Re: [PATCH v3 00/34] New page table range API
From: Ryan Roberts
Date: Thu Mar 09 2023 - 06:14:20 EST
On 28/02/2023 21:37, Matthew Wilcox (Oracle) wrote:
> This patchset changes the API used by the MM to set up page table entries.
> The four APIs are:
> set_ptes(mm, addr, ptep, pte, nr)
> update_mmu_cache_range(vma, addr, ptep, nr)
> flush_dcache_folio(folio)
> flush_icache_pages(vma, page, nr)
>
> flush_dcache_folio() isn't technically new, but no architecture
> implemented it, so I've done that for you. The old APIs remain around
> but are mostly implemented by calling the new interfaces.
>
> The new APIs are based around setting up N page table entries at once.
> The N entries belong to the same PMD, the same folio and the same VMA,
> so ptep++ is a legitimate operation, and locking is taken care of for
> you. Some architectures can do a better job of it than just a loop,
> but I have hesitated to make too deep a change to architectures I don't
> understand well.
>
> One thing I have changed in every architecture is that PG_arch_1 is now a
> per-folio bit instead of a per-page bit. This was something that would
> have to happen eventually, and it makes sense to do it now rather than
> iterate over every page involved in a cache flush and figure out if it
> needs to happen.
>
> The point of all this is better performance, and Fengwei Yin has
> measured improvement on x86. I suspect you'll see improvement on
> your architecture too. Try the new will-it-scale test mentioned here:
> https://lore.kernel.org/linux-mm/20230206140639.538867-5-fengwei.yin@xxxxxxxxx/
> You'll need to run it on an XFS filesystem and have
> CONFIG_TRANSPARENT_HUGEPAGE set.
>
> For testing, I've only run the code on x86. If an x86->foo compiler
> exists in Debian, I've built defconfig. I'm relying on the buildbots
> to tell me what I missed, and people who actually have the hardware to
> tell me if it actually works.
>
> I'd like to get this into the MM tree soon after the current merge window
> closes, so quick feedback would be appreciated.
I've boot-tested the series (with the Yin's typo fix for patch 32) on arm64 FVP
and Ampere Altra. On the Altra, I also ran page_fault4 from will-it-scale, and
see ~35% improvement from this series. So:
Tested-by: Ryan Roberts <ryan.roberts@xxxxxxx>
Thanks,
Ryan