From: Peter Zijlstra <a.p.zijlstra@chello.nl>
To: linux-mm@kvack.org, linux-kernel@vger.kernel.org, netdev@vger.kernel.org
Cc: Linus Torvalds <torvalds@osdl.org>, Andrew Morton <akpm@osdl.org>,
David Miller <davem@davemloft.net>,
Rik van Riel <riel@redhat.com>,
Daniel Phillips <phillips@google.com>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Mike Christie <michaelc@cs.wisc.edu>
Subject: [PATCH 18/20] netlink: add SOCK_VMIO support to AF_NETLINK
Date: Tue, 12 Sep 2006 17:25:49 +0200 [thread overview]
Message-ID: <20060912144904.998523000@chello.nl> (raw)
In-Reply-To: 20060912143049.278065000@chello.nl
[-- Attachment #1: netlink_vmio.patch --]
[-- Type: text/plain, Size: 4624 bytes --]
Modify the netlink code so that SOCK_VMIO has the desired effect on the
user-space side of the connection.
Modify sys_{send,recv}msg to use sk->sk_allocation instead of GFP_KERNEL,
this should not change existing behaviour because the default of
sk->sk_allocation is GFP_KERNEL, and no user-space exposed socket would
have it any different at this time.
This change allows the system calls to succeed for SOCK_VMIO sockets
(who have sk->sk_allocation |= GFP_EMERGENCY) even under extreme memory
pressure.
Since netlink_sendmsg is used to transfer msgs from user- to kernel-space
treat the skb allocation there as a receive allocation.
Also export netlink_lookup, this is needed to locate the kernel side struct
sock object associated with the user-space netlink socket.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
CC: David Miller <davem@davemloft.net>
CC: Mike Christie <michaelc@cs.wisc.edu>
---
include/linux/netlink.h | 1 +
net/compat.c | 2 +-
net/netlink/af_netlink.c | 8 +++++---
net/socket.c | 6 +++---
4 files changed, 10 insertions(+), 7 deletions(-)
Index: linux-2.6/net/netlink/af_netlink.c
===================================================================
--- linux-2.6.orig/net/netlink/af_netlink.c
+++ linux-2.6/net/netlink/af_netlink.c
@@ -199,7 +199,7 @@ netlink_unlock_table(void)
wake_up(&nl_table_wait);
}
-static __inline__ struct sock *netlink_lookup(int protocol, u32 pid)
+__inline__ struct sock *netlink_lookup(int protocol, u32 pid)
{
struct nl_pid_hash *hash = &nl_table[protocol].hash;
struct hlist_head *head;
@@ -1147,7 +1147,7 @@ static int netlink_sendmsg(struct kiocb
if (len > sk->sk_sndbuf - 32)
goto out;
err = -ENOBUFS;
- skb = alloc_skb(len, GFP_KERNEL);
+ skb = __alloc_skb(len, GFP_KERNEL, SKB_ALLOC_RX);
if (skb==NULL)
goto out;
@@ -1178,7 +1178,8 @@ static int netlink_sendmsg(struct kiocb
if (dst_group) {
atomic_inc(&skb->users);
- netlink_broadcast(sk, skb, dst_pid, dst_group, GFP_KERNEL);
+ netlink_broadcast(sk, skb, dst_pid, dst_group,
+ sk->sk_allocation);
}
err = netlink_unicast(sk, skb, dst_pid, msg->msg_flags&MSG_DONTWAIT);
@@ -1788,6 +1789,7 @@ panic:
core_initcall(netlink_proto_init);
+EXPORT_SYMBOL(netlink_lookup);
EXPORT_SYMBOL(netlink_ack);
EXPORT_SYMBOL(netlink_run_queue);
EXPORT_SYMBOL(netlink_queue_skip);
Index: linux-2.6/net/socket.c
===================================================================
--- linux-2.6.orig/net/socket.c
+++ linux-2.6/net/socket.c
@@ -1790,7 +1790,7 @@ asmlinkage long sys_sendmsg(int fd, stru
err = -ENOMEM;
iov_size = msg_sys.msg_iovlen * sizeof(struct iovec);
