Commit ffa47032 authored by Coly Li's avatar Coly Li Committed by Jens Axboe

bcache: add bucket_size_hi into struct cache_sb_disk for large bucket

The large bucket feature is to extend bucket_size from 16bit to 32bit.

When create cache device on zoned device (e.g. zoned NVMe SSD), making
a single bucket cover one or more zones of the zoned device is the
simplest way to support zoned device as cache by bcache.

But current maximum bucket size is 16MB and a typical zone size of zoned
device is 256MB, this is the major motiviation to extend bucket size to
a larger bit width.

This patch is the basic and first change to support large bucket size,
the major changes it makes are,
- Add BCH_FEATURE_INCOMPAT_LARGE_BUCKET for the large bucket feature,
  INCOMPAT means it introduces incompatible on-disk format change.
- Add BCH_FEATURE_INCOMPAT_FUNCS(large_bucket, LARGE_BUCKET) routines.
- Adds __le16 bucket_size_hi into struct cache_sb_disk at offset 0x8d0
  for the on-disk super block format.
- For the in-memory super block struct cache_sb, member bucket_size is
  extended from __u16 to __32.
- Add get_bucket_size() to combine the bucket_size and bucket_size_hi
  from struct cache_sb_disk into an unsigned int value.

Since we already have large bucket size helpers meta_bucket_pages(),
meta_bucket_bytes() and alloc_meta_bucket_pages(), they make sure when
bucket size > 8MB, the memory allocation for bcache meta data bucket
won't fail no matter how large the bucket size extended. So these meta
data buckets are handled properly when the bucket size width increase
from 16bit to 32bit, we don't need to worry about them.
Signed-off-by: default avatarColy Li <colyli@suse.de>
Signed-off-by: default avatarJens Axboe <axboe@kernel.dk>
parent f9c32a5a
...@@ -87,7 +87,7 @@ void bch_rescale_priorities(struct cache_set *c, int sectors) ...@@ -87,7 +87,7 @@ void bch_rescale_priorities(struct cache_set *c, int sectors)
{ {
struct cache *ca; struct cache *ca;
struct bucket *b; struct bucket *b;
unsigned int next = c->nbuckets * c->sb.bucket_size / 1024; unsigned long next = c->nbuckets * c->sb.bucket_size / 1024;
unsigned int i; unsigned int i;
int r; int r;
......
// SPDX-License-Identifier: GPL-2.0
/*
* Feature set bits and string conversion.
* Inspired by ext4's features compat/incompat/ro_compat related code.
*
* Copyright 2020 Coly Li <colyli@suse.de>
*
*/
#include <linux/bcache.h>
#include "bcache.h"
struct feature {
int compat;
unsigned int mask;
const char *string;
};
static struct feature feature_list[] = {
{BCH_FEATURE_INCOMPAT, BCH_FEATURE_INCOMPAT_LARGE_BUCKET,
"large_bucket"},
{0, 0, 0 },
};
...@@ -11,9 +11,13 @@ ...@@ -11,9 +11,13 @@
#define BCH_FEATURE_INCOMPAT 2 #define BCH_FEATURE_INCOMPAT 2
#define BCH_FEATURE_TYPE_MASK 0x03 #define BCH_FEATURE_TYPE_MASK 0x03
/* Feature set definition */
/* Incompat feature set */
#define BCH_FEATURE_INCOMPAT_LARGE_BUCKET 0x0001 /* 32bit bucket size */
#define BCH_FEATURE_COMPAT_SUUP 0 #define BCH_FEATURE_COMPAT_SUUP 0
#define BCH_FEATURE_RO_COMPAT_SUUP 0 #define BCH_FEATURE_RO_COMPAT_SUUP 0
#define BCH_FEATURE_INCOMPAT_SUUP 0 #define BCH_FEATURE_INCOMPAT_SUUP BCH_FEATURE_INCOMPAT_LARGE_BUCKET
#define BCH_HAS_COMPAT_FEATURE(sb, mask) \ #define BCH_HAS_COMPAT_FEATURE(sb, mask) \
((sb)->feature_compat & (mask)) ((sb)->feature_compat & (mask))
...@@ -22,8 +26,6 @@ ...@@ -22,8 +26,6 @@
#define BCH_HAS_INCOMPAT_FEATURE(sb, mask) \ #define BCH_HAS_INCOMPAT_FEATURE(sb, mask) \
((sb)->feature_incompat & (mask)) ((sb)->feature_incompat & (mask))
/* Feature set definition */
#define BCH_FEATURE_COMPAT_FUNCS(name, flagname) \ #define BCH_FEATURE_COMPAT_FUNCS(name, flagname) \
static inline int bch_has_feature_##name(struct cache_sb *sb) \ static inline int bch_has_feature_##name(struct cache_sb *sb) \
{ \ { \
