Programming Languages Research Group: Git

2014-10-01	Chandler Carruth	[x86] Teach both sext and zext vector tests to cover...	commit \| commitdiff \| tree
2014-10-01	Chandler Carruth	[x86] Sort the ISA-specific RUN lines for vector-sext...	commit \| commitdiff \| tree
2014-10-01	Chandler Carruth	[x86] Rename avx-{s,z}ext.ll to vector-{s,z}ext.ll.	commit \| commitdiff \| tree
2014-10-01	Chandler Carruth	[x86] Cleanup and re-generate the checks for avx-zext...	commit \| commitdiff \| tree
2014-10-01	Chandler Carruth	[x86] Generate the FileCheck assertions for avx-blend...	commit \| commitdiff \| tree
2014-10-01	Chandler Carruth	[x86] Clean up and generate detailed FileCheck assertions for	commit \| commitdiff \| tree
2014-10-01	Chandler Carruth	[x86] Fix a few more tiny patterns with the new vector...	commit \| commitdiff \| tree
2014-10-01	Chandler Carruth	[x86] Delete some extraneous logic from the new vector...	commit \| commitdiff \| tree
2014-10-01	Chandler Carruth	[x86] Teach the new vector shuffle lowering to be even...	commit \| commitdiff \| tree
2014-10-01	Chandler Carruth	[x86] Hoist the zext-lowering up in the v4i32 lowering...	commit \| commitdiff \| tree
2014-10-01	Chandler Carruth	[x86] Teach the new vector shuffle lowering about VBROADCAST and	commit \| commitdiff \| tree
2014-09-30	Chandler Carruth	[x86] Add AVX1 and AVX2 testing to all of the 128-bit...	commit \| commitdiff \| tree
2014-09-30	Chandler Carruth	[x86] Update the exact FileCheck syntax of the 256...	commit \| commitdiff \| tree
2014-09-30	Chandler Carruth	[x86] Rework all of the 128-bit vector shuffle tests...	commit \| commitdiff \| tree
2014-09-30	Chandler Carruth	[x86] Revert r218588, r218589, and r218600. These patches...	commit \| commitdiff \| tree
2014-09-30	Chandler Carruth	[x86] Add some vector-register broadcast operations...	commit \| commitdiff \| tree
2014-09-29	Chandler Carruth	[x86] Make the new vector shuffle lowering lower blends...	commit \| commitdiff \| tree
2014-09-29	Chandler Carruth	[x86] Delete a bunch of really bad and totally unnecessary...	commit \| commitdiff \| tree
2014-09-29	Chandler Carruth	[x86] Refactor all of the VSELECT-as-blend lowering...	commit \| commitdiff \| tree
2014-09-29	Chandler Carruth	[x86] Improve naming and comments for VSELECT lowering.	commit \| commitdiff \| tree
2014-09-29	Chandler Carruth	[x86] Add the dispatch skeleton to the new vector shuffle...	commit \| commitdiff \| tree
2014-09-29	Chandler Carruth	[x86] Make the split-and-lower routine fully generic...	commit \| commitdiff \| tree
2014-09-28	Chandler Carruth	[x86] Teach the new vector shuffle lowering to fall...	commit \| commitdiff \| tree
2014-09-28	Chandler Carruth	[x86] Fix the new vector shuffle lowering's use of...	commit \| commitdiff \| tree
2014-09-28	Chandler Carruth	[x86] Fix a really silly bug that I introduced fixing...	commit \| commitdiff \| tree
2014-09-28	Chandler Carruth	[x86] Fix yet another bug in the new vector shuffle...	commit \| commitdiff \| tree
2014-09-27	Chandler Carruth	[x86] Fix yet another issue with widening vector shuffle...	commit \| commitdiff \| tree
2014-09-27	Chandler Carruth	[x86] Fix terrible bugs everywhere in the new vector...	commit \| commitdiff \| tree
2014-09-27	Chandler Carruth	[x86] Flip the sentinel values used in the target shuffle...	commit \| commitdiff \| tree
2014-09-26	Chandler Carruth	[x86] Fix a moderately terrifying bug in the new 128...	commit \| commitdiff \| tree
2014-09-26	Chandler Carruth	[x86] The mnemonic is SHUFPS not SHUPFS. =[ I'm very...	commit \| commitdiff \| tree
2014-09-26	Chandler Carruth	[x86] In the new vector shuffle lowering, when trying...	commit \| commitdiff \| tree
2014-09-26	Chandler Carruth	[x86] Fix a large collection of bugs that crept in...	commit \| commitdiff \| tree
2014-09-25	Chandler Carruth	[x86] Teach the new vector shuffle lowering to use...	commit \| commitdiff \| tree
2014-09-25	Chandler Carruth	[x86] Teach the new vector shuffle lowering a fancier...	commit \| commitdiff \| tree
2014-09-25	Chandler Carruth	[x86] Fix an oversight in the v8i32 path of the new...	commit \| commitdiff \| tree
2014-09-25	Chandler Carruth	[x86] Rearrange the code for v16i16 lowering a bit...	commit \| commitdiff \| tree
2014-09-25	Chandler Carruth	[x86] Implement AVX2 support for v32i8 in the new vector...	commit \| commitdiff \| tree
2014-09-25	Chandler Carruth	[x86] More tweaks to the v32i8 test cases.	commit \| commitdiff \| tree
2014-09-25	Chandler Carruth	[x86] Re-work a bunch of the v32i8 test cases to actually...	commit \| commitdiff \| tree
2014-09-25	Chandler Carruth	[x86] Remove the defunct X86ISD::BLENDV entry -- we...	commit \| commitdiff \| tree
2014-09-25	Chandler Carruth	[x86] Fix the v16i16 blend logic I added in the prior...	commit \| commitdiff \| tree
2014-09-25	Chandler Carruth	[x86] Implement v16i16 support with AVX2 in the new...	commit \| commitdiff \| tree
2014-09-24	Chandler Carruth	[x86] Factor out the logic to generically decombose...	commit \| commitdiff \| tree
2014-09-24	Chandler Carruth	[x86] Teach the instruction lowering to add comments...	commit \| commitdiff \| tree
2014-09-24	Chandler Carruth	[x86] More refactoring of the shuffle comment emission...	commit \| commitdiff \| tree
2014-09-24	Chandler Carruth	[x86] Bypass the shuffle mask comment generation when...	commit \| commitdiff \| tree
2014-09-24	Chandler Carruth	[x86] Hoist the logic for extracting the relevant bits...	commit \| commitdiff \| tree
2014-09-24	Chandler Carruth	[x86] Start refactoring the comment printing logic...	commit \| commitdiff \| tree
2014-09-24	Chandler Carruth	[x86] Teach the new vector shuffle lowering to lower...	commit \| commitdiff \| tree
2014-09-24	Chandler Carruth	[x86] Fix a really terrible bug in the repeated 128...	commit \| commitdiff \| tree
2014-09-23	Chandler Carruth	[x86] Teach the new vector shuffle lowering to lower...	commit \| commitdiff \| tree
2014-09-23	Chandler Carruth	[x86] Teach the rest of the 'target shuffle' machinery...	commit \| commitdiff \| tree
2014-09-23	Chandler Carruth	[x86] Teach the new shuffle lowering's blend functionality...	commit \| commitdiff \| tree
2014-09-23	Chandler Carruth	[x86] Teach the vector comment parsing and printing...	commit \| commitdiff \| tree
2014-09-23	Chandler Carruth	[x86] Teach the AVX1 path of the new vector shuffle...	commit \| commitdiff \| tree
2014-09-22	Chandler Carruth	[x86] Rename X86ISD::VPERMILP to X86ISD::VPERMILPI...	commit \| commitdiff \| tree
2014-09-22	Chandler Carruth	[x86] Stub out the integer lowering of 256-bit vectors...	commit \| commitdiff \| tree
2014-09-22	Chandler Carruth	[x86] Introduce tests covering the gamut of 256-bit...	commit \| commitdiff \| tree
2014-09-22	Chandler Carruth	[x86] Move the AVX v4i64 test cases down to group them...	commit \| commitdiff \| tree
2014-09-22	Chandler Carruth	[x86] Back out a bad choice about lowering v4i64 and...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Teach the new vector shuffle lowering how to...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Add a bunch of test cases where we have different...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] With the stronger canonicalization of shuffles...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Teach the new vector shuffle lowering to re-use...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Refactor the logic to form SHUFPS instruction...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Teach the new vector shuffle lowering the basics...