From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Message-ID: <516C4E14.1080708@amd.com>
Date: Mon, 15 Apr 2013 13:59:32 -0500
From: Aravind <aravind.gopalakrishnan@amd.com>
MIME-Version: 1.0
To: Borislav Petkov <bp@alien8.de>
CC: Bjorn Helgaas <bhelgaas@google.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Doug Thompson <dougthompson@xmission.com>,
	Jesse Barnes <jbarnes@virtuousgeek.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-pci@vger.kernel.org" <linux-pci@vger.kernel.org>,
	<linux-edac@vger.kernel.org>
Subject: Re: [PATCH] edac: Handle EDAC ECC errors for Family 16h
References: <1366039059-15772-1-git-send-email-Aravind.Gopalakrishnan@amd.com> <CAErSpo6cD4WtPhTbcBEyfOLNASJzr+cU2Lch=fVyqn-gh15drQ@mail.gmail.com> <20130415161121.GA18440@pd.tnic> <516C2EA2.1050806@amd.com>
In-Reply-To: <516C2EA2.1050806@amd.com>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-pci.vger.kernel.org>

On 04/15/2013 11:45 AM, Aravind wrote:
> On 04/15/2013 11:11 AM, Borislav Petkov wrote:
>> On Mon, Apr 15, 2013 at 09:56:08AM -0600, Bjorn Helgaas wrote:
>>> On Mon, Apr 15, 2013 at 9:17 AM, Aravind Gopalakrishnan
>>>> @@ -172,7 +176,12 @@
>>>>    */
>>>>   #define PCI_DEVICE_ID_AMD_15H_NB_F1    0x1601
>>>>   #define PCI_DEVICE_ID_AMD_15H_NB_F2    0x1602
>>>> -
>>>> +#define PCI_DEVICE_ID_AMD_16H_NB_F0    0x1530
>>>> +#define PCI_DEVICE_ID_AMD_16H_NB_F1    0x1531
>>>> +#define PCI_DEVICE_ID_AMD_16H_NB_F2    0x1532
>>>> +#define PCI_DEVICE_ID_AMD_16H_NB_F3    0x1533
>>>> +#define PCI_DEVICE_ID_AMD_16H_NB_F4    0x1534
>>>> +#define PCI_DEVICE_ID_AMD_16H_NB_F5    0x1535
>>>>
>>>>   /*
>>>>    * Function 1 - Address Map
>>>> @@ -300,6 +309,7 @@ enum amd_families {
>>>>          K8_CPUS = 0,
>>>>          F10_CPUS,
>>>>          F15_CPUS,
>>>> +       F16_CPUS,
>>>>          NUM_FAMILIES,
>>>>   };
>>>>
>>>> diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
>>>> index 1679ff6..59f732f 100644
>>>> --- a/include/linux/pci_ids.h
>>>> +++ b/include/linux/pci_ids.h
>>>> @@ -519,6 +519,8 @@
>>>>   #define PCI_DEVICE_ID_AMD_11H_NB_LINK  0x1304
>>>>   #define PCI_DEVICE_ID_AMD_15H_NB_F3    0x1603
>>>>   #define PCI_DEVICE_ID_AMD_15H_NB_F4    0x1604
>>>> +#define PCI_DEVICE_ID_AMD_16H_NB_F3    0x1533
>>>> +#define PCI_DEVICE_ID_AMD_16H_NB_F4    0x1534
>>> What is the point of adding identical #defines both here and in
>>> amd64_edac.h?  Also, read the note at the top of
>>> include/linux/pci_ids.h.
>> Yeah, they should be in pci_ids.h since they're used in amd_nb.c and
>> amd64_edac.c.
>>
>> But, what shouldn't be is adding unused defines (*_F0 and *_F5).
>> Aravind, please drop them.
>
>     Yes, I have made the changes and sent it out as version 2 of the 
> PATCH.
>     Do have a look and let me know if I need to make any changes.
>
>
>> I'll review the rest of the patch when I get around to it this week.
>>
>

      Correction: Fam16 does not support GART. I had mistakenly included 
it in amd_nb.c (Apologies)
      I have fixed it up and made few other cosmetic changes to the 
patch. I have tested it once again to
      make certain it works fine and it does..
      Sending it out as V3 of the patch..