aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/fs
diff options
context:
space:
mode:
Diffstat (limited to 'fs')
-rw-r--r--fs/afs/dir.c108
-rw-r--r--fs/afs/dir_silly.c22
-rw-r--r--fs/afs/fsclient.c27
-rw-r--r--fs/afs/yfsclient.c26
-rw-r--r--fs/btrfs/block-group.c1
-rw-r--r--fs/btrfs/file.c15
-rw-r--r--fs/btrfs/reflink.c1
-rw-r--r--fs/btrfs/relocation.c23
-rw-r--r--fs/btrfs/space-info.c20
-rw-r--r--fs/btrfs/tree-log.c93
-rw-r--r--fs/ceph/dir.c4
-rw-r--r--fs/ceph/file.c4
-rw-r--r--fs/ceph/mds_client.h2
-rw-r--r--fs/io_uring.c301
-rw-r--r--fs/nfs/pnfs.c3
-rw-r--r--fs/proc/root.c7
16 files changed, 352 insertions, 305 deletions
diff --git a/fs/afs/dir.c b/fs/afs/dir.c
index 5c794f4b051a..d1e1caa23c8b 100644
--- a/fs/afs/dir.c
+++ b/fs/afs/dir.c
@@ -1032,7 +1032,7 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags)
struct dentry *parent;
struct inode *inode;
struct key *key;
- afs_dataversion_t dir_version;
+ afs_dataversion_t dir_version, invalid_before;
long de_version;
int ret;
@@ -1084,8 +1084,8 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags)
if (de_version == (long)dir_version)
goto out_valid_noupdate;
- dir_version = dir->invalid_before;
- if (de_version - (long)dir_version >= 0)
+ invalid_before = dir->invalid_before;
+ if (de_version - (long)invalid_before >= 0)
goto out_valid;
_debug("dir modified");
@@ -1275,6 +1275,7 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
struct afs_fs_cursor fc;
struct afs_vnode *dvnode = AFS_FS_I(dir);
struct key *key;
+ afs_dataversion_t data_version;
int ret;
mode |= S_IFDIR;
@@ -1295,7 +1296,7 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
- afs_dataversion_t data_version = dvnode->status.data_version + 1;
+ data_version = dvnode->status.data_version + 1;
while (afs_select_fileserver(&fc)) {
fc.cb_break = afs_calc_vnode_cb_break(dvnode);
@@ -1316,10 +1317,14 @@ static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
goto error_key;
}
- if (ret == 0 &&
- test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
- afs_edit_dir_add(dvnode, &dentry->d_name, &iget_data.fid,
- afs_edit_dir_for_create);
+ if (ret == 0) {
+ down_write(&dvnode->validate_lock);
+ if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+ dvnode->status.data_version == data_version)
+ afs_edit_dir_add(dvnode, &dentry->d_name, &iget_data.fid,
+ afs_edit_dir_for_create);
+ up_write(&dvnode->validate_lock);
+ }
key_put(key);
kfree(scb);
@@ -1360,6 +1365,7 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry)
struct afs_fs_cursor fc;
struct afs_vnode *dvnode = AFS_FS_I(dir), *vnode = NULL;
struct key *key;
+ afs_dataversion_t data_version;
int ret;
_enter("{%llx:%llu},{%pd}",
@@ -1391,7 +1397,7 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry)
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
- afs_dataversion_t data_version = dvnode->status.data_version + 1;
+ data_version = dvnode->status.data_version + 1;
while (afs_select_fileserver(&fc)) {
fc.cb_break = afs_calc_vnode_cb_break(dvnode);
@@ -1404,9 +1410,12 @@ static int afs_rmdir(struct inode *dir, struct dentry *dentry)
ret = afs_end_vnode_operation(&fc);
if (ret == 0) {
afs_dir_remove_subdir(dentry);
- if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
+ down_write(&dvnode->validate_lock);
+ if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+ dvnode->status.data_version == data_version)
afs_edit_dir_remove(dvnode, &dentry->d_name,
afs_edit_dir_for_rmdir);
+ up_write(&dvnode->validate_lock);
}
}
@@ -1544,10 +1553,15 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry)
ret = afs_end_vnode_operation(&fc);
if (ret == 0 && !(scb[1].have_status || scb[1].have_error))
ret = afs_dir_remove_link(dvnode, dentry, key);
- if (ret == 0 &&
- test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
- afs_edit_dir_remove(dvnode, &dentry->d_name,
- afs_edit_dir_for_unlink);
+
+ if (ret == 0) {
+ down_write(&dvnode->validate_lock);
+ if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+ dvnode->status.data_version == data_version)
+ afs_edit_dir_remove(dvnode, &dentry->d_name,
+ afs_edit_dir_for_unlink);
+ up_write(&dvnode->validate_lock);
+ }
}
if (need_rehash && ret < 0 && ret != -ENOENT)
@@ -1573,6 +1587,7 @@ static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
struct afs_status_cb *scb;
struct afs_vnode *dvnode = AFS_FS_I(dir);
struct key *key;
+ afs_dataversion_t data_version;
int ret;
mode |= S_IFREG;
@@ -1597,7 +1612,7 @@ static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
- afs_dataversion_t data_version = dvnode->status.data_version + 1;
+ data_version = dvnode->status.data_version + 1;
while (afs_select_fileserver(&fc)) {
fc.cb_break = afs_calc_vnode_cb_break(dvnode);
@@ -1618,9 +1633,12 @@ static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
goto error_key;
}
- if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
+ down_write(&dvnode->validate_lock);
+ if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+ dvnode->status.data_version == data_version)
afs_edit_dir_add(dvnode, &dentry->d_name, &iget_data.fid,
afs_edit_dir_for_create);
+ up_write(&dvnode->validate_lock);
kfree(scb);
key_put(key);
@@ -1648,6 +1666,7 @@ static int afs_link(struct dentry *from, struct inode *dir,
struct afs_vnode *dvnode = AFS_FS_I(dir);
struct afs_vnode *vnode = AFS_FS_I(d_inode(from));
struct key *key;
+ afs_dataversion_t data_version;
int ret;
_enter("{%llx:%llu},{%llx:%llu},{%pd}",
@@ -1672,7 +1691,7 @@ static int afs_link(struct dentry *from, struct inode *dir,
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
- afs_dataversion_t data_version = dvnode->status.data_version + 1;
+ data_version = dvnode->status.data_version + 1;
if (mutex_lock_interruptible_nested(&vnode->io_lock, 1) < 0) {
afs_end_vnode_operation(&fc);
@@ -1702,9 +1721,12 @@ static int afs_link(struct dentry *from, struct inode *dir,
goto error_key;
}
- if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
+ down_write(&dvnode->validate_lock);
+ if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+ dvnode->status.data_version == data_version)
afs_edit_dir_add(dvnode, &dentry->d_name, &vnode->fid,
afs_edit_dir_for_link);
+ up_write(&dvnode->validate_lock);
key_put(key);
kfree(scb);
@@ -1732,6 +1754,7 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry,
struct afs_status_cb *scb;
struct afs_vnode *dvnode = AFS_FS_I(dir);
struct key *key;
+ afs_dataversion_t data_version;
int ret;
_enter("{%llx:%llu},{%pd},%s",
@@ -1759,7 +1782,7 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry,
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
- afs_dataversion_t data_version = dvnode->status.data_version + 1;
+ data_version = dvnode->status.data_version + 1;
while (afs_select_fileserver(&fc)) {
fc.cb_break = afs_calc_vnode_cb_break(dvnode);
@@ -1780,9 +1803,12 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry,
goto error_key;
}
- if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
+ down_write(&dvnode->validate_lock);
