Attachment 89694 Details for Bug 71944 – Output with R600_DEBUG=vs,sbdump

Output with R600_DEBUG=vs,sbdump

sbdump (text/plain), 176.12 KB, created by Martin Andersson on 2013-11-23 21:07:52 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Martin Andersson

Created: 2013-11-23 21:07:52 UTC

Size: 176.12 KB

patch

obsolete

>
>
>###### after parse
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>    CF_END     
>}  
>
>
>###### after if_conversion
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after peephole
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after gvn
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after def_use
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after dce_cleanup
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after gcm
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after parse
>{  
>    ALU     
>    <  
>        [  
>            INTERP_LOAD_P0     R1.x,    Param0x
>            INTERP_LOAD_P0     R1.y,    Param0y
>            INTERP_LOAD_P0     R1.z,    Param0z
>            INTERP_LOAD_P0     R1.w,    Param0w
>        ]  
>        [  
>            MOV     R2.x,    R1.x
>            MOV     R2.y,    R1.y
>            MOV     R2.z,    R1.z
>            MOV     R2.w,    R1.w
>        ]  
>    >  
>    EXPORT  PIXEL 0     R2.x, R2.y, R2.z, R2.w
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{  
>    ALU     
>    <  
>        [  
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>        ]  
>        [  
>            MOV     R2.x.1,    R1.x.1
>            MOV     R2.y.1,    R1.y.1
>            MOV     R2.z.1,    R1.z.1
>            MOV     R2.w.1,    R1.w.1
>        ]  
>    >  
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>    CF_END     
>}  
>
>
>###### after if_conversion
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after peephole
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after gvn
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after def_use
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after dce_cleanup
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after gcm
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            (copy) MOV     t4,    R1.w.1
>            (copy) MOV     t3,    R1.z.1
>            (copy) MOV     t2,    R1.y.1
>            (copy) MOV     t1,    R1.x.1
>        >  
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            (copy) MOV     t4,    R1.w.1
>            (copy) MOV     t3,    R1.z.1
>            (copy) MOV     t2,    R1.y.1
>            (copy) MOV     t1,    R1.x.1
>        >  live_after: [t1 t2 t3 t4 ]
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1FP@R0.w,    Param0w
>            INTERP_LOAD_P0     R1.z.1FP@R0.z,    Param0z
>            INTERP_LOAD_P0     R1.y.1FP@R0.y,    Param0y
>            INTERP_LOAD_P0     R1.x.1FP@R0.x,    Param0x
>            (copy) MOV     t4FP@R0.w,    R1.w.1FP@R0.w
>            (copy) MOV     t3FP@R0.z,    R1.z.1FP@R0.z
>            (copy) MOV     t2FP@R0.y,    R1.y.1FP@R0.y
>            (copy) MOV     t1FP@R0.x,    R1.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1FP@R0.w,    Param0w
>            INTERP_LOAD_P0     R1.z.1FP@R0.z,    Param0z
>            INTERP_LOAD_P0     R1.y.1FP@R0.y,    Param0y
>            INTERP_LOAD_P0     R1.x.1FP@R0.x,    Param0x
>            (copy) MOV     t4FP@R0.w,    R1.w.1FP@R0.w
>            (copy) MOV     t3FP@R0.z,    R1.z.1FP@R0.z
>            (copy) MOV     t2FP@R0.y,    R1.y.1FP@R0.y
>            (copy) MOV     t1FP@R0.x,    R1.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            [  
>                INTERP_LOAD_P0     R1.x.1||FP@R0.x,    Param0x
>                INTERP_LOAD_P0     R1.y.1||FP@R0.y,    Param0y
>                INTERP_LOAD_P0     R1.z.1||FP@R0.z,    Param0z
>                INTERP_LOAD_P0     R1.w.1||FP@R0.w,    Param0w
>            ]  
>        >  
>        EXPORT  PIXEL 0     t1||FP@R0.x, t2||FP@R0.y, t3||FP@R0.z, t4||FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after parse
>{  
>    ALU     
>    <  
>        [  
>            INTERP_LOAD_P0     R1.x,    Param0x
>            INTERP_LOAD_P0     R1.y,    Param0y
>            INTERP_LOAD_P0     R1.z,    Param0z
>            INTERP_LOAD_P0     R1.w,    Param0w
>        ]  
>        [  
>            MOV     R2.x,    R1.x
>            MOV     R2.y,    R1.y
>            MOV     R2.z,    R1.z
>            MOV     R2.w,    R1.w
>        ]  
>    >  
>    EXPORT  PIXEL 0     R2.x, R2.y, R2.z, R2.w
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{  
>    ALU     
>    <  
>        [  
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>        ]  
>        [  
>            MOV     R2.x.1,    R1.x.1
>            MOV     R2.y.1,    R1.y.1
>            MOV     R2.z.1,    R1.z.1
>            MOV     R2.w.1,    R1.w.1
>        ]  
>    >  
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>    CF_END     
>}  
>
>
>###### after if_conversion
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after peephole
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after gvn
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after def_use
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after dce_cleanup
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after gcm
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            (copy) MOV     t4,    R1.w.1
>            (copy) MOV     t3,    R1.z.1
>            (copy) MOV     t2,    R1.y.1
>            (copy) MOV     t1,    R1.x.1
>        >  
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            (copy) MOV     t4,    R1.w.1
>            (copy) MOV     t3,    R1.z.1
>            (copy) MOV     t2,    R1.y.1
>            (copy) MOV     t1,    R1.x.1
>        >  live_after: [t1 t2 t3 t4 ]
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1FP@R0.w,    Param0w
>            INTERP_LOAD_P0     R1.z.1FP@R0.z,    Param0z
>            INTERP_LOAD_P0     R1.y.1FP@R0.y,    Param0y
>            INTERP_LOAD_P0     R1.x.1FP@R0.x,    Param0x
>            (copy) MOV     t4FP@R0.w,    R1.w.1FP@R0.w
>            (copy) MOV     t3FP@R0.z,    R1.z.1FP@R0.z
>            (copy) MOV     t2FP@R0.y,    R1.y.1FP@R0.y
>            (copy) MOV     t1FP@R0.x,    R1.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1FP@R0.w,    Param0w
>            INTERP_LOAD_P0     R1.z.1FP@R0.z,    Param0z
>            INTERP_LOAD_P0     R1.y.1FP@R0.y,    Param0y
>            INTERP_LOAD_P0     R1.x.1FP@R0.x,    Param0x
>            (copy) MOV     t4FP@R0.w,    R1.w.1FP@R0.w
>            (copy) MOV     t3FP@R0.z,    R1.z.1FP@R0.z
>            (copy) MOV     t2FP@R0.y,    R1.y.1FP@R0.y
>            (copy) MOV     t1FP@R0.x,    R1.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            [  
>                INTERP_LOAD_P0     R1.x.1||FP@R0.x,    Param0x
>                INTERP_LOAD_P0     R1.y.1||FP@R0.y,    Param0y
>                INTERP_LOAD_P0     R1.z.1||FP@R0.z,    Param0z
>                INTERP_LOAD_P0     R1.w.1||FP@R0.w,    Param0w
>            ]  
>        >  
>        EXPORT  PIXEL 0     t1||FP@R0.x, t2||FP@R0.y, t3||FP@R0.z, t4||FP@R0.w
>    } end BB_0  
>}  
>--------------------------------------------------------------
>VERT
>DCL IN[0]
>DCL IN[1]
>DCL OUT[0], POSITION
>DCL OUT[1], GENERIC[0]
>  0: MOV OUT[0], IN[0]
>  1: MOV OUT[1], IN[1]
>  2: END
>
>===== SHADER #9 ========================================= VS/CAYMAN/CAYMAN =====
>===== 26 dw ===== 5 gprs ===== 1 stack =========================================
>0000  00000000 84c00000 CALL_FS @0
>0002  00000005 a01c0000 ALU 8 @10
> 0010  00000001 00600c90     1      x: MOV                R3.x,  R1.x
> 0012  00000401 20600c90            y: MOV                R3.y,  R1.y
> 0014  00000801 40600c90            z: MOV                R3.z,  R1.z
> 0016  80000c01 60600c90            w: MOV                R3.w,  R1.w
> 0018  00000002 00800c90     2      x: MOV                R4.x,  R2.x
> 0020  00000402 20800c90            y: MOV                R4.y,  R2.y
> 0022  00000802 40800c90            z: MOV                R4.z,  R2.z
> 0024  80000c02 60800c90            w: MOV                R4.w,  R2.w
>0004  c001a03c 95000688 EXPORT_DONE        POS   60    R3.xyzw
>0006  c0024000 95000688 EXPORT_DONE        PARAM 0     R4.xyzw
>0008  00000000 88000000 CF_END @0
>===== SHADER_END ===============================================================
>
>
>
>###### after parse
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.xF@R1.x, R1.yF@R1.y, R1.zF@R1.z, R1.wF@R1.w, R2.xF@R2.x, R2.yF@R2.y, R2.zF@R2.z, R2.wF@R2.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    ALU     
>    <  
>        [  
>            MOV     R3.x,    R1.xF@R1.x
>            MOV     R3.y,    R1.yF@R1.y
>            MOV     R3.z,    R1.zF@R1.z
>            MOV     R3.w,    R1.wF@R1.w
>        ]  
>        [  
>            MOV     R4.x,    R2.xF@R2.x
>            MOV     R4.y,    R2.yF@R2.y
>            MOV     R4.z,    R2.zF@R2.z
>            MOV     R4.w,    R2.wF@R2.w
>        ]  
>    >  
>    EXPORT  POS   60     R3.x, R3.y, R3.z, R3.w
>    EXPORT  PARAM 0     R4.x, R4.y, R4.z, R4.w
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1, R2.x.1, R2.y.1, R2.z.1, R2.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    ALU     
>    <  
>        [  
>            MOV     R3.x.1,    R1.x.1
>            MOV     R3.y.1,    R1.y.1
>            MOV     R3.z.1,    R1.z.1
>            MOV     R3.w.1,    R1.w.1
>        ]  
>        [  
>            MOV     R4.x.1,    R2.x.1
>            MOV     R4.y.1,    R2.y.1
>            MOV     R4.z.1,    R2.z.1
>            MOV     R4.w.1,    R2.w.1
>        ]  
>    >  
>    EXPORT  POS   60     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>    EXPORT  PARAM 0     R4.x.1, R4.y.1, R4.z.1, R4.w.1
>    CF_END     
>}  
>
>
>###### after if_conversion
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1, R2.x.1, R2.y.1, R2.z.1, R2.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R3.x.1,    R1.x.1
>    MOV     R3.y.1,    R1.y.1
>    MOV     R3.z.1,    R1.z.1
>    MOV     R3.w.1,    R1.w.1
>    MOV     R4.x.1,    R2.x.1
>    MOV     R4.y.1,    R2.y.1
>    MOV     R4.z.1,    R2.z.1
>    MOV     R4.w.1,    R2.w.1
>    EXPORT  POS   60     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>    EXPORT  PARAM 0     R4.x.1, R4.y.1, R4.z.1, R4.w.1
>}  
>
>
>###### after peephole
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1, R2.x.1, R2.y.1, R2.z.1, R2.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R3.x.1,    R1.x.1
>    MOV     R3.y.1,    R1.y.1
>    MOV     R3.z.1,    R1.z.1
>    MOV     R3.w.1,    R1.w.1
>    MOV     R4.x.1,    R2.x.1
>    MOV     R4.y.1,    R2.y.1
>    MOV     R4.z.1,    R2.z.1
>    MOV     R4.w.1,    R2.w.1
>    EXPORT  POS   60     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>    EXPORT  PARAM 0     R4.x.1, R4.y.1, R4.z.1, R4.w.1
>}  
>
>
>###### after gvn
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1, R2.x.1, R2.y.1, R2.z.1, R2.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R3.x.1,    R1.x.1
>    MOV     R3.y.1,    R1.y.1
>    MOV     R3.z.1,    R1.z.1
>    MOV     R3.w.1,    R1.w.1
>    MOV     R4.x.1,    R2.x.1
>    MOV     R4.y.1,    R2.y.1
>    MOV     R4.z.1,    R2.z.1
>    MOV     R4.w.1,    R2.w.1
>    EXPORT  POS   60     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>    EXPORT  PARAM 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after def_use
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1, R2.x.1, R2.y.1, R2.z.1, R2.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R3.x.1,    R1.x.1
>    MOV     R3.y.1,    R1.y.1
>    MOV     R3.z.1,    R1.z.1
>    MOV     R3.w.1,    R1.w.1
>    MOV     R4.x.1,    R2.x.1
>    MOV     R4.y.1,    R2.y.1
>    MOV     R4.z.1,    R2.z.1
>    MOV     R4.w.1,    R2.w.1
>    EXPORT  POS   60     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>    EXPORT  PARAM 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1, R2.x.1, R2.y.1, R2.z.1, R2.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    EXPORT  POS   60     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>    EXPORT  PARAM 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after gcm
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  
>        CALL_FS     t9F@R1.x, t10F@R1.y, t11F@R1.z, t12F@R1.w, t13F@R2.x, t14F@R2.y, t15F@R2.z, t16F@R2.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  
>            (copy) MOV     R1.w.1,    t12F@R1.w
>            (copy) MOV     t8,    R1.w.1
>            (copy) MOV     R1.z.1,    t11F@R1.z
>            (copy) MOV     t7,    R1.z.1
>            (copy) MOV     R1.y.1,    t10F@R1.y
>            (copy) MOV     t6,    R1.y.1
>            (copy) MOV     R1.x.1,    t9F@R1.x
>            (copy) MOV     t5,    R1.x.1
>            (copy) MOV     R2.w.1,    t16F@R2.w
>            (copy) MOV     t4,    R2.w.1
>            (copy) MOV     R2.z.1,    t15F@R2.z
>            (copy) MOV     t3,    R2.z.1
>            (copy) MOV     R2.y.1,    t14F@R2.y
>            (copy) MOV     t2,    R2.y.1
>            (copy) MOV     R2.x.1,    t13F@R2.x
>            (copy) MOV     t1,    R2.x.1
>        >  
>        EXPORT  PARAM 0     t1, t2, t3, t4
>        EXPORT  POS   60     t5, t6, t7, t8
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t9F@R1.x, t10F@R1.y, t11F@R1.z, t12F@R1.w, t13F@R2.x, t14F@R2.y, t15F@R2.z, t16F@R2.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t9F@R1.x t10F@R1.y t11F@R1.z t12F@R1.w t13F@R2.x t14F@R2.y t15F@R2.z t16F@R2.w ]
>            (copy) MOV     R1.w.1,    t12F@R1.w
>            (copy) MOV     t8,    R1.w.1
>            (copy) MOV     R1.z.1,    t11F@R1.z
>            (copy) MOV     t7,    R1.z.1
>            (copy) MOV     R1.y.1,    t10F@R1.y
>            (copy) MOV     t6,    R1.y.1
>            (copy) MOV     R1.x.1,    t9F@R1.x
>            (copy) MOV     t5,    R1.x.1
>            (copy) MOV     R2.w.1,    t16F@R2.w
>            (copy) MOV     t4,    R2.w.1
>            (copy) MOV     R2.z.1,    t15F@R2.z
>            (copy) MOV     t3,    R2.z.1
>            (copy) MOV     R2.y.1,    t14F@R2.y
>            (copy) MOV     t2,    R2.y.1
>            (copy) MOV     R2.x.1,    t13F@R2.x
>            (copy) MOV     t1,    R2.x.1
>        >  live_after: [t1 t2 t3 t4 t5 t6 t7 t8 ]
>        EXPORT  PARAM 0     t1, t2, t3, t4
>        EXPORT  POS   60     t5, t6, t7, t8
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t9FP@R1.x, t10FP@R1.y, t11FP@R1.z, t12FP@R1.w, t13FP@R2.x, t14FP@R2.y, t15FP@R2.z, t16FP@R2.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t9FP@R1.x t10FP@R1.y t11FP@R1.z t12FP@R1.w t13FP@R2.x t14FP@R2.y t15FP@R2.z t16FP@R2.w ]
>            (copy) MOV     R1.w.1FP@R1.w,    t12FP@R1.w
>            (copy) MOV     t8FP@R1.w,    R1.w.1FP@R1.w
>            (copy) MOV     R1.z.1FP@R1.z,    t11FP@R1.z
>            (copy) MOV     t7FP@R1.z,    R1.z.1FP@R1.z
>            (copy) MOV     R1.y.1FP@R1.y,    t10FP@R1.y
>            (copy) MOV     t6FP@R1.y,    R1.y.1FP@R1.y
>            (copy) MOV     R1.x.1FP@R1.x,    t9FP@R1.x
>            (copy) MOV     t5FP@R1.x,    R1.x.1FP@R1.x
>            (copy) MOV     R2.w.1FP@R2.w,    t16FP@R2.w
>            (copy) MOV     t4FP@R2.w,    R2.w.1FP@R2.w
>            (copy) MOV     R2.z.1FP@R2.z,    t15FP@R2.z
>            (copy) MOV     t3FP@R2.z,    R2.z.1FP@R2.z
>            (copy) MOV     R2.y.1FP@R2.y,    t14FP@R2.y
>            (copy) MOV     t2FP@R2.y,    R2.y.1FP@R2.y
>            (copy) MOV     R2.x.1FP@R2.x,    t13FP@R2.x
>            (copy) MOV     t1FP@R2.x,    R2.x.1FP@R2.x
>        >  live_after: [t1FP@R2.x t2FP@R2.y t3FP@R2.z t4FP@R2.w t5FP@R1.x t6FP@R1.y t7FP@R1.z t8FP@R1.w ]
>        EXPORT  PARAM 0     t1FP@R2.x, t2FP@R2.y, t3FP@R2.z, t4FP@R2.w
>        EXPORT  POS   60     t5FP@R1.x, t6FP@R1.y, t7FP@R1.z, t8FP@R1.w
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t9FP@R1.x, t10FP@R1.y, t11FP@R1.z, t12FP@R1.w, t13FP@R2.x, t14FP@R2.y, t15FP@R2.z, t16FP@R2.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t9FP@R1.x t10FP@R1.y t11FP@R1.z t12FP@R1.w t13FP@R2.x t14FP@R2.y t15FP@R2.z t16FP@R2.w ]
>            (copy) MOV     R1.w.1FP@R1.w,    t12FP@R1.w
>            (copy) MOV     t8FP@R1.w,    R1.w.1FP@R1.w
>            (copy) MOV     R1.z.1FP@R1.z,    t11FP@R1.z
>            (copy) MOV     t7FP@R1.z,    R1.z.1FP@R1.z
>            (copy) MOV     R1.y.1FP@R1.y,    t10FP@R1.y
>            (copy) MOV     t6FP@R1.y,    R1.y.1FP@R1.y
>            (copy) MOV     R1.x.1FP@R1.x,    t9FP@R1.x
>            (copy) MOV     t5FP@R1.x,    R1.x.1FP@R1.x
>            (copy) MOV     R2.w.1FP@R2.w,    t16FP@R2.w
>            (copy) MOV     t4FP@R2.w,    R2.w.1FP@R2.w
>            (copy) MOV     R2.z.1FP@R2.z,    t15FP@R2.z
>            (copy) MOV     t3FP@R2.z,    R2.z.1FP@R2.z
>            (copy) MOV     R2.y.1FP@R2.y,    t14FP@R2.y
>            (copy) MOV     t2FP@R2.y,    R2.y.1FP@R2.y
>            (copy) MOV     R2.x.1FP@R2.x,    t13FP@R2.x
>            (copy) MOV     t1FP@R2.x,    R2.x.1FP@R2.x
>        >  live_after: [t1FP@R2.x t2FP@R2.y t3FP@R2.z t4FP@R2.w t5FP@R1.x t6FP@R1.y t7FP@R1.z t8FP@R1.w ]
>        EXPORT  PARAM 0     t1FP@R2.x, t2FP@R2.y, t3FP@R2.z, t4FP@R2.w
>        EXPORT  POS   60     t5FP@R1.x, t6FP@R1.y, t7FP@R1.z, t8FP@R1.w
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t9||FP@R1.x, t10||FP@R1.y, t11||FP@R1.z, t12||FP@R1.w, t13||FP@R2.x, t14||FP@R2.y, t15||FP@R2.z, t16||FP@R2.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        EXPORT  PARAM 0     t1||FP@R2.x, t2||FP@R2.y, t3||FP@R2.z, t4||FP@R2.w
>        EXPORT  POS   60     t5||FP@R1.x, t6||FP@R1.y, t7||FP@R1.z, t8||FP@R1.w
>    } end BB_0  
>}  
>
>===== SHADER #9 OPT ===================================== VS/CAYMAN/CAYMAN =====
>===== 8 dw ===== 3 gprs ===== 1 stack ==========================================
>0000  00000000 84c00000 CALL_FS @0
>0002  c0014000 95000688 EXPORT_DONE        PARAM 0     R2.xyzw
>0004  c000a03c 95000688 EXPORT_DONE        POS   60    R1.xyzw
>0006  00000000 88000000 CF_END @0
>===== SHADER_END ===============================================================
>
>--------------------------------------------------------------
>VERT
>DCL IN[0]
>DCL OUT[0], POSITION
>  0: MOV OUT[0], IN[0]
>  1: END
>STREAMOUT
>  0: MEM_STREAM0_BUF0[0..0] <- OUT[0].x
>
>===== SHADER #10 ======================================== VS/CAYMAN/CAYMAN =====
>===== 20 dw ===== 3 gprs ===== 1 stack =========================================
>0000  00000000 84c00000 CALL_FS @0
>0002  00000006 a00c0000 ALU 4 @12
> 0012  00000001 00400c90     1      x: MOV                R2.x,  R1.x
> 0014  00000401 20400c90            y: MOV                R2.y,  R1.y
> 0016  00000801 40400c90            z: MOV                R2.z,  R1.z
> 0018  80000c01 60400c90            w: MOV                R2.w,  R1.w
>0004  40010000 90001fff MEM_STREAM0_BUF0   WRITE 0     R2.x___  ES:1
>0006  c001203c 95000688 EXPORT_DONE        POS   60    R2.xyzw
>0008  c0004000 95000fff EXPORT_DONE        PARAM 0     R0.____
