Help is available by moving the cursor above any 
symbol or by checking MAQAO website.
[ 4 / 4 ] Application profile is long enough (307.06 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 compiled with -g and -fno-omit-frame-pointer
-g option gives access to debugging informations, such are source locations. -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.
[ 3 / 3 ] Architecture specific option -x Host is used
[ 2 / 2 ] Application is correctly profiled ("Others" category represents 14.04 % 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.
[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (70.00%)
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 99.45% of observed threads are actually active
[ 4 / 4 ] CPU activity is good
CPU cores are active 99.45% of time
[ 4 / 4 ] Loop profile is not flat
At least one loop coverage is greater than 4% (66.50%), representing an hotspot for the application
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (69.44%)
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.
[ 2 / 4 ] Affinity stability is lower than 90% (69.52%)
Threads are often migrating to other CPU cores/threads. For OpenMP, typically set (OMP_PLACES=cores OMP_PROC_BIND=close) or (OMP_PLACES=threads OMP_PROC_BIND=spread). With OpenMPI + OpenMP, use --bind-to core --map-by node:PE=$OMP_NUM_THREADS --report-bindings. With IntelMPI + OpenMP, set I_MPI_PIN_DOMAIN=omp:compact or I_MPI_PIN_DOMAIN=omp:scatter and use -print-rank-map.
[ 3 / 3 ] Less than 10% (0.00%) is spend in BLAS1 operations
It could be more efficient to inline by hand BLAS1 operations
[ 3 / 3 ] Functions mostly use all threads
Functions running on a reduced number of threads (typically sequential code) cover less than 10% of application walltime (0.00%)
[ 3 / 3 ] Cumulative Outermost/In between loops coverage (0.56%) lower than cumulative innermost loop coverage (69.44%)
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)
| Loop ID | Analysis | Penalty Score |
|---|---|---|
| ►Loop 29 - kmeans-icpx-Ofast | Execution Time: 66 % - Vectorization Ratio: 57.89 % - Vector Length Use: 18.86 % | |
| ►Loop Computation Issues | 4 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ►Control Flow Issues | 260 | |
| ○ | [SA] Too many paths (256 paths) - Simplify control structure. There are 256 issues ( = paths) costing 1 point each with a malus of 4 points. | 260 |
| ►Data Access Issues | 8 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 8 issues (= instructions) costing 1 point each. | 8 |
| ►Vectorization Roadblocks | 260 | |
| ○ | [SA] Too many paths (256 paths) - Simplify control structure. There are 256 issues ( = paths) costing 1 point each with a malus of 4 points. | 260 |
| ►Inefficient Vectorization | 8 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 8 issues (= instructions) costing 1 point each. | 8 |
| ►Loop 22 - kmeans-icpx-Ofast | Execution Time: 2 % - Vectorization Ratio: 0.00 % - Vector Length Use: 11.61 % | |
| ►Loop Computation Issues | 6 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ►Data Access Issues | 12 | |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 3 issues ( = indirect data accesses) costing 4 point each. | 12 |
| ►Vectorization Roadblocks | 12 | |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 3 issues ( = indirect data accesses) costing 4 point each. | 12 |
| ►Loop 28 - kmeans-icpx-Ofast | Execution Time: 0 % - Vectorization Ratio: 94.12 % - Vector Length Use: 23.90 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 262 | |
| ○ | [SA] Too many paths (256 paths) - Simplify control structure. There are 256 issues ( = paths) costing 1 point each with a malus of 4 points. | 260 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Vectorization Roadblocks | 262 | |
| ○ | [SA] Too many paths (256 paths) - Simplify control structure. There are 256 issues ( = paths) costing 1 point each with a malus of 4 points. | 260 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 0 - kmeans-icpx-Ofast | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 11.28 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 1 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ►Data Access Issues | 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 1 issues ( = data accesses) costing 2 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. | 2 |
| ►Vectorization Roadblocks | 3 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [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 1 issues ( = data accesses) costing 2 point each. | 2 |
[ 4 / 4 ] Application profile is long enough (270.68 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 compiled with -g and -fno-omit-frame-pointer
-g option gives access to debugging informations, such are source locations. -fno-omit-frame-pointer improve 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.
