Help is available by moving the cursor above any 
symbol or by checking MAQAO website.
[ 4 / 4 ] Application profile is long enough (591.50 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.
[ 0 / 3 ] Architecture specific options are unknown for some functions (compilation options info are not available)
Architecture specific options are needed to produce efficient code for a specific processor ( -mcpu=native ).
[ 0 / 3 ] Most of time spent in analyzed modules comes from functions without compilation options informations
Functions without compilation options information cumulate 61.47% of the time spent in analyzed modules. For engine_linuxa64_nvidia_ompi, NVHPC cannot record compilation options.
[ 0 / 3 ] Optimization level is unknown for some functions (compilation options info are not available)
To have better performances, it is advised to help the compiler by using a proper optimization level (-O2 of higher). Warning, depending on compilers, faster optimization levels can decrease numeric accuracy.
[ 2 / 3 ] Security settings from the host restrict profiling. Some metrics will be missing or incomplete.
Current value for kernel.perf_event_paranoid is 2. If possible, set it to 1 or check with your system administrator which flag can be used to achieve this.
[ 2 / 2 ] Application is correctly profiled ("Others" category represents 2.00 % 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 (56.43%)
If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.
[ 4 / 4 ] Threads activity is good
On average, more than 95.16% of observed threads are actually active
[ 4 / 4 ] CPU activity is good
CPU cores are active 94.84% of time
[ 4 / 4 ] Loop profile is not flat
No hotspot found in the application (greatest loop coverage is 2.39%), but the twenty hottest loops cumulated coverage is representative enough (26.95% > 20%)
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (51.62%)
If the time spent in analyzed innermost loops is less than 15%, standard innermost loop optimizations such as vectorisation will have a limited impact on application performances.
[ 4 / 4 ] Affinity is good (98.05%)
Threads are not migrating to CPU cores: probably successfully pinned
[ 3 / 3 ] Less than 10% (0.00%) is spend in BLAS1 operations
It could be more efficient to inline by hand BLAS1 operations
[ 3 / 3 ] Functions mostly use all threads
Functions running on a reduced number of threads (typically sequential code) cover less than 10% of application walltime (0.00%)
[ 3 / 3 ] Cumulative Outermost/In between loops coverage (4.81%) lower than cumulative innermost loop coverage (51.62%)
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 11554 - engine_linuxa64_nvidia_ompi | Execution Time: 2 % - Vectorization Ratio: 2.73 % - Vector Length Use: 51.37 % | |
| ►Loop Computation Issues | 10 | |
| ○ | [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] Large loop body: over microp cache size - Perform loop splitting or reduce unrolling. There are 1 issues (= chunks of 50 instructions) costing 2 point each. | 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 |
| ○ | [SA] Bottleneck in the front end - If loop size is very small (rare occurrences), perform unroll and jam. If loop size is large, perform loop splitting. This issue costs 2 points. | 2 |
| ►Control Flow Issues | 16 | |
| ○ | [SA] Too many paths (12 paths) - Simplify control structure. There are 12 issues ( = paths) costing 1 point each with a malus of 4 points. | 16 |
| ►Vectorization Roadblocks | 16 | |
| ○ | [SA] Too many paths (12 paths) - Simplify control structure. There are 12 issues ( = paths) costing 1 point each with a malus of 4 points. | 16 |
| ►Loop 11293 - engine_linuxa64_nvidia_ompi | Execution Time: 2 % - Vectorization Ratio: 0.23 % - Vector Length Use: 50.11 % | |
| ►Loop Computation Issues | 10 | |
| ○ | [SA] Presence of expensive FP instructions - Perform hoisting, change algorithm, use SVML or proper numerical library or perform value profiling (count the number of distinct input values). There are 1 issues (= instructions) costing 4 points each. | 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 |
| ○ | [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 | 0 | |
| ►Vectorization Roadblocks | 1000 | |
| ○ | [SA] Too many paths (at least 1000 paths) - Simplify control structure. There are at least 1000 issues ( = paths) costing 1 point. | 1000 |
| ►Loop 9831 - engine_linuxa64_nvidia_ompi | Execution Time: 1 % - Vectorization Ratio: 24.24 % - Vector Length Use: 64.14 % | |
| ►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 | 78 | |
| ○ | [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 29 issues ( = data accesses) costing 2 point each. | 58 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 5 issues ( = indirect data accesses) costing 4 point each. | 20 |
| ►Vectorization Roadblocks | 78 | |
