OV - exec - Function sortAtomsInCell

	Function: sortAtomsInCell	Module: exec	Source: haloExchange.c:614-645	Coverage (incl. loops): 3.19% \| (excl. loops): 0.04%

/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: }

0x20c320 PUSH	%RBP

0x20c321 MOV	%RSP,%RBP

0x20c324 PUSH	%R15

0x20c326 PUSH	%R14

0x20c328 PUSH	%R13

0x20c32a PUSH	%R12

0x20c32c PUSH	%RBX

0x20c32d SUB	$0x38,%RSP

0x20c331 MOV	0x78(%RSI),%RAX

0x20c335 MOVSXD	%EDX,%RCX

0x20c338 MOV	%RSP,%R14

0x20c33b MOVSXD	(%RAX,%RCX,4),%RAX

0x20c33f MOV	%EAX,%ESI

0x20c341 IMUL	$0x38,%RSI,%RCX

0x20c345 ADD	$0xf,%RCX

0x20c349 AND	$-0x10,%RCX

0x20c34d SUB	%RCX,%R14

0x20c350 MOV	%R14,%RSP

0x20c353 TEST	%RAX,%RAX

0x20c356 JLE	20c43e

0x20c35c MOV	%RDI,%RBX

0x20c35f SAL	$0x6,%EDX

0x20c362 MOV	0x8(%RDI),%RCX

0x20c366 MOV	0x10(%RDI),%RDI

0x20c36a MOV	0x18(%RBX),%R8

0x20c36e MOV	0x20(%RBX),%R9

0x20c372 ADD	%EDX,%EAX

0x20c374 MOVSXD	%EDX,%R12

0x20c377 MOV	%R14,%RDX

0x20c37a MOVSXD	%EAX,%R15

0x20c37d LEA	(%R12,%R12,1),%RAX

0x20c381 MOV	%R12,%R10

0x20c384 LEA	0x4(%RAX,%RAX,2),%RAX

0x20c389 NOPL	(%RAX)

(59) 0x20c390 MOV	(%RCX,%R10,4),%R11D

(59) 0x20c394 MOV	%R11D,(%RDX)

(59) 0x20c397 MOV	(%RDI,%R10,4),%R11D

(59) 0x20c39b INC	%R10

(59) 0x20c39e MOV	%R11D,0x4(%RDX)

(59) 0x20c3a2 VMOVSD	-0x10(%R8,%RAX,4),%XMM0

(59) 0x20c3a9 VMOVSD	%XMM0,0x8(%RDX)

(59) 0x20c3ae VMOVSD	-0x8(%R8,%RAX,4),%XMM0

(59) 0x20c3b5 VMOVSD	%XMM0,0x10(%RDX)

(59) 0x20c3ba VMOVSD	(%R8,%RAX,4),%XMM0

(59) 0x20c3c0 VMOVSD	%XMM0,0x18(%RDX)

(59) 0x20c3c5 VMOVSD	-0x10(%R9,%RAX,4),%XMM0

(59) 0x20c3cc VMOVSD	%XMM0,0x20(%RDX)

(59) 0x20c3d1 VMOVSD	-0x8(%R9,%RAX,4),%XMM0

(59) 0x20c3d8 VMOVSD	%XMM0,0x28(%RDX)

(59) 0x20c3dd VMOVSD	(%R9,%RAX,4),%XMM0

(59) 0x20c3e3 ADD	$0x6,%RAX

(59) 0x20c3e7 VMOVSD	%XMM0,0x30(%RDX)

