netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Gilad Naaman <gnaaman@drivenets.com>
To: kuniyu@amazon.com
Cc: davem@davemloft.net, edumazet@google.com, gnaaman@drivenets.com,
	kuba@kernel.org, netdev@vger.kernel.org, pabeni@redhat.com
Subject: Re: [PATCH net-next v4 3/6] Convert neigh_* seq_file functions to use hlist
Date: Wed, 16 Oct 2024 09:11:52 +0000	[thread overview]
Message-ID: <20241016091152.3504685-1-gnaaman@drivenets.com> (raw)
In-Reply-To: <20241015232529.67605-1-kuniyu@amazon.com>

> > -		if (++state->bucket >= (1 << nht->hash_shift))
> > -			break;
> 
> Let's keep this,
> 
> 
> > +		while (!n && ++state->bucket < (1 << nht->hash_shift))
> > +			n = neigh_first_rcu(&nht->hash_heads[state->bucket]);
> >  
> > -		n = rcu_dereference(nht->hash_buckets[state->bucket]);
> 
> and simply fetch neigh_first_entry().
> 
> Then, we can let the next loop handle the termination condition
> and keep the code simple.

Unfortunately `hlist_for_each_entry_continue_rcu` dereferences `n`
first thing using `hlist_next_rcu`, before checking for NULL:

    #define hlist_for_each_entry_continue_rcu(pos, member)			\
    	for (pos = hlist_entry_safe(rcu_dereference_raw(hlist_next_rcu( \
    			&(pos)->member)), typeof(*(pos)), member);	\
    	     pos;							\
    	     pos = hlist_entry_safe(rcu_dereference_raw(hlist_next_rcu(	\
    			&(pos)->member)), typeof(*(pos)), member))

If I'm using it, I have to add a null-check after calling `neigh_first_entry`.

Another alternative is to use `hlist_for_each_entry_from_rcu`,
but it'll require calling `next` one time before the loop.

Is this preferable?

  reply	other threads:[~2024-10-16  9:12 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-15 16:59 [PATCH net-next v4 0/6] Improve neigh_flush_dev performance Gilad Naaman
2024-10-15 16:59 ` [PATCH net-next v4 1/6] Add hlist_node to struct neighbour Gilad Naaman
2024-10-15 22:49   ` Kuniyuki Iwashima
2024-10-15 16:59 ` [PATCH net-next v4 2/6] Define neigh_for_each Gilad Naaman
2024-10-15 22:57   ` Kuniyuki Iwashima
2024-10-15 16:59 ` [PATCH net-next v4 3/6] Convert neigh_* seq_file functions to use hlist Gilad Naaman
2024-10-15 23:25   ` Kuniyuki Iwashima
2024-10-16  9:11     ` Gilad Naaman [this message]
2024-10-16 20:54       ` Kuniyuki Iwashima
2024-10-17  6:31         ` Gilad Naaman
2024-10-15 16:59 ` [PATCH net-next v4 4/6] Convert neighbour iteration to use hlist+macro Gilad Naaman
2024-10-15 23:38   ` Kuniyuki Iwashima
2024-10-16  9:18     ` Gilad Naaman
2024-10-15 16:59 ` [PATCH net-next v4 5/6] Remove bare neighbour::next pointer Gilad Naaman
2024-10-15 16:59 ` [PATCH net-next v4 6/6] Create netdev->neighbour association Gilad Naaman
2024-10-15 23:54   ` Kuniyuki Iwashima

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=20241016091152.3504685-1-gnaaman@drivenets.com \
    --to=gnaaman@drivenets.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=kuba@kernel.org \
    --cc=kuniyu@amazon.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.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 NNTP newsgroup(s).