xref: /openbmc/linux/drivers/md/Kconfig (revision 6197e5b7)
1# SPDX-License-Identifier: GPL-2.0-only
2#
3# Block device driver configuration
4#
5
6menuconfig MD
7	bool "Multiple devices driver support (RAID and LVM)"
8	depends on BLOCK
9	select SRCU
10	help
11	  Support multiple physical spindles through a single logical device.
12	  Required for RAID and logical volume management.
13
14if MD
15
16config BLK_DEV_MD
17	tristate "RAID support"
18	help
19	  This driver lets you combine several hard disk partitions into one
20	  logical block device. This can be used to simply append one
21	  partition to another one or to combine several redundant hard disks
22	  into a RAID1/4/5 device so as to provide protection against hard
23	  disk failures. This is called "Software RAID" since the combining of
24	  the partitions is done by the kernel. "Hardware RAID" means that the
25	  combining is done by a dedicated controller; if you have such a
26	  controller, you do not need to say Y here.
27
28	  More information about Software RAID on Linux is contained in the
29	  Software RAID mini-HOWTO, available from
30	  <https://www.tldp.org/docs.html#howto>. There you will also learn
31	  where to get the supporting user space utilities raidtools.
32
33	  If unsure, say N.
34
35config MD_AUTODETECT
36	bool "Autodetect RAID arrays during kernel boot"
37	depends on BLK_DEV_MD=y
38	default y
39	help
40	  If you say Y here, then the kernel will try to autodetect raid
41	  arrays as part of its boot process.
42
43	  If you don't use raid and say Y, this autodetection can cause
44	  a several-second delay in the boot time due to various
45	  synchronisation steps that are part of this step.
46
47	  If unsure, say Y.
48
49config MD_LINEAR
50	tristate "Linear (append) mode"
51	depends on BLK_DEV_MD
52	help
53	  If you say Y here, then your multiple devices driver will be able to
54	  use the so-called linear mode, i.e. it will combine the hard disk
55	  partitions by simply appending one to the other.
56
57	  To compile this as a module, choose M here: the module
58	  will be called linear.
59
60	  If unsure, say Y.
61
62config MD_RAID0
63	tristate "RAID-0 (striping) mode"
64	depends on BLK_DEV_MD
65	help
66	  If you say Y here, then your multiple devices driver will be able to
67	  use the so-called raid0 mode, i.e. it will combine the hard disk
68	  partitions into one logical device in such a fashion as to fill them
69	  up evenly, one chunk here and one chunk there. This will increase
70	  the throughput rate if the partitions reside on distinct disks.
71
72	  Information about Software RAID on Linux is contained in the
73	  Software-RAID mini-HOWTO, available from
74	  <https://www.tldp.org/docs.html#howto>. There you will also
75	  learn where to get the supporting user space utilities raidtools.
76
77	  To compile this as a module, choose M here: the module
78	  will be called raid0.
79
80	  If unsure, say Y.
81
82config MD_RAID1
83	tristate "RAID-1 (mirroring) mode"
84	depends on BLK_DEV_MD
85	help
86	  A RAID-1 set consists of several disk drives which are exact copies
87	  of each other.  In the event of a mirror failure, the RAID driver
88	  will continue to use the operational mirrors in the set, providing
89	  an error free MD (multiple device) to the higher levels of the
90	  kernel.  In a set with N drives, the available space is the capacity
91	  of a single drive, and the set protects against a failure of (N - 1)
92	  drives.
93
94	  Information about Software RAID on Linux is contained in the
95	  Software-RAID mini-HOWTO, available from
96	  <https://www.tldp.org/docs.html#howto>.  There you will also
97	  learn where to get the supporting user space utilities raidtools.
98
99	  If you want to use such a RAID-1 set, say Y.  To compile this code
100	  as a module, choose M here: the module will be called raid1.
101
102	  If unsure, say Y.
103
104config MD_RAID10
105	tristate "RAID-10 (mirrored striping) mode"
106	depends on BLK_DEV_MD
107	help
108	  RAID-10 provides a combination of striping (RAID-0) and
109	  mirroring (RAID-1) with easier configuration and more flexible
110	  layout.
111	  Unlike RAID-0, but like RAID-1, RAID-10 requires all devices to
112	  be the same size (or at least, only as much as the smallest device
113	  will be used).
114	  RAID-10 provides a variety of layouts that provide different levels
115	  of redundancy and performance.
