/*
* CDDL HEADER START
*
* The contents of this file are subject to the terms of the
* Common Development and Distribution License (the "License").
* You may not use this file except in compliance with the License.
*
* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
* See the License for the specific language governing permissions
* and limitations under the License.
*
* When distributing Covered Code, include this CDDL HEADER in each
* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
* If applicable, add the following below this CDDL HEADER, with the
* fields enclosed by brackets "[]" replaced with your own identifying
* information: Portions Copyright [yyyy] [name of copyright owner]
*
* CDDL HEADER END
*/
/*
* Copyright (c) 2012, Joyent, Inc. All rights reserved.
* Copyright (c) 2011, 2016 by Delphix. All rights reserved.
* Copyright (c) 2014 by Saso Kiselkov. All rights reserved.
* Copyright 2015 Nexenta Systems, Inc. All rights reserved.
*/
/*
* DVA-based Adjustable Replacement Cache
*
* While much of the theory of operation used here is
* based on the self-tuning, low overhead replacement cache
* presented by Megiddo and Modha at FAST 2003, there are some
* significant differences:
*
* 1. The Megiddo and Modha model assumes any page is evictable.
* Pages in its cache cannot be "locked" into memory. This makes
* the eviction algorithm simple: evict the last page in the list.
* This also make the performance characteristics easy to reason
* about. Our cache is not so simple. At any given moment, some
* subset of the blocks in the cache are un-evictable because we
* have handed out a reference to them. Blocks are only evictable
* when there are no external references active. This makes
* eviction far more problematic: we choose to evict the evictable
* blocks that are the "lowest" in the list.
*
* There are times when it is not possible to evict the requested
* space. In these circumstances we are unable to adjust the cache
* size. To prevent the cache growing unbounded at these times we
* implement a "cache throttle" that slows the flow of new data
* into the cache until we can make space available.
*
* 2. The Megiddo and Modha model assumes a fixed cache size.
* Pages are evicted when the cache is full and there is a cache
* miss. Our model has a variable sized cache. It grows with
* high use, but also tries to react to memory pressure from the
* operating system: decreasing its size when system memory is
* tight.
*
* 3. The Megiddo and Modha model assumes a fixed page size. All
* elements of the cache are therefore exactly the same size. So
* when adjusting the cache size following a cache miss, its simply
* a matter of choosing a single page to evict. In our model, we
* have variable sized cache blocks (rangeing from 512 bytes to
* 128K bytes). We therefore choose a set of blocks to evict to make
* space for a cache miss that approximates as closely as possible
* the space used by the new block.
*
* See also: "ARC: A Self-Tuning, Low Overhead Replacement Cache"
* by N. Megiddo & D. Modha, FAST 2003
*/
/*
* The locking model:
*
* A new reference to a cache buffer can be obtained in two
* ways: 1) via a hash table lookup using the DVA as a key,
* or 2) via one of the ARC lists. The arc_read() interface
* uses method 1, while the internal ARC algorithms for
* adjusting the cache use method 2. We therefore provide two
* types of locks: 1) the hash table lock array, and 2) the
* ARC list locks.
*
* Buffers do not have their own mutexes, rather they rely on the
* hash table mutexes for the bulk of their protection (i.e. most
* fields in the arc_buf_hdr_t are protected by these mutexes).
*
* buf_hash_find() returns the appropriate mutex (held) when it
* locates the requested buffer in the hash table. It returns
* NULL for the mutex if the buffer was not in the table.
*
* buf_hash_remove() expects the appropriate hash mutex to be
* already held before it is invoked.
*
* Each ARC state also has a mutex which is used to protect the
* buffer list associated with the state. When attempting to
* obtain a hash table lock while holding an ARC list lock you
* must use: mutex_tryenter() to avoid deadlock. Also note that
* the active state mutex must be held before the ghost state mutex.
*
* Note that the majority of the performance stats are manipulated
* with atomic operations.
*
* The L2ARC uses the l2ad_mtx on each vdev for the following:
*
* - L2ARC buflist creation
* - L2ARC buflist eviction
* - L2ARC write completion, which walks L2ARC buflists
* - ARC header destruction, as it removes from L2ARC buflists
* - ARC header release, as it removes from L2ARC buflists
*/
/*
* ARC operation:
*
* Every block that is in the ARC is tracked by an arc_buf_hdr_t structure.
* This structure can point either to a block that is still in the cache or to
* one that is only accessible in an L2 ARC device, or it can provide
* information about a block that was recently evicted. If a block is
* only accessible in the L2ARC, then the arc_buf_hdr_t only has enough
* information to retrieve it from the L2ARC device. This information is
* stored in the l2arc_buf_hdr_t sub-structure of the arc_buf_hdr_t. A block
* that is in this state cannot access the data directly.
*
* Blocks that are actively being referenced or have not been evicted
* are cached in the L1ARC. The L1ARC (l1arc_buf_hdr_t) is a structure within
* the arc_buf_hdr_t that will point to the data block in memory. A block can
* only be read by a consumer if it has an l1arc_buf_hdr_t. The L1ARC
* caches data in two ways -- in a list of ARC buffers (arc_buf_t) and
* also in the arc_buf_hdr_t's private physical data block pointer (b_pdata).
*
* The L1ARC's data pointer may or may not be uncompressed. The ARC has the
* ability to store the physical data (b_pdata) associated with the DVA of the
* arc_buf_hdr_t. Since the b_pdata is a copy of the on-disk physical block,
* it will match its on-disk compression characteristics. This behavior can be
* disabled by setting 'zfs_compressed_arc_enabled' to B_FALSE. When the
* compressed ARC functionality is disabled, the b_pdata will point to an
* uncompressed version of the on-disk data.
*
* Data in the L1ARC is not accessed by consumers of the ARC directly. Each
* arc_buf_hdr_t can have multiple ARC buffers (arc_buf_t) which reference it.
* Each ARC buffer (arc_buf_t) is being actively accessed by a specific ARC
* consumer. The ARC will provide references to this data and will keep it
* cached until it is no longer in use. The ARC caches only the L1ARC's physical
* data block and will evict any arc_buf_t that is no longer referenced. The
* amount of memory consumed by the arc_buf_ts' data buffers can be seen via the
* "overhead_size" kstat.
*
* Depending on the consumer, an arc_buf_t can be requested in uncompressed or
* compressed form. The typical case is that consumers will want uncompressed
* data, and when that happens a new data buffer is allocated where the data is
* decompressed for them to use. Currently the only consumer who wants
* compressed arc_buf_t's is "zfs send", when it streams data exactly as it
* exists on disk. When this happens, the arc_buf_t's data buffer is shared
* with the arc_buf_hdr_t.
*
* Here is a diagram showing an arc_buf_hdr_t referenced by two arc_buf_t's. The
* first one is owned by a compressed send consumer (and therefore references
* the same compressed data buffer as the arc_buf_hdr_t) and the second could be
* used by any other consumer (and has its own uncompressed copy of the data
* buffer).
*
* arc_buf_hdr_t
* +-----------+
* | fields |
* | common to |
* | L1- and |
* | L2ARC |
* +-----------+
* | l2arc_buf_hdr_t
* | |
* +-----------+
* | l1arc_buf_hdr_t
* | | arc_buf_t
* | b_buf +------------>+-----------+ arc_buf_t
* | b_pdata +-+ |b_next +---->+-----------+
* +-----------+ | |-----------| |b_next +-->NULL
* | |b_comp = T | +-----------+
* | |b_data +-+ |b_comp = F |
* | +-----------+ | |b_data +-+
* +->+------+ | +-----------+ |
* compressed | | | |
* data | |<--------------+ | uncompressed
* +------+ compressed, | data
* shared +-->+------+
* data | |
* | |
* +------+
*
* When a consumer reads a block, the ARC must first look to see if the
* arc_buf_hdr_t is cached. If the hdr is cached then the ARC allocates a new
* arc_buf_t and either copies uncompressed data into a new data buffer from an
* existing uncompressed arc_buf_t, decompresses the hdr's b_pdata buffer into a
* new data buffer, or shares the hdr's b_pdata buffer, depending on whether the
* hdr is compressed and the desired compression characteristics of the
* arc_buf_t consumer. If the arc_buf_t ends up sharing data with the
* arc_buf_hdr_t and both of them are uncompressed then the arc_buf_t must be
* the last buffer in the hdr's b_buf list, however a shared compressed buf can
* be anywhere in the hdr's list.
*
* The diagram below shows an example of an uncompressed ARC hdr that is
* sharing its data with an arc_buf_t (note that the shared uncompressed buf is
* the last element in the buf list):
*
* arc_buf_hdr_t
* +-----------+
* | |
* | |
* | |
* +-----------+
* l2arc_buf_hdr_t| |
* | |
* +-----------+
* l1arc_buf_hdr_t| |
* | | arc_buf_t (shared)
* | b_buf +------------>+---------+ arc_buf_t
* | | |b_next +---->+---------+
* | b_pdata +-+ |---------| |b_next +-->NULL
* +-----------+ | | | +---------+
* | |b_data +-+ | |
* | +---------+ | |b_data +-+
* +->+------+ | +---------+ |
* | | | |
* uncompressed | | | |
* data +------+ | |
* ^ +->+------+ |
* | uncompressed | | |
* | data | | |
* | +------+ |
* +---------------------------------+
*
* Writing to the ARC requires that the ARC first discard the hdr's b_pdata
* since the physical block is about to be rewritten. The new data contents
* will be contained in the arc_buf_t. As the I/O pipeline performs the write,
* it may compress the data before writing it to disk. The ARC will be called
* with the transformed data and will bcopy the transformed on-disk block into
* a newly allocated b_pdata. Writes are always done into buffers which have
* either been loaned (and hence are new and don't have other readers) or
* buffers which have been released (and hence have their own hdr, if there
* were originally other readers of the buf's original hdr). This ensures that
* the ARC only needs to update a single buf and its hdr after a write occurs.
*
* When the L2ARC is in use, it will also take advantage of the b_pdata. The
* L2ARC will always write the contents of b_pdata to the L2ARC. This means
* that when compressed ARC is enabled that the L2ARC blocks are identical
* to the on-disk block in the main data pool. This provides a significant
* advantage since the ARC can leverage the bp's checksum when reading from the
* L2ARC to determine if the contents are valid. However, if the compressed
* ARC is disabled, then the L2ARC's block must be transformed to look
* like the physical block in the main data pool before comparing the
* checksum and determining its validity.
*/
#include <sys/spa_impl.h>
#include <sys/zio_compress.h>
#include <sys/zio_checksum.h>
#include <sys/zfs_context.h>
#include <sys/refcount.h>
#include <sys/vdev_impl.h>
#include <sys/dsl_pool.h>
#include <sys/multilist.h>
#ifdef _KERNEL
#endif
#include <zfs_fletcher.h>
#ifndef _KERNEL
/* set with ZFS_DEBUG=watch, to enable watchpoints on frozen buffers */
int arc_procfd;
#endif
/*
* The number of headers to evict in arc_evict_state_impl() before
* dropping the sublist lock and evicting from another sublist. A lower
* value means we're more likely to evict the "correct" header (i.e. the
* oldest header in the arc state), but comes with higher overhead
* (i.e. more invocations of arc_evict_state_impl()).
*/
/*
* The number of sublists used for each of the arc state lists. If this
* is not set to a suitable value by the user, it will be configured to
* the number of CPUs on the system in arc_init().
*/
int zfs_arc_num_sublists_per_state = 0;
/* number of seconds before growing cache again */
/* shift of arc_c for calculating overflow limit in arc_get_data_buf */
/* shift of arc_c for calculating both min and max arc_p */
/* log2(fraction of arc to reclaim) */
/*
* log2(fraction of ARC which must be free to allow growing).
* I.e. If there is less than arc_c >> arc_no_grow_shift free memory,
* when reading a new block into the ARC, we will evict an equal-sized block
* from the ARC.
*
* This must be less than arc_shrink_shift, so that when we shrink the ARC,
* we will still not allow it to grow.
*/
/*
* minimum lifespan of a prefetch block in clock ticks
* (initialized in arc_init())
*/
static int arc_min_prefetch_lifespan;
/*
* If this percent of memory is free, don't throttle.
*/
static int arc_dead;
/*
* The arc has filled available memory and has now warmed up.
*/
/*
* log2 fraction of the zio arena to keep free.
*/
/*
* These tunables are for performance analysis.
*/
int zfs_arc_grow_retry = 0;
int zfs_arc_shrink_shift = 0;
int zfs_arc_p_min_shift = 0;
/*
* Note that buffers can be in one of 6 states:
* ARC_anon - anonymous (discussed below)
* ARC_mru - recently used, currently cached
* ARC_mru_ghost - recentely used, no longer in cache
* ARC_mfu - frequently used, currently cached
* ARC_mfu_ghost - frequently used, no longer in cache
* ARC_l2c_only - exists in L2ARC but not other states
* When there are no active references to the buffer, they are
* are linked onto a list in one of these arc states. These are
* the only buffers that can be evicted or deleted. Within each
* state there are multiple lists, one for meta-data and one for
* non-meta-data. Meta-data (indirect blocks, blocks of dnodes,
* etc.) is tracked separately so that it can be managed more
* explicitly: favored over data, limited explicitly.
*
* Anonymous buffers are buffers that are not associated with
* a DVA. These are buffers that hold dirty block copies
* before they are written to stable storage. By definition,
* they are "ref'd" and are considered part of arc_mru
* that cannot be freed. Generally, they will aquire a DVA
* as they are written and migrate onto the arc_mru list.
