	*************************************************
	*                                               *
	*          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/179-075-5378/intel/TeaLeaf/run/oneview_runs/compilers/gcc_2/oneview_results_1790757542 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/179-075-5378/intel/TeaLeaf/run/oneview_runs/compilers/gcc_2/oneview_results_1790757542


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


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

  Application:			/beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/run/binaries/gcc_2/exec
  Timestamp:			2026-09-30 10:39:02
  Universal Timestamp:		1790757542
  Experiment Type:		MPI; OpenMP; Throughput; 
  Machine:			ins01.benchmarkcenter.megware.com
  Architecture:			x86_64
  Micro Architecture:		ZEN_V4
  Model Name:			AMD EPYC 9654 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: GNU C++17 11.5.0 20240719 (Red Hat 11.5.0-14) -march=znver4 -mprefer-vector-width=512 -g -O3 -O3 -std=c++17 -funroll-loops -ffast-math -fno-omit-frame-pointer -fcf-protection=none -fopenmp 
  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:				22.92 s
  Max (Thread Active Time):		17.43 s
  Average Active Time:			17.15 s
  Activity Ratio:			98.3 %
  Average number of active threads:	143.668
  Affinity Stability:			99.5 %
  Time spent in analyzed loops:		42.8 %
  Time spent in analyzed innermost loops: 42.0 %
  Time spent in user code:		43.3 %
  Compilation Options Score:		100
  Array Access Efficiency:		97.1 %

   Potential Speedups
  ----------------------------------------------------
  Perfect Flow Complexity:		1.00
  Perfect OpenMP/MPI/Pthread/TBB:	1.67
  Perfect OpenMP/MPI/Pthread/TBB + Load Distribution:	2.34
  If No Scalar Integer:
      Potential Speedup:		1.01
      Nb Loops to get 80%:		4
  If FP Vectorized:
      Potential Speedup:		1.00
      Nb Loops to get 80%:		2
  If Fully Vectorized:
      Potential Speedup:		1.00
      Nb Loops to get 80%:		5
  If Only FP Arithmetic:
      Potential Speedup:		1.12
      Nb Loops to get 80%:		2




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

  If No Scalar Integer:
      Number of loops   | 1      | 6      | 12     | 17     | 23     | 
      Cumulated Speedup | 1.0037 | 1.0058 | 1.0064 | 1.0064 | 1.0064 | 
  Top 5 loops:
    exec - 20:	1.0037
    exec - 22:	1.0044
    exec - 18:	1.0049
    exec - 26:	1.0053
    exec - 98:	1.0055

  If FP Vectorized:
      Number of loops   | 1      | 6      | 12     | 17     | 23     | 
      Cumulated Speedup | 1.0002 | 1.0004 | 1.0004 | 1.0004 | 1.0004 | 
  Top 5 loops:
    exec - 22:	1.0002
    exec - 18:	1.0004
    exec - 26:	1.0004
    exec - 130:	1.0004
    exec - 20:	1.0004

  If Fully Vectorized:
      Number of loops   | 1      | 6      | 12     | 17     | 23     | 
      Cumulated Speedup | 1.0010 | 1.0026 | 1.0031 | 1.0031 | 1.0031 | 
  Top 5 loops:
    exec - 22:	1.001
    exec - 18:	1.0015
    exec - 26:	1.0019
    exec - 70:	1.0022
    exec - 74:	1.0025

  If Only FP Arithmetic:
      Number of loops   | 1      | 6      | 12     | 17     | 23     | 
      Cumulated Speedup | 1.0790 | 1.1175 | 1.1202 | 1.1211 | 1.1212 | 
  Top 5 loops:
    exec - 20:	1.079
    exec - 27:	1.1127
    exec - 22:	1.114
    exec - 93:	1.1153
    exec - 96:	1.1165



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


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

  [4 / 4] Application profile is long enough (17.43 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.

