Hello Everyone,
This is RFC v3 for DMA buffer sharing mechanism - changes from v2 are in the changelog below.
Various subsystems - V4L2, GPU-accessors, DRI to name a few - have felt the need to have a common mechanism to share memory buffers across different devices - ARM, video hardware, GPU.
This need comes forth from a variety of use cases including cameras, image processing, video recorders, sound processing, DMA engines, GPU and display buffers, and others.
This RFC is an attempt to define such a buffer sharing mechanism- it is the result of discussions from a couple of memory-management mini-summits held by Linaro to understand and address common needs around memory management. [1]
A new dma_buf buffer object is added, with operations and API to allow easy sharing of this buffer object across devices.
The framework allows: - a new buffer-object to be created with fixed size. - different devices to 'attach' themselves to this buffer, to facilitate backing storage negotiation, using dma_buf_attach() API. - association of a file pointer with each user-buffer and associated allocator-defined operations on that buffer. This operation is called the 'export' operation. - this exported buffer-object to be shared with the other entity by asking for its 'file-descriptor (fd)', and sharing the fd across. - a received fd to get the buffer object back, where it can be accessed using the associated exporter-defined operations. - the exporter and user to share the scatterlist using map_dma_buf and unmap_dma_buf operations.
Documentation present in the patch-set gives more details.
This is based on design suggestions from many people at the mini-summits, most notably from Arnd Bergmann arnd@arndb.de, Rob Clark rob@ti.com and Daniel Vetter daniel@ffwll.ch.
The implementation is inspired from proof-of-concept patch-set from Tomasz Stanislawski t.stanislaws@samsung.com, who demonstrated buffer sharing between two v4l2 devices. [2]
References: [1]: https://wiki.linaro.org/OfficeofCTO/MemoryManagement [2]: http://lwn.net/Articles/454389
Patchset based on top of 3.2-rc3, the current version can be found at
http://git.linaro.org/gitweb?p=people/sumitsemwal/linux-3.x.git Branch: dma-buf-upstr-v2
Earlier versions: v2 at: https://lkml.org/lkml/2011/12/2/53 v1 at: https://lkml.org/lkml/2011/10/11/92
Best regards, ~Sumit Semwal
History:
v3: - Review comments incorporated: - from Konrad Rzeszutek Wilk [https://lkml.org/lkml/2011/12/3/45] - replaced BUG_ON with WARN_ON - various places - added some error-checks - replaced EXPORT_SYMBOL with EXPORT_SYMBOL_GPL - some cosmetic / documentation comments
- from Arnd Bergmann, Daniel Vetter, Rob Clark [https://lkml.org/lkml/2011/12/5/321] - removed mmap() fop and dma_buf_op, also the sg_sync* operations, and documented that mmap is not allowed for exported buffer - updated documentation to clearly state when migration is allowed - changed kconfig - some error code checks
- from Rob Clark [https://lkml.org/lkml/2011/12/5/572] - update documentation to allow map_dma_buf to return -EINTR
v2: - Review comments incorporated: - from Tomasz Stanislawski [https://lkml.org/lkml/2011/10/14/136] - kzalloc moved out of critical section - corrected some in-code comments
- from Dave Airlie [https://lkml.org/lkml/2011/11/25/123]
- from Daniel Vetter and Rob Clark [https://lkml.org/lkml/2011/11/26/53] - use struct sg_table in place of struct scatterlist - rename {get,put}_scatterlist to {map,unmap}_dma_buf - add new wrapper APIs dma_buf_{map,unmap}_attachment for ease of users
- documentation updates as per review comments from Randy Dunlap [https://lkml.org/lkml/2011/10/12/439]
v1: original
Sumit Semwal (2): dma-buf: Introduce dma buffer sharing mechanism dma-buf: Documentation for buffer sharing framework
Documentation/dma-buf-sharing.txt | 222 ++++++++++++++++++++++++++++ drivers/base/Kconfig | 10 ++ drivers/base/Makefile | 1 + drivers/base/dma-buf.c | 289 +++++++++++++++++++++++++++++++++++++ include/linux/dma-buf.h | 172 ++++++++++++++++++++++ 5 files changed, 694 insertions(+), 0 deletions(-) create mode 100644 Documentation/dma-buf-sharing.txt create mode 100644 drivers/base/dma-buf.c create mode 100644 include/linux/dma-buf.h
Hello Everyone,
This is RFC v3 for DMA buffer sharing mechanism - changes from v2 are in the changelog below.
Various subsystems - V4L2, GPU-accessors, DRI to name a few - have felt the need to have a common mechanism to share memory buffers across different devices - ARM, video hardware, GPU.
This need comes forth from a variety of use cases including cameras, image processing, video recorders, sound processing, DMA engines, GPU and display buffers, and others.
This RFC is an attempt to define such a buffer sharing mechanism- it is the result of discussions from a couple of memory-management mini-summits held by Linaro to understand and address common needs around memory management. [1]
A new dma_buf buffer object is added, with operations and API to allow easy sharing of this buffer object across devices.
The framework allows: - a new buffer-object to be created with fixed size. - different devices to 'attach' themselves to this buffer, to facilitate backing storage negotiation, using dma_buf_attach() API. - association of a file pointer with each user-buffer and associated allocator-defined operations on that buffer. This operation is called the 'export' operation. - this exported buffer-object to be shared with the other entity by asking for its 'file-descriptor (fd)', and sharing the fd across. - a received fd to get the buffer object back, where it can be accessed using the associated exporter-defined operations. - the exporter and user to share the scatterlist using map_dma_buf and unmap_dma_buf operations.
Documentation present in the patch-set gives more details.
This is based on design suggestions from many people at the mini-summits, most notably from Arnd Bergmann arnd@arndb.de, Rob Clark rob@ti.com and Daniel Vetter daniel@ffwll.ch.
The implementation is inspired from proof-of-concept patch-set from Tomasz Stanislawski t.stanislaws@samsung.com, who demonstrated buffer sharing between two v4l2 devices. [2]
Some sample implementations and WIP for dma-buf as user and exporter are available at [3] and [4]. [These are not being submitted for discussion / inclusion right now, but are for reference only]
References: [1]: https://wiki.linaro.org/OfficeofCTO/MemoryManagement [2]: http://lwn.net/Articles/454389 [3]: Dave Airlie's: http://cgit.freedesktop.org/~airlied/linux/log/?h=drm-prime-dmabuf [4]: Rob Clark's: https://github.com/robclark/kernel-omap4/commits/dmabuf
Patchset based on top of 3.2-rc3, the current version can be found at
http://git.linaro.org/gitweb?p=people/sumitsemwal/linux-3.x.git Branch: dma-buf-upstr-v3
Earlier versions: v2 at: https://lkml.org/lkml/2011/12/2/53 v1 at: https://lkml.org/lkml/2011/10/11/92
Best regards, ~Sumit Semwal
History:
v3: - Review comments incorporated: - from Konrad Rzeszutek Wilk [https://lkml.org/lkml/2011/12/3/45] - replaced BUG_ON with WARN_ON - various places - added some error-checks - replaced EXPORT_SYMBOL with EXPORT_SYMBOL_GPL - some cosmetic / documentation comments
- from Arnd Bergmann, Daniel Vetter, Rob Clark [https://lkml.org/lkml/2011/12/5/321] - removed mmap() fop and dma_buf_op, also the sg_sync* operations, and documented that mmap is not allowed for exported buffer - updated documentation to clearly state when migration is allowed - changed kconfig - some error code checks
- from Rob Clark [https://lkml.org/lkml/2011/12/5/572] - update documentation to allow map_dma_buf to return -EINTR
v2: - Review comments incorporated: - from Tomasz Stanislawski [https://lkml.org/lkml/2011/10/14/136] - kzalloc moved out of critical section - corrected some in-code comments
- from Dave Airlie [https://lkml.org/lkml/2011/11/25/123]
- from Daniel Vetter and Rob Clark [https://lkml.org/lkml/2011/11/26/53] - use struct sg_table in place of struct scatterlist - rename {get,put}_scatterlist to {map,unmap}_dma_buf - add new wrapper APIs dma_buf_{map,unmap}_attachment for ease of users
- documentation updates as per review comments from Randy Dunlap [https://lkml.org/lkml/2011/10/12/439]
v1: original
Sumit Semwal (2): dma-buf: Introduce dma buffer sharing mechanism dma-buf: Documentation for buffer sharing framework
Documentation/dma-buf-sharing.txt | 222 ++++++++++++++++++++++++++++ drivers/base/Kconfig | 10 ++ drivers/base/Makefile | 1 + drivers/base/dma-buf.c | 289 +++++++++++++++++++++++++++++++++++++ include/linux/dma-buf.h | 172 ++++++++++++++++++++++ 5 files changed, 694 insertions(+), 0 deletions(-) create mode 100644 Documentation/dma-buf-sharing.txt create mode 100644 drivers/base/dma-buf.c create mode 100644 include/linux/dma-buf.h
This is the first step in defining a dma buffer sharing mechanism.
A new buffer object dma_buf is added, with operations and API to allow easy sharing of this buffer object across devices.
The framework allows: - different devices to 'attach' themselves to this buffer, to facilitate backing storage negotiation, using dma_buf_attach() API. - association of a file pointer with each user-buffer and associated allocator-defined operations on that buffer. This operation is called the 'export' operation. - this exported buffer-object to be shared with the other entity by asking for its 'file-descriptor (fd)', and sharing the fd across. - a received fd to get the buffer object back, where it can be accessed using the associated exporter-defined operations. - the exporter and user to share the scatterlist using map_dma_buf and unmap_dma_buf operations.
Atleast one 'attach()' call is required to be made prior to calling the map_dma_buf() operation.
Couple of building blocks in map_dma_buf() are added to ease introduction of sync'ing across exporter and users, and late allocation by the exporter.
More details are there in the documentation patch.
This is based on design suggestions from many people at the mini-summits[1], most notably from Arnd Bergmann arnd@arndb.de, Rob Clark rob@ti.com and Daniel Vetter daniel@ffwll.ch.
