xref: /openbmc/linux/drivers/md/Kconfig (revision e847c767)
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	help
10	  Support multiple physical spindles through a single logical device.
11	  Required for RAID and logical volume management.
12
13if MD
14
15config BLK_DEV_MD
16	tristate "RAID support"
17	select BLOCK_HOLDER_DEPRECATED if SYSFS
18	# BLOCK_LEGACY_AUTOLOAD requirement should be removed
19	# after relevant mdadm enhancements - to make "names=yes"
20	# the default - are widely available.
21	select BLOCK_LEGACY_AUTOLOAD
22	help
23	  This driver lets you combine several hard disk partitions into one
24	  logical block device. This can be used to simply append one
25	  partition to another one or to combine several redundant hard disks
26	  into a RAID1/4/5 device so as to provide protection against hard
27	  disk failures. This is called "Software RAID" since the combining of
28	  the partitions is done by the kernel. "Hardware RAID" means that the
29	  combining is done by a dedicated controller; if you have such a
30	  controller, you do not need to say Y here.
31
32	  More information about Software RAID on Linux is contained in the
33	  Software RAID mini-HOWTO, available from
34	  <https://www.tldp.org/docs.html#howto>. There you will also learn
35	  where to get the supporting user space utilities raidtools.
36
37	  If unsure, say N.
38
39config MD_AUTODETECT
40	bool "Autodetect RAID arrays during kernel boot"
41	depends on BLK_DEV_MD=y
42	default y
43	help
44	  If you say Y here, then the kernel will try to autodetect raid
45	  arrays as part of its boot process.
46
47	  If you don't use raid and say Y, this autodetection can cause
48	  a several-second delay in the boot time due to various
49	  synchronisation steps that are part of this step.
50
51	  If unsure, say Y.
52
53config MD_LINEAR
54	tristate "Linear (append) mode (deprecated)"
55	depends on BLK_DEV_MD
56	help
57	  If you say Y here, then your multiple devices driver will be able to
58	  use the so-called linear mode, i.e. it will combine the hard disk
59	  partitions by simply appending one to the other.
60
61	  To compile this as a module, choose M here: the module
62	  will be called linear.
63
64	  If unsure, say Y.
65
66config MD_RAID0
67	tristate "RAID-0 (striping) mode"
68	depends on BLK_DEV_MD
69	help
70	  If you say Y here, then your multiple devices driver will be able to
71	  use the so-called raid0 mode, i.e. it will combine the hard disk
72	  partitions into one logical device in such a fashion as to fill them
73	  up evenly, one chunk here and one chunk there. This will increase
74	  the throughput rate if the partitions reside on distinct disks.
75
76	  Information about Software RAID on Linux is contained in the
77	  Software-RAID mini-HOWTO, available from
78	  <https://www.tldp.org/docs.html#howto>. There you will also
79	  learn where to get the supporting user space utilities raidtools.
80
81	  To compile this as a module, choose M here: the module
82	  will be called raid0.
83
84	  If unsure, say Y.
85
86config MD_RAID1
87	tristate "RAID-1 (mirroring) mode"
88	depends on BLK_DEV_MD
89	help
90	  A RAID-1 set consists of several disk drives which are exact copies
91	  of each other.  In the event of a mirror failure, the RAID driver
92	  will continue to use the operational mirrors in the set, providing
93	  an error free MD (multiple device) to the higher levels of the
94	  kernel.  In a set with N drives, the available space is the capacity
95	  of a single drive, and the set protects against a failure of (N - 1)
96	  drives.
97
98	  Information about Software RAID on Linux is contained in the
99	  Software-RAID mini-HOWTO, available from
100	  <https://www.tldp.org/docs.html#howto>.  There you will also
101	  learn where to get the supporting user space utilities raidtools.
102
103	  If you want to use such a RAID-1 set, say Y.  To compile this code
104	  as a module, choose M here: the module will be called raid1.
