| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Allow the JIT ExecutionEngine to report details about the generated machine ↵ | Argyrios Kyrtzidis | 2009-05-18 | 3 | -6/+38 |
| | | | | | | | | | | | code. Introduce a new class (MachineCodeInfo) that the JIT can fill in with details. Right now, just the address and the size of the machine code are reported. Patch by Evan Phoenix! llvm-svn: 72040 | ||||
| * | Fix CodePlacementOpt::OptimizeIntraLoopEdges so that its return value | Bob Wilson | 2009-05-18 | 1 | -0/+2 |
| | | | | | | | correctly indicates whether it changed the code. llvm-svn: 72038 | ||||
| * | Fix pr4202: Disable CodePlacementOpt for ARM. The ARMConstantIslandPass has | Bob Wilson | 2009-05-18 | 1 | -1/+4 |
| | | | | | | | | | | to run last because it needs to know the exact size and position of every basic block. Currently CodePlacementOpt is set up to run last. It might be worthwhile to investigate reordering these passes, but for now, let's just make it work. llvm-svn: 72037 | ||||
| * | New Spiller interface and trivial implementation. | Lang Hames | 2009-05-18 | 3 | -2/+262 |
| | | | | | llvm-svn: 72030 | ||||
| * | Revert r72025. It is possible for clients to convert between signed types | Dan Gohman | 2009-05-18 | 1 | -3/+0 |
| | | | | | | | | and pointer types safely if they only do so when the sizes are the same. llvm-gcc is such a client. llvm-svn: 72029 | ||||
| * | Revert last commit. It was wrong. | Bill Wendling | 2009-05-18 | 1 | -2/+1 |
| | | | | | llvm-svn: 72026 | ||||
| * | Add assertions to CastInst::getCastOpcode to catch attempted conversions | Dan Gohman | 2009-05-18 | 1 | -0/+3 |
| | | | | | | | | | between integers and pointers when the source type is marked signed, since inttoptr and ptrtoint always use zero-extension when the destination is larger than the source. llvm-svn: 72025 | ||||
| * | Don't call RegionInlinedFnEnd if our optimization level isn't -O0. | Bill Wendling | 2009-05-18 | 1 | -1/+2 |
| | | | | | llvm-svn: 72024 | ||||
| * | Fill in the missing patterns for ADC and SBB. | Dale Johannesen | 2009-05-18 | 1 | -14/+97 |
| | | | | | | | Some comment cleanup. llvm-svn: 72022 | ||||
| * | termios.h contains the winsize structure we need to determine the | Douglas Gregor | 2009-05-18 | 1 | -1/+4 |
| | | | | | | | | width of a terminal. Don't try to get the width of a terminal if we don't have this header. llvm-svn: 72018 | ||||
| * | Rename UseTy to AccessTy, for consistency with getAccessType, and to | Dan Gohman | 2009-05-18 | 1 | -24/+25 |
| | | | | | | | avoid ambiguity with the word "use" in IVStrideUse. llvm-svn: 72012 | ||||
| * | Silence Release-Asserts warnings. | Daniel Dunbar | 2009-05-18 | 3 | -6/+11 |
| | | | | | llvm-svn: 72011 | ||||
| * | Teach ScalarEvolution to recognize x^-1 in the case where non-demanded | Dan Gohman | 2009-05-18 | 1 | -0/+13 |
| | | | | | | | bits have been stripped out by instcombine. llvm-svn: 72010 | ||||
| * | Delete a redundant 'else'. | Dan Gohman | 2009-05-18 | 1 | -1/+1 |
| | | | | | llvm-svn: 72009 | ||||
| * | Fix ScalarEvolution::isLoopGuardedByCond to accept a null Loop*, for | Dan Gohman | 2009-05-18 | 1 | -0/+4 |
| | | | | | | | | consistency with other routines that use a null Loop* to mean code not contained by any loop. llvm-svn: 72008 | ||||
