OV - exec - Function sortAtomsInCell

	Function: sortAtomsInCell	Module: exec	Source: haloExchange.c:614-645	Coverage (incl. loops): 1.05% \| (excl. loops): 0.03%

/home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/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: }

0x20bae0 PUSH	%RBP

0x20bae1 MOV	%RSP,%RBP

0x20bae4 PUSH	%R15

0x20bae6 PUSH	%R14

0x20bae8 PUSH	%R13

0x20baea PUSH	%R12

0x20baec PUSH	%RBX

0x20baed SUB	$0x38,%RSP

0x20baf1 MOV	0x78(%RSI),%RAX

0x20baf5 MOVSXD	%EDX,%RCX

0x20baf8 MOV	%RSP,%R14

0x20bafb MOVSXD	(%RAX,%RCX,4),%RAX

0x20baff MOV	%EAX,%ESI

0x20bb01 IMUL	$0x38,%RSI,%RCX

0x20bb05 ADD	$0xf,%RCX

0x20bb09 AND	$-0x10,%RCX

0x20bb0d SUB	%RCX,%R14

0x20bb10 MOV	%R14,%RSP

0x20bb13 TEST	%RAX,%RAX

0x20bb16 JLE	20bbfe

0x20bb1c MOV	%RDI,%RBX

0x20bb1f SAL	$0x6,%EDX

0x20bb22 MOV	0x8(%RDI),%RCX

0x20bb26 MOV	0x10(%RDI),%RDI

0x20bb2a MOV	0x18(%RBX),%R8

0x20bb2e MOV	0x20(%RBX),%R9

0x20bb32 ADD	%EDX,%EAX

0x20bb34 MOVSXD	%EDX,%R12

0x20bb37 MOV	%R14,%RDX

0x20bb3a MOVSXD	%EAX,%R15

0x20bb3d LEA	(%R12,%R12,1),%RAX

0x20bb41 MOV	%R12,%R10

0x20bb44 LEA	0x4(%RAX,%RAX,2),%RAX

0x20bb49 NOPL	(%RAX)

(59) 0x20bb50 MOV	(%RCX,%R10,4),%R11D

(59) 0x20bb54 MOV	%R11D,(%RDX)

(59) 0x20bb57 MOV	(%RDI,%R10,4),%R11D

(59) 0x20bb5b INC	%R10

(59) 0x20bb5e MOV	%R11D,0x4(%RDX)

(59) 0x20bb62 VMOVSD	-0x10(%R8,%RAX,4),%XMM0

(59) 0x20bb69 VMOVSD	%XMM0,0x8(%RDX)

(59) 0x20bb6e VMOVSD	-0x8(%R8,%RAX,4),%XMM0

(59) 0x20bb75 VMOVSD	%XMM0,0x10(%RDX)

(59) 0x20bb7a VMOVSD	(%R8,%RAX,4),%XMM0

(59) 0x20bb80 VMOVSD	%XMM0,0x18(%RDX)

(59) 0x20bb85 VMOVSD	-0x10(%R9,%RAX,4),%XMM0

(59) 0x20bb8c VMOVSD	%XMM0,0x20(%RDX)

(59) 0x20bb91 VMOVSD	-0x8(%R9,%RAX,4),%XMM0

(59) 0x20bb98 VMOVSD	%XMM0,0x28(%RDX)

(59) 0x20bb9d VMOVSD	(%R9,%RAX,4),%XMM0

(59) 0x20bba3 ADD	$0x6,%RAX

(59) 0x20bba7 VMOVSD	%XMM0,0x30(%RDX)

(59) 0x20bbac ADD	$0x38,%RDX

(59) 0x20bbb0 CMP	%R15,%R10

(59) 0x20bbb3 JL	20bb50

0x20bbb5 LEA	0x524(%RIP),%RCX

0x20bbbc MOV	$0x38,%EDX

0x20bbc1 MOV	%R14,%RDI

0x20bbc4 CALL	2125d0 <@plt_start@+0x1d0>

0x20bbc9 MOV	0x8(%RBX),%RAX

0x20bbcd MOV	0x10(%RBX),%RCX

0x20bbd1 MOV	0x18(%RBX),%RDX

0x20bbd5 MOV	0x20(%RBX),%RSI

0x20bbd9 MOV	%R12,%RBX

0x20bbdc OR	$0x1,%RBX

0x20bbe0 MOV	%R15,%R9

0x20bbe3 CMP	%R15,%RBX

0x20bbe6 CMOVG	%RBX,%R9

0x20bbea MOV	%R9,%RDI

0x20bbed SUB	%R12,%RDI

0x20bbf0 CMP	$0x3,%RDI

0x20bbf4 JAE	20bc1a

0x20bbf6 XOR	%R8D,%R8D

0x20bbf9 JMP	20c040

0x20bbfe LEA	0x4db(%RIP),%RCX

0x20bc05 MOV	$0x38,%EDX

0x20bc0a MOV	%R14,%RDI

0x20bc0d MOV	%RAX,%RSI

0x20bc10 CALL	2125d0 <@plt_start@+0x1d0>

0x20bc15 JMP	20c0c5

0x20bc1a CMP	%R15,%RBX

0x20bc1d LEA	(%RAX,%R12,4),%R8

0x20bc21 CMOVLE	%R15,%RBX

0x20bc25 IMUL	$0x38,%R12,%R10

0x20bc29 MOV	%R8,-0x40(%RBP)

