• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 53 subscribers
  • Views 735 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Compiler/TMS570LS3137: Which optimization techniques work or being used for each predefined optimization levels in CCS project configuration? (For TI 18.12.3 LTS Compiler)

harun saybak
Prodigy 30 points
Part Number: TMS570LS3137

Tool/software: TI C/C++ Compiler

In the document titled as  "ARM Optimizing C/C++ Compiler v18.12.0.LTS", there exist general information about optimization levels. However, ı could not find any specific information about which optimization techniques are being used or work for which optimization level. 

For example: In for which optimization levels is the technique called "Data flow optimizations" being used?

The related section published below for better understanding.

Could you help me about this situation, please?

Thanks,

 "ARM Optimizing C/C++ Compiler v18.12.0.LTS" : (http://www.ti.com/lit/ug/spnu151t/spnu151t.pdf)

  • 0
    TI__Guru**** 244300 points

    This post uses several different technical terms.  In all cases, the definitions given are not precise.  Rather, a general and intuitive description is given.

    Data flow optimizations is a general term for a group of optimization techniques.  Among these techniques is one named copy propagation.  Copy propagation is separated into different variants.  One is termed local and another is termed global.  The difference between local and global is how much code is considered, at one time, when applying the optimization.  A local optimization is applied only within blocks, where a block is a set of statements performed in sequence, usually ending in a branch or return from the function.  A global optimization is applied across blocks.  Under --opt_level=1, local copy propagation is performed.  Under --opt_level=2, both local and global copy propagation is performed.

    It is a similar story with other data flow optimizations such as common subexpression elimination and redundant assignments.

    Thanks and regards,

    -George