1 Buffer Sharing and Synchronization
2 ==================================
4 The dma-buf subsystem provides the framework for sharing buffers for
5 hardware (DMA) access across multiple device drivers and subsystems, and
6 for synchronizing asynchronous hardware access.
8 This is used, for example, by drm "prime" multi-GPU support, but is of
9 course not limited to GPU use cases.
11 The three main components of this are: (1) dma-buf, representing a
12 sg_table and exposed to userspace as a file descriptor to allow passing
13 between devices, (2) fence, which provides a mechanism to signal when
14 one device has finished access, and (3) reservation, which manages the
15 shared or exclusive fence(s) associated with the buffer.
20 This document serves as a guide to device-driver writers on what is the dma-buf
21 buffer sharing API, how to use it for exporting and using shared buffers.
23 Any device driver which wishes to be a part of DMA buffer sharing, can do so as
24 either the 'exporter' of buffers, or the 'user' or 'importer' of buffers.
26 Say a driver A wants to use buffers created by driver B, then we call B as the
27 exporter, and A as buffer-user/importer.
31 - implements and manages operations in :c:type:`struct dma_buf_ops
32 <dma_buf_ops>` for the buffer,
33 - allows other users to share the buffer by using dma_buf sharing APIs,
34 - manages the details of buffer allocation, wrapped in a :c:type:`struct
36 - decides about the actual backing storage where this allocation happens,
37 - and takes care of any migration of scatterlist - for all (shared) users of
42 - is one of (many) sharing users of the buffer.
43 - doesn't need to worry about how the buffer is allocated, or where.
44 - and needs a mechanism to get access to the scatterlist that makes up this
45 buffer in memory, mapped into its own address space, so it can access the
46 same area of memory. This interface is provided by :c:type:`struct
47 dma_buf_attachment <dma_buf_attachment>`.
49 Any exporters or users of the dma-buf buffer sharing framework must have a
50 'select DMA_SHARED_BUFFER' in their respective Kconfigs.
52 Userspace Interface Notes
53 ~~~~~~~~~~~~~~~~~~~~~~~~~
55 Mostly a DMA buffer file descriptor is simply an opaque object for userspace,
56 and hence the generic interface exposed is very minimal. There's a few things to
59 - Since kernel 3.12 the dma-buf FD supports the llseek system call, but only
60 with offset=0 and whence=SEEK_END|SEEK_SET. SEEK_SET is supported to allow
61 the usual size discover pattern size = SEEK_END(0); SEEK_SET(0). Every other
62 llseek operation will report -EINVAL.
64 If llseek on dma-buf FDs isn't support the kernel will report -ESPIPE for all
65 cases. Userspace can use this to detect support for discovering the dma-buf
68 - In order to avoid fd leaks on exec, the FD_CLOEXEC flag must be set
69 on the file descriptor. This is not just a resource leak, but a
70 potential security hole. It could give the newly exec'd application
71 access to buffers, via the leaked fd, to which it should otherwise
72 not be permitted access.
74 The problem with doing this via a separate fcntl() call, versus doing it
75 atomically when the fd is created, is that this is inherently racy in a
76 multi-threaded app[3]. The issue is made worse when it is library code
77 opening/creating the file descriptor, as the application may not even be
80 To avoid this problem, userspace must have a way to request O_CLOEXEC
81 flag be set when the dma-buf fd is created. So any API provided by
82 the exporting driver to create a dmabuf fd must provide a way to let
83 userspace control setting of O_CLOEXEC flag passed in to dma_buf_fd().
85 - Memory mapping the contents of the DMA buffer is also supported. See the
86 discussion below on `CPU Access to DMA Buffer Objects`_ for the full details.
