Loop Id: 544 | Module: exec | Source: advec_mom_kernel.f90-pp.f90:215-241 [...] | Coverage: 7.71% |
---|
Loop Id: 544 | Module: exec | Source: advec_mom_kernel.f90-pp.f90:215-241 [...] | Coverage: 7.71% |
---|
0x41fd2c FMSUB D2, D2, D6, D6 |
0x41fd30 SUB W3, W3, #1 |
0x41fd34 CMP W3, #1 |
0x41fd38 FADD D2, D2, D3 |
0x41fd3c FMUL D1, D2, D1 |
0x41fd40 STR D1, [X13, X4,LSL #3] [9] |
0x41fd44 ADD X4, X4, #1 |
0x41fd48 B.LE 41fd00 |
0x41fd4c LDR D1, [X16, X4,LSL #3] [2] |
0x41fd50 MOVI D6, #0 |
0x41fd54 FCMP D1, #0 |
0x41fd58 FABS D2, D1 |
0x41fd5c CSINC W6, W8, W8, #10 |
0x41fd60 CSEL W5, W1, W2, #10 |
0x41fd64 SBFM X6, X6, #0, #31 |
0x41fd68 SBFM X5, X5, #0, #31 |
0x41fd6c MADD X6, X10, X6, X4 |
0x41fd70 MADD X5, X10, X5, X4 |
0x41fd74 LDR D3, [X11, X6,LSL #3] [1] |
0x41fd78 LDR D4, [X12, X5,LSL #3] [4] |
0x41fd7c CSINC X5, X8, X8, #11 |
0x41fd80 MADD X5, X10, X5, X4 |
0x41fd84 FDIV D2, D2, D3 |
0x41fd88 LDR D3, [X12, X6,LSL #3] [6] |
0x41fd8c FSUB D5, D3, S4 |
0x41fd90 LDR D4, [X12, X5,LSL #3] [3] |
0x41fd94 FSUB D4, D4, S3 |
0x41fd98 FMUL D7, D4, D5 |
0x41fd9c FCMP D7, #0 |
0x41fda0 B.LE 41fd2c |
0x41fda4 LDP X6, X7, [X20, #312] [8] |
0x41fda8 FCMP D1, #0 |
0x41fdac FABS D5, D5 |
0x41fdb0 FABS D7, D4 |
0x41fdb4 CSEL W5, W1, W0, #10 |
0x41fdb8 FMADD D16, D2, D5, D5 |
0x41fdbc FCMP D4, #0 |
0x41fdc0 SBFM X5, X5, #0, #31 |
0x41fdc4 LDR X7, [X7] [7] |
0x41fdc8 SUB X5, X5, X7 |
0x41fdcc SUB X19, X8, X7 |
0x41fdd0 LDR D17, [X6, X5,LSL #3] [5] |
0x41fdd4 LDR D6, [X6, X19,LSL #3] [10] |
0x41fdd8 FDIV D16, D16, D17 |
0x41fddc FSUB D17, D0, S2 |
0x41fde0 FMUL D17, D7, D17 |
0x41fde4 FDIV D17, D17, D6 |
0x41fde8 FADD D16, D16, D17 |
0x41fdec FMOV D17, X18 |
0x41fdf0 FMUL D6, D6, D17 |
0x41fdf4 FMUL D6, D6, D16 |
0x41fdf8 FMINNM D5, D6, D5 |
0x41fdfc FMINNM D5, D5, D7 |
0x41fe00 FNEG D6, D5 |
0x41fe04 FCSEL D6, D5, D6, #12 |
0x41fe08 B 41fd2c |
/home/hbollore/qaas-runs/170-307-1706/intel/CloverLeafFC/build/armclang_5/CMakeFiles/clover_leaf.dir/CloverLeaf_ref/kernels/advec_mom_kernel.f90-pp.f90: 215 - 241 |
-------------------------------------------------------------------------------- |
215: DO j=x_min,x_max+1 |
[...] |
227: # 227 "/home/hbollore/qaas-runs/170-307-1706/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/advec_mom_kernel.f90" |
228: sigma=ABS(node_flux(j,k))/(node_mass_pre(j,donor)) |
229: width=celldy(k) |
230: vdiffuw=vel1(j,donor)-vel1(j,upwind) |
231: vdiffdw=vel1(j,downwind)-vel1(j,donor) |
232: limiter=0.0 |
233: IF(vdiffuw*vdiffdw.GT.0.0)THEN |
234: auw=ABS(vdiffuw) |
235: adw=ABS(vdiffdw) |
236: wind=1.0_8 |
237: IF(vdiffdw.LE.0.0) wind=-1.0_8 |
238: limiter=wind*MIN(width*((2.0_8-sigma)*adw/width+(1.0_8+sigma)*auw/celldy(dif))/6.0_8,auw,adw) |
239: ENDIF |
240: advec_vel_s=vel1(j,donor)+(1.0_8-sigma)*limiter |
241: mom_flux(j,k)=advec_vel_s*node_flux(j,k) |
Path / |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 2.20 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 5.33 - 7.04 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.10 |
Bottlenecks | |
Function | advec_mom_kernel |
Source | advec_mom_kernel.f90-pp.f90:215-215,advec_mom_kernel.f90-pp.f90:227-241 |
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 | 5.50 |
CQA cycles if no scalar integer | 2.50 |
CQA cycles if FP arith vectorized | 5.50 |
CQA cycles if fully vectorized | 1.03 - 0.78 |
Front-end cycles | 5.38 |
DIV/SQRT cycles | 1.25 |
P0 cycles | 1.25 |
P1 cycles | 3.25 |
P2 cycles | 3.25 |
P3 cycles | 3.38 |
P4 cycles | 3.13 |
P5 cycles | 5.13 |
P6 cycles | 5.13 |
P7 cycles | 5.13 |
P8 cycles | 5.13 |
P9 cycles | 2.83 |
P10 cycles | 2.50 |
P11 cycles | 2.67 |
P12 cycles | 0.00 |
