On Sat, Apr 06, 2013 at 12:04:52PM +0200, Marco Stornelli wrote:
> In every place where sb_start_write was called now we must manage
> the error code and return -EINTR.
Now go and look for callers of mnt_want_write() ;-/ The really
painful one is in do_last(), but kern_path_create() is not much
b
Il 06/04/2013 15:17, Matthew Wilcox ha scritto:
On Sat, Apr 06, 2013 at 12:04:52PM +0200, Marco Stornelli wrote:
In every place where sb_start_write was called now we must manage
the error code and return -EINTR.
If we must manage the error code, then these functions should be marked
__must_ch
On Sat, Apr 06, 2013 at 12:04:52PM +0200, Marco Stornelli wrote:
> In every place where sb_start_write was called now we must manage
> the error code and return -EINTR.
If we must manage the error code, then these functions should be marked
__must_check.
--
Matthew Wilcox
In every place where sb_start_write was called now we must manage
the error code and return -EINTR.
Signed-off-by: Marco Stornelli
---
fs/btrfs/file.c|9 +++--
fs/cifs/file.c |4 +++-
fs/ext4/mmp.c |3 ++-
fs/ext4/super.c|4 +++-
fs/fuse/file.c |4 +++