| /linux/Documentation/admin-guide/device-mapper/ |
| H A D | cache.rst | 2 Cache title
8 dm-cache is a device mapper target written by Joe Thornber, Heinz
11 It aims to improve performance of a block device (eg, a spindle) by
15 This device-mapper solution allows us to insert this caching at
17 a thin-provisioning pool. Caching solutions that are integrated more
20 The target reuses the metadata library used in the thin-provisioning
23 The decision as to what data to migrate and when is left to a plug-in
32 Movement of the primary copy of a logical block from one
39 The origin device always contains a copy of the logical block, which
40 may be out of date or kept in sync with the copy on the cache device
[all …]
|
| H A D | cache-policies.rst | 21 doesn't update states (eg, hit counts) for a block more than once
26 Overview of supplied cache replacement policies
30 ---------------
43 ---------------------------
47 The stochastic multi-queue (smq) policy addresses some of the problems
55 DM table that is using the cache target. Doing so will cause all of the
56 mq policy's hints to be dropped. Also, performance of the cache may
63 The mq policy used a lot of memory; 88 bytes per cache block on a 64
67 pointers. It avoids storing an explicit hit count for each block. It
68 has a 'hotspot' queue, rather than a pre-cache, which uses a quarter of
[all …]
|
| H A D | writecache.rst | 6 doesn't cache reads because reads are supposed to be cached in page cache
14 1. type of the cache device - "p" or "s"
15 - p - persistent memory
16 - s - SSD
18 3. the cache device
19 4. block size (4096 is recommended; the maximum block size is the page
25 offset from the start of cache device in 512-byte sectors
45 applicable only to persistent memory - use the FUA flag
49 applicable only to persistent memory - don't use the FUA
53 - some underlying devices perform better with fua, some
[all …]
|
| H A D | era.rst | 2 dm-era
8 dm-era is a target that behaves similar to the linear target. In
11 maintains the current era as a monotonically increasing 32-bit
15 partially invalidating the contents of a cache to restore cache
21 era <metadata dev> <origin dev> <block size>
26 block size block size of origin data device, granularity that is
36 ----------
43 ------------------
48 ------------------
55 <metadata block size> <#used metadata blocks>/<#total metadata blocks>
[all …]
|
| /linux/fs/squashfs/ |
| H A D | cache.c | 1 // SPDX-License-Identifier: GPL-2.0-or-later
3 * Squashfs - a compressed read only filesystem for Linux
8 * cache.c
15 * This file implements a generic cache implementation used for both caches,
16 * plus functions layered ontop of the generic cache implementation to
19 * To avoid out of memory and fragmentation issues with vmalloc the cache
22 * It should be noted that the cache is not used for file datablocks, these
23 * are decompressed and cached in the page-cache in the normal way. The
24 * cache is only used to temporarily cache fragment and metadata blocks
29 * have been packed with it, these because of locality-of-reference may be read
[all …]
|
| H A D | file.c | 1 // SPDX-License-Identifier: GPL-2.0-or-later
3 * Squashfs - a compressed read only filesystem for Linux
14 * compressed fragment block (tail-end packed block). The compressed size
15 * of each datablock is stored in a block list contained within the
19 * larger), the code implements an index cache that caches the mapping from
20 * block index to datablock location on disk.
22 * The index cache allows Squashfs to handle large files (up to 1.75 TiB) while
23 * retaining a simple and space-efficient block list on disk. The cache
26 * The index cache is designed to be memory efficient, and by default uses
46 * Locate cache slot in range [offset, index] for specified inode. If
[all …]
|
| H A D | Kconfig | 1 # SPDX-License-Identifier: GPL-2.0-only
3 tristate "SquashFS 4.0 - Squashed file system support"
4 depends on BLOCK
7 Read-Only File System). Squashfs is a highly compressed read-only
11 Block sizes greater than 4K are supported up to a maximum of 1 Mbytes
12 (default block size 128K). SquashFS 4.0 supports 64 bit filesystems
16 Squashfs is intended for general read-only filesystem use, for
19 and tools are available from github.com/plougher/squashfs-tools.
