Help is available by moving the cursor above any symbol or by checking MAQAO website.
[ 4 / 4 ] Application profile is long enough (45.79 s)
To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.
[ 3 / 3 ] Optimization level option -O2 is used
To have better performances, it is advised to help the compiler by using a proper optimization level (-O2)
[ 2 / 3 ] Helper debug compilation option -fno-omit-frame-pointer is missing
-fno-omit-frame-pointer improves the accuracy of callchains found during the application profiling.
[ 0 / 3 ] Architecture specific options are not used
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 0.02 % 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 (67.09%)
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% (60.02%), representing an hotspot for the application
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (66.40%)
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 (0.69%) lower than cumulative innermost loop coverage (66.4%)
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% (0.47%) 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 (%) |
---|---|---|---|---|---|---|
►917 | miniqmc | Partial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access. | 18 | 60.02 | 45.71 | 49.29 |
○ | [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 9 issues ( = data accesses) costing 2 point each. | 18 | ||||
►2010 | miniqmc | Partial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access. | 44 | 1.73 | 97.18 | 97.89 |
○ | [SA] Inefficient vectorization: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 9 issues ( = arrays) costing 2 points each | 18 | ||||
○ | [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] 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 | ||||
○ | [SA] Presence of special instructions executing on a single port (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 | ||||
○ | [DA] The ratio FP/LS (floating point / memory accesses) is between 0.8 and 1.2 (1.03) - Both arithmetic and data access have to be optimized simultaneously. | 0 | ||||
►1268 | miniqmc | Partial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access. | 64 | 0.82 | 95.8 | 96.85 |
○ | [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] Inefficient vectorization: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 9 issues ( = arrays) costing 2 points each | 18 | ||||
○ | [SA] Presence of special instructions executing on a single port (BROADCAST) - Simplify data access and try to get stride 1 access. There are 6 issues (= instructions) costing 1 point each. | 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] 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 | ||||
○ | [DA] The ratio FP/LS (floating point / memory accesses) is greater than 1.2 (1.37) - Focus on optimizing arithmetic. | 0 | ||||
►259 | miniqmc | Partial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access. | 85 | 0.65 | 0 | 21.53 |
○ | [SA] Too many paths (81 paths) - Simplify control structure. There are 81 issues ( = paths) costing 1 point each with a malus of 4 points. | 85 | ||||
○ | Warning! Some static analysis are missing because the loop has too many paths. Use a higher value for --maximal_path_number option. | 0 | ||||
►918 | miniqmc | Partial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access. | 24 | 0.59 | 20 | 27.5 |
○ | [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 11 issues ( = data accesses) costing 2 point each. | 22 | ||||
○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 | ||||
►1770 | miniqmc | Partial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access. | 14 | 0.5 | 54.55 | 31.82 |
○ | [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 5 issues ( = data accesses) costing 2 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 | ||||
►845 | miniqmc | Partial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access. | 4 | 0.34 | 56.52 | 41.3 |
○ | [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 | ||||
►836 | miniqmc | Partial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access. | 4 | 0.32 | 56.52 | 41.3 |
○ | [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 | ||||
►983 | miniqmc | Partial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access. | 14 | 0.32 | 14.29 | 22.32 |
○ | [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 | ||||
►623 | miniqmc | Inefficient vectorization. | 14 | 0.29 | 100 | 100 |
○ | [SA] Inefficient vectorization: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 5 issues ( = arrays) costing 2 points each | 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 | ||||
○ | [DA] The ratio FP/LS (floating point / memory accesses) is smaller than 0.8 (0.33) - Focus on optimizing data accesses. | 0 |