x86: vc1dsp: Convert vc1_inv_trans_*_dc to NASM format

3 files changed, 111 insertions, 207 deletions

diff --git a/libavcodec/x86/vc1dsp.asm b/libavcodec/x86/vc1dsp.asm
index 6415a83..91a1991 100644
--- a/libavcodec/x86/vc1dsp.asm
+++ b/libavcodec/x86/vc1dsp.asm
@@ -395,3 +395,101 @@ cglobal vc1_put_ver_16b_shift2, 4,7,0, dst, src, stride
         jnz         .loop
     REP_RET
 %endif ; HAVE_MMX_INLINE
+
+%macro INV_TRANS_INIT 0
+    movsxdifnidn linesizeq, linesized
+    movd       m0, blockd
+    SPLATW     m0, m0
+    pxor       m1, m1
+    psubw      m1, m0
+    packuswb   m0, m0
+    packuswb   m1, m1
+
+    DEFINE_ARGS dest, linesize, linesize3
+    lea    linesize3q, [linesizeq*3]
+%endmacro
+
+%macro INV_TRANS_PROCESS 1
+    mov%1                  m2, [destq+linesizeq*0]
+    mov%1                  m3, [destq+linesizeq*1]
+    mov%1                  m4, [destq+linesizeq*2]
+    mov%1                  m5, [destq+linesize3q]
+    paddusb                m2, m0
+    paddusb                m3, m0
+    paddusb                m4, m0
+    paddusb                m5, m0
+    psubusb                m2, m1
+    psubusb                m3, m1
+    psubusb                m4, m1
+    psubusb                m5, m1
+    mov%1 [linesizeq*0+destq], m2
+    mov%1 [linesizeq*1+destq], m3
+    mov%1 [linesizeq*2+destq], m4
+    mov%1 [linesize3q +destq], m5
+%endmacro
+
+; ff_vc1_inv_trans_?x?_dc_mmxext(uint8_t *dest, int linesize, int16_t *block)
+INIT_MMX mmxext
+cglobal vc1_inv_trans_4x4_dc, 3,4,0, dest, linesize, block
+    movsx         r3d, WORD [blockq]
+    mov        blockd, r3d             ; dc
+    shl        blockd, 4               ; 16 * dc
+    lea        blockd, [blockq+r3+4]   ; 17 * dc + 4
+    sar        blockd, 3               ; >> 3
+    mov           r3d, blockd          ; dc
+    shl        blockd, 4               ; 16 * dc
+    lea        blockd, [blockq+r3+64]  ; 17 * dc + 64
+    sar        blockd, 7               ; >> 7
+
+    INV_TRANS_INIT
+
+    INV_TRANS_PROCESS h
+    RET
+
+INIT_MMX mmxext
+cglobal vc1_inv_trans_4x8_dc, 3,4,0, dest, linesize, block
+    movsx         r3d, WORD [blockq]
+    mov        blockd, r3d             ; dc
+    shl        blockd, 4               ; 16 * dc
+    lea        blockd, [blockq+r3+4]   ; 17 * dc + 4
+    sar        blockd, 3               ; >> 3
+    shl        blockd, 2               ;  4 * dc
+    lea        blockd, [blockq*3+64]   ; 12 * dc + 64
+    sar        blockd, 7               ; >> 7
+
+    INV_TRANS_INIT
+
+    INV_TRANS_PROCESS h
+    lea         destq, [destq+linesizeq*4]
+    INV_TRANS_PROCESS h
+    RET
+
+INIT_MMX mmxext
+cglobal vc1_inv_trans_8x4_dc, 3,4,0, dest, linesize, block
