netdev - Re: [PATCH net-next 3/5] tipc: Optimizations to bearer enabling logic

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20101015011139.GB5048@windriver.com>
Date:	Thu, 14 Oct 2010 21:11:51 -0400
From:	Paul Gortmaker <paul.gortmaker@...driver.com>
To:	Neil Horman <nhorman@...driver.com>
Cc:	davem@...emloft.net, netdev@...r.kernel.org,
	allan.stephens@...driver.com
Subject: Re: [PATCH net-next 3/5] tipc: Optimizations to bearer enabling
 logic

[Re: [PATCH net-next 3/5] tipc: Optimizations to bearer enabling logic] On 13/10/2010 (Wed 10:58) Neil Horman wrote:

> On Tue, Oct 12, 2010 at 08:25:56PM -0400, Paul Gortmaker wrote:
> > From: Allan Stephens <allan.stephens@...driver.com>
> > 
> > Introduces "enabling" state during activation of a new TIPC bearer,
> > which supplements the existing "disabled" and "enabled" states.
> > This change allows the new bearer to be added without having to
> > temporarily block the processing of incoming packets on existing
> > bearers during the binding of the new bearer to its associated
> > interface. It also makes it unnecessary to zero out the entire
> > bearer structure at the start of activation.
> > 

[...]

> > +	b_ptr->state = BEARER_ENABLING;
> >  	strcpy(b_ptr->publ.name, name);
> > +	b_ptr->priority = priority;
> > +
> > +	write_unlock_bh(&tipc_net_lock);
> Why the 3rd state?  Doesn't seem needed. 

I'm a bit disappointed in myself for also not noticing that it
was set but never tested for.  The following should give the
same end result but without the obfuscation of an extra state.

This one also doesn't explicitly depend on any other changes,
so if it is now OK, the option is there for it to be applied
independently of the others that haven't been reworked yet.

Thanks,
Paul.


>From 86d0d5c92439d0a3f5a0f165aa8bd842d377dae9 Mon Sep 17 00:00:00 2001
From: Allan Stephens <allan.stephens@...driver.com>
Date: Thu, 14 Oct 2010 16:09:23 -0400
Subject: [PATCH] tipc: Optimizations to bearer enabling logic

Allow new bearers to be added without having to temporarily block
the processing of incoming packets on existing bearers during the
binding of the new bearer to its associated interface. Eliminates
zeroing out of the new bearer structure at the start of activation,
since it is already in that state.

Signed-off-by: Allan Stephens <allan.stephens@...driver.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@...driver.com>
---
 net/tipc/bearer.c |    7 ++++---
 1 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c
index fd9c06c..2ff8181 100644
--- a/net/tipc/bearer.c
+++ b/net/tipc/bearer.c
@@ -556,14 +556,15 @@ restart:
 	}
 
 	b_ptr = &tipc_bearers[bearer_id];
-	memset(b_ptr, 0, sizeof(struct bearer));
-
 	strcpy(b_ptr->publ.name, name);
+
+	write_unlock_bh(&tipc_net_lock);
 	res = m_ptr->enable_bearer(&b_ptr->publ);
 	if (res) {
 		warn("Bearer <%s> rejected, enable failure (%d)\n", name, -res);
-		goto failed;
+		return res;
 	}
+	write_lock_bh(&tipc_net_lock);
 
 	b_ptr->identity = bearer_id;
 	b_ptr->media = m_ptr;
-- 
1.7.2.1

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html