On Thu, Jun 24, 2021 at 03:00:39AM +0100, Matthew Wilcox wrote:

On Wed, Jun 23, 2021 at 02:58:32PM -0700, Omar Sandoval wrote:

quoted

But what you described would look more like:

	// Needs to be large enough for maximum returned header + data.
	char buffer[...];
	struct iovec iov[] = {
		{ buffer, sizeof(buffer) },
	};
	preadv2(fd, iov, 2, -1, RWF_ENCODED);
	// We should probably align the buffer.
	struct encoded_iov *encoded_iov = (void *)buffer;
	char *data = buffer + encoded_iov->size;

That's a little uglier, but it should work, and allows for arbitrary
extensions. So, among these three alternatives (fixed size structure
with reserved space, variable size structure like above, or ioctl),
which would you prefer?

Does that work for O_DIRECT and the required 512-byte alignment?

I suppose the kernel could pad the encoded_iov structure with zeroes to
the next sector boundary, since zeroes are effectively noops for
encoded_iov. (As an aside, RWF_ENCODED is always "direct I/O" in the
sense that it bypasses the page cache, but not necessarily in the sense
that it does DMA to/from the user buffers. The Btrfs implementation
doesn't do the latter yet.)