| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Fix PR1146: parameter attributes are longer part of | Duncan Sands | 2007-11-27 | 6 | -97/+105 |
| | | | | | | | | | | | | | the function type, instead they belong to functions and function calls. This is an updated and slightly corrected version of Reid Spencer's original patch. The only known problem is that auto-upgrading of bitcode files doesn't seem to work properly (see test/Bitcode/AutoUpgradeIntrinsics.ll). Hopefully a bitcode guru (who might that be? :) ) will fix it. llvm-svn: 44359 | ||||
| * | Fix PR1816. If a bitcast of a function only exists because of a | Duncan Sands | 2007-11-25 | 2 | -31/+51 |
| | | | | | | | | | | trivial difference in function attributes, allow calls to it to be converted to direct calls. Based on a patch by Török Edwin. While there, move the various lists of mutually incompatible parameters etc out of the verifier and into ParameterAttributes.h. llvm-svn: 44315 | ||||
| * | Rename the 'const' parameter attribute to 'readnone', | Duncan Sands | 2007-11-22 | 2 | -6/+6 |
| | | | | | | | | and the 'pure' parameter attribute to 'readonly'. Names suggested by DannyB. llvm-svn: 44273 | ||||
| * | In order for parameter attribute uniquing to make | Duncan Sands | 2007-11-20 | 1 | -2/+6 |
| | | | | | | | | | | | | | any sense it is important that ParamAttr::None gets treated the same as not supplying an attribute at all. Rather than stripping ParamAttr::None out of the list of attributes, assert if ParamAttr::None is seen. Fix up the bitcode reader which liked to insert ParamAttr::None all over the place. Patch based on one by Török Edwin. llvm-svn: 44250 | ||||
| * | Remove meaningless qualifiers from return types, avoiding compiler warnings. | Dan Gohman | 2007-11-19 | 1 | -1/+1 |
| | | | | | llvm-svn: 44240 | ||||
| * | print a call to a fastcc function as: | Chris Lattner | 2007-11-18 | 1 | -2/+2 |
| | | | | | | | | | | | call x86_fastcallcc void @func( i32* %X, i64 0 ) not: callx86_fastcallcc void @func( i32* %X, i64 0 ) This fixes Codegen/X86/fast-cc-merge-stack-adj.ll llvm-svn: 44217 | ||||
| * | Add support for vectors to int <-> float casts. | Nate Begeman | 2007-11-17 | 4 | -17/+91 |
| | | | | | llvm-svn: 44204 | ||||
| * | Basic non-power-of-2 vector support | Nate Begeman | 2007-11-15 | 1 | -1/+0 |
| | | | | | llvm-svn: 44181 | ||||
| * | I discover array_lengthof, thanks to gabor on #llvm. | Duncan Sands | 2007-11-14 | 1 | -2/+1 |
| | | | | | llvm-svn: 44139 | ||||
| * | Simplify the attribute verification code. | Duncan Sands | 2007-11-14 | 1 | -34/+24 |
| | | | | | llvm-svn: 44116 | ||||
| * | Add pure/const attributes. Documentation will follow. | Anton Korobeynikov | 2007-11-14 | 2 | -4/+17 |
| | | | | | llvm-svn: 44109 | ||||
| * | Don't barf on empty basic blocks. Do not rely on assert | Duncan Sands | 2007-11-01 | 1 | -9/+21 |
| | | | | | | | | | | | doing something - this needs to work for release builds too. I chose to just abort rather than following the fancy logic of abortIfBroken, because (1) it is a pain to do otherwise, and (2) nothing is going to work if the module is this broken. llvm-svn: 43611 | ||||
| * | remove verifier command line option: this should be part of the API, not | Chris Lattner | 2007-11-01 | 1 | -11/+0 |
| | | | | | | | a command line optn. llvm-svn: 43603 | ||||
