9p: transport API reorganization
Eric Van Hensbergen [Thu, 7 Feb 2008 01:25:03 +0000 (19:25 -0600)]
This merges the mux.c (including the connection interface) with trans_fd
in preparation for transport API changes.  Ultimately, trans_fd will need
to be rewritten to clean it up and simplify the implementation, but this
reorganization is viewed as the first step.

Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>

fs/9p/v9fs.c
fs/9p/v9fs.h
include/net/9p/client.h
include/net/9p/conn.h [deleted file]
include/net/9p/transport.h
net/9p/Makefile
net/9p/client.c
net/9p/mod.c
net/9p/mux.c [deleted file]
net/9p/trans_fd.c
net/9p/trans_virtio.c

index fbb12da..9b0f022 100644 (file)
@@ -3,7 +3,7 @@
  *
  *  This file contains functions assisting in mapping VFS to 9P2000
  *
- *  Copyright (C) 2004 by Eric Van Hensbergen <ericvh@gmail.com>
+ *  Copyright (C) 2004-2008 by Eric Van Hensbergen <ericvh@gmail.com>
  *  Copyright (C) 2002 by Ron Minnich <rminnich@lanl.gov>
  *
  *  This program is free software; you can redistribute it and/or modify
@@ -31,7 +31,6 @@
 #include <linux/idr.h>
 #include <net/9p/9p.h>
 #include <net/9p/transport.h>
-#include <net/9p/conn.h>
 #include <net/9p/client.h>
 #include "v9fs.h"
 #include "v9fs_vfs.h"
 
 enum {
        /* Options that take integer arguments */
-       Opt_debug, Opt_msize, Opt_dfltuid, Opt_dfltgid, Opt_afid,
+       Opt_debug, Opt_dfltuid, Opt_dfltgid, Opt_afid,
        /* String options */
        Opt_uname, Opt_remotename, Opt_trans,
        /* Options that take no arguments */
-       Opt_legacy, Opt_nodevmap,
+       Opt_nodevmap,
        /* Cache options */
        Opt_cache_loose,
        /* Access options */
@@ -58,14 +57,11 @@ enum {
 
 static match_table_t tokens = {
        {Opt_debug, "debug=%x"},
-       {Opt_msize, "msize=%u"},
        {Opt_dfltuid, "dfltuid=%u"},
        {Opt_dfltgid, "dfltgid=%u"},
        {Opt_afid, "afid=%u"},
        {Opt_uname, "uname=%s"},
        {Opt_remotename, "aname=%s"},
-       {Opt_trans, "trans=%s"},
-       {Opt_legacy, "noextend"},
        {Opt_nodevmap, "nodevmap"},
        {Opt_cache_loose, "cache=loose"},
        {Opt_cache_loose, "loose"},
@@ -85,16 +81,14 @@ static void v9fs_parse_options(struct v9fs_session_info *v9ses)
        char *options;
        substring_t args[MAX_OPT_ARGS];
        char *p;
-       int option;
-       int ret;
+       int option = 0;
        char *s, *e;
+       int ret;
 
        /* setup defaults */
-       v9ses->maxdata = 8192;
        v9ses->afid = ~0;
        v9ses->debug = 0;
        v9ses->cache = 0;
-       v9ses->trans = v9fs_default_trans();
 
        if (!v9ses->options)
                return;
@@ -106,7 +100,8 @@ static void v9fs_parse_options(struct v9fs_session_info *v9ses)
                        continue;
                token = match_token(p, tokens, args);
                if (token < Opt_uname) {
-                       if ((ret = match_int(&args[0], &option)) < 0) {
+                       ret = match_int(&args[0], &option);
+                       if (ret < 0) {
                                P9_DPRINTK(P9_DEBUG_ERROR,
                                        "integer field, but no integer?\n");
                                continue;
@@ -119,9 +114,7 @@ static void v9fs_parse_options(struct v9fs_session_info *v9ses)
                        p9_debug_level = option;
 #endif
                        break;
-               case Opt_msize:
-                       v9ses->maxdata = option;
-                       break;
+
                case Opt_dfltuid:
                        v9ses->dfltuid = option;
                        break;
@@ -131,18 +124,12 @@ static void v9fs_parse_options(struct v9fs_session_info *v9ses)
                case Opt_afid:
                        v9ses->afid = option;
                        break;
-               case Opt_trans:
-                       v9ses->trans = v9fs_match_trans(&args[0]);
-                       break;
                case Opt_uname:
                        match_strcpy(v9ses->uname, &args[0]);
                        break;
                case Opt_remotename:
                        match_strcpy(v9ses->aname, &args[0]);
                        break;
-               case Opt_legacy:
-                       v9ses->flags &= ~V9FS_EXTENDED;
-                       break;
                case Opt_nodevmap:
                        v9ses->nodev = 1;
                        break;
@@ -185,7 +172,6 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses,
                  const char *dev_name, char *data)
 {
        int retval = -EINVAL;
-       struct p9_trans *trans = NULL;
        struct p9_fid *fid;
 
        v9ses->uname = __getname();
@@ -207,24 +193,7 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses,
        v9ses->options = kstrdup(data, GFP_KERNEL);
        v9fs_parse_options(v9ses);
 
-       if (v9ses->trans == NULL) {
-               retval = -EPROTONOSUPPORT;
-               P9_DPRINTK(P9_DEBUG_ERROR,
-                               "No transport defined or default transport\n");
-               goto error;
-       }
-
-       trans = v9ses->trans->create(dev_name, v9ses->options);
-       if (IS_ERR(trans)) {
-               retval = PTR_ERR(trans);
-               trans = NULL;
-               goto error;
-       }
-       if ((v9ses->maxdata+P9_IOHDRSZ) > v9ses->trans->maxsize)
-               v9ses->maxdata = v9ses->trans->maxsize-P9_IOHDRSZ;
-
-       v9ses->clnt = p9_client_create(trans, v9ses->maxdata+P9_IOHDRSZ,
-               v9fs_extended(v9ses));
+       v9ses->clnt = p9_client_create(dev_name, v9ses->options);
 
        if (IS_ERR(v9ses->clnt)) {
                retval = PTR_ERR(v9ses->clnt);
@@ -236,6 +205,8 @@ struct p9_fid *v9fs_session_init(struct v9fs_session_info *v9ses,
        if (!v9ses->clnt->dotu)
                v9ses->flags &= ~V9FS_EXTENDED;
 
+       v9ses->maxdata = v9ses->clnt->msize;
+
        /* for legacy mode, fall back to V9FS_ACCESS_ANY */
        if (!v9fs_extended(v9ses) &&
                ((v9ses->flags&V9FS_ACCESS_MASK) == V9FS_ACCESS_USER)) {
index db4b419..7d3a101 100644 (file)
@@ -1,7 +1,7 @@
 /*
  * V9FS definitions.
  *
- *  Copyright (C) 2004 by Eric Van Hensbergen <ericvh@gmail.com>
+ *  Copyright (C) 2004-2008 by Eric Van Hensbergen <ericvh@gmail.com>
  *  Copyright (C) 2002 by Ron Minnich <rminnich@lanl.gov>
  *
  *  This program is free software; you can redistribute it and/or modify
@@ -28,7 +28,6 @@
 
 struct v9fs_session_info {
        /* options */
-       unsigned int maxdata;
        unsigned char flags;    /* session flags */
        unsigned char nodev;    /* set to 1 if no disable device mapping */
        unsigned short debug;   /* debug level */
@@ -38,10 +37,10 @@ struct v9fs_session_info {
        char *options;          /* copy of mount options */
        char *uname;            /* user name to mount as */
        char *aname;            /* name of remote hierarchy being mounted */
+       unsigned int maxdata;   /* max data for client interface */
        unsigned int dfltuid;   /* default uid/muid for legacy support */
        unsigned int dfltgid;   /* default gid for legacy support */
        u32 uid;                /* if ACCESS_SINGLE, the uid that has access */
-       struct p9_trans_module *trans; /* 9p transport */
        struct p9_client *clnt; /* 9p client */
        struct dentry *debugfs_dir;
 };
index 9b9221a..e52f93d 100644 (file)
@@ -3,6 +3,7 @@
  *
  * 9P Client Definitions
  *
+ *  Copyright (C) 2008 by Eric Van Hensbergen <ericvh@gmail.com>
  *  Copyright (C) 2007 by Latchesar Ionkov <lucho@ionkov.net>
  *
  *  This program is free software; you can redistribute it and/or modify
@@ -29,6 +30,7 @@ struct p9_client {
        spinlock_t lock; /* protect client structure */
        int msize;
        unsigned char dotu;
+       struct p9_trans_module *trans_mod;
        struct p9_trans *trans;
        struct p9_conn *conn;
 
