From: Mel Gorman <mgorman@techsingularity.net>
To: Jesper Dangaard Brouer <brouer@redhat.com>
Cc: linux-mm@kvack.org, chuck.lever@oracle.com,
netdev@vger.kernel.org, linux-nfs@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH RFC net-next 3/3] mm: make zone->free_area[order] access faster
Date: Thu, 25 Feb 2021 11:28:49 +0000 [thread overview]
Message-ID: <20210225112849.GM3697@techsingularity.net> (raw)
In-Reply-To: <161419301128.2718959.4838557038019199822.stgit@firesoul>
As a side-node, I didn't pick up the other patches as there is review
feedback and I didn't have strong opinions either way. Patch 3 is curious
though, it probably should be split out and sent separetly but still;
On Wed, Feb 24, 2021 at 07:56:51PM +0100, Jesper Dangaard Brouer wrote:
> Avoid multiplication (imul) operations when accessing:
> zone->free_area[order].nr_free
>
> This was really tricky to find. I was puzzled why perf reported that
> rmqueue_bulk was using 44% of the time in an imul operation:
>
> ??? del_page_from_free_list():
> 44,54 ??? e2: imul $0x58,%rax,%rax
>
> This operation was generated (by compiler) because the struct free_area have
> size 88 bytes or 0x58 hex. The compiler cannot find a shift operation to use
> and instead choose to use a more expensive imul, to find the offset into the
> array free_area[].
>
> The patch align struct free_area to a cache-line, which cause the
> compiler avoid the imul operation. The imul operation is very fast on
> modern Intel CPUs. To help fast-path that decrement 'nr_free' move the
> member 'nr_free' to be first element, which saves one 'add' operation.
>
> Looking up instruction latency this exchange a 3-cycle imul with a
> 1-cycle shl, saving 2-cycles. It does trade some space to do this.
>
> Used: gcc (GCC) 9.3.1 20200408 (Red Hat 9.3.1-2)
>
I'm having some trouble parsing this and matching it to the patch itself.
First off, on my system (x86-64), the size of struct free area is 72,
not 88 bytes. For either size, cache-aligning the structure is a big
increase in the struct size.
struct free_area {
struct list_head free_list[4]; /* 0 64 */
/* --- cacheline 1 boundary (64 bytes) --- */
long unsigned int nr_free; /* 64 8 */
/* size: 72, cachelines: 2, members: 2 */
/* last cacheline: 8 bytes */
};
Are there other patches in the tree? What does pahole say?
With gcc-9, I'm also not seeing the imul instruction outputted like you
described in rmqueue_pcplist which inlines rmqueue_bulk. At the point
where it calls get_page_from_free_area, it's using shl for the page list
operation. This might be a compiler glitch but given that free_area is a
different size, I'm less certain and wonder if something else is going on.
Finally, moving nr_free to the end and cache aligning it will make the
started of each free_list cache-aligned because of its location in the
struct zone so what purpose does __pad_to_align_free_list serve?
--
Mel Gorman
SUSE Labs
next prev parent reply other threads:[~2021-02-25 11:31 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-24 10:26 [RFC PATCH 0/3] Introduce a bulk order-0 page allocator for sunrpc Mel Gorman
2021-02-24 10:26 ` [PATCH 1/3] SUNRPC: Set rq_page_end differently Mel Gorman
2021-02-24 10:26 ` [PATCH 2/3] mm, page_alloc: Add a bulk page allocator Mel Gorman
2021-02-24 10:26 ` [PATCH 3/3] SUNRPC: Refresh rq_pages using " Mel Gorman
2021-02-24 11:27 ` [RFC PATCH 0/3] Introduce a bulk order-0 page allocator for sunrpc Jesper Dangaard Brouer
2021-02-24 11:55 ` Mel Gorman
2021-02-24 13:20 ` Chuck Lever
2021-02-24 18:56 ` [PATCH RFC net-next 0/3] Use bulk order-0 page allocator API for page_pool Jesper Dangaard Brouer
2021-02-24 18:56 ` [PATCH RFC net-next 1/3] net: page_pool: refactor dma_map into own function page_pool_dma_map Jesper Dangaard Brouer
2021-02-24 20:11 ` Ilias Apalodimas
2021-02-24 18:56 ` [PATCH RFC net-next 2/3] net: page_pool: use alloc_pages_bulk in refill code path Jesper Dangaard Brouer
2021-02-24 20:15 ` Ilias Apalodimas
2021-02-26 14:31 ` Jesper Dangaard Brouer
2021-02-24 18:56 ` [PATCH RFC net-next 3/3] mm: make zone->free_area[order] access faster Jesper Dangaard Brouer
2021-02-25 11:28 ` Mel Gorman [this message]
2021-02-25 15:16 ` Jesper Dangaard Brouer
2021-02-25 15:38 ` Mel Gorman
2021-02-26 14:34 ` Jesper Dangaard Brouer
2021-03-01 13:29 ` [PATCH RFC V2 net-next 0/2] Use bulk order-0 page allocator API for page_pool Jesper Dangaard Brouer
2021-03-01 13:29 ` [PATCH RFC V2 net-next 1/2] net: page_pool: refactor dma_map into own function page_pool_dma_map Jesper Dangaard Brouer
2021-03-01 13:29 ` [PATCH RFC V2 net-next 2/2] net: page_pool: use alloc_pages_bulk in refill code path Jesper Dangaard Brouer
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210225112849.GM3697@techsingularity.net \
--to=mgorman@techsingularity.net \
--cc=brouer@redhat.com \
--cc=chuck.lever@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-nfs@vger.kernel.org \
--cc=netdev@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).