...@@ -75,4 +77,5 @@ static inline void bch_clear_feature_##name(struct cache_sb *sb) \ ...@@ -75,4 +77,5 @@ static inline void bch_clear_feature_##name(struct cache_sb *sb) \
~BCH##_FEATURE_INCOMPAT_##flagname; \ ~BCH##_FEATURE_INCOMPAT_##flagname; \
} }
BCH_FEATURE_INCOMPAT_FUNCS(large_bucket, LARGE_BUCKET);
#endif #endif
...@@ -206,8 +206,8 @@ void bch_moving_gc(struct cache_set *c) ...@@ -206,8 +206,8 @@ void bch_moving_gc(struct cache_set *c)
mutex_lock(&c->bucket_lock); mutex_lock(&c->bucket_lock);
for_each_cache(ca, c, i) { for_each_cache(ca, c, i) {
unsigned int sectors_to_move = 0; unsigned long sectors_to_move = 0;
unsigned int reserve_sectors = ca->sb.bucket_size * unsigned long reserve_sectors = ca->sb.bucket_size *
fifo_used(&ca->free[RESERVE_MOVINGGC]); fifo_used(&ca->free[RESERVE_MOVINGGC]);
ca->heap.used = 0; ca->heap.used = 0;
......
...@@ -60,6 +60,17 @@ struct workqueue_struct *bch_journal_wq; ...@@ -60,6 +60,17 @@ struct workqueue_struct *bch_journal_wq;
/* Superblock */ /* Superblock */
static unsigned int get_bucket_size(struct cache_sb *sb, struct cache_sb_disk *s)
{
unsigned int bucket_size = le16_to_cpu(s->bucket_size);
if (sb->version >= BCACHE_SB_VERSION_CDEV_WITH_FEATURES &&
bch_has_feature_large_bucket(sb))
bucket_size |= le16_to_cpu(s->bucket_size_hi) << 16;
return bucket_size;
}
static const char *read_super_common(struct cache_sb *sb, struct block_device *bdev, static const char *read_super_common(struct cache_sb *sb, struct block_device *bdev,
struct cache_sb_disk *s) struct cache_sb_disk *s)
{ {
...@@ -68,7 +79,7 @@ static const char *read_super_common(struct cache_sb *sb, struct block_device * ...@@ -68,7 +79,7 @@ static const char *read_super_common(struct cache_sb *sb, struct block_device *
sb->first_bucket= le16_to_cpu(s->first_bucket); sb->first_bucket= le16_to_cpu(s->first_bucket);
sb->nbuckets = le64_to_cpu(s->nbuckets); sb->nbuckets = le64_to_cpu(s->nbuckets);
sb->bucket_size = le16_to_cpu(s->bucket_size); sb->bucket_size = get_bucket_size(sb, s);
sb->nr_in_set = le16_to_cpu(s->nr_in_set); sb->nr_in_set = le16_to_cpu(s->nr_in_set);
sb->nr_this_dev = le16_to_cpu(s->nr_this_dev); sb->nr_this_dev = le16_to_cpu(s->nr_this_dev);
...@@ -210,12 +221,16 @@ static const char *read_super(struct cache_sb *sb, struct block_device *bdev, ...@@ -210,12 +221,16 @@ static const char *read_super(struct cache_sb *sb, struct block_device *bdev,
goto err; goto err;
break; break;
case BCACHE_SB_VERSION_CDEV_WITH_FEATURES: case BCACHE_SB_VERSION_CDEV_WITH_FEATURES:
err = read_super_common(sb, bdev, s); /*
if (err) * Feature bits are needed in read_super_common(),
goto err; * convert them firstly.
*/
sb->feature_compat = le64_to_cpu(s->feature_compat); sb->feature_compat = le64_to_cpu(s->feature_compat);
sb->feature_incompat = le64_to_cpu(s->feature_incompat); sb->feature_incompat = le64_to_cpu(s->feature_incompat);
sb->feature_ro_compat = le64_to_cpu(s->feature_ro_compat); sb->feature_ro_compat = le64_to_cpu(s->feature_ro_compat);
err = read_super_common(sb, bdev, s);
if (err)
goto err;
break; break;
default: default:
err = "Unsupported superblock version"; err = "Unsupported superblock version";
......
...@@ -213,6 +213,7 @@ struct cache_sb_disk { ...@@ -213,6 +213,7 @@ struct cache_sb_disk {
__le16 keys; __le16 keys;
}; };
__le64 d[SB_JOURNAL_BUCKETS]; /* journal buckets */ __le64 d[SB_JOURNAL_BUCKETS]; /* journal buckets */
__le16 bucket_size_hi;
}; };
/* /*
...@@ -247,9 +248,9 @@ struct cache_sb { ...@@ -247,9 +248,9 @@ struct cache_sb {
__u64 nbuckets; /* device size */ __u64 nbuckets; /* device size */
__u16 block_size; /* sectors */ __u16 block_size; /* sectors */
__u16 bucket_size; /* sectors */
__u16 nr_in_set; __u16 nr_in_set;
__u16 nr_this_dev; __u16 nr_this_dev;
__u32 bucket_size; /* sectors */
}; };
struct { struct {
/* Backing devices */ /* Backing devices */
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment