Re: [PATCH 1/4] xfs: drop submit side trans alloc for append ioends

[Brian Foster <bfoster@redhat.com>]

On Wed, Apr 07, 2021 at 07:33:21AM +0100, Christoph Hellwig wrote:
> > @@ -182,12 +155,10 @@ xfs_end_ioend(
> >  		error = xfs_reflink_end_cow(ip, offset, size);
> >  	else if (ioend->io_type == IOMAP_UNWRITTEN)
> >  		error = xfs_iomap_write_unwritten(ip, offset, size, false);
> > -	else
> > -		ASSERT(!xfs_ioend_is_append(ioend) || ioend->io_private);
> 
> I first though we'd now call xfs_setfilesize for unwritten extents
> as well, but as those have already updated di_size we are fine here.
> 
> As a future enhancement it would be useful to let xfs_reflink_end_cow
> update the file size similar to what we do for the unwritten case.
> 

Agreed. I noticed that when first passing through the code but didn't
want to get into further functional changes.

> >  done:
> > -	if (ioend->io_private)
> > -		error = xfs_setfilesize_ioend(ioend, error);
> > +	if (!error && xfs_ioend_is_append(ioend))
> > +		error = xfs_setfilesize(ip, ioend->io_offset, ioend->io_size);
> >  	iomap_finish_ioends(ioend, error);
> 
> The done label can move after the call to xfs_setfilesize now.
> 

Fixed.

> Otherwise looks good:
> 
> Reviewed-by: Christoph Hellwig <hch@lst.de>
> 

Thanks.

Brian