All of lore.kernel.org
 help / color / mirror / Atom feed
From: Olga Kornievskaia <aglo@umich.edu>
To: Chuck Lever <chuck.lever@oracle.com>
Cc: linux-nfs <linux-nfs@vger.kernel.org>
Subject: Re: [PATCH v1 01/11] NFSD: Shrink size of struct nfsd4_copy_notify
Date: Mon, 25 Jul 2022 10:19:33 -0400	[thread overview]
Message-ID: <CAN-5tyGJFpBum6ZtO=9r0o7yNf1tsgNSEZBPJWj_qJUyLNtBWQ@mail.gmail.com> (raw)
In-Reply-To: <165852114280.11403.7277687995924103645.stgit@manet.1015granger.net>

Hi Chuck,

A general question: what's the rule for deciding whether to allocate
statically or dynamically? I thought that "small" structures it's
better to preallocate (statically) for performance reasons. Or is the
idea here that copy_notify/copy are rare operations that instead they
should be allocated dynamically and so that other operations doesn't
consume more memory in nfsd4_op structure?

On Fri, Jul 22, 2022 at 4:35 PM Chuck Lever <chuck.lever@oracle.com> wrote:
>
> struct nfsd4_copy_notify is part of struct nfsd4_op, which resides
> in an 8-element array.
>
> sizeof(struct nfsd4_op):
> Before: /* size: 2208, cachelines: 35, members: 5 */
> After:  /* size: 1696, cachelines: 27, members: 5 */
>
> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
> ---
>  fs/nfsd/nfs4proc.c |    4 ++--
>  fs/nfsd/nfs4xdr.c  |   12 ++++++++++--
>  fs/nfsd/xdr4.h     |    4 ++--
>  3 files changed, 14 insertions(+), 6 deletions(-)
>
> diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
> index 3895eb52d2b1..22c5ccb83d20 100644
> --- a/fs/nfsd/nfs4proc.c
> +++ b/fs/nfsd/nfs4proc.c
> @@ -1953,9 +1953,9 @@ nfsd4_copy_notify(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
>         /* For now, only return one server address in cpn_src, the
>          * address used by the client to connect to this server.
>          */
> -       cn->cpn_src.nl4_type = NL4_NETADDR;
> +       cn->cpn_src->nl4_type = NL4_NETADDR;
>         status = nfsd4_set_netaddr((struct sockaddr *)&rqstp->rq_daddr,
> -                                &cn->cpn_src.u.nl4_addr);
> +                                &cn->cpn_src->u.nl4_addr);
>         WARN_ON_ONCE(status);
>         if (status) {
>                 nfs4_put_cpntf_state(nn, cps);
> diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
> index 358b3338c4cc..335431199077 100644
> --- a/fs/nfsd/nfs4xdr.c
> +++ b/fs/nfsd/nfs4xdr.c
> @@ -1952,10 +1952,17 @@ nfsd4_decode_copy_notify(struct nfsd4_compoundargs *argp,
>  {
>         __be32 status;
>
> +       cn->cpn_src = svcxdr_tmpalloc(argp, sizeof(*cn->cpn_src));
> +       if (cn->cpn_src == NULL)
> +               return nfserrno(-ENOMEM);       /* XXX: jukebox? */
> +       cn->cpn_dst = svcxdr_tmpalloc(argp, sizeof(*cn->cpn_dst));
> +       if (cn->cpn_dst == NULL)
> +               return nfserrno(-ENOMEM);       /* XXX: jukebox? */
> +
>         status = nfsd4_decode_stateid4(argp, &cn->cpn_src_stateid);
>         if (status)
>                 return status;
> -       return nfsd4_decode_nl4_server(argp, &cn->cpn_dst);
> +       return nfsd4_decode_nl4_server(argp, cn->cpn_dst);
>  }
>
>  static __be32
> @@ -4898,7 +4905,8 @@ nfsd4_encode_copy_notify(struct nfsd4_compoundres *resp, __be32 nfserr,
>
>         *p++ = cpu_to_be32(1);
>
> -       return nfsd42_encode_nl4_server(resp, &cn->cpn_src);
> +       nfserr = nfsd42_encode_nl4_server(resp, cn->cpn_src);
> +       return nfserr;
>  }
>
>  static __be32
> diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h
> index 6e6a89008ce1..f253fc3f4708 100644
> --- a/fs/nfsd/xdr4.h
> +++ b/fs/nfsd/xdr4.h
> @@ -595,13 +595,13 @@ struct nfsd4_offload_status {
>  struct nfsd4_copy_notify {
>         /* request */
>         stateid_t               cpn_src_stateid;
> -       struct nl4_server       cpn_dst;
> +       struct nl4_server       *cpn_dst;
>
>         /* response */
>         stateid_t               cpn_cnr_stateid;
>         u64                     cpn_sec;
>         u32                     cpn_nsec;
> -       struct nl4_server       cpn_src;
> +       struct nl4_server       *cpn_src;
>  };
>
>  struct nfsd4_op {
>
>

