All of lore.kernel.org
 help / color / mirror / Atom feed
From: Stefan Hajnoczi <stefanha@gmail.com>
To: Hannes Reinecke <hare@suse.de>
Cc: qemu-devel@nongnu.org, Paolo Bonzini <pbonzini@redhat.com>,
	Stefan Haynoczi <stefanha@linux.vnet.ibm.com>,
	kvm@vger.kernel.org, Alexander Graf <agraf@suse.de>
Subject: Re: [Qemu-devel] [PATCH 5/5] megasas: LSI Megaraid SAS emulation
Date: Tue, 5 Jul 2011 16:21:41 +0100	[thread overview]
Message-ID: <CAJSP0QVDh1+1Z2Z5h4d+7MsYTN17gNmuHvjONnmD1AzD=n4hvg@mail.gmail.com> (raw)
In-Reply-To: <1309863815-28236-6-git-send-email-hare@suse.de>

On Tue, Jul 5, 2011 at 12:03 PM, Hannes Reinecke <hare@suse.de> wrote:
> +static void megasas_unmap_sgl(struct megasas_cmd_t *cmd)
> +{
> +    uint16_t flags = le16_to_cpu(cmd->frame->header.flags);
> +    int i, is_write = (flags & MFI_FRAME_DIR_WRITE) ? 1 : 0;
> +
> +    for (i = 0; i < cmd->frame->header.sge_count; i++) {
> +        cpu_physical_memory_unmap(cmd->iov[i].iov_base, cmd->iov[i].iov_len,
> +                                  is_write, cmd->iov[i].iov_len);
> +    }

We cannot map control structures from guest memory and treating them
as valid request state later on.

A malicious guest can issue the request, then change the fields the
control structure while QEMU is processing the I/O, and then this
function will execute with is_write/sge_count no longer the same as
when the request started.

Good practice would be to copy in any request state needed instead of
reaching into guest memory at later points of the request lifecycle.
This way a malicious guest can never cause QEMU to crash or do
something due to inconsistent state.

The particular problem I see here is starting the request with
sge_count=1 and then setting it to sge_count=255.  We will perform
invalid iov[] accesses.

Stefan

WARNING: multiple messages have this Message-ID (diff)
From: Stefan Hajnoczi <stefanha@gmail.com>
To: Hannes Reinecke <hare@suse.de>
Cc: Paolo Bonzini <pbonzini@redhat.com>,
	Alexander Graf <agraf@suse.de>,
	qemu-devel@nongnu.org, kvm@vger.kernel.org,
	Stefan Haynoczi <stefanha@linux.vnet.ibm.com>
Subject: Re: [Qemu-devel] [PATCH 5/5] megasas: LSI Megaraid SAS emulation
Date: Tue, 5 Jul 2011 16:21:41 +0100	[thread overview]
Message-ID: <CAJSP0QVDh1+1Z2Z5h4d+7MsYTN17gNmuHvjONnmD1AzD=n4hvg@mail.gmail.com> (raw)
In-Reply-To: <1309863815-28236-6-git-send-email-hare@suse.de>

On Tue, Jul 5, 2011 at 12:03 PM, Hannes Reinecke <hare@suse.de> wrote:
> +static void megasas_unmap_sgl(struct megasas_cmd_t *cmd)
> +{
> +    uint16_t flags = le16_to_cpu(cmd->frame->header.flags);
> +    int i, is_write = (flags & MFI_FRAME_DIR_WRITE) ? 1 : 0;
> +
> +    for (i = 0; i < cmd->frame->header.sge_count; i++) {
> +        cpu_physical_memory_unmap(cmd->iov[i].iov_base, cmd->iov[i].iov_len,
> +                                  is_write, cmd->iov[i].iov_len);
> +    }

We cannot map control structures from guest memory and treating them
as valid request state later on.

A malicious guest can issue the request, then change the fields the
control structure while QEMU is processing the I/O, and then this
function will execute with is_write/sge_count no longer the same as
when the request started.

Good practice would be to copy in any request state needed instead of
reaching into guest memory at later points of the request lifecycle.
This way a malicious guest can never cause QEMU to crash or do
something due to inconsistent state.

The particular problem I see here is starting the request with
sge_count=1 and then setting it to sge_count=255.  We will perform
invalid iov[] accesses.

Stefan

  parent reply	other threads:[~2011-07-05 15:21 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-07-05 11:03 [PATCH 0/5][v6] Megasas HBA emulation Hannes Reinecke
2011-07-05 11:03 ` [Qemu-devel] " Hannes Reinecke
2011-07-05 11:03 ` [PATCH 1/5] iov: Update parameter usage in iov_(to|from)_buf() Hannes Reinecke
2011-07-05 11:03   ` [Qemu-devel] " Hannes Reinecke
2011-07-05 11:03   ` [PATCH 2/5] scsi: Add 'hba_private' to SCSIRequest Hannes Reinecke
2011-07-05 11:03     ` [Qemu-devel] " Hannes Reinecke
2011-07-05 11:03     ` [PATCH 3/5] scsi-disk: Fixup debugging statement Hannes Reinecke
2011-07-05 11:03       ` [Qemu-devel] " Hannes Reinecke
2011-07-05 11:03       ` [PATCH 4/5] scsi-disk: Mask out serial number EVPD Hannes Reinecke
2011-07-05 11:03         ` [Qemu-devel] " Hannes Reinecke
2011-07-05 11:03         ` [PATCH 5/5] megasas: LSI Megaraid SAS emulation Hannes Reinecke
2011-07-05 11:03           ` [Qemu-devel] " Hannes Reinecke
2011-07-05 13:06           ` Alexander Graf
2011-07-05 13:06             ` [Qemu-devel] " Alexander Graf
2011-07-05 13:38           ` Alexander Graf
2011-07-05 13:38             ` [Qemu-devel] " Alexander Graf
2011-07-05 13:59             ` Paolo Bonzini
2011-07-05 13:59               ` [Qemu-devel] " Paolo Bonzini
2011-07-05 14:05               ` Alexander Graf
2011-07-05 14:05                 ` [Qemu-devel] " Alexander Graf
2011-07-05 15:21           ` Stefan Hajnoczi [this message]
2011-07-05 15:21             ` Stefan Hajnoczi
2011-07-06  6:20             ` Hannes Reinecke
2011-07-06  6:20               ` Hannes Reinecke
2011-07-06  8:39               ` Paolo Bonzini
2011-07-06  8:39                 ` Paolo Bonzini
2011-07-05 11:06         ` [PATCH 4/5] scsi-disk: Mask out serial number EVPD Paolo Bonzini
2011-07-05 11:06           ` [Qemu-devel] " Paolo Bonzini
2011-07-05 11:05       ` [PATCH 3/5] scsi-disk: Fixup debugging statement Paolo Bonzini
2011-07-05 11:05         ` [Qemu-devel] " Paolo Bonzini
2011-07-05 13:01 ` [PATCH 0/5][v6] Megasas HBA emulation Alexander Graf
2011-07-05 13:01   ` [Qemu-devel] " Alexander Graf

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='CAJSP0QVDh1+1Z2Z5h4d+7MsYTN17gNmuHvjONnmD1AzD=n4hvg@mail.gmail.com' \
    --to=stefanha@gmail.com \
    --cc=agraf@suse.de \
    --cc=hare@suse.de \
    --cc=kvm@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@linux.vnet.ibm.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.