OV - exec - Function hypre_ParCSRRelaxThreads.omp

	Function: hypre_ParCSRRelaxThreads.omp_outlined.27	Module: exec	Source: ams.c:3662-3682	Coverage (incl. loops): 32.42% \| (excl. loops): 0.00%

/home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/ams.c: 3662 - 3682

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

3662: #pragma omp parallel for private(i,ii,jj,res) HYPRE_SMP_SCHEDULE

3663: #endif

3664:       for (i = 0; i < n; i++)

3665:       {

3666:          /*-----------------------------------------------------------

3667:           * If diagonal is nonzero, relax point i; otherwise, skip it.

3668:           *-----------------------------------------------------------*/

3669:          if (A_diag_data[A_diag_i[i]] != zero)

3670:          {

3671:             res = f_data[i];

3672:             for (jj = A_diag_i[i]; jj < A_diag_i[i+1]; jj++)

3673:             {

3674:                ii = A_diag_j[jj];

3675:                res -= A_diag_data[jj] * Vtemp_data[ii];

3676:             }

3677:             for (jj = A_offd_i[i]; jj < A_offd_i[i+1]; jj++)

3678:             {

3679:                ii = A_offd_j[jj];

3680:                res -= A_offd_data[jj] * Vext_data[ii];

3681:             }

3682:             u_data[i] += (relax_weight*res)/l1_norms[i];

0x47b3e0 SUB	SP, SP, #112

0x47b3e4 STP	X29, X30, [SP, #32]

0x47b3e8 STR	X25, [SP, #48]

0x47b3ec STP	X24, X23, [SP, #64]

0x47b3f0 STP	X22, X21, [SP, #80]

0x47b3f4 STP	X20, X19, [SP, #96]

0x47b3f8 ADD	X29, SP, #32

0x47b3fc LDR	X8, [X2]

0x47b400 SUBS	X25, X8, #1

0x47b404 B.LT	47b680

0x47b408 LDR	W20, [X0]

0x47b40c ORR	X21, XZR, X7

0x47b410 ORR	X22, XZR, X6

0x47b414 ORR	X19, XZR, X5

0x47b418 ORR	X23, XZR, X4

0x47b41c ORR	X24, XZR, X3

0x47b420 MOVZ	W8, #1

0x47b424 ADRP	X0,

0x47b428 ADD	X0, X0, #1248

0x47b42c ADD	X3, SP, #12

0x47b430 ADD	X4, X29, #24

0x47b434 SUB	X5, X29, #8

0x47b438 ADD	X6, SP, #16

0x47b43c MOVZ	W2, #34

0x47b440 STR	XZR, [X29, #24]

0x47b444 MOVZ	W7, #1

0x47b448 STUR	X25, [X29, #504]

0x47b44c STR	X8, [SP, #16]

0x47b450 STR	WZR, [SP, #12]

0x47b454 ORR	W1, WZR, W20

0x47b458 STR	X8, [SP]

0x47b45c BL	4103d0

0x47b460 LDUR	X8, [X29, #504]

0x47b464 LDR	X9, [X29, #24]

0x47b468 CMP	X8, X25

0x47b46c CSEL	X1, X8, X25, #11

0x47b470 CMP	X9, X1

0x47b474 STUR	X1, [X29, #504]

0x47b478 B.GT	47b670

0x47b47c LDP	X16, X17, [X29, #96]

0x47b480 LDP	X8, X0, [X29, #128]

0x47b484 ADD	X1, X1, #1

0x47b488 LDR	X10, [X24]

0x47b48c ADD	X3, X10, #16

0x47b490 LDR	X13, [X21]

0x47b494 LDP	X18, X2, [X29, #112]

0x47b498 LDP	X14, X15, [X29, #80]

0x47b49c LDR	X16, [X16]

0x47b4a0 LDR	X17, [X17]

0x47b4a4 LDR	X11, [X23]

0x47b4a8 LDR	X12, [X22]

0x47b4ac LDR	X14, [X14]

0x47b4b0 ADD	X4, X13, #16

0x47b4b4 ADD	X5, X17, #16

0x47b4b8 LDR	X15, [X15]

0x47b4bc LDR	X18, [X18]

0x47b4c0 LDR	X0, [X0]

0x47b4c4 LDR	X2, [X2]

0x47b4c8 ADD	X6, X16, #16

0x47b4cc B	47b508

0x47b4d0 HINT	#0

0x47b4d4 HINT	#0

0x47b4d8 HINT	#0

0x47b4dc HINT	#0

  (2588) 0x47b4e0 LDR	D1, [X8]

  (2588) 0x47b4e4 FMUL	D0, D0, D1

  (2588) 0x47b4e8 LDR	D1, [X0, X9,LSL #3]

  (2588) 0x47b4ec FDIV	D0, D0, D1

  (2588) 0x47b4f0 LDR	D1, [X2, X9,LSL #3]

  (2588) 0x47b4f4 FADD	D0, D0, D1

  (2588) 0x47b4f8 STR	D0, [X2, X9,LSL #3]

  (2588) 0x47b4fc CMP	X7, X1

  (2588) 0x47b500 ORR	X9, XZR, X7

  (2588) 0x47b504 B.EQ	47b670

  (2588) 0x47b508 LDR	X21, [X11, X9,LSL #3]

  (2588) 0x47b50c LDR	D1, [X19]

  (2588) 0x47b510 LDR	D0, [X10, X21,LSL #3]

  (2588) 0x47b514 FCMP	D0, D1

  (2588) 0x47b518 B.EQ	47b660

  (2588) 0x47b51c LDR	D0, [X12, X9,LSL #3]

  (2588) 0x47b520 ADD	X7, X9, #1

  (2588) 0x47b524 LDR	X22, [X11, X7,LSL #3]

  (2588) 0x47b528 SUBS	X23, X21, X22

  (2588) 0x47b52c B.GE	47b5b8

  (2588) 0x47b530 SUB	W24, W22, W21

  (2588) 0x47b534 ANDS	X24, X24, #0x3

  (2588) 0x47b538 B.EQ	47b55c

  (2588) 0x47b53c HINT	#0

    (2592) 0x47b540 LDR	X25, [X13, X21,LSL #3]

    (2592) 0x47b544 LDR	D1, [X10, X21,LSL #3]

    (2592) 0x47b548 ADD	X21, X21, #1

    (2592) 0x47b54c SUBS	X24, X24, #1

    (2592) 0x47b550 LDR	D2, [X14, X25,LSL #3]

    (2592) 0x47b554 FMSUB	D0, D1, D2, D0

    (2592) 0x47b558 B.NE	47b540

  (2588) 0x47b55c CMN	X23, #4

  (2588) 0x47b560 B.HI	47b5b8

  (2588) 0x47b564 SUB	X22, X22, X21

  (2588) 0x47b568 ADD	X23, X3, X21,LSL #3

  (2588) 0x47b56c ADD	X21, X4, X21,LSL #3

  (2588) 0x47b570 HINT	#0

  (2588) 0x47b574 HINT	#0

  (2588) 0x47b578 HINT	#0

  (2588) 0x47b57c HINT	#0

    (2591) 0x47b580 LDP	X24, X25, [X21, #1008]

    (2591) 0x47b584 LDP	D2, D3, [X23, #1008]

    (2591) 0x47b588 SUBS	X22, X22, #4

    (2591) 0x47b58c LDR	D1, [X14, X24,LSL #3]

    (2591) 0x47b590 FMSUB	D0, D2, D1, D0

    (2591) 0x47b594 LDR	D1, [X14, X25,LSL #3]

    (2591) 0x47b598 FMSUB	D0, D3, D1, D0

    (2591) 0x47b59c LDP	X24, X25, [X21], #32

    (2591) 0x47b5a0 LDR	D2, [X14, X24,LSL #3]

    (2591) 0x47b5a4 LDP	D1, D3, [X23], #32

    (2591) 0x47b5a8 FMSUB	D0, D1, D2, D0

    (2591) 0x47b5ac LDR	D1, [X14, X25,LSL #3]

    (2591) 0x47b5b0 FMSUB	D0, D3, D1, D0

    (2591) 0x47b5b4 B.NE	47b580

  (2588) 0x47b5b8 LDR	X21, [X15, X9,LSL #3]

