summaryrefslogtreecommitdiffstats
path: root/llvm
Commit message (Collapse)AuthorAgeFilesLines
* elf writer really wants the size of the global, not the size Chris Lattner2009-01-041-3/+3
| | | | | | of the pointer to the global. llvm-svn: 61630
* Set a few more vimrc indentation options.Dan Gohman2009-01-041-2/+4
| | | | llvm-svn: 61628
* Add utility functions for marking parameters as noalias or nocapture.Nick Lewycky2009-01-041-12/+39
| | | | | | | Clean up some of the existing code by making it use hasFnAttr/addFnAttr and round it off by creating removeFnAttr. llvm-svn: 61627
* XFAIL this test. The xform was removed.Bill Wendling2009-01-041-0/+1
| | | | llvm-svn: 61624
* Revert this transform. It was causing some dramatic slowdowns in a few ↵Bill Wendling2009-01-041-31/+0
| | | | | | tests. See PR3266. llvm-svn: 61623
* Add several more unimplemented operator overloads to ilist_iteratorDan Gohman2009-01-041-4/+16
| | | | | | to help catch errors. llvm-svn: 61622
* The llvm::ELFWriter::EmitGlobal() method is calling theBill Wendling2009-01-041-0/+1
| | | | | | | | | | llvm::PATypeHolder::get() method when LLVM is self-hosted in Release mode. Before the parser changed, there was a definition of llvm::PAHolder::get() in llvmAsmParser.y. This was probably a bug that no-one noticed. Explicitly #include the Type.h file as a temporary fix for now. llvm-svn: 61620
* add #include guards, thanks Dan.Chris Lattner2009-01-041-0/+5
| | | | llvm-svn: 61619
* Use softtabstop instead of redefining tabstop.Dan Gohman2009-01-041-1/+1
| | | | llvm-svn: 61618
* Switch the vimrc file from smartindent to cindent, which isDan Gohman2009-01-041-2/+3
| | | | | | | | smarter about C-ish syntax, and supports the cinoptions variable. Set cinoptions to suppress the extra indentation for switch case labels. llvm-svn: 61617
* Revert r61598 as it does nothing.Bill Wendling2009-01-031-1/+2
| | | | llvm-svn: 61614
* Fix a DAGCombiner abort on an invalid shift count constant. This fixes PR3250.Dan Gohman2009-01-032-0/+19
| | | | llvm-svn: 61613
* CommuteNodesToReducePressure() is now removed.Dan Gohman2009-01-031-1/+0
| | | | llvm-svn: 61612
* Remove the code from the scheduler that commuted two-addressDan Gohman2009-01-033-74/+0
| | | | | | | | | | | | | | instructions to avoid copies, because TwoAddressInstructionPass also does this optimization. The scheduler's version didn't account for live-out values, which resulted in spurious commutes and missed opportunities. Now, TwoAddressInstructionPass handles all the opportunities, instead of just those that the scheduler missed. The result is usually the same, though there are occasional trivial differences resulting from the avoidance of spurious commutes. llvm-svn: 61611
* Any void readonly functions are provably dead, don't waste time adding Nick Lewycky2009-01-031-14/+0
| | | | | | nocapture attributes to them. llvm-svn: 61610
* improve test and address Misha's commentsNuno Lopes2009-01-031-43/+34
| | | | llvm-svn: 61609
* Add Intel processors core i7 and atom.Evan Cheng2009-01-032-1/+4
| | | | llvm-svn: 61603
* Fix PR3210: Detect more Intel processors. Patch by Torok Edwin.Evan Cheng2009-01-031-2/+15
| | | | llvm-svn: 61602
* * Wrap command line flag in <tt>Misha Brukman2009-01-031-3/+7
| | | | | | * Wrapped long lines in code section llvm-svn: 61601
* We know it's always a SCEVConstant if it gets here, so just cast it andNick Lewycky2009-01-031-6/+2
| | | | | | inline the only use of isNegative. Fixes warning reported by Mike Stump. llvm-svn: 61600
* Setting BUILD_ARCHIVE to 0 has the same effect as defining it to 1.Bill Wendling2009-01-031-2/+0
| | | | llvm-svn: 61599
* Make sure that 'ranlib' runs only after 'ar' is completed.Bill Wendling2009-01-031-2/+1
| | | | llvm-svn: 61598
* CellSPU:Scott Michel2009-01-034-110/+123
| | | | | | | | | - Remove custom lowering for BRCOND - Add remaining functionality for branches in SPUInstrInfo, such as branch condition reversal and load/store folding. Updated BrCond test to reflect branch reversal. llvm-svn: 61597
* Reassign the buffer to the pointer so that we don't overwrite memory.Bill Wendling2009-01-021-0/+2
| | | | llvm-svn: 61596
* Alphabetized #includes.Misha Brukman2009-01-022-4/+3
| | | | llvm-svn: 61595
* Down with trailing whitespace!Misha Brukman2009-01-025-67/+66
| | | | llvm-svn: 61594
* VerifyLineLength() actually takes a max length parameter.Misha Brukman2009-01-021-1/+3
| | | | llvm-svn: 61593
* Added some basic lint tools for C++ and generic lint tool applicable to allMisha Brukman2009-01-024-0/+168
| | | | | | types of files (TableGen, LLVM assembly, HTML files, etc.) llvm-svn: 61592
* - Make copyRegToReg use the "LR" assembler synonym for "OR". Makes findingScott Michel2009-01-021-38/+50
| | | | | | | register copies a little easier to pick out from the output. - Fix bug 3192. llvm-svn: 61591
