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;
Date:   Thu, 23 Sep 2021 14:29:07 +0000
From:   Yazen Ghannam <yazen.ghannam@amd.com>
To:     Mukul Joshi <mukul.joshi@amd.com>
Cc:     linux-edac@vger.kernel.org, x86@kernel.org,
        linux-kernel@vger.kernel.org, bp@alien8.de, mingo@redhat.com,
        mchehab@kernel.org, amd-gfx@lists.freedesktop.org
Subject: Re: [PATCHv3 2/2] drm/amdgpu: Register MCE notifier for Aldebaran RAS
Message-ID: <YUyPM7VfYFG/PJmu@yaz-ubuntu>
References: <20210913021311.12896-2-mukul.joshi@amd.com>
 <20210922193620.15925-1-mukul.joshi@amd.com>
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210922193620.15925-1-mukul.joshi@amd.com>
MIME-Version: 1.0
X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-MessageData-0: =?us-ascii?Q?mLAuCd6Dvl05kLOBCYiYt7MMCi7PPIwz7BRwW9hvPsOwKVPPGk4pbGzY0HF6?=
 =?us-ascii?Q?k3m4lXFB14l2499lGrQR/EZI/Jn7ryv31eHGWakz5ahG46lUjSvlRO1RMVPu?=
 =?us-ascii?Q?fNQxwaIow2YG0D7dj89j5DW9elSpmEAZdrGsRrPZF0QkG5nGpvk2V3sEi4MI?=
 =?us-ascii?Q?pm2ItEmYqU/oD/Jj2w2szNG52EvRhgh9bVmf/zRl+4HdL2jd5+cPdFl0W//M?=
 =?us-ascii?Q?XPDFEXU6Jlz7O0P4X1iIhiOnZP4pomL8ke/FZTB2U4mJouXW/4zCziz89EBr?=
 =?us-ascii?Q?VVJDDUsJjVJsRdRi8yx2PDrNYwg3GTGLCZc9hXbiDAmLpPY5l8d6qtlAsu9f?=
 =?us-ascii?Q?wTzxvSnWeZuikezuJEBYEe4XUaY7JBtqrrtuvTY05wNUfgu/6bKMKpPcKbe3?=
 =?us-ascii?Q?dyBSXxgYviJAHEn7obUfIpudngAPZVIrHRjov8hLESCr/nTYjZb4KZz/O/hP?=
 =?us-ascii?Q?W+ARY6q8TAU8s9pV9whxVnVqpjsiC1qyMnnhuAfrM5M1+ZGbDhzZ9liNQ6Ac?=
 =?us-ascii?Q?d+g1iOHF+PYdUqX6SAW5dNPcqjLykobTmcHcIgRh7cs0KDKHW/W8j1E07XQd?=
 =?us-ascii?Q?iMY+ZGJZ3qWiOSNtGBvO0f+tHEA3f8U/RJE3TZXz+ZyeiOleL6Ppik5i4qQs?=
 =?us-ascii?Q?ifan2LokW7SNEO16+kQ7Pgl4WT2aamUDLfeZUudT7LDN1r/w34u3UAIuKJ4t?=
 =?us-ascii?Q?wrtDUEszZ7hE3UiYwrDXrf5EhenQuL722m1u/bBXTGZ58rCb1LfkhBG+MiQd?=
 =?us-ascii?Q?6MPpCVGkBaDBp/5QyFDXWDhnBL/lu8L4EEcpOqcU4fJsJwwkw1pNysfZ8Gsk?=
 =?us-ascii?Q?oe2bM4OlD+VEur+6HcrF9wNjWohilglhNCaKkeiegmmOGtmp3qaW+l/A/qQw?=
 =?us-ascii?Q?+B/E7BIE3t03oyzHXT0C/v4seBbZsiyPBRJyw+Nt1M5in1a7QV36Xrz38WOx?=
 =?us-ascii?Q?xm13AFHqhwiWDTKRx97WFytDJVP6PS9idufUMGwNJhFc9osV39gWlRv/gsEr?=
 =?us-ascii?Q?6ZxiN4PgJ0m0Faqm5zDW10wkoP0e8Tdx9SaAilwDSDlRrb75+REnLyEJ1rHW?=
 =?us-ascii?Q?e0em8FoGCj2FmmaLy+Sb+kwGqbwtyPZUc6wM7va3u3iZW0hyZoqPShYRWMly?=
 =?us-ascii?Q?7fwBRnYf0Oqn6kIkdhVejjAbc8QaYZIEcqnIWc/xfkLPrr1NlPgsBhEJ5wTf?=
 =?us-ascii?Q?JMfrNDH9oFmvesywWBj6W67Wbm2jlZwjnCS8jsOJGwh1tmv38cEu1wfSeDul?=
 =?us-ascii?Q?IVm2UCopF7XPL7dxLgEVofJr65fOQOqpqWqWZQqyvIg3yZ83R/WjoabQC5dG?=
 =?us-ascii?Q?P/yeIbKvZAIp0WkE905a2LPj?=
X-OriginatorOrg: amd.com
X-MS-Exchange-CrossTenant-Network-Message-Id: e12d8585-2fef-4206-0ea1-08d97e9e867d
X-MS-Exchange-CrossTenant-AuthSource: BN8PR12MB3108.namprd12.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 23 Sep 2021 14:29:17.7829
 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 3dd8961f-e488-4e60-8e11-a82d994e183d
X-MS-Exchange-CrossTenant-MailboxType: HOSTED
X-MS-Exchange-CrossTenant-UserPrincipalName: BWU4ZSm1Svft4YtB9cI+mrbxPsonKXQSwt5cSJ4StnxJKNMJq7QyC0Ml1LJNc+A+7h/VO9TcHfgo66FIgcgk3g==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: BN6PR1201MB0020
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Wed, Sep 22, 2021 at 03:36:20PM -0400, Mukul Joshi wrote:
> On Aldebaran, GPU driver will handle bad page retirement
> even though UMC is host managed. As a result, register a
> bad page retirement handler on the mce notifier chain to
> retire bad pages on Aldebaran.
> 

I think this should state that the driver will do page retirement for
GPU-managed memory. As written, it implies that the driver do page retirement
in general for the system.

...

> +
> +static int amdgpu_bad_page_notifier(struct notifier_block *nb,
> +				    unsigned long val, void *data)
> +{
> +	struct mce *m = (struct mce *)data;
> +	struct amdgpu_device *adev = NULL;
> +	uint32_t gpu_id = 0;
> +	uint32_t umc_inst = 0;
> +	uint32_t ch_inst, channel_index = 0;
> +	struct ras_err_data err_data = {0, 0, 0, NULL};
> +	struct eeprom_table_record err_rec;
> +	uint64_t retired_page;
> +
> +	/*
> +	 * If the error was generated in UMC_V2, which belongs to GPU UMCs,
> +	 * and error occurred in DramECC (Extended error code = 0) then only
> +	 * process the error, else bail out.
> +	 */
> +	if (!m || !((smca_get_bank_type(m->bank) == SMCA_UMC_V2) &&
> +		    (XEC(m->status, 0x1f) == 0x0)))

The MCA_STATUS[ErrorCodeExt] field is bits [21:16], so the mask should be
0x3f.

> +		return NOTIFY_DONE;
> +
> +	/*
> +	 * If it is correctable error, return.
> +	 */
> +	if (mce_is_correctable(m))
> +		return NOTIFY_OK;

Shouldn't this be "NOTIFY_DONE" if "don't care" about this error?

Thanks,
Yazen