Re: [PATCH v5 07/14] mm: shmem: allow THP support determination at folio allocation time

[Luiz Capitulino]

On 2026-06-01 08:09, Baolin Wang wrote:
> On 5/29/26 10:55 PM, Luiz Capitulino wrote:
> > In order to enable THP support in shmem today, besides the user
> > configuration required, the CPU must support PMD-sized pages. This
> > is the case because of the following has_transparent_hugepage()
> > usage:
> >
> > - shmem_parse_one() and shmem_parse_huge(): Check if THP is built-in and
> >    if the CPU supports PMD-sized pages
> >
> > - shmem_init(): Since the CONFIG_TRANSPARENT_HUGEPAGE guard is outside
> >    the code block calling has_transparent_hugepage(), the
> >    has_transparent_hugepage() call is exclusively checking if the CPU
> >    supports PMD-sized pages
> >
> > While it's necessary to check if CONFIG_TRANSPARENT_HUGEPAGE is enabled
> > in all cases, shmem can determine THP size support at folio allocation
> > time. Therefore, drop the has_transparent_hugepage() usage listed above
> > while keeping the CONFIG_TRANSPARENT_HUGEPAGE checks.
> >
> > Additionally, we need to check if PMD size order is supported in
> > shmem_getattr(). Use pgtable_has_pmd_leaves() for that.
> >
> > Signed-off-by: Luiz Capitulino <luizcap@xxxxxxxxxx>
> > ---
> >   mm/shmem.c | 9 +++++----
> >   1 file changed, 5 insertions(+), 4 deletions(-)
> >
> > diff --git a/mm/shmem.c b/mm/shmem.c
> > index 88bc0c292f1a..108dc969b92a 100644
> > --- a/mm/shmem.c
> > +++ b/mm/shmem.c
> > @@ -689,7 +689,7 @@ static int shmem_parse_huge(const char *str)
> >       else
> >           return -EINVAL;
> > -    if (!has_transparent_hugepage() &&
> > +    if (!IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) &&
> >           huge != SHMEM_HUGE_NEVER && huge != SHMEM_HUGE_DENY)
> >           return -EINVAL;
> > @@ -1303,6 +1303,8 @@ static int shmem_getattr(struct mnt_idmap *idmap,
> >       generic_fillattr(idmap, request_mask, inode, stat);
> >       orders = shmem_huge_global_enabled(inode, 0, 0, false, NULL, 0);
> > +    if (!pgtable_has_pmd_leaves())
> > +        orders &= ~BIT(PMD_ORDER);
>
> Ah, you did in this patch, but this change should belong in patch 14?

If we move it, then this patch will add a temporary issue until patch 14
where shmem_getattr() may report PMD-size for blksize even when the CPU
doesn't support PMD-sized pages.