| Commit message (Collapse) | Author | Age | Files | Lines | ||
|---|---|---|---|---|---|---|
| ... | ||||||
| * | Match changes to unix build system. | Jeff Cohen | 2006-02-14 | 1 | -1/+1 | |
| | | | | | llvm-svn: 26167 | |||||
| * | now that libcalls don't suck, we can remove this hack | Chris Lattner | 2006-02-14 | 1 | -6/+2 | |
| | | | | | llvm-svn: 26164 | |||||
| * | Adjust to new style "generated files in CVS" mechanism for lex output | Chris Lattner | 2006-02-14 | 2 | -0/+201 | |
| | | | | | llvm-svn: 26163 | |||||
| * | adjust to new style of handling lexer changes | Chris Lattner | 2006-02-14 | 2 | -111/+466 | |
| | | | | | llvm-svn: 26162 | |||||
| * | Adjust to new form of handling lexer dependencies, this way shouldn't have | Chris Lattner | 2006-02-14 | 3 | -40/+273 | |
| | | | | | | | the problems the old way did. llvm-svn: 26161 | |||||
| * | Implement an alternative way of handling generated lex files in CVS. This | Chris Lattner | 2006-02-14 | 1 | -5/+15 | |
| | | | | | | | | should solve the "updating cvs when .l files change give me conflict markers that break my build" issue. llvm-svn: 26160 | |||||
| * | Wrap a couple more long lines | Chris Lattner | 2006-02-14 | 1 | -4/+5 | |
| | | | | | llvm-svn: 26159 | |||||
| * | wrap long lines | Chris Lattner | 2006-02-14 | 1 | -9/+16 | |
| | | | | | llvm-svn: 26158 | |||||
| * | Use statistics to keep track of what flavors of loops we are unswitching | Chris Lattner | 2006-02-14 | 1 | -7/+19 | |
| | | | | | llvm-svn: 26157 | |||||
| * | Fix a latent bug in the call sequence handling stuff. Some targets (e.g. x86) | Chris Lattner | 2006-02-14 | 1 | -4/+11 | |
| | | | | | | | create these nodes with flag results. Remember that we legalized them. llvm-svn: 26156 | |||||
| * | Implement Instcombine/and.ll:test34 | Chris Lattner | 2006-02-13 | 1 | -0/+12 | |
| | | | | | llvm-svn: 26155 | |||||
| * | new testcase | Chris Lattner | 2006-02-13 | 1 | -0/+6 | |
| | | | | | llvm-svn: 26154 | |||||
| * | Fix typo that caused build failures for things trying to use m_Or. | Chris Lattner | 2006-02-13 | 1 | -2/+2 | |
| | | | | | llvm-svn: 26153 | |||||
| * | If any of the sign extended bits are demanded, the input sign bit is demanded | Chris Lattner | 2006-02-13 | 1 | -3/+9 | |
| | | | | | | | | | for a sign extension. This fixes InstCombine/2006-02-13-DemandedMiscompile.ll and Ptrdist/bc. llvm-svn: 26152 | |||||
| * | testcase reduced from Ptrdist/bc that is miscompiled by recent instcombine ↵ | Chris Lattner | 2006-02-13 | 1 | -0/+9 | |
| | | | | | | | changes. llvm-svn: 26151 | |||||
| * | Another hack due to allowing multiple symbols with the same name. | Chris Lattner | 2006-02-13 | 1 | -11/+52 | |
| | | | | | llvm-svn: 26150 | |||||
| * | Another work around for the 'symbols with different types can have the same | Chris Lattner | 2006-02-13 | 1 | -0/+4 | |
| | | | | | | | name' issue. llvm-svn: 26149 | |||||
| * | improved zap discovery | Andrew Lenharth | 2006-02-13 | 1 | -0/+33 | |
| | | | | | llvm-svn: 26148 | |||||
| * | test for improved zap discovery | Andrew Lenharth | 2006-02-13 | 1 | -0/+10 | |
| | | | | | llvm-svn: 26147 | |||||
| * | Sync up the tag numbers with gcc4. | Jim Laskey | 2006-02-13 | 1 | -9/+1 | |
| | | | | | llvm-svn: 26146 | |||||
| * | Rename to better reflect usage (current and planned.) | Jim Laskey | 2006-02-13 | 3 | -5/+7 | |
| | | | | | llvm-svn: 26145 | |||||
| * | Completely rewrite libcall insertion by the legalizer, providing the | Chris Lattner | 2006-02-13 | 1 | -198/+217 | |
| | | | | | | | | | | | | | | | | | following handy-dandy properties: 1. it is always correct now 2. it is much faster than before 3. it is easier to understand This implementation builds off of the recent simplifications of the legalizer that made it single-pass instead of iterative. This fixes JM/lencod, JM/ldecod, and CodeGen/Generic/2006-02-12-InsertLibcall.ll (at least on PPC). llvm-svn: 26144 | |||||
| * | Switch targets over to using SelectionDAG::getCALLSEQ_START to create | Chris Lattner | 2006-02-13 | 4 | -12/+8 | |
| | | | | | | | CALLSEQ_START nodes. llvm-svn: 26143 | |||||
| * | Switch to using getCALLSEQ_START instead of using our own creation calls | Chris Lattner | 2006-02-13 | 1 | -4/+4 | |
| | | | | | llvm-svn: 26142 | |||||
| * | Add a method | Chris Lattner | 2006-02-13 | 1 | -0/+12 | |
| | | | | | llvm-svn: 26141 | |||||
| * | this passes now, due to Nate's recent efforts | Chris Lattner | 2006-02-13 | 1 | -1/+0 | |
| | | | | | llvm-svn: 26140 | |||||
| * | Reduce this testcase a bit more, with the help of llvm-extract and some hand ↵ | Chris Lattner | 2006-02-13 | 1 | -12/+8 | |
| | | | | | | | tweaks llvm-svn: 26139 | |||||
| * | Be careful not to request or look at bits shifted in from outside the size | Chris Lattner | 2006-02-13 | 1 | -3/+9 | |
| | | | | | | | of the input. This fixes the mediabench/gsm/toast failure last night. llvm-svn: 26138 | |||||
| * | Added a test case for a libcall insertion bug. | Evan Cheng | 2006-02-12 | 1 | -0/+64 | |
| | | | | | llvm-svn: 26137 | |||||
| * | Add missing patterns for andi. and andis., fixing test/Regression/CodeGen/ | Nate Begeman | 2006-02-12 | 1 | -2/+4 | |
| | | | | | | | PowerPC/and-imm.ll llvm-svn: 26136 | |||||
| * | remove some more dead special case code | Chris Lattner | 2006-02-12 | 1 | -35/+8 | |
| | | | | | llvm-svn: 26135 | |||||
| * | Eliminate special case hacks that are superceded by general purpose hacks | Chris Lattner | 2006-02-12 | 1 | -139/+51 | |
| | | | | | llvm-svn: 26134 | |||||
| * | tweaks | Chris Lattner | 2006-02-12 | 2 | -1/+10 | |
| | | | | | llvm-svn: 26133 | |||||
| * | Three changes: | Chris Lattner | 2006-02-12 | 1 | -6/+135 | |
| | | | | | | | | | | | | | 1. Teach GetConstantInType to handle boolean constants. 2. Teach instcombine to fold (compare X, CST) when X has known 0/1 bits. Testcase here: set.ll:test22 3. Improve the "(X >> c1) & C2 == 0" folding code to allow a noop cast between the shift and and. More aggressive bitfolding for other reasons was turning signed shr's into unsigned shr's, leaving the noop cast in the way. llvm-svn: 26131 | |||||
| * | new testcase | Chris Lattner | 2006-02-12 | 1 | -1/+9 | |
| | | | | | llvm-svn: 26130 | |||||
| * | move a failing testcase from bit-tracking.ll to narrow.ll, and move the | Chris Lattner | 2006-02-12 | 2 | -22/+18 | |
| | | | | | | | xfail marker with it llvm-svn: 26129 | |||||
| * | Revert my last patch. It too breaks stuff | Chris Lattner | 2006-02-12 | 1 | -12/+6 | |
| | | | | | llvm-svn: 26128 | |||||
| * | Make these tests fail if opt crashes. | Chris Lattner | 2006-02-12 | 2 | -0/+2 | |
| | | | | | llvm-svn: 26127 | |||||
| * | Fix for my previously reverted patch | Chris Lattner | 2006-02-11 | 1 | -6/+12 | |
| | | | | | llvm-svn: 26126 | |||||
| * | Update comments to be actually accurate | Chris Lattner | 2006-02-11 | 1 | -13/+19 | |
| | | | | | llvm-svn: 26124 | |||||
| * | This is implemented by the simplify-libcalls pass, not instcombine | Chris Lattner | 2006-02-11 | 1 | -23/+0 | |
| | | | | | llvm-svn: 26123 | |||||
| * | Port the recent innovations in ComputeMaskedBits to SimplifyDemandedBits. | Chris Lattner | 2006-02-11 | 1 | -211/+425 | |
| | | | | | | | | | | | | This allows us to simplify on conditions where bits are not known, but they are not demanded either! This also fixes a couple of bugs in ComputeMaskedBits that were exposed during this work. In the future, swaths of instcombine should be removed, as this code subsumes a bunch of ad-hockery. llvm-svn: 26122 | |||||
| * | revert my previous change, it exposed other problems. | Chris Lattner | 2006-02-11 | 1 | -1/+1 | |
| | | | | | llvm-svn: 26121 | |||||
| * | fix storing booleans (grawp missed this one) | Duraid Madina | 2006-02-11 | 1 | -3/+3 | |
| | | | | | llvm-svn: 26120 | |||||
| * | now short immediates will get matched (previously constants were all | Duraid Madina | 2006-02-11 | 1 | -5/+3 | |
| | | | | | | | triggering movl 64bit imm fat instructions) llvm-svn: 26119 | |||||
| * | Make this check stricter. Disallow loop exit blocks from being shared by | Chris Lattner | 2006-02-11 | 1 | -4/+7 | |
| | | | | | | | loops and their subloops. llvm-svn: 26118 | |||||
| * | Prevent certain nodes that have already been selected from being folded into | Evan Cheng | 2006-02-11 | 1 | -19/+40 | |
| | | | | | | | | | X86 addressing mode. Currently we do not allow any node whose target node produces a chain as well as any node that is at the root of the addressing mode expression tree. llvm-svn: 26117 | |||||
| * | remove dead expr | Chris Lattner | 2006-02-11 | 1 | -1/+0 | |
| | | | | | llvm-svn: 26116 | |||||
| * | Reorg for integration with gcc4. Old style debug info will not be passed though | Jim Laskey | 2006-02-11 | 3 | -193/+436 | |
| | | | | | | | to SelIDAG. llvm-svn: 26115 | |||||
| * | implement unswitching of loops with switch stmts and selects in them | Chris Lattner | 2006-02-11 | 1 | -94/+135 | |
| | | | | | llvm-svn: 26114 | |||||

