* a couple questions about drivers/net/macv*.c
From: Robert P. J. Day @ 2010-07-10 23:44 UTC (permalink / raw)
To: netdev
first (trivial) point, i notice that both macvlan.c and macvtap.c
are still both listed as EXPERIMENTAL -- is that still accurate?
bigger issue in macvtap.c -- notice this declaration:
static unsigned int macvtap_major;
that seems like a violation of coding style since that variable is
used later on in:
devt = MKDEV(MAJOR(macvtap_major), dev->ifindex);
and
err = alloc_chrdev_region(&macvtap_major, 0,
MACVTAP_NUM_DEVS, "macvtap");
where its type should simply be the typedef "dev_t" for transparency,
should it not? hardcoding that variable as an unsigned int seems like
a bad idea.
rday
--
========================================================================
Robert P. J. Day Waterloo, Ontario, CANADA
Top-notch, inexpensive online Linux/OSS/kernel courses
http://crashcourse.ca
Twitter: http://twitter.com/rpjday
LinkedIn: http://ca.linkedin.com/in/rpjday
========================================================================
^ permalink raw reply
* Re: a couple questions about drivers/net/macv*.c
From: David Miller @ 2010-07-11 2:26 UTC (permalink / raw)
To: rpjday; +Cc: netdev
In-Reply-To: <alpine.DEB.2.00.1007101652030.7658@lynx>
From: "Robert P. J. Day" <rpjday@crashcourse.ca>
Date: Sat, 10 Jul 2010 19:44:59 -0400 (EDT)
>
> first (trivial) point, i notice that both macvlan.c and macvtap.c
> are still both listed as EXPERIMENTAL -- is that still accurate?
Probably the tag should be removed, a lot of people use this
facility and it works quite well as far as I can tell.
> bigger issue in macvtap.c -- notice this declaration:
>
> static unsigned int macvtap_major;
>
> that seems like a violation of coding style since that variable is
> used later on in:
>
> devt = MKDEV(MAJOR(macvtap_major), dev->ifindex);
> and
>
> err = alloc_chrdev_region(&macvtap_major, 0,
> MACVTAP_NUM_DEVS, "macvtap");
>
> where its type should simply be the typedef "dev_t" for transparency,
> should it not?
Yep, I'll check in the following to net-next-2.6, thanks.
--------------------
macvtap: Use dev_t for macvtap_major.
Reported-by: "Robert P. J. Day" <rpjday@crashcourse.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
drivers/net/macvtap.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
index a8a94e2..2b4d59b 100644
--- a/drivers/net/macvtap.c
+++ b/drivers/net/macvtap.c
@@ -58,7 +58,7 @@ static struct proto macvtap_proto = {
* only has one tap, the interface numbers assure that the
* device nodes are unique.
*/
-static unsigned int macvtap_major;
+static dev_t macvtap_major;
#define MACVTAP_NUM_DEVS 65536
static struct class *macvtap_class;
static struct cdev macvtap_cdev;
--
1.7.1.1
^ permalink raw reply related
* Re: a couple questions about drivers/net/macv*.c
From: Robert P. J. Day @ 2010-07-11 2:31 UTC (permalink / raw)
To: David Miller; +Cc: netdev
In-Reply-To: <20100710.192610.232749731.davem@davemloft.net>
On Sat, 10 Jul 2010, David Miller wrote:
> From: "Robert P. J. Day" <rpjday@crashcourse.ca>
> Date: Sat, 10 Jul 2010 19:44:59 -0400 (EDT)
>
> >
> > first (trivial) point, i notice that both macvlan.c and macvtap.c
> > are still both listed as EXPERIMENTAL -- is that still accurate?
>
> Probably the tag should be removed, a lot of people use this
> facility and it works quite well as far as I can tell.
>
> > bigger issue in macvtap.c -- notice this declaration:
> >
> > static unsigned int macvtap_major;
> >
> > that seems like a violation of coding style since that variable is
> > used later on in:
> >
> > devt = MKDEV(MAJOR(macvtap_major), dev->ifindex);
> > and
> >
> > err = alloc_chrdev_region(&macvtap_major, 0,
> > MACVTAP_NUM_DEVS, "macvtap");
> >
> > where its type should simply be the typedef "dev_t" for transparency,
> > should it not?
>
> Yep, I'll check in the following to net-next-2.6, thanks.
>
> --------------------
> macvtap: Use dev_t for macvtap_major.
>
> Reported-by: "Robert P. J. Day" <rpjday@crashcourse.ca>
> Signed-off-by: David S. Miller <davem@davemloft.net>
> ---
> drivers/net/macvtap.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/drivers/net/macvtap.c b/drivers/net/macvtap.c
> index a8a94e2..2b4d59b 100644
> --- a/drivers/net/macvtap.c
> +++ b/drivers/net/macvtap.c
> @@ -58,7 +58,7 @@ static struct proto macvtap_proto = {
> * only has one tap, the interface numbers assure that the
> * device nodes are unique.
> */
> -static unsigned int macvtap_major;
> +static dev_t macvtap_major;
technically, i would drop the "_major" suffix since that variable
doesn't represent simply the major device number but the entire dev_t
typedef, but that's your call.
rday
--
========================================================================
Robert P. J. Day Waterloo, Ontario, CANADA
Top-notch, inexpensive online Linux/OSS/kernel courses
http://crashcourse.ca
Twitter: http://twitter.com/rpjday
LinkedIn: http://ca.linkedin.com/in/rpjday
========================================================================
^ permalink raw reply
* Re: oops in tcp_xmit_retransmit_queue() w/ v2.6.32.15
From: David Miller @ 2010-07-11 2:36 UTC (permalink / raw)
To: tj; +Cc: linux-kernel, netdev, henning.fehrmann, carsten.aulbert
In-Reply-To: <4C358AAA.9080400@kernel.org>
From: Tejun Heo <tj@kernel.org>
Date: Thu, 08 Jul 2010 10:22:02 +0200
> We've been seeing oops in tcp_xmit_retransmit_queue() w/ 2.6.32.15.
...
> Does anything ring a bell?
A long time ago we had a packet scheduler bug that could corrupt
the TCP socket queues, but that was fixed in 2.6.27 so would
definitely be fixed in your kernel.
--------------------
commit 69747650c814a8a79fef412c7416adf823293a3e
Author: David S. Miller <davem@davemloft.net>
Date: Sun Aug 17 23:55:36 2008 -0700
pkt_sched: Fix return value corruption in HTB and TBF.
Based upon a bug report by Josip Rodin.
Packet schedulers should only return NET_XMIT_DROP iff
the packet really was dropped. If the packet does reach
the device after we return NET_XMIT_DROP then TCP can
crash because it depends upon the enqueue path return
values being accurate.
Signed-off-by: David S. Miller <davem@davemloft.net>
--------------------
Nothing else jumps to mind, sorry.
^ permalink raw reply
* Re: [PATCH net-next-2.6] be2net: Patch to determine if function is VF while running in guest OS.
From: David Miller @ 2010-07-11 2:42 UTC (permalink / raw)
To: sarveshwarb; +Cc: netdev
In-Reply-To: <20100709114344.GA2392@serverengines.com>
From: Sarveshwar Bandi <sarveshwarb@serverengines.com>
Date: Fri, 9 Jul 2010 17:13:55 +0530
> When driver is loaded in guest OS, the pci variables is_virtfn and is_physfn are
> both set to 0. This change uses registers in controller to determine the same.
>
> Signed-off-by: Sarveshwar Bandi <sarveshwarb@serverengines.com>
Applied.
^ permalink raw reply
* Re: [PATCH 1/2] 82596: do not panic on out of memory
From: David Miller @ 2010-07-11 2:42 UTC (permalink / raw)
To: segooon
Cc: kernel-janitors, jpirko, shemminger, eric.dumazet, andre.goddard,
netdev
In-Reply-To: <1278678323-6848-1-git-send-email-segooon@gmail.com>
From: Kulikov Vasiliy <segooon@gmail.com>
Date: Fri, 9 Jul 2010 16:25:22 +0400
> If dev_alloc_skb() failed then free already allocated skbs.
> remove_rx_bufs() can be called multiple times, so set rbd->skb to NULL
> to avoid double free. remove_rx_bufs() was moved upwards to be seen by
> init_rx_bufs().
>
> Signed-off-by: Kulikov Vasiliy <segooon@gmail.com>
Applied to net-next-2.6
^ permalink raw reply
* Re: [PATCH 2/2] 82596: free resources on error
From: David Miller @ 2010-07-11 2:42 UTC (permalink / raw)
To: segooon
Cc: kernel-janitors, jpirko, shemminger, eric.dumazet, andre.goddard,
netdev
In-Reply-To: <1278678340-6889-1-git-send-email-segooon@gmail.com>
From: Kulikov Vasiliy <segooon@gmail.com>
Date: Fri, 9 Jul 2010 16:25:40 +0400
> IRQ 56 was not freed anywhere (neither in i596_open() on error nor in
> i596_close()), rx_bufs were not freed if init_i596_mem() fails,
> netif_stop_queue() was not called.
>
> Signed-off-by: Kulikov Vasiliy <segooon@gmail.com>
Applied to net-next-2.6
^ permalink raw reply
* Re: [PATCH kernel 2.6.35-rc3-git7] axnet_cs: use spin_lock_irqsave in ax_interrupt
From: Ben Hutchings @ 2010-07-11 2:46 UTC (permalink / raw)
To: Ken Kawasaki; +Cc: netdev
In-Reply-To: <20100710201813.a3e5c79c.ken_kawasaki@spring.nifty.jp>
On Sat, 2010-07-10 at 20:18 +0900, Ken Kawasaki wrote:
> axnet_cs:
> use spin_lock_irqsave instead of spin_lock in ax_interrupt
[...]
I assume this is because it's now called from ei_watchdog() and not only
from interrupt context. Perhaps you should explain that in the commit
message.
Ben.