>0010  00000000 88000000 CF_END @0
>===== SHADER_END ===============================================================
>
>
>
>###### after parse
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.xF@R1.x, R1.yF@R1.y, R1.zF@R1.z, R1.wF@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    ALU     
>    <  
>        [  
>            MOV     R2.x,    R1.xF@R1.x
>            MOV     R2.y,    R1.yF@R1.y
>            MOV     R2.z,    R1.zF@R1.z
>            MOV     R2.w,    R1.wF@R1.w
>        ]  
>    >  
>    MEM_STREAM0_BUF0  WRITE 0   ES:1     R2.x, __, __, __
>    EXPORT  POS   60     R2.x, R2.y, R2.z, R2.w
>    EXPORT  PARAM 0     __, __, __, __
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    ALU     
>    <  
>        [  
>            MOV     R2.x.1,    R1.x.1
>            MOV     R2.y.1,    R1.y.1
>            MOV     R2.z.1,    R1.z.1
>            MOV     R2.w.1,    R1.w.1
>        ]  
>    >  
>    MEM_STREAM0_BUF0  WRITE 0   ES:1     R2.x.1, __, __, __
>    EXPORT  POS   60     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>    EXPORT  PARAM 0     __, __, __, __
>    CF_END     
>}  
>
>
>###### after if_conversion
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    MEM_STREAM0_BUF0  WRITE 0   ES:1     R2.x.1, __, __, __
>    EXPORT  POS   60     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>    EXPORT  PARAM 0     __, __, __, __
>}  
>
>
>###### after peephole
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    MEM_STREAM0_BUF0  WRITE 0   ES:1     R2.x.1, __, __, __
>    EXPORT  POS   60     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>    EXPORT  PARAM 0     __, __, __, __
>}  
>
>
>###### after gvn
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    MEM_STREAM0_BUF0  WRITE 0   ES:1     R1.x.1, __, __, __
>    EXPORT  POS   60     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>    EXPORT  PARAM 0     __, __, __, __
>}  
>
>
>###### after def_use
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    MEM_STREAM0_BUF0  WRITE 0   ES:1     R1.x.1, __, __, __
>    EXPORT  POS   60     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>    EXPORT  PARAM 0     __, __, __, __
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MEM_STREAM0_BUF0  WRITE 0   ES:1     R1.x.1, __, __, __
>    EXPORT  POS   60     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>    EXPORT  PARAM 0     __, __, __, __
>}  
>
>
>###### after gcm
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  
>        CALL_FS     t6F@R1.x, t7F@R1.y, t8F@R1.z, t9F@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  
>            (copy) MOV     R1.x.1,    t6F@R1.x
>            (copy) MOV     t5,    R1.x.1
>            (copy) MOV     R1.w.1,    t9F@R1.w
>            (copy) MOV     t4,    R1.w.1
>            (copy) MOV     R1.z.1,    t8F@R1.z
>            (copy) MOV     t3,    R1.z.1
>            (copy) MOV     R1.y.1,    t7F@R1.y
>            (copy) MOV     t2,    R1.y.1
>            (copy) MOV     t1,    R1.x.1
>        >  
>        EXPORT  POS   60     t1, t2, t3, t4
>        MEM_STREAM0_BUF0  WRITE 0   ES:1     t5, __, __, __
>        EXPORT  PARAM 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t6F@R1.x, t7F@R1.y, t8F@R1.z, t9F@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t6F@R1.x t7F@R1.y t8F@R1.z t9F@R1.w ]
>            (copy) MOV     R1.x.1,    t6F@R1.x
>            (copy) MOV     t5,    R1.x.1
>            (copy) MOV     R1.w.1,    t9F@R1.w
>            (copy) MOV     t4,    R1.w.1
>            (copy) MOV     R1.z.1,    t8F@R1.z
>            (copy) MOV     t3,    R1.z.1
>            (copy) MOV     R1.y.1,    t7F@R1.y
>            (copy) MOV     t2,    R1.y.1
>            (copy) MOV     t1,    R1.x.1
>        >  live_after: [t1 t2 t3 t4 t5 ]
>        EXPORT  POS   60     t1, t2, t3, t4
>        MEM_STREAM0_BUF0  WRITE 0   ES:1     t5, __, __, __
>        EXPORT  PARAM 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t6FP@R1.x, t7FP@R1.y, t8FP@R1.z, t9FP@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t6FP@R1.x t7FP@R1.y t8FP@R1.z t9FP@R1.w ]
>            (copy) MOV     R1.x.1FP@R1.x,    t6FP@R1.x
>            (copy) MOV     t5FP@R1.x,    R1.x.1FP@R1.x
>            (copy) MOV     R1.w.1FP@R1.w,    t9FP@R1.w
>            (copy) MOV     t4FP@R1.w,    R1.w.1FP@R1.w
>            (copy) MOV     R1.z.1FP@R1.z,    t8FP@R1.z
>            (copy) MOV     t3FP@R1.z,    R1.z.1FP@R1.z
>            (copy) MOV     R1.y.1FP@R1.y,    t7FP@R1.y
>            (copy) MOV     t2FP@R1.y,    R1.y.1FP@R1.y
>            (copy) MOV     t1FP@R1.x,    R1.x.1FP@R1.x
>        >  live_after: [t1FP@R1.x t2FP@R1.y t3FP@R1.z t4FP@R1.w t5FP@R1.x ]
>        EXPORT  POS   60     t1FP@R1.x, t2FP@R1.y, t3FP@R1.z, t4FP@R1.w
>        MEM_STREAM0_BUF0  WRITE 0   ES:1     t5FP@R1.x, __, __, __
>        EXPORT  PARAM 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t6FP@R1.x, t7FP@R1.y, t8FP@R1.z, t9FP@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t6FP@R1.x t7FP@R1.y t8FP@R1.z t9FP@R1.w ]
>            (copy) MOV     R1.x.1FP@R1.x,    t6FP@R1.x
>            (copy) MOV     t5FP@R1.x,    R1.x.1FP@R1.x
>            (copy) MOV     R1.w.1FP@R1.w,    t9FP@R1.w
>            (copy) MOV     t4FP@R1.w,    R1.w.1FP@R1.w
>            (copy) MOV     R1.z.1FP@R1.z,    t8FP@R1.z
>            (copy) MOV     t3FP@R1.z,    R1.z.1FP@R1.z
>            (copy) MOV     R1.y.1FP@R1.y,    t7FP@R1.y
>            (copy) MOV     t2FP@R1.y,    R1.y.1FP@R1.y
>            (copy) MOV     t1FP@R1.x,    R1.x.1FP@R1.x
>        >  live_after: [t1FP@R1.x t2FP@R1.y t3FP@R1.z t4FP@R1.w t5FP@R1.x ]
>        EXPORT  POS   60     t1FP@R1.x, t2FP@R1.y, t3FP@R1.z, t4FP@R1.w
>        MEM_STREAM0_BUF0  WRITE 0   ES:1     t5FP@R1.x, __, __, __
>        EXPORT  PARAM 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t6||FP@R1.x, t7||FP@R1.y, t8||FP@R1.z, t9||FP@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        EXPORT  POS   60     t1||FP@R1.x, t2||FP@R1.y, t3||FP@R1.z, t4||FP@R1.w
>        MEM_STREAM0_BUF0  WRITE 0   ES:1     t5||FP@R1.x, __, __, __
>        EXPORT  PARAM 0     __, __, __, __
>    } end BB_0  
>}  
>
>===== SHADER #10 OPT ==================================== VS/CAYMAN/CAYMAN =====
>===== 10 dw ===== 2 gprs ===== 1 stack =========================================
>0000  00000000 84c00000 CALL_FS @0
>0002  c000a03c 95000688 EXPORT_DONE        POS   60    R1.xyzw
>0004  40008000 90001fff MEM_STREAM0_BUF0   WRITE 0     R1.x___  ES:1
>0006  c0004000 95000fff EXPORT_DONE        PARAM 0     R0.____
>0008  00000000 88000000 CF_END @0
>===== SHADER_END ===============================================================
>
>
>
>###### after parse
>{  
>    ALU     
>    <  
>        [  
>            INTERP_LOAD_P0     R1.x,    Param0x
>            INTERP_LOAD_P0     R1.y,    Param0y
>            INTERP_LOAD_P0     R1.z,    Param0z
>            INTERP_LOAD_P0     R1.w,    Param0w
>        ]  
>    >  
>    EXPORT  PIXEL 0     __, __, __, __
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{  
>    ALU     
>    <  
>        [  
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>        ]  
>    >  
>    EXPORT  PIXEL 0     __, __, __, __
>    CF_END     
>}  
>
>
>###### after if_conversion
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after peephole
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after gvn
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after def_use
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after dce_cleanup
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after gcm
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after parse
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>    CF_END     
>}  
>
>
>###### after if_conversion
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after peephole
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after gvn
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after def_use
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after dce_cleanup
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after gcm
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after parse
>{  
>    ALU     
>    <  
>        [  
>            INTERP_LOAD_P0     R1.x,    Param0x
>            INTERP_LOAD_P0     R1.y,    Param0y
>            INTERP_LOAD_P0     R1.z,    Param0z
>            INTERP_LOAD_P0     R1.w,    Param0w
>        ]  
>        [  
>            MOV     R2.x,    R1.x
>            MOV     R2.y,    R1.y
>            MOV     R2.z,    R1.z
>            MOV     R2.w,    R1.w
>        ]  
>    >  
>    EXPORT  PIXEL 0     R2.x, R2.y, R2.z, R2.w
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{  
>    ALU     
>    <  
>        [  
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>        ]  
>        [  
>            MOV     R2.x.1,    R1.x.1
>            MOV     R2.y.1,    R1.y.1
>            MOV     R2.z.1,    R1.z.1
>            MOV     R2.w.1,    R1.w.1
>        ]  
>    >  
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>    CF_END     
>}  
>
>
>###### after if_conversion
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after peephole
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after gvn
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after def_use
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after dce_cleanup
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after gcm
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            (copy) MOV     t4,    R1.w.1
>            (copy) MOV     t3,    R1.z.1
>            (copy) MOV     t2,    R1.y.1
>            (copy) MOV     t1,    R1.x.1
>        >  
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            (copy) MOV     t4,    R1.w.1
>            (copy) MOV     t3,    R1.z.1
>            (copy) MOV     t2,    R1.y.1
>            (copy) MOV     t1,    R1.x.1
>        >  live_after: [t1 t2 t3 t4 ]
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1FP@R0.w,    Param0w
>            INTERP_LOAD_P0     R1.z.1FP@R0.z,    Param0z
>            INTERP_LOAD_P0     R1.y.1FP@R0.y,    Param0y
>            INTERP_LOAD_P0     R1.x.1FP@R0.x,    Param0x
>            (copy) MOV     t4FP@R0.w,    R1.w.1FP@R0.w
>            (copy) MOV     t3FP@R0.z,    R1.z.1FP@R0.z
>            (copy) MOV     t2FP@R0.y,    R1.y.1FP@R0.y
>            (copy) MOV     t1FP@R0.x,    R1.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1FP@R0.w,    Param0w
>            INTERP_LOAD_P0     R1.z.1FP@R0.z,    Param0z
>            INTERP_LOAD_P0     R1.y.1FP@R0.y,    Param0y
>            INTERP_LOAD_P0     R1.x.1FP@R0.x,    Param0x
>            (copy) MOV     t4FP@R0.w,    R1.w.1FP@R0.w
>            (copy) MOV     t3FP@R0.z,    R1.z.1FP@R0.z
>            (copy) MOV     t2FP@R0.y,    R1.y.1FP@R0.y
>            (copy) MOV     t1FP@R0.x,    R1.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            [  
>                INTERP_LOAD_P0     R1.x.1||FP@R0.x,    Param0x
>                INTERP_LOAD_P0     R1.y.1||FP@R0.y,    Param0y
>                INTERP_LOAD_P0     R1.z.1||FP@R0.z,    Param0z
>                INTERP_LOAD_P0     R1.w.1||FP@R0.w,    Param0w
>            ]  
>        >  
>        EXPORT  PIXEL 0     t1||FP@R0.x, t2||FP@R0.y, t3||FP@R0.z, t4||FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after parse
>{  
>    ALU     
>    <  
>        [  
>            INTERP_LOAD_P0     R1.x,    Param0x
>            INTERP_LOAD_P0     R1.y,    Param0y
>            INTERP_LOAD_P0     R1.z,    Param0z
>            INTERP_LOAD_P0     R1.w,    Param0w
>        ]  
>        [  
>            MOV     R2.x,    R1.x
>            MOV     R2.y,    R1.y
>            MOV     R2.z,    R1.z
>            MOV     R2.w,    R1.w
>        ]  
>    >  
>    EXPORT  PIXEL 0     R2.x, R2.y, R2.z, R2.w
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{  
>    ALU     
>    <  
>        [  
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>        ]  
>        [  
>            MOV     R2.x.1,    R1.x.1
>            MOV     R2.y.1,    R1.y.1
>            MOV     R2.z.1,    R1.z.1
>            MOV     R2.w.1,    R1.w.1
>        ]  
>    >  
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>    CF_END     
>}  
>
>
>###### after if_conversion
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after peephole
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after gvn
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after def_use
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after dce_cleanup
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after gcm
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            (copy) MOV     t4,    R1.w.1
>            (copy) MOV     t3,    R1.z.1
>            (copy) MOV     t2,    R1.y.1
>            (copy) MOV     t1,    R1.x.1
>        >  
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            (copy) MOV     t4,    R1.w.1
>            (copy) MOV     t3,    R1.z.1
>            (copy) MOV     t2,    R1.y.1
>            (copy) MOV     t1,    R1.x.1
>        >  live_after: [t1 t2 t3 t4 ]
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1FP@R0.w,    Param0w
>            INTERP_LOAD_P0     R1.z.1FP@R0.z,    Param0z
>            INTERP_LOAD_P0     R1.y.1FP@R0.y,    Param0y
>            INTERP_LOAD_P0     R1.x.1FP@R0.x,    Param0x
>            (copy) MOV     t4FP@R0.w,    R1.w.1FP@R0.w
>            (copy) MOV     t3FP@R0.z,    R1.z.1FP@R0.z
>            (copy) MOV     t2FP@R0.y,    R1.y.1FP@R0.y
>            (copy) MOV     t1FP@R0.x,    R1.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1FP@R0.w,    Param0w
>            INTERP_LOAD_P0     R1.z.1FP@R0.z,    Param0z
>            INTERP_LOAD_P0     R1.y.1FP@R0.y,    Param0y
>            INTERP_LOAD_P0     R1.x.1FP@R0.x,    Param0x
>            (copy) MOV     t4FP@R0.w,    R1.w.1FP@R0.w
>            (copy) MOV     t3FP@R0.z,    R1.z.1FP@R0.z
>            (copy) MOV     t2FP@R0.y,    R1.y.1FP@R0.y
>            (copy) MOV     t1FP@R0.x,    R1.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            [  
>                INTERP_LOAD_P0     R1.x.1||FP@R0.x,    Param0x
>                INTERP_LOAD_P0     R1.y.1||FP@R0.y,    Param0y
>                INTERP_LOAD_P0     R1.z.1||FP@R0.z,    Param0z
>                INTERP_LOAD_P0     R1.w.1||FP@R0.w,    Param0w
>            ]  
>        >  
>        EXPORT  PIXEL 0     t1||FP@R0.x, t2||FP@R0.y, t3||FP@R0.z, t4||FP@R0.w
>    } end BB_0  
>}  
>--------------------------------------------------------------
>VERT
>DCL IN[0]
>DCL IN[1]
>DCL OUT[0], POSITION
>DCL OUT[1], GENERIC[0]
>  0: MOV OUT[0], IN[0]
>  1: MOV OUT[1], IN[1]
>  2: END
>
>===== SHADER #20 ======================================== VS/CAYMAN/CAYMAN =====
>===== 26 dw ===== 5 gprs ===== 1 stack =========================================
>0000  00000000 84c00000 CALL_FS @0
>0002  00000005 a01c0000 ALU 8 @10
> 0010  00000001 00600c90     1      x: MOV                R3.x,  R1.x
> 0012  00000401 20600c90            y: MOV                R3.y,  R1.y
> 0014  00000801 40600c90            z: MOV                R3.z,  R1.z
> 0016  80000c01 60600c90            w: MOV                R3.w,  R1.w
> 0018  00000002 00800c90     2      x: MOV                R4.x,  R2.x
> 0020  00000402 20800c90            y: MOV                R4.y,  R2.y
> 0022  00000802 40800c90            z: MOV                R4.z,  R2.z
> 0024  80000c02 60800c90            w: MOV                R4.w,  R2.w
>0004  c001a03c 95000688 EXPORT_DONE        POS   60    R3.xyzw
>0006  c0024000 95000688 EXPORT_DONE        PARAM 0     R4.xyzw
>0008  00000000 88000000 CF_END @0
>===== SHADER_END ===============================================================
>
>
>
>###### after parse
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.xF@R1.x, R1.yF@R1.y, R1.zF@R1.z, R1.wF@R1.w, R2.xF@R2.x, R2.yF@R2.y, R2.zF@R2.z, R2.wF@R2.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    ALU     
>    <  
>        [  
>            MOV     R3.x,    R1.xF@R1.x
>            MOV     R3.y,    R1.yF@R1.y
>            MOV     R3.z,    R1.zF@R1.z
>            MOV     R3.w,    R1.wF@R1.w
>        ]  
>        [  
>            MOV     R4.x,    R2.xF@R2.x
>            MOV     R4.y,    R2.yF@R2.y
>            MOV     R4.z,    R2.zF@R2.z
>            MOV     R4.w,    R2.wF@R2.w
>        ]  
>    >  
>    EXPORT  POS   60     R3.x, R3.y, R3.z, R3.w
>    EXPORT  PARAM 0     R4.x, R4.y, R4.z, R4.w
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1, R2.x.1, R2.y.1, R2.z.1, R2.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    ALU     
>    <  
>        [  
>            MOV     R3.x.1,    R1.x.1
>            MOV     R3.y.1,    R1.y.1
>            MOV     R3.z.1,    R1.z.1
>            MOV     R3.w.1,    R1.w.1
>        ]  
>        [  
>            MOV     R4.x.1,    R2.x.1
>            MOV     R4.y.1,    R2.y.1
>            MOV     R4.z.1,    R2.z.1
>            MOV     R4.w.1,    R2.w.1
>        ]  
>    >  
>    EXPORT  POS   60     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>    EXPORT  PARAM 0     R4.x.1, R4.y.1, R4.z.1, R4.w.1
>    CF_END     
>}  
>
>
>###### after if_conversion
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1, R2.x.1, R2.y.1, R2.z.1, R2.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R3.x.1,    R1.x.1
>    MOV     R3.y.1,    R1.y.1
>    MOV     R3.z.1,    R1.z.1
>    MOV     R3.w.1,    R1.w.1
>    MOV     R4.x.1,    R2.x.1
>    MOV     R4.y.1,    R2.y.1
>    MOV     R4.z.1,    R2.z.1
>    MOV     R4.w.1,    R2.w.1
>    EXPORT  POS   60     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>    EXPORT  PARAM 0     R4.x.1, R4.y.1, R4.z.1, R4.w.1
>}  
>
>
>###### after peephole
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1, R2.x.1, R2.y.1, R2.z.1, R2.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R3.x.1,    R1.x.1
>    MOV     R3.y.1,    R1.y.1
>    MOV     R3.z.1,    R1.z.1
>    MOV     R3.w.1,    R1.w.1
>    MOV     R4.x.1,    R2.x.1
>    MOV     R4.y.1,    R2.y.1
>    MOV     R4.z.1,    R2.z.1
>    MOV     R4.w.1,    R2.w.1
>    EXPORT  POS   60     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>    EXPORT  PARAM 0     R4.x.1, R4.y.1, R4.z.1, R4.w.1
>}  
>
>
>###### after gvn
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1, R2.x.1, R2.y.1, R2.z.1, R2.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R3.x.1,    R1.x.1
>    MOV     R3.y.1,    R1.y.1
>    MOV     R3.z.1,    R1.z.1
>    MOV     R3.w.1,    R1.w.1
>    MOV     R4.x.1,    R2.x.1
>    MOV     R4.y.1,    R2.y.1
>    MOV     R4.z.1,    R2.z.1
>    MOV     R4.w.1,    R2.w.1
>    EXPORT  POS   60     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>    EXPORT  PARAM 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after def_use
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1, R2.x.1, R2.y.1, R2.z.1, R2.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R3.x.1,    R1.x.1
>    MOV     R3.y.1,    R1.y.1
>    MOV     R3.z.1,    R1.z.1
>    MOV     R3.w.1,    R1.w.1
>    MOV     R4.x.1,    R2.x.1
>    MOV     R4.y.1,    R2.y.1
