On 02/14/2018 03:50 AM, Dongwon Kim wrote:

Upload of intial version of core framework in hyper_DMABUF driver enabling DMA_BUF exchange between two different VMs in virtualized platform based on Hypervisor such as XEN.

Hyper_DMABUF drv's primary role is to import a DMA_BUF from originator then re-export it to another Linux VM so that it can be mapped and accessed in there.

This driver has two layers, one is so called, "core framework", which contains driver interface and core functions handling export/import of new hyper_DMABUF and its maintenance. This part of the driver is independent from Hypervisor so can work as is with any Hypervisor.

The other layer is called "Hypervisor Backend". This layer represents the interface between "core framework" and actual Hypervisor, handling memory sharing and communication. Not like "core framework", every Hypervisor needs it's own backend interface designed using its native mechanism for memory sharing and inter-VM communication.

This patch contains the first part, "core framework", which consists of 7 source files and 11 header files. Some brief description of these source code are attached below:

hyper_dmabuf_drv.c

• Linux driver interface and initialization/cleaning-up routines

hyper_dmabuf_ioctl.c

• IOCTLs calls for export/import of DMA-BUF comm channel's creation and destruction.

hyper_dmabuf_sgl_proc.c

• Provides methods to managing DMA-BUF for exporing and importing. For exporting, extraction of pages, sharing pages via procedures in "Backend" and notifying importing VM exist. For importing, all operations related to the reconstruction of DMA-BUF (with shared pages) on importer's side are defined.

hyper_dmabuf_ops.c

• Standard DMA-BUF operations for hyper_DMABUF reconstructed on importer's side.

hyper_dmabuf_list.c

• Lists for storing exported and imported hyper_DMABUF to keep track of remote usage of hyper_DMABUF currently being shared.

hyper_dmabuf_msg.c

• Defines messages exchanged between VMs (exporter and importer) and function calls for sending and parsing (when received) those.

hyper_dmabuf_id.c

• Contains methods to generate and manage "hyper_DMABUF id" for each hyper_DMABUF being exported. It is a global handle for a hyper_DMABUF, which another VM needs to know to import it.

hyper_dmabuf_struct.h

• Contains data structures of importer or exporter hyper_DMABUF

include/uapi/linux/hyper_dmabuf.h

• Contains definition of data types and structures referenced by user application to interact with driver

Signed-off-by: Dongwon Kim dongwon.kim@intel.com Signed-off-by: Mateusz Polrola mateuszx.potrola@intel.com

---

drivers/dma-buf/Kconfig | 2 + drivers/dma-buf/Makefile | 1 + drivers/dma-buf/hyper_dmabuf/Kconfig | 23 + drivers/dma-buf/hyper_dmabuf/Makefile | 34 ++ drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_drv.c | 254 ++++++++ drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_drv.h | 111 ++++ drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_id.c | 135 +++++ drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_id.h | 53 ++ drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ioctl.c | 672 +++++++++++++++++++++ drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ioctl.h | 52 ++ drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_list.c | 294 +++++++++ drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_list.h | 73 +++ drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_msg.c | 320 ++++++++++ drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_msg.h | 87 +++ drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ops.c | 264 ++++++++ drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ops.h | 34 ++ .../dma-buf/hyper_dmabuf/hyper_dmabuf_sgl_proc.c | 256 ++++++++ .../dma-buf/hyper_dmabuf/hyper_dmabuf_sgl_proc.h | 43 ++ drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_struct.h | 131 ++++ include/uapi/linux/hyper_dmabuf.h | 87 +++ 20 files changed, 2926 insertions(+) create mode 100644 drivers/dma-buf/hyper_dmabuf/Kconfig create mode 100644 drivers/dma-buf/hyper_dmabuf/Makefile create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_drv.c create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_drv.h create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_id.c create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_id.h create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ioctl.c create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ioctl.h create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_list.c create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_list.h create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_msg.c create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_msg.h create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ops.c create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ops.h create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_sgl_proc.c create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_sgl_proc.h create mode 100644 drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_struct.h create mode 100644 include/uapi/linux/hyper_dmabuf.h

diff --git a/drivers/dma-buf/Kconfig b/drivers/dma-buf/Kconfig index ed3b785bae37..09ccac1768e3 100644 --- a/drivers/dma-buf/Kconfig +++ b/drivers/dma-buf/Kconfig @@ -30,4 +30,6 @@ config SW_SYNC WARNING: improper use of this can result in deadlocking kernel drivers from userspace. Intended for test and debug only. +source "drivers/dma-buf/hyper_dmabuf/Kconfig"

• endmenu

diff --git a/drivers/dma-buf/Makefile b/drivers/dma-buf/Makefile index c33bf8863147..445749babb19 100644 --- a/drivers/dma-buf/Makefile +++ b/drivers/dma-buf/Makefile @@ -1,3 +1,4 @@ obj-y := dma-buf.o dma-fence.o dma-fence-array.o reservation.o seqno-fence.o obj-$(CONFIG_SYNC_FILE) += sync_file.o obj-$(CONFIG_SW_SYNC) += sw_sync.o sync_debug.o +obj-$(CONFIG_HYPER_DMABUF) += ./hyper_dmabuf/ diff --git a/drivers/dma-buf/hyper_dmabuf/Kconfig b/drivers/dma-buf/hyper_dmabuf/Kconfig new file mode 100644 index 000000000000..5ebf516d65eb --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/Kconfig @@ -0,0 +1,23 @@ +menu "HYPER_DMABUF"

+config HYPER_DMABUF

• tristate "Enables hyper dmabuf driver"
• default y

Not sure you want this enabled by default

• help

•  This option enables Hyper_DMABUF driver.
  

•  This driver works as abstraction layer that export and import
  

•  DMA_BUF from/to another virtual OS running on the same HW platform
  

•  powered by a hypervisor
  

+config HYPER_DMABUF_SYSFS

• bool "Enable sysfs information about hyper DMA buffers"
• default y

Ditto

• depends on HYPER_DMABUF
• help

•  Expose run-time information about currently imported and exported buffers
  

•  registered in EXPORT and IMPORT list in Hyper_DMABUF driver.
  

•  The location of sysfs is under "...."
  

+endmenu diff --git a/drivers/dma-buf/hyper_dmabuf/Makefile b/drivers/dma-buf/hyper_dmabuf/Makefile new file mode 100644 index 000000000000..3908522b396a --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/Makefile @@ -0,0 +1,34 @@ +TARGET_MODULE:=hyper_dmabuf

+# If we running by kernel building system +ifneq ($(KERNELRELEASE),)

Not sure why you need this

• $(TARGET_MODULE)-objs := hyper_dmabuf_drv.o \

•                             hyper_dmabuf_ioctl.o \
  

•                             hyper_dmabuf_list.o \
  

• 		 hyper_dmabuf_sgl_proc.o \
  

• 		 hyper_dmabuf_ops.o \
  

• 		 hyper_dmabuf_msg.o \
  

• 		 hyper_dmabuf_id.o \
  

+obj-$(CONFIG_HYPER_DMABUF) := $(TARGET_MODULE).o

+# If we are running without kernel build system

Ditto

+else +BUILDSYSTEM_DIR?=../../../ +PWD:=$(shell pwd)

+all : +# run kernel build system to make module +$(MAKE) -C $(BUILDSYSTEM_DIR) M=$(PWD) modules

+clean: +# run kernel build system to cleanup in current directory +$(MAKE) -C $(BUILDSYSTEM_DIR) M=$(PWD) clean

+load:

• insmod ./$(TARGET_MODULE).ko

+unload:

• rmmod ./$(TARGET_MODULE).ko

This seems to be some helper code you use while doing development which needs to be removed

+endif diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_drv.c b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_drv.c new file mode 100644 index 000000000000..18c1cd735ea2 --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_drv.c @@ -0,0 +1,254 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• • Authors:
• • Dongwon Kim dongwon.kim@intel.com
• • Mateusz Polrola mateuszx.potrola@intel.com
• */

+#include <linux/init.h> +#include <linux/module.h> +#include <linux/miscdevice.h> +#include <linux/workqueue.h> +#include <linux/slab.h> +#include <linux/device.h> +#include <linux/uaccess.h> +#include <linux/poll.h> +#include <linux/dma-buf.h> +#include "hyper_dmabuf_drv.h" +#include "hyper_dmabuf_ioctl.h" +#include "hyper_dmabuf_list.h" +#include "hyper_dmabuf_id.h"

+MODULE_LICENSE("GPL and additional rights"); +MODULE_AUTHOR("Intel Corporation");

+struct hyper_dmabuf_private *hy_drv_priv;

instead of using a global symbol here you might want to first allocate misc device and then use devm_kzalloc to allocate your private data

+static void force_free(struct exported_sgt_info *exported,

•        void *attr)
  

+{

• struct ioctl_hyper_dmabuf_unexport unexport_attr;
• struct file *filp = (struct file *)attr;
• if (!filp || !exported)

• return;
  

• if (exported->filp == filp) {

• dev_dbg(hy_drv_priv->dev,
  

• 	"Forcefully releasing buffer {id:%d key:%d %d %d}\n",
  

• 	 exported->hid.id, exported->hid.rng_key[0],
  

• 	 exported->hid.rng_key[1], exported->hid.rng_key[2]);
  

• unexport_attr.hid = exported->hid;
  

• unexport_attr.delay_ms = 0;
  

• hyper_dmabuf_unexport_ioctl(filp, &unexport_attr);
  

• }

+}

+static int hyper_dmabuf_open(struct inode *inode, struct file *filp) +{

• int ret = 0;
• /* Do not allow exclusive open */
• if (filp->f_flags & O_EXCL)

• return -EBUSY;
  

• return ret;

+}

+static int hyper_dmabuf_release(struct inode *inode, struct file *filp) +{

• hyper_dmabuf_foreach_exported(force_free, filp);
• return 0;

+}

+static const struct file_operations hyper_dmabuf_driver_fops = {

• .owner = THIS_MODULE,
• .open = hyper_dmabuf_open,
• .release = hyper_dmabuf_release,
• .unlocked_ioctl = hyper_dmabuf_ioctl,

+};

+static struct miscdevice hyper_dmabuf_miscdev = {

• .minor = MISC_DYNAMIC_MINOR,
• .name = "hyper_dmabuf",

Can this string be a constant through the driver?

• .fops = &hyper_dmabuf_driver_fops,

+};

+static int register_device(void) +{

• int ret = 0;
• ret = misc_register(&hyper_dmabuf_miscdev);
• if (ret) {

• pr_err("hyper_dmabuf: driver can't be registered\n");
  

• return ret;
  

• }
• hy_drv_priv->dev = hyper_dmabuf_miscdev.this_device;
• /* TODO: Check if there is a different way to initialize dma mask */
• dma_coerce_mask_and_coherent(hy_drv_priv->dev, DMA_BIT_MASK(64));
• return ret;

+}

+static void unregister_device(void) +{

• dev_info(hy_drv_priv->dev,

• "hyper_dmabuf: %s is called\n", __func__);
  

• misc_deregister(&hyper_dmabuf_miscdev);

+}

+static int __init hyper_dmabuf_drv_init(void) +{

• int ret = 0;
• pr_notice("hyper_dmabuf_starting: Initialization started\n");
• hy_drv_priv = kcalloc(1, sizeof(struct hyper_dmabuf_private),

• 	      GFP_KERNEL);
  

• if (!hy_drv_priv)

• return -ENOMEM;
  

• ret = register_device();
• if (ret < 0) {

• kfree(hy_drv_priv);
  

• return ret;
  

• }
• hy_drv_priv->bknd_ops = NULL;
• if (hy_drv_priv->bknd_ops == NULL) {

• pr_err("Hyper_dmabuf: no backend found\n");
  

• kfree(hy_drv_priv);
  

• return -1;
  

• }
• mutex_init(&hy_drv_priv->lock);
• mutex_lock(&hy_drv_priv->lock);

Why do you need to immediately lock here?

• hy_drv_priv->initialized = false;

kcalloc allocates zeroed memory, so you might rely on that fact

• dev_info(hy_drv_priv->dev,

•  "initializing database for imported/exported dmabufs\n");
  

• hy_drv_priv->work_queue = create_workqueue("hyper_dmabuf_wqueue");
• ret = hyper_dmabuf_table_init();
• if (ret < 0) {

• dev_err(hy_drv_priv->dev,
  

• 	"fail to init table for exported/imported entries\n");
  

• mutex_unlock(&hy_drv_priv->lock);
  

• kfree(hy_drv_priv);
  

• return ret;
  

• }

+#ifdef CONFIG_HYPER_DMABUF_SYSFS

• ret = hyper_dmabuf_register_sysfs(hy_drv_priv->dev);
• if (ret < 0) {

• dev_err(hy_drv_priv->dev,
  

• 	"failed to initialize sysfs\n");
  

• mutex_unlock(&hy_drv_priv->lock);
  

• kfree(hy_drv_priv);
  

• return ret;
  

• }

+#endif

• if (hy_drv_priv->bknd_ops->init) {

• ret = hy_drv_priv->bknd_ops->init();
  

• if (ret < 0) {
  

• 	dev_dbg(hy_drv_priv->dev,
  

• 		"failed to initialize backend.\n");
  

• 	mutex_unlock(&hy_drv_priv->lock);
  

• 	kfree(hy_drv_priv);
  

unregister sysfs?

• 	return ret;
  

• }
  

• }
• hy_drv_priv->domid = hy_drv_priv->bknd_ops->get_vm_id();

This seems to be a bit inconsistent, e.g. domid vs vm_id

• ret = hy_drv_priv->bknd_ops->init_comm_env();
• if (ret < 0) {

• dev_dbg(hy_drv_priv->dev,
  

• 	"failed to initialize comm-env.\n");
  

bknd_ops->cleanup?