@@ -52,8 +54,7 @@ struct p9_fid {
        struct list_head dlist; /* list of all fids attached to a dentry */
 };
 
-struct p9_client *p9_client_create(struct p9_trans *trans, int msize,
-                                                               int dotu);
+struct p9_client *p9_client_create(const char *dev_name, char *options);
 void p9_client_destroy(struct p9_client *clnt);
 void p9_client_disconnect(struct p9_client *clnt);
 struct p9_fid *p9_client_attach(struct p9_client *clnt, struct p9_fid *afid,
diff --git a/include/net/9p/conn.h b/include/net/9p/conn.h
deleted file mode 100644 (file)
index 756d878..0000000
+++ /dev/null
@@ -1,57 +0,0 @@
-/*
- * include/net/9p/conn.h
- *
- * Connection Definitions
- *
- *  Copyright (C) 2005 by Latchesar Ionkov <lucho@ionkov.net>
- *  Copyright (C) 2004 by Eric Van Hensbergen <ericvh@gmail.com>
- *
- *  This program is free software; you can redistribute it and/or modify
- *  it under the terms of the GNU General Public License version 2
- *  as published by the Free Software Foundation.
- *
- *  This program is distributed in the hope that it will be useful,
- *  but WITHOUT ANY WARRANTY; without even the implied warranty of
- *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- *  GNU General Public License for more details.
- *
- *  You should have received a copy of the GNU General Public License
- *  along with this program; if not, write to:
- *  Free Software Foundation
- *  51 Franklin Street, Fifth Floor
- *  Boston, MA  02111-1301  USA
- *
- */
-
-#ifndef NET_9P_CONN_H
-#define NET_9P_CONN_H
-
-#undef P9_NONBLOCK
-
-struct p9_conn;
-struct p9_req;
-
-/**
- * p9_mux_req_callback - callback function that is called when the
- * response of a request is received. The callback is called from
- * a workqueue and shouldn't block.
- *
- * @req - request
- * @a - the pointer that was specified when the request was send to be
- *      passed to the callback
- */
-typedef void (*p9_conn_req_callback)(struct p9_req *req, void *a);
-
-struct p9_conn *p9_conn_create(struct p9_trans *trans, int msize,
-                                                       unsigned char *dotu);
-void p9_conn_destroy(struct p9_conn *);
-int p9_conn_rpc(struct p9_conn *m, struct p9_fcall *tc, struct p9_fcall **rc);
-
-#ifdef P9_NONBLOCK
-int p9_conn_rpcnb(struct p9_conn *m, struct p9_fcall *tc,
-       p9_conn_req_callback cb, void *a);
-#endif /* P9_NONBLOCK */
-
-void p9_conn_cancel(struct p9_conn *m, int err);
-
-#endif /* NET_9P_CONN_H */
index f9f362e..d2209ae 100644 (file)
@@ -4,7 +4,7 @@
  * Transport Definition
  *
  *  Copyright (C) 2005 by Latchesar Ionkov <lucho@ionkov.net>
- *  Copyright (C) 2004 by Eric Van Hensbergen <ericvh@gmail.com>
+ *  Copyright (C) 2004-2008 by Eric Van Hensbergen <ericvh@gmail.com>
  *
  *  This program is free software; you can redistribute it and/or modify
  *  it under the terms of the GNU General Public License version 2
@@ -34,13 +34,12 @@ enum p9_trans_status {
 
 struct p9_trans {
        enum p9_trans_status status;
+       int msize;
+       unsigned char extended;
        void *priv;
-       int (*write) (struct p9_trans *, void *, int);
-       int (*read) (struct p9_trans *, void *, int);
        void (*close) (struct p9_trans *);
-       unsigned int (*poll)(struct p9_trans *, struct poll_table_struct *);
        int (*rpc) (struct p9_trans *t, struct p9_fcall *tc,
-                               struct p9_fcall **rc, int msize, int dotu);
+                                                       struct p9_fcall **rc);
 };
 
 struct p9_trans_module {
@@ -48,7 +47,7 @@ struct p9_trans_module {
        char *name;             /* name of transport */
        int maxsize;            /* max message size of transport */
        int def;                /* this transport should be default */
-       struct p9_trans * (*create)(const char *devname, char *options);
+       struct p9_trans * (*create)(const char *, char *, int, unsigned char);
 };
 
 void v9fs_register_trans(struct p9_trans_module *m);
index d3abb24..8a10511 100644 (file)
@@ -4,7 +4,6 @@ obj-$(CONFIG_NET_9P_VIRTIO) += 9pnet_virtio.o
 
 9pnet-objs := \
        mod.o \
-       mux.o \
        client.o \
        conv.o \
        error.o \
index 433b307..84e087e 100644 (file)
@@ -3,6 +3,7 @@
  *
  * 9P Client
  *
+ *  Copyright (C) 2008 by Eric Van Hensbergen <ericvh@gmail.com>
  *  Copyright (C) 2007 by Latchesar Ionkov <lucho@ionkov.net>
  *
  *  This program is free software; you can redistribute it and/or modify
@@ -25,6 +26,7 @@
 #include <linux/module.h>
 #include <linux/errno.h>
 #include <linux/fs.h>
+#include <linux/poll.h>
 #include <linux/idr.h>
 #include <linux/mutex.h>
 #include <linux/sched.h>
 #include <net/9p/9p.h>
 #include <linux/parser.h>
 #include <net/9p/transport.h>
-#include <net/9p/conn.h>
 #include <net/9p/client.h>
 
 static struct p9_fid *p9_fid_create(struct p9_client *clnt);
 static void p9_fid_destroy(struct p9_fid *fid);
 static struct p9_stat *p9_clone_stat(struct p9_stat *st, int dotu);
 
+/*
+  * Client Option Parsing (code inspired by NFS code)
+  *  - a little lazy - parse all client options
+  */
+
+enum {
+       Opt_msize,
+       Opt_trans,
+       Opt_legacy,
+       Opt_err,
+};
+
+static match_table_t tokens = {
+       {Opt_msize, "msize=%u"},
+       {Opt_legacy, "noextend"},
+       {Opt_trans, "trans=%s"},
+       {Opt_err, NULL},
+};
+
+/**
+ * v9fs_parse_options - parse mount options into session structure
+ * @options: options string passed from mount
+ * @v9ses: existing v9fs session information
+ *
+ */
+
+static void parse_opts(char *options, struct p9_client *clnt)
+{
+       char *p;
+       substring_t args[MAX_OPT_ARGS];
+       int option;
+       int ret;
+
+       clnt->trans_mod = v9fs_default_trans();
+       clnt->dotu = 1;
+       clnt->msize = 8192;
+
+       if (!options)
+               return;
+
+       while ((p = strsep(&options, ",")) != NULL) {
+               int token;
+               if (!*p)
+                       continue;
+               token = match_token(p, tokens, args);
+               if (token < Opt_trans) {
+                       ret = match_int(&args[0], &option);
+                       if (ret < 0) {
+                               P9_DPRINTK(P9_DEBUG_ERROR,
+                                       "integer field, but no integer?\n");
+                               continue;
+                       }
+               }
+               switch (token) {
+               case Opt_msize:
+                       clnt->msize = option;
+                       break;
+               case Opt_trans:
+                       clnt->trans_mod = v9fs_match_trans(&args[0]);
+                       break;
+               case Opt_legacy:
+                       clnt->dotu = 0;
+                       break;
+               default:
+                       continue;
+               }
+       }
+}
+
+
 /**
  * p9_client_rpc - sends 9P request and waits until a response is available.
  *      The function can be interrupted.
@@ -50,14 +121,10 @@ int
 p9_client_rpc(struct p9_client *c, struct p9_fcall *tc,
        struct p9_fcall **rc)
 {
-       if (c->trans->rpc)
-               return c->trans->rpc(c->trans, tc, rc, c->msize, c->dotu);
-       else
-               return p9_conn_rpc(c->conn, tc, rc);
+       return c->trans->rpc(c->trans, tc, rc);
 }
 
-struct p9_client *p9_client_create(struct p9_trans *trans, int msize,
-                                                                  int dotu)
+struct p9_client *p9_client_create(const char *dev_name, char *options)
 {
        int err, n;
        struct p9_client *clnt;
@@ -71,12 +138,7 @@ struct p9_client *p9_client_create(struct p9_trans *trans, int msize,
        if (!clnt)
                return ERR_PTR(-ENOMEM);
 
