SPRUIV4D May   2020  – May 2024

 

  1.   1
  2.   Read This First
    1.     About This Manual
    2.     Related Documentation
    3.     Trademarks
  3. 2Introduction
    1. 2.1 C7000 Digital Signal Processor CPU Architecture Overview
    2. 2.2 C7000 Split Datapath and Functional Units
  4. 3C7000 C/C++ Compiler Options
    1. 3.1 Overview
    2. 3.2 Selecting Compiler Options for Performance
    3. 3.3 Understanding Compiler Optimization
      1. 3.3.1 Software Pipelining
      2. 3.3.2 Vectorization and Vector Predication
      3. 3.3.3 Automatic Use of Streaming Engine and Streaming Address Generator
      4. 3.3.4 Loop Collapsing and Loop Coalescing
      5. 3.3.5 Automatic Inlining
      6. 3.3.6 If Conversion
  5. 4Basic Code Optimization
    1. 4.1  Signed Types for Iteration Counters and Limits
    2. 4.2  Floating-Point Division
    3. 4.3  Loop-Carried Dependencies and the Restrict Keyword
      1. 4.3.1 Loop-Carried Dependencies
      2. 4.3.2 The Restrict Keyword
      3. 4.3.3 Run-Time Alias Disambiguation
    4. 4.4  Function Calls and Inlining
    5. 4.5  MUST_ITERATE and PROB_ITERATE Pragmas and Attributes
    6. 4.6  If Statements and Nested If Statements
    7. 4.7  Intrinsics
    8. 4.8  Vector Types
    9. 4.9  C++ Features to Use and Avoid
    10. 4.10 Streaming Engine
    11. 4.11 Streaming Address Generator
    12. 4.12 Optimized Libraries
    13. 4.13 Memory Optimizations
  6. 5Understanding the Assembly Comment Blocks
    1. 5.1 Software Pipelining Processing Stages
    2. 5.2 Software Pipeline Information Comment Block
      1. 5.2.1 Loop and Iteration Count Information
      2. 5.2.2 Dependency and Resource Bounds
      3. 5.2.3 Initiation Interval (ii) and Iterations
      4. 5.2.4 Constant Extensions
      5. 5.2.5 Resources Used and Register Tables
      6. 5.2.6 Stage Collapsing
      7. 5.2.7 Memory Bank Conflicts
      8. 5.2.8 Loop Duration Formula
    3. 5.3 Single Scheduled Iteration Comment Block
    4. 5.4 Identifying Pipeline Failures and Performance Issues
      1. 5.4.1 Issues that Prevent a Loop from Being Software Pipelined
      2. 5.4.2 Software Pipeline Failure Messages
      3. 5.4.3 Performance Issues
  7. 6Revision History

3.3.4 Loop Collapsing and Loop Coalescing

The compiler attempts to collapse or coalesce nested loops if it is legal and can improve performance. A nested loop is a set of two loops where one loop resides inside of another enclosing loop. Both collapsing and coalescing involve transforming a nested loop into a single loop. Collapsing takes place when there is no code in the outer loop. Coalescing takes place when there is code in the outer loop.

After the two nested loops are combined into one loop, the code that was in the body of the outer loop must be transformed so that it conditionally executes only when necessary. Collapsing and coalescing can have performance benefits because only one pipe-up and pipe-down are executed when the loop nest is executed, instead of a pipe-down and pipe-up of the inner loop every time the outer loop executes when loop coalescing/collapsing is not performed.

In order to perform loop collapsing or loop coalescing, the combined loop must be able to be software pipelined. This means that the loop nest must not contain function calls. The loops must each have a signed counting iterator that iterates a fixed amount each time. That is, the inner loop must not iterate a different number of times depending on which outer loop iteration execution is in. Also, the outer loop must not contain too much code, otherwise the transformation will not improve performance. If the outer loop carries a memory dependence, loop coalescing and loop collapsing likely will not be performed.

When loop collapsing or loop coalescing take place, the software pipelined loop indicates the beginning loop source line ("Loop source line") near the top of the software information comment block. When this source line number references an outer loop, this indicates that the inner loop has been fully unrolled or the compiler has performed loop coalescing or collapsing. In cases of loop coalescing, the compiler uses special instructions, such as NLCINIT, TICK, GETP, and BNL. A description of these hardware features, encompassing what is known as the "NLC", is beyond the scope of this document. More details of the NLC may be found in the C71x DSP CPU, Instruction Set, and Matrix Multiply Accelerator Technical Reference Manual (SPRUIP0).