On 04/07/2014 10:53 PM, Alexey Kardashevskiy wrote:
> Current guest kernels try allocating as many vectors as the quota is.
> For example, in the case of virtio-net (which has just 3 vectors)
> the guest requests 4 vectors (that is the quota in the test) and
> the existing ibm,change-msi handler returns 4. But before it returns,
> it calls msix_set_message() in a loop and corrupts memory behind
> the end of msix_table.
>
> This limits the number of vectors returned by ibm,change-msi to
> the maximum supported by the actual device.
>
> Signed-off-by: Alexey Kardashevskiy <a...@ozlabs.ru>
> ---
>
>
> This might go to 2.0 actually.
Ping? Ok, not 2.0 but ppc-next-2.1 or ppc-next? :)
>
> ---
> hw/ppc/spapr_pci.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c
> index cbef095..cdfa369 100644
> --- a/hw/ppc/spapr_pci.c
> +++ b/hw/ppc/spapr_pci.c
> @@ -343,6 +343,9 @@ static void rtas_ibm_change_msi(PowerPCCPU *cpu,
> sPAPREnvironment *spapr,
>
> /* There is no cached config, allocate MSIs */
> if (!phb->msi_table[ndev].nvec) {
> + if (req_num > pdev->msix_entries_nr) {
> + req_num = pdev->msix_entries_nr;
> + }
> irq = spapr_allocate_irq_block(req_num, false,
> ret_intr_type == RTAS_TYPE_MSI);
> if (irq < 0) {
>
--
Alexey
