[04/13] x86: qpel: Move fullpel and l2 functions to a separate file

Message ID 1364995179-98359-4-git-send-email-martin@martin.st
State Committed
Headers show

Commit Message

Martin Storsjö April 3, 2013, 1:19 p.m.
From: "Ronald S. Bultje" <rsbultje@gmail.com>

This way, they can be shared between mpeg4qpel and h264qpel without
requiring either one to be compiled unconditionally.
---
 libavcodec/x86/Makefile     |    4 +-
 libavcodec/x86/dsputil.asm  |  192 --------------------------------
 libavcodec/x86/hpeldsp.asm  |   41 -------
 libavcodec/x86/qpelbase.asm |  256 +++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 259 insertions(+), 234 deletions(-)
 create mode 100644 libavcodec/x86/qpelbase.asm

Comments

Luca Barbato April 3, 2013, 1:35 p.m. | #1
On 03/04/13 15:19, Martin Storsjö wrote:
> From: "Ronald S. Bultje" <rsbultje@gmail.com>
> 
> This way, they can be shared between mpeg4qpel and h264qpel without
> requiring either one to be compiled unconditionally.
> ---
>  libavcodec/x86/Makefile     |    4 +-
>  libavcodec/x86/dsputil.asm  |  192 --------------------------------
>  libavcodec/x86/hpeldsp.asm  |   41 -------
>  libavcodec/x86/qpelbase.asm |  256 +++++++++++++++++++++++++++++++++++++++++++
>  4 files changed, 259 insertions(+), 234 deletions(-)
>  create mode 100644 libavcodec/x86/qpelbase.asm
> 
> diff --git a/libavcodec/x86/Makefile b/libavcodec/x86/Makefile
> index a759e6e..7087c4f 100644
> --- a/libavcodec/x86/Makefile
> +++ b/libavcodec/x86/Makefile
> @@ -57,7 +57,8 @@ YASM-OBJS-$(CONFIG_H264DSP)            += x86/h264_deblock.o            \
>  YASM-OBJS-$(CONFIG_H264PRED)           += x86/h264_intrapred.o          \
>                                            x86/h264_intrapred_10bit.o
>  YASM-OBJS-$(CONFIG_H264QPEL)           += x86/h264_qpel_8bit.o          \
> -                                          x86/h264_qpel_10bit.o
> +                                          x86/h264_qpel_10bit.o         \
> +                                          x86/qpelbase.o
>  YASM-OBJS-$(CONFIG_MPEGAUDIODSP)       += x86/imdct36.o
>  YASM-OBJS-$(CONFIG_PNG_DECODER)        += x86/pngdsp.o
>  YASM-OBJS-$(CONFIG_PRORES_DECODER)     += x86/proresdsp.o
> @@ -76,3 +77,4 @@ YASM-OBJS                              += x86/dsputil.o                 \
>                                            x86/fmtconvert.o              \
>                                            x86/hpeldsp.o                 \
>                                            x86/mpeg4qpel.o               \
> +                                          x86/qpelbase.o

something strange here, x86/qpelbase gets compiled unconditionally AND
when h264qpel is needed? I'd drop the h264qpel duplicate.

lu
Martin Storsjö April 3, 2013, 1:43 p.m. | #2
On Wed, 3 Apr 2013, Luca Barbato wrote:

