mirror of
https://github.com/torvalds/linux.git
synced 2024-11-23 04:31:50 +00:00
ec4edd7b9d
bcachefs btree nodes are big - typically 256k - and btree roots are pinned in memory. As we're now up to 18 btrees, we now have significant memory overhead in mostly empty btree roots. And in the future we're going to start enforcing that certain btree node boundaries exist, to solve lock contention issues - analagous to XFS's AGIs. Thus, we need to start allocating smaller btree node buffers when we can. This patch changes code that refers to the filesystem constant c->opts.btree_node_size to refer to the btree node buffer size - btree_buf_bytes() - where appropriate. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
1598 lines
40 KiB
C
1598 lines
40 KiB
C
// SPDX-License-Identifier: GPL-2.0
|
|
/*
|
|
* Code for working with individual keys, and sorted sets of keys with in a
|
|
* btree node
|
|
*
|
|
* Copyright 2012 Google, Inc.
|
|
*/
|
|
|
|
#include "bcachefs.h"
|
|
#include "btree_cache.h"
|
|
#include "bset.h"
|
|
#include "eytzinger.h"
|
|
#include "trace.h"
|
|
#include "util.h"
|
|
|
|
#include <asm/unaligned.h>
|
|
#include <linux/console.h>
|
|
#include <linux/random.h>
|
|
#include <linux/prefetch.h>
|
|
|
|
static inline void __bch2_btree_node_iter_advance(struct btree_node_iter *,
|
|
struct btree *);
|
|
|
|
static inline unsigned __btree_node_iter_used(struct btree_node_iter *iter)
|
|
{
|
|
unsigned n = ARRAY_SIZE(iter->data);
|
|
|
|
while (n && __btree_node_iter_set_end(iter, n - 1))
|
|
--n;
|
|
|
|
return n;
|
|
}
|
|
|
|
struct bset_tree *bch2_bkey_to_bset(struct btree *b, struct bkey_packed *k)
|
|
{
|
|
return bch2_bkey_to_bset_inlined(b, k);
|
|
}
|
|
|
|
/*
|
|
* There are never duplicate live keys in the btree - but including keys that
|
|
* have been flagged as deleted (and will be cleaned up later) we _will_ see
|
|
* duplicates.
|
|
*
|
|
* Thus the sort order is: usual key comparison first, but for keys that compare
|
|
* equal the deleted key(s) come first, and the (at most one) live version comes
|
|
* last.
|
|
*
|
|
* The main reason for this is insertion: to handle overwrites, we first iterate
|
|
* over keys that compare equal to our insert key, and then insert immediately
|
|
* prior to the first key greater than the key we're inserting - our insert
|
|
* position will be after all keys that compare equal to our insert key, which
|
|
* by the time we actually do the insert will all be deleted.
|
|
*/
|
|
|
|
void bch2_dump_bset(struct bch_fs *c, struct btree *b,
|
|
struct bset *i, unsigned set)
|
|
{
|
|
struct bkey_packed *_k, *_n;
|
|
struct bkey uk, n;
|
|
struct bkey_s_c k;
|
|
struct printbuf buf = PRINTBUF;
|
|
|
|
if (!i->u64s)
|
|
return;
|
|
|
|
for (_k = i->start;
|
|
_k < vstruct_last(i);
|
|
_k = _n) {
|
|
_n = bkey_p_next(_k);
|
|
|
|
if (!_k->u64s) {
|
|
printk(KERN_ERR "block %u key %5zu - u64s 0? aieee!\n", set,
|
|
_k->_data - i->_data);
|
|
break;
|
|
}
|
|
|
|
k = bkey_disassemble(b, _k, &uk);
|
|
|
|
printbuf_reset(&buf);
|
|
if (c)
|
|
bch2_bkey_val_to_text(&buf, c, k);
|
|
else
|
|
bch2_bkey_to_text(&buf, k.k);
|
|
printk(KERN_ERR "block %u key %5zu: %s\n", set,
|
|
_k->_data - i->_data, buf.buf);
|
|
|
|
if (_n == vstruct_last(i))
|
|
continue;
|
|
|
|
n = bkey_unpack_key(b, _n);
|
|
|
|
if (bpos_lt(n.p, k.k->p)) {
|
|
printk(KERN_ERR "Key skipped backwards\n");
|
|
continue;
|
|
}
|
|
|
|
if (!bkey_deleted(k.k) && bpos_eq(n.p, k.k->p))
|
|
printk(KERN_ERR "Duplicate keys\n");
|
|
}
|
|
|
|
printbuf_exit(&buf);
|
|
}
|
|
|
|
void bch2_dump_btree_node(struct bch_fs *c, struct btree *b)
|
|
{
|
|
struct bset_tree *t;
|
|
|
|
console_lock();
|
|
for_each_bset(b, t)
|
|
bch2_dump_bset(c, b, bset(b, t), t - b->set);
|
|
console_unlock();
|
|
}
|
|
|
|
void bch2_dump_btree_node_iter(struct btree *b,
|
|
struct btree_node_iter *iter)
|
|
{
|
|
struct btree_node_iter_set *set;
|
|
struct printbuf buf = PRINTBUF;
|
|
|
|
printk(KERN_ERR "btree node iter with %u/%u sets:\n",
|
|
__btree_node_iter_used(iter), b->nsets);
|
|
|
|
btree_node_iter_for_each(iter, set) {
|
|
struct bkey_packed *k = __btree_node_offset_to_key(b, set->k);
|
|
struct bset_tree *t = bch2_bkey_to_bset(b, k);
|
|
struct bkey uk = bkey_unpack_key(b, k);
|
|
|
|
printbuf_reset(&buf);
|
|
bch2_bkey_to_text(&buf, &uk);
|
|
printk(KERN_ERR "set %zu key %u: %s\n",
|
|
t - b->set, set->k, buf.buf);
|
|
}
|
|
|
|
printbuf_exit(&buf);
|
|
}
|
|
|
|
#ifdef CONFIG_BCACHEFS_DEBUG
|
|
|
|
void __bch2_verify_btree_nr_keys(struct btree *b)
|
|
{
|
|
struct bset_tree *t;
|
|
struct bkey_packed *k;
|
|
struct btree_nr_keys nr = { 0 };
|
|
|
|
for_each_bset(b, t)
|
|
bset_tree_for_each_key(b, t, k)
|
|
if (!bkey_deleted(k))
|
|
btree_keys_account_key_add(&nr, t - b->set, k);
|
|
|
|
BUG_ON(memcmp(&nr, &b->nr, sizeof(nr)));
|
|
}
|
|
|
|
static void bch2_btree_node_iter_next_check(struct btree_node_iter *_iter,
|
|
struct btree *b)
|
|
{
|
|
struct btree_node_iter iter = *_iter;
|
|
const struct bkey_packed *k, *n;
|
|
|
|
k = bch2_btree_node_iter_peek_all(&iter, b);
|
|
__bch2_btree_node_iter_advance(&iter, b);
|
|
n = bch2_btree_node_iter_peek_all(&iter, b);
|
|
|
|
bkey_unpack_key(b, k);
|
|
|
|
if (n &&
|
|
bkey_iter_cmp(b, k, n) > 0) {
|
|
struct btree_node_iter_set *set;
|
|
struct bkey ku = bkey_unpack_key(b, k);
|
|
struct bkey nu = bkey_unpack_key(b, n);
|
|
struct printbuf buf1 = PRINTBUF;
|
|
struct printbuf buf2 = PRINTBUF;
|
|
|
|
bch2_dump_btree_node(NULL, b);
|
|
bch2_bkey_to_text(&buf1, &ku);
|
|
bch2_bkey_to_text(&buf2, &nu);
|
|
printk(KERN_ERR "out of order/overlapping:\n%s\n%s\n",
|
|
buf1.buf, buf2.buf);
|
|
printk(KERN_ERR "iter was:");
|
|
|
|
btree_node_iter_for_each(_iter, set) {
|
|
struct bkey_packed *k2 = __btree_node_offset_to_key(b, set->k);
|
|
struct bset_tree *t = bch2_bkey_to_bset(b, k2);
|
|
