| Message ID | 20220825152425.6296-6-logang@deltatee.com |
|---|---|
| State | New |
| Headers | show |
| Series | Userspace P2PDMA with O_DIRECT NVMe devices | expand |
On Thu, Aug 25, 2022 at 09:24:22AM -0600, Logan Gunthorpe wrote: > + if (bio->bi_bdev && bio->bi_bdev->bd_disk && > + blk_queue_pci_p2pdma(bio->bi_bdev->bd_disk->queue)) bdev->bd_disk is never NULL. Otherwise looks good: Reviewed-by: Christoph Hellwig <hch@lst.de>
On 8/25/22 08:24, Logan Gunthorpe wrote: > When a bio's queue supports PCI P2PDMA, set FOLL_PCI_P2PDMA for > iov_iter_get_pages_flags(). This allows PCI P2PDMA pages to be passed > from userspace and enables the O_DIRECT path in iomap based filesystems > and direct to block devices. Oh great, more O_DIRECT code paths. :) > > Signed-off-by: Logan Gunthorpe <logang@deltatee.com> > --- > block/bio.c | 10 ++++++++-- > 1 file changed, 8 insertions(+), 2 deletions(-) > > diff --git a/block/bio.c b/block/bio.c > index 969607bc1f4d..ca09d79a0683 100644 > --- a/block/bio.c > +++ b/block/bio.c > @@ -1200,6 +1200,7 @@ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) > unsigned short entries_left = bio->bi_max_vecs - bio->bi_vcnt; > struct bio_vec *bv = bio->bi_io_vec + bio->bi_vcnt; > struct page **pages = (struct page **)bv; > + unsigned int flags = 0; It would be nice to name this gup_flags, instead. > ssize_t size, left; > unsigned len, i = 0; > size_t offset, trim; > @@ -1213,6 +1214,10 @@ 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); > > + if (bio->bi_bdev && bio->bi_bdev->bd_disk && > + blk_queue_pci_p2pdma(bio->bi_bdev->bd_disk->queue)) > + flags |= FOLL_PCI_P2PDMA; > + > /* > * 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 > @@ -1220,8 +1225,9 @@ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) > * 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. > */ > - size = iov_iter_get_pages2(iter, pages, UINT_MAX - bio->bi_iter.bi_size, > - nr_pages, &offset); > + size = iov_iter_get_pages_flags(iter, pages, > + UINT_MAX - bio->bi_iter.bi_size, > + nr_pages, &offset, flags); > if (unlikely(size <= 0)) > return size ? size : -EFAULT; > Looks good. After applying Christoph's tweak, and optionally the gup_flags rename as well, please feel free to add: Reviewed-by: John Hubbard <jhubbard@nvidia.com> thanks,
diff --git a/block/bio.c b/block/bio.c index 969607bc1f4d..ca09d79a0683 100644 --- a/block/bio.c +++ b/block/bio.c @@ -1200,6 +1200,7 @@ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) unsigned short entries_left = bio->bi_max_vecs - bio->bi_vcnt; struct bio_vec *bv = bio->bi_io_vec + bio->bi_vcnt; struct page **pages = (struct page **)bv; + unsigned int flags = 0; ssize_t size, left; unsigned len, i = 0; size_t offset, trim; @@ -1213,6 +1214,10 @@ 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); + if (bio->bi_bdev && bio->bi_bdev->bd_disk && + blk_queue_pci_p2pdma(bio->bi_bdev->bd_disk->queue)) + flags |= FOLL_PCI_P2PDMA; + /* * 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 @@ -1220,8 +1225,9 @@ static int __bio_iov_iter_get_pages(struct bio *bio, struct iov_iter *iter) * 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. */ - size = iov_iter_get_pages2(iter, pages, UINT_MAX - bio->bi_iter.bi_size, - nr_pages, &offset); + size = iov_iter_get_pages_flags(iter, pages, + UINT_MAX - bio->bi_iter.bi_size, + nr_pages, &offset, flags); if (unlikely(size <= 0)) return size ? size : -EFAULT;
When a bio's queue supports PCI P2PDMA, set FOLL_PCI_P2PDMA for iov_iter_get_pages_flags(). This allows PCI P2PDMA pages to be passed from userspace and enables the O_DIRECT path in iomap based filesystems and direct to block devices. Signed-off-by: Logan Gunthorpe <logang@deltatee.com> --- block/bio.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-)