Help is available by moving the cursor above any 
symbol or by checking MAQAO website.
[ 4 / 4 ] Application profile is long enough (400.43 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 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.
[ 3 / 3 ] Optimization level option is correctly used
[ 3 / 3 ] Host configuration allows retrieval of all necessary metrics.
[ 3 / 3 ] Architecture specific option -march=znver5 is used
[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.64 % 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 (98.04%)
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 99.91% of observed threads are actually active
[ 4 / 4 ] CPU activity is good
CPU cores are active 99.92% of time
[ 4 / 4 ] Loop profile is not flat
At least one loop coverage is greater than 4% (91.91%), representing an hotspot for the application
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (96.40%)
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.
[ 4 / 4 ] Affinity is good (99.92%)
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 (1.63%) lower than cumulative innermost loop coverage (96.40%)
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 97 - exec | Execution Time: 91 % - Vectorization Ratio: 35.93 % - Vector Length Use: 16.99 % | |
| ►Loop Computation Issues | 4 | |
| ○ | [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 |
| ►Control Flow Issues | 3 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 3 |
| ►Data Access Issues | 4 | |
| ○ | [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 1 issues ( = data accesses) costing 2 point each. | 2 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 issues (= instructions) costing 1 point each. | 2 |
| ►Vectorization Roadblocks | 5 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 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 1 issues ( = data accesses) costing 2 point each. | 2 |
| ►Inefficient Vectorization | 2 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 issues (= instructions) costing 1 point each. | 2 |
| ►Loop 104 - exec | Execution Time: 1 % - Vectorization Ratio: 97.22 % - Vector Length Use: 72.57 % | |
| ►Data Access Issues | 39 | |
| ○ | [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 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 6 issues (= instructions) costing 4 points each. | 24 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Vectorization Roadblocks | 8 | |
| ○ | [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 |
| ►Inefficient Vectorization | 31 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 6 issues (= instructions) costing 4 points each. | 24 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Loop 66 - exec | Execution Time: 1 % - Vectorization Ratio: 97.22 % - Vector Length Use: 60.07 % | |
| ►Data Access Issues | 43 | |
| ○ | [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: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 0 issues ( = arrays) costing 2 points each | 0 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 8 issues (= instructions) costing 4 points each. | 32 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Vectorization Roadblocks | 4 | |
| ○ | [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 |
| ►Inefficient Vectorization | 41 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 8 issues (= instructions) costing 4 points each. | 32 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ○ | [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points. | 2 |
| ►Loop 96 - exec | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 6.25 % | |
| ►Control Flow Issues | 5 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 3 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 4 | |
| ○ | [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 |
| ►Vectorization Roadblocks | 9 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 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 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 90 - exec | Execution Time: 0 % - Vectorization Ratio: 17.39 % - Vector Length Use: 12.50 % | |
| ►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 | 334 | |
| ○ | [SA] Too many paths (330 paths) - Simplify control structure. There are 330 issues ( = paths) costing 1 point each with a malus of 4 points. | 334 |
| ►Vectorization Roadblocks | 334 | |
| ○ | [SA] Too many paths (330 paths) - Simplify control structure. There are 330 issues ( = paths) costing 1 point each with a malus of 4 points. | 334 |
| ►Loop 99 - exec | Execution Time: 0 % - Vectorization Ratio: 33.33 % - Vector Length Use: 12.50 % | |
| ►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 |
| ►Loop 107 - exec | Execution Time: 0 % - Vectorization Ratio: 82.18 % - Vector Length Use: 67.57 % | |
| ►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 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [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 | 20 | |
| ○ | [SA] Too many paths (14 paths) - Simplify control structure. There are 14 issues ( = paths) costing 1 point each with a malus of 4 points. | 18 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 77 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 14 issues (= instructions) costing 4 points each. | 56 |
| ○ | [SA] Presence of special instructions executing on a single port (INSERT/EXTRACT, SHUFFLE/PERM, BROADCAST, MASKMOV/MOV2M) - Simplify data access and try to get stride 1 access. There are 19 issues (= instructions) costing 1 point each. | 19 |
| ○ | [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 | 20 | |
| ○ | [SA] Too many paths (14 paths) - Simplify control structure. There are 14 issues ( = paths) costing 1 point each with a malus of 4 points. | 18 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Inefficient Vectorization | 77 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 14 issues (= instructions) costing 4 points each. | 56 |
| ○ | [SA] Presence of special instructions executing on a single port (INSERT/EXTRACT, SHUFFLE/PERM, BROADCAST, MASKMOV/MOV2M) - Simplify data access and try to get stride 1 access. There are 19 issues (= instructions) costing 1 point each. | 19 |
| ○ | [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points. | 2 |
| ►Loop 95 - exec | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 10.16 % | |
| ►Control Flow Issues | 6 | |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 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 | 6 | |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 103 - exec | Execution Time: 0 % - Vectorization Ratio: 50.00 % - Vector Length Use: 18.75 % | |
| ►Loop Computation Issues | 5 | |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Control Flow Issues | 5 | |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Loop 109 - exec | Execution Time: 0 % - Vectorization Ratio: 100.00 % - Vector Length Use: 95.24 % | |
| ►Loop Computation Issues | 8 | |
| ○ | [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 |
| ►Data Access Issues | 48 | |
| ○ | [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: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 0 issues ( = arrays) costing 2 points each | 0 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 36 issues (= instructions) costing 1 point each. | 36 |
| ►Vectorization Roadblocks | 4 | |
| ○ | [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 |
| ►Inefficient Vectorization | 44 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 36 issues (= instructions) costing 1 point each. | 36 |
[ 4 / 4 ] Application profile is long enough (56.83 s)
To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.