• } else {

• hy_drv_priv->initialized = true;
  

• }
• mutex_unlock(&hy_drv_priv->lock);
• dev_info(hy_drv_priv->dev,

• "Finishing up initialization of hyper_dmabuf drv\n");
  

• /* interrupt for comm should be registered here: */
• return ret;

+}

+static void hyper_dmabuf_drv_exit(void)

__exit?

+{ +#ifdef CONFIG_HYPER_DMABUF_SYSFS

• hyper_dmabuf_unregister_sysfs(hy_drv_priv->dev);

+#endif

• mutex_lock(&hy_drv_priv->lock);
• /* hash tables for export/import entries and ring_infos */
• hyper_dmabuf_table_destroy();
• hy_drv_priv->bknd_ops->destroy_comm();
• if (hy_drv_priv->bknd_ops->cleanup) {

• hy_drv_priv->bknd_ops->cleanup();
  

• };
• /* destroy workqueue */
• if (hy_drv_priv->work_queue)

• destroy_workqueue(hy_drv_priv->work_queue);
  

• /* destroy id_queue */
• if (hy_drv_priv->id_queue)

• hyper_dmabuf_free_hid_list();
  

• mutex_unlock(&hy_drv_priv->lock);
• dev_info(hy_drv_priv->dev,

•  "hyper_dmabuf driver: Exiting\n");
  

• kfree(hy_drv_priv);
• unregister_device();

+}

+module_init(hyper_dmabuf_drv_init); +module_exit(hyper_dmabuf_drv_exit); diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_drv.h b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_drv.h new file mode 100644 index 000000000000..46119d762430 --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_drv.h @@ -0,0 +1,111 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• */

+#ifndef __LINUX_HYPER_DMABUF_DRV_H__ +#define __LINUX_HYPER_DMABUF_DRV_H__

+#include <linux/device.h> +#include <linux/hyper_dmabuf.h>

+struct hyper_dmabuf_req;

+struct hyper_dmabuf_private {

• struct device *dev;
• /* VM(domain) id of current VM instance */
• int domid;
• /* workqueue dedicated to hyper_dmabuf driver */
• struct workqueue_struct *work_queue;
• /* list of reusable hyper_dmabuf_ids */
• struct list_reusable_id *id_queue;
• /* backend ops - hypervisor specific */
• struct hyper_dmabuf_bknd_ops *bknd_ops;
• /* device global lock */
• /* TODO: might need a lock per resource (e.g. EXPORT LIST) */
• struct mutex lock;
• /* flag that shows whether backend is initialized */
• bool initialized;
• /* # of pending events */
• int pending;

+};

+struct list_reusable_id {

• hyper_dmabuf_id_t hid;
• struct list_head list;

+};

+struct hyper_dmabuf_bknd_ops {

• /* backend initialization routine (optional) */
• int (*init)(void);
• /* backend cleanup routine (optional) */
• int (*cleanup)(void);
• /* retreiving id of current virtual machine */
• int (*get_vm_id)(void);
• /* get pages shared via hypervisor-specific method */
• int (*share_pages)(struct page **pages, int vm_id,

• 	   int nents, void **refs_info);
  

• /* make shared pages unshared via hypervisor specific method */
• int (*unshare_pages)(void **refs_info, int nents);
• /* map remotely shared pages on importer's side via

• * hypervisor-specific method
  

• */
  

• struct page ** (*map_shared_pages)(unsigned long ref, int vm_id,

• 			   int nents, void **refs_info);
  

• /* unmap and free shared pages on importer's side via

• * hypervisor-specific method
  

• */
  

• int (*unmap_shared_pages)(void **refs_info, int nents);
• /* initialize communication environment */
• int (*init_comm_env)(void);
• void (*destroy_comm)(void);
• /* upstream ch setup (receiving and responding) */
• int (*init_rx_ch)(int vm_id);
• /* downstream ch setup (transmitting and parsing responses) */
• int (*init_tx_ch)(int vm_id);
• int (*send_req)(int vm_id, struct hyper_dmabuf_req *req, int wait);

+};

+/* exporting global drv private info */ +extern struct hyper_dmabuf_private *hy_drv_priv;

+#endif /* __LINUX_HYPER_DMABUF_DRV_H__ */ diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_id.c b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_id.c new file mode 100644 index 000000000000..f2e994a4957d --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_id.c @@ -0,0 +1,135 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• • Authors:
• • Dongwon Kim dongwon.kim@intel.com
• • Mateusz Polrola mateuszx.potrola@intel.com
• */

+#include <linux/list.h> +#include <linux/slab.h> +#include <linux/random.h> +#include "hyper_dmabuf_drv.h" +#include "hyper_dmabuf_id.h"

Common notes: - I think even if hy_drv_priv is global you shouldn't touch it directly, but pass it as function parameter. - Don't you need to protect reusable list with lock?

+void hyper_dmabuf_store_hid(hyper_dmabuf_id_t hid) +{

• struct list_reusable_id *reusable_head = hy_drv_priv->id_queue;
• struct list_reusable_id *new_reusable;
• new_reusable = kmalloc(sizeof(*new_reusable), GFP_KERNEL);
• if (!new_reusable)

• return;
  

• new_reusable->hid = hid;
• list_add(&new_reusable->list, &reusable_head->list);

+}

+static hyper_dmabuf_id_t get_reusable_hid(void) +{

• struct list_reusable_id *reusable_head = hy_drv_priv->id_queue;
• hyper_dmabuf_id_t hid = {-1, {0, 0, 0} };
• /* check there is reusable id */
• if (!list_empty(&reusable_head->list)) {

• reusable_head = list_first_entry(&reusable_head->list,
  

• 				 struct list_reusable_id,
  

• 				 list);
  

• list_del(&reusable_head->list);
  

• hid = reusable_head->hid;
  

• kfree(reusable_head);
  

• }
• return hid;

+}

+void hyper_dmabuf_free_hid_list(void) +{

• struct list_reusable_id *reusable_head = hy_drv_priv->id_queue;
• struct list_reusable_id *temp_head;
• if (reusable_head) {

• /* freeing mem space all reusable ids in the stack */
  

• while (!list_empty(&reusable_head->list)) {
  

• 	temp_head = list_first_entry(&reusable_head->list,
  

• 				     struct list_reusable_id,
  

• 				     list);
  

• 	list_del(&temp_head->list);
  

• 	kfree(temp_head);
  

• }
  

• /* freeing head */
  

• kfree(reusable_head);
  

• }

+}

+hyper_dmabuf_id_t hyper_dmabuf_get_hid(void) +{

• static int count;

could you please explicitly initialize this?

• hyper_dmabuf_id_t hid;
• struct list_reusable_id *reusable_head;
• /* first call to hyper_dmabuf_get_id */
• if (count == 0) {

• reusable_head = kmalloc(sizeof(*reusable_head), GFP_KERNEL);
  

• if (!reusable_head)
  

• 	return (hyper_dmabuf_id_t){-1, {0, 0, 0} };
  

• /* list head has an invalid count */
  

• reusable_head->hid.id = -1;
  

• INIT_LIST_HEAD(&reusable_head->list);
  

• hy_drv_priv->id_queue = reusable_head;
  

• }
• hid = get_reusable_hid();
• /*creating a new H-ID only if nothing in the reusable id queue

start the comment from a new line

• * and count is less than maximum allowed
  

• */
  

• if (hid.id == -1 && count < HYPER_DMABUF_ID_MAX)

• hid.id = HYPER_DMABUF_ID_CREATE(hy_drv_priv->domid, count++);
  

• /* random data embedded in the id for security */
• get_random_bytes(&hid.rng_key[0], 12);

can magic 12 be a defined constant?

• return hid;

+}

+bool hyper_dmabuf_hid_keycomp(hyper_dmabuf_id_t hid1, hyper_dmabuf_id_t hid2) +{

• int i;
• /* compare keys */
• for (i = 0; i < 3; i++) {

can magic 3 be defined as a constant please?

• if (hid1.rng_key[i] != hid2.rng_key[i])
  

• 	return false;
  

• }
• return true;

+} diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_id.h b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_id.h new file mode 100644 index 000000000000..11f530e2c8f6 --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_id.h @@ -0,0 +1,53 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• */

+#ifndef __HYPER_DMABUF_ID_H__ +#define __HYPER_DMABUF_ID_H__

+#define HYPER_DMABUF_ID_CREATE(domid, cnt) \

• ((((domid) & 0xFF) << 24) | ((cnt) & 0xFFFFFF))

I would define hyper_dmabuf_id_t.id as a union or 2 separate fields to avoid his magic

+#define HYPER_DMABUF_DOM_ID(hid) \

• (((hid.id) >> 24) & 0xFF)

+/* currently maximum number of buffers shared

• • at any given moment is limited to 1000
• */

+#define HYPER_DMABUF_ID_MAX 1000

Why 1000? Is it just to limit or is dictated by some use-cases/experiments?

+/* adding freed hid to the reusable list */ +void hyper_dmabuf_store_hid(hyper_dmabuf_id_t hid);

+/* freeing the reusasble list */ +void hyper_dmabuf_free_hid_list(void);

+/* getting a hid available to use. */ +hyper_dmabuf_id_t hyper_dmabuf_get_hid(void);

+/* comparing two different hid */ +bool hyper_dmabuf_hid_keycomp(hyper_dmabuf_id_t hid1, hyper_dmabuf_id_t hid2);

+#endif /*__HYPER_DMABUF_ID_H*/ diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ioctl.c b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ioctl.c new file mode 100644 index 000000000000..020a5590a254 --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ioctl.c @@ -0,0 +1,672 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• • Authors:
• • Dongwon Kim dongwon.kim@intel.com
• • Mateusz Polrola mateuszx.potrola@intel.com
• */

+#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/uaccess.h> +#include <linux/dma-buf.h> +#include "hyper_dmabuf_drv.h" +#include "hyper_dmabuf_id.h" +#include "hyper_dmabuf_struct.h" +#include "hyper_dmabuf_ioctl.h" +#include "hyper_dmabuf_list.h" +#include "hyper_dmabuf_msg.h" +#include "hyper_dmabuf_sgl_proc.h" +#include "hyper_dmabuf_ops.h"

Here and below: please do not touch global hy_drv_priv

+static int hyper_dmabuf_tx_ch_setup_ioctl(struct file *filp, void *data) +{

• struct ioctl_hyper_dmabuf_tx_ch_setup *tx_ch_attr;
• struct hyper_dmabuf_bknd_ops *bknd_ops = hy_drv_priv->bknd_ops;
• int ret = 0;
• if (!data) {

• dev_err(hy_drv_priv->dev, "user data is NULL\n");
  

• return -EINVAL;
  

• }
• tx_ch_attr = (struct ioctl_hyper_dmabuf_tx_ch_setup *)data;
• ret = bknd_ops->init_tx_ch(tx_ch_attr->remote_domain);
• return ret;

+}

+static int hyper_dmabuf_rx_ch_setup_ioctl(struct file *filp, void *data) +{

• struct ioctl_hyper_dmabuf_rx_ch_setup *rx_ch_attr;
• struct hyper_dmabuf_bknd_ops *bknd_ops = hy_drv_priv->bknd_ops;
• int ret = 0;
• if (!data) {

• dev_err(hy_drv_priv->dev, "user data is NULL\n");
  

• return -EINVAL;
  

• }
• rx_ch_attr = (struct ioctl_hyper_dmabuf_rx_ch_setup *)data;
• ret = bknd_ops->init_rx_ch(rx_ch_attr->source_domain);
• return ret;

+}

+static int send_export_msg(struct exported_sgt_info *exported,

• 	   struct pages_info *pg_info)
  

+{

• struct hyper_dmabuf_bknd_ops *bknd_ops = hy_drv_priv->bknd_ops;
• struct hyper_dmabuf_req *req;
• int op[MAX_NUMBER_OF_OPERANDS] = {0};
• int ret, i;
• /* now create request for importer via ring */
• op[0] = exported->hid.id;
• for (i = 0; i < 3; i++)

• op[i+1] = exported->hid.rng_key[i];
  

• if (pg_info) {

heh, can we have a well defined structures for requests/responses, so we don't have to put all these magics?

• op[4] = pg_info->nents;
  

• op[5] = pg_info->frst_ofst;
  

• op[6] = pg_info->last_len;
  

• op[7] = bknd_ops->share_pages(pg_info->pgs, exported->rdomid,
  

• 			 pg_info->nents, &exported->refs_info);
  

ret?

• if (op[7] < 0) {
  

• 	dev_err(hy_drv_priv->dev, "pages sharing failed\n");
  

• 	return op[7];
  

• }
  

• }
• req = kcalloc(1, sizeof(*req), GFP_KERNEL);
• if (!req)

• return -ENOMEM;
  

• /* composing a message to the importer */
• hyper_dmabuf_create_req(req, HYPER_DMABUF_EXPORT, &op[0]);
• ret = bknd_ops->send_req(exported->rdomid, req, true);

can we allocate req on stack? and don't use kcalloc?

• kfree(req);
• return ret;

+}

+/* Fast path exporting routine in case same buffer is already exported.

• • If same buffer is still valid and exist in EXPORT LIST it returns 0 so
• • that remaining normal export process can be skipped.
• • If "unexport" is scheduled for the buffer, it cancels it since the buffer
• • is being re-exported.
• • return '1' if reexport is needed, return '0' if succeeds, return
• • Kernel error code if something goes wrong
• */

+static int fastpath_export(hyper_dmabuf_id_t hid) +{

• int reexport = 1;
• int ret = 0;

why do you need these two variables?

• struct exported_sgt_info *exported;
• exported = hyper_dmabuf_find_exported(hid);
• if (!exported)

• return reexport;
  

• if (exported->valid == false)

• return reexport;
  

• /*

• * Check if unexport is already scheduled for that buffer,
  

• * if so try to cancel it. If that will fail, buffer needs
  

• * to be reexport once again.
  

• */
  

• if (exported->unexport_sched) {

• if (!cancel_delayed_work_sync(&exported->unexport))
  

• 	return reexport;
  

• exported->unexport_sched = false;
  

• }
• return ret;

+}

+static int hyper_dmabuf_export_remote_ioctl(struct file *filp, void *data) +{

• struct ioctl_hyper_dmabuf_export_remote *export_remote_attr =

• 	(struct ioctl_hyper_dmabuf_export_remote *)data;
  

• struct dma_buf *dma_buf;
• struct dma_buf_attachment *attachment;
• struct sg_table *sgt;
• struct pages_info *pg_info;
• struct exported_sgt_info *exported;
• hyper_dmabuf_id_t hid;
• int ret = 0;
• if (hy_drv_priv->domid == export_remote_attr->remote_domain) {

• dev_err(hy_drv_priv->dev,
  

• 	"exporting to the same VM is not permitted\n");
  

• return -EINVAL;
  

• }
• dma_buf = dma_buf_get(export_remote_attr->dmabuf_fd);
• if (IS_ERR(dma_buf)) {

• dev_err(hy_drv_priv->dev, "Cannot get dma buf\n");
  

• return PTR_ERR(dma_buf);
  

• }
• /* we check if this specific attachment was already exported

• * to the same domain and if yes and it's valid sgt_info,
  

• * it returns hyper_dmabuf_id of pre-exported sgt_info
  

• */
  

• hid = hyper_dmabuf_find_hid_exported(dma_buf,

• 			     export_remote_attr->remote_domain);
  

• if (hid.id != -1) {

• ret = fastpath_export(hid);
  

• /* return if fastpath_export succeeds or
  

•  * gets some fatal error
  

•  */
  

• if (ret <= 0) {
  

• 	dma_buf_put(dma_buf);
  

• 	export_remote_attr->hid = hid;
  

• 	return ret;
  

• }
  

• }
• attachment = dma_buf_attach(dma_buf, hy_drv_priv->dev);
• if (IS_ERR(attachment)) {

• dev_err(hy_drv_priv->dev, "cannot get attachment\n");
  

• ret = PTR_ERR(attachment);
  

here and below - if you have dma-buf from fastpath don't you need to release/handle it on error path here? E.g. fastpath may have canceled unexport work for this buffer

• goto fail_attach;
  

• }
• sgt = dma_buf_map_attachment(attachment, DMA_BIDIRECTIONAL);
• if (IS_ERR(sgt)) {

• dev_err(hy_drv_priv->dev, "cannot map attachment\n");
  

• ret = PTR_ERR(sgt);
  

• goto fail_map_attachment;
  

• }
• exported = kcalloc(1, sizeof(*exported), GFP_KERNEL);
• if (!exported) {

• ret = -ENOMEM;
  

• goto fail_sgt_info_creation;
  

• }
• exported->hid = hyper_dmabuf_get_hid();
• /* no more exported dmabuf allowed */
• if (exported->hid.id == -1) {

• dev_err(hy_drv_priv->dev,
  

• 	"exceeds allowed number of dmabuf to be exported\n");
  

• ret = -ENOMEM;
  

• goto fail_sgt_info_creation;
  

• }
• exported->rdomid = export_remote_attr->remote_domain;
• exported->dma_buf = dma_buf;
• exported->valid = true;
• exported->active_sgts = kmalloc(sizeof(struct sgt_list), GFP_KERNEL);
• if (!exported->active_sgts) {

• ret = -ENOMEM;
  

• goto fail_map_active_sgts;
  

• }
• exported->active_attached = kmalloc(sizeof(struct attachment_list),

• 			    GFP_KERNEL);
  

• if (!exported->active_attached) {

• ret = -ENOMEM;
  

• goto fail_map_active_attached;
  

