Loop Id: 106 | Module: libparcsr_mv.so | Source: par_csr_matop.c:865-989 [...] | Coverage: 0.03% |
---|
Loop Id: 106 | Module: libparcsr_mv.so | Source: par_csr_matop.c:865-989 [...] | Coverage: 0.03% |
---|
0xc570 MOV %RCX,-0x30(%RBP) |
0xc574 MOV %RAX,%R14 |
0xc577 MOV -0x48(%RBP),%RCX |
0xc57b CMP %RCX,%R12 |
0xc57e JGE c8cd |
0xc584 MOV %R12,%RDX |
0xc587 CMPQ $0,-0xb8(%RBP) |
0xc58f JE c5a9 |
0xc591 MOV %R14,(%RDI,%RDX,8) |
0xc595 MOVQ $0,(%R15,%R14,8) |
0xc59d MOV -0x50(%RBP),%RAX |
0xc5a1 MOV %RDX,(%RAX,%R14,8) |
0xc5a5 LEA 0x1(%R14),%RAX |
0xc5a9 CMPQ $0,-0xb0(%RBP) |
0xc5b1 JE c740 |
0xc5b7 MOV -0x58(%RBP),%RCX |
0xc5bb MOV (%RCX,%RDX,8),%RSI |
0xc5bf LEA 0x1(%RDX),%R12 |
0xc5c3 CMP 0x8(%RCX,%RDX,8),%RSI |
0xc5c8 JGE c744 |
0xc5ce MOV %R12,-0x38(%RBP) |
0xc5d2 MOV -0x30(%RBP),%RCX |
0xc5d6 MOV %RDX,-0x40(%RBP) |
0xc5da JMP c5f9 |
(110) 0xc5e0 MOV %R12,%RSI |
(110) 0xc5e3 INC %RSI |
(110) 0xc5e6 MOV -0x58(%RBP),%RDX |
(110) 0xc5ea MOV -0x40(%RBP),%R12 |
(110) 0xc5ee CMP 0x8(%RDX,%R12,8),%RSI |
(110) 0xc5f3 JGE c750 |
(110) 0xc5f9 MOV -0xc0(%RBP),%RDX |
(110) 0xc600 MOV (%RDX,%RSI,8),%R8 |
(110) 0xc604 MOV -0xc8(%RBP),%RDX |
(110) 0xc60b MOV %RSI,%R12 |
(110) 0xc60e VMOVSD (%RDX,%RSI,8),%XMM0 |
(110) 0xc613 MOV -0x78(%RBP),%RDX |
(110) 0xc617 MOV (%RDX,%R8,8),%R9 |
(110) 0xc61b MOV 0x8(%RDX,%R8,8),%R10 |
(110) 0xc620 CMP %R10,%R9 |
(110) 0xc623 JL c665 |
(110) 0xc625 JMP c69e |
(112) 0xc630 MOV %RCX,(%RDI,%RSI,8) |
(112) 0xc634 MOV -0xa0(%RBP),%RDX |
(112) 0xc63b VMULSD (%RDX,%R9,8),%XMM0,%XMM1 |
(112) 0xc641 VMOVSD %XMM1,(%RBX,%RCX,8) |
(112) 0xc646 MOV -0x90(%RBP),%RDX |
(112) 0xc64d MOV %R11,(%RDX,%RCX,8) |
(112) 0xc651 INC %RCX |
(112) 0xc654 MOV -0x78(%RBP),%RDX |
(112) 0xc658 MOV 0x8(%RDX,%R8,8),%R10 |
(112) 0xc65d INC %R9 |
(112) 0xc660 CMP %R10,%R9 |
(112) 0xc663 JGE c69e |
(112) 0xc665 MOV -0x100(%RBP),%RDX |
(112) 0xc66c MOV (%RDX,%R9,8),%R11 |
(112) 0xc670 LEA (%R11,%R13,1),%RSI |
(112) 0xc674 MOV (%RDI,%RSI,8),%RDX |
(112) 0xc678 CMP -0x30(%RBP),%RDX |
(112) 0xc67c JL c630 |
(112) 0xc67e MOV -0xa0(%RBP),%RSI |
(112) 0xc685 VMOVSD (%RSI,%R9,8),%XMM1 |
(112) 0xc68b VFMADD213SD (%RBX,%RDX,8),%XMM0,%XMM1 |
(112) 0xc691 VMOVSD %XMM1,(%RBX,%RDX,8) |
(112) 0xc696 INC %R9 |
(112) 0xc699 CMP %R10,%R9 |
(112) 0xc69c JL c665 |
(110) 0xc69e MOV -0x70(%RBP),%RDX |
(110) 0xc6a2 MOV (%RDX,%R8,8),%R9 |
(110) 0xc6a6 MOV 0x8(%RDX,%R8,8),%R10 |
(110) 0xc6ab CMP %R10,%R9 |
(110) 0xc6ae JL c6f7 |
(110) 0xc6b0 JMP c5e0 |
(111) 0xc6c0 MOV %RAX,(%RDI,%R11,8) |
(111) 0xc6c4 MOV -0x98(%RBP),%RDX |
(111) 0xc6cb VMULSD (%RDX,%R9,8),%XMM0,%XMM1 |
(111) 0xc6d1 VMOVSD %XMM1,(%R15,%RAX,8) |
(111) 0xc6d7 MOV -0x50(%RBP),%RDX |
(111) 0xc6db MOV %R11,(%RDX,%RAX,8) |
(111) 0xc6df INC %RAX |
(111) 0xc6e2 MOV -0x70(%RBP),%RDX |
(111) 0xc6e6 MOV 0x8(%RDX,%R8,8),%R10 |
(111) 0xc6eb INC %R9 |
(111) 0xc6ee CMP %R10,%R9 |
(111) 0xc6f1 JGE c5e0 |
(111) 0xc6f7 MOV -0xf8(%RBP),%RDX |