[ 3.00 / 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.
[ 3 / 3 ] Optimization level option is correctly used
[ 3 / 3 ] Host configuration allows retrieval of all necessary metrics.
[ 3.00 / 3 ] Architecture specific option -march=znver5 is used
[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.38 % 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 (90.51%)
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 496.21% of observed threads are actually active
[ 4 / 4 ] CPU activity is good
CPU cores are active 99.87% of time
[ 4 / 4 ] Loop profile is not flat
At least one loop coverage is greater than 4% (82.43%), representing an hotspot for the application
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (88.90%)
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 (99.90%)
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 (0.47%)
[ 3 / 3 ] Cumulative Outermost/In between loops coverage (1.61%) lower than cumulative innermost loop coverage (88.90%)
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 97 - exec | Execution Time: 82 % - Vectorization Ratio: 35.93 % - Vector Length Use: 16.99 % | |
| ►Loop Computation Issues | 4 | |
| ○ | [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 |
| ►Control Flow Issues | 3 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 3 |
| ►Data Access Issues | 4 | |
| ○ | [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 1 issues ( = data accesses) costing 2 point each. | 2 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 issues (= instructions) costing 1 point each. | 2 |
| ►Vectorization Roadblocks | 5 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 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 1 issues ( = data accesses) costing 2 point each. | 2 |
| ►Inefficient Vectorization | 2 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 issues (= instructions) costing 1 point each. | 2 |
| ►Loop 104 - exec | Execution Time: 1 % - Vectorization Ratio: 97.22 % - Vector Length Use: 72.57 % | |
| ►Data Access Issues | 39 | |
| ○ | [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 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 6 issues (= instructions) costing 4 points each. | 24 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Vectorization Roadblocks | 8 | |
| ○ | [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 |
| ►Inefficient Vectorization | 31 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 6 issues (= instructions) costing 4 points each. | 24 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Loop 99 - exec | Execution Time: 1 % - Vectorization Ratio: 33.33 % - Vector Length Use: 12.50 % | |
| ►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 |
| ►Loop 66 - exec | Execution Time: 1 % - Vectorization Ratio: 97.22 % - Vector Length Use: 60.07 % | |
| ►Data Access Issues | 43 | |
| ○ | [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: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 0 issues ( = arrays) costing 2 points each | 0 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 8 issues (= instructions) costing 4 points each. | 32 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Vectorization Roadblocks | 4 | |
| ○ | [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 |
| ►Inefficient Vectorization | 41 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 8 issues (= instructions) costing 4 points each. | 32 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ○ | [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points. | 2 |
| ►Loop 90 - exec | Execution Time: 1 % - Vectorization Ratio: 17.39 % - Vector Length Use: 12.50 % | |
| ►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 | 334 | |
| ○ | [SA] Too many paths (330 paths) - Simplify control structure. There are 330 issues ( = paths) costing 1 point each with a malus of 4 points. | 334 |
| ►Vectorization Roadblocks | 334 | |
| ○ | [SA] Too many paths (330 paths) - Simplify control structure. There are 330 issues ( = paths) costing 1 point each with a malus of 4 points. | 334 |
| ►Loop 96 - exec | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 6.25 % | |
| ►Control Flow Issues | 5 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 3 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 4 | |
| ○ | [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 |
| ►Vectorization Roadblocks | 9 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 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 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 107 - exec | Execution Time: 0 % - Vectorization Ratio: 82.18 % - Vector Length Use: 67.57 % | |
| ►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 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [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 | 20 | |
| ○ | [SA] Too many paths (14 paths) - Simplify control structure. There are 14 issues ( = paths) costing 1 point each with a malus of 4 points. | 18 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 77 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 14 issues (= instructions) costing 4 points each. | 56 |
| ○ | [SA] Presence of special instructions executing on a single port (INSERT/EXTRACT, SHUFFLE/PERM, BROADCAST, MASKMOV/MOV2M) - Simplify data access and try to get stride 1 access. There are 19 issues (= instructions) costing 1 point each. | 19 |
| ○ | [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 | 20 | |
| ○ | [SA] Too many paths (14 paths) - Simplify control structure. There are 14 issues ( = paths) costing 1 point each with a malus of 4 points. | 18 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Inefficient Vectorization | 77 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 14 issues (= instructions) costing 4 points each. | 56 |
| ○ | [SA] Presence of special instructions executing on a single port (INSERT/EXTRACT, SHUFFLE/PERM, BROADCAST, MASKMOV/MOV2M) - Simplify data access and try to get stride 1 access. There are 19 issues (= instructions) costing 1 point each. | 19 |
| ○ | [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points. | 2 |
| ►Loop 103 - exec | Execution Time: 0 % - Vectorization Ratio: 50.00 % - Vector Length Use: 18.75 % | |
| ►Loop Computation Issues | 5 | |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Control Flow Issues | 5 | |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Loop 95 - exec | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 10.16 % | |
| ►Control Flow Issues | 6 | |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 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 | 6 | |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 109 - exec | Execution Time: 0 % - Vectorization Ratio: 100.00 % - Vector Length Use: 95.24 % | |
| ►Loop Computation Issues | 8 | |
| ○ | [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 |
| ►Data Access Issues | 48 | |
| ○ | [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: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 0 issues ( = arrays) costing 2 points each | 0 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 36 issues (= instructions) costing 1 point each. | 36 |
| ►Vectorization Roadblocks | 4 | |
| ○ | [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 |
| ►Inefficient Vectorization | 44 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 36 issues (= instructions) costing 1 point each. | 36 |
[ 4 / 4 ] Application profile is long enough (42.91 s)
To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.
[ 3.00 / 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.
[ 3 / 3 ] Optimization level option is correctly used
[ 3 / 3 ] Host configuration allows retrieval of all necessary metrics.
[ 3.00 / 3 ] Architecture specific option -march=znver5 is used
[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.36 % 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 (85.25%)
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 890.54% of observed threads are actually active
[ 4 / 4 ] CPU activity is good
CPU cores are active 99.90% of time
[ 4 / 4 ] Loop profile is not flat
At least one loop coverage is greater than 4% (76.83%), representing an hotspot for the application
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (83.66%)
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 (99.90%)
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
[ 0 / 3 ] Too many functions do not use all threads
Functions running on a reduced number of threads (typically sequential code) cover at least 10% of application walltime (12.15%). Check both "Max Inclusive Time Over Threads" and "Nb Threads" in Functions or Loops tabs and consider parallelizing sequential regions or improving parallelization of regions running on a reduced number of threads
[ 3 / 3 ] Cumulative Outermost/In between loops coverage (1.59%) lower than cumulative innermost loop coverage (83.66%)
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 97 - exec | Execution Time: 76 % - Vectorization Ratio: 35.93 % - Vector Length Use: 16.99 % | |
| ►Loop Computation Issues | 4 | |
| ○ | [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 |
| ►Control Flow Issues | 3 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 3 |
| ►Data Access Issues | 4 | |
| ○ | [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 1 issues ( = data accesses) costing 2 point each. | 2 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 issues (= instructions) costing 1 point each. | 2 |
| ►Vectorization Roadblocks | 5 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 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 1 issues ( = data accesses) costing 2 point each. | 2 |
| ►Inefficient Vectorization | 2 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 issues (= instructions) costing 1 point each. | 2 |
| ►Loop 99 - exec | Execution Time: 2 % - Vectorization Ratio: 33.33 % - Vector Length Use: 12.50 % | |
| ►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 |
| ►Loop 104 - exec | Execution Time: 1 % - Vectorization Ratio: 97.22 % - Vector Length Use: 72.57 % | |
| ►Data Access Issues | 39 | |
| ○ | [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 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 6 issues (= instructions) costing 4 points each. | 24 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Vectorization Roadblocks | 8 | |
| ○ | [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 |
| ►Inefficient Vectorization | 31 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 6 issues (= instructions) costing 4 points each. | 24 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Loop 66 - exec | Execution Time: 1 % - Vectorization Ratio: 97.22 % - Vector Length Use: 60.07 % | |
| ►Data Access Issues | 43 | |
| ○ | [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: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 0 issues ( = arrays) costing 2 points each | 0 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 8 issues (= instructions) costing 4 points each. | 32 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Vectorization Roadblocks | 4 | |
| ○ | [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 |
| ►Inefficient Vectorization | 41 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 8 issues (= instructions) costing 4 points each. | 32 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ○ | [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points. | 2 |
| ►Loop 90 - exec | Execution Time: 0 % - Vectorization Ratio: 17.39 % - Vector Length Use: 12.50 % | |
| ►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 | 334 | |
| ○ | [SA] Too many paths (330 paths) - Simplify control structure. There are 330 issues ( = paths) costing 1 point each with a malus of 4 points. | 334 |
| ►Vectorization Roadblocks | 334 | |
| ○ | [SA] Too many paths (330 paths) - Simplify control structure. There are 330 issues ( = paths) costing 1 point each with a malus of 4 points. | 334 |
| ►Loop 96 - exec | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 6.25 % | |
| ►Control Flow Issues | 5 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 3 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 4 | |
| ○ | [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 |
| ►Vectorization Roadblocks | 9 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 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 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 107 - exec | Execution Time: 0 % - Vectorization Ratio: 82.18 % - Vector Length Use: 67.57 % | |
| ►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 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [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 | 20 | |
| ○ | [SA] Too many paths (14 paths) - Simplify control structure. There are 14 issues ( = paths) costing 1 point each with a malus of 4 points. | 18 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 77 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 14 issues (= instructions) costing 4 points each. | 56 |
| ○ | [SA] Presence of special instructions executing on a single port (INSERT/EXTRACT, SHUFFLE/PERM, BROADCAST, MASKMOV/MOV2M) - Simplify data access and try to get stride 1 access. There are 19 issues (= instructions) costing 1 point each. | 19 |
| ○ | [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 | 20 | |
| ○ | [SA] Too many paths (14 paths) - Simplify control structure. There are 14 issues ( = paths) costing 1 point each with a malus of 4 points. | 18 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Inefficient Vectorization | 77 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 14 issues (= instructions) costing 4 points each. | 56 |
| ○ | [SA] Presence of special instructions executing on a single port (INSERT/EXTRACT, SHUFFLE/PERM, BROADCAST, MASKMOV/MOV2M) - Simplify data access and try to get stride 1 access. There are 19 issues (= instructions) costing 1 point each. | 19 |
| ○ | [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points. | 2 |
| ►Loop 103 - exec | Execution Time: 0 % - Vectorization Ratio: 50.00 % - Vector Length Use: 18.75 % | |
| ►Loop Computation Issues | 5 | |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Control Flow Issues | 5 | |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Loop 95 - exec | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 10.16 % | |
| ►Control Flow Issues | 6 | |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 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 | 6 | |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 109 - exec | Execution Time: 0 % - Vectorization Ratio: 100.00 % - Vector Length Use: 95.24 % | |
| ►Loop Computation Issues | 8 | |
| ○ | [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 |
| ►Data Access Issues | 48 | |
| ○ | [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: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 0 issues ( = arrays) costing 2 points each | 0 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 36 issues (= instructions) costing 1 point each. | 36 |
| ►Vectorization Roadblocks | 4 | |
| ○ | [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 |
| ►Inefficient Vectorization | 44 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 36 issues (= instructions) costing 1 point each. | 36 |
[ 4 / 4 ] Application profile is long enough (35.69 s)
To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.
[ 3.00 / 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.
[ 3 / 3 ] Optimization level option is correctly used
[ 3 / 3 ] Host configuration allows retrieval of all necessary metrics.
