OV - exec - Function getBoxFromTuple

	Function: getBoxFromTuple	Module: exec	Source: linkCells.c:204-250	Coverage (incl. loops): 0.01% \| (excl. loops): 0.01%

/home/eoseret/qaas_runs_GNR/173-814-1858/intel/CoMD/build/CoMD/CoMD/src-openmp/linkCells.c: 204 - 250

--------------------------------------------------------------------------------

204: {

205:    int iBox = 0;

206:    const int* gridSize = boxes->gridSize; // alias

207:

208:    // Halo in Z+

209:    if (iz == gridSize[2])

210:    {

211:       iBox = boxes->nLocalBoxes + 2*gridSize[2]*gridSize[1] + 2*gridSize[2]*(gridSize[0]+2) +

212:          (gridSize[0]+2)*(gridSize[1]+2) + (gridSize[0]+2)*(iy+1) + (ix+1);

213:    }

214:    // Halo in Z-

215:    else if (iz == -1)

216:    {

217:       iBox = boxes->nLocalBoxes + 2*gridSize[2]*gridSize[1] + 2*gridSize[2]*(gridSize[0]+2) +

218:          (gridSize[0]+2)*(iy+1) + (ix+1);

219:    }

220:    // Halo in Y+

221:    else if (iy == gridSize[1])

222:    {

223:       iBox = boxes->nLocalBoxes + 2*gridSize[2]*gridSize[1] + gridSize[2]*(gridSize[0]+2) +

224:          (gridSize[0]+2)*iz + (ix+1);

225:    }

226:    // Halo in Y-

227:    else if (iy == -1)

228:    {

229:       iBox = boxes->nLocalBoxes + 2*gridSize[2]*gridSize[1] + iz*(gridSize[0]+2) + (ix+1);

230:    }

231:    // Halo in X+

232:    else if (ix == gridSize[0])

233:    {

234:       iBox = boxes->nLocalBoxes + gridSize[1]*gridSize[2] + iz*gridSize[1] + iy;

235:    }

236:    // Halo in X-

237:    else if (ix == -1)

238:    {

239:       iBox = boxes->nLocalBoxes + iz*gridSize[1] + iy;

240:    }

241:    // local link celll.

242:    else

243:    {

244:       iBox = ix + gridSize[0]*iy + gridSize[0]*gridSize[1]*iz;

245:    }

246:    assert(iBox >= 0);

247:    assert(iBox < boxes->nTotalBoxes);

248:

249:    return iBox;

250: }

0x40c550 PUSH	%RBP

0x40c551 MOV	%EDX,%EAX

0x40c553 MOV	%ECX,%R9D

0x40c556 MOV	0x8(%RDI),%EDX

0x40c559 MOV	%ESI,%R8D

0x40c55c MOV	(%RDI),%ECX

0x40c55e MOV	0x4(%RDI),%ESI

0x40c561 MOV	%RSP,%RBP

0x40c564 CMP	%R9D,%EDX

0x40c567 JE	40c5e0

0x40c569 CMP	$-0x1,%R9D

0x40c56d JE	40c5a8

0x40c56f CMP	%ESI,%EAX

0x40c571 JE	40c630

0x40c577 CMP	$-0x1,%EAX

0x40c57a JE	40c610

0x40c580 CMP	%ECX,%R8D

0x40c583 JE	40c658

0x40c589 CMP	$-0x1,%R8D

0x40c58d JE	40c670

0x40c593 IMUL	%ECX,%ESI

0x40c596 IMUL	%ECX,%EAX

0x40c599 IMUL	%R9D,%ESI

0x40c59d ADD	%R8D,%EAX

0x40c5a0 ADD	%ESI,%EAX

0x40c5a2 JMP	40c5c9

0x40c5a4 NOPL	(%RAX)

