On 24/01/2020 4:52, Jason Gunthorpe wrote: > On Fri, Jan 24, 2020 at 12:40:18AM +0000, Saleem, Shiraz wrote: >> It would be good to get the debug data to back this or prove it wrong. >> But if this is indeed what's happening, then ORing in the sgl->length for the >> first sge to restrict the page size might cut it. So something like, > > or'ing in the sgl length is a nonsense thing to do, the length has > nothing to do with the restriction, which is entirely based on IOVA > bits which can't be passed through. The weekend runs passed with Leon's proposed patch. Leon, can you please submit it so I can drop this revert? Thanks
