	*************************************************
	*                                               *
	*          ONE-View report generation           *
	*                                               *
	*************************************************

[MAQAO] Info: Experiment configuration summary is available adding -dbg=1 in command line

* [MAQAO] Warning: Experiment directory /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/run/oneview_runs/defaults/icx/oneview_results_1786546675 already exists and is reused.
           It can be replaced using --replace in the command line.
[MAQAO] Info: 
[MAQAO] Info: START THE APPLICATION PROFILING
[MAQAO] Info: -> RUNNING THE PROFILER...
[MAQAO] Info:   LPROF has already been run
[MAQAO] Info: STOP THE APPLICATION PROFILING
[MAQAO] Info: 
[MAQAO] Info: START FUNCTIONS AND LOOPS ANALYSIS ...
[MAQAO] Info: STOP FUNCTIONS AND LOOPS ANALYSIS ...
[MAQAO] Info: 
[MAQAO] Info: START THE REPORT GENERATION
[MAQAO] Info: -> ONE-VIEW EXPERIMENT DIRECTORY: /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/run/oneview_runs/defaults/icx/oneview_results_1786546675


+====================================================================================================================+
+                                                    1  -  GLOBAL                                                    +
+====================================================================================================================+


+--------------------------------------------------------------------------------------------------------------------+
+                                             1.1  -  Experiment Summary                                             +
+--------------------------------------------------------------------------------------------------------------------+

  Application:			/beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/run/base_runs/defaults/icx/exec
  Timestamp:			2026-08-12 16:57:55
  Universal Timestamp:		1786546675
  Experiment Type:		MPI; OpenMP; Throughput; 
  Machine:			gmz17.benchmarkcenter.megware.com
  Architecture:			x86_64
  Micro Architecture:		ZEN_V5
  Model Name:			AMD EPYC 9655 96-Core Processor
  Cache Size:			1024 KB
  Number of Cores:		96
  OS Version:			Linux 5.14.0-687.29.1.el9_8.x86_64 #1 SMP PREEMPT_DYNAMIC Thu Jul 23 16:18:48 EDT 2026
  Compilation Options:		
		exec:  --driver-mode=g++ --intel -D ADD_ -D H5_USE_16_API -D HAVE_CONFIG_H -D HAVE_MKL -D MPICH_SKIP_MPICXX -D OMPI_SKIP_MPICXX -D OPENMP_NO_COMPLEX -D _MPICC_H -D restrict=__restrict__ -I /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src -I /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/icx/src -I /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Particle -I /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Utilities -I /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Platforms -I /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Platforms/Host -O3 -march=native -g -fno-omit-frame-pointer -fcf-protection=none -no-pie -grecord-command-line -fiopenmp -fstrict-aliasing -O3 -D NDEBUG -std=c++17 -MD -MT src/QMCWaveFunctions/CMakeFiles/qmcwfs.dir/SPOSet_builder.cpp.o -MF src/QMCWaveFunctions/CMakeFiles/qmcwfs.dir/SPOSet_builder.cpp.o.d -o src/QMCWaveFunctions/CMakeFiles/qmcwfs.dir/SPOSet_builder.cpp.o -c /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/SPOSet_builder.cpp -I /cluster/hpcx/2.22/ompi5-ifx-mt/include -I /cluster/hpcx/2.22/ompi5-ifx-mt/include/openmpi -fveclib=SVML 
  Number of processes observed:	8
  Number of threads observed:	192
  MAQAO version:		2026.1.0
  MAQAO build:			6d1be1d51c1e63266254997eb301734a7264775d::20260810-150026




+--------------------------------------------------------------------------------------------------------------------+
+                                               1.2  -  Global Metrics                                               +
+--------------------------------------------------------------------------------------------------------------------+

  Total Time:				130.74 s
  Max (Thread Active Time):		125.32 s
  Average Active Time:			118.45 s
  Activity Ratio:			98.1 %
  Average number of active threads:	173.958
  Affinity Stability:			99.5 %
  Time spent in analyzed loops:		49.1 %
  Time spent in analyzed innermost loops: 48.9 %
  Time spent in user code:		49.3 %
  Compilation Options Score:		100
  Array Access Efficiency:		95.0 %

   Potential Speedups
  ----------------------------------------------------
  Perfect Flow Complexity:		1.00
  Perfect OpenMP/MPI/Pthread/TBB:	1.03
  Perfect OpenMP/MPI/Pthread/TBB + Load Distribution:	1.07
  If No Scalar Integer:
      Potential Speedup:		1.02
      Nb Loops to get 80%:		2
  If FP Vectorized:
      Potential Speedup:		1.04
      Nb Loops to get 80%:		1
  If Fully Vectorized:
      Potential Speedup:		1.06
      Nb Loops to get 80%:		2
  If Only FP Arithmetic:
      Potential Speedup:		1.03
      Nb Loops to get 80%:		3




+--------------------------------------------------------------------------------------------------------------------+
+                                             1.3  -  Potential Speedups                                             +
+--------------------------------------------------------------------------------------------------------------------+

  If No Scalar Integer:
      Number of loops   | 1      | 8      | 16     | 23     | 32     | 
      Cumulated Speedup | 1.0125 | 1.0203 | 1.0204 | 1.0204 | 1.0204 | 
  Top 5 loops:
    exec - 1976:	1.0125
    exec - 463:	1.0171
    exec - 1039:	1.0191
    exec - 423:	1.0197
    exec - 100:	1.02

  If FP Vectorized:
      Number of loops   | 1      | 8      | 16     | 23     | 32     | 
      Cumulated Speedup | 1.0364 | 1.0422 | 1.0422 | 1.0422 | 1.0422 | 
  Top 5 loops:
    exec - 2215:	1.0364
    exec - 1158:	1.0397
    exec - 1143:	1.0416
    exec - 1141:	1.0421
    exec - 1145:	1.0422

  If Fully Vectorized:
      Number of loops   | 1      | 8      | 16     | 23     | 32     | 
      Cumulated Speedup | 1.0418 | 1.0630 | 1.0634 | 1.0634 | 1.0634 | 
  Top 5 loops:
    exec - 2215:	1.0418
    exec - 1976:	1.0561
    exec - 1039:	1.0603
    exec - 1158:	1.0614
    exec - 1143:	1.062

  If Only FP Arithmetic:
      Number of loops   | 1      | 8      | 16     | 23     | 32     | 
      Cumulated Speedup | 1.0130 | 1.0309 | 1.0318 | 1.0318 | 1.0318 | 
  Top 5 loops:
    exec - 1976:	1.013
    exec - 463:	1.0236
    exec - 1039:	1.0271
    exec - 423:	1.0289
    exec - 1122:	1.0297



+====================================================================================================================+
+                                                   2  -  SUMMARY                                                    +
+====================================================================================================================+


+--------------------------------------------------------------------------------------------------------------------+
+                                             2.1  -  EXPERIMENT QUALITY                                             +
+--------------------------------------------------------------------------------------------------------------------+

  [4 / 4] Application profile is long enough (125.32 s)
To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.

  [3 / 3] Most of time spent in analyzed modules comes from functions with source/debug info
-g option gives access to debugging informations, such are source locations.

  [2.9748425255818 / 3] Most of time spent in analyzed modules (99.16%) comes from functions compiled with architecture specialization option
-march=native


  [3 / 3] Most of time spent in analyzed modules comes from functions with compilation options informations and
-fno-omit-frame-pointer is present
-fno-omit-frame-pointer improves the accuracy of callchains found during the application profiling.

  [3 / 3] Optimization level option is correctly used


  [3 / 3] Host configuration allows retrieval of all necessary metrics.


  [2 / 2] Application is correctly profiled ("Others" category represents 0.06 % of the execution time)
To have a representative profiling, it is advised that the category "Others" represents less than 20% of the execution
time in order to analyze as much as possible of the user code

  [1 / 1] Lstopo present. The Topology lstopo report will be generated.



+--------------------------------------------------------------------------------------------------------------------+
+                                                2.2  -  CODE QUALITY                                                +
+--------------------------------------------------------------------------------------------------------------------+

  [4 / 4] Enough time of the experiment time spent in analyzed loops (49.09%)
If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on
application performances.

  [4 / 4] Threads activity is good
On average, more than 90.60% of observed threads are actually active 

  [4 / 4] CPU activity is good
CPU cores are active 98.13% of time

  [4 / 4] Loop profile is not flat
At least one loop coverage is greater than 4% (27.06%), representing an hotspot for the application

  [4 / 4] Enough time of the experiment time spent in analyzed innermost loops (48.94%)
If the time spent in analyzed innermost loops is less than 15%, standard innermost loop optimizations such as
vectorisation will have a limited impact on application performances.

  [4 / 4] Affinity is good (99.52%)
Threads are not migrating to CPU cores: probably successfully pinned

  [3 / 3] Less than 10% (0.00%) is spend in BLAS1 operations
It could be more efficient to inline by hand BLAS1 operations

  [3 / 3] Functions mostly use all threads
Functions running on a reduced number of threads (typically sequential code) cover less than 10% of application
walltime (4.88%)

  [3 / 3] Cumulative Outermost/In between loops coverage (0.15%) lower than cumulative innermost loop coverage (48.94%)
Having cumulative Outermost/In between loops coverage greater than cumulative innermost loop coverage will make loop
optimization more complex

  [2 / 2] Less than 10% (0.00%) is spend in BLAS2 operations
BLAS2 calls usually could make a poor cache usage and could benefit from inlining.

  [2 / 2] Less than 10% (0.00%) is spend in Libm/SVML (special functions)



+--------------------------------------------------------------------------------------------------------------------+
+                                               2.3  -  LOOPS OVERVIEW                                               +
+--------------------------------------------------------------------------------------------------------------------+

  Top 5 loops:
   + exec - 1037:
     analysis: Execution Time: 27 % - Vectorization Ratio: 100.00 % - Vector Length Use: 100.00 %

   + exec - 1048:
     analysis: Execution Time: 11 % - Vectorization Ratio: 100.00 % - Vector Length Use: 100.00 %
     Data Access Issues: 0
        [0] [SA] Inefficient vectorization: more than 10% of the vector loads instructions are unaligned - When
            allocating arrays, don’t forget to align them. There are 0 issues ( = arrays) costing 2 points each

   + exec - 2215:
     analysis: Execution Time: 4 % - Vectorization Ratio: 12.77 % - Vector Length Use: 14.10 %
     Loop Computation Issues: 4
        [4] [SA] Presence of expensive FP instructions - Perform hoisting, change algorithm, use SVML or proper
            numerical library or perform value profiling (count the number of distinct input values). There are 1
            issues (= instructions) costing 4 points each.