[ 3 / 3 ] Architecture specific option -x Host is used
[ 2 / 2 ] Application is correctly profiled ("Others" category represents 9.65 % 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.
[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (43.54%)
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 164.88% of observed threads are actually active
[ 4 / 4 ] CPU activity is good
CPU cores are active 99.65% of time
[ 4 / 4 ] Loop profile is not flat
At least one loop coverage is greater than 4% (36.21%), representing an hotspot for the application
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (38.16%)
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.
[ 3 / 4 ] Affinity stability is lower than 90% (79.12%)
Threads are often migrating to other CPU cores/threads. For OpenMP, typically set (OMP_PLACES=cores OMP_PROC_BIND=close) or (OMP_PLACES=threads OMP_PROC_BIND=spread). With OpenMPI + OpenMP, use --bind-to core --map-by node:PE=$OMP_NUM_THREADS --report-bindings. With IntelMPI + OpenMP, set I_MPI_PIN_DOMAIN=omp:compact or I_MPI_PIN_DOMAIN=omp:scatter and use -print-rank-map.
[ 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 (34.06%). 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 (5.38%) lower than cumulative innermost loop coverage (38.16%)
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)
| Loop ID | Analysis | Penalty Score |
|---|---|---|
| ►Loop 27 - kmeans-icpx-Ofast | Execution Time: 36 % - Vectorization Ratio: 57.14 % - Vector Length Use: 18.75 % | |
| ►Loop Computation Issues | 6 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ►Control Flow Issues | 20 | |
| ○ | [SA] Too many paths (16 paths) - Simplify control structure. There are 16 issues ( = paths) costing 1 point each with a malus of 4 points. | 20 |
| ►Data Access Issues | 4 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 4 issues (= instructions) costing 1 point each. | 4 |
| ►Vectorization Roadblocks | 20 | |
| ○ | [SA] Too many paths (16 paths) - Simplify control structure. There are 16 issues ( = paths) costing 1 point each with a malus of 4 points. | 20 |
| ►Inefficient Vectorization | 4 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 4 issues (= instructions) costing 1 point each. | 4 |
| ►Loop 21 - kmeans-icpx-Ofast | Execution Time: 2 % - Vectorization Ratio: 0.00 % - Vector Length Use: 8.68 % | |
| ►Loop Computation Issues | 6 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ►Control Flow Issues | 7 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 18 | |
| ○ | [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. | 4 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 3 issues ( = indirect data accesses) costing 4 point each. | 12 |
| ○ | [SA] More than 20% of the loads are accessing the stack - Perform loop splitting to decrease pressure on registers. This issue costs 2 points. | 2 |
| ►Vectorization Roadblocks | 23 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 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. | 4 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 3 issues ( = indirect data accesses) costing 4 point each. | 12 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 26 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 58.57 % - Vector Length Use: 19.38 % | |
| ►Loop Computation Issues | 11 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Control Flow Issues | 7 | |
| ○ | [SA] Several paths (2 paths) - Simplify control structure or force the compiler to use masked instructions. There are 2 issues ( = paths) costing 1 point each. | 2 |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Data Access Issues | 1 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 1 issues (= instructions) costing 1 point each. | 1 |
| ►Vectorization Roadblocks | 2 | |
| ○ | [SA] Several paths (2 paths) - Simplify control structure or force the compiler to use masked instructions. There are 2 issues ( = paths) costing 1 point each. | 2 |
| ►Inefficient Vectorization | 1 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 1 issues (= instructions) costing 1 point each. | 1 |
| ►Loop 24 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 0.00 % - Vector Length Use: 9.90 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 1003 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Too many paths (13364 paths) - Simplify control structure. There are 13364 issues ( = paths) costing 1 point, limited to 1000. | 1000 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 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. | 2 |
| ►Vectorization Roadblocks | 1003 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Too many paths (13364 paths) - Simplify control structure. There are 13364 issues ( = paths) costing 1 point, limited to 1000. | 1000 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 25 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 42.86 % - Vector Length Use: 16.07 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 57 | |
| ○ | [SA] Too many paths (51 paths) - Simplify control structure. There are 51 issues ( = paths) costing 1 point each with a malus of 4 points. | 55 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Vectorization Roadblocks | 57 | |
| ○ | [SA] Too many paths (51 paths) - Simplify control structure. There are 51 issues ( = paths) costing 1 point each with a malus of 4 points. | 55 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 0 - kmeans-icpx-Ofast | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 11.28 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 1 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ►Data Access Issues | 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 1 issues ( = data accesses) costing 2 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. | 2 |
| ►Vectorization Roadblocks | 3 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [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 1 issues ( = data accesses) costing 2 point each. | 2 |
[ 4 / 4 ] Application profile is long enough (181.82 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 compiled with -g and -fno-omit-frame-pointer
-g option gives access to debugging informations, such are source locations. -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.