  [2.9987593138845 / 3] Most of time spent in analyzed modules (99.96%) comes from functions compiled with architecture specialization option
-march=znver4


  [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 0.01 % 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 (42.80%)
If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on
application performances.

  [3 / 4] A significant amount of threads are idle (25.17%)
On average, more than 10% of observed threads are idle. Such threads are probably IO/sync waiting. Some hints: use
faster filesystems to read/write data, improve parallel load balancing and/or scheduling.

  [4 / 4] CPU activity is good
CPU cores are active 98.33% of time

  [4 / 4] Loop profile is not flat
At least one loop coverage is greater than 4% (20.51%), representing an hotspot for the application

  [4 / 4] Enough time of the experiment time spent in analyzed innermost loops (42.03%)
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 (99.48%)
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

  [0 / 3] Too many functions do not use all threads
Functions running on a reduced number of threads (typically sequential code) cover at least 10% of application
walltime (20.75%). Check both "Max Inclusive Time Over Threads" and "Nb Threads" in Functions or Loops tabs and
consider parallelizing sequential regions or improving parallelization of regions running on a reduced number of
threads

  [3 / 3] Cumulative Outermost/In between loops coverage (0.77%) lower than cumulative innermost loop coverage (42.03%)
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 - 20  :
     analysis: Execution Time: 20 % - Vectorization Ratio: 100.00 % - Vector Length Use: 100.00 %

   + exec - 24  :
     analysis: Execution Time: 14 % - Vectorization Ratio: 100.00 % - Vector Length Use: 100.00 %

   + exec - 27  :
     analysis: Execution Time: 7 % - Vectorization Ratio: 100.00 % - Vector Length Use: 100.00 %
     Data Access Issues: 4
        [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.
     Vectorization Roadblocks: 4
        [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.

   + exec - 22  :
     analysis: Execution Time: 0 % - Vectorization Ratio: 23.72 % - Vector Length Use: 26.48 %
     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.
     Control Flow Issues: 99
        [97] [SA] Too many paths (93 paths) - Simplify control structure. There are 93 issues ( = paths) costing 1
            point each with a malus of 4 points.
        [2] [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points.
     Data Access Issues: 7
        [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 (INSERT/EXTRACT, BLEND/MERGE) - Simplify
            data access and try to get stride 1 access. There are 5 issues (= instructions) costing 1 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: 99
        [97] [SA] Too many paths (93 paths) - Simplify control structure. There are 93 issues ( = paths) costing 1
            point each with a malus of 4 points.
        [2] [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points.
     Inefficient Vectorization: 5
        [5] [SA] Presence of special instructions executing on a single port (INSERT/EXTRACT, BLEND/MERGE) - Simplify
            data access and try to get stride 1 access. There are 5 issues (= instructions) costing 1 point each.

   + exec - 26  :
     analysis: Execution Time: 0 % - Vectorization Ratio: 28.92 % - Vector Length Use: 35.32 %
     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.
     Control Flow Issues: 160
        [158] [SA] Too many paths (154 paths) - Simplify control structure. There are 154 issues ( = paths) costing 1
            point each with a malus of 4 points.
        [2] [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points.
     Data Access Issues: 2
        [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
        [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: 160
        [158] [SA] Too many paths (154 paths) - Simplify control structure. There are 154 issues ( = paths) costing 1
            point each with a malus of 4 points.
        [2] [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points.



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


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

   Category | IO     | Exe    | System  | Others  | Memory | String | MPI   | TBB   | OMP   | Pthread | Math  |
  ----------+--------+--------+---------+---------+--------+--------+-------+-------+-------+---------+-------+
   Time (%) | 0.00   | 43.32  | 0.09    | 0.01    | 0.00   | 0.09   | 0.52  | 0.00  | 55.96 | 0.00    | 0.00  |




