	*************************************************
	*                                               *
	*          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/178-654-6841/intel/miniqmc/run/oneview_runs/defaults/gcc/oneview_results_1786548085 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/178-654-6841/intel/miniqmc/run/oneview_runs/defaults/gcc/oneview_results_1786548085


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


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

  Application:			/beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/run/base_runs/defaults/gcc/exec
  Timestamp:			2026-08-12 17:21:25
  Universal Timestamp:		1786548085
  Experiment Type:		MPI; OpenMP; Throughput; 
  Machine:			isix07.benchmarkcenter.megware.com
  Architecture:			x86_64
  Micro Architecture:		GRANITE_RAPIDS
  Model Name:			Intel(R) Xeon(R) 6972P
  Cache Size:			491520 KB
  Number of Cores:		96
  OS Version:			Linux 5.14.0-687.31.1.el9_8.x86_64 #1 SMP PREEMPT_DYNAMIC Sat Aug 1 05:38:01 EDT 2026
  Compilation Options:		
		exec: GNU C++17 15.1.0 -march=graniterapids -mmmx -mpopcnt -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -mavx -mavx2 -mno-sse4a -mno-fma4 -mno-xop -mfma -mavx512f -mbmi -mbmi2 -maes -mpclmul -mavx512vl -mavx512bw -mavx512dq -mavx512cd -mavx512vbmi -mavx512ifma -mavx512vpopcntdq -mavx512vbmi2 -mgfni -mvpclmulqdq -mavx512vnni -mavx512bitalg -mavx512bf16 -mno-avx512vp2intersect -mno-3dnow -madx -mabm -mcldemote -mclflushopt -mclwb -mno-clzero -mcx16 -menqcmd -mf16c -mfsgsbase -mfxsr -mno-hle -msahf -mno-lwp -mlzcnt -mmovbe -mmovdir64b -mmovdiri -mno-mwaitx -mpconfig -mpku -mprfchw -mptwrite -mrdpid -mrdrnd -mrdseed -mno-rtm -mserialize -msgx -msha -mshstk -mno-tbm -mtsxldtrk -mvaes -mwaitpkg -mwbnoinvd -mxsave -mxsavec -mxsaveopt -mxsaves -mamx-tile -mamx-int8 -mamx-bf16 -muintr -mhreset -mno-kl -mno-widekl -mavxvnni -mavx512fp16 -mno-avxifma -mno-avxvnniint8 -mno-avxneconvert -mno-cmpccxadd -mamx-fp16 -mprefetchi -mno-raoint -mno-amx-complex -mno-avxvnniint16 -mno-sm3 -mno-sha512 -mno-sm4 -mno-apxf -mno-usermsr -mavx10.1-256 -mavx10.1-512 -mno-avx10.2 -mno-amx-avx512 -mno-amx-tf32 -mno-amx-transpose -mno-amx-fp8 -mno-movrs -mno-amx-movrs --param=l1-cache-size=48 --param=l1-cache-line-size=64 --param=l2-cache-size=491520 -mtune=graniterapids -g -O3 -O3 -std=c++17 -fno-omit-frame-pointer -fcf-protection=none -fopenmp -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math 
  Number of processes observed:	6
  Number of threads observed:	192
  MAQAO version:		2026.1.0
  MAQAO build:			6d1be1d51c1e63266254997eb301734a7264775d::20260810-150026




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

  Total Time:				87.20 s
  Max (Thread Active Time):		86.25 s
  Average Active Time:			83.63 s
  Activity Ratio:			96.4 %
  Average number of active threads:	184.141
  Affinity Stability:			99.0 %
  Time spent in analyzed loops:		64.8 %
  Time spent in analyzed innermost loops: 64.7 %
  Time spent in user code:		65.1 %
  Compilation Options Score:		100
  Array Access Efficiency:		93.1 %

   Potential Speedups
  ----------------------------------------------------
  Perfect Flow Complexity:		1.03
  Perfect OpenMP/MPI/Pthread/TBB:	1.00
  Perfect OpenMP/MPI/Pthread/TBB + Load Distribution:	1.03
  If No Scalar Integer:
      Potential Speedup:		1.05
      Nb Loops to get 80%:		1
  If FP Vectorized:
      Potential Speedup:		1.32
      Nb Loops to get 80%:		3
  If Fully Vectorized:
      Potential Speedup:		1.87
      Nb Loops to get 80%:		5
  If Only FP Arithmetic:
      Potential Speedup:		1.18
      Nb Loops to get 80%:		1