  reply	other threads:[~2022-07-25 14:19 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-07-22 20:18 [PATCH v1 00/11] Put struct nfsd4_copy on a diet Chuck Lever
2022-07-22 20:19 ` [PATCH v1 01/11] NFSD: Shrink size of struct nfsd4_copy_notify Chuck Lever
2022-07-25 14:19   ` Olga Kornievskaia [this message]
2022-07-25 14:36     ` Chuck Lever III
2022-07-22 20:19 ` [PATCH v1 02/11] NFSD: Shrink size of struct nfsd4_copy Chuck Lever
2022-07-24  6:30   ` kernel test robot
2022-07-22 20:19 ` [PATCH v1 03/11] NFSD: Reorder the fields in struct nfsd4_op Chuck Lever
2022-07-22 20:19 ` [PATCH v1 04/11] NFSD: Make nfs4_put_copy() static Chuck Lever
2022-07-22 20:19 ` [PATCH v1 05/11] NFSD: Make boolean fields in struct nfsd4_copy into atomic bit flags Chuck Lever
2022-07-22 20:19 ` [PATCH v1 06/11] NFSD: Refactor nfsd4_cleanup_inter_ssc() (1/2) Chuck Lever
2022-07-22 20:19 ` [PATCH v1 07/11] NFSD: Refactor nfsd4_cleanup_inter_ssc() (2/2) Chuck Lever
2022-07-22 20:19 ` [PATCH v1 08/11] NFSD: Refactor nfsd4_do_copy() Chuck Lever
2022-07-22 20:19 ` [PATCH v1 09/11] NFSD: Remove kmalloc from nfsd4_do_async_copy() Chuck Lever
2022-07-22 20:19 ` [PATCH v1 10/11] NFSD: Add nfsd4_send_cb_offload() Chuck Lever
2022-07-22 20:20 ` [PATCH v1 11/11] NFSD: Move copy offload callback arguments into a separate structure Chuck Lever
2022-07-26 19:45 ` [PATCH v1 00/11] Put struct nfsd4_copy on a diet Olga Kornievskaia
2022-07-27 16:18   ` Olga Kornievskaia
2022-07-27 16:38     ` dai.ngo
2022-07-27 17:15     ` Chuck Lever III
2022-07-27 17:52       ` Olga Kornievskaia
2022-07-27 18:04         ` Chuck Lever III
2022-07-27 18:21           ` Olga Kornievskaia
2022-07-27 18:48             ` Olga Kornievskaia
2022-07-27 19:23               ` dai.ngo

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='CAN-5tyGJFpBum6ZtO=9r0o7yNf1tsgNSEZBPJWj_qJUyLNtBWQ@mail.gmail.com' \
    --to=aglo@umich.edu \
    --cc=chuck.lever@oracle.com \
    --cc=linux-nfs@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.