- 19f5f71 Revert "Use MachineInstr::readsWritesVirtualRegister to determine if a register is read." by Jakob Stoklund Olesen · 15 years ago
- 00c53ca Use MachineInstr::readsWritesVirtualRegister to determine if a register is read. by Jakob Stoklund Olesen · 15 years ago
- 63e6a48 If the first definition of a virtual register is a partial redef, add an by Jakob Stoklund Olesen · 15 years ago
- 761fd4c Fix typo in comment. by Nick Lewycky · 15 years ago
- ad6c5a2 Yes, if the redef is a copy, update the old val# with the copy. But make sure to clear the copy field if the redef is not a copy. by Evan Cheng · 15 years ago
- 87565c1 improve portability to systems that don't have powf/modf (e.g. solaris 9) by Chris Lattner · 15 years ago
- 61d5310 A partial re-def instruction may be a copy. by Evan Cheng · 15 years ago
- 623d3c1 Re-defined valno is always valno even for partial re-def's. by Evan Cheng · 15 years ago
- 3749943 Teach liveintervalanalysis about virtual registers which are defined by reg_sequence instructions that are formed by registers defined by distinct instructions. e.g. by Evan Cheng · 15 years ago
- afff40a Teach PHI elimination to remove REG_SEQUENCE instructions and update references of the source operands with references of the destination with subreg indices. e.g. by Evan Cheng · 15 years ago
- ad98f79 Print basic block numbers in live interval debug output. Since the rest of the by Bob Wilson · 15 years ago
- 6691a89 - Really preserve dbg_value instructions when the register is spilled. by Evan Cheng · 15 years ago
- 6fa7636 Re-enable 102323 with fix: do not update dbg_value's with incorrect frame indices when the live interval are being re-materialized. by Evan Cheng · 15 years ago
- 6cd4b34 Temporary disable spiller modifying dbg_value. It's breaking build. by Evan Cheng · 15 years ago
- 962021b - Move TargetLowering::EmitTargetCodeForFrameDebugValue to TargetInstrInfo and rename it to emitFrameIndexDebugValue. by Evan Cheng · 15 years ago
- 81bf03e Eliminate MachineBasicBlock::const_livein_iterator and make by Dan Gohman · 15 years ago
- 75361b6 rename llvm::llvm_report_error -> llvm::report_fatal_error by Chris Lattner · 15 years ago
- 991de14 Introduce SpecificBumpPtrAllocator, a wrapper for BumpPtrAllocator which allows by Benjamin Kramer · 15 years ago
- 01cb1b6 Reapply r99881 with some fixes: only call destructor in releaseMemory! by Torok Edwin · 15 years ago
- 226dd2b Revert 99881, it brooke smooshlab's llvm-gcc-i386-darwin9. by Torok Edwin · 15 years ago
- 2a22004 Introduce another Reset() method in BumpPtrAllocator that calls a destructor by Torok Edwin · 15 years ago
- 28a1e48 Avoid being influenced by the presence of dbg_value instructions. by Evan Cheng · 15 years ago
- d6a6b3b Revert Edwin's change that is breaking MultiSource/Applications/ClamAV/clamscan. by Bob Wilson · 15 years ago
- 1541738 Fix memory leak in liveintervals: the destructor for VNInfos must be called, by Torok Edwin · 15 years ago
- 4507f08 Fix liveintervals handling of dbg_value instructions. by Evan Cheng · 15 years ago
- 826cbac The check for coalescing a virtual register to a physical register, e.g. by Evan Cheng · 15 years ago
- a2f767b Fix a bug in DEBUG_VALUE handling Devang ran into. I'll get this loop right yet. by Dale Johannesen · 15 years ago
- e5d9041 Use methods to determine if a LiveInterval is spillable. by Jakob Stoklund Olesen · 15 years ago
- dcfe5f3 Keep track of phi join registers explicitly in LiveVariables. by Jakob Stoklund Olesen · 15 years ago
- 352d352 Always normalize spill weights, also for intervals created by spilling. by Jakob Stoklund Olesen · 15 years ago
- e21765d When I rewrote this loop per Chris' preference I changed its behavior. Oops. by Dale Johannesen · 16 years ago
