summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Move SplitBlockPredecessors out of loopsimplify into BasicBlockUtils.hChris Lattner2008-04-213-107/+128
| | | | | | | | | as a global helper function. At the same type, switch it from taking a vector of predecessors to an arbitrary sequential input. This allows us to switch LoopSimplify to use a SmallVector for various temporary vectors that it passed into SplitBlockPredecessors. llvm-svn: 50020
* Move domtree/frontier updating earlier, allowing us to use it to update phi Chris Lattner2008-04-211-31/+18
| | | | | | nodes, removing a hack. llvm-svn: 50019
* Factor dominator tree and frontier updating into SplitBlockPredecessorsChris Lattner2008-04-211-18/+14
| | | | | | instead of doing it after every call. llvm-svn: 50018
* fit some more code in 80 cols.Chris Lattner2008-04-211-12/+14
| | | | llvm-svn: 50016
* simplify code, fit in 80 cols.Chris Lattner2008-04-211-65/+67
| | | | llvm-svn: 50015
* fit in 80 colsChris Lattner2008-04-211-6/+6
| | | | llvm-svn: 50014
* Fix improper dereference of end() iterator. Patch by Argiris Kirtzidis!Ted Kremenek2008-04-201-2/+5
| | | | llvm-svn: 50012
* Be pessimistic in computing the buffer size when aligning.Nicolas Geoffray2008-04-202-7/+11
| | | | llvm-svn: 50008
* Don't forget to update the current operand when getting the size of an ↵Nicolas Geoffray2008-04-201-4/+14
| | | | | | instruction. llvm-svn: 50007
* finish the first cut of a jump threading pass implementation.Chris Lattner2008-04-201-21/+141
| | | | llvm-svn: 50006
* replace a slow and verbose version of Instruction::isUsedOutsideOfBlock withChris Lattner2008-04-201-34/+4
| | | | | | a call to Instruction::isUsedOutsideOfBlock. llvm-svn: 50005
* add a handy helper method to instruction, useful for determiningChris Lattner2008-04-202-0/+30
| | | | | | | whether it is used outside of some block. This can be used to see if there are any non-local references, for example. llvm-svn: 50004
* we can only thread blocks when there is a pred we can determine the succ of.Chris Lattner2008-04-201-2/+17
| | | | llvm-svn: 50003
* improve comments, infrastructure, and add some validity checks for threading.Chris Lattner2008-04-201-9/+106
| | | | | | Add a cost function. llvm-svn: 50002
* Fix PR2236Nate Begeman2008-04-203-8/+26
| | | | | | | Add test Implement feedback from Chris re: PR1998 llvm-svn: 50001
* Add a new Jump Threading pass, which will handle casesChris Lattner2008-04-203-0/+60
| | | | | | | such as those in PR2235. Right now the pass is not very effective. :) llvm-svn: 50000
* hopefully resolve PR2240Chris Lattner2008-04-201-1/+1
| | | | llvm-svn: 49999
* Two improvements to initializer parsing:Chris Lattner2008-04-202-5/+27
| | | | | | | | 1. If we hit a semantic error, try harder to recover to emit diagnostics for later initializer errors (PR2241). 2. Don't leak parsed initializers on an error. llvm-svn: 49998
* for exponent-related errors, emit the diagnostic on the 'e' or 'p'.Chris Lattner2008-04-201-4/+7
| | | | llvm-svn: 49997
* when emitting diagnostics about some character in a numericChris Lattner2008-04-201-20/+26
| | | | | | | | | | | literal, indicate what character the error is about or where it would be. For example: pr2241.c:1:17: error: exponent has no digits float f[] = { 1e , 1ee0 }; ^ llvm-svn: 49996
* rearrange some code, simplify handling of shifts.Chris Lattner2008-04-201-60/+56
| | | | llvm-svn: 49995
* Check we aren't trying to convert PPC long double.Dale Johannesen2008-04-202-7/+7
| | | | | | This fixes the testsuite failure on ppcf128-4.ll. llvm-svn: 49994
* Cosmetic changes, as suggested by Evan. No functionality changes.Nicolas Geoffray2008-04-204-104/+81
| | | | llvm-svn: 49993