The implementation is inspired from proof-of-concept patch-set from Tomasz Stanislawski t.stanislaws@samsung.com, who demonstrated buffer sharing between two v4l2 devices. [2]
[1]: https://wiki.linaro.org/OfficeofCTO/MemoryManagement [2]: http://lwn.net/Articles/454389
Signed-off-by: Sumit Semwal sumit.semwal@linaro.org Signed-off-by: Sumit Semwal sumit.semwal@ti.com --- drivers/base/Kconfig | 10 ++ drivers/base/Makefile | 1 + drivers/base/dma-buf.c | 289 +++++++++++++++++++++++++++++++++++++++++++++++ include/linux/dma-buf.h | 172 ++++++++++++++++++++++++++++ 4 files changed, 472 insertions(+), 0 deletions(-) create mode 100644 drivers/base/dma-buf.c create mode 100644 include/linux/dma-buf.h
diff --git a/drivers/base/Kconfig b/drivers/base/Kconfig index 21cf46f..8a0e87f 100644 --- a/drivers/base/Kconfig +++ b/drivers/base/Kconfig @@ -174,4 +174,14 @@ config SYS_HYPERVISOR
source "drivers/base/regmap/Kconfig"
+config DMA_SHARED_BUFFER + bool "Buffer framework to be shared between drivers" + default n + select ANON_INODES + help + This option enables the framework for buffer-sharing between + multiple drivers. A buffer is associated with a file using driver + APIs extension; the file's descriptor can then be passed on to other + driver. + endmenu diff --git a/drivers/base/Makefile b/drivers/base/Makefile index 99a375a..d0df046 100644 --- a/drivers/base/Makefile +++ b/drivers/base/Makefile @@ -8,6 +8,7 @@ obj-$(CONFIG_DEVTMPFS) += devtmpfs.o obj-y += power/ obj-$(CONFIG_HAS_DMA) += dma-mapping.o obj-$(CONFIG_HAVE_GENERIC_DMA_COHERENT) += dma-coherent.o +obj-$(CONFIG_DMA_SHARED_BUFFER) += dma-buf.o obj-$(CONFIG_ISA) += isa.o obj-$(CONFIG_FW_LOADER) += firmware_class.o obj-$(CONFIG_NUMA) += node.o diff --git a/drivers/base/dma-buf.c b/drivers/base/dma-buf.c new file mode 100644 index 0000000..e920709 --- /dev/null +++ b/drivers/base/dma-buf.c @@ -0,0 +1,289 @@ +/* + * Framework for buffer objects that can be shared across devices/subsystems. + * + * Copyright(C) 2011 Linaro Limited. All rights reserved. + * Author: Sumit Semwal sumit.semwal@ti.com + * + * Many thanks to linaro-mm-sig list, and specially + * Arnd Bergmann arnd@arndb.de, Rob Clark rob@ti.com and + * Daniel Vetter daniel@ffwll.ch for their support in creation and + * refining of this idea. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published by + * the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see http://www.gnu.org/licenses/. + */ + +#include <linux/fs.h> +#include <linux/slab.h> +#include <linux/dma-buf.h> +#include <linux/anon_inodes.h> +#include <linux/export.h> + +static inline int is_dma_buf_file(struct file *); + +static int dma_buf_release(struct inode *inode, struct file *file) +{ + struct dma_buf *dmabuf; + + if (!is_dma_buf_file(file)) + return -EINVAL; + + dmabuf = file->private_data; + + dmabuf->ops->release(dmabuf); + kfree(dmabuf); + return 0; +} + +static const struct file_operations dma_buf_fops = { + .release = dma_buf_release, +}; + +/* + * is_dma_buf_file - Check if struct file* is associated with dma_buf + */ +static inline int is_dma_buf_file(struct file *file) +{ + return file->f_op == &dma_buf_fops; +} + +/** + * dma_buf_export - Creates a new dma_buf, and associates an anon file + * with this buffer, so it can be exported. + * Also connect the allocator specific data and ops to the buffer. + * + * @priv: [in] Attach private data of allocator to this buffer + * @ops: [in] Attach allocator-defined dma buf ops to the new buffer. + * @size: [in] Size of the buffer + * @flags: [in] mode flags for the file. + * + * Returns, on success, a newly created dma_buf object, which wraps the + * supplied private data and operations for dma_buf_ops. On either missing + * ops, or error in allocating struct dma_buf, will return negative error. + * + */ +struct dma_buf *dma_buf_export(void *priv, struct dma_buf_ops *ops, + size_t size, int flags) +{ + struct dma_buf *dmabuf; + struct file *file; + + if (WARN_ON(!priv || !ops + || !ops->map_dma_buf + || !ops->unmap_dma_buf + || !ops->release)) { + return ERR_PTR(-EINVAL); + } + + dmabuf = kzalloc(sizeof(struct dma_buf), GFP_KERNEL); + if (dmabuf == NULL) + return ERR_PTR(-ENOMEM); + + dmabuf->priv = priv; + dmabuf->ops = ops; + dmabuf->size = size; + + file = anon_inode_getfile("dmabuf", &dma_buf_fops, dmabuf, flags); + + dmabuf->file = file; + + mutex_init(&dmabuf->lock); + INIT_LIST_HEAD(&dmabuf->attachments); + + return dmabuf; +} +EXPORT_SYMBOL_GPL(dma_buf_export); + + +/** + * dma_buf_fd - returns a file descriptor for the given dma_buf + * @dmabuf: [in] pointer to dma_buf for which fd is required. + * + * On success, returns an associated 'fd'. Else, returns error. + */ +int dma_buf_fd(struct dma_buf *dmabuf) +{ + int error, fd; + + if (!dmabuf || !dmabuf->file) + return -EINVAL; + + error = get_unused_fd(); + if (error < 0) + return error; + fd = error; + + fd_install(fd, dmabuf->file); + + return fd; +} +EXPORT_SYMBOL_GPL(dma_buf_fd); + +/** + * dma_buf_get - returns the dma_buf structure related to an fd + * @fd: [in] fd associated with the dma_buf to be returned + * + * On success, returns the dma_buf structure associated with an fd; uses + * file's refcounting done by fget to increase refcount. returns ERR_PTR + * otherwise. + */ +struct dma_buf *dma_buf_get(int fd) +{ + struct file *file; + + file = fget(fd); + + if (!file) + return ERR_PTR(-EBADF); + + if (!is_dma_buf_file(file)) { + fput(file); + return ERR_PTR(-EINVAL); + } + + return file->private_data; +} +EXPORT_SYMBOL_GPL(dma_buf_get); + +/** + * dma_buf_put - decreases refcount of the buffer + * @dmabuf: [in] buffer to reduce refcount of + * + * Uses file's refcounting done implicitly by fput() + */ +void dma_buf_put(struct dma_buf *dmabuf) +{ + if (WARN_ON(!dmabuf || !dmabuf->file)) + return; + + fput(dmabuf->file); +} +EXPORT_SYMBOL_GPL(dma_buf_put); + +/** + * dma_buf_attach - Add the device to dma_buf's attachments list; optionally, + * calls attach() of dma_buf_ops to allow device-specific attach functionality + * @dmabuf: [in] buffer to attach device to. + * @dev: [in] device to be attached. + * + * Returns struct dma_buf_attachment * for this attachment; may return negative + * error codes. + * + */ +struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf, + struct device *dev) +{ + struct dma_buf_attachment *attach; + int ret; + + if (WARN_ON(!dmabuf || !dev || !dmabuf->ops)) + return ERR_PTR(-EINVAL); + + attach = kzalloc(sizeof(struct dma_buf_attachment), GFP_KERNEL); + if (attach == NULL) + goto err_alloc; + + mutex_lock(&dmabuf->lock); + + attach->dev = dev; + attach->dmabuf = dmabuf; + if (dmabuf->ops->attach) { + ret = dmabuf->ops->attach(dmabuf, dev, attach); + if (ret) + goto err_attach; + } + list_add(&attach->node, &dmabuf->attachments); + + mutex_unlock(&dmabuf->lock); + return attach; + +err_alloc: + return ERR_PTR(-ENOMEM); +err_attach: + kfree(attach); + mutex_unlock(&dmabuf->lock); + return ERR_PTR(ret); +} +EXPORT_SYMBOL_GPL(dma_buf_attach); + +/** + * dma_buf_detach - Remove the given attachment from dmabuf's attachments list; + * optionally calls detach() of dma_buf_ops for device-specific detach + * @dmabuf: [in] buffer to detach from. + * @attach: [in] attachment to be detached; is free'd after this call. + * + */ +void dma_buf_detach(struct dma_buf *dmabuf, struct dma_buf_attachment *attach) +{ + if (WARN_ON(!dmabuf || !attach || !dmabuf->ops)) + return; + + mutex_lock(&dmabuf->lock); + list_del(&attach->node); + if (dmabuf->ops->detach) + dmabuf->ops->detach(dmabuf, attach); + + mutex_unlock(&dmabuf->lock); + kfree(attach); +} +EXPORT_SYMBOL_GPL(dma_buf_detach); + +/** + * dma_buf_map_attachment - Returns the scatterlist table of the attachment; + * mapped into _device_ address space. Is a wrapper for map_dma_buf() of the + * dma_buf_ops. + * @attach: [in] attachment whose scatterlist is to be returned + * @direction: [in] direction of DMA transfer + * + * Returns sg_table containing the scatterlist to be returned; may return NULL + * or ERR_PTR. + * + */ +struct sg_table *dma_buf_map_attachment(struct dma_buf_attachment *attach, + enum dma_data_direction direction) +{ + struct sg_table *sg_table = ERR_PTR(-EINVAL); + + if (WARN_ON(!attach || !attach->dmabuf || !attach->dmabuf->ops)) + return ERR_PTR(-EINVAL); + + mutex_lock(&attach->dmabuf->lock); + if (attach->dmabuf->ops->map_dma_buf) + sg_table = attach->dmabuf->ops->map_dma_buf(attach, direction); + mutex_unlock(&attach->dmabuf->lock); + + return sg_table; +} +EXPORT_SYMBOL_GPL(dma_buf_map_attachment); + +/** + * dma_buf_unmap_attachment - unmaps and decreases usecount of the buffer;might + * deallocate the scatterlist associated. Is a wrapper for unmap_dma_buf() of + * dma_buf_ops. + * @attach: [in] attachment to unmap buffer from + * @sg_table: [in] scatterlist info of the buffer to unmap + * + */ +void dma_buf_unmap_attachment(struct dma_buf_attachment *attach, + struct sg_table *sg_table) +{ + if (WARN_ON(!attach || !attach->dmabuf || !sg_table + || !attach->dmabuf->ops)) + return; + + mutex_lock(&attach->dmabuf->lock); + if (attach->dmabuf->ops->unmap_dma_buf) + attach->dmabuf->ops->unmap_dma_buf(attach, sg_table); + mutex_unlock(&attach->dmabuf->lock); + +} +EXPORT_SYMBOL_GPL(dma_buf_unmap_attachment); diff --git a/include/linux/dma-buf.h b/include/linux/dma-buf.h new file mode 100644 index 0000000..3e3ecf3 --- /dev/null +++ b/include/linux/dma-buf.h @@ -0,0 +1,172 @@ +/* + * Header file for dma buffer sharing framework. + * + * Copyright(C) 2011 Linaro Limited. All rights reserved. + * Author: Sumit Semwal sumit.semwal@ti.com + * + * Many thanks to linaro-mm-sig list, and specially + * Arnd Bergmann arnd@arndb.de, Rob Clark rob@ti.com and + * Daniel Vetter daniel@ffwll.ch for their support in creation and + * refining of this idea. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published by + * the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see http://www.gnu.org/licenses/. + */ +#ifndef __DMA_BUF_H__ +#define __DMA_BUF_H__ + +#include <linux/file.h> +#include <linux/err.h> +#include <linux/device.h> +#include <linux/scatterlist.h> +#include <linux/list.h> +#include <linux/dma-mapping.h> + +struct dma_buf; + +/** + * struct dma_buf_attachment - holds device-buffer attachment data + * @dmabuf: buffer for this attachment. + * @dev: device attached to the buffer. + * @node: list_head to allow manipulation of list of dma_buf_attachment. + * @priv: exporter-specific attachment data. + */ +struct dma_buf_attachment { + struct dma_buf *dmabuf; + struct device *dev; + struct list_head node; + void *priv; +}; + +/** + * struct dma_buf_ops - operations possible on struct dma_buf + * @attach: allows different devices to 'attach' themselves to the given + * buffer. It might return -EBUSY to signal that backing storage + * is already allocated and incompatible with the requirements + * of requesting device. [optional] + * @detach: detach a given device from this buffer. [optional] + * @map_dma_buf: returns list of scatter pages allocated, increases usecount + * of the buffer. Requires atleast one attach to be called + * before. Returned sg list should already be mapped into + * _device_ address space. This call may sleep. May also return + * -EINTR. + * @unmap_dma_buf: decreases usecount of buffer, might deallocate scatter + * pages. + * @release: release this buffer; to be called after the last dma_buf_put. + * @sync_sg_for_cpu: sync the sg list for cpu. + * @sync_sg_for_device: synch the sg list for device. + */ +struct dma_buf_ops { + int (*attach)(struct dma_buf *, struct device *, + struct dma_buf_attachment *); + + void (*detach)(struct dma_buf *, struct dma_buf_attachment *); + + /* For {map,unmap}_dma_buf below, any specific buffer attributes + * required should get added to device_dma_parameters accessible + * via dev->dma_params. + */ + struct sg_table * (*map_dma_buf)(struct dma_buf_attachment *, + enum dma_data_direction); + void (*unmap_dma_buf)(struct dma_buf_attachment *, + struct sg_table *); + /* TODO: Add try_map_dma_buf version, to return immed with -EBUSY + * if the call would block. + */ + + /* after final dma_buf_put() */ + void (*release)(struct dma_buf *); + +}; + +/** + * struct dma_buf - shared buffer object + * @file: file pointer used for sharing buffers across, and for refcounting. + * @attachments: list of dma_buf_attachment that denotes all devices attached. + * @ops: dma_buf_ops associated with this buffer object + * @priv: user specific private data + */ +struct dma_buf { + size_t size; + struct file *file; + struct list_head attachments; + const struct dma_buf_ops *ops; + /* mutex to serialize list manipulation and other ops */ + struct mutex lock; + void *priv; +}; + +#ifdef CONFIG_DMA_SHARED_BUFFER +struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf, + struct device *dev); +void dma_buf_detach(struct dma_buf *dmabuf, + struct dma_buf_attachment *dmabuf_attach); +struct dma_buf *dma_buf_export(void *priv, struct dma_buf_ops *ops, + size_t size, int flags); +int dma_buf_fd(struct dma_buf *dmabuf); +struct dma_buf *dma_buf_get(int fd); +void dma_buf_put(struct dma_buf *dmabuf); + +struct sg_table *dma_buf_map_attachment(struct dma_buf_attachment *, + enum dma_data_direction); +void dma_buf_unmap_attachment(struct dma_buf_attachment *, struct sg_table *); +#else + +static inline struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf, + struct device *dev) +{ + return ERR_PTR(-ENODEV); +} + +static inline void dma_buf_detach(struct dma_buf *dmabuf, + struct dma_buf_attachment *dmabuf_attach) +{ + return; +} + +static inline struct dma_buf *dma_buf_export(void *priv, + struct dma_buf_ops *ops, + size_t size, int flags) +{ + return ERR_PTR(-ENODEV); +} + +static inline int dma_buf_fd(struct dma_buf *dmabuf) +{ + return -ENODEV; +} + +static inline struct dma_buf *dma_buf_get(int fd) +{ + return ERR_PTR(-ENODEV); +} + +static inline void dma_buf_put(struct dma_buf *dmabuf) +{ + return; +} + +static inline struct sg_table *dma_buf_map_attachment( + struct dma_buf_attachment *attach, enum dma_data_direction write) +{ + return ERR_PTR(-ENODEV); +} + +static inline void dma_buf_unmap_attachment(struct dma_buf_attachment *attach, + struct sg_table *sg) +{ + return; +} + +#endif /* CONFIG_DMA_SHARED_BUFFER */ + +#endif /* __DMA_BUF_H__ */
On Mon, Dec 19, 2011 at 02:03:30PM +0530, Sumit Semwal wrote:
This is the first step in defining a dma buffer sharing mechanism.
