Lines Matching refs:nblocks

581   PetscInt           nblocks, blocksize = 512; /* blocksize is fixed to be 512 */  in MatMult_SeqSELLHIP()  local
604 nblocks = 1 + (nrows - 1) / sliceheight; in MatMult_SeqSELLHIP()
606 …matmult_seqsell_tiled_kernel9<2><<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, acolidx, ava… in MatMult_SeqSELLHIP()
608 …matmult_seqsell_tiled_kernel9<4><<<nblocks, dim3(WARP_SIZE, 4)>>>(nrows, sliceheight, acolidx, ava… in MatMult_SeqSELLHIP()
610 …matmult_seqsell_tiled_kernel9<8><<<nblocks, dim3(WARP_SIZE, 8)>>>(nrows, sliceheight, acolidx, ava… in MatMult_SeqSELLHIP()
612 …matmult_seqsell_tiled_kernel9<16><<<nblocks, dim3(WARP_SIZE, 16)>>>(nrows, sliceheight, acolidx, a… in MatMult_SeqSELLHIP()
614 …matmult_seqsell_tiled_kernel9<2><<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, acolidx, ava… in MatMult_SeqSELLHIP()
618 nblocks = 1 + (nrows - 1) / (hipstruct->blocky * sliceheight); in MatMult_SeqSELLHIP()
620 …matmult_seqsell_tiled_kernel7<<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, acolidx, aval, … in MatMult_SeqSELLHIP()
622 …matmult_seqsell_tiled_kernel7<<<nblocks, dim3(WARP_SIZE, 4)>>>(nrows, sliceheight, acolidx, aval, … in MatMult_SeqSELLHIP()
624 …matmult_seqsell_tiled_kernel7<<<nblocks, dim3(WARP_SIZE, 8)>>>(nrows, sliceheight, acolidx, aval, … in MatMult_SeqSELLHIP()
626 …matmult_seqsell_tiled_kernel7<<<nblocks, dim3(WARP_SIZE, 16)>>>(nrows, sliceheight, acolidx, aval,… in MatMult_SeqSELLHIP()
628 nblocks = 1 + (nrows - 1) / (2 * sliceheight); in MatMult_SeqSELLHIP()
629 …matmult_seqsell_tiled_kernel7<<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, acolidx, aval, … in MatMult_SeqSELLHIP()
634 nblocks = 1 + (nrows - 1) / (blocksize / 32); /* 1 slice per block if sliceheight=32 */ in MatMult_SeqSELLHIP()
635 …matmult_seqsell_tiled_kernel6<<<nblocks, block32>>>(nrows, sliceheight, acolidx, aval, sliidx, x, … in MatMult_SeqSELLHIP()
638 nblocks = 1 + (nrows - 1) / (blocksize / 16); /* 2 slices per block if sliceheight=32*/ in MatMult_SeqSELLHIP()
639 …matmult_seqsell_tiled_kernel5<<<nblocks, block16>>>(nrows, sliceheight, acolidx, aval, sliidx, x, … in MatMult_SeqSELLHIP()
642 nblocks = 1 + (nrows - 1) / (blocksize / 8); /* 4 slices per block if sliceheight=32 */ in MatMult_SeqSELLHIP()
643 …matmult_seqsell_tiled_kernel4<<<nblocks, block8>>>(nrows, sliceheight, acolidx, aval, sliidx, x, y… in MatMult_SeqSELLHIP()
646 nblocks = 1 + (nrows - 1) / (blocksize / 4); /* 8 slices per block if sliceheight=32 */ in MatMult_SeqSELLHIP()
647 …matmult_seqsell_tiled_kernel3<<<nblocks, block4>>>(nrows, sliceheight, acolidx, aval, sliidx, x, y… in MatMult_SeqSELLHIP()
650 nblocks = 1 + (nrows - 1) / (blocksize / 2); in MatMult_SeqSELLHIP()
651 …matmult_seqsell_tiled_kernel2<<<nblocks, block2>>>(nrows, sliceheight, acolidx, aval, sliidx, x, y… in MatMult_SeqSELLHIP()
654 nblocks = 1 + (nrows - 1) / blocksize; in MatMult_SeqSELLHIP()
655 …matmult_seqsell_basic_kernel<<<nblocks, blocksize>>>(nrows, sliceheight, acolidx, aval, sliidx, x,… in MatMult_SeqSELLHIP()
665 nblocks = 1 + (nchunks - 1) / chunksperblock; in MatMult_SeqSELLHIP()
668 …matmult_seqsell_tiled_kernel8<2><<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, chunksperblo… in MatMult_SeqSELLHIP()
