#include "libavutil/mips/generic_macros_msa.h"
#include "h264dsp_mips.h"
-static void avc_wgt_4x2_msa(uint8_t *data, int32_t stride,
+static void avc_wgt_4x2_msa(uint8_t *data, ptrdiff_t stride,
int32_t log2_denom, int32_t src_weight,
int32_t offset_in)
{
tmp0 = __msa_srlr_h(tmp0, denom);
tmp0 = (v8i16) __msa_sat_u_h((v8u16) tmp0, 7);
src0 = (v16u8) __msa_pckev_b((v16i8) tmp0, (v16i8) tmp0);
- ST4x2_UB(src0, data, stride);
+ ST_W2(src0, 0, 1, data, stride);
}
-static void avc_wgt_4x4_msa(uint8_t *data, int32_t stride, int32_t log2_denom,
- int32_t src_weight, int32_t offset_in)
+static void avc_wgt_4x4_msa(uint8_t *data, ptrdiff_t stride,
+ int32_t log2_denom, int32_t src_weight,
+ int32_t offset_in)
{
uint32_t tp0, tp1, tp2, tp3, offset_val;
v16u8 src0 = { 0 };
tmp1 = __msa_srlr_h(tmp1, denom);
SAT_UH2_SH(tmp0, tmp1, 7);
src0 = (v16u8) __msa_pckev_b((v16i8) tmp1, (v16i8) tmp0);
- ST4x4_UB(src0, src0, 0, 1, 2, 3, data, stride);
+ ST_W4(src0, 0, 1, 2, 3, data, stride);
}
-static void avc_wgt_4x8_msa(uint8_t *data, int32_t stride, int32_t log2_denom,
- int32_t src_weight, int32_t offset_in)
+static void avc_wgt_4x8_msa(uint8_t *data, ptrdiff_t stride,
+ int32_t log2_denom, int32_t src_weight,
+ int32_t offset_in)
{
uint32_t tp0, tp1, tp2, tp3, offset_val;
v16u8 src0 = { 0 }, src1 = { 0 };
SRLR_H4_SH(tmp0, tmp1, tmp2, tmp3, denom);
SAT_UH4_SH(tmp0, tmp1, tmp2, tmp3, 7);
PCKEV_B2_UB(tmp1, tmp0, tmp3, tmp2, src0, src1);
- ST4x8_UB(src0, src1, data, stride);
+ ST_W8(src0, src1, 0, 1, 2, 3, 0, 1, 2, 3, data, stride);
}
-static void avc_wgt_8x4_msa(uint8_t *data, int32_t stride, int32_t log2_denom,
- int32_t src_weight, int32_t offset_in)
+static void avc_wgt_8x4_msa(uint8_t *data, ptrdiff_t stride,
+ int32_t log2_denom, int32_t src_weight,
+ int32_t offset_in)
{
uint32_t offset_val;
uint64_t tp0, tp1, tp2, tp3;
SRLR_H4_SH(tmp0, tmp1, tmp2, tmp3, denom);
SAT_UH4_SH(tmp0, tmp1, tmp2, tmp3, 7);
PCKEV_B2_UB(tmp1, tmp0, tmp3, tmp2, src0, src1);
- ST8x4_UB(src0, src1, data, stride);
+ ST_D4(src0, src1, 0, 1, 0, 1, data, stride);
}
-static void avc_wgt_8x8_msa(uint8_t *data, int32_t stride, int32_t log2_denom,
+static void avc_wgt_8x8_msa(uint8_t *data, ptrdiff_t stride, int32_t log2_denom,
int32_t src_weight, int32_t offset_in)
{
uint32_t offset_val;
SAT_UH8_SH(tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7, 7);
PCKEV_B4_UB(tmp1, tmp0, tmp3, tmp2, tmp5, tmp4, tmp7, tmp6, src0, src1,
src2, src3);
- ST8x8_UB(src0, src1, src2, src3, data, stride);
+ ST_D8(src0, src1, src2, src3, 0, 1, 0, 1, 0, 1, 0, 1, data, stride);
}
-static void avc_wgt_8x16_msa(uint8_t *data, int32_t stride, int32_t log2_denom,
- int32_t src_weight, int32_t offset_in)
+static void avc_wgt_8x16_msa(uint8_t *data, ptrdiff_t stride,
+ int32_t log2_denom, int32_t src_weight,
+ int32_t offset_in)
{
uint32_t offset_val, cnt;
uint64_t tp0, tp1, tp2, tp3;
SAT_UH8_SH(tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7, 7);
PCKEV_B4_UB(tmp1, tmp0, tmp3, tmp2, tmp5, tmp4, tmp7, tmp6, src0, src1,
src2, src3);
- ST8x8_UB(src0, src1, src2, src3, data, stride);
+ ST_D8(src0, src1, src2, src3, 0, 1, 0, 1, 0, 1, 0, 1, data, stride);
data += 8 * stride;
}
}
-static void avc_biwgt_4x2_msa(uint8_t *src, uint8_t *dst, int32_t stride,
+static void avc_biwgt_4x2_msa(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
int32_t log2_denom, int32_t src_weight,
int32_t dst_weight, int32_t offset_in)
{
tmp0 = __msa_maxi_s_h(tmp0, 0);
tmp0 = __msa_min_s_h(max255, tmp0);
dst0 = (v16u8) __msa_pckev_b((v16i8) tmp0, (v16i8) tmp0);
- ST4x2_UB(dst0, dst, stride);
+ ST_W2(dst0, 0, 1, dst, stride);
}
-static void avc_biwgt_4x4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
+static void avc_biwgt_4x4_msa(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
int32_t log2_denom, int32_t src_weight,
int32_t dst_weight, int32_t offset_in)
{
tmp1 >>= denom;
CLIP_SH2_0_255(tmp0, tmp1);
dst0 = (v16u8) __msa_pckev_b((v16i8) tmp1, (v16i8) tmp0);
- ST4x4_UB(dst0, dst0, 0, 1, 2, 3, dst, stride);
+ ST_W4(dst0, 0, 1, 2, 3, dst, stride);
}
-static void avc_biwgt_4x8_msa(uint8_t *src, uint8_t *dst, int32_t stride,
+static void avc_biwgt_4x8_msa(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
int32_t log2_denom, int32_t src_weight,
int32_t dst_weight, int32_t offset_in)
{
SRA_4V(tmp0, tmp1, tmp2, tmp3, denom);
CLIP_SH4_0_255(tmp0, tmp1, tmp2, tmp3);
PCKEV_B2_UB(tmp1, tmp0, tmp3, tmp2, dst0, dst1);
- ST4x8_UB(dst0, dst1, dst, stride);
+ ST_W8(dst0, dst1, 0, 1, 2, 3, 0, 1, 2, 3, dst, stride);
}
-static void avc_biwgt_8x4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
+static void avc_biwgt_8x4_msa(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
int32_t log2_denom, int32_t src_weight,
int32_t dst_weight, int32_t offset_in)
{
SRA_4V(tmp0, tmp1, tmp2, tmp3, denom);
CLIP_SH4_0_255(tmp0, tmp1, tmp2, tmp3);
PCKEV_B2_UB(tmp1, tmp0, tmp3, tmp2, dst0, dst1);
- ST8x4_UB(dst0, dst1, dst, stride);
+ ST_D4(dst0, dst1, 0, 1, 0, 1, dst, stride);
}
-static void avc_biwgt_8x8_msa(uint8_t *src, uint8_t *dst, int32_t stride,
+static void avc_biwgt_8x8_msa(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
int32_t log2_denom, int32_t src_weight,
int32_t dst_weight, int32_t offset_in)
{
tmp7 = __msa_dpadd_s_h(offset, wgt, vec7);
SRA_4V(tmp0, tmp1, tmp2, tmp3, denom);
SRA_4V(tmp4, tmp5, tmp6, tmp7, denom);
- CLIP_SH4_0_255(tmp0, tmp1, tmp2, tmp3);
- CLIP_SH4_0_255(tmp4, tmp5, tmp6, tmp7);
+ CLIP_SH8_0_255(tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7);
PCKEV_B2_UB(tmp1, tmp0, tmp3, tmp2, dst0, dst1);
PCKEV_B2_UB(tmp5, tmp4, tmp7, tmp6, dst2, dst3);
- ST8x8_UB(dst0, dst1, dst2, dst3, dst, stride);
+ ST_D8(dst0, dst1, dst2, dst3, 0, 1, 0, 1, 0, 1, 0, 1, dst, stride);
}
-static void avc_biwgt_8x16_msa(uint8_t *src, uint8_t *dst, int32_t stride,
+static void avc_biwgt_8x16_msa(uint8_t *src, uint8_t *dst, ptrdiff_t stride,
int32_t log2_denom, int32_t src_weight,
int32_t dst_weight, int32_t offset_in)
{
SRA_4V(temp0, temp1, temp2, temp3, denom);
SRA_4V(temp4, temp5, temp6, temp7, denom);
- CLIP_SH4_0_255(temp0, temp1, temp2, temp3);
- CLIP_SH4_0_255(temp4, temp5, temp6, temp7);
+ CLIP_SH8_0_255(temp0, temp1, temp2, temp3, temp4, temp5, temp6, temp7);
PCKEV_B4_UB(temp1, temp0, temp3, temp2, temp5, temp4, temp7, temp6,
dst0, dst1, dst2, dst3);
- ST8x8_UB(dst0, dst1, dst2, dst3, dst, stride);
+ ST_D8(dst0, dst1, dst2, dst3, 0, 1, 0, 1, 0, 1, 0, 1, dst, stride);
dst += 8 * stride;
}
}
temp = p1_or_q1_org_in << 1; \
clip3 = clip3 - temp; \
clip3 = __msa_ave_s_h(p2_or_q2_org_in, clip3); \
- clip3 = CLIP_SH(clip3, negate_tc_in, tc_in); \
+ CLIP_SH(clip3, negate_tc_in, tc_in); \
p1_or_q1_out = p1_or_q1_org_in + clip3; \
}
delta = q0_sub_p0 + p1_sub_q1; \
delta >>= 3; \
\
- delta = CLIP_SH(delta, negate_threshold_in, threshold_in); \
+ CLIP_SH(delta, negate_threshold_in, threshold_in); \
\
p0_or_q0_out = p0_or_q0_org_in + delta; \
q0_or_p0_out = q0_or_p0_org_in - delta; \
delta = q0_sub_p0 + p1_sub_q1; \
delta = __msa_srari_h(delta, 3); \
\
- delta = CLIP_SH(delta, -tc, tc); \
+ CLIP_SH(delta, -tc, tc); \
\
ILVR_B2_SH(zeros, src1, zeros, src2, res0_r, res1_r); \
\
\
out0 = (v16u8) __msa_ilvr_b((v16i8) in1, (v16i8) in0); \
out1 = (v16u8) __msa_sldi_b(zero_m, (v16i8) out0, 2); \
- SLDI_B2_0_UB(out1, out2, out2, out3, 2); \
+ SLDI_B2_UB(zero_m, out1, zero_m, out2, 2, out2, out3); \
}
#define AVC_LPF_H_2BYTE_CHROMA_422(src, stride, tc_val, alpha, beta, res) \
