Loop Id: 109 | Module: exec | Source: ljForce.c:191-216 [...] | Coverage: 30.22% |
---|
Loop Id: 109 | Module: exec | Source: ljForce.c:191-216 [...] | Coverage: 30.22% |
---|
0x2104c0 INC %R8 |
0x2104c3 ADD $0x18,%RBX |
0x2104c7 CMP %R10,%R8 |
0x2104ca JGE 210480 |
0x2104cc VMOVUPD (%RSI),%XMM7 [3] |
0x2104d0 VSUBPD -0x10(%RBX),%XMM7,%XMM7 [10] |
0x2104d5 VMOVAPD %XMM7,%XMM8 |
0x2104d9 VFMADD213SD %XMM0,%XMM7,%XMM8 |
0x2104de VPERMILPD $0x1,%XMM7,%XMM9 |
0x2104e4 VFMADD213SD %XMM8,%XMM9,%XMM9 |
0x2104e9 VMOVSD 0x10(%RSI),%XMM8 [3] |
0x2104ee VSUBSD (%RBX),%XMM8,%XMM8 [10] |
0x2104f2 VFMADD231SD %XMM8,%XMM8,%XMM9 |
0x2104f7 VUCOMISD %XMM0,%XMM9 |
0x2104fb JBE 2104c0 |
0x2104fd VMOVSD (%R15),%XMM10 [1] |
0x210502 VUCOMISD %XMM9,%XMM10 |
0x210507 JB 2104c0 |
0x210509 VDIVSD %XMM9,%XMM1,%XMM9 |
0x21050e MOV 0x30(%R9),%R11 [8] |
0x210512 VMOVSD (%R11,%RAX,8),%XMM12 [2] |
0x210518 VMULSD %XMM9,%XMM9,%XMM10 |
0x21051d VMULSD %XMM10,%XMM9,%XMM10 |
0x210522 VMULSD (%R14),%XMM10,%XMM10 [4] |
0x210527 VADDSD %XMM2,%XMM10,%XMM11 |
0x21052b VFMSUB213SD (%RDX),%XMM10,%XMM11 [7] |
0x210530 VFMADD231SD %XMM3,%XMM11,%XMM12 |
0x210535 VMOVSD %XMM12,(%R11,%RAX,8) [2] |
0x21053b VFMADD213SD -0x38(%RBP),%XMM3,%XMM11 [9] |
0x210541 VMOVSD %XMM11,-0x38(%RBP) [9] |
0x210546 VMULSD (%RCX),%XMM4,%XMM11 [6] |
0x21054a MOV 0x28(%R9),%R11 [8] |
0x21054e VMULSD %XMM11,%XMM10,%XMM11 |
0x210553 VFMADD213SD %XMM6,%XMM5,%XMM10 |
0x210558 VMULSD %XMM11,%XMM9,%XMM9 |
0x21055d VMULSD %XMM9,%XMM10,%XMM9 |
0x210562 VMOVDDUP %XMM9,%XMM10 |
0x210567 VFNMADD213PD (%R11,%R13,8),%XMM7,%XMM10 [5] |
0x21056d VMOVUPD %XMM10,(%R11,%R13,8) [5] |
0x210573 VFNMADD213SD 0x10(%R11,%R13,8),%XMM9,%XMM8 [5] |
0x21057a VMOVSD %XMM8,0x10(%R11,%R13,8) [5] |
0x210581 JMP 2104c0 |
/beegfs/hackathon/users/eoseret/qaas_runs/170-850-7424/intel/CoMD/build/CoMD/CoMD/src-openmp/ljForce.c: 191 - 216 |
-------------------------------------------------------------------------------- |
191: for (int jOff=jBox*MAXATOMS; jOff<(jBox*MAXATOMS+nJBox); jOff++) |
[...] |
197: dr[m] = s->atoms->r[iOff][m]-s->atoms->r[jOff][m]; |
198: r2+=dr[m]*dr[m]; |
199: } |
200: |
201: if ( r2 <= rCut2 && r2 > 0.0) |
202: { |
203: |
204: // Important note: |
205: // from this point on r actually refers to 1.0/r |
206: r2 = 1.0/r2; |
207: real_t r6 = s6 * (r2*r2*r2); |
208: real_t eLocal = r6 * (r6 - 1.0) - eShift; |
209: s->atoms->U[iOff] += 0.5*eLocal; |
210: ePot += 0.5*eLocal; |
211: |
212: // different formulation to avoid sqrt computation |
213: real_t fr = - 4.0*epsilon*r6*r2*(12.0*r6 - 6.0); |
214: for (int m=0; m<3; m++) |
215: { |
216: s->atoms->f[iOff][m] -= dr[m]*fr; |
Path / |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.02 |
CQA speedup if FP arith vectorized | 1.71 |
CQA speedup if fully vectorized | 4.32 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.26 |
Bottlenecks | |
Function | .omp_outlined..5.120 |
Source | ljForce.c:191-191,ljForce.c:197-216 |
Source loop unroll info | not unrolled or unrolled with no peel/tail loop |
Source loop unroll confidence level | max |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 4.83 |
CQA cycles if no scalar integer | 4.72 |
CQA cycles if FP arith vectorized | 2.83 |
CQA cycles if fully vectorized | 1.12 |
Front-end cycles | 4.44 |
DIV/SQRT cycles | 1.50 |
P0 cycles | 1.00 |
P1 cycles | 1.00 |
P2 cycles | 1.00 |
P3 cycles | 1.50 |
P4 cycles | 3.00 |
P5 cycles | 3.00 |
P6 cycles | 3.00 |
P7 cycles | 3.83 |
P8 cycles | 4.17 |
P9 cycles | 2.56 |
P10 cycles | 2.44 |
P11 cycles | 1.50 |
P12 cycles | 1.50 |
P13 cycles | 1.67 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | NA |
Stall cycles (UFS) | NA |