105
106	  If unsure, say Y.
107
108config MD_RAID10
109	tristate "RAID-10 (mirrored striping) mode"
110	depends on BLK_DEV_MD
111	help
112	  RAID-10 provides a combination of striping (RAID-0) and
113	  mirroring (RAID-1) with easier configuration and more flexible
114	  layout.
115	  Unlike RAID-0, but like RAID-1, RAID-10 requires all devices to
116	  be the same size (or at least, only as much as the smallest device
117	  will be used).
118	  RAID-10 provides a variety of layouts that provide different levels
119	  of redundancy and performance.
120
121	  RAID-10 requires mdadm-1.7.0 or later, available at:
122
123	  https://www.kernel.org/pub/linux/utils/raid/mdadm/
124
125	  If unsure, say Y.
126
127config MD_RAID456
128	tristate "RAID-4/RAID-5/RAID-6 mode"
129	depends on BLK_DEV_MD
130	select RAID6_PQ
131	select LIBCRC32C
132	select ASYNC_MEMCPY
133	select ASYNC_XOR
134	select ASYNC_PQ
135	select ASYNC_RAID6_RECOV
136	help
137	  A RAID-5 set of N drives with a capacity of C MB per drive provides
138	  the capacity of C * (N - 1) MB, and protects against a failure
139	  of a single drive. For a given sector (row) number, (N - 1) drives
140	  contain data sectors, and one drive contains the parity protection.
141	  For a RAID-4 set, the parity blocks are present on a single drive,
142	  while a RAID-5 set distributes the parity across the drives in one
143	  of the available parity distribution methods.
144
145	  A RAID-6 set of N drives with a capacity of C MB per drive
146	  provides the capacity of C * (N - 2) MB, and protects
147	  against a failure of any two drives. For a given sector
148	  (row) number, (N - 2) drives contain data sectors, and two
149	  drives contains two independent redundancy syndromes.  Like
150	  RAID-5, RAID-6 distributes the syndromes across the drives
151	  in one of the available parity distribution methods.
152
153	  Information about Software RAID on Linux is contained in the
154	  Software-RAID mini-HOWTO, available from
155	  <https://www.tldp.org/docs.html#howto>. There you will also
156	  learn where to get the supporting user space utilities raidtools.
157
158	  If you want to use such a RAID-4/RAID-5/RAID-6 set, say Y.  To
159	  compile this code as a module, choose M here: the module
160	  will be called raid456.
161
162	  If unsure, say Y.
163
164config MD_MULTIPATH
165	tristate "Multipath I/O support (deprecated)"
166	depends on BLK_DEV_MD
167	help
168	  MD_MULTIPATH provides a simple multi-path personality for use
169	  the MD framework.  It is not under active development.  New
170	  projects should consider using DM_MULTIPATH which has more
171	  features and more testing.
172
173	  If unsure, say N.
174
175config MD_FAULTY
176	tristate "Faulty test module for MD (deprecated)"
177	depends on BLK_DEV_MD
178	help
179	  The "faulty" module allows for a block device that occasionally returns
180	  read or write errors.  It is useful for testing.
181
182	  In unsure, say N.
183
184
185config MD_CLUSTER
186	tristate "Cluster Support for MD"
187	depends on BLK_DEV_MD
188	depends on DLM
189	default n
190	help
191	Clustering support for MD devices. This enables locking and
192	synchronization across multiple systems on the cluster, so all
193	nodes in the cluster can access the MD devices simultaneously.
194
195	This brings the redundancy (and uptime) of RAID levels across the
196	nodes of the cluster. Currently, it can work with raid1 and raid10
197	(limited support).
198
199	If unsure, say N.
200
201source "drivers/md/bcache/Kconfig"
202
203config BLK_DEV_DM_BUILTIN
204	bool
205
206config BLK_DEV_DM
207	tristate "Device mapper support"
208	select BLOCK_HOLDER_DEPRECATED if SYSFS
209	select BLK_DEV_DM_BUILTIN
210	select BLK_MQ_STACKING
211	depends on DAX || DAX=n
212	help
213	  Device-mapper is a low level volume manager.  It works by allowing
214	  people to specify mappings for ranges of logical sectors.  Various
215	  mapping types are available, in addition people may write their own
216	  modules containing custom mappings if they wish.