0x20bc2d LEA	(%RAX,%RBX,4),%R8

0x20bc31 MOV	%R8,-0x48(%RBP)

0x20bc35 LEA	(%RCX,%R12,4),%R8

0x20bc39 MOV	%R8,-0x50(%RBP)

0x20bc3d LEA	(%RCX,%RBX,4),%R8

0x20bc41 MOV	%R8,-0x38(%RBP)

0x20bc45 IMUL	$0x38,%RBX,%R8

0x20bc49 SAL	$0x3,%RBX

0x20bc4d LEA	(%RBX,%RBX,2),%RBX

0x20bc51 LEA	(%RDX,%RBX,1),%R11

0x20bc55 SUB	%R10,%R8

0x20bc58 ADD	%RSI,%RBX

0x20bc5b LEA	(,%R12,8),%R10

0x20bc63 MOV	%R11,-0x58(%RBP)

0x20bc67 MOV	-0x40(%RBP),%R11

0x20bc6b LEA	-0x30(%R14,%R8,1),%R13

0x20bc70 ADD	%R14,%R8

0x20bc73 CMP	-0x38(%RBP),%R11

0x20bc77 MOV	-0x48(%RBP),%R11

0x20bc7b LEA	(%R10,%R10,2),%R10

0x20bc7f SETBB	-0x2a(%RBP)

0x20bc83 CMP	%R11,-0x50(%RBP)

0x20bc87 LEA	(%RDX,%R10,1),%R11

0x20bc8b SETBB	-0x29(%RBP)

0x20bc8f CMP	%R13,-0x40(%RBP)

0x20bc93 SETBB	-0x2c(%RBP)

0x20bc97 CMP	-0x48(%RBP),%R14

0x20bc9b SETBB	-0x2b(%RBP)

0x20bc9f CMP	%R13,-0x50(%RBP)

0x20bca3 LEA	(%RSI,%R10,1),%R13

0x20bca7 SETBB	-0x2e(%RBP)

0x20bcab CMP	-0x38(%RBP),%R14

0x20bcaf SETBB	-0x2d(%RBP)

0x20bcb3 CMP	%RBX,%R11

0x20bcb6 SETBB	-0x48(%RBP)

0x20bcba CMP	-0x58(%RBP),%R13

0x20bcbe SETBB	-0x40(%RBP)

0x20bcc2 CMP	%R8,%R11

0x20bcc5 LEA	0x8(%R14),%R11

0x20bcc9 SETBB	-0x38(%RBP)

0x20bccd CMP	-0x58(%RBP),%R11

0x20bcd1 SETBB	-0x50(%RBP)

0x20bcd5 CMP	%R8,%R13

0x20bcd8 SETB	%R13B

0x20bcdc CMP	%RBX,%R11

0x20bcdf SETB	%R11B

0x20bce3 XOR	%R8D,%R8D

0x20bce6 TEST	%R11B,%R13B

0x20bce9 JNE	20c040

0x20bcef MOVZX	-0x50(%RBP),%R11D

0x20bcf4 AND	%R11B,-0x38(%RBP)

0x20bcf8 JNE	20c040

0x20bcfe MOVZX	-0x40(%RBP),%R11D

0x20bd03 AND	%R11B,-0x48(%RBP)

0x20bd07 JNE	20c040

0x20bd0d MOVZX	-0x2d(%RBP),%R11D

0x20bd12 AND	%R11B,-0x2e(%RBP)

0x20bd16 JNE	20c040

0x20bd1c MOVZX	-0x2b(%RBP),%R11D

0x20bd21 AND	%R11B,-0x2c(%RBP)

0x20bd25 JNE	20c040

0x20bd2b MOVZX	-0x29(%RBP),%R11D

0x20bd30 AND	%R11B,-0x2a(%RBP)

0x20bd34 JNE	20c040

0x20bd3a CMP	$0x9,%RDI

0x20bd3e JAE	20bd48

0x20bd40 XOR	%R8D,%R8D

0x20bd43 JMP	20bf70

0x20bd48 VMOVAPD	-0x7352(%RIP),%ZMM0

0x20bd52 VMOVAPD	-0x731c(%RIP),%ZMM1

0x20bd5c VMOVAPD	-0x72e6(%RIP),%ZMM2

0x20bd66 VMOVAPD	-0x72b0(%RIP),%ZMM3

0x20bd70 VMOVAPD	-0x727a(%RIP),%ZMM4

0x20bd7a VMOVAPD	-0x7244(%RIP),%ZMM5

0x20bd84 VMOVAPD	-0x720e(%RIP),%ZMM6

0x20bd8e VMOVAPD	-0x71d8(%RIP),%ZMM7

0x20bd98 VMOVAPD	-0x71a2(%RIP),%ZMM8

0x20bda2 VMOVAPD	-0x716c(%RIP),%ZMM9

0x20bdac VMOVAPD	-0x7136(%RIP),%ZMM10

0x20bdb6 MOV	%R9D,%R8D

0x20bdb9 AND	$0x7,%R8D

0x20bdbd MOV	$0x8,%R13D

0x20bdc3 MOV	%R9,%R11

0x20bdc6 CMOVNE	%R8,%R13

0x20bdca MOV	%RDI,%R8

0x20bdcd SUB	%R13,%R8

0x20bdd0 SUB	%R13,%R11

0x20bdd3 MOV	%R13,%RBX

0x20bdd6 MOV	%R13,-0x38(%RBP)