printk(" [%zi %zi]", t - b->set,
|
|
k2->_data - bset(b, t)->_data);
|
|
}
|
|
panic("\n");
|
|
}
|
|
}
|
|
|
|
void bch2_btree_node_iter_verify(struct btree_node_iter *iter,
|
|
struct btree *b)
|
|
{
|
|
struct btree_node_iter_set *set, *s2;
|
|
struct bkey_packed *k, *p;
|
|
struct bset_tree *t;
|
|
|
|
if (bch2_btree_node_iter_end(iter))
|
|
return;
|
|
|
|
/* Verify no duplicates: */
|
|
btree_node_iter_for_each(iter, set) {
|
|
BUG_ON(set->k > set->end);
|
|
btree_node_iter_for_each(iter, s2)
|
|
BUG_ON(set != s2 && set->end == s2->end);
|
|
}
|
|
|
|
/* Verify that set->end is correct: */
|
|
btree_node_iter_for_each(iter, set) {
|
|
for_each_bset(b, t)
|
|
if (set->end == t->end_offset)
|
|
goto found;
|
|
BUG();
|
|
found:
|
|
BUG_ON(set->k < btree_bkey_first_offset(t) ||
|
|
set->k >= t->end_offset);
|
|
}
|
|
|
|
/* Verify iterator is sorted: */
|
|
btree_node_iter_for_each(iter, set)
|
|
BUG_ON(set != iter->data &&
|
|
btree_node_iter_cmp(b, set[-1], set[0]) > 0);
|
|
|
|
k = bch2_btree_node_iter_peek_all(iter, b);
|
|
|
|
for_each_bset(b, t) {
|
|
if (iter->data[0].end == t->end_offset)
|
|
continue;
|
|
|
|
p = bch2_bkey_prev_all(b, t,
|
|
bch2_btree_node_iter_bset_pos(iter, b, t));
|
|
|
|
BUG_ON(p && bkey_iter_cmp(b, k, p) < 0);
|
|
}
|
|
}
|
|
|
|
void bch2_verify_insert_pos(struct btree *b, struct bkey_packed *where,
|
|
struct bkey_packed *insert, unsigned clobber_u64s)
|
|
{
|
|
struct bset_tree *t = bch2_bkey_to_bset(b, where);
|
|
struct bkey_packed *prev = bch2_bkey_prev_all(b, t, where);
|
|
struct bkey_packed *next = (void *) ((u64 *) where->_data + clobber_u64s);
|
|
struct printbuf buf1 = PRINTBUF;
|
|
struct printbuf buf2 = PRINTBUF;
|
|
#if 0
|
|
BUG_ON(prev &&
|
|
bkey_iter_cmp(b, prev, insert) > 0);
|
|
#else
|
|
if (prev &&
|
|
bkey_iter_cmp(b, prev, insert) > 0) {
|
|
struct bkey k1 = bkey_unpack_key(b, prev);
|
|
struct bkey k2 = bkey_unpack_key(b, insert);
|
|
|
|
bch2_dump_btree_node(NULL, b);
|
|
bch2_bkey_to_text(&buf1, &k1);
|
|
bch2_bkey_to_text(&buf2, &k2);
|
|
|
|
panic("prev > insert:\n"
|
|
"prev key %s\n"
|
|
"insert key %s\n",
|
|
buf1.buf, buf2.buf);
|
|
}
|
|
#endif
|
|
#if 0
|
|
BUG_ON(next != btree_bkey_last(b, t) &&
|
|
bkey_iter_cmp(b, insert, next) > 0);
|
|
#else
|
|
if (next != btree_bkey_last(b, t) &&
|
|
bkey_iter_cmp(b, insert, next) > 0) {
|
|
struct bkey k1 = bkey_unpack_key(b, insert);
|
|
struct bkey k2 = bkey_unpack_key(b, next);
|
|
|
|
bch2_dump_btree_node(NULL, b);
|
|
bch2_bkey_to_text(&buf1, &k1);
|
|
bch2_bkey_to_text(&buf2, &k2);
|
|
|
|
panic("insert > next:\n"
|
|
"insert key %s\n"
|
|
"next key %s\n",
|
|
buf1.buf, buf2.buf);
|
|
}
|
|
#endif
|
|
}
|
|
|
|
#else
|
|
|
|
static inline void bch2_btree_node_iter_next_check(struct btree_node_iter *iter,
|
|
struct btree *b) {}
|
|
|
|
#endif
|
|
|
|
/* Auxiliary search trees */
|
|
|
|
#define BFLOAT_FAILED_UNPACKED U8_MAX
|
|
#define BFLOAT_FAILED U8_MAX
|
|
|
|
struct bkey_float {
|
|
u8 exponent;
|
|
u8 key_offset;
|
|
u16 mantissa;
|
|
};
|
|
#define BKEY_MANTISSA_BITS 16
|
|
|
|
static unsigned bkey_float_byte_offset(unsigned idx)
|
|
{
|
|
return idx * sizeof(struct bkey_float);
|
|
}
|
|
|
|
struct ro_aux_tree {
|
|
u8 nothing[0];
|
|
struct bkey_float f[];
|
|
};
|
|
|
|
struct rw_aux_tree {
|
|
u16 offset;
|
|
struct bpos k;
|
|
};
|
|
|
|
static unsigned bset_aux_tree_buf_end(const struct bset_tree *t)
|
|
{
|
|
BUG_ON(t->aux_data_offset == U16_MAX);
|
|
|
|
switch (bset_aux_tree_type(t)) {
|
|
case BSET_NO_AUX_TREE:
|
|
return t->aux_data_offset;
|
|
case BSET_RO_AUX_TREE:
|
|
return t->aux_data_offset +
|
|
DIV_ROUND_UP(t->size * sizeof(struct bkey_float) +
|
|
t->size * sizeof(u8), 8);
|
|
case BSET_RW_AUX_TREE:
|
|
return t->aux_data_offset +
|
|
DIV_ROUND_UP(sizeof(struct rw_aux_tree) * t->size, 8);
|
|
default:
|
|
BUG();
|
|
}
|
|
}
|
|
|
|
static unsigned bset_aux_tree_buf_start(const struct btree *b,
|
|
const struct bset_tree *t)
|
|
{
|
|
return t == b->set
|
|
? DIV_ROUND_UP(b->unpack_fn_len, 8)
|
|
: bset_aux_tree_buf_end(t - 1);
|
|
}
|
|
|
|
static void *__aux_tree_base(const struct btree *b,
|
|
const struct bset_tree *t)
|
|
{
|
|
return b->aux_data + t->aux_data_offset * 8;
|
|
}
|
|
|
|
static struct ro_aux_tree *ro_aux_tree_base(const struct btree *b,
|
|
const struct bset_tree *t)
|
|
{
|
|
EBUG_ON(bset_aux_tree_type(t) != BSET_RO_AUX_TREE);
|
|
|
|
return __aux_tree_base(b, t);
|
|
}
|
|
|
|
static u8 *ro_aux_tree_prev(const struct btree *b,
|
|
const struct bset_tree *t)
|
|
{
|
|
EBUG_ON(bset_aux_tree_type(t) != BSET_RO_AUX_TREE);
|
|
|
|
return __aux_tree_base(b, t) + bkey_float_byte_offset(t->size);
|
|
}
|
|
|
|
static struct bkey_float *bkey_float(const struct btree *b,
|
|
const struct bset_tree *t,
|
|
unsigned idx)
|
|
{
|
|
return ro_aux_tree_base(b, t)->f + idx;
|
|
}
|
|
|
|
static void bset_aux_tree_verify(const struct btree *b)
|
|
{
|
|
#ifdef CONFIG_BCACHEFS_DEBUG
|
|
const struct bset_tree *t;
|
|
|
|
for_each_bset(b, t) {
|
|
if (t->aux_data_offset == U16_MAX)
|
|
continue;
|
|
|
|
BUG_ON(t != b->set &&
|
|
t[-1].aux_data_offset == U16_MAX);
|
|
|
|
BUG_ON(t->aux_data_offset < bset_aux_tree_buf_start(b, t));
|
|
BUG_ON(t->aux_data_offset > btree_aux_data_u64s(b));
|
|
BUG_ON(bset_aux_tree_buf_end(t) > btree_aux_data_u64s(b));
|
|
}
|
|
#endif
|
|
}
|
|
|
|
void bch2_btree_keys_init(struct btree *b)
|
|
{
|
|
unsigned i;
|
|
|
|
b->nsets = 0;
|
|
memset(&b->nr, 0, sizeof(b->nr));
|
|
|
|
for (i = 0; i < MAX_BSETS; i++)
|
|
b->set[i].data_offset = U16_MAX;
|
|
|
|
bch2_bset_set_no_aux_tree(b, b->set);
|
|
}
|
|
|
|
/* Binary tree stuff for auxiliary search trees */
|
|
|
|
/*
|
|
* Cacheline/offset <-> bkey pointer arithmetic:
|
|
*
|
|
* t->tree is a binary search tree in an array; each node corresponds to a key
|
|
* in one cacheline in t->set (BSET_CACHELINE bytes).