--
Ben Hutchings, Senior Software Engineer, Solarflare Communications
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply
* Re: [PATCH kernel 2.6.35-rc3-git7] axnet_cs: use spin_lock_irqsave in ax_interrupt
From: David Miller @ 2010-07-11 2:49 UTC (permalink / raw)
To: bhutchings; +Cc: ken_kawasaki, netdev
In-Reply-To: <1278816388.6139.10.camel@localhost>
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Sun, 11 Jul 2010 03:46:28 +0100
> On Sat, 2010-07-10 at 20:18 +0900, Ken Kawasaki wrote:
>> axnet_cs:
>> use spin_lock_irqsave instead of spin_lock in ax_interrupt
> [...]
>
> I assume this is because it's now called from ei_watchdog() and not only
> from interrupt context. Perhaps you should explain that in the commit
> message.
No, interrupt handlers in general may not assume that interrupts
are off or on when they are invoked.
Therefore they must use irqflags saving/restoring.
^ permalink raw reply
* Re: linux-next: build warning after merge of the net tree
From: David Miller @ 2010-07-11 2:52 UTC (permalink / raw)
To: joe; +Cc: sfr, netdev, linux-next, linux-kernel, gregkh
In-Reply-To: <1278562423.1712.43.camel@Joe-Laptop>
From: Joe Perches <joe@perches.com>
Date: Wed, 07 Jul 2010 21:13:42 -0700
> On Wed, 2010-07-07 at 18:18 -0700, David Miller wrote:
>> It looks like there are just a hand-ful of cases, so maybe we can tweak
>> them by hand. For example, in the sym53c8xx_2 driver bits we can replace
>> the NULL labels passed to sym_print_msg() with a real string and therefore
>> remove the "" case.
>>
>> Joe, any better ideas?
>
> You're right there are just a few cases where dev_info
> is uses as a preface for a hex_dump style display.
>
> Maybe it'd be OK to simply add a trailing space to the
> preface and remove any leading spaces from the subsequent
> initial printks.
>
> dev_info(dev, " ");
That might work.
The sym53c8xx_2 doesn't even need this, like I said, you could
just remove the NULL 'label' argument cases and then have that
bit cured.
Could you take a stab at this and the other scsi bits that
trigger this warning?
Thanks Joe!
^ permalink raw reply
* Re: [PATCH] tproxy: nf_tproxy_assign_sock() can handle tw sockets
From: Felipe W Damasio @ 2010-07-11 3:11 UTC (permalink / raw)
To: David Miller; +Cc: eric.dumazet, kaber, linux-kernel, netdev
In-Reply-To: <20100710.123011.260095467.davem@davemloft.net>
Hi Mr. Miller,
2010/7/10 David Miller <davem@davemloft.net>:
> It could be corruption from elsewhere. Those last four hex
> digits (0x5d415d41) are "]A]A" in ascii, but that could just
> be coincidence.
What do you mean "from elsewhere"? You mean elsewhere on the network code?
Since the function that had the problem was tcp_recvmsg and we're
talking about a squid process, we're either talking about a typical
webserver-objet response, or about about an incorrect/faulty http
request from the user.
Like I told Mr. Dumazet, since on the squid logs I got a:
2010/07/08 14:51:10| clientTryParseRequest: FD 6088
(187.16.240.122:2035) Invalid Request
Only a second before the bug entry on syslog, I suppose that this
invalid request caused the problem (more like a guess, really).
If you think there's a way I can help reproduce/trigger and fix this
bug, please let me know, since the production machine is down until I
can ensure my bosses that this particular crash won't happen again.
Thanks,
Felipe Damasio
^ permalink raw reply
* Re: [PATCH kernel 2.6.35-rc3-git7] axnet_cs: use spin_lock_irqsave in ax_interrupt
From: Ben Hutchings @ 2010-07-11 3:12 UTC (permalink / raw)
To: David Miller; +Cc: ken_kawasaki, netdev
In-Reply-To: <20100710.194928.116377282.davem@davemloft.net>
On Sat, 2010-07-10 at 19:49 -0700, David Miller wrote:
> From: Ben Hutchings <bhutchings@solarflare.com>
> Date: Sun, 11 Jul 2010 03:46:28 +0100
>
> > On Sat, 2010-07-10 at 20:18 +0900, Ken Kawasaki wrote:
> >> axnet_cs:
> >> use spin_lock_irqsave instead of spin_lock in ax_interrupt
> > [...]
> >
> > I assume this is because it's now called from ei_watchdog() and not only
> > from interrupt context. Perhaps you should explain that in the commit
> > message.
>
> No, interrupt handlers in general may not assume that interrupts
> are off or on when they are invoked.
>
> Therefore they must use irqflags saving/restoring.
But an interrupt handler will not be called recursively for the same
IRQ. Since this device only uses one IRQ, surely it was OK to use
spin_lock() in this function so long as it was only called from the
interrupt handler.
Ben.
--
Ben Hutchings, Senior Software Engineer, Solarflare Communications
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply
* [PATCH] inet, inet6: make tcp_sendmsg() and tcp_sendpage() through inet_sendmsg() and inet_sendpage()
From: Changli Gao @ 2010-07-11 3:44 UTC (permalink / raw)
To: David S. Miller
Cc: Alexey Kuznetsov, Pekka Savola (ipv6), James Morris,
Hideaki YOSHIFUJI, Patrick McHardy, Tom Herbert, netdev,
Changli Gao
inet, inet6: make tcp_sendmsg() and tcp_sendpage() through inet_sendmsg() and
inet_sendpage()
a new boolean flag no_autobind is added to structure proto to avoid the autobind
calls when the protocol is TCP. Then sock_rps_record_flow() is called int the
TCP's sendmsg() and sendpage() pathes.
Signed-off-by: Changli Gao <xiaosuo@gmail.com>
----
include/net/inet_common.h | 53 +++++++++++++++++++---------------------------
include/net/sock.h | 1
include/net/tcp.h | 47 ++++++++++++++++++----------------------
net/ipv4/af_inet.c | 15 +++++++------
net/ipv4/tcp.c | 9 +++----
net/ipv4/tcp_ipv4.c | 3 ++
net/ipv6/af_inet6.c | 8 +++---
net/ipv6/tcp_ipv6.c | 3 ++
8 files changed, 67 insertions(+), 72 deletions(-)
diff --git a/include/net/inet_common.h b/include/net/inet_common.h
index 18c7732..519790e 100644
--- a/include/net/inet_common.h
+++ b/include/net/inet_common.h
@@ -13,37 +13,28 @@ struct sock;
struct sockaddr;
struct socket;
-extern int inet_release(struct socket *sock);
-extern int inet_stream_connect(struct socket *sock,
- struct sockaddr * uaddr,
- int addr_len, int flags);
-extern int inet_dgram_connect(struct socket *sock,
- struct sockaddr * uaddr,
- int addr_len, int flags);
-extern int inet_accept(struct socket *sock,
- struct socket *newsock, int flags);
-extern int inet_sendmsg(struct kiocb *iocb,
- struct socket *sock,
- struct msghdr *msg,
- size_t size);
-extern int inet_shutdown(struct socket *sock, int how);
-extern int inet_listen(struct socket *sock, int backlog);
-
-extern void inet_sock_destruct(struct sock *sk);
-
-extern int inet_bind(struct socket *sock,
- struct sockaddr *uaddr, int addr_len);
-extern int inet_getname(struct socket *sock,
- struct sockaddr *uaddr,
- int *uaddr_len, int peer);
-extern int inet_ioctl(struct socket *sock,
- unsigned int cmd, unsigned long arg);
-
-extern int inet_ctl_sock_create(struct sock **sk,
- unsigned short family,
- unsigned short type,
- unsigned char protocol,
- struct net *net);
+extern int inet_release(struct socket *sock);
+extern int inet_stream_connect(struct socket *sock, struct sockaddr * uaddr,
+ int addr_len, int flags);
+extern int inet_dgram_connect(struct socket *sock, struct sockaddr * uaddr,
+ int addr_len, int flags);
+extern int inet_accept(struct socket *sock, struct socket *newsock, int flags);
+extern int inet_sendmsg(struct kiocb *iocb, struct socket *sock,
+ struct msghdr *msg, size_t size);
+extern int inet_recvmsg(struct kiocb *iocb, struct socket *sock,
+ struct msghdr *msg, size_t size, int flags);
+extern ssize_t inet_sendpage(struct socket *sock, struct page *page, int offset,
+ size_t size, int flags);
+extern int inet_shutdown(struct socket *sock, int how);
+extern int inet_listen(struct socket *sock, int backlog);
+extern void inet_sock_destruct(struct sock *sk);
+extern int inet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len);
+extern int inet_getname(struct socket *sock, struct sockaddr *uaddr,
+ int *uaddr_len, int peer);
+extern int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg);
+extern int inet_ctl_sock_create(struct sock **sk, unsigned short family,
+ unsigned short type, unsigned char protocol,
+ struct net *net);
static inline void inet_ctl_sock_destroy(struct sock *sk)
{
diff --git a/include/net/sock.h b/include/net/sock.h
index 4f26f2f..3100e71 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -772,6 +772,7 @@ struct proto {
int *sysctl_wmem;
int *sysctl_rmem;
int max_header;
+ bool no_autobind;
struct kmem_cache *slab;
unsigned int obj_size;
diff --git a/include/net/tcp.h b/include/net/tcp.h
index c2f96c2..dd72895 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -296,45 +296,40 @@ extern struct proto tcp_prot;
#define TCP_ADD_STATS_USER(net, field, val) SNMP_ADD_STATS_USER((net)->mib.tcp_statistics, field, val)
#define TCP_ADD_STATS(net, field, val) SNMP_ADD_STATS((net)->mib.tcp_statistics, field, val)
