OV - exec - Loop 2364

	Loop Id: 2364	Module: exec	Source: par_csr_matop.c:946-965 [...]	Coverage: 0.86%

0x50fe40 MOV	(%R11,%RAX,8),%RCX    [3]

0x50fe44 VMULSD	(%R13,%RAX,8),%XMM7,%XMM8    [5]

0x50fe4b LEA	(%R8,%RCX,8),%R9

0x50fe4f MOV	(%R9),%R15    [1]

0x50fe52 CMP	%R12,%R15

0x50fe55 JL	510060

0x50fe5b LEA	(%RBX,%R15,8),%RCX

0x50fe5f VADDSD	(%RCX),%XMM8,%XMM9    [8]

0x50fe63 INC	%RAX

0x50fe66 VMOVSD	%XMM9,(%RCX)    [8]

0x50fe6a CMP	%R10,%RAX

0x50fe6d JL	50fe40

0x510060 MOV	-0x38(%RBP),%R10    [4]

0x510064 MOV	%RDX,(%R9)    [1]

0x510067 VMOVSD	%XMM8,(%RBX,%RDX,8)    [7]

0x51006c MOV	%RCX,(%R10,%RDX,8)    [2]

0x510070 INC	%RAX

0x510073 INC	%RDX

0x510076 MOV	(%R14),%R10    [6]

0x510079 CMP	%RAX,%R10

0x51007c JG	50fe40

/home/eoseret/qaas_runs_GNR/173-927-0874/intel/AMG/build/AMG/AMG/parcsr_mv/par_csr_matop.c: 946 - 965

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

946:          for (jj3 = B_diag_i[i2]; jj3 < B_diag_i[i2+1]; jj3++)

947:          {

948:             i3 = B_diag_j[jj3];

[...]

956:             if (B_marker[i3] < jj_row_begin_diag)

957:             {

958:                B_marker[i3] = jj_count_diag;

959:                C_diag_data[jj_count_diag] = a_entry*B_diag_data[jj3];

960:                C_diag_j[jj_count_diag] = i3;

961:                jj_count_diag++;

962:             }

963:             else

964:             {

965:                C_diag_data[B_marker[i3]] += a_entry*B_diag_data[jj3];

Coverage (%)	Name	Source Location	Module
○97.53	gomp_thread_start	team.c:130	libgomp.so.1.0.0
○2.47	GOMP_parallel	libgomp.h:982	libgomp.so.1.0.0

Path /

Metric	Value
CQA speedup if no scalar integer	1.50
CQA speedup if FP arith vectorized	2.00
CQA speedup if fully vectorized	8.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.42
Bottlenecks
Function	hypre_ParMatmul._omp_fn.3
Source	par_csr_matop.c:946-948,par_csr_matop.c:956-965
Source loop unroll info	not unrolled or unrolled with no peel/tail loop
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	2.25
CQA cycles if no scalar integer	1.50
CQA cycles if FP arith vectorized	1.13
CQA cycles if fully vectorized	0.28
Front-end cycles	2.25
P0 cycles	1.50
P1 cycles	0.75
P2 cycles	1.50
P3 cycles	1.50
P4 cycles	1.00
P5 cycles	0.65
P6 cycles	1.00
P7 cycles	1.00
P8 cycles	1.00
P9 cycles	1.00
P10 cycles	0.60
P11 cycles	1.50
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	13.50
Nb uops	12.50
Nb loads	4.50
Nb stores	2.00
Nb stack references	0.50
FLOP/cycle	0.67
Nb FLOP add-sub	0.50
Nb FLOP mul	1.00
Nb FLOP fma	0.00
Nb FLOP div	0.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	22.80
Bytes prefetched	0.00
Bytes loaded	36.00
Bytes stored	16.00
Stride 0	1.00
Stride 1	3.00
Stride n	0.00
Stride unknown	1.50
Stride indirect	0.00
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	0.00
Vectorization ratio mul	0.00
Vectorization ratio add_sub	0.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	12.50
Vector-efficiency ratio load	12.50
Vector-efficiency ratio store	12.50
Vector-efficiency ratio mul	12.50
Vector-efficiency ratio add_sub	12.50
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	12.50