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

   Buckets                    | Nb Functions              | Coverage                  | Cumulated Coverage        |
  ----------------------------+---------------------------+---------------------------+---------------------------+
   > 8%                       | 3                         | 89.27                     | 89.27                     |
   4% to 8%                   | 1                         | 7.19                      | 96.46                     |
   2% to 4%                   | 0                         | 0.00                      | 96.46                     |
   1% to 2%                   | 1                         | 1.47                      | 97.92                     |
   0.5% to 1%                 | 0                         | 0.00                      | 97.92                     |
   0.25% to 0.5%              | 1                         | 0.25                      | 98.17                     |
   0.125% to 0.25%            | 2                         | 0.41                      | 98.59                     |
   < 0.125%                   | 126                       | 1.41                      | 100.00                    |




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

   Buckets                    | Nb Loops                  | Coverage                  | Cumulated Coverage        |
  ----------------------------+---------------------------+---------------------------+---------------------------+
   > 8%                       | 2                         | 34.58                     | 34.58                     |
   4% to 8%                   | 1                         | 7.00                      | 41.59                     |
   2% to 4%                   | 0                         | 0.00                      | 41.59                     |
   1% to 2%                   | 0                         | 0.00                      | 41.59                     |
   0.5% to 1%                 | 0                         | 0.00                      | 41.59                     |
   0.25% to 0.5%              | 0                         | 0.00                      | 41.59                     |
   0.125% to 0.25%            | 0                         | 0.00                      | 41.59                     |
   < 0.125%                   | 32                        | 0.44                      | 42.03                     |


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


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

   Function                                               | Module              | Coverage (%)   | Time (s)       |
  --------------------------------------------------------+---------------------+----------------+----------------+
   omp_get_num_procs                                      | libgomp.so.1.0.0    | 54.18          | 9.29           |
   cg_calc_w(int, int, int, double*, double const*, do... | exec                | 20.70          | 3.55           |
   cg_calc_ur(int, int, int, double, double*, double*,... | exec                | 14.39          | 2.47           |
   cg_calc_p(int, int, int, double, double*, double co... | exec                | 7.19           | 1.23           |
   omp_fulfill_event                                      | libgomp.so.1.0.0    | 1.47           | 0.25           |
   mca_btl_sm_component_progress                          | libopen-pal.so.8... | 0.25           | 1.04           |
   unknown_kernel_region                                  | kernel              | 0.25           | 0.05           |
   pack_top(int, int, int, int, double const*, double*... | exec                | 0.16           | 0.05           |
   pack_bottom(int, int, int, int, double const*, doub... | exec                | 0.12           | 0.07           |
   omp_get_num_threads                                    | libgomp.so.1.0.0    | 0.11           | 0.02           |


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


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

   Loop Id        | Module              | Source Location                                        | Coverage (%)   |
  ----------------+---------------------+--------------------------------------------------------+----------------+
   20             | exec                | cg.cpp:86-90                                           | 20.51          |
   24             | exec                | cg.cpp:108-113                                         | 14.08          |
   27             | exec                | cg.cpp:128-131                                         | 7.00           |
   22             | exec                | cg.cpp:105-105,cg.cpp:108-113                          | 0.24           |
   26             | exec                | cg.cpp:125-125,cg.cpp:128-131                          | 0.15           |
   18             | exec                | cg.cpp:83-90                                           | 0.13           |
   93             | exec                | pack_halos.cpp:51-53                                   | 0.12           |
   96             | exec                | pack_halos.cpp:69-71                                   | 0.11           |
   98             | exec                | pack_halos.cpp:83-83,pack_halos.cpp:86-88              | 0.05           |
   101            | exec                | pack_halos.cpp:100-100,pack_halos.cpp:103-105          | 0.05           |





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


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





      6.1.1  -  Loop 20 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/build/TeaLeaf/src/omp/cg.cpp:86-90.

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

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

53% of peak computational performance is used (12.86 out of 24.00 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:
 - 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 18.67 to 12.00 cycles (1.56x speedup).


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





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

Detected 64 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).




