aboutsummaryrefslogtreecommitdiff
path: root/fs/reiserfs/bitmap.c
diff options
context:
space:
mode:
Diffstat (limited to 'fs/reiserfs/bitmap.c')
-rw-r--r--fs/reiserfs/bitmap.c742
1 files changed, 472 insertions, 270 deletions
diff --git a/fs/reiserfs/bitmap.c b/fs/reiserfs/bitmap.c
index 909f71e9a30..dc198bc64c6 100644
--- a/fs/reiserfs/bitmap.c
+++ b/fs/reiserfs/bitmap.c
@@ -3,16 +3,15 @@
*/
/* Reiserfs block (de)allocator, bitmap-based. */
-#include <linux/config.h>
#include <linux/time.h>
-#include <linux/reiserfs_fs.h>
+#include "reiserfs.h"
#include <linux/errno.h>
#include <linux/buffer_head.h>
#include <linux/kernel.h>
#include <linux/pagemap.h>
-#include <linux/reiserfs_fs_sb.h>
-#include <linux/reiserfs_fs_i.h>
+#include <linux/vmalloc.h>
#include <linux/quotaops.h>
+#include <linux/seq_file.h>
#define PREALLOCATION_SIZE 9
@@ -40,83 +39,86 @@
#define SET_OPTION(optname) \
do { \
- reiserfs_warning(s, "reiserfs: option \"%s\" is set", #optname); \
- set_bit(_ALLOC_ ## optname , &SB_ALLOC_OPTS(s)); \
+ reiserfs_info(s, "block allocator option \"%s\" is set", #optname); \
+ set_bit(_ALLOC_ ## optname , &SB_ALLOC_OPTS(s)); \
} while(0)
#define TEST_OPTION(optname, s) \
test_bit(_ALLOC_ ## optname , &SB_ALLOC_OPTS(s))
static inline void get_bit_address(struct super_block *s,
- b_blocknr_t block, int *bmap_nr, int *offset)
+ b_blocknr_t block,
+ unsigned int *bmap_nr,
+ unsigned int *offset)
{
- /* It is in the bitmap block number equal to the block
- * number divided by the number of bits in a block. */
- *bmap_nr = block / (s->s_blocksize << 3);
+ /*
+ * It is in the bitmap block number equal to the block
+ * number divided by the number of bits in a block.
+ */
+ *bmap_nr = block >> (s->s_blocksize_bits + 3);
/* Within that bitmap block it is located at bit offset *offset. */
*offset = block & ((s->s_blocksize << 3) - 1);
- return;
}
-#ifdef CONFIG_REISERFS_CHECK
int is_reusable(struct super_block *s, b_blocknr_t block, int bit_value)
{
- int i, j;
+ unsigned int bmap, offset;
+ unsigned int bmap_count = reiserfs_bmap_count(s);
if (block == 0 || block >= SB_BLOCK_COUNT(s)) {
- reiserfs_warning(s,
- "vs-4010: is_reusable: block number is out of range %lu (%u)",
- block, SB_BLOCK_COUNT(s));
+ reiserfs_error(s, "vs-4010",
+ "block number is out of range %lu (%u)",
+ block, SB_BLOCK_COUNT(s));
return 0;
}
- /* it can't be one of the bitmap blocks */
- for (i = 0; i < SB_BMAP_NR(s); i++)
- if (block == SB_AP_BITMAP(s)[i].bh->b_blocknr) {
- reiserfs_warning(s, "vs: 4020: is_reusable: "
- "bitmap block %lu(%u) can't be freed or reused",
- block, SB_BMAP_NR(s));
+ get_bit_address(s, block, &bmap, &offset);
+
+ /*
+ * Old format filesystem? Unlikely, but the bitmaps are all
+ * up front so we need to account for it.
+ */
+ if (unlikely(test_bit(REISERFS_OLD_FORMAT,
+ &REISERFS_SB(s)->s_properties))) {
+ b_blocknr_t bmap1 = REISERFS_SB(s)->s_sbh->b_blocknr + 1;
+ if (block >= bmap1 &&
+ block <= bmap1 + bmap_count) {
+ reiserfs_error(s, "vs-4019", "bitmap block %lu(%u) "
+ "can't be freed or reused",
+ block, bmap_count);
+ return 0;
+ }
+ } else {
+ if (offset == 0) {
+ reiserfs_error(s, "vs-4020", "bitmap block %lu(%u) "
+ "can't be freed or reused",
+ block, bmap_count);
return 0;
}
-
- get_bit_address(s, block, &i, &j);
-
- if (i >= SB_BMAP_NR(s)) {
- reiserfs_warning(s,
- "vs-4030: is_reusable: there is no so many bitmap blocks: "
- "block=%lu, bitmap_nr=%d", block, i);
- return 0;
}
- if ((bit_value == 0 &&
- reiserfs_test_le_bit(j, SB_AP_BITMAP(s)[i].bh->b_data)) ||
- (bit_value == 1 &&
- reiserfs_test_le_bit(j, SB_AP_BITMAP(s)[i].bh->b_data) == 0)) {
- reiserfs_warning(s,
- "vs-4040: is_reusable: corresponding bit of block %lu does not "
- "match required value (i==%d, j==%d) test_bit==%d",
- block, i, j, reiserfs_test_le_bit(j,
- SB_AP_BITMAP
- (s)[i].bh->
- b_data));
-
+ if (bmap >= bmap_count) {
+ reiserfs_error(s, "vs-4030", "bitmap for requested block "
+ "is out of range: block=%lu, bitmap_nr=%u",
+ block, bmap);
return 0;
}
if (bit_value == 0 && block == SB_ROOT_BLOCK(s)) {
- reiserfs_warning(s,
- "vs-4050: is_reusable: this is root block (%u), "
- "it must be busy", SB_ROOT_BLOCK(s));
+ reiserfs_error(s, "vs-4050", "this is root block (%u), "
+ "it must be busy", SB_ROOT_BLOCK(s));
return 0;
}
return 1;
}
-#endif /* CONFIG_REISERFS_CHECK */
-/* searches in journal structures for a given block number (bmap, off). If block
- is found in reiserfs journal it suggests next free block candidate to test. */
-static inline int is_block_in_journal(struct super_block *s, int bmap, int
- off, int *next)
+/*
+ * Searches in journal structures for a given block number (bmap, off).
