* implement slice vectorization. Because it uses unaligned

packet access, it is not certain that it will bring a performance
  improvement: benchmarking needed.
* improve logic choosing slice vectorization.
* fix typo in SSE packet math, causing crash in unaligned case.
* fix bug in Product, causing crash in unaligned case.
* add TEST_SSE3 CMake option.
This commit is contained in:
Benoit Jacob
2008-06-22 15:02:05 +00:00
parent 8cef541b5a
commit 8a967fb17c
5 changed files with 64 additions and 34 deletions

View File

@@ -22,6 +22,10 @@ IF(CMAKE_COMPILER_IS_GNUCXX)
SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -msse2")
MESSAGE("Enabling SSE2 in tests/examples")
ENDIF(TEST_SSE2)
IF(TEST_SSE3)
SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -msse3")
MESSAGE("Enabling SSE3 in tests/examples")
ENDIF(TEST_SSE3)
IF(TEST_ALTIVEC)
SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -maltivec -mabi=altivec")
MESSAGE("Enabling AltiVec in tests/examples")