q0_sub_p0 <<= 2; \
delta = q0_sub_p0 + p1_sub_q1; \
delta = __msa_srari_h(delta, 3); \
- delta = CLIP_SH(delta, -tc, tc); \
+ CLIP_SH(delta, -tc, tc); \
\
ILVR_B2_SH(zeros, src1, zeros, src2, res0_r, res1_r); \
\
static void avc_loopfilter_luma_intra_edge_hor_msa(uint8_t *data,
uint8_t alpha_in,
uint8_t beta_in,
- uint32_t img_width)
+ ptrdiff_t img_width)
{
v16u8 p0_asub_q0, p1_asub_p0, q1_asub_q0;
v16u8 is_less_than, is_less_than_beta, is_less_than_alpha;
static void avc_loopfilter_luma_intra_edge_ver_msa(uint8_t *data,
uint8_t alpha_in,
uint8_t beta_in,
- uint32_t img_width)
+ ptrdiff_t img_width)
{
uint8_t *src = data - 4;
v16u8 alpha, beta, p0_asub_q0;
ILVRL_H2_SH(tp3, tp2, tmp6, tmp7);
src = data - 3;
- ST4x4_UB(tmp3, tmp3, 0, 1, 2, 3, src, img_width);
- ST2x4_UB(tmp2, 0, src + 4, img_width);
+ ST_W4(tmp3, 0, 1, 2, 3, src, img_width);
+ ST_H4(tmp2, 0, 1, 2, 3, src + 4, img_width);
src += 4 * img_width;
- ST4x4_UB(tmp4, tmp4, 0, 1, 2, 3, src, img_width);
- ST2x4_UB(tmp2, 4, src + 4, img_width);
+ ST_W4(tmp4, 0, 1, 2, 3, src, img_width);
+ ST_H4(tmp2, 4, 5, 6, 7, src + 4, img_width);
src += 4 * img_width;
- ST4x4_UB(tmp6, tmp6, 0, 1, 2, 3, src, img_width);
- ST2x4_UB(tmp5, 0, src + 4, img_width);
+ ST_W4(tmp6, 0, 1, 2, 3, src, img_width);
+ ST_H4(tmp5, 0, 1, 2, 3, src + 4, img_width);
src += 4 * img_width;
- ST4x4_UB(tmp7, tmp7, 0, 1, 2, 3, src, img_width);
- ST2x4_UB(tmp5, 4, src + 4, img_width);
+ ST_W4(tmp7, 0, 1, 2, 3, src, img_width);
+ ST_H4(tmp5, 4, 5, 6, 7, src + 4, img_width);
}
}
}
-static void avc_h_loop_filter_luma_mbaff_intra_msa(uint8_t *src, int32_t stride,
+static void avc_h_loop_filter_luma_mbaff_intra_msa(uint8_t *src,
+ ptrdiff_t stride,
int32_t alpha_in,
int32_t beta_in)
{
ILVR_W2_SB(tmp2, tmp0, tmp3, tmp1, src6, src3);
ILVL_W2_SB(tmp2, tmp0, tmp3, tmp1, src1, src5);
- SLDI_B4_0_SB(src6, src1, src3, src5, src0, src2, src4, src7, 8);
+ SLDI_B4_SB(zeros, src6, zeros, src1, zeros, src3, zeros, src5,
+ 8, src0, src2, src4, src7);
p0_asub_q0 = __msa_asub_u_b((v16u8) src2, (v16u8) src3);
p1_asub_p0 = __msa_asub_u_b((v16u8) src1, (v16u8) src2);
ILVRL_H2_SH(zeros, dst2_x, tmp2, tmp3);
ILVR_W2_UB(tmp2, tmp0, tmp3, tmp1, dst0, dst4);
- SLDI_B2_0_UB(dst0, dst4, dst1, dst5, 8);
+ SLDI_B2_UB(zeros, dst0, zeros, dst4, 8, dst1, dst5);
dst2_x = (v16u8) __msa_ilvl_w((v4i32) tmp2, (v4i32) tmp0);
dst2_y = (v16u8) __msa_ilvl_w((v4i32) tmp3, (v4i32) tmp1);
- SLDI_B2_0_UB(dst2_x, dst2_y, dst3_x, dst3_y, 8);
