Programming Languages Research Group: Git

2013-04-08	Nadav Rotem	c++ new operators are not malloc-like functions because...	commit \| commitdiff \| tree
2013-04-08	Nadav Rotem	Update the docs about the fact that the loop vectorizer...	commit \| commitdiff \| tree
2013-04-06	Nadav Rotem	typo	commit \| commitdiff \| tree
2013-04-04	Nadav Rotem	Document the return value of SmallSet insert.	commit \| commitdiff \| tree
2013-04-01	Nadav Rotem	Add support for vector data types in the LLVM interpreter.	commit \| commitdiff \| tree
2013-03-29	Nadav Rotem	Fix a typo	commit \| commitdiff \| tree
2013-03-28	Nadav Rotem	Add the X86 FMAs to the scheduling model.	commit \| commitdiff \| tree
2013-03-28	Nadav Rotem	Add the Haswell machine model.	commit \| commitdiff \| tree
2013-03-28	Nadav Rotem	Remove the unused port from the SandyBridge machine...	commit \| commitdiff \| tree
2013-03-20	Nadav Rotem	When computing the demanded bits of Load SDNodes, make...	commit \| commitdiff \| tree
2013-03-19	Nadav Rotem	Optimize sext <4 x i8> and <4 x i16> to <4 x i64>.	commit \| commitdiff \| tree
2013-03-15	Nadav Rotem	Add a triple to the test.	commit \| commitdiff \| tree
2013-03-14	Nadav Rotem	Unaligned loads should use the VMOVUPS opcode.	commit \| commitdiff \| tree
2013-03-07	Nadav Rotem	No need to go through int64 and APInt when generating...	commit \| commitdiff \| tree
2013-03-07	Nadav Rotem	ArrayRef has a OneElt constructor. Beautify the code.	commit \| commitdiff \| tree
2013-03-07	Nadav Rotem	Switch from std::vector to ArrayRef. Speedup FoldBitCast...	commit \| commitdiff \| tree
2013-03-02	Nadav Rotem	PR14448 - prevent the loop vectorizer from vectorizing...	commit \| commitdiff \| tree
2013-02-27	Nadav Rotem	Silence the unused variable warning.	commit \| commitdiff \| tree
2013-02-27	Nadav Rotem	The FastISEL should be fast. But when we record statistics...	commit \| commitdiff \| tree
2013-02-27	Nadav Rotem	Revert r176166 because it broke one of the lit tests.	commit \| commitdiff \| tree
2013-02-27	Nadav Rotem	For each function that we optimize we initialize a...	commit \| commitdiff \| tree
2013-02-27	Nadav Rotem	std::string to StringRef.	commit \| commitdiff \| tree
2013-02-24	Nadav Rotem	Revert r169638 because it broke Mesa llvmpipe tests.	commit \| commitdiff \| tree
2013-02-22	Nadav Rotem	SelectionDAG compile time improvement.	commit \| commitdiff \| tree
2013-02-19	Nadav Rotem	Fix a bug in mayHaveSideEffects. Functions that do...	commit \| commitdiff \| tree
2013-02-19	Nadav Rotem	Fix a bug that was found by the clang static analyzer...	commit \| commitdiff \| tree
2013-02-14	Nadav Rotem	Dont merge consecutive loads/stores into vectors when...	commit \| commitdiff \| tree
2013-02-14	Nadav Rotem	80-col	commit \| commitdiff \| tree
2013-02-07	Nadav Rotem	fix 80-col violation and fix the docs.	commit \| commitdiff \| tree
2013-02-07	Nadav Rotem	Update Release notes regarding TTI.	commit \| commitdiff \| tree
2013-02-07	Nadav Rotem	Document the loop vectorizer changes.	commit \| commitdiff \| tree
2013-02-01	Nadav Rotem	Fix errant fallthrough in the generation of the lifetime...	commit \| commitdiff \| tree
2013-02-01	Nadav Rotem	Revert r174152. The shift amount may overflow and in...	commit \| commitdiff \| tree
2013-02-01	Nadav Rotem	Optimize shift lefts of a constant by a value plus...	commit \| commitdiff \| tree