* (re)fix handling of UGT. Pointed out by Nick Lewycky.Dale Johannesen2008-04-201-1/+2
| | | | llvm-svn: 49991
* g++-4.3 build-fix: CHAR_BIT requires <climits>.Torok Edwin2008-04-201-0/+1
| | | | llvm-svn: 49989
* Do not hold the JIT lock when materializing a function and verify if theNicolas Geoffray2008-04-201-1/+6
| | | | | | | function has already been codegen'd. This is required by the Java class loading mechanism which executes Java code when materializing a function. llvm-svn: 49988
* Only generate code for static functions and global variables that are ↵Nate Begeman2008-04-203-5/+78
| | | | | | | | | actually used. Warn about unused static functions and variables. This fixes PR1998. llvm-svn: 49987
* A better fix for my previous patch, MOVZQI2PQIrr just requires SSE2.Chris Lattner2008-04-202-3/+3
| | | | llvm-svn: 49986
* Not all x86-64 machines have sse3 apparently.Chris Lattner2008-04-201-2/+2
| | | | llvm-svn: 49985
* Restructure and simplify Sema::ObjCQualifiedIdTypesAreCompatible,Chris Lattner2008-04-201-65/+64
| | | | | | | | | | deleting dead code, making other code more obvious, and fixing a bug. This fixes these (bogus) warnings in release mode: SemaExprObjC.cpp:383: warning: 'RHSProtoE' may be used uninitialized in this function SemaExprObjC.cpp:383: warning: 'RHSProtoI' may be used uninitialized in this function llvm-svn: 49984
* Check that APFloat::convert is not trying to targetDale Johannesen2008-04-201-0/+1
| | | | | | | | | ppc long double, which doesn't work. This may break some stuff temporarily, but I want to avoid the propagation of code that assumes this works. llvm-svn: 49983
* html::EscapeText (using the Rewriter) now replaces tabs by default.Ted Kremenek2008-04-202-11/+3
| | | | llvm-svn: 49982
* implement a fixme, add support for ++/-- on long double.Chris Lattner2008-04-201-5/+6
| | | | llvm-svn: 49980
* use simplified ConstantFP:get method.Chris Lattner2008-04-202-6/+4
| | | | llvm-svn: 49979
* regenerateChris Lattner2008-04-203-276/+276
| | | | llvm-svn: 49978
* Switch to using Simplified ConstantFP::get API.Chris Lattner2008-04-209-66/+60
| | | | llvm-svn: 49977
* Use simplified ConstantFP::get method, fix a bug handling frem x, 0 with ↵Chris Lattner2008-04-202-16/+19
| | | | | | long doubles. llvm-svn: 49976
* replace form feeds with an <hr> tag.Chris Lattner2008-04-191-1/+5
| | | | llvm-svn: 49975
* simplify the return of InitializePreprocessor to be bool instead of a fileidChris Lattner2008-04-191-10/+9
| | | | llvm-svn: 49974
* simplify ownership of the predefines buffer.Chris Lattner2008-04-194-30/+15
| | | | llvm-svn: 49973
* free(null) -> noopChris Lattner2008-04-191-2/+2
| | | | llvm-svn: 49972
* rename *.llx -> *.ll, last batch.Chris Lattner2008-04-1946-0/+0
| | | | llvm-svn: 49971
* rename *.llx -> *.llChris Lattner2008-04-1929-0/+0
| | | | llvm-svn: 49970
* rename *.llx -> *.llChris Lattner2008-04-1936-0/+0
| | | | llvm-svn: 49969
* merge of 49966 from branches/ggreif/use-diet to trunk. these are already ↵Gabor Greif2008-04-198-48/+48
| | | | | | active API changes llvm-svn: 49968
* Implement PR2206.Chris Lattner2008-04-192-8/+21
| | | | llvm-svn: 49967
* refactor handling of symbolic constant folding, picking upChris Lattner2008-04-192-66/+66
| | | | | | | a few new cases( see Integer/a1.ll), but not anything that would happen in practice. llvm-svn: 49965
* indentation fix.Chris Lattner2008-04-191-14/+14
| | | | llvm-svn: 49964
* Some clarifications suggested by Jon SargeantChris Lattner2008-04-191-8/+12
| | | | llvm-svn: 49963
* Allow argpromote to promote struct arguments with a specified numberChris Lattner2008-04-192-12/+20
| | | | | | of elements. Patch by Matthijs Kooijman! llvm-svn: 49962
OpenPOWER on IntegriCloud