  (2588) 0x47b5bc LDR	X22, [X15, X7,LSL #3]

  (2588) 0x47b5c0 SUBS	X23, X21, X22

  (2588) 0x47b5c4 B.GE	47b4e0

  (2588) 0x47b5c8 SUB	W24, W22, W21

  (2588) 0x47b5cc ANDS	X24, X24, #0x3

  (2588) 0x47b5d0 B.EQ	47b5fc

  (2588) 0x47b5d4 HINT	#0

  (2588) 0x47b5d8 HINT	#0

  (2588) 0x47b5dc HINT	#0

    (2590) 0x47b5e0 LDR	X25, [X16, X21,LSL #3]

    (2590) 0x47b5e4 LDR	D1, [X17, X21,LSL #3]

    (2590) 0x47b5e8 ADD	X21, X21, #1

    (2590) 0x47b5ec SUBS	X24, X24, #1

    (2590) 0x47b5f0 LDR	D2, [X18, X25,LSL #3]

    (2590) 0x47b5f4 FMSUB	D0, D1, D2, D0

    (2590) 0x47b5f8 B.NE	47b5e0

  (2588) 0x47b5fc CMN	X23, #4

  (2588) 0x47b600 B.HI	47b4e0

  (2588) 0x47b604 SUB	X22, X22, X21

  (2588) 0x47b608 ADD	X23, X5, X21,LSL #3

  (2588) 0x47b60c ADD	X21, X6, X21,LSL #3

  (2588) 0x47b610 HINT	#0

  (2588) 0x47b614 HINT	#0

  (2588) 0x47b618 HINT	#0

  (2588) 0x47b61c HINT	#0

    (2589) 0x47b620 LDP	X24, X25, [X21, #1008]

    (2589) 0x47b624 LDP	D2, D3, [X23, #1008]

    (2589) 0x47b628 SUBS	X22, X22, #4

    (2589) 0x47b62c LDR	D1, [X18, X24,LSL #3]

    (2589) 0x47b630 FMSUB	D0, D2, D1, D0

    (2589) 0x47b634 LDR	D1, [X18, X25,LSL #3]

    (2589) 0x47b638 FMSUB	D0, D3, D1, D0

    (2589) 0x47b63c LDP	X24, X25, [X21], #32

    (2589) 0x47b640 LDR	D2, [X18, X24,LSL #3]

    (2589) 0x47b644 LDP	D1, D3, [X23], #32

    (2589) 0x47b648 FMSUB	D0, D1, D2, D0

    (2589) 0x47b64c LDR	D1, [X18, X25,LSL #3]

    (2589) 0x47b650 FMSUB	D0, D3, D1, D0

    (2589) 0x47b654 B.NE	47b620

  (2588) 0x47b658 B	47b4e0

0x47b65c HINT	#0

(2587) 0x47b660 ADD	X7, X9, #1

(2587) 0x47b664 CMP	X7, X1

(2587) 0x47b668 ORR	X9, XZR, X7

(2587) 0x47b66c B.NE	47b508

0x47b670 ADRP	X0,

0x47b674 ADD	X0, X0, #1272

0x47b678 ORR	W1, WZR, W20

0x47b67c BL	410150

0x47b680 LDP	X20, X19, [SP, #96]

0x47b684 LDP	X22, X21, [SP, #80]

0x47b688 LDR	X25, [SP, #48]

0x47b68c LDP	X24, X23, [SP, #64]

0x47b690 LDP	X29, X30, [SP, #32]

0x47b694 ADD	SP, SP, #112

0x47b698 RET

0x47b69c HINT	#0

Coverage (%)	Name	Source Location	Module
►98.49+	__kmp_invoke_microtask		libomp.so
○	__kmp_invoke_task_func		libomp.so
○	__kmp_launch_thread		libomp.so
○	__kmp_launch_worker(void*)		libomp.so
○	start_thread		libc.so.6
○	thread_start		libc.so.6
►1.44+	__kmp_invoke_microtask		libomp.so
○	__kmp_invoke_task_func		libomp.so
○	__kmp_fork_call		libomp.so
○	__kmpc_fork_call		libomp.so
○	hypre_ParCSRRelaxThreads	ams.c:3685	exec
○	hypre_BoomerAMGCycle	par_cycle.c:334	exec
○	hypre_BoomerAMGSolve	par_amg_solve.c:274	exec
○	hypre_PCGSolve	pcg.c:545	exec
○	main	amg.c:419	exec
○	__libc_start_call_main		libc.so.6
○	__libc_start_main		libc.so.6
○	_start		exec