*
* The ARC_l2c_only state is for buffers that are in the second
* level ARC but no longer in any of the ARC_m* lists. The second
* level ARC itself may also contain buffers that are in any of
* the ARC_m* states - meaning that a buffer can exist in two
* places. The reason for the ARC_l2c_only state is to keep the
* buffer header in the hash table, so that reads that hit the
* second level ARC benefit from these fast lookups.
*/
typedef struct arc_state {
/*
* list of evictable buffers
*/
/*
* total amount of evictable data in this state
*/
/*
* total amount of data in this state; this includes: evictable,
* non-evictable, ARC_BUFC_DATA, and ARC_BUFC_METADATA.
*/
} arc_state_t;
/* The 6 states: */
typedef struct arc_stats {
/*
* Number of buffers that could not be evicted because the hash lock
* was held by another thread. The lock may not necessarily be held
* by something using the same buffer, since hash locks are shared
* by multiple buffers.
*/
/*
* Number of buffers skipped because they have I/O in progress, are
* indrect prefetch buffers that have not lived long enough, or are
* not from the spa we're trying to evict from.
*/
/*
* Number of times arc_evict_state() was unable to evict enough
* buffers to reach it's target amount.
*/
/*
* Number of compressed bytes stored in the arc_buf_hdr_t's b_pdata.
* Note that the compressed bytes may match the uncompressed bytes
* if the block is either not compressed or compressed arc is disabled.
*/
/*
* Uncompressed size of the data stored in b_pdata. If compressed
* arc is disabled then this value will be identical to the stat
* above.
*/
/*
* Number of bytes stored in all the arc_buf_t's. This is classified
* as "overhead" since this data is typically short-lived and will
* be evicted from the arc when it becomes unreferenced unless the
* zfs_keep_uncompressed_metadata or zfs_keep_uncompressed_level
* values have been set (see comment in dbuf.c for more information).
*/
/*
* Number of bytes consumed by internal ARC structures necessary
* for tracking purposes; these structures are not actually
* backed by ARC buffers. This includes arc_buf_hdr_t structures
* (allocated via arc_buf_hdr_t_full and arc_buf_hdr_t_l2only
* caches), and arc_buf_t structures (allocated via arc_buf_t
* cache).
*/
/*
* Number of bytes consumed by ARC buffers of type equal to
* ARC_BUFC_DATA. This is generally consumed by buffers backing
* on disk user data (e.g. plain file contents).
*/
/*
* Number of bytes consumed by ARC buffers of type equal to
* ARC_BUFC_METADATA. This is generally consumed by buffers
* backing on disk data that is used for internal ZFS
* structures (e.g. ZAP, dnode, indirect blocks, etc).
*/
/*
* Number of bytes consumed by various buffers and structures
* not actually backed with ARC buffers. This includes bonus
* buffers (allocated directly via zio_buf_* functions),
* dmu_buf_impl_t structures (allocated via dmu_buf_impl_t
* cache), and dnode_t structures (allocated via dnode_t cache).
*/
/*
* Total number of bytes consumed by ARC buffers residing in the
* arc_anon state. This includes *all* buffers in the arc_anon
* state; e.g. data, metadata, evictable, and unevictable buffers
* are all included in this value.
*/
/*
* Number of bytes consumed by ARC buffers that meet the
* following criteria: backing buffers of type ARC_BUFC_DATA,
* residing in the arc_anon state, and are eligible for eviction
* (e.g. have no outstanding holds on the buffer).
*/
/*
* Number of bytes consumed by ARC buffers that meet the
* following criteria: backing buffers of type ARC_BUFC_METADATA,
* residing in the arc_anon state, and are eligible for eviction
* (e.g. have no outstanding holds on the buffer).
*/
/*
* Total number of bytes consumed by ARC buffers residing in the
* arc_mru state. This includes *all* buffers in the arc_mru
* state; e.g. data, metadata, evictable, and unevictable buffers
* are all included in this value.
*/
/*
* Number of bytes consumed by ARC buffers that meet the
* following criteria: backing buffers of type ARC_BUFC_DATA,
* residing in the arc_mru state, and are eligible for eviction
* (e.g. have no outstanding holds on the buffer).
*/
/*
* Number of bytes consumed by ARC buffers that meet the
* following criteria: backing buffers of type ARC_BUFC_METADATA,
* residing in the arc_mru state, and are eligible for eviction
* (e.g. have no outstanding holds on the buffer).
*/
/*
* Total number of bytes that *would have been* consumed by ARC
* buffers in the arc_mru_ghost state. The key thing to note
* here, is the fact that this size doesn't actually indicate
* RAM consumption. The ghost lists only consist of headers and
* don't actually have ARC buffers linked off of these headers.
* Thus, *if* the headers had associated ARC buffers, these
* buffers *would have* consumed this number of bytes.
*/
/*
* Number of bytes that *would have been* consumed by ARC
* buffers that are eligible for eviction, of type
* ARC_BUFC_DATA, and linked off the arc_mru_ghost state.
*/
/*
* Number of bytes that *would have been* consumed by ARC
* buffers that are eligible for eviction, of type
* ARC_BUFC_METADATA, and linked off the arc_mru_ghost state.
*/
/*
* Total number of bytes consumed by ARC buffers residing in the
* arc_mfu state. This includes *all* buffers in the arc_mfu
* state; e.g. data, metadata, evictable, and unevictable buffers
* are all included in this value.
*/
/*
* Number of bytes consumed by ARC buffers that are eligible for
* eviction, of type ARC_BUFC_DATA, and reside in the arc_mfu
* state.
*/
/*
* Number of bytes consumed by ARC buffers that are eligible for
* eviction, of type ARC_BUFC_METADATA, and reside in the
* arc_mfu state.
*/
/*
* Total number of bytes that *would have been* consumed by ARC
* buffers in the arc_mfu_ghost state. See the comment above
* arcstat_mru_ghost_size for more details.
*/
/*
* Number of bytes that *would have been* consumed by ARC
* buffers that are eligible for eviction, of type
* ARC_BUFC_DATA, and linked off the arc_mfu_ghost state.
*/
/*
* Number of bytes that *would have been* consumed by ARC
* buffers that are eligible for eviction, of type
* ARC_BUFC_METADATA, and linked off the arc_mru_ghost state.
*/
} arc_stats_t;
{ "hits", KSTAT_DATA_UINT64 },
{ "misses", KSTAT_DATA_UINT64 },
{ "demand_data_hits", KSTAT_DATA_UINT64 },
{ "demand_data_misses", KSTAT_DATA_UINT64 },
{ "demand_metadata_hits", KSTAT_DATA_UINT64 },
{ "demand_metadata_misses", KSTAT_DATA_UINT64 },
{ "prefetch_data_hits", KSTAT_DATA_UINT64 },
{ "prefetch_data_misses", KSTAT_DATA_UINT64 },
{ "prefetch_metadata_hits", KSTAT_DATA_UINT64 },
{ "prefetch_metadata_misses", KSTAT_DATA_UINT64 },
{ "mru_hits", KSTAT_DATA_UINT64 },
{ "mru_ghost_hits", KSTAT_DATA_UINT64 },
{ "mfu_hits", KSTAT_DATA_UINT64 },
{ "mfu_ghost_hits", KSTAT_DATA_UINT64 },
{ "deleted", KSTAT_DATA_UINT64 },
{ "mutex_miss", KSTAT_DATA_UINT64 },
{ "evict_skip", KSTAT_DATA_UINT64 },
{ "evict_not_enough", KSTAT_DATA_UINT64 },
{ "evict_l2_cached", KSTAT_DATA_UINT64 },
{ "evict_l2_eligible", KSTAT_DATA_UINT64 },
{ "evict_l2_ineligible", KSTAT_DATA_UINT64 },
{ "evict_l2_skip", KSTAT_DATA_UINT64 },
{ "hash_elements", KSTAT_DATA_UINT64 },
{ "hash_elements_max", KSTAT_DATA_UINT64 },
{ "hash_collisions", KSTAT_DATA_UINT64 },
{ "hash_chains", KSTAT_DATA_UINT64 },
{ "hash_chain_max", KSTAT_DATA_UINT64 },
{ "p", KSTAT_DATA_UINT64 },
{ "c", KSTAT_DATA_UINT64 },
{ "c_min", KSTAT_DATA_UINT64 },
{ "c_max", KSTAT_DATA_UINT64 },
{ "size", KSTAT_DATA_UINT64 },
{ "compressed_size", KSTAT_DATA_UINT64 },
{ "uncompressed_size", KSTAT_DATA_UINT64 },
{ "overhead_size", KSTAT_DATA_UINT64 },
{ "hdr_size", KSTAT_DATA_UINT64 },
{ "data_size", KSTAT_DATA_UINT64 },
{ "metadata_size", KSTAT_DATA_UINT64 },
{ "other_size", KSTAT_DATA_UINT64 },
{ "anon_size", KSTAT_DATA_UINT64 },
{ "anon_evictable_data", KSTAT_DATA_UINT64 },
{ "anon_evictable_metadata", KSTAT_DATA_UINT64 },
{ "mru_size", KSTAT_DATA_UINT64 },
{ "mru_evictable_data", KSTAT_DATA_UINT64 },
{ "mru_evictable_metadata", KSTAT_DATA_UINT64 },
{ "mru_ghost_size", KSTAT_DATA_UINT64 },
{ "mru_ghost_evictable_data", KSTAT_DATA_UINT64 },
{ "mru_ghost_evictable_metadata", KSTAT_DATA_UINT64 },
{ "mfu_size", KSTAT_DATA_UINT64 },
{ "mfu_evictable_data", KSTAT_DATA_UINT64 },
{ "mfu_evictable_metadata", KSTAT_DATA_UINT64 },
{ "mfu_ghost_size", KSTAT_DATA_UINT64 },
{ "mfu_ghost_evictable_data", KSTAT_DATA_UINT64 },
{ "mfu_ghost_evictable_metadata", KSTAT_DATA_UINT64 },
{ "l2_hits", KSTAT_DATA_UINT64 },
{ "l2_misses", KSTAT_DATA_UINT64 },
{ "l2_feeds", KSTAT_DATA_UINT64 },
{ "l2_rw_clash", KSTAT_DATA_UINT64 },
{ "l2_read_bytes", KSTAT_DATA_UINT64 },
{ "l2_write_bytes", KSTAT_DATA_UINT64 },
{ "l2_writes_sent", KSTAT_DATA_UINT64 },
{ "l2_writes_done", KSTAT_DATA_UINT64 },
{ "l2_writes_error", KSTAT_DATA_UINT64 },
{ "l2_writes_lock_retry", KSTAT_DATA_UINT64 },
{ "l2_evict_lock_retry", KSTAT_DATA_UINT64 },
{ "l2_evict_reading", KSTAT_DATA_UINT64 },
{ "l2_evict_l1cached", KSTAT_DATA_UINT64 },
{ "l2_free_on_write", KSTAT_DATA_UINT64 },
{ "l2_abort_lowmem", KSTAT_DATA_UINT64 },
{ "l2_cksum_bad", KSTAT_DATA_UINT64 },
{ "l2_io_error", KSTAT_DATA_UINT64 },
{ "l2_size", KSTAT_DATA_UINT64 },
{ "l2_asize", KSTAT_DATA_UINT64 },
{ "l2_hdr_size", KSTAT_DATA_UINT64 },
{ "memory_throttle_count", KSTAT_DATA_UINT64 },
{ "arc_meta_used", KSTAT_DATA_UINT64 },
{ "arc_meta_limit", KSTAT_DATA_UINT64 },
{ "arc_meta_max", KSTAT_DATA_UINT64 },
{ "arc_meta_min", KSTAT_DATA_UINT64 },
{ "sync_wait_for_async", KSTAT_DATA_UINT64 },
{ "demand_hit_predictive_prefetch", KSTAT_DATA_UINT64 },
};
uint64_t m; \
continue; \
}
/*
* two separate conditions, giving a total of four different subtypes for
* each of hits and misses (so eight statistics total).
*/
if (cond1) { \
if (cond2) { \
} else { \
} \
} else { \
if (cond2) { \
} else { \
} \
}
/*
* There are several ARC variables that are critical to export as kstats --
* but we don't want to have to grovel around in the kstat whenever we wish to
* manipulate them. For these variables, we therefore define them to be in
* terms of the statistic variable. This assures that we are not introducing
* the possibility of inconsistency by having shadow copies of the variables,
* while still allowing the code to be readable.
*/
/* compressed size of entire arc */
/* uncompressed size of entire arc */
/* number of bytes in the arc from arc_buf_t's */
struct arc_callback {
void *acb_private;
};
struct arc_write_callback {
void *awcb_private;
};
/*
* ARC buffers are separated into multiple structs as a memory saving measure:
* - Common fields struct, always defined, and embedded within it:
* - L2-only fields, always allocated but undefined when not in L2ARC
* - L1-only fields, only allocated when in L1ARC
*
* Buffer in L1 Buffer only in L2
* +------------------------+ +------------------------+
* | arc_buf_hdr_t | | arc_buf_hdr_t |
* | | | |
* | | | |
* | | | |
* +------------------------+ +------------------------+
* | l2arc_buf_hdr_t | | l2arc_buf_hdr_t |
* | (undefined if L1-only) | | |
* +------------------------+ +------------------------+
* | l1arc_buf_hdr_t |
* | |
* | |
* | |
* | |
* +------------------------+
*
* Because it's possible for the L2ARC to become extremely large, we can wind
* up eating a lot of memory in L2ARC buffer headers, so the size of a header
* is minimized by only allocating the fields necessary for an L1-cached buffer
* when a header is actually in the L1 cache. The sub-headers (l1arc_buf_hdr and
* l2arc_buf_hdr) are embedded rather than allocated separately to save a couple
* words in pointers. arc_hdr_realloc() is used to switch a header between
* these two allocation states.
