OV - exec - Function sortAtomsInCell

	Function: sortAtomsInCell	Module: exec	Source: haloExchange.c:614-645	Coverage (incl. loops): 3.38% \| (excl. loops): 0.45%

/beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c: 614 - 645

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

614: {

615:    int nAtoms = boxes->nAtoms[iBox];

616:

617:    AtomMsg tmp[nAtoms];

618:

619:    int begin = iBox*MAXATOMS;

620:    int end = begin + nAtoms;

621:    for (int ii=begin, iTmp=0; ii<end; ++ii, ++iTmp)

622:    {

623:       tmp[iTmp].gid  = atoms->gid[ii];

624:       tmp[iTmp].type = atoms->iSpecies[ii];

625:       tmp[iTmp].rx =   atoms->r[ii][0];

626:       tmp[iTmp].ry =   atoms->r[ii][1];

627:       tmp[iTmp].rz =   atoms->r[ii][2];

628:       tmp[iTmp].px =   atoms->p[ii][0];

629:       tmp[iTmp].py =   atoms->p[ii][1];

630:       tmp[iTmp].pz =   atoms->p[ii][2];

631:    }

632:    qsort(&tmp, nAtoms, sizeof(AtomMsg), sortAtomsById);

633:    for (int ii=begin, iTmp=0; ii<end; ++ii, ++iTmp)

634:    {

635:       atoms->gid[ii]   = tmp[iTmp].gid;

636:       atoms->iSpecies[ii] = tmp[iTmp].type;

637:       atoms->r[ii][0]  = tmp[iTmp].rx;

638:       atoms->r[ii][1]  = tmp[iTmp].ry;

639:       atoms->r[ii][2]  = tmp[iTmp].rz;

640:       atoms->p[ii][0]  = tmp[iTmp].px;

641:       atoms->p[ii][1]  = tmp[iTmp].py;

642:       atoms->p[ii][2]  = tmp[iTmp].pz;

643:    }

644:

645: }

0x408950 PUSH	%RBP

0x408951 MOVSXD	%EDX,%RDX

0x408954 MOV	%RSP,%RBP

0x408957 PUSH	%R15

0x408959 PUSH	%R14

0x40895b PUSH	%R13

0x40895d PUSH	%R12

0x40895f PUSH	%RBX

0x408960 MOV	%RDX,%RBX

0x408963 SUB	$0x28,%RSP

0x408967 MOV	0x78(%RSI),%RAX

0x40896b SAL	$0x6,%EBX

0x40896e MOVSXD	(%RAX,%RDX,4),%R9

0x408972 IMUL	$0x38,%R9,%RCX

0x408976 MOV	%R9,%R12

0x408979 ADD	%EBX,%R12D

0x40897c ADD	$0xf,%RCX

0x408980 AND	$-0x10,%RCX

0x408984 SUB	%RCX,%RSP

0x408987 MOV	%RSP,%R14

0x40898a CMP	%R12D,%EBX

0x40898d JGE	408d70

0x408993 MOV	0x8(%RDI),%R15

0x408997 MOVSXD	%EBX,%RBX

0x40899a MOV	0x10(%RDI),%R11

0x40899e MOV	%RDI,%R10

0x4089a1 MOV	0x20(%R10),%RSI

0x4089a5 MOV	0x18(%RDI),%RDI

0x4089a9 LEA	(%RDX,%RDX,2),%R13

0x4089ad MOV	%RBX,%RCX

0x4089b0 NOT	%RCX

0x4089b3 LEA	0x38(%RSP),%RAX

0x4089b8 SAL	$0x9,%R13

0x4089bc MOV	(%R15,%RBX,4),%R8D

0x4089c0 ADD	%R12D,%ECX

0x4089c3 VMOVSD	(%RDI,%R13,1),%XMM2

0x4089c9 AND	$0x1,%ECX

0x4089cc MOV	%R8D,(%RSP)

0x4089d0 MOV	(%R11,%RBX,4),%EDX

0x4089d4 LEA	0x1(%RBX),%R8

0x4089d8 VMOVSD	%XMM2,0x8(%RSP)

0x4089de VMOVSD	0x8(%RDI,%R13,1),%XMM0

0x4089e5 VMOVSD	%XMM0,0x10(%RSP)

0x4089eb VMOVSD	0x10(%RDI,%R13,1),%XMM1

0x4089f2 VMOVSD	%XMM1,0x18(%RSP)

0x4089f8 VMOVSD	(%RSI,%R13,1),%XMM3

0x4089fe VMOVSD	%XMM3,0x20(%RSP)

0x408a04 VMOVSD	0x8(%RSI,%R13,1),%XMM4

0x408a0b VMOVSD	%XMM4,0x28(%RSP)

0x408a11 VMOVSD	0x10(%RSI,%R13,1),%XMM5

0x408a18 VMOVSD	%XMM5,0x30(%RSP)

