Trying to port openblas for android. the openblas for android took from here https://sourceforge.net/projects/openblas/files/v0.2.8-arm/openblas-v0.2.8-android-rc1.tar.gz/download. This is my time measures for cblas_sgemm
1. android samsung s6
0.003 s for 100x100 sgemm
3.130 s for 1000X1000 sgemm
2. windows x64 intel i5
0.001 s for 100x100 sgemm
0.080 s for 1000X1000 sgemm
What causing to openblas run so slow with 1000X1000 matrix in android. How i can improve it?