diff mbox

[FFmpeg-devel] avcodec/mips: Improve avc chroma copy and avg vert mc msa functions

Message ID 70293ACCC3BA6A4E81FFCA024C7A86E1E0595AF4@PUMAIL01.pu.imgtec.org
State New
Headers show

Commit Message

Manojkumar Bhosale Oct. 24, 2017, 11:21 a.m. UTC 
LGTM

-----Original Message-----
From: ffmpeg-devel [mailto:ffmpeg-devel-bounces@ffmpeg.org] On Behalf Of kaustubh.raste@imgtec.com

Sent: Tuesday, October 24, 2017 12:41 PM
To: ffmpeg-devel@ffmpeg.org
Cc: Kaustubh Raste
Subject: [FFmpeg-devel] [PATCH] avcodec/mips: Improve avc chroma copy and avg vert mc msa functions

From: Kaustubh Raste <kaustubh.raste@imgtec.com>


Replace generic with block size specific function.
Load the specific destination bytes instead of MSA load and pack.

Signed-off-by: Kaustubh Raste <kaustubh.raste@imgtec.com>

---
 libavcodec/mips/h264chroma_msa.c |  627 +++++++++++++++++---------------------
 1 file changed, 275 insertions(+), 352 deletions(-)

     res = (v8i16) __msa_pckev_b((v16i8) res_r, (v16i8) res_r);
     res = (v8i16) __msa_aver_u_b((v16u8) res, dst_data);
 
-    ST2x4_UB(res, 0, dst, dst_stride);
-    dst += (4 * dst_stride);
-}
-
-static void avc_chroma_vt_and_aver_dst_2x8_msa(uint8_t *src, int32_t src_stride,
-                                               uint8_t *dst, int32_t dst_stride,
-                                               uint32_t coeff0, uint32_t coeff1)
-{
-    uint32_t load0, load1, load2, load3;
-    v16i8 src0, src1, src2, src3, src4, src5, src6, src7, src8;
-    v16u8 tmp0, tmp1, tmp2, tmp3;
-    v8i16 res;
-    v8u16 res_r;
-    v16i8 coeff_vec0 = __msa_fill_b(coeff0);
-    v16i8 coeff_vec1 = __msa_fill_b(coeff1);
-    v16u8 coeff_vec = (v16u8) __msa_ilvr_b(coeff_vec0, coeff_vec1);
-    v16u8 dst_data0 = { 0 };
-    v16u8 dst_data1 = { 0 };
-
-    LD_SB5(src, src_stride, src0, src1, src2, src3, src4);
-    src += (5 * src_stride);
-    LD_SB4(src, src_stride, src5, src6, src7, src8);
-
-    LW4(dst, dst_stride, load0, load1, load2, load3);
-
-    dst_data0 = (v16u8) __msa_insert_h((v8i16) dst_data0, 0, load0);
-    dst_data0 = (v16u8) __msa_insert_h((v8i16) dst_data0, 1, load1);
-    dst_data0 = (v16u8) __msa_insert_h((v8i16) dst_data0, 2, load2);
-    dst_data0 = (v16u8) __msa_insert_h((v8i16) dst_data0, 3, load3);
-
-    LW4(dst + 4 * dst_stride, dst_stride, load0, load1, load2, load3);
-
-    dst_data1 = (v16u8) __msa_insert_h((v8i16) dst_data1, 0, load0);
-    dst_data1 = (v16u8) __msa_insert_h((v8i16) dst_data1, 1, load1);
-    dst_data1 = (v16u8) __msa_insert_h((v8i16) dst_data1, 2, load2);
-    dst_data1 = (v16u8) __msa_insert_h((v8i16) dst_data1, 3, load3);
-
-    ILVR_B4_UB(src1, src0, src2, src1, src3, src2, src4, src3,
-               tmp0, tmp1, tmp2, tmp3);
-
-    ILVR_W2_UB(tmp1, tmp0, tmp3, tmp2, tmp0, tmp2);
-
-    tmp0 = (v16u8) __msa_ilvr_d((v2i64) tmp2, (v2i64) tmp0);
-
-    res_r = __msa_dotp_u_h(tmp0, coeff_vec);
-    res_r <<= 3;
-    res_r = (v8u16) __msa_srari_h((v8i16) res_r, 6);
-    res_r = __msa_sat_u_h(res_r, 7);
-
-    res = (v8i16) __msa_pckev_b((v16i8) res_r, (v16i8) res_r);
-    res = (v8i16) __msa_aver_u_b((v16u8) res, dst_data0);
-
-    ST2x4_UB(res, 0, dst, dst_stride);
-    dst += (4 * dst_stride);
-
-    ILVR_B4_UB(src5, src4, src6, src5, src7, src6, src8, src7,
-               tmp0, tmp1, tmp2, tmp3);
-
-    ILVR_W2_UB(tmp1, tmp0, tmp3, tmp2, tmp0, tmp2);
-
-    tmp0 = (v16u8) __msa_ilvr_d((v2i64) tmp2, (v2i64) tmp0);
-
-    res_r = __msa_dotp_u_h(tmp0, coeff_vec);
-    res_r <<= 3;
-    res_r = (v8u16) __msa_srari_h((v8i16) res_r, 6);
-    res_r = __msa_sat_u_h(res_r, 7);
-
-    res = (v8i16) __msa_pckev_b((v16i8) res_r, (v16i8) res_r);
-    res = (v8i16) __msa_aver_u_b((v16u8) res, dst_data1);
-
-    ST2x4_UB(res, 0, dst, dst_stride);
+    ST2x4_UB(res, 0, dst, stride);
 }
 