*/
typedef struct l1arc_buf_hdr {
#ifdef ZFS_DEBUG
/*
* Used for debugging with kmem_flags - by allocating and freeing
* b_thawed when the buffer is thawed, we get a record of the stack
* trace that thawed it.
*/
void *b_thawed;
#endif
/* for waiting on writes to complete */
/* protected by arc state mutex */
/* updated atomically */
/* self protecting */
void *b_pdata;
typedef struct l2arc_buf_hdr {
/* protected by arc_buf_hdr mutex */
struct arc_buf_hdr {
/* protected by hash lock */
/*
* This field stores the size of the data buffer after
* compression, and is set in the arc's zio completion handlers.
* It is in units of SPA_MINBLOCKSIZE (e.g. 1 == 512 bytes).
*
* While the block pointers can store up to 32MB in their psize
* field, we can only store up to 32MB minus 512B. This is due
* to the bp using a bias of 1, whereas we use a bias of 0 (i.e.
* a field of zeros represents 512B in the bp). We can't use a
* bias of 1 since we need to reserve a psize of zero, here, to
* represent holes and embedded blocks.
*
* This isn't a problem in practice, since the maximum size of a
* buffer is limited to 16MB, so we never need to store 32MB in
* this field. Even in the upstream illumos code base, the
* maximum size of a buffer is limited to 16MB.
*/
/*
* This field stores the size of the data buffer before
* compression, and cannot change once set. It is in units
* of SPA_MINBLOCKSIZE (e.g. 2 == 1024 bytes)
*/
/* L2ARC fields. Undefined when not in L2ARC. */
/* L1ARC fields. Undefined when in l2arc_only state */
};
(state) == arc_l2c_only)
/* For storing compression mode in b_flags */
/*
* Other sizes
*/
/*
* Hash table routines
*/
struct ht_lock {
#ifdef _KERNEL
#endif
};
typedef struct buf_hash_table {
/*
* Level 2 ARC
*/
/*
* If we discover during ARC scan any buffers to be compressed, we boost
* our headroom for the next scanning cycle by this percentage multiple.
*/
/* L2ARC Performance Tunables */
/*
* L2ARC Internals
*/
struct l2arc_dev {
};
typedef struct l2arc_read_callback {
typedef struct l2arc_write_callback {
typedef struct l2arc_data_free {
/* protected by l2arc_free_on_write_mtx */
void *l2df_data;
static void arc_hdr_alloc_pdata(arc_buf_hdr_t *);
static boolean_t arc_is_overflowing();
static void arc_buf_watch(arc_buf_t *);
static void l2arc_read_done(zio_t *);
static uint64_t
{
int i;
for (i = 0; i < sizeof (dva_t); i++)
return (crc);
}
static void
{
}
static arc_buf_hdr_t *
{
return (hdr);
}
}
return (NULL);
}
/*
* Insert an entry into the hash table. If there is already an element
* equal to elem in the hash table, then the already existing element
* will be returned and the new element will not be inserted.
* Otherwise returns NULL.
* If lockp == NULL, the caller is assumed to already hold the hash lock.
*/
static arc_buf_hdr_t *
{
uint32_t i;
} else {
}
return (fhdr);
}
/* collect some hash table performance data */
if (i > 0) {
if (i == 1)
}
return (NULL);
}
static void
{
}
/* collect some hash table performance data */
}
/*
* Global data structures and functions for the buf kmem cache.
*/
static void
buf_fini(void)
{
int i;
for (i = 0; i < BUF_LOCKS; i++)
}
/*
* Constructor callback - called when the cache is empty
* and a new buf is requested.
*/
/* ARGSUSED */
static int
{
return (0);
}
/* ARGSUSED */
static int
{
return (0);
}
/* ARGSUSED */
static int
{
return (0);
}
/*
* Destructor callback - called when a cached buf is
* no longer required.
*/
/* ARGSUSED */
static void
{
}
/* ARGSUSED */
static void
{
}
/* ARGSUSED */
static void
{
}
/*
* Reclaim callback -- invoked when memory is low.
*/
/* ARGSUSED */
static void
{
dprintf("hdr_recl called\n");
/*
* umem calls the reclaim func when we destroy the buf cache,
* which is after we do arc_fini().
*/
if (!arc_dead)
}
static void
buf_init(void)
{
int i, j;
/*
* The hash table is big enough to fill all of physical memory
* with an average block size of zfs_arc_average_blocksize (default 8K).
* By default, the table will take up
* totalmem * sizeof(void*) / 8K (1MB per GB with 8-byte pointers).
*/
hsize <<= 1;
hsize >>= 1;
goto retry;
}
for (i = 0; i < 256; i++)
for (i = 0; i < BUF_LOCKS; i++) {
}
}
/*
* This is the size that the buf occupies in memory. If the buf is compressed,
* it will correspond to the compressed size. You should use this method of
* getting the buf size unless you explicitly need the logical size.
*/
{
return (ARC_BUF_COMPRESSED(buf) ?
}
{
}
enum zio_compress
{
return (ARC_BUF_COMPRESSED(buf) ?
}
static inline boolean_t
{
/*
* It would be nice to assert arc_can_share() too, but the "hdr isn't
* already being shared" requirement prevents us from doing that.
*/
return (shared);
}
/*
* Free the checksum associated with this header. If there is no checksum, this
* is a no-op.
*/
static inline void
{
}
}
/*
* Return true iff at least one of the bufs on hdr is not compressed.
*/
static boolean_t
{
if (!ARC_BUF_COMPRESSED(b)) {
return (B_TRUE);
}
}
return (B_FALSE);
}
/*
* If we've turned on the ZFS_DEBUG_MODIFY flag, verify that the buf's data
* matches the checksum that is stored in the hdr. If there is no checksum,
* or if the buf is compressed, this is a no-op.
*/
static void
{
if (!(zfs_flags & ZFS_DEBUG_MODIFY))
return;
if (ARC_BUF_COMPRESSED(buf)) {
return;
}
return;
}
panic("buffer modified while frozen!");
}
static boolean_t
{
/*
* We rely on the blkptr's checksum to determine if the block
* is valid or not. When compressed arc is enabled, the l2arc
* writes the block to the l2arc just as it appears in the pool.
* This allows us to use the blkptr's checksum to validate the
* data that we just read off of the l2arc without having to store
* a separate checksum in the arc_buf_hdr_t. However, if compressed
* arc is disabled, then the data written to the l2arc is always
* uncompressed and won't match the block as it exists in the main
* pool. When this is the case, we must first compress it if it is
* compressed on the main pool before we can validate the checksum.
*/
/*
* Compressed blocks are always a multiple of the
* smallest ashift in the pool. Ideally, we would
* like to round up the csize to the next
* spa_min_ashift but that value may have changed
* since the block was last written. Instead,
* we rely on the fact that the hdr's psize
* was set to the psize of the block when it was
* last written. We set the csize to that value
* and zero out any part that should not contain
* data.
*/
}
}
/*
* Block pointers always store the checksum for the logical data.
* If the block pointer has the gang bit set, then the checksum
* it represents is for the reconstituted data and not for an
* individual gang member. The zio pipeline, however, must be able to
* determine the checksum of each of the gang constituents so it
* treats the checksum comparison differently than what we need
* for l2arc blocks. This prevents us from using the
* zio_checksum_error() interface directly. Instead we must call the
* zio_checksum_error_impl() so that we can ensure the checksum is
* generated using the correct checksum algorithm and accounts for the
* logical I/O size and not just a gang fragment.
*/
return (valid_cksum);
}
/*
* Given a buf full of data, if ZFS_DEBUG_MODIFY is enabled this computes a
* checksum and attaches it to the buf's hdr so that we can ensure that the buf
* isn't modified later on. If buf is compressed or there is already a checksum
* on the hdr, this is a no-op (we only checksum uncompressed bufs).
*/
static void
{
if (!(zfs_flags & ZFS_DEBUG_MODIFY))
return;
return;
} else if (ARC_BUF_COMPRESSED(buf)) {
return;
}
KM_SLEEP);
}
#ifndef _KERNEL
typedef struct procctl {
long cmd;
} procctl_t;
#endif
/* ARGSUSED */
static void
{
#ifndef _KERNEL
if (arc_watch) {
int result;
}
#endif
}
/* ARGSUSED */
static void
{
#ifndef _KERNEL
if (arc_watch) {
int result;
}
#endif
}
static arc_buf_contents_t
{
if (HDR_ISTYPE_METADATA(hdr)) {
} else {
}
return (type);
}
{
}
static uint32_t
{
switch (type) {
case ARC_BUFC_DATA:
/* metadata field is 0 if buffer contains normal data */
return (0);
case ARC_BUFC_METADATA:
return (ARC_FLAG_BUFC_METADATA);
default:
break;
}
panic("undefined ARC buffer type!");
return ((uint32_t)-1);
}
void
{
/*
* Compressed buffers do not manipulate the b_freeze_cksum or
* allocate b_thawed.
*/
if (ARC_BUF_COMPRESSED(buf)) {
return;
}
#ifdef ZFS_DEBUG
if (zfs_flags & ZFS_DEBUG_MODIFY) {
}
#endif
}
void
{
if (!(zfs_flags & ZFS_DEBUG_MODIFY))
return;
if (ARC_BUF_COMPRESSED(buf)) {
return;
}
}
/*
* The arc_buf_hdr_t's b_flags should never be modified directly. Instead,
* the following functions should be used to ensure that the flags are
* updated in a thread-safe way. When manipulating the flags either
* the hash_lock must be held or the hdr must be undiscoverable. This
* ensures that we're not racing with any other threads when updating
* the flags.
*/
static inline void
{
}
static inline void
{
}
/*
* Setting the compression bits in the arc_buf_hdr_t's b_flags is
* done in a special way since we have to clear and set bits
* at the same time. Consumers that wish to set the compression bits
* must use this function to ensure that the flags are updated in
* thread-safe manner.
*/
static void
{
/*
* Holes and embedded blocks will always have a psize = 0 so
* we ignore the compression of the blkptr and set the
* arc_buf_hdr_t's compression to ZIO_COMPRESS_OFF.
* Holes and embedded blocks remain anonymous so we don't
* want to uncompress them. Mark them as uncompressed.
*/
} else {
}
}
/*
* Looks for another buf on the same hdr which has the data decompressed, copies
* from it, and returns true. If no such buf exists, returns false.
*/
static boolean_t
{
/* can't use our own data buffer */
continue;
}
if (!ARC_BUF_COMPRESSED(from)) {
break;
}
}
/*
* There were no decompressed bufs, so there should not be a
* checksum on the hdr either.
*/
return (copied);
}
/*
* Given a buf that has a data buffer attached to it, this function will
* efficiently fill the buf with data of the specified compression setting from
* the hdr and update the hdr's b_freeze_cksum if necessary. If the buf and hdr
* are already sharing a data buf, no copy is performed.
*
* If the buf is marked as compressed but uncompressed data was requested, this
* will allocate a new data buffer for the buf, remove that flag, and fill the
* buf with uncompressed data. You can't request a compressed buf on a hdr with
* uncompressed data, and (since we haven't added support for it yet) if you
* want compressed data your buf must already be marked as compressed and have
* the correct-sized data buffer.
*/
static int
{
if (hdr_compressed == compressed) {
if (!arc_buf_is_shared(buf)) {
arc_buf_size(buf));
}
} else {
ASSERT(!compressed);
/*
* If the buf is sharing its data with the hdr, unlink it and
* allocate a new data buffer for the buf.
*/
if (arc_buf_is_shared(buf)) {
/* We need to give the buf it's own b_data */
/* Previously overhead was 0; just add new overhead */
} else if (ARC_BUF_COMPRESSED(buf)) {
/* We need to reallocate the buf's b_data */
buf);
/* We increased the size of b_data; update overhead */
}
/*
* Regardless of the buf's previous compression settings, it
* should not be compressed at the end of this function.
*/
/*
* Try copying the data from another buf which already has a
* decompressed version. If that's not possible, it's time to
* bite the bullet and decompress the data from the hdr.
*/
/* Skip byteswapping and checksumming (already done) */
return (0);
} else {
/*
* Absent hardware errors or software bugs, this should
* be impossible, but log it anyway so we can debug it.
*/
if (error != 0) {
"hdr %p, compress %d, psize %d, lsize %d",
}
}
}
/* Byteswap the buf's data if necessary */
if (bswap != DMU_BSWAP_NUMFUNCS) {
}
/* Compute the hdr's checksum if necessary */
return (0);
}
int
{
}
/*
* Return the size of the block, b_pdata, that is stored in the arc_buf_hdr_t.
*/
static uint64_t
{
HDR_GET_PSIZE(hdr) > 0) {
} else {
}
return (size);
}
/*
* Increment the amount of evictable space in the arc_state_t's refcount.
* We account for the space used by the hdr and the arc buf individually
* so that we can add and remove them from the refcount individually.
*/
static void
{
if (GHOST_STATE(state)) {
return;
}
}
if (arc_buf_is_shared(buf))
continue;
}
}
/*
* Decrement the amount of evictable space in the arc_state_t's refcount.
* We account for the space used by the hdr and the arc buf individually
* so that we can add and remove them from the refcount individually.
*/
static void
{
if (GHOST_STATE(state)) {
return;
}
}
if (arc_buf_is_shared(buf))
continue;
}
}
/*
* Add a reference to this hdr indicating that someone is actively
* referencing that memory. When the refcount transitions from 0 to 1,
* we remove it from the respective arc_state_t list to indicate that
* it is not evictable.