   + exec - 1976:
     analysis: Execution Time: 1 % - Vectorization Ratio: 0.00 % - Vector Length Use: 12.50 %
     Loop Computation Issues: 2
        [2] [SA] Presence of a large number of scalar integer instructions - Simplify loop structure, perform loop
            splitting or perform unroll and jam. This issue costs 2 points.
     Data Access Issues: 22
        [4] [SA] Presence of constant non unit stride data access - Use array restructuring, perform loop interchange
            or use gather instructions to lower a bit the cost. There are 2 issues ( = data accesses) costing 2 point
            each.
        [16] [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost.
            There are 4 issues ( = indirect data accesses) costing 4 point each.
        [2] [SA] More than 20% of the loads are accessing the stack - Perform loop splitting to decrease pressure on
            registers. This issue costs 2 points.
     Vectorization Roadblocks: 20
        [4] [SA] Presence of constant non unit stride data access - Use array restructuring, perform loop interchange
            or use gather instructions to lower a bit the cost. There are 2 issues ( = data accesses) costing 2 point
            each.
        [16] [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost.
            There are 4 issues ( = indirect data accesses) costing 4 point each.

   + exec - 463 :
     analysis: Execution Time: 1 % - Vectorization Ratio: 94.12 % - Vector Length Use: 94.49 %
     Loop Computation Issues: 2
        [2] [SA] Presence of a large number of scalar integer instructions - Simplify loop structure, perform loop
            splitting or perform unroll and jam. This issue costs 2 points.
     Data Access Issues: 17
        [12] [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost.
            There are 3 issues ( = indirect data accesses) costing 4 point each.
        [0] [SA] Inefficient vectorization: more than 10% of the vector loads instructions are unaligned - When
            allocating arrays, don’t forget to align them. There are 0 issues ( = arrays) costing 2 points each
        [5] [SA] Presence of special instructions executing on a single port (COMPRESS/EXPAND, BROADCAST) - Simplify
            data access and try to get stride 1 access. There are 5 issues (= instructions) costing 1 point each.
     Vectorization Roadblocks: 12
        [12] [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost.
            There are 3 issues ( = indirect data accesses) costing 4 point each.
     Inefficient Vectorization: 7
        [5] [SA] Presence of special instructions executing on a single port (COMPRESS/EXPAND, BROADCAST) - Simplify
            data access and try to get stride 1 access. There are 5 issues (= instructions) costing 1 point each.
        [2] [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2
            points.



+====================================================================================================================+
+                                                 3  -  APPLICATION                                                  +
+====================================================================================================================+


+--------------------------------------------------------------------------------------------------------------------+
+                                               3.1  -  Categorization                                               +
+--------------------------------------------------------------------------------------------------------------------+

   Category | IO     | Exe    | System  | Others  | Memory | String | MPI   | TBB   | OMP   | Pthread | Math  |
  ----------+--------+--------+---------+---------+--------+--------+-------+-------+-------+---------+-------+
   Time (%) | 0.00   | 49.34  | 0.55    | 0.06    | 0.39   | 0.43   | 0.11  | 0.00  | 0.92  | 0.00    | 48.19 |




+--------------------------------------------------------------------------------------------------------------------+
+                                          3.2  -  Function Based Profiling                                          +
+--------------------------------------------------------------------------------------------------------------------+

   Buckets                    | Nb Functions              | Coverage                  | Cumulated Coverage        |
  ----------------------------+---------------------------+---------------------------+---------------------------+
   > 8%                       | 5                         | 84.67                     | 84.67                     |
   4% to 8%                   | 1                         | 4.62                      | 89.30                     |
   2% to 4%                   | 0                         | 0.00                      | 89.30                     |
   1% to 2%                   | 2                         | 2.81                      | 92.11                     |
   0.5% to 1%                 | 6                         | 4.11                      | 96.22                     |
   0.25% to 0.5%              | 4                         | 1.53                      | 97.75                     |
   0.125% to 0.25%            | 5                         | 0.91                      | 98.66                     |
   < 0.125%                   | 82                        | 1.18                      | 99.83                     |




+--------------------------------------------------------------------------------------------------------------------+
+                                            3.3  -  Loop Based Profiling                                            +
+--------------------------------------------------------------------------------------------------------------------+

   Buckets                    | Nb Loops                  | Coverage                  | Cumulated Coverage        |
  ----------------------------+---------------------------+---------------------------+---------------------------+
   > 8%                       | 2                         | 38.16                     | 38.16                     |
   4% to 8%                   | 1                         | 4.58                      | 42.74                     |
   2% to 4%                   | 0                         | 0.00                      | 42.74                     |
   1% to 2%                   | 2                         | 2.77                      | 45.51                     |
   0.5% to 1%                 | 2                         | 1.34                      | 46.85                     |
   0.25% to 0.5%              | 2                         | 0.78                      | 47.63                     |
   0.125% to 0.25%            | 4                         | 0.65                      | 48.29                     |
   < 0.125%                   | 40                        | 0.65                      | 48.94                     |


+====================================================================================================================+
+                                                  4  -  FUNCTIONS                                                   +
+====================================================================================================================+


+--------------------------------------------------------------------------------------------------------------------+
+                                              4.1  -  Top 10 Functions                                              +
+--------------------------------------------------------------------------------------------------------------------+