• }
• exported->va_kmapped = kmalloc(sizeof(struct kmap_vaddr_list),

• 		       GFP_KERNEL);
  

• if (!exported->va_kmapped) {

• ret = -ENOMEM;
  

• goto fail_map_va_kmapped;
  

• }
• exported->va_vmapped = kmalloc(sizeof(struct vmap_vaddr_list),

• 		       GFP_KERNEL);
  

• if (!exported->va_vmapped) {

• ret = -ENOMEM;
  

• goto fail_map_va_vmapped;
  

• }
• exported->active_sgts->sgt = sgt;
• exported->active_attached->attach = attachment;
• exported->va_kmapped->vaddr = NULL;
• exported->va_vmapped->vaddr = NULL;
• /* initialize list of sgt, attachment and vaddr for dmabuf sync

• * via shadow dma-buf
  

• */
  

• INIT_LIST_HEAD(&exported->active_sgts->list);
• INIT_LIST_HEAD(&exported->active_attached->list);
• INIT_LIST_HEAD(&exported->va_kmapped->list);
• INIT_LIST_HEAD(&exported->va_vmapped->list);
• if (ret) {

• dev_err(hy_drv_priv->dev,
  

• 	"failed to load private data\n");
  

• ret = -EINVAL;
  

• goto fail_export;
  

• }
• pg_info = hyper_dmabuf_ext_pgs(sgt);
• if (!pg_info) {

• dev_err(hy_drv_priv->dev,
  

• 	"failed to construct pg_info\n");
  

• ret = -ENOMEM;
  

• goto fail_export;
  

• }
• exported->nents = pg_info->nents;
• /* now register it to export list */
• hyper_dmabuf_register_exported(exported);
• export_remote_attr->hid = exported->hid;
• ret = send_export_msg(exported, pg_info);
• if (ret < 0) {

• dev_err(hy_drv_priv->dev,
  

• 	"failed to send out the export request\n");
  

• goto fail_send_request;
  

• }
• /* free pg_info */
• kfree(pg_info->pgs);
• kfree(pg_info);
• exported->filp = filp;
• return ret;

+/* Clean-up if error occurs */

+fail_send_request:

• hyper_dmabuf_remove_exported(exported->hid);
• /* free pg_info */
• kfree(pg_info->pgs);
• kfree(pg_info);

+fail_export:

• kfree(exported->va_vmapped);

+fail_map_va_vmapped:

• kfree(exported->va_kmapped);

+fail_map_va_kmapped:

• kfree(exported->active_attached);

+fail_map_active_attached:

• kfree(exported->active_sgts);
• kfree(exported);

+fail_map_active_sgts: +fail_sgt_info_creation:

• dma_buf_unmap_attachment(attachment, sgt,

• 		 DMA_BIDIRECTIONAL);
  

+fail_map_attachment:

• dma_buf_detach(dma_buf, attachment);

+fail_attach:

• dma_buf_put(dma_buf);
• return ret;

+}

+static int hyper_dmabuf_export_fd_ioctl(struct file *filp, void *data) +{

• struct ioctl_hyper_dmabuf_export_fd *export_fd_attr =

• 	(struct ioctl_hyper_dmabuf_export_fd *)data;
  

• struct hyper_dmabuf_bknd_ops *bknd_ops = hy_drv_priv->bknd_ops;
• struct imported_sgt_info *imported;
• struct hyper_dmabuf_req *req;
• struct page **data_pgs;
• int op[4];

don't you have hyper_dmabuf_id_t for that?

• int i;
• int ret = 0;
• dev_dbg(hy_drv_priv->dev, "%s entry\n", __func__);
• /* look for dmabuf for the id */
• imported = hyper_dmabuf_find_imported(export_fd_attr->hid);
• /* can't find sgt from the table */
• if (!imported) {

• dev_err(hy_drv_priv->dev, "can't find the entry\n");
  

• return -ENOENT;
  

• }
• mutex_lock(&hy_drv_priv->lock);
• imported->importers++;
• /* send notification for export_fd to exporter */
• op[0] = imported->hid.id;
• for (i = 0; i < 3; i++)

• op[i+1] = imported->hid.rng_key[i];
  

• dev_dbg(hy_drv_priv->dev, "Export FD of buffer {id:%d key:%d %d %d}\n",

• imported->hid.id, imported->hid.rng_key[0],
  

• imported->hid.rng_key[1], imported->hid.rng_key[2]);
  

• req = kcalloc(1, sizeof(*req), GFP_KERNEL);

can you have req allocated on stack?

• if (!req) {

• mutex_unlock(&hy_drv_priv->lock);
  

• return -ENOMEM;
  

• }
• hyper_dmabuf_create_req(req, HYPER_DMABUF_EXPORT_FD, &op[0]);
• ret = bknd_ops->send_req(HYPER_DMABUF_DOM_ID(imported->hid), req, true);
• if (ret < 0) {

• /* in case of timeout other end eventually will receive request,
  

•  * so we need to undo it
  

•  */
  

and what if there is a race condition? at the time you delete the buffer the corresponding response comes in?

• hyper_dmabuf_create_req(req, HYPER_DMABUF_EXPORT_FD_FAILED,
  

• 			&op[0]);
  

• bknd_ops->send_req(HYPER_DMABUF_DOM_ID(imported->hid),
  

• 		   req, false);
  

• kfree(req);
  

• dev_err(hy_drv_priv->dev,
  

• 	"Failed to create sgt or notify exporter\n");
  

• imported->importers--;
  

• mutex_unlock(&hy_drv_priv->lock);
  

• return ret;
  

• }
• kfree(req);
• if (ret == HYPER_DMABUF_REQ_ERROR) {

• dev_err(hy_drv_priv->dev,
  

• 	"Buffer invalid {id:%d key:%d %d %d}, cannot import\n",
  

• 	imported->hid.id, imported->hid.rng_key[0],
  

• 	imported->hid.rng_key[1], imported->hid.rng_key[2]);
  

• imported->importers--;
  

• mutex_unlock(&hy_drv_priv->lock);
  

• return -EINVAL;
  

• }
• ret = 0;
• dev_dbg(hy_drv_priv->dev,

• "Found buffer gref %d off %d\n",
  

• imported->ref_handle, imported->frst_ofst);
  

• dev_dbg(hy_drv_priv->dev,

• "last len %d nents %d domain %d\n",
  

• imported->last_len, imported->nents,
  

• HYPER_DMABUF_DOM_ID(imported->hid));
  

• if (!imported->sgt) {

• dev_dbg(hy_drv_priv->dev,
  

• 	"buffer {id:%d key:%d %d %d} pages not mapped yet\n",
  

• 	imported->hid.id, imported->hid.rng_key[0],
  

• 	imported->hid.rng_key[1], imported->hid.rng_key[2]);
  

• data_pgs = bknd_ops->map_shared_pages(imported->ref_handle,
  

• 			HYPER_DMABUF_DOM_ID(imported->hid),
  

• 			imported->nents,
  

• 			&imported->refs_info);
  

• if (!data_pgs) {
  

• 	dev_err(hy_drv_priv->dev,
  

• 		"can't map pages hid {id:%d key:%d %d %d}\n",
  

• 		imported->hid.id, imported->hid.rng_key[0],
  

• 		imported->hid.rng_key[1],
  

• 		imported->hid.rng_key[2]);
  

• 	imported->importers--;
  

• 	req = kcalloc(1, sizeof(*req), GFP_KERNEL);
  

• 	if (!req) {
  

• 		mutex_unlock(&hy_drv_priv->lock);
  

• 		return -ENOMEM;
  

• 	}
  

• 	hyper_dmabuf_create_req(req,
  

• 				HYPER_DMABUF_EXPORT_FD_FAILED,
  

• 				&op[0]);
  

• 	bknd_ops->send_req(HYPER_DMABUF_DOM_ID(imported->hid),
  

• 			   req, false);
  

• 	kfree(req);
  

• 	mutex_unlock(&hy_drv_priv->lock);
  

• 	return -EINVAL;
  

• }
  

• imported->sgt = hyper_dmabuf_create_sgt(data_pgs,
  

• 					imported->frst_ofst,
  

• 					imported->last_len,
  

• 					imported->nents);
  

• }
• export_fd_attr->fd = hyper_dmabuf_export_fd(imported,

• 				    export_fd_attr->flags);
  

• if (export_fd_attr->fd < 0) {

• /* fail to get fd */
  

• ret = export_fd_attr->fd;
  

why don't you send HYPER_DMABUF_EXPORT_FD_FAILED in this case?

• }
• mutex_unlock(&hy_drv_priv->lock);
• dev_dbg(hy_drv_priv->dev, "%s exit\n", __func__);
• return ret;

+}

+/* unexport dmabuf from the database and send int req to the source domain

• • to unmap it.
• */

+static void delayed_unexport(struct work_struct *work) +{

• struct hyper_dmabuf_req *req;
• struct hyper_dmabuf_bknd_ops *bknd_ops = hy_drv_priv->bknd_ops;
• struct exported_sgt_info *exported =

• container_of(work, struct exported_sgt_info, unexport.work);
  

• int op[4];

use the struct defined for this

• int i, ret;
• if (!exported)

• return;
  

• dev_dbg(hy_drv_priv->dev,

• "Marking buffer {id:%d key:%d %d %d} as invalid\n",
  

• exported->hid.id, exported->hid.rng_key[0],
  

• exported->hid.rng_key[1], exported->hid.rng_key[2]);
  

• /* no longer valid */
• exported->valid = false;
• req = kcalloc(1, sizeof(*req), GFP_KERNEL);
• if (!req)

will we leak the buffer because we return here?

• return;
  

• op[0] = exported->hid.id;
• for (i = 0; i < 3; i++)

• op[i+1] = exported->hid.rng_key[i];
  

• hyper_dmabuf_create_req(req, HYPER_DMABUF_NOTIFY_UNEXPORT, &op[0]);
• /* Now send unexport request to remote domain, marking

• * that buffer should not be used anymore
  

• */
  

• ret = bknd_ops->send_req(exported->rdomid, req, true);
• if (ret < 0) {

• dev_err(hy_drv_priv->dev,
  

• 	"unexport message for buffer {id:%d key:%d %d %d} failed\n",
  

• 	exported->hid.id, exported->hid.rng_key[0],
  

• 	exported->hid.rng_key[1], exported->hid.rng_key[2]);
  

• }
• kfree(req);
• exported->unexport_sched = false;
• /* Immediately clean-up if it has never been exported by importer

• * (so no SGT is constructed on importer).
  

• * clean it up later in remote sync when final release ops
  

• * is called (importer does this only when there's no
  

• * no consumer of locally exported FDs)
  

• */
  

• if (exported->active == 0) {

• dev_dbg(hy_drv_priv->dev,
  

• 	"claning up buffer {id:%d key:%d %d %d} completly\n",
  

• 	exported->hid.id, exported->hid.rng_key[0],
  

• 	exported->hid.rng_key[1], exported->hid.rng_key[2]);
  

• hyper_dmabuf_cleanup_sgt_info(exported, false);
  

• hyper_dmabuf_remove_exported(exported->hid);
  

• /* register hyper_dmabuf_id to the list for reuse */
  

• hyper_dmabuf_store_hid(exported->hid);
  

• kfree(exported);
  

• }

+}

+/* Schedule unexport of dmabuf.

• */

+int hyper_dmabuf_unexport_ioctl(struct file *filp, void *data) +{

• struct ioctl_hyper_dmabuf_unexport *unexport_attr =

• 	(struct ioctl_hyper_dmabuf_unexport *)data;
  

• struct exported_sgt_info *exported;
• dev_dbg(hy_drv_priv->dev, "%s entry\n", __func__);
• /* find dmabuf in export list */
• exported = hyper_dmabuf_find_exported(unexport_attr->hid);
• dev_dbg(hy_drv_priv->dev,

• "scheduling unexport of buffer {id:%d key:%d %d %d}\n",
  

• unexport_attr->hid.id, unexport_attr->hid.rng_key[0],
  

• unexport_attr->hid.rng_key[1], unexport_attr->hid.rng_key[2]);
  

• /* failed to find corresponding entry in export list */
• if (exported == NULL) {

• unexport_attr->status = -ENOENT;
  

• return -ENOENT;
  

• }
• if (exported->unexport_sched)

• return 0;
  

• exported->unexport_sched = true;
• INIT_DELAYED_WORK(&exported->unexport, delayed_unexport);

why can't you just wait for the buffer to be unexported?

• schedule_delayed_work(&exported->unexport,

• 	      msecs_to_jiffies(unexport_attr->delay_ms));
  

• dev_dbg(hy_drv_priv->dev, "%s exit\n", __func__);
• return 0;

+}

+const struct hyper_dmabuf_ioctl_desc hyper_dmabuf_ioctls[] = {

• HYPER_DMABUF_IOCTL_DEF(IOCTL_HYPER_DMABUF_TX_CH_SETUP,

• 	       hyper_dmabuf_tx_ch_setup_ioctl, 0),
  

• HYPER_DMABUF_IOCTL_DEF(IOCTL_HYPER_DMABUF_RX_CH_SETUP,

• 	       hyper_dmabuf_rx_ch_setup_ioctl, 0),
  

• HYPER_DMABUF_IOCTL_DEF(IOCTL_HYPER_DMABUF_EXPORT_REMOTE,

• 	       hyper_dmabuf_export_remote_ioctl, 0),
  

• HYPER_DMABUF_IOCTL_DEF(IOCTL_HYPER_DMABUF_EXPORT_FD,

• 	       hyper_dmabuf_export_fd_ioctl, 0),
  

• HYPER_DMABUF_IOCTL_DEF(IOCTL_HYPER_DMABUF_UNEXPORT,

• 	       hyper_dmabuf_unexport_ioctl, 0),
  

+};

+long hyper_dmabuf_ioctl(struct file *filp,

• 	unsigned int cmd, unsigned long param)
  

+{

• const struct hyper_dmabuf_ioctl_desc *ioctl = NULL;
• unsigned int nr = _IOC_NR(cmd);
• int ret;
• hyper_dmabuf_ioctl_t func;
• char *kdata;
• if (nr > ARRAY_SIZE(hyper_dmabuf_ioctls)) {

• dev_err(hy_drv_priv->dev, "invalid ioctl\n");
  

• return -EINVAL;
  

• }
• ioctl = &hyper_dmabuf_ioctls[nr];
• func = ioctl->func;
• if (unlikely(!func)) {

• dev_err(hy_drv_priv->dev, "no function\n");
  

• return -EINVAL;
  

• }
• kdata = kmalloc(_IOC_SIZE(cmd), GFP_KERNEL);
• if (!kdata)

• return -ENOMEM;
  

• if (copy_from_user(kdata, (void __user *)param,

• 	   _IOC_SIZE(cmd)) != 0) {
  

• dev_err(hy_drv_priv->dev,
  

• 	"failed to copy from user arguments\n");
  

• ret = -EFAULT;
  

• goto ioctl_error;
  

• }
• ret = func(filp, kdata);
• if (copy_to_user((void __user *)param, kdata,

• 	 _IOC_SIZE(cmd)) != 0) {
  

• dev_err(hy_drv_priv->dev,
  

• 	"failed to copy to user arguments\n");
  

• ret = -EFAULT;
  

• goto ioctl_error;
  

• }

+ioctl_error:

• kfree(kdata);
• return ret;

+} diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ioctl.h b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ioctl.h new file mode 100644 index 000000000000..d8090900ffa2 --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ioctl.h @@ -0,0 +1,52 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• */

+#ifndef __HYPER_DMABUF_IOCTL_H__ +#define __HYPER_DMABUF_IOCTL_H__

+typedef int (*hyper_dmabuf_ioctl_t)(struct file *filp, void *data);

