| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Fix JIT encoding of conditional branches | Nate Begeman | 2005-08-26 | 2 | -27/+19 |
| | | | | | llvm-svn: 23076 | ||||
| * | add initial support for converting select_cc -> fsel in the legalizer | Chris Lattner | 2005-08-26 | 3 | -2/+67 |
| | | | | | | | | | | instead of in the backend. This currently handles fsel cases with registers, but doesn't have the 0.0 and -0.0 optimization enabled yet. Once this is finished, special hack for fp immediates can go away. llvm-svn: 23075 | ||||
| * | SUBFIC produces two results, not one. | Nate Begeman | 2005-08-26 | 1 | -4/+4 |
| | | | | | llvm-svn: 23073 | ||||
| * | Implement SHL_PARTS and SRL_PARTS | Nate Begeman | 2005-08-26 | 1 | -0/+38 |
| | | | | | llvm-svn: 23072 | ||||
| * | Emit the lo/hi parts in the right order :) | Chris Lattner | 2005-08-25 | 1 | -1/+1 |
| | | | | | llvm-svn: 23068 | ||||
| * | implement support for 64-bit add/sub, fix a broken assertion for 64-bit | Chris Lattner | 2005-08-25 | 1 | -2/+58 |
| | | | | | | | | return. Allow the udiv breaker-upper to work with any non-zero constant operand. llvm-svn: 23066 | ||||
| * | simplify the add/sub_parts code | Chris Lattner | 2005-08-25 | 1 | -16/+24 |
| | | | | | llvm-svn: 23065 | ||||
| * | Finish implementing SDIV/UDIV by copying over the majik constant code from | Chris Lattner | 2005-08-25 | 1 | -3/+180 |
| | | | | | | | ISelPattern llvm-svn: 23062 | ||||
| * | Simplify some code. It's not clear why the UDIV expanded sequence | Chris Lattner | 2005-08-25 | 1 | -13/+7 |
| | | | | | | | doesn't work for large uint constants, but we'll keep the current behavior llvm-svn: 23061 | ||||
| * | Implement setcc correctly for G5 and non-G5 systems | Chris Lattner | 2005-08-25 | 1 | -11/+17 |
| | | | | | llvm-svn: 23060 | ||||
| * | implement setcc on the G5. We're still missing the non-g5 specific bits, but | Chris Lattner | 2005-08-25 | 1 | -0/+121 |
| | | | | | | | they will come later. llvm-svn: 23059 | ||||
| * | Remove option to make SetCC illegal on PowerPC after long discussion with | Nate Begeman | 2005-08-25 | 1 | -11/+0 |
| | | | | | | | | Chris. This will be accomplished through correctly modeling CR's and subregs. llvm-svn: 23056 | ||||
| * | Add support for sdiv by 2^k and -2^k. Producing code like: | Chris Lattner | 2005-08-25 | 1 | -1/+27 |
| | | | | | | | | | | _test: srawi r2, r3, 2 addze r3, r2 blr llvm-svn: 23052 | ||||
| * | fit in 80 cols | Chris Lattner | 2005-08-25 | 1 | -1/+1 |
| | | | | | llvm-svn: 23051 | ||||
| * | add an enum value | Chris Lattner | 2005-08-25 | 1 | -2/+3 |
| | | | | | llvm-svn: 23048 | ||||
| * | Implement support for taking the address of constant pool indices, which | Chris Lattner | 2005-08-25 | 1 | -0/+10 |
| | | | | | | | | is used by the int -> FP code among other things. This gets 2005-05-12-Int64ToFP past that failure, to dying on lack of support for add_parts llvm-svn: 23042 | ||||
| * | Add support for FP constants, fixing UnitTests/2004-02-02-NegativeZero | Chris Lattner | 2005-08-25 | 1 | -6/+25 |
| | | | | | llvm-svn: 23038 | ||||
| * | Fully implement frame index, so that we can pass the address of alloca's | Chris Lattner | 2005-08-25 | 1 | -4/+12 |
| | | | | | | | around to functions and stuff llvm-svn: 23036 | ||||
| * | implement unconditional branches, fixing UnitTests/2003-05-02-DependentPHI.c | Chris Lattner | 2005-08-25 | 1 | -1/+4 |
| | | | | | llvm-svn: 23034 | ||||
| * | LFS/STFS load and store FP values, not integer ones. This change allows us | Chris Lattner | 2005-08-25 | 1 | -4/+4 |
| | | | | | | | | | | | | | | | | | | | | | | | | to codegen this: float foo() { return 1.245; } into this: _foo: lis r2, ha16(.CPI_foo_0) lfs f1, lo16(.CPI_foo_0)(r2) blr instead of this: _foo: lis r2, ha16(.CPI_foo_0) lfs r2, lo16(.CPI_foo_0)(r2) <-- ouch or f1, r2, r2 <-- ouch blr with the dag isel. llvm-svn: 23033 | ||||
| * | Fix a broken assertion | Chris Lattner | 2005-08-25 | 1 | -1/+1 |
| | | | | | llvm-svn: 23032 | ||||
| * | Fix a warning | Chris Lattner | 2005-08-25 | 1 | -0/+1 |
| | | | | | llvm-svn: 23031 | ||||
| * | fix a warning in optimized build | Chris Lattner | 2005-08-25 | 1 | -1/+1 |
| | | | | | llvm-svn: 23030 | ||||
| * | Fix some warnings | Chris Lattner | 2005-08-25 | 2 | -0/+2 |
| | | | | | llvm-svn: 23029 | ||||
| * | Split IMPLICIT_DEF into IMPLICIT_DEF_GPR and IMPLICIT_DEF_FP, so that the | Chris Lattner | 2005-08-24 | 5 | -7/+21 |
