| Loop Id: 509 | Module: exec | Source: calc_dt_kernel.f90-pp.f90:96-131 | Coverage: 1.85% |
|---|
| Loop Id: 509 | Module: exec | Source: calc_dt_kernel.f90-pp.f90:96-131 | Coverage: 1.85% |
|---|
0x4215a0 LDR X22, [X20, #304] |
0x4215a4 LDR D21, [X22] |
(510) 0x4215a8 FMUL D17, D17, D4 |
(510) 0x4215ac SUB W21, W21, #1 |
(510) 0x4215b0 ADD X13, X13, #8 |
(510) 0x4215b4 FDIV D16, D17, D16 |
(510) 0x4215b8 FMUL D17, D18, D5 |
(510) 0x4215bc FMUL D18, D18, D6 |
(510) 0x4215c0 FDIV D17, D17, D19 |
(510) 0x4215c4 FDIV D18, D18, D20 |
(510) 0x4215c8 FCMP D8, D16 |
(510) 0x4215cc FCSEL D16, D16, D8, #10 |
(510) 0x4215d0 FCMP D16, D17 |
(510) 0x4215d4 FCSEL D16, D17, D16, #10 |
(510) 0x4215d8 FCMP D16, D18 |
(510) 0x4215dc FCSEL D16, D18, D16, #10 |
(510) 0x4215e0 FCMP D16, D21 |
(510) 0x4215e4 FCSEL D8, D21, D16, #10 |
(510) 0x4215e8 CMP W21, #1 |
(510) 0x4215ec B.LE 4214a4 |
(510) 0x4215f0 LDR D18, [X0, X13] |
(510) 0x4215f4 LDR D19, [X18, X13] |
(510) 0x4215f8 LDR D16, [X1, X13] |
(510) 0x4215fc ADD X22, X2, X13 |
(510) 0x421600 FADD D18, D18, D18 |
(510) 0x421604 LDR Q20, [X4, X13] |
(510) 0x421608 LDP D24, D23, [X22, #1016] |
(510) 0x42160c LDR D22, [X16, X13] |
(510) 0x421610 ADD X22, X6, X13 |
(510) 0x421614 LDR D17, [X15, X13] |
(510) 0x421618 FDIV D18, D18, D19 |
(510) 0x42161c LDR Q19, [X7, X13] |
(510) 0x421620 FADD D23, D23, D24 |
(510) 0x421624 LDP D25, D24, [X22, #1016] |
(510) 0x421628 FMUL D22, D22, D23 |
(510) 0x42162c LDR D23, [X5, X13] |
(510) 0x421630 FADD D24, D24, D25 |
(510) 0x421634 FADD V19.2D, V19.2D, V20.2D |
(510) 0x421638 FMUL D23, D23, D24 |
(510) 0x42163c FMADD D16, D16, D16, D18 |
(510) 0x421640 LDR Q18, [X3, X13] |
(510) 0x421644 FSQRT D16, D16 |
(510) 0x421648 FMUL V19.2D, V18.2D, V19.2D |
(510) 0x42164c MOV D20, V19.D[1] |
(510) 0x421650 FADD D18, D20, D0 |
(510) 0x421654 FABS D20, D20 |
(510) 0x421658 FSUB D21, D18, S19 |
(510) 0x42165c LDR D18, [X17, X13] |
(510) 0x421660 FABS D19, D19 |
(510) 0x421664 FADD D21, D21, D23 |
(510) 0x421668 FABS D23, D23 |
(510) 0x42166c FSUB D21, D21, S22 |
(510) 0x421670 FABS D22, D22 |
(510) 0x421674 FCMP D16, D3 |
(510) 0x421678 FCSEL D16, D3, D16, #9 |
(510) 0x42167c FCMP D17, D2 |
(510) 0x421680 FCSEL D17, D2, D17, #10 |
(510) 0x421684 FCMP D19, D20 |
(510) 0x421688 FCSEL D19, D20, D19, #9 |
(510) 0x42168c FMUL D20, D3, D18 |
(510) 0x421690 FCMP D19, D20 |
(510) 0x421694 FCSEL D19, D20, D19, #9 |
(510) 0x421698 FCMP D22, D23 |
(510) 0x42169c FCSEL D22, D23, D22, #9 |
(510) 0x4216a0 FCMP D22, D20 |
(510) 0x4216a4 FCSEL D20, D20, D22, #9 |
(510) 0x4216a8 FADD D22, D18, D18 |
(510) 0x4216ac FDIV D21, D21, D22 |
(510) 0x4216b0 FCMP D21, D7 |
(510) 0x4216b4 B.PL 4215a0 |
(510) 0x4216b8 FDIV D21, D1, D21 |
(510) 0x4216bc LDR X22, [X20, #296] |
(510) 0x4216c0 LDR D22, [X22] |
(510) 0x4216c4 FMUL D21, D21, D22 |
(510) 0x4216c8 B 4215a8 |
/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/armclang_1/CMakeFiles/clover_leaf.dir/CloverLeaf_ref/kernels/calc_dt_kernel.f90-pp.f90: 96 - 131 |
-------------------------------------------------------------------------------- |
96: # 96 "/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/calc_dt_kernel.f90" |
97: dsx=celldx(j) |
98: dsy=celldy(k) |
99: # 99 "/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/calc_dt_kernel.f90" |
100: cc=soundspeed(j,k)*soundspeed(j,k) |
101: cc=cc+2.0_8*viscosity_a(j,k)/density0(j,k) |
102: cc=MAX(SQRT(cc),g_small) |
103: # 103 "/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/calc_dt_kernel.f90" |
104: dtct=dtc_safe*MIN(dsx,dsy)/cc |
105: # 105 "/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/calc_dt_kernel.f90" |
106: div=0.0 |
107: # 107 "/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/calc_dt_kernel.f90" |
