[ODE] From unstable to trunk...

[Adam D. Moss]

Peter Kyme wrote:
> Rodrigo Hernandez wrote:
>> Tanguy Fautre wrote:
>>> IIRC It's a 'define' Rodrigez added and that is not defined in the 
>>> old Makefile. Hence compiler errors on this 'undefined define'.
>>>
>>> ode/src/collision_trimesh_box.cpp:1270: error: `dEpsilon' undeclared 
>>> (first use this function)
>>
>> Yes, that was a quick bugfix someone submited, it's on the archives I 
>> am sure, in theory, all it would take would be to add a check on 
>> configurator.c similar to the one for dInfinity
>>
> 
> Here's a patch for the above.

Great, thanks, I've put this on the trunk, though of course
nothing there uses dEpsilon yet.

I've also applied several patches from the last week or
so that I said I'd apply if there were no objections (because
there were no objections):

* more reliable gcc opcode workaround trigger
* always use GetContactStatus() in trimesh<->*
* a quickstep optimization
* all-int dRandInt()

--adam
-- 
Adam D. Moss   -   adam at gimp.org