+ * If block is found in reiserfs journal it suggests next free block
+ * candidate to test.
+ */
+static inline int is_block_in_journal(struct super_block *s, unsigned int bmap,
+ int off, int *next)
{
b_blocknr_t tmp;
@@ -125,7 +127,7 @@ static inline int is_block_in_journal(struct super_block *s, int bmap, int
*next = tmp;
PROC_INFO_INC(s, scan_bitmap.in_journal_hint);
} else {
- (*next) = off + 1; /* inc offset to avoid looping. */
+ (*next) = off + 1; /* inc offset to avoid looping. */
PROC_INFO_INC(s, scan_bitmap.in_journal_nohint);
}
PROC_INFO_INC(s, scan_bitmap.retry);
@@ -134,48 +136,52 @@ static inline int is_block_in_journal(struct super_block *s, int bmap, int
return 0;
}
-/* it searches for a window of zero bits with given minimum and maximum lengths in one bitmap
- * block; */
+/*
+ * Searches for a window of zero bits with given minimum and maximum
+ * lengths in one bitmap block
+ */
static int scan_bitmap_block(struct reiserfs_transaction_handle *th,
- int bmap_n, int *beg, int boundary, int min,
- int max, int unfm)
+ unsigned int bmap_n, int *beg, int boundary,
+ int min, int max, int unfm)
{
struct super_block *s = th->t_super;
struct reiserfs_bitmap_info *bi = &SB_AP_BITMAP(s)[bmap_n];
+ struct buffer_head *bh;
int end, next;
int org = *beg;
BUG_ON(!th->t_trans_id);
-
- RFALSE(bmap_n >= SB_BMAP_NR(s), "Bitmap %d is out of range (0..%d)",
- bmap_n, SB_BMAP_NR(s) - 1);
+ RFALSE(bmap_n >= reiserfs_bmap_count(s), "Bitmap %u is out of "
+ "range (0..%u)", bmap_n, reiserfs_bmap_count(s) - 1);
PROC_INFO_INC(s, scan_bitmap.bmap);
-/* this is unclear and lacks comments, explain how journal bitmaps
- work here for the reader. Convey a sense of the design here. What
- is a window? */
-/* - I mean `a window of zero bits' as in description of this function - Zam. */
if (!bi) {
- reiserfs_warning(s, "NULL bitmap info pointer for bitmap %d",
- bmap_n);
+ reiserfs_error(s, "jdm-4055", "NULL bitmap info pointer "
+ "for bitmap %d", bmap_n);
return 0;
}
- if (buffer_locked(bi->bh)) {
- PROC_INFO_INC(s, scan_bitmap.wait);
- __wait_on_buffer(bi->bh);
- }
+
+ bh = reiserfs_read_bitmap_block(s, bmap_n);
+ if (bh == NULL)
+ return 0;
while (1) {
- cont:
- if (bi->free_count < min)
- return 0; // No free blocks in this bitmap
+cont:
+ if (bi->free_count < min) {
+ brelse(bh);
+ return 0; /* No free blocks in this bitmap */
+ }
- /* search for a first zero bit -- beggining of a window */
+ /* search for a first zero bit -- beginning of a window */
*beg = reiserfs_find_next_zero_le_bit
- ((unsigned long *)(bi->bh->b_data), boundary, *beg);
+ ((unsigned long *)(bh->b_data), boundary, *beg);
- if (*beg + min > boundary) { /* search for a zero bit fails or the rest of bitmap block
- * cannot contain a zero window of minimum size */
+ /*
+ * search for a zero bit fails or the rest of bitmap block
+ * cannot contain a zero window of minimum size
+ */
+ if (*beg + min > boundary) {
+ brelse(bh);
return 0;
}
@@ -184,54 +190,79 @@ static int scan_bitmap_block(struct reiserfs_transaction_handle *th,
/* first zero bit found; we check next bits */
for (end = *beg + 1;; end++) {
if (end >= *beg + max || end >= boundary
- || reiserfs_test_le_bit(end, bi->bh->b_data)) {
+ || reiserfs_test_le_bit(end, bh->b_data)) {
next = end;
break;
}
- /* finding the other end of zero bit window requires looking into journal structures (in
- * case of searching for free blocks for unformatted nodes) */
+
+ /*
+ * finding the other end of zero bit window requires
+ * looking into journal structures (in case of
+ * searching for free blocks for unformatted nodes)
+ */
if (unfm && is_block_in_journal(s, bmap_n, end, &next))
break;
}
- /* now (*beg) points to beginning of zero bits window,
- * (end) points to one bit after the window end */
- if (end - *beg >= min) { /* it seems we have found window of proper size */
+ /*
+ * now (*beg) points to beginning of zero bits window,
+ * (end) points to one bit after the window end
+ */
+
+ /* found window of proper size */
+ if (end - *beg >= min) {
int i;
- reiserfs_prepare_for_journal(s, bi->bh, 1);
- /* try to set all blocks used checking are they still free */
+ reiserfs_prepare_for_journal(s, bh, 1);
+ /*
+ * try to set all blocks used checking are
+ * they still free
+ */
for (i = *beg; i < end; i++) {
- /* It seems that we should not check in journal again. */
+ /* Don't check in journal again. */
if (reiserfs_test_and_set_le_bit
- (i, bi->bh->b_data)) {
- /* bit was set by another process
- * while we slept in prepare_for_journal() */
+ (i, bh->b_data)) {
+ /*
+ * bit was set by another process while
+ * we slept in prepare_for_journal()
+ */
PROC_INFO_INC(s, scan_bitmap.stolen);
- if (i >= *beg + min) { /* we can continue with smaller set of allocated blocks,
- * if length of this set is more or equal to `min' */
+
+ /*
+ * we can continue with smaller set
+ * of allocated blocks, if length of
+ * this set is more or equal to `min'
+ */
+ if (i >= *beg + min) {
end = i;
break;
}
- /* otherwise we clear all bit were set ... */
+
+ /*
+ * otherwise we clear all bit
+ * were set ...
