Thread (28 messages) 28 messages, 6 authors, 2022-11-09

RE: [PATCH 05/12] x86/hyperv: Change vTOM handling to use standard coco mechanisms

From: Michael Kelley (LINUX) <hidden>
Date: 2022-11-03 14:12:41
Also in: linux-arch, linux-hyperv, linux-iommu, linux-pci, lkml

From: Tianyu Lan <redacted> Sent: Thursday, November 3, 2022 7:01 AM
On 10/21/2022 1:57 AM, Michael Kelley wrote:
quoted
+
+void __init hv_vtom_init(void)
+{
+	cc_set_vendor(CC_VENDOR_HYPERV);
+	cc_set_mask(ms_hyperv.shared_gpa_boundary);
+	physical_mask &= ms_hyperv.shared_gpa_boundary - 1;

hv_vtom_init() also works for VBS case. VBS doesn't have vTOM support
and so shared_gpa_boundary should not be applied for VBS. Here seems to
assume ms_hyperv.shared_gpa_boundary to 0 for VBS, right?
Yes, that is correct.  In fact, you'll notice that this patch slightly changes
the code in ms_hyperv_init_platform() so that shared_gpa_boundary is
set only when it is known to be valid; otherwise it is left as zero.  This
change is specifically so that shared_gpa_boundary is zero for the VBS
case.

Michael
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help