2005-01-25 |
Chris Lattner | We can fold promoted and non-promoted loads into divs... |
tree | commitdiff |
2005-01-25 |
Chris Lattner | Fold promoted loads into binary ops for FP, allowing... |
tree | commitdiff |
2005-01-23 |
Chris Lattner | Silence a warning. |
tree | commitdiff |
2005-01-23 |
Chris Lattner | Allow the FP stackifier to completely ignore functions... |
tree | commitdiff |
2005-01-23 |
Reid Spencer | Support Cygwin assembly generation. The cygwin version... |
tree | commitdiff |
2005-01-21 |
Chris Lattner | Speed up folding operations into loads. |
tree | commitdiff |
2005-01-21 |
Chris Lattner | The ever-important vanity pass name :) |
tree | commitdiff |
2005-01-21 |
Chris Lattner | Fix a FIXME: realize that argument stores are all indep... |
tree | commitdiff |
2005-01-20 |
Chris Lattner | Implement ADD_PARTS/SUB_PARTS so that 64-bit integer... |
tree | commitdiff |
2005-01-20 |
Chris Lattner | Fix a crash compiling 134.perl. |
tree | commitdiff |
2005-01-19 |
Chris Lattner | Fix a problem where were were literally selecting for... |
tree | commitdiff |
2005-01-19 |
Chris Lattner | When commuting these instructions, make sure to actuall... |
tree | commitdiff |
2005-01-19 |
Chris Lattner | Implement Regression/CodeGen/X86/rotate.ll: emit rotate... |
tree | commitdiff |
2005-01-19 |
Chris Lattner | Add rotate instructions. |
tree | commitdiff |
2005-01-19 |
Chris Lattner | Match 16-bit shld/shrd instructions as well, implementi... |
tree | commitdiff |
2005-01-19 |
Chris Lattner | Improve coverage of the X86 instruction set by adding... |
tree | commitdiff |
2005-01-19 |
Chris Lattner | Teach the code generator that shrd/shld is commutable... |
tree | commitdiff |
2005-01-19 |
Chris Lattner | Codegen long >> 2 to this: |
tree | commitdiff |
2005-01-19 |
Chris Lattner | X86 shifts mask the amount. |
tree | commitdiff |
2005-01-18 |
Chris Lattner | Code to handle FP_EXTEND is dead now. X86 doesn't... |
tree | commitdiff |
2005-01-18 |
Chris Lattner | Remove more dead code. |
tree | commitdiff |
2005-01-18 |
Chris Lattner | The selection dag code handles the promotions from... |
tree | commitdiff |
2005-01-18 |
Chris Lattner | Fix 124.m88ksim. |
tree | commitdiff |
2005-01-18 |
Chris Lattner | Do not emit loads multiple times, potentially in the... |
tree | commitdiff |
2005-01-18 |
Chris Lattner | Eliminate bad assertions. |
tree | commitdiff |
2005-01-18 |
Chris Lattner | * Eliminate the TokenSet and just use the ExprMap for... |
tree | commitdiff |
2005-01-18 |
Chris Lattner | Rely on the code in MatchAddress to do this work. ... |
tree | commitdiff |
2005-01-18 |
Chris Lattner | Fix a problem where probing for addressing modes caused... |
tree | commitdiff |
2005-01-17 |
Chris Lattner | Fix indentation. |
tree | commitdiff |
2005-01-17 |
Chris Lattner | Don't bother using max here. |
tree | commitdiff |
2005-01-17 |
Chris Lattner | Do not give token factor nodes outrageous weights |
tree | commitdiff |
2005-01-17 |
Chris Lattner | Two changes: |
tree | commitdiff |
2005-01-17 |
Chris Lattner | Refactor load/op/store folding into it's own method... |
tree | commitdiff |
2005-01-17 |
Chris Lattner | Fix a major regression last night that prevented us... |
tree | commitdiff |
2005-01-17 |
Chris Lattner | Codegen this: |
tree | commitdiff |
2005-01-17 |
Chris Lattner | Fix test/Regression/CodeGen/X86/2005-01-17-CycleInDAG... |
tree | commitdiff |
2005-01-17 |
Chris Lattner | Do not fold a load into a comparison that is used by... |
tree | commitdiff |
2005-01-17 |
Chris Lattner | Do not codegen 'xor bool, true' as 'not reg'. not... |
tree | commitdiff |
2005-01-17 |
Chris Lattner | Set up the shift and setcc types. |
tree | commitdiff |
2005-01-16 |
Chris Lattner | * Adjust to changes in TargetLowering interfaces. |
tree | commitdiff |
2005-01-15 |
Chris Lattner | Add support for truncstore and *extload. |
tree | commitdiff |
2005-01-14 |
Chris Lattner | Adjust to CopyFromREg changes. |
tree | commitdiff |
2005-01-13 |
Chris Lattner | Add new ImplicitDef node, rename CopyRegSDNode class... |
tree | commitdiff |
2005-01-13 |
Chris Lattner | Codegen factor nodes more intelligently according to... |
tree | commitdiff |
2005-01-13 |
Chris Lattner | Initial trivial (but stupid) codegen for this node. |
tree | commitdiff |
2005-01-13 |
Chris Lattner | Add some really pedantic assertions to the load folding... |
tree | commitdiff |
2005-01-12 |
Chris Lattner | We can only fold a load into an op if there is exactly... |
tree | commitdiff |
2005-01-12 |
Chris Lattner | Try both ways to fold an add together. This allows... |
tree | commitdiff |
2005-01-12 |
Chris Lattner | Fix a major miscompilation where we were overwriting... |
tree | commitdiff |
2005-01-12 |
Chris Lattner | Do not use the type of the RHS constant to determine... |
tree | commitdiff |
2005-01-12 |
Chris Lattner | Do not lose the offset from teh global when peephole... |