-extern void tcp_v4_err(struct sk_buff *skb, u32);
+extern void tcp_v4_err(struct sk_buff *skb, u32);
-extern void tcp_shutdown (struct sock *sk, int how);
+extern void tcp_shutdown (struct sock *sk, int how);
-extern int tcp_v4_rcv(struct sk_buff *skb);
+extern int tcp_v4_rcv(struct sk_buff *skb);
-extern int tcp_v4_remember_stamp(struct sock *sk);
+extern int tcp_v4_remember_stamp(struct sock *sk);
-extern int tcp_v4_tw_remember_stamp(struct inet_timewait_sock *tw);
+extern int tcp_v4_tw_remember_stamp(struct inet_timewait_sock *tw);
-extern int tcp_sendmsg(struct kiocb *iocb, struct socket *sock,
- struct msghdr *msg, size_t size);
-extern ssize_t tcp_sendpage(struct socket *sock, struct page *page, int offset, size_t size, int flags);
+extern int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
+ size_t size);
+extern int tcp_sendpage(struct sock *sk, struct page *page, int offset,
+ size_t size, int flags);
-extern int tcp_ioctl(struct sock *sk,
- int cmd,
- unsigned long arg);
+extern int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg);
-extern int tcp_rcv_state_process(struct sock *sk,
- struct sk_buff *skb,
- struct tcphdr *th,
- unsigned len);
+extern int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
+ struct tcphdr *th, unsigned len);
-extern int tcp_rcv_established(struct sock *sk,
- struct sk_buff *skb,
- struct tcphdr *th,
- unsigned len);
+extern int tcp_rcv_established(struct sock *sk, struct sk_buff *skb,
+ struct tcphdr *th, unsigned len);
-extern void tcp_rcv_space_adjust(struct sock *sk);
+extern void tcp_rcv_space_adjust(struct sock *sk);
-extern void tcp_cleanup_rbuf(struct sock *sk, int copied);
+extern void tcp_cleanup_rbuf(struct sock *sk, int copied);
-extern int tcp_twsk_unique(struct sock *sk,
- struct sock *sktw, void *twp);
+extern int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp);
-extern void tcp_twsk_destructor(struct sock *sk);
+extern void tcp_twsk_destructor(struct sock *sk);
-extern ssize_t tcp_splice_read(struct socket *sk, loff_t *ppos,
- struct pipe_inode_info *pipe, size_t len, unsigned int flags);
+extern ssize_t tcp_splice_read(struct socket *sk, loff_t *ppos,
+ struct pipe_inode_info *pipe, size_t len,
+ unsigned int flags);
static inline void tcp_dec_quickack_mode(struct sock *sk,
const unsigned int pkts)
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 3ceb025..1e5bad1 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -727,28 +727,31 @@ int inet_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
sock_rps_record_flow(sk);
/* We may need to bind the socket. */
- if (!inet_sk(sk)->inet_num && inet_autobind(sk))
+ if (!sk->sk_prot->no_autobind &&
+ !inet_sk(sk)->inet_num && inet_autobind(sk))
return -EAGAIN;
return sk->sk_prot->sendmsg(iocb, sk, msg, size);
}
EXPORT_SYMBOL(inet_sendmsg);
-static ssize_t inet_sendpage(struct socket *sock, struct page *page, int offset,
- size_t size, int flags)
+ssize_t inet_sendpage(struct socket *sock, struct page *page, int offset,
+ size_t size, int flags)
{
struct sock *sk = sock->sk;
sock_rps_record_flow(sk);
/* We may need to bind the socket. */
- if (!inet_sk(sk)->inet_num && inet_autobind(sk))
+ if (!sk->sk_prot->no_autobind &&
+ !inet_sk(sk)->inet_num && inet_autobind(sk))
return -EAGAIN;
if (sk->sk_prot->sendpage)
return sk->sk_prot->sendpage(sk, page, offset, size, flags);
return sock_no_sendpage(sock, page, offset, size, flags);
}
+EXPORT_SYMBOL(inet_sendpage);
int inet_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
size_t size, int flags)
@@ -894,10 +897,10 @@ const struct proto_ops inet_stream_ops = {
.shutdown = inet_shutdown,
.setsockopt = sock_common_setsockopt,
.getsockopt = sock_common_getsockopt,
- .sendmsg = tcp_sendmsg,
+ .sendmsg = inet_sendmsg,
.recvmsg = inet_recvmsg,
.mmap = sock_no_mmap,
- .sendpage = tcp_sendpage,
+ .sendpage = inet_sendpage,
.splice_read = tcp_splice_read,
#ifdef CONFIG_COMPAT
.compat_setsockopt = compat_sock_common_setsockopt,
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 4e6ddfb..b0e0e7c 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -856,15 +856,15 @@ out_err:
return sk_stream_error(sk, flags, err);
}
-ssize_t tcp_sendpage(struct socket *sock, struct page *page, int offset,
+int tcp_sendpage(struct sock *sk, struct page *page, int offset,
size_t size, int flags)
{
ssize_t res;
- struct sock *sk = sock->sk;
if (!(sk->sk_route_caps & NETIF_F_SG) ||
!(sk->sk_route_caps & NETIF_F_ALL_CSUM))
- return sock_no_sendpage(sock, page, offset, size, flags);
+ return sock_no_sendpage(sk->sk_socket, page, offset, size,
+ flags);
lock_sock(sk);
TCP_CHECK_TIMER(sk);
@@ -897,10 +897,9 @@ static inline int select_size(struct sock *sk, int sg)
return tmp;
}
-int tcp_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
+int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
size_t size)
{
- struct sock *sk = sock->sk;
struct iovec *iov;
struct tcp_sock *tp = tcp_sk(sk);
struct sk_buff *skb;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 8fa32f5..f7eb3a6 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2595,6 +2595,8 @@ struct proto tcp_prot = {
.setsockopt = tcp_setsockopt,
.getsockopt = tcp_getsockopt,
.recvmsg = tcp_recvmsg,
+ .sendmsg = tcp_sendmsg,
+ .sendpage = tcp_sendpage,
.backlog_rcv = tcp_v4_do_rcv,
.hash = inet_hash,
.unhash = inet_unhash,
@@ -2613,6 +2615,7 @@ struct proto tcp_prot = {
.twsk_prot = &tcp_timewait_sock_ops,
.rsk_prot = &tcp_request_sock_ops,
.h.hashinfo = &tcp_hashinfo,
+ .no_autobind = true,
#ifdef CONFIG_COMPAT
.compat_setsockopt = compat_tcp_setsockopt,
.compat_getsockopt = compat_tcp_getsockopt,
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index e830cd4..56b9bf2 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -522,10 +522,10 @@ const struct proto_ops inet6_stream_ops = {
.shutdown = inet_shutdown, /* ok */
.setsockopt = sock_common_setsockopt, /* ok */
.getsockopt = sock_common_getsockopt, /* ok */
- .sendmsg = tcp_sendmsg, /* ok */
- .recvmsg = sock_common_recvmsg, /* ok */
+ .sendmsg = inet_sendmsg, /* ok */
+ .recvmsg = inet_recvmsg, /* ok */
.mmap = sock_no_mmap,
- .sendpage = tcp_sendpage,
+ .sendpage = inet_sendpage,
.splice_read = tcp_splice_read,
#ifdef CONFIG_COMPAT
.compat_setsockopt = compat_sock_common_setsockopt,
@@ -549,7 +549,7 @@ const struct proto_ops inet6_dgram_ops = {
.setsockopt = sock_common_setsockopt, /* ok */
.getsockopt = sock_common_getsockopt, /* ok */
.sendmsg = inet_sendmsg, /* ok */
- .recvmsg = sock_common_recvmsg, /* ok */
+ .recvmsg = inet_recvmsg, /* ok */
.mmap = sock_no_mmap,
.sendpage = sock_no_sendpage,
#ifdef CONFIG_COMPAT
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 5ebc27e..fe6d404 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2142,6 +2142,8 @@ struct proto tcpv6_prot = {
.setsockopt = tcp_setsockopt,
.getsockopt = tcp_getsockopt,
.recvmsg = tcp_recvmsg,
+ .sendmsg = tcp_sendmsg,
+ .sendpage = tcp_sendpage,
.backlog_rcv = tcp_v6_do_rcv,
.hash = tcp_v6_hash,
.unhash = inet_unhash,
@@ -2160,6 +2162,7 @@ struct proto tcpv6_prot = {
.twsk_prot = &tcp6_timewait_sock_ops,
.rsk_prot = &tcp6_request_sock_ops,
.h.hashinfo = &tcp_hashinfo,
+ .no_autobind = true,
#ifdef CONFIG_COMPAT
.compat_setsockopt = compat_tcp_setsockopt,
.compat_getsockopt = compat_tcp_getsockopt,
^ permalink raw reply related
* Re: [PATCH] inet, inet6: make tcp_sendmsg() and tcp_sendpage() through inet_sendmsg() and inet_sendpage()
From: Dimitris Michailidis @ 2010-07-11 4:45 UTC (permalink / raw)
To: Changli Gao
Cc: David S. Miller, Alexey Kuznetsov, Pekka Savola (ipv6),
James Morris, Hideaki YOSHIFUJI, Patrick McHardy, Tom Herbert,
netdev
In-Reply-To: <1278819894-2861-1-git-send-email-xiaosuo@gmail.com>
Changli Gao wrote:
> diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
> index 3ceb025..1e5bad1 100644
> --- a/net/ipv4/af_inet.c
> +++ b/net/ipv4/af_inet.c
> @@ -727,28 +727,31 @@ int inet_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
> sock_rps_record_flow(sk);
>
> /* We may need to bind the socket. */
> - if (!inet_sk(sk)->inet_num && inet_autobind(sk))
> + if (!sk->sk_prot->no_autobind &&
> + !inet_sk(sk)->inet_num && inet_autobind(sk))
> return -EAGAIN;
I think it's better to test inet_num first since usually it's not 0. With
the above ordering protocols with no_autobind==false end up making two
comparisons.
^ permalink raw reply
* [PATCH net-next] drivers/scsi: Remove warnings after vsprintf %pV introduction
From: Joe Perches @ 2010-07-11 5:08 UTC (permalink / raw)
To: David Miller
Cc: sfr, netdev, linux-next, linux-kernel, gregkh,
James E.J. Bottomley, Matthew Wilcox, linux-scsi
In-Reply-To: <20100710.195210.200371025.davem@davemloft.net>
On Sat, 2010-07-10 at 19:52 -0700, David Miller wrote:
> Could you take a stab at this and the other scsi bits that
> trigger this warning?
