Re: [PATCH] RDMA/irdma: Check for null udata during reg_user_mr

[Leon Romanovsky <leon@kernel.org>]

On Sun, Jul 05, 2026 at 05:09:40PM +0300, Leon Romanovsky wrote:
> On Tue, Jun 30, 2026 at 03:47:19PM +0000, Jacob Moroni wrote:
> > The irdma driver requires udata for reg_user_mr. Previously,
> > it was assuming that it was always non-null, but it can actually
> > be null if the user intentionally triggers the UVERBS_METHOD_REG_MR
> > ioctl.
> > 
> > Signed-off-by: Jacob Moroni <jmoroni@google.com>
> > ---
> >  drivers/infiniband/hw/irdma/verbs.c | 3 +++
> >  1 file changed, 3 insertions(+)
> > 
> > diff --git a/drivers/infiniband/hw/irdma/verbs.c b/drivers/infiniband/hw/irdma/verbs.c
> > index cb54c7c8fcd..3bcac68a612 100644
> > --- a/drivers/infiniband/hw/irdma/verbs.c
> > +++ b/drivers/infiniband/hw/irdma/verbs.c
> > @@ -3519,6 +3519,9 @@ static struct ib_mr *irdma_reg_user_mr(struct ib_pd *pd, u64 start, u64 len,
> >  	if (dmah)
> >  		return ERR_PTR(-EOPNOTSUPP);
> >  
> > +	if (!udata)
> > +		return ERR_PTR(-EINVAL);
> 
> Is this check specific to iRDMA?

Ohh, I see that it is fixed in  011199f46f44 ("RDMA/uverbs: Add
UVERBS_ATTR_UHW to UVERBS_METHOD_REG_MR").

Thanks