- 1d0aeab Rewrite loop to suit Chris' preference. by Dale Johannesen · 16 years ago
- bd63520 Skip DBG_VALUE many places in live intervals and by Dale Johannesen · 16 years ago
- 518bb53 move target-independent opcodes out of TargetInstrInfo by Chris Lattner · 16 years ago
- 00a99a3 Run codegen dce pass for all targets at all optimization levels. Previously it's by Evan Cheng · 16 years ago
- 1caedd0 Ignore DEBUG_VALUE when building live intervals; by Dale Johannesen · 16 years ago
- 8a34229 by David Greene · 16 years ago
- 74ab5ee Changed slot index ranges for MachineBasicBlocks to be exclusive of endpoint. by Lang Hames · 16 years ago
- 74215fc Reuse lowered phi nodes. by Jakob Stoklund Olesen · 16 years ago
- ac94863 Coalesce insert_subreg undef, x first to avoid phase ordering issue. by Evan Cheng · 16 years ago
- cf97036 Also attempt trivial coalescing for live intervals that end in a copy. by Jakob Stoklund Olesen · 16 years ago
- 6194569 Added a new "splitting" spiller. by Lang Hames · 16 years ago
- dc492e0 Temporarily revert r90502. It was causing the llvm-gcc bootstrap on PPC to fail. by Bill Wendling · 16 years ago
- 6b74e50 Also attempt trivial coalescing for live intervals that end in a copy. by Jakob Stoklund Olesen · 16 years ago
- f4811a9 Clean up some loop logic. by Jakob Stoklund Olesen · 16 years ago
- 9836a26 Fix PR5391: support early clobber physical register def tied with a use (ewwww) by Evan Cheng · 16 years ago
- ce7a663 New virtual registers created for spill intervals should inherit allocation hints from the original register. by Jakob Stoklund Olesen · 16 years ago
- 6cd8103 More consistent labelling of basic blocks in debug output by Jakob Stoklund Olesen · 16 years ago
- 324da76 Add MachineBasicBlock::getName, and use it in place of getBasicBlock()->getName. by Jakob Stoklund Olesen · 16 years ago
- 1e78aa4 Remove the -early-coalescing option by Jakob Stoklund Olesen · 16 years ago
- df8ed02 Hide a couple of options. by Evan Cheng · 16 years ago
- 3de23e6 Fix inverted conflict test in -early-coalesce. by Jakob Stoklund Olesen · 16 years ago
- 233a60e The Indexes Patch. by Lang Hames · 16 years ago
- 5f4fb86 Trim unnecessary includes. by Evan Cheng · 16 years ago
- 0222a8c If the physical register being spilled does not have an interval, spill its sub-registers instead. by Evan Cheng · 16 years ago
- a70dca1 Factor out LiveIntervalAnalysis' code to determine whether an instruction by Dan Gohman · 16 years ago
- 599a87a isTriviallyReMaterializable checks the by Dan Gohman · 16 years ago
- 2627e08 Replace some code for aggressive-remat with MachineInstr::isInvariantLoad, and by Dan Gohman · 16 years ago
- 6cc91e3 Oops. Renamed remaining MachineInstrIndex references. by Lang Hames · 16 years ago
- cc3b065 Renamed MachineInstrIndex to LiveIndex. by Lang Hames · 16 years ago
- 295a808 Remove unused variable. by Daniel Dunbar · 16 years ago
- 2bf0649 Simplify a few more uses of reg_iterator. by Dan Gohman · 16 years ago
- c76909a Improve MachineMemOperand handling. by Dan Gohman · 16 years ago
- 296925d Fix PR5024. LiveVariables physical register defs should *commit* only after all by Evan Cheng · 16 years ago
- b0f5973 Fix PR4986. "r1024 = insert_subreg r1024, undef, 2" cannot be turned in an implicit_def. Instead, it's an identity copy so it should be eliminated. Also make sure to update livevariable kill information. by Evan Cheng · 16 years ago
- 39faac2 When computing live intervals for earlyclobber operands, by Dale Johannesen · 16 years ago
- 504f9a6 Fix -Asserts warning. by Daniel Dunbar · 16 years ago
- 3f85549 Another try at early partial coalescing. Identity phi source copies (their sources are defined by phi join def) are coalesced. And the phi join copy is backward copy propagated into the other copies. by Evan Cheng · 16 years ago
