summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Fix a bug reported by Kelly Wilson, where we incorrectlyChris Lattner2008-06-302-4/+17
| | | | | | rejected FP immediates like 08.123 llvm-svn: 52890
* refactor some code out into a new method.Chris Lattner2008-06-302-83/+106
| | | | llvm-svn: 52889
* check in anton's patch to make inlining happen in a determinstic orderChris Lattner2008-06-301-4/+6
| | | | | | | and fix the bug that it uncovers: inlining a pattern fragment could bring in other pattern fragments if the inlinee hadn't already been inlined. llvm-svn: 52888
* Implement split and scalarize for SELECT_CC, fixing PR2504Chris Lattner2008-06-302-0/+32
| | | | llvm-svn: 52887
* Because of the laziness, the required passes must exist even afterNick Lewycky2008-06-301-2/+2
| | | | | | | | | LoopVR::runOnFunction runs. This should accomplish that, but it doesn't. I think that's a PassManager bug, but without a consumer of LoopVR in the tree, I can't give steps to reproduce. llvm-svn: 52886
* Add a value range analysis that lazily computes ranges using ScalarEvolutions.Nick Lewycky2008-06-303-0/+381
| | | | llvm-svn: 52885
* Revert (52748 and friends):Anton Korobeynikov2008-06-2911-141/+158
| | | | | | | | | | | | Move GetConstantStringInfo to lib/Analysis. Remove string output routine from Constant. Update all callers. Change debug intrinsic api slightly to accomodate move of routine, these now return values instead of strings. This unbreaks llvm-gcc bootstrap. llvm-svn: 52884
* make type attribute processing static instead of methods on Sema.Chris Lattner2008-06-292-48/+43
| | | | llvm-svn: 52881
* make the rest of the decl attribute processing methods beChris Lattner2008-06-292-179/+160
| | | | | | static functions instead of methods on sema. llvm-svn: 52880
* make most of Sema public. Sema is a class private to the Sema library Chris Lattner2008-06-292-39/+37
| | | | | | | anyway, so there is no real loss here. Start making attribute processing methods static functions instead of methods on Sema. llvm-svn: 52879
* move a few methods, no other change.Chris Lattner2008-06-291-74/+75
| | | | llvm-svn: 52878
* handle type attributes when converting types, not when processing decls.Chris Lattner2008-06-292-16/+6
| | | | llvm-svn: 52877
* Make ProcessDeclAttributes walk the declarator structure pulling Chris Lattner2008-06-294-21/+60
| | | | | | | | | | | | | decl attributes out of the various places they can hide. This makes us correctly reject things like this: t.c:2:22: error: mode attribute only supported for integer and floating-point types int **__attribute((mode(HI)))* i32; ^ because you can't make a pointer be HImode. llvm-svn: 52876
* move some code from all callers of ProcessDeclAttributes into Chris Lattner2008-06-293-18/+11
| | | | | | the implementation of ProcessDeclAttributes. llvm-svn: 52875
* more attribute refactoring/renaming, no functionality change.Chris Lattner2008-06-283-48/+53
| | | | llvm-svn: 52874
* more minor tidiness.Chris Lattner2008-06-281-17/+12
| | | | llvm-svn: 52873
* more cleanups, refactor HandleVectorTypeAttributeChris Lattner2008-06-283-44/+46
| | | | | | | | to work list the rest of the attr handlers. Also, rename it to HandleVectorSizeAttribute to match its attr name. No functionality change. llvm-svn: 52872
* adjust the prototypes of a bunch of decl processing methods to take Chris Lattner2008-06-283-157/+158
| | | | | | | | | | | the single attribute they look at by reference instead of by pointer. This is a subtle indicator that they take the specified attribute, not a whole list of them. This also make HandleExtVectorTypeAttribute work the same way as the rest of the attributes, adds some comments etc. No functionality change. llvm-svn: 52871
* Start refactoring of asmprinters: provide a TAI hook, which will select a ↵Anton Korobeynikov2008-06-282-2/+69
| | | | | | 'section kind' for a global. llvm-svn: 52868
