Cong Hou 6fe6cafdd5 [X86][SSE] Detect AVG pattern during instruction combine for SSE2/AVX2/AVX512BW. ... This patch detects the AVG pattern in vectorized code, which is simply c = (a + b + 1) / 2, where a, b, and c have the same type which are vectors of either unsigned i8 or unsigned i16. In the IR, i8/i16 will be promoted to i32 before any arithmetic operations. The following IR shows such an example: %1 = zext <N x i8> %a to <N x i32> %2 = zext <N x i8> %b to <N x i32> %3 = add nuw nsw <N x i32> %1, <i32 1 x N> %4 = add nuw nsw <N x i32> %3, %2 %5 = lshr <N x i32> %N, <i32 1 x N> %6 = trunc <N x i32> %5 to <N x i8> and with this patch it will be converted to a X86ISD::AVG instruction. The pattern recognition is done when combining instructions just before type legalization during instruction selection. We do it here because after type legalization, it is much more difficult to do pattern recognition based on many instructions that are doing type conversions. Therefore, for target-specific instructions (like X86ISD::AVG), we need to take care of type legalization by ourselves. However, as X86ISD::AVG behaves similarly to ISD::ADD, I am wondering if there is a way to legalize operands and result types of X86ISD::AVG together with ISD::ADD. It seems that the current design doesn't support this idea. Tests are added for SSE2, AVX2, and AVX512BW and both i8 and i16 types of variant vector sizes. Differential revision: http://reviews.llvm.org/D14761 llvm-svn: 253952		2015-11-24 05:44:19 +00:00
..
Analysis	Revert "Change memcpy/memset/memmove to have dest and source alignments."	2015-11-19 05:56:52 +00:00
Assembler	[IR] Add support for empty tokens	2015-11-11 21:57:16 +00:00
Bindings	Disable Go bindings test with MSan, it has tons of linker errors	2015-11-19 00:05:20 +00:00
Bitcode	Revert "Change memcpy/memset/memmove to have dest and source alignments."	2015-11-19 05:56:52 +00:00
BugPoint	[bugpoint] Add a named metadata (+their operands) reducer	2015-11-06 00:12:50 +00:00
CodeGen	[X86][SSE] Detect AVG pattern during instruction combine for SSE2/AVX2/AVX512BW.	2015-11-24 05:44:19 +00:00
DebugInfo	Removing specific target from the generic test	2015-11-19 12:24:47 +00:00
Examples
ExecutionEngine	[RuntimeDyld] Don't allocate unnecessary stub buffer space	2015-11-23 21:47:51 +00:00
Feature	[FunctionAttrs] Add handling for operand bundles	2015-11-07 01:56:00 +00:00
FileCheck
Instrumentation	Revert "Change memcpy/memset/memmove to have dest and source alignments."	2015-11-19 05:56:52 +00:00
Integer
JitListener	DI: Reverse direction of subprogram -> function edge.	2015-11-05 22:03:56 +00:00
LibDriver
Linker	Have a single way for creating unique value names.	2015-11-22 00:16:24 +00:00
LTO	[LTO] Add options to llvm-lto to select output format and dump merged module	2015-11-20 00:13:05 +00:00
MC	[mips] .ent and .end should also set the type and size of the symbol respectively.	2015-11-23 16:08:03 +00:00
Object	Revert "Change memcpy/memset/memmove to have dest and source alignments."	2015-11-19 05:56:52 +00:00
Other	Revert "Change memcpy/memset/memmove to have dest and source alignments."	2015-11-19 05:56:52 +00:00
SymbolRewriter
TableGen	Revert r253923.	2015-11-23 22:19:57 +00:00
tools	[PGO] Add --text option for llvm-profdata show|merge commands	2015-11-23 20:47:38 +00:00
Transforms	SamplePGO - Add coverage tracking for samples.	2015-11-23 20:12:21 +00:00
Unit
Verifier	Revert "Change memcpy/memset/memmove to have dest and source alignments."	2015-11-19 05:56:52 +00:00
YAMLParser
.clang-format
CMakeLists.txt	sancov test suite	2015-11-12 19:34:21 +00:00
lit.cfg	Disable Go bindings test with MSan, it has tons of linker errors	2015-11-19 00:05:20 +00:00
lit.site.cfg.in
Makefile
Makefile.tests
TestRunner.sh