| commit | a41ad394a03b802497958d7c98a9dcf607266645 | |
| author | Josef Bacik <josef@redhat.com> | |
| Mon, 31 Jan 2011 20:30:16 +0000 (31 15:30 -0500) | ||
| committer | Josef Bacik <josef@redhat.com> | |
| Thu, 17 Mar 2011 18:21:22 +0000 (17 14:21 -0400) | ||
| tree | f18d97b7feb5c3dcf0d3c1a670d4eaf6b94e8b79 | treesnapshot (tar.gz zip) |
| parent | dc89e9824464e91fa0b06267864ceabe3186fd8b | commitdiff |
Btrfs: convert to the new truncate sequence
->truncate() is going away, instead all of the work needs to be done in
->setattr(). So this converts us over to do this. It's fairly straightforward,
just get rid of our .truncate inode operation and call btrfs_truncate() directly
from btrfs_setsize. This works out better for us since truncate can technically
return ENOSPC, and before we had no way of letting anybody know. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
->truncate() is going away, instead all of the work needs to be done in
->setattr(). So this converts us over to do this. It's fairly straightforward,
just get rid of our .truncate inode operation and call btrfs_truncate() directly
from btrfs_setsize. This works out better for us since truncate can technically
return ENOSPC, and before we had no way of letting anybody know. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
| fs/btrfs/ctree.h | diffblobblamehistory | |
| fs/btrfs/file.c | diffblobblamehistory | |
| fs/btrfs/inode.c | diffblobblamehistory |