0x20bdda MOV	%R12,%R13

0x20bddd MOV	%R14,%RBX

(61) 0x20bde0 VMOVD	(%RBX),%XMM11

(61) 0x20bde4 VMOVD	0xe0(%RBX),%XMM12

(61) 0x20bdec VPINSRD	$0x1,0x38(%RBX),%XMM11,%XMM11

(61) 0x20bdf3 VPINSRD	$0x1,0x118(%RBX),%XMM12,%XMM12

(61) 0x20bdfd VPINSRD	$0x2,0x70(%RBX),%XMM11,%XMM11

(61) 0x20be04 VPINSRD	$0x2,0x150(%RBX),%XMM12,%XMM12

(61) 0x20be0e VPINSRD	$0x3,0xa8(%RBX),%XMM11,%XMM11

(61) 0x20be18 VPINSRD	$0x3,0x188(%RBX),%XMM12,%XMM12

(61) 0x20be22 VMOVDQU	%XMM12,0x10(%RAX,%R13,4)

(61) 0x20be29 VMOVDQU	%XMM11,(%RAX,%R13,4)

(61) 0x20be2f VMOVD	0x4(%RBX),%XMM11

(61) 0x20be34 VMOVD	0xe4(%RBX),%XMM12

(61) 0x20be3c VPINSRD	$0x1,0x3c(%RBX),%XMM11,%XMM11

(61) 0x20be43 VPINSRD	$0x1,0x11c(%RBX),%XMM12,%XMM12

(61) 0x20be4d VPINSRD	$0x2,0x74(%RBX),%XMM11,%XMM11

(61) 0x20be54 VPINSRD	$0x2,0x154(%RBX),%XMM12,%XMM12

(61) 0x20be5e VPINSRD	$0x3,0xac(%RBX),%XMM11,%XMM11

(61) 0x20be68 VPINSRD	$0x3,0x18c(%RBX),%XMM12,%XMM12

(61) 0x20be72 VMOVDQU	%XMM12,0x10(%RCX,%R13,4)

(61) 0x20be79 VMOVDQU	%XMM11,(%RCX,%R13,4)

(61) 0x20be7f ADD	$0x8,%R13

(61) 0x20be83 VMOVUPD	0x8(%RBX),%ZMM11

(61) 0x20be8d VMOVUPD	0x88(%RBX),%ZMM13

(61) 0x20be97 VMOVUPD	0x148(%RBX),%ZMM17

(61) 0x20bea1 VMOVUPD	0x48(%RBX),%ZMM12

(61) 0x20beab VMOVUPD	0xc8(%RBX),%ZMM14

(61) 0x20beb5 VMOVUPD	0x108(%RBX),%ZMM16

(61) 0x20bebf VMOVUPD	0x188(%RBX),%ZMM15

(61) 0x20bec9 ADD	$0x1c0,%RBX

(61) 0x20bed0 VMOVAPD	%ZMM17,%ZMM18

(61) 0x20bed6 VMOVAPD	%ZMM11,%ZMM20

(61) 0x20bedc VPERMT2PD	%ZMM16,%ZMM0,%ZMM18

(61) 0x20bee2 VMOVAPD	%ZMM13,%ZMM19

(61) 0x20bee8 VPERMT2PD	%ZMM14,%ZMM2,%ZMM19

(61) 0x20beee VPERMT2PD	%ZMM16,%ZMM5,%ZMM17

(61) 0x20bef4 VPERMT2PD	%ZMM12,%ZMM7,%ZMM11

(61) 0x20befa VPERMT2PD	%ZMM13,%ZMM9,%ZMM14

(61) 0x20bf00 VPERMT2PD	%ZMM12,%ZMM4,%ZMM20

(61) 0x20bf06 VPERMT2PD	%ZMM15,%ZMM1,%ZMM18

(61) 0x20bf0c VPERMT2PD	%ZMM16,%ZMM3,%ZMM19

(61) 0x20bf12 VPERMT2PD	%ZMM15,%ZMM6,%ZMM17

(61) 0x20bf18 VPERMT2PD	%ZMM13,%ZMM8,%ZMM11

(61) 0x20bf1e VPERMT2PD	%ZMM16,%ZMM10,%ZMM14

(61) 0x20bf24 VMOVUPD	%ZMM20,(%RDX,%R10,1)

(61) 0x20bf2b VMOVUPD	%ZMM18,0x80(%RDX,%R10,1)

(61) 0x20bf33 VMOVUPD	%ZMM19,0x40(%RDX,%R10,1)

(61) 0x20bf3b VMOVUPD	%ZMM17,0x80(%RSI,%R10,1)

(61) 0x20bf43 VMOVUPD	%ZMM14,0x40(%RSI,%R10,1)

(61) 0x20bf4b VMOVUPD	%ZMM11,(%RSI,%R10,1)

