avcodec/x86/h264_qpel: Remove constant parameters from shift5 funcs

They are constant since the size 16 version is no longer emulated
via the size 8 version.

Reviewed-by: James Almer <[email protected]>
Signed-off-by: Andreas Rheinhardt <[email protected]>

2 files changed, 15 insertions, 17 deletions

diff --git a/libavcodec/x86/h264_qpel.c b/libavcodec/x86/h264_qpel.c
index b782d32bea..e6035b8de5 100644
--- a/libavcodec/x86/h264_qpel.c
+++ b/libavcodec/x86/h264_qpel.c
@@ -67,9 +67,9 @@ void ff_ ## OPNAME ## _h264_qpel4_hv_lowpass_h_mmxext(int16_t *tmp, uint8_t *dst
 void ff_ ## OPNAME ## _h264_qpel8or16_hv1_lowpass_op_sse2(const uint8_t *src, int16_t *tmp, int srcStride, int size);\
 void ff_ ## OPNAME ## _h264_qpel8or16_hv2_lowpass_op_mmxext(uint8_t *dst, int16_t *tmp, int dstStride, int unused, int h);\
 void ff_ ## OPNAME ## _h264_qpel8or16_hv2_lowpass_ssse3(uint8_t *dst, int16_t *tmp, int dstStride, int tmpStride, int size);\
-void ff_ ## OPNAME ## _pixels4_l2_shift5_mmxext(uint8_t *dst, const int16_t *src16, const uint8_t *src8, int dstStride, int src8Stride, int h);\
-void ff_ ## OPNAME ## _pixels8_l2_shift5_mmxext(uint8_t *dst, const int16_t *src16, const uint8_t *src8, int dstStride, int src8Stride, int h);\
-void ff_ ## OPNAME ## _pixels16_l2_shift5_sse2(uint8_t *dst, const int16_t *src16, const uint8_t *src8, int dstStride, int src8Stride, int h);\
+void ff_ ## OPNAME ## _pixels4_l2_shift5_mmxext(uint8_t *dst, const int16_t *src16, const uint8_t *src8, int dstStride);\
+void ff_ ## OPNAME ## _pixels8_l2_shift5_mmxext(uint8_t *dst, const int16_t *src16, const uint8_t *src8, int dstStride);\
+void ff_ ## OPNAME ## _pixels16_l2_shift5_sse2(uint8_t *dst, const int16_t *src16, const uint8_t *src8, int dstStride);\
 
 DEF_QPEL(avg)
 DEF_QPEL(put)
@@ -309,7 +309,7 @@ static void OPNAME ## h264_qpel ## SIZE ## _mc12_ ## MMX(uint8_t *dst, const uin
     int16_t * const halfV= (int16_t*)(temp + SIZE*SIZE);\
     av_assert2(((uintptr_t)temp & 7) == 0);\
     ff_put_h264_qpel ## SIZE ## _hv_lowpass_ ## MMX(halfHV, halfV, src, SIZE, SIZE, stride);\
-    ff_ ## OPNAME ## pixels ## SIZE ## _l2_shift5_mmxext(dst, halfV+2, halfHV, stride, SIZE, SIZE);\
+    ff_ ## OPNAME ## pixels ## SIZE ## _l2_shift5_mmxext(dst, halfV+2, halfHV, stride);\
 }\
 \
 static void OPNAME ## h264_qpel ## SIZE ## _mc32_ ## MMX(uint8_t *dst, const uint8_t *src, ptrdiff_t stride)\
@@ -319,7 +319,7 @@ static void OPNAME ## h264_qpel ## SIZE ## _mc32_ ## MMX(uint8_t *dst, const uin
     int16_t * const halfV= (int16_t*)(temp + SIZE*SIZE);\
     av_assert2(((uintptr_t)temp & 7) == 0);\
     ff_put_h264_qpel ## SIZE ## _hv_lowpass_ ## MMX(halfHV, halfV, src, SIZE, SIZE, stride);\
-    ff_ ## OPNAME ## pixels ## SIZE ## _l2_shift5_mmxext(dst, halfV+3, halfHV, stride, SIZE, SIZE);\
+    ff_ ## OPNAME ## pixels ## SIZE ## _l2_shift5_mmxext(dst, halfV+3, halfHV, stride);\
 }\
 
