+++ /dev/null
-Index: linux-stage/fs/ext4/ext4.h
-===================================================================
---- linux-stage.orig/fs/ext4/ext4.h
-+++ linux-stage/fs/ext4/ext4.h
-@@ -1136,11 +1136,14 @@ struct ext4_sb_info {
-
- /* tunables */
- unsigned long s_stripe;
-- unsigned int s_mb_stream_request;
-+ unsigned long s_mb_small_req;
-+ unsigned long s_mb_large_req;
- unsigned int s_mb_max_to_scan;
- unsigned int s_mb_min_to_scan;
- unsigned int s_mb_stats;
- unsigned int s_mb_order2_reqs;
-+ unsigned long *s_mb_prealloc_table;
-+ unsigned long s_mb_prealloc_table_size;
- unsigned int s_mb_group_prealloc;
- unsigned int s_max_writeback_mb_bump;
- /* where last allocation was done - for stream allocation */
-Index: linux-stage/fs/ext4/mballoc.c
-===================================================================
---- linux-stage.orig/fs/ext4/mballoc.c
-+++ linux-stage/fs/ext4/mballoc.c
-@@ -1838,6 +1838,26 @@ void ext4_mb_complex_scan_group(struct e
- ext4_mb_check_limits(ac, e4b, 1);
- }
-
-+static int ext4_mb_prealloc_table_add(struct ext4_sb_info *sbi, int value)
-+{
-+ int i;
-+
-+ if (value > (sbi->s_blocks_per_group - 1 - 1 - sbi->s_itb_per_group))
-+ return -1;
-+
-+ for (i = 0; i < sbi->s_mb_prealloc_table_size; i++) {
-+ if (sbi->s_mb_prealloc_table[i] == 0) {
-+ sbi->s_mb_prealloc_table[i] = value;
-+ return 0;
-+ }
-+
-+ /* they should add values in order */
-+ if (value <= sbi->s_mb_prealloc_table[i])
-+ return -1;
-+ }
-+ return -1;
-+}
-+
- /*
- * This is a special case for storages like raid5
- * we try to find stripe-aligned chunks for stripe-size requests
-@@ -2155,6 +2175,82 @@ static const struct seq_operations ext4_
- .show = ext4_mb_seq_groups_show,
- };
-
-+#define EXT4_MB_PREALLOC_TABLE "prealloc_table"
-+
-+static int ext4_mb_prealloc_table_proc_read(char *page, char **start, off_t off,
-+ int count, int *eof, void *data)
-+{
-+ struct ext4_sb_info *sbi = data;
-+ int len = 0;
-+ int i;
-+
-+ *eof = 1;
-+ if (off != 0)
-+ return 0;
-+
-+ for (i = 0; i < sbi->s_mb_prealloc_table_size; i++)
-+ len += sprintf(page + len, "%ld ",
-+ sbi->s_mb_prealloc_table[i]);
-+ len += sprintf(page + len, "\n");
-+
-+ *start = page;
-+ return len;
-+}
-+
-+static int ext4_mb_prealloc_table_proc_write(struct file *file,
-+ const char __user *buf,
-+ unsigned long cnt, void *data)
-+{
-+ struct ext4_sb_info *sbi = data;
-+ unsigned long value;
-+ unsigned long prev = 0;
-+ char str[128];
-+ char *cur;
-+ char *end;
-+ unsigned long *new_table;
-+ int num = 0;
-+ int i = 0;
-+
-+ if (cnt >= sizeof(str))
-+ return -EINVAL;
-+ if (copy_from_user(str, buf, cnt))
-+ return -EFAULT;
-+
-+ num = 0;
-+ cur = str;
-+ end = str + cnt;
-+ while (cur < end) {
-+ while ((cur < end) && (*cur == ' ')) cur++;
-+ value = simple_strtol(cur, &cur, 0);
-+ if (value == 0)
-+ break;
-+ if (value <= prev)
-+ return -EINVAL;
-+ prev = value;
-+ num++;
-+ }
-+
-+ new_table = kmalloc(num * sizeof(*new_table), GFP_KERNEL);
-+ if (new_table == NULL)
-+ return -ENOMEM;
-+ kfree(sbi->s_mb_prealloc_table);
-+ memset(new_table, 0, num * sizeof(*new_table));
-+ sbi->s_mb_prealloc_table = new_table;
-+ sbi->s_mb_prealloc_table_size = num;
-+ cur = str;
-+ end = str + cnt;
