tags:

views:

162

answers:

1
+1  Q: 

SSE2: How to reduce a _m128 to a word

Hello

What's the best way ( sse2 ) to reduce a _m128 ( 4 words a b c d) to one word? I want the low part of each _m128 components:

int result = ( _m128.a & 0x000000ff ) <<  24
        | ( _m128.b & 0x000000ff ) << 16
        | ( _m128.c & 0x000000ff ) << 8
        | ( _m128.d & 0x000000ff ) << 0

Is there an intrinsics for that ? thanks !

+1  A: 

FYI, the sse3 intrinsics _mm_shuffle_epi8 do the job: (with the mask 0x0004080c in this case )

benoitj  2009-11-18 17:27:52
`_mm_shuffle_epi8` (`PSHUFB`) is **SSSE3** (aka **SSE3.5** aka **MNI**). Otherwise good answer: +1.
Paul R  2010-03-12 21:29:16

related questions

Benchmarking SSE instructions
Java performance in numerical algorithms
SSE register return with SSE disabled
Why is SSE scalar sqrt(x) slower than rsqrt(x) * x?
Whats a good place to start learning assembly?
What is the maximum theoretical speed-up due to SSE for a simple binary subtraction?
SIMD programming languages
Calling constructor from another class
Getting started with SSE
Resources for (Manual and Automatic) Loop Vectorization
How to determine SSE prefetch instruction size?
Get GCC to preserve an SSE register throughout a function that uses inline asm
What's the most efficient way to multiply 4 floats by 4 floats using SSE ?
How do modern compilers use mmx/3dnow/sse instructions?
glibc and SSE functionality
Using SSE instructions
developing for new instruction sets
Using SSE in c# is it possible?
SSE4 instructions in VS2005?
Best resource for learning about prefetching a buffer in C on Intel/AMD 64 bit
difference in speed between char and integer arrays?
How to get GCC to use more than two SIMD registers when using intrinsics?
How much speed-up from converting 3D maths to SSE or other SIMD?
What is the meaning of "non temporal" memory accesses in x86