Re: [PATCH v5 02/11] pcie_sriov: Validate NumVFs

[Akihiko Odaki]

On 2024/02/19 2:36, Michael S. Tsirkin wrote:

On Sun, Feb 18, 2024 at 01:56:07PM +0900, Akihiko Odaki wrote:

The guest may write NumVFs greater than TotalVFs and that can lead
to buffer overflow in VF implementations.

Cc: qemu-sta...@nongnu.org
Fixes: 7c0fa8dff811 ("pcie: Add support for Single Root I/O Virtualization 
(SR/IOV)")
Signed-off-by: Akihiko Odaki 
---
  hw/pci/pcie_sriov.c | 3 +++
  1 file changed, 3 insertions(+)

diff --git a/hw/pci/pcie_sriov.c b/hw/pci/pcie_sriov.c
index a1fe65f5d801..da209b7f47fd 100644
--- a/hw/pci/pcie_sriov.c
+++ b/hw/pci/pcie_sriov.c
@@ -176,6 +176,9 @@ static void register_vfs(PCIDevice *dev)
  
  assert(sriov_cap > 0);

  num_vfs = pci_get_word(dev->config + sriov_cap + PCI_SRIOV_NUM_VF);
+if (num_vfs > pci_get_word(dev->config + sriov_cap + PCI_SRIOV_TOTAL_VF)) {
+return;
+}
  
  dev->exp.sriov_pf.vf = g_new(PCIDevice *, num_vfs);



This reminds me: how is this num_vfs value set on migration?


That's a good point... Actually no consideration of migration is made 
and SR-IOV is completely broken with it.

Re: [PATCH v5 02/11] pcie_sriov: Validate NumVFs

[Michael S. Tsirkin]

On Sun, Feb 18, 2024 at 01:56:07PM +0900, Akihiko Odaki wrote:
> The guest may write NumVFs greater than TotalVFs and that can lead
> to buffer overflow in VF implementations.
> 
> Cc: qemu-sta...@nongnu.org
> Fixes: 7c0fa8dff811 ("pcie: Add support for Single Root I/O Virtualization 
> (SR/IOV)")
> Signed-off-by: Akihiko Odaki 
> ---
>  hw/pci/pcie_sriov.c | 3 +++
>  1 file changed, 3 insertions(+)
> 
> diff --git a/hw/pci/pcie_sriov.c b/hw/pci/pcie_sriov.c
> index a1fe65f5d801..da209b7f47fd 100644
> --- a/hw/pci/pcie_sriov.c
> +++ b/hw/pci/pcie_sriov.c
> @@ -176,6 +176,9 @@ static void register_vfs(PCIDevice *dev)
>  
>  assert(sriov_cap > 0);
>  num_vfs = pci_get_word(dev->config + sriov_cap + PCI_SRIOV_NUM_VF);
> +if (num_vfs > pci_get_word(dev->config + sriov_cap + 
> PCI_SRIOV_TOTAL_VF)) {
> +return;
> +}
>  
>  dev->exp.sriov_pf.vf = g_new(PCIDevice *, num_vfs);


This reminds me: how is this num_vfs value set on migration?


> 
> -- 
> 2.43.1