[ 3.00 / 3 ] Architecture specific option -march=znver5 is used
[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.30 % 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 (82.92%)
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 1283.87% of observed threads are actually active
[ 4 / 4 ] CPU activity is good
CPU cores are active 99.92% of time
[ 4 / 4 ] Loop profile is not flat
At least one loop coverage is greater than 4% (74.14%), representing an hotspot for the application
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (81.31%)
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 (99.92%)
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
[ 0 / 3 ] Too many functions do not use all threads
Functions running on a reduced number of threads (typically sequential code) cover at least 10% of application walltime (15.06%). Check both "Max Inclusive Time Over Threads" and "Nb Threads" in Functions or Loops tabs and consider parallelizing sequential regions or improving parallelization of regions running on a reduced number of threads
[ 3 / 3 ] Cumulative Outermost/In between loops coverage (1.61%) lower than cumulative innermost loop coverage (81.31%)
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 97 - exec | Execution Time: 74 % - Vectorization Ratio: 35.93 % - Vector Length Use: 16.99 % | |
| ►Loop Computation Issues | 4 | |
| ○ | [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 |
| ►Control Flow Issues | 3 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 3 |
| ►Data Access Issues | 4 | |
| ○ | [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 1 issues ( = data accesses) costing 2 point each. | 2 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 issues (= instructions) costing 1 point each. | 2 |
| ►Vectorization Roadblocks | 5 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 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 1 issues ( = data accesses) costing 2 point each. | 2 |
| ►Inefficient Vectorization | 2 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 issues (= instructions) costing 1 point each. | 2 |
| ►Loop 99 - exec | Execution Time: 2 % - Vectorization Ratio: 33.33 % - Vector Length Use: 12.50 % | |
| ►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 |
| ►Loop 104 - exec | Execution Time: 1 % - Vectorization Ratio: 97.22 % - Vector Length Use: 72.57 % | |
| ►Data Access Issues | 39 | |
| ○ | [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 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 6 issues (= instructions) costing 4 points each. | 24 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Vectorization Roadblocks | 8 | |
| ○ | [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 |
| ►Inefficient Vectorization | 31 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 6 issues (= instructions) costing 4 points each. | 24 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Loop 66 - exec | Execution Time: 1 % - Vectorization Ratio: 97.22 % - Vector Length Use: 60.07 % | |
| ►Data Access Issues | 43 | |
| ○ | [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: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 0 issues ( = arrays) costing 2 points each | 0 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 8 issues (= instructions) costing 4 points each. | 32 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Vectorization Roadblocks | 4 | |
| ○ | [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 |
| ►Inefficient Vectorization | 41 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 8 issues (= instructions) costing 4 points each. | 32 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ○ | [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points. | 2 |
| ►Loop 96 - exec | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 6.25 % | |
| ►Control Flow Issues | 5 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 3 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 4 | |
| ○ | [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 |
| ►Vectorization Roadblocks | 9 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 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 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 90 - exec | Execution Time: 0 % - Vectorization Ratio: 17.39 % - Vector Length Use: 12.50 % | |
| ►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 | 334 | |
| ○ | [SA] Too many paths (330 paths) - Simplify control structure. There are 330 issues ( = paths) costing 1 point each with a malus of 4 points. | 334 |
| ►Vectorization Roadblocks | 334 | |
| ○ | [SA] Too many paths (330 paths) - Simplify control structure. There are 330 issues ( = paths) costing 1 point each with a malus of 4 points. | 334 |
| ►Loop 107 - exec | Execution Time: 0 % - Vectorization Ratio: 82.18 % - Vector Length Use: 67.57 % | |
| ►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 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [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 | 20 | |
| ○ | [SA] Too many paths (14 paths) - Simplify control structure. There are 14 issues ( = paths) costing 1 point each with a malus of 4 points. | 18 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 77 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 14 issues (= instructions) costing 4 points each. | 56 |
| ○ | [SA] Presence of special instructions executing on a single port (INSERT/EXTRACT, SHUFFLE/PERM, BROADCAST, MASKMOV/MOV2M) - Simplify data access and try to get stride 1 access. There are 19 issues (= instructions) costing 1 point each. | 19 |
| ○ | [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 | 20 | |
| ○ | [SA] Too many paths (14 paths) - Simplify control structure. There are 14 issues ( = paths) costing 1 point each with a malus of 4 points. | 18 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Inefficient Vectorization | 77 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 14 issues (= instructions) costing 4 points each. | 56 |
| ○ | [SA] Presence of special instructions executing on a single port (INSERT/EXTRACT, SHUFFLE/PERM, BROADCAST, MASKMOV/MOV2M) - Simplify data access and try to get stride 1 access. There are 19 issues (= instructions) costing 1 point each. | 19 |
| ○ | [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points. | 2 |
| ►Loop 103 - exec | Execution Time: 0 % - Vectorization Ratio: 50.00 % - Vector Length Use: 18.75 % | |
| ►Loop Computation Issues | 5 | |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Control Flow Issues | 5 | |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Loop 95 - exec | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 10.16 % | |
| ►Control Flow Issues | 6 | |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 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 | 6 | |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 109 - exec | Execution Time: 0 % - Vectorization Ratio: 100.00 % - Vector Length Use: 95.24 % | |
| ►Loop Computation Issues | 8 | |
| ○ | [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 |
| ►Data Access Issues | 48 | |
| ○ | [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: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 0 issues ( = arrays) costing 2 points each | 0 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 36 issues (= instructions) costing 1 point each. | 36 |
| ►Vectorization Roadblocks | 4 | |
| ○ | [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 |
| ►Inefficient Vectorization | 44 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 36 issues (= instructions) costing 1 point each. | 36 |
[ 4 / 4 ] Application profile is long enough (33.22 s)
To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.