      6.1.1.1.4  -  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: 2 occurrences<<list_path_1_complex_1>>



      6.1.1.1.5  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

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



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

The binary loop is composed of 240 FP arithmetical operations:
 - 144: addition or subtraction (64 inside FMA instructions)
 - 96: multiply (64 inside FMA instructions)
The binary loop is loading 1680 bytes (210 double precision FP elements).
The binary loop is storing 128 bytes (16 double precision FP elements).


      6.1.1.1.7  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

Arithmetic intensity is 0.13 FP operations per loaded or stored byte.







      6.1.2  -  Loop 24 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/build/TeaLeaf/src/omp/cg.cpp:108-113.

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

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

50% of peak computational performance is used (12.00 out of 24.00 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
  ----------------------------------------------------------------------------------------------------------

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




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

Detected 96 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: 8 occurrences<<list_path_1_complex_1>>



      6.1.2.1.5  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

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



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

The binary loop is composed of 192 FP arithmetical operations:
 - 96: addition or subtraction (all inside FMA instructions)
 - 96: multiply (all inside FMA instructions)
The binary loop is loading 1032 bytes (129 double precision FP elements).
The binary loop is storing 512 bytes (64 double precision FP elements).


      6.1.2.1.7  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

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







      6.1.3  -  Loop 27 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/build/TeaLeaf/src/omp/cg.cpp:128-131.

It is main loop of related source loop which is unrolled by 8 (including vectorization).

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

40% of peak computational performance is used (9.60 out of 24.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.3.1.1  -  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).



      6.1.3.1.2  -  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 13.33 to 8.00 cycles (1.67x speedup).


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





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

Detected 64 FMA (fused multiply-add) operations.




      6.1.3.1.4  -  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: 8 occurrences<<list_path_1_complex_1>>



      6.1.3.1.5  -  Slow data structures access
  ----------------------------------------------------------------------------------------------------------

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

Details
 - Constant non-unit stride: 2 occurrence(s)
Non-unit stride (uncontiguous) accesses are not efficiently using data caches


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



      6.1.3.1.6  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

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



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

The binary loop is composed of 128 FP arithmetical operations:
 - 64: addition or subtraction (all inside FMA instructions)
 - 64: multiply (all inside FMA instructions)
The binary loop is loading 1024 bytes (128 double precision FP elements).
The binary loop is storing 512 bytes (64 double precision FP elements).


      6.1.3.1.8  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

Arithmetic intensity is 0.08 FP operations per loaded or stored byte.







      6.1.4  -  Loop 22 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/build/TeaLeaf/src/omp/cg.cpp:105-113.

Analyzed code is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/build/TeaLeaf/src/omp/cg.cpp:105,108-113.

Warnings:
 - Non-innermost loop: analyzing only self part (ignoring child loops).
 - Ignoring paths for analysis
 - Too many paths. If you really need to analyze all of the 93 paths individually, rerun with max-paths=93
 - 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 93 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)


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

Warnings:
The number of fused uops of the instruction [CLTQ] is unknown
21% of peak computational performance is used (5.17 out of 24.00 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 46.00 to 32.00 cycles (1.44x 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 poorly vectorized.
Only 26% 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 46.00 to 27.19 cycles (1.69x speedup).

Details
23% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):
 - 25% of SSE/AVX loads are used in vector version.
 - 26% of SSE/AVX stores are used in vector version.
 - 41% of SSE/AVX addition or subtraction instructions are used in vector version.
 - 26% of SSE/AVX fused multiply-add instructions are used in vector version.
 - 9% 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
 - 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
  ----------------------------------------------------------------------------------------------------------

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




      6.1.4.1.4  -  FMA
  ----------------------------------------------------------------------------------------------------------

Detected 110 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: 1.00x.
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).




      6.1.4.1.5  -  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.
 - LEA: 5 occurrences<<list_path_1_complex_1>>
 - VMOVUPD: 6 occurrences<<list_path_1_complex_2>>



      6.1.4.1.6  -  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>>


      6.1.4.1.7  -  Conversion instructions
  ----------------------------------------------------------------------------------------------------------

Detected expensive conversion instructions.

Details
 - CLTQ: 2 occurrences<<list_path_1_cvt_1>>


Workaround
Avoid mixing data with different types. In particular, check if the type of constants is the same as array elements.


      6.1.4.1.8  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

32 SSE or AVX instructions are processing arithmetic or math operations on double precision FP elements in scalar mode (one at a time).
4 SSE or AVX instructions are processing arithmetic or math operations on double precision FP elements in vector mode (two at a time).
4 AVX instructions are processing arithmetic or math operations on double precision FP elements in vector mode (four at a time).
9 AVX-512 instructions are processing arithmetic or math operations on double precision FP elements in vector mode (eight at a time).



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

The binary loop is composed of 238 FP arithmetical operations:
 - 124: addition or subtraction (110 inside FMA instructions)
 - 114: multiply (110 inside FMA instructions)
The binary loop is loading 1305 bytes (163 double precision FP elements).
The binary loop is storing 625 bytes (78 double precision FP elements).


      6.1.4.1.10  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

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







      6.1.5  -  Loop 26 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/build/TeaLeaf/src/omp/cg.cpp:125-131.

Analyzed code is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/build/TeaLeaf/src/omp/cg.cpp:125,128-131.

Warnings:
 - Non-innermost loop: analyzing only self part (ignoring child loops).
 - Ignoring paths for analysis
 - Too many paths. If you really need to analyze all of the 154 paths individually, rerun with max-paths=154
 - 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 154 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
  ----------------------------------------------------------------------------------------------------------

Warnings:
The number of fused uops of the instruction [CLTQ] is unknown
20% of peak computational performance is used (4.83 out of 24.00 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 29.00 to 22.67 cycles (1.28x 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 poorly vectorized.
Only 35% 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 29.00 to 20.24 cycles (1.43x speedup).

Details
28% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):
 - 41% of SSE/AVX loads are used in vector version.
 - 44% of SSE/AVX stores are used in vector version.
 - 0% of SSE/AVX addition or subtraction instructions are used in vector version.
 - 44% of SSE/AVX fused multiply-add instructions 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
 - 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)



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

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




      6.1.5.1.4  -  FMA
  ----------------------------------------------------------------------------------------------------------

Detected 70 FMA (fused multiply-add) operations.




      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.
 - VMOVUPD: 7 occurrences<<list_path_1_complex_1>>



      6.1.5.1.6  -  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>>


      6.1.5.1.7  -  Conversion instructions
  ----------------------------------------------------------------------------------------------------------

Detected expensive conversion instructions.

Details
 - CLTQ: 2 occurrences<<list_path_1_cvt_1>>


Workaround
Avoid mixing data with different types. In particular, check if the type of constants is the same as array elements.


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

10 SSE or AVX instructions are processing arithmetic or math operations on double precision FP elements in scalar mode (one at a time).
1 AVX instructions are processing arithmetic or math operations on double precision FP elements in vector mode (four at a time).
7 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 is composed of 140 FP arithmetical operations:
 - 70: addition or subtraction (all inside FMA instructions)
 - 70: multiply (all inside FMA instructions)
The binary loop is loading 1160 bytes (145 double precision FP elements).
The binary loop is storing 560 bytes (70 double precision FP elements).


      6.1.5.1.10  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

Arithmetic intensity is 0.08 FP operations per loaded or stored byte.







      6.1.6  -  Loop 18 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/build/TeaLeaf/src/omp/cg.cpp:83-90.

Warnings:
 - Non-innermost loop: analyzing only self part (ignoring child loops).
 - Ignoring paths for analysis
 - Too many paths. If you really need to analyze all of the 45 paths individually, rerun with max-paths=45
 - 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 45 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)


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

Warnings:
The number of fused uops of the instruction [CLTQ] is unknown
19% of peak computational performance is used (4.68 out of 24.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.6.1.1  -  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 59.83 to 35.67 cycles (1.68x speedup).

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



      6.1.6.1.2  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is poorly vectorized.
Only 23% 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 59.83 to 37.99 cycles (1.57x speedup).

Details
21% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):
 - 26% of SSE/AVX loads are used in vector version.
 - 13% of SSE/AVX stores are used in vector version.
 - 31% of SSE/AVX addition or subtraction instructions are used in vector version.
 - 29% of SSE/AVX multiply instructions are used in vector version.
 - 22% of SSE/AVX fused multiply-add instructions are used in vector version.
 - 7% 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
 - 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)



      6.1.6.1.3  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

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




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

Detected 68 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: 1.04x.
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).




      6.1.6.1.5  -  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.
 - ADD: 1 occurrences<<list_path_1_complex_1>>
 - INC: 1 occurrences<<list_path_1_complex_2>>
 - LEA: 11 occurrences<<list_path_1_complex_3>>
 - VMOVUPD: 1 occurrences<<list_path_1_complex_4>>



      6.1.6.1.6  -  Vector unaligned load/store instructions
  ----------------------------------------------------------------------------------------------------------

Detected 5 optimal vector unaligned load/store instructions.


Details
 - VMOVUPD: 5 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>>


      6.1.6.1.7  -  Conversion instructions
  ----------------------------------------------------------------------------------------------------------

Detected expensive conversion instructions.

Details
 - CLTQ: 1 occurrences<<list_path_1_cvt_1>>


Workaround
Avoid mixing data with different types. In particular, check if the type of constants is the same as array elements.


      6.1.6.1.8  -  Type of elements and instruction set
  ----------------------------------------------------------------------------------------------------------

68 SSE or AVX instructions are processing arithmetic or math operations on double precision FP elements in scalar mode (one at a time).
4 SSE or AVX instructions are processing arithmetic or math operations on double precision FP elements in vector mode (two at a time).
12 AVX instructions are processing arithmetic or math operations on double precision FP elements in vector mode (four at a time).
11 AVX-512 instructions are processing arithmetic or math operations on double precision FP elements in vector mode (eight at a time).



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

The binary loop is composed of 280 FP arithmetical operations:
 - 172: addition or subtraction (68 inside FMA instructions)
 - 108: multiply (68 inside FMA instructions)
The binary loop is loading 1853 bytes (231 double precision FP elements).
The binary loop is storing 192 bytes (24 double precision FP elements).


      6.1.6.1.10  -  Arithmetic intensity
  ----------------------------------------------------------------------------------------------------------

Arithmetic intensity is 0.14 FP operations per loaded or stored byte.







      6.1.7  -  Loop 93 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/build/TeaLeaf/src/omp/pack_halos.cpp:51-53.

It is main loop of related source loop which is unrolled by 8 (including vectorization).

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

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

      6.1.7.1.1  -  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).



      6.1.7.1.2  -  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)


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



      6.1.7.1.3  -  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: 8 occurrences<<list_path_1_complex_1>>



      6.1.7.1.4  -  Slow data structures access
  ----------------------------------------------------------------------------------------------------------

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

Details
 - Constant non-unit stride: 2 occurrence(s)
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)



      6.1.7.1.5  -  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.


      6.1.7.1.6  -  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 512 bytes.
The binary loop is storing 512 bytes.







      6.1.8  -  Loop 96 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/build/TeaLeaf/src/omp/pack_halos.cpp:69-71.

It is main loop of related source loop which is unrolled by 8 (including vectorization).

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

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

      6.1.8.1.1  -  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).



      6.1.8.1.2  -  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)


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



      6.1.8.1.3  -  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: 8 occurrences<<list_path_1_complex_1>>



      6.1.8.1.4  -  Slow data structures access
  ----------------------------------------------------------------------------------------------------------

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

Details
 - Constant non-unit stride: 2 occurrence(s)
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)



      6.1.8.1.5  -  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.