0x408a1e MOV	%EDX,0x4(%RSP)

0x408a22 LEA	0x18(%R13),%RDX

0x408a26 CMP	%R8D,%R12D

0x408a29 JLE	408b11

0x408a2f TEST	%ECX,%ECX

0x408a31 JNE	408d00

0x408a37 MOV	%R13,-0x48(%RBP)

0x408a3b MOV	%R10,%R13

0x408a3e MOV	%R11,%R10

0x408a41 MOV	%R9,-0x38(%RBP)

0x408a45 MOV	%RBX,-0x40(%RBP)

(59) 0x408a49 MOV	(%R15,%R8,4),%R9D

(59) 0x408a4d LEA	0x1(%R8),%R11

(59) 0x408a51 VMOVSD	(%RDI,%RDX,1),%XMM12

(59) 0x408a56 ADD	$0x70,%RAX

(59) 0x408a5a MOV	%R9D,-0x70(%RAX)

(59) 0x408a5e MOV	(%R10,%R8,4),%EBX

(59) 0x408a62 VMOVSD	%XMM12,-0x68(%RAX)

(59) 0x408a67 VMOVSD	0x8(%RDI,%RDX,1),%XMM13

(59) 0x408a6d VMOVSD	%XMM13,-0x60(%RAX)

(59) 0x408a72 VMOVSD	0x10(%RDI,%RDX,1),%XMM14

(59) 0x408a78 VMOVSD	%XMM14,-0x58(%RAX)

(59) 0x408a7d VMOVSD	(%RSI,%RDX,1),%XMM15

(59) 0x408a82 VMOVSD	%XMM15,-0x50(%RAX)

(59) 0x408a87 VMOVSD	0x8(%RSI,%RDX,1),%XMM2

(59) 0x408a8d VMOVSD	%XMM2,-0x48(%RAX)

(59) 0x408a92 VMOVSD	0x10(%RSI,%RDX,1),%XMM0

(59) 0x408a98 ADD	$0x2,%R8

(59) 0x408a9c VMOVSD	%XMM0,-0x40(%RAX)

(59) 0x408aa1 VMOVSD	0x18(%RDI,%RDX,1),%XMM1

(59) 0x408aa7 MOV	%EBX,-0x6c(%RAX)

(59) 0x408aaa MOV	(%R15,%R11,4),%ECX

(59) 0x408aae MOV	%ECX,-0x38(%RAX)

(59) 0x408ab1 MOV	(%R10,%R11,4),%R9D

(59) 0x408ab5 VMOVSD	%XMM1,-0x30(%RAX)

(59) 0x408aba VMOVSD	0x20(%RDI,%RDX,1),%XMM3

(59) 0x408ac0 VMOVSD	%XMM3,-0x28(%RAX)

(59) 0x408ac5 VMOVSD	0x28(%RDI,%RDX,1),%XMM4

(59) 0x408acb VMOVSD	%XMM4,-0x20(%RAX)

(59) 0x408ad0 VMOVSD	0x18(%RSI,%RDX,1),%XMM5

(59) 0x408ad6 VMOVSD	%XMM5,-0x18(%RAX)

(59) 0x408adb VMOVSD	0x20(%RSI,%RDX,1),%XMM6

(59) 0x408ae1 VMOVSD	%XMM6,-0x10(%RAX)

(59) 0x408ae6 VMOVSD	0x28(%RSI,%RDX,1),%XMM7

(59) 0x408aec ADD	$0x30,%RDX

(59) 0x408af0 VMOVSD	%XMM7,-0x8(%RAX)

(59) 0x408af5 MOV	%R9D,-0x34(%RAX)

(59) 0x408af9 CMP	%R8D,%R12D

(59) 0x408afc JG	408a49

0x408b02 MOV	%R13,%R10

0x408b05 MOV	-0x38(%RBP),%R9

0x408b09 MOV	-0x40(%RBP),%RBX

0x408b0d MOV	-0x48(%RBP),%R13

0x408b11 MOV	%R9,%RSI

0x408b14 MOV	$0x407c00,%ECX

0x408b19 MOV	$0x38,%EDX

0x408b1e MOV	%R10,-0x38(%RBP)

0x408b22 MOV	%R14,%RDI

0x408b25 CALL	4032e0 <qsort@plt>

0x408b2a MOV	-0x38(%RBP),%RSI

0x408b2e MOV	(%R14),%EAX

0x408b31 MOV	%RBX,%RCX

0x408b34 NOT	%RCX

0x408b37 LEA	0x1(%RBX),%R8

0x408b3b VMOVSD	0x8(%R14),%XMM8

0x408b41 ADD	%R12D,%ECX

0x408b44 AND	$0x1,%ECX

0x408b47 MOV	0x8(%RSI),%R15

0x408b4b MOV	0x10(%RSI),%R11

0x408b4f MOV	0x18(%RSI),%RDI

0x408b53 MOV	0x20(%RSI),%R9

0x408b57 MOV	%EAX,(%R15,%RBX,4)