|
|
*
|
|
* This means we don't have to store the full index of the key that a node in
|
|
* the binary tree points to; eytzinger1_to_inorder() gives us the cacheline, and
|
|
* then bkey_float->m gives us the offset within that cacheline, in units of 8
|
|
* bytes.
|
|
*
|
|
* cacheline_to_bkey() and friends abstract out all the pointer arithmetic to
|
|
* make this work.
|
|
*
|
|
* To construct the bfloat for an arbitrary key we need to know what the key
|
|
* immediately preceding it is: we have to check if the two keys differ in the
|
|
* bits we're going to store in bkey_float->mantissa. t->prev[j] stores the size
|
|
* of the previous key so we can walk backwards to it from t->tree[j]'s key.
|
|
*/
|
|
|
|
static inline void *bset_cacheline(const struct btree *b,
|
|
const struct bset_tree *t,
|
|
unsigned cacheline)
|
|
{
|
|
return (void *) round_down((unsigned long) btree_bkey_first(b, t),
|
|
L1_CACHE_BYTES) +
|
|
cacheline * BSET_CACHELINE;
|
|
}
|
|
|
|
static struct bkey_packed *cacheline_to_bkey(const struct btree *b,
|
|
const struct bset_tree *t,
|
|
unsigned cacheline,
|
|
unsigned offset)
|
|
{
|
|
return bset_cacheline(b, t, cacheline) + offset * 8;
|
|
}
|
|
|
|
static unsigned bkey_to_cacheline(const struct btree *b,
|
|
const struct bset_tree *t,
|
|
const struct bkey_packed *k)
|
|
{
|
|
return ((void *) k - bset_cacheline(b, t, 0)) / BSET_CACHELINE;
|
|
}
|
|
|
|
static ssize_t __bkey_to_cacheline_offset(const struct btree *b,
|
|
const struct bset_tree *t,
|
|
unsigned cacheline,
|
|
const struct bkey_packed *k)
|
|
{
|
|
return (u64 *) k - (u64 *) bset_cacheline(b, t, cacheline);
|
|
}
|
|
|
|
static unsigned bkey_to_cacheline_offset(const struct btree *b,
|
|
const struct bset_tree *t,
|
|
unsigned cacheline,
|
|
const struct bkey_packed *k)
|
|
{
|
|
size_t m = __bkey_to_cacheline_offset(b, t, cacheline, k);
|
|
|
|
EBUG_ON(m > U8_MAX);
|
|
return m;
|
|
}
|
|
|
|
static inline struct bkey_packed *tree_to_bkey(const struct btree *b,
|
|
const struct bset_tree *t,
|
|
unsigned j)
|
|
{
|
|
return cacheline_to_bkey(b, t,
|
|
__eytzinger1_to_inorder(j, t->size - 1, t->extra),
|
|
bkey_float(b, t, j)->key_offset);
|
|
}
|
|
|
|
static struct bkey_packed *tree_to_prev_bkey(const struct btree *b,
|
|
const struct bset_tree *t,
|
|
unsigned j)
|
|
{
|
|
unsigned prev_u64s = ro_aux_tree_prev(b, t)[j];
|
|
|
|
return (void *) ((u64 *) tree_to_bkey(b, t, j)->_data - prev_u64s);
|
|
}
|
|
|
|
static struct rw_aux_tree *rw_aux_tree(const struct btree *b,
|
|
const struct bset_tree *t)
|
|
{
|
|
EBUG_ON(bset_aux_tree_type(t) != BSET_RW_AUX_TREE);
|
|
|
|
return __aux_tree_base(b, t);
|
|
}
|
|
|
|
/*
|
|
* For the write set - the one we're currently inserting keys into - we don't
|
|
* maintain a full search tree, we just keep a simple lookup table in t->prev.
|
|
*/
|
|
static struct bkey_packed *rw_aux_to_bkey(const struct btree *b,
|
|
struct bset_tree *t,
|
|
unsigned j)
|
|
{
|
|
return __btree_node_offset_to_key(b, rw_aux_tree(b, t)[j].offset);
|
|
}
|
|
|
|
static void rw_aux_tree_set(const struct btree *b, struct bset_tree *t,
|
|
unsigned j, struct bkey_packed *k)
|
|
{
|
|
EBUG_ON(k >= btree_bkey_last(b, t));
|
|
|
|
rw_aux_tree(b, t)[j] = (struct rw_aux_tree) {
|
|
.offset = __btree_node_key_to_offset(b, k),
|
|
.k = bkey_unpack_pos(b, k),
|
|
};
|
|
}
|
|
|
|
static void bch2_bset_verify_rw_aux_tree(struct btree *b,
|
|
struct bset_tree *t)
|
|
{
|
|
struct bkey_packed *k = btree_bkey_first(b, t);
|
|
unsigned j = 0;
|
|
|
|
if (!bch2_expensive_debug_checks)
|
|
return;
|
|
|
|
BUG_ON(bset_has_ro_aux_tree(t));
|
|
|
|
if (!bset_has_rw_aux_tree(t))
|
|
return;
|
|
|
|
BUG_ON(t->size < 1);
|
|
BUG_ON(rw_aux_to_bkey(b, t, j) != k);
|
|
|
|
goto start;
|
|
while (1) {
|
|
if (rw_aux_to_bkey(b, t, j) == k) {
|
|
BUG_ON(!bpos_eq(rw_aux_tree(b, t)[j].k,
|
|
bkey_unpack_pos(b, k)));
|
|
start:
|
|
if (++j == t->size)
|
|
break;
|
|
|
|
BUG_ON(rw_aux_tree(b, t)[j].offset <=
|
|
rw_aux_tree(b, t)[j - 1].offset);
|
|
}
|
|
|
|
k = bkey_p_next(k);
|
|
BUG_ON(k >= btree_bkey_last(b, t));
|
|
}
|
|
}
|
|
|
|
/* returns idx of first entry >= offset: */
|
|
static unsigned rw_aux_tree_bsearch(struct btree *b,
|
|
struct bset_tree *t,
|
|
unsigned offset)
|
|
{
|
|
unsigned bset_offs = offset - btree_bkey_first_offset(t);
|
|
unsigned bset_u64s = t->end_offset - btree_bkey_first_offset(t);
|
|
unsigned idx = bset_u64s ? bset_offs * t->size / bset_u64s : 0;
|
|
|
|
EBUG_ON(bset_aux_tree_type(t) != BSET_RW_AUX_TREE);
|
|
EBUG_ON(!t->size);
|
|
EBUG_ON(idx > t->size);
|
|
|
|
while (idx < t->size &&
|
|
rw_aux_tree(b, t)[idx].offset < offset)
|
|
idx++;
|
|
|
|
while (idx &&
|
|
rw_aux_tree(b, t)[idx - 1].offset >= offset)
|
|
idx--;
|
|
|
|
EBUG_ON(idx < t->size &&
|
|
rw_aux_tree(b, t)[idx].offset < offset);
|
|
EBUG_ON(idx && rw_aux_tree(b, t)[idx - 1].offset >= offset);
|
|
EBUG_ON(idx + 1 < t->size &&
|
|
rw_aux_tree(b, t)[idx].offset ==
|
|
rw_aux_tree(b, t)[idx + 1].offset);
|
|
|
|
return idx;
|
|
}
|
|
|
|
static inline unsigned bkey_mantissa(const struct bkey_packed *k,
|
|
const struct bkey_float *f,
|
|
unsigned idx)
|
|
{
|
|
u64 v;
|
|
|
|
EBUG_ON(!bkey_packed(k));
|
|
|
|
v = get_unaligned((u64 *) (((u8 *) k->_data) + (f->exponent >> 3)));
|
|
|
|
/*
|
|
* In little endian, we're shifting off low bits (and then the bits we
|
|
* want are at the low end), in big endian we're shifting off high bits
|
|
* (and then the bits we want are at the high end, so we shift them
|
|
* back down):
|
|
*/
|
|
#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
|
|
v >>= f->exponent & 7;
|
|
#else
|
|
v >>= 64 - (f->exponent & 7) - BKEY_MANTISSA_BITS;
|
|
#endif
|
|
return (u16) v;
|
|
}
|
|
|
|
static __always_inline void make_bfloat(struct btree *b, struct bset_tree *t,
|
|
unsigned j,
|
|
struct bkey_packed *min_key,
|
|
struct bkey_packed *max_key)
|
|
{
|
|
struct bkey_float *f = bkey_float(b, t, j);
|
|
struct bkey_packed *m = tree_to_bkey(b, t, j);
|
|
struct bkey_packed *l = is_power_of_2(j)
|
|
? min_key
|
|
: tree_to_prev_bkey(b, t, j >> ffs(j));
|
|
struct bkey_packed *r = is_power_of_2(j + 1)
|
|
? max_key
|
|
: tree_to_bkey(b, t, j >> (ffz(j) + 1));
|
|
unsigned mantissa;
|
|
int shift, exponent, high_bit;
|
|
|
|
/*
|
|
* for failed bfloats, the lookup code falls back to comparing against
|
|
* the original key.