*/
static void
{
}
/* We don't use the L2-only state list. */
if (state != arc_l2c_only) {
hdr);
}
/* remove the prefetch flag if we get a reference */
}
}
/*
* Remove a reference from this hdr. When the reference transitions from
* 1 to 0 and we're not anonymous, then we add this hdr to the arc_state_t's
* list making it eligible for eviction.
*/
static int
{
int cnt;
/*
* arc_l2c_only counts as a ghost state so we don't need to explicitly
* check to prevent usage of the arc_l2c_only list.
*/
}
return (cnt);
}
/*
* Move the supplied buffer to the indicated state. The hash lock
* for the buffer must be held by the caller.
*/
static void
{
/*
* We almost always have an L1 hdr here, since we call arc_hdr_realloc()
* in arc_read() when bringing a buffer out of the L2ARC. However, the
* L1 hdr doesn't always exist when we change state to arc_anon before
* destroying a header, in which case reallocating to add the L1 hdr is
* pointless.
*/
if (HDR_HAS_L1HDR(hdr)) {
} else {
refcnt = 0;
bufcnt = 0;
}
/*
* If this buffer is evictable, transfer it from the
* old state list to the new state list.
*/
if (refcnt == 0) {
if (GHOST_STATE(old_state)) {
update_old = B_TRUE;
}
}
/*
* An L1 header always exists here, since if we're
* moving to some L1-cached state (i.e. not l2c_only or
* anonymous), we realloc the header to add an L1hdr
* beforehand.
*/
if (GHOST_STATE(new_state)) {
update_new = B_TRUE;
}
}
}
/* adjust state sizes (ignore arc_l2c_only) */
if (GHOST_STATE(new_state)) {
/*
* When moving a header to a ghost state, we first
* remove all arc buffers. Thus, we'll have a
* bufcnt of zero, and no arc buffer to use for
* the reference. As a result, we use the arc
* header pointer for the reference.
*/
} else {
/*
* Each individual buffer holds a unique reference,
* thus we must remove each of these references one
* at a time.
*/
buffers++;
/*
* When the arc_buf_t is sharing the data
* block with the hdr, the owner of the
* reference belongs to the hdr. Only
* add to the refcount if the arc_buf_t is
* not shared.
*/
if (arc_buf_is_shared(buf))
continue;
}
} else {
}
}
}
if (GHOST_STATE(old_state)) {
/*
* When moving a header off of a ghost state,
* the header will not contain any arc buffers.
* We use the arc header pointer for the reference
* which is exactly what we did when we put the
* header on the ghost state.
*/
} else {
/*
* Each individual buffer holds a unique reference,
* thus we must remove each of these references one
* at a time.
*/
buffers++;
/*
* When the arc_buf_t is sharing the data
* block with the hdr, the owner of the
* reference belongs to the hdr. Only
* add to the refcount if the arc_buf_t is
* not shared.
*/
if (arc_buf_is_shared(buf))
continue;
(void) refcount_remove_many(
buf);
}
(void) refcount_remove_many(
}
}
if (HDR_HAS_L1HDR(hdr))
/*
* L2 headers should never be on the L2 state list since they don't
* have L1 headers allocated.
*/
}
void
{
switch (type) {
case ARC_SPACE_DATA:
break;
case ARC_SPACE_META:
break;
case ARC_SPACE_OTHER:
break;
case ARC_SPACE_HDRS:
break;
case ARC_SPACE_L2HDRS:
break;
}
if (type != ARC_SPACE_DATA)
}
void
{
switch (type) {
case ARC_SPACE_DATA:
break;
case ARC_SPACE_META:
break;
case ARC_SPACE_OTHER:
break;
case ARC_SPACE_HDRS:
break;
case ARC_SPACE_L2HDRS:
break;
}
if (type != ARC_SPACE_DATA) {
if (arc_meta_max < arc_meta_used)
}
}
/*
* Given a hdr and a buf, returns whether that buf can share its b_data buffer
* with the hdr's b_pdata.
*/
static boolean_t
{
/*
* The criteria for sharing a hdr's data are:
* 1. the hdr's compression matches the buf's compression
* 2. the hdr doesn't need to be byteswapped
* 3. the hdr isn't already being shared
* 4. the buf is either compressed or it is the last buf in the hdr list
*
* Criterion #4 maintains the invariant that shared uncompressed
* bufs must be the final buf in the hdr's b_buf list. Reading this, you
* might ask, "if a compressed buf is allocated first, won't that be the
* last thing in the list?", but in that case it's impossible to create
* a shared uncompressed buf anyway (because the hdr must be compressed
* to have the compressed buf). You might also think that #3 is
* sufficient to make this guarantee, however it's possible
* (specifically in the rare L2ARC write race mentioned in
* arc_buf_alloc_impl()) there will be an existing uncompressed buf that
* is sharable, but wasn't at the time of its allocation. Rather than
* allow a new shared uncompressed buf to be created and then shuffle
* the list around to make it the last element, this simply disallows
* sharing if the new buf isn't the first to be added.
*/
return (buf_compressed == hdr_compressed &&
!HDR_SHARED_DATA(hdr) &&
}
/*
* Allocate a buf for this hdr. If you care about the data that's in the hdr,
* or if you want a compressed buffer, pass those flags in. Returns 0 if the
* copy was made successfully, or an error code otherwise.
*/
static int
{
/*
* We're about to change the hdr's b_flags. We must either
* hold the hash_lock or be undiscoverable.
*/
/*
* Only honor requests for compressed bufs if the hdr is actually
* compressed.
*/
/*
* If the hdr's data can be shared then we share the data buffer and
* set the appropriate bit in the hdr's b_flags to indicate the hdr is
* sharing it's b_pdata with the arc_buf_t. Otherwise, we allocate a new
* buffer to store the buf's data.
*
* There is one additional restriction here because we're sharing
* hdr -> buf instead of the usual buf -> hdr: the hdr can't be actively
* involved in an L2ARC write, because if this buf is used by an
* arc_write() then the hdr's data buffer will be released when the
* write completes, even though the L2ARC write might still be using it.
*/
/* Set up b_data and sharing */
if (can_share) {
} else {
}
/*
* If the user wants the data from the hdr, we need to either copy or
* decompress the data.
*/
if (fill) {
}
return (0);
}
static inline void
{
/* assert that it did not wrap around */
}
/*
* Loan out an anonymous arc buffer. Loaned buffers are not counted as in
* flight data by arc_tempreserve_space() until they are "returned". Loaned
* buffers must be returned to the arc before they can be used by the DMU or
* freed.
*/
{
return (buf);
}
enum zio_compress compression_type)
{
return (buf);
}
/*
* Return a loaned arc buffer to the arc.
*/
void
{
}
/* Detach an arc_buf from a dbuf (tag) */
void
{
}
static void
{
}
static void
{
/* protected by hash lock, if in the hash table */
}
}
/*
* Share the arc_buf_t's data with the hdr. Whenever we are sharing the
* data buffer, we transfer the refcount ownership to the hdr and update
* the appropriate kstats.
*/
static void
{
/*
* Start sharing the data buffer. We transfer the
* refcount ownership to the hdr since it always owns
* the refcount whenever an arc_buf_t is shared.
*/
/*
* Since we've transferred ownership to the hdr we need
* to increment its compressed and uncompressed kstats and
* decrement the overhead size.
*/
}
static void
{
/*
* We are no longer sharing this buffer so we need
* to transfer its ownership to the rightful owner.
*/
/*
* Since the buffer is no longer shared between
* the arc buf and the hdr, count it as overhead.
*/
}
/*
* Remove an arc_buf_t from the hdr's buf list and return the last
* arc_buf_t on the list. If no buffers remain on the list then return
* NULL.
*/
static arc_buf_t *
{
/*
* Remove the buf from the hdr list and locate the last
* remaining buffer on the list.
*/
/*
* If we've removed a buffer in the middle of
* the list then update the lastbuf and update
* bufp.
*/
}
}
return (lastbuf);
}
/*
* Free up buf->b_data and pull the arc_buf_t off of the the arc_buf_hdr_t's
* list and free it.
*/
static void
{
/*
* Free up the data associated with the buf but only if we're not
* sharing this with the hdr. If we are sharing it with the hdr, the
* hdr is responsible for doing the free.
*/
/*
* We're about to change the hdr's b_flags. We must either
* hold the hash_lock or be undiscoverable.
*/
if (arc_buf_is_shared(buf)) {
} else {
}
}
/*
* If the current arc_buf_t is sharing its data buffer with the
* hdr, then reassign the hdr's b_pdata to share it with the new
* buffer at the end of the list. The shared buffer is always
* the last one on the hdr's buffer list.
*
* There is an equivalent case for compressed bufs, but since
* they aren't guaranteed to be the last buf in the list and
* that is an exceedingly rare case, we just allow that space be
* wasted temporarily.
*/
/* Only one buf can be shared at once */
/* hdr is uncompressed so can't have compressed buf */
/*
* We must setup a new shared block between the
* last buffer and the hdr. The data would have
* been allocated by the arc buf so we need to transfer
* ownership to the hdr since it's now being shared.
*/
}
} else if (HDR_SHARED_DATA(hdr)) {
/*
* Uncompressed shared buffers are always at the end
* of the list. Compressed buffers don't have the
* same requirements. This makes it hard to
* simply assert that the lastbuf is shared so
* we rely on the hdr's compression flags to determine
* if we have a compressed, shared buffer.
*/
}
/*
* Free the checksum if we're removing the last uncompressed buf from
* this hdr.
*/
if (!arc_hdr_has_uncompressed_buf(hdr)) {
}
/* clean up the buf */
}
static void
{
}
static void
{
/*
* If the hdr is currently being written to the l2arc then
* we defer freeing the data by adding it to the l2arc_free_on_write
* list. The l2arc will free the data once it's finished
* writing it to the l2arc device.
*/
if (HDR_L2_WRITING(hdr)) {
} else {
}
}
static arc_buf_hdr_t *
{
/*
* Allocate the hdr's buffer. This will contain either
* the compressed or uncompressed data depending on the block
* it references and compressed arc enablement.
*/
return (hdr);
}
/*
* Transition between the two allocation states for the arc_buf_hdr struct.
* The arc_buf_hdr struct can be allocated with (hdr_full_cache) or without
* (hdr_l2only_cache) the fields necessary for the L1 cache - the smaller
* version is used when a cache buffer is only in the L2ARC in order to reduce
* memory usage.
*/
static arc_buf_hdr_t *
{
if (new == hdr_full_cache) {
/*
* arc_access and arc_change_state need to be aware that a
* header has just come out of L2ARC, so we set its state to
* l2c_only even though it's about to change.
*/
/* Verify previous threads set to NULL before freeing */
} else {
/*
* If we've reached here, We must have been called from
* arc_evict_hdr(), as such we should have already been
* removed from any ghost list we were previously on
* (which protects us from racing with arc_evict_state),
* thus no locking is needed during this check.
*/
/*
* A buffer must not be moved into the arc_l2c_only
* state if it's not finished being written out to the
* l2arc device. Otherwise, the b_l1hdr.b_pdata field
* might try to be accessed, even though it was removed.
*/
#ifdef ZFS_DEBUG
}
#endif
}
/*
* The header has been reallocated so we need to re-insert it into any
* lists it was on.
*/
/*
* We must place the realloc'ed header back into the list at
* the same spot. Otherwise, if it's placed earlier in the list,
* l2arc_write_buffers() could find it during the function's
* write phase, and try to write it out to the l2arc.
*/
/*
* Since we're using the pointer address as the tag when
* incrementing and decrementing the l2ad_alloc refcount, we
* must remove the old pointer (that we're about to destroy) and
* add the new pointer to the refcount. Otherwise we'd remove
* the wrong pointer address when calling arc_hdr_destroy() later.
*/
return (nhdr);
}
/*
* Allocate a new arc_buf_hdr_t and arc_buf_t and return the buf to the caller.
* The buf is returned thawed since we expect the consumer to modify it.
*/
{
return (buf);
}
/*
* Allocate a compressed buf in the same manner as arc_alloc_buf. Don't use this
* for bufs containing metadata.
*/
enum zio_compress compression_type)
{
return (buf);
}
static void
{
}
static void
{
if (HDR_HAS_L1HDR(hdr)) {
}
if (HDR_HAS_L2HDR(hdr)) {
if (!buflist_held)
/*
* Even though we checked this conditional above, we
* need to check this again now that we have the
* l2ad_mtx. This is because we could be racing with
* another thread calling l2arc_evict() which might have
* destroyed this header's L2 portion as we were waiting
* to acquire the l2ad_mtx. If that happens, we don't
* want to re-destroy the header's L2 portion.
*/
if (HDR_HAS_L2HDR(hdr))
if (!buflist_held)
}
if (HDR_HAS_L1HDR(hdr)) {
#ifdef ZFS_DEBUG
}
#endif
}
}
if (HDR_HAS_L1HDR(hdr)) {
} else {
}
}
void
{
return;
}
}
/*
* Evict the arc_buf_hdr that is provided as a parameter. The resultant
* state of the header is dependent on it's state prior to entering this
* function. The following transitions are possible:
*
* - arc_mru -> arc_mru_ghost
* - arc_mfu -> arc_mfu_ghost
* - arc_mru_ghost -> arc_l2c_only
* - arc_mru_ghost -> deleted
* - arc_mfu_ghost -> arc_l2c_only
* - arc_mfu_ghost -> deleted
*/
static int64_t
{
if (GHOST_STATE(state)) {
/*
* l2arc_write_buffers() relies on a header's L1 portion
* (i.e. its b_pdata field) during its write phase.
* Thus, we cannot push a header onto the arc_l2c_only
* state (removing it's L1 piece) until the header is
* done being written to the l2arc.
*/
return (bytes_evicted);
}
if (HDR_HAS_L2HDR(hdr)) {
/*
* This buffer is cached on the 2nd Level ARC;
* don't destroy the header.
*/
/*
* dropping from L1+L2 cached to L2-only,
* realloc to remove the L1 header.
*/
} else {
}
return (bytes_evicted);
}
/* prefetch buffers have a minimum lifespan */
if (HDR_IO_IN_PROGRESS(hdr) ||
return (bytes_evicted);
}
break;
}
}
if (HDR_HAS_L2HDR(hdr)) {
} else {
HDR_GET_LSIZE(hdr));
} else {
HDR_GET_LSIZE(hdr));
}
}
/*
* If this hdr is being evicted and has a compressed
* buffer then we discard it here before we change states.
* This ensures that the accounting is updated correctly
* in arc_free_data_buf().
*/
}
return (bytes_evicted);
}
static uint64_t
{
int evict_count = 0;
break;
/*
* To keep our iteration location, move the marker
* forward. Since we're not holding hdr's hash lock, we
* must be very careful and not remove 'hdr' from the
* sublist. Otherwise, other consumers might mistake the
* 'hdr' as not being on a sublist when they call the
* multilist_link_active() function (they all rely on
* the hash lock protecting concurrent insertions and
* removals). multilist_sublist_move_forward() was
* specifically implemented to ensure this is the case
* (only 'marker' will be removed and re-inserted).
*/
/*
* The only case where the b_spa field should ever be
* zero, is the marker headers inserted by
* arc_evict_state(). It's possible for multiple threads
* to be calling arc_evict_state() concurrently (e.g.
* dsl_pool_close() and zio_inject_fault()), so we must
* skip any markers we see from these other threads.
*/
continue;
/* we're only interested in evicting buffers of a certain spa */
continue;
}
/*
* We aren't calling this function from any code path
* that would already be holding a hash lock, so we're
* asserting on this assumption to be defensive in case
* this ever changes. Without this check, it would be
* possible to incorrectly increment arcstat_mutex_miss
* below (e.g. if the code changed such that we called
* this function with a hash lock held).
*/
if (mutex_tryenter(hash_lock)) {
bytes_evicted += evicted;
/*
* If evicted is zero, arc_evict_hdr() must have
* decided to skip this header, don't increment
* evict_count in this case.
*/
if (evicted != 0)
evict_count++;
/*
* If arc_size isn't overflowing, signal any
* threads that might happen to be waiting.
*
* For each header evicted, we wake up a single
* thread. If we used cv_broadcast, we could
* wake up "too many" threads causing arc_size
* to significantly overflow arc_c; since
* arc_get_data_buf() doesn't check for overflow
* when it's woken up (it doesn't because it's
* possible for the ARC to be overflowing while
* full of un-evictable buffers, and the
* function should proceed in this case).
*
* If threads are left sleeping, due to not
* using cv_broadcast, they will be woken up
* just before arc_reclaim_thread() sleeps.
*/
if (!arc_is_overflowing())
} else {
}
}
return (bytes_evicted);
}
/*
* Evict buffers from the given arc state, until we've removed the
* specified number of bytes. Move the removed buffers to the
* appropriate evict state.
*
* This function makes a "best effort". It skips over any buffers
* it can't get a hash_lock on, and so, may not catch all candidates.
* It may also return without evicting as much space as requested.
*
* If bytes is specified using the special value ARC_EVICT_ALL, this
* will evict all available (i.e. unlocked and evictable) buffers from
* the given arc state; which is used by arc_flush().
*/
static uint64_t
{
int num_sublists;
/*
* If we've tried to evict from each sublist, made some
* progress, but still have not hit the target number of bytes
* to evict, we want to keep trying. The markers allow us to
* pick up where we left off for each individual sublist, rather
* than starting from the tail each time.
*/
for (int i = 0; i < num_sublists; i++) {
/*
* A b_spa of 0 is used to indicate that this header is
* a marker. This fact is used in arc_adjust_type() and
* arc_evict_state_impl().
*/
}
/*
* While we haven't hit our target number of bytes to evict, or
* we're evicting all available buffers.
*/
/*
* Start eviction using a randomly selected sublist,
* this is to try and evenly balance eviction across all
* sublists. Always starting at the same sublist
* (e.g. index 0) would cause evictions to favor certain
* sublists over others.
*/
for (int i = 0; i < num_sublists; i++) {
if (bytes == ARC_EVICT_ALL)
else if (total_evicted < bytes)
else
break;
/* we've reached the end, wrap to the beginning */
if (++sublist_idx >= num_sublists)
sublist_idx = 0;
}
/*
* If we didn't evict anything during this scan, we have
* no reason to believe we'll evict more during another
* scan, so break the loop.
*/
if (scan_evicted == 0) {
/* This isn't possible, let's make that obvious */
/*
* When bytes is ARC_EVICT_ALL, the only way to
* break the loop is when scan_evicted is zero.
* In that case, we actually have evicted enough,
* so we don't want to increment the kstat.
*/
if (bytes != ARC_EVICT_ALL) {
}
break;
}
}
for (int i = 0; i < num_sublists; i++) {
}
return (total_evicted);
}
/*
* Flush all "evictable" data of the given type from the arc state
* specified. This will not evict any "active" buffers (i.e. referenced).
*
* When 'retry' is set to B_FALSE, the function will make a single pass
* over the state and evict any buffers that it can. Since it doesn't
* continually retry the eviction, it might end up leaving some buffers
* in the ARC due to lock misses.
*
* When 'retry' is set to B_TRUE, the function will continually retry the
* eviction until *all* evictable buffers have been removed from the
* state. As a result, if concurrent insertions into the state are
* allowed (e.g. if the ARC isn't shutting down), this function might
* wind up in an infinite loop, continually trying to evict buffers.
*/
static uint64_t
{
if (!retry)
break;
}
return (evicted);
}
/*
* Evict the specified number of bytes from the state specified,
* restricting eviction to the spa and type given. This function
* prevents us from trying to evict more from a state's list than
* is "evictable", and to skip evicting altogether when passed a
* negative value for "bytes". In contrast, arc_evict_state() will
* evict everything it can, when passed a negative value for "bytes".
*/
static uint64_t
{
}
return (0);
}
/*
* Evict metadata buffers from the cache, such that arc_meta_used is
* capped by the arc_meta_limit tunable.
*/
static uint64_t
arc_adjust_meta(void)
{
/*
* If we're over the meta limit, we want to evict enough
* metadata to get back under the meta limit. We don't want to
* evict so much that we drop the MRU below arc_p, though. If
* we're over the meta limit more than we're over arc_p, we
* evict some from the MRU here, and some from the MFU below.
*/
/*
* Similar to the above, we want to evict enough bytes to get us
* below the meta limit, but not so much as to drop us below the
* space allotted to the MFU (which is defined as arc_c - arc_p).
*/
return (total_evicted);
}
/*
* Return the type of the oldest buffer in the given arc state
*
* This function will select a random sublist of type ARC_BUFC_DATA and
* a random sublist of type ARC_BUFC_METADATA. The tail of each sublist
* is compared, and the type which contains the "older" buffer will be
* returned.
*/
static arc_buf_contents_t
{
/*
* We keep the sublist lock until we're finished, to prevent
* the headers from being destroyed via arc_evict_state().
*/
/*
* These two loops are to ensure we skip any markers that
* might be at the tail of the lists due to arc_evict_state().
*/
break;
}
break;
}
} else {
/* The headers can't be on the sublist without an L1 header */
} else {
}
}
return (type);
}
/*
* Evict buffers from the cache, such that arc_size is capped by arc_c.
*/
static uint64_t
arc_adjust(void)
{
/*
* If we're over arc_meta_limit, we want to correct that before
* potentially evicting data buffers below.
*/
total_evicted += arc_adjust_meta();
/*
* Adjust MRU size
*
* If we're over the target cache size, we want to evict enough
* from the list to get back to our target size. We don't want
* to evict too much from the MRU, such that it drops below
* arc_p. So, if we're over our target cache size more than
* the MRU is over arc_p, we'll evict enough to get back to
* arc_p here, and then evict more from the MFU below.
*/
/*
* If we're below arc_meta_min, always prefer to evict data.
* Otherwise, try to satisfy the requested number of bytes to
* evict from the type which contains older buffers; in an
* effort to keep newer buffers in the cache regardless of their
* type. If we cannot satisfy the number of bytes from this
* type, spill over into the next type.
*/
total_evicted += bytes;
/*
* If we couldn't evict our target number of bytes from
* metadata, we try to get the rest from data.
*/
} else {
total_evicted += bytes;
/*
* If we couldn't evict our target number of bytes from
* data, we try to get the rest from metadata.
*/
}
/*
* Adjust MFU size
*
* Now that we've tried to evict enough from the MRU to get its
* size back to arc_p, if we're still above the target cache
* size, we evict the rest from the MFU.
*/
total_evicted += bytes;
/*
* If we couldn't evict our target number of bytes from
* metadata, we try to get the rest from data.
*/
} else {
total_evicted += bytes;
/*
* If we couldn't evict our target number of bytes from
* data, we try to get the rest from data.
*/
}
/*
* Adjust ghost lists
*
* In addition to the above, the ARC also defines target values
* for the ghost lists. The sum of the mru list and mru ghost
* list should never exceed the target size of the cache, and
* the sum of the mru list, mfu list, mru ghost list, and mfu
* ghost list should never exceed twice the target size of the
* cache. The following logic enforces these limits on the ghost
* caches, and evicts from them as needed.
*/
total_evicted += bytes;
/*
* We assume the sum of the mru list and mfu list is less than
* or equal to arc_c (we enforced this above), which means we
* can use the simpler of the two equations below:
*
* mru + mfu + mru ghost + mfu ghost <= 2 * arc_c
* mru ghost + mfu ghost <= arc_c
*/
total_evicted += bytes;
return (total_evicted);
}
void
{
/*
* If retry is B_TRUE, a spa must not be specified since we have
* no good way to determine if all of a spa's buffers have been
* evicted from an arc state.
*/
}
void
{
else
}
(void) arc_adjust();
}
typedef enum free_memory_reason_t {
/*
* Additional reserve of pages for pp_reserve.
*/
/*
* Additional reserve of pages for swapfs.
*/
/*
* Return the amount of memory that can be consumed before reclaim will be
* needed. Positive if there is sufficient free memory, negative indicates
* the amount of memory that needs to be freed up.
*/
static int64_t
arc_available_memory(void)
{
int64_t n;
#ifdef _KERNEL
if (needfree > 0) {
if (n < lowest) {
lowest = n;
r = FMR_NEEDFREE;
}
}
/*
* check that we're out of range of the pageout scanner. It starts to
* schedule paging if freemem is less than lotsfree and needfree.
* lotsfree is the high-water mark for pageout, and needfree is the
* number of needed free pages. We add extra pages here to make sure
* the scanner doesn't start up while we're freeing memory.
*/
if (n < lowest) {
lowest = n;
r = FMR_LOTSFREE;
}
/*
* check to make sure that swapfs has enough space so that anon
* reservations can still succeed. anon_resvmem() checks that the
* availrmem is greater than swapfs_minfree, and the number of reserved
* swap pages. We also add a bit of extra here just to prevent
* circumstances from getting really dire.
*/
if (n < lowest) {
lowest = n;
r = FMR_SWAPFS_MINFREE;
}
/*
* Check that we have enough availrmem that memory locking (e.g., via
* mlock(3C) or memcntl(2)) can still succeed. (pages_pp_maximum
* stores the number of pages that cannot be locked; when availrmem
* drops below pages_pp_maximum, page locking mechanisms such as
* page_pp_lock() will fail.)
*/
if (n < lowest) {
lowest = n;
r = FMR_PAGES_PP_MAXIMUM;
}
#if defined(__i386)
/*
* If we're on an i386 platform, it's possible that we'll exhaust the
* kernel heap space before we ever run out of available physical
* memory. Most checks of the size of the heap_area compare against
* tune.t_minarmem, which is the minimum available real memory that we
* can have in the system. However, this is generally fixed at 25 pages
* which is so low that it's useless. In this comparison, we seek to
* calculate the total heap-size, and reclaim if more than 3/4ths of the
* heap is allocated. (Or, in the calculation, if less than 1/4th is
* free)
*/
if (n < lowest) {
lowest = n;
r = FMR_HEAP_ARENA;
}
#endif
/*
* If zio data pages are being allocated out of a separate heap segment,
* then enforce that the size of available vmem for this arena remains
* above about 1/4th (1/(2^arc_zio_arena_free_shift)) free.
*
* Note that reducing the arc_zio_arena_free_shift keeps more virtual
* memory (in the zio_arena) free, which can avoid memory
* fragmentation issues.
*/
if (n < lowest) {
lowest = n;
r = FMR_ZIO_ARENA;
}
}
#else
/* Every 100 calls, free a small amount */
if (spa_get_random(100) == 0)
lowest = -1024;
#endif
last_free_reason = r;
return (lowest);
}
/*
* Determine if the system is under memory pressure and is asking
* to reclaim memory. A return value of B_TRUE indicates that the system
* is under memory pressure and that the arc should adjust accordingly.
*/
static boolean_t
arc_reclaim_needed(void)
{
return (arc_available_memory() < 0);
}
static void
arc_kmem_reap_now(void)
{
size_t i;
extern kmem_cache_t *zio_buf_cache[];
extern kmem_cache_t *zio_data_buf_cache[];
extern kmem_cache_t *range_seg_cache;
#ifdef _KERNEL
if (arc_meta_used >= arc_meta_limit) {
/*
* We are exceeding our meta-data cache limit.