116
117	  RAID-10 requires mdadm-1.7.0 or later, available at:
118
119	  https://www.kernel.org/pub/linux/utils/raid/mdadm/
120
121	  If unsure, say Y.
122
123config MD_RAID456
124	tristate "RAID-4/RAID-5/RAID-6 mode"
125	depends on BLK_DEV_MD
126	select RAID6_PQ
127	select LIBCRC32C
128	select ASYNC_MEMCPY
129	select ASYNC_XOR
130	select ASYNC_PQ
131	select ASYNC_RAID6_RECOV
132	help
133	  A RAID-5 set of N drives with a capacity of C MB per drive provides
134	  the capacity of C * (N - 1) MB, and protects against a failure
135	  of a single drive. For a given sector (row) number, (N - 1) drives
136	  contain data sectors, and one drive contains the parity protection.
137	  For a RAID-4 set, the parity blocks are present on a single drive,
138	  while a RAID-5 set distributes the parity across the drives in one
139	  of the available parity distribution methods.
140
141	  A RAID-6 set of N drives with a capacity of C MB per drive
142	  provides the capacity of C * (N - 2) MB, and protects
143	  against a failure of any two drives. For a given sector
144	  (row) number, (N - 2) drives contain data sectors, and two
145	  drives contains two independent redundancy syndromes.  Like
146	  RAID-5, RAID-6 distributes the syndromes across the drives
147	  in one of the available parity distribution methods.
148
149	  Information about Software RAID on Linux is contained in the
150	  Software-RAID mini-HOWTO, available from
151	  <https://www.tldp.org/docs.html#howto>. There you will also
152	  learn where to get the supporting user space utilities raidtools.
153
154	  If you want to use such a RAID-4/RAID-5/RAID-6 set, say Y.  To
155	  compile this code as a module, choose M here: the module
156	  will be called raid456.
157
158	  If unsure, say Y.
159
160config MD_MULTIPATH
161	tristate "Multipath I/O support"
162	depends on BLK_DEV_MD
163	help
164	  MD_MULTIPATH provides a simple multi-path personality for use
165	  the MD framework.  It is not under active development.  New
166	  projects should consider using DM_MULTIPATH which has more
167	  features and more testing.
168
169	  If unsure, say N.
170
171config MD_FAULTY
172	tristate "Faulty test module for MD"
173	depends on BLK_DEV_MD
174	help
175	  The "faulty" module allows for a block device that occasionally returns
176	  read or write errors.  It is useful for testing.
177
178	  In unsure, say N.
179
180
181config MD_CLUSTER
182	tristate "Cluster Support for MD"
183	depends on BLK_DEV_MD
184	depends on DLM
185	default n
186	help
187	Clustering support for MD devices. This enables locking and
188	synchronization across multiple systems on the cluster, so all
189	nodes in the cluster can access the MD devices simultaneously.
190
191	This brings the redundancy (and uptime) of RAID levels across the
192	nodes of the cluster. Currently, it can work with raid1 and raid10
193	(limited support).
194
195	If unsure, say N.
196
197source "drivers/md/bcache/Kconfig"
198
199config BLK_DEV_DM_BUILTIN
200	bool
201
202config BLK_DEV_DM
203	tristate "Device mapper support"
204	select BLK_DEV_DM_BUILTIN
205	depends on DAX || DAX=n
206	help
207	  Device-mapper is a low level volume manager.  It works by allowing
208	  people to specify mappings for ranges of logical sectors.  Various
209	  mapping types are available, in addition people may write their own
210	  modules containing custom mappings if they wish.
211
212	  Higher level volume managers such as LVM2 use this driver.
213
214	  To compile this as a module, choose M here: the module will be
215	  called dm-mod.
216
217	  If unsure, say N.
218
219config DM_DEBUG
220	bool "Device mapper debugging support"
221	depends on BLK_DEV_DM
222	help
223	  Enable this for messages that may help debug device-mapper problems.
224
225	  If unsure, say N.
226
227config DM_BUFIO
228       tristate
229       depends on BLK_DEV_DM
230	help
231	 This interface allows you to do buffered I/O on a device and acts
232	 as a cache, holding recently-read blocks in memory and performing
233	 delayed writes.
234
235config DM_DEBUG_BLOCK_MANAGER_LOCKING
236       bool "Block manager locking"
237       depends on DM_BUFIO
238	help
239	 Block manager locking can catch various metadata corruption issues.
240
241	 If unsure, say N.
242
243config DM_DEBUG_BLOCK_STACK_TRACING
244       bool "Keep stack trace of persistent data block lock holders"
245       depends on STACKTRACE_SUPPORT && DM_DEBUG_BLOCK_MANAGER_LOCKING
246       select STACKTRACE
247	help
248	 Enable this for messages that may help debug problems with the
249	 block manager locking used by thin provisioning and caching.
250
251	 If unsure, say N.
252
253config DM_BIO_PRISON
254       tristate
255       depends on BLK_DEV_DM
256	help
257	 Some bio locking schemes used by other device-mapper targets
258	 including thin provisioning.
259
260source "drivers/md/persistent-data/Kconfig"
261
262config DM_UNSTRIPED
263       tristate "Unstriped target"
264       depends on BLK_DEV_DM
265	help
266	  Unstripes I/O so it is issued solely on a single drive in a HW
267	  RAID0 or dm-striped target.
268
269config DM_CRYPT
270	tristate "Crypt target support"
271	depends on BLK_DEV_DM
272	depends on (ENCRYPTED_KEYS || ENCRYPTED_KEYS=n)
273	select CRYPTO
274	select CRYPTO_CBC
275	select CRYPTO_ESSIV
276	help
277	  This device-mapper target allows you to create a device that
278	  transparently encrypts the data on it. You'll need to activate
279	  the ciphers you're going to use in the cryptoapi configuration.
280
281	  For further information on dm-crypt and userspace tools see:
282	  <https://gitlab.com/cryptsetup/cryptsetup/wikis/DMCrypt>
283
284	  To compile this code as a module, choose M here: the module will
285	  be called dm-crypt.
286
287	  If unsure, say N.
288
289config DM_SNAPSHOT
290       tristate "Snapshot target"
291       depends on BLK_DEV_DM
292       select DM_BUFIO
293	help
294	 Allow volume managers to take writable snapshots of a device.
295
296config DM_THIN_PROVISIONING
297       tristate "Thin provisioning target"
298       depends on BLK_DEV_DM
299       select DM_PERSISTENT_DATA
300       select DM_BIO_PRISON
301	help
302	 Provides thin provisioning and snapshots that share a data store.
303
304config DM_CACHE
305       tristate "Cache target (EXPERIMENTAL)"
306       depends on BLK_DEV_DM
307       default n
308       select DM_PERSISTENT_DATA
309       select DM_BIO_PRISON
310	help
311	 dm-cache attempts to improve performance of a block device by
312	 moving frequently used data to a smaller, higher performance
313	 device.  Different 'policy' plugins can be used to change the
314	 algorithms used to select which blocks are promoted, demoted,
315	 cleaned etc.  It supports writeback and writethrough modes.
316
317config DM_CACHE_SMQ
318       tristate "Stochastic MQ Cache Policy (EXPERIMENTAL)"
319       depends on DM_CACHE
320       default y
321	help
322	 A cache policy that uses a multiqueue ordered by recent hits
323	 to select which blocks should be promoted and demoted.
324	 This is meant to be a general purpose policy.  It prioritises
325	 reads over writes.  This SMQ policy (vs MQ) offers the promise
326	 of less memory utilization, improved performance and increased
327	 adaptability in the face of changing workloads.
328
329config DM_WRITECACHE
330	tristate "Writecache target"
331	depends on BLK_DEV_DM
332	help
333	   The writecache target caches writes on persistent memory or SSD.
334	   It is intended for databases or other programs that need extremely
335	   low commit latency.
336
337	   The writecache target doesn't cache reads because reads are supposed
338	   to be cached in standard RAM.
339
340config DM_EBS
341	tristate "Emulated block size target (EXPERIMENTAL)"
342	depends on BLK_DEV_DM
343	select DM_BUFIO
344	help
345	  dm-ebs emulates smaller logical block size on backing devices
346	  with larger ones (e.g. 512 byte sectors on 4K native disks).
347
348config DM_ERA
349       tristate "Era target (EXPERIMENTAL)"
350       depends on BLK_DEV_DM
351       default n
352       select DM_PERSISTENT_DATA
353       select DM_BIO_PRISON
354	help
355	 dm-era tracks which parts of a block device are written to
356	 over time.  Useful for maintaining cache coherency when using
357	 vendor snapshots.
358
359config DM_CLONE
360       tristate "Clone target (EXPERIMENTAL)"
361       depends on BLK_DEV_DM
362       default n
363       select DM_PERSISTENT_DATA
364	help
365	 dm-clone produces a one-to-one copy of an existing, read-only source
366	 device into a writable destination device. The cloned device is
367	 visible/mountable immediately and the copy of the source device to the
368	 destination device happens in the background, in parallel with user
369	 I/O.
370
371	 If unsure, say N.
372
373config DM_MIRROR
374       tristate "Mirror target"
375       depends on BLK_DEV_DM
376	help
377	 Allow volume managers to mirror logical volumes, also
378	 needed for live data migration tools such as 'pvmove'.
379
380config DM_LOG_USERSPACE
381	tristate "Mirror userspace logging"
382	depends on DM_MIRROR && NET
383	select CONNECTOR
384	help
385	  The userspace logging module provides a mechanism for
386	  relaying the dm-dirty-log API to userspace.  Log designs
387	  which are more suited to userspace implementation (e.g.
388	  shared storage logs) or experimental logs can be implemented
389	  by leveraging this framework.
390
391config DM_RAID
392       tristate "RAID 1/4/5/6/10 target"
393       depends on BLK_DEV_DM
394       select MD_RAID0
395       select MD_RAID1
396       select MD_RAID10
397       select MD_RAID456
398       select BLK_DEV_MD
399	help
400	 A dm target that supports RAID1, RAID10, RAID4, RAID5 and RAID6 mappings
401
402	 A RAID-5 set of N drives with a capacity of C MB per drive provides
403	 the capacity of C * (N - 1) MB, and protects against a failure
404	 of a single drive. For a given sector (row) number, (N - 1) drives
405	 contain data sectors, and one drive contains the parity protection.
406	 For a RAID-4 set, the parity blocks are present on a single drive,
407	 while a RAID-5 set distributes the parity across the drives in one
408	 of the available parity distribution methods.
409
410	 A RAID-6 set of N drives with a capacity of C MB per drive
411	 provides the capacity of C * (N - 2) MB, and protects
412	 against a failure of any two drives. For a given sector
413	 (row) number, (N - 2) drives contain data sectors, and two
414	 drives contains two independent redundancy syndromes.  Like
415	 RAID-5, RAID-6 distributes the syndromes across the drives
416	 in one of the available parity distribution methods.
417
418config DM_ZERO
419	tristate "Zero target"
420	depends on BLK_DEV_DM
421	help
422	  A target that discards writes, and returns all zeroes for
423	  reads.  Useful in some recovery situations.
424
425config DM_MULTIPATH
426	tristate "Multipath target"
427	depends on BLK_DEV_DM
428	# nasty syntax but means make DM_MULTIPATH independent
429	# of SCSI_DH if the latter isn't defined but if
430	# it is, DM_MULTIPATH must depend on it.  We get a build
431	# error if SCSI_DH=m and DM_MULTIPATH=y
432	depends on !SCSI_DH || SCSI
433	help
434	  Allow volume managers to support multipath hardware.
435
436config DM_MULTIPATH_QL
437	tristate "I/O Path Selector based on the number of in-flight I/Os"
438	depends on DM_MULTIPATH
439	help
440	  This path selector is a dynamic load balancer which selects
441	  the path with the least number of in-flight I/Os.
442
443	  If unsure, say N.
444
445config DM_MULTIPATH_ST
446	tristate "I/O Path Selector based on the service time"
447	depends on DM_MULTIPATH
448	help
449	  This path selector is a dynamic load balancer which selects
450	  the path expected to complete the incoming I/O in the shortest
451	  time.
452
453	  If unsure, say N.
454
455config DM_MULTIPATH_HST
456	tristate "I/O Path Selector based on historical service time"
457	depends on DM_MULTIPATH
458	help
459	  This path selector is a dynamic load balancer which selects
460	  the path expected to complete the incoming I/O in the shortest
461	  time by comparing estimated service time (based on historical
462	  service time).
463
464	  If unsure, say N.
465
466config DM_MULTIPATH_IOA
467	tristate "I/O Path Selector based on CPU submission"
468	depends on DM_MULTIPATH
469	help
470	  This path selector selects the path based on the CPU the IO is
471	  executed on and the CPU to path mapping setup at path addition time.
472
473	  If unsure, say N.
474
475config DM_DELAY
476	tristate "I/O delaying target"
477	depends on BLK_DEV_DM
478	help
479	A target that delays reads and/or writes and can send
480	them to different devices.  Useful for testing.
481
482	If unsure, say N.
483
484config DM_DUST
485	tristate "Bad sector simulation target"
486	depends on BLK_DEV_DM
487	help
488	A target that simulates bad sector behavior.
489	Useful for testing.
490
491	If unsure, say N.
492
493config DM_INIT
494	bool "DM \"dm-mod.create=\" parameter support"
495	depends on BLK_DEV_DM=y
496	help
497	Enable "dm-mod.create=" parameter to create mapped devices at init time.
498	This option is useful to allow mounting rootfs without requiring an
499	initramfs.
500	See Documentation/admin-guide/device-mapper/dm-init.rst for dm-mod.create="..."
501	format.
