optimization/8049: SSE unaligned vector stores crash with -O0
Ian Ollmann
iano@cco.caltech.edu
Thu Oct 10 17:56:00 GMT 2002
I think the problem is that the compiler is not properly padding data.
_mm_store_ps is nearly unique in that its argument order is
_mm_store_ps( float*, vector );
The 0x4 offset is to protect the float* that is written to the stack a few
bytes earlier. The compiler needs to insert an extra 12 bytes of padding.
In fact, it I swap the order of the arguments as follows:
_mm_store_ps( vector, float*)
then the crash goes away.
Ian
---------------------------------------------------
Ian Ollmann, Ph.D. iano@cco.caltech.edu
---------------------------------------------------
More information about the Gcc-bugs
mailing list