| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Don't attempt load PRE when there is no real redundancy (i.e., the load is in | Bob Wilson | 2010-03-02 | 1 | -4/+6 |
| | | | | | | | a loop and is itself the only dependency). llvm-svn: 97526 | ||||
| * | refactor some code out of OPC_EmitMergeInputChains into a | Chris Lattner | 2010-03-02 | 1 | -21/+37 |
| | | | | | | | new helper function. llvm-svn: 97525 | ||||
| * | When GVN needs to split critical edges for load PRE, check all of the | Bob Wilson | 2010-03-01 | 1 | -1/+4 |
| | | | | | | | | | predecessors before returning. Otherwise, if multiple predecessor edges need splitting, we only get one of them per iteration. This makes a small but measurable compile time improvement with -enable-full-load-pre. llvm-svn: 97521 | ||||
| * | remove a little hack I did for the old isel, not needed | Chris Lattner | 2010-03-01 | 1 | -16/+2 |
| | | | | | | | now that it is gone. llvm-svn: 97516 | ||||
| * | MemoryDepAnalysis is not used if redundant load processing is disabled. | Evan Cheng | 2010-03-01 | 1 | -1/+1 |
| | | | | | llvm-svn: 97512 | ||||
| * | remove all but one version of SelectionDAG::MorphNodeTo | Chris Lattner | 2010-03-01 | 1 | -81/+0 |
| | | | | | | | (the most general) the others are dead. llvm-svn: 97511 | ||||
| * | Remove the optimize for code size limitation on r67917. Optimize 64-bit imul ↵ | Evan Cheng | 2010-03-01 | 1 | -4/+0 |
| | | | | | | | by constants into leas + shl regardless if optimizing for code size. The size saving from using imulq isn't worth it. Also, the lea and shl instructions may expose further optimization. llvm-svn: 97507 | ||||
| * | remove a terrible hack that disabled assertions from this file because of ↵ | Chris Lattner | 2010-03-01 | 1 | -9/+0 |
| | | | | | | | | | build time problems. rdar://7697850. llvm-svn: 97500 | ||||
| * | Use methods to determine if a LiveInterval is spillable. | Jakob Stoklund Olesen | 2010-03-01 | 1 | -18/+26 |
| | | | | | | | Don't accidentally produce unspillable intervals for deeply nested loops. llvm-svn: 97496 | ||||
| * | Update CMake build. | Ted Kremenek | 2010-03-01 | 1 | -1/+0 |
| | | | | | llvm-svn: 97488 | ||||
| * | stop using generated sdnodexforms. | Chris Lattner | 2010-03-01 | 1 | -1/+5 |
| | | | | | llvm-svn: 97485 | ||||
| * | remove anders-aa from mainline, it isn't maintained and is | Chris Lattner | 2010-03-01 | 1 | -2868/+0 |
| | | | | | | | tantalyzing enough that people keep trying to use it. llvm-svn: 97483 | ||||
| * | Added STRHT for disassembly only and fixed a bug in AI3sthpo class where the W | Johnny Chen | 2010-03-01 | 2 | -2/+10 |
| | | | | | | | bit should be set to 0 instead of 1. llvm-svn: 97481 | ||||
| * | Accelerate isel dispatch for tables that start with a top-level | Chris Lattner | 2010-03-01 | 1 | -1/+36 |
| | | | | | | | | OPC_SwitchOpcode to use a table lookup instead of having to go through the interpreter for this. llvm-svn: 97469 | ||||
| * | Don't potentially read past the end of the fill data when making a NaN from | John McCall | 2010-03-01 | 1 | -1/+2 |
| | | | | | | | an APInt. llvm-svn: 97467 | ||||
| * | Fix optimization of ISD::TRUNCATE on vector operands. Based on a patch | Dan Gohman | 2010-03-01 | 1 | -4/+5 |
