Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
In-Reply-To: <871sdck9ds.fsf@notabene.neil.brown.name>
References: <152782754287.30340.4395718227884933670.stgit@noble>
 <152782824964.30340.6329146982899668633.stgit@noble> <20180602154851.pfy4wryezuhxp76v@gondor.apana.org.au>
 <87y3fvpf40.fsf@notabene.neil.brown.name> <CALx6S37AYrFoJNrdToRc84J=B4f57jaHS077UsxdEoQ-xwyCeg@mail.gmail.com>
 <87sh63pakb.fsf@notabene.neil.brown.name> <CALx6S35sf4oj=x73SfhUxi1e9-u4Stxgr9BAj5E7KKmbufntwg@mail.gmail.com>
 <87r2lmnj2c.fsf@notabene.neil.brown.name> <CALx6S36Ce-rXQMzmFYZVPGD10Bo6udvRAHiZ5gWwnzVwoTVv0w@mail.gmail.com>
 <87in6wo636.fsf@notabene.neil.brown.name> <871sdjnwkr.fsf@notabene.neil.brown.name>
 <87y3frmhyx.fsf@notabene.neil.brown.name> <CALx6S35GgUOd0dPgv7P96wNNTv5pN7fij0pcAoccqcSWZhvY7Q@mail.gmail.com>
 <871sdck9ds.fsf@notabene.neil.brown.name>
From:   Tom Herbert <tom@herbertland.com>
Date:   Thu, 14 Jun 2018 10:41:05 -0700
Message-ID: <CALx6S35Yr8th_LFoCLjEqfzti0cp7+7nKZC7KO3dzeQkU7c0nw@mail.gmail.com>
Subject: Re: [PATCH RFC v2] rhashtable: implement rhashtable_walk_peek() using rhashtable_walk_last_seen()
To:     NeilBrown <neilb@suse.com>
Cc:     Herbert Xu <herbert@gondor.apana.org.au>,
        Thomas Graf <tgraf@suug.ch>,
        Linux Kernel Network Developers <netdev@vger.kernel.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Tom Herbert <tom@quantonium.net>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Jun 11, 2018 at 7:48 PM, NeilBrown <neilb@suse.com> wrote:
>
> rhashtable_walk_last_seen() does most of the work that
> rhashtable_walk_peek() needs done, so use it and put
> it in a "static inline".
> Also update the documentation for rhashtable_walk_peek() to clarify
> the expected use case.
>
> Signed-off-by: NeilBrown <neilb@suse.com>

Acked-by: Tom Herbert <tom@quantonium.net>

> ---
>
> v2 as static-inline - suggested by Tom.
>
> Thanks,
> NeilBrown
>
>  include/linux/rhashtable.h | 29 ++++++++++++++++++++++++++++-
>  lib/rhashtable.c           | 34 ----------------------------------
>  2 files changed, 28 insertions(+), 35 deletions(-)
>
> diff --git a/include/linux/rhashtable.h b/include/linux/rhashtable.h
> index d63b472e9d50..96ebc2690027 100644
> --- a/include/linux/rhashtable.h
> +++ b/include/linux/rhashtable.h
> @@ -247,10 +247,37 @@ static inline void rhashtable_walk_start(struct rhashtable_iter *iter)
>  }
>
>  void *rhashtable_walk_next(struct rhashtable_iter *iter);
> -void *rhashtable_walk_peek(struct rhashtable_iter *iter);
>  void *rhashtable_walk_last_seen(struct rhashtable_iter *iter);
>  void rhashtable_walk_stop(struct rhashtable_iter *iter) __releases(RCU);
>
> +/**
> + * rhashtable_walk_peek - Return the next object to use in an interrupted walk
> + * @iter:      Hash table iterator
> + *
> + * Returns the "current" object or NULL when the end of the table is reached.
> + * When an rhashtable_walk is interrupted with rhashtable_walk_stop(),
> + * it is often because an object was found that could not be processed
> + * immediately, possible because there is no more space to encode details
> + * of the object (e.g. when producing a seq_file from the table).
> + * When the walk is restarted, the same object needs to be processed again,
> + * if possible.  The object might have been removed from the table while
> + * the walk was paused, so it might not be available.  In that case, the
> + * normal "next" object should be treated as "current".
> + *
> + * To support this common case, rhashtable_walk_peek() returns the
> + * appropriate object to process after an interrupted walk, either the
> + * one that was most recently returned, or if that doesn't exist - the
> + * next one.
> + *
> + * Returns -EAGAIN if resize event occurred.  In that case the iterator
> + * will rewind back to the beginning and you may continue to use it.
> + */
> +static inline void *rhashtable_walk_peek(struct rhashtable_iter *iter)
> +{
> +       return rhashtable_walk_last_seen(iter) ?:
> +               rhashtable_walk_next(iter);
> +}
> +
>  void rhashtable_free_and_destroy(struct rhashtable *ht,
>                                  void (*free_fn)(void *ptr, void *arg),
>                                  void *arg);
> diff --git a/lib/rhashtable.c b/lib/rhashtable.c
> index 45f2554399a5..354275037df3 100644
> --- a/lib/rhashtable.c
> +++ b/lib/rhashtable.c
> @@ -915,40 +915,6 @@ void *rhashtable_walk_next(struct rhashtable_iter *iter)
>  }
>  EXPORT_SYMBOL_GPL(rhashtable_walk_next);
>
> -/**
> - * rhashtable_walk_peek - Return the next object but don't advance the iterator
> - * @iter:      Hash table iterator
> - *
> - * Returns the next object or NULL when the end of the table is reached.
> - *
> - * Returns -EAGAIN if resize event occurred.  Note that the iterator
> - * will rewind back to the beginning and you may continue to use it.
> - */
> -void *rhashtable_walk_peek(struct rhashtable_iter *iter)
> -{
> -       struct rhlist_head *list = iter->list;
> -       struct rhashtable *ht = iter->ht;
> -       struct rhash_head *p = iter->p;
> -
> -       if (p)
> -               return rht_obj(ht, ht->rhlist ? &list->rhead : p);
> -
> -       /* No object found in current iter, find next one in the table. */
> -
> -       if (iter->skip) {
> -               /* A nonzero skip value points to the next entry in the table
> -                * beyond that last one that was found. Decrement skip so
> -                * we find the current value. __rhashtable_walk_find_next
> -                * will restore the original value of skip assuming that
> -                * the table hasn't changed.
> -                */
> -               iter->skip--;
> -       }
> -
> -       return __rhashtable_walk_find_next(iter);
> -}
> -EXPORT_SYMBOL_GPL(rhashtable_walk_peek);
> -
>  /**
>   * rhashtable_walk_last_seen - Return the previously returned object, if available
>   * @iter:      Hash table iterator
> --
> 2.14.0.rc0.dirty
>