summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
Diffstat
-rw-r--r--Documentation/ABI/testing/sysfs-fs-f2fs4
-rw-r--r--Documentation/filesystems/f2fs.rst6
-rw-r--r--fs/f2fs/checkpoint.c15
-rw-r--r--fs/f2fs/compress.c89
-rw-r--r--fs/f2fs/data.c93
-rw-r--r--fs/f2fs/debug.c64
-rw-r--r--fs/f2fs/dir.c2
-rw-r--r--fs/f2fs/extent_cache.c18
-rw-r--r--fs/f2fs/f2fs.h81
-rw-r--r--fs/f2fs/file.c264
-rw-r--r--fs/f2fs/gc.c75
-rw-r--r--fs/f2fs/inline.c21
-rw-r--r--fs/f2fs/inode.c4
-rw-r--r--fs/f2fs/namei.c18
-rw-r--r--fs/f2fs/node.c38
-rw-r--r--fs/f2fs/recovery.c12
-rw-r--r--fs/f2fs/segment.c129
-rw-r--r--fs/f2fs/segment.h10
-rw-r--r--fs/f2fs/super.c68
-rw-r--r--fs/f2fs/sysfs.c23
-rw-r--r--fs/f2fs/verity.c6
-rw-r--r--fs/f2fs/xattr.c4
-rw-r--r--include/trace/events/f2fs.h63
23 files changed, 815 insertions, 292 deletions
diff --git a/Documentation/ABI/testing/sysfs-fs-f2fs b/Documentation/ABI/testing/sysfs-fs-f2fs
index 4bb93a06d8ab..7f730c4c8df2 100644
--- a/Documentation/ABI/testing/sysfs-fs-f2fs
+++ b/Documentation/ABI/testing/sysfs-fs-f2fs
@@ -229,7 +229,9 @@ Date: August 2017
Contact: "Jaegeuk Kim" <jaegeuk@kernel.org>
Description: Do background GC agressively when set. When gc_urgent = 1,
background thread starts to do GC by given gc_urgent_sleep_time
- interval. It is set to 0 by default.
+ interval. When gc_urgent = 2, F2FS will lower the bar of
+ checking idle in order to process outstanding discard commands
+ and GC a little bit aggressively. It is set to 0 by default.
What: /sys/fs/f2fs/<disk>/gc_urgent_sleep_time
Date: August 2017
diff --git a/Documentation/filesystems/f2fs.rst b/Documentation/filesystems/f2fs.rst
index a11d329542f9..ec8d99703ecb 100644
--- a/Documentation/filesystems/f2fs.rst
+++ b/Documentation/filesystems/f2fs.rst
@@ -258,6 +258,8 @@ compress_extension=%s Support adding specified extension, so that f2fs can enab
on compression extension list and enable compression on
these file by default rather than to enable it via ioctl.
For other files, we can still enable compression via ioctl.
+ Note that, there is one reserved special extension '*', it
+ can be set to enable compression for all files.
inlinecrypt When possible, encrypt/decrypt the contents of encrypted
files using the blk-crypto framework rather than
filesystem-layer encryption. This allows the use of
@@ -743,8 +745,8 @@ Compression implementation
- In order to eliminate write amplification during overwrite, F2FS only
support compression on write-once file, data can be compressed only when
- all logical blocks in file are valid and cluster compress ratio is lower
- than specified threshold.
+ all logical blocks in cluster contain valid data and compress ratio of
+ cluster data is lower than specified threshold.
- To enable compression on regular inode, there are three ways:
diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index 236064930251..ff807e14c891 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -523,7 +523,7 @@ void f2fs_remove_ino_entry(struct f2fs_sb_info *sbi, nid_t ino, int type)
__remove_ino_entry(sbi, ino, type);
}
-/* mode should be APPEND_INO or UPDATE_INO */
+/* mode should be APPEND_INO, UPDATE_INO or TRANS_DIR_INO */
bool f2fs_exist_written_data(struct f2fs_sb_info *sbi, nid_t ino, int mode)
{
struct inode_management *im = &sbi->im[mode];
@@ -1258,8 +1258,6 @@ void f2fs_wait_on_all_pages(struct f2fs_sb_info *sbi, int type)
DEFINE_WAIT(wait);
for (;;) {
- prepare_to_wait(&sbi->cp_wait, &wait, TASK_UNINTERRUPTIBLE);
-
if (!get_pages(sbi, type))
break;
@@ -1269,6 +1267,10 @@ void f2fs_wait_on_all_pages(struct f2fs_sb_info *sbi, int type)
if (type == F2FS_DIRTY_META)
f2fs_sync_meta_pages(sbi, META, LONG_MAX,
FS_CP_META_IO);
+ else if (type == F2FS_WB_CP_DATA)
+ f2fs_submit_merged_write(sbi, DATA);
+
+ prepare_to_wait(&sbi->cp_wait, &wait, TASK_UNINTERRUPTIBLE);
io_schedule_timeout(DEFAULT_IO_TIMEOUT);
}
finish_wait(&sbi->cp_wait, &wait);
@@ -1415,7 +1417,7 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
curseg_alloc_type(sbi, i + CURSEG_HOT_DATA);
}
- /* 2 cp + n data seg summary + orphan inode blocks */
+ /* 2 cp + n data seg summary + orphan inode blocks */
data_sum_blocks = f2fs_npages_for_summary_flush(sbi, false);
spin_lock_irqsave(&sbi->cp_lock, flags);
if (data_sum_blocks < NR_CURSEG_DATA_TYPE)
@@ -1515,9 +1517,10 @@ static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
/*
* invalidate intermediate page cache borrowed from meta inode which are
- * used for migration of encrypted or verity inode's blocks.
+ * used for migration of encrypted, verity or compressed inode's blocks.
*/
- if (f2fs_sb_has_encrypt(sbi) || f2fs_sb_has_verity(sbi))
+ if (f2fs_sb_has_encrypt(sbi) || f2fs_sb_has_verity(sbi) ||
+ f2fs_sb_has_compression(sbi))
invalidate_mapping_pages(META_MAPPING(sbi),
MAIN_BLKADDR(sbi), MAX_BLKADDR(sbi) - 1);
diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c
index 29e50fbe7eca..1dfb126a0cb2 100644
--- a/fs/f2fs/compress.c
+++ b/fs/f2fs/compress.c
@@ -49,6 +49,13 @@ bool f2fs_is_compressed_page(struct page *page)
return false;
if (IS_ATOMIC_WRITTEN_PAGE(page) || IS_DUMMY_WRITTEN_PAGE(page))
return false;
+ /*
+ * page->private may be set with pid.
+ * pid_max is enough to check if it is traced.
+ */
+ if (IS_IO_TRACED_PAGE(page))
+ return false;
+
f2fs_bug_on(F2FS_M_SB(page->mapping),
*((u32 *)page_private(page)) != F2FS_COMPRESSED_PAGE_MAGIC);
return true;
@@ -506,7 +513,7 @@ bool f2fs_is_compress_backend_ready(struct inode *inode)
return f2fs_cops[F2FS_I(inode)->i_compress_algorithm];
}
-static mempool_t *compress_page_pool = NULL;
+static mempool_t *compress_page_pool;
static int num_compress_pages = 512;
module_param(num_compress_pages, uint, 0444);
MODULE_PARM_DESC(num_compress_pages,
@@ -663,6 +670,7 @@ void f2fs_decompress_pages(struct bio *bio, struct page *page, bool verity)
const struct f2fs_compress_ops *cops =
f2fs_cops[fi->i_compress_algorithm];
int ret;
+ int i;
dec_page_count(sbi, F2FS_RD_DATA);
@@ -681,6 +689,26 @@ void f2fs_decompress_pages(struct bio *bio, struct page *page, bool verity)
goto out_free_dic;
}
+ dic->tpages = f2fs_kzalloc(sbi, sizeof(struct page *) *
+ dic->cluster_size, GFP_NOFS);
+ if (!dic->tpages) {
+ ret = -ENOMEM;
+ goto out_free_dic;
+ }
+
+ for (i = 0; i < dic->cluster_size; i++) {
+ if (dic->rpages[i]) {
+ dic->tpages[i] = dic->rpages[i];
+ continue;
+ }
+
+ dic->tpages[i] = f2fs_compress_alloc_page();
+ if (!dic->tpages[i]) {
+ ret = -ENOMEM;
+ goto out_free_dic;
+ }
+ }
+
if (cops->init_decompress_ctx) {
ret = cops->init_decompress_ctx(dic);
if (ret)
@@ -821,7 +849,7 @@ static int f2fs_compressed_blocks(struct compress_ctx *cc)
}
/* return # of valid blocks in compressed cluster */
-static int f2fs_cluster_blocks(struct compress_ctx *cc, bool compr)
+static int f2fs_cluster_blocks(struct compress_ctx *cc)
{
return __f2fs_cluster_blocks(cc, false);
}
@@ -835,7 +863,7 @@ int f2fs_is_compressed_cluster(struct inode *inode, pgoff_t index)
.cluster_idx = index >> F2FS_I(inode)->i_log_cluster_size,
};
- return f2fs_cluster_blocks(&cc, false);
+ return f2fs_cluster_blocks(&cc);
}
static bool cluster_may_compress(struct compress_ctx *cc)
@@ -886,7 +914,7 @@ static int prepare_compress_overwrite(struct compress_ctx *cc,
bool prealloc;
retry:
- ret = f2fs_cluster_blocks(cc, false);
+ ret = f2fs_cluster_blocks(cc);
if (ret <= 0)
return ret;
@@ -949,7 +977,7 @@ retry:
}
if (prealloc) {
- __do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, true);
+ f2fs_do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, true);
set_new_dnode(&dn, cc->inode, NULL, NULL, 0);
@@ -964,7 +992,7 @@ retry:
break;
}
- __do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, false);
+ f2fs_do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, false);
}
if (likely(!ret)) {
@@ -1096,8 +1124,16 @@ static int f2fs_write_compressed_pages(struct compress_ctx *cc,
loff_t psize;
int i, err;
- if (!IS_NOQUOTA(inode) && !f2fs_trylock_op(sbi))
+ if (IS_NOQUOTA(inode)) {
+ /*
+ * We need to wait for node_write to avoid block allocation during
+ * checkpoint. This can only happen to quota writes which can cause
+ * the below discard race condition.
+ */
+ down_read(&sbi->node_write);
+ } else if (!f2fs_trylock_op(sbi)) {
return -EAGAIN;
+ }
set_new_dnode(&dn, cc->inode, NULL, NULL, 0);
@@ -1137,6 +1173,13 @@ static int f2fs_write_compressed_pages(struct compress_ctx *cc,
f2fs_set_compressed_page(cc->cpages[i], inode,
cc->rpages[i + 1]->index, cic);
fio.compressed_page = cc->cpages[i];
+
+ fio.old_blkaddr = data_blkaddr(dn.inode, dn.node_page,
+ dn.ofs_in_node + i + 1);
+
+ /* wait for GCed page writeback via META_MAPPING */
+ f2fs_wait_on_block_writeback(inode, fio.old_blkaddr);
+
if (fio.encrypted) {
fio.page = cc->rpages[i + 1];
err = f2fs_encrypt_one_page(&fio);
@@ -1203,7 +1246,9 @@ unlock_continue:
set_inode_flag(inode, FI_FIRST_BLOCK_WRITTEN);
f2fs_put_dnode(&dn);
- if (!IS_NOQUOTA(inode))
+ if (IS_NOQUOTA(inode))
+ up_read(&sbi->node_write);
+ else
f2fs_unlock_op(sbi);
spin_lock(&fi->i_size_lock);
@@ -1230,7 +1275,9 @@ out_put_cic:
out_put_dnode:
f2fs_put_dnode(&dn);
out_unlock_op:
- if (!IS_NOQUOTA(inode))
+ if (IS_NOQUOTA(inode))
+ up_read(&sbi->node_write);
+ else
f2fs_unlock_op(sbi);
return -EAGAIN;
}
@@ -1310,6 +1357,12 @@ retry_write:
congestion_wait(BLK_RW_ASYNC,
DEFAULT_IO_TIMEOUT);
lock_page(cc->rpages[i]);
+
+ if (!PageDirty(cc->rpages[i])) {
+ unlock_page(cc->rpages[i]);
+ continue;
+ }
+
clear_page_dirty_for_io(cc->rpages[i]);
goto retry_write;
}
@@ -1353,6 +1406,8 @@ int f2fs_write_multi_pages(struct compress_ctx *cc,
err = f2fs_write_compressed_pages(cc, submitted,
wbc, io_type);
cops->destroy_compress_ctx(cc);
+ kfree(cc->cpages);
+ cc->cpages = NULL;
if (!err)
return 0;
f2fs_bug_on(F2FS_I_SB(cc->inode), err != -EAGAIN);
@@ -1415,22 +1470,6 @@ struct decompress_io_ctx *f2fs_alloc_dic(struct compress_ctx *cc)
dic->cpages[i] = page;
}
- dic->tpages = f2fs_kzalloc(sbi, sizeof(struct page *) *
- dic->cluster_size, GFP_NOFS);
- if (!dic->tpages)
- goto out_free;
-
- for (i = 0; i < dic->cluster_size; i++) {
- if (cc->rpages[i]) {
- dic->tpages[i] = cc->rpages[i];
- continue;
- }
-
- dic->tpages[i] = f2fs_compress_alloc_page();
- if (!dic->tpages[i])
- goto out_free;
- }
-
return dic;
out_free:
diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 5f527073143e..ed2bca0fce92 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -87,7 +87,7 @@ static bool __is_cp_guaranteed(struct page *page)
sbi = F2FS_I_SB(inode);
if (inode->i_ino == F2FS_META_INO(sbi) ||
- inode->i_ino == F2FS_NODE_INO(sbi) ||
+ inode->i_ino == F2FS_NODE_INO(sbi) ||
S_ISDIR(inode->i_mode) ||
(S_ISREG(inode->i_mode) &&
(f2fs_is_atomic_file(inode) || IS_NOQUOTA(inode))) ||
@@ -1073,12 +1073,13 @@ static void f2fs_release_read_bio(struct bio *bio)
/* This can handle encryption stuffs */
static int f2fs_submit_page_read(struct inode *inode, struct page *page,
- block_t blkaddr, bool for_write)
+ block_t blkaddr, int op_flags, bool for_write)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
struct bio *bio;
- bio = f2fs_grab_read_bio(inode, blkaddr, 1, 0, page->index, for_write);
+ bio = f2fs_grab_read_bio(inode, blkaddr, 1, op_flags,
+ page->index, for_write);
if (IS_ERR(bio))
return PTR_ERR(bio);
@@ -1193,7 +1194,7 @@ int f2fs_reserve_block(struct dnode_of_data *dn, pgoff_t index)
int f2fs_get_block(struct dnode_of_data *dn, pgoff_t index)
{
- struct extent_info ei = {0,0,0};
+ struct extent_info ei = {0, 0, 0};
struct inode *inode = dn->inode;
if (f2fs_lookup_extent_cache(inode, index, &ei)) {
@@ -1265,7 +1266,8 @@ got_it:
return page;
}
- err = f2fs_submit_page_read(inode, page, dn.data_blkaddr, for_write);
+ err = f2fs_submit_page_read(inode, page, dn.data_blkaddr,
+ op_flags, for_write);
if (err)
goto put_err;
return page;
@@ -1414,7 +1416,7 @@ alloc:
set_summary(&sum, dn->nid, dn->ofs_in_node, ni.version);
old_blkaddr = dn->data_blkaddr;
f2fs_allocate_data_block(sbi, NULL, old_blkaddr, &dn->data_blkaddr,
- &sum, seg_type, NULL, false);
+ &sum, seg_type, NULL);
if (GET_SEGNO(sbi, old_blkaddr) != NULL_SEGNO)
invalidate_mapping_pages(META_MAPPING(sbi),
old_blkaddr, old_blkaddr);
@@ -1474,7 +1476,7 @@ map_blocks:
return err;
}
-void __do_map_lock(struct f2fs_sb_info *sbi, int flag, bool lock)
+void f2fs_do_map_lock(struct f2fs_sb_info *sbi, int flag, bool lock)
{
if (flag == F2FS_GET_BLOCK_PRE_AIO) {
if (lock)
@@ -1539,7 +1541,7 @@ int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map,
next_dnode:
if (map->m_may_create)
- __do_map_lock(sbi, flag, true);
+ f2fs_do_map_lock(sbi, flag, true);
/* When reading holes, we need its node page */
set_new_dnode(&dn, inode, NULL, NULL, 0);
@@ -1688,7 +1690,7 @@ skip:
f2fs_put_dnode(&dn);
if (map->m_may_create) {
- __do_map_lock(sbi, flag, false);
+ f2fs_do_map_lock(sbi, flag, false);
f2fs_balance_fs(sbi, dn.node_changed);
}
goto next_dnode;
@@ -1714,7 +1716,7 @@ sync_out:
f2fs_put_dnode(&dn);
unlock_out:
if (map->m_may_create) {
- __do_map_lock(sbi, flag, false);
+ f2fs_do_map_lock(sbi, flag, false);
f2fs_balance_fs(sbi, dn.node_changed);
}
out:
@@ -1861,6 +1863,7 @@ static int f2fs_xattr_fiemap(struct inode *inode,
flags |= FIEMAP_EXTENT_LAST;
err = fiemap_fill_next_extent(fieinfo, 0, phys, len, flags);
+ trace_f2fs_fiemap(inode, 0, phys, len, flags, err);
if (err || err == 1)
return err;
}
@@ -1884,8 +1887,10 @@ static int f2fs_xattr_fiemap(struct inode *inode,
flags = FIEMAP_EXTENT_LAST;
}
- if (phys)
+ if (phys) {
err = fiemap_fill_next_extent(fieinfo, 0, phys, len, flags);
+ trace_f2fs_fiemap(inode, 0, phys, len, flags, err);
+ }
return (err < 0 ? err : 0);
}
@@ -1979,6 +1984,7 @@ next:
ret = fiemap_fill_next_extent(fieinfo, logical,
phys, size, flags);
+ trace_f2fs_fiemap(inode, logical, phys, size, flags, ret);
if (ret)
goto out;
size = 0;
@@ -2213,9 +2219,7 @@ int f2fs_read_multi_pages(struct compress_ctx *cc, struct bio **bio_ret,
if (ret)
goto out;
- /* cluster was overwritten as normal cluster */
- if (dn.data_blkaddr != COMPRESS_ADDR)
- goto out;
+ f2fs_bug_on(sbi, dn.data_blkaddr != COMPRESS_ADDR);
for (i = 1; i < cc->cluster_size; i++) {
block_t blkaddr;
@@ -2342,6 +2346,7 @@ static int f2fs_mpage_readpages(struct inode *inode,
unsigned nr_pages = rac ? readahead_count(rac) : 1;
unsigned max_nr_pages = nr_pages;
int ret = 0;
+ bool drop_ra = false;
map.m_pblk = 0;
map.m_lblk = 0;
@@ -2352,10 +2357,26 @@ static int f2fs_mpage_readpages(struct inode *inode,
map.m_seg_type = NO_CHECK_TYPE;
map.m_may_create = false;
+ /*
+ * Two readahead threads for same address range can cause race condition
+ * which fragments sequential read IOs. So let's avoid each other.
+ */
+ if (rac && readahead_count(rac)) {
+ if (READ_ONCE(F2FS_I(inode)->ra_offset) == readahead_index(rac))
+ drop_ra = true;
+ else
+ WRITE_ONCE(F2FS_I(inode)->ra_offset,
+ readahead_index(rac));
+ }
+
for (; nr_pages; nr_pages--) {
if (rac) {
page = readahead_page(rac);
prefetchw(&page->flags);
+ if (drop_ra) {
+ f2fs_put_page(page, 1);
+ continue;
+ }
}
#ifdef CONFIG_F2FS_FS_COMPRESSION
@@ -2418,6 +2439,9 @@ next_page:
}
if (bio)
__submit_bio(F2FS_I_SB(inode), bio, DATA);
+
+ if (rac && readahead_count(rac) && !drop_ra)
+ WRITE_ONCE(F2FS_I(inode)->ra_offset, -1);
return ret;
}
@@ -2772,8 +2796,20 @@ write:
/* Dentry/quota blocks are controlled by checkpoint */
if (S_ISDIR(inode->i_mode) || IS_NOQUOTA(inode)) {
+ /*
+ * We need to wait for node_write to avoid block allocation during
+ * checkpoint. This can only happen to quota writes which can cause
+ * the below discard race condition.
+ */
+ if (IS_NOQUOTA(inode))
+ down_read(&sbi->node_write);
+
fio.need_lock = LOCK_DONE;
err = f2fs_do_write_data_page(&fio);
+
+ if (IS_NOQUOTA(inode))
+ up_read(&sbi->node_write);
+
goto done;
}
@@ -3268,7 +3304,7 @@ static int prepare_write_begin(struct f2fs_sb_info *sbi,
if (f2fs_has_inline_data(inode) ||
(pos & PAGE_MASK) >= i_size_read(inode)) {
- __do_map_lock(sbi, flag, true);
+ f2fs_do_map_lock(sbi, flag, true);
locked = true;
}
@@ -3305,7 +3341,7 @@ restart:
err = f2fs_get_dnode_of_data(&dn, index, LOOKUP_NODE);
if (err || dn.data_blkaddr == NULL_ADDR) {
f2fs_put_dnode(&dn);
- __do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO,
+ f2fs_do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO,
true);
WARN_ON(flag != F2FS_GET_BLOCK_PRE_AIO);
locked = true;
@@ -3321,7 +3357,7 @@ out:
f2fs_put_dnode(&dn);
unlock_out:
if (locked)
- __do_map_lock(sbi, flag, false);
+ f2fs_do_map_lock(sbi, flag, false);
return err;
}
@@ -3433,7 +3469,7 @@ repeat:
err = -EFSCORRUPTED;
goto fail;
}
- err = f2fs_submit_page_read(inode, page, blkaddr, true);
+ err = f2fs_submit_page_read(inode, page, blkaddr, 0, true);
if (err)
goto fail;
@@ -3483,6 +3519,10 @@ static int f2fs_write_end(struct file *file,
if (f2fs_compressed_file(inode) && fsdata) {
f2fs_compress_write_end(inode, fsdata, page->index, copied);
f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
+
+ if (pos + copied > i_size_read(inode) &&
+ !f2fs_verity_in_progress(inode))
+ f2fs_i_size_write(inode, pos + copied);
return copied;
}
#endif
@@ -3742,10 +3782,9 @@ static sector_t f2fs_bmap_compress(struct inode *inode, sector_t block)
}
f2fs_put_dnode(&dn);
-
return blknr;
#else
- return -EOPNOTSUPP;
+ return 0;
#endif
}
@@ -3753,18 +3792,26 @@ static sector_t f2fs_bmap_compress(struct inode *inode, sector_t block)
static sector_t f2fs_bmap(struct address_space *mapping, sector_t block)
{
struct inode *inode = mapping->host;
+ struct buffer_head tmp = {
+ .b_size = i_blocksize(inode),
+ };
+ sector_t blknr = 0;
if (f2fs_has_inline_data(inode))
- return 0;
+ goto out;
/* make sure allocating whole blocks */
if (mapping_tagged(mapping, PAGECACHE_TAG_DIRTY))
filemap_write_and_wait(mapping);
if (f2fs_compressed_file(inode))
- return f2fs_bmap_compress(inode, block);
+ blknr = f2fs_bmap_compress(inode, block);
- return generic_block_bmap(mapping, block, get_data_block_bmap);
+ if (!get_data_block_bmap(inode, block, &tmp, 0))
+ blknr = tmp.b_blocknr;
+out:
+ trace_f2fs_bmap(inode, block, blknr);
+ return blknr;
}
#ifdef CONFIG_MIGRATION
diff --git a/fs/f2fs/debug.c b/fs/f2fs/debug.c
index 0dbcb0f9c019..4276c0f79beb 100644
--- a/fs/f2fs/debug.c
+++ b/fs/f2fs/debug.c
@@ -174,6 +174,26 @@ static void update_general_status(struct f2fs_sb_info *sbi)
for (i = META_CP; i < META_MAX; i++)
si->meta_count[i] = atomic_read(&sbi->meta_count[i]);
+ for (i = 0; i < NO_CHECK_TYPE; i++) {
+ si->dirty_seg[i] = 0;
+ si->full_seg[i] = 0;
+ si->valid_blks[i] = 0;
+ }
+
+ for (i = 0; i < MAIN_SEGS(sbi); i++) {
+ int blks = get_seg_entry(sbi, i)->valid_blocks;
+ int type = get_seg_entry(sbi, i)->type;
+
+ if (!blks)
+ continue;
+
+ if (blks == sbi->blocks_per_seg)
+ si->full_seg[type]++;
+ else
+ si->dirty_seg[type]++;
+ si->valid_blks[type] += blks;
+ }
+
for (i = 0; i < 2; i++) {
si->segment_count[i] = sbi->segment_count[i];
si->block_count[i] = sbi->block_count[i];
@@ -329,30 +349,50 @@ static int stat_show(struct seq_file *s, void *v)
seq_printf(s, "\nMain area: %d segs, %d secs %d zones\n",
si->main_area_segs, si->main_area_sections,
si->main_area_zones);
- seq_printf(s, " - COLD data: %d, %d, %d\n",
+ seq_printf(s, " TYPE %8s %8s %8s %10s %10s %10s\n",
+ "segno", "secno", "zoneno", "dirty_seg", "full_seg", "valid_blk");
+ seq_printf(s, " - COLD data: %8d %8d %8d %10u %10u %10u\n",
si->curseg[CURSEG_COLD_DATA],
si->cursec[CURSEG_COLD_DATA],
- si->curzone[CURSEG_COLD_DATA]);
- seq_printf(s, " - WARM data: %d, %d, %d\n",
+ si->curzone[CURSEG_COLD_DATA],
+ si->dirty_seg[CURSEG_COLD_DATA],
+ si->full_seg[CURSEG_COLD_DATA],
+ si->valid_blks[CURSEG_COLD_DATA]);
+ seq_printf(s, " - WARM data: %8d %8d %8d %10u %10u %10u\n",
si->curseg[CURSEG_WARM_DATA],
si->cursec[CURSEG_WARM_DATA],
- si->curzone[CURSEG_WARM_DATA]);
- seq_printf(s, " - HOT data: %d, %d, %d\n",
+ si->curzone[CURSEG_WARM_DATA],
+ si->dirty_seg[CURSEG_WARM_DATA],
+ si->full_seg[CURSEG_WARM_DATA],
+ si->valid_blks[CURSEG_WARM_DATA]);
+ seq_printf(s, " - HOT data: %8d %8d %8d %10u %10u %10u\n",
si->curseg[CURSEG_HOT_DATA],
si->cursec[CURSEG_HOT_DATA],
- si->curzone[CURSEG_HOT_DATA]);
- seq_printf(s, " - Dir dnode: %d, %d, %d\n",
+ si->curzone[CURSEG_HOT_DATA],
+ si->dirty_seg[CURSEG_HOT_DATA],
+ si->full_seg[CURSEG_HOT_DATA],
+ si->valid_blks[CURSEG_HOT_DATA]);
+ seq_printf(s, " - Dir dnode: %8d %8d %8d %10u %10u %10u\n",
si->curseg[CURSEG_HOT_NODE],
si->cursec[CURSEG_HOT_NODE],
- si->curzone[CURSEG_HOT_NODE]);
- seq_printf(s, " - File dnode: %d, %d, %d\n",
+ si->curzone[CURSEG_HOT_NODE],
+ si->dirty_seg[CURSEG_HOT_NODE],
+ si->full_seg[CURSEG_HOT_NODE],
+ si->valid_blks[CURSEG_HOT_NODE]);
+ seq_printf(s, " - File dnode: %8d %8d %8d %10u %10u %10u\n",
si->curseg[CURSEG_WARM_NODE],
si->cursec[CURSEG_WARM_NODE],
- si->curzone[CURSEG_WARM_NODE]);
- seq_printf(s, " - Indir nodes: %d, %d, %d\n",
+ si->curzone[CURSEG_WARM_NODE],
+ si->dirty_seg[CURSEG_WARM_NODE],
+ si->full_seg[CURSEG_WARM_NODE],
+ si->valid_blks[CURSEG_WARM_NODE]);
+ seq_printf(s, " - Indir nodes: %8d %8d %8d %10u %10u %10u\n",
si->curseg[CURSEG_COLD_NODE],
si->cursec[CURSEG_COLD_NODE],
- si->curzone[CURSEG_COLD_NODE]);
+ si->curzone[CURSEG_COLD_NODE],
+ si->dirty_seg[CURSEG_COLD_NODE],
+ si->full_seg[CURSEG_COLD_NODE],
+ si->valid_blks[CURSEG_COLD_NODE]);
seq_printf(s, "\n - Valid: %d\n - Dirty: %d\n",
si->main_area_segs - si->dirty_count -
si->prefree_count - si->free_segs,
diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
index d35976785e8c..069f498af1e3 100644
--- a/fs/f2fs/dir.c
+++ b/fs/f2fs/dir.c
@@ -779,7 +779,7 @@ int f2fs_do_add_link(struct inode *dir, const struct qstr *name,
return err;
/*
- * An immature stakable filesystem shows a race condition between lookup
+ * An immature stackable filesystem shows a race condition between lookup
* and create. If we have same task when doing lookup and create, it's
* definitely fine as expected by VFS normally. Otherwise, let's just
* verify on-disk dentry one more time, which guarantees filesystem
diff --git a/fs/f2fs/extent_cache.c b/fs/f2fs/extent_cache.c
index e60078460ad1..686c68b98610 100644
--- a/fs/f2fs/extent_cache.c
+++ b/fs/f2fs/extent_cache.c
@@ -325,9 +325,10 @@ static void __drop_largest_extent(struct extent_tree *et,
}
/* return true, if inode page is changed */
-static bool __f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
+static void __f2fs_init_extent_tree(struct inode *inode, struct page *ipage)
{
struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+ struct f2fs_extent *i_ext = ipage ? &F2FS_INODE(ipage)->i_ext : NULL;
struct extent_tree *et;
struct extent_node *en;
struct extent_info ei;
@@ -335,16 +336,18 @@ static bool __f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_e
if (!f2fs_may_extent_tree(inode)) {
/* drop largest extent */
if (i_ext && i_ext->len) {
+ f2fs_wait_on_page_writeback(ipage, NODE, true, true);
i_ext->len = 0;
- return true;
+ set_page_dirty(ipage);
+ return;
}
- return false;
+ return;
}
et = __grab_extent_tree(inode);
if (!i_ext || !i_ext->len)
- return false;
+ return;
get_extent_info(&ei, i_ext);
@@ -360,17 +363,14 @@ static bool __f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_e
}
out:
write_unlock(&et->lock);
- return false;
}
-bool f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
+void f2fs_init_extent_tree(struct inode *inode, struct page *ipage)
{
- bool ret = __f2fs_init_extent_tree(inode, i_ext);
+ __f2fs_init_extent_tree(inode, ipage);
if (!F2FS_I(inode)->extent_tree)
set_inode_flag(inode, FI_NO_EXTENT);
-
- return ret;
}
static bool f2fs_lookup_extent_tree(struct inode *inode, pgoff_t pgofs,
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index b35a50f4953c..16322ea5b463 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -402,12 +402,8 @@ static inline bool __has_cursum_space(struct f2fs_journal *journal,
}
/*
- * ioctl commands
+ * f2fs-specific ioctl commands
*/
-#define F2FS_IOC_GETFLAGS FS_IOC_GETFLAGS
-#define F2FS_IOC_SETFLAGS FS_IOC_SETFLAGS
-#define F2FS_IOC_GETVERSION FS_IOC_GETVERSION
-
#define F2FS_IOCTL_MAGIC 0xf5
#define F2FS_IOC_START_ATOMIC_WRITE _IO(F2FS_IOCTL_MAGIC, 1)
#define F2FS_IOC_COMMIT_ATOMIC_WRITE _IO(F2FS_IOCTL_MAGIC, 2)
@@ -434,13 +430,8 @@ static inline bool __has_cursum_space(struct f2fs_journal *journal,
_IOR(F2FS_IOCTL_MAGIC, 18, __u64)
#define F2FS_IOC_RESERVE_COMPRESS_BLOCKS \
_IOR(F2FS_IOCTL_MAGIC, 19, __u64)
-
-#define F2FS_IOC_GET_VOLUME_NAME FS_IOC_GETFSLABEL
-#define F2FS_IOC_SET_VOLUME_NAME FS_IOC_SETFSLABEL
-
-#define F2FS_IOC_SET_ENCRYPTION_POLICY FS_IOC_SET_ENCRYPTION_POLICY
-#define F2FS_IOC_GET_ENCRYPTION_POLICY FS_IOC_GET_ENCRYPTION_POLICY
-#define F2FS_IOC_GET_ENCRYPTION_PWSALT FS_IOC_GET_ENCRYPTION_PWSALT
+#define F2FS_IOC_SEC_TRIM_FILE _IOW(F2FS_IOCTL_MAGIC, 20, \
+ struct f2fs_sectrim_range)
/*
* should be same as XFS_IOC_GOINGDOWN.
@@ -453,17 +444,12 @@ static inline bool __has_cursum_space(struct f2fs_journal *journal,
#define F2FS_GOING_DOWN_METAFLUSH 0x3 /* going down with meta flush */
#define F2FS_GOING_DOWN_NEED_FSCK 0x4 /* going down to trigger fsck */
-#if defined(__KERNEL__) && defined(CONFIG_COMPAT)
/*
- * ioctl commands in 32 bit emulation
+ * Flags used by F2FS_IOC_SEC_TRIM_FILE
*/
-#define F2FS_IOC32_GETFLAGS FS_IOC32_GETFLAGS
-#define F2FS_IOC32_SETFLAGS FS_IOC32_SETFLAGS
-#define F2FS_IOC32_GETVERSION FS_IOC32_GETVERSION
-#endif
-
-#define F2FS_IOC_FSGETXATTR FS_IOC_FSGETXATTR
-#define F2FS_IOC_FSSETXATTR FS_IOC_FSSETXATTR
+#define F2FS_TRIM_FILE_DISCARD 0x1 /* send discard command */
+#define F2FS_TRIM_FILE_ZEROOUT 0x2 /* zero out */
+#define F2FS_TRIM_FILE_MASK 0x3
struct f2fs_gc_range {
u32 sync;
@@ -488,6 +474,12 @@ struct f2fs_flush_device {
u32 segments; /* # of segments to flush */
};
+struct f2fs_sectrim_range {
+ u64 start;
+ u64 len;
+ u64 flags;
+};
+
/* for inline stuff */
#define DEF_INLINE_RESERVED_SIZE 1
static inline int get_extra_isize(struct inode *inode);
@@ -794,6 +786,7 @@ struct f2fs_inode_info {
struct list_head inmem_pages; /* inmemory pages managed by f2fs */
struct task_struct *inmem_task; /* store inmemory task */
struct mutex inmem_lock; /* lock for inmemory pages */
+ pgoff_t ra_offset; /* ongoing readahead offset */
struct extent_tree *extent_tree; /* cached extent_tree entry */
/* avoid racing between foreground op and gc */
@@ -1267,7 +1260,8 @@ enum {
GC_NORMAL,
GC_IDLE_CB,
GC_IDLE_GREEDY,
- GC_URGENT,
+ GC_URGENT_HIGH,
+ GC_URGENT_LOW,
};
enum {
@@ -1313,6 +1307,14 @@ enum fsync_mode {
#define IS_DUMMY_WRITTEN_PAGE(page) \
(page_private(page) == (unsigned long)DUMMY_WRITTEN_PAGE)
+#ifdef CONFIG_F2FS_IO_TRACE
+#define IS_IO_TRACED_PAGE(page) \
+ (page_private(page) > 0 && \
+ page_private(page) < (unsigned long)PID_MAX_LIMIT)
+#else
+#define IS_IO_TRACED_PAGE(page) (0)
+#endif
+
#ifdef CONFIG_FS_ENCRYPTION
#define DUMMY_ENCRYPTION_ENABLED(sbi) \
(unlikely(F2FS_OPTION(sbi).dummy_enc_ctx.ctx != NULL))
@@ -1438,7 +1440,7 @@ struct f2fs_sb_info {
unsigned long last_time[MAX_TIME]; /* to store time in jiffies */
long interval_time[MAX_TIME]; /* to store thresholds */
- struct inode_management im[MAX_INO_ENTRY]; /* manage inode cache */
+ struct inode_management im[MAX_INO_ENTRY]; /* manage inode cache */
spinlock_t fsync_node_lock; /* for node entry lock */
struct list_head fsync_node_list; /* node list head */
@@ -1516,8 +1518,9 @@ struct f2fs_sb_info {
unsigned int cur_victim_sec; /* current victim section num */
unsigned int gc_mode; /* current GC state */
unsigned int next_victim_seg[2]; /* next segment in victim section */
+
/* for skip statistic */
- unsigned int atomic_files; /* # of opened atomic file */
+ unsigned int atomic_files; /* # of opened atomic file */
unsigned long long skipped_atomic_files[2]; /* FG_GC and BG_GC */
unsigned long long skipped_gc_rwsem; /* FG_GC only */
@@ -2456,7 +2459,7 @@ static inline void *f2fs_kmem_cache_alloc(struct kmem_cache *cachep,
static inline bool is_idle(struct f2fs_sb_info *sbi, int type)
{
- if (sbi->gc_mode == GC_URGENT)
+ if (sbi->gc_mode == GC_URGENT_HIGH)
return true;
if (get_pages(sbi, F2FS_RD_DATA) || get_pages(sbi, F2FS_RD_NODE) ||
@@ -2474,6 +2477,10 @@ static inline bool is_idle(struct f2fs_sb_info *sbi, int type)
atomic_read(&SM_I(sbi)->fcc_info->queued_flush))
return false;
+ if (sbi->gc_mode == GC_URGENT_LOW &&
+ (type == DISCARD_TIME || type == GC_TIME))
+ return true;
+
return f2fs_time_over(sbi, type);
}
@@ -2649,7 +2656,7 @@ static inline void __mark_inode_dirty_flag(struct inode *inode,
static inline void set_inode_flag(struct inode *inode, int flag)
{
- test_and_set_bit(flag, F2FS_I(inode)->flags);
+ set_bit(flag, F2FS_I(inode)->flags);
__mark_inode_dirty_flag(inode, flag, true);
}
@@ -2660,7 +2667,7 @@ static inline int is_inode_flag_set(struct inode *inode, int flag)
static inline void clear_inode_flag(struct inode *inode, int flag)
{
- test_and_clear_bit(flag, F2FS_I(inode)->flags);
+ clear_bit(flag, F2FS_I(inode)->flags);
__mark_inode_dirty_flag(inode, flag, false);
}
@@ -3275,7 +3282,7 @@ void f2fs_ra_node_page(struct f2fs_sb_info *sbi, nid_t nid);
struct page *f2fs_get_node_page(struct f2fs_sb_info *sbi, pgoff_t nid);
struct page *f2fs_get_node_page_ra(struct page *parent, int start);
int f2fs_move_node_page(struct page *node_page, int gc_type);
-int f2fs_flush_inline_data(struct f2fs_sb_info *sbi);
+void f2fs_