Re: [IDEA+RFC] Possible solution for min()/max() war

[David Wagner]

Richard B. Johnson wrote:
>The min() macro is not really used for the mathematical min, anywhere
>I've found it in the kernel. It's used as a whatever_will_fit() macro
>where the writer wanted to prevent a buffer overflow. [...]
>
>I suggest we just leave the damn thing alone and fix any
>bugs found in the normal way, i.e., "If it ain't broke, don't fix it."

The problem is that this tends not to work out so well when it
comes to security.  Patch-and-pray gets pretty painful after a
while; proactive measures would seem to be called for, if security
is on the line.