+ */
while (--i >= *beg)
- reiserfs_test_and_clear_le_bit
- (i, bi->bh->b_data);
- reiserfs_restore_prepared_buffer(s,
- bi->
- bh);
+ reiserfs_clear_le_bit
+ (i, bh->b_data);
+ reiserfs_restore_prepared_buffer(s, bh);
*beg = org;
- /* ... and search again in current block from beginning */
+
+ /*
+ * Search again in current block
+ * from beginning
+ */
goto cont;
}
}
bi->free_count -= (end - *beg);
- journal_mark_dirty(th, s, bi->bh);
+ journal_mark_dirty(th, bh);
+ brelse(bh);
/* free block count calculation */
reiserfs_prepare_for_journal(s, SB_BUFFER_WITH_SB(s),
1);
PUT_SB_FREE_BLOCKS(s, SB_FREE_BLOCKS(s) - (end - *beg));
- journal_mark_dirty(th, s, SB_BUFFER_WITH_SB(s));
+ journal_mark_dirty(th, SB_BUFFER_WITH_SB(s));
return end - (*beg);
} else {
@@ -251,12 +282,12 @@ static int bmap_hash_id(struct super_block *s, u32 id)
} else {
hash_in = (char *)(&id);
hash = keyed_hash(hash_in, 4);
- bm = hash % SB_BMAP_NR(s);
+ bm = hash % reiserfs_bmap_count(s);
if (!bm)
bm = 1;
}
/* this can only be true when SB_BMAP_NR = 1 */
- if (bm >= SB_BMAP_NR(s))
+ if (bm >= reiserfs_bmap_count(s))
bm = 0;
return bm;
}
@@ -267,9 +298,22 @@ static int bmap_hash_id(struct super_block *s, u32 id)
*/
static inline int block_group_used(struct super_block *s, u32 id)
{
- int bm;
- bm = bmap_hash_id(s, id);
- if (SB_AP_BITMAP(s)[bm].free_count > ((s->s_blocksize << 3) * 60 / 100)) {
+ int bm = bmap_hash_id(s, id);
+ struct reiserfs_bitmap_info *info = &SB_AP_BITMAP(s)[bm];
+
+ /*
+ * If we don't have cached information on this bitmap block, we're
+ * going to have to load it later anyway. Loading it here allows us
+ * to make a better decision. This favors long-term performance gain
+ * with a better on-disk layout vs. a short term gain of skipping the
+ * read and potentially having a bad placement.
+ */
+ if (info->free_count == UINT_MAX) {
+ struct buffer_head *bh = reiserfs_read_bitmap_block(s, bm);
+ brelse(bh);
+ }
+
+ if (info->free_count > ((s->s_blocksize << 3) * 60 / 100)) {
return 0;
}
return 1;
@@ -296,35 +340,36 @@ __le32 reiserfs_choose_packing(struct inode * dir)
return packing;
}
-/* Tries to find contiguous zero bit window (given size) in given region of
- * bitmap and place new blocks there. Returns number of allocated blocks. */
+/*
+ * Tries to find contiguous zero bit window (given size) in given region of
+ * bitmap and place new blocks there. Returns number of allocated blocks.
+ */
static int scan_bitmap(struct reiserfs_transaction_handle *th,
b_blocknr_t * start, b_blocknr_t finish,
- int min, int max, int unfm, unsigned long file_block)
+ int min, int max, int unfm, sector_t file_block)
{
int nr_allocated = 0;
struct super_block *s = th->t_super;
- /* find every bm and bmap and bmap_nr in this file, and change them all to bitmap_blocknr
- * - Hans, it is not a block number - Zam. */
-
- int bm, off;
- int end_bm, end_off;
- int off_max = s->s_blocksize << 3;
+ unsigned int bm, off;
+ unsigned int end_bm, end_off;
+ unsigned int off_max = s->s_blocksize << 3;
BUG_ON(!th->t_trans_id);
-
PROC_INFO_INC(s, scan_bitmap.call);
+
+ /* No point in looking for more free blocks */
if (SB_FREE_BLOCKS(s) <= 0)
- return 0; // No point in looking for more free blocks
+ return 0;
get_bit_address(s, *start, &bm, &off);
get_bit_address(s, finish, &end_bm, &end_off);
- if (bm > SB_BMAP_NR(s))
+ if (bm > reiserfs_bmap_count(s))
return 0;
- if (end_bm > SB_BMAP_NR(s))
- end_bm = SB_BMAP_NR(s);
+ if (end_bm > reiserfs_bmap_count(s))
+ end_bm = reiserfs_bmap_count(s);
- /* When the bitmap is more than 10% free, anyone can allocate.
+ /*
+ * When the bitmap is more than 10% free, anyone can allocate.
* When it's less than 10% free, only files that already use the
* bitmap are allowed. Once we pass 80% full, this restriction
* is lifted.
