volk: adding unaligned byteswap kernel.
volk: use loadu for unaligned volk_32f_x2_dot_prod_32f_u_sse*
volk: fix volk_32f_x2_dot_prod_32f_u_sse tail case
Merge branch 'maint'
volk: don't initialize phase in rotator
volk: replace (__m128) with volk cast for portability
volk: fixing some volk kernels.
This should fix some problems with gr-filter QA tests. Also removes some warnings.
volk: added missing avx header include
Merge branch 'gr_filter'
filter: adding ssc and fsf versions of filter with associated new Volk kernels.
These routines work and pass QA. They could use some performance work. the FSF is just slightly slower than before; the SCC version is more noticably slower.
Both could benefit, probably, by using SSE2 intrinsics to handle the shorts.
Also available in: