summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Change hte lexer to start a start pointer to the underlying Chris Lattner2007-07-222-10/+12
| | | | | | | | memorybuffer instead of a pointer to the memorybuffer itself. This reduces coupling and eliminates a pointer dereference on a hot path. This speeds up -Eonly on 483.xalancbmk by 2.1% llvm-svn: 40394
* split the slow path out of Lexer::getSourceLocation and do not let theChris Lattner2007-07-221-14/+27
| | | | | | compiler inline it. This speeds up -Eonly on 483.xalancbmk by about 1% llvm-svn: 40393
* Remove -O1 in favour of making llvm-gcc-4.2 a bitDuncan Sands2007-07-221-2/+1
| | | | | | smarter. llvm-svn: 40391
* Implement a simple cache in headersearch. This speeds upChris Lattner2007-07-222-1/+35
| | | | | | | preprocessing 483.xalancbmk by about 10%, reducing the number of file lookup queries from 2139411 to 199466 (over 10x) llvm-svn: 40390
* Refactor switch analysis to make it possible to detect duplicate case valuesAnders Carlsson2007-07-229-42/+150
| | | | llvm-svn: 40388
* remove redundant testChris Lattner2007-07-221-2/+0
| | | | llvm-svn: 40386
* avoid a small bit of string traffic by not storing the ""'s around a string ↵Chris Lattner2007-07-221-2/+6
| | | | | | in CurFilename llvm-svn: 40385
* avoid recursion between SkipBCPLComment and SkipWhitespace. In cases like this:Chris Lattner2007-07-221-24/+22
| | | | | | | | | | | // foo // bar // baz we'd get two levels of call (bcpl & whitespace) for each line, leading to some seriously deep stacks in some cases. llvm-svn: 40384
* when running in -E mode on multiple files, there is no reason to accumulate Chris Lattner2007-07-222-0/+20
| | | | | | | fileid's and macroid's across files. Clearing between files keeps the tables smaller and slightly speeds up compilation. llvm-svn: 40383
* Catch goto's with a missing identifier, patch by Neil Booth.Chris Lattner2007-07-222-0/+10
| | | | llvm-svn: 40381
* Fix a really subtle bug in the macro expander caching code, whereChris Lattner2007-07-223-5/+15
| | | | | | | redefinition of a macro could cause invalid memory to be deleted. Found preprocessing 253.perlbmk. llvm-svn: 40380
* Fix a lexer bug where we incorrectly rejectedChris Lattner2007-07-212-1/+15
| | | | | | | | int i = /*/ */ 1; Thanks to Neil for pointing this out. llvm-svn: 40379
* This xform isn't safe, removing it.Chris Lattner2007-07-211-10/+0
| | | | llvm-svn: 40378
* Minor simplification to Expr::isLvalue().Steve Naroff2007-07-212-2/+2
| | | | llvm-svn: 40375
* Update for changes in library.shReid Spencer2007-07-211-1/+1
| | | | llvm-svn: 40371
* Add support for reusing macroid's with negative physical loc deltas. ThisChris Lattner2007-07-212-14/+24
| | | | | | keeps the MacroInfo table more compact. llvm-svn: 40281
* implement getSourceRange for ImplicitCastChris Lattner2007-07-211-1/+1
| | | | llvm-svn: 40264
* Return an exit code of 1 if errors occur, not an exit code equal to the # ↵Chris Lattner2007-07-211-1/+1
| | | | | | errors :) llvm-svn: 40189
* minor simplificationsChris Lattner2007-07-211-2/+2
| | | | llvm-svn: 40176
* testcase from neilChris Lattner2007-07-211-0/+6
| | | | llvm-svn: 40173
* recover from void argument types more nicely.Chris Lattner2007-07-211-2/+3
| | | | llvm-svn: 40169
* better wording for an error, suggested by Neil.Chris Lattner2007-07-212-3/+3
| | | | llvm-svn: 40163
* Implement code generation for __func__, __FUNCTION__ and __PRETTY_FUNCTION__Anders Carlsson2007-07-219-10/+102
| | | | llvm-svn: 40162
* improve and simplify error recovery for calls, fix a crash when diagnosingChris Lattner2007-07-212-17/+15
| | | | | | invalid arguments. llvm-svn: 40161
* fix a warning on some compilersChris Lattner2007-07-211-0/+1
| | | | llvm-svn: 40139
* add an evil macro expansion perf test from Neil.Chris Lattner2007-07-211-0/+47
| | | | llvm-svn: 40138
* Fix off-by-one error when emitting diagnostics. Also, make diagnosticChris Lattner2007-07-212-2/+18
