tint/writer/glsl: Set default precision to highp

WGSL states that `f32` is a IEEE-754 binary32. mediump precision guarantees are much smaller than highp, and using mediump can break code generated by tint.

Change-Id: I8a129c4c8ac078c785e0d5d7d21f19ed6b3620b5
Reviewed-on: https://dawn-review.googlesource.com/c/dawn/+/123400
Kokoro: Kokoro <noreply+kokoro@google.com>
Commit-Queue: Ben Clayton <bclayton@google.com>
Reviewed-by: Stephen White <senorblanco@chromium.org>
This commit is contained in:
Ben Clayton
2023-03-09 19:58:23 +00:00
committed by Dawn LUCI CQ
parent 4fea9d0fbb
commit f1f56b2d3c
2871 changed files with 2908 additions and 2908 deletions

View File

@@ -1,6 +1,6 @@
#version 310 es
#extension GL_AMD_gpu_shader_half_float : require
precision mediump float;
precision highp float;
layout(location = 0) out vec4 value;
vec4 tint_symbol() {

View File

@@ -1,6 +1,6 @@
#version 310 es
#extension GL_AMD_gpu_shader_half_float : require
precision mediump float;
precision highp float;
layout(location = 0) out vec4 value;
vec4 tint_symbol() {