On Tue, Apr 29, 2025 at 7:49 AM Zi Yan <z...@nvidia.com> wrote:
>
> On 28 Apr 2025, at 14:29, Nico Pache wrote:
>
> > setting /transparent_hugepages/enabled=always allows applications
> > to benefit from THPs without having to madvise. However, the pf handler
>
> s/pf/page fault
>
> > takes very few considerations to decide weather or not to actually use a
>
> s/weather/whether
>
> > THP. This can lead to a lot of wasted memory. khugepaged only operates
> > on memory that was either allocated with enabled=always or MADV_HUGEPAGE.
> >
> > Introduce the ability to set enabled=defer, which will prevent THPs from
> > being allocated by the page fault handler unless madvise is set,
> > leaving it up to khugepaged to decide which allocations will collapse to a
> > THP. This should allow applications to benefits from THPs, while curbing
> > some of the memory waste.
> >
> > Co-developed-by: Rafael Aquini <raqu...@redhat.com>
> > Signed-off-by: Rafael Aquini <raqu...@redhat.com>
> > Signed-off-by: Nico Pache <npa...@redhat.com>
> > ---
> > include/linux/huge_mm.h | 15 +++++++++++++--
> > mm/huge_memory.c | 31 +++++++++++++++++++++++++++----
> > 2 files changed, 40 insertions(+), 6 deletions(-)
> >
> > diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
> > index e3d15c737008..57e6c962afb1 100644
> > --- a/include/linux/huge_mm.h
> > +++ b/include/linux/huge_mm.h
> > @@ -48,6 +48,7 @@ enum transparent_hugepage_flag {
> > TRANSPARENT_HUGEPAGE_UNSUPPORTED,
> > TRANSPARENT_HUGEPAGE_FLAG,
> > TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG,
> > + TRANSPARENT_HUGEPAGE_DEFER_PF_INST_FLAG,
>
> What does INST mean here? Can you add one sentence on this new flag
> in the commit log to explain what it is short for?
"INSERT". Someone else commented on the length of this FLAG name. I
forgot to update it.
I can shorten it to something like ..DEFER_FLAG or DEFER_PF_FLAG
>
>
> > TRANSPARENT_HUGEPAGE_DEFRAG_DIRECT_FLAG,
> > TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_FLAG,
> > TRANSPARENT_HUGEPAGE_DEFRAG_KSWAPD_OR_MADV_FLAG,
> > @@ -186,6 +187,7 @@ static inline bool hugepage_global_enabled(void)
> > {
> > return transparent_hugepage_flags &
> > ((1<<TRANSPARENT_HUGEPAGE_FLAG) |
> > + (1<<TRANSPARENT_HUGEPAGE_DEFER_PF_INST_FLAG) |
> > (1<<TRANSPARENT_HUGEPAGE_REQ_MADV_FLAG));
> > }
> >
> > @@ -195,6 +197,12 @@ static inline bool hugepage_global_always(void)
> > (1<<TRANSPARENT_HUGEPAGE_FLAG);
> > }
> >
> > +static inline bool hugepage_global_defer(void)
> > +{
> > + return transparent_hugepage_flags &
> > + (1<<TRANSPARENT_HUGEPAGE_DEFER_PF_INST_FLAG);
> > +}
> > +
> > static inline int highest_order(unsigned long orders)
> > {
> > return fls_long(orders) - 1;
> > @@ -291,13 +299,16 @@ unsigned long thp_vma_allowable_orders(struct
> > vm_area_struct *vma,
> > unsigned long tva_flags,
> > unsigned long orders)
> > {
> > + if ((tva_flags & TVA_IN_PF) && hugepage_global_defer() &&
> > + !(vm_flags & VM_HUGEPAGE))
> > + return 0;
> > +
> > /* Optimization to check if required orders are enabled early. */
> > if ((tva_flags & TVA_ENFORCE_SYSFS) && vma_is_anonymous(vma)) {
> > unsigned long mask = READ_ONCE(huge_anon_orders_always);
> > -
>
> This newline should stay, right?
Yes, I can fix that.
>
> The rest looks good to me. Thanks. Acked-by: Zi Yan <z...@nvidia.com>
Thank you!
-- Nico
>
> Best Regards,
> Yan, Zi
>