35 intermediate buffer and then memcopied it into the page cache.
37 the page cache.
[all …]
|
| /linux/fs/btrfs/ |
| H A D | block-group.c | 1 // SPDX-License-Identifier: GPL-2.0
7 #include "block-group.h"
8 #include "space-info.h"
9 #include "disk-io.h"
10 #include "free-space-cache.h"
11 #include "free-space-tree.h"
14 #include "ref-verify.h"
16 #include "tree-log.h"
17 #include "delalloc-space.h"
23 #include "extent-tree.h"
[all …]
|
| H A D | zoned.c | 1 // SPDX-License-Identifier: GPL-2.0
12 #include "disk-io.h"
13 #include "block-group.h"
14 #include "dev-replace.h"
15 #include "space-info.h"
25 #define WP_MISSING_DEV ((u64)-1)
27 #define WP_CONVENTIONAL ((u64)-2)
32 * - primary superblock: 0B (zone 0)
33 * - first copy: 512G (zone starting at that offset)
34 * - second copy: 4T (zone starting at that offset)
[all …]
|
| /linux/arch/riscv/boot/dts/sifive/ |
| H A D | fu540-c000.dtsi | 1 // SPDX-License-Identifier: (GPL-2.0 OR MIT)
2 /* Copyright (c) 2018-2019 SiFive, Inc */
4 /dts-v1/;
6 #include <dt-bindings/clock/sifive-fu540-prci.h>
9 #address-cells = <2>;
10 #size-cells = <2>;
11 compatible = "sifive,fu540-c000", "sifive,fu540";
23 #address-cells = <1>;
24 #size-cells = <0>;
28 i-cache-block-size = <64>;
[all …]
|
| H A D | fu740-c000.dtsi | 1 // SPDX-License-Identifier: (GPL-2.0 OR MIT)
4 /dts-v1/;
6 #include <dt-bindings/clock/sifive-fu740-prci.h>
9 #address-cells = <2>;
10 #size-cells = <2>;
11 compatible = "sifive,fu740-c000", "sifive,fu740";
23 #address-cells = <1>;
24 #size-cells = <0>;
28 i-cache-block-size = <64>;
29 i-cache-sets = <128>;
[all …]
|
| /linux/drivers/md/ |
| H A D | dm-cache-policy.h | 1 /* SPDX-License-Identifier: GPL-2.0-only */
11 #include "dm-cache-block-types.h"
13 #include <linux/device-mapper.h>
15 /*----------------------------------------------------------------*/
18 * The cache policy makes the important decisions about which blocks get to
19 * live on the faster cache device.
37 * The cache policy object. It is envisaged that this structure will be
47 * Find the location of a block.
49 * Must not block.
51 * Returns 0 if in cache (cblock will be set), -ENOENT if not, < 0 for
[all …]
|
| /linux/Documentation/block/ |
| H A D | writeback_cache_control.rst | 2 Explicit volatile write back cache control
6 ------------
10 operating system before data actually has hit the non-volatile storage. This
12 system needs to force data out to the non-volatile storage when it performs
15 The Linux block layer provides two simple mechanisms that let filesystems
17 a forced cache flush, and the Force Unit Access (FUA) flag for requests.
20 Explicit cache flushes
21 ----------------------
24 the filesystem and will make sure the volatile cache of the storage device
26 guarantees that previously completed write requests are on non-volatile
[all …]
|
| /linux/fs/ocfs2/ |
| H A D | uptodate.c | 1 // SPDX-License-Identifier: GPL-2.0-or-later
5 * Tracking the up-to-date-ness of a local buffer_head with respect to
11 * in a clustered environment - a buffer may be marked up to date on
15 * memory as possible - we never pin buffer_head structures in order
16 * to cache them.
