From: James Bottomley <James.Bottomley@HansenPartnership.com>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: linux-kernel@vger.kernel.org, linux-scsi@vger.kernel.org,
Hu Tao <hutao@cn.fujitsu.com>
Subject: Re: [PATCH for 3.4] virtio-scsi: fix TMF use-after-free
Date: Wed, 18 Apr 2012 18:09:08 +0400 [thread overview]
Message-ID: <1334758148.4410.51.camel@dabdike> (raw)
In-Reply-To: <1334756761-12312-1-git-send-email-pbonzini@redhat.com>
On Wed, 2012-04-18 at 15:46 +0200, Paolo Bonzini wrote:
> Fix a race in TMF path, where cmd may have been already freed
> by virtscsi_complete_free after waking up from the completion.
There's no may about this; the command will be freed long before the
completion waiter is awoken. The description could be clearer.
The problem is a use after free in virtscsi_tmf because the
virtio_scsi_command is freed before the completion returns.
The fix is to make callers specifying a completion responsible for
freeing the command in all cases.
James
> Cc: James Bottomley <JBottomley@parallels.com>
> Cc: linux-scsi@vger.kernel.org
> Signed-off-by: Hu Tao <hutao@cn.fujitsu.com>
> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
> ---
> drivers/scsi/virtio_scsi.c | 24 +++++++++++++-----------
> 1 file changed, 13 insertions(+), 11 deletions(-)
>
> diff --git a/drivers/scsi/virtio_scsi.c b/drivers/scsi/virtio_scsi.c
> index efccd72..1b38431 100644
> --- a/drivers/scsi/virtio_scsi.c
> +++ b/drivers/scsi/virtio_scsi.c
> @@ -175,7 +175,8 @@ static void virtscsi_complete_free(void *buf)
>
> if (cmd->comp)
> complete_all(cmd->comp);
> - mempool_free(cmd, virtscsi_cmd_pool);
> + else
> + mempool_free(cmd, virtscsi_cmd_pool);
> }
>
> static void virtscsi_ctrl_done(struct virtqueue *vq)
> @@ -311,21 +312,22 @@ out:
> static int virtscsi_tmf(struct virtio_scsi *vscsi, struct virtio_scsi_cmd *cmd)
> {
> DECLARE_COMPLETION_ONSTACK(comp);
> - int ret;
> + int ret = FAILED;
>
> cmd->comp = ∁
> - ret = virtscsi_kick_cmd(vscsi, vscsi->ctrl_vq, cmd,
> - sizeof cmd->req.tmf, sizeof cmd->resp.tmf,
> - GFP_NOIO);
> - if (ret < 0)
> - return FAILED;
> + if (virtscsi_kick_cmd(vscsi, vscsi->ctrl_vq, cmd,
> + sizeof cmd->req.tmf, sizeof cmd->resp.tmf,
> + GFP_NOIO) < 0)
> + goto out;
>
> wait_for_completion(&comp);
> - if (cmd->resp.tmf.response != VIRTIO_SCSI_S_OK &&
> - cmd->resp.tmf.response != VIRTIO_SCSI_S_FUNCTION_SUCCEEDED)
> - return FAILED;
> + if (cmd->resp.tmf.response == VIRTIO_SCSI_S_OK ||
> + cmd->resp.tmf.response == VIRTIO_SCSI_S_FUNCTION_SUCCEEDED)
> + ret = SUCCESS;
>
> - return SUCCESS;
> +out:
> + mempool_free(cmd, virtscsi_cmd_pool);
> + return ret;
> }
>
> static int virtscsi_device_reset(struct scsi_cmnd *sc)
next prev parent reply other threads:[~2012-04-18 14:09 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-04-18 13:46 [PATCH for 3.4] virtio-scsi: fix TMF use-after-free Paolo Bonzini
2012-04-18 14:01 ` Zhi Yong Wu
2012-04-18 14:09 ` James Bottomley [this message]
2012-05-01 8:09 ` Paolo Bonzini
2012-05-01 8:52 ` James Bottomley
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1334758148.4410.51.camel@dabdike \
--to=james.bottomley@hansenpartnership.com \
--cc=hutao@cn.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=pbonzini@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).