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Teach the new vector shuffle lowering how to...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Add test cases for UNPCK instructions with v8f32...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Begin teaching the new vector shuffle lowering...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Regenerate this test case now that I've improved...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Explicitly lower to a blend early if it is trivial...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Teach the new vector shuffle lowering of v4f64...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Add some more comprehensive tests for v4f64 blending.	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Re-generate a bunch of the v4f64 test cases with...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Switch the blend implementation to use a MVT...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Remove some essentially lying comments from the...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Fix a helper to reflect that what we actually...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Teach the new vector shuffle lowering the first...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Add some more test cases covering specific blend...	commit \| commitdiff \| tree
2014-09-21	Chandler Carruth	[x86] Add the beginnings of some tests for our v8f32...	commit \| commitdiff \| tree
2014-09-20	Chandler Carruth	[x86] Teach the new vector shuffle lowering to use...	commit \| commitdiff \| tree
2014-09-20	Chandler Carruth	[x86] Add an AVX run to the 128-bit v2 tests, teach...	commit \| commitdiff \| tree
2014-09-20	Chandler Carruth	[x86] Teach the new vector shuffle lowering to use...	commit \| commitdiff \| tree
2014-09-20	Chandler Carruth	[x86] Start moving to a fancier check syntax to reduce...	commit \| commitdiff \| tree
2014-09-20	Chandler Carruth	[x86] Teach the v4f32 path of the new shuffle lowering...	commit \| commitdiff \| tree
2014-09-20	Chandler Carruth	[x86] Refactor the code for emitting INSERTPS to reuse...	commit \| commitdiff \| tree
2014-09-20	Chandler Carruth	[x86] Generalize the single-element insertion lowering...	commit \| commitdiff \| tree
2014-09-20	Chandler Carruth	[x86] Replace some duplicated logic reasoning about...	commit \| commitdiff \| tree
2014-09-19	Chandler Carruth	[x86] Hoist a function up to the rest of the non-type...	commit \| commitdiff \| tree
2014-09-19	Chandler Carruth	[x86] Hoist the actual lowering logic into a helper...	commit \| commitdiff \| tree
2014-09-19	Chandler Carruth	[x86] Fully generalize the zext lowering in the new...	commit \| commitdiff \| tree
2014-09-19	Chandler Carruth	[x86] Recognize that we can use duplication to widen...	commit \| commitdiff \| tree
2014-09-19	Chandler Carruth	[x86] Actually test the SSE2 lowering for most of the...	commit \| commitdiff \| tree
2014-09-19	Chandler Carruth	[x86] Teach the new vector shuffle lowering to also...	commit \| commitdiff \| tree
2014-09-19	Chandler Carruth	[x86] Add a dedicated lowering path for zext-compatible...	commit \| commitdiff \| tree
2014-09-19	Chandler Carruth	[x86] Extend this test to cover SSE4.1. Nothing interesting...	commit \| commitdiff \| tree
2014-09-18	Chandler Carruth	[x86] Use PALIGNR for v4i32 and v2i64 blends when appropriate.	commit \| commitdiff \| tree
2014-09-18	Chandler Carruth	[x86] Add an SSSE3 run and check mode to the 128-bit...	commit \| commitdiff \| tree
2014-09-18	Chandler Carruth	[x86] Add an SSSE3 run to the v4 shuffle test.	commit \| commitdiff \| tree
next

C/C++ LLVM-based compilers that forbids OOTA behaviors

RSS Atom