(61) 0x20bf52 ADD	$0xc0,%R10

(61) 0x20bf59 CMP	%R13,%R11

(61) 0x20bf5c JNE	20bde0

0x20bf62 CMPL	$0x2,-0x38(%RBP)

0x20bf66 JA	20bf70

0x20bf68 ADD	%R8,%R12

0x20bf6b JMP	20c040

0x20bf70 VMOVAPD	-0x75f8(%RIP),%YMM0

0x20bf78 AND	$0x1,%R9D

0x20bf7c MOV	$0x2,%R10D

0x20bf82 CMOVNE	%R9,%R10

0x20bf86 SUB	%R10,%RDI

0x20bf89 IMUL	$0x38,%R8,%R10

0x20bf8d ADD	%R12,%R8

0x20bf90 LEA	(,%R8,8),%R11

0x20bf98 LEA	(%RDI,%R12,1),%R9

0x20bf9c LEA	(%R11,%R11,2),%R11

0x20bfa0 ADD	%R14,%R10

0x20bfa3 NOPW	%CS:(%RAX,%RAX,1)

(62) 0x20bfb0 VMOVD	(%R10),%XMM1

(62) 0x20bfb5 VPINSRD	$0x1,0x38(%R10),%XMM1,%XMM1

(62) 0x20bfbc VMOVQ	%XMM1,(%RAX,%R8,4)

(62) 0x20bfc2 VMOVD	0x4(%R10),%XMM1

(62) 0x20bfc8 VPINSRD	$0x1,0x3c(%R10),%XMM1,%XMM1

(62) 0x20bfcf VMOVQ	%XMM1,(%RCX,%R8,4)

(62) 0x20bfd5 ADD	$0x2,%R8

(62) 0x20bfd9 VMOVUPD	0x8(%R10),%ZMM1

(62) 0x20bfe3 VMOVUPD	0x48(%R10),%YMM3

(62) 0x20bfe9 VMOVUPD	0x8(%R10),%YMM2

(62) 0x20bfef VMOVUPD	0x58(%R10),%XMM5

(62) 0x20bff5 VMOVUPD	0x48(%R10),%XMM4

(62) 0x20bffb VBLENDPD	$0x8,0x28(%R10),%YMM2,%YMM2

(62) 0x20c002 VSHUFPD	$0x1,0x68(%R10),%XMM5,%XMM5

(62) 0x20c009 ADD	$0x70,%R10

(62) 0x20c00d VPERMT2PD	%ZMM3,%ZMM0,%ZMM1

(62) 0x20c013 VMOVUPD	%XMM4,0x20(%RDX,%R11,1)

(62) 0x20c01a VMOVUPD	%YMM2,(%RDX,%R11,1)

(62) 0x20c020 VMOVUPD	%XMM5,0x20(%RSI,%R11,1)

(62) 0x20c027 VMOVUPD	%YMM1,(%RSI,%R11,1)

(62) 0x20c02d ADD	$0x30,%R11

(62) 0x20c031 CMP	%R8,%R9

(62) 0x20c034 JNE	20bfb0

0x20c03a MOV	%R9,%R12

0x20c03d MOV	%RDI,%R8

0x20c040 LEA	(%R12,%R12,2),%RDI

0x20c044 IMUL	$0x38,%R8,%R8

0x20c048 LEA	0x10(,%RDI,8),%RDI

0x20c050 ADD	%R8,%R14

0x20c053 NOPW	%CS:(%RAX,%RAX,1)

(60) 0x20c060 MOV	(%R14),%R8D

(60) 0x20c063 MOV	%R8D,(%RAX,%R12,4)

(60) 0x20c067 MOV	0x4(%R14),%R8D

(60) 0x20c06b MOV	%R8D,(%RCX,%R12,4)

(60) 0x20c06f INC	%R12

(60) 0x20c072 VMOVSD	0x8(%R14),%XMM0

(60) 0x20c078 VMOVSD	%XMM0,-0x10(%RDX,%RDI,1)

(60) 0x20c07e VMOVSD	0x10(%R14),%XMM0

(60) 0x20c084 VMOVSD	%XMM0,-0x8(%RDX,%RDI,1)

(60) 0x20c08a VMOVSD	0x18(%R14),%XMM0

(60) 0x20c090 VMOVSD	%XMM0,(%RDX,%RDI,1)

(60) 0x20c095 VMOVSD	0x20(%R14),%XMM0

(60) 0x20c09b VMOVSD	%XMM0,-0x10(%RSI,%RDI,1)

(60) 0x20c0a1 VMOVSD	0x28(%R14),%XMM0

(60) 0x20c0a7 VMOVSD	%XMM0,-0x8(%RSI,%RDI,1)

(60) 0x20c0ad VMOVSD	0x30(%R14),%XMM0

(60) 0x20c0b3 ADD	$0x38,%R14

(60) 0x20c0b7 VMOVSD	%XMM0,(%RSI,%RDI,1)

(60) 0x20c0bc ADD	$0x18,%RDI

(60) 0x20c0c0 CMP	%R15,%R12

(60) 0x20c0c3 JL	20c060

0x20c0c5 LEA	-0x28(%RBP),%RSP

0x20c0c9 POP	%RBX

0x20c0ca POP	%R12

0x20c0cc POP	%R13

0x20c0ce POP	%R14

0x20c0d0 POP	%R15

0x20c0d2 POP	%RBP

0x20c0d3 VZEROUPPER

0x20c0d6 RET

0x20c0d7 NOPW	(%RAX,%RAX,1)

Coverage (%)	Name	Source Location	Module

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 0.03% of application time for run orig_0

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

The function is defined in /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/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 23.75 to 5.50 cycles (4.32x speedup).

