From: John Levon <levon@mo...> - 2004-11-01 17:31:27
On Mon, Nov 01, 2004 at 08:41:32PM +1100, Greg Banks wrote:
> Introduce check_user_page_readable() as to check if a user page is
> present and readable, and use it instead of follow_page() in the i386
> oprofile backtracing code. This fixes a bug where a sample which
> walked a user stack into a PROT_NONE page would oops.
I don't think we need to EXPORT_SYMBOL(follow_page); now. Otherwise,
On Mon, Nov 01, 2004 at 05:31:24PM +0000, John Levon wrote:
> On Mon, Nov 01, 2004 at 08:41:32PM +1100, Greg Banks wrote:
> > Introduce check_user_page_readable() [...]
> I don't think we need to EXPORT_SYMBOL(follow_page); now.
You're right, no potentially-modular code in the kernel needs it
anymore. I'll drop the EXPORT_SYMBOL().
Greg Banks, R&D Software Engineer, SGI Australian Software Group.
I don't speak for SGI.
Get latest updates about Open Source Projects, Conferences and News.