND_NEGOTIATE2(_sk, _size, signal_proto) \
+ vsock_vmci_send_control_pkt(_sk, VSOCK_PACKET_TYPE_NEGOTIATE2, \
+ _size, 0, NULL, signal_proto, \
+ VMCI_INVALID_HANDLE)
+#define VSOCK_SEND_QP_OFFER(_sk, _handle) \
+ vsock_vmci_send_control_pkt(_sk, VSOCK_PACKET_TYPE_OFFER, \
+ 0, 0, NULL, VSOCK_PROTO_INVALID, _handle)
+#define VSOCK_SEND_CONN_REQUEST(_sk, _size) \
+ vsock_vmci_send_control_pkt(_sk, VSOCK_PACKET_TYPE_REQUEST, \
+ _size, 0, NULL, VSOCK_PROTO_INVALID, \
+ VMCI_INVALID_HANDLE)
+#define VSOCK_SEND_CONN_REQUEST2(_sk, _size, signal_proto) \
+ vsock_vmci_send_control_pkt(_sk, VSOCK_PACKET_TYPE_REQUEST2, \
+ _size, 0, NULL, signal_proto, \
+ VMCI_INVALID_HANDLE)
+#define VSOCK_SEND_ATTACH(_sk, _handle) \
+ vsock_vmci_send_control_pkt(_sk, VSOCK_PACKET_TYPE_ATTACH, \
+ 0, 0, NULL, VSOCK_PROTO_INVALID, _handle)
+#define VSOCK_SEND_WROTE(_sk) \
+ vsock_vmci_send_control_pkt(_sk, VSOCK_PACKET_TYPE_WROTE, \
+ 0, 0, NULL, VSOCK_PROTO_INVALID, \
+ VMCI_INVALID_HANDLE)
+#define VSOCK_SEND_READ(_sk) \
+ vsock_vmci_send_control_pkt(_sk, VSOCK_PACKET_TYPE_READ, \
+ 0, 0, NULL, VSOCK_PROTO_INVALID, \
+ VMCI_INVALID_HANDLE)
+#define VSOCK_SEND_SHUTDOWN(_sk, _mode) \
+ vsock_vmci_send_control_pkt(_sk, VSOCK_PACKET_TYPE_SHUTDOWN, \
+ 0, _mode, NULL, VSOCK_PROTO_INVALID, \
+ VMCI_INVALID_HANDLE)
+#define VSOCK_SEND_WAITING_WRITE(_sk, _wait_info) \
+ vsock_vmci_send_control_pkt(_sk, VSOCK_PACKET_TYPE_WAITING_WRITE, \
+ 0, 0, _wait_info, VSOCK_PROTO_INVALID, \
+ VMCI_INVALID_HANDLE)
+#define VSOCK_SEND_WAITING_READ(_sk, _wait_info) \
+ vsock_vmci_send_control_pkt(_sk, VSOCK_PACKET_TYPE_WAITING_READ, \
+ 0, 0, _wait_info, VSOCK_PROTO_INVALID, \
+ VMCI_INVALID_HANDLE)
+#define VSOCK_REPLY_RESET(_pkt) \
+ vsock_vmci_reply_control_pkt_fast(_pkt, VSOCK_PACKET_TYPE_RST, \
+ 0, 0, NULL, VMCI_INVALID_HANDLE)
+
+#endif /* __AF_VSOCK_H__ */
^ permalink raw reply related
* [PATCH 2/6] VSOCK: vsock address implementaion.
From: George Zhang @ 2012-11-07 18:45 UTC (permalink / raw)
To: netdev, linux-kernel, georgezhang, virtualization
Cc: pv-drivers, gregkh, davem
In-Reply-To: <20121107184340.10612.3372.stgit@promb-2n-dhcp175.eng.vmware.com>
VSOCK linux address code implementation.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
net/vmw_vsock/vsock_addr.c | 264 ++++++++++++++++++++++++++++++++++++++++++++
net/vmw_vsock/vsock_addr.h | 40 +++++++
2 files changed, 304 insertions(+), 0 deletions(-)
create mode 100644 net/vmw_vsock/vsock_addr.c
create mode 100644 net/vmw_vsock/vsock_addr.h
diff --git a/net/vmw_vsock/vsock_addr.c b/net/vmw_vsock/vsock_addr.c
new file mode 100644
index 0000000..c93a174
--- /dev/null
+++ b/net/vmw_vsock/vsock_addr.c
@@ -0,0 +1,264 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2007-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * vsockAddr.c --
+ *
+ * VSockets address implementation.
+ */
+
+/*
+ * These includes come before vsockCommon.h to ensure that VMware's ASSERT
+ * macro is used instead of Linux's irda.h definition.
+ */
+#include <linux/types.h>
+#include <linux/socket.h>
+#include <linux/stddef.h> /* for NULL */
+#include <net/sock.h>
+
+#include "vsock_common.h"
+
+/*
+ *
+ * vsock_addr_init --
+ *
+ * Initialize the given address with the given context id and port. This will
+ * clear the address, set the correct family, and add the given values.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+void vsock_addr_init(struct sockaddr_vm *addr, u32 cid, u32 port)
+{
+ memset(addr, 0, sizeof *addr);
+
+ addr->svm_family = AF_VSOCK;
+ addr->svm_cid = cid;
+ addr->svm_port = port;
+
+ BUG_ON(vsock_addr_validate(addr) != 0);
+}
+
+/*
+ *
+ * vsock_addr_validate --
+ *
+ * Try to validate the given address. The address must not be null and must
+ * have the correct address family. Any reserved fields must be zero.
+ *
+ * Results: 0 on success, EFAULT if the address is null, EAFNOSUPPORT if the
+ * address is of the wrong family, and EINVAL if the reserved fields are not
+ * zero.
+ *
+ * Side effects: None.
+ */
+
+int vsock_addr_validate(const struct sockaddr_vm *addr)
+{
+ if (!addr)
+ return -EFAULT;
+
+ if (addr->svm_family != AF_VSOCK)
+ return -EAFNOSUPPORT;
+
+ if (addr->svm_zero[0] != 0)
+ return -EINVAL;
+
+ return 0;
+}
+
+/*
+ *
+ * vsock_addr_bound --
+ *
+ * Determines whether the provided address is bound.
+ *
+ * Results: TRUE if the address structure is bound, FALSE otherwise.
+ *
+ * Side effects: None.
+ */
+
+bool vsock_addr_bound(const struct sockaddr_vm *addr)
+{
+ BUG_ON(!addr);
+
+ return addr->svm_port != VMADDR_PORT_ANY;
+}
+
+/*
+ *
+ * vsock_addr_unbind --
+ *
+ * Unbind the given addresss.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+void vsock_addr_unbind(struct sockaddr_vm *addr)
+{
+ vsock_addr_init(addr, VMADDR_CID_ANY, VMADDR_PORT_ANY);
+}
+
+/*
+ *
+ * vsock_addr_equals_addr --
+ *
+ * Determine if the given addresses are equal.
+ *
+ * Results: TRUE if the addresses are equal, FALSE otherwise.
+ *
+ * Side effects: None.
+ */
+
+bool vsock_addr_equals_addr(const struct sockaddr_vm *addr,
+ const struct sockaddr_vm *other)
+{
+ BUG_ON(vsock_addr_validate(addr) != 0);
+ BUG_ON(vsock_addr_validate(other) != 0);
+
+ return addr->svm_cid == other->svm_cid &&
+ addr->svm_port == other->svm_port;
+}
+
+/*
+ *
+ * vsock_addr_equals_addr_any --
+ *
+ * Determine if the given addresses are equal. Will accept either an exact
+ * match or one where the rids match and that either the cids match or are set
+ * to VMADDR_CID_ANY.
+ *
+ * Results: TRUE if the addresses are equal, FALSE otherwise.
+ *
+ * Side effects: None.
+ */
+
+bool vsock_addr_equals_addr_any(const struct sockaddr_vm *addr,
+ const struct sockaddr_vm *other)
+{
+ BUG_ON(vsock_addr_validate(addr) != 0);
+ BUG_ON(vsock_addr_validate(other) != 0);
+
+ return (addr->svm_cid == VMADDR_CID_ANY ||
+ other->svm_cid == VMADDR_CID_ANY ||
+ addr->svm_cid == other->svm_cid) &&
+ addr->svm_port == other->svm_port;
+}
+
+/*
+ *
+ * vsock_addr_equals_handle_port --
+ *
+ * Determines if the given address matches the given handle and port.
+ *
+ * Results: TRUE if the address matches the handle and port, FALSE otherwise.
+ *
+ * Side effects: None.
+ */
+
+bool vsock_addr_equals_handle_port(const struct sockaddr_vm *addr,
+ struct vmci_handle handle, u32 port)
+{
+ BUG_ON(vsock_addr_validate(addr) != 0);
+
+ return addr->svm_cid == VMCI_HANDLE_TO_CONTEXT_ID(handle) &&
+ addr->svm_port == port;
+}
+
+/*
+ *
+ * vsock_addr_cast --
+ *
+ * Try to cast the given generic address to a VM address. The given length
+ * must match that of a VM address and the address must be valid. The
+ * "out_addr" parameter contains the address if successful.
+ *
+ * Results: 0 on success, EFAULT if the length is too small. See
+ * vsock_addr_validate() for other possible return codes.
+ *
+ * Side effects: None.
+ */
+
+int vsock_addr_cast(const struct sockaddr *addr,
+ size_t len, struct sockaddr_vm **out_addr)
+{
+ BUG_ON(!out_addr);
+
+ if (len < sizeof **out_addr)
+ return -EFAULT;
+
+ *out_addr = (struct sockaddr_vm *)addr;
+ return vsock_addr_validate(*out_addr);
+}
+
+/*
+ *
+ * vsock_addr_socket_context_stream --
+ *
+ * Determines whether the provided context id represents a context that
+ * contains a stream socket endpoints.
+ *
+ * Results: TRUE if the context does have socket endpoints, FALSE otherwise.
+ *
+ * Side effects: None.
+ */
+
+bool vsock_addr_socket_context_stream(u32 cid)
+{
+ static const vmci_id non_socket_contexts[] = {
+ VMCI_HYPERVISOR_CONTEXT_ID,
+ VMCI_WELL_KNOWN_CONTEXT_ID,
+ };
+ int i;
+
+ BUILD_BUG_ON(sizeof cid != sizeof *non_socket_contexts);
+
+ for (i = 0; i < ARRAY_SIZE(non_socket_contexts); i++) {
+ if (cid == non_socket_contexts[i])
+ return false;
+
+ }
+
+ return true;
+}
+
+/*
+ *
+ * vsock_addr_socket_context_dgram --
+ *
+ * Determines whether the provided <context id, resource id> represent a
+ * protected datagram endpoint.
+ *
+ * Results: TRUE if the context does have socket endpoints, FALSE otherwise.
+ *
+ * Side effects: None.
+ */
+
+bool vsock_addr_socket_context_dgram(u32 cid, u32 rid)
+{
+ if (cid == VMCI_HYPERVISOR_CONTEXT_ID) {
+ /*
+ * Registrations of PBRPC Servers do not modify VMX/Hypervisor
+ * state and are allowed.
+ */
+ return rid == VMCI_UNITY_PBRPC_REGISTER;
+ }
+
+ return true;
+}
diff --git a/net/vmw_vsock/vsock_addr.h b/net/vmw_vsock/vsock_addr.h
new file mode 100644
index 0000000..18f023d
--- /dev/null
+++ b/net/vmw_vsock/vsock_addr.h
@@ -0,0 +1,40 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2007-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * vsockAddr.h --
+ *
+ * VSockets address constants, types and functions.
+ */
+
+#ifndef _VSOCK_ADDR_H_
+#define _VSOCK_ADDR_H_
+
+void vsock_addr_init(struct sockaddr_vm *addr, u32 cid, u32 port);
+int vsock_addr_validate(const struct sockaddr_vm *addr);
+bool vsock_addr_bound(const struct sockaddr_vm *addr);
+void vsock_addr_unbind(struct sockaddr_vm *addr);
+bool vsock_addr_equals_addr(const struct sockaddr_vm *addr,
+ const struct sockaddr_vm *other);
+bool vsock_addr_equals_addr_any(const struct sockaddr_vm *addr,
+ const struct sockaddr_vm *other);
+bool vsock_addr_equals_handle_port(const struct sockaddr_vm *addr,
+ struct vmci_handle handle, u32 port);
+int vsock_addr_cast(const struct sockaddr *addr, size_t len,
+ struct sockaddr_vm **out_addr);
+bool vsock_addr_socket_context_stream(u32 cid);
+bool vsock_addr_socket_context_dgram(u32 cid, u32 rid);
+
+#endif
^ permalink raw reply related
* [PATCH 3/6] VSOCK: notification implementation.
From: George Zhang @ 2012-11-07 18:45 UTC (permalink / raw)
To: netdev, linux-kernel, georgezhang, virtualization
Cc: pv-drivers, gregkh, davem
In-Reply-To: <20121107184340.10612.3372.stgit@promb-2n-dhcp175.eng.vmware.com>
VSOCK control notifications for VMCI Stream Sockets protocol.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
net/vmw_vsock/notify.c | 984 ++++++++++++++++++++++++++++++++++++++++++++++++
net/vmw_vsock/notify.h | 130 ++++++
2 files changed, 1114 insertions(+), 0 deletions(-)
create mode 100644 net/vmw_vsock/notify.c
create mode 100644 net/vmw_vsock/notify.h
diff --git a/net/vmw_vsock/notify.c b/net/vmw_vsock/notify.c
new file mode 100644
index 0000000..5c69c7c
--- /dev/null
+++ b/net/vmw_vsock/notify.c
@@ -0,0 +1,984 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2009-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * notify.c --
+ *
+ * Linux control notifications for the VMCI Stream Sockets protocol.
+ */
+
+#include <linux/types.h>
+
+#include <linux/socket.h>
+#include <linux/stddef.h> /* for NULL */
+#include <net/sock.h>
+
+#include "notify.h"
+#include "af_vsock.h"
+
+#define PKT_FIELD(vsk, field_name) \
+ (vsk)->notify.pkt.field_name
+
+#define VSOCK_MAX_DGRAM_RESENDS 10
+
+/*
+ *
+ * vsock_vmci_notify_waiting_write --
+ *
+ * Determines if the conditions have been met to notify a waiting writer.
+ *
+ * Results: true if a notification should be sent, false otherwise.
+ *
+ * Side effects: None.
+ */
+
+static bool vsock_vmci_notify_waiting_write(vsock_vmci_sock *vsk)
+{
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+ bool retval;
+ u64 notify_limit;
+
+ if (!PKT_FIELD(vsk, peer_waiting_write))
+ return false;
+
+#ifdef VSOCK_OPTIMIZATION_FLOW_CONTROL
+ /*
+ * When the sender blocks, we take that as a sign that the sender is
+ * faster than the receiver. To reduce the transmit rate of the sender,
+ * we delay the sending of the read notification by decreasing the
+ * write_notify_window. The notification is delayed until the number of
+ * bytes used in the queue drops below the write_notify_window.
+ */
+
+ if (!PKT_FIELD(vsk, peer_waiting_write_detected)) {
+ PKT_FIELD(vsk, peer_waiting_write_detected) = true;
+ if (PKT_FIELD(vsk, write_notify_window) < PAGE_SIZE) {
+ PKT_FIELD(vsk, write_notify_window) =
+ PKT_FIELD(vsk, write_notify_min_window);
+ } else {
+ PKT_FIELD(vsk, write_notify_window) -= PAGE_SIZE;
+ if (PKT_FIELD(vsk, write_notify_window) <
+ PKT_FIELD(vsk, write_notify_min_window))
+ PKT_FIELD(vsk, write_notify_window) =
+ PKT_FIELD(vsk, write_notify_min_window);
+
+ }
+ }
+ notify_limit = vsk->consume_size - PKT_FIELD(vsk, write_notify_window);
+#else
+ notify_limit = 0;
+#endif
+
+ /*
+ * For now we ignore the wait information and just see if the free
+ * space exceeds the notify limit. Note that improving this function
+ * to be more intelligent will not require a protocol change and will
+ * retain compatibility between endpoints with mixed versions of this
+ * function.
+ *
+ * The notify_limit is used to delay notifications in the case where
+ * flow control is enabled. Below the test is expressed in terms of
+ * free space in the queue: if free_space > ConsumeSize -
+ * write_notify_window then notify An alternate way of expressing this
+ * is to rewrite the expression to use the data ready in the receive
+ * queue: if write_notify_window > bufferReady then notify as
+ * free_space == ConsumeSize - bufferReady.
+ */
+ retval = vmci_qpair_consume_free_space(vsk->qpair) > notify_limit;
+#ifdef VSOCK_OPTIMIZATION_FLOW_CONTROL
+ if (retval) {
+ /*
+ * Once we notify the peer, we reset the detected flag so the
+ * next wait will again cause a decrease in the window size.
+ */
+
+ PKT_FIELD(vsk, peer_waiting_write_detected) = false;
+ }
+#endif
+ return retval;
+#else
+ return true;
+#endif
+}
+
+/*
+ *
+ * vsock_vmci_notify_waiting_read --
+ *
+ * Determines if the conditions have been met to notify a waiting reader.
+ *
+ * Results: true if a notification should be sent, false otherwise.
+ *
+ * Side effects: None.
+ */
+
+static bool vsock_vmci_notify_waiting_read(vsock_vmci_sock *vsk)
+{
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+ if (!PKT_FIELD(vsk, peer_waiting_read))
+ return false;
+
+ /*
+ * For now we ignore the wait information and just see if there is any
+ * data for our peer to read. Note that improving this function to be
+ * more intelligent will not require a protocol change and will retain
+ * compatibility between endpoints with mixed versions of this
+ * function.
+ */
+ return vmci_qpair_produce_buf_ready(vsk->qpair) > 0;
+#else
+ return true;
+#endif
+}
+
+/*
+ *
+ * vsock_vmci_handle_waiting_read --
+ *
+ * Handles an incoming waiting read message.
+ *
+ * Results: None.
+ *
+ * Side effects: May send a notification to the peer, may update socket's wait
+ * info structure.
+ */
+
+static void
+vsock_vmci_handle_waiting_read(struct sock *sk,
+ vsock_packet *pkt,
+ bool bottom_half,
+ struct sockaddr_vm *dst, struct sockaddr_vm *src)
+{
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+ vsock_vmci_sock *vsk;
+
+ vsk = vsock_sk(sk);
+
+ PKT_FIELD(vsk, peer_waiting_read) = true;
+ memcpy(&PKT_FIELD(vsk, peer_waiting_read_info), &pkt->u.wait,
+ sizeof PKT_FIELD(vsk, peer_waiting_read_info));
+
+ if (vsock_vmci_notify_waiting_read(vsk)) {
+ bool sent;
+
+ if (bottom_half)
+ sent = VSOCK_SEND_WROTE_BH(dst, src) > 0;
+ else
+ sent = VSOCK_SEND_WROTE(sk) > 0;
+
+ if (sent)
+ PKT_FIELD(vsk, peer_waiting_read) = false;
+
+ }
+#endif
+}
+
+/*
+ *
+ * vsock_vmci_handle_waiting_write --
+ *
+ * Handles an incoming waiting write message.