(111) 0xc6fe MOV (%RDX,%R9,8),%R11 |
(111) 0xc702 MOV (%RDI,%R11,8),%RDX |
(111) 0xc706 CMP %R14,%RDX |
(111) 0xc709 JL c6c0 |
(111) 0xc70b MOV -0x98(%RBP),%RSI |
(111) 0xc712 VMOVSD (%RSI,%R9,8),%XMM1 |
(111) 0xc718 VFMADD213SD (%R15,%RDX,8),%XMM0,%XMM1 |
(111) 0xc71e VMOVSD %XMM1,(%R15,%RDX,8) |
(111) 0xc724 INC %R9 |
(111) 0xc727 CMP %R10,%R9 |
(111) 0xc72a JL c6f7 |
(110) 0xc72c JMP c5e0 |
0xc740 LEA 0x1(%RDX),%R12 |
0xc744 MOV -0x30(%RBP),%RCX |
0xc748 JMP c757 |
0xc750 MOV %R12,%RDX |
0xc753 MOV -0x38(%RBP),%R12 |
0xc757 MOV -0x60(%RBP),%RSI |
0xc75b MOV (%RSI,%RDX,8),%RDX |
0xc75f CMP (%RSI,%R12,8),%RDX |
0xc763 JGE c570 |
0xc769 MOV %R12,-0x38(%RBP) |
0xc76d JMP c785 |
(107) 0xc770 INC %RDX |
(107) 0xc773 MOV -0x60(%RBP),%RSI |
(107) 0xc777 MOV -0x38(%RBP),%R12 |
(107) 0xc77b CMP (%RSI,%R12,8),%RDX |
(107) 0xc77f JGE c570 |
(107) 0xc785 MOV -0xe0(%RBP),%RSI |
(107) 0xc78c MOV (%RSI,%RDX,8),%RSI |
(107) 0xc790 MOV -0xd8(%RBP),%R8 |
(107) 0xc797 VMOVSD (%R8,%RDX,8),%XMM0 |
(107) 0xc79d MOV -0x80(%RBP),%R9 |
(107) 0xc7a1 MOV (%R9,%RSI,8),%R8 |
(107) 0xc7a5 MOV 0x8(%R9,%RSI,8),%R9 |
(107) 0xc7aa CMP %R9,%R8 |
(107) 0xc7ad JL c7f3 |
(107) 0xc7af JMP c828 |
(109) 0xc7c0 MOV %RAX,(%RDI,%R10,8) |
(109) 0xc7c4 MOV -0xa8(%RBP),%R9 |
(109) 0xc7cb VMULSD (%R9,%R8,8),%XMM0,%XMM1 |
(109) 0xc7d1 VMOVSD %XMM1,(%R15,%RAX,8) |
(109) 0xc7d7 MOV -0x50(%RBP),%R9 |
(109) 0xc7db MOV %R10,(%R9,%RAX,8) |
(109) 0xc7df INC %RAX |
(109) 0xc7e2 MOV -0x80(%RBP),%R9 |
(109) 0xc7e6 MOV 0x8(%R9,%RSI,8),%R9 |
(109) 0xc7eb INC %R8 |
(109) 0xc7ee CMP %R9,%R8 |
(109) 0xc7f1 JGE c828 |
(109) 0xc7f3 MOV -0x108(%RBP),%R10 |
(109) 0xc7fa MOV (%R10,%R8,8),%R10 |
(109) 0xc7fe MOV (%RDI,%R10,8),%R11 |
(109) 0xc802 CMP %R14,%R11 |
(109) 0xc805 JL c7c0 |
(109) 0xc807 MOV -0xa8(%RBP),%R10 |
(109) 0xc80e VMOVSD (%R10,%R8,8),%XMM1 |
(109) 0xc814 VFMADD213SD (%R15,%R11,8),%XMM0,%XMM1 |
(109) 0xc81a VMOVSD %XMM1,(%R15,%R11,8) |
(109) 0xc820 INC %R8 |
(109) 0xc823 CMP %R9,%R8 |
(109) 0xc826 JL c7f3 |
(107) 0xc828 CMPQ $0,-0xd0(%RBP) |
(107) 0xc830 JE c770 |
(107) 0xc836 MOV -0x68(%RBP),%R9 |
(107) 0xc83a MOV (%R9,%RSI,8),%R8 |
(107) 0xc83e MOV 0x8(%R9,%RSI,8),%R9 |
(107) 0xc843 JMP c86f |
(108) 0xc850 MOV -0x88(%RBP),%R10 |
(108) 0xc857 VMOVSD (%R10,%R8,8),%XMM1 |
(108) 0xc85d VFMADD213SD (%RBX,%R11,8),%XMM0,%XMM1 |
(108) 0xc863 VMOVSD %XMM1,(%RBX,%R11,8) |
(108) 0xc869 MOV %R12,%R13 |
(108) 0xc86c INC %R8 |
(108) 0xc86f CMP %R9,%R8 |
(108) 0xc872 JGE c770 |
(108) 0xc878 MOV -0xe8(%RBP),%R10 |
(108) 0xc87f MOV (%R10,%R8,8),%R10 |
(108) 0xc883 MOV -0xf0(%RBP),%R11 |
(108) 0xc88a MOV (%R11,%R10,8),%R10 |
(108) 0xc88e MOV %R13,%R12 |
(108) 0xc891 ADD %R10,%R13 |
(108) 0xc894 MOV (%RDI,%R13,8),%R11 |
(108) 0xc898 CMP -0x30(%RBP),%R11 |
(108) 0xc89c JGE c850 |
(108) 0xc89e MOV %RCX,(%RDI,%R13,8) |
(108) 0xc8a2 MOV -0x88(%RBP),%R9 |
(108) 0xc8a9 VMULSD (%R9,%R8,8),%XMM0,%XMM1 |
(108) 0xc8af VMOVSD %XMM1,(%RBX,%RCX,8) |
(108) 0xc8b4 MOV -0x90(%RBP),%R9 |