0x40c5a8 IMUL	%EDX,%ESI

0x40c5ab ADD	$0x2,%ECX

0x40c5ae IMUL	%ECX,%EDX

0x40c5b1 MOV	0xc(%RDI),%R9D

0x40c5b5 INC	%EAX

0x40c5b7 IMUL	%ECX,%EAX

0x40c5ba LEA	(%R9,%RSI,2),%R10D

0x40c5be LEA	(%R10,%RDX,2),%EDX

0x40c5c2 ADD	%EDX,%EAX

0x40c5c4 LEA	0x1(%R8,%RAX,1),%EAX

0x40c5c9 TEST	%EAX,%EAX

0x40c5cb JS	40c67e

0x40c5d1 CMP	%EAX,0x14(%RDI)

0x40c5d4 JLE	40c697

0x40c5da POP	%RBP

0x40c5db RET

0x40c5dc NOPL	(%RAX)

0x40c5e0 MOV	%EDX,%R11D

0x40c5e3 ADD	$0x2,%ECX

0x40c5e6 IMUL	%ESI,%R11D

0x40c5ea IMUL	%ECX,%EDX

0x40c5ed MOV	0xc(%RDI),%R9D

0x40c5f1 ADD	$0x2,%ESI

0x40c5f4 IMUL	%ECX,%ESI

0x40c5f7 INC	%EAX

0x40c5f9 IMUL	%ECX,%EAX

0x40c5fc LEA	(%R9,%R11,2),%R10D

0x40c600 LEA	(%R10,%RDX,2),%EDX

0x40c604 ADD	%EDX,%ESI

0x40c606 ADD	%ESI,%EAX

0x40c608 LEA	0x1(%R8,%RAX,1),%EAX

0x40c60d JMP	40c5c9

0x40c60f NOP

0x40c610 IMUL	%ESI,%EDX

0x40c613 ADD	$0x2,%ECX

0x40c616 MOV	0xc(%RDI),%EAX

0x40c619 IMUL	%R9D,%ECX

0x40c61d LEA	(%RAX,%RDX,2),%R10D

0x40c621 ADD	%ECX,%R10D

0x40c624 LEA	0x1(%R8,%R10,1),%EAX

0x40c629 JMP	40c5c9

0x40c62b NOPL	(%RAX,%RAX,1)

0x40c630 IMUL	%EDX,%EAX

0x40c633 ADD	$0x2,%ECX

0x40c636 IMUL	%ECX,%EDX

0x40c639 MOV	0xc(%RDI),%R11D

0x40c63d IMUL	%R9D,%ECX

0x40c641 LEA	(%R11,%RAX,2),%ESI

0x40c645 ADD	%ESI,%EDX

0x40c647 ADD	%ECX,%EDX

0x40c649 LEA	0x1(%R8,%RDX,1),%EAX

0x40c64e JMP	40c5c9

0x40c653 NOPL	(%RAX,%RAX,1)

0x40c658 IMUL	%ESI,%EDX

0x40c65b IMUL	%R9D,%ESI

0x40c65f ADD	0xc(%RDI),%EDX

0x40c662 ADD	%ESI,%EDX

0x40c664 ADD	%EDX,%EAX

0x40c666 JMP	40c5c9

0x40c66b NOPL	(%RAX,%RAX,1)

