On 07.09.2020 09:40, Paul Durrant wrote:
> From: Paul Durrant <pdurr...@amazon.com>
>
> This patch adds a full I/O TLB flush to the error paths of iommu_map() and
> iommu_unmap().
>
> Without this change callers need constructs such as:
>
> rc = iommu_map/unmap(...)
> err = iommu_flush(...)
> if ( !rc )
> rc = err;
>
> With this change, it can be simplified to:
>
> rc = iommu_map/unmap(...)
> if ( !rc )
> rc = iommu_flush(...)
>
> because, if the map or unmap fails the flush will be unnecessary. This saves
> a stack variable and generally makes the call sites tidier.
>
> Signed-off-by: Paul Durrant <pdurr...@amazon.com>
Reviewed-by: Jan Beulich <jbeul...@suse.com>
with one cosmetic issue taken care of (perhaps while committing):
> @@ -338,14 +346,8 @@ int iommu_legacy_unmap(struct domain *d, dfn_t dfn,
> unsigned int page_order)
> unsigned int flush_flags = 0;
> int rc = iommu_unmap(d, dfn, page_order, &flush_flags);
>
> - if ( !this_cpu(iommu_dont_flush_iotlb) )
> - {
> - int err = iommu_iotlb_flush(d, dfn, (1u << page_order),
> - flush_flags);
> -
> - if ( !rc )
> - rc = err;
> - }
> + if ( !this_cpu(iommu_dont_flush_iotlb) && ! rc )
There's a stray blank after the latter ! here.
Jan