+struct hyper_dmabuf_ioctl_desc {

• unsigned int cmd;
• int flags;
• hyper_dmabuf_ioctl_t func;
• const char *name;

+};

+#define HYPER_DMABUF_IOCTL_DEF(ioctl, _func, _flags) \

• [_IOC_NR(ioctl)] = { \

• 	.cmd = ioctl,			\
  

• 	.func = _func,			\
  

• 	.flags = _flags,		\
  

• 	.name = #ioctl			\
  

• }

+long hyper_dmabuf_ioctl(struct file *filp,

• 	unsigned int cmd, unsigned long param);
  

+int hyper_dmabuf_unexport_ioctl(struct file *filp, void *data);

+#endif //__HYPER_DMABUF_IOCTL_H__ diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_list.c b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_list.c new file mode 100644 index 000000000000..f2f65a8ec47f --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_list.c @@ -0,0 +1,294 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• • Authors:
• • Dongwon Kim dongwon.kim@intel.com
• • Mateusz Polrola mateuszx.potrola@intel.com
• */

+#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/cdev.h> +#include <linux/hashtable.h> +#include "hyper_dmabuf_drv.h" +#include "hyper_dmabuf_list.h" +#include "hyper_dmabuf_id.h"

+DECLARE_HASHTABLE(hyper_dmabuf_hash_imported, MAX_ENTRY_IMPORTED); +DECLARE_HASHTABLE(hyper_dmabuf_hash_exported, MAX_ENTRY_EXPORTED);

+#ifdef CONFIG_HYPER_DMABUF_SYSFS +static ssize_t hyper_dmabuf_imported_show(struct device *drv,

• 			  struct device_attribute *attr,
  

• 			  char *buf)
  

+{

• struct list_entry_imported *info_entry;
• int bkt;
• ssize_t count = 0;
• size_t total = 0;
• hash_for_each(hyper_dmabuf_hash_imported, bkt, info_entry, node) {

• hyper_dmabuf_id_t hid = info_entry->imported->hid;
  

• int nents = info_entry->imported->nents;
  

• bool valid = info_entry->imported->valid;
  

• int num_importers = info_entry->imported->importers;
  

• total += nents;
  

• count += scnprintf(buf + count, PAGE_SIZE - count,
  

• 		"hid:{%d %d %d %d}, nent:%d, v:%c, numi:%d\n",
  

• 		hid.id, hid.rng_key[0], hid.rng_key[1],
  

• 		hid.rng_key[2], nents, (valid ? 't' : 'f'),
  

• 		num_importers);
  

• }
• count += scnprintf(buf + count, PAGE_SIZE - count,

• 	   "total nents: %lu\n", total);
  

• return count;

+}

+static ssize_t hyper_dmabuf_exported_show(struct device *drv,

• 			  struct device_attribute *attr,
  

• 			  char *buf)
  

+{

• struct list_entry_exported *info_entry;
• int bkt;
• ssize_t count = 0;
• size_t total = 0;
• hash_for_each(hyper_dmabuf_hash_exported, bkt, info_entry, node) {

• hyper_dmabuf_id_t hid = info_entry->exported->hid;
  

• int nents = info_entry->exported->nents;
  

• bool valid = info_entry->exported->valid;
  

• int importer_exported = info_entry->exported->active;
  

• total += nents;
  

• count += scnprintf(buf + count, PAGE_SIZE - count,
  

• 		   "hid:{%d %d %d %d}, nent:%d, v:%c, ie:%d\n",
  

• 		   hid.id, hid.rng_key[0], hid.rng_key[1],
  

• 		   hid.rng_key[2], nents, (valid ? 't' : 'f'),
  

• 		   importer_exported);
  

• }
• count += scnprintf(buf + count, PAGE_SIZE - count,

• 	   "total nents: %lu\n", total);
  

• return count;

+}

+static DEVICE_ATTR(imported, 0400, hyper_dmabuf_imported_show, NULL); +static DEVICE_ATTR(exported, 0400, hyper_dmabuf_exported_show, NULL);

+int hyper_dmabuf_register_sysfs(struct device *dev) +{

• int err;
• err = device_create_file(dev, &dev_attr_imported);
• if (err < 0)

• goto err1;
  

• err = device_create_file(dev, &dev_attr_exported);
• if (err < 0)

• goto err2;
  

• return 0;

+err2:

• device_remove_file(dev, &dev_attr_imported);

+err1:

• return -1;

+}

+int hyper_dmabuf_unregister_sysfs(struct device *dev) +{

• device_remove_file(dev, &dev_attr_imported);
• device_remove_file(dev, &dev_attr_exported);
• return 0;

+}

+#endif

+int hyper_dmabuf_table_init(void) +{

• hash_init(hyper_dmabuf_hash_imported);
• hash_init(hyper_dmabuf_hash_exported);
• return 0;

+}

+int hyper_dmabuf_table_destroy(void) +{

• /* TODO: cleanup hyper_dmabuf_hash_imported

• * and hyper_dmabuf_hash_exported
  

• */
  

• return 0;

+}

+int hyper_dmabuf_register_exported(struct exported_sgt_info *exported) +{

• struct list_entry_exported *info_entry;
• info_entry = kmalloc(sizeof(*info_entry), GFP_KERNEL);
• if (!info_entry)

• return -ENOMEM;
  

• info_entry->exported = exported;
• hash_add(hyper_dmabuf_hash_exported, &info_entry->node,

•  info_entry->exported->hid.id);
  

• return 0;

+}

+int hyper_dmabuf_register_imported(struct imported_sgt_info *imported) +{

• struct list_entry_imported *info_entry;
• info_entry = kmalloc(sizeof(*info_entry), GFP_KERNEL);
• if (!info_entry)

• return -ENOMEM;
  

• info_entry->imported = imported;
• hash_add(hyper_dmabuf_hash_imported, &info_entry->node,

•  info_entry->imported->hid.id);
  

• return 0;

+}

+struct exported_sgt_info *hyper_dmabuf_find_exported(hyper_dmabuf_id_t hid) +{

• struct list_entry_exported *info_entry;
• int bkt;
• hash_for_each(hyper_dmabuf_hash_exported, bkt, info_entry, node)

• /* checking hid.id first */
  

• if (info_entry->exported->hid.id == hid.id) {
  

• 	/* then key is compared */
  

• 	if (hyper_dmabuf_hid_keycomp(info_entry->exported->hid,
  

• 				    hid))
  

• 		return info_entry->exported;
  

• 	/* if key is unmatched, given HID is invalid,
  

• 	 * so returning NULL
  

• 	 */
  

• 	break;
  

• }
  

• return NULL;

+}

+/* search for pre-exported sgt and return id of it if it exist */ +hyper_dmabuf_id_t hyper_dmabuf_find_hid_exported(struct dma_buf *dmabuf,

• 				 int domid)
  

+{

• struct list_entry_exported *info_entry;
• hyper_dmabuf_id_t hid = {-1, {0, 0, 0} };
• int bkt;
• hash_for_each(hyper_dmabuf_hash_exported, bkt, info_entry, node)

• if (info_entry->exported->dma_buf == dmabuf &&
  

•     info_entry->exported->rdomid == domid)
  

• 	return info_entry->exported->hid;
  

• return hid;

+}

+struct imported_sgt_info *hyper_dmabuf_find_imported(hyper_dmabuf_id_t hid) +{

• struct list_entry_imported *info_entry;
• int bkt;
• hash_for_each(hyper_dmabuf_hash_imported, bkt, info_entry, node)

• /* checking hid.id first */
  

• if (info_entry->imported->hid.id == hid.id) {
  

• 	/* then key is compared */
  

• 	if (hyper_dmabuf_hid_keycomp(info_entry->imported->hid,
  

• 				    hid))
  

• 		return info_entry->imported;
  

• 	/* if key is unmatched, given HID is invalid,
  

• 	 * so returning NULL
  

• 	 */
  

• 	break;
  

• }
  

• return NULL;

+}

+int hyper_dmabuf_remove_exported(hyper_dmabuf_id_t hid) +{

• struct list_entry_exported *info_entry;
• int bkt;
• hash_for_each(hyper_dmabuf_hash_exported, bkt, info_entry, node)

• /* checking hid.id first */
  

• if (info_entry->exported->hid.id == hid.id) {
  

• 	/* then key is compared */
  

• 	if (hyper_dmabuf_hid_keycomp(info_entry->exported->hid,
  

• 				    hid)) {
  

• 		hash_del(&info_entry->node);
  

• 		kfree(info_entry);
  

• 		return 0;
  

• 	}
  

• 	break;
  

• }
  

• return -ENOENT;

+}

+int hyper_dmabuf_remove_imported(hyper_dmabuf_id_t hid) +{

• struct list_entry_imported *info_entry;
• int bkt;
• hash_for_each(hyper_dmabuf_hash_imported, bkt, info_entry, node)

• /* checking hid.id first */
  

• if (info_entry->imported->hid.id == hid.id) {
  

• 	/* then key is compared */
  

• 	if (hyper_dmabuf_hid_keycomp(info_entry->imported->hid,
  

• 				    hid)) {
  

• 		hash_del(&info_entry->node);
  

• 		kfree(info_entry);
  

• 		return 0;
  

• 	}
  

• 	break;
  

• }
  

• return -ENOENT;

+}

+void hyper_dmabuf_foreach_exported(

• void (*func)(struct exported_sgt_info *, void *attr),
• void *attr)

+{

• struct list_entry_exported *info_entry;
• struct hlist_node *tmp;
• int bkt;
• hash_for_each_safe(hyper_dmabuf_hash_exported, bkt, tmp,

• 	info_entry, node) {
  

• func(info_entry->exported, attr);
  

• }

+} diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_list.h b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_list.h new file mode 100644 index 000000000000..3c6a23ef80c6 --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_list.h @@ -0,0 +1,73 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• */

+#ifndef __HYPER_DMABUF_LIST_H__ +#define __HYPER_DMABUF_LIST_H__

+#include "hyper_dmabuf_struct.h"

+/* number of bits to be used for exported dmabufs hash table */ +#define MAX_ENTRY_EXPORTED 7 +/* number of bits to be used for imported dmabufs hash table */ +#define MAX_ENTRY_IMPORTED 7

+struct list_entry_exported {

• struct exported_sgt_info *exported;
• struct hlist_node node;

+};

+struct list_entry_imported {

• struct imported_sgt_info *imported;
• struct hlist_node node;

+};

+int hyper_dmabuf_table_init(void);

+int hyper_dmabuf_table_destroy(void);

+int hyper_dmabuf_register_exported(struct exported_sgt_info *info);

+/* search for pre-exported sgt and return id of it if it exist */ +hyper_dmabuf_id_t hyper_dmabuf_find_hid_exported(struct dma_buf *dmabuf,

• 				 int domid);
  

+int hyper_dmabuf_register_imported(struct imported_sgt_info *info);

+struct exported_sgt_info *hyper_dmabuf_find_exported(hyper_dmabuf_id_t hid);

+struct imported_sgt_info *hyper_dmabuf_find_imported(hyper_dmabuf_id_t hid);

+int hyper_dmabuf_remove_exported(hyper_dmabuf_id_t hid);

+int hyper_dmabuf_remove_imported(hyper_dmabuf_id_t hid);

+void hyper_dmabuf_foreach_exported(void (*func)(struct exported_sgt_info *,

• 		   void *attr), void *attr);
  

+int hyper_dmabuf_register_sysfs(struct device *dev); +int hyper_dmabuf_unregister_sysfs(struct device *dev);

+#endif /* __HYPER_DMABUF_LIST_H__ */ diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_msg.c b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_msg.c new file mode 100644 index 000000000000..129b2ff2af2b --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_msg.c @@ -0,0 +1,320 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• • Authors:
• • Dongwon Kim dongwon.kim@intel.com
• • Mateusz Polrola mateuszx.potrola@intel.com
• */

+#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/workqueue.h> +#include "hyper_dmabuf_drv.h" +#include "hyper_dmabuf_msg.h" +#include "hyper_dmabuf_list.h"

+struct cmd_process {

• struct work_struct work;
• struct hyper_dmabuf_req *rq;
• int domid;

+};

+void hyper_dmabuf_create_req(struct hyper_dmabuf_req *req,

• 	     enum hyper_dmabuf_command cmd, int *op)
  

can we have structures for all the types of requests/responses defined in some protocol header file? so we avoid hardcoding?

+{

• int i;
• req->stat = HYPER_DMABUF_REQ_NOT_RESPONDED;
• req->cmd = cmd;
• switch (cmd) {
• /* as exporter, commands to importer */
• case HYPER_DMABUF_EXPORT:

• /* exporting pages for dmabuf */
  

• /* command : HYPER_DMABUF_EXPORT,
  

•  * op0~op3 : hyper_dmabuf_id
  

•  * op4 : number of pages to be shared
  

•  * op5 : offset of data in the first page
  

•  * op6 : length of data in the last page
  

•  * op7 : top-level reference number for shared pages
  

•  */
  

• memcpy(&req->op[0], &op[0], 8 * sizeof(int) + op[8]);
  

• break;
  

• case HYPER_DMABUF_NOTIFY_UNEXPORT:

• /* destroy sg_list for hyper_dmabuf_id on remote side */
  

• /* command : DMABUF_DESTROY,
  

•  * op0~op3 : hyper_dmabuf_id_t hid
  

•  */
  

• for (i = 0; i < 4; i++)
  

• 	req->op[i] = op[i];
  

• break;
  

• case HYPER_DMABUF_EXPORT_FD:
• case HYPER_DMABUF_EXPORT_FD_FAILED:

• /* dmabuf fd is being created on imported side or importing
  

•  * failed
  

•  *
  

•  * command : HYPER_DMABUF_EXPORT_FD or
  

•  *	     HYPER_DMABUF_EXPORT_FD_FAILED,
  

•  * op0~op3 : hyper_dmabuf_id
  

•  */
  

• for (i = 0; i < 4; i++)
  

• 	req->op[i] = op[i];
  

• break;
  

• default:

• /* no command found */
  

• return;
  

• }

+}

+static void cmd_process_work(struct work_struct *work) +{

• struct imported_sgt_info *imported;
• struct cmd_process *proc = container_of(work,

• 				struct cmd_process, work);
  

• struct hyper_dmabuf_req *req;
• int domid;
• int i;
• req = proc->rq;
• domid = proc->domid;
• switch (req->cmd) {
• case HYPER_DMABUF_EXPORT:

• /* exporting pages for dmabuf */
  

• /* command : HYPER_DMABUF_EXPORT,
  

•  * op0~op3 : hyper_dmabuf_id
  

•  * op4 : number of pages to be shared
  

•  * op5 : offset of data in the first page
  

•  * op6 : length of data in the last page
  

•  * op7 : top-level reference number for shared pages
  

•  */
  

• /* if nents == 0, it means it is a message only for
  

•  * priv synchronization. for existing imported_sgt_info
  

•  * so not creating a new one
  

•  */
  

• if (req->op[4] == 0) {
  

• 	hyper_dmabuf_id_t exist = {req->op[0],
  

• 				   {req->op[1], req->op[2],
  

• 				   req->op[3] } };
  

• 	imported = hyper_dmabuf_find_imported(exist);
  

• 	if (!imported) {
  

• 		dev_err(hy_drv_priv->dev,
  

• 			"Can't find imported sgt_info\n");
  

• 		break;
  

• 	}
  

• 	break;
  

• }
  

• imported = kcalloc(1, sizeof(*imported), GFP_KERNEL);
  

• if (!imported)
  

• 	break;
  

• imported->hid.id = req->op[0];
  

• for (i = 0; i < 3; i++)
  

• 	imported->hid.rng_key[i] = req->op[i+1];
  

• imported->nents = req->op[4];
  

• imported->frst_ofst = req->op[5];
  

• imported->last_len = req->op[6];
  

• imported->ref_handle = req->op[7];
  

• dev_dbg(hy_drv_priv->dev, "DMABUF was exported\n");
  

• dev_dbg(hy_drv_priv->dev, "\thid{id:%d key:%d %d %d}\n",
  

• 	req->op[0], req->op[1], req->op[2],
  

• 	req->op[3]);
  

• dev_dbg(hy_drv_priv->dev, "\tnents %d\n", req->op[4]);
  

• dev_dbg(hy_drv_priv->dev, "\tfirst offset %d\n", req->op[5]);
  

• dev_dbg(hy_drv_priv->dev, "\tlast len %d\n", req->op[6]);
  

• dev_dbg(hy_drv_priv->dev, "\tgrefid %d\n", req->op[7]);
  

heh, and what if you have to insert something at index 1, for example? you'll end up changing all the hardcodes... Please have the protocol and its constants, structures etc. defined somewhere

• imported->valid = true;
  

• hyper_dmabuf_register_imported(imported);
  

• break;
  

• default:

• /* shouldn't get here */
  

• break;
  

• }
• kfree(req);
• kfree(proc);

+}

+int hyper_dmabuf_msg_parse(int domid, struct hyper_dmabuf_req *req)

This seems to be a hyper_dmabuf_msg_*handle* rather than parse...

