| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Add the last part that is needed for vectorization of if-converted code. | Nadav Rotem | 2012-12-04 | 1 | -162/+251 |
| * | LoopVectorize.cpp: Suppress a warning. [-Wunused-variable] | NAKAMURA Takumi | 2012-12-04 | 1 | -2/+2 |
| * | Fix whitespace. | NAKAMURA Takumi | 2012-12-04 | 1 | -4/+3 |
| * | minor renaming, documentation and cleanups. | Nadav Rotem | 2012-12-03 | 1 | -25/+21 |
| * | IF-conversion: teach the cost-model how to grade if-converted loops. | Nadav Rotem | 2012-12-03 | 1 | -15/+26 |
| * | Now that we have a basic if-conversion infrastructure we can rename the | Nadav Rotem | 2012-12-03 | 1 | -18/+18 |
| * | Add initial support for IF-conversion. This patch implements the first 1/3, | Nadav Rotem | 2012-12-03 | 1 | -134/+253 |
| * | Use the new script to sort the includes of every file under lib. | Chandler Carruth | 2012-12-03 | 1 | -18/+19 |
| * | minor cleanups | Nadav Rotem | 2012-11-30 | 1 | -8/+3 |
| * | Remove the use of LPPassManager. We can remove LPM because we dont need to ru... | Nadav Rotem | 2012-11-30 | 1 | -10/+10 |
| * | When broadcasting invariant scalars into vectors, place the broadcast code in... | Nadav Rotem | 2012-11-29 | 1 | -11/+29 |
| * | Move the code that uses SCEVs prior to creating the new loops. | Nadav Rotem | 2012-11-26 | 1 | -32/+35 |
| * | Move the max vector width to a constant parameter. No functionality change. | Nadav Rotem | 2012-11-25 | 1 | -5/+7 |
| * | Fix the document style. | Nadav Rotem | 2012-11-25 | 1 | -3/+3 |
| * | Refactor the ptr runtime check generation code. No functionality change. | Nadav Rotem | 2012-11-25 | 1 | -50/+66 |
| * | Rename method. No functionality change. | Nadav Rotem | 2012-11-25 | 1 | -3/+3 |
| * | The induction-pointer work is inspired by a research paper. This commit adds ... | Nadav Rotem | 2012-11-25 | 1 | -0/+4 |
| * | Add support for pointer induction variables even when there is no integer ind... | Nadav Rotem | 2012-11-25 | 1 | -105/+182 |
| * | LoopVectorizer: Add initial support for pointer induction variables (for exam... | Nadav Rotem | 2012-11-17 | 1 | -33/+126 |
| * | LoopVectorize: Division reductions generate incorrect code. Remove the part o... | Nadav Rotem | 2012-11-16 | 1 | -2/+0 |
| * | use the getSplat API. Patch by Paul Redmond. | Nadav Rotem | 2012-11-14 | 1 | -7/+1 |
| * | Fix a comment typo and add comments. | Nadav Rotem | 2012-11-11 | 1 | -4/+5 |
| * | Add support for memory runtime check. When we can, we calculate array bounds. | Nadav Rotem | 2012-11-09 | 1 | -33/+191 |
| * | Fix sign compare warning. Patch by Mahesha HS. | Chandler Carruth | 2012-11-02 | 1 | -2/+2 |
| * | LoopVectorize: Preserve NSW, NUW and IsExact flags. | Nadav Rotem | 2012-10-31 | 1 | -1/+12 |
| * | Put the threshold magic number in a variable. | Nadav Rotem | 2012-10-31 | 1 | -1/+4 |
| * | Remove enum values since they are not used anymore. | Nadav Rotem | 2012-10-31 | 1 | -6/+6 |
| * | LoopVectorize: Do not vectorize loops with tiny constant trip counts. | Nadav Rotem | 2012-10-31 | 1 | -0/+8 |
| * | Add support for loops that don't start with Zero. | Nadav Rotem | 2012-10-31 | 1 | -12/+19 |
| * | Add documentation. | Nadav Rotem | 2012-10-30 | 1 | -0/+5 |
| * | LoopVectorize: Add support for write-only loops when the write destination is... | Nadav Rotem | 2012-10-30 | 1 | -0/+7 |
| * | LoopVectorize: Fix a bug in the initialization of reduction variables. AND ne... | Nadav Rotem | 2012-10-30 | 1 | -7/+21 |
| * | LoopVectorizer: change debug prints: Print the module identifier when decidin... | Nadav Rotem | 2012-10-30 | 1 | -4/+6 |
| * | LoopVectorize: Update and preserve the dominator tree info. | Nadav Rotem | 2012-10-29 | 1 | -9/+37 |
| * | 1. Fix a bug in getTypeConversion. When a *simple* type is split, we need to ... | Nadav Rotem | 2012-10-27 | 1 | -1/+1 |
| * | Refactor the VectorTargetTransformInfo interface. | Nadav Rotem | 2012-10-26 | 1 | -9/+52 |
| * | LoopVectorize: Teach the cost model to query scalar costs as scalar types and... | Nadav Rotem | 2012-10-25 | 1 | -41/+61 |
| * | Add support for additional reduction variables: AND, OR, XOR. | Nadav Rotem | 2012-10-25 | 1 | -7/+42 |
| * | Implement a basic cost model for vector and scalar instructions. | Nadav Rotem | 2012-10-24 | 1 | -17/+32 |
| * | whitespace | Nadav Rotem | 2012-10-24 | 1 | -3/+3 |
| * | LoopVectorizer: Add a basic cost model which uses the VTTI interface. | Nadav Rotem | 2012-10-24 | 1 | -30/+273 |
| * | Back out r166591, not sure why this made it through since I cancelled the com... | Micah Villmow | 2012-10-24 | 1 | -2/+2 |
| * | Delete a directory that wasn't supposed to be checked in yet. | Micah Villmow | 2012-10-24 | 1 | -2/+2 |
| * | Use the AliasAnalysis isIdentifiedObj because it also understands mallocs and... | Nadav Rotem | 2012-10-23 | 1 | -19/+2 |
| * | Don't crash if the load/store pointer is not a GEP. | Nadav Rotem | 2012-10-22 | 1 | -1/+1 |
| * | Rename a variable. | Nadav Rotem | 2012-10-22 | 1 | -13/+13 |
| * | Vectorizer: optimize the generation of selects. If the condition is uniform, ... | Nadav Rotem | 2012-10-22 | 1 | -6/+16 |
| * | Update the loop vectorizer docs. | Nadav Rotem | 2012-10-22 | 1 | -17/+38 |
| * | Avoid an extra hash lookup when inserting a value into the widen map. | Anders Carlsson | 2012-10-21 | 1 | -4/+4 |
| * | Simplify code. No functionality change. | Jakub Staszak | 2012-10-21 | 1 | -1/+1 |