Workaround

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

Vectorization

Your function is probably not vectorized. Only 33% 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 23.75 to 20.59 cycles (1.15x speedup).

Details

Store and arithmetical 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.

unknown: 2 occurrences►
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:632
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:632

Complex instructions

Detected COMPLEX INSTRUCTIONS.

Details

These instructions generate more than one micro-operation and only one of them can be decoded during a cycle and the extra micro-operations increase pressure on execution units.

CMOVG: 1 occurrences►
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
CMOVLE: 1 occurrences►
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
LEA: 2 occurrences►
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:621
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633

Conversion instructions

Detected expensive conversion instructions.

Details

MOVZX: 5 occurrences►
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633

Workaround

Avoid mixing data with different types. In particular, check if the type of constants is the same as array elements.

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 906 bytes. The binary function is storing 63 bytes.

General properties

nb instructions	188
nb uops	190
loop length	810
used x86 registers	16
used mmx registers	0
used xmm registers	0
used ymm registers	1
used zmm registers	11
nb stack references	12

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	23.75 cycles
front end	23.75 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	20.00	20.00	20.00	20.00	20.00	20.00	15.75	15.75	15.75	15.75	0.00	0.00	0.00	0.00	0.00	0.00
cycles	20.00	20.00	20.00	20.00	20.00	20.00	15.75	15.75	15.75	15.75	0.00	0.00	0.00	0.00	0.00	0.00

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	23.75
Dispatch	20.00
Overall L1	23.75

Vectorization ratios

INT

all	2%
load	0%
store	0%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	NA (no fma vectorizable/vectorized instructions)
other	3%

all	100%
load	100%
store	NA (no store vectorizable/vectorized instructions)
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	27%
load	54%
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	3%

Vector efficiency ratios

INT

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

all	95%
load	95%
store	NA (no store vectorizable/vectorized instructions)
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	33%
load	57%
store	12%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	12%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	12%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 23.75 cycles. At this rate:

14% of peak load performance is reached (38.15 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
2% of peak store performance is reached (2.65 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). By removing all these bottlenecks, you can lower the cost of an iteration from 23.75 to 20.00 cycles (1.19x speedup).