-static void avc_chroma_vt_and_aver_dst_2w_msa(uint8_t *src, int32_t src_stride,
-                                              uint8_t *dst, int32_t dst_stride,
-                                              uint32_t coeff0, uint32_t coeff1,
-                                              int32_t height)
+static void avc_chroma_vt_and_aver_dst_2w_msa(uint8_t *src, uint8_t *dst,
+                                              int32_t stride, uint32_t coeff0,
+                                              uint32_t coeff1, int32_t 
+height)
 {
     if (2 == height) {
-        avc_chroma_vt_and_aver_dst_2x2_msa(src, src_stride, dst, dst_stride,
-                                           coeff0, coeff1);
+        avc_chroma_vt_and_aver_dst_2x2_msa(src, dst, stride, coeff0, 
+ coeff1);
     } else if (4 == height) {
-        avc_chroma_vt_and_aver_dst_2x4_msa(src, src_stride, dst, dst_stride,
-                                           coeff0, coeff1);
-    } else if (8 == height) {
-        avc_chroma_vt_and_aver_dst_2x8_msa(src, src_stride, dst, dst_stride,
-                                           coeff0, coeff1);
+        avc_chroma_vt_and_aver_dst_2x4_msa(src, dst, stride, coeff0, 
+ coeff1);
     }
 }
 
-static void avc_chroma_vt_and_aver_dst_4x2_msa(uint8_t *src, int32_t src_stride,
-                                               uint8_t *dst, int32_t dst_stride,
-                                               uint32_t coeff0, uint32_t coeff1)
+static void avc_chroma_vt_and_aver_dst_4x2_msa(uint8_t *src, uint8_t *dst,
+                                               int32_t stride, uint32_t coeff0,
+                                               uint32_t coeff1)
 {
     uint32_t load0, load1;
-    v16i8 src0, src1, src2, tmp0, tmp1;
+    v16u8 src0, src1, src2, tmp0, tmp1;
     v16u8 dst_data = { 0 };
     v8u16 res_r;
     v16u8 res;
@@ -1305,128 +1227,184 @@ static void avc_chroma_vt_and_aver_dst_4x2_msa(uint8_t *src, int32_t src_stride,
     v16i8 coeff_vec1 = __msa_fill_b(coeff1);
     v16u8 coeff_vec = (v16u8) __msa_ilvr_b(coeff_vec0, coeff_vec1);
 
-    LD_SB3(src, src_stride, src0, src1, src2);
+    LD_UB3(src, stride, src0, src1, src2);
 
-    load0 = LW(dst);
-    load1 = LW(dst + dst_stride);
+    LW2(dst, stride, load0, load1);
 
     INSERT_W2_UB(load0, load1, dst_data);
-    ILVR_B2_SB(src1, src0, src2, src1, tmp0, tmp1);
+    ILVR_B2_UB(src1, src0, src2, src1, tmp0, tmp1);
 
-    tmp0 = (v16i8) __msa_ilvr_d((v2i64) tmp1, (v2i64) tmp0);
+    tmp0 = (v16u8) __msa_ilvr_d((v2i64) tmp1, (v2i64) tmp0);
 
-    res_r = __msa_dotp_u_h((v16u8) tmp0, coeff_vec);
+    res_r = __msa_dotp_u_h(tmp0, coeff_vec);
     res_r <<= 3;
     res_r = (v8u16) __msa_srari_h((v8i16) res_r, 6);
     res_r = __msa_sat_u_h(res_r, 7);
     res = (v16u8) __msa_pckev_b((v16i8) res_r, (v16i8) res_r);
     res = __msa_aver_u_b(res, dst_data);
 
-    ST4x2_UB(res, dst, dst_stride);
+    ST4x2_UB(res, dst, stride);
 }
 
-static void avc_chroma_vt_and_aver_dst_4x4mul_msa(uint8_t *src,
-                                                  int32_t src_stride,
-                                                  uint8_t *dst,
-                                                  int32_t dst_stride,
-                                                  uint32_t coeff0,
-                                                  uint32_t coeff1,
-                                                  int32_t height)
+static void avc_chroma_vt_and_aver_dst_4x4_msa(uint8_t *src, uint8_t *dst,
+                                               int32_t stride, uint32_t coeff0,
+                                               uint32_t coeff1)
 {
-    uint32_t load0, load1, row;
-    v16i8 src0, src1, src2, src3, src4;
+    uint32_t tp0, tp1, tp2, tp3;
+    v16u8 src0, src1, src2, src3, src4;
     v16u8 tmp0, tmp1, tmp2, tmp3;
     v16u8 dst0 = { 0 };
-    v16u8 dst1 = { 0 };
     v8u16 res0_r, res1_r;
-    v16u8 res0, res1;
+    v16u8 out;
     v16i8 coeff_vec0 = __msa_fill_b(coeff0);
     v16i8 coeff_vec1 = __msa_fill_b(coeff1);
     v16u8 coeff_vec = (v16u8) __msa_ilvr_b(coeff_vec0, coeff_vec1);
 
-    src0 = LD_SB(src);
-    src += src_stride;
-
-    for (row = (height >> 2); row--;) {
-        LD_SB4(src, src_stride, src1, src2, src3, src4);
-        src += (4 * src_stride);
-
-        load0 = LW(dst);
-        load1 = LW(dst + dst_stride);
-
-        INSERT_W2_UB(load0, load1, dst0);
-        load0 = LW(dst + 2 * dst_stride);
-        load1 = LW(dst + 3 * dst_stride);
-        INSERT_W2_UB(load0, load1, dst1);
-
-        ILVR_B4_UB(src1, src0, src2, src1, src3, src2, src4, src3,
-                   tmp0, tmp1, tmp2, tmp3);
-        ILVR_D2_UB(tmp1, tmp0, tmp3, tmp2, tmp0, tmp2);
-        DOTP_UB2_UH(tmp0, tmp2, coeff_vec, coeff_vec, res0_r, res1_r);
-
-        res0_r <<= 3;
-        res1_r <<= 3;
+    LD_UB5(src, stride, src0, src1, src2, src3, src4);
+    LW4(dst, stride, tp0, tp1, tp2, tp3);
+    INSERT_W4_UB(tp0, tp1, tp2, tp3, dst0);
+    ILVR_B4_UB(src1, src0, src2, src1, src3, src2, src4, src3, tmp0, tmp1, tmp2,
+               tmp3);
+    ILVR_D2_UB(tmp1, tmp0, tmp3, tmp2, tmp0, tmp2);
+    DOTP_UB2_UH(tmp0, tmp2, coeff_vec, coeff_vec, res0_r, res1_r);
+    res0_r <<= 3;
+    res1_r <<= 3;
+    SRARI_H2_UH(res0_r, res1_r, 6);
+    SAT_UH2_UH(res0_r, res1_r, 7);
+    out = (v16u8) __msa_pckev_b((v16i8) res1_r, (v16i8) res0_r);
+    out = __msa_aver_u_b(out, dst0);
+    ST4x4_UB(out, out, 0, 1, 2, 3, dst, stride); }
 
-        SRARI_H2_UH(res0_r, res1_r, 6);
-        SAT_UH2_UH(res0_r, res1_r, 7);
-        PCKEV_B2_UB(res0_r, res0_r, res1_r, res1_r, res0, res1);
-        AVER_UB2_UB(res0, dst0, res1, dst1, res0, res1);
+static void avc_chroma_vt_and_aver_dst_4x8_msa(uint8_t *src, uint8_t *dst,
+                                               int32_t stride, uint32_t coeff0,
+                                               uint32_t coeff1) {
+    uint32_t tp0, tp1, tp2, tp3;
+    v16u8 src0, src1, src2, src3, src4, src5, src6, src7, src8;
+    v16u8 tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7, out0, out1;
+    v16u8 dst0 = { 0 }, dst1 = { 0 };
+    v8u16 res0, res1, res2, res3;
+    v16i8 coeff_vec0 = __msa_fill_b(coeff0);
+    v16i8 coeff_vec1 = __msa_fill_b(coeff1);
+    v16u8 coeff_vec = (v16u8) __msa_ilvr_b(coeff_vec0, coeff_vec1);
 
-        ST4x4_UB(res0, res1, 0, 1, 0, 1, dst, dst_stride);
-        dst += (4 * dst_stride);
-        src0 = src4;
-    }
+    LD_UB5(src, stride, src0, src1, src2, src3, src4);
+    src += (5 * stride);
+    LD_UB4(src, stride, src5, src6, src7, src8);
+    LW4(dst, stride, tp0, tp1, tp2, tp3);
+    INSERT_W4_UB(tp0, tp1, tp2, tp3, dst0);
+    LW4(dst + 4 * stride, stride, tp0, tp1, tp2, tp3);
+    INSERT_W4_UB(tp0, tp1, tp2, tp3, dst1);
+    ILVR_B4_UB(src1, src0, src2, src1, src3, src2, src4, src3, tmp0, tmp1, tmp2,
+               tmp3);
+    ILVR_B4_UB(src5, src4, src6, src5, src7, src6, src8, src7, tmp4, tmp5, tmp6,
+               tmp7);
+    ILVR_D2_UB(tmp1, tmp0, tmp3, tmp2, tmp0, tmp2);
+    ILVR_D2_UB(tmp5, tmp4, tmp7, tmp6, tmp4, tmp6);
+    DOTP_UB2_UH(tmp0, tmp2, coeff_vec, coeff_vec, res0, res1);
+    DOTP_UB2_UH(tmp4, tmp6, coeff_vec, coeff_vec, res2, res3);
+    SLLI_4V(res0, res1, res2, res3, 3);
+    SRARI_H4_UH(res0, res1, res2, res3, 6);
+    SAT_UH4_UH(res0, res1, res2, res3, 7);
+    PCKEV_B2_UB(res1, res0, res3, res2, out0, out1);
+    AVER_UB2_UB(out0, dst0, out1, dst1, out0, out1);
+    ST4x8_UB(out0, out1, dst, stride);
 }
 
-static void avc_chroma_vt_and_aver_dst_4w_msa(uint8_t *src, int32_t src_stride,
-                                              uint8_t *dst, int32_t dst_stride,
-                                              uint32_t coeff0, uint32_t coeff1,
-                                              int32_t height)
+static void avc_chroma_vt_and_aver_dst_4w_msa(uint8_t *src, uint8_t *dst,
+                                              int32_t stride, uint32_t coeff0,
+                                              uint32_t coeff1, int32_t 
+height)
 {
     if (2 == height) {
-        avc_chroma_vt_and_aver_dst_4x2_msa(src, src_stride, dst, dst_stride,
-                                           coeff0, coeff1);
-    } else {
-        avc_chroma_vt_and_aver_dst_4x4mul_msa(src, src_stride, dst, dst_stride,
-                                              coeff0, coeff1, height);
+        avc_chroma_vt_and_aver_dst_4x2_msa(src, dst, stride, coeff0, coeff1);
+    } else if (4 == height) {
+        avc_chroma_vt_and_aver_dst_4x4_msa(src, dst, stride, coeff0, coeff1);
+    } else if (8 == height) {
+        avc_chroma_vt_and_aver_dst_4x8_msa(src, dst, stride, coeff0, 
+ coeff1);
     }
 }
 
