LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
* [PATCH V2] Allow compaction of unevictable pages
@ 2015-03-09 17:12 Eric B Munson
  2015-03-09 19:57 ` David Rientjes
  0 siblings, 1 reply; 3+ messages in thread
From: Eric B Munson @ 2015-03-09 17:12 UTC (permalink / raw)
  To: Andrew Morton
  Cc: Eric B Munson, Vlastimil Babka, Thomas Gleixner,
	Christoph Lameter, Peter Zijlstra, Mel Gorman, David Rientjes,
	linux-mm, linux-kernel

Currently, pages which are marked as unevictable are protected from
compaction, but not from other types of migration.  The mlock
desctription does not promise that all page faults will be avoided, only
major ones so this protection is not necessary.  This extra protection
can cause problems for applications that are using mlock to avoid
swapping pages out, but require order > 0 allocations to continue to
succeed in a fragmented environment.  This patch removes the
ISOLATE_UNEVICTABLE mode and the check for it in __isolate_lru_page().

To illustrate this problem I wrote a quick test program that mmaps a
large number of 1MB files filled with random data.  These maps are
created locked and read only.  Then every other mmap is unmapped and I
attempt to allocate huge pages to the static huge page pool.  Without
this patch I am unable to allocate any huge pages after  fragmenting
memory.  With it, I can allocate almost all the space freed by unmapping
as huge pages.

Signed-off-by: Eric B Munson <emunson@akamai.com>

Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Christoph Lameter <cl@linux.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: David Rientjes <rientjes@google.com>
Cc: linux-mm@kvack.org
Cc: linux-kernel@vger.kernel.org

---
 include/linux/mmzone.h |    2 --
 mm/compaction.c        |    3 +--
 mm/vmscan.c            |    4 ----
 3 files changed, 1 insertion(+), 8 deletions(-)

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index f279d9c..599fb01 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -232,8 +232,6 @@ struct lruvec {
 #define ISOLATE_UNMAPPED	((__force isolate_mode_t)0x2)
 /* Isolate for asynchronous migration */
 #define ISOLATE_ASYNC_MIGRATE	((__force isolate_mode_t)0x4)
-/* Isolate unevictable pages */
-#define ISOLATE_UNEVICTABLE	((__force isolate_mode_t)0x8)
 
 /* LRU Isolation modes. */
 typedef unsigned __bitwise__ isolate_mode_t;
diff --git a/mm/compaction.c b/mm/compaction.c
index 8c0d945..4a8ea87 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -872,8 +872,7 @@ isolate_migratepages_range(struct compact_control *cc, unsigned long start_pfn,
 		if (!pageblock_pfn_to_page(pfn, block_end_pfn, cc->zone))
 			continue;
 
-		pfn = isolate_migratepages_block(cc, pfn, block_end_pfn,
-							ISOLATE_UNEVICTABLE);
+		pfn = isolate_migratepages_block(cc, pfn, block_end_pfn, 0);
 
 		/*
 		 * In case of fatal failure, release everything that might
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 5e8eadd..3b2a444 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1234,10 +1234,6 @@ int __isolate_lru_page(struct page *page, isolate_mode_t mode)
 	if (!PageLRU(page))
 		return ret;
 
-	/* Compaction should not handle unevictable pages but CMA can do so */
-	if (PageUnevictable(page) && !(mode & ISOLATE_UNEVICTABLE))
-		return ret;
-
 	ret = -EBUSY;
 
 	/*
-- 
1.7.9.5


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH V2] Allow compaction of unevictable pages
  2015-03-09 17:12 [PATCH V2] Allow compaction of unevictable pages Eric B Munson
@ 2015-03-09 19:57 ` David Rientjes
  2015-03-09 20:39   ` Eric B Munson
  0 siblings, 1 reply; 3+ messages in thread
From: David Rientjes @ 2015-03-09 19:57 UTC (permalink / raw)
  To: Eric B Munson
  Cc: Andrew Morton, Vlastimil Babka, Thomas Gleixner,
	Christoph Lameter, Peter Zijlstra, Mel Gorman, linux-mm,
	linux-kernel

On Mon, 9 Mar 2015, Eric B Munson wrote:

> diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
> index f279d9c..599fb01 100644
> --- a/include/linux/mmzone.h
> +++ b/include/linux/mmzone.h
> @@ -232,8 +232,6 @@ struct lruvec {
>  #define ISOLATE_UNMAPPED	((__force isolate_mode_t)0x2)
>  /* Isolate for asynchronous migration */
>  #define ISOLATE_ASYNC_MIGRATE	((__force isolate_mode_t)0x4)
> -/* Isolate unevictable pages */
> -#define ISOLATE_UNEVICTABLE	((__force isolate_mode_t)0x8)
>  
>  /* LRU Isolation modes. */
>  typedef unsigned __bitwise__ isolate_mode_t;
> diff --git a/mm/compaction.c b/mm/compaction.c
> index 8c0d945..4a8ea87 100644
> --- a/mm/compaction.c
> +++ b/mm/compaction.c
> @@ -872,8 +872,7 @@ isolate_migratepages_range(struct compact_control *cc, unsigned long start_pfn,
>  		if (!pageblock_pfn_to_page(pfn, block_end_pfn, cc->zone))
>  			continue;
>  
> -		pfn = isolate_migratepages_block(cc, pfn, block_end_pfn,
> -							ISOLATE_UNEVICTABLE);
> +		pfn = isolate_migratepages_block(cc, pfn, block_end_pfn, 0);
>  
>  		/*
>  		 * In case of fatal failure, release everything that might
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 5e8eadd..3b2a444 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -1234,10 +1234,6 @@ int __isolate_lru_page(struct page *page, isolate_mode_t mode)
>  	if (!PageLRU(page))
>  		return ret;
>  
> -	/* Compaction should not handle unevictable pages but CMA can do so */
> -	if (PageUnevictable(page) && !(mode & ISOLATE_UNEVICTABLE))
> -		return ret;
> -
>  	ret = -EBUSY;
>  
>  	/*