23 * - buffer_jbd is used liberally - if a bh is in the journal on
25 * - the standard buffer_uptodate() macro is used to detect buffers
58 BUG_ON(!ci || !ci->ci_ops); in ocfs2_metadata_cache_owner()
60 return ci->ci_ops->co_owner(ci); in ocfs2_metadata_cache_owner()
65 BUG_ON(!ci || !ci->ci_ops); in ocfs2_metadata_cache_get_super()
[all …]
|
| /linux/include/linux/ |
| H A D | dm-bufio.h | 1 /* SPDX-License-Identifier: GPL-2.0-only */
3 * Copyright (C) 2009-2011 Red Hat, Inc.
16 /*----------------------------------------------------------------*/
27 * Create a buffered IO cache on a given device
37 * Release a buffered IO cache.
53 * - At most one thread can hold at most "reserved_buffers" simultaneously.
54 * - Each other threads can hold at most one buffer.
55 * - Threads which call only dm_bufio_get can hold unlimited number of
60 * Read a given block from disk. Returns pointer to data. Returns a
64 void *dm_bufio_read(struct dm_bufio_client *c, sector_t block,
[all …]
|
| /linux/drivers/mtd/ |
| H A D | mtdblock.c | 1 // SPDX-License-Identifier: GPL-2.0-or-later
3 * Direct MTD block device access
5 * Copyright © 1999-2010 David Woodhouse <dwmw2@infradead.org>
6 * Copyright © 2000-2003 Nicolas Pitre <nico@fluxnic.net>
35 * Cache stuff...
38 * buffer cache can handle, we must implement read-modify-write on flash
39 * sectors for each block write requests. To avoid over-erasing flash sectors
40 * and to speed things up, we locally cache a whole flash sector while it is
52 * First, let's erase the flash block. in erase_write()
61 pos, len, mtd->name); in erase_write()
[all …]
|
| /linux/drivers/base/regmap/ |
| H A D | regcache-rbtree.c | 1 // SPDX-License-Identifier: GPL-2.0
3 // Register cache access API - rbtree caching support
22 /* block of adjacent registers */
23 void *block; member
26 /* base register handled by this block */
28 /* number of registers available in the block */
30 /* the actual rbtree node holding this block */
44 *base = rbnode->base_reg; in regcache_rbtree_get_base_top_reg()
45 *top = rbnode->base_reg + ((rbnode->blklen - 1) * map->reg_stride); in regcache_rbtree_get_base_top_reg()
51 return regcache_get_val(map, rbnode->block, idx); in regcache_rbtree_get_register()
[all …]
|
| /linux/Documentation/devicetree/bindings/cache/ |
| H A D | sifive,ccache0.yaml | 1 # SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
4 ---
5 $id: http://devicetree.org/schemas/cache/sifive,ccache0.yaml#
6 $schema: http://devicetree.org/meta-schemas/core.yaml#
8 title: SiFive Composable Cache Controller
11 - Paul Walmsley <paul.walmsley@sifive.com>
14 The SiFive Composable Cache Controller is used to provide access to fast copies
15 of memory for masters in a Core Complex. The Composable Cache Controller also
16 acts as directory-based coherency manager.
24 - sifive,ccache0
[all …]
|
| H A D | starfive,jh8100-starlink-cache.yaml | 1 # SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
3 ---
4 $id: http://devicetree.org/schemas/cache/starfive,jh8100-starlink-cache.yaml#
5 $schema: http://devicetree.org/meta-schemas/core.yaml#
7 title: StarFive StarLink Cache Controller
10 - Joshua Yeong <joshua.yeong@starfivetech.com>
13 StarFive's StarLink Cache Controller manages the L3 cache shared between
14 clusters of CPU cores. The cache driver enables RISC-V non-standard cache
15 management as an alternative to instructions in the RISC-V Zicbom extension.