@@ -362,7 +407,7 @@ static int scan_bitmap(struct reiserfs_transaction_handle *th,
nr_allocated =
scan_bitmap_block(th, bm, &off, end_off + 1, min, max, unfm);
- ret:
+ret:
*start = bm * off_max + off;
return nr_allocated;
@@ -374,45 +419,49 @@ static void _reiserfs_free_block(struct reiserfs_transaction_handle *th,
{
struct super_block *s = th->t_super;
struct reiserfs_super_block *rs;
- struct buffer_head *sbh;
+ struct buffer_head *sbh, *bmbh;
struct reiserfs_bitmap_info *apbi;
- int nr, offset;
+ unsigned int nr, offset;
BUG_ON(!th->t_trans_id);
-
PROC_INFO_INC(s, free_block);
-
rs = SB_DISK_SUPER_BLOCK(s);
sbh = SB_BUFFER_WITH_SB(s);
apbi = SB_AP_BITMAP(s);
get_bit_address(s, block, &nr, &offset);
- if (nr >= sb_bmap_nr(rs)) {
- reiserfs_warning(s, "vs-4075: reiserfs_free_block: "
- "block %lu is out of range on %s",
- block, reiserfs_bdevname(s));
+ if (nr >= reiserfs_bmap_count(s)) {
+ reiserfs_error(s, "vs-4075", "block %lu is out of range",
+ block);
return;
}
- reiserfs_prepare_for_journal(s, apbi[nr].bh, 1);
+ bmbh = reiserfs_read_bitmap_block(s, nr);
+ if (!bmbh)
+ return;
+
+ reiserfs_prepare_for_journal(s, bmbh, 1);
/* clear bit for the given block in bit map */
- if (!reiserfs_test_and_clear_le_bit(offset, apbi[nr].bh->b_data)) {
- reiserfs_warning(s, "vs-4080: reiserfs_free_block: "
- "free_block (%s:%lu)[dev:blocknr]: bit already cleared",
- reiserfs_bdevname(s), block);
+ if (!reiserfs_test_and_clear_le_bit(offset, bmbh->b_data)) {
+ reiserfs_error(s, "vs-4080",
+ "block %lu: bit already cleared", block);
}
apbi[nr].free_count++;
- journal_mark_dirty(th, s, apbi[nr].bh);
+ journal_mark_dirty(th, bmbh);
+ brelse(bmbh);
reiserfs_prepare_for_journal(s, sbh, 1);
/* update super block */
set_sb_free_blocks(rs, sb_free_blocks(rs) + 1);
- journal_mark_dirty(th, s, sbh);
- if (for_unformatted)
- DQUOT_FREE_BLOCK_NODIRTY(inode, 1);
+ journal_mark_dirty(th, sbh);
+ if (for_unformatted) {
+ int depth = reiserfs_write_unlock_nested(s);
+ dquot_free_block_nodirty(inode, 1);
+ reiserfs_write_lock_nested(s, depth);
+ }
}
void reiserfs_free_block(struct reiserfs_transaction_handle *th,
@@ -422,10 +471,17 @@ void reiserfs_free_block(struct reiserfs_transaction_handle *th,
struct super_block *s = th->t_super;
BUG_ON(!th->t_trans_id);
-
RFALSE(!s, "vs-4061: trying to free block on nonexistent device");
- RFALSE(is_reusable(s, block, 1) == 0,
- "vs-4071: can not free such block");
+ if (!is_reusable(s, block, 1))
+ return;
+
+ if (block > sb_block_count(REISERFS_SB(s)->s_rs)) {
+ reiserfs_error(th->t_super, "bitmap-4072",
+ "Trying to free block outside file system "
+ "boundaries (%lu > %lu)",
+ block, sb_block_count(REISERFS_SB(s)->s_rs));
+ return;
+ }
/* mark it before we clear it, just in case */
journal_mark_freed(th, s, block);
_reiserfs_free_block(th, inode, block, for_unformatted);
@@ -435,11 +491,11 @@ void reiserfs_free_block(struct reiserfs_transaction_handle *th,
static void reiserfs_free_prealloc_block(struct reiserfs_transaction_handle *th,
struct inode *inode, b_blocknr_t block)
{
+ BUG_ON(!th->t_trans_id);
RFALSE(!th->t_super,
"vs-4060: trying to free block on nonexistent device");
- RFALSE(is_reusable(th->t_super, block, 1) == 0,
- "vs-4070: can not free such block");
- BUG_ON(!th->t_trans_id);
+ if (!is_reusable(th->t_super, block, 1))
+ return;
_reiserfs_free_block(th, inode, block, 1);
}
@@ -449,12 +505,12 @@ static void __discard_prealloc(struct reiserfs_transaction_handle *th,
unsigned long save = ei->i_prealloc_block;
int dirty = 0;
struct inode *inode = &ei->vfs_inode;
+
BUG_ON(!th->t_trans_id);
#ifdef CONFIG_REISERFS_CHECK
if (ei->i_prealloc_count < 0)
- reiserfs_warning(th->t_super,
- "zam-4001:%s: inode has negative prealloc blocks count.",
- __FUNCTION__);
+ reiserfs_error(th->t_super, "zam-4001",
+ "inode has negative prealloc blocks count.");
#endif
while (ei->i_prealloc_count > 0) {
reiserfs_free_prealloc_block(th, inode, ei->i_prealloc_block);
@@ -465,7 +521,7 @@ static void __discard_prealloc(struct reiserfs_transaction_handle *th,
if (dirty)
reiserfs_update_sd(th, inode);
ei->i_prealloc_block = save;
- list_del_init(&(ei->i_prealloc_list));
+ list_del_init(&ei->i_prealloc_list);
}
/* FIXME: It should be inline function */
@@ -473,6 +529,7 @@ void reiserfs_discard_prealloc(struct reiserfs_transaction_handle *th,
struct inode *inode)
{
struct reiserfs_inode_info *ei = REISERFS_I(inode);
+
BUG_ON(!th->t_trans_id);
if (ei->i_prealloc_count)
__discard_prealloc(th, ei);
@@ -483,16 +540,15 @@ void reiserfs_discard_all_prealloc(struct reiserfs_transaction_handle *th)
struct list_head *plist = &SB_JOURNAL(th->t_super)->j_prealloc_list;
BUG_ON(!th->t_trans_id);
-
while (!list_empty(plist)) {
struct reiserfs_inode_info *ei;
ei = list_entry(plist->next, struct reiserfs_inode_info,
i_prealloc_list);
#ifdef CONFIG_REISERFS_CHECK
if (!ei->i_prealloc_count) {
- reiserfs_warning(th->t_super,
- "zam-4001:%s: inode is in prealloc list but has no preallocated blocks.",
- __FUNCTION__);
+ reiserfs_error(th->t_super, "zam-4001",
+ "inode is in prealloc list but has "
+ "no preallocated blocks.");
}
#endif
__discard_prealloc(th, ei);
@@ -511,7 +567,8 @@ int reiserfs_parse_alloc_options(struct super_block *s, char *options)
{
char *this_char, *value;
- REISERFS_SB(s)->s_alloc_options.bits = 0; /* clear default settings */
+ /* clear default settings */
+ REISERFS_SB(s)->s_alloc_options.bits = 0;
while ((this_char = strsep(&options, ":")) != NULL) {
