From: "Paul E. McKenney" <paulmck@us.ibm.com>
To: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Linus Torvalds <torvalds@osdl.org>, Andrew Morton <akpm@osdl.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Suzanne Wood <suzannew@cs.pdx.edu>
Subject: Re: [LIST] Add missing rcu_dereference on first element
Date: Fri, 14 Oct 2005 19:03:25 -0700 [thread overview]
Message-ID: <20051015020324.GL1302@us.ibm.com> (raw)
In-Reply-To: <20051015002649.GA28555@gondor.apana.org.au>
On Sat, Oct 15, 2005 at 10:26:49AM +1000, Herbert Xu wrote:
> Hi:
>
> It seems that all the list_*_rcu primitives are missing a memory barrier
> on the very first dereference. For example,
>
> #define list_for_each_rcu(pos, head) \
> for (pos = (head)->next; prefetch(pos->next), pos != (head); \
> pos = rcu_dereference(pos->next))
>
> It will go something like:
>
> pos = (head)->next
>
> prefetch(pos->next)
>
> pos != (head)
>
> do stuff
>
> We're missing a barrier here.
>
> pos = rcu_dereference(pos->next)
>
> fetch pos->next
>
> barrier given by rcu_dereference(pos->next)
>
> store pos
>
> Without the missing barrier, the pos->next value may turn out to be
> stale. In fact, if "do stuff" were also dereferencing pos and relying
> on list_for_each_rcu to provide the barrier then it may also break.
>
> So here is a patch to make sure that we have a barrier for the first
> element in the list.
Good catch!!! I wonder if Alpha SMP Linux machines have noticed...
Some comments interspersed.
> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
>
> Cheers,
> --
> Visit Openswan at http://www.openswan.org/
> Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
> Home Page: http://gondor.apana.org.au/~herbert/
> PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
> diff --git a/include/linux/list.h b/include/linux/list.h
> --- a/include/linux/list.h
> +++ b/include/linux/list.h
> @@ -442,12 +442,15 @@ static inline void list_splice_init(stru
> * as long as the traversal is guarded by rcu_read_lock().
> */
> #define list_for_each_rcu(pos, head) \
> - for (pos = (head)->next; prefetch(pos->next), pos != (head); \
> - pos = rcu_dereference(pos->next))
> + for (pos = (head)->next; \
> + pos = rcu_dereference(pos), \
> + prefetch(pos->next), pos != (head); \
> + pos = pos->next)
Why not something like the following? Seems a bit simpler to me.
#define list_for_each_rcu(pos, head) \
for (pos = rcu_dereference((head)->next); \
prefetch(pos->next), pos != (head); \
pos = rcu_dereference(pos->next))
> #define __list_for_each_rcu(pos, head) \
> - for (pos = (head)->next; pos != (head); \
> - pos = rcu_dereference(pos->next))
> + for (pos = (head)->next; \
> + rcu_dereference(pos) != (head); \
> + pos = pos->next)
>
> /**
> * list_for_each_safe_rcu - iterate over an rcu-protected list safe
> @@ -461,8 +464,9 @@ static inline void list_splice_init(stru
> * as long as the traversal is guarded by rcu_read_lock().
> */
> #define list_for_each_safe_rcu(pos, n, head) \
> - for (pos = (head)->next, n = pos->next; pos != (head); \
> - pos = rcu_dereference(n), n = pos->next)
> + for (pos = (head)->next; \
> + n = rcu_dereference(pos)->next, pos != (head); \
> + pos = n)
>
> /**
> * list_for_each_entry_rcu - iterate over rcu list of given type
> @@ -474,11 +478,11 @@ static inline void list_splice_init(stru
> * the _rcu list-mutation primitives such as list_add_rcu()
> * as long as the traversal is guarded by rcu_read_lock().
> */
> -#define list_for_each_entry_rcu(pos, head, member) \
> - for (pos = list_entry((head)->next, typeof(*pos), member); \
> - prefetch(pos->member.next), &pos->member != (head); \
> - pos = rcu_dereference(list_entry(pos->member.next, \
> - typeof(*pos), member)))
> +#define list_for_each_entry_rcu(pos, head, member) \
> + for (pos = list_entry((head)->next, typeof(*pos), member); \
> + pos = rcu_dereference(pos), \
> + prefetch(pos->member.next), &pos->member != (head); \
> + pos = list_entry(pos->member.next, typeof(*pos), member))
>
>
> /**
> @@ -492,8 +496,10 @@ static inline void list_splice_init(stru
> * as long as the traversal is guarded by rcu_read_lock().
> */
> #define list_for_each_continue_rcu(pos, head) \
> - for ((pos) = (pos)->next; prefetch((pos)->next), (pos) != (head); \
> - (pos) = rcu_dereference((pos)->next))
> + for ((pos) = (pos)->next; \
> + (pos) = rcu_dereference((pos)), \
> + prefetch((pos)->next), (pos) != (head); \
> + (pos) = (pos)->next)
The above hurts my head -- childhood trauma due to having to use a
FORTRAN compiler that required "I=I" at odd intervals in order to
generate correct code... How about the following?
#define list_for_each_continue_rcu(pos, head) \
for ((pos) = (pos)->next; \
prefetch(rcu_dereference(pos)->next), (pos) != (head); \
(pos) = (pos)->next)
> /*
> * Double linked lists with a single pointer list head.
> @@ -696,8 +702,9 @@ static inline void hlist_add_after_rcu(s
> pos = n)
>
> #define hlist_for_each_rcu(pos, head) \
> - for ((pos) = (head)->first; pos && ({ prefetch((pos)->next); 1; }); \
> - (pos) = rcu_dereference((pos)->next))
> + for ((pos) = (head)->first; \
> + rcu_dereference((pos)) && ({ prefetch((pos)->next); 1; }); \
> + (pos) = (pos)->next)
>
> /**
> * hlist_for_each_entry - iterate over list of given type
> @@ -762,9 +769,9 @@ static inline void hlist_add_after_rcu(s
> */
> #define hlist_for_each_entry_rcu(tpos, pos, head, member) \
> for (pos = (head)->first; \
> - pos && ({ prefetch(pos->next); 1;}) && \
> + rcu_dereference(pos) && ({ prefetch(pos->next); 1;}) && \
> ({ tpos = hlist_entry(pos, typeof(*tpos), member); 1;}); \
> - pos = rcu_dereference(pos->next))
> + pos = pos->next)
>
> #else
> #warning "don't include kernel headers in userspace"
next prev parent reply other threads:[~2005-10-15 2:02 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-10-15 0:26 [LIST] Add missing rcu_dereference on first element Herbert Xu
2005-10-15 2:03 ` Paul E. McKenney [this message]
2005-10-15 2:39 ` Herbert Xu
2005-10-15 3:22 ` Herbert Xu
2005-10-16 20:58 ` Paul E. McKenney
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=20051015020324.GL1302@us.ibm.com \
--to=paulmck@us.ibm.com \
--cc=akpm@osdl.org \
--cc=herbert@gondor.apana.org.au \
--cc=linux-kernel@vger.kernel.org \
--cc=suzannew@cs.pdx.edu \
--cc=torvalds@osdl.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.