| * | Minor code cleanups. Do more of the work before the if statements | Dan Gohman | 2009-05-18 | 1 | -16/+21 |
| | | | | | | | instead of within their controlling expressions. llvm-svn: 72007 | ||||
| * | Add assertion checks to the SCEV operator creation methods to catch | Dan Gohman | 2009-05-18 | 1 | -0/+34 |
| | | | | | | | type mismatches. llvm-svn: 72006 | ||||
| * | Make ScalarEvolution::isLoopGuardedByCond work even when the edge | Dan Gohman | 2009-05-18 | 1 | -13/+27 |
| | | | | | | | entering a loop is a non-split critical edge. llvm-svn: 72004 | ||||
| * | Add an isOne() utility function to ScalarEvolution, similar to isZero() | Dan Gohman | 2009-05-18 | 1 | -1/+6 |
| | | | | | | | and similar to ConstantInt's isOne(). llvm-svn: 72003 | ||||
| * | Tweak MemoryBuffer::getSTDIN so that it returns after the first EOF. | Eli Friedman | 2009-05-18 | 1 | -3/+6 |
| | | | | | | | | It doesn't matter for piped input, but it's annoying when typing at the console. llvm-svn: 71998 | ||||
| * | Prevented reg0 from being added to MBB live-in set, which was causing issues | Lang Hames | 2009-05-17 | 1 | -0/+5 |
| | | | | | | | for PostRAScheduler. llvm-svn: 71991 | ||||
| * | Rename MachineVerifier pass to avoid command line collision. | Jakob Stoklund Olesen | 2009-05-17 | 1 | -1/+1 |
| | | | | | llvm-svn: 71987 | ||||
| * | Mark rotl/rotr as expand. This generates pretty ugly code, but this is ↵ | Anton Korobeynikov | 2009-05-17 | 1 | -0/+4 |
| | | | | | | | better than nothing. llvm-svn: 71976 | ||||
| * | Typo | Anton Korobeynikov | 2009-05-17 | 2 | -3/+3 |
| | | | | | llvm-svn: 71975 | ||||
| * | Fix a missing def-flag on a Mips epilogue load. | Jakob Stoklund Olesen | 2009-05-16 | 1 | -4/+4 |
| | | | | | llvm-svn: 71935 | ||||
| * | Remember to set def-flag on register loaded from stack slot in CellSPU. | Jakob Stoklund Olesen | 2009-05-16 | 1 | -1/+1 |
| | | | | | llvm-svn: 71934 | ||||
| * | Verify that explicit definitions in the TargetInstrDesc are matched by | Jakob Stoklund Olesen | 2009-05-16 | 1 | -1/+12 |
| | | | | | | | explicit register define operands. llvm-svn: 71933 | ||||
| * | Allow redefinition of reserved registers. | Jakob Stoklund Olesen | 2009-05-16 | 1 | -2/+3 |
| | | | | | llvm-svn: 71932 | ||||
| * | Put back a bit of expensive checking logic that | Duncan Sands | 2009-05-16 | 1 | -1/+1 |
| | | | | | | | was overenthusiastically deleted in r70234. llvm-svn: 71926 | ||||
| * | Pacify gcc-4.3, which suggests explicit braces here | Duncan Sands | 2009-05-16 | 1 | -1/+2 |
| | | | | | | | to avoid an ambiguous else. llvm-svn: 71924 | ||||
| * | Pass to verify generated machine code. | Jakob Stoklund Olesen | 2009-05-16 | 3 | -27/+704 |
| | | | | | | | | | | | | | | | | | | | | The following is checked: * Operand counts: All explicit operands must be present. * Register classes: All physical and virtual register operands must be compatible with the register class required by the instruction descriptor. * Register live intervals: Registers must be defined only once, and must be defined before use. The machine code verifier is enabled with the command-line option '-verify-machineinstrs', or by defining the environment variable LLVM_VERIFY_MACHINEINSTRS to the name of a file that will receive all the verifier errors. llvm-svn: 71918 | ||||
| * | Split out the DwarfDebug module from the DwarfWriter module. | Bill Wendling | 2009-05-15 | 7 | -2804/+3188 |
| | | | | | | | Again, no intendtional functionality change. llvm-svn: 71854 | ||||
| * | Mark class as hidden. | Bill Wendling | 2009-05-15 | 1 | -1/+1 |
| | | | | | llvm-svn: 71850 | ||||
| * | Fix PR4210. Rewritter should track and update kills of sub-registers as well. | Evan Cheng | 2009-05-15 | 1 | -42/+63 |
| | | | | | llvm-svn: 71848 | ||||
| * | Move c'tor/d'tor to top of file. | Bill Wendling | 2009-05-15 | 1 | -17/+14 |
| | | | | | llvm-svn: 71828 | ||||
| * | Split out the DwarfException class into its own file. No functionality change, | Bill Wendling | 2009-05-15 | 3 | -769/+894 |
| | | | | | | | though the classes have been marked with "VISIBILITY_HIDDEN". llvm-svn: 71827 | ||||
| * | Split out the Dwarf writer stuff into separate files. This is a much more | Bill Wendling | 2009-05-15 | 7 | -1278/+1584 |
| | | | | | | | | | | logical/sane approach to organizing all of the stuff that goes into writing out DWARF information. Honestly? even this is too complex for what it's supposed to be doing. Trivia: It *looks* like there would be functionality changes, however there aren't! llvm-svn: 71821 | ||||
| * | Reflow to fit 80-col. | Mike Stump | 2009-05-14 | 1 | -2/+2 |
| | | | | | llvm-svn: 71813 | ||||
| * | Reflow to fit 80-col. | Mike Stump | 2009-05-14 | 1 | -2/+2 |
| | | | | | llvm-svn: 71812 | ||||
| * | Down with static variables! | Owen Anderson | 2009-05-14 | 1 | -4/+8 |
| | | | | | | | Part one of many. llvm-svn: 71785 | ||||
| * | Tag blocks with DW_AT_APPLE_block. | Mike Stump | 2009-05-14 | 1 | -0/+7 |
| | | | | | llvm-svn: 71784 | ||||
| * | Reuse existing getUnderlyingObject instead of | Dale Johannesen | 2009-05-14 | 1 | -24/+6 |
| | | | | | | | adding another copy. llvm-svn: 71783 | ||||
| * | Non-functionality changes: | Bill Wendling | 2009-05-14 | 1 | -110/+85 |
| | | | | | | | | | - Reformatting. - Use while() instead of do-while(). - Move simple constructors into .h file. llvm-svn: 71782 | ||||
| * | Clean up this file fixing 80-column violations, bad formatting, etc. No ↵ | Bill Wendling | 2009-05-14 | 1 | -60/+69 |
| | | | | | | | functionality change. llvm-svn: 71781 | ||||
| * | Revert r71744. I must not have understood this correctly, because the | Bob Wilson | 2009-05-14 | 1 | -4/+9 |
| | | | | | | | assertion is failing for some tests. llvm-svn: 71779 | ||||
| * | Use abs64 in one more place. | Dale Johannesen | 2009-05-14 | 1 | -1/+1 |
| | | | | | llvm-svn: 71775 | ||||
| * | Fix for PR4124. Make TwoAddressFormPass::FindLastUseInMBB return the real ↵ | Lang Hames | 2009-05-14 | 1 | -2/+5 |
| | | | | | | | last use. llvm-svn: 71769 | ||||
| * | Update the names of the exception handling sjlj instrinsics to | Jim Grosbach | 2009-05-14 | 4 | -14/+16 |
| | | | | | | | | | | llvm.eh.sjlj.* for better clarity as to their purpose and scope. Add a description of llvm.eh.sjlj.setjmp to ExceptionHandling.html. (llvm.eh.sjlj.longjmp documentation coming when that implementation is added). llvm-svn: 71758 | ||||
| * | Add an assert to turn a segfault on an unsupported inline | Dan Gohman | 2009-05-14 | 1 | -0/+2 |
| | | | | | | | asm construct into an assertion failure. llvm-svn: 71757 | ||||
| * | Add dumping support for DW_AT_APPLE_isa and DW_AT_APPLE_block. | Mike Stump | 2009-05-14 | 1 | -0/+2 |
| | | | | | | | Radar 6867696 llvm-svn: 71750 | ||||