if ((value = strchr(this_char, '=')) != NULL)
@@ -541,7 +598,7 @@ int reiserfs_parse_alloc_options(struct super_block *s, char *options)
if (!strcmp(this_char, "displacing_new_packing_localities")) {
SET_OPTION(displacing_new_packing_localities);
continue;
- };
+ }
if (!strcmp(this_char, "old_hashed_relocation")) {
SET_OPTION(old_hashed_relocation);
@@ -606,23 +663,114 @@ int reiserfs_parse_alloc_options(struct super_block *s, char *options)
continue;
}
- reiserfs_warning(s, "zam-4001: %s : unknown option - %s",
- __FUNCTION__, this_char);
+ reiserfs_warning(s, "zam-4001", "unknown option - %s",
+ this_char);
return 1;
}
- reiserfs_warning(s, "allocator options = [%08x]\n", SB_ALLOC_OPTS(s));
+ reiserfs_info(s, "allocator options = [%08x]\n", SB_ALLOC_OPTS(s));
return 0;
}
+static void print_sep(struct seq_file *seq, int *first)
+{
+ if (!*first)
+ seq_puts(seq, ":");
+ else
+ *first = 0;
+}
+
+void show_alloc_options(struct seq_file *seq, struct super_block *s)
+{
+ int first = 1;
+
+ if (SB_ALLOC_OPTS(s) == ((1 << _ALLOC_skip_busy) |
+ (1 << _ALLOC_dirid_groups) | (1 << _ALLOC_packing_groups)))
+ return;
+
+ seq_puts(seq, ",alloc=");
+
+ if (TEST_OPTION(concentrating_formatted_nodes, s)) {
+ print_sep(seq, &first);
+ if (REISERFS_SB(s)->s_alloc_options.border != 10) {
+ seq_printf(seq, "concentrating_formatted_nodes=%d",
+ 100 / REISERFS_SB(s)->s_alloc_options.border);
+ } else
+ seq_puts(seq, "concentrating_formatted_nodes");
+ }
+ if (TEST_OPTION(displacing_large_files, s)) {
+ print_sep(seq, &first);
+ if (REISERFS_SB(s)->s_alloc_options.large_file_size != 16) {
+ seq_printf(seq, "displacing_large_files=%lu",
+ REISERFS_SB(s)->s_alloc_options.large_file_size);
+ } else
+ seq_puts(seq, "displacing_large_files");
+ }
+ if (TEST_OPTION(displacing_new_packing_localities, s)) {
+ print_sep(seq, &first);
+ seq_puts(seq, "displacing_new_packing_localities");
+ }
+ if (TEST_OPTION(old_hashed_relocation, s)) {
+ print_sep(seq, &first);
+ seq_puts(seq, "old_hashed_relocation");
+ }
+ if (TEST_OPTION(new_hashed_relocation, s)) {
+ print_sep(seq, &first);
+ seq_puts(seq, "new_hashed_relocation");
+ }
+ if (TEST_OPTION(dirid_groups, s)) {
+ print_sep(seq, &first);
+ seq_puts(seq, "dirid_groups");
+ }
+ if (TEST_OPTION(oid_groups, s)) {
+ print_sep(seq, &first);
+ seq_puts(seq, "oid_groups");
+ }
+ if (TEST_OPTION(packing_groups, s)) {
+ print_sep(seq, &first);
+ seq_puts(seq, "packing_groups");
+ }
+ if (TEST_OPTION(hashed_formatted_nodes, s)) {
+ print_sep(seq, &first);
+ seq_puts(seq, "hashed_formatted_nodes");
+ }
+ if (TEST_OPTION(skip_busy, s)) {
+ print_sep(seq, &first);
+ seq_puts(seq, "skip_busy");
+ }
+ if (TEST_OPTION(hundredth_slices, s)) {
+ print_sep(seq, &first);
+ seq_puts(seq, "hundredth_slices");
+ }
+ if (TEST_OPTION(old_way, s)) {
+ print_sep(seq, &first);
+ seq_puts(seq, "old_way");
+ }
+ if (TEST_OPTION(displace_based_on_dirid, s)) {
+ print_sep(seq, &first);
+ seq_puts(seq, "displace_based_on_dirid");
+ }
+ if (REISERFS_SB(s)->s_alloc_options.preallocmin != 0) {
+ print_sep(seq, &first);
+ seq_printf(seq, "preallocmin=%d",
+ REISERFS_SB(s)->s_alloc_options.preallocmin);
+ }
+ if (REISERFS_SB(s)->s_alloc_options.preallocsize != 17) {
+ print_sep(seq, &first);
+ seq_printf(seq, "preallocsize=%d",
+ REISERFS_SB(s)->s_alloc_options.preallocsize);
+ }
+}
+
static inline void new_hashed_relocation(reiserfs_blocknr_hint_t * hint)
{
char *hash_in;
+
if (hint->formatted_node) {
hash_in = (char *)&hint->key.k_dir_id;
} else {
if (!hint->inode) {
- //hint->search_start = hint->beg;
+ /*hint->search_start = hint->beg;*/
hash_in = (char *)&hint->key.k_dir_id;
} else
if (TEST_OPTION(displace_based_on_dirid, hint->th->t_super))
@@ -638,7 +786,7 @@ static inline void new_hashed_relocation(reiserfs_blocknr_hint_t * hint)
/*
* Relocation based on dirid, hashing them into a given bitmap block
- * files. Formatted nodes are unaffected, a seperate policy covers them
+ * files. Formatted nodes are unaffected, a separate policy covers them
*/
static void dirid_groups(reiserfs_blocknr_hint_t * hint)
{
@@ -646,6 +794,7 @@ static void dirid_groups(reiserfs_blocknr_hint_t * hint)
__u32 dirid = 0;
int bm = 0;
struct super_block *sb = hint->th->t_super;
+
if (hint->inode)
dirid = le32_to_cpu(INODE_PKEY(hint->inode)->k_dir_id);
else if (hint->formatted_node)
@@ -663,7 +812,7 @@ static void dirid_groups(reiserfs_blocknr_hint_t * hint)
/*
* Relocation based on oid, hashing them into a given bitmap block
- * files. Formatted nodes are unaffected, a seperate policy covers them
+ * files. Formatted nodes are unaffected, a separate policy covers them
*/
static void oid_groups(reiserfs_blocknr_hint_t * hint)
{
@@ -675,7 +824,8 @@ static void oid_groups(reiserfs_blocknr_hint_t * hint)
dirid = le32_to_cpu(INODE_PKEY(hint->inode)->k_dir_id);
- /* keep the root dir and it's first set of subdirs close to
+ /*
+ * keep the root dir and it's first set of subdirs close to
* the start of the disk
*/
if (dirid <= 2)
@@ -689,37 +839,42 @@ static void oid_groups(reiserfs_blocknr_hint_t * hint)
}
}
-/* returns 1 if it finds an indirect item and gets valid hint info
+/*
+ * returns 1 if it finds an indirect item and gets valid hint info
* from it, otherwise 0
*/
static int get_left_neighbor(reiserfs_blocknr_hint_t * hint)
{
- struct path *path;
+ struct treepath *path;
struct buffer_head *bh;
struct item_head *ih;
int pos_in_item;
__le32 *item;
int ret = 0;