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 0.00% of application time for run orig_0

Source file and lines	ams.c:3662-3682
Module	exec

The function is defined in /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/ams.c:3662-3682.

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Vectorization

Your function is not vectorized. Only 26% of vector register length is used (average across all VPU instructions). By vectorizing your function, you can lower the cost of an iteration from 11.83 to 2.69 cycles (4.40x speedup).

Details

All VPU 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)

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/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/ams.c:3662
- /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/ams.c:3662

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 3 bytes. The binary function is storing 3 bytes.

General properties

nb instructions	77
nb uops	71
loop length	308
used w registers	6
used x registers	29
used b registers	0
used h registers	0
used s registers	1
used d registers	0
used q registers	0
used v registers	0
used z registers	0
nb stack references	13

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	8.88 cycles
front end	8.88 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	3.00	3.00	8.00	8.00	8.00	8.00	0.00	0.00	0.00	0.00	11.83	11.50	11.67	5.50	5.50
cycles	3.00	3.00	8.00	8.00	8.00	8.00	0.00	0.00	0.00	0.00	11.83	11.50	11.67	5.50	5.50

Execution ports to units layout:

P0: BRU
P1: BRU
P2: ALU
P3: ALU
P4: ALU
P5: ALU
P6 (128 bits): VPU, FP store data, ALU, DIV/SQRT
P7 (128 bits): VPU, ALU, FP store data
P8 (128 bits): VPU, ALU, DIV/SQRT
P9 (128 bits): ALU, VPU
P10 (256 bits): store address, load
P11 (256 bits): store address, load
P12 (256 bits): load
P13 (64 bits): store data
P14 (64 bits): store data

Cycles executing div or sqrt instructions

Cycles summary

Front-end	8.88
Dispatch	11.83
Overall L1	11.83

Vectorization ratios

all	0%
load	0%
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	0%

Vector efficiency ratios

all	26%
load	26%
store	32%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	25%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	18%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 11.83 cycles. At this rate:

0% of peak load performance is reached (0.25 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
0% of peak store performance is reached (0.25 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 47b3e0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
SUB SP, SP, #112	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
STP X29, X30, [SP, #32]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
STR X25, [SP, #48]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
STP X24, X23, [SP, #64]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
STP X22, X21, [SP, #80]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
STP X20, X19, [SP, #96]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
ADD X29, SP, #32	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
LDR X8, [X2]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
SUBS X25, X8, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.LT 47b680 <hypre_ParCSRRelaxThreads.omp_outlined.27+0x2a0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR W20, [X0]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (12.5%)
ORR X21, XZR, X7	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ORR X22, XZR, X6	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ORR X19, XZR, X5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ORR X23, XZR, X4	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ORR X24, XZR, X3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
MOVZ W8, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
ADRP X0, <4ee424>	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ADD X0, X0, #1248	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ADD X3, SP, #12	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ADD X4, X29, #24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
SUB X5, X29, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ADD X6, SP, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
MOVZ W2, #34	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
STR XZR, [X29, #24]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
MOVZ W7, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
STUR X25, [X29, #504]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
STR X8, [SP, #16]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
STR WZR, [SP, #12]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (12.5%)
ORR W1, WZR, W20	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
STR X8, [SP]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
BL 4103d0 <@plt_start@+0x3b0>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDUR X8, [X29, #504]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
LDR X9, [X29, #24]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
CMP X8, X25	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
CSEL X1, X8, X25, #11	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
CMP X9, X1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
STUR X1, [X29, #504]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
B.GT 47b670 <hypre_ParCSRRelaxThreads.omp_outlined.27+0x290>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDP X16, X17, [X29, #96]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X8, X0, [X29, #128]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
ADD X1, X1, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
LDR X10, [X24]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
ADD X3, X10, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
LDR X13, [X21]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
LDP X18, X2, [X29, #112]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X14, X15, [X29, #80]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDR X16, [X16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR X17, [X17]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR X11, [X23]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
LDR X12, [X22]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
LDR X14, [X14]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X4, X13, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ADD X5, X17, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
LDR X15, [X15]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR X18, [X18]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR X0, [X0]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR X2, [X2]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X6, X16, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
B 47b508 <hypre_ParCSRRelaxThreads.omp_outlined.27+0x128>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
HINT #0																			N/A
HINT #0																			N/A
HINT #0																			N/A
HINT #0																			N/A
HINT #0																			N/A
ADRP X0, <4ee670>	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ADD X0, X0, #1272	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ORR W1, WZR, W20	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
BL 410150 <@plt_start@+0x130>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDP X20, X19, [SP, #96]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
LDP X22, X21, [SP, #80]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDR X25, [SP, #48]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
LDP X24, X23, [SP, #64]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X29, X30, [SP, #32]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
ADD SP, SP, #112	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
RET	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
HINT #0																			N/A

The code analyzed by CQA in that panel excludes loops and represents 0.00% of application time for run orig_0

Source file and lines	ams.c:3662-3682
Module	exec

The function is defined in /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/ams.c:3662-3682.

Warnings:
Ignoring paths for analysis

gain
potential
hint
expert

Vectorization

Details

All VPU 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)

No data for this section

CALL instructions

Detected function call instructions.

Details

unknown: 2 occurrences►
- /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/ams.c:3662
- /home/eoseret/qaas/qaas_runs/178-188-3659/intel/AMG/build/AMG/AMG/parcsr_ls/ams.c:3662

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 3 bytes. The binary function is storing 3 bytes.

General properties

nb instructions	77
nb uops	71
loop length	308
used w registers	6
used x registers	29
used b registers	0
used h registers	0
used s registers	1
used d registers	0
used q registers	0
used v registers	0
used z registers	0
nb stack references	13

Front-end

MACRO FUSION NOT POSSIBLE FIT IN UOP CACHE

micro-operation queue	8.88 cycles
front end	8.88 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14
uops	3.00	3.00	8.00	8.00	8.00	8.00	0.00	0.00	0.00	0.00	11.83	11.50	11.67	5.50	5.50
cycles	3.00	3.00	8.00	8.00	8.00	8.00	0.00	0.00	0.00	0.00	11.83	11.50	11.67	5.50	5.50

Execution ports to units layout:

P0: BRU
P1: BRU
P2: ALU
P3: ALU
P4: ALU
P5: ALU
P6 (128 bits): VPU, FP store data, ALU, DIV/SQRT
P7 (128 bits): VPU, ALU, FP store data
P8 (128 bits): VPU, ALU, DIV/SQRT
P9 (128 bits): ALU, VPU
P10 (256 bits): store address, load
P11 (256 bits): store address, load
P12 (256 bits): load
P13 (64 bits): store data
P14 (64 bits): store data

Cycles executing div or sqrt instructions

Cycles summary

Front-end	8.88
Dispatch	11.83
Overall L1	11.83

Vectorization ratios

all	0%
load	0%
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	0%

Vector efficiency ratios

all	26%
load	26%
store	32%
mul	NA (no mul vectorizable/vectorized instructions)
add-sub	25%
fma	NA (no fma vectorizable/vectorized instructions)
div/sqrt	NA (no div/sqrt vectorizable/vectorized instructions)
other	18%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each call to the function takes 11.83 cycles. At this rate:

0% of peak load performance is reached (0.25 out of 96.00 bytes loaded per cycle (GB/s @ 1GHz))
0% of peak store performance is reached (0.25 out of 48.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

Found no such bottlenecks.

