2005-10-14 |
Chris Lattner | Rename PPC32*.h to PPC*.h |
blob | commitdiff | raw |
2005-10-14 |
Chris Lattner | Rename PowerPC*.h to PPC*.h |
blob | commitdiff | raw | diff to current |
2005-10-14 |
Chris Lattner | Rename PowerPCInstrBuilder.h -> PPC* |
blob | commitdiff | raw | diff to current |
2005-10-14 |
Chris Lattner | Nuke PowerPCInstrFormats.h, its contents are dead.... |
blob | commitdiff | raw | diff to current |
2005-10-02 |
Chris Lattner | another solution to the fsel issue. Instead of having... |
blob | commitdiff | raw | diff to current |
2005-10-02 |
Chris Lattner | fsel can take a different FP type for the comparison... |
blob | commitdiff | raw | diff to current |
2005-10-01 |
Chris Lattner | Modify the ppc backend to use two register classes... |
blob | commitdiff | raw | diff to current |
2005-09-29 |
Chris Lattner | consistency with other cases, no functionality change |
blob | commitdiff | raw | diff to current |
2005-09-28 |
Chris Lattner | Add FP versions of the binary operators, keeping the... |
blob | commitdiff | raw | diff to current |
2005-09-10 |
Chris Lattner | PowerPC cannot truncstore i1 natively |
blob | commitdiff | raw | diff to current |
2005-09-06 |
Nate Begeman | Implement i64<->fp using the fctidz/fcfid instructions... |
blob | commitdiff | raw | diff to current |
2005-08-31 |
Chris Lattner | Move FCTIWZ handling out of the instruction selectors... |
blob | commitdiff | raw | diff to current |
2005-08-31 |
Chris Lattner | Remove dead code |
blob | commitdiff | raw | diff to current |
2005-08-31 |
Chris Lattner | Remove code that is now dead from the pattern isel. |
blob | commitdiff | raw | diff to current |
2005-08-31 |
Chris Lattner | Handle AssertSext/AssertZext nodes, fixing the regressi... |
blob | commitdiff | raw | diff to current |
2005-08-26 |
Chris Lattner | Make fsel emission work with both the pattern and dag... |
blob | commitdiff | raw | diff to current |
2005-08-26 |
Nate Begeman | Remove some code made dead by the fsel patch |
blob | commitdiff | raw | diff to current |
2005-08-26 |
Chris Lattner | now that fsel is formed during legalization, this code... |
blob | commitdiff | raw | diff to current |
2005-08-26 |
Chris Lattner | Change ConstantPoolSDNode to actually hold the Constant... |
blob | commitdiff | raw | diff to current |
2005-08-26 |
Chris Lattner | Fix some warnings in an optimized build |
blob | commitdiff | raw | diff to current |
2005-08-26 |
Chris Lattner | add initial support for converting select_cc -> fsel... |
blob | commitdiff | raw | diff to current |
2005-08-25 |
Chris Lattner | simplify the add/sub_parts code |
blob | commitdiff | raw | diff to current |
2005-08-25 |
Chris Lattner | Simplify some code. It's not clear why the UDIV expand... |
blob | commitdiff | raw | diff to current |
2005-08-25 |
Chris Lattner | fit in 80 cols |
blob | commitdiff | raw | diff to current |
2005-08-24 |
Chris Lattner | Split IMPLICIT_DEF into IMPLICIT_DEF_GPR and IMPLICIT_D... |
blob | commitdiff | raw | diff to current |
2005-08-24 |
Chris Lattner | Remove some dead cases. |
blob | commitdiff | raw | diff to current |
2005-08-24 |
Nate Begeman | Whoops, fix a thinko. All cases except SETNE are now... |
blob | commitdiff | raw | diff to current |
2005-08-24 |
Nate Begeman | Add the "ppc specific" setcc-equivalent select_cc cases |
blob | commitdiff | raw | diff to current |
2005-08-22 |
Chris Lattner | Make sure expressions only have one use before emitting... |
blob | commitdiff | raw | diff to current |
2005-08-21 |
Chris Lattner | Simplify the logic for BRTWOWAY_CC handling. The isel... |
blob | commitdiff | raw | diff to current |
2005-08-21 |
Chris Lattner | If the false value for a select_cc is really simple... |
blob | commitdiff | raw | diff to current |
2005-08-19 |
Nate Begeman | Fix a bug where we were passing the wrong number of... |
blob | commitdiff | raw | diff to current |
2005-08-18 |
Chris Lattner | MFLR doesn't take an operand, the LR register is implicit |
blob | commitdiff | raw | diff to current |
2005-08-18 |
Jim Laskey | More optimal solution for loading constants. |
blob | commitdiff | raw | diff to current |
2005-08-18 |
Nate Begeman | Fix int foo() { return 65535; } by using the top 16... |
blob | commitdiff | raw | diff to current |
2005-08-18 |
Nate Begeman | Improve ISD::Constant codegen. |
blob | commitdiff | raw | diff to current |
2005-08-18 |
Chris Lattner | replace switch stmt with an assert, generate li 0 inste... |
blob | commitdiff | raw | diff to current |
2005-08-18 |
Jim Laskey | Handle loading of 0x????0000 constants with a single... |
blob | commitdiff | raw | diff to current |
2005-08-18 |
Jim Laskey | Better version of isIntImmediate. |
blob | commitdiff | raw | diff to current |
2005-08-17 |
Chris Lattner | Fix a few small typos I noticed when converting this... |
blob | commitdiff | raw | diff to current |
2005-08-17 |
Jim Laskey | Removed UINT_TO_FP and SINT_TO_FP from ISel outright. |
blob | commitdiff | raw | diff to current |
2005-08-17 |
Jim Laskey | Remove ISel code generation for UINT_TO_FP and SINT_TO_... |
blob | commitdiff | raw | diff to current |
2005-08-17 |
Nate Begeman | Implement a couple improvements: |
blob | commitdiff | raw | diff to current |
2005-08-16 |
Chris Lattner | updates for changes in nodes |
blob | commitdiff | raw | diff to current |
2005-08-16 |
Nate Begeman | Implement BR_CC and BRTWOWAY_CC. This allows the remov... |
blob | commitdiff | raw | diff to current |
2005-08-16 |
Chris Lattner | Pull the LLVM -> DAG lowering code out of the pattern... |
blob | commitdiff | raw | diff to current |
2005-08-15 |
Jim Laskey | Broke 80 column rule. |
blob | commitdiff | raw | diff to current |
2005-08-15 |
Jim Laskey | Changed code gen for int to f32 to use rounding. This... |
blob | commitdiff | raw | diff to current |
2005-08-14 |
Nate Begeman | Fix last night's PPC32 regressions by |
blob | commitdiff | raw | diff to current |
2005-08-14 |
Nate Begeman | Make FP_TO_UINT Illegal. This allows us to generate... |
blob | commitdiff | raw | diff to current |
2005-08-12 |
Jim Laskey | Fix for 2005-08-12-rlwimi-crash.ll. Make allowance... |
blob | commitdiff | raw | diff to current |
2005-08-12 |
Jim Laskey | 1. This changes handles the cases of (~x)&y and x&... |
blob | commitdiff | raw | diff to current |
2005-08-11 |
Jim Laskey | 1. Added the function isOpcWithIntImmediate to simplify... |
blob | commitdiff | raw | diff to current |
2005-08-11 |
Chris Lattner | Tidied up the use of dyn_cast<ConstantSDNode> by using... |
blob | commitdiff | raw | diff to current |
2005-08-11 |
Chris Lattner | Use a more efficient method of creating integer and... |
blob | commitdiff | raw | diff to current |
2005-08-10 |
Nate Begeman | Make SELECT illegal on PPC32, switch to using SELECT_CC... |
blob | commitdiff | raw | diff to current |
2005-08-10 |
Chris Lattner | Changes for PPC32ISelPattern.cpp |
blob | commitdiff | raw | diff to current |
2005-08-10 |
Chris Lattner | Changed the XOR case to use the isOprNot predicate. |
blob | commitdiff | raw | diff to current |
2005-08-10 |
Chris Lattner | 1. Refactored handling of integer immediate values... |
blob | commitdiff | raw | diff to current |
2005-08-10 |
Chris Lattner | Fix a bug compiling: select (i32 < i32), f32, f32 |
blob | commitdiff | raw | diff to current |
2005-08-09 |
Chris Lattner | Update the targets to the new SETCC/CondCodeSDNode... |
blob | commitdiff | raw | diff to current |
2005-08-09 |
Chris Lattner | Minor cleanup patch, no functionality changes. Written... |
blob | commitdiff | raw | diff to current |
2005-08-09 |
Chris Lattner | Fix CodeGen/Generic/div-neg-power-2.ll, a regression... |
blob | commitdiff | raw | diff to current |
2005-08-08 |
Nate Begeman | Factor out some common code, and be smarter about when... |
blob | commitdiff | raw | diff to current |
2005-08-08 |
Chris Lattner | Remove getImmediateForOpcode, which is now dead. |
blob | commitdiff | raw | diff to current |
2005-08-08 |
Chris Lattner | Add new immediate handling support for mul/div. |
blob | commitdiff | raw | diff to current |
2005-08-08 |
Chris Lattner | Add support for OR/XOR/SUB immediates that are handled... |
blob | commitdiff | raw | diff to current |