P13 cycles | 0.00 |
P14 cycles | 2.00 - 1.00 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | NA |
Stall cycles (UFS) | NA |
Nb insns | 43.00 |
Nb uops | 43.00 |
Nb loads | NA |
Nb stores | 1.00 |
Nb stack references | 0.00 |
FLOP/cycle | 2.27 |
Nb FLOP add-sub | 4.00 |
Nb FLOP mul | 3.50 |
Nb FLOP fma | 1.50 |
Nb FLOP div | 2.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 12.47 |
Bytes prefetched | 0.00 |
Bytes loaded | 60.00 |
Bytes stored | 8.00 |
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 | 0.00 |
Vectorization ratio fma | NA |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 0.00 |
Vector-efficiency ratio all | 22.47 |
Vector-efficiency ratio load | 25.00 |
Vector-efficiency ratio store | 25.00 |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | 12.50 |
Vector-efficiency ratio fma | NA |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 18.75 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 2.56 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 4.83 - 7.25 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.04 |
Bottlenecks | P6, P7, P8, P9, |
Function | advec_mom_kernel |
Source | advec_mom_kernel.f90-pp.f90:215-215,advec_mom_kernel.f90-pp.f90:227-241 |
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 | 7.25 |
CQA cycles if no scalar integer | 2.83 |
CQA cycles if FP arith vectorized | 7.25 |
CQA cycles if fully vectorized | 1.50 - 1.00 |
Front-end cycles | 7.00 |
DIV/SQRT cycles | 1.50 |
P0 cycles | 1.50 |
P1 cycles | 3.75 |
P2 cycles | 3.75 |
P3 cycles | 3.75 |
P4 cycles | 3.75 |
P5 cycles | 7.25 |
P6 cycles | 7.25 |
P7 cycles | 7.25 |
P8 cycles | 7.25 |
P9 cycles | 3.50 |
P10 cycles | 3.17 |
P11 cycles | 3.33 |
P12 cycles | 0.00 |
P13 cycles | 0.00 |
P14 cycles | 3.00 - 1.50 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | NA |
Stall cycles (UFS) | NA |
Nb insns | 56.00 |
Nb uops | 56.00 |
Nb loads | NA |
Nb stores | 1.00 |
Nb stack references | 0.00 |
FLOP/cycle | 2.34 |
Nb FLOP add-sub | 5.00 |
Nb FLOP mul | 5.00 |
Nb FLOP fma | 2.00 |
Nb FLOP div | 3.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 12.14 |
Bytes prefetched | 0.00 |
Bytes loaded | 80.00 |
Bytes stored | 8.00 |
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 | 0.00 |
Vectorization ratio fma | NA |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 0.00 |
Vector-efficiency ratio all | 22.73 |
Vector-efficiency ratio load | 25.00 |
Vector-efficiency ratio store | 25.00 |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | 12.50 |
Vector-efficiency ratio fma | NA |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 18.75 |
Metric | Value |
---|---|
CQA speedup if no scalar integer | 1.73 |
CQA speedup if FP arith vectorized | 1.00 |
CQA speedup if fully vectorized | 6.67 |
CQA speedup if no inter-iteration dependency | NA |
CQA speedup if next bottleneck killed | 1.25 |
Bottlenecks | micro-operation queue, |
Function | advec_mom_kernel |
Source | advec_mom_kernel.f90-pp.f90:215-215,advec_mom_kernel.f90-pp.f90:227-241 |
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 | 3.75 |
CQA cycles if no scalar integer | 2.17 |
CQA cycles if FP arith vectorized | 3.75 |
CQA cycles if fully vectorized | 0.56 |
Front-end cycles | 3.75 |
DIV/SQRT cycles | 1.00 |
P0 cycles | 1.00 |
P1 cycles | 2.75 |
P2 cycles | 2.75 |
P3 cycles | 3.00 |
P4 cycles | 2.50 |
P5 cycles | 3.00 |
P6 cycles | 3.00 |
P7 cycles | 3.00 |
P8 cycles | 3.00 |
P9 cycles | 2.17 |
P10 cycles | 1.83 |
P11 cycles | 2.00 |
P12 cycles | 0.00 |
P13 cycles | 0.00 |
P14 cycles | 1.00 - 0.50 |
Inter-iter dependencies cycles | 1 |
FE+BE cycles (UFS) | NA |