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

  If No Scalar Integer:
      Number of loops   | 1      | 10     | 20     | 29     | 40     | 
      Cumulated Speedup | 1.0470 | 1.0485 | 1.0485 | 1.0485 | 1.0485 | 
  Top 5 loops:
    exec - 2660:	1.047
    exec - 587:	1.0474
    exec - 151:	1.0479
    exec - 285:	1.0482
    exec - 335:	1.0483

  If FP Vectorized:
      Number of loops   | 1      | 10     | 20     | 29     | 40     | 
      Cumulated Speedup | 1.1066 | 1.3120 | 1.3165 | 1.3173 | 1.3173 | 
  Top 5 loops:
    exec - 962:	1.1066
    exec - 1026:	1.2151
    exec - 3166:	1.2976
    exec - 2666:	1.3011
    exec - 2667:	1.3042

  If Fully Vectorized:
      Number of loops   | 1      | 10     | 20     | 29     | 40     | 
      Cumulated Speedup | 1.1432 | 1.8014 | 1.8546 | 1.8687 | 1.8738 | 
  Top 5 loops:
    exec - 2660:	1.1432
    exec - 1026:	1.3209
    exec - 962:	1.5198
    exec - 3166:	1.6821
    exec - 334:	1.7447

  If Only FP Arithmetic:
      Number of loops   | 1      | 10     | 20     | 29     | 40     | 
      Cumulated Speedup | 1.1456 | 1.1737 | 1.1752 | 1.1752 | 1.1752 | 
  Top 5 loops:
    exec - 2660:	1.1456
    exec - 334:	1.1555
    exec - 1031:	1.1621
    exec - 1138:	1.1658
    exec - 1130:	1.1689



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


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

  [4 / 4] Application profile is long enough (86.25 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.

  [2.9981495668496 / 3] Most of time spent in analyzed modules (99.94%) comes from functions compiled with architecture specialization option
-march=graniterapids


  [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 0.03 % 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 (64.83%)
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 95.91% of observed threads are actually active 

  [4 / 4] CPU activity is good
CPU cores are active 96.43% of time

  [4 / 4] Loop profile is not flat
At least one loop coverage is greater than 4% (19.82%), representing an hotspot for the application

  [4 / 4] Enough time of the experiment time spent in analyzed innermost loops (64.66%)
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 (98.98%)
Threads are not migrating to CPU cores: probably successfully pinned

  [3 / 3] Less than 10% (0.04%) 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 (0.76%)

  [3 / 3] Cumulative Outermost/In between loops coverage (0.17%) lower than cumulative innermost loop coverage (64.66%)
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% (0.01%) is spend in Libm/SVML (special functions)



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

  Top 5 loops:
   + exec - 962 :
     analysis: Execution Time: 19 % - Vectorization Ratio: 100.00 % - Vector Length Use: 50.00 %

   + exec - 2660:
     analysis: Execution Time: 14 % - Vectorization Ratio: 27.27 % - Vector Length Use: 15.91 %
     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: 20
        [20] [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 10 issues ( = data accesses) costing 2
            point each.
     Vectorization Roadblocks: 20
        [20] [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 10 issues ( = data accesses) costing 2
            point each.

   + exec - 1026:
     analysis: Execution Time: 13 % - Vectorization Ratio: 15.22 % - Vector Length Use: 14.40 %

   + exec - 3166:
     analysis: Execution Time: 7 % - Vectorization Ratio: 12.24 % - Vector Length Use: 14.03 %
     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 - 334 :
     analysis: Execution Time: 2 % - Vectorization Ratio: 0.00 % - Vector Length Use: 10.00 %
     Control Flow Issues: 1000
        [1000] [SA] Too many paths (6561 paths) - Simplify control structure. There are 6561 issues ( = paths) costing 1
            point, limited to 1000.
     Vectorization Roadblocks: 1000
        [1000] [SA] Too many paths (6561 paths) - Simplify control structure. There are 6561 issues ( = paths) costing 1
            point, limited to 1000.