* Purge some DNLC entries to release holds on meta-data.
*/
}
#if defined(__i386)
/*
* Reclaim unused memory from all kmem caches.
*/
kmem_reap();
#endif
#endif
for (i = 0; i < SPA_MAXBLOCKSIZE >> SPA_MINBLOCKSHIFT; i++) {
if (zio_buf_cache[i] != prev_cache) {
prev_cache = zio_buf_cache[i];
}
if (zio_data_buf_cache[i] != prev_data_cache) {
}
}
/*
* Ask the vmem arena to reclaim unused memory from its
* quantum caches.
*/
}
}
/*
* Threads can block in arc_get_data_buf() waiting for this thread to evict
* enough data and signal them to proceed. When this happens, the threads in
* arc_get_data_buf() are sleeping while holding the hash lock for their
* particular arc header. Thus, we must be careful to never sleep on a
* hash lock in this thread. This is to prevent the following deadlock:
*
* - Thread A sleeps on CV in arc_get_data_buf() holding hash lock "L",
* waiting for the reclaim thread to signal it.
*
* - arc_reclaim_thread() tries to acquire hash lock "L" using mutex_enter,
* fails, and goes to sleep forever.
*
* This possible deadlock is avoided by always acquiring a hash lock
* using mutex_tryenter() from arc_reclaim_thread().
*/
static void
arc_reclaim_thread(void)
{
while (!arc_reclaim_thread_exit) {
/*
* This is necessary in order for the mdb ::arc dcmd to
* show up to date information. Since the ::arc command
* does not call the kstat's update function, without
* this call, the command may show stale stats for the
* anon, mru, mru_ghost, mfu, and mfu_ghost lists. Even
* with this change, the data might be up to 1 second
* out of date; but that should suffice. The arc_state_t
* structures can be queried directly if more accurate
* information is needed.
*/
/*
* We call arc_adjust() before (possibly) calling
* arc_kmem_reap_now(), so that we can wake up
* arc_get_data_buf() sooner.
*/
evicted = arc_adjust();
if (free_memory < 0) {
/*
* Wait at least zfs_grow_retry (default 60) seconds
* before considering growing.
*/
/*
* If we are still low on memory, shrink the ARC
* so that we have arc_shrink_min free space.
*/
if (to_free > 0) {
#ifdef _KERNEL
#endif
}
}
/*
* If evicted is zero, we couldn't evict anything via
* arc_adjust(). This could be due to hash lock
* collisions, but more likely due to the majority of
* arc buffers being unevictable. Therefore, even if
* arc_size is above arc_c, another pass is unlikely to
* be helpful and could potentially cause us to enter an
* infinite loop.
*/
/*
* We're either no longer overflowing, or we
* can't evict anything more, so we should wake
* up any threads before we go to sleep.
*/
/*
* Block until signaled, or after one second (we
* might need to perform arc_kmem_reap_now()
* even if we aren't being signalled)
*/
(void) cv_timedwait_hires(&arc_reclaim_thread_cv,
}
}
thread_exit();
}
/*
* Adapt arc info given the number of bytes we are trying to add and
* the state that we are comming from. This function is only called
* when we are adding new content to the cache.
*/
static void
{
int mult;
if (state == arc_l2c_only)
return;
/*
* Adapt the target size of the MRU list:
* - if we just hit in the MRU ghost list, then increase
* the target size of the MRU list.
* - if we just hit in the MFU ghost list, then increase
* the target size of the MFU list by decreasing the
* target size of the MRU list.
*/
if (state == arc_mru_ghost) {
} else if (state == arc_mfu_ghost) {
}
if (arc_reclaim_needed()) {
return;
}
if (arc_no_grow)
return;
return;
/*
* If we're within (2 * maxblocksize) bytes of the target
* cache size, increment the target cache size
*/
}
}
/*
* Check if arc_size has grown past our upper threshold, determined by
* zfs_arc_overflow_shift.
*/
static boolean_t
arc_is_overflowing(void)
{
/* Always allow at least one block of overflow */
}
/*
* Allocate a block and return it to the caller. If we are hitting the
* hard limit for the cache size, we must sleep, waiting for the eviction
* thread to catch up. If we're past the target size but below the hard
* limit, we'll only signal the reclaim thread and continue on.
*/
static void *
{
/*
* If arc_size is currently overflowing, and has grown past our
* upper limit, we must be adding data faster than the evict
* thread can evict. Thus, to ensure we don't compound the
* problem by adding more data and forcing arc_size to grow even
* further past it's target size, we halt and wait for the
* eviction thread to catch up.
*
* It's also possible that the reclaim thread is unable to evict
* enough buffers to get arc_size below the overflow limit (e.g.
* due to buffers being un-evictable, or hash lock collisions).
* In this case, we want to proceed regardless if we're
* overflowing; thus we don't use a while loop here.
*/
if (arc_is_overflowing()) {
/*
* Now that we've acquired the lock, we may no longer be
* over the overflow limit, lets check.
*
* We're ignoring the case of spurious wake ups. If that
* were to happen, it'd let this thread consume an ARC
* buffer before it should have (i.e. before we're under
* the overflow limit and were signalled by the reclaim
* thread). As long as that is a rare occurrence, it
* shouldn't cause any harm.
*/
if (arc_is_overflowing()) {
}
}
if (type == ARC_BUFC_METADATA) {
} else {
}
/*
* Update the state size. Note that ghost states have a
* "ghost size" and so don't need to be updated.
*/
if (!GHOST_STATE(state)) {
/*
* If this is reached via arc_read, the link is
* protected by the hash lock. If reached via
* arc_buf_alloc, the header should not be accessed by
* any other thread. And, if reached via arc_read_done,
* the hash lock will protect it if it's found in the
* hash table; otherwise no other thread should be
* trying to [add|remove]_reference it.
*/
}
/*
* If we are growing the cache, and we are adding anonymous
* data, and we have outgrown arc_p, update arc_p
*/
}
return (datap);
}
/*
* Free the arc data buffer.
*/
static void
{
/* protected by hash lock, if in the hash table */
}
if (type == ARC_BUFC_METADATA) {
} else {
}
}
/*
* This routine is called whenever a buffer is accessed.
* NOTE: the hash lock is dropped in this function.
*/
static void
{
/*
* This buffer is not in the cache, and does not
* appear in our "ghost" list. Add the new buffer
* to the MRU state.
*/
now = ddi_get_lbolt();
/*
* If this buffer is here because of a prefetch, then either:
* - clear the flag if this is a "referencing" read
* (any subsequent access will bump this into the MFU state).
* or
* - move the buffer to the head of the list if this is
* another prefetch (to make it less likely to be evicted).
*/
if (HDR_PREFETCH(hdr)) {
/* link protected by hash lock */
} else {
}
return;
}
/*
* This buffer has been "accessed" only once so far,
* but it is still in the cache. Move it to the MFU
* state.
*/
/*
* More than 125ms have passed since we
* instantiated this buffer. Move it to the
* most frequently used state.
*/
}
/*
* This buffer has been "accessed" recently, but
* was evicted from the cache. Move it to the
* MFU state.
*/
if (HDR_PREFETCH(hdr)) {
} else {
}
/*
* This buffer has been accessed more than once and is
* still in the cache. Keep it in the MFU state.
*
* NOTE: an add_reference() that occurred when we did
* the arc_read() will have kicked this off the list.
* If it was a prefetch, we will explicitly move it to
* the head of the list now.
*/
if ((HDR_PREFETCH(hdr)) != 0) {
/* link protected by hash_lock */
}
/*
* This buffer has been accessed more than once but has
* been evicted from the cache. Move it back to the
* MFU state.
*/
if (HDR_PREFETCH(hdr)) {
/*
* This is a prefetch access...
* move this block back to the MRU state.
*/
}
/*
* This buffer is on the 2nd Level ARC.
*/
} else {
ASSERT(!"invalid arc state");
}
}
/* a generic arc_done_func_t which you can use */
/* ARGSUSED */
void
{
}
/* a generic arc_done_func_t */
void
{
} else {
}
}
static void
{
} else {
if (HDR_COMPRESSION_ENABLED(hdr)) {
}
}
}
static void
{
/*
* The hdr was inserted into hash-table and removed from lists
* prior to starting I/O. We should find this header, since
* it's in the hash table, and it should be legit since it's
* not possible to evict it during the I/O. The only possible
* reason for it not to be found is if we were freed during the
* read.
*/
if (HDR_IN_HASH_TABLE(hdr)) {
&hash_lock);
}
if (no_zio_error) {
/* byteswap if necessary */
} else {
}
} else {
}
}
/*
* Only call arc_access on anonymous buffers. This is because
* if we've issued an I/O for an evicted buffer, we've already
* called arc_access (to prevent any simultaneous readers from
* getting confused).
*/
}
/*
* If a read request has a callback (i.e. acb_done is not NULL), then we
* make a buf containing the data according to the parameters which were
* passed in. The implementation of arc_buf_alloc_impl() ensures that we
* aren't needlessly decompressing the data multiple times.
*/
int callback_cnt = 0;
continue;
/* This is a demand read since prefetches don't use callbacks */
callback_cnt++;
if (no_zio_error) {
}
}
if (callback_cnt == 0) {
}
if (no_zio_error) {
} else {
if (HDR_IN_HASH_TABLE(hdr))
}
/*
* Broadcast before we drop the hash_lock to avoid the possibility
* that the hdr (and hence the cv) might be freed before we get to
* the cv_broadcast().
*/
} else {
/*
* This block was freed while we waited for the read to
* complete. It has been removed from the hash table and
* moved to the anonymous state (so that it won't show up
* in the cache).
*/
}
/* execute each callback and free its structure */
}
}
if (freeable)
}
/*
* "Read" the block at the specified DVA (in bp) via the
* cache. If the block is found in the cache, invoke the provided
* callback immediately and return. Note that the `zio' parameter
* in the callback will be NULL in this case, since no IO was
* required. If the block is not in the cache pass the read request
* on to the spa with a substitute callback function, so that the
* requested block will be added to the cache.
*
* If a read request arrives for a block that has a read in-progress,
* either wait for the in-progress read to complete (and return the
* results); or, if this is a read with a "done" func, add a record
* to the read to invoke the "done" func when the read completes,
* and return; or just return.
*
* arc_read_done() will invoke all the requested "done" functions
* for readers of this block.
*/
int
{
top:
if (!BP_IS_EMBEDDED(bp)) {
/*
* Embedded BP's have no DVA and require no I/O to "read".
* Create an anonymous arc buf to back it.
*/
}
*arc_flags |= ARC_FLAG_CACHED;
if (HDR_IO_IN_PROGRESS(hdr)) {
/*
* This sync read must wait for an
* in-progress async read (e.g. a predictive
* prefetch). Async reads are queued
* separately at the vdev_queue layer, so
* this is a form of priority inversion.
* Ideally, we would "inherit" the demand
* i/o's priority by moving the i/o from
* the async queue to the synchronous queue,
* but there is currently no mechanism to do
* so. Track this so that we can evaluate
* the magnitude of this potential performance
* problem.
*
* Note that if the prefetch i/o is already
* active (has been issued to the device),
* the prefetch improved performance, because
* we issued it sooner than we would have
* without the prefetch.
*/
arc_buf_hdr_t *, hdr);
}
}
if (*arc_flags & ARC_FLAG_WAIT) {
goto top;
}
if (done) {
KM_SLEEP);
return (0);
}
return (0);
}
if (done) {
/*
* This is a demand read which does not have to
* wait for i/o because we did a predictive
* prefetch i/o for it, which has completed.
*/
arc_buf_hdr_t *, hdr);
}
/* Get a buf with the desired data in it. */
} else if (*arc_flags & ARC_FLAG_PREFETCH &&
}
if (*arc_flags & ARC_FLAG_L2CACHE)
if (done)
} else {
/* this block is not in the cache */
if (!BP_IS_EMBEDDED(bp)) {
}
/* somebody beat us to the hash insert */
goto top; /* restart the IO request */
}
} else {
/*
* This block is in the ghost cache. If it was L2-only
* (and thus didn't have an L1 hdr), we realloc the
* header to add an L1 hdr.
*/
if (!HDR_HAS_L1HDR(hdr)) {
}
/*
* This is a delicate dance that we play here.
* This hdr is in the ghost list so we access it
* to move it out of the ghost list before we
* initiate the read. If it's a prefetch then
* it won't have a callback so we'll remove the
* reference that arc_buf_alloc_impl() created. We
* do this after we've called arc_access() to
* avoid hitting an assert in remove_reference().
*/
}
/*
* If compression is enabled on the hdr, then will do
* RAW I/O and will store the compressed data in the hdr's
* data block. Otherwise, the hdr's data block will contain
* the uncompressed data.
*/
}
if (*arc_flags & ARC_FLAG_PREFETCH)
if (*arc_flags & ARC_FLAG_L2CACHE)
if (BP_GET_LEVEL(bp) > 0)
if (HDR_HAS_L2HDR(hdr) &&
/*
* Lock out device removal.
*/
if (vdev_is_dead(vd) ||
}
if (priority == ZIO_PRIORITY_ASYNC_READ)
else
/*
* At this point, we have a level 1 cache miss. Try again in
* L2ARC if possible.
*/
/*
* Read from the L2ARC if the following are true:
* 1. The L2ARC vdev was previously cached.
* 2. This buffer still has L2ARC metadata.
* 3. This buffer isn't currently writing to the L2ARC.
* 4. The L2ARC entry wasn't evicted, which may
* also have invalidated the vdev.
* 5. This isn't prefetch and l2arc_noprefetch is set.
