aboutsummaryrefslogtreecommitdiffstats
path: root/Documentation/filesystems
diff options
context:
space:
mode:
authorTheodore Ts'o <tytso@mit.edu>2009-01-06 14:53:06 -0500
committerTheodore Ts'o <tytso@mit.edu>2009-01-06 14:53:06 -0500
commit8e1a4857cd92e32e642b3e7184c7f6bf85c96e2e (patch)
tree2d82db0e86e080bd1673572b77ae6da2cc870067 /Documentation/filesystems
parent59e315b4c410b00a9acd0f24a00dbadbe81ce692 (diff)
Update Documentation/filesystems/ext4.txt
Fix paragraph with recommendations on how to tune ext4 for benchmarks. Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Diffstat (limited to 'Documentation/filesystems')
-rw-r--r--Documentation/filesystems/ext4.txt42
1 files changed, 32 insertions, 10 deletions
diff --git a/Documentation/filesystems/ext4.txt b/Documentation/filesystems/ext4.txt
index 174eaff7ded9..f75ab101c00a 100644
--- a/Documentation/filesystems/ext4.txt
+++ b/Documentation/filesystems/ext4.txt
@@ -58,13 +58,22 @@ Note: More extensive information for getting started with ext4 can be
58 58
59 # mount -t ext4 /dev/hda1 /wherever 59 # mount -t ext4 /dev/hda1 /wherever
60 60
61 - When comparing performance with other filesystems, remember that 61 - When comparing performance with other filesystems, it's always
62 ext3/4 by default offers higher data integrity guarantees than most. 62 important to try multiple workloads; very often a subtle change in a
63 So when comparing with a metadata-only journalling filesystem, such 63 workload parameter can completely change the ranking of which
64 as ext3, use `mount -o data=writeback'. And you might as well use 64 filesystems do well compared to others. When comparing versus ext3,
65 `mount -o nobh' too along with it. Making the journal larger than 65 note that ext4 enables write barriers by default, while ext3 does
66 the mke2fs default often helps performance with metadata-intensive 66 not enable write barriers by default. So it is useful to use
67 workloads. 67 explicitly specify whether barriers are enabled or not when via the
68 '-o barriers=[0|1]' mount option for both ext3 and ext4 filesystems
69 for a fair comparison. When tuning ext3 for best benchmark numbers,
70 it is often worthwhile to try changing the data journaling mode; '-o
71 data=writeback,nobh' can be faster for some workloads. (Note
72 however that running mounted with data=writeback can potentially
73 leave stale data exposed in recently written files in case of an
74 unclean shutdown, which could be a security exposure in some
75 situations.) Configuring the filesystem with a large journal can
76 also be helpful for metadata-intensive workloads.
68 77
692. Features 782. Features
70=========== 79===========
@@ -74,7 +83,7 @@ Note: More extensive information for getting started with ext4 can be
74* ability to use filesystems > 16TB (e2fsprogs support not available yet) 83* ability to use filesystems > 16TB (e2fsprogs support not available yet)
75* extent format reduces metadata overhead (RAM, IO for access, transactions) 84* extent format reduces metadata overhead (RAM, IO for access, transactions)
76* extent format more robust in face of on-disk corruption due to magics, 85* extent format more robust in face of on-disk corruption due to magics,
77* internal redunancy in tree 86* internal redundancy in tree
78* improved file allocation (multi-block alloc) 87* improved file allocation (multi-block alloc)
79* fix 32000 subdirectory limit 88* fix 32000 subdirectory limit
80* nsec timestamps for mtime, atime, ctime, create time 89* nsec timestamps for mtime, atime, ctime, create time
@@ -116,6 +125,12 @@ grouping of bitmaps and inode tables. Some test results available here:
116When mounting an ext4 filesystem, the following option are accepted: 125When mounting an ext4 filesystem, the following option are accepted:
117(*) == default 126(*) == default
118 127
128ro Mount filesystem read only. Note that ext4 will
129 replay the journal (and thus write to the
130 partition) even when mounted "read only". The
131 mount options "ro,noload" can be used to prevent
132 writes to the filesystem.
133
119extents (*) ext4 will use extents to address file data. The 134extents (*) ext4 will use extents to address file data. The
120 file system will no longer be mountable by ext3. 135 file system will no longer be mountable by ext3.
121 136
@@ -144,7 +159,11 @@ journal_dev=devnum When the external journal device's major/minor numbers
144 identified through its new major/minor numbers encoded 159 identified through its new major/minor numbers encoded
145 in devnum. 160 in devnum.
146 161
147noload Don't load the journal on mounting. 162noload Don't load the journal on mounting. Note that
163 if the filesystem was not unmounted cleanly,
164 skipping the journal replay will lead to the
165 filesystem containing inconsistencies that can
166 lead to any number of problems.
148 167
149data=journal All data are committed into the journal prior to being 168data=journal All data are committed into the journal prior to being
150 written into the main file system. 169 written into the main file system.
@@ -219,9 +238,12 @@ minixdf Make 'df' act like Minix.
219 238
220debug Extra debugging information is sent to syslog. 239debug Extra debugging information is sent to syslog.
221 240
222errors=remount-ro(*) Remount the filesystem read-only on an error. 241errors=remount-ro Remount the filesystem read-only on an error.
223errors=continue Keep going on a filesystem error. 242errors=continue Keep going on a filesystem error.
224errors=panic Panic and halt the machine if an error occurs. 243errors=panic Panic and halt the machine if an error occurs.
244 (These mount options override the errors behavior
245 specified in the superblock, which can be configured
246 using tune2fs)
225 247
226data_err=ignore(*) Just print an error message if an error occurs 248data_err=ignore(*) Just print an error message if an error occurs
227 in a file data buffer in ordered mode. 249 in a file data buffer in ordered mode.