(108) 0xc8bb MOV %R10,(%R9,%RCX,8) |
(108) 0xc8bf INC %RCX |
(108) 0xc8c2 MOV -0x68(%RBP),%R9 |
(108) 0xc8c6 MOV 0x8(%R9,%RSI,8),%R9 |
(108) 0xc8cb JMP c869 |
/home/eoseret/qaas_runs_CPU_9468/171-147-2675/intel/AMG/build/AMG/AMG/parcsr_mv/par_csr_matop.c: 865 - 989 |
-------------------------------------------------------------------------------- |
865: for (i1 = ns; i1 < ne; i1++) |
[...] |
874: if ( allsquare ) |
875: { |
876: B_marker[i1] = jj_count_diag; |
877: C_diag_data[jj_count_diag] = zero; |
878: C_diag_j[jj_count_diag] = i1; |
879: jj_count_diag++; |
[...] |
886: if (num_cols_offd_A) |
887: { |
888: for (jj2 = A_offd_i[i1]; jj2 < A_offd_i[i1+1]; jj2++) |
889: { |
890: i2 = A_offd_j[jj2]; |
891: a_entry = A_offd_data[jj2]; |
[...] |
897: for (jj3 = B_ext_offd_i[i2]; jj3 < B_ext_offd_i[i2+1]; jj3++) |
898: { |
899: i3 = num_cols_diag_B+B_ext_offd_j[jj3]; |
[...] |
907: if (B_marker[i3] < jj_row_begin_offd) |
908: { |
909: B_marker[i3] = jj_count_offd; |
910: C_offd_data[jj_count_offd] = a_entry*B_ext_offd_data[jj3]; |
911: C_offd_j[jj_count_offd] = i3-num_cols_diag_B; |
912: jj_count_offd++; |
913: } |
914: else |
915: C_offd_data[B_marker[i3]] += a_entry*B_ext_offd_data[jj3]; |
916: } |
917: for (jj3 = B_ext_diag_i[i2]; jj3 < B_ext_diag_i[i2+1]; jj3++) |
918: { |
919: i3 = B_ext_diag_j[jj3]; |
920: if (B_marker[i3] < jj_row_begin_diag) |
921: { |
922: B_marker[i3] = jj_count_diag; |
923: C_diag_data[jj_count_diag] = a_entry*B_ext_diag_data[jj3]; |
924: C_diag_j[jj_count_diag] = i3; |
925: jj_count_diag++; |
926: } |
927: else |
928: C_diag_data[B_marker[i3]] += a_entry*B_ext_diag_data[jj3]; |
[...] |
937: for (jj2 = A_diag_i[i1]; jj2 < A_diag_i[i1+1]; jj2++) |
938: { |
939: i2 = A_diag_j[jj2]; |
940: a_entry = A_diag_data[jj2]; |
[...] |
946: for (jj3 = B_diag_i[i2]; jj3 < B_diag_i[i2+1]; jj3++) |
947: { |
948: i3 = B_diag_j[jj3]; |
[...] |
956: if (B_marker[i3] < jj_row_begin_diag) |
957: { |
958: B_marker[i3] = jj_count_diag; |
959: C_diag_data[jj_count_diag] = a_entry*B_diag_data[jj3]; |
960: C_diag_j[jj_count_diag] = i3; |
961: jj_count_diag++; |
962: } |
963: else |
964: { |
965: C_diag_data[B_marker[i3]] += a_entry*B_diag_data[jj3]; |
966: } |
967: } |
968: if (num_cols_offd_B) |
969: { |
970: for (jj3 = B_offd_i[i2]; jj3 < B_offd_i[i2+1]; jj3++) |
971: { |
972: i3 = num_cols_diag_B+map_B_to_C[B_offd_j[jj3]]; |
[...] |
980: if (B_marker[i3] < jj_row_begin_offd) |
981: { |
982: B_marker[i3] = jj_count_offd; |
983: C_offd_data[jj_count_offd] = a_entry*B_offd_data[jj3]; |
984: C_offd_j[jj_count_offd] = i3-num_cols_diag_B; |
985: jj_count_offd++; |
986: } |
987: else |
988: { |
989: C_offd_data[B_marker[i3]] += a_entry*B_offd_data[jj3]; |
Path / |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.00 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 12.44 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.42 |
Bottlenecks | micro-operation queue, |
