Received-SPF: pass (google.com: domain of linux-kernel+bounces-174958-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
References: <fa2d39cf-b0df-4674-979d-b775d5077bce@deltatee.com>
 <20240509174204.GV4718@ziepe.ca>
User-agent: mu4e 1.10.8; emacs 29.1
From: Alistair Popple <apopple@nvidia.com>
To: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Logan Gunthorpe <logang@deltatee.com>, Martin Oliveira
 <Martin.Oliveira@eideticom.com>, Christoph Hellwig <hch@lst.de>, Dan
 Williams <dan.j.williams@intel.com>, LKML <linux-kernel@vger.kernel.org>,
 "linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>,
 "linux-mm@kvack.org" <linux-mm@kvack.org>
Subject: Re: P2PDMA in Userspace RDMA
Date: Thu, 09 May 2024 13:48:31 -0700
In-reply-to: <20240509174204.GV4718@ziepe.ca>
Message-ID: <87cypuvh2i.fsf@nvdebian.thelocal>
Content-Type: text/plain
Precedence: bulk
MIME-Version: 1.0
X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-MessageData-0:
	=?us-ascii?Q?MnMmbHkPzFwhjZ2KMi6sAO35Xbl8JQJRkSg0WBOclBSTua6lRXq5bxrvfIXk?=
 =?us-ascii?Q?vmzOhAQgKO9FqMJs4GCs2NUyKRNzLnekOfA6BI4h2Gz44msLSsZ1ggBv+HC+?=
 =?us-ascii?Q?hVdkM4ULqkMjZ5HX+Pi5Byw0U+MkdBNVcpEA0mIYFDRcKqPuUappUBURkGCP?=
 =?us-ascii?Q?pJN1UHYKgpXT6AVWoWh9kpGERCVXgOH1XhD7iyXedDuUbGWmy0GOVwfjV3GP?=
 =?us-ascii?Q?Qq+pigVb8UBPAPqLtHa6OHrOmDAYKuwhECUpFjwrXCPXNG5EiYb1WEhmQHWC?=
 =?us-ascii?Q?WMlH2Q67Xnq82vnyctUfPOT8uKQZNJqcCbDIEgjQIMOI63cEXTgol8Hb5XAh?=
 =?us-ascii?Q?LJnRZvtMc9Hjt/3SoIxbcsfbC+hNJL4DZ168dRVqHl77FuNY08g6B4cXzPsl?=
 =?us-ascii?Q?8bh24bTT2XK/Xr5QcO9ERuz1lyd+d8+31yQDUs/YXxO8EEu5rhMm6wG/yk0h?=
 =?us-ascii?Q?uArJkhyhW0vTYfb3QxITDOeDG2rOnh1PJE3nN1webgXRp1UAsW82vudoQB30?=
 =?us-ascii?Q?bxom2qSJd2jm8LeuwktYztteBV5oiUea3X32CSn+vlVWzK30YDNGYfDnKzHp?=
 =?us-ascii?Q?3qJmt2ZhqFnve7/2ZW7QApdIxHFwykm/CCxdwSCL9rq1uHpenUs+UCrsGjJU?=
 =?us-ascii?Q?gx42RGBj5rqrGulUeOe+6Mzy853q0zK/1q7i4eBxJ68QU3IZ3JZuB/Ex6OzM?=
 =?us-ascii?Q?/aUDPo9Vo8yv/cX0mlbsyPAt1xnaqd59RPJvpdE3/JIC8DyHDN2mvhIDh6bq?=
 =?us-ascii?Q?mGQnFUZxz9EZeqpEqQhoEoKHihrQcvp/GBUGk9VTZ8xUtT6Ttqfg8KrMwTdr?=
 =?us-ascii?Q?MCNJ68Jfa5CMZAF5XycaXFDZyv1OGSM/h7E7vrl0E/Qq9Rakjzjd/DEcsqJ1?=
 =?us-ascii?Q?hhB/I8lvSdYY0KknzwQTVqGcL32/TOZbrAKhlRD487huCUfmq1LFyicKF0Do?=
 =?us-ascii?Q?MSqCDJW8rN9d6oeG/ljv9YYYROeo/Ttrls36os2saDXTYspmFzCQc7n+ujGJ?=
 =?us-ascii?Q?vWkKBGZBoc4vpCrgV+W22bTTS0CCR9lgPUwbkyCtoweSq9yJPiOG+1XWwY8r?=
 =?us-ascii?Q?om4Xp1IRvnH6wShDdyvwh6zgfBRDHdYaflCQVv8eEn0TjylSDiOkEBuy3q6h?=
 =?us-ascii?Q?cN/eqc9rbzEc2ffWr65g7SX27HPAwHzY8TNirOjn/TKMiM6KVlLSpihQVyIn?=
 =?us-ascii?Q?x4ctZ2h0LhD9iLAXtuLQdKYUGCbBUqfMc6dG9pcMv3G59EicSUH2HVyT78fY?=
 =?us-ascii?Q?BfjXn/BiqeDLlOYqfzfB4hj6SbjYtiy0EIKT44HJJsph5D0K5xkSYQVO8dw/?=
 =?us-ascii?Q?T0yox23yWiZXhhLiu3LKaqkD7iLGRdIweAxNCIxIJAJQ4dZ5hosgyXM5adhW?=
 =?us-ascii?Q?FDBKm5uw/sDaR/rAjC+o+5V4utekCkmVIGRqObL3Jzrj1tSirWIO5CTIz2fa?=
 =?us-ascii?Q?zy219FysG1JIc9HiOg05yD5vWEMnwCLgI+/Dr5asjJblvBfZaYmfH8x1Y5E8?=
 =?us-ascii?Q?UiuSPg5oaSleVC+sZT5PucDiUMVlUcEvfVzOFNA8dCqeP6Z/OI+npW/FnYLv?=
 =?us-ascii?Q?F9LXbOKarsBFBfhi9iSZd7JsTlg91UFobxAlOWnb?=
X-OriginatorOrg: Nvidia.com
X-MS-Exchange-CrossTenant-Network-Message-Id: 7ee3c650-0993-42c2-1f45-08dc706abbe9
X-MS-Exchange-CrossTenant-AuthSource: DS0PR12MB7726.namprd12.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 09 May 2024 20:58:10.6060
 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 43083d15-7273-40c1-b7db-39efd9ccc17a
X-MS-Exchange-CrossTenant-MailboxType: HOSTED
X-MS-Exchange-CrossTenant-UserPrincipalName: 3ybgJm0wThpGbkCHBtKHrueOCZwwVGfdVKWJcv44DtqBERYiXU6T8Tn/ZTK5mrBuJ+T//bDW09uX1WeMWlyo1A==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: SJ0PR12MB7007


