The cost of stack traces

Wed May 10 17:59:00 GMT 2006

On Wed, 2006年05月10日 at 14:47 -0400, Andrew Haley wrote:
> As some of you may be aware, I've been doing benchmarking of servers
> with a mind to improving gcj performance.

Is it worth considering caching stack trace results? I'm not sure
exactly how it would work, but I imagine there must be some way to take
advantage of the fact that many stack traces will share a common bottom
portion of the stack, so we would only have to trace the unknown
portion.
AG