On 2022/05/19 6:56, Keith Busch wrote:

On Wed, May 18, 2022 at 08:27:31PM -0700, Eric Biggers wrote:

quoted

So the bio ends up with a total length that is a multiple of the logical block
size, but the lengths of the individual bvecs in the bio are *not* necessarily
multiples of the logical block size.  That's the problem.

I'm surely missing something here. I know the bvecs are not necessarily lbs
aligned, but why does that matter? Is there some driver that can only take
exactly 1 bvec, but allows it to be unaligned? If so, we could take the segment
queue limit into account, but I am not sure that we need to.

For direct IO, the first bvec will always be aligned to a logical block size.
See __blkdev_direct_IO() and __blkdev_direct_IO_simple():

        if ((pos | iov_iter_alignment(iter)) &
            (bdev_logical_block_size(bdev) - 1))
                return -EINVAL;

And given that, all bvecs should also be LBA aligned since the LBA size is
always a divisor of the page size. Since splitting is always done on an LBA
boundary, I do not see how we can ever get bvecs that are not LBA aligned.
Or I am missing something too...

quoted

Note, there's also lots of code that assumes that bio_vec::bv_len is a multiple
of 512.  

Could you point me to some examples?

quoted

That was implied by it being a multiple of the logical block size.  But
the DMA alignment can be much lower, like 8 bytes (see nvme_set_queue_limits()).

That's the driver this was tested on, though I just changed it to 4 bytes for
5.19.

-- 
Damien Le Moal
Western Digital Research