| 2015-02-26 |
Chandler Carruth | [x86] Restructure the comments and the conditions for...
|
commit | commitdiff | tree |
| 2015-02-26 |
Chandler Carruth | [x86] Re-order the combines of select in the X86 backend...
|
commit | commitdiff | tree |
| 2015-02-26 |
Chandler Carruth | [x86] Add an assert to catch if we ever try to blend...
|
commit | commitdiff | tree |
| 2015-02-26 |
Chandler Carruth | [x86] Sink the single-input v8i16 lowering code that...
|
commit | commitdiff | tree |
| 2015-02-26 |
Chandler Carruth | [x86] Remove a SimpleTy usage. No need for it here...
|
commit | commitdiff | tree |
| 2015-02-26 |
Chandler Carruth | [x86] Make the vector shuffle helpers order the SDLoc...
|
commit | commitdiff | tree |
| 2015-02-20 |
Chandler Carruth | [x86] Switching the shuffle equivalence test to a variadic...
|
commit | commitdiff | tree |
| 2015-02-20 |
Chandler Carruth | [x86] Remove the old vector shuffle lowering code and...
|
commit | commitdiff | tree |
| 2015-02-20 |
Chandler Carruth | [x86] Now that the new vector shuffle legality is enabled...
|
commit | commitdiff | tree |
| 2015-02-20 |
Chandler Carruth | [x86] Make the new vector shuffle legality test on...
|
commit | commitdiff | tree |
| 2015-02-20 |
Chandler Carruth | [x86] Clean up a couple of test cases with the new...
|
commit | commitdiff | tree |
| 2015-02-20 |
Chandler Carruth | Revert r229944: EH: Prune unreachable resume instructions...
|
commit | commitdiff | tree |
| 2015-02-19 |
Chandler Carruth | [x86] Delete still more piles of complex code now that...
|
commit | commitdiff | tree |
| 2015-02-19 |
Chandler Carruth | [x86] Teach the unpack lowering how to lower with an...
|
commit | commitdiff | tree |
| 2015-02-19 |
Chandler Carruth | [x86] Dramatically improve v8i16 shuffle lowering by...
|
commit | commitdiff | tree |
| 2015-02-19 |
Chandler Carruth | [x86] Remove the final fallback in the v8i16 lowering...
|
commit | commitdiff | tree |
| 2015-02-19 |
Chandler Carruth | [x86] Simplify the pre-SSSE3 v16i8 lowering significantly...
|
commit | commitdiff | tree |
| 2015-02-19 |
Chandler Carruth | [x86] Remove the insanely over-aggressive unpack lowering...
|
commit | commitdiff | tree |
| 2015-02-19 |
Chandler Carruth | [x86] The SELECT x86 DAG combine also does legalization...
|
commit | commitdiff | tree |
| 2015-02-19 |
Chandler Carruth | [x86] Add support for bit-wise blending and use it...
|
commit | commitdiff | tree |
| 2015-02-19 |
Chandler Carruth | [x86,sdag] Two interrelated changes to the x86 and...
|
commit | commitdiff | tree |
| 2015-02-18 |
Chandler Carruth | [x86] Merge checks for a recently added test case that...
|
commit | commitdiff | tree |
| 2015-02-18 |
Chandler Carruth | [x86] Tighten the assertions to document that canonicalizati...
|
commit | commitdiff | tree |
| 2015-02-18 |
Chandler Carruth | [x86] Switch an if which is trivially true to an assert...
|
commit | commitdiff | tree |
| 2015-02-18 |
Chandler Carruth | [x86] Remove some more 'bit' nomenclature from the...
|
commit | commitdiff | tree |
| 2015-02-18 |
Chandler Carruth | [x86] Fold together the two shift lowering strategies...
|
commit | commitdiff | tree |
| 2015-02-18 |
Chandler Carruth | [x86] Refactor the bit shift code the same as I just...
|
commit | commitdiff | tree |
| 2015-02-18 |
Chandler Carruth | [x86] Rewrite the byte shift detection to not use boolean...
|
commit | commitdiff | tree |
| 2015-02-18 |
Chandler Carruth | [shuffles] Tweak my shufflevector fuzz test generation...
|
commit | commitdiff | tree |
| 2015-02-17 |
Chandler Carruth | [x86] Teach the unpack lowering to try wider element...
|
commit | commitdiff | tree |
| 2015-02-16 |
Chandler Carruth | [x86] Add a generic unpack-targeted lowering technique...
|
commit | commitdiff | tree |
| 2015-02-16 |
Chandler Carruth | [x86] Switch this test to use checks generated by my...
|
commit | commitdiff | tree |
| 2015-02-16 |
Chandler Carruth | [x86] Add initial basic support for forming blends...
|
commit | commitdiff | tree |
| 2015-02-16 |
Chandler Carruth | [x86] Add some more test cases for i8 vector blends.
|
commit | commitdiff | tree |
| 2015-02-16 |
Chandler Carruth | [x86] Switch my usage of VariadicFunction to a "normal...
|
commit | commitdiff | tree |
| 2015-02-16 |
Chandler Carruth | Switch our index sequence away from template aliases...
|
commit | commitdiff | tree |
| 2015-02-16 |
Chandler Carruth | [x86] Teach the 128-bit vector shuffle lowering routines...
