https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Fast-data-buffer-initialisation-routine/m-p/810448#M3856 Hi,<BR /><BR />On C/C++ you can use standard<BR /> Function: void * memset (void *BLOCK, int C, size_t SIZE)<BR /> This function copies the value of C (converted to an `unsigned<BR /> char') into each of the first SIZE bytes of the object beginning<BR /> at BLOCK. It returns the value of BLOCK.<BR /><BR />And in case of using Intel compilers it will be highly optimized. Wed, 02 Feb 2011 04:42:19 GMT barragan_villanueva_ 2011-02-02T04:42:19Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Fast-data-buffer-initialisation-routine/m-p/810447#M3855 I can't find a fast data buffer initialisation routine in MKL, VML etc (Typical use would be to set all array elements to zero.) I would have thought that this would be a useful routine. I did a lot of work with the Digital Extended Maths Library some years ago which from memory had a fast initialisation routine.<BR /><BR />Apologies if there is one and I've missed it<BR /><BR />David<BR /><BR /> Tue, 01 Feb 2011 23:25:33 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Fast-data-buffer-initialisation-routine/m-p/810447#M3855 dehvidc1 2011-02-01T23:25:33Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Fast-data-buffer-initialisation-routine/m-p/810448#M3856 Hi,<BR /><BR />On C/C++ you can use standard<BR /> Function: void * memset (void *BLOCK, int C, size_t SIZE)<BR /> This function copies the value of C (converted to an `unsigned<BR /> char') into each of the first SIZE bytes of the object beginning<BR /> at BLOCK. It returns the value of BLOCK.<BR /><BR />And in case of using Intel compilers it will be highly optimized. Wed, 02 Feb 2011 04:42:19 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Fast-data-buffer-initialisation-routine/m-p/810448#M3856 barragan_villanueva_ 2011-02-02T04:42:19Z