| Commit message (Collapse) | Author | Age | Files | Lines | ||
|---|---|---|---|---|---|---|
| ... | ||||||
| * | remove "Slot", it is dead | Chris Lattner | 2006-03-08 | 1 | -2/+25 | |
| | | | | | llvm-svn: 26609 | |||||
| * | Codegen copysign[f] into a FCOPYSIGN node | Chris Lattner | 2006-03-05 | 1 | -2/+44 | |
| | | | | | llvm-svn: 26542 | |||||
| * | Add more vector NodeTypes: VSDIV, VUDIV, VAND, VOR, and VXOR. | Evan Cheng | 2006-03-03 | 1 | -4/+14 | |
| | | | | | llvm-svn: 26504 | |||||
| * | remove the read/write port/io intrinsics. | Chris Lattner | 2006-03-03 | 1 | -62/+0 | |
| | | | | | llvm-svn: 26479 | |||||
| * | Vector ops lowering. | Evan Cheng | 2006-03-01 | 1 | -45/+65 | |
| | | | | | llvm-svn: 26436 | |||||
| * | Fix a problem on itanium with memset. The value to set has been promoted to | Chris Lattner | 2006-02-20 | 1 | -2/+6 | |
| | | | | | | | i64 before this code, so zero_ext doesn't work. llvm-svn: 26290 | |||||
| * | kill ADD_PARTS & SUB_PARTS and replace them with fancy new ADDC, ADDE, SUBC | Nate Begeman | 2006-02-17 | 1 | -14/+43 | |
| | | | | | | | | and SUBE nodes that actually expose what's going on and allow for significant simplifications in the targets. llvm-svn: 26255 | |||||
| * | Fix another miscompilation exposed by lencode, where we lowered i64->f32 | Chris Lattner | 2006-02-17 | 1 | -1/+1 | |
| | | | | | | | | conversions to __floatdidf instead of __floatdisf on targets that support f32 but not i64 (e.g. sparc). llvm-svn: 26254 | |||||
| * | Should not combine ISD::LOCATIONs until we have scheme to remove from | Jim Laskey | 2006-02-15 | 1 | -2/+2 | |
| | | | | | | | MachineDebugInfo tables. llvm-svn: 26216 | |||||
| * | Fix a latent bug in the call sequence handling stuff. Some targets (e.g. x86) | Chris Lattner | 2006-02-14 | 1 | -4/+11 | |
| | | | | | | | create these nodes with flag results. Remember that we legalized them. llvm-svn: 26156 | |||||
| * | Completely rewrite libcall insertion by the legalizer, providing the | Chris Lattner | 2006-02-13 | 1 | -198/+217 | |
| | | | | | | | | | | | | | | | | | following handy-dandy properties: 1. it is always correct now 2. it is much faster than before 3. it is easier to understand This implementation builds off of the recent simplifications of the legalizer that made it single-pass instead of iterative. This fixes JM/lencod, JM/ldecod, and CodeGen/Generic/2006-02-12-InsertLibcall.ll (at least on PPC). llvm-svn: 26144 | |||||
| * | *** empty log message *** | Nate Begeman | 2006-02-01 | 1 | -0/+1 | |
| | | | | | llvm-svn: 25879 | |||||
| * | Fix some of the stuff in the PPC README file, and clean up legalization | Nate Begeman | 2006-02-01 | 1 | -219/+205 | |
| | | | | | | | of the SELECT_CC, BR_CC, and BRTWOWAY_CC nodes. llvm-svn: 25875 | |||||
| * | Allow custom lowering of fabs. I forgot to check in this change which | Evan Cheng | 2006-01-31 | 1 | -1/+6 | |
| | | | | | | | caused several test failures. llvm-svn: 25852 | |||||
| * | Only insert an AND when converting from BR_COND to BRCC if needed. | Chris Lattner | 2006-01-31 | 1 | -1/+3 | |
| | | | | | llvm-svn: 25832 | |||||
| * | Fix a bug in my legalizer reworking that caused the X86 backend to not get | Chris Lattner | 2006-01-30 | 1 | -2/+1 | |
| | | | | | | | | a chance to custom legalize setcc, which broke a bunch of C++ Codes. Testcase here: CodeGen/X86/2006-01-30-LongSetcc.ll llvm-svn: 25821 | |||||
