https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/performance-in-libmkl-mc-so/m-p/788605#M2016 Profiling tools tells me that about 40% sampling time is spent on LN4_M4_K_LOOPgas_1 in libmkl_mc.so lib. <BR /><BR />I wonder whether this is any performance optimization that I can apply here.<BR /><BR />Thanks Mon, 21 Nov 2011 19:44:10 GMT missing__zlw 2011-11-21T19:44:10Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/performance-in-libmkl-mc-so/m-p/788605#M2016 Profiling tools tells me that about 40% sampling time is spent on LN4_M4_K_LOOPgas_1 in libmkl_mc.so lib. <BR /><BR />I wonder whether this is any performance optimization that I can apply here.<BR /><BR />Thanks Mon, 21 Nov 2011 19:44:10 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/performance-in-libmkl-mc-so/m-p/788605#M2016 missing__zlw 2011-11-21T19:44:10Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/performance-in-libmkl-mc-so/m-p/788606#M2017 Hi,<BR /><BR />Could you provide us with small testcase (with describing all your environment)to see and analyze this problem on our side? Tue, 22 Nov 2011 04:16:23 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/performance-in-libmkl-mc-so/m-p/788606#M2017 barragan_villanueva_ 2011-11-22T04:16:23Z