- if (!hint->path) /* reiserfs code can call this function w/o pointer to path
- * structure supplied; then we rely on supplied search_start */
+ /*
+ * reiserfs code can call this function w/o pointer to path
+ * structure supplied; then we rely on supplied search_start
+ */
+ if (!hint->path)
return 0;
path = hint->path;
bh = get_last_bh(path);
RFALSE(!bh, "green-4002: Illegal path specified to get_left_neighbor");
- ih = get_ih(path);
+ ih = tp_item_head(path);
pos_in_item = path->pos_in_item;
- item = get_item(path);
+ item = tp_item_body(path);
hint->search_start = bh->b_blocknr;
+ /*
+ * for indirect item: go to left and look for the first non-hole entry
+ * in the indirect item
+ */
if (!hint->formatted_node && is_indirect_le_ih(ih)) {
- /* for indirect item: go to left and look for the first non-hole entry
- in the indirect item */
if (pos_in_item == I_UNFM_NUM(ih))
pos_in_item--;
-// pos_in_item = I_UNFM_NUM (ih) - 1;
while (pos_in_item >= 0) {
int t = get_block_num(item, pos_in_item);
if (t) {
@@ -735,10 +890,12 @@ static int get_left_neighbor(reiserfs_blocknr_hint_t * hint)
return ret;
}
-/* should be, if formatted node, then try to put on first part of the device
- specified as number of percent with mount option device, else try to put
- on last of device. This is not to say it is good code to do so,
- but the effect should be measured. */
+/*
+ * should be, if formatted node, then try to put on first part of the device
+ * specified as number of percent with mount option device, else try to put
+ * on last of device. This is not to say it is good code to do so,
+ * but the effect should be measured.
+ */
static inline void set_border_in_hint(struct super_block *s,
reiserfs_blocknr_hint_t * hint)
{
@@ -864,21 +1021,27 @@ static void determine_search_start(reiserfs_blocknr_hint_t * hint,
set_border_in_hint(s, hint);
#ifdef DISPLACE_NEW_PACKING_LOCALITIES
- /* whenever we create a new directory, we displace it. At first we will
- hash for location, later we might look for a moderately empty place for
- it */
+ /*
+ * whenever we create a new directory, we displace it. At first
+ * we will hash for location, later we might look for a moderately
+ * empty place for it
+ */
if (displacing_new_packing_localities(s)
&& hint->th->displace_new_blocks) {
displace_new_packing_locality(hint);
- /* we do not continue determine_search_start,
- * if new packing locality is being displaced */
+ /*
+ * we do not continue determine_search_start,
+ * if new packing locality is being displaced
+ */
return;
}
#endif
- /* all persons should feel encouraged to add more special cases here and
- * test them */
+ /*
+ * all persons should feel encouraged to add more special cases
+ * here and test them
+ */
if (displacing_large_files(s) && !hint->formatted_node
&& this_blocknr_allocation_would_make_it_a_large_file(hint)) {
@@ -886,8 +1049,10 @@ static void determine_search_start(reiserfs_blocknr_hint_t * hint,
return;
}
- /* if none of our special cases is relevant, use the left neighbor in the
- tree order of the new node we are allocating for */
+ /*
+ * if none of our special cases is relevant, use the left
+ * neighbor in the tree order of the new node we are allocating for
+ */
if (hint->formatted_node && TEST_OPTION(hashed_formatted_nodes, s)) {
hash_formatted_node(hint);
return;
@@ -895,10 +1060,13 @@ static void determine_search_start(reiserfs_blocknr_hint_t * hint,
unfm_hint = get_left_neighbor(hint);
- /* Mimic old block allocator behaviour, that is if VFS allowed for preallocation,
- new blocks are displaced based on directory ID. Also, if suggested search_start
- is less than last preallocated block, we start searching from it, assuming that
- HDD dataflow is faster in forward direction */
+ /*
+ * Mimic old block allocator behaviour, that is if VFS allowed for
+ * preallocation, new blocks are displaced based on directory ID.
+ * Also, if suggested search_start is less than last preallocated
+ * block, we start searching from it, assuming that HDD dataflow
+ * is faster in forward direction
+ */
if (TEST_OPTION(old_way, s)) {
if (!hint->formatted_node) {
if (!reiserfs_hashed_relocation(s))
@@ -927,11 +1095,13 @@ static void determine_search_start(reiserfs_blocknr_hint_t * hint,
TEST_OPTION(old_hashed_relocation, s)) {
old_hashed_relocation(hint);
}
+
/* new_hashed_relocation works with both formatted/unformatted nodes */
if ((!unfm_hint || hint->formatted_node) &&
TEST_OPTION(new_hashed_relocation, s)) {
new_hashed_relocation(hint);
}
+
/* dirid grouping works only on unformatted nodes */
if (!unfm_hint && !hint->formatted_node && TEST_OPTION(dirid_groups, s)) {
dirid_groups(hint);
@@ -969,8 +1139,6 @@ static int determine_prealloc_size(reiserfs_blocknr_hint_t * hint)
return CARRY_ON;
}
-/* XXX I know it could be merged with upper-level function;
- but may be result function would be too complex. */
static inline int allocate_without_wrapping_disk(reiserfs_blocknr_hint_t * hint,
b_blocknr_t * new_blocknrs,
b_blocknr_t start,
@@ -998,7 +1166,10 @@ static inline int allocate_without_wrapping_disk(reiserfs_blocknr_hint_t * hint,
/* do we have something to fill prealloc. array also ? */
if (nr_allocated > 0) {
- /* it means prealloc_size was greater that 0 and we do preallocation */
+ /*
+ * it means prealloc_size was greater that 0 and
+ * we do preallocation
+ */
list_add(&REISERFS_I(hint->inode)->i_prealloc_list,
&SB_JOURNAL(hint->th->t_super)->
j_prealloc_list);