>    MOV     R4.z.1,    R2.z.1
>    MOV     R4.w.1,    R2.w.1
>    EXPORT  POS   60     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>    EXPORT  PARAM 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1, R2.x.1, R2.y.1, R2.z.1, R2.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    EXPORT  POS   60     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>    EXPORT  PARAM 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after gcm
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  
>        CALL_FS     t9F@R1.x, t10F@R1.y, t11F@R1.z, t12F@R1.w, t13F@R2.x, t14F@R2.y, t15F@R2.z, t16F@R2.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  
>            (copy) MOV     R1.w.1,    t12F@R1.w
>            (copy) MOV     t8,    R1.w.1
>            (copy) MOV     R1.z.1,    t11F@R1.z
>            (copy) MOV     t7,    R1.z.1
>            (copy) MOV     R1.y.1,    t10F@R1.y
>            (copy) MOV     t6,    R1.y.1
>            (copy) MOV     R1.x.1,    t9F@R1.x
>            (copy) MOV     t5,    R1.x.1
>            (copy) MOV     R2.w.1,    t16F@R2.w
>            (copy) MOV     t4,    R2.w.1
>            (copy) MOV     R2.z.1,    t15F@R2.z
>            (copy) MOV     t3,    R2.z.1
>            (copy) MOV     R2.y.1,    t14F@R2.y
>            (copy) MOV     t2,    R2.y.1
>            (copy) MOV     R2.x.1,    t13F@R2.x
>            (copy) MOV     t1,    R2.x.1
>        >  
>        EXPORT  PARAM 0     t1, t2, t3, t4
>        EXPORT  POS   60     t5, t6, t7, t8
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t9F@R1.x, t10F@R1.y, t11F@R1.z, t12F@R1.w, t13F@R2.x, t14F@R2.y, t15F@R2.z, t16F@R2.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t9F@R1.x t10F@R1.y t11F@R1.z t12F@R1.w t13F@R2.x t14F@R2.y t15F@R2.z t16F@R2.w ]
>            (copy) MOV     R1.w.1,    t12F@R1.w
>            (copy) MOV     t8,    R1.w.1
>            (copy) MOV     R1.z.1,    t11F@R1.z
>            (copy) MOV     t7,    R1.z.1
>            (copy) MOV     R1.y.1,    t10F@R1.y
>            (copy) MOV     t6,    R1.y.1
>            (copy) MOV     R1.x.1,    t9F@R1.x
>            (copy) MOV     t5,    R1.x.1
>            (copy) MOV     R2.w.1,    t16F@R2.w
>            (copy) MOV     t4,    R2.w.1
>            (copy) MOV     R2.z.1,    t15F@R2.z
>            (copy) MOV     t3,    R2.z.1
>            (copy) MOV     R2.y.1,    t14F@R2.y
>            (copy) MOV     t2,    R2.y.1
>            (copy) MOV     R2.x.1,    t13F@R2.x
>            (copy) MOV     t1,    R2.x.1
>        >  live_after: [t1 t2 t3 t4 t5 t6 t7 t8 ]
>        EXPORT  PARAM 0     t1, t2, t3, t4
>        EXPORT  POS   60     t5, t6, t7, t8
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t9FP@R1.x, t10FP@R1.y, t11FP@R1.z, t12FP@R1.w, t13FP@R2.x, t14FP@R2.y, t15FP@R2.z, t16FP@R2.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t9FP@R1.x t10FP@R1.y t11FP@R1.z t12FP@R1.w t13FP@R2.x t14FP@R2.y t15FP@R2.z t16FP@R2.w ]
>            (copy) MOV     R1.w.1FP@R1.w,    t12FP@R1.w
>            (copy) MOV     t8FP@R1.w,    R1.w.1FP@R1.w
>            (copy) MOV     R1.z.1FP@R1.z,    t11FP@R1.z
>            (copy) MOV     t7FP@R1.z,    R1.z.1FP@R1.z
>            (copy) MOV     R1.y.1FP@R1.y,    t10FP@R1.y
>            (copy) MOV     t6FP@R1.y,    R1.y.1FP@R1.y
>            (copy) MOV     R1.x.1FP@R1.x,    t9FP@R1.x
>            (copy) MOV     t5FP@R1.x,    R1.x.1FP@R1.x
>            (copy) MOV     R2.w.1FP@R2.w,    t16FP@R2.w
>            (copy) MOV     t4FP@R2.w,    R2.w.1FP@R2.w
>            (copy) MOV     R2.z.1FP@R2.z,    t15FP@R2.z
>            (copy) MOV     t3FP@R2.z,    R2.z.1FP@R2.z
>            (copy) MOV     R2.y.1FP@R2.y,    t14FP@R2.y
>            (copy) MOV     t2FP@R2.y,    R2.y.1FP@R2.y
>            (copy) MOV     R2.x.1FP@R2.x,    t13FP@R2.x
>            (copy) MOV     t1FP@R2.x,    R2.x.1FP@R2.x
>        >  live_after: [t1FP@R2.x t2FP@R2.y t3FP@R2.z t4FP@R2.w t5FP@R1.x t6FP@R1.y t7FP@R1.z t8FP@R1.w ]
>        EXPORT  PARAM 0     t1FP@R2.x, t2FP@R2.y, t3FP@R2.z, t4FP@R2.w
>        EXPORT  POS   60     t5FP@R1.x, t6FP@R1.y, t7FP@R1.z, t8FP@R1.w
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t9FP@R1.x, t10FP@R1.y, t11FP@R1.z, t12FP@R1.w, t13FP@R2.x, t14FP@R2.y, t15FP@R2.z, t16FP@R2.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t9FP@R1.x t10FP@R1.y t11FP@R1.z t12FP@R1.w t13FP@R2.x t14FP@R2.y t15FP@R2.z t16FP@R2.w ]
>            (copy) MOV     R1.w.1FP@R1.w,    t12FP@R1.w
>            (copy) MOV     t8FP@R1.w,    R1.w.1FP@R1.w
>            (copy) MOV     R1.z.1FP@R1.z,    t11FP@R1.z
>            (copy) MOV     t7FP@R1.z,    R1.z.1FP@R1.z
>            (copy) MOV     R1.y.1FP@R1.y,    t10FP@R1.y
>            (copy) MOV     t6FP@R1.y,    R1.y.1FP@R1.y
>            (copy) MOV     R1.x.1FP@R1.x,    t9FP@R1.x
>            (copy) MOV     t5FP@R1.x,    R1.x.1FP@R1.x
>            (copy) MOV     R2.w.1FP@R2.w,    t16FP@R2.w
>            (copy) MOV     t4FP@R2.w,    R2.w.1FP@R2.w
>            (copy) MOV     R2.z.1FP@R2.z,    t15FP@R2.z
>            (copy) MOV     t3FP@R2.z,    R2.z.1FP@R2.z
>            (copy) MOV     R2.y.1FP@R2.y,    t14FP@R2.y
>            (copy) MOV     t2FP@R2.y,    R2.y.1FP@R2.y
>            (copy) MOV     R2.x.1FP@R2.x,    t13FP@R2.x
>            (copy) MOV     t1FP@R2.x,    R2.x.1FP@R2.x
>        >  live_after: [t1FP@R2.x t2FP@R2.y t3FP@R2.z t4FP@R2.w t5FP@R1.x t6FP@R1.y t7FP@R1.z t8FP@R1.w ]
>        EXPORT  PARAM 0     t1FP@R2.x, t2FP@R2.y, t3FP@R2.z, t4FP@R2.w
>        EXPORT  POS   60     t5FP@R1.x, t6FP@R1.y, t7FP@R1.z, t8FP@R1.w
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t9||FP@R1.x, t10||FP@R1.y, t11||FP@R1.z, t12||FP@R1.w, t13||FP@R2.x, t14||FP@R2.y, t15||FP@R2.z, t16||FP@R2.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        EXPORT  PARAM 0     t1||FP@R2.x, t2||FP@R2.y, t3||FP@R2.z, t4||FP@R2.w
>        EXPORT  POS   60     t5||FP@R1.x, t6||FP@R1.y, t7||FP@R1.z, t8||FP@R1.w
>    } end BB_0  
>}  
>
>===== SHADER #20 OPT ==================================== VS/CAYMAN/CAYMAN =====
>===== 8 dw ===== 3 gprs ===== 1 stack ==========================================
>0000  00000000 84c00000 CALL_FS @0
>0002  c0014000 95000688 EXPORT_DONE        PARAM 0     R2.xyzw
>0004  c000a03c 95000688 EXPORT_DONE        POS   60    R1.xyzw
>0006  00000000 88000000 CF_END @0
>===== SHADER_END ===============================================================
>
>--------------------------------------------------------------
>VERT
>DCL IN[0]
>DCL OUT[0], POSITION
>  0: MOV OUT[0], IN[0]
>  1: END
>STREAMOUT
>  0: MEM_STREAM0_BUF0[0..0] <- OUT[0].x
>
>===== SHADER #21 ======================================== VS/CAYMAN/CAYMAN =====
>===== 20 dw ===== 3 gprs ===== 1 stack =========================================
>0000  00000000 84c00000 CALL_FS @0
>0002  00000006 a00c0000 ALU 4 @12
> 0012  00000001 00400c90     1      x: MOV                R2.x,  R1.x
> 0014  00000401 20400c90            y: MOV                R2.y,  R1.y
> 0016  00000801 40400c90            z: MOV                R2.z,  R1.z
> 0018  80000c01 60400c90            w: MOV                R2.w,  R1.w
>0004  40010000 90001fff MEM_STREAM0_BUF0   WRITE 0     R2.x___  ES:1
>0006  c001203c 95000688 EXPORT_DONE        POS   60    R2.xyzw
>0008  c0004000 95000fff EXPORT_DONE        PARAM 0     R0.____
>0010  00000000 88000000 CF_END @0
>===== SHADER_END ===============================================================
>
>
>
>###### after parse
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.xF@R1.x, R1.yF@R1.y, R1.zF@R1.z, R1.wF@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    ALU     
>    <  
>        [  
>            MOV     R2.x,    R1.xF@R1.x
>            MOV     R2.y,    R1.yF@R1.y
>            MOV     R2.z,    R1.zF@R1.z
>            MOV     R2.w,    R1.wF@R1.w
>        ]  
>    >  
>    MEM_STREAM0_BUF0  WRITE 0   ES:1     R2.x, __, __, __
>    EXPORT  POS   60     R2.x, R2.y, R2.z, R2.w
>    EXPORT  PARAM 0     __, __, __, __
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    ALU     
>    <  
>        [  
>            MOV     R2.x.1,    R1.x.1
>            MOV     R2.y.1,    R1.y.1
>            MOV     R2.z.1,    R1.z.1
>            MOV     R2.w.1,    R1.w.1
>        ]  
>    >  
>    MEM_STREAM0_BUF0  WRITE 0   ES:1     R2.x.1, __, __, __
>    EXPORT  POS   60     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>    EXPORT  PARAM 0     __, __, __, __
>    CF_END     
>}  
>
>
>###### after if_conversion
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    MEM_STREAM0_BUF0  WRITE 0   ES:1     R2.x.1, __, __, __
>    EXPORT  POS   60     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>    EXPORT  PARAM 0     __, __, __, __
>}  
>
>
>###### after peephole
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    MEM_STREAM0_BUF0  WRITE 0   ES:1     R2.x.1, __, __, __
>    EXPORT  POS   60     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>    EXPORT  PARAM 0     __, __, __, __
>}  
>
>
>###### after gvn
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    MEM_STREAM0_BUF0  WRITE 0   ES:1     R1.x.1, __, __, __
>    EXPORT  POS   60     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>    EXPORT  PARAM 0     __, __, __, __
>}  
>
>
>###### after def_use
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    MEM_STREAM0_BUF0  WRITE 0   ES:1     R1.x.1, __, __, __
>    EXPORT  POS   60     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>    EXPORT  PARAM 0     __, __, __, __
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MEM_STREAM0_BUF0  WRITE 0   ES:1     R1.x.1, __, __, __
>    EXPORT  POS   60     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>    EXPORT  PARAM 0     __, __, __, __
>}  
>
>
>###### after gcm
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  
>        CALL_FS     t6F@R1.x, t7F@R1.y, t8F@R1.z, t9F@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  
>            (copy) MOV     R1.x.1,    t6F@R1.x
>            (copy) MOV     t5,    R1.x.1
>            (copy) MOV     R1.w.1,    t9F@R1.w
>            (copy) MOV     t4,    R1.w.1
>            (copy) MOV     R1.z.1,    t8F@R1.z
>            (copy) MOV     t3,    R1.z.1
>            (copy) MOV     R1.y.1,    t7F@R1.y
>            (copy) MOV     t2,    R1.y.1
>            (copy) MOV     t1,    R1.x.1
>        >  
>        EXPORT  POS   60     t1, t2, t3, t4
>        MEM_STREAM0_BUF0  WRITE 0   ES:1     t5, __, __, __
>        EXPORT  PARAM 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t6F@R1.x, t7F@R1.y, t8F@R1.z, t9F@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t6F@R1.x t7F@R1.y t8F@R1.z t9F@R1.w ]
>            (copy) MOV     R1.x.1,    t6F@R1.x
>            (copy) MOV     t5,    R1.x.1
>            (copy) MOV     R1.w.1,    t9F@R1.w
>            (copy) MOV     t4,    R1.w.1
>            (copy) MOV     R1.z.1,    t8F@R1.z
>            (copy) MOV     t3,    R1.z.1
>            (copy) MOV     R1.y.1,    t7F@R1.y
>            (copy) MOV     t2,    R1.y.1
>            (copy) MOV     t1,    R1.x.1
>        >  live_after: [t1 t2 t3 t4 t5 ]
>        EXPORT  POS   60     t1, t2, t3, t4
>        MEM_STREAM0_BUF0  WRITE 0   ES:1     t5, __, __, __
>        EXPORT  PARAM 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t6FP@R1.x, t7FP@R1.y, t8FP@R1.z, t9FP@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t6FP@R1.x t7FP@R1.y t8FP@R1.z t9FP@R1.w ]
>            (copy) MOV     R1.x.1FP@R1.x,    t6FP@R1.x
>            (copy) MOV     t5FP@R1.x,    R1.x.1FP@R1.x
>            (copy) MOV     R1.w.1FP@R1.w,    t9FP@R1.w
>            (copy) MOV     t4FP@R1.w,    R1.w.1FP@R1.w
>            (copy) MOV     R1.z.1FP@R1.z,    t8FP@R1.z
>            (copy) MOV     t3FP@R1.z,    R1.z.1FP@R1.z
>            (copy) MOV     R1.y.1FP@R1.y,    t7FP@R1.y
>            (copy) MOV     t2FP@R1.y,    R1.y.1FP@R1.y
>            (copy) MOV     t1FP@R1.x,    R1.x.1FP@R1.x
>        >  live_after: [t1FP@R1.x t2FP@R1.y t3FP@R1.z t4FP@R1.w t5FP@R1.x ]
>        EXPORT  POS   60     t1FP@R1.x, t2FP@R1.y, t3FP@R1.z, t4FP@R1.w
>        MEM_STREAM0_BUF0  WRITE 0   ES:1     t5FP@R1.x, __, __, __
>        EXPORT  PARAM 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t6FP@R1.x, t7FP@R1.y, t8FP@R1.z, t9FP@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t6FP@R1.x t7FP@R1.y t8FP@R1.z t9FP@R1.w ]
>            (copy) MOV     R1.x.1FP@R1.x,    t6FP@R1.x
>            (copy) MOV     t5FP@R1.x,    R1.x.1FP@R1.x
>            (copy) MOV     R1.w.1FP@R1.w,    t9FP@R1.w
>            (copy) MOV     t4FP@R1.w,    R1.w.1FP@R1.w
>            (copy) MOV     R1.z.1FP@R1.z,    t8FP@R1.z
>            (copy) MOV     t3FP@R1.z,    R1.z.1FP@R1.z
>            (copy) MOV     R1.y.1FP@R1.y,    t7FP@R1.y
>            (copy) MOV     t2FP@R1.y,    R1.y.1FP@R1.y
>            (copy) MOV     t1FP@R1.x,    R1.x.1FP@R1.x
>        >  live_after: [t1FP@R1.x t2FP@R1.y t3FP@R1.z t4FP@R1.w t5FP@R1.x ]
>        EXPORT  POS   60     t1FP@R1.x, t2FP@R1.y, t3FP@R1.z, t4FP@R1.w
>        MEM_STREAM0_BUF0  WRITE 0   ES:1     t5FP@R1.x, __, __, __
>        EXPORT  PARAM 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t6||FP@R1.x, t7||FP@R1.y, t8||FP@R1.z, t9||FP@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        EXPORT  POS   60     t1||FP@R1.x, t2||FP@R1.y, t3||FP@R1.z, t4||FP@R1.w
>        MEM_STREAM0_BUF0  WRITE 0   ES:1     t5||FP@R1.x, __, __, __
>        EXPORT  PARAM 0     __, __, __, __
>    } end BB_0  
>}  
>
>===== SHADER #21 OPT ==================================== VS/CAYMAN/CAYMAN =====
>===== 10 dw ===== 2 gprs ===== 1 stack =========================================
>0000  00000000 84c00000 CALL_FS @0
>0002  c000a03c 95000688 EXPORT_DONE        POS   60    R1.xyzw
>0004  40008000 90001fff MEM_STREAM0_BUF0   WRITE 0     R1.x___  ES:1
>0006  c0004000 95000fff EXPORT_DONE        PARAM 0     R0.____
>0008  00000000 88000000 CF_END @0
>===== SHADER_END ===============================================================
>
>
>
>###### after parse
>{  
>    ALU     
>    <  
>        [  
>            INTERP_LOAD_P0     R1.x,    Param0x
>            INTERP_LOAD_P0     R1.y,    Param0y
>            INTERP_LOAD_P0     R1.z,    Param0z
>            INTERP_LOAD_P0     R1.w,    Param0w
>        ]  
>    >  
>    EXPORT  PIXEL 0     __, __, __, __
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{  
>    ALU     
>    <  
>        [  
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>        ]  
>    >  
>    EXPORT  PIXEL 0     __, __, __, __
>    CF_END     
>}  
>
>
>###### after if_conversion
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after peephole
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after gvn
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after def_use
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after dce_cleanup
>{  
>    EXPORT  PIXEL 0     __, __, __, __
>}  
>
>
>###### after gcm
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{  
>    { BB_0    loop_level = 0  
>        EXPORT  PIXEL 0     __, __, __, __
>    } end BB_0  
>}  
>
>
>###### after parse
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    ALU     
>    <  
>        [  
>            INTERP_ZW       
>                4S  INTERP_ZW     __,    R0.yF@R0.y, Param0x
>                4S  INTERP_ZW     __,    R0.xF@R0.x, Param0y
>                4S  INTERP_ZW     R2.z,    R0.yF@R0.y, Param0z
>                4S  INTERP_ZW     R2.w,    R0.xF@R0.x, Param0w
>        ]  
>        [  
>            INTERP_XY       
>                4S  INTERP_XY     R2.x,    R0.yF@R0.y, Param0x
>                4S  INTERP_XY     R2.y,    R0.xF@R0.x, Param0y
>                4S  INTERP_XY     __,    R0.yF@R0.y, Param0z
>                4S  INTERP_XY     __,    R0.xF@R0.x, Param0w
>        ]  
>        [  
>            MOV     R3.x,    R2.x
>            MOV     R3.y,    R2.y
>            MOV     R3.z,    R2.z
>            MOV     R3.w,    R2.w
>        ]  
>    >  
>    EXPORT  PIXEL 0     R3.x, R3.y, R3.z, R3.w
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    ALU     
>    <  
>        [  
>            INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>        ]  
>        [  
>            INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>        ]  
>        [  
>            MOV     R3.x.1,    R2.x.1
>            MOV     R3.y.1,    R2.y.1
>            MOV     R3.z.1,    R2.z.1
>            MOV     R3.w.1,    R2.w.1
>        ]  
>    >  
>    EXPORT  PIXEL 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>    CF_END     
>}  
>
>
>###### after if_conversion
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    MOV     R3.x.1,    R2.x.1
>    MOV     R3.y.1,    R2.y.1
>    MOV     R3.z.1,    R2.z.1
>    MOV     R3.w.1,    R2.w.1
>    EXPORT  PIXEL 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>}  
>
>
>###### after peephole
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    MOV     R3.x.1,    R2.x.1
>    MOV     R3.y.1,    R2.y.1
>    MOV     R3.z.1,    R2.z.1
>    MOV     R3.w.1,    R2.w.1
>    EXPORT  PIXEL 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>}  
>
>
>###### after gvn
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    MOV     R3.x.1,    R2.x.1
>    MOV     R3.y.1,    R2.y.1
>    MOV     R3.z.1,    R2.z.1
>    MOV     R3.w.1,    R2.w.1
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after def_use
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    MOV     R3.x.1,    R2.x.1
>    MOV     R3.y.1,    R2.y.1
>    MOV     R3.z.1,    R2.z.1
>    MOV     R3.w.1,    R2.w.1
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after gcm
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            (copy) MOV     t4,    R2.w.1
>            (copy) MOV     t3,    R2.z.1
>            (copy) MOV     t2,    R2.y.1
>            (copy) MOV     t1,    R2.x.1
>        >  
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>        ALU     
>        <  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>            INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            (copy) MOV     t4,    R2.w.1
>            (copy) MOV     t3,    R2.z.1
>            (copy) MOV     t2,    R2.y.1
>            (copy) MOV     t1,    R2.x.1
>        >  live_after: [t1 t2 t3 t4 ]
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>        ALU     
>        <  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>            INTERP_ZW     __, __, R2.z.1FP@R0.z, R2.w.1FP@R0.w,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            INTERP_XY     R2.x.1FP@R0.x, R2.y.1FP@R0.y, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            (copy) MOV     t4FP@R0.w,    R2.w.1FP@R0.w
>            (copy) MOV     t3FP@R0.z,    R2.z.1FP@R0.z
>            (copy) MOV     t2FP@R0.y,    R2.y.1FP@R0.y
>            (copy) MOV     t1FP@R0.x,    R2.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>        ALU     
>        <  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>            INTERP_ZW     __, __, R2.z.1FP@R0.z, R2.w.1FP@R0.w,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            INTERP_XY     R2.x.1FP@R0.x, R2.y.1FP@R0.y, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            (copy) MOV     t4FP@R0.w,    R2.w.1FP@R0.w
