Loop Id: 6544 | Module: exec | Source: csr_matvec.c:560-567 | Coverage: 0.01% |
---|
Loop Id: 6544 | Module: exec | Source: csr_matvec.c:560-567 | Coverage: 0.01% |
---|
0x578e80 KXNORW %K0,%K0,%K1 |
0x578e84 VXORPD %XMM7,%XMM7,%XMM7 |
0x578e88 VPSLLQ $0x3,%ZMM5,%ZMM6 |
0x578e8f ADD $0x8,%RAX |
0x578e93 VGATHERQPD (%RBX,%ZMM5,8),%ZMM7{%K1} |
0x578e9a VPADDQ %ZMM6,%ZMM0,%ZMM6 |
0x578ea0 KXNORW %K0,%K0,%K1 |
0x578ea4 VADDPD %ZMM4,%ZMM7,%ZMM4 |
0x578eaa VSCATTERQPD %ZMM4,(,%ZMM6,1){%K1} |
0x578eb5 CMP %RSI,%RAX |
0x578eb8 JA 578f30 |
0x578eba VBROADCASTSD (%R15,%R8,8),%ZMM4 |
0x578ec1 LEA (%R10,%RAX,1),%RDX |
0x578ec5 VPADDQ (%R14,%RDX,8),%ZMM1,%ZMM5 |
0x578ecc VPCONFLICTQ %ZMM5,%ZMM6 |
0x578ed2 VPTESTMQ %ZMM6,%ZMM6,%K1 |
0x578ed8 KORTESTB %K1,%K1 |
0x578edc VMULPD (%R12,%RDX,8),%ZMM4,%ZMM4 |
0x578ee3 JE 578e80 |
0x578ee5 VPLZCNTQ %ZMM6,%ZMM6 |
0x578eeb VPSUBQ %ZMM6,%ZMM2,%ZMM6 |
0x578ef1 NOPW %CS:(%RAX,%RAX,1) |
(6545) 0x578f00 VPERMQ %ZMM4,%ZMM6,%ZMM7 |
(6545) 0x578f06 VPERMQ %ZMM6,%ZMM6,%ZMM6{%K1} |
(6545) 0x578f0c VADDPD %ZMM7,%ZMM4,%ZMM4{%K1} |
(6545) 0x578f12 VPCMPNEQQ %ZMM3,%ZMM6,%K1 |
(6545) 0x578f19 KORTESTB %K1,%K1 |
(6545) 0x578f1d JNE 578f00 |
0x578f1f JMP 578e80 |
/home/eoseret/qaas_runs_CPU_9468/172-019-1763/intel/AMG/build/AMG/AMG/seq_mv/csr_matvec.c: 560 - 567 |
-------------------------------------------------------------------------------- |
560: #pragma omp for HYPRE_SMP_SCHEDULE |
561: #endif |
562: for (i = 0; i < num_rows; i++) |
563: { |
564: for (jj = A_i[i]; jj < A_i[i+1]; jj++) |
565: { |
566: j = A_j[jj]; |
567: y_data_expand[offset + j] += A_data[jj] * x_data[i]; |
Coverage (%) | Name | Source Location | Module |
---|---|---|---|
►100.00+ | __kmp_invoke_microtask | libiomp5.so | |
○ | __kmp_invoke_task_func | libiomp5.so |
Path / |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.05 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 1.04 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.34 |
Bottlenecks | |
Function | hypre_CSRMatrixMatvecT.extracted.49 |
Source | csr_matvec.c:560-560,csr_matvec.c:564-567 |
Source loop unroll info | NA |
Source loop unroll confidence level | NA |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 19.42 |
CQA cycles if no scalar integer | 18.50 |
CQA cycles if FP arith vectorized | 19.42 |
CQA cycles if fully vectorized | 18.60 |
Front-end cycles | 19.42 |
DIV/SQRT cycles | 1.25 |
P0 cycles | 1.00 |
P1 cycles | 1.00 |
P2 cycles | 1.00 |
P3 cycles | 1.25 |
P4 cycles | 1.67 |
P5 cycles | 1.67 |
P6 cycles | 1.67 |
P7 cycles | 9.50 |
P8 cycles | 8.96 |
P9 cycles | 9.25 |
P10 cycles | 9.29 |
P11 cycles | 14.50 |
P12 cycles | 14.50 |
P13 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | NA |
Stall cycles (UFS) | NA |
Nb insns | 21.00 |
Nb uops | 116.50 |
Nb loads | 4.00 |
Nb stores | 1.00 |
Nb stack references | 0.00 |
FLOP/cycle | 0.82 |
Nb FLOP add-sub | 8.00 |
Nb FLOP mul | 8.00 |
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 | 13.60 |
Bytes prefetched | 0.00 |
Bytes loaded | 200.00 |
Bytes stored | 64.00 |
Stride 0 | 1.00 |
Stride 1 | 2.00 |
Stride n | 0.00 |
Stride unknown | 0.00 |
Stride indirect | 2.00 |
Vectorization ratio all | 91.61 |
Vectorization ratio load | 75.00 |
Vectorization ratio store | 100.00 |
Vectorization ratio mul | 100.00 |
Vectorization ratio add_sub | 100.00 |
Vectorization ratio fma | NA |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 84.52 |
