| commit | face15025ffdf664de95e86ae831544154d26c9c | [log] [tgz] |
|---|---|---|
| author | J. Bruce Fields <bfields@redhat.com> | Fri Oct 26 16:12:31 2012 -0400 |
| committer | J. Bruce Fields <bfields@redhat.com> | Wed Nov 07 19:31:34 2012 -0500 |
| tree | 7b848b1a55182da4368e182ab5bc9e3f52043263 | |
| parent | fae5096ad217db2e3368e980c1d86223f786856b [diff] |
nfsd: use vfs_fsync_range(), not O_SYNC, for stable writes NFSv4 shares the same struct file across multiple writes. (And we'd like NFSv2 and NFSv3 to do that as well some day.) So setting O_SYNC on the struct file as a way to request a synchronous write doesn't work. Instead, do a vfs_fsync_range() in that case. Reported-by: Peter Staubach <pstaubach@exagrid.com> Signed-off-by: J. Bruce Fields <bfields@redhat.com>