x86/mm: further simplify cleanup_page_mappings()
PATCH 1/1 x86/mm: further simplify cleanup_page_mappings()
With the removal of update_xen_mappings() there's no need anymore for a 2nd error code variable to transiently hold the IOMMU unmap return value.
Signed-off-by Jan Beulich <[email protected]>
---
I have to admit that I was tempted to get rid of PAGE_ORDER_4K at this occasion, as it feels awkward to me to have such in clearly x86-only code.
xen/arch/x86/mm.c +1 -6
@@ -2470,12 +2470,7 @@ static int cleanup_page_mappings(structstruct domain *d =page_get_owner (page);if ( d &&unlikely (need_iommu_pt_sync (d)) &&is_pv_domain (d) ) - { -int rc2 =iommu_legacy_unmap (d,_dfn (mfn),1u <<PAGE_ORDER_4K ); - -if ( !rc ) - rc = rc2; - } + rc =iommu_legacy_unmap (d,_dfn (mfn),1u <<PAGE_ORDER_4K );if (likely (!is_special_page (page)) ) {
Quoted 5 lines · Jan Beulich
With the removal of
update_xen_mappings()there's no need anymore for a 2nd error code variable to transiently hold the IOMMU unmap return value.Signed-off-by Jan Beulich <[email protected]>
Oh - I'd not even spotted that simplification.
Acked-by Andrew Cooper <[email protected]>
Quoted 4 lines
---
I have to admit that I was tempted to get rid ofPAGE_ORDER_4Kat this occasion, as it feels awkward to me to have such in clearly x86-only code.
Happy for that to go too.
~Andrew