Re: [PATCH 00/12] mm: add bitmap VMA flag helpers and convert all mmap_prepare to use them

On Mon, Jan 19, 2026 at 11:33:02AM -0800, Andrew Morton wrote:
> On Mon, 19 Jan 2026 14:48:51 +0000 Lorenzo Stoakes <[email protected]> wrote:
>
> > We introduced the bitmap VMA type vma_flags_t in the aptly named commit
> > 9ea35a25d51b ("mm: introduce VMA flags bitmap type") in order to permit
> > future growth in VMA flags and to prevent the asinine requirement that VMA
> > flags be available to 64-bit kernels only if they happened to use a bit
> > number about 32-bits.
> >
> > This is a long-term project as there are very many users of VMA flags
> > within the kernel that need to be updated in order to utilise this new
> > type.
>
> Thanks, let's give this a run in mm-new for a few days, see if that
> helps shake anything out.  I didn't add [11/12] due to a significant
> merge clash, but it compiles!

I mean I'm not sure what testing is running in mm-new anyway other than
David's private bot, but that patch is key to the series, albeit thankfully
only affecting VMA userland tests.

However unfortunately _nobody_ is running VMA userlands tests except me
locally (we did ask, more than once kernelci people but it seems they are
too busy), so I guess we may as well leave there FWIW I guess.

If the change that conflicts is in mm-unstable by tomorrow I can rebase and
respin.

I can't base on mm-new as the criteria for inclusion are extremely
confusing right now and everything there is untested so it's like playing
whack-a-mole a bit. Hopefully your forthcoming documentation will sort that
out (and hopefully it aligns with what the community would like).

Anyway I don't have expectation that this series will be taken this cycle,
rather it's for:

- Review
- Testing
- To get the painful merge issues dealt with ahead of time

As a result it'd be good to get into mm-unstable sooner rather than later
though since that's what's for linux-next I guess during merge window?
Though it becomes a little unclear what the tree state is then so not sure.

The intent with patch 11/12 re: vma userland test changes is to make it
MUCH easier to maintain and to avoid this kind of thing in future. So
short-term pain for long-term gain.

Thanks, Lorenzo.

--
To unsubscribe, send a message with 'unsubscribe linux-aio' in
the body to [email protected].  For more info on Linux AIO,
see: http://www.kvack.org/aio/
Don't email: <a href=mailto:"[email protected]">[email protected]</a>