18 - $ref: /schemas/cache-controller.yaml#
[all …]
|
| /linux/arch/riscv/boot/dts/microchip/ |
| H A D | mpfs.dtsi | 1 // SPDX-License-Identifier: (GPL-2.0 OR MIT)
2 /* Copyright (c) 2020-2021 Microchip Technology Inc */
4 /dts-v1/;
5 #include "dt-bindings/clock/microchip,mpfs-clock.h"
8 #address-cells = <2>;
9 #size-cells = <2>;
14 #address-cells = <1>;
15 #size-cells = <0>;
16 timebase-frequency = <1000000>;
21 i-cache-block-size = <64>;
[all …]
|
| /linux/arch/powerpc/kernel/ |
| H A D | cacheinfo.c | 1 // SPDX-License-Identifier: GPL-2.0-only
3 * Processor cache information made available to userspace via sysfs;
26 /* per-cpu object for tracking:
27 * - a "cache" kobject for the top-level directory
28 * - a list of "index" objects representing the cpu's local cache hierarchy
31 struct kobject *kobj; /* bare (not embedded) kobject for cache
36 /* "index" object: each cpu's cache directory has an index
37 * subdirectory corresponding to a cache object associated with the
43 struct cache *cache; member
47 * cache type */
[all …]
|
| /linux/arch/sparc/include/uapi/asm/ |
| H A D | asi.h | 1 /* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
32 /* VAC Cache flushing on sun4c and sun4 */
37 /* SPARCstation-5: only 6 bits are decoded. */
52 #define ASI_M_TXTC_TAG 0x0C /* Instruction Cache Tag; rw, ss */
53 #define ASI_M_TXTC_DATA 0x0D /* Instruction Cache Data; rw, ss */
54 #define ASI_M_DATAC_TAG 0x0E /* Data Cache Tag; rw, ss */
55 #define ASI_M_DATAC_DATA 0x0F /* Data Cache Data; rw, ss */
57 /* The following cache flushing ASIs work only with the 'sta'
63 #define ASI_M_FLUSH_PAGE 0x10 /* Flush I&D Cache Line (page); wo, ss */
64 #define ASI_M_FLUSH_SEG 0x11 /* Flush I&D Cache Line (seg); wo, ss */
[all …]
|
| /linux/tools/lib/python/kdoc/ |
| H A D | kdoc_output.py | 2 # SPDX-License-Identifier: GPL-2.0
8 Classes to implement output filters to print kernel-doc documentation.
28 function_pointer = KernRe(r"([^\(]*\(\*)\s*\)\s*\(([^\)]*)\)", cache=False)
31 type_constant = KernRe(r"\b``([^\`]+)``\b", cache=False)
32 type_constant2 = KernRe(r"\%([-_*\w]+)", cache=False)
33 type_func = KernRe(r"(\w+)\(\)", cache=False)
34 type_param_ref = KernRe(r"([\!~\*]?)\@(\w*((\.\w+)|(->\
112 highlight_block(self, block) global() argument
663 output_highlight(self, block) global() argument
[all...] |
| /linux/Documentation/driver-api/md/ |
| H A D | raid5-cache.rst | 2 RAID 4/5/6 cache
5 Raid 4/5/6 could include an extra disk for data cache besides normal RAID
6 disks. The role of RAID disks isn't changed with the cache disk. The cache disk
7 caches data to the RAID disks. The cache can be in write-through (supported
8 since 4.4) or write-back mode (supported since 4.10). mdadm (supported since
9 3.4) has a new option '--write-journal' to create array with cache. Please
10 refer to mdadm manual for details. By default (RAID array starts), the cache is
11 in write-through mode. A user can switch it to write-back mode by::
13 echo "write-back" > /sys/block/md0/md/journal_mode
15 And switch it back to write-through mode by::
[all …]
|
| /linux/arch/sparc/lib/ |
| H A D | M7memset.S | 15 * Fast assembler language version of the following C-program for memset
16 * which represents the `standard' for the C-library.
25 * } while (--n != 0);
35 * Then store as many 4-byte chunks, followed by trailing bytes.
39 * store 8-bytes chunks to align the address on 64 byte boundary
42 * 64-byte cache line to zero which will also clear the
43 * other seven long words of the cache line.
47 * ST_CHUNK cache lines (64 bytes each) before the main
49 * In the main loop, continue pre-setting the first long
50 * word of each cache line ST_CHUNK lines in advance while
[all …]
|