[ 3 / 3 ] Architecture specific option -x Host is used
[ 2 / 2 ] Application is correctly profiled ("Others" category represents 9.53 % 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.
[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (43.11%)
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 245.33% of observed threads are actually active
[ 4 / 4 ] CPU activity is good
CPU cores are active 99.65% of time
[ 4 / 4 ] Loop profile is not flat
At least one loop coverage is greater than 4% (35.81%), representing an hotspot for the application
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (37.71%)
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.
[ 3 / 4 ] Affinity stability is lower than 90% (79.17%)
Threads are often migrating to other CPU cores/threads. For OpenMP, typically set (OMP_PLACES=cores OMP_PROC_BIND=close) or (OMP_PLACES=threads OMP_PROC_BIND=spread). With OpenMPI + OpenMP, use --bind-to core --map-by node:PE=$OMP_NUM_THREADS --report-bindings. With IntelMPI + OpenMP, set I_MPI_PIN_DOMAIN=omp:compact or I_MPI_PIN_DOMAIN=omp:scatter and use -print-rank-map.
[ 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 (50.51%). 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 (5.40%) lower than cumulative innermost loop coverage (37.71%)
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)
| Loop ID | Analysis | Penalty Score |
|---|---|---|
| ►Loop 27 - kmeans-icpx-Ofast | Execution Time: 35 % - Vectorization Ratio: 57.14 % - Vector Length Use: 18.75 % | |
| ►Loop Computation Issues | 6 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ►Control Flow Issues | 20 | |
| ○ | [SA] Too many paths (16 paths) - Simplify control structure. There are 16 issues ( = paths) costing 1 point each with a malus of 4 points. | 20 |
| ►Data Access Issues | 4 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 4 issues (= instructions) costing 1 point each. | 4 |
| ►Vectorization Roadblocks | 20 | |
| ○ | [SA] Too many paths (16 paths) - Simplify control structure. There are 16 issues ( = paths) costing 1 point each with a malus of 4 points. | 20 |
| ►Inefficient Vectorization | 4 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 4 issues (= instructions) costing 1 point each. | 4 |
| ►Loop 21 - kmeans-icpx-Ofast | Execution Time: 2 % - Vectorization Ratio: 0.00 % - Vector Length Use: 8.68 % | |
| ►Loop Computation Issues | 6 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ►Control Flow Issues | 7 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 18 | |
| ○ | [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. | 4 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 3 issues ( = indirect data accesses) costing 4 point each. | 12 |
| ○ | [SA] More than 20% of the loads are accessing the stack - Perform loop splitting to decrease pressure on registers. This issue costs 2 points. | 2 |
| ►Vectorization Roadblocks | 23 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 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. | 4 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 3 issues ( = indirect data accesses) costing 4 point each. | 12 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 26 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 58.57 % - Vector Length Use: 19.38 % | |
| ►Loop Computation Issues | 11 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Control Flow Issues | 7 | |
| ○ | [SA] Several paths (2 paths) - Simplify control structure or force the compiler to use masked instructions. There are 2 issues ( = paths) costing 1 point each. | 2 |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Data Access Issues | 1 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 1 issues (= instructions) costing 1 point each. | 1 |
| ►Vectorization Roadblocks | 2 | |
| ○ | [SA] Several paths (2 paths) - Simplify control structure or force the compiler to use masked instructions. There are 2 issues ( = paths) costing 1 point each. | 2 |
| ►Inefficient Vectorization | 1 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 1 issues (= instructions) costing 1 point each. | 1 |
| ►Loop 24 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 0.00 % - Vector Length Use: 9.90 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 1003 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Too many paths (13364 paths) - Simplify control structure. There are 13364 issues ( = paths) costing 1 point, limited to 1000. | 1000 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 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. | 2 |
