[PATCH v3] KVM: kvmclock: Fix vCPUs > 64 can't be online/hotpluged

[Wanpeng Li <kernellwp@gmail.com>]

From: Wanpeng Li <wanpengli@tencent.com>

The per-cpu vsyscall pvclock data pointer assigns either an element of the 
static array hv_clock_boot (#vCPU <= 64) or dynamically allocated memory 
hvclock_mem (vCPU > 64), the dynamically memory will not be allocated if 
kvmclock vsyscall is disabled, this can result in cpu hotpluged fails in 
kvmclock_setup_percpu() which returns -ENOMEM. It's broken for no-vsyscall
and sometimes you end up with vsyscall disabled if the host does something 
strange. This patch fixes it by allocating this dynamically memory 
unconditionally even if vsyscall is disabled.

Fixes: 6a1cac56f4 ("x86/kvm: Use __bss_decrypted attribute in shared variables")
Reported-by: Zelin Deng <zelin.deng@linux.alibaba.com>
Tested-by: Haiwei Li <lihaiwei@tencent.com>
Cc: Brijesh Singh <brijesh.singh@amd.com>
Cc: stable@vger.kernel.org#v4.19-rc5+
Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
---
v2 -> v3:
 * allocate dynamically memory unconditionally
v1 -> v2:
 * add code comments

 arch/x86/kernel/kvmclock.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c
index aa59374..a72b16e 100644
--- a/arch/x86/kernel/kvmclock.c
+++ b/arch/x86/kernel/kvmclock.c
@@ -268,6 +268,8 @@ static void __init kvmclock_init_mem(void)
 
 static int __init kvm_setup_vsyscall_timeinfo(void)
 {
+	kvmclock_init_mem();
+
 #ifdef CONFIG_X86_64
 	u8 flags;
 
@@ -281,8 +283,6 @@ static int __init kvm_setup_vsyscall_timeinfo(void)
 	kvm_clock.vdso_clock_mode = VDSO_CLOCKMODE_PVCLOCK;
 #endif
 
-	kvmclock_init_mem();
-
 	return 0;
 }
 early_initcall(kvm_setup_vsyscall_timeinfo);
-- 
2.7.4