KVM: SEV: Do cache maintenance on the source VM during intra-host migrationgrafted

Manually perform cache maintenance on the source VM during intra-host
migration to ensure no stale data is left in CPU caches after the VM is
destroyed.  Because the source VM is "converted" to a non-SEV VM, KVM's
memory reclaim flows won't trigger cache maintenance, e.g. when all guest
memory is reclaimed in response to detaching from the mmu_notifier.

Note, relying on the destination VM to do cache maintenance isn't an option
as KVM doesn't require identical guest memory configurations, i.e. the
source VM may have access to memory that the destination VM does not.
Enforcing equivalent memory configurations is infeasible, as it would
require a *deep* comparison of memslots, e.g. to verify that not only are
the memslot identical, but what the memslots point at is also identical.

Fixes: b56639318bb2 ("KVM: SEV: Add support for SEV intra host migration")
Cc: stable@vger.kernel.org
Reported-by: Stefan Teodorescu <fane@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-ID: <20260923163721.1584779-3-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

1 files changed, 157 insertions, 0 deletions

diff --git a/Documentation/ABI/testing/sysfs-devices-edac b/Documentation/ABI/testing/sysfs-devices-edac
new file mode 100644
index 000000000..256a9e990
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-devices-edac
@@ -0,0 +1,157 @@
+What:		/sys/devices/system/edac/mc/mc*/reset_counters
+Date:		January 2006
+Contact:	linux-edac@vger.kernel.org
+Description:	This write-only control file will zero all the statistical
+		counters for UE and CE errors on the given memory controller.
+		Zeroing the counters will also reset the timer indicating how
+		long since the last counter were reset. This is useful for
+		computing errors/time.  Since the counters are always reset
+		at driver initialization time, no module/kernel parameter
+		is available.
+
+What:		/sys/devices/system/edac/mc/mc*/seconds_since_reset
+Date:		January 2006
+Contact:	linux-edac@vger.kernel.org
+Description:	This attribute file displays how many seconds have elapsed
+		since the last counter reset. This can be used with the error
+		counters to measure error rates.
+
+What:		/sys/devices/system/edac/mc/mc*/mc_name
+Date:		January 2006
+Contact:	linux-edac@vger.kernel.org
+Description:	This attribute file displays the type of memory controller
+		that is being utilized.
+
+What:		/sys/devices/system/edac/mc/mc*/size_mb
+Date:		January 2006
+Contact:	linux-edac@vger.kernel.org
+Description:	This attribute file displays, in count of megabytes, of memory
+		that this memory controller manages.
+
+What:		/sys/devices/system/edac/mc/mc*/ue_count
+Date:		January 2006
+Contact:	linux-edac@vger.kernel.org
+Description:	This attribute file displays the total count of uncorrectable
+		errors that have occurred on this memory controller. If
+		panic_on_ue is set, this counter will not have a chance to
+		increment, since EDAC will panic the system
+
+What:		/sys/devices/system/edac/mc/mc*/ue_noinfo_count
+Date:		January 2006
+Contact:	linux-edac@vger.kernel.org
+Description:	This attribute file displays the number of UEs that have
+		occurred on this memory controller with no information as to
+		which DIMM slot is having errors.
+
+What:		/sys/devices/system/edac/mc/mc*/ce_count
+Date:		January 2006
+Contact:	linux-edac@vger.kernel.org
+Description:	This attribute file displays the total count of correctable
+		errors that have occurred on this memory controller. This
+		count is very important to examine. CEs provide early
+		indications that a DIMM is beginning to fail. This count
+		field should be monitored for non-zero values and report
+		such information to the system administrator.
+
+What:		/sys/devices/system/edac/mc/mc*/ce_noinfo_count
+Date:		January 2006
+Contact:	linux-edac@vger.kernel.org
+Description:	This attribute file displays the number of CEs that
+		have occurred on this memory controller wherewith no
+		information as to which DIMM slot is having errors. Memory is
+		handicapped, but operational, yet no information is available
+		to indicate which slot the failing memory is in. This count
+		field should be also be monitored for non-zero values.
+
+What:		/sys/devices/system/edac/mc/mc*/sdram_scrub_rate
+Date:		February 2007
