| 2009-03-25 |
Dan Gohman | Revert r67655 and r67656, as they are breaking the...
|
commit | commitdiff | tree |
| 2009-03-25 |
Dan Gohman | LLVM's master Makefile.common is in LLVM_SRC_ROOT,...
|
commit | commitdiff | tree |
| 2009-03-24 |
Dan Gohman | Regenerate configure.
|
commit | commitdiff | tree |
| 2009-03-24 |
Dan Gohman | Fix paths; AutoRegen.sh changes its current working...
|
commit | commitdiff | tree |
| 2009-03-24 |
Dan Gohman | Update for autoconf 2.6x;
|
commit | commitdiff | tree |
| 2009-03-24 |
Dan Gohman | Fix bash-isms.
|
commit | commitdiff | tree |
| 2009-03-24 |
Dan Gohman | Set the svn:mime-type to text/x-sh, so that diffs work.
|
commit | commitdiff | tree |
| 2009-03-24 |
Dan Gohman | Add a testcase for the scheduling heuristic introduced...
|
commit | commitdiff | tree |
| 2009-03-24 |
Dan Gohman | I was convinced that it's ok to allow a second i8 return...
|
commit | commitdiff | tree |
| 2009-03-24 |
Dan Gohman | Minor compile-time optimization; don't bother checking
|
commit | commitdiff | tree |
| 2009-03-24 |
Dan Gohman | Add a pre-pass to the burr-list scheduler which makes...
|
commit | commitdiff | tree |
| 2009-03-23 |
Dan Gohman | When unfolding a load during scheduling, the new operator...
|
commit | commitdiff | tree |
| 2009-03-23 |
Dan Gohman | Don't set SUnit::hasPhysRegDefs to true unless the...
|
commit | commitdiff | tree |
| 2009-03-23 |
Dan Gohman | Fix canClobberPhysRegDefs to check all SDNodes grouped...
|
commit | commitdiff | tree |
| 2009-03-23 |
Dan Gohman | Add a new bit to SUnit to record whether a node has...
|
commit | commitdiff | tree |
| 2009-03-23 |
Dan Gohman | Now that errs() is properly non-buffered, there's no...
|
commit | commitdiff | tree |
| 2009-03-23 |
Dan Gohman | Clarify a comment.
|
commit | commitdiff | tree |
| 2009-03-23 |
Dan Gohman | LoopVR is not CFGOnly.
|
commit | commitdiff | tree |
| 2009-03-23 |
Dan Gohman | Enhance LiveValues to work on PHI operands.
|
commit | commitdiff | tree |
| 2009-03-23 |
Dan Gohman | Make getOperandNumForIncomingValue and getOperandNumForIncom...
|
commit | commitdiff | tree |
| 2009-03-23 |
Dan Gohman | Correct some comments. Operand numbers start at 0.
|
commit | commitdiff | tree |
| 2009-03-23 |
Dan Gohman | Fix a grammaro in a comment that Bill noticed.
|
commit | commitdiff | tree |
| 2009-03-23 |
Dan Gohman | Add comments explaining why there's only one register for
|
commit | commitdiff | tree |
| 2009-03-20 |
Dan Gohman | Simplify this code; use a while instead of an if and...
|
commit | commitdiff | tree |
| 2009-03-20 |
Dan Gohman | The last use in a block that doesn't have successors
|
commit | commitdiff | tree |
| 2009-03-19 |
Dan Gohman | SADDO and UADDO are commutative.
|
commit | commitdiff | tree |
| 2009-03-19 |
Dan Gohman | Add a liveness analysis pass for LLVM IR values. This...
|
commit | commitdiff | tree |
| 2009-03-17 |
Dan Gohman | Recognize bswapl as bswap too.
|
commit | commitdiff | tree |
| 2009-03-17 |
Dan Gohman | Recognize "bswapq" as an alternate spelling for the...
|
commit | commitdiff | tree |
| 2009-03-14 |
Dan Gohman | Apply a patch by Micah Villmow to fix AsmParser to...
|
commit | commitdiff | tree |
| 2009-03-14 |
Dan Gohman | Add a testcase that covers a wide variety of ABI isel...
|
commit | commitdiff | tree |
| 2009-03-14 |
Dan Gohman | Use %rip-relative addressing on x86-64 whenever practical, as
|
commit | commitdiff | tree |
| 2009-03-14 |
Dan Gohman | Don't forego folding of loads into 64-bit adds when...
|
commit | commitdiff | tree |
| 2009-03-13 |
Dan Gohman | Add a few more ptrtoint/inttoptr cast tests.
