libFLAC/stream_encoder_intrin_*.c: Refactor abs calculation
Patch-from: lvqcl <lvqcl.mail@gmail.com>
diff --git a/src/libFLAC/stream_encoder_intrin_sse2.c b/src/libFLAC/stream_encoder_intrin_sse2.c
index be8395a..b60a3db 100644
--- a/src/libFLAC/stream_encoder_intrin_sse2.c
+++ b/src/libFLAC/stream_encoder_intrin_sse2.c
@@ -43,6 +43,17 @@
#include <stdlib.h> /* for abs() */
#include <emmintrin.h> /* SSE2 */
#include "FLAC/assert.h"
+#include "share/compat.h"
+
+FLAC__SSE_TARGET("sse2")
+static inline __m128i local_abs_epi32(__m128i val)
+{
+ __m128i mask = _mm_srai_epi32(val, 31);
+ val = _mm_xor_si128(val, mask);
+ val = _mm_sub_epi32(val, mask);
+ return val;
+}
+
FLAC__SSE_TARGET("sse2")
void FLAC__precompute_partition_info_sums_intrin_sse2(const FLAC__int32 residual[], FLAC__uint64 abs_residual_partition_sums[],
@@ -71,25 +82,19 @@
/* assumption: residual[] is properly aligned so (residual + e1) is properly aligned too and _mm_loadu_si128() is fast */
for( ; residual_sample < e1; residual_sample++) {
__m128i mm_res = _mm_cvtsi32_si128(residual[residual_sample]);
- __m128i mm_mask = _mm_srai_epi32(mm_res, 31);
- mm_res = _mm_xor_si128(mm_res, mm_mask);
- mm_res = _mm_sub_epi32(mm_res, mm_mask); /* abs(INT_MIN) is undefined, but if the residual is INT_MIN we have bigger problems */
+ mm_res = local_abs_epi32(mm_res);
mm_sum = _mm_add_epi32(mm_sum, mm_res);
}
for( ; residual_sample < e3; residual_sample+=4) {
__m128i mm_res = _mm_loadu_si128((const __m128i*)(residual+residual_sample));
- __m128i mm_mask = _mm_srai_epi32(mm_res, 31);
- mm_res = _mm_xor_si128(mm_res, mm_mask);
- mm_res = _mm_sub_epi32(mm_res, mm_mask);
+ mm_res = local_abs_epi32(mm_res);
mm_sum = _mm_add_epi32(mm_sum, mm_res);
}
for( ; residual_sample < end; residual_sample++) {
__m128i mm_res = _mm_cvtsi32_si128(residual[residual_sample]);
- __m128i mm_mask = _mm_srai_epi32(mm_res, 31);
- mm_res = _mm_xor_si128(mm_res, mm_mask);
- mm_res = _mm_sub_epi32(mm_res, mm_mask);
+ mm_res = local_abs_epi32(mm_res);
mm_sum = _mm_add_epi32(mm_sum, mm_res);
}
@@ -109,26 +114,20 @@
for( ; residual_sample < e1; residual_sample++) {
__m128i mm_res = _mm_cvtsi32_si128(residual[residual_sample]); /* 0 0 0 r0 */
- __m128i mm_mask = _mm_srai_epi32(mm_res, 31);
- mm_res = _mm_xor_si128(mm_res, mm_mask);
- mm_res = _mm_sub_epi32(mm_res, mm_mask); /* 0 0 0 |r0| == 00 |r0_64| */
+ mm_res = local_abs_epi32(mm_res); /* 0 0 0 |r0| == 00 |r0_64| */
mm_sum = _mm_add_epi64(mm_sum, mm_res);
}
for( ; residual_sample < e3; residual_sample+=2) {
__m128i mm_res = _mm_loadl_epi64((const __m128i*)(residual+residual_sample)); /* 0 0 r1 r0 */
- __m128i mm_mask = _mm_srai_epi32(mm_res, 31);
- mm_res = _mm_xor_si128(mm_res, mm_mask);
- mm_res = _mm_sub_epi32(mm_res, mm_mask); /* 0 0 |r1| |r0| */
+ mm_res = local_abs_epi32(mm_res); /* 0 0 |r1| |r0| */
mm_res = _mm_shuffle_epi32(mm_res, _MM_SHUFFLE(3,1,2,0)); /* 0 |r1| 0 |r0| == |r1_64| |r0_64| */
mm_sum = _mm_add_epi64(mm_sum, mm_res);
}
for( ; residual_sample < end; residual_sample++) {
__m128i mm_res = _mm_cvtsi32_si128(residual[residual_sample]);
- __m128i mm_mask = _mm_srai_epi32(mm_res, 31);
- mm_res = _mm_xor_si128(mm_res, mm_mask);
- mm_res = _mm_sub_epi32(mm_res, mm_mask);
+ mm_res = local_abs_epi32(mm_res);
mm_sum = _mm_add_epi64(mm_sum, mm_res);
}