if (msg_sys.msg_iovlen > UIO_FASTIOV) {
- iov = sock_kmalloc(sock->sk, iov_size, GFP_KERNEL);
+ iov = sock_kmalloc(sock->sk, iov_size, sock->sk->sk_allocation);
if (!iov)
goto out_put;
}
@@ -1818,7 +1818,7 @@ asmlinkage long sys_sendmsg(int fd, stru
} else if (ctl_len) {
if (ctl_len > sizeof(ctl))
{
- ctl_buf = sock_kmalloc(sock->sk, ctl_len, GFP_KERNEL);
+ ctl_buf = sock_kmalloc(sock->sk, ctl_len, sock->sk->sk_allocation);
if (ctl_buf == NULL)
goto out_freeiov;
}
@@ -1891,7 +1891,7 @@ asmlinkage long sys_recvmsg(int fd, stru
err = -ENOMEM;
iov_size = msg_sys.msg_iovlen * sizeof(struct iovec);
if (msg_sys.msg_iovlen > UIO_FASTIOV) {
- iov = sock_kmalloc(sock->sk, iov_size, GFP_KERNEL);
+ iov = sock_kmalloc(sock->sk, iov_size, sock->sk->sk_allocation);
if (!iov)
goto out_put;
}
Index: linux-2.6/include/linux/netlink.h
===================================================================
--- linux-2.6.orig/include/linux/netlink.h
+++ linux-2.6/include/linux/netlink.h
@@ -150,6 +150,7 @@ struct netlink_skb_parms
#define NETLINK_CREDS(skb) (&NETLINK_CB((skb)).creds)
+extern struct sock *netlink_lookup(int protocol, __u32 pid);
extern struct sock *netlink_kernel_create(int unit, unsigned int groups, void (*input)(struct sock *sk, int len), struct module *module);
extern void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err);
extern int netlink_has_listeners(struct sock *sk, unsigned int group);
Index: linux-2.6/net/compat.c
===================================================================
--- linux-2.6.orig/net/compat.c
+++ linux-2.6/net/compat.c
@@ -170,7 +170,7 @@ int cmsghdr_from_user_compat_to_kern(str
* from the user.
*/
if (kcmlen > stackbuf_size)
- kcmsg_base = kcmsg = sock_kmalloc(sk, kcmlen, GFP_KERNEL);
+ kcmsg_base = kcmsg = sock_kmalloc(sk, kcmlen, sk->sk_allocation);
if (kcmsg == NULL)
return -ENOBUFS;
--
WARNING: multiple messages have this Message-ID (diff)
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
To: linux-mm@kvack.org, linux-kernel@vger.kernel.org, netdev@vger.kernel.org
Cc: Linus Torvalds <torvalds@osdl.org>, Andrew Morton <akpm@osdl.org>,
David Miller <davem@davemloft.net>,
Rik van Riel <riel@redhat.com>,
Daniel Phillips <phillips@google.com>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Mike Christie <michaelc@cs.wisc.edu>
Subject: [PATCH 18/20] netlink: add SOCK_VMIO support to AF_NETLINK
Date: Tue, 12 Sep 2006 17:25:49 +0200 [thread overview]
Message-ID: <20060912144904.998523000@chello.nl> (raw)
In-Reply-To: 20060912143049.278065000@chello.nl
[-- Attachment #1: netlink_vmio.patch --]
[-- Type: text/plain, Size: 4849 bytes --]
Modify the netlink code so that SOCK_VMIO has the desired effect on the
user-space side of the connection.
Modify sys_{send,recv}msg to use sk->sk_allocation instead of GFP_KERNEL,
this should not change existing behaviour because the default of
sk->sk_allocation is GFP_KERNEL, and no user-space exposed socket would
have it any different at this time.
This change allows the system calls to succeed for SOCK_VMIO sockets
(who have sk->sk_allocation |= GFP_EMERGENCY) even under extreme memory
pressure.
Since netlink_sendmsg is used to transfer msgs from user- to kernel-space
treat the skb allocation there as a receive allocation.
