OpenCloudOS-Kernel/arch
Haitao Shan 7de33b0fc9 KVM: x86: Fix lapic timer interrupt lost after loading a snapshot.
commit 9cfec6d097c607e36199cf0cfbb8cf5acbd8e9b2 upstream.

When running android emulator (which is based on QEMU 2.12) on
certain Intel hosts with kernel version 6.3-rc1 or above, guest
will freeze after loading a snapshot. This is almost 100%
reproducible. By default, the android emulator will use snapshot
to speed up the next launching of the same android guest. So
this breaks the android emulator badly.

I tested QEMU 8.0.4 from Debian 12 with an Ubuntu 22.04 guest by
running command "loadvm" after "savevm". The same issue is
observed. At the same time, none of our AMD platforms is impacted.
More experiments show that loading the KVM module with
"enable_apicv=false" can workaround it.

The issue started to show up after commit 8e6ed96cdd ("KVM: x86:
fire timer when it is migrated and expired, and in oneshot mode").
However, as is pointed out by Sean Christopherson, it is introduced
by commit 967235d320 ("KVM: vmx: clear pending interrupts on
KVM_SET_LAPIC"). commit 8e6ed96cdd ("KVM: x86: fire timer when
it is migrated and expired, and in oneshot mode") just makes it
easier to hit the issue.

Having both commits, the oneshot lapic timer gets fired immediately
inside the KVM_SET_LAPIC call when loading the snapshot. On Intel
platforms with APIC virtualization and posted interrupt processing,
this eventually leads to setting the corresponding PIR bit. However,
the whole PIR bits get cleared later in the same KVM_SET_LAPIC call
by apicv_post_state_restore. This leads to timer interrupt lost.

The fix is to move vmx_apicv_post_state_restore to the beginning of
the KVM_SET_LAPIC call and rename to vmx_apicv_pre_state_restore.
What vmx_apicv_post_state_restore does is actually clearing any
former apicv state and this behavior is more suitable to carry out
in the beginning.

Fixes: 967235d320 ("KVM: vmx: clear pending interrupts on KVM_SET_LAPIC")
Cc: stable@vger.kernel.org
Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Haitao Shan <hshan@google.com>
Link: https://lore.kernel.org/r/20230913000215.478387-1-hshan@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2023-11-28 17:19:56 +00:00
..
alpha Kbuild updates for v6.6 2023-09-05 11:01:47 -07:00
arc ARC updates for v6.6 2023-09-04 15:38:24 -07:00
arm ARM: 9320/1: fix stack depot IRQ stack filter 2023-11-28 17:19:43 +00:00
arm64 arm64: dts: ls208xa: use a pseudo-bus to constrain usb dma size 2023-11-28 17:19:42 +00:00
csky arch/csky 2nd patches for 6.6 2023-09-01 08:02:45 -07:00
hexagon Add x86 shadow stack support 2023-08-31 12:20:12 -07:00
ia64 cpu-hotplug: Provide prototypes for arch CPU registration 2023-10-11 14:27:37 +02:00
loongarch LoongArch: Disable WUC for pgprot_writecombine() like ioremap_wc() 2023-10-18 08:42:52 +08:00
m68k ata changes for 6.6 2023-09-05 12:37:28 -07:00
microblaze Microblaze patches for 6.6-rc1 2023-09-05 10:15:22 -07:00
mips KVM: MIPS: fix -Wunused-but-set-variable warning 2023-10-12 11:25:40 -04:00
nios2 Add x86 shadow stack support 2023-08-31 12:20:12 -07:00
openrisc OpenRISC updates for 6.6 2023-09-05 10:09:31 -07:00
parisc parisc architecture fixes for kernel v6.6-rc5: 2023-10-07 13:05:43 -07:00
powerpc powerpc/perf: Fix disabling BHRB and instruction sampling 2023-11-28 17:19:55 +00:00
riscv riscv: split cache ops out of dma-noncoherent.c 2023-11-28 17:19:49 +00:00
s390 s390 updates for 6.6-rc7 2023-10-21 10:11:11 -07:00
sh sh: bios: Revive earlyprintk support 2023-11-20 11:59:22 +01:00
sparc sparc32: fix a braino in fault handling in csum_and_copy_..._user() 2023-10-27 20:06:06 -04:00
um This pull request contains the following changes for UML: 2023-09-04 11:32:21 -07:00
x86 KVM: x86: Fix lapic timer interrupt lost after loading a snapshot. 2023-11-28 17:19:56 +00:00
xtensa xtensa: boot/lib: fix function prototypes 2023-09-20 05:03:30 -07:00
.gitignore
Kconfig Add x86 shadow stack support 2023-08-31 12:20:12 -07:00