[Bf-committers] Re: Bf-committers Digest, Vol 16, Issue 47

[Alexander Ewering]

On Fri, 25 Nov 2005, Jonathan Merritt wrote:

> very similar) where there would be no recourse but for Blender to quit.  In 
> these situations, a "malloc or die" function would be desirable, I think.

Also, I'm not sure anyone is willing to go through all *600* mallocs

intrr at time:~/bf-cvs/blender$ egrep -rI "MEM_malloc" *|wc -l 
599

and construct a sane fallback for each of them failing (thus investigating
all surrounding code, the caller stack, giving appropriate UI feedback) ;)

So I guess this is really a non-issue and the discussion can be closed.

What I *do* agree on is check for NULL on special, easily handlable cases
where huge chunks of memory are to be malloced (say for "Start
fluid simulation" or "Start Render" (shadow buffers, maybe)).

| alexander ewering              instinctive mediaworks
| ae[@]instinctive[.]de   http://www[.]instinctive[.]de