options

exec - 2024-01-23 21:48:08 - MAQAO 2.19.0

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

Stylizer  

[ 4 / 4 ] Application profile is long enough (123.47 s)

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

[ 2.75 / 3 ] Optimization level option is correctly used

[ 2.74 / 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.

[ 2.74 / 3 ] Architecture specific option -xCORE is used

[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.1 % 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

Strategizer  

[ 4 / 4 ] Enough time of the experiment time spent in analyzed loops (46.40%)

If the time spent in analyzed loops is less than 30%, standard loop optimizations will have a limited impact on application performances.

[ 4 / 4 ] Loop profile is not flat

At least one loop coverage is greater than 4% (4.05%), representing an hotspot for the application

[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (38.53%)

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.

[ 0 / 3 ] More than 10% (18.22%) is spend in BLAS1 operations

It could be more efficient to inline by hand BLAS1 operations

[ 3 / 3 ] Cumulative Outermost/In between loops coverage (7.87%) lower than cumulative innermost loop coverage (38.53%)

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)

[ 0 / 2 ] More than 10% (19.16%) is spend in BLAS2 operations

BLAS2 calls usually make a poor cache usage. Try blocking. Such calls should probably benefit from inlining. Such inlining will have to be hand made.

Optimizer

Loop IDModuleAnalysisPenalty ScoreCoverage (%)Vectorization
Ratio (%)
Vector Length
Use (%)
2338execPartial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access.284.05012.5
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 7 issues ( = indirect data accesses) costing 4 point each.28
3176execPartial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access.293.2518.1813.64
[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 14 issues ( = data accesses) costing 2 point each.28
[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
2997execInefficient vectorization.23.2510050
[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
4649execPartial or unexisting vectorization - No issue detected02.5406.25
2944execInefficient vectorization.22.1910050
[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
2063execInefficient vectorization.21.810050
[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
2336execPartial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access.121.68012.5
[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 6 issues ( = data accesses) costing 2 point each.12
4599execPartial or unexisting vectorization - No issue detected01.4306.25
2235execPartial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access.81.33012.5
[SA] Presence of indirect accesses - Use array restructuring or gather instructions to lower the cost. There are 2 issues ( = indirect data accesses) costing 4 point each.8
2248execPartial or unexisting vectorization - Use pragma to force vectorization and check potential dependencies between array access.81.2311.1113.89
[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 4 issues ( = data accesses) costing 2 point each.8
×