LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: David Hildenbrand <david@redhat.com>
To: Mike Rapoport <rppt@kernel.org>, linux-mm@kvack.org
Cc: Andrew Morton <akpm@linux-foundation.org>,
	Andy Lutomirski <luto@kernel.org>,
	Dave Hansen <dave.hansen@linux.intel.com>,
	Ira Weiny <ira.weiny@intel.com>,
	Kees Cook <keescook@chromium.org>,
	Mike Rapoport <rppt@linux.ibm.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Rick Edgecombe <rick.p.edgecombe@intel.com>,
	Vlastimil Babka <vbabka@suse.cz>,
	x86@kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [RFC PATCH 3/4] mm/page_alloc: introduce __GFP_PTE_MAPPED flag to allocate pte-mapped pages
Date: Wed, 25 Aug 2021 10:43:14 +0200	[thread overview]
Message-ID: <0a6341d6-c5be-ade3-9fb9-66cc8ebe1619@redhat.com> (raw)
In-Reply-To: <20210823132513.15836-4-rppt@kernel.org>

On 23.08.21 15:25, Mike Rapoport wrote:
> From: Mike Rapoport <rppt@linux.ibm.com>
> 
> When __GFP_PTE_MAPPED flag is passed to an allocation request of order 0,
> the allocated page will be mapped at PTE level in the direct map.
> 
> To reduce the direct map fragmentation, maintain a cache of 4K pages that
> are already mapped at PTE level in the direct map. Whenever the cache
> should be replenished, try to allocate 2M page and split it to 4K pages
> to localize shutter of the direct map. If the allocation of 2M page fails,
> fallback to a single page allocation at expense of the direct map
> fragmentation.
> 
> The cache registers a shrinker that releases free pages from the cache to
> the page allocator.
> 
> The __GFP_PTE_MAPPED and caching of 4K pages are enabled only if an
> architecture selects ARCH_WANTS_PTE_MAPPED_CACHE in its Kconfig.
> 
> [
> cache management are mostly copied from
> https://lore.kernel.org/lkml/20210505003032.489164-4-rick.p.edgecombe@intel.com/
> ]
> 
> Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
> ---
>   arch/Kconfig                    |   8 +
>   arch/x86/Kconfig                |   1 +
>   include/linux/gfp.h             |  11 +-
>   include/linux/mm.h              |   2 +
>   include/linux/pageblock-flags.h |  26 ++++
>   init/main.c                     |   1 +
>   mm/internal.h                   |   3 +-
>   mm/page_alloc.c                 | 261 +++++++++++++++++++++++++++++++-
>   8 files changed, 309 insertions(+), 4 deletions(-)
> 
> diff --git a/arch/Kconfig b/arch/Kconfig
> index 129df498a8e1..2db95331201b 100644
> --- a/arch/Kconfig
> +++ b/arch/Kconfig
> @@ -243,6 +243,14 @@ config ARCH_HAS_SET_MEMORY
>   config ARCH_HAS_SET_DIRECT_MAP
>   	bool

[...]

> +static int __pte_mapped_cache_init(struct pte_mapped_cache *cache)
> +{
> +	int err;
> +
> +	err = list_lru_init(&cache->lru);
> +	if (err)
> +		return err;
> +
> +	cache->shrinker.count_objects = pte_mapped_cache_shrink_count;
> +	cache->shrinker.scan_objects = pte_mapped_cache_shrink_scan;
> +	cache->shrinker.seeks = DEFAULT_SEEKS;
> +	cache->shrinker.flags = SHRINKER_NUMA_AWARE;
> +
> +	err = register_shrinker(&cache->shrinker);
> +	if (err)
> +		goto err_list_lru_destroy;

With a shrinker in place, it really does somewhat feel like this should 
be a cache outside of the buddy. Or at least moved outside of 
page_alloc.c with a clean interface to work with the buddy.

But I only had a quick glimpse over this patch.

-- 
Thanks,

David / dhildenb


  parent reply	other threads:[~2021-08-25  8:43 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-23 13:25 [RFC PATCH 0/4] mm/page_alloc: cache pte-mapped allocations Mike Rapoport
2021-08-23 13:25 ` [RFC PATCH 1/4] list: Support getting most recent element in list_lru Mike Rapoport
2021-08-23 13:25 ` [RFC PATCH 2/4] list: Support list head not in object for list_lru Mike Rapoport
2021-08-23 13:25 ` [RFC PATCH 3/4] mm/page_alloc: introduce __GFP_PTE_MAPPED flag to allocate pte-mapped pages Mike Rapoport
2021-08-23 20:29   ` Edgecombe, Rick P
2021-08-24 13:02     ` Mike Rapoport
2021-08-24 16:38       ` Edgecombe, Rick P
2021-08-24 16:54         ` Mike Rapoport
2021-08-24 17:23           ` Edgecombe, Rick P
2021-08-24 17:37             ` Mike Rapoport
2021-08-24 16:12   ` Vlastimil Babka
2021-08-25  8:43   ` David Hildenbrand [this message]
2021-08-23 13:25 ` [RFC PATCH 4/4] x86/mm: write protect (most) page tables Mike Rapoport
2021-08-23 20:08   ` Edgecombe, Rick P
2021-08-23 23:50   ` Dave Hansen
2021-08-24  3:34     ` Andy Lutomirski
2021-08-25 14:59       ` Dave Hansen
2021-08-24 13:32     ` Mike Rapoport
2021-08-25  8:38     ` David Hildenbrand
2021-08-26  8:02     ` Mike Rapoport
2021-08-26  9:01       ` Vlastimil Babka
     [not found]   ` <FB6C09CD-9CEA-4FE8-B179-98DB63EBDD68@gmail.com>
2021-08-24  5:34     ` Nadav Amit
2021-08-24 13:36       ` Mike Rapoport
2021-08-23 20:02 ` [RFC PATCH 0/4] mm/page_alloc: cache pte-mapped allocations Edgecombe, Rick P
2021-08-24 13:03   ` Mike Rapoport
2021-08-24 16:09 ` Vlastimil Babka
2021-08-29  7:06   ` Mike Rapoport

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=0a6341d6-c5be-ade3-9fb9-66cc8ebe1619@redhat.com \
    --to=david@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=dave.hansen@linux.intel.com \
    --cc=ira.weiny@intel.com \
    --cc=keescook@chromium.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=luto@kernel.org \
    --cc=peterz@infradead.org \
    --cc=rick.p.edgecombe@intel.com \
    --cc=rppt@kernel.org \
    --cc=rppt@linux.ibm.com \
    --cc=vbabka@suse.cz \
    --cc=x86@kernel.org \
    --subject='Re: [RFC PATCH 3/4] mm/page_alloc: introduce __GFP_PTE_MAPPED flag to allocate pte-mapped pages' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).