> On 03/04/13 15:19, Martin Storsjö wrote:
>> From: "Ronald S. Bultje" <rsbultje@gmail.com>
>> 
>> This way, they can be shared between mpeg4qpel and h264qpel without
>> requiring either one to be compiled unconditionally.
>> ---
>>  libavcodec/x86/Makefile     |    4 +-
>>  libavcodec/x86/dsputil.asm  |  192 --------------------------------
>>  libavcodec/x86/hpeldsp.asm  |   41 -------
>>  libavcodec/x86/qpelbase.asm |  256 +++++++++++++++++++++++++++++++++++++++++++
>>  4 files changed, 259 insertions(+), 234 deletions(-)
>>  create mode 100644 libavcodec/x86/qpelbase.asm
>> 
>> diff --git a/libavcodec/x86/Makefile b/libavcodec/x86/Makefile
>> index a759e6e..7087c4f 100644
>> --- a/libavcodec/x86/Makefile
>> +++ b/libavcodec/x86/Makefile
>> @@ -57,7 +57,8 @@ YASM-OBJS-$(CONFIG_H264DSP)            += x86/h264_deblock.o            \
>>  YASM-OBJS-$(CONFIG_H264PRED)           += x86/h264_intrapred.o          \
>>                                            x86/h264_intrapred_10bit.o
>>  YASM-OBJS-$(CONFIG_H264QPEL)           += x86/h264_qpel_8bit.o          \
>> -                                          x86/h264_qpel_10bit.o
>> +                                          x86/h264_qpel_10bit.o         \
>> +                                          x86/qpelbase.o
>>  YASM-OBJS-$(CONFIG_MPEGAUDIODSP)       += x86/imdct36.o
>>  YASM-OBJS-$(CONFIG_PNG_DECODER)        += x86/pngdsp.o
>>  YASM-OBJS-$(CONFIG_PRORES_DECODER)     += x86/proresdsp.o
>> @@ -76,3 +77,4 @@ YASM-OBJS                              += x86/dsputil.o                 \
>>                                            x86/fmtconvert.o              \
>>                                            x86/hpeldsp.o                 \
>>                                            x86/mpeg4qpel.o               \
>> +                                          x86/qpelbase.o
>
> something strange here, x86/qpelbase gets compiled unconditionally AND
> when h264qpel is needed? I'd drop the h264qpel duplicate.

Indeed, fixed locally. (By fixing this, I also change the later patch 
where dsputil becomes optional, readding this file to h264qpel in that 
case.)

// Martin
Luca Barbato April 3, 2013, 1:55 p.m. | #3
On 03/04/13 15:43, Martin Storsjö wrote:
> On Wed, 3 Apr 2013, Luca Barbato wrote:
> 
>> On 03/04/13 15:19, Martin Storsjö wrote:
>>> From: "Ronald S. Bultje" <rsbultje@gmail.com>
>>>
>>> This way, they can be shared between mpeg4qpel and h264qpel without
>>> requiring either one to be compiled unconditionally.
>>> ---
>>>  libavcodec/x86/Makefile     |    4 +-
>>>  libavcodec/x86/dsputil.asm  |  192 --------------------------------
>>>  libavcodec/x86/hpeldsp.asm  |   41 -------
>>>  libavcodec/x86/qpelbase.asm |  256
>>> +++++++++++++++++++++++++++++++++++++++++++
>>>  4 files changed, 259 insertions(+), 234 deletions(-)
>>>  create mode 100644 libavcodec/x86/qpelbase.asm
>>>
>>> diff --git a/libavcodec/x86/Makefile b/libavcodec/x86/Makefile
>>> index a759e6e..7087c4f 100644
>>> --- a/libavcodec/x86/Makefile
>>> +++ b/libavcodec/x86/Makefile
>>> @@ -57,7 +57,8 @@ YASM-OBJS-$(CONFIG_H264DSP)            +=
>>> x86/h264_deblock.o            \
>>>  YASM-OBJS-$(CONFIG_H264PRED)           +=
>>> x86/h264_intrapred.o          \
>>>                                            x86/h264_intrapred_10bit.o
>>>  YASM-OBJS-$(CONFIG_H264QPEL)           +=
>>> x86/h264_qpel_8bit.o          \
>>> -                                          x86/h264_qpel_10bit.o
>>> +                                         
>>> x86/h264_qpel_10bit.o         \
>>> +                                          x86/qpelbase.o
>>>  YASM-OBJS-$(CONFIG_MPEGAUDIODSP)       += x86/imdct36.o
>>>  YASM-OBJS-$(CONFIG_PNG_DECODER)        += x86/pngdsp.o
>>>  YASM-OBJS-$(CONFIG_PRORES_DECODER)     += x86/proresdsp.o
>>> @@ -76,3 +77,4 @@ YASM-OBJS                              +=
>>> x86/dsputil.o                 \
>>>                                           
>>> x86/fmtconvert.o              \
>>>                                           
>>> x86/hpeldsp.o                 \
>>>                                           
>>> x86/mpeg4qpel.o               \
>>> +                                          x86/qpelbase.o
>>
>> something strange here, x86/qpelbase gets compiled unconditionally AND
>> when h264qpel is needed? I'd drop the h264qpel duplicate.
> 
> Indeed, fixed locally. (By fixing this, I also change the later patch
> where dsputil becomes optional, readding this file to h264qpel in that
> case.)
> 