[ 3.00 / 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.
[ 3 / 3 ] Optimization level option is correctly used
[ 3 / 3 ] Host configuration allows retrieval of all necessary metrics.
[ 3.00 / 3 ] Architecture specific option -march=znver5 is used
[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.26 % 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 (76.70%)
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 1677.15% of observed threads are actually active
[ 4 / 4 ] CPU activity is good
CPU cores are active 99.93% of time
[ 4 / 4 ] Loop profile is not flat
At least one loop coverage is greater than 4% (67.92%), representing an hotspot for the application
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (75.13%)
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 (99.92%)
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
[ 0 / 3 ] Too many functions do not use all threads
Functions running on a reduced number of threads (typically sequential code) cover at least 10% of application walltime (16.42%). Check both "Max Inclusive Time Over Threads" and "Nb Threads" in Functions or Loops tabs and consider parallelizing sequential regions or improving parallelization of regions running on a reduced number of threads
[ 3 / 3 ] Cumulative Outermost/In between loops coverage (1.56%) lower than cumulative innermost loop coverage (75.13%)
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 97 - exec | Execution Time: 67 % - Vectorization Ratio: 35.93 % - Vector Length Use: 16.99 % | |
| ►Loop Computation Issues | 4 | |
| ○ | [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 |
| ►Control Flow Issues | 3 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 3 |
| ►Data Access Issues | 4 | |
| ○ | [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 1 issues ( = data accesses) costing 2 point each. | 2 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 issues (= instructions) costing 1 point each. | 2 |
| ►Vectorization Roadblocks | 5 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 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 1 issues ( = data accesses) costing 2 point each. | 2 |
| ►Inefficient Vectorization | 2 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 issues (= instructions) costing 1 point each. | 2 |
| ►Loop 99 - exec | Execution Time: 2 % - Vectorization Ratio: 33.33 % - Vector Length Use: 12.50 % | |
| ►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 |
| ►Loop 104 - exec | Execution Time: 1 % - Vectorization Ratio: 97.22 % - Vector Length Use: 72.57 % | |
| ►Data Access Issues | 39 | |
| ○ | [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 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 6 issues (= instructions) costing 4 points each. | 24 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Vectorization Roadblocks | 8 | |
| ○ | [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 |
| ►Inefficient Vectorization | 31 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 6 issues (= instructions) costing 4 points each. | 24 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Loop 66 - exec | Execution Time: 1 % - Vectorization Ratio: 97.22 % - Vector Length Use: 60.07 % | |
| ►Data Access Issues | 43 | |
| ○ | [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: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 0 issues ( = arrays) costing 2 points each | 0 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 8 issues (= instructions) costing 4 points each. | 32 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Vectorization Roadblocks | 4 | |
| ○ | [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 |
| ►Inefficient Vectorization | 41 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 8 issues (= instructions) costing 4 points each. | 32 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ○ | [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points. | 2 |
| ►Loop 96 - exec | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 6.25 % | |
| ►Control Flow Issues | 5 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 3 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 4 | |
| ○ | [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 |
| ►Vectorization Roadblocks | 9 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 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 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 90 - exec | Execution Time: 0 % - Vectorization Ratio: 17.39 % - Vector Length Use: 12.50 % | |
| ►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 | 334 | |
| ○ | [SA] Too many paths (330 paths) - Simplify control structure. There are 330 issues ( = paths) costing 1 point each with a malus of 4 points. | 334 |
| ►Vectorization Roadblocks | 334 | |
| ○ | [SA] Too many paths (330 paths) - Simplify control structure. There are 330 issues ( = paths) costing 1 point each with a malus of 4 points. | 334 |
| ►Loop 107 - exec | Execution Time: 0 % - Vectorization Ratio: 82.18 % - Vector Length Use: 67.57 % | |
| ►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 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [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 | 20 | |
| ○ | [SA] Too many paths (14 paths) - Simplify control structure. There are 14 issues ( = paths) costing 1 point each with a malus of 4 points. | 18 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 77 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 14 issues (= instructions) costing 4 points each. | 56 |
| ○ | [SA] Presence of special instructions executing on a single port (INSERT/EXTRACT, SHUFFLE/PERM, BROADCAST, MASKMOV/MOV2M) - Simplify data access and try to get stride 1 access. There are 19 issues (= instructions) costing 1 point each. | 19 |
| ○ | [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 | 20 | |
| ○ | [SA] Too many paths (14 paths) - Simplify control structure. There are 14 issues ( = paths) costing 1 point each with a malus of 4 points. | 18 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Inefficient Vectorization | 77 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 14 issues (= instructions) costing 4 points each. | 56 |
| ○ | [SA] Presence of special instructions executing on a single port (INSERT/EXTRACT, SHUFFLE/PERM, BROADCAST, MASKMOV/MOV2M) - Simplify data access and try to get stride 1 access. There are 19 issues (= instructions) costing 1 point each. | 19 |
| ○ | [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points. | 2 |
| ►Loop 103 - exec | Execution Time: 0 % - Vectorization Ratio: 50.00 % - Vector Length Use: 18.75 % | |
| ►Loop Computation Issues | 5 | |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Control Flow Issues | 5 | |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Loop 109 - exec | Execution Time: 0 % - Vectorization Ratio: 100.00 % - Vector Length Use: 95.24 % | |
| ►Loop Computation Issues | 8 | |
| ○ | [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 |
| ►Data Access Issues | 48 | |
| ○ | [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: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 0 issues ( = arrays) costing 2 points each | 0 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 36 issues (= instructions) costing 1 point each. | 36 |
| ►Vectorization Roadblocks | 4 | |
| ○ | [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 |
| ►Inefficient Vectorization | 44 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 36 issues (= instructions) costing 1 point each. | 36 |
| ►Loop 95 - exec | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 10.16 % | |
| ►Control Flow Issues | 6 | |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 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 | 6 | |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
[ 4 / 4 ] Application profile is long enough (30.44 s)
To have good quality measurements, it is advised that the application profiling time is greater than 10 seconds.
[ 3.00 / 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.
[ 3 / 3 ] Optimization level option is correctly used
[ 3 / 3 ] Host configuration allows retrieval of all necessary metrics.
[ 3.00 / 3 ] Architecture specific option -march=znver5 is used
[ 2 / 2 ] Application is correctly profiled ("Others" category represents 0.25 % 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 (74.68%)
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 2069.65% of observed threads are actually active
[ 4 / 4 ] CPU activity is good
CPU cores are active 99.91% of time
[ 4 / 4 ] Loop profile is not flat
At least one loop coverage is greater than 4% (65.78%), representing an hotspot for the application
[ 4 / 4 ] Enough time of the experiment time spent in analyzed innermost loops (73.18%)
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 (99.93%)
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
[ 0 / 3 ] Too many functions do not use all threads
Functions running on a reduced number of threads (typically sequential code) cover at least 10% of application walltime (18.63%). Check both "Max Inclusive Time Over Threads" and "Nb Threads" in Functions or Loops tabs and consider parallelizing sequential regions or improving parallelization of regions running on a reduced number of threads
[ 3 / 3 ] Cumulative Outermost/In between loops coverage (1.50%) lower than cumulative innermost loop coverage (73.18%)
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 97 - exec | Execution Time: 65 % - Vectorization Ratio: 35.93 % - Vector Length Use: 16.99 % | |
| ►Loop Computation Issues | 4 | |
| ○ | [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 |
| ►Control Flow Issues | 3 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 3 |
| ►Data Access Issues | 4 | |
| ○ | [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 1 issues ( = data accesses) costing 2 point each. | 2 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 issues (= instructions) costing 1 point each. | 2 |
| ►Vectorization Roadblocks | 5 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 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 1 issues ( = data accesses) costing 2 point each. | 2 |
| ►Inefficient Vectorization | 2 | |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 2 issues (= instructions) costing 1 point each. | 2 |
| ►Loop 99 - exec | Execution Time: 2 % - Vectorization Ratio: 33.33 % - Vector Length Use: 12.50 % | |
| ►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 |
| ►Loop 104 - exec | Execution Time: 1 % - Vectorization Ratio: 97.22 % - Vector Length Use: 72.57 % | |
| ►Data Access Issues | 39 | |
| ○ | [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 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 6 issues (= instructions) costing 4 points each. | 24 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Vectorization Roadblocks | 8 | |
| ○ | [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 |
| ►Inefficient Vectorization | 31 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 6 issues (= instructions) costing 4 points each. | 24 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Loop 66 - exec | Execution Time: 1 % - Vectorization Ratio: 97.22 % - Vector Length Use: 60.07 % | |
| ►Data Access Issues | 43 | |