+ SLDI_B2_UB(zeros, dst2_x, zeros, dst2_y, 8, dst3_x, dst3_y);
out0 = __msa_copy_u_w((v4i32) dst0, 0);
out1 = __msa_copy_u_h((v8i16) dst0, 2);
static void avc_loopfilter_cb_or_cr_intra_edge_hor_msa(uint8_t *data_cb_or_cr,
uint8_t alpha_in,
uint8_t beta_in,
- uint32_t img_width)
+ ptrdiff_t img_width)
{
v16u8 alpha, beta;
v16u8 is_less_than;
static void avc_loopfilter_cb_or_cr_intra_edge_ver_msa(uint8_t *data_cb_or_cr,
uint8_t alpha_in,
uint8_t beta_in,
- uint32_t img_width)
+ ptrdiff_t img_width)
{
v8i16 tmp1;
v16u8 alpha, beta, is_less_than;
tmp1 = (v8i16) __msa_ilvr_b((v16i8) q0_or_p0_org, (v16i8) p0_or_q0_org);
data_cb_or_cr -= 1;
- ST2x4_UB(tmp1, 0, data_cb_or_cr, img_width);
+ ST_H4(tmp1, 0, 1, 2, 3, data_cb_or_cr, img_width);
data_cb_or_cr += 4 * img_width;
- ST2x4_UB(tmp1, 4, data_cb_or_cr, img_width);
+ ST_H4(tmp1, 4, 5, 6, 7, data_cb_or_cr, img_width);
}
}
uint8_t tc2, uint8_t tc3,
uint8_t alpha_in,
uint8_t beta_in,
- uint32_t img_width)
+ ptrdiff_t img_width)
{
v16u8 tmp_vec, bs = { 0 };
uint8_t tc2, uint8_t tc3,
uint8_t alpha_in,
uint8_t beta_in,
- uint32_t image_width)
+ ptrdiff_t image_width)
{
v16u8 tmp_vec;
v16u8 bs = { 0 };
}
}
-static void avc_h_loop_filter_luma_mbaff_msa(uint8_t *in, int32_t stride,
+static void avc_h_loop_filter_luma_mbaff_msa(uint8_t *in, ptrdiff_t stride,
int32_t alpha_in, int32_t beta_in,
int8_t *tc0)
{
v8i16 tc, tc_orig_r, tc_plus1;
v16u8 is_tc_orig1, is_tc_orig2, tc_orig = { 0 };
v8i16 p0_ilvr_q0, p0_add_q0, q0_sub_p0, p1_sub_q1;
- v8u16 src2_r, src3_r;
+ v8i16 src2_r, src3_r;
v8i16 p2_r, p1_r, q2_r, q1_r;
v16u8 p2, q2, p0, q0;
v4i32 dst0, dst1;
tc_orig_r = (v8i16) __msa_ilvr_b(zeros, (v16i8) tc_orig);
tc = tc_orig_r;
- p2_r = CLIP_SH(p2_r, -tc_orig_r, tc_orig_r);
- q2_r = CLIP_SH(q2_r, -tc_orig_r, tc_orig_r);
+ CLIP_SH(p2_r, -tc_orig_r, tc_orig_r);
+ CLIP_SH(q2_r, -tc_orig_r, tc_orig_r);
p2_r += p1_r;
q2_r += q1_r;
(v16i8) is_less_than_beta2);
tc = (v8i16) __msa_bmnz_v((v16u8) tc, (v16u8) tc_plus1, is_less_than_beta2);
- q0_sub_p0 = CLIP_SH(q0_sub_p0, -tc, tc);
+ CLIP_SH(q0_sub_p0, -tc, tc);
- ILVR_B2_UH(zeros, src2, zeros, src3, src2_r, src3_r);
+ ILVR_B2_SH(zeros, src2, zeros, src3, src2_r, src3_r);
src2_r += q0_sub_p0;
src3_r -= q0_sub_p0;
- src2_r = (v8u16) CLIP_SH_0_255(src2_r);
- src3_r = (v8u16) CLIP_SH_0_255(src3_r);
+ CLIP_SH2_0_255(src2_r, src3_r);
PCKEV_B2_UB(src2_r, src2_r, src3_r, src3_r, p0, q0);
uint8_t tc2, uint8_t tc3,
uint8_t alpha_in,
