| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Update CMake build. | Ted Kremenek | 2010-01-15 | 1 | -1/+0 |
| | | | | | llvm-svn: 93571 | ||||
| * | Update CMake build. | Ted Kremenek | 2010-01-15 | 1 | -1/+2 |
| | | | | | llvm-svn: 93570 | ||||
| * | add testcase for r93564 | Jim Grosbach | 2010-01-15 | 1 | -0/+20 |
| | | | | | llvm-svn: 93567 | ||||
| * | Add comments to the dump() and dumpr() routines. | Dan Gohman | 2010-01-15 | 1 | -0/+8 |
| | | | | | llvm-svn: 93566 | ||||
| * | DEBUG_VALUE is now variable sized, as it has a | Dale Johannesen | 2010-01-15 | 2 | -1/+25 |
| | | | | | | | | | target-dependent memory address representation in it. Restore X86 printing of DEBUG_VALUE; lowering is done in X86RegisterInfo using the normal algorithm. llvm-svn: 93565 | ||||
| * | Update Thumb1 storeRegToStackSlot() and loadRegFromStackSlot() to properly | Jim Grosbach | 2010-01-15 | 1 | -2/+6 |
| | | | | | | | | handle physical registers R0-R7 when described as having a non-tGPR register class. llvm-svn: 93564 | ||||
| * | Fix a typo that Anton noticed. | Dan Gohman | 2010-01-15 | 1 | -1/+1 |
| | | | | | llvm-svn: 93563 | ||||
| * | Add FIXME. | Devang Patel | 2010-01-15 | 1 | -0/+1 |
| | | | | | llvm-svn: 93562 | ||||
| * | Fix a comment typo. | Bob Wilson | 2010-01-15 | 1 | -1/+1 |
| | | | | | llvm-svn: 93560 | ||||
| * | move "Metadata Nodes and Metadata Strings" section to the right place in the | Chris Lattner | 2010-01-15 | 1 | -38/+32 |
| | | | | | | | document and edit it. llvm-svn: 93559 | ||||
| * | Remove the InlineHint attribute. There are no current or planned | Eric Christopher | 2010-01-15 | 12 | -21/+4 |
| | | | | | | | users. llvm-svn: 93558 | ||||
| * | zext / truncate is free on msp430. Inform codegen about this. | Anton Korobeynikov | 2010-01-15 | 2 | -0/+42 |
| | | | | | llvm-svn: 93556 | ||||
| * | Reenable tests | Anton Korobeynikov | 2010-01-15 | 2 | -8/+6 |
| | | | | | llvm-svn: 93555 | ||||
| * | Add branch relaxation pass (shamelessly stolen from PPC). | Anton Korobeynikov | 2010-01-15 | 7 | -1/+256 |
| | | | | | llvm-svn: 93554 | ||||
| * | Provide instruction sizes & encoding. No opcodes yet (but not needed so far). | Anton Korobeynikov | 2010-01-15 | 3 | -612/+839 |
| | | | | | llvm-svn: 93553 | ||||
| * | Enable bit tests and setcc stuff. | Anton Korobeynikov | 2010-01-15 | 2 | -20/+25 |
| | | | | | llvm-svn: 93552 | ||||
| * | Add micro-optimization which allows us to fold imm into cmp. This allows us ↵ | Anton Korobeynikov | 2010-01-15 | 1 | -0/+32 |
| | | | | | | | to save 1 word (sometimes) and reduce register pressure. llvm-svn: 93551 | ||||
| * | Add instruction formats & support stuff | Anton Korobeynikov | 2010-01-15 | 1 | -10/+115 |
| | | | | | llvm-svn: 93550 | ||||
| * | Don't make changes to the MBB in MachineBasicBlock::canFallThrough(). | Jakob Stoklund Olesen | 2010-01-15 | 1 | -1/+1 |
| | | | | | | | | This fixes the regression for -pre-regalloc-taildup in MultiSource/Applications/lambda-0.1.3. llvm-svn: 93541 | ||||
| * | Simplify logic. Any functional change is unintended. | Jakob Stoklund Olesen | 2010-01-15 | 1 | -6/+2 |
| | | | | | llvm-svn: 93540 | ||||