+ *
+ * Results: None.
+ *
+ * Side effects: May send a notification to the peer, may update socket's wait
+ * info structure.
+ */
+
+static void
+vsock_vmci_handle_waiting_write(struct sock *sk,
+ vsock_packet *pkt,
+ bool bottom_half,
+ struct sockaddr_vm *dst,
+ struct sockaddr_vm *src)
+{
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+ vsock_vmci_sock *vsk;
+
+ vsk = vsock_sk(sk);
+
+ PKT_FIELD(vsk, peer_waiting_write) = true;
+ memcpy(&PKT_FIELD(vsk, peer_waiting_write_info), &pkt->u.wait,
+ sizeof PKT_FIELD(vsk, peer_waiting_write_info));
+
+ if (vsock_vmci_notify_waiting_write(vsk)) {
+ bool sent;
+
+ if (bottom_half)
+ sent = VSOCK_SEND_READ_BH(dst, src) > 0;
+ else
+ sent = VSOCK_SEND_READ(sk) > 0;
+
+ if (sent)
+ PKT_FIELD(vsk, peer_waiting_write) = false;
+
+ }
+#endif
+}
+
+/*
+ *
+ * vsock_vmci_handle_read --
+ *
+ * Handles an incoming read message.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void
+vsock_vmci_handle_read(struct sock *sk,
+ vsock_packet *pkt,
+ bool bottom_half,
+ struct sockaddr_vm *dst, struct sockaddr_vm *src)
+{
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+ vsock_vmci_sock *vsk;
+
+ vsk = vsock_sk(sk);
+ PKT_FIELD(vsk, sent_waiting_write) = false;
+#endif
+
+ sk->sk_write_space(sk);
+}
+
+/*
+ *
+ * vsock_vmci_send_waiting_read --
+ *
+ * Sends a waiting read notification to this socket's peer.
+ *
+ * Results: true if the datagram is sent successfully, false otherwise.
+ *
+ * Side effects: Our peer will notify us when there is data to read from our
+ * consume queue.
+ */
+
+static bool vsock_vmci_send_waiting_read(struct sock *sk, u64 room_needed)
+{
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+ vsock_vmci_sock *vsk;
+ vsock_waiting_info waiting_info;
+ u64 tail;
+ u64 head;
+ u64 room_left;
+ bool ret;
+
+ vsk = vsock_sk(sk);
+
+ if (PKT_FIELD(vsk, sent_waiting_read))
+ return true;
+
+ if (PKT_FIELD(vsk, write_notify_window) < vsk->consume_size)
+ PKT_FIELD(vsk, write_notify_window) =
+ min(PKT_FIELD(vsk, write_notify_window) + PAGE_SIZE,
+ vsk->consume_size);
+
+ vmci_qpair_get_consume_indexes(vsk->qpair, &tail, &head);
+ room_left = vsk->consume_size - head;
+ if (room_needed >= room_left) {
+ waiting_info.offset = room_needed - room_left;
+ waiting_info.generation =
+ PKT_FIELD(vsk, consume_q_generation) + 1;
+ } else {
+ waiting_info.offset = head + room_needed;
+ waiting_info.generation = PKT_FIELD(vsk, consume_q_generation);
+ }
+
+ ret = VSOCK_SEND_WAITING_READ(sk, &waiting_info) > 0;
+ if (ret)
+ PKT_FIELD(vsk, sent_waiting_read) = true;
+
+ return ret;
+#else
+ return true;
+#endif
+}
+
+/*
+ *
+ * vsock_vmci_send_waiting_write --
+ *
+ * Sends a waiting write notification to this socket's peer.
+ *
+ * Results: true if the datagram is sent successfully or does not need to be
+ * sent. false otherwise.
+ *
+ * Side effects: Our peer will notify us when there is room to write in to our
+ * produce queue.
+ */
+
+static bool vsock_vmci_send_waiting_write(struct sock *sk, u64 room_needed)
+{
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+ vsock_vmci_sock *vsk;
+ vsock_waiting_info waiting_info;
+ u64 tail;
+ u64 head;
+ u64 room_left;
+ bool ret;
+
+ vsk = vsock_sk(sk);
+
+ if (PKT_FIELD(vsk, sent_waiting_write))
+ return true;
+
+ vmci_qpair_get_produce_indexes(vsk->qpair, &tail, &head);
+ room_left = vsk->produce_size - tail;
+ if (room_needed + 1 >= room_left) {
+ /* Wraps around to current generation. */
+ waiting_info.offset = room_needed + 1 - room_left;
+ waiting_info.generation = PKT_FIELD(vsk, produce_q_generation);
+ } else {
+ waiting_info.offset = tail + room_needed + 1;
+ waiting_info.generation =
+ PKT_FIELD(vsk, produce_q_generation) - 1;
+ }
+
+ ret = VSOCK_SEND_WAITING_WRITE(sk, &waiting_info) > 0;
+ if (ret)
+ PKT_FIELD(vsk, sent_waiting_write) = true;
+
+ return ret;
+#else
+ return true;
+#endif
+}
+
+/*
+ *
+ * vsock_vmci_send_read_notification --
+ *
+ * Sends a read notification to this socket's peer.
+ *
+ * Results: >= 0 if the datagram is sent successfully, negative error value
+ * otherwise.
+ *
+ * Side effects: None.
+ */
+
+static int vsock_vmci_send_read_notification(struct sock *sk)
+{
+ vsock_vmci_sock *vsk;
+ bool sent_read;
+ unsigned int retries;
+ int err;
+
+ vsk = vsock_sk(sk);
+ sent_read = false;
+ retries = 0;
+ err = 0;
+
+ if (vsock_vmci_notify_waiting_write(vsk)) {
+ /*
+ * Notify the peer that we have read, retrying the send on
+ * failure up to our maximum value. XXX For now we just log
+ * the failure, but later we should schedule a work item to
+ * handle the resend until it succeeds. That would require
+ * keeping track of work items in the vsk and cleaning them up
+ * upon socket close.
+ */
+ while (!(vsk->peer_shutdown & RCV_SHUTDOWN) &&
+ !sent_read && retries < VSOCK_MAX_DGRAM_RESENDS) {
+ err = VSOCK_SEND_READ(sk);
+ if (err >= 0)
+ sent_read = true;
+
+ retries++;
+ }
+
+ if (retries >= VSOCK_MAX_DGRAM_RESENDS)
+ printk
+ ("%p unable to send read notify to peer.\n",
+ sk);
+ else
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+ PKT_FIELD(vsk, peer_waiting_write) = false;
+#endif
+
+ }
+ return err;
+}
+
+/*
+ *
+ * vsock_vmci_handle_wrote --
+ *
+ * Handles an incoming wrote message.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void
+vsock_vmci_handle_wrote(struct sock *sk,
+ vsock_packet *pkt,
+ bool bottom_half,
+ struct sockaddr_vm *dst, struct sockaddr_vm *src)
+{
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+ vsock_vmci_sock *vsk;
+
+ vsk = vsock_sk(sk);
+ PKT_FIELD(vsk, sent_waiting_read) = false;
+#endif
+
+ sk->sk_data_ready(sk, 0);
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_socket_init --
+ *
+ * Function that is called after a socket is created and before any notify ops
+ * are used.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void vsock_vmci_notify_pkt_socket_init(struct sock *sk)
+{
+ vsock_vmci_sock *vsk;
+ vsk = vsock_sk(sk);
+
+ PKT_FIELD(vsk, write_notify_window) = PAGE_SIZE;
+ PKT_FIELD(vsk, write_notify_min_window) = PAGE_SIZE;
+ PKT_FIELD(vsk, peer_waiting_read) = false;
+ PKT_FIELD(vsk, peer_waiting_write) = false;
+ PKT_FIELD(vsk, peer_waiting_write_detected) = false;
+ PKT_FIELD(vsk, sent_waiting_read) = false;
+ PKT_FIELD(vsk, sent_waiting_write) = false;
+ PKT_FIELD(vsk, produce_q_generation) = 0;
+ PKT_FIELD(vsk, consume_q_generation) = 0;
+
+ memset(&PKT_FIELD(vsk, peer_waiting_read_info), 0,
+ sizeof PKT_FIELD(vsk, peer_waiting_read_info));
+ memset(&PKT_FIELD(vsk, peer_waiting_write_info), 0,
+ sizeof PKT_FIELD(vsk, peer_waiting_write_info));
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_socket_destruct --
+ *
+ * Function that is called when the socket is being released.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void vsock_vmci_notify_pkt_socket_destruct(struct sock *sk)
+{
+ return;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_poll_in --
+ *
+ * Called by the poll function to figure out if there is data to read and to
+ * setup future notifications if needed. Only called on sockets that aren't
+ * shutdown for recv.
+ *
+ * Results: 0 on success. Negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_poll_in(struct sock *sk,
+ size_t target, bool *data_ready_now)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ if (vsock_vmci_stream_has_data(vsk)) {
+ *data_ready_now = true;
+ } else {
+ /*
+ * We can't read right now because there is nothing in the
+ * queue. Ask for notifications when there is something to
+ * read.
+ */
+ if (sk->sk_state == SS_CONNECTED) {
+ if (!vsock_vmci_send_waiting_read(sk, 1))
+ return -1;
+
+ }
+ *data_ready_now = false;
+ }
+
+ return 0;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_poll_out
+ *
+ * Called by the poll function to figure out if there is space to write and to
+ * setup future notifications if needed. Only called on a connected socket that
+ * isn't shutdown for send.
+ *
+ * Results: 0 on success. Negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_poll_out(struct sock *sk,
+ size_t target, bool *space_avail_now)
+{
+ s64 produce_q_free_space;
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ produce_q_free_space = vsock_vmci_stream_has_space(vsk);
+ if (produce_q_free_space > 0) {
+ *space_avail_now = true;
+ return 0;
+ } else if (produce_q_free_space == 0) {
+ /*
+ * This is a connected socket but we can't currently send data.
+ * Notify the peer that we are waiting if the queue is full. We
+ * only send a waiting write if the queue is full because
+ * otherwise we end up in an infinite WAITING_WRITE, READ,
+ * WAITING_WRITE, READ, etc. loop. Treat failing to send the
+ * notification as a socket error, passing that back through
+ * the mask.
+ */
+ if (!vsock_vmci_send_waiting_write(sk, 1))
+ return -1;
+
+ *space_avail_now = false;
+ }
+
+ return 0;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_recv_init --
+ *
+ * Called at the start of a stream recv call with the socket lock held.
+ *
+ * Results: 0 on success. Negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_recv_init(struct sock *sk,
+ size_t target,
+ vsock_vmci_recv_notify_data *data)
+{
+ vsock_vmci_sock *vsk;
+
+ vsk = vsock_sk(sk);
+
+#ifdef VSOCK_OPTIMIZATION_WAITING_NOTIFY
+ data->consume_head = 0;
+ data->produce_tail = 0;
+#ifdef VSOCK_OPTIMIZATION_FLOW_CONTROL
+ data->notify_on_block = false;
+
+ if (PKT_FIELD(vsk, write_notify_min_window) < target + 1) {
+ ASSERT(target < vsk->consume_size);
+ PKT_FIELD(vsk, write_notify_min_window) = target + 1;
+ if (PKT_FIELD(vsk, write_notify_window) <
+ PKT_FIELD(vsk, write_notify_min_window)) {
+ /*
+ * If the current window is smaller than the new
+ * minimal window size, we need to reevaluate whether
+ * we need to notify the sender. If the number of ready
+ * bytes are smaller than the new window, we need to
+ * send a notification to the sender before we block.
+ */
+
+ PKT_FIELD(vsk, write_notify_window) =
+ PKT_FIELD(vsk, write_notify_min_window);
+ data->notify_on_block = true;
+ }
+ }
+#endif
+#endif
+
+ return 0;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_recv_pre_block --
+ *
+ * Called right before a socket is about to block with the socket lock held.
+ * The socket lock may have been released between the entry function and the
+ * preblock call.
+ *
+ * Note: This function may be called multiple times before the post block
+ * function is called.
+ *
+ * Results: 0 on success. Negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_recv_pre_block(struct sock *sk,
+ size_t target,
+ vsock_vmci_recv_notify_data *data)
+{
+ int err = 0;
+
+ /* Notify our peer that we are waiting for data to read. */
+ if (!vsock_vmci_send_waiting_read(sk, target)) {
+ err = -EHOSTUNREACH;
+ return err;
+ }
+#ifdef VSOCK_OPTIMIZATION_FLOW_CONTROL
+ if (data->notify_on_block) {
+ err = vsock_vmci_send_read_notification(sk);
+ if (err < 0)
+ return err;
+
+ data->notify_on_block = false;
+ }
+#endif
+
+ return err;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_recv_pre_dequeue --
+ *
+ * Called right before we dequeue / peek data from a socket.
+ *
+ * Results: 0 on success. Negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_recv_pre_dequeue(struct sock *sk,
+ size_t target,
+ vsock_vmci_recv_notify_data *data)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ /*
+ * Now consume up to len bytes from the queue. Note that since we have
+ * the socket locked we should copy at least ready bytes.
+ */
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+ vmci_qpair_get_consume_indexes(vsk->qpair,
+ &data->produce_tail,
+ &data->consume_head);
+#endif
+
+ return 0;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_recv_post_dequeue --
+ *
+ * Called right after we dequeue / peek data from a socket.
+ *
+ * Results: 0 on success. Negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_recv_post_dequeue(struct sock *sk,
+ size_t target,
+ ssize_t copied,
+ bool data_read,
+ vsock_vmci_recv_notify_data *data)
+{
+ vsock_vmci_sock *vsk;
+ int err;
+
+ vsk = vsock_sk(sk);
+ err = 0;
+
+ if (data_read) {
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+ /*
+ * Detect a wrap-around to maintain queue generation. Note
+ * that this is safe since we hold the socket lock across the
+ * two queue pair operations.
+ */
+ if (copied >= vsk->consume_size - data->consume_head)
+ PKT_FIELD(vsk, consume_q_generation)++;
+
+#endif
+
+ err = vsock_vmci_send_read_notification(sk);
+ if (err < 0)
+ return err;
+
+ }
+ return err;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_send_init --
+ *
+ * Called at the start of a stream send call with the socket lock held.
+ *
+ * Results: 0 on success. A negative error code on failure.
+ *
+ * Side effects:
+ */
+
+static int
+vsock_vmci_notify_pkt_send_init(struct sock *sk,
+ vsock_vmci_send_notify_data *data)
+{
+#ifdef VSOCK_OPTIMIZATION_WAITING_NOTIFY
+ data->consume_head = 0;
+ data->produce_tail = 0;
+#endif
+
+ return 0;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_send_pre_block --
+ *
+ * Called right before a socket is about to block with the socket lock held.
+ * The socket lock may have been released between the entry function and the
+ * preblock call.
+ *
+ * Note: This function may be called multiple times before the post block
+ * function is called.
+ *
+ * Results. 0 on success. A negative error code on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_send_pre_block(struct sock *sk,
+ vsock_vmci_send_notify_data *data)
+{
+ /* Notify our peer that we are waiting for room to write. */
+ if (!vsock_vmci_send_waiting_write(sk, 1))
+ return -EHOSTUNREACH;
+
+ return 0;
+}
+
+/*
+ *
+ * vsock_vmci_notifySendPreEnqueue --
+ *
+ * Called right before we Enqueue to a socket.
+ *
+ * Results: 0 on success. Negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_send_pre_enqueue(struct sock *sk,
+ vsock_vmci_send_notify_data *data)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+ vmci_qpair_get_produce_indexes(vsk->qpair,
+ &data->produce_tail,
+ &data->consume_head);
+#endif
+
+ return 0;
+}
+
+/*
+ *
+ * vsock_vmci_notifySendPostEnqueue --
+ *
+ * Called right after we enqueue data to a socket.
+ *
+ * Results: 0 on success. Negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_send_post_enqueue(struct sock *sk,
+ ssize_t written,
+ vsock_vmci_send_notify_data *data)
+{
+ int err = 0;
+ vsock_vmci_sock *vsk;
+ bool sent_wrote = false;
+ int retries = 0;
+
+ vsk = vsock_sk(sk);
+
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+ /*
+ * Detect a wrap-around to maintain queue generation. Note that this
+ * is safe since we hold the socket lock across the two queue pair
+ * operations.
+ */
+ if (written >= vsk->produce_size - data->produce_tail)
+ PKT_FIELD(vsk, produce_q_generation)++;
+
+#endif
+
+ if (vsock_vmci_notify_waiting_read(vsk)) {
+ /*
+ * Notify the peer that we have written, retrying the send on
+ * failure up to our maximum value. See the XXX comment for the
+ * corresponding piece of code in StreamRecvmsg() for potential
+ * improvements.
+ */
+ while (!(vsk->peer_shutdown & RCV_SHUTDOWN) &&
+ !sent_wrote && retries < VSOCK_MAX_DGRAM_RESENDS) {
+ err = VSOCK_SEND_WROTE(sk);
+ if (err >= 0)
+ sent_wrote = true;
+
+ retries++;
+ }
+
+ if (retries >= VSOCK_MAX_DGRAM_RESENDS) {
+ printk
+ (" %p unable to send wrote notify to peer.\n",
+ sk);
+ return err;
+ } else {
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+ PKT_FIELD(vsk, peer_waiting_read) = false;
+#endif
+ }
+ }
+ return err;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_handle_pkt
+ *
+ * Called when a notify packet is recieved for a socket in the connected state.