(59) 0x20c3ec ADD	$0x38,%RDX

(59) 0x20c3f0 CMP	%R15,%R10

(59) 0x20c3f3 JL	20c390

0x20c3f5 LEA	0x524(%RIP),%RCX

0x20c3fc MOV	$0x38,%EDX

0x20c401 MOV	%R14,%RDI

0x20c404 CALL	212e30 <@plt_start@+0x1d0>

0x20c409 MOV	0x8(%RBX),%RAX

0x20c40d MOV	0x10(%RBX),%RCX

0x20c411 MOV	0x18(%RBX),%RDX

0x20c415 MOV	0x20(%RBX),%RSI

0x20c419 MOV	%R12,%RBX

0x20c41c OR	$0x1,%RBX

0x20c420 MOV	%R15,%R9

0x20c423 CMP	%R15,%RBX

0x20c426 CMOVG	%RBX,%R9

0x20c42a MOV	%R9,%RDI

0x20c42d SUB	%R12,%RDI

0x20c430 CMP	$0x3,%RDI

0x20c434 JAE	20c45a

0x20c436 XOR	%R8D,%R8D

0x20c439 JMP	20c880

0x20c43e LEA	0x4db(%RIP),%RCX

0x20c445 MOV	$0x38,%EDX

0x20c44a MOV	%R14,%RDI

0x20c44d MOV	%RAX,%RSI

0x20c450 CALL	212e30 <@plt_start@+0x1d0>

0x20c455 JMP	20c905

0x20c45a CMP	%R15,%RBX

0x20c45d LEA	(%RAX,%R12,4),%R8

0x20c461 CMOVLE	%R15,%RBX

0x20c465 IMUL	$0x38,%R12,%R10

0x20c469 MOV	%R8,-0x40(%RBP)

0x20c46d LEA	(%RAX,%RBX,4),%R8

0x20c471 MOV	%R8,-0x48(%RBP)

0x20c475 LEA	(%RCX,%R12,4),%R8

0x20c479 MOV	%R8,-0x50(%RBP)

0x20c47d LEA	(%RCX,%RBX,4),%R8

0x20c481 MOV	%R8,-0x38(%RBP)

0x20c485 IMUL	$0x38,%RBX,%R8

0x20c489 SAL	$0x3,%RBX

0x20c48d LEA	(%RBX,%RBX,2),%RBX

0x20c491 LEA	(%RDX,%RBX,1),%R11

0x20c495 SUB	%R10,%R8

0x20c498 ADD	%RSI,%RBX

0x20c49b LEA	(,%R12,8),%R10

0x20c4a3 MOV	%R11,-0x58(%RBP)

0x20c4a7 MOV	-0x40(%RBP),%R11

0x20c4ab LEA	-0x30(%R14,%R8,1),%R13

0x20c4b0 ADD	%R14,%R8

0x20c4b3 CMP	-0x38(%RBP),%R11

0x20c4b7 MOV	-0x48(%RBP),%R11

0x20c4bb LEA	(%R10,%R10,2),%R10

0x20c4bf SETBB	-0x2a(%RBP)

0x20c4c3 CMP	%R11,-0x50(%RBP)

0x20c4c7 LEA	(%RDX,%R10,1),%R11

0x20c4cb SETBB	-0x29(%RBP)

0x20c4cf CMP	%R13,-0x40(%RBP)

0x20c4d3 SETBB	-0x2c(%RBP)

0x20c4d7 CMP	-0x48(%RBP),%R14

0x20c4db SETBB	-0x2b(%RBP)

0x20c4df CMP	%R13,-0x50(%RBP)

0x20c4e3 LEA	(%RSI,%R10,1),%R13

0x20c4e7 SETBB	-0x2e(%RBP)

0x20c4eb CMP	-0x38(%RBP),%R14

0x20c4ef SETBB	-0x2d(%RBP)

0x20c4f3 CMP	%RBX,%R11

0x20c4f6 SETBB	-0x48(%RBP)

0x20c4fa CMP	-0x58(%RBP),%R13

0x20c4fe SETBB	-0x40(%RBP)

0x20c502 CMP	%R8,%R11

0x20c505 LEA	0x8(%R14),%R11

0x20c509 SETBB	-0x38(%RBP)

0x20c50d CMP	-0x58(%RBP),%R11

0x20c511 SETBB	-0x50(%RBP)

0x20c515 CMP	%R8,%R13

0x20c518 SETB	%R13B

0x20c51c CMP	%RBX,%R11

0x20c51f SETB	%R11B

0x20c523 XOR	%R8D,%R8D

0x20c526 TEST	%R11B,%R13B

0x20c529 JNE	20c880

0x20c52f MOVZX	-0x50(%RBP),%R11D

0x20c534 AND	%R11B,-0x38(%RBP)

