Re: [PATCH v4 0/2] fix MADV_COLLAPSE issue if THP settings are disabled

[Lorenzo Stoakes <lorenzo.stoakes@oracle.com>]

On Wed, Jun 25, 2025 at 10:40:23AM +0200, David Hildenbrand wrote:
>
> I just looked at the original use cases [1], such a use case is not
> mentioned.
>
> But it did add process_madvise(MADV_COLLAPSE) in
> 876b4a1896646cc85ec6b1fc1c9270928b7e0831 where we document
>
> "
>     This is useful for the development of userspace agents that seek to
>     optimize THP utilization system-wide by using userspace signals to
>     prioritize what memory is most deserving of being THP-backed.
> "
>
> The "prioritize" might indicate that this is used in combination with
> "madvise", not with "never"/
>
>
> So yeah, it all boils down to
>
> (1) If there is no such use case, "never can mean never". Because there
>     is nothing to break, really.

Yeah and as I said in my first response to Hugh, I think the (unintentional)
vagueness of the docs means we are not breaking documented behaviour anyway.

>
> (2) If there is such a use case, we might be breaking it.
>
> [1]
> https://lore.kernel.org/linux-mm/20220706235936.2197195-1-zokeefe@google.com/

Agreed.

This speaks to the 'I want THP now' being the intent of MADV_COLLAPSE. Not a
never/manual (but immediate) control over THP assignment.

Which aligns with the docs:

"Perform a best-effort synchronous collapse of the native pages mapped by the
 memory range into Transparent Huge Pages (THPs)."