+{

• struct cmd_process *proc;
• struct hyper_dmabuf_req *temp_req;
• struct imported_sgt_info *imported;
• struct exported_sgt_info *exported;
• hyper_dmabuf_id_t hid;
• if (!req) {

• dev_err(hy_drv_priv->dev, "request is NULL\n");
  

• return -EINVAL;
  

• }
• hid.id = req->op[0];
• hid.rng_key[0] = req->op[1];
• hid.rng_key[1] = req->op[2];
• hid.rng_key[2] = req->op[3];
• if ((req->cmd < HYPER_DMABUF_EXPORT) ||

• (req->cmd > HYPER_DMABUF_NOTIFY_UNEXPORT)) {
  

• dev_err(hy_drv_priv->dev, "invalid command\n");
  

• return -EINVAL;
  

• }
• req->stat = HYPER_DMABUF_REQ_PROCESSED;
• /* HYPER_DMABUF_DESTROY requires immediate

• * follow up so can't be processed in workqueue
  

• */
  

• if (req->cmd == HYPER_DMABUF_NOTIFY_UNEXPORT) {

• /* destroy sg_list for hyper_dmabuf_id on remote side */
  

• /* command : HYPER_DMABUF_NOTIFY_UNEXPORT,
  

•  * op0~3 : hyper_dmabuf_id
  

•  */
  

• dev_dbg(hy_drv_priv->dev,
  

• 	"processing HYPER_DMABUF_NOTIFY_UNEXPORT\n");
  

• imported = hyper_dmabuf_find_imported(hid);
  

• if (imported) {
  

• 	/* if anything is still using dma_buf */
  

• 	if (imported->importers) {
  

• 		/* Buffer is still in  use, just mark that
  

• 		 * it should not be allowed to export its fd
  

• 		 * anymore.
  

• 		 */
  

• 		imported->valid = false;
  

• 	} else {
  

• 		/* No one is using buffer, remove it from
  

• 		 * imported list
  

• 		 */
  

• 		hyper_dmabuf_remove_imported(hid);
  

• 		kfree(imported);
  

• 	}
  

• } else {
  

• 	req->stat = HYPER_DMABUF_REQ_ERROR;
  

• }
  

• return req->cmd;
  

• }
• /* synchronous dma_buf_fd export */
• if (req->cmd == HYPER_DMABUF_EXPORT_FD) {

• /* find a corresponding SGT for the id */
  

• dev_dbg(hy_drv_priv->dev,
  

• 	"HYPER_DMABUF_EXPORT_FD for {id:%d key:%d %d %d}\n",
  

• 	hid.id, hid.rng_key[0], hid.rng_key[1], hid.rng_key[2]);
  

• exported = hyper_dmabuf_find_exported(hid);
  

• if (!exported) {
  

• 	dev_err(hy_drv_priv->dev,
  

• 		"buffer {id:%d key:%d %d %d} not found\n",
  

• 		hid.id, hid.rng_key[0], hid.rng_key[1],
  

• 		hid.rng_key[2]);
  

• 	req->stat = HYPER_DMABUF_REQ_ERROR;
  

• } else if (!exported->valid) {
  

• 	dev_dbg(hy_drv_priv->dev,
  

• 		"Buffer no longer valid {id:%d key:%d %d %d}\n",
  

• 		hid.id, hid.rng_key[0], hid.rng_key[1],
  

• 		hid.rng_key[2]);
  

• 	req->stat = HYPER_DMABUF_REQ_ERROR;
  

• } else {
  

• 	dev_dbg(hy_drv_priv->dev,
  

• 		"Buffer still valid {id:%d key:%d %d %d}\n",
  

• 		hid.id, hid.rng_key[0], hid.rng_key[1],
  

• 		hid.rng_key[2]);
  

• 	exported->active++;
  

• 	req->stat = HYPER_DMABUF_REQ_PROCESSED;
  

• }
  

• return req->cmd;
  

• }
• if (req->cmd == HYPER_DMABUF_EXPORT_FD_FAILED) {

• dev_dbg(hy_drv_priv->dev,
  

• 	"HYPER_DMABUF_EXPORT_FD_FAILED for {id:%d key:%d %d %d}\n",
  

• 	hid.id, hid.rng_key[0], hid.rng_key[1], hid.rng_key[2]);
  

• exported = hyper_dmabuf_find_exported(hid);
  

• if (!exported) {
  

• 	dev_err(hy_drv_priv->dev,
  

• 		"buffer {id:%d key:%d %d %d} not found\n",
  

• 		hid.id, hid.rng_key[0], hid.rng_key[1],
  

• 		hid.rng_key[2]);
  

• 	req->stat = HYPER_DMABUF_REQ_ERROR;
  

• } else {
  

• 	exported->active--;
  

• 	req->stat = HYPER_DMABUF_REQ_PROCESSED;
  

• }
  

• return req->cmd;
  

• }
• dev_dbg(hy_drv_priv->dev,

• "%s: putting request to workqueue\n", __func__);
  

• temp_req = kmalloc(sizeof(*temp_req), GFP_KERNEL);
• if (!temp_req)

• return -ENOMEM;
  

• memcpy(temp_req, req, sizeof(*temp_req));
• proc = kcalloc(1, sizeof(struct cmd_process), GFP_KERNEL);
• if (!proc) {

• kfree(temp_req);
  

• return -ENOMEM;
  

• }
• proc->rq = temp_req;
• proc->domid = domid;
• INIT_WORK(&(proc->work), cmd_process_work);

Why do you need to be so asynchronous and schedule a work for processing rather than handle it now?

• queue_work(hy_drv_priv->work_queue, &(proc->work));
• return req->cmd;

+} diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_msg.h b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_msg.h new file mode 100644 index 000000000000..59f1528e9b1e --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_msg.h @@ -0,0 +1,87 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• */

+#ifndef __HYPER_DMABUF_MSG_H__ +#define __HYPER_DMABUF_MSG_H__

+#define MAX_NUMBER_OF_OPERANDS 8

+struct hyper_dmabuf_req {

• unsigned int req_id;
• unsigned int stat;
• unsigned int cmd;
• unsigned int op[MAX_NUMBER_OF_OPERANDS];

+};

+struct hyper_dmabuf_resp {

• unsigned int resp_id;
• unsigned int stat;
• unsigned int cmd;
• unsigned int op[MAX_NUMBER_OF_OPERANDS];

+};

The structures above are of size of 11 * sizeof(int) == 44 bytes Can these be aligned to 64 for example, From Xen POV: these will be sent over the shared ring, which is of PAGE_SIZE size, so 4096 / 44...

+enum hyper_dmabuf_command {

• HYPER_DMABUF_EXPORT = 0x10,
• HYPER_DMABUF_EXPORT_FD,
• HYPER_DMABUF_EXPORT_FD_FAILED,
• HYPER_DMABUF_NOTIFY_UNEXPORT,

+};

+enum hyper_dmabuf_ops {

• HYPER_DMABUF_OPS_ATTACH = 0x1000,
• HYPER_DMABUF_OPS_DETACH,
• HYPER_DMABUF_OPS_MAP,
• HYPER_DMABUF_OPS_UNMAP,
• HYPER_DMABUF_OPS_RELEASE,
• HYPER_DMABUF_OPS_BEGIN_CPU_ACCESS,
• HYPER_DMABUF_OPS_END_CPU_ACCESS,
• HYPER_DMABUF_OPS_KMAP_ATOMIC,
• HYPER_DMABUF_OPS_KUNMAP_ATOMIC,
• HYPER_DMABUF_OPS_KMAP,
• HYPER_DMABUF_OPS_KUNMAP,
• HYPER_DMABUF_OPS_MMAP,
• HYPER_DMABUF_OPS_VMAP,
• HYPER_DMABUF_OPS_VUNMAP,

+};

+enum hyper_dmabuf_req_feedback {

This rather seems to be a status

• HYPER_DMABUF_REQ_PROCESSED = 0x100,
• HYPER_DMABUF_REQ_NEEDS_FOLLOW_UP,
• HYPER_DMABUF_REQ_ERROR,
• HYPER_DMABUF_REQ_NOT_RESPONDED

+};

+/* create a request packet with given command and operands */ +void hyper_dmabuf_create_req(struct hyper_dmabuf_req *req,

• 		 enum hyper_dmabuf_command command,
  

• 		 int *operands);
  

+/* parse incoming request packet (or response) and take

• • appropriate actions for those
• */

+int hyper_dmabuf_msg_parse(int domid, struct hyper_dmabuf_req *req);

+#endif // __HYPER_DMABUF_MSG_H__ diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ops.c b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ops.c new file mode 100644 index 000000000000..b4d3c2caad73 --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ops.c @@ -0,0 +1,264 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• • Authors:
• • Dongwon Kim dongwon.kim@intel.com
• • Mateusz Polrola mateuszx.potrola@intel.com
• */

+#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/dma-buf.h> +#include "hyper_dmabuf_drv.h" +#include "hyper_dmabuf_struct.h" +#include "hyper_dmabuf_ops.h" +#include "hyper_dmabuf_sgl_proc.h" +#include "hyper_dmabuf_id.h" +#include "hyper_dmabuf_msg.h" +#include "hyper_dmabuf_list.h"

+#define WAIT_AFTER_SYNC_REQ 0 +#define REFS_PER_PAGE (PAGE_SIZE/sizeof(grant_ref_t))

+static int dmabuf_refcount(struct dma_buf *dma_buf) +{

• if ((dma_buf != NULL) && (dma_buf->file != NULL))

• return file_count(dma_buf->file);
  

• return -EINVAL;

+}

+static int hyper_dmabuf_ops_attach(struct dma_buf *dmabuf,

• 		   struct device *dev,
  

• 		   struct dma_buf_attachment *attach)
  

+{

• return 0;

+}

+static void hyper_dmabuf_ops_detach(struct dma_buf *dmabuf,

• 		    struct dma_buf_attachment *attach)
  

+{ +}

+static struct sg_table *hyper_dmabuf_ops_map(

• 		struct dma_buf_attachment *attachment,
  

• 		enum dma_data_direction dir)
  

+{

• struct sg_table *st;
• struct imported_sgt_info *imported;
• struct pages_info *pg_info;
• if (!attachment->dmabuf->priv)

• return NULL;
  

• imported = (struct imported_sgt_info *)attachment->dmabuf->priv;
• /* extract pages from sgt */
• pg_info = hyper_dmabuf_ext_pgs(imported->sgt);
• if (!pg_info)

• return NULL;
  

• /* create a new sg_table with extracted pages */
• st = hyper_dmabuf_create_sgt(pg_info->pgs, pg_info->frst_ofst,

• 		     pg_info->last_len, pg_info->nents);
  

• if (!st)

• goto err_free_sg;
  

• if (!dma_map_sg(attachment->dev, st->sgl, st->nents, dir))

• goto err_free_sg;
  

• kfree(pg_info->pgs);
• kfree(pg_info);
• return st;

+err_free_sg:

• if (st) {

• sg_free_table(st);
  

• kfree(st);
  

• }
• kfree(pg_info->pgs);
• kfree(pg_info);
• return NULL;

+}

+static void hyper_dmabuf_ops_unmap(struct dma_buf_attachment *attachment,

• 		   struct sg_table *sg,
  

• 		   enum dma_data_direction dir)
  

+{

• struct imported_sgt_info *imported;
• if (!attachment->dmabuf->priv)

• return;
  

• imported = (struct imported_sgt_info *)attachment->dmabuf->priv;
• dma_unmap_sg(attachment->dev, sg->sgl, sg->nents, dir);
• sg_free_table(sg);
• kfree(sg);

+}

+static void hyper_dmabuf_ops_release(struct dma_buf *dma_buf) +{

• struct imported_sgt_info *imported;
• struct hyper_dmabuf_bknd_ops *bknd_ops = hy_drv_priv->bknd_ops;
• int finish;
• if (!dma_buf->priv)

• return;
  

• imported = (struct imported_sgt_info *)dma_buf->priv;
• if (!dmabuf_refcount(imported->dma_buf))

• imported->dma_buf = NULL;
  

• imported->importers--;
• if (imported->importers == 0) {

• bknd_ops->unmap_shared_pages(&imported->refs_info,
  

• 			     imported->nents);
  

• if (imported->sgt) {
  

• 	sg_free_table(imported->sgt);
  

• 	kfree(imported->sgt);
  

• 	imported->sgt = NULL;
  

• }
  

• }
• finish = imported && !imported->valid &&

•  !imported->importers;
  

• /*

• * Check if buffer is still valid and if not remove it
  

• * from imported list. That has to be done after sending
  

• * sync request
  

• */
  

• if (finish) {

• hyper_dmabuf_remove_imported(imported->hid);
  

• kfree(imported);
  

• }

+}

+static int hyper_dmabuf_ops_begin_cpu_access(struct dma_buf *dmabuf,

• 			     enum dma_data_direction dir)
  

+{

• return 0;

+}

+static int hyper_dmabuf_ops_end_cpu_access(struct dma_buf *dmabuf,

• 			   enum dma_data_direction dir)
  

+{

• return 0;

+}

+static void *hyper_dmabuf_ops_kmap_atomic(struct dma_buf *dmabuf,

• 			  unsigned long pgnum)
  

+{

• /* TODO: NULL for now. Need to return the addr of mapped region */
• return NULL;

+}

+static void hyper_dmabuf_ops_kunmap_atomic(struct dma_buf *dmabuf,

• 			   unsigned long pgnum, void *vaddr)
  

+{ +}

+static void *hyper_dmabuf_ops_kmap(struct dma_buf *dmabuf, unsigned long pgnum) +{

• /* for now NULL.. need to return the address of mapped region */
• return NULL;

+}

+static void hyper_dmabuf_ops_kunmap(struct dma_buf *dmabuf, unsigned long pgnum,

• 		    void *vaddr)
  

+{ +}

+static int hyper_dmabuf_ops_mmap(struct dma_buf *dmabuf,

• 		 struct vm_area_struct *vma)
  

+{

• return 0;

+}

+static void *hyper_dmabuf_ops_vmap(struct dma_buf *dmabuf) +{

• return NULL;

