On Thu, Jan 31, 2019 at 08:03:36AM +0800, Qu Wenruo wrote: > On 2019/1/30 下午10:59, Nikolay Borisov wrote: > > On 30.01.19 г. 16:57 ч., David Sterba wrote: > >> On Fri, Jan 25, 2019 at 01:09:16PM +0800, Qu Wenruo wrote: > >>> Just add one extra line to show when the corruption is detected. > >>> Currently only read time detection is possible. > >>> > >>> Signed-off-by: Qu Wenruo <w...@suse.com> > >>> Reviewed-by: Nikolay Borisov <nbori...@suse.com> > >>> --- > >>> fs/btrfs/disk-io.c | 2 ++ > >>> 1 file changed, 2 insertions(+) > >>> > >>> diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c > >>> index 794d5bb7fe33..426e9f450f70 100644 > >>> --- a/fs/btrfs/disk-io.c > >>> +++ b/fs/btrfs/disk-io.c > >>> @@ -658,6 +658,8 @@ static int btree_readpage_end_io_hook(struct > >>> btrfs_io_bio *io_bio, > >>> > >>> if (!ret) > >>> set_extent_buffer_uptodate(eb); > >>> + else > >>> + btrfs_err(fs_info, "read time tree block corrupted detected"); > >> > >> I'm not sure the 'read time' is clear enoug, my suggestion is to use > >> 'post-read' (and pre-write analogicaly). What do you think? > > > > > > How about "error during tree block reading" or "error reading treeblock"? > > Nikolay's suggestion looks more straightforward to me. > > +1 for his idea. > > The 'post-read' still could confuse end-user IMHO.
The idea is to distinguish if the error was because the block can't be read or because the data it contains are wrong.
