From: Dennis Zhou <dennis@kernel.org>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Yu Ma <yu.ma@intel.com>,
dennis@kernel.org, Liam.Howlett@oracle.com,
dan.j.williams@intel.com, dave.hansen@intel.com,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
lipeng.zhu@intel.com, pan.deng@intel.com, shakeelb@google.com,
tianyou.li@intel.com, tim.c.chen@intel.com,
tim.c.chen@linux.intel.com
Subject: Re: [PATCH v3] percpu-internal/pcpu_chunk: Re-layout pcpu_chunk structure to reduce false sharing
Date: Mon, 12 Jun 2023 14:55:44 -0700 [thread overview]
Message-ID: <ZIeUYNd8sAVm1xE8@V92F7Y9K0C.lan> (raw)
In-Reply-To: <20230612144331.b1d069bce4ba3800fdd62738@linux-foundation.org>
Hi Andrew,
On Mon, Jun 12, 2023 at 02:43:31PM -0700, Andrew Morton wrote:
> On Fri, 9 Jun 2023 23:07:30 -0400 Yu Ma <yu.ma@intel.com> wrote:
>
> > When running UnixBench/Execl throughput case, false sharing is observed
> > due to frequent read on base_addr and write on free_bytes, chunk_md.
> >
> > UnixBench/Execl represents a class of workload where bash scripts
> > are spawned frequently to do some short jobs. It will do system call on
> > execl frequently, and execl will call mm_init to initialize mm_struct
> > of the process. mm_init will call __percpu_counter_init for
> > percpu_counters initialization. Then pcpu_alloc is called to read
> > the base_addr of pcpu_chunk for memory allocation. Inside pcpu_alloc,
> > it will call pcpu_alloc_area to allocate memory from a specified chunk.
> > This function will update "free_bytes" and "chunk_md" to record the
> > rest free bytes and other meta data for this chunk. Correspondingly,
> > pcpu_free_area will also update these 2 members when free memory.
> > Call trace from perf is as below:
> > + 57.15% 0.01% execl [kernel.kallsyms] [k] __percpu_counter_init
> > + 57.13% 0.91% execl [kernel.kallsyms] [k] pcpu_alloc
> > - 55.27% 54.51% execl [kernel.kallsyms] [k] osq_lock
> > - 53.54% 0x654278696e552f34
> > main
> > __execve
> > entry_SYSCALL_64_after_hwframe
> > do_syscall_64
> > __x64_sys_execve
> > do_execveat_common.isra.47
> > alloc_bprm
> > mm_init
> > __percpu_counter_init
> > pcpu_alloc
> > - __mutex_lock.isra.17
> >
> > In current pcpu_chunk layout, ‘base_addr’ is in the same cache line
> > with ‘free_bytes’ and ‘chunk_md’, and ‘base_addr’ is at the
> > last 8 bytes. This patch moves ‘bound_map’ up to ‘base_addr’,
> > to let ‘base_addr’ locate in a new cacheline.
> >
> > With this change, on Intel Sapphire Rapids 112C/224T platform,
> > based on v6.4-rc4, the 160 parallel score improves by 24%.
>
> Well that's nice.
>
> >
> > ...
> >
> > --- a/mm/percpu-internal.h
> > +++ b/mm/percpu-internal.h
> > @@ -41,10 +41,17 @@ struct pcpu_chunk {
> > struct list_head list; /* linked to pcpu_slot lists */
> > int free_bytes; /* free bytes in the chunk */
> > struct pcpu_block_md chunk_md;
> > - void *base_addr; /* base address of this chunk */
> > + unsigned long *bound_map; /* boundary map */
> > +
> > + /*
> > + * base_addr is the base address of this chunk.
> > + * To reduce false sharing, current layout is optimized to make sure
> > + * base_addr locate in the different cacheline with free_bytes and
> > + * chunk_md.
> > + */
> > + void *base_addr ____cacheline_aligned_in_smp;
> >
> > unsigned long *alloc_map; /* allocation map */
> > - unsigned long *bound_map; /* boundary map */
> > struct pcpu_block_md *md_blocks; /* metadata blocks */
> >
> > void *data; /* chunk data */
>
> This will of course consume more memory. Do we have a feel for the
> worst-case impact of this?
>
The pcpu_chunk struct is a backing data structure per chunk, so the
additional memory should not be dramatic. A chunk covers ballpark
between 64kb and 512kb memory depending on some config and boot time
stuff, so I believe the additional memory used here is nominal at best.
Working the #s on my desktop:
Percpu: 58624 kB
28 cores -> ~2.1MB of percpu memory.
At say ~128KB per chunk -> 33 chunks, generously 40 chunks.
Adding alignment might bump the chunk size ~64 bytes, so in total ~2KB
of overhead?
I believe we can do a little better to avoid eating that full padding,
so likely less than that.
Acked-by: Dennis Zhou <dennis@kernel.org>
Thanks,
Dennis
next prev parent reply other threads:[~2023-06-12 21:55 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-06-06 12:54 [PATCH] percpu-internal/pcpu_chunk: Re-layout pcpu_chunk structure to reduce false sharing Yu Ma
2023-06-06 19:21 ` Liam R. Howlett
2023-06-06 21:25 ` Dennis Zhou
2023-06-07 12:50 ` Ma, Yu
2023-06-07 14:50 ` [PATCH v2] " Yu Ma
2023-06-07 15:02 ` Ma, Yu
2023-06-09 18:20 ` Dennis Zhou
2023-06-10 0:12 ` Ma, Yu
2023-06-10 3:07 ` [PATCH v3] " Yu Ma
2023-06-12 21:43 ` Andrew Morton
2023-06-12 21:55 ` Dennis Zhou [this message]
2023-06-13 17:41 ` Ma, Yu
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=ZIeUYNd8sAVm1xE8@V92F7Y9K0C.lan \
--to=dennis@kernel.org \
--cc=Liam.Howlett@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=dan.j.williams@intel.com \
--cc=dave.hansen@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lipeng.zhu@intel.com \
--cc=pan.deng@intel.com \
--cc=shakeelb@google.com \
--cc=tianyou.li@intel.com \
--cc=tim.c.chen@intel.com \
--cc=tim.c.chen@linux.intel.com \
--cc=yu.ma@intel.com \
/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 read-only IMAP folder(s) and NNTP newsgroup(s).