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.

1 files changed, 3 insertions, 15 deletions

diff --git a/compiler/nativeGen/RegAlloc/Graph/Main.hs b/compiler/nativeGen/RegAlloc/Graph/Main.hs
index 5fa771cfb8..35ec879eb2 100644
--- a/compiler/nativeGen/RegAlloc/Graph/Main.hs
+++ b/compiler/nativeGen/RegAlloc/Graph/Main.hs
@@ -380,25 +380,13 @@ seqNode node
 	`seq` (seqVirtualRegList (uniqSetToList (Color.nodeCoalesce node)))
 
 seqVirtualReg :: VirtualReg -> ()
-seqVirtualReg reg
- = case reg of
-	VirtualRegI _	-> ()
-	VirtualRegHi _	-> ()
-	VirtualRegF _	-> ()
-	VirtualRegD _	-> ()
+seqVirtualReg reg = reg `seq` ()
 
 seqRealReg :: RealReg -> ()
-seqRealReg reg
- = case reg of
- 	RealRegSingle _ -> ()
-	RealRegPair _ _	-> ()
+seqRealReg reg = reg `seq` ()
 
 seqRegClass :: RegClass -> ()
-seqRegClass c
- = case c of
- 	RcInteger	-> ()
-	RcFloat		-> ()
-	RcDouble	-> ()
+seqRegClass c = c `seq` ()
 
 seqMaybeRealReg :: Maybe RealReg -> ()
 seqMaybeRealReg mr