A new buffer object dma_buf is added, with operations and API to allow easy sharing of this buffer object across devices.
The framework allows:
- different devices to 'attach' themselves to this buffer, to facilitate backing storage negotiation, using dma_buf_attach() API.
Any thoughts of adding facility to track them? So you can see who is using what?
- association of a file pointer with each user-buffer and associated allocator-defined operations on that buffer. This operation is called the 'export' operation.
'create'? or 'alloc' ?
export implies an import somwhere and I don't think that is the case here.
- this exported buffer-object to be shared with the other entity by asking for its 'file-descriptor (fd)', and sharing the fd across.
- a received fd to get the buffer object back, where it can be accessed using the associated exporter-defined operations.
- the exporter and user to share the scatterlist using map_dma_buf and unmap_dma_buf operations.
Atleast one 'attach()' call is required to be made prior to calling the map_dma_buf() operation.
for the whole memory region or just for the device itself?
Couple of building blocks in map_dma_buf() are added to ease introduction of sync'ing across exporter and users, and late allocation by the exporter.
More details are there in the documentation patch.
This is based on design suggestions from many people at the mini-summits[1], most notably from Arnd Bergmann arnd@arndb.de, Rob Clark rob@ti.com and Daniel Vetter daniel@ffwll.ch.
The implementation is inspired from proof-of-concept patch-set from Tomasz Stanislawski t.stanislaws@samsung.com, who demonstrated buffer sharing between two v4l2 devices. [2]
Signed-off-by: Sumit Semwal sumit.semwal@linaro.org Signed-off-by: Sumit Semwal sumit.semwal@ti.com
drivers/base/Kconfig | 10 ++ drivers/base/Makefile | 1 + drivers/base/dma-buf.c | 289 +++++++++++++++++++++++++++++++++++++++++++++++ include/linux/dma-buf.h | 172 ++++++++++++++++++++++++++++ 4 files changed, 472 insertions(+), 0 deletions(-) create mode 100644 drivers/base/dma-buf.c create mode 100644 include/linux/dma-buf.h
diff --git a/drivers/base/Kconfig b/drivers/base/Kconfig index 21cf46f..8a0e87f 100644 --- a/drivers/base/Kconfig +++ b/drivers/base/Kconfig @@ -174,4 +174,14 @@ config SYS_HYPERVISOR source "drivers/base/regmap/Kconfig" +config DMA_SHARED_BUFFER
- bool "Buffer framework to be shared between drivers"
- default n
- select ANON_INODES
- help
This option enables the framework for buffer-sharing betweenmultiple drivers. A buffer is associated with a file using driverAPIs extension; the file's descriptor can then be passed on to otherdriver.endmenu diff --git a/drivers/base/Makefile b/drivers/base/Makefile index 99a375a..d0df046 100644 --- a/drivers/base/Makefile +++ b/drivers/base/Makefile @@ -8,6 +8,7 @@ obj-$(CONFIG_DEVTMPFS) += devtmpfs.o obj-y += power/ obj-$(CONFIG_HAS_DMA) += dma-mapping.o obj-$(CONFIG_HAVE_GENERIC_DMA_COHERENT) += dma-coherent.o +obj-$(CONFIG_DMA_SHARED_BUFFER) += dma-buf.o obj-$(CONFIG_ISA) += isa.o obj-$(CONFIG_FW_LOADER) += firmware_class.o obj-$(CONFIG_NUMA) += node.o diff --git a/drivers/base/dma-buf.c b/drivers/base/dma-buf.c new file mode 100644 index 0000000..e920709 --- /dev/null +++ b/drivers/base/dma-buf.c @@ -0,0 +1,289 @@ +/*
- Framework for buffer objects that can be shared across devices/subsystems.
- Copyright(C) 2011 Linaro Limited. All rights reserved.
- Author: Sumit Semwal sumit.semwal@ti.com
- Many thanks to linaro-mm-sig list, and specially
- Arnd Bergmann arnd@arndb.de, Rob Clark rob@ti.com and
- Daniel Vetter daniel@ffwll.ch for their support in creation and
- refining of this idea.
- This program is free software; you can redistribute it and/or modify it
- under the terms of the GNU General Public License version 2 as published by
- the Free Software Foundation.
- This program is distributed in the hope that it will be useful, but WITHOUT
- ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
- more details.
- You should have received a copy of the GNU General Public License along with
- this program. If not, see http://www.gnu.org/licenses/.
- */
+#include <linux/fs.h> +#include <linux/slab.h> +#include <linux/dma-buf.h> +#include <linux/anon_inodes.h> +#include <linux/export.h>
+static inline int is_dma_buf_file(struct file *);
+static int dma_buf_release(struct inode *inode, struct file *file) +{
- struct dma_buf *dmabuf;
- if (!is_dma_buf_file(file))
return -EINVAL;- dmabuf = file->private_data;
- dmabuf->ops->release(dmabuf);
- kfree(dmabuf);
- return 0;
+}
+static const struct file_operations dma_buf_fops = {
- .release = dma_buf_release,
+};
+/*
- is_dma_buf_file - Check if struct file* is associated with dma_buf
- */
+static inline int is_dma_buf_file(struct file *file) +{
- return file->f_op == &dma_buf_fops;
+}
+/**
Wrong kerneldoc.
- dma_buf_export - Creates a new dma_buf, and associates an anon file
- with this buffer, so it can be exported.
- Also connect the allocator specific data and ops to the buffer.
- @priv: [in] Attach private data of allocator to this buffer
- @ops: [in] Attach allocator-defined dma buf ops to the new buffer.
- @size: [in] Size of the buffer
- @flags: [in] mode flags for the file.
- Returns, on success, a newly created dma_buf object, which wraps the
- supplied private data and operations for dma_buf_ops. On either missing
- ops, or error in allocating struct dma_buf, will return negative error.
- */
+struct dma_buf *dma_buf_export(void *priv, struct dma_buf_ops *ops,
size_t size, int flags)+{
- struct dma_buf *dmabuf;
- struct file *file;
- if (WARN_ON(!priv || !ops
|| !ops->map_dma_buf|| !ops->unmap_dma_buf|| !ops->release)) {return ERR_PTR(-EINVAL);- }
- dmabuf = kzalloc(sizeof(struct dma_buf), GFP_KERNEL);
- if (dmabuf == NULL)
return ERR_PTR(-ENOMEM);- dmabuf->priv = priv;
- dmabuf->ops = ops;
- dmabuf->size = size;
- file = anon_inode_getfile("dmabuf", &dma_buf_fops, dmabuf, flags);
- dmabuf->file = file;
- mutex_init(&dmabuf->lock);
- INIT_LIST_HEAD(&dmabuf->attachments);
- return dmabuf;
+} +EXPORT_SYMBOL_GPL(dma_buf_export);
+/**
- dma_buf_fd - returns a file descriptor for the given dma_buf
- @dmabuf: [in] pointer to dma_buf for which fd is required.
- On success, returns an associated 'fd'. Else, returns error.
- */
+int dma_buf_fd(struct dma_buf *dmabuf) +{
- int error, fd;
- if (!dmabuf || !dmabuf->file)
return -EINVAL;- error = get_unused_fd();
- if (error < 0)
return error;- fd = error;
- fd_install(fd, dmabuf->file);
- return fd;
+} +EXPORT_SYMBOL_GPL(dma_buf_fd);
+/**
- dma_buf_get - returns the dma_buf structure related to an fd
- @fd: [in] fd associated with the dma_buf to be returned
- On success, returns the dma_buf structure associated with an fd; uses
- file's refcounting done by fget to increase refcount. returns ERR_PTR
- otherwise.
- */
+struct dma_buf *dma_buf_get(int fd) +{
- struct file *file;
- file = fget(fd);
- if (!file)
return ERR_PTR(-EBADF);- if (!is_dma_buf_file(file)) {
fput(file);return ERR_PTR(-EINVAL);- }
- return file->private_data;
+} +EXPORT_SYMBOL_GPL(dma_buf_get);
+/**
- dma_buf_put - decreases refcount of the buffer
- @dmabuf: [in] buffer to reduce refcount of
- Uses file's refcounting done implicitly by fput()
- */
+void dma_buf_put(struct dma_buf *dmabuf) +{
- if (WARN_ON(!dmabuf || !dmabuf->file))
return;- fput(dmabuf->file);
+} +EXPORT_SYMBOL_GPL(dma_buf_put);
+/**
- dma_buf_attach - Add the device to dma_buf's attachments list; optionally,
- calls attach() of dma_buf_ops to allow device-specific attach functionality
- @dmabuf: [in] buffer to attach device to.
- @dev: [in] device to be attached.
- Returns struct dma_buf_attachment * for this attachment; may return negative
- error codes.
- */
+struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf,
struct device *dev)+{
- struct dma_buf_attachment *attach;
- int ret;
- if (WARN_ON(!dmabuf || !dev || !dmabuf->ops))
return ERR_PTR(-EINVAL);- attach = kzalloc(sizeof(struct dma_buf_attachment), GFP_KERNEL);
- if (attach == NULL)
goto err_alloc;- mutex_lock(&dmabuf->lock);
- attach->dev = dev;
- attach->dmabuf = dmabuf;
- if (dmabuf->ops->attach) {
ret = dmabuf->ops->attach(dmabuf, dev, attach);if (ret)goto err_attach;- }
- list_add(&attach->node, &dmabuf->attachments);
- mutex_unlock(&dmabuf->lock);
- return attach;
+err_alloc:
- return ERR_PTR(-ENOMEM);
+err_attach:
- kfree(attach);
- mutex_unlock(&dmabuf->lock);
- return ERR_PTR(ret);
+} +EXPORT_SYMBOL_GPL(dma_buf_attach);
+/**
- dma_buf_detach - Remove the given attachment from dmabuf's attachments list;
- optionally calls detach() of dma_buf_ops for device-specific detach
- @dmabuf: [in] buffer to detach from.
- @attach: [in] attachment to be detached; is free'd after this call.
- */
+void dma_buf_detach(struct dma_buf *dmabuf, struct dma_buf_attachment *attach) +{
- if (WARN_ON(!dmabuf || !attach || !dmabuf->ops))
return;- mutex_lock(&dmabuf->lock);
- list_del(&attach->node);
- if (dmabuf->ops->detach)
dmabuf->ops->detach(dmabuf, attach);- mutex_unlock(&dmabuf->lock);
- kfree(attach);
+} +EXPORT_SYMBOL_GPL(dma_buf_detach);
+/**
- dma_buf_map_attachment - Returns the scatterlist table of the attachment;
- mapped into _device_ address space. Is a wrapper for map_dma_buf() of the
- dma_buf_ops.
- @attach: [in] attachment whose scatterlist is to be returned
- @direction: [in] direction of DMA transfer
- Returns sg_table containing the scatterlist to be returned; may return NULL
- or ERR_PTR.
- */
+struct sg_table *dma_buf_map_attachment(struct dma_buf_attachment *attach,
enum dma_data_direction direction)+{
- struct sg_table *sg_table = ERR_PTR(-EINVAL);
- if (WARN_ON(!attach || !attach->dmabuf || !attach->dmabuf->ops))
return ERR_PTR(-EINVAL);- mutex_lock(&attach->dmabuf->lock);
- if (attach->dmabuf->ops->map_dma_buf)
sg_table = attach->dmabuf->ops->map_dma_buf(attach, direction);- mutex_unlock(&attach->dmabuf->lock);
- return sg_table;
+} +EXPORT_SYMBOL_GPL(dma_buf_map_attachment);
+/**
- dma_buf_unmap_attachment - unmaps and decreases usecount of the buffer;might
- deallocate the scatterlist associated. Is a wrapper for unmap_dma_buf() of
- dma_buf_ops.