Nb insns | 25.00 |
Nb uops | 26.67 |
Nb loads | 7.67 |
Nb stores | 1.33 |
Nb stack references | 0.33 |
FLOP/cycle | 3.45 |
Nb FLOP add-sub | 3.33 |
Nb FLOP mul | 2.33 |
Nb FLOP fma | 5.33 |
Nb FLOP div | 0.33 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 18.50 |
Bytes prefetched | 0.00 |
Bytes loaded | 80.00 |
Bytes stored | 13.33 |
Stride 0 | 3.33 |
Stride 1 | 0.00 |
Stride n | 1.00 |
Stride unknown | 0.00 |
Stride indirect | 0.67 |
Vectorization ratio all | 30.51 |
Vectorization ratio load | 38.33 |
Vectorization ratio store | 25.00 |
Vectorization ratio mul | 0.00 |
Vectorization ratio add_sub | 44.44 |
Vectorization ratio fma | 3.70 |
Vectorization ratio div_sqrt | 0.00 |
Vectorization ratio other | 52.22 |
Vector-efficiency ratio all | 16.31 |
Vector-efficiency ratio load | 17.29 |
Vector-efficiency ratio store | 15.63 |
Vector-efficiency ratio mul | 12.50 |
Vector-efficiency ratio add_sub | 18.06 |
Vector-efficiency ratio fma | 12.96 |
Vector-efficiency ratio div_sqrt | 12.50 |
Vector-efficiency ratio other | 19.03 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.07 |
CQA speedup if FP arith vectorized | 2.00 |
CQA speedup if fully vectorized | 5.12 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.45 |
Bottlenecks | micro-operation queue, |
Function | .omp_outlined..5.120 |
Source | ljForce.c:191-191,ljForce.c:197-216 |
Source loop unroll info | not unrolled or unrolled with no peel/tail loop |
Source loop unroll confidence level | max |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 2.67 |
CQA cycles if no scalar integer | 2.50 |
CQA cycles if FP arith vectorized | 1.33 |
CQA cycles if fully vectorized | 0.52 |
Front-end cycles | 2.67 |
DIV/SQRT cycles | 1.00 |
P0 cycles | 1.00 |
P1 cycles | 1.00 |
P2 cycles | 1.00 |
P3 cycles | 1.00 |
P4 cycles | 1.33 |
P5 cycles | 1.33 |
P6 cycles | 1.33 |
P7 cycles | 1.50 |
P8 cycles | 1.83 |
P9 cycles | 1.83 |
P10 cycles | 1.83 |
P11 cycles | 0.50 |
P12 cycles | 0.50 |
P13 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | NA |
Stall cycles (UFS) | NA |
Nb insns | 15.00 |
Nb uops | 16.00 |
Nb loads | 4.00 |
Nb stores | 0.00 |
Nb stack references | 0.00 |
FLOP/cycle | 3.37 |
Nb FLOP add-sub | 3.00 |
Nb FLOP mul | 0.00 |
Nb FLOP fma | 3.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 18.00 |
Bytes prefetched | 0.00 |
Bytes loaded | 48.00 |
Bytes stored | 0.00 |
Stride 0 | 1.00 |
Stride 1 | 0.00 |
Stride n | 1.00 |
Stride unknown | 0.00 |
Stride indirect | 0.00 |
Vectorization ratio all | 40.00 |
Vectorization ratio load | 50.00 |
Vectorization ratio store | NA |
Vectorization ratio mul | NA |
Vectorization ratio add_sub | 50.00 |
Vectorization ratio fma | 0.00 |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 66.67 |
Vector-efficiency ratio all | 17.50 |
Vector-efficiency ratio load | 18.75 |
Vector-efficiency ratio store | NA |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | 18.75 |
Vector-efficiency ratio fma | 12.50 |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 20.83 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.00 |
CQA speedup if FP arith vectorized | 1.55 |
CQA speedup if fully vectorized | 3.78 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.16 |
Bottlenecks | P8, P9, |
Function | .omp_outlined..5.120 |
Source | ljForce.c:191-191,ljForce.c:197-216 |
Source loop unroll info | not unrolled or unrolled with no peel/tail loop |
Source loop unroll confidence level | max |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 8.50 |
CQA cycles if no scalar integer | 8.50 |
CQA cycles if FP arith vectorized | 5.50 |
CQA cycles if fully vectorized | 2.25 |
Front-end cycles | 7.33 |
DIV/SQRT cycles | 2.00 |
P0 cycles | 1.00 |
P1 cycles | 1.00 |
P2 cycles | 1.00 |
P3 cycles | 2.00 |
P4 cycles | 6.00 |
P5 cycles | 6.00 |
P6 cycles | 6.00 |
P7 cycles | 8.50 |
P8 cycles | 8.50 |
P9 cycles | 3.50 |
P10 cycles | 3.50 |
P11 cycles | 3.00 |
P12 cycles | 3.00 |
P13 cycles | 5.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | NA |
Stall cycles (UFS) | NA |
Nb insns | 42.00 |
Nb uops | 44.00 |
Nb loads | 14.00 |
Nb stores | 4.00 |
Nb stack references | 1.00 |
FLOP/cycle | 3.76 |
Nb FLOP add-sub | 4.00 |
Nb FLOP mul | 7.00 |
Nb FLOP fma | 10.00 |
Nb FLOP div | 1.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 20.71 |
Bytes prefetched | 0.00 |
Bytes loaded | 136.00 |
Bytes stored | 40.00 |
Stride 0 | 7.00 |
Stride 1 | 0.00 |
Stride n | 1.00 |
Stride unknown | 0.00 |
Stride indirect | 2.00 |
Vectorization ratio all | 18.18 |
Vectorization ratio load | 25.00 |
Vectorization ratio store | 25.00 |
Vectorization ratio mul | 0.00 |
Vectorization ratio add_sub | 33.33 |
Vectorization ratio fma | 11.11 |
Vectorization ratio div_sqrt | 0.00 |
Vectorization ratio other | 40.00 |
Vector-efficiency ratio all | 14.77 |
Vector-efficiency ratio load | 15.63 |
Vector-efficiency ratio store | 15.63 |
Vector-efficiency ratio mul | 12.50 |
Vector-efficiency ratio add_sub | 16.67 |
Vector-efficiency ratio fma | 13.89 |
Vector-efficiency ratio div_sqrt | 12.50 |
Vector-efficiency ratio other | 17.50 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.05 |
CQA speedup if FP arith vectorized | 2.00 |
CQA speedup if fully vectorized | 5.71 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.43 |
Bottlenecks | micro-operation queue, |
Function | .omp_outlined..5.120 |
Source | ljForce.c:191-191,ljForce.c:197-216 |
Source loop unroll info | not unrolled or unrolled with no peel/tail loop |
Source loop unroll confidence level | max |
Unroll/vectorization loop type | NA |
Unroll factor | NA |
CQA cycles | 3.33 |
CQA cycles if no scalar integer | 3.17 |
CQA cycles if FP arith vectorized | 1.67 |
CQA cycles if fully vectorized | 0.58 |
Front-end cycles | 3.33 |
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 | 1.50 |
P8 cycles | 2.17 |
P9 cycles | 2.33 |
P10 cycles | 2.00 |
P11 cycles | 1.00 |
P12 cycles | 1.00 |
P13 cycles | 0.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | NA |
Stall cycles (UFS) | NA |
Nb insns | 18.00 |
Nb uops | 20.00 |
Nb loads | 5.00 |
Nb stores | 0.00 |
Nb stack references | 0.00 |
FLOP/cycle | 2.70 |
Nb FLOP add-sub | 3.00 |
Nb FLOP mul | 0.00 |
Nb FLOP fma | 3.00 |
Nb FLOP div | 0.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 16.80 |
Bytes prefetched | 0.00 |
Bytes loaded | 56.00 |
Bytes stored | 0.00 |
Stride 0 | 2.00 |
Stride 1 | 0.00 |
Stride n | 1.00 |
Stride unknown | 0.00 |
Stride indirect | 0.00 |
Vectorization ratio all | 33.33 |
Vectorization ratio load | 40.00 |
Vectorization ratio store | NA |
Vectorization ratio mul | NA |
Vectorization ratio add_sub | 50.00 |
Vectorization ratio fma | 0.00 |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 50.00 |
Vector-efficiency ratio all | 16.67 |
Vector-efficiency ratio load | 17.50 |
Vector-efficiency ratio store | NA |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | 18.75 |
Vector-efficiency ratio fma | 12.50 |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 18.75 |
Path / |
Function | .omp_outlined..5.120 |
Source file and lines | ljForce.c:191-216 |
Module | exec |
nb instructions | 25 |
nb uops | 26.67 |
loop length | 110.67 |
used x86 registers | 7.33 |
used mmx registers | 0 |
used xmm registers | 7.33 |
used ymm registers | 0 |
used zmm registers | 0 |
nb stack references | 0.33 |
micro-operation queue | 4.44 cycles |
front end | 4.44 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 | 3.00 | 3.00 | 3.00 | 3.83 | 4.17 | 2.56 | 2.44 | 1.50 | 1.50 |
cycles | 1.50 | 1.00 | 1.00 | 1.00 | 1.50 | 3.00 | 3.00 | 3.00 | 3.83 | 4.17 | 2.56 | 2.44 | 1.50 | 1.50 |
Cycles executing div or sqrt instructions | 1.67 |
Longest recurrence chain latency (RecMII) | 1.00 |
Front-end | 4.44 |
Dispatch | 4.22 |
DIV/SQRT | 1.67 |
Data deps. | 1.00 |
Overall L1 | 4.83 |
all | 30% |
load | 38% |
store | 25% |
mul | 0% |
add-sub | 44% |
fma | 3% |
div/sqrt | 0% |
other | 52% |
all | 16% |
load | 17% |
store | 15% |
mul | 12% |
add-sub | 18% |
fma | 12% |
div/sqrt | 12% |
other | 19% |
Function | .omp_outlined..5.120 |
Source file and lines | ljForce.c:191-216 |
Module | exec |
nb instructions | 15 |
nb uops | 16 |
loop length | 61 |
used x86 registers | 4 |
used mmx registers | 0 |
used xmm registers | 4 |
used ymm registers | 0 |
used zmm registers | 0 |
nb stack references | 0 |
micro-operation queue | 2.67 cycles |
front end | 2.67 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 | 1.33 | 1.33 | 1.33 | 1.50 | 1.83 | 1.83 | 1.83 | 0.50 | 0.50 |
cycles | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 | 1.33 | 1.33 | 1.33 | 1.50 | 1.83 | 1.83 | 1.83 | 0.50 | 0.50 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
Front-end | 2.67 |
Dispatch | 1.83 |
Data deps. | 1.00 |
Overall L1 | 2.67 |
all | 40% |
load | 50% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 50% |
fma | 0% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 66% |
all | 17% |
load | 18% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 18% |
fma | 12% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 20% |
Instruction | Nb FU | ALU0/BRU0 | ALU1 | ALU2 | ALU3 | BRU1 | AGU0 | AGU1 | AGU2 | FP0 | FP1 | FP2 | FP3 | FP4 | FP5 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
INC %R8 | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
ADD $0x18,%RBX | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
CMP %R10,%R8 | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
JGE 210480 <.omp_outlined..5.120+0x270> | 1 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50-1 |
VMOVUPD (%RSI),%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 0.50 |
VSUBPD -0x10(%RBX),%XMM7,%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 3 | 0.50 |
VMOVAPD %XMM7,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
VFMADD213SD %XMM0,%XMM7,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VPERMILPD $0x1,%XMM7,%XMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 1 | 0.33 |
VFMADD213SD %XMM8,%XMM9,%XMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VMOVSD 0x10(%RSI),%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 |
VSUBSD (%RBX),%XMM8,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 3 | 0.50 |
VFMADD231SD %XMM8,%XMM8,%XMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VUCOMISD %XMM0,%XMM9 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0.50 | 6 | 1 |
JBE 2104c0 <.omp_outlined..5.120+0x2b0> | 1 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50-1 |
Function | .omp_outlined..5.120 |
Source file and lines | ljForce.c:191-216 |
Module | exec |
nb instructions | 42 |
nb uops | 44 |
loop length | 198 |
used x86 registers | 13 |
used mmx registers | 0 |
used xmm registers | 13 |
used ymm registers | 0 |
used zmm registers | 0 |
nb stack references | 1 |
ADD-SUB / MUL ratio | 0.43 |
micro-operation queue | 7.33 cycles |
front end | 7.33 cycles |
ALU0/BRU0 | ALU1 | ALU2 | ALU3 | BRU1 | AGU0 | AGU1 | AGU2 | FP0 | FP1 | FP2 | FP3 | FP4 | FP5 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 2.00 | 1.00 | 1.00 | 1.00 | 2.00 | 6.00 | 6.00 | 6.00 | 8.50 | 8.50 | 3.50 | 3.50 | 3.00 | 3.00 |
cycles | 2.00 | 1.00 | 1.00 | 1.00 | 2.00 | 6.00 | 6.00 | 6.00 | 8.50 | 8.50 | 3.50 | 3.50 | 3.00 | 3.00 |
Cycles executing div or sqrt instructions | 5.00 |
Longest recurrence chain latency (RecMII) | 1.00 |
Front-end | 7.33 |
Dispatch | 8.50 |
DIV/SQRT | 5.00 |
Data deps. | 1.00 |
Overall L1 | 8.50 |
all | 18% |
load | 25% |
store | 25% |
mul | 0% |
add-sub | 33% |
fma | 11% |
div/sqrt | 0% |
other | 40% |
all | 14% |
load | 15% |
store | 15% |
mul | 12% |
add-sub | 16% |
fma | 13% |
div/sqrt | 12% |
other | 17% |
Instruction | Nb FU | ALU0/BRU0 | ALU1 | ALU2 | ALU3 | BRU1 | AGU0 | AGU1 | AGU2 | FP0 | FP1 | FP2 | FP3 | FP4 | FP5 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
INC %R8 | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
ADD $0x18,%RBX | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
CMP %R10,%R8 | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
JGE 210480 <.omp_outlined..5.120+0x270> | 1 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50-1 |
VMOVUPD (%RSI),%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 0.50 |
VSUBPD -0x10(%RBX),%XMM7,%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 3 | 0.50 |
VMOVAPD %XMM7,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
VFMADD213SD %XMM0,%XMM7,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VPERMILPD $0x1,%XMM7,%XMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 1 | 0.33 |
VFMADD213SD %XMM8,%XMM9,%XMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VMOVSD 0x10(%RSI),%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 |
VSUBSD (%RBX),%XMM8,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 3 | 0.50 |
VFMADD231SD %XMM8,%XMM8,%XMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VUCOMISD %XMM0,%XMM9 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0.50 | 6 | 1 |
JBE 2104c0 <.omp_outlined..5.120+0x2b0> | 1 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50-1 |
VMOVSD (%R15),%XMM10 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 |
VUCOMISD %XMM9,%XMM10 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0.50 | 6 | 1 |
JB 2104c0 <.omp_outlined..5.120+0x2b0> | 1 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50-1 |
VDIVSD %XMM9,%XMM1,%XMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 13 | 5 |
MOV 0x30(%R9),%R11 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 0.33 |
VMOVSD (%R11,%RAX,8),%XMM12 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 |
VMULSD %XMM9,%XMM9,%XMM10 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 3 | 0.50 |
VMULSD %XMM10,%XMM9,%XMM10 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 3 | 0.50 |
VMULSD (%R14),%XMM10,%XMM10 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 3 | 0.50 |
VADDSD %XMM2,%XMM10,%XMM11 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 3 | 0.50 |
VFMSUB213SD (%RDX),%XMM10,%XMM11 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VFMADD231SD %XMM3,%XMM11,%XMM12 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VMOVSD %XMM12,(%R11,%RAX,8) | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 1 | 1 |
VFMADD213SD -0x38(%RBP),%XMM3,%XMM11 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VMOVSD %XMM11,-0x38(%RBP) | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 1 | 1 |
VMULSD (%RCX),%XMM4,%XMM11 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 3 | 0.50 |
MOV 0x28(%R9),%R11 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 0.33 |
VMULSD %XMM11,%XMM10,%XMM11 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 3 | 0.50 |
VFMADD213SD %XMM6,%XMM5,%XMM10 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VMULSD %XMM11,%XMM9,%XMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 3 | 0.50 |
VMULSD %XMM9,%XMM10,%XMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 3 | 0.50 |
VMOVDDUP %XMM9,%XMM10 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 1 | 0.33 |
VFNMADD213PD (%R11,%R13,8),%XMM7,%XMM10 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VMOVUPD %XMM10,(%R11,%R13,8) | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 4 | 1 |
VFNMADD213SD 0x10(%R11,%R13,8),%XMM9,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VMOVSD %XMM8,0x10(%R11,%R13,8) | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 1 | 1 |
JMP 2104c0 <.omp_outlined..5.120+0x2b0> | 1 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
Function | .omp_outlined..5.120 |
Source file and lines | ljForce.c:191-216 |
Module | exec |
nb instructions | 18 |
nb uops | 20 |
loop length | 73 |
used x86 registers | 5 |
used mmx registers | 0 |
used xmm registers | 5 |
used ymm registers | 0 |
used zmm registers | 0 |
nb stack references | 0 |
micro-operation queue | 3.33 cycles |
front end | 3.33 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 | 1.67 | 1.67 | 1.67 | 1.50 | 2.17 | 2.33 | 2.00 | 1.00 | 1.00 |
cycles | 1.50 | 1.00 | 1.00 | 1.00 | 1.50 | 1.67 | 1.67 | 1.67 | 1.50 | 2.17 | 2.33 | 2.00 | 1.00 | 1.00 |
Cycles executing div or sqrt instructions | NA |
Longest recurrence chain latency (RecMII) | 1.00 |
Front-end | 3.33 |
Dispatch | 2.33 |
Data deps. | 1.00 |
Overall L1 | 3.33 |
all | 33% |
load | 40% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 50% |
fma | 0% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 50% |
all | 16% |
load | 17% |
store | NA (no store vectorizable/vectorized instructions) |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 18% |
fma | 12% |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 18% |
Instruction | Nb FU | ALU0/BRU0 | ALU1 | ALU2 | ALU3 | BRU1 | AGU0 | AGU1 | AGU2 | FP0 | FP1 | FP2 | FP3 | FP4 | FP5 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
INC %R8 | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
ADD $0x18,%RBX | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
CMP %R10,%R8 | 1 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
JGE 210480 <.omp_outlined..5.120+0x270> | 1 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50-1 |
VMOVUPD (%RSI),%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 3 | 0.50 |
VSUBPD -0x10(%RBX),%XMM7,%XMM7 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 3 | 0.50 |
VMOVAPD %XMM7,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.17 |
VFMADD213SD %XMM0,%XMM7,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VPERMILPD $0x1,%XMM7,%XMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 1 | 0.33 |
VFMADD213SD %XMM8,%XMM9,%XMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VMOVSD 0x10(%RSI),%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 |
VSUBSD (%RBX),%XMM8,%XMM8 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 3 | 0.50 |
VFMADD231SD %XMM8,%XMM8,%XMM9 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 4 | 0.50 |
VUCOMISD %XMM0,%XMM9 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0.50 | 6 | 1 |
JBE 2104c0 <.omp_outlined..5.120+0x2b0> | 1 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50-1 |
VMOVSD (%R15),%XMM10 | 1 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 |
VUCOMISD %XMM9,%XMM10 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0.50 | 0.50 | 6 | 1 |
JB 2104c0 <.omp_outlined..5.120+0x2b0> | 1 | 0.50 | 0 | 0 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50-1 |