|
|
*/
|
|
|
|
if (!bkey_packed(l) || !bkey_packed(r) || !bkey_packed(m) ||
|
|
!b->nr_key_bits) {
|
|
f->exponent = BFLOAT_FAILED_UNPACKED;
|
|
return;
|
|
}
|
|
|
|
/*
|
|
* The greatest differing bit of l and r is the first bit we must
|
|
* include in the bfloat mantissa we're creating in order to do
|
|
* comparisons - that bit always becomes the high bit of
|
|
* bfloat->mantissa, and thus the exponent we're calculating here is
|
|
* the position of what will become the low bit in bfloat->mantissa:
|
|
*
|
|
* Note that this may be negative - we may be running off the low end
|
|
* of the key: we handle this later:
|
|
*/
|
|
high_bit = max(bch2_bkey_greatest_differing_bit(b, l, r),
|
|
min_t(unsigned, BKEY_MANTISSA_BITS, b->nr_key_bits) - 1);
|
|
exponent = high_bit - (BKEY_MANTISSA_BITS - 1);
|
|
|
|
/*
|
|
* Then we calculate the actual shift value, from the start of the key
|
|
* (k->_data), to get the key bits starting at exponent:
|
|
*/
|
|
#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
|
|
shift = (int) (b->format.key_u64s * 64 - b->nr_key_bits) + exponent;
|
|
|
|
EBUG_ON(shift + BKEY_MANTISSA_BITS > b->format.key_u64s * 64);
|
|
#else
|
|
shift = high_bit_offset +
|
|
b->nr_key_bits -
|
|
exponent -
|
|
BKEY_MANTISSA_BITS;
|
|
|
|
EBUG_ON(shift < KEY_PACKED_BITS_START);
|
|
#endif
|
|
EBUG_ON(shift < 0 || shift >= BFLOAT_FAILED);
|
|
|
|
f->exponent = shift;
|
|
mantissa = bkey_mantissa(m, f, j);
|
|
|
|
/*
|
|
* If we've got garbage bits, set them to all 1s - it's legal for the
|
|
* bfloat to compare larger than the original key, but not smaller:
|
|
*/
|
|
if (exponent < 0)
|
|
mantissa |= ~(~0U << -exponent);
|
|
|
|
f->mantissa = mantissa;
|
|
}
|
|
|
|
/* bytes remaining - only valid for last bset: */
|
|
static unsigned __bset_tree_capacity(const struct btree *b, const struct bset_tree *t)
|
|
{
|
|
bset_aux_tree_verify(b);
|
|
|
|
return btree_aux_data_bytes(b) - t->aux_data_offset * sizeof(u64);
|
|
}
|
|
|
|
static unsigned bset_ro_tree_capacity(const struct btree *b, const struct bset_tree *t)
|
|
{
|
|
return __bset_tree_capacity(b, t) /
|
|
(sizeof(struct bkey_float) + sizeof(u8));
|
|
}
|
|
|
|
static unsigned bset_rw_tree_capacity(const struct btree *b, const struct bset_tree *t)
|
|
{
|
|
return __bset_tree_capacity(b, t) / sizeof(struct rw_aux_tree);
|
|
}
|
|
|
|
static noinline void __build_rw_aux_tree(struct btree *b, struct bset_tree *t)
|
|
{
|
|
struct bkey_packed *k;
|
|
|
|
t->size = 1;
|
|
t->extra = BSET_RW_AUX_TREE_VAL;
|
|
rw_aux_tree(b, t)[0].offset =
|
|
__btree_node_key_to_offset(b, btree_bkey_first(b, t));
|
|
|
|
bset_tree_for_each_key(b, t, k) {
|
|
if (t->size == bset_rw_tree_capacity(b, t))
|
|
break;
|
|
|
|
if ((void *) k - (void *) rw_aux_to_bkey(b, t, t->size - 1) >
|
|
L1_CACHE_BYTES)
|
|
rw_aux_tree_set(b, t, t->size++, k);
|
|
}
|
|
}
|
|
|
|
static noinline void __build_ro_aux_tree(struct btree *b, struct bset_tree *t)
|
|
{
|
|
struct bkey_packed *prev = NULL, *k = btree_bkey_first(b, t);
|
|
struct bkey_i min_key, max_key;
|
|
unsigned cacheline = 1;
|
|
|
|
t->size = min(bkey_to_cacheline(b, t, btree_bkey_last(b, t)),
|
|
bset_ro_tree_capacity(b, t));
|
|
retry:
|
|
if (t->size < 2) {
|
|
t->size = 0;
|
|
t->extra = BSET_NO_AUX_TREE_VAL;
|
|
return;
|
|
}
|
|
|
|
t->extra = (t->size - rounddown_pow_of_two(t->size - 1)) << 1;
|
|
|
|
/* First we figure out where the first key in each cacheline is */
|
|
eytzinger1_for_each(j, t->size - 1) {
|
|
while (bkey_to_cacheline(b, t, k) < cacheline)
|
|
prev = k, k = bkey_p_next(k);
|
|
|
|
if (k >= btree_bkey_last(b, t)) {
|
|
/* XXX: this path sucks */
|
|
t->size--;
|
|
goto retry;
|
|
}
|
|
|
|
ro_aux_tree_prev(b, t)[j] = prev->u64s;
|
|
bkey_float(b, t, j)->key_offset =
|
|
bkey_to_cacheline_offset(b, t, cacheline++, k);
|
|
|
|
EBUG_ON(tree_to_prev_bkey(b, t, j) != prev);
|
|
EBUG_ON(tree_to_bkey(b, t, j) != k);
|
|
}
|
|
|
|
while (k != btree_bkey_last(b, t))
|
|
prev = k, k = bkey_p_next(k);
|
|
|
|
if (!bkey_pack_pos(bkey_to_packed(&min_key), b->data->min_key, b)) {
|
|
bkey_init(&min_key.k);
|
|
min_key.k.p = b->data->min_key;
|
|
}
|
|
|
|
if (!bkey_pack_pos(bkey_to_packed(&max_key), b->data->max_key, b)) {
|
|
bkey_init(&max_key.k);
|
|
max_key.k.p = b->data->max_key;
|
|
}
|
|
|
|
/* Then we build the tree */
|
|
eytzinger1_for_each(j, t->size - 1)
|
|
make_bfloat(b, t, j,
|
|
bkey_to_packed(&min_key),
|
|
bkey_to_packed(&max_key));
|
|
}
|
|
|
|
static void bset_alloc_tree(struct btree *b, struct bset_tree *t)
|
|
{
|
|
struct bset_tree *i;
|
|
|
|
for (i = b->set; i != t; i++)
|
|
BUG_ON(bset_has_rw_aux_tree(i));
|
|
|
|
bch2_bset_set_no_aux_tree(b, t);
|
|
|
|
/* round up to next cacheline: */
|
|
t->aux_data_offset = round_up(bset_aux_tree_buf_start(b, t),
|
|
SMP_CACHE_BYTES / sizeof(u64));
|
|
|
|
bset_aux_tree_verify(b);
|
|
}
|
|
|
|
void bch2_bset_build_aux_tree(struct btree *b, struct bset_tree *t,
|
|
bool writeable)
|
|
{
|
|
if (writeable
|
|
? bset_has_rw_aux_tree(t)
|
|
: bset_has_ro_aux_tree(t))
|
|
return;
|
|
|
|
bset_alloc_tree(b, t);
|
|
|
|
if (!__bset_tree_capacity(b, t))
|
|
return;
|
|
|
|
if (writeable)
|
|
__build_rw_aux_tree(b, t);
|
|
else
|
|
__build_ro_aux_tree(b, t);
|
|
|
|
bset_aux_tree_verify(b);
|
|
}
|
|
|
|
void bch2_bset_init_first(struct btree *b, struct bset *i)
|
|
{
|
|
struct bset_tree *t;
|
|
|
|
BUG_ON(b->nsets);
|
|
|
|
memset(i, 0, sizeof(*i));
|
|
get_random_bytes(&i->seq, sizeof(i->seq));
|
|
SET_BSET_BIG_ENDIAN(i, CPU_BIG_ENDIAN);
|
|
|
|
t = &b->set[b->nsets++];
|
|
set_btree_bset(b, t, i);
|
|
}
|
|
|
|
void bch2_bset_init_next(struct btree *b, struct btree_node_entry *bne)
|
|
{
|
|
struct bset *i = &bne->keys;
|
|
struct bset_tree *t;
|
|
|
|
BUG_ON(bset_byte_offset(b, bne) >= btree_buf_bytes(b));
|
|
BUG_ON((void *) bne < (void *) btree_bkey_last(b, bset_tree_last(b)));
|
|
BUG_ON(b->nsets >= MAX_BSETS);
|
|
|
|
memset(i, 0, sizeof(*i));
|
|
i->seq = btree_bset_first(b)->seq;
|
|
SET_BSET_BIG_ENDIAN(i, CPU_BIG_ENDIAN);
|
|
|
|
t = &b->set[b->nsets++];
|
|
set_btree_bset(b, t, i);
|
|
}
|
|
|
|
/*
|
|
* find _some_ key in the same bset as @k that precedes @k - not necessarily the
|
|
* immediate predecessor:
|
|
*/
|
|
static struct bkey_packed *__bkey_prev(struct btree *b, struct bset_tree *t,
|
|
struct bkey_packed *k)
|
|
{
|
|
struct bkey_packed *p;
|
|
unsigned offset;
|
|
int j;
|
|
|
|
EBUG_ON(k < btree_bkey_first(b, t) ||
|
|
k > btree_bkey_last(b, t));
|
|
|
|
if (k == btree_bkey_first(b, t))
|
|
return NULL;
|
|
|
|
switch (bset_aux_tree_type(t)) {
|
|
case BSET_NO_AUX_TREE:
|
|
p = btree_bkey_first(b, t);
|
|
break;
|
|
case BSET_RO_AUX_TREE:
|
|
j = min_t(unsigned, t->size - 1, bkey_to_cacheline(b, t, k));
|
|
|
|
do {
|
|
p = j ? tree_to_bkey(b, t,
|
|
__inorder_to_eytzinger1(j--,
|
|
t->size - 1, t->extra))
|
|
: btree_bkey_first(b, t);
|
|
} while (p >= k);
|
|
break;
|
|
case BSET_RW_AUX_TREE:
|
|
offset = __btree_node_key_to_offset(b, k);
|
|
j = rw_aux_tree_bsearch(b, t, offset);
|
|
p = j ? rw_aux_to_bkey(b, t, j - 1)
|
|
: btree_bkey_first(b, t);
|
|
break;
|
|
}
|
|
|
|
return p;
|
|
}
|
|
|
|
struct bkey_packed *bch2_bkey_prev_filter(struct btree *b,
|
|
struct bset_tree *t,
|
|
struct bkey_packed *k,
|
|
unsigned min_key_type)
|
|
{
|
|
struct bkey_packed *p, *i, *ret = NULL, *orig_k = k;
|
|
|
|
while ((p = __bkey_prev(b, t, k)) && !ret) {
|
|
for (i = p; i != k; i = bkey_p_next(i))
|
|
if (i->type >= min_key_type)
|
|
ret = i;
|
|
|
|
k = p;
|
|
}
|
|
|
|
if (bch2_expensive_debug_checks) {
|
|
BUG_ON(ret >= orig_k);
|
|
|
|
for (i = ret
|
|
? bkey_p_next(ret)
|
|
: btree_bkey_first(b, t);
|
|
i != orig_k;
|
|
i = bkey_p_next(i))
|
|
BUG_ON(i->type >= min_key_type);
|
|
}
|
|
|
|
return ret;
|
|
}
|
|
|
|
/* Insert */
|
|
|
|
static void bch2_bset_fix_lookup_table(struct btree *b,
|
|
struct bset_tree *t,
|
|
struct bkey_packed *_where,
|
|
unsigned clobber_u64s,
|
|
unsigned new_u64s)
|
|
{
|
|
int shift = new_u64s - clobber_u64s;
|
|
unsigned l, j, where = __btree_node_key_to_offset(b, _where);
|
|
|
|
EBUG_ON(bset_has_ro_aux_tree(t));
|
|
|
|
if (!bset_has_rw_aux_tree(t))
|
|
return;
|
|
|
|
/* returns first entry >= where */
|
|
l = rw_aux_tree_bsearch(b, t, where);
|
|
|
|
if (!l) /* never delete first entry */
|
|
l++;
|
|
else if (l < t->size &&
|
|
where < t->end_offset &&
|
|
rw_aux_tree(b, t)[l].offset == where)
|
|
rw_aux_tree_set(b, t, l++, _where);
|
|
|
|
/* l now > where */
|
|
|
|
for (j = l;
|
|
j < t->size &&
|
|
rw_aux_tree(b, t)[j].offset < where + clobber_u64s;
|
|
j++)
|
|
;
|
|
|
|
if (j < t->size &&
|
|
rw_aux_tree(b, t)[j].offset + shift ==
|
|
rw_aux_tree(b, t)[l - 1].offset)
|
|
j++;
|
|
|
|
memmove(&rw_aux_tree(b, t)[l],
|
|
&rw_aux_tree(b, t)[j],
|
|
(void *) &rw_aux_tree(b, t)[t->size] -
|
|
(void *) &rw_aux_tree(b, t)[j]);
|
|
t->size -= j - l;
|
|
|
|
for (j = l; j < t->size; j++)
|
|
rw_aux_tree(b, t)[j].offset += shift;
|
|
|
|
EBUG_ON(l < t->size &&
|
|
rw_aux_tree(b, t)[l].offset ==
|
|
rw_aux_tree(b, t)[l - 1].offset);
|
|
|
|
if (t->size < bset_rw_tree_capacity(b, t) &&
|
|
(l < t->size
|
|
? rw_aux_tree(b, t)[l].offset
|
|
: t->end_offset) -
|
|
rw_aux_tree(b, t)[l - 1].offset >
|
|
L1_CACHE_BYTES / sizeof(u64)) {
|
|
struct bkey_packed *start = rw_aux_to_bkey(b, t, l - 1);
|
|
struct bkey_packed *end = l < t->size
|
|
? rw_aux_to_bkey(b, t, l)
|
|
: btree_bkey_last(b, t);
|
|
struct bkey_packed *k = start;
|
|
|
|
while (1) {
|
|
k = bkey_p_next(k);
|
|
if (k == end)
|
|
break;
|
|
|
|
if ((void *) k - (void *) start >= L1_CACHE_BYTES) {
|
|
memmove(&rw_aux_tree(b, t)[l + 1],
|
|
&rw_aux_tree(b, t)[l],
|
|
(void *) &rw_aux_tree(b, t)[t->size] -
|
|
(void *) &rw_aux_tree(b, t)[l]);
|
|
t->size++;
|
|
rw_aux_tree_set(b, t, l, k);
|
|
break;
|
|
}
|
|
}
|
|
}
|
|
|
|
bch2_bset_verify_rw_aux_tree(b, t);
|
|
bset_aux_tree_verify(b);
|
|
}
|
|
|
|
void bch2_bset_insert(struct btree *b,
|
|
struct btree_node_iter *iter,
|
|
struct bkey_packed *where,
|
|
struct bkey_i *insert,
|
|
unsigned clobber_u64s)
|
|
{
|
|
struct bkey_format *f = &b->format;
|
|
struct bset_tree *t = bset_tree_last(b);
|
|
