> From: Jan Beulich [mailto:jbeul...@suse.com]
> Sent: Wednesday, September 4, 2019 9:28 PM
>
> The two uses of pci_get_pdev_by_domain() lack proper locking, but are
> also only used to get hold of a NUMA node ID. Calculate and store the
> node ID earlier on and remove the lookups (in lieu of fixi
On 04.09.2019 17:45, Andrew Cooper wrote:
> On 04/09/2019 14:27, Jan Beulich wrote:
>> --- a/xen/drivers/passthrough/vtd/iommu.h
>> +++ b/xen/drivers/passthrough/vtd/iommu.h
>> @@ -530,6 +530,7 @@ struct intel_iommu {
>> struct ir_ctrl ir_ctrl;
>> struct iommu_flush flush;
>> struct
On 04/09/2019 14:27, Jan Beulich wrote:
> --- a/xen/drivers/passthrough/vtd/iommu.h
> +++ b/xen/drivers/passthrough/vtd/iommu.h
> @@ -530,6 +530,7 @@ struct intel_iommu {
> struct ir_ctrl ir_ctrl;
> struct iommu_flush flush;
> struct acpi_drhd_unit *drhd;
> +nodeid_t node;
> };
The two uses of pci_get_pdev_by_domain() lack proper locking, but are
also only used to get hold of a NUMA node ID. Calculate and store the
node ID earlier on and remove the lookups (in lieu of fixing the
locking).
While doing this it became apparent that iommu_alloc()'s use of
alloc_pgtable_maddr