      6.1.8.1.6  -  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 512 bytes.
The binary loop is storing 512 bytes.







      6.1.9  -  Loop 98 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/build/TeaLeaf/src/omp/pack_halos.cpp:83-88.

Analyzed code is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/build/TeaLeaf/src/omp/pack_halos.cpp:83,86-88.

Warnings:
 - Non-innermost loop: analyzing only self part (ignoring child loops).
 - Ignoring paths for analysis
 - Too many paths. If you really need to analyze all of the 154 paths individually, rerun with max-paths=154
 - 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 154 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)


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

Warnings:
The number of fused uops of the instruction [CLTQ] is unknown
0% of peak computational performance is used (0.00 out of 48.00 FLOP per cycle (GFLOPS @ 1GHz))

      6.1.9.1.1  -  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 27.67 to 14.67 cycles (1.89x speedup).

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



      6.1.9.1.2  -  Vectorization
  ----------------------------------------------------------------------------------------------------------

Your loop is poorly vectorized.
Only 31% 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 27.67 to 20.58 cycles (1.34x speedup).

Details
25% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):
 - 33% of SSE/AVX loads are used in vector version.
 - 44% of SSE/AVX stores are used in vector version.
 - 0% of SSE/AVX addition or subtraction instructions 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
 - 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)



      6.1.9.1.3  -  Execution units bottlenecks
  ----------------------------------------------------------------------------------------------------------

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



No data for this section



      6.1.9.1.4  -  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: 7 occurrences<<list_path_1_complex_1>>



      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  -  Conversion instructions
  ----------------------------------------------------------------------------------------------------------

Detected expensive conversion instructions.

Details
 - CLTQ: 1 occurrences<<list_path_1_cvt_1>>


Workaround
Avoid mixing data with different types. In particular, check if the type of constants is the same as array elements.


      6.1.9.1.7  -  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.


      6.1.9.1.8  -  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 636 bytes.
The binary loop is storing 560 bytes.







      6.1.10  -  Loop 101 from exec
  ==========================================================================================================

The loop is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/build/TeaLeaf/src/omp/pack_halos.cpp:100-105.

Analyzed code is defined in /beegfs/hackathon/users/eoseret/qaas_runs_test/179-075-5378/intel/TeaLeaf/build/TeaLeaf/src/omp/pack_halos.cpp:100,103-105.

Warnings:
 - Non-innermost loop: analyzing only self part (ignoring child loops).
 - Ignoring paths for analysis
 - Too many paths. If you really need to analyze all of the 154 paths individually, rerun with max-paths=154
 - 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 154 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)


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

Warnings:
The number of fused uops of the instruction [CLTQ] 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 27.67 to 14.67 cycles (1.89x 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 poorly vectorized.
Only 32% 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 27.67 to 20.58 cycles (1.34x speedup).

Details
25% of SSE/AVX instructions are used in vector version (process two or more data elements in vector registers):
 - 32% of SSE/AVX loads are used in vector version.
 - 44% of SSE/AVX stores are used in vector version.
 - 0% of SSE/AVX addition or subtraction instructions 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
 - 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)



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

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



No data for this section



      6.1.10.1.4  -  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: 7 occurrences<<list_path_1_complex_1>>



      6.1.10.1.5  -  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>>


      6.1.10.1.6  -  Conversion instructions
  ----------------------------------------------------------------------------------------------------------

Detected expensive conversion instructions.

Details
 - CLTQ: 1 occurrences<<list_path_1_cvt_1>>


Workaround
Avoid mixing data with different types. In particular, check if the type of constants is the same as array elements.


      6.1.10.1.7  -  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.


      6.1.10.1.8  -  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 644 bytes.
The binary loop is storing 560 bytes.





[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/179-075-5378/intel/TeaLeaf/run/oneview_runs/compilers/gcc_2/oneview_run_1790757542"
[MAQAO] Info: 
