| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Update comment | Chris Lattner | 2004-02-02 | 1 | -1/+1 |
| | | | | | llvm-svn: 11082 | ||||
| * | No need to declare implicit uses/defs of ST0 | Chris Lattner | 2004-02-02 | 1 | -1/+1 |
| | | | | | llvm-svn: 11081 | ||||
| * | Bug fixed | Chris Lattner | 2004-02-02 | 1 | -1/+2 |
| | | | | | llvm-svn: 11080 | ||||
| * | Make deadarghaX0r warning louder. | Brian Gaeke | 2004-02-02 | 1 | -1/+1 |
| | | | | | | | (I just love typing haX0r. haX0r haX0r haX0r.) llvm-svn: 11079 | ||||
| * | Generate the fchs instruction to negate a floating point number | Chris Lattner | 2004-02-02 | 2 | -1/+15 |
| | | | | | llvm-svn: 11078 | ||||
| * | Add support for OneArgFPRW instructions, fix a couple of typeos | Chris Lattner | 2004-02-02 | 1 | -12/+43 |
| | | | | | llvm-svn: 11077 | ||||
| * | Include PowerPC.h. Flesh out the stub versions of addPassesToEmitAssembly() | Brian Gaeke | 2004-02-02 | 1 | -3/+13 |
| | | | | | | | and addPassesToJITCompile() slightly. llvm-svn: 11076 | ||||
| * | Add a -march=powerpc option. Automatically select it if this looks like a | Brian Gaeke | 2004-02-02 | 1 | -1/+10 |
| | | | | | | | | big-endian, 32-bit module, or if __ppc__, __POWERPC__, or __APPLE__ are defined. llvm-svn: 11075 | ||||
| * | Link in the PowerPC back-end. | Brian Gaeke | 2004-02-02 | 1 | -0/+1 |
| | | | | | llvm-svn: 11074 | ||||
| * | Add comments describing how you would add prototypes for factory methods for | Brian Gaeke | 2004-02-02 | 1 | -0/+15 |
| | | | | | | | PowerPC-specific passes here. llvm-svn: 11073 | ||||
| * | Add prototype for llvm::allocatePowerPCTargetMachine(). | Brian Gaeke | 2004-02-02 | 1 | -0/+8 |
| | | | | | llvm-svn: 11072 | ||||
| * | Codegen -0.0 correctly. Do not use fldz! This is another -0.0 == +0.0 ↵ | Chris Lattner | 2004-02-02 | 1 | -4/+4 |
| | | | | | | | problem, arg. llvm-svn: 11070 | ||||
| * | Correct the method I just added to actually return false sometimes | Chris Lattner | 2004-02-02 | 1 | -1/+1 |
| | | | | | llvm-svn: 11069 | ||||
| * | Add a new method to ConstantFP | Chris Lattner | 2004-02-02 | 1 | -0/+18 |
| | | | | | llvm-svn: 11068 | ||||
| * | Bug fixed | Chris Lattner | 2004-02-02 | 1 | -0/+1 |
| | | | | | llvm-svn: 11067 | ||||
| * | Create an object for tracking physical register usage. This will look | Alkis Evlogimenos | 2004-02-02 | 1 | -106/+116 |
| | | | | | | | much better when I get rid of the reserved registers. llvm-svn: 11066 | ||||
| * | Bug fixed | Chris Lattner | 2004-02-01 | 1 | -0/+1 |
| | | | | | llvm-svn: 11065 | ||||
| * | Second half of a fix for PR218 & ↵ | Chris Lattner | 2004-02-01 | 1 | -4/+43 |
| | | | | | | | | | | | test/Regression/Assembler/2004-02-01-NegativeZero.llx. Basically we store floating point values as their integral components, instead of relying on the semantics of floating point < to differentiate between values. This is likely to make the map search be faster anyway. llvm-svn: 11064 | ||||
| * | The first half of a fix for PR218 & ↵ | Chris Lattner | 2004-02-01 | 1 | -3/+12 |
| | | | | | | | test/Regression/Assembler/2004-02-01-NegativeZero.llx llvm-svn: 11063 | ||||
| * | Add another test | Chris Lattner | 2004-02-01 | 1 | -1/+2 |
| | | | | | llvm-svn: 11062 | ||||
| * | Change weight array into a vector and make it as big as the number of | Alkis Evlogimenos | 2004-02-01 | 1 | -17/+25 |
| | | | | | | | | | | | | registers (not as the max number of registers). Change toSpill from a std::set into a std::vector<bool>. Use the reverse iterator adapter to do a reverse scan of allocatable registers. llvm-svn: 11061 | ||||
| * | Use std::map::count() instead of std::map::find() != std::map::end() | Alkis Evlogimenos | 2004-02-01 | 1 | -5/+3 |
| | | | | | | | where appropriate. llvm-svn: 11060 | ||||
| * | New testcase for PR218 | Chris Lattner | 2004-02-01 | 1 | -0/+5 |
| | | | | | llvm-svn: 11059 | ||||
| * | Add MRegisterInfo::getNumRegs(). | Alkis Evlogimenos | 2004-02-01 | 1 | -0/+6 |
| | | | | | llvm-svn: 11058 | ||||
| * | Change string for joined intervals. | Alkis Evlogimenos | 2004-02-01 | 1 | -1/+1 |
| | | | | | llvm-svn: 11057 | ||||
| * | FpMOV is also a move instruction. | Alkis Evlogimenos | 2004-02-01 | 1 | -1/+2 |
| | | | | | llvm-svn: 11055 | ||||
| * | Change xor to ^. | Alkis Evlogimenos | 2004-02-01 | 1 | -1/+1 |
| | | | | | llvm-svn: 11054 | ||||