Remove warnings introduced by conversions of dev_<level>
macros to functions.
Compile tested only.
Signed-off-by: Joe Perches <joe@perches.com>
---
drivers/scsi/constants.c | 63 ++++++++++++++++++++---------------
drivers/scsi/sd.c | 6 ++--
drivers/scsi/sym53c8xx_2/sym_hipd.c | 10 ++---
3 files changed, 43 insertions(+), 36 deletions(-)
diff --git a/drivers/scsi/constants.c b/drivers/scsi/constants.c
index cd05e04..f95de51 100644
--- a/drivers/scsi/constants.c
+++ b/drivers/scsi/constants.c
@@ -1226,29 +1226,38 @@ scsi_extd_sense_format(unsigned char asc, unsigned char ascq) {
}
EXPORT_SYMBOL(scsi_extd_sense_format);
+static void scsi_show_extd_sense_args(const char *fmt, ...)
+{
+ va_list args;
+ struct va_format vaf;
+
+ va_start(args, fmt);
+
+ vaf.fmt = fmt;
+ vaf.va = &args;
+
+ printk(KERN_CONT "Add. Sense: %pV\n", &vaf);
+
+ va_end(args);
+}
+
void
scsi_show_extd_sense(unsigned char asc, unsigned char ascq)
{
const char *extd_sense_fmt = scsi_extd_sense_format(asc, ascq);
if (extd_sense_fmt) {
- if (strstr(extd_sense_fmt, "%x")) {
- printk("Add. Sense: ");
- printk(extd_sense_fmt, ascq);
- } else
- printk("Add. Sense: %s", extd_sense_fmt);
+ scsi_show_extd_sense_args(extd_sense_fmt, ascq);
} else {
if (asc >= 0x80)
- printk("<<vendor>> ASC=0x%x ASCQ=0x%x", asc,
- ascq);
+ printk(KERN_CONT "<<vendor>> ASC=0x%x ASCQ=0x%x",
+ asc, ascq);
if (ascq >= 0x80)
- printk("ASC=0x%x <<vendor>> ASCQ=0x%x", asc,
- ascq);
+ printk(KERN_CONT "ASC=0x%x <<vendor>> ASCQ=0x%x\n",
+ asc, ascq);
else
- printk("ASC=0x%x ASCQ=0x%x", asc, ascq);
+ printk(KERN_CONT "ASC=0x%x ASCQ=0x%x\n", asc, ascq);
}
-
- printk("\n");
}
EXPORT_SYMBOL(scsi_show_extd_sense);
@@ -1310,15 +1319,15 @@ scsi_decode_sense_buffer(const unsigned char *sense_buffer, int sense_len,
if (0 == res) {
/* this may be SCSI-1 sense data */
num = (sense_len < 32) ? sense_len : 32;
- printk("Unrecognized sense data (in hex):");
+ printk(KERN_CONT "Unrecognized sense data (in hex):");
for (k = 0; k < num; ++k) {
if (0 == (k % 16)) {
- printk("\n");
- printk(KERN_INFO " ");
+ printk(KERN_CONT "\n");
+ printk(KERN_INFO " ");
}
- printk("%02x ", sense_buffer[k]);
+ printk(KERN_CONT " %02x", sense_buffer[k]);
}
- printk("\n");
+ printk(KERN_CONT "\n");
return;
}
}
@@ -1364,22 +1373,22 @@ scsi_decode_sense_extras(const unsigned char *sense_buffer, int sense_len,
res += snprintf(buff + res, blen - res, "ILI");
}
if (res > 0)
- printk("%s\n", buff);
+ printk(KERN_CONT "%s\n", buff);
} else if (sshdr->additional_length > 0) {
/* descriptor format with sense descriptors */
num = 8 + sshdr->additional_length;
num = (sense_len < num) ? sense_len : num;
- printk("Descriptor sense data with sense descriptors "
+ printk(KERN_CONT "Descriptor sense data with sense descriptors "
"(in hex):");
for (k = 0; k < num; ++k) {
if (0 == (k % 16)) {
- printk("\n");
- printk(KERN_INFO " ");
+ printk(KERN_CONT "\n");
+ printk(KERN_INFO " ");
}
- printk("%02x ", sense_buffer[k]);
+ printk(KERN_CONT " %02x", sense_buffer[k]);
}
- printk("\n");
+ printk(KERN_CONT "\n");
}
}
@@ -1404,13 +1413,13 @@ void scsi_print_sense(char *name, struct scsi_cmnd *cmd)
{
struct scsi_sense_hdr sshdr;
- scmd_printk(KERN_INFO, cmd, "");
+ scmd_printk(KERN_INFO, cmd, " ");
scsi_decode_sense_buffer(cmd->sense_buffer, SCSI_SENSE_BUFFERSIZE,
&sshdr);
scsi_show_sense_hdr(&sshdr);
scsi_decode_sense_extras(cmd->sense_buffer, SCSI_SENSE_BUFFERSIZE,
&sshdr);
- scmd_printk(KERN_INFO, cmd, "");
+ scmd_printk(KERN_INFO, cmd, " ");
scsi_show_extd_sense(sshdr.asc, sshdr.ascq);
}
EXPORT_SYMBOL(scsi_print_sense);
@@ -1443,7 +1452,7 @@ void scsi_show_result(int result)
void scsi_show_result(int result)
{
- printk("Result: hostbyte=0x%02x driverbyte=0x%02x\n",
+ printk(KERN_CONT "Result: hostbyte=0x%02x driverbyte=0x%02x\n",
host_byte(result), driver_byte(result));
}
@@ -1453,7 +1462,7 @@ EXPORT_SYMBOL(scsi_show_result);
void scsi_print_result(struct scsi_cmnd *cmd)
{
- scmd_printk(KERN_INFO, cmd, "");
+ scmd_printk(KERN_INFO, cmd, " ");
scsi_show_result(cmd->result);
}
EXPORT_SYMBOL(scsi_print_result);
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index 829cc37..2fddadd 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -2550,15 +2550,15 @@ module_exit(exit_sd);
static void sd_print_sense_hdr(struct scsi_disk *sdkp,
struct scsi_sense_hdr *sshdr)
{
- sd_printk(KERN_INFO, sdkp, "");
+ sd_printk(KERN_INFO, sdkp, " ");
scsi_show_sense_hdr(sshdr);
- sd_printk(KERN_INFO, sdkp, "");
+ sd_printk(KERN_INFO, sdkp, " ");
scsi_show_extd_sense(sshdr->asc, sshdr->ascq);
}
static void sd_print_result(struct scsi_disk *sdkp, int result)
{
- sd_printk(KERN_INFO, sdkp, "");
+ sd_printk(KERN_INFO, sdkp, " ");
scsi_show_result(result);
}
diff --git a/drivers/scsi/sym53c8xx_2/sym_hipd.c b/drivers/scsi/sym53c8xx_2/sym_hipd.c
index a7bc8b7..d740a5b 100644
--- a/drivers/scsi/sym53c8xx_2/sym_hipd.c
+++ b/drivers/scsi/sym53c8xx_2/sym_hipd.c
@@ -72,10 +72,7 @@ static void sym_printl_hex(u_char *p, int n)
static void sym_print_msg(struct sym_ccb *cp, char *label, u_char *msg)
{
- if (label)
- sym_print_addr(cp->cmd, "%s: ", label);
- else
- sym_print_addr(cp->cmd, "");
+ sym_print_addr(cp->cmd, "%s: ", label);
spi_print_msg(msg);
printf("\n");
@@ -4558,7 +4555,8 @@ static void sym_int_sir(struct sym_hcb *np)
switch (np->msgin [2]) {
case M_X_MODIFY_DP:
if (DEBUG_FLAGS & DEBUG_POINTER)
- sym_print_msg(cp, NULL, np->msgin);
+ sym_print_msg(cp, "extended msg ",
+ np->msgin);
tmp = (np->msgin[3]<<24) + (np->msgin[4]<<16) +
(np->msgin[5]<<8) + (np->msgin[6]);
sym_modify_dp(np, tp, cp, tmp);
@@ -4585,7 +4583,7 @@ static void sym_int_sir(struct sym_hcb *np)
*/
case M_IGN_RESIDUE:
if (DEBUG_FLAGS & DEBUG_POINTER)
- sym_print_msg(cp, NULL, np->msgin);
+ sym_print_msg(cp, "half byte ", np->msgin);
if (cp->host_flags & HF_SENSE)
OUTL_DSP(np, SCRIPTA_BA(np, clrack));
else
^ permalink raw reply related
* Re: [PATCH] tproxy: nf_tproxy_assign_sock() can handle tw sockets
From: Avi Kivity @ 2010-07-11 5:19 UTC (permalink / raw)
To: Eric Dumazet
Cc: Felipe W Damasio, David Miller, Patrick McHardy, linux-kernel,
netdev
In-Reply-To: <1278742649.2538.17.camel@edumazet-laptop>
On 07/10/2010 09:17 AM, Eric Dumazet wrote:
>
> Strange thing with your crash report is CR2 value, with unexpected value
> of 000000000b388000 while RAX value is dce8dce85d415d41
>
> Faulting instruction is :
>
> 48 83 b8 b0 00 00 00 00 cmpq $0x0,0xb0(%rax)
>
> So I would have expected CR2 being RAX+0xb0, but its not.
>
Nothing strange about it. You only get page faults and valid cr2 for
canonical addresses (17 high order bits all equal). In this case
rax+0xb0 is not a canonical address, so you got a general protection
fault instead, with cr2 unchanged.
--
I have a truly marvellous patch that fixes the bug which this
signature is too narrow to contain.