Thanks =)
Diego Biurrun April 4, 2013, 1:59 p.m. | #4
On Wed, Apr 03, 2013 at 04:19:30PM +0300, Martin Storsjö wrote:
> From: "Ronald S. Bultje" <rsbultje@gmail.com>
> 
> This way, they can be shared between mpeg4qpel and h264qpel without
> requiring either one to be compiled unconditionally.
> ---
>  libavcodec/x86/Makefile     |    4 +-
>  libavcodec/x86/dsputil.asm  |  192 --------------------------------
>  libavcodec/x86/hpeldsp.asm  |   41 -------
>  libavcodec/x86/qpelbase.asm |  256 +++++++++++++++++++++++++++++++++++++++++++
>  4 files changed, 259 insertions(+), 234 deletions(-)
>  create mode 100644 libavcodec/x86/qpelbase.asm

I think the name should be qpel.asm, we don't use a "base" suffix
anywhere else.

> --- a/libavcodec/x86/Makefile
> +++ b/libavcodec/x86/Makefile
> @@ -57,7 +57,8 @@ YASM-OBJS-$(CONFIG_H264DSP)            += x86/h264_deblock.o            \
>  YASM-OBJS-$(CONFIG_H264PRED)           += x86/h264_intrapred.o          \
>                                            x86/h264_intrapred_10bit.o
>  YASM-OBJS-$(CONFIG_H264QPEL)           += x86/h264_qpel_8bit.o          \
> -                                          x86/h264_qpel_10bit.o
> +                                          x86/h264_qpel_10bit.o         \
> +                                          x86/qpelbase.o
>  YASM-OBJS-$(CONFIG_MPEGAUDIODSP)       += x86/imdct36.o
>  YASM-OBJS-$(CONFIG_PNG_DECODER)        += x86/pngdsp.o
>  YASM-OBJS-$(CONFIG_PRORES_DECODER)     += x86/proresdsp.o
> @@ -76,3 +77,4 @@ YASM-OBJS                              += x86/dsputil.o                 \
>                                            x86/fmtconvert.o              \
>                                            x86/hpeldsp.o                 \
>                                            x86/mpeg4qpel.o               \
> +                                          x86/qpelbase.o

duplicate entry

Diego
Martin Storsjö April 4, 2013, 3:30 p.m. | #5
On Thu, 4 Apr 2013, Diego Biurrun wrote:

> On Wed, Apr 03, 2013 at 04:19:30PM +0300, Martin Storsjö wrote:
>> From: "Ronald S. Bultje" <rsbultje@gmail.com>
>>
>> This way, they can be shared between mpeg4qpel and h264qpel without
>> requiring either one to be compiled unconditionally.
>> ---
>>  libavcodec/x86/Makefile     |    4 +-
>>  libavcodec/x86/dsputil.asm  |  192 --------------------------------
>>  libavcodec/x86/hpeldsp.asm  |   41 -------
>>  libavcodec/x86/qpelbase.asm |  256 +++++++++++++++++++++++++++++++++++++++++++
>>  4 files changed, 259 insertions(+), 234 deletions(-)
>>  create mode 100644 libavcodec/x86/qpelbase.asm
>
> I think the name should be qpel.asm, we don't use a "base" suffix
> anywhere else.

Renamed locally.