+ if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+ dvnode->status.data_version == data_version)
afs_edit_dir_add(dvnode, &dentry->d_name, &iget_data.fid,
afs_edit_dir_for_symlink);
+ up_write(&dvnode->validate_lock);
key_put(key);
kfree(scb);
@@ -1812,6 +1838,8 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
struct dentry *tmp = NULL, *rehash = NULL;
struct inode *new_inode;
struct key *key;
+ afs_dataversion_t orig_data_version;
+ afs_dataversion_t new_data_version;
bool new_negative = d_is_negative(new_dentry);
int ret;
@@ -1890,10 +1918,6 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
ret = -ERESTARTSYS;
if (afs_begin_vnode_operation(&fc, orig_dvnode, key, true)) {
- afs_dataversion_t orig_data_version;
- afs_dataversion_t new_data_version;
- struct afs_status_cb *new_scb = &scb[1];
-
orig_data_version = orig_dvnode->status.data_version + 1;
if (orig_dvnode != new_dvnode) {
@@ -1904,7 +1928,6 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
new_data_version = new_dvnode->status.data_version + 1;
} else {
new_data_version = orig_data_version;
- new_scb = &scb[0];
}
while (afs_select_fileserver(&fc)) {
@@ -1912,7 +1935,7 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
fc.cb_break_2 = afs_calc_vnode_cb_break(new_dvnode);
afs_fs_rename(&fc, old_dentry->d_name.name,
new_dvnode, new_dentry->d_name.name,
- &scb[0], new_scb);
+ &scb[0], &scb[1]);
}
afs_vnode_commit_status(&fc, orig_dvnode, fc.cb_break,
@@ -1930,18 +1953,25 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
if (ret == 0) {
if (rehash)
d_rehash(rehash);
- if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags))
- afs_edit_dir_remove(orig_dvnode, &old_dentry->d_name,
- afs_edit_dir_for_rename_0);
+ down_write(&orig_dvnode->validate_lock);
+ if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags) &&
+ orig_dvnode->status.data_version == orig_data_version)
+ afs_edit_dir_remove(orig_dvnode, &old_dentry->d_name,
+ afs_edit_dir_for_rename_0);
+ if (orig_dvnode != new_dvnode) {
+ up_write(&orig_dvnode->validate_lock);
- if (!new_negative &&
- test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags))
- afs_edit_dir_remove(new_dvnode, &new_dentry->d_name,
- afs_edit_dir_for_rename_1);
+ down_write(&new_dvnode->validate_lock);
+ }
+ if (test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags) &&
+ orig_dvnode->status.data_version == new_data_version) {
+ if (!new_negative)
+ afs_edit_dir_remove(new_dvnode, &new_dentry->d_name,
+ afs_edit_dir_for_rename_1);
- if (test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags))
afs_edit_dir_add(new_dvnode, &new_dentry->d_name,
&vnode->fid, afs_edit_dir_for_rename_2);
+ }
new_inode = d_inode(new_dentry);
if (new_inode) {
@@ -1957,14 +1987,10 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry,
* Note that if we ever implement RENAME_EXCHANGE, we'll have
* to update both dentries with opposing dir versions.
*/
- if (new_dvnode != orig_dvnode) {
- afs_update_dentry_version(&fc, old_dentry, &scb[1]);
- afs_update_dentry_version(&fc, new_dentry, &scb[1]);
- } else {
- afs_update_dentry_version(&fc, old_dentry, &scb[0]);
- afs_update_dentry_version(&fc, new_dentry, &scb[0]);
- }
+ afs_update_dentry_version(&fc, old_dentry, &scb[1]);
+ afs_update_dentry_version(&fc, new_dentry, &scb[1]);
d_move(old_dentry, new_dentry);
+ up_write(&new_dvnode->validate_lock);
goto error_tmp;
}
diff --git a/fs/afs/dir_silly.c b/fs/afs/dir_silly.c
index 361088a5edb9..d94e2b7cddff 100644
--- a/fs/afs/dir_silly.c
+++ b/fs/afs/dir_silly.c
@@ -21,6 +21,7 @@ static int afs_do_silly_rename(struct afs_vnode *dvnode, struct afs_vnode *vnode
{
struct afs_fs_cursor fc;
struct afs_status_cb *scb;
+ afs_dataversion_t dir_data_version;
int ret = -ERESTARTSYS;
_enter("%pd,%pd", old, new);
@@ -31,7 +32,7 @@ static int afs_do_silly_rename(struct afs_vnode *dvnode, struct afs_vnode *vnode
trace_afs_silly_rename(vnode, false);
if (afs_begin_vnode_operation(&fc, dvnode, key, true)) {
- afs_dataversion_t dir_data_version = dvnode->status.data_version + 1;
+ dir_data_version = dvnode->status.data_version + 1;
while (afs_select_fileserver(&fc)) {
fc.cb_break = afs_calc_vnode_cb_break(dvnode);
@@ -54,12 +55,15 @@ static int afs_do_silly_rename(struct afs_vnode *dvnode, struct afs_vnode *vnode
dvnode->silly_key = key_get(key);
}
- if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
+ down_write(&dvnode->validate_lock);
+ if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+ dvnode->status.data_version == dir_data_version) {
afs_edit_dir_remove(dvnode, &old->d_name,
afs_edit_dir_for_silly_0);
- if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
afs_edit_dir_add(dvnode, &new->d_name,
&vnode->fid, afs_edit_dir_for_silly_1);
+ }
+ up_write(&dvnode->validate_lock);
}
kfree(scb);
@@ -181,10 +185,14 @@ static int afs_do_silly_unlink(struct afs_vnode *dvnode, struct afs_vnode *vnode
clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags);
}
}
- if (ret == 0 &&
- test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags))
- afs_edit_dir_remove(dvnode, &dentry->d_name,
- afs_edit_dir_for_unlink);
+ if (ret == 0) {
+ down_write(&dvnode->validate_lock);
+ if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) &&
+ dvnode->status.data_version == dir_data_version)
+ afs_edit_dir_remove(dvnode, &dentry->d_name,
+ afs_edit_dir_for_unlink);
+ up_write(&dvnode->validate_lock);
+ }
}
kfree(scb);
diff --git a/fs/afs/fsclient.c b/fs/afs/fsclient.c
index 1f9c5d8e6fe5..68fc46634346 100644
--- a/fs/afs/fsclient.c
+++ b/fs/afs/fsclient.c
@@ -65,6 +65,7 @@ static int xdr_decode_AFSFetchStatus(const __be32 **_bp,
bool inline_error = (call->operation_ID == afs_FS_InlineBulkStatus);
u64 data_version, size;
u32 type, abort_code;
+ int ret;
abort_code = ntohl(xdr->abort_code);
@@ -78,7 +79,7 @@ static int xdr_decode_AFSFetchStatus(const __be32 **_bp,
*/
status->abort_code = abort_code;
scb->have_error = true;
- return 0;
+ goto good;
}
pr_warn("Unknown AFSFetchStatus version %u\n", ntohl(xdr->if_version));
@@ -87,7 +88,8 @@ static int xdr_decode_AFSFetchStatus(const __be32 **_bp,
if (abort_code != 0 && inline_error) {
status->abort_code = abort_code;
- return 0;
+ scb->have_error = true;
+ goto good;
}
type = ntohl(xdr->type);
@@ -123,13 +125,16 @@ static int xdr_decode_AFSFetchStatus(const __be32 **_bp,
data_version |= (u64)ntohl(xdr->data_version_hi) << 32;
status->data_version = data_version;
scb->have_status = true;
-
+good:
+ ret = 0;
+advance:
*_bp = (const void *)*_bp + sizeof(*xdr);
- return 0;
+ return ret;
bad:
xdr_dump_bad(*_bp);
- return afs_protocol_error(call, -EBADMSG, afs_eproto_bad_status);
+ ret = afs_protocol_error(call, -EBADMSG, afs_eproto_bad_status);
+ goto advance;
}
static time64_t xdr_decode_expiry(struct afs_call *call, u32 expiry)
@@ -981,16 +986,16 @@ static int afs_deliver_fs_rename(struct afs_call *call)
if (ret < 0)
return ret;
- /* unmarshall the reply once we've received all of it */
+ /* If the two dirs are the same, we have two copies of the same status
+ * report, so we just decode it twice.