0x20c538 JNE	20c880

0x20c53e MOVZX	-0x40(%RBP),%R11D

0x20c543 AND	%R11B,-0x48(%RBP)

0x20c547 JNE	20c880

0x20c54d MOVZX	-0x2d(%RBP),%R11D

0x20c552 AND	%R11B,-0x2e(%RBP)

0x20c556 JNE	20c880

0x20c55c MOVZX	-0x2b(%RBP),%R11D

0x20c561 AND	%R11B,-0x2c(%RBP)

0x20c565 JNE	20c880

0x20c56b MOVZX	-0x29(%RBP),%R11D

0x20c570 AND	%R11B,-0x2a(%RBP)

0x20c574 JNE	20c880

0x20c57a CMP	$0x9,%RDI

0x20c57e JAE	20c588

0x20c580 XOR	%R8D,%R8D

0x20c583 JMP	20c7b0

0x20c588 VMOVAPD	-0x7352(%RIP),%ZMM0

0x20c592 VMOVAPD	-0x731c(%RIP),%ZMM1

0x20c59c VMOVAPD	-0x72e6(%RIP),%ZMM2

0x20c5a6 VMOVAPD	-0x72b0(%RIP),%ZMM3

0x20c5b0 VMOVAPD	-0x727a(%RIP),%ZMM4

0x20c5ba VMOVAPD	-0x7244(%RIP),%ZMM5

0x20c5c4 VMOVAPD	-0x720e(%RIP),%ZMM6

0x20c5ce VMOVAPD	-0x71d8(%RIP),%ZMM7

0x20c5d8 VMOVAPD	-0x71a2(%RIP),%ZMM8

0x20c5e2 VMOVAPD	-0x716c(%RIP),%ZMM9

0x20c5ec VMOVAPD	-0x7136(%RIP),%ZMM10

0x20c5f6 MOV	%R9D,%R8D

0x20c5f9 AND	$0x7,%R8D

0x20c5fd MOV	$0x8,%R13D

0x20c603 MOV	%R9,%R11

0x20c606 CMOVNE	%R8,%R13

0x20c60a MOV	%RDI,%R8

0x20c60d SUB	%R13,%R8

0x20c610 SUB	%R13,%R11

0x20c613 MOV	%R13,%RBX

0x20c616 MOV	%R13,-0x38(%RBP)

0x20c61a MOV	%R12,%R13

0x20c61d MOV	%R14,%RBX

(61) 0x20c620 VMOVD	(%RBX),%XMM11

(61) 0x20c624 VMOVD	0xe0(%RBX),%XMM12

(61) 0x20c62c VPINSRD	$0x1,0x38(%RBX),%XMM11,%XMM11

(61) 0x20c633 VPINSRD	$0x1,0x118(%RBX),%XMM12,%XMM12

(61) 0x20c63d VPINSRD	$0x2,0x70(%RBX),%XMM11,%XMM11

(61) 0x20c644 VPINSRD	$0x2,0x150(%RBX),%XMM12,%XMM12

(61) 0x20c64e VPINSRD	$0x3,0xa8(%RBX),%XMM11,%XMM11

(61) 0x20c658 VPINSRD	$0x3,0x188(%RBX),%XMM12,%XMM12

(61) 0x20c662 VMOVDQU	%XMM12,0x10(%RAX,%R13,4)

(61) 0x20c669 VMOVDQU	%XMM11,(%RAX,%R13,4)

(61) 0x20c66f VMOVD	0x4(%RBX),%XMM11

(61) 0x20c674 VMOVD	0xe4(%RBX),%XMM12

(61) 0x20c67c VPINSRD	$0x1,0x3c(%RBX),%XMM11,%XMM11

(61) 0x20c683 VPINSRD	$0x1,0x11c(%RBX),%XMM12,%XMM12

(61) 0x20c68d VPINSRD	$0x2,0x74(%RBX),%XMM11,%XMM11

(61) 0x20c694 VPINSRD	$0x2,0x154(%RBX),%XMM12,%XMM12

(61) 0x20c69e VPINSRD	$0x3,0xac(%RBX),%XMM11,%XMM11

(61) 0x20c6a8 VPINSRD	$0x3,0x18c(%RBX),%XMM12,%XMM12

(61) 0x20c6b2 VMOVDQU	%XMM12,0x10(%RCX,%R13,4)