>            (copy) MOV     t3FP@R0.z,    R2.z.1FP@R0.z
>            (copy) MOV     t2FP@R0.y,    R2.y.1FP@R0.y
>            (copy) MOV     t1FP@R0.x,    R2.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{   preloaded inputs [R0.x||FP@R0.x, R0.y||FP@R0.y]  
>    { BB_0    loop_level = 0  live_before: [R0.x||FP@R0.x R0.y||FP@R0.y ]
>        ALU     
>        <  
>            [  
>                4S  INTERP_ZW     __,    R0.y||FP@R0.y, Param0x
>                4S  INTERP_ZW     __,    R0.x||FP@R0.x, Param0y
>                4S  INTERP_ZW     R2.z.1||FP@R0.z,    R0.y||FP@R0.y, Param0z
>                4S  INTERP_ZW     R2.w.1||FP@R0.w,    R0.x||FP@R0.x, Param0w
>            ]  
>            [  
>                4S  INTERP_XY     R2.x.1||FP@R0.x,    R0.y||FP@R0.y, Param0x
>                4S  INTERP_XY     R2.y.1||FP@R0.y,    R0.x||FP@R0.x, Param0y
>                4S  INTERP_XY     __,    R0.y||FP@R0.y, Param0z
>                4S  INTERP_XY     __,    R0.x||FP@R0.x, Param0w
>            ]  
>        >  
>        EXPORT  PIXEL 0     t1||FP@R0.x, t2||FP@R0.y, t3||FP@R0.z, t4||FP@R0.w
>    } end BB_0  
>}  
>--------------------------------------------------------------
>VERT
>DCL IN[0]
>DCL OUT[0], POSITION
>DCL OUT[1], GENERIC[19]
>DCL CONST[0..6]
>DCL TEMP[0], LOCAL
>DCL TEMP[1..3], ARRAY(1), LOCAL
>DCL TEMP[4..12], ARRAY(2), LOCAL
>DCL TEMP[13..15], ARRAY(3), LOCAL
>DCL TEMP[16..18], ARRAY(4), LOCAL
>DCL TEMP[19..20], LOCAL
>DCL ADDR[0]
>IMM[0] FLT32 {    1.0000,     2.0000,     3.0000,     0.0000}
>IMM[1] FLT32 {    4.0000,     5.0000,     6.0000,    25.0000}
>IMM[2] FLT32 {    7.0000,     8.0000,     9.0000,    25.0000}
>IMM[3] FLT32 {   10.0000,    11.0000,    12.0000,    25.0000}
>IMM[4] FLT32 {   13.0000,    14.0000,    15.0000,    25.0000}
>IMM[5] FLT32 {   16.0000,    17.0000,    18.0000,    25.0000}
>IMM[6] FLT32 {   19.0000,    20.0000,    21.0000,    25.0000}
>IMM[7] FLT32 {   22.0000,    23.0000,    24.0000,    25.0000}
>IMM[8] FLT32 {   25.0000,    26.0000,    27.0000,     0.0000}
>IMM[9] INT32 {3, 0, 0, 0}
>  0: MUL TEMP[0], CONST[3], IN[0].xxxx
>  1: MAD TEMP[0], CONST[4], IN[0].yyyy, TEMP[0]
>  2: MAD TEMP[0], CONST[5], IN[0].zzzz, TEMP[0]
>  3: MAD TEMP[0], CONST[6], IN[0].wwww, TEMP[0]
>  4: MOV TEMP[1], IMM[0].xyzx
>  5: MOV TEMP[2], IMM[1].xyzx
>  6: MOV TEMP[3], IMM[2].xyzx
>  7: MOV TEMP[4], TEMP[1].xyzz
>  8: MOV TEMP[5], TEMP[2].xyzz
>  9: MOV TEMP[6], TEMP[3].xyzz
> 10: MOV TEMP[13], IMM[3].xyzx
> 11: MOV TEMP[14], IMM[4].xyzx
> 12: MOV TEMP[15], IMM[5].xyzx
> 13: MOV TEMP[7], TEMP[13].xyzz
> 14: MOV TEMP[8], TEMP[14].xyzz
> 15: MOV TEMP[9], TEMP[15].xyzz
> 16: MOV TEMP[16], IMM[6].xyzx
> 17: MOV TEMP[17], IMM[7].xyzx
> 18: MOV TEMP[18], IMM[8].xyzx
> 19: MOV TEMP[10], TEMP[16].xyzz
> 20: MOV TEMP[11], TEMP[17].xyzz
> 21: MOV TEMP[12], TEMP[18].xyzz
> 22: UMUL TEMP[19].x, CONST[0].xxxx, IMM[9].xxxx
> 23: UADD TEMP[19].x, CONST[1].xxxx, TEMP[19].xxxx
> 24: UARL ADDR[0].x, TEMP[19].xxxx
> 25: FSEQ TEMP[20].xyz, TEMP[ADDR[0].x+4](2).xyzz, CONST[2].xyzz
> 26: AND TEMP[20].y, TEMP[20].yyyy, TEMP[20].zzzz
> 27: AND TEMP[19].x, TEMP[20].xxxx, TEMP[20].yyyy
> 28: UIF TEMP[19].xxxx :0
> 29:   MOV TEMP[19], IMM[0].wxwx
> 30: ELSE :0
> 31:   MOV TEMP[19], IMM[0].xwwx
> 32: ENDIF
> 33: MOV OUT[1], TEMP[19]
> 34: MOV OUT[0], TEMP[0]
> 35: END
>
>===== SHADER #24 ======================================== VS/CAYMAN/CAYMAN =====
>===== 292 dw ===== 27 gprs ===== 1 stack =======================================
>0000  00000000 84c00000 CALL_FS @0
>0002  4000000b a1bc0000 ALU 112 @22 KC0[CB0:0-15]
> 0022  00002083 00800090     1      x: MUL                R4.x,  KC0[3].x, R1.x
> 0024  00002483 20800090            y: MUL                R4.y,  KC0[3].y, R1.x
> 0026  00002883 40800090            z: MUL                R4.z,  KC0[3].z, R1.x
> 0028  80002c83 60800090            w: MUL                R4.w,  KC0[3].w, R1.x
> 0030  00802084 008280fe     2      x: MULADD             R4.x,  KC0[4].x, R1.y, PV.x
> 0032  00802484 208284fe            y: MULADD             R4.y,  KC0[4].y, R1.y, PV.y
> 0034  00802884 408288fe            z: MULADD             R4.z,  KC0[4].z, R1.y, PV.z
> 0036  80802c84 60828cfe            w: MULADD             R4.w,  KC0[4].w, R1.y, PV.w
> 0038  01002085 008280fe     3      x: MULADD             R4.x,  KC0[5].x, R1.z, PV.x
> 0040  01002485 208284fe            y: MULADD             R4.y,  KC0[5].y, R1.z, PV.y
> 0042  01002885 408288fe            z: MULADD             R4.z,  KC0[5].z, R1.z, PV.z
> 0044  81002c85 60828cfe            w: MULADD             R4.w,  KC0[5].w, R1.z, PV.w
> 0046  01802086 008280fe     4      x: MULADD             R4.x,  KC0[6].x, R1.w, PV.x
> 0048  01802486 208284fe            y: MULADD             R4.y,  KC0[6].y, R1.w, PV.y
> 0050  01802886 408288fe            z: MULADD             R4.z,  KC0[6].z, R1.w, PV.z
> 0052  81802c86 60828cfe            w: MULADD             R4.w,  KC0[6].w, R1.w, PV.w
> 0054  000000f9 00a00c90     5      x: MOV                R5.x,  1.0
> 0056  000000fd 20a00c90            y: MOV                R5.y,  [0x40000000 2].x
> 0058  000004fd 40a00c90            z: MOV                R5.z,  [0x40400000 3].y
> 0060  800000f9 60a00c90            w: MOV                R5.w,  1.0
> 0062  40000000 
> 0063  40400000 
> 0064  000000fd 00c00c90     6      x: MOV                R6.x,  [0x40800000 4].x
> 0066  000004fd 20c00c90            y: MOV                R6.y,  [0x40a00000 5].y
> 0068  000008fd 40c00c90            z: MOV                R6.z,  [0x40c00000 6].z
> 0070  800000fd 60c00c90            w: MOV                R6.w,  [0x40800000 4].x
> 0072  40800000 
> 0073  40a00000 
> 0074  40c00000 
> 0076  000000fd 00e00c90     7      x: MOV                R7.x,  [0x40e00000 7].x
> 0078  000004fd 20e00c90            y: MOV                R7.y,  [0x41000000 8].y
> 0080  000008fd 40e00c90            z: MOV                R7.z,  [0x41100000 9].z
> 0082  800000fd 60e00c90            w: MOV                R7.w,  [0x40e00000 7].x
> 0084  40e00000 
> 0085  41000000 
> 0086  41100000 
> 0088  00000005 01000c90     8      x: MOV                R8.x,  R5.x
> 0090  00000405 21000c90            y: MOV                R8.y,  R5.y
> 0092  00000805 41000c90            z: MOV                R8.z,  R5.z
> 0094  80000805 61000c90            w: MOV                R8.w,  R5.z
> 0096  00000006 01200c90     9      x: MOV                R9.x,  R6.x
> 0098  00000406 21200c90            y: MOV                R9.y,  R6.y
> 0100  00000806 41200c90            z: MOV                R9.z,  R6.z
> 0102  80000806 61200c90            w: MOV                R9.w,  R6.z
> 0104  00000007 01400c90    10      x: MOV                R10.x,  R7.x
> 0106  00000407 21400c90            y: MOV                R10.y,  R7.y
> 0108  00000807 41400c90            z: MOV                R10.z,  R7.z
> 0110  80000807 61400c90            w: MOV                R10.w,  R7.z
> 0112  000000fd 02200c90    11      x: MOV                R17.x,  [0x41200000 10].x
> 0114  000004fd 22200c90            y: MOV                R17.y,  [0x41300000 11].y
> 0116  000008fd 42200c90            z: MOV                R17.z,  [0x41400000 12].z
> 0118  800000fd 62200c90            w: MOV                R17.w,  [0x41200000 10].x
> 0120  41200000 
> 0121  41300000 
> 0122  41400000 
> 0124  000000fd 02400c90    12      x: MOV                R18.x,  [0x41500000 13].x
> 0126  000004fd 22400c90            y: MOV                R18.y,  [0x41600000 14].y
> 0128  000008fd 42400c90            z: MOV                R18.z,  [0x41700000 15].z
> 0130  800000fd 62400c90            w: MOV                R18.w,  [0x41500000 13].x
> 0132  41500000 
> 0133  41600000 
> 0134  41700000 
> 0136  000000fd 02600c90    13      x: MOV                R19.x,  [0x41800000 16].x
> 0138  000004fd 22600c90            y: MOV                R19.y,  [0x41880000 17].y
> 0140  000008fd 42600c90            z: MOV                R19.z,  [0x41900000 18].z
> 0142  800000fd 62600c90            w: MOV                R19.w,  [0x41800000 16].x
> 0144  41800000 
> 0145  41880000 
> 0146  41900000 
> 0148  00000011 01600c90    14      x: MOV                R11.x,  R17.x
> 0150  00000411 21600c90            y: MOV                R11.y,  R17.y
> 0152  00000811 41600c90            z: MOV                R11.z,  R17.z
> 0154  80000811 61600c90            w: MOV                R11.w,  R17.z
> 0156  00000012 01800c90    15      x: MOV                R12.x,  R18.x
> 0158  00000412 21800c90            y: MOV                R12.y,  R18.y
> 0160  00000812 41800c90            z: MOV                R12.z,  R18.z
> 0162  80000812 61800c90            w: MOV                R12.w,  R18.z
> 0164  00000013 01a00c90    16      x: MOV                R13.x,  R19.x
> 0166  00000413 21a00c90            y: MOV                R13.y,  R19.y
> 0168  00000813 41a00c90            z: MOV                R13.z,  R19.z
> 0170  80000813 61a00c90            w: MOV                R13.w,  R19.z
> 0172  000000fd 02800c90    17      x: MOV                R20.x,  [0x41980000 19].x
> 0174  000004fd 22800c90            y: MOV                R20.y,  [0x41a00000 20].y
> 0176  000008fd 42800c90            z: MOV                R20.z,  [0x41a80000 21].z
> 0178  800000fd 62800c90            w: MOV                R20.w,  [0x41980000 19].x
> 0180  41980000 
> 0181  41a00000 
> 0182  41a80000 
> 0184  000000fd 02a00c90    18      x: MOV                R21.x,  [0x41b00000 22].x
> 0186  000004fd 22a00c90            y: MOV                R21.y,  [0x41b80000 23].y
> 0188  000008fd 42a00c90            z: MOV                R21.z,  [0x41c00000 24].z
> 0190  800000fd 62a00c90            w: MOV                R21.w,  [0x41b00000 22].x
> 0192  41b00000 
> 0193  41b80000 
> 0194  41c00000 
> 0196  000000fd 02c00c90    19      x: MOV                R22.x,  [0x41c80000 25].x
> 0198  000004fd 22c00c90            y: MOV                R22.y,  [0x41d00000 26].y
> 0200  000008fd 42c00c90            z: MOV                R22.z,  [0x41d80000 27].z
> 0202  800000fd 62c00c90            w: MOV                R22.w,  [0x41c80000 25].x
> 0204  41c80000 
> 0205  41d00000 
> 0206  41d80000 
> 0208  00000014 01c00c90    20      x: MOV                R14.x,  R20.x
> 0210  00000414 21c00c90            y: MOV                R14.y,  R20.y
> 0212  00000814 41c00c90            z: MOV                R14.z,  R20.z
> 0214  80000814 61c00c90            w: MOV                R14.w,  R20.z
> 0216  00000015 01e00c90    21      x: MOV                R15.x,  R21.x
> 0218  00000415 21e00c90            y: MOV                R15.y,  R21.y
> 0220  00000815 41e00c90            z: MOV                R15.z,  R21.z
> 0222  80000815 61e00c90            w: MOV                R15.w,  R21.z
> 0224  00000016 02000c90    22      x: MOV                R16.x,  R22.x
> 0226  00000416 22000c90            y: MOV                R16.y,  R22.y
> 0228  00000816 42000c90            z: MOV                R16.z,  R22.z
> 0230  80000816 62000c90            w: MOV                R16.w,  R22.z
> 0232  001fa080 02e04790    23      x: MULLO_INT          R23.x,  KC0[0].x, [0x00000003 4.2039e-45].x
> 0234  001fa080 22e04780            y: MULLO_INT          __.y,  KC0[0].x, [0x00000003 4.2039e-45].x
> 0236  001fa080 42e04780            z: MULLO_INT          __.z,  KC0[0].x, [0x00000003 4.2039e-45].x
> 0238  801fa080 62e04780            w: MULLO_INT          __.w,  KC0[0].x, [0x00000003 4.2039e-45].x
> 0240  00000003 
> 0242  801fc081 02e01a10    24      x: ADD_INT            R23.x,  KC0[1].x, PV.x
> 0244  800000fe 03200c90    25      x: MOV                R25.x,  PV.x
>0004  4000007b a4180000 ALU_PUSH_BEFORE 7 @246 KC0[CB0:0-15]
> 0246  80000019 00006600    26      x: MOVA_INT           __.x,  R25.x
> 0248  00104208 03000610    27      x: SETE_DX10          R24.x,  R[8+AR].x, KC0[2].x
> 0250  00904608 23000610            y: SETE_DX10          R24.y,  R[8+AR].y, KC0[2].y
> 0252  81104a08 43000610            z: SETE_DX10          R24.z,  R[8+AR].z, KC0[2].z
> 0254  811fc4fe 23001810    28      y: AND_INT            R24.y,  PV.y, PV.z
> 0256  809fc018 02e01810    29      x: AND_INT            R23.x,  R24.x, PV.y
> 0258  801f00fe 0340229c    30 MP   x: PRED_SETNE_INT     R26.x,  PV.x, 0
>0006  00000005 82800000 JUMP @10
>0008  00000082 a00c0000 ALU 4 @260
> 0260  00000cf8 02e00c90    31      x: MOV                R23.x,  0
> 0262  000000f9 22e00c90            y: MOV                R23.y,  1.0
> 0264  00000cf8 42e00c90            z: MOV                R23.z,  0
> 0266  800000f9 62e00c90            w: MOV                R23.w,  1.0
>0010  00000007 83400001 ELSE @14 POP:1
>0012  00000086 a80c0000 ALU_POP_AFTER 4 @268
> 0268  000000f9 02e00c90    32      x: MOV                R23.x,  1.0
> 0270  00000cf8 22e00c90            y: MOV                R23.y,  0
> 0272  00000cf8 42e00c90            z: MOV                R23.z,  0
> 0274  800000f9 62e00c90            w: MOV                R23.w,  1.0
>0014  0000008a a01c0000 ALU 8 @276
> 0276  00000017 00600c90    33      x: MOV                R3.x,  R23.x
> 0278  00000417 20600c90            y: MOV                R3.y,  R23.y
> 0280  00000817 40600c90            z: MOV                R3.z,  R23.z
> 0282  80000c17 60600c90            w: MOV                R3.w,  R23.w
> 0284  00000004 00400c90    34      x: MOV                R2.x,  R4.x
> 0286  00000404 20400c90            y: MOV                R2.y,  R4.y
> 0288  00000804 40400c90            z: MOV                R2.z,  R4.z
> 0290  80000c04 60400c90            w: MOV                R2.w,  R4.w
>0016  c001203c 95000688 EXPORT_DONE        POS   60    R2.xyzw
>0018  c001c000 95000688 EXPORT_DONE        PARAM 0     R3.xyzw
>0020  00000000 88000000 CF_END @0
>===== SHADER_END ===============================================================
>
>add_gpr_array: @5.x [3]
>add_gpr_array: @5.y [3]
>add_gpr_array: @5.z [3]
>add_gpr_array: @5.w [3]
>add_gpr_array: @8.x [9]
>add_gpr_array: @8.y [9]
>add_gpr_array: @8.z [9]
>add_gpr_array: @8.w [9]
>add_gpr_array: @17.x [3]
>add_gpr_array: @17.y [3]
>add_gpr_array: @17.z [3]
>add_gpr_array: @17.w [3]
>add_gpr_array: @20.x [3]
>add_gpr_array: @20.y [3]
>add_gpr_array: @20.z [3]
>add_gpr_array: @20.w [3]
>
>
>###### after parse
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.xF@R1.x, R1.yF@R1.y, R1.zF@R1.z, R1.wF@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    ALU     
>    <  
>        [  
>            MUL     R4.x,    C3.x, R1.xF@R1.x
>            MUL     R4.y,    C3.y, R1.xF@R1.x
>            MUL     R4.z,    C3.z, R1.xF@R1.x
>            MUL     R4.w,    C3.w, R1.xF@R1.x
>        ]  
>        [  
>            MULADD     R4.x,    C4.x, R1.yF@R1.y, R4.x
>            MULADD     R4.y,    C4.y, R1.yF@R1.y, R4.y
>            MULADD     R4.z,    C4.z, R1.yF@R1.y, R4.z
>            MULADD     R4.w,    C4.w, R1.yF@R1.y, R4.w
>        ]  
>        [  
>            MULADD     R4.x,    C5.x, R1.zF@R1.z, R4.x
>            MULADD     R4.y,    C5.y, R1.zF@R1.z, R4.y
>            MULADD     R4.z,    C5.z, R1.zF@R1.z, R4.z
>            MULADD     R4.w,    C5.w, R1.zF@R1.z, R4.w
>        ]  
>        [  
>            MULADD     R4.x,    C6.x, R1.wF@R1.w, R4.x
>            MULADD     R4.y,    C6.y, R1.wF@R1.w, R4.y
>            MULADD     R4.z,    C6.z, R1.wF@R1.w, R4.z
>            MULADD     R4.w,    C6.w, R1.wF@R1.w, R4.w
>        ]  
>        [  
>            MOV     R5.x,    1|3f800000
>            MOV     R5.y,    2|40000000
>            MOV     R5.z,    3|40400000
>            MOV     R5.w,    1|3f800000
>        ]  
>        [  
>            MOV     R6.x,    4|40800000
>            MOV     R6.y,    5|40a00000
>            MOV     R6.z,    6|40c00000
>            MOV     R6.w,    4|40800000
>        ]  
>        [  
>            MOV     R7.x,    7|40e00000
>            MOV     R7.y,    8|41000000
>            MOV     R7.z,    9|41100000
>            MOV     R7.w,    7|40e00000
>        ]  
>        [  
>            MOV     R8.x,    R5.x
>            MOV     R8.y,    R5.y
>            MOV     R8.z,    R5.z
>            MOV     R8.w,    R5.z
>        ]  
>        [  
>            MOV     R9.x,    R6.x
>            MOV     R9.y,    R6.y
>            MOV     R9.z,    R6.z
>            MOV     R9.w,    R6.z
>        ]  
>        [  
>            MOV     R10.x,    R7.x
>            MOV     R10.y,    R7.y
>            MOV     R10.z,    R7.z
>            MOV     R10.w,    R7.z
>        ]  
>        [  
>            MOV     R17.x,    10|41200000
>            MOV     R17.y,    11|41300000
>            MOV     R17.z,    12|41400000
>            MOV     R17.w,    10|41200000
>        ]  
>        [  
>            MOV     R18.x,    13|41500000
>            MOV     R18.y,    14|41600000
>            MOV     R18.z,    15|41700000
>            MOV     R18.w,    13|41500000
>        ]  
>        [  
>            MOV     R19.x,    16|41800000
>            MOV     R19.y,    17|41880000
>            MOV     R19.z,    18|41900000
>            MOV     R19.w,    16|41800000
>        ]  
>        [  
>            MOV     R11.x,    R17.x
>            MOV     R11.y,    R17.y
>            MOV     R11.z,    R17.z
>            MOV     R11.w,    R17.z
>        ]  
>        [  
>            MOV     R12.x,    R18.x
>            MOV     R12.y,    R18.y
>            MOV     R12.z,    R18.z
>            MOV     R12.w,    R18.z
>        ]  
>        [  
>            MOV     R13.x,    R19.x
>            MOV     R13.y,    R19.y
>            MOV     R13.z,    R19.z
>            MOV     R13.w,    R19.z
>        ]  
>        [  
>            MOV     R20.x,    19|41980000
>            MOV     R20.y,    20|41a00000
>            MOV     R20.z,    21|41a80000
>            MOV     R20.w,    19|41980000
>        ]  
>        [  
>            MOV     R21.x,    22|41b00000
>            MOV     R21.y,    23|41b80000
>            MOV     R21.z,    24|41c00000
>            MOV     R21.w,    22|41b00000
>        ]  
>        [  
>            MOV     R22.x,    25|41c80000
>            MOV     R22.y,    26|41d00000
>            MOV     R22.z,    27|41d80000
>            MOV     R22.w,    25|41c80000
>        ]  
>        [  
>            MOV     R14.x,    R20.x
>            MOV     R14.y,    R20.y
>            MOV     R14.z,    R20.z
>            MOV     R14.w,    R20.z
>        ]  
>        [  
>            MOV     R15.x,    R21.x
>            MOV     R15.y,    R21.y
>            MOV     R15.z,    R21.z
>            MOV     R15.w,    R21.z
>        ]  
>        [  
>            MOV     R16.x,    R22.x
>            MOV     R16.y,    R22.y
>            MOV     R16.z,    R22.z
>            MOV     R16.w,    R22.z
>        ]  
>        [  
>            MULLO_INT       
>                4S  MULLO_INT     R23.x,    C0.x, 4.2039e-45|00000003
>                4S  MULLO_INT     __,    C0.x, 4.2039e-45|00000003
>                4S  MULLO_INT     __,    C0.x, 4.2039e-45|00000003
