[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH V1 10/16] xen/mm: Handle properly reference in set_foreign_p2m_entry() on Arm
- To: Jan Beulich <jbeulich@xxxxxxxx>
- From: Julien Grall <julien@xxxxxxx>
- Date: Wed, 16 Sep 2020 09:55:20 +0100
- Cc: Oleksandr Tyshchenko <olekstysh@xxxxxxxxx>, xen-devel@xxxxxxxxxxxxxxxxxxxx, Oleksandr Tyshchenko <oleksandr_tyshchenko@xxxxxxxx>, Stefano Stabellini <sstabellini@xxxxxxxxxx>, Volodymyr Babchuk <Volodymyr_Babchuk@xxxxxxxx>, Andrew Cooper <andrew.cooper3@xxxxxxxxxx>, George Dunlap <george.dunlap@xxxxxxxxxx>, Ian Jackson <ian.jackson@xxxxxxxxxxxxx>, Wei Liu <wl@xxxxxxx>, Roger Pau Monné <roger.pau@xxxxxxxxxx>, Julien Grall <julien.grall@xxxxxxx>
- Delivery-date: Wed, 16 Sep 2020 08:55:35 +0000
- List-id: Xen developer discussion <xen-devel.lists.xenproject.org>
On 16/09/2020 09:52, Jan Beulich wrote:
On 16.09.2020 10:50, Julien Grall wrote:
On 16/09/2020 08:17, Jan Beulich wrote:
On 10.09.2020 22:22, Oleksandr Tyshchenko wrote:
for ( i = 0; !rc && i < xmar.nr_frames; i++ )
{
- rc = set_foreign_p2m_entry(currd, gfn_list[i],
+ rc = set_foreign_p2m_entry(currd, d, gfn_list[i],
_mfn(mfn_list[i]));
Is it going to lead to proper behavior when d == currd, specifically
for Arm but also in the abstract model? If you expose this to other
than Dom0, this case needs at least considering (and hence mentioning
in the description of why it's safe to permit if you don't reject
such attempts).
This is already rejected by rcu_lock_remote_domain_by_id().
Oh, yes, I'm sorry for overlooking this.
That's fine, I also overlooked it when I originally wrote the code.
@oleksandr, it might be worth mentioning in the commit message and maybe
in the code this subtlety.
Cheers,
--
Julien Grall
|
