Abstract


 
 
 Code size is a primary concern in the embedded computing community. Minimizing physical memory requirements reduces total system cost and improves performance and power efficiency. VLIW processors rely on the compiler to statically encode the ILP in the program before its execution, and because of this, code size is larger relative to other processors. In this paper we describe the co-design of compiler optimizations and processor architecture features that have progressively reduced code size across three generations of a VLIW processor.
 
 

Highlights

  • VLIW processors are well-suited for high performance embedded applications, which are characterized by mathematically oriented loop kernels and abundant ILP

  • NOP compression has almost doubled the code-size reduction to 32%. – C6X-3(VLI): Compiler transformations to exploit variable length instructions have degraded the speedup by 2.6%, but the code-size reduction has improved to 43%. – C6X-3(VLI+MLB): The 0.8% speedup is because the modulo loop buffer allows the compiler to aggressively software pipeline loops even when compiling for size

  • Code size is a primary concern in the embedded computing community

Read more

Summary

Introduction

VLIW processors are well-suited for high performance embedded applications, which are characterized by mathematically oriented loop kernels and abundant ILP. The C6X-2 processors increase the size of register file by providing an additional 32 static general-purpose registers. If the p-bit of instruction i is 0, instruction i + 1 is part of the execute packet. The least significant bits (LSBs) of the program memory address shows how the fetch packets are laid out in memory. On the C6X-1 processors, execute packets cannot span a fetch packet boundary. The op field encodes the operation and functional unit, and the s-bit specifies the cluster that the instruction executes on.

NOP Compression
Software-pipelined Loop Collapsing
Modulo Loop Buffer
Variable Length Instructions
The Fetch Packet Header
Results
Summary
Related Work
Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.