| ►Vectorization Roadblocks | 1003 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Too many paths (13364 paths) - Simplify control structure. There are 13364 issues ( = paths) costing 1 point, limited to 1000. | 1000 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 25 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 42.86 % - Vector Length Use: 16.07 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 57 | |
| ○ | [SA] Too many paths (51 paths) - Simplify control structure. There are 51 issues ( = paths) costing 1 point each with a malus of 4 points. | 55 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Vectorization Roadblocks | 57 | |
| ○ | [SA] Too many paths (51 paths) - Simplify control structure. There are 51 issues ( = paths) costing 1 point each with a malus of 4 points. | 55 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 0 - kmeans-icpx-Ofast | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 11.28 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 1 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ►Data Access Issues | 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 1 issues ( = data accesses) costing 2 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. | 2 |
| ►Vectorization Roadblocks | 3 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [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 1 issues ( = data accesses) costing 2 point each. | 2 |
[ 4 / 4 ] Application profile is long enough (136.98 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 compiled with -g and -fno-omit-frame-pointer
-g option gives access to debugging informations, such are source locations. -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.
[ 3 / 3 ] Architecture specific option -x Host is used
[ 2 / 2 ] Application is correctly profiled ("Others" category represents 9.47 % 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.
[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (43.44%)
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 324.24% of observed threads are actually active
[ 4 / 4 ] CPU activity is good
CPU cores are active 99.64% of time
[ 4 / 4 ] Loop profile is not flat
At least one loop coverage is greater than 4% (36.26%), representing an hotspot for the application
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (38.11%)
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.
[ 3 / 4 ] Affinity stability is lower than 90% (79.15%)
Threads are often migrating to other CPU cores/threads. For OpenMP, typically set (OMP_PLACES=cores OMP_PROC_BIND=close) or (OMP_PLACES=threads OMP_PROC_BIND=spread). With OpenMPI + OpenMP, use --bind-to core --map-by node:PE=$OMP_NUM_THREADS --report-bindings. With IntelMPI + OpenMP, set I_MPI_PIN_DOMAIN=omp:compact or I_MPI_PIN_DOMAIN=omp:scatter and use -print-rank-map.
[ 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 (66.72%). 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 (5.33%) lower than cumulative innermost loop coverage (38.11%)
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)
| Loop ID | Analysis | Penalty Score |
|---|---|---|
| ►Loop 27 - kmeans-icpx-Ofast | Execution Time: 36 % - Vectorization Ratio: 57.14 % - Vector Length Use: 18.75 % | |
| ►Loop Computation Issues | 6 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ►Control Flow Issues | 20 | |
| ○ | [SA] Too many paths (16 paths) - Simplify control structure. There are 16 issues ( = paths) costing 1 point each with a malus of 4 points. | 20 |
| ►Data Access Issues | 4 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 4 issues (= instructions) costing 1 point each. | 4 |
| ►Vectorization Roadblocks | 20 | |
| ○ | [SA] Too many paths (16 paths) - Simplify control structure. There are 16 issues ( = paths) costing 1 point each with a malus of 4 points. | 20 |
| ►Inefficient Vectorization | 4 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 4 issues (= instructions) costing 1 point each. | 4 |
| ►Loop 21 - kmeans-icpx-Ofast | Execution Time: 2 % - Vectorization Ratio: 0.00 % - Vector Length Use: 8.68 % | |
| ►Loop Computation Issues | 6 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ►Control Flow Issues | 7 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 18 | |
| ○ | [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. | 4 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 3 issues ( = indirect data accesses) costing 4 point each. | 12 |
| ○ | [SA] More than 20% of the loads are accessing the stack - Perform loop splitting to decrease pressure on registers. This issue costs 2 points. | 2 |