ASM code

In the binary file, the address of the function is: 20bae0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOV %RSP,%RBP	1	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	1	0.50	N/A
PUSH %R14	1	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	1	0.50	N/A
PUSH %R12	1	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	1	0.50	N/A
SUB $0x38,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x78(%RSI),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOVSXD %EDX,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %RSP,%R14	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOVSXD (%RAX,%RCX,4),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	N/A
MOV %EAX,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
IMUL $0x38,%RSI,%RCX	1	0.33	0.33	0.33	0	0	0	0	0	0	0	3	0.33	N/A
ADD $0xf,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
AND $-0x10,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
SUB %RCX,%R14	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %R14,%RSP	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
TEST %RAX,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JLE 20bbfe <sortAtomsInCell+0x11e>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOV %RDI,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
SAL $0x6,%EDX	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.33	N/A
MOV 0x8(%RDI),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x10(%RDI),%RDI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x18(%RBX),%R8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x20(%RBX),%R9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (12.5%)
ADD %EDX,%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOVSXD %EDX,%R12	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %R14,%RDX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOVSXD %EAX,%R15	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
LEA (%R12,%R12,1),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %R12,%R10	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA 0x4(%RAX,%RAX,2),%RAX	2	0	0	0.50	0.50	0	0	0.25	0.25	0.25	0.25	2	0.50	N/A
NOPL (%RAX)	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
LEA 0x524(%RIP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV $0x38,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %R14,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CALL 2125d0 <@plt_start@+0x1d0>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0.50	N/A
MOV 0x8(%RBX),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x10(%RBX),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x18(%RBX),%RDX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x20(%RBX),%RSI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV %R12,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
OR $0x1,%RBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %R15,%R9	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
CMP %R15,%RBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
CMOVG %RBX,%R9	2	0.33	0.33	0.33	0.33	0.33	0.33	0	0	0	0	1	0.33	scal (12.5%)
MOV %R9,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
SUB %R12,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
CMP $0x3,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JAE 20bc1a <sortAtomsInCell+0x13a>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0.25	N/A
JMP 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A
LEA 0x4db(%RIP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV $0x38,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %R14,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %RAX,%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CALL 2125d0 <@plt_start@+0x1d0>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0.50	N/A
JMP 20c0c5 <sortAtomsInCell+0x5e5>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A
CMP %R15,%RBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
LEA (%RAX,%R12,4),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
CMOVLE %R15,%RBX	2	0.33	0.33	0.33	0.33	0.33	0.33	0	0	0	0	1	0.33	N/A
IMUL $0x38,%R12,%R10	1	0.33	0.33	0.33	0	0	0	0	0	0	0	3	0.33	N/A
MOV %R8,-0x40(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
LEA (%RAX,%RBX,4),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %R8,-0x48(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
LEA (%RCX,%R12,4),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %R8,-0x50(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
LEA (%RCX,%RBX,4),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %R8,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
IMUL $0x38,%RBX,%R8	1	0.33	0.33	0.33	0	0	0	0	0	0	0	3	0.33	N/A
SAL $0x3,%RBX	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.33	N/A
LEA (%RBX,%RBX,2),%RBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA (%RDX,%RBX,1),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
SUB %R10,%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
ADD %RSI,%RBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA (,%R12,8),%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %R11,-0x58(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOV -0x40(%RBP),%R11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
LEA -0x30(%R14,%R8,1),%R13	2	0	0	0.50	0.50	0	0	0.25	0.25	0.25	0.25	2	0.50	N/A
ADD %R14,%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
CMP -0x38(%RBP),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
MOV -0x48(%RBP),%R11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
LEA (%R10,%R10,2),%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
SETBB -0x2a(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP %R11,-0x50(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
LEA (%RDX,%R10,1),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
SETBB -0x29(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP %R13,-0x40(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
SETBB -0x2c(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP -0x48(%RBP),%R14	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
SETBB -0x2b(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP %R13,-0x50(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
LEA (%RSI,%R10,1),%R13	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
SETBB -0x2e(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP -0x38(%RBP),%R14	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
SETBB -0x2d(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP %RBX,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
SETBB -0x48(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP -0x58(%RBP),%R13	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
SETBB -0x40(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP %R8,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
LEA 0x8(%R14),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
SETBB -0x38(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP -0x58(%RBP),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
SETBB -0x50(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP %R8,%R13	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
SETB %R13B	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
CMP %RBX,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
SETB %R11B	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0.25	N/A
TEST %R11B,%R13B	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
JNE 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOVZX -0x50(%RBP),%R11D	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	N/A
AND %R11B,-0x38(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
JNE 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOVZX -0x40(%RBP),%R11D	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	N/A
AND %R11B,-0x48(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
JNE 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOVZX -0x2d(%RBP),%R11D	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	N/A
AND %R11B,-0x2e(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
JNE 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOVZX -0x2b(%RBP),%R11D	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	N/A
AND %R11B,-0x2c(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
JNE 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOVZX -0x29(%RBP),%R11D	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	N/A
AND %R11B,-0x2a(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
JNE 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
CMP $0x9,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JAE 20bd48 <sortAtomsInCell+0x268>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0.25	N/A
JMP 20bf70 <sortAtomsInCell+0x490>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A
VMOVAPD -0x7352(%RIP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x731c(%RIP),%ZMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x72e6(%RIP),%ZMM2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x72b0(%RIP),%ZMM3	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x727a(%RIP),%ZMM4	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x7244(%RIP),%ZMM5	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x720e(%RIP),%ZMM6	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x71d8(%RIP),%ZMM7	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x71a2(%RIP),%ZMM8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x716c(%RIP),%ZMM9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x7136(%RIP),%ZMM10	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
MOV %R9D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
AND $0x7,%R8D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $0x8,%R13D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
MOV %R9,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CMOVNE %R8,%R13	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
MOV %RDI,%R8	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
SUB %R13,%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
SUB %R13,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %R13,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R13,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOV %R12,%R13	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
MOV %R14,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CMPL $0x2,-0x38(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (6.3%)
JA 20bf70 <sortAtomsInCell+0x490>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
ADD %R8,%R12	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
JMP 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A
VMOVAPD -0x75f8(%RIP),%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (50.0%)
AND $0x1,%R9D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
MOV $0x2,%R10D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
CMOVNE %R9,%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
SUB %R10,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
IMUL $0x38,%R8,%R10	1	0.33	0.33	0.33	0	0	0	0	0	0	0	3	0.33	N/A
ADD %R12,%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA (,%R8,8),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA (%RDI,%R12,1),%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA (%R11,%R11,2),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
ADD %R14,%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
MOV %R9,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %RDI,%R8	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA (%R12,%R12,2),%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
IMUL $0x38,%R8,%R8	1	0.33	0.33	0.33	0	0	0	0	0	0	0	3	0.33	N/A
LEA 0x10(,%RDI,8),%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
ADD %R8,%R14	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
NOPW %CS:(%RAX,%RAX,1)	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	1	0.25	N/A
POP %RBX	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R12	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R13	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R14	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R15	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %RBP	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
RET	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50	N/A
NOPW (%RAX,%RAX,1)	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 0.03% of application time for run orig_0

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

The function is defined in /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/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 23.75 to 5.50 cycles (4.32x speedup).

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

unknown: 2 occurrences►
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:632
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:632

Complex instructions

Detected COMPLEX INSTRUCTIONS.

Details

These instructions generate more than one micro-operation and only one of them can be decoded during a cycle and the extra micro-operations increase pressure on execution units.

CMOVG: 1 occurrences►
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
CMOVLE: 1 occurrences►
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
LEA: 2 occurrences►
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:621
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633

Conversion instructions

Detected expensive conversion instructions.

Details

MOVZX: 5 occurrences►
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /home/eoseret/qaas_runs_ZEN5/174-043-3878/intel/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633

Workaround

Avoid mixing data with different types. In particular, check if the type of constants is the same as array elements.

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 906 bytes. The binary function is storing 63 bytes.