| ○ | [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 29 issues ( = data accesses) costing 2 point each. | 58 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 5 issues ( = indirect data accesses) costing 4 point each. | 20 |
| ►Loop 32132 - engine_linuxa64_nvidia_ompi | Execution Time: 1 % - Vectorization Ratio: 0.00 % - Vector Length Use: 47.46 % | |
| ►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 | 0 | |
| ►Vectorization Roadblocks | 1000 | |
| ○ | [SA] Too many paths (at least 1000 paths) - Simplify control structure. There are at least 1000 issues ( = paths) costing 1 point. | 1000 |
| ►Loop 23938 - engine_linuxa64_nvidia_ompi | Execution Time: 1 % - Vectorization Ratio: 0.00 % - Vector Length Use: 49.96 % | |
| ►Loop Computation Issues | 54 | |
| ○ | [SA] Presence of expensive FP instructions - Perform hoisting, change algorithm, use SVML or proper numerical library or perform value profiling (count the number of distinct input values). There are 12 issues (= instructions) costing 4 points each. | 48 |
| ○ | [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 | 0 | |
| ►Vectorization Roadblocks | 1000 | |
| ○ | [SA] Too many paths (at least 1000 paths) - Simplify control structure. There are at least 1000 issues ( = paths) costing 1 point. | 1000 |
| ►Loop 43510 - engine_linuxa64_nvidia_ompi | Execution Time: 1 % - Vectorization Ratio: 31.25 % - Vector Length Use: 65.63 % | |
| ►Data Access Issues | 12 | |
| ○ | [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 6 issues ( = data accesses) costing 2 point each. | 12 |
| ►Vectorization Roadblocks | 12 | |
| ○ | [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 6 issues ( = data accesses) costing 2 point each. | 12 |
| ►Loop 5037 - engine_linuxa64_nvidia_ompi | Execution Time: 1 % - Vectorization Ratio: 37.50 % - Vector Length Use: 67.71 % | |
| ►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 |
| ►Loop 9471 - engine_linuxa64_nvidia_ompi | Execution Time: 1 % - Vectorization Ratio: 0.00 % - Vector Length Use: 50.00 % | |
| ►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 |
| ►Data Access Issues | 78 | |
| ○ | [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 21 issues ( = data accesses) costing 2 point each. | 42 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 9 issues ( = indirect data accesses) costing 4 point each. | 36 |
| ►Vectorization Roadblocks | 78 | |
| ○ | [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 21 issues ( = data accesses) costing 2 point each. | 42 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 9 issues ( = indirect data accesses) costing 4 point each. | 36 |
| ►Loop 43502 - engine_linuxa64_nvidia_ompi | Execution Time: 1 % - Vectorization Ratio: 41.86 % - Vector Length Use: 67.65 % | |
| ►Loop Computation Issues | 264 | |
| ○ | [SA] Presence of expensive FP instructions - Perform hoisting, change algorithm, use SVML or proper numerical library or perform value profiling (count the number of distinct input values). There are 8 issues (= instructions) costing 4 points each. | 32 |
| ○ | [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] Large loop body: over microp cache size - Perform loop splitting or reduce unrolling. There are 112 issues (= chunks of 50 instructions) costing 2 point each. | 224 |
| ○ | [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] Bottleneck in the front end - If loop size is very small (rare occurrences), perform unroll and jam. If loop size is large, perform loop splitting. This issue costs 2 points. | 2 |
| ►Control Flow Issues | 13 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 11 issues (= calls) costing 1 point each. | 11 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Vectorization Roadblocks | 1013 | |
| ○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 11 issues (= calls) costing 1 point each. | 11 |
| ○ | [SA] Too many paths (at least 1000 paths) - Simplify control structure. There are at least 1000 issues ( = paths) costing 1 point. | 1000 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 43613 - engine_linuxa64_nvidia_ompi | Execution Time: 1 % - Vectorization Ratio: 0.00 % - Vector Length Use: 50.00 % | |
| ►Loop Computation Issues | 14 | |
| ○ | [SA] Presence of expensive FP instructions - Perform hoisting, change algorithm, use SVML or proper numerical library or perform value profiling (count the number of distinct input values). There are 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [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 | 34 | |
| ○ | [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 7 issues ( = data accesses) costing 2 point each. | 14 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 5 issues ( = indirect data accesses) costing 4 point each. | 20 |
| ►Vectorization Roadblocks | 34 | |
| ○ | [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 7 issues ( = data accesses) costing 2 point each. | 14 |
| ○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 5 issues ( = indirect data accesses) costing 4 point each. | 20 |