| | | | | | | | by Micah Villmow for PR6335. llvm-svn: 97461 | ||||
| * | Add explicit keywords. | Dan Gohman | 2010-03-01 | 2 | -2/+3 |
| | | | | | llvm-svn: 97460 | ||||
| * | Add a comment. | Dan Gohman | 2010-03-01 | 1 | -1/+5 |
| | | | | | llvm-svn: 97459 | ||||
| * | Add some debug output to LoopSimplify. | Dan Gohman | 2010-03-01 | 1 | -0/+30 |
| | | | | | llvm-svn: 97458 | ||||
| * | Spelling fixes. | Dan Gohman | 2010-03-01 | 5 | -37/+37 |
| | | | | | llvm-svn: 97453 | ||||
| * | This is now done. | Dan Gohman | 2010-03-01 | 1 | -5/+0 |
| | | | | | llvm-svn: 97450 | ||||
| * | Fix a missing newline in debug output. | Dan Gohman | 2010-03-01 | 1 | -1/+1 |
| | | | | | llvm-svn: 97449 | ||||
| * | Prune #includes. | Dan Gohman | 2010-03-01 | 1 | -1/+0 |
| | | | | | llvm-svn: 97448 | ||||
| * | Don't print "Modified" for passes which haven't modified anything. | Dan Gohman | 2010-03-01 | 1 | -6/+12 |
| | | | | | llvm-svn: 97446 | ||||
| * | Add JIT support to the TODO list (test commit) | Nathan Keynes | 2010-03-01 | 1 | -0/+1 |
| | | | | | llvm-svn: 97443 | ||||
| * | some trivial microoptimizations. | Chris Lattner | 2010-03-01 | 2 | -9/+18 |
| | | | | | llvm-svn: 97441 | ||||
| * | eliminate the CheckMultiOpcodeMatcher code and have each | Chris Lattner | 2010-03-01 | 1 | -9/+0 |
| | | | | | | | | | | ComplexPattern at the root be generated multiple times, once for each opcode they are part of. This encourages factoring because the opcode checks get treated just like everything else in the matcher. llvm-svn: 97439 | ||||
| * | add a new OPC_SwitchOpcode which is semantically equivalent | Chris Lattner | 2010-03-01 | 1 | -0/+30 |
| | | | | | | | | | | | | | to a scope where every child starts with a CheckOpcode, but executes more efficiently. Enhance DAGISelMatcherOpt to form it. This also fixes a bug in CheckOpcode: apparently the SDNodeInfo objects are not pointer comparable, we have to compare the enum name. llvm-svn: 97438 | ||||
| * | 80-col violations/trailing whitespace. | Mikhail Glushenkov | 2010-02-28 | 1 | -11/+12 |
| | | | | | llvm-svn: 97427 | ||||
| * | eliminate GetInt1/2 | Chris Lattner | 2010-02-28 | 1 | -16/+3 |
| | | | | | llvm-svn: 97426 | ||||
| * | hoist the new isel interpreter out of DAGISelHeader.h | Chris Lattner | 2010-02-28 | 1 | -0/+859 |
| | | | | | | | | | (which gets #included into the middle of each target's DAGISel class) into a .cpp file where it is only compiled once. llvm-svn: 97425 | ||||
| * | enhance the new isel to handle the 'node already exists' | Chris Lattner | 2010-02-28 | 1 | -1/+1 |
| | | | | | | | case of MorphNodeTo directly. llvm-svn: 97417 | ||||
| * | simplify this code, return only ever has zero or one operands. | Chris Lattner | 2010-02-28 | 1 | -8/+6 |
| | | | | | llvm-svn: 97408 | ||||
| * | Properly clear all the extra bits in a significand when making a NaN from an | John McCall | 2010-02-28 | 1 | -8/+28 |
| | | | | | | | | APInt. Be certain to set the integer bit in an x87 extended-precision significand so that we don't accidentally make a pseudo-NaN. llvm-svn: 97382 | ||||
| * | Add an override to StringRef::getAsInteger which parses into an APInt. | John McCall | 2010-02-28 | 1 | -12/+93 |