Vector-efficiency ratio all | 86.36 |
Vector-efficiency ratio load | 78.13 |
Vector-efficiency ratio store | 100.00 |
Vector-efficiency ratio mul | 100.00 |
Vector-efficiency ratio add_sub | 100.00 |
Vector-efficiency ratio fma | NA |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 74.85 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.05 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 1.04 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.36 |
Bottlenecks | micro-operation queue, |
Function | hypre_CSRMatrixMatvecT.extracted.49 |
Source | csr_matvec.c:560-560,csr_matvec.c:564-567 |
Source loop unroll info | NA |
Source loop unroll confidence level | NA |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 19.67 |
CQA cycles if no scalar integer | 18.67 |
CQA cycles if FP arith vectorized | 19.67 |
CQA cycles if fully vectorized | 18.92 |
Front-end cycles | 19.67 |
DIV/SQRT cycles | 1.50 |
P0 cycles | 1.00 |
P1 cycles | 1.00 |
P2 cycles | 1.00 |
P3 cycles | 1.50 |
P4 cycles | 1.67 |
P5 cycles | 1.67 |
P6 cycles | 1.67 |
P7 cycles | 10.00 |
P8 cycles | 9.08 |
P9 cycles | 9.50 |
P10 cycles | 9.42 |
P11 cycles | 14.50 |
P12 cycles | 14.50 |
P13 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | NA |
Stall cycles (UFS) | NA |
Nb insns | 23.00 |
Nb uops | 118.00 |
Nb loads | 4.00 |
Nb stores | 1.00 |
Nb stack references | 0.00 |
FLOP/cycle | 0.81 |
Nb FLOP add-sub | 8.00 |
Nb FLOP mul | 8.00 |
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 | 13.42 |
Bytes prefetched | 0.00 |
Bytes loaded | 200.00 |
Bytes stored | 64.00 |
Stride 0 | 1.00 |
Stride 1 | 2.00 |
Stride n | 0.00 |
Stride unknown | 0.00 |
Stride indirect | 2.00 |
Vectorization ratio all | 92.31 |
Vectorization ratio load | 75.00 |
Vectorization ratio store | 100.00 |
Vectorization ratio mul | 100.00 |
Vectorization ratio add_sub | 100.00 |
Vectorization ratio fma | NA |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 85.71 |
Vector-efficiency ratio all | 87.50 |
Vector-efficiency ratio load | 78.13 |
Vector-efficiency ratio store | 100.00 |
Vector-efficiency ratio mul | 100.00 |
Vector-efficiency ratio add_sub | 100.00 |
Vector-efficiency ratio fma | NA |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 76.79 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.05 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 1.05 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.32 |
Bottlenecks | micro-operation queue, |
Function | hypre_CSRMatrixMatvecT.extracted.49 |
Source | csr_matvec.c:560-560,csr_matvec.c:564-567 |
Source loop unroll info | NA |
Source loop unroll confidence level | NA |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 19.17 |
CQA cycles if no scalar integer | 18.33 |
CQA cycles if FP arith vectorized | 19.17 |
CQA cycles if fully vectorized | 18.28 |
Front-end cycles | 19.17 |
DIV/SQRT cycles | 1.00 |
P0 cycles | 1.00 |
P1 cycles | 1.00 |
P2 cycles | 1.00 |
P3 cycles | 1.00 |
P4 cycles | 1.67 |
P5 cycles | 1.67 |
P6 cycles | 1.67 |
P7 cycles | 9.00 |
P8 cycles | 8.83 |
P9 cycles | 9.00 |
P10 cycles | 9.17 |
P11 cycles | 14.50 |
P12 cycles | 14.50 |
P13 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | NA |
Stall cycles (UFS) | NA |
Nb insns | 19.00 |
Nb uops | 115.00 |
Nb loads | 4.00 |
Nb stores | 1.00 |
Nb stack references | 0.00 |
FLOP/cycle | 0.83 |
Nb FLOP add-sub | 8.00 |
Nb FLOP mul | 8.00 |
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 | 13.77 |
Bytes prefetched | 0.00 |
Bytes loaded | 200.00 |
Bytes stored | 64.00 |
Stride 0 | 1.00 |
Stride 1 | 2.00 |
Stride n | 0.00 |
Stride unknown | 0.00 |
Stride indirect | 2.00 |
Vectorization ratio all | 90.91 |
Vectorization ratio load | 75.00 |
Vectorization ratio store | 100.00 |
Vectorization ratio mul | 100.00 |
Vectorization ratio add_sub | 100.00 |
Vectorization ratio fma | NA |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 83.33 |
Vector-efficiency ratio all | 85.23 |
Vector-efficiency ratio load | 78.13 |
Vector-efficiency ratio store | 100.00 |
Vector-efficiency ratio mul | 100.00 |
Vector-efficiency ratio add_sub | 100.00 |
Vector-efficiency ratio fma | NA |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 72.92 |
Path / |
Function | hypre_CSRMatrixMatvecT.extracted.49 |
Source file and lines | csr_matvec.c:560-567 |
Module | exec |
nb instructions | 21 |
nb uops | 116.50 |
loop length | 117 |
used x86 registers | 9 |
used mmx registers | 0 |
used xmm registers | 1 |
used ymm registers | 0 |
used zmm registers | 6.50 |
nb stack references | 0 |
ADD-SUB / MUL ratio | 1.00 |
micro-operation queue | 19.42 cycles |
front end | 19.42 cycles |
ALU0/BRU0 | ALU1 | ALU2 | ALU3 | BRU1 | AGU0 | AGU1 | AGU2 | FP0 | FP1 | FP2 | FP3 | FP4 | FP5 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 1.25 | 1.00 | 1.00 | 1.00 | 1.25 | 1.00 | 1.00 | 1.00 | 8.50 | 8.96 | 8.75 | 8.79 | 14.50 | 14.50 |
cycles | 1.25 | 1.00 | 1.00 | 1.00 | 1.25 | 1.67 | 1.67 | 1.67 | 9.50 | 8.96 | 9.25 | 9.29 | 14.50 | 14.50 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
Front-end | 19.42 |
Dispatch | 14.50 |
Data deps. | 1.00 |
Overall L1 | 19.42 |
all | 100% |
load | 100% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 100% |
all | 83% |
load | 66% |
store | 100% |
mul | 100% |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 66% |
all | 91% |
load | 75% |
store | 100% |
mul | 100% |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 84% |
all | 100% |
load | 100% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 100% |
all | 72% |
load | 70% |
store | 100% |
mul | 100% |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 45% |
all | 86% |
load | 78% |
store | 100% |
mul | 100% |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 74% |
Function | hypre_CSRMatrixMatvecT.extracted.49 |
Source file and lines | csr_matvec.c:560-567 |
Module | exec |
nb instructions | 23 |
nb uops | 118 |
loop length | 133 |
used x86 registers | 9 |
used mmx registers | 0 |
used xmm registers | 1 |
used ymm registers | 0 |
used zmm registers | 7 |
nb stack references | 0 |
ADD-SUB / MUL ratio | 1.00 |
micro-operation queue | 19.67 cycles |
front end | 19.67 cycles |
ALU0/BRU0 | ALU1 | ALU2 | ALU3 | BRU1 | AGU0 | AGU1 | AGU2 | FP0 | FP1 | FP2 | FP3 | FP4 | FP5 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 1.50 | 1.00 | 1.00 | 1.00 | 1.50 | 1.00 | 1.00 | 1.00 | 9.00 | 9.08 | 9.00 | 8.92 | 14.50 | 14.50 |
cycles | 1.50 | 1.00 | 1.00 | 1.00 | 1.50 | 1.67 | 1.67 | 1.67 | 10.00 | 9.08 | 9.50 | 9.42 | 14.50 | 14.50 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
Front-end | 19.67 |
Dispatch | 14.50 |
Data deps. | 1.00 |
Overall L1 | 19.67 |
all | 100% |
load | 100% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 100% |
all | 83% |
load | 66% |
store | 100% |
mul | 100% |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 66% |
all | 92% |
load | 75% |
store | 100% |
mul | 100% |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 85% |
all | 100% |
load | 100% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 100% |
all | 72% |
load | 70% |
store | 100% |
mul | 100% |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 45% |
all | 87% |
load | 78% |
store | 100% |
mul | 100% |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 76% |
Instruction | Nb FU | ALU0/BRU0 | ALU1 | ALU2 | ALU3 | BRU1 | AGU0 | AGU1 | AGU2 | FP0 | FP1 | FP2 | FP3 | FP4 | FP5 | Latency | Recip. throughput | Vectorization |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
KXNORW %K0,%K0,%K1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 | N/A |
VXORPD %XMM7,%XMM7,%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 | vect (25.0%) |
VPSLLQ $0x3,%ZMM5,%ZMM6 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 1 | 1 | vect (100.0%) |
ADD $0x8,%RAX | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 | N/A |
VGATHERQPD (%RBX,%ZMM5,8),%ZMM7{%K1} | 48 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.75 | 4.58 | 5.08 | 3.58 | 5 | 5 | 0-16 | 8.94 | vect (100.0%) |
VPADDQ %ZMM6,%ZMM0,%ZMM6 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 1 | 0.50 | vect (100.0%) |
KXNORW %K0,%K0,%K1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 | N/A |
VADDPD %ZMM4,%ZMM7,%ZMM4 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 3 | 1 | vect (100.0%) |
VSCATTERQPD %ZMM4,(,%ZMM6,1){%K1} | 48 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 3.50 | 1.50 | 9 | 9 | 1-40 | 12.14 | vect (100.0%) |
CMP %RSI,%RAX | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 | N/A |
JA 578f30 <hypre_CSRMatrixMatvecT.extracted.49+0x890> | 1 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50-1 | N/A |
VBROADCASTSD (%R15,%R8,8),%ZMM4 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 5 | 1 | scal (12.5%) |
LEA (%R10,%RAX,1),%RDX | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 | N/A |
VPADDQ (%R14,%RDX,8),%ZMM1,%ZMM5 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 1 | 1 | vect (100.0%) |
VPCONFLICTQ %ZMM5,%ZMM6 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 1 | 0.50 | 0 | 0 | 0 | 7 | 1.33 | vect (100.0%) |
VPTESTMQ %ZMM6,%ZMM6,%K1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5 | 1 | vect (100.0%) |
KORTESTB %K1,%K1 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0.50 | 6 | 1 | N/A |
VMULPD (%R12,%RDX,8),%ZMM4,%ZMM4 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 3 | 1 | vect (100.0%) |
JE 578e80 <hypre_CSRMatrixMatvecT.extracted.49+0x7e0> | 1 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50-1 | N/A |
VPLZCNTQ %ZMM6,%ZMM6 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 2 | 1 | vect (100.0%) |
VPSUBQ %ZMM6,%ZMM2,%ZMM6 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 1 | 0.50 | vect (100.0%) |
NOPW %CS:(%RAX,%RAX,1) | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.09 | N/A |
JMP 578e80 <hypre_CSRMatrixMatvecT.extracted.49+0x7e0> | 1 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | N/A |
Function | hypre_CSRMatrixMatvecT.extracted.49 |
Source file and lines | csr_matvec.c:560-567 |
Module | exec |
nb instructions | 19 |
nb uops | 115 |
loop length | 101 |
used x86 registers | 9 |
used mmx registers | 0 |
used xmm registers | 1 |
used ymm registers | 0 |
used zmm registers | 6 |
nb stack references | 0 |
ADD-SUB / MUL ratio | 1.00 |
micro-operation queue | 19.17 cycles |
front end | 19.17 cycles |
ALU0/BRU0 | ALU1 | ALU2 | ALU3 | BRU1 | AGU0 | AGU1 | AGU2 | FP0 | FP1 | FP2 | FP3 | FP4 | FP5 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 8.00 | 8.83 | 8.50 | 8.67 | 14.50 | 14.50 |
cycles | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.67 | 1.67 | 1.67 | 9.00 | 8.83 | 9.00 | 9.17 | 14.50 | 14.50 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
Front-end | 19.17 |
Dispatch | 14.50 |
Data deps. | 1.00 |
Overall L1 | 19.17 |
all | 100% |
load | 100% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 100% |
all | 83% |
load | 66% |
store | 100% |
mul | 100% |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 66% |
all | 90% |
load | 75% |
store | 100% |
mul | 100% |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 83% |
all | 100% |
load | 100% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
other | 100% |
all | 72% |
load | 70% |
store | 100% |
mul | 100% |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 45% |
all | 85% |
load | 78% |
store | 100% |
mul | 100% |
add-sub | 100% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 72% |
Instruction | Nb FU | ALU0/BRU0 | ALU1 | ALU2 | ALU3 | BRU1 | AGU0 | AGU1 | AGU2 | FP0 | FP1 | FP2 | FP3 | FP4 | FP5 | Latency | Recip. throughput | Vectorization |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
KXNORW %K0,%K0,%K1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 | N/A |
VXORPD %XMM7,%XMM7,%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 | vect (25.0%) |
VPSLLQ $0x3,%ZMM5,%ZMM6 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 1 | 1 | vect (100.0%) |
ADD $0x8,%RAX | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 | N/A |
VGATHERQPD (%RBX,%ZMM5,8),%ZMM7{%K1} | 48 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.75 | 4.58 | 5.08 | 3.58 | 5 | 5 | 0-16 | 8.94 | vect (100.0%) |
VPADDQ %ZMM6,%ZMM0,%ZMM6 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 1 | 0.50 | vect (100.0%) |
KXNORW %K0,%K0,%K1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 1 | 0.50 | N/A |
VADDPD %ZMM4,%ZMM7,%ZMM4 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 3 | 1 | vect (100.0%) |
VSCATTERQPD %ZMM4,(,%ZMM6,1){%K1} | 48 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 3.50 | 1.50 | 9 | 9 | 1-40 | 12.14 | vect (100.0%) |
CMP %RSI,%RAX | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 | N/A |
JA 578f30 <hypre_CSRMatrixMatvecT.extracted.49+0x890> | 1 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50-1 | N/A |
VBROADCASTSD (%R15,%R8,8),%ZMM4 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 5 | 1 | scal (12.5%) |
LEA (%R10,%RAX,1),%RDX | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 | N/A |
VPADDQ (%R14,%RDX,8),%ZMM1,%ZMM5 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 1 | 1 | vect (100.0%) |
VPCONFLICTQ %ZMM5,%ZMM6 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 1 | 0.50 | 0 | 0 | 0 | 7 | 1.33 | vect (100.0%) |
VPTESTMQ %ZMM6,%ZMM6,%K1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 5 | 1 | vect (100.0%) |
KORTESTB %K1,%K1 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0.50 | 6 | 1 | N/A |
VMULPD (%R12,%RDX,8),%ZMM4,%ZMM4 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 3 | 1 | vect (100.0%) |
JE 578e80 <hypre_CSRMatrixMatvecT.extracted.49+0x7e0> | 1 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50-1 | N/A |