+ */
bp = call->buffer;
ret = xdr_decode_AFSFetchStatus(&bp, call, call->out_dir_scb);
if (ret < 0)
return ret;
- if (call->out_dir_scb != call->out_scb) {
- ret = xdr_decode_AFSFetchStatus(&bp, call, call->out_scb);
- if (ret < 0)
- return ret;
- }
+ ret = xdr_decode_AFSFetchStatus(&bp, call, call->out_scb);
+ if (ret < 0)
+ return ret;
xdr_decode_AFSVolSync(&bp, call->out_volsync);
_leave(" = 0 [done]");
diff --git a/fs/afs/yfsclient.c b/fs/afs/yfsclient.c
index a26126ac7bf1..b5b45c57e1b1 100644
--- a/fs/afs/yfsclient.c
+++ b/fs/afs/yfsclient.c
@@ -165,15 +165,15 @@ static void xdr_dump_bad(const __be32 *bp)
int i;
pr_notice("YFS XDR: Bad status record\n");
- for (i = 0; i < 5 * 4 * 4; i += 16) {
+ for (i = 0; i < 6 * 4 * 4; i += 16) {
memcpy(x, bp, 16);
bp += 4;
pr_notice("%03x: %08x %08x %08x %08x\n",
i, ntohl(x[0]), ntohl(x[1]), ntohl(x[2]), ntohl(x[3]));
}
- memcpy(x, bp, 4);
- pr_notice("0x50: %08x\n", ntohl(x[0]));
+ memcpy(x, bp, 8);
+ pr_notice("0x60: %08x %08x\n", ntohl(x[0]), ntohl(x[1]));
}
/*
@@ -186,13 +186,14 @@ static int xdr_decode_YFSFetchStatus(const __be32 **_bp,
const struct yfs_xdr_YFSFetchStatus *xdr = (const void *)*_bp;
struct afs_file_status *status = &scb->status;
u32 type;
+ int ret;
status->abort_code = ntohl(xdr->abort_code);
if (status->abort_code != 0) {
if (status->abort_code == VNOVNODE)
status->nlink = 0;
scb->have_error = true;
- return 0;
+ goto good;
}
type = ntohl(xdr->type);
@@ -220,13 +221,16 @@ static int xdr_decode_YFSFetchStatus(const __be32 **_bp,
status->size = xdr_to_u64(xdr->size);
status->data_version = xdr_to_u64(xdr->data_version);
scb->have_status = true;
-
+good:
+ ret = 0;
+advance:
*_bp += xdr_size(xdr);
- return 0;
+ return ret;
bad:
xdr_dump_bad(*_bp);
- return afs_protocol_error(call, -EBADMSG, afs_eproto_bad_status);
+ ret = afs_protocol_error(call, -EBADMSG, afs_eproto_bad_status);
+ goto advance;
}
/*
@@ -1153,11 +1157,9 @@ static int yfs_deliver_fs_rename(struct afs_call *call)
ret = xdr_decode_YFSFetchStatus(&bp, call, call->out_dir_scb);
if (ret < 0)
return ret;
- if (call->out_dir_scb != call->out_scb) {
- ret = xdr_decode_YFSFetchStatus(&bp, call, call->out_scb);
- if (ret < 0)
- return ret;
- }
+ ret = xdr_decode_YFSFetchStatus(&bp, call, call->out_scb);
+ if (ret < 0)
+ return ret;
xdr_decode_YFSVolSync(&bp, call->out_volsync);
_leave(" = 0 [done]");
diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c
index 786849fcc319..47f66c6a7d7f 100644
--- a/fs/btrfs/block-group.c
+++ b/fs/btrfs/block-group.c
@@ -3370,6 +3370,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info)
space_info->bytes_reserved > 0 ||
space_info->bytes_may_use > 0))
btrfs_dump_space_info(info, space_info, 0, 0);
+ WARN_ON(space_info->reclaim_size > 0);
list_del(&space_info->list);
btrfs_sysfs_remove_space_info(space_info);
}
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 8a144f9cb7ac..719e68ab552c 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -2098,6 +2098,21 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
atomic_inc(&root->log_batch);
/*
+ * If the inode needs a full sync, make sure we use a full range to
+ * avoid log tree corruption, due to hole detection racing with ordered
+ * extent completion for adjacent ranges and races between logging and
+ * completion of ordered extents for adjancent ranges - both races
+ * could lead to file extent items in the log with overlapping ranges.
+ * Do this while holding the inode lock, to avoid races with other
+ * tasks.
+ */
+ if (test_bit(BTRFS_INODE_NEEDS_FULL_SYNC,
+ &BTRFS_I(inode)->runtime_flags)) {
+ start = 0;
+ end = LLONG_MAX;
+ }
+
+ /*
* Before we acquired the inode's lock, someone may have dirtied more
* pages in the target range. We need to make sure that writeback for
* any such pages does not start while we are logging the inode, because
diff --git a/fs/btrfs/reflink.c b/fs/btrfs/reflink.c
index d1973141d3bb..040009d1cc31 100644
--- a/fs/btrfs/reflink.c
+++ b/fs/btrfs/reflink.c
@@ -264,6 +264,7 @@ copy_inline_extent:
size);
inode_add_bytes(dst, datal);
set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(dst)->runtime_flags);
+ ret = btrfs_inode_set_file_extent_range(BTRFS_I(dst), 0, aligned_end);
out:
if (!ret && !trans) {
/*
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index f65595602aa8..d35936c934ab 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -611,8 +611,8 @@ static int should_ignore_root(struct btrfs_root *root)
if (!reloc_root)
return 0;
- if (btrfs_root_last_snapshot(&reloc_root->root_item) ==
- root->fs_info->running_transaction->transid - 1)
+ if (btrfs_header_generation(reloc_root->commit_root) ==
+ root->fs_info->running_transaction->transid)
return 0;
/*
* if there is reloc tree and it was created in previous
@@ -1527,8 +1527,7 @@ int btrfs_init_reloc_root(struct btrfs_trans_handle *trans,
int clear_rsv = 0;
int ret;
- if (!rc || !rc->create_reloc_tree ||
- root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID)
+ if (!rc)
return 0;
/*
@@ -1538,12 +1537,28 @@ int btrfs_init_reloc_root(struct btrfs_trans_handle *trans,
if (reloc_root_is_dead(root))
return 0;
+ /*
+ * This is subtle but important. We do not do
+ * record_root_in_transaction for reloc roots, instead we record their
+ * corresponding fs root, and then here we update the last trans for the
+ * reloc root. This means that we have to do this for the entire life
+ * of the reloc root, regardless of which stage of the relocation we are
+ * in.
+ */
if (root->reloc_root) {
reloc_root = root->reloc_root;
reloc_root->last_trans = trans->transid;
return 0;
}
+ /*
+ * We are merging reloc roots, we do not need new reloc trees. Also
+ * reloc trees never need their own reloc tree.
+ */
+ if (!rc->create_reloc_tree ||
+ root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID)
+ return 0;
+
if (!trans->reloc_reserved) {
rsv = trans->block_rsv;
trans->block_rsv = rc->block_rsv;
diff --git a/fs/btrfs/space-info.c b/fs/btrfs/space-info.c
index 8b0fe053a25d..ff17a4420358 100644
--- a/fs/btrfs/space-info.c
+++ b/fs/btrfs/space-info.c
@@ -361,6 +361,16 @@ int btrfs_can_overcommit(struct btrfs_fs_info *fs_info,
return 0;
}
+static void remove_ticket(struct btrfs_space_info *space_info,
+ struct reserve_ticket *ticket)
+{
+ if (!list_empty(&ticket->list)) {
+ list_del_init(&ticket->list);
+ ASSERT(space_info->reclaim_size >= ticket->bytes);
+ space_info->reclaim_size -= ticket->bytes;
+ }
+}
+
/*
* This is for space we already have accounted in space_info->bytes_may_use, so
* basically when we're returning space from block_rsv's.
@@ -388,9 +398,7 @@ again:
btrfs_space_info_update_bytes_may_use(fs_info,
space_info,
ticket->bytes);
- list_del_init(&ticket->list);
- ASSERT(space_info->reclaim_size >= ticket->bytes);
- space_info->reclaim_size -= ticket->bytes;
+ remove_ticket(space_info, ticket);
ticket->bytes = 0;
space_info->tickets_id++;
wake_up(&ticket->wait);
@@ -899,7 +907,7 @@ static bool maybe_fail_all_tickets(struct btrfs_fs_info *fs_info,
btrfs_info(fs_info, "failing ticket with %llu bytes",
ticket->bytes);
- list_del_init(&ticket->list);
+ remove_ticket(space_info, ticket);
ticket->error = -ENOSPC;
wake_up(&ticket->wait);
@@ -1063,7 +1071,7 @@ static void wait_reserve_ticket(struct btrfs_fs_info *fs_info,
* despite getting an error, resulting in a space leak
* (bytes_may_use counter of our space_info).
*/
- list_del_init(&ticket->list);
+ remove_ticket(space_info, ticket);
ticket->error = -EINTR;
break;
}
@@ -1121,7 +1129,7 @@ static int handle_reserve_ticket(struct btrfs_fs_info *fs_info,
* either the async reclaim job deletes the ticket from the list
* or we delete it ourselves at wait_reserve_ticket().
*/
- list_del_init(&ticket->list);
+ remove_ticket(space_info, ticket);
if (!ret)
ret = -ENOSPC;
}
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 58c111474ba5..ec36a7c6ba3d 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -96,8 +96,8 @@ enum {
static int btrfs_log_inode(struct btrfs_trans_handle *trans,
struct btrfs_root *root, struct btrfs_inode *inode,
int inode_only,
- u64 start,
- u64 end,
+ const loff_t start,
+ const loff_t end,
struct btrfs_log_ctx *ctx);
static int link_to_fixup_dir(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
@@ -4533,15 +4533,13 @@ static int btrfs_log_all_xattrs(struct btrfs_trans_handle *trans,
static int btrfs_log_holes(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_inode *inode,
- struct btrfs_path *path,
- const u64 start,
- const u64 end)
+ struct btrfs_path *path)
{
struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_key key;
const u64 ino = btrfs_ino(inode);
const u64 i_size = i_size_read(&inode->vfs_inode);
- u64 prev_extent_end = start;
+ u64 prev_extent_end = 0;
int ret;
if (!btrfs_fs_incompat(fs_info, NO_HOLES) || i_size == 0)
@@ -4549,21 +4547,14 @@ static int btrfs_log_holes(struct btrfs_trans_handle *trans,
key.objectid = ino;
key.type = BTRFS_EXTENT_DATA_KEY;
- key.offset = start;
+ key.offset = 0;
ret = btrfs_search_slot(NULL, root, &key, path, 0, 0);
if (ret < 0)
return ret;
- if (ret > 0 && path->slots[0] > 0) {
- btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0] - 1);
- if (key.objectid == ino && key.type == BTRFS_EXTENT_DATA_KEY)
- path->slots[0]--;
- }
-
while (true) {
struct extent_buffer *leaf = path->nodes[0];
- u64 extent_end;
if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) {
ret = btrfs_next_leaf(root, path);
@@ -4580,18 +4571,9 @@ static int btrfs_log_holes(struct btrfs_trans_handle *trans,
if (key.objectid != ino || key.type != BTRFS_EXTENT_DATA_KEY)
break;
- extent_end = btrfs_file_extent_end(path);
- if (extent_end <= start)
- goto next_slot;
-
/* We have a hole, log it. */
if (prev_extent_end < key.offset) {
- u64 hole_len;
-
- if (key.offset >= end)
- hole_len = end - prev_extent_end;
- else
- hole_len = key.offset - prev_extent_end;
+ const u64 hole_len = key.offset - prev_extent_end;
/*
* Release the path to avoid deadlocks with other code
@@ -4621,20 +4603,16 @@ static int btrfs_log_holes(struct btrfs_trans_handle *trans,
leaf = path->nodes[0];
}
- prev_extent_end = min(extent_end, end);
- if (extent_end >= end)
- break;
-next_slot:
+ prev_extent_end = btrfs_file_extent_end(path);
path->slots[0]++;
cond_resched();
}
- if (prev_extent_end < end && prev_extent_end < i_size) {
+ if (prev_extent_end < i_size) {
u64 hole_len;
btrfs_release_path(path);
- hole_len = min(ALIGN(i_size, fs_info->sectorsize), end);
- hole_len -= prev_extent_end;
+ hole_len = ALIGN(i_size - prev_extent_end, fs_info->sectorsize);
ret = btrfs_insert_file_extent(trans, root->log_root,
ino, prev_extent_end, 0, 0,
hole_len, 0, hole_len,
@@ -4971,8 +4949,6 @@ static int copy_inode_items_to_log(struct btrfs_trans_handle *trans,
const u64 logged_isize,
const bool recursive_logging,
const int inode_only,
- const u64 start,
- const u64 end,
struct btrfs_log_ctx *ctx,
bool *need_log_inode_item)
{
@@ -4981,21 +4957,6 @@ static int copy_inode_items_to_log(struct btrfs_trans_handle *trans,
int ins_nr = 0;
int ret;
- /*
- * We must make sure we don't copy extent items that are entirely out of
- * the range [start, end - 1]. This is not just an optimization to avoid
- * copying but also needed to avoid a corruption where we end up with
- * file extent items in the log tree that have overlapping ranges - this
- * can happen if we race with ordered extent completion for ranges that
- * are outside our target range. For example we copy an extent item and
- * when we move to the next leaf, that extent was trimmed and a new one
- * covering a subrange of it, but with a higher key, was inserted - we
- * would then copy this other extent too, resulting in a log tree with
- * 2 extent items that represent overlapping ranges.
- *
- * We can copy the entire extents at the range bondaries however, even
- * if they cover an area outside the target range. That's ok.
- */
while (1) {
ret = btrfs_search_forward(root, min_key, path, trans->transid);
if (ret < 0)
@@ -5063,29 +5024,6 @@ again:
goto next_slot;
}
- if (min_key->type == BTRFS_EXTENT_DATA_KEY) {
- const u64 extent_end = btrfs_file_extent_end(path);
-
- if (extent_end <= start) {
- if (ins_nr > 0) {
- ret = copy_items(trans, inode, dst_path,
- path, ins_start_slot,
- ins_nr, inode_only,
- logged_isize);
- if (ret < 0)
- return ret;
- ins_nr = 0;
- }
- goto next_slot;
- }
- if (extent_end >= end) {
- ins_nr++;
- if (ins_nr == 1)
- ins_start_slot = path->slots[0];
- break;
- }
- }
-
if (ins_nr && ins_start_slot + ins_nr == path->slots[0]) {
ins_nr++;
goto next_slot;
@@ -5151,8 +5089,8 @@ next_key:
static int btrfs_log_inode(struct btrfs_trans_handle *trans,
struct btrfs_root *root, struct btrfs_inode *inode,
int inode_only,
- u64 start,
- u64 end,
+ const loff_t start,
+ const loff_t end,
struct btrfs_log_ctx *ctx)
{
struct btrfs_fs_info *fs_info = root->fs_info;
@@ -5180,9 +5118,6 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
return -ENOMEM;
}
- start = ALIGN_DOWN(start, fs_info->sectorsize);
- end = ALIGN(end, fs_info->sectorsize);
-
min_key.objectid = ino;
min_key.type = BTRFS_INODE_ITEM_KEY;
min_key.offset = 0;
@@ -5298,8 +5233,8 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
err = copy_inode_items_to_log(trans, inode, &min_key, &max_key,
path, dst_path, logged_isize,
- recursive_logging, inode_only,
- start, end, ctx, &need_log_inode_item);
+ recursive_logging, inode_only, ctx,
+ &need_log_inode_item);
if (err)
goto out_unlock;
@@ -5312,7 +5247,7 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans,
if (max_key.type >= BTRFS_EXTENT_DATA_KEY && !fast_search) {
btrfs_release_path(path);
btrfs_release_path(dst_path);
- err = btrfs_log_holes(trans, root, inode, path, start, end);
+ err = btrfs_log_holes(trans, root, inode, path);
if (err)
goto out_unlock;
}
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index d594c2627430..4c4202c93b71 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -1051,8 +1051,8 @@ static void ceph_async_unlink_cb(struct ceph_mds_client *mdsc,
/* If op failed, mark everyone involved for errors */
if (result) {
- int pathlen;
- u64 base;
+ int pathlen = 0;
+ u64 base = 0;
char *path = ceph_mdsc_build_path(req->r_dentry, &pathlen,
&base, 0);
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 4a5ccbb7e808..afdfca965a7f 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -527,8 +527,8 @@ static void ceph_async_create_cb(struct ceph_mds_client *mdsc,
if (result) {
struct dentry *dentry = req->r_dentry;
- int pathlen;
- u64 base;
+ int pathlen = 0;
+ u64 base = 0;
char *path = ceph_mdsc_build_path(req->r_dentry, &pathlen,
&base, 0);
diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
index 4e5be79bf080..903d9edfd4bf 100644
--- a/fs/ceph/mds_client.h
+++ b/fs/ceph/mds_client.h
@@ -521,7 +521,7 @@ extern void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc);
static inline void ceph_mdsc_free_path(char *path, int len)
{
- if (path)
+ if (!IS_ERR_OR_NULL(path))
__putname(path - (PATH_MAX - 1 - len));
}
diff --git a/fs/io_uring.c b/fs/io_uring.c
index 5190bfb6a665..381d50becd04 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -357,7 +357,6 @@ struct io_timeout_data {
struct hrtimer timer;
struct timespec64 ts;
enum hrtimer_mode mode;
- u32 seq_offset;
};
struct io_accept {
@@ -385,7 +384,7 @@ struct io_timeout {
struct file *file;
u64 addr;
int flags;
- unsigned count;
+ u32 count;
};
struct io_rw {
@@ -508,6 +507,7 @@ enum {
REQ_F_FORCE_ASYNC_BIT = IOSQE_ASYNC_BIT,
REQ_F_BUFFER_SELECT_BIT = IOSQE_BUFFER_SELECT_BIT,
+ REQ_F_LINK_HEAD_BIT,
REQ_F_LINK_NEXT_BIT,
REQ_F_FAIL_LINK_BIT,
REQ_F_INFLIGHT_BIT,
@@ -543,6 +543,8 @@ enum {
/* IOSQE_BUFFER_SELECT */
REQ_F_BUFFER_SELECT = BIT(REQ_F_BUFFER_SELECT_BIT),
+ /* head of a link */
+ REQ_F_LINK_HEAD = BIT(REQ_F_LINK_HEAD_BIT),
/* already grabbed next link */
REQ_F_LINK_NEXT = BIT(REQ_F_LINK_NEXT_BIT),
/* fail rest of links */
@@ -955,8 +957,8 @@ static inline bool __req_need_defer(struct io_kiocb *req)
{
struct io_ring_ctx *ctx = req->ctx;
- return req->sequence != ctx->cached_cq_tail + ctx->cached_sq_dropped
- + atomic_read(&ctx->cached_cq_overflow);
+ return req->sequence != ctx->cached_cq_tail
+ + atomic_read(&ctx->cached_cq_overflow);
}
static inline bool req_need_defer(struct io_kiocb *req)
@@ -1437,7 +1439,7 @@ static bool io_link_cancel_timeout(struct io_kiocb *req)
if (ret != -1) {
io_cqring_fill_event(req, -ECANCELED);
io_commit_cqring(ctx);
- req->flags &= ~REQ_F_LINK;
+ req->flags &= ~REQ_F_LINK_HEAD;
io_put_req(req);
return true;
}
@@ -1473,7 +1475,7 @@ static void io_req_link_next(struct io_kiocb *req, struct io_kiocb **nxtptr)
list_del_init(&req->link_list);
if (!list_empty(&nxt->link_list))
- nxt->flags |= REQ_F_LINK;
+ nxt->flags |= REQ_F_LINK_HEAD;
*nxtptr = nxt;
break;
}
@@ -1484,7 +1486,7 @@ static void io_req_link_next(struct io_kiocb *req, struct io_kiocb **nxtptr)
}
/*
- * Called if REQ_F_LINK is set, and we fail the head request
+ * Called if REQ_F_LINK_HEAD is set, and we fail the head request
*/
static void io_fail_links(struct io_kiocb *req)
{
@@ -1517,7 +1519,7 @@ static void io_fail_links(struct io_kiocb *req)
static void io_req_find_next(struct io_kiocb *req, struct io_kiocb **nxt)
{
- if (likely(!(req->flags & REQ_F_LINK)))
+ if (likely(!(req->flags & REQ_F_LINK_HEAD)))
return;
/*
@@ -1669,7 +1671,7 @@ static inline unsigned int io_sqring_entries(struct io_ring_ctx *ctx)
static inline bool io_req_multi_free(struct req_batch *rb, struct io_kiocb *req)
{
- if ((req->flags & REQ_F_LINK) || io_is_fallback_req(req))
+ if ((req->flags & REQ_F_LINK_HEAD) || io_is_fallback_req(req))
return false;
if (!(req->flags & REQ_F_FIXED_FILE) || req->io)
@@ -2562,7 +2564,7 @@ static int io_read(struct io_kiocb *req, bool force_nonblock)
req->result = 0;
io_size = ret;
- if (req->flags & REQ_F_LINK)
+ if (req->flags & REQ_F_LINK_HEAD)
req->result = io_size;
/*
@@ -2653,7 +2655,7 @@ static int io_write(struct io_kiocb *req, bool force_nonblock)
req->result = 0;
io_size = ret;
- if (req->flags & REQ_F_LINK)
+ if (req->flags & REQ_F_LINK_HEAD)
req->result = io_size;
/*
@@ -2760,7 +2762,7 @@ static bool io_splice_punt(struct file *file)
return false;
if (!io_file_supports_async(file))
return true;
- return !(file->f_mode & O_NONBLOCK);
+ return !(file->f_flags & O_NONBLOCK);
}
static int io_splice(struct io_kiocb *req, bool force_nonblock)
@@ -4153,20 +4155,57 @@ static int __io_async_wake(struct io_kiocb *req, struct io_poll_iocb *poll,
return 1;
}
+static bool io_poll_rewait(struct io_kiocb *req, struct io_poll_iocb *poll)
+ __acquires(&req->ctx->completion_lock)
+{
+ struct io_ring_ctx *ctx = req->ctx;
+
+ if (!req->result && !READ_ONCE(poll->canceled)) {
+ struct poll_table_struct pt = { ._key = poll->events };
+
+ req->result = vfs_poll(req->file, &pt) & poll->events;
+ }
+
+ spin_lock_irq(&ctx->completion_lock);
+ if (!req->result && !READ_ONCE(poll->canceled)) {
+ add_wait_queue(poll->head, &poll->wait);
+ return true;
+ }
+
+ return false;
+}
+
static void io_async_task_func(struct callback_head *cb)
{
struct io_kiocb *req = container_of(cb, struct io_kiocb, task_work);
struct async_poll *apoll = req->apoll;
struct io_ring_ctx *ctx = req->ctx;
+ bool canceled;
trace_io_uring_task_run(req->ctx, req->opcode, req->user_data);
- WARN_ON_ONCE(!list_empty(&req->apoll->poll.wait.entry));
+ if (io_poll_rewait(req, &apoll->poll)) {
+ spin_unlock_irq(&ctx->completion_lock);
+ return;
+ }
- if (hash_hashed(&req->hash_node)) {
- spin_lock_irq(&ctx->completion_lock);
+ if (hash_hashed(&req->hash_node))
hash_del(&req->hash_node);
- spin_unlock_irq(&ctx->completion_lock);
+
+ canceled = READ_ONCE(apoll->poll.canceled);
+ if (canceled) {
+ io_cqring_fill_event(req, -ECANCELED);
+ io_commit_cqring(ctx);
+ }
+
+ spin_unlock_irq(&ctx->completion_lock);
+
+ if (canceled) {
+ kfree(apoll);
+ io_cqring_ev_posted(ctx);
+ req_set_fail_links(req);
+ io_put_req(req);
+ return;
}
/* restore ->work in case we need to retry again */
@@ -4315,11 +4354,13 @@ static bool __io_poll_remove_one(struct io_kiocb *req,
static bool io_poll_remove_one(struct io_kiocb *req)
{
+ struct async_poll *apoll = NULL;
bool do_complete;
if (req->opcode == IORING_OP_POLL_ADD) {
do_complete = __io_poll_remove_one(req, &req->poll);
} else {
+ apoll = req->apoll;
/* non-poll requests have submit ref still */
do_complete = __io_poll_remove_one(req, &req->apoll->poll);
if (do_complete)
@@ -4328,6 +4369,14 @@ static bool io_poll_remove_one(struct io_kiocb *req)
hash_del(&req->hash_node);
+ if (apoll) {
+ /*
+ * restore ->work because we need to call io_req_work_drop_env.
+ */
+ memcpy(&req->work, &apoll->work, sizeof(req->work));
+ kfree(apoll);
+ }
+
if (do_complete) {
io_cqring_fill_event(req, -ECANCELED);
io_commit_cqring(req->ctx);
@@ -4342,7 +4391,7 @@ static void io_poll_remove_all(struct io_ring_ctx *ctx)
{
struct hlist_node *tmp;
struct io_kiocb *req;
- int i;
+ int posted = 0, i;
spin_lock_irq(&ctx->completion_lock);
for (i = 0; i < (1U << ctx->cancel_hash_bits); i++) {
@@ -4350,11 +4399,12 @@ static void io_poll_remove_all(struct io_ring_ctx *ctx)
list = &ctx->cancel_hash[i];
hlist_for_each_entry_safe(req, tmp, list, hash_node)
- io_poll_remove_one(req);
+ posted += io_poll_remove_one(req);
}
spin_unlock_irq(&ctx->completion_lock);
- io_cqring_ev_posted(ctx);
+ if (posted)
+ io_cqring_ev_posted(ctx);
}
static int io_poll_cancel(struct io_ring_ctx *ctx, __u64 sqe_addr)
@@ -4423,18 +4473,11 @@ static void io_poll_task_handler(struct io_kiocb *req, struct io_kiocb **nxt)
struct io_ring_ctx *ctx = req->ctx;
struct io_poll_iocb *poll = &req->poll;
- if (!req->result && !READ_ONCE(poll->canceled)) {
- struct poll_table_struct pt = { ._key = poll->events };
-
- req->result = vfs_poll(req->file, &pt) & poll->events;
- }
-
- spin_lock_irq(&ctx->completion_lock);
- if (!req->result && !READ_ONCE(poll->canceled)) {
- add_wait_queue(poll->head, &poll->wait);
+ if (io_poll_rewait(req, poll)) {
spin_unlock_irq(&ctx->completion_lock);
return;
}
+
hash_del(&req->hash_node);
io_poll_complete(req, req->result, 0);
req->flags |= REQ_F_COMP_LOCKED;
@@ -4665,11 +4708,12 @@ static int io_timeout_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe,
static int io_timeout(struct io_kiocb *req)
{
- unsigned count;
struct io_ring_ctx *ctx = req->ctx;
struct io_timeout_data *data;
struct list_head *entry;
unsigned span = 0;
+ u32 count = req->timeout.count;
+ u32 seq = req->sequence;
data = &req->io->timeout;
@@ -4678,7 +4722,6 @@ static int io_timeout(struct io_kiocb *req)
* timeout event to be satisfied. If it isn't set, then this is
* a pure timeout request, sequence isn't used.
*/
- count = req->timeout.count;
if (!count) {
req->flags |= REQ_F_TIMEOUT_NOSEQ;
spin_lock_irq(&ctx->completion_lock);
@@ -4686,8 +4729,7 @@ static int io_timeout(struct io_kiocb *req)
goto add;
}
- req->sequence = ctx->cached_sq_head + count - 1;
- data->seq_offset = count;
+ req->sequence = seq + count;
/*
* Insertion sort, ensuring the first entry in the list is always
@@ -4696,26 +4738,26 @@ static int io_timeout(struct io_kiocb *req)
spin_lock_irq(&ctx->completion_lock);
list_for_each_prev(entry, &ctx->timeout_list) {
struct io_kiocb *nxt = list_entry(entry, struct io_kiocb, list);
- unsigned nxt_sq_head;
+ unsigned nxt_seq;
long long tmp, tmp_nxt;
- u32 nxt_offset = nxt->io->timeout.seq_offset;
+ u32 nxt_offset = nxt->timeout.count;
if (nxt->flags & REQ_F_TIMEOUT_NOSEQ)
continue;
/*
- * Since cached_sq_head + count - 1 can overflow, use type long
+ * Since seq + count can overflow, use type long
* long to store it.
*/
- tmp = (long long)ctx->cached_sq_head + count - 1;
- nxt_sq_head = nxt->sequence - nxt_offset + 1;
- tmp_nxt = (long long)nxt_sq_head + nxt_offset - 1;
+ tmp = (long long)seq + count;
+ nxt_seq = nxt->sequence - nxt_offset;
+ tmp_nxt = (long long)nxt_seq + nxt_offset;
/*
* cached_sq_head may overflow, and it will never overflow twice
* once there is some timeout req still be valid.
*/
- if (ctx->cached_sq_head < nxt_sq_head)
+ if (seq < nxt_seq)
tmp += UINT_MAX;
if (tmp > tmp_nxt)
@@ -5476,7 +5518,7 @@ static struct io_kiocb *io_prep_linked_timeout(struct io_kiocb *req)
{
struct io_kiocb *nxt;
- if (!(req->flags & REQ_F_LINK))
+ if (!(req->flags & REQ_F_LINK_HEAD))
return NULL;
/* for polled retry, if flag is set, we already went through here */
if (req->flags & REQ_F_POLLED)
@@ -5604,54 +5646,11 @@ static inline void io_queue_link_head(struct io_kiocb *req)
io_queue_sqe(req, NULL);
}
-#define SQE_VALID_FLAGS (IOSQE_FIXED_FILE|IOSQE_IO_DRAIN|IOSQE_IO_LINK| \
- IOSQE_IO_HARDLINK | IOSQE_ASYNC | \
- IOSQE_BUFFER_SELECT)
-
-static bool io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe,
+static int io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe,
struct io_submit_state *state, struct io_kiocb **link)
{
struct io_ring_ctx *ctx = req->ctx;
- unsigned int sqe_flags;
- int ret, id, fd;
-
- sqe_flags = READ_ONCE(sqe->flags);
-
- /* enforce forwards compatibility on users */
- if (unlikely(sqe_flags & ~SQE_VALID_FLAGS)) {
- ret = -EINVAL;
- goto err_req;
- }
-
- if ((sqe_flags & IOSQE_BUFFER_SELECT) &&
- !io_op_defs[req->opcode].buffer_select) {
- ret = -EOPNOTSUPP;
- goto err_req;
- }
-
- id = READ_ONCE(sqe->personality);
- if (id) {
- req->work.creds = idr_find(&ctx->personality_idr, id);
- if (unlikely(!req->work.creds)) {
- ret = -EINVAL;
- goto err_req;
- }
- get_cred(req->work.creds);
- }
-
- /* same numerical values with corresponding REQ_F_*, safe to copy */
- req->flags |= sqe_flags & (IOSQE_IO_DRAIN | IOSQE_IO_HARDLINK |
- IOSQE_ASYNC | IOSQE_FIXED_FILE |
- IOSQE_BUFFER_SELECT);
-
- fd = READ_ONCE(sqe->fd);
- ret = io_req_set_file(state, req, fd, sqe_flags);
- if (unlikely(ret)) {
-err_req:
- io_cqring_add_event(req, ret);
- io_double_put_req(req);
- return false;
- }
+ int ret;
/*
* If we already have a head request, queue this one for async
@@ -5670,42 +5669,39 @@ err_req:
* next after the link request. The last one is done via
* drain_next flag to persist the effect across calls.
*/
- if (sqe_flags & IOSQE_IO_DRAIN) {
+ if (req->flags & REQ_F_IO_DRAIN) {
head->flags |= REQ_F_IO_DRAIN;
ctx->drain_next = 1;
}
- if (io_alloc_async_ctx(req)) {
- ret = -EAGAIN;
- goto err_req;
- }
+ if (io_alloc_async_ctx(req))
+ return -EAGAIN;
ret = io_req_defer_prep(req, sqe);
if (ret) {
/* fail even hard links since we don't submit */
head->flags |= REQ_F_FAIL_LINK;
- goto err_req;
+ return ret;
}
trace_io_uring_link(ctx, req, head);
list_add_tail(&req->link_list, &head->link_list);
/* last request of a link, enqueue the link */
- if (!(sqe_flags & (IOSQE_IO_LINK|IOSQE_IO_HARDLINK))) {
+ if (!(req->flags & (REQ_F_LINK | REQ_F_HARDLINK))) {
io_queue_link_head(head);
*link = NULL;
}
} else {
if (unlikely(ctx->drain_next)) {
req->flags |= REQ_F_IO_DRAIN;
- req->ctx->drain_next = 0;
+ ctx->drain_next = 0;
}
- if (sqe_flags & (IOSQE_IO_LINK|IOSQE_IO_HARDLINK)) {
- req->flags |= REQ_F_LINK;
+ if (req->flags & (REQ_F_LINK | REQ_F_HARDLINK)) {
+ req->flags |= REQ_F_LINK_HEAD;
INIT_LIST_HEAD(&req->link_list);
- if (io_alloc_async_ctx(req)) {
- ret = -EAGAIN;
- goto err_req;
- }
+ if (io_alloc_async_ctx(req))
+ return -EAGAIN;
+
ret = io_req_defer_prep(req, sqe);
if (ret)
req->flags |= REQ_F_FAIL_LINK;
@@ -5715,7 +5711,7 @@ err_req:
}
}
- return true;
+ return 0;
}
/*
@@ -5789,15 +5785,23 @@ static inline void io_consume_sqe(struct io_ring_ctx *ctx)
ctx->cached_sq_head++;
}
-static void io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req,
- const struct io_uring_sqe *sqe)
+#define SQE_VALID_FLAGS (IOSQE_FIXED_FILE|IOSQE_IO_DRAIN|IOSQE_IO_LINK| \
+ IOSQE_IO_HARDLINK | IOSQE_ASYNC | \
+ IOSQE_BUFFER_SELECT)
+
+static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req,
+ const struct io_uring_sqe *sqe,
+ struct io_submit_state *state, bool async)
{
+ unsigned int sqe_flags;
+ int id, fd;
+
/*
* All io need record the previous position, if LINK vs DARIN,
* it can be used to mark the position of the first IO in the
* link list.
