[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-12-05 Thread Chris Lattner
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.303 -> 1.304 X86InstrSSE.td updated: 1.172 -> 1.173 --- Log message: If we have ScalarSSE, we can select bitconvert into single instructions. This compiles bitcast.ll:test3/test4 into: _test3: movd %xmm0, %eax

[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-06-15 Thread Evan Cheng
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.229 -> 1.230 X86InstrSSE.td updated: 1.122 -> 1.123 --- Log message: Type of vector extract / insert index operand should be iPTR. --- Diffs of the changes: (+18 -18) X86ISelLowering.cpp | 10 +- X86InstrSSE.

[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-05-05 Thread Chris Lattner
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.198 -> 1.199 X86InstrSSE.td updated: 1.115 -> 1.116 --- Log message: Teach the code generator to use cvtss2sd as extload f32 -> f64 --- Diffs of the changes: (+1 -5) X86ISelLowering.cpp |4 X86InstrSSE.td

[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-04-17 Thread Evan Cheng
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.172 -> 1.173 X86InstrSSE.td updated: 1.99 -> 1.100 --- Log message: Use movss to insert_vector_elt(v, s, 0). --- Diffs of the changes: (+37 -19) X86ISelLowering.cpp | 51 -

[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-04-12 Thread Evan Cheng
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.162 -> 1.163 X86InstrSSE.td updated: 1.78 -> 1.79 --- Log message: Promote v4i32, v8i16, v16i8 load to v2i64 load. --- Diffs of the changes: (+41 -57) X86ISelLowering.cpp |7 +--- X86InstrSSE.td | 91 ++

[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-04-10 Thread Evan Cheng
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.160 -> 1.161 X86InstrSSE.td updated: 1.68 -> 1.69 --- Log message: Conditional move of vector types. --- Diffs of the changes: (+63 -37) X86ISelLowering.cpp | 85 +---

[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-04-03 Thread Evan Cheng
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.152 -> 1.153 X86InstrSSE.td updated: 1.54 -> 1.55 --- Log message: Use movlpd to: store lower f64 extracted from v2f64. Use movhpd to: store upper f64 extracted from v2f64. --- Diffs of the changes: (+10 -6) X86ISelLo

[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-03-30 Thread Evan Cheng
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.145 -> 1.146 X86InstrSSE.td updated: 1.49 -> 1.50 --- Log message: Make sure all possible shuffles are matched. Use pshufd, pshuhw, and pshulw to shuffle v4f32 if shufps doesn't match. Use shufps to shuffle v4f32 if pshufd

[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-03-29 Thread Evan Cheng
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.143 -> 1.144 X86InstrSSE.td updated: 1.43 -> 1.44 --- Log message: Need to special case splat after all. Make the second operand of splat vector_shuffle undef. --- Diffs of the changes: (+18 -15) X86ISelLowering.cpp |

[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-03-28 Thread Evan Cheng
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.142 -> 1.143 X86InstrSSE.td updated: 1.41 -> 1.42 --- Log message: - More shuffle related bug fixes. - Whenever possible use ops of the right packed types for vector shuffles / splats. --- Diffs of the changes: (+30 -

[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-03-28 Thread Evan Cheng
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.141 -> 1.142 X86InstrSSE.td updated: 1.40 -> 1.41 --- Log message: - Only use pshufd for v4i32 vector shuffles. - Other shuffle related fixes. --- Diffs of the changes: (+83 -61) X86ISelLowering.cpp | 51 +++

[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-03-26 Thread Evan Cheng
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.134 -> 1.135 X86InstrSSE.td updated: 1.32 -> 1.33 --- Log message: Use pcmpeq to generate vector of all ones. --- Diffs of the changes: (+15 -0) X86ISelLowering.cpp |4 X86InstrSSE.td | 11 +++

[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-03-24 Thread Evan Cheng
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.130 -> 1.131 X86InstrSSE.td updated: 1.28 -> 1.29 --- Log message: Added 128-bit packed integer subtraction. --- Diffs of the changes: (+26 -0) X86ISelLowering.cpp |3 +++ X86InstrSSE.td | 23 ++

[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-03-22 Thread Evan Cheng
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.122 -> 1.123 X86InstrSSE.td updated: 1.18 -> 1.19 --- Log message: Add 128-bit integer vector load and add (for testing). --- Diffs of the changes: (+51 -5) X86ISelLowering.cpp |8 +++- X86InstrSSE.td |

[llvm-commits] CVS: llvm/lib/Target/X86/X86ISelLowering.cpp X86InstrSSE.td

2006-03-22 Thread Evan Cheng
Changes in directory llvm/lib/Target/X86: X86ISelLowering.cpp updated: 1.119 -> 1.120 X86InstrSSE.td updated: 1.16 -> 1.17 --- Log message: - Supposely movlhps is faster / better than unpcklpd. - Don't forget pshufd is only available with sse2. --- Diffs of the changes: (+16 -11) X86ISelLo