>
>> --- a/libavcodec/x86/Makefile
>> +++ b/libavcodec/x86/Makefile
>> @@ -57,7 +57,8 @@ YASM-OBJS-$(CONFIG_H264DSP)            += x86/h264_deblock.o            \
>>  YASM-OBJS-$(CONFIG_H264PRED)           += x86/h264_intrapred.o          \
>>                                            x86/h264_intrapred_10bit.o
>>  YASM-OBJS-$(CONFIG_H264QPEL)           += x86/h264_qpel_8bit.o          \
>> -                                          x86/h264_qpel_10bit.o
>> +                                          x86/h264_qpel_10bit.o         \
>> +                                          x86/qpelbase.o
>>  YASM-OBJS-$(CONFIG_MPEGAUDIODSP)       += x86/imdct36.o
>>  YASM-OBJS-$(CONFIG_PNG_DECODER)        += x86/pngdsp.o
>>  YASM-OBJS-$(CONFIG_PRORES_DECODER)     += x86/proresdsp.o
>> @@ -76,3 +77,4 @@ YASM-OBJS                              += x86/dsputil.o                 \
>>                                            x86/fmtconvert.o              \
>>                                            x86/hpeldsp.o                 \
>>                                            x86/mpeg4qpel.o               \
>> +                                          x86/qpelbase.o
>
> duplicate entry

Already pointed out by Luca and already fixed locally.

// Martin

Patch

diff --git a/libavcodec/x86/Makefile b/libavcodec/x86/Makefile
index a759e6e..7087c4f 100644
--- a/libavcodec/x86/Makefile
+++ b/libavcodec/x86/Makefile
@@ -57,7 +57,8 @@  YASM-OBJS-$(CONFIG_H264DSP)            += x86/h264_deblock.o            \
 YASM-OBJS-$(CONFIG_H264PRED)           += x86/h264_intrapred.o          \
                                           x86/h264_intrapred_10bit.o
 YASM-OBJS-$(CONFIG_H264QPEL)           += x86/h264_qpel_8bit.o          \
-                                          x86/h264_qpel_10bit.o
+                                          x86/h264_qpel_10bit.o         \
+                                          x86/qpelbase.o
 YASM-OBJS-$(CONFIG_MPEGAUDIODSP)       += x86/imdct36.o
 YASM-OBJS-$(CONFIG_PNG_DECODER)        += x86/pngdsp.o
 YASM-OBJS-$(CONFIG_PRORES_DECODER)     += x86/proresdsp.o
@@ -76,3 +77,4 @@  YASM-OBJS                              += x86/dsputil.o                 \
                                           x86/fmtconvert.o              \
                                           x86/hpeldsp.o                 \
                                           x86/mpeg4qpel.o               \
+                                          x86/qpelbase.o
diff --git a/libavcodec/x86/dsputil.asm b/libavcodec/x86/dsputil.asm
index 078f58c..5d73ff8 100644
--- a/libavcodec/x86/dsputil.asm
+++ b/libavcodec/x86/dsputil.asm
@@ -648,195 +648,3 @@  BSWAP32_BUF
 