^ permalink raw reply
* Re: [PATCH] inet, inet6: make tcp_sendmsg() and tcp_sendpage() through inet_sendmsg() and inet_sendpage()
From: David Miller @ 2010-07-11 6:07 UTC (permalink / raw)
To: dm; +Cc: xiaosuo, kuznet, pekkas, jmorris, yoshfuji, kaber, therbert,
netdev
In-Reply-To: <4C394C5E.2000805@chelsio.com>
From: Dimitris Michailidis <dm@chelsio.com>
Date: Sat, 10 Jul 2010 21:45:18 -0700
> Changli Gao wrote:
>
>> diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
>> index 3ceb025..1e5bad1 100644
>> --- a/net/ipv4/af_inet.c
>> +++ b/net/ipv4/af_inet.c
>> @@ -727,28 +727,31 @@ int inet_sendmsg(struct kiocb *iocb, struct
>> socket *sock, struct msghdr *msg,
>> sock_rps_record_flow(sk);
>> /* We may need to bind the socket. */
>> - if (!inet_sk(sk)->inet_num && inet_autobind(sk))
>> + if (!sk->sk_prot->no_autobind &&
>> + !inet_sk(sk)->inet_num && inet_autobind(sk))
>> return -EAGAIN;
>
> I think it's better to test inet_num first since usually it's not 0.
> With the above ordering protocols with no_autobind==false end up
> making two comparisons.
Agreed, Changli please make this change.
Thanky ou.
^ permalink raw reply
* Re: [PATCH net-next] drivers/scsi: Remove warnings after vsprintf %pV introduction
From: David Miller @ 2010-07-11 6:10 UTC (permalink / raw)
To: joe
Cc: sfr, netdev, linux-next, linux-kernel, gregkh, James.Bottomley,
matthew, linux-scsi
In-Reply-To: <1278824921.1501.37.camel@Joe-Laptop.home>
From: Joe Perches <joe@perches.com>
Date: Sat, 10 Jul 2010 22:08:41 -0700
> On Sat, 2010-07-10 at 19:52 -0700, David Miller wrote:
>> Could you take a stab at this and the other scsi bits that
>> trigger this warning?
>
> Remove warnings introduced by conversions of dev_<level>
> macros to functions.
>
> Compile tested only.
>
> Signed-off-by: Joe Perches <joe@perches.com>
SCSI folks, the background is that we have moved the dev_*() printk
macros to external functions, so that the prefixing printf strings
don't get emitting at every call site.
As a consequence, dev_*() calls that try to use an empty string as the
printf format emit a warning from gcc since an empty constant string
is not a valid printf format.
That's what this change is all about.
Anyways:
Acked-by: David S. Miller <davem@davemloft.net>
^ permalink raw reply
* [PATCH 1/2] net: cleanups
From: Changli Gao @ 2010-07-11 6:41 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Changli Gao
net: cleanups
remove useless blanks.
Signed-off-by: Changli Gao <xiaosuo@gmail.com>
----
include/net/inet_common.h | 55 ++++-------
include/net/tcp.h | 222 +++++++++++++++++-----------------------------
include/net/udp.h | 38 +++----
3 files changed, 123 insertions(+), 192 deletions(-)
diff --git a/include/net/inet_common.h b/include/net/inet_common.h
index 18c7732..140c1ec 100644
--- a/include/net/inet_common.h
+++ b/include/net/inet_common.h
@@ -1,8 +1,8 @@
#ifndef _INET_COMMON_H
#define _INET_COMMON_H
-extern const struct proto_ops inet_stream_ops;
-extern const struct proto_ops inet_dgram_ops;
+extern const struct proto_ops inet_stream_ops;
+extern const struct proto_ops inet_dgram_ops;
/*
* INET4 prototypes used by INET6
@@ -13,37 +13,24 @@ struct sock;
struct sockaddr;
struct socket;
-extern int inet_release(struct socket *sock);
-extern int inet_stream_connect(struct socket *sock,
- struct sockaddr * uaddr,
- int addr_len, int flags);
-extern int inet_dgram_connect(struct socket *sock,
- struct sockaddr * uaddr,
- int addr_len, int flags);
-extern int inet_accept(struct socket *sock,
- struct socket *newsock, int flags);
-extern int inet_sendmsg(struct kiocb *iocb,
- struct socket *sock,
- struct msghdr *msg,
- size_t size);
-extern int inet_shutdown(struct socket *sock, int how);
-extern int inet_listen(struct socket *sock, int backlog);
-
-extern void inet_sock_destruct(struct sock *sk);
-
-extern int inet_bind(struct socket *sock,
- struct sockaddr *uaddr, int addr_len);
-extern int inet_getname(struct socket *sock,
- struct sockaddr *uaddr,
- int *uaddr_len, int peer);
-extern int inet_ioctl(struct socket *sock,
- unsigned int cmd, unsigned long arg);
-
-extern int inet_ctl_sock_create(struct sock **sk,
- unsigned short family,
- unsigned short type,
- unsigned char protocol,
- struct net *net);
+extern int inet_release(struct socket *sock);
+extern int inet_stream_connect(struct socket *sock, struct sockaddr * uaddr,
+ int addr_len, int flags);
+extern int inet_dgram_connect(struct socket *sock, struct sockaddr * uaddr,
+ int addr_len, int flags);
+extern int inet_accept(struct socket *sock, struct socket *newsock, int flags);
+extern int inet_sendmsg(struct kiocb *iocb, struct socket *sock,
+ struct msghdr *msg, size_t size);
+extern int inet_shutdown(struct socket *sock, int how);
+extern int inet_listen(struct socket *sock, int backlog);
+extern void inet_sock_destruct(struct sock *sk);
+extern int inet_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len);
+extern int inet_getname(struct socket *sock, struct sockaddr *uaddr,
+ int *uaddr_len, int peer);
+extern int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg);
+extern int inet_ctl_sock_create(struct sock **sk, unsigned short family,
+ unsigned short type, unsigned char protocol,
+ struct net *net);
static inline void inet_ctl_sock_destroy(struct sock *sk)
{
@@ -51,5 +38,3 @@ static inline void inet_ctl_sock_destroy(struct sock *sk)
}
#endif
-
-
diff --git a/include/net/tcp.h b/include/net/tcp.h
index c2f96c2..33ce524 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -296,45 +296,30 @@ extern struct proto tcp_prot;
#define TCP_ADD_STATS_USER(net, field, val) SNMP_ADD_STATS_USER((net)->mib.tcp_statistics, field, val)
#define TCP_ADD_STATS(net, field, val) SNMP_ADD_STATS((net)->mib.tcp_statistics, field, val)
-extern void tcp_v4_err(struct sk_buff *skb, u32);
-
-extern void tcp_shutdown (struct sock *sk, int how);
-
-extern int tcp_v4_rcv(struct sk_buff *skb);
-
-extern int tcp_v4_remember_stamp(struct sock *sk);
-
-extern int tcp_v4_tw_remember_stamp(struct inet_timewait_sock *tw);
-
-extern int tcp_sendmsg(struct kiocb *iocb, struct socket *sock,
- struct msghdr *msg, size_t size);
-extern ssize_t tcp_sendpage(struct socket *sock, struct page *page, int offset, size_t size, int flags);
-
-extern int tcp_ioctl(struct sock *sk,
- int cmd,
- unsigned long arg);
-
-extern int tcp_rcv_state_process(struct sock *sk,
- struct sk_buff *skb,
- struct tcphdr *th,
- unsigned len);
-
-extern int tcp_rcv_established(struct sock *sk,
- struct sk_buff *skb,
- struct tcphdr *th,
- unsigned len);
-
-extern void tcp_rcv_space_adjust(struct sock *sk);
-
-extern void tcp_cleanup_rbuf(struct sock *sk, int copied);
-
-extern int tcp_twsk_unique(struct sock *sk,
- struct sock *sktw, void *twp);
-
-extern void tcp_twsk_destructor(struct sock *sk);
-
-extern ssize_t tcp_splice_read(struct socket *sk, loff_t *ppos,
- struct pipe_inode_info *pipe, size_t len, unsigned int flags);
+extern void tcp_v4_err(struct sk_buff *skb, u32);
+
+extern void tcp_shutdown (struct sock *sk, int how);
+
+extern int tcp_v4_rcv(struct sk_buff *skb);
+
+extern int tcp_v4_remember_stamp(struct sock *sk);
+extern int tcp_v4_tw_remember_stamp(struct inet_timewait_sock *tw);
+extern int tcp_sendmsg(struct kiocb *iocb, struct socket *sock,
+ struct msghdr *msg, size_t size);
+extern ssize_t tcp_sendpage(struct socket *sock, struct page *page, int offset,
+ size_t size, int flags);
+extern int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg);
+extern int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
+ struct tcphdr *th, unsigned len);
+extern int tcp_rcv_established(struct sock *sk, struct sk_buff *skb,
+ struct tcphdr *th, unsigned len);
+extern void tcp_rcv_space_adjust(struct sock *sk);
+extern void tcp_cleanup_rbuf(struct sock *sk, int copied);
+extern int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp);
+extern void tcp_twsk_destructor(struct sock *sk);
+extern ssize_t tcp_splice_read(struct socket *sk, loff_t *ppos,
+ struct pipe_inode_info *pipe, size_t len,
+ unsigned int flags);
static inline void tcp_dec_quickack_mode(struct sock *sk,
const unsigned int pkts)
@@ -372,88 +357,59 @@ enum tcp_tw_status {
};
-extern enum tcp_tw_status tcp_timewait_state_process(struct inet_timewait_sock *tw,
- struct sk_buff *skb,
- const struct tcphdr *th);
-
-extern struct sock * tcp_check_req(struct sock *sk,struct sk_buff *skb,
- struct request_sock *req,
- struct request_sock **prev);
-extern int tcp_child_process(struct sock *parent,
- struct sock *child,
- struct sk_buff *skb);
-extern int tcp_use_frto(struct sock *sk);
-extern void tcp_enter_frto(struct sock *sk);
-extern void tcp_enter_loss(struct sock *sk, int how);
-extern void tcp_clear_retrans(struct tcp_sock *tp);
-extern void tcp_update_metrics(struct sock *sk);
-
-extern void tcp_close(struct sock *sk,
- long timeout);
-extern unsigned int tcp_poll(struct file * file, struct socket *sock, struct poll_table_struct *wait);
-
-extern int tcp_getsockopt(struct sock *sk, int level,
- int optname,
- char __user *optval,
- int __user *optlen);
-extern int tcp_setsockopt(struct sock *sk, int level,
- int optname, char __user *optval,
- unsigned int optlen);
-extern int compat_tcp_getsockopt(struct sock *sk,
- int level, int optname,
- char __user *optval, int __user *optlen);
-extern int compat_tcp_setsockopt(struct sock *sk,
- int level, int optname,
- char __user *optval, unsigned int optlen);
-extern void tcp_set_keepalive(struct sock *sk, int val);
-extern void tcp_syn_ack_timeout(struct sock *sk,
- struct request_sock *req);
-extern int tcp_recvmsg(struct kiocb *iocb, struct sock *sk,