0x408b5b MOV	0x4(%R14),%EDX

0x408b5f LEA	0x38(%R14),%RAX

0x408b63 MOV	%EDX,(%R11,%RBX,4)

0x408b67 VMOVSD	%XMM8,(%RDI,%R13,1)

0x408b6d LEA	0x18(%R13),%RDX

0x408b71 VMOVSD	0x10(%R14),%XMM9

0x408b77 VMOVSD	%XMM9,0x8(%RDI,%R13,1)

0x408b7e VMOVSD	0x18(%R14),%XMM10

0x408b84 VMOVSD	%XMM10,0x10(%RDI,%R13,1)

0x408b8b VMOVSD	0x20(%R14),%XMM11

0x408b91 VMOVSD	%XMM11,(%R9,%R13,1)

0x408b97 VMOVSD	0x28(%R14),%XMM12

0x408b9d VMOVSD	%XMM12,0x8(%R9,%R13,1)

0x408ba4 VMOVSD	0x30(%R14),%XMM13

0x408baa VMOVSD	%XMM13,0x10(%R9,%R13,1)

0x408bb1 CMP	%R8D,%R12D

0x408bb4 JLE	408cee

0x408bba TEST	%ECX,%ECX

0x408bbc JE	408c30

0x408bbe MOV	(%RAX),%R10D

0x408bc1 VMOVSD	0x8(%RAX),%XMM14

0x408bc6 MOV	%R10D,(%R15,%R8,4)

0x408bca MOV	0x4(%RAX),%ESI

0x408bcd MOV	%ESI,(%R11,%R8,4)

0x408bd1 LEA	0x2(%RBX),%R8

0x408bd5 VMOVSD	%XMM14,(%RDI,%RDX,1)

0x408bda VMOVSD	0x10(%RAX),%XMM15

0x408bdf VMOVSD	%XMM15,0x8(%RDI,%RDX,1)

0x408be5 VMOVSD	0x18(%RAX),%XMM2

0x408bea VMOVSD	%XMM2,0x10(%RDI,%RDX,1)

0x408bf0 VMOVSD	0x20(%RAX),%XMM0

0x408bf5 VMOVSD	%XMM0,(%R9,%RDX,1)

0x408bfb VMOVSD	0x28(%RAX),%XMM1

0x408c00 VMOVSD	%XMM1,0x8(%R9,%RDX,1)

0x408c07 VMOVSD	0x30(%RAX),%XMM3

0x408c0c LEA	0x70(%R14),%RAX

0x408c10 VMOVSD	%XMM3,0x10(%R9,%RDX,1)

0x408c17 LEA	0x30(%R13),%RDX

0x408c1b CMP	%R8D,%R12D

0x408c1e JLE	408cee

0x408c24 NOPW	%CS:(%RAX,%RAX,1)

0x408c2f NOP

(58) 0x408c30 MOV	(%RAX),%R14D

(58) 0x408c33 LEA	0x1(%R8),%R13

(58) 0x408c37 VMOVSD	0x8(%RAX),%XMM4

(58) 0x408c3c ADD	$0x70,%RAX

(58) 0x408c40 MOV	%R14D,(%R15,%R8,4)

(58) 0x408c44 MOV	-0x6c(%RAX),%EBX

(58) 0x408c47 MOV	%EBX,(%R11,%R8,4)

(58) 0x408c4b MOV	-0x38(%RAX),%ECX

(58) 0x408c4e VMOVSD	%XMM4,(%RDI,%RDX,1)

(58) 0x408c53 VMOVSD	-0x60(%RAX),%XMM5

(58) 0x408c58 VMOVSD	%XMM5,0x8(%RDI,%RDX,1)

(58) 0x408c5e VMOVSD	-0x58(%RAX),%XMM6

(58) 0x408c63 VMOVSD	%XMM6,0x10(%RDI,%RDX,1)

(58) 0x408c69 VMOVSD	-0x50(%RAX),%XMM7

(58) 0x408c6e VMOVSD	%XMM7,(%R9,%RDX,1)

(58) 0x408c74 VMOVSD	-0x48(%RAX),%XMM8

(58) 0x408c79 VMOVSD	%XMM8,0x8(%R9,%RDX,1)

(58) 0x408c80 VMOVSD	-0x40(%RAX),%XMM9

(58) 0x408c85 VMOVSD	%XMM9,0x10(%R9,%RDX,1)

(58) 0x408c8c ADD	$0x2,%R8

(58) 0x408c90 VMOVSD	-0x30(%RAX),%XMM10

(58) 0x408c95 MOV	%ECX,(%R15,%R13,4)