-static void avc_chroma_vt_and_aver_dst_8w_msa(uint8_t *src, int32_t src_stride,
-                                              uint8_t *dst, int32_t dst_stride,
-                                              uint32_t coeff0, uint32_t coeff1,
-                                              int32_t height)
+static void avc_chroma_vt_and_aver_dst_8x4_msa(uint8_t *src, uint8_t *dst,
+                                               int32_t stride, uint32_t coeff0,
+                                               uint32_t coeff1)
 {
-    uint32_t row;
+    uint64_t tp0, tp1, tp2, tp3;
     v16u8 src0, src1, src2, src3, src4;
     v16u8 out0, out1;
     v8u16 res0, res1, res2, res3;
-    v16u8 dst0, dst1, dst2, dst3;
+    v16u8 dst0 = { 0 }, dst1 = { 0 };
     v16i8 coeff_vec0 = __msa_fill_b(coeff0);
     v16i8 coeff_vec1 = __msa_fill_b(coeff1);
     v16u8 coeff_vec = (v16u8) __msa_ilvr_b(coeff_vec0, coeff_vec1);
 
-    src0 = LD_UB(src);
-    src += src_stride;
+    LD_UB5(src, stride, src0, src1, src2, src3, src4);
+    LD4(dst, stride, tp0, tp1, tp2, tp3);
+    INSERT_D2_UB(tp0, tp1, dst0);
+    INSERT_D2_UB(tp2, tp3, dst1);
+    ILVR_B4_UB(src1, src0, src2, src1, src3, src2, src4, src3,
+               src0, src1, src2, src3);
+    DOTP_UB4_UH(src0, src1, src2, src3, coeff_vec, coeff_vec, coeff_vec,
+                coeff_vec, res0, res1, res2, res3);
+    SLLI_4V(res0, res1, res2, res3, 3);
+    SRARI_H4_UH(res0, res1, res2, res3, 6);
+    SAT_UH4_UH(res0, res1, res2, res3, 7);
+    PCKEV_B2_UB(res1, res0, res3, res2, out0, out1);
+    AVER_UB2_UB(out0, dst0, out1, dst1, out0, out1);
+    ST8x4_UB(out0, out1, dst, stride);
+}
 
-    for (row = height >> 2; row--;) {
-        LD_UB4(src, src_stride, src1, src2, src3, src4);
-        src += (4 * src_stride);
-        LD_UB4(dst, dst_stride, dst0, dst1, dst2, dst3);
-        ILVR_B4_UB(src1, src0, src2, src1, src3, src2, src4, src3,
-                   src0, src1, src2, src3);
-        DOTP_UB4_UH(src0, src1, src2, src3, coeff_vec, coeff_vec, coeff_vec,
-                    coeff_vec, res0, res1, res2, res3);
-        SLLI_4V(res0, res1, res2, res3, 3);
-        SRARI_H4_UH(res0, res1, res2, res3, 6);
-        SAT_UH4_UH(res0, res1, res2, res3, 7);
-        PCKEV_B2_UB(res1, res0, res3, res2, out0, out1);
-        PCKEV_D2_UB(dst1, dst0, dst3, dst2, dst0, dst1);
-        AVER_UB2_UB(out0, dst0, out1, dst1, out0, out1);
-        ST8x4_UB(out0, out1, dst, dst_stride);
+static void avc_chroma_vt_and_aver_dst_8x8_msa(uint8_t *src, uint8_t *dst,
+                                               int32_t stride, uint32_t coeff0,
+                                               uint32_t coeff1) {
+    uint64_t tp0, tp1, tp2, tp3;
+    v16u8 src0, src1, src2, src3, src4, src5, src6, src7, src8;
+    v16u8 out0, out1, out2, out3;
+    v16u8 dst0 = { 0 }, dst1 = { 0 }, dst2 = { 0 }, dst3 = { 0 };
+    v8u16 res0, res1, res2, res3, res4, res5, res6, res7;
+    v16i8 coeff_vec0 = __msa_fill_b(coeff0);
+    v16i8 coeff_vec1 = __msa_fill_b(coeff1);
+    v16u8 coeff_vec = (v16u8) __msa_ilvr_b(coeff_vec0, coeff_vec1);
 
-        dst += (4 * dst_stride);
-        src0 = src4;
+    LD_UB5(src, stride, src0, src1, src2, src3, src4);
+    src += (5 * stride);
+    LD_UB4(src, stride, src5, src6, src7, src8);
+    LD4(dst, stride, tp0, tp1, tp2, tp3);
+    INSERT_D2_UB(tp0, tp1, dst0);
+    INSERT_D2_UB(tp2, tp3, dst1);
+    LD4(dst + 4 * stride, stride, tp0, tp1, tp2, tp3);
+    INSERT_D2_UB(tp0, tp1, dst2);
+    INSERT_D2_UB(tp2, tp3, dst3);
+    ILVR_B4_UB(src1, src0, src2, src1, src3, src2, src4, src3,
+               src0, src1, src2, src3);
+    ILVR_B4_UB(src5, src4, src6, src5, src7, src6, src8, src7,
+               src4, src5, src6, src7);
+    DOTP_UB4_UH(src0, src1, src2, src3, coeff_vec, coeff_vec, coeff_vec,
+                coeff_vec, res0, res1, res2, res3);
+    DOTP_UB4_UH(src4, src5, src6, src7, coeff_vec, coeff_vec, coeff_vec,
+                coeff_vec, res4, res5, res6, res7);
+    SLLI_4V(res0, res1, res2, res3, 3);
+    SLLI_4V(res4, res5, res6, res7, 3);
+    SRARI_H4_UH(res0, res1, res2, res3, 6);
+    SRARI_H4_UH(res4, res5, res6, res7, 6);
+    SAT_UH4_UH(res0, res1, res2, res3, 7);
+    SAT_UH4_UH(res0, res1, res2, res3, 7);
+    PCKEV_B2_UB(res1, res0, res3, res2, out0, out1);
+    PCKEV_B2_UB(res5, res4, res7, res6, out2, out3);
+    AVER_UB2_UB(out0, dst0, out1, dst1, out0, out1);
+    AVER_UB2_UB(out2, dst2, out3, dst3, out2, out3);
+    ST8x8_UB(out0, out1, out2, out3, dst, stride); }
+
+static void avc_chroma_vt_and_aver_dst_8w_msa(uint8_t *src, uint8_t *dst,
+                                              int32_t stride, uint32_t coeff0,
+                                              uint32_t coeff1, int32_t 
+height) {
+    if (4 == height) {
+        avc_chroma_vt_and_aver_dst_8x4_msa(src, dst, stride, coeff0, coeff1);
+    } else if (8 == height) {
+        avc_chroma_vt_and_aver_dst_8x8_msa(src, dst, stride, coeff0, 
+coeff1);
     }
 }
 
