On Thu, Oct 11, 2018 at 11:05:36AM -0700, Viacheslav Dubeyko wrote:
> On Wed, 2018-10-10 at 15:23 -0300, Ernesto A. Fernández wrote:
> > Direct writes to empty inodes fail with EIO. The generic direct-io code
> > is in part to blame (a patch has been submitted as "direct-io: allow
> > direct writes to empty inodes"), but hfsplus is worse affected than the
> > other filesystems because the fallback to buffered I/O doesn't happen.
> >
>
> Could you please share more detailed explanation of the patch that
> affects the HFS+ behavior? It's hard to follow what patch you mean.
It was sent to linux-fsdevel: "direct-io: allow direct writes to empty
inodes."
>
> Thanks,
> Vyacheslav Dubeyko.
>
> > The problem is the return value of hfsplus_get_block() when called with
> > !create. Change it to be more consistent with the other modules.
> >
> > Signed-off-by: Ernesto A. Fernández <ernesto.mnd.fernandez@xxxxxxxxx>
> > ---
> > fs/hfsplus/extents.c | 4 +++-
> > 1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/fs/hfsplus/extents.c b/fs/hfsplus/extents.c
> > index 8a8893d522ef..a930ddd15681 100644
> > --- a/fs/hfsplus/extents.c
> > +++ b/fs/hfsplus/extents.c
> > @@ -237,7 +237,9 @@ int hfsplus_get_block(struct inode *inode, sector_t iblock,
> > ablock = iblock >> sbi->fs_shift;
> >
> > if (iblock >= hip->fs_blocks) {
> > - if (iblock > hip->fs_blocks || !create)
> > + if (!create)
> > + return 0;
> > + if (iblock > hip->fs_blocks)
> > return -EIO;
> > if (ablock >= hip->alloc_blocks) {
> > res = hfsplus_file_extend(inode, false);
>
>