+    movsx      blockd, WORD [blockq]   ; dc
+    lea        blockd, [blockq*3+1]    ;  3 * dc + 1
+    sar        blockd, 1               ; >> 1
+    mov           r3d, blockd          ; dc
+    shl        blockd, 4               ; 16 * dc
+    lea        blockd, [blockq+r3+64]  ; 17 * dc + 64
+    sar        blockd, 7               ; >> 7
+
+    INV_TRANS_INIT
+
+    INV_TRANS_PROCESS a
+    RET
+
+INIT_MMX mmxext
+cglobal vc1_inv_trans_8x8_dc, 3,3,0, dest, linesize, block
+    movsx      blockd, WORD [blockq]   ; dc
+    lea        blockd, [blockq*3+1]    ;  3 * dc + 1
+    sar        blockd, 1               ; >> 1
+    lea        blockd, [blockq*3+16]   ;  3 * dc + 16
+    sar        blockd, 5               ; >> 5
+
+    INV_TRANS_INIT
+
+    INV_TRANS_PROCESS a
+    lea         destq, [destq+linesizeq*4]
+    INV_TRANS_PROCESS a
+    RET
diff --git a/libavcodec/x86/vc1dsp_init.c b/libavcodec/x86/vc1dsp_init.c
index 1747305..c8943fa 100644
--- a/libavcodec/x86/vc1dsp_init.c
+++ b/libavcodec/x86/vc1dsp_init.c
@@ -92,6 +92,14 @@ void ff_put_vc1_chroma_mc8_nornd_ssse3(uint8_t *dst, uint8_t *src,
                                        int stride, int h, int x, int y);
 void ff_avg_vc1_chroma_mc8_nornd_ssse3(uint8_t *dst, uint8_t *src,
                                        int stride, int h, int x, int y);
+void ff_vc1_inv_trans_4x4_dc_mmxext(uint8_t *dest, int linesize,
+                                    int16_t *block);
+void ff_vc1_inv_trans_4x8_dc_mmxext(uint8_t *dest, int linesize,
+                                    int16_t *block);
+void ff_vc1_inv_trans_8x4_dc_mmxext(uint8_t *dest, int linesize,
+                                    int16_t *block);
+void ff_vc1_inv_trans_8x8_dc_mmxext(uint8_t *dest, int linesize,
+                                    int16_t *block);
 
 
 av_cold void ff_vc1dsp_init_x86(VC1DSPContext *dsp)
@@ -130,6 +138,11 @@ av_cold void ff_vc1dsp_init_x86(VC1DSPContext *dsp)
 
         dsp->avg_vc1_mspel_pixels_tab[1][0]      = avg_vc1_mspel_mc00_8_mmxext;
         dsp->avg_vc1_mspel_pixels_tab[0][0]      = avg_vc1_mspel_mc00_16_mmxext;
+
+        dsp->vc1_inv_trans_8x8_dc                = ff_vc1_inv_trans_8x8_dc_mmxext;
+        dsp->vc1_inv_trans_4x8_dc                = ff_vc1_inv_trans_4x8_dc_mmxext;
+        dsp->vc1_inv_trans_8x4_dc                = ff_vc1_inv_trans_8x4_dc_mmxext;
+        dsp->vc1_inv_trans_4x4_dc                = ff_vc1_inv_trans_4x4_dc_mmxext;
     }
     if (EXTERNAL_SSE2(cpu_flags)) {
         dsp->vc1_v_loop_filter8  = ff_vc1_v_loop_filter8_sse2;
diff --git a/libavcodec/x86/vc1dsp_mmx.c b/libavcodec/x86/vc1dsp_mmx.c
index c268cc6..ff13d9b 100644
--- a/libavcodec/x86/vc1dsp_mmx.c
