On Mon, May 23, 2022 at 02:01:19PM -0700, Keith Busch wrote:
> Removed iomap support for now
Do you plan to add a separate patch for it? It would be a shame to
miss it especially as you said you already tested xfs.
> + struct request_queue *q = bdev_get_queue(bio->bi_bdev);
> struct bio_vec *bv = bio->bi_io_vec + bio->bi_vcnt;
> struct page **pages = (struct page **)bv;
> ssize_t size, left;
> @@ -1219,7 +1220,19 @@ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter)
> BUILD_BUG_ON(PAGE_PTRS_PER_BVEC < 2);
> pages += entries_left * (PAGE_PTRS_PER_BVEC - 1);
>
> + /*
> + * Each segment in the iov is required to be a block size multiple.
> + * However, we may not be able to get the entire segment if it spans
> + * more pages than bi_max_vecs allows, so we have to ALIGN_DOWN the
> + * result to ensure the bio's total size is correct. The remainder of
> + * the iov data will be picked up in the next bio iteration.
> + *
> + * If the result is ever 0, that indicates the iov fails the segment
> + * size requirement and is an error.
> + */
> size = iov_iter_get_pages(iter, pages, LONG_MAX, nr_pages, &offset);
> + if (size > 0)
> + size = ALIGN_DOWN(size, queue_logical_block_size(q));
I think we can simply use bdev_logical_block_size here and remove the need
for the q local variable.
Given that bio_iov_iter_get_pages is used by more than the block device
direct I/O code maybe split this up further?
> @@ -42,6 +42,16 @@ static unsigned int dio_bio_write_op(struct kiocb *iocb)
> return op;
> }
>
> +static int blkdev_dio_aligned(struct block_device *bdev, loff_t pos,
> + struct iov_iter *iter)
> +{
> + if ((pos | iov_iter_count(iter)) & (bdev_logical_block_size(bdev) - 1))
> + return -EINVAL;
> + if (iov_iter_alignment(iter) & bdev_dma_alignment(bdev))
> + return -EINVAL;
> + return 0;
> +}
I'd also split adding this helper into another prep patch to see
the actual change in behavior more easily.
