On Mon, 18 Sept 2023 at 17:22, Christian Brauner [off-list ref] wrote:

quoted

So to be clear about your proposal: .mnt_root and .mountpoint are
initialized by the caller to buffers that the kernel can copy paths
into?

Yeah, u64 pointer to a buffer and a size (see e.g., @set_tid and
@set_tid_size for struct clone_args, @log_buf and @log_size and other
args in there).

quoted

If there's an overflow (one of the buffers was too small) the syscall
returns -EOVERFLOW?

Yeah, I mean we have to make some things their problem.

To me that is an acceptable compromise.

Okay, so there are now (at least) two buffers, and on overflow the
caller cannot know which one got overflown.  It can resize both, but
that doesn't make the caller any simpler to implement.

Also the interface is kind of weird in that some struct members are
out, some are in (the pointers and the lengths).

I'd prefer the single buffer interface, which has none of the above issues.

Thanks,
Miklos