0x40c670 IMUL	%R9D,%ESI

0x40c674 ADD	0xc(%RDI),%ESI

0x40c677 ADD	%ESI,%EAX

0x40c679 JMP	40c5c9

0x40c67e MOV	$0x413000,%ECX

0x40c683 MOV	$0xf6,%EDX

0x40c688 MOV	$0x412ef0,%ESI

0x40c68d MOV	$0x412fbf,%EDI

0x40c692 CALL	4030c0 <__assert_fail@plt>

0x40c697 MOV	$0x413000,%ECX

0x40c69c MOV	$0xf7,%EDX

0x40c6a1 MOV	$0x412ef0,%ESI

0x40c6a6 MOV	$0x412fc9,%EDI

0x40c6ab CALL	4030c0 <__assert_fail@plt>

Coverage (%)	Name	Source Location	Module
►21.74+	putAtomInBox	linkCells.c:378	exec
○	unloadAtomsBuffer	haloExchange.c:414	exec
○	haloExchange	haloExchange.c:303	exec
○	timestep	timestep.c:150	exec
○	main	CoMD.c:125	exec
►17.39+	putAtomInBox	linkCells.c:378	exec
○	unloadAtomsBuffer	haloExchange.c:414	exec
○	haloExchange	haloExchange.c:302	exec
○	timestep	timestep.c:150	exec
○	main	CoMD.c:125	exec
►13.04+	unloadAtomsBuffer	haloExchange.c:417	exec
○	haloExchange	haloExchange.c:303	exec
○	timestep	timestep.c:150	exec
○	main	CoMD.c:125	exec
○	__libc_start_call_main		libc.so.6
►8.70+	putAtomInBox	linkCells.c:378	exec
○	unloadAtomsBuffer	haloExchange.c:417	exec
○	haloExchange	haloExchange.c:302	exec
○	timestep	timestep.c:150	exec
○	main	CoMD.c:125	exec
►8.70+	unloadAtomsBuffer	haloExchange.c:417	exec
○	haloExchange	haloExchange.c:303	exec
○	timestep	timestep.c:150	exec
○	main	CoMD.c:125	exec
►8.69+	unloadAtomsBuffer	haloExchange.c:414	exec
○	haloExchange	haloExchange.c:302	exec
○	timestep	timestep.c:150	exec
○	main	CoMD.c:125	exec
►4.35+	putAtomInBox	linkCells.c:378	exec
○	unloadAtomsBuffer	haloExchange.c:414	exec
○	haloExchange	haloExchange.c:302	exec
○	timestep	timestep.c:150	exec
○	main	CoMD.c:125	exec
○	__libc_start_call_main		libc.so.6
►4.35+	unloadAtomsBuffer	haloExchange.c:417	exec
○	haloExchange	haloExchange.c:303	exec
○	redistributeAtoms	timestep.c:150	exec
○	main	CoMD.c:207	exec
○	__libc_start_call_main		libc.so.6
►4.34+	unloadAtomsBuffer	haloExchange.c:417	exec
○	haloExchange	haloExchange.c:302	exec
○	timestep	timestep.c:150	exec
○	main	CoMD.c:125	exec
►4.34+	putAtomInBox	linkCells.c:378	exec
○	createFccLattice	initAtoms.c:96	exec
○	main	CoMD.c:199	exec
○	__libc_start_call_main		libc.so.6
►4.34+	unloadAtomsBuffer	haloExchange.c:414	exec
○	haloExchange	haloExchange.c:302	exec
○	timestep	timestep.c:150	exec
○	main	CoMD.c:125	exec
○	__libc_start_call_main		libc.so.6

Path /

Average path: Display a virtual path defined by average values of all real paths

The code analyzed by CQA in that panel excludes loops and represents 0.01% of application time for run gcc_3

Source file and lines	linkCells.c:204-250
Module	exec

The function is defined in /home/eoseret/qaas_runs_GNR/173-814-1858/intel/CoMD/build/CoMD/CoMD/src-openmp/linkCells.c:204-250.

Warnings:

Ignoring paths for analysis
Too many paths. Rerun with max-paths=21

Try to simplify control and/or increase the maximum number of paths per function/loop through the 'max-paths-nb' option.

This function has 21 execution paths.

The presence of multiple execution paths is typically the main/first bottleneck.
Try to simplify control inside function: ideally, try to remove all conditional expressions, for example by (if applicable):

hoisting them (moving them outside the function)
turning them into conditional moves, MIN or MAX

Ex: if (x<0) x=0 => x = (x<0 ? 0 : x) (or MAX(0,x) after defining the corresponding macro)

gain
potential
hint
expert

Vectorization

Your function is not vectorized. Only 6% of vector register length is used (average across all SSE/AVX instructions). By vectorizing your function, you can lower the cost of an iteration from 30.40 to 1.90 cycles (16.00x speedup).

