- e0cf075 Get the asminfo for the target most closely matching the module and use that for inline asm by Andrew Lenharth · 18 years ago
- 6c0695f X86 asm -> gcc asm translation table (incomplete) by Andrew Lenharth · 18 years ago
- 3655de6 Add per-target support for asm translation in the cbe by Andrew Lenharth · 18 years ago
- 7fc4a94 Remove debug code. by Jim Laskey · 18 years ago
- 2ada085 Prime text sections to improve branch locality in large object files. by Jim Laskey · 18 years ago
- e9bd7b2 32-bit int space was not accounted for properly in lowerCall. by Jim Laskey · 18 years ago
- 861d9d6 Regenerate. by Reid Spencer · 18 years ago
- b78b908 Implement signedness caching for values, value lists, constants and by Reid Spencer · 18 years ago
- 5d4dbd1 Remove 4 FIXME's from the CAST patch now that the back end is correctly by Reid Spencer · 18 years ago
- 5e1452c Add an -append-exit-code option to bugpoint. This will cause bugpoint to by Reid Spencer · 18 years ago
- 2c2148b Fix a typo. by Reid Spencer · 18 years ago
- 38b0e7b Put the #include for a module first. by Bill Wendling · 18 years ago
- 8bcb042 Clean up. by Evan Cheng · 18 years ago
- afed814 Fix indentation. by Evan Cheng · 18 years ago
- 8f48766 Changed to using LLVM streams. by Bill Wendling · 18 years ago
- d9fd2ac Changed to using llvm streams. by Bill Wendling · 18 years ago
- eba4be8 Added helper function to transition to using llvm streams. by Bill Wendling · 18 years ago
- 7bb64e8 Fix JIT encoding bugs for shift / rotate by one ops. by Evan Cheng · 18 years ago
- 7b09487 testcase for pr1016 by Chris Lattner · 18 years ago
- b6c8060 Fix PR1016 by Chris Lattner · 18 years ago
- 480f093 Removed #include <iostream> and replaced streams with llvm streams. by Bill Wendling · 18 years ago
- 29c277f Preliminary support for inline asm in the cbe. The target specific ugliness by Andrew Lenharth · 18 years ago
- c0f64ff Change MachineInstr ctor's to take a TargetInstrDescriptor reference instead by Evan Cheng · 18 years ago
- 722a0ca remove dead tag by Chris Lattner · 18 years ago
- 3687c1a Fix the dag combiner bug corresponding to PR1014. by Chris Lattner · 18 years ago
- 779d212 this doesn't pass with CVS HEAD. by Chris Lattner · 18 years ago
- f2f1643 Fix PR1014 and InstCombine/2006-11-27-XorBug.ll. by Chris Lattner · 18 years ago
- e434f4a Add 3 more test cases for ptrtoint, fptoui and fptosi by Reid Spencer · 18 years ago
- f357daf testcase for PR1014 by Chris Lattner · 18 years ago
- 639cf7d When truncating to bool, it is necessary to & with 1 for all casts that by Reid Spencer · 18 years ago
- eb73138 Protect against null streams. by Bill Wendling · 18 years ago
- 41edad7 Removed #include <iostream> and replaced with llvm_* streams. by Bill Wendling · 18 years ago
- e16746d Alter these tests to ensure they match a "test $1, X" X86 instruction that by Reid Spencer · 18 years ago
- df2a4ff Add a test to ensure a bitcast/and/trunc combination eliminates the bitcast. by Reid Spencer · 18 years ago
- bc4cf8d For better or worse, load from i1 is assumed to be zero extended. Do not by Chris Lattner · 18 years ago
- f990817 If a brcond condition is promoted, make sure to zero extend it, even if not by Chris Lattner · 18 years ago
- 3da59db For PR950: by Reid Spencer · 18 years ago
- 5fed9b9 Don't use bitcast yet! by Reid Spencer · 18 years ago
- 0270546 Make this test case easier to read. No functional changes. by Reid Spencer · 18 years ago
- fe6b146 Removed #include <iostream> and replace with llvm_* streams. by Bill Wendling · 18 years ago
- 7fe9518 Add method that will take a function pointer so that it can handle things by Bill Wendling · 18 years ago
- 0d45a09 Remove #include <iostream> and use llvm_* streams instead. by Bill Wendling · 18 years ago
- 0a81aac Replace #include <iostream> with llvm_* streams. by Bill Wendling · 18 years ago
- b742703 Removed #include <iostream> and replaced with llvm_* streams. by Bill Wendling · 18 years ago
- 62c804a Removed #include <iostream> and used the llvm_cerr/DOUT streams instead. by Bill Wendling · 18 years ago
- f525eb9 fix storing bools to mem and unordered FP ops by Duraid Madina · 18 years ago
- 9df1246 Make the absolute/relative tolerance information easier to read/understand. by Reid Spencer · 18 years ago
- 24652d1 Add newline at end of file. by Reid Spencer · 18 years ago
- ba07a69 Only show the first few lines of difference. by Reid Spencer · 18 years ago
- 2bae1f5 Add an option to enable lli (interpreter mode) testing. by Reid Spencer · 18 years ago