uint8_t beta_in,
- uint32_t img_width)
+ ptrdiff_t img_width)
{
v16u8 alpha, beta;
v8i16 tmp_vec;
uint8_t tc2, uint8_t tc3,
uint8_t alpha_in,
uint8_t beta_in,
- uint32_t img_width)
+ ptrdiff_t img_width)
{
uint8_t *src;
v16u8 alpha, beta;
q0_org = __msa_bmnz_v(q0_org, q0, is_less_than);
tmp1 = (v8i16) __msa_ilvr_b((v16i8) q0_org, (v16i8) p0_org);
src = data - 1;
- ST2x4_UB(tmp1, 0, src, img_width);
+ ST_H4(tmp1, 0, 1, 2, 3, src, img_width);
src += 4 * img_width;
- ST2x4_UB(tmp1, 4, src, img_width);
+ ST_H4(tmp1, 4, 5, 6, 7, src, img_width);
}
}
}
-static void avc_h_loop_filter_chroma422_msa(uint8_t *src, int32_t stride,
+static void avc_h_loop_filter_chroma422_msa(uint8_t *src, ptrdiff_t stride,
int32_t alpha_in, int32_t beta_in,
int8_t *tc0)
{
}
AVC_LPF_H_CHROMA_422(src, stride, tc_val, alpha, beta, res);
- ST2x4_UB(res, 0, (src - 1), stride);
+ ST_H4(res, 0, 1, 2, 3, (src - 1), stride);
src += (4 * stride);
}
}
-static void avc_h_loop_filter_chroma422_mbaff_msa(uint8_t *src, int32_t stride,
+static void avc_h_loop_filter_chroma422_mbaff_msa(uint8_t *src,
+ ptrdiff_t stride,
int32_t alpha_in,
int32_t beta_in,
int8_t *tc0)
}
}
-void ff_h264_h_lpf_luma_inter_msa(uint8_t *data, int img_width,
+void ff_h264_h_lpf_luma_inter_msa(uint8_t *data, ptrdiff_t img_width,
int alpha, int beta, int8_t *tc)
{
uint8_t bs0 = 1;
alpha, beta, img_width);
}
-void ff_h264_v_lpf_luma_inter_msa(uint8_t *data, int img_width,
+void ff_h264_v_lpf_luma_inter_msa(uint8_t *data, ptrdiff_t img_width,
int alpha, int beta, int8_t *tc)
{
alpha, beta, img_width);
}
-void ff_h264_h_lpf_chroma_inter_msa(uint8_t *data, int img_width,
+void ff_h264_h_lpf_chroma_inter_msa(uint8_t *data, ptrdiff_t img_width,
int alpha, int beta, int8_t *tc)
{
uint8_t bs0 = 1;
alpha, beta, img_width);
}
-void ff_h264_v_lpf_chroma_inter_msa(uint8_t *data, int img_width,
+void ff_h264_v_lpf_chroma_inter_msa(uint8_t *data, ptrdiff_t img_width,
int alpha, int beta, int8_t *tc)
{
uint8_t bs0 = 1;
alpha, beta, img_width);
}
-void ff_h264_h_lpf_luma_intra_msa(uint8_t *data, int img_width,
+void ff_h264_h_lpf_luma_intra_msa(uint8_t *data, ptrdiff_t img_width,
int alpha, int beta)
{
avc_loopfilter_luma_intra_edge_ver_msa(data, (uint8_t) alpha,
(uint8_t) beta,
- (unsigned int) img_width);
+ img_width);
}
-void ff_h264_v_lpf_luma_intra_msa(uint8_t *data, int img_width,
+void ff_h264_v_lpf_luma_intra_msa(uint8_t *data, ptrdiff_t img_width,
int alpha, int beta)
{