@@ -1771,160 +1749,115 @@ static void avc_chroma_hv_and_aver_dst_8w_msa(uint8_t *src, int32_t src_stride,
     }
 }
 
-static void copy_width8_msa(uint8_t *src, int32_t src_stride,
-                            uint8_t *dst, int32_t dst_stride,
+static void copy_width4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
                             int32_t height)  {
-    int32_t cnt;
-    uint64_t out0, out1, out2, out3, out4, out5, out6, out7;
-    v16u8 src0, src1, src2, src3, src4, src5, src6, src7;
+    uint32_t tp0, tp1, tp2, tp3, tp4, tp5, tp6, tp7;
 
-    if (0 == height % 12) {
-        for (cnt = (height / 12); cnt--;) {
-            LD_UB8(src, src_stride,
-                   src0, src1, src2, src3, src4, src5, src6, src7);
-            src += (8 * src_stride);
-
-            out0 = __msa_copy_u_d((v2i64) src0, 0);
-            out1 = __msa_copy_u_d((v2i64) src1, 0);
-            out2 = __msa_copy_u_d((v2i64) src2, 0);
-            out3 = __msa_copy_u_d((v2i64) src3, 0);
-            out4 = __msa_copy_u_d((v2i64) src4, 0);
-            out5 = __msa_copy_u_d((v2i64) src5, 0);
-            out6 = __msa_copy_u_d((v2i64) src6, 0);
-            out7 = __msa_copy_u_d((v2i64) src7, 0);
-
-            SD4(out0, out1, out2, out3, dst, dst_stride);
-            dst += (4 * dst_stride);
-            SD4(out4, out5, out6, out7, dst, dst_stride);
-            dst += (4 * dst_stride);
-
-            LD_UB4(src, src_stride, src0, src1, src2, src3);
-            src += (4 * src_stride);
-
-            out0 = __msa_copy_u_d((v2i64) src0, 0);
-            out1 = __msa_copy_u_d((v2i64) src1, 0);
-            out2 = __msa_copy_u_d((v2i64) src2, 0);
-            out3 = __msa_copy_u_d((v2i64) src3, 0);
-
-            SD4(out0, out1, out2, out3, dst, dst_stride);
-            dst += (4 * dst_stride);
-        }
-    } else if (0 == height % 8) {
-        for (cnt = height >> 3; cnt--;) {
-            LD_UB8(src, src_stride,
-                   src0, src1, src2, src3, src4, src5, src6, src7);
-            src += (8 * src_stride);
-
-            out0 = __msa_copy_u_d((v2i64) src0, 0);
-            out1 = __msa_copy_u_d((v2i64) src1, 0);
-            out2 = __msa_copy_u_d((v2i64) src2, 0);
-            out3 = __msa_copy_u_d((v2i64) src3, 0);
-            out4 = __msa_copy_u_d((v2i64) src4, 0);
-            out5 = __msa_copy_u_d((v2i64) src5, 0);
-            out6 = __msa_copy_u_d((v2i64) src6, 0);
-            out7 = __msa_copy_u_d((v2i64) src7, 0);
-
-            SD4(out0, out1, out2, out3, dst, dst_stride);
-            dst += (4 * dst_stride);
-            SD4(out4, out5, out6, out7, dst, dst_stride);
-            dst += (4 * dst_stride);
-        }
-    } else if (0 == height % 4) {
-        for (cnt = (height / 4); cnt--;) {
-            LD_UB4(src, src_stride, src0, src1, src2, src3);
-            src += (4 * src_stride);
-            out0 = __msa_copy_u_d((v2i64) src0, 0);
-            out1 = __msa_copy_u_d((v2i64) src1, 0);
-            out2 = __msa_copy_u_d((v2i64) src2, 0);
-            out3 = __msa_copy_u_d((v2i64) src3, 0);
-
-            SD4(out0, out1, out2, out3, dst, dst_stride);
-            dst += (4 * dst_stride);
-        }
-    } else if (0 == height % 2) {
-        for (cnt = (height / 2); cnt--;) {
-            LD_UB2(src, src_stride, src0, src1);
-            src += (2 * src_stride);
-            out0 = __msa_copy_u_d((v2i64) src0, 0);
-            out1 = __msa_copy_u_d((v2i64) src1, 0);
-
-            SD(out0, dst);
-            dst += dst_stride;
-            SD(out1, dst);
-            dst += dst_stride;
-        }
+    if (8 == height) {
+        LW4(src, stride, tp0, tp1, tp2, tp3);
+        src += 4 * stride;
+        LW4(src, stride, tp4, tp5, tp6, tp7);
+        SW4(tp0, tp1, tp2, tp3, dst, stride);
+        dst += 4 * stride;
+        SW4(tp4, tp5, tp6, tp7, dst, stride);
+    } else if (4 == height) {
+        LW4(src, stride, tp0, tp1, tp2, tp3);
+        SW4(tp0, tp1, tp2, tp3, dst, stride);
+    } else if (2 == height) {
+        LW2(src, stride, tp0, tp1);
+        SW(tp0, dst);
+        dst += stride;
+        SW(tp1, dst);
     }
 }
 
