| 2006-04-11 |
Evan Cheng | movnt* and maskmovdqu intrinsics
|
commit | commitdiff | tree |
| 2006-04-11 |
Evan Cheng | Only get Tmp2 for cases where number of operands is...
|
commit | commitdiff | tree |
| 2006-04-11 |
Evan Cheng | Added support for _mm_move_ss and _mm_move_sd.
|
commit | commitdiff | tree |
| 2006-04-10 |
Evan Cheng | Remove some bogus patterns; clean up.
|
commit | commitdiff | tree |
| 2006-04-10 |
Evan Cheng | Remove an entry that is now done.
|
commit | commitdiff | tree |
| 2006-04-10 |
Evan Cheng | Added some missing shuffle patterns.
|
commit | commitdiff | tree |
| 2006-04-10 |
Evan Cheng | Correct an entry
|
commit | commitdiff | tree |
| 2006-04-10 |
Evan Cheng | movups / movupd
|
commit | commitdiff | tree |
| 2006-04-10 |
Evan Cheng | __builtin_ia32_loadup{s|d}, __builtin_ia32_storeup...
|
commit | commitdiff | tree |
| 2006-04-10 |
Evan Cheng | Missing break
|
commit | commitdiff | tree |
| 2006-04-10 |
Evan Cheng | Add a vselect test case.
|
commit | commitdiff | tree |
| 2006-04-10 |
Evan Cheng | Conditional move of vector types.
|
commit | commitdiff | tree |
| 2006-04-10 |
Evan Cheng | New entries
|
commit | commitdiff | tree |
| 2006-04-10 |
Evan Cheng | Use movaps to do VR128 reg-to-reg copies for now. It...
|
commit | commitdiff | tree |
| 2006-04-08 |
Evan Cheng | INSERT_VECTOR_ELT lowering bug:
|
commit | commitdiff | tree |
| 2006-04-08 |
Evan Cheng | ldmxcsr and stmxcsr.
|
commit | commitdiff | tree |
| 2006-04-08 |
Evan Cheng | Fix int_x86_sse_stmxcsr and int_x86_sse_ldmxcsr. Not...
|
commit | commitdiff | tree |
| 2006-04-07 |
Evan Cheng | Code clean up.
|
commit | commitdiff | tree |
| 2006-04-07 |
Evan Cheng | Doh!
|
commit | commitdiff | tree |
| 2006-04-07 |
Evan Cheng | Added patterns for MOVHPSmr and MOVLPSmr.
|
commit | commitdiff | tree |
| 2006-04-07 |
Evan Cheng | Keep track of an Mac OS X / x86 ABI bug.
|
commit | commitdiff | tree |
| 2006-04-07 |
Evan Cheng | Remove int_x86_sse_storeh_ps and int_x86_sse_storel_ps...
|
commit | commitdiff | tree |
| 2006-04-07 |
Evan Cheng | A saner workaround. I hope.
|
commit | commitdiff | tree |
| 2006-04-07 |
Evan Cheng | Temporary workaround for a Mac OSX specific issue.
|
commit | commitdiff | tree |
| 2006-04-07 |
Evan Cheng | Added more shuffle tests
|
commit | commitdiff | tree |
| 2006-04-07 |
Evan Cheng | Add code to RemapOperand() to handle Instruction::ExtractEle...
|
commit | commitdiff | tree |
| 2006-04-06 |
Evan Cheng | A MOVPS2SSmr, i.e. _mm_store_ss, encoding bug.
|
commit | commitdiff | tree |
| 2006-04-06 |
Evan Cheng | Remove int_x86_sse_loadh_ps and int_x86_sse_loadl_ps...
|
commit | commitdiff | tree |
| 2006-04-06 |
Evan Cheng | - movlp{s|d} and movhp{s|d} support.
|
commit | commitdiff | tree |
| 2006-04-06 |
Evan Cheng | New entries.
|
commit | commitdiff | tree |
| 2006-04-06 |
Evan Cheng | 1. If both vector operands of a vector_shuffle are...
|
commit | commitdiff | tree |
| 2006-04-06 |
Evan Cheng | POR encoded as PAND, yikes.
|
commit | commitdiff | tree |
| 2006-04-05 |
Evan Cheng | An entry about comi / ucomi intrinsics.
|
commit | commitdiff | tree |
| 2006-04-05 |
Evan Cheng | Support for comi / ucomi intrinsics.
