Re: [PATCH v4] mm/zswap: store <PAGE_SIZE compression failed page as-is

[Chris Li]

On Tue, Aug 19, 2025 at 6:21 PM Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx> wrote:
>
> On Wed, Aug 20, 2025 at 01:13:13PM +1200, Barry Song wrote:
> >
> > Let’s sync with Herbert: have we reached the stage where all drivers
> > reliably return -ENOSPC when dst_buf is PAGE_SIZE but the compressed
> > size would exceed it?
>
> It doesn't matter.  Software compression should never fail, and
> if it does fail due to a bug, that's not something that the user
> of the API should worry about.
>
> Hardware compression should always fall back to software compression
> in case of failure.
>
> IOW all compression errors should be treated as incompressible
> data.

In that case, you seem to suggest that the crypto library can NEVER
return an error other than the dst buffer is too small. In that case I
would suggest the crypto library remove that free form error return
code. Just return true for success compression or false on failure.

It is also strange to return a freeform error code and declare the
caller shouldn't check on it. Ideally the caller shouldn't warn or bug
on it, so it should handle the possible error other than -ENOSPC.

Chris