[Mesa-dev] [PATCH 2/3] nir/algebraic: Add more lowering

Iago Toral itoral at igalia.com
Thu Jan 7 07:21:16 PST 2016


On Wed, 2016-01-06 at 15:30 -0800, Jason Ekstrand wrote:
> This commit adds lowering options for the following opcodes:
> 
>  - nir_op_fmod
>  - nir_op_bitfield_insert
>  - nir_op_uadd_carry
>  - nir_op_usub_borrow
> ---
>  src/glsl/nir/nir.h                | 4 ++++
>  src/glsl/nir/nir_opt_algebraic.py | 6 ++++++
>  2 files changed, 10 insertions(+)
> 
> diff --git a/src/glsl/nir/nir.h b/src/glsl/nir/nir.h
> index 59f6f68..61e51da 100644
> --- a/src/glsl/nir/nir.h
> +++ b/src/glsl/nir/nir.h
> @@ -1509,6 +1509,10 @@ typedef struct nir_shader_compiler_options {
>     bool lower_fpow;
>     bool lower_fsat;
>     bool lower_fsqrt;
> +   bool lower_fmod;
> +   bool lower_bitfield_insert;
> +   bool lower_uadd_carry;
> +   bool lower_usub_borrow;
>     /** lowers fneg and ineg to fsub and isub. */
>     bool lower_negate;
>     /** lowers fsub and isub to fadd+fneg and iadd+ineg. */
> diff --git a/src/glsl/nir/nir_opt_algebraic.py b/src/glsl/nir/nir_opt_algebraic.py
> index a5a4841..f4a8632 100644
> --- a/src/glsl/nir/nir_opt_algebraic.py
> +++ b/src/glsl/nir/nir_opt_algebraic.py
> @@ -222,6 +222,12 @@ optimizations = [
>     (('iadd', a, ('isub', 0, b)), ('isub', a, b)),
>     (('fabs', ('fsub', 0.0, a)), ('fabs', a)),
>     (('iabs', ('isub', 0, a)), ('iabs', a)),
> +
> +   # Misc. lowering
> +   (('fmod', a, b), ('fsub', a, ('fmul', b, ('ffloor', ('fdiv', a, b)))), 'options->lower_fmod'),
> +   (('bitfield_insert', a, b, c, d), ('bfi', ('bfm', d, c), b, a), 'options->lower_bitfield_insert'),
> +   (('uadd_carry', a, b), ('ult', ('iadd', a, b), a), 'options->lower_uadd_carry'),
> +   (('usub_borrow', a, b), ('ult', a, b), 'options->lower_usub_borrow'),

Don't we need to use b2i with uadd_carry and usub_borrow? ult seems to
return NIR_TRUE (~0u) instead of 1.

With that change (or confirmation that is not needed):
Reviewed-by: Iago Toral Quiroga <itoral at igalia.com>

Iago

>  ]
>  
>  # Add optimizations to handle the case where the result of a ternary is




More information about the mesa-dev mailing list