avc_loopfilter_luma_intra_edge_hor_msa(data, (uint8_t) alpha,
(uint8_t) beta,
- (unsigned int) img_width);
+ img_width);
}
-void ff_h264_h_lpf_chroma_intra_msa(uint8_t *data, int img_width,
+void ff_h264_h_lpf_chroma_intra_msa(uint8_t *data, ptrdiff_t img_width,
int alpha, int beta)
{
avc_loopfilter_cb_or_cr_intra_edge_ver_msa(data, (uint8_t) alpha,
(uint8_t) beta,
- (unsigned int) img_width);
+ img_width);
}
-void ff_h264_v_lpf_chroma_intra_msa(uint8_t *data, int img_width,
+void ff_h264_v_lpf_chroma_intra_msa(uint8_t *data, ptrdiff_t img_width,
int alpha, int beta)
{
avc_loopfilter_cb_or_cr_intra_edge_hor_msa(data, (uint8_t) alpha,
(uint8_t) beta,
- (unsigned int) img_width);
+ img_width);
}
void ff_h264_h_loop_filter_chroma422_msa(uint8_t *src,
- int32_t ystride,
+ ptrdiff_t ystride,
int32_t alpha, int32_t beta,
int8_t *tc0)
{
}
void ff_h264_h_loop_filter_chroma422_mbaff_msa(uint8_t *src,
- int32_t ystride,
+ ptrdiff_t ystride,
int32_t alpha,
int32_t beta,
int8_t *tc0)
}
void ff_h264_h_loop_filter_luma_mbaff_msa(uint8_t *src,
- int32_t ystride,
+ ptrdiff_t ystride,
int32_t alpha,
int32_t beta,
int8_t *tc0)
}
void ff_h264_h_loop_filter_luma_mbaff_intra_msa(uint8_t *src,
- int32_t ystride,
+ ptrdiff_t ystride,
int32_t alpha,
int32_t beta)
{
SRA_4V(tmp4, tmp5, tmp6, tmp7, denom);
SRA_4V(tmp8, tmp9, tmp10, tmp11, denom);
SRA_4V(tmp12, tmp13, tmp14, tmp15, denom);
- CLIP_SH4_0_255(tmp0, tmp1, tmp2, tmp3);
- CLIP_SH4_0_255(tmp4, tmp5, tmp6, tmp7);
- CLIP_SH4_0_255(tmp8, tmp9, tmp10, tmp11);
- CLIP_SH4_0_255(tmp12, tmp13, tmp14, tmp15);
+ CLIP_SH8_0_255(tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7);
+ CLIP_SH8_0_255(tmp8, tmp9, tmp10, tmp11, tmp12, tmp13, tmp14, tmp15);
PCKEV_B4_UB(tmp1, tmp0, tmp3, tmp2, tmp5, tmp4, tmp7, tmp6, dst0, dst1,
dst2, dst3);
PCKEV_B4_UB(tmp9, tmp8, tmp11, tmp10, tmp13, tmp12, tmp15, tmp14, dst4,
SRA_4V(tmp4, tmp5, tmp6, tmp7, denom);
SRA_4V(tmp8, tmp9, tmp10, tmp11, denom);
SRA_4V(tmp12, tmp13, tmp14, tmp15, denom);
- CLIP_SH4_0_255(tmp0, tmp1, tmp2, tmp3);
- CLIP_SH4_0_255(tmp4, tmp5, tmp6, tmp7);
- CLIP_SH4_0_255(tmp8, tmp9, tmp10, tmp11);
- CLIP_SH4_0_255(tmp12, tmp13, tmp14, tmp15);
+ CLIP_SH8_0_255(tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7);
+ CLIP_SH8_0_255(tmp8, tmp9, tmp10, tmp11, tmp12, tmp13, tmp14, tmp15);
PCKEV_B4_UB(tmp1, tmp0, tmp3, tmp2, tmp5, tmp4, tmp7, tmp6, dst0, dst1,
dst2, dst3);
PCKEV_B4_UB(tmp9, tmp8, tmp11, tmp10, tmp13, tmp12, tmp15, tmp14, dst4,
projects / ffmpeg / blobdiff