+ * Note this might be called from a bottom half.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void
+vsock_vmci_notify_pkt_handle_pkt(struct sock *sk,
+ vsock_packet *pkt,
+ bool bottom_half,
+ struct sockaddr_vm *dst,
+ struct sockaddr_vm *src, bool *pkt_processed)
+{
+ bool processed = false;
+
+ switch (pkt->type) {
+ case VSOCK_PACKET_TYPE_WROTE:
+ vsock_vmci_handle_wrote(sk, pkt, bottom_half, dst, src);
+ processed = true;
+ break;
+ case VSOCK_PACKET_TYPE_READ:
+ vsock_vmci_handle_read(sk, pkt, bottom_half, dst, src);
+ processed = true;
+ break;
+ case VSOCK_PACKET_TYPE_WAITING_WRITE:
+ vsock_vmci_handle_waiting_write(sk, pkt, bottom_half, dst, src);
+ processed = true;
+ break;
+
+ case VSOCK_PACKET_TYPE_WAITING_READ:
+ vsock_vmci_handle_waiting_read(sk, pkt, bottom_half, dst, src);
+ processed = true;
+ break;
+ }
+
+ if (pkt_processed)
+ *pkt_processed = processed;
+
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_process_request
+ *
+ * Called near the end of process request.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void vsock_vmci_notify_pkt_process_request(struct sock *sk)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ PKT_FIELD(vsk, write_notify_window) = vsk->consume_size;
+ if (vsk->consume_size < PKT_FIELD(vsk, write_notify_min_window))
+ PKT_FIELD(vsk, write_notify_min_window) = vsk->consume_size;
+
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_process_negotiate
+ *
+ * Called near the end of process negotiate.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void vsock_vmci_notify_pkt_process_negotiate(struct sock *sk)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ PKT_FIELD(vsk, write_notify_window) = vsk->consume_size;
+ if (vsk->consume_size < PKT_FIELD(vsk, write_notify_min_window))
+ PKT_FIELD(vsk, write_notify_min_window) = vsk->consume_size;
+
+}
+
+/* Socket control packet based operations. */
+vsock_vmci_notify_ops vsock_vmci_notify_pkt_ops = {
+ vsock_vmci_notify_pkt_socket_init,
+ vsock_vmci_notify_pkt_socket_destruct,
+ vsock_vmci_notify_pkt_poll_in,
+ vsock_vmci_notify_pkt_poll_out,
+ vsock_vmci_notify_pkt_handle_pkt,
+ vsock_vmci_notify_pkt_recv_init,
+ vsock_vmci_notify_pkt_recv_pre_block,
+ vsock_vmci_notify_pkt_recv_pre_dequeue,
+ vsock_vmci_notify_pkt_recv_post_dequeue,
+ vsock_vmci_notify_pkt_send_init,
+ vsock_vmci_notify_pkt_send_pre_block,
+ vsock_vmci_notify_pkt_send_pre_enqueue,
+ vsock_vmci_notify_pkt_send_post_enqueue,
+ vsock_vmci_notify_pkt_process_request,
+ vsock_vmci_notify_pkt_process_negotiate,
+};
diff --git a/net/vmw_vsock/notify.h b/net/vmw_vsock/notify.h
new file mode 100644
index 0000000..fe39442
--- /dev/null
+++ b/net/vmw_vsock/notify.h
@@ -0,0 +1,130 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2009-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * notify.h --
+ *
+ * Notify functions for Linux VSocket module.
+ */
+
+#ifndef __NOTIFY_H__
+#define __NOTIFY_H__
+
+#include <linux/types.h>
+
+#include "vsock_common.h"
+#include "vsock_packet.h"
+
+/* Comment this out to compare with old protocol. */
+#define VSOCK_OPTIMIZATION_WAITING_NOTIFY 1
+#if defined(VSOCK_OPTIMIZATION_WAITING_NOTIFY)
+/* Comment this out to remove flow control for "new" protocol */
+#define VSOCK_OPTIMIZATION_FLOW_CONTROL 1
+#endif
+
+#define VSOCK_MAX_DGRAM_RESENDS 10
+
+#define NOTIFYCALLRET(vsk, rv, mod_fn, args...) \
+ do { \
+ if (vsk->notify_ops && \
+ vsk->notify_ops->mod_fn != NULL) \
+ rv = (vsk->notify_ops->mod_fn)(args); \
+ else \
+ rv = 0; \
+ \
+ } while (0)
+
+#define NOTIFYCALL(vsk, mod_fn, args...) \
+ do { \
+ if (vsk->notify_ops && \
+ vsk->notify_ops->mod_fn != NULL) \
+ (vsk->notify_ops->mod_fn)(args); \
+ \
+ } while (0)
+
+typedef struct vsock_vmci_notify_pkt {
+ u64 write_notify_window;
+ u64 write_notify_min_window;
+ bool peer_waiting_read;
+ bool peer_waiting_write;
+ bool peer_waiting_write_detected;
+ bool sent_waiting_read;
+ bool sent_waiting_write;
+ vsock_waiting_info peer_waiting_read_info;
+ vsock_waiting_info peer_waiting_write_info;
+ u64 produce_q_generation;
+ u64 consume_q_generation;
+} vsock_vmci_notify_pkt;
+
+typedef struct vsock_vmci_notify_pkt_q_state {
+ u64 write_notify_window;
+ u64 write_notify_min_window;
+ bool peer_waiting_write;
+ bool peer_waiting_write_detected;
+} vsock_vmci_notify_pkt_q_state;
+
+typedef union vsock_vmci_notify {
+ vsock_vmci_notify_pkt pkt;
+ vsock_vmci_notify_pkt_q_state pkt_q_state;
+} vsock_vmci_notify;
+
+typedef struct vsock_vmci_recv_notify_data {
+ u64 consume_head;
+ u64 produce_tail;
+ bool notify_on_block;
+} vsock_vmci_recv_notify_data;
+
+typedef struct vsock_vmci_send_notify_data {
+ u64 consume_head;
+ u64 produce_tail;
+} vsock_vmci_send_notify_data;
+
+/* Socket notification callbacks. */
+typedef struct vsock_vmci_notify_ops {
+ void (*socket_init) (struct sock *sk);
+ void (*socket_destruct) (struct sock *sk);
+ int (*poll_in) (struct sock *sk, size_t target,
+ bool *data_ready_now);
+ int (*poll_out) (struct sock *sk, size_t target,
+ bool *space_avail_now);
+ void (*handle_notify_pkt) (struct sock *sk, vsock_packet *pkt,
+ bool bottom_half, struct sockaddr_vm *dst,
+ struct sockaddr_vm *src,
+ bool *pkt_processed);
+ int (*recv_init) (struct sock *sk, size_t target,
+ vsock_vmci_recv_notify_data *data);
+ int (*recv_pre_block) (struct sock *sk, size_t target,
+ vsock_vmci_recv_notify_data *data);
+ int (*recv_pre_dequeue) (struct sock *sk, size_t target,
+ vsock_vmci_recv_notify_data *data);
+ int (*recv_post_dequeue) (struct sock *sk, size_t target,
+ ssize_t copied, bool data_read,
+ vsock_vmci_recv_notify_data *data);
+ int (*send_init) (struct sock *sk,
+ vsock_vmci_send_notify_data *data);
+ int (*send_pre_block) (struct sock *sk,
+ vsock_vmci_send_notify_data *data);
+ int (*send_pre_enqueue) (struct sock *sk,
+ vsock_vmci_send_notify_data *data);
+ int (*send_post_enqueue) (struct sock *sk, ssize_t written,
+ vsock_vmci_send_notify_data *data);
+ void (*process_request) (struct sock *sk);
+ void (*process_negotiate) (struct sock *sk);
+} vsock_vmci_notify_ops;
+
+extern vsock_vmci_notify_ops vsock_vmci_notify_pkt_ops;
+extern vsock_vmci_notify_ops vsock_vmci_notify_pkt_q_state_ops;
+
+#endif /* __NOTIFY_H__ */
^ permalink raw reply related
* [PATCH 4/6] VSOCK: statistics implementation.
From: George Zhang @ 2012-11-07 18:45 UTC (permalink / raw)
To: netdev, linux-kernel, georgezhang, virtualization
Cc: pv-drivers, gregkh, davem
In-Reply-To: <20121107184340.10612.3372.stgit@promb-2n-dhcp175.eng.vmware.com>
VSOCK stats for VMCI Stream Sockets protocol.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
net/vmw_vsock/stats.c | 37 ++++++++
net/vmw_vsock/stats.h | 219 +++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 256 insertions(+), 0 deletions(-)
create mode 100644 net/vmw_vsock/stats.c
create mode 100644 net/vmw_vsock/stats.h
diff --git a/net/vmw_vsock/stats.c b/net/vmw_vsock/stats.c
new file mode 100644
index 0000000..2d172d5
--- /dev/null
+++ b/net/vmw_vsock/stats.c
@@ -0,0 +1,37 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2009-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * stats.c --
+ *
+ * Linux stats for the VMCI Stream Sockets protocol.
+ */
+
+#include <linux/types.h>
+
+#include <linux/socket.h>
+#include <linux/stddef.h> /* for NULL */
+#include <net/sock.h>
+
+#include "af_vsock.h"
+#include "stats.h"
+
+#ifdef VSOCK_GATHER_STATISTICS
+u64 vsock_stats_ctl_pkt_count[VSOCK_PACKET_TYPE_MAX];
+u64 vsock_stats_consume_queue_hist[VSOCK_NUM_QUEUE_LEVEL_BUCKETS];
+u64 vsock_stats_produce_queue_hist[VSOCK_NUM_QUEUE_LEVEL_BUCKETS];
+atomic64_t vsock_stats_consume_total;
+atomic64_t vsock_stats_produce_total;
+#endif
diff --git a/net/vmw_vsock/stats.h b/net/vmw_vsock/stats.h
new file mode 100644
index 0000000..2f9d44c
--- /dev/null
+++ b/net/vmw_vsock/stats.h
@@ -0,0 +1,219 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2009-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * stats.h --
+ *
+ * Stats functions for Linux vsock module.
+ */
+
+#ifndef __STATS_H__
+#define __STATS_H__
+
+#include <linux/types.h>
+
+#include "vsock_common.h"
+#include "vsock_packet.h"
+
+/*
+ * Define VSOCK_GATHER_STATISTICS to turn on statistics gathering. Currently
+ * this consists of 3 types of stats: 1. The number of control datagram
+ * messages sent. 2. The level of queuepair fullness (in 10% buckets) whenever
+ * data is about to be enqueued or dequeued from the queuepair. 3. The total
+ * number of bytes enqueued/dequeued.
+ */
+
+#ifdef VSOCK_GATHER_STATISTICS
+
+#define VSOCK_NUM_QUEUE_LEVEL_BUCKETS 10
+extern u64 vsock_stats_ctl_pkt_count[VSOCK_PACKET_TYPE_MAX];
+extern u64 vsock_stats_consume_queue_hist[VSOCK_NUM_QUEUE_LEVEL_BUCKETS];
+extern u64 vsock_stats_produce_queue_hist[VSOCK_NUM_QUEUE_LEVEL_BUCKETS];
+extern atomic64_t vsock_stats_consume_total;
+extern atomic64_t vsock_stats_produce_total;
+
+#define VSOCK_STATS_STREAM_CONSUME_HIST(vsk) \
+ vsock_vmci_stats_update_queue_bucket_count((vsk)->qpair, \
+ (vsk)->consume_size, \
+ vmci_qpair_consume_buf_ready((vsk)->qpair), \
+ vsock_stats_consume_queue_hist)
+#define VSOCK_STATS_STREAM_PRODUCE_HIST(vsk) \
+ vsock_vmci_stats_update_queue_bucket_count((vsk)->qpair, \
+ (vsk)->produce_size, \
+ vmci_qpair_produce_buf_ready((vsk)->qpair), \
+ vsock_stats_produce_queue_hist)
+#define VSOCK_STATS_CTLPKT_LOG(pkt_type) \
+ do { \
+ ++vsock_stats_ctl_pkt_count[pkt_type]; \
+ } while (0)
+#define VSOCK_STATS_STREAM_CONSUME(bytes) \
+ atomic64_add(&vsock_stats_consume_total, bytes)
+#define VSOCK_STATS_STREAM_PRODUCE(bytes) \
+ atomic64_add(&vsock_stats_produce_total, bytes)
+#define VSOCK_STATS_CTLPKT_DUMP_ALL() vsock_vmci_stats_ctl_pkt_dump_all()
+#define VSOCK_STATS_HIST_DUMP_ALL() vsock_vmci_stats_hist_dump_all()
+#define VSOCK_STATS_TOTALS_DUMP_ALL() vsock_vmci_stats_totals_dump_all()
+#define VSOCK_STATS_RESET() vsock_vmci_stats_reset()
+
+/*
+ *
+ * vsock_vmci_stats_update_queue_bucket_count --
+ *
+ * Given a queue, determine how much data is enqueued and add that to the
+ * specified queue level statistic bucket.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static inline void
+vsock_vmci_stats_update_queue_bucket_count(vmci_qpair *qpair,
+ u64 queue_size,
+ u64 data_ready,
+ u64 queue_hist[])
+{
+ u64 bucket = 0;
+ u32 remainder = 0;
+
+ /*
+ * We can't do 64 / 64 = 64 bit divides on linux because it requires a
+ * libgcc which is not linked into the kernel module. Since this code
+ * is only used by developers we just limit the queue_size to be less
+ * than MAX_UINT for now.
+ */
+ ASSERT(queue_size <= MAX_UINT32);
+ Div643264(data_ready * 10, queue_size, &bucket, &remainder);
+ ASSERT(bucket < VSOCK_NUM_QUEUE_LEVEL_BUCKETS);
+ ++queue_hist[bucket];
+}
+
+/*
+ *
+ * vsock_vmci_stats_ctl_pkt_dump_all --
+ *
+ * Prints all stream control packet counts out to the console using the
+ * appropriate platform logging.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static inline void vsock_vmci_stats_ctl_pkt_dump_all(void)
+{
+ int index;
+
+ ASSERT_ON_COMPILE(VSOCK_PACKET_TYPE_MAX ==
+ ARRAY_SIZE(vsock_stats_ctl_pkt_count));
+
+ for (index = 0; index < ARRAY_SIZE(vsock_stats_ctl_pkt_count);
+ index++) {
+ pr_info("Control packet: Type = %u, Count = %" FMT64
+ "u\n", index, vsock_stats_ctl_pkt_count[index]);
+ }
+}
+
+/*
+ *
+ * vsock_vmci_stats_hist_dump_all --
+ *
+ * Prints the produce and consume queue histograms to the console.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static inline void vsock_vmci_stats_hist_dump_all(void)
+{
+ int index;
+
+#define VSOCK_DUMP_HIST(strname, name) do { \
+ for (index = 0; index < ARRAY_SIZE(name); index++) { \
+ printk(strname " Bucket count %u = %"FMT64"u\n", \
+ index, name[index]); \
+ } \
+ } while (0)
+
+ VSOCK_DUMP_HIST("Produce Queue", vsock_stats_produce_queue_hist);
+ VSOCK_DUMP_HIST("Consume Queue", vsock_stats_consume_queue_hist);
+
+#undef VSOCK_DUMP_HIST
+}
+
+/*
+ *
+ * vsock_vmci_stats_totals_dump_all --
+ *
+ * Prints the produce and consume totals.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static inline void vsock_vmci_stats_totals_dump_all(void)
+{
+ pr_info("Produced %" FMT64 "u total bytes\n",
+ atomic64_read(&vsock_stats_produce_total));
+ pr_info("Consumed %" FMT64 "u total bytes\n",
+ atomic64_read(&vsock_stats_consume_total));
+}
+
+/*
+ *
+ * vsock_vmci_stats_reset --
+ *
+ * Reset all VSock statistics.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static inline void vsock_vmci_stats_reset(void)
+{
+ int index;
+
+#define VSOCK_RESET_ARRAY(name) do { \
+ for (index = 0; index < ARRAY_SIZE(name); index++) { \
+ name[index] = 0; \
+ } \
+ } while (0)
+
+ VSOCK_RESET_ARRAY(vsock_stats_ctl_pkt_count);
+ VSOCK_RESET_ARRAY(vsock_stats_produce_queue_hist);
+ VSOCK_RESET_ARRAY(vsock_stats_consume_queue_hist);
+
+#undef VSOCK_RESET_ARRAY
+
+ atomic64_set(&vsock_stats_consume_total, 0);
+ atomic64_set(&vsock_stats_produce_total, 0);
+}
+
+#else
+#define VSOCK_STATS_STREAM_CONSUME_HIST(vsk)
+#define VSOCK_STATS_STREAM_PRODUCE_HIST(vsk)
+#define VSOCK_STATS_STREAM_PRODUCE(bytes)
+#define VSOCK_STATS_STREAM_CONSUME(bytes)
+#define VSOCK_STATS_CTLPKT_LOG(pkt_type)
+#define VSOCK_STATS_CTLPKT_DUMP_ALL()
+#define VSOCK_STATS_HIST_DUMP_ALL()
+#define VSOCK_STATS_TOTALS_DUMP_ALL()
+#define VSOCK_STATS_RESET()
+#endif
+
+#endif
^ permalink raw reply related
* [PATCH 5/6] VSOCK: utility functions.
From: George Zhang @ 2012-11-07 18:46 UTC (permalink / raw)
To: netdev, linux-kernel, georgezhang, virtualization
Cc: pv-drivers, gregkh, davem
In-Reply-To: <20121107184340.10612.3372.stgit@promb-2n-dhcp175.eng.vmware.com>
VSOCK utility functions for Linux VSocket module.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
net/vmw_vsock/util.c | 626 ++++++++++++++++++++++++++++++++++++++++++++++++++
net/vmw_vsock/util.h | 312 +++++++++++++++++++++++++
2 files changed, 938 insertions(+), 0 deletions(-)
create mode 100644 net/vmw_vsock/util.c
create mode 100644 net/vmw_vsock/util.h
diff --git a/net/vmw_vsock/util.c b/net/vmw_vsock/util.c
new file mode 100644
index 0000000..1c42b23
--- /dev/null
+++ b/net/vmw_vsock/util.c
@@ -0,0 +1,626 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2007-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * util.c --
+ *
+ * Utility functions for Linux VSocket module.