|
commit | commitdiff | tree |
| 2009-03-13 |
Dan Gohman | Improve FastISel's handling of truncates to i1, and...
|
commit | commitdiff | tree |
| 2009-03-13 |
Dan Gohman | Fix FastISel's assumption that i1 values are always...
|
commit | commitdiff | tree |
| 2009-03-13 |
Dan Gohman | Enhance address-mode folding of ISD::ADD to handle...
|
commit | commitdiff | tree |
| 2009-03-12 |
Dan Gohman | Fix a typo in a comment.
|
commit | commitdiff | tree |
| 2009-03-11 |
Dan Gohman | Revert r66024. The JIT encoding for CALLpcrel32 is...
|
commit | commitdiff | tree |
| 2009-03-10 |
Dan Gohman | Add an explicit keyword.
|
commit | commitdiff | tree |
| 2009-03-10 |
Dan Gohman | Change these sed lines to behave correctly when the...
|
commit | commitdiff | tree |
| 2009-03-10 |
Dan Gohman | Use WriteAsOperand instead of manually decorating the...
|
commit | commitdiff | tree |
| 2009-03-10 |
Dan Gohman | Fix a post-RA scheduling liveness bug. When a basic...
|
commit | commitdiff | tree |
| 2009-03-10 |
Dan Gohman | Add more information to the EFLAGS note.
|
commit | commitdiff | tree |
| 2009-03-09 |
Dan Gohman | Add a note about EFLAGS optimization.
|
commit | commitdiff | tree |
| 2009-03-09 |
Dan Gohman | Don't record the increment instruction; just recompute...
|
commit | commitdiff | tree |
| 2009-03-09 |
Dan Gohman | Fix a few more places where induction variable types...
|
commit | commitdiff | tree |
| 2009-03-09 |
Dan Gohman | Use ReplacedTy instead of recomputing the same value.
|
commit | commitdiff | tree |
| 2009-03-09 |
Dan Gohman | Use LoopInfo's getLoopLatch() instead of doing what...
|
commit | commitdiff | tree |
| 2009-03-09 |
Dan Gohman | Don't use an induction variable type as a memory access...
|
commit | commitdiff | tree |
| 2009-03-09 |
Dan Gohman | Factor out the code that determines the memory access...
|
commit | commitdiff | tree |
| 2009-03-09 |
Dan Gohman | Move the sorting of the StrideOrder array earlier so...
|
commit | commitdiff | tree |
| 2009-03-09 |
Dan Gohman | Delete the isOnlyStride argument, which is unused.
|
commit | commitdiff | tree |
| 2009-03-09 |
Dan Gohman | Tidy some LSR debug output: announce the loop it's...
|
commit | commitdiff | tree |
| 2009-03-07 |
Dan Gohman | Arithmetic instructions don't set EFLAGS bits OF and...
|
commit | commitdiff | tree |
| 2009-03-06 |
Dan Gohman | Use c_str() to force the string to be nul-terminated.
|
commit | commitdiff | tree |
| 2009-03-06 |
Dan Gohman | Fix ScheduleDAGRRList::CopyAndMoveSuccessors' handling...
|
commit | commitdiff | tree |
| 2009-03-06 |
Dan Gohman | Use CloneModule's ValueMap to avoid needing to look up
|
commit | commitdiff | tree |
| 2009-03-05 |
Dan Gohman | Fix a bugpoint bug on anonymous functions. Instead...
|
commit | commitdiff | tree |
| 2009-03-05 |
Dan Gohman | Don't use plain INC32 and DEC32 on x86-64; it needs
|
commit | commitdiff | tree |
| 2009-03-05 |
Dan Gohman | When creating X86ISD::INC and X86ISD::DEC nodes, only...
|
commit | commitdiff | tree |
| 2009-03-05 |
Dan Gohman | Fix the "test" optimization to recognize "dec" as an...
|
commit | commitdiff | tree |
| 2009-03-05 |
Dan Gohman | Make this test more thorough. Not only should there...
|
commit | commitdiff | tree |
| 2009-03-04 |
Dan Gohman | Fix this comment.
|
commit | commitdiff | tree |
| 2009-03-04 |
Dan Gohman | Add an assertion for a condition that's always true...
|
commit | commitdiff | tree |
| 2009-03-04 |
Dan Gohman | Re-apply 66008, now that the unfoldMemoryOperand bug...
|
commit | commitdiff | tree |
| 2009-03-04 |
Dan Gohman | Correct this comment.
|
commit | commitdiff | tree |
| 2009-03-04 |
Dan Gohman | When using MachineInstr operand indices on SDNodes...
