Re: Re: Re: Re: Re: [PATCH] RDMA/siw: Fix compiler warnings on 32-bit due to u64/pointer abuse

From: Jason Gunthorpe
Date: Mon Aug 19 2019 - 12:35:25 EST

On Mon, Aug 19, 2019 at 04:29:11PM +0000, Bernard Metzler wrote:
> >To: "Bernard Metzler" <BMT@xxxxxxxxxxxxxx>
> >From: "Jason Gunthorpe" <jgg@xxxxxxxx>
> >Date: 08/19/2019 06:05PM
> >Cc: "Geert Uytterhoeven" <geert@xxxxxxxxxxxxxx>, "Doug Ledford"
> ><dledford@xxxxxxxxxx>, linux-rdma@xxxxxxxxxxxxxxx,
> >linux-kernel@xxxxxxxxxxxxxxx
> >Subject: [EXTERNAL] Re: Re: Re: Re: [PATCH] RDMA/siw: Fix compiler
> >warnings on 32-bit due to u64/pointer abuse
> >
> >On Mon, Aug 19, 2019 at 03:54:56PM +0000, Bernard Metzler wrote:
> >
> >> Absolutely. But these addresses are conveyed through the
> >> API as unsigned 64 during post_send(), and land in the siw
> >> send queue as is. During send queue processing, these addresses
> >> must be interpreted according to its context and transformed
> >> (casted) back to the callers intention. I frankly do not
> >> know what we can do differently... The representation of
> >> all addresses as unsigned 64 is given. Sorry for the confusion.
> >
> >send work does not have pointers in it, so I'm confused what this is
> >about. Does siw allow userspace to stick an ordinary pointer for the
> >SG list?
> Right a user references a buffer by address and local key it
> got during reservation of that buffer. The user can provide any
> VA between start of that buffer and registered length.

Oh gross, it overloads the IOVA in the WR with a kernel void * ??

> >The code paths here must be totally different, so there should be
> >different types and functions for each case.
> Yes, there is a function to process application memory (siw_rx_umem),
> to process a kernel PBL (siw_rx_pbl), and one to process kernel
> addresses (siw_rx_kva). Before running that function, the API
> representation of the current SGE gets translated into target
> buffer representation.

Why does siw_pbl_get_buffer not return a void *??

Still looks like two types have been crammed together.

The kernel can't ever store anything into the user WQE buffer, so I
would think it would copy the buffer to kernel space, transform it
properly and then refer to it as a kernel buffer. Kernel sourced
buffers just skip the transofmration.