Things like this are best sorted out with atomic_store, otherwise you
may fall victim of various compiler mischief (even if it happens to be
harmless for the time being).

iow all ->staleness accesses should be converted to atomic_*.

grep finds this in ng_bridge_timeout:

                        if (++host->staleness >= priv->conf.maxStaleness) {

the code should probably explain how this is synchronized against said zeroing

On 2/9/21, Lutz Donnerhacke <[email protected]> wrote:
> The branch main has been updated by donner:
>
> URL:
> https://cgit.FreeBSD.org/src/commit/?id=011b7317dbb5038a95b9b4fca050325a62f3991e
>
> commit 011b7317dbb5038a95b9b4fca050325a62f3991e
> Author:     Lutz Donnerhacke <[email protected]>
> AuthorDate: 2021-02-09 11:32:46 +0000
> Commit:     Lutz Donnerhacke <[email protected]>
> CommitDate: 2021-02-09 19:09:05 +0000
>
>     netgraph/ng_bridge: Document staleness in multithreaded operation
>
>     In the data path of ng_bridge(4), the only value of the host struct,
>     which needs to be modified, is the staleness, which is reset every
>     time a frame is received.  It's save to leave the code as it is.
>
>     This patch is part of a series to make ng_bridge(4) multithreaded.
>
>     Reviewed by:    kp
>     MFC after:      2 weeks
>     Differential Revision: https://reviews.freebsd.org/D28546
> ---
>  sys/netgraph/ng_bridge.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/sys/netgraph/ng_bridge.c b/sys/netgraph/ng_bridge.c
> index 6e3f993d591b..29ab2c715aa7 100644
> --- a/sys/netgraph/ng_bridge.c
> +++ b/sys/netgraph/ng_bridge.c
> @@ -739,7 +739,10 @@ ng_bridge_rcvdata(hook_p hook, item_p item)
>
>       /* Look up packet's source Ethernet address in hashtable */
>       if ((host = ng_bridge_get(priv, eh->ether_shost)) != NULL) {
> -             /* Update time since last heard from this host */
> +             /* Update time since last heard from this host.
> +              * This is safe without locking, because it's
> +              * the only operation during shared access.
> +              */
>               host->staleness = 0;
>
>               /* Did host jump to a different link? */
> _______________________________________________
> [email protected] mailing list
> https://lists.freebsd.org/mailman/listinfo/dev-commits-src-all
> To unsubscribe, send any mail to
> "[email protected]"
>


-- 
Mateusz Guzik <mjguzik gmail.com>
_______________________________________________
[email protected] mailing list
https://lists.freebsd.org/mailman/listinfo/dev-commits-src-main
To unsubscribe, send any mail to "[email protected]"

Reply via email to