hv: lapic: fix MISRA-C violation of potential numeric overflow

This patch fixes the MISRA-C violations in arch/x86/lapic.c, change local variable from
uint32_t to uint64_t to avoid potential numeric overflow.

Tracked-On: #861
Signed-off-by: Yan, Like <like.yan@intel.com>
Acked-by: Eddie Dong <eddie.dong@intel.com>

Author: lyan3

hypervisor/arch/x86/lapic.c

@@ -154,15 +154,15 @@ static void restore_lapic(const struct lapic_regs *regs)
 
 void suspend_lapic(void)
 {
-	uint32_t val;
+	uint64_t val;
 
 	saved_lapic_base_msr.value = msr_read(MSR_IA32_APIC_BASE);
 	save_lapic(&saved_lapic_regs);
 
 	/* disable APIC with software flag */
-	val = (uint32_t) msr_read(MSR_IA32_EXT_APIC_SIVR);
-	val = (~LAPIC_SVR_APIC_ENABLE_MASK) & val;
-	msr_write(MSR_IA32_EXT_APIC_SIVR, (uint64_t) val);
+	val = msr_read(MSR_IA32_EXT_APIC_SIVR);
+	val = (~(uint64_t)LAPIC_SVR_APIC_ENABLE_MASK) & val;
+	msr_write(MSR_IA32_EXT_APIC_SIVR, val);
 }
 
 void resume_lapic(void)