| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Fix PR2553 | Chris Lattner | 2008-07-17 | 1 | -1/+2 |
| | | | | | llvm-svn: 53715 | ||||
| * | When lowering certain atomics, we need to copy the memoperand from the old | Mon P Wang | 2008-07-17 | 1 | -1/+5 |
| | | | | | | | atomic operation to the new one. llvm-svn: 53714 | ||||
| * | Factorize some code for determining which libcall to use. | Duncan Sands | 2008-07-17 | 4 | -562/+204 |
| | | | | | llvm-svn: 53713 | ||||
| * | Inliner tweak. Function calls should cost more than one instruction! | Evan Cheng | 2008-07-17 | 1 | -0/+5 |
| | | | | | llvm-svn: 53712 | ||||
| * | Factor MergeBlockIntoPredecessor out into BasicBlockUtils. | Owen Anderson | 2008-07-17 | 2 | -51/+56 |
| | | | | | llvm-svn: 53705 | ||||
| * | Mark function used by asm block as used, otherwise optimizer may not see the ↵ | Devang Patel | 2008-07-16 | 1 | -1/+3 |
| | | | | | | | use and may delete the function. llvm-svn: 53692 | ||||
| * | There's no need to iterate block merging and PRE. In fact, iterating the latter | Owen Anderson | 2008-07-16 | 1 | -14/+12 |
| | | | | | | | could cause problems for memdep when it breaks critical edges. llvm-svn: 53691 | ||||
| * | Somehow, custom lowering of i64 multiplications got dropped along the way. | Scott Michel | 2008-07-16 | 1 | -104/+105 |
| | | | | | llvm-svn: 53689 | ||||
| * | Fix the result type of X86's truncate to i8. | Dan Gohman | 2008-07-16 | 1 | -5/+3 |
| | | | | | llvm-svn: 53688 | ||||
| * | Fix the result type of a VECTOR_SHUFFLE+BIT_CONVERT dagcombine. This | Dan Gohman | 2008-07-16 | 1 | -3/+2 |
| | | | | | | | | was turned up by some new SelectionDAG assertion checks that I'm working on. llvm-svn: 53687 | ||||
| * | Add support for promoting and expanding AssertZext | Duncan Sands | 2008-07-16 | 2 | -0/+42 |
| | | | | | | | | and AssertSext. Needed when passing huge integer parameters with the zeroext or signext attributes. llvm-svn: 53684 | ||||
| * | Fix a comment to say nonnegative instead of positive. | Dan Gohman | 2008-07-16 | 1 | -1/+1 |
| | | | | | llvm-svn: 53681 | ||||
| * | Add an assert to check for empty flags for MachineMemOperand. | Dan Gohman | 2008-07-16 | 1 | -0/+1 |
| | | | | | llvm-svn: 53680 | ||||
| * | Redo InstCombiner::visitExtractValueInst. Instead of using the (complicate) | Matthijs Kooijman | 2008-07-16 | 1 | -5/+80 |
| | | | | | | | | FindInsertedValue, it now performs a number of simple transformations that should result in the same effect when applied iteratively. llvm-svn: 53673 | ||||
| * | Reorder methods alphabetically. No functionality change. | Duncan Sands | 2008-07-16 | 3 | -922/+913 |
| | | | | | | | | | While this is not a wonderful organizing principle, it does make it easy to find routines, and clear where to insert new ones. llvm-svn: 53672 | ||||
| * | Turn on LegalizeTypes by default. | Duncan Sands | 2008-07-16 | 1 | -4/+6 |
| | | | | | llvm-svn: 53671 | ||||
| * | Don't use ++idx_begin when I actually mean idx_begin + 1, especially since we | Matthijs Kooijman | 2008-07-16 | 1 | -1/+1 |
| | | | | | | | | | also use *idx_begin in the same expression, giving unpredictable results. This fixes this bug: http://lists.cs.uiuc.edu/pipermail/llvmdev/2008-July/015877.html llvm-svn: 53670 | ||||
| * | Fix PR2296. Do not transform x86_sse2_storel_dq into a full-width store. | Evan Cheng | 2008-07-16 | 1 | -1/+0 |
