On 05/07/21 09:15, Greg KH wrote:
On Mon, Jun 28, 2021 at 09:10:54PM +0000, Alper Gun wrote:
commit 934002cd660b035b926438244b4294e647507e13 upstream.
Send SEV_CMD_DECOMMISSION command to PSP firmware if ASID binding
fails. If a failure happens after a successful LAUNCH_START command,
a decommission command should be executed. Otherwise, guest context
will be unfreed inside the AMD SP. After the firmware will not have
memory to allocate more SEV guest context, LAUNCH_START command will
begin to fail with SEV_RET_RESOURCE_LIMIT error.
The existing code calls decommission inside sev_unbind_asid, but it is
not called if a failure happens before guest activation succeeds. If
sev_bind_asid fails, decommission is never called. PSP firmware has a
limit for the number of guests. If sev_asid_binding fails many times,
PSP firmware will not have resources to create another guest context.
Cc: stable@xxxxxxxxxxxxxxx
Fixes: 59414c989220 ("KVM: SVM: Add support for KVM_SEV_LAUNCH_START command")
Reported-by: Peter Gonda <pgonda@xxxxxxxxxx>
Signed-off-by: Alper Gun <alpergun@xxxxxxxxxx>
Reviewed-by: Marc Orr <marcorr@xxxxxxxxxx>
Signed-off-by: Paolo Bonzini <pbonzini@xxxxxxxxxx>
Message-Id: <20210610174604.2554090-1-alpergun@xxxxxxxxxx>
Message-id? Odd...
Not that much, see "git log -- drivers | grep Message-Id". A link to
lore.kernel.org is getting more popular these days, but Message-Id is
what "git am" knows about.
---
arch/x86/kvm/svm.c | 32 +++++++++++++++++++++-----------
1 file changed, 21 insertions(+), 11 deletions(-)
<snip>
Can you also provide working backports for the newer kernel trees as
well? We would need this in 5.10 and 5.12, right?
Already queued:
https://lore.kernel.org/stable/20210628141828.31757-102-sashal@xxxxxxxxxx/
for 5.12
https://lore.kernel.org/stable/20210628142607.32218-94-sashal@xxxxxxxxxx/
for 5.10
For this one:
Acked-by: Paolo Bonzini <pbonzini@xxxxxxxxxx>
Paolo