@@ -1020,7 +1191,7 @@ static inline int blocknrs_and_prealloc_arrays_from_search_start
b_blocknr_t finish = SB_BLOCK_COUNT(s) - 1;
int passno = 0;
int nr_allocated = 0;
- int bigalloc = 0;
+ int depth;
determine_prealloc_size(hint);
if (!hint->formatted_node) {
@@ -1030,45 +1201,29 @@ static inline int blocknrs_and_prealloc_arrays_from_search_start
"reiserquota: allocating %d blocks id=%u",
amount_needed, hint->inode->i_uid);
#endif
+ depth = reiserfs_write_unlock_nested(s);
quota_ret =
- DQUOT_ALLOC_BLOCK_NODIRTY(hint->inode, amount_needed);
- if (quota_ret) /* Quota exceeded? */
+ dquot_alloc_block_nodirty(hint->inode, amount_needed);
+ if (quota_ret) { /* Quota exceeded? */
+ reiserfs_write_lock_nested(s, depth);
return QUOTA_EXCEEDED;
+ }
if (hint->preallocate && hint->prealloc_size) {
#ifdef REISERQUOTA_DEBUG
reiserfs_debug(s, REISERFS_DEBUG_CODE,
"reiserquota: allocating (prealloc) %d blocks id=%u",
hint->prealloc_size, hint->inode->i_uid);
#endif
- quota_ret =
- DQUOT_PREALLOC_BLOCK_NODIRTY(hint->inode,
+ quota_ret = dquot_prealloc_block_nodirty(hint->inode,
hint->prealloc_size);
if (quota_ret)
hint->preallocate = hint->prealloc_size = 0;
}
/* for unformatted nodes, force large allocations */
- bigalloc = amount_needed;
+ reiserfs_write_lock_nested(s, depth);
}
do {
- /* in bigalloc mode, nr_allocated should stay zero until
- * the entire allocation is filled
- */
- if (unlikely(bigalloc && nr_allocated)) {
- reiserfs_warning(s, "bigalloc is %d, nr_allocated %d\n",
- bigalloc, nr_allocated);
- /* reset things to a sane value */
- bigalloc = amount_needed - nr_allocated;
- }
- /*
- * try pass 0 and pass 1 looking for a nice big
- * contiguous allocation. Then reset and look
- * for anything you can find.
- */
- if (passno == 2 && bigalloc) {
- passno = 0;
- bigalloc = 0;
- }
switch (passno++) {
case 0: /* Search from hint->search_start to end of disk */
start = hint->search_start;
@@ -1082,7 +1237,8 @@ static inline int blocknrs_and_prealloc_arrays_from_search_start
start = 0;
finish = hint->beg;
break;
- default: /* We've tried searching everywhere, not enough space */
+ default:
+ /* We've tried searching everywhere, not enough space */
/* Free the blocks */
if (!hint->formatted_node) {
#ifdef REISERQUOTA_DEBUG
@@ -1093,7 +1249,12 @@ static inline int blocknrs_and_prealloc_arrays_from_search_start
nr_allocated,
hint->inode->i_uid);
#endif
- DQUOT_FREE_BLOCK_NODIRTY(hint->inode, amount_needed + hint->prealloc_size - nr_allocated); /* Free not allocated blocks */
+ /* Free not allocated blocks */
+ depth = reiserfs_write_unlock_nested(s);
+ dquot_free_block_nodirty(hint->inode,
+ amount_needed + hint->prealloc_size -
+ nr_allocated);
+ reiserfs_write_lock_nested(s, depth);
}
while (nr_allocated--)
reiserfs_free_block(hint->th, hint->inode,
@@ -1106,8 +1267,7 @@ static inline int blocknrs_and_prealloc_arrays_from_search_start
new_blocknrs +
nr_allocated,
start, finish,
- bigalloc ?
- bigalloc : 1,
+ 1,
amount_needed -
nr_allocated,
hint->
@@ -1125,10 +1285,13 @@ static inline int blocknrs_and_prealloc_arrays_from_search_start
REISERFS_I(hint->inode)->i_prealloc_count,
hint->inode->i_uid);
#endif
- DQUOT_FREE_BLOCK_NODIRTY(hint->inode, amount_needed +
+
+ depth = reiserfs_write_unlock_nested(s);
+ dquot_free_block_nodirty(hint->inode, amount_needed +
hint->prealloc_size - nr_allocated -
REISERFS_I(hint->inode)->
i_prealloc_count);
+ reiserfs_write_lock_nested(s, depth);
}
return CARRY_ON;
@@ -1160,8 +1323,11 @@ static int use_preallocated_list_if_available(reiserfs_blocknr_hint_t * hint,
return amount_needed;
}
-int reiserfs_allocate_blocknrs(reiserfs_blocknr_hint_t * hint, b_blocknr_t * new_blocknrs, int amount_needed, int reserved_by_us /* Amount of blocks we have
- already reserved */ )
+int reiserfs_allocate_blocknrs(reiserfs_blocknr_hint_t *hint,
+ b_blocknr_t *new_blocknrs,
+ int amount_needed,
+ /* Amount of blocks we have already reserved */
+ int reserved_by_us)
{
int initial_amount_needed = amount_needed;
int ret;
@@ -1173,15 +1339,21 @@ int reiserfs_allocate_blocknrs(reiserfs_blocknr_hint_t * hint, b_blocknr_t * new
return NO_DISK_SPACE;
/* should this be if !hint->inode && hint->preallocate? */
/* do you mean hint->formatted_node can be removed ? - Zam */
- /* hint->formatted_node cannot be removed because we try to access
- inode information here, and there is often no inode assotiated with
- metadata allocations - green */
+ /*
+ * hint->formatted_node cannot be removed because we try to access
+ * inode information here, and there is often no inode associated with
+ * metadata allocations - green
+ */
if (!hint->formatted_node && hint->preallocate) {
amount_needed = use_preallocated_list_if_available
(hint, new_blocknrs, amount_needed);
- if (amount_needed == 0) /* all blocknrs we need we got from
- prealloc. list */
+
+ /*
+ * We have all the block numbers we need from the
+ * prealloc list
+ */
+ if (amount_needed == 0)
return CARRY_ON;
new_blocknrs += (initial_amount_needed - amount_needed);
}
@@ -1195,10 +1367,12 @@ int reiserfs_allocate_blocknrs(reiserfs_blocknr_hint_t * hint, b_blocknr_t * new
ret = blocknrs_and_prealloc_arrays_from_search_start
(hint, new_blocknrs, amount_needed);
- /* we used prealloc. list to fill (partially) new_blocknrs array. If final allocation fails we
- * need to return blocks back to prealloc. list or just free them. -- Zam (I chose second
- * variant) */
-
+ /*
+ * We used prealloc. list to fill (partially) new_blocknrs array.