2005-08-08 |
Chris Lattner | Modify the ISD::AND opcode case to use new immediate... |
blob | commitdiff | raw | diff to current |
2005-08-08 |
Chris Lattner | Modify the ISD::ADD opcode case to use new immediate... |
blob | commitdiff | raw | diff to current |
2005-08-08 |
Chris Lattner | Modify existing support functions to use new immediate... |
blob | commitdiff | raw | diff to current |
2005-08-08 |
Chris Lattner | Add support predicates for future immediate constant... |
blob | commitdiff | raw | diff to current |
2005-08-08 |
Chris Lattner | Move IsRunOfOnes to a more logical place and rename... |
blob | commitdiff | raw | diff to current |
2005-08-05 |
Chris Lattner | Consolidate the GPOpt stuff to all use the Subtarget... |
blob | commitdiff | raw | diff to current |
2005-08-03 |
Misha Brukman | Fix grammar: apostrophe-s ('s) is possessive, not plura... |
blob | commitdiff | raw | diff to current |
2005-08-03 |
Chris Lattner | minor capitalization thing, patch by Jim Laskey |
blob | commitdiff | raw | diff to current |
2005-08-02 |
Chris Lattner | This hunk accidentally got dropped. Patch by Jim Laskey |
blob | commitdiff | raw | diff to current |
2005-08-02 |
Chris Lattner | Update to use the new MathExtras.h support for log2... |
blob | commitdiff | raw | diff to current |
2005-08-02 |
Chris Lattner | add a pass name to make debugging dumps nicer |
blob | commitdiff | raw | diff to current |
2005-07-28 |
Chris Lattner | Eliminate an extra copy from R1 that Nate noticed on... |
blob | commitdiff | raw | diff to current |
2005-07-28 |
Chris Lattner | Specify the correct number of operands |
blob | commitdiff | raw | diff to current |
2005-07-28 |
Nate Begeman | Fold constant adds into loads and stores to frame indices. |
blob | commitdiff | raw | diff to current |
2005-07-27 |
Nate Begeman | Fix some comments |
blob | commitdiff | raw | diff to current |
2005-07-27 |
Jeff Cohen | Eliminate all remaining tabs and trailing spaces. |
blob | commitdiff | raw | diff to current |
2005-07-25 |
Nate Begeman | Fix an optimization put in for accessing static globals... |
blob | commitdiff | raw | diff to current |
2005-07-22 |
Chris Lattner | PowerPC no-pic code is not quite ready for prime-time |
blob | commitdiff | raw | diff to current |
2005-07-21 |
Nate Begeman | Support building non-PIC |
blob | commitdiff | raw | diff to current |
2005-07-20 |
Nate Begeman | Generate mfocrf when targeting g5. Generate fsqrt... |
blob | commitdiff | raw | diff to current |
2005-07-19 |
Nate Begeman | Integrate SelectFPExpr into SelectExpr. This gets... |
blob | commitdiff | raw | diff to current |
2005-07-10 |
Chris Lattner | Change *EXTLOAD to use an VTSDNode operand instead... |
blob | commitdiff | raw | diff to current |
2005-07-10 |
Chris Lattner | Change TRUNCSTORE to use a VTSDNode operand instead... |
blob | commitdiff | raw | diff to current |
2005-07-05 |
Chris Lattner | Make several cleanups to Andrews varargs change: |
blob | commitdiff | raw | diff to current |
2005-07-05 |
Chris Lattner | Fix PowerPC varargs |
blob | commitdiff | raw | diff to current |
2005-07-01 |
Chris Lattner | Varargs is apparently currently broken on PPC. This... |
blob | commitdiff | raw | diff to current |
2005-06-15 |
Nate Begeman | Commit fix for generating conditional branch pseudo... |
blob | commitdiff | raw | diff to current |
2005-06-14 |
Nate Begeman | Commit a small improvement that is already in the x86... |
blob | commitdiff | raw | diff to current |
2005-06-08 |
Nate Begeman | Handle some more real world cases of rlwimi. These... |
blob | commitdiff | raw | diff to current |
2005-05-15 |
Chris Lattner | Fix andrews changes to fit in 80 columns |
blob | commitdiff | raw | diff to current |
2005-05-13 |
Chris Lattner | treat TAILCALL nodes identically to CALL nodes |
blob | commitdiff | raw | diff to current |
2005-05-13 |
Chris Lattner | Add an isTailCall flag to LowerCallTo |
blob | commitdiff | raw | diff to current |
2005-05-13 |
Chris Lattner | Realize that we don't support fmod directly, fixing... |
blob | commitdiff | raw | diff to current |
next |