|
commit | commitdiff | tree |
| 2015-02-16 |
Chandler Carruth | [x86] Clean up a few test cases with the update script...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] Add the test case from PR22412, we now get this...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] Teach the decomposed shuffle/blend lowering to...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [SDAG] Teach the SelectionDAG to canonicalize vector...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] Teach the shuffle mask equivalence test to look...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] Tweak the ordering of unpack matching vs. element...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] Stop shuffling zero vectors. =]
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] Use a more helpful parenthesizing of these comparisons...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] When splitting 256-bit vectors into 128-bit vectors...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] Make computing the zeroable elements slightly...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] Update some tests with the latest version of...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] Add a slight variation on some of the other generic...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] Add a test case for PR22390 which was a dup of...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] Fix PR22377, a regression with the new vector...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] Switch a collection of tests explicitly to the...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] Remove the now-default-on flag for the new vector...
|
commit | commitdiff | tree |
| 2015-02-15 |
Chandler Carruth | [x86] Teach my test updating script about another quirk...
|
commit | commitdiff | tree |
| 2015-02-14 |
Chandler Carruth | [gold] Consolidate the gold plugin options and actually...
|
commit | commitdiff | tree |
| 2015-02-14 |
Chandler Carruth | Remove a variable only used in an assert and sink its...
|
commit | commitdiff | tree |
| 2015-02-14 |
Chandler Carruth | Back out two accidental changes that snuck in with...
|
commit | commitdiff | tree |
| 2015-02-14 |
Chandler Carruth | [lit] Make the gold plugin support testing work with...
|
commit | commitdiff | tree |
| 2015-02-14 |
Chandler Carruth | Revert r229224: Make the 'llvm-lit' utility defend...
|
commit | commitdiff | tree |
| 2015-02-14 |
Chandler Carruth | [lit] Make the 'llvm-lit' utility defend against a...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [PM] Fix a compile error I introduced in r229094 and...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [PM] Update the examples to reflect the removal of the
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [PM] Remove the old 'PassManager.h' header file at...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | Re-sort the #include lines in bindings and examples...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | Re-sort #include lines using my handy dandy ./utils...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | Revert a series of commits starting at r228886 which...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Concede defeat and disable the unroll analyzer...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Merge the simplification and DCE estimation...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Remove pointless dyn_cast<>s to Instruction...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Don't check the loop set for whether an instruction is
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Change the other worklist in the unroll analyzer...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Replace a boolean, for loop, condition, and...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Directly query for dead instructions.
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Replace a linear time check for no uses with...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Rather than an operand set, use a setvector...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Extract the code to enqueue operansd for the...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Use a small set to de-duplicate operands prior...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Make the unroll cost analysis terminate determinist...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Make range based for loops a bit more explicit...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [IC] Fix a bug with the instcombine canonicalizing...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Avoid the "Insn" abbreviation of Instruction...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Tidy up the integer we use to accumululate...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Don't use a map from pointer to bool. Use...
|
commit | commitdiff | tree |
| 2015-02-13 |
Chandler Carruth | [unroll] Update the new analysis logic from r228265...
|
commit | commitdiff | tree |
| 2015-02-12 |
Chandler Carruth | [slp] Fix a nasty bug in the SLP vectorizer that Joerg...
|
commit | commitdiff | tree |
| 2015-02-10 |
Chandler Carruth | Revert r228556: InstCombine: propagate nonNull through...
|
commit | commitdiff | tree |
| 2015-02-10 |
Chandler Carruth | [x86] Fix PR22524: the DAG combiner was incorrectly...
|
commit | commitdiff | tree |
| 2015-02-07 |
Chandler Carruth | Consistently use override rather than virtual.
|
commit | commitdiff | tree |
| 2015-02-04 |
Chandler Carruth | [x86] Give movss and movsd execution domains in the...
|
commit | commitdiff | tree |
| 2015-02-04 |
Chandler Carruth | [x86] Remove a low-value test that was just checking...
|
commit | commitdiff | tree |
| 2015-02-04 |
Chandler Carruth | [x86] Mechanically update a bunch of tests' check lines...
|
commit | commitdiff | tree |
| 2015-02-04 |
Chandler Carruth | [x86] Teach the test update script to strip trailing...
|
commit | commitdiff | tree |
| 2015-02-04 |
Chandler Carruth | [x86] Include the destination register in the check...
|
commit | commitdiff | tree |
| 2015-02-04 |
Chandler Carruth | [x86] Add some tests I missed in the prior commit to...
|
commit | commitdiff | tree |
| 2015-02-04 |
Chandler Carruth | [x86] Start to introduce bit-masking based blend lowering.
|
commit | commitdiff | tree |
| 2015-02-04 |
Chandler Carruth | [x86] Add missing patterns for andps, orps, xorps,...
|
commit | commitdiff | tree |
| 2015-02-04 |
Chandler Carruth | [x86] Add tests for blends-with-zero on 4-element vectors.
|
commit | commitdiff | tree |
| 2015-02-04 |
Chandler Carruth | [x86] Refresh the checks of a number of tests using
|
commit | commitdiff | tree |
| 2015-02-04 |
Chandler Carruth | [x86] Switch to using the long '--check-prefix' form...
|
commit | commitdiff | tree |
| 2015-02-04 |
Chandler Carruth | [x86] Port this test to use utils/update_llc_test_checks.py.
|
commit | commitdiff | tree |
| next |
projects / oota-llvm.git / search