| 2006-12-12 |
Jim Laskey | Rollback changes to take a different tack. |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Chris Lattner | this can be trunc or bitcast, per line 3092. |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Chris Lattner | Fix regression on 400.perlbench last night. |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Rafael Espindola | more general matching of the MVN instruction |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Jim Laskey | Honor the command line specification for machine type. |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Rafael Espindola | don't use "ordinary" addressing mode 1 when mvn is... |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Jim Laskey | Reduce number of instructions to load 64-bit constants. |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Reid Spencer | Fix numerous inferred casts. |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Reid Spencer | Get even more accurate on the casting. |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Reid Spencer | Fix the casting for the computation of the Malloc size. |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Evan Cheng | Expand formal arguments and call arguments recursively... |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Reid Spencer | Don't create usless casts for same-bith-width floating... |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Chris Lattner | fit in 80 cols |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Chris Lattner | this can only be fptrunc. |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Chris Lattner | This case isn't needed with recent changes to ConstantI... |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Chris Lattner | split up inttoptr from ptrtoint handling, the cases... |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Chris Lattner | the operand of a bitcast is always the right size,... |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Reid Spencer | Change inferred getCast into specific getCast. Passes... |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Chris Lattner | Patch for PR1045 and Transforms/ScalarRepl/2006-12... |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Chris Lattner | new testcase for pr1045 |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Chris Lattner | Revert Nate's patch to fix X86/store-fp-constant.ll... |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Chris Lattner | make this code more aggressive about turning store... |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Chris Lattner | This should check for darwin literal sections as well. |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Chris Lattner | teach scev to analyze X*4|1 like X*4+c. This allows... |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Reid Spencer | Tidy up a bit. |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Reid Spencer | Generate the correct cast opcode for constant expressions. |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Reid Spencer | Change inferred cast creation calls to more specific... |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Rafael Espindola | use MVN to handle small negative constants |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Reid Spencer | Don't bother upgrading this. |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Reid Spencer | Implement getIntegerCast and getFPCast for ConstantExpr... |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Reid Spencer | Implement createIntegerCast and createFPCast factory... |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Reid Spencer | Fix PR1040: |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Rafael Espindola | add mvn |
commit | commitdiff | tree | snapshot |
| 2006-12-12 |
Reid Spencer | Fix the BitCastUnion type for 32-bit targets. |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Rafael Espindola | add note |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | Another step forward in PPC64 JIT support: we now no... |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Reid Spencer | Create the cast constant expression that was read inste... |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Reid Spencer | Test case for ensuring correct reading back of cast... |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | getInstrItineraryData shouldn't copy the itineraries |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Reid Spencer | Fix constant folding of FP->int due to cut & paste... |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Reid Spencer | Add a test for constant folding of bitcast to/from... |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Reid Spencer | Provide a test for int<->float and long<->double bitcasts. |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Reid Spencer | Implement correct bitcast of int<->float and long<... |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Evan Cheng | Re-apply changes that were backed out and fix a naughty... |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
John Criswell | It seems the llvm::OStream class does not handle stream... |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | Revert changes that broke oggenc on ppc |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Jim Laskey | Missing opcode. |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | Implement correct constant folding of bitcast. This... |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | constant folding of bitcast is broken. |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Jim Laskey | Layout proper frame for ppc64. |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Reid Spencer | Provide a script that can track down which optimization... |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Evan Cheng | Don't convert store double C, Ptr to store long C,... |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Evan Cheng | f32 / f64 node is expanded to one i32 / i64 node. |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Evan Cheng | Clean up some bad code. |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Nate Begeman | Move something that should be in the dag combiner from... |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Nate Begeman | Properly mangles symbol table names |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Nate Begeman | Add a default relocation type, and allow clients to... |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | fix typo |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | Update note, with the SROA change, we now produce: |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | trunc to integer, not to FP. |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | the xform is done by scalarrepl not the code generator. |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | this is done |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | Evan implemented the machineinstr improvements. |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | implement promotion of unions containing two packed... |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | new testcase |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | clarify some comments, simplify some checks, fix: |
commit | commitdiff | tree | snapshot |
| 2006-12-11 |
Chris Lattner | testcase for a regression introduced by the recent... |
commit | commitdiff | tree | snapshot |
| 2006-12-10 |
Chris Lattner | * Eliminate calls to CastInst::createInferredCast. |
commit | commitdiff | tree | snapshot |
| 2006-12-10 |
Chris Lattner | new testcase |
commit | commitdiff | tree | snapshot |
| 2006-12-10 |
Reid Spencer | Regenerate. |
commit | commitdiff | tree | snapshot |
| 2006-12-10 |
Anton Korobeynikov | Cleaned setjmp/longjmp lowering interfaces. Now we... |
commit | commitdiff | tree | snapshot |
| 2006-12-10 |
Chris Lattner | make statistics and timing info print even if the JIT... |
commit | commitdiff | tree | snapshot |
| 2006-12-10 |
Jim Laskey | Reverting until finding the cause of secondary bugs. |
commit | commitdiff | tree | snapshot |
| 2006-12-10 |
Jim Laskey | __PPC64CompilationCallback code was allowing registers... |
commit | commitdiff | tree | snapshot |
| 2006-12-10 |
Chris Lattner | fix PR1039 by making timing info be destroyed by llvm_s... |
commit | commitdiff | tree | snapshot |
| 2006-12-10 |
Reid Spencer | Don't bother with setting the path before running llvm... |
commit | commitdiff | tree | snapshot |
| 2006-12-10 |
Reid Spencer | For PR739: |
commit | commitdiff | tree | snapshot |
| 2006-12-10 |
Rafael Espindola | .align is in bits |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Rafael Espindola | fix test/Regression/CodeGen/X86/weak.ll |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Rafael Espindola | use a . instead of a % in the grep expression |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Reid Spencer | Regenerate. |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Reid Spencer | Fix test/Regression/Assembler/2006-12-09-Cast-To-Bool.ll |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Reid Spencer | Fix the run line and make this test more specific. |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Reid Spencer | Add a check to catch an incorrect upgrade. |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Reid Spencer | Regenerate. |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Reid Spencer | When upgrading cast to bool to a setne, generate icmp... |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Reid Spencer | Too soon to commit, revert. |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Reid Spencer | New test cases for regressions in the SetCC -> ICmp... |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Reid Spencer | Allow the input of the test program to be specified. |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Evan Cheng | Preliminary soft float support. |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Evan Cheng | Added option -soft-float to generate SW fp library... |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Bill Wendling | Fixed some formatting. |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Bill Wendling | Another example of using the llvm IO streams. |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Bill Wendling | Add documentation for how to use the new LLVM streams. |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Devang Patel | Analysis resolver now manages AnalysisImpls for the... |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Devang Patel | Top level pass manager keeps track of other managers... |
commit | commitdiff | tree | snapshot |
| 2006-12-09 |
Devang Patel | Use analysis resolver to find the info. |
commit | commitdiff | tree | snapshot |
| 2006-12-08 |
Devang Patel | Do not drop ImmutablePasses on the floor. |
commit | commitdiff | tree | snapshot |
| 2006-12-08 |
Devang Patel | Set AnalysisResolver for the passes when they are inser... |
commit | commitdiff | tree | snapshot |
| 2006-12-08 |
Devang Patel | Add AnalysisResolver_New. |
commit | commitdiff | tree | snapshot |
| next |
projects / oota-llvm.git / shortlog