xref: /linux/drivers/md/Kconfig (revision cbac924200b838cfb8d8b1415113d788089dc50b)
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	select BLOCK_HOLDER_DEPRECATED if SYSFS
19	help
20	  This driver lets you combine several hard disk partitions into one
21	  logical block device. This can be used to simply append one
22	  partition to another one or to combine several redundant hard disks
23	  into a RAID1/4/5 device so as to provide protection against hard
24	  disk failures. This is called "Software RAID" since the combining of
25	  the partitions is done by the kernel. "Hardware RAID" means that the
26	  combining is done by a dedicated controller; if you have such a
27	  controller, you do not need to say Y here.
28
29	  More information about Software RAID on Linux is contained in the
30	  Software RAID mini-HOWTO, available from
31	  <https://www.tldp.org/docs.html#howto>. There you will also learn
32	  where to get the supporting user space utilities raidtools.
33
34	  If unsure, say N.
35
36config MD_AUTODETECT
37	bool "Autodetect RAID arrays during kernel boot"
38	depends on BLK_DEV_MD=y
39	default y
40	help
41	  If you say Y here, then the kernel will try to autodetect raid
42	  arrays as part of its boot process.
43
44	  If you don't use raid and say Y, this autodetection can cause
45	  a several-second delay in the boot time due to various
46	  synchronisation steps that are part of this step.
47
48	  If unsure, say Y.
49
50config MD_LINEAR
51	tristate "Linear (append) mode (deprecated)"
52	depends on BLK_DEV_MD
53	help
54	  If you say Y here, then your multiple devices driver will be able to
55	  use the so-called linear mode, i.e. it will combine the hard disk
56	  partitions by simply appending one to the other.
57
58	  To compile this as a module, choose M here: the module
59	  will be called linear.
60
61	  If unsure, say Y.
62
63config MD_RAID0
64	tristate "RAID-0 (striping) mode"
65	depends on BLK_DEV_MD
66	help
67	  If you say Y here, then your multiple devices driver will be able to
68	  use the so-called raid0 mode, i.e. it will combine the hard disk
69	  partitions into one logical device in such a fashion as to fill them
70	  up evenly, one chunk here and one chunk there. This will increase
71	  the throughput rate if the partitions reside on distinct disks.
72
73	  Information about Software RAID on Linux is contained in the
74	  Software-RAID mini-HOWTO, available from
75	  <https://www.tldp.org/docs.html#howto>. There you will also
76	  learn where to get the supporting user space utilities raidtools.
77
78	  To compile this as a module, choose M here: the module
79	  will be called raid0.
80
81	  If unsure, say Y.
82
83config MD_RAID1
84	tristate "RAID-1 (mirroring) mode"
85	depends on BLK_DEV_MD
86	help
87	  A RAID-1 set consists of several disk drives which are exact copies
88	  of each other.  In the event of a mirror failure, the RAID driver
89	  will continue to use the operational mirrors in the set, providing
90	  an error free MD (multiple device) to the higher levels of the
91	  kernel.  In a set with N drives, the available space is the capacity
92	  of a single drive, and the set protects against a failure of (N - 1)
93	  drives.
94
95	  Information about Software RAID on Linux is contained in the
96	  Software-RAID mini-HOWTO, available from
97	  <https://www.tldp.org/docs.html#howto>.  There you will also
98	  learn where to get the supporting user space utilities raidtools.
99
100	  If you want to use such a RAID-1 set, say Y.  To compile this code
101	  as a module, choose M here: the module will be called raid1.
102
103	  If unsure, say Y.
104
105config MD_RAID10
106	tristate "RAID-10 (mirrored striping) mode"
107	depends on BLK_DEV_MD
108	help
109	  RAID-10 provides a combination of striping (RAID-0) and
110	  mirroring (RAID-1) with easier configuration and more flexible
111	  layout.