+ * If final allocation fails we need to return blocks back to
+ * prealloc. list or just free them. -- Zam (I chose second
+ * variant)
+ */
if (ret != CARRY_ON) {
while (amount_needed++ < initial_amount_needed) {
reiserfs_free_block(hint->th, hint->inode,
@@ -1208,59 +1382,87 @@ int reiserfs_allocate_blocknrs(reiserfs_blocknr_hint_t * hint, b_blocknr_t * new
return ret;
}
-/* These 2 functions are here to provide blocks reservation to the rest of kernel */
-/* Reserve @blocks amount of blocks in fs pointed by @sb. Caller must make sure
- there are actually this much blocks on the FS available */
-void reiserfs_claim_blocks_to_be_allocated(struct super_block *sb, /* super block of
- filesystem where
- blocks should be
- reserved */
- int blocks /* How much to reserve */
- )
+void reiserfs_cache_bitmap_metadata(struct super_block *sb,
+ struct buffer_head *bh,
+ struct reiserfs_bitmap_info *info)
{
+ unsigned long *cur = (unsigned long *)(bh->b_data + bh->b_size);
- /* Fast case, if reservation is zero - exit immediately. */
- if (!blocks)
- return;
+ /* The first bit must ALWAYS be 1 */
+ if (!reiserfs_test_le_bit(0, (unsigned long *)bh->b_data))
+ reiserfs_error(sb, "reiserfs-2025", "bitmap block %lu is "
+ "corrupted: first bit must be 1", bh->b_blocknr);
+
+ info->free_count = 0;
- spin_lock(&REISERFS_SB(sb)->bitmap_lock);
- REISERFS_SB(sb)->reserved_blocks += blocks;
- spin_unlock(&REISERFS_SB(sb)->bitmap_lock);
+ while (--cur >= (unsigned long *)bh->b_data) {
+ /* 0 and ~0 are special, we can optimize for them */
+ if (*cur == 0)
+ info->free_count += BITS_PER_LONG;
+ else if (*cur != ~0L) /* A mix, investigate */
+ info->free_count += BITS_PER_LONG - hweight_long(*cur);
+ }
}
-/* Unreserve @blocks amount of blocks in fs pointed by @sb */
-void reiserfs_release_claimed_blocks(struct super_block *sb, /* super block of
- filesystem where
- blocks should be
- reserved */
- int blocks /* How much to unreserve */
- )
+struct buffer_head *reiserfs_read_bitmap_block(struct super_block *sb,
+ unsigned int bitmap)
{
+ b_blocknr_t block = (sb->s_blocksize << 3) * bitmap;
+ struct reiserfs_bitmap_info *info = SB_AP_BITMAP(sb) + bitmap;
+ struct buffer_head *bh;
- /* Fast case, if unreservation is zero - exit immediately. */
- if (!blocks)
- return;
+ /*
+ * Way old format filesystems had the bitmaps packed up front.
+ * I doubt there are any of these left, but just in case...
+ */
+ if (unlikely(test_bit(REISERFS_OLD_FORMAT,
+ &REISERFS_SB(sb)->s_properties)))
+ block = REISERFS_SB(sb)->s_sbh->b_blocknr + 1 + bitmap;
+ else if (bitmap == 0)
+ block = (REISERFS_DISK_OFFSET_IN_BYTES >> sb->s_blocksize_bits) + 1;
+
+ bh = sb_bread(sb, block);
+ if (bh == NULL)
+ reiserfs_warning(sb, "sh-2029: %s: bitmap block (#%u) "
+ "reading failed", __func__, block);
+ else {
+ if (buffer_locked(bh)) {
+ int depth;
+ PROC_INFO_INC(sb, scan_bitmap.wait);
+ depth = reiserfs_write_unlock_nested(sb);
+ __wait_on_buffer(bh);
+ reiserfs_write_lock_nested(sb, depth);
+ }
+ BUG_ON(!buffer_uptodate(bh));
+ BUG_ON(atomic_read(&bh->b_count) == 0);
+
+ if (info->free_count == UINT_MAX)
+ reiserfs_cache_bitmap_metadata(sb, bh, info);
+ }
- spin_lock(&REISERFS_SB(sb)->bitmap_lock);
- REISERFS_SB(sb)->reserved_blocks -= blocks;
- spin_unlock(&REISERFS_SB(sb)->bitmap_lock);
- RFALSE(REISERFS_SB(sb)->reserved_blocks < 0,
- "amount of blocks reserved became zero?");
+ return bh;
}
-/* This function estimates how much pages we will be able to write to FS
- used for reiserfs_file_write() purposes for now. */
-int reiserfs_can_fit_pages(struct super_block *sb /* superblock of filesystem
- to estimate space */ )
+int reiserfs_init_bitmap_cache(struct super_block *sb)
{
- int space;
+ struct reiserfs_bitmap_info *bitmap;
+ unsigned int bmap_nr = reiserfs_bmap_count(sb);
+
+ bitmap = vmalloc(sizeof(*bitmap) * bmap_nr);
+ if (bitmap == NULL)
+ return -ENOMEM;
+
+ memset(bitmap, 0xff, sizeof(*bitmap) * bmap_nr);
+
+ SB_AP_BITMAP(sb) = bitmap;
- spin_lock(&REISERFS_SB(sb)->bitmap_lock);
- space =
- (SB_FREE_BLOCKS(sb) -
- REISERFS_SB(sb)->reserved_blocks) >> (PAGE_CACHE_SHIFT -
- sb->s_blocksize_bits);
- spin_unlock(&REISERFS_SB(sb)->bitmap_lock);
+ return 0;
+}
- return space > 0 ? space : 0;
+void reiserfs_free_bitmap_cache(struct super_block *sb)
+{
+ if (SB_AP_BITMAP(sb)) {
+ vfree(SB_AP_BITMAP(sb));
+ SB_AP_BITMAP(sb) = NULL;
+ }
}