217
218	  Higher level volume managers such as LVM2 use this driver.
219
220	  To compile this as a module, choose M here: the module will be
221	  called dm-mod.
222
223	  If unsure, say N.
224
225config DM_DEBUG
226	bool "Device mapper debugging support"
227	depends on BLK_DEV_DM
228	help
229	  Enable this for messages that may help debug device-mapper problems.
230
231	  If unsure, say N.
232
233config DM_BUFIO
234       tristate
235       depends on BLK_DEV_DM
236	help
237	 This interface allows you to do buffered I/O on a device and acts
238	 as a cache, holding recently-read blocks in memory and performing
239	 delayed writes.
240
241config DM_DEBUG_BLOCK_MANAGER_LOCKING
242       bool "Block manager locking"
243       depends on DM_BUFIO
244	help
245	 Block manager locking can catch various metadata corruption issues.
246
247	 If unsure, say N.
248
249config DM_DEBUG_BLOCK_STACK_TRACING
250       bool "Keep stack trace of persistent data block lock holders"
251       depends on STACKTRACE_SUPPORT && DM_DEBUG_BLOCK_MANAGER_LOCKING
252       select STACKTRACE
253	help
254	 Enable this for messages that may help debug problems with the
255	 block manager locking used by thin provisioning and caching.
256
257	 If unsure, say N.
258
259config DM_BIO_PRISON
260       tristate
261       depends on BLK_DEV_DM
262	help
263	 Some bio locking schemes used by other device-mapper targets
264	 including thin provisioning.
265
266source "drivers/md/persistent-data/Kconfig"
267
268config DM_UNSTRIPED
269       tristate "Unstriped target"
270       depends on BLK_DEV_DM
271	help
272	  Unstripes I/O so it is issued solely on a single drive in a HW
273	  RAID0 or dm-striped target.
274
275config DM_CRYPT
276	tristate "Crypt target support"
277	depends on BLK_DEV_DM
278	depends on (ENCRYPTED_KEYS || ENCRYPTED_KEYS=n)
279	depends on (TRUSTED_KEYS || TRUSTED_KEYS=n)
280	select CRYPTO
281	select CRYPTO_CBC
282	select CRYPTO_ESSIV
283	help
284	  This device-mapper target allows you to create a device that
285	  transparently encrypts the data on it. You'll need to activate
286	  the ciphers you're going to use in the cryptoapi configuration.
287
288	  For further information on dm-crypt and userspace tools see:
289	  <https://gitlab.com/cryptsetup/cryptsetup/wikis/DMCrypt>
290
291	  To compile this code as a module, choose M here: the module will
292	  be called dm-crypt.
293
294	  If unsure, say N.
295
296config DM_SNAPSHOT
297       tristate "Snapshot target"
298       depends on BLK_DEV_DM
299       select DM_BUFIO
300	help
301	 Allow volume managers to take writable snapshots of a device.
302
303config DM_THIN_PROVISIONING
304       tristate "Thin provisioning target"
305       depends on BLK_DEV_DM
306       select DM_PERSISTENT_DATA
307       select DM_BIO_PRISON
308	help
309	 Provides thin provisioning and snapshots that share a data store.
310
311config DM_CACHE
312       tristate "Cache target (EXPERIMENTAL)"
313       depends on BLK_DEV_DM
314       default n
315       select DM_PERSISTENT_DATA
316       select DM_BIO_PRISON
317	help
318	 dm-cache attempts to improve performance of a block device by
319	 moving frequently used data to a smaller, higher performance
320	 device.  Different 'policy' plugins can be used to change the
321	 algorithms used to select which blocks are promoted, demoted,
322	 cleaned etc.  It supports writeback and writethrough modes.
