Lines Matching full:performance
38 \fBllvm\-mca\fP is a performance analysis tool that uses information
39 available in LLVM (e.g. scheduling models) to statically measure the performance
42 Performance is measured in terms of throughput as well as processor resource
46 The main goal of this tool is not just to predict the performance of the code
47 when run on the target, but also help with diagnosing potential performance
90 If you see that the performance report is not accurate for a processor,
326 is analyzed in isolation, and the final performance report is the union of all
538 to generate a performance report.
545 Here is an example of a performance report generated by the tool for a
626 few performance numbers; the goal of this section is to give a very quick
627 overview of the performance throughput. Important performance indicators are
651 field is an indicator of a performance issue. In the absence of loop\-carried
663 \fIBlock RThroughput\fP) is an indicator of a performance bottleneck caused by the
671 an indicator of a performance bottleneck caused by the lack of hardware
812 help diagnose performance bottlenecks caused by long data dependencies and
858 Table \fIAverage Wait times\fP helps diagnose performance issues that are caused by
864 When the performance is limited by data dependencies and/or long latency
869 instructions. When performance is mostly limited by the lack of hardware
876 performance bottlenecks.
927 performance. By construction, the accuracy of this analysis is strongly
933 .SS Extra Statistics to Further Diagnose Performance Issues
935 The \fB\-all\-stats\fP command line option enables extra statistics and performance
1096 decode stages are not modeled. Performance bottlenecks in the frontend are not