Re: [PATCH v2] zsmalloc: merge size_class to reduce fragmentation

[Joonsoo Kim]

On Wed, Sep 24, 2014 at 05:12:20PM +0900, Minchan Kim wrote:
> Hi Joonsoo,
> 
> On Wed, Sep 24, 2014 at 03:03:46PM +0900, Joonsoo Kim wrote:
> > zsmalloc has many size_classes to reduce fragmentation and they are
> > in 16 bytes unit, for example, 16, 32, 48, etc., if PAGE_SIZE is 4096.
> > And, zsmalloc has constraint that each zspage has 4 pages at maximum.
> > 
> > In this situation, we can see interesting aspect.
> > Let's think about size_class for 1488, 1472, ..., 1376.
> > To prevent external fragmentation, they uses 4 pages per zspage and
> > so all they can contain 11 objects at maximum.
> > 
> > 16384 (4096 * 4) = 1488 * 11 + remains
> > 16384 (4096 * 4) = 1472 * 11 + remains
> > 16384 (4096 * 4) = ...
> > 16384 (4096 * 4) = 1376 * 11 + remains
> > 
> > It means that they have same characteristics and classification between
> > them isn't needed. If we use one size_class for them, we can reduce
> > fragementation and save some memory. For this purpose, this patch
> > implement size_class merging. If there is size_class that have
> > same pages_per_zspage and same number of objects per zspage with previous
> > size_class, we don't create and use new size_class. Instead, we use
> > previous, same characteristic size_class. With this way, above example
> > sizes (1488, 1472, ..., 1376) use just one size_class so we can get much
> > more memory utilization.
> > 
> > Below is result of my simple test.
> > 
> > TEST ENV: EXT4 on zram, mount with discard option
> > WORKLOAD: untar kernel source code, remove directory in descending order
> > in size. (drivers arch fs sound include net Documentation firmware
> > kernel tools)
> > 
> > Each line represents orig_data_size, compr_data_size, mem_used_total,
> > fragmentation overhead (mem_used - compr_data_size) and overhead ratio
> > (overhead to compr_data_size), respectively, after untar and remove
> > operation is executed.
> > 
> > * untar-nomerge.out
> > 
> > orig_size compr_size used_size overhead overhead_ratio
> > 525.88MB 199.16MB 210.23MB  11.08MB 5.56%
> > 288.32MB  97.43MB 105.63MB   8.20MB 8.41%
> > 177.32MB  61.12MB  69.40MB   8.28MB 13.55%
> > 146.47MB  47.32MB  56.10MB   8.78MB 18.55%
> > 124.16MB  38.85MB  48.41MB   9.55MB 24.58%
> > 103.93MB  31.68MB  40.93MB   9.25MB 29.21%
> >  84.34MB  22.86MB  32.72MB   9.86MB 43.13%
> >  66.87MB  14.83MB  23.83MB   9.00MB 60.70%
> >  60.67MB  11.11MB  18.60MB   7.49MB 67.48%
> >  55.86MB   8.83MB  16.61MB   7.77MB 88.03%
> >  53.32MB   8.01MB  15.32MB   7.31MB 91.24%
> > 
> > * untar-merge.out
> > 
> > orig_size compr_size used_size overhead overhead_ratio
> > 526.23MB 199.18MB 209.81MB  10.64MB 5.34%
> > 288.68MB  97.45MB 104.08MB   6.63MB 6.80%
> > 177.68MB  61.14MB  66.93MB   5.79MB 9.47%
> > 146.83MB  47.34MB  52.79MB   5.45MB 11.51%
> > 124.52MB  38.87MB  44.30MB   5.43MB 13.96%
> > 104.29MB  31.70MB  36.83MB   5.13MB 16.19%
> >  84.70MB  22.88MB  27.92MB   5.04MB 22.04%
> >  67.11MB  14.83MB  19.26MB   4.43MB 29.86%
> >  60.82MB  11.10MB  14.90MB   3.79MB 34.17%
> >  55.90MB   8.82MB  12.61MB   3.79MB 42.97%
> >  53.32MB   8.01MB  11.73MB   3.73MB 46.53%
> > 
> > As you can see above result, merged one has better utilization (overhead
> > ratio, 5th column) and uses less memory (mem_used_total, 3rd column).
> 
> What a nice patch!

Thank you. :)

> Below is minor but want to fix.
> 
> > 
> > Changed from v1:
> > - More commit description about what to do in this patch.
> > - Remove nr_obj in size_class, because it isn't need after initialization.
> > - Rename __size_class to size_class, size_class to merged_size_class.
> > - Add code comment for merged_size_class of struct zs_pool.
> > - Add code comment how merging works in zs_create_pool().
> > 
> > Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx>
> > ---
> >  mm/zsmalloc.c |   57 ++++++++++++++++++++++++++++++++++++++++++++++++---------
> >  1 file changed, 48 insertions(+), 9 deletions(-)
> > 
> > diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
> > index c4a9157..586c19d 100644
> > --- a/mm/zsmalloc.c
> > +++ b/mm/zsmalloc.c
> > @@ -214,6 +214,11 @@ struct link_free {
> >  };
> >  
> >  struct zs_pool {
> > +	/*
> > +	 * Each merge_size_class is pointing to one of size_class that have
> > +	 * same characteristics. See zs_create_pool() for more information.
> > +	 */
> > +	struct size_class *merged_size_class[ZS_SIZE_CLASSES];
> >  	struct size_class size_class[ZS_SIZE_CLASSES];
> 
> I don't like the naming.
> My preference is to keep size_class as it is. Instead, please use
> __size_class or raw_size_class for internal size class.

I think that Dan's proposal in another mail is better that is just
allocating real size_class by kmalloc and removing this ambiguity. It can
also save memory. I will go that way.

> >  
> >  	gfp_t flags;	/* allocation flags used when growing pool */
> > @@ -468,7 +473,7 @@ static enum fullness_group fix_fullness_group(struct zs_pool *pool,
> >  	if (newfg == currfg)
> >  		goto out;
> >  
> > -	class = &pool->size_class[class_idx];
> > +	class = pool->merged_size_class[class_idx];
> >  	remove_zspage(page, class, currfg);
> >  	insert_zspage(page, class, newfg);
> >  	set_zspage_mapping(page, class_idx, newfg);
> > @@ -929,6 +934,22 @@ fail:
> >  	return notifier_to_errno(ret);
> >  }
> >  
> > +static unsigned int objs_per_zspage(struct size_class *class)
> 
> It seems to be the number of objects in the class but it doesn't.
> Please use get_maxobj_per_zspage for consistent with get_pages_per_zspage.

Will do.

Thanks.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/