From: Paolo Bonzini Date: Wed, 29 Jul 2015 09:06:34 +0000 (+0200) Subject: KVM: x86: remove unnecessary memory barriers for shared MSRs X-Git-Url: http://git.cdn.openwrt.org/?a=commitdiff_plain;h=c847fe8895fded71d0b9a0ba58d3828320fd1824;p=openwrt%2Fstaging%2Fblogic.git KVM: x86: remove unnecessary memory barriers for shared MSRs There is no smp_rmb matching the smp_wmb. shared_msr_update is called from hardware_enable, which in turn is called via on_each_cpu. on_each_cpu and must imply a read memory barrier (on x86 the rmb is achieved simply through asm volatile in native_apic_mem_write). Signed-off-by: Paolo Bonzini --- diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index c675ea3351cf..2d62229aac26 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -222,11 +222,9 @@ static void shared_msr_update(unsigned slot, u32 msr) void kvm_define_shared_msr(unsigned slot, u32 msr) { BUG_ON(slot >= KVM_NR_SHARED_MSRS); + shared_msrs_global.msrs[slot] = msr; if (slot >= shared_msrs_global.nr) shared_msrs_global.nr = slot + 1; - shared_msrs_global.msrs[slot] = msr; - /* we need ensured the shared_msr_global have been updated */ - smp_wmb(); } EXPORT_SYMBOL_GPL(kvm_define_shared_msr);