Help is available by moving the cursor above any symbol or by checking MAQAO website.
[ 0 / 4 ] Application profile is too short (0.96 s)
If the overall application profiling time is less than 10 seconds, many of the measurements at function or loop level will very likely be under the measurement quality threshold (0,1 seconds). Rerun to increase runtime duration: for example use a larger dataset or include a repetition loop.
[ 3 / 3 ] Optimization level option -O3 is used
To have better performances, it is advised to help the compiler by using a proper optimization level (-O3)
[ 3 / 3 ] Helper debug compilation options -g and -fno-omit-frame-pointer are used
-g option gives access to debugging informations, such are source locations and -fno-omit-frame-pointer improve the accuracy of callchains found during the application profiling.
[ 3 / 3 ] Architecture specific option -march=native is used
[ 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
[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (86.44%)
If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.
[ 4 / 4 ] Loop profile is not flat
At least one loop coverage is greater than 4% (30.21%), representing an hotspot for the application
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (85.92%)
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 / 3 ] Less than 10% (0%) is spend in BLAS1 operations
It could be more efficient to inline by hand BLAS1 operations
[ 3 / 3 ] Cumulative Outermost/In between loops coverage (0.52%) lower than cumulative innermost loop coverage (85.92%)
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%) is spend in Libm/SVML (special functions)
[ 2 / 2 ] Less than 10% (1.04%) is spend in BLAS2 operations
BLAS2 calls usually could make a poor cache usage and could benefit from inlining.
Loop ID | Module | Analysis | Penalty Score | Coverage (%) | Vectorization Ratio (%) | Vector Length Use (%) |
---|---|---|---|---|---|---|
►831 | exec | Inefficient vectorization. | 2 | 30.21 | 100 | 50 |
○ | [SA] Inefficient vectorization: use of shorter than available vector length - Force compiler to use proper vector length. CAUTION: use of 512 bits vectors could be more expensive than 256 bits on some processors. Use intrinsics (costly and not portable). The issue costs 2 points. | 2 | ||||
○ | Warning! There is no dynamic data for this loop. Some checks can not been performed. | 0 | ||||
►1167 | exec | Partial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access. | 48 | 23.44 | 92.54 | 46.22 |
○ | [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 4 issues (= instructions) costing 4 points each. | 16 | ||||
○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 3 issues (= instructions) costing 4 points each. | 12 | ||||
○ | [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 10 issues (= instructions) costing 1 point each. | 10 | ||||
○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 1 issues ( = indirect data accesses) costing 4 point each. | 4 | ||||
○ | [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points. | 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 | ||||
○ | [SA] Inefficient vectorization: use of shorter than available vector length - Force compiler to use proper vector length. CAUTION: use of 512 bits vectors could be more expensive than 256 bits on some processors. Use intrinsics (costly and not portable). The issue costs 2 points. | 2 | ||||
○ | Warning! There is no dynamic data for this loop. Some checks can not been performed. | 0 | ||||
►840 | exec | Inefficient vectorization. | 2 | 20.31 | 100 | 50 |
○ | [SA] Inefficient vectorization: use of shorter than available vector length - Force compiler to use proper vector length. CAUTION: use of 512 bits vectors could be more expensive than 256 bits on some processors. Use intrinsics (costly and not portable). The issue costs 2 points. | 2 | ||||
○ | Warning! There is no dynamic data for this loop. Some checks can not been performed. | 0 | ||||
►812 | exec | Inefficient vectorization. | 6 | 1.56 | 100 | 50 |
○ | [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] Inefficient vectorization: use of shorter than available vector length - Force compiler to use proper vector length. CAUTION: use of 512 bits vectors could be more expensive than 256 bits on some processors. Use intrinsics (costly and not portable). The issue costs 2 points. | 2 | ||||
○ | Warning! There is no dynamic data for this loop. Some checks can not been performed. | 0 | ||||
►1469 | exec | Inefficient vectorization. | 2 | 1.56 | 100 | 50 |
○ | [SA] Inefficient vectorization: use of shorter than available vector length - Force compiler to use proper vector length. CAUTION: use of 512 bits vectors could be more expensive than 256 bits on some processors. Use intrinsics (costly and not portable). The issue costs 2 points. | 2 | ||||
○ | Warning! There is no dynamic data for this loop. Some checks can not been performed. | 0 | ||||
►1246 | exec | Partial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access. | 9 | 1.56 | 0 | 12.5 |
○ | [SA] Too many paths (5 paths) - Simplify control structure. There are 5 issues ( = paths) costing 1 point each with a malus of 4 points. | 9 | ||||
○ | Warning! Some static analysis are missing because the loop has too many paths. Use a higher value for --maximal_path_number option. | 0 | ||||
►302 | exec | Partial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access. | 23 | 1.04 | 87.88 | 38.92 |
○ | [SA] Presence of special instructions executing on a single port (INSERT/EXTRACT, COMPRESS/EXPAND, BLEND/MERGE, SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 9 issues (= instructions) costing 1 point each. | 9 | ||||
○ | [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 3 issues ( = data accesses) costing 2 point each. | 6 | ||||
○ | [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points. | 2 | ||||
○ | [SA] Inefficient vectorization: use of shorter than available vector length - Force compiler to use proper vector length. CAUTION: use of 512 bits vectors could be more expensive than 256 bits on some processors. Use intrinsics (costly and not portable). The issue costs 2 points. | 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 | ||||
○ | [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 | ||||
○ | Warning! There is no dynamic data for this loop. Some checks can not been performed. | 0 | ||||
►930 | exec | Partial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access. | 11 | 0.52 | 85.71 | 41.07 |
○ | [SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 1 issues ( = indirect data accesses) costing 4 point each. | 4 | ||||
○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 1 issues (= instructions) costing 4 points each. | 4 | ||||
○ | [SA] Inefficient vectorization: use of shorter than available vector length - Force compiler to use proper vector length. CAUTION: use of 512 bits vectors could be more expensive than 256 bits on some processors. Use intrinsics (costly and not portable). The issue costs 2 points. | 2 | ||||
○ | [SA] Presence of special instructions executing on a single port (BROADCAST) - Simplify data access and try to get stride 1 access. There are 1 issues (= instructions) costing 1 point each. | 1 | ||||
○ | Warning! There is no dynamic data for this loop. Some checks can not been performed. | 0 | ||||
►240 | exec | Inefficient vectorization. | 2 | 0.52 | 100 | 50 |
○ | [SA] Inefficient vectorization: use of shorter than available vector length - Force compiler to use proper vector length. CAUTION: use of 512 bits vectors could be more expensive than 256 bits on some processors. Use intrinsics (costly and not portable). The issue costs 2 points. | 2 | ||||
○ | Warning! There is no dynamic data for this loop. Some checks can not been performed. | 0 | ||||
►1187 | exec | Partial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access. | 15 | 0.52 | 42.62 | 17.42 |
○ | [SA] Too many paths (8 paths) - Simplify control structure. There are 8 issues ( = paths) costing 1 point each with a malus of 4 points. | 12 | ||||
○ | [SA] Presence of calls - Inline either by compiler or by hand and use SVML for libm calls. There are 3 issues (= calls) costing 1 point each. | 3 | ||||
○ | Warning! Some static analysis are missing because the loop has too many paths. Use a higher value for --maximal_path_number option. | 0 |