| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Add QPR_VFP2 regclass and add copy_to_regclass nodes, where needed to | Anton Korobeynikov | 2009-09-12 | 1 | -4/+8 |
| * | Unbreak getOnesVector() / getZeroVector() to use valid ARM extended imm's. | Anton Korobeynikov | 2009-09-08 | 1 | -6/+10 |
| * | Add NEON 'laned' operations. This fixes another bunch of gcc testsuite fails and | Anton Korobeynikov | 2009-09-08 | 1 | -1/+364 |
| * | More missed vdup patterns | Anton Korobeynikov | 2009-09-02 | 1 | -0/+12 |
| * | Add support for generating code for vst{234}lane intrinsics. | Bob Wilson | 2009-09-01 | 1 | -0/+34 |
| * | Generate code for vld{234}_lane intrinsics. | Bob Wilson | 2009-09-01 | 1 | -0/+38 |
| * | Add missed pattern | Anton Korobeynikov | 2009-08-30 | 1 | -0/+2 |
| * | Add missed extract_element pattern | Anton Korobeynikov | 2009-08-28 | 1 | -1/+3 |
| * | Forgot about actual change :) | Anton Korobeynikov | 2009-08-27 | 1 | -0/+20 |
| * | Transform float scalar_to_vector into subreg accesses. | Anton Korobeynikov | 2009-08-27 | 1 | -0/+7 |
| * | Remove some unused SDNode definitions. | Bob Wilson | 2009-08-25 | 1 | -27/+0 |
| * | Expose the instruction contraint string as an argument to the NLdSt class. | Bob Wilson | 2009-08-25 | 1 | -22/+18 |
| * | Rename ARM "lane_cst" operands to "nohash_imm" since they are used for | Bob Wilson | 2009-08-21 | 1 | -10/+10 |
| * | Fix some typos and use type-based isel for VZIP/VUZP/VTRN | Anton Korobeynikov | 2009-08-21 | 1 | -9/+3 |
| * | Add nodes & dummy matchers for some v{zip,uzp,trn} instructions | Anton Korobeynikov | 2009-08-21 | 1 | -0/+12 |
| * | Provide vext.{16,32} | Anton Korobeynikov | 2009-08-21 | 1 | -10/+23 |
| * | Add support for Neon VEXT (vector extract) shuffles. | Bob Wilson | 2009-08-19 | 1 | -0/+19 |
| * | Create a new ARM-specific DAG node, VDUP, to represent a splat from a | Bob Wilson | 2009-08-14 | 1 | -22/+10 |
| * | During legalization, change Neon vdup_lane operations from shuffles to | Bob Wilson | 2009-08-14 | 1 | -17/+24 |
| * | Revert r78852 for now. I want to do this differently, but I don't have time | Bob Wilson | 2009-08-13 | 1 | -13/+22 |
| * | Recognize Neon VDUP shuffles during legalization instead of selection. | Bob Wilson | 2009-08-12 | 1 | -22/+13 |
| * | Recognize Neon VREV shuffles during legalization instead of selection. | Bob Wilson | 2009-08-12 | 1 | -25/+12 |
| * | Generate Neon VTBL and VTBX instructions from the corresponding intrinsics. | Bob Wilson | 2009-08-12 | 1 | -0/+53 |
| * | Fix TableGen warnings. This partly reverts my previous change to this file, | Bob Wilson | 2009-08-12 | 1 | -14/+18 |
| * | Add missing chain operands for VLD* and VST* instructions. | Bob Wilson | 2009-08-12 | 1 | -14/+16 |
| * | Split EVT into MVT and EVT, the former representing _just_ a primitive type, ... | Owen Anderson | 2009-08-11 | 1 | -9/+9 |
| * | Use vAny type to get rid of Neon intrinsics that differed only in whether | Bob Wilson | 2009-08-11 | 1 | -36/+36 |
| * | Use new EVT::vAny type to combine Neon intrinsics for VPADD. | Bob Wilson | 2009-08-11 | 1 | -4/+4 |
| * | Fix bug in NEON convert for single-precision FP. This also fixes the tblgen w... | David Goodwin | 2009-08-11 | 1 | -4/+4 |
| * | Rename MVT to EVT, in preparation for splitting SimpleValueType out into its ... | Owen Anderson | 2009-08-10 | 1 | -9/+9 |
| * | Use NEON for single-precision int<->FP conversions. | David Goodwin | 2009-08-10 | 1 | -13/+47 |
| * | Use subclassing to print lane-like immediates (w/o hash) eliminating | Anton Korobeynikov | 2009-08-08 | 1 | -20/+20 |
| * | Add insert_elt / extract_elt patterns for v4f32 stuff. | Anton Korobeynikov | 2009-08-08 | 1 | -19/+30 |
| * | Lane number should be printed w/o hash | Anton Korobeynikov | 2009-08-08 | 1 | -10/+10 |
| * | Use VLDM / VSTM to spill/reload 128-bit Neon registers | Anton Korobeynikov | 2009-08-08 | 1 | -5/+4 |
| * | Implement Neon VZIP and VUZP instructions. These are very similar to VTRN, | Bob Wilson | 2009-08-08 | 1 | -16/+37 |
| * | Implement Neon VTRN instructions. For now, anyway, these are selected | Bob Wilson | 2009-08-08 | 1 | -0/+21 |
| * | Unbreak the stuff | Anton Korobeynikov | 2009-08-07 | 1 | -2/+2 |
| * | 2 more vdup.32 cases | Anton Korobeynikov | 2009-08-07 | 1 | -0/+14 |
| * | Fix support to use NEON for single precision fp math. | Evan Cheng | 2009-08-07 | 1 | -41/+83 |
| * | Implement Neon VST[234] operations. | Bob Wilson | 2009-08-06 | 1 | -24/+67 |
| * | Add parameter to pattern classes to enable an itinerary to be specified for i... | David Goodwin | 2009-08-06 | 1 | -68/+107 |
| * | Neon does not actually have VLD{234}.64 instructions. | Bob Wilson | 2009-08-06 | 1 | -3/+0 |
| * | When using NEON for single-precision FP, the NEON result must be placed in D0... | David Goodwin | 2009-08-05 | 1 | -14/+20 |
| * | 80 col violations. | Evan Cheng | 2009-08-05 | 1 | -9/+15 |
| * | Change DAG nodes for Neon VLD2/3/4 operations to return multiple results. | Bob Wilson | 2009-08-05 | 1 | -7/+43 |
| * | Replace dregsingle operand modifier with explicit escaped curly brackets. | Bob Wilson | 2009-08-04 | 1 | -2/+2 |
| * | Add NEON single-precision FP support for fabs and fneg. | David Goodwin | 2009-08-04 | 1 | -0/+8 |
| * | Match common pattern for FNMAC. Add NEON SP support. | David Goodwin | 2009-08-04 | 1 | -0/+1 |
| * | Initial support for single-precision FP using NEON. Added "neonfp" attribute ... | David Goodwin | 2009-08-04 | 1 | -0/+28 |
