[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <063D6719AE5E284EB5DD2968C1650D6D0F6E93E1@AcuExch.aculab.com>
Date: Wed, 26 Mar 2014 09:49:38 +0000
From: David Laight <David.Laight@...LAB.COM>
To: "'Eric W. Biederman'" <ebiederm@...ssion.com>,
Eric Dumazet <eric.dumazet@...il.com>
CC: David Miller <davem@...emloft.net>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
"xiyou.wangcong@...il.com" <xiyou.wangcong@...il.com>,
"mpm@...enic.com" <mpm@...enic.com>,
"satyam.sharma@...il.com" <satyam.sharma@...il.com>
Subject: RE: [PATCH 01/54] uml/net_kern: Call dev_consume_skb_any instead of
dev_kfree_skb.
From: Of Eric W. Biederman
...
> I really don't think using enum skb_free_reason makes any sense
> whatsoever. Not in the implementation of dev_kfree_skb_any and
> dev_kfree_skb_irq and certainly not in a driver. What
> net/core/drop_monitor.c wants is the address of the function where drops
> occur (so we can track down and debug why the kernel is dropping
> packets) and the existing implementation of dev_kfree_skb_any and
> dev_kfree_skb_irq loose that information. The use of enum
> skb_free_reason is a big part of the reason why we loose that
> information. (We should be using a (void *) so that we can capture
> __builtin_return_address(0) instead...
Maybe more useful to allow a literal string be given.
Easier to find in the source tree than the return address.
Or (OTT) create a linkset data item containing info about the
call site and a counter....
David
--
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
Powered by blists - more mailing lists