| * | don't insert an and node if it isn't needed here, this can prevent folding | Chris Lattner | 2006-01-30 | 1 | -1/+3 | |
| | | | | | | | of lowered target nodes. llvm-svn: 25804 | |||||
| * | Fix RET of promoted values on targets that custom expand RET to a target node. | Chris Lattner | 2006-01-29 | 1 | -8/+9 | |
| | | | | | llvm-svn: 25794 | |||||
| * | cleanups to the ValueTypeActions interface | Chris Lattner | 2006-01-29 | 1 | -7/+5 | |
| | | | | | llvm-svn: 25785 | |||||
| * | Remove some special case hacks for CALLSEQ_*, using UpdateNodeOperands | Chris Lattner | 2006-01-29 | 1 | -16/+21 | |
| | | | | | | | instead. llvm-svn: 25780 | |||||
| * | Allow custom expansion of ConstantVec nodes. PPC will use this in the future. | Chris Lattner | 2006-01-29 | 1 | -22/+35 | |
| | | | | | llvm-svn: 25774 | |||||
| * | Legalize ConstantFP into TargetConstantFP when the target allows. Implement | Chris Lattner | 2006-01-29 | 1 | -1/+18 | |
| | | | | | | | custom expansion of ConstantFP nodes. llvm-svn: 25772 | |||||
| * | eliminate uses of SelectionDAG::getBR2Way_CC | Chris Lattner | 2006-01-29 | 1 | -6/+14 | |
| | | | | | llvm-svn: 25767 | |||||
| * | Use the new "UpdateNodeOperands" method to simplify LegalizeDAG and make it | Chris Lattner | 2006-01-28 | 1 | -317/+194 | |
| | | | | | | | | | | | | | | | faster. This cuts about 120 lines of code out of the legalizer (mostly code checking to see if operands have changed). It also fixes an ugly performance issue, where the legalizer cloned the entire graph after any change. Now the "UpdateNodeOperands" method gives it a chance to reuse nodes if the operands of a node change but not its opcode or valuetypes. This speeds up instruction selection time on kimwitu++ by about 8.2% with a release build. llvm-svn: 25746 | |||||
| * | minor tweaks | Chris Lattner | 2006-01-28 | 1 | -21/+21 | |
| | | | | | llvm-svn: 25740 | |||||
| * | move a bunch of code, no other change. | Chris Lattner | 2006-01-28 | 1 | -326/+324 | |
| | | | | | llvm-svn: 25739 | |||||
| * | remove a couple more now-extraneous legalizeop's | Chris Lattner | 2006-01-28 | 1 | -9/+5 | |
| | | | | | llvm-svn: 25738 | |||||
| * | fix a bug | Chris Lattner | 2006-01-28 | 1 | -2/+1 | |
| | | | | | llvm-svn: 25737 | |||||
| * | Several major changes: | Chris Lattner | 2006-01-28 | 1 | -764/+488 | |
| | | | | | | | | | | | | 1. Pull out the expand cases for BSWAP and CT* into a separate function, reducing the size of LegalizeOp. 2. Fix a bug where expand(bswap i64) was wrong when i64 is legal. 3. Changed LegalizeOp/PromoteOp so that the legalizer never needs to be iterative. It now operates in a single pass over the nodes. 4. Simplify a LOT of code, with a net reduction of ~280 lines. llvm-svn: 25736 | |||||
| * | Eliminate the need for ExpandOp to set 'needsanotheriteration', as it already | Chris Lattner | 2006-01-28 | 1 | -131/+80 | |
| | | | | | | | | | | | | relegalizes the stuff it returns. Add the ability to custom expand ADD/SUB, so that targets don't need to deal with ADD_PARTS/SUB_PARTS if they don't want. Fix some obscure potential bugs and simplify code. llvm-svn: 25732 | |||||
| * | Instead of making callers of ExpandLibCall legalize the result, make | Chris Lattner | 2006-01-28 | 1 | -6/+7 | |
| | | | | | | | ExpandLibCall do it itself. llvm-svn: 25731 | |||||
