On Fri, Dec 20, 2013 at 03:58:08PM +0100, Paolo Bonzini wrote: > Linux prefers WRITE SAME to UNMAP if the count is zero, and WRITE > SAME does not discard anything unless the device can guarantee that > the resulting block is zero. > > This fixes thin provisioning on glusterfs.
I don't see discard requests reaching GlusterFS even after this patch. bdrv_co_discard or bdrv_aio_discard isn't getting called for the gluster drive. Regards, Bharata.
