patch applied (ghc): Implement SSE2 floating-point support in the x86 native code generator (#594)

[Simon Marlow]

Thu Feb  4 02:48:49 PST 2010  Simon Marlow <marlowsd at gmail.com>
  * Implement SSE2 floating-point support in the x86 native code generator (#594)
  
  The new flag -msse2 enables code generation for SSE2 on x86.  It
  results in substantially faster floating-point performance; the main
  reason for doing this was that our x87 code generation is appallingly
  bad, and since we plan to drop -fvia-C soon, we need a way to generate
  half-decent floating-point code.
  
  The catch is that SSE2 is only available on CPUs that support it (P4+,
  AMD K8+).  We'll have to think hard about whether we should enable it
  by default for the libraries we ship.  In the meantime, at least
  -msse2 should be an acceptable replacement for "-fvia-C
  -optc-ffast-math -fexcess-precision".
  
  SSE2 also has the advantage of performing all operations at the
  correct precision, so floating-point results are consistent with other
  platforms.
  
  I also tweaked the x87 code generation a bit while I was here, now
  it's slighlty less bad than before.

    M ./compiler/main/DynFlags.hs +4
    M ./compiler/nativeGen/PPC/Ppr.hs +1
    M ./compiler/nativeGen/PPC/Regs.hs -23 +6
    M ./compiler/nativeGen/Reg.hs +5
    M ./compiler/nativeGen/RegAlloc/Graph/Main.hs -15 +3
    M ./compiler/nativeGen/RegAlloc/Graph/TrivColorable.hs -2 +16
    M ./compiler/nativeGen/RegClass.hs +3
    M ./compiler/nativeGen/SPARC/Instr.hs +3
    M ./compiler/nativeGen/SPARC/Ppr.hs +1
    M ./compiler/nativeGen/SPARC/Regs.hs -7 +7
    M ./compiler/nativeGen/X86/CodeGen.hs -412 +380
    M ./compiler/nativeGen/X86/Instr.hs -61 +27
    M ./compiler/nativeGen/X86/Ppr.hs -43 +50
    M ./compiler/nativeGen/X86/RegInfo.hs -16 +18
    M ./compiler/nativeGen/X86/Regs.hs -165 +78
    M ./docs/users_guide/flags.xml +8
    M ./docs/users_guide/using.xml -1 +19

View patch online:
http://darcs.haskell.org/ghc/_darcs/patches/20100204104849-12142-a1cb7bcdfb7ec0aec38f612fad44ad0ab6dab294.gz