lk/linux-kernel-test
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Btrfs: Do metadata checksums for reads via a workqueue

Browse Source 
Before, metadata checksumming was done by the callers of read_tree_block,
which would set EXTENT_CSUM bits in the extent tree to show that a given
range of pages was already checksummed and didn't need to be verified
again.

But, those bits could go away via try_to_releasepage, and the end
result was bogus checksum failures on pages that never left the cache.

The new code validates checksums when the page is read.  It is a little
tricky because metadata blocks can span pages and a single read may
end up going via multiple bios.

Signed-off-by: Chris Mason <chris.mason@oracle.com>
This commit is contained in:
Chris Mason
2008-04-09 16:28:12 -04:00
parent 728131d8e4
commit ce9adaa5a7
7 changed files with 310 additions and 67 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
fs/btrfs/ctree.c
View File

@@ -158,9 +158,6 @@ int __btrfs_cow_block(struct btrfs_trans_handle *trans,
} else {
root_gen = 0;
}
if (!(buf->flags & EXTENT_CSUM))
WARN_ON(1);
WARN_ON(root->ref_cows && trans->transid !=
root->fs_info->running_transaction->transid);
WARN_ON(root->ref_cows && trans->transid != root->last_trans);
@@ -247,8 +244,6 @@ int btrfs_cow_block(struct btrfs_trans_handle *trans,
root->fs_info->generation);
WARN_ON(1);
}
if (!(buf->flags & EXTENT_CSUM))
WARN_ON(1);
header_trans = btrfs_header_generation(buf);
spin_lock(&root->fs_info->hash_lock);