The AMD CodeAnalyst Performance Analyzer for Linux and the AMD Simulation Utilities form a suite of tools designed to assist software developers in optimizing their software for AMD microprocessors. First, profile your system or target code using the AMD CodeAnalyst Performance Analyzer to identify performance areas and issues. Then the AMD Simulation Utilities 2.1 can be used to identify the causes of the stalls and bottlenecks. This tool suite is designed to support 32-bit Linux distributions based around the 2.4 or 2.6 kernel series on x86 architecture. Earlier 2.2 series kernels are not supported. Distributions of 64-bit Linux on x86_64 architecture are supported. Please send feedback and questions to: codeanalyst.support@amd.com.

AMD CodeAnalyst Performance Analyzer 2.3 Beta for Linux

AMD CodeAnalyst 2.3 has changed Oprofile driver and daemon in order to profile Java application. AMD CodeAnalyst 2.3 supports SUSE® LINUX with 2.4 kernel today and will support SUSE LINUX 2.6 and Red Hat Linux in upcoming revisions. Download the BETA and send feedback or questions to: codeanalyst.support@amd.com

AMD CodeAnalyst Performance Analyzer 2.2 for Linux

AMD CodeAnalyst Performance Analyzer 2.2 is an open source front-end graphical user interface to Oprofile. The graphical user interface simplifies the profile configuration and enhances the post profile data analysis. Four performance counters can be configured from a large set of processor performance events for simultaneous profiling. The profiling data is provided in both tabular and chart formats through hierarchical drill down tabs.

AMD Simulation Utilities 2.1 for Linux

• tracegen utility
  The tracegen utility is used to generate a trace file of the target code module which is used by the simulate tool.
  
  
• simulate utility
  The simulate utility runs a simulation of a trace on a variety of processors. Processor simulation is used to extract extremely detailed processor cycle information on each instruction executed in a chosen block of code. The data presented is at the assembly instruction level and is not intended to assist a programmer working in a high-level language. The detailed data on the execution of each instruction takes into account the previous instructions executed and the state of the processor caches. The data is obtained by running the target block of code, then using the debug capabilities of the processor to single step through each opcode to obtain an execution trace. This execution trace is then fed into a Processor Simulation that analyzes the execution. Processor Simulation only supports single processor execution.

  Processor simulation supports the simulation of 32-bit code, offering a choice of:

  • AMD Athlon™ processor
  • AMD Athlon™ XP processor
  • AMD Opteron™ processor
  • AMD Athlon™ 64 processor

  Processor simulation also supports the simulation of 64-bit code, offering a choice of:

  • AMD Opteron processor
  • AMD Athlon 64 processor
    
    
• simreport
  The simulation report is used to identify software stalls for possible optimizations. The simulation data is saved, so reports can be generated at a later time. Data from two or more reports can be manually compared to verify progress on the optimization. Alternatively, optimization trends may be viewed from a series of reports.