struct bkey_packed packed, *src = bkey_to_packed(insert);
|
|
|
|
bch2_bset_verify_rw_aux_tree(b, t);
|
|
bch2_verify_insert_pos(b, where, bkey_to_packed(insert), clobber_u64s);
|
|
|
|
if (bch2_bkey_pack_key(&packed, &insert->k, f))
|
|
src = &packed;
|
|
|
|
if (!bkey_deleted(&insert->k))
|
|
btree_keys_account_key_add(&b->nr, t - b->set, src);
|
|
|
|
if (src->u64s != clobber_u64s) {
|
|
u64 *src_p = (u64 *) where->_data + clobber_u64s;
|
|
u64 *dst_p = (u64 *) where->_data + src->u64s;
|
|
|
|
EBUG_ON((int) le16_to_cpu(bset(b, t)->u64s) <
|
|
(int) clobber_u64s - src->u64s);
|
|
|
|
memmove_u64s(dst_p, src_p, btree_bkey_last(b, t)->_data - src_p);
|
|
le16_add_cpu(&bset(b, t)->u64s, src->u64s - clobber_u64s);
|
|
set_btree_bset_end(b, t);
|
|
}
|
|
|
|
memcpy_u64s_small(where, src,
|
|
bkeyp_key_u64s(f, src));
|
|
memcpy_u64s(bkeyp_val(f, where), &insert->v,
|
|
bkeyp_val_u64s(f, src));
|
|
|
|
if (src->u64s != clobber_u64s)
|
|
bch2_bset_fix_lookup_table(b, t, where, clobber_u64s, src->u64s);
|
|
|
|
bch2_verify_btree_nr_keys(b);
|
|
}
|
|
|
|
void bch2_bset_delete(struct btree *b,
|
|
struct bkey_packed *where,
|
|
unsigned clobber_u64s)
|
|
{
|
|
struct bset_tree *t = bset_tree_last(b);
|
|
u64 *src_p = (u64 *) where->_data + clobber_u64s;
|
|
u64 *dst_p = where->_data;
|
|
|
|
bch2_bset_verify_rw_aux_tree(b, t);
|
|
|
|
EBUG_ON(le16_to_cpu(bset(b, t)->u64s) < clobber_u64s);
|
|
|
|
memmove_u64s_down(dst_p, src_p, btree_bkey_last(b, t)->_data - src_p);
|
|
le16_add_cpu(&bset(b, t)->u64s, -clobber_u64s);
|
|
set_btree_bset_end(b, t);
|
|
|
|
bch2_bset_fix_lookup_table(b, t, where, clobber_u64s, 0);
|
|
}
|
|
|
|
/* Lookup */
|
|
|
|
__flatten
|
|
static struct bkey_packed *bset_search_write_set(const struct btree *b,
|
|
struct bset_tree *t,
|
|
struct bpos *search)
|
|
{
|
|
unsigned l = 0, r = t->size;
|
|
|
|
while (l + 1 != r) {
|
|
unsigned m = (l + r) >> 1;
|
|
|
|
if (bpos_lt(rw_aux_tree(b, t)[m].k, *search))
|
|
l = m;
|
|
else
|
|
r = m;
|
|
}
|
|
|
|
return rw_aux_to_bkey(b, t, l);
|
|
}
|
|
|
|
static inline void prefetch_four_cachelines(void *p)
|
|
{
|
|
#ifdef CONFIG_X86_64
|
|
asm("prefetcht0 (-127 + 64 * 0)(%0);"
|
|
"prefetcht0 (-127 + 64 * 1)(%0);"
|
|
"prefetcht0 (-127 + 64 * 2)(%0);"
|
|
"prefetcht0 (-127 + 64 * 3)(%0);"
|
|
:
|
|
: "r" (p + 127));
|
|
#else
|
|
prefetch(p + L1_CACHE_BYTES * 0);
|
|
prefetch(p + L1_CACHE_BYTES * 1);
|
|
prefetch(p + L1_CACHE_BYTES * 2);
|
|
prefetch(p + L1_CACHE_BYTES * 3);
|
|
#endif
|
|
}
|
|
|
|
static inline bool bkey_mantissa_bits_dropped(const struct btree *b,
|
|
const struct bkey_float *f,
|
|
unsigned idx)
|
|
{
|
|
#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
|
|
unsigned key_bits_start = b->format.key_u64s * 64 - b->nr_key_bits;
|
|
|
|
return f->exponent > key_bits_start;
|
|
#else
|
|
unsigned key_bits_end = high_bit_offset + b->nr_key_bits;
|
|
|
|
return f->exponent + BKEY_MANTISSA_BITS < key_bits_end;
|
|
#endif
|
|
}
|
|
|
|
__flatten
|
|
static struct bkey_packed *bset_search_tree(const struct btree *b,
|
|
const struct bset_tree *t,
|
|
const struct bpos *search,
|
|
const struct bkey_packed *packed_search)
|
|
{
|
|
struct ro_aux_tree *base = ro_aux_tree_base(b, t);
|
|
struct bkey_float *f;
|
|
struct bkey_packed *k;
|
|
unsigned inorder, n = 1, l, r;
|
|
int cmp;
|
|
|
|
do {
|
|
if (likely(n << 4 < t->size))
|
|
prefetch(&base->f[n << 4]);
|
|
|
|
f = &base->f[n];
|
|
if (unlikely(f->exponent >= BFLOAT_FAILED))
|
|
goto slowpath;
|
|
|
|
l = f->mantissa;
|
|
r = bkey_mantissa(packed_search, f, n);
|
|
|
|
if (unlikely(l == r) && bkey_mantissa_bits_dropped(b, f, n))
|
|
goto slowpath;
|
|
|
|
n = n * 2 + (l < r);
|
|
continue;
|
|
slowpath:
|
|
k = tree_to_bkey(b, t, n);
|
|
cmp = bkey_cmp_p_or_unp(b, k, packed_search, search);
|
|
if (!cmp)
|
|
return k;
|
|
|
|
n = n * 2 + (cmp < 0);
|
|
} while (n < t->size);
|
|
|
|
inorder = __eytzinger1_to_inorder(n >> 1, t->size - 1, t->extra);
|
|
|
|
/*
|
|
* n would have been the node we recursed to - the low bit tells us if
|
|
* we recursed left or recursed right.
|
|
*/
|
|
if (likely(!(n & 1))) {
|
|
--inorder;
|
|
if (unlikely(!inorder))
|
|
return btree_bkey_first(b, t);
|
|
|
|
f = &base->f[eytzinger1_prev(n >> 1, t->size - 1)];
|
|
}
|
|
|
|
return cacheline_to_bkey(b, t, inorder, f->key_offset);
|
|
}
|
|
|
|
static __always_inline __flatten
|
|
struct bkey_packed *__bch2_bset_search(struct btree *b,
|
|
struct bset_tree *t,
|
|
struct bpos *search,
|
|
const struct bkey_packed *lossy_packed_search)
|
|
{
|
|
|
|
/*
|
|
* First, we search for a cacheline, then lastly we do a linear search
|
|
* within that cacheline.
|
|
*
|
|
* To search for the cacheline, there's three different possibilities:
|
|
* * The set is too small to have a search tree, so we just do a linear
|
|
* search over the whole set.
|
|
* * The set is the one we're currently inserting into; keeping a full
|
|
* auxiliary search tree up to date would be too expensive, so we
|
|
* use a much simpler lookup table to do a binary search -
|
|
* bset_search_write_set().