- @attach: [in] attachment to unmap buffer from
- @sg_table: [in] scatterlist info of the buffer to unmap
- */
+void dma_buf_unmap_attachment(struct dma_buf_attachment *attach,
struct sg_table *sg_table)+{
- if (WARN_ON(!attach || !attach->dmabuf || !sg_table
|| !attach->dmabuf->ops))return;- mutex_lock(&attach->dmabuf->lock);
- if (attach->dmabuf->ops->unmap_dma_buf)
attach->dmabuf->ops->unmap_dma_buf(attach, sg_table);- mutex_unlock(&attach->dmabuf->lock);
+} +EXPORT_SYMBOL_GPL(dma_buf_unmap_attachment); diff --git a/include/linux/dma-buf.h b/include/linux/dma-buf.h new file mode 100644 index 0000000..3e3ecf3 --- /dev/null +++ b/include/linux/dma-buf.h @@ -0,0 +1,172 @@ +/*
- Header file for dma buffer sharing framework.
- Copyright(C) 2011 Linaro Limited. All rights reserved.
- Author: Sumit Semwal sumit.semwal@ti.com
- Many thanks to linaro-mm-sig list, and specially
- Arnd Bergmann arnd@arndb.de, Rob Clark rob@ti.com and
- Daniel Vetter daniel@ffwll.ch for their support in creation and
- refining of this idea.
- This program is free software; you can redistribute it and/or modify it
- under the terms of the GNU General Public License version 2 as published by
- the Free Software Foundation.
- This program is distributed in the hope that it will be useful, but WITHOUT
- ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
- more details.
- You should have received a copy of the GNU General Public License along with
- this program. If not, see http://www.gnu.org/licenses/.
- */
+#ifndef __DMA_BUF_H__ +#define __DMA_BUF_H__
+#include <linux/file.h> +#include <linux/err.h> +#include <linux/device.h> +#include <linux/scatterlist.h> +#include <linux/list.h> +#include <linux/dma-mapping.h>
+struct dma_buf;
+/**
- struct dma_buf_attachment - holds device-buffer attachment data
OK, but what is the purpose of it?
- @dmabuf: buffer for this attachment.
- @dev: device attached to the buffer.
^^^ this
- @node: list_head to allow manipulation of list of dma_buf_attachment.
Just say: "list of dma_buf_attachment"'
- @priv: exporter-specific attachment data.
That "exporter-specific.." brings to my mind custom decleration forms. But maybe that is me.
- */
+struct dma_buf_attachment {
- struct dma_buf *dmabuf;
- struct device *dev;
- struct list_head node;
- void *priv;
+};
Why don't you move the decleration of this below 'struct dma_buf'? It would easier than to read this structure..
+/**
- struct dma_buf_ops - operations possible on struct dma_buf
- @attach: allows different devices to 'attach' themselves to the given
register?
buffer. It might return -EBUSY to signal that backing storage
is already allocated and incompatible with the requirements
Wait.. allocated or attached?
of requesting device. [optional]
What is optional? The return value? Or the 'attach' call? If the later , say that in the first paragraph.
- @detach: detach a given device from this buffer. [optional]
- @map_dma_buf: returns list of scatter pages allocated, increases usecount
of the buffer. Requires atleast one attach to be called
before. Returned sg list should already be mapped into
_device_ address space. This call may sleep. May also return
Ok, there is some __might_sleep macro you should put on the function.
-EINTR.
Ok. What is the return code if attach has _not_ been called?
- @unmap_dma_buf: decreases usecount of buffer, might deallocate scatter
pages.
- @release: release this buffer; to be called after the last dma_buf_put.
- @sync_sg_for_cpu: sync the sg list for cpu.
- @sync_sg_for_device: synch the sg list for device.
Not seeing those two.
- */
+struct dma_buf_ops {
- int (*attach)(struct dma_buf *, struct device *,
struct dma_buf_attachment *);- void (*detach)(struct dma_buf *, struct dma_buf_attachment *);
- /* For {map,unmap}_dma_buf below, any specific buffer attributes
* required should get added to device_dma_parameters accessible* via dev->dma_params.*/- struct sg_table * (*map_dma_buf)(struct dma_buf_attachment *,
enum dma_data_direction);- void (*unmap_dma_buf)(struct dma_buf_attachment *,
struct sg_table *);- /* TODO: Add try_map_dma_buf version, to return immed with -EBUSY
Ewww. Why? Why not just just the 'map_dma_buf' and return that?
* if the call would block.*/- /* after final dma_buf_put() */
- void (*release)(struct dma_buf *);
+};
+/**
- struct dma_buf - shared buffer object
Missing the 'size'.
- @file: file pointer used for sharing buffers across, and for refcounting.
- @attachments: list of dma_buf_attachment that denotes all devices attached.
- @ops: dma_buf_ops associated with this buffer object
- @priv: user specific private data
Can you elaborate on this? Is this the "exporter" using this? Or is it for the "user" using it? If so, why provide it? Wouldn't the user of this have something like this:
struct my_dma_bufs { struct dma_buf[20]; void *priv; }
Anyhow?
- */
+struct dma_buf {
- size_t size;
- struct file *file;
- struct list_head attachments;
- const struct dma_buf_ops *ops;
- /* mutex to serialize list manipulation and other ops */
- struct mutex lock;
- void *priv;
+};
+#ifdef CONFIG_DMA_SHARED_BUFFER +struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf,
struct device *dev);+void dma_buf_detach(struct dma_buf *dmabuf,
struct dma_buf_attachment *dmabuf_attach);+struct dma_buf *dma_buf_export(void *priv, struct dma_buf_ops *ops,
size_t size, int flags);+int dma_buf_fd(struct dma_buf *dmabuf); +struct dma_buf *dma_buf_get(int fd); +void dma_buf_put(struct dma_buf *dmabuf);
+struct sg_table *dma_buf_map_attachment(struct dma_buf_attachment *,
enum dma_data_direction);+void dma_buf_unmap_attachment(struct dma_buf_attachment *, struct sg_table *); +#else
+static inline struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf,
struct device *dev)+{
- return ERR_PTR(-ENODEV);
+}
+static inline void dma_buf_detach(struct dma_buf *dmabuf,
struct dma_buf_attachment *dmabuf_attach)+{
- return;
+}
+static inline struct dma_buf *dma_buf_export(void *priv,
struct dma_buf_ops *ops,size_t size, int flags)+{
- return ERR_PTR(-ENODEV);
+}
+static inline int dma_buf_fd(struct dma_buf *dmabuf) +{
- return -ENODEV;
+}
+static inline struct dma_buf *dma_buf_get(int fd) +{
- return ERR_PTR(-ENODEV);
+}
+static inline void dma_buf_put(struct dma_buf *dmabuf) +{
- return;
+}
+static inline struct sg_table *dma_buf_map_attachment(
- struct dma_buf_attachment *attach, enum dma_data_direction write)
+{
- return ERR_PTR(-ENODEV);
+}
+static inline void dma_buf_unmap_attachment(struct dma_buf_attachment *attach,
struct sg_table *sg)+{
- return;
+}
+#endif /* CONFIG_DMA_SHARED_BUFFER */
+#endif /* __DMA_BUF_H__ */
1.7.4.1
dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
On Tue, Dec 20, 2011 at 10:36 AM, Konrad Rzeszutek Wilk konrad.wilk@oracle.com wrote:
On Mon, Dec 19, 2011 at 02:03:30PM +0530, Sumit Semwal wrote:
This is the first step in defining a dma buffer sharing mechanism.
A new buffer object dma_buf is added, with operations and API to allow easy sharing of this buffer object across devices.
The framework allows:
- different devices to 'attach' themselves to this buffer, to facilitate
backing storage negotiation, using dma_buf_attach() API.
Any thoughts of adding facility to track them? So you can see who is using what?
- association of a file pointer with each user-buffer and associated
allocator-defined operations on that buffer. This operation is called the 'export' operation.
'create'? or 'alloc' ?
export implies an import somwhere and I don't think that is the case here.
Well, the import is the attach/map stuff. The userspace facing export ioctl would call dma_buf_export().. the userspace facing import ioctl of the importing device would call dma_buf_attach/map().
Perhaps the language should be a "create dma buf" API used by the exporting driver. I guess the dmabuf is being created, the backing memory/pages are being exported.
- this exported buffer-object to be shared with the other entity by asking for
its 'file-descriptor (fd)', and sharing the fd across.
- a received fd to get the buffer object back, where it can be accessed using
the associated exporter-defined operations.
- the exporter and user to share the scatterlist using map_dma_buf and
unmap_dma_buf operations.
Atleast one 'attach()' call is required to be made prior to calling the map_dma_buf() operation.
for the whole memory region or just for the device itself?
Attach should be called per buffer. It is basically registering intent of the importing device to user the buffer 1 or more times. Detach is letting the exporter know that the importer won't be using the buffer again (at least for a while).
For example: attach() map() // buffer is backed and pinned do dma to/from buffer unmap() // exporter could mark the buffer evictable here map() // buffer is pinned do dma to/from buffer unmap() ... detach()
Also, in case of multiple importers, the exporter could also wait to allocate backing pages until first map() in case one of the importers has stricter memory requirements (physically contiguous, certain memory range, etc) than the others. Some exporters, like v4l, might have some restrictions about the buffer being backed/pinned earlier, for those attach() might be basically a no-op. But the idea is to give exporters that are a bit more dynamic w/ memory management some flexibility.
BR, -R
Couple of building blocks in map_dma_buf() are added to ease introduction of sync'ing across exporter and users, and late allocation by the exporter.
More details are there in the documentation patch.
This is based on design suggestions from many people at the mini-summits[1], most notably from Arnd Bergmann arnd@arndb.de, Rob Clark rob@ti.com and Daniel Vetter daniel@ffwll.ch.
The implementation is inspired from proof-of-concept patch-set from Tomasz Stanislawski t.stanislaws@samsung.com, who demonstrated buffer sharing between two v4l2 devices. [2]
Signed-off-by: Sumit Semwal sumit.semwal@linaro.org Signed-off-by: Sumit Semwal sumit.semwal@ti.com
drivers/base/Kconfig | 10 ++ drivers/base/Makefile | 1 + drivers/base/dma-buf.c | 289 +++++++++++++++++++++++++++++++++++++++++++++++ include/linux/dma-buf.h | 172 ++++++++++++++++++++++++++++ 4 files changed, 472 insertions(+), 0 deletions(-) create mode 100644 drivers/base/dma-buf.c create mode 100644 include/linux/dma-buf.h
diff --git a/drivers/base/Kconfig b/drivers/base/Kconfig index 21cf46f..8a0e87f 100644 --- a/drivers/base/Kconfig +++ b/drivers/base/Kconfig @@ -174,4 +174,14 @@ config SYS_HYPERVISOR
source "drivers/base/regmap/Kconfig"
+config DMA_SHARED_BUFFER
- bool "Buffer framework to be shared between drivers"
- default n
- select ANON_INODES
- help
- This option enables the framework for buffer-sharing between
- multiple drivers. A buffer is associated with a file using driver
- APIs extension; the file's descriptor can then be passed on to other
- driver.
endmenu diff --git a/drivers/base/Makefile b/drivers/base/Makefile index 99a375a..d0df046 100644 --- a/drivers/base/Makefile +++ b/drivers/base/Makefile @@ -8,6 +8,7 @@ obj-$(CONFIG_DEVTMPFS) += devtmpfs.o obj-y += power/ obj-$(CONFIG_HAS_DMA) += dma-mapping.o obj-$(CONFIG_HAVE_GENERIC_DMA_COHERENT) += dma-coherent.o +obj-$(CONFIG_DMA_SHARED_BUFFER) += dma-buf.o obj-$(CONFIG_ISA) += isa.o obj-$(CONFIG_FW_LOADER) += firmware_class.o obj-$(CONFIG_NUMA) += node.o diff --git a/drivers/base/dma-buf.c b/drivers/base/dma-buf.c new file mode 100644 index 0000000..e920709 --- /dev/null +++ b/drivers/base/dma-buf.c @@ -0,0 +1,289 @@ +/*
- Framework for buffer objects that can be shared across devices/subsystems.
- Copyright(C) 2011 Linaro Limited. All rights reserved.
- Author: Sumit Semwal sumit.semwal@ti.com
- Many thanks to linaro-mm-sig list, and specially
- Arnd Bergmann arnd@arndb.de, Rob Clark rob@ti.com and
- Daniel Vetter daniel@ffwll.ch for their support in creation and
- refining of this idea.
- This program is free software; you can redistribute it and/or modify it
- under the terms of the GNU General Public License version 2 as published by
- the Free Software Foundation.
- This program is distributed in the hope that it will be useful, but WITHOUT
- ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
- more details.
- You should have received a copy of the GNU General Public License along with
- this program. If not, see http://www.gnu.org/licenses/.