+Contact:	linux-edac@vger.kernel.org
+Description:	Read/Write attribute file that controls memory scrubbing.
+		The scrubbing rate used by the memory controller is set by
+		writing a minimum bandwidth in bytes/sec to the attribute file.
+		The rate will be translated to an internal value that gives at
+		least the specified rate.
+		Reading the file will return the actual scrubbing rate employed.
+		If configuration fails or memory scrubbing is not implemented,
+		the value of the attribute file will be -1.
+
+What:		/sys/devices/system/edac/mc/mc*/max_location
+Date:		April 2012
+Contact:	Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
+		linux-edac@vger.kernel.org
+Description:	This attribute file displays the information about the last
+		available memory slot in this memory controller. It is used by
+		userspace tools in order to display the memory filling layout.
+
+What:		/sys/devices/system/edac/mc/mc*/(dimm|rank)*/size
+Date:		April 2012
+Contact:	Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
+		linux-edac@vger.kernel.org
+Description:	This attribute file will display the size of dimm or rank.
+		For dimm*/size, this is the size, in MB of the DIMM memory
+		stick. For rank*/size, this is the size, in MB for one rank
+		of the DIMM memory stick. On single rank memories (1R), this
+		is also the total size of the dimm. On dual rank (2R) memories,
+		this is half the size of the total DIMM memories.
+
+What:		/sys/devices/system/edac/mc/mc*/(dimm|rank)*/dimm_dev_type
+Date:		April 2012
+Contact:	Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
+		linux-edac@vger.kernel.org
+Description:	This attribute file will display what type of DRAM device is
+		being utilized on this DIMM (x1, x2, x4, x8, ...).
+
+What:		/sys/devices/system/edac/mc/mc*/(dimm|rank)*/dimm_edac_mode
+Date:		April 2012
+Contact:	Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
+		linux-edac@vger.kernel.org
+Description:	This attribute file will display what type of Error detection
+		and correction is being utilized. For example: S4ECD4ED would
+		mean a Chipkill with x4 DRAM.
+
+What:		/sys/devices/system/edac/mc/mc*/(dimm|rank)*/dimm_label
+Date:		April 2012
+Contact:	Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
+		linux-edac@vger.kernel.org
+Description:	This control file allows this DIMM to have a label assigned
+		to it. With this label in the module, when errors occur
+		the output can provide the DIMM label in the system log.
+		This becomes vital for panic events to isolate the
+		cause of the UE event.
+		DIMM Labels must be assigned after booting, with information
+		that correctly identifies the physical slot with its
+		silk screen label. This information is currently very
+		motherboard specific and determination of this information
+		must occur in userland at this time.
+
+What:		/sys/devices/system/edac/mc/mc*/(dimm|rank)*/dimm_location
+Date:		April 2012
+Contact:	Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
+		linux-edac@vger.kernel.org
+Description:	This attribute file will display the location (csrow/channel,
+		branch/channel/slot or channel/slot) of the dimm or rank.
+
+What:		/sys/devices/system/edac/mc/mc*/(dimm|rank)*/dimm_mem_type
+Date:		April 2012
+Contact:	Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
+		linux-edac@vger.kernel.org
+Description:	This attribute file will display what type of memory is
+		currently on this csrow. Normally, either buffered or
+		unbuffered memory (for example, Unbuffered-DDR3).
+
+What:		/sys/devices/system/edac/mc/mc*/(dimm|rank)*/dimm_ce_count
+Date:		October 2016
+Contact:	linux-edac@vger.kernel.org
+Description:	This attribute file displays the total count of correctable
+		errors that have occurred on this DIMM. This count is very important
+		to examine. CEs provide early indications that a DIMM is beginning
+		to fail. This count field should be monitored for non-zero values
+		and report such information to the system administrator.
+
+What:		/sys/devices/system/edac/mc/mc*/(dimm|rank)*/dimm_ue_count
+Date:		October 2016
+Contact:	linux-edac@vger.kernel.org
+Description:	This attribute file displays the total count of uncorrectable
+		errors that have occurred on this DIMM. If panic_on_ue is set, this
+		counter will not have a chance to increment, since EDAC will panic the
+		system