+}

+static void hyper_dmabuf_ops_vunmap(struct dma_buf *dmabuf, void *vaddr) +{ +}

+static const struct dma_buf_ops hyper_dmabuf_ops = {

• .attach = hyper_dmabuf_ops_attach,
• .detach = hyper_dmabuf_ops_detach,
• .map_dma_buf = hyper_dmabuf_ops_map,
• .unmap_dma_buf = hyper_dmabuf_ops_unmap,
• .release = hyper_dmabuf_ops_release,
• .begin_cpu_access = (void *)hyper_dmabuf_ops_begin_cpu_access,
• .end_cpu_access = (void *)hyper_dmabuf_ops_end_cpu_access,
• .map_atomic = hyper_dmabuf_ops_kmap_atomic,
• .unmap_atomic = hyper_dmabuf_ops_kunmap_atomic,
• .map = hyper_dmabuf_ops_kmap,
• .unmap = hyper_dmabuf_ops_kunmap,
• .mmap = hyper_dmabuf_ops_mmap,
• .vmap = hyper_dmabuf_ops_vmap,
• .vunmap = hyper_dmabuf_ops_vunmap,

+};

+/* exporting dmabuf as fd */ +int hyper_dmabuf_export_fd(struct imported_sgt_info *imported, int flags) +{

• int fd = -1;
• /* call hyper_dmabuf_export_dmabuf and create

• * and bind a handle for it then release
  

• */
  

• hyper_dmabuf_export_dma_buf(imported);
• if (imported->dma_buf)

• fd = dma_buf_fd(imported->dma_buf, flags);
  

• return fd;

+}

+void hyper_dmabuf_export_dma_buf(struct imported_sgt_info *imported) +{

• DEFINE_DMA_BUF_EXPORT_INFO(exp_info);
• exp_info.ops = &hyper_dmabuf_ops;
• /* multiple of PAGE_SIZE, not considering offset */
• exp_info.size = imported->sgt->nents * PAGE_SIZE;

Here and below: it can be that PAGE_SIZE differs across VMs

• exp_info.flags = /* not sure about flag */ 0;
• exp_info.priv = imported;
• imported->dma_buf = dma_buf_export(&exp_info);

+} diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ops.h b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ops.h new file mode 100644 index 000000000000..b30367f2836b --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_ops.h @@ -0,0 +1,34 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• */

+#ifndef __HYPER_DMABUF_OPS_H__ +#define __HYPER_DMABUF_OPS_H__

+int hyper_dmabuf_export_fd(struct imported_sgt_info *imported, int flags);

+void hyper_dmabuf_export_dma_buf(struct imported_sgt_info *imported);

+#endif /* __HYPER_DMABUF_IMP_H__ */ diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_sgl_proc.c b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_sgl_proc.c new file mode 100644 index 000000000000..d92ae13d8a30 --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_sgl_proc.c @@ -0,0 +1,256 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• • Authors:
• • Dongwon Kim dongwon.kim@intel.com
• • Mateusz Polrola mateuszx.potrola@intel.com
• */

+#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/dma-buf.h> +#include "hyper_dmabuf_drv.h" +#include "hyper_dmabuf_struct.h" +#include "hyper_dmabuf_sgl_proc.h"

+#define REFS_PER_PAGE (PAGE_SIZE/sizeof(grant_ref_t))

+/* return total number of pages referenced by a sgt

• • for pre-calculation of # of pages behind a given sgt
• */

+static int get_num_pgs(struct sg_table *sgt) +{

• struct scatterlist *sgl;
• int length, i;
• /* at least one page */
• int num_pages = 1;
• sgl = sgt->sgl;
• length = sgl->length - PAGE_SIZE + sgl->offset;
• /* round-up */
• num_pages += ((length + PAGE_SIZE - 1)/PAGE_SIZE);

DIV_ROUND_UP

• for (i = 1; i < sgt->nents; i++) {

• sgl = sg_next(sgl);
  

• /* round-up */
  

• num_pages += ((sgl->length + PAGE_SIZE - 1) /
  

• 	     PAGE_SIZE); /* round-up */
  

Ditto

• }
• return num_pages;

+}

+/* extract pages directly from struct sg_table */ +struct pages_info *hyper_dmabuf_ext_pgs(struct sg_table *sgt) +{

• struct pages_info *pg_info;
• int i, j, k;
• int length;
• struct scatterlist *sgl;
• pg_info = kmalloc(sizeof(*pg_info), GFP_KERNEL);
• if (!pg_info)

• return NULL;
  

• pg_info->pgs = kmalloc_array(get_num_pgs(sgt),

• 		     sizeof(struct page *),
  

• 		     GFP_KERNEL);
  

• if (!pg_info->pgs) {

• kfree(pg_info);
  

• return NULL;
  

• }
• sgl = sgt->sgl;
• pg_info->nents = 1;
• pg_info->frst_ofst = sgl->offset;
• pg_info->pgs[0] = sg_page(sgl);
• length = sgl->length - PAGE_SIZE + sgl->offset;
• i = 1;
• while (length > 0) {

• pg_info->pgs[i] = nth_page(sg_page(sgl), i);
  

• length -= PAGE_SIZE;
  

• pg_info->nents++;
  

• i++;
  

• }
• for (j = 1; j < sgt->nents; j++) {

• sgl = sg_next(sgl);
  

• pg_info->pgs[i++] = sg_page(sgl);
  

• length = sgl->length - PAGE_SIZE;
  

• pg_info->nents++;
  

• k = 1;
  

• while (length > 0) {
  

• 	pg_info->pgs[i++] = nth_page(sg_page(sgl), k++);
  

• 	length -= PAGE_SIZE;
  

• 	pg_info->nents++;
  

• }
  

• }
• /*

• * lenght at that point will be 0 or negative,
  

• * so to calculate last page size just add it to PAGE_SIZE
  

• */
  

• pg_info->last_len = PAGE_SIZE + length;
• return pg_info;

+}

+/* create sg_table with given pages and other parameters */ +struct sg_table *hyper_dmabuf_create_sgt(struct page **pgs,

• 			 int frst_ofst, int last_len,
  

• 			 int nents)
  

+{

• struct sg_table *sgt;
• struct scatterlist *sgl;
• int i, ret;
• sgt = kmalloc(sizeof(struct sg_table), GFP_KERNEL);
• if (!sgt)

• return NULL;
  

• ret = sg_alloc_table(sgt, nents, GFP_KERNEL);
• if (ret) {

• if (sgt) {
  

• 	sg_free_table(sgt);
  

• 	kfree(sgt);
  

• }
  

• return NULL;
  

• }
• sgl = sgt->sgl;
• sg_set_page(sgl, pgs[0], PAGE_SIZE-frst_ofst, frst_ofst);
• for (i = 1; i < nents-1; i++) {

• sgl = sg_next(sgl);
  

• sg_set_page(sgl, pgs[i], PAGE_SIZE, 0);
  

• }
• if (nents > 1) /* more than one page */ {

• sgl = sg_next(sgl);
  

• sg_set_page(sgl, pgs[i], last_len, 0);
  

• }
• return sgt;

+}

+int hyper_dmabuf_cleanup_sgt_info(struct exported_sgt_info *exported,

• 		  int force)
  

+{

• struct sgt_list *sgtl;
• struct attachment_list *attachl;
• struct kmap_vaddr_list *va_kmapl;
• struct vmap_vaddr_list *va_vmapl;
• struct hyper_dmabuf_bknd_ops *bknd_ops = hy_drv_priv->bknd_ops;
• if (!exported) {

• dev_err(hy_drv_priv->dev, "invalid hyper_dmabuf_id\n");
  

• return -EINVAL;
  

• }
• /* if force != 1, sgt_info can be released only if

• * there's no activity on exported dma-buf on importer
  

• * side.
  

• */
  

• if (!force &&

•    exported->active) {
  

• dev_warn(hy_drv_priv->dev,
  

• 	 "dma-buf is used by importer\n");
  

• return -EPERM;
  

• }
• /* force == 1 is not recommended */
• while (!list_empty(&exported->va_kmapped->list)) {

• va_kmapl = list_first_entry(&exported->va_kmapped->list,
  

• 			    struct kmap_vaddr_list, list);
  

• dma_buf_kunmap(exported->dma_buf, 1, va_kmapl->vaddr);
  

• list_del(&va_kmapl->list);
  

• kfree(va_kmapl);
  

• }
• while (!list_empty(&exported->va_vmapped->list)) {

• va_vmapl = list_first_entry(&exported->va_vmapped->list,
  

• 			    struct vmap_vaddr_list, list);
  

• dma_buf_vunmap(exported->dma_buf, va_vmapl->vaddr);
  

• list_del(&va_vmapl->list);
  

• kfree(va_vmapl);
  

• }
• while (!list_empty(&exported->active_sgts->list)) {

• attachl = list_first_entry(&exported->active_attached->list,
  

• 			   struct attachment_list, list);
  

• sgtl = list_first_entry(&exported->active_sgts->list,
  

• 			struct sgt_list, list);
  

• dma_buf_unmap_attachment(attachl->attach, sgtl->sgt,
  

• 			 DMA_BIDIRECTIONAL);
  

• list_del(&sgtl->list);
  

• kfree(sgtl);
  

• }
• while (!list_empty(&exported->active_sgts->list)) {

• attachl = list_first_entry(&exported->active_attached->list,
  

• 			   struct attachment_list, list);
  

• dma_buf_detach(exported->dma_buf, attachl->attach);
  

• list_del(&attachl->list);
  

• kfree(attachl);
  

• }
• /* Start cleanup of buffer in reverse order to exporting */
• bknd_ops->unshare_pages(&exported->refs_info, exported->nents);

is the above synchronous? can it be delayed?

• /* unmap dma-buf */
• dma_buf_unmap_attachment(exported->active_attached->attach,

• 		 exported->active_sgts->sgt,
  

• 		 DMA_BIDIRECTIONAL);
  

if the above is asynchronous then this might make troubles as we are unmapping yet shared pages

• /* detatch dma-buf */
• dma_buf_detach(exported->dma_buf, exported->active_attached->attach);
• /* close connection to dma-buf completely */
• dma_buf_put(exported->dma_buf);
• exported->dma_buf = NULL;
• kfree(exported->active_sgts);
• kfree(exported->active_attached);
• kfree(exported->va_kmapped);
• kfree(exported->va_vmapped);
• return 0;

+} diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_sgl_proc.h b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_sgl_proc.h new file mode 100644 index 000000000000..8dbc9c3dfda4 --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_sgl_proc.h @@ -0,0 +1,43 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• */

+#ifndef __HYPER_DMABUF_IMP_H__ +#define __HYPER_DMABUF_IMP_H__

+/* extract pages directly from struct sg_table */ +struct pages_info *hyper_dmabuf_ext_pgs(struct sg_table *sgt);

+/* create sg_table with given pages and other parameters */ +struct sg_table *hyper_dmabuf_create_sgt(struct page **pgs,

• 			 int frst_ofst, int last_len,
  

• 			 int nents);
  

+int hyper_dmabuf_cleanup_sgt_info(struct exported_sgt_info *exported,

• 		  int force);
  

+void hyper_dmabuf_free_sgt(struct sg_table *sgt);

+#endif /* __HYPER_DMABUF_IMP_H__ */ diff --git a/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_struct.h b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_struct.h new file mode 100644 index 000000000000..144e3821fbc2 --- /dev/null +++ b/drivers/dma-buf/hyper_dmabuf/hyper_dmabuf_struct.h @@ -0,0 +1,131 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• • SPDX-License-Identifier: (MIT OR GPL-2.0)
• */

+#ifndef __HYPER_DMABUF_STRUCT_H__ +#define __HYPER_DMABUF_STRUCT_H__

+/* stack of mapped sgts */ +struct sgt_list {

• struct sg_table *sgt;
• struct list_head list;

+};

+/* stack of attachments */ +struct attachment_list {

• struct dma_buf_attachment *attach;
• struct list_head list;

+};

+/* stack of vaddr mapped via kmap */ +struct kmap_vaddr_list {

• void *vaddr;
• struct list_head list;

+};

+/* stack of vaddr mapped via vmap */ +struct vmap_vaddr_list {

• void *vaddr;
• struct list_head list;

+};

+/* Exporter builds pages_info before sharing pages */ +struct pages_info {

• int frst_ofst;
• int last_len;
• int nents;
• struct page **pgs;

+};

+/* Exporter stores references to sgt in a hash table

• • Exporter keeps these references for synchronization
• • and tracking purposes
• */

+struct exported_sgt_info {

• hyper_dmabuf_id_t hid;
• /* VM ID of importer */
• int rdomid;
• struct dma_buf *dma_buf;
• int nents;
• /* list for tracking activities on dma_buf */
• struct sgt_list *active_sgts;
• struct attachment_list *active_attached;
• struct kmap_vaddr_list *va_kmapped;
• struct vmap_vaddr_list *va_vmapped;
• /* set to 0 when unexported. Importer doesn't

• * do a new mapping of buffer if valid == false
  

• */
  

• bool valid;
• /* active == true if the buffer is actively used

• * (mapped) by importer
  

• */
  

• int active;
• /* hypervisor specific reference data for shared pages */
• void *refs_info;
• struct delayed_work unexport;
• bool unexport_sched;
• /* list for file pointers associated with all user space

• * application that have exported this same buffer to
  

• * another VM. This needs to be tracked to know whether
  

• * the buffer can be completely freed.
  

• */
  

• struct file *filp;

+};

+/* imported_sgt_info contains information about imported DMA_BUF

• • this info is kept in IMPORT list and asynchorously retrieved and
• • used to map DMA_BUF on importer VM's side upon export fd ioctl
• • request from user-space
• */

+struct imported_sgt_info {

• hyper_dmabuf_id_t hid; /* unique id for shared dmabuf imported */
• /* hypervisor-specific handle to pages */
• int ref_handle;
• /* offset and size info of DMA_BUF */
• int frst_ofst;
• int last_len;
• int nents;
• struct dma_buf *dma_buf;
• struct sg_table *sgt;
• void *refs_info;
• bool valid;
• int importers;

+};

+#endif /* __HYPER_DMABUF_STRUCT_H__ */ diff --git a/include/uapi/linux/hyper_dmabuf.h b/include/uapi/linux/hyper_dmabuf.h new file mode 100644 index 000000000000..caaae2da9d4d --- /dev/null +++ b/include/uapi/linux/hyper_dmabuf.h @@ -0,0 +1,87 @@ +/*

• • Copyright © 2018 Intel Corporation
• • Permission is hereby granted, free of charge, to any person obtaining a
• • copy of this software and associated documentation files (the "Software"),
• • to deal in the Software without restriction, including without limitation
• • the rights to use, copy, modify, merge, publish, distribute, sublicense,
• • and/or sell copies of the Software, and to permit persons to whom the
• • Software is furnished to do so, subject to the following conditions:
• • The above copyright notice and this permission notice (including the next
• • paragraph) shall be included in all copies or substantial portions of the
• • Software.
• • THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
• • IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
• • FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
• • THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
• • LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
• • FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
• • IN THE SOFTWARE.
• */

+#ifndef __LINUX_PUBLIC_HYPER_DMABUF_H__ +#define __LINUX_PUBLIC_HYPER_DMABUF_H__

+typedef struct {

• int id;

can this be defined as a union as you seem to store count and vm_id in this field?