 INIT_XMM ssse3
 BSWAP32_BUF
-
-
-; FIXME: All of the code below should be put back in h264_qpel_8bit.asm.
-; Unfortunately it is unconditionally used from dsputil_mmx.c since 71155d7 ..
-
-%macro op_avgh 3
-    movh   %3, %2
-    pavgb  %1, %3
-    movh   %2, %1
-%endmacro
-
-%macro op_avg 2
-    pavgb  %1, %2
-    mova   %2, %1
-%endmacro
-
-%macro op_puth 2-3
-    movh   %2, %1
-%endmacro
-
-%macro op_put 2
-    mova   %2, %1
-%endmacro
-
-; void pixels4_l2_mmxext(uint8_t *dst, uint8_t *src1, uint8_t *src2, int dstStride, int src1Stride, int h)
-%macro PIXELS4_L2 1
-%define OP op_%1h
-cglobal %1_pixels4_l2, 6,6
-    movsxdifnidn r3, r3d
-    movsxdifnidn r4, r4d
-    test        r5d, 1
-    je        .loop
-    movd         m0, [r1]
-    movd         m1, [r2]
-    add          r1, r4
-    add          r2, 4
-    pavgb        m0, m1
-    OP           m0, [r0], m3
-    add          r0, r3
-    dec         r5d
-.loop:
-    mova         m0, [r1]
-    mova         m1, [r1+r4]
-    lea          r1, [r1+2*r4]
-    pavgb        m0, [r2]
-    pavgb        m1, [r2+4]
-    OP           m0, [r0], m3
-    OP           m1, [r0+r3], m3
-    lea          r0, [r0+2*r3]
-    mova         m0, [r1]
-    mova         m1, [r1+r4]
-    lea          r1, [r1+2*r4]
-    pavgb        m0, [r2+8]
-    pavgb        m1, [r2+12]
-    OP           m0, [r0], m3
-    OP           m1, [r0+r3], m3
-    lea          r0, [r0+2*r3]
-    add          r2, 16
-    sub         r5d, 4
-    jne       .loop
-    REP_RET
-%endmacro
-
-INIT_MMX mmxext
-PIXELS4_L2 put
-PIXELS4_L2 avg
-
-; void pixels8_l2_mmxext(uint8_t *dst, uint8_t *src1, uint8_t *src2, int dstStride, int src1Stride, int h)
-%macro PIXELS8_L2 1
-%define OP op_%1
-cglobal %1_pixels8_l2, 6,6
-    movsxdifnidn r3, r3d
-    movsxdifnidn r4, r4d
-    test        r5d, 1
-    je        .loop
-    mova         m0, [r1]
-    mova         m1, [r2]
-    add          r1, r4
-    add          r2, 8
-    pavgb        m0, m1
-    OP           m0, [r0]
-    add          r0, r3
-    dec         r5d
-.loop:
-    mova         m0, [r1]
-    mova         m1, [r1+r4]
-    lea          r1, [r1+2*r4]
-    pavgb        m0, [r2]
-    pavgb        m1, [r2+8]
-    OP           m0, [r0]
-    OP           m1, [r0+r3]
-    lea          r0, [r0+2*r3]
-    mova         m0, [r1]
-    mova         m1, [r1+r4]
-    lea          r1, [r1+2*r4]
-    pavgb        m0, [r2+16]
-    pavgb        m1, [r2+24]
-    OP           m0, [r0]
-    OP           m1, [r0+r3]
-    lea          r0, [r0+2*r3]
-    add          r2, 32
-    sub         r5d, 4
-    jne       .loop
-    REP_RET
-%endmacro
-
-INIT_MMX mmxext
-PIXELS8_L2 put
-PIXELS8_L2 avg
-
-; void pixels16_l2_mmxext(uint8_t *dst, uint8_t *src1, uint8_t *src2, int dstStride, int src1Stride, int h)
-%macro PIXELS16_L2 1
-%define OP op_%1
-cglobal %1_pixels16_l2, 6,6
-    movsxdifnidn r3, r3d
-    movsxdifnidn r4, r4d
-    test        r5d, 1
-    je        .loop
-    mova         m0, [r1]
-    mova         m1, [r1+8]
-    pavgb        m0, [r2]
-    pavgb        m1, [r2+8]
-    add          r1, r4
-    add          r2, 16
-    OP           m0, [r0]
-    OP           m1, [r0+8]
-    add          r0, r3
-    dec         r5d