Looks better!

I think there's one more cleanup we can do now thanks to your patch: 
dropping the isolate_mode_t formal from isolate_migratepages_block() 
entirely since that function can now just do

	const isolate_mode_t isolate_mode =
		(cc->mode == MIGRATE_ASYNC ? ISOLATE_ASYNC_MIGRATE : 0);

since we already pass in the struct compact_control and isolate_mode only 
depends on MIGRATE_ASYNC or not.

If you'd like to fold that change into this patch because it's logically 
allowed by it, feel free to add my enthusiastic

	Acked-by: David Rientjes <rientjes@google.com>

Otherwise, I'll just send a change on top of it if you don't have time.

Thanks!

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH V2] Allow compaction of unevictable pages
  2015-03-09 19:57 ` David Rientjes
@ 2015-03-09 20:39   ` Eric B Munson
  0 siblings, 0 replies; 3+ messages in thread
From: Eric B Munson @ 2015-03-09 20:39 UTC (permalink / raw)
  To: David Rientjes
  Cc: Andrew Morton, Vlastimil Babka, Thomas Gleixner,
	Christoph Lameter, Peter Zijlstra, Mel Gorman, linux-mm,
	linux-kernel

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 03/09/2015 03:57 PM, David Rientjes wrote:
> On Mon, 9 Mar 2015, Eric B Munson wrote:
> 
>> diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h 
>> index f279d9c..599fb01 100644 --- a/include/linux/mmzone.h +++
>> b/include/linux/mmzone.h @@ -232,8 +232,6 @@ struct lruvec { 
>> #define ISOLATE_UNMAPPED	((__force isolate_mode_t)0x2) /* Isolate
>> for asynchronous migration */ #define ISOLATE_ASYNC_MIGRATE
>> ((__force isolate_mode_t)0x4) -/* Isolate unevictable pages */ 
>> -#define ISOLATE_UNEVICTABLE	((__force isolate_mode_t)0x8)
>> 
>> /* LRU Isolation modes. */ typedef unsigned __bitwise__
>> isolate_mode_t; diff --git a/mm/compaction.c b/mm/compaction.c 
>> index 8c0d945..4a8ea87 100644 --- a/mm/compaction.c +++
>> b/mm/compaction.c @@ -872,8 +872,7 @@
>> isolate_migratepages_range(struct compact_control *cc, unsigned
>> long start_pfn, if (!pageblock_pfn_to_page(pfn, block_end_pfn,
>> cc->zone)) continue;
>> 
>> -		pfn = isolate_migratepages_block(cc, pfn, block_end_pfn, -
>> ISOLATE_UNEVICTABLE); +		pfn = isolate_migratepages_block(cc,
>> pfn, block_end_pfn, 0);
>> 
>> /* * In case of fatal failure, release everything that might diff
>> --git a/mm/vmscan.c b/mm/vmscan.c index 5e8eadd..3b2a444 100644 
>> --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1234,10 +1234,6 @@ int
>> __isolate_lru_page(struct page *page, isolate_mode_t mode) if
>> (!PageLRU(page)) return ret;
>> 
>> -	/* Compaction should not handle unevictable pages but CMA can
>> do so */ -	if (PageUnevictable(page) && !(mode &
>> ISOLATE_UNEVICTABLE)) -		return ret; - ret = -EBUSY;
>> 
>> /*
> 
> Looks better!
> 
> I think there's one more cleanup we can do now thanks to your
> patch: dropping the isolate_mode_t formal from
> isolate_migratepages_block() entirely since that function can now
> just do
> 
> const isolate_mode_t isolate_mode = (cc->mode == MIGRATE_ASYNC ?
> ISOLATE_ASYNC_MIGRATE : 0);
> 
> since we already pass in the struct compact_control and
> isolate_mode only depends on MIGRATE_ASYNC or not.
> 
> If you'd like to fold that change into this patch because it's
> logically allowed by it, feel free to add my enthusiastic
> 
> Acked-by: David Rientjes <rientjes@google.com>
> 
> Otherwise, I'll just send a change on top of it if you don't have
> time.

I'll V3 out shortly with that change.

Thanks for the review.

Eric
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
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=suBI
-----END PGP SIGNATURE-----

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2015-03-09 20:39 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-03-09 17:12 [PATCH V2] Allow compaction of unevictable pages Eric B Munson
2015-03-09 19:57 ` David Rientjes
2015-03-09 20:39   ` Eric B Munson

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).