[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <f04c1409-dd6c-15c0-4fea-e448e614aab9@amd.com>
Date: Thu, 1 Nov 2018 14:55:36 +0000
From: "Moger, Babu" <Babu.Moger@....com>
To: Fenghua Yu <fenghua.yu@...el.com>
CC: "tglx@...utronix.de" <tglx@...utronix.de>,
"mingo@...hat.com" <mingo@...hat.com>,
"bp@...en8.de" <bp@...en8.de>, "corbet@....net" <corbet@....net>,
"reinette.chatre@...el.com" <reinette.chatre@...el.com>,
"peterz@...radead.org" <peterz@...radead.org>,
"gregkh@...uxfoundation.org" <gregkh@...uxfoundation.org>,
"davem@...emloft.net" <davem@...emloft.net>,
"akpm@...ux-foundation.org" <akpm@...ux-foundation.org>,
"hpa@...or.com" <hpa@...or.com>, "x86@...nel.org" <x86@...nel.org>,
"mchehab+samsung@...nel.org" <mchehab+samsung@...nel.org>,
"arnd@...db.de" <arnd@...db.de>,
"kstewart@...uxfoundation.org" <kstewart@...uxfoundation.org>,
"pombredanne@...b.com" <pombredanne@...b.com>,
"rafael@...nel.org" <rafael@...nel.org>,
"kirill.shutemov@...ux.intel.com" <kirill.shutemov@...ux.intel.com>,
"tony.luck@...el.com" <tony.luck@...el.com>,
"qianyue.zj@...baba-inc.com" <qianyue.zj@...baba-inc.com>,
"xiaochen.shen@...el.com" <xiaochen.shen@...el.com>,
"pbonzini@...hat.com" <pbonzini@...hat.com>,
"Singh, Brijesh" <brijesh.singh@....com>,
"Hurwitz, Sherry" <sherry.hurwitz@....com>,
"dwmw2@...radead.org" <dwmw2@...radead.org>,
"Lendacky, Thomas" <Thomas.Lendacky@....com>,
"luto@...nel.org" <luto@...nel.org>,
"joro@...tes.org" <joro@...tes.org>,
"jannh@...gle.com" <jannh@...gle.com>,
"vkuznets@...hat.com" <vkuznets@...hat.com>,
"rian@...m.mit.edu" <rian@...m.mit.edu>,
"jpoimboe@...hat.com" <jpoimboe@...hat.com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"linux-doc@...r.kernel.org" <linux-doc@...r.kernel.org>
Subject: RE: [PATCH v5 00/13] arch/x86: AMD QoS support
Fenghua,
> -----Original Message-----
> From: linux-kernel-owner@...r.kernel.org <linux-kernel-
> owner@...r.kernel.org> On Behalf Of Fenghua Yu
> Sent: Wednesday, October 31, 2018 4:55 PM
> To: Moger, Babu <Babu.Moger@....com>
> Cc: tglx@...utronix.de; mingo@...hat.com; bp@...en8.de; corbet@....net;
> reinette.chatre@...el.com; peterz@...radead.org;
> gregkh@...uxfoundation.org; davem@...emloft.net; akpm@...ux-
> foundation.org; hpa@...or.com; x86@...nel.org;
> mchehab+samsung@...nel.org; arnd@...db.de;
> kstewart@...uxfoundation.org; pombredanne@...b.com;
> rafael@...nel.org; kirill.shutemov@...ux.intel.com; tony.luck@...el.com;
> qianyue.zj@...baba-inc.com; xiaochen.shen@...el.com;
> pbonzini@...hat.com; Singh, Brijesh <brijesh.singh@....com>; Hurwitz,
> Sherry <sherry.hurwitz@....com>; dwmw2@...radead.org; Lendacky,
> Thomas <Thomas.Lendacky@....com>; luto@...nel.org; joro@...tes.org;
> jannh@...gle.com; vkuznets@...hat.com; rian@...m.mit.edu;
> jpoimboe@...hat.com; linux-kernel@...r.kernel.org; linux-
> doc@...r.kernel.org
> Subject: Re: [PATCH v5 00/13] arch/x86: AMD QoS support
>
> On Thu, Oct 18, 2018 at 10:52:00PM +0000, Moger, Babu wrote:
> > Changes from v4 -> v5:
> > b. The functions update_mba_bw and set_mba_sc is not required for
> AMD.
> > Removed all the changes related to these functions.
>
> Hi, Babu,
>
> In v4, you says AMD won't support MBA software controller.
>
> In v5, does AMD support MBA software controller or not? The v5 patches
No, AMD does not support MBA software controller. set_mba_sc will always
return -EINVAL on AMD(because delay_linear is set false). That is the
reason, I did not add this check. I will add this check in next refresh.
Will add Suggested-by you. Thanks
> show the feature is supported now. If that's a code bug, you may need
> the following patch?
>
> x86/resctrl: Only Intel RDT supports MBA software controller
>
> AMD doesn't support the feature.
>
> Signed-off-by: Fenghua Yu <fenghua.yu@...el.com>
>
> diff --git a/arch/x86/kernel/cpu/resctrl_rdtgroup.c
> b/arch/x86/kernel/cpu/resctrl_rdtgroup.c
> index 8b6b4a8bb7ca..89dd9b7c9dd7 100644
> --- a/arch/x86/kernel/cpu/resctrl_rdtgroup.c
> +++ b/arch/x86/kernel/cpu/resctrl_rdtgroup.c
> @@ -1878,9 +1878,11 @@ static int parse_rdtgroupfs_options(char *data)
> if (ret)
> goto out;
> } else if (!strcmp(token, "mba_MBps")) {
> - ret = set_mba_sc(true);
> - if (ret)
> - goto out;
> + if (boot_cpu_data.x86_vendor ==
> X86_VENDOR_INTEL) {
> + ret = set_mba_sc(true);
> + if (ret)
> + goto out;
> + }
> } else {
> ret = -EINVAL;
> goto out;
Powered by blists - more mailing lists