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


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

   Category | IO     | Exe    | System  | Others  | Memory | String | MPI   | TBB   | OMP   | Pthread | Math  |
  ----------+--------+--------+---------+---------+--------+--------+-------+-------+-------+---------+-------+
   Time (%) | 0.00   | 65.11  | 0.47    | 0.03    | 0.03   | 1.46   | 0.00  | 0.00  | 0.05  | 0.00    | 32.85 |




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

   Buckets                    | Nb Functions              | Coverage                  | Cumulated Coverage        |
  ----------------------------+---------------------------+---------------------------+---------------------------+
   > 8%                       | 5                         | 77.75                     | 77.75                     |
   4% to 8%                   | 1                         | 7.48                      | 85.23                     |
   2% to 4%                   | 1                         | 2.46                      | 87.69                     |
   1% to 2%                   | 3                         | 3.56                      | 91.25                     |
   0.5% to 1%                 | 8                         | 5.65                      | 96.90                     |
   0.25% to 0.5%              | 3                         | 0.93                      | 97.83                     |
   0.125% to 0.25%            | 7                         | 1.16                      | 98.99                     |
   < 0.125%                   | 83                        | 0.92                      | 99.91                     |




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

   Buckets                    | Nb Loops                  | Coverage                  | Cumulated Coverage        |
  ----------------------------+---------------------------+---------------------------+---------------------------+
   > 8%                       | 3                         | 48.22                     | 48.22                     |
   4% to 8%                   | 1                         | 7.47                      | 55.69                     |
   2% to 4%                   | 1                         | 2.33                      | 58.02                     |
   1% to 2%                   | 0                         | 0.00                      | 58.02                     |
   0.5% to 1%                 | 5                         | 2.94                      | 60.96                     |
   0.25% to 0.5%              | 6                         | 2.02                      | 62.98                     |
   0.125% to 0.25%            | 4                         | 0.81                      | 63.79                     |
   < 0.125%                   | 43                        | 0.87                      | 64.66                     |


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


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

   Function                                               | Module              | Coverage (%)   | Time (s)       |
  --------------------------------------------------------+---------------------+----------------+----------------+
   miniqmcreference::einspline_spo_ref<double>::evalua... | exec                | 19.83          | 16.59          |
   mkl_blas_avx512_dgemm_kernel_nocopy_tn_b1              | libmkl_avx512.so.3  | 16.77          | 14.02          |
   qmcplusplus::SoaDistanceTableAAOMPTarget<double, 3u... | exec                | 14.96          | 12.51          |
   void miniqmcreference::MultiBsplineEvalRef::evaluat... | exec                | 13.93          | 11.65          |
   mkl_blas_avx512_dgemm_kernel_0                         | libmkl_avx512.so.3  | 12.27          | 10.26          |
   qmcplusplus::SoaDistanceTableABOMPTarget<double, 3u... | exec                | 7.48           | 6.26           |
   miniqmcreference::TwoBodyJastrowRef<qmcplusplus::Bs... | exec                | 2.46           | 2.05           |
   __GI___strcasecmp_l_sse2                               | libc.so.6           | 1.46           | 1.22           |
   mkl_blas_avx512_dgemm_kernel_nocopy_tn_b0              | libmkl_avx512.so.3  | 1.08           | 0.90           |
   mkl_blas_avx512_xdgemv_kernel_t                        | libmkl_avx512.so.3  | 1.02           | 0.85           |


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


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

   Loop Id        | Module              | Source Location                                        | Coverage (%)   |
  ----------------+---------------------+--------------------------------------------------------+----------------+
   962            | exec                | MultiBsplineRef.hpp:68-71                              | 19.82          |
   2660           | exec                | VectorSoAContainer.h:244-244,VectorSoAContainer.h:2... | 14.95          |
   1026           | exec                | MultiBsplineRef.hpp:242-262                            | 13.45          |
   3166           | exec                | ParticleBConds3DSoa.h:280-298,SoaDistanceTableABOMP... | 7.47           |
   334            | exec                | BsplineFunctor.h:236-241                               | 2.33           |
   592            | exec                | TwoBodyJastrowRef.h:342-347                            | 0.70           |
   1138           | exec                | OperatorTags.h:63-63,OperatorTags.h:94-94,inner_pro... | 0.64           |
   1031           | exec                | einspline_spo_ref.hpp:223-227,TinyVector.h:145-145     | 0.57           |
   1130           | exec                | OperatorTags.h:63-63,OperatorTags.h:94-94,inner_pro... | 0.52           |
   2666           | exec                | ParticleBConds3DSoa.h:237-255                          | 0.52           |





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


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





      6.1.1  -  Loop 962 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/Numerics/Spline2/MultiBsplineRef.hpp:68-71.

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

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

37% of peak computational performance is used (12.00 out of 32.00 FLOP per cycle (GFLOPS @ 1GHz))

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

Your loop is vectorized, but using only 256 out of 512 bits (AVX/AVX2 instructions on AVX-512 processors).
<<image_4x64_512>>

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


Workaround
Read the "512-bits vectorization" report at "Potential" confidence level.


      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 12.00 to 10.00 cycles (1.20x speedup).


Workaround
Reduce the number of FP multiply/FMA instructions




      6.1.1.1.3  -  512-bits vectorization
  ----------------------------------------------------------------------------------------------------------

On some x86 processors supporting 512-bits vectorization, compilers are often too conservative and limit vectorization to 256 bits. Performance can then be improved by enforcing 512-bits vectorization, especially with many vectorized and high trip count loops. 512-bits vectorization performance overhead (compared to 256-bits) is generally lower on newer processors.


Workaround
Recompile with -mprefer-vector-width=512


      6.1.1.1.4  -  FMA
  ----------------------------------------------------------------------------------------------------------

Detected 48 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.
Estimated speedup by perfect pairing: 1.20x.
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).




      6.1.1.1.5  -  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>>


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

24 AVX instructions are processing arithmetic or math operations on double precision FP elements in vector mode (four 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 144 FP arithmetical operations:
 - 64: addition or subtraction (48 inside FMA instructions)
 - 80: multiply (48 inside FMA instructions)
The binary loop is loading 584 bytes (73 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 0.22 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 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







      6.1.2  -  Loop 2660 from exec
  ==========================================================================================================

The loop is defined in:
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/VectorSoAContainer.h:244,263
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/Particle/SoaDistanceTableAAOMPTarget.h:440-442
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVector.h:182


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

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

0% of peak computational performance is used (0.00 out of 64.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.2.1.1  -  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 6.67 to 4.67 cycles (1.43x speedup).

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



      6.1.2.1.2  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is probably not vectorized.
Only 15% 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 6.67 to 1.08 cycles (6.15x speedup).

Details
Store and arithmetical SSE/AVX instructions are used in scalar version (process only one data element in vector registers).
Since your execution units are vector units, only a vectorized 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)<<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)



      6.1.2.1.3  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

Performance is limited by reading data from caches/RAM (load units are a bottleneck).


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




No data for this section



      6.1.2.1.4  -  Slow data structures access
  ----------------------------------------------------------------------------------------------------------

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

Details
 - Constant unknown stride: 4 occurrence(s)
 - Constant non-unit stride: 6 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)



      6.1.2.1.5  -  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.


      6.1.2.1.6  -  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 208 bytes.
The binary loop is storing 64 bytes.


      6.1.2.1.7  -  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 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







      6.1.3  -  Loop 1026 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/Numerics/Spline2/MultiBsplineRef.hpp:242-262.

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

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

10% of peak computational performance is used (3.28 out of 32.00 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 12.50 to 1.56 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
 - 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 12.50 to 11.00 cycles (1.14x speedup).


Workaround
Reduce the number of FP multiply/FMA instructions




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

Detected 16 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.
Estimated speedup by perfect pairing: 1.14x.
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
  ----------------------------------------------------------------------------------------------------------

25 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 41 FP arithmetical operations:
 - 19: addition or subtraction (16 inside FMA instructions)
 - 22: multiply (16 inside FMA instructions)
The binary loop is loading 136 bytes (17 double precision FP elements).
The binary loop is storing 80 bytes (10 double precision FP elements).


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

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







      6.1.4  -  Loop 3166 from exec
  ==========================================================================================================

The loop is defined in:
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/Particle/Lattice/ParticleBConds3DSoa.h:280-298
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/Particle/SoaDistanceTableABOMPTarget.h:228


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

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

9% of peak computational performance is used (3.00 out of 32.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.4.1.1  -  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 15.00 to 2.25 cycles (6.67x 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
 - 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)



      6.1.4.1.2  -  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 15.00 to 10.50 cycles (1.43x speedup).


Workaround
Reduce the number of FP multiply/FMA instructions




      6.1.4.1.3  -  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.
Estimated speedup by perfect pairing: 1.07x.
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).




      6.1.4.1.4  -  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.
 - VRNDSCALESD: 3 occurrences<<list_path_1_complex_1>>



      6.1.4.1.5  -  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).



      6.1.4.1.6  -  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 40 bytes (5 double precision FP elements).
The binary loop is storing 32 bytes (4 double precision FP elements).


      6.1.4.1.7  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

Arithmetic intensity is 0.62 FP operations per loaded or stored byte.







      6.1.5  -  Loop 334 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/BsplineFunctor.h:236-241.

The related source loop is not unrolled or unrolled with no peel/tail loop.
Warnings:
 - Ignoring paths for analysis
 - Too many paths. If you really need to analyze all of the 6561 paths individually, rerun with max-paths=6561
 - RecMII not computed since number of paths is unknown or > max_paths
 - Streams not analyzed since number of paths is unknown or > max_paths

Try to simplify control and/or increase the maximum number of paths per function/loop through the 'max-paths-nb' option.

This loop has 6561 execution paths.

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)


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

0% of peak computational performance is used (0.00 out of 32.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.5.1.1  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is not vectorized.
Only 10% 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 12.50 to 1.03 cycles (12.12x 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)<<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)



      6.1.5.1.2  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

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



No data for this section



      6.1.5.1.3  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

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



      6.1.5.1.4  -  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 64 bytes (8 double precision FP elements).
The binary loop is storing 64 bytes (8 double precision FP elements).







      6.1.6  -  Loop 592 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/Jastrow/TwoBodyJastrowRef.h:342-347.

It is main loop of related source loop which is unrolled by 2 (including vectorization).

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

18% of peak computational performance is used (6.00 out of 32.00 FLOP per cycle (GFLOPS @ 1GHz))

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

Your loop is vectorized, but using only 256 out of 512 bits (AVX/AVX2 instructions on AVX-512 processors).
<<image_4x64_512>>

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


Workaround
Read the "512-bits vectorization" report at "Potential" confidence level.


      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  -  512-bits vectorization
  ----------------------------------------------------------------------------------------------------------

On some x86 processors supporting 512-bits vectorization, compilers are often too conservative and limit vectorization to 256 bits. Performance can then be improved by enforcing 512-bits vectorization, especially with many vectorized and high trip count loops. 512-bits vectorization performance overhead (compared to 256-bits) is generally lower on newer processors.


Workaround
Recompile with -mprefer-vector-width=512


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

Detected 48 FMA (fused multiply-add) operations.




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

Detected 20 optimal vector unaligned load/store instructions.


Details
 - VMOVUPD: 20 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
  ----------------------------------------------------------------------------------------------------------

12 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 96 FP arithmetical operations:
 - 48: addition or subtraction (all inside FMA instructions)
 - 48: multiply (all inside FMA instructions)
The binary loop is loading 640 bytes (80 double precision FP elements).
The binary loop is storing 128 bytes (16 double precision FP elements).


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

Arithmetic intensity is 0.12 FP operations per loaded or stored byte.







      6.1.7  -  Loop 1138 from exec
  ==========================================================================================================

The loop is defined in:
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h:63,94
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/Platforms/CPU/SIMD/inner_product.hpp:155


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

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

10% of peak computational performance is used (3.43 out of 32.00 FLOP per cycle (GFLOPS @ 1GHz))

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

Your loop is vectorized, but using only 256 out of 512 bits (AVX/AVX2 instructions on AVX-512 processors).
<<image_4x64_512>>

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


Workaround
Read the "512-bits vectorization" report at "Potential" confidence level.


      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 28.00 to 12.67 cycles (2.21x speedup).


Workaround
Reduce the number of FP multiply/FMA instructions




      6.1.7.1.3  -  512-bits vectorization
  ----------------------------------------------------------------------------------------------------------

On some x86 processors supporting 512-bits vectorization, compilers are often too conservative and limit vectorization to 256 bits. Performance can then be improved by enforcing 512-bits vectorization, especially with many vectorized and high trip count loops. 512-bits vectorization performance overhead (compared to 256-bits) is generally lower on newer processors.


Workaround
Recompile with -mprefer-vector-width=512


      6.1.7.1.4  -  FMA
  ----------------------------------------------------------------------------------------------------------

Detected 48 FMA (fused multiply-add) operations.




      6.1.7.1.5  -  Vector unaligned load/store instructions
  ----------------------------------------------------------------------------------------------------------

Detected 16 optimal vector unaligned load/store instructions.


Details
 - VMOVUPD: 16 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>>


      6.1.7.1.6  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

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



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

The binary loop is composed of 96 FP arithmetical operations:
 - 48: addition or subtraction (all inside FMA instructions)
 - 48: multiply (all inside FMA instructions)
The binary loop is loading 512 bytes (64 double precision FP elements).


      6.1.7.1.8  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

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







      6.1.8  -  Loop 1031 from exec
  ==========================================================================================================

The loop is defined in:
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/QMCWaveFunctions/einspline_spo_ref.hpp:223-227
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/Numerics/OhmmsPETE/TinyVector.h:145


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 64.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.8.1.1  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is not 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 8.00 to 1.25 cycles (6.40x speedup).