Also export netlink_lookup, this is needed to locate the kernel side struct
sock object associated with the user-space netlink socket.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
CC: David Miller <davem@davemloft.net>
CC: Mike Christie <michaelc@cs.wisc.edu>
---
include/linux/netlink.h | 1 +
net/compat.c | 2 +-
net/netlink/af_netlink.c | 8 +++++---
net/socket.c | 6 +++---
4 files changed, 10 insertions(+), 7 deletions(-)
Index: linux-2.6/net/netlink/af_netlink.c
===================================================================
--- linux-2.6.orig/net/netlink/af_netlink.c
+++ linux-2.6/net/netlink/af_netlink.c
@@ -199,7 +199,7 @@ netlink_unlock_table(void)
wake_up(&nl_table_wait);
}
-static __inline__ struct sock *netlink_lookup(int protocol, u32 pid)
+__inline__ struct sock *netlink_lookup(int protocol, u32 pid)
{
struct nl_pid_hash *hash = &nl_table[protocol].hash;
struct hlist_head *head;
@@ -1147,7 +1147,7 @@ static int netlink_sendmsg(struct kiocb
if (len > sk->sk_sndbuf - 32)
goto out;
err = -ENOBUFS;
- skb = alloc_skb(len, GFP_KERNEL);
+ skb = __alloc_skb(len, GFP_KERNEL, SKB_ALLOC_RX);
if (skb==NULL)
goto out;
@@ -1178,7 +1178,8 @@ static int netlink_sendmsg(struct kiocb
if (dst_group) {
atomic_inc(&skb->users);
- netlink_broadcast(sk, skb, dst_pid, dst_group, GFP_KERNEL);
+ netlink_broadcast(sk, skb, dst_pid, dst_group,
+ sk->sk_allocation);
}
err = netlink_unicast(sk, skb, dst_pid, msg->msg_flags&MSG_DONTWAIT);
@@ -1788,6 +1789,7 @@ panic:
core_initcall(netlink_proto_init);
+EXPORT_SYMBOL(netlink_lookup);
EXPORT_SYMBOL(netlink_ack);
EXPORT_SYMBOL(netlink_run_queue);
EXPORT_SYMBOL(netlink_queue_skip);
Index: linux-2.6/net/socket.c
===================================================================
--- linux-2.6.orig/net/socket.c
+++ linux-2.6/net/socket.c
@@ -1790,7 +1790,7 @@ asmlinkage long sys_sendmsg(int fd, stru
err = -ENOMEM;
iov_size = msg_sys.msg_iovlen * sizeof(struct iovec);
if (msg_sys.msg_iovlen > UIO_FASTIOV) {
- iov = sock_kmalloc(sock->sk, iov_size, GFP_KERNEL);
+ iov = sock_kmalloc(sock->sk, iov_size, sock->sk->sk_allocation);
if (!iov)
goto out_put;
}
@@ -1818,7 +1818,7 @@ asmlinkage long sys_sendmsg(int fd, stru
} else if (ctl_len) {
if (ctl_len > sizeof(ctl))
{
- ctl_buf = sock_kmalloc(sock->sk, ctl_len, GFP_KERNEL);
+ ctl_buf = sock_kmalloc(sock->sk, ctl_len, sock->sk->sk_allocation);
if (ctl_buf == NULL)
goto out_freeiov;
}
@@ -1891,7 +1891,7 @@ asmlinkage long sys_recvmsg(int fd, stru
err = -ENOMEM;
iov_size = msg_sys.msg_iovlen * sizeof(struct iovec);
if (msg_sys.msg_iovlen > UIO_FASTIOV) {
- iov = sock_kmalloc(sock->sk, iov_size, GFP_KERNEL);
+ iov = sock_kmalloc(sock->sk, iov_size, sock->sk->sk_allocation);
if (!iov)
goto out_put;
}
Index: linux-2.6/include/linux/netlink.h
===================================================================
--- linux-2.6.orig/include/linux/netlink.h
+++ linux-2.6/include/linux/netlink.h
@@ -150,6 +150,7 @@ struct netlink_skb_parms
#define NETLINK_CREDS(skb) (&NETLINK_CB((skb)).creds)
+extern struct sock *netlink_lookup(int protocol, __u32 pid);
extern struct sock *netlink_kernel_create(int unit, unsigned int groups, void (*input)(struct sock *sk, int len), struct module *module);
extern void netlink_ack(struct sk_buff *in_skb, struct nlmsghdr *nlh, int err);
extern int netlink_has_listeners(struct sock *sk, unsigned int group);
Index: linux-2.6/net/compat.c
===================================================================
--- linux-2.6.orig/net/compat.c
+++ linux-2.6/net/compat.c
@@ -170,7 +170,7 @@ int cmsghdr_from_user_compat_to_kern(str
* from the user.