+++ b/libavcodec/x86/vc1dsp_mmx.c
@@ -481,208 +481,6 @@ DECLARE_FUNCTION(3, 1)
 DECLARE_FUNCTION(3, 2)
 DECLARE_FUNCTION(3, 3)
 
-static void vc1_inv_trans_4x4_dc_mmxext(uint8_t *dest, int linesize,
-                                        int16_t *block)
-{
-    int dc = block[0];
-    dc = (17 * dc +  4) >> 3;
-    dc = (17 * dc + 64) >> 7;
-    __asm__ volatile(
-        "movd          %0, %%mm0 \n\t"
-        "pshufw $0, %%mm0, %%mm0 \n\t"
-        "pxor       %%mm1, %%mm1 \n\t"
-        "psubw      %%mm0, %%mm1 \n\t"
-        "packuswb   %%mm0, %%mm0 \n\t"
-        "packuswb   %%mm1, %%mm1 \n\t"
-        ::"r"(dc)
-    );
-    __asm__ volatile(
-        "movd          %0, %%mm2 \n\t"
-        "movd          %1, %%mm3 \n\t"
-        "movd          %2, %%mm4 \n\t"
-        "movd          %3, %%mm5 \n\t"
-        "paddusb    %%mm0, %%mm2 \n\t"
-        "paddusb    %%mm0, %%mm3 \n\t"
-        "paddusb    %%mm0, %%mm4 \n\t"
-        "paddusb    %%mm0, %%mm5 \n\t"
-        "psubusb    %%mm1, %%mm2 \n\t"
-        "psubusb    %%mm1, %%mm3 \n\t"
-        "psubusb    %%mm1, %%mm4 \n\t"
-        "psubusb    %%mm1, %%mm5 \n\t"
-        "movd       %%mm2, %0    \n\t"
-        "movd       %%mm3, %1    \n\t"
-        "movd       %%mm4, %2    \n\t"
-        "movd       %%mm5, %3    \n\t"
-        :"+m"(*(uint32_t*)(dest+0*linesize)),
-         "+m"(*(uint32_t*)(dest+1*linesize)),
-         "+m"(*(uint32_t*)(dest+2*linesize)),
-         "+m"(*(uint32_t*)(dest+3*linesize))
-    );
-}
-
-static void vc1_inv_trans_4x8_dc_mmxext(uint8_t *dest, int linesize,
-                                        int16_t *block)
-{
-    int dc = block[0];
-    dc = (17 * dc +  4) >> 3;
-    dc = (12 * dc + 64) >> 7;
-    __asm__ volatile(
-        "movd          %0, %%mm0 \n\t"
-        "pshufw $0, %%mm0, %%mm0 \n\t"
-        "pxor       %%mm1, %%mm1 \n\t"
-        "psubw      %%mm0, %%mm1 \n\t"
-        "packuswb   %%mm0, %%mm0 \n\t"
-        "packuswb   %%mm1, %%mm1 \n\t"
-        ::"r"(dc)
-    );
-    __asm__ volatile(
-        "movd          %0, %%mm2 \n\t"
-        "movd          %1, %%mm3 \n\t"
-        "movd          %2, %%mm4 \n\t"
-        "movd          %3, %%mm5 \n\t"
-        "paddusb    %%mm0, %%mm2 \n\t"
-        "paddusb    %%mm0, %%mm3 \n\t"
-        "paddusb    %%mm0, %%mm4 \n\t"
-        "paddusb    %%mm0, %%mm5 \n\t"
-        "psubusb    %%mm1, %%mm2 \n\t"
-        "psubusb    %%mm1, %%mm3 \n\t"
-        "psubusb    %%mm1, %%mm4 \n\t"
-        "psubusb    %%mm1, %%mm5 \n\t"
-        "movd       %%mm2, %0    \n\t"
-        "movd       %%mm3, %1    \n\t"
-        "movd       %%mm4, %2    \n\t"
-        "movd       %%mm5, %3    \n\t"