>                4S  MULLO_INT     __,    C0.x, 4.2039e-45|00000003
>        ]  
>        [  
>            ADD_INT     R23.x,    C1.x, R23.x
>        ]  
>        [  
>            MOV     R25.x,    R23.x
>        ]  
>    >  
>    ALU_PUSH_BEFORE     
>    <  
>        [  
>            MOVA_INT     AR,    R25.x
>        ]  
>        [  
>            SETE_DX10     R24.x,    A8.x[AR]_135, C2.x
>					    rels: A8.x[AR]_135 :  <= R8.x, R9.x, R10.x, R11.x, R12.x, R13.x, R14.x, R15.x, R16.x
>            SETE_DX10     R24.y,    A8.y[AR]_138, C2.y
>					    rels: A8.y[AR]_138 :  <= R8.y, R9.y, R10.y, R11.y, R12.y, R13.y, R14.y, R15.y, R16.y
>            SETE_DX10     R24.z,    A8.z[AR]_141, C2.z
>					    rels: A8.z[AR]_141 :  <= R8.z, R9.z, R10.z, R11.z, R12.z, R13.z, R14.z, R15.z, R16.z
>        ]  
>        [  
>            AND_INT     R24.y,    R24.y, R24.z
>        ]  
>        [  
>            AND_INT     R23.x,    R24.x, R24.y
>        ]  
>        [  
>            PRED_SETNE_INT     R26.x, PR, EM,    R23.x, 0|00000000
>        ]  
>    >  
>    region #0   
>        depart region #0 after {    
>            if EM       
>            {
>                depart region #0 after {    
>                    JUMP      @10
>                    ALU     
>                    <  
>                        [  
>                            MOV     R23.x,    0|00000000
>                            MOV     R23.y,    1|3f800000
>                            MOV     R23.z,    0|00000000
>                            MOV     R23.w,    1|3f800000
>                        ]  
>                    >  
>                } end_depart   
>            } endif   
>            ELSE      @14
>            ALU_POP_AFTER     
>            <  
>                [  
>                    MOV     R23.x,    1|3f800000
>                    MOV     R23.y,    0|00000000
>                    MOV     R23.z,    0|00000000
>                    MOV     R23.w,    1|3f800000
>                ]  
>            >  
>        } end_depart   
>    
>    ALU     
>    <  
>        [  
>            MOV     R3.x,    R23.x
>            MOV     R3.y,    R23.y
>            MOV     R3.z,    R23.z
>            MOV     R3.w,    R23.w
>        ]  
>        [  
>            MOV     R2.x,    R4.x
>            MOV     R2.y,    R4.y
>            MOV     R2.z,    R4.z
>            MOV     R2.w,    R4.w
>        ]  
>    >  
>    EXPORT  POS   60     R2.x, R2.y, R2.z, R2.w
>    EXPORT  PARAM 0     R3.x, R3.y, R3.z, R3.w
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    ALU     
>    <  
>        [  
>            MUL     R4.x.1,    C3.x, R1.x.1
>            MUL     R4.y.1,    C3.y, R1.x.1
>            MUL     R4.z.1,    C3.z, R1.x.1
>            MUL     R4.w.1,    C3.w, R1.x.1
>        ]  
>        [  
>            MULADD     R4.x.2,    C4.x, R1.y.1, R4.x.1
>            MULADD     R4.y.2,    C4.y, R1.y.1, R4.y.1
>            MULADD     R4.z.2,    C4.z, R1.y.1, R4.z.1
>            MULADD     R4.w.2,    C4.w, R1.y.1, R4.w.1
>        ]  
>        [  
>            MULADD     R4.x.3,    C5.x, R1.z.1, R4.x.2
>            MULADD     R4.y.3,    C5.y, R1.z.1, R4.y.2
>            MULADD     R4.z.3,    C5.z, R1.z.1, R4.z.2
>            MULADD     R4.w.3,    C5.w, R1.z.1, R4.w.2
>        ]  
>        [  
>            MULADD     R4.x.4,    C6.x, R1.w.1, R4.x.3
>            MULADD     R4.y.4,    C6.y, R1.w.1, R4.y.3
>            MULADD     R4.z.4,    C6.z, R1.w.1, R4.z.3
>            MULADD     R4.w.4,    C6.w, R1.w.1, R4.w.3
>        ]  
>        [  
>            MOV     R5.x.1,    1|3f800000
>            MOV     R5.y.1,    2|40000000
>            MOV     R5.z.1,    3|40400000
>            MOV     R5.w.1,    1|3f800000
>        ]  
>        [  
>            MOV     R6.x.1,    4|40800000
>            MOV     R6.y.1,    5|40a00000
>            MOV     R6.z.1,    6|40c00000
>            MOV     R6.w.1,    4|40800000
>        ]  
>        [  
>            MOV     R7.x.1,    7|40e00000
>            MOV     R7.y.1,    8|41000000
>            MOV     R7.z.1,    9|41100000
>            MOV     R7.w.1,    7|40e00000
>        ]  
>        [  
>            MOV     R8.x.1,    R5.x.1
>            MOV     R8.y.1,    R5.y.1
>            MOV     R8.z.1,    R5.z.1
>            MOV     R8.w.1,    R5.z.1
>        ]  
>        [  
>            MOV     R9.x.1,    R6.x.1
>            MOV     R9.y.1,    R6.y.1
>            MOV     R9.z.1,    R6.z.1
>            MOV     R9.w.1,    R6.z.1
>        ]  
>        [  
>            MOV     R10.x.1,    R7.x.1
>            MOV     R10.y.1,    R7.y.1
>            MOV     R10.z.1,    R7.z.1
>            MOV     R10.w.1,    R7.z.1
>        ]  
>        [  
>            MOV     R17.x.1,    10|41200000
>            MOV     R17.y.1,    11|41300000
>            MOV     R17.z.1,    12|41400000
>            MOV     R17.w.1,    10|41200000
>        ]  
>        [  
>            MOV     R18.x.1,    13|41500000
>            MOV     R18.y.1,    14|41600000
>            MOV     R18.z.1,    15|41700000
>            MOV     R18.w.1,    13|41500000
>        ]  
>        [  
>            MOV     R19.x.1,    16|41800000
>            MOV     R19.y.1,    17|41880000
>            MOV     R19.z.1,    18|41900000
>            MOV     R19.w.1,    16|41800000
>        ]  
>        [  
>            MOV     R11.x.1,    R17.x.1
>            MOV     R11.y.1,    R17.y.1
>            MOV     R11.z.1,    R17.z.1
>            MOV     R11.w.1,    R17.z.1
>        ]  
>        [  
>            MOV     R12.x.1,    R18.x.1
>            MOV     R12.y.1,    R18.y.1
>            MOV     R12.z.1,    R18.z.1
>            MOV     R12.w.1,    R18.z.1
>        ]  
>        [  
>            MOV     R13.x.1,    R19.x.1
>            MOV     R13.y.1,    R19.y.1
>            MOV     R13.z.1,    R19.z.1
>            MOV     R13.w.1,    R19.z.1
>        ]  
>        [  
>            MOV     R20.x.1,    19|41980000
>            MOV     R20.y.1,    20|41a00000
>            MOV     R20.z.1,    21|41a80000
>            MOV     R20.w.1,    19|41980000
>        ]  
>        [  
>            MOV     R21.x.1,    22|41b00000
>            MOV     R21.y.1,    23|41b80000
>            MOV     R21.z.1,    24|41c00000
>            MOV     R21.w.1,    22|41b00000
>        ]  
>        [  
>            MOV     R22.x.1,    25|41c80000
>            MOV     R22.y.1,    26|41d00000
>            MOV     R22.z.1,    27|41d80000
>            MOV     R22.w.1,    25|41c80000
>        ]  
>        [  
>            MOV     R14.x.1,    R20.x.1
>            MOV     R14.y.1,    R20.y.1
>            MOV     R14.z.1,    R20.z.1
>            MOV     R14.w.1,    R20.z.1
>        ]  
>        [  
>            MOV     R15.x.1,    R21.x.1
>            MOV     R15.y.1,    R21.y.1
>            MOV     R15.z.1,    R21.z.1
>            MOV     R15.w.1,    R21.z.1
>        ]  
>        [  
>            MOV     R16.x.1,    R22.x.1
>            MOV     R16.y.1,    R22.y.1
>            MOV     R16.z.1,    R22.z.1
>            MOV     R16.w.1,    R22.z.1
>        ]  
>        [  
>            MULLO_INT     R23.x.1, __, __, __,       C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003  
>        ]  
>        [  
>            ADD_INT     R23.x.2,    C1.x, R23.x.1
>        ]  
>        [  
>            MOV     R25.x.1,    R23.x.2
>        ]  
>    >  
>    ALU_PUSH_BEFORE     
>    <  
>        [  
>            MOVA_INT     AR.1,    R25.x.1
>        ]  
>        [  
>            SETE_DX10     R24.x.1,    A8.x[AR.1]_135, C2.x
>					    rels: A8.x[AR.1]_135 :  <= R8.x.1, R9.x.1, R10.x.1, R11.x.1, R12.x.1, R13.x.1, R14.x.1, R15.x.1, R16.x.1
>            SETE_DX10     R24.y.1,    A8.y[AR.1]_138, C2.y
>					    rels: A8.y[AR.1]_138 :  <= R8.y.1, R9.y.1, R10.y.1, R11.y.1, R12.y.1, R13.y.1, R14.y.1, R15.y.1, R16.y.1
>            SETE_DX10     R24.z.1,    A8.z[AR.1]_141, C2.z
>					    rels: A8.z[AR.1]_141 :  <= R8.z.1, R9.z.1, R10.z.1, R11.z.1, R12.z.1, R13.z.1, R14.z.1, R15.z.1, R16.z.1
>        ]  
>        [  
>            AND_INT     R24.y.2,    R24.y.1, R24.z.1
>        ]  
>        [  
>            AND_INT     R23.x.3,    R24.x.1, R24.y.2
>        ]  
>        [  
>            PRED_SETNE_INT     R26.x.1, PR.1, EM.1,    R23.x.3, 0|00000000
>        ]  
>    >  
>    region #0   
>        depart region #0 after {    
>            if EM.1       
>            {
>                depart region #0 after {    
>                    JUMP      @10
>                    ALU     
>                    <  
>                        [  
>                            MOV     R23.x.4,    0|00000000
>                            MOV     R23.y.1,    1|3f800000
>                            MOV     R23.z.1,    0|00000000
>                            MOV     R23.w.1,    1|3f800000
>                        ]  
>                    >  
>                } end_depart   
>            } endif   
>            ELSE      @14
>            ALU_POP_AFTER     
>            <  
>                [  
>                    MOV     R23.x.5,    1|3f800000
>                    MOV     R23.y.2,    0|00000000
>                    MOV     R23.z.2,    0|00000000
>                    MOV     R23.w.2,    1|3f800000
>                ]  
>            >  
>        } end_depart   
>    {  
>        * phi     R23.x.6,       R23.x.4, R23.x.5
>        * phi     R23.y.3,       R23.y.1, R23.y.2
>        * phi     R23.z.3,       R23.z.1, R23.z.2
>        * phi     R23.w.3,       R23.w.1, R23.w.2
>    }  
>    
>    ALU     
>    <  
>        [  
>            MOV     R3.x.1,    R23.x.6
>            MOV     R3.y.1,    R23.y.3
>            MOV     R3.z.1,    R23.z.3
>            MOV     R3.w.1,    R23.w.3
>        ]  
>        [  
>            MOV     R2.x.1,    R4.x.4
>            MOV     R2.y.1,    R4.y.4
>            MOV     R2.z.1,    R4.z.4
>            MOV     R2.w.1,    R4.w.4
>        ]  
>    >  
>    EXPORT  POS   60     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>    EXPORT  PARAM 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>    CF_END     
>}  
>
>
>###### after if_conversion
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MUL     R4.x.1,    C3.x, R1.x.1
>    MUL     R4.y.1,    C3.y, R1.x.1
>    MUL     R4.z.1,    C3.z, R1.x.1
>    MUL     R4.w.1,    C3.w, R1.x.1
>    MULADD     R4.x.2,    C4.x, R1.y.1, R4.x.1
>    MULADD     R4.y.2,    C4.y, R1.y.1, R4.y.1
>    MULADD     R4.z.2,    C4.z, R1.y.1, R4.z.1
>    MULADD     R4.w.2,    C4.w, R1.y.1, R4.w.1
>    MULADD     R4.x.3,    C5.x, R1.z.1, R4.x.2
>    MULADD     R4.y.3,    C5.y, R1.z.1, R4.y.2
>    MULADD     R4.z.3,    C5.z, R1.z.1, R4.z.2
>    MULADD     R4.w.3,    C5.w, R1.z.1, R4.w.2
>    MULADD     R4.x.4,    C6.x, R1.w.1, R4.x.3
>    MULADD     R4.y.4,    C6.y, R1.w.1, R4.y.3
>    MULADD     R4.z.4,    C6.z, R1.w.1, R4.z.3
>    MULADD     R4.w.4,    C6.w, R1.w.1, R4.w.3
>    MOV     R5.x.1,    1|3f800000
>    MOV     R5.y.1,    2|40000000
>    MOV     R5.z.1,    3|40400000
>    MOV     R6.x.1,    4|40800000
>    MOV     R6.y.1,    5|40a00000
>    MOV     R6.z.1,    6|40c00000
>    MOV     R7.x.1,    7|40e00000
>    MOV     R7.y.1,    8|41000000
>    MOV     R7.z.1,    9|41100000
>    MOV     R8.x.1,    R5.x.1
>    MOV     R8.y.1,    R5.y.1
>    MOV     R8.z.1,    R5.z.1
>    MOV     R9.x.1,    R6.x.1
>    MOV     R9.y.1,    R6.y.1
>    MOV     R9.z.1,    R6.z.1
>    MOV     R10.x.1,    R7.x.1
>    MOV     R10.y.1,    R7.y.1
>    MOV     R10.z.1,    R7.z.1
>    MOV     R17.x.1,    10|41200000
>    MOV     R17.y.1,    11|41300000
>    MOV     R17.z.1,    12|41400000
>    MOV     R18.x.1,    13|41500000
>    MOV     R18.y.1,    14|41600000
>    MOV     R18.z.1,    15|41700000
>    MOV     R19.x.1,    16|41800000
>    MOV     R19.y.1,    17|41880000
>    MOV     R19.z.1,    18|41900000
>    MOV     R11.x.1,    R17.x.1
>    MOV     R11.y.1,    R17.y.1
>    MOV     R11.z.1,    R17.z.1
>    MOV     R12.x.1,    R18.x.1
>    MOV     R12.y.1,    R18.y.1
>    MOV     R12.z.1,    R18.z.1
>    MOV     R13.x.1,    R19.x.1
>    MOV     R13.y.1,    R19.y.1
>    MOV     R13.z.1,    R19.z.1
>    MOV     R20.x.1,    19|41980000
>    MOV     R20.y.1,    20|41a00000
>    MOV     R20.z.1,    21|41a80000
>    MOV     R21.x.1,    22|41b00000
>    MOV     R21.y.1,    23|41b80000
>    MOV     R21.z.1,    24|41c00000
>    MOV     R22.x.1,    25|41c80000
>    MOV     R22.y.1,    26|41d00000
>    MOV     R22.z.1,    27|41d80000
>    MOV     R14.x.1,    R20.x.1
>    MOV     R14.y.1,    R20.y.1
>    MOV     R14.z.1,    R20.z.1
>    MOV     R15.x.1,    R21.x.1
>    MOV     R15.y.1,    R21.y.1
>    MOV     R15.z.1,    R21.z.1
>    MOV     R16.x.1,    R22.x.1
>    MOV     R16.y.1,    R22.y.1
>    MOV     R16.z.1,    R22.z.1
>    MULLO_INT     R23.x.1, __, __, __,       C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003  
>    ADD_INT     R23.x.2,    C1.x, R23.x.1
>    MOV     R25.x.1,    R23.x.2
>    MOVA_INT     AR.1,    R25.x.1
>    SETE_DX10     R24.x.1,    A8.x[AR.1]_135, C2.x
>					    rels: A8.x[AR.1]_135 :  <= R8.x.1, R9.x.1, R10.x.1, R11.x.1, R12.x.1, R13.x.1, R14.x.1, R15.x.1, R16.x.1
>    SETE_DX10     R24.y.1,    A8.y[AR.1]_138, C2.y
>					    rels: A8.y[AR.1]_138 :  <= R8.y.1, R9.y.1, R10.y.1, R11.y.1, R12.y.1, R13.y.1, R14.y.1, R15.y.1, R16.y.1
>    SETE_DX10     R24.z.1,    A8.z[AR.1]_141, C2.z
>					    rels: A8.z[AR.1]_141 :  <= R8.z.1, R9.z.1, R10.z.1, R11.z.1, R12.z.1, R13.z.1, R14.z.1, R15.z.1, R16.z.1
>    AND_INT     R24.y.2,    R24.y.1, R24.z.1
>    AND_INT     R23.x.3,    R24.x.1, R24.y.2
>    PRED_SETNE_INT     __, __, EM.1,    R23.x.3, 0|00000000
>    SETE_INT     t1,    R23.x.3, 0|00000000
>    MOV     R23.x.4,    0|00000000
>    MOV     R23.y.1,    1|3f800000
>    MOV     R23.z.1,    0|00000000
>    MOV     R23.w.1,    1|3f800000
>    MOV     R23.x.5,    1|3f800000
>    MOV     R23.y.2,    0|00000000
>    MOV     R23.z.2,    0|00000000
>    MOV     R23.w.2,    1|3f800000
>    CNDE_INT     R23.w.3,    t1, R23.w.1, R23.w.2
>    CNDE_INT     R23.z.3,    t1, R23.z.1, R23.z.2
>    CNDE_INT     R23.y.3,    t1, R23.y.1, R23.y.2
>    CNDE_INT     R23.x.6,    t1, R23.x.4, R23.x.5
>    MOV     R3.x.1,    R23.x.6
>    MOV     R3.y.1,    R23.y.3
>    MOV     R3.z.1,    R23.z.3
>    MOV     R3.w.1,    R23.w.3
>    MOV     R2.x.1,    R4.x.4
>    MOV     R2.y.1,    R4.y.4
>    MOV     R2.z.1,    R4.z.4
>    MOV     R2.w.1,    R4.w.4
>    EXPORT  POS   60     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>    EXPORT  PARAM 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>}  
>
>
>###### after peephole
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MUL     R4.x.1,    C3.x, R1.x.1
>    MUL     R4.y.1,    C3.y, R1.x.1
>    MUL     R4.z.1,    C3.z, R1.x.1
>    MUL     R4.w.1,    C3.w, R1.x.1
>    MULADD     R4.x.2,    C4.x, R1.y.1, R4.x.1
>    MULADD     R4.y.2,    C4.y, R1.y.1, R4.y.1
>    MULADD     R4.z.2,    C4.z, R1.y.1, R4.z.1
>    MULADD     R4.w.2,    C4.w, R1.y.1, R4.w.1
>    MULADD     R4.x.3,    C5.x, R1.z.1, R4.x.2
>    MULADD     R4.y.3,    C5.y, R1.z.1, R4.y.2
>    MULADD     R4.z.3,    C5.z, R1.z.1, R4.z.2
>    MULADD     R4.w.3,    C5.w, R1.z.1, R4.w.2
>    MULADD     R4.x.4,    C6.x, R1.w.1, R4.x.3
>    MULADD     R4.y.4,    C6.y, R1.w.1, R4.y.3
>    MULADD     R4.z.4,    C6.z, R1.w.1, R4.z.3
>    MULADD     R4.w.4,    C6.w, R1.w.1, R4.w.3
>    MOV     R5.x.1,    1|3f800000
>    MOV     R5.y.1,    2|40000000
>    MOV     R5.z.1,    3|40400000
>    MOV     R6.x.1,    4|40800000
>    MOV     R6.y.1,    5|40a00000
>    MOV     R6.z.1,    6|40c00000
>    MOV     R7.x.1,    7|40e00000
>    MOV     R7.y.1,    8|41000000
>    MOV     R7.z.1,    9|41100000
>    MOV     R8.x.1,    R5.x.1
>    MOV     R8.y.1,    R5.y.1
>    MOV     R8.z.1,    R5.z.1
>    MOV     R9.x.1,    R6.x.1
>    MOV     R9.y.1,    R6.y.1
>    MOV     R9.z.1,    R6.z.1
>    MOV     R10.x.1,    R7.x.1
>    MOV     R10.y.1,    R7.y.1
>    MOV     R10.z.1,    R7.z.1
>    MOV     R17.x.1,    10|41200000
>    MOV     R17.y.1,    11|41300000
>    MOV     R17.z.1,    12|41400000
>    MOV     R18.x.1,    13|41500000
>    MOV     R18.y.1,    14|41600000
>    MOV     R18.z.1,    15|41700000
>    MOV     R19.x.1,    16|41800000
>    MOV     R19.y.1,    17|41880000
>    MOV     R19.z.1,    18|41900000
>    MOV     R11.x.1,    R17.x.1
>    MOV     R11.y.1,    R17.y.1
>    MOV     R11.z.1,    R17.z.1
>    MOV     R12.x.1,    R18.x.1
>    MOV     R12.y.1,    R18.y.1
>    MOV     R12.z.1,    R18.z.1
>    MOV     R13.x.1,    R19.x.1
>    MOV     R13.y.1,    R19.y.1
>    MOV     R13.z.1,    R19.z.1
>    MOV     R20.x.1,    19|41980000
>    MOV     R20.y.1,    20|41a00000
>    MOV     R20.z.1,    21|41a80000
>    MOV     R21.x.1,    22|41b00000
>    MOV     R21.y.1,    23|41b80000
>    MOV     R21.z.1,    24|41c00000
>    MOV     R22.x.1,    25|41c80000
>    MOV     R22.y.1,    26|41d00000
>    MOV     R22.z.1,    27|41d80000
>    MOV     R14.x.1,    R20.x.1
>    MOV     R14.y.1,    R20.y.1
>    MOV     R14.z.1,    R20.z.1
>    MOV     R15.x.1,    R21.x.1
>    MOV     R15.y.1,    R21.y.1
>    MOV     R15.z.1,    R21.z.1
>    MOV     R16.x.1,    R22.x.1
>    MOV     R16.y.1,    R22.y.1
>    MOV     R16.z.1,    R22.z.1
>    MULLO_INT     R23.x.1, __, __, __,       C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003  
>    ADD_INT     R23.x.2,    C1.x, R23.x.1
>    MOV     R25.x.1,    R23.x.2
>    MOVA_INT     AR.1,    R25.x.1
>    SETE_DX10     R24.x.1,    A8.x[AR.1]_135, C2.x
>					    rels: A8.x[AR.1]_135 :  <= R8.x.1, R9.x.1, R10.x.1, R11.x.1, R12.x.1, R13.x.1, R14.x.1, R15.x.1, R16.x.1
>    SETE_DX10     R24.y.1,    A8.y[AR.1]_138, C2.y
>					    rels: A8.y[AR.1]_138 :  <= R8.y.1, R9.y.1, R10.y.1, R11.y.1, R12.y.1, R13.y.1, R14.y.1, R15.y.1, R16.y.1
>    SETE_DX10     R24.z.1,    A8.z[AR.1]_141, C2.z
>					    rels: A8.z[AR.1]_141 :  <= R8.z.1, R9.z.1, R10.z.1, R11.z.1, R12.z.1, R13.z.1, R14.z.1, R15.z.1, R16.z.1
>    AND_INT     R24.y.2,    R24.y.1, R24.z.1
>    AND_INT     R23.x.3,    R24.x.1, R24.y.2
>    PRED_SETNE_INT     __, __, EM.1,    R23.x.3, 0|00000000
>    SETE_INT     t1,    R23.x.3, 0|00000000
>    MOV     R23.x.4,    0|00000000
>    MOV     R23.y.1,    1|3f800000
>    MOV     R23.z.1,    0|00000000
>    MOV     R23.w.1,    1|3f800000
>    MOV     R23.x.5,    1|3f800000
>    MOV     R23.y.2,    0|00000000
>    MOV     R23.z.2,    0|00000000
>    MOV     R23.w.2,    1|3f800000
>    CNDE_INT     R23.w.3,    R23.x.3, R23.w.2, R23.w.1
>    CNDE_INT     R23.z.3,    R23.x.3, R23.z.2, R23.z.1
>    CNDE_INT     R23.y.3,    R23.x.3, R23.y.2, R23.y.1
>    CNDE_INT     R23.x.6,    R23.x.3, R23.x.5, R23.x.4
>    MOV     R3.x.1,    R23.x.6
>    MOV     R3.y.1,    R23.y.3
>    MOV     R3.z.1,    R23.z.3
>    MOV     R3.w.1,    R23.w.3
>    MOV     R2.x.1,    R4.x.4
>    MOV     R2.y.1,    R4.y.4
>    MOV     R2.z.1,    R4.z.4
>    MOV     R2.w.1,    R4.w.4
>    EXPORT  POS   60     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>    EXPORT  PARAM 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>}  
>
>
