On (22/11/28 11:16), Nhat Pham wrote: > Currently, zsmalloc has a hierarchy of locks, which includes a > pool-level migrate_lock, and a lock for each size class. We have to > obtain both locks in the hotpath in most cases anyway, except for > zs_malloc. This exception will no longer exist when we introduce a LRU > into the zs_pool for the new writeback functionality - we will need to > obtain a pool-level lock to synchronize LRU handling even in zs_malloc. > > In preparation for zsmalloc writeback, consolidate these locks into a > single pool-level lock, which drastically reduces the complexity of > synchronization in zsmalloc. > > We have also benchmarked the lock consolidation to see the performance > effect of this change on zram. > > First, we ran a synthetic FS workload on a server machine with 36 cores > (same machine for all runs), using > > fs_mark -d ../zram1mnt -s 100000 -n 2500 -t 32 -k > > before and after for btrfs and ext4 on zram (FS usage is 80%). > > Here is the result (unit is file/second): > > With lock consolidation (btrfs): > Average: 13520.2, Median: 13531.0, Stddev: 137.5961482019028 > > Without lock consolidation (btrfs): > Average: 13487.2, Median: 13575.0, Stddev: 309.08283679298665 > > With lock consolidation (ext4): > Average: 16824.4, Median: 16839.0, Stddev: 89.97388510006668 > > Without lock consolidation (ext4) > Average: 16958.0, Median: 16986.0, Stddev: 194.7370021336469 > > As you can see, we observe a 0.3% regression for btrfs, and a 0.9% > regression for ext4. This is a small, barely measurable difference in my > opinion. > > For a more realistic scenario, we also tries building the kernel on zram. > Here is the time it takes (in seconds): > > With lock consolidation (btrfs): > real > Average: 319.6, Median: 320.0, Stddev: 0.8944271909999159 > user > Average: 6894.2, Median: 6895.0, Stddev: 25.528415540334656 > sys > Average: 521.4, Median: 522.0, Stddev: 1.51657508881031 > > Without lock consolidation (btrfs): > real > Average: 319.8, Median: 320.0, Stddev: 0.8366600265340756 > user > Average: 6896.6, Median: 6899.0, Stddev: 16.04057355583023 > sys > Average: 520.6, Median: 521.0, Stddev: 1.140175425099138 > > With lock consolidation (ext4): > real > Average: 320.0, Median: 319.0, Stddev: 1.4142135623730951 > user > Average: 6896.8, Median: 6878.0, Stddev: 28.621670111997307 > sys > Average: 521.2, Median: 521.0, Stddev: 1.7888543819998317 > > Without lock consolidation (ext4) > real > Average: 319.6, Median: 319.0, Stddev: 0.8944271909999159 > user > Average: 6886.2, Median: 6887.0, Stddev: 16.93221781102523 > sys > Average: 520.4, Median: 520.0, Stddev: 1.140175425099138 > > The difference is entirely within the noise of a typical run on zram. This > hardly justifies the complexity of maintaining both the pool lock and > the class lock. In fact, for writeback, we would need to introduce yet > another lock to prevent data races on the pool's LRU, further > complicating the lock handling logic. IMHO, it is just better to > collapse all of these into a single pool-level lock. > > Suggested-by: Johannes Weiner <hannes@xxxxxxxxxxx> > Signed-off-by: Nhat Pham <nphamcs@xxxxxxxxx> > Acked-by: Minchan Kim <minchan@xxxxxxxxxx> > Acked-by: Johannes Weiner <hannes@xxxxxxxxxxx> Reviewed-by: Sergey Senozhatsky <senozhatsky@xxxxxxxxxxxx>