- */
+#include <linux/fs.h> +#include <linux/slab.h> +#include <linux/dma-buf.h> +#include <linux/anon_inodes.h> +#include <linux/export.h>
+static inline int is_dma_buf_file(struct file *);
+static int dma_buf_release(struct inode *inode, struct file *file) +{
- struct dma_buf *dmabuf;
- if (!is_dma_buf_file(file))
- return -EINVAL;
- dmabuf = file->private_data;
- dmabuf->ops->release(dmabuf);
- kfree(dmabuf);
- return 0;
+}
+static const struct file_operations dma_buf_fops = {
- .release = dma_buf_release,
+};
+/*
- is_dma_buf_file - Check if struct file* is associated with dma_buf
- */
+static inline int is_dma_buf_file(struct file *file) +{
- return file->f_op == &dma_buf_fops;
+}
+/**
Wrong kerneldoc.
- dma_buf_export - Creates a new dma_buf, and associates an anon file
- with this buffer, so it can be exported.
- Also connect the allocator specific data and ops to the buffer.
- @priv: [in] Attach private data of allocator to this buffer
- @ops: [in] Attach allocator-defined dma buf ops to the new buffer.
- @size: [in] Size of the buffer
- @flags: [in] mode flags for the file.
- Returns, on success, a newly created dma_buf object, which wraps the
- supplied private data and operations for dma_buf_ops. On either missing
- ops, or error in allocating struct dma_buf, will return negative error.
- */
+struct dma_buf *dma_buf_export(void *priv, struct dma_buf_ops *ops,
- size_t size, int flags)
+{
- struct dma_buf *dmabuf;
- struct file *file;
- if (WARN_ON(!priv || !ops
- || !ops->map_dma_buf
- || !ops->unmap_dma_buf
- || !ops->release)) {
- return ERR_PTR(-EINVAL);
- }
- dmabuf = kzalloc(sizeof(struct dma_buf), GFP_KERNEL);
- if (dmabuf == NULL)
- return ERR_PTR(-ENOMEM);
- dmabuf->priv = priv;
- dmabuf->ops = ops;
- dmabuf->size = size;
- file = anon_inode_getfile("dmabuf", &dma_buf_fops, dmabuf, flags);
- dmabuf->file = file;
- mutex_init(&dmabuf->lock);
- INIT_LIST_HEAD(&dmabuf->attachments);
- return dmabuf;
+} +EXPORT_SYMBOL_GPL(dma_buf_export);
+/**
- dma_buf_fd - returns a file descriptor for the given dma_buf
- @dmabuf: [in] pointer to dma_buf for which fd is required.
- On success, returns an associated 'fd'. Else, returns error.
- */
+int dma_buf_fd(struct dma_buf *dmabuf) +{
- int error, fd;
- if (!dmabuf || !dmabuf->file)
- return -EINVAL;
- error = get_unused_fd();
- if (error < 0)
- return error;
- fd = error;
- fd_install(fd, dmabuf->file);
- return fd;
+} +EXPORT_SYMBOL_GPL(dma_buf_fd);
+/**
- dma_buf_get - returns the dma_buf structure related to an fd
- @fd: [in] fd associated with the dma_buf to be returned
- On success, returns the dma_buf structure associated with an fd; uses
- file's refcounting done by fget to increase refcount. returns ERR_PTR
- otherwise.
- */
+struct dma_buf *dma_buf_get(int fd) +{
- struct file *file;
- file = fget(fd);
- if (!file)
- return ERR_PTR(-EBADF);
- if (!is_dma_buf_file(file)) {
- fput(file);
- return ERR_PTR(-EINVAL);
- }
- return file->private_data;
+} +EXPORT_SYMBOL_GPL(dma_buf_get);
+/**
- dma_buf_put - decreases refcount of the buffer
- @dmabuf: [in] buffer to reduce refcount of
- Uses file's refcounting done implicitly by fput()
- */
+void dma_buf_put(struct dma_buf *dmabuf) +{
- if (WARN_ON(!dmabuf || !dmabuf->file))
- return;
- fput(dmabuf->file);
+} +EXPORT_SYMBOL_GPL(dma_buf_put);
+/**
- dma_buf_attach - Add the device to dma_buf's attachments list; optionally,
- calls attach() of dma_buf_ops to allow device-specific attach functionality
- @dmabuf: [in] buffer to attach device to.
- @dev: [in] device to be attached.
- Returns struct dma_buf_attachment * for this attachment; may return negative
- error codes.
- */
+struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf,
- struct device *dev)
+{
- struct dma_buf_attachment *attach;
- int ret;
- if (WARN_ON(!dmabuf || !dev || !dmabuf->ops))
- return ERR_PTR(-EINVAL);
- attach = kzalloc(sizeof(struct dma_buf_attachment), GFP_KERNEL);
- if (attach == NULL)
- goto err_alloc;
- mutex_lock(&dmabuf->lock);
- attach->dev = dev;
- attach->dmabuf = dmabuf;
- if (dmabuf->ops->attach) {
- ret = dmabuf->ops->attach(dmabuf, dev, attach);
- if (ret)
- goto err_attach;
- }
- list_add(&attach->node, &dmabuf->attachments);
- mutex_unlock(&dmabuf->lock);
- return attach;
+err_alloc:
- return ERR_PTR(-ENOMEM);
+err_attach:
- kfree(attach);
- mutex_unlock(&dmabuf->lock);
- return ERR_PTR(ret);
+} +EXPORT_SYMBOL_GPL(dma_buf_attach);
+/**
- dma_buf_detach - Remove the given attachment from dmabuf's attachments list;
- optionally calls detach() of dma_buf_ops for device-specific detach
- @dmabuf: [in] buffer to detach from.
- @attach: [in] attachment to be detached; is free'd after this call.
- */
+void dma_buf_detach(struct dma_buf *dmabuf, struct dma_buf_attachment *attach) +{
- if (WARN_ON(!dmabuf || !attach || !dmabuf->ops))
- return;
- mutex_lock(&dmabuf->lock);
- list_del(&attach->node);
- if (dmabuf->ops->detach)
- dmabuf->ops->detach(dmabuf, attach);
- mutex_unlock(&dmabuf->lock);
- kfree(attach);
+} +EXPORT_SYMBOL_GPL(dma_buf_detach);
+/**
- dma_buf_map_attachment - Returns the scatterlist table of the attachment;
- mapped into _device_ address space. Is a wrapper for map_dma_buf() of the
- dma_buf_ops.
- @attach: [in] attachment whose scatterlist is to be returned
- @direction: [in] direction of DMA transfer
- Returns sg_table containing the scatterlist to be returned; may return NULL
- or ERR_PTR.
- */
+struct sg_table *dma_buf_map_attachment(struct dma_buf_attachment *attach,
- enum dma_data_direction direction)
+{
- struct sg_table *sg_table = ERR_PTR(-EINVAL);
- if (WARN_ON(!attach || !attach->dmabuf || !attach->dmabuf->ops))
- return ERR_PTR(-EINVAL);
- mutex_lock(&attach->dmabuf->lock);
- if (attach->dmabuf->ops->map_dma_buf)
- sg_table = attach->dmabuf->ops->map_dma_buf(attach, direction);
- mutex_unlock(&attach->dmabuf->lock);
- return sg_table;
+} +EXPORT_SYMBOL_GPL(dma_buf_map_attachment);
+/**
- dma_buf_unmap_attachment - unmaps and decreases usecount of the buffer;might
- deallocate the scatterlist associated. Is a wrapper for unmap_dma_buf() of
- dma_buf_ops.
- @attach: [in] attachment to unmap buffer from
- @sg_table: [in] scatterlist info of the buffer to unmap
- */
+void dma_buf_unmap_attachment(struct dma_buf_attachment *attach,
- struct sg_table *sg_table)
+{
- if (WARN_ON(!attach || !attach->dmabuf || !sg_table
- || !attach->dmabuf->ops))
- return;
- mutex_lock(&attach->dmabuf->lock);
- if (attach->dmabuf->ops->unmap_dma_buf)
- attach->dmabuf->ops->unmap_dma_buf(attach, sg_table);
- mutex_unlock(&attach->dmabuf->lock);
+} +EXPORT_SYMBOL_GPL(dma_buf_unmap_attachment); diff --git a/include/linux/dma-buf.h b/include/linux/dma-buf.h new file mode 100644 index 0000000..3e3ecf3 --- /dev/null +++ b/include/linux/dma-buf.h @@ -0,0 +1,172 @@ +/*
- Header file for dma buffer sharing framework.
- Copyright(C) 2011 Linaro Limited. All rights reserved.
- Author: Sumit Semwal sumit.semwal@ti.com
- Many thanks to linaro-mm-sig list, and specially
- Arnd Bergmann arnd@arndb.de, Rob Clark rob@ti.com and
- Daniel Vetter daniel@ffwll.ch for their support in creation and
- refining of this idea.
- This program is free software; you can redistribute it and/or modify it
- under the terms of the GNU General Public License version 2 as published by
- the Free Software Foundation.
- This program is distributed in the hope that it will be useful, but WITHOUT
- ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
- more details.
- You should have received a copy of the GNU General Public License along with
- this program. If not, see http://www.gnu.org/licenses/.
- */
+#ifndef __DMA_BUF_H__ +#define __DMA_BUF_H__
+#include <linux/file.h> +#include <linux/err.h> +#include <linux/device.h> +#include <linux/scatterlist.h> +#include <linux/list.h> +#include <linux/dma-mapping.h>
+struct dma_buf;
+/**
- struct dma_buf_attachment - holds device-buffer attachment data
OK, but what is the purpose of it?
- @dmabuf: buffer for this attachment.
- @dev: device attached to the buffer.
^^^ this
- @node: list_head to allow manipulation of list of dma_buf_attachment.
Just say: "list of dma_buf_attachment"'
- @priv: exporter-specific attachment data.
That "exporter-specific.." brings to my mind custom decleration forms. But maybe that is me.
- */
+struct dma_buf_attachment {
- struct dma_buf *dmabuf;
- struct device *dev;
- struct list_head node;
- void *priv;
+};
Why don't you move the decleration of this below 'struct dma_buf'? It would easier than to read this structure..
+/**
- struct dma_buf_ops - operations possible on struct dma_buf
- @attach: allows different devices to 'attach' themselves to the given
register?
- buffer. It might return -EBUSY to signal that backing storage
- is already allocated and incompatible with the requirements
Wait.. allocated or attached?
- of requesting device. [optional]
What is optional? The return value? Or the 'attach' call? If the later , say that in the first paragraph.
- @detach: detach a given device from this buffer. [optional]
- @map_dma_buf: returns list of scatter pages allocated, increases usecount
- of the buffer. Requires atleast one attach to be called
- before. Returned sg list should already be mapped into
- _device_ address space. This call may sleep. May also return
Ok, there is some __might_sleep macro you should put on the function.
- -EINTR.
Ok. What is the return code if attach has _not_ been called?
- @unmap_dma_buf: decreases usecount of buffer, might deallocate scatter
- pages.
- @release: release this buffer; to be called after the last dma_buf_put.
- @sync_sg_for_cpu: sync the sg list for cpu.
- @sync_sg_for_device: synch the sg list for device.
Not seeing those two.
- */
+struct dma_buf_ops {
- int (*attach)(struct dma_buf *, struct device *,
- struct dma_buf_attachment *);
- void (*detach)(struct dma_buf *, struct dma_buf_attachment *);
- /* For {map,unmap}_dma_buf below, any specific buffer attributes
- * required should get added to device_dma_parameters accessible
- * via dev->dma_params.
- */
- struct sg_table * (*map_dma_buf)(struct dma_buf_attachment *,
- enum dma_data_direction);
- void (*unmap_dma_buf)(struct dma_buf_attachment *,
- struct sg_table *);
- /* TODO: Add try_map_dma_buf version, to return immed with -EBUSY
Ewww. Why? Why not just just the 'map_dma_buf' and return that?
- * if the call would block.
- */
- /* after final dma_buf_put() */
- void (*release)(struct dma_buf *);
+};
+/**
- struct dma_buf - shared buffer object
Missing the 'size'.
- @file: file pointer used for sharing buffers across, and for refcounting.
- @attachments: list of dma_buf_attachment that denotes all devices attached.
- @ops: dma_buf_ops associated with this buffer object
- @priv: user specific private data
Can you elaborate on this? Is this the "exporter" using this? Or is it for the "user" using it? If so, why provide it? Wouldn't the user of this have something like this:
struct my_dma_bufs { struct dma_buf[20]; void *priv; }
Anyhow?
- */
+struct dma_buf {
- size_t size;
- struct file *file;
- struct list_head attachments;
- const struct dma_buf_ops *ops;
- /* mutex to serialize list manipulation and other ops */
- struct mutex lock;
- void *priv;
+};
+#ifdef CONFIG_DMA_SHARED_BUFFER +struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf,
- struct device *dev);
+void dma_buf_detach(struct dma_buf *dmabuf,
- struct dma_buf_attachment *dmabuf_attach);
+struct dma_buf *dma_buf_export(void *priv, struct dma_buf_ops *ops,
- size_t size, int flags);
+int dma_buf_fd(struct dma_buf *dmabuf); +struct dma_buf *dma_buf_get(int fd); +void dma_buf_put(struct dma_buf *dmabuf);
+struct sg_table *dma_buf_map_attachment(struct dma_buf_attachment *,
- enum dma_data_direction);
+void dma_buf_unmap_attachment(struct dma_buf_attachment *, struct sg_table *); +#else
+static inline struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf,
- struct device *dev)
+{
- return ERR_PTR(-ENODEV);
+}
+static inline void dma_buf_detach(struct dma_buf *dmabuf,
- struct dma_buf_attachment *dmabuf_attach)
+{
- return;
+}
+static inline struct dma_buf *dma_buf_export(void *priv,
- struct dma_buf_ops *ops,
- size_t size, int flags)
+{
- return ERR_PTR(-ENODEV);
+}
+static inline int dma_buf_fd(struct dma_buf *dmabuf) +{
- return -ENODEV;
+}
+static inline struct dma_buf *dma_buf_get(int fd) +{
- return ERR_PTR(-ENODEV);
+}
+static inline void dma_buf_put(struct dma_buf *dmabuf) +{
- return;
+}
+static inline struct sg_table *dma_buf_map_attachment(
- struct dma_buf_attachment *attach, enum dma_data_direction write)
+{
- return ERR_PTR(-ENODEV);
+}
+static inline void dma_buf_unmap_attachment(struct dma_buf_attachment *attach,
- struct sg_table *sg)
+{
- return;
+}
+#endif /* CONFIG_DMA_SHARED_BUFFER */
+#endif /* __DMA_BUF_H__ */
1.7.4.1
dri-devel mailing list dri-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/dri-devel
-- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi Konrad,
On Tue, Dec 20, 2011 at 10:06 PM, Konrad Rzeszutek Wilk konrad.wilk@oracle.com wrote:
On Mon, Dec 19, 2011 at 02:03:30PM +0530, Sumit Semwal wrote:
This is the first step in defining a dma buffer sharing mechanism.
A new buffer object dma_buf is added, with operations and API to allow easy sharing of this buffer object across devices.
The framework allows:
- different devices to 'attach' themselves to this buffer, to facilitate
backing storage negotiation, using dma_buf_attach() API.
Any thoughts of adding facility to track them? So you can see who is using what?
Not for version 1, but it would be a useful addition once we start using this mechanism.
- association of a file pointer with each user-buffer and associated
allocator-defined operations on that buffer. This operation is called the 'export' operation.
'create'? or 'alloc' ?
export implies an import somwhere and I don't think that is the case here.
I will rephrase it as suggested by Rob as well.
- this exported buffer-object to be shared with the other entity by asking for
its 'file-descriptor (fd)', and sharing the fd across.
- a received fd to get the buffer object back, where it can be accessed using
the associated exporter-defined operations.
- the exporter and user to share the scatterlist using map_dma_buf and
unmap_dma_buf operations.
Atleast one 'attach()' call is required to be made prior to calling the map_dma_buf() operation.
for the whole memory region or just for the device itself?
Rob has very eloquently and kindly explained it in his reply.
<snip>
+/*
- is_dma_buf_file - Check if struct file* is associated with dma_buf
- */
+static inline int is_dma_buf_file(struct file *file) +{
- return file->f_op == &dma_buf_fops;
+}
+/**
Wrong kerneldoc.
I looked into scripts/kernel-doc, and Documentation/kernel-doc-na-HOWTO.txt => both these places mention that the kernel-doc comments have to start with /**, and I couldn't spot an error in what's wrong with my usage - would you please elaborate on what you think is not right?
<snip>
+/**
- struct dma_buf_attachment - holds device-buffer attachment data
OK, but what is the purpose of it?
I will add that in the comments.
- @dmabuf: buffer for this attachment.
- @dev: device attached to the buffer.
^^^ this
- @node: list_head to allow manipulation of list of dma_buf_attachment.
Just say: "list of dma_buf_attachment"'
ok.
- @priv: exporter-specific attachment data.
That "exporter-specific.." brings to my mind custom decleration forms. But maybe that is me.
:) well, in context of dma-buf 'exporter', it makes sense.
- */
+struct dma_buf_attachment {
- struct dma_buf *dmabuf;
- struct device *dev;
- struct list_head node;
- void *priv;
+};
Why don't you move the decleration of this below 'struct dma_buf'? It would easier than to read this structure..
I could do that, but then anyways I will have to do a forward-declaration of dma_buf_attachment, since I have to use it in dma_buf_ops. If it improves readability, I am happy to move it below struct dma_buf.
+/**
- struct dma_buf_ops - operations possible on struct dma_buf
- @attach: allows different devices to 'attach' themselves to the given
register?
- buffer. It might return -EBUSY to signal that backing storage
- is already allocated and incompatible with the requirements
Wait.. allocated or attached?
This and above comment on 'register' are already answered by Rob in his explanation of the sequence in earlier reply. [the Documentation patch [2/2] also tries to explain it]
- of requesting device. [optional]
What is optional? The return value? Or the 'attach' call? If the later , say that in the first paragraph.
ok, sure. it is meant for the attach op.
- @detach: detach a given device from this buffer. [optional]
- @map_dma_buf: returns list of scatter pages allocated, increases usecount
- of the buffer. Requires atleast one attach to be called
- before. Returned sg list should already be mapped into
- _device_ address space. This call may sleep. May also return
Ok, there is some __might_sleep macro you should put on the function.
That's a nice suggestion; I will add it to the wrapper function for map_dma_buf().
- -EINTR.
Ok. What is the return code if attach has _not_ been called?
Will document it to return -EINVAL if atleast on attach() hasn't been called.
- @unmap_dma_buf: decreases usecount of buffer, might deallocate scatter
- pages.
- @release: release this buffer; to be called after the last dma_buf_put.
- @sync_sg_for_cpu: sync the sg list for cpu.
- @sync_sg_for_device: synch the sg list for device.
Not seeing those two.
Oops; removed in v3 - will correct.
- */
<snip>
- /* TODO: Add try_map_dma_buf version, to return immed with -EBUSY
Ewww. Why? Why not just just the 'map_dma_buf' and return that?
Requirement is to allow for blocking and non-blocking versions of map_dma_buf. try_map_dma_buf could be used for the non-blocking version.
<snip>
+/**
- struct dma_buf - shared buffer object
Missing the 'size'.
Will add.
- @file: file pointer used for sharing buffers across, and for refcounting.
- @attachments: list of dma_buf_attachment that denotes all devices attached.
- @ops: dma_buf_ops associated with this buffer object
- @priv: user specific private data
Can you elaborate on this? Is this the "exporter" using this? Or is it for the "user" using it? If so, why provide it? Wouldn't the user of this have something like this:
struct my_dma_bufs { struct dma_buf[20]; void *priv; }
Anyhow?
My bad - it is meant for the exporter - exporter gives this as one of the parameters to 'dma_buf_export()' API. I will correct the comment.
Thanks for your review! Best regards, ~Sumit.
On Fri, Dec 23, 2011 at 03:22:35PM +0530, Semwal, Sumit wrote:
Hi Konrad,
On Tue, Dec 20, 2011 at 10:06 PM, Konrad Rzeszutek Wilk konrad.wilk@oracle.com wrote:
On Mon, Dec 19, 2011 at 02:03:30PM +0530, Sumit Semwal wrote:
This is the first step in defining a dma buffer sharing mechanism.
A new buffer object dma_buf is added, with operations and API to allow easy sharing of this buffer object across devices.
The framework allows:
- different devices to 'attach' themselves to this buffer, to facilitate
backing storage negotiation, using dma_buf_attach() API.
Any thoughts of adding facility to track them? So you can see who is using what?
Not for version 1, but it would be a useful addition once we start using this mechanism.
OK. Looking forward to V2.
- association of a file pointer with each user-buffer and associated
allocator-defined operations on that buffer. This operation is called the 'export' operation.
'create'? or 'alloc' ?
export implies an import somwhere and I don't think that is the case here.
I will rephrase it as suggested by Rob as well.
- this exported buffer-object to be shared with the other entity by asking for
its 'file-descriptor (fd)', and sharing the fd across.
- a received fd to get the buffer object back, where it can be accessed using
the associated exporter-defined operations.
- the exporter and user to share the scatterlist using map_dma_buf and
unmap_dma_buf operations.
Atleast one 'attach()' call is required to be made prior to calling the map_dma_buf() operation.
for the whole memory region or just for the device itself?
Rob has very eloquently and kindly explained it in his reply.
Can you include his words of wisdom in the git description?
<snip> >> +/* >> + * is_dma_buf_file - Check if struct file* is associated with dma_buf >> + */ >> +static inline int is_dma_buf_file(struct file *file) >> +{ >> + return file->f_op == &dma_buf_fops; >> +} >> + >> +/** > > Wrong kerneldoc. I looked into scripts/kernel-doc, and Documentation/kernel-doc-na-HOWTO.txt => both these places mention that the kernel-doc comments have to start with /**, and I couldn't spot an error in what's wrong with my usage - would you please elaborate on what you think is not right?
The issue I had was with '/**' but let me double-check where I learned that /** was a bad. Either way, it is a style-guide thing and the Documentation/* trumps what I recall.
<snip> >> +/** >> + * struct dma_buf_attachment - holds device-buffer attachment data > > OK, but what is the purpose of it? I will add that in the comments. > >> + * @dmabuf: buffer for this attachment. >> + * @dev: device attached to the buffer. > ^^^ this >> + * @node: list_head to allow manipulation of list of dma_buf_attachment. > > Just say: "list of dma_buf_attachment"' ok. > >> + * @priv: exporter-specific attachment data. > > That "exporter-specific.." brings to my mind custom decleration forms. But maybe that is me. :) well, in context of dma-buf 'exporter', it makes sense.
Or just private contents of the backend driver. But the naming is not that important to inhibit this patch from being merged.
- */
+struct dma_buf_attachment {
- struct dma_buf *dmabuf;
- struct device *dev;
- struct list_head node;
- void *priv;
+};
Why don't you move the decleration of this below 'struct dma_buf'? It would easier than to read this structure..
I could do that, but then anyways I will have to do a forward-declaration of dma_buf_attachment, since I have to use it in dma_buf_ops. If it improves readability, I am happy to move it below struct dma_buf
It is more of just making the readability easier. As in reading from top bottom one. But if it is too ugly, don't bother.
+/**
- struct dma_buf_ops - operations possible on struct dma_buf
- @attach: allows different devices to 'attach' themselves to the given
register?
- buffer. It might return -EBUSY to signal that backing storage
- is already allocated and incompatible with the requirements
Wait.. allocated or attached?
This and above comment on 'register' are already answered by Rob in his explanation of the sequence in earlier reply. [the Documentation patch [2/2] also tries to explain it]
OK. Might want to mention the user to look in the Documentation, in case you don't already have it.
- of requesting device. [optional]
What is optional? The return value? Or the 'attach' call? If the later , say that in the first paragraph.
ok, sure. it is meant for the attach op.
- @detach: detach a given device from this buffer. [optional]
- @map_dma_buf: returns list of scatter pages allocated, increases usecount
- of the buffer. Requires atleast one attach to be called
- before. Returned sg list should already be mapped into
- _device_ address space. This call may sleep. May also return
Ok, there is some __might_sleep macro you should put on the function.
That's a nice suggestion; I will add it to the wrapper function for map_dma_buf().
- -EINTR.
Ok. What is the return code if attach has _not_ been called?
Will document it to return -EINVAL if atleast on attach() hasn't been called.
- @unmap_dma_buf: decreases usecount of buffer, might deallocate scatter
- pages.
- @release: release this buffer; to be called after the last dma_buf_put.
- @sync_sg_for_cpu: sync the sg list for cpu.
- @sync_sg_for_device: synch the sg list for device.
Not seeing those two.
Oops; removed in v3 - will correct.
- */
<snip> >> + /* TODO: Add try_map_dma_buf version, to return immed with -EBUSY > > Ewww. Why? Why not just just the 'map_dma_buf' and return that? Requirement is to allow for blocking and non-blocking versions of map_dma_buf. try_map_dma_buf could be used for the non-blocking version.
Ok. What about just adding that in as a nop function operation. And have an initial implementation in the code base that does .. well, nothing except return -ENOSPC and move the TODO to the code instead of it being in the header?
<snip> >> +/** >> + * struct dma_buf - shared buffer object > > Missing the 'size'. Will add. > >> + * @file: file pointer used for sharing buffers across, and for refcounting. >> + * @attachments: list of dma_buf_attachment that denotes all devices attached. >> + * @ops: dma_buf_ops associated with this buffer object >> + * @priv: user specific private data > > > Can you elaborate on this? Is this the "exporter" using this? Or is > it for the "user" using it? If so, why provide it? Wouldn't the > user of this have something like this: > > struct my_dma_bufs { > struct dma_buf[20]; > void *priv; > } > > Anyhow? My bad - it is meant for the exporter - exporter gives this as one of the parameters to 'dma_buf_export()' API. I will correct the comment. > Thanks for your review!