108: dv1=(xvel0(j ,k)+xvel0(j ,k+1))*xarea(j ,k) |
109: dv2=(xvel0(j+1,k)+xvel0(j+1,k+1))*xarea(j+1,k) |
110: # 110 "/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/calc_dt_kernel.f90" |
111: div=div+dv2-dv1 |
112: # 112 "/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/calc_dt_kernel.f90" |
113: dtut=dtu_safe*2.0_8*volume(j,k)/MAX(ABS(dv1),ABS(dv2),g_small*volume(j,k)) |
114: # 114 "/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/calc_dt_kernel.f90" |
115: dv1=(yvel0(j,k )+yvel0(j+1,k ))*yarea(j,k ) |
116: dv2=(yvel0(j,k+1)+yvel0(j+1,k+1))*yarea(j,k+1) |
117: # 117 "/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/calc_dt_kernel.f90" |
118: div=div+dv2-dv1 |
119: # 119 "/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/calc_dt_kernel.f90" |
120: dtvt=dtv_safe*2.0_8*volume(j,k)/MAX(ABS(dv1),ABS(dv2),g_small*volume(j,k)) |
121: # 121 "/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/calc_dt_kernel.f90" |
122: div=div/(2.0_8*volume(j,k)) |
123: # 123 "/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/calc_dt_kernel.f90" |
124: IF(div.LT.-g_small)THEN |
125: dtdivt=dtdiv_safe*(-1.0_8/div) |
126: ELSE |
127: dtdivt=g_big |
128: ENDIF |
129: # 129 "/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/calc_dt_kernel.f90" |
130: dt_min_val=MIN(dt_min_val,dtct,dtut,dtvt,dtdivt) |
131: # 131 "/home/hbollore/qaas-runs/170-308-3604/intel/CloverLeafFC/build/CloverLeafFC/CloverLeaf_ref/kernels/calc_dt_kernel.f90" |
| Path / |
| Metric | Value |
|---|---|
| CQA speedup if no scalar integer | NA |
| CQA speedup if FP arith vectorized | NA |
| CQA speedup if fully vectorized | NA |
| CQA speedup if no inter-iteration dependency | NA |
| CQA speedup if next bottleneck killed | NA |
| Bottlenecks | NA |
| Function | __nv_calc_dt_kernel_module_calc_dt_kernel__F1L89_1_ |
| Source | calc_dt_kernel.f90-pp.f90:126-126 |
| Source loop unroll info | NA |
| Source loop unroll confidence level | NA |
| Unroll/vectorization loop type | NA |
| Unroll factor | NA |
| CQA cycles | NA |
| CQA cycles if no scalar integer | NA |
| CQA cycles if FP arith vectorized | NA |
| CQA cycles if fully vectorized | NA |
| Front-end cycles | NA |
| DIV/SQRT cycles | NA |
| P0 cycles | NA |
| P1 cycles | NA |
| P2 cycles | NA |
| P3 cycles | NA |
| P4 cycles | NA |
| P5 cycles | NA |
| P6 cycles | NA |
| P7 cycles | NA |
| P8 cycles | NA |
| P9 cycles | NA |
| P10 cycles | NA |
| P11 cycles | NA |
| P12 cycles | NA |
| P13 cycles | NA |
| P14 cycles | NA |
| Inter-iter dependencies cycles | NA |
| FE+BE cycles (UFS) | NA |
| Stall cycles (UFS) | NA |
| Nb insns | NA |
| Nb uops | NA |
| Nb loads | NA |
| Nb stores | NA |
| Nb stack references | NA |
| FLOP/cycle | NA |
| Nb FLOP add-sub | NA |
| Nb FLOP mul | NA |
| Nb FLOP fma | NA |
| Nb FLOP div | NA |
| Nb FLOP rcp | NA |
| Nb FLOP sqrt | NA |
| Nb FLOP rsqrt | NA |
| Bytes/cycle | NA |
| Bytes prefetched | NA |
| Bytes loaded | NA |
| Bytes stored | NA |
| Stride 0 | NA |
| Stride 1 | NA |
| Stride n | NA |
| Stride unknown | NA |
| Stride indirect | NA |
| Vectorization ratio all | NA |
| Vectorization ratio load | NA |
| Vectorization ratio store | NA |
| Vectorization ratio mul | NA |
| Vectorization ratio add_sub | NA |
| Vectorization ratio fma | NA |
| Vectorization ratio div_sqrt | NA |
| Vectorization ratio other | NA |
| Vector-efficiency ratio all | NA |
| Vector-efficiency ratio load | NA |
| Vector-efficiency ratio store | NA |
| 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 | NA |
| Path / |
| Function | __nv_calc_dt_kernel_module_calc_dt_kernel__F1L89_1_ |
| Source file and lines | calc_dt_kernel.f90-pp.f90:126-126 |
| Module | exec |