| * | Now with less tabs! | Owen Anderson | 2007-11-01 | 1 | -5/+5 |
| | | | | | llvm-svn: 43601 | ||||
| * | Fix InstCombine/2007-10-31-StringCrash.ll by removing an obvious | Chris Lattner | 2007-11-01 | 1 | -12/+8 |
| | | | | | | | (in hindsight) infinite recursion. Simplify the code. llvm-svn: 43597 | ||||
| * | Fix a regression in test/CodeGen/X86/2007-04-24-VectorCrash.ll introduced | Dan Gohman | 2007-10-31 | 1 | -4/+3 |
| | | | | | | | | by r43510. Gracefully handle constants with vector type that aren't ConstantVector or ConstantAggregateZero. llvm-svn: 43579 | ||||
| * | Add a preverifier pass to check that every basic block ends in a terminator, ↵ | Owen Anderson | 2007-10-31 | 1 | -1/+20 |
| | | | | | | | so that we don't segfault when verifying invalid code. llvm-svn: 43578 | ||||
| * | Add support for folding binary operators with vector zero operands. | Dan Gohman | 2007-10-30 | 1 | -19/+29 |
| | | | | | llvm-svn: 43510 | ||||
| * | I am not sure this is a good idea to be an option, but rename the option for | Chris Lattner | 2007-10-29 | 1 | -1/+1 |
| | | | | | | | now. It conflicts with clang's -pedantic flag. llvm-svn: 43431 | ||||
| * | Add 'pedantic' mode to verifier rejecting syntactically valid, but 'bad' due ↵ | Anton Korobeynikov | 2007-10-28 | 1 | -3/+14 |
| | | | | | | | to other reasons code llvm-svn: 43424 | ||||
| * | Make DomTree and PostDomTree thin wrappers around DomTreeBase, rather than ↵ | Owen Anderson | 2007-10-23 | 1 | -11/+1 |
| | | | | | | | inheriting from it. llvm-svn: 43259 | ||||
| * | The #include <iterator> isn't needed in this header. | Dan Gohman | 2007-10-22 | 1 | -3/+0 |
| | | | | | llvm-svn: 43232 | ||||
| * | this doesn't need dynamic_cast. | Chris Lattner | 2007-10-18 | 1 | -1/+1 |
| | | | | | llvm-svn: 43133 | ||||
| * | Reduce reliance on rtti info | Chris Lattner | 2007-10-18 | 1 | -2/+1 |
| | | | | | llvm-svn: 43130 | ||||
| * | fix typo | Chris Lattner | 2007-10-18 | 1 | -1/+1 |
| | | | | | llvm-svn: 43129 | ||||
| * | Define a helper function ConstantVector::getSplatValue for testing for | Dan Gohman | 2007-10-17 | 1 | -0/+11 |
| | | | | | | | and working with broadcasted constants. llvm-svn: 43076 | ||||
| * | Move splitBlock into DomTreeBase from DomTree. | Owen Anderson | 2007-10-17 | 1 | -82/+0 |
| | | | | | llvm-svn: 43059 | ||||
| * | Disable attempts to constant fold PPC f128. | Dale Johannesen | 2007-10-16 | 1 | -0/+4 |
| | | | | | | | | Remove the assumption that this will happen from various places. llvm-svn: 43053 | ||||
| * | Template DominatorTreeBase by node type. This is the next major step towards | Owen Anderson | 2007-10-16 | 1 | -199/+5 |
| | | | | | | | having dominator information on MBB's. llvm-svn: 43036 | ||||
| * | Initial infrastructure for arbitrary precision integer | Duncan Sands | 2007-10-16 | 1 | -15/+7 |
| | | | | | | | | | | | codegen support. This should have no effect on codegen for other types. Debatable bits: (1) the use (abuse?) of a set in SDNode::getValueTypeList; (2) the length of getTypeToTransformTo, which maybe should be refactored with a non-inline part for extended value types. llvm-svn: 43030 | ||||
| * | avoid an APFloat copy. | Chris Lattner | 2007-10-15 | 1 | -1/+1 |
| | | | | | llvm-svn: 42979 | ||||
| * | Disable some compile-time optimizations on PPC | Dale Johannesen | 2007-10-14 | 1 | -0/+12 |
| | | | | | | | long double. llvm-svn: 42958 | ||||
