Hi, All,
In page allocator, PCP (Per-CPU Pageset) is refilled and drained in
batches to increase page allocation throughput, reduce page
allocation/freeing latency per page, and reduce zone lock contention.
But too large batch size will cause too long maximal
allocation/freeing latency, which may punish arbitrary users. So the
default batch size is chosen carefully (in zone_batchsize(), the value
is 63 now) to avoid that.
In commit 3b12e7e97938 ("mm/page_alloc: scale the number of pages that
are batch freed"), the batch size will be scaled for large number of
page freeing to improve page freeing performance and reduce zone lock
contention. Similar optimization are advised for large number of
pages allocation too in [1].
[1] https://lore.kernel.org/linux-mm/20230714140710.5xbesq6xguhcbyvi@xxxxxxxxxxxxxxxxxxx/
To find out a suitable max batch scale factor (that is, max effective
batch size), we have done some tests and measurement on some machines
as follows.
A set of debug patches are implemented as follows,
- Set PCP high to be 2 * batch to reduce the effect of PCP high
- Disable free batch size scaling to get the raw performance.
- The code with zone lock held is extracted from rmqueue_bulk() and
free_pcppages_bulk() to 2 separate functions to make it easy to
measure the function run time with ftrace function_graph tracer.
- The batch size is hard coded to be 63 (default), 127, 255, 511,
1023, 2047, 4095.
Then will-it-scale/page_fault1 is used to generate the page
allocation/freeing workload. The page allocation/freeing throughput
(page/s) is measured via will-it-scale. The page allocation/freeing
average latency (alloc/free latency avg, in us) and allocation/freeing
latency at 99 percentile (alloc/free latency 99%, in us) are measured
with ftrace function_graph tracer.
The test results are as follows,
Sapphire Rapids Server
======================
Batch throughput free latency free latency alloc latency alloc latency
page/s avg / us 99% / us avg / us 99% / us
----- ---------- ------------ ------------ ------------- -------------
63 513633.4 2.33 3.57 2.67 6.83
127 517616.7 4.35 6.65 4.22 13.03
255 520822.8 8.29 13.32 7.52 25.24
511 524122.0 15.79 23.42 14.02 49.35
1023 525980.5 30.25 44.19 25.36 94.88
2047 526793.6 59.39 84.50 45.22 140.81
Ice Lake Server
===============
Batch throughput free latency free latency alloc latency alloc latency
page/s avg / us 99% / us avg / us 99% / us
----- ---------- ------------ ------------ ------------- -------------
63 620210.3 2.21 3.68 2.02 4.35
127 627003.0 4.09 6.86 3.51 8.28
255 630777.5 7.70 13.50 6.17 15.97
511 633651.5 14.85 22.62 11.66 31.08
1023 637071.1 28.55 42.02 20.81 54.36
2047 638089.7 56.54 84.06 39.28 91.68
Cascade Lake Server
===================
Batch throughput free latency free latency alloc latency alloc latency
page/s avg / us 99% / us avg / us 99% / us
----- ---------- ------------ ------------ ------------- -------------
63 404706.7 3.29 5.03 3.53 4.75
127 422475.2 6.12 9.09 6.36 8.76
255 411522.2 11.68 16.97 10.90 16.39
511 428124.1 22.54 31.28 19.86 32.25
1023 414718.4 43.39 62.52 40.00 66.33
2047 429848.7 86.64 120.34 71.14 106.08
Commet Lake Desktop
===================
Batch throughput free latency free latency alloc latency alloc latency
page/s avg / us 99% / us avg / us 99% / us
----- ---------- ------------ ------------ ------------- -------------
63 795183.13 2.18 3.55 2.03 3.05
127 803067.85 3.91 6.56 3.85 5.52
255 812771.10 7.35 10.80 7.14 10.20
511 817723.48 14.17 27.54 13.43 30.31
1023 818870.19 27.72 40.10 27.89 46.28
Coffee Lake Desktop
===================
Batch throughput free latency free latency alloc latency alloc latency
page/s avg / us 99% / us avg / us 99% / us
----- ---------- ------------ ------------ ------------- -------------
63 510542.8 3.13 4.40 2.48 3.43
127 514288.6 5.97 7.89 4.65 6.04
255 516889.7 11.86 15.58 8.96 12.55
511 519802.4 23.10 28.81 16.95 26.19
1023 520802.7 45.30 52.51 33.19 45.95
2047 519997.1 90.63 104.00 65.26 81.74
>From the above data, it can be found that
- If the max batch size is 1023, the allocate/free latency at 99
percentile will be less than 100 us
- If the max batch size is 2047, the allocate/free latency at 99
percentile will be less than 100 us at most cases, but the max value
can reach 140 us.
So, if we can accept 100 us latency at most, batch size 1023 or 2047
is acceptable. This translates to batch scale factor 4 or 5. If so,
we should restrict the max batch scale factor to be 4 or 5.
What do you think about this? Do we need to collect more data?
--
Best Regards,
Huang, Ying