Stall cycles (UFS) | NA |
Nb insns | 30.00 |
Nb uops | 30.00 |
Nb loads | NA |
Nb stores | 1.00 |
Nb stack references | 0.00 |
FLOP/cycle | 2.13 |
Nb FLOP add-sub | 3.00 |
Nb FLOP mul | 2.00 |
Nb FLOP fma | 1.00 |
Nb FLOP div | 1.00 |
Nb FLOP rcp | 0.00 |
Nb FLOP sqrt | 0.00 |
Nb FLOP rsqrt | 0.00 |
Bytes/cycle | 12.80 |
Bytes prefetched | 0.00 |
Bytes loaded | 40.00 |
Bytes stored | 8.00 |
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 | 0.00 |
Vectorization ratio fma | NA |
Vectorization ratio div_sqrt | NA |
Vectorization ratio other | 0.00 |
Vector-efficiency ratio all | 22.22 |
Vector-efficiency ratio load | 25.00 |
Vector-efficiency ratio store | 25.00 |
Vector-efficiency ratio mul | NA |
Vector-efficiency ratio add_sub | 12.50 |
Vector-efficiency ratio fma | NA |
Vector-efficiency ratio div_sqrt | NA |
Vector-efficiency ratio other | 18.75 |
Path / |
Function | advec_mom_kernel |
Source file and lines | advec_mom_kernel.f90-pp.f90:215-241 |
Module | exec |
nb instructions | 43 |
loop length | 172 |
nb stack references | 0 |
front end | 5.38 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | P12 | P13 | P14 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 1.25 | 1.25 | 3.25 | 3.25 | 3.38 | 3.13 | 5.13 | 5.13 | 5.13 | 5.13 | 2.83 | 2.50 | 2.67 | 0.00 | 0.00 |
cycles | 1.25 | 1.25 | 3.25 | 3.25 | 3.38 | 3.13 | 5.13 | 5.13 | 5.13 | 5.13 | 2.83 | 2.50 | 2.67 | 0.00 | 0.00 |
Cycles executing div or sqrt instructions | 2.00-1.00 |
Longest recurrence chain latency (RecMII) | 1.00 |
Front-end | 5.38 |
Data deps. | 1.00 |
Overall L1 | 5.50 |
all | 0% |
load | 0% |
store | 0% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 0% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 0% |
Function | advec_mom_kernel |
Source file and lines | advec_mom_kernel.f90-pp.f90:215-241 |
Module | exec |
nb instructions | 56 |
loop length | 224 |
nb stack references | 0 |
front end | 7.00 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | P12 | P13 | P14 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 1.50 | 1.50 | 3.75 | 3.75 | 3.75 | 3.75 | 7.25 | 7.25 | 7.25 | 7.25 | 3.50 | 3.17 | 3.33 | 0.00 | 0.00 |
cycles | 1.50 | 1.50 | 3.75 | 3.75 | 3.75 | 3.75 | 7.25 | 7.25 | 7.25 | 7.25 | 3.50 | 3.17 | 3.33 | 0.00 | 0.00 |
Cycles executing div or sqrt instructions | 3.00-1.50 |
Longest recurrence chain latency (RecMII) | 1.00 |
Front-end | 7.00 |
Data deps. | 1.00 |
Overall L1 | 7.25 |
all | 0% |
load | 0% |
store | 0% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 0% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 0% |
Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | P12 | P13 | P14 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
FMSUB D2, D2, D6, D6 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 4 | 0.25 |
SUB W3, W3, #1 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
CMP W3, #1 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.33 |
FADD D2, D2, D3 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
FMUL D1, D2, D1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 3 | 0.25 |
STR D1, [X13, X4,LSL #3] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 2 | 0.50 |
ADD X4, X4, #1 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
B.LE 41fd00 <__nv_advec_mom_kernel_mod_advec_mom_kernel__F1L79_1_+0x18e0> | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 |
LDR D1, [X16, X4,LSL #3] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 6 | 0.33 |
MOVI D6, #0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
FCMP D1, #0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
FABS D2, D1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