|
|
* * Or we use the auxiliary search tree we constructed earlier -
|
|
* bset_search_tree()
|
|
*/
|
|
|
|
switch (bset_aux_tree_type(t)) {
|
|
case BSET_NO_AUX_TREE:
|
|
return btree_bkey_first(b, t);
|
|
case BSET_RW_AUX_TREE:
|
|
return bset_search_write_set(b, t, search);
|
|
case BSET_RO_AUX_TREE:
|
|
return bset_search_tree(b, t, search, lossy_packed_search);
|
|
default:
|
|
BUG();
|
|
}
|
|
}
|
|
|
|
static __always_inline __flatten
|
|
struct bkey_packed *bch2_bset_search_linear(struct btree *b,
|
|
struct bset_tree *t,
|
|
struct bpos *search,
|
|
struct bkey_packed *packed_search,
|
|
const struct bkey_packed *lossy_packed_search,
|
|
struct bkey_packed *m)
|
|
{
|
|
if (lossy_packed_search)
|
|
while (m != btree_bkey_last(b, t) &&
|
|
bkey_iter_cmp_p_or_unp(b, m,
|
|
lossy_packed_search, search) < 0)
|
|
m = bkey_p_next(m);
|
|
|
|
if (!packed_search)
|
|
while (m != btree_bkey_last(b, t) &&
|
|
bkey_iter_pos_cmp(b, m, search) < 0)
|
|
m = bkey_p_next(m);
|
|
|
|
if (bch2_expensive_debug_checks) {
|
|
struct bkey_packed *prev = bch2_bkey_prev_all(b, t, m);
|
|
|
|
BUG_ON(prev &&
|
|
bkey_iter_cmp_p_or_unp(b, prev,
|
|
packed_search, search) >= 0);
|
|
}
|
|
|
|
return m;
|
|
}
|
|
|
|
/* Btree node iterator */
|
|
|
|
static inline void __bch2_btree_node_iter_push(struct btree_node_iter *iter,
|
|
struct btree *b,
|
|
const struct bkey_packed *k,
|
|
const struct bkey_packed *end)
|
|
{
|
|
if (k != end) {
|
|
struct btree_node_iter_set *pos;
|
|
|
|
btree_node_iter_for_each(iter, pos)
|
|
;
|
|
|
|
BUG_ON(pos >= iter->data + ARRAY_SIZE(iter->data));
|
|
*pos = (struct btree_node_iter_set) {
|
|
__btree_node_key_to_offset(b, k),
|
|
__btree_node_key_to_offset(b, end)
|
|
};
|
|
}
|
|
}
|
|
|
|
void bch2_btree_node_iter_push(struct btree_node_iter *iter,
|
|
struct btree *b,
|
|
const struct bkey_packed *k,
|
|
const struct bkey_packed *end)
|
|
{
|
|
__bch2_btree_node_iter_push(iter, b, k, end);
|
|
bch2_btree_node_iter_sort(iter, b);
|
|
}
|
|
|
|
noinline __flatten __cold
|
|
static void btree_node_iter_init_pack_failed(struct btree_node_iter *iter,
|
|
struct btree *b, struct bpos *search)
|
|
{
|
|
struct bkey_packed *k;
|
|
|
|
trace_bkey_pack_pos_fail(search);
|
|
|
|
bch2_btree_node_iter_init_from_start(iter, b);
|
|
|
|
while ((k = bch2_btree_node_iter_peek(iter, b)) &&
|
|
bkey_iter_pos_cmp(b, k, search) < 0)
|
|
bch2_btree_node_iter_advance(iter, b);
|
|
}
|
|
|
|
/**
|
|
* bch2_btree_node_iter_init - initialize a btree node iterator, starting from a
|
|
* given position
|
|
*
|
|
* @iter: iterator to initialize
|
|
* @b: btree node to search
|
|
* @search: search key
|
|
*
|
|
* Main entry point to the lookup code for individual btree nodes:
|
|
*
|
|
* NOTE:
|
|
*
|
|
* When you don't filter out deleted keys, btree nodes _do_ contain duplicate
|
|
* keys. This doesn't matter for most code, but it does matter for lookups.
|
|
*
|
|
* Some adjacent keys with a string of equal keys:
|
|
* i j k k k k l m
|
|
*
|
|
* If you search for k, the lookup code isn't guaranteed to return you any
|
|
* specific k. The lookup code is conceptually doing a binary search and
|
|
* iterating backwards is very expensive so if the pivot happens to land at the
|
|
* last k that's what you'll get.
|
|
*
|
|
* This works out ok, but it's something to be aware of:
|
|
*
|
|
* - For non extents, we guarantee that the live key comes last - see
|
|
* btree_node_iter_cmp(), keys_out_of_order(). So the duplicates you don't
|
|
* see will only be deleted keys you don't care about.
|
|
*
|
|
* - For extents, deleted keys sort last (see the comment at the top of this
|
|
* file). But when you're searching for extents, you actually want the first
|
|
* key strictly greater than your search key - an extent that compares equal
|
|
* to the search key is going to have 0 sectors after the search key.
|
|
*
|
|
* But this does mean that we can't just search for
|
|
* bpos_successor(start_of_range) to get the first extent that overlaps with
|
|
* the range we want - if we're unlucky and there's an extent that ends
|
|
* exactly where we searched, then there could be a deleted key at the same
|
|
* position and we'd get that when we search instead of the preceding extent
|
|
* we needed.
|
|
*
|
|
* So we've got to search for start_of_range, then after the lookup iterate
|
|
* past any extents that compare equal to the position we searched for.
|
|
*/
|
|
__flatten
|
|
void bch2_btree_node_iter_init(struct btree_node_iter *iter,
|
|
struct btree *b, struct bpos *search)
|
|
{
|
|
struct bkey_packed p, *packed_search = NULL;
|
|
struct btree_node_iter_set *pos = iter->data;
|
|
struct bkey_packed *k[MAX_BSETS];
|
|
unsigned i;
|
|
|
|
EBUG_ON(bpos_lt(*search, b->data->min_key));
|
|
EBUG_ON(bpos_gt(*search, b->data->max_key));
|
|
bset_aux_tree_verify(b);
|
|
|
|
memset(iter, 0, sizeof(*iter));
|
|
|
|
switch (bch2_bkey_pack_pos_lossy(&p, *search, b)) {
|
|
case BKEY_PACK_POS_EXACT:
|
|
packed_search = &p;
|
|
break;
|
|
case BKEY_PACK_POS_SMALLER:
|
|
packed_search = NULL;
|
|
break;
|
|
case BKEY_PACK_POS_FAIL:
|
|
btree_node_iter_init_pack_failed(iter, b, search);
|
|
return;
|
|
}
|
|
|
|
for (i = 0; i < b->nsets; i++) {
|
|
k[i] = __bch2_bset_search(b, b->set + i, search, &p);
|
|
prefetch_four_cachelines(k[i]);
|
|
}
|
|
|
|
for (i = 0; i < b->nsets; i++) {
|
|
struct bset_tree *t = b->set + i;
|
|
struct bkey_packed *end = btree_bkey_last(b, t);
|
|
|
|
k[i] = bch2_bset_search_linear(b, t, search,
|
|
packed_search, &p, k[i]);
|
|
if (k[i] != end)
|
|
*pos++ = (struct btree_node_iter_set) {
|
|
__btree_node_key_to_offset(b, k[i]),
|
|
__btree_node_key_to_offset(b, end)
|
|
};
|
|
}
|
|
|
|
bch2_btree_node_iter_sort(iter, b);
|
|
}
|
|
|
|
void bch2_btree_node_iter_init_from_start(struct btree_node_iter *iter,
