[Linux-ia64] race conditions in ia32 subsystem

[Linux-ia64] race conditions in ia32 subsystem

[David Mosberger]

A while we discussed the race conditions that are present in the ia32
subsystem when the page size is greater than 4KB.  It occurred to me
that these problems can be fixed easily and cleanly by using the
get_user_pages() routine (this routine was introduced in 2.4.17, if
memory serves).

The basic idea would be to use the address space's mmap_sem instead of
the global ia32_mmap_sem and to use get_user_pages() instead of
copy_from_user()/copy_to_user() to access memory.

Does someone want to take a stab at doing this?

	--david

Re: [Linux-ia64] race conditions in ia32 subsystem

[n0ano]

David-

I'm up for taking a stab at this.  It might be a week or so, I'm pretty
busy right now, but I'll get to it.

On Fri, Jan 18, 2002 at 10:14:08AM -0800, David Mosberger wrote:
> A while we discussed the race conditions that are present in the ia32
> subsystem when the page size is greater than 4KB.  It occurred to me
> that these problems can be fixed easily and cleanly by using the
> get_user_pages() routine (this routine was introduced in 2.4.17, if
> memory serves).
> 
> The basic idea would be to use the address space's mmap_sem instead of
> the global ia32_mmap_sem and to use get_user_pages() instead of
> copy_from_user()/copy_to_user() to access memory.
> 
> Does someone want to take a stab at doing this?
> 
> 	--david
> 
> _______________________________________________
> Linux-IA64 mailing list
> Linux-IA64@linuxia64.org
> http://lists.linuxia64.org/lists/listinfo/linux-ia64

-- 
Don Dugger
"Censeo Toto nos in Kansa esse decisse." - D. Gale
n0ano@indstorage.com
Ph: 303/652-0870x117