[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <56E19B21.8090501@redhat.com>
Date: Thu, 10 Mar 2016 17:04:49 +0100
From: Paolo Bonzini <pbonzini@...hat.com>
To: Xiao Guangrong <guangrong.xiao@...ux.intel.com>,
Lan Tianyu <tianyu.lan@...el.com>,
Thomas Gleixner <tglx@...utronix.de>
Cc: gleb@...nel.org, mingo@...hat.com, hpa@...or.com, x86@...nel.org,
kvm@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] KVM: Remove redundant smp_mb() in the
kvm_mmu_commit_zap_page()
On 10/03/2016 16:45, Xiao Guangrong wrote:
>>
>>> Compared to smp_load_acquire(), smp_mb() adds an ordering between stores
>>> and loads.
>>
>> Here, the ordering is load-store, hence...
>
> Yes, this is why i put smp_mb() in the code. :)
Here is a table of barriers:
'. after| |
before '. | load | store
__________'.|___________________|________________________
| |
| smp_rmb | smp_load_acquire
load | smp_load_acquire | smp_store_release XX
| smp_mb | smp_mb
____________|___________________|________________________
| |
| | smp_wmb
store | smp_mb | smp_store_release
| | smp_mb
| |
Your case is the one marked with XX, so a smp_load_acquire() is
enough---and it's preferrable, because it's cheaper than smp_mb() and
more self-documenting.
Paolo
Powered by blists - more mailing lists