General properties

nb instructions	188
nb uops	190
loop length	810
used x86 registers	16
used mmx registers	0
used xmm registers	0
used ymm registers	1
used zmm registers	11
nb stack references	12

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	23.75 cycles
front end	23.75 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	P15
uops	20.00	20.00	20.00	20.00	20.00	20.00	15.75	15.75	15.75	15.75	0.00	0.00	0.00	0.00	0.00	0.00
cycles	20.00	20.00	20.00	20.00	20.00	20.00	15.75	15.75	15.75	15.75	0.00	0.00	0.00	0.00	0.00	0.00

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	23.75
Dispatch	20.00
Overall L1	23.75

Vectorization ratios

INT

all	2%
load	0%
store	0%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	0%
fma	NA (no fma vectorizable/vectorized instructions)
other	3%

all	100%
load	100%
store	NA (no store vectorizable/vectorized instructions)
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	27%
load	54%
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	3%

Vector efficiency ratios

INT

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

all	95%
load	95%
store	NA (no store vectorizable/vectorized instructions)
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	33%
load	57%
store	12%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	12%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	12%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 23.75 cycles. At this rate:

14% of peak load performance is reached (38.15 out of 256.00 bytes loaded per cycle (GB/s @ 1GHz))
2% of peak store performance is reached (2.65 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