Details

All SSE/AVX instructions are used in scalar version (process only one data element in vector registers). Since your execution units are vector units, only a vectorized function can use their full power.

Workaround

Try another compiler or update/tune your current one
Make array accesses unit-stride:
- If your function streams arrays of structures (AoS), try to use structures of arrays instead (SoA): for(i) a[i].x = b[i].x; (slow, non stride 1) => for(i) a.x[i] = b.x[i]; (fast, stride 1)

Execution units bottlenecks

Found no such bottlenecks but see expert reports for more complex bottlenecks.

No data for this section

CALL instructions

Detected function call instructions.

Details

Calling (and then returning from) a function prevents many compiler optimizations (like vectorization), breaks control flow (which reduces pipeline performance) and executes extra instructions to save/restore the registers used inside it, which is very expensive (dozens of cycles). Consider to inline small functions.

__assert_fail@plt: 2 occurrences►
- /home/eoseret/qaas_runs_GNR/173-814-1858/intel/CoMD/build/CoMD/CoMD/src-openmp/linkCells.c:246
- /home/eoseret/qaas_runs_GNR/173-814-1858/intel/CoMD/build/CoMD/CoMD/src-openmp/linkCells.c:247

Type of elements and instruction set

No instructions are processing arithmetic or math operations on FP elements. This function is probably writing/copying data or processing integer elements.

Matching between your function (in the source code) and the binary function

The binary function does not contain any FP arithmetical operations. The binary function is loading 40 bytes.

General properties

nb instructions	101
nb uops	103
loop length	352
used x86 registers	11
used mmx registers	0
used xmm registers	0
used ymm registers	0
used zmm registers	0
nb stack references	0

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	17.17 cycles
front end	17.17 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	12.70	22.00	4.00	4.00	1.50	12.80	12.70	1.50	1.50	1.50	12.80	4.00
cycles	12.70	30.40	4.00	4.00	1.50	12.80	12.70	1.50	1.50	1.50	12.80	4.00

Execution ports to units layout:

P0 (256 bits): VPU, BRU, ALU, DIV/SQRT
P1 (256 bits): ALU, VPU
P2 (512 bits): load
P3 (512 bits): load
P4 (512 bits): store data
P5 (512 bits): ALU, VPU
P6: ALU, BRU
P7: store address
P8: store address
P9 (512 bits): store data
P10: ALU
P11 (512 bits): load

Cycles executing div or sqrt instructions

Cycles summary

Front-end	17.17
Dispatch	30.40
Overall L1	30.40

Vectorization ratios

all	0%
load	0%
store	NA (no store vectorizable/vectorized instructions)
mul	0%
add-sub	0%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	0%

Vector efficiency ratios

all	6%
load	6%
store	NA (no store vectorizable/vectorized instructions)
mul	6%
add-sub	6%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	6%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 30.40 cycles. At this rate:

0% of peak load performance is reached (1.32 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 40c550