323
324config DM_CACHE_SMQ
325       tristate "Stochastic MQ Cache Policy (EXPERIMENTAL)"
326       depends on DM_CACHE
327       default y
328	help
329	 A cache policy that uses a multiqueue ordered by recent hits
330	 to select which blocks should be promoted and demoted.
331	 This is meant to be a general purpose policy.  It prioritises
332	 reads over writes.  This SMQ policy (vs MQ) offers the promise
333	 of less memory utilization, improved performance and increased
334	 adaptability in the face of changing workloads.
335
336config DM_WRITECACHE
337	tristate "Writecache target"
338	depends on BLK_DEV_DM
339	help
340	   The writecache target caches writes on persistent memory or SSD.
341	   It is intended for databases or other programs that need extremely
342	   low commit latency.
343
344	   The writecache target doesn't cache reads because reads are supposed
345	   to be cached in standard RAM.
346
347config DM_EBS
348	tristate "Emulated block size target (EXPERIMENTAL)"
349	depends on BLK_DEV_DM && !HIGHMEM
350	select DM_BUFIO
351	help
352	  dm-ebs emulates smaller logical block size on backing devices
353	  with larger ones (e.g. 512 byte sectors on 4K native disks).
354
355config DM_ERA
356       tristate "Era target (EXPERIMENTAL)"
357       depends on BLK_DEV_DM
358       default n
359       select DM_PERSISTENT_DATA
360       select DM_BIO_PRISON
361	help
362	 dm-era tracks which parts of a block device are written to
363	 over time.  Useful for maintaining cache coherency when using
364	 vendor snapshots.
365
366config DM_CLONE
367       tristate "Clone target (EXPERIMENTAL)"
368       depends on BLK_DEV_DM
369       default n
370       select DM_PERSISTENT_DATA
371	help
372	 dm-clone produces a one-to-one copy of an existing, read-only source
373	 device into a writable destination device. The cloned device is
374	 visible/mountable immediately and the copy of the source device to the
375	 destination device happens in the background, in parallel with user
376	 I/O.
377
378	 If unsure, say N.
379
380config DM_MIRROR
381       tristate "Mirror target"
382       depends on BLK_DEV_DM
383	help
384	 Allow volume managers to mirror logical volumes, also
385	 needed for live data migration tools such as 'pvmove'.
386
387config DM_LOG_USERSPACE
388	tristate "Mirror userspace logging"
389	depends on DM_MIRROR && NET
390	select CONNECTOR
391	help
392	  The userspace logging module provides a mechanism for
393	  relaying the dm-dirty-log API to userspace.  Log designs
394	  which are more suited to userspace implementation (e.g.
395	  shared storage logs) or experimental logs can be implemented
396	  by leveraging this framework.
397
398config DM_RAID
399       tristate "RAID 1/4/5/6/10 target"
400       depends on BLK_DEV_DM
401       select MD_RAID0
402       select MD_RAID1
403       select MD_RAID10
404       select MD_RAID456
405       select BLK_DEV_MD
406	help
407	 A dm target that supports RAID1, RAID10, RAID4, RAID5 and RAID6 mappings
408
409	 A RAID-5 set of N drives with a capacity of C MB per drive provides
410	 the capacity of C * (N - 1) MB, and protects against a failure
411	 of a single drive. For a given sector (row) number, (N - 1) drives
412	 contain data sectors, and one drive contains the parity protection.
413	 For a RAID-4 set, the parity blocks are present on a single drive,
414	 while a RAID-5 set distributes the parity across the drives in one
415	 of the available parity distribution methods.
416
417	 A RAID-6 set of N drives with a capacity of C MB per drive
418	 provides the capacity of C * (N - 2) MB, and protects
419	 against a failure of any two drives. For a given sector
420	 (row) number, (N - 2) drives contain data sectors, and two
421	 drives contains two independent redundancy syndromes.  Like
422	 RAID-5, RAID-6 distributes the syndromes across the drives
423	 in one of the available parity distribution methods.
