I am not sure I am posting this message in the appropriate group.
Anyway, here is the problem I am facing. I am trying to use the Intel
SIMD-SEE3 instructions in an __asm block to accelerate some signal
processing code and I just can't make it work. Here is an example that
generates the problem:
movsldup xmm0, xmm1
... and what I get in the assembly output is this:
59: movsldup xmm0, xmm1
00401191 movhlps xmm0,xmm1
.... which, of course, does not work.
I only have this problem with SSE3 instructions. Any clue what the problem
I am using the Intel 8.0 Compiler under Visual C++ 6.0.