[PATCH 4/4] nfs: deadlock prevention for NFS

[Peter Zijlstra]

Provide a proper a_ops->swapfile() implementation for NFS. This will
set the NFS socket to SOCK_VMIO and put the socket reconnection under
PF_MEMALLOC (I hope this is enough, otherwise more work needs to be done).

Signed-off-by: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>
---
 fs/nfs/file.c               |   21 ++++++++++++++++++++-
 include/linux/sunrpc/xprt.h |    4 +++-
 net/sunrpc/xprtsock.c       |   16 ++++++++++++++++
 3 files changed, 39 insertions(+), 2 deletions(-)

Index: linux-2.6/fs/nfs/file.c
===================================================================
--- linux-2.6.orig/fs/nfs/file.c
+++ linux-2.6/fs/nfs/file.c
@@ -27,6 +27,7 @@
 #include <linux/slab.h>
 #include <linux/pagemap.h>
 #include <linux/smp_lock.h>
+#include <net/sock.h>
 
 #include <asm/uaccess.h>
 #include <asm/system.h>
@@ -317,7 +318,25 @@ static int nfs_release_page(struct page 
 
 static int nfs_swapfile(struct address_space *mapping, int enable)
 {
-	return 0;
+	int err = -EINVAL;
+	struct rpc_clnt *client = NFS_CLIENT(mapping->host);
+	struct sock *sk = client->cl_xprt->inet;
+
+	if (enable) {
+		client->cl_xprt->swapper = 1;
+		/*
+		 * keep one extra sock reference so the reserve won't dip
+		 * when the socket gets reconnected.
+		 */
+		sk_adjust_memalloc(1, 1);
+		err = sk_set_vmio(sk);
+	} else if (client->cl_xprt->swapper) {
+		client->cl_xprt->swapper = 0;
+		sk_adjust_memalloc(-1, -1);
+		err = sk_clear_vmio(sk);
+	}
+
+	return err;
 }
 
 const struct address_space_operations nfs_file_aops = {
Index: linux-2.6/net/sunrpc/xprtsock.c
===================================================================
--- linux-2.6.orig/net/sunrpc/xprtsock.c
+++ linux-2.6/net/sunrpc/xprtsock.c
@@ -1014,6 +1014,7 @@ static void xs_udp_connect_worker(void *
 {
 	struct rpc_xprt *xprt = (struct rpc_xprt *) args;
 	struct socket *sock = xprt->sock;
+	unsigned long pflags = current->flags;
 	int err, status = -EIO;
 
 	if (xprt->shutdown || xprt->addr.sin_port == 0)
@@ -1021,6 +1022,9 @@ static void xs_udp_connect_worker(void *
 
 	dprintk("RPC:      xs_udp_connect_worker for xprt %p\n", xprt);
 
+	if (xprt->swapper)
+		current->flags |= PF_MEMALLOC;
+
 	/* Start by resetting any existing state */
 	xs_close(xprt);
 
@@ -1054,6 +1058,9 @@ static void xs_udp_connect_worker(void *
 		xprt->sock = sock;
 		xprt->inet = sk;
 
+		if (xprt->swapper)
+			sk_set_vmio(sk);
+
 		write_unlock_bh(&sk->sk_callback_lock);
 	}
 	xs_udp_do_set_buffer_size(xprt);
@@ -1061,6 +1068,7 @@ static void xs_udp_connect_worker(void *
 out:
 	xprt_wake_pending_tasks(xprt, status);
 	xprt_clear_connecting(xprt);
+	current->flags = pflags;
 }
 
 /*
@@ -1097,11 +1105,15 @@ static void xs_tcp_connect_worker(void *
 {
 	struct rpc_xprt *xprt = (struct rpc_xprt *)args;
 	struct socket *sock = xprt->sock;
+	unsigned long pflags = current->flags;
 	int err, status = -EIO;
 
 	if (xprt->shutdown || xprt->addr.sin_port == 0)
 		goto out;
 
+	if (xprt->swapper)
+		current->flags |= PF_MEMALLOC;
+
 	dprintk("RPC:      xs_tcp_connect_worker for xprt %p\n", xprt);
 
 	if (!xprt->sock) {
@@ -1170,10 +1182,14 @@ static void xs_tcp_connect_worker(void *
 				break;
 		}
 	}
+
+	if (xprt->swapper)
+		sk_set_vmio(xprt->inet);
 out:
 	xprt_wake_pending_tasks(xprt, status);
 out_clear:
 	xprt_clear_connecting(xprt);
+	current->flags = pflags;
 }
 
 /**
Index: linux-2.6/include/linux/sunrpc/xprt.h
===================================================================
--- linux-2.6.orig/include/linux/sunrpc/xprt.h
+++ linux-2.6/include/linux/sunrpc/xprt.h
@@ -147,7 +147,9 @@ struct rpc_xprt {
 	unsigned int		max_reqs;	/* total slots */
 	unsigned long		state;		/* transport state */
 	unsigned char		shutdown   : 1,	/* being shut down */
-				resvport   : 1; /* use a reserved port */
+				resvport   : 1, /* use a reserved port */
+				swapper    : 1; /* we're swapping over this
+						   transport */
 
 	/*
 	 * XID
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/