112	  Unlike RAID-0, but like RAID-1, RAID-10 requires all devices to
113	  be the same size (or at least, only as much as the smallest device
114	  will be used).
115	  RAID-10 provides a variety of layouts that provide different levels
116	  of redundancy and performance.
117
118	  RAID-10 requires mdadm-1.7.0 or later, available at:
119
120	  https://www.kernel.org/pub/linux/utils/raid/mdadm/
121
122	  If unsure, say Y.
123
124config MD_RAID456
125	tristate "RAID-4/RAID-5/RAID-6 mode"
126	depends on BLK_DEV_MD
127	select RAID6_PQ
128	select LIBCRC32C
129	select ASYNC_MEMCPY
130	select ASYNC_XOR
131	select ASYNC_PQ
132	select ASYNC_RAID6_RECOV
133	help
134	  A RAID-5 set of N drives with a capacity of C MB per drive provides
135	  the capacity of C * (N - 1) MB, and protects against a failure
136	  of a single drive. For a given sector (row) number, (N - 1) drives
137	  contain data sectors, and one drive contains the parity protection.
138	  For a RAID-4 set, the parity blocks are present on a single drive,
139	  while a RAID-5 set distributes the parity across the drives in one
140	  of the available parity distribution methods.
141
142	  A RAID-6 set of N drives with a capacity of C MB per drive
143	  provides the capacity of C * (N - 2) MB, and protects
144	  against a failure of any two drives. For a given sector
145	  (row) number, (N - 2) drives contain data sectors, and two
146	  drives contains two independent redundancy syndromes.  Like
147	  RAID-5, RAID-6 distributes the syndromes across the drives
148	  in one of the available parity distribution methods.
149
150	  Information about Software RAID on Linux is contained in the
151	  Software-RAID mini-HOWTO, available from
152	  <https://www.tldp.org/docs.html#howto>. There you will also
153	  learn where to get the supporting user space utilities raidtools.
154
155	  If you want to use such a RAID-4/RAID-5/RAID-6 set, say Y.  To
156	  compile this code as a module, choose M here: the module
157	  will be called raid456.
158
159	  If unsure, say Y.
160
161config MD_MULTIPATH
162	tristate "Multipath I/O support (deprecated)"
163	depends on BLK_DEV_MD
164	help
165	  MD_MULTIPATH provides a simple multi-path personality for use
166	  the MD framework.  It is not under active development.  New
167	  projects should consider using DM_MULTIPATH which has more
168	  features and more testing.
169
170	  If unsure, say N.
171
172config MD_FAULTY
173	tristate "Faulty test module for MD (deprecated)"
174	depends on BLK_DEV_MD
175	help
176	  The "faulty" module allows for a block device that occasionally returns
177	  read or write errors.  It is useful for testing.
178
179	  In unsure, say N.
180
181
182config MD_CLUSTER
183	tristate "Cluster Support for MD"
184	depends on BLK_DEV_MD
185	depends on DLM
186	default n
187	help
188	Clustering support for MD devices. This enables locking and
189	synchronization across multiple systems on the cluster, so all
190	nodes in the cluster can access the MD devices simultaneously.
191
192	This brings the redundancy (and uptime) of RAID levels across the
193	nodes of the cluster. Currently, it can work with raid1 and raid10
194	(limited support).
195
196	If unsure, say N.
197
198source "drivers/md/bcache/Kconfig"
199
200config BLK_DEV_DM_BUILTIN
201	bool
202
203config BLK_DEV_DM
204	tristate "Device mapper support"
205	select BLOCK_HOLDER_DEPRECATED if SYSFS
206	select BLK_DEV_DM_BUILTIN
207	select BLK_MQ_STACKING
208	depends on DAX || DAX=n
209	help
210	  Device-mapper is a low level volume manager.  It works by allowing
211	  people to specify mappings for ranges of logical sectors.  Various
212	  mapping types are available, in addition people may write their own
213	  modules containing custom mappings if they wish.