2013-01-30	Nadav Rotem	InstCombine: canonicalize sext-and --> select	commit \| commitdiff \| tree
2013-01-28	Nadav Rotem	A bugfix for tblgen, in the function ‘emitSourceFileHeader’.	commit \| commitdiff \| tree
2013-01-25	Nadav Rotem	Fix a warning in the new DWARFheader. Add a new line...	commit \| commitdiff \| tree
2013-01-25	Nadav Rotem	LoopVectorize: Refactor the code that vectorizes loads...	commit \| commitdiff \| tree
2013-01-25	Nadav Rotem	LoopVectorizer: Refactor more code to use the IRBuilder.	commit \| commitdiff \| tree
2013-01-25	Nadav Rotem	Refactor some code to use the IRBuilder.	commit \| commitdiff \| tree
2013-01-23	Nadav Rotem	Add support for reverse pointer induction variables...	commit \| commitdiff \| tree
2013-01-21	Nadav Rotem	Fix a comment. Induction vars dont need to start at...	commit \| commitdiff \| tree
2013-01-20	Nadav Rotem	Fix a build error.	commit \| commitdiff \| tree
2013-01-20	Nadav Rotem	Revert 172708.	commit \| commitdiff \| tree
2013-01-20	Nadav Rotem	Update the gcc-loops benchmark	commit \| commitdiff \| tree
2013-01-20	Nadav Rotem	Update the linpack benchmark with different array sizes.	commit \| commitdiff \| tree
2013-01-20	Nadav Rotem	LoopVectorizer: Implement a new heuristics for selecting...	commit \| commitdiff \| tree
2013-01-20	Nadav Rotem	Change the cpu type in the test.	commit \| commitdiff \| tree
2013-01-19	Nadav Rotem	On Sandybridge split unaligned 256bit stores into two...	commit \| commitdiff \| tree
2013-01-18	Nadav Rotem	On Sandybridge loading unaligned 256bits using two...	commit \| commitdiff \| tree
2013-01-15	Nadav Rotem	Teach InstCombine to optimize extract of a value from...	commit \| commitdiff \| tree
2013-01-15	Nadav Rotem	LoopVectorizer cost model. Honor the user command line...	commit \| commitdiff \| tree
2013-01-15	Nadav Rotem	LTO: Also init TTI for codegen passes.	commit \| commitdiff \| tree
2013-01-13	Nadav Rotem	Fix PR14547. Handle induction variables of small sizes...	commit \| commitdiff \| tree
2013-01-11	Nadav Rotem	PPC: Implement efficient lowering of sign_extend_inreg.	commit \| commitdiff \| tree
2013-01-11	Nadav Rotem	ARM Cost Model: Modify the target independent cost...	commit \| commitdiff \| tree
2013-01-11	Nadav Rotem	ARM Cost Model: We need to detect the max bitwidth...	commit \| commitdiff \| tree
2013-01-10	Nadav Rotem	LoopVectorizer: Fix a bug in the vectorization of BinaryOper...	commit \| commitdiff \| tree
2013-01-09	Nadav Rotem	ARM Cost model: Use the size of vector registers and...	commit \| commitdiff \| tree
2013-01-09	Nadav Rotem	add -march to the test	commit \| commitdiff \| tree
2013-01-09	Nadav Rotem	Efficient lowering of vector sdiv when the divisor...	commit \| commitdiff \| tree
2013-01-09	Nadav Rotem	ARM Cost Model: Add a basic vectorization unrolling...	commit \| commitdiff \| tree
2013-01-09	Nadav Rotem	Remove the -licm pass from the loop vectorizer test...	commit \| commitdiff \| tree
2013-01-09	Nadav Rotem	Cost Model: Move the 'max unroll factor' variable to...	commit \| commitdiff \| tree
2013-01-08	Nadav Rotem	Docs: mention that we support float reductions when...	commit \| commitdiff \| tree
2013-01-08	Nadav Rotem	Code cleanup: refactor the switch statements in the...	commit \| commitdiff \| tree
2013-01-08	Nadav Rotem	Rename the enum members to match the LLVM coding style.	commit \| commitdiff \| tree