*/
if (HDR_HAS_L2HDR(hdr) &&
KM_SLEEP);
/*
* l2arc read. The SCL_L2ARC lock will be
* released by l2arc_read_done().
* Issue a null zio if the underlying buffer
* was squashed to zero size by compression.
*/
if (*arc_flags & ARC_FLAG_NOWAIT) {
return (0);
}
return (0);
/* l2arc read error; goto zio_read() */
} else {
arc_buf_hdr_t *, hdr);
if (HDR_L2_WRITING(hdr))
}
} else {
if (l2arc_ndev != 0) {
arc_buf_hdr_t *, hdr);
}
}
if (*arc_flags & ARC_FLAG_WAIT)
}
return (0);
}
/*
* Notify the arc that a block was freed, and thus will never be used again.
*/
void
{
return;
/*
* We might be trying to free a block that is still doing I/O
* (i.e. prefetch) or has a reference (i.e. a dedup-ed,
* dmu_sync-ed block). If this block is being prefetched, then it
* would still have the ARC_FLAG_IO_IN_PROGRESS flag set on the hdr
* until the I/O completes. A block may also have a reference if it is
* part of a dedup-ed, dmu_synced write. The dmu_sync() function would
* have written the new block to its final resting place on disk but
* without the dedup flag set. This would have left the hdr in the MRU
* state and discoverable. When the txg finally syncs it detects that
* the block was overridden in open context and issues an override I/O.
* Since this is a dedup block, the override I/O will determine if the
* block is already in the DDT. If so, then it will replace the io_bp
* with the bp from the DDT and allow the I/O to finish. When the I/O
* reaches the done callback, dbuf_write_override_done, it will
* check to see if the io_bp and io_bp_override are identical.
* If they are not, then it indicates that the bp was replaced with
* the bp in the DDT and the override bp is freed. This allows
* us to arrive here with a reference on a block that is being
* freed. So if we have an I/O in progress, or a reference to
* this hdr, then we don't destroy the hdr.
*/
} else {
}
}
/*
* Release this buffer from the cache, making it an anonymous buffer. This
* must be done after a read and prior to modifying the buffer contents.
* If the buffer has more than one reference, we must make
* a new hdr for the buffer.
*/
void
{
/*
* It would be nice to assert that if it's DMU metadata (level >
* 0 || it's the dnode file), then it must be syncing context.
* But we don't know that information at this level.
*/
/*
* We don't grab the hash lock prior to this check, because if
* the buffer's header is in the arc_anon state, it won't be
* linked into the hash table.
*/
/*
* If the buf is being overridden then it may already
* have a hdr that is not empty.
*/
return;
}
/*
* This assignment is only valid as long as the hash_lock is
* held, we must be careful not to reference state or the
* b_state field after dropping the lock.
*/
/* this buffer is not on any list */
if (HDR_HAS_L2HDR(hdr)) {
/*
* We have to recheck this conditional again now that
* we're holding the l2ad_mtx to prevent a race with
* another thread which might be concurrently calling
* l2arc_evict(). In that case, l2arc_evict() might have
* destroyed the header's L2 portion as we were waiting
* to acquire the l2ad_mtx.
*/
if (HDR_HAS_L2HDR(hdr))
}
/*
* Do we have more than one buf?
*/
}
/*
* Pull the data off of this hdr and attach it to
* a new anonymous hdr. Also find the last buffer
* in the hdr's buffer list.
*/
/*
* If the current arc_buf_t and the hdr are sharing their data
* buffer, then we must stop sharing that block.
*/
if (arc_buf_is_shared(buf)) {
/*
* First, sever the block sharing relationship between
* buf and the arc_buf_hdr_t.
*/
/*
* Now we need to recreate the hdr's b_pdata. Since we
* have lastbuf handy, we try to share with it, but if
* we can't then we allocate a new b_pdata and copy the
* data from buf into it.
*/
} else {
}
} else if (HDR_SHARED_DATA(hdr)) {
/*
* Uncompressed shared buffers are always at the end
* of the list. Compressed buffers don't have the
* same requirements. This makes it hard to
* simply assert that the lastbuf is shared so
* we rely on the hdr's compression flags to determine
* if we have a compressed, shared buffer.
*/
}
}
/*
* Allocate a new hdr. The new hdr will contain a b_pdata
* buffer which will be freed in arc_write().
*/
} else {
/* protected by hash lock, or hdr is on arc_anon */
}
}
int
{
int released;
return (released);
}
#ifdef ZFS_DEBUG
int
{
int referenced;
return (referenced);
}
#endif
static void
{
/*
* If we're reexecuting this zio because the pool suspended, then
* cleanup any state that was previously set the first time the
* callback was invoked.
*/
if (arc_buf_is_shared(buf)) {
} else {
}
}
}
if (HDR_IO_IN_PROGRESS(hdr))
} else {
}
/*
* If the hdr is compressed, then copy the compressed
* zio contents into arc_buf_hdr_t. Otherwise, copy the original
* data buf into the hdr. Ideally, we would like to always copy the
* io_data into b_pdata but the user may have disabled compressed
* arc thus the on-disk block may or may not match what we maintain
* in the hdr's b_pdata field.
*/
!ARC_BUF_COMPRESSED(buf)) {
} else {
/*
* This hdr is not compressed so we're able to share
* the arc_buf_t data buffer with the hdr.
*/
HDR_GET_LSIZE(hdr)));
}
}
static void
{
}
/*
* The SPA calls this callback for each physical write that happens on behalf
* of a logical write. See the comment in dbuf_write_physdone() for details.
*/
static void
{
}
static void
{
} else {
}
} else {
}
/*
* If the block to be written was all-zero or compressed enough to be
* embedded in the BP, no write was performed so there will be no
* (and uncached).
*/
/*
* This can only happen if we overwrite for
* sync-to-convergence, because we remove
* buffers from the hash table when we arc_free().
*/
panic("bad overwrite, hdr=%p exists=%p",
/* nopwrite */
panic("bad nopwrite, hdr=%p exists=%p",
} else {
/* Dedup */
}
}
/* if it's not anon, we are doing a scrub */
} else {
}
}
zio_t *
{
if (l2arc)
if (ARC_BUF_COMPRESSED(buf)) {
}
/*
* The hdr's b_pdata is now stale, free it now. A new data block
* will be allocated when the zio pipeline calls arc_write_ready().
*/
/*
* If the buf is currently sharing the data block with
* the hdr then we need to break that relationship here.
* The hdr will remain with a NULL data pointer and the
* buf will take sole ownership of the block.
*/
if (arc_buf_is_shared(buf)) {
} else {
}
}
return (zio);
}
static int
{
#ifdef _KERNEL
#if defined(__i386)
#endif
return (0);
page_load = 0;
}
/*
* If we are in pageout, we know that memory is already tight,
* the arc is already going to be evicting, so we just want to
* continue to let page writes occur as quickly as possible.
*/
if (curproc == proc_pageout) {
/* Note: reserve is inflated, so we deflate */
return (0);
} else if (page_load > 0 && arc_reclaim_needed()) {
/* memory is low, delay before restarting */
}
page_load = 0;
#endif
return (0);
}
void
{
}
int
{
int error;
/*
* Don't count loaned bufs as in flight dirty data to prevent long
* network delays from blocking transactions that are ready to be
* assigned to a txg.
*/
/* assert that it has not wrapped around */
arc_loaned_bytes), 0);
/*
* Writes will, almost always, require additional memory allocations
* make sure that there is sufficient available memory for this.
*/
if (error != 0)
return (error);
/*
* Throttle writes when the amount of dirty data in the cache
* gets too large. We try to keep the cache less than half full
* of dirty blocks so that our sync times don't grow too large.
* Note: if two requests come in concurrently, we might let them
* both succeed, when one of them should fail. Not a huge deal.
*/
dprintf("failing, arc_tempreserve=%lluK anon_meta=%lluK "
"anon_data=%lluK tempreserve=%lluK arc_c=%lluK\n",
}
return (0);
}
static void
{
}
static int
{
if (rw == KSTAT_WRITE) {
return (EACCES);
} else {
}
return (0);
}
/*
* This function *must* return indices evenly distributed between all
* sublists of the multilist. This is needed due to how the ARC eviction
* code is laid out; arc_evict_state() assumes ARC buffers are evenly
* distributed between all sublists and uses this assumption when
* deciding which sublist to evict from and how much to evict from it.
*/
unsigned int
{
/*
* We rely on b_dva to generate evenly distributed index
* numbers using buf_hash below. So, as an added precaution,
* let's make sure we never add empty buffers to the arc lists.
*/
/*
* The assumption here, is the hash value for a given
* arc_buf_hdr_t will remain constant throughout it's lifetime
* (i.e. it's b_spa, b_dva, and b_birth fields don't change).
* Thus, we don't need to store the header's sublist index
* on insertion, as this index can be recalculated on removal.
*
* Also, the low order bits of the hash value are thought to be
* distributed evenly. Otherwise, in the case that the multilist
* has a power of two number of sublists, each sublists' usage
* would not be evenly distributed.
*/
}
static void
arc_state_init(void)
{
sizeof (arc_buf_hdr_t),
sizeof (arc_buf_hdr_t),
sizeof (arc_buf_hdr_t),
sizeof (arc_buf_hdr_t),
sizeof (arc_buf_hdr_t),
sizeof (arc_buf_hdr_t),
sizeof (arc_buf_hdr_t),
sizeof (arc_buf_hdr_t),
sizeof (arc_buf_hdr_t),
sizeof (arc_buf_hdr_t),
}
static void
arc_state_fini(void)
{
}
arc_max_bytes(void)
{
return (arc_c_max);
}
void
arc_init(void)
{
/*
* allmem is "all memory that we could possibly use".
*/
#ifdef _KERNEL
#else
#endif
/* Convert seconds to clock ticks */
/* set min cache to 1/32 of all memory, or 64MB, whichever is more */
/* set max to 3/4 of all memory, or all but 1GB, whichever is more */
else
/*
* In userland, there's only the memory pressure that we artificially
* create (see arc_available_memory()). Don't let arc_c get too
* small, because it can cause transactions to be larger than
* arc_c, causing arc_tempreserve_space() to fail.
*/
#ifndef _KERNEL
#endif
/*
* Allow the tunables to override our calculations if they are
* reasonable (ie. over 64MB)
*/
}
arc_size = 0;
/* limit meta-data to 1/4 of the arc capacity */
#ifdef _KERNEL
/*
* Metadata is stored in the kernel's heap. Don't let us
* use more than half the heap for the ARC.
*/
#endif
/* Allow the tunable to override if it is reasonable */
if (zfs_arc_meta_min > 0) {
} else {
}
if (zfs_arc_grow_retry > 0)
if (zfs_arc_shrink_shift > 0)
/*
* Ensure that arc_no_grow_shift is less than arc_shrink_shift.
*/
if (arc_no_grow_shift >= arc_shrink_shift)
if (zfs_arc_p_min_shift > 0)
if (zfs_arc_num_sublists_per_state < 1)
/* if kmem_flags are set, lets try to use less memory */
if (kmem_debugging())
buf_init();
}
/*
* Calculate maximum amount of dirty data per pool.
*
* Otherwise, use a percentage of physical memory defined by
* zfs_dirty_data_max_percent (default 10%) with a cap at
* zfs_dirty_data_max_max (default 4GB).
*/
if (zfs_dirty_data_max == 0) {
zfs_dirty_data_max_percent / 100;
}
}
void
arc_fini(void)
{
/*
* The reclaim thread will set arc_reclaim_thread_exit back to
* B_FALSE when it is finished exiting; we're waiting for that.
*/
while (arc_reclaim_thread_exit) {
}
/* Use B_TRUE to ensure *all* buffers are evicted */
}
buf_fini();
}
/*
* Level 2 ARC
*
* The level 2 ARC (L2ARC) is a cache layer in-between main memory and disk.
* It uses dedicated storage devices to hold cached data, which are populated
* using large infrequent writes. The main role of this cache is to boost
* the performance of random read workloads. The intended L2ARC devices
* include short-stroked disks, solid state disks, and other media with
* substantially faster read latency than disk.
*
* +-----------------------+
* | ARC |
* +-----------------------+
* | ^ ^
* | | |
* l2arc_feed_thread() arc_read()
* | | |
* | l2arc read |
* V | |
* +---------------+ |
* | L2ARC | |
* +---------------+ |
* | ^ |
* l2arc_write() | |
* | | |
* V | |
* +-------+ +-------+
* | vdev | | vdev |
* | cache | | cache |
* +-------+ +-------+
* +=========+ .-----.
* : L2ARC : |-_____-|
* : devices : | Disks |
* +=========+ `-_____-'
*
* Read requests are satisfied from the following sources, in order:
*
* 1) ARC
* 2) vdev cache of L2ARC devices
* 3) L2ARC devices
* 4) vdev cache of disks
* 5) disks
*
* Some L2ARC device types exhibit extremely slow write performance.
* To accommodate for this there are some significant differences between
* the L2ARC and traditional cache design:
*
* 1. There is no eviction path from the ARC to the L2ARC. Evictions from
* the ARC behave as usual, freeing buffers and placing headers on ghost
* lists. The ARC does not send buffers to the L2ARC during eviction as
* this would add inflated write latencies for all ARC memory pressure.
*
* 2. The L2ARC attempts to cache data from the ARC before it is evicted.