424
425config DM_ZERO
426	tristate "Zero target"
427	depends on BLK_DEV_DM
428	help
429	  A target that discards writes, and returns all zeroes for
430	  reads.  Useful in some recovery situations.
431
432config DM_MULTIPATH
433	tristate "Multipath target"
434	depends on BLK_DEV_DM
435	# nasty syntax but means make DM_MULTIPATH independent
436	# of SCSI_DH if the latter isn't defined but if
437	# it is, DM_MULTIPATH must depend on it.  We get a build
438	# error if SCSI_DH=m and DM_MULTIPATH=y
439	depends on !SCSI_DH || SCSI
440	help
441	  Allow volume managers to support multipath hardware.
442
443config DM_MULTIPATH_QL
444	tristate "I/O Path Selector based on the number of in-flight I/Os"
445	depends on DM_MULTIPATH
446	help
447	  This path selector is a dynamic load balancer which selects
448	  the path with the least number of in-flight I/Os.
449
450	  If unsure, say N.
451
452config DM_MULTIPATH_ST
453	tristate "I/O Path Selector based on the service time"
454	depends on DM_MULTIPATH
455	help
456	  This path selector is a dynamic load balancer which selects
457	  the path expected to complete the incoming I/O in the shortest
458	  time.
459
460	  If unsure, say N.
461
462config DM_MULTIPATH_HST
463	tristate "I/O Path Selector based on historical service time"
464	depends on DM_MULTIPATH
465	help
466	  This path selector is a dynamic load balancer which selects
467	  the path expected to complete the incoming I/O in the shortest
468	  time by comparing estimated service time (based on historical
469	  service time).
470
471	  If unsure, say N.
472
473config DM_MULTIPATH_IOA
474	tristate "I/O Path Selector based on CPU submission"
475	depends on DM_MULTIPATH
476	help
477	  This path selector selects the path based on the CPU the IO is
478	  executed on and the CPU to path mapping setup at path addition time.
479
480	  If unsure, say N.
481
482config DM_DELAY
483	tristate "I/O delaying target"
484	depends on BLK_DEV_DM
485	help
486	A target that delays reads and/or writes and can send
487	them to different devices.  Useful for testing.
488
489	If unsure, say N.
490
491config DM_DUST
492	tristate "Bad sector simulation target"
493	depends on BLK_DEV_DM
494	help
495	A target that simulates bad sector behavior.
496	Useful for testing.
497
498	If unsure, say N.
499
500config DM_INIT
501	bool "DM \"dm-mod.create=\" parameter support"
502	depends on BLK_DEV_DM=y
503	help
504	Enable "dm-mod.create=" parameter to create mapped devices at init time.
505	This option is useful to allow mounting rootfs without requiring an
506	initramfs.
507	See Documentation/admin-guide/device-mapper/dm-init.rst for dm-mod.create="..."
508	format.
509
510	If unsure, say N.
511
512config DM_UEVENT
513	bool "DM uevents"
514	depends on BLK_DEV_DM
515	help
516	Generate udev events for DM events.
517
518config DM_FLAKEY
519       tristate "Flakey target"
520       depends on BLK_DEV_DM
521	help
522	 A target that intermittently fails I/O for debugging purposes.
523
524config DM_VERITY
525	tristate "Verity target support"
526	depends on BLK_DEV_DM
527	select CRYPTO
528	select CRYPTO_HASH
529	select DM_BUFIO
530	help
531	  This device-mapper target creates a read-only device that
532	  transparently validates the data on one underlying device against
533	  a pre-generated tree of cryptographic checksums stored on a second
534	  device.
535
536	  You'll need to activate the digests you're going to use in the
537	  cryptoapi configuration.
538
539	  To compile this code as a module, choose M here: the module will
540	  be called dm-verity.