Function | hypre_ParMatmul.extracted.12 |
Source | par_csr_matop.c:865-865,par_csr_matop.c:874-879,par_csr_matop.c:886-888,par_csr_matop.c:937-937 |
Source loop unroll info | NA |
Source loop unroll confidence level | NA |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 6.17 |
CQA cycles if no scalar integer | 6.17 |
CQA cycles if FP arith vectorized | 6.17 |
CQA cycles if fully vectorized | 0.50 |
Front-end cycles | 6.17 |
DIV/SQRT cycles | 2.50 |
P0 cycles | 1.80 |
P1 cycles | 4.33 |
P2 cycles | 4.33 |
P3 cycles | 3.50 |
P4 cycles | 1.60 |
P5 cycles | 2.50 |
P6 cycles | 3.50 |
P7 cycles | 3.50 |
P8 cycles | 3.50 |
P9 cycles | 1.60 |
P10 cycles | 4.33 |
P11 cycles | 0.00 |
Inter-iter dependencies cycles | NA |
FE+BE cycles (UFS) | 6.32 |
Stall cycles (UFS) | 0.00 |
Nb insns | 35.00 |
Nb uops | 35.00 |
Nb loads | 13.00 |
Nb stores | 7.00 |
Nb stack references | 9.00 |
FLOP/cycle | 0.00 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 0.00 |
Nb FLOP fma | 0.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 25.95 |
Bytes prefetched | 0.00 |
Bytes loaded | 104.00 |
Bytes stored | 56.00 |
Stride 0 | NA |
Stride 1 | NA |
Stride n | NA |
Stride unknown | NA |
Stride indirect | NA |
Vectorization ratio all | 0.00 |
Vectorization ratio load | 0.00 |
Vectorization ratio store | 0.00 |
Vectorization ratio mul | NA |
Vectorization ratio add_sub | NA |
Vectorization ratio fma | NA |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 0.00 |
Vector-efficiency ratio all | 11.93 |
Vector-efficiency ratio load | 12.50 |
Vector-efficiency ratio store | 11.61 |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | NA |
Vector-efficiency ratio fma | NA |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 12.50 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.00 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 12.44 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.42 |
Bottlenecks | micro-operation queue, |
Function | hypre_ParMatmul.extracted.12 |
Source | par_csr_matop.c:865-865,par_csr_matop.c:874-879,par_csr_matop.c:886-888,par_csr_matop.c:937-937 |
Source loop unroll info | NA |
Source loop unroll confidence level | NA |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 6.17 |
CQA cycles if no scalar integer | 6.17 |
CQA cycles if FP arith vectorized | 6.17 |
CQA cycles if fully vectorized | 0.50 |
Front-end cycles | 6.17 |
DIV/SQRT cycles | 2.50 |
P0 cycles | 1.80 |
P1 cycles | 4.33 |
P2 cycles | 4.33 |
P3 cycles | 3.50 |
P4 cycles | 1.60 |
P5 cycles | 2.50 |
P6 cycles | 3.50 |
P7 cycles | 3.50 |
P8 cycles | 3.50 |
P9 cycles | 1.60 |
P10 cycles | 4.33 |
P11 cycles | 0.00 |
Inter-iter dependencies cycles | NA |
FE+BE cycles (UFS) | 6.32 |
Stall cycles (UFS) | 0.00 |
Nb insns | 35.00 |
Nb uops | 35.00 |
Nb loads | 13.00 |
Nb stores | 7.00 |
Nb stack references | 9.00 |
FLOP/cycle | 0.00 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 0.00 |