+ */
+
+#include <linux/types.h>
+#include <linux/list.h>
+#include <linux/socket.h>
+#include <linux/stddef.h> /* for NULL */
+#include <net/sock.h>
+
+#include "af_vsock.h"
+#include "util.h"
+
+struct list_head vsock_bind_table[VSOCK_HASH_SIZE + 1];
+struct list_head vsock_connected_table[VSOCK_HASH_SIZE];
+
+DEFINE_SPINLOCK(vsock_table_lock);
+
+/*
+ *
+ * vsock_vmci_log_pkt --
+ *
+ * Logs the provided packet.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+void vsock_vmci_log_pkt(char const *function, u32 line, vsock_packet *pkt)
+{
+ char buf[256];
+ char *cur = buf;
+ int left = sizeof buf;
+ int written = 0;
+ char *type_strings[] = {
+ [VSOCK_PACKET_TYPE_INVALID] = "INVALID",
+ [VSOCK_PACKET_TYPE_REQUEST] = "REQUEST",
+ [VSOCK_PACKET_TYPE_NEGOTIATE] = "NEGOTIATE",
+ [VSOCK_PACKET_TYPE_OFFER] = "OFFER",
+ [VSOCK_PACKET_TYPE_ATTACH] = "ATTACH",
+ [VSOCK_PACKET_TYPE_WROTE] = "WROTE",
+ [VSOCK_PACKET_TYPE_READ] = "READ",
+ [VSOCK_PACKET_TYPE_RST] = "RST",
+ [VSOCK_PACKET_TYPE_SHUTDOWN] = "SHUTDOWN",
+ [VSOCK_PACKET_TYPE_WAITING_WRITE] = "WAITING_WRITE",
+ [VSOCK_PACKET_TYPE_WAITING_READ] = "WAITING_READ",
+ [VSOCK_PACKET_TYPE_REQUEST2] = "REQUEST2",
+ [VSOCK_PACKET_TYPE_NEGOTIATE2] = "NEGOTIATE2",
+ };
+
+ written = snprintf(cur, left, "PKT: %u:%u -> %u:%u",
+ VMCI_HANDLE_TO_CONTEXT_ID(pkt->dg.src),
+ pkt->src_port,
+ VMCI_HANDLE_TO_CONTEXT_ID(pkt->dg.dst),
+ pkt->dst_port);
+ if (written >= left)
+ goto error;
+
+ left -= written;
+ cur += written;
+
+ switch (pkt->type) {
+ case VSOCK_PACKET_TYPE_REQUEST:
+ case VSOCK_PACKET_TYPE_NEGOTIATE:
+ written = snprintf(cur, left, ", %s, size = %" FMT64 "u",
+ type_strings[pkt->type], pkt->u.size);
+ break;
+
+ case VSOCK_PACKET_TYPE_OFFER:
+ case VSOCK_PACKET_TYPE_ATTACH:
+ written = snprintf(cur, left, ", %s, handle = %u:%u",
+ type_strings[pkt->type],
+ VMCI_HANDLE_TO_CONTEXT_ID(pkt->u.handle),
+ VMCI_HANDLE_TO_RESOURCE_ID(pkt->u.handle));
+ break;
+
+ case VSOCK_PACKET_TYPE_WROTE:
+ case VSOCK_PACKET_TYPE_READ:
+ case VSOCK_PACKET_TYPE_RST:
+ written = snprintf(cur, left, ", %s", type_strings[pkt->type]);
+ break;
+ case VSOCK_PACKET_TYPE_SHUTDOWN: {
+ bool recv;
+ bool send;
+
+ recv = pkt->u.mode & RCV_SHUTDOWN;
+ send = pkt->u.mode & SEND_SHUTDOWN;
+ written = snprintf(cur, left, ", %s, mode = %c%c",
+ type_strings[pkt->type],
+ recv ? 'R' : ' ', send ? 'S' : ' ');
+ }
+ break;
+
+ case VSOCK_PACKET_TYPE_WAITING_WRITE:
+ case VSOCK_PACKET_TYPE_WAITING_READ:
+ written = snprintf(cur, left, ", %s, generation = %" FMT64 "u, "
+ "offset = %" FMT64 "u",
+ type_strings[pkt->type],
+ pkt->u.wait.generation, pkt->u.wait.offset);
+
+ break;
+
+ case VSOCK_PACKET_TYPE_REQUEST2:
+ case VSOCK_PACKET_TYPE_NEGOTIATE2:
+ written = snprintf(cur, left, ", %s, size = %" FMT64 "u, "
+ "proto = %u",
+ type_strings[pkt->type], pkt->u.size,
+ pkt->proto);
+ break;
+
+ default:
+ written = snprintf(cur, left, ", unrecognized type");
+ }
+
+ if (written >= left)
+ goto error;
+
+ left -= written;
+ cur += written;
+
+ written = snprintf(cur, left, " [%s:%u]\n", function, line);
+ if (written >= left)
+ goto error;
+
+ return;
+
+error:
+ pr_err("could not log packet\n");
+}
+
+/*
+ *
+ * vsock_vmci_init_tables --
+ *
+ * Initializes the tables used for socket lookup.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+void vsock_vmci_init_tables(void)
+{
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(vsock_bind_table); i++)
+ INIT_LIST_HEAD(&vsock_bind_table[i]);
+
+ for (i = 0; i < ARRAY_SIZE(vsock_connected_table); i++)
+ INIT_LIST_HEAD(&vsock_connected_table[i]);
+}
+
+/*
+ *
+ * __vsock_vmci_insert_bound --
+ *
+ * Inserts socket into the bound table.
+ *
+ * Note that this assumes any necessary locks are held.
+ *
+ * Results: None.
+ *
+ * Side effects: The reference count for sk is incremented.
+ */
+
+void __vsock_vmci_insert_bound(struct list_head *list, struct sock *sk)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ sock_hold(sk);
+ list_add(&vsk->bound_table, list);
+}
+
+/*
+ *
+ * __vsock_vmci_insert_connected --
+ *
+ * Inserts socket into the connected table.
+ *
+ * Note that this assumes any necessary locks are held.
+ *
+ * Results: None.
+ *
+ * Side effects: The reference count for sk is incremented.
+ */
+
+void __vsock_vmci_insert_connected(struct list_head *list, struct sock *sk)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ sock_hold(sk);
+ list_add(&vsk->connected_table, list);
+}
+
+/*
+ *
+ * __vsock_vmci_remove_bound --
+ *
+ * Removes socket from the bound table.
+ *
+ * Note that this assumes any necessary locks are held.
+ *
+ * Results: None.
+ *
+ * Side effects: The reference count for sk is decremented.
+ */
+
+void __vsock_vmci_remove_bound(struct sock *sk)
+{
+ vsock_vmci_sock *vsk;
+
+ ASSERT(__vsock_vmci_in_bound_table(sk));
+
+ vsk = vsock_sk(sk);
+
+ list_del_init(&vsk->bound_table);
+ sock_put(sk);
+}
+
+/*
+ *
+ * __vsock_vmci_remove_connected --
+ *
+ * Removes socket from the connected table.
+ *
+ * Note that this assumes any necessary locks are held.
+ *
+ * Results: None.
+ *
+ * Side effects: The reference count for sk is decremented.
+ */
+
+void __vsock_vmci_remove_connected(struct sock *sk)
+{
+ vsock_vmci_sock *vsk;
+
+ ASSERT(__vsock_vmci_in_connected_table(sk));
+
+ vsk = vsock_sk(sk);
+
+ list_del_init(&vsk->connected_table);
+ sock_put(sk);
+}
+
+/*
+ *
+ * __vsock_vmci_find_bound_socket --
+ *
+ * Finds the socket corresponding to the provided address in the bound sockets
+ * hash table.
+ *
+ * Note that this assumes any necessary locks are held.
+ *
+ * Results: The sock structure if found, NULL if not found.
+ *
+ * Side effects: None.
+ */
+
+struct sock *__vsock_vmci_find_bound_socket(struct sockaddr_vm *addr)
+{
+ vsock_vmci_sock *vsk;
+
+ list_for_each_entry(vsk, vsock_bound_sockets(addr), bound_table) {
+ if (vsock_addr_equals_addr_any(addr, &vsk->local_addr)) {
+ return sk_vsock(vsk);
+ }
+ }
+
+ return NULL;
+}
+
+/*
+ *
+ * __vsock_vmci_find_connected_socket --
+ *
+ * Finds the socket corresponding to the provided addresses in the connected
+ * sockets hash table.
+ *
+ * Note that this assumes any necessary locks are held.
+ *
+ * Results: The sock structure if found, NULL if not found.
+ *
+ * Side effects: None.
+ */
+
+struct sock *__vsock_vmci_find_connected_socket(struct sockaddr_vm *src,
+ struct sockaddr_vm *dst)
+{
+ vsock_vmci_sock *vsk;
+
+ list_for_each_entry(vsk, vsock_connected_sockets(src, dst),
+ connected_table) {
+ if (vsock_addr_equals_addr(src, &vsk->remote_addr)
+ && vsock_addr_equals_addr(dst, &vsk->local_addr)) {
+ return sk_vsock(vsk);
+ }
+ }
+
+ return NULL;
+}
+
+/*
+ *
+ * __vsock_vmci_in_bound_table --
+ *
+ * Determines whether the provided socket is in the bound table.
+ *
+ * Results: TRUE is socket is in bound table, FALSE otherwise.
+ *
+ * Side effects: None.
+ */
+
+bool __vsock_vmci_in_bound_table(struct sock *sk)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ return !list_empty(&vsk->bound_table);
+}
+
+/*
+ *
+ * __vsock_vmci_in_connected_table --
+ *
+ * Determines whether the provided socket is in the connected table.
+ *
+ * Results: TRUE is socket is in connected table, FALSE otherwise.
+ *
+ * Side effects: None.
+ */
+
+bool __vsock_vmci_in_connected_table(struct sock *sk)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ return !list_empty(&vsk->connected_table);
+}
+
+/*
+ *
+ * vsock_vmci_get_pending --
+ *
+ * Retrieves a pending connection that matches the addresses specified in the
+ * provided packet.
+ *
+ * Assumes the socket lock is held for listener.
+ *
+ * Results: Socket of the pending connection on success, NULL if not found.
+ *
+ * Side effects: A reference is held on the socket until the release function
+ * is called.
+ */
+
+struct sock *vsock_vmci_get_pending(struct sock *listener, vsock_packet *pkt)
+{
+ vsock_vmci_sock *vlistener;
+ vsock_vmci_sock *vpending;
+ struct sock *pending;
+
+ vlistener = vsock_sk(listener);
+
+ list_for_each_entry(vpending, &vlistener->pending_links,
+ pending_links) {
+ struct sockaddr_vm src;
+ struct sockaddr_vm dst;
+
+ vsock_addr_init(&src, VMCI_HANDLE_TO_CONTEXT_ID(pkt->dg.src),
+ pkt->src_port);
+ vsock_addr_init(&dst, VMCI_HANDLE_TO_CONTEXT_ID(pkt->dg.dst),
+ pkt->dst_port);
+
+ if (vsock_addr_equals_addr(&src, &vpending->remote_addr) &&
+ vsock_addr_equals_addr(&dst, &vpending->local_addr)) {
+ pending = sk_vsock(vpending);
+ sock_hold(pending);
+ goto found;
+ }
+ }
+
+ pending = NULL;
+found:
+ return pending;
+
+}
+
+/*
+ *
+ * vsock_vmci_release_pending --
+ *
+ * Releases the reference on a socket previously obtained by a call to
+ * vsock_vmci_get_pending().
+ *
+ * Results: None.
+ *
+ * Side effects: The socket may be freed if this was the last reference.
+ */
+
+void vsock_vmci_release_pending(struct sock *pending)
+{
+ sock_put(pending);
+}
+
+/*
+ *
+ * vsock_vmci_add_pending --
+ *
+ * Adds a pending connection on listener's pending list.
+ *
+ * Assumes the socket lock is held for listener. Assumes the socket lock is
+ * held for pending.
+ *
+ * Results: None.
+ *
+ * Side effects: The reference count of the sockets is incremented.
+ */
+
+void vsock_vmci_add_pending(struct sock *listener, struct sock *pending)
+{
+ vsock_vmci_sock *vlistener;
+ vsock_vmci_sock *vpending;
+
+ vlistener = vsock_sk(listener);
+ vpending = vsock_sk(pending);
+
+ sock_hold(pending);
+ sock_hold(listener);
+ list_add_tail(&vpending->pending_links, &vlistener->pending_links);
+}
+
+/*
+ *
+ * vsock_vmci_remove_pending --
+ *
+ * Removes a pending connection from the listener's pending list.
+ *
+ * Assumes the socket lock is held for listener. Assumes the socket lock is
+ * held for pending.
+ *
+ * Results: None.
+ *
+ * Side effects: The reference count of the sockets is decremented.
+ */
+
+void vsock_vmci_remove_pending(struct sock *listener, struct sock *pending)
+{
+ vsock_vmci_sock *vpending = vsock_sk(pending);
+
+ list_del_init(&vpending->pending_links);
+ sock_put(listener);
+ sock_put(pending);
+}
+
+/*
+ *
+ * vsock_vmci_enqueue_accept --
+ *
+ * Enqueues the connected socket on the listening socket's accepting queue.
+ *
+ * Assumes the socket lock is held for listener. Assumes the socket lock is
+ * held for connected.
+ *
+ * Results: None.
+ *
+ * Side effects: The sockets' reference counts are incremented.
+ */
+
+void vsock_vmci_enqueue_accept(struct sock *listener, struct sock *connected)
+{
+ vsock_vmci_sock *vlistener;
+ vsock_vmci_sock *vconnected;
+
+ vlistener = vsock_sk(listener);
+ vconnected = vsock_sk(connected);
+
+ sock_hold(connected);
+ sock_hold(listener);
+ list_add_tail(&vconnected->accept_queue, &vlistener->accept_queue);
+}
+
+/*
+ *
+ * vsock_vmci_dequeue_accept --
+ *
+ * Dequeues the next connected socket from the listening socket's accept queue.
+ *
+ * Assumes the socket lock is held for listener.
+ *
+ * Note that the caller must call sock_put() on the returned socket once it is
+ * done with the socket.
+ *
+ * Results: The next socket from the queue, or NULL if the queue is empty.
+ *
+ * Side effects: The reference count of the listener is decremented.
+ */
+
+struct sock *vsock_vmci_dequeue_accept(struct sock *listener)
+{
+ vsock_vmci_sock *vlistener;
+ vsock_vmci_sock *vconnected;
+
+ vlistener = vsock_sk(listener);
+
+ if (list_empty(&vlistener->accept_queue))
+ return NULL;
+
+ vconnected = list_entry(vlistener->accept_queue.next,
+ vsock_vmci_sock, accept_queue);
+ ASSERT(vconnected);
+
+ list_del_init(&vconnected->accept_queue);
+ sock_put(listener);
+ /*
+ * The caller will need a reference on the connected socket so we let
+ * it call sock_put().
+ */
+
+ ASSERT(sk_vsock(vconnected));
+ return sk_vsock(vconnected);
+}
+
+/*
+ *
+ * vsock_vmci_remove_accept --
+ *
+ * Removes a socket from the accept queue of a listening socket.
+ *
+ * Assumes the socket lock is held for listener. Assumes the socket lock is
+ * held for connected.
+ *
+ * Results: None.
+ *
+ * Side effects: The sockets' reference counts are decremented.
+ */
+
+void vsock_vmci_remove_accept(struct sock *listener, struct sock *connected)
+{
+ vsock_vmci_sock *vconnected;
+
+ if (!vsock_vmci_in_accept_queue(connected))
+ return;
+
+ vconnected = vsock_sk(connected);
+ ASSERT(vconnected->listener == listener);
+
+ list_del_init(&vconnected->accept_queue);
+ sock_put(listener);
+ sock_put(connected);
+}
+
+/*
+ *
+ * vsock_vmci_in_accept_queue --
+ *
+ * Determines whether a socket is on an accept queue.
+ *
+ * Assumes the socket lock is held for sk.
+ *
+ * Results: TRUE if the socket is in an accept queue, FALSE otherwise.
+ *
+ * Side effects: None.
+ */
+
+bool vsock_vmci_in_accept_queue(struct sock *sk)
+{
+ /*
+ * If our accept queue isn't empty, it means we're linked into some
+ * listener socket's accept queue.
+ */
+ return !vsock_vmci_is_accept_queue_empty(sk);
+}
+
+/*
+ *
+ * vsock_vmci_is_accept_queue_empty --
+ *
+ * Determines whether the provided socket's accept queue is empty.
+ *
+ * Assumes the socket lock is held for sk.
+ *
+ * Results: TRUE if the socket's accept queue is empty, FALSE otherwsise.
+ *
+ * Side effects: None.
+ *
+ */
+
+bool vsock_vmci_is_accept_queue_empty(struct sock *sk)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+ return list_empty(&vsk->accept_queue);
+}
+
+/*
+ *
+ * vsock_vmci_is_pending --
+ *
+ * Determines whether a socket is pending.
+ *
+ * Assumes the socket lock is held for sk.
+ *
+ * Results: TRUE if the socket is pending, FALSE otherwise.
+ *
+ * Side effects: None.
+ */
+
+bool vsock_vmci_is_pending(struct sock *sk)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+ return !list_empty(&vsk->pending_links);
+}
diff --git a/net/vmw_vsock/util.h b/net/vmw_vsock/util.h
new file mode 100644
index 0000000..50ed6c8
--- /dev/null
+++ b/net/vmw_vsock/util.h
@@ -0,0 +1,312 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2007-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * util.h --
+ *
+ * Utility functions for Linux VSocket module.
+ */
+
+#ifndef __UTIL_H__
+#define __UTIL_H__
+
+#include <linux/types.h>
+#include <linux/stddef.h> /* for NULL */
+#include <net/sock.h>
+#include <linux/spinlock.h>
+
+#include "vsock_common.h"
+#include "vsock_packet.h"
+
+/*
+ * Each bound VSocket is stored in the bind hash table and each connected
+ * VSocket is stored in the connected hash table.
+ *
+ * Unbound sockets are all put on the same list attached to the end of the hash
+ * table (vsock_unbound_sockets). Bound sockets are added to the hash table in
+ * the bucket that their local address hashes to (vsock_bound_sockets(addr)
+ * represents the list that addr hashes to).
+ *
+ * Specifically, we initialize the vsock_bind_table array to a size of
+ * VSOCK_HASH_SIZE + 1 so that vsock_bind_table[0] through
+ * vsock_bind_table[VSOCK_HASH_SIZE - 1] are for bound sockets and
+ * vsock_bind_table[VSOCK_HASH_SIZE] is for unbound sockets. The hash function
+ * mods with VSOCK_HASH_SIZE - 1 to ensure this.
+ */
+#define VSOCK_HASH_SIZE 251
+#define LAST_RESERVED_PORT 1023
+#define MAX_PORT_RETRIES 24
+
+extern struct list_head vsock_bind_table[VSOCK_HASH_SIZE + 1];
+extern struct list_head vsock_connected_table[VSOCK_HASH_SIZE];
+
+extern spinlock_t vsock_table_lock;
+
+#define VSOCK_HASH(addr) ((addr)->svm_port % (VSOCK_HASH_SIZE - 1))
+#define vsock_bound_sockets(addr) (&vsock_bind_table[VSOCK_HASH(addr)])
+#define vsock_unbound_sockets (&vsock_bind_table[VSOCK_HASH_SIZE])
+
+/* XXX This can probably be implemented in a better way. */
+#define VSOCK_CONN_HASH(src, dst) \
+ (((src)->svm_cid ^ (dst)->svm_port) % (VSOCK_HASH_SIZE - 1))
+#define vsock_connected_sockets(src, dst) \
+ (&vsock_connected_table[VSOCK_CONN_HASH(src, dst)])
+#define vsock_connected_sockets_vsk(vsk) \
+ vsock_connected_sockets(&(vsk)->remote_addr, &(vsk)->local_addr)
+
+/*
+ * Prototypes.
+ */
+
+void vsock_vmci_log_pkt(char const *function, u32 line, vsock_packet *pkt);
+
+void vsock_vmci_init_tables(void);
+void __vsock_vmci_insert_bound(struct list_head *list, struct sock *sk);
+void __vsock_vmci_insert_connected(struct list_head *list, struct sock *sk);
+void __vsock_vmci_remove_bound(struct sock *sk);
+void __vsock_vmci_remove_connected(struct sock *sk);
+struct sock *__vsock_vmci_find_bound_socket(struct sockaddr_vm *addr);
+struct sock *__vsock_vmci_find_connected_socket(struct sockaddr_vm *src,
+ struct sockaddr_vm *dst);
+bool __vsock_vmci_in_bound_table(struct sock *sk);
+bool __vsock_vmci_in_connected_table(struct sock *sk);
+
+struct sock *vsock_vmci_get_pending(struct sock *listener, vsock_packet *pkt);
+void vsock_vmci_release_pending(struct sock *pending);
+void vsock_vmci_add_pending(struct sock *listener, struct sock *pending);
+void vsock_vmci_remove_pending(struct sock *listener, struct sock *pending);
+void vsock_vmci_enqueue_accept(struct sock *listener, struct sock *connected);
+struct sock *vsock_vmci_dequeue_accept(struct sock *listener);
+void vsock_vmci_remove_accept(struct sock *listener, struct sock *connected);
+bool vsock_vmci_in_accept_queue(struct sock *sk);
+bool vsock_vmci_is_accept_queue_empty(struct sock *sk);
+bool vsock_vmci_is_pending(struct sock *sk);
+
+static inline void vsock_vmci_insert_bound(struct list_head *list,
+ struct sock *sk);
+static inline void vsock_vmci_insert_connected(struct list_head *list,
+ struct sock *sk);
+static inline void vsock_vmci_remove_bound(struct sock *sk);
+static inline void vsock_vmci_remove_connected(struct sock *sk);
+static inline struct sock *vsock_vmci_find_bound_socket(struct sockaddr_vm
+ *addr);
+static inline struct sock *vsock_vmci_find_connected_socket(struct sockaddr_vm
+ *src,
+ struct sockaddr_vm
+ *dst);
+static inline bool vsock_vmci_in_bound_table(struct sock *sk);
+static inline bool vsock_vmci_in_connected_table(struct sock *sk);
+
+/*
+ *
+ * vsock_vmci_insert_bound --
+ *
+ * Inserts socket into the bound table.
+ *
+ * Note that it is important to invoke the bottom-half versions of the spinlock
+ * functions since these may be called from tasklets.
+ *
+ * Results: None.
+ *
+ * Side effects: vsock_table_lock is acquired and released.
+ */
+
+static inline void
+vsock_vmci_insert_bound(struct list_head *list, struct sock *sk)
+{
+ spin_lock_bh(&vsock_table_lock);
+ __vsock_vmci_insert_bound(list, sk);
+ spin_unlock_bh(&vsock_table_lock);
+}
+
+/*
+ *
+ * vsock_vmci_insert_connected --
+ *
+ * Inserts socket into the connected table.
+ *
+ * Note that it is important to invoke the bottom-half versions of the spinlock
+ * functions since these may be called from tasklets.
+ *
+ * Results: None.
+ *
+ * Side effects: vsock_table_lock is acquired and released.
+ */
+
+static inline void
+vsock_vmci_insert_connected(struct list_head *list, struct sock *sk)
+{
+ spin_lock_bh(&vsock_table_lock);
+ __vsock_vmci_insert_connected(list, sk);
+ spin_unlock_bh(&vsock_table_lock);
+}
+
+/*
+ *
+ * vsock_vmci_remove_bound --
+ *
+ * Removes socket from the bound list.
+ *
+ * Note that it is important to invoke the bottom-half versions of the spinlock
+ * functions since these may be called from tasklets.
+ *
+ * Results: None.
+ *
+ * Side effects: vsock_table_lock is acquired and released.
+ */
+
+static inline void vsock_vmci_remove_bound(struct sock *sk)
+{
+ spin_lock_bh(&vsock_table_lock);
+ __vsock_vmci_remove_bound(sk);
+ spin_unlock_bh(&vsock_table_lock);
+}
+
+/*
+ *
+ * vsock_vmci_remove_connected --
+ *
+ * Removes socket from the connected list.
+ *
+ * Note that it is important to invoke the bottom-half versions of the spinlock
+ * functions since these may be called from tasklets.
+ *
+ * Results: None.
+ *
+ * Side effects: vsock_table_lock is acquired and released.
+ */
+
+static inline void vsock_vmci_remove_connected(struct sock *sk)
+{
+ spin_lock_bh(&vsock_table_lock);
+ __vsock_vmci_remove_connected(sk);
+ spin_unlock_bh(&vsock_table_lock);
+}
+
+/*
+ *
+ * vsock_vmci_find_bound_socket --
+ *
+ * Finds the socket corresponding to the provided address in the bound sockets
+ * hash table.
+ *
+ * Note that it is important to invoke the bottom-half versions of the spinlock
+ * functions since these are called from tasklets.
+ *
+ * Results: The sock structure if found, NULL on failure.
+ *
+ * Side effects: vsock_table_lock is acquired and released. The socket's
+ * reference count is increased.
+ */
+
+static inline struct sock *vsock_vmci_find_bound_socket(struct sockaddr_vm
+ *addr)
+{
+ struct sock *sk;
+
+ spin_lock_bh(&vsock_table_lock);
+ sk = __vsock_vmci_find_bound_socket(addr);
+ if (sk)
+ sock_hold(sk);
+
+ spin_unlock_bh(&vsock_table_lock);
+
+ return sk;
+}
+
+/*
+ *
+ * vsock_vmci_find_connected_socket --
+ *
+ * Finds the socket corresponding to the provided address in the connected
+ * sockets hash table.
+ *
+ * Note that it is important to invoke the bottom-half versions of the spinlock
+ * functions since these are called from tasklets.
+ *
+ * Results: The sock structure if found, NULL on failure.
+ *
+ * Side effects: vsock_table_lock is acquired and released. The socket's
+ * reference count is increased.
+ */
+
+static inline struct sock *vsock_vmci_find_connected_socket(struct sockaddr_vm
+ *src,
+ struct sockaddr_vm
+ *dst)
+{
+ struct sock *sk;
+
+ spin_lock_bh(&vsock_table_lock);
+ sk = __vsock_vmci_find_connected_socket(src, dst);
+ if (sk)
+ sock_hold(sk);
+
+ spin_unlock_bh(&vsock_table_lock);
+
+ return sk;
+}
+
+/*
+ *
+ * vsock_vmci_in_bound_table --
+ *
+ * Determines whether the provided socket is in the bound table.
+ *
+ * Note that it is important to invoke the bottom-half versions of the spinlock
+ * functions since these may be called from tasklets.
+ *
+ * Results: TRUE is socket is in bound table, FALSE otherwise.
+ *
+ * Side effects: vsock_table_lock is acquired and released.
+ */
+
+static inline bool vsock_vmci_in_bound_table(struct sock *sk)
+{
+ bool ret;
+
+ spin_lock_bh(&vsock_table_lock);
+ ret = __vsock_vmci_in_bound_table(sk);
+ spin_unlock_bh(&vsock_table_lock);
+
+ return ret;
+}
+
+/*
+ *
+ * vsock_vmci_in_connected_table --
+ *
+ * Determines whether the provided socket is in the connected table.
+ *
+ * Note that it is important to invoke the bottom-half versions of the spinlock
+ * functions since these may be called from tasklets.
+ *
+ * Results: TRUE is socket is in connected table, FALSE otherwise.
+ *
+ * Side effects: vsock_table_lock is acquired and released.
+ */
+
+static inline bool vsock_vmci_in_connected_table(struct sock *sk)
+{
+ bool ret;
+
+ spin_lock_bh(&vsock_table_lock);
+ ret = __vsock_vmci_in_connected_table(sk);
+ spin_unlock_bh(&vsock_table_lock);
+
+ return ret;
+}
+
+#endif /* __UTIL_H__ */
^ permalink raw reply related
* [PATCH 6/6] VSOCK: header and config files.
From: George Zhang @ 2012-11-07 18:46 UTC (permalink / raw)
To: netdev, linux-kernel, georgezhang, virtualization
Cc: pv-drivers, gregkh, davem
In-Reply-To: <20121107184340.10612.3372.stgit@promb-2n-dhcp175.eng.vmware.com>
VSOCK header files, Makefiles and Kconfig systems for Linux VSocket module.
Signed-off-by: George Zhang <georgezhang@vmware.com>
---
include/linux/socket.h | 4
net/Kconfig | 1
net/Makefile | 1
net/vmw_vsock/Kconfig | 14 +
net/vmw_vsock/Makefile | 4
net/vmw_vsock/notify_qstate.c | 627 +++++++++++++++++++++++++++++++++++
net/vmw_vsock/vmci_sockets.h | 517 +++++++++++++++++++++++++++++
net/vmw_vsock/vmci_sockets_packet.h | 106 ++++++
net/vmw_vsock/vsock_common.h | 130 +++++++
net/vmw_vsock/vsock_packet.h | 131 +++++++
net/vmw_vsock/vsock_version.h | 29 ++
11 files changed, 1563 insertions(+), 1 deletions(-)
create mode 100644 net/vmw_vsock/Kconfig
create mode 100644 net/vmw_vsock/Makefile
create mode 100644 net/vmw_vsock/notify_qstate.c
create mode 100644 net/vmw_vsock/vmci_sockets.h
create mode 100644 net/vmw_vsock/vmci_sockets_packet.h
create mode 100644 net/vmw_vsock/vsock_common.h
create mode 100644 net/vmw_vsock/vsock_packet.h
create mode 100644 net/vmw_vsock/vsock_version.h
diff --git a/include/linux/socket.h b/include/linux/socket.h
index 25d6322..57bc85e 100644
--- a/include/linux/socket.h
+++ b/include/linux/socket.h
@@ -195,7 +195,8 @@ struct ucred {
#define AF_CAIF 37 /* CAIF sockets */
#define AF_ALG 38 /* Algorithm sockets */
#define AF_NFC 39 /* NFC sockets */
-#define AF_MAX 40 /* For now.. */
+#define AF_VSOCK 40 /* VMCI sockets */
+#define AF_MAX 41 /* For now.. */
/* Protocol families, same as address families. */
#define PF_UNSPEC AF_UNSPEC
@@ -238,6 +239,7 @@ struct ucred {
#define PF_CAIF AF_CAIF
#define PF_ALG AF_ALG
#define PF_NFC AF_NFC
+#define PF_VSOCK AF_VSOCK
#define PF_MAX AF_MAX
/* Maximum queue length specifiable by listen. */
diff --git a/net/Kconfig b/net/Kconfig
index 245831b..75b8d5e 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -216,6 +216,7 @@ source "net/dcb/Kconfig"
source "net/dns_resolver/Kconfig"
source "net/batman-adv/Kconfig"
source "net/openvswitch/Kconfig"
+source "net/vmw_vsock/Kconfig"
config RPS
boolean
diff --git a/net/Makefile b/net/Makefile
index 4f4ee08..cae59f4 100644
--- a/net/Makefile
+++ b/net/Makefile
@@ -70,3 +70,4 @@ obj-$(CONFIG_CEPH_LIB) += ceph/
obj-$(CONFIG_BATMAN_ADV) += batman-adv/
obj-$(CONFIG_NFC) += nfc/
obj-$(CONFIG_OPENVSWITCH) += openvswitch/
+obj-$(CONFIG_VMWARE_VSOCK) += vmw_vsock/
diff --git a/net/vmw_vsock/Kconfig b/net/vmw_vsock/Kconfig
new file mode 100644
index 0000000..95e2568
--- /dev/null
+++ b/net/vmw_vsock/Kconfig
@@ -0,0 +1,14 @@
+#
+# Vsock protocol
+#
+
+config VMWARE_VSOCK
+ tristate "Virtual Socket protocol"
+ depends on VMWARE_VMCI
+ help
+ Virtual Socket Protocol is a socket protocol similar to TCP/IP
+ allowing comunication between Virtual Machines and VMware
+ hypervisor.
+
+ To compile this driver as a module, choose M here: the module
+ will be called vsock. If unsure, say N.
diff --git a/net/vmw_vsock/Makefile b/net/vmw_vsock/Makefile
new file mode 100644
index 0000000..4e940fe
--- /dev/null
+++ b/net/vmw_vsock/Makefile
@@ -0,0 +1,4 @@
+obj-$(CONFIG_VMWARE_VSOCK) += vmw_vsock.o
+
+vmw_vsock-y += af_vsock.o notify.o notify_qstate.o stats.o util.o \
+ vsock_addr.o
diff --git a/net/vmw_vsock/notify_qstate.c b/net/vmw_vsock/notify_qstate.c
new file mode 100644
index 0000000..9cc4b58
--- /dev/null
+++ b/net/vmw_vsock/notify_qstate.c
@@ -0,0 +1,627 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2009-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * notifyQState.c --
+ *
+ * Linux control notifications based on Queuepair state for the VMCI Stream
+ * Sockets protocol.
+ */
+
+#include <linux/types.h>
+
+#include <linux/socket.h>
+
+#include <linux/stddef.h> /* for NULL */
+#include <net/sock.h>
+
+#include "notify.h"
+#include "af_vsock.h"
+
+#define PKT_FIELD(vsk, field_name) \
+ (vsk)->notify.pkt_q_state.field_name
+
+/*
+ *
+ * vsock_vmci_notify_waiting_write --
+ *
+ * Determines if the conditions have been met to notify a waiting writer.
+ *
+ * Results: true if a notification should be sent, false otherwise.
+ *
+ * Side effects: None.
+ */
+
+static bool vsock_vmci_notify_waiting_write(vsock_vmci_sock *vsk)
+{
+ bool retval;
+ u64 notify_limit;
+
+ if (!PKT_FIELD(vsk, peer_waiting_write))
+ return false;
+
+ /*
+ * When the sender blocks, we take that as a sign that the sender is
+ * faster than the receiver. To reduce the transmit rate of the sender,
+ * we delay the sending of the read notification by decreasing the
+ * write_notify_window. The notification is delayed until the number of
+ * bytes used in the queue drops below the write_notify_window.
+ */
+
+ if (!PKT_FIELD(vsk, peer_waiting_write_detected)) {
+ PKT_FIELD(vsk, peer_waiting_write_detected) = true;
+ if (PKT_FIELD(vsk, write_notify_window) < PAGE_SIZE) {
+ PKT_FIELD(vsk, write_notify_window) =
+ PKT_FIELD(vsk, write_notify_min_window);
+ } else {
+ PKT_FIELD(vsk, write_notify_window) -= PAGE_SIZE;
+ if (PKT_FIELD(vsk, write_notify_window) <
+ PKT_FIELD(vsk, write_notify_min_window))
+ PKT_FIELD(vsk, write_notify_window) =
+ PKT_FIELD(vsk, write_notify_min_window);
+
+ }
+ }
+ notify_limit = vsk->consume_size - PKT_FIELD(vsk, write_notify_window);
+
+ /*
+ * The notify_limit is used to delay notifications in the case where
+ * flow control is enabled. Below the test is expressed in terms of
+ * free space in the queue: if free_space > ConsumeSize -
+ * write_notify_window then notify An alternate way of expressing this
+ * is to rewrite the expression to use the data ready in the receive
+ * queue: if write_notify_window > bufferReady then notify as
+ * free_space == ConsumeSize - bufferReady.
+ */
+
+ retval = vmci_qpair_consume_free_space(vsk->qpair) > notify_limit;
+
+ if (retval) {
+ /*
+ * Once we notify the peer, we reset the detected flag so the
+ * next wait will again cause a decrease in the window size.
+ */
+
+ PKT_FIELD(vsk, peer_waiting_write_detected) = false;
+ }
+ return retval;
+}
+
+/*
+ *
+ * vsock_vmci_handle_read --
+ *
+ * Handles an incoming read message.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void
+vsock_vmci_handle_read(struct sock *sk,
+ vsock_packet *pkt,
+ bool bottom_half,
+ struct sockaddr_vm *dst, struct sockaddr_vm *src)
+{
+
+ sk->sk_write_space(sk);
+}
+
+/*
+ *
+ * vsock_vmci_handle_wrote --
+ *
+ * Handles an incoming wrote message.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void
+vsock_vmci_handle_wrote(struct sock *sk,
+ vsock_packet *pkt,
+ bool bottom_half,
+ struct sockaddr_vm *dst, struct sockaddr_vm *src)
+{
+ sk->sk_data_ready(sk, 0);
+}
+
+/*
+ *
+ * vsock_vmci_block_update_write_window --
+ *
+ * Updates the write window when we are blocking for data.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void vsock_vmci_block_update_write_window(struct sock *sk)
+{
+ vsock_vmci_sock *vsk;
+
+ vsk = vsock_sk(sk);
+
+ if (PKT_FIELD(vsk, write_notify_window) < vsk->consume_size)
+ PKT_FIELD(vsk, write_notify_window) =
+ min(PKT_FIELD(vsk, write_notify_window) + PAGE_SIZE,
+ vsk->consume_size);
+
+}
+
+/*
+ *
+ * vsock_vmci_send_read_notification --
+ *
+ * Sends a read notification to this socket's peer.
+ *
+ * Results: >= 0 if the datagram is sent successfully, negative error value
+ * otherwise.
+ *
+ * Side effects: None.
+ */
+
+static int vsock_vmci_send_read_notification(struct sock *sk)
+{
+ vsock_vmci_sock *vsk;
+ bool sent_read;
+ unsigned int retries;
+ int err;
+
+ vsk = vsock_sk(sk);
+ sent_read = false;
+ retries = 0;
+ err = 0;
+
+ if (vsock_vmci_notify_waiting_write(vsk)) {
+ /*
+ * Notify the peer that we have read, retrying the send on
+ * failure up to our maximum value. XXX For now we just log
+ * the failure, but later we should schedule a work item to
+ * handle the resend until it succeeds. That would require
+ * keeping track of work items in the vsk and cleaning them up
+ * upon socket close.
+ */
+ while (!(vsk->peer_shutdown & RCV_SHUTDOWN) &&
+ !sent_read && retries < VSOCK_MAX_DGRAM_RESENDS) {
+ err = VSOCK_SEND_READ(sk);
+ if (err >= 0)
+ sent_read = true;
+
+ retries++;
+ }
+
+ if (retries >= VSOCK_MAX_DGRAM_RESENDS && !sent_read)
+ printk
+ ("%p unable to send read notification to peer.\n",
+ sk);
+ else
+ PKT_FIELD(vsk, peer_waiting_write) = false;
+
+ }
+ return err;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_socket_init --
+ *
+ * Function that is called after a socket is created and before any notify ops
+ * are used.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void vsock_vmci_notify_pkt_socket_init(struct sock *sk)
+{
+ vsock_vmci_sock *vsk;
+ vsk = vsock_sk(sk);
+
+ PKT_FIELD(vsk, write_notify_window) = PAGE_SIZE;
+ PKT_FIELD(vsk, write_notify_min_window) = PAGE_SIZE;
+ PKT_FIELD(vsk, peer_waiting_write) = false;
+ PKT_FIELD(vsk, peer_waiting_write_detected) = false;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_socket_destruct --
+ *
+ * Function that is called when the socket is being released.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void vsock_vmci_notify_pkt_socket_destruct(struct sock *sk)
+{
+ vsock_vmci_sock *vsk;
+ vsk = vsock_sk(sk);
+
+ PKT_FIELD(vsk, write_notify_window) = PAGE_SIZE;
+ PKT_FIELD(vsk, write_notify_min_window) = PAGE_SIZE;
+ PKT_FIELD(vsk, peer_waiting_write) = false;
+ PKT_FIELD(vsk, peer_waiting_write_detected) = false;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_poll_in --
+ *
+ * Called by the poll function to figure out if there is data to read and to
+ * setup future notifications if needed. Only called on sockets that aren't
+ * shutdown for recv.
+ *
+ * Results: 0 on success. Negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_poll_in(struct sock *sk,
+ size_t target, bool *data_ready_now)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ if (vsock_vmci_stream_has_data(vsk)) {
+ *data_ready_now = true;
+ } else {
+ /*
+ * We can't read right now because there is nothing in the
+ * queue. Ask for notifications when there is something to
+ * read.
