Misaki/Misaki.HighPerformance
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(numeric): add scalar conversions and quaternion fix

Browse Source 
Added [NumericConvertable] attributes for scalar type conversions to all vector structs. Fixed bitwise mask calculation in quaternion.cs for improved correctness. Updated Program.cs with quaternion test code and bumped project version to 1.3.2.
This commit is contained in:
Misaki
2026-04-07 20:09:37 +09:00
parent 9824c1ed19
commit 81eb5cb4cf
7 changed files with 28 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
Misaki.HighPerformance.Mathematics/quaternion.cs
View File

@@ -57,14 +57,14 @@ public partial struct quaternion : IEquatable<quaternion>
var u_sign = (asuint(u.x) & 0x80000000);
var t = v.y + asfloat(asuint(w.z) ^ u_sign);
var u_mask = new uint4(u_sign >> 31);
var t_mask = new uint4(asuint(t) >> 31);
var u_mask = new uint4((int)u_sign >> 31);
var t_mask = new uint4(asint(t) >> 31);
var tr = 1.0f + abs(u.x);
float tr = 1.0f + abs(u.x);
var sign_flips = new uint4(0x00000000, 0x80000000, 0x80000000, 0x80000000) ^ (u_mask & new uint4(0x00000000, 0x80000000, 0x00000000, 0x80000000)) ^ (t_mask & new uint4(0x80000000, 0x80000000, 0x80000000, 0x00000000));
value = new float4(tr, u.y, w.x, v.z) + asfloat(asuint(new float4(t, v.x, u.z, w.y)) ^ sign_flips); // +---, +++-, ++-+, +-++
value = float4(tr, u.y, w.x, v.z) + asfloat(asuint(float4(t, v.x, u.z, w.y)) ^ sign_flips); // +---, +++-, ++-+, +-++
value = asfloat((asuint(value) & ~u_mask) | (asuint(value.zwxy) & u_mask));
value = asfloat((asuint(value.wzyx) & ~t_mask) | (asuint(value) & t_mask));
@@ -81,8 +81,8 @@ public partial struct quaternion : IEquatable<quaternion>
var u_sign = (asuint(u.x) & 0x80000000);
var t = v.y + asfloat(asuint(w.z) ^ u_sign);
var u_mask = new uint4(u_sign >> 31);
var t_mask = new uint4(asuint(t) >> 31);
var u_mask = new uint4((int)u_sign >> 31);
var t_mask = new uint4(asint(t) >> 31);
var tr = 1.0f + abs(u.x);