Contents

• Introduction
  • Contributing your improvement to ATLAS
  • Directory terminology
  • Coding conventions for contributed code
  • A note about ATLAS kernels
  
  
• Speeding up the Level 3 BLAS
  • Building the General Matrix Multiply From gemmK
  • The Main GEMM Kernel, gemmK
  • Putting it together with some examples
  • More timing info
  • Complex gemmK
  • What to do if you are writing in assembler
  • Providing ATLAS with kernel cleanup code
  • gemmK usage notes
  • Getting ATLAS to use your kernel
  • Contributing a complete GEMM implementation
  
  
• Speeding up the Level 2 BLAS
  • Testing and timing mvn_k
  • Testing and timing mvt_k
  • Testing and timing ger_k
  • Testing and timing ger2_k
  • OLDSTUFF
  • Speeding Up GEMV, HEMV, SYMV, TRMV and TRSV
  • Speeding Up GER, GERU, GERC, HER, HER2, SYR and SYR2
  
  
• Speeding up the Level 1 BLAS
  • General comments for Level 1 optimization
  • Testing a kernel
  • Timing a kernel
  • What to do if you are writing in assembler
  • Ramblings on special cases
  • Notes for ROTG, ROTMG, ROT, ROTM
  • Notes for ROT
  • Notes for ASUM
  • Notes for AXPBY
  • Notes for AXPY
  • Notes for COPY
  • Notes for CPSC
  • Notes for DOT
  • Notes for IAMAX
  • Notes for NRM2
  • Notes for SCAL
  • Notes for SET
  • Notes for SWAP
  • Getting your new kernel used
  
  
• Using special compilers and flags for kernel compilation
  • Specifying all new compilers and flags
  • Specifying additional flags for the default compiler
  • Using a binary kernel
  
  
• A quick reference to ATLAS programming resources
  • ATLAS's prefetch header file
  
  
• Conclusion
• Some notes on using assembly
  • Some notes on x86-32 assembler
  • Some notes on x86-64 assembler
  • Some notes on Sparc assembler
  • Some notes on PowerPC assembler
  • Some notes on HP PA-RISC assembler
  • Some notes on MIPS assembler