+ */
+ if (sk->sk_state == SS_CONNECTED)
+ vsock_vmci_block_update_write_window(sk);
+
+ *data_ready_now = false;
+ }
+
+ return 0;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_poll_out
+ *
+ * Called by the poll function to figure out if there is space to write and to
+ * setup future notifications if needed. Only called on a connected socket that
+ * isn't shutdown for send.
+ *
+ * Results: 0 on success. Negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_poll_out(struct sock *sk,
+ size_t target, bool *space_avail_now)
+{
+ s64 produce_q_free_space;
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ produce_q_free_space = vsock_vmci_stream_has_space(vsk);
+ if (produce_q_free_space > 0) {
+ *space_avail_now = true;
+ return 0;
+ } else if (produce_q_free_space == 0) {
+ /*
+ * This is a connected socket but we can't currently send data.
+ * Nothing else to do.
+ */
+ *space_avail_now = false;
+ }
+
+ return 0;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_recv_init --
+ *
+ * Called at the start of a stream recv call with the socket lock held.
+ *
+ * Results: 0 on success. Negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_recv_init(struct sock *sk,
+ size_t target,
+ vsock_vmci_recv_notify_data *data)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ data->consume_head = 0;
+ data->produce_tail = 0;
+ data->notify_on_block = false;
+
+ if (PKT_FIELD(vsk, write_notify_min_window) < target + 1) {
+ ASSERT(target < vsk->consume_size);
+ PKT_FIELD(vsk, write_notify_min_window) = target + 1;
+ if (PKT_FIELD(vsk, write_notify_window) <
+ PKT_FIELD(vsk, write_notify_min_window)) {
+ /*
+ * If the current window is smaller than the new
+ * minimal window size, we need to reevaluate whether
+ * we need to notify the sender. If the number of ready
+ * bytes are smaller than the new window, we need to
+ * send a notification to the sender before we block.
+ */
+
+ PKT_FIELD(vsk, write_notify_window) =
+ PKT_FIELD(vsk, write_notify_min_window);
+ data->notify_on_block = true;
+ }
+ }
+
+ return 0;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_recv_pre_block --
+ *
+ * Called right before a socket is about to block with the socket lock held.
+ * The socket lock may have been released between the entry function and the
+ * preblock call.
+ *
+ * Note: This function may be called multiple times before the post block
+ * function is called.
+ *
+ * Results: 0 on success. Negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_recv_pre_block(struct sock *sk,
+ size_t target,
+ vsock_vmci_recv_notify_data *data)
+{
+ int err = 0;
+
+ vsock_vmci_block_update_write_window(sk);
+
+ if (data->notify_on_block) {
+ err = vsock_vmci_send_read_notification(sk);
+ if (err < 0)
+ return err;
+
+ data->notify_on_block = false;
+ }
+
+ return err;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_recv_post_dequeue --
+ *
+ * Called right after we dequeue / peek data from a socket.
+ *
+ * Results: 0 on success. Negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_recv_post_dequeue(struct sock *sk,
+ size_t target,
+ ssize_t copied,
+ bool data_read,
+ vsock_vmci_recv_notify_data *data)
+{
+ vsock_vmci_sock *vsk;
+ int err;
+ bool was_full = false;
+ u64 free_space;
+
+ vsk = vsock_sk(sk);
+ err = 0;
+
+ if (data_read) {
+ smp_mb();
+
+ free_space = vmci_qpair_consume_free_space(vsk->qpair);
+ was_full = free_space == copied;
+
+ if (was_full)
+ PKT_FIELD(vsk, peer_waiting_write) = true;
+
+ err = vsock_vmci_send_read_notification(sk);
+ if (err < 0)
+ return err;
+
+ /* See the comment in vsock_vmci_notify_pkt_send_post_enqueue */
+ sk->sk_data_ready(sk, 0);
+ }
+
+ return err;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_send_init --
+ *
+ * Called at the start of a stream send call with the socket lock held.
+ *
+ * Results: 0 on success. A negative error code on failure.
+ *
+ * Side effects:
+ */
+
+static int
+vsock_vmci_notify_pkt_send_init(struct sock *sk,
+ vsock_vmci_send_notify_data *data)
+{
+ data->consume_head = 0;
+ data->produce_tail = 0;
+
+ return 0;
+}
+
+/*
+ *
+ * vsock_vmci_notifySendPostEnqueue --
+ *
+ * Called right after we enqueue data to a socket.
+ *
+ * Results: 0 on success. Negative error on failure.
+ *
+ * Side effects: None.
+ */
+
+static int
+vsock_vmci_notify_pkt_send_post_enqueue(struct sock *sk,
+ ssize_t written,
+ vsock_vmci_send_notify_data *data)
+{
+ int err = 0;
+ vsock_vmci_sock *vsk;
+ bool sent_wrote = false;
+ bool was_empty;
+ int retries = 0;
+
+ vsk = vsock_sk(sk);
+
+ smp_mb();
+
+ was_empty = (vmci_qpair_produce_buf_ready(vsk->qpair) == written);
+ if (was_empty) {
+ while (!(vsk->peer_shutdown & RCV_SHUTDOWN) &&
+ !sent_wrote && retries < VSOCK_MAX_DGRAM_RESENDS) {
+ err = VSOCK_SEND_WROTE(sk);
+ if (err >= 0)
+ sent_wrote = true;
+
+ retries++;
+ }
+ }
+
+ if (retries >= VSOCK_MAX_DGRAM_RESENDS && !sent_wrote) {
+ printk
+ ("%p unable to send wrote notification to peer.\n",
+ sk);
+ return err;
+ }
+
+ return err;
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_handle_pkt
+ *
+ * Called when a notify packet is recieved for a socket in the connected state.
+ * Note this might be called from a bottom half.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void
+vsock_vmci_notify_pkt_handle_pkt(struct sock *sk,
+ vsock_packet *pkt,
+ bool bottom_half,
+ struct sockaddr_vm *dst,
+ struct sockaddr_vm *src, bool *pkt_processed)
+{
+ bool processed = false;
+
+ switch (pkt->type) {
+ case VSOCK_PACKET_TYPE_WROTE:
+ vsock_vmci_handle_wrote(sk, pkt, bottom_half, dst, src);
+ processed = true;
+ break;
+ case VSOCK_PACKET_TYPE_READ:
+ vsock_vmci_handle_read(sk, pkt, bottom_half, dst, src);
+ processed = true;
+ break;
+ }
+
+ if (pkt_processed)
+ *pkt_processed = processed;
+
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_process_request
+ *
+ * Called near the end of process request.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void vsock_vmci_notify_pkt_process_request(struct sock *sk)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ PKT_FIELD(vsk, write_notify_window) = vsk->consume_size;
+ if (vsk->consume_size < PKT_FIELD(vsk, write_notify_min_window))
+ PKT_FIELD(vsk, write_notify_min_window) = vsk->consume_size;
+
+}
+
+/*
+ *
+ * vsock_vmci_notify_pkt_process_negotiate
+ *
+ * Called near the end of process negotiate.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static void vsock_vmci_notify_pkt_process_negotiate(struct sock *sk)
+{
+ vsock_vmci_sock *vsk = vsock_sk(sk);
+
+ PKT_FIELD(vsk, write_notify_window) = vsk->consume_size;
+ if (vsk->consume_size < PKT_FIELD(vsk, write_notify_min_window))
+ PKT_FIELD(vsk, write_notify_min_window) = vsk->consume_size;
+
+}
+
+/* Socket always on control packet based operations. */
+vsock_vmci_notify_ops vsock_vmci_notify_pkt_q_state_ops = {
+ vsock_vmci_notify_pkt_socket_init,
+ vsock_vmci_notify_pkt_socket_destruct,
+ vsock_vmci_notify_pkt_poll_in,
+ vsock_vmci_notify_pkt_poll_out,
+ vsock_vmci_notify_pkt_handle_pkt,
+ vsock_vmci_notify_pkt_recv_init,
+ vsock_vmci_notify_pkt_recv_pre_block,
+ NULL, /* recv_pre_dequeue */
+ vsock_vmci_notify_pkt_recv_post_dequeue,
+ vsock_vmci_notify_pkt_send_init,
+ NULL, /* send_pre_block */
+ NULL, /* send_pre_enqueue */
+ vsock_vmci_notify_pkt_send_post_enqueue,
+ vsock_vmci_notify_pkt_process_request,
+ vsock_vmci_notify_pkt_process_negotiate,
+};
diff --git a/net/vmw_vsock/vmci_sockets.h b/net/vmw_vsock/vmci_sockets.h
new file mode 100644
index 0000000..6e6fd98
--- /dev/null
+++ b/net/vmw_vsock/vmci_sockets.h
@@ -0,0 +1,517 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2007-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * vmci_sockets.h --
+ *
+ * VMCI sockets public constants and types.
+ */
+
+#ifndef _VMCI_SOCKETS_H_
+#define _VMCI_SOCKETS_H_
+
+#if !defined(__KERNEL__)
+#include <sys/socket.h>
+#endif
+
+/*
+ * \brief Option name for STREAM socket buffer size.
+ *
+ * Use as the option name in \c setsockopt(3) or \c getsockopt(3) to set or get
+ * an \c unsigned \c long \c long that specifies the size of the buffer
+ * underlying a vSockets STREAM socket.
+ *
+ * \note Value is clamped to the MIN and MAX.
+ *
+ * \see vmci_sock_get_af_value_fd() \see SO_VMCI_BUFFER_MIN_SIZE \see
+ * SO_VMCI_BUFFER_MAX_SIZE
+ *
+ * An example is given below.
+ *
+ * \code int vmciFd; int af = vmci_sock_get_af_value_fd(&vmciFd); unsigned long
+ * long val = 0x1000; int fd = socket(af, SOCK_STREAM, 0); setsockopt(fd, af,
+ * SO_VMCI_BUFFER_SIZE, &val, sizeof val); ... close(fd);
+ * vmci_sock_release_af_value_fd(vmciFd); \endcode
+ */
+
+#define SO_VMCI_BUFFER_SIZE 0
+
+/*
+ * \brief Option name for STREAM socket minimum buffer size.
+ *
+ * Use as the option name in \c setsockopt(3) or \c getsockopt(3) to set or get
+ * an \c unsigned \c long \c long that specifies the minimum size allowed for
+ * the buffer underlying a vSockets STREAM socket.
+ *
+ * \see vmci_sock_get_af_value_fd() \see SO_VMCI_BUFFER_SIZE \see
+ * SO_VMCI_BUFFER_MAX_SIZE
+ *
+ * An example is given below.
+ *
+ * \code int vmciFd; int af = vmci_sock_get_af_value_fd(&vmciFd); unsigned long
+ * long val = 0x500; int fd = socket(af, SOCK_STREAM, 0); setsockopt(fd, af,
+ * SO_VMCI_BUFFER_MIN_SIZE, &val, sizeof val); ... close(fd);
+ * vmci_sock_release_af_value_fd(vmciFd); \endcode
+ */
+
+#define SO_VMCI_BUFFER_MIN_SIZE 1
+
+/*
+ * \brief Option name for STREAM socket maximum buffer size.
+ *
+ * Use as the option name in \c setsockopt(3) or \c getsockopt(3) to set or get
+ * an unsigned long long that specifies the maximum size allowed for the buffer
+ * underlying a vSockets STREAM socket.
+ *
+ * \see vmci_sock_get_af_value_fd() \see SO_VMCI_BUFFER_SIZE \see
+ * SO_VMCI_BUFFER_MIN_SIZE
+ *
+ * An example is given below.
+ *
+ * \code int vmciFd; int af = vmci_sock_get_af_value_fd(&vmciFd); unsigned long
+ * long val = 0x4000; int fd = socket(af, SOCK_STREAM, 0); setsockopt(fd, af,
+ * SO_VMCI_BUFFER_MAX_SIZE, &val, sizeof val); ... close(fd);
+ * vmci_sock_release_af_value_fd(vmciFd); \endcode
+ */
+
+#define SO_VMCI_BUFFER_MAX_SIZE 2
+
+/*
+ * \brief Option name for socket peer's host-specific VM ID.
+ *
+ * Use as the option name in \c getsockopt(3) to get a host-specific identifier
+ * for the peer endpoint's VM. The identifier is a signed integer.
+ *
+ * \note Only available for ESX (VMKernel/userworld) endpoints.
+ *
+ * An example is given below.
+ *
+ * \code int vmciFd; int af = vmci_sock_get_af_value_fd(&vmciFd); int id;
+ * socklen_t len = sizeof id; int fd = socket(af, SOCK_DGRAM, 0); getsockopt(fd,
+ * af, SO_VMCI_PEER_HOST_VM_ID, &id, &len); ... close(fd);
+ * vmci_sock_release_af_value_fd(vmciFd); \endcode
+ */
+
+#define SO_VMCI_PEER_HOST_VM_ID 3
+
+/*
+ * \brief Option name for socket's service label.
+ *
+ * Use as the option name in \c setsockopt(3) or \c getsockopt(3) to set or get
+ * the service label for a socket. The service label is a C-style
+ * NUL-terminated string.
+ *
+ * \note Only available for ESX (VMkernel/userworld) endpoints.
+ */
+
+#define SO_VMCI_SERVICE_LABEL 4
+
+/*
+ * \brief Option name for determining if a socket is trusted.
+ *
+ * Use as the option name in \c getsockopt(3) to determine if a socket is
+ * trusted. The value is a signed integer.
+ *
+ * An example is given below.
+ *
+ * \code int vmciFd; int af = vmci_sock_get_af_value_fd(&vmciFd); int trusted;
+ * socklen_t len = sizeof trusted; int fd = socket(af, SOCK_DGRAM, 0);
+ * getsockopt(fd, af, SO_VMCI_TRUSTED, &trusted, &len); ... close(fd);
+ * vmci_sock_release_af_value_fd(vmciFd); \endcode
+ */
+
+#define SO_VMCI_TRUSTED 5
+
+/*
+ * \brief Option name for STREAM socket connection timeout.
+ *
+ * Use as the option name in \c setsockopt(3) or \c getsockopt(3) to set or get
+ * the connection timeout for a STREAM socket. The value is platform
+ * dependent. On ESX, Linux and Mac OS, it is a \c struct \c timeval. On
+ * Windows, it is a \c DWORD.
+ *
+ * An example is given below.
+ *
+ * \code int vmciFd; int af = vmci_sock_get_af_value_fd(&vmciFd); struct
+ * timeval t = { 5, 100000 }; int fd = socket(af, SOCK_STREAM, 0);
+ * setsockopt(fd, af, SO_VMCI_CONNECT_TIMEOUT, &t, sizeof t); ... close(fd);
+ * vmci_sock_release_af_value_fd(vmciFd); \endcode
+ */
+
+#define SO_VMCI_CONNECT_TIMEOUT 6
+
+/*
+ * \brief Option name for using non-blocking send/receive.
+ *
+ * Use as the option name for \c setsockopt(3) or \c getsockopt(3) to set or
+ * get the non-blocking transmit/receive flag for a STREAM socket. This flag
+ * determines whether \c send() and \c recv() can be called in non-blocking
+ * contexts for the given socket. The value is a signed integer.
+ *
+ * This option is only relevant to kernel endpoints, where descheduling the
+ * thread of execution is not allowed, for example, while holding a spinlock.
+ * It is not to be confused with conventional non-blocking socket operations.
+ *
+ * \note Only available for VMKernel endpoints.
+ *
+ * An example is given below.
+ *
+ * \code int vmciFd; int af = vmci_sock_get_af_value_fd(&vmciFd); int nonblock;
+ * socklen_t len = sizeof nonblock; int fd = socket(af, SOCK_STREAM, 0);
+ * getsockopt(fd, af, SO_VMCI_NONBLOCK_TXRX, &nonblock, &len); ... close(fd);
+ * vmci_sock_release_af_value_fd(vmciFd); \endcode
+ */
+
+#define SO_VMCI_NONBLOCK_TXRX 7
+
+/*
+ * \brief The vSocket equivalent of INADDR_ANY.
+ *
+ * This works for the \c svm_cid field of sockaddr_vm and indicates the context
+ * ID of the current endpoint.
+ *
+ * \see sockaddr_vm
+ *
+ * An example is given below.
+ *
+ * \code int vmciFd; int af = vmci_sock_get_af_value_fd(&vmciFd); struct
+ * sockaddr_vm addr; int fd = socket(af, SOCK_DGRAM, 0); addr.svm_family = af;
+ * addr.svm_cid = VMADDR_CID_ANY; addr.svm_port = 2000; bind(fd, &addr, sizeof
+ * addr); ... close(fd); vmci_sock_release_af_value_fd(vmciFd); \endcode
+ */
+
+#define VMADDR_CID_ANY ((unsigned int)-1)
+
+/*
+ * \brief Bind to any available port.
+ *
+ * Works for the \c svm_port field of sockaddr_vm.
+ *
+ * \see sockaddr_vm
+ *
+ * An example is given below.
+ *
+ * \code int vmciFd; int af = vmci_sock_get_af_value_fd(&vmciFd); struct
+ * sockaddr_vm addr; int fd = socket(af, SOCK_DGRAM, 0); addr.svm_family = af;
+ * addr.svm_cid = VMADDR_CID_ANY; addr.svm_port = VMADDR_PORT_ANY; bind(fd,
+ * &addr, sizeof addr); ... close(fd); vmci_sock_release_af_value_fd(vmciFd);
+ * \endcode
+ */
+
+#define VMADDR_PORT_ANY ((unsigned int)-1)
+
+/*
+ * \brief Invalid vSockets version.
+ *
+ * \see VMCISock_Version()
+ */
+
+#define VMCI_SOCKETS_INVALID_VERSION ((unsigned int)-1)
+
+/*
+ * \brief The epoch (first) component of the vSockets version.
+ *
+ * A single byte representing the epoch component of the vSockets version.
+ *
+ * \see VMCISock_Version()
+ *
+ * An example is given below.
+ *
+ * \code unsigned int ver = VMCISock_Version(); unsigned char epoch =
+ * VMCI_SOCKETS_VERSION_EPOCH(ver); \endcode
+ */
+
+#define VMCI_SOCKETS_VERSION_EPOCH(_v) (((_v) & 0xFF000000) >> 24)
+
+/*
+ * \brief The major (second) component of the vSockets version.
+ *
+ * A single byte representing the major component of the vSockets version.
+ * Typically changes for every major release of a product.
+ *
+ * \see VMCISock_Version()
+ *
+ * An example is given below.
+ *
+ * \code unsigned int ver = VMCISock_Version(); unsigned char major =
+ * VMCI_SOCKETS_VERSION_MAJOR(ver); \endcode
+ */
+
+#define VMCI_SOCKETS_VERSION_MAJOR(_v) (((_v) & 0x00FF0000) >> 16)
+
+/*
+ * \brief The minor (third) component of the vSockets version.
+ *
+ * Two bytes representing the minor component of the vSockets version.
+ *
+ * \see VMCISock_Version()
+ *
+ * An example is given below.
+ *
+ * \code unsigned int ver = VMCISock_Version(); unsigned short minor =
+ * VMCI_SOCKETS_VERSION_MINOR(ver); \endcode
+ */
+
+#define VMCI_SOCKETS_VERSION_MINOR(_v) (((_v) & 0x0000FFFF))
+
+/** \cond PRIVATE */
+
+/** \endcond */
+
+/*
+ * \brief Address structure for vSockets.