Metric	Value
CQA speedup if no scalar integer	1.33
CQA speedup if FP arith vectorized	2.67
CQA speedup if fully vectorized	8.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.33
Bottlenecks	micro-operation queue,
Function	hypre_ParMatmul._omp_fn.3
Source	par_csr_matop.c:946-948,par_csr_matop.c:956-965
Source loop unroll info	not unrolled or unrolled with no peel/tail loop
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	2.00
CQA cycles if no scalar integer	1.50
CQA cycles if FP arith vectorized	0.75
CQA cycles if fully vectorized	0.25
Front-end cycles	2.00
P0 cycles	1.50
P1 cycles	1.00
P2 cycles	1.33
P3 cycles	1.33
P4 cycles	0.50
P5 cycles	0.70
P6 cycles	1.00
P7 cycles	0.50
P8 cycles	0.50
P9 cycles	0.50
P10 cycles	0.80
P11 cycles	1.33
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	12.00
Nb uops	11.00
Nb loads	4.00
Nb stores	1.00
Nb stack references	0.00
FLOP/cycle	1.00
Nb FLOP add-sub	1.00
Nb FLOP mul	1.00
Nb FLOP fma	0.00
Nb FLOP div	0.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	20.00
Bytes prefetched	0.00
Bytes loaded	32.00
Bytes stored	8.00
Stride 0	0.00
Stride 1	2.00
Stride n	0.00
Stride unknown	2.00
Stride indirect	0.00
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	0.00
Vectorization ratio mul	0.00
Vectorization ratio add_sub	0.00
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	12.50
Vector-efficiency ratio load	12.50
Vector-efficiency ratio store	12.50
Vector-efficiency ratio mul	12.50
Vector-efficiency ratio add_sub	12.50
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	12.50

Metric	Value
CQA speedup if no scalar integer	1.67
CQA speedup if FP arith vectorized	1.67
CQA speedup if fully vectorized	8.00
CQA speedup if no inter-iteration dependency	NA
CQA speedup if next bottleneck killed	1.50
Bottlenecks	micro-operation queue,
Function	hypre_ParMatmul._omp_fn.3
Source	par_csr_matop.c:946-948,par_csr_matop.c:956-965
Source loop unroll info	not unrolled or unrolled with no peel/tail loop
Source loop unroll confidence level	max
Unroll/vectorization loop type	NA
Unroll factor	NA
CQA cycles	2.50
CQA cycles if no scalar integer	1.50
CQA cycles if FP arith vectorized	1.50
CQA cycles if fully vectorized	0.31
Front-end cycles	2.50
P0 cycles	1.50
P1 cycles	0.50
P2 cycles	1.67
P3 cycles	1.67
P4 cycles	1.50
P5 cycles	0.60
P6 cycles	1.00
P7 cycles	1.50
P8 cycles	1.50
P9 cycles	1.50
P10 cycles	0.40
P11 cycles	1.67
DIV/SQRT cycles	0.00
Inter-iter dependencies cycles	1
FE+BE cycles (UFS)	NA
Stall cycles (UFS)	NA
Nb insns	15.00
Nb uops	14.00
Nb loads	5.00
Nb stores	3.00
Nb stack references	1.00
FLOP/cycle	0.40
Nb FLOP add-sub	0.00
Nb FLOP mul	1.00
Nb FLOP fma	0.00
Nb FLOP div	0.00
Nb FLOP rcp	0.00
Nb FLOP sqrt	0.00
Nb FLOP rsqrt	0.00
Bytes/cycle	25.60
Bytes prefetched	0.00
Bytes loaded	40.00
Bytes stored	24.00
Stride 0	2.00
Stride 1	4.00
Stride n	0.00
Stride unknown	1.00
Stride indirect	0.00
Vectorization ratio all	0.00
Vectorization ratio load	0.00
Vectorization ratio store	0.00
Vectorization ratio mul	0.00
Vectorization ratio add_sub	NA
Vectorization ratio fma	NA
Vectorization ratio div_sqrt	NA
Vectorization ratio other	0.00
Vector-efficiency ratio all	12.50
Vector-efficiency ratio load	12.50
Vector-efficiency ratio store	12.50
Vector-efficiency ratio mul	12.50
Vector-efficiency ratio add_sub	NA
Vector-efficiency ratio fma	NA
Vector-efficiency ratio div_sqrt	NA
Vector-efficiency ratio other	12.50

Path /

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

Function	hypre_ParMatmul._omp_fn.3
Source file and lines	par_csr_matop.c:946-965
Module	exec

The loop is defined in /home/eoseret/qaas_runs_GNR/173-927-0874/intel/AMG/build/AMG/AMG/parcsr_mv/par_csr_matop.c:946-948,956-965.

The related source loop is not unrolled or unrolled with no peel/tail loop.
The structure of this loop is probably <if then [else] end>.

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

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

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

gain
potential
hint
expert

Vectorization

Your loop is not vectorized. 8 data elements could be processed at once in vector registers.