(58) 0x408c99 MOV	-0x34(%RAX),%R10D

(58) 0x408c9d MOV	%R10D,(%R11,%R13,4)

(58) 0x408ca1 VMOVSD	%XMM10,0x18(%RDI,%RDX,1)

(58) 0x408ca7 VMOVSD	-0x28(%RAX),%XMM11

(58) 0x408cac VMOVSD	%XMM11,0x20(%RDI,%RDX,1)

(58) 0x408cb2 VMOVSD	-0x20(%RAX),%XMM12

(58) 0x408cb7 VMOVSD	%XMM12,0x28(%RDI,%RDX,1)

(58) 0x408cbd VMOVSD	-0x18(%RAX),%XMM13

(58) 0x408cc2 VMOVSD	%XMM13,0x18(%R9,%RDX,1)

(58) 0x408cc9 VMOVSD	-0x10(%RAX),%XMM14

(58) 0x408cce VMOVSD	%XMM14,0x20(%R9,%RDX,1)

(58) 0x408cd5 VMOVSD	-0x8(%RAX),%XMM15

(58) 0x408cda VMOVSD	%XMM15,0x28(%R9,%RDX,1)

(58) 0x408ce1 ADD	$0x30,%RDX

(58) 0x408ce5 CMP	%R8D,%R12D

(58) 0x408ce8 JG	408c30

0x408cee LEA	-0x28(%RBP),%RSP

0x408cf2 POP	%RBX

0x408cf3 POP	%R12

0x408cf5 POP	%R13

0x408cf7 POP	%R14

0x408cf9 POP	%R15

0x408cfb POP	%RBP

0x408cfc RET

0x408cfd NOPL	(%RAX)

0x408d00 MOV	(%R15,%R8,4),%ECX

0x408d04 VMOVSD	(%RDI,%RDX,1),%XMM6

0x408d09 MOV	%ECX,(%RAX)

0x408d0b MOV	(%R11,%R8,4),%R8D

0x408d0f VMOVSD	%XMM6,0x8(%RAX)

0x408d14 VMOVSD	0x8(%RDI,%RDX,1),%XMM7

0x408d1a VMOVSD	%XMM7,0x10(%RAX)

0x408d1f VMOVSD	0x10(%RDI,%RDX,1),%XMM8

0x408d25 VMOVSD	%XMM8,0x18(%RAX)

0x408d2a VMOVSD	(%RSI,%RDX,1),%XMM9

0x408d2f VMOVSD	%XMM9,0x20(%RAX)

0x408d34 VMOVSD	0x8(%RSI,%RDX,1),%XMM10

0x408d3a VMOVSD	%XMM10,0x28(%RAX)

0x408d3f VMOVSD	0x10(%RSI,%RDX,1),%XMM11

0x408d45 LEA	0x30(%R13),%RDX

0x408d49 VMOVSD	%XMM11,0x30(%RAX)

0x408d4e MOV	%R8D,0x4(%RAX)

0x408d52 LEA	0x2(%RBX),%R8

0x408d56 LEA	0x70(%RSP),%RAX

0x408d5b CMP	%R8D,%R12D

0x408d5e JG	408a37

0x408d64 JMP	408b11

0x408d69 NOPL	(%RAX)

0x408d70 MOV	%RSP,%RDI

0x408d73 MOV	$0x407c00,%ECX

0x408d78 MOV	$0x38,%EDX

0x408d7d MOV	%R9,%RSI

0x408d80 CALL	4032e0 <qsort@plt>

0x408d85 LEA	-0x28(%RBP),%RSP

0x408d89 POP	%RBX

0x408d8a POP	%R12

0x408d8c POP	%R13

0x408d8e POP	%R14

0x408d90 POP	%R15

0x408d92 POP	%RBP

0x408d93 RET

0x408d94 NOPW	%CS:(%RAX,%RAX,1)

0x408d9e XCHG	%AX,%AX

Coverage (%)	Name	Source Location	Module
►12.53+	redistributeAtoms._omp_fn.0	timestep.c:154	exec
○	gomp_thread_start	team.c:130	libgomp.so.1.0.0
►12.51+	redistributeAtoms._omp_fn.0	timestep.c:154	exec
○	gomp_thread_start	team.c:130	libgomp.so.1.0.0
►12.49+	redistributeAtoms._omp_fn.0	timestep.c:154	exec
○	gomp_thread_start	team.c:130	libgomp.so.1.0.0
►12.49+	redistributeAtoms._omp_fn.0	timestep.c:154	exec
○	gomp_thread_start	team.c:130	libgomp.so.1.0.0
►12.49+	redistributeAtoms._omp_fn.0	timestep.c:154	exec
○	gomp_thread_start	team.c:130	libgomp.so.1.0.0
►12.12+	redistributeAtoms._omp_fn.0	timestep.c:154	exec
○	gomp_thread_start	team.c:130	libgomp.so.1.0.0
►11.96+	redistributeAtoms._omp_fn.0	timestep.c:154	exec
○	gomp_thread_start	team.c:130	libgomp.so.1.0.0
►11.77+	redistributeAtoms._omp_fn.0	timestep.c:154	exec
○	gomp_thread_start	team.c:130	libgomp.so.1.0.0

min	med	avg	max

Percentile Index	10	20	30	40	50	60	70	80	90	100
Value

min	med	avg	max

Percentile Index	10	20	30	40	50	60	70	80	90	100
Value

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 0.45% of application time for run gcc_10

