On (24/06/04 17:53), Yosry Ahmed wrote: > Zswap creates multiple zpools to improve concurrency. Each zsmalloc > zpool creates its own 'zs_handle' and 'zspage' slab caches. Currently we > end up with 32 slab caches of each type. > > Since each slab cache holds some free objects, we end up with a lot of > free objects distributed among the separate zpool caches. Slab caches > are designed to handle concurrent allocations by using percpu > structures, so having a single instance of each cache should be enough, > and avoids wasting more memory than needed due to fragmentation. > > Additionally, having more slab caches than needed unnecessarily slows > down code paths that iterate slab_caches. > > In the results reported by Eric in [1], the amount of unused slab memory > in these caches goes down from 242808 bytes to 29216 bytes (-88%). This > is calculated by (num_objs - active_objs) * objsize for each 'zs_handle' > and 'zspage' cache. Although this patch did not help with the allocation > failure reported by Eric with zswap + zsmalloc, I think it is still > worth merging on its own. > > [1]https://lore.kernel.org/lkml/20240604134458.3ae4396a@yea/ > > Signed-off-by: Yosry Ahmed <yosryahmed@xxxxxxxxxx> Makes perfect sense, thanks. Reviewed-by: Sergey Senozhatsky <senozhatsky@xxxxxxxxxxxx>