|
commit | commitdiff | tree |
| 2006-04-05 |
Evan Cheng | Added comi and ucomi SSE intrinsics.
|
commit | commitdiff | tree |
| 2006-04-05 |
Evan Cheng | Handle canonical form of e.g.
|
commit | commitdiff | tree |
| 2006-04-05 |
Evan Cheng | Bogus assert
|
commit | commitdiff | tree |
| 2006-04-05 |
Evan Cheng | Fallthrough to expand if a VECTOR_SHUFFLE cannot be...
|
commit | commitdiff | tree |
| 2006-04-05 |
Evan Cheng | Exapnd a VECTOR_SHUFFLE to a BUILD_VECTOR if target...
|
commit | commitdiff | tree |
| 2006-04-05 |
Evan Cheng | Separate out to 2 test cases
|
commit | commitdiff | tree |
| 2006-04-05 |
Evan Cheng | Handle v8i16 shuffle that must be broken into a pair...
|
commit | commitdiff | tree |
| 2006-04-05 |
Evan Cheng | Add a new shuffle test case that requires pshuflw ...
|
commit | commitdiff | tree |
| 2006-04-04 |
Evan Cheng | Added pslldq and psrldq.
|
commit | commitdiff | tree |
| 2006-04-04 |
Evan Cheng | Added intrinsics to match __builtin_ia32_pslldqi128 and
|
commit | commitdiff | tree |
| 2006-04-04 |
Evan Cheng | Minor fixes + naming changes.
|
commit | commitdiff | tree |
| 2006-04-04 |
Evan Cheng | PSHUF* encoding bugs.
|
commit | commitdiff | tree |
| 2006-04-04 |
Evan Cheng | cmpps / cmppd encoding bug
|
commit | commitdiff | tree |
| 2006-04-04 |
Evan Cheng | Compact some intrinsic definitions.
|
commit | commitdiff | tree |
| 2006-04-03 |
Evan Cheng | Some SSE1 intrinsics: min, max, sqrt, etc.
|
commit | commitdiff | tree |
| 2006-04-03 |
Evan Cheng | Use movlpd to: store lower f64 extracted from v2f64.
|
commit | commitdiff | tree |
| 2006-04-03 |
Evan Cheng | - More efficient extract_vector_elt with shuffle and...
|
commit | commitdiff | tree |
| 2006-03-31 |
Evan Cheng | An entry about packed type alignments.
|
commit | commitdiff | tree |
| 2006-03-31 |
Evan Cheng | TargetData.cpp::getTypeInfo() was returning alignment...
|
commit | commitdiff | tree |
| 2006-03-31 |
Evan Cheng | Use a X86 target specific node X86ISD::PINSRW instead...
|
commit | commitdiff | tree |
| 2006-03-31 |
Evan Cheng | Added support for SSE3 horizontal ops: haddp{s|d} and...
|
commit | commitdiff | tree |
| 2006-03-31 |
Evan Cheng | Added haddp{s|d} and hsubp{s|d} intrinsics.
|
commit | commitdiff | tree |
| 2006-03-31 |
Evan Cheng | Add support to use pextrw and pinsrw to extract and...
|
commit | commitdiff | tree |
| 2006-03-31 |
Evan Cheng | Add vector_extract and vector_insert nodes.
|
commit | commitdiff | tree |
| 2006-03-31 |
Evan Cheng | Expand all INSERT_VECTOR_ELT (obviously bad) for now.
|
commit | commitdiff | tree |
| 2006-03-31 |
Evan Cheng | Expand INSERT_VECTOR_ELT to store vec, sp; store elt...
|
commit | commitdiff | tree |
| 2006-03-31 |
Evan Cheng | Typo
|
commit | commitdiff | tree |
| 2006-03-31 |
Evan Cheng | Ok for vector_shuffle mask to contain undef elements.
|
commit | commitdiff | tree |
| 2006-03-30 |
Evan Cheng | Make sure all possible shuffles are matched.
|
commit | commitdiff | tree |
| 2006-03-30 |
Evan Cheng | More logical ops patterns
|
commit | commitdiff | tree |
| 2006-03-30 |
Evan Cheng | Add support for _mm_cmp{cc}_ss and _mm_cmp{cc}_ps intrinsics
|
commit | commitdiff | tree |
| 2006-03-30 |
Evan Cheng | Add 128-bit pmovmskb intrinsic support.
|
commit | commitdiff | tree |
| 2006-03-29 |
Evan Cheng | Change SSE pack operation definitions to fit what the...
|
commit | commitdiff | tree |
| 2006-03-29 |
Evan Cheng | Add SSE2 integer pack with saturation intrinsics.
|
commit | commitdiff | tree |
| 2006-03-29 |
Evan Cheng | - Added some SSE2 128-bit packed integer ops.
|
commit | commitdiff | tree |
| 2006-03-29 |
Evan Cheng | Add a pshufhw test case.
|
commit | commitdiff | tree |
| 2006-03-29 |
Evan Cheng | Need to special case splat after all. Make the second...
|
commit | commitdiff | tree |
| 2006-03-29 |
Evan Cheng | Use unpcklpd for v2f64 splat.
|
commit | commitdiff | tree |
| 2006-03-29 |
Evan Cheng | Floating point logical operation patterns should match...
|
commit | commitdiff | tree |
| 2006-03-29 |
Evan Cheng | Add more SSE intrinsics
|
commit | commitdiff | tree |
| 2006-03-29 |
Evan Cheng | - More shuffle related bug fixes.
|
commit | commitdiff | tree |
| 2006-03-29 |
Evan Cheng | Another entry about shuffles.
|
commit | commitdiff | tree |
| 2006-03-29 |
Evan Cheng | - Only use pshufd for v4i32 vector shuffles.
|
commit | commitdiff | tree |
| 2006-03-28 |
Evan Cheng | Added aliases to scalar SSE instructions, e.g. addss...
|
commit | commitdiff | tree |
| 2006-03-28 |
Evan Cheng | Fixing buggy code.
|
commit | commitdiff | tree |
| 2006-03-28 |
Evan Cheng | Don't sort the names before outputing the intrinsic...
|
commit | commitdiff | tree |
| 2006-03-28 |
Evan Cheng | Missed X86::isUNPCKHMask
|
commit | commitdiff | tree |
| 2006-03-28 |
Evan Cheng | movlps and movlpd should be modeled as two address...
|
commit | commitdiff | tree |
| 2006-03-28 |
Evan Cheng | Update
|
commit | commitdiff | tree |
| 2006-03-28 |
Evan Cheng | Typo
|
commit | commitdiff | tree |
| 2006-03-28 |
Evan Cheng | * Prefer using operation of matching types. e.g unpcklpd...
|
commit | commitdiff | tree |
| 2006-03-28 |
Evan Cheng | Use movhpd is even better than movlhps.
|
commit | commitdiff | tree |
| 2006-03-28 |
Evan Cheng | Added a couple of entries about movhps and movlhps.
|
commit | commitdiff | tree |
| 2006-03-28 |
Evan Cheng | All unpack cases are now being handled.
|
commit | commitdiff | tree |
| 2006-03-28 |
Evan Cheng | - Clean up / consoladate various shuffle masks.
|
commit | commitdiff | tree |
| 2006-03-28 |
Evan Cheng | getVectorTyppe(MVT::i64, 2) ==> MVT::v2i64.
|
commit | commitdiff | tree |
| 2006-03-28 |
Evan Cheng | Model unpack lower and interleave as vector_shuffle...
|
commit | commitdiff | tree |
| 2006-03-27 |
Evan Cheng | Intrinsics naming convention change.
|
commit | commitdiff | tree |
| 2006-03-27 |
Evan Cheng | Try again
|
commit | commitdiff | tree |
| 2006-03-27 |
Evan Cheng | Incorrect check for FP all one's
|
commit | commitdiff | tree |
| 2006-03-27 |
Evan Cheng | Use pcmpeq to generate vector of all ones.
|
commit | commitdiff | tree |
| 2006-03-27 |
Evan Cheng | Changed isBuildVectorAllOnesInteger to isBuildVectorAllOnes.
|
commit | commitdiff | tree |
| 2006-03-27 |
Evan Cheng | Change isBuildVectorAllOnesInteger to isBuildVectorAllOnes...
|
commit | commitdiff | tree |
| 2006-03-26 |
Evan Cheng | Remove X86:isZeroVector, use ISD::isBuildVectorAllZeros...
|
commit | commitdiff | tree |
| 2006-03-26 |
Evan Cheng | Remove PPC:isZeroVector, use ISD::isBuildVectorAllZeros...
|
commit | commitdiff | tree |
| 2006-03-26 |
Evan Cheng | Add immAllZerosV helper
|
commit | commitdiff | tree |
| next |
projects / oota-llvm.git / search