* * Alphabetized system headers per the style guideMisha Brukman2009-01-024-12/+10
| | | | | | * Minor spacing and comment cleanups llvm-svn: 61590
* Don't try to analyze this "backward" case. This is overly conservative Nick Lewycky2009-01-022-0/+52
| | | | | | pending a correct solution. llvm-svn: 61589
* So *this* is where all the trailing whitespace in file header comments comeMisha Brukman2009-01-021-2/+2
| | | | | | | | from: copy-pasted straight from the coding standards doc! Oh, the irony. llvm-svn: 61586
* Remove comma at end of enumerator list.Daniel Dunbar2009-01-021-1/+1
| | | | llvm-svn: 61585
* Remove bison specific Makefile bits for AsmParser.Daniel Dunbar2009-01-021-12/+0
| | | | llvm-svn: 61584
* Deleted trailing whitespace; no functional changes.Misha Brukman2009-01-021-61/+61
| | | | llvm-svn: 61583
* vimrcMisha Brukman2009-01-023-7/+20
| | | | | | | | | | | | * Fixed cursors in terminal by setting nocompatible (sorry, vi users) * Enable syntax highlighting so that this file can stand on its own * Highlight trailing whitespace * Fixed commands to delete trailing whitespaces and convert tabs to spaces llvm.vim and tablegen.vim * Removed trailing whitespace, as it's now very visible llvm-svn: 61582
* We also removed gtest-all.cc from the distribution tarball.Misha Brukman2009-01-021-3/+2
| | | | llvm-svn: 61581
* Load tracking means that the value analyzed mayDuncan Sands2009-01-021-2/+8
| | | | | | | | | | | | not have pointer type. In particular, it may be the condition argument for a select or a GEP index. While I was unable to construct a testcase for which some bits of the original pointer are captured due to one of these, it's very very close to being possible - so play safe and exclude these possibilities. llvm-svn: 61580
* fist short at a new unit test for ImmutableSets. no bugs found, though :PNuno Lopes2009-01-021-0/+208
| | | | llvm-svn: 61576
* make 'make clean' remove test binaries as wellNuno Lopes2009-01-021-0/+3
| | | | llvm-svn: 61572
* When calculating 'nocapture' argument attributes, allowDuncan Sands2009-01-022-21/+83
| | | | | | | | | | | | the argument to be stored to an alloca by tracking uses of the alloca. This occurs 4 times (out of 7121, 0.05%) in MultiSource/Applications, so may not be worth it. On the other hand, it is easy to do and fairly cheap. The functions it helps are: W_addcom and W_addlit in spiff; process_args (argv) in d (make_dparser); ercPixConcealIMB in JM/ldecod. llvm-svn: 61570
* Improve comments and reorganize a bit - no functionalityDuncan Sands2009-01-021-56/+44
| | | | | | change. llvm-svn: 61569
* Fix a really horrible typo, which caused undefined behavior.Chris Lattner2009-01-021-1/+1
| | | | llvm-svn: 61566
* minor cleanups and comment improvements.Chris Lattner2009-01-022-196/+149
| | | | llvm-svn: 61564
* add a #include to hopefully get the x86-64-linux buildbot building.Chris Lattner2009-01-021-0/+1
| | | | llvm-svn: 61563
* Remove lex/bison support from makefile.rules.Chris Lattner2009-01-021-111/+5
| | | | llvm-svn: 61562
* update the cmakefile. This is a "best guess", I haven't tested this.Chris Lattner2009-01-021-28/+2
| | | | llvm-svn: 61561
* lex and bison aren't needed to build llvm anymore.Chris Lattner2009-01-024-31/+2
| | | | llvm-svn: 61560
* new non-ambiguous syntax.Chris Lattner2009-01-021-1/+1
| | | | llvm-svn: 61559
* Reimplement the old and horrible bison parser for .ll files with a niceChris Lattner2009-01-0237-15779/+3956
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | and clean recursive descent parser. This change has a couple of ramifications: 1. The parser code is about 400 lines shorter (in what we maintain, not including what is autogenerated). 2. The code should be significantly faster than the old code because we don't have to work around bison's poor handling of datatypes with ctors/dtors. This also makes the code much more resistant to memory leaks. 3. We now get caret diagnostics from the .ll parser, woo. 4. The actual diagnostics emited from the parser are completely different so a bunch of testcases had to be updated. 5. I now disallow "%ty = type opaque %ty = type i32". There was no good reason to support this, it was just an accident of the old implementation. I have no reason to think that anyone is actually using this. 6. The syntax for sticking a global variable has changed to make it unambiguous. I don't think anyone is depending on this since only clang supports this and it is not solid yet, so I'm not worried about anything breaking. 7. This gets rid of the last use of bison, and along with it the .cvs files. I'll prune this from the makefiles as a subsequent commit. There are a few minor cleanups that can be done after this commit (suggestions welcome!) but this passes dejagnu testing and is ready for its time in the limelight. llvm-svn: 61558
OpenPOWER on IntegriCloud