-        :"+m"(*(uint32_t*)(dest+0*linesize)),
-         "+m"(*(uint32_t*)(dest+1*linesize)),
-         "+m"(*(uint32_t*)(dest+2*linesize)),
-         "+m"(*(uint32_t*)(dest+3*linesize))
-    );
-    dest += 4*linesize;
-    __asm__ volatile(
-        "movd          %0, %%mm2 \n\t"
-        "movd          %1, %%mm3 \n\t"
-        "movd          %2, %%mm4 \n\t"
-        "movd          %3, %%mm5 \n\t"
-        "paddusb    %%mm0, %%mm2 \n\t"
-        "paddusb    %%mm0, %%mm3 \n\t"
-        "paddusb    %%mm0, %%mm4 \n\t"
-        "paddusb    %%mm0, %%mm5 \n\t"
-        "psubusb    %%mm1, %%mm2 \n\t"
-        "psubusb    %%mm1, %%mm3 \n\t"
-        "psubusb    %%mm1, %%mm4 \n\t"
-        "psubusb    %%mm1, %%mm5 \n\t"
-        "movd       %%mm2, %0    \n\t"
-        "movd       %%mm3, %1    \n\t"
-        "movd       %%mm4, %2    \n\t"
-        "movd       %%mm5, %3    \n\t"
-        :"+m"(*(uint32_t*)(dest+0*linesize)),
-         "+m"(*(uint32_t*)(dest+1*linesize)),
-         "+m"(*(uint32_t*)(dest+2*linesize)),
-         "+m"(*(uint32_t*)(dest+3*linesize))
-    );
-}
-
-static void vc1_inv_trans_8x4_dc_mmxext(uint8_t *dest, int linesize,
-                                        int16_t *block)
-{
-    int dc = block[0];
-    dc = ( 3 * dc +  1) >> 1;
-    dc = (17 * dc + 64) >> 7;
-    __asm__ volatile(
-        "movd          %0, %%mm0 \n\t"
-        "pshufw $0, %%mm0, %%mm0 \n\t"
-        "pxor       %%mm1, %%mm1 \n\t"
-        "psubw      %%mm0, %%mm1 \n\t"
-        "packuswb   %%mm0, %%mm0 \n\t"
-        "packuswb   %%mm1, %%mm1 \n\t"
-        ::"r"(dc)
-    );
-    __asm__ volatile(
-        "movq          %0, %%mm2 \n\t"
-        "movq          %1, %%mm3 \n\t"
-        "movq          %2, %%mm4 \n\t"
-        "movq          %3, %%mm5 \n\t"
-        "paddusb    %%mm0, %%mm2 \n\t"
-        "paddusb    %%mm0, %%mm3 \n\t"
-        "paddusb    %%mm0, %%mm4 \n\t"
-        "paddusb    %%mm0, %%mm5 \n\t"
-        "psubusb    %%mm1, %%mm2 \n\t"
-        "psubusb    %%mm1, %%mm3 \n\t"
-        "psubusb    %%mm1, %%mm4 \n\t"
-        "psubusb    %%mm1, %%mm5 \n\t"
-        "movq       %%mm2, %0    \n\t"
-        "movq       %%mm3, %1    \n\t"
-        "movq       %%mm4, %2    \n\t"
-        "movq       %%mm5, %3    \n\t"
-        :"+m"(*(uint32_t*)(dest+0*linesize)),
-         "+m"(*(uint32_t*)(dest+1*linesize)),
-         "+m"(*(uint32_t*)(dest+2*linesize)),
-         "+m"(*(uint32_t*)(dest+3*linesize))
-    );
-}
-
-static void vc1_inv_trans_8x8_dc_mmxext(uint8_t *dest, int linesize,
-                                        int16_t *block)
-{
-    int dc = block[0];
-    dc = (3 * dc +  1) >> 1;
-    dc = (3 * dc + 16) >> 5;
-    __asm__ volatile(
-        "movd          %0, %%mm0 \n\t"