541
542	  If unsure, say N.
543
544config DM_VERITY_VERIFY_ROOTHASH_SIG
545	def_bool n
546	bool "Verity data device root hash signature verification support"
547	depends on DM_VERITY
548	select SYSTEM_DATA_VERIFICATION
549	help
550	  Add ability for dm-verity device to be validated if the
551	  pre-generated tree of cryptographic checksums passed has a pkcs#7
552	  signature file that can validate the roothash of the tree.
553
554	  By default, rely on the builtin trusted keyring.
555
556	  If unsure, say N.
557
558config DM_VERITY_VERIFY_ROOTHASH_SIG_SECONDARY_KEYRING
559	bool "Verity data device root hash signature verification with secondary keyring"
560	depends on DM_VERITY_VERIFY_ROOTHASH_SIG
561	depends on SECONDARY_TRUSTED_KEYRING
562	help
563	  Rely on the secondary trusted keyring to verify dm-verity signatures.
564
565	  If unsure, say N.
566
567config DM_VERITY_FEC
568	bool "Verity forward error correction support"
569	depends on DM_VERITY
570	select REED_SOLOMON
571	select REED_SOLOMON_DEC8
572	help
573	  Add forward error correction support to dm-verity. This option
574	  makes it possible to use pre-generated error correction data to
575	  recover from corrupted blocks.
576
577	  If unsure, say N.
578
579config DM_SWITCH
580	tristate "Switch target support (EXPERIMENTAL)"
581	depends on BLK_DEV_DM
582	help
583	  This device-mapper target creates a device that supports an arbitrary
584	  mapping of fixed-size regions of I/O across a fixed set of paths.
585	  The path used for any specific region can be switched dynamically
586	  by sending the target a message.
587
588	  To compile this code as a module, choose M here: the module will
589	  be called dm-switch.
590
591	  If unsure, say N.
592
593config DM_LOG_WRITES
594	tristate "Log writes target support"
595	depends on BLK_DEV_DM
596	help
597	  This device-mapper target takes two devices, one device to use
598	  normally, one to log all write operations done to the first device.
599	  This is for use by file system developers wishing to verify that
600	  their fs is writing a consistent file system at all times by allowing
601	  them to replay the log in a variety of ways and to check the
602	  contents.
603
604	  To compile this code as a module, choose M here: the module will
605	  be called dm-log-writes.
606
607	  If unsure, say N.
608
609config DM_INTEGRITY
610	tristate "Integrity target support"
611	depends on BLK_DEV_DM
612	select BLK_DEV_INTEGRITY
613	select DM_BUFIO
614	select CRYPTO
615	select CRYPTO_SKCIPHER
616	select ASYNC_XOR
617	select DM_AUDIT if AUDIT
618	help
619	  This device-mapper target emulates a block device that has
620	  additional per-sector tags that can be used for storing
621	  integrity information.
622
623	  This integrity target is used with the dm-crypt target to
624	  provide authenticated disk encryption or it can be used
625	  standalone.
626
627	  To compile this code as a module, choose M here: the module will
628	  be called dm-integrity.
629
630config DM_ZONED
631	tristate "Drive-managed zoned block device target support"
632	depends on BLK_DEV_DM
633	depends on BLK_DEV_ZONED
634	select CRC32
635	help
636	  This device-mapper target takes a host-managed or host-aware zoned
637	  block device and exposes most of its capacity as a regular block
638	  device (drive-managed zoned block device) without any write
639	  constraints. This is mainly intended for use with file systems that
640	  do not natively support zoned block devices but still want to
641	  benefit from the increased capacity offered by SMR disks. Other uses
642	  by applications using raw block devices (for example object stores)
643	  are also possible.
644
645	  To compile this code as a module, choose M here: the module will
646	  be called dm-zoned.
647
648	  If unsure, say N.
649
650config DM_AUDIT
651	bool "DM audit events"
652	depends on AUDIT
653	help
654	  Generate audit events for device-mapper.
655
656	  Enables audit logging of several security relevant events in the
657	  particular device-mapper targets, especially the integrity target.
658
659endif # MD
660