On 11/21/22 4:40 PM, Eric Farman wrote:
> The intention is to read the first IDAW to determine the starting
> location of an I/O operation, knowing that the second and any/all
> subsequent IDAWs will be aligned per architecture. But, this read
> receives 64-bits of data, which is the size of a format-2 IDAW.
>
> In the event that Format-1 IDAWs are presented, discard the lower
> 32 bits as they contain the second IDAW in such a list.
>
> Signed-off-by: Eric Farman <farman@xxxxxxxxxxxxx>
> ---
> drivers/s390/cio/vfio_ccw_cp.c | 8 ++++++--
> 1 file changed, 6 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/s390/cio/vfio_ccw_cp.c b/drivers/s390/cio/vfio_ccw_cp.c
> index 34a133d962d1..53246f4f95f7 100644
> --- a/drivers/s390/cio/vfio_ccw_cp.c
> +++ b/drivers/s390/cio/vfio_ccw_cp.c
> @@ -516,11 +516,15 @@ static int ccw_count_idaws(struct ccw1 *ccw,
> bytes = ccw->count;
>
> if (ccw_is_idal(ccw)) {
> - /* Read first IDAW to see if it's 4K-aligned or not. */
> - /* All subsequent IDAws will be 4K-aligned. */
> + /* Read first IDAW to check its starting address. */
> + /* All subsequent IDAWs will be 2K- or 4K-aligned. */
> ret = vfio_dma_rw(vdev, ccw->cda, &iova, sizeof(iova), false);
> if (ret)
> return ret;
> +
> + /* Format-1 IDAWs only occupy the first int */
> + if (!cp->orb.cmd.c64)
> + iova = iova >> 32;
Rather than read 8B and discarding 4B, can't we check this format value first and only read 4B for format-1?
> } else {
> iova = ccw->cda;
> }
