1.. SPDX-License-Identifier: GPL-2.0 2 3=========================================== 4Userspace block device driver (ublk driver) 5=========================================== 6 7Overview 8======== 9 10ublk is a generic framework for implementing block device logic from userspace. 11The motivation behind it is that moving virtual block drivers into userspace, 12such as loop, nbd and similar can be very helpful. It can help to implement 13new virtual block device such as ublk-qcow2 (there are several attempts of 14implementing qcow2 driver in kernel). 15 16Userspace block devices are attractive because: 17 18- They can be written many programming languages. 19- They can use libraries that are not available in the kernel. 20- They can be debugged with tools familiar to application developers. 21- Crashes do not kernel panic the machine. 22- Bugs are likely to have a lower security impact than bugs in kernel 23 code. 24- They can be installed and updated independently of the kernel. 25- They can be used to simulate block device easily with user specified 26 parameters/setting for test/debug purpose 27 28ublk block device (``/dev/ublkb*``) is added by ublk driver. Any IO request 29on the device will be forwarded to ublk userspace program. For convenience, 30in this document, ``ublk server`` refers to generic ublk userspace 31program. ``ublksrv`` [#userspace]_ is one of such implementation. It 32provides ``libublksrv`` [#userspace_lib]_ library for developing specific 33user block device conveniently, while also generic type block device is 34included, such as loop and null. Richard W.M. Jones wrote userspace nbd device 35``nbdublk`` [#userspace_nbdublk]_ based on ``libublksrv`` [#userspace_lib]_. 36 37After the IO is handled by userspace, the result is committed back to the 38driver, thus completing the request cycle. This way, any specific IO handling 39logic is totally done by userspace, such as loop's IO handling, NBD's IO 40communication, or qcow2's IO mapping. 41 42``/dev/ublkb*`` is driven by blk-mq request-based driver. Each request is 43assigned by one queue wide unique tag. ublk server assigns unique tag to each 44IO too, which is 1:1 mapped with IO of ``/dev/ublkb*``. 45 46Both the IO request forward and IO handling result committing are done via 47``io_uring`` passthrough command; that is why ublk is also one io_uring based 48block driver. It has been observed that using io_uring passthrough command can 49give better IOPS than block IO; which is why ublk is one of high performance 50implementation of userspace block device: not only IO request communication is 51done by io_uring, but also the preferred IO handling in ublk server is io_uring 52based approach too. 53 54ublk provides control interface to set/get ublk block device parameters. 55The interface is extendable and kabi compatible: basically any ublk request 56queue's parameter or ublk generic feature parameters can be set/get via the 57interface. Thus, ublk is generic userspace block device framework. 58For example, it is easy to setup a ublk device with specified block 59parameters from userspace. 60 61Using ublk 62========== 63 64ublk requires userspace ublk server to handle real block device logic. 65 66Below is example of using ``ublksrv`` to provide ublk-based loop device. 67 68- add a device:: 69 70 ublk add -t loop -f ublk-loop.img 71 72- format with xfs, then use it:: 73 74 mkfs.xfs /dev/ublkb0 75 mount /dev/ublkb0 /mnt 76 # do anything. all IOs are handled by io_uring 77 ... 78 umount /mnt 79 80- list the devices with their info:: 81 82 ublk list 83 84- delete the device:: 85 86 ublk del -a 87 ublk del -n $ublk_dev_id 88 89See usage details in README of ``ublksrv`` [#userspace_readme]_. 90 91Design 92====== 93 94Control plane 95------------- 96 97ublk driver provides global misc device node (``/dev/ublk-control``) for 98managing and controlling ublk devices with help of several control commands: 99 100- ``UBLK_CMD_ADD_DEV`` 101 102 Add a ublk char device (``/dev/ublkc*``) which is talked with ublk server 103 WRT IO command communication. Basic device info is sent together with this 104 command. It sets UAPI structure of ``ublksrv_ctrl_dev_info``, 105 such as ``nr_hw_queues``, ``queue_depth``, and max IO request buffer size, 106 for which the info is negotiated with the driver and sent back to the server. 