apeforest commented on a change in pull request #13191: [MXNET-1185] [WIP]
Support large array in several operators
URL: https://github.com/apache/incubator-mxnet/pull/13191#discussion_r236553150
##########
File path: src/operator/mxnet_op.h
##########
@@ -536,22 +536,22 @@ struct Kernel<OP, cpu> {
* \param args Varargs to eventually pass to the OP::Map() functoion
*/
template<typename PRIMITIVE_OP, typename DType, typename ...Args>
- static void LaunchTuned(mshadow::Stream<cpu> *, const int N, Args... args) {
+ static void LaunchTuned(mshadow::Stream<cpu> *, const size_t N, Args...
args) {
#ifdef _OPENMP
const int omp_threads =
engine::OpenMP::Get()->GetRecommendedOMPThreadCount();
if (omp_threads < 2 || !tuned_op<PRIMITIVE_OP, DType>::UseOMP(
- static_cast<size_t>(N), static_cast<size_t>(omp_threads))) {
- for (int i = 0; i < N; ++i) {
+ N, static_cast<size_t>(omp_threads))) {
+ for (size_t i = 0; i < N; ++i) {
Review comment:
Because omp parallel block requires index to be signed integer, while size_t
is unsigned.
----------------------------------------------------------------
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:
[email protected]
With regards,
Apache Git Services