-        "pshufw $0, %%mm0, %%mm0 \n\t"
-        "pxor       %%mm1, %%mm1 \n\t"
-        "psubw      %%mm0, %%mm1 \n\t"
-        "packuswb   %%mm0, %%mm0 \n\t"
-        "packuswb   %%mm1, %%mm1 \n\t"
-        ::"r"(dc)
-    );
-    __asm__ volatile(
-        "movq          %0, %%mm2 \n\t"
-        "movq          %1, %%mm3 \n\t"
-        "movq          %2, %%mm4 \n\t"
-        "movq          %3, %%mm5 \n\t"
-        "paddusb    %%mm0, %%mm2 \n\t"
-        "paddusb    %%mm0, %%mm3 \n\t"
-        "paddusb    %%mm0, %%mm4 \n\t"
-        "paddusb    %%mm0, %%mm5 \n\t"
-        "psubusb    %%mm1, %%mm2 \n\t"
-        "psubusb    %%mm1, %%mm3 \n\t"
-        "psubusb    %%mm1, %%mm4 \n\t"
-        "psubusb    %%mm1, %%mm5 \n\t"
-        "movq       %%mm2, %0    \n\t"
-        "movq       %%mm3, %1    \n\t"
-        "movq       %%mm4, %2    \n\t"
-        "movq       %%mm5, %3    \n\t"
-        :"+m"(*(uint32_t*)(dest+0*linesize)),
-         "+m"(*(uint32_t*)(dest+1*linesize)),
-         "+m"(*(uint32_t*)(dest+2*linesize)),
-         "+m"(*(uint32_t*)(dest+3*linesize))
-    );
-    dest += 4*linesize;
-    __asm__ volatile(
-        "movq          %0, %%mm2 \n\t"
-        "movq          %1, %%mm3 \n\t"
-        "movq          %2, %%mm4 \n\t"
-        "movq          %3, %%mm5 \n\t"
-        "paddusb    %%mm0, %%mm2 \n\t"
-        "paddusb    %%mm0, %%mm3 \n\t"
-        "paddusb    %%mm0, %%mm4 \n\t"
-        "paddusb    %%mm0, %%mm5 \n\t"
-        "psubusb    %%mm1, %%mm2 \n\t"
-        "psubusb    %%mm1, %%mm3 \n\t"
-        "psubusb    %%mm1, %%mm4 \n\t"
-        "psubusb    %%mm1, %%mm5 \n\t"
-        "movq       %%mm2, %0    \n\t"
-        "movq       %%mm3, %1    \n\t"
-        "movq       %%mm4, %2    \n\t"
-        "movq       %%mm5, %3    \n\t"
-        :"+m"(*(uint32_t*)(dest+0*linesize)),
-         "+m"(*(uint32_t*)(dest+1*linesize)),
-         "+m"(*(uint32_t*)(dest+2*linesize)),
-         "+m"(*(uint32_t*)(dest+3*linesize))
-    );
-}
-
 #define FN_ASSIGN(OP, X, Y, INSN) \
     dsp->OP##vc1_mspel_pixels_tab[1][X+4*Y] = OP##vc1_mspel_mc##X##Y##INSN; \
     dsp->OP##vc1_mspel_pixels_tab[0][X+4*Y] = OP##vc1_mspel_mc##X##Y##_16##INSN
@@ -729,10 +527,5 @@ av_cold void ff_vc1dsp_init_mmxext(VC1DSPContext *dsp)
     FN_ASSIGN(avg_, 3, 1, _mmxext);
     FN_ASSIGN(avg_, 3, 2, _mmxext);
     FN_ASSIGN(avg_, 3, 3, _mmxext);
-
-    dsp->vc1_inv_trans_8x8_dc = vc1_inv_trans_8x8_dc_mmxext;
-    dsp->vc1_inv_trans_4x8_dc = vc1_inv_trans_4x8_dc_mmxext;
-    dsp->vc1_inv_trans_8x4_dc = vc1_inv_trans_8x4_dc_mmxext;
-    dsp->vc1_inv_trans_4x4_dc = vc1_inv_trans_4x4_dc_mmxext;
 }
 #endif /* HAVE_6REGS && HAVE_INLINE_ASM && HAVE_MMX_EXTERNAL */