Jason Gunthorpe <jgg@ziepe.ca> writes:

> On Thu, May 09, 2024 at 11:31:33AM -0600, Logan Gunthorpe wrote:
>> Hi Jason,
>> 
>> We've become interested again in enabling P2PDMA transactions with
>> userspace RDMA and the NVMe CMBs we are already exporting to userspace
>> from our previous work.
>> 
>> Enabling FOLL_PCI_P2PDMA in ib_umem_get is almost a trivial change, but
>> there are two issues holding us back.
>> 
>> The biggest issue is that we disallowed FOLL_LONGTERM with
>> FOLL_PCI_P2PDMA out of concern that P2PDMA had the same problem as
>> fs-dax.
>
> Yeah, it was not a great outcome of that issue.

I recently came across this trying to do something similar with io_uring
to mapped P2PDMA pages for testing purposes. I commented out the
FOLL_PCI_P2PDMA check but it hung/stalled during process teardown so
there might be something there.

>> See [1] to review the discussion from 2 years ago. However, in
>> trying to understand the problem again, I'm not sure that concern was
>> valid. In P2PDMA, unmap_mapping_range() is strictly only called on
>> driver unbind when everything is getting torn down[2]. The next thing
>> that happens immediately after the unmap is the tear down of the pgmap
>> which drops the elevated reference on the pages and waits for all page's
>> reference counts to go back to zero. This will effectively wait until
>> all longterm pins involving the memory have been released. This can
>> cause a hang on unbind but, in your words, its "annoying not critical".
>
> Yes
>
> But you are looking at the code as it is right now, and stuff has been
> quitely fixed with the pgmap refcount area since. I think it is
> probably good now. IIRC it was pushed over the finish line when the
> ZONE_DEVICE/PRIVATE pages were converted to have normal reference
> counting.

But P2P DMA pages are still in the off-by-one reference counting
scheme. My RFR series[1] (aims) to fix that, although the pgmap
refcounting scheme needs a closer look because I think doing the page
refcounting properly allows some cleanups there.

[1] - https://lore.kernel.org/linux-mm/cover.fe275e9819458a4bbb9451b888cafb88af8867d4.1712796818.git-series.apopple@nvidia.com/

> If p2p is following the new ZONE_DEVICE scheme then it should be fine.
>
> It would be good to read over Alistair's latest series fixing up fsdax
> refcounts to see if anything pops out as problematic specifically with
> the P2P case.
>
> Otherwise a careful check through is probably all that is needed.
>
>> The other issue we hit when enabling this feature is the check for
>> vma_needs_dirty_tracking() in writable_file_mapping_allowed() during the
>> gup flow. This hits because the p2pdma code is using the common
>> sysfs/kernfs infrastructure to create the VMA which installs a
>> page_mkwrite operator()[4] to change the file update time on write. 
>
> Ah.
>
>> I don't think this feature really makes any sense for the P2PDMA
>> sysfs file which is really operating as an allocator in userspace --
>> the time on the file does not really need to reflect the last write
>> of some process that wrote to memory allocated using it. 
>
> Right, you shouldn't have mkwrite for these pages.
>
> Jason