-+ while (cur < end && i < num) {
-+ while ((cur < end) && (*cur == ' ')) cur++;
-+ value = simple_strtol(cur, &cur, 0);
-+ if (ext4_mb_prealloc_table_add(sbi, value) == 0)
-+ ++i;
-+ }
-+ if (i != num)
-+ sbi->s_mb_prealloc_table_size = i;
-+
-+ return cnt;
-+}
-+
- static int ext4_mb_seq_groups_open(struct inode *inode, struct file *file)
- {
- struct super_block *sb = PDE(inode)->data;
-@@ -2346,7 +2442,7 @@ err_freesgi:
- int ext4_mb_init(struct super_block *sb, int needs_recovery)
- {
- struct ext4_sb_info *sbi = EXT4_SB(sb);
-- unsigned i, j;
-+ unsigned i, j, k, l;
- unsigned offset;
- unsigned max;
- int ret;
-@@ -2380,26 +2476,61 @@ int ext4_mb_init(struct super_block *sb,
- i++;
- } while (i <= sb->s_blocksize_bits + 1);
-
-- /* init file for buddy data */
-- ret = ext4_mb_init_backend(sb);
-- if (ret != 0) {
-- kfree(sbi->s_mb_offsets);
-- kfree(sbi->s_mb_maxs);
-- return ret;
-- }
--
- spin_lock_init(&sbi->s_md_lock);
- spin_lock_init(&sbi->s_bal_lock);
-
- sbi->s_mb_max_to_scan = MB_DEFAULT_MAX_TO_SCAN;
- sbi->s_mb_min_to_scan = MB_DEFAULT_MIN_TO_SCAN;
- sbi->s_mb_stats = MB_DEFAULT_STATS;
-- sbi->s_mb_stream_request = MB_DEFAULT_STREAM_THRESHOLD;
- sbi->s_mb_order2_reqs = MB_DEFAULT_ORDER2_REQS;
-- sbi->s_mb_group_prealloc = MB_DEFAULT_GROUP_PREALLOC;
-+
-+ if (sbi->s_stripe == 0) {
-+ sbi->s_mb_prealloc_table_size = 10;
-+ i = sbi->s_mb_prealloc_table_size * sizeof(unsigned long);
-+ sbi->s_mb_prealloc_table = kmalloc(i, GFP_NOFS);
-+ if (sbi->s_mb_prealloc_table == NULL) {
-+ kfree(sbi->s_mb_offsets);
-+ kfree(sbi->s_mb_maxs);
-+ return -ENOMEM;
-+ }
-+ memset(sbi->s_mb_prealloc_table, 0, i);
-+
-+ for (k = 0, l = 4; k <= 9; ++k, l *= 2) {
-+ if (ext4_mb_prealloc_table_add(sbi, l) < 0) {
-+ sbi->s_mb_prealloc_table_size = k;
-+ break;
-+ }
-+ }
-+
-+ sbi->s_mb_small_req = 256;
-+ sbi->s_mb_large_req = 1024;
-+ sbi->s_mb_group_prealloc = 512;
-+ } else {
-+ sbi->s_mb_prealloc_table_size = 3;
-+ i = sbi->s_mb_prealloc_table_size * sizeof(unsigned long);
-+ sbi->s_mb_prealloc_table = kmalloc(i, GFP_NOFS);
-+ if (sbi->s_mb_prealloc_table == NULL) {
-+ kfree(sbi->s_mb_offsets);
-+ kfree(sbi->s_mb_maxs);
-+ return -ENOMEM;
-+ }
-+ memset(sbi->s_mb_prealloc_table, 0, i);
-+
-+ for (k = 0, l = sbi->s_stripe; k <= 2; ++k, l *= 2) {
-+ if (ext4_mb_prealloc_table_add(sbi, l) < 0) {
-+ sbi->s_mb_prealloc_table_size = k;
-+ break;
-+ }
-+ }
-+
-+ sbi->s_mb_small_req = sbi->s_stripe;
-+ sbi->s_mb_large_req = sbi->s_stripe * 8;
-+ sbi->s_mb_group_prealloc = sbi->s_stripe * 4;
-+ }
-
- sbi->s_locality_groups = alloc_percpu(struct ext4_locality_group);
- if (sbi->s_locality_groups == NULL) {
-+ kfree(sbi->s_mb_prealloc_table);
- kfree(sbi->s_mb_offsets);
- kfree(sbi->s_mb_maxs);
- return -ENOMEM;
-@@ -2413,9 +2544,27 @@ int ext4_mb_init(struct super_block *sb,
- spin_lock_init(&lg->lg_prealloc_lock);
- }
-
-- if (sbi->s_proc)
-+ /* init file for buddy data */
-+ ret = ext4_mb_init_backend(sb);
-+ if (ret != 0) {
-+ kfree(sbi->s_mb_prealloc_table);
-+ kfree(sbi->s_mb_offsets);
-+ kfree(sbi->s_mb_maxs);
-+ return ret;
-+ }
-+
-+ if (sbi->s_proc) {
-+ struct proc_dir_entry *p;
- proc_create_data("mb_groups", S_IRUGO, sbi->s_proc,
- &ext4_mb_seq_groups_fops, sb);
-+ p = create_proc_entry(EXT4_MB_PREALLOC_TABLE, S_IFREG |
-+ S_IRUGO | S_IWUSR, sbi->s_proc);
-+ if (p) {
-+ p->data = sbi;
-+ p->read_proc = ext4_mb_prealloc_table_proc_read;
-+ p->write_proc = ext4_mb_prealloc_table_proc_write;
-+ }
-+ }
-
- if (sbi->s_journal)
- sbi->s_journal->j_commit_callback = release_blocks_on_commit;
-@@ -2448,8 +2597,10 @@ int ext4_mb_release(struct super_block *
- struct ext4_group_info *grinfo;
- struct ext4_sb_info *sbi = EXT4_SB(sb);
-
-- if (sbi->s_proc)
-+ if (sbi->s_proc) {
- remove_proc_entry("mb_groups", sbi->s_proc);
-+ remove_proc_entry(EXT4_MB_PREALLOC_TABLE, sbi->s_proc);
-+ }
-
- if (sbi->s_group_info) {
- for (i = 0; i < ngroups; i++) {
-@@ -2469,6 +2620,7 @@ int ext4_mb_release(struct super_block *
- kfree(sbi->s_group_info[i]);
- ext4_kvfree(sbi->s_group_info);
- }
-+ kfree(sbi->s_mb_prealloc_table);
- kfree(sbi->s_mb_offsets);
- kfree(sbi->s_mb_maxs);
- if (sbi->s_buddy_cache)
-@@ -2798,11 +2950,12 @@ static noinline_for_stack void
- ext4_mb_normalize_request(struct ext4_allocation_context *ac,
- struct ext4_allocation_request *ar)
- {
-- int bsbits, max;
-+ int bsbits, i, wind;
- ext4_lblk_t end;
-- loff_t size, orig_size, start_off;
-+ loff_t size, orig_size;
- ext4_lblk_t start, orig_start;
- struct ext4_inode_info *ei = EXT4_I(ac->ac_inode);
-+ struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb);
- struct ext4_prealloc_space *pa;
-
- /* do normalize only data requests, metadata requests
-@@ -2832,49 +2985,35 @@ ext4_mb_normalize_request(struct ext4_al
- size = size << bsbits;
- if (size < i_size_read(ac->ac_inode))
- size = i_size_read(ac->ac_inode);
-+ size = (size + ac->ac_sb->s_blocksize - 1) >> bsbits;
-
-- /* max size of free chunks */
-- max = 2 << bsbits;
-+ start = wind = 0;
-
--#define NRL_CHECK_SIZE(req, size, max, chunk_size) \
-- (req <= (size) || max <= (chunk_size))
-+ /* let's choose preallocation window depending on file size */
-+ for (i = 0; i < sbi->s_mb_prealloc_table_size; i++) {
-+ if (size <= sbi->s_mb_prealloc_table[i]) {
-+ wind = sbi->s_mb_prealloc_table[i];
-+ break;
-+ }
-+ }
-+ size = wind;
-
-- /* first, try to predict filesize */
-- /* XXX: should this table be tunable? */
-- start_off = 0;
-- if (size <= 16 * 1024) {
-- size = 16 * 1024;
-- } else if (size <= 32 * 1024) {
-- size = 32 * 1024;
-- } else if (size <= 64 * 1024) {
-- size = 64 * 1024;
-- } else if (size <= 128 * 1024) {
-- size = 128 * 1024;
-- } else if (size <= 256 * 1024) {
-- size = 256 * 1024;
-- } else if (size <= 512 * 1024) {
-- size = 512 * 1024;
-- } else if (size <= 1024 * 1024) {
-- size = 1024 * 1024;
-- } else if (NRL_CHECK_SIZE(size, 4 * 1024 * 1024, max, 2 * 1024)) {
-- start_off = ((loff_t)ac->ac_o_ex.fe_logical >>
-- (21 - bsbits)) << 21;
-- size = 2 * 1024 * 1024;
-- } else if (NRL_CHECK_SIZE(size, 8 * 1024 * 1024, max, 4 * 1024)) {
-- start_off = ((loff_t)ac->ac_o_ex.fe_logical >>
-- (22 - bsbits)) << 22;
-- size = 4 * 1024 * 1024;
-- } else if (NRL_CHECK_SIZE(ac->ac_o_ex.fe_len,
-- (8<<20)>>bsbits, max, 8 * 1024)) {
-- start_off = ((loff_t)ac->ac_o_ex.fe_logical >>
-- (23 - bsbits)) << 23;
-- size = 8 * 1024 * 1024;
-- } else {
-- start_off = (loff_t)ac->ac_o_ex.fe_logical << bsbits;
-- size = ac->ac_o_ex.fe_len << bsbits;
-+ if (wind == 0) {
-+ __u64 tstart, tend;
-+ /* file is quite large, we now preallocate with
-+ * the biggest configured window with regart to
-+ * logical offset */
-+ wind = sbi->s_mb_prealloc_table[i - 1];
-+ tstart = ac->ac_o_ex.fe_logical;
-+ do_div(tstart, wind);
-+ start = tstart * wind;
-+ tend = ac->ac_o_ex.fe_logical + ac->ac_o_ex.fe_len - 1;
-+ do_div(tend, wind);
-+ tend = tend * wind + wind;
-+ size = tend - start;
- }
-- orig_size = size = size >> bsbits;
-- orig_start = start = start_off >> bsbits;
-+ orig_size = size;
-+ orig_start = start;
-
- /* don't cover already allocated blocks in selected range */
- if (ar->pleft && start <= ar->lleft) {
-@@ -2946,7 +3085,6 @@ ext4_mb_normalize_request(struct ext4_al
- }
- BUG_ON(start + size <= ac->ac_o_ex.fe_logical &&
- start > ac->ac_o_ex.fe_logical);
-- BUG_ON(size <= 0 || size > EXT4_BLOCKS_PER_GROUP(ac->ac_sb));
-
- /* now prepare goal request */
-
-@@ -3930,11 +4068,19 @@ static void ext4_mb_group_or_file(struct
-
- /* don't use group allocation for large files */
- size = max(size, isize);
-- if (size > sbi->s_mb_stream_request) {
-+ if ((ac->ac_o_ex.fe_len >= sbi->s_mb_small_req) ||
-+ (size >= sbi->s_mb_large_req)) {
- ac->ac_flags |= EXT4_MB_STREAM_ALLOC;
- return;
- }
-
-+ /*
-+ * request is so large that we don't care about
-+ * streaming - it overweights any possible seek
-+ */
-+ if (ac->ac_o_ex.fe_len >= sbi->s_mb_large_req)
-+ return;
-+
- BUG_ON(ac->ac_lg != NULL);
- /*
- * locality group prealloc space are per cpu. The reason for having
-Index: linux-stage/fs/ext4/super.c
-===================================================================
---- linux-stage.orig/fs/ext4/super.c
-+++ linux-stage/fs/ext4/super.c
-@@ -2377,7 +2377,8 @@ EXT4_RW_ATTR_SBI_UI(mb_stats, s_mb_stats
- EXT4_RW_ATTR_SBI_UI(mb_max_to_scan, s_mb_max_to_scan);
- EXT4_RW_ATTR_SBI_UI(mb_min_to_scan, s_mb_min_to_scan);
- EXT4_RW_ATTR_SBI_UI(mb_order2_req, s_mb_order2_reqs);
--EXT4_RW_ATTR_SBI_UI(mb_stream_req, s_mb_stream_request);
-+EXT4_RW_ATTR_SBI_UI(mb_small_req, s_mb_small_req);
-+EXT4_RW_ATTR_SBI_UI(mb_large_req, s_mb_large_req);
- EXT4_RW_ATTR_SBI_UI(mb_group_prealloc, s_mb_group_prealloc);
- EXT4_RW_ATTR_SBI_UI(max_writeback_mb_bump, s_max_writeback_mb_bump);
-
-@@ -2391,7 +2392,8 @@ static struct attribute *ext4_attrs[] =
- ATTR_LIST(mb_max_to_scan),
- ATTR_LIST(mb_min_to_scan),
- ATTR_LIST(mb_order2_req),
-- ATTR_LIST(mb_stream_req),
-+ ATTR_LIST(mb_small_req),
-+ ATTR_LIST(mb_large_req),
- ATTR_LIST(mb_group_prealloc),
- ATTR_LIST(max_writeback_mb_bump),
- NULL,
-Index: linux-stage/fs/ext4/inode.c
-===================================================================
---- linux-stage.orig/fs/ext4/inode.c
-+++ linux-stage/fs/ext4/inode.c
-@@ -3070,6 +3070,11 @@ static int ext4_da_writepages(struct add
- if (unlikely(sbi->s_mount_flags & EXT4_MF_FS_ABORTED))
- return -EROFS;
-
-+ if (wbc->nr_to_write < sbi->s_mb_small_req) {
-+ nr_to_writebump = sbi->s_mb_small_req - wbc->nr_to_write;
-+ wbc->nr_to_write = sbi->s_mb_small_req;
-+ }
-+
- if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX)
- range_whole = 1;
-
git://git.whamcloud.com / fs / lustre-release.git / blobdiff