By vectorizing your loop, you can lower the cost of an iteration from 2.25 to 0.28 cycles (8.00x speedup).

Details

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

Workaround

Try another compiler or update/tune your current one
Remove inter-iterations dependences from your loop and make it unit-stride:
- If your arrays have 2 or more dimensions, check whether elements are accessed contiguously and, otherwise, try to permute loops accordingly: C storage order is row-major: for(i) for(j) a[j][i] = b[j][i]; (slow, non stride 1) => for(i) for(j) a[i][j] = b[i][j]; (fast, stride 1)
- If your loop 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)

FMA

Presence of both ADD/SUB and MUL operations.

Workaround

Recompile with march=sapphirerapids. CQA target is Granite_Rapids_ad (Intel(R) Xeon(R) Granite Rapids) but specialization flags are -march=graniterapids
Try to change order in which elements are evaluated (using parentheses) in arithmetic expressions containing both ADD/SUB and MUL operations to enable your compiler to generate FMA instructions wherever possible. For instance a + b*c is a valid FMA (MUL then ADD). However (a+b)* c cannot be translated into an FMA (ADD then MUL).

Slow data structures access

Detected data structures (typically arrays) that cannot be efficiently read/written

Details

Constant unknown stride: 1 occurrence(s)

Non-unit stride (uncontiguous) accesses are not efficiently using data caches

Workaround

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

Type of elements and instruction set

1 SSE or AVX instructions are processing arithmetic or math operations on double precision FP elements in scalar mode (one at a time).

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

The binary loop is composed of 1.5 FP arithmetical operations:

0.50: addition or subtraction
1: multiply

The binary loop is loading 36 bytes (4 double precision FP elements). The binary loop is storing 16 bytes (2 double precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.03 FP operations per loaded or stored byte.

Unroll opportunity

Loop is data access bound.

Workaround

Unroll your loop if trip count is significantly higher than target unroll factor and if some data references are common to consecutive iterations. This can be done manually. Or by recompiling with -funroll-loops and/or -floop-unroll-and-jam. Or with the unroll (resp. unroll_and_jam) directive on top of the inner (resp. surrounding) loop. You can enforce an unroll factor: #pragma GCC unroll N

General properties

nb instructions	13.50
nb uops	12.50
loop length	54
used x86 registers	11.50
used mmx registers	0
used xmm registers	2.50
used ymm registers	0
used zmm registers	0
nb stack references	0.50
ADD-SUB / MUL ratio	1.00

Front-end

MACRO FUSION NOT POSSIBLE

micro-operation queue	2.25 cycles
front end	2.25 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	1.50	0.75	1.50	1.50	1.00	0.65	1.00	1.00	1.00	1.00	0.60	1.50
cycles	1.50	0.75	1.50	1.50	1.00	0.65	1.00	1.00	1.00	1.00	0.60	1.50

Execution ports to units layout:

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

Cycles executing div or sqrt instructions	NA
Longest recurrence chain latency (RecMII)	1.00

Cycles summary

Front-end	2.25
Dispatch	1.58
Data deps.	1.00
Overall L1	2.25

Vectorization ratios

INT

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

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

INT+FP

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

Vector efficiency ratios

INT

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

all	12%
load	12%
store	12%
mul	12%
add-sub	12%
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	12%
load	12%
store	12%
mul	12%
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 iteration of the binary loop takes 2.25 cycles. At this rate:

8% of peak load performance is reached (16.00 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
5% of peak store performance is reached (6.80 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Function	hypre_ParMatmul._omp_fn.3
Source file and lines	par_csr_matop.c:946-965
Module	exec

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

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

gain
potential
hint
expert

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 2.00 to 1.50 cycles (1.33x speedup).

Workaround

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

Vectorization

Your loop is not vectorized. 8 data elements could be processed at once in vector registers.

By vectorizing your loop, you can lower the cost of an iteration from 2.00 to 0.25 cycles (8.00x speedup).

Details

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

Workaround

Try another compiler or update/tune your current one
Remove inter-iterations dependences from your loop and make it unit-stride:
- If your arrays have 2 or more dimensions, check whether elements are accessed contiguously and, otherwise, try to permute loops accordingly: C storage order is row-major: for(i) for(j) a[j][i] = b[j][i]; (slow, non stride 1) => for(i) for(j) a[i][j] = b[i][j]; (fast, stride 1)
- If your loop 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.

FMA

Presence of both ADD/SUB and MUL operations.

Workaround

Recompile with march=sapphirerapids. CQA target is Granite_Rapids_ad (Intel(R) Xeon(R) Granite Rapids) but specialization flags are -march=graniterapids
Try to change order in which elements are evaluated (using parentheses) in arithmetic expressions containing both ADD/SUB and MUL operations to enable your compiler to generate FMA instructions wherever possible. For instance a + b*c is a valid FMA (MUL then ADD). However (a+b)* c cannot be translated into an FMA (ADD then MUL).

Slow data structures access

Detected data structures (typically arrays) that cannot be efficiently read/written

Details

Constant unknown stride: 2 occurrence(s)

Non-unit stride (uncontiguous) accesses are not efficiently using data caches

Workaround

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

Type of elements and instruction set

2 SSE or AVX instructions are processing arithmetic or math operations on double precision FP elements in scalar mode (one at a time).

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

The binary loop is composed of 2 FP arithmetical operations:

1: addition or subtraction
1: multiply

The binary loop is loading 32 bytes (4 double precision FP elements). The binary loop is storing 8 bytes (1 double precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.05 FP operations per loaded or stored byte.

Unroll opportunity

Loop is potentially data access bound.

Workaround

General properties

nb instructions	12
nb uops	11
loop length	47
used x86 registers	10
used mmx registers	0
used xmm registers	3
used ymm registers	0
used zmm registers	0
nb stack references	0
ADD-SUB / MUL ratio	1.00

Front-end

ASSUMED MACRO FUSION FIT IN UOP CACHE

micro-operation queue	2.00 cycles
front end	2.00 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	1.50	1.00	1.33	1.33	0.50	0.70	1.00	0.50	0.50	0.50	0.80	1.33
cycles	1.50	1.00	1.33	1.33	0.50	0.70	1.00	0.50	0.50	0.50	0.80	1.33

Execution ports to units layout:

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

Cycles executing div or sqrt instructions	NA
Longest recurrence chain latency (RecMII)	1.00

Cycles summary

Front-end	2.00
Dispatch	1.50
Data deps.	1.00
Overall L1	2.00

Vectorization ratios

INT

all	0%
load	NA (no load vectorizable/vectorized instructions)
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)
other	0%

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

INT+FP

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

Vector efficiency ratios

INT

all	12%
load	NA (no load vectorizable/vectorized instructions)
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)
other	12%

all	12%
load	12%
store	12%
mul	12%
add-sub	12%
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	12%
load	12%
store	12%
mul	12%
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 iteration of the binary loop takes 2.00 cycles. At this rate:

8% of peak load performance is reached (16.00 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
3% of peak store performance is reached (4.00 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 2.00 to 1.50 cycles (1.33x speedup).

ASM code

In the binary file, the address of the loop is: 50fe40

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput	Vectorization
MOV (%R11,%RAX,8),%RCX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
VMULSD (%R13,%RAX,8),%XMM7,%XMM8	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
LEA (%R8,%RCX,8),%R9	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV (%R9),%R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
CMP %R12,%R15	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JL 510060 <hypre_ParMatmul._omp_fn.3+0x4d0>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
LEA (%RBX,%R15,8),%RCX	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
VADDSD (%RCX),%XMM8,%XMM9	1	0	0.50	0.33	0.33	0	0.50	0	0	0	0	0	0.33	3	0.50	scal (12.5%)
INC %RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
VMOVSD %XMM9,(%RCX)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
CMP %R10,%RAX	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
JL 50fe40 <hypre_ParMatmul._omp_fn.3+0x2b0>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A

Function	hypre_ParMatmul._omp_fn.3
Source file and lines	par_csr_matop.c:946-965
Module	exec

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

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

gain
potential
hint
expert

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 2.50 to 1.50 cycles (1.67x speedup).

Workaround

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

Vectorization

Your loop is not vectorized. 8 data elements could be processed at once in vector registers.

By vectorizing your loop, you can lower the cost of an iteration from 2.50 to 0.31 cycles (8.00x speedup).

Details

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

Workaround

Try another compiler or update/tune your current one
Remove inter-iterations dependences from your loop and make it unit-stride:
- If your arrays have 2 or more dimensions, check whether elements are accessed contiguously and, otherwise, try to permute loops accordingly: C storage order is row-major: for(i) for(j) a[j][i] = b[j][i]; (slow, non stride 1) => for(i) for(j) a[i][j] = b[i][j]; (fast, stride 1)
- If your loop 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

Slow data structures access

Detected data structures (typically arrays) that cannot be efficiently read/written

Details

Constant unknown stride: 1 occurrence(s)

Non-unit stride (uncontiguous) accesses are not efficiently using data caches

Workaround

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

Type of elements and instruction set

1 SSE or AVX instructions are processing arithmetic or math operations on double precision FP elements in scalar mode (one at a time).

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

The binary loop is composed of 1 FP arithmetical operations:

1: multiply

The binary loop is loading 40 bytes (5 double precision FP elements). The binary loop is storing 24 bytes (3 double precision FP elements).

Arithmetic intensity

Arithmetic intensity is 0.02 FP operations per loaded or stored byte.

Unroll opportunity

Loop is data access bound.

Workaround

General properties

nb instructions	15
nb uops	14
loop length	61
used x86 registers	13
used mmx registers	0
used xmm registers	2
used ymm registers	0
used zmm registers	0
nb stack references	1

Front-end

ASSUMED MACRO FUSION FIT IN UOP CACHE

micro-operation queue	2.50 cycles
front end	2.50 cycles

Back-end

	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11
uops	1.50	0.50	1.67	1.67	1.50	0.60	1.00	1.50	1.50	1.50	0.40	1.67
cycles	1.50	0.50	1.67	1.67	1.50	0.60	1.00	1.50	1.50	1.50	0.40	1.67

Execution ports to units layout:

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

Cycles executing div or sqrt instructions	NA
Longest recurrence chain latency (RecMII)	1.00

Cycles summary

Front-end	2.50
Dispatch	1.67
Data deps.	1.00
Overall L1	2.50

Vectorization ratios

INT

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

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

INT+FP

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

Vector efficiency ratios

INT

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

all	12%
load	12%
store	12%
mul	12%
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	12%
load	12%
store	12%
mul	12%
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	12%

Cycles and memory resources usage

Assuming all data fit into the L1 cache, each iteration of the binary loop takes 2.50 cycles. At this rate:

8% of peak load performance is reached (16.00 out of 192.00 bytes loaded per cycle (GB/s @ 1GHz))
7% of peak store performance is reached (9.60 out of 128.00 bytes stored per cycle (GB/s @ 1GHz))

Front-end bottlenecks

ASM code

In the binary file, the address of the loop is: 50fe40

Instruction	Nb FU	P0	P1	P2	P3	P4	P5	P6	P7	P8	P9	P10	P11	Latency	Recip. throughput	Vectorization
MOV (%R11,%RAX,8),%RCX	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
VMULSD (%R13,%RAX,8),%XMM7,%XMM8	1	0.50	0.50	0.33	0.33	0	0	0	0	0	0	0	0.33	4	0.50	scal (12.5%)
LEA (%R8,%RCX,8),%R9	1	0	0	0	0	0	0	0	0	0	0	0	0	0	0.17	N/A
MOV (%R9),%R15	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	scal (12.5%)
CMP %R12,%R15	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	scal (12.5%)
JL 510060 <hypre_ParMatmul._omp_fn.3+0x4d0>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A
MOV -0x38(%RBP),%R10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
MOV %RDX,(%R9)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
VMOVSD %XMM8,(%RBX,%RDX,8)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
MOV %RCX,(%R10,%RDX,8)	1	0	0	0	0	0.50	0	0	0.50	0.50	0.50	0	0	1	0.50	scal (12.5%)
INC %RAX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
INC %RDX	1	0	0	0	0	0	0	0	0	0	0	0	0	1	0.17	N/A
MOV (%R14),%R10	1	0	0	0.33	0.33	0	0	0	0	0	0	0	0.33	1	0.33	N/A
CMP %RAX,%R10	1	0.20	0.20	0	0	0	0.20	0.20	0	0	0	0.20	0	1	0.20	N/A
JG 50fe40 <hypre_ParMatmul._omp_fn.3+0x2b0>	1	0.50	0	0	0	0	0	0.50	0	0	0	0	0	0	0.50	N/A

Report Configuration

Vectorization

Details

Workaround

FMA

Workaround

Slow data structures access

Details

Workaround

Type of elements and instruction set

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

Arithmetic intensity

Unroll opportunity

Workaround

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Code clean check

Workaround

Vectorization

Details

Workaround

Execution units bottlenecks

FMA

Workaround

Slow data structures access

Details

Workaround

Type of elements and instruction set

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

Arithmetic intensity

Unroll opportunity

Workaround

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

Slow data structures access

Details

Workaround

Type of elements and instruction set

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

Arithmetic intensity

Unroll opportunity

Workaround

General properties

Front-end

Back-end

Cycles summary

Vectorization ratios

Vector efficiency ratios

Cycles and memory resources usage

Front-end bottlenecks

ASM code