Displaying 2 results from an estimated 2 matches for "saturate_float_to_16bit".
2011 Sep 01
0
[PATCH 5/5] resample: Add NEON optimized inner_product_single for floating point
...s when len % 4 == 0 */
static inline int32_t inner_product_single(const int16_t *a, const int16_t *b, unsigned int len)
@@ -97,4 +121,81 @@ static inline int32_t inner_product_single(const int16_t *a, const int16_t *b, u
return ret;
}
+#elif defined(FLOATING_POINT)
+
+static inline int32_t saturate_float_to_16bit(float a) {
+ int32_t ret;
+ asm ("vmov.f32 d0[0], %[a]\n"
+ "vcvt.s32.f32 d0, d0, #15\n"
+ "vqrshrn.s32 d0, q0, #15\n"
+ "vmov.s16 %[ret], d0[0]\n"
+ : [ret] "=&r" (ret)
+ : [a] "r" (a)
+...
2011 Sep 01
6
[PATCH 0/5] ARM NEON optimization for samplerate converter
From: Jyri Sarha <jsarha at ti.com>
I optimized Speex resampler for NEON capable ARM CPUs. The first patch
should speed up resampling on any platform that can spare the
increased memory usage. It would be nice to have these merged to the
master branch. Please let me know if there is anything I can do to
help the the merge. The patches have been rebased on top of master
branch in