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOV %EDX,%EAX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %ECX,%R9D	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x8(%RDI),%EDX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %ESI,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV (%RDI),%ECX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (6.3%)
MOV 0x4(%RDI),%ESI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
CMP %R9D,%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
JE 40c5e0 <getBoxFromTuple+0x90>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $-0x1,%R9D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
JE 40c5a8 <getBoxFromTuple+0x58>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP %ESI,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
JE 40c630 <getBoxFromTuple+0xe0>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $-0x1,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
JE 40c610 <getBoxFromTuple+0xc0>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP %ECX,%R8D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
JE 40c658 <getBoxFromTuple+0x108>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $-0x1,%R8D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
JE 40c670 <getBoxFromTuple+0x120>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
IMUL %ECX,%ESI	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
IMUL %ECX,%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
IMUL %R9D,%ESI	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
ADD %R8D,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
ADD %ESI,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
JMP 40c5c9 <getBoxFromTuple+0x79>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
NOPL (%RAX)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
IMUL %EDX,%ESI	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
ADD $0x2,%ECX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
IMUL %ECX,%EDX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
MOV 0xc(%RDI),%R9D	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
INC %EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
IMUL %ECX,%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
LEA (%R9,%RSI,2),%R10D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
LEA (%R10,%RDX,2),%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
ADD %EDX,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
LEA 0x1(%R8,%RAX,1),%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
TEST %EAX,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (6.3%)
JS 40c67e <getBoxFromTuple+0x12e>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP %EAX,0x14(%RDI)	1	0.20	0.20	0.33	0.33	0	0.20	0.20	0	0	0	0.20	0.33	1	0.33	scal (6.3%)
JLE 40c697 <getBoxFromTuple+0x147>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
POP %RBP	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
RET	1	0.50	0	0.33	0.33	0	0	0.50	0	0	0	0	0.33	0	2.13	N/A
NOPL (%RAX)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %EDX,%R11D	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD $0x2,%ECX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
IMUL %ESI,%R11D	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
IMUL %ECX,%EDX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
MOV 0xc(%RDI),%R9D	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
ADD $0x2,%ESI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
IMUL %ECX,%ESI	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
INC %EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
IMUL %ECX,%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
LEA (%R9,%R11,2),%R10D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
LEA (%R10,%RDX,2),%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
ADD %EDX,%ESI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
ADD %ESI,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
LEA 0x1(%R8,%RAX,1),%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
JMP 40c5c9 <getBoxFromTuple+0x79>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
NOP	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
IMUL %ESI,%EDX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
ADD $0x2,%ECX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
MOV 0xc(%RDI),%EAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
IMUL %R9D,%ECX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (6.3%)
LEA (%RAX,%RDX,2),%R10D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
ADD %ECX,%R10D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
LEA 0x1(%R8,%R10,1),%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
JMP 40c5c9 <getBoxFromTuple+0x79>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
NOPL (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
IMUL %EDX,%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
ADD $0x2,%ECX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
IMUL %ECX,%EDX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
MOV 0xc(%RDI),%R11D	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
IMUL %R9D,%ECX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (6.3%)
LEA (%R11,%RAX,2),%ESI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
ADD %ESI,%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
ADD %ECX,%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
LEA 0x1(%R8,%RDX,1),%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
JMP 40c5c9 <getBoxFromTuple+0x79>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
NOPL (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
IMUL %ESI,%EDX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
IMUL %R9D,%ESI	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
ADD 0xc(%RDI),%EDX	1	0.20	0.20	0.33	0.33	0	0.20	0.20	0	0	0	0.20	0.33	1	0.33	N/A
ADD %ESI,%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
ADD %EDX,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
JMP 40c5c9 <getBoxFromTuple+0x79>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
NOPL (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
IMUL %R9D,%ESI	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
ADD 0xc(%RDI),%ESI	1	0.20	0.20	0.33	0.33	0	0.20	0.20	0	0	0	0.20	0.33	1	0.33	N/A
ADD %ESI,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
JMP 40c5c9 <getBoxFromTuple+0x79>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
MOV $0x413000,%ECX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
MOV $0xf6,%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV $0x412ef0,%ESI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV $0x412fbf,%EDI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
CALL 4030c0 <__assert_fail@plt>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV $0x413000,%ECX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
MOV $0xf7,%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV $0x412ef0,%ESI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV $0x412fc9,%EDI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
CALL 4030c0 <__assert_fail@plt>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A

The code analyzed by CQA in that panel excludes loops and represents 0.01% of application time for run gcc_3

Source file and lines	linkCells.c:204-250
Module	exec

The function is defined in /home/eoseret/qaas_runs_GNR/173-814-1858/intel/CoMD/build/CoMD/CoMD/src-openmp/linkCells.c:204-250.

