Re: [PATCH v8 09/14] powerpc: move has_transparent_hugepage() out of THP guard

[Christophe Leroy (CS GROUP)]

Le 05/10/2026 à 22:40, Luiz Capitulino a écrit :
> On 10/5/26 12:52 AM, Christophe Leroy (CS GROUP) wrote:
> > Hi,
> >
> > Don't forget <linuxppc-dev@xxxxxxxxxxxxxxxx> when you address powerpc 
> > architecture.
>
> Will do.
>
> > Le 03/10/2026 à 17:44, Luiz Capitulino a écrit :
> > > On 10/2/26 3:28 PM, David Hildenbrand (Arm) wrote:
> > > > On 9/18/26 03:45, Luiz Capitulino wrote:
> > > > > A future commit will introduce a kernel API to allow for checking 
> > > > > if the
> > > > > CPU supports PMD-sized pages. This API will be based on the
> > > > > has_transparent_hugepage() implementation but will be orthogonal to 
> > > > > THP
> > > > > and therefore must work when CONFIG_TRANSPARENT_HUGEPAGE=n.
> > > > >
> > > > > Move its definition out of the THP guard.
> > > > >
> > > > > Signed-off-by: Luiz Capitulino <luizcap@xxxxxxxxxx>
> > > > > ---
> > > > >   arch/powerpc/include/asm/book3s/64/hash-4k.h  |  2 +-
> > > > >   arch/powerpc/include/asm/book3s/64/hash-64k.h |  2 +-
> > > > >   arch/powerpc/include/asm/book3s/64/pgtable.h  | 18 ++++++++ 
> > > > > +---------
> > > > >   arch/powerpc/include/asm/book3s/64/radix.h    | 14 +++++++-------
> > > > >   arch/powerpc/mm/book3s64/hash_pgtable.c       |  4 ++--
> > > > >   5 files changed, 20 insertions(+), 20 deletions(-)
> > > > >
> > > > > diff --git a/arch/powerpc/include/asm/book3s/64/hash-4k.h b/arch/ 
> > > > > powerpc/include/asm/book3s/64/hash-4k.h
> > > > > index 8e5bd9902bed..79511e6abfca 100644
> > > > > --- a/arch/powerpc/include/asm/book3s/64/hash-4k.h
> > > > > +++ b/arch/powerpc/include/asm/book3s/64/hash-4k.h
> > > > > @@ -165,9 +165,9 @@ extern void 
> > > > > hash__pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp,
> > > > >   extern pgtable_t hash__pgtable_trans_huge_withdraw(struct 
> > > > > mm_struct *mm, pmd_t *pmdp);
> > > > >   extern pmd_t hash__pmdp_huge_get_and_clear(struct mm_struct *mm,
> > > > >                          unsigned long addr, pmd_t *pmdp);
> > > > > -extern int hash__has_transparent_hugepage(void);
> > > > >   #endif
> > > > > +extern int hash__has_transparent_hugepage(void);
> > > > >   #endif /* !__ASSEMBLER__ */
> > > > >   #endif /* _ASM_POWERPC_BOOK3S_64_HASH_4K_H */
> > > > > diff --git a/arch/powerpc/include/asm/book3s/64/hash-64k.h b/arch/ 
> > > > > powerpc/include/asm/book3s/64/hash-64k.h
> > > > > index 7deb3a66890b..a4a44a112ff9 100644
> > > > > --- a/arch/powerpc/include/asm/book3s/64/hash-64k.h
> > > > > +++ b/arch/powerpc/include/asm/book3s/64/hash-64k.h
> > > > > @@ -278,9 +278,9 @@ extern void 
> > > > > hash__pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp,
> > > > >   extern pgtable_t hash__pgtable_trans_huge_withdraw(struct 
> > > > > mm_struct *mm, pmd_t *pmdp);
> > > > >   extern pmd_t hash__pmdp_huge_get_and_clear(struct mm_struct *mm,
> > > > >                          unsigned long addr, pmd_t *pmdp);
> > > > > -extern int hash__has_transparent_hugepage(void);
> > > >
> > > > Without some of these helpers in place, I assume actually using PMD 
> > > > leafs
> > > > without THP would require some more work. (which is not the goal of 
> > > > this series,
> > > > just asking).
> > >
> > > Yes, you're right.
> > >
> > > > I do wonder whether the architecture should instead simply say "not 
> > > > supported"
> > > > if !CONFIG_TRANSPARENT_HUGEPAGE?
> > > >
> > > > That should still enable your series: using mTHP without PMD support.
> > >
> > > Yes, it would. However, I think this introduces an inconsistency.
> >
> > powerpc has two types of MMU (HASH and RADIX) with different page 
> > layout. MMU type is selected at boottime based on the capabilities of 
> > the CPU. For transparent page you have:
> >
> > static inline int has_transparent_hugepage(void)
> > {
> >      if (radix_enabled())
> >          return radix__has_transparent_hugepage();
> >      return hash__has_transparent_hugepage();
> > }
>
> Right and, unless I'm misunderstanding your comment, this supports my
> position that the new arch_has_pmd_leaves() API is all about reporting a
> hardware capability and not tied to THP support.

Yes my comment was in reaction of comment below, I wanted to say that 
CONFIG_TRANSPARENT_HUGEPAGE is not always enough to tell if a powerpc 
actually has transparent hugepages or not.

>>>> I do wonder whether the architecture should instead simply say "not
>>>> supported"
>>>> if !CONFIG_TRANSPARENT_HUGEPAGE?
>>>>
>>>> That should still enable your series: using mTHP without PMD support.