|
commit | commitdiff | tree |
| 2009-03-04 |
Dan Gohman | Revert r66004 for now; it's causing a variety of test...
|
commit | commitdiff | tree |
| 2009-03-04 |
Dan Gohman | Teach the x86 backend to eliminate "test" instructions...
|
commit | commitdiff | tree |
| 2009-03-03 |
Dan Gohman | Add '(implicit EFLAGS)' for AND, OR, XOR, NEG, INC...
|
commit | commitdiff | tree |
| 2009-03-03 |
Dan Gohman | Make a comment less terse.
|
commit | commitdiff | tree |
| 2009-03-03 |
Dan Gohman | Fix a bunch of Doxygen syntax issues. Escape special...
|
commit | commitdiff | tree |
| 2009-02-27 |
Dan Gohman | Update another test for the LoopInfo::print changes.
|
commit | commitdiff | tree |
| 2009-02-27 |
Dan Gohman | Update this test for the LoopInfo::print changes.
|
commit | commitdiff | tree |
| 2009-02-27 |
Dan Gohman | Make LoopInfo::print() identify header, exit, and latch...
|
commit | commitdiff | tree |
| 2009-02-24 |
Dan Gohman | Rename ScalarEvolution's getIterationCount to getBackedgeTak...
|
commit | commitdiff | tree |
| 2009-02-24 |
Dan Gohman | Add a testcase for the problem fixed in r65289.
|
commit | commitdiff | tree |
| 2009-02-24 |
Dan Gohman | Fix a ValueTracking rule: RHS means operand 1, not...
|
commit | commitdiff | tree |
| 2009-02-24 |
Dan Gohman | Generalize the ChangeCompareStride code, in preparation for
|
commit | commitdiff | tree |
| 2009-02-24 |
Dan Gohman | Preserve the DominanceFrontier analysis in the LoopDeletion...
|
commit | commitdiff | tree |
| 2009-02-23 |
Dan Gohman | Back out the change in 64918 that used sign-extensions...
|
commit | commitdiff | tree |
| 2009-02-23 |
Dan Gohman | Fast-isel can't do TLS yet, so it should fall back...
|
commit | commitdiff | tree |
| 2009-02-23 |
Dan Gohman | LoopDeletion needs to inform ScalarEvolution when a...
|
commit | commitdiff | tree |
| 2009-02-23 |
Dan Gohman | Use the -stack-alignment option instead of using a...
|
commit | commitdiff | tree |
| 2009-02-23 |
Dan Gohman | IndVarSimplify preserves ScalarEvolution. In the
|
commit | commitdiff | tree |
| 2009-02-22 |
Dan Gohman | Revert the part of 64623 that attempted to align the...
|
commit | commitdiff | tree |
| 2009-02-22 |
Dan Gohman | Properly parenthesize this expression, fixing a real...
|
commit | commitdiff | tree |
| 2009-02-20 |
Dan Gohman | Fix a bug that David Greene found in the DAGCombiner...
|
commit | commitdiff | tree |
| 2009-02-20 |
Dan Gohman | Simplify code and reduce indentation. No functionality...
|
commit | commitdiff | tree |
| 2009-02-20 |
Dan Gohman | Fix 80-column violations.
|
commit | commitdiff | tree |
| 2009-02-20 |
Dan Gohman | It's not necessary to check if Base is null here.
|
commit | commitdiff | tree |
| 2009-02-20 |
Dan Gohman | Add a comment about how Imm can be used for loop-variant...
|
commit | commitdiff | tree |
| 2009-02-20 |
Dan Gohman | Implement "superhero" strength reduction, or full strength
|
commit | commitdiff | tree |
| 2009-02-19 |
Dan Gohman | Use DEBUG() instead of passing *DOUT to WriteAsOperand,
|
commit | commitdiff | tree |
| 2009-02-19 |
Dan Gohman | Make the debug output of LSR less cryptic and more...
|
commit | commitdiff | tree |
| 2009-02-19 |
Dan Gohman | Fix the logic in this assertion to properly validate...
|
commit | commitdiff | tree |
| 2009-02-19 |
Dan Gohman | Reapply r57340. VMKit does not presently rely on materialize...
|
commit | commitdiff | tree |
| 2009-02-18 |
Dan Gohman | Use a sign-extend instead of a zero-extend when promoting a
|
commit | commitdiff | tree |
| 2009-02-18 |
Dan Gohman | Simplify by using dyn_cast instead of isa and cast.
|
commit | commitdiff | tree |
| next |
projects / oota-llvm.git / search