[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <47673195.80004@redhat.com>
Date: Mon, 17 Dec 2007 21:33:57 -0500
From: Hideo AOKI <haoki@...hat.com>
To: David Miller <davem@...emloft.net>,
Herbert Xu <herbert@...dor.apana.org.au>,
netdev <netdev@...r.kernel.org>
CC: haoki@...hat.com, Takahiro Yasui <tyasui@...hat.com>,
Masami Hiramatsu <mhiramat@...hat.com>,
Satoshi Oshima <satoshi.oshima.fk@...achi.com>,
Bill Fink <billfink@...dspring.com>,
Andi Kleen <andi@...stfloor.org>,
Evgeniy Polyakov <johnpol@....mipt.ru>,
Stephen Hemminger <shemminger@...ux-foundation.org>,
yoshfuji@...ux-ipv6.org,
Yumiko Sugita <yumiko.sugita.yf@...achi.com>
Subject: [PATCH 0/4] [UDP]: memory accounting and limitation (take 11)
Hello,
I updated patch set of UDP memory accounting and limitation.
The spin lock that I used in previous take was removed from datagram
memory accounting functions. As David commented, I used socket lock
and backlog processing to keep consistency of memory accounting like
TCP. I added socket lock to places where skbuff is freed, since the
locking is needed to change sk_forward_alloc only.
Moreover, I revised memory accounting functions. As Herbert commented,
I stopped using large inline function and tried to reduce amount of
inline functions.
The patch set was tested on net-2.6 tree.
Changelog take 10 -> take 11:
* stopped using spin lock in memory accounting function
* socket lock and backlog processing were used to avoid conflict
between receive system call processing and BH
* revised memory accounting functions
* stooped changing sock_queue_rcv_skb() and skb_set_owner_r()
* added __udp_queue_rcv_skb to set proper destructor
* removed udp_set_owner_r()
* removed reclaim in inet_sock_destruct()
Changelog take 9 -> take 10:
* supported using sk_forward_alloc
* introduced several memory accounting functions with spin lock
* changed detagram receive functions to be able to customize
destructor
* fixed accounting bugs in previous takes
Changelog take 8 -> take 9:
* introduced mem_schdeule functions for datargram protocols
* removed protocol check function, from patch set
* restructured patch set
Changelog take 7 -> take 8:
* sk_datagram_pages(): avoided using divide instruction
* udp_recvmsg(): fixed referring released truesize in accounting
Best regards,
Hideo Aoki
--
Hitachi Computer Products (America) Inc.
--
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