> People are always quick to state as fact things that they have
> not looked into. I don't know what the current status is on
> this, but as I understand it, trampolines work with these
> patches.
Well, sort of. You get a trap on each trampoline call and the kernel
emulates the trampoline call. This sounds slower than actually is---an
emulated trampoline call is only about twice as slow as a direct
trampoline call (if called in a very tight loop). The influence on a
large Ada program (the GNAT compiler) was hardly observable.
The bad thing is, however, that Solar Designer's patch doesn't recognize
all trampolines. For example, those created by GCC 2.95.2 aren't
handled properly.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/