Re: [PATCH][COMPAT] compat_sys_fcntl{,64} 1/9 Generic part

[Andi Kleen <ak@muc.de>]

On Fri, Feb 28, 2003 at 05:33:49AM +0100, Stephen Rothwell wrote:
> +static int get_compat_flock(struct flock *kfl, struct compat_flock *ufl)
> +{
> +	if (!access_ok(VERIFY_READ, ufl, sizeof(*ufl)) ||
> +	    __get_user(kfl->l_type, &ufl->l_type) ||
> +	    __get_user(kfl->l_whence, &ufl->l_whence) ||
> +	    __get_user(kfl->l_start, &ufl->l_start) ||
> +	    __get_user(kfl->l_len, &ufl->l_len) ||
> +	    __get_user(kfl->l_pid, &ufl->l_pid))

Perhaps there should be really a big fat comment on top of compat.c
that it depends on a hole on __PAGE_OFFSET if the arch allows passing
64bit pointers to the compat functions.

> +
> +asmlinkage long compat_sys_fcntl(unsigned int fd, unsigned int cmd,
> +		unsigned long arg)
> +{
> +	if ((cmd == F_GETLK64) || (cmd == F_SETLK64) || (cmd == F_SETLKW64))
> +		return -EINVAL;

That won't work for IA32 emulation. There are programs that call
old style fcntl() with F_*LK64. Just drop the if here.

> +	return compat_sys_fcntl64(fd, cmd, arg);

-Andi