(61) 0x20c6b9 VMOVDQU	%XMM11,(%RCX,%R13,4)

(61) 0x20c6bf ADD	$0x8,%R13

(61) 0x20c6c3 VMOVUPD	0x8(%RBX),%ZMM11

(61) 0x20c6cd VMOVUPD	0x88(%RBX),%ZMM13

(61) 0x20c6d7 VMOVUPD	0x148(%RBX),%ZMM17

(61) 0x20c6e1 VMOVUPD	0x48(%RBX),%ZMM12

(61) 0x20c6eb VMOVUPD	0xc8(%RBX),%ZMM14

(61) 0x20c6f5 VMOVUPD	0x108(%RBX),%ZMM16

(61) 0x20c6ff VMOVUPD	0x188(%RBX),%ZMM15

(61) 0x20c709 ADD	$0x1c0,%RBX

(61) 0x20c710 VMOVAPD	%ZMM17,%ZMM18

(61) 0x20c716 VMOVAPD	%ZMM11,%ZMM20

(61) 0x20c71c VPERMT2PD	%ZMM16,%ZMM0,%ZMM18

(61) 0x20c722 VMOVAPD	%ZMM13,%ZMM19

(61) 0x20c728 VPERMT2PD	%ZMM14,%ZMM2,%ZMM19

(61) 0x20c72e VPERMT2PD	%ZMM16,%ZMM5,%ZMM17

(61) 0x20c734 VPERMT2PD	%ZMM12,%ZMM7,%ZMM11

(61) 0x20c73a VPERMT2PD	%ZMM13,%ZMM9,%ZMM14

(61) 0x20c740 VPERMT2PD	%ZMM12,%ZMM4,%ZMM20

(61) 0x20c746 VPERMT2PD	%ZMM15,%ZMM1,%ZMM18

(61) 0x20c74c VPERMT2PD	%ZMM16,%ZMM3,%ZMM19

(61) 0x20c752 VPERMT2PD	%ZMM15,%ZMM6,%ZMM17

(61) 0x20c758 VPERMT2PD	%ZMM13,%ZMM8,%ZMM11

(61) 0x20c75e VPERMT2PD	%ZMM16,%ZMM10,%ZMM14

(61) 0x20c764 VMOVUPD	%ZMM20,(%RDX,%R10,1)

(61) 0x20c76b VMOVUPD	%ZMM18,0x80(%RDX,%R10,1)

(61) 0x20c773 VMOVUPD	%ZMM19,0x40(%RDX,%R10,1)

(61) 0x20c77b VMOVUPD	%ZMM17,0x80(%RSI,%R10,1)

(61) 0x20c783 VMOVUPD	%ZMM14,0x40(%RSI,%R10,1)

(61) 0x20c78b VMOVUPD	%ZMM11,(%RSI,%R10,1)

(61) 0x20c792 ADD	$0xc0,%R10

(61) 0x20c799 CMP	%R13,%R11

(61) 0x20c79c JNE	20c620

0x20c7a2 CMPL	$0x2,-0x38(%RBP)

0x20c7a6 JA	20c7b0

0x20c7a8 ADD	%R8,%R12

0x20c7ab JMP	20c880

0x20c7b0 VMOVAPD	-0x75d8(%RIP),%YMM0

0x20c7b8 AND	$0x1,%R9D

0x20c7bc MOV	$0x2,%R10D

0x20c7c2 CMOVNE	%R9,%R10

0x20c7c6 SUB	%R10,%RDI

0x20c7c9 IMUL	$0x38,%R8,%R10

0x20c7cd ADD	%R12,%R8

0x20c7d0 LEA	(,%R8,8),%R11

0x20c7d8 LEA	(%RDI,%R12,1),%R9

0x20c7dc LEA	(%R11,%R11,2),%R11

0x20c7e0 ADD	%R14,%R10

0x20c7e3 NOPW	%CS:(%RAX,%RAX,1)