| * | unbreak the build, grr symlinks. | Chris Lattner | 2010-01-15 | 1 | -1/+1 |
| | | | | | llvm-svn: 93539 | ||||
| * | Add some debug routines to SelectionDAG to dump full DAGs. | David Greene | 2010-01-15 | 3 | -1/+63 |
| | | | | | | | | | | | | print/dumpWithDepth allows one to dump a DAG up to N levels deep. dump/printWithFullDepth prints the whole DAG, subject to a depth limit on 100 in the default case (to prevent infinite recursion). Have CannotYetSelect to a dumpWithFullDepth so it is clearer exactly what the non-matching DAG looks like. llvm-svn: 93538 | ||||
| * | fix a bug in range information for $42, eliminate an | Chris Lattner | 2010-01-15 | 4 | -14/+10 |
| | | | | | | | unneeded argument from ParseExpression. llvm-svn: 93536 | ||||
| * | add range information for mem X86Operand's, now all | Chris Lattner | 2010-01-15 | 1 | -7/+11 |
| | | | | | | | X86Operand's have range info. llvm-svn: 93535 | ||||
| * | extend MCAsmParser::ParseExpression and ParseParenExpression | Chris Lattner | 2010-01-15 | 5 | -42/+70 |
| | | | | | | | | to return range information for subexpressions. Use this to provide range info for several new X86Operands. llvm-svn: 93534 | ||||
| * | give X86Operand a ctor and start passing SMLoc's into it. | Chris Lattner | 2010-01-15 | 1 | -11/+13 |
| | | | | | llvm-svn: 93532 | ||||
| * | Improve llvm.dbg.declare intrinsic by referring directly to the storage in ↵ | Victor Hernandez | 2010-01-15 | 11 | -57/+63 |
| | | | | | | | | | | its first argument, via function-local metadata (instead of via a bitcast). This patch also cleans up code that expects there to be a bitcast in the first argument and testcases that call llvm.dbg.declare. It also strips old llvm.dbg.declare intrinsics that did not pass metadata as the first argument. llvm-svn: 93531 | ||||
| * | Revert 93499. After discussion with Chris we agreed | Dale Johannesen | 2010-01-15 | 2 | -29/+0 |
| | | | | | | | | | | FrameIndexes should be lowered, but the same way as everything else (target dependent) rather than in a special hacked way. The lowering needs to be done for eventual purposes of Dwarf generation. llvm-svn: 93530 | ||||
| * | add range location info for registers, change | Chris Lattner | 2010-01-15 | 1 | -33/+42 |
| | | | | | | | | X86Operand::Create* implementations to avoid copy ctor use. llvm-svn: 93528 | ||||
| * | Hook up llc's -filetype=obj to use MCStreamer if an MCCodeEmitter is available. | Nate Begeman | 2010-01-15 | 10 | -1556/+144 |
| | | | | | | | | | | Remove most of old Mach-O Writer support, it has been replaced by MCMachOStreamer Further refactoring to completely remove MachOWriter and drive the object file writer with the AsmPrinter MCInst/MCSection logic is forthcoming. llvm-svn: 93527 | ||||
| * | clean up the memory management of the operands. | Chris Lattner | 2010-01-15 | 1 | -53/+52 |
| | | | | | llvm-svn: 93526 | ||||
| * | refactor ParseRegister to avoid using X86Operand as a temporary | Chris Lattner | 2010-01-15 | 1 | -15/+12 |
| | | | | | | | datastructure when parsing a mem operand. llvm-svn: 93521 | ||||
| * | mention that unwind isn't to be trusted, patch by Dustin Laurence | Chris Lattner | 2010-01-15 | 1 | -0/+6 |
| | | | | | llvm-svn: 93518 | ||||
| * | Revert r93504 because older uses of llvm.dbg.declare intrinsics need to be ↵ | Victor Hernandez | 2010-01-15 | 9 | -49/+55 |
| | | | | | | | auto-upgraded llvm-svn: 93515 | ||||
| * | Test case for http://llvm.org/PR6028. | Jay Foad | 2010-01-15 | 1 | -0/+9 |
| | | | | | llvm-svn: 93511 | ||||
| * | Fix http://llvm.org/PR6028, an assertion failure when an UndefValue of | Jay Foad | 2010-01-15 | 1 | -2/+16 |
| | | | | | | | integer type is used. llvm-svn: 93509 | ||||
| * | Change pre-regalloc tail duplication to only duplicate indirect branch blocks. | Bob Wilson | 2010-01-15 | 1 | -4/+11 |
| | | | | | | | | | | The pre-regalloc pass caused some regressions in both compile time and performance of the generated code, and it did not improve performance, except for indirect branches. I also moved the check for single-block loops to speed up the common case when running the taildup pass before reg allocation. llvm-svn: 93505 | ||||
| * | Improve llvm.dbg.declare intrinsic by referring directly to the storage in ↵ | Victor Hernandez | 2010-01-15 | 9 | -55/+49 |
| | | | | | | | | | its first argument, via function-local metadata (instead of via a bitcast). This patch also cleans up code that expects there to be a bitcast in the first argument and testcases that call llvm.dbg.declare. llvm-svn: 93504 | ||||
| * | Temporary disable tests | Anton Korobeynikov | 2010-01-15 | 2 | -0/+2 |
| | | | | | llvm-svn: 93501 | ||||
| * | Lower FrameIndex operand of DEBUG_VALUE (specially) and | Dale Johannesen | 2010-01-15 | 2 | -0/+29 |
| | | | | | | | print it as a comment on X86. llvm-svn: 93499 | ||||
| * | Remove DEBUG_DECLARE, looks like we don't need it. | Dale Johannesen | 2010-01-15 | 5 | -29/+6 |
| | | | | | | | Also, DEBUG_VALUE has side effects. llvm-svn: 93498 | ||||
| * | Fix cmp emission on msp430: we definitely should turn stuff like | Anton Korobeynikov | 2010-01-15 | 2 | -25/+27 |
| | | | | | | | "icmp lhs, rhs" into "cmp rhs, lhs". This should fix PR5979. llvm-svn: 93496 | ||||
| * | Do not use AT_specification die for static variables. It confuses gdb. | Devang Patel | 2010-01-15 | 2 | -3/+18 |
| | | | | | llvm-svn: 93494 | ||||
| * | Make sure include/llvm/MC/MCParsedAsmOperand.h can be compiled alone. | Jeffrey Yasskin | 2010-01-15 | 1 | -0/+1 |
| | | | | | llvm-svn: 93492 | ||||
| * | fix 80-column violations | Jim Grosbach | 2010-01-15 | 1 | -12/+14 |
| | | | | | llvm-svn: 93487 | ||||
| * | new test case for r93485. | Devang Patel | 2010-01-15 | 1 | -0/+4 |
| | | | | | llvm-svn: 93486 | ||||
| * | Fix 80 column violations and clean up whitespace | Jim Grosbach | 2010-01-15 | 1 | -5/+4 |
| | | | | | llvm-svn: 93484 | ||||
| * | Do not emit multiple AT_container_type attributes. | Devang Patel | 2010-01-15 | 1 | -1/+2 |
| | | | | | | | We need to find a better way to emit this info. llvm-svn: 93481 | ||||
| * | Name change for consistency. No functional change. | Jim Grosbach | 2010-01-15 | 1 | -7/+7 |
| | | | | | llvm-svn: 93480 | ||||
| * | EmitAtomicCmpSwap() custome inserter needs to delete the MI passed in. ↵ | Jim Grosbach | 2010-01-15 | 1 | -0/+3 |
| | | | | | | | EmitAtomicBinary() already does this. llvm-svn: 93479 | ||||

