From mboxrd@z Thu Jan  1 00:00:00 1970
From: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Subject: Re: [PATCH v5 1/6] x86/hvm: Collect information of TSC
 scaling ratio
Date: Wed, 24 Feb 2016 09:49:20 -0500
Message-ID: <56CDC2F0.3060201@oracle.com>
References: <1456193104-12761-1-git-send-email-haozhong.zhang@intel.com>
	<1456193104-12761-2-git-send-email-haozhong.zhang@intel.com>
	<56CC65F4.4070803@oracle.com>
	<56CC764F02000078000D53C9@prv-mh.provo.novell.com>
	<56CC69AC.1030109@oracle.com>
	<56CC8ACC02000078000D5480@prv-mh.provo.novell.com>
	<20160224060020.GA3974@hz-desktop.sh.intel.com>
	<20160224134606.GA10229@hz-desktop.sh.intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; Format="flowed"
Content-Transfer-Encoding: 7bit
Return-path: <xen-devel-bounces@lists.xen.org>
In-Reply-To: <20160224134606.GA10229@hz-desktop.sh.intel.com>
List-Unsubscribe: <http://lists.xen.org/cgi-bin/mailman/options/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=unsubscribe>
List-Post: <mailto:xen-devel@lists.xen.org>
List-Help: <mailto:xen-devel-request@lists.xen.org?subject=help>
List-Subscribe: <http://lists.xen.org/cgi-bin/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=subscribe>
Sender: xen-devel-bounces@lists.xen.org
Errors-To: xen-devel-bounces@lists.xen.org
To: Jan Beulich <JBeulich@suse.com>, Aravind Gopalakrishnan <Aravind.Gopalakrishnan@amd.com>, Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>, Andrew Cooper <andrew.cooper3@citrix.com>, Jun Nakajima <jun.nakajima@intel.com>, Kevin Tian <kevin.tian@intel.com>, xen-devel@lists.xen.org, Keir Fraser <keir@xen.org>
List-Id: xen-devel@lists.xenproject.org

On 02/24/2016 08:46 AM, Haozhong Zhang wrote:
> Sorry, forgot sending the last reply to Boris.
>
> On 02/24/16 14:00, Haozhong Zhang wrote:
>> On 02/23/16 08:37, Jan Beulich wrote:
>>>>>> On 23.02.16 at 15:16, <boris.ostrovsky@oracle.com> wrote:
>>>> On 02/23/2016 09:10 AM, Jan Beulich wrote:
>>>>>>>> On 23.02.16 at 15:00, <boris.ostrovsky@oracle.com> wrote:
>>>>>> On 02/22/2016 09:04 PM, Haozhong Zhang wrote:
>>>>>>
>>>>>>>     
>>>>>>> +    if ( cpu_has_tsc_ratio )
>>>>>>> +        svm_function_table.tsc_scaling.ratio_frac_bits = 32;
>>>>>>> +
>>>>>>>     
>>>>>>> +#define hvm_tsc_scaling_supported \
>>>>>>> +    (!!hvm_funcs.tsc_scaling.ratio_frac_bits)
>>>>>>> +
>>>>>> What is the difference (in usage) between cpu_has_tsc_ratio and
>>>>>> hvm_tsc_scaling_supported? Isn't the first imply the second (and if yes
>>>>>> then what's the reason for having the latter)?
>>>>> Iiuc cpu_has_tsc_ratio is AMD/SVM specific, while
>>>>> hvm_tsc_scaling_supported is meant to be vendor independent.
>> Yes, it's to be vendor independent. Earlier versions of this patch
>> series set a field tsc_scaling_supported in hvm_function_table if
>> cpu_has_vmx_tsc_scaling or cpu_has_tsc_ratio. Jan suggested we could get
>> the same information if some of other fields are initialized
>> conditionally, and no extra field (tsc_scaling_supported) would be
>> needed any more.
>>
>>>> Ah, OK. Can we then
>>>>
>>>> #define hvm_tsc_scaling_supported (cpu_has_vmx_tsc_scaling ||
>>>> cpu_has_tsc_ratio)
>>> Why would we? The above is doing precisely (but implicitly) that,
>>> just with only one memory access instead of two.
>>>
>> Boris, does the current one look fine for you, as it does the same thing
>> as your suggested one?

Yes, this is fine. IMO using cpu_has_* would be slightly  more logical 
but as Jan pointed it is also potentially more costly.

Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>

-boris