aboutsummaryrefslogtreecommitdiffstats
path: root/fs/f2fs/node.c
diff options
context:
space:
mode:
Diffstat (limited to 'fs/f2fs/node.c')
-rw-r--r--fs/f2fs/node.c591
1 files changed, 395 insertions, 196 deletions
diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
index 9d02cdcdbb07..983572f23896 100644
--- a/fs/f2fs/node.c
+++ b/fs/f2fs/node.c
@@ -8,7 +8,7 @@
#include <linux/fs.h>
#include <linux/f2fs_fs.h>
#include <linux/mpage.h>
-#include <linux/backing-dev.h>
+#include <linux/sched/mm.h>
#include <linux/blkdev.h>
#include <linux/pagevec.h>
#include <linux/swap.h>
@@ -17,7 +17,7 @@
#include "node.h"
#include "segment.h"
#include "xattr.h"
-#include "trace.h"
+#include "iostat.h"
#include <trace/events/f2fs.h>
#define on_f2fs_build_free_nids(nmi) mutex_is_locked(&(nm_i)->build_lock)
@@ -36,6 +36,7 @@ int f2fs_check_nid_range(struct f2fs_sb_info *sbi, nid_t nid)
set_sbi_flag(sbi, SBI_NEED_FSCK);
f2fs_warn(sbi, "%s: out-of-range nid=%x, run fsck to fix.",
__func__, nid);
+ f2fs_handle_error(sbi, ERROR_CORRUPTED_INODE);
return -EFSCORRUPTED;
}
return 0;
@@ -44,11 +45,15 @@ int f2fs_check_nid_range(struct f2fs_sb_info *sbi, nid_t nid)
bool f2fs_available_free_memory(struct f2fs_sb_info *sbi, int type)
{
struct f2fs_nm_info *nm_i = NM_I(sbi);
+ struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
struct sysinfo val;
unsigned long avail_ram;
unsigned long mem_size = 0;
bool res = false;
+ if (!nm_i)
+ return true;
+
si_meminfo(&val);
/* only uses low memory */
@@ -62,8 +67,8 @@ bool f2fs_available_free_memory(struct f2fs_sb_info *sbi, int type)
sizeof(struct free_nid)) >> PAGE_SHIFT;
res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 2);
} else if (type == NAT_ENTRIES) {
- mem_size = (nm_i->nat_cnt * sizeof(struct nat_entry)) >>
- PAGE_SHIFT;
+ mem_size = (nm_i->nat_cnt[TOTAL_NAT] *
+ sizeof(struct nat_entry)) >> PAGE_SHIFT;
res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 2);
if (excess_cached_nats(sbi))
res = false;
@@ -86,10 +91,24 @@ bool f2fs_available_free_memory(struct f2fs_sb_info *sbi, int type)
atomic_read(&sbi->total_ext_node) *
sizeof(struct extent_node)) >> PAGE_SHIFT;
res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 1);
- } else if (type == INMEM_PAGES) {
- /* it allows 20% / total_ram for inmemory pages */
- mem_size = get_pages(sbi, F2FS_INMEM_PAGES);
- res = mem_size < (val.totalram / 5);
+ } else if (type == DISCARD_CACHE) {
+ mem_size = (atomic_read(&dcc->discard_cmd_cnt) *
+ sizeof(struct discard_cmd)) >> PAGE_SHIFT;
+ res = mem_size < (avail_ram * nm_i->ram_thresh / 100);
+ } else if (type == COMPRESS_PAGE) {
+#ifdef CONFIG_F2FS_FS_COMPRESSION
+ unsigned long free_ram = val.freeram;
+
+ /*
+ * free memory is lower than watermark or cached page count
+ * exceed threshold, deny caching compress page.
+ */
+ res = (free_ram > avail_ram * sbi->compress_watermark / 100) &&
+ (COMPRESS_MAPPING(sbi)->nrpages <
+ free_ram * sbi->compress_percent / 100);
+#else
+ res = false;
+#endif
} else {
if (!sbi->sb->s_bdi->wb.dirty_exceeded)
return true;
@@ -109,7 +128,7 @@ static void clear_node_page_dirty(struct page *page)
static struct page *get_current_nat_page(struct f2fs_sb_info *sbi, nid_t nid)
{
- return f2fs_get_meta_page_nofail(sbi, current_nat_addr(sbi, nid));
+ return f2fs_get_meta_page_retry(sbi, current_nat_addr(sbi, nid));
}
static struct page *get_next_nat_page(struct f2fs_sb_info *sbi, nid_t nid)
@@ -141,14 +160,13 @@ static struct page *get_next_nat_page(struct f2fs_sb_info *sbi, nid_t nid)
return dst_page;
}
-static struct nat_entry *__alloc_nat_entry(nid_t nid, bool no_fail)
+static struct nat_entry *__alloc_nat_entry(struct f2fs_sb_info *sbi,
+ nid_t nid, bool no_fail)
{
struct nat_entry *new;
- if (no_fail)
- new = f2fs_kmem_cache_alloc(nat_entry_slab, GFP_F2FS_ZERO);
- else
- new = kmem_cache_alloc(nat_entry_slab, GFP_F2FS_ZERO);
+ new = f2fs_kmem_cache_alloc(nat_entry_slab,
+ GFP_F2FS_ZERO, no_fail, sbi);
if (new) {
nat_set_nid(new, nid);
nat_reset_flag(new);
@@ -177,7 +195,8 @@ static struct nat_entry *__init_nat_entry(struct f2fs_nm_info *nm_i,
list_add_tail(&ne->list, &nm_i->nat_entries);
spin_unlock(&nm_i->nat_list_lock);
- nm_i->nat_cnt++;
+ nm_i->nat_cnt[TOTAL_NAT]++;
+ nm_i->nat_cnt[RECLAIMABLE_NAT]++;
return ne;
}
@@ -207,7 +226,8 @@ static unsigned int __gang_lookup_nat_cache(struct f2fs_nm_info *nm_i,
static void __del_from_nat_cache(struct f2fs_nm_info *nm_i, struct nat_entry *e)
{
radix_tree_delete(&nm_i->nat_root, nat_get_nid(e));
- nm_i->nat_cnt--;
+ nm_i->nat_cnt[TOTAL_NAT]--;
+ nm_i->nat_cnt[RECLAIMABLE_NAT]--;
__free_nat_entry(e);
}
@@ -219,7 +239,8 @@ static struct nat_entry_set *__grab_nat_entry_set(struct f2fs_nm_info *nm_i,
head = radix_tree_lookup(&nm_i->nat_set_root, set);
if (!head) {
- head = f2fs_kmem_cache_alloc(nat_entry_set_slab, GFP_NOFS);
+ head = f2fs_kmem_cache_alloc(nat_entry_set_slab,
+ GFP_NOFS, true, NULL);
INIT_LIST_HEAD(&head->entry_list);
INIT_LIST_HEAD(&head->set_list);
@@ -253,7 +274,8 @@ static void __set_nat_cache_dirty(struct f2fs_nm_info *nm_i,
if (get_nat_flag(ne, IS_DIRTY))
goto refresh_list;
- nm_i->dirty_nat_cnt++;
+ nm_i->nat_cnt[DIRTY_NAT]++;
+ nm_i->nat_cnt[RECLAIMABLE_NAT]--;
set_nat_flag(ne, IS_DIRTY, true);