Details
11% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):
 - 0% of SSE/AVX loads are used in vector version.
 - 25% of SSE/AVX stores 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
 - 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)



      6.1.8.1.2  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

Performance is limited by writing data to caches/RAM (the store unit is a bottleneck).


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




No data for this section



      6.1.8.1.3  -  Slow data structures access
  ----------------------------------------------------------------------------------------------------------

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

Details
 - Constant non-unit 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)



      6.1.8.1.4  -  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.


      6.1.8.1.5  -  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 160 bytes.
The binary loop is storing 160 bytes.


      6.1.8.1.6  -  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







      6.1.9  -  Loop 1130 from exec
  ==========================================================================================================

The loop is defined in:
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/Numerics/PETE/OperatorTags.h:63,94
 - /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/Platforms/CPU/SIMD/inner_product.hpp:155


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

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

10% of peak computational performance is used (3.43 out of 32.00 FLOP per cycle (GFLOPS @ 1GHz))

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

Your loop is vectorized, but using only 256 out of 512 bits (AVX/AVX2 instructions on AVX-512 processors).
<<image_4x64_512>>

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


Workaround
Read the "512-bits vectorization" report at "Potential" confidence level.


      6.1.9.1.2  -  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 28.00 to 12.83 cycles (2.18x speedup).


Workaround
Reduce the number of FP multiply/FMA instructions




      6.1.9.1.3  -  512-bits vectorization
  ----------------------------------------------------------------------------------------------------------

On some x86 processors supporting 512-bits vectorization, compilers are often too conservative and limit vectorization to 256 bits. Performance can then be improved by enforcing 512-bits vectorization, especially with many vectorized and high trip count loops. 512-bits vectorization performance overhead (compared to 256-bits) is generally lower on newer processors.


Workaround
Recompile with -mprefer-vector-width=512


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

Detected 48 FMA (fused multiply-add) operations.




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

Detected 16 optimal vector unaligned load/store instructions.


Details
 - VMOVUPD: 16 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
  ----------------------------------------------------------------------------------------------------------

12 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 96 FP arithmetical operations:
 - 48: addition or subtraction (all inside FMA instructions)
 - 48: multiply (all inside FMA instructions)
The binary loop is loading 512 bytes (64 double precision FP elements).


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

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







      6.1.10  -  Loop 2666 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/178-654-6841/intel/miniqmc/build/miniqmc/src/Particle/Lattice/ParticleBConds3DSoa.h:237-255.

The related source loop is multi-versionned.

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

37% of peak computational performance is used (12.00 out of 32.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.10.1.1  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is vectorized, but using only 256 out of 512 bits (AVX/AVX2 instructions on AVX-512 processors).
<<image_4x64_512>>

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


Workaround
Read the "512-bits vectorization" report at "Potential" confidence level.


      6.1.10.1.2  -  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 15.00 to 10.50 cycles (1.43x speedup).


Workaround
Reduce the number of FP multiply/FMA instructions




      6.1.10.1.3  -  512-bits vectorization
  ----------------------------------------------------------------------------------------------------------

On some x86 processors supporting 512-bits vectorization, compilers are often too conservative and limit vectorization to 256 bits. Performance can then be improved by enforcing 512-bits vectorization, especially with many vectorized and high trip count loops. 512-bits vectorization performance overhead (compared to 256-bits) is generally lower on newer processors.


Workaround
Recompile with -mprefer-vector-width=512


      6.1.10.1.4  -  FMA
  ----------------------------------------------------------------------------------------------------------

Detected 56 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.
Estimated speedup by perfect pairing: 1.07x.
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).




      6.1.10.1.5  -  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.
 - VRNDSCALEPD: 3 occurrences<<list_path_1_complex_1>>



      6.1.10.1.6  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

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



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

The binary loop is composed of 180 FP arithmetical operations:
 - 92: addition or subtraction (56 inside FMA instructions)
 - 84: multiply (56 inside FMA instructions)
 - 4: square root
The binary loop is loading 304 bytes (38 double precision FP elements).
The binary loop is storing 128 bytes (16 double precision FP elements).


      6.1.10.1.8  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

Arithmetic intensity is 0.42 FP operations per loaded or stored byte.


      6.1.10.1.9  -  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 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





[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/178-654-6841/intel/miniqmc/run/oneview_runs/defaults/gcc/oneview_run_1786548085"
[MAQAO] Info: 