+ *
+ * The address family should be set to whatever vmci_sock_get_af_value_fd()
+ * returns. The structure members should all align on their natural boundaries
+ * without resorting to compiler packing directives. The total size of this
+ * structure should be exactly the same as that of \c struct \c sockaddr.
+ *
+ * \see vmci_sock_get_af_value_fd()
+ */
+
+struct sockaddr_vm {
+
+ /** \brief Address family. \see vmci_sock_get_af_value_fd() */
+ sa_family_t svm_family;
+
+ /** \cond PRIVATE */
+ unsigned short svm_reserved1;
+ /** \endcond */
+
+ /** \brief Port. \see VMADDR_PORT_ANY */
+ unsigned int svm_port;
+
+ /** \brief Context ID. \see VMADDR_CID_ANY */
+ unsigned int svm_cid;
+
+ /** \cond PRIVATE */
+ unsigned char svm_zero[sizeof(struct sockaddr) -
+ sizeof(sa_family_t) -
+ sizeof(unsigned short) -
+ sizeof(unsigned int) - sizeof(unsigned int)];
+ /** \endcond */
+};
+
+#if defined(linux) && defined(__KERNEL__)
+int vmci_sock_get_local_c_id(void);
+#else
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <sys/ioctl.h>
+#include <unistd.h>
+
+#include <stdio.h>
+
+/** \cond PRIVATE */
+#define VMCI_SOCKETS_DEFAULT_DEVICE "/dev/vsock"
+#define VMCI_SOCKETS_CLASSIC_ESX_DEVICE "/vmfs/devices/char/vsock/vsock"
+#define VMCI_SOCKETS_VERSION 1972
+#define VMCI_SOCKETS_GET_AF_VALUE 1976
+#define VMCI_SOCKETS_GET_LOCAL_CID 1977
+/** \endcond */
+
+ /*
+ * VMCISock_Version
+ *
+ * \brief Retrieve the vSockets version.
+ *
+ * Returns the current version of vSockets. The version is a 32-bit
+ * unsigned integer that consist of three components: the epoch, the major
+ * version, and the minor version. Use the \c VMCI_SOCKETS_VERSION macros
+ * to extract the components.
+ *
+ * \see VMCI_SOCKETS_VERSION_EPOCH() \see VMCI_SOCKETS_VERSION_MAJOR() \see
+ * VMCI_SOCKETS_VERSION_MINOR()
+ *
+ * \retval VMCI_SOCKETS_INVALID_VERSION Not available. \retval other The
+ * current version.
+ *
+ * An example is given below.
+ *
+ * \code unsigned int ver = VMCISock_Version(); if (ver !=
+ * VMCI_SOCKETS_INVALID_VERSION) { printf("vSockets version=%d.%d.%d\n",
+ * VMCI_SOCKETS_VERSION_EPOCH(ver), VMCI_SOCKETS_VERSION_MAJOR(ver),
+ * VMCI_SOCKETS_VERSION_MINOR(ver)); } \endcode
+ */
+
+static inline unsigned int VMCISock_Version(void)
+{
+ int fd;
+ unsigned int version;
+
+ fd = open(VMCI_SOCKETS_DEFAULT_DEVICE, O_RDWR);
+ if (fd < 0) {
+ fd = open(VMCI_SOCKETS_CLASSIC_ESX_DEVICE, O_RDWR);
+ if (fd < 0)
+ return VMCI_SOCKETS_INVALID_VERSION;
+
+ }
+
+ if (ioctl(fd, VMCI_SOCKETS_VERSION, &version) < 0)
+ version = VMCI_SOCKETS_INVALID_VERSION;
+
+ close(fd);
+ return version;
+}
+
+ /*
+ * vmci_sock_get_af_value_fd
+ *
+ * \brief Retrieve the address family value for vSockets.
+ *
+ * Returns the value to be used for the VMCI Sockets address family. This
+ * value should be used as the domain argument to \c socket(2) (when you
+ * might otherwise use \c AF_INET). For VMCI Socket-specific options, this
+ * value should also be used for the level argument to \c setsockopt(2)
+ * (when you might otherwise use \c SOL_TCP).
+ *
+ * \see vmci_sock_release_af_value_fd() \see sockaddr_vm
+ *
+ * \param[out] out_fd File descriptor to the VMCI device. The address
+ * family value is valid until this descriptor is closed. This parameter is
+ * only valid if the return value is not -1. Call
+ * vmci_sock_release_af_value_fd() to close this descriptor.
+ *
+ * \retval -1 Not available. \retval other The address family value.
+ *
+ * An example is given below.
+ *
+ * \code int vmciFd; int af = vmci_sock_get_af_value_fd(&vmciFd); if (af !=
+ * -1) { int fd = socket(af, SOCK_STREAM, 0); ... close(fd); close(vmciFd); }
+ * \endcode
+ */
+
+static inline int vmci_sock_get_af_value_fd(int *out_fd)
+{
+ int fd;
+ int family;
+
+ fd = open(VMCI_SOCKETS_DEFAULT_DEVICE, O_RDWR);
+ if (fd < 0) {
+ fd = open(VMCI_SOCKETS_CLASSIC_ESX_DEVICE, O_RDWR);
+ if (fd < 0)
+ return -1;
+
+ }
+
+ if (ioctl(fd, VMCI_SOCKETS_GET_AF_VALUE, &family) < 0)
+ family = -1;
+
+ if (family < 0)
+ close(fd);
+ else if (out_fd)
+ *out_fd = fd;
+
+ return family;
+}
+
+ /** \cond PRIVATE */
+ /*
+ * vmci_sock_get_af_value
+ *
+ * \brief Retrieve the address family value for vSockets.
+ *
+ * Returns the value to be used for the VMCI Sockets address family. This
+ * value should be used as the domain argument to \c socket(2) (when you
+ * might otherwise use \c AF_INET). For VMCI Socket-specific options, this
+ * value should also be used for the level argument to \c setsockopt(2)
+ * (when you might otherwise use \c SOL_TCP).
+ *
+ * \note This function leaves its descriptor to the vsock device open so
+ * that the socket implementation knows that the socket family is still in
+ * use. This is done because the address family is registered with the
+ * kernel on-demand and a notification is needed to unregister the address
+ * family. Use of this function is thus discouraged; please use
+ * vmci_sock_get_af_value_fd() instead.
+ *
+ * \see vmci_sock_get_af_value_fd() \see sockaddr_vm
+ *
+ * \retval -1 Not available. \retval other The address family value.
+ *
+ * An example is given below.
+ *
+ * \code int af = vmci_sock_get_af_value(); if (af != -1) { int fd =
+ * socket(af, SOCK_STREAM, 0); ... close(fd); } \endcode
+ */
+
+static inline int vmci_sock_get_af_value(void)
+{
+ return vmci_sock_get_af_value_fd(NULL);
+}
+
+ /** \endcond PRIVATE */
+
+ /*
+ * vmci_sock_release_af_value_fd
+ *
+ * \brief Release the file descriptor obtained when retrieving the address
+ * family value.
+ *
+ * Use this to release the file descriptor obtained by calling
+ * vmci_sock_get_af_value_fd().
+ *
+ * \see vmci_sock_get_af_value_fd()
+ *
+ * \param[in] fd File descriptor to the VMCI device.
+ */
+
+static inline void vmci_sock_release_af_value_fd(int fd)
+{
+ if (fd >= 0)
+ close(fd);
+
+}
+
+ /*
+ * vmci_sock_get_local_c_id
+ *
+ * \brief Retrieve the current context ID.
+ *
+ * \see VMADDR_CID_ANY
+ *
+ * \retval VMADDR_CID_ANY Not available. \retval other The current context
+ * ID.
+ *
+ * An example is given below.
+ *
+ * \code int vmciFd; int af = vmci_sock_get_af_value_fd(&vmciFd); struct
+ * sockaddr_vm addr; addr.svm_family = af; addr.svm_cid =
+ * vmci_sock_get_local_c_id(); vmci_sock_release_af_value_fd(vmciFd);
+ * \endcode
+ */
+
+static inline unsigned int vmci_sock_get_local_c_id(void)
+{
+ int fd;
+ unsigned int context_id;
+
+ fd = open(VMCI_SOCKETS_DEFAULT_DEVICE, O_RDWR);
+ if (fd < 0) {
+ fd = open(VMCI_SOCKETS_CLASSIC_ESX_DEVICE, O_RDWR);
+ if (fd < 0)
+ return VMADDR_CID_ANY;
+
+ }
+
+ if (ioctl(fd, VMCI_SOCKETS_GET_LOCAL_CID, &context_id) < 0)
+ context_id = VMADDR_CID_ANY;
+
+ close(fd);
+ return context_id;
+}
+#endif
+
+#endif
diff --git a/net/vmw_vsock/vmci_sockets_packet.h b/net/vmw_vsock/vmci_sockets_packet.h
new file mode 100644
index 0000000..fc9f434
--- /dev/null
+++ b/net/vmw_vsock/vmci_sockets_packet.h
@@ -0,0 +1,106 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * vmci_sockets_packet.h --
+ *
+ * Definition of VMCI Sockets packet format, constants, and types.
+ */
+
+#ifndef _VMCI_SOCKETS_PACKET_H_
+#define _VMCI_SOCKETS_PACKET_H_
+
+#include <linux/vmw_vmci_defs.h>
+#include <linux/vmw_vmci_api.h>
+
+/*
+ * STREAM control packets.
+ */
+
+/* If the packet format changes in a release then this should change too. */
+#define VSOCK_PACKET_VERSION 1
+
+/* The resource ID on which control packets are sent. */
+#define VSOCK_PACKET_RID 1
+
+/*
+ * Assert that the given packet is valid. We check that the two original
+ * reserved fields equal zero because the version of the common code that
+ * shipped with ESX 4.0 and WS 6.5 did so and will return a RST packet if they
+ * aren't set that way. For newer packet types added after that release we
+ * don't do this.
+ */
+#define VSOCK_PACKET_ASSERT(_p) \
+ do { \
+ ASSERT((_p)->type < VSOCK_PACKET_TYPE_MAX); \
+ if ((_p)->type < VSOCK_PACKET_TYPE_REQUEST2) { \
+ ASSERT(0 == (_p)->proto); \
+ ASSERT(0 == (_p)->_reserved2); \
+ } \
+ } while (0)
+
+typedef enum vsock_packet_type {
+ VSOCK_PACKET_TYPE_INVALID = 0,
+ VSOCK_PACKET_TYPE_REQUEST,
+ VSOCK_PACKET_TYPE_NEGOTIATE,
+ VSOCK_PACKET_TYPE_OFFER,
+ VSOCK_PACKET_TYPE_ATTACH,
+ VSOCK_PACKET_TYPE_WROTE,
+ VSOCK_PACKET_TYPE_READ,
+ VSOCK_PACKET_TYPE_RST,
+ VSOCK_PACKET_TYPE_SHUTDOWN,
+ VSOCK_PACKET_TYPE_WAITING_WRITE,
+ VSOCK_PACKET_TYPE_WAITING_READ,
+ VSOCK_PACKET_TYPE_REQUEST2,
+ VSOCK_PACKET_TYPE_NEGOTIATE2,
+ VSOCK_PACKET_TYPE_MAX
+} vsock_packet_type;
+
+typedef u16 vsock_proto_version;
+#define VSOCK_PROTO_INVALID 0
+#define VSOCK_PROTO_PKT_ON_NOTIFY (1 << 0)
+
+#define VSOCK_PROTO_ALL_SUPPORTED (VSOCK_PROTO_PKT_ON_NOTIFY)
+
+typedef struct vsock_waiting_info {
+ u64 generation;
+ u64 offset;
+} vsock_waiting_info;
+
+/*
+ * Control packet type for STREAM sockets. DGRAMs have no control packets nor
+ * special packet header for data packets, they are just raw VMCI DGRAM
+ * messages. For STREAMs, control packets are sent over the control channel
+ * while data is written and read directly from queue pairs with no packet
+ * format.
+ */
+typedef struct vsock_packet {
+ struct vmci_datagram dg;
+ u8 version;
+ u8 type;
+ vsock_proto_version proto;
+
+ u32 src_port;
+ u32 dst_port;
+ u32 _reserved2;
+ union {
+ u64 size;
+ u64 mode;
+ struct vmci_handle handle;
+ vsock_waiting_info wait;
+ } u;
+} vsock_packet;
+
+#endif
diff --git a/net/vmw_vsock/vsock_common.h b/net/vmw_vsock/vsock_common.h
new file mode 100644
index 0000000..fd7f2cd
--- /dev/null
+++ b/net/vmw_vsock/vsock_common.h
@@ -0,0 +1,130 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2007-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * vsockCommon.h --
+ *
+ * VSockets common constants, types and functions.
+ */
+
+#ifndef _VSOCK_COMMON_H_
+#define _VSOCK_COMMON_H_
+
+/*
+ * vmci_sock_get_af_value_int is defined separately from vmci_sock_get_af_value
+ * because it is used in several different contexts. In particular it is called
+ * from vsock_addr.c which gets compiled into both our kernel modules as well
+ * as the user level vsock library. In the linux kernel we need different
+ * behavior than external kernel modules using VMCI Sockets api inside the
+ * kernel. FIXME
+ */
+
+#if defined __KERNEL__
+#include <linux/types.h>
+#include <linux/mm.h>
+#include <asm/page.h>
+#else
+/* In userland, just use the normal exported userlevel api. */
+#define vmci_sock_get_af_value_int() vmci_sock_get_af_value()
+#endif
+
+#include <linux/vmw_vmci_defs.h>
+#include <linux/vmw_vmci_api.h>
+
+#include "vmci_sockets.h"
+#include "vsock_addr.h"
+
+#define ASSERT(cond) BUG_ON(!(cond))
+#define ASSERT_ON_COMPILE(cond) BUILD_BUG_ON(!(cond))
+
+#ifdef __x86_64__
+#define FMT64 "ll"
+#else
+#define FMT64 "L"
+#endif
+
+#ifndef PAGE_SHIFT
+#if defined __i386__
+#define PAGE_SHIFT 12
+#elif defined __arm__
+#define PAGE_SHIFT 12
+#else
+#error
+#endif
+#endif
+
+#ifndef PAGE_SIZE
+#define PAGE_SIZE (1<<PAGE_SHIFT)
+#endif
+
+#define MAX_UINT32 ((u32)0xffffffff)
+
+#ifndef ESYSNOTREADY
+#define ESYSNOTREADY EOPNOTSUPP
+#endif
+
+#define sockerr() errno
+#define sockerr2err(_e) (((_e) > 0) ? -(_e) : (_e))
+#define SS_LISTEN 255
+
+extern vmci_id vmci_get_context_id(void);
+
+/*
+ * Helper function to determine if the given handle points to the local context.
+ * Returns TRUE if the given handle is for the local context, FALSE otherwise.
+ */
+
+static inline bool vsock_vmci_is_local(struct vmci_handle handle)
+{
+ return vmci_get_context_id() == VMCI_HANDLE_TO_CONTEXT_ID(handle);
+}
+
+/*
+ * Helper function to convert from a VMCI error code to a VSock error code.
+ */
+
+static inline s32 vsock_vmci_error_to_vsock_error(s32 vmci_error)
+{
+ int err;
+
+ switch (vmci_error) {
+ case VMCI_ERROR_NO_MEM:
+ err = ENOMEM;
+ break;
+ case VMCI_ERROR_DUPLICATE_ENTRY:
+ case VMCI_ERROR_ALREADY_EXISTS:
+ err = EADDRINUSE;
+ break;
+ case VMCI_ERROR_NO_ACCESS:
+ err = EPERM;
+ break;
+ case VMCI_ERROR_NO_RESOURCES:
+ err = ENOBUFS;
+ break;
+ case VMCI_ERROR_INVALID_RESOURCE:
+ err = EHOSTUNREACH;
+ break;
+ case VMCI_ERROR_MODULE_NOT_LOADED:
+ err = ESYSNOTREADY;
+ break;
+ case VMCI_ERROR_INVALID_ARGS:
+ default:
+ err = EINVAL;
+ }
+
+ return sockerr2err(err);
+}
+
+#endif
diff --git a/net/vmw_vsock/vsock_packet.h b/net/vmw_vsock/vsock_packet.h
new file mode 100644
index 0000000..76bcac0
--- /dev/null
+++ b/net/vmw_vsock/vsock_packet.h
@@ -0,0 +1,131 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2007-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * vsockPacket.h --
+ *
+ * Packet constants, types and functions.
+ */
+
+#ifndef _VSOCK_PACKET_H_
+#define _VSOCK_PACKET_H_
+
+#include "vmci_sockets_packet.h"
+
+/*
+ *
+ * vsock_packet_init --
+ *
+ * Initialize the given packet. The packet version is set and the fields are
+ * filled out. Reserved fields are cleared.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static inline void
+vsock_packet_init(vsock_packet *pkt,
+ struct sockaddr_vm *src,
+ struct sockaddr_vm *dst,
+ u8 type,
+ u64 size,
+ u64 mode,
+ vsock_waiting_info *wait,
+ vsock_proto_version proto, struct vmci_handle handle)
+{
+ BUG_ON(!pkt);
+ BUG_ON(vsock_addr_validate(src) != 0);
+ BUG_ON(vsock_addr_validate(dst) != 0);
+
+ /*
+ * We register the stream control handler as an any cid handle so we
+ * must always send from a source address of VMADDR_CID_ANY
+ */
+ pkt->dg.src = VMCI_MAKE_HANDLE(VMADDR_CID_ANY, VSOCK_PACKET_RID);
+ pkt->dg.dst = VMCI_MAKE_HANDLE(dst->svm_cid, VSOCK_PACKET_RID);
+ pkt->dg.payload_size = sizeof *pkt - sizeof pkt->dg;
+ pkt->version = VSOCK_PACKET_VERSION;
+ pkt->type = type;
+ pkt->src_port = src->svm_port;
+ pkt->dst_port = dst->svm_port;
+ memset(&pkt->proto, 0, sizeof pkt->proto);
+ memset(&pkt->_reserved2, 0, sizeof pkt->_reserved2);
+
+ switch (pkt->type) {
+ case VSOCK_PACKET_TYPE_INVALID:
+ pkt->u.size = 0;
+ break;
+
+ case VSOCK_PACKET_TYPE_REQUEST:
+ case VSOCK_PACKET_TYPE_NEGOTIATE:
+ pkt->u.size = size;
+ break;
+
+ case VSOCK_PACKET_TYPE_OFFER:
+ case VSOCK_PACKET_TYPE_ATTACH:
+ pkt->u.handle = handle;
+ break;
+
+ case VSOCK_PACKET_TYPE_WROTE:
+ case VSOCK_PACKET_TYPE_READ:
+ case VSOCK_PACKET_TYPE_RST:
+ pkt->u.size = 0;
+ break;
+
+ case VSOCK_PACKET_TYPE_SHUTDOWN:
+ pkt->u.mode = mode;
+ break;
+
+ case VSOCK_PACKET_TYPE_WAITING_READ:
+ case VSOCK_PACKET_TYPE_WAITING_WRITE:
+ ASSERT(wait);
+ memcpy(&pkt->u.wait, wait, sizeof pkt->u.wait);
+ break;
+
+ case VSOCK_PACKET_TYPE_REQUEST2:
+ case VSOCK_PACKET_TYPE_NEGOTIATE2:
+ pkt->u.size = size;
+ pkt->proto = proto;
+ break;
+ }
+
+ VSOCK_PACKET_ASSERT(pkt);
+}
+
+/*
+ *
+ * vsock_packet_get_addresses --
+ *
+ * Get the local and remote addresses from the given packet.
