Re: [PATCHv2 1/2] zsmalloc: introduce zs_huge_object() function

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Some more nitpicks :)

On Sat, Feb 10, 2018 at 05:23:21PM +0900, Sergey Senozhatsky wrote:
> Not every object can be share its zspage with other objects, e.g.
> when the object is as big as zspage or nearly as big a zspage.
> For such objects zsmalloc has a so called huge class - every object
> which belongs to huge class consumes the entire zspage (which
> consists of a physical page). On x86_64, PAGE_SHIFT 12 box, the
> first non-huge class size is 3264, so starting down from size 3264,
> objects can share page(-s) and thus minimize memory wastage.
> 
> ZRAM, however, has its own statically defined watermark for huge
> objects - "3 * PAGE_SIZE / 4 = 3072", and forcibly stores every
> object larger than this watermark (3072) as a PAGE_SIZE object,
> in other words, to a huge class, while zsmalloc can keep some of
> those objects in non-huge classes. This results in increased
> memory consumption.
> 
> zsmalloc knows better if the object is huge or not. Introduce
> zs_huge_object() function which tells if the given object can be
> stored in one of non-huge classes or not. This will let us to drop
> ZRAM's huge object watermark and fully rely on zsmalloc when we
> decide if the object is huge.
> 
> Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@xxxxxxxxx>
> ---
>  include/linux/zsmalloc.h |  2 ++
>  mm/zsmalloc.c            | 26 ++++++++++++++++++++++++++
>  2 files changed, 28 insertions(+)
> 
> diff --git a/include/linux/zsmalloc.h b/include/linux/zsmalloc.h
> index 57a8e98f2708..9a1baf673cc1 100644
> --- a/include/linux/zsmalloc.h
> +++ b/include/linux/zsmalloc.h
> @@ -47,6 +47,8 @@ void zs_destroy_pool(struct zs_pool *pool);
>  unsigned long zs_malloc(struct zs_pool *pool, size_t size, gfp_t flags);
>  void zs_free(struct zs_pool *pool, unsigned long obj);
> 
> +bool zs_huge_object(size_t sz);
> +
>  void *zs_map_object(struct zs_pool *pool, unsigned long handle,
>  			enum zs_mapmode mm);
>  void zs_unmap_object(struct zs_pool *pool, unsigned long handle);
> diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
> index c3013505c305..922180183ca3 100644
> --- a/mm/zsmalloc.c
> +++ b/mm/zsmalloc.c
> @@ -192,6 +192,7 @@ static struct vfsmount *zsmalloc_mnt;
>   * (see: fix_fullness_group())
>   */
>  static const int fullness_threshold_frac = 4;
> +static size_t zs_huge_class_size;
> 
>  struct size_class {
>  	spinlock_t lock;
> @@ -1417,6 +1418,28 @@ void zs_unmap_object(struct zs_pool *pool, unsigned long handle)
>  }
>  EXPORT_SYMBOL_GPL(zs_unmap_object);
> 
> +/**
> + * zs_huge_object() - Test if a compressed object's size is too big for normal
> + *                    zspool classes and it shall be stored in a huge class.

I think "is should be stored" is more appropriate

> + * @sz: Size of the compressed object (in bytes).
> + *
> + * The function checks if the object's size falls into huge_class
> + * area. We must take handle size into account and test the actual
> + * size we are going to use, because zs_malloc() unconditionally
> + * adds %ZS_HANDLE_SIZE before it performs %size_class lookup.

                                            ^ &size_class ;-)

> + *
> + * Context: Any context.
> + *
> + * Return:
> + * * true  - The object's size is too big, it will be stored in a huge class.
> + * * false - The object will be store in normal zspool classes.
> + */
> +bool zs_huge_object(size_t sz)
> +{
> +	return sz + ZS_HANDLE_SIZE >= zs_huge_class_size;
> +}
> +EXPORT_SYMBOL_GPL(zs_huge_object);
> +
>  static unsigned long obj_malloc(struct size_class *class,
>  				struct zspage *zspage, unsigned long handle)
>  {
> @@ -2404,6 +2427,9 @@ struct zs_pool *zs_create_pool(const char *name)
>  			INIT_LIST_HEAD(&class->fullness_list[fullness]);
> 
>  		prev_class = class;
> +		if (pages_per_zspage == 1 && objs_per_zspage == 1
> +				&& !zs_huge_class_size)
> +			zs_huge_class_size = size;
>  	}
> 
>  	/* debug only, don't abort if it fails */
> -- 
> 2.16.1
> 

-- 
Sincerely yours,
Mike.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>



[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux