Re: [PATCH v5] userfaultfd: opportunistic TLB-flush batching for present pages in MOVE

[Barry Song <21cnbao@gmail.com>]

On Thu, Aug 14, 2025 at 10:02 AM Lokesh Gidra <lokeshgidra@google.com> wrote:
>
> On Wed, Aug 13, 2025 at 2:47 PM Andrew Morton <akpm@linux-foundation.org> wrote:
> >
> > On Wed, 13 Aug 2025 12:30:24 -0700 Lokesh Gidra <lokeshgidra@google.com> wrote:
> >
> > > MOVE ioctl's runtime is dominated by TLB-flush cost, which is required
> > > for moving present pages. Mitigate this cost by opportunistically
> > > batching present contiguous pages for TLB flushing.
> > >
> > > Without batching, in our testing on an arm64 Android device with UFFD GC,
> > > which uses MOVE ioctl for compaction, we observed that out of the total
> > > time spent in move_pages_pte(), over 40% is in ptep_clear_flush(), and
> > > ~20% in vm_normal_folio().
> > >
> > > With batching, the proportion of vm_normal_folio() increases to over
> > > 70% of move_pages_pte() without any changes to vm_normal_folio().
> > > Furthermore, time spent within move_pages_pte() is only ~20%, which
> > > includes TLB-flush overhead.
> > >
> > > When the GC intensive benchmark, which was used to gather the above
> > > numbers, is run on cuttlefish (qemu android instance on x86_64), the
> > > completion time of the benchmark went down from ~45mins to ~20mins.
> > >
> > > Furthermore, system_server, one of the most performance critical system
> > > processes on android, saw over 50% reduction in GC compaction time on an
> > > arm64 android device.
> >
> > Were these inefficiencies a regression relative to an earlier kernel?
> No, these inefficiencies existed since MOVE ioctl came into existence.
> IOW, the patch is improving performance, rather than fixing a
> regression.

It would be nice if Lokesh can backport it to older Android common kernel such
as 6.6, 6.12 etc.

Thanks
Barry