- 752195e Add early coalescing to liveintervals. This is work in progress and is known to miscompute some tests. Read it at your own rish, I have aged 10 year while writing this. by Evan Cheng · 16 years ago
- 35f291d Moved some more index operations over to LiveIntervals. by Lang Hames · 16 years ago
- 2173111 80 col violations. by Evan Cheng · 16 years ago
- 8651125 Replaces uses of unsigned for indexes in LiveInterval and VNInfo with by Lang Hames · 16 years ago
- 45cfe54 Change Pass::print to take a raw ostream instead of std::ostream, by Chris Lattner · 16 years ago
- 705e07f remove various std::ostream version of printing methods from by Chris Lattner · 16 years ago
- 77942d4 remove a dead class. by Chris Lattner · 16 years ago
- 8e6179f Convert DOUT to DEBUG(errs()...). by Bill Wendling · 16 years ago
- 52c1afc Modified VNInfo. The "copy" member is now a union which holds the copy for a register interval, or the defining register for a stack interval. Access is via getCopy/setCopy and getReg/setReg. by Lang Hames · 16 years ago
- 6ade93b Turn some insert_subreg, extract_subreg, subreg_to_reg into implicit_defs. by Evan Cheng · 16 years ago
- b525766 by David Greene · 16 years ago
- 845012e Use setPreservesAll and setPreservesCFG in CodeGen passes. by Dan Gohman · 16 years ago
- ce63ffb More migration to raw_ostream, the water has dried up around the iostream hole. by Daniel Dunbar · 16 years ago
- 3f0e830 Move to raw_ostream. by Daniel Dunbar · 16 years ago
- 340482d by David Greene · 16 years ago
- 3380d5c revert r76602, 76603, and r76615, pending design discussions. by Chris Lattner · 16 years ago
- a358c1d by David Greene · 16 years ago
- d129d73 Simplify the coalescer (finally!) by making LiveIntervals::processImplicitDefs a little more aggressive and teaching liveintervals to make use of isUndef marker on MachineOperands. by Evan Cheng · 16 years ago
- 5f15992 Changed my mind. We now allow remat of instructions whose defs have subreg indices. by Evan Cheng · 16 years ago
- 3784453 Let callers decide the sub-register index on the def operand of rematerialized instructions. by Evan Cheng · 16 years ago
- c23197a llvm_unreachable->llvm_unreachable(0), LLVM_UNREACHABLE->llvm_unreachable. by Torok Edwin · 16 years ago
- c25e758 assert(0) -> LLVM_UNREACHABLE. by Torok Edwin · 16 years ago
- 7d696d8 Convert more assert(0)+abort() -> LLVM_UNREACHABLE, by Torok Edwin · 16 years ago
- 413a15e Avoid compiler warnings if assertions turned off. by Duncan Sands · 16 years ago
- ffd1326 Improved tracking of value number kills. VN kills are now represented by Lang Hames · 16 years ago
- 459a7c6 Remove special handling of implicit_def. Fix a couple more bugs in liveintervalanalysis and coalescer handling of implicit_def. by Evan Cheng · 16 years ago
- 2578ba2 Handle IMPLICIT_DEF with isUndef operand marker, part 2. This patch moves the code to annotate machineoperands to LiveIntervalAnalysis. It also add markers for implicit_def that define physical registers. The rest, is just a lot of details. by Evan Cheng · 16 years ago
- 4784f1f Add a bit IsUndef to MachineOperand. This indicates the def / use register operand is defined by an implicit_def. That means it can def / use any register and passes (e.g. register scavenger) can feel free to ignore them. by Evan Cheng · 16 years ago
- 1873d0c When doing remat, don't consider uses of non-allocatable physregs. Patch by Chris Lattner · 16 years ago
- 10382fb More VNInfo tweaking, plus a little progress on intra-block splitting. by Lang Hames · 16 years ago
- d21c316 Improved PHI def marking, replaced some gotos with breaks. by Lang Hames · 16 years ago
- 857c4e0 VNInfo cleanup. by Lang Hames · 16 years ago