| Commit message (Expand) | Author | Age | Files | Lines |
* | Proper encoding for VLDM and VSTM instructions. The register lists for these | Bill Wendling | 2010-11-17 | 1 | -0/+4 |
* | Use new neon_vector_type and neon_polyvector_type attributes for Neon vectors. | Bob Wilson | 2010-11-16 | 1 | -7/+11 |
* | Change Neon polynomial types to be signed to match GCC. | Bob Wilson | 2010-11-16 | 1 | -2/+2 |
* | Refactor to new GetNumElements function. | Bob Wilson | 2010-11-16 | 1 | -35/+38 |
* | Tidy up some things in <arm_neon.h>. | Bob Wilson | 2010-11-16 | 1 | -94/+33 |
* | Reapply "Stop using struct wrappers for Neon vector types in <arm_neon.h>." | Bob Wilson | 2010-11-16 | 1 | -65/+48 |
* | Revert "Stop using struct wrappers for Neon vector types in <arm_neon.h>." | Bob Wilson | 2010-11-16 | 1 | -48/+65 |
* | This is the first step in adding sane error handling support to LLVMSystem. | Michael J. Spencer | 2010-11-16 | 4 | -1192/+1 |
* | Stop using struct wrappers for Neon vector types in <arm_neon.h>. | Bob Wilson | 2010-11-16 | 1 | -65/+48 |
* | Encode the multi-load/store instructions with their respective modes ('ia', | Bill Wendling | 2010-11-16 | 1 | -5/+5 |
* | pull the code to get the operand value out of the loop. | Chris Lattner | 2010-11-15 | 1 | -48/+58 |
* | split the giant encoder loop into two new helper functions. | Chris Lattner | 2010-11-15 | 2 | -112/+129 |
* | reduce nesting and minor cleanups, no functionality change. | Chris Lattner | 2010-11-15 | 1 | -87/+88 |
* | add fields to the .td files unconditionally, simplifying tblgen a bit. | Chris Lattner | 2010-11-15 | 2 | -5/+4 |
* | FileCheck: Eliminate DOSish \r from input file. | NAKAMURA Takumi | 2010-11-14 | 1 | -0/+5 |
* | ARM fixup encoding for direct call instructions (BL). | Jim Grosbach | 2010-11-11 | 1 | -0/+2 |
* | Add support for specifying a PostEncoderMethod, which can perform post-proces... | Owen Anderson | 2010-11-11 | 1 | -0/+4 |
* | Rename AccessesArguments and AccessesArgumentsReadonly, and rewrite | Dan Gohman | 2010-11-10 | 1 | -2/+2 |
* | Translate IntrReadArgMem to AccessesArgumentsReadonly. | Dan Gohman | 2010-11-09 | 1 | -0/+2 |
* | pass literals like $$1 through to the asm matcher. This isn't right yet, but... | Chris Lattner | 2010-11-06 | 1 | -9/+10 |
* | add (and document) the ability for alias results to have | Chris Lattner | 2010-11-06 | 4 | -11/+75 |
* | generalize alias support to allow the result of an alias to | Chris Lattner | 2010-11-06 | 3 | -21/+71 |
* | fix a bug where we had an implicit assumption that the | Chris Lattner | 2010-11-06 | 3 | -2/+26 |
* | fix some bugs in the alias support, unblocking changing of "clr" aliases | Chris Lattner | 2010-11-06 | 2 | -18/+65 |
* | Reimplement BuildResultOperands to be in terms of the result instruction's | Chris Lattner | 2010-11-06 | 3 | -26/+20 |
* | implement more checking to reject things like: | Chris Lattner | 2010-11-06 | 2 | -34/+25 |
* | decode and validate instruction alias result definitions. | Chris Lattner | 2010-11-06 | 1 | -0/+33 |
* | simplify | Chris Lattner | 2010-11-06 | 1 | -6/+8 |
* | fix another fixme, replacing a string with a semantic pointer. | Chris Lattner | 2010-11-06 | 1 | -10/+9 |
* | disolve a hack, having CodeGenInstAlias decode the alias in the .td | Chris Lattner | 2010-11-06 | 3 | -13/+31 |
* | In the calling convention logic, ValVT is always a legal type, | Duncan Sands | 2010-11-04 | 1 | -2/+2 |
* | partition operand processing between aliases and instructions. | Chris Lattner | 2010-11-04 | 1 | -8/+65 |
* | pull name slicing out of BuildInstructionOperandReference so | Chris Lattner | 2010-11-04 | 1 | -16/+12 |
* | cleanups. | Chris Lattner | 2010-11-04 | 1 | -18/+22 |
* | replace SrcOpNum with SrcOpName, eliminating a numering dependency | Chris Lattner | 2010-11-04 | 1 | -26/+23 |
* | strength reduce some code, resolving a fixme. | Chris Lattner | 2010-11-04 | 1 | -66/+59 |
* | take a big step to making aliases more general and less of a hack: | Chris Lattner | 2010-11-04 | 1 | -112/+197 |
* | Support generating an MC'ized CodeEmitter directly. Maintain a reference to the | Jim Grosbach | 2010-11-03 | 1 | -5/+18 |
* | rename Operand -> AsmOperand for clarity. | Chris Lattner | 2010-11-03 | 1 | -23/+17 |
* | Inside the calling convention logic LocVT is always a simple | Duncan Sands | 2010-11-03 | 1 | -4/+4 |
* | Break ARM addrmode4 (load/store multiple base address) into its constituent | Jim Grosbach | 2010-11-03 | 1 | -2/+2 |
* | Completely reject instructions that have an operand in their | Chris Lattner | 2010-11-02 | 1 | -18/+12 |
* | make MatchableInfo::Validate reject instructions (like LDR_PRE in ARM) | Chris Lattner | 2010-11-02 | 1 | -6/+31 |
* | rewrite EmitConvertToMCInst to iterate over the MCInst operands, | Chris Lattner | 2010-11-02 | 2 | -99/+76 |
* | merge two large parallel loops in EmitConvertToMCInst, no change | Chris Lattner | 2010-11-02 | 1 | -64/+40 |
* | a bunch of random cleanup, move a helper to CGT where it belongs. | Chris Lattner | 2010-11-02 | 3 | -50/+52 |
* | Revert r114340 (improvements in Darwin function prologue/epilogue), as it broke | Jim Grosbach | 2010-11-02 | 1 | -0/+6 |
* | add and update comments. | Chris Lattner | 2010-11-02 | 1 | -2/+9 |
* | refactor/cleanup MatchableInfo by eliminating the Tokens array, | Chris Lattner | 2010-11-02 | 1 | -126/+127 |
* | rename operands -> asmoperands to be more descriptive. | Chris Lattner | 2010-11-02 | 1 | -28/+30 |