   Function                                               | Module              | Coverage (%)   | Time (s)       |
  --------------------------------------------------------+---------------------+----------------+----------------+
   miniqmcreference::einspline_spo_ref<double>::evalua... | exec                | 27.07          | 32.07          |
   mkl_blas_def_dgemm_kernel_zen                          | libmkl_def.so.3     | 21.03          | 24.91          |
   mkl_blas_def_dgemm_copybn_bdz                          | libmkl_def.so.3     | 13.03          | 15.43          |
   mkl_blas_def_dgemm_pst                                 | libmkl_def.so.3     | 11.99          | 14.20          |
   miniqmcreference::einspline_spo_ref<double>::evalua... | exec                | 11.56          | 13.69          |
   qmcplusplus::SoaDistanceTableABOMPTarget<double, 3u... | exec                | 4.62           | 5.48           |
   qmcplusplus::SoaDistanceTableAAOMPTarget<double, 3u... | exec                | 1.49           | 1.76           |
   miniqmcreference::TwoBodyJastrowRef<qmcplusplus::Bs... | exec                | 1.33           | 1.57           |
   kmp_flag_64<false, true>::wait(kmp_info*, int, void*)  | libiomp5.so         | 0.87           | 1.03           |
   mkl_blas_def_dgemm_copyan_bdz                          | libmkl_def.so.3     | 0.76           | 0.90           |


+====================================================================================================================+
+                                                    5  -  LOOPS                                                     +
+====================================================================================================================+


+--------------------------------------------------------------------------------------------------------------------+
+                                                5.1  -  Top 10 Loops                                                +
+--------------------------------------------------------------------------------------------------------------------+

   Loop Id        | Module              | Source Location                                        | Coverage (%)   |
  ----------------+---------------------+--------------------------------------------------------+----------------+
   1037           | exec                | MultiBsplineRef.hpp:68-70                              | 27.06          |
   1048           | exec                | MultiBsplineRef.hpp:242-262                            | 11.10          |
   2215           | exec                | SoaDistanceTableABOMPTarget.h:228-228,ParticleBCond... | 4.58           |
   1976           | exec                | SoaDistanceTableAAOMPTarget.h:440-442,VectorSoACont... | 1.49           |
   463            | exec                | BsplineFunctor.h:236-241                               | 1.28           |
   1158           | exec                | OperatorTags.h:63-63,inner_product.hpp:82-83,DiracD... | 0.67           |
   1514           | exec                | ParticleBConds3DSoa.h:235-256                          | 0.67           |
   1039           | exec                | TinyVector.h:145-145,VectorSoAContainer.h:231-231,V... | 0.41           |
   1143           | exec                | TinyVector.h:61-61,OperatorTags.h:63-63,inner_produ... | 0.38           |
   423            | exec                | BsplineFunctor.h:291-297                               | 0.20           |





+====================================================================================================================+
+                                                     6  -  CQA                                                      +
+====================================================================================================================+


+--------------------------------------------------------------------------------------------------------------------+
+                                                   6.1  -  Loops                                                    +
+--------------------------------------------------------------------------------------------------------------------+





      6.1.1  -  Loop 1037 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Numerics/Spline2/MultiBsplineRef.hpp:68-70.

The related source loop is not unrolled or unrolled with no peel/tail loop.

      6.1.1.1  -  Path 1
  ----------------------------------------------------------------------------------------------------------

60% of peak computational performance is used (28.80 out of 48.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.1.1.1  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is fully vectorized, using full register length.


Details
All SSE/AVX instructions are used in vector version (process two or more data elements in vector registers).



      6.1.1.1.2  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

Performance is limited by execution of FP multiply or FMA (fused multiply-add) operations (the FP multiply/FMA unit is a bottleneck).

By removing all these bottlenecks, you can lower the cost of an iteration from 2.50 to 1.50 cycles (1.67x speedup).


Workaround
Reduce the number of FP multiply/FMA instructions




      6.1.1.1.3  -  FMA
  ----------------------------------------------------------------------------------------------------------

Detected 32 FMA (fused multiply-add) operations.




      6.1.1.1.4  -  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.
 - VMOVUPD: 1 occurrences<<list_path_1_complex_1>>



      6.1.1.1.5  -  Vector unaligned load/store instructions
  ----------------------------------------------------------------------------------------------------------

Detected 1 optimal vector unaligned load/store instructions.


Details
 - VMOVUPD: 1 occurrences<<list_path_1_vec_align_1>>


Workaround
Use vector aligned instructions:
 1) align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
 2) inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.
<<image_vec_align>>


      6.1.1.1.6  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

5 AVX-512 instructions are processing arithmetic or math operations on double precision FP elements in vector mode (eight at a time).



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

The binary loop is composed of 72 FP arithmetical operations:
 - 32: addition or subtraction (all inside FMA instructions)
 - 40: multiply (32 inside FMA instructions)
The binary loop is loading 320 bytes (40 double precision FP elements).
The binary loop is storing 64 bytes (8 double precision FP elements).


      6.1.1.1.8  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

Arithmetic intensity is 0.19 FP operations per loaded or stored byte.


      6.1.1.1.9  -  Unroll opportunity
  ----------------------------------------------------------------------------------------------------------

Loop is potentially 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 with the unroll (resp. unroll_and_jam) directive on top of the inner (resp. surrounding) loop. You can enforce an unroll factor: #pragma unroll_and_jam N, unroll_and_jam(N), unroll N or unroll(N)







      6.1.2  -  Loop 1048 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Numerics/Spline2/MultiBsplineRef.hpp:242-262.

It is main loop of related source loop which is unrolled by 16 (including vectorization).

