| Age | Commit message (Collapse) | Author |
|---|
|
when fasttimers are amazingly consistent, we remove all data as outliers
instead of removing nothing. Probably something else going on here,
but this should fix the crash so we can analyze better.
|
|
|
|
|
|
reviewed by davep.
|
|
|
|
Updated header blocks for llphysicsshapebuilderutil classes.
|
|
|
|
The reason was to manually resolve a couple of merge problems, as upstream has had several changes since this was applied, and some of the changes overlapped.
|
|
|
|
|
|
|
|
TODO) around some extremely obscure code that needs further attention but which is outside this scope.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
where sensible.
Not all instances of dist_vec() were squared, only those where it wouldn't (hopefully) change the functionality.
|
|
|
|
|
|
|
|
|
|
data from gDEBugger.
Reviewed by Nyx.
|
|
|
|
|
|
silhouette segments pointing at center of screen.
|
|
|
|
|
|
|
|
Backed out SH-659 since merge was messy; this will be merged in manually later.
|
|
|
|
spinner and the manipulation tools etc.
|
|
|
|
|
|
Viewer attempting to load precached images in file types that are not being used.)
|
|
|
|
platform-dependent ways.
|
|
dependencies or architecture on the build machines? Trying updating expected float values to see if it begins to pass.
Updated expected values to match result of query on WolframAlpha.com (mathematica):
N[Orthogonalize[{{1,4,3},{1,2,0},{2,4,2}}],8]
|
|
|
|
|
|
|
|
|
|
|
|
|
