aboutsummaryrefslogtreecommitdiff
path: root/lib/Transforms/Scalar/InstructionCombining.cpp
AgeCommit message (Collapse)Author
2007-10-17Reverting r43070 for now. It's causing llc test failures.Evan Cheng
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43103 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-17Apply "Instead of loading small c string constant, use integer constant ↵Devang Patel
directly" transformation while processing load instruction. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43070 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-16Use immediate stores.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@43055 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-15Achieve same result but use fewer lines of code.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42985 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-12Dest type is always i8 *. This allows some simplification.Devang Patel
Do not filter memmove. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42930 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-12Fix a bug in my patch last night that broke InstCombine/2007-10-12-Crash.llChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42920 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-12eliminate warningGabor Greif
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42892 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-12Fix some 80 column violations.Chris Lattner
Fix DecomposeSimpleLinearExpr to handle simple constants better. Don't nuke gep(bitcast(allocation)) if the bitcast(allocation) will fold the allocation. This fixes PR1728 and Instcombine/malloc3.ll git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42891 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-11Lower memcpy if it makes sense.Devang Patel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42864 91177308-0d34-0410-b5e6-96231b3b80d8
2007-10-03Tone down an overzealous optimization.Dale Johannesen
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42582 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-19Improve comment.Duncan Sands
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42132 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-19A global variable with external weak linkage can be null, whileDuncan Sands
an alias could alias such a global variable. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42130 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-17Instcombine x-((x/y)*y) into a remainder operator.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42035 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-17Factor the trampoline transformation into a subroutine.Duncan Sands
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@42021 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-14Remove the assumption that FP's are either float orDale Johannesen
double from some of the many places in the optimizers it appears, and do something reasonable with x86 long double. Make APInt::dump() public, remove newline, use it to dump ConstantSDNode's. Allow APFloats in FoldingSet. Expand X86 backend handling of long doubles (conversions to/from int, mostly). git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41967 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-14silence a bogus gcc warning.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41949 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-11Turn calls to trampolines into calls to the underlyingDuncan Sands
nested function. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41844 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-10remove some dead code, this is handled by constant folding.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41819 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-07Don't zap back to back volatile load/storesChris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41759 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-06Next round of APFloat changes.Dale Johannesen
Use APFloat in UpgradeParser and AsmParser. Change all references to ConstantFP to use the APFloat interface rather than double. Remove the ConstantFP double interfaces. Use APFloat functions for constant folding arithmetic and comparisons. (There are still way too many places APFloat is just a wrapper around host float/double, but we're getting there.) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41747 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-06Use isTrueWhenEqual. Thanks Chris!Nick Lewycky
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41741 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-06When the two operands of an icmp are equal, there are five possible predicatesNick Lewycky
that would make the icmp true. Fixes PR1637. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41740 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-05Forgot to obey 80 column rule. Fixing that.Chuck Rose III
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41725 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-05Added default parameters to GetElementPtrInstr constructor call. Visual ↵Chuck Rose III
Studio 2k5 was getting confused and was unable to compile it. Suspected compiler error. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41721 91177308-0d34-0410-b5e6-96231b3b80d8
2007-09-04Update GEP constructors to use an iterator interface to fixDavid Greene
GLIBCXX_DEBUG issues. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41697 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-28Cut off crazy computation. This helps PR1622 slightly.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41522 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-27Update InvokeInst to work like CallInstDavid Greene
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41506 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-11Transform a load from an undef/zero global into an undef/global even if weChris Lattner