ASM code

In the binary file, the address of the function is: 20bae0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	Latency	Recip. throughput	Vectorization
PUSH %RBP	1	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
MOV %RSP,%RBP	1	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	1	0.50	N/A
PUSH %R14	1	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	1	0.50	N/A
PUSH %R12	1	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	1	0.50	N/A
SUB $0x38,%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
MOV 0x78(%RSI),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOVSXD %EDX,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %RSP,%R14	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOVSXD (%RAX,%RCX,4),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	N/A
MOV %EAX,%ESI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
IMUL $0x38,%RSI,%RCX	1	0.33	0.33	0.33	0	0	0	0	0	0	0	3	0.33	N/A
ADD $0xf,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
AND $-0x10,%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
SUB %RCX,%R14	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %R14,%RSP	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
TEST %RAX,%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JLE 20bbfe <sortAtomsInCell+0x11e>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOV %RDI,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
SAL $0x6,%EDX	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.33	N/A
MOV 0x8(%RDI),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x10(%RDI),%RDI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x18(%RBX),%R8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x20(%RBX),%R9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	scal (12.5%)
ADD %EDX,%EAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOVSXD %EDX,%R12	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %R14,%RDX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOVSXD %EAX,%R15	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
LEA (%R12,%R12,1),%RAX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %R12,%R10	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA 0x4(%RAX,%RAX,2),%RAX	2	0	0	0.50	0.50	0	0	0.25	0.25	0.25	0.25	2	0.50	N/A
NOPL (%RAX)	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
LEA 0x524(%RIP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV $0x38,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %R14,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CALL 2125d0 <@plt_start@+0x1d0>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0.50	N/A
MOV 0x8(%RBX),%RAX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x10(%RBX),%RCX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x18(%RBX),%RDX	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV 0x20(%RBX),%RSI	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
MOV %R12,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
OR $0x1,%RBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %R15,%R9	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
CMP %R15,%RBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
CMOVG %RBX,%R9	2	0.33	0.33	0.33	0.33	0.33	0.33	0	0	0	0	1	0.33	scal (12.5%)
MOV %R9,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
SUB %R12,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
CMP $0x3,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JAE 20bc1a <sortAtomsInCell+0x13a>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0.25	N/A
JMP 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A
LEA 0x4db(%RIP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV $0x38,%EDX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %R14,%RDI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %RAX,%RSI	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CALL 2125d0 <@plt_start@+0x1d0>	2	0	0	0	0.33	0.33	0.33	0.25	0.25	0.25	0.25	0	0.50	N/A
JMP 20c0c5 <sortAtomsInCell+0x5e5>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A
CMP %R15,%RBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
LEA (%RAX,%R12,4),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
CMOVLE %R15,%RBX	2	0.33	0.33	0.33	0.33	0.33	0.33	0	0	0	0	1	0.33	N/A
IMUL $0x38,%R12,%R10	1	0.33	0.33	0.33	0	0	0	0	0	0	0	3	0.33	N/A
MOV %R8,-0x40(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
LEA (%RAX,%RBX,4),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %R8,-0x48(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
LEA (%RCX,%R12,4),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %R8,-0x50(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
LEA (%RCX,%RBX,4),%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %R8,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
IMUL $0x38,%RBX,%R8	1	0.33	0.33	0.33	0	0	0	0	0	0	0	3	0.33	N/A
SAL $0x3,%RBX	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.33	N/A
LEA (%RBX,%RBX,2),%RBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA (%RDX,%RBX,1),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
SUB %R10,%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
ADD %RSI,%RBX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA (,%R12,8),%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
MOV %R11,-0x58(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOV -0x40(%RBP),%R11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
LEA -0x30(%R14,%R8,1),%R13	2	0	0	0.50	0.50	0	0	0.25	0.25	0.25	0.25	2	0.50	N/A
ADD %R14,%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
CMP -0x38(%RBP),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
MOV -0x48(%RBP),%R11	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.25	N/A
LEA (%R10,%R10,2),%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
SETBB -0x2a(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP %R11,-0x50(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
LEA (%RDX,%R10,1),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
SETBB -0x29(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP %R13,-0x40(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
SETBB -0x2c(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP -0x48(%RBP),%R14	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
SETBB -0x2b(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP %R13,-0x50(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
LEA (%RSI,%R10,1),%R13	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
SETBB -0x2e(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP -0x38(%RBP),%R14	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
SETBB -0x2d(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP %RBX,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
SETBB -0x48(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP -0x58(%RBP),%R13	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
SETBB -0x40(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP %R8,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
LEA 0x8(%R14),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
SETBB -0x38(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP -0x58(%RBP),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (12.5%)
SETBB -0x50(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
CMP %R8,%R13	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
SETB %R13B	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
CMP %RBX,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
SETB %R11B	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0.25	N/A
TEST %R11B,%R13B	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
JNE 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOVZX -0x50(%RBP),%R11D	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	N/A
AND %R11B,-0x38(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
JNE 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOVZX -0x40(%RBP),%R11D	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	N/A
AND %R11B,-0x48(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
JNE 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOVZX -0x2d(%RBP),%R11D	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	N/A
AND %R11B,-0x2e(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
JNE 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOVZX -0x2b(%RBP),%R11D	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	N/A
AND %R11B,-0x2c(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
JNE 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
MOVZX -0x29(%RBP),%R11D	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	N/A
AND %R11B,-0x2a(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.50	N/A
JNE 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
CMP $0x9,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
JAE 20bd48 <sortAtomsInCell+0x268>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0.25	N/A
JMP 20bf70 <sortAtomsInCell+0x490>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A
VMOVAPD -0x7352(%RIP),%ZMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x731c(%RIP),%ZMM1	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x72e6(%RIP),%ZMM2	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x72b0(%RIP),%ZMM3	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x727a(%RIP),%ZMM4	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x7244(%RIP),%ZMM5	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x720e(%RIP),%ZMM6	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x71d8(%RIP),%ZMM7	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x71a2(%RIP),%ZMM8	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x716c(%RIP),%ZMM9	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
VMOVAPD -0x7136(%RIP),%ZMM10	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (100.0%)
MOV %R9D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
AND $0x7,%R8D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV $0x8,%R13D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
MOV %R9,%R11	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CMOVNE %R8,%R13	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (12.5%)
MOV %RDI,%R8	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
SUB %R13,%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
SUB %R13,%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
MOV %R13,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %R13,-0x38(%RBP)	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	4	0.50	scal (12.5%)
MOV %R12,%R13	1	0	0	0	0	0	0	0	0	0	0	0	0.17	scal (12.5%)
MOV %R14,%RBX	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
CMPL $0x2,-0x38(%RBP)	1	0.17	0.17	0.17	0.17	0.17	0.17	0.25	0.25	0.25	0.25	1	0.25	scal (6.3%)
JA 20bf70 <sortAtomsInCell+0x490>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50-1	N/A
ADD %R8,%R12	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
JMP 20c040 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	1	N/A
VMOVAPD -0x75f8(%RIP),%YMM0	1	0	0	0	0	0	0	0.25	0.25	0.25	0.25	3	0.50	vect (50.0%)
AND $0x1,%R9D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	scal (6.3%)
MOV $0x2,%R10D	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
CMOVNE %R9,%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
SUB %R10,%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
IMUL $0x38,%R8,%R10	1	0.33	0.33	0.33	0	0	0	0	0	0	0	3	0.33	N/A
ADD %R12,%R8	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
LEA (,%R8,8),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA (%RDI,%R12,1),%R9	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
LEA (%R11,%R11,2),%R11	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
ADD %R14,%R10	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
NOPW %CS:(%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A
MOV %R9,%R12	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV %RDI,%R8	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
LEA (%R12,%R12,2),%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
IMUL $0x38,%R8,%R8	1	0.33	0.33	0.33	0	0	0	0	0	0	0	3	0.33	N/A
LEA 0x10(,%RDI,8),%RDI	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.25	N/A
ADD %R8,%R14	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
NOPW %CS:(%RAX,%RAX,1)	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	1	0.25	N/A
POP %RBX	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R12	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R13	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R14	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %R15	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
POP %RBP	1	0	0	0	0	0	0	0	0	0	0	1	0.33	N/A
VZEROUPPER	1	0	0	0	0	0	0	0	0	0	0	0	1	vect (25.0%)
RET	1	0	0	0	0.33	0.33	0.33	0	0	0	0	1	0.50	N/A
NOPW (%RAX,%RAX,1)	0	0	0	0	0	0	0	0	0	0	0	0	0.09	N/A

Name	Coverage (%)	Time (s)
▼sortAtomsInCell–	1.05	0.30
○Loop 59 - haloExchange.c:621-630 - exec	1.01	0.28
○Loop 61 - haloExchange.c:633-642 - exec	0.01	0.01
○Loop 62 - haloExchange.c:633-642 - exec	0.00	0.01
○Loop 60 - haloExchange.c:633-642 - exec	0.00	0.01

Report Configuration

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

CALL instructions

Details

Complex instructions

Details

Conversion instructions

Details

Workaround

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

Complex instructions

Details

Conversion instructions

Details

Workaround

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