-       P9_DPRINTK(P9_DEBUG_9P, "clnt %p trans %p msize %d dotu %d\n",
-               clnt, trans, msize, dotu);
        spin_lock_init(&clnt->lock);
-       clnt->trans = trans;
-       clnt->msize = msize;
-       clnt->dotu = dotu;
        INIT_LIST_HEAD(&clnt->fidlist);
        clnt->fidpool = p9_idpool_create();
        if (!clnt->fidpool) {
@@ -85,13 +147,29 @@ struct p9_client *p9_client_create(struct p9_trans *trans, int msize,
                goto error;
        }
 
-       clnt->conn = p9_conn_create(clnt->trans, clnt->msize, &clnt->dotu);
-       if (IS_ERR(clnt->conn)) {
-               err = PTR_ERR(clnt->conn);
-               clnt->conn = NULL;
+       parse_opts(options, clnt);
+       if (clnt->trans_mod == NULL) {
+               err = -EPROTONOSUPPORT;
+               P9_DPRINTK(P9_DEBUG_ERROR,
+                               "No transport defined or default transport\n");
+               goto error;
+       }
+
+       P9_DPRINTK(P9_DEBUG_9P, "clnt %p trans %p msize %d dotu %d\n",
+               clnt, clnt->trans_mod, clnt->msize, clnt->dotu);
+
+
+       clnt->trans = clnt->trans_mod->create(dev_name, options, clnt->msize,
+                                                               clnt->dotu);
+       if (IS_ERR(clnt->trans)) {
+               err = PTR_ERR(clnt->trans);
+               clnt->trans = NULL;
                goto error;
        }
 
+       if ((clnt->msize+P9_IOHDRSZ) > clnt->trans_mod->maxsize)
+               clnt->msize = clnt->trans_mod->maxsize-P9_IOHDRSZ;
+
        tc = p9_create_tversion(clnt->msize, clnt->dotu?"9P2000.u":"9P2000");
        if (IS_ERR(tc)) {
                err = PTR_ERR(tc);
@@ -134,10 +212,6 @@ void p9_client_destroy(struct p9_client *clnt)
        struct p9_fid *fid, *fidptr;
 
        P9_DPRINTK(P9_DEBUG_9P, "clnt %p\n", clnt);
-       if (clnt->conn) {
-               p9_conn_destroy(clnt->conn);
-               clnt->conn = NULL;
-       }
 
        if (clnt->trans) {
                clnt->trans->close(clnt->trans);
@@ -159,7 +233,6 @@ void p9_client_disconnect(struct p9_client *clnt)
 {
        P9_DPRINTK(P9_DEBUG_9P, "clnt %p\n", clnt);
        clnt->trans->status = Disconnected;
-       p9_conn_cancel(clnt->conn, -EIO);
 }
 EXPORT_SYMBOL(p9_client_disconnect);
 
index 8f9763a..c285aab 100644 (file)
@@ -106,15 +106,10 @@ EXPORT_SYMBOL(v9fs_default_trans);
  */
 static int __init init_p9(void)
 {
-       int ret;
+       int ret = 0;
 
        p9_error_init();
        printk(KERN_INFO "Installing 9P2000 support\n");
-       ret = p9_mux_global_init();
-       if (ret) {
-               printk(KERN_WARNING "9p: starting mux failed\n");
-               return ret;
-       }
 
        return ret;
 }
@@ -126,7 +121,7 @@ static int __init init_p9(void)
 
 static void __exit exit_p9(void)
 {
-       p9_mux_global_exit();
+       printk(KERN_INFO "Unloading 9P2000 support\n");
 }
 
