- Feb 25, 2008
-
-
Dale Johannesen authored
llvm-svn: 47573
-
Dan Gohman authored
and StoreSDNode into LSBaseSDNode. llvm-svn: 47570
-
Dan Gohman authored
{S,U}MUL_LOHI with an unused high value. llvm-svn: 47569
-
Dan Gohman authored
result into a MUL late in the X86 codegen process. ISD::MUL is once again Legal on X86, so this is no longer needed. And, the hack was suboptimal; see PR1874 for details. llvm-svn: 47567
-
Dan Gohman authored
llvm-svn: 47566
-
Chris Lattner authored
llvm-svn: 47563
-
Dan Gohman authored
a SignBitIsZero function to simplify a common use case. llvm-svn: 47561
-
Chris Lattner authored
llvm-svn: 47558
-
Evan Cheng authored
All remat'ed loads cannot be folded into two-address code. Not just argument loads. This change doesn't really have any impact on codegen. llvm-svn: 47557
-
Chris Lattner authored
llvm-svn: 47556
-
Chris Lattner authored
llvm-svn: 47555
-
Dale Johannesen authored
of TokenFactor underneath chain (seems to be enough) llvm-svn: 47554
-
Dan Gohman authored
that combines getValueType and MVT::getSizeInBits, since this occurrs frequently. llvm-svn: 47552
-
Duncan Sands authored
result and operand types are legal. llvm-svn: 47546
-
Evan Cheng authored
llvm-svn: 47545
-
Owen Anderson authored
not safe. This is fixed by more aggressively checking that the return slot is not used elsewhere in the function. llvm-svn: 47544
-
Evan Cheng authored
llvm-svn: 47542
-
Owen Anderson authored
Fix an issue where GVN would try to use an instruction before its definition when performing return slot optimization. llvm-svn: 47541
-
- Feb 24, 2008
-
-
Duncan Sands authored
out of illegal elements (BUILD_VECTOR). Uses and beefs up BUILD_PAIR, though it didn't really have to. Like most of LegalizeTypes, does not support soft-float. This cures all "make check" vector building failures. llvm-svn: 47537
-
Bill Wendling authored
%r3 on PPC) in their ASM files. However, it's hard for humans to read during debugging. Adding a new field to the register data that lets you specify a different name to be printed than the one that goes into the ASM file -- %x3 instead of %r3, for instance. llvm-svn: 47534
-
- Feb 23, 2008
-
-
Evan Cheng authored
llvm-svn: 47533
-
Scott Michel authored
for CellSPU modifications: - SPUInstrInfo.td refactoring: "multiclass" really is _your_ friend. - Other improvements based on refactoring effort in SPUISelLowering.cpp, esp. in SPUISelLowering::PerformDAGCombine(), where zero amount shifts and rotates are now eliminiated, other scalar-to-vector-to-scalar silliness is also eliminated. - 64-bit operations are being implemented, _muldi3.c gcc runtime now compiles and generates the right code. More work still needs to be done. llvm-svn: 47532
-
Zhou Sheng authored
llvm-svn: 47531
-
Evan Cheng authored
llvm-svn: 47529
-
Evan Cheng authored
llvm-svn: 47527
-
Evan Cheng authored
If remating a machine instr with virtual register operand, make sure the vr is avaliable at all uses regardless of whether it would be folded. llvm-svn: 47526
-
Evan Cheng authored
llvm-svn: 47525
-
Evan Cheng authored
llvm-svn: 47524
-
Dan Gohman authored
invalid command-line options. llvm-svn: 47523
-
Evan Cheng authored
No need recognize load from a fixed argument slot as re-materializable. LiveIntervalAnalysis already handles it as a special case. llvm-svn: 47522
-
Devang Patel authored
llvm-svn: 47521
-
Evan Cheng authored
Recognize loads of arguments as re-materializable first. Therefore if isReallyTriviallyReMaterializable() returns true it doesn't confuse it as a "normal" re-materializable instruction. llvm-svn: 47520
-
Devang Patel authored
llvm-svn: 47519
-
Devang Patel authored
llvm-svn: 47518
-
Ted Kremenek authored
llvm-svn: 47517
-
Devang Patel authored
llvm-svn: 47516
-
Devang Patel authored
llvm-svn: 47514
-
Evan Cheng authored
llvm-svn: 47513
-
Devang Patel authored
llvm-svn: 47511
-
Evan Cheng authored
llvm-svn: 47510
-