107 When this command is completed, the basic device info is immutable. 108 109- ``UBLK_CMD_SET_PARAMS`` / ``UBLK_CMD_GET_PARAMS`` 110 111 Set or get parameters of the device, which can be either generic feature 112 related, or request queue limit related, but can't be IO logic specific, 113 because the driver does not handle any IO logic. This command has to be 114 sent before sending ``UBLK_CMD_START_DEV``. 115 116- ``UBLK_CMD_START_DEV`` 117 118 After the server prepares userspace resources (such as creating per-queue 119 pthread & io_uring for handling ublk IO), this command is sent to the 120 driver for allocating & exposing ``/dev/ublkb*``. Parameters set via 121 ``UBLK_CMD_SET_PARAMS`` are applied for creating the device. 122 123- ``UBLK_CMD_STOP_DEV`` 124 125 Halt IO on ``/dev/ublkb*`` and remove the device. When this command returns, 126 ublk server will release resources (such as destroying per-queue pthread & 127 io_uring). 128 129- ``UBLK_CMD_DEL_DEV`` 130 131 Remove ``/dev/ublkc*``. When this command returns, the allocated ublk device 132 number can be reused. 133 134- ``UBLK_CMD_GET_QUEUE_AFFINITY`` 135 136 When ``/dev/ublkc`` is added, the driver creates block layer tagset, so 137 that each queue's affinity info is available. The server sends 138 ``UBLK_CMD_GET_QUEUE_AFFINITY`` to retrieve queue affinity info. It can 139 set up the per-queue context efficiently, such as bind affine CPUs with IO 140 pthread and try to allocate buffers in IO thread context. 141 142- ``UBLK_CMD_GET_DEV_INFO`` 143 144 For retrieving device info via ``ublksrv_ctrl_dev_info``. It is the server's 145 responsibility to save IO target specific info in userspace. 146 147- ``UBLK_CMD_START_USER_RECOVERY`` 148 149 This command is valid if ``UBLK_F_USER_RECOVERY`` feature is enabled. This 150 command is accepted after the old process has exited, ublk device is quiesced 151 and ``/dev/ublkc*`` is released. User should send this command before he starts 152 a new process which re-opens ``/dev/ublkc*``. When this command returns, the 153 ublk device is ready for the new process. 154 155- ``UBLK_CMD_END_USER_RECOVERY`` 156 157 This command is valid if ``UBLK_F_USER_RECOVERY`` feature is enabled. This 158 command is accepted after ublk device is quiesced and a new process has 159 opened ``/dev/ublkc*`` and get all ublk queues be ready. When this command 160 returns, ublk device is unquiesced and new I/O requests are passed to the 161 new process. 162 163- user recovery feature description 164 165 Two new features are added for user recovery: ``UBLK_F_USER_RECOVERY`` and 166 ``UBLK_F_USER_RECOVERY_REISSUE``. 167 168 With ``UBLK_F_USER_RECOVERY`` set, after one ubq_daemon(ublk server's io 169 handler) is dying, ublk does not delete ``/dev/ublkb*`` during the whole 170 recovery stage and ublk device ID is kept. It is ublk server's 171 responsibility to recover the device context by its own knowledge. 172 Requests which have not been issued to userspace are requeued. Requests 173 which have been issued to userspace are aborted. 174 175 With ``UBLK_F_USER_RECOVERY_REISSUE`` set, after one ubq_daemon(ublk 176 server's io handler) is dying, contrary to ``UBLK_F_USER_RECOVERY``, 177 requests which have been issued to userspace are requeued and will be 178 re-issued to the new process after handling ``UBLK_CMD_END_USER_RECOVERY``. 179 ``UBLK_F_USER_RECOVERY_REISSUE`` is designed for backends who tolerate 180 double-write since the driver may issue the same I/O request twice. It 181 might be useful to a read-only FS or a VM backend. 182 183Data plane 184---------- 185 186ublk server needs to create per-queue IO pthread & io_uring for handling IO 187commands via io_uring passthrough. The per-queue IO pthread 188focuses on IO handling and shouldn't handle any control & management 189tasks. 190 191The's IO is assigned by a unique tag, which is 1:1 mapping with IO 192request of ``/dev/ublkb*``. 193 194UAPI structure of ``ublksrv_io_desc`` is defined for describing each IO from 195the driver. A fixed mmaped area (array) on ``/dev/ublkc*`` is provided for 196exporting IO info to the server; such as IO offset, length, OP/flags and 197buffer address. Each ``ublksrv_io_desc`` instance can be indexed via queue id 198and IO tag directly. 