- struct msghdr *msg,
- size_t len, int nonblock,
- int flags, int *addr_len);
-
-extern void tcp_parse_options(struct sk_buff *skb,
- struct tcp_options_received *opt_rx,
- u8 **hvpp,
- int estab);
-
-extern u8 *tcp_parse_md5sig_option(struct tcphdr *th);
+extern enum tcp_tw_status tcp_timewait_state_process(struct inet_timewait_sock *tw,
+ struct sk_buff *skb,
+ const struct tcphdr *th);
+extern struct sock * tcp_check_req(struct sock *sk,struct sk_buff *skb,
+ struct request_sock *req,
+ struct request_sock **prev);
+extern int tcp_child_process(struct sock *parent, struct sock *child,
+ struct sk_buff *skb);
+extern int tcp_use_frto(struct sock *sk);
+extern void tcp_enter_frto(struct sock *sk);
+extern void tcp_enter_loss(struct sock *sk, int how);
+extern void tcp_clear_retrans(struct tcp_sock *tp);
+extern void tcp_update_metrics(struct sock *sk);
+extern void tcp_close(struct sock *sk, long timeout);
+extern unsigned int tcp_poll(struct file * file, struct socket *sock,
+ struct poll_table_struct *wait);
+extern int tcp_getsockopt(struct sock *sk, int level, int optname,
+ char __user *optval, int __user *optlen);
+extern int tcp_setsockopt(struct sock *sk, int level, int optname,
+ char __user *optval, unsigned int optlen);
+extern int compat_tcp_getsockopt(struct sock *sk, int level, int optname,
+ char __user *optval, int __user *optlen);
+extern int compat_tcp_setsockopt(struct sock *sk, int level, int optname,
+ char __user *optval, unsigned int optlen);
+extern void tcp_set_keepalive(struct sock *sk, int val);
+extern void tcp_syn_ack_timeout(struct sock *sk, struct request_sock *req);
+extern int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
+ size_t len, int nonblock, int flags, int *addr_len);
+extern void tcp_parse_options(struct sk_buff *skb,
+ struct tcp_options_received *opt_rx, u8 **hvpp,
+ int estab);
+extern u8 *tcp_parse_md5sig_option(struct tcphdr *th);
/*
* TCP v4 functions exported for the inet6 API
*/
-extern void tcp_v4_send_check(struct sock *sk,
- struct sk_buff *skb);
-
-extern int tcp_v4_conn_request(struct sock *sk,
- struct sk_buff *skb);
-
-extern struct sock * tcp_create_openreq_child(struct sock *sk,
- struct request_sock *req,
- struct sk_buff *skb);
-
-extern struct sock * tcp_v4_syn_recv_sock(struct sock *sk,
- struct sk_buff *skb,
- struct request_sock *req,
- struct dst_entry *dst);
-
-extern int tcp_v4_do_rcv(struct sock *sk,
+extern void tcp_v4_send_check(struct sock *sk, struct sk_buff *skb);
+extern int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb);
+extern struct sock * tcp_create_openreq_child(struct sock *sk,
+ struct request_sock *req,
struct sk_buff *skb);
-
-extern int tcp_v4_connect(struct sock *sk,
- struct sockaddr *uaddr,
- int addr_len);
-
-extern int tcp_connect(struct sock *sk);
-
-extern struct sk_buff * tcp_make_synack(struct sock *sk,
- struct dst_entry *dst,
- struct request_sock *req,
- struct request_values *rvp);
-
-extern int tcp_disconnect(struct sock *sk, int flags);
+extern struct sock * tcp_v4_syn_recv_sock(struct sock *sk, struct sk_buff *skb,
+ struct request_sock *req,
+ struct dst_entry *dst);
+extern int tcp_v4_do_rcv(struct sock *sk, struct sk_buff *skb);
+extern int tcp_v4_connect(struct sock *sk, struct sockaddr *uaddr,
+ int addr_len);
+extern int tcp_connect(struct sock *sk);
+extern struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst,
+ struct request_sock *req,
+ struct request_values *rvp);
+extern int tcp_disconnect(struct sock *sk, int flags);
/* From syncookies.c */
@@ -485,10 +441,10 @@ extern int tcp_fragment(struct sock *, struct sk_buff *, u32, unsigned int);
extern void tcp_send_probe0(struct sock *);
extern void tcp_send_partial(struct sock *);
-extern int tcp_write_wakeup(struct sock *);
+extern int tcp_write_wakeup(struct sock *);
extern void tcp_send_fin(struct sock *sk);
extern void tcp_send_active_reset(struct sock *sk, gfp_t priority);
-extern int tcp_send_synack(struct sock *);
+extern int tcp_send_synack(struct sock *);
extern void tcp_push_one(struct sock *, unsigned int mss_now);
extern void tcp_send_ack(struct sock *sk);
extern void tcp_send_delayed_ack(struct sock *sk);
@@ -592,7 +548,7 @@ static inline u32 tcp_receive_window(const struct tcp_sock *tp)
* scaling applied to the result. The caller does these things
* if necessary. This is a "raw" window selection.
*/
-extern u32 __tcp_select_window(struct sock *sk);
+extern u32 __tcp_select_window(struct sock *sk);
/* TCP timestamps are only 32-bits, this causes a slight
* complication on 64-bit systems since we store a snapshot
@@ -1174,22 +1130,14 @@ struct tcp_md5sig_pool {
#define TCP_MD5SIG_MAXKEYS (~(u32)0) /* really?! */
/* - functions */
-extern int tcp_v4_md5_hash_skb(char *md5_hash,
- struct tcp_md5sig_key *key,
- struct sock *sk,
- struct request_sock *req,
- struct sk_buff *skb);
-
-extern struct tcp_md5sig_key *tcp_v4_md5_lookup(struct sock *sk,
- struct sock *addr_sk);
-
-extern int tcp_v4_md5_do_add(struct sock *sk,
- __be32 addr,
- u8 *newkey,
- u8 newkeylen);
-
-extern int tcp_v4_md5_do_del(struct sock *sk,
- __be32 addr);
+extern int tcp_v4_md5_hash_skb(char *md5_hash, struct tcp_md5sig_key *key,
+ struct sock *sk, struct request_sock *req,
+ struct sk_buff *skb);
+extern struct tcp_md5sig_key * tcp_v4_md5_lookup(struct sock *sk,
+ struct sock *addr_sk);
+extern int tcp_v4_md5_do_add(struct sock *sk, __be32 addr, u8 *newkey,
+ u8 newkeylen);
+extern int tcp_v4_md5_do_del(struct sock *sk, __be32 addr);
#ifdef CONFIG_TCP_MD5SIG
#define tcp_twsk_md5_key(twsk) ((twsk)->tw_md5_keylen ? \
@@ -1202,10 +1150,10 @@ extern int tcp_v4_md5_do_del(struct sock *sk,
#endif
extern struct tcp_md5sig_pool * __percpu *tcp_alloc_md5sig_pool(struct sock *);
-extern void tcp_free_md5sig_pool(void);
+extern void tcp_free_md5sig_pool(void);
extern struct tcp_md5sig_pool *tcp_get_md5sig_pool(void);
-extern void tcp_put_md5sig_pool(void);
+extern void tcp_put_md5sig_pool(void);
extern int tcp_md5_hash_header(struct tcp_md5sig_pool *, struct tcphdr *);
extern int tcp_md5_hash_skb_data(struct tcp_md5sig_pool *, struct sk_buff *,
@@ -1433,7 +1381,7 @@ extern int tcp_gro_complete(struct sk_buff *skb);
extern int tcp4_gro_complete(struct sk_buff *skb);
#ifdef CONFIG_PROC_FS
-extern int tcp4_proc_init(void);
+extern int tcp4_proc_init(void);
extern void tcp4_proc_exit(void);
#endif
diff --git a/include/net/udp.h b/include/net/udp.h
index 5348d80..7abdf30 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -157,30 +157,28 @@ static inline void udp_lib_close(struct sock *sk, long timeout)
sk_common_release(sk);
}
-extern int udp_lib_get_port(struct sock *sk, unsigned short snum,
- int (*)(const struct sock *,const struct sock *),
- unsigned int hash2_nulladdr);
+extern int udp_lib_get_port(struct sock *sk, unsigned short snum,
+ int (*)(const struct sock *,const struct sock *),
+ unsigned int hash2_nulladdr);
/* net/ipv4/udp.c */
-extern int udp_get_port(struct sock *sk, unsigned short snum,
- int (*saddr_cmp)(const struct sock *, const struct sock *));
-extern void udp_err(struct sk_buff *, u32);
-
-extern int udp_sendmsg(struct kiocb *iocb, struct sock *sk,
+extern int udp_get_port(struct sock *sk, unsigned short snum,
+ int (*saddr_cmp)(const struct sock *,
+ const struct sock *));
+extern void udp_err(struct sk_buff *, u32);
+extern int udp_sendmsg(struct kiocb *iocb, struct sock *sk,
struct msghdr *msg, size_t len);
-extern void udp_flush_pending_frames(struct sock *sk);
-
-extern int udp_rcv(struct sk_buff *skb);
-extern int udp_ioctl(struct sock *sk, int cmd, unsigned long arg);
-extern int udp_disconnect(struct sock *sk, int flags);
+extern void udp_flush_pending_frames(struct sock *sk);
+extern int udp_rcv(struct sk_buff *skb);
+extern int udp_ioctl(struct sock *sk, int cmd, unsigned long arg);
+extern int udp_disconnect(struct sock *sk, int flags);
extern unsigned int udp_poll(struct file *file, struct socket *sock,
poll_table *wait);
-extern int udp_lib_getsockopt(struct sock *sk, int level, int optname,
- char __user *optval, int __user *optlen);
-extern int udp_lib_setsockopt(struct sock *sk, int level, int optname,
- char __user *optval, unsigned int optlen,
- int (*push_pending_frames)(struct sock *));
-
+extern int udp_lib_getsockopt(struct sock *sk, int level, int optname,
+ char __user *optval, int __user *optlen);
+extern int udp_lib_setsockopt(struct sock *sk, int level, int optname,
+ char __user *optval, unsigned int optlen,
+ int (*push_pending_frames)(struct sock *));
extern struct sock *udp4_lib_lookup(struct net *net, __be32 saddr, __be16 sport,
__be32 daddr, __be16 dport,
int dif);
@@ -236,7 +234,7 @@ struct udp_iter_state {
extern int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo);
extern void udp_proc_unregister(struct net *net, struct udp_seq_afinfo *afinfo);
-extern int udp4_proc_init(void);
+extern int udp4_proc_init(void);
extern void udp4_proc_exit(void);
#endif
^ permalink raw reply related
* [PATCH 2/2] inet, inet6: make tcp_sendmsg() and tcp_sendpage() through inet_sendmsg() and inet_sendpage()
From: Changli Gao @ 2010-07-11 6:41 UTC (permalink / raw)
To: David S. Miller
Cc: Alexey Kuznetsov, Pekka Savola (ipv6), James Morris,
Hideaki YOSHIFUJI, Patrick McHardy, Tom Herbert, netdev,
Changli Gao
inet, inet6: make tcp_sendmsg() and tcp_sendpage() through inet_sendmsg() and
inet_sendpage()
a new boolean flag no_autobind is added to structure proto to avoid the autobind
calls when the protocol is TCP. Then sock_rps_record_flow() is called int the
TCP's sendmsg() and sendpage() pathes.