214
215	  Higher level volume managers such as LVM2 use this driver.
216
217	  To compile this as a module, choose M here: the module will be
218	  called dm-mod.
219
220	  If unsure, say N.
221
222config DM_DEBUG
223	bool "Device mapper debugging support"
224	depends on BLK_DEV_DM
225	help
226	  Enable this for messages that may help debug device-mapper problems.
227
228	  If unsure, say N.
229
230config DM_BUFIO
231       tristate
232       depends on BLK_DEV_DM
233	help
234	 This interface allows you to do buffered I/O on a device and acts
235	 as a cache, holding recently-read blocks in memory and performing
236	 delayed writes.
237
238config DM_DEBUG_BLOCK_MANAGER_LOCKING
239       bool "Block manager locking"
240       depends on DM_BUFIO
241	help
242	 Block manager locking can catch various metadata corruption issues.
243
244	 If unsure, say N.
245
246config DM_DEBUG_BLOCK_STACK_TRACING
247       bool "Keep stack trace of persistent data block lock holders"
248       depends on STACKTRACE_SUPPORT && DM_DEBUG_BLOCK_MANAGER_LOCKING
249       select STACKTRACE
250	help
251	 Enable this for messages that may help debug problems with the
252	 block manager locking used by thin provisioning and caching.
253
254	 If unsure, say N.
255
256config DM_BIO_PRISON
257       tristate
258       depends on BLK_DEV_DM
259	help
260	 Some bio locking schemes used by other device-mapper targets
261	 including thin provisioning.
262
263source "drivers/md/persistent-data/Kconfig"
264
265config DM_UNSTRIPED
266       tristate "Unstriped target"
267       depends on BLK_DEV_DM
268	help
269	  Unstripes I/O so it is issued solely on a single drive in a HW
270	  RAID0 or dm-striped target.
271
272config DM_CRYPT
273	tristate "Crypt target support"
274	depends on BLK_DEV_DM
275	depends on (ENCRYPTED_KEYS || ENCRYPTED_KEYS=n)
276	depends on (TRUSTED_KEYS || TRUSTED_KEYS=n)
277	select CRYPTO
278	select CRYPTO_CBC
279	select CRYPTO_ESSIV
280	help
281	  This device-mapper target allows you to create a device that
282	  transparently encrypts the data on it. You'll need to activate
283	  the ciphers you're going to use in the cryptoapi configuration.
284
285	  For further information on dm-crypt and userspace tools see:
286	  <https://gitlab.com/cryptsetup/cryptsetup/wikis/DMCrypt>
287
288	  To compile this code as a module, choose M here: the module will
289	  be called dm-crypt.
290
291	  If unsure, say N.
292
293config DM_SNAPSHOT
294       tristate "Snapshot target"
295       depends on BLK_DEV_DM
296       select DM_BUFIO
297	help
298	 Allow volume managers to take writable snapshots of a device.
299
300config DM_THIN_PROVISIONING
301       tristate "Thin provisioning target"
302       depends on BLK_DEV_DM
303       select DM_PERSISTENT_DATA
304       select DM_BIO_PRISON
305	help
306	 Provides thin provisioning and snapshots that share a data store.
307
308config DM_CACHE
309       tristate "Cache target (EXPERIMENTAL)"
310       depends on BLK_DEV_DM
311       default n
312       select DM_PERSISTENT_DATA
313       select DM_BIO_PRISON
314	help
315	 dm-cache attempts to improve performance of a block device by
316	 moving frequently used data to a smaller, higher performance
317	 device.  Different 'policy' plugins can be used to change the
318	 algorithms used to select which blocks are promoted, demoted,
319	 cleaned etc.  It supports writeback and writethrough modes.
320
321config DM_CACHE_SMQ
322       tristate "Stochastic MQ Cache Policy (EXPERIMENTAL)"
323       depends on DM_CACHE
324       default y
325	help
326	 A cache policy that uses a multiqueue ordered by recent hits
327	 to select which blocks should be promoted and demoted.
328	 This is meant to be a general purpose policy.  It prioritises
329	 reads over writes.  This SMQ policy (vs MQ) offers the promise
330	 of less memory utilization, improved performance and increased
331	 adaptability in the face of changing workloads.
332
333config DM_WRITECACHE
334	tristate "Writecache target"
335	depends on BLK_DEV_DM
336	help
337	   The writecache target caches writes on persistent memory or SSD.
338	   It is intended for databases or other programs that need extremely
339	   low commit latency.
340
341	   The writecache target doesn't cache reads because reads are supposed
342	   to be cached in standard RAM.
343
344config DM_EBS
345	tristate "Emulated block size target (EXPERIMENTAL)"
346	depends on BLK_DEV_DM && !HIGHMEM
347	select DM_BUFIO
348	help
349	  dm-ebs emulates smaller logical block size on backing devices
350	  with larger ones (e.g. 512 byte sectors on 4K native disks).
351
352config DM_ERA
353       tristate "Era target (EXPERIMENTAL)"
354       depends on BLK_DEV_DM
355       default n
356       select DM_PERSISTENT_DATA
357       select DM_BIO_PRISON
358	help
359	 dm-era tracks which parts of a block device are written to
360	 over time.  Useful for maintaining cache coherency when using
361	 vendor snapshots.
362
363config DM_CLONE
364       tristate "Clone target (EXPERIMENTAL)"
365       depends on BLK_DEV_DM
366       default n
367       select DM_PERSISTENT_DATA
368	help
369	 dm-clone produces a one-to-one copy of an existing, read-only source
370	 device into a writable destination device. The cloned device is
371	 visible/mountable immediately and the copy of the source device to the
372	 destination device happens in the background, in parallel with user
373	 I/O.
374
375	 If unsure, say N.
376
377config DM_MIRROR
378       tristate "Mirror target"
379       depends on BLK_DEV_DM
380	help
381	 Allow volume managers to mirror logical volumes, also
382	 needed for live data migration tools such as 'pvmove'.
383
384config DM_LOG_USERSPACE
385	tristate "Mirror userspace logging"
386	depends on DM_MIRROR && NET
387	select CONNECTOR
388	help
389	  The userspace logging module provides a mechanism for
390	  relaying the dm-dirty-log API to userspace.  Log designs
391	  which are more suited to userspace implementation (e.g.
392	  shared storage logs) or experimental logs can be implemented
393	  by leveraging this framework.
394
395config DM_RAID
396       tristate "RAID 1/4/5/6/10 target"
397       depends on BLK_DEV_DM
398       select MD_RAID0
399       select MD_RAID1
400       select MD_RAID10
401       select MD_RAID456
402       select BLK_DEV_MD
403	help
404	 A dm target that supports RAID1, RAID10, RAID4, RAID5 and RAID6 mappings
405
406	 A RAID-5 set of N drives with a capacity of C MB per drive provides
407	 the capacity of C * (N - 1) MB, and protects against a failure
408	 of a single drive. For a given sector (row) number, (N - 1) drives
409	 contain data sectors, and one drive contains the parity protection.
410	 For a RAID-4 set, the parity blocks are present on a single drive,
411	 while a RAID-5 set distributes the parity across the drives in one
412	 of the available parity distribution methods.
413
414	 A RAID-6 set of N drives with a capacity of C MB per drive
415	 provides the capacity of C * (N - 2) MB, and protects
416	 against a failure of any two drives. For a given sector
417	 (row) number, (N - 2) drives contain data sectors, and two
418	 drives contains two independent redundancy syndromes.  Like
419	 RAID-5, RAID-6 distributes the syndromes across the drives
420	 in one of the available parity distribution methods.