CSINC W6, W8, W8, #10 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
CSEL W5, W1, W2, #10 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
SBFM X6, X6, #0, #31 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
SBFM X5, X5, #0, #31 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
MADD X6, X10, X6, X4 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
MADD X5, X10, X5, X4 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
LDR D3, [X11, X6,LSL #3] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 6 | 0.33 |
LDR D4, [X12, X5,LSL #3] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 6 | 0.33 |
CSINC X5, X8, X8, #11 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
MADD X5, X10, X5, X4 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
FDIV D2, D2, D3 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 7-15 | 1-0.50 |
LDR D3, [X12, X6,LSL #3] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 6 | 0.33 |
FSUB D5, D3, S4 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
LDR D4, [X12, X5,LSL #3] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 6 | 0.33 |
FSUB D4, D4, S3 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
FMUL D7, D4, D5 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 3 | 0.25 |
FCMP D7, #0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
B.LE 41fd2c <__nv_advec_mom_kernel_mod_advec_mom_kernel__F1L79_1_+0x190c> | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 |
LDP X6, X7, [X20, #312] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 4 | 1 |
FCMP D1, #0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
FABS D5, D5 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
FABS D7, D4 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
CSEL W5, W1, W0, #10 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
FMADD D16, D2, D5, D5 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 4 | 0.25 |
FCMP D4, #0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
SBFM X5, X5, #0, #31 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
LDR X7, [X7] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 4 | 0.33 |
SUB X5, X5, X7 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
SUB X19, X8, X7 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
LDR D17, [X6, X5,LSL #3] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 6 | 0.33 |
LDR D6, [X6, X19,LSL #3] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 6 | 0.33 |
FDIV D16, D16, D17 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 7-15 | 1-0.50 |
FSUB D17, D0, S2 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
FMUL D17, D7, D17 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 3 | 0.25 |
FDIV D17, D17, D6 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 7-15 | 1-0.50 |
FADD D16, D16, D17 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
FMOV D17, X18 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
FMUL D6, D6, D17 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 3 | 0.25 |
FMUL D6, D6, D16 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 3 | 0.25 |
FMINNM D5, D6, D5 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
FMINNM D5, D5, D7 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
FNEG D6, D5 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
FCSEL D6, D5, D6, #12 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 0.50 |
B 41fd2c <__nv_advec_mom_kernel_mod_advec_mom_kernel__F1L79_1_+0x190c> | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 |
Function | advec_mom_kernel |
Source file and lines | advec_mom_kernel.f90-pp.f90:215-241 |
Module | exec |
nb instructions | 30 |
