linux-kernel - Re: [PATCH] ceph: Reorder fields in 'struct ceph_snapid

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <f6b869ea-979c-efda-d454-8dc688d1986b@redhat.com>
Date:   Fri, 28 Apr 2023 08:53:52 +0800
From:   Xiubo Li <xiubli@...hat.com>
To:     Christophe JAILLET <christophe.jaillet@...adoo.fr>,
        Ilya Dryomov <idryomov@...il.com>,
        Jeff Layton <jlayton@...nel.org>
Cc:     linux-kernel@...r.kernel.org, kernel-janitors@...r.kernel.org,
        ceph-devel@...r.kernel.org
Subject: Re: [PATCH] ceph: Reorder fields in 'struct ceph_snapid_map'


On 4/28/23 02:05, Christophe JAILLET wrote:
> Group some variables based on their sizes to reduce holes.
> On x86_64, this shrinks the size of 'struct ceph_snapid_map' from 72 to 64
> bytes.
>
> When such a structure is allocated, because of the way memory allocation
> works, when 72 bytes were requested, 96 bytes were allocated.
>
> So, on x86_64, this change saves 32 bytes per allocation and has the
> structure fit in a single cacheline.
>
> Signed-off-by: Christophe JAILLET <christophe.jaillet@...adoo.fr>
> ---
> Using pahole
>
> Before:
> ======
> struct ceph_snapid_map {
> 	struct rb_node             node __attribute__((__aligned__(8))); /*     0    24 */
> 	struct list_head           lru;                  /*    24    16 */
> 	atomic_t                   ref;                  /*    40     4 */
>
> 	/* XXX 4 bytes hole, try to pack */
>
> 	u64                        snap;                 /*    48     8 */
> 	dev_t                      dev;                  /*    56     4 */
>
> 	/* XXX 4 bytes hole, try to pack */
>
> 	/* --- cacheline 1 boundary (64 bytes) --- */
> 	long unsigned int          last_used;            /*    64     8 */
>
> 	/* size: 72, cachelines: 2, members: 6 */
> 	/* sum members: 64, holes: 2, sum holes: 8 */
> 	/* forced alignments: 1 */
> 	/* last cacheline: 8 bytes */
> } __attribute__((__aligned__(8)));
> ---
>   fs/ceph/mds_client.h | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
> index 0598faa50e2e..2328dbda5ab6 100644
> --- a/fs/ceph/mds_client.h
> +++ b/fs/ceph/mds_client.h
> @@ -355,8 +355,8 @@ struct ceph_snapid_map {
>   	struct rb_node node;
>   	struct list_head lru;
>   	atomic_t ref;
> -	u64 snap;
>   	dev_t dev;
> +	u64 snap;
>   	unsigned long last_used;
>   };
>   

This looks good to me. Thanks.

Will apply it to the testing branch.

- Xiubo