+ *
+ * Results: None.
+ *
+ * Side effects: None.
+ */
+
+static inline void
+vsock_packet_get_addresses(vsock_packet *pkt,
+ struct sockaddr_vm *local,
+ struct sockaddr_vm *remote)
+{
+ VSOCK_PACKET_ASSERT(pkt);
+ vsock_addr_init(local, VMCI_HANDLE_TO_CONTEXT_ID(pkt->dg.dst),
+ pkt->dst_port);
+ vsock_addr_init(remote, VMCI_HANDLE_TO_CONTEXT_ID(pkt->dg.src),
+ pkt->src_port);
+}
+
+#endif
diff --git a/net/vmw_vsock/vsock_version.h b/net/vmw_vsock/vsock_version.h
new file mode 100644
index 0000000..9a6a63b
--- /dev/null
+++ b/net/vmw_vsock/vsock_version.h
@@ -0,0 +1,29 @@
+/*
+ * VMware vSockets Driver
+ *
+ * Copyright (C) 2011-2012 VMware, Inc. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation version 2 and no later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ */
+
+/*
+ * vsock_version.h --
+ *
+ * Version definitions for the Linux vsock driver.
+ */
+
+#ifndef _VSOCK_VERSION_H_
+#define _VSOCK_VERSION_H_
+
+#define VSOCK_DRIVER_VERSION 1.0.0.0-k
+#define VSOCK_DRIVER_VERSION_COMMAS 1,0,0,0
+#define VSOCK_DRIVER_VERSION_STRING "1.0.0.0-k"
+
+#endif /* _VSOCK_VERSION_H_ */
^ permalink raw reply related
* Re: [PATCH v11 1/7] mm: adjust address_space_operations.migratepage() return code
From: Andrew Morton @ 2012-11-07 19:56 UTC (permalink / raw)
To: Rafael Aquini
Cc: Rik van Riel, Konrad Rzeszutek Wilk, Michael S. Tsirkin,
linux-kernel, virtualization, linux-mm, Peter Zijlstra,
Andi Kleen, Minchan Kim, Paul E. McKenney
In-Reply-To: <74bc30697313206e1225f6fc658bc5952b588dcc.1352256085.git.aquini@redhat.com>
On Wed, 7 Nov 2012 01:05:48 -0200
Rafael Aquini <aquini@redhat.com> wrote:
> This patch introduces MIGRATEPAGE_SUCCESS as the default return code
> for address_space_operations.migratepage() method and documents the
> expected return code for the same method in failure cases.
I hit a large number of rejects applying this against linux-next. Due
to the increasingly irritating sched/numa code in there.
I attempted to fix it up and also converted some (but not all) of the
implicit tests of `rc' against zero.
Please check the result very carefully - more changes will be needed.
All those
- if (rc)
+ if (rc != MIGRATEPAGE_SUCCESS)
changes are a pain. Perhaps we shouldn't bother.
^ permalink raw reply
* Re: [PATCH v11 5/7] virtio_balloon: introduce migration primitives to balloon pages
From: Andrew Morton @ 2012-11-07 19:58 UTC (permalink / raw)
To: Rafael Aquini
Cc: Rik van Riel, Konrad Rzeszutek Wilk, Michael S. Tsirkin,
linux-kernel, virtualization, linux-mm, Peter Zijlstra,
Andi Kleen, Minchan Kim, Paul E. McKenney
In-Reply-To: <265aaff9a79f503672f0cdcdff204114b5b5ba5b.1352256088.git.aquini@redhat.com>
On Wed, 7 Nov 2012 01:05:52 -0200
Rafael Aquini <aquini@redhat.com> wrote:
> Memory fragmentation introduced by ballooning might reduce significantly
> the number of 2MB contiguous memory blocks that can be used within a guest,
> thus imposing performance penalties associated with the reduced number of
> transparent huge pages that could be used by the guest workload.
>
> Besides making balloon pages movable at allocation time and introducing
> the necessary primitives to perform balloon page migration/compaction,
> this patch also introduces the following locking scheme, in order to
> enhance the syncronization methods for accessing elements of struct
> virtio_balloon, thus providing protection against concurrent access
> introduced by parallel memory migration threads.
>
> ...
>
> @@ -122,18 +128,25 @@ static void set_page_pfns(u32 pfns[], struct page *page)
>
> static void fill_balloon(struct virtio_balloon *vb, size_t num)
> {
> + struct balloon_dev_info *vb_dev_info = vb->vb_dev_info;
> +
> + static DEFINE_RATELIMIT_STATE(fill_balloon_rs,
> + DEFAULT_RATELIMIT_INTERVAL,
> + DEFAULT_RATELIMIT_BURST);
> +
> /* We can only do one array worth at a time. */
> num = min(num, ARRAY_SIZE(vb->pfns));
>
> + mutex_lock(&vb->balloon_lock);
> for (vb->num_pfns = 0; vb->num_pfns < num;
> vb->num_pfns += VIRTIO_BALLOON_PAGES_PER_PAGE) {
> - struct page *page = alloc_page(GFP_HIGHUSER | __GFP_NORETRY |
> - __GFP_NOMEMALLOC | __GFP_NOWARN);
> + struct page *page = balloon_page_enqueue(vb_dev_info);
> +
> if (!page) {
> - if (printk_ratelimit())
> + if (__ratelimit(&fill_balloon_rs))
> dev_printk(KERN_INFO, &vb->vdev->dev,
> "Out of puff! Can't get %zu pages\n",
> - num);
> + VIRTIO_BALLOON_PAGES_PER_PAGE);
> /* Sleep for at least 1/5 of a second before retry. */
> msleep(200);
> break;
linux-next's fill_balloon() has already been converted to
dev_info_ratelimited(). I fixed everything up. Please check the result.
^ permalink raw reply
* Re: [PATCH v11 3/7] mm: introduce a common interface for balloon pages mobility
From: Andrew Morton @ 2012-11-07 21:02 UTC (permalink / raw)
To: Rafael Aquini
Cc: Rik van Riel, Konrad Rzeszutek Wilk, Michael S. Tsirkin,
linux-kernel, virtualization, linux-mm, Peter Zijlstra,
Andi Kleen, Minchan Kim, Paul E. McKenney
In-Reply-To: <4ea10ef1eb1544e12524c8ca7df20cf621395463.1352256087.git.aquini@redhat.com>
On Wed, 7 Nov 2012 01:05:50 -0200
Rafael Aquini <aquini@redhat.com> wrote:
> Memory fragmentation introduced by ballooning might reduce significantly
> the number of 2MB contiguous memory blocks that can be used within a guest,
> thus imposing performance penalties associated with the reduced number of
> transparent huge pages that could be used by the guest workload.
>
> This patch introduces a common interface to help a balloon driver on
> making its page set movable to compaction, and thus allowing the system
> to better leverage the compation efforts on memory defragmentation.
mm/migrate.c: In function 'unmap_and_move':
mm/migrate.c:899: error: 'COMPACTBALLOONRELEASED' undeclared (first use in this function)
mm/migrate.c:899: error: (Each undeclared identifier is reported only once
mm/migrate.c:899: error: for each function it appears in.)
You've been bad - you didn't test with your feature disabled.
Please do that. And not just compilation testing.
We can fix this one with a sucky macro. I think that's better than
unconditionally defining the enums.
--- a/include/linux/balloon_compaction.h~mm-introduce-a-common-interface-for-balloon-pages-mobility-fix
+++ a/include/linux/balloon_compaction.h
@@ -207,10 +207,8 @@ static inline gfp_t balloon_mapping_gfp_
return GFP_HIGHUSER;
}
-static inline void balloon_event_count(enum vm_event_item item)
-{
- return;
-}
+/* A macro, to avoid generating references to the undefined COMPACTBALLOON* */
+#define balloon_event_count(item) do { } while (0)
static inline bool balloon_compaction_check(void)
{
^ permalink raw reply
* Re: [PATCH v11 3/7] mm: introduce a common interface for balloon pages mobility
From: Rafael Aquini @ 2012-11-07 21:34 UTC (permalink / raw)
To: Andrew Morton
Cc: Rik van Riel, Konrad Rzeszutek Wilk, Michael S. Tsirkin,
linux-kernel, virtualization, linux-mm, Peter Zijlstra,
Andi Kleen, Minchan Kim, Paul E. McKenney
In-Reply-To: <20121107130207.214f16ea.akpm@linux-foundation.org>
On Wed, Nov 07, 2012 at 01:02:07PM -0800, Andrew Morton wrote:
> On Wed, 7 Nov 2012 01:05:50 -0200
> Rafael Aquini <aquini@redhat.com> wrote:
>
> > Memory fragmentation introduced by ballooning might reduce significantly
> > the number of 2MB contiguous memory blocks that can be used within a guest,
> > thus imposing performance penalties associated with the reduced number of
> > transparent huge pages that could be used by the guest workload.
> >
> > This patch introduces a common interface to help a balloon driver on
> > making its page set movable to compaction, and thus allowing the system
> > to better leverage the compation efforts on memory defragmentation.
>
>
> mm/migrate.c: In function 'unmap_and_move':
> mm/migrate.c:899: error: 'COMPACTBALLOONRELEASED' undeclared (first use in this function)
> mm/migrate.c:899: error: (Each undeclared identifier is reported only once
> mm/migrate.c:899: error: for each function it appears in.)
>
> You've been bad - you didn't test with your feature disabled.
> Please do that. And not just compilation testing.
>
Gasp... Shame on me, indeed. balloon_event_count() was a macro and I had it
tested a couple of review rounds earlier. You requested me to get rid of
preprocessor pirotech, and I did but I miserably failed on re-testing it.
I'm pretty sure Santa is not going to visit me this year.
Do you want me to resubmit this?
>
> We can fix this one with a sucky macro. I think that's better than
> unconditionally defining the enums.
>
> --- a/include/linux/balloon_compaction.h~mm-introduce-a-common-interface-for-balloon-pages-mobility-fix
> +++ a/include/linux/balloon_compaction.h
> @@ -207,10 +207,8 @@ static inline gfp_t balloon_mapping_gfp_
> return GFP_HIGHUSER;
> }
>
> -static inline void balloon_event_count(enum vm_event_item item)
> -{
> - return;
> -}
> +/* A macro, to avoid generating references to the undefined COMPACTBALLOON* */
> +#define balloon_event_count(item) do { } while (0)
>
> static inline bool balloon_compaction_check(void)
> {
>
^ permalink raw reply
* Re: [PATCH v11 1/7] mm: adjust address_space_operations.migratepage() return code
From: Rafael Aquini @ 2012-11-07 21:39 UTC (permalink / raw)
To: Andrew Morton
Cc: Rik van Riel, Konrad Rzeszutek Wilk, Michael S. Tsirkin,
linux-kernel, virtualization, linux-mm, Peter Zijlstra,
Andi Kleen, Minchan Kim, Paul E. McKenney
In-Reply-To: <20121107115610.c0cb650c.akpm@linux-foundation.org>
On Wed, Nov 07, 2012 at 11:56:10AM -0800, Andrew Morton wrote:
> On Wed, 7 Nov 2012 01:05:48 -0200
> Rafael Aquini <aquini@redhat.com> wrote:
>
> > This patch introduces MIGRATEPAGE_SUCCESS as the default return code
> > for address_space_operations.migratepage() method and documents the
> > expected return code for the same method in failure cases.
>
> I hit a large number of rejects applying this against linux-next. Due
> to the increasingly irritating sched/numa code in there.
>
> I attempted to fix it up and also converted some (but not all) of the
> implicit tests of `rc' against zero.
>
> Please check the result very carefully - more changes will be needed.
>
> All those
>
> - if (rc)
> + if (rc != MIGRATEPAGE_SUCCESS)
>
> changes are a pain. Perhaps we shouldn't bother.
Thanks for doing that.
This hunk at migrate_pages(), however, is not necessary:
@@ -1001,7 +1001,7 @@ out:
if (!swapwrite)
current->flags &= ~PF_SWAPWRITE;
- if (rc)
+ if (rc != MIGRATEPAGE_SUCCESS)
return rc;
Here, migrate_pages() is not testing rc for the migration success, but it's just trying to
devise the flow if it has to return -ENOMEM, actually.
I guess, a change to make that snippet more clear could be:
diff --git a/mm/migrate.c b/mm/migrate.c
index 77ed2d7..6562aee 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -987,7 +987,7 @@ int migrate_pages(struct list_head *from,
case -EAGAIN:
retry++;
break;
- case 0:
+ case MIGRATEPAGE_SUCCESS:
break;
default:
/* Permanent failure */
@@ -996,15 +996,12 @@ int migrate_pages(struct list_head *from,
}
}
}
- rc = 0;
+ rc = nr_failed + retry;
out:
if (!swapwrite)
current->flags &= ~PF_SWAPWRITE;
- if (rc)
- return rc;
-
- return nr_failed + retry;
+ return rc;
}
I can rebase this patch and resubmit if you prefer
^ permalink raw reply related
* Re: [PATCH v11 5/7] virtio_balloon: introduce migration primitives to balloon pages
From: Rafael Aquini @ 2012-11-07 22:02 UTC (permalink / raw)
To: Andrew Morton
Cc: Rik van Riel, Konrad Rzeszutek Wilk, Michael S. Tsirkin,
linux-kernel, virtualization, linux-mm, Peter Zijlstra,
Andi Kleen, Minchan Kim, Paul E. McKenney
In-Reply-To: <20121107115810.1ae286ac.akpm@linux-foundation.org>
On Wed, Nov 07, 2012 at 11:58:10AM -0800, Andrew Morton wrote:
> On Wed, 7 Nov 2012 01:05:52 -0200
> Rafael Aquini <aquini@redhat.com> wrote:
>
> > Memory fragmentation introduced by ballooning might reduce significantly
> > the number of 2MB contiguous memory blocks that can be used within a guest,
> > thus imposing performance penalties associated with the reduced number of
> > transparent huge pages that could be used by the guest workload.
> >
> > Besides making balloon pages movable at allocation time and introducing
> > the necessary primitives to perform balloon page migration/compaction,
> > this patch also introduces the following locking scheme, in order to
> > enhance the syncronization methods for accessing elements of struct
> > virtio_balloon, thus providing protection against concurrent access
> > introduced by parallel memory migration threads.
> >
> > ...
> >
> > @@ -122,18 +128,25 @@ static void set_page_pfns(u32 pfns[], struct page *page)
> >
> > static void fill_balloon(struct virtio_balloon *vb, size_t num)
> > {
> > + struct balloon_dev_info *vb_dev_info = vb->vb_dev_info;
> > +
> > + static DEFINE_RATELIMIT_STATE(fill_balloon_rs,
> > + DEFAULT_RATELIMIT_INTERVAL,
> > + DEFAULT_RATELIMIT_BURST);
> > +
> > /* We can only do one array worth at a time. */
> > num = min(num, ARRAY_SIZE(vb->pfns));
> >
> > + mutex_lock(&vb->balloon_lock);
> > for (vb->num_pfns = 0; vb->num_pfns < num;
> > vb->num_pfns += VIRTIO_BALLOON_PAGES_PER_PAGE) {
> > - struct page *page = alloc_page(GFP_HIGHUSER | __GFP_NORETRY |
> > - __GFP_NOMEMALLOC | __GFP_NOWARN);
> > + struct page *page = balloon_page_enqueue(vb_dev_info);
> > +
> > if (!page) {
> > - if (printk_ratelimit())
> > + if (__ratelimit(&fill_balloon_rs))
> > dev_printk(KERN_INFO, &vb->vdev->dev,
> > "Out of puff! Can't get %zu pages\n",
> > - num);
> > + VIRTIO_BALLOON_PAGES_PER_PAGE);
> > /* Sleep for at least 1/5 of a second before retry. */
> > msleep(200);
> > break;
>
> linux-next's fill_balloon() has already been converted to
> dev_info_ratelimited(). I fixed everything up. Please check the result.
Looks great, thanks for doing it
>
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@kvack.org. For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply
* Re: [PATCH] virtio-mmio: Fix irq parsing in command line parameter
From: Rusty Russell @ 2012-11-07 22:22 UTC (permalink / raw)
To: Lee Jones, Pawel Moll; +Cc: linux-kernel, virtualization
In-Reply-To: <20121105134413.GC5220@gmail.com>
Lee Jones <lee.jones@linaro.org> writes:
> On Mon, 05 Nov 2012, Pawel Moll wrote:
>
>> On 64-bit machines resource_size_t is a 64-bit value, while
>> sscanf() format for this argument was defined as "%u". Fixed
>> by using an intermediate local value of a known length.
>>
>> Also added cleaned up the resource creation and adde extra
>> comments to make the parameters parsing easier to follow.
>>
>> Reported-by: Lee Jones <lee.jones@linaro.org>
>> Signed-off-by: Pawel Moll <pawel.moll@arm.com>
>> ---
>> drivers/virtio/virtio_mmio.c | 26 ++++++++++++++++++--------
>> 1 file changed, 18 insertions(+), 8 deletions(-)
>
> Resorted to poaching now have we Pawel? ;)
I hope you were joking! Doing your work for you isn't poaching.
Your correct response was "thanks" followed by 'Tested-by:'.
Cheers,
Rusty.
^ permalink raw reply
* Re: [PATCH v2] virtio-mmio: Fix irq parsing in the command line
From: Rusty Russell @ 2012-11-07 22:47 UTC (permalink / raw)
Cc: Pawel Moll, virtualization
In-Reply-To: <1352297892-27838-1-git-send-email-pawel.moll@arm.com>
Pawel Moll <pawel.moll@arm.com> writes:
> On 64-bit machines resource_size_t is a 64-bit value, while
> sscanf() format for this argument was defined as "%u". Fixed
> by using an intermediate local value of a known length.
Actually, on 32-bit machines, too (eg. x86 with PAE). Otherwise we'd
just change it to %lu.
> + /* Get "@<base>:<irq>[:<id>]" chunks */
> processed = sscanf(str, "@%lli:%u%n:%d%n",
> - &base, &resources[1].start, &consumed,
> + &base, &irq, &consumed,
> &vm_cmdline_id, &consumed);
>
> + /*
> + * sscanf() processes 3 chunks if "<id>" is given, 2 if not;
> + * also there must be no extra characters after the last
> + * chunk, so str[consumed] should be '\0'
> + */
> if (processed < 2 || processed > 3 || str[consumed])
> return -EINVAL;
I would drop the > 3 case. It's unnecessary, and you're assuming
consumed doesn't add to the count, which may be true but is a documented
sscanf weirdness so I don't like to rely on it.
Thanks,
Rusty.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox