7bdb0d18bf
Currently, the default behavior of O_DIRECT writes was allowing concurrent writing among nodes to the same file, with no cluster coherency guaranteed (no EX lock held). This can leave stale data in the cache for buffered reads on other nodes. The new mount option introduce a chance to choose two different behaviors for O_DIRECT writes: * coherency=full, as the default value, will disallow concurrent O_DIRECT writes by taking EX locks. * coherency=buffered, allow concurrent O_DIRECT writes without EX lock among nodes, which gains high performance at risk of getting stale data on other nodes. Signed-off-by: Tristan Ye <tristan.ye@oracle.com> Signed-off-by: Joel Becker <joel.becker@oracle.com> |
||
---|---|---|
.. | ||
caching | ||
configfs | ||
nfs | ||
pohmelfs | ||
9p.txt | ||
00-INDEX | ||
adfs.txt | ||
affs.txt | ||
afs.txt | ||
autofs4-mount-control.txt | ||
automount-support.txt | ||
befs.txt | ||
bfs.txt | ||
btrfs.txt | ||
ceph.txt | ||
cifs.txt | ||
coda.txt | ||
cramfs.txt | ||
debugfs.txt | ||
dentry-locking.txt | ||
devpts.txt | ||
directory-locking | ||
dlmfs.txt | ||
dnotify_test.c | ||
dnotify.txt | ||
ecryptfs.txt | ||
exofs.txt | ||
ext2.txt | ||
ext3.txt | ||
ext4.txt | ||
fiemap.txt | ||
files.txt | ||
fuse.txt | ||
gfs2-glocks.txt | ||
gfs2-uevents.txt | ||
gfs2.txt | ||
hfs.txt | ||
hfsplus.txt | ||
hpfs.txt | ||
inotify.txt | ||
isofs.txt | ||
jfs.txt | ||
Locking | ||
locks.txt | ||
logfs.txt | ||
Makefile | ||
mandatory-locking.txt | ||
ncpfs.txt | ||
nilfs2.txt | ||
ntfs.txt | ||
ocfs2.txt | ||
omfs.txt | ||
porting | ||
proc.txt | ||
quota.txt | ||
ramfs-rootfs-initramfs.txt | ||
relay.txt | ||
romfs.txt | ||
seq_file.txt | ||
sharedsubtree.txt | ||
smbfs.txt | ||
spufs.txt | ||
squashfs.txt | ||
sysfs-pci.txt | ||
sysfs-tagging.txt | ||
sysfs.txt | ||
sysv-fs.txt | ||
tmpfs.txt | ||
ubifs.txt | ||
udf.txt | ||
ufs.txt | ||
vfat.txt | ||
vfs.txt | ||
xfs-delayed-logging-design.txt | ||
xfs.txt | ||
xip.txt |