lavc/h264dsp: factor some mostly identical R-V V code

libavcodec/riscv/h264idct_rvv.S

@@ -418,8 +418,8 @@ const ff_h264_scan8
 endconst
 
 #if (__riscv_xlen == 64)
-.irp    depth, 8, 16
-func ff_h264_idct_add16_\depth\()_rvv, zve32x
+.macro  idct4_adds type, depth
+func ff_h264_idct_add\type\()_\depth\()_rvv, zve32x
         csrwi   vxrm, 0
         addi    sp, sp, -96
         lla     t0, ff_h264_scan8
@@ -455,9 +455,13 @@ func ff_h264_idct_add16_\depth\()_rvv, zve32x
 .endif
         vmsne.vi  v1, v16, 0
         vsetvli   zero, zero, e8, m1, ta, ma
+.ifc \type, 16
         vmseq.vi  v2, v12, 1
+.endif
         vmsne.vi  v0, v12, 0
+.ifc \type, 16
         vmand.mm  v1, v1, v2
+.endif
         vsetvli   zero, zero, e16, m2, ta, ma
         vmv.x.s   s2, v0
         vmv.x.s   s3, v1
@@ -470,96 +474,9 @@ func ff_h264_idct_add16_\depth\()_rvv, zve32x
         andi    t0, s2, 1
         addi    s1, s1, -1
         srli    s2, s2, 1
+.ifc \type, 16
         beqz    t0, 3f     # if (nnz)
-        lw      t2, (s5)   # block_offset[i]
-        andi    t1, s3, 1
-        mv      a1, s6
-        mv      a2, s7
-        add     a0, s4, t2
-.if \depth > 8
-        mv      a5, s8
 .endif
-        bnez    t1, 2f    # if (nnz == 1 && block[i * 16])
-        jal     .Lidct_add4_\depth\()_rvv
-        j       3f
-2:
-.if \depth == 8
-        call    ff_h264_idct_dc_add_\depth\()_c
-.else
-        jalr    s9
-.endif
-3:
-        srli    s3, s3, 1
-        addi    s5, s5, 4
-        addi    s6, s6, 16 * 2 * (\depth / 8)
-        bnez    s1, 1b
-
-.if \depth > 8
-        ld      s9, 80(sp)
-        ld      s8, 72(sp)
-.endif
-        ld      s7, 64(sp)
-        ld      s6, 56(sp)
-        ld      s5, 48(sp)
-        ld      s4, 40(sp)
-        ld      s3, 32(sp)
-        ld      s2, 24(sp)
-        ld      s1, 16(sp)
-        ld      ra,  8(sp)
-        ld      s0,  0(sp)
-        addi    sp, sp, 96
-        ret
-endfunc
-
-func ff_h264_idct_add16intra_\depth\()_rvv, zve32x
-        csrwi   vxrm, 0
-        addi    sp, sp, -96
-        lla     t0, ff_h264_scan8
-        sd      s0,   (sp)
-        li      t1, 32 * (\depth / 8)
-        mv      s0, sp
-        sd      ra,  8(sp)
-        sd      s1, 16(sp)
-        sd      s2, 24(sp)
-        sd      s3, 32(sp)
-        sd      s4, 40(sp)
-        sd      s5, 48(sp)
-        sd      s6, 56(sp)
-        sd      s7, 64(sp)
-.if \depth > 8
-        sd      s8, 72(sp)
-        sd      s9, 80(sp)
-        mv      s8, a5
-        mv      s9, a6
-.endif
-        vsetivli  zero, 16, e8, m1, ta, ma
-        vle8.v    v8, (t0)
-.if \depth == 8
-        vlse16.v  v16, (a2), t1
-.else
-        vlse32.v  v16, (a2), t1
-.endif
-        vluxei8.v v12, (a4), v8
-.if \depth == 8
-        vsetvli   zero, zero, e16, m2, ta, ma
-.else
-        vsetvli   zero, zero, e32, m4, ta, ma
-.endif
-        vmsne.vi  v1, v16, 0
-        vsetvli   zero, zero, e8, m1, ta, ma
-        vmsne.vi  v0, v12, 0
-        vsetvli   zero, zero, e16, m2, ta, ma
-        vmv.x.s   s2, v0
-        vmv.x.s   s3, v1
-        li      s1, 16
-        mv      s4, a0
-        mv      s5, a1
-        mv      s6, a2
-        mv      s7, a3
-1:
-        andi    t0, s2, 1
-        addi    s1, s1, -1
-        srli    s2, s2, 1
         lw      t2, (s5)   # block_offset[i]
         andi    t1, s3, 1
         mv      a1, s6
@@ -568,11 +485,17 @@ func ff_h264_idct_add16intra_\depth\()_rvv, zve32x
 .if \depth > 8
         mv      a5, s8
 .endif
+.ifc \type, 16
+        bnez    t1, 2f     # if (nnz == 1 && block[i * 16])
+.else
         beqz    t0, 2f     # if (nnzc[scan8[i]])
+.endif
         jal     .Lidct_add4_\depth\()_rvv
         j       3f
 2:
+.ifnc \type, 16
         beqz    t1, 3f    # if (block[i * 16])
+.endif
 .if \depth == 8
         call    ff_h264_idct_dc_add_\depth\()_c
 .else
@@ -600,6 +523,11 @@ func ff_h264_idct_add16intra_\depth\()_rvv, zve32x
         addi    sp, sp, 96
         ret
 endfunc
+.endm
+
+.irp    depth, 8, 16
+idct4_adds 16, \depth
+idct4_adds 16intra, \depth
 
 func ff_h264_idct8_add4_\depth\()_rvv, zve32x
         csrwi       vxrm, 0