| ○ | [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: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 0 issues ( = arrays) costing 2 points each | 0 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 8 issues (= instructions) costing 4 points each. | 32 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ►Vectorization Roadblocks | 4 | |
| ○ | [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 |
| ►Inefficient Vectorization | 41 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 8 issues (= instructions) costing 4 points each. | 32 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM, BROADCAST) - Simplify data access and try to get stride 1 access. There are 7 issues (= instructions) costing 1 point each. | 7 |
| ○ | [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points. | 2 |
| ►Loop 96 - exec | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 6.25 % | |
| ►Control Flow Issues | 5 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 3 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 4 | |
| ○ | [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 |
| ►Vectorization Roadblocks | 9 | |
| ○ | [SA] Several paths (3 paths) - Simplify control structure or force the compiler to use masked instructions. There are 3 issues ( = paths) costing 1 point each. | 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 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Loop 107 - exec | Execution Time: 0 % - Vectorization Ratio: 82.18 % - Vector Length Use: 67.57 % | |
| ►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 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [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 | 20 | |
| ○ | [SA] Too many paths (14 paths) - Simplify control structure. There are 14 issues ( = paths) costing 1 point each with a malus of 4 points. | 18 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 77 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 14 issues (= instructions) costing 4 points each. | 56 |
| ○ | [SA] Presence of special instructions executing on a single port (INSERT/EXTRACT, SHUFFLE/PERM, BROADCAST, MASKMOV/MOV2M) - Simplify data access and try to get stride 1 access. There are 19 issues (= instructions) costing 1 point each. | 19 |
| ○ | [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 | 20 | |
| ○ | [SA] Too many paths (14 paths) - Simplify control structure. There are 14 issues ( = paths) costing 1 point each with a malus of 4 points. | 18 |
| ○ | [SA] Non innermost loop (Outermost) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Inefficient Vectorization | 77 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 14 issues (= instructions) costing 4 points each. | 56 |
| ○ | [SA] Presence of special instructions executing on a single port (INSERT/EXTRACT, SHUFFLE/PERM, BROADCAST, MASKMOV/MOV2M) - Simplify data access and try to get stride 1 access. There are 19 issues (= instructions) costing 1 point each. | 19 |
| ○ | [SA] Inefficient vectorization: use of masked instructions - Simplify control structure. The issue costs 2 points. | 2 |
| ►Loop 90 - exec | Execution Time: 0 % - Vectorization Ratio: 17.39 % - Vector Length Use: 12.50 % | |
| ►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 | 334 | |
| ○ | [SA] Too many paths (330 paths) - Simplify control structure. There are 330 issues ( = paths) costing 1 point each with a malus of 4 points. | 334 |
| ►Vectorization Roadblocks | 334 | |
| ○ | [SA] Too many paths (330 paths) - Simplify control structure. There are 330 issues ( = paths) costing 1 point each with a malus of 4 points. | 334 |
| ►Loop 103 - exec | Execution Time: 0 % - Vectorization Ratio: 50.00 % - Vector Length Use: 18.75 % | |
| ►Loop Computation Issues | 5 | |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Control Flow Issues | 5 | |
| ○ | [SA] Peel/tail loop, considered having a low iteration count - Perform full unroll. Force compiler to use masked instructions. This issue costs 5 points. | 5 |
| ►Loop 109 - exec | Execution Time: 0 % - Vectorization Ratio: 100.00 % - Vector Length Use: 95.24 % | |
| ►Loop Computation Issues | 8 | |
| ○ | [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 |
| ►Data Access Issues | 48 | |
| ○ | [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: more than 10% of the vector loads instructions are unaligned - When allocating arrays, don’t forget to align them. There are 0 issues ( = arrays) costing 2 points each | 0 |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 36 issues (= instructions) costing 1 point each. | 36 |
| ►Vectorization Roadblocks | 4 | |
| ○ | [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 |
| ►Inefficient Vectorization | 44 | |
| ○ | [SA] Presence of expensive instructions (GATHER/SCATTER) - Use array restructuring. There are 2 issues (= instructions) costing 4 points each. | 8 |
| ○ | [SA] Presence of special instructions executing on a single port (SHUFFLE/PERM) - Simplify data access and try to get stride 1 access. There are 36 issues (= instructions) costing 1 point each. | 36 |
| ►Loop 95 - exec | Execution Time: 0 % - Vectorization Ratio: 0.00 % - Vector Length Use: 10.16 % | |
| ►Control Flow Issues | 6 | |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |
| ►Data Access Issues | 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 | 6 | |
| ○ | [SA] Several paths (4 paths) - Simplify control structure or force the compiler to use masked instructions. There are 4 issues ( = paths) costing 1 point each. | 4 |
| ○ | [SA] Non innermost loop (InBetween) - Collapse loop with innermost ones. This issue costs 2 points. | 2 |