On Thu, 12 Mar 2015, Eric B Munson wrote:

> 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 adds a sysctl entry
> that will be used to allow root to enable compaction of unevictable
> pages.
> 
> 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.  When the
> compact_unevictable sysctl is 0, I cannot allocate hugepages after
> fragmenting memory.  When the value is set to 1, allocations succeed.
> 
> Signed-off-by: Eric B Munson <[email protected]>
> Cc: Vlastimil Babka <[email protected]>
> Cc: Thomas Gleixner <[email protected]>
> Cc: Christoph Lameter <[email protected]>
> Cc: Peter Zijlstra <[email protected]>
> Cc: Mel Gorman <[email protected]>
> Cc: David Rientjes <[email protected]>
> Cc: Rik van Riel <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> ---
> Changes from V3:
> Instead of removing the ISOLATE_UNEVICTABLE mode and checks, allow the
> sysadmin to control if compaction of unevictable pages is allowable.
> 
>  include/linux/compaction.h |    1 +
>  kernel/sysctl.c            |    7 +++++++
>  mm/compaction.c            |    3 +++
>  3 files changed, 11 insertions(+)
> 
> diff --git a/include/linux/compaction.h b/include/linux/compaction.h
> index a014559..9dd7e7c 100644
> --- a/include/linux/compaction.h
> +++ b/include/linux/compaction.h
> @@ -34,6 +34,7 @@ extern int sysctl_compaction_handler(struct ctl_table 
> *table, int write,
>  extern int sysctl_extfrag_threshold;
>  extern int sysctl_extfrag_handler(struct ctl_table *table, int write,
>                       void __user *buffer, size_t *length, loff_t *ppos);
> +extern int sysctl_compact_unevictable;
>  
>  extern int fragmentation_index(struct zone *zone, unsigned int order);
>  extern unsigned long try_to_compact_pages(gfp_t gfp_mask, unsigned int order,
> diff --git a/kernel/sysctl.c b/kernel/sysctl.c
> index 88ea2d6..cc1a678 100644
> --- a/kernel/sysctl.c
> +++ b/kernel/sysctl.c
> @@ -1313,6 +1313,13 @@ static struct ctl_table vm_table[] = {
>               .extra1         = &min_extfrag_threshold,
>               .extra2         = &max_extfrag_threshold,
>       },
> +     {
> +             .procname       = "compact_unevictable",
> +             .data           = &sysctl_compact_unevictable,
> +             .maxlen         = sizeof(int),
> +             .mode           = 0644,
> +             .proc_handler   = proc_dointvec,
> +     },
>  
>  #endif /* CONFIG_COMPACTION */
>       {
> diff --git a/mm/compaction.c b/mm/compaction.c
> index 8c0d945..b2c1e4e 100644
> --- a/mm/compaction.c
> +++ b/mm/compaction.c
> @@ -1046,6 +1046,8 @@ typedef enum {
>       ISOLATE_SUCCESS,        /* Pages isolated, migrate */
>  } isolate_migrate_t;
>  
> +int sysctl_compact_unevictable;
> +
>  /*
>   * Isolate all pages that can be migrated from the first suitable block,
>   * starting at the block pointed to by the migrate scanner pfn within
> @@ -1057,6 +1059,7 @@ static isolate_migrate_t isolate_migratepages(struct 
> zone *zone,
>       unsigned long low_pfn, end_pfn;
>       struct page *page;
>       const isolate_mode_t isolate_mode =
> +             (sysctl_compact_unevictable ? ISOLATE_UNEVICTABLE: 0) |

Sorry, missed the space following the :, if this idea is acceptable, I
will send a patch with the correct whitespace.

>               (cc->mode == MIGRATE_ASYNC ? ISOLATE_ASYNC_MIGRATE : 0);
>  
>       /*
> -- 
> 1.7.9.5
> 

Attachment: signature.asc
Description: Digital signature

Reply via email to