-.loop:
-    mova         m0, [r1]
-    mova         m1, [r1+8]
-    add          r1, r4
-    pavgb        m0, [r2]
-    pavgb        m1, [r2+8]
-    OP           m0, [r0]
-    OP           m1, [r0+8]
-    add          r0, r3
-    mova         m0, [r1]
-    mova         m1, [r1+8]
-    add          r1, r4
-    pavgb        m0, [r2+16]
-    pavgb        m1, [r2+24]
-    OP           m0, [r0]
-    OP           m1, [r0+8]
-    add          r0, r3
-    add          r2, 32
-    sub         r5d, 2
-    jne       .loop
-    REP_RET
-%endmacro
-
-INIT_MMX mmxext
-PIXELS16_L2 put
-PIXELS16_L2 avg
-
-INIT_MMX mmxext
-; void pixels(uint8_t *block, const uint8_t *pixels, int line_size, int h)
-%macro PIXELS48 2
-%if %2 == 4
-%define OP movh
-%else
-%define OP mova
-%endif
-cglobal %1_pixels%2, 4,5
-    movsxdifnidn r2, r2d
-    lea          r4, [r2*3]
-.loop:
-    OP           m0, [r1]
-    OP           m1, [r1+r2]
-    OP           m2, [r1+r2*2]
-    OP           m3, [r1+r4]
-    lea          r1, [r1+r2*4]
-%ifidn %1, avg
-    pavgb        m0, [r0]
-    pavgb        m1, [r0+r2]
-    pavgb        m2, [r0+r2*2]
-    pavgb        m3, [r0+r4]
-%endif
-    OP         [r0], m0
-    OP      [r0+r2], m1
-    OP    [r0+r2*2], m2
-    OP      [r0+r4], m3
-    sub         r3d, 4
-    lea          r0, [r0+r2*4]
-    jne       .loop
-    RET
-%endmacro
-
-PIXELS48 put, 4
-PIXELS48 avg, 4
-PIXELS48 put, 8
-PIXELS48 avg, 8
diff --git a/libavcodec/x86/hpeldsp.asm b/libavcodec/x86/hpeldsp.asm
index d38186c..ec04d99 100644
--- a/libavcodec/x86/hpeldsp.asm
+++ b/libavcodec/x86/hpeldsp.asm
@@ -452,44 +452,3 @@  INIT_MMX mmxext
 AVG_PIXELS8_XY2
 INIT_MMX 3dnow
 AVG_PIXELS8_XY2
-
-INIT_XMM sse2
-; void put_pixels16_sse2(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
-cglobal put_pixels16, 4,5,4
-    lea          r4, [r2*3]
-.loop:
-    movu         m0, [r1]
-    movu         m1, [r1+r2]
-    movu         m2, [r1+r2*2]
-    movu         m3, [r1+r4]
-    lea          r1, [r1+r2*4]
-    mova       [r0], m0
-    mova    [r0+r2], m1
-    mova  [r0+r2*2], m2
-    mova    [r0+r4], m3
-    sub         r3d, 4
-    lea          r0, [r0+r2*4]
-    jnz       .loop
-    REP_RET
-
-; void avg_pixels16_sse2(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
-cglobal avg_pixels16, 4,5,4
-    lea          r4, [r2*3]
-.loop:
-    movu         m0, [r1]
-    movu         m1, [r1+r2]
-    movu         m2, [r1+r2*2]
-    movu         m3, [r1+r4]
-    lea          r1, [r1+r2*4]
-    pavgb        m0, [r0]
-    pavgb        m1, [r0+r2]
-    pavgb        m2, [r0+r2*2]
-    pavgb        m3, [r0+r4]
-    mova       [r0], m0
-    mova    [r0+r2], m1
-    mova  [r0+r2*2], m2
-    mova    [r0+r4], m3
-    sub         r3d, 4
-    lea          r0, [r0+r2*4]
-    jnz       .loop
-    REP_RET
diff --git a/libavcodec/x86/qpelbase.asm b/libavcodec/x86/qpelbase.asm
new file mode 100644
index 0000000..ba97e2a
--- /dev/null
+++ b/libavcodec/x86/qpelbase.asm
@@ -0,0 +1,256 @@ 
+;******************************************************************************
+;* MMX optimized DSP utils
+;* Copyright (c) 2008 Loren Merritt
+;* Copyright (c) 2003-2013 Michael Niedermayer
+;* Copyright (c) 2013 Daniel Kang
+;*
