Loop Id: 219 | Module: libseq_mv.so | Source: vector.c:449-452 | Coverage: 1.23% |
---|
Loop Id: 219 | Module: libseq_mv.so | Source: vector.c:449-452 | Coverage: 1.23% |
---|
0xb640 VMOVUPD (%R8,%R10,8),%YMM1 [1] |
0xb646 VBROADCASTSD %XMM0,%YMM2 |
0xb64b VFMADD213PD (%RDI,%R10,8),%YMM1,%YMM2 [2] |
0xb651 VMOVUPD %YMM2,(%RDI,%R10,8) [2] |
0xb657 ADD $0x4,%R10 |
0xb65b CMP %R9,%R10 |
0xb65e JLE b640 |
/home/eoseret/qaas_runs_CPU_9468/171-147-2675/intel/AMG/build/AMG/AMG/seq_mv/vector.c: 449 - 452 |
-------------------------------------------------------------------------------- |
449: #pragma omp parallel for private(i) HYPRE_SMP_SCHEDULE |
450: #endif |
451: for (i = 0; i < size; i++) |
452: y_data[i] += alpha * x_data[i]; |
Path / |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.00 |
CQA speedup if FP arith vectorized | 1.17 |
CQA speedup if fully vectorized | 3.50 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.17 |
Bottlenecks | micro-operation queue, |
Function | hypre_SeqVectorAxpy.extracted |
Source | vector.c:449-452 |
Source loop unroll info | unrolled by 4 |
Source loop unroll confidence level | max |
Unroll/vectorization loop type | main |
Unroll factor | 4 |
CQA cycles | 1.17 |
CQA cycles if no scalar integer | 1.17 |
CQA cycles if FP arith vectorized | 1.00 |
CQA cycles if fully vectorized | 0.33 |
Front-end cycles | 1.17 |
DIV/SQRT cycles | 1.00 |
P0 cycles | 0.50 |
P1 cycles | 0.67 |
P2 cycles | 0.67 |
P3 cycles | 0.50 |
P4 cycles | 1.00 |
P5 cycles | 0.50 |
P6 cycles | 0.50 |
P7 cycles | 0.50 |
P8 cycles | 0.50 |
P9 cycles | 0.00 |
P10 cycles | 0.67 |
P11 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | 1.32 |
Stall cycles (UFS) | 0.00 |
Nb insns | 7.00 |
Nb uops | 6.00 |
Nb loads | 2.00 |
Nb stores | 1.00 |
Nb stack references | 0.00 |
FLOP/cycle | 6.86 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 0.00 |
Nb FLOP fma | 4.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 82.29 |
Bytes prefetched | 0.00 |
Bytes loaded | 64.00 |
Bytes stored | 32.00 |
Stride 0 | 0.00 |
Stride 1 | 2.00 |
Stride n | 0.00 |
Stride unknown | 0.00 |
Stride indirect | 0.00 |
Vectorization ratio all | 75.00 |
Vectorization ratio load | 100.00 |
Vectorization ratio store | 100.00 |
Vectorization ratio mul | NA |
Vectorization ratio add_sub | NA |
Vectorization ratio fma | 100.00 |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 0.00 |
Vector-efficiency ratio all | 40.63 |
Vector-efficiency ratio load | 50.00 |
Vector-efficiency ratio store | 50.00 |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | NA |
Vector-efficiency ratio fma | 50.00 |
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 | 1.17 |
CQA speedup if fully vectorized | 3.50 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.17 |
Bottlenecks | micro-operation queue, |
Function | hypre_SeqVectorAxpy.extracted |
Source | vector.c:449-452 |
Source loop unroll info | unrolled by 4 |
Source loop unroll confidence level | max |
Unroll/vectorization loop type | main |
Unroll factor | 4 |
CQA cycles | 1.17 |
CQA cycles if no scalar integer | 1.17 |
CQA cycles if FP arith vectorized | 1.00 |
CQA cycles if fully vectorized | 0.33 |
Front-end cycles | 1.17 |
DIV/SQRT cycles | 1.00 |
P0 cycles | 0.50 |
P1 cycles | 0.67 |
P2 cycles | 0.67 |
P3 cycles | 0.50 |
P4 cycles | 1.00 |
P5 cycles | 0.50 |
P6 cycles | 0.50 |
P7 cycles | 0.50 |
P8 cycles | 0.50 |
P9 cycles | 0.00 |
P10 cycles | 0.67 |
P11 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | 1.32 |
Stall cycles (UFS) | 0.00 |
Nb insns | 7.00 |
Nb uops | 6.00 |
Nb loads | 2.00 |
Nb stores | 1.00 |
Nb stack references | 0.00 |
FLOP/cycle | 6.86 |
Nb FLOP add-sub | 0.00 |
Nb FLOP mul | 0.00 |
Nb FLOP fma | 4.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 82.29 |
Bytes prefetched | 0.00 |
Bytes loaded | 64.00 |
Bytes stored | 32.00 |
Stride 0 | 0.00 |
Stride 1 | 2.00 |
Stride n | 0.00 |
Stride unknown | 0.00 |
Stride indirect | 0.00 |
Vectorization ratio all | 75.00 |
Vectorization ratio load | 100.00 |
Vectorization ratio store | 100.00 |
Vectorization ratio mul | NA |
Vectorization ratio add_sub | NA |
Vectorization ratio fma | 100.00 |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 0.00 |
Vector-efficiency ratio all | 40.63 |
Vector-efficiency ratio load | 50.00 |
Vector-efficiency ratio store | 50.00 |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | NA |
Vector-efficiency ratio fma | 50.00 |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 12.50 |
Path / |
Function | hypre_SeqVectorAxpy.extracted |
Source file and lines | vector.c:449-452 |
Module | libseq_mv.so |
nb instructions | 7 |
nb uops | 6 |
loop length | 32 |
used x86 registers | 4 |
used mmx registers | 0 |
used xmm registers | 1 |
used ymm registers | 2 |
used zmm registers | 0 |
nb stack references | 0 |
micro-operation queue | 1.17 cycles |
front end | 1.17 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 1.00 | 0.50 | 0.67 | 0.67 | 0.50 | 1.00 | 0.50 | 0.50 | 0.50 | 0.50 | 0.00 | 0.67 |
cycles | 1.00 | 0.50 | 0.67 | 0.67 | 0.50 | 1.00 | 0.50 | 0.50 | 0.50 | 0.50 | 0.00 | 0.67 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
FE+BE cycles | 1.32 |
Stall cycles | 0.00 |
Front-end | 1.17 |
Dispatch | 1.00 |
Data deps. | 1.00 |
Overall L1 | 1.17 |
all | 75% |
load | 100% |
store | 100% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 0% |
all | 40% |
load | 50% |
store | 50% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | 50% |
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 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
VMOVUPD (%R8,%R10,8),%YMM1 | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0-1 | 0.33 |
VBROADCASTSD %XMM0,%YMM2 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 |
VFMADD213PD (%RDI,%R10,8),%YMM1,%YMM2 | 1 | 0.50 | 0.50 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 4 | 0.50 |
VMOVUPD %YMM2,(%RDI,%R10,8) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0-1 | 0.50 |
ADD $0x4,%R10 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
CMP %R9,%R10 | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JLE b640 <hypre_SeqVectorAxpy.extracted+0xd0> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |
Function | hypre_SeqVectorAxpy.extracted |
Source file and lines | vector.c:449-452 |
Module | libseq_mv.so |
nb instructions | 7 |
nb uops | 6 |
loop length | 32 |
used x86 registers | 4 |
used mmx registers | 0 |
used xmm registers | 1 |
used ymm registers | 2 |
used zmm registers | 0 |
nb stack references | 0 |
micro-operation queue | 1.17 cycles |
front end | 1.17 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 1.00 | 0.50 | 0.67 | 0.67 | 0.50 | 1.00 | 0.50 | 0.50 | 0.50 | 0.50 | 0.00 | 0.67 |
cycles | 1.00 | 0.50 | 0.67 | 0.67 | 0.50 | 1.00 | 0.50 | 0.50 | 0.50 | 0.50 | 0.00 | 0.67 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
FE+BE cycles | 1.32 |
Stall cycles | 0.00 |
Front-end | 1.17 |
Dispatch | 1.00 |
Data deps. | 1.00 |
Overall L1 | 1.17 |
all | 75% |
load | 100% |
store | 100% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | 100% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 0% |
all | 40% |
load | 50% |
store | 50% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | NA (no add-sub vectorizable/vectorized instructions) |
fma | 50% |
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 |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
VMOVUPD (%R8,%R10,8),%YMM1 | 1 | 0 | 0 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0-1 | 0.33 |
VBROADCASTSD %XMM0,%YMM2 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 1 |
VFMADD213PD (%RDI,%R10,8),%YMM1,%YMM2 | 1 | 0.50 | 0.50 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 4 | 0.50 |
VMOVUPD %YMM2,(%RDI,%R10,8) | 1 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0 | 0 | 0-1 | 0.50 |
ADD $0x4,%R10 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.17 |
CMP %R9,%R10 | 1 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0.20 | 0 | 0 | 0 | 0.20 | 0 | 1 | 0.20 |
JLE b640 <hypre_SeqVectorAxpy.extracted+0xd0> | 1 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 |