421
422config DM_ZERO
423	tristate "Zero target"
424	depends on BLK_DEV_DM
425	help
426	  A target that discards writes, and returns all zeroes for
427	  reads.  Useful in some recovery situations.
428
429config DM_MULTIPATH
430	tristate "Multipath target"
431	depends on BLK_DEV_DM
432	# nasty syntax but means make DM_MULTIPATH independent
433	# of SCSI_DH if the latter isn't defined but if
434	# it is, DM_MULTIPATH must depend on it.  We get a build
435	# error if SCSI_DH=m and DM_MULTIPATH=y
436	depends on !SCSI_DH || SCSI
437	help
438	  Allow volume managers to support multipath hardware.
439
440config DM_MULTIPATH_QL
441	tristate "I/O Path Selector based on the number of in-flight I/Os"
442	depends on DM_MULTIPATH
443	help
444	  This path selector is a dynamic load balancer which selects
445	  the path with the least number of in-flight I/Os.
446
447	  If unsure, say N.
448
449config DM_MULTIPATH_ST
450	tristate "I/O Path Selector based on the service time"
451	depends on DM_MULTIPATH
452	help
453	  This path selector is a dynamic load balancer which selects
454	  the path expected to complete the incoming I/O in the shortest
455	  time.
456
457	  If unsure, say N.
458
459config DM_MULTIPATH_HST
460	tristate "I/O Path Selector based on historical service time"
461	depends on DM_MULTIPATH
462	help
463	  This path selector is a dynamic load balancer which selects
464	  the path expected to complete the incoming I/O in the shortest
465	  time by comparing estimated service time (based on historical
466	  service time).
467
468	  If unsure, say N.
469
470config DM_MULTIPATH_IOA
471	tristate "I/O Path Selector based on CPU submission"
472	depends on DM_MULTIPATH
473	help
474	  This path selector selects the path based on the CPU the IO is
475	  executed on and the CPU to path mapping setup at path addition time.
476
477	  If unsure, say N.
478
479config DM_DELAY
480	tristate "I/O delaying target"
481	depends on BLK_DEV_DM
482	help
483	A target that delays reads and/or writes and can send
484	them to different devices.  Useful for testing.
485
486	If unsure, say N.
487
488config DM_DUST
489	tristate "Bad sector simulation target"
490	depends on BLK_DEV_DM
491	help
492	A target that simulates bad sector behavior.
493	Useful for testing.
494
495	If unsure, say N.
496
497config DM_INIT
498	bool "DM \"dm-mod.create=\" parameter support"
499	depends on BLK_DEV_DM=y
500	help
501	Enable "dm-mod.create=" parameter to create mapped devices at init time.
502	This option is useful to allow mounting rootfs without requiring an
503	initramfs.
504	See Documentation/admin-guide/device-mapper/dm-init.rst for dm-mod.create="..."
505	format.
506
507	If unsure, say N.
508
509config DM_UEVENT
510	bool "DM uevents"
511	depends on BLK_DEV_DM
512	help
513	Generate udev events for DM events.
514
515config DM_FLAKEY
516       tristate "Flakey target"
517       depends on BLK_DEV_DM
518	help
519	 A target that intermittently fails I/O for debugging purposes.
520
521config DM_VERITY
522	tristate "Verity target support"
523	depends on BLK_DEV_DM
524	select CRYPTO
525	select CRYPTO_HASH
526	select DM_BUFIO
527	help
528	  This device-mapper target creates a read-only device that
529	  transparently validates the data on one underlying device against
530	  a pre-generated tree of cryptographic checksums stored on a second
531	  device.
532
533	  You'll need to activate the digests you're going to use in the
534	  cryptoapi configuration.
535
536	  To compile this code as a module, choose M here: the module will
537	  be called dm-verity.