* It does this by periodically scanning buffers from the eviction-end of
* the MFU and MRU ARC lists, copying them to the L2ARC devices if they are
* not already there. It scans until a headroom of buffers is satisfied,
* which itself is a buffer for ARC eviction. If a compressible buffer is
* found during scanning and selected for writing to an L2ARC device, we
* temporarily boost scanning headroom during the next scan cycle to make
* sure we adapt to compression effects (which might significantly reduce
* the data volume we write to L2ARC). The thread that does this is
* l2arc_feed_thread(), illustrated below; example sizes are included to
* provide a better sense of ratio than this diagram:
*
* head --> tail
* +---------------------+----------+
* ARC_mfu |:::::#:::::::::::::::|o#o###o###|-->. # already on L2ARC
* +---------------------+----------+ | o L2ARC eligible
* ARC_mru |:#:::::::::::::::::::|#o#ooo####|-->| : ARC buffer
* +---------------------+----------+ |
* 15.9 Gbytes ^ 32 Mbytes |
* headroom |
* l2arc_feed_thread()
* |
* l2arc write hand <--[oooo]--'
* | 8 Mbyte
* | write max
* V
* +==============================+
* L2ARC dev |####|#|###|###| |####| ... |
* +==============================+
* 32 Gbytes
*
* 3. If an ARC buffer is copied to the L2ARC but then hit instead of
* evicted, then the L2ARC has cached a buffer much sooner than it probably
* needed to, potentially wasting L2ARC device bandwidth and storage. It is
* safe to say that this is an uncommon case, since buffers at the end of
* the ARC lists have moved there due to inactivity.
*
* 4. If the ARC evicts faster than the L2ARC can maintain a headroom,
* then the L2ARC simply misses copying some buffers. This serves as a
* pressure valve to prevent heavy read workloads from both stalling the ARC
* with waits and clogging the L2ARC with writes. This also helps prevent
* the potential for the L2ARC to churn if it attempts to cache content too
* quickly, such as during backups of the entire pool.
*
* 5. After system boot and before the ARC has filled main memory, there are
* no evictions from the ARC and so the tails of the ARC_mfu and ARC_mru
* lists can remain mostly static. Instead of searching from tail of these
* lists as pictured, the l2arc_feed_thread() will search from the list heads
* for eligible buffers, greatly increasing its chance of finding them.
*
* The L2ARC device write speed is also boosted during this time so that
* the L2ARC warms up faster. Since there have been no ARC evictions yet,
* there are no L2ARC reads, and no fear of degrading read performance
* through increased writes.
*
* 6. Writes to the L2ARC devices are grouped and sent in-sequence, so that
* the vdev queue can aggregate them into larger and fewer writes. Each
* device is written to in a rotor fashion, sweeping writes through
* available space then repeating.
*
* 7. The L2ARC does not store dirty content. It never needs to flush
* write buffers back to disk based storage.
*
* 8. If an ARC buffer is written (and dirtied) which also exists in the
* L2ARC, the now stale L2ARC buffer is immediately dropped.
*
* The performance of the L2ARC can be tweaked by a number of tunables, which
* may be necessary for different workloads:
*
* l2arc_write_max max write bytes per interval
* l2arc_write_boost extra write bytes during device warmup
* l2arc_noprefetch skip caching prefetched buffers
* l2arc_headroom number of max device writes to precache
* l2arc_headroom_boost when we find compressed buffers during ARC
* scanning, we multiply headroom by this
* percentage factor for the next scan cycle,
* since more compressed buffers are likely to
* be present
* l2arc_feed_secs seconds between L2ARC writing
*
* Tunables may be removed or added as future performance improvements are
* integrated, and also may become zpool properties.
*
* There are three key functions that control how the L2ARC warms up:
*
* l2arc_write_eligible() check if a buffer is eligible to cache
* l2arc_write_size() calculate how much to write
* l2arc_write_interval() calculate sleep delay between writes
*
* These three functions determine what to write, how much, and how quickly
* to send writes.
*/
static boolean_t
{
/*
* A buffer is *not* eligible for the L2ARC if it:
* 1. belongs to a different spa.
* 2. is already cached on the L2ARC.
* 3. has an I/O in progress (it may be an incomplete read).
* 4. is flagged not eligible (zfs property).
*/
return (B_FALSE);
return (B_TRUE);
}
static uint64_t
l2arc_write_size(void)
{
/*
* Make sure our globals have meaningful values in case the user
* altered them.
*/
if (size == 0) {
"be greater than zero, resetting it to the default (%d)",
}
return (size);
}
static clock_t
{
/*
* If the ARC lists are busy, increase our write rate; if the
* lists are stale, idle back. This is achieved by checking
* how much we previously wrote - if it was more than half of
* what we wanted, schedule the next write much sooner.
*/
else
now = ddi_get_lbolt();
return (next);
}
/*
* Cycle through L2ARC devices. This is how L2ARC load balances.
* If a device is returned, this also returns holding the spa config lock.
*/
static l2arc_dev_t *
l2arc_dev_get_next(void)
{
/*
* Lock out the removal of spas (spa_namespace_lock), then removal
* of cache devices (l2arc_dev_mtx). Once a device has been selected,
* both locks will be dropped and a spa config lock held instead.
*/
/* if there are no vdevs, there is nothing to do */
if (l2arc_ndev == 0)
goto out;
do {
/* loop around the list looking for a non-faulted vdev */
} else {
}
/* if we have come back to the start, bail out */
break;
/* if we were unable to find any usable vdevs, return NULL */
out:
/*
* Grab the config lock to prevent the 'next' device from being
* removed while we are writing to it.
*/
return (next);
}
/*
* Free buffers that were tagged for destruction.
*/
static void
{
} else {
}
}
}
/*
* A write to a cache device has completed. Update all headers to allow
* reads from these buffers to begin.
*/
static void
{
l2arc_write_callback_t *, cb);
/*
* All writes completed, or an error was hit.
*/
top:
/*
* We cannot use mutex_enter or else we can deadlock
* with l2arc_write_buffers (due to swapping the order
* the hash lock and l2ad_mtx are taken).
*/
if (!mutex_tryenter(hash_lock)) {
/*
* Missed the hash lock. We must retry so we
* don't leave the ARC_FLAG_L2_WRITING bit set.
*/
/*
* We don't want to rescan the headers we've
* already marked as having been written out, so
* we reinsert the head node so we can pick up
* where we left off.
*/
/*
* We wait for the hash lock to become available
* to try and prevent busy waiting, and increase
* the chance we'll be able to acquire the lock
* the next time around.
*/
goto top;
}
/*
* We could not have been moved into the arc_l2c_only
* state while in-flight due to our ARC_FLAG_L2_WRITING
* bit being set. Let's just ensure that's being enforced.
*/
/*
* Error - drop L2ARC entry.
*/
}
/*
* Allow ARC to begin reads and ghost list evictions to
* this L2ARC entry.
*/
}
}
/*
* A read to a cache device completed. Validate buffer contents before
* handing over to the regular ARC routines.
*/
static void
{
/*
* Check this survived the L2ARC journey.
*/
} else {
/*
* Buffer didn't survive caching. Increment stats and
* reissue to the original storage device.
*/
} else {
}
if (!valid_cksum)
/*
* If there's no waiter, issue an async i/o to the primary
* storage now. If there *is* a waiter, the caller must
* issue the i/o in a context where it's OK to block.
*/
}
}
}
/*
* This is the list priority from which the L2ARC will search for pages to
* cache. This is used within loops (0..3) to cycle through lists in the
* desired order. This order can have a significant effect on cache
* performance.
*
* Currently the metadata lists are hit first, MFU then MRU, followed by
* the data lists. This function returns a locked list, and also returns
* the lock pointer.
*/
static multilist_sublist_t *
{
unsigned int idx;
switch (list_num) {
case 0:
break;
case 1:
break;
case 2:
break;
case 3:
break;
}
/*
* Return a randomly-selected sublist. This is acceptable
* because the caller feeds only a little bit of data for each
* call (8MB). Subsequent calls will result in different
* sublists being selected.
*/
}
/*
* Evict buffers from the device write hand to the distance specified in
* bytes. This distance may span populated buffers, it may span nothing.
* This is clearing a region on the L2ARC device ready for writing.
* If the 'all' boolean is set, every buffer is evicted.
*/
static void
{
/*
* This is the first sweep through the device. There is
* nothing to evict.
*/
return;
}
/*
* When nearing the end of the device, evict to the end
* before the device write hand jumps to the start.
*/
} else {
}
top:
/*
* We cannot use mutex_enter or else we can deadlock
* with l2arc_write_buffers (due to swapping the order
* the hash lock and l2ad_mtx are taken).
*/
if (!mutex_tryenter(hash_lock)) {
/*
* Missed the hash lock. Retry.
*/
goto top;
}
if (HDR_L2_WRITE_HEAD(hdr)) {
/*
* We hit a write head node. Leave it for
* l2arc_write_done().
*/
continue;
}
/*
* We've evicted to the target address,
* or the end of the device.
*/
break;
}
if (!HDR_HAS_L1HDR(hdr)) {
/*
* This doesn't exist in the ARC. Destroy.
* arc_hdr_destroy() will call list_remove()
* and decrement arcstat_l2_size.
*/
} else {
/*
* Invalidate issued or about to be issued
* reads, since we may be about to write
* over this location.
*/
if (HDR_L2_READING(hdr)) {
}
/* Ensure this header has finished being written */
}
}
}
/*
* Find and write ARC buffers to the L2ARC device.
*
* An ARC_FLAG_L2_WRITING flag is set so that the L2ARC buffers are not valid
* for reading until they have completed writing.
* The headroom_boost is an in-out parameter used to maintain headroom boost
* state between calls to this function.
*
* Returns the number of bytes actually written (which may be smaller than
* the delta by which the device hand has changed due to alignment).
*/
static uint64_t
{
/*
* Copy buffers for L2ARC writing.
*/
/*
* L2ARC fast warmup.
*
* Until the ARC is warm and starts to evict, read from the
* head of the ARC lists rather than the tail.
*/
else
else
if (!mutex_tryenter(hash_lock)) {
/*
* Skip this buffer rather than waiting.
*/
continue;
}
/*
* Searched too far.
*/
break;
}
continue;
}
break;
}
/*
* Insert a dummy header on the buflist so
* l2arc_write_done() can find where the
* write buffers begin without searching.
*/
cb = kmem_alloc(
sizeof (l2arc_write_callback_t), KM_SLEEP);
}
/*
* We rely on the L1 portion of the header below, so
* it's invalid for this header to have been evicted out
* of the ghost cache, prior to being written out. The
* ARC_FLAG_L2_WRITING bit ensures this won't happen.
*/
/*
* Normally the L2ARC can use the hdr's data, but if
* we're sharing data between the hdr and one of its
* bufs, L2ARC needs its own copy of the data so that
* the ZIO below can't race with the buf consumer. To
* ensure that this copy will be available for the
* lifetime of the ZIO and be cleaned up afterwards, we
* add it to the l2arc_free_on_write queue.
*/
void *to_write;
if (!HDR_SHARED_DATA(hdr)) {
} else {
if (type == ARC_BUFC_METADATA) {
} else {
}
}
write_asize += size;
/*
* Keep the clock hand suitably device-aligned.
*/
size);
write_psize += asize;
(void) zio_nowait(wzio);
}
break;
}
/* No buffers selected for writing? */
return (0);
}
/*
* Bump device hand to the device start if it is approaching the end.
* l2arc_evict() will already have evicted ahead for this case.
*/
}
return (write_asize);
}
/*
* This thread feeds the L2ARC at regular intervals. This is the beating
* heart of the L2ARC.
*/
static void
l2arc_feed_thread(void)
{
while (l2arc_thread_exit == 0) {
next);
/*
* Quick check for L2ARC devices.
*/
if (l2arc_ndev == 0) {
continue;
}
begin = ddi_get_lbolt();
/*
* This selects the next l2arc device to write to, and in
* doing so the next spa to feed from: dev->l2ad_spa. This
* will return NULL if there are now no l2arc devices or if
* they are all faulted.
*
* If a device is returned, its spa's config lock is also
* held to prevent device removal. l2arc_dev_get_next()
* will grab and release l2arc_dev_mtx.
*/
continue;
/*
* If the pool is read-only then force the feed thread to
* sleep a little longer.
*/
if (!spa_writeable(spa)) {
continue;
}
/*
* Avoid contributing to memory pressure.
*/
if (arc_reclaim_needed()) {
continue;
}
size = l2arc_write_size();
/*
* Evict L2ARC buffers that will be overwritten.
*/
/*
* Write ARC buffers.
*/
/*
* Calculate interval between writes.
*/
}
l2arc_thread_exit = 0;
thread_exit();
}
{
break;
}
}
/*
* Add a vdev for use by the L2ARC. By this point the spa has already
* validated the vdev and opened it.
*/
void
{
/*
* Create a new l2arc device entry.
*/
/*
* This is a list of all ARC buffers that are still valid on the
* device.
*/
/*
* Add device to global list
*/
}
/*
* Remove a vdev from the L2ARC.
*/
void
{
/*
* Find the device by vdev
*/
break;
}
}
/*
* Remove device from global list
*/
/*
* Clear all buflists and ARC references. L2ARC device flush.
*/
}
void
l2arc_init(void)
{
l2arc_thread_exit = 0;
l2arc_ndev = 0;
l2arc_writes_sent = 0;
l2arc_writes_done = 0;
}
void
l2arc_fini(void)
{
/*
* This is called from dmu_fini(), which is called from spa_fini();
* Because of this, we can assume that all l2arc devices have
* already been removed when the pools themselves were removed.
*/
}
void
l2arc_start(void)
{
if (!(spa_mode_global & FWRITE))
return;
}
void
l2arc_stop(void)
{
if (!(spa_mode_global & FWRITE))
return;
l2arc_thread_exit = 1;
while (l2arc_thread_exit != 0)
}