(62) 0x20c7f0 VMOVD	(%R10),%XMM1

(62) 0x20c7f5 VPINSRD	$0x1,0x38(%R10),%XMM1,%XMM1

(62) 0x20c7fc VMOVQ	%XMM1,(%RAX,%R8,4)

(62) 0x20c802 VMOVD	0x4(%R10),%XMM1

(62) 0x20c808 VPINSRD	$0x1,0x3c(%R10),%XMM1,%XMM1

(62) 0x20c80f VMOVQ	%XMM1,(%RCX,%R8,4)

(62) 0x20c815 ADD	$0x2,%R8

(62) 0x20c819 VMOVUPD	0x8(%R10),%ZMM1

(62) 0x20c823 VMOVUPD	0x48(%R10),%YMM3

(62) 0x20c829 VMOVUPD	0x8(%R10),%YMM2

(62) 0x20c82f VMOVUPD	0x58(%R10),%XMM5

(62) 0x20c835 VMOVUPD	0x48(%R10),%XMM4

(62) 0x20c83b VBLENDPD	$0x8,0x28(%R10),%YMM2,%YMM2

(62) 0x20c842 VSHUFPD	$0x1,0x68(%R10),%XMM5,%XMM5

(62) 0x20c849 ADD	$0x70,%R10

(62) 0x20c84d VPERMT2PD	%ZMM3,%ZMM0,%ZMM1

(62) 0x20c853 VMOVUPD	%XMM4,0x20(%RDX,%R11,1)

(62) 0x20c85a VMOVUPD	%YMM2,(%RDX,%R11,1)

(62) 0x20c860 VMOVUPD	%XMM5,0x20(%RSI,%R11,1)

(62) 0x20c867 VMOVUPD	%YMM1,(%RSI,%R11,1)

(62) 0x20c86d ADD	$0x30,%R11

(62) 0x20c871 CMP	%R8,%R9

(62) 0x20c874 JNE	20c7f0

0x20c87a MOV	%R9,%R12

0x20c87d MOV	%RDI,%R8

0x20c880 LEA	(%R12,%R12,2),%RDI

0x20c884 IMUL	$0x38,%R8,%R8

0x20c888 LEA	0x10(,%RDI,8),%RDI

0x20c890 ADD	%R8,%R14

0x20c893 NOPW	%CS:(%RAX,%RAX,1)

(60) 0x20c8a0 MOV	(%R14),%R8D

(60) 0x20c8a3 MOV	%R8D,(%RAX,%R12,4)

(60) 0x20c8a7 MOV	0x4(%R14),%R8D

(60) 0x20c8ab MOV	%R8D,(%RCX,%R12,4)

(60) 0x20c8af INC	%R12

(60) 0x20c8b2 VMOVSD	0x8(%R14),%XMM0

(60) 0x20c8b8 VMOVSD	%XMM0,-0x10(%RDX,%RDI,1)

(60) 0x20c8be VMOVSD	0x10(%R14),%XMM0

(60) 0x20c8c4 VMOVSD	%XMM0,-0x8(%RDX,%RDI,1)

(60) 0x20c8ca VMOVSD	0x18(%R14),%XMM0

(60) 0x20c8d0 VMOVSD	%XMM0,(%RDX,%RDI,1)

(60) 0x20c8d5 VMOVSD	0x20(%R14),%XMM0

(60) 0x20c8db VMOVSD	%XMM0,-0x10(%RSI,%RDI,1)

(60) 0x20c8e1 VMOVSD	0x28(%R14),%XMM0

(60) 0x20c8e7 VMOVSD	%XMM0,-0x8(%RSI,%RDI,1)

(60) 0x20c8ed VMOVSD	0x30(%R14),%XMM0

(60) 0x20c8f3 ADD	$0x38,%R14

(60) 0x20c8f7 VMOVSD	%XMM0,(%RSI,%RDI,1)

(60) 0x20c8fc ADD	$0x18,%RDI

(60) 0x20c900 CMP	%R15,%R12

(60) 0x20c903 JL	20c8a0

0x20c905 LEA	-0x28(%RBP),%RSP

0x20c909 POP	%RBX

0x20c90a POP	%R12

0x20c90c POP	%R13

0x20c90e POP	%R14

0x20c910 POP	%R15

0x20c912 POP	%RBP

0x20c913 VZEROUPPER

0x20c916 RET

0x20c917 NOPW	(%RAX,%RAX,1)

Coverage (%)	Name	Source Location	Module
►99.98+	redistributeAtoms.omp_outlined	timestep.c:153	exec
○	__kmp_invoke_microtask		libomp.so
○	__kmp_invoke_task_func		libomp.so

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 0.04% of application time for run orig_0

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 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►
- /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

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►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
CMOVLE: 1 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
LEA: 2 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:621
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633

Conversion instructions

Detected expensive conversion instructions.

Details

MOVZX: 5 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/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: 20c320

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 20c43e <sortAtomsInCell+0x11e>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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	0.17	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	0.13	N/A
LEA 0x524(%RIP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	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 212e30 <@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 20c45a <sortAtomsInCell+0x13a>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
JMP 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.50	N/A
LEA 0x4db(%RIP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	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 212e30 <@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 20c905 <sortAtomsInCell+0x5e5>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.50	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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 20c588 <sortAtomsInCell+0x268>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
JMP 20c7b0 <sortAtomsInCell+0x490>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.50	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 20c7b0 <sortAtomsInCell+0x490>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.50	N/A
VMOVAPD -0x75d8(%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	0.17	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	0.17	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	0.17	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	0.13	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	0.17	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	0.17	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	0.13	N/A
LEA -0x28(%RBP),%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
POP %RBX	1	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	1	0.25	N/A
POP %R13	1	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	1	0.25	N/A
POP %R15	1	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	1	0.25	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	0.13	N/A

The code analyzed by CQA in that panel excludes loops and represents 0.04% of application time for run orig_0

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

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►
- /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

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►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
CMOVLE: 1 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
LEA: 2 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:621
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633

Conversion instructions

Detected expensive conversion instructions.

Details

MOVZX: 5 occurrences►
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/CoMD/build/CoMD/CoMD/src-openmp/haloExchange.c:633
- /beegfs/hackathon/users/eoseret/qaas_runs_test/gmz12.benchmarkcenter.megware.com/177-374-1600/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: 20c320

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 20c43e <sortAtomsInCell+0x11e>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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	0.17	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	0.13	N/A
LEA 0x524(%RIP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	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 212e30 <@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 20c45a <sortAtomsInCell+0x13a>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
JMP 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.50	N/A
LEA 0x4db(%RIP),%RCX	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	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 212e30 <@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 20c905 <sortAtomsInCell+0x5e5>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.50	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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	0.17	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 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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 20c588 <sortAtomsInCell+0x268>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	N/A
XOR %R8D,%R8D	1	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
JMP 20c7b0 <sortAtomsInCell+0x490>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.50	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 20c7b0 <sortAtomsInCell+0x490>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.33-1	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 20c880 <sortAtomsInCell+0x560>	1	0	0	0	0.33	0.33	0.33	0	0	0	0	0	0.50	N/A
VMOVAPD -0x75d8(%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	0.17	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	0.17	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	0.17	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	0.13	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	0.17	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	0.17	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	0.13	N/A
LEA -0x28(%RBP),%RSP	1	0.17	0.17	0.17	0.17	0.17	0.17	0	0	0	0	1	0.17	N/A
POP %RBX	1	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	1	0.25	N/A
POP %R13	1	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	1	0.25	N/A
POP %R15	1	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	1	0.25	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	0.13	N/A

Name	Coverage (%)	Time (s)
▼sortAtomsInCell–	3.19	0.24
○Loop 59 - haloExchange.c:621-630 - exec	3.14	0.23
○Loop 61 - haloExchange.c:633-642 - exec	0.00	0.01
○Loop 62 - haloExchange.c:633-642 - exec	0.00	0.01
○Loop 60 - haloExchange.c:633-642 - exec	0.00	0.00

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