538
539	  If unsure, say N.
540
541config DM_VERITY_VERIFY_ROOTHASH_SIG
542	def_bool n
543	bool "Verity data device root hash signature verification support"
544	depends on DM_VERITY
545	select SYSTEM_DATA_VERIFICATION
546	help
547	  Add ability for dm-verity device to be validated if the
548	  pre-generated tree of cryptographic checksums passed has a pkcs#7
549	  signature file that can validate the roothash of the tree.
550
551	  By default, rely on the builtin trusted keyring.
552
553	  If unsure, say N.
554
555config DM_VERITY_VERIFY_ROOTHASH_SIG_SECONDARY_KEYRING
556	bool "Verity data device root hash signature verification with secondary keyring"
557	depends on DM_VERITY_VERIFY_ROOTHASH_SIG
558	depends on SECONDARY_TRUSTED_KEYRING
559	help
560	  Rely on the secondary trusted keyring to verify dm-verity signatures.
561
562	  If unsure, say N.
563
564config DM_VERITY_FEC
565	bool "Verity forward error correction support"
566	depends on DM_VERITY
567	select REED_SOLOMON
568	select REED_SOLOMON_DEC8
569	help
570	  Add forward error correction support to dm-verity. This option
571	  makes it possible to use pre-generated error correction data to
572	  recover from corrupted blocks.
573
574	  If unsure, say N.
575
576config DM_SWITCH
577	tristate "Switch target support (EXPERIMENTAL)"
578	depends on BLK_DEV_DM
579	help
580	  This device-mapper target creates a device that supports an arbitrary
581	  mapping of fixed-size regions of I/O across a fixed set of paths.
582	  The path used for any specific region can be switched dynamically
583	  by sending the target a message.
584
585	  To compile this code as a module, choose M here: the module will
586	  be called dm-switch.
587
588	  If unsure, say N.
589
590config DM_LOG_WRITES
591	tristate "Log writes target support"
592	depends on BLK_DEV_DM
593	help
594	  This device-mapper target takes two devices, one device to use
595	  normally, one to log all write operations done to the first device.
596	  This is for use by file system developers wishing to verify that
597	  their fs is writing a consistent file system at all times by allowing
598	  them to replay the log in a variety of ways and to check the
599	  contents.
600
601	  To compile this code as a module, choose M here: the module will
602	  be called dm-log-writes.
603
604	  If unsure, say N.
605
606config DM_INTEGRITY
607	tristate "Integrity target support"
608	depends on BLK_DEV_DM
609	select BLK_DEV_INTEGRITY
610	select DM_BUFIO
611	select CRYPTO
612	select CRYPTO_SKCIPHER
613	select ASYNC_XOR
614	select DM_AUDIT if AUDIT
615	help
616	  This device-mapper target emulates a block device that has
617	  additional per-sector tags that can be used for storing
618	  integrity information.
619
620	  This integrity target is used with the dm-crypt target to
621	  provide authenticated disk encryption or it can be used
622	  standalone.
623
624	  To compile this code as a module, choose M here: the module will
625	  be called dm-integrity.
626
627config DM_ZONED
628	tristate "Drive-managed zoned block device target support"
629	depends on BLK_DEV_DM
630	depends on BLK_DEV_ZONED
631	select CRC32
632	help
633	  This device-mapper target takes a host-managed or host-aware zoned
634	  block device and exposes most of its capacity as a regular block
635	  device (drive-managed zoned block device) without any write
636	  constraints. This is mainly intended for use with file systems that
637	  do not natively support zoned block devices but still want to
638	  benefit from the increased capacity offered by SMR disks. Other uses
639	  by applications using raw block devices (for example object stores)
640	  are also possible.
641
642	  To compile this code as a module, choose M here: the module will
643	  be called dm-zoned.
644
645	  If unsure, say N.
646
647config DM_AUDIT
648	bool "DM audit events"
649	depends on AUDIT
650	help
651	  Generate audit events for device-mapper.
652
653	  Enables audit logging of several security relevant events in the
654	  particular device-mapper targets, especially the integrity target.
655
656endif # MD
657