| Commit message (Collapse) | Author | Age | Files | Lines | ||
|---|---|---|---|---|---|---|
| ... | ||||||
| * | Add llvm::sys::getHostTriple and remove | Daniel Dunbar | 2009-03-31 | 3 | -32/+36 | |
| | | | | | | | | | | | | llvm::sys::getOS{Name,Version}. Right now the implementation just derives from LLVM_HOSTTRIPLE (which is wrong, but it doesn't look like we have a define for the target triple). Ideally this routine would actually be able to compute the triple for targets we care about. llvm-svn: 68118 | |||||
| * | Minor top-level comment fix. | Dan Gohman | 2009-03-31 | 1 | -1/+1 | |
| | | | | | llvm-svn: 68113 | |||||
| * | Tidy up some comments. | Dan Gohman | 2009-03-31 | 1 | -3/+3 | |
| | | | | | llvm-svn: 68112 | |||||
| * | Add a comment. | Dan Gohman | 2009-03-31 | 1 | -1/+2 | |
| | | | | | llvm-svn: 68111 | |||||
| * | remove unused arguments. | Rafael Espindola | 2009-03-31 | 1 | -14/+12 | |
| | | | | | llvm-svn: 68109 | |||||
| * | Really temporarily revert r68073. | Bill Wendling | 2009-03-31 | 2 | -15/+1 | |
| | | | | | llvm-svn: 68100 | |||||
| * | Oy! When reverting r68073, I added in experimental code. Sorry... | Bill Wendling | 2009-03-31 | 6 | -96/+47 | |
| | | | | | llvm-svn: 68099 | |||||
| * | Remove the "fast" cases for spill and restore point determination, as these ↵ | Owen Anderson | 2009-03-31 | 2 | -127/+60 | |
| | | | | | | | | | were subtlely wrong in obscure cases. Patch the testcase to account for this change. llvm-svn: 68093 | |||||
| * | Revert r68073. It's causing a failure in the Apple-style builds. | Bill Wendling | 2009-03-31 | 6 | -47/+96 | |
| | | | | | llvm-svn: 68092 | |||||
| * | shrink subclassid, liberating some bits for future (ab)use. | Chris Lattner | 2009-03-31 | 1 | -1/+1 | |
| | | | | | llvm-svn: 68087 | |||||
| * | Fix live-out reg logic to not insert over-aggressive AssertZExt | Dan Gohman | 2009-03-31 | 2 | -3/+23 | |
| | | | | | | | instructions. This fixes lua. llvm-svn: 68083 | |||||
| * | X86 address mode isel tweak. If the base of the address is also used by a ↵ | Evan Cheng | 2009-03-31 | 1 | -1/+25 | |
| | | | | | | | | | CopyToReg (i.e. it's likely live-out), do not fold the sub-expressions into the addressing mode to avoid computing the address twice. The CopyToReg use will be isel'ed to a LEA, re-use it for address instead. This is not yet enabled. llvm-svn: 68082 | |||||
| * | Really, really fix PointerUnion3::is | Douglas Gregor | 2009-03-31 | 1 | -1/+1 | |
| | | | | | llvm-svn: 68079 | |||||
| * | Except in asm-verbose mode, avoid printing labels for blocks that are | Dan Gohman | 2009-03-30 | 2 | -1/+15 | |
| | | | | | | | | | only reachable via fall-through edges. This dramatically reduces the number of labels printed, and thus also the number of labels the assembler must parse and remember. llvm-svn: 68073 | |||||
| * | Loop Index Split can eliminate a loop if it can determin if loop body is ↵ | Devang Patel | 2009-03-30 | 2 | -8/+48 | |
| | | | | | | | executed only once. There was a bug in determining IV based value of the iteration for which the loop body is executed. Fix it. llvm-svn: 68071 | |||||
| * | Make PointerUnion3::get work properly | Douglas Gregor | 2009-03-30 | 1 | -3/+5 | |
| | | | | | llvm-svn: 68067 | |||||
| * | When optimzing a mul by immediate into two, the resulting mul's should get a ↵ | Evan Cheng | 2009-03-30 | 5 | -5/+12 | |
| | | | | | | | x86 specific node to avoid dag combiner from hacking on them further. llvm-svn: 68066 | |||||
| * | Turn a 2-address instruction into a 3-address one when it's profitable even ↵ | Evan Cheng | 2009-03-30 | 2 | -22/+69 | |
| | | | | | | | | | | | | | | if the two-address operand is killed. e.g. %reg1024<def> = MOV r1 %reg1025<def> = ADD %reg1024, %reg1026 r0 = MOV %reg1025 If it's not possible / profitable to commute ADD, then turning ADD into a LEA saves a copy. llvm-svn: 68065 | |||||
| * | update comment. | Chris Lattner | 2009-03-30 | 1 | -1/+1 | |
| | | | | | llvm-svn: 68060 | |||||
| * | Balance out quote in debug output. | Bill Wendling | 2009-03-30 | 1 | -1/+1 | |
| | | | | | llvm-svn: 68059 | |||||
| * | Fix grammar-o in comment. | Bill Wendling | 2009-03-30 | 1 | -1/+1 | |
| | | | | | llvm-svn: 68057 | |||||
| * | add a PointerUnion3 class and generalize PointerUnion to work with | Chris Lattner | 2009-03-30 | 1 | -15/+120 | |