ASM code

In the binary file, the address of the function is: 47b3e0

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	P12	P13	P14	Latency	Recip. throughput	Vectorization
SUB SP, SP, #112	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
STP X29, X30, [SP, #32]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
STR X25, [SP, #48]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
STP X24, X23, [SP, #64]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
STP X22, X21, [SP, #80]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
STP X20, X19, [SP, #96]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (50.0%)
ADD X29, SP, #32	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
LDR X8, [X2]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
SUBS X25, X8, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
B.LT 47b680 <hypre_ParCSRRelaxThreads.omp_outlined.27+0x2a0>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDR W20, [X0]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (12.5%)
ORR X21, XZR, X7	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ORR X22, XZR, X6	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ORR X19, XZR, X5	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ORR X23, XZR, X4	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ORR X24, XZR, X3	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
MOVZ W8, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
ADRP X0, <4ee424>	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ADD X0, X0, #1248	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ADD X3, SP, #12	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ADD X4, X29, #24	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
SUB X5, X29, #8	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ADD X6, SP, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
MOVZ W2, #34	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
STR XZR, [X29, #24]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
MOVZ W7, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
STUR X25, [X29, #504]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
STR X8, [SP, #16]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
STR WZR, [SP, #12]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (12.5%)
ORR W1, WZR, W20	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
STR X8, [SP]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
BL 4103d0 <@plt_start@+0x3b0>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDUR X8, [X29, #504]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
LDR X9, [X29, #24]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
CMP X8, X25	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
CSEL X1, X8, X25, #11	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
CMP X9, X1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.33	scal (25.0%)
STUR X1, [X29, #504]	1	0	0	0	0	0	0	0	0	0	0	0.50	0.50	0	0.50	0.50	1	0.50	scal (25.0%)
B.GT 47b670 <hypre_ParCSRRelaxThreads.omp_outlined.27+0x290>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDP X16, X17, [X29, #96]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X8, X0, [X29, #128]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
ADD X1, X1, #1	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
LDR X10, [X24]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
ADD X3, X10, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
LDR X13, [X21]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
LDP X18, X2, [X29, #112]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X14, X15, [X29, #80]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDR X16, [X16]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR X17, [X17]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR X11, [X23]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
LDR X12, [X22]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
LDR X14, [X14]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X4, X13, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
ADD X5, X17, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
LDR X15, [X15]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR X18, [X18]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR X0, [X0]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
LDR X2, [X2]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	N/A
ADD X6, X16, #16	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (25.0%)
B 47b508 <hypre_ParCSRRelaxThreads.omp_outlined.27+0x128>	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
HINT #0																			N/A
HINT #0																			N/A
HINT #0																			N/A
HINT #0																			N/A
HINT #0																			N/A
ADRP X0, <4ee670>	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ADD X0, X0, #1272	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
ORR W1, WZR, W20	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	scal (12.5%)
BL 410150 <@plt_start@+0x130>	1	0.50	0.50	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
LDP X20, X19, [SP, #96]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	scal (50.0%)
LDP X22, X21, [SP, #80]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDR X25, [SP, #48]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	0.33	scal (25.0%)
LDP X24, X23, [SP, #64]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
LDP X29, X30, [SP, #32]	1	0	0	0	0	0	0	0	0	0	0	0.33	0.33	0.33	0	0	4	1	N/A
ADD SP, SP, #112	1	0	0	0.25	0.25	0.25	0.25	0	0	0	0	0	0	0	0	0	1	0.25	N/A
RET	1	0.50	0.50	0	0	0	0	0	0	0	0	0	0	0	0	0	1	0.50	N/A
HINT #0																			N/A

Name	Coverage (%)	Time (s)
▼hypre_ParCSRRelaxThreads.omp_outlined.27–	32.42	15.46
▼Loop 2587 - ams.c:3664-3682 - exec–	0.00	0.00
▼Loop 2588 - ams.c:3664-3682 - exec–	2.01	0.69
○Loop 2591 - ams.c:3672-3675 - exec	27.71	9.53
○Loop 2592 - ams.c:3672-3675 - exec	2.70	0.93
○Loop 2589 - ams.c:3677-3680 - exec	0.00	0.00
○Loop 2590 - ams.c:3677-3680 - exec	0.00	0.00

Report Configuration

Vectorization

Details

Workaround

CALL instructions

Details

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

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code

Vectorization

Details

Workaround

CALL instructions

Details

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