Message-ID: <088e40fd-3fc7-77dd-a3de-0a2b097d3717@kernel.dk>
Date:   Mon, 30 Jan 2023 15:11:39 -0700
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux aarch64; rv:102.0) Gecko/20100101
 Thunderbird/102.6.0
Subject: Re: [GIT PULL] iov_iter: Improve page extraction (pin or just list)
Content-Language: en-US
To:     John Hubbard <jhubbard@nvidia.com>,
        David Howells <dhowells@redhat.com>
Cc:     Al Viro <viro@zeniv.linux.org.uk>,
        Christoph Hellwig <hch@infradead.org>,
        Matthew Wilcox <willy@infradead.org>, Jan Kara <jack@suse.cz>,
        David Hildenbrand <david@redhat.com>,
        Jason Gunthorpe <jgg@nvidia.com>,
        Logan Gunthorpe <logang@deltatee.com>,
        Jeff Layton <jlayton@kernel.org>, linux-block@vger.kernel.org,
        linux-fsdevel@vger.kernel.org, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org
References: <3351099.1675077249@warthog.procyon.org.uk>
 <fd0003a0-a133-3daf-891c-ba7deafad768@kernel.dk>
 <f57ee72f-38e9-6afa-182f-2794638eadcb@kernel.dk>
 <e8480b18-08af-d101-a721-50d213893492@kernel.dk>
 <e68c5cab-c3a6-1872-98fa-9f909f23be79@nvidia.com>
From:   Jens Axboe <axboe@kernel.dk>
In-Reply-To: <e68c5cab-c3a6-1872-98fa-9f909f23be79@nvidia.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 1/30/23 3:02?PM, John Hubbard wrote:
> On 1/30/23 13:57, Jens Axboe wrote:
>>> This does cause about a 2.7% regression for me, using O_DIRECT on a raw
>>> block device. Looking at a perf diff, here's the top:
>>>
>>>                 +2.71%  [kernel.vmlinux]  [k] mod_node_page_state
>>>                 +2.22%  [kernel.vmlinux]  [k] iov_iter_extract_pages
>>>
>>> and these two are gone:
>>>
>>>       2.14%             [kernel.vmlinux]  [k] __iov_iter_get_pages_alloc
>>>       1.53%             [kernel.vmlinux]  [k] iov_iter_get_pages
>>>
>>> rest is mostly in the noise, but mod_node_page_state() sticks out like
>>> a sore thumb. They seem to be caused by the node stat accounting done
>>> in gup.c for FOLL_PIN.
>>
>> Confirmed just disabling the node_stat bits in mm/gup.c and now the
>> performance is back to the same levels as before.
>>
>> An almost 3% regression is a bit hard to swallow...
> 
> This is something that we say when adding pin_user_pages_fast(),
> yes. I doubt that I can quickly find the email thread, but we
> measured it and weren't immediately able to come up with a way
> to make it faster.
> 
> At this point, it's a good time to consider if there is any
> way to speed it up. But I wanted to confirm that you're absolutely
> right: the measurement sounds about right, and that's also the
> hotspot that we say, too.

From spending all of 5 minutes on this, it must be due to exceeding the
pcp stat_threashold, as we then end up doing two atomic_long_adds().
Looking at proc, looks like it's 108. And with this test, then we're
hitting that slow path ~80k/second. Uhm...

-- 
Jens Axboe