The Transmeta Code Morphing/spl trade/ Software: using speculation, recovery, and adaptive retranslation to address real-life challenges

Transmeta's Crusoe microprocessor is a full, system-level implementation of the x86 architecture, comprising a native VLIW microprocessor with a software layer, the Code Morphing Software (CMS), that combines an interpreter, dynamic binary translator, optimizer, and run-time system. In its general structure, CMS resembles other binary translation systems described in the literature, but it is unique in several respects. The wide range of PC workloads that CMS must handle gracefully in real-life operation, plus the need for full system-level x86 compatibility, expose several issues that have received little or no attention in previous literature, such as exceptions and interrupts, I/O, DMA, and self-modifying code. In this paper we discuss some of the challenges raised by these issues, and present the techniques developed in Crusoe and CMS to meet those challenges. The key to these solutions is the Crusoe paradigm of aggressive speculation, recovery to a consistent x86 state using unique hardware commit-and-rollback support, and adaptive retranslation when exceptions occur too often to be handled efficiently by interpretation.

[1]  David Keppel,et al.  Shade: a fast instruction-set simulator for execution profiling , 1994, SIGMETRICS.

[2]  Paolo Faraboschi,et al.  DELI: a new run-time control point , 2002, MICRO.

[3]  Paolo Faraboschi,et al.  DELI: a new run-time control point , 2002, 35th Annual IEEE/ACM International Symposium on Microarchitecture, 2002. (MICRO-35). Proceedings..

[4]  Michael Gschwind,et al.  Precise Exception Semantics in Dynamic Compilation , 2002, CC.

[5]  K. Ebcioglu,et al.  Daisy: Dynamic Compilation For 10o?40 Architectural Compatibility , 1997, Conference Proceedings. The 24th Annual International Symposium on Computer Architecture.

[6]  Kemal Ebcioglu,et al.  An architectural framework for supporting heterogeneous instruction-set architectures , 1993, Computer.

[7]  John Yates,et al.  FX!32 a profile-directed binary translator , 1998, IEEE Micro.

[8]  Erik R. Altman,et al.  Daisy: Dynamic Compilation For 10o?40 Architectural Compatibility , 1997, Conference Proceedings. The 24th Annual International Symposium on Computer Architecture.

[9]  Richard L. Sites,et al.  Binary translation , 1993, CACM.

[10]  Erik R. Altman,et al.  Welcome to the Opportunities of Binary Translation , 2000, Computer.

[11]  Frank G. Soltis.  Inside the As/400 , 1996 .

[12]  Michael Gschwind,et al.  Dynamic and Transparent Binary Translation , 2000, Computer.

[13]  Michael Gschwind,et al.  Dynamic Binary Translation and Optimization , 2001, IEEE Trans. Computers.

[14]  Frank Yellin,et al.  The Java Virtual Machine Specification , 1996 .

[15]  Colin B. Hunter,et al.  DOS at RISC , 1989 .

[16]  Michael D. Smith,et al.  Efficient superscalar performance through boosting , 1992, ASPLOS V.

[17]  Vasanth Bala,et al.  Transparent Dynamic Optimization: The Design and Implementation of Dynamo , 1999 .

[18]  Erik R. Altman,et al.  Advances and future challenges in binary translation and optimization , 2001, Proc. IEEE.

[19]  Erik R. Altman,et al.  LaTTe: a Java VM just-in-time compiler with fast and efficient register allocation , 1999, 1999 International Conference on Parallel Architectures and Compilation Techniques (Cat. No.PR00425).

[20]  Cindy Zheng,et al.  PA-RISC to IA-64: Transparent Execution, No Recompilation , 2000, Computer.

[21]  Richard L. Sites,et al.  Binary translation : Digital's alpha chip project , 1993 .

[22]  Scott A. Mahlke,et al.  Dynamic memory disambiguation using the memory conflict buffer , 1994, ASPLOS VI.

[23]  Thomas R. Gross,et al.  Avoidance and suppression of compensation code in a trace scheduling compiler , 1994, TOPL.