Alexandre Julliard wrote:
> Peter Ganten <[EMAIL PROTECTED]> writes:
>
> > At least this option works :-) I used the appendend patch to test it.
> > If nobody objects, I will submit it to wine-patches.
>
> I'd prefer that we remove the IsBadReadPtr hack in the PE loader,
> rather than adding another hack on top of it (simply commenting out
> the IsBadReadPtr checks should work fine, unless you want to run
> native USER).
Well, I guess we should implement shared PE sections, then the loader
hack would be superfluous :-/
But independent of this, I'd consider the behaviour of IsBadReadPtr
buggy: if the default rlimits are in place, the stack will be
automatically grown to *unlimited* size, and therefore a --in principle
perfectly legal-- call to IsBadReadPtr might permanently waste hundreds
of megabytes of virtual address space ...
But instead of the hack of using a dummy page, I guess we could simply
use the intended interface and call setrlimit(RLIMIT_STACK, ...) to
limit the automatic stack growth to a reasonable size. ;-)
Bye,
Ulrich
--
Dr. Ulrich Weigand
[EMAIL PROTECTED]