670 …matmult_seqsell_tiled_kernel8<4><<<nblocks, dim3(WARP_SIZE, 4)>>>(nrows, sliceheight, chunksperblo… in MatMult_SeqSELLHIP()
672 …matmult_seqsell_tiled_kernel8<8><<<nblocks, dim3(WARP_SIZE, 8)>>>(nrows, sliceheight, chunksperblo… in MatMult_SeqSELLHIP()
674 …matmult_seqsell_tiled_kernel8<16><<<nblocks, dim3(WARP_SIZE, 16)>>>(nrows, sliceheight, chunksperb… in MatMult_SeqSELLHIP()
676 …matmult_seqsell_tiled_kernel8<2><<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, chunksperblo… in MatMult_SeqSELLHIP()
682 nblocks = 1 + (nrows - 1) / (2 * sliceheight); /* two slices per block */ in MatMult_SeqSELLHIP()
683 …matmult_seqsell_tiled_kernel7<<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, acolidx, aval, … in MatMult_SeqSELLHIP()
685 nblocks = 1 + (nrows - 1) / sliceheight; in MatMult_SeqSELLHIP()
686 …matmult_seqsell_tiled_kernel9<2><<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, acolidx, ava… in MatMult_SeqSELLHIP()
689 nblocks = 1 + (nrows - 1) / sliceheight; in MatMult_SeqSELLHIP()
690 …matmult_seqsell_tiled_kernel9<8><<<nblocks, dim3(WARP_SIZE, 8)>>>(nrows, sliceheight, acolidx, ava… in MatMult_SeqSELLHIP()
692 nblocks = 1 + (nrows - 1) / sliceheight; in MatMult_SeqSELLHIP()
693 …matmult_seqsell_tiled_kernel9<16><<<nblocks, dim3(WARP_SIZE, 16)>>>(nrows, sliceheight, acolidx, a… in MatMult_SeqSELLHIP()
729 PetscInt nblocks, blocksize = 512; in MatMultAdd_SeqSELLHIP() local
739 nblocks = 1 + (nrows - 1) / sliceheight; in MatMultAdd_SeqSELLHIP()
741 …matmultadd_seqsell_tiled_kernel9<2><<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, acolidx, … in MatMultAdd_SeqSELLHIP()
743 …matmultadd_seqsell_tiled_kernel9<4><<<nblocks, dim3(WARP_SIZE, 4)>>>(nrows, sliceheight, acolidx, … in MatMultAdd_SeqSELLHIP()
745 …matmultadd_seqsell_tiled_kernel9<8><<<nblocks, dim3(WARP_SIZE, 8)>>>(nrows, sliceheight, acolidx, … in MatMultAdd_SeqSELLHIP()
747 …matmultadd_seqsell_tiled_kernel9<16><<<nblocks, dim3(WARP_SIZE, 16)>>>(nrows, sliceheight, acolidx… in MatMultAdd_SeqSELLHIP()
749 …matmultadd_seqsell_tiled_kernel9<2><<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, acolidx, … in MatMultAdd_SeqSELLHIP()
757 nblocks = 1 + (nchunks - 1) / chunksperblock; in MatMultAdd_SeqSELLHIP()
760 …matmultadd_seqsell_tiled_kernel8<2><<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, chunksper… in MatMultAdd_SeqSELLHIP()
762 …matmultadd_seqsell_tiled_kernel8<4><<<nblocks, dim3(WARP_SIZE, 4)>>>(nrows, sliceheight, chunksper… in MatMultAdd_SeqSELLHIP()
764 …matmultadd_seqsell_tiled_kernel8<8><<<nblocks, dim3(WARP_SIZE, 8)>>>(nrows, sliceheight, chunksper… in MatMultAdd_SeqSELLHIP()
766 …matmultadd_seqsell_tiled_kernel8<16><<<nblocks, dim3(WARP_SIZE, 16)>>>(nrows, sliceheight, chunksp… in MatMultAdd_SeqSELLHIP()
768 …matmultadd_seqsell_tiled_kernel8<2><<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, chunksper… in MatMultAdd_SeqSELLHIP()
772 nblocks = 1 + (nrows - 1) / (blocky * sliceheight); in MatMultAdd_SeqSELLHIP()
774 …matmultadd_seqsell_tiled_kernel7<<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, acolidx, ava… in MatMultAdd_SeqSELLHIP()
776 …matmultadd_seqsell_tiled_kernel7<<<nblocks, dim3(WARP_SIZE, 4)>>>(nrows, sliceheight, acolidx, ava… in MatMultAdd_SeqSELLHIP()
