| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Turn optimize compares back on with fix. We needed to test that a machine op was | Bill Wendling | 2010-08-10 | 1 | -1/+1 |
| * | Delete some unused instructions. | Evan Cheng | 2010-08-10 | 3 | -99/+0 |
| * | Re-apply r110655 with fixes. Epilogue must restore sp from fp if the function... | Evan Cheng | 2010-08-10 | 5 | -63/+73 |
| * | Revert r110655, "Fix ARM hasFP() semantics. It should return true whenever FP | Daniel Dunbar | 2010-08-10 | 5 | -68/+63 |
| * | Fix ARM hasFP() semantics. It should return true whenever FP register is | Evan Cheng | 2010-08-10 | 5 | -63/+68 |
| * | Add AVX movnt{pd,ps,dq} 256-bit intrinsics | Bruno Cardoso Lopes | 2010-08-10 | 1 | -0/+7 |
| * | Add AVX movmsk 256-bit intrinsics | Bruno Cardoso Lopes | 2010-08-10 | 1 | -8/+6 |
| * | Support AVX 256-bit load and store intrinsics | Bruno Cardoso Lopes | 2010-08-10 | 2 | -6/+20 |
| * | Patterns to match AVX cmp instructions | Bruno Cardoso Lopes | 2010-08-10 | 1 | -10/+8 |
| * | Add matching patterns for vblend AVX intrinsics | Bruno Cardoso Lopes | 2010-08-10 | 1 | -11/+6 |
| * | Wording. | Eric Christopher | 2010-08-09 | 1 | -1/+1 |
| * | ARMBaseRegisterInfo::hasFP() has been broken for a while now. :-( | Evan Cheng | 2010-08-09 | 1 | -1/+2 |
| * | Add VCVTPD2PS, VCVTPS2DQ, VCVTPS2PDY, VCVTTPD2DQY, VCVTTPS2DQ and VCVTPD2DQ 2... | Bruno Cardoso Lopes | 2010-08-09 | 1 | -0/+30 |
| * | Add patterns to AVX conversions instructions. Do that instead of declaring mo... | Bruno Cardoso Lopes | 2010-08-09 | 1 | -9/+20 |
| * | CMake: eliminated unnecessary target_link_libraries. | Oscar Fuentes | 2010-08-09 | 1 | -1/+0 |
| * | Explicitly initialize SlowFPBrcc and Pref32BitThumb to false. | Evan Cheng | 2010-08-09 | 1 | -0/+2 |
| * | Change -prefer-32bit-thumb to attribute -mattr=+32bit instead to disable more... | Evan Cheng | 2010-08-09 | 4 | -8/+10 |
| * | Memory version of vcvtdq2pd intrinsic | Bruno Cardoso Lopes | 2010-08-09 | 1 | -1/+3 |
| * | Patterns to match vinsert, vbroadcast, vmovmask and vcvtdq2pd AVX intrinsics | Bruno Cardoso Lopes | 2010-08-09 | 1 | -20/+59 |
| * | Add an option to disable 32 -> 16-bit Thumb2 size reduction pass for experime... | Evan Cheng | 2010-08-09 | 1 | -2/+7 |
| * | Have SPU handle halfvec stores aligned by 8 bytes. | Kalle Raiskila | 2010-08-09 | 5 | -9/+57 |
| * | Add optimization to Target/README.txt. | Nick Lewycky | 2010-08-08 | 1 | -0/+16 |
| * | Use the "isCompare" machine instruction attribute instead of calling the | Bill Wendling | 2010-08-08 | 3 | -10/+11 |
| * | Use sdmem and sse_load_f64 (etc.) for the vector | Dale Johannesen | 2010-08-07 | 1 | -8/+13 |
| * | Patterns to match AVX 256-bit vzero intrinsics | Bruno Cardoso Lopes | 2010-08-06 | 1 | -4/+4 |
| * | Patterns to match AVX 256-bit permutation intrinsics | Bruno Cardoso Lopes | 2010-08-06 | 2 | -13/+48 |
| * | Remove empty processFunctionBeforeFrameFinalized(). The default | Jim Grosbach | 2010-08-06 | 2 | -6/+0 |
| * | Reapply r110396, with fixes to appease the Linux buildbot gods. | Owen Anderson | 2010-08-06 | 31 | -36/+36 |
| * | Fix eabi calling convention when a 64 bit value shadows r3. | Rafael Espindola | 2010-08-06 | 1 | -1/+1 |
| * | Patterns to match AVX 256-bit horizontal arithmetic intrinsics | Bruno Cardoso Lopes | 2010-08-06 | 1 | -14/+12 |
| * | Patterns to match AVX 256-bit arithmetic intrinsics | Bruno Cardoso Lopes | 2010-08-06 | 1 | -30/+75 |
| * | Add the Optimize Compares pass (disabled by default). | Bill Wendling | 2010-08-06 | 2 | -0/+67 |
| * | Revert r110396 to fix buildbots. | Owen Anderson | 2010-08-06 | 31 | -36/+36 |
| * | Add an option to always emit realignment code for a particular module. | Eric Christopher | 2010-08-05 | 1 | -1/+34 |
| * | Don't use PassInfo* as a type identifier for passes. Instead, use the addres... | Owen Anderson | 2010-08-05 | 31 | -36/+36 |
| * | Remove IntrWriteMem, as it's the default. Rename IntrWriteArgMem | Dan Gohman | 2010-08-05 | 1 | -9/+3 |
| * | Support very basic (doesn't include ABI support in the front-end, varags, ...... | Bruno Cardoso Lopes | 2010-08-05 | 2 | -4/+30 |
| * | Handle the memory barrier pseudo that goes to nothing for the JIT. | Eric Christopher | 2010-08-05 | 1 | -1/+7 |
| * | Set hasSideEffects on the 64-bit no-sse memory barrier. | Eric Christopher | 2010-08-05 | 1 | -1/+1 |
| * | For local variables in functions with a frame pointer, use FP as a base | Jim Grosbach | 2010-08-05 | 2 | -18/+39 |
| * | Fix indentation. | Bob Wilson | 2010-08-05 | 1 | -6/+6 |
| * | Add an ARM RSCrr instruction for disassembly only. | Bob Wilson | 2010-08-05 | 1 | -0/+8 |
| * | Be a little bit more specific about target for the memory barrier | Eric Christopher | 2010-08-05 | 2 | -2/+4 |
| * | Handle the pseudo in MCInstLower. | Eric Christopher | 2010-08-05 | 1 | -0/+6 |
| * | Add an ARM RSBrr instruction for disassembly only. | Bob Wilson | 2010-08-05 | 1 | -1/+9 |
| * | Silence a GCC warning about && and || without explicit parentheses. This | Chandler Carruth | 2010-08-05 | 1 | -3/+3 |
| * | ARM "rrx" shift operands do not have an immediate. PR7790. | Bob Wilson | 2010-08-05 | 2 | -24/+18 |
| * | Make x86-64 membarriers work without sse and clean up some of the | Eric Christopher | 2010-08-04 | 3 | -4/+14 |
| * | and back in. false alarm on the tests from another unrelated local change. | Jim Grosbach | 2010-08-04 | 1 | -2/+7 |
| * | PR7814: Truncates cannot be ignored for signed comparisons. | Eli Friedman | 2010-08-04 | 1 | -1/+2 |
