https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/question-mkl-dcsrmv-matdescra/m-p/908367#M11900 <DIV style="margin:0px;"> <DIV id="quote_reply" style="width: 100%; margin-top: 5px;"> <DIV style="margin-left:2px;margin-right:2px;">Quoting - <A href="https://community.intel.com/en-us/profile/429860">kimjoonshikgmail.com</A></DIV> <DIV style="background-color:#E5E5E5; padding:5px;border: 1px; border-style: inset;margin-left:2px;margin-right:2px;"><EM>Would someone help me ? <BR />I need to do matrix-vector multiplication and for that tried to use mkl_dcsrmv( ) functions. My sparse matrix is compressed sparse row format and NOT square matrix. not upper triangular not lower triangular not diagonal. That is just general sparse matrix.<BR /><BR />then what 'matdescra' options do I have to use ? for<BR />matdescra[0], matdescra[1], ~ matdescra[3] ?<BR /><BR />for matdecra[1], In the manual I can find only 'L', or 'U' options for lower and upper triangular matrix and there are not an option for neither of them.<BR /><BR />Thank you.<BR />Joon.<BR /></EM></DIV> </DIV> </DIV> For general matrix you just put 'G' as first element, then 2nd, and 3rd are ignored anyway, so put anything, and 'F' (for 1-based indexing) or 'C' (for zero-based) as the last one.<BR /><BR />See Table <EM>Possible Combinations of Element Values of the Parameter matdescra</EM>.<BR /><BR />A.<BR /><BR /> Tue, 28 Jul 2009 03:33:49 GMT ArturGuzik 2009-07-28T03:33:49Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/question-mkl-dcsrmv-matdescra/m-p/908366#M11899 Would someone help me ? <BR />I need to do matrix-vector multiplication and for that tried to use mkl_dcsrmv( ) functions. My sparse matrix is compressed sparse row format and NOT square matrix. not upper triangular not lower triangular not diagonal. That is just general sparse matrix.<BR /><BR />then what 'matdescra' options do I have to use ? for<BR />matdescra[0], matdescra[1], ~ matdescra[3] ?<BR /><BR />for matdecra[1], In the manual I can find only 'L', or 'U' options for lower and upper triangular matrix and there are not an option for neither of them.<BR /><BR />Thank you.<BR />Joon.<BR /> Tue, 28 Jul 2009 02:20:31 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/question-mkl-dcsrmv-matdescra/m-p/908366#M11899 kimjoonshikgmail_com 2009-07-28T02:20:31Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/question-mkl-dcsrmv-matdescra/m-p/908367#M11900 <DIV style="margin:0px;"> <DIV id="quote_reply" style="width: 100%; margin-top: 5px;"> <DIV style="margin-left:2px;margin-right:2px;">Quoting - <A href="https://community.intel.com/en-us/profile/429860">kimjoonshikgmail.com</A></DIV> <DIV style="background-color:#E5E5E5; padding:5px;border: 1px; border-style: inset;margin-left:2px;margin-right:2px;"><EM>Would someone help me ? <BR />I need to do matrix-vector multiplication and for that tried to use mkl_dcsrmv( ) functions. My sparse matrix is compressed sparse row format and NOT square matrix. not upper triangular not lower triangular not diagonal. That is just general sparse matrix.<BR /><BR />then what 'matdescra' options do I have to use ? for<BR />matdescra[0], matdescra[1], ~ matdescra[3] ?<BR /><BR />for matdecra[1], In the manual I can find only 'L', or 'U' options for lower and upper triangular matrix and there are not an option for neither of them.<BR /><BR />Thank you.<BR />Joon.<BR /></EM></DIV> </DIV> </DIV> For general matrix you just put 'G' as first element, then 2nd, and 3rd are ignored anyway, so put anything, and 'F' (for 1-based indexing) or 'C' (for zero-based) as the last one.<BR /><BR />See Table <EM>Possible Combinations of Element Values of the Parameter matdescra</EM>.<BR /><BR />A.<BR /><BR /> Tue, 28 Jul 2009 03:33:49 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/question-mkl-dcsrmv-matdescra/m-p/908367#M11900 ArturGuzik 2009-07-28T03:33:49Z