*/
if (kcmlen > stackbuf_size)
- kcmsg_base = kcmsg = sock_kmalloc(sk, kcmlen, GFP_KERNEL);
+ kcmsg_base = kcmsg = sock_kmalloc(sk, kcmlen, sk->sk_allocation);
if (kcmsg == NULL)
return -ENOBUFS;
--
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2006-09-12 15:54 UTC|newest]
Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-09-12 15:25 [PATCH 00/20] vm deadlock avoidance for NFS, NBD and iSCSI (take 7) Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 04/20] mm: methods for teaching filesystems about PG_swapcache pages Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 11/20] nbd: request_fn fixup Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 22:47 ` Jens Axboe
2006-09-12 22:47 ` Jens Axboe
2006-09-13 0:21 ` Jeff Garzik
2006-09-13 0:21 ` Jeff Garzik
2006-09-13 6:14 ` Jens Axboe
2006-09-13 6:14 ` Jens Axboe
2006-09-12 15:25 ` [PATCH 19/20] mm: a process flags to avoid blocking allocations Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 16/20] iscsi: add session context to ep_connect Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 13/20] nbd: use swapdev hook to make swap deadlock free Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 06/20] nfs: teach the NFS client how to treat PG_swapcache pages Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 15/20] iscsi: kernel side tcp connect Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 05/20] uml: rename arch/um remove_mapping() Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 08/20] nfs: enable swap on NFS Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 01/20] mm: serialize access to min_free_kbytes Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 09/20] nfs: make swap on NFS robust Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra [this message]
2006-09-12 15:25 ` [PATCH 18/20] netlink: add SOCK_VMIO support to AF_NETLINK Peter Zijlstra
2006-09-12 15:25 ` [PATCH 02/20] net: vm deadlock avoidance core Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 10/20] mm: block device swap notification Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 20/20] iscsi: support for swapping over iSCSI Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-13 20:50 ` Mike Christie
2006-09-13 20:50 ` Mike Christie
2006-09-14 6:17 ` Peter Zijlstra
2006-09-14 6:17 ` Peter Zijlstra
2006-09-14 19:22 ` Mike Christie
2006-09-14 19:22 ` Mike Christie
2006-09-14 20:35 ` Peter Zijlstra
2006-09-14 20:35 ` Peter Zijlstra
2006-09-14 20:46 ` Peter Zijlstra
2006-09-14 20:46 ` Peter Zijlstra
2006-09-14 21:09 ` Mike Christie
2006-09-14 21:09 ` Mike Christie
2006-09-14 21:28 ` Mike Christie
2006-09-14 21:28 ` Mike Christie
2006-09-14 21:00 ` Mike Christie
2006-09-14 21:00 ` Mike Christie
2006-09-14 21:03 ` Mike Christie
2006-09-14 21:03 ` Mike Christie
2006-09-14 21:18 ` Peter Zijlstra
2006-09-14 21:18 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 17/20] scsi: propagate the swapdev hook into the scsi stack Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 03/20] mm: add support for non block device backed swap files Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 07/20] nfs: add a comment explaining the use of PG_private in the NFS client Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 14/20] uml: enable scsi and add iscsi config Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 15:25 ` [PATCH 12/20] nbd: limit blk_queue Peter Zijlstra
2006-09-12 15:25 ` Peter Zijlstra
2006-09-12 22:47 ` Jens Axboe
2006-09-12 22:47 ` Jens Axboe
2006-09-12 16:37 ` [PATCH 00/20] vm deadlock avoidance for NFS, NBD and iSCSI (take 7) Linus Torvalds
2006-09-12 16:37 ` Linus Torvalds
2006-09-12 23:58 ` Nate Diller
2006-09-12 23:58 ` Nate Diller
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20060912144904.998523000@chello.nl \
--to=a.p.zijlstra@chello.nl \
--cc=akpm@osdl.org \
--cc=davem@davemloft.net \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=michaelc@cs.wisc.edu \
--cc=netdev@vger.kernel.org \
--cc=phillips@google.com \
--cc=riel@redhat.com \
--cc=torvalds@osdl.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.