Nb FLOP fma | 0.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 25.95 |
Bytes prefetched | 0.00 |
Bytes loaded | 104.00 |
Bytes stored | 56.00 |
Stride 0 | NA |
Stride 1 | NA |
Stride n | NA |
Stride unknown | NA |
Stride indirect | NA |
Vectorization ratio all | 0.00 |
Vectorization ratio load | 0.00 |
Vectorization ratio store | 0.00 |
Vectorization ratio mul | NA |
Vectorization ratio add_sub | NA |
Vectorization ratio fma | NA |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 0.00 |
Vector-efficiency ratio all | 11.93 |
Vector-efficiency ratio load | 12.50 |
Vector-efficiency ratio store | 11.61 |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | NA |
Vector-efficiency ratio fma | NA |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 12.50 |
Path / |
Function | hypre_ParMatmul.extracted.12 |
Source file and lines | par_csr_matop.c:865-989 |
Module | libparcsr_mv.so |
nb instructions | 35 |
nb uops | 35 |
loop length | 149 |
used x86 registers | 9 |
used mmx registers | 0 |
used xmm registers | 0 |
used ymm registers | 0 |
used zmm registers | 0 |
nb stack references | 9 |
micro-operation queue | 6.17 cycles |
front end | 6.17 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 2.50 | 1.80 | 4.33 | 4.33 | 3.50 | 1.60 | 2.50 | 3.50 | 3.50 | 3.50 | 1.60 | 4.33 |
cycles | 2.50 | 1.80 | 4.33 | 4.33 | 3.50 | 1.60 | 2.50 | 3.50 | 3.50 | 3.50 | 1.60 | 4.33 |
Cycles executing div or sqrt instructions | NA |
FE+BE cycles | 6.32 |
Stall cycles | 0.00 |
Front-end | 6.17 |
Dispatch | 4.33 |
Overall L1 | 6.17 |
all | 0% |
load | 0% |
store | 0% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 0% |
all | 11% |
load | 12% |
store | 11% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 12% |
Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
MOV %RCX,-0x30(%RBP) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
MOV %RAX,%R14 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
MOV -0x48(%RBP),%RCX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
CMP %RCX,%R12 | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JGE c8cd <hypre_ParMatmul.extracted.12+0x57d> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
MOV %R12,%RDX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
CMPQ $0,-0xb8(%RBP) | 1 | 0.20 | 0.20 | 0.33 | 0.33 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.33 | 1 | 0.33 |
JE c5a9 <hypre_ParMatmul.extracted.12+0x259> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
MOV %R14,(%RDI,%RDX,8) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
MOVQ $0,(%R15,%R14,8) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
MOV -0x50(%RBP),%RAX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV %RDX,(%RAX,%R14,8) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
LEA 0x1(%R14),%RAX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
CMPQ $0,-0xb0(%RBP) | 1 | 0.20 | 0.20 | 0.33 | 0.33 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.33 | 1 | 0.33 |