| | | | | | llvm-svn: 53666 | ||||
| * | x86-64 PIC JIT fixes: do not generate the extra load for external GV's. | Evan Cheng | 2008-07-16 | 2 | -8/+5 |
| | | | | | llvm-svn: 53661 | ||||
| * | X86-64 PIC jump table values are different from x86-32 cases, they are dest ↵ | Evan Cheng | 2008-07-16 | 1 | -0/+4 |
| | | | | | | | - table base. llvm-svn: 53660 | ||||
| * | TargetAsmInfo::SectionForGlobal showed up in a profile. Simplify it a little. | Dan Gohman | 2008-07-15 | 1 | -5/+3 |
| | | | | | llvm-svn: 53639 | ||||
| * | SelectionDAG::AssignNodeIds is unused. | Dan Gohman | 2008-07-15 | 1 | -11/+0 |
| | | | | | llvm-svn: 53636 | ||||
| * | Don't sort SDNodes by their addresses in SelectionDAG::dump. Instead, | Dan Gohman | 2008-07-15 | 1 | -9/+5 |
| | | | | | | | | just use the AllNodes order, which is at least relatively stable across runs. llvm-svn: 53632 | ||||
| * | Revert this, as it seems to still be broken. | Owen Anderson | 2008-07-15 | 1 | -1/+1 |
| | | | | | llvm-svn: 53627 | ||||
| * | Enable local PRE by default. | Owen Anderson | 2008-07-15 | 1 | -1/+1 |
| | | | | | llvm-svn: 53616 | ||||
| * | Have GVN do a pre-pass over the CFG that folds away unconditional branches ↵ | Owen Anderson | 2008-07-15 | 1 | -2/+63 |
| | | | | | | | where possible. This allows local PRE to be more aggressive. llvm-svn: 53615 | ||||
| * | LegalizeTypes support for fabs on ppc long double. | Duncan Sands | 2008-07-15 | 2 | -0/+15 |
| | | | | | llvm-svn: 53613 | ||||
| * | Allow deadargelim to change return types even though now values were dead. This | Matthijs Kooijman | 2008-07-15 | 1 | -11/+1 |
| | | | | | | | again canonicalizes {i32} into i32 and {} into void. llvm-svn: 53610 | ||||
| * | Revert r53606. It turns out that explicitely tracking the liveness of the | Matthijs Kooijman | 2008-07-15 | 1 | -97/+103 |
| | | | | | | | | | | | | return value as a whole in deadargelim is really not needed now that we simply rebuild the old return value and actually prevents some canonicalization from taking place. This revert stops deadargelim from changing {i32} into i32 for now, but I'll fix that next. llvm-svn: 53609 | ||||
| * | Make deadargelim a bit less smart, so it doesn't choke on nested structs as | Matthijs Kooijman | 2008-07-15 | 1 | -38/+21 |
| | | | | | | | | | | | | | return values that are still (partially) live. Instead of updating all uses of a call instruction after removing some elements, it now just rebuilds the original struct (With undef gaps where the unused values were) and leaves it to instcombine to clean this up. The added testcase still fails currently, but this is due to instcombine which isn't good enough yet. I will fix that part next. llvm-svn: 53608 | ||||
| * | Don't use isa when we can reuse a previous dyn_cast. | Matthijs Kooijman | 2008-07-15 | 1 | -3/+2 |
| | | | | | llvm-svn: 53607 | ||||
| * | Make DeadArgElim keep liveness of the return value as a whole in addition to | Matthijs Kooijman | 2008-07-15 | 1 | -126/+124 |
| | | | | | | | | | | | | only the liveness of partial return values (for functions returning a struct). This is more explicit to prevent unwanted changes in the return value. In particular, deadargelim now canonicalizes a function returning {i32} to returning i32 and {} to void, if the struct returned is not used in its entirety, but only the single element is used. llvm-svn: 53606 | ||||
