chenlin/Mujoco_WASM
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Improve mju_cross implementation.

Browse Source 
- Allow in-place cross product a = a x b using `mju_cross(a, a, b)`.
- Easier for compiler to optimize since tmp cannot alias a or b (verified with godbolt).

PiperOrigin-RevId: 547265908
Change-Id: If965867e08c1bc980e71729e38f6855e32de18b2
This commit is contained in:
Yuval Tassa
2023-07-11 12:34:10 -07:00
committed by Copybara-Service
parent e2677d6057
commit 77b3533d90
1 changed files with 8 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/engine/engine_util_spatial.c
+8 -3
View File
@@ -323,9 +323,14 @@ void mju_trnVecPose(mjtNum res[3], const mjtNum pos[3], const mjtNum quat[4], co
// vector cross-product, 3D
void mju_cross(mjtNum res[3], const mjtNum a[3], const mjtNum b[3]) {
res[0] = a[1]*b[2] - a[2]*b[1];
res[1] = a[2]*b[0] - a[0]*b[2];
res[2] = a[0]*b[1] - a[1]*b[0];
mjtNum tmp[3] = {
a[1]*b[2] - a[2]*b[1],
a[2]*b[0] - a[0]*b[2],
a[0]*b[1] - a[1]*b[0]
};
res[0] = tmp[0];
res[1] = tmp[1];
res[2] = tmp[2];
}