Re: [PATCH 1/1] net/rds: Track user mapped pages through special API

[santosh.shilimkar@oracle.com]

On 2/12/20 9:51 AM, Leon Romanovsky wrote:
> On Wed, Feb 12, 2020 at 09:31:51AM -0800, santosh.shilimkar@oracle.com wrote:
>> On 2/11/20 7:03 PM, John Hubbard wrote:
>>> From: Leon Romanovsky <leonro@mellanox.com>
>>>
>>> Convert net/rds to use the newly introduces pin_user_pages() API,
>>> which properly sets FOLL_PIN. Setting FOLL_PIN is now required for
>>> code that requires tracking of pinned pages.
>>>
>>> Note that this effectively changes the code's behavior: it now
>>> ultimately calls set_page_dirty_lock(), instead of set_page_dirty().
>>> This is probably more accurate.
>>>
>>> As Christoph Hellwig put it, "set_page_dirty() is only safe if we are
>>> dealing with a file backed page where we have reference on the inode it
>>> hangs off." [1]
>>>
>>> [1] https://urldefense.com/v3/__https://lore.kernel.org/r/20190723153640.GB720@lst.de__;!!GqivPVa7Brio!OJHuecs9Iup5ig3kQBi_423uMMuskWhBQAdOICrY3UQ_ZfEaxt9ySY7E8y32Q7pk5tByyA$
>>>
>>> Cc: Hans Westgaard Ry <hans.westgaard.ry@oracle.com>
>>> Cc: Santosh Shilimkar <santosh.shilimkar@oracle.com>
>>> Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
>>> Signed-off-by: John Hubbard <jhubbard@nvidia.com>
>>> ---
>> Change looks fine to me. Just on safer side, we will try
>> to test this change with regression suite to make sure it
>> works as expected.
> 
> Thanks Santosh,
> I wrote this patch before John's series was merged into the tree,
> but back then, Hans tested it and it worked, hope that it still works. :)
> 
I see. Wasn't aware of it. In that case, its should be fine.

Regards,
Santosh