| Commit message (Expand) | Author | Age | Files | Lines |
* | Do not isel load folding bt instructions for pentium m, core, core2, and AMD ... | Evan Cheng | 2009-01-02 | 1 | -2/+3 |
* | Fix some JIT encodings. | Chris Lattner | 2008-12-25 | 1 | -6/+6 |
* | BT memory operands load from their address operand. | Chris Lattner | 2008-12-25 | 1 | -2/+2 |
* | Add instruction patterns and encodings for the x86 bt instructions. | Dan Gohman | 2008-12-23 | 1 | -0/+24 |
* | - Use patterns instead of creating completely new instruction matching patterns, | Bill Wendling | 2008-12-12 | 1 | -368/+335 |
* | Redo the arithmetic with overflow architecture. I was changing the semantics of | Bill Wendling | 2008-12-12 | 1 | -128/+416 |
* | Add sub/mul overflow intrinsics. This currently doesn't have a | Bill Wendling | 2008-12-09 | 1 | -41/+65 |
* | Fix typo, psuedo -> pseudo. | Nick Lewycky | 2008-12-07 | 1 | -27/+27 |
* | Rename isSimpleLoad to canFoldAsLoad, to better reflect its meaning. | Dan Gohman | 2008-12-03 | 1 | -2/+2 |
* | Reapply r60382. This time, don't mark "ADC" nodes with "implicit EFLAGS". | Bill Wendling | 2008-12-02 | 1 | -20/+90 |
* | Temporarily revert r60382. It caused CodeGen/X86/i2k.ll and others to fail. | Bill Wendling | 2008-12-01 | 1 | -103/+26 |
* | - Have "ADD" instructions return an implicit EFLAGS. | Bill Wendling | 2008-12-01 | 1 | -26/+103 |
* | Generate something sensible for an [SU]ADDO op when the overflow/carry flag is | Bill Wendling | 2008-11-26 | 1 | -3/+9 |
* | Don't set neverHasSideEffects on x86's divide instructions, since | Dan Gohman | 2008-11-18 | 1 | -1/+1 |
* | Generate code for TLS instructions. | Nicolas Geoffray | 2008-10-25 | 1 | -4/+5 |
* | Add implicit defs of XMM8 to XMM15 on 32-bit call instructions. While this is... | Evan Cheng | 2008-10-17 | 1 | -1/+2 |
* | Fun x86 encoding tricks: when adding an immediate value of 128, | Dan Gohman | 2008-10-17 | 1 | -0/+11 |
* | Define patterns for shld and shrd that match immediate | Dan Gohman | 2008-10-17 | 1 | -0/+92 |
* | Now that predicates can be composed, simplify several of | Dan Gohman | 2008-10-15 | 1 | -12/+4 |
* | Change CALLSEQ_BEGIN and CALLSEQ_END to take TargetConstant's as | Chris Lattner | 2008-10-11 | 1 | -2/+2 |
* | Model hardwired inputs & outputs of x86 8-bit divides correctly. | Dale Johannesen | 2008-10-07 | 1 | -4/+4 |
* | Make atomic Swap work, 64-bit on x86-32. | Dale Johannesen | 2008-10-03 | 1 | -0/+6 |
* | Pass MemOperand through for 64-bit atomics on 32-bit, | Dale Johannesen | 2008-10-03 | 1 | -0/+1 |
* | Handle some 64-bit atomics on x86-32, some of the time. | Dale Johannesen | 2008-10-02 | 1 | -2/+45 |
* | Split x86's ADJCALLSTACK instructions into 32-bit and 64-bit forms. | Dan Gohman | 2008-10-01 | 1 | -7/+12 |
* | Mark CALL instructions as having a Use of ESP/RSP. | Dan Gohman | 2008-10-01 | 1 | -2/+3 |
* | Fix PR2835. Do not change the width of a volatile load. | Evan Cheng | 2008-09-29 | 1 | -2/+18 |
* | Implement "punpckldq %xmm0, $xmm0" as "pshufd $0x50, %xmm0, %xmm" unless op... | Evan Cheng | 2008-09-26 | 1 | -0/+1 |
* | Fix patterns for SSE4.1 move and sign extend instructions. Also add instructi... | Evan Cheng | 2008-09-24 | 1 | -0/+10 |
* | Reverting r56249. On further investigation, this functionality isn't needed. | Bill Wendling | 2008-09-16 | 1 | -1/+1 |
* | - Change "ExternalSymbolSDNode" to "SymbolSDNode". | Bill Wendling | 2008-09-16 | 1 | -1/+1 |
* | Rename ConstantSDNode::getValue to getZExtValue, for consistency | Dan Gohman | 2008-09-12 | 1 | -2/+2 |
* | Transform (x << (y&31)) -> (x << y). This takes advantage of the fact x86 shi... | Evan Cheng | 2008-08-30 | 1 | -0/+40 |
* | Split the ATOMIC NodeType's to include the size, e.g. | Dale Johannesen | 2008-08-28 | 1 | -20/+20 |
* | Reverting r55190, r55191, and r55192. They broke the build with this error me... | Bill Wendling | 2008-08-22 | 1 | -13/+8 |
* | Anyext tweaks for x86. When extloading a value to i32 or i64, choose | Dan Gohman | 2008-08-22 | 1 | -8/+13 |
* | Move the handling of ANY_EXTEND, SIGN_EXTEND_INREG, and TRUNCATE | Dan Gohman | 2008-08-20 | 1 | -14/+38 |
* | Tablegen generated code already tests the opcode value, so it's not | Dan Gohman | 2008-08-20 | 1 | -18/+16 |
* | Revert r55018 and apply the correct "fix" for the 64-bit sub_and_fetch atomic. | Bill Wendling | 2008-08-20 | 1 | -16/+0 |
* | Add support for the __sync_sub_and_fetch atomics and friends for X86. The code | Bill Wendling | 2008-08-19 | 1 | -0/+16 |
* | Add support for 8 and 16 bit forms of __sync | Dale Johannesen | 2008-08-19 | 1 | -8/+45 |
* | Re-introduce the 8-bit subreg zext-inreg patterns for x86-32, | Dan Gohman | 2008-08-06 | 1 | -0/+10 |
* | xchg does not modify FLAGS. | Dan Gohman | 2008-08-06 | 1 | -1/+1 |
* | Reapply r54147 with a constraint to only use the 8-bit | Dan Gohman | 2008-07-30 | 1 | -10/+22 |
* | Revert 54147. | Dan Gohman | 2008-07-29 | 1 | -10/+0 |
* | Add x86 isel patterns to match what would be a ZERO_EXTEND_INREG operation, | Dan Gohman | 2008-07-28 | 1 | -0/+10 |
* | Fix encoding of atomic compare and swap for i64 | Anton Korobeynikov | 2008-07-22 | 1 | -1/+1 |
* | Added MemOperands to Atomic operations since Atomics touches memory. | Mon P Wang | 2008-06-25 | 1 | -5/+5 |
* | XOR32rr, etc. are not AsCheapAsMove, but MOV32ri, etc. are. | Evan Cheng | 2008-06-18 | 1 | -4/+3 |
* | add missing atomic intrinsic from gcc | Andrew Lenharth | 2008-06-14 | 1 | -23/+3 |