Signed-off-by: Changli Gao <xiaosuo@gmail.com>
----
include/net/inet_common.h | 4 ++++
include/net/sock.h | 1 +
include/net/tcp.h | 8 ++++----
net/ipv4/af_inet.c | 15 +++++++++------
net/ipv4/tcp.c | 11 +++++------
net/ipv4/tcp_ipv4.c | 3 +++
net/ipv6/af_inet6.c | 8 ++++----
net/ipv6/tcp_ipv6.c | 3 +++
8 files changed, 33 insertions(+), 20 deletions(-)
diff --git a/include/net/inet_common.h b/include/net/inet_common.h
index 140c1ec..22fac98 100644
--- a/include/net/inet_common.h
+++ b/include/net/inet_common.h
@@ -21,6 +21,10 @@ extern int inet_dgram_connect(struct socket *sock, struct sockaddr * uaddr,
extern int inet_accept(struct socket *sock, struct socket *newsock, int flags);
extern int inet_sendmsg(struct kiocb *iocb, struct socket *sock,
struct msghdr *msg, size_t size);
+extern ssize_t inet_sendpage(struct socket *sock, struct page *page, int offset,
+ size_t size, int flags);
+extern int inet_recvmsg(struct kiocb *iocb, struct socket *sock,
+ struct msghdr *msg, size_t size, int flags);
extern int inet_shutdown(struct socket *sock, int how);
extern int inet_listen(struct socket *sock, int backlog);
extern void inet_sock_destruct(struct sock *sk);
diff --git a/include/net/sock.h b/include/net/sock.h
index 4f26f2f..3100e71 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -772,6 +772,7 @@ struct proto {
int *sysctl_wmem;
int *sysctl_rmem;
int max_header;
+ bool no_autobind;
struct kmem_cache *slab;
unsigned int obj_size;
diff --git a/include/net/tcp.h b/include/net/tcp.h
index 33ce524..468b01f 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -304,10 +304,10 @@ extern int tcp_v4_rcv(struct sk_buff *skb);
extern int tcp_v4_remember_stamp(struct sock *sk);
extern int tcp_v4_tw_remember_stamp(struct inet_timewait_sock *tw);
-extern int tcp_sendmsg(struct kiocb *iocb, struct socket *sock,
- struct msghdr *msg, size_t size);
-extern ssize_t tcp_sendpage(struct socket *sock, struct page *page, int offset,
- size_t size, int flags);
+extern int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
+ size_t size);
+extern int tcp_sendpage(struct sock *sk, struct page *page, int offset,
+ size_t size, int flags);
extern int tcp_ioctl(struct sock *sk, int cmd, unsigned long arg);
extern int tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb,
struct tcphdr *th, unsigned len);
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 3ceb025..6a1100c 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -727,28 +727,31 @@ int inet_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
sock_rps_record_flow(sk);
/* We may need to bind the socket. */
- if (!inet_sk(sk)->inet_num && inet_autobind(sk))
+ if (!inet_sk(sk)->inet_num && !sk->sk_prot->no_autobind &&
+ inet_autobind(sk))
return -EAGAIN;
return sk->sk_prot->sendmsg(iocb, sk, msg, size);
}
EXPORT_SYMBOL(inet_sendmsg);
-static ssize_t inet_sendpage(struct socket *sock, struct page *page, int offset,
- size_t size, int flags)
+ssize_t inet_sendpage(struct socket *sock, struct page *page, int offset,
+ size_t size, int flags)
{
struct sock *sk = sock->sk;
sock_rps_record_flow(sk);
/* We may need to bind the socket. */
- if (!inet_sk(sk)->inet_num && inet_autobind(sk))
+ if (!inet_sk(sk)->inet_num && !sk->sk_prot->no_autobind &&
+ inet_autobind(sk))
return -EAGAIN;
if (sk->sk_prot->sendpage)
return sk->sk_prot->sendpage(sk, page, offset, size, flags);
return sock_no_sendpage(sock, page, offset, size, flags);
}
+EXPORT_SYMBOL(inet_sendpage);
int inet_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
size_t size, int flags)
@@ -894,10 +897,10 @@ const struct proto_ops inet_stream_ops = {
.shutdown = inet_shutdown,
.setsockopt = sock_common_setsockopt,
.getsockopt = sock_common_getsockopt,
- .sendmsg = tcp_sendmsg,
+ .sendmsg = inet_sendmsg,
.recvmsg = inet_recvmsg,
.mmap = sock_no_mmap,
- .sendpage = tcp_sendpage,
+ .sendpage = inet_sendpage,
.splice_read = tcp_splice_read,
#ifdef CONFIG_COMPAT
.compat_setsockopt = compat_sock_common_setsockopt,
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 4e6ddfb..f21e03c 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -856,15 +856,15 @@ out_err:
return sk_stream_error(sk, flags, err);
}
-ssize_t tcp_sendpage(struct socket *sock, struct page *page, int offset,
- size_t size, int flags)
+int tcp_sendpage(struct sock *sk, struct page *page, int offset,
+ size_t size, int flags)
{
ssize_t res;
- struct sock *sk = sock->sk;
if (!(sk->sk_route_caps & NETIF_F_SG) ||
!(sk->sk_route_caps & NETIF_F_ALL_CSUM))
- return sock_no_sendpage(sock, page, offset, size, flags);
+ return sock_no_sendpage(sk->sk_socket, page, offset, size,
+ flags);
lock_sock(sk);
TCP_CHECK_TIMER(sk);
@@ -897,10 +897,9 @@ static inline int select_size(struct sock *sk, int sg)
return tmp;
}
-int tcp_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
+int tcp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
size_t size)
{
- struct sock *sk = sock->sk;
struct iovec *iov;
struct tcp_sock *tp = tcp_sk(sk);
struct sk_buff *skb;
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index 8fa32f5..f7eb3a6 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -2595,6 +2595,8 @@ struct proto tcp_prot = {
.setsockopt = tcp_setsockopt,
.getsockopt = tcp_getsockopt,
.recvmsg = tcp_recvmsg,
+ .sendmsg = tcp_sendmsg,
+ .sendpage = tcp_sendpage,
.backlog_rcv = tcp_v4_do_rcv,
.hash = inet_hash,
.unhash = inet_unhash,
@@ -2613,6 +2615,7 @@ struct proto tcp_prot = {
.twsk_prot = &tcp_timewait_sock_ops,
.rsk_prot = &tcp_request_sock_ops,
.h.hashinfo = &tcp_hashinfo,
+ .no_autobind = true,
#ifdef CONFIG_COMPAT
.compat_setsockopt = compat_tcp_setsockopt,
.compat_getsockopt = compat_tcp_getsockopt,
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index e830cd4..56b9bf2 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -522,10 +522,10 @@ const struct proto_ops inet6_stream_ops = {
.shutdown = inet_shutdown, /* ok */
.setsockopt = sock_common_setsockopt, /* ok */
.getsockopt = sock_common_getsockopt, /* ok */
- .sendmsg = tcp_sendmsg, /* ok */
- .recvmsg = sock_common_recvmsg, /* ok */
+ .sendmsg = inet_sendmsg, /* ok */
+ .recvmsg = inet_recvmsg, /* ok */
.mmap = sock_no_mmap,
- .sendpage = tcp_sendpage,
+ .sendpage = inet_sendpage,
.splice_read = tcp_splice_read,
#ifdef CONFIG_COMPAT
.compat_setsockopt = compat_sock_common_setsockopt,
@@ -549,7 +549,7 @@ const struct proto_ops inet6_dgram_ops = {
.setsockopt = sock_common_setsockopt, /* ok */
.getsockopt = sock_common_getsockopt, /* ok */
.sendmsg = inet_sendmsg, /* ok */
- .recvmsg = sock_common_recvmsg, /* ok */
+ .recvmsg = inet_recvmsg, /* ok */
.mmap = sock_no_mmap,
.sendpage = sock_no_sendpage,
#ifdef CONFIG_COMPAT
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 5ebc27e..fe6d404 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -2142,6 +2142,8 @@ struct proto tcpv6_prot = {
.setsockopt = tcp_setsockopt,
.getsockopt = tcp_getsockopt,
.recvmsg = tcp_recvmsg,
+ .sendmsg = tcp_sendmsg,
+ .sendpage = tcp_sendpage,
.backlog_rcv = tcp_v6_do_rcv,
.hash = tcp_v6_hash,
.unhash = inet_unhash,
@@ -2160,6 +2162,7 @@ struct proto tcpv6_prot = {
.twsk_prot = &tcp6_timewait_sock_ops,
.rsk_prot = &tcp6_request_sock_ops,
.h.hashinfo = &tcp_hashinfo,
+ .no_autobind = true,
#ifdef CONFIG_COMPAT
.compat_setsockopt = compat_tcp_setsockopt,
.compat_getsockopt = compat_tcp_getsockopt,
^ permalink raw reply related
* Re: [PATCH] tproxy: nf_tproxy_assign_sock() can handle tw sockets
From: Felipe W Damasio @ 2010-07-11 7:11 UTC (permalink / raw)
To: David Miller; +Cc: eric.dumazet, kaber, linux-kernel, netdev
In-Reply-To: <AANLkTilQbM8F27a5SLZLMzndLYfz-eRhzyxP18KCFz2K@mail.gmail.com>
Hi,
2010/7/11 Felipe W Damasio <felipewd@gmail.com>:
> What do you mean "from elsewhere"? You mean elsewhere on the network code?