Source file and lines	haloExchange.c:614-645
Module	exec

The function is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:614-645.

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Code clean check

Detected a slowdown caused by scalar integer instructions (typically used for address computation). By removing them, you can lower the cost of an iteration from 21.25 to 12.75 cycles (1.67x speedup).

Workaround

Try to reorganize arrays of structures to structures of arrays
Consider to permute loops (see vectorization gain report)

Vectorization

Your function is not vectorized. Only 10% 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 21.25 to 2.12 cycles (10.00x 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.

qsort@plt: 2 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:632
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:632

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 332 bytes. The binary function is storing 256 bytes.

General properties

nb instructions	174
nb uops	170
loop length	729
used x86 registers	16
used mmx registers	0
used xmm registers	16
used ymm registers	0
used zmm registers	0
nb stack references	14

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	21.25 cycles
front end	21.25 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	9.17	9.17	9.17	9.17	9.17	9.17	21.00	21.00	21.00	21.00	0.00	0.00	0.00	0.00	12.00	12.00
cycles	9.17	9.17	9.17	9.17	9.17	9.17	21.00	21.00	21.00	21.00	0.00	0.00	0.00	0.00	12.00	12.00

Execution ports to units layout:

P0: ALU
P1: ALU
P2: ALU
P3: ALU, BRU
P4: ALU, BRU
P5: ALU, BRU
P6 (512 bits): store address, load
P7 (512 bits): store address, load
P8 (512 bits): store address, load
P9 (512 bits): store address, load
P10 (512 bits): VPU, DIV/SQRT
P11 (512 bits): VPU, DIV/SQRT
P12 (512 bits): VPU
P13 (512 bits): VPU
P14 (512 bits): FP store data
P15 (512 bits): FP store data

Cycles executing div or sqrt instructions

Cycles summary

Front-end	21.25
Dispatch	21.00
Overall L1	21.25

Vectorization ratios

INT

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

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

INT+FP

all	0%
load	0%
store	0%
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

INT

all	8%
load	6%
store	8%
mul	12%
add-sub	8%
fma	NA (no fma vectorizable/vectorized instructions)
other	8%

all	12%
load	12%
store	12%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	NA (no other vectorizable/vectorized instructions)

INT+FP

all	10%
load	12%
store	11%
mul	12%
add-sub	8%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	8%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 21.25 cycles. At this rate:

6% of peak load performance is reached (15.62 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
9% of peak store performance is reached (12.05 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Performance is limited by instruction throughput (loading/decoding program instructions to execution core) (front-end is a bottleneck).