refresh_list:
spin_lock(&nm_i->nat_list_lock);
@@ -273,7 +295,8 @@ static void __clear_nat_cache_dirty(struct f2fs_nm_info *nm_i,
set_nat_flag(ne, IS_DIRTY, false);
set->entry_cnt--;
- nm_i->dirty_nat_cnt--;
+ nm_i->nat_cnt[DIRTY_NAT]--;
+ nm_i->nat_cnt[RECLAIMABLE_NAT]++;
}
static unsigned int __gang_lookup_nat_set(struct f2fs_nm_info *nm_i,
@@ -304,7 +327,8 @@ static unsigned int f2fs_add_fsync_node_entry(struct f2fs_sb_info *sbi,
unsigned long flags;
unsigned int seq_id;
- fn = f2fs_kmem_cache_alloc(fsync_node_entry_slab, GFP_NOFS);
+ fn = f2fs_kmem_cache_alloc(fsync_node_entry_slab,
+ GFP_NOFS, true, NULL);
get_page(page);
fn->page = page;
@@ -355,14 +379,14 @@ int f2fs_need_dentry_mark(struct f2fs_sb_info *sbi, nid_t nid)
struct nat_entry *e;
bool need = false;
- down_read(&nm_i->nat_tree_lock);
+ f2fs_down_read(&nm_i->nat_tree_lock);
e = __lookup_nat_cache(nm_i, nid);
if (e) {
if (!get_nat_flag(e, IS_CHECKPOINTED) &&
!get_nat_flag(e, HAS_FSYNCED_INODE))
need = true;
}
- up_read(&nm_i->nat_tree_lock);
+ f2fs_up_read(&nm_i->nat_tree_lock);
return need;
}
@@ -372,11 +396,11 @@ bool f2fs_is_checkpointed_node(struct f2fs_sb_info *sbi, nid_t nid)
struct nat_entry *e;
bool is_cp = true;
- down_read(&nm_i->nat_tree_lock);
+ f2fs_down_read(&nm_i->nat_tree_lock);
e = __lookup_nat_cache(nm_i, nid);
if (e && !get_nat_flag(e, IS_CHECKPOINTED))
is_cp = false;
- up_read(&nm_i->nat_tree_lock);
+ f2fs_up_read(&nm_i->nat_tree_lock);
return is_cp;
}
@@ -386,13 +410,13 @@ bool f2fs_need_inode_block_update(struct f2fs_sb_info *sbi, nid_t ino)
struct nat_entry *e;
bool need_update = true;
- down_read(&nm_i->nat_tree_lock);
+ f2fs_down_read(&nm_i->nat_tree_lock);
e = __lookup_nat_cache(nm_i, ino);
if (e && get_nat_flag(e, HAS_LAST_FSYNC) &&
(get_nat_flag(e, IS_CHECKPOINTED) ||
get_nat_flag(e, HAS_FSYNCED_INODE)))
need_update = false;
- up_read(&nm_i->nat_tree_lock);
+ f2fs_up_read(&nm_i->nat_tree_lock);
return need_update;
}
@@ -403,11 +427,15 @@ static void cache_nat_entry(struct f2fs_sb_info *sbi, nid_t nid,
struct f2fs_nm_info *nm_i = NM_I(sbi);
struct nat_entry *new, *e;
- new = __alloc_nat_entry(nid, false);
+ /* Let's mitigate lock contention of nat_tree_lock during checkpoint */
+ if (f2fs_rwsem_is_locked(&sbi->cp_global_sem))
+ return;
+
+ new = __alloc_nat_entry(sbi, nid, false);
if (!new)
return;
- down_write(&nm_i->nat_tree_lock);
+ f2fs_down_write(&nm_i->nat_tree_lock);
e = __lookup_nat_cache(nm_i, nid);
if (!e)
e = __init_nat_entry(nm_i, new, ne, false);
@@ -416,7 +444,7 @@ static void cache_nat_entry(struct f2fs_sb_info *sbi, nid_t nid,
nat_get_blkaddr(e) !=
le32_to_cpu(ne->block_addr) ||
nat_get_version(e) != ne->version);
- up_write(&nm_i->nat_tree_lock);
+ f2fs_up_write(&nm_i->nat_tree_lock);
if (e != new)
__free_nat_entry(new);
}
@@ -426,9 +454,9 @@ static void set_node_addr(struct f2fs_sb_info *sbi, struct node_info *ni,
{
struct f2fs_nm_info *nm_i = NM_I(sbi);
struct nat_entry *e;
- struct nat_entry *new = __alloc_nat_entry(ni->nid, true);
+ struct nat_entry *new = __alloc_nat_entry(sbi, ni->nid, true);
- down_write(&nm_i->nat_tree_lock);
+ f2fs_down_write(&nm_i->nat_tree_lock);
e = __lookup_nat_cache(nm_i, ni->nid);
if (!e) {
e = __init_nat_entry(nm_i, new, NULL, true);
@@ -459,6 +487,7 @@ static void set_node_addr(struct f2fs_sb_info *sbi, struct node_info *ni,
/* increment version no as node is removed */
if (nat_get_blkaddr(e) != NEW_ADDR && new_blkaddr == NULL_ADDR) {
unsigned char version = nat_get_version(e);
+
nat_set_version(e, inc_node_version(version));
}
@@ -476,7 +505,7 @@ static void set_node_addr(struct f2fs_sb_info *sbi, struct node_info *ni,
set_nat_flag(e, HAS_FSYNCED_INODE, true);
set_nat_flag(e, HAS_LAST_FSYNC, fsync_done);
}
- up_write(&nm_i->nat_tree_lock);
+ f2fs_up_write(&nm_i->nat_tree_lock);
}
int f2fs_try_to_free_nats(struct f2fs_sb_info *sbi, int nr_shrink)
@@ -484,7 +513,7 @@ int f2fs_try_to_free_nats(struct f2fs_sb_info *sbi, int nr_shrink)
struct f2fs_nm_info *nm_i = NM_I(sbi);
int nr = nr_shrink;
- if (!down_write_trylock(&nm_i->nat_tree_lock))
+ if (!f2fs_down_write_trylock(&nm_i->nat_tree_lock))
return 0;
spin_lock(&nm_i->nat_list_lock);
@@ -506,15 +535,12 @@ int f2fs_try_to_free_nats(struct f2fs_sb_info *sbi, int nr_shrink)
}
spin_unlock(&nm_i->nat_list_lock);
- up_write(&nm_i->nat_tree_lock);
+ f2fs_up_write(&nm_i->nat_tree_lock);
return nr - nr_shrink;
}
-/*
- * This function always returns success
- */
int f2fs_get_node_info(struct f2fs_sb_info *sbi, nid_t nid,
- struct node_info *ni)
+ struct node_info *ni, bool checkpoint_context)
{
struct f2fs_nm_info *nm_i = NM_I(sbi);
struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
@@ -529,22 +555,32 @@ int f2fs_get_node_info(struct f2fs_sb_info *sbi, nid_t nid,
int i;
ni->nid = nid;
-
+retry:
/* Check nat cache */
- down_read(&nm_i->nat_tree_lock);
+ f2fs_down_read(&nm_i->nat_tree_lock);
e = __lookup_nat_cache(nm_i, nid);
if (e) {
ni->ino = nat_get_ino(e);
ni->blk_addr = nat_get_blkaddr(e);
ni->version = nat_get_version(e);
- up_read(&nm_i->nat_tree_lock);
+ f2fs_up_read(&nm_i->nat_tree_lock);
return 0;
}
- memset(&ne, 0, sizeof(struct f2fs_nat_entry));
+ /*
+ * Check current segment summary by trying to grab journal_rwsem first.
+ * This sem is on the critical path on the checkpoint requiring the above
+ * nat_tree_lock. Therefore, we should retry, if we failed to grab here
+ * while not bothering checkpoint.
+ */
+ if (!f2fs_rwsem_is_locked(&sbi->cp_global_sem) || checkpoint_context) {
+ down_read(&curseg->journal_rwsem);
+ } else if (f2fs_rwsem_is_contended(&nm_i->nat_tree_lock) ||
+ !down_read_trylock(&curseg->journal_rwsem)) {
+ f2fs_up_read(&nm_i->nat_tree_lock);
+ goto retry;
+ }
- /* Check current segment summary */
- down_read(&curseg->journal_rwsem);
i = f2fs_lookup_journal_in_cursum(journal, NAT_JOURNAL, nid, 0);
if (i >= 0) {
ne = nat_in_journal(journal, i);
@@ -552,13 +588,13 @@ int f2fs_get_node_info(struct f2fs_sb_info *sbi, nid_t nid,
}
up_read(&curseg->journal_rwsem);
if (i >= 0) {
- up_read(&nm_i->nat_tree_lock);
+ f2fs_up_read(&nm_i->nat_tree_lock);
goto cache;
}
/* Fill node_info from nat page */
index = current_nat_addr(sbi, nid);
- up_read(&nm_i->nat_tree_lock);
+ f2fs_up_read(&nm_i->nat_tree_lock);
page = f2fs_get_meta_page(sbi, index);
if (IS_ERR(page))
@@ -621,10 +657,10 @@ pgoff_t f2fs_get_next_page_offset(struct dnode_of_data *dn, pgoff_t pgofs)
switch (dn->max_level) {
case 3:
base += 2 * indirect_blks;
- /* fall through */
+ fallthrough;
case 2:
base += 2 * direct_blks;
- /* fall through */
+ fallthrough;
case 1:
base += direct_index;
break;
@@ -716,8 +752,7 @@ got:
/*
* Caller should call f2fs_put_dnode(dn).
* Also, it should grab and release a rwsem by calling f2fs_lock_op() and
- * f2fs_unlock_op() only if ro is not set RDONLY_NODE.
- * In the case of RDONLY_NODE, we don't need to care about mutex.
+ * f2fs_unlock_op() only if mode is set with ALLOC_NODE.
*/
int f2fs_get_dnode_of_data(struct dnode_of_data *dn, pgoff_t index, int mode)
{
@@ -809,8 +844,27 @@ int f2fs_get_dnode_of_data(struct dnode_of_data *dn, pgoff_t index, int mode)
dn->nid = nids[level];
dn->ofs_in_node = offset[level];
dn->node_page = npage[level];
- dn->data_blkaddr = datablock_addr(dn->inode,
- dn->node_page, dn->ofs_in_node);
+ dn->data_blkaddr = f2fs_data_blkaddr(dn);
+
+ if (is_inode_flag_set(dn->inode, FI_COMPRESSED_FILE) &&
+ f2fs_sb_has_readonly(sbi)) {
+ unsigned int c_len = f2fs_cluster_blocks_are_contiguous(dn);
+ block_t blkaddr;
+
+ if (!c_len)
+ goto out;
+
+ blkaddr = f2fs_data_blkaddr(dn);
+ if (blkaddr == COMPRESS_ADDR)
+ blkaddr = data_blkaddr(dn->inode, dn->node_page,
+ dn->ofs_in_node + 1);
+
+ f2fs_update_extent_tree_range_compressed(dn->inode,
+ index, blkaddr,
+ F2FS_I(dn->inode)->i_cluster_size,
+ c_len);
+ }
+out:
return 0;
release_pages:
@@ -835,7 +889,7 @@ static int truncate_node(struct dnode_of_data *dn)
int err;
pgoff_t index;
- err = f2fs_get_node_info(sbi, dn->nid, &ni);
+ err = f2fs_get_node_info(sbi, dn->nid, &ni, false);
if (err)
return err;
@@ -1046,8 +1100,10 @@ int f2fs_truncate_inode_blocks(struct inode *inode, pgoff_t from)
trace_f2fs_truncate_inode_blocks_enter(inode, from);
level = get_node_path(inode, from, offset, noffset);
- if (level < 0)
+ if (level < 0) {
+ trace_f2fs_truncate_inode_blocks_exit(inode, level);
return level;
+ }
page = f2fs_get_node_page(sbi, inode->i_ino);
if (IS_ERR(page)) {
@@ -1188,8 +1244,9 @@ int f2fs_remove_inode_page(struct inode *inode)
}
if (unlikely(inode->i_blocks != 0 && inode->i_blocks != 8)) {
- f2fs_warn(F2FS_I_SB(inode), "Inconsistent i_blocks, ino:%lu, iblocks:%llu",
- inode->i_ino, (unsigned long long)inode->i_blocks);
+ f2fs_warn(F2FS_I_SB(inode),
+ "f2fs_remove_inode_page: inconsistent i_blocks, ino:%lu, iblocks:%llu",
+ inode->i_ino, (unsigned long long)inode->i_blocks);
set_sbi_flag(F2FS_I_SB(inode), SBI_NEED_FSCK);
}
@@ -1231,12 +1288,17 @@ struct page *f2fs_new_node_page(struct dnode_of_data *dn, unsigned int ofs)
goto fail;
#ifdef CONFIG_F2FS_CHECK_FS
- err = f2fs_get_node_info(sbi, dn->nid, &new_ni);
+ err = f2fs_get_node_info(sbi, dn->nid, &new_ni, false);
if (err) {
dec_valid_node_count(sbi, dn->inode, !ofs);
goto fail;
}
- f2fs_bug_on(sbi, new_ni.blk_addr != NULL_ADDR);
+ if (unlikely(new_ni.blk_addr != NULL_ADDR)) {
+ err = -EFSCORRUPTED;
+ set_sbi_flag(sbi, SBI_NEED_FSCK);
+ f2fs_handle_error(sbi, ERROR_INVALID_BLKADDR);
+ goto fail;
+ }
#endif
new_ni.nid = dn->nid;
new_ni.ino = dn->inode->i_ino;
@@ -1271,7 +1333,7 @@ fail:
* 0: f2fs_put_page(page, 0)
* LOCKED_PAGE or error: f2fs_put_page(page, 1)
*/
-static int read_node_page(struct page *page, int op_flags)
+static int read_node_page(struct page *page, blk_opf_t op_flags)
{
struct f2fs_sb_info *sbi = F2FS_P_SB(page);
struct node_info ni;
@@ -1293,18 +1355,25 @@ static int read_node_page(struct page *page, int op_flags)
return LOCKED_PAGE;
}
- err = f2fs_get_node_info(sbi, page->index, &ni);
+ err = f2fs_get_node_info(sbi, page->index, &ni, false);
if (err)
return err;
- if (unlikely(ni.blk_addr == NULL_ADDR) ||
+ /* NEW_ADDR can be seen, after cp_error drops some dirty node pages */
+ if (unlikely(ni.blk_addr == NULL_ADDR || ni.blk_addr == NEW_ADDR) ||
is_sbi_flag_set(sbi, SBI_IS_SHUTDOWN)) {
ClearPageUptodate(page);
return -ENOENT;
}
fio.new_blkaddr = fio.old_blkaddr = ni.blk_addr;
- return f2fs_submit_page_bio(&fio);
+
+ err = f2fs_submit_page_bio(&fio);
+
+ if (!err)
+ f2fs_update_iostat(sbi, NULL, FS_NODE_READ_IO, F2FS_BLKSIZE);
+
+ return err;
}
/*
@@ -1349,8 +1418,7 @@ repeat:
err = read_node_page(page, 0);
if (err < 0) {
- f2fs_put_page(page, 1);
- return ERR_PTR(err);
+ goto out_put_err;
} else if (err == LOCKED_PAGE) {
err = 0;
goto page_hit;
@@ -1376,18 +1444,23 @@ repeat:
goto out_err;
}
page_hit:
- if(unlikely(nid != nid_of_node(page))) {
- f2fs_warn(sbi, "inconsistent node block, nid:%lu, node_footer[nid:%u,ino:%u,ofs:%u,cpver:%llu,blkaddr:%u]",
+ if (likely(nid == nid_of_node(page)))
+ return page;
+
+ f2fs_warn(sbi, "inconsistent node block, nid:%lu, node_footer[nid:%u,ino:%u,ofs:%u,cpver:%llu,blkaddr:%u]",
nid, nid_of_node(page), ino_of_node(page),
ofs_of_node(page), cpver_of_node(page),
next_blkaddr_of_node(page));
- err = -EINVAL;
+ set_sbi_flag(sbi, SBI_NEED_FSCK);
+ err = -EINVAL;
out_err:
- ClearPageUptodate(page);
- f2fs_put_page(page, 1);
- return ERR_PTR(err);
- }
- return page;
+ ClearPageUptodate(page);
+out_put_err:
+ /* ENOENT comes from read_node_page which is not an error. */
+ if (err != -ENOENT)
+ f2fs_handle_page_eio(sbi, page->index, NODE);
+ f2fs_put_page(page, 1);
+ return ERR_PTR(err);
}
struct page *f2fs_get_node_page(struct f2fs_sb_info *sbi, pgoff_t nid)
@@ -1518,8 +1591,12 @@ static int __write_node_page(struct page *page, bool atomic, bool *submitted,
trace_f2fs_writepage(page, NODE);
- if (unlikely(f2fs_cp_error(sbi)))
- goto redirty_out;
+ if (unlikely(f2fs_cp_error(sbi))) {
+ ClearPageUptodate(page);
+ dec_page_count(sbi, F2FS_DIRTY_NODES);
+ unlock_page(page);
+ return 0;
+ }
if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
goto redirty_out;
@@ -1533,21 +1610,21 @@ static int __write_node_page(struct page *page, bool atomic, bool *submitted,
nid = nid_of_node(page);
f2fs_bug_on(sbi, page->index != nid);
- if (f2fs_get_node_info(sbi, nid, &ni))
+ if (f2fs_get_node_info(sbi, nid, &ni, !do_balance))
goto redirty_out;
if (wbc->for_reclaim) {
- if (!down_read_trylock(&sbi->node_write))
+ if (!f2fs_down_read_trylock(&sbi->node_write))
goto redirty_out;
} else {
- down_read(&sbi->node_write);
+ f2fs_down_read(&sbi->node_write);
}
/* This page is already truncated */
if (unlikely(ni.blk_addr == NULL_ADDR)) {
ClearPageUptodate(page);
dec_page_count(sbi, F2FS_DIRTY_NODES);
- up_read(&sbi->node_write);
+ f2fs_up_read(&sbi->node_write);
unlock_page(page);
return 0;
}
@@ -1555,27 +1632,28 @@ static int __write_node_page(struct page *page, bool atomic, bool *submitted,
if (__is_valid_data_blkaddr(ni.blk_addr) &&
!f2fs_is_valid_blkaddr(sbi, ni.blk_addr,
DATA_GENERIC_ENHANCE)) {
- up_read(&sbi->node_write);
+ f2fs_up_read(&sbi->node_write);
goto redirty_out;
}
- if (atomic && !test_opt(sbi, NOBARRIER))
+ if (atomic && !test_opt(sbi, NOBARRIER) && !f2fs_sb_has_blkzoned(sbi))
fio.op_flags |= REQ_PREFLUSH | REQ_FUA;
- set_page_writeback(page);
- ClearPageError(page);
-
+ /* should add to global list before clearing PAGECACHE status */
if (f2fs_in_warm_node_list(sbi, page)) {
seq = f2fs_add_fsync_node_entry(sbi, page);
if (seq_id)
*seq_id = seq;
}
+ set_page_writeback(page);
+ ClearPageError(page);
+
fio.old_blkaddr = ni.blk_addr;
f2fs_do_write_node_page(nid, &fio);
set_node_addr(sbi, &ni, fio.new_blkaddr, is_fsync_dnode(page));
dec_page_count(sbi, F2FS_DIRTY_NODES);
- up_read(&sbi->node_write);
+ f2fs_up_read(&sbi->node_write);
if (wbc->for_reclaim) {
f2fs_submit_merged_write_cond(sbi, NULL, page, 0, NODE);
@@ -1709,6 +1787,7 @@ continue_unlock:
if (!atomic || page == last_page) {
set_fsync_mark(page, 1);
+ percpu_counter_inc(&sbi->rf_node_block_count);
if (IS_INODE(page)) {
if (is_inode_flag_set(inode,
FI_DIRTY_INODE))
@@ -1716,7 +1795,7 @@ continue_unlock:
set_dentry_mark(page,
f2fs_need_dentry_mark(sbi, ino));
}
- /* may be written by other thread */
+ /* may be written by other thread */
if (!PageDirty(page))
set_page_dirty(page);
}
@@ -1760,7 +1839,7 @@ continue_unlock:
out:
if (nwritten)
f2fs_submit_merged_write_cond(sbi, NULL, NULL, ino, NODE);
- return ret ? -EIO: 0;
+ return ret ? -EIO : 0;
}
static int f2fs_match_ino(struct inode *inode, unsigned long ino, void *data)
@@ -1804,6 +1883,51 @@ static bool flush_dirty_inode(struct page *page)
return true;
}
+void f2fs_flush_inline_data(struct f2fs_sb_info *sbi)
+{
+ pgoff_t index = 0;
+ struct pagevec pvec;
+ int nr_pages;
+
+ pagevec_init(&pvec);
+
+ while ((nr_pages = pagevec_lookup_tag(&pvec,
+ NODE_MAPPING(sbi), &index, PAGECACHE_TAG_DIRTY))) {
+ int i;
+
+ for (i = 0; i < nr_pages; i++) {
+ struct page *page = pvec.pages[i];
+
+ if (!IS_DNODE(page))
+ continue;
+
+ lock_page(page);
+
+ if (unlikely(page->mapping != NODE_MAPPING(sbi))) {
+continue_unlock:
+ unlock_page(page);
+ continue;
+ }
+
+ if (!PageDirty(page)) {
+ /* someone wrote it for us */
+ goto continue_unlock;
+ }
+
+ /* flush inline_data, if it's async context. */
+ if (page_private_inline(page)) {
+ clear_page_private_inline(page);
+ unlock_page(page);
+ flush_inline_data(sbi, ino_of_node(page));
+ continue;
+ }
+ unlock_page(page);
+ }
+ pagevec_release(&pvec);
+ cond_resched();
+ }
+}
+
int f2fs_sync_node_pages(struct f2fs_sb_info *sbi,
struct writeback_control *wbc,
bool do_balance, enum iostat_type io_type)
@@ -1827,7 +1951,6 @@ next_step:
for (i = 0; i < nr_pages; i++) {
struct page *page = pvec.pages[i];
bool submitted = false;
- bool may_dirty = true;
/* give a priority to WB_SYNC threads */
if (atomic_read(&sbi->wb_sync_req[NODE]) &&
@@ -1867,21 +1990,22 @@ continue_unlock:
goto continue_unlock;
}
+ /* flush inline_data/inode, if it's async context. */
+ if (!do_balance)
+ goto write_node;
+
/* flush inline_data */
- if (is_inline_node(page)) {
- clear_inline_node(page);
+ if (page_private_inline(page)) {
+ clear_page_private_inline(page);
unlock_page(page);
flush_inline_data(sbi, ino_of_node(page));
goto lock_node;
}
/* flush dirty inode */
- if (IS_INODE(page) && may_dirty) {
- may_dirty = false;
- if (flush_dirty_inode(page))
- goto lock_node;
- }
-
+ if (IS_INODE(page) && flush_dirty_inode(page))
+ goto lock_node;
+write_node:
f2fs_wait_on_page_writeback(page, NODE, true, true);
if (!clear_page_dirty_for_io(page))
@@ -1979,7 +2103,7 @@ static int f2fs_write_node_pages(struct address_space *mapping,
goto skip_write;
/* balancing f2fs's metadata in background */
- f2fs_balance_fs_bg(sbi);
+ f2fs_balance_fs_bg(sbi, true);
/* collect a number of dirty node pages and write together */
if (wbc->sync_mode != WB_SYNC_ALL &&
@@ -1989,8 +2113,12 @@ static int f2fs_write_node_pages(struct address_space *mapping,
if (wbc->sync_mode == WB_SYNC_ALL)
atomic_inc(&sbi->wb_sync_req[NODE]);
- else if (atomic_read(&sbi->wb_sync_req[NODE]))
+ else if (atomic_read(&sbi->wb_sync_req[NODE])) {
+ /* to avoid potential deadlock */
+ if (current->plug)
+ blk_finish_plug(current->plug);
goto skip_write;
+ }
trace_f2fs_writepages(mapping->host, wbc, NODE);
@@ -2010,24 +2138,23 @@ skip_write:
return 0;
}
-static int f2fs_set_node_page_dirty(struct page *page)
+static bool f2fs_dirty_node_folio(struct address_space *mapping,
+ struct folio *folio)
{
- trace_f2fs_set_page_dirty(page, NODE);
+ trace_f2fs_set_page_dirty(&folio->page, NODE);
- if (!PageUptodate(page))
- SetPageUptodate(page);
+ if (!folio_test_uptodate(folio))
+ folio_mark_uptodate(folio);
#ifdef CONFIG_F2FS_CHECK_FS
- if (IS_INODE(page))
- f2fs_inode_chksum_set(F2FS_P_SB(page), page);
+ if (IS_INODE(&folio->page))
+ f2fs_inode_chksum_set(F2FS_M_SB(mapping), &folio->page);
#endif
- if (!PageDirty(page)) {
- __set_page_dirty_nobuffers(page);
- inc_page_count(F2FS_P_SB(page), F2FS_DIRTY_NODES);
- f2fs_set_page_private(page, 0);
- f2fs_trace_pid(page);
- return 1;
+ if (filemap_dirty_folio(mapping, folio)) {
+ inc_page_count(F2FS_M_SB(mapping), F2FS_DIRTY_NODES);
+ set_page_private_reference(&folio->page);
+ return true;
}
- return 0;
+ return false;
}
/*
@@ -2036,12 +2163,10 @@ static int f2fs_set_node_page_dirty(struct page *page)
const struct address_space_operations f2fs_node_aops = {
.writepage = f2fs_write_node_page,
.writepages = f2fs_write_node_pages,
- .set_page_dirty = f2fs_set_node_page_dirty,
- .invalidatepage = f2fs_invalidate_page,
- .releasepage = f2fs_release_page,
-#ifdef CONFIG_MIGRATION
- .migratepage = f2fs_migrate_page,
-#endif
+ .dirty_folio = f2fs_dirty_node_folio,
+ .invalidate_folio = f2fs_invalidate_folio,
+ .release_folio = f2fs_release_folio,
+ .migrate_folio = filemap_migrate_folio,
};
static struct free_nid *__lookup_free_nid_list(struct f2fs_nm_info *nm_i,
@@ -2051,18 +2176,16 @@ static struct free_nid *__lookup_free_nid_list(struct f2fs_nm_info *nm_i,
}
static int __insert_free_nid(struct f2fs_sb_info *sbi,
- struct free_nid *i, enum nid_state state)
+ struct free_nid *i)
{
struct f2fs_nm_info *nm_i = NM_I(sbi);
-
int err = radix_tree_insert(&nm_i->free_nid_root, i->nid, i);
+
if (err)
return err;
- f2fs_bug_on(sbi, state != i->state);
- nm_i->nid_cnt[state]++;
- if (state == FREE_NID)
- list_add_tail(&i->list, &nm_i->free_nid_list);
+ nm_i->nid_cnt[FREE_NID]++;
+ list_add_tail(&i->list, &nm_i->free_nid_list);
return 0;
}
@@ -2100,6 +2223,24 @@ static void __move_free_nid(struct f2fs_sb_info *sbi, struct free_nid *i,
}
}
+bool f2fs_nat_bitmap_enabled(struct f2fs_sb_info *sbi)
+{
+ struct f2fs_nm_info *nm_i = NM_I(sbi);
+ unsigned int i;
+ bool ret = true;
+
+ f2fs_down_read(&nm_i->nat_tree_lock);
+ for (i = 0; i < nm_i->nat_blocks; i++) {
+ if (!test_bit_le(i, nm_i->nat_block_bitmap)) {
+ ret = false;
+ break;
+ }
+ }
+ f2fs_up_read(&nm_i->nat_tree_lock);
+
+ return ret;
+}
+
static void update_free_nid_bitmap(struct f2fs_sb_info *sbi, nid_t nid,
bool set, bool build)
{
@@ -2141,7 +2282,7 @@ static bool add_free_nid(struct f2fs_sb_info *sbi,
if (unlikely(f2fs_check_nid_range(sbi, nid)))
return false;
- i = f2fs_kmem_cache_alloc(free_nid_slab, GFP_NOFS);
+ i = f2fs_kmem_cache_alloc(free_nid_slab, GFP_NOFS, true, NULL);
i->nid = nid;
i->state = FREE_NID;
@@ -2184,7 +2325,7 @@ static bool add_free_nid(struct f2fs_sb_info *sbi,
}
}
ret = true;
- err = __insert_free_nid(sbi, i, FREE_NID);
+ err = __insert_free_nid(sbi, i);
err_out:
if (update) {
update_free_nid_bitmap(sbi, nid, ret, build);
@@ -2278,7 +2419,7 @@ static void scan_free_nid_bits(struct f2fs_sb_info *sbi)
unsigned int i, idx;
nid_t nid;
- down_read(&nm_i->nat_tree_lock);
+ f2fs_down_read(&nm_i->nat_tree_lock);
for (i = 0; i < nm_i->nat_blocks; i++) {
if (!test_bit_le(i, nm_i->nat_block_bitmap))
@@ -2301,7 +2442,7 @@ static void scan_free_nid_bits(struct f2fs_sb_info *sbi)
out:
scan_curseg_cache(sbi);
- up_read(&nm_i->nat_tree_lock);
+ f2fs_up_read(&nm_i->nat_tree_lock);
}
static int __f2fs_build_free_nids(struct f2fs_sb_info *sbi,
@@ -2314,6 +2455,9 @@ static int __f2fs_build_free_nids(struct f2fs_sb_info *sbi,
if (unlikely(nid >= nm_i->max_nid))
nid = 0;
+ if (unlikely(nid % NAT_ENTRY_PER_BLOCK))
+ nid = NAT_BLOCK_OFFSET(nid) * NAT_ENTRY_PER_BLOCK;
+
/* Enough entries */
if (nm_i->nid_cnt[FREE_NID] >= NAT_ENTRY_PER_BLOCK)
return 0;
@@ -2333,7 +2477,7 @@ static int __f2fs_build_free_nids(struct f2fs_sb_info *sbi,
f2fs_ra_meta_pages(sbi, NAT_BLOCK_OFFSET(nid), FREE_NID_PAGES,
META_NAT, true);
- down_read(&nm_i->nat_tree_lock);
+ f2fs_down_read(&nm_i->nat_tree_lock);
while (1) {
if (!test_bit_le(NAT_BLOCK_OFFSET(nid),
@@ -2348,7 +2492,7 @@ static int __f2fs_build_free_nids(struct f2fs_sb_info *sbi,
}
if (ret) {
- up_read(&nm_i->nat_tree_lock);
+ f2fs_up_read(&nm_i->nat_tree_lock);
f2fs_err(sbi, "NAT is corrupt, run fsck to fix it");
return ret;
}
@@ -2368,7 +2512,7 @@ static int __f2fs_build_free_nids(struct f2fs_sb_info *sbi,
/* find free nids from current sum_pages */
scan_curseg_cache(sbi);
- up_read(&nm_i->nat_tree_lock);
+ f2fs_up_read(&nm_i->nat_tree_lock);
f2fs_ra_meta_pages(sbi, NAT_BLOCK_OFFSET(nm_i->next_scan_nid),
nm_i->ra_nid_pages, META_NAT, false);
@@ -2485,7 +2629,6 @@ void f2fs_alloc_nid_failed(struct f2fs_sb_info *sbi, nid_t nid)
int f2fs_try_to_free_nids(struct f2fs_sb_info *sbi, int nr_shrink)
{
struct f2fs_nm_info *nm_i = NM_I(sbi);
- struct free_nid *i, *next;
int nr = nr_shrink;
if (nm_i->nid_cnt[FREE_NID] <= MAX_FREE_NIDS)
@@ -2494,23 +2637,29 @@ int f2fs_try_to_free_nids(struct f2fs_sb_info *sbi, int nr_shrink)
if (!mutex_trylock(&nm_i->build_lock))
return 0;
- spin_lock(&nm_i->nid_list_lock);
- list_for_each_entry_safe(i, next, &nm_i->free_nid_list, list) {
- if (nr_shrink <= 0 ||
- nm_i->nid_cnt[FREE_NID] <= MAX_FREE_NIDS)
- break;
+ while (nr_shrink && nm_i->nid_cnt[FREE_NID] > MAX_FREE_NIDS) {
+ struct free_nid *i, *next;
+ unsigned int batch = SHRINK_NID_BATCH_SIZE;
- __remove_free_nid(sbi, i, FREE_NID);
- kmem_cache_free(free_nid_slab, i);
- nr_shrink--;
+ spin_lock(&nm_i->nid_list_lock);
+ list_for_each_entry_safe(i, next, &nm_i->free_nid_list, list) {
+ if (!nr_shrink || !batch ||
+ nm_i->nid_cnt[FREE_NID] <= MAX_FREE_NIDS)
+ break;
+ __remove_free_nid(sbi, i, FREE_NID);
+ kmem_cache_free(free_nid_slab, i);
+ nr_shrink--;
+ batch--;
+ }
+ spin_unlock(&nm_i->nid_list_lock);
}
- spin_unlock(&nm_i->nid_list_lock);
+
mutex_unlock(&nm_i->build_lock);
return nr - nr_shrink;
}
-void f2fs_recover_inline_xattr(struct inode *inode, struct page *page)
+int f2fs_recover_inline_xattr(struct inode *inode, struct page *page)
{
void *src_addr, *dst_addr;
size_t inline_size;
@@ -2518,13 +2667,20 @@ void f2fs_recover_inline_xattr(struct inode *inode, struct page *page)
struct f2fs_inode *ri;
ipage = f2fs_get_node_page(F2FS_I_SB(inode), inode->i_ino);
- f2fs_bug_on(F2FS_I_SB(inode), IS_ERR(ipage));
+ if (IS_ERR(ipage))
+ return PTR_ERR(ipage);
ri = F2FS_INODE(page);
if (ri->i_inline & F2FS_INLINE_XATTR) {
- set_inode_flag(inode, FI_INLINE_XATTR);
+ if (!f2fs_has_inline_xattr(inode)) {
+ set_inode_flag(inode, FI_INLINE_XATTR);
+ stat_inc_inline_xattr(inode);
+ }
} else {
- clear_inode_flag(inode, FI_INLINE_XATTR);
+ if (f2fs_has_inline_xattr(inode)) {
+ stat_dec_inline_xattr(inode);
+ clear_inode_flag(inode, FI_INLINE_XATTR);
+ }
goto update_inode;
}
@@ -2537,6 +2693,7 @@ void f2fs_recover_inline_xattr(struct inode *inode, struct page *page)
update_inode:
f2fs_update_inode(inode, ipage);
f2fs_put_page(ipage, 1);
+ return 0;
}
int f2fs_recover_xattr_data(struct inode *inode, struct page *page)
@@ -2553,7 +2710,7 @@ int f2fs_recover_xattr_data(struct inode *inode, struct page *page)
goto recover_xnid;
/* 1: invalidate the previous xattr nid */
- err = f2fs_get_node_info(sbi, prev_xnid, &ni);
+ err = f2fs_get_node_info(sbi, prev_xnid, &ni, false);
if (err)
return err;
@@ -2593,7 +2750,7 @@ int f2fs_recover_inode_page(struct f2fs_sb_info *sbi, struct page *page)
struct page *ipage;
int err;
- err = f2fs_get_node_info(sbi, ino, &old_ni);
+ err = f2fs_get_node_info(sbi, ino, &old_ni, false);
if (err)
return err;
@@ -2602,7 +2759,7 @@ int f2fs_recover_inode_page(struct f2fs_sb_info *sbi, struct page *page)
retry:
ipage = f2fs_grab_cache_page(NODE_MAPPING(sbi), ino, false);
if (!ipage) {
- congestion_wait(BLK_RW_ASYNC, HZ/50);
+ memalloc_retry_wait(GFP_NOFS);
goto retry;
}
@@ -2617,7 +2774,7 @@ retry:
src = F2FS_INODE(page);
dst = F2FS_INODE(ipage);
- memcpy(dst, src, (unsigned long)&src->i_ext - (unsigned long)src);
+ memcpy(dst, src, offsetof(struct f2fs_inode, i_ext));
dst->i_size = 0;
dst->i_blocks = cpu_to_le64(1);
dst->i_links = cpu_to_le32(1);
@@ -2670,7 +2827,7 @@ int f2fs_restore_node_summary(struct f2fs_sb_info *sbi,
sum_entry = &sum->entries[0];
for (i = 0; i < last_offset; i += nrpages, addr += nrpages) {
- nrpages = min(last_offset - i, BIO_MAX_PAGES);
+ nrpages = bio_max_segs(last_offset - i);
/* readahead node pages */
f2fs_ra_meta_pages(sbi, addr, nrpages, META_POR, true);
@@ -2708,11 +2865,14 @@ static void remove_nats_in_journal(struct f2fs_sb_info *sbi)
struct f2fs_nat_entry raw_ne;
nid_t nid = le32_to_cpu(nid_in_journal(journal, i));
+ if (f2fs_check_nid_range(sbi, nid))
+ continue;
+
raw_ne = nat_in_journal(journal, i);
ne = __lookup_nat_cache(nm_i, nid);
if (!ne) {
- ne = __alloc_nat_entry(nid, true);
+ ne = __alloc_nat_entry(sbi, nid, true);
__init_nat_entry(nm_i, ne, &raw_ne, true);
}
@@ -2752,7 +2912,23 @@ add_out:
list_add_tail(&nes->set_list, head);
}
-static void __update_nat_bits(struct f2fs_sb_info *sbi, nid_t start_nid,
+static void __update_nat_bits(struct f2fs_nm_info *nm_i, unsigned int nat_ofs,
+ unsigned int valid)
+{
+ if (valid == 0) {
+ __set_bit_le(nat_ofs, nm_i->empty_nat_bits);
+ __clear_bit_le(nat_ofs, nm_i->full_nat_bits);
+ return;
+ }
+
+ __clear_bit_le(nat_ofs, nm_i->empty_nat_bits);
+ if (valid == NAT_ENTRY_PER_BLOCK)
+ __set_bit_le(nat_ofs, nm_i->full_nat_bits);
+ else
+ __clear_bit_le(nat_ofs, nm_i->full_nat_bits);
+}
+
+static void update_nat_bits(struct f2fs_sb_info *sbi, nid_t start_nid,
struct page *page)
{
struct f2fs_nm_info *nm_i = NM_I(sbi);
@@ -2761,7 +2937,7 @@ static void __update_nat_bits(struct f2fs_sb_info *sbi, nid_t start_nid,
int valid = 0;
int i = 0;
- if (!enabled_nat_bits(sbi, NULL))
+ if (!is_set_ckpt_flags(sbi, CP_NAT_BITS_FLAG))
return;
if (nat_index == 0) {
@@ -2772,17 +2948,36 @@ static void __update_nat_bits(struct f2fs_sb_info *sbi, nid_t start_nid,
if (le32_to_cpu(nat_blk->entries[i].block_addr) != NULL_ADDR)
valid++;
}
- if (valid == 0) {
- __set_bit_le(nat_index, nm_i->empty_nat_bits);
- __clear_bit_le(nat_index, nm_i->full_nat_bits);
- return;
+
+ __update_nat_bits(nm_i, nat_index, valid);
+}
+
+void f2fs_enable_nat_bits(struct f2fs_sb_info *sbi)
+{
+ struct f2fs_nm_info *nm_i = NM_I(sbi);
+ unsigned int nat_ofs;
+
+ f2fs_down_read(&nm_i->nat_tree_lock);
+
+ for (nat_ofs = 0; nat_ofs < nm_i->nat_blocks; nat_ofs++) {
+ unsigned int valid = 0, nid_ofs = 0;
+
+ /* handle nid zero due to it should never be used */
+ if (unlikely(nat_ofs == 0)) {
+ valid = 1;
+ nid_ofs = 1;
+ }
+
+ for (; nid_ofs < NAT_ENTRY_PER_BLOCK; nid_ofs++) {
+ if (!test_bit_le(nid_ofs,
+ nm_i->free_nid_bitmap[nat_ofs]))
+ valid++;
+ }
+
+ __update_nat_bits(nm_i, nat_ofs, valid);
}
- __clear_bit_le(nat_index, nm_i->empty_nat_bits);
- if (valid == NAT_ENTRY_PER_BLOCK)
- __set_bit_le(nat_index, nm_i->full_nat_bits);
- else
- __clear_bit_le(nat_index, nm_i->full_nat_bits);
+ f2fs_up_read(&nm_i->nat_tree_lock);
}
static int __flush_nat_entry_set(struct f2fs_sb_info *sbi,
@@ -2801,7 +2996,7 @@ static int __flush_nat_entry_set(struct f2fs_sb_info *sbi,
* #1, flush nat entries to journal in current hot data summary block.
* #2, flush nat entries to nat page.
*/
- if (enabled_nat_bits(sbi, cpc) ||
+ if ((cpc->reason & CP_UMOUNT) ||
!__has_cursum_space(journal, set->entry_cnt, NAT_JOURNAL))
to_journal = false;
@@ -2848,7 +3043,7 @@ static int __flush_nat_entry_set(struct f2fs_sb_info *sbi,
if (to_journal) {
up_write(&curseg->journal_rwsem);
} else {
- __update_nat_bits(sbi, start_nid, page);
+ update_nat_bits(sbi, start_nid, page);
f2fs_put_page(page, 1);
}
@@ -2875,30 +3070,35 @@ int f2fs_flush_nat_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc)
LIST_HEAD(sets);
int err = 0;
- /* during unmount, let's flush nat_bits before checking dirty_nat_cnt */
- if (enabled_nat_bits(sbi, cpc)) {
- down_write(&nm_i->nat_tree_lock);
+ /*
+ * during unmount, let's flush nat_bits before checking
+ * nat_cnt[DIRTY_NAT].
+ */
+ if (cpc->reason & CP_UMOUNT) {
+ f2fs_down_write(&nm_i->nat_tree_lock);
remove_nats_in_journal(sbi);
- up_write(&nm_i->nat_tree_lock);
+ f2fs_up_write(&nm_i->nat_tree_lock);
}
- if (!nm_i->dirty_nat_cnt)
+ if (!nm_i->nat_cnt[DIRTY_NAT])
return 0;
- down_write(&nm_i->nat_tree_lock);
+ f2fs_down_write(&nm_i->nat_tree_lock);
/*
* if there are no enough space in journal to store dirty nat
* entries, remove all entries from journal and merge them
* into nat entry set.
*/
- if (enabled_nat_bits(sbi, cpc) ||
- !__has_cursum_space(journal, nm_i->dirty_nat_cnt, NAT_JOURNAL))
+ if (cpc->reason & CP_UMOUNT ||
+ !__has_cursum_space(journal,
+ nm_i->nat_cnt[DIRTY_NAT], NAT_JOURNAL))
remove_nats_in_journal(sbi);
while ((found = __gang_lookup_nat_set(nm_i,
set_idx, SETVEC_SIZE, setvec))) {
unsigned idx;
+
set_idx = setvec[found - 1]->set + 1;
for (idx = 0; idx < found; idx++)
__adjust_nat_entry_set(setvec[idx], &sets,
@@ -2912,7 +3112,7 @@ int f2fs_flush_nat_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc)
break;
}
- up_write(&nm_i->nat_tree_lock);
+ f2fs_up_write(&nm_i->nat_tree_lock);
/* Allow dirty nats by node block allocation in write_begin */
return err;
@@ -2927,15 +3127,18 @@ static int __get_nat_bitmaps(struct f2fs_sb_info *sbi)
__u64 cp_ver = cur_cp_version(ckpt);
block_t nat_bits_addr;
- if (!enabled_nat_bits(sbi, NULL))
- return 0;
-
nm_i->nat_bits_blocks = F2FS_BLK_ALIGN((nat_bits_bytes << 1) + 8);
- nm_i->nat_bits = f2fs_kzalloc(sbi,
+ nm_i->nat_bits = f2fs_kvzalloc(sbi,
nm_i->nat_bits_blocks << F2FS_BLKSIZE_BITS, GFP_KERNEL);
if (!nm_i->nat_bits)
return -ENOMEM;
+ nm_i->full_nat_bits = nm_i->nat_bits + 8;
+ nm_i->empty_nat_bits = nm_i->full_nat_bits + nat_bits_bytes;
+
+ if (!is_set_ckpt_flags(sbi, CP_NAT_BITS_FLAG))
+ return 0;
+
nat_bits_addr = __start_cp_addr(sbi) + sbi->blocks_per_seg -
nm_i->nat_bits_blocks;
for (i = 0; i < nm_i->nat_bits_blocks; i++) {
@@ -2952,13 +3155,12 @@ static int __get_nat_bitmaps(struct f2fs_sb_info *sbi)
cp_ver |= (cur_cp_crc(ckpt) << 32);
if (cpu_to_le64(cp_ver) != *(__le64 *)nm_i->nat_bits) {
- disable_nat_bits(sbi, true);
+ clear_ckpt_flags(sbi, CP_NAT_BITS_FLAG);
+ f2fs_notice(sbi, "Disable nat_bits due to incorrect cp_ver (%llu, %llu)",
+ cp_ver, le64_to_cpu(*(__le64 *)nm_i->nat_bits));
return 0;
}
- nm_i->full_nat_bits = nm_i->nat_bits + 8;
- nm_i->empty_nat_bits = nm_i->full_nat_bits + nat_bits_bytes;
-
f2fs_notice(sbi, "Found nat_bits in checkpoint");
return 0;
}
@@ -2969,7 +3171,7 @@ static inline void load_free_nid_bitmap(struct f2fs_sb_info *sbi)
unsigned int i = 0;
nid_t nid, last_nid;
- if (!enabled_nat_bits(sbi, NULL))
+ if (!is_set_ckpt_flags(sbi, CP_NAT_BITS_FLAG))
return;
for (i = 0; i < nm_i->nat_blocks; i++) {
@@ -3017,10 +3219,10 @@ static int init_node_manager(struct f2fs_sb_info *sbi)
F2FS_RESERVED_NODE_NUM;
nm_i->nid_cnt[FREE_NID] = 0;
nm_i->nid_cnt[PREALLOC_NID] = 0;
- nm_i->nat_cnt = 0;
nm_i->ram_thresh = DEF_RAM_THRESHOLD;
nm_i->ra_nid_pages = DEF_RA_NID_PAGES;
nm_i->dirty_nats_ratio = DEF_DIRTY_NAT_RATIO_THRESHOLD;
+ nm_i->max_rf_node_blocks = DEF_RF_NODE_BLOCKS;
INIT_RADIX_TREE(&nm_i->free_nid_root, GFP_ATOMIC);
INIT_LIST_HEAD(&nm_i->free_nid_list);
@@ -3031,14 +3233,11 @@ static int init_node_manager(struct f2fs_sb_info *sbi)
mutex_init(&nm_i->build_lock);
spin_lock_init(&nm_i->nid_list_lock);
- init_rwsem(&nm_i->nat_tree_lock);
+ init_f2fs_rwsem(&nm_i->nat_tree_lock);
nm_i->next_scan_nid = le32_to_cpu(sbi->ckpt->next_free_nid);
nm_i->bitmap_size = __bitmap_size(sbi, NAT_BITMAP);
version_bitmap = __bitmap_ptr(sbi, NAT_BITMAP);
- if (!version_bitmap)
- return -EFAULT;
-
nm_i->nat_bitmap = kmemdup(version_bitmap, nm_i->bitmap_size,
GFP_KERNEL);
if (!nm_i->nat_bitmap)
@@ -3064,9 +3263,9 @@ static int init_free_nid_cache(struct f2fs_sb_info *sbi)
int i;
nm_i->free_nid_bitmap =
- f2fs_kzalloc(sbi, array_size(sizeof(unsigned char *),
- nm_i->nat_blocks),
- GFP_KERNEL);
+ f2fs_kvzalloc(sbi, array_size(sizeof(unsigned char *),
+ nm_i->nat_blocks),
+ GFP_KERNEL);
if (!nm_i->free_nid_bitmap)
return -ENOMEM;
@@ -3140,7 +3339,7 @@ void f2fs_destroy_node_manager(struct f2fs_sb_info *sbi)
spin_unlock(&nm_i->nid_list_lock);
/* destroy nat cache */
- down_write(&nm_i->nat_tree_lock);
+ f2fs_down_write(&nm_i->nat_tree_lock);
while ((found = __gang_lookup_nat_cache(nm_i,
nid, NATVEC_SIZE, natvec))) {
unsigned idx;
@@ -3154,7 +3353,7 @@ void f2fs_destroy_node_manager(struct f2fs_sb_info *sbi)
__del_from_nat_cache(nm_i, natvec[idx]);
}
}
- f2fs_bug_on(sbi, nm_i->nat_cnt);
+ f2fs_bug_on(sbi, nm_i->nat_cnt[TOTAL_NAT]);
/* destroy nat set cache */
nid = 0;
@@ -3170,7 +3369,7 @@ void f2fs_destroy_node_manager(struct f2fs_sb_info *sbi)
kmem_cache_free(nat_entry_set_slab, setvec[idx]);
}
}
- up_write(&nm_i->nat_tree_lock);
+ f2fs_up_write(&nm_i->nat_tree_lock);
kvfree(nm_i->nat_block_bitmap);
if (nm_i->free_nid_bitmap) {
@@ -3188,27 +3387,27 @@ void f2fs_destroy_node_manager(struct f2fs_sb_info *sbi)
kvfree(nm_i->nat_bitmap_mir);
#endif
sbi->nm_info = NULL;
- kvfree(nm_i);
+ kfree(nm_i);
}
int __init f2fs_create_node_manager_caches(void)
{
- nat_entry_slab = f2fs_kmem_cache_create("nat_entry",
+ nat_entry_slab = f2fs_kmem_cache_create("f2fs_nat_entry",
sizeof(struct nat_entry));
if (!nat_entry_slab)
goto fail;
- free_nid_slab = f2fs_kmem_cache_create("free_nid",
+ free_nid_slab = f2fs_kmem_cache_create("f2fs_free_nid",
sizeof(struct free_nid));
if (!free_nid_slab)
goto destroy_nat_entry;
- nat_entry_set_slab = f2fs_kmem_cache_create("nat_entry_set",
+ nat_entry_set_slab = f2fs_kmem_cache_create("f2fs_nat_entry_set",
sizeof(struct nat_entry_set));
if (!nat_entry_set_slab)
goto destroy_free_nid;
- fsync_node_entry_slab = f2fs_kmem_cache_create("fsync_node_entry",
+ fsync_node_entry_slab = f2fs_kmem_cache_create("f2fs_fsync_node_entry",
sizeof(struct fsync_node_entry));
if (!fsync_node_entry_slab)
goto destroy_nat_entry_set;
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.