tree | commitdiff |
2005-01-12 |
Jeff Cohen | Fix C++ more compilatiom errors |
tree | commitdiff |
2005-01-12 |
Chris Lattner | Fix a compile error with VC++, which things that static... |
tree | commitdiff |
2005-01-12 |
Chris Lattner | Fix a bug that caused us to crash on povray. We weren... |
tree | commitdiff |
2005-01-12 |
Chris Lattner | Print a load of a null pointer (in intel mode) like... |
tree | commitdiff |
2005-01-12 |
Chris Lattner | Print a load of a null pointer like this: |
tree | commitdiff |
2005-01-12 |
Chris Lattner | Fix a crash compiling povray on UINT_TO_FP from i16. |
tree | commitdiff |
2005-01-12 |
Chris Lattner | There are no [mem] op= reg instructions for FP, so... |
tree | commitdiff |
2005-01-12 |
Chris Lattner | Fix a bug where we didn't insert FP_REG_KILL instructio... |
tree | commitdiff |
2005-01-12 |
Chris Lattner | Fold TRUNCATE (LOAD P) into a smaller load from P. |
tree | commitdiff |
2005-01-12 |
Chris Lattner | Be more careful about order of arg evalution for CopyTo... |
tree | commitdiff |
2005-01-12 |
Chris Lattner | Fold some more [mem] op= val operators. This allows... |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Fold loads into sign/zero extends. instead of: |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Comment out debug code :) |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Handle the global address case here, not just the offse... |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Treat int constants as not requiring a register, since... |
tree | commitdiff |
2005-01-11 |
Chris Lattner | * Factor a bunch of binary operator cases into shared... |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Fold multiplies by 3,5,9 into addressing modes when... |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Instead of generating stuff like this: |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Implement MEMCPY natively in terms of rep movs* |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Implement memset -> rep stos* |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Announce that we don't support mem ops yet. |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Teach the address selector to make 'reg+reg' addressing... |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Emit NOT instructions. |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Fix a bug emitting branches that broke a lot of programs. |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Be more careful where we set ContainsFPCode. We were... |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Fix a major bug in setcc/cmov folding, where we acciden... |
tree | commitdiff |
2005-01-11 |
Chris Lattner | Take register pressure into account when we have to... |
tree | commitdiff |
2005-01-10 |
Chris Lattner | Fold setcc instructions into selects. |
tree | commitdiff |
2005-01-10 |
Chris Lattner | Add conditional moves for the parity flag. |
tree | commitdiff |
2005-01-10 |
Chris Lattner | Implement 8-bit multiply for X86. |
tree | commitdiff |
2005-01-09 |
Chris Lattner | Codegen (Reg|imm)+&GV as an LEA, because we cannot... |
tree | commitdiff |
2005-01-09 |
Chris Lattner | Fix copy and pasto's for FP -> Int. This fixes fldry |
tree | commitdiff |
2005-01-09 |
Chris Lattner | Initial implementation of FP->INT and INT->FP casts |
tree | commitdiff |
2005-01-09 |
Chris Lattner | Fix a subtle bug involving constant expr casts from... |
tree | commitdiff |
2005-01-09 |
Chris Lattner | Implement varargs and returnaddress/frameaddress intrin... |
tree | commitdiff |
2005-01-08 |
Chris Lattner | Okay 15th time is the charm. Looking at the vector... |
tree | commitdiff |
2005-01-08 |
Chris Lattner | Okay, my off by one was actually off by two. This... |
tree | commitdiff |
2005-01-08 |
Chris Lattner | Fix off by one error |
tree | commitdiff |
2005-01-08 |
Chris Lattner | Adjust to changes in LowerCallTo interface |
tree | commitdiff |
2005-01-08 |
Chris Lattner | Wrap long line. |
tree | commitdiff |
2005-01-08 |
Chris Lattner | The X86 instruction selector already handles codegen of: |
tree | commitdiff |
2005-01-07 |
Chris Lattner | Allow the selection-dag based selector to be diabled... |
tree | commitdiff |
2005-01-07 |
Chris Lattner | Reimplementation of the X86 pattern isel. This is... |
tree | commitdiff |
2005-01-07 |
Chris Lattner | This file is now dead. |
tree | commitdiff |
2005-01-07 |
Chris Lattner | Add a new prototype |
tree | commitdiff |
2005-01-06 |
Chris Lattner | Codegen -1 and -0.0 more efficiently. This implements... |
tree | commitdiff |
2005-01-05 |
Chris Lattner | 1. If a double FP constant must be put into a constant... |
tree | commitdiff |
2005-01-05 |
Chris Lattner | Minor optimization to allocate R8 registers in a better... |
tree | commitdiff |
2005-01-03 |
Jeff Cohen | Revert elimination of global variable hack... still... |
tree | commitdiff |
next |