Any update on this? I've worked around it (painfully) for now, but would like to see a fix. If someone could point me in the right direction I could take a look at it myself, but don't really have a lot of time to spend on this at the moment (trying to get product out the door).
The not-so painful workaround for me was to declare the function pointer static. When it's not automatic, it's not on the stack, and not addressed register-indexed. As for fixing it, for someone who's not familiar with the gcc backend (machine definition and that stuff) it's probably quite a bit of learning first. My shallow knowledge of it made me change my mind about trying to fix it after half an hour. Heiko