*/
- req->sequence = ctx->cached_sq_head;
+ req->sequence = ctx->cached_sq_head - ctx->cached_sq_dropped;
req->opcode = READ_ONCE(sqe->opcode);
req->user_data = READ_ONCE(sqe->user_data);
req->io = NULL;
@@ -5808,17 +5812,50 @@ static void io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req,
refcount_set(&req->refs, 2);
req->task = NULL;
req->result = 0;
+ req->needs_fixed_file = async;
INIT_IO_WORK(&req->work, io_wq_submit_work);
+
+ if (unlikely(req->opcode >= IORING_OP_LAST))
+ return -EINVAL;
+
+ if (io_op_defs[req->opcode].needs_mm && !current->mm) {
+ if (unlikely(!mmget_not_zero(ctx->sqo_mm)))
+ return -EFAULT;
+ use_mm(ctx->sqo_mm);
+ }
+
+ sqe_flags = READ_ONCE(sqe->flags);
+ /* enforce forwards compatibility on users */
+ if (unlikely(sqe_flags & ~SQE_VALID_FLAGS))
+ return -EINVAL;
+
+ if ((sqe_flags & IOSQE_BUFFER_SELECT) &&
+ !io_op_defs[req->opcode].buffer_select)
+ return -EOPNOTSUPP;
+
+ id = READ_ONCE(sqe->personality);
+ if (id) {
+ req->work.creds = idr_find(&ctx->personality_idr, id);
+ if (unlikely(!req->work.creds))
+ return -EINVAL;
+ get_cred(req->work.creds);
+ }
+
+ /* same numerical values with corresponding REQ_F_*, safe to copy */
+ req->flags |= sqe_flags & (IOSQE_IO_DRAIN | IOSQE_IO_HARDLINK |
+ IOSQE_ASYNC | IOSQE_FIXED_FILE |
+ IOSQE_BUFFER_SELECT | IOSQE_IO_LINK);
+
+ fd = READ_ONCE(sqe->fd);
+ return io_req_set_file(state, req, fd, sqe_flags);
}
static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
- struct file *ring_file, int ring_fd,
- struct mm_struct **mm, bool async)
+ struct file *ring_file, int ring_fd, bool async)
{
struct io_submit_state state, *statep = NULL;
struct io_kiocb *link = NULL;
int i, submitted = 0;
- bool mm_fault = false;
/* if we have a backlog and couldn't flush it all, return BUSY */
if (test_bit(0, &ctx->sq_check_overflow)) {
@@ -5858,34 +5895,23 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr,
break;
}
- io_init_req(ctx, req, sqe);
+ err = io_init_req(ctx, req, sqe, statep, async);
io_consume_sqe(ctx);
/* will complete beyond this point, count as submitted */
submitted++;
- if (unlikely(req->opcode >= IORING_OP_LAST)) {
- err = -EINVAL;
+ if (unlikely(err)) {
fail_req:
io_cqring_add_event(req, err);
io_double_put_req(req);
break;
}
- if (io_op_defs[req->opcode].needs_mm && !*mm) {
- mm_fault = mm_fault || !mmget_not_zero(ctx->sqo_mm);
- if (unlikely(mm_fault)) {
- err = -EFAULT;
- goto fail_req;
- }
- use_mm(ctx->sqo_mm);
- *mm = ctx->sqo_mm;
- }
-
- req->needs_fixed_file = async;
trace_io_uring_submit_sqe(ctx, req->opcode, req->user_data,
true, async);
- if (!io_submit_sqe(req, sqe, statep, &link))
- break;
+ err = io_submit_sqe(req, sqe, statep, &link);
+ if (err)
+ goto fail_req;
}
if (unlikely(submitted != nr)) {
@@ -5904,10 +5930,19 @@ fail_req:
return submitted;
}
+static inline void io_sq_thread_drop_mm(struct io_ring_ctx *ctx)
+{
+ struct mm_struct *mm = current->mm;
+
+ if (mm) {
+ unuse_mm(mm);
+ mmput(mm);
+ }
+}
+
static int io_sq_thread(void *data)
{
struct io_ring_ctx *ctx = data;
- struct mm_struct *cur_mm = NULL;
const struct cred *old_cred;
mm_segment_t old_fs;
DEFINE_WAIT(wait);
@@ -5948,11 +5983,7 @@ static int io_sq_thread(void *data)
* adding ourselves to the waitqueue, as the unuse/drop
* may sleep.
*/
- if (cur_mm) {
- unuse_mm(cur_mm);
- mmput(cur_mm);
- cur_mm = NULL;
- }
+ io_sq_thread_drop_mm(ctx);
/*
* We're polling. If we're within the defined idle
@@ -6016,7 +6047,7 @@ static int io_sq_thread(void *data)
}
mutex_lock(&ctx->uring_lock);
- ret = io_submit_sqes(ctx, to_submit, NULL, -1, &cur_mm, true);
+ ret = io_submit_sqes(ctx, to_submit, NULL, -1, true);
mutex_unlock(&ctx->uring_lock);
timeout = jiffies + ctx->sq_thread_idle;
}
@@ -6025,10 +6056,7 @@ static int io_sq_thread(void *data)
task_work_run();
set_fs(old_fs);
- if (cur_mm) {
- unuse_mm(cur_mm);
- mmput(cur_mm);
- }
+ io_sq_thread_drop_mm(ctx);
revert_creds(old_cred);
kthread_parkme();
@@ -7509,13 +7537,8 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit,
wake_up(&ctx->sqo_wait);
submitted = to_submit;
} else if (to_submit) {
- struct mm_struct *cur_mm;
-
mutex_lock(&ctx->uring_lock);
- /* already have mm, so io_submit_sqes() won't try to grab it */
- cur_mm = ctx->sqo_mm;
- submitted = io_submit_sqes(ctx, to_submit, f.file, fd,
- &cur_mm, false);
+ submitted = io_submit_sqes(ctx, to_submit, f.file, fd, false);
mutex_unlock(&ctx->uring_lock);
if (submitted != to_submit)
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index f2dc35c22964..b8d78f393365 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -2023,6 +2023,7 @@ lookup_again:
goto lookup_again;
}
+ spin_unlock(&ino->i_lock);
first = true;
status = nfs4_select_rw_stateid(ctx->state,
iomode == IOMODE_RW ? FMODE_WRITE : FMODE_READ,
@@ -2032,12 +2033,12 @@ lookup_again:
trace_pnfs_update_layout(ino, pos, count,
iomode, lo, lseg,
PNFS_UPDATE_LAYOUT_INVALID_OPEN);
- spin_unlock(&ino->i_lock);
nfs4_schedule_stateid_recovery(server, ctx->state);
pnfs_clear_first_layoutget(lo);
pnfs_put_layout_hdr(lo);
goto lookup_again;
}
+ spin_lock(&ino->i_lock);
} else {
nfs4_stateid_copy(&stateid, &lo->plh_stateid);
}
diff --git a/fs/proc/root.c b/fs/proc/root.c
index 2633f10446c3..cdbe9293ea55 100644
--- a/fs/proc/root.c
+++ b/fs/proc/root.c
@@ -196,6 +196,13 @@ static void proc_kill_sb(struct super_block *sb)
if (ns->proc_thread_self)
dput(ns->proc_thread_self);
kill_anon_super(sb);
+
+ /* Make the pid namespace safe for the next mount of proc */
+ ns->proc_self = NULL;
+ ns->proc_thread_self = NULL;
+ ns->pid_gid = GLOBAL_ROOT_GID;
+ ns->hide_pid = 0;
+
put_pid_ns(ns);
}
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.