| | | | | | | | | | | | It gets its own implementation totally divorced from the (presumably performance-sensitive) routines which parse into a uint64_t. Add APInt::operator|=(uint64_t), which is situationally much better than using a full APInt. llvm-svn: 97381 | ||||
| * | Add support for global variables in an address space for llvm-c and ocaml. | Erick Tryzelaar | 2010-02-28 | 1 | -0/+8 |
| | | | | | llvm-svn: 97377 | ||||
| * | Add indirect br support to llvm-c and ocaml. | Erick Tryzelaar | 2010-02-28 | 1 | -0/+13 |
| | | | | | llvm-svn: 97376 | ||||
| * | Add metadata functions to llvm-c and ocaml. | Erick Tryzelaar | 2010-02-28 | 1 | -0/+56 |
| | | | | | llvm-svn: 97375 | ||||
| * | remove redundant instruction. | Chris Lattner | 2010-02-28 | 1 | -2/+2 |
| | | | | | llvm-svn: 97374 | ||||
| * | Add the new builder arthmetic instructions to llvm-c and ocaml. | Erick Tryzelaar | 2010-02-28 | 1 | -0/+81 |
| | | | | | llvm-svn: 97372 | ||||
| * | Add the new union arthmetic instructions to llvm-c and ocaml. | Erick Tryzelaar | 2010-02-28 | 1 | -1/+3 |
| | | | | | llvm-svn: 97371 | ||||
| * | Add generic binary op and cast builder functions. | Erick Tryzelaar | 2010-02-28 | 1 | -0/+13 |
| | | | | | llvm-svn: 97370 | ||||
| * | Revert r97245 which seems to be causing performance problems. | Bob Wilson | 2010-02-28 | 1 | -6/+5 |
| | | | | | llvm-svn: 97366 | ||||
| * | Teach APFloat how to create both QNaNs and SNaNs and with arbitrary-width | John McCall | 2010-02-28 | 2 | -12/+45 |
| | | | | | | | | | payloads. APFloat's internal folding routines always make QNaNs now, instead of sometimes making QNaNs and sometimes SNaNs depending on the type. llvm-svn: 97364 | ||||
| * | Don't try to replace physical registers when doing CSE. | Dan Gohman | 2010-02-28 | 1 | -1/+12 |
| | | | | | llvm-svn: 97360 | ||||
| * | Add a flag to addPassesToEmit* to disable the Verifier pass run | Dan Gohman | 2010-02-28 | 6 | -14/+30 |
| | | | | | | | after LSR, so that clients can opt in. llvm-svn: 97357 | ||||
| * | Implement XMM subregs. | Dan Gohman | 2010-02-28 | 5 | -163/+168 |
| | | | | | | | | | | | | Extracting the low element of a vector is now done with EXTRACT_SUBREG, and the zero-extension performed by load movss is now modeled with SUBREG_TO_REG, and so on. Register-to-register movss and movsd are no longer considered copies; they are two-address instructions which insert a scalar into a vector. llvm-svn: 97354 | ||||
| * | Don't unconditionally suppress hoisting of instructions with implicit | Dan Gohman | 2010-02-28 | 1 | -26/+4 |
| | | | | | | | | | defs or uses. The regular def and use checking below covers them, and can be more precise. It's safe to hoist an instruction with a dead implicit def if the register isn't live into the loop header. llvm-svn: 97352 | ||||
| * | The mayHaveSideEffects flag is no longer used. | Dan Gohman | 2010-02-27 | 12 | -26/+24 |
| | | | | | llvm-svn: 97348 | ||||
| * | remove a bogus pattern, which had the same pattern as STDU | Chris Lattner | 2010-02-27 | 2 | -8/+1 |
| | | | | | | | | | | but codegen'd differently. This really wanted to use some sort of subreg to get the low 4 bytes of the G8RC register or something. However, it's invalid and nothing is testing it, so I'm just zapping the bogosity. llvm-svn: 97345 | ||||

