Re: [PATCH v4 3/8] mm: vmscan: make memcg slab shrink lockless

[Vlastimil Babka <vbabka@suse.cz>]

On 3/7/23 07:56, Qi Zheng wrote:
> Like global slab shrink, this commit also uses SRCU to make
> memcg slab shrink lockless.
> 
> We can reproduce the down_read_trylock() hotspot through the
> following script:
> 
> ```
> 
> DIR="/root/shrinker/memcg/mnt"
> 
> do_create()
> {
>     mkdir -p /sys/fs/cgroup/memory/test
>     mkdir -p /sys/fs/cgroup/perf_event/test
>     echo 4G > /sys/fs/cgroup/memory/test/memory.limit_in_bytes
>     for i in `seq 0 $1`;
>     do
>         mkdir -p /sys/fs/cgroup/memory/test/$i;
>         echo $$ > /sys/fs/cgroup/memory/test/$i/cgroup.procs;
>         echo $$ > /sys/fs/cgroup/perf_event/test/cgroup.procs;
>         mkdir -p $DIR/$i;
>     done
> }
> 
> do_mount()
> {
>     for i in `seq $1 $2`;
>     do
>         mount -t tmpfs $i $DIR/$i;
>     done
> }
> 
> do_touch()
> {
>     for i in `seq $1 $2`;
>     do
>         echo $$ > /sys/fs/cgroup/memory/test/$i/cgroup.procs;
>         echo $$ > /sys/fs/cgroup/perf_event/test/cgroup.procs;
>             dd if=/dev/zero of=$DIR/$i/file$i bs=1M count=1 &
>     done
> }
> 
> case "$1" in
>   touch)
>     do_touch $2 $3
>     ;;
>   test)
>       do_create 4000
>     do_mount 0 4000
>     do_touch 0 3000
>     ;;
>   *)
>     exit 1
>     ;;
> esac
> ```
> 
> Save the above script, then run test and touch commands.
> Then we can use the following perf command to view hotspots:
> 
> perf top -U -F 999
> 
> 1) Before applying this patchset:
> 
>   32.31%  [kernel]           [k] down_read_trylock
>   19.40%  [kernel]           [k] pv_native_safe_halt
>   16.24%  [kernel]           [k] up_read
>   15.70%  [kernel]           [k] shrink_slab
>    4.69%  [kernel]           [k] _find_next_bit
>    2.62%  [kernel]           [k] shrink_node
>    1.78%  [kernel]           [k] shrink_lruvec
>    0.76%  [kernel]           [k] do_shrink_slab
> 
> 2) After applying this patchset:
> 
>   27.83%  [kernel]           [k] _find_next_bit
>   16.97%  [kernel]           [k] shrink_slab
>   15.82%  [kernel]           [k] pv_native_safe_halt
>    9.58%  [kernel]           [k] shrink_node
>    8.31%  [kernel]           [k] shrink_lruvec
>    5.64%  [kernel]           [k] do_shrink_slab
>    3.88%  [kernel]           [k] mem_cgroup_iter
> 
> At the same time, we use the following perf command to capture
> IPC information:
> 
> perf stat -e cycles,instructions -G test -a --repeat 5 -- sleep 10
> 
> 1) Before applying this patchset:
> 
>  Performance counter stats for 'system wide' (5 runs):
> 
>       454187219766      cycles                    test                    ( +-  1.84% )
>        78896433101      instructions              test #    0.17  insn per cycle           ( +-  0.44% )
> 
>         10.0020430 +- 0.0000366 seconds time elapsed  ( +-  0.00% )
> 
> 2) After applying this patchset:
> 
>  Performance counter stats for 'system wide' (5 runs):
> 
>       841954709443      cycles                    test                    ( +- 15.80% )  (98.69%)
>       527258677936      instructions              test #    0.63  insn per cycle           ( +- 15.11% )  (98.68%)
> 
>           10.01064 +- 0.00831 seconds time elapsed  ( +-  0.08% )
> 
> We can see that IPC drops very seriously when calling
> down_read_trylock() at high frequency. After using SRCU,
> the IPC is at a normal level.

The interpretation looks somewhat weird to me. I'd say the workload is
stalled a lot as it fails the trylock (there might be some optimistic
spinning perhaps) and then goes to sleep. See how "pv_native_safe_halt" is
also more prominent in before. And because of that sleeping, there's less
instructions executed in the same amount of cycles (as it's a system wide
collection, otherwise it wouldn't be collecting the sleeping processes).

> 
> Signed-off-by: Qi Zheng <zhengqi.arch@bytedance.com>

Other than that:

Acked-by: Vlastimil Babka <Vbabka@suse.cz>

A small thing below:

> ---
>  mm/vmscan.c | 46 +++++++++++++++++++++++++++-------------------
>  1 file changed, 27 insertions(+), 19 deletions(-)
> 
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 8515ac40bcaf..1de9bc3e5aa2 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -57,6 +57,7 @@
>  #include <linux/khugepaged.h>
>  #include <linux/rculist_nulls.h>
>  #include <linux/random.h>
> +#include <linux/srcu.h>

I guess this should have been in patch 2/8 already? It may work accidentaly
because some other header pulls it transitively...