| * | LegalizeTypes support for promotion of bswap. | Duncan Sands | 2008-07-15 | 2 | -0/+12 |
| | | | | | | | | | | | | | In LegalizeDAG the value is zero-extended to the new type before byte swapping. It doesn't matter how the extension is done since the new bits are shifted off anyway after the swap, so extend by any old rubbish bits. This results in the final assembler for the testcase being one line shorter. llvm-svn: 53604 | ||||
| * | LegalizeTypes support for promotion of SIGN_EXTEND_INREG. | Duncan Sands | 2008-07-15 | 2 | -0/+9 |
| | | | | | llvm-svn: 53603 | ||||
| * | Reorder the integer promotion methods alphabetically. | Duncan Sands | 2008-07-15 | 1 | -365/+356 |
| | | | | | | | No change in functionality. llvm-svn: 53602 | ||||
| * | Let DAE keep a list of live functions, instead of simply marking all arguments | Matthijs Kooijman | 2008-07-15 | 1 | -6/+14 |
| | | | | | | | | and return values live for those functions. This doesn't change anything yet, but prepares for the coming commits. llvm-svn: 53601 | ||||
| * | Split DAE::MarkLive into MarkLive and PropagateLiveness. | Matthijs Kooijman | 2008-07-15 | 1 | -0/+6 |
| | | | | | llvm-svn: 53600 | ||||
| * | Pass around const RetOrArg references instead of copying values. Also, mark | Matthijs Kooijman | 2008-07-15 | 1 | -3/+3 |
| | | | | | | | RetOrArg::getDescription() as const. llvm-svn: 53599 | ||||
| * | Simplify debug code by using RetOrArg::getDescription(). | Matthijs Kooijman | 2008-07-15 | 1 | -6/+1 |
| | | | | | llvm-svn: 53598 | ||||
| * | Fix indentation (intentionally left out of the previous commit). | Matthijs Kooijman | 2008-07-15 | 1 | -51/+51 |
| | | | | | llvm-svn: 53592 | ||||
| * | Move the deadargelim code for intrinsically alive functions into its own | Matthijs Kooijman | 2008-07-15 | 1 | -23/+25 |
| | | | | | | | method, to slightly simplify control flow. llvm-svn: 53591 | ||||
| * | Fixed potential bug if the source and target of a bit convert have different ↵ | Mon P Wang | 2008-07-15 | 1 | -1/+3 |
| | | | | | | | alignment llvm-svn: 53590 | ||||
| * | Correct this inversion! | Nick Lewycky | 2008-07-15 | 1 | -1/+1 |
| | | | | | | | I swear that didn't show up in svn diff... llvm-svn: 53587 | ||||
| * | Fix up comments. | Nick Lewycky | 2008-07-15 | 1 | -5/+7 |
| | | | | | llvm-svn: 53586 | ||||
| * | Fixed call stack alignment. Improved AsmPrinter alignment issues. | Bruno Cardoso Lopes | 2008-07-15 | 5 | -18/+47 |
| | | | | | llvm-svn: 53585 | ||||
| * | Reapply 53476 and 53480, with a fix so that it properly updates | Dan Gohman | 2008-07-14 | 5 | -35/+75 |
| | | | | | | | | the BB member to the current basic block after emitting instructions. llvm-svn: 53567 | ||||
| * | Fix uninitialized use of the Changed variable. | Dan Gohman | 2008-07-14 | 1 | -0/+2 |
| | | | | | llvm-svn: 53564 | ||||
| * | Improve debug output for MemOperandSDNode. PseudoSourceValue nodes | Dan Gohman | 2008-07-14 | 1 | -4/+7 |
| | | | | | | | | don't have value names, so use print instead of getName() to get a useful string. llvm-svn: 53563 | ||||
| * | Fix edito in the PseudoSourceValue name list. | Dan Gohman | 2008-07-14 | 1 | -1/+1 |
| | | | | | llvm-svn: 53562 | ||||
| * | I don't think BUILD_PAIR can have a vector result. | Duncan Sands | 2008-07-14 | 1 | -12/+0 |
| | | | | | | | Remove support for this. llvm-svn: 53559 | ||||