| | | | | | | | | instructions take a consistent reg class. Implement ISD::UNDEF in the dag->dag selector to generate this, fixing UnitTests/2003-07-06-IntOverflow. llvm-svn: 23028 | ||||
| * | implement support for calls | Chris Lattner | 2005-08-24 | 1 | -3/+99 |
| | | | | | llvm-svn: 23026 | ||||
| * | Remove some dead cases. | Chris Lattner | 2005-08-24 | 1 | -7/+1 |
| | | | | | | | | | | | | | | Emit the indcall sequence as: mtctr inreg mr R12, inreg btctr If inreg and R12 aren't coallesced, this reduces the odds of having the mtctr and btctr in the same dispatch group. :) llvm-svn: 23023 | ||||
| * | add an idea | Chris Lattner | 2005-08-24 | 1 | -0/+30 |
| | | | | | llvm-svn: 23020 | ||||
| * | Adjust to new interface | Chris Lattner | 2005-08-24 | 1 | -4/+4 |
| | | | | | llvm-svn: 23010 | ||||
| * | Whoops, fix a thinko. All cases except SETNE are now handled by the | Nate Begeman | 2005-08-24 | 1 | -11/+3 |
| | | | | | | | target independent code in SelectionDAG.cpp llvm-svn: 23002 | ||||
| * | Remove unused statistic | Nate Begeman | 2005-08-24 | 1 | -3/+5 |
| | | | | | | | Prefer 'neg X' to 'subfic 0, X' since neg does not set XER[CA] llvm-svn: 23001 | ||||
| * | Add the "ppc specific" setcc-equivalent select_cc cases | Nate Begeman | 2005-08-24 | 1 | -4/+25 |
| | | | | | | | Prefer 'neg X' to 'subfic 0, X' since it does not set XER[CA] llvm-svn: 23000 | ||||
| * | Add callseq_begin/end support | Chris Lattner | 2005-08-24 | 1 | -0/+10 |
| | | | | | | | Call stil not supported yet llvm-svn: 22998 | ||||
| * | Adjust to new livevars interface | Chris Lattner | 2005-08-23 | 1 | -4/+4 |
| | | | | | llvm-svn: 22991 | ||||
| * | Simplify this code by using LiveVariables::KillsRegister | Chris Lattner | 2005-08-23 | 1 | -35/+11 |
| | | | | | llvm-svn: 22988 | ||||
| * | Fix PR618 and Regression/CodeGen/CBackend/2005-08-23-Fmod.ll by not emitting | Chris Lattner | 2005-08-23 | 1 | -0/+14 |
| | | | | | | | x%y for 'rem' on fp values. llvm-svn: 22984 | ||||
| * | add a note | Chris Lattner | 2005-08-23 | 1 | -0/+20 |
| | | | | | llvm-svn: 22982 | ||||
| * | Ack, typo | Nate Begeman | 2005-08-23 | 1 | -1/+1 |
| | | | | | llvm-svn: 22981 | ||||
| * | Add an option to make SetCC illegal as a beta option | Nate Begeman | 2005-08-23 | 1 | -0/+10 |
| | | | | | llvm-svn: 22979 | ||||
| * | Remove some instructions we no longer generate | Nate Begeman | 2005-08-23 | 1 | -24/+0 |
| | | | | | llvm-svn: 22976 | ||||
| * | Remove some regs that are not used. | Chris Lattner | 2005-08-22 | 2 | -8/+1 |
| | | | | | llvm-svn: 22975 | ||||
| * | Nate noticed that 30% of the malloc/frees in llc come from calls to ↵ | Chris Lattner | 2005-08-22 | 2 | -44/+44 |
| | | | | | | | | | | | LowercaseString in the asmprinter. This changes the .td files to use lower case register names, avoiding the need to do this call. This speeds up the asmprinter from 1.52s to 1.06s on kc++ in a release build. llvm-svn: 22974 | ||||
| * | Fix a crash I introduced into the IA64 backend with my copyfromreg change. | Chris Lattner | 2005-08-22 | 1 | -1/+3 |
| | | | | | | | It used to crash on any function that took float arguments. llvm-svn: 22973 | ||||
| * | Add a pass name for -time-passes output | Chris Lattner | 2005-08-22 | 1 | -0/+1 |
| | | | | | llvm-svn: 22970 | ||||
| * | Implement stores. | Chris Lattner | 2005-08-22 | 1 | -0/+26 |
| | | | | | llvm-svn: 22963 | ||||
| * | Fix compilation of: | Chris Lattner | 2005-08-22 | 1 | -1/+1 |
| | | | | | | | | | | | | | float %test2(float* %P) { %Q = load float* %P %R = add float %Q, %Q ret float %R } By returning the right result. llvm-svn: 22961 | ||||
| * | Make sure expressions only have one use before emitting them into a place ↵ | Chris Lattner | 2005-08-22 | 1 | -3/+4 |
| | | | | | | | that is conditionally executed llvm-svn: 22960 | ||||
| * | Implement most of load support. There is still a bug though. | Chris Lattner | 2005-08-21 | 1 | -8/+96 |
| | | | | | llvm-svn: 22959 | ||||
| * | Don't print out the MBB label for the entry mbb | Chris Lattner | 2005-08-21 | 1 | -2/+6 |
| | | | | | llvm-svn: 22953 | ||||
| * | Simplify the logic for BRTWOWAY_CC handling. The isel code already | Chris Lattner | 2005-08-21 | 1 | -16/+7 |
| | | | | | | | | | simplifies BRTWOWAY into BR if one of the results is a fall-through. Unless I'm missing something, there is no reason to duplicate this in the target-specific code. llvm-svn: 22952 | ||||

