Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <1f6d32e14f121a8ccf8807b8343597c3ae88c7d2.1587196252.git.yu.c.chen@intel.com>
 <20210308134957.23852-1-youling257@gmail.com> <CAAYoRsXec2eq=t-pSn5TOqNt0G6kfZCKkDuEhCnX5SgL0zgkBg@mail.gmail.com>
 <20210308161548.GA37664@chenyu-desktop>
In-Reply-To: <20210308161548.GA37664@chenyu-desktop>
From:   Doug Smythies <dsmythies@telus.net>
Date:   Wed, 10 Mar 2021 16:03:31 -0800
Message-ID: <CAAYoRsVkrHberSgM42dqYjdVwz8vumURJ1_DGeV1R5-=LMdjVA@mail.gmail.com>
Subject: Re: [3/3,v3] tools/power turbostat: Enable accumulate RAPL display
To:     Chen Yu <yu.c.chen@intel.com>, bas@basnieuwenhuizen.nl,
        erwanaliasr1@gmail.com, owen.si@ucloud.cn
Cc:     youling257 <youling257@gmail.com>, lenb@kernel.org,
        rjw@rjwysocki.net, linux-kernel@vger.kernel.org,
        Len Brown <len.brown@intel.com>,
        dsmythies <dsmythies@telus.net>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Hi Yu,

I am just resending your e-mail, adjusting the "To:" list to
include the 3 others that have submitted similar patches.

... Doug

On Mon, Mar 8, 2021 at 8:11 AM Chen Yu <yu.c.chen@intel.com> wrote:
>
> Hi,
> On Mon, Mar 08, 2021 at 07:37:07AM -0800, Doug Smythies wrote:
> > On Mon, Mar 8, 2021 at 5:50 AM youling257 <youling257@gmail.com> wrote:
> > >
> > > this cause turbostat not work on amd cpu.
> > >
> > > root@localhost:~# /turbostat
> > > turbostat version 20.09.30 - Len Brown <lenb@kernel.org>
> > > CPUID(0): AuthenticAMD 0xd CPUID levels; 0x8000001f xlevels; family:model:stepping 0x17:18:1 (23:24:1)
> >
> > There are already two fixes for this in the queue.
> > https://marc.info/?l=linux-pm&m=161382097503925&w=2
> > https://marc.info/?l=linux-pm&m=161141701219263&w=2
> Thanks for reporting and pointing this out. I assume these two patches are both fixing the
> same issue? It looks like these two patches should be merged into one:
> 1. Bingsong's patch access MSR_PKG_ENERGY_STAT only when RAPL_AMD_F17H is set,
>    which is consistent with the original context.
> 2. Bas Nieuwenhuizen's patch also fixes the issue in idx_valid()
>    in case RAPL_PKG was not set but with RAPL_AMD_F17H set.
>
> thanks,
> Chenyu