| /openbmc/linux/fs/btrfs/ |
| H A D | extent_map.h | diff 890871be854b5f5e43e7ba2475f706209906cc24 Wed Sep 02 15:24:52 CDT 2009 Chris Mason <chris.mason@oracle.com> Btrfs: switch extent_map to a rw lock
There are two main users of the extent_map tree. The
first is regular file inodes, where it is evenly spread
between readers and writers.
The second is the chunk allocation tree, which maps blocks from
logical addresses to phyiscal ones, and it is 99.99% reads.
The mapping tree is a point of lock contention during heavy IO
workloads, so this commit switches things to a rw lock.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
|
| H A D | extent_map.c | diff 890871be854b5f5e43e7ba2475f706209906cc24 Wed Sep 02 15:24:52 CDT 2009 Chris Mason <chris.mason@oracle.com> Btrfs: switch extent_map to a rw lock
There are two main users of the extent_map tree. The
first is regular file inodes, where it is evenly spread
between readers and writers.
The second is the chunk allocation tree, which maps blocks from
logical addresses to phyiscal ones, and it is 99.99% reads.
The mapping tree is a point of lock contention during heavy IO
workloads, so this commit switches things to a rw lock.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
|
| H A D | compression.c | diff 890871be854b5f5e43e7ba2475f706209906cc24 Wed Sep 02 15:24:52 CDT 2009 Chris Mason <chris.mason@oracle.com> Btrfs: switch extent_map to a rw lock
There are two main users of the extent_map tree. The
first is regular file inodes, where it is evenly spread
between readers and writers.
The second is the chunk allocation tree, which maps blocks from
logical addresses to phyiscal ones, and it is 99.99% reads.
The mapping tree is a point of lock contention during heavy IO
workloads, so this commit switches things to a rw lock.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
|
| H A D | relocation.c | diff 890871be854b5f5e43e7ba2475f706209906cc24 Wed Sep 02 15:24:52 CDT 2009 Chris Mason <chris.mason@oracle.com> Btrfs: switch extent_map to a rw lock
There are two main users of the extent_map tree. The
first is regular file inodes, where it is evenly spread
between readers and writers.
The second is the chunk allocation tree, which maps blocks from
logical addresses to phyiscal ones, and it is 99.99% reads.
The mapping tree is a point of lock contention during heavy IO
workloads, so this commit switches things to a rw lock.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
|
| H A D | file.c | diff 890871be854b5f5e43e7ba2475f706209906cc24 Wed Sep 02 15:24:52 CDT 2009 Chris Mason <chris.mason@oracle.com> Btrfs: switch extent_map to a rw lock
There are two main users of the extent_map tree. The
first is regular file inodes, where it is evenly spread
between readers and writers.
The second is the chunk allocation tree, which maps blocks from
logical addresses to phyiscal ones, and it is 99.99% reads.
The mapping tree is a point of lock contention during heavy IO
workloads, so this commit switches things to a rw lock.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
|
| H A D | extent_io.c | diff 890871be854b5f5e43e7ba2475f706209906cc24 Wed Sep 02 15:24:52 CDT 2009 Chris Mason <chris.mason@oracle.com> Btrfs: switch extent_map to a rw lock
There are two main users of the extent_map tree. The
first is regular file inodes, where it is evenly spread
between readers and writers.
The second is the chunk allocation tree, which maps blocks from
logical addresses to phyiscal ones, and it is 99.99% reads.
The mapping tree is a point of lock contention during heavy IO
workloads, so this commit switches things to a rw lock.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
|
| H A D | volumes.c | diff 890871be854b5f5e43e7ba2475f706209906cc24 Wed Sep 02 15:24:52 CDT 2009 Chris Mason <chris.mason@oracle.com> Btrfs: switch extent_map to a rw lock
There are two main users of the extent_map tree. The
first is regular file inodes, where it is evenly spread
between readers and writers.
The second is the chunk allocation tree, which maps blocks from
logical addresses to phyiscal ones, and it is 99.99% reads.
The mapping tree is a point of lock contention during heavy IO
workloads, so this commit switches things to a rw lock.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
|
| H A D | extent-tree.c | diff 890871be854b5f5e43e7ba2475f706209906cc24 Wed Sep 02 15:24:52 CDT 2009 Chris Mason <chris.mason@oracle.com> Btrfs: switch extent_map to a rw lock
There are two main users of the extent_map tree. The
first is regular file inodes, where it is evenly spread
between readers and writers.
The second is the chunk allocation tree, which maps blocks from
logical addresses to phyiscal ones, and it is 99.99% reads.
The mapping tree is a point of lock contention during heavy IO
workloads, so this commit switches things to a rw lock.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
|
| H A D | disk-io.c | diff 890871be854b5f5e43e7ba2475f706209906cc24 Wed Sep 02 15:24:52 CDT 2009 Chris Mason <chris.mason@oracle.com> Btrfs: switch extent_map to a rw lock
There are two main users of the extent_map tree. The
first is regular file inodes, where it is evenly spread
between readers and writers.
The second is the chunk allocation tree, which maps blocks from
logical addresses to phyiscal ones, and it is 99.99% reads.
The mapping tree is a point of lock contention during heavy IO
workloads, so this commit switches things to a rw lock.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
|
| H A D | inode.c | diff 890871be854b5f5e43e7ba2475f706209906cc24 Wed Sep 02 15:24:52 CDT 2009 Chris Mason <chris.mason@oracle.com> Btrfs: switch extent_map to a rw lock
There are two main users of the extent_map tree. The
first is regular file inodes, where it is evenly spread
between readers and writers.
The second is the chunk allocation tree, which maps blocks from
logical addresses to phyiscal ones, and it is 99.99% reads.
The mapping tree is a point of lock contention during heavy IO
workloads, so this commit switches things to a rw lock.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
|