778 …matmultadd_seqsell_tiled_kernel7<<<nblocks, dim3(WARP_SIZE, 8)>>>(nrows, sliceheight, acolidx, ava… in MatMultAdd_SeqSELLHIP()
780 …matmultadd_seqsell_tiled_kernel7<<<nblocks, dim3(WARP_SIZE, 16)>>>(nrows, sliceheight, acolidx, av… in MatMultAdd_SeqSELLHIP()
782 nblocks = 1 + (nrows - 1) / (2 * sliceheight); in MatMultAdd_SeqSELLHIP()
783 …matmultadd_seqsell_tiled_kernel7<<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, acolidx, ava… in MatMultAdd_SeqSELLHIP()
788 nblocks = 1 + (nrows - 1) / (blocksize / 32); in MatMultAdd_SeqSELLHIP()
789 …matmultadd_seqsell_tiled_kernel6<<<nblocks, block32>>>(nrows, sliceheight, acolidx, aval, sliidx, … in MatMultAdd_SeqSELLHIP()
792 nblocks = 1 + (nrows - 1) / (blocksize / 16); in MatMultAdd_SeqSELLHIP()
793 …matmultadd_seqsell_tiled_kernel5<<<nblocks, block16>>>(nrows, sliceheight, acolidx, aval, sliidx, … in MatMultAdd_SeqSELLHIP()
796 nblocks = 1 + (nrows - 1) / (blocksize / 8); in MatMultAdd_SeqSELLHIP()
797 …matmultadd_seqsell_tiled_kernel4<<<nblocks, block8>>>(nrows, sliceheight, acolidx, aval, sliidx, x… in MatMultAdd_SeqSELLHIP()
800 nblocks = 1 + (nrows - 1) / (blocksize / 4); in MatMultAdd_SeqSELLHIP()
801 …matmultadd_seqsell_tiled_kernel3<<<nblocks, block4>>>(nrows, sliceheight, acolidx, aval, sliidx, x… in MatMultAdd_SeqSELLHIP()
804 nblocks = 1 + (nrows - 1) / (blocksize / 2); in MatMultAdd_SeqSELLHIP()
805 …matmultadd_seqsell_tiled_kernel2<<<nblocks, block2>>>(nrows, sliceheight, acolidx, aval, sliidx, x… in MatMultAdd_SeqSELLHIP()
808 nblocks = 1 + (nrows - 1) / blocksize; in MatMultAdd_SeqSELLHIP()
809 …matmultadd_seqsell_basic_kernel<<<nblocks, blocksize>>>(nrows, sliceheight, acolidx, aval, sliidx,… in MatMultAdd_SeqSELLHIP()
819 nblocks = 1 + (nchunks - 1) / chunksperblock; in MatMultAdd_SeqSELLHIP()
822 …matmultadd_seqsell_tiled_kernel8<2><<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, chunksper… in MatMultAdd_SeqSELLHIP()
824 …matmultadd_seqsell_tiled_kernel8<4><<<nblocks, dim3(WARP_SIZE, 4)>>>(nrows, sliceheight, chunksper… in MatMultAdd_SeqSELLHIP()
826 …matmultadd_seqsell_tiled_kernel8<8><<<nblocks, dim3(WARP_SIZE, 8)>>>(nrows, sliceheight, chunksper… in MatMultAdd_SeqSELLHIP()
828 …matmultadd_seqsell_tiled_kernel8<16><<<nblocks, dim3(WARP_SIZE, 16)>>>(nrows, sliceheight, chunksp… in MatMultAdd_SeqSELLHIP()
830 …matmultadd_seqsell_tiled_kernel8<2><<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, chunksper… in MatMultAdd_SeqSELLHIP()
836 nblocks = 1 + (nrows - 1) / (2 * sliceheight); /* two slices per block */ in MatMultAdd_SeqSELLHIP()
837 …matmultadd_seqsell_tiled_kernel7<<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, acolidx, ava… in MatMultAdd_SeqSELLHIP()
839 nblocks = 1 + (nrows - 1) / sliceheight; in MatMultAdd_SeqSELLHIP()
840 …matmultadd_seqsell_tiled_kernel9<2><<<nblocks, dim3(WARP_SIZE, 2)>>>(nrows, sliceheight, acolidx, … in MatMultAdd_SeqSELLHIP()
843 nblocks = 1 + (nrows - 1) / sliceheight; in MatMultAdd_SeqSELLHIP()
844 …matmultadd_seqsell_tiled_kernel9<8><<<nblocks, dim3(WARP_SIZE, 8)>>>(nrows, sliceheight, acolidx, … in MatMultAdd_SeqSELLHIP()
846 nblocks = 1 + (nrows - 1) / sliceheight; in MatMultAdd_SeqSELLHIP()
847 …matmultadd_seqsell_tiled_kernel9<16><<<nblocks, dim3(WARP_SIZE, 16)>>>(nrows, sliceheight, acolidx… in MatMultAdd_SeqSELLHIP()