88 - The DMA buffer FD is also pollable, see `Implicit Fence Poll Support`_ below for
91 Basic Operation and Device DMA Access
92 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
94 .. kernel-doc:: drivers/dma-buf/dma-buf.c
95 :doc: dma buf device access
97 CPU Access to DMA Buffer Objects
98 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
100 .. kernel-doc:: drivers/dma-buf/dma-buf.c
103 Implicit Fence Poll Support
104 ~~~~~~~~~~~~~~~~~~~~~~~~~~~
106 .. kernel-doc:: drivers/dma-buf/dma-buf.c
107 :doc: implicit fence polling
109 Kernel Functions and Structures Reference
110 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
112 .. kernel-doc:: drivers/dma-buf/dma-buf.c
115 .. kernel-doc:: include/linux/dma-buf.h
118 Buffer Mapping Helpers
119 ~~~~~~~~~~~~~~~~~~~~~~
121 .. kernel-doc:: include/linux/dma-buf-map.h
124 .. kernel-doc:: include/linux/dma-buf-map.h
130 .. kernel-doc:: drivers/dma-buf/dma-resv.c
131 :doc: Reservation Object Overview
133 .. kernel-doc:: drivers/dma-buf/dma-resv.c
136 .. kernel-doc:: include/linux/dma-resv.h
142 .. kernel-doc:: drivers/dma-buf/dma-fence.c
143 :doc: DMA fences overview
145 DMA Fence Cross-Driver Contract
146 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
148 .. kernel-doc:: drivers/dma-buf/dma-fence.c
149 :doc: fence cross-driver contract
151 DMA Fence Signalling Annotations
152 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
154 .. kernel-doc:: drivers/dma-buf/dma-fence.c
155 :doc: fence signalling annotation
157 DMA Fences Functions Reference
158 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
160 .. kernel-doc:: drivers/dma-buf/dma-fence.c
163 .. kernel-doc:: include/linux/dma-fence.h
166 Seqno Hardware Fences
167 ~~~~~~~~~~~~~~~~~~~~~
169 .. kernel-doc:: include/linux/seqno-fence.h
175 .. kernel-doc:: drivers/dma-buf/dma-fence-array.c
178 .. kernel-doc:: include/linux/dma-fence-array.h
181 DMA Fence uABI/Sync File
182 ~~~~~~~~~~~~~~~~~~~~~~~~
184 .. kernel-doc:: drivers/dma-buf/sync_file.c
187 .. kernel-doc:: include/linux/sync_file.h
190 Indefinite DMA Fences
191 ~~~~~~~~~~~~~~~~~~~~~
193 At various times struct dma_fence with an indefinite time until dma_fence_wait()
194 finishes have been proposed. Examples include:
196 * Future fences, used in HWC1 to signal when a buffer isn't used by the display
197 any longer, and created with the screen update that makes the buffer visible.
198 The time this fence completes is entirely under userspace's control.
200 * Proxy fences, proposed to handle &drm_syncobj for which the fence has not yet
201 been set. Used to asynchronously delay command submission.
203 * Userspace fences or gpu futexes, fine-grained locking within a command buffer
204 that userspace uses for synchronization across engines or with the CPU, which
205 are then imported as a DMA fence for integration into existing winsys
208 * Long-running compute command buffers, while still using traditional end of
209 batch DMA fences for memory management instead of context preemption DMA
210 fences which get reattached when the compute job is rescheduled.
212 Common to all these schemes is that userspace controls the dependencies of these
213 fences and controls when they fire. Mixing indefinite fences with normal
214 in-kernel DMA fences does not work, even when a fallback timeout is included to
215 protect against malicious userspace:
217 * Only the kernel knows about all DMA fence dependencies, userspace is not aware
218 of dependencies injected due to memory management or scheduler decisions.
220 * Only userspace knows about all dependencies in indefinite fences and when
221 exactly they will complete, the kernel has no visibility.
223 Furthermore the kernel has to be able to hold up userspace command submission
224 for memory management needs, which means we must support indefinite fences being
225 dependent upon DMA fences. If the kernel also support indefinite fences in the
226 kernel like a DMA fence, like any of the above proposal would, there is the
227 potential for deadlocks.
229 .. kernel-render:: DOT
230 :alt: Indefinite Fencing Dependency Cycle
231 :caption: Indefinite Fencing Dependency Cycle
233 digraph "Fencing Cycle" {
234 node [shape=box bgcolor=grey style=filled]
235 kernel [label="Kernel DMA Fences"]
236 userspace [label="userspace controlled fences"]
237 kernel -> userspace [label="memory management"]
238 userspace -> kernel [label="Future fence, fence proxy, ..."]
240 { rank=same; kernel userspace }
243 This means that the kernel might accidentally create deadlocks
244 through memory management dependencies which userspace is unaware of, which
245 randomly hangs workloads until the timeout kicks in. Workloads, which from
246 userspace's perspective, do not contain a deadlock. In such a mixed fencing
247 architecture there is no single entity with knowledge of all dependencies.
248 Thefore preventing such deadlocks from within the kernel is not possible.
250 The only solution to avoid dependencies loops is by not allowing indefinite
251 fences in the kernel. This means:
253 * No future fences, proxy fences or userspace fences imported as DMA fences,
254 with or without a timeout.
256 * No DMA fences that signal end of batchbuffer for command submission where
257 userspace is allowed to use userspace fencing or long running compute
258 workloads. This also means no implicit fencing for shared buffers in these