      6.1.2.1  -  Path 1
  ----------------------------------------------------------------------------------------------------------

60% of peak computational performance is used (29.22 out of 48.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.2.1.1  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is fully vectorized, using full register length.


Details
All SSE/AVX instructions are used in vector version (process two or more data elements in vector registers).



      6.1.2.1.2  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

Performance is limited by execution of FP multiply or FMA (fused multiply-add) operations (the FP multiply/FMA unit is a bottleneck).

By removing all these bottlenecks, you can lower the cost of an iteration from 23.00 to 20.00 cycles (1.15x speedup).


Workaround
Reduce the number of FP multiply/FMA instructions




      6.1.2.1.3  -  FMA
  ----------------------------------------------------------------------------------------------------------

Detected 304 FMA (fused multiply-add) operations.




      6.1.2.1.4  -  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.
 - VMOVUPD: 20 occurrences<<list_path_1_complex_1>>



      6.1.2.1.5  -  Vector unaligned load/store instructions
  ----------------------------------------------------------------------------------------------------------

Detected 42 optimal vector unaligned load/store instructions.


Details
 - VMOVUPD: 42 occurrences<<list_path_1_vec_align_1>>


Workaround
Use vector aligned instructions:
 1) align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
 2) inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.
<<image_vec_align>>


      6.1.2.1.6  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

46 AVX-512 instructions are processing arithmetic or math operations on double precision FP elements in vector mode (eight at a time).



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

The binary loop is composed of 672 FP arithmetical operations:
 - 304: addition or subtraction (all inside FMA instructions)
 - 368: multiply (304 inside FMA instructions)
The binary loop is loading 1800 bytes (225 double precision FP elements).
The binary loop is storing 1280 bytes (160 double precision FP elements).


      6.1.2.1.8  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

Arithmetic intensity is 0.22 FP operations per loaded or stored byte.







      6.1.3  -  Loop 2215 from exec
  ==========================================================================================================

The loop is defined in:
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Particle/SoaDistanceTableABOMPTarget.h:228
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Particle/Lattice/ParticleBConds3DSoa.h:280-298


The related source loop is not unrolled or unrolled with no peel/tail loop.

      6.1.3.1  -  Path 1
  ----------------------------------------------------------------------------------------------------------

8% of peak computational performance is used (4.09 out of 48.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.3.1.1  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is probably not vectorized.
Only 14% of vector register length is used (average across all SSE/AVX instructions).
By vectorizing your loop, you can lower the cost of an iteration from 11.00 to 1.37 cycles (8.00x 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 loop can use their full power.


Workaround
 - Try another compiler or update/tune your current one:
  * recompile with ffast-math (included in Ofast) to extend loop vectorization to FP reductions.
 - 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)<<image_row_maj>>
  * 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)



      6.1.3.1.2  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

Performance is limited by execution of FP multiply or FMA (fused multiply-add) operations (the FP multiply/FMA unit is a bottleneck).

By removing all these bottlenecks, you can lower the cost of an iteration from 11.00 to 7.50 cycles (1.47x speedup).


Workaround
Reduce the number of FP multiply/FMA instructions




      6.1.3.1.3  -  FMA
  ----------------------------------------------------------------------------------------------------------

Detected 14 FMA (fused multiply-add) operations.
Presence of both ADD/SUB and MUL operations.

Workaround
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).




      6.1.3.1.4  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

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



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

The binary loop is composed of 45 FP arithmetical operations:
 - 23: addition or subtraction (14 inside FMA instructions)
 - 21: multiply (14 inside FMA instructions)
 - 1: square root
The binary loop is loading 24 bytes (3 double precision FP elements).
The binary loop is storing 32 bytes (4 double precision FP elements).


      6.1.3.1.6  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

Arithmetic intensity is 0.80 FP operations per loaded or stored byte.







      6.1.4  -  Loop 1976 from exec
  ==========================================================================================================

The loop is defined in:
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Particle/SoaDistanceTableAAOMPTarget.h:440-442
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h:244,263
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:223


The related source loop is not unrolled or unrolled with no peel/tail loop.

      6.1.4.1  -  Path 1
  ----------------------------------------------------------------------------------------------------------

0% of peak computational performance is used (0.00 out of 96.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.4.1.1  -  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 7.25 to 1.25 cycles (5.80x speedup).

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



      6.1.4.1.2  -  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 7.25 to 0.91 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:
  * recompile with ffast-math (included in Ofast) to extend loop vectorization to FP reductions.
 - 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)<<image_row_maj>>
  * 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)



      6.1.4.1.3  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

Performance is limited by:
 - reading data from caches/RAM (load units are a bottleneck)
 - writing data to caches/RAM (the store unit is a bottleneck)

By removing all these bottlenecks, you can lower the cost of an iteration from 7.25 to 5.38 cycles (1.35x speedup).


Workaround
 - Read less array elements
 - Write less array elements
 - Provide more information to your compiler:
  * hardcode the bounds of the corresponding 'for' loop




No data for this section



      6.1.4.1.4  -  Slow data structures access
  ----------------------------------------------------------------------------------------------------------

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

Details
 - Constant non-unit stride: 2 occurrence(s)
 - Irregular (variable stride) or indirect: 4 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)
 - Try to remove indirect accesses. If applicable, precompute elements out of the innermost loop.



      6.1.4.1.5  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

No instructions are processing arithmetic or math operations on FP elements. This loop is probably writing/copying data or processing integer elements.


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

The binary loop does not contain any FP arithmetical operations.
The binary loop is loading 168 bytes.
The binary loop is storing 64 bytes.


      6.1.4.1.7  -  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 with the unroll (resp. unroll_and_jam) directive on top of the inner (resp. surrounding) loop. You can enforce an unroll factor: #pragma unroll_and_jam N, unroll_and_jam(N), unroll N or unroll(N)







      6.1.5  -  Loop 463 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/BsplineFunctor.h:236-241.

It is main loop of related source loop which is unrolled by 32 (including vectorization).

      6.1.5.1  -  Path 1
  ----------------------------------------------------------------------------------------------------------

Warnings:
 - The number of fused uops of the instruction [VPCMPNEQD	%ZMM1,%ZMM8,%K0] is unknown
 - The number of fused uops of the instruction [VPCMPNEQD	%ZMM1,%ZMM9,%K1] is unknown

0% of peak computational performance is used (0.00 out of 48.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.5.1.1  -  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 10.00 to 6.50 cycles (1.54x speedup).

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



      6.1.5.1.2  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is highly vectorized.
94% of vector register length is used (average across all SSE/AVX instructions).


Details
94% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):
 - 85% of SSE/AVX instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.



      6.1.5.1.3  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

Found no such bottlenecks but see expert reports for more complex bottlenecks.




      6.1.5.1.4  -  Masked instructions
  ----------------------------------------------------------------------------------------------------------

Detected masked instructions.

Details
Vector registers are partially exploited, which is expected if your loop is irregular or mixes elements of different sizes.

Workaround
If your loop is irregular, try to remove or hoist conditional structures out of your loop. If it mixes elements of different sizes, try to uniformize them.




      6.1.5.1.5  -  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.
 - VCOMPRESSPD: 4 occurrences<<list_path_1_complex_1>>
 - VPBROADCASTD: 1 occurrences<<list_path_1_complex_2>>



      6.1.5.1.6  -  Slow data structures access
  ----------------------------------------------------------------------------------------------------------

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

Details
 - Irregular (variable stride) or indirect: 3 occurrence(s)
Non-unit stride (uncontiguous) accesses are not efficiently using data caches


Workaround
Try to remove indirect accesses. If applicable, precompute elements out of the innermost loop.


      6.1.5.1.7  -  Vector unaligned load/store instructions
  ----------------------------------------------------------------------------------------------------------

Detected 4 optimal vector unaligned load/store instructions.


Details
 - VMOVUPD: 4 occurrences<<list_path_1_vec_align_1>>


Workaround
Use vector aligned instructions:
 1) align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
 2) inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.
<<image_vec_align>>


      6.1.5.1.8  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

4 AVX-512 instructions are processing arithmetic or math operations on double precision FP elements in vector mode (eight at a time).



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

The binary loop does not contain any FP arithmetical operations.
The binary loop is loading 256 bytes (32 double precision FP elements).
The binary loop is storing 256 bytes (32 double precision FP elements).







      6.1.6  -  Loop 1158 from exec
  ==========================================================================================================

The loop is defined in:
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h:63
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Platforms/CPU/SIMD/inner_product.hpp:82-83
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/DiracDeterminantRef.cpp:157


It is main loop of related source loop which is unrolled by 4 (including vectorization).

      6.1.6.1  -  Path 1
  ----------------------------------------------------------------------------------------------------------

16% of peak computational performance is used (8.00 out of 48.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.6.1.1  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is vectorized, but using 79% register length (average across all SSE/AVX instructions).
By fully vectorizing your loop, you can lower the cost of an iteration from 4.00 to 3.37 cycles (1.19x speedup).

Details
All SSE/AVX instructions are used in vector version (process two or more data elements in vector registers).
Since your execution units are vector units, only a fully vectorized loop can use their full power.


Workaround
 - Pass to your compiler a micro-architecture specialization option:
  * use march=native
 - Use vector aligned instructions:
  1) align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
  2) inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.



      6.1.6.1.2  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

Found no such bottlenecks but see expert reports for more complex bottlenecks.




      6.1.6.1.3  -  Masked instructions
  ----------------------------------------------------------------------------------------------------------

Detected masked instructions.

Details
Vector registers are partially exploited, which is expected if your loop is irregular or mixes elements of different sizes.

Workaround
If your loop is irregular, try to remove or hoist conditional structures out of your loop. If it mixes elements of different sizes, try to uniformize them.


      6.1.6.1.4  -  FMA
  ----------------------------------------------------------------------------------------------------------

Detected 16 FMA (fused multiply-add) operations.




      6.1.6.1.5  -  Vector unaligned load/store instructions
  ----------------------------------------------------------------------------------------------------------

Detected 2 optimal vector unaligned load/store instructions.


Details
 - VMOVUPD: 2 occurrences<<list_path_1_vec_align_1>>


Workaround
Use vector aligned instructions:
 1) align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
 2) inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.
<<image_vec_align>>


      6.1.6.1.6  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

4 AVX instructions are processing arithmetic or math operations on double precision FP elements in vector mode (four at a time).



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

The binary loop is composed of 32 FP arithmetical operations:
 - 16: addition or subtraction (all inside FMA instructions)
 - 16: multiply (all inside FMA instructions)
The binary loop is loading 192 bytes (24 double precision FP elements).


      6.1.6.1.8  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

Arithmetic intensity is 0.17 FP operations per loaded or stored byte.







      6.1.7  -  Loop 1514 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Particle/Lattice/ParticleBConds3DSoa.h:235-256.

It is main loop of related source loop which is unrolled by 32 (including vectorization).

      6.1.7.1  -  Path 1
  ----------------------------------------------------------------------------------------------------------

68% of peak computational performance is used (32.73 out of 48.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.7.1.1  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is highly vectorized.
92% of vector register length is used (average across all SSE/AVX instructions).


Details
91% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):
 - 40% of SSE/AVX loads are used in vector version.
 - 66% of SSE/AVX instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.



      6.1.7.1.2  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

Performance is limited by execution of FP multiply or FMA (fused multiply-add) operations (the FP multiply/FMA unit is a bottleneck).

By removing all these bottlenecks, you can lower the cost of an iteration from 44.00 to 30.00 cycles (1.47x speedup).


Workaround
Reduce the number of FP multiply/FMA instructions




      6.1.7.1.3  -  FMA
  ----------------------------------------------------------------------------------------------------------

Detected 448 FMA (fused multiply-add) operations.
Presence of both ADD/SUB and MUL operations.

Workaround
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).




      6.1.7.1.4  -  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.
 - VMOVUPD: 16 occurrences<<list_path_1_complex_1>>



      6.1.7.1.5  -  Slow data structures access
  ----------------------------------------------------------------------------------------------------------

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

Details
 - Irregular (variable stride) or indirect: 7 occurrence(s)
Non-unit stride (uncontiguous) accesses are not efficiently using data caches


Workaround
Try to remove indirect accesses. If applicable, precompute elements out of the innermost loop.


      6.1.7.1.6  -  Vector unaligned load/store instructions
  ----------------------------------------------------------------------------------------------------------

Detected 28 optimal vector unaligned load/store instructions.


Details
 - VMOVUPD: 28 occurrences<<list_path_1_vec_align_1>>


Workaround
Use vector aligned instructions:
 1) align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
 2) inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.
<<image_vec_align>>


      6.1.7.1.7  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

136 AVX-512 instructions are processing arithmetic or math operations on double precision FP elements in vector mode (eight at a time).



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

The binary loop is composed of 1440 FP arithmetical operations:
 - 736: addition or subtraction (448 inside FMA instructions)
 - 672: multiply (448 inside FMA instructions)
 - 32: square root
The binary loop is loading 912 bytes (114 double precision FP elements).
The binary loop is storing 1024 bytes (128 double precision FP elements).


      6.1.7.1.9  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

Arithmetic intensity is 0.74 FP operations per loaded or stored byte.







      6.1.8  -  Loop 1039 from exec
  ==========================================================================================================

The loop is defined in:
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVector.h:145
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h:231,271
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/einspline_spo_ref.hpp:223-227
 - /usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_vector.h:1046
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/OhmmsVector.h:223
 - /usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_algobase.h:235


The related source loop is not unrolled or unrolled with no peel/tail loop.

      6.1.8.1  -  Path 1
  ----------------------------------------------------------------------------------------------------------

0% of peak computational performance is used (0.00 out of 96.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.8.1.1  -  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 5.75 to 3.00 cycles (1.92x speedup).

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



      6.1.8.1.2  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is poorly vectorized.
Only 13% of vector register length is used (average across all SSE/AVX instructions).
By fully vectorizing your loop, you can lower the cost of an iteration from 5.75 to 0.53 cycles (10.82x speedup).

Details
20% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):
 - 20% of SSE/AVX loads are used in vector version.
 - 25% of SSE/AVX stores are used in vector version.
 - 0% of SSE/AVX instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.
Since your execution units are vector units, only a fully vectorized loop can use their full power.


Workaround
 - Try another compiler or update/tune your current one:
  * recompile with ffast-math (included in Ofast) to extend loop vectorization to FP reductions.
 - 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)<<image_row_maj>>
  * 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)



      6.1.8.1.3  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

Performance is limited by:
 - reading data from caches/RAM (load units are a bottleneck)
 - writing data to caches/RAM (the store unit is a bottleneck)

By removing all these bottlenecks, you can lower the cost of an iteration from 5.75 to 4.75 cycles (1.21x speedup).


Workaround
 - Read less array elements
 - Write less array elements
 - Provide more information to your compiler:
  * hardcode the bounds of the corresponding 'for' loop




No data for this section



      6.1.8.1.4  -  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.
 - CMOVL: 1 occurrences<<list_path_1_complex_1>>
 - LEA: 1 occurrences<<list_path_1_complex_2>>



      6.1.8.1.5  -  Slow data structures access
  ----------------------------------------------------------------------------------------------------------

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

Details
 - Constant unknown stride: 4 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)



      6.1.8.1.6  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

No instructions are processing arithmetic or math operations on FP elements. This loop is probably writing/copying data or processing integer elements.


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

The binary loop does not contain any FP arithmetical operations.
The binary loop is loading 132 bytes.
The binary loop is storing 40 bytes.


      6.1.8.1.8  -  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 with the unroll (resp. unroll_and_jam) directive on top of the inner (resp. surrounding) loop. You can enforce an unroll factor: #pragma unroll_and_jam N, unroll_and_jam(N), unroll N or unroll(N)







      6.1.9  -  Loop 1143 from exec
  ==========================================================================================================

The loop is defined in:
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVector.h:61
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h:63
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/Platforms/CPU/SIMD/inner_product.hpp:155


It is main loop of related source loop which is unrolled by 5 (including vectorization).