| ►Vectorization Roadblocks | 23 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 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. | 4 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 3 issues ( = indirect data accesses) costing 4 point each. | 12 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 26 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 58.57 % - Vector Length Use: 19.38 % | |
| ►Loop Computation Issues | 11 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Control Flow Issues | 7 | |
| ○ | [SA] Several paths (2 paths) - Simplify control structure or force the compiler to use masked instructions. There are 2 issues ( = paths) costing 1 point each. | 2 |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Data Access Issues | 1 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 1 issues (= instructions) costing 1 point each. | 1 |
| ►Vectorization Roadblocks | 2 | |
| ○ | [SA] Several paths (2 paths) - Simplify control structure or force the compiler to use masked instructions. There are 2 issues ( = paths) costing 1 point each. | 2 |
| ►Inefficient Vectorization | 1 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 1 issues (= instructions) costing 1 point each. | 1 |
| ►Loop 24 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 0.00 % - Vector Length Use: 9.90 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 1003 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Too many paths (13364 paths) - Simplify control structure. There are 13364 issues ( = paths) costing 1 point, limited to 1000. | 1000 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 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. | 2 |
| ►Vectorization Roadblocks | 1003 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Too many paths (13364 paths) - Simplify control structure. There are 13364 issues ( = paths) costing 1 point, limited to 1000. | 1000 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 25 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 42.86 % - Vector Length Use: 16.07 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 57 | |
| ○ | [SA] Too many paths (51 paths) - Simplify control structure. There are 51 issues ( = paths) costing 1 point each with a malus of 4 points. | 55 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Vectorization Roadblocks | 57 | |
| ○ | [SA] Too many paths (51 paths) - Simplify control structure. There are 51 issues ( = paths) costing 1 point each with a malus of 4 points. | 55 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 0 - kmeans-icpx-Ofast | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 11.28 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 1 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ►Data Access Issues | 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 1 issues ( = data accesses) costing 2 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. | 2 |
| ►Vectorization Roadblocks | 3 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [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 1 issues ( = data accesses) costing 2 point each. | 2 |
[ 4 / 4 ] Application profile is long enough (114.77 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 compiled with -g and -fno-omit-frame-pointer
-g option gives access to debugging informations, such are source locations. -fno-omit-frame-pointer improve 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.
[ 3 / 3 ] Architecture specific option -x Host is used
[ 2 / 2 ] Application is correctly profiled ("Others" category represents 9.42 % 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.
[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (43.15%)
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 386.76% of observed threads are actually active
[ 4 / 4 ] CPU activity is good
CPU cores are active 99.64% of time
[ 4 / 4 ] Loop profile is not flat
At least one loop coverage is greater than 4% (36.01%), representing an hotspot for the application
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (37.90%)
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.
[ 3 / 4 ] Affinity stability is lower than 90% (79.15%)
Threads are often migrating to other CPU cores/threads. For OpenMP, typically set (OMP_PLACES=cores OMP_PROC_BIND=close) or (OMP_PLACES=threads OMP_PROC_BIND=spread). With OpenMPI + OpenMP, use --bind-to core --map-by node:PE=$OMP_NUM_THREADS --report-bindings. With IntelMPI + OpenMP, set I_MPI_PIN_DOMAIN=omp:compact or I_MPI_PIN_DOMAIN=omp:scatter and use -print-rank-map.
[ 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 (79.55%). 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 (5.25%) lower than cumulative innermost loop coverage (37.90%)
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)
| Loop ID | Analysis | Penalty Score |
|---|---|---|
| ►Loop 27 - kmeans-icpx-Ofast | Execution Time: 36 % - Vectorization Ratio: 57.14 % - Vector Length Use: 18.75 % | |
| ►Loop Computation Issues | 6 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ►Control Flow Issues | 20 | |
| ○ | [SA] Too many paths (16 paths) - Simplify control structure. There are 16 issues ( = paths) costing 1 point each with a malus of 4 points. | 20 |
| ►Data Access Issues | 4 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 4 issues (= instructions) costing 1 point each. | 4 |
| ►Vectorization Roadblocks | 20 | |
| ○ | [SA] Too many paths (16 paths) - Simplify control structure. There are 16 issues ( = paths) costing 1 point each with a malus of 4 points. | 20 |
| ►Inefficient Vectorization | 4 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 4 issues (= instructions) costing 1 point each. | 4 |
| ►Loop 21 - kmeans-icpx-Ofast | Execution Time: 2 % - Vectorization Ratio: 0.00 % - Vector Length Use: 8.68 % | |
| ►Loop Computation Issues | 6 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ►Control Flow Issues | 7 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 18 | |
| ○ | [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. | 4 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 3 issues ( = indirect data accesses) costing 4 point each. | 12 |
| ○ | [SA] More than 20% of the loads are accessing the stack - Perform loop splitting to decrease pressure on registers. This issue costs 2 points. | 2 |
| ►Vectorization Roadblocks | 23 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 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. | 4 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 3 issues ( = indirect data accesses) costing 4 point each. | 12 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 26 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 58.57 % - Vector Length Use: 19.38 % | |
| ►Loop Computation Issues | 11 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Control Flow Issues | 7 | |
| ○ | [SA] Several paths (2 paths) - Simplify control structure or force the compiler to use masked instructions. There are 2 issues ( = paths) costing 1 point each. | 2 |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Data Access Issues | 1 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 1 issues (= instructions) costing 1 point each. | 1 |
| ►Vectorization Roadblocks | 2 | |
| ○ | [SA] Several paths (2 paths) - Simplify control structure or force the compiler to use masked instructions. There are 2 issues ( = paths) costing 1 point each. | 2 |
| ►Inefficient Vectorization | 1 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 1 issues (= instructions) costing 1 point each. | 1 |
| ►Loop 24 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 0.00 % - Vector Length Use: 9.90 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 1003 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Too many paths (13364 paths) - Simplify control structure. There are 13364 issues ( = paths) costing 1 point, limited to 1000. | 1000 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 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. | 2 |
| ►Vectorization Roadblocks | 1003 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Too many paths (13364 paths) - Simplify control structure. There are 13364 issues ( = paths) costing 1 point, limited to 1000. | 1000 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 25 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 42.86 % - Vector Length Use: 16.07 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 57 | |
| ○ | [SA] Too many paths (51 paths) - Simplify control structure. There are 51 issues ( = paths) costing 1 point each with a malus of 4 points. | 55 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Vectorization Roadblocks | 57 | |
| ○ | [SA] Too many paths (51 paths) - Simplify control structure. There are 51 issues ( = paths) costing 1 point each with a malus of 4 points. | 55 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 0 - kmeans-icpx-Ofast | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 11.28 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 1 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ►Data Access Issues | 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 1 issues ( = data accesses) costing 2 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. | 2 |
| ►Vectorization Roadblocks | 3 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [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 1 issues ( = data accesses) costing 2 point each. | 2 |
[ 4 / 4 ] Application profile is long enough (106.22 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 compiled with -g and -fno-omit-frame-pointer
-g option gives access to debugging informations, such are source locations. -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.
[ 3 / 3 ] Architecture specific option -x Host is used
[ 2 / 2 ] Application is correctly profiled ("Others" category represents 9.48 % 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.
[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (43.40%)
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 418.22% of observed threads are actually active
[ 4 / 4 ] CPU activity is good
CPU cores are active 99.62% of time
[ 4 / 4 ] Loop profile is not flat
At least one loop coverage is greater than 4% (36.00%), representing an hotspot for the application
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (37.91%)
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.
[ 3 / 4 ] Affinity stability is lower than 90% (79.13%)
Threads are often migrating to other CPU cores/threads. For OpenMP, typically set (OMP_PLACES=cores OMP_PROC_BIND=close) or (OMP_PLACES=threads OMP_PROC_BIND=spread). With OpenMPI + OpenMP, use --bind-to core --map-by node:PE=$OMP_NUM_THREADS --report-bindings. With IntelMPI + OpenMP, set I_MPI_PIN_DOMAIN=omp:compact or I_MPI_PIN_DOMAIN=omp:scatter and use -print-rank-map.
