options

exec - 2024-04-27 02:28:20 - MAQAO 2.20.0

Help is available by moving the cursor above any symbol or by checking MAQAO website.

Stylizer  

[ 4 / 4 ] Application profile is long enough (85.39 s)

To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.

[ 2.02 / 3 ] Some functions are compiled with a low optimization level (O0 or O1)

To have good 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.

[ 2.02 / 3 ] Most of time spent in analyzed modules comes from functions without compilation information

Functions without compilation information (typically not compiled with -g) cumulate 32.79% of the time spent in analyzed modules. Check that -g is present. Remark: if -g is indeed used, this can also be due to some compiler built-in functions (typically math) or statically linked libraries. This warning can be ignored in that case.

[ 2.02 / 3 ] Compilation of some functions is not optimized for the target processor

Architecture specific options are needed to produce efficient code for a specific processor ( -x(target) or -ax(target) ).

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0 % 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

Optimizer

Loop IDAnalysisPenalty Score
Loop 246 - exec+Execution Time: 4 % - Vectorization Ratio: 100.00 % - Vector Length Use: 82.69 %
Loop Computation Issues+66
[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 16 issues (= instructions) costing 4 points each.64
[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+2
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.2
Data Access Issues+218
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 26 issues ( = indirect data accesses) costing 4 point each.104
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 28 issues (= instructions) costing 4 points each.112
[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+106
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.2
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 26 issues ( = indirect data accesses) costing 4 point each.104
Inefficient Vectorization+112
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 28 issues (= instructions) costing 4 points each.112
Loop 248 - exec+Execution Time: 4 % - Vectorization Ratio: 99.12 % - Vector Length Use: 78.65 %
Loop Computation Issues+66
[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 16 issues (= instructions) costing 4 points each.64
[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+2
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.2
Data Access Issues+150
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 17 issues ( = indirect data accesses) costing 4 point each.68
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 20 issues (= instructions) costing 4 points each.80
[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+70
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.2
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 17 issues ( = indirect data accesses) costing 4 point each.68
Inefficient Vectorization+80
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 20 issues (= instructions) costing 4 points each.80
Loop 192 - exec+Execution Time: 3 % - Vectorization Ratio: 97.44 % - Vector Length Use: 86.06 %
Loop Computation Issues+68
[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 16 issues (= instructions) costing 4 points each.64
[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+2
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.2
Data Access Issues+36
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 4 issues ( = indirect data accesses) costing 4 point each.16
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 5 issues (= instructions) costing 4 points each.20
Vectorization Roadblocks+18
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.2
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 4 issues ( = indirect data accesses) costing 4 point each.16
Inefficient Vectorization+20
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 5 issues (= instructions) costing 4 points each.20
Loop 95 - exec+Execution Time: 3 % - Vectorization Ratio: 98.82 % - Vector Length Use: 76.91 %
Loop Computation Issues+34
[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] 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+2
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.2
Data Access Issues+254
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 23 issues ( = indirect data accesses) costing 4 point each.92
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 40 issues (= instructions) costing 4 points each.160
[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+94
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.2
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 23 issues ( = indirect data accesses) costing 4 point each.92
Inefficient Vectorization+160
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 40 issues (= instructions) costing 4 points each.160
Loop 115 - exec+Execution Time: 3 % - Vectorization Ratio: 99.35 % - Vector Length Use: 82.59 %
Loop Computation Issues+64
[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 16 issues (= instructions) costing 4 points each.64
Control Flow Issues+4
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.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
Data Access Issues+56
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 6 issues ( = indirect data accesses) costing 4 point each.24
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 7 issues (= instructions) costing 4 points each.28
[SA] Presence of special instructions executing on a single port (BLEND/MERGE, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 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.2
Vectorization Roadblocks+28
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.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
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 6 issues ( = indirect data accesses) costing 4 point each.24
Inefficient Vectorization+32
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 7 issues (= instructions) costing 4 points each.28
[SA] Presence of special instructions executing on a single port (BLEND/MERGE, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 issues (= instructions) costing 1 point each.2
[SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points.2
Loop 123 - exec+Execution Time: 2 % - Vectorization Ratio: 99.34 % - Vector Length Use: 81.16 %
Loop Computation Issues+64
[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 16 issues (= instructions) costing 4 points each.64
Control Flow Issues+4
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.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
Data Access Issues+64
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 8 issues ( = indirect data accesses) costing 4 point each.32
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 7 issues (= instructions) costing 4 points each.28
[SA] Presence of special instructions executing on a single port (BLEND/MERGE, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 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.2
Vectorization Roadblocks+36
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.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
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 8 issues ( = indirect data accesses) costing 4 point each.32
Inefficient Vectorization+32
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 7 issues (= instructions) costing 4 points each.28
[SA] Presence of special instructions executing on a single port (BLEND/MERGE, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 issues (= instructions) costing 1 point each.2
[SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points.2
Loop 168 - exec+Execution Time: 2 % - Vectorization Ratio: 98.73 % - Vector Length Use: 76.11 %
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+2
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.2
Data Access Issues+94
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 11 issues ( = indirect data accesses) costing 4 point each.44
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 12 issues (= instructions) costing 4 points each.48
[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+46
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.2
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 11 issues ( = indirect data accesses) costing 4 point each.44
Inefficient Vectorization+48
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 12 issues (= instructions) costing 4 points each.48
Loop 150 - exec+Execution Time: 2 % - Vectorization Ratio: 99.24 % - Vector Length Use: 83.30 %
Loop Computation Issues+224
[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 56 issues (= instructions) costing 4 points each.224
Control Flow Issues+2
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.2
Data Access Issues+142
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 17 issues ( = indirect data accesses) costing 4 point each.68
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 18 issues (= instructions) costing 4 points each.72
[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+70
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.2
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 17 issues ( = indirect data accesses) costing 4 point each.68
Inefficient Vectorization+74
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 18 issues (= instructions) costing 4 points each.72
[SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points.2
Loop 665 - exec+Execution Time: 2 % - Vectorization Ratio: 98.29 % - Vector Length Use: 84.16 %
Loop Computation Issues+208
[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 52 issues (= instructions) costing 4 points each.208
Control Flow Issues+4
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.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
Data Access Issues+137
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 15 issues ( = indirect data accesses) costing 4 point each.60
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 18 issues (= instructions) costing 4 points each.72
[SA] Presence of special instructions executing on a single port (BROADCAST) - Simplify data access and try to get stride 1 access. There are 3 issues (= instructions) costing 1 point each.3
[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+64
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.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
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 15 issues ( = indirect data accesses) costing 4 point each.60
Inefficient Vectorization+77
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 18 issues (= instructions) costing 4 points each.72
[SA] Presence of special instructions executing on a single port (BROADCAST) - Simplify data access and try to get stride 1 access. There are 3 issues (= instructions) costing 1 point each.3
[SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points.2
Loop 121 - exec+Execution Time: 2 % - Vectorization Ratio: 97.56 % - Vector Length Use: 76.98 %
Loop Computation Issues+34
[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] 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+2
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.2
Data Access Issues+44
[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
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 6 issues (= instructions) costing 4 points each.24
Vectorization Roadblocks+22
[SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 2 issues (= calls) costing 1 point each.2
[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
Inefficient Vectorization+24
[SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 6 issues (= instructions) costing 4 points each.24
×