Warnings:

Ignoring paths for analysis
Too many paths. Rerun with max-paths=21

hoisting them (moving them outside the function)
turning them into conditional moves, MIN or MAX

Ex: if (x<0) x=0 => x = (x<0 ? 0 : x) (or MAX(0,x) after defining the corresponding macro)

gain
potential
hint
expert

Vectorization

Details

Workaround

Try another compiler or update/tune your current one
Make array accesses unit-stride:
- If your function streams arrays of structures (AoS), try to use structures of arrays instead (SoA): for(i) a[i].x = b[i].x; (slow, non stride 1) => for(i) a.x[i] = b.x[i]; (fast, stride 1)

Execution units bottlenecks

Found no such bottlenecks but see expert reports for more complex bottlenecks.

No data for this section

CALL instructions

Detected function call instructions.

Details

__assert_fail@plt: 2 occurrences►
- /home/eoseret/qaas_runs_GNR/173-814-1858/intel/CoMD/build/CoMD/CoMD/src-openmp/linkCells.c:246
- /home/eoseret/qaas_runs_GNR/173-814-1858/intel/CoMD/build/CoMD/CoMD/src-openmp/linkCells.c:247

Type of elements and instruction set

No instructions are processing arithmetic or math operations on FP elements. This function is probably writing/copying data or processing integer elements.

Matching between your function (in the source code) and the binary function

The binary function does not contain any FP arithmetical operations. The binary function is loading 40 bytes.

General properties

nb instructions	101
nb uops	103
loop length	352
used x86 registers	11
used mmx registers	0
used xmm registers	0
used ymm registers	0
used zmm registers	0
nb stack references	0

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	17.17 cycles
front end	17.17 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	12.70	22.00	4.00	4.00	1.50	12.80	12.70	1.50	1.50	1.50	12.80	4.00
cycles	12.70	30.40	4.00	4.00	1.50	12.80	12.70	1.50	1.50	1.50	12.80	4.00

Execution ports to units layout:

P0 (256 bits): VPU, BRU, ALU, DIV/SQRT
P1 (256 bits): ALU, VPU
P2 (512 bits): load
P3 (512 bits): load
P4 (512 bits): store data
P5 (512 bits): ALU, VPU
P6: ALU, BRU
P7: store address
P8: store address
P9 (512 bits): store data
P10: ALU
P11 (512 bits): load

Cycles executing div or sqrt instructions

Cycles summary

Front-end	17.17
Dispatch	30.40
Overall L1	30.40

Vectorization ratios

all	0%
load	0%
store	NA (no store vectorizable/vectorized instructions)
mul	0%
add-sub	0%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	0%

Vector efficiency ratios

all	6%
load	6%
store	NA (no store vectorizable/vectorized instructions)
mul	6%
add-sub	6%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	6%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 30.40 cycles. At this rate:

0% of peak load performance is reached (1.32 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 40c550