[ 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 (85.90%). 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 (5.49%) lower than cumulative innermost loop coverage (37.91%)
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)
| Loop ID | Analysis | Penalty Score |
|---|---|---|
| ►Loop 27 - kmeans-icpx-Ofast | Execution Time: 36 % - Vectorization Ratio: 57.14 % - Vector Length Use: 18.75 % | |
| ►Loop Computation Issues | 6 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ►Control Flow Issues | 20 | |
| ○ | [SA] Too many paths (16 paths) - Simplify control structure. There are 16 issues ( = paths) costing 1 point each with a malus of 4 points. | 20 |
| ►Data Access Issues | 4 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 4 issues (= instructions) costing 1 point each. | 4 |
| ►Vectorization Roadblocks | 20 | |
| ○ | [SA] Too many paths (16 paths) - Simplify control structure. There are 16 issues ( = paths) costing 1 point each with a malus of 4 points. | 20 |
| ►Inefficient Vectorization | 4 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 4 issues (= instructions) costing 1 point each. | 4 |
| ►Loop 21 - kmeans-icpx-Ofast | Execution Time: 2 % - Vectorization Ratio: 0.00 % - Vector Length Use: 8.68 % | |
| ►Loop Computation Issues | 6 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ►Control Flow Issues | 7 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 18 | |
| ○ | [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. | 4 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 3 issues ( = indirect data accesses) costing 4 point each. | 12 |
| ○ | [SA] More than 20% of the loads are accessing the stack - Perform loop splitting to decrease pressure on registers. This issue costs 2 points. | 2 |
| ►Vectorization Roadblocks | 23 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 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. | 4 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 3 issues ( = indirect data accesses) costing 4 point each. | 12 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 26 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 58.57 % - Vector Length Use: 19.38 % | |
| ►Loop Computation Issues | 11 | |
| ○ | [SA] Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA - Reorganize arithmetic expressions to exhibit potential for FMA. This issue costs 4 points. | 4 |
| ○ | [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. | 2 |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Control Flow Issues | 7 | |
| ○ | [SA] Several paths (2 paths) - Simplify control structure or force the compiler to use masked instructions. There are 2 issues ( = paths) costing 1 point each. | 2 |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Data Access Issues | 1 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 1 issues (= instructions) costing 1 point each. | 1 |
| ►Vectorization Roadblocks | 2 | |
| ○ | [SA] Several paths (2 paths) - Simplify control structure or force the compiler to use masked instructions. There are 2 issues ( = paths) costing 1 point each. | 2 |
| ►Inefficient Vectorization | 1 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 1 issues (= instructions) costing 1 point each. | 1 |
| ►Loop 24 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 0.00 % - Vector Length Use: 9.90 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 1003 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Too many paths (13364 paths) - Simplify control structure. There are 13364 issues ( = paths) costing 1 point, limited to 1000. | 1000 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 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. | 2 |
| ►Vectorization Roadblocks | 1003 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [SA] Too many paths (13364 paths) - Simplify control structure. There are 13364 issues ( = paths) costing 1 point, limited to 1000. | 1000 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 25 - kmeans-icpx-Ofast | Execution Time: 1 % - Vectorization Ratio: 42.86 % - Vector Length Use: 16.07 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 57 | |
| ○ | [SA] Too many paths (51 paths) - Simplify control structure. There are 51 issues ( = paths) costing 1 point each with a malus of 4 points. | 55 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Vectorization Roadblocks | 57 | |
| ○ | [SA] Too many paths (51 paths) - Simplify control structure. There are 51 issues ( = paths) costing 1 point each with a malus of 4 points. | 55 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 0 - kmeans-icpx-Ofast | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 11.28 % | |
| ►Loop Computation Issues | 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. | 2 |
| ►Control Flow Issues | 1 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ►Data Access Issues | 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 1 issues ( = data accesses) costing 2 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. | 2 |
| ►Vectorization Roadblocks | 3 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 1 issues (= calls) costing 1 point each. | 1 |
| ○ | [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 1 issues ( = data accesses) costing 2 point each. | 2 |