SHA256
1
0
forked from pool/xen
xen/531d8e34-x86-HVM-consolidate-passthrough-handling-in-epte_get_entry_emt.patch
Charles Arnold e46082b3ea - Upstream patches from Jan
530b27fd-x86-MCE-Fix-race-condition-in-mctelem_reserve.patch
  530b2880-Nested-VMX-update-nested-paging-mode-on-vmexit.patch
  530b28c5-x86-MSI-don-t-risk-division-by-zero.patch
  530c54c3-x86-mce-Reduce-boot-time-logspam.patch
  5310bac3-mm-ensure-useful-progress-in-decrease_reservation.patch
  5315a254-IOMMU-generalize-and-correct-softirq-processing.patch
  5315a3bb-x86-don-t-propagate-acpi_skip_timer_override-do-Dom0.patch
  5315a43a-x86-ACPI-also-print-address-space-for-PM1x-fields.patch
  531d8db1-x86-hvm-refine-the-judgment-on-IDENT_PT-for-EMT.patch
  531d8e09-x86-HVM-fix-memory-type-merging-in-epte_get_entry_emt.patch
  531d8e34-x86-HVM-consolidate-passthrough-handling-in-epte_get_entry_emt.patch
  531d8fd0-kexec-identify-which-cpu-the-kexec-image-is-being-executed-on.patch
  531dc0e2-xmalloc-handle-correctly-page-allocation-when-align-size.patch

- Add conversion tool for migrating xend/xm managed VMs to libvirt 
  xen2libvirt.py (Jim Fehlig)

OBS-URL: https://build.opensuse.org/package/show/Virtualization/xen?expand=0&rev=304
2014-03-13 23:46:35 +00:00

60 lines
2.2 KiB
Diff

# Commit 3089a6d82bdf3112ccb1dd074ce34a8cbdc4ccd8
# Date 2014-03-10 11:04:36 +0100
# Author Jan Beulich <jbeulich@suse.com>
# Committer Jan Beulich <jbeulich@suse.com>
x86/HVM: consolidate passthrough handling in epte_get_entry_emt()
It is inconsistent to depend on iommu_enabled alone: For a guest
without devices passed through to it, it is of no concern whether the
IOMMU is enabled.
There's one rather special case to take care of: VMX code marks the
LAPIC access page as MMIO. The added assertion needs to take this into
consideration, and the subsequent handling of the direct MMIO case was
inconsistent too: That page would have been WB in the absence of an
IOMMU, but UC in the presence of it, while in fact the cachabilty of
this page is entirely unrelated to an IOMMU being in use.
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: "Xu, Dongxiao" <dongxiao.xu@intel.com>
Acked-by: Keir Fraser <keir@xen.org>
--- a/xen/arch/x86/hvm/mtrr.c
+++ b/xen/arch/x86/hvm/mtrr.c
@@ -698,14 +698,20 @@ uint8_t epte_get_entry_emt(struct domain
if ( hvm_get_mem_pinned_cacheattr(d, gfn, &type) )
return type;
- if ( !iommu_enabled )
+ if ( !iommu_enabled ||
+ (rangeset_is_empty(d->iomem_caps) &&
+ rangeset_is_empty(d->arch.ioport_caps) &&
+ !has_arch_pdevs(d)) )
{
+ ASSERT(!direct_mmio ||
+ mfn_x(mfn) == d->arch.hvm_domain.vmx.apic_access_mfn);
*ipat = 1;
return MTRR_TYPE_WRBACK;
}
if ( direct_mmio )
- return MTRR_TYPE_UNCACHABLE;
+ return mfn_x(mfn) != d->arch.hvm_domain.vmx.apic_access_mfn
+ ? MTRR_TYPE_UNCACHABLE : MTRR_TYPE_WRBACK;
if ( iommu_snoop )
{
--- a/xen/arch/x86/hvm/vmx/vmx.c
+++ b/xen/arch/x86/hvm/vmx/vmx.c
@@ -2038,9 +2038,9 @@ static int vmx_alloc_vlapic_mapping(stru
if ( apic_va == NULL )
return -ENOMEM;
share_xen_page_with_guest(virt_to_page(apic_va), d, XENSHARE_writable);
+ d->arch.hvm_domain.vmx.apic_access_mfn = virt_to_mfn(apic_va);
set_mmio_p2m_entry(d, paddr_to_pfn(APIC_DEFAULT_PHYS_BASE),
_mfn(virt_to_mfn(apic_va)));
- d->arch.hvm_domain.vmx.apic_access_mfn = virt_to_mfn(apic_va);
return 0;
}