>###### after gvn
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MUL     R4.x.1,    C3.x, R1.x.1
>    MUL     R4.y.1,    C3.y, R1.x.1
>    MUL     R4.z.1,    C3.z, R1.x.1
>    MUL     R4.w.1,    C3.w, R1.x.1
>    MULADD     R4.x.2,    C4.x, R1.y.1, R4.x.1
>    MULADD     R4.y.2,    C4.y, R1.y.1, R4.y.1
>    MULADD     R4.z.2,    C4.z, R1.y.1, R4.z.1
>    MULADD     R4.w.2,    C4.w, R1.y.1, R4.w.1
>    MULADD     R4.x.3,    C5.x, R1.z.1, R4.x.2
>    MULADD     R4.y.3,    C5.y, R1.z.1, R4.y.2
>    MULADD     R4.z.3,    C5.z, R1.z.1, R4.z.2
>    MULADD     R4.w.3,    C5.w, R1.z.1, R4.w.2
>    MULADD     R4.x.4,    C6.x, R1.w.1, R4.x.3
>    MULADD     R4.y.4,    C6.y, R1.w.1, R4.y.3
>    MULADD     R4.z.4,    C6.z, R1.w.1, R4.z.3
>    MULADD     R4.w.4,    C6.w, R1.w.1, R4.w.3
>    MOV     R5.x.1,    1|3f800000
>    MOV     R5.y.1,    2|40000000
>    MOV     R5.z.1,    3|40400000
>    MOV     R6.x.1,    4|40800000
>    MOV     R6.y.1,    5|40a00000
>    MOV     R6.z.1,    6|40c00000
>    MOV     R7.x.1,    7|40e00000
>    MOV     R7.y.1,    8|41000000
>    MOV     R7.z.1,    9|41100000
>    MOV     R8.x.1,    1|3f800000
>    MOV     R8.y.1,    2|40000000
>    MOV     R8.z.1,    3|40400000
>    MOV     R9.x.1,    4|40800000
>    MOV     R9.y.1,    5|40a00000
>    MOV     R9.z.1,    6|40c00000
>    MOV     R10.x.1,    7|40e00000
>    MOV     R10.y.1,    8|41000000
>    MOV     R10.z.1,    9|41100000
>    MOV     R17.x.1,    10|41200000
>    MOV     R17.y.1,    11|41300000
>    MOV     R17.z.1,    12|41400000
>    MOV     R18.x.1,    13|41500000
>    MOV     R18.y.1,    14|41600000
>    MOV     R18.z.1,    15|41700000
>    MOV     R19.x.1,    16|41800000
>    MOV     R19.y.1,    17|41880000
>    MOV     R19.z.1,    18|41900000
>    MOV     R11.x.1,    10|41200000
>    MOV     R11.y.1,    11|41300000
>    MOV     R11.z.1,    12|41400000
>    MOV     R12.x.1,    13|41500000
>    MOV     R12.y.1,    14|41600000
>    MOV     R12.z.1,    15|41700000
>    MOV     R13.x.1,    16|41800000
>    MOV     R13.y.1,    17|41880000
>    MOV     R13.z.1,    18|41900000
>    MOV     R20.x.1,    19|41980000
>    MOV     R20.y.1,    20|41a00000
>    MOV     R20.z.1,    21|41a80000
>    MOV     R21.x.1,    22|41b00000
>    MOV     R21.y.1,    23|41b80000
>    MOV     R21.z.1,    24|41c00000
>    MOV     R22.x.1,    25|41c80000
>    MOV     R22.y.1,    26|41d00000
>    MOV     R22.z.1,    27|41d80000
>    MOV     R14.x.1,    19|41980000
>    MOV     R14.y.1,    20|41a00000
>    MOV     R14.z.1,    21|41a80000
>    MOV     R15.x.1,    22|41b00000
>    MOV     R15.y.1,    23|41b80000
>    MOV     R15.z.1,    24|41c00000
>    MOV     R16.x.1,    25|41c80000
>    MOV     R16.y.1,    26|41d00000
>    MOV     R16.z.1,    27|41d80000
>    MULLO_INT     R23.x.1, __, __, __,       C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003  
>    ADD_INT     R23.x.2,    C1.x, R23.x.1
>    MOV     R25.x.1,    R23.x.2
>    MOVA_INT     AR.1,    R23.x.2
>    SETE_DX10     R24.x.1,    A8.x[R23.x.2]_135, C2.x
>					    rels: A8.x[R23.x.2]_135 :  <= R8.x.1, R9.x.1, R10.x.1, R11.x.1, R12.x.1, R13.x.1, R14.x.1, R15.x.1, R16.x.1
>    SETE_DX10     R24.y.1,    A8.y[R23.x.2]_138, C2.y
>					    rels: A8.y[R23.x.2]_138 :  <= R8.y.1, R9.y.1, R10.y.1, R11.y.1, R12.y.1, R13.y.1, R14.y.1, R15.y.1, R16.y.1
>    SETE_DX10     R24.z.1,    A8.z[R23.x.2]_141, C2.z
>					    rels: A8.z[R23.x.2]_141 :  <= R8.z.1, R9.z.1, R10.z.1, R11.z.1, R12.z.1, R13.z.1, R14.z.1, R15.z.1, R16.z.1
>    AND_INT     R24.y.2,    R24.y.1, R24.z.1
>    AND_INT     R23.x.3,    R24.x.1, R24.y.2
>    PRED_SETNE_INT     __, __, EM.1,    R23.x.3, 0|00000000
>    SETE_INT     t1,    R23.x.3, 0|00000000
>    MOV     R23.x.4,    0|00000000
>    MOV     R23.y.1,    1|3f800000
>    MOV     R23.z.1,    0|00000000
>    MOV     R23.w.1,    1|3f800000
>    MOV     R23.x.5,    1|3f800000
>    MOV     R23.y.2,    0|00000000
>    MOV     R23.z.2,    0|00000000
>    MOV     R23.w.2,    1|3f800000
>    MOV     R23.w.3,    1|3f800000
>    MOV     R23.z.3,    0|00000000
>    CNDE_INT     R23.y.3,    R23.x.3, 0|00000000, 1|3f800000
>    CNDE_INT     R23.x.6,    R23.x.3, 1|3f800000, 0|00000000
>    MOV     R3.x.1,    R23.x.6
>    MOV     R3.y.1,    R23.y.3
>    MOV     R3.z.1,    0|00000000
>    MOV     R3.w.1,    1|3f800000
>    MOV     R2.x.1,    R4.x.4
>    MOV     R2.y.1,    R4.y.4
>    MOV     R2.z.1,    R4.z.4
>    MOV     R2.w.1,    R4.w.4
>    EXPORT  POS   60     R4.x.4, R4.y.4, R4.z.4, R4.w.4
>    EXPORT  PARAM 0     R23.x.6, R23.y.3, 0|00000000, 1|3f800000
>}  
>
>
>###### after def_use
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MUL     R4.x.1,    C3.x, R1.x.1
>    MUL     R4.y.1,    C3.y, R1.x.1
>    MUL     R4.z.1,    C3.z, R1.x.1
>    MUL     R4.w.1,    C3.w, R1.x.1
>    MULADD     R4.x.2,    C4.x, R1.y.1, R4.x.1
>    MULADD     R4.y.2,    C4.y, R1.y.1, R4.y.1
>    MULADD     R4.z.2,    C4.z, R1.y.1, R4.z.1
>    MULADD     R4.w.2,    C4.w, R1.y.1, R4.w.1
>    MULADD     R4.x.3,    C5.x, R1.z.1, R4.x.2
>    MULADD     R4.y.3,    C5.y, R1.z.1, R4.y.2
>    MULADD     R4.z.3,    C5.z, R1.z.1, R4.z.2
>    MULADD     R4.w.3,    C5.w, R1.z.1, R4.w.2
>    MULADD     R4.x.4,    C6.x, R1.w.1, R4.x.3
>    MULADD     R4.y.4,    C6.y, R1.w.1, R4.y.3
>    MULADD     R4.z.4,    C6.z, R1.w.1, R4.z.3
>    MULADD     R4.w.4,    C6.w, R1.w.1, R4.w.3
>    MOV     R5.x.1,    1|3f800000
>    MOV     R5.y.1,    2|40000000
>    MOV     R5.z.1,    3|40400000
>    MOV     R6.x.1,    4|40800000
>    MOV     R6.y.1,    5|40a00000
>    MOV     R6.z.1,    6|40c00000
>    MOV     R7.x.1,    7|40e00000
>    MOV     R7.y.1,    8|41000000
>    MOV     R7.z.1,    9|41100000
>    MOV     R8.x.1,    1|3f800000
>    MOV     R8.y.1,    2|40000000
>    MOV     R8.z.1,    3|40400000
>    MOV     R9.x.1,    4|40800000
>    MOV     R9.y.1,    5|40a00000
>    MOV     R9.z.1,    6|40c00000
>    MOV     R10.x.1,    7|40e00000
>    MOV     R10.y.1,    8|41000000
>    MOV     R10.z.1,    9|41100000
>    MOV     R17.x.1,    10|41200000
>    MOV     R17.y.1,    11|41300000
>    MOV     R17.z.1,    12|41400000
>    MOV     R18.x.1,    13|41500000
>    MOV     R18.y.1,    14|41600000
>    MOV     R18.z.1,    15|41700000
>    MOV     R19.x.1,    16|41800000
>    MOV     R19.y.1,    17|41880000
>    MOV     R19.z.1,    18|41900000
>    MOV     R11.x.1,    10|41200000
>    MOV     R11.y.1,    11|41300000
>    MOV     R11.z.1,    12|41400000
>    MOV     R12.x.1,    13|41500000
>    MOV     R12.y.1,    14|41600000
>    MOV     R12.z.1,    15|41700000
>    MOV     R13.x.1,    16|41800000
>    MOV     R13.y.1,    17|41880000
>    MOV     R13.z.1,    18|41900000
>    MOV     R20.x.1,    19|41980000
>    MOV     R20.y.1,    20|41a00000
>    MOV     R20.z.1,    21|41a80000
>    MOV     R21.x.1,    22|41b00000
>    MOV     R21.y.1,    23|41b80000
>    MOV     R21.z.1,    24|41c00000
>    MOV     R22.x.1,    25|41c80000
>    MOV     R22.y.1,    26|41d00000
>    MOV     R22.z.1,    27|41d80000
>    MOV     R14.x.1,    19|41980000
>    MOV     R14.y.1,    20|41a00000
>    MOV     R14.z.1,    21|41a80000
>    MOV     R15.x.1,    22|41b00000
>    MOV     R15.y.1,    23|41b80000
>    MOV     R15.z.1,    24|41c00000
>    MOV     R16.x.1,    25|41c80000
>    MOV     R16.y.1,    26|41d00000
>    MOV     R16.z.1,    27|41d80000
>    MULLO_INT     R23.x.1, __, __, __,       C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003  
>    ADD_INT     R23.x.2,    C1.x, R23.x.1
>    MOV     R25.x.1,    R23.x.2
>    MOVA_INT     AR.1,    R23.x.2
>    SETE_DX10     R24.x.1,    A8.x[R23.x.2]_135, C2.x
>					    rels: A8.x[R23.x.2]_135 :  <= R8.x.1, R9.x.1, R10.x.1, R11.x.1, R12.x.1, R13.x.1, R14.x.1, R15.x.1, R16.x.1
>    SETE_DX10     R24.y.1,    A8.y[R23.x.2]_138, C2.y
>					    rels: A8.y[R23.x.2]_138 :  <= R8.y.1, R9.y.1, R10.y.1, R11.y.1, R12.y.1, R13.y.1, R14.y.1, R15.y.1, R16.y.1
>    SETE_DX10     R24.z.1,    A8.z[R23.x.2]_141, C2.z
>					    rels: A8.z[R23.x.2]_141 :  <= R8.z.1, R9.z.1, R10.z.1, R11.z.1, R12.z.1, R13.z.1, R14.z.1, R15.z.1, R16.z.1
>    AND_INT     R24.y.2,    R24.y.1, R24.z.1
>    AND_INT     R23.x.3,    R24.x.1, R24.y.2
>    PRED_SETNE_INT     __, __, EM.1,    R23.x.3, 0|00000000
>    SETE_INT     t1,    R23.x.3, 0|00000000
>    MOV     R23.x.4,    0|00000000
>    MOV     R23.y.1,    1|3f800000
>    MOV     R23.z.1,    0|00000000
>    MOV     R23.w.1,    1|3f800000
>    MOV     R23.x.5,    1|3f800000
>    MOV     R23.y.2,    0|00000000
>    MOV     R23.z.2,    0|00000000
>    MOV     R23.w.2,    1|3f800000
>    MOV     R23.w.3,    1|3f800000
>    MOV     R23.z.3,    0|00000000
>    CNDE_INT     R23.y.3,    R23.x.3, 0|00000000, 1|3f800000
>    CNDE_INT     R23.x.6,    R23.x.3, 1|3f800000, 0|00000000
>    MOV     R3.x.1,    R23.x.6
>    MOV     R3.y.1,    R23.y.3
>    MOV     R3.z.1,    0|00000000
>    MOV     R3.w.1,    1|3f800000
>    MOV     R2.x.1,    R4.x.4
>    MOV     R2.y.1,    R4.y.4
>    MOV     R2.z.1,    R4.z.4
>    MOV     R2.w.1,    R4.w.4
>    EXPORT  POS   60     R4.x.4, R4.y.4, R4.z.4, R4.w.4
>    EXPORT  PARAM 0     R23.x.6, R23.y.3, 0|00000000, 1|3f800000
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    CALL_FS     R1.x.1, R1.y.1, R1.z.1, R1.w.1,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>    MUL     R4.x.1,    C3.x, R1.x.1
>    MUL     R4.y.1,    C3.y, R1.x.1
>    MUL     R4.z.1,    C3.z, R1.x.1
>    MUL     R4.w.1,    C3.w, R1.x.1
>    MULADD     R4.x.2,    C4.x, R1.y.1, R4.x.1
>    MULADD     R4.y.2,    C4.y, R1.y.1, R4.y.1
>    MULADD     R4.z.2,    C4.z, R1.y.1, R4.z.1
>    MULADD     R4.w.2,    C4.w, R1.y.1, R4.w.1
>    MULADD     R4.x.3,    C5.x, R1.z.1, R4.x.2
>    MULADD     R4.y.3,    C5.y, R1.z.1, R4.y.2
>    MULADD     R4.z.3,    C5.z, R1.z.1, R4.z.2
>    MULADD     R4.w.3,    C5.w, R1.z.1, R4.w.2
>    MULADD     R4.x.4,    C6.x, R1.w.1, R4.x.3
>    MULADD     R4.y.4,    C6.y, R1.w.1, R4.y.3
>    MULADD     R4.z.4,    C6.z, R1.w.1, R4.z.3
>    MULADD     R4.w.4,    C6.w, R1.w.1, R4.w.3
>    MOV     R8.x.1,    1|3f800000
>    MOV     R8.y.1,    2|40000000
>    MOV     R8.z.1,    3|40400000
>    MOV     R9.x.1,    4|40800000
>    MOV     R9.y.1,    5|40a00000
>    MOV     R9.z.1,    6|40c00000
>    MOV     R10.x.1,    7|40e00000
>    MOV     R10.y.1,    8|41000000
>    MOV     R10.z.1,    9|41100000
>    MOV     R11.x.1,    10|41200000
>    MOV     R11.y.1,    11|41300000
>    MOV     R11.z.1,    12|41400000
>    MOV     R12.x.1,    13|41500000
>    MOV     R12.y.1,    14|41600000
>    MOV     R12.z.1,    15|41700000
>    MOV     R13.x.1,    16|41800000
>    MOV     R13.y.1,    17|41880000
>    MOV     R13.z.1,    18|41900000
>    MOV     R14.x.1,    19|41980000
>    MOV     R14.y.1,    20|41a00000
>    MOV     R14.z.1,    21|41a80000
>    MOV     R15.x.1,    22|41b00000
>    MOV     R15.y.1,    23|41b80000
>    MOV     R15.z.1,    24|41c00000
>    MOV     R16.x.1,    25|41c80000
>    MOV     R16.y.1,    26|41d00000
>    MOV     R16.z.1,    27|41d80000
>    MULLO_INT     R23.x.1, __, __, __,       C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003  
>    ADD_INT     R23.x.2,    C1.x, R23.x.1
>    SETE_DX10     R24.x.1,    A8.x[R23.x.2]_135, C2.x
>					    rels: A8.x[R23.x.2]_135 :  <= R8.x.1, R9.x.1, R10.x.1, R11.x.1, R12.x.1, R13.x.1, R14.x.1, R15.x.1, R16.x.1
>    SETE_DX10     R24.y.1,    A8.y[R23.x.2]_138, C2.y
>					    rels: A8.y[R23.x.2]_138 :  <= R8.y.1, R9.y.1, R10.y.1, R11.y.1, R12.y.1, R13.y.1, R14.y.1, R15.y.1, R16.y.1
>    SETE_DX10     R24.z.1,    A8.z[R23.x.2]_141, C2.z
>					    rels: A8.z[R23.x.2]_141 :  <= R8.z.1, R9.z.1, R10.z.1, R11.z.1, R12.z.1, R13.z.1, R14.z.1, R15.z.1, R16.z.1
>    AND_INT     R24.y.2,    R24.y.1, R24.z.1
>    AND_INT     R23.x.3,    R24.x.1, R24.y.2
>    CNDE_INT     R23.y.3,    R23.x.3, 0|00000000, 1|3f800000
>    CNDE_INT     R23.x.6,    R23.x.3, 1|3f800000, 0|00000000
>    EXPORT  POS   60     R4.x.4, R4.y.4, R4.z.4, R4.w.4
>    EXPORT  PARAM 0     R23.x.6, R23.y.3, 0|00000000, 1|3f800000
>}  
>
>
>###### after gcm
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  
>        CALL_FS     t8F@R1.x, t9F@R1.y, t10F@R1.z, t11F@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  
>            MULLO_INT     R23.x.1, __, __, __,       C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003  
>            MOV     R8.y.1,    2|40000000
>            MOV     R9.y.1,    5|40a00000
>            MOV     R10.y.1,    8|41000000
>            MOV     R11.y.1,    11|41300000
>            MOV     R12.y.1,    14|41600000
>            MOV     R13.y.1,    17|41880000
>            MOV     R14.y.1,    20|41a00000
>            MOV     R15.y.1,    23|41b80000
>            MOV     R16.y.1,    26|41d00000
>            ADD_INT     R23.x.2,    C1.x, R23.x.1
>            MOV     R8.z.1,    3|40400000
>            MOV     R9.z.1,    6|40c00000
>            MOV     R10.z.1,    9|41100000
>            MOV     R11.z.1,    12|41400000
>            MOV     R12.z.1,    15|41700000
>            MOV     R13.z.1,    18|41900000
>            MOV     R14.z.1,    21|41a80000
>            MOV     R15.z.1,    24|41c00000
>            MOV     R16.z.1,    27|41d80000
>            (copy) MOV     R1.x.1,    t8F@R1.x
>            MUL     R4.w.1,    C3.w, R1.x.1
>            MUL     R4.z.1,    C3.z, R1.x.1
>            MUL     R4.y.1,    C3.y, R1.x.1
>            MUL     R4.x.1,    C3.x, R1.x.1
>            MOV     R8.x.1,    1|3f800000
>            MOV     R9.x.1,    4|40800000
>            MOV     R10.x.1,    7|40e00000
>            MOV     R11.x.1,    10|41200000
>            MOV     R12.x.1,    13|41500000
>            MOV     R13.x.1,    16|41800000
>            MOV     R14.x.1,    19|41980000
>            MOV     R15.x.1,    22|41b00000
>            MOV     R16.x.1,    25|41c80000
>            SETE_DX10     R24.y.1,    A8.y[R23.x.2]_138, C2.y
>					    rels: A8.y[R23.x.2]_138 :  <= R8.y.1, R9.y.1, R10.y.1, R11.y.1, R12.y.1, R13.y.1, R14.y.1, R15.y.1, R16.y.1
>            SETE_DX10     R24.z.1,    A8.z[R23.x.2]_141, C2.z
>					    rels: A8.z[R23.x.2]_141 :  <= R8.z.1, R9.z.1, R10.z.1, R11.z.1, R12.z.1, R13.z.1, R14.z.1, R15.z.1, R16.z.1
>            (copy) MOV     R1.y.1,    t9F@R1.y
>            MULADD     R4.w.2,    C4.w, R1.y.1, R4.w.1
>            MULADD     R4.z.2,    C4.z, R1.y.1, R4.z.1
>            MULADD     R4.y.2,    C4.y, R1.y.1, R4.y.1
>            MULADD     R4.x.2,    C4.x, R1.y.1, R4.x.1
>            SETE_DX10     R24.x.1,    A8.x[R23.x.2]_135, C2.x
>					    rels: A8.x[R23.x.2]_135 :  <= R8.x.1, R9.x.1, R10.x.1, R11.x.1, R12.x.1, R13.x.1, R14.x.1, R15.x.1, R16.x.1
>            AND_INT     R24.y.2,    R24.y.1, R24.z.1
>            (copy) MOV     R1.z.1,    t10F@R1.z
>            MULADD     R4.w.3,    C5.w, R1.z.1, R4.w.2
>            MULADD     R4.z.3,    C5.z, R1.z.1, R4.z.2
>            MULADD     R4.y.3,    C5.y, R1.z.1, R4.y.2
>            MULADD     R4.x.3,    C5.x, R1.z.1, R4.x.2
>            AND_INT     R23.x.3,    R24.x.1, R24.y.2
>            (copy) MOV     R1.w.1,    t11F@R1.w
>            MULADD     R4.w.4,    C6.w, R1.w.1, R4.w.3
>            MULADD     R4.z.4,    C6.z, R1.w.1, R4.z.3
>            MULADD     R4.y.4,    C6.y, R1.w.1, R4.y.3
>            MULADD     R4.x.4,    C6.x, R1.w.1, R4.x.3
>            CNDE_INT     R23.y.3,    R23.x.3, 0|00000000, 1|3f800000
>            CNDE_INT     R23.x.6,    R23.x.3, 1|3f800000, 0|00000000
>            (copy) MOV     t7,    R4.w.4
>            (copy) MOV     t6,    R4.z.4
>            (copy) MOV     t5,    R4.y.4
>            (copy) MOV     t4,    R4.x.4
>            (copy) MOV     t3,    R23.y.3
>            (copy) MOV     t2,    R23.x.6
>        >  
>        EXPORT  PARAM 0     t2, t3, 0|00000000, 1|3f800000
>        EXPORT  POS   60     t4, t5, t6, t7
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t8F@R1.x, t9F@R1.y, t10F@R1.z, t11F@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t8F@R1.x t9F@R1.y t10F@R1.z t11F@R1.w ]
>            MULLO_INT     R23.x.1, __, __, __,       C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003  
>            MOV     R8.y.1,    2|40000000
>            MOV     R9.y.1,    5|40a00000
>            MOV     R10.y.1,    8|41000000
>            MOV     R11.y.1,    11|41300000
>            MOV     R12.y.1,    14|41600000
>            MOV     R13.y.1,    17|41880000
>            MOV     R14.y.1,    20|41a00000
>            MOV     R15.y.1,    23|41b80000
>            MOV     R16.y.1,    26|41d00000
>            ADD_INT     R23.x.2,    C1.x, R23.x.1
>            MOV     R8.z.1,    3|40400000
>            MOV     R9.z.1,    6|40c00000
>            MOV     R10.z.1,    9|41100000
>            MOV     R11.z.1,    12|41400000
>            MOV     R12.z.1,    15|41700000
>            MOV     R13.z.1,    18|41900000
>            MOV     R14.z.1,    21|41a80000
>            MOV     R15.z.1,    24|41c00000
>            MOV     R16.z.1,    27|41d80000
>            (copy) MOV     R1.x.1,    t8F@R1.x
>            MUL     R4.w.1,    C3.w, R1.x.1
>            MUL     R4.z.1,    C3.z, R1.x.1
>            MUL     R4.y.1,    C3.y, R1.x.1
>            MUL     R4.x.1,    C3.x, R1.x.1
>            MOV     R8.x.1,    1|3f800000
>            MOV     R9.x.1,    4|40800000
>            MOV     R10.x.1,    7|40e00000
>            MOV     R11.x.1,    10|41200000
>            MOV     R12.x.1,    13|41500000
>            MOV     R13.x.1,    16|41800000
>            MOV     R14.x.1,    19|41980000
>            MOV     R15.x.1,    22|41b00000
>            MOV     R16.x.1,    25|41c80000
>            SETE_DX10     R24.y.1,    A8.y[R23.x.2]_138, C2.y
>					    rels: A8.y[R23.x.2]_138 :  <= R8.y.1, R9.y.1, R10.y.1, R11.y.1, R12.y.1, R13.y.1, R14.y.1, R15.y.1, R16.y.1
>            SETE_DX10     R24.z.1,    A8.z[R23.x.2]_141, C2.z
>					    rels: A8.z[R23.x.2]_141 :  <= R8.z.1, R9.z.1, R10.z.1, R11.z.1, R12.z.1, R13.z.1, R14.z.1, R15.z.1, R16.z.1
>            (copy) MOV     R1.y.1,    t9F@R1.y
>            MULADD     R4.w.2,    C4.w, R1.y.1, R4.w.1
>            MULADD     R4.z.2,    C4.z, R1.y.1, R4.z.1
>            MULADD     R4.y.2,    C4.y, R1.y.1, R4.y.1
>            MULADD     R4.x.2,    C4.x, R1.y.1, R4.x.1
>            SETE_DX10     R24.x.1,    A8.x[R23.x.2]_135, C2.x
>					    rels: A8.x[R23.x.2]_135 :  <= R8.x.1, R9.x.1, R10.x.1, R11.x.1, R12.x.1, R13.x.1, R14.x.1, R15.x.1, R16.x.1
>            AND_INT     R24.y.2,    R24.y.1, R24.z.1
>            (copy) MOV     R1.z.1,    t10F@R1.z
>            MULADD     R4.w.3,    C5.w, R1.z.1, R4.w.2
>            MULADD     R4.z.3,    C5.z, R1.z.1, R4.z.2
>            MULADD     R4.y.3,    C5.y, R1.z.1, R4.y.2
>            MULADD     R4.x.3,    C5.x, R1.z.1, R4.x.2
>            AND_INT     R23.x.3,    R24.x.1, R24.y.2
>            (copy) MOV     R1.w.1,    t11F@R1.w
>            MULADD     R4.w.4,    C6.w, R1.w.1, R4.w.3
>            MULADD     R4.z.4,    C6.z, R1.w.1, R4.z.3
>            MULADD     R4.y.4,    C6.y, R1.w.1, R4.y.3
>            MULADD     R4.x.4,    C6.x, R1.w.1, R4.x.3
>            CNDE_INT     R23.y.3,    R23.x.3, 0|00000000, 1|3f800000