-static void avg_width4_msa(uint8_t *src, int32_t src_stride,
-                           uint8_t *dst, int32_t dst_stride,
-                           int32_t height)
+static void copy_width8_msa(uint8_t *src, uint8_t *dst, int32_t stride,
+                            int32_t height)
 {
-    int32_t cnt;
-    uint32_t out0, out1, out2, out3;
-    v16u8 src0, src1, src2, src3;
-    v16u8 dst0, dst1, dst2, dst3;
-
-    if (0 == (height % 4)) {
-        for (cnt = (height / 4); cnt--;) {
-            LD_UB4(src, src_stride, src0, src1, src2, src3);
-            src += (4 * src_stride);
-
-            LD_UB4(dst, dst_stride, dst0, dst1, dst2, dst3);
+    uint64_t src0, src1, src2, src3, src4, src5, src6, src7;
 
-            AVER_UB4_UB(src0, dst0, src1, dst1, src2, dst2, src3, dst3,
-                        dst0, dst1, dst2, dst3);
-
-            out0 = __msa_copy_u_w((v4i32) dst0, 0);
-            out1 = __msa_copy_u_w((v4i32) dst1, 0);
-            out2 = __msa_copy_u_w((v4i32) dst2, 0);
-            out3 = __msa_copy_u_w((v4i32) dst3, 0);
-            SW4(out0, out1, out2, out3, dst, dst_stride);
-            dst += (4 * dst_stride);
-        }
-    } else if (0 == (height % 2)) {
-        for (cnt = (height / 2); cnt--;) {
-            LD_UB2(src, src_stride, src0, src1);
-            src += (2 * src_stride);
-
-            LD_UB2(dst, dst_stride, dst0, dst1);
-
-            AVER_UB2_UB(src0, dst0, src1, dst1, dst0, dst1);
-
-            out0 = __msa_copy_u_w((v4i32) dst0, 0);
-            out1 = __msa_copy_u_w((v4i32) dst1, 0);
-            SW(out0, dst);
-            dst += dst_stride;
-            SW(out1, dst);
-            dst += dst_stride;
-        }
+    if (8 == height) {
+        LD4(src, stride, src0, src1, src2, src3);
+        src += 4 * stride;
+        LD4(src, stride, src4, src5, src6, src7);
+        SD4(src0, src1, src2, src3, dst, stride);
+        dst += 4 * stride;
+        SD4(src4, src5, src6, src7, dst, stride);
+    } else if (4 == height) {
+        LD4(src, stride, src0, src1, src2, src3);
+        SD4(src0, src1, src2, src3, dst, stride);
     }
 }
 
-static void avg_width8_msa(uint8_t *src, int32_t src_stride,
-                           uint8_t *dst, int32_t dst_stride,
+static void avg_width4_msa(uint8_t *src, uint8_t *dst, int32_t stride,
                            int32_t height)  {
-    int32_t cnt;
-    uint64_t out0, out1, out2, out3;
-    v16u8 src0, src1, src2, src3;
-    v16u8 dst0, dst1, dst2, dst3;
-
-    for (cnt = (height / 4); cnt--;) {
-        LD_UB4(src, src_stride, src0, src1, src2, src3);
-        src += (4 * src_stride);
-        LD_UB4(dst, dst_stride, dst0, dst1, dst2, dst3);
+    uint32_t tp0, tp1, tp2, tp3;
+    v16u8 src0 = { 0 }, src1 = { 0 }, dst0 = { 0 }, dst1 = { 0 };
+
+    if (8 == height) {
+        LW4(src, stride, tp0, tp1, tp2, tp3);
+        src += 4 * stride;
+        INSERT_W4_UB(tp0, tp1, tp2, tp3, src0);
+        LW4(src, stride, tp0, tp1, tp2, tp3);
+        INSERT_W4_UB(tp0, tp1, tp2, tp3, src1);
+        LW4(dst, stride, tp0, tp1, tp2, tp3);
+        INSERT_W4_UB(tp0, tp1, tp2, tp3, dst0);
+        LW4(dst + 4 * stride, stride, tp0, tp1, tp2, tp3);
+        INSERT_W4_UB(tp0, tp1, tp2, tp3, dst1);
+        AVER_UB2_UB(src0, dst0, src1, dst1, dst0, dst1);
+        ST4x8_UB(dst0, dst1, dst, stride);
+    } else if (4 == height) {
+        LW4(src, stride, tp0, tp1, tp2, tp3);
+        INSERT_W4_UB(tp0, tp1, tp2, tp3, src0);
+        LW4(dst, stride, tp0, tp1, tp2, tp3);
+        INSERT_W4_UB(tp0, tp1, tp2, tp3, dst0);
+        dst0 = __msa_aver_u_b(src0, dst0);
+        ST4x4_UB(dst0, dst0, 0, 1, 2, 3, dst, stride);
+    } else if (2 == height) {
+        LW2(src, stride, tp0, tp1);
+        INSERT_W2_UB(tp0, tp1, src0);
+        LW2(dst, stride, tp0, tp1);
+        INSERT_W2_UB(tp0, tp1, dst0);
+        dst0 = __msa_aver_u_b(src0, dst0);
+        ST4x2_UB(dst0, dst, stride);
+    }
+}
 
-        AVER_UB4_UB(src0, dst0, src1, dst1, src2, dst2, src3, dst3,
-                    dst0, dst1, dst2, dst3);
+static void avg_width8_msa(uint8_t *src, uint8_t *dst, int32_t stride,
+                           int32_t height) {
+    uint64_t tp0, tp1, tp2, tp3, tp4, tp5, tp6, tp7;
+    v16u8 src0 = { 0 }, src1 = { 0 }, src2 = { 0 }, src3 = { 0 };
+    v16u8 dst0 = { 0 }, dst1 = { 0 }, dst2 = { 0 }, dst3 = { 0 };
 