      6.1.9.1  -  Path 1
  ----------------------------------------------------------------------------------------------------------

12% of peak computational performance is used (6.00 out of 48.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.9.1.1  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is vectorized, but using 81% register length (average across all SSE/AVX instructions).
By fully vectorizing your loop, you can lower the cost of an iteration from 4.00 to 3.47 cycles (1.15x speedup).

Details
All SSE/AVX instructions are used in vector version (process two or more data elements in vector registers).
Since your execution units are vector units, only a fully vectorized loop can use their full power.


Workaround
 - Pass to your compiler a micro-architecture specialization option:
  * use march=native
 - Use vector aligned instructions:
  1) align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
  2) inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.



      6.1.9.1.2  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

Found no such bottlenecks but see expert reports for more complex bottlenecks.




      6.1.9.1.3  -  Masked instructions
  ----------------------------------------------------------------------------------------------------------

Detected masked instructions.

Details
Vector registers are partially exploited, which is expected if your loop is irregular or mixes elements of different sizes.

Workaround
If your loop is irregular, try to remove or hoist conditional structures out of your loop. If it mixes elements of different sizes, try to uniformize them.


      6.1.9.1.4  -  FMA
  ----------------------------------------------------------------------------------------------------------

Detected 12 FMA (fused multiply-add) operations.




      6.1.9.1.5  -  Vector unaligned load/store instructions
  ----------------------------------------------------------------------------------------------------------

Detected 2 optimal vector unaligned load/store instructions.


Details
 - VMOVUPD: 2 occurrences<<list_path_1_vec_align_1>>


Workaround
Use vector aligned instructions:
 1) align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
 2) inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.
<<image_vec_align>>


      6.1.9.1.6  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

3 AVX instructions are processing arithmetic or math operations on double precision FP elements in vector mode (four at a time).



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

The binary loop is composed of 24 FP arithmetical operations:
 - 12: addition or subtraction (all inside FMA instructions)
 - 12: multiply (all inside FMA instructions)
The binary loop is loading 160 bytes (20 double precision FP elements).


      6.1.9.1.8  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

Arithmetic intensity is 0.15 FP operations per loaded or stored byte.







      6.1.10  -  Loop 423 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/BsplineFunctor.h:291-297.

It is main loop of related source loop which is unrolled by 32 (including vectorization).

      6.1.10.1  -  Path 1
  ----------------------------------------------------------------------------------------------------------

Warnings:
 - The number of fused uops of the instruction [VPCMPNEQD	%ZMM3,%ZMM11,%K2] is unknown
 - The number of fused uops of the instruction [VPCMPNEQD	%ZMM3,%ZMM10,%K4] is unknown

0% of peak computational performance is used (0.00 out of 48.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.10.1.1  -  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 12.50 to 9.00 cycles (1.39x speedup).

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



      6.1.10.1.2  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is partially vectorized.
88% of vector register length is used (average across all SSE/AVX instructions).


Details
86% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):
 - 66% of SSE/AVX instructions that are not load, store, addition, subtraction nor multiply instructions are used in vector version.



      6.1.10.1.3  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

Found no such bottlenecks but see expert reports for more complex bottlenecks.




      6.1.10.1.4  -  Masked instructions
  ----------------------------------------------------------------------------------------------------------

Detected masked instructions.

Details
Vector registers are partially exploited, which is expected if your loop is irregular or mixes elements of different sizes.

Workaround
If your loop is irregular, try to remove or hoist conditional structures out of your loop. If it mixes elements of different sizes, try to uniformize them.




      6.1.10.1.5  -  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.
 - KMOVW: 1 occurrences<<list_path_1_complex_1>>
 - VCOMPRESSPD: 4 occurrences<<list_path_1_complex_2>>
 - VPBROADCASTD: 2 occurrences<<list_path_1_complex_3>>
 - VPCOMPRESSD: 2 occurrences<<list_path_1_complex_4>>



      6.1.10.1.6  -  Slow data structures access
  ----------------------------------------------------------------------------------------------------------

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

Details
 - Irregular (variable stride) or indirect: 5 occurrence(s)
Non-unit stride (uncontiguous) accesses are not efficiently using data caches


Workaround
Try to remove indirect accesses. If applicable, precompute elements out of the innermost loop.


      6.1.10.1.7  -  Vector unaligned load/store instructions
  ----------------------------------------------------------------------------------------------------------

Detected 4 optimal vector unaligned load/store instructions.


Details
 - VMOVUPD: 4 occurrences<<list_path_1_vec_align_1>>


Workaround
Use vector aligned instructions:
 1) align your arrays on 64 bytes boundaries: replace { void *p = malloc (size); } with { void *p; posix_memalign (&p, 64, size); }.
 2) inform your compiler that your arrays are vector aligned: if array 'foo' is 64 bytes-aligned, define a pointer 'p_foo' as __builtin_assume_aligned (foo, 64) and use it instead of 'foo' in the loop.
<<image_vec_align>>


      6.1.10.1.8  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

4 AVX-512 instructions are processing arithmetic or math operations on double precision FP elements in vector mode (eight at a time).



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

The binary loop does not contain any FP arithmetical operations.
The binary loop is loading 258 bytes (32 double precision FP elements).
The binary loop is storing 386 bytes (48 double precision FP elements).





[MAQAO] Info: STOP THE REPORT GENERATION
[MAQAO] Info: 
[MAQAO] Info: If your application produces files, they can be found in directory "/beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-5599/intel/miniqmc/run/oneview_runs/defaults/icx/oneview_run_1786546675"
[MAQAO] Info: 
