On Wed, May 29, 2019 at 9:33 PM Darrick J. Wong <darrick.wong@xxxxxxxxxx> wrote: > > On Wed, May 29, 2019 at 08:43:12PM +0300, Amir Goldstein wrote: > > Because generic_copy_file_range doesn't hold the destination inode lock > > throughout the copy, strip setuid bits before and after copy. > > > > The destination inode mtime is updated before and after the copy and the > > source inode atime is updated after the copy, similar to > > generic_file_read_iter(). > > > > Signed-off-by: Amir Goldstein <amir73il@xxxxxxxxx> > > Looks reasonable, Actually, it isn't reasonable. I'd like to recall this patch :-/ As one might expect, splice_direct_to_actor() already has file_accessed() and "file_modified" is the responsibility of filesystem's ->write_iter(). Thanks, Amir.