199 200The following IO commands are communicated via io_uring passthrough command, 201and each command is only for forwarding the IO and committing the result 202with specified IO tag in the command data: 203 204- ``UBLK_IO_FETCH_REQ`` 205 206 Sent from the server IO pthread for fetching future incoming IO requests 207 destined to ``/dev/ublkb*``. This command is sent only once from the server 208 IO pthread for ublk driver to setup IO forward environment. 209 210- ``UBLK_IO_COMMIT_AND_FETCH_REQ`` 211 212 When an IO request is destined to ``/dev/ublkb*``, the driver stores 213 the IO's ``ublksrv_io_desc`` to the specified mapped area; then the 214 previous received IO command of this IO tag (either ``UBLK_IO_FETCH_REQ`` 215 or ``UBLK_IO_COMMIT_AND_FETCH_REQ)`` is completed, so the server gets 216 the IO notification via io_uring. 217 218 After the server handles the IO, its result is committed back to the 219 driver by sending ``UBLK_IO_COMMIT_AND_FETCH_REQ`` back. Once ublkdrv 220 received this command, it parses the result and complete the request to 221 ``/dev/ublkb*``. In the meantime setup environment for fetching future 222 requests with the same IO tag. That is, ``UBLK_IO_COMMIT_AND_FETCH_REQ`` 223 is reused for both fetching request and committing back IO result. 224 225- ``UBLK_IO_NEED_GET_DATA`` 226 227 With ``UBLK_F_NEED_GET_DATA`` enabled, the WRITE request will be firstly 228 issued to ublk server without data copy. Then, IO backend of ublk server 229 receives the request and it can allocate data buffer and embed its addr 230 inside this new io command. After the kernel driver gets the command, 231 data copy is done from request pages to this backend's buffer. Finally, 232 backend receives the request again with data to be written and it can 233 truly handle the request. 234 235 ``UBLK_IO_NEED_GET_DATA`` adds one additional round-trip and one 236 io_uring_enter() syscall. Any user thinks that it may lower performance 237 should not enable UBLK_F_NEED_GET_DATA. ublk server pre-allocates IO 238 buffer for each IO by default. Any new project should try to use this 239 buffer to communicate with ublk driver. However, existing project may 240 break or not able to consume the new buffer interface; that's why this 241 command is added for backwards compatibility so that existing projects 242 can still consume existing buffers. 243 244- data copy between ublk server IO buffer and ublk block IO request 245 246 The driver needs to copy the block IO request pages into the server buffer 247 (pages) first for WRITE before notifying the server of the coming IO, so 248 that the server can handle WRITE request. 249 250 When the server handles READ request and sends 251 ``UBLK_IO_COMMIT_AND_FETCH_REQ`` to the server, ublkdrv needs to copy 252 the server buffer (pages) read to the IO request pages. 253 254Future development 255================== 256 257Container-aware ublk deivice 258---------------------------- 259 260ublk driver doesn't handle any IO logic. Its function is well defined 261for now and very limited userspace interfaces are needed, which is also 262well defined too. It is possible to make ublk devices container-aware block 263devices in future as Stefan Hajnoczi suggested [#stefan]_, by removing 264ADMIN privilege. 265 266Zero copy 267--------- 268 269Zero copy is a generic requirement for nbd, fuse or similar drivers. A 270problem [#xiaoguang]_ Xiaoguang mentioned is that pages mapped to userspace 271can't be remapped any more in kernel with existing mm interfaces. This can 272occurs when destining direct IO to ``/dev/ublkb*``. Also, he reported that 273big requests (IO size >= 256 KB) may benefit a lot from zero copy. 274 275 276References 277========== 278 279.. [#userspace] https://github.com/ming1/ubdsrv 280 281.. [#userspace_lib] https://github.com/ming1/ubdsrv/tree/master/lib 282 283.. [#userspace_nbdublk] https://gitlab.com/rwmjones/libnbd/-/tree/nbdublk 284 285.. [#userspace_readme] https://github.com/ming1/ubdsrv/blob/master/README 286 287.. [#stefan] https://lore.kernel.org/linux-block/YoOr6jBfgVm8GvWg@stefanha-x1.localdomain/ 288 289.. [#xiaoguang] https://lore.kernel.org/linux-block/YoOr6jBfgVm8GvWg@stefanha-x1.localdomain/ 290