ASM code

In the binary file, the address of the function is: 408950

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P14	P15	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOVSXD %EDX,%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
PUSH %R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R14	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R13	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOV %RDX,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
SUB $0x28,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV 0x78(%RSI),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
SAL $0x6,%EBX	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	1	0.33	N/A
MOVSXD (%RAX,%RDX,4),%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	0	0	1	0.25	N/A
IMUL $0x38,%R9,%RCX	1	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	3	0.33	scal (12.5%)
MOV %R9,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
ADD %EBX,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
ADD $0xf,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (12.5%)
AND $-0x10,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (12.5%)
SUB %RCX,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV %RSP,%R14	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CMP %R12D,%EBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
JGE 408d70 <sortAtomsInCell+0x420>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.33-1	N/A
MOV 0x8(%RDI),%R15	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOVSXD %EBX,%RBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV 0x10(%RDI),%R11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV %RDI,%R10	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV 0x20(%R10),%RSI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV 0x18(%RDI),%RDI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
LEA (%RDX,%RDX,2),%R13	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV %RBX,%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
NOT %RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (12.5%)
LEA 0x38(%RSP),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
SAL $0x9,%R13	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	1	0.33	N/A
MOV (%R15,%RBX,4),%R8D	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
ADD %R12D,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
VMOVSD (%RDI,%R13,1),%XMM2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
AND $0x1,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
MOV %R8D,(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
MOV (%R11,%RBX,4),%EDX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
LEA 0x1(%RBX),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
VMOVSD %XMM2,0x8(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x8(%RDI,%R13,1),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM0,0x10(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x10(%RDI,%R13,1),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM1,0x18(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD (%RSI,%R13,1),%XMM3	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM3,0x20(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x8(%RSI,%R13,1),%XMM4	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM4,0x28(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x10(%RSI,%R13,1),%XMM5	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM5,0x30(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
MOV %EDX,0x4(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
LEA 0x18(%R13),%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
CMP %R8D,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
JLE 408b11 <sortAtomsInCell+0x1c1>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.33-1	N/A
TEST %ECX,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
JNE 408d00 <sortAtomsInCell+0x3b0>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.33-1	N/A
MOV %R13,-0x48(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (12.5%)
MOV %R10,%R13	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R11,%R10	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R9,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (12.5%)
MOV %RBX,-0x40(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (12.5%)
MOV %R13,%R10	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV -0x38(%RBP),%R9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV -0x40(%RBP),%RBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV -0x48(%RBP),%R13	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV %R9,%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV $0x407c00,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
MOV $0x38,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV %R10,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (12.5%)
MOV %R14,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CALL 4032e0 <qsort@plt>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0.50	N/A
MOV -0x38(%RBP),%RSI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV (%R14),%EAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV %RBX,%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
NOT %RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (12.5%)
LEA 0x1(%RBX),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
VMOVSD 0x8(%R14),%XMM8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
ADD %R12D,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
AND $0x1,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
MOV 0x8(%RSI),%R15	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV 0x10(%RSI),%R11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV 0x18(%RSI),%RDI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV 0x20(%RSI),%R9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV %EAX,(%R15,%RBX,4)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
MOV 0x4(%R14),%EDX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
LEA 0x38(%R14),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV %EDX,(%R11,%RBX,4)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
VMOVSD %XMM8,(%RDI,%R13,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
LEA 0x18(%R13),%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
VMOVSD 0x10(%R14),%XMM9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM9,0x8(%RDI,%R13,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x18(%R14),%XMM10	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM10,0x10(%RDI,%R13,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x20(%R14),%XMM11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM11,(%R9,%R13,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x28(%R14),%XMM12	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM12,0x8(%R9,%R13,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x30(%R14),%XMM13	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM13,0x10(%R9,%R13,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
CMP %R8D,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
JLE 408cee <sortAtomsInCell+0x39e>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.33-1	N/A
TEST %ECX,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
JE 408c30 <sortAtomsInCell+0x2e0>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.33-1	N/A
MOV (%RAX),%R10D	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
VMOVSD 0x8(%RAX),%XMM14	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
MOV %R10D,(%R15,%R8,4)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
MOV 0x4(%RAX),%ESI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV %ESI,(%R11,%R8,4)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
LEA 0x2(%RBX),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
VMOVSD %XMM14,(%RDI,%RDX,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x10(%RAX),%XMM15	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM15,0x8(%RDI,%RDX,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x18(%RAX),%XMM2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM2,0x10(%RDI,%RDX,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x20(%RAX),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM0,(%R9,%RDX,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x28(%RAX),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM1,0x8(%R9,%RDX,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x30(%RAX),%XMM3	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
LEA 0x70(%R14),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
VMOVSD %XMM3,0x10(%R9,%RDX,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
LEA 0x30(%R13),%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
CMP %R8D,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
JLE 408cee <sortAtomsInCell+0x39e>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.33-1	N/A
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
NOP	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
LEA -0x28(%RBP),%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
POP %RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R13	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R14	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
RET	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	1	0.50	N/A
NOPL (%RAX)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
MOV (%R15,%R8,4),%ECX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	scal (6.3%)
VMOVSD (%RDI,%RDX,1),%XMM6	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
MOV %ECX,(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
MOV (%R11,%R8,4),%R8D	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
VMOVSD %XMM6,0x8(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x8(%RDI,%RDX,1),%XMM7	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM7,0x10(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x10(%RDI,%RDX,1),%XMM8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM8,0x18(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD (%RSI,%RDX,1),%XMM9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM9,0x20(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x8(%RSI,%RDX,1),%XMM10	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM10,0x28(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x10(%RSI,%RDX,1),%XMM11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
LEA 0x30(%R13),%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
VMOVSD %XMM11,0x30(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
MOV %R8D,0x4(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
LEA 0x2(%RBX),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
LEA 0x70(%RSP),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
CMP %R8D,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
JG 408a37 <sortAtomsInCell+0xe7>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.33-1	N/A
JMP 408b11 <sortAtomsInCell+0x1c1>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.50	N/A
NOPL (%RAX)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
MOV %RSP,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV $0x407c00,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
MOV $0x38,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV %R9,%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CALL 4032e0 <qsort@plt>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0.50	N/A
LEA -0x28(%RBP),%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
POP %RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R13	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R14	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
RET	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	1	0.50	N/A
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
XCHG %AX,%AX	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A

The code analyzed by CQA in that panel excludes loops and represents 0.45% of application time for run gcc_10

Source file and lines	haloExchange.c:614-645
Module	exec

gain
potential
hint
expert

Code clean check

Workaround

Try to reorganize arrays of structures to structures of arrays
Consider to permute loops (see vectorization gain report)

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

qsort@plt: 2 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:632
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:632

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 332 bytes. The binary function is storing 256 bytes.