>            CNDE_INT     R23.x.6,    R23.x.3, 1|3f800000, 0|00000000
>            (copy) MOV     t7,    R4.w.4
>            (copy) MOV     t6,    R4.z.4
>            (copy) MOV     t5,    R4.y.4
>            (copy) MOV     t4,    R4.x.4
>            (copy) MOV     t3,    R23.y.3
>            (copy) MOV     t2,    R23.x.6
>        >  live_after: [t2 t3 t4 t5 t6 t7 ]
>        EXPORT  PARAM 0     t2, t3, 0|00000000, 1|3f800000
>        EXPORT  POS   60     t4, t5, t6, t7
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t8F@R1.x, t9F@R1.y, t10F@R1.z, t11F@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t8F@R1.x t9F@R1.y t10F@R1.z t11F@R1.w ]
>            MULLO_INT     R23.x.1, __, __, __,       C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003  
>            MOV     R8.y.1,    2|40000000
>            MOV     R9.y.1,    5|40a00000
>            MOV     R10.y.1,    8|41000000
>            MOV     R11.y.1,    11|41300000
>            MOV     R12.y.1,    14|41600000
>            MOV     R13.y.1,    17|41880000
>            MOV     R14.y.1,    20|41a00000
>            MOV     R15.y.1,    23|41b80000
>            MOV     R16.y.1,    26|41d00000
>            ADD_INT     R23.x.2,    C1.x, R23.x.1
>            MOV     R8.z.1,    3|40400000
>            MOV     R9.z.1,    6|40c00000
>            MOV     R10.z.1,    9|41100000
>            MOV     R11.z.1,    12|41400000
>            MOV     R12.z.1,    15|41700000
>            MOV     R13.z.1,    18|41900000
>            MOV     R14.z.1,    21|41a80000
>            MOV     R15.z.1,    24|41c00000
>            MOV     R16.z.1,    27|41d80000
>            (copy) MOV     R1.x.1F@R1.x,    t8F@R1.x
>            MUL     R4.w.1,    C3.w, R1.x.1F@R1.x
>            MUL     R4.z.1,    C3.z, R1.x.1F@R1.x
>            MUL     R4.y.1,    C3.y, R1.x.1F@R1.x
>            MUL     R4.x.1,    C3.x, R1.x.1F@R1.x
>            MOV     R8.x.1,    1|3f800000
>            MOV     R9.x.1,    4|40800000
>            MOV     R10.x.1,    7|40e00000
>            MOV     R11.x.1,    10|41200000
>            MOV     R12.x.1,    13|41500000
>            MOV     R13.x.1,    16|41800000
>            MOV     R14.x.1,    19|41980000
>            MOV     R15.x.1,    22|41b00000
>            MOV     R16.x.1,    25|41c80000
>            SETE_DX10     R24.y.1,    A8.y[R23.x.2]_138, C2.y
>					    rels: A8.y[R23.x.2]_138 :  <= R8.y.1, R9.y.1, R10.y.1, R11.y.1, R12.y.1, R13.y.1, R14.y.1, R15.y.1, R16.y.1
>            SETE_DX10     R24.z.1,    A8.z[R23.x.2]_141, C2.z
>					    rels: A8.z[R23.x.2]_141 :  <= R8.z.1, R9.z.1, R10.z.1, R11.z.1, R12.z.1, R13.z.1, R14.z.1, R15.z.1, R16.z.1
>            (copy) MOV     R1.y.1F@R1.y,    t9F@R1.y
>            MULADD     R4.w.2,    C4.w, R1.y.1F@R1.y, R4.w.1
>            MULADD     R4.z.2,    C4.z, R1.y.1F@R1.y, R4.z.1
>            MULADD     R4.y.2,    C4.y, R1.y.1F@R1.y, R4.y.1
>            MULADD     R4.x.2,    C4.x, R1.y.1F@R1.y, R4.x.1
>            SETE_DX10     R24.x.1,    A8.x[R23.x.2]_135, C2.x
>					    rels: A8.x[R23.x.2]_135 :  <= R8.x.1, R9.x.1, R10.x.1, R11.x.1, R12.x.1, R13.x.1, R14.x.1, R15.x.1, R16.x.1
>            AND_INT     R24.y.2,    R24.y.1, R24.z.1
>            (copy) MOV     R1.z.1F@R1.z,    t10F@R1.z
>            MULADD     R4.w.3,    C5.w, R1.z.1F@R1.z, R4.w.2
>            MULADD     R4.z.3,    C5.z, R1.z.1F@R1.z, R4.z.2
>            MULADD     R4.y.3,    C5.y, R1.z.1F@R1.z, R4.y.2
>            MULADD     R4.x.3,    C5.x, R1.z.1F@R1.z, R4.x.2
>            AND_INT     R23.x.3,    R24.x.1, R24.y.2
>            (copy) MOV     R1.w.1F@R1.w,    t11F@R1.w
>            MULADD     R4.w.4FP@R0.w,    C6.w, R1.w.1F@R1.w, R4.w.3
>            MULADD     R4.z.4FP@R0.z,    C6.z, R1.w.1F@R1.w, R4.z.3
>            MULADD     R4.y.4FP@R0.y,    C6.y, R1.w.1F@R1.w, R4.y.3
>            MULADD     R4.x.4FP@R0.x,    C6.x, R1.w.1F@R1.w, R4.x.3
>            CNDE_INT     R23.y.3FP@R1.y,    R23.x.3, 0|00000000, 1|3f800000
>            CNDE_INT     R23.x.6FP@R1.x,    R23.x.3, 1|3f800000, 0|00000000
>            (copy) MOV     t7FP@R0.w,    R4.w.4FP@R0.w
>            (copy) MOV     t6FP@R0.z,    R4.z.4FP@R0.z
>            (copy) MOV     t5FP@R0.y,    R4.y.4FP@R0.y
>            (copy) MOV     t4FP@R0.x,    R4.x.4FP@R0.x
>            (copy) MOV     t3FP@R1.y,    R23.y.3FP@R1.y
>            (copy) MOV     t2FP@R1.x,    R23.x.6FP@R1.x
>        >  live_after: [t2FP@R1.x t3FP@R1.y t4FP@R0.x t5FP@R0.y t6FP@R0.z t7FP@R0.w ]
>        EXPORT  PARAM 0     t2FP@R1.x, t3FP@R1.y, 0|00000000, 1|3f800000
>        EXPORT  POS   60     t4FP@R0.x, t5FP@R0.y, t6FP@R0.z, t7FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t8F@R1.x, t9F@R1.y, t10F@R1.z, t11F@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  live_before: [t8F@R1.x t9F@R1.y t10F@R1.z t11F@R1.w ]
>            MULLO_INT     R23.x.1@R0.x, __, __, __,       C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003, C0.x, 4.2039e-45|00000003  
>            MOV     R8.y.1F,    2|40000000
>            MOV     R9.y.1F,    5|40a00000
>            MOV     R10.y.1F,    8|41000000
>            MOV     R11.y.1F,    11|41300000
>            MOV     R12.y.1F,    14|41600000
>            MOV     R13.y.1F,    17|41880000
>            MOV     R14.y.1F,    20|41a00000
>            MOV     R15.y.1F,    23|41b80000
>            MOV     R16.y.1F,    26|41d00000
>            ADD_INT     R23.x.2@R0.y,    C1.x, R23.x.1@R0.x
>            MOV     R8.z.1F,    3|40400000
>            MOV     R9.z.1F,    6|40c00000
>            MOV     R10.z.1F,    9|41100000
>            MOV     R11.z.1F,    12|41400000
>            MOV     R12.z.1F,    15|41700000
>            MOV     R13.z.1F,    18|41900000
>            MOV     R14.z.1F,    21|41a80000
>            MOV     R15.z.1F,    24|41c00000
>            MOV     R16.z.1F,    27|41d80000
>            (copy) MOV     R1.x.1F@R1.x,    t8F@R1.x
>            MUL     R4.w.1@R0.z,    C3.w, R1.x.1F@R1.x
>            MUL     R4.z.1@R0.w,    C3.z, R1.x.1F@R1.x
>            MUL     R4.y.1@R9.x,    C3.y, R1.x.1F@R1.x
>            MUL     R4.x.1@R11.y,    C3.x, R1.x.1F@R1.x
>            MOV     R8.x.1F,    1|3f800000
>            MOV     R9.x.1F,    4|40800000
>            MOV     R10.x.1F,    7|40e00000
>            MOV     R11.x.1F,    10|41200000
>            MOV     R12.x.1F,    13|41500000
>            MOV     R13.x.1F,    16|41800000
>            MOV     R14.x.1F,    19|41980000
>            MOV     R15.x.1F,    22|41b00000
>            MOV     R16.x.1F,    25|41c80000
>            SETE_DX10     R24.y.1@R11.z,    A8.y[R23.x.2@R0.y]_138F@R2.y, C2.y
>					    rels: A8.y[R23.x.2@R0.y]_138F@R2.y :  <= R8.y.1F, R9.y.1F, R10.y.1F, R11.y.1F, R12.y.1F, R13.y.1F, R14.y.1F, R15.y.1F, R16.y.1F
>            SETE_DX10     R24.z.1@R2.w,    A8.z[R23.x.2@R0.y]_141F@R2.z, C2.z
>					    rels: A8.z[R23.x.2@R0.y]_141F@R2.z :  <= R8.z.1F, R9.z.1F, R10.z.1F, R11.z.1F, R12.z.1F, R13.z.1F, R14.z.1F, R15.z.1F, R16.z.1F
>            (copy) MOV     R1.y.1F@R1.y,    t9F@R1.y
>            MULADD     R4.w.2@R10.x,    C4.w, R1.y.1F@R1.y, R4.w.1@R0.z
>            MULADD     R4.z.2@R2.y,    C4.z, R1.y.1F@R1.y, R4.z.1@R0.w
>            MULADD     R4.y.2@R0.z,    C4.y, R1.y.1F@R1.y, R4.y.1@R9.x
>            MULADD     R4.x.2@R0.w,    C4.x, R1.y.1F@R1.y, R4.x.1@R11.y
>            SETE_DX10     R24.x.1@R0.x,    A8.x[R23.x.2@R0.y]_135F@R0.x, C2.x
>					    rels: A8.x[R23.x.2@R0.y]_135F@R0.x :  <= R8.x.1F, R9.x.1F, R10.x.1F, R11.x.1F, R12.x.1F, R13.x.1F, R14.x.1F, R15.x.1F, R16.x.1F
>            AND_INT     R24.y.2@R0.y,    R24.y.1@R11.z, R24.z.1@R2.w
>            (copy) MOV     R1.z.1F@R1.z,    t10F@R1.z
>            MULADD     R4.w.3@R2.z,    C5.w, R1.z.1F@R1.z, R4.w.2@R10.x
>            MULADD     R4.z.3@R2.w,    C5.z, R1.z.1F@R1.z, R4.z.2@R2.y
>            MULADD     R4.y.3@R1.x,    C5.y, R1.z.1F@R1.z, R4.y.2@R0.z
>            MULADD     R4.x.3@R1.y,    C5.x, R1.z.1F@R1.z, R4.x.2@R0.w
>            AND_INT     R23.x.3@R1.z,    R24.x.1@R0.x, R24.y.2@R0.y
>            (copy) MOV     R1.w.1F@R1.w,    t11F@R1.w
>            MULADD     R4.w.4FP@R0.w,    C6.w, R1.w.1F@R1.w, R4.w.3@R2.z
>            MULADD     R4.z.4FP@R0.z,    C6.z, R1.w.1F@R1.w, R4.z.3@R2.w
>            MULADD     R4.y.4FP@R0.y,    C6.y, R1.w.1F@R1.w, R4.y.3@R1.x
>            MULADD     R4.x.4FP@R0.x,    C6.x, R1.w.1F@R1.w, R4.x.3@R1.y
>            CNDE_INT     R23.y.3FP@R1.y,    R23.x.3@R1.z, 0|00000000, 1|3f800000
>            CNDE_INT     R23.x.6FP@R1.x,    R23.x.3@R1.z, 1|3f800000, 0|00000000
>            (copy) MOV     t7FP@R0.w,    R4.w.4FP@R0.w
>            (copy) MOV     t6FP@R0.z,    R4.z.4FP@R0.z
>            (copy) MOV     t5FP@R0.y,    R4.y.4FP@R0.y
>            (copy) MOV     t4FP@R0.x,    R4.x.4FP@R0.x
>            (copy) MOV     t3FP@R1.y,    R23.y.3FP@R1.y
>            (copy) MOV     t2FP@R1.x,    R23.x.6FP@R1.x
>        >  live_after: [t2FP@R1.x t3FP@R1.y t4FP@R0.x t5FP@R0.y t6FP@R0.z t7FP@R0.w ]
>        EXPORT  PARAM 0     t2FP@R1.x, t3FP@R1.y, 0|00000000, 1|3f800000
>        EXPORT  POS   60     t4FP@R0.x, t5FP@R0.y, t6FP@R0.z, t7FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y R0.zF@R0.z R0.wF@R0.w ]
>        CALL_FS     t8||FP@R1.x, t9||FP@R1.y, t10||FP@R1.z, t11||FP@R1.w,       R0.xF@R0.x, R0.yF@R0.y, R0.zF@R0.z, R0.wF@R0.w
>        ALU     
>        <  
>            [  
>                4S  MULLO_INT     R23.x.1F@R124.x,    C0.x, 4.2039e-45|00000003
>                4S  MULLO_INT     __,    C0.x, 4.2039e-45|00000003
>                4S  MULLO_INT     __,    C0.x, 4.2039e-45|00000003
>                4S  MULLO_INT     __,    C0.x, 4.2039e-45|00000003
>            ]  
>            [  
>                ADD_INT     R23.x.2F@R124.y,    C1.x, R23.x.1F@R124.x
>                MOV     R16.z.1F,    27|41d80000
>            ]  
>            [  
>                MOVA_INT     __,    R23.x.2F@R124.y
>            ]  
>            [  
>                MUL     R4.y.1F@R125.x,    C3.y, R1.x.1||FP@R1.x
>                MOV     R16.y.1F,    26|41d00000
>                MOV     R15.z.1F,    24|41c00000
>            ]  
>            [  
>                MOV     R16.x.1F,    25|41c80000
>                MOV     R15.y.1F,    23|41b80000
>                MOV     R14.z.1F,    21|41a80000
>            ]  
>            [  
>                MOV     R15.x.1F,    22|41b00000
>                MOV     R14.y.1F,    20|41a00000
>                MOV     R13.z.1F,    18|41900000
>            ]  
>            [  
>                MOV     R14.x.1F,    19|41980000
>                MOV     R13.y.1F,    17|41880000
>                MOV     R12.z.1F,    15|41700000
>            ]  
>            [  
>                MOV     R13.x.1F,    16|41800000
>                MOV     R12.y.1F,    14|41600000
>                MOV     R11.z.1F,    12|41400000
>            ]  
>            [  
>                MOV     R12.x.1F,    13|41500000
>                MOV     R11.y.1F,    11|41300000
>                MOV     R10.z.1F,    9|41100000
>            ]  
>            [  
>                MOV     R11.x.1F,    10|41200000
>                MOV     R10.y.1F,    8|41000000
>                MOV     R9.z.1F,    6|40c00000
>            ]  
>            [  
>                MOV     R10.x.1F,    7|40e00000
>                MOV     R9.y.1F,    5|40a00000
>                MOV     R8.z.1F,    3|40400000
>            ]  
>            [  
>                MOV     R9.x.1F,    4|40800000
>                MOV     R8.y.1F,    2|40000000
>                MUL     R4.w.1F@R124.z,    C3.w, R1.x.1||FP@R1.x
>                MUL     R4.z.1F@R125.w,    C3.z, R1.x.1||FP@R1.x
>            ]  
>            [  
>                MOV     R8.x.1F,    1|3f800000
>                MUL     R4.x.1F@R124.y,    C3.x, R1.x.1||FP@R1.x
>                SETE_DX10     R24.z.1F@R124.w,    A8.z[R23.x.2F@R124.y]_141F@R2.z, C2.z
>					    rels: A8.z[R23.x.2F@R124.y]_141F@R2.z :  <= R8.z.1F, R9.z.1F, R10.z.1F, R11.z.1F, R12.z.1F, R13.z.1F, R14.z.1F, R15.z.1F, R16.z.1F
>            ]  
>            [  
>                MULADD     R4.w.2F@R124.x,    C4.w, R1.y.1||FP@R1.y, R4.w.1F@R124.z
>                MULADD     R4.z.2F@R126.y,    C4.z, R1.y.1||FP@R1.y, R4.z.1F@R125.w
>                SETE_DX10     R24.y.1F@R124.z,    A8.y[R23.x.2F@R124.y]_138F@R2.y, C2.y
>					    rels: A8.y[R23.x.2F@R124.y]_138F@R2.y :  <= R8.y.1F, R9.y.1F, R10.y.1F, R11.y.1F, R12.y.1F, R13.y.1F, R14.y.1F, R15.y.1F, R16.y.1F
>            ]  
>            [  
>                SETE_DX10     R24.x.1F@R125.x,    A8.x[R23.x.2F@R124.y]_135F@R0.x, C2.x
>					    rels: A8.x[R23.x.2F@R124.y]_135F@R0.x :  <= R8.x.1F, R9.x.1F, R10.x.1F, R11.x.1F, R12.x.1F, R13.x.1F, R14.x.1F, R15.x.1F, R16.x.1F
>                AND_INT     R24.y.2F@R125.y,    R24.y.1F@R124.z, R24.z.1F@R124.w
>                MULADD     R4.y.2F@R124.z,    C4.y, R1.y.1||FP@R1.y, R4.y.1F@R125.x
>                MULADD     R4.x.2F@R124.w,    C4.x, R1.y.1||FP@R1.y, R4.x.1F@R124.y
>            ]  
>            [  
>                MULADD     R4.y.3F@R124.x,    C5.y, R1.z.1||FP@R1.z, R4.y.2F@R124.z
>                MULADD     R4.x.3F@R124.y,    C5.x, R1.z.1||FP@R1.z, R4.x.2F@R124.w
>                MULADD     R4.w.3F@R125.z,    C5.w, R1.z.1||FP@R1.z, R4.w.2F@R124.x
>            ]  
>            [  
>                MULADD     R4.x.4||FP@R0.x,    C6.x, R1.w.1||FP@R1.w, R4.x.3F@R124.y
>                MULADD     R4.y.4||FP@R0.y,    C6.y, R1.w.1||FP@R1.w, R4.y.3F@R124.x
>                AND_INT     R23.x.3F@R124.z,    R24.x.1F@R125.x, R24.y.2F@R125.y
>                MULADD     R4.z.3F@R124.w,    C5.z, R1.z.1||FP@R1.z, R4.z.2F@R126.y
>            ]  
>            [  
>                CNDE_INT     R23.x.6||FP@R1.x,    R23.x.3F@R124.z, 1|3f800000, 0|00000000
>                CNDE_INT     R23.y.3||FP@R1.y,    R23.x.3F@R124.z, 0|00000000, 1|3f800000
>                MULADD     R4.z.4||FP@R0.z,    C6.z, R1.w.1||FP@R1.w, R4.z.3F@R124.w
>                MULADD     R4.w.4||FP@R0.w,    C6.w, R1.w.1||FP@R1.w, R4.w.3F@R125.z
>            ]  
>        >  
>        EXPORT  PARAM 0     t2||FP@R1.x, t3||FP@R1.y, 0|00000000, 1|3f800000
>        EXPORT  POS   60     t4||FP@R0.x, t5||FP@R0.y, t6||FP@R0.z, t7||FP@R0.w
>    } end BB_0  
>}  
>
>===== SHADER #24 OPT ==================================== VS/CAYMAN/CAYMAN =====
>===== 158 dw ===== 11 gprs ===== 1 stack =======================================
>0000  00000000 84c00000 CALL_FS @0
>0002  40000005 a1240000 ALU 74 @10 KC0[CB0:0-15]
> 0010  001fa080 0f804790     1      x: MULLO_INT          T0.x,  KC0[0].x, [0x00000003 4.2039e-45].x
> 0012  001fa080 20004780            y: MULLO_INT          __.y,  KC0[0].x, [0x00000003 4.2039e-45].x
> 0014  001fa080 40004780            z: MULLO_INT          __.z,  KC0[0].x, [0x00000003 4.2039e-45].x
> 0016  801fa080 60004780            w: MULLO_INT          __.w,  KC0[0].x, [0x00000003 4.2039e-45].x
> 0018  00000003 
> 0020  000f8081 2f801a10     2      y: ADD_INT            T0.y,  KC0[1].x, T0.x
> 0022  800000fd 41400c90            z: MOV                R10.z,  [0x41d80000 27].x
> 0024  41d80000 
> 0026  8000047c 00006600     3      x: MOVA_INT           __.x,  T0.y
> 0028  00002483 0fa00090     4      x: MUL                T1.x,  KC0[3].y, R1.x
> 0030  000000fd 21400c90            y: MOV                R10.y,  [0x41d00000 26].x
> 0032  800004fd 41200c90            z: MOV                R9.z,  [0x41c00000 24].y
> 0034  41d00000 
> 0035  41c00000 
> 0036  000000fd 01000c90     5      x: MOV                R8.x,  [0x41c80000 25].x
> 0038  000004fd 21200c90            y: MOV                R9.y,  [0x41b80000 23].y
> 0040  800008fd 41000c90            z: MOV                R8.z,  [0x41a80000 21].z
> 0042  41c80000 
> 0043  41b80000 
> 0044  41a80000 
> 0046  000000fd 00e00c90     6      x: MOV                R7.x,  [0x41b00000 22].x
> 0048  000004fd 21000c90            y: MOV                R8.y,  [0x41a00000 20].y
> 0050  800008fd 40e00c90            z: MOV                R7.z,  [0x41900000 18].z
> 0052  41b00000 
> 0053  41a00000 
> 0054  41900000 
> 0056  000000fd 00c00c90     7      x: MOV                R6.x,  [0x41980000 19].x
> 0058  000004fd 20e00c90            y: MOV                R7.y,  [0x41880000 17].y
> 0060  800008fd 40c00c90            z: MOV                R6.z,  [0x41700000 15].z
> 0062  41980000 
> 0063  41880000 
> 0064  41700000 
> 0066  000000fd 00a00c90     8      x: MOV                R5.x,  [0x41800000 16].x
> 0068  000004fd 20c00c90            y: MOV                R6.y,  [0x41600000 14].y
> 0070  800008fd 40a00c90            z: MOV                R5.z,  [0x41400000 12].z
> 0072  41800000 
> 0073  41600000 
> 0074  41400000 
> 0076  000000fd 00800c90     9      x: MOV                R4.x,  [0x41500000 13].x
> 0078  000004fd 20a00c90            y: MOV                R5.y,  [0x41300000 11].y
> 0080  800008fd 40800c90            z: MOV                R4.z,  [0x41100000 9].z
> 0082  41500000 
> 0083  41300000 
> 0084  41100000 
> 0086  000000fd 00600c90    10      x: MOV                R3.x,  [0x41200000 10].x
> 0088  000004fd 20800c90            y: MOV                R4.y,  [0x41000000 8].y
> 0090  800008fd 40600c90            z: MOV                R3.z,  [0x40c00000 6].z
> 0092  41200000 
> 0093  41000000 
> 0094  40c00000 
> 0096  000000fd 00400c90    11      x: MOV                R2.x,  [0x40e00000 7].x
> 0098  000004fd 20600c90            y: MOV                R3.y,  [0x40a00000 5].y
> 0100  800008fd 40400c90            z: MOV                R2.z,  [0x40400000 3].z
> 0102  40e00000 
> 0103  40a00000 
> 0104  40400000 
> 0106  000000fd 00200c90    12      x: MOV                R1.x,  [0x40800000 4].x
> 0108  000004fd 20400c90            y: MOV                R2.y,  [0x40000000 2].y
> 0110  00002c83 4f800090            z: MUL                T0.z,  KC0[3].w, R1.x
> 0112  80002883 6fa00090            w: MUL                T1.w,  KC0[3].z, R1.x
> 0114  40800000 
> 0115  40000000 
> 0116  000000f9 00000c90    13      x: MOV                R0.x,  1.0
> 0118  00002083 2f800090            y: MUL                T0.y,  KC0[3].x, R1.x
> 0120  81104a02 6f800610            w: SETE_DX10          T0.w,  R[2+AR].z, KC0[2].z
> 0122  00802c84 0f82887c    14      x: MULADD             T0.x,  KC0[4].w, R1.y, T0.z
> 0124  00802884 2fc28c7d            y: MULADD             T2.y,  KC0[4].z, R1.y, T1.w
> 0126  80904602 4f800610            z: SETE_DX10          T0.z,  R[2+AR].y, KC0[2].y
> 0128  00104200 0fa00610    15      x: SETE_DX10          T1.x,  R[0+AR].x, KC0[2].x
> 0130  018f887c 2fa01810            y: AND_INT            T1.y,  T0.z, T0.w
> 0132  00802484 4f82807d            z: MULADD             T0.z,  KC0[4].y, R1.y, T1.x
> 0134  80802084 6f82847c            w: MULADD             T0.w,  KC0[4].x, R1.y, T0.y
> 0136  01002485 0f82887c    16      x: MULADD             T0.x,  KC0[5].y, R1.z, T0.z
> 0138  01002085 2f828c7c            y: MULADD             T0.y,  KC0[5].x, R1.z, T0.w
> 0140  81002c85 4fa2807c            z: MULADD             T1.z,  KC0[5].w, R1.z, T0.x
> 0142  01802086 0002847c    17      x: MULADD             R0.x,  KC0[6].x, R1.w, T0.y
> 0144  01802486 2002807c            y: MULADD             R0.y,  KC0[6].y, R1.w, T0.x
> 0146  008fa07d 4f801810            z: AND_INT            T0.z,  T1.x, T1.y
> 0148  81002885 6f8a847e            w: MULADD             T0.w,  KC0[5].z, R1.z, T2.y   VEC_120
> 0150  001f287c 002380f8    18      x: CNDE_INT           R1.x,  T0.z, 1.0, 0
> 0152  001f087c 202380f9            y: CNDE_INT           R1.y,  T0.z, 0, 1.0
> 0154  01802886 40028c7c            z: MULADD             R0.z,  KC0[6].z, R1.w, T0.w
> 0156  81802c86 6002887d            w: MULADD             R0.w,  KC0[6].w, R1.w, T1.z
>0004  c000c000 95000b08 EXPORT_DONE        PARAM 0     R1.xy01
>0006  c000203c 95000688 EXPORT_DONE        POS   60    R0.xyzw
>0008  00000000 88000000 CF_END @0
>===== SHADER_END ===============================================================
>
>
>
>###### after parse
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    ALU     
>    <  
>        [  
>            INTERP_ZW       
>                4S  INTERP_ZW     __,    R0.yF@R0.y, Param0x
>                4S  INTERP_ZW     __,    R0.xF@R0.x, Param0y
>                4S  INTERP_ZW     R2.z,    R0.yF@R0.y, Param0z
>                4S  INTERP_ZW     R2.w,    R0.xF@R0.x, Param0w
>        ]  
>        [  
>            INTERP_XY       
>                4S  INTERP_XY     R2.x,    R0.yF@R0.y, Param0x
>                4S  INTERP_XY     R2.y,    R0.xF@R0.x, Param0y
>                4S  INTERP_XY     __,    R0.yF@R0.y, Param0z
>                4S  INTERP_XY     __,    R0.xF@R0.x, Param0w
>        ]  
>        [  
>            MOV     R3.x,    R2.x
>            MOV     R3.y,    R2.y
>            MOV     R3.z,    R2.z
>            MOV     R3.w,    R2.w
>        ]  
>    >  
>    EXPORT  PIXEL 0     R3.x, R3.y, R3.z, R3.w
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    ALU     
>    <  
>        [  
>            INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>        ]  
>        [  
>            INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>        ]  
>        [  
>            MOV     R3.x.1,    R2.x.1
>            MOV     R3.y.1,    R2.y.1
>            MOV     R3.z.1,    R2.z.1
>            MOV     R3.w.1,    R2.w.1
>        ]  
>    >  
>    EXPORT  PIXEL 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>    CF_END     
>}  
>
>
>###### after if_conversion
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    MOV     R3.x.1,    R2.x.1
>    MOV     R3.y.1,    R2.y.1
>    MOV     R3.z.1,    R2.z.1
>    MOV     R3.w.1,    R2.w.1
>    EXPORT  PIXEL 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>}  
>
>
>###### after peephole
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    MOV     R3.x.1,    R2.x.1
>    MOV     R3.y.1,    R2.y.1
>    MOV     R3.z.1,    R2.z.1
>    MOV     R3.w.1,    R2.w.1
>    EXPORT  PIXEL 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>}  
>
>
>###### after gvn
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    MOV     R3.x.1,    R2.x.1
>    MOV     R3.y.1,    R2.y.1
>    MOV     R3.z.1,    R2.z.1
>    MOV     R3.w.1,    R2.w.1
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after def_use
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    MOV     R3.x.1,    R2.x.1
>    MOV     R3.y.1,    R2.y.1
>    MOV     R3.z.1,    R2.z.1
>    MOV     R3.w.1,    R2.w.1
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after gcm
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            (copy) MOV     t4,    R2.w.1
>            (copy) MOV     t3,    R2.z.1
>            (copy) MOV     t2,    R2.y.1
>            (copy) MOV     t1,    R2.x.1
>        >  
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>        ALU     
>        <  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>            INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            (copy) MOV     t4,    R2.w.1
>            (copy) MOV     t3,    R2.z.1
>            (copy) MOV     t2,    R2.y.1
>            (copy) MOV     t1,    R2.x.1
>        >  live_after: [t1 t2 t3 t4 ]
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>        ALU     
>        <  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>            INTERP_ZW     __, __, R2.z.1FP@R0.z, R2.w.1FP@R0.w,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            INTERP_XY     R2.x.1FP@R0.x, R2.y.1FP@R0.y, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            (copy) MOV     t4FP@R0.w,    R2.w.1FP@R0.w
>            (copy) MOV     t3FP@R0.z,    R2.z.1FP@R0.z
>            (copy) MOV     t2FP@R0.y,    R2.y.1FP@R0.y
>            (copy) MOV     t1FP@R0.x,    R2.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>        ALU     
>        <  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>            INTERP_ZW     __, __, R2.z.1FP@R0.z, R2.w.1FP@R0.w,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            INTERP_XY     R2.x.1FP@R0.x, R2.y.1FP@R0.y, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            (copy) MOV     t4FP@R0.w,    R2.w.1FP@R0.w
>            (copy) MOV     t3FP@R0.z,    R2.z.1FP@R0.z
>            (copy) MOV     t2FP@R0.y,    R2.y.1FP@R0.y
>            (copy) MOV     t1FP@R0.x,    R2.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{   preloaded inputs [R0.x||FP@R0.x, R0.y||FP@R0.y]  
>    { BB_0    loop_level = 0  live_before: [R0.x||FP@R0.x R0.y||FP@R0.y ]
>        ALU     
>        <  
>            [  
>                4S  INTERP_ZW     __,    R0.y||FP@R0.y, Param0x
>                4S  INTERP_ZW     __,    R0.x||FP@R0.x, Param0y
>                4S  INTERP_ZW     R2.z.1||FP@R0.z,    R0.y||FP@R0.y, Param0z
>                4S  INTERP_ZW     R2.w.1||FP@R0.w,    R0.x||FP@R0.x, Param0w
>            ]  
>            [  
>                4S  INTERP_XY     R2.x.1||FP@R0.x,    R0.y||FP@R0.y, Param0x
>                4S  INTERP_XY     R2.y.1||FP@R0.y,    R0.x||FP@R0.x, Param0y
>                4S  INTERP_XY     __,    R0.y||FP@R0.y, Param0z
>                4S  INTERP_XY     __,    R0.x||FP@R0.x, Param0w
>            ]  
>        >  
>        EXPORT  PIXEL 0     t1||FP@R0.x, t2||FP@R0.y, t3||FP@R0.z, t4||FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after parse
>{  
>    ALU     
>    <  
>        [  
>            INTERP_LOAD_P0     R1.x,    Param0x
>            INTERP_LOAD_P0     R1.y,    Param0y
>            INTERP_LOAD_P0     R1.z,    Param0z
>            INTERP_LOAD_P0     R1.w,    Param0w
>        ]  
>        [  
>            MOV     R2.x,    R1.x
>            MOV     R2.y,    R1.y
>            MOV     R2.z,    R1.z
>            MOV     R2.w,    R1.w
>        ]  
>    >  
>    EXPORT  PIXEL 0     R2.x, R2.y, R2.z, R2.w
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{  
>    ALU     
>    <  
>        [  
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>        ]  
>        [  
>            MOV     R2.x.1,    R1.x.1
>            MOV     R2.y.1,    R1.y.1
>            MOV     R2.z.1,    R1.z.1
>            MOV     R2.w.1,    R1.w.1
>        ]  
>    >  
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>    CF_END     
>}  
>
>
>###### after if_conversion
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after peephole
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R2.x.1, R2.y.1, R2.z.1, R2.w.1
>}  
>
>
>###### after gvn
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after def_use
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    MOV     R2.x.1,    R1.x.1
>    MOV     R2.y.1,    R1.y.1
>    MOV     R2.z.1,    R1.z.1
>    MOV     R2.w.1,    R1.w.1
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after dce_cleanup
>{  
>    INTERP_LOAD_P0     R1.x.1,    Param0x
>    INTERP_LOAD_P0     R1.y.1,    Param0y
>    INTERP_LOAD_P0     R1.z.1,    Param0z
>    INTERP_LOAD_P0     R1.w.1,    Param0w
>    EXPORT  PIXEL 0     R1.x.1, R1.y.1, R1.z.1, R1.w.1
>}  
>
>
>###### after gcm
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            (copy) MOV     t4,    R1.w.1
>            (copy) MOV     t3,    R1.z.1
>            (copy) MOV     t2,    R1.y.1
>            (copy) MOV     t1,    R1.x.1
>        >  
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1,    Param0w
>            INTERP_LOAD_P0     R1.z.1,    Param0z
>            INTERP_LOAD_P0     R1.y.1,    Param0y
>            INTERP_LOAD_P0     R1.x.1,    Param0x
>            (copy) MOV     t4,    R1.w.1
>            (copy) MOV     t3,    R1.z.1
>            (copy) MOV     t2,    R1.y.1
>            (copy) MOV     t1,    R1.x.1
>        >  live_after: [t1 t2 t3 t4 ]
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1FP@R0.w,    Param0w
>            INTERP_LOAD_P0     R1.z.1FP@R0.z,    Param0z
>            INTERP_LOAD_P0     R1.y.1FP@R0.y,    Param0y
>            INTERP_LOAD_P0     R1.x.1FP@R0.x,    Param0x
>            (copy) MOV     t4FP@R0.w,    R1.w.1FP@R0.w
>            (copy) MOV     t3FP@R0.z,    R1.z.1FP@R0.z
>            (copy) MOV     t2FP@R0.y,    R1.y.1FP@R0.y
>            (copy) MOV     t1FP@R0.x,    R1.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_LOAD_P0     R1.w.1FP@R0.w,    Param0w
>            INTERP_LOAD_P0     R1.z.1FP@R0.z,    Param0z
>            INTERP_LOAD_P0     R1.y.1FP@R0.y,    Param0y
>            INTERP_LOAD_P0     R1.x.1FP@R0.x,    Param0x
>            (copy) MOV     t4FP@R0.w,    R1.w.1FP@R0.w
>            (copy) MOV     t3FP@R0.z,    R1.z.1FP@R0.z
>            (copy) MOV     t2FP@R0.y,    R1.y.1FP@R0.y
>            (copy) MOV     t1FP@R0.x,    R1.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            [  
>                INTERP_LOAD_P0     R1.x.1||FP@R0.x,    Param0x
>                INTERP_LOAD_P0     R1.y.1||FP@R0.y,    Param0y
>                INTERP_LOAD_P0     R1.z.1||FP@R0.z,    Param0z
>                INTERP_LOAD_P0     R1.w.1||FP@R0.w,    Param0w
>            ]  
>        >  
>        EXPORT  PIXEL 0     t1||FP@R0.x, t2||FP@R0.y, t3||FP@R0.z, t4||FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after parse
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    ALU     
>    <  
>        [  
>            INTERP_ZW       
>                4S  INTERP_ZW     __,    R0.yF@R0.y, Param0x
>                4S  INTERP_ZW     __,    R0.xF@R0.x, Param0y
>                4S  INTERP_ZW     R2.z,    R0.yF@R0.y, Param0z
>                4S  INTERP_ZW     R2.w,    R0.xF@R0.x, Param0w
>        ]  
>        [  
>            INTERP_XY       
>                4S  INTERP_XY     R2.x,    R0.yF@R0.y, Param0x
>                4S  INTERP_XY     R2.y,    R0.xF@R0.x, Param0y
>                4S  INTERP_XY     __,    R0.yF@R0.y, Param0z
>                4S  INTERP_XY     __,    R0.xF@R0.x, Param0w
>        ]  
>    >  
>    TEX     
>    <  
>        SAMPLE     R3.x, R3.y, R3.z, R3.w,       R2.x, R2.y, __, __
>    >  
>    EXPORT  PIXEL 0     R3.x, R3.y, R3.z, R3.w
>    CF_END     
>}  
>
>
>###### after ssa_rename
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    ALU     
>    <  
>        [  
>            INTERP_ZW     __, __, R2.z.1, R2.w.1,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>        ]  
>        [  
>            INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>        ]  
>    >  
>    TEX     
>    <  
>        SAMPLE     R3.x.1, R3.y.1, R3.z.1, R3.w.1,       R2.x.1, R2.y.1, __, __
>    >  
>    EXPORT  PIXEL 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>    CF_END     
>}  
>
>
>###### after if_conversion
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    SAMPLE     R3.x.1, R3.y.1, R3.z.1, R3.w.1,       R2.x.1, R2.y.1, __, __
>    EXPORT  PIXEL 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>}  
>
>
>###### after peephole
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    SAMPLE     R3.x.1, R3.y.1, R3.z.1, R3.w.1,       R2.x.1, R2.y.1, __, __
>    EXPORT  PIXEL 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>}  
>
>
>###### after gvn
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    SAMPLE     R3.x.1, R3.y.1, R3.z.1, R3.w.1,       R2.x.1, R2.y.1, __, __
>    EXPORT  PIXEL 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>}  
>
>
>###### after def_use
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    SAMPLE     R3.x.1, R3.y.1, R3.z.1, R3.w.1,       R2.x.1, R2.y.1, __, __
>    EXPORT  PIXEL 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>    SAMPLE     R3.x.1, R3.y.1, R3.z.1, R3.w.1,       R2.x.1, R2.y.1, __, __
>    EXPORT  PIXEL 0     R3.x.1, R3.y.1, R3.z.1, R3.w.1
>}  
>
>
>###### after gcm
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    { BB_0    loop_level = 0  
>        ALU     
>        <  
>            INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            (copy) MOV     t6,    R2.y.1
>            (copy) MOV     t5,    R2.x.1
>        >  
>        TEX     
>        <  
>            SAMPLE     t7, t8, t9, t10,       t5, t6, __, __
>        >  
>        ALU     
>        <  
>            (copy) MOV     R3.w.1,    t10
>            (copy) MOV     t4,    R3.w.1
>            (copy) MOV     R3.z.1,    t9
>            (copy) MOV     t3,    R3.z.1
>            (copy) MOV     R3.y.1,    t8
>            (copy) MOV     t2,    R3.y.1
>            (copy) MOV     R3.x.1,    t7
>            (copy) MOV     t1,    R3.x.1
>        >  
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after dce_cleanup
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>        ALU     
>        <  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>            INTERP_XY     R2.x.1, R2.y.1, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            (copy) MOV     t6,    R2.y.1
>            (copy) MOV     t5,    R2.x.1
>        >  live_after: [t5 t6 ]
>        TEX     
>        <  live_before: [t5 t6 ]
>            SAMPLE     t7, t8, t9, t10,       t5, t6, __, __
>        >  live_after: [t7 t8 t9 t10 ]
>        ALU     
>        <  live_before: [t7 t8 t9 t10 ]
>            (copy) MOV     R3.w.1,    t10
>            (copy) MOV     t4,    R3.w.1
>            (copy) MOV     R3.z.1,    t9
>            (copy) MOV     t3,    R3.z.1
>            (copy) MOV     R3.y.1,    t8
>            (copy) MOV     t2,    R3.y.1
>            (copy) MOV     R3.x.1,    t7
>            (copy) MOV     t1,    R3.x.1
>        >  live_after: [t1 t2 t3 t4 ]
>        EXPORT  PIXEL 0     t1, t2, t3, t4
>    } end BB_0  
>}  
>
>
>###### after ra_coalesce
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>        ALU     
>        <  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>            INTERP_XY     R2.x.1FP@R0.x, R2.y.1FP@R0.y, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            (copy) MOV     t6FP@R0.y,    R2.y.1FP@R0.y
>            (copy) MOV     t5FP@R0.x,    R2.x.1FP@R0.x
>        >  live_after: [t5FP@R0.x t6FP@R0.y ]
>        TEX     
>        <  live_before: [t5FP@R0.x t6FP@R0.y ]
>            SAMPLE     t7FP@R0.x, t8FP@R0.y, t9FP@R0.z, t10FP@R0.w,       t5FP@R0.x, t6FP@R0.y, __, __
>        >  live_after: [t7FP@R0.x t8FP@R0.y t9FP@R0.z t10FP@R0.w ]
>        ALU     
>        <  live_before: [t7FP@R0.x t8FP@R0.y t9FP@R0.z t10FP@R0.w ]
>            (copy) MOV     R3.w.1FP@R0.w,    t10FP@R0.w
>            (copy) MOV     t4FP@R0.w,    R3.w.1FP@R0.w
>            (copy) MOV     R3.z.1FP@R0.z,    t9FP@R0.z
>            (copy) MOV     t3FP@R0.z,    R3.z.1FP@R0.z
>            (copy) MOV     R3.y.1FP@R0.y,    t8FP@R0.y
>            (copy) MOV     t2FP@R0.y,    R3.y.1FP@R0.y
>            (copy) MOV     R3.x.1FP@R0.x,    t7FP@R0.x
>            (copy) MOV     t1FP@R0.x,    R3.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after ra_init
>{   preloaded inputs [R0.xF@R0.x, R0.yF@R0.y]  
>    { BB_0    loop_level = 0  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>        ALU     
>        <  live_before: [R0.xF@R0.x R0.yF@R0.y ]
>            INTERP_XY     R2.x.1FP@R0.x, R2.y.1FP@R0.y, __, __,       R0.yF@R0.y, Param0x, R0.xF@R0.x, Param0y, R0.yF@R0.y, Param0z, R0.xF@R0.x, Param0w  
>            (copy) MOV     t6FP@R0.y,    R2.y.1FP@R0.y
>            (copy) MOV     t5FP@R0.x,    R2.x.1FP@R0.x
>        >  live_after: [t5FP@R0.x t6FP@R0.y ]
>        TEX     
>        <  live_before: [t5FP@R0.x t6FP@R0.y ]
>            SAMPLE     t7FP@R0.x, t8FP@R0.y, t9FP@R0.z, t10FP@R0.w,       t5FP@R0.x, t6FP@R0.y, __, __
>        >  live_after: [t7FP@R0.x t8FP@R0.y t9FP@R0.z t10FP@R0.w ]
>        ALU     
>        <  live_before: [t7FP@R0.x t8FP@R0.y t9FP@R0.z t10FP@R0.w ]
>            (copy) MOV     R3.w.1FP@R0.w,    t10FP@R0.w
>            (copy) MOV     t4FP@R0.w,    R3.w.1FP@R0.w
>            (copy) MOV     R3.z.1FP@R0.z,    t9FP@R0.z
>            (copy) MOV     t3FP@R0.z,    R3.z.1FP@R0.z
>            (copy) MOV     R3.y.1FP@R0.y,    t8FP@R0.y
>            (copy) MOV     t2FP@R0.y,    R3.y.1FP@R0.y
>            (copy) MOV     R3.x.1FP@R0.x,    t7FP@R0.x
>            (copy) MOV     t1FP@R0.x,    R3.x.1FP@R0.x
>        >  live_after: [t1FP@R0.x t2FP@R0.y t3FP@R0.z t4FP@R0.w ]
>        EXPORT  PIXEL 0     t1FP@R0.x, t2FP@R0.y, t3FP@R0.z, t4FP@R0.w
>    } end BB_0  
>}  
>
>
>###### after post_scheduler
>{   preloaded inputs [R0.x||FP@R0.x, R0.y||FP@R0.y]  
>    { BB_0    loop_level = 0  live_before: [R0.x||FP@R0.x R0.y||FP@R0.y ]
>        ALU     
>        <  
>            [  
>                4S  INTERP_XY     R2.x.1||FP@R0.x,    R0.y||FP@R0.y, Param0x
>                4S  INTERP_XY     R2.y.1||FP@R0.y,    R0.x||FP@R0.x, Param0y
>                4S  INTERP_XY     __,    R0.y||FP@R0.y, Param0z
>                4S  INTERP_XY     __,    R0.x||FP@R0.x, Param0w
>            ]  
>        >  
>        TEX     
>        <  live_before: [t5||FP@R0.x t6||FP@R0.y ]
>            SAMPLE     t7||FP@R0.x, t8||FP@R0.y, t9||FP@R0.z, t10||FP@R0.w,       t5||FP@R0.x, t6||FP@R0.y, __, __
>        >  live_after: [t7||FP@R0.x t8||FP@R0.y t9||FP@R0.z t10||FP@R0.w ]
>        EXPORT  PIXEL 0     t1||FP@R0.x, t2||FP@R0.y, t3||FP@R0.z, t4||FP@R0.w
>    } end BB_0  
>}  
>Probe color at (10,10)
>  Expected: 0.000000 1.000000 0.000000
>  Observed: 0.501961 0.501961 0.501961
>Probe color at (25,10)
>  Expected: 0.000000 1.000000 0.000000
>  Observed: 0.501961 0.501961 0.501961
>Probe color at (40,10)
>  Expected: 0.000000 1.000000 0.000000
>  Observed: 0.501961 0.501961 0.501961
>Probe color at (65,10)
>  Expected: 0.000000 1.000000 0.000000
>  Observed: 0.501961 0.501961 0.501961
>Probe color at (80,10)
>  Expected: 0.000000 1.000000 0.000000
>  Observed: 0.501961 0.501961 0.501961
>Probe color at (95,10)
>  Expected: 0.000000 1.000000 0.000000
>  Observed: 0.501961 0.501961 0.501961
>Probe color at (120,10)
>  Expected: 0.000000 1.000000 0.000000
>  Observed: 0.501961 0.501961 0.501961
>Probe color at (135,10)
>  Expected: 0.000000 1.000000 0.000000
>  Observed: 0.501961 0.501961 0.501961
>Probe color at (150,10)
>  Expected: 0.000000 1.000000 0.000000
>  Observed: 0.501961 0.501961 0.501961
>PIGLIT: {'result': 'fail' }

Actions: View

Attachments on bug 71944: 89688 | 89689 | 89694 | 89964

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.