+;* This file is part of Libav.
+;*
+;* Libav is free software; you can redistribute it and/or
+;* modify it under the terms of the GNU Lesser General Public
+;* License as published by the Free Software Foundation; either
+;* version 2.1 of the License, or (at your option) any later version.
+;*
+;* Libav is distributed in the hope that it will be useful,
+;* but WITHOUT ANY WARRANTY; without even the implied warranty of
+;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+;* Lesser General Public License for more details.
+;*
+;* You should have received a copy of the GNU Lesser General Public
+;* License along with Libav; if not, write to the Free Software
+;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+;******************************************************************************
+
+%include "libavutil/x86/x86util.asm"
+
+SECTION .text
+
+%macro op_avgh 3
+    movh   %3, %2
+    pavgb  %1, %3
+    movh   %2, %1
+%endmacro
+
+%macro op_avg 2
+    pavgb  %1, %2
+    mova   %2, %1
+%endmacro
+
+%macro op_puth 2-3
+    movh   %2, %1
+%endmacro
+
+%macro op_put 2
+    mova   %2, %1
+%endmacro
+
+; void pixels4_l2_mmxext(uint8_t *dst, uint8_t *src1, uint8_t *src2, int dstStride, int src1Stride, int h)
+%macro PIXELS4_L2 1
+%define OP op_%1h
+cglobal %1_pixels4_l2, 6,6
+    movsxdifnidn r3, r3d
+    movsxdifnidn r4, r4d
+    test        r5d, 1
+    je        .loop
+    movd         m0, [r1]
+    movd         m1, [r2]
+    add          r1, r4
+    add          r2, 4
+    pavgb        m0, m1
+    OP           m0, [r0], m3
+    add          r0, r3
+    dec         r5d
+.loop:
+    mova         m0, [r1]
+    mova         m1, [r1+r4]
+    lea          r1, [r1+2*r4]
+    pavgb        m0, [r2]
+    pavgb        m1, [r2+4]
+    OP           m0, [r0], m3
+    OP           m1, [r0+r3], m3
+    lea          r0, [r0+2*r3]
+    mova         m0, [r1]
+    mova         m1, [r1+r4]
+    lea          r1, [r1+2*r4]
+    pavgb        m0, [r2+8]
+    pavgb        m1, [r2+12]
+    OP           m0, [r0], m3
+    OP           m1, [r0+r3], m3
+    lea          r0, [r0+2*r3]
+    add          r2, 16
+    sub         r5d, 4
+    jne       .loop
+    REP_RET
+%endmacro
+
+INIT_MMX mmxext
+PIXELS4_L2 put
+PIXELS4_L2 avg
+
+; void pixels8_l2_mmxext(uint8_t *dst, uint8_t *src1, uint8_t *src2, int dstStride, int src1Stride, int h)
+%macro PIXELS8_L2 1
+%define OP op_%1
+cglobal %1_pixels8_l2, 6,6
+    movsxdifnidn r3, r3d
+    movsxdifnidn r4, r4d
+    test        r5d, 1
+    je        .loop
+    mova         m0, [r1]
+    mova         m1, [r2]
+    add          r1, r4
+    add          r2, 8
+    pavgb        m0, m1
+    OP           m0, [r0]
+    add          r0, r3
+    dec         r5d
+.loop:
+    mova         m0, [r1]
+    mova         m1, [r1+r4]
+    lea          r1, [r1+2*r4]
+    pavgb        m0, [r2]
+    pavgb        m1, [r2+8]
+    OP           m0, [r0]
+    OP           m1, [r0+r3]
+    lea          r0, [r0+2*r3]
+    mova         m0, [r1]
+    mova         m1, [r1+r4]
+    lea          r1, [r1+2*r4]
+    pavgb        m0, [r2+16]
+    pavgb        m1, [r2+24]
+    OP           m0, [r0]
+    OP           m1, [r0+r3]
+    lea          r0, [r0+2*r3]
+    add          r2, 32
