Re: [PATCH] iommu/vt-d: Flush old iotlb for kdump when the device gets context mapped
From: Xunlei Pang
Date: Wed Nov 30 2016 - 21:14:39 EST
On 11/30/2016 at 10:26 PM, Joerg Roedel wrote:
> On Wed, Nov 30, 2016 at 06:23:34PM +0800, Baoquan He wrote:
>> OK, talked with Xunlei. The old cache could be entry with present bit
>> set.
> -EPARSE
>
> Anyway, what I was trying to say is, that the IOMMU TLB is tagged with
> domain-ids, and that there is also a context-cache which maps device-ids
> to domain-ids.
>
> If we update the context entry then we need to flush only the context
> entry, as it will point to a new domain-id then and future IOTLB lookups
> in the IOMMU will be using the new domain-id and do not match the old
> entries.
Hi Joerg,
Thanks for the explanation, and we still need to flush context cache using old domain-id, right?
How about the following update?
index 3965e73..624eac9 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -2024,6 +2024,25 @@ static int domain_context_mapping_one(struct dmar_domain *domain,
if (context_present(context))
goto out_unlock;
+ /*
+ * For kdump cases, old valid entries may be cached due to the
+ * in-flight DMA and copied pgtable, but there is no unmapping
+ * behaviour for them, thus we need an explicit cache flush for
+ * the newly-mapped device. For kdump, at this point, the device
+ * is supposed to finish reset at its driver probe stage, so no
+ * in-flight DMA will exist, and we don't need to worry anymore
+ * hereafter.
+ */
+ if (context_copied(context)) {
+ u16 did_old = context_domain_id(context);
+
+ if (did_old >= 0 && did_old < cap_ndoms(iommu->cap))
+ iommu->flush.flush_context(iommu, did_old,
+ (((u16)bus) << 8) | devfn,
+ DMA_CCMD_MASK_NOBIT,
+ DMA_CCMD_DEVICE_INVL);
+ }
+
pgd = domain->pgd;