| * | Now that tailduplication does not bork SSA form, run mem2reg earlier in gccas. | Chris Lattner | 2004-02-01 | 1 | -2/+4 |
| | | | | | | | | | | | This tremendously improves the code generated by the LLVM optimizer, primarily by making the inliner more aggressive. For example, it improves the stepanov benchmark from 55.56 mega-additions/sec to 98.04 Ma/s. It also improves the oopack/iterator benchmark from 338.3MFLOPS/s to 1103.4MFLOPS/s. Less noteworthy, it improves oopack/matrix from 573 -> 641 MFLOPS/s. llvm-svn: 11053 | ||||
| * | Disable tail duplication in any "hard" cases, where it might break SSA form. | Chris Lattner | 2004-02-01 | 1 | -1/+27 |
| | | | | | llvm-svn: 11052 | ||||
| * | Fix a bug in a recent checkin | Chris Lattner | 2004-02-01 | 1 | -1/+1 |
| | | | | | llvm-svn: 11050 | ||||
| * | Fix the count of the number of instructions removed | Chris Lattner | 2004-02-01 | 1 | -0/+1 |
| | | | | | llvm-svn: 11049 | ||||
| * | Missed one silly assert :-) | Alkis Evlogimenos | 2004-02-01 | 1 | -1/+1 |
| | | | | | llvm-svn: 11048 | ||||
| * | Simplify joinIntervals() code. | Alkis Evlogimenos | 2004-02-01 | 1 | -62/+57 |
| | | | | | llvm-svn: 11047 | ||||
| * | Remove all of the annoying statistics now that I'm finished (for the near | Chris Lattner | 2004-02-01 | 3 | -89/+0 |
| | | | | | | | term) working on bytecode size stuff. llvm-svn: 11046 | ||||
| * | Use MRegisterInfo::isPhysicalRegister and | Alkis Evlogimenos | 2004-02-01 | 2 | -33/+33 |
| | | | | | | | MRegisterInfo::isVirtualRegister. llvm-svn: 11045 | ||||
| * | Fix a crasher bug in my constant folding rewrite | Chris Lattner | 2004-02-01 | 1 | -3/+10 |
| | | | | | llvm-svn: 11044 | ||||
| * | Print an error message if we can't materialize the bytecode file | Chris Lattner | 2004-02-01 | 1 | -5/+11 |
| | | | | | llvm-svn: 11043 | ||||
| * | Add comments | Chris Lattner | 2004-02-01 | 1 | -1/+4 |
| | | | | | llvm-svn: 11042 | ||||
| * | Print an error message if there is an error materialize the bc file. | Chris Lattner | 2004-02-01 | 1 | -1/+6 |
| | | | | | llvm-svn: 11041 | ||||
| * | Apply final part of Chris' patch. | Alkis Evlogimenos | 2004-01-31 | 1 | -11/+9 |
| | | | | | llvm-svn: 11040 | ||||
| * | Be a little smarter on the way we handle physical register defs. | Alkis Evlogimenos | 2004-01-31 | 1 | -26/+13 |
| | | | | | llvm-svn: 11038 | ||||
| * | Finegrainify namespacification, use new MRegisterInfo::isVirtualRegister | Chris Lattner | 2004-01-31 | 2 | -9/+7 |
| | | | | | | | method llvm-svn: 11037 | ||||
| * | Fix, correctly this time, the computation of the return value | Chris Lattner | 2004-01-31 | 1 | -15/+16 |
| | | | | | | | | | | Fix a spello Tighten up the assertion checking No functionality changes. llvm-svn: 11036 | ||||
| * | * Fix incorrect computation of the runOnMachineFunction return value | Chris Lattner | 2004-01-31 | 1 | -30/+23 |
| | | | | | | | * Turn a bunch of instance variables into automatic variables llvm-svn: 11035 | ||||
| * | Remove unneeded #includes | Chris Lattner | 2004-01-31 | 1 | -12/+10 |
| | | | | | | | | | Move Passes.h (which defines the interface to this file) to the top. Move statistics to the top of the file. Add a comment llvm-svn: 11034 | ||||
| * | Avoid referencing deleted DSgraphs when merging an SCC into a larger SCC. This | Chris Lattner | 2004-01-31 | 1 | -9/+11 |
| | | | | | | | fixes the crash in 176.gcc. llvm-svn: 11033 | ||||
| * | Merge safe parts from last night's buggy commit. These do not break | Alkis Evlogimenos | 2004-01-31 | 2 | -67/+78 |
| | | | | | | | any test cases :-) llvm-svn: 11032 | ||||
| * | Add two static methods to avoid having client code explicitly compare against | Chris Lattner | 2004-01-31 | 1 | -0/+12 |
| | | | | | | | FirstVirtualRegister llvm-svn: 11031 | ||||
| * | Optimize liveAt() and overlaps(). We now use a binary search instead | Alkis Evlogimenos | 2004-01-31 | 1 | -20/+30 |
| | | | | | | | | | of a linear search to find the first range for comparisons. This cuts down the linear scan register allocator running time by a factor of 3 in 254.perlbmk and by a factor of 2.2 in 176.gcc. llvm-svn: 11030 | ||||
| * | Revert last night's changes as they broke some tests. Will remerge parts of ↵ | Alkis Evlogimenos | 2004-01-31 | 2 | -115/+115 |
| | | | | | | | the patch. llvm-svn: 11029 | ||||