| | | | | | | | anything pointer-like, which may or may not actually be a pointer. llvm-svn: 68056 | |||||
| * | fix the PointerLikeTypeTraits specialization for PointerIntPair to | Chris Lattner | 2009-03-30 | 1 | -2/+4 | |
| | | | | | | | allow the traits to be specified as well. llvm-svn: 68055 | |||||
| * | Constify arguments in isSuccessor and isLayoutSuccessor. | Dan Gohman | 2009-03-30 | 2 | -4/+4 | |
| | | | | | llvm-svn: 68054 | |||||
| * | Update the polygen grammer to reflect that zext and sext are no longer | Dan Gohman | 2009-03-30 | 1 | -2/+0 | |
| | | | | | | | valid argument attributes (zeroext and signext are). llvm-svn: 68053 | |||||
| * | docs/TestingGuide.html: correction to prev. text (objdir!=srcdir required ↵ | John Mosby | 2009-03-30 | 1 | -7/+3 | |
| | | | | | | | for running test-suite), removed refs to llvm-test llvm-svn: 68051 | |||||
| * | Fix comment to match function name. | Bob Wilson | 2009-03-30 | 1 | -1/+1 | |
| | | | | | llvm-svn: 68050 | |||||
| * | getEntryFor() may invalidate DenseMap iterator. | Devang Patel | 2009-03-30 | 1 | -0/+1 | |
| | | | | | | | Walking an invalidated iterator is not a good idea. llvm-svn: 68047 | |||||
| * | Add ccc back for now. | Mike Stump | 2009-03-30 | 1 | -1/+1 | |
| | | | | | llvm-svn: 68038 | |||||
| * | Clearify local/global relocations wording | Anton Korobeynikov | 2009-03-30 | 1 | -1/+10 | |
| | | | | | llvm-svn: 68037 | |||||
| * | Fix thinko: put stuff with both global and local relocations into ↵ | Anton Korobeynikov | 2009-03-30 | 1 | -3/+3 | |
| | | | | | | | data.rel{.ro}, not .local llvm-svn: 68036 | |||||
| * | Tweak test for recent relro stuff | Anton Korobeynikov | 2009-03-30 | 1 | -9/+17 | |
| | | | | | llvm-svn: 68035 | |||||
| * | Fix infinite looping | Anton Korobeynikov | 2009-03-30 | 1 | -0/+2 | |
| | | | | | llvm-svn: 68034 | |||||
| * | Properly propagate Kind. | Anton Korobeynikov | 2009-03-30 | 1 | -1/+1 | |
| | | | | | llvm-svn: 68033 | |||||
| * | Do not propagate ELF-specific stuff (data.rel) into other targets. This ↵ | Anton Korobeynikov | 2009-03-30 | 5 | -25/+37 | |
| | | | | | | | simplifies code and also ensures correctness. llvm-svn: 68032 | |||||
| * | Add data.rel stuff | Anton Korobeynikov | 2009-03-30 | 6 | -14/+63 | |
| | | | | | llvm-svn: 68031 | |||||
| * | fix some validation problems. | Chris Lattner | 2009-03-30 | 1 | -7/+2 | |
| | | | | | llvm-svn: 68026 | |||||
| * | Forgot this test. | Evan Cheng | 2009-03-30 | 1 | -0/+12 | |
| | | | | | llvm-svn: 68025 | |||||
| * | Clarify section on setting up and running test-suite | John Mosby | 2009-03-30 | 1 | -14/+27 | |
| | | | | | llvm-svn: 68023 | |||||
| * | Updated the comment for isArithmeticShift() to match reality. | Misha Brukman | 2009-03-29 | 1 | -2/+1 | |
| | | | | | llvm-svn: 68016 | |||||
| * | Constify check. This fixes PR3900. | Bill Wendling | 2009-03-29 | 1 | -1/+1 | |
| | | | | | llvm-svn: 68013 | |||||
| * | Testcase for recent ro/relocs stuff | Anton Korobeynikov | 2009-03-29 | 1 | -0/+15 | |
| | | | | | llvm-svn: 68008 | |||||
| * | IA64 is as weird as Alpha wrt r/o relocs :) | Anton Korobeynikov | 2009-03-29 | 2 | -0/+8 | |
| | | | | | llvm-svn: 68007 | |||||
| * | Alpha always requires global relocations to be r/w regardless of PIC. | Anton Korobeynikov | 2009-03-29 | 2 | -1/+8 | |
| | | | | | llvm-svn: 68006 | |||||
| * | Honour relocation behaviour stuff for ro objects | Anton Korobeynikov | 2009-03-29 | 3 | -3/+29 | |
| | | | | | llvm-svn: 68005 | |||||
| * | Extend the relocation tracker handler, so we can filter on different 'kinds' ↵ | Anton Korobeynikov | 2009-03-29 | 2 | -7/+28 | |
| | | | | | | | of relocations required. llvm-svn: 68004 | |||||
| * | Fix PR3899: add support for extracting floats from vectors | Duncan Sands | 2009-03-29 | 4 | -0/+32 | |
| | | | | | | | | when using -soft-float. Based on a patch by Jakob Stoklund Olesen. llvm-svn: 67996 | |||||
| * | add missing space. | Chris Lattner | 2009-03-29 | 1 | -2/+2 | |
| | | | | | llvm-svn: 67995 | |||||
| * | add some comments, add a dyn_cast method. | Chris Lattner | 2009-03-29 | 1 | -0/+16 | |
| | | | | | llvm-svn: 67992 | |||||
| * | When forming sentinels for empty/tombstone, make sure to respect the | Chris Lattner | 2009-03-29 | 2 | -6/+17 | |
| | | | | | | | | | pointer's expected number of zero low-bits. This should fix the breakage I introduced recently. llvm-svn: 67990 | |||||