Sure. And please put 'Reviewed-by: Konrad Rzeszutek Wilk konrad.wilk@oracle.com'.
Best regards, ~Sumit.
Add documentation for dma buffer sharing framework, explaining the various operations, members and API of the dma buffer sharing framework.
Signed-off-by: Sumit Semwal sumit.semwal@linaro.org Signed-off-by: Sumit Semwal sumit.semwal@ti.com --- Documentation/dma-buf-sharing.txt | 222 +++++++++++++++++++++++++++++++++++++ 1 files changed, 222 insertions(+), 0 deletions(-) create mode 100644 Documentation/dma-buf-sharing.txt
diff --git a/Documentation/dma-buf-sharing.txt b/Documentation/dma-buf-sharing.txt new file mode 100644 index 0000000..3a2a35f --- /dev/null +++ b/Documentation/dma-buf-sharing.txt @@ -0,0 +1,222 @@ + DMA Buffer Sharing API Guide + ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + + Sumit Semwal + <sumit dot semwal at linaro dot org> + <sumit dot semwal at ti dot com> + +This document serves as a guide to device-driver writers on what is the dma-buf +buffer sharing API, how to use it for exporting and using shared buffers. + +Any device driver which wishes to be a part of DMA buffer sharing, can do so as +either the 'exporter' of buffers, or the 'user' of buffers. + +Say a driver A wants to use buffers created by driver B, then we call B as the +exporter, and A as buffer-user. + +The exporter +- implements and manages operations[1] for the buffer +- allows other users to share the buffer by using dma_buf sharing APIs, +- manages the details of buffer allocation, +- decides about the actual backing storage where this allocation happens, +- takes care of any migration of scatterlist - for all (shared) users of this + buffer, + +The buffer-user +- is one of (many) sharing users of the buffer. +- doesn't need to worry about how the buffer is allocated, or where. +- needs a mechanism to get access to the scatterlist that makes up this buffer + in memory, mapped into its own address space, so it can access the same area + of memory. + +*IMPORTANT*: A buffer shared using the dma_buf sharing API *must not* be +exported to user space using "mmap". [see https://lkml.org/lkml/2011/12/2/53 +for more on the discussion] + +The dma_buf buffer sharing API usage contains the following steps: + +1. Exporter announces that it wishes to export a buffer +2. Userspace gets the file descriptor associated with the exported buffer, and + passes it around to potential buffer-users based on use case +3. Each buffer-user 'connects' itself to the buffer +4. When needed, buffer-user requests access to the buffer from exporter +5. When finished with its use, the buffer-user notifies end-of-DMA to exporter +6. when buffer-user is done using this buffer completely, it 'disconnects' + itself from the buffer. + + +1. Exporter's announcement of buffer export + + The buffer exporter announces its wish to export a buffer. In this, it + connects its own private buffer data, provides implementation for operations + that can be performed on the exported dma_buf, and flags for the file + associated with this buffer. + + Interface: + struct dma_buf *dma_buf_export(void *priv, struct dma_buf_ops *ops, + int flags) + + If this succeeds, dma_buf_export allocates a dma_buf structure, and returns a + pointer to the same. It also associates an anonymous file with this buffer, + so it can be exported. On failure to allocate the dma_buf object, it returns + NULL. + +2. Userspace gets a handle to pass around to potential buffer-users + + Userspace entity requests for a file-descriptor (fd) which is a handle to the + anonymous file associated with the buffer. It can then share the fd with other + drivers and/or processes. + + Interface: + int dma_buf_fd(struct dma_buf *dmabuf) + + This API installs an fd for the anonymous file associated with this buffer; + returns either 'fd', or error. + +3. Each buffer-user 'connects' itself to the buffer + + Each buffer-user now gets a reference to the buffer, using the fd passed to + it. + + Interface: + struct dma_buf *dma_buf_get(int fd) + + This API will return a reference to the dma_buf, and increment refcount for + it. + + After this, the buffer-user needs to attach its device with the buffer, which + helps the exporter to know of device buffer constraints. + + Interface: + struct dma_buf_attachment *dma_buf_attach(struct dma_buf *dmabuf, + struct device *dev) + + This API returns reference to an attachment structure, which is then used + for scatterlist operations. It will optionally call the 'attach' dma_buf + operation, if provided by the exporter. + + The dma-buf sharing framework does the bookkeeping bits related to managing + the list of all attachments to a buffer. + +Until this stage, the buffer-exporter has the option to choose not to actually +allocate the backing storage for this buffer, but wait for the first buffer-user +to request use of buffer for allocation. + + +4. When needed, buffer-user requests access to the buffer + + Whenever a buffer-user wants to use the buffer for any DMA, it asks for + access to the buffer using dma_buf_map_attachment API. At least one attach to + the buffer must have happened before map_dma_buf can be called. + + Interface: + struct sg_table * dma_buf_map_attachment(struct dma_buf_attachment *, + enum dma_data_direction); + + This is a wrapper to dma_buf->ops->map_dma_buf operation, which hides the + "dma_buf->ops->" indirection from the users of this interface. + + In struct dma_buf_ops, map_dma_buf is defined as + struct sg_table * (*map_dma_buf)(struct dma_buf_attachment *, + enum dma_data_direction); + + It is one of the buffer operations that must be implemented by the exporter. + It should return the sg_table containing scatterlist for this buffer, mapped + into caller's address space. + + If this is being called for the first time, the exporter can now choose to + scan through the list of attachments for this buffer, collate the requirements + of the attached devices, and choose an appropriate backing storage for the + buffer. + + Based on enum dma_data_direction, it might be possible to have multiple users + accessing at the same time (for reading, maybe), or any other kind of sharing + that the exporter might wish to make available to buffer-users. + + map_dma_buf() operation can return -EINTR if it is interrupted by a signal. + + +5. When finished, the buffer-user notifies end-of-DMA to exporter + + Once the DMA for the current buffer-user is over, it signals 'end-of-DMA' to + the exporter using the dma_buf_unmap_attachment API. + + Interface: + void dma_buf_unmap_attachment(struct dma_buf_attachment *, + struct sg_table *); + + This is a wrapper to dma_buf->ops->unmap_dma_buf() operation, which hides the + "dma_buf->ops->" indirection from the users of this interface. + + In struct dma_buf_ops, unmap_dma_buf is defined as + void (*unmap_dma_buf)(struct dma_buf_attachment *, struct sg_table *); + + unmap_dma_buf signifies the end-of-DMA for the attachment provided. Like + map_dma_buf, this API also must be implemented by the exporter. + + +6. when buffer-user is done using this buffer, it 'disconnects' itself from the + buffer. + + After the buffer-user has no more interest in using this buffer, it should + disconnect itself from the buffer: + + - it first detaches itself from the buffer. + + Interface: + void dma_buf_detach(struct dma_buf *dmabuf, + struct dma_buf_attachment *dmabuf_attach); + + This API removes the attachment from the list in dmabuf, and optionally calls + dma_buf->ops->detach(), if provided by exporter, for any housekeeping bits. + + - Then, the buffer-user returns the buffer reference to exporter. + + Interface: + void dma_buf_put(struct dma_buf *dmabuf); + + This API then reduces the refcount for this buffer. + + If, as a result of this call, the refcount becomes 0, the 'release' file + operation related to this fd is called. It calls the dmabuf->ops->release() + operation in turn, and frees the memory allocated for dmabuf when exported. + +NOTES: +- Importance of attach-detach and {map,unmap}_dma_buf operation pairs + The attach-detach calls allow the exporter to figure out backing-storage + constraints for the currently-interested devices. This allows preferential + allocation, and/or migration of pages across different types of storage + available, if possible. + + Bracketing of DMA access with {map,unmap}_dma_buf operations is essential + to allow just-in-time backing of storage, and migration mid-way through a + use-case. + +- Migration of backing storage if needed + If after + - at least one map_dma_buf has happened, + - and the backing storage has been allocated for this buffer, + another new buffer-user intends to attach itself to this buffer, it might + be allowed, if possible for the exporter. + + In case it is allowed by the exporter: + if the new buffer-user has stricter 'backing-storage constraints', and the + exporter can handle these constraints, the exporter can just stall on the + map_dma_buf until all outstanding access is completed (as signalled by + unmap_dma_buf). + Once all users have finished accessing and have unmapped this buffer, the + exporter could potentially move the buffer to the stricter backing-storage, + and then allow further {map,unmap}_dma_buf operations from any buffer-user + from the migrated backing-storage. + + If the exporter cannot fulfil the backing-storage constraints of the new + buffer-user device as requested, dma_buf_attach() would return an error to + denote non-compatibility of the new buffer-sharing request with the current + buffer. + + If the exporter chooses not to allow an attach() operation once a + map_dma_buf() API has been called, it simply returns an error. + +References: +[1] struct dma_buf_ops in include/linux/dma-buf.h +[2] All interfaces mentioned above defined in include/linux/dma-buf.h
On Mon, Dec 19, 2011 at 02:03:28PM +0530, Sumit Semwal wrote:
Hello Everyone,
This is RFC v3 for DMA buffer sharing mechanism - changes from v2 are in the changelog below.
Various subsystems - V4L2, GPU-accessors, DRI to name a few - have felt the need to have a common mechanism to share memory buffers across different devices - ARM, video hardware, GPU.
This need comes forth from a variety of use cases including cameras, image processing, video recorders, sound processing, DMA engines, GPU and display buffers, and others.
This RFC is an attempt to define such a buffer sharing mechanism- it is the result of discussions from a couple of memory-management mini-summits held by Linaro to understand and address common needs around memory management. [1]
A new dma_buf buffer object is added, with operations and API to allow easy sharing of this buffer object across devices.
The framework allows:
- a new buffer-object to be created with fixed size.
- different devices to 'attach' themselves to this buffer, to facilitate backing storage negotiation, using dma_buf_attach() API.
- association of a file pointer with each user-buffer and associated allocator-defined operations on that buffer. This operation is called the 'export' operation.
- this exported buffer-object to be shared with the other entity by asking for its 'file-descriptor (fd)', and sharing the fd across.
- a received fd to get the buffer object back, where it can be accessed using the associated exporter-defined operations.
- the exporter and user to share the scatterlist using map_dma_buf and unmap_dma_buf operations.
Documentation present in the patch-set gives more details.
This is based on design suggestions from many people at the mini-summits, most notably from Arnd Bergmann arnd@arndb.de, Rob Clark rob@ti.com and Daniel Vetter daniel@ffwll.ch.
The implementation is inspired from proof-of-concept patch-set from Tomasz Stanislawski t.stanislaws@samsung.com, who demonstrated buffer sharing between two v4l2 devices. [2]
References: [1]: https://wiki.linaro.org/OfficeofCTO/MemoryManagement [2]: http://lwn.net/Articles/454389
Patchset based on top of 3.2-rc3, the current version can be found at
http://git.linaro.org/gitweb?p=people/sumitsemwal/linux-3.x.git Branch: dma-buf-upstr-v2
Earlier versions: v2 at: https://lkml.org/lkml/2011/12/2/53 v1 at: https://lkml.org/lkml/2011/10/11/92
Best regards, ~Sumit Semwal
I think this is a really good v1 version of dma_buf. It contains all the required bits (with well-specified semantics in the doc patch) to implement some basic use-cases and start fleshing out the integration with various subsystem (like drm and v4l). All the things still under discussion like - userspace mmap support - more advanced (and more strictly specified) coherency models - and shared infrastructure for implementing exporters are imo much clearer once we have a few example drivers at hand and a better understanding of some of the insane corner cases we need to be able to handle.
And I think any risk that the resulting clarifications will break a basic use-case is really minimal, so I think it'd be great if this could go into 3.3 (maybe as some kind of staging/experimental infrastructure).
Hence for both patches: Reviewed-by: Daniel Vetter daniel.vetter@ffwll.ch
This is RFC v3 for DMA buffer sharing mechanism - changes from v2 are in the changelog below.
Various subsystems - V4L2, GPU-accessors, DRI to name a few - have felt the need to have a common mechanism to share memory buffers across different devices - ARM, video hardware, GPU.
This need comes forth from a variety of use cases including cameras, image processing, video recorders, sound processing, DMA engines, GPU and display buffers, and others.
This RFC is an attempt to define such a buffer sharing mechanism- it is the result of discussions from a couple of memory-management mini-summits held by Linaro to understand and address common needs around memory management. [1]
A new dma_buf buffer object is added, with operations and API to allow easy sharing of this buffer object across devices.