2013-01-07	Nadav Rotem	LoopVectorizer: Add support for floating point reductions	commit \| commitdiff \| tree
2013-01-07	Nadav Rotem	LoopVectorizer: When we vectorizer and widen loops...	commit \| commitdiff \| tree
2013-01-05	Nadav Rotem	Revert revision 171524. Original message:	commit \| commitdiff \| tree
2013-01-05	Nadav Rotem	Fix a typo. Remove the duplicated test.	commit \| commitdiff \| tree
2013-01-05	Nadav Rotem	iLoopVectorize: Non commutative operators can be used...	commit \| commitdiff \| tree
2013-01-05	Nadav Rotem	Force a fixed unroll count on the target independent...	commit \| commitdiff \| tree
2013-01-04	Nadav Rotem	Fix a warning	commit \| commitdiff \| tree
2013-01-04	Nadav Rotem	Update the gcc-loops benchmarks results with the new...	commit \| commitdiff \| tree
2013-01-04	Nadav Rotem	Add linpack-pc bench	commit \| commitdiff \| tree
2013-01-04	Nadav Rotem	Change the default number of registers to prevent unrolling...	commit \| commitdiff \| tree
2013-01-04	Nadav Rotem	Move the loop vectorizer from O2 to O3. It looks like...	commit \| commitdiff \| tree
2013-01-04	Nadav Rotem	Update the docs about the new unroll features.	commit \| commitdiff \| tree
2013-01-04	Nadav Rotem	LoopVectorizer:	commit \| commitdiff \| tree
2013-01-04	Nadav Rotem	Revert revision: 171467. This transformation is incorrect...	commit \| commitdiff \| tree
2013-01-03	Nadav Rotem	Reformat the rst text.	commit \| commitdiff \| tree
2013-01-03	Nadav Rotem	LoopVectorizer: Test the unrolling flag.	commit \| commitdiff \| tree
2013-01-03	Nadav Rotem	LoopVectorizer: Document the unrolling feature.	commit \| commitdiff \| tree
2013-01-03	Nadav Rotem	LoopVectorizer: Add support for loop-unrolling during...	commit \| commitdiff \| tree
2013-01-02	Nadav Rotem	Avoid vectorization when the function has the "noimplicitflo...	commit \| commitdiff \| tree
2013-01-02	Nadav Rotem	AVX: Fix a bug in WidenMaskArithmetic.	commit \| commitdiff \| tree
2013-01-02	Nadav Rotem	AVX: Fix a bug in WidenMaskArithmetic.	commit \| commitdiff \| tree
2013-01-01	Nadav Rotem	Make opt grab the triple from the module and use it...	commit \| commitdiff \| tree
2012-12-30	Nadav Rotem	LoopVectorizer: Fix a bug in the code that updates...	commit \| commitdiff \| tree
2012-12-28	Nadav Rotem	CostModel: initial checkin for code that estimates...	commit \| commitdiff \| tree
2012-12-28	Nadav Rotem	wrap 80-col lines.	commit \| commitdiff \| tree
2012-12-28	Nadav Rotem	AVX: Move the ZEXT/ANYEXT DAGCo optimizations to the...	commit \| commitdiff \| tree
2012-12-27	Nadav Rotem	Reverse the 'if' condition and reduce the indentation.	commit \| commitdiff \| tree
2012-12-27	Nadav Rotem	AVX/AVX2: Move the SEXT lowering code from a target...	commit \| commitdiff \| tree
2012-12-27	Nadav Rotem	DAGCombinerInformation: add a getter that exposes the...	commit \| commitdiff \| tree
2012-12-27	Nadav Rotem	docs: Update the benchmark with updated perf numbers.	commit \| commitdiff \| tree
2012-12-27	Nadav Rotem	On AVX/AVX2 the type v8i1 is legalized to v8i16, which...	commit \| commitdiff \| tree
2012-12-27	Nadav Rotem	AVX/AVX2: Move the code that lowers vector-trunc from...	commit \| commitdiff \| tree
2012-12-27	Nadav Rotem	Refactor DAGCombinerInfo. Change the different booleans...	commit \| commitdiff \| tree
next

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

RSS Atom