Re: [PATCH 1/4] block: bio-integrity: add support for user buffers

[Christoph Hellwig <hch@lst.de>]

int bio_integrity_map_user(struct bio *bio, void __user *ubuf, unsigned int len,
> +			   u32 seed, u32 maxvecs)
> +{
> +	struct request_queue *q = bdev_get_queue(bio->bi_bdev);
> +	unsigned long align = q->dma_pad_mask | queue_dma_alignment(q);
> +	struct page *stack_pages[UIO_FASTIOV];
> +	size_t offset = offset_in_page(ubuf);
> +	unsigned long ptr = (uintptr_t)ubuf;
> +	struct page **pages = stack_pages;
> +	struct bio_integrity_payload *bip;
> +	int npages, ret, i;
> +
> +	if (bio_integrity(bio) || ptr & align || maxvecs > UIO_FASTIOV)
> +		return -EINVAL;

We also need to check the length for the dma alignment/pad, not
just the start.  (The undocumented iov_iter_alignment_iovec helper
obsfucateѕ this for the data path).

> +	bip = bio_integrity_alloc(bio, GFP_KERNEL, maxvecs);
> +	if (IS_ERR(bip))
> +		return PTR_ERR(bip);
> +
> +	ret = pin_user_pages_fast(ptr, UIO_FASTIOV, FOLL_WRITE, pages);
> +	if (unlikely(ret < 0))
> +		goto free_bip;
> +
> +	npages = ret;
> +	for (i = 0; i < npages; i++) {
> +		u32 bytes = min_t(u32, len, PAGE_SIZE - offset);
> +		ret = bio_integrity_add_page(bio, pages[i], bytes, offset);
> +		if (ret != bytes) {
> +			ret = -EINVAL;
> +			goto release_pages;
> +		}
> +		len -= ret;
> +		offset = 0;
> +	}

Any reason to not use the bio_vec array as the buffer, similar to the
data size here?

> +EXPORT_SYMBOL(bio_integrity_map_user);

Everything that just thinly wraps get_user_pages_fast needs to be
EXPORT_SYMBOL_GPL.