The framework allows:
- a new buffer-object to be created with fixed size.
- different devices to 'attach' themselves to this buffer, to facilitate
backing storage negotiation, using dma_buf_attach() API.
- association of a file pointer with each user-buffer and associated
allocator-defined operations on that buffer. This operation is called the 'export' operation.
- this exported buffer-object to be shared with the other entity by asking for
its 'file-descriptor (fd)', and sharing the fd across.
- a received fd to get the buffer object back, where it can be accessed using
the associated exporter-defined operations.
- the exporter and user to share the scatterlist using map_dma_buf and
unmap_dma_buf operations.
Documentation present in the patch-set gives more details.
This is based on design suggestions from many people at the mini-summits, most notably from Arnd Bergmann arnd@arndb.de, Rob Clark rob@ti.com and Daniel Vetter daniel@ffwll.ch.
The implementation is inspired from proof-of-concept patch-set from Tomasz Stanislawski t.stanislaws@samsung.com, who demonstrated buffer sharing between two v4l2 devices. [2]
References: [1]: https://wiki.linaro.org/OfficeofCTO/MemoryManagement [2]: http://lwn.net/Articles/454389
Patchset based on top of 3.2-rc3, the current version can be found at
http://git.linaro.org/gitweb?p=people/sumitsemwal/linux-3.x.git Branch: dma-buf-upstr-v2
Earlier versions: v2 at: https://lkml.org/lkml/2011/12/2/53 v1 at: https://lkml.org/lkml/2011/10/11/92
Best regards, ~Sumit Semwal
I think this is a really good v1 version of dma_buf. It contains all the required bits (with well-specified semantics in the doc patch) to implement some basic use-cases and start fleshing out the integration with various subsystem (like drm and v4l). All the things still under discussion like
- userspace mmap support
- more advanced (and more strictly specified) coherency models
- and shared infrastructure for implementing exporters
are imo much clearer once we have a few example drivers at hand and a better understanding of some of the insane corner cases we need to be able to handle.
And I think any risk that the resulting clarifications will break a basic use-case is really minimal, so I think it'd be great if this could go into 3.3 (maybe as some kind of staging/experimental infrastructure).
Hence for both patches: Reviewed-by: Daniel Vetter daniel.vetter@ffwll.ch
Yeah I'm with Daniel, I like this one, I can definitely build the drm buffer sharing layer on top of this.
How do we see this getting merged? I'm quite happy to push it to Linus if we don't have an identified path, though it could go via a Linaro tree as well.
so feel free to add: Reviewed-by: Dave Airlie airlied@redhat.com
Dave.
On Tue, Dec 20, 2011 at 2:20 PM, Dave Airlie airlied@gmail.com wrote:
This is RFC v3 for DMA buffer sharing mechanism - changes from v2 are in the changelog below.
Various subsystems - V4L2, GPU-accessors, DRI to name a few - have felt the need to have a common mechanism to share memory buffers across different devices - ARM, video hardware, GPU.
This need comes forth from a variety of use cases including cameras, image processing, video recorders, sound processing, DMA engines, GPU and display buffers, and others.
This RFC is an attempt to define such a buffer sharing mechanism- it is the result of discussions from a couple of memory-management mini-summits held by Linaro to understand and address common needs around memory management. [1]
A new dma_buf buffer object is added, with operations and API to allow easy sharing of this buffer object across devices.
The framework allows:
- a new buffer-object to be created with fixed size.
- different devices to 'attach' themselves to this buffer, to facilitate
backing storage negotiation, using dma_buf_attach() API.
- association of a file pointer with each user-buffer and associated
allocator-defined operations on that buffer. This operation is called the 'export' operation.
- this exported buffer-object to be shared with the other entity by asking for
its 'file-descriptor (fd)', and sharing the fd across.
- a received fd to get the buffer object back, where it can be accessed using
the associated exporter-defined operations.
- the exporter and user to share the scatterlist using map_dma_buf and
unmap_dma_buf operations.
Documentation present in the patch-set gives more details.
This is based on design suggestions from many people at the mini-summits, most notably from Arnd Bergmann arnd@arndb.de, Rob Clark rob@ti.com and Daniel Vetter daniel@ffwll.ch.
The implementation is inspired from proof-of-concept patch-set from Tomasz Stanislawski t.stanislaws@samsung.com, who demonstrated buffer sharing between two v4l2 devices. [2]
References: [1]: https://wiki.linaro.org/OfficeofCTO/MemoryManagement [2]: http://lwn.net/Articles/454389
Patchset based on top of 3.2-rc3, the current version can be found at
http://git.linaro.org/gitweb?p=people/sumitsemwal/linux-3.x.git Branch: dma-buf-upstr-v2
Earlier versions: v2 at: https://lkml.org/lkml/2011/12/2/53 v1 at: https://lkml.org/lkml/2011/10/11/92
Best regards, ~Sumit Semwal
I think this is a really good v1 version of dma_buf. It contains all the required bits (with well-specified semantics in the doc patch) to implement some basic use-cases and start fleshing out the integration with various subsystem (like drm and v4l). All the things still under discussion like
- userspace mmap support
- more advanced (and more strictly specified) coherency models
- and shared infrastructure for implementing exporters
are imo much clearer once we have a few example drivers at hand and a better understanding of some of the insane corner cases we need to be able to handle.
And I think any risk that the resulting clarifications will break a basic use-case is really minimal, so I think it'd be great if this could go into 3.3 (maybe as some kind of staging/experimental infrastructure).
Hence for both patches: Reviewed-by: Daniel Vetter daniel.vetter@ffwll.ch
Yeah I'm with Daniel, I like this one, I can definitely build the drm buffer sharing layer on top of this.
How do we see this getting merged? I'm quite happy to push it to Linus if we don't have an identified path, though it could go via a Linaro tree as well.
so feel free to add: Reviewed-by: Dave Airlie airlied@redhat.com
fwiw, patches to share buffers between drm and v4l2 are here:
https://github.com/robclark/kernel-omap4/commits/drmplane-dmabuf
(need a bit of cleanup before the vb2 patches are submitted.. but that is unrelated to the dmabuf patches)
so,
Reviewed-and-Tested-by: Rob Clark rob.clark@linaro.org
Dave.
To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Wed, Dec 21, 2011 at 3:56 AM, Rob Clark rob@ti.com wrote:
On Tue, Dec 20, 2011 at 2:20 PM, Dave Airlie airlied@gmail.com wrote:
<snip>
I think this is a really good v1 version of dma_buf. It contains all the required bits (with well-specified semantics in the doc patch) to implement some basic use-cases and start fleshing out the integration with various subsystem (like drm and v4l). All the things still under discussion like
- userspace mmap support
- more advanced (and more strictly specified) coherency models
- and shared infrastructure for implementing exporters
are imo much clearer once we have a few example drivers at hand and a better understanding of some of the insane corner cases we need to be able to handle.
And I think any risk that the resulting clarifications will break a basic use-case is really minimal, so I think it'd be great if this could go into 3.3 (maybe as some kind of staging/experimental infrastructure).
Hence for both patches: Reviewed-by: Daniel Vetter daniel.vetter@ffwll.ch
Yeah I'm with Daniel, I like this one, I can definitely build the drm buffer sharing layer on top of this.
How do we see this getting merged? I'm quite happy to push it to Linus if we don't have an identified path, though it could go via a Linaro tree as well.
so feel free to add: Reviewed-by: Dave Airlie airlied@redhat.com
fwiw, patches to share buffers between drm and v4l2 are here:
https://github.com/robclark/kernel-omap4/commits/drmplane-dmabuf
(need a bit of cleanup before the vb2 patches are submitted.. but that is unrelated to the dmabuf patches)
so,
Reviewed-and-Tested-by: Rob Clark rob.clark@linaro.org
Thanks Daniel, Dave, and Rob! BR, Sumit.
Dave.
To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Wed, Dec 21, 2011 at 1:50 AM, Dave Airlie airlied@gmail.com wrote: <snip>
I think this is a really good v1 version of dma_buf. It contains all the required bits (with well-specified semantics in the doc patch) to implement some basic use-cases and start fleshing out the integration with various subsystem (like drm and v4l). All the things still under discussion like
- userspace mmap support
- more advanced (and more strictly specified) coherency models
- and shared infrastructure for implementing exporters
are imo much clearer once we have a few example drivers at hand and a better understanding of some of the insane corner cases we need to be able to handle.
And I think any risk that the resulting clarifications will break a basic use-case is really minimal, so I think it'd be great if this could go into 3.3 (maybe as some kind of staging/experimental infrastructure).
Hence for both patches: Reviewed-by: Daniel Vetter daniel.vetter@ffwll.ch
Yeah I'm with Daniel, I like this one, I can definitely build the drm buffer sharing layer on top of this.
How do we see this getting merged? I'm quite happy to push it to Linus if we don't have an identified path, though it could go via a Linaro tree as well.
so feel free to add: Reviewed-by: Dave Airlie airlied@redhat.com
Thanks Daniel and Dave!
I guess we can start with staging for 3.3, and see how it shapes up. I will post the latest patch version pretty soon.
Arnd, Dave: do you have any preference on the path it takes to get merged? In my mind, Linaro tree might make more sense, but I would leave it upto you gentlemen.
Dave.
Best regards, ~Sumit.
On Fri, Dec 23, 2011 at 4:08 AM, Semwal, Sumit sumit.semwal@ti.com wrote:
On Wed, Dec 21, 2011 at 1:50 AM, Dave Airlie airlied@gmail.com wrote:
<snip> >> >> I think this is a really good v1 version of dma_buf. It contains all the >> required bits (with well-specified semantics in the doc patch) to >> implement some basic use-cases and start fleshing out the integration with >> various subsystem (like drm and v4l). All the things still under >> discussion like >> - userspace mmap support >> - more advanced (and more strictly specified) coherency models >> - and shared infrastructure for implementing exporters >> are imo much clearer once we have a few example drivers at hand and a >> better understanding of some of the insane corner cases we need to be able >> to handle. >> >> And I think any risk that the resulting clarifications will break a basic >> use-case is really minimal, so I think it'd be great if this could go into >> 3.3 (maybe as some kind of staging/experimental infrastructure). >> >> Hence for both patches: >> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> > > Yeah I'm with Daniel, I like this one, I can definitely build the drm > buffer sharing layer on top of this. > > How do we see this getting merged? I'm quite happy to push it to Linus > if we don't have an identified path, though it could go via a Linaro > tree as well. > > so feel free to add: > Reviewed-by: Dave Airlie <airlied@redhat.com> Thanks Daniel and Dave!
I guess we can start with staging for 3.3, and see how it shapes up. I will post the latest patch version pretty soon.
not sure about staging, but could make sense to mark as experimental.
Arnd, Dave: do you have any preference on the path it takes to get merged? In my mind, Linaro tree might make more sense, but I would leave it upto you gentlemen.
Looks like Dave is making some progress on drm usage of buffer sharing between gpu's.. if that is ready to go in at the same time, it might be a bit logistically simpler for him to put dmabuf in the same pull req. I don't have strong preference one way or another, so do what is collectively simpler ;-)
BR, -R
Dave.
Best regards, ~Sumit. -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Fri, Dec 23, 2011 at 10:50 PM, Rob Clark rob@ti.com wrote:
On Fri, Dec 23, 2011 at 4:08 AM, Semwal, Sumit sumit.semwal@ti.com wrote:
On Wed, Dec 21, 2011 at 1:50 AM, Dave Airlie airlied@gmail.com wrote:
<snip> >> >> Hence for both patches: >> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> > > Yeah I'm with Daniel, I like this one, I can definitely build the drm > buffer sharing layer on top of this. > > How do we see this getting merged? I'm quite happy to push it to Linus > if we don't have an identified path, though it could go via a Linaro > tree as well. > > so feel free to add: > Reviewed-by: Dave Airlie <airlied@redhat.com> Thanks Daniel and Dave!
I guess we can start with staging for 3.3, and see how it shapes up. I will post the latest patch version pretty soon.
not sure about staging, but could make sense to mark as experimental.
Thanks, I will mark it experimental for the first version; we can remove that once it is more widely used and tested.
Arnd, Dave: do you have any preference on the path it takes to get merged? In my mind, Linaro tree might make more sense, but I would leave it upto you gentlemen.
Looks like Dave is making some progress on drm usage of buffer sharing between gpu's.. if that is ready to go in at the same time, it might be a bit logistically simpler for him to put dmabuf in the same pull req. I don't have strong preference one way or another, so do what is collectively simpler ;-)
:) Right - I am quite happy for it to get merged in either ways :)
BR, -R
Best regards, ~Sumit.
linaro-mm-sig@lists.linaro.org
-
Daniel Vetter -
Dave Airlie -
Konrad Rzeszutek Wilk -
Rob Clark -
Semwal, Sumit -
Sumit Semwal