[FFmpeg-devel] [PATCH] x86: vc1dsp: Convert vc1_inv_trans_*_dc to NASM format

[James Almer]

On 1/31/2016 6:18 PM, James Almer wrote:
> On 1/31/2016 4:48 PM, Timothy Gu wrote:
>> +; 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]
> 
> Can this value be negative? Because you're using it as an argument
> for lea using native size after movsx sign extended the value to 32
> bits, which means that on x86_64 the upper bits of the register will
> be zeroed.
> 
> If it can you'll have to use blockq/r3q everywhere, and if it can't
> then use movzx and shr.

Or not... Seems to work as is and using movzx breaks it.

> 
>> +    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
>