JE c740 <hypre_ParMatmul.extracted.12+0x3f0> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
MOV -0x58(%RBP),%RCX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV (%RCX,%RDX,8),%RSI | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
LEA 0x1(%RDX),%R12 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
CMP 0x8(%RCX,%RDX,8),%RSI | 1 | 0.20 | 0.20 | 0.33 | 0.33 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.33 | 1 | 0.33 |
JGE c744 <hypre_ParMatmul.extracted.12+0x3f4> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
MOV %R12,-0x38(%RBP) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
MOV -0x30(%RBP),%RCX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV %RDX,-0x40(%RBP) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
JMP c5f9 <hypre_ParMatmul.extracted.12+0x2a9> | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 5.84 |
LEA 0x1(%RDX),%R12 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
MOV -0x30(%RBP),%RCX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
JMP c757 <hypre_ParMatmul.extracted.12+0x407> | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 5.84 |
MOV %R12,%RDX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
MOV -0x38(%RBP),%R12 | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV -0x60(%RBP),%RSI | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV (%RSI,%RDX,8),%RDX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
CMP (%RSI,%R12,8),%RDX | 1 | 0.20 | 0.20 | 0.33 | 0.33 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.33 | 1 | 0.33 |
JGE c570 <hypre_ParMatmul.extracted.12+0x220> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
MOV %R12,-0x38(%RBP) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
JMP c785 <hypre_ParMatmul.extracted.12+0x435> | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 5.84 |
Function | hypre_ParMatmul.extracted.12 |
Source file and lines | par_csr_matop.c:865-989 |
Module | libparcsr_mv.so |
nb instructions | 35 |
nb uops | 35 |
loop length | 149 |
used x86 registers | 9 |
used mmx registers | 0 |
used xmm registers | 0 |
used ymm registers | 0 |
used zmm registers | 0 |
nb stack references | 9 |
micro-operation queue | 6.17 cycles |
front end | 6.17 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 2.50 | 1.80 | 4.33 | 4.33 | 3.50 | 1.60 | 2.50 | 3.50 | 3.50 | 3.50 | 1.60 | 4.33 |
cycles | 2.50 | 1.80 | 4.33 | 4.33 | 3.50 | 1.60 | 2.50 | 3.50 | 3.50 | 3.50 | 1.60 | 4.33 |
Cycles executing div or sqrt instructions | NA |
FE+BE cycles | 6.32 |
Stall cycles | 0.00 |
Front-end | 6.17 |
Dispatch | 4.33 |
Overall L1 | 6.17 |
all | 0% |
load | 0% |
store | 0% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 0% |
all | 11% |
load | 12% |
store | 11% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 12% |
Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
MOV %RCX,-0x30(%RBP) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
