| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Fix a bogus comment | Eli Bendersky | 2012-12-13 | 1 | -3/+3 |
| * | Sorry about the churn. One more change to getOptimalMemOpType() hook. Did I | Evan Cheng | 2012-12-12 | 2 | -14/+12 |
| * | - Rename isLegalMemOpType to isSafeMemOpType. "Legal" is a very overloade term. | Evan Cheng | 2012-12-12 | 2 | -11/+12 |
| * | Avoid using lossy load / stores for memcpy / memset expansion. e.g. | Evan Cheng | 2012-12-12 | 2 | -0/+15 |
| * | Revert EVT->MVT changes, r169836-169851, due to buildbot failures. | Patrik Hagglund | 2012-12-11 | 3 | -10/+10 |
| * | Change TargetLowering::getTypeForExtArgOrReturn to take and return | Patrik Hagglund | 2012-12-11 | 2 | -5/+5 |
| * | Change TargetLowering::RegisterTypeForVT to contain MVTs, instead of | Patrik Hagglund | 2012-12-11 | 1 | -2/+2 |
| * | Change TargetLowering::findRepresentativeClass to take an MVT, instead | Patrik Hagglund | 2012-12-11 | 2 | -3/+3 |
| * | Fall back to the selection dag isel to select tail calls. | Chad Rosier | 2012-12-11 | 1 | -0/+4 |
| * | Some enhancements for memcpy / memset inline expansion. | Evan Cheng | 2012-12-10 | 2 | -4/+10 |
| * | Revert "Make '-mtune=x86_64' assume fast unaligned memory accesses." | Chandler Carruth | 2012-12-10 | 1 | -2/+1 |
| * | Make '-mtune=x86_64' assume fast unaligned memory accesses. | Chandler Carruth | 2012-12-10 | 1 | -1/+2 |
| * | Fix a typo in my previous commit -- bloomfield is 0x1A not 0x2A. | Chandler Carruth | 2012-12-10 | 1 | -1/+1 |
| * | Address a FIXME and update the fast unaligned memory feature for newer | Chandler Carruth | 2012-12-10 | 2 | -13/+21 |
| * | - Re-enable population count loop idiom recognization | Shuxin Yang | 2012-12-09 | 3 | -1/+20 |
| * | Revert the patches adding a popcount loop idiom recognition pass. | Chandler Carruth | 2012-12-08 | 3 | -20/+1 |
| * | s/AttrListPtr/AttributeSet/g to better label what this class is going to be i... | Bill Wendling | 2012-12-07 | 1 | -1/+1 |
| * | When we use the BLEND instruction that uses the MSB as a mask, we can remove | Nadav Rotem | 2012-12-07 | 1 | -1/+6 |
| * | X86: Prefer using VPSHUFD over VPERMIL because it has better throughput. | Nadav Rotem | 2012-12-07 | 1 | -3/+4 |
| * | Replace r169459 with something safer. Rather than having computeMaskedBits to | Evan Cheng | 2012-12-06 | 2 | -38/+25 |
| * | Remove unneeded function, since PR8156 was fixed over a year ago. | Jakub Staszak | 2012-12-06 | 1 | -18/+1 |
| * | Simplify code. | Jakub Staszak | 2012-12-06 | 1 | -3/+1 |
| * | Remove intrinsic specific instructions for (V)MOVQUmr with patterns pointing ... | Craig Topper | 2012-12-06 | 1 | -13/+5 |
| * | Mark MOVDQ(A/U)rm as ReMaterializable. Mark all MOVDQ(A/U) instructions as ne... | Craig Topper | 2012-12-06 | 2 | -4/+9 |
| * | Let targets provide hooks that compute known zero and ones for any_extend | Evan Cheng | 2012-12-06 | 2 | -0/+38 |
| * | Added a option to the disassembler to print immediates as hex. | Kevin Enderby | 2012-12-05 | 1 | -4/+4 |
| * | Simplified BLEND pattern matching for shuffles. | Elena Demikhovsky | 2012-12-05 | 4 | -78/+68 |
| * | Add x86 isel lowering logic to form bit test with inverted condition. e.g. | Evan Cheng | 2012-12-05 | 1 | -6/+17 |
| * | Make NaCl naming consistent. The triple OSType is called NaCl and is represented | Eli Bendersky | 2012-12-04 | 1 | -1/+1 |
| * | Sort includes for all of the .h files under the 'lib' tree. These were | Chandler Carruth | 2012-12-04 | 6 | -13/+13 |
| * | Use the new script to sort the includes of every file under lib. | Chandler Carruth | 2012-12-03 | 21 | -95/+94 |
| * | rdar://12100355 (part 1) | Shuxin Yang | 2012-11-29 | 3 | -1/+20 |
| * | I changed hasAVX() to hasFp256() and hasAVX2() to hasInt256() in X86IselLower... | Elena Demikhovsky | 2012-11-29 | 2 | -102/+104 |
| * | Remove all references to TargetInstrInfoImpl. | Jakob Stoklund Olesen | 2012-11-28 | 1 | -3/+3 |
| * | X86: do not fold load instructions such as [V]MOVS[S|D] to other instructions | Manman Ren | 2012-11-27 | 1 | -0/+15 |
| * | Remove the X86 Maximal Stack Alignment Check pass as it is no longer necessary. | Chad Rosier | 2012-11-26 | 3 | -50/+0 |
| * | Normalize splat 256bit vectors with 8 elements. | Jakub Staszak | 2012-11-26 | 1 | -1/+1 |
| * | Decouple MCInstBuilder from the streamer per Eli's request. | Benjamin Kramer | 2012-11-26 | 1 | -17/+14 |
| * | Add MCInstBuilder, a utility class to simplify MCInst creation similar to Mac... | Benjamin Kramer | 2012-11-26 | 1 | -39/+22 |
| * | Fix execution domain for packed FMA4 instructions. | Craig Topper | 2012-11-21 | 2 | -28/+30 |
| * | Mark ISD::FMA as Legal instead of custom for x86 with FMA3/FMA4. Needed so th... | Craig Topper | 2012-11-21 | 1 | -6/+6 |
| * | Make calcLiveInMask method static. | Jakub Staszak | 2012-11-21 | 1 | -1/+1 |
| * | Make isScratchReg and isFPCopy methods static. | Jakub Staszak | 2012-11-21 | 1 | -2/+2 |
| * | Add obvious constantness. | Jakub Staszak | 2012-11-20 | 1 | -3/+3 |
| * | Intel OCL built-ins calling conventions now support MacOS 32-bit. | Elena Demikhovsky | 2012-11-20 | 1 | -23/+24 |
| * | Add the Erlang/HiPE calling convention, patch by Yiannis Tsiouris. | Duncan Sands | 2012-11-16 | 3 | -5/+63 |
| * | Use roundps/pd for llvm.ceil, llvm.trunc, llvm.rint, and llvm.nearbyint of ve... | Craig Topper | 2012-11-16 | 2 | -0/+68 |
| * | Return 0 instead of false. | Jakub Staszak | 2012-11-15 | 1 | -3/+3 |
| * | Simplify code. | Jakub Staszak | 2012-11-15 | 1 | -1/+1 |
| * | Add llvm.ceil, llvm.trunc, llvm.rint, llvm.nearbyint intrinsics. | Craig Topper | 2012-11-15 | 1 | -1/+5 |
