Revision 3657cf5f363fd64aeaf06432e62e3960800927b0 authored by Andrew Adams on 26 January 2024, 17:26:12 UTC, committed by GitHub on 26 January 2024, 17:26:12 UTC
* Fix bounds_of_nested_lanes bounds_of_nested_lanes assumed that one layer of nested vectorization could be removed at a time. When faced with the expression: min(ramp(x8(a), x8(b), 5), x40(27)) It panicked, because on the left hand side it reduced the bounds to x8(a) ... x8(a) + x8(b) * 4, and on the right hand side it reduced the bounds to 27. It then attempted to take a min of mismatched types. In general we can't assume that binary operators on nested vectors have the same nesting structure on both sides, so I just rewrote it to reduce directly to a scalar. Fixes #8038
1 parent 4590a09
Lerp.h
#ifndef HALIDE_LERP_H
#define HALIDE_LERP_H
/** \file
* Defines methods for converting a lerp intrinsic into Halide IR.
*/
#include "Expr.h"
namespace Halide {
struct Target;
namespace Internal {
/** Build Halide IR that computes a lerp. Use by codegen targets that don't have
* a native lerp. The lerp is done in the type of the zero value. The final_type
* is a cast that should occur after the lerp. It's included because in some
* cases you can incorporate a final cast into the lerp math. */
Expr lower_lerp(Type final_type, Expr zero_val, Expr one_val, const Expr &weight, const Target &target);
} // namespace Internal
} // namespace Halide
#endif
Computing file changes ...
