diff options
Diffstat (limited to 'chromium/third_party/libvpx/source/libvpx/vpx_dsp/mips/loopfilter_msa.h')
-rw-r--r-- | chromium/third_party/libvpx/source/libvpx/vpx_dsp/mips/loopfilter_msa.h | 235 |
1 files changed, 81 insertions, 154 deletions
diff --git a/chromium/third_party/libvpx/source/libvpx/vpx_dsp/mips/loopfilter_msa.h b/chromium/third_party/libvpx/source/libvpx/vpx_dsp/mips/loopfilter_msa.h index 4063e5e6b87..49fd74c25a4 100644 --- a/chromium/third_party/libvpx/source/libvpx/vpx_dsp/mips/loopfilter_msa.h +++ b/chromium/third_party/libvpx/source/libvpx/vpx_dsp/mips/loopfilter_msa.h @@ -13,144 +13,71 @@ #include "vpx_dsp/mips/macros_msa.h" -#define VP9_LPF_FILTER4_8W(p1_in, p0_in, q0_in, q1_in, mask_in, hev_in, \ - p1_out, p0_out, q0_out, q1_out) \ - { \ - v16i8 p1_m, p0_m, q0_m, q1_m, q0_sub_p0, filt_sign; \ - v16i8 filt, filt1, filt2, cnst4b, cnst3b; \ - v8i16 q0_sub_p0_r, filt_r, cnst3h; \ - \ - p1_m = (v16i8)__msa_xori_b(p1_in, 0x80); \ - p0_m = (v16i8)__msa_xori_b(p0_in, 0x80); \ - q0_m = (v16i8)__msa_xori_b(q0_in, 0x80); \ - q1_m = (v16i8)__msa_xori_b(q1_in, 0x80); \ - \ - filt = __msa_subs_s_b(p1_m, q1_m); \ - filt = filt & (v16i8)hev_in; \ - q0_sub_p0 = q0_m - p0_m; \ - filt_sign = __msa_clti_s_b(filt, 0); \ - \ - cnst3h = __msa_ldi_h(3); \ - q0_sub_p0_r = (v8i16)__msa_ilvr_b(q0_sub_p0, q0_sub_p0); \ - q0_sub_p0_r = __msa_dotp_s_h((v16i8)q0_sub_p0_r, (v16i8)cnst3h); \ - filt_r = (v8i16)__msa_ilvr_b(filt_sign, filt); \ - filt_r += q0_sub_p0_r; \ - filt_r = __msa_sat_s_h(filt_r, 7); \ - \ - /* combine left and right part */ \ - filt = __msa_pckev_b((v16i8)filt_r, (v16i8)filt_r); \ - \ - filt = filt & (v16i8)mask_in; \ - cnst4b = __msa_ldi_b(4); \ - filt1 = __msa_adds_s_b(filt, cnst4b); \ - filt1 >>= 3; \ - \ - cnst3b = __msa_ldi_b(3); \ - filt2 = __msa_adds_s_b(filt, cnst3b); \ - filt2 >>= 3; \ - \ - q0_m = __msa_subs_s_b(q0_m, filt1); \ - q0_out = __msa_xori_b((v16u8)q0_m, 0x80); \ - p0_m = __msa_adds_s_b(p0_m, filt2); \ - p0_out = __msa_xori_b((v16u8)p0_m, 0x80); \ - \ - filt = __msa_srari_b(filt1, 1); \ - hev_in = __msa_xori_b((v16u8)hev_in, 0xff); \ - filt = filt & (v16i8)hev_in; \ - \ - q1_m = __msa_subs_s_b(q1_m, filt); \ - q1_out = __msa_xori_b((v16u8)q1_m, 0x80); \ - p1_m = __msa_adds_s_b(p1_m, filt); \ - p1_out = __msa_xori_b((v16u8)p1_m, 0x80); \ - } - -#define VP9_LPF_FILTER4_4W(p1_in, p0_in, q0_in, q1_in, mask_in, hev_in, \ - p1_out, p0_out, q0_out, q1_out) \ - { \ - v16i8 p1_m, p0_m, q0_m, q1_m, q0_sub_p0, filt_sign; \ - v16i8 