* UnbreakAnton Korobeynikov2008-06-281-2/+2
| | | | llvm-svn: 52866
* Temporary rever invalid commitAnton Korobeynikov2008-06-282-7/+9
| | | | llvm-svn: 52865
* Move printing of module-level GVs into dedicated helperAnton Korobeynikov2008-06-282-197/+200
| | | | llvm-svn: 52864
* Use common naming conventionAnton Korobeynikov2008-06-282-4/+4
| | | | llvm-svn: 52863
* Factor out stuff into helper functionAnton Korobeynikov2008-06-282-33/+39
| | | | llvm-svn: 52862
* CleanupAnton Korobeynikov2008-06-281-6/+3
| | | | llvm-svn: 52861
* Remove X86SharedAsmPrinterAnton Korobeynikov2008-06-284-529/+487
| | | | llvm-svn: 52860
* whitespace cleanupAnton Korobeynikov2008-06-282-25/+25
| | | | llvm-svn: 52859
* Make intel asmprinter child of generic asmprinter, not x86 shared asm ↵Anton Korobeynikov2008-06-284-27/+138
| | | | | | printer. This leads to some code duplication, which will be resolved later. llvm-svn: 52858
* CleanupAnton Korobeynikov2008-06-281-10/+5
| | | | llvm-svn: 52857
* Whitespace cleanupAnton Korobeynikov2008-06-281-22/+22
| | | | llvm-svn: 52856
* Handle unnamed bitfields when parsing C++ classes.Argyrios Kyrtzidis2008-06-282-26/+31
| | | | llvm-svn: 52855
* Really fix the bootstrap failure.Chris Lattner2008-06-281-9/+5
| | | | llvm-svn: 52854
* Fix for PR2501; this patch makes usual arithmetic conversions for Eli Friedman2008-06-286-5/+97
| | | | | | | | integers which have the same width and different signedness work correctly. (The testcase in PR2501 uses a comparison between long and unsigned int). llvm-svn: 52853
* Replace CurFunctionDecl and CurMethodDecl with methods getCurFunctionDecl() ↵Argyrios Kyrtzidis2008-06-288-52/+50
| | | | | | and getCurMethodDecl() that return the appropriate Decl through CurContext. llvm-svn: 52852
* Add back the capability to include nul characters in strings with Chris Lattner2008-06-283-8/+14
| | | | | | | GetConstantStringInfo. This will hopefully restore llvm-gcc to happy bootstrap land. llvm-svn: 52851
* Tighten up checking.Chris Lattner2008-06-281-4/+10
| | | | llvm-svn: 52850
* When folding a bitcast into a load or store, preserve the alignmentDan Gohman2008-06-281-2/+2
| | | | | | | information of the original load or store, which is checked to be at least as good, and possibly better. llvm-svn: 52849
* Updated checker build to checker-47.Ted Kremenek2008-06-281-1/+1
| | | | llvm-svn: 52848
* Fix GetMainExecutable. Patch by Sam Bishop.Seo Sanghyeon2008-06-271-4/+2
| | | | llvm-svn: 52847
* Fix a bug where we didn't promote 'const float' (or typedefs) toChris Lattner2008-06-272-5/+12
| | | | | | double in some places. llvm-svn: 52846
* the 'mode' attribute is a decl attribute, not a type attribute. Move it toChris Lattner2008-06-274-137/+132
| | | | | | SemaDeclAttr, and do some cleanups. llvm-svn: 52844
* Looks like this condition is inverted.Evan Cheng2008-06-271-2/+2
| | | | llvm-svn: 52841
* Nuno points out that my numbers were out of dateChris Lattner2008-06-271-1/+1
| | | | llvm-svn: 52840
* Add a NextPowerOf2 function to calculate the next power of two greater than ↵Owen Anderson2008-06-271-0/+12
| | | | | | a given integer. llvm-svn: 52839
* implement some fixme's by making "autorenaming" in the value symbol table notChris Lattner2008-06-272-32/+11
| | | | | | thrash the heap with string stuff (e.g. utostr). llvm-svn: 52838
* simplify some code to avoid string thrashing.Chris Lattner2008-06-271-3/+5
| | | | llvm-svn: 52837
* Use StringSet instead of std::set<std::string>Anton Korobeynikov2008-06-273-26/+25
| | | | llvm-svn: 52836
* add a helper.Chris Lattner2008-06-271-0/+2
| | | | llvm-svn: 52835
* Add a new version of Module::getFunction that takes a const char* insteadChris Lattner2008-06-274-0/+15
| | | | | | | of a std::string. This avoids copying the string to the heap in common cases. Patch by Pratik Solanki! llvm-svn: 52834
* making progress!Chris Lattner2008-06-271-1/+1
| | | | llvm-svn: 52833
OpenPOWER on IntegriCloud