MOV %RAX,%R14 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
MOV -0x48(%RBP),%RCX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
CMP %RCX,%R12 | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JGE c8cd <hypre_ParMatmul.extracted.12+0x57d> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
MOV %R12,%RDX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
CMPQ $0,-0xb8(%RBP) | 1 | 0.20 | 0.20 | 0.33 | 0.33 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.33 | 1 | 0.33 |
JE c5a9 <hypre_ParMatmul.extracted.12+0x259> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
MOV %R14,(%RDI,%RDX,8) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
MOVQ $0,(%R15,%R14,8) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
MOV -0x50(%RBP),%RAX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV %RDX,(%RAX,%R14,8) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
LEA 0x1(%R14),%RAX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
CMPQ $0,-0xb0(%RBP) | 1 | 0.20 | 0.20 | 0.33 | 0.33 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.33 | 1 | 0.33 |
JE c740 <hypre_ParMatmul.extracted.12+0x3f0> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
MOV -0x58(%RBP),%RCX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV (%RCX,%RDX,8),%RSI | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
LEA 0x1(%RDX),%R12 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
CMP 0x8(%RCX,%RDX,8),%RSI | 1 | 0.20 | 0.20 | 0.33 | 0.33 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.33 | 1 | 0.33 |
JGE c744 <hypre_ParMatmul.extracted.12+0x3f4> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
MOV %R12,-0x38(%RBP) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
MOV -0x30(%RBP),%RCX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV %RDX,-0x40(%RBP) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
JMP c5f9 <hypre_ParMatmul.extracted.12+0x2a9> | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 5.84 |
LEA 0x1(%RDX),%R12 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
MOV -0x30(%RBP),%RCX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
JMP c757 <hypre_ParMatmul.extracted.12+0x407> | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 5.84 |
MOV %R12,%RDX | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
MOV -0x38(%RBP),%R12 | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV -0x60(%RBP),%RSI | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
MOV (%RSI,%RDX,8),%RDX | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 1 | 0.33 |
CMP (%RSI,%R12,8),%RDX | 1 | 0.20 | 0.20 | 0.33 | 0.33 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.33 | 1 | 0.33 |
JGE c570 <hypre_ParMatmul.extracted.12+0x220> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
MOV %R12,-0x38(%RBP) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 |
JMP c785 <hypre_ParMatmul.extracted.12+0x435> | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 5.84 |