General properties

nb instructions	174
nb uops	170
loop length	729
used x86 registers	16
used mmx registers	0
used xmm registers	16
used ymm registers	0
used zmm registers	0
nb stack references	14

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	21.25 cycles
front end	21.25 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	9.17	9.17	9.17	9.17	9.17	9.17	21.00	21.00	21.00	21.00	0.00	0.00	0.00	0.00	12.00	12.00
cycles	9.17	9.17	9.17	9.17	9.17	9.17	21.00	21.00	21.00	21.00	0.00	0.00	0.00	0.00	12.00	12.00

Execution ports to units layout:

P0: ALU
P1: ALU
P2: ALU
P3: ALU, BRU
P4: ALU, BRU
P5: ALU, BRU
P6 (512 bits): store address, load
P7 (512 bits): store address, load
P8 (512 bits): store address, load
P9 (512 bits): store address, load
P10 (512 bits): VPU, DIV/SQRT
P11 (512 bits): VPU, DIV/SQRT
P12 (512 bits): VPU
P13 (512 bits): VPU
P14 (512 bits): FP store data
P15 (512 bits): FP store data

Cycles executing div or sqrt instructions

Cycles summary

Front-end	21.25
Dispatch	21.00
Overall L1	21.25

Vectorization ratios

INT

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

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

INT+FP

all	0%
load	0%
store	0%
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

INT

all	8%
load	6%
store	8%
mul	12%
add-sub	8%
fma	NA (no fma vectorizable/vectorized instructions)
other	8%

all	12%
load	12%
store	12%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	NA (no add-sub vectorizable/vectorized instructions)
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	NA (no other vectorizable/vectorized instructions)

INT+FP

all	10%
load	12%
store	11%
mul	12%
add-sub	8%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	8%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 21.25 cycles. At this rate:

6% of peak load performance is reached (15.62 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
9% of peak store performance is reached (12.05 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Performance is limited by instruction throughput (loading/decoding program instructions to execution core) (front-end is a bottleneck).