| | | | | | | | | | | | | | a bit nicer for people who pass lots of extra arguments to calls by selecting them all instead of just the first one: arg-duplicate.c:13:13: error: too many arguments to function f3 (1, 1, 2, 3, 4); // expected-error {{too many arguments to function}} ^~~~~~~ This implements test/Sema/arg-duplicate.c, thanks to Neil for pointing out this crash. llvm-svn: 40136
* move some casts up to the entry of the function for clarity.Chris Lattner2007-07-211-23/+23
| | | | llvm-svn: 40135
* Two fixes: Chris Lattner2007-07-213-18/+36
| | | | | | | | | | 1) fix a crash on test/Sema/default.c by making sure that the switch scope is non-null. 2) if there is an error sema'ing a default or case stmt, make sure to return the substmt up, so that the error recovery code has more acurate info to continue with. llvm-svn: 40134
* Apply temporary work around to fix llvm mis-compilationDevang Patel2007-07-211-1/+3
| | | | | | reported in PR 1556. llvm-svn: 40133
* No more noResults.Evan Cheng2007-07-2114-57/+48
| | | | llvm-svn: 40132
* Don't assume that only Uses can be kills. Defs are marked as kills initiallyDan Gohman2007-07-203-11/+9
| | | | | | | | when there are no uses. This fixes a dangling-pointer bug, where pointers to deleted instructions were not removed from kills lists. More info here: http://lists.cs.uiuc.edu/pipermail/llvmdev/2007-July/009749.html llvm-svn: 40131
* Simplify the logic for setVolatile.Dan Gohman2007-07-201-2/+2
| | | | llvm-svn: 40130
* Need -O1 or better to have these builtins lowered toDuncan Sands2007-07-201-1/+2
| | | | | | | | llvm intrinsics in llvm-gcc-4.2. This is because get_pointer_alignment bails out: it relies on TER to compute accurate alignment information. llvm-svn: 40128
* zext(undef) = 0 and sext(undef) = 0, not undef.Chris Lattner2007-07-201-1/+6
| | | | | | This hopefully fixes a miscompilation of TargetData.cpp when self hosting. llvm-svn: 40125
* this xform is already done by the constant folder.Chris Lattner2007-07-201-6/+1
| | | | llvm-svn: 40124
* Added -print-emitted-asm to print out JIT generated asm to cerr.Evan Cheng2007-07-2010-23/+41
| | | | llvm-svn: 40123
* Update to include clean and install commands.Reid Spencer2007-07-201-0/+2
| | | | llvm-svn: 40119
* No longer referencing "shared memory" as this can confuse people. The memory is Chandler Carruth2007-07-201-13/+13
| | | | | | | clearly shared between processors if these instructions are being used, no further specification of what type of memory is necessary. llvm-svn: 40118
* Silly HTMLReid Spencer2007-07-201-1/+3
| | | | llvm-svn: 40117
* Fix validation errors.Reid Spencer2007-07-201-16/+10
| | | | llvm-svn: 40116
* This introduces the atomic operation intrinsics into the documentation. This is Chandler Carruth2007-07-201-0/+305
| | | | | | | a preview for the intrinsics that are going to be implemented over the next few weeks. llvm-svn: 40115
* Make the heuristic for shrinking DenseMap smarter.Owen Anderson2007-07-201-2/+4
| | | | llvm-svn: 40114
* Fix a valgrind error noticed by Benoit BoissinotChris Lattner2007-07-203-11/+14
| | | | llvm-svn: 40113
* remove some old cruftChris Lattner2007-07-201-27/+1
| | | | llvm-svn: 40111
* fix a nasty bug Owen noticed in a gcc warning.Chris Lattner2007-07-201-4/+4
| | | | llvm-svn: 40110
* Use SmallVector instead of std::vector.Devang Patel2007-07-202-12/+12
| | | | llvm-svn: 40109
* improve comments, implement a trivial single-entry cache in Chris Lattner2007-07-203-24/+39
| | | | | | | | | SourceManager::getInstantiationLoc. With this change, every token expanded from a macro doesn't get its own MacroID. :) This reduces # macro IDs in carbon.h from 16805 to 9197 llvm-svn: 40108
* Fixing some differences between CVS and SVN diff'ing. Reid fixed these already, Chandler Carruth2007-07-201-2/+2
| | | | | | but I think it got lost in the conversion mess. llvm-svn: 40107
* Update project for LexerToken.h -> Token.hChris Lattner2007-07-201-4/+4
| | | | llvm-svn: 40106
OpenPOWER on IntegriCloud