Re: [PATCH v1 00/11] mm, sparse-vmemmap: Introduce compound pagemaps

From: Joao Martins <joao.m.martins@oracle.com>
To: linux-mm@kvack.org
Cc: linux-nvdimm@lists.01.org, Matthew Wilcox <willy@infradead.org>,
	Jason Gunthorpe <jgg@ziepe.ca>,
	Muchun Song <songmuchun@bytedance.com>,
	Mike Kravetz <mike.kravetz@oracle.com>,
	Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [PATCH v1 00/11] mm, sparse-vmemmap: Introduce compound pagemaps
Date: Thu, 1 Apr 2021 10:38:04 +0100	[thread overview]
Message-ID: <a2b7dc3a-3f1d-b8e1-4770-e261055f2b17@oracle.com> (raw)
In-Reply-To: <20210325230938.30752-1-joao.m.martins@oracle.com>

On 3/25/21 11:09 PM, Joao Martins wrote:

[...]

> Patch 11: Optimize grabbing page refcount changes given that we
> are working with compound pages i.e. we do 1 increment to the head
> page for a given set of N subpages compared as opposed to N individual writes.
> {get,pin}_user_pages_fast() for zone_device with compound pagemap consequently
> improves considerably with DRAM stored struct pages. It also *greatly*
> improves pinning with altmap. Results with gup_test:
> 
>                                                    before     after
>     (16G get_user_pages_fast 2M page size)         ~59 ms -> ~6.1 ms
>     (16G pin_user_pages_fast 2M page size)         ~87 ms -> ~6.2 ms
>     (16G get_user_pages_fast altmap 2M page size) ~494 ms -> ~9 ms
>     (16G pin_user_pages_fast altmap 2M page size) ~494 ms -> ~10 ms
> 
>     altmap performance gets specially interesting when pinning a pmem dimm:
> 
>                                                    before     after
>     (128G get_user_pages_fast 2M page size)         ~492 ms -> ~49 ms
>     (128G pin_user_pages_fast 2M page size)         ~493 ms -> ~50 ms
>     (128G get_user_pages_fast altmap 2M page size)  ~3.91 ms -> ~70 ms
>     (128G pin_user_pages_fast altmap 2M page size)  ~3.97 ms -> ~74 ms
> 

Quick correction: These last two 3.91 and 3.97 on the left column are in *seconds* not
milliseconds. By mistake I added an extra 'm'. Sorry about that.

> The unpinning improvement patches are in mmotm/linux-next so removed from this
> series.
> 
> I have deferred the __get_user_pages() patch to outside this series
> (https://lore.kernel.org/linux-mm/20201208172901.17384-11-joao.m.martins@oracle.com/),
> as I found an simpler way to address it and that is also applicable to
> THP. But will submit that as a follow up of this.
> 
> Patches apply on top of linux-next tag next-20210325 (commit b4f20b70784a).
> 
> Comments and suggestions very much appreciated!
> 
> Changelog,
> 
>  RFC -> v1:
>  (New patches 1-3, 5-8 but the diffstat is that different)
>  * Fix hwpoisoning of devmap pages reported by Jane (Patch 1 is new in v1)
>  * Fix/Massage commit messages to be more clear and remove the 'we' occurences (Dan, John, Matthew)

I just noticed that I haven't fully removed the 'we' occurrences. Patches 7 and 8 had
their commit messages rewritten and I mistakenly re-introduced remnants of 'we'. I will
have it fixed for (v2) albeit I'll still wait for comments on this series before following up.

>  * Use pfn_align to be clear it's nr of pages for @align value (John, Dan)
>  * Add two helpers pgmap_align() and pgmap_pfn_align() as accessors of pgmap->align;
>  * Remove the gup_device_compound_huge special path and have the same code
>    work both ways while special casing when devmap page is compound (Jason, John)
>  * Avoid usage of vmemmap_populate_basepages() and introduce a first class
>    loop that doesn't care about passing an altmap for memmap reuse. (Dan)
>  * Completely rework the vmemmap_populate_compound() to avoid the sparse_add_section
>    hack into passing block across sparse_add_section calls. It's a lot easier to
>    follow and more explicit in what it does.
>  * Replace the vmemmap refactoring with adding a @pgmap argument and moving
>    parts of the vmemmap_populate_base_pages(). (Patch 5 and 6 are new as a result)
>  * Add PMD tail page vmemmap area reuse for 1GB pages. (Patch 8 is new)
>  * Improve memmap_init_zone_device() to initialize compound pages when
>    struct pages are cache warm. That lead to a even further speed up further
>    from RFC series from 190ms -> 80-120ms. Patches 2 and 3 are the new ones
>    as a result (Dan)
>  * Remove PGMAP_COMPOUND and use @align as the property to detect whether
>    or not to reuse vmemmap areas (Dan)
> 
> Thanks,
> 	Joao
> 
> Joao Martins (11):
>   memory-failure: fetch compound_head after pgmap_pfn_valid()
>   mm/page_alloc: split prep_compound_page into head and tail subparts
>   mm/page_alloc: refactor memmap_init_zone_device() page init
>   mm/memremap: add ZONE_DEVICE support for compound pages
>   mm/sparse-vmemmap: add a pgmap argument to section activation
>   mm/sparse-vmemmap: refactor vmemmap_populate_basepages()
>   mm/sparse-vmemmap: populate compound pagemaps
>   mm/sparse-vmemmap: use hugepages for PUD compound pagemaps
>   mm/page_alloc: reuse tail struct pages for compound pagemaps
>   device-dax: compound pagemap support
>   mm/gup: grab head page refcount once for group of subpages
> 
>  drivers/dax/device.c           |  58 +++++++--
>  include/linux/memory_hotplug.h |   5 +-
>  include/linux/memremap.h       |  13 ++
>  include/linux/mm.h             |   8 +-
>  mm/gup.c                       |  52 +++++---
>  mm/memory-failure.c            |   2 +
>  mm/memory_hotplug.c            |   3 +-
>  mm/memremap.c                  |   9 +-
>  mm/page_alloc.c                | 126 +++++++++++++------
>  mm/sparse-vmemmap.c            | 221 +++++++++++++++++++++++++++++----
>  mm/sparse.c                    |  24 ++--
>  11 files changed, 406 insertions(+), 115 deletions(-)
> 
_______________________________________________
Linux-nvdimm mailing list -- linux-nvdimm@lists.01.org
To unsubscribe send an email to linux-nvdimm-leave@lists.01.org