On Fri, 17 Dec 2021 12:38:46 +0200 Mika Westerberg <mika.westerberg@xxxxxxxxxxxxxxx> wrote: > Hi, > > On Thu, Dec 16, 2021 at 05:43:47PM +0100, Boris Brezillon wrote: > > Why no just > > > > if (iop->mem_op.addr.nbytes != op->addr.nbytes || > > iop->mem_op.addr.dtr != op->addr.dtr) > > return false; > > > > then? > > That would be too simple ;-) > > I'll do this change in v5. I didn't review the driver thoroughly, but it looks clean an tidy. Feel free to add Reviewed-by: Boris Brezillon <boris.brezillon@xxxxxxxxxxxxx> One more thing. I think it'd be good to implement the dirmap hooks (in a follow-up series, not here), so you don't have to iterate over the ops table every time a read/write request is issued. Should be as simple as storing the exec_op() pointer in the driver-private dirmap pointer, and calling this function when ->dirmap_{read,write}() is called.