have complex pointer manipulation going on. This allows us to compile stuff like this: __m128i foo(__m128i x){ static const unsigned int c_0[4] = { 0, 0, 0, 0 }; __m128i v_Zero = _mm_loadu_si128((__m128i*)c_0); x = _mm_unpacklo_epi8(x, v_Zero); return x; } into: _foo: xorps %xmm1, %xmm1 punpcklbw %xmm1, %xmm0 ret git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41022 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-09when we see a unaligned load from an insufficiently aligned global orChris Lattner
alloca, increase the alignment of the load, turning it into an aligned load. This allows us to compile: #include <xmmintrin.h> __m128i foo(__m128i x){ static const unsigned int c_0[4] = { 0, 0, 0, 0 }; __m128i v_Zero = _mm_loadu_si128((__m128i*)c_0); x = _mm_unpacklo_epi8(x, v_Zero); return x; } into: _foo: punpcklbw _c_0.5944, %xmm0 ret .data .lcomm _c_0.5944,16,4 # c_0.5944 instead of: _foo: movdqu _c_0.5944, %xmm1 punpcklbw %xmm1, %xmm0 ret .data .lcomm _c_0.5944,16,2 # c_0.5944 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40971 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-06It's safe to fold not of fcmp.Nick Lewycky
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40870 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-05at the end of instcombine, explicitly clear WorklistMap.Chris Lattner
This shrinks it down to something small. On the testcase from PR1432, this speeds up instcombine from 0.7959s to 0.5000s, (59%) git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40840 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-04This is the patch to provide clean intrinsic function overloading support in ↵Chandler Carruth
LLVM. It cleans up the intrinsic definitions and generally smooths the process for more complicated intrinsic writing. It will be used by the upcoming atomic intrinsics as well as vector and float intrinsics in the future. This also changes the syntax for llvm.bswap, llvm.part.set, llvm.part.select, and llvm.ct* intrinsics. They are automatically upgraded by both the LLVM ASM reader and the bitcode reader. The test cases have been updated, with special tests added to ensure the automatic upgrading is supported. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40807 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-02Replacing a cast with another one does not reduce the number of Chris Lattner
casts in the input. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40741 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-02Disable an xform that causes an infinite loop. This fixes PR1594Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40739 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-02wrap some long lines. Major offenders that are left includeChris Lattner
gvn, gvnpre, dse, and predsimplify. To see these, use: make check-line-length git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40738 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-02Enhance instcombine to be more aggressive about folding casts ofChris Lattner
operations of casts. This implements InstCombine/zext-fold.ll git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40726 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-01New CallInst interface to address GLIBCXX_DEBUG errors caused byDavid Greene
indexing an empty std::vector. Updates to all clients. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40660 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-31Fix a bug in GetKnownAlignment of packed structs.Lauro Ramos Venancio
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40649 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-30Fix a typo/thinko.Reid Spencer
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40599 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-23completely remove a transformation that is unsafe in the face ofChris Lattner
undefs. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40439 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-21Apply temporary work around to fix llvm mis-compilationDevang Patel
reported in PR 1556. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40133 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-20this xform is already done by the constant folder.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40124 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-20Optimize alignment of loads and stores.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@40102 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-16Fix comments about vectors to use the current wording.Dan Gohman
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@39921 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-16Repair a regression in Transforms/InstCombine/mul.ll that Reid noticed.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@39896 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-15Implement shift-simplify.ll:test[45].Chris Lattner
First teach instcombine that sign bit checks only demand the sign bit, this allows simplify demanded bits to hack on expressions better. Second, teach instcombine that ashr is useless if only the sign bit is demanded. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@39880 91177308-0d34-0410-b5e6-96231b3b80d8
2007-07-15Implement shift-simplify.ll:test3, turning:Chris Lattner
(X << 31) <s 0 --> (X&1) != 0 This happens dozens of times in the CFE. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@39879 91177308-0d34-0410-b5e6-96231b3b80d8
2007-06-21Significantly improve the documentation of the instcombine divide/compareChris Lattner
transformation. Also, keep track of which end of the integer interval overflows occur on. This fixes Transforms/InstCombine/2007-06-21-DivCompareMiscomp.ll and rdar://5278853, a miscompilation of perl. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@37692 91177308-0d34-0410-b5e6-96231b3b80d8
2007-06-20refactor a bunch of code out of visitICmpInstWithInstAndIntCst into its ownChris Lattner
routine. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@37679 91177308-0d34-0410-b5e6-96231b3b80d8
2007-06-19silence a bogus warning Duraid ran into.Chris Lattner
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@37649 91177308-0d34-0410-b5e6-96231b3b80d8