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	5-12	0.50	N/A
MOV %EDX,%EAX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV %ECX,%R9D	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV 0x8(%RDI),%EDX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %ESI,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV (%RDI),%ECX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (6.3%)
MOV 0x4(%RDI),%ESI	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	scal (12.5%)
CMP %R9D,%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
JE 40c5e0 <getBoxFromTuple+0x90>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $-0x1,%R9D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
JE 40c5a8 <getBoxFromTuple+0x58>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP %ESI,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
JE 40c630 <getBoxFromTuple+0xe0>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $-0x1,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
JE 40c610 <getBoxFromTuple+0xc0>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP %ECX,%R8D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
JE 40c658 <getBoxFromTuple+0x108>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP $-0x1,%R8D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
JE 40c670 <getBoxFromTuple+0x120>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
IMUL %ECX,%ESI	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
IMUL %ECX,%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
IMUL %R9D,%ESI	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
ADD %R8D,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
ADD %ESI,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
JMP 40c5c9 <getBoxFromTuple+0x79>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
NOPL (%RAX)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
IMUL %EDX,%ESI	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
ADD $0x2,%ECX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
IMUL %ECX,%EDX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
MOV 0xc(%RDI),%R9D	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
INC %EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
IMUL %ECX,%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
LEA (%R9,%RSI,2),%R10D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
LEA (%R10,%RDX,2),%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
ADD %EDX,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
LEA 0x1(%R8,%RAX,1),%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
TEST %EAX,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	2	0.20	scal (6.3%)
JS 40c67e <getBoxFromTuple+0x12e>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
CMP %EAX,0x14(%RDI)	1	0.20	0.20	0.33	0.33	0	0.20	0.20	0	0	0	0.20	0.33	1	0.33	scal (6.3%)
JLE 40c697 <getBoxFromTuple+0x147>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
POP %RBP	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1-6	0.33	N/A
RET	1	0.50	0	0.33	0.33	0	0	0.50	0	0	0	0	0.33	0	2.13	N/A
NOPL (%RAX)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %EDX,%R11D	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
ADD $0x2,%ECX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
IMUL %ESI,%R11D	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
IMUL %ECX,%EDX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
MOV 0xc(%RDI),%R9D	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
ADD $0x2,%ESI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
IMUL %ECX,%ESI	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
INC %EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
IMUL %ECX,%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
LEA (%R9,%R11,2),%R10D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
LEA (%R10,%RDX,2),%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
ADD %EDX,%ESI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
ADD %ESI,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
LEA 0x1(%R8,%RAX,1),%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
JMP 40c5c9 <getBoxFromTuple+0x79>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
NOP	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
IMUL %ESI,%EDX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
ADD $0x2,%ECX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
MOV 0xc(%RDI),%EAX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
IMUL %R9D,%ECX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (6.3%)
LEA (%RAX,%RDX,2),%R10D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
ADD %ECX,%R10D	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
LEA 0x1(%R8,%R10,1),%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
JMP 40c5c9 <getBoxFromTuple+0x79>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	5.84	N/A
NOPL (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
IMUL %EDX,%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
ADD $0x2,%ECX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
IMUL %ECX,%EDX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
MOV 0xc(%RDI),%R11D	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
IMUL %R9D,%ECX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	scal (6.3%)
LEA (%R11,%RAX,2),%ESI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1-2	0.20	N/A
ADD %ESI,%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
ADD %ECX,%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
LEA 0x1(%R8,%RDX,1),%EAX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
JMP 40c5c9 <getBoxFromTuple+0x79>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
NOPL (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
IMUL %ESI,%EDX	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
IMUL %R9D,%ESI	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
ADD 0xc(%RDI),%EDX	1	0.20	0.20	0.33	0.33	0	0.20	0.20	0	0	0	0.20	0.33	1	0.33	N/A
ADD %ESI,%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
ADD %EDX,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
JMP 40c5c9 <getBoxFromTuple+0x79>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
NOPL (%RAX,%RAX,1)	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
IMUL %R9D,%ESI	1	0	1	0	0	0	0	0	0	0	0	0	0	3	1	N/A
ADD 0xc(%RDI),%ESI	1	0.20	0.20	0.33	0.33	0	0.20	0.20	0	0	0	0.20	0.33	1	0.33	N/A
ADD %ESI,%EAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
JMP 40c5c9 <getBoxFromTuple+0x79>	1	0	0	0	0	0	0	0	0	0	0	0	0	0	2.08	N/A
MOV $0x413000,%ECX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
MOV $0xf6,%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV $0x412ef0,%ESI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV $0x412fbf,%EDI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
CALL 4030c0 <__assert_fail@plt>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A
MOV $0x413000,%ECX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (6.3%)
MOV $0xf7,%EDX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV $0x412ef0,%ESI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
MOV $0x412fc9,%EDI	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
CALL 4030c0 <__assert_fail@plt>	2	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	0	1	N/A

Name	Coverage (%)	Time (s)
○getBoxFromTuple	0.01	0.00

Report Configuration

Vectorization

Details

Workaround

Execution units bottlenecks

CALL instructions

Details

Type of elements and instruction set

Matching between your function (in the source code) and the binary function

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code

Vectorization

Details

Workaround

Execution units bottlenecks

CALL instructions

Details

Type of elements and instruction set

Matching between your function (in the source code) and the binary function

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code