Re: copy_from_user() fix

H. Peter Anvin (hpa@transmeta.com)
Mon, 24 Aug 1998 15:29:08 -0700 (PDT)

> > You propose to slow down every syscall return by test
> > if (retcode == -EFAULT) raise(SIGSEGV).
> >
> > My argument is that if you do it from within kernel there's no such
> > slowdown.
>
> But there is no normal demand for this fairly weird feature to justify
> bloating the generic kernel with that
>

Actually, I'd argue that EFAULT is a fundamentally broken thing. The
reason is that the difference between library functions and system
calls really should be invisible to the user, and EFAULT makes it
visible.

This is bad. This is VERY bad. This means a slight change in the
partitioning in libc can break programs.

-hpa

-
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.altern.org/andrebalsa/doc/lkml-faq.html