loop length | 120 |
nb stack references | 0 |
front end | 3.75 cycles |
P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | P12 | P13 | P14 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
uops | 1.00 | 1.00 | 2.75 | 2.75 | 3.00 | 2.50 | 3.00 | 3.00 | 3.00 | 3.00 | 2.17 | 1.83 | 2.00 | 0.00 | 0.00 |
cycles | 1.00 | 1.00 | 2.75 | 2.75 | 3.00 | 2.50 | 3.00 | 3.00 | 3.00 | 3.00 | 2.17 | 1.83 | 2.00 | 0.00 | 0.00 |
Cycles executing div or sqrt instructions | 1.00-0.50 |
Longest recurrence chain latency (RecMII) | 1.00 |
Front-end | 3.75 |
Data deps. | 1.00 |
Overall L1 | 3.75 |
all | 0% |
load | 0% |
store | 0% |
mul | NA (no mul vectorizable/vectorized instructions) |
add-sub | 0% |
fma | NA (no fma vectorizable/vectorized instructions) |
div/sqrt | NA (no div/sqrt vectorizable/vectorized instructions) |
other | 0% |
Instruction | Nb FU | P0 | P1 | P2 | P3 | P4 | P5 | P6 | P7 | P8 | P9 | P10 | P11 | P12 | P13 | P14 | Latency | Recip. throughput |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
FMSUB D2, D2, D6, D6 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 4 | 0.25 |
SUB W3, W3, #1 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
CMP W3, #1 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.33 |
FADD D2, D2, D3 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
FMUL D1, D2, D1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 3 | 0.25 |
STR D1, [X13, X4,LSL #3] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0.50 | 0.50 | 0 | 0 | 0 | 2 | 0.50 |
ADD X4, X4, #1 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
B.LE 41fd00 <__nv_advec_mom_kernel_mod_advec_mom_kernel__F1L79_1_+0x18e0> | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 |
LDR D1, [X16, X4,LSL #3] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 6 | 0.33 |
MOVI D6, #0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
FCMP D1, #0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
FABS D2, D1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
CSINC W6, W8, W8, #10 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
CSEL W5, W1, W2, #10 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
SBFM X6, X6, #0, #31 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
SBFM X5, X5, #0, #31 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
MADD X6, X10, X6, X4 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
MADD X5, X10, X5, X4 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
LDR D3, [X11, X6,LSL #3] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 6 | 0.33 |
LDR D4, [X12, X5,LSL #3] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 6 | 0.33 |
CSINC X5, X8, X8, #11 | 1 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.25 |
MADD X5, X10, X5, X4 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
FDIV D2, D2, D3 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.50 | 0 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 7-15 | 1-0.50 |
LDR D3, [X12, X6,LSL #3] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 6 | 0.33 |
FSUB D5, D3, S4 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
LDR D4, [X12, X5,LSL #3] | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.33 | 0.33 | 0.33 | 0 | 0 | 6 | 0.33 |
FSUB D4, D4, S3 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 2 | 0.25 |
FMUL D7, D4, D5 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.25 | 0.25 | 0.25 | 0.25 | 0 | 0 | 0 | 0 | 0 | 3 | 0.25 |
FCMP D7, #0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 |
B.LE 41fd2c <__nv_advec_mom_kernel_mod_advec_mom_kernel__F1L79_1_+0x190c> | 1 | 0.50 | 0.50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.50 |