netdev - [PATCH v2 0/2] vhost: fix ranges when call vhost_iotlb_itree

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:   Wed,  9 Nov 2022 11:25:01 +0100
From:   Stefano Garzarella <sgarzare@...hat.com>
To:     virtualization@...ts.linux-foundation.org
Cc:     kvm@...r.kernel.org, "Michael S. Tsirkin" <mst@...hat.com>,
        Jason Wang <jasowang@...hat.com>, linux-kernel@...r.kernel.org,
        netdev@...r.kernel.org, Stefano Garzarella <sgarzare@...hat.com>
Subject: [PATCH v2 0/2] vhost: fix ranges when call vhost_iotlb_itree_first()

v2:
- Patch 2: Replaced Fixes tag with the right one [Jason]

v1: https://lore.kernel.org/virtualization/20221108103437.105327-1-sgarzare@redhat.com/

While I was working on vringh to support VA in vringh_*_iotlb()
I saw that the range we use in iotlb_translate() when we call
vhost_iotlb_itree_first() was not correct IIUC.
So I looked at all the calls and found that in vhost.c as well.

I didn't observe a failure and I don't have a reproducer because
I noticed the problem by looking at the code.

Maybe we didn't have a problem, because a shorter range was being
returned anyway and the loop stopped taking into account the total
amount of bytes translated, but I think it's better to fix.

Thanks,
Stefano

Stefano Garzarella (2):
  vringh: fix range used in iotlb_translate()
  vhost: fix range used in translate_desc()

 drivers/vhost/vhost.c  | 4 ++--
 drivers/vhost/vringh.c | 5 ++---
 2 files changed, 4 insertions(+), 5 deletions(-)

-- 
2.38.1