|
|
struct btree *b)
|
|
{
|
|
struct bset_tree *t;
|
|
|
|
memset(iter, 0, sizeof(*iter));
|
|
|
|
for_each_bset(b, t)
|
|
__bch2_btree_node_iter_push(iter, b,
|
|
btree_bkey_first(b, t),
|
|
btree_bkey_last(b, t));
|
|
bch2_btree_node_iter_sort(iter, b);
|
|
}
|
|
|
|
struct bkey_packed *bch2_btree_node_iter_bset_pos(struct btree_node_iter *iter,
|
|
struct btree *b,
|
|
struct bset_tree *t)
|
|
{
|
|
struct btree_node_iter_set *set;
|
|
|
|
btree_node_iter_for_each(iter, set)
|
|
if (set->end == t->end_offset)
|
|
return __btree_node_offset_to_key(b, set->k);
|
|
|
|
return btree_bkey_last(b, t);
|
|
}
|
|
|
|
static inline bool btree_node_iter_sort_two(struct btree_node_iter *iter,
|
|
struct btree *b,
|
|
unsigned first)
|
|
{
|
|
bool ret;
|
|
|
|
if ((ret = (btree_node_iter_cmp(b,
|
|
iter->data[first],
|
|
iter->data[first + 1]) > 0)))
|
|
swap(iter->data[first], iter->data[first + 1]);
|
|
return ret;
|
|
}
|
|
|
|
void bch2_btree_node_iter_sort(struct btree_node_iter *iter,
|
|
struct btree *b)
|
|
{
|
|
/* unrolled bubble sort: */
|
|
|
|
if (!__btree_node_iter_set_end(iter, 2)) {
|
|
btree_node_iter_sort_two(iter, b, 0);
|
|
btree_node_iter_sort_two(iter, b, 1);
|
|
}
|
|
|
|
if (!__btree_node_iter_set_end(iter, 1))
|
|
btree_node_iter_sort_two(iter, b, 0);
|
|
}
|
|
|
|
void bch2_btree_node_iter_set_drop(struct btree_node_iter *iter,
|
|
struct btree_node_iter_set *set)
|
|
{
|
|
struct btree_node_iter_set *last =
|
|
iter->data + ARRAY_SIZE(iter->data) - 1;
|
|
|
|
memmove(&set[0], &set[1], (void *) last - (void *) set);
|
|
*last = (struct btree_node_iter_set) { 0, 0 };
|
|
}
|
|
|
|
static inline void __bch2_btree_node_iter_advance(struct btree_node_iter *iter,
|
|
struct btree *b)
|
|
{
|
|
iter->data->k += __bch2_btree_node_iter_peek_all(iter, b)->u64s;
|
|
|
|
EBUG_ON(iter->data->k > iter->data->end);
|
|
|
|
if (unlikely(__btree_node_iter_set_end(iter, 0))) {
|
|
/* avoid an expensive memmove call: */
|
|
iter->data[0] = iter->data[1];
|
|
iter->data[1] = iter->data[2];
|
|
iter->data[2] = (struct btree_node_iter_set) { 0, 0 };
|
|
return;
|
|
}
|
|
|
|
if (__btree_node_iter_set_end(iter, 1))
|
|
return;
|
|
|
|
if (!btree_node_iter_sort_two(iter, b, 0))
|
|
return;
|
|
|
|
if (__btree_node_iter_set_end(iter, 2))
|
|
return;
|
|
|
|
btree_node_iter_sort_two(iter, b, 1);
|
|
}
|
|
|
|
void bch2_btree_node_iter_advance(struct btree_node_iter *iter,
|
|
struct btree *b)
|
|
{
|
|
if (bch2_expensive_debug_checks) {
|
|
bch2_btree_node_iter_verify(iter, b);
|
|
bch2_btree_node_iter_next_check(iter, b);
|
|
}
|
|
|
|
__bch2_btree_node_iter_advance(iter, b);
|
|
}
|
|
|
|
/*
|
|
* Expensive:
|
|
*/
|
|
struct bkey_packed *bch2_btree_node_iter_prev_all(struct btree_node_iter *iter,
|
|
struct btree *b)
|
|
{
|
|
struct bkey_packed *k, *prev = NULL;
|
|
struct btree_node_iter_set *set;
|
|
struct bset_tree *t;
|
|
unsigned end = 0;
|
|
|
|
if (bch2_expensive_debug_checks)
|
|
bch2_btree_node_iter_verify(iter, b);
|
|
|
|
for_each_bset(b, t) {
|
|
k = bch2_bkey_prev_all(b, t,
|
|
bch2_btree_node_iter_bset_pos(iter, b, t));
|
|
if (k &&
|
|
(!prev || bkey_iter_cmp(b, k, prev) > 0)) {
|
|
prev = k;
|
|
end = t->end_offset;
|
|
}
|
|
}
|
|
|
|
if (!prev)
|
|
return NULL;
|
|
|
|
/*
|
|
* We're manually memmoving instead of just calling sort() to ensure the
|
|
* prev we picked ends up in slot 0 - sort won't necessarily put it
|
|
* there because of duplicate deleted keys:
|
|
*/
|
|
btree_node_iter_for_each(iter, set)
|
|
if (set->end == end)
|
|
goto found;
|
|
|
|
BUG_ON(set != &iter->data[__btree_node_iter_used(iter)]);
|
|
found:
|
|
BUG_ON(set >= iter->data + ARRAY_SIZE(iter->data));
|
|
|
|
memmove(&iter->data[1],
|
|
&iter->data[0],
|
|
(void *) set - (void *) &iter->data[0]);
|
|
|
|
iter->data[0].k = __btree_node_key_to_offset(b, prev);
|
|
iter->data[0].end = end;
|
|
|
|
if (bch2_expensive_debug_checks)
|
|
bch2_btree_node_iter_verify(iter, b);
|
|
return prev;
|
|
}
|
|
|
|
struct bkey_packed *bch2_btree_node_iter_prev(struct btree_node_iter *iter,
|
|
struct btree *b)
|
|
{
|
|
struct bkey_packed *prev;
|
|
|
|
do {
|
|
prev = bch2_btree_node_iter_prev_all(iter, b);
|
|
} while (prev && bkey_deleted(prev));
|
|
|
|
return prev;
|
|
}
|
|
|
|
struct bkey_s_c bch2_btree_node_iter_peek_unpack(struct btree_node_iter *iter,
|
|
struct btree *b,
|
|
struct bkey *u)
|
|
{
|
|
struct bkey_packed *k = bch2_btree_node_iter_peek(iter, b);
|
|
|
|
return k ? bkey_disassemble(b, k, u) : bkey_s_c_null;
|
|
}
|
|
|
|
/* Mergesort */
|
|
|
|
void bch2_btree_keys_stats(const struct btree *b, struct bset_stats *stats)
|
|
{
|
|
const struct bset_tree *t;
|
|
|
|
for_each_bset(b, t) {
|
|
enum bset_aux_tree_type type = bset_aux_tree_type(t);
|
|
size_t j;
|
|
|
|
stats->sets[type].nr++;
|
|
stats->sets[type].bytes += le16_to_cpu(bset(b, t)->u64s) *
|
|
sizeof(u64);
|
|
|
|
if (bset_has_ro_aux_tree(t)) {
|
|
stats->floats += t->size - 1;
|
|
|
|
for (j = 1; j < t->size; j++)
|
|
stats->failed +=
|
|
bkey_float(b, t, j)->exponent ==
|
|
BFLOAT_FAILED;
|
|
}
|
|
}
|
|
}
|
|
|
|
void bch2_bfloat_to_text(struct printbuf *out, struct btree *b,
|
|
struct bkey_packed *k)
|
|
{
|
|
struct bset_tree *t = bch2_bkey_to_bset(b, k);
|
|
struct bkey uk;
|
|
unsigned j, inorder;
|
|
|
|
if (!bset_has_ro_aux_tree(t))
|
|
return;
|
|
|
|
inorder = bkey_to_cacheline(b, t, k);
|
|
if (!inorder || inorder >= t->size)
|
|
return;
|
|
|
|
j = __inorder_to_eytzinger1(inorder, t->size - 1, t->extra);
|
|
if (k != tree_to_bkey(b, t, j))
|
|
return;
|
|
|
|
switch (bkey_float(b, t, j)->exponent) {
|
|
case BFLOAT_FAILED:
|
|
uk = bkey_unpack_key(b, k);
|
|
prt_printf(out,
|
|
" failed unpacked at depth %u\n"
|
|
"\t",
|
|
ilog2(j));
|
|
bch2_bpos_to_text(out, uk.p);
|
|
prt_printf(out, "\n");
|
|
break;
|
|
}
|
|
}
|