options

Statistics

AnalysisCountPercentageWeighted Count
Loop Computation Issues95
Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA5083.330.69
Presence of a large number of scalar integer instructions2948.330.45
Presence of expensive FP instructions1220.000.11
Large loop body over microp cache size23.330.03
Bottleneck in the front-end23.330.03
Control Flow Issues25
Presence of 2 to 4 paths1525.000.16
Presence of calls46.670.04
Non-innermost loop46.670.04
Presence of more than 4 paths23.330.02
Data Access Issues46
Presence of constant non-unit stride data access3761.670.64
Presence of indirect access915.000.18
Vectorization Roadblocks76
Presence of constant non-unit stride data access3761.670.64
Presence of 2 to 4 paths1525.000.16
Presence of indirect access915.000.18
Presence of more than 4 paths711.670.06
Non-innermost loop46.670.04
Presence of calls46.670.04

Details

Analysisr_1r_2
Loop Computation IssuesPresence of expensive FP instructions84
Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA2624
Large loop body over microp cache size11
Presence of a large number of scalar integer instructions1217
Bottleneck in the front-end11
Control Flow IssuesPresence of calls31
Presence of 2 to 4 paths78
Presence of more than 4 paths11
Non-innermost loop31
Data Access IssuesPresence of constant non-unit stride data access2512
Presence of indirect access27
Vectorization RoadblocksPresence of calls31
Presence of 2 to 4 paths78
Presence of more than 4 paths52
Non-innermost loop31
Presence of constant non-unit stride data access2512
Presence of indirect access27
×