-        out0 = __msa_copy_u_d((v2i64) dst0, 0);
-        out1 = __msa_copy_u_d((v2i64) dst1, 0);
-        out2 = __msa_copy_u_d((v2i64) dst2, 0);
-        out3 = __msa_copy_u_d((v2i64) dst3, 0);
-        SD4(out0, out1, out2, out3, dst, dst_stride);
-        dst += (4 * dst_stride);
+    if (8 == height) {
+        LD4(src, stride, tp0, tp1, tp2, tp3);
+        src += 4 * stride;
+        LD4(src, stride, tp4, tp5, tp6, tp7);
+        INSERT_D2_UB(tp0, tp1, src0);
+        INSERT_D2_UB(tp2, tp3, src1);
+        INSERT_D2_UB(tp4, tp5, src2);
+        INSERT_D2_UB(tp6, tp7, src3);
+        LD4(dst, stride, tp0, tp1, tp2, tp3);
+        LD4(dst + 4 * stride, stride, tp4, tp5, tp6, tp7);
+        INSERT_D2_UB(tp0, tp1, dst0);
+        INSERT_D2_UB(tp2, tp3, dst1);
+        INSERT_D2_UB(tp4, tp5, dst2);
+        INSERT_D2_UB(tp6, tp7, dst3);
+        AVER_UB4_UB(src0, dst0, src1, dst1, src2, dst2, src3, dst3, dst0, dst1,
+                    dst2, dst3);
+        ST8x8_UB(dst0, dst1, dst2, dst3, dst, stride);
+    } else if (4 == height) {
+        LD4(src, stride, tp0, tp1, tp2, tp3);
+        INSERT_D2_UB(tp0, tp1, src0);
+        INSERT_D2_UB(tp2, tp3, src1);
+        LD4(dst, stride, tp0, tp1, tp2, tp3);
+        INSERT_D2_UB(tp0, tp1, dst0);
+        INSERT_D2_UB(tp2, tp3, dst1);
+        AVER_UB2_UB(src0, dst0, src1, dst1, dst0, dst1);
+        ST8x4_UB(dst0, dst1, dst, stride);
     }
 }
 
@@ -1940,15 +1873,13 @@ void ff_put_h264_chroma_mc8_msa(uint8_t *dst, uint8_t *src,
     } else if (y) {
         avc_chroma_vt_8w_msa(src, dst, stride, y, (8 - y), height);
     } else {
-        copy_width8_msa(src, stride, dst, stride, height);
+        copy_width8_msa(src, dst, stride, height);
     }
 }
 
 void ff_put_h264_chroma_mc4_msa(uint8_t *dst, uint8_t *src,
                                 ptrdiff_t stride, int height, int x, int y)  {
-    int32_t cnt;
-
     av_assert2(x < 8 && y < 8 && x >= 0 && y >= 0);
 
     if (x && y) {
@@ -1958,12 +1889,7 @@ void ff_put_h264_chroma_mc4_msa(uint8_t *dst, uint8_t *src,
     } else if (y) {
         avc_chroma_vt_4w_msa(src, dst, stride, y, (8 - y), height);
     } else {
-        for (cnt = height; cnt--;) {
-            *((uint32_t *) dst) = *((uint32_t *) src);
-
-            src += stride;
-            dst += stride;
-        }
+        copy_width4_msa(src, dst, stride, height);
     }
 }
 
@@ -2003,10 +1929,9 @@ void ff_avg_h264_chroma_mc8_msa(uint8_t *dst, uint8_t *src,
     } else if (x) {
         avc_chroma_hz_and_aver_dst_8w_msa(src, dst, stride, x, (8 - x), height);
     } else if (y) {
-        avc_chroma_vt_and_aver_dst_8w_msa(src, stride, dst,
-                                          stride, y, (8 - y), height);
+        avc_chroma_vt_and_aver_dst_8w_msa(src, dst, stride, y, (8 - y), 
+ height);
     } else {
-        avg_width8_msa(src, stride, dst, stride, height);
+        avg_width8_msa(src, dst, stride, height);
     }
 }
 
@@ -2022,10 +1947,9 @@ void ff_avg_h264_chroma_mc4_msa(uint8_t *dst, uint8_t *src,
     } else if (x) {
         avc_chroma_hz_and_aver_dst_4w_msa(src, dst, stride, x, (8 - x), height);
     } else if (y) {
-        avc_chroma_vt_and_aver_dst_4w_msa(src, stride, dst,
-                                          stride, y, (8 - y), height);
+        avc_chroma_vt_and_aver_dst_4w_msa(src, dst, stride, y, (8 - y), 
+ height);
     } else {
-        avg_width4_msa(src, stride, dst, stride, height);
+        avg_width4_msa(src, dst, stride, height);
     }
 }
 
@@ -2043,8 +1967,7 @@ void ff_avg_h264_chroma_mc2_msa(uint8_t *dst, uint8_t *src,
     } else if (x) {
         avc_chroma_hz_and_aver_dst_2w_msa(src, dst, stride, x, (8 - x), height);
     } else if (y) {
-        avc_chroma_vt_and_aver_dst_2w_msa(src, stride, dst,
-                                          stride, y, (8 - y), height);
+        avc_chroma_vt_and_aver_dst_2w_msa(src, dst, stride, y, (8 - y), 
+ height);
     } else {
         for (cnt = height; cnt--;) {
             dst[0] = (dst[0] + src[0] + 1) >> 1;
--
1.7.9.5