| * | Eliminate the need to do another iteration of the legalizer after inserting | Chris Lattner | 2006-01-28 | 1 | -3/+2 | |
| | | | | | | | a libcall. llvm-svn: 25730 | |||||
| * | Implement Promote for VAARG, and allow it to be custom promoted for people | Nate Begeman | 2006-01-28 | 1 | -1/+25 | |
| | | | | | | | who don't want the default behavior (Alpha). llvm-svn: 25726 | |||||
| * | Remove the ISD::CALL and ISD::TAILCALL nodes | Chris Lattner | 2006-01-28 | 1 | -80/+0 | |
| | | | | | llvm-svn: 25721 | |||||
| * | initial selectiondag support for new INLINEASM node. Note that inline asms | Chris Lattner | 2006-01-26 | 1 | -0/+22 | |
| | | | | | | | with outputs or inputs are not supported yet. :) llvm-svn: 25664 | |||||
| * | First part of bug 680: | Nate Begeman | 2006-01-25 | 1 | -6/+151 | |
| | | | | | | | | Remove TLI.LowerVA* and replace it with SDNodes that are lowered the same way as everything else. llvm-svn: 25606 | |||||
| * | Fix an infinite loop I caused by making sure to legalize the flag operand | Chris Lattner | 2006-01-24 | 1 | -1/+9 | |
| | | | | | | | of CALLSEQ_* nodes llvm-svn: 25582 | |||||
| * | Fix Regression/CodeGen/SparcV8/2006-01-22-BitConvertLegalize.ll by making | Chris Lattner | 2006-01-23 | 1 | -2/+4 | |
| | | | | | | | sure that the result of expanding a BIT_CONVERT node is itself legalized. llvm-svn: 25538 | |||||
| * | Fix CodeGen/PowerPC/2006-01-20-ShiftPartsCrash.ll | Chris Lattner | 2006-01-21 | 1 | -15/+15 | |
| | | | | | llvm-svn: 25496 | |||||
| * | remove some unintentionally committed code | Chris Lattner | 2006-01-20 | 1 | -3/+2 | |
| | | | | | llvm-svn: 25483 | |||||
| * | If the target doesn't support f32 natively, insert the FP_EXTEND in target-indep | Chris Lattner | 2006-01-20 | 1 | -2/+3 | |
| | | | | | | | code, so that the LowerReturn code doesn't have to handle it. llvm-svn: 25482 | |||||
| * | Another typo | Evan Cheng | 2006-01-19 | 1 | -1/+1 | |
| | | | | | llvm-svn: 25440 | |||||
| * | was ignoring the legalized chain in this case, fixed SPASS on alpha | Andrew Lenharth | 2006-01-18 | 1 | -1/+1 | |
| | | | | | llvm-svn: 25428 | |||||
| * | Bug fix: missing LegalizeOp() on newly created nodes. | Evan Cheng | 2006-01-17 | 1 | -1/+6 | |
| | | | | | llvm-svn: 25401 | |||||
| * | Adding basic support for Dwarf line number debug information. | Jim Laskey | 2006-01-17 | 1 | -1/+1 | |
| | | | | | | | I promise to keep future commits smaller. llvm-svn: 25396 | |||||
| * | Expand case for 64b Legalize, even though no one should end up using this | Nate Begeman | 2006-01-16 | 1 | -4/+43 | |
| | | | | | | | | (itanium supports bswap natively, alpha should custom lower it using the VAX floating point swapload, ha ha). llvm-svn: 25356 | |||||
| * | Allow the target to specify 'expand' if they just require the amount to | Chris Lattner | 2006-01-15 | 1 | -9/+19 | |
| | | | | | | | be subtracted from the stack pointer. llvm-svn: 25331 | |||||
| * | Fix custom lowering of dynamic_stackalloc | Chris Lattner | 2006-01-15 | 1 | -6/+9 | |
| | | | | | llvm-svn: 25329 | |||||
| * | Token chain results are not always the first or last result. Consider ↵ | Chris Lattner | 2006-01-14 | 1 | -3/+17 | |
| | | | | | | | copyfromreg nodes, where they are the middle result (the flag result is last) llvm-svn: 25325 | |||||
| * | bswap implementation | Nate Begeman | 2006-01-14 | 1 | -0/+68 | |
| | | | | | llvm-svn: 25312 | |||||