 #define H264_MC(QPEL, SIZE, MMX, ALIGN)\
diff --git a/libavcodec/x86/h264_qpel_8bit.asm b/libavcodec/x86/h264_qpel_8bit.asm
index fefa3aff01..7dbdf41d0a 100644
--- a/libavcodec/x86/h264_qpel_8bit.asm
+++ b/libavcodec/x86/h264_qpel_8bit.asm
@@ -701,9 +701,8 @@ QPEL8OR16_HV2_LOWPASS_OP_XMM avg
 
 
 %macro PIXELS4_L2_SHIFT5 1
-cglobal %1_pixels4_l2_shift5,6,6 ; dst, src16, src8, dstStride, src8Stride, h
+cglobal %1_pixels4_l2_shift5,4,4 ; dst, src16, src8, dstStride
     movsxdifnidn  r3, r3d
-    movsxdifnidn  r4, r4d
     mova          m0, [r1]
     mova          m1, [r1+24]
     psraw         m0, 5
@@ -711,10 +710,9 @@ cglobal %1_pixels4_l2_shift5,6,6 ; dst, src16, src8, dstStride, src8Stride, h
     packuswb      m0, m0
     packuswb      m1, m1
     pavgb         m0, [r2]
-    pavgb         m1, [r2+r4]
+    pavgb         m1, [r2+4]
     op_%1h        m0, [r0], m4
     op_%1h        m1, [r0+r3], m5
-    lea           r2, [r2+r4*2]
     lea           r0, [r0+r3*2]
     mova          m0, [r1+48]
     mova          m1, [r1+72]
@@ -722,8 +720,8 @@ cglobal %1_pixels4_l2_shift5,6,6 ; dst, src16, src8, dstStride, src8Stride, h
     psraw         m1, 5
     packuswb      m0, m0
     packuswb      m1, m1
-    pavgb         m0, [r2]
-    pavgb         m1, [r2+r4]
+    pavgb         m0, [r2+2*4]
+    pavgb         m1, [r2+3*4]
     op_%1h        m0, [r0], m4
     op_%1h        m1, [r0+r3], m5
     RET
@@ -736,12 +734,12 @@ PIXELS4_L2_SHIFT5 avg
 
 %macro PIXELS_L2_SHIFT5 2
 %if cpuflag(sse2)
-cglobal %1_pixels%2_l2_shift5, 6, 6, 4 ; dst, src16, src8, dstStride, src8Stride, h
+cglobal %1_pixels%2_l2_shift5, 5, 5, 4 ; dst, src16, src8, dstStride
 %else
-cglobal %1_pixels%2_l2_shift5, 6, 6 ; dst, src16, src8, dstStride, src8Stride, h
+cglobal %1_pixels%2_l2_shift5, 5, 5 ; dst, src16, src8, dstStride
 %endif
     movsxdifnidn  r3, r3d
-    movsxdifnidn  r4, r4d
+    mov          r4d, %2
 .loop:
     movu          m0, [r1]
     movu          m1, [r1+%2]
@@ -754,13 +752,13 @@ cglobal %1_pixels%2_l2_shift5, 6, 6 ; dst, src16, src8, dstStride, src8Stride, h
     packuswb      m0, m1
     packuswb      m2, m3
     pavgb         m0, [r2]
-    pavgb         m2, [r2+r4]
+    pavgb         m2, [r2+%2]
     op_%1         m0, [r0], m1
     op_%1         m2, [r0+r3], m1
-    lea           r2, [r2+2*r4]
+    lea           r2, [r2+2*%2]
     add           r1, 48*2
     lea           r0, [r0+2*r3]
-    sub          r5d, 2
+    sub          r4d, 2
     jne        .loop
     RET
 %endmacro