| * | Next PPC long double bits. First cut at constants. | Dale Johannesen | 2007-10-11 | 2 | -2/+13 |
| | | | | | | | | | | No compile-time support for constant operations yet, just format transformations. Make readers and writers work. Split constants into 2 doubles in Legalize. llvm-svn: 42865 | ||||
| * | C and Objective Caml bindings for PHINode::addIncoming etc. | Gordon Henriksen | 2007-10-08 | 1 | -1/+21 |
| | | | | | llvm-svn: 42760 | ||||
| * | Remove debugging printf to get build going on x86-64. | Evan Cheng | 2007-10-08 | 1 | -2/+0 |
| | | | | | llvm-svn: 42757 | ||||
| * | Add explicit keywords. | Dan Gohman | 2007-10-08 | 1 | -5/+7 |
| | | | | | llvm-svn: 42747 | ||||
| * | Begin the process of allowing DomTree on MBB's. Step One: template ↵ | Owen Anderson | 2007-10-08 | 1 | -16/+0 |
| | | | | | | | DomTreeNode by making it a typedef of a templated DomTreeNodeBase. llvm-svn: 42743 | ||||
| * | C and Objective Caml bindings for getFunction and getNamedGlobal. Also enhanced | Gordon Henriksen | 2007-10-08 | 1 | -0/+8 |
| | | | | | | | | the Objective Caml 'declare_*' functions to behave more or less like getOrInsertFunction. llvm-svn: 42740 | ||||
| * | C and Objective Caml bindings for GlobalVariable::isConstant. | Gordon Henriksen | 2007-10-07 | 1 | -0/+11 |
| | | | | | llvm-svn: 42736 | ||||
| * | convertFromInteger, as originally written, expected sign-extended | Neil Booth | 2007-10-07 | 1 | -1/+1 |
| | | | | | | | | | input. APInt unfortunately zero-extends signed integers, so Dale modified the function to expect zero-extended input. Make this assumption explicit in the function name. llvm-svn: 42732 | ||||
| * | C and Objective Caml bindings for PATypeHolder. | Gordon Henriksen | 2007-10-07 | 1 | -0/+18 |
| | | | | | llvm-svn: 42713 | ||||
| * | Adopting a uniform naming convention for type constructors in bindings. | Gordon Henriksen | 2007-10-06 | 1 | -10/+10 |
| | | | | | llvm-svn: 42698 | ||||
| * | Adopting a uniform naming convention for constant constructors in bindings. | Gordon Henriksen | 2007-10-06 | 1 | -13/+12 |
| | | | | | llvm-svn: 42697 | ||||
| * | Adding C and Ocaml bindings for ConstantExpr. | Gordon Henriksen | 2007-10-06 | 2 | -2/+203 |
| | | | | | llvm-svn: 42696 | ||||
| * | Wrapping Value::dump. | Gordon Henriksen | 2007-10-06 | 1 | -0/+4 |
| | | | | | llvm-svn: 42668 | ||||
| * | Removing the silly CHelpers header by rolling wrap and unwrap into the C | Gordon Henriksen | 2007-10-05 | 1 | -14/+0 |
| | | | | | | | bindings headers themselves, hidden behind #ifdef __cplusplus. llvm-svn: 42666 | ||||
| * | Fix bug in updating dominance frontier after loop | Devang Patel | 2007-10-05 | 1 | -2/+5 |
| | | | | | | | | unswitch when frontier includes basic blocks that are not inside loop. llvm-svn: 42654 | ||||
| * | Completely merge the implementation details of DomTree and PostDomTree. | Owen Anderson | 2007-10-03 | 2 | -108/+5 |
| | | | | | | | Also, add a FIXME for a bug in PostDomTree calculation I noticed while writing this, llvm-svn: 42593 | ||||
| * | Use empty() member functions when that's what's being tested for instead | Dan Gohman | 2007-10-03 | 1 | -2/+2 |
| | | | | | | | of comparing begin() and end(). llvm-svn: 42585 | ||||
| * | Fix a using namespace llvm; in a header file. | Dan Gohman | 2007-10-03 | 1 | -0/+1 |
| | | | | | llvm-svn: 42584 | ||||