502
503	If unsure, say N.
504
505config DM_UEVENT
506	bool "DM uevents"
507	depends on BLK_DEV_DM
508	help
509	Generate udev events for DM events.
510
511config DM_FLAKEY
512       tristate "Flakey target"
513       depends on BLK_DEV_DM
514	help
515	 A target that intermittently fails I/O for debugging purposes.
516
517config DM_VERITY
518	tristate "Verity target support"
519	depends on BLK_DEV_DM
520	select CRYPTO
521	select CRYPTO_HASH
522	select DM_BUFIO
523	help
524	  This device-mapper target creates a read-only device that
525	  transparently validates the data on one underlying device against
526	  a pre-generated tree of cryptographic checksums stored on a second
527	  device.
528
529	  You'll need to activate the digests you're going to use in the
530	  cryptoapi configuration.
531
532	  To compile this code as a module, choose M here: the module will
533	  be called dm-verity.
534
535	  If unsure, say N.
536
537config DM_VERITY_VERIFY_ROOTHASH_SIG
538	def_bool n
539	bool "Verity data device root hash signature verification support"
540	depends on DM_VERITY
541	select SYSTEM_DATA_VERIFICATION
542	help
543	  Add ability for dm-verity device to be validated if the
544	  pre-generated tree of cryptographic checksums passed has a pkcs#7
545	  signature file that can validate the roothash of the tree.
546
547	  By default, rely on the builtin trusted keyring.
548
549	  If unsure, say N.
550
551config DM_VERITY_VERIFY_ROOTHASH_SIG_SECONDARY_KEYRING
552	bool "Verity data device root hash signature verification with secondary keyring"
553	depends on DM_VERITY_VERIFY_ROOTHASH_SIG
554	depends on SECONDARY_TRUSTED_KEYRING
555	help
556	  Rely on the secondary trusted keyring to verify dm-verity signatures.
557
558	  If unsure, say N.
559
560config DM_VERITY_FEC
561	bool "Verity forward error correction support"
562	depends on DM_VERITY
563	select REED_SOLOMON
564	select REED_SOLOMON_DEC8
565	help
566	  Add forward error correction support to dm-verity. This option
567	  makes it possible to use pre-generated error correction data to
568	  recover from corrupted blocks.
569
570	  If unsure, say N.
571
572config DM_SWITCH
573	tristate "Switch target support (EXPERIMENTAL)"
574	depends on BLK_DEV_DM
575	help
576	  This device-mapper target creates a device that supports an arbitrary
577	  mapping of fixed-size regions of I/O across a fixed set of paths.
578	  The path used for any specific region can be switched dynamically
579	  by sending the target a message.
580
581	  To compile this code as a module, choose M here: the module will
582	  be called dm-switch.
583
584	  If unsure, say N.
585
586config DM_LOG_WRITES
587	tristate "Log writes target support"
588	depends on BLK_DEV_DM
589	help
590	  This device-mapper target takes two devices, one device to use
591	  normally, one to log all write operations done to the first device.
592	  This is for use by file system developers wishing to verify that
593	  their fs is writing a consistent file system at all times by allowing
594	  them to replay the log in a variety of ways and to check the
595	  contents.
596
597	  To compile this code as a module, choose M here: the module will
598	  be called dm-log-writes.
599
600	  If unsure, say N.
601
602config DM_INTEGRITY
603	tristate "Integrity target support"
604	depends on BLK_DEV_DM
605	select BLK_DEV_INTEGRITY
606	select DM_BUFIO
607	select CRYPTO
608	select CRYPTO_SKCIPHER
609	select ASYNC_XOR
610	help
611	  This device-mapper target emulates a block device that has
612	  additional per-sector tags that can be used for storing
613	  integrity information.
614
615	  This integrity target is used with the dm-crypt target to
616	  provide authenticated disk encryption or it can be used
617	  standalone.
618
619	  To compile this code as a module, choose M here: the module will
620	  be called dm-integrity.
621
622config DM_ZONED
623	tristate "Drive-managed zoned block device target support"
624	depends on BLK_DEV_DM
625	depends on BLK_DEV_ZONED
626	select CRC32
627	help
628	  This device-mapper target takes a host-managed or host-aware zoned
629	  block device and exposes most of its capacity as a regular block
630	  device (drive-managed zoned block device) without any write
631	  constraints. This is mainly intended for use with file systems that
632	  do not natively support zoned block devices but still want to
633	  benefit from the increased capacity offered by SMR disks. Other uses
634	  by applications using raw block devices (for example object stores)
635	  are also possible.
636
637	  To compile this code as a module, choose M here: the module will
638	  be called dm-zoned.
639
640	  If unsure, say N.
641
642endif # MD
643