ASM code

In the binary file, the address of the function is: 408950

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P14	P15	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOVSXD %EDX,%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV %RSP,%RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
PUSH %R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R14	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R13	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
PUSH %RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOV %RDX,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
SUB $0x28,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV 0x78(%RSI),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
SAL $0x6,%EBX	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	1	0.33	N/A
MOVSXD (%RAX,%RDX,4),%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	0	0	1	0.25	N/A
IMUL $0x38,%R9,%RCX	1	0.33	0.33	0.33	0	0	0	0	0	0	0	0	0	3	0.33	scal (12.5%)
MOV %R9,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
ADD %EBX,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
ADD $0xf,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (12.5%)
AND $-0x10,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (12.5%)
SUB %RCX,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV %RSP,%R14	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CMP %R12D,%EBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
JGE 408d70 <sortAtomsInCell+0x420>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.33-1	N/A
MOV 0x8(%RDI),%R15	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOVSXD %EBX,%RBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV 0x10(%RDI),%R11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV %RDI,%R10	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV 0x20(%R10),%RSI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV 0x18(%RDI),%RDI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
LEA (%RDX,%RDX,2),%R13	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV %RBX,%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
NOT %RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (12.5%)
LEA 0x38(%RSP),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
SAL $0x9,%R13	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	1	0.33	N/A
MOV (%R15,%RBX,4),%R8D	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
ADD %R12D,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
VMOVSD (%RDI,%R13,1),%XMM2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
AND $0x1,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
MOV %R8D,(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
MOV (%R11,%RBX,4),%EDX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
LEA 0x1(%RBX),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
VMOVSD %XMM2,0x8(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x8(%RDI,%R13,1),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM0,0x10(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x10(%RDI,%R13,1),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM1,0x18(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD (%RSI,%R13,1),%XMM3	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM3,0x20(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x8(%RSI,%R13,1),%XMM4	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM4,0x28(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x10(%RSI,%R13,1),%XMM5	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM5,0x30(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
MOV %EDX,0x4(%RSP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
LEA 0x18(%R13),%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
CMP %R8D,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
JLE 408b11 <sortAtomsInCell+0x1c1>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.33-1	N/A
TEST %ECX,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
JNE 408d00 <sortAtomsInCell+0x3b0>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.33-1	N/A
MOV %R13,-0x48(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (12.5%)
MOV %R10,%R13	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R11,%R10	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R9,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (12.5%)
MOV %RBX,-0x40(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (12.5%)
MOV %R13,%R10	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV -0x38(%RBP),%R9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV -0x40(%RBP),%RBX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV -0x48(%RBP),%R13	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV %R9,%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV $0x407c00,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
MOV $0x38,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV %R10,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (12.5%)
MOV %R14,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CALL 4032e0 <qsort@plt>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0.50	N/A
MOV -0x38(%RBP),%RSI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV (%R14),%EAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV %RBX,%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
NOT %RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (12.5%)
LEA 0x1(%RBX),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
VMOVSD 0x8(%R14),%XMM8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
ADD %R12D,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
AND $0x1,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
MOV 0x8(%RSI),%R15	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV 0x10(%RSI),%R11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV 0x18(%RSI),%RDI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV 0x20(%RSI),%R9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV %EAX,(%R15,%RBX,4)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
MOV 0x4(%R14),%EDX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
LEA 0x38(%R14),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV %EDX,(%R11,%RBX,4)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
VMOVSD %XMM8,(%RDI,%R13,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
LEA 0x18(%R13),%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
VMOVSD 0x10(%R14),%XMM9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM9,0x8(%RDI,%R13,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x18(%R14),%XMM10	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM10,0x10(%RDI,%R13,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x20(%R14),%XMM11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM11,(%R9,%R13,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x28(%R14),%XMM12	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM12,0x8(%R9,%R13,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x30(%R14),%XMM13	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM13,0x10(%R9,%R13,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
CMP %R8D,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
JLE 408cee <sortAtomsInCell+0x39e>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.33-1	N/A
TEST %ECX,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
JE 408c30 <sortAtomsInCell+0x2e0>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.33-1	N/A
MOV (%RAX),%R10D	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
VMOVSD 0x8(%RAX),%XMM14	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
MOV %R10D,(%R15,%R8,4)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
MOV 0x4(%RAX),%ESI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
MOV %ESI,(%R11,%R8,4)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
LEA 0x2(%RBX),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
VMOVSD %XMM14,(%RDI,%RDX,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x10(%RAX),%XMM15	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM15,0x8(%RDI,%RDX,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x18(%RAX),%XMM2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM2,0x10(%RDI,%RDX,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x20(%RAX),%XMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM0,(%R9,%RDX,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x28(%RAX),%XMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM1,0x8(%R9,%RDX,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x30(%RAX),%XMM3	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
LEA 0x70(%R14),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
VMOVSD %XMM3,0x10(%R9,%RDX,1)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
LEA 0x30(%R13),%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
CMP %R8D,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
JLE 408cee <sortAtomsInCell+0x39e>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.33-1	N/A
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
NOP	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
LEA -0x28(%RBP),%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
POP %RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R13	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R14	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
RET	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	1	0.50	N/A
NOPL (%RAX)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
MOV (%R15,%R8,4),%ECX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	scal (6.3%)
VMOVSD (%RDI,%RDX,1),%XMM6	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
MOV %ECX,(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
MOV (%R11,%R8,4),%R8D	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	3	0.25	N/A
VMOVSD %XMM6,0x8(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x8(%RDI,%RDX,1),%XMM7	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM7,0x10(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x10(%RDI,%RDX,1),%XMM8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM8,0x18(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD (%RSI,%RDX,1),%XMM9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM9,0x20(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x8(%RSI,%RDX,1),%XMM10	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM10,0x28(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
VMOVSD 0x10(%RSI,%RDX,1),%XMM11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	1	0.50	scal (12.5%)
LEA 0x30(%R13),%RDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
VMOVSD %XMM11,0x30(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0.50	0.50	1	0.50	scal (12.5%)
MOV %R8D,0x4(%RAX)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	0	0	4	0.50	scal (6.3%)
LEA 0x2(%RBX),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
LEA 0x70(%RSP),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
CMP %R8D,%R12D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
JG 408a37 <sortAtomsInCell+0xe7>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.33-1	N/A
JMP 408b11 <sortAtomsInCell+0x1c1>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	0	0.50	N/A
NOPL (%RAX)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
MOV %RSP,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV $0x407c00,%ECX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	scal (6.3%)
MOV $0x38,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
MOV %R9,%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CALL 4032e0 <qsort@plt>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0	0	0.50	N/A
LEA -0x28(%RBP),%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	0	0	1	0.17	N/A
POP %RBX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R12	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R13	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R14	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %R15	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
POP %RBP	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.25	N/A
RET	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0	1	0.50	N/A
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.13	N/A
XCHG %AX,%AX	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A

Name	Coverage (%)	Time (s)
▼sortAtomsInCell–	3.38	0.24
○Loop 59 - haloExchange.c:621-630 - exec	2.42	0.11
○Loop 58 - haloExchange.c:633-642 - exec	0.52	0.03

Report Configuration

Code clean check

Workaround

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

Code clean check

Workaround

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