• int rng_key[3]; /* 12bytes long random number */

+} hyper_dmabuf_id_t;

+#define IOCTL_HYPER_DMABUF_TX_CH_SETUP \ +_IOC(_IOC_NONE, 'G', 0, sizeof(struct ioctl_hyper_dmabuf_tx_ch_setup)) +struct ioctl_hyper_dmabuf_tx_ch_setup {

• /* IN parameters */
• /* Remote domain id */
• int remote_domain;

+};

+#define IOCTL_HYPER_DMABUF_RX_CH_SETUP \ +_IOC(_IOC_NONE, 'G', 1, sizeof(struct ioctl_hyper_dmabuf_rx_ch_setup)) +struct ioctl_hyper_dmabuf_rx_ch_setup {

• /* IN parameters */
• /* Source domain id */
• int source_domain;

+};

+#define IOCTL_HYPER_DMABUF_EXPORT_REMOTE \ +_IOC(_IOC_NONE, 'G', 2, sizeof(struct ioctl_hyper_dmabuf_export_remote)) +struct ioctl_hyper_dmabuf_export_remote {

• /* IN parameters */
• /* DMA buf fd to be exported */
• int dmabuf_fd;
• /* Domain id to which buffer should be exported */
• int remote_domain;
• /* exported dma buf id */
• hyper_dmabuf_id_t hid;

+};

+#define IOCTL_HYPER_DMABUF_EXPORT_FD \ +_IOC(_IOC_NONE, 'G', 3, sizeof(struct ioctl_hyper_dmabuf_export_fd)) +struct ioctl_hyper_dmabuf_export_fd {

• /* IN parameters */
• /* hyper dmabuf id to be imported */
• hyper_dmabuf_id_t hid;
• /* flags */
• int flags;
• /* OUT parameters */
• /* exported dma buf fd */
• int fd;

+};

+#define IOCTL_HYPER_DMABUF_UNEXPORT \ +_IOC(_IOC_NONE, 'G', 4, sizeof(struct ioctl_hyper_dmabuf_unexport)) +struct ioctl_hyper_dmabuf_unexport {

• /* IN parameters */
• /* hyper dmabuf id to be unexported */
• hyper_dmabuf_id_t hid;
• /* delay in ms by which unexport processing will be postponed */
• int delay_ms;
• /* OUT parameters */
• /* Status of request */
• int status;

+};

+#endif //__LINUX_PUBLIC_HYPER_DMABUF_H__

On Tue, Feb 13, 2018 at 05:50:01PM -0800, Dongwon Kim wrote:

Reference document for hyper_DMABUF driver

Documentation/hyper-dmabuf-sharing.txt

This should likely be patch 1 in order for reviewers to have the appropriate context.

Signed-off-by: Dongwon Kim dongwon.kim@intel.com

Documentation/hyper-dmabuf-sharing.txt | 734 +++++++++++++++++++++++++++++++++ 1 file changed, 734 insertions(+) create mode 100644 Documentation/hyper-dmabuf-sharing.txt

diff --git a/Documentation/hyper-dmabuf-sharing.txt b/Documentation/hyper-dmabuf-sharing.txt new file mode 100644 index 000000000000..928e411931e3 --- /dev/null +++ b/Documentation/hyper-dmabuf-sharing.txt @@ -0,0 +1,734 @@ +Linux Hyper DMABUF Driver

+------------------------------------------------------------------------------ +Section 1. Overview +------------------------------------------------------------------------------

+Hyper_DMABUF driver is a Linux device driver running on multiple Virtual +achines (VMs), which expands DMA-BUF sharing capability to the VM environment +where multiple different OS instances need to share same physical data without +data-copy across VMs.

+To share a DMA_BUF across VMs, an instance of the Hyper_DMABUF drv on the +exporting VM (so called, “exporter”) imports a local DMA_BUF from the original +producer of the buffer,

The usage of export and import in the above sentence makes it almost impossible to understand.

then re-exports it with an unique ID, hyper_dmabuf_id +for the buffer to the importing VM (so called, “importer”).

And this is even worse.

Maybe it would help to have some kind of flow diagram of all this import/export operations, but please read below.

+Another instance of the Hyper_DMABUF driver on importer registers +a hyper_dmabuf_id together with reference information for the shared physical +pages associated with the DMA_BUF to its database when the export happens.

+The actual mapping of the DMA_BUF on the importer’s side is done by +the Hyper_DMABUF driver when user space issues the IOCTL command to access +the shared DMA_BUF. The Hyper_DMABUF driver works as both an importing and +exporting driver as is, that is, no special configuration is required. +Consequently, only a single module per VM is needed to enable cross-VM DMA_BUF +exchange.

IMHO I need a more generic view of the problem you are trying to solve in the overview section. I've read the full overview, and I still have no idea why you need all this.

I think the overview should contain at least:

1. A description of the problem you are trying to solve. 2. A high level description of the proposed solution. 3. How the proposed solution deals with the problem described in 1.

This overview is not useful for people that don't know which problem you are trying to solve, like myself.

Thanks, Roger.

Sorry for top-posting

Can we have all this go into some header file which

will not only describe the structures/commands/responses/etc,

but will also allow drivers to use those directly without

defining the same one more time in the code? For example,

this is how it is done in Xen [1]. This way, you can keep

documentation and the protocol implementation in sync easily

On 02/14/2018 03:50 AM, Dongwon Kim wrote:

Reference document for hyper_DMABUF driver

Documentation/hyper-dmabuf-sharing.txt

Signed-off-by: Dongwon Kim dongwon.kim@intel.com

Documentation/hyper-dmabuf-sharing.txt | 734 +++++++++++++++++++++++++++++++++ 1 file changed, 734 insertions(+) create mode 100644 Documentation/hyper-dmabuf-sharing.txt

diff --git a/Documentation/hyper-dmabuf-sharing.txt b/Documentation/hyper-dmabuf-sharing.txt new file mode 100644 index 000000000000..928e411931e3 --- /dev/null +++ b/Documentation/hyper-dmabuf-sharing.txt @@ -0,0 +1,734 @@ +Linux Hyper DMABUF Driver

+------------------------------------------------------------------------------ +Section 1. Overview +------------------------------------------------------------------------------

+Hyper_DMABUF driver is a Linux device driver running on multiple Virtual +achines (VMs), which expands DMA-BUF sharing capability to the VM environment +where multiple different OS instances need to share same physical data without +data-copy across VMs.

+To share a DMA_BUF across VMs, an instance of the Hyper_DMABUF drv on the +exporting VM (so called, “exporter”) imports a local DMA_BUF from the original +producer of the buffer, then re-exports it with an unique ID, hyper_dmabuf_id +for the buffer to the importing VM (so called, “importer”).

+Another instance of the Hyper_DMABUF driver on importer registers +a hyper_dmabuf_id together with reference information for the shared physical +pages associated with the DMA_BUF to its database when the export happens.

+The actual mapping of the DMA_BUF on the importer’s side is done by +the Hyper_DMABUF driver when user space issues the IOCTL command to access +the shared DMA_BUF. The Hyper_DMABUF driver works as both an importing and +exporting driver as is, that is, no special configuration is required. +Consequently, only a single module per VM is needed to enable cross-VM DMA_BUF +exchange.

+------------------------------------------------------------------------------ +Section 2. Architecture +------------------------------------------------------------------------------

+1. Hyper_DMABUF ID

+hyper_dmabuf_id is a global handle for shared DMA BUFs, which is compatible +across VMs. It is a key used by the importer to retrieve information about +shared Kernel pages behind the DMA_BUF structure from the IMPORT list. When +a DMA_BUF is exported to another domain, its hyper_dmabuf_id and META data +are also kept in the EXPORT list by the exporter for further synchronization +of control over the DMA_BUF.

+hyper_dmabuf_id is “targeted”, meaning it is valid only in exporting (owner of +the buffer) and importing VMs, where the corresponding hyper_dmabuf_id is +stored in their database (EXPORT and IMPORT lists).

+A user-space application specifies the targeted VM id in the user parameter +when it calls the IOCTL command to export shared DMA_BUF to another VM.

+hyper_dmabuf_id_t is a data type for hyper_dmabuf_id. It is defined as 16-byte +data structure, and it contains id and rng_key[3] as elements for +the structure.

+typedef struct {

•    int id;
  

•    int rng_key[3]; /* 12bytes long random number */
  

+} hyper_dmabuf_id_t;

+The first element in the hyper_dmabuf_id structure, int id is combined data of +a count number generated by the driver running on the exporter and +the exporter’s ID. The VM’s ID is a one byte value and located at the field’s +SB in int id. The remaining three bytes in int id are reserved for a count +number.

+However, there is a limit related to this count number, which is 1000. +Therefore, only little more than a byte starting from the LSB is actually used +for storing this count number.

+#define HYPER_DMABUF_ID_CREATE(domid, id) \

•    ((((domid) & 0xFF) << 24) | ((id) & 0xFFFFFF))
  

+This limit on the count number directly means the maximum number of DMA BUFs +that can be shared simultaneously by one VM. The second element of +hyper_dmabuf_id, that is int rng_key[3], is an array of three integers. These +numbers are generated by Linux’s native random number generation mechanism. +This field is added to enhance the security of the Hyper DMABUF driver by +maximizing the entropy of hyper_dmabuf_id (that is, preventing it from being +guessed by a security attacker).

+Once DMA_BUF is no longer shared, the hyper_dmabuf_id associated with +the DMA_BUF is released, but the count number in hyper_dmabuf_id is saved in +the ID list for reuse. However, random keys stored in int rng_key[3] are not +reused. Instead, those keys are always filled with freshly generated random +keys for security.

+2. IOCTLs

+a. IOCTL_HYPER_DMABUF_TX_CH_SETUP

+This type of IOCTL is used for initialization of a one-directional transmit +communication channel with a remote domain.

+The user space argument for this type of IOCTL is defined as:

+struct ioctl_hyper_dmabuf_tx_ch_setup {

• /* IN parameters */
• /* Remote domain id */
• int remote_domain;

+};

+b. IOCTL_HYPER_DMABUF_RX_CH_SETUP

+This type of IOCTL is used for initialization of a one-directional receive +communication channel with a remote domain.

+The user space argument for this type of IOCTL is defined as:

+struct ioctl_hyper_dmabuf_rx_ch_setup {

• /* IN parameters */
• /* Source domain id */
• int source_domain;

+};

+c. IOCTL_HYPER_DMABUF_EXPORT_REMOTE

+This type of IOCTL is used to export a DMA BUF to another VM. When a user +space application makes this call to the driver, it extracts Kernel pages +associated with the DMA_BUF, then makes those shared with the importing VM.

+All reference information for this shared pages and hyper_dmabuf_id is +created, then passed to the importing domain through a communications +channel for synchronous registration. In the meantime, the hyper_dmabuf_id +for the shared DMA_BUF is also returned to user-space application.

+This IOCTL can accept a reference to “user-defined” data as well as a FD +for the DMA BUF. This private data is then attached to the DMA BUF and +exported together with it.

+More details regarding this private data can be found in chapter for +“Hyper_DMABUF Private Data”.

+The user space argument for this type of IOCTL is defined as:

+struct ioctl_hyper_dmabuf_export_remote {

• /* IN parameters */
• /* DMA buf fd to be exported */
• int dmabuf_fd;
• /* Domain id to which buffer should be exported */
• int remote_domain;
• /* exported dma buf id */
• hyper_dmabuf_id_t hid;
• /* size of private data */
• int sz_priv;
• /* ptr to the private data for Hyper_DMABUF */
• char *priv;

+};

+d. IOCTL_HYPER_DMABUF_EXPORT_FD

+The importing VM uses this IOCTL to import and re-export a shared DMA_BUF +locally to the end-consumer using the standard Linux DMA_BUF framework. +Upon IOCTL call, the Hyper_DMABUF driver finds the reference information +of the shared DMA_BUF with the given hyper_dmabuf_id, then maps all shared +pages in its own Kernel space. The driver then constructs a scatter-gather +list with those mapped pages and creates a brand-new DMA_BUF with the list, +which is eventually exported with a file descriptor to the local consumer.

+The user space argument for this type of IOCTL is defined as:

+struct ioctl_hyper_dmabuf_export_fd {

• /* IN parameters */
• /* hyper dmabuf id to be imported */
• int hyper_dmabuf_id;
• /* flags */
• int flags;
• /* OUT parameters */
• /* exported dma buf fd */
• int fd;

+};

+e. IOCTL_HYPER_DMABUF_UNEXPORT

+This type of IOCTL is used when it is necessary to terminate the current +sharing of a DMA_BUF. When called, the driver first checks if there are any +consumers actively using the DMA_BUF. Then, it unexports it if it is not +mapped or used by any consumers. Otherwise, it postpones unexporting, but +makes the buffer invalid to prevent any further import of the same DMA_BUF. +DMA_BUF is completely unexported after the last consumer releases it.

+”Unexport” means removing all reference information about the DMA_BUF from the +LISTs and make all pages private again.

+The user space argument for this type of IOCTL is defined as:

+struct ioctl_hyper_dmabuf_unexport {

• /* IN parameters */
• /* hyper dmabuf id to be unexported */
• int hyper_dmabuf_id;
• /* delay in ms by which unexport processing will be postponed */
• int delay_ms;
• /* OUT parameters */
• /* Status of request */
• int status;

+};

+f. IOCTL_HYPER_DMABUF_QUERY

+This IOCTL is used to retrieve specific information about a DMA_BUF that +is being shared.

+The user space argument for this type of IOCTL is defined as:

+struct ioctl_hyper_dmabuf_query {

• /* in parameters */
• /* hyper dmabuf id to be queried */
• int hyper_dmabuf_id;
• /* item to be queried */
• int item;
• /* OUT parameters */
• /* output of query */
• /* info can be either value or reference */
• unsigned long info;

+};

+<Available Queries>

+HYPER_DMABUF_QUERY_TYPE

• • Return the type of DMA_BUF from the current domain, Exported or Imported.

+HYPER_DMABUF_QUERY_EXPORTER

• • Return the exporting domain’s ID of a shared DMA_BUF.

+HYPER_DMABUF_QUERY_IMPORTER

• • Return the importing domain’s ID of a shared DMA_BUF.

+HYPER_DMABUF_QUERY_SIZE

• • Return the size of a shared DMA_BUF in bytes.

+HYPER_DMABUF_QUERY_BUSY

• • Return ‘true’ if a shared DMA_BUF is currently used
• (mapped by the end-consumer).

+HYPER_DMABUF_QUERY_UNEXPORTED

• • Return ‘true’ if a shared DMA_BUF is not valid anymore
• (so it does not allow a new consumer to map it).

+HYPER_DMABUF_QUERY_DELAYED_UNEXPORTED

• • Return ‘true’ if a shared DMA_BUF is scheduled to be unexported
• (but is still valid) within a fixed time.

+HYPER_DMABUF_QUERY_PRIV_INFO

• • Return ‘private’ data attached to shared DMA_BUF to the user space.
• ‘unsigned long info’ is the user space pointer for the buffer, where
• private data will be copied to.

+HYPER_DMABUF_QUERY_PRIV_INFO_SIZE

• • Return the size of the private data attached to the shared DMA_BUF.

+3. Event Polling

+Event-polling can be enabled optionally by selecting the Kernel config option, +Enable event-generation and polling operation under xen/hypervisor in Kernel’s +menuconfig. The event-polling mechanism includes the generation of +an import-event, adding it to the event-queue and providing a notification to +the application so that it can retrieve the event data from the queue.

+For this mechanism, “Poll” and “Read” operations are added to the Hyper_DMABUF +driver. A user application that polls the driver goes into a sleep state until +there is a new event added to the queue. An application uses “Read” to retrieve +event data from the event queue. Event data contains the hyper_dmabuf_id and +the private data of the buffer that has been received by the importer.

+For more information on private data, refer to Section 3.5). +Using this method, it is possible to lower the risk of the hyper_dmabuf_id and +other sensitive information about the shared buffer (for example, meta-data +for shared images) being leaked while being transferred to the importer because +all of this data is shared as “private info” at the driver level. However, +please note there should be a way for the importer to find the correct DMA_BUF +in this case when there are multiple Hyper_DMABUFs being shared simultaneously. +For example, the surface name or the surface ID of a specific rendering surface +needs to be sent to the importer in advance before it is exported in a surface- +sharing use-case.

+Each event data given to the user-space consists of a header and the private +information of the buffer. The data type is defined as follows:

+struct hyper_dmabuf_event_hdr {

•    int event_type; /* one type only for now - new import */
  

•    hyper_dmabuf_id_t hid; /* hyper_dmabuf_id of specific hyper_dmabuf */
  

•    int size; /* size of data */
  

+};

+struct hyper_dmabuf_event_data {

•    struct hyper_dmabuf_event_hdr hdr;
  

•    void *data; /* private data */
  

+};

+4. Hyper_DMABUF Private Data

+Each Hyper_DMABUF can come with private data, the size of which can be up to +AX_SIZE_PRIV_DATA (currently 192 byte). This private data is just a chunk of +plain data attached to every Hyper_DMABUF. It is guaranteed to be synchronized +across VMs, exporter and importer. This private data does not have any specific +structure defined at the driver level, so any “user-defined” format or +structure can be used. In addition, there is no dedicated use-case for this +data. It can be used virtually for any purpose. For example, it can be used to +share meta-data such as dimension and color formats for shared images in +a surface sharing model. Another example is when we share protected media +contents.

+This private data can be used to transfer flags related to content protection +information on streamed media to the importer.

+Private data is initially generated when a buffer is exported for the first +time. Then, it is updated whenever the same buffer is re-exported. During the +re-exporting process, the Hyper_DMABUF driver only updates private data on +both sides with new data from user-space since the same buffer already exists +on both the IMPORT LIST and EXPORT LIST.

+There are two different ways to retrieve this private data from user-space. +The first way is to use “Read” on the Hyper_DMABUF driver. “Read” returns the +data of events containing private data of the buffer. The second way is to +make a query to Hyper_DMABUF. There are two query items, +HYPER_DMABUF_QUERY_PRIV_INFO and HYPER_DMABUF_QUERY_PRIV_INFO_SIZE available +for retrieving private data and its size.

+5. Scatter-Gather List Table (SGT) Management

+SGT management is the core part of the Hyper_DMABUF driver that manages an +SGT, a representation of the group of kernel pages associated with a DMA_BUF. +This block includes four different sub-blocks:

+a. Hyper_DMABUF_id Manager

+This ID manager is responsible for generating a hyper_dmabuf_id for an +exported DMA_BUF. When an ID is requested, the ID Manager first checks if +there are any reusable IDs left in the list and returns one of those, +if available. Otherwise, it creates the next count number and returns it +to the caller.

+b. SGT Creator

+The SGT (struct sg_table) contains information about the DMA_BUF such as +references to all kernel pages for the buffer and their connections. The +SGT Creator creates a new SGT on the importer side with pages shared by +the hypervisor.

+c. Kernel Page Extractor

+The Page Extractor extracts pages from a given SGT before those pages +are shared.

+d. List Manager Interface

+The SGT manger also interacts with export and import list managers. It +sends out information (for example, hyper_dmabuf_id, reference, and +DMA_BUF information) about the exported or imported DMA_BUFs to the +list manager. Also, on IOCTL request, it asks the list manager to find +and return the information for a corresponding DMA_BUF in the list.

+6. DMA-BUF Interface

+The DMA-BUF interface provides standard methods to manage DMA_BUFs +reconstructed by the Hyper_DMABUF driver from shared pages. All of the +relevant operations are listed in struct dma_buf_ops. These operations +are standard DMA_BUF operations, therefore they follow standard DMA BUF +protocols.

+Each DMA_BUF operation communicates with the exporter at the end of the +routine for “indirect DMA_BUF synchronization”.

+7. Export/Import List Management

+Whenever a DMA_BUF is shared and exported, its information is added to the +database (EXPORT-list) on the exporting VM. Similarly, information about an +imported DMA_BUF is added to the importing database (IMPORT list) on the +importing VM, when the export happens.

+All of the entries in the lists are needed to manage the exported/imported +DMA_BUF more efficiently. Both lists are implemented as Linux hash tables. +The key to the list is hyper_dmabuf_id and the output is the information of +the DMA_BUF. The List Manager manages all requests from other blocks and +transactions within lists to ensure that all entries are up-to-date and +that the list structure is consistent.

+The List Manager provides basic functionality, such as:

+- Adding to the List +- Removal from the List +- Finding information about a DMA_BUF, given the hyper_dmabuf_id

+8. Page Sharing by Hypercalls

+The Hyper_DMABUF driver assumes that there is a native page-by-page memory +sharing mechanism available on the hypervisor. Referencing a group of pages +that are being shared is what the driver expects from “backend” APIs or the +hypervisor itself.

+For the example, xen backend integrated in current code base utilizes Xen’s +grant-table interface for sharing the underlying kernel pages (struct *page).

+More details about grant-table interface can be found at the following locations:

+https://wiki.xen.org/wiki/Grant_Table +https://xenbits.xen.org/docs/4.6-testing/misc/grant-tables.txt

+9. Message Handling

+The exporter and importer can each create a message that consists of an opcode +(command) and operands (parameters) and send it to each other.

+The message format is defined as:

+struct hyper_dmabuf_req {

•    unsigned int req_id; /* Sequence number. Used for RING BUF
  

•                            synchronization */
  

•    unsigned int stat; /* Status.Response from receiver. */
  

•    unsigned int cmd;  /* Opcode */
  

•    unsigned int op[MAX_NUMBER_OF_OPERANDS]; /* Operands */
  

+};

+The following table gives the list of opcodes:

+<Opcodes in Message to Exporter/Importer>

+HYPER_DMABUF_EXPORT (exporter --> importer)

• • Export a DMA_BUF to the importer. The importer registers the corresponding
• DMA_BUF in its IMPORT LIST when the message is received.

+HYPER_DMABUF_EXPORT_FD (importer --> exporter)

• • Locally exported as FD. The importer sends out this command to the exporter
• to notify that the buffer is now locally exported (mapped and used).

+HYPER_DMABUF_EXPORT_FD_FAILED (importer --> exporter)

• • Failed while exporting locally. The importer sends out this command to the
• exporter to notify the exporter that the EXPORT_FD failed.

+HYPER_DMABUF_NOTIFY_UNEXPORT (exporter --> importer)

• • Termination of sharing. The exporter notifies the importer that the DMA_BUF
• has been unexported.

+HYPER_DMABUF_OPS_TO_REMOTE (importer --> exporter)

• • Not implemented yet.

+HYPER_DMABUF_OPS_TO_SOURCE (exporter --> importer)

• • DMA_BUF ops to the exporter, for DMA_BUF upstream synchronization.
• Note: Implemented but it is done asynchronously due to performance issues.

+The following table shows the list of operands for each opcode.

+<Operands in Message to Exporter/Importer>

+- HYPER_DMABUF_EXPORT

+op0 to op3 – hyper_dmabuf_id +op4 – number of pages to be shared +op5 – offset of data in the first page +op6 – length of data in the last page +op7 – reference number for the group of shared pages +op8 – size of private data +op9 to (op9+op8) – private data

+- HYPER_DMABUF_EXPORT_FD

+op0 to op3 – hyper_dmabuf_id

+- HYPER_DMABUF_EXPORT_FD_FAILED

+op0 to op3 – hyper_dmabuf_id

+- HYPER_DMABUF_NOTIFY_UNEXPORT

+op0 to op3 – hyper_dmabuf_id

+- HYPER_DMABUF_OPS_TO_REMOTE(Not implemented)

+- HYPER_DMABUF_OPS_TO_SOURCE

+op0 to op3 – hyper_dmabuf_id +op4 – type of DMA_BUF operation

+9. Inter VM (Domain) Communication

+Two different types of inter-domain communication channels are required, +one in kernel space and the other in user space. The communication channel +in user space is for transmitting or receiving the hyper_dmabuf_id. Since +there is no specific security (for example, encryption) involved in the +generation of a global id at the driver level, it is highly recommended that +the customer’s user application set up a very secure channel for exchanging +hyper_dmabuf_id between VMs.

+The communication channel in kernel space is required for exchanging messages +from “message management” block between two VMs. In the current reference +backend for Xen hypervisor, Xen ring-buffer and event-channel mechanisms are +used for message exchange between impoter and exporter.

+10. What are required in hypervisor

+emory sharing and message communication between VMs

+------------------------------------------------------------------------------ +Section 3. Hyper DMABUF Sharing Flow +------------------------------------------------------------------------------

+1. Exporting

+To export a DMA_BUF to another VM, user space has to call an IOCTL +(IOCTL_HYPER_DMABUF_EXPORT_REMOTE) with a file descriptor for the buffer given +by the original exporter. The Hyper_DMABUF driver maps a DMA_BUF locally, then +issues a hyper_dmabuf_id and SGT for the DMA_BUF, which is registered to the +EXPORT list. Then, all pages for the SGT are extracted and each individual +page is shared via a hypervisor-specific memory sharing mechanism +(for example, in Xen this is grant-table).

+One important requirement on this memory sharing method is that it needs to +create a single integer value that represents the list of pages, which can +then be used by the importer for retrieving the group of shared pages. For +this, the “Backend” in the reference driver utilizes the multiple level +addressing mechanism.

+Once the integer reference to the list of pages is created, the exporter +builds the “export” command and sends it to the importer, then notifies the +importer.

+2. Importing

+The Import process is divided into two sections. One is the registration +of DMA_BUF from the exporter. The other is the actual mapping of the buffer +before accessing the data in the buffer. The former (termed “Registration”) +happens on an export event (that is, the export command with an interrupt) +in the exporter.

+The latter (termed “Mapping”) is done asynchronously when the driver gets the +IOCTL call from user space. When the importer gets an interrupt from the +exporter, it checks the command in the receiving queue and if it is an +“export” command, the registration process is started. It first finds +hyper_dmabuf_id and the integer reference for the shared pages, then stores +all of that information together with the “domain id” of the exporting domain +in the IMPORT LIST.

+In the case where “event-polling” is enabled (Kernel Config - Enable event- +generation and polling operation), a “new sharing available” event is +generated right after the reference info for the new shared DMA_BUF is +registered to the IMPORT LIST. This event is added to the event-queue.

+The user process that polls Hyper_DMABUF driver wakes up when this event-queue +is not empty and is able to read back event data from the queue using the +driver’s “Read” function. Once the user-application calls EXPORT_FD IOCTL with +the proper parameters including hyper_dmabuf_id, the Hyper_DMABUF driver +retrieves information about the matched DMA_BUF from the IMPORT LIST. Then, it +maps all pages shared (referenced by the integer reference) in its kernel +space and creates its own DMA_BUF referencing the same shared pages. After +this, it exports this new DMA_BUF to the other drivers with a file descriptor. +DMA_BUF can then be used just in the same way a local DMA_BUF is.

+3. Indirect Synchronization of DMA_BUF

+Synchronization of a DMA_BUF within a single OS is automatically achieved +because all of importer’s DMA_BUF operations are done using functions defined +on the exporter’s side, which means there is one central place that has full +control over the DMA_BUF. In other words, any primary activities such as +attaching/detaching and mapping/un-mapping are all captured by the exporter, +meaning that the exporter knows basic information such as who is using the +DMA_BUF and how it is being used. This, however, is not applicable if this +sharing is done beyond a single OS because kernel space (where the exporter’s +DMA_BUF operations reside) is simply not visible to the importing VM.

+Therefore, “indirect synchronization” was introduced as an alternative solution, +which is now implemented in the Hyper_DMABUF driver. This technique makes +the exporter create a shadow DMA_BUF when the end-consumer of the buffer maps +the DMA_BUF, then duplicates any DMA_BUF operations performed on +the importer’s side. Through this “indirect synchronization”, the exporter is +able to virtually track all activities done by the consumer (mostly reference +counter) as if those are done in exporter’s local system.

+------------------------------------------------------------------------------ +Section 4. Hypervisor Backend Interface +------------------------------------------------------------------------------

+The Hyper_DMABUF driver has a standard “Backend” structure that contains +mappings to various functions designed for a specific Hypervisor. Most of +these API functions should provide a low-level implementation of communication +and memory sharing capability that utilize a Hypervisor’s native mechanisms.

+struct hyper_dmabuf_backend_ops {

•    /* retreiving id of current virtual machine */
  

•    int (*get_vm_id)(void);
  

•    /* get pages shared via hypervisor-specific method */
  

•    int (*share_pages)(struct page **, int, int, void **);
  

•    /* make shared pages unshared via hypervisor specific method */
  

•    int (*unshare_pages)(void **, int);
  

•    /* map remotely shared pages on importer's side via
  

•     *  hypervisor-specific method
  

•     */
  

•    struct page ** (*map_shared_pages)(int, int, int, void **);
  

•    /* unmap and free shared pages on importer's side via
  

•     *  hypervisor-specific method
  

•     */
  

•    int (*unmap_shared_pages)(void **, int);
  

•    /* initialize communication environment */
  

•    int (*init_comm_env)(void);
  

•    /* destroy communication channel */
  

•    void (*destroy_comm)(void);
  

•    /* upstream ch setup (receiving and responding) */
  

•    int (*init_rx_ch)(int);
  

•    /* downstream ch setup (transmitting and parsing responses) */
  

•    int (*init_tx_ch)(int);
  

•    /* send msg via communication ch */
  

•    int (*send_req)(int, struct hyper_dmabuf_req *, int);
  

+};

+<Hypervisor-specific Backend Structure>

+1. get_vm_id

• Returns the VM (domain) ID
• Input:

• -ID of the current domain
  

• Output:

• None
  

+2. share_pages

• Get pages shared via hypervisor-specific method and return one reference
• ID that represents the complete list of shared pages
• Input:

• -Array of pages
  

• -ID of importing VM
  

• -Number of pages
  

• -Hypervisor specific Representation of reference info of shared
  

•  pages
  

• Output:

• -Hypervisor specific integer value that represents all of
  

•  the shared pages
  

+3. unshare_pages

• Stop sharing pages
• Input:

• -Hypervisor specific Representation of reference info of shared
  

•  pages
  

• -Number of shared pages
  

• Output:

• 0
  

+4. map_shared_pages

• Map shared pages locally using a hypervisor-specific method
• Input:

• -Reference number that represents all of shared pages
  

• -ID of exporting VM, Number of pages
  

• -Reference information for any purpose
  

• Output:

• -An array of shared pages (struct page**)
  

+5. unmap_shared_pages

• Unmap shared pages
• Input:

• -Hypervisor specific Representation of reference info of shared pages
  

• Output:

• -0 (successful) or one of Standard Kernel errors
  

+6. init_comm_env

• Setup infrastructure needed for communication channel
• Input:

• None
  

• Output:

• None
  

+7. destroy_comm

• Cleanup everything done via init_comm_env
• Input:

• None
  

• Output:

• None
  

+8. init_rx_ch

• Configure receive channel
• Input:

• -ID of VM on the other side of the channel
  

• Output:

• -0 (successful) or one of Standard Kernel errors
  

+9. init_tx_ch

• Configure transmit channel
• Input:

• -ID of VM on the other side of the channel
  

• Output:

• -0 (success) or one of Standard Kernel errors
  

+10. send_req

• Send message to other VM
• Input:

• -ID of VM that receives the message
  

• -Message
  

• Output:

• -0 (success) or one of Standard Kernel errors
  

+------------------------------------------------------------------------------- +-------------------------------------------------------------------------------

[1] https://elixir.bootlin.com/linux/v4.16.1/source/include/xen/interface/io/kbd...