+    sub         r5d, 4
+    jne       .loop
+    REP_RET
+%endmacro
+
+INIT_MMX mmxext
+PIXELS8_L2 put
+PIXELS8_L2 avg
+
+; void pixels16_l2_mmxext(uint8_t *dst, uint8_t *src1, uint8_t *src2, int dstStride, int src1Stride, int h)
+%macro PIXELS16_L2 1
+%define OP op_%1
+cglobal %1_pixels16_l2, 6,6
+    movsxdifnidn r3, r3d
+    movsxdifnidn r4, r4d
+    test        r5d, 1
+    je        .loop
+    mova         m0, [r1]
+    mova         m1, [r1+8]
+    pavgb        m0, [r2]
+    pavgb        m1, [r2+8]
+    add          r1, r4
+    add          r2, 16
+    OP           m0, [r0]
+    OP           m1, [r0+8]
+    add          r0, r3
+    dec         r5d
+.loop:
+    mova         m0, [r1]
+    mova         m1, [r1+8]
+    add          r1, r4
+    pavgb        m0, [r2]
+    pavgb        m1, [r2+8]
+    OP           m0, [r0]
+    OP           m1, [r0+8]
+    add          r0, r3
+    mova         m0, [r1]
+    mova         m1, [r1+8]
+    add          r1, r4
+    pavgb        m0, [r2+16]
+    pavgb        m1, [r2+24]
+    OP           m0, [r0]
+    OP           m1, [r0+8]
+    add          r0, r3
+    add          r2, 32
+    sub         r5d, 2
+    jne       .loop
+    REP_RET
+%endmacro
+
+INIT_MMX mmxext
+PIXELS16_L2 put
+PIXELS16_L2 avg
+
+INIT_MMX mmxext
+; void pixels(uint8_t *block, const uint8_t *pixels, int line_size, int h)
+%macro PIXELS48 2
+%if %2 == 4
+%define OP movh
+%else
+%define OP mova
+%endif
+cglobal %1_pixels%2, 4,5
+    movsxdifnidn r2, r2d
+    lea          r4, [r2*3]
+.loop:
+    OP           m0, [r1]
+    OP           m1, [r1+r2]
+    OP           m2, [r1+r2*2]
+    OP           m3, [r1+r4]
+    lea          r1, [r1+r2*4]
+%ifidn %1, avg
+    pavgb        m0, [r0]
+    pavgb        m1, [r0+r2]
+    pavgb        m2, [r0+r2*2]
+    pavgb        m3, [r0+r4]
+%endif
+    OP         [r0], m0
+    OP      [r0+r2], m1
+    OP    [r0+r2*2], m2
+    OP      [r0+r4], m3
+    sub         r3d, 4
+    lea          r0, [r0+r2*4]
+    jne       .loop
+    RET
+%endmacro
+
+PIXELS48 put, 4
+PIXELS48 avg, 4
+PIXELS48 put, 8
+PIXELS48 avg, 8
+
+
+INIT_XMM sse2
+; void put_pixels16_sse2(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
+cglobal put_pixels16, 4,5,4
+    lea          r4, [r2*3]
+.loop:
+    movu         m0, [r1]
+    movu         m1, [r1+r2]
+    movu         m2, [r1+r2*2]
+    movu         m3, [r1+r4]
+    lea          r1, [r1+r2*4]
+    mova       [r0], m0
+    mova    [r0+r2], m1
+    mova  [r0+r2*2], m2
+    mova    [r0+r4], m3
+    sub         r3d, 4
+    lea          r0, [r0+r2*4]
+    jnz       .loop
+    REP_RET
+
+; void avg_pixels16_sse2(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)
+cglobal avg_pixels16, 4,5,4
+    lea          r4, [r2*3]
+.loop:
+    movu         m0, [r1]
+    movu         m1, [r1+r2]
+    movu         m2, [r1+r2*2]
+    movu         m3, [r1+r4]
+    lea          r1, [r1+r2*4]
+    pavgb        m0, [r0]
+    pavgb        m1, [r0+r2]
+    pavgb        m2, [r0+r2*2]
+    pavgb        m3, [r0+r4]
+    mova       [r0], m0
+    mova    [r0+r2], m1
+    mova  [r0+r2*2], m2
+    mova    [r0+r4], m3
+    sub         r3d, 4
+    lea          r0, [r0+r2*4]
+    jnz       .loop
+    REP_RET