You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2018/11/28 01:24:58 UTC
[GitHub] rongzha1 commented on a change in pull request #12866: Optimization
for embedding OP for CPU
rongzha1 commented on a change in pull request #12866: Optimization for embedding OP for CPU
URL: https://github.com/apache/incubator-mxnet/pull/12866#discussion_r236910849
##########
File path: src/operator/tensor/indexing_op.cc
##########
@@ -61,10 +61,14 @@ template<typename DType>
bool CheckIndexOutOfBound(const DType* data_ptr, size_t data_size,
const DType min, const DType max) {
bool is_valid = true;
- for (size_t i = 0; i < data_size; i++) {
+ // to avoid Jenkins omp check error
+ int64_t size = data_size;
+ int64_t check_block_size = dmlc::GetEnv("MXNET_CPU_PARALLEL_CHECK_SIZE", 14000);
Review comment:
I have done some test to make sure the magic number size.
size | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | max time
-- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | --
2000 | 0.017 | 0.052 | 0.072 | 0.103 | 0.104 | 0.042 | 0.097 | 0.099 | 0.097 | 0.128 | 0.079 | 0.099 | 0.08 | 0.113 | 0.081 | 0.095 | 0.081 | 0.085 | 0.084 | 0.081 | 0.116 | 0.082 | 0.085 | 0.083 | 0.041 | 0.082 | 0.082 | 0.083 | 0.128
4000 | 0.01 | 0.006 | 0.015 | 0.015 | 0.007 | 0.007 | 0.007 | 0.007 | 0.01 | 0.006 | 0.01 | 0.007 | 0.007 | 0.007 | 0.007 | 0.007 | 0.007 | 0.009 | 0.007 | 0.008 | 0.009 | 0.008 | 0.009 | 0.008 | 0.009 | 0.009 | 0.009 | 0.013 | 0.015
6000 | 0.014 | 0.008 | 0.02 | 0.008 | 0.008 | 0.009 | 0.008 | 0.007 | 0.007 | 0.01 | 0.008 | 0.011 | 0.009 | 0.01 | 0.008 | 0.009 | 0.009 | 0.009 | 0.008 | 0.008 | 0.009 | 0.008 | 0.01 | 0.01 | 0.011 | 0.009 | 0.009 | 0.011 | 0.02
8000 | 0.018 | 0.01 | 0.025 | 0.01 | 0.007 | 0.008 | 0.008 | 0.011 | 0.01 | 0.013 | 0.009 | 0.012 | 0.009 | 0.013 | 0.009 | 0.01 | 0.009 | 0.01 | 0.008 | 0.009 | 0.01 | 0.009 | 0.009 | 0.009 | 0.011 | 0.01 | 0.01 | 0.011 | 0.025
10000 | 0.02 | 0.012 | 0.029 | 0.01 | 0.011 | 0.008 | 0.021 | 0.011 | 0.008 | 0.015 | 0.009 | 0.01 | 0.009 | 0.01 | 0.012 | 0.011 | 0.009 | 0.011 | 0.01 | 0.008 | 0.01 | 0.01 | 0.01 | 0.013 | 0.01 | 0.011 | 0.011 | 0.02 | 0.029
12000 | 0.024 | 0.015 | 0.035 | 0.011 | 0.01 | 0.01 | 0.008 | 0.009 | 0.012 | 0.012 | 0.015 | 0.012 | 0.009 | 0.01 | 0.009 | 0.009 | 0.01 | 0.011 | 0.009 | 0.01 | 0.012 | 0.012 | 0.011 | 0.011 | 0.014 | 0.011 | 0.011 | 0.012 | 0.035
14000 | 0.028 | 0.017 | 0.04 | 0.013 | 0.011 | 0.01 | 0.01 | 0.014 | 0.013 | 0.012 | 0.012 | 0.009 | 0.011 | 0.011 | 0.009 | 0.011 | 0.012 | 0.011 | 0.012 | 0.01 | 0.014 | 0.015 | 0.013 | 0.012 | 0.013 | 0.011 | 0.011 | 0.013 | 0.04
16000 | 0.032 | 0.018 | 0.018 | 0.014 | 0.012 | 0.011 | 0.01 | 0.011 | 0.01 | 0.01 | 0.011 | 0.011 | 0.011 | 0.01 | 0.011 | 0.011 | 0.011 | 0.016 | 0.014 | 0.01 | 0.018 | 0.015 | 0.011 | 0.011 | 0.018 | 0.013 | 0.013 | 0.014 | 0.032
18000 | 0.038 | 0.022 | 0.017 | 0.013 | 0.013 | 0.013 | 0.019 | 0.014 | 0.013 | 0.017 | 0.01 | 0.012 | 0.01 | 0.011 | 0.011 | 0.011 | 0.017 | 0.017 | 0.01 | 0.012 | 0.013 | 0.011 | 0.029 | 0.011 | 0.013 | 0.014 | 0.014 | 0.015 | 0.038
20000 | 0.042 | 0.023 | 0.019 | 0.015 | 0.019 | 0.013 | 0.014 | 0.012 | 0.011 | 0.012 | 0.011 | 0.012 | 0.011 | 0.01 | 0.011 | 0.01 | 0.01 | 0.011 | 0.012 | 0.012 | 0.012 | 0.013 | 0.013 | 0.016 | 0.013 | 0.017 | 0.017 | 0.014 | 0.042
we can find that : after size=14000, function CheckIndexOutOfBound will benefit from omp thread.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services