filt, filt1, filt2, cnst4b, cnst3b; \ - v8i16 q0_sub_p0_r, q0_sub_p0_l, filt_l, filt_r, cnst3h; \ - \ - p1_m = (v16i8)__msa_xori_b(p1_in, 0x80); \ - p0_m = (v16i8)__msa_xori_b(p0_in, 0x80); \ - q0_m = (v16i8)__msa_xori_b(q0_in, 0x80); \ - q1_m = (v16i8)__msa_xori_b(q1_in, 0x80); \ - \ - filt = __msa_subs_s_b(p1_m, q1_m); \ - \ - filt = filt & (v16i8)hev_in; \ - \ - q0_sub_p0 = q0_m - p0_m; \ - filt_sign = __msa_clti_s_b(filt, 0); \ - \ - cnst3h = __msa_ldi_h(3); \ - q0_sub_p0_r = (v8i16)__msa_ilvr_b(q0_sub_p0, q0_sub_p0); \ - q0_sub_p0_r = __msa_dotp_s_h((v16i8)q0_sub_p0_r, (v16i8)cnst3h); \ - filt_r = (v8i16)__msa_ilvr_b(filt_sign, filt); \ - filt_r += q0_sub_p0_r; \ - filt_r = __msa_sat_s_h(filt_r, 7); \ - \ - q0_sub_p0_l = (v8i16)__msa_ilvl_b(q0_sub_p0, q0_sub_p0); \ - q0_sub_p0_l = __msa_dotp_s_h((v16i8)q0_sub_p0_l, (v16i8)cnst3h); \ - filt_l = (v8i16)__msa_ilvl_b(filt_sign, filt); \ - filt_l += q0_sub_p0_l; \ - filt_l = __msa_sat_s_h(filt_l, 7); \ - \ - filt = __msa_pckev_b((v16i8)filt_l, (v16i8)filt_r); \ - filt = filt & (v16i8)mask_in; \ - \ - cnst4b = __msa_ldi_b(4); \ - filt1 = __msa_adds_s_b(filt, cnst4b); \ - filt1 >>= 3; \ - \ - cnst3b = __msa_ldi_b(3); \ - filt2 = __msa_adds_s_b(filt, cnst3b); \ - filt2 >>= 3; \ - \ - q0_m = __msa_subs_s_b(q0_m, filt1); \ - q0_out = __msa_xori_b((v16u8)q0_m, 0x80); \ - p0_m = __msa_adds_s_b(p0_m, filt2); \ - p0_out = __msa_xori_b((v16u8)p0_m, 0x80); \ - \ - filt = __msa_srari_b(filt1, 1); \ - hev_in = __msa_xori_b((v16u8)hev_in, 0xff); \ - filt = filt & (v16i8)hev_in; \ - \ - q1_m = __msa_subs_s_b(q1_m, filt); \ - q1_out = __msa_xori_b((v16u8)q1_m, 0x80); \ - p1_m = __msa_adds_s_b(p1_m, filt); \ - p1_out = __msa_xori_b((v16u8)p1_m, 0x80); \ +#define VP9_LPF_FILTER4_4W(p1_in, p0_in, q0_in, q1_in, mask, hev, p1_out, \ + p0_out, q0_out, q1_out) \ + { \ + v16i8 p1_m, p0_m, q0_m, q1_m, filt, q0_sub_p0, t1, t2; \ + const v16i8 cnst4b = __msa_ldi_b(4); \ + const v16i8 cnst3b = __msa_ldi_b(3); \ + \ + p1_m = (v16i8)__msa_xori_b(p1_in, 0x80); \ + p0_m = (v16i8)__msa_xori_b(p0_in, 0x80); \ + q0_m = (v16i8)__msa_xori_b(q0_in, 0x80); \ + q1_m = (v16i8)__msa_xori_b(q1_in, 0x80); \ + \ + filt = __msa_subs_s_b(p1_m, q1_m); \ + filt &= hev; \ + q0_sub_p0 = __msa_subs_s_b(q0_m, p0_m); \ + filt = __msa_adds_s_b(filt, q0_sub_p0); \ + filt = __msa_adds_s_b(filt, q0_sub_p0); \ + filt = __msa_adds_s_b(filt, q0_sub_p0); \ + filt &= mask; \ + t1 = __msa_adds_s_b(filt, cnst4b); \ + t1 >>= cnst3b; \ + t2 = __msa_adds_s_b(filt, cnst3b); \ + t2 >>= cnst3b; \ + q0_m = __msa_subs_s_b(q0_m, t1); \ + q0_out = __msa_xori_b((v16u8)q0_m, 0x80); \ + p0_m = __msa_adds_s_b(p0_m, t2); \ + p0_out = __msa_xori_b((v16u8)p0_m, 0x80); \ + filt = __msa_srari_b(t1, 1); \ + hev = __msa_xori_b(hev, 0xff); \ + filt &= hev; \ + q1_m = __msa_subs_s_b(q1_m, filt); \ + q1_out = __msa_xori_b((v16u8)q1_m, 0x80); \ + p1_m = __msa_adds_s_b(p1_m, filt); \ + p1_out = __msa_xori_b((v16u8)p1_m, 0x80); \ } -#define VP9_FLAT4(p3_in, p2_in, p0_in, q0_in, q2_in, q3_in, flat_out) \ - { \ - v16u8 tmp, p2_a_sub_p0, q2_a_sub_q0, p3_a_sub_p0, q3_a_sub_q0; \ - v16u8 zero_in = { 0 }; \ - \ - tmp = __msa_ori_b(zero_in, 1); \ - p2_a_sub_p0 = __msa_asub_u_b(p2_in, p0_in); \ - q2_a_sub_q0 = __msa_asub_u_b(q2_in, q0_in); \ - p3_a_sub_p0 = __msa_asub_u_b(p3_in, p0_in); \ - q3_a_sub_q0 = __msa_asub_u_b(q3_in, q0_in); \ - \ - p2_a_sub_p0 = __msa_max_u_b(p2_a_sub_p0, q2_a_sub_q0); \ - flat_out = __msa_max_u_b(p2_a_sub_p0, flat_out); \ - p3_a_sub_p0 = __msa_max_u_b(p3_a_sub_p0, q3_a_sub_q0); \ - flat_out = __msa_max_u_b(p3_a_sub_p0, flat_out); \ - \ - flat_out = (tmp < (v16u8)flat_out); \ - flat_out = __msa_xori_b(flat_out, 0xff); \ - flat_out = flat_out & (mask); \ +#define VP9_FLAT4(p3_in, p2_in, p0_in, q0_in, q2_in, q3_in, flat_out) \ + { \ + v16u8 tmp_flat4, p2_a_sub_p0, q2_a_sub_q0, p3_a_sub_p0, q3_a_sub_q0; \ + v16u8 zero_in = { 0 }; \ + \ + tmp_flat4 = __msa_ori_b(zero_in, 1); \ + p2_a_sub_p0 = __msa_asub_u_b(p2_in, p0_in); \ + q2_a_sub_q0 = __msa_asub_u_b(q2_in, q0_in); \ + p3_a_sub_p0 = __msa_asub_u_b(p3_in, p0_in); \ + q3_a_sub_q0 = __msa_asub_u_b(q3_in, q0_in); \ + \ + p2_a_sub_p0 = __msa_max_u_b(p2_a_sub_p0, q2_a_sub_q0); \ + flat_out = __msa_max_u_b(p2_a_sub_p0, flat_out); \ + p3_a_sub_p0 = __msa_max_u_b(p3_a_sub_p0, q3_a_sub_q0); \ + flat_out = __msa_max_u_b(p3_a_sub_p0, flat_out); \ + \ + flat_out = (tmp_flat4 < (v16u8)flat_out); \ + flat_out = __msa_xori_b(flat_out, 0xff); \ + flat_out = flat_out & (mask); \ } #define VP9_FLAT5(p7_in, p6_in, p5_in, p4_in, p0_in, q0_in, q4_in, q5_in, \ q6_in, q7_in, flat_in, flat2_out) \ { \ - v16u8 tmp, zero_in = { 0 }; \ + v16u8 tmp_flat5, zero_in = { 0 }; \ v16u8 p4_a_sub_p0, q4_a_sub_q0, p5_a_sub_p0, q5_a_sub_q0; \ v16u8 p6_a_sub_p0, q6_a_sub_q0, p7_a_sub_p0, q7_a_sub_q0; \ \ - tmp = __msa_ori_b(zero_in, 1); \ + tmp_flat5 = __msa_ori_b(zero_in, 1); \ p4_a_sub_p0 = __msa_asub_u_b(p4_in, p0_in); \ q4_a_sub_q0 = __msa_asub_u_b(q4_in, q0_in); \ p5_a_sub_p0 = __msa_asub_u_b(p5_in, p0_in); \ @@ -168,7 +95,7 @@ p7_a_sub_p0 = __msa_max_u_b(p7_a_sub_p0, q7_a_sub_q0); \ flat2_out = __msa_max_u_b(p7_a_sub_p0, flat2_out); \ \ - flat2_out = (tmp < (v16u8)flat2_out); \ + flat2_out = (tmp_flat5 < (v16u8)flat2_out); \ flat2_out = __msa_xori_b(flat2_out, 0xff); \ flat2_out = flat2_out & flat_in; \ } @@ -177,38 +104,38 @@ p2_filt8_out, p1_filt8_out, p0_filt8_out, q0_filt8_out, \ q1_filt8_out, q2_filt8_out) \ { \ - v8u16 tmp0, tmp1, tmp2; \ + v8u16 tmp_filt8_0, tmp_filt8_1, tmp_filt8_2; \ \ - tmp2 = p2_in + p1_in + p0_in; \ - tmp0 = p3_in << 1; \ + tmp_filt8_2 = p2_in + p1_in + p0_in; \ + tmp_filt8_0 = p3_in << 1; \ \ - tmp0 = tmp0 + tmp2 + q0_in; \ - tmp1 = tmp0 + p3_in + p2_in; \ - p2_filt8_out = (v8i16)__msa_srari_h((v8i16)tmp1, 3); \ + tmp_filt8_0 = tmp_filt8_0 + tmp_filt8_2 + q0_in; \ + tmp_filt8_1 = tmp_filt8_0 + p3_in + p2_in; \ + p2_filt8_out = (v8i16)__msa_srari_h((v8i16)tmp_filt8_1, 3); \ \ - tmp1 = tmp0 + p1_in + q1_in; \ - p1_filt8_out = (v8i16)__msa_srari_h((v8i16)tmp1, 3); \ + tmp_filt8_1 = tmp_filt8_0 + p1_in + q1_in; \ + p1_filt8_out = (v8i16)__msa_srari_h((v8i16)tmp_filt8_1, 3); \ \ - tmp1 = q2_in + q1_in + q0_in; \ - tmp2 = tmp2 + tmp1; \ - tmp0 = tmp2 + (p0_in); \ - tmp0 = tmp0 + (p3_in); \ - p0_filt8_out = (v8i16)__msa_srari_h((v8i16)tmp0, 3); \ + tmp_filt8_1 = q2_in + q1_in + q0_in; \ + tmp_filt8_2 = tmp_filt8_2 + tmp_filt8_1; \ + tmp_filt8_0 = tmp_filt8_2 + (p0_in); \ + tmp_filt8_0 = tmp_filt8_0 + (p3_in); \ + p0_filt8_out = (v8i16)__msa_srari_h((v8i16)tmp_filt8_0, 3); \ \ - tmp0 = q2_in + q3_in; \ - tmp0 = p0_in + tmp1 + tmp0; \ - tmp1 = q3_in + q3_in; \ - tmp1 = tmp1 + tmp0; \ - q2_filt8_out = (v8i16)__msa_srari_h((v8i16)tmp1, 3); \ + tmp_filt8_0 = q2_in + q3_in; \ + tmp_filt8_0 = p0_in + tmp_filt8_1 + tmp_filt8_0; \ + tmp_filt8_1 = q3_in + q3_in; \ + tmp_filt8_1 = tmp_filt8_1 + tmp_filt8_0; \ + q2_filt8_out = (v8i16)__msa_srari_h((v8i16)tmp_filt8_1, 3); \ \ - tmp0 = tmp2 + q3_in; \ - tmp1 = tmp0 + q0_in; \ - q0_filt8_out = (v8i16)__msa_srari_h((v8i16)tmp1, 3); \ + tmp_filt8_0 = tmp_filt8_2 + q3_in; \ + tmp_filt8_1 = tmp_filt8_0 + q0_in; \ + q0_filt8_out = (v8i16)__msa_srari_h((v8i16)tmp_filt8_1, 3); \ \ - tmp1 = tmp0 - p2_in; \ - tmp0 = q1_in + q3_in; \ - tmp1 = tmp0 + tmp1; \ - q1_filt8_out = (v8i16)__msa_srari_h((v8i16)tmp1, 3); \ + tmp_filt8_1 = tmp_filt8_0 - p2_in; \ + tmp_filt8_0 = q1_in + q3_in; \ + tmp_filt8_1 = tmp_filt8_0 + tmp_filt8_1; \ + q1_filt8_out = (v8i16)__msa_srari_h((v8i16)tmp_filt8_1, 3); \ } #define LPF_MASK_HEV(p3_in, p2_in, p1_in, p0_in, q0_in, q1_in, q2_in, q3_in, \ |