_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-devel

Comments

Michael Niedermayer Oct. 25, 2017, 5:58 p.m. UTC | #1 
On Tue, Oct 24, 2017 at 11:21:32AM +0000, Manojkumar Bhosale wrote:
> LGTM

will apply

thanks

[...]
diff mbox

Patch

diff --git a/libavcodec/mips/h264chroma_msa.c b/libavcodec/mips/h264chroma_msa.c
index 2a54675..a5c3334 100644
--- a/libavcodec/mips/h264chroma_msa.c
+++ b/libavcodec/mips/h264chroma_msa.c
@@ -1124,24 +1124,25 @@  static void avc_chroma_hz_and_aver_dst_8w_msa(uint8_t *src, uint8_t *dst,
     }
 }
 
-static void avc_chroma_vt_and_aver_dst_2x2_msa(uint8_t *src, int32_t src_stride,
-                                               uint8_t *dst, int32_t dst_stride,
-                                               uint32_t coeff0, uint32_t coeff1)
+static void avc_chroma_vt_and_aver_dst_2x2_msa(uint8_t *src, uint8_t *dst,
+                                               int32_t stride, uint32_t coeff0,
+                                               uint32_t coeff1)
 {
     uint16_t out0, out1;
-    uint32_t load0, load1;
     v16i8 src0, src1, src2, tmp0, tmp1, res;
     v16u8 dst_data = { 0 };
+    v8i16 out;
     v8u16 res_r;
     v16i8 coeff_vec0 = __msa_fill_b(coeff0);
     v16i8 coeff_vec1 = __msa_fill_b(coeff1);
     v16u8 coeff_vec = (v16u8) __msa_ilvr_b(coeff_vec0, coeff_vec1);
 
-    LD_SB3(src, src_stride, src0, src1, src2);
-    load0 = LW(dst);
-    load1 = LW(dst + dst_stride);
+    LD_SB3(src, stride, src0, src1, src2);
+    out0 = LH(dst);
+    out1 = LH(dst + stride);
 
-    INSERT_W2_UB(load0, load1, dst_data);
+    dst_data = (v16u8) __msa_insert_h((v8i16) dst_data, 0, out0);
+    dst_data = (v16u8) __msa_insert_h((v8i16) dst_data, 2, out1);
 
     ILVR_B2_SB(src1, src0, src2, src1, tmp0, tmp1);
 
@@ -1151,20 +1152,20 @@  static void avc_chroma_vt_and_aver_dst_2x2_msa(uint8_t *src, int32_t src_stride,
     res_r = (v8u16) __msa_srari_h((v8i16) res_r, 6);
     res_r = __msa_sat_u_h(res_r, 7);
     res = __msa_pckev_b((v16i8) res_r, (v16i8) res_r);
-    dst_data = __msa_aver_u_b((v16u8) res, dst_data);
-    out0 = __msa_copy_u_h((v8i16) dst_data, 0);
-    out1 = __msa_copy_u_h((v8i16) dst_data, 2);
+    out = (v8i16) __msa_aver_u_b((v16u8) res, dst_data);
+    out0 = __msa_copy_u_h(out, 0);
+    out1 = __msa_copy_u_h(out, 2);
 
     SH(out0, dst);
-    dst += dst_stride;
+    dst += stride;
     SH(out1, dst);
 }
 
-static void avc_chroma_vt_and_aver_dst_2x4_msa(uint8_t *src, int32_t src_stride,
-                                               uint8_t *dst, int32_t dst_stride,
-                                               uint32_t coeff0, uint32_t coeff1)
+static void avc_chroma_vt_and_aver_dst_2x4_msa(uint8_t *src, uint8_t *dst,
+                                               int32_t stride, uint32_t coeff0,
+                                               uint32_t coeff1)
 {
-    uint32_t load0, load1;
+    uint16_t tp0, tp1, tp2, tp3;
     v16i8 src0, src1, src2, src3, src4;
     v16u8 tmp0, tmp1, tmp2, tmp3;
     v8u16 res_r;
@@ -1174,19 +1175,16 @@  static void avc_chroma_vt_and_aver_dst_2x4_msa(uint8_t *src, int32_t src_stride,
     v16u8 coeff_vec = (v16u8) __msa_ilvr_b(coeff_vec0, coeff_vec1);
     v16u8 dst_data = { 0 };
 
-    LD_SB5(src, src_stride, src0, src1, src2, src3, src4);
-
-    load0 = LW(dst);
-    load1 = LW(dst + dst_stride);
-
-    dst_data = (v16u8) __msa_insert_h((v8i16) dst_data, 0, load0);
-    dst_data = (v16u8) __msa_insert_h((v8i16) dst_data, 1, load1);
+    LD_SB5(src, stride, src0, src1, src2, src3, src4);
 
-    load0 = LW(dst + 2 * dst_stride);
-    load1 = LW(dst + 3 * dst_stride);
-
-    dst_data = (v16u8) __msa_insert_h((v8i16) dst_data, 2, load0);
-    dst_data = (v16u8) __msa_insert_h((v8i16) dst_data, 3, load1);
+    tp0 = LH(dst);
+    tp1 = LH(dst + stride);
+    tp2 = LH(dst + 2 * stride);
+    tp3 = LH(dst + 3 * stride);
+    dst_data = (v16u8) __msa_insert_h((v8i16) dst_data, 0, tp0);
+    dst_data = (v16u8) __msa_insert_h((v8i16) dst_data, 1, tp1);
+    dst_data = (v16u8) __msa_insert_h((v8i16) dst_data, 2, tp2);
+    dst_data = (v16u8) __msa_insert_h((v8i16) dst_data, 3, tp3);
 
     ILVR_B4_UB(src1, src0, src2, src1, src3, src2, src4, src3,
                tmp0, tmp1, tmp2, tmp3); @@ -1202,102 +1200,26 @@ static void avc_chroma_vt_and_aver_dst_2x4_msa(uint8_t *src, int32_t src_stride,