options

Statistics

AnalysisCountPercentageWeighted Count
Loop Computation Issues94
Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA4981.670.63
Presence of a large number of scalar integer instructions2846.670.42
Presence of expensive FP instructions1321.670.11
Large loop body over microp cache size23.330.03
Bottleneck in the front-end23.330.03
Control Flow Issues25
Presence of 2 to 4 paths1423.330.13
Presence of calls46.670.04
Non-innermost loop46.670.04
Presence of more than 4 paths35.000.02
Data Access Issues39
Presence of constant non-unit stride data access2745.000.50
Presence of indirect access1220.000.22
Vectorization Roadblocks69
Presence of constant non-unit stride data access2745.000.50
Presence of 2 to 4 paths1423.330.13
Presence of indirect access1220.000.22
Presence of more than 4 paths813.330.07
Non-innermost loop46.670.04
Presence of calls46.670.04

Details

Analysisr_1r_2
Loop Computation IssuesPresence of expensive FP instructions67
Less than 10% of the FP ADD/SUB/MUL arithmetic operations are performed using FMA2425
Large loop body over microp cache size11
Presence of a large number of scalar integer instructions1612
Bottleneck in the front-end11
Control Flow IssuesPresence of calls13
Presence of 2 to 4 paths77
Presence of more than 4 paths21
Non-innermost loop13
Data Access IssuesPresence of constant non-unit stride data access1215
Presence of indirect access57
Vectorization RoadblocksPresence of calls13
Presence of 2 to 4 paths77
Presence of more than 4 paths35
Non-innermost loop13
Presence of constant non-unit stride data access1215
Presence of indirect access57
×