- 0b44d6c Fix some typos. by Reid Spencer · 18 years ago
- e5d4efa Promote GEP ubyte indices to uint. Backwards compatibility for 1.2 and by Reid Spencer · 18 years ago
- 565706b Update to new predicate simplifier VRP design. Fixes PR966 and PR967. by Nick Lewycky · 18 years ago
- 6d0fbd4 Stop early if there is no mis-optimization. by Reid Spencer · 18 years ago
- b52818e Documented the --disable-compression option in gccld and gccas. by John Criswell · 18 years ago
- d452599 Added the --disable-compression option which controls whether the by John Criswell · 18 years ago
- 13c654a This xform is handled by FoldOpIntoPhi in visitCastInst in a more elegant way. by Chris Lattner · 18 years ago
- 12c49af Refactored *GVRequiresExtraLoad() to Subtarget method. by Anton Korobeynikov · 18 years ago
- e04bf54 Test for PR1013. by Jim Laskey · 18 years ago
- bdde0d9 new testcase by Chris Lattner · 18 years ago
- e9c9f98 in ppc64-mode, don't allocate the 32-bit version of r13 either. by Chris Lattner · 18 years ago
- 0374248 Global label not handled correctly. by Jim Laskey · 18 years ago
- bdc571b r13 is the thread pointer on darwin/ppc64, don't allocate it. by Chris Lattner · 18 years ago
- 1dcf5e8 Fix PR1005 by Chris Lattner · 18 years ago
- f7560ed Fix PR1001, patch by Nikhil Patil! by Chris Lattner · 18 years ago
- 3b6f497 Fix codegen for x86-64 on systems (like ppc or i386) that don't have 64-bit by Chris Lattner · 18 years ago
- 32ba1aa Fix PR1011 and CodeGen/Generic/2006-11-20-DAGCombineCrash.ll by Chris Lattner · 18 years ago
- f2d2a5a Testcase for PR1011 by Chris Lattner · 18 years ago
- a138c6c setOperand should not zap the operand list or add implicit operands to an by Chris Lattner · 18 years ago
- aa9c549 Testcase for pr1012 by Chris Lattner · 18 years ago
- a598685 We should perform extra load for dllimported calls by Anton Korobeynikov · 18 years ago
- 134693a Fixed typo by Tanya Lattner · 18 years ago
- f7b6be0 Adding 1.9 release and removing references to cfrontend by Tanya Lattner · 18 years ago
- c9aee64 Changing cfrontend dir to llvm-gcc3.4 as the example. by Tanya Lattner · 18 years ago
- 45fb3f3 For PR950: by Reid Spencer · 18 years ago
- 36699ca For PR950: by Reid Spencer · 18 years ago
- b7f2628 Corrections from review. by Reid Spencer · 18 years ago
- 4b10591 Don't transform in another bad case: if the block is empty, it should be by Chris Lattner · 18 years ago
- f10a56a Fix another case we *don't* want to do this xform. by Chris Lattner · 18 years ago
- 36d68e4 Update a reference to the old setcc instruction. by Reid Spencer · 18 years ago
- f3a70a6 For PR950: by Reid Spencer · 18 years ago
- 9b9cc5a MachineInstr includes this by Chris Lattner · 18 years ago
- 1ccc468 Implement operator<< for machine basic blocks to make it easier to dump them. by Chris Lattner · 18 years ago
- 154e104 make the previous change more aggressive, moving any block with no fallthrough. by Chris Lattner · 18 years ago
- a7bef4a Minor code layout tweak: if we have something like this: by Chris Lattner · 18 years ago
- 822a879 Do not convert massive blocks on phi nodes into select statements. Instead by Chris Lattner · 18 years ago
- e6cc288 converting massive blocks of phis into selects like this is silly. by Chris Lattner · 18 years ago
- 161a2fb Make the awk filter print out which files its eliminating. by Reid Spencer · 18 years ago
- 35d958c This is a utility for preparing patches against LLVM. It runs cvs diff with by Reid Spencer · 18 years ago
- 5b31cf0 Provide a list of gccld optimization switches. by Reid Spencer · 18 years ago
- 52d0d66 Fix bugs in this. What was I thinking??? :) by Reid Spencer · 18 years ago
- 9413de3 Movind newly-intorudced dll* linkage stuff & windows CC to "supported" list by Anton Korobeynikov · 18 years ago
- c53e506 fix bug by Chris Lattner · 18 years ago
- 2629922 first draft of 1.9 release notes by Chris Lattner · 18 years ago
- 64fc341 Fail even if opt doesn't print anything. by Reid Spencer · 18 years ago
- 1faa549 Add a simple test to make sure getModRefInfo is 1/2 way sane. by Reid Spencer · 18 years ago
- 967413f Have ConstantExprs upgrade opcodes the same way as instructions. by Reid Spencer · 18 years ago
- b1eb987 on ppc64, float arguments take 8-byte stack slots not 4-byte stack slots. by Chris Lattner · 18 years ago
- ff79089 make sure to safe LR8 in the right stack slot for PPC64 by Chris Lattner · 18 years ago
- 566c1b1 Pretty print 'rldicr r2, r2, 2, 61' as 'sldi r2, r2, 2'. by Chris Lattner · 18 years ago