>
> Since the function that had the problem was tcp_recvmsg and we're
> talking about a squid process, we're either talking about a typical
> webserver-objet response, or about about an incorrect/faulty http
> request from the user.
I'm trying to understand the network code to see if I can help track
this down...
So I looked at the code of tcp_recvmsg, and I saw that the function
that calls __kfree_skb is sk_eat_skb, which is called in tcp_recvmsg:
if (tcp_hdr(skb)->fin)
goto found_fin_ok;
if (!(flags & MSG_PEEK)) {
sk_eat_skb(sk, skb, copied_early);
copied_early = 0;
}
continue;
found_fin_ok:
/* Process the FIN. */
++*seq;
if (!(flags & MSG_PEEK)) {
sk_eat_skb(sk, skb, copied_early);
copied_early = 0;
}
break;
Now, I'm no kernel programmer, but I saw the "Process the FIN" thing
on the code. I'm tuning these proc parameters....maybe some of these
are triggering the bug?
The production machine has 8GB of RAM:
echo 1 > /proc/sys/net/ipv4/ip_forward
echo 0 > /proc/sys/net/ipv4/conf/lo/rp_filter
echo 1 > /proc/sys/net/ipv4/ip_nonlocal_bind
echo 1 > /proc/sys/net/ipv4/tcp_low_latency
echo 0 > /proc/sys/net/ipv4/conf/eth2/rp_filter
echo 0 > /proc/sys/net/ipv4/conf/eth1/rp_filter
echo 0 > /proc/sys/net/ipv4/conf/eth0/rp_filter
echo 0 > /proc/sys/net/ipv4/conf/br0/rp_filter
echo 1 > /proc/sys/net/ipv4/conf/all/forwarding
echo 1 > /proc/sys/net/ipv4/conf/all/send_redirects
echo 1 > /proc/sys/net/ipv4/conf/eth0/send_redirects
echo 16384 > /proc/sys/net/ipv4/neigh/default/gc_thresh1
echo 32768 > /proc/sys/net/ipv4/neigh/default/gc_thresh2
echo 65535 > /proc/sys/net/ipv4/neigh/default/gc_thresh3
echo 1024 65535 > /proc/sys/net/ipv4/ip_local_port_range
echo 0 > /proc/sys/net/ipv4/conf/all/rp_filter
echo 0 > /proc/sys/net/ipv4/tcp_ecn
echo 1 > /proc/sys/net/ipv4/tcp_low_latency
echo 100000 > /proc/sys/net/core/netdev_max_backlog
echo 409600 > /proc/sys/net/ipv4/tcp_max_syn_backlog
echo 3 > /proc/sys/net/ipv4/tcp_fin_timeout
echo 15 > /proc/sys/net/ipv4/tcp_keepalive_intvl
echo 3 > /proc/sys/net/ipv4/tcp_keepalive_probes
echo 65536 > /proc/sys/vm/min_free_kbytes
echo "262144 1024000 4194304" > /proc/sys/net/ipv4/tcp_rmem
echo "262144 1024000 4194304" > /proc/sys/net/ipv4/tcp_wmem
echo "1024000" > /proc/sys/net/core/rmem_max
echo "1024000" > /proc/sys/net/core/wmem_max
echo "512000" > /proc/sys/net/core/rmem_default
echo "512000" > /proc/sys/net/core/wmem_default
echo "524288" > /proc/sys/net/ipv4/netfilter/ip_conntrack_max
echo "2" > /proc/sys/net/ipv4/tcp_synack_retries
echo "2" > /proc/sys/net/ipv4/tcp_syn_retries
echo "262144" > /proc/sys/net/ipv4/tcp_max_orphans
echo "262144" > /proc/sys/net/core/somaxconn
Maybe these can help?
Cheers,
Felipe Damasio
^ permalink raw reply
* Re: [PATCH] tproxy: nf_tproxy_assign_sock() can handle tw sockets
From: Felipe W Damasio @ 2010-07-11 7:13 UTC (permalink / raw)
To: David Miller; +Cc: eric.dumazet, kaber, linux-kernel, netdev
In-Reply-To: <AANLkTikqNLp5HN0Puu00_bbJEQCUNCCtSsOFxuxfX5g0@mail.gmail.com>
2010/7/11 Felipe W Damasio <felipewd@gmail.com>:
> The production machine has 8GB of RAM:
I'm sorry, this is not right. The production machine has 16GB of RAM.
Don't know if that matters regarding those proc parameters, though.
Cheers,
Felipe Damasio
^ permalink raw reply
* Re: [PATCH] tproxy: nf_tproxy_assign_sock() can handle tw sockets
From: Eric Dumazet @ 2010-07-11 8:02 UTC (permalink / raw)
To: Avi Kivity
Cc: Felipe W Damasio, David Miller, Patrick McHardy, linux-kernel,
netdev
In-Reply-To: <4C395459.6080407@redhat.com>
Le dimanche 11 juillet 2010 à 08:19 +0300, Avi Kivity a écrit :
> On 07/10/2010 09:17 AM, Eric Dumazet wrote:
> >
> > Strange thing with your crash report is CR2 value, with unexpected value
> > of 000000000b388000 while RAX value is dce8dce85d415d41
> >
> > Faulting instruction is :
> >
> > 48 83 b8 b0 00 00 00 00 cmpq $0x0,0xb0(%rax)
> >
> > So I would have expected CR2 being RAX+0xb0, but its not.
> >
>
> Nothing strange about it. You only get page faults and valid cr2 for
> canonical addresses (17 high order bits all equal). In this case
> rax+0xb0 is not a canonical address, so you got a general protection
> fault instead, with cr2 unchanged.
>
OK, thanks Avi for this information, as I was not aware of this.
So something overwrote sk->sk_prot pointer (or skb->sk pointer) with
some data.
tcp sockets are allocated from a dedicated kmem_cache (because of
SLAB_DESTROY_RCU attribute). Their sk->sk_prot should never change in
normal operation, since underlying memory cannot be reused by another
object type in kernel. It should be NULL or &tcp_prot
Felipe, please describe your configuration as much as possible.
It might be a driver bug with with special kind of network frames.
lsmod
lspci -v
ethtool -k eth0
ethtool -k eth1 (if applicable)
^ permalink raw reply
* Re: [PATCH] tproxy: nf_tproxy_assign_sock() can handle tw sockets
From: Eric Dumazet @ 2010-07-11 8:36 UTC (permalink / raw)
To: David Miller; +Cc: felipewd, kaber, linux-kernel, netdev
In-Reply-To: <20100710.123011.260095467.davem@davemloft.net>
Le samedi 10 juillet 2010 à 12:30 -0700, David Miller a écrit :
> From: Eric Dumazet <eric.dumazet@gmail.com>
> Date: Sat, 10 Jul 2010 08:17:29 +0200
>
> > Strange thing with your crash report is CR2 value, with unexpected value
> > of 000000000b388000 while RAX value is dce8dce85d415d41
> >
> > Faulting instruction is :
> >
> > 48 83 b8 b0 00 00 00 00 cmpq $0x0,0xb0(%rax)
> >
> > So I would have expected CR2 being RAX+0xb0, but its not.
>
> It could be corruption from elsewhere. Those last four hex
> digits (0x5d415d41) are "]A]A" in ascii, but that could just
> be coincidence.
>
x86 being litle endian, string is "A]A]" followed by another "XYXY"
pattern (non ASCII chars : 0xE8, 0xDC, 0xE8, 0xDC, "èÜèÜ" in ISO8859)
^ permalink raw reply
* [PATCH net-next-2.6] net: sock_free() optimizations
From: Eric Dumazet @ 2010-07-11 8:45 UTC (permalink / raw)
To: David Miller; +Cc: netdev
Avoid two extra instructions in sock_free(), to reload
skb->truesize and skb->sk
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
---
diff --git a/net/core/sock.c b/net/core/sock.c
index fef2434..363bc26 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1339,9 +1339,10 @@ EXPORT_SYMBOL(sock_wfree);
void sock_rfree(struct sk_buff *skb)
{
struct sock *sk = skb->sk;
+ unsigned int len = skb->truesize;
- atomic_sub(skb->truesize, &sk->sk_rmem_alloc);
- sk_mem_uncharge(skb->sk, skb->truesize);
+ atomic_sub(len, &sk->sk_rmem_alloc);
+ sk_mem_uncharge(sk, len);
}
EXPORT_SYMBOL(sock_rfree);
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox