On Mon, 03 Mar 2008 22:55:56 +0900
Tejun Heo <htejun@xxxxxxxxx> wrote:
> FUJITA Tomonori wrote:
> >>>> FUJITA Tomonori wrote:
> >>>>>>> I can't see what changing the meaning of rq->data_len (and
> >>>>>>> investigating all the block drivers) gives us.
> >>>>>> No matter which way you go, you change the meaning of rq->data_len and
> >>>>>> you MUST inspect rq->data_len usage whichever way you go.
> >>>>> The patch doens't change that rq->data_len means the true data
> >>>>> length. But yeah, it breaks rq->data_len == sum(sg). So it might break
> >>>>> some drivers.
> >>>> Yeah, that's what I was saying. You end up breaking one of the two
> >>>> assumptions. As sglist is getting modified for any driver if it has DMA
> >>>> alignment set, whether rq->data_len is adjusted together or not, sglist
> >>>> and data_len usages have to be audited.
> >>> My patch (well, James' original approach) doesn't affect drivers that
> >>> don't use drain buffer. rq->data_len still means the true data length
> >>> and rq->data_len is equal to sum(sg) for them. So right now we need to
> >>> audit only libata.
> >> Your patch does change sglist for any driver which sets DMA alignment.
> >
> > I overlook it. Where does it changes sglist?
>
> At the end of blk_rq_map_user() together with data_len / extra_len
> mangling or were you talking about James' original patch?
With my patch, at the end of blk_rq_map_user, we have:
if (len & queue_dma_alignment(q)) {
unsigned int pad_len = (queue_dma_alignment(q) & ~len) + 1;
rq->extra_len += pad_len;
}
So no change as compared with 2.6.24?
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
