| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| ... | |
| * | Change MRegisterDesc::AliasSet, TargetInstrDescriptor::ImplicitDefs | Alkis Evlogimenos | 2003-10-08 | 4 | -44/+54 |
| * | whoops, don't accidentally lose variable names | Chris Lattner | 2003-10-07 | 1 | -1/+1 |
| * | Fix bug: InstCombine/cast.ll:test11 / PR#7 | Chris Lattner | 2003-10-07 | 1 | -1/+23 |
| * | Refactor code a bit | Chris Lattner | 2003-10-07 | 1 | -4/+12 |
| * | Add a sanity check for constant expression casts | Chris Lattner | 2003-10-07 | 1 | -0/+2 |
| * | Fix bugzilla bug #5 | Chris Lattner | 2003-10-07 | 1 | -1/+6 |
| * | Bill contributed this major rewrite of the -lowerswitch pass to make it generate | Chris Lattner | 2003-10-07 | 1 | -44/+144 |
| * | Fix bug in previous checkin | Chris Lattner | 2003-10-07 | 1 | -0/+1 |
| * | Actually _PASS IN_ NO_RESERVE if we have it. | Chris Lattner | 2003-10-06 | 1 | -1/+1 |
| * | ToolRunner.h has been moved from include/Support to include/llvm/Support. | Misha Brukman | 2003-10-06 | 1 | -1/+1 |
| * | Users can never be null | Chris Lattner | 2003-10-06 | 1 | -1/+1 |
| * | Minor speedups for the instcombine pass | Chris Lattner | 2003-10-06 | 1 | -17/+26 |
| * | Speed up the predicate used to decide when to inline by caching the size | Chris Lattner | 2003-10-06 | 1 | -13/+28 |
| * | Add # of printed instructions statistic to both the SPARC and X86 LLC backends. | Brian Gaeke | 2003-10-06 | 3 | -1/+12 |
| * | Avoid doing pointless work. Amazingly, this makes us go faster. | Chris Lattner | 2003-10-06 | 2 | -8/+5 |
| * | Implement the NamedRegionTimer class | Chris Lattner | 2003-10-06 | 1 | -0/+18 |
| * | All of our supported operating systems (so far) and FreeBSD technically | Brian Gaeke | 2003-10-06 | 1 | -0/+1 |
| * | This changes the PromoteMemToReg function to create "pruned" SSA form, not | Chris Lattner | 2003-10-05 | 1 | -24/+109 |
| * | Change the interface to PromoteMemToReg to also take a DominatorTree | Chris Lattner | 2003-10-05 | 4 | -5/+14 |
| * | Speed up the mem2reg transform for allocas which are only read/written in a s... | Chris Lattner | 2003-10-05 | 1 | -5/+84 |
| * | Move support/lib/Support back to lib/Support | Chris Lattner | 2003-10-05 | 1 | -1/+1 |
| * | Add support for the Invoke instruction by using the LowerInvoke pass | Chris Lattner | 2003-10-05 | 1 | -0/+6 |
| * | Instead of hacking in custom support for Invoke/Unwind, use the LowerInvoke pass | Chris Lattner | 2003-10-05 | 3 | -62/+14 |
| * | Initial checkin of the LLVM->LLVM transform to support code generators which | Chris Lattner | 2003-10-05 | 1 | -0/+74 |
| * | simplify-cfg is really a function pass | Chris Lattner | 2003-10-05 | 1 | -1/+1 |
| * | Be more careful handling PHI nodes, which might be of potentially high degree. | Chris Lattner | 2003-10-05 | 1 | -93/+80 |
| * | The first PHI node may be null, scan for the first non-null one | Chris Lattner | 2003-10-05 | 1 | -1/+4 |
| * | The VersionNumbers vector is only used during PHI placement. Turn it into an... | Chris Lattner | 2003-10-05 | 1 | -10/+6 |
| * | * Update file header comment | Chris Lattner | 2003-10-05 | 1 | -44/+64 |
| * | Simplify the loop a bit | Chris Lattner | 2003-10-05 | 1 | -10/+9 |
| * | There is no need for separate WriteSets and PhiNodeBlocks lists. It is just a | Chris Lattner | 2003-10-05 | 1 | -18/+8 |
| * | The PhiNodes 2D vector is only used during PHI node placement. It doesn't | Chris Lattner | 2003-10-05 | 1 | -11/+10 |
| * | * Document instance vars better | Chris Lattner | 2003-10-05 | 1 | -20/+26 |
| * | Two small cleanups/speedups: | Chris Lattner | 2003-10-05 | 1 | -29/+31 |
| * | * Minor cleanups | Chris Lattner | 2003-10-05 | 1 | -42/+28 |
| * | A couple of minor code cleanups. | Chris Lattner | 2003-10-05 | 2 | -42/+32 |
| * | Type tables are now AbstractTypeUsers. This allows them to merge together | Chris Lattner | 2003-10-05 | 1 | -135/+217 |
| * | Rename AbstractModuleProvider -> ModuleProvider, to match the header file name, | Chris Lattner | 2003-10-04 | 3 | -10/+8 |
| * | Transform two methods to return pointers directly instead of returning them | Chris Lattner | 2003-10-04 | 3 | -66/+42 |
| * | Use V for values, not D. | Chris Lattner | 2003-10-04 | 1 | -16/+15 |
| * | Do not leak the ModuleProvider if releaseModule() throws. | Chris Lattner | 2003-10-04 | 1 | -11/+6 |
| * | Minor random speedup: make Constant::getNullValue only call ConstantFOO::get | Chris Lattner | 2003-10-03 | 1 | -13/+45 |
| * | Explicit copy ctors are no longer needed | Chris Lattner | 2003-10-03 | 1 | -21/+0 |
| * | Minor cleanups | Chris Lattner | 2003-10-03 | 1 | -31/+17 |
| * | This checkin basically amounts to a complete rewrite of the type-resolution | Chris Lattner | 2003-10-03 | 2 | -284/+138 |
| * | These methods are dead, remove them. Because the bodies will soon be | Chris Lattner | 2003-10-03 | 1 | -2/+8 |
| * | Dramatically simplify DerivedType::refineAbstractTypeToInternal | Chris Lattner | 2003-10-03 | 1 | -44/+25 |
| * | Make the PATypeHolder use a simple union-find implementation to handle | Chris Lattner | 2003-10-02 | 1 | -22/+30 |
| * | There is no need for BytecodeParser to be an AbstractTypeUser. Instead, it | Chris Lattner | 2003-10-02 | 2 | -28/+3 |
| * | There is no reason for Value to be an AbstractTypeUser. This just makes things | Chris Lattner | 2003-10-02 | 2 | -18/+2 |