 module_init(init_p9)
diff --git a/net/9p/mux.c b/net/9p/mux.c
deleted file mode 100644 (file)
index c9f0805..0000000
+++ /dev/null
@@ -1,1060 +0,0 @@
-/*
- * net/9p/mux.c
- *
- * Protocol Multiplexer
- *
- *  Copyright (C) 2004 by Eric Van Hensbergen <ericvh@gmail.com>
- *  Copyright (C) 2004-2005 by Latchesar Ionkov <lucho@ionkov.net>
- *
- *  This program is free software; you can redistribute it and/or modify
- *  it under the terms of the GNU General Public License version 2
- *  as published by the Free Software Foundation.
- *
- *  This program is distributed in the hope that it will be useful,
- *  but WITHOUT ANY WARRANTY; without even the implied warranty of
- *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- *  GNU General Public License for more details.
- *
- *  You should have received a copy of the GNU General Public License
- *  along with this program; if not, write to:
- *  Free Software Foundation
- *  51 Franklin Street, Fifth Floor
- *  Boston, MA  02111-1301  USA
- *
- */
-
-#include <linux/module.h>
-#include <linux/errno.h>
-#include <linux/fs.h>
-#include <linux/poll.h>
-#include <linux/kthread.h>
-#include <linux/idr.h>
-#include <linux/mutex.h>
-#include <net/9p/9p.h>
-#include <linux/parser.h>
-#include <net/9p/transport.h>
-#include <net/9p/conn.h>
-
-#define ERREQFLUSH     1
-#define SCHED_TIMEOUT  10
-#define MAXPOLLWADDR   2
-
-enum {
-       Rworksched = 1,         /* read work scheduled or running */
-       Rpending = 2,           /* can read */
-       Wworksched = 4,         /* write work scheduled or running */
-       Wpending = 8,           /* can write */
-};
-
-enum {
-       None,
-       Flushing,
-       Flushed,
-};
-
-struct p9_mux_poll_task;
-
-struct p9_req {
-       spinlock_t lock; /* protect request structure */
-       int tag;
-       struct p9_fcall *tcall;
-       struct p9_fcall *rcall;
-       int err;
-       p9_conn_req_callback cb;
-       void *cba;
-       int flush;
-       struct list_head req_list;
-};
-
-struct p9_conn {
-       spinlock_t lock; /* protect lock structure */
-       struct list_head mux_list;
-       struct p9_mux_poll_task *poll_task;
-       int msize;
-       unsigned char *extended;
-       struct p9_trans *trans;
-       struct p9_idpool *tagpool;
-       int err;
-       wait_queue_head_t equeue;
-       struct list_head req_list;
-       struct list_head unsent_req_list;
-       struct p9_fcall *rcall;
-       int rpos;
-       char *rbuf;
-       int wpos;
-       int wsize;
-       char *wbuf;
-       wait_queue_t poll_wait[MAXPOLLWADDR];
-       wait_queue_head_t *poll_waddr[MAXPOLLWADDR];
-       poll_table pt;
-       struct work_struct rq;
-       struct work_struct wq;
-       unsigned long wsched;
-};
-
-struct p9_mux_poll_task {
-       struct task_struct *task;
-       struct list_head mux_list;
-       int muxnum;
-};
-
-struct p9_mux_rpc {
-       struct p9_conn *m;
-       int err;
-       struct p9_fcall *tcall;
-       struct p9_fcall *rcall;
-       wait_queue_head_t wqueue;
-};
-
-static int p9_poll_proc(void *);
-static void p9_read_work(struct work_struct *work);
-static void p9_write_work(struct work_struct *work);
-static void p9_pollwait(struct file *filp, wait_queue_head_t *wait_address,
-                         poll_table * p);
-static u16 p9_mux_get_tag(struct p9_conn *);
-static void p9_mux_put_tag(struct p9_conn *, u16);
-
-static DEFINE_MUTEX(p9_mux_task_lock);
-static struct workqueue_struct *p9_mux_wq;
-
-static int p9_mux_num;
-static int p9_mux_poll_task_num;
-static struct p9_mux_poll_task p9_mux_poll_tasks[100];
-
-int p9_mux_global_init(void)
-{
-       int i;
-
-       for (i = 0; i < ARRAY_SIZE(p9_mux_poll_tasks); i++)
-               p9_mux_poll_tasks[i].task = NULL;
-
-       p9_mux_wq = create_workqueue("v9fs");
-       if (!p9_mux_wq) {
-               printk(KERN_WARNING "v9fs: mux: creating workqueue failed\n");
-               return -ENOMEM;
-       }
-
-       return 0;
-}
-
-void p9_mux_global_exit(void)
-{
-       destroy_workqueue(p9_mux_wq);
-}
-
-/**
- * p9_mux_calc_poll_procs - calculates the number of polling procs
- * based on the number of mounted v9fs filesystems.
- *
- * The current implementation returns sqrt of the number of mounts.
- */
-static int p9_mux_calc_poll_procs(int muxnum)
-{
-       int n;
-
-       if (p9_mux_poll_task_num)
-               n = muxnum / p9_mux_poll_task_num +
-                   (muxnum % p9_mux_poll_task_num ? 1 : 0);
-       else
-               n = 1;
-
-       if (n > ARRAY_SIZE(p9_mux_poll_tasks))
-               n = ARRAY_SIZE(p9_mux_poll_tasks);
-
-       return n;
-}
-
-static int p9_mux_poll_start(struct p9_conn *m)
-{
-       int i, n;
-       struct p9_mux_poll_task *vpt, *vptlast;
-       struct task_struct *pproc;
-
-       P9_DPRINTK(P9_DEBUG_MUX, "mux %p muxnum %d procnum %d\n", m, p9_mux_num,
-               p9_mux_poll_task_num);
-       mutex_lock(&p9_mux_task_lock);
-
-       n = p9_mux_calc_poll_procs(p9_mux_num + 1);
-       if (n > p9_mux_poll_task_num) {
-               for (i = 0; i < ARRAY_SIZE(p9_mux_poll_tasks); i++) {
-                       if (p9_mux_poll_tasks[i].task == NULL) {
-                               vpt = &p9_mux_poll_tasks[i];
-                               P9_DPRINTK(P9_DEBUG_MUX, "create proc %p\n",
-                                                                       vpt);
-                               pproc = kthread_create(p9_poll_proc, vpt,
-                                                               "v9fs-poll");
-
-                               if (!IS_ERR(pproc)) {
-                                       vpt->task = pproc;
-                                       INIT_LIST_HEAD(&vpt->mux_list);
-                                       vpt->muxnum = 0;
-                                       p9_mux_poll_task_num++;
-                                       wake_up_process(vpt->task);
-                               }
-                               break;
-                       }
-               }
-
-               if (i >= ARRAY_SIZE(p9_mux_poll_tasks))
-                       P9_DPRINTK(P9_DEBUG_ERROR,
-                                       "warning: no free poll slots\n");
-       }
-
-       n = (p9_mux_num + 1) / p9_mux_poll_task_num +
-           ((p9_mux_num + 1) % p9_mux_poll_task_num ? 1 : 0);
-
-       vptlast = NULL;
-       for (i = 0; i < ARRAY_SIZE(p9_mux_poll_tasks); i++) {
-               vpt = &p9_mux_poll_tasks[i];
-               if (vpt->task != NULL) {
-                       vptlast = vpt;
-                       if (vpt->muxnum < n) {
-                               P9_DPRINTK(P9_DEBUG_MUX, "put in proc %d\n", i);
-                               list_add(&m->mux_list, &vpt->mux_list);
-                               vpt->muxnum++;
-                               m->poll_task = vpt;
-                               memset(&m->poll_waddr, 0,
-                                                       sizeof(m->poll_waddr));
-                               init_poll_funcptr(&m->pt, p9_pollwait);
-                               break;
-                       }
-               }
-       }
-
-       if (i >= ARRAY_SIZE(p9_mux_poll_tasks)) {
-               if (vptlast == NULL) {
-                       mutex_unlock(&p9_mux_task_lock);
-                       return -ENOMEM;
-               }
-
-               P9_DPRINTK(P9_DEBUG_MUX, "put in proc %d\n", i);
-               list_add(&m->mux_list, &vptlast->mux_list);
-               vptlast->muxnum++;
-               m->poll_task = vptlast;
-               memset(&m->poll_waddr, 0, sizeof(m->poll_waddr));
-               init_poll_funcptr(&m->pt, p9_pollwait);
-       }
-
-       p9_mux_num++;
-       mutex_unlock(&p9_mux_task_lock);
-
-       return 0;
-}
-
-static void p9_mux_poll_stop(struct p9_conn *m)
-{
-       int i;
-       struct p9_mux_poll_task *vpt;
-
-       mutex_lock(&p9_mux_task_lock);
-       vpt = m->poll_task;
-       list_del(&m->mux_list);
-       for (i = 0; i < ARRAY_SIZE(m->poll_waddr); i++) {
-               if (m->poll_waddr[i] != NULL) {
-                       remove_wait_queue(m->poll_waddr[i], &m->poll_wait[i]);
-                       m->poll_waddr[i] = NULL;
-               }
-       }
-       vpt->muxnum--;
-       if (!vpt->muxnum) {
-               P9_DPRINTK(P9_DEBUG_MUX, "destroy proc %p\n", vpt);
-               kthread_stop(vpt->task);
-               vpt->task = NULL;
-               p9_mux_poll_task_num--;
-       }
-       p9_mux_num--;
-       mutex_unlock(&p9_mux_task_lock);
-}
-
-/**
- * p9_conn_create - allocate and initialize the per-session mux data
- * Creates the polling task if this is the first session.
- *
- * @trans - transport structure
- * @msize - maximum message size
- * @extended - pointer to the extended flag
- */
-struct p9_conn *p9_conn_create(struct p9_trans *trans, int msize,
-                                   unsigned char *extended)
-{
-       int i, n;
-       struct p9_conn *m, *mtmp;
-
-       P9_DPRINTK(P9_DEBUG_MUX, "transport %p msize %d\n", trans, msize);
-       m = kmalloc(sizeof(struct p9_conn), GFP_KERNEL);
-       if (!m)
-               return ERR_PTR(-ENOMEM);
-
-       spin_lock_init(&m->lock);
-       INIT_LIST_HEAD(&m->mux_list);
-       m->msize = msize;
-       m->extended = extended;
-       m->trans = trans;
-       m->tagpool = p9_idpool_create();
-       if (IS_ERR(m->tagpool)) {
-               mtmp = ERR_PTR(-ENOMEM);
-               kfree(m);
-               return mtmp;
-       }
-
-       m->err = 0;
-       init_waitqueue_head(&m->equeue);
-       INIT_LIST_HEAD(&m->req_list);
-       INIT_LIST_HEAD(&m->unsent_req_list);
-       m->rcall = NULL;
-       m->rpos = 0;
-       m->rbuf = NULL;
-       m->wpos = m->wsize = 0;
-       m->wbuf = NULL;
-       INIT_WORK(&m->rq, p9_read_work);
-       INIT_WORK(&m->wq, p9_write_work);
-       m->wsched = 0;
-       memset(&m->poll_waddr, 0, sizeof(m->poll_waddr));
-       m->poll_task = NULL;
-       n = p9_mux_poll_start(m);
-       if (n) {
-               kfree(m);
-               return ERR_PTR(n);
-       }
-
-       n = trans->poll(trans, &m->pt);
-       if (n & POLLIN) {
-               P9_DPRINTK(P9_DEBUG_MUX, "mux %p can read\n", m);
-               set_bit(Rpending, &m->wsched);
-       }
-
-       if (n & POLLOUT) {
-               P9_DPRINTK(P9_DEBUG_MUX, "mux %p can write\n", m);
-               set_bit(Wpending, &m->wsched);
-       }
-
-       for (i = 0; i < ARRAY_SIZE(m->poll_waddr); i++) {
-               if (IS_ERR(m->poll_waddr[i])) {
-                       p9_mux_poll_stop(m);
-                       mtmp = (void *)m->poll_waddr;   /* the error code */
-                       kfree(m);
-                       m = mtmp;
-                       break;
-               }
-       }
-
-       return m;
-}
-EXPORT_SYMBOL(p9_conn_create);
-
-/**
- * p9_mux_destroy - cancels all pending requests and frees mux resources
- */
-void p9_conn_destroy(struct p9_conn *m)
-{
-       P9_DPRINTK(P9_DEBUG_MUX, "mux %p prev %p next %p\n", m,
-               m->mux_list.prev, m->mux_list.next);
-       p9_conn_cancel(m, -ECONNRESET);
-
-       if (!list_empty(&m->req_list)) {
-               /* wait until all processes waiting on this session exit */
-               P9_DPRINTK(P9_DEBUG_MUX,
-                       "mux %p waiting for empty request queue\n", m);
-               wait_event_timeout(m->equeue, (list_empty(&m->req_list)), 5000);
-               P9_DPRINTK(P9_DEBUG_MUX, "mux %p request queue empty: %d\n", m,
-                       list_empty(&m->req_list));
-       }
-
-       p9_mux_poll_stop(m);
-       m->trans = NULL;
-       p9_idpool_destroy(m->tagpool);
-       kfree(m);
-}
-EXPORT_SYMBOL(p9_conn_destroy);
-
-/**
- * p9_pollwait - called by files poll operation to add v9fs-poll task
- *     to files wait queue
- */
-static void
-p9_pollwait(struct file *filp, wait_queue_head_t *wait_address,
-             poll_table * p)
-{
-       int i;
-       struct p9_conn *m;
-
-       m = container_of(p, struct p9_conn, pt);
-       for (i = 0; i < ARRAY_SIZE(m->poll_waddr); i++)
-               if (m->poll_waddr[i] == NULL)
-                       break;
-
-       if (i >= ARRAY_SIZE(m->poll_waddr)) {
-               P9_DPRINTK(P9_DEBUG_ERROR, "not enough wait_address slots\n");
-               return;
-       }
-
-       m->poll_waddr[i] = wait_address;
-
-       if (!wait_address) {
-               P9_DPRINTK(P9_DEBUG_ERROR, "no wait_address\n");
-               m->poll_waddr[i] = ERR_PTR(-EIO);
-               return;
-       }
-
-       init_waitqueue_entry(&m->poll_wait[i], m->poll_task->task);
-       add_wait_queue(wait_address, &m->poll_wait[i]);
-}
-
-/**
- * p9_poll_mux - polls a mux and schedules read or write works if necessary
- */
-static void p9_poll_mux(struct p9_conn *m)
-{
-       int n;
-
-       if (m->err < 0)
-               return;
-
-       n = m->trans->poll(m->trans, NULL);
-       if (n < 0 || n & (POLLERR | POLLHUP | POLLNVAL)) {
-               P9_DPRINTK(P9_DEBUG_MUX, "error mux %p err %d\n", m, n);
-               if (n >= 0)
-                       n = -ECONNRESET;
-               p9_conn_cancel(m, n);
-       }
-
-       if (n & POLLIN) {
-               set_bit(Rpending, &m->wsched);
-               P9_DPRINTK(P9_DEBUG_MUX, "mux %p can read\n", m);
-               if (!test_and_set_bit(Rworksched, &m->wsched)) {
-                       P9_DPRINTK(P9_DEBUG_MUX, "schedule read work %p\n", m);
-                       queue_work(p9_mux_wq, &m->rq);
-               }
-       }
-
-       if (n & POLLOUT) {
-               set_bit(Wpending, &m->wsched);
-               P9_DPRINTK(P9_DEBUG_MUX, "mux %p can write\n", m);
-               if ((m->wsize || !list_empty(&m->unsent_req_list))
-                   && !test_and_set_bit(Wworksched, &m->wsched)) {
-                       P9_DPRINTK(P9_DEBUG_MUX, "schedule write work %p\n", m);
-                       queue_work(p9_mux_wq, &m->wq);
-               }
-       }
-}
-
-/**
- * p9_poll_proc - polls all v9fs transports for new events and queues
- *     the appropriate work to the work queue
- */
-static int p9_poll_proc(void *a)
-{
-       struct p9_conn *m, *mtmp;
-       struct p9_mux_poll_task *vpt;
-
-       vpt = a;
-       P9_DPRINTK(P9_DEBUG_MUX, "start %p %p\n", current, vpt);
-       while (!kthread_should_stop()) {
-               set_current_state(TASK_INTERRUPTIBLE);
-
-               list_for_each_entry_safe(m, mtmp, &vpt->mux_list, mux_list) {
-                       p9_poll_mux(m);
-               }
-
-               P9_DPRINTK(P9_DEBUG_MUX, "sleeping...\n");
-               schedule_timeout(SCHED_TIMEOUT * HZ);
-       }
-
-       __set_current_state(TASK_RUNNING);
-       P9_DPRINTK(P9_DEBUG_MUX, "finish\n");
-       return 0;
-}
-
-/**
- * p9_write_work - called when a transport can send some data
- */
-static void p9_write_work(struct work_struct *work)
-{
-       int n, err;
-       struct p9_conn *m;
-       struct p9_req *req;
-
-       m = container_of(work, struct p9_conn, wq);
-
-       if (m->err < 0) {
-               clear_bit(Wworksched, &m->wsched);
-               return;
-       }
-
-       if (!m->wsize) {
-               if (list_empty(&m->unsent_req_list)) {
-                       clear_bit(Wworksched, &m->wsched);
-                       return;
-               }
-
-               spin_lock(&m->lock);
-again:
-               req = list_entry(m->unsent_req_list.next, struct p9_req,
-                              req_list);
-               list_move_tail(&req->req_list, &m->req_list);
-               if (req->err == ERREQFLUSH)
-                       goto again;
-
-               m->wbuf = req->tcall->sdata;
-               m->wsize = req->tcall->size;
-               m->wpos = 0;
-               spin_unlock(&m->lock);
-       }
-
-       P9_DPRINTK(P9_DEBUG_MUX, "mux %p pos %d size %d\n", m, m->wpos,
-                                                               m->wsize);
-       clear_bit(Wpending, &m->wsched);
-       err = m->trans->write(m->trans, m->wbuf + m->wpos, m->wsize - m->wpos);
-       P9_DPRINTK(P9_DEBUG_MUX, "mux %p sent %d bytes\n", m, err);
-       if (err == -EAGAIN) {
-               clear_bit(Wworksched, &m->wsched);
-               return;
-       }
-
-       if (err < 0)
-               goto error;
-       else if (err == 0) {
-               err = -EREMOTEIO;
-               goto error;
-       }
-
-       m->wpos += err;
-       if (m->wpos == m->wsize)
-               m->wpos = m->wsize = 0;
-
-       if (m->wsize == 0 && !list_empty(&m->unsent_req_list)) {
-               if (test_and_clear_bit(Wpending, &m->wsched))
-                       n = POLLOUT;
-               else
-                       n = m->trans->poll(m->trans, NULL);
-
-               if (n & POLLOUT) {
-                       P9_DPRINTK(P9_DEBUG_MUX, "schedule write work %p\n", m);
-                       queue_work(p9_mux_wq, &m->wq);
-               } else
-                       clear_bit(Wworksched, &m->wsched);
-       } else
-               clear_bit(Wworksched, &m->wsched);
-
-       return;
-
-error:
-       p9_conn_cancel(m, err);
-       clear_bit(Wworksched, &m->wsched);
-}
-
-static void process_request(struct p9_conn *m, struct p9_req *req)
-{
-       int ecode;
-       struct p9_str *ename;
-
-       if (!req->err && req->rcall->id == P9_RERROR) {
-               ecode = req->rcall->params.rerror.errno;
-               ename = &req->rcall->params.rerror.error;
-
-               P9_DPRINTK(P9_DEBUG_MUX, "Rerror %.*s\n", ename->len,
-                                                               ename->str);
-
-               if (*m->extended)
-                       req->err = -ecode;
-
-               if (!req->err) {
-                       req->err = p9_errstr2errno(ename->str, ename->len);
-
-                       if (!req->err) {        /* string match failed */
-                               PRINT_FCALL_ERROR("unknown error", req->rcall);
-                       }
-
-                       if (!req->err)
-                               req->err = -ESERVERFAULT;
-               }
-       } else if (req->tcall && req->rcall->id != req->tcall->id + 1) {
-               P9_DPRINTK(P9_DEBUG_ERROR,
-                               "fcall mismatch: expected %d, got %d\n",
-                               req->tcall->id + 1, req->rcall->id);
-               if (!req->err)
-                       req->err = -EIO;
-       }
-}
-
-/**
- * p9_read_work - called when there is some data to be read from a transport
- */
-static void p9_read_work(struct work_struct *work)
-{
-       int n, err;
-       struct p9_conn *m;
-       struct p9_req *req, *rptr, *rreq;
-       struct p9_fcall *rcall;
-       char *rbuf;
-
-       m = container_of(work, struct p9_conn, rq);
-
-       if (m->err < 0)
-               return;
-
-       rcall = NULL;
-       P9_DPRINTK(P9_DEBUG_MUX, "start mux %p pos %d\n", m, m->rpos);
-
-       if (!m->rcall) {
-               m->rcall =
-                   kmalloc(sizeof(struct p9_fcall) + m->msize, GFP_KERNEL);
-               if (!m->rcall) {
-                       err = -ENOMEM;
-                       goto error;
-               }
-
-               m->rbuf = (char *)m->rcall + sizeof(struct p9_fcall);
-               m->rpos = 0;
-       }
-
-       clear_bit(Rpending, &m->wsched);
-       err = m->trans->read(m->trans, m->rbuf + m->rpos, m->msize - m->rpos);
-       P9_DPRINTK(P9_DEBUG_MUX, "mux %p got %d bytes\n", m, err);
-       if (err == -EAGAIN) {
-               clear_bit(Rworksched, &m->wsched);
-               return;
-       }
-
-       if (err <= 0)
-               goto error;
-
-       m->rpos += err;
-       while (m->rpos > 4) {
-               n = le32_to_cpu(*(__le32 *) m->rbuf);
-               if (n >= m->msize) {
-                       P9_DPRINTK(P9_DEBUG_ERROR,
-                               "requested packet size too big: %d\n", n);
-                       err = -EIO;
-                       goto error;
-               }
-
-               if (m->rpos < n)
-                       break;
-
-               err =
-                   p9_deserialize_fcall(m->rbuf, n, m->rcall, *m->extended);
-               if (err < 0) {
-                       goto error;
-               }
-
-#ifdef CONFIG_NET_9P_DEBUG
-               if ((p9_debug_level&P9_DEBUG_FCALL) == P9_DEBUG_FCALL) {
-                       char buf[150];
-
-                       p9_printfcall(buf, sizeof(buf), m->rcall,
-                               *m->extended);
-                       printk(KERN_NOTICE ">>> %p %s\n", m, buf);
-               }
-#endif
-
-               rcall = m->rcall;
-               rbuf = m->rbuf;
-               if (m->rpos > n) {
-                       m->rcall = kmalloc(sizeof(struct p9_fcall) + m->msize,
-                                          GFP_KERNEL);
-                       if (!m->rcall) {
-                               err = -ENOMEM;
-                               goto error;
-                       }
-
-                       m->rbuf = (char *)m->rcall + sizeof(struct p9_fcall);
-                       memmove(m->rbuf, rbuf + n, m->rpos - n);
-                       m->rpos -= n;
-               } else {
-                       m->rcall = NULL;
-                       m->rbuf = NULL;
-                       m->rpos = 0;
-               }
-
-               P9_DPRINTK(P9_DEBUG_MUX, "mux %p fcall id %d tag %d\n", m,
-                                                       rcall->id, rcall->tag);
-
-               req = NULL;
-               spin_lock(&m->lock);
-               list_for_each_entry_safe(rreq, rptr, &m->req_list, req_list) {
-                       if (rreq->tag == rcall->tag) {
-                               req = rreq;
-                               if (req->flush != Flushing)
-                                       list_del(&req->req_list);
-                               break;
-                       }
-               }
-               spin_unlock(&m->lock);
-
-               if (req) {
-                       req->rcall = rcall;
-                       process_request(m, req);
-
-                       if (req->flush != Flushing) {
-                               if (req->cb)
-                                       (*req->cb) (req, req->cba);
-                               else
-                                       kfree(req->rcall);
-
-                               wake_up(&m->equeue);
-                       }
-               } else {
-                       if (err >= 0 && rcall->id != P9_RFLUSH)
-                               P9_DPRINTK(P9_DEBUG_ERROR,
-                                 "unexpected response mux %p id %d tag %d\n",
-                                 m, rcall->id, rcall->tag);
-                       kfree(rcall);
-               }
-       }
-
-       if (!list_empty(&m->req_list)) {
-               if (test_and_clear_bit(Rpending, &m->wsched))
-                       n = POLLIN;
-               else
-                       n = m->trans->poll(m->trans, NULL);
-
-               if (n & POLLIN) {
-                       P9_DPRINTK(P9_DEBUG_MUX, "schedule read work %p\n", m);
-                       queue_work(p9_mux_wq, &m->rq);
-               } else
-                       clear_bit(Rworksched, &m->wsched);
-       } else
-               clear_bit(Rworksched, &m->wsched);
-
-       return;
-
-error:
-       p9_conn_cancel(m, err);
-       clear_bit(Rworksched, &m->wsched);
-}
-
-/**
- * p9_send_request - send 9P request
- * The function can sleep until the request is scheduled for sending.
- * The function can be interrupted. Return from the function is not
- * a guarantee that the request is sent successfully. Can return errors
- * that can be retrieved by PTR_ERR macros.
- *
- * @m: mux data
- * @tc: request to be sent
- * @cb: callback function to call when response is received
- * @cba: parameter to pass to the callback function
- */
-static struct p9_req *p9_send_request(struct p9_conn *m,
-                                         struct p9_fcall *tc,
-                                         p9_conn_req_callback cb, void *cba)
-{
-       int n;
-       struct p9_req *req;
-
-       P9_DPRINTK(P9_DEBUG_MUX, "mux %p task %p tcall %p id %d\n", m, current,
-               tc, tc->id);
-       if (m->err < 0)
-               return ERR_PTR(m->err);
-
-       req = kmalloc(sizeof(struct p9_req), GFP_KERNEL);
-       if (!req)
-               return ERR_PTR(-ENOMEM);
-
-       if (tc->id == P9_TVERSION)
-               n = P9_NOTAG;
-       else
-               n = p9_mux_get_tag(m);
-
-       if (n < 0)
-               return ERR_PTR(-ENOMEM);
-
-       p9_set_tag(tc, n);
-
-#ifdef CONFIG_NET_9P_DEBUG
-       if ((p9_debug_level&P9_DEBUG_FCALL) == P9_DEBUG_FCALL) {
-               char buf[150];
-
-               p9_printfcall(buf, sizeof(buf), tc, *m->extended);
-               printk(KERN_NOTICE "<<< %p %s\n", m, buf);
-       }
-#endif
-
-       spin_lock_init(&req->lock);
-       req->tag = n;
-       req->tcall = tc;
-       req->rcall = NULL;
-       req->err = 0;
-       req->cb = cb;
-       req->cba = cba;
-       req->flush = None;
-
-       spin_lock(&m->lock);
-       list_add_tail(&req->req_list, &m->unsent_req_list);
-       spin_unlock(&m->lock);
-
-       if (test_and_clear_bit(Wpending, &m->wsched))
-               n = POLLOUT;
-       else
-               n = m->trans->poll(m->trans, NULL);
-
-       if (n & POLLOUT && !test_and_set_bit(Wworksched, &m->wsched))
-               queue_work(p9_mux_wq, &m->wq);
-
-       return req;
-}
-
-static void p9_mux_free_request(struct p9_conn *m, struct p9_req *req)
-{
-       p9_mux_put_tag(m, req->tag);
-       kfree(req);
-}
-
-static void p9_mux_flush_cb(struct p9_req *freq, void *a)
-{
-       p9_conn_req_callback cb;
-       int tag;
-       struct p9_conn *m;
-       struct p9_req *req, *rreq, *rptr;
-
-       m = a;
-       P9_DPRINTK(P9_DEBUG_MUX, "mux %p tc %p rc %p err %d oldtag %d\n", m,
-               freq->tcall, freq->rcall, freq->err,
-               freq->tcall->params.tflush.oldtag);
-
-       spin_lock(&m->lock);
-       cb = NULL;
-       tag = freq->tcall->params.tflush.oldtag;
-       req = NULL;
-       list_for_each_entry_safe(rreq, rptr, &m->req_list, req_list) {
-               if (rreq->tag == tag) {
-                       req = rreq;
-                       list_del(&req->req_list);
-                       break;
-               }
-       }
-       spin_unlock(&m->lock);
-
-       if (req) {
-               spin_lock(&req->lock);
-               req->flush = Flushed;
-               spin_unlock(&req->lock);
-
-               if (req->cb)
-                       (*req->cb) (req, req->cba);
-               else
-                       kfree(req->rcall);
-
-               wake_up(&m->equeue);
-       }
-
-       kfree(freq->tcall);
-       kfree(freq->rcall);
-       p9_mux_free_request(m, freq);
-}
-
-static int
-p9_mux_flush_request(struct p9_conn *m, struct p9_req *req)
-{
-       struct p9_fcall *fc;
-       struct p9_req *rreq, *rptr;
-
-       P9_DPRINTK(P9_DEBUG_MUX, "mux %p req %p tag %d\n", m, req, req->tag);
-
-       /* if a response was received for a request, do nothing */
-       spin_lock(&req->lock);
-       if (req->rcall || req->err) {
-               spin_unlock(&req->lock);
-               P9_DPRINTK(P9_DEBUG_MUX,
-                       "mux %p req %p response already received\n", m, req);
-               return 0;
-       }
-
-       req->flush = Flushing;
-       spin_unlock(&req->lock);
-
-       spin_lock(&m->lock);
-       /* if the request is not sent yet, just remove it from the list */
-       list_for_each_entry_safe(rreq, rptr, &m->unsent_req_list, req_list) {
-               if (rreq->tag == req->tag) {
-                       P9_DPRINTK(P9_DEBUG_MUX,
-                          "mux %p req %p request is not sent yet\n", m, req);
-                       list_del(&rreq->req_list);
-                       req->flush = Flushed;
-                       spin_unlock(&m->lock);
-                       if (req->cb)
-                               (*req->cb) (req, req->cba);
-                       return 0;
-               }
-       }
-       spin_unlock(&m->lock);
-
-       clear_thread_flag(TIF_SIGPENDING);
-       fc = p9_create_tflush(req->tag);
-       p9_send_request(m, fc, p9_mux_flush_cb, m);
-       return 1;
-}
-
-static void
-p9_conn_rpc_cb(struct p9_req *req, void *a)
-{
-       struct p9_mux_rpc *r;
-
-       P9_DPRINTK(P9_DEBUG_MUX, "req %p r %p\n", req, a);
-       r = a;
-       r->rcall = req->rcall;
-       r->err = req->err;
-
-       if (req->flush != None && !req->err)
-               r->err = -ERESTARTSYS;
-
-       wake_up(&r->wqueue);
-}
-
-/**
- * p9_mux_rpc - sends 9P request and waits until a response is available.
- *     The function can be interrupted.
- * @m: mux data
- * @tc: request to be sent
- * @rc: pointer where a pointer to the response is stored
- */
-int
-p9_conn_rpc(struct p9_conn *m, struct p9_fcall *tc,
-            struct p9_fcall **rc)
-{
-       int err, sigpending;
-       unsigned long flags;
-       struct p9_req *req;
-       struct p9_mux_rpc r;
-
-       r.err = 0;
-       r.tcall = tc;
-       r.rcall = NULL;
-       r.m = m;
-       init_waitqueue_head(&r.wqueue);
-
-       if (rc)
-               *rc = NULL;
-
-       sigpending = 0;
-       if (signal_pending(current)) {
-               sigpending = 1;
-               clear_thread_flag(TIF_SIGPENDING);
-       }
-
-       req = p9_send_request(m, tc, p9_conn_rpc_cb, &r);
-       if (IS_ERR(req)) {
-               err = PTR_ERR(req);
-               P9_DPRINTK(P9_DEBUG_MUX, "error %d\n", err);
-               return err;
-       }
-
-       err = wait_event_interruptible(r.wqueue, r.rcall != NULL || r.err < 0);
-       if (r.err < 0)
-               err = r.err;
-
-       if (err == -ERESTARTSYS && m->trans->status == Connected
-                                                       && m->err == 0) {
-               if (p9_mux_flush_request(m, req)) {
-                       /* wait until we get response of the flush message */
-                       do {
-                               clear_thread_flag(TIF_SIGPENDING);
-                               err = wait_event_interruptible(r.wqueue,
-                                       r.rcall || r.err);
-                       } while (!r.rcall && !r.err && err == -ERESTARTSYS &&
-                               m->trans->status == Connected && !m->err);
-
-                       err = -ERESTARTSYS;
-               }
-               sigpending = 1;
-       }
-
-       if (sigpending) {
-               spin_lock_irqsave(&current->sighand->siglock, flags);
-               recalc_sigpending();
-               spin_unlock_irqrestore(&current->sighand->siglock, flags);
-       }
-
-       if (rc)
-               *rc = r.rcall;
-       else
-               kfree(r.rcall);
-
-       p9_mux_free_request(m, req);
-       if (err > 0)
-               err = -EIO;
-
-       return err;
-}
-EXPORT_SYMBOL(p9_conn_rpc);
-
-#ifdef P9_NONBLOCK
-/**
- * p9_conn_rpcnb - sends 9P request without waiting for response.
- * @m: mux data
- * @tc: request to be sent
- * @cb: callback function to be called when response arrives
- * @cba: value to pass to the callback function
- */
-int p9_conn_rpcnb(struct p9_conn *m, struct p9_fcall *tc,
-                  p9_conn_req_callback cb, void *a)
-{
-       int err;
-       struct p9_req *req;
-
-       req = p9_send_request(m, tc, cb, a);
-       if (IS_ERR(req)) {
-               err = PTR_ERR(req);
-               P9_DPRINTK(P9_DEBUG_MUX, "error %d\n", err);
-               return PTR_ERR(req);
-       }
-
-       P9_DPRINTK(P9_DEBUG_MUX, "mux %p tc %p tag %d\n", m, tc, req->tag);
-       return 0;
-}
-EXPORT_SYMBOL(p9_conn_rpcnb);
-#endif /* P9_NONBLOCK */
-
-/**
- * p9_conn_cancel - cancel all pending requests with error
- * @m: mux data
- * @err: error code
- */
-void p9_conn_cancel(struct p9_conn *m, int err)
-{
-       struct p9_req *req, *rtmp;
-       LIST_HEAD(cancel_list);
-
-       P9_DPRINTK(P9_DEBUG_ERROR, "mux %p err %d\n", m, err);
-       m->err = err;
-       spin_lock(&m->lock);
-       list_for_each_entry_safe(req, rtmp, &m->req_list, req_list) {
-               list_move(&req->req_list, &cancel_list);
-       }
-       list_for_each_entry_safe(req, rtmp, &m->unsent_req_list, req_list) {
-               list_move(&req->req_list, &cancel_list);
-       }
-       spin_unlock(&m->lock);
-
-       list_for_each_entry_safe(req, rtmp, &cancel_list, req_list) {
-               list_del(&req->req_list);
-               if (!req->err)
-                       req->err = err;
-
-               if (req->cb)
-                       (*req->cb) (req, req->cba);
-               else
-                       kfree(req->rcall);
-       }
-
-       wake_up(&m->equeue);
-}
-EXPORT_SYMBOL(p9_conn_cancel);
-
-static u16 p9_mux_get_tag(struct p9_conn *m)
-{
-       int tag;
-
-       tag = p9_idpool_get(m->tagpool);
-       if (tag < 0)
-               return P9_NOTAG;
-       else
-               return (u16) tag;
-}
-
-static void p9_mux_put_tag(struct p9_conn *m, u16 tag)
-{
-       if (tag != P9_NOTAG && p9_idpool_check(tag, m->tagpool))
-               p9_idpool_put(tag, m->tagpool);
-}
index 62332ed..1aa9d51 100644 (file)
@@ -5,7 +5,7 @@
  *
  *  Copyright (C) 2006 by Russ Cox <rsc@swtch.com>
  *  Copyright (C) 2004-2005 by Latchesar Ionkov <lucho@ionkov.net>
- *  Copyright (C) 2004-2007 by Eric Van Hensbergen <ericvh@gmail.com>
+ *  Copyright (C) 2004-2008 by Eric Van Hensbergen <ericvh@gmail.com>
  *  Copyright (C) 1997-2002 by Ron Minnich <rminnich@sarnoff.com>
  *
  *  This program is free software; you can redistribute it and/or modify
@@ -29,6 +29,7 @@
 #include <linux/module.h>
 #include <linux/net.h>
 #include <linux/ipv6.h>
+#include <linux/kthread.h>
 #include <linux/errno.h>
 #include <linux/kernel.h>
 #include <linux/un.h>
@@ -42,7 +43,9 @@
 
 #define P9_PORT 564
 #define MAX_SOCK_BUF (64*1024)
-
+#define ERREQFLUSH     1
+#define SCHED_TIMEOUT  10
+#define MAXPOLLWADDR   2
 
 struct p9_fd_opts {
        int rfd;
@@ -53,6 +56,7 @@ struct p9_fd_opts {
 struct p9_trans_fd {
        struct file *rd;
        struct file *wr;
+       struct p9_conn *conn;
 };
 
 /*
@@ -72,6 +76,1028 @@ static match_table_t tokens = {
        {Opt_err, NULL},
 };
 
+enum {
+       Rworksched = 1,         /* read work scheduled or running */
+       Rpending = 2,           /* can read */
+       Wworksched = 4,         /* write work scheduled or running */
+       Wpending = 8,           /* can write */
+};
+
+enum {
+       None,
+       Flushing,
+       Flushed,
+};
+
+struct p9_req;
+
+typedef void (*p9_conn_req_callback)(struct p9_req *req, void *a);
+struct p9_req {
+       spinlock_t lock; /* protect request structure */
+       int tag;
+       struct p9_fcall *tcall;
+       struct p9_fcall *rcall;
+       int err;
+       p9_conn_req_callback cb;
+       void *cba;
+       int flush;
+       struct list_head req_list;
+};
+
+struct p9_mux_poll_task;
+
+struct p9_conn {
+       spinlock_t lock; /* protect lock structure */
+       struct list_head mux_list;
+       struct p9_mux_poll_task *poll_task;
+       int msize;
+       unsigned char extended;
+       struct p9_trans *trans;
+       struct p9_idpool *tagpool;
+       int err;
+       wait_queue_head_t equeue;
+       struct list_head req_list;
+       struct list_head unsent_req_list;
+       struct p9_fcall *rcall;
+       int rpos;
+       char *rbuf;
+       int wpos;
+       int wsize;
+       char *wbuf;
+       wait_queue_t poll_wait[MAXPOLLWADDR];
+       wait_queue_head_t *poll_waddr[MAXPOLLWADDR];
+       poll_table pt;
+       struct work_struct rq;
+       struct work_struct wq;
+       unsigned long wsched;
+};
+
+struct p9_mux_poll_task {
+       struct task_struct *task;
+       struct list_head mux_list;
+       int muxnum;
+};
+
+struct p9_mux_rpc {
+       struct p9_conn *m;
+       int err;
+       struct p9_fcall *tcall;
+       struct p9_fcall *rcall;
+       wait_queue_head_t wqueue;
+};
+
+static int p9_poll_proc(void *);
+static void p9_read_work(struct work_struct *work);
+static void p9_write_work(struct work_struct *work);
+static void p9_pollwait(struct file *filp, wait_queue_head_t *wait_address,
+                                                               poll_table *p);
+static int p9_fd_write(struct p9_trans *trans, void *v, int len);
+static int p9_fd_read(struct p9_trans *trans, void *v, int len);
+
+static DEFINE_MUTEX(p9_mux_task_lock);
+static struct workqueue_struct *p9_mux_wq;
+
+static int p9_mux_num;
+static int p9_mux_poll_task_num;
+static struct p9_mux_poll_task p9_mux_poll_tasks[100];
+
+static void p9_conn_destroy(struct p9_conn *);
+static unsigned int p9_fd_poll(struct p9_trans *trans,
+                                               struct poll_table_struct *pt);
+
+#ifdef P9_NONBLOCK
+static int p9_conn_rpcnb(struct p9_conn *m, struct p9_fcall *tc,
+       p9_conn_req_callback cb, void *a);
+#endif /* P9_NONBLOCK */
+
+static void p9_conn_cancel(struct p9_conn *m, int err);
+
+static int p9_mux_global_init(void)
+{
+       int i;
+
+       for (i = 0; i < ARRAY_SIZE(p9_mux_poll_tasks); i++)
+               p9_mux_poll_tasks[i].task = NULL;
+
+       p9_mux_wq = create_workqueue("v9fs");
+       if (!p9_mux_wq) {
+               printk(KERN_WARNING "v9fs: mux: creating workqueue failed\n");
+               return -ENOMEM;
+       }
+
+       return 0;
+}
+
+static u16 p9_mux_get_tag(struct p9_conn *m)
+{
+       int tag;
+
+       tag = p9_idpool_get(m->tagpool);
+       if (tag < 0)
+               return P9_NOTAG;
+       else
+               return (u16) tag;
+}
+
+static void p9_mux_put_tag(struct p9_conn *m, u16 tag)
+{
+       if (tag != P9_NOTAG && p9_idpool_check(tag, m->tagpool))
+               p9_idpool_put(tag, m->tagpool);
+}
+
+/**
+ * p9_mux_calc_poll_procs - calculates the number of polling procs
+ * based on the number of mounted v9fs filesystems.
+ *
+ * The current implementation returns sqrt of the number of mounts.
+ */
+static int p9_mux_calc_poll_procs(int muxnum)
+{
+       int n;
+
+       if (p9_mux_poll_task_num)
+               n = muxnum / p9_mux_poll_task_num +
+                   (muxnum % p9_mux_poll_task_num ? 1 : 0);
+       else
+               n = 1;
+
+       if (n > ARRAY_SIZE(p9_mux_poll_tasks))
+               n = ARRAY_SIZE(p9_mux_poll_tasks);
+
+       return n;
+}
+
+static int p9_mux_poll_start(struct p9_conn *m)
+{
+       int i, n;
+       struct p9_mux_poll_task *vpt, *vptlast;
+       struct task_struct *pproc;
+
+       P9_DPRINTK(P9_DEBUG_MUX, "mux %p muxnum %d procnum %d\n", m, p9_mux_num,
+               p9_mux_poll_task_num);
+       mutex_lock(&p9_mux_task_lock);
+
+       n = p9_mux_calc_poll_procs(p9_mux_num + 1);
+       if (n > p9_mux_poll_task_num) {
+               for (i = 0; i < ARRAY_SIZE(p9_mux_poll_tasks); i++) {
+                       if (p9_mux_poll_tasks[i].task == NULL) {
+                               vpt = &p9_mux_poll_tasks[i];
+                               P9_DPRINTK(P9_DEBUG_MUX, "create proc %p\n",
+                                                                       vpt);
+                               pproc = kthread_create(p9_poll_proc, vpt,
+                                                               "v9fs-poll");
+
+                               if (!IS_ERR(pproc)) {
+                                       vpt->task = pproc;
+                                       INIT_LIST_HEAD(&vpt->mux_list);
+                                       vpt->muxnum = 0;
+                                       p9_mux_poll_task_num++;
+                                       wake_up_process(vpt->task);
+                               }
+                               break;
+                       }
+               }
+
+               if (i >= ARRAY_SIZE(p9_mux_poll_tasks))
+                       P9_DPRINTK(P9_DEBUG_ERROR,
+                                       "warning: no free poll slots\n");
+       }
+
+       n = (p9_mux_num + 1) / p9_mux_poll_task_num +
+           ((p9_mux_num + 1) % p9_mux_poll_task_num ? 1 : 0);
+
+       vptlast = NULL;
+       for (i = 0; i < ARRAY_SIZE(p9_mux_poll_tasks); i++) {
+               vpt = &p9_mux_poll_tasks[i];
+               if (vpt->task != NULL) {
+                       vptlast = vpt;
+                       if (vpt->muxnum < n) {
+                               P9_DPRINTK(P9_DEBUG_MUX, "put in proc %d\n", i);
+                               list_add(&m->mux_list, &vpt->mux_list);
+                               vpt->muxnum++;
+                               m->poll_task = vpt;
+                               memset(&m->poll_waddr, 0,
+                                                       sizeof(m->poll_waddr));
+                               init_poll_funcptr(&m->pt, p9_pollwait);
+                               break;
+                       }
+               }
+       }
+
+       if (i >= ARRAY_SIZE(p9_mux_poll_tasks)) {
+               if (vptlast == NULL) {
+                       mutex_unlock(&p9_mux_task_lock);
+                       return -ENOMEM;
+               }
+
+               P9_DPRINTK(P9_DEBUG_MUX, "put in proc %d\n", i);
+               list_add(&m->mux_list, &vptlast->mux_list);
+               vptlast->muxnum++;
+               m->poll_task = vptlast;
+               memset(&m->poll_waddr, 0, sizeof(m->poll_waddr));
+               init_poll_funcptr(&m->pt, p9_pollwait);
+       }
+
+       p9_mux_num++;
+       mutex_unlock(&p9_mux_task_lock);
+
+       return 0;
+}
+
+static void p9_mux_poll_stop(struct p9_conn *m)
+{
+       int i;
+       struct p9_mux_poll_task *vpt;
+
+       mutex_lock(&p9_mux_task_lock);
+       vpt = m->poll_task;
+       list_del(&m->mux_list);
+       for (i = 0; i < ARRAY_SIZE(m->poll_waddr); i++) {
+               if (m->poll_waddr[i] != NULL) {
+                       remove_wait_queue(m->poll_waddr[i], &m->poll_wait[i]);
+                       m->poll_waddr[i] = NULL;
+               }
+       }
+       vpt->muxnum--;
+       if (!vpt->muxnum) {
+               P9_DPRINTK(P9_DEBUG_MUX, "destroy proc %p\n", vpt);
+               kthread_stop(vpt->task);
+               vpt->task = NULL;
+               p9_mux_poll_task_num--;
+       }
+       p9_mux_num--;
+       mutex_unlock(&p9_mux_task_lock);
+}
+
+/**
+ * p9_conn_create - allocate and initialize the per-session mux data
+ * Creates the polling task if this is the first session.
+ *
+ * @trans - transport structure
+ * @msize - maximum message size
+ * @extended - extended flag
+ */
+static struct p9_conn *p9_conn_create(struct p9_trans *trans)
+{
+       int i, n;
+       struct p9_conn *m, *mtmp;
+
+       P9_DPRINTK(P9_DEBUG_MUX, "transport %p msize %d\n", trans,
+                                                               trans->msize);
+       m = kmalloc(sizeof(struct p9_conn), GFP_KERNEL);
+       if (!m)
+               return ERR_PTR(-ENOMEM);
+
+       spin_lock_init(&m->lock);
+       INIT_LIST_HEAD(&m->mux_list);
+       m->msize = trans->msize;
+       m->extended = trans->extended;
+       m->trans = trans;
+       m->tagpool = p9_idpool_create();
+       if (IS_ERR(m->tagpool)) {
+               mtmp = ERR_PTR(-ENOMEM);
+               kfree(m);
+               return mtmp;
+       }
+
+       m->err = 0;
+       init_waitqueue_head(&m->equeue);
+       INIT_LIST_HEAD(&m->req_list);
+       INIT_LIST_HEAD(&m->unsent_req_list);
+       m->rcall = NULL;
+       m->rpos = 0;
+       m->rbuf = NULL;
+       m->wpos = m->wsize = 0;
+       m->wbuf = NULL;
+       INIT_WORK(&m->rq, p9_read_work);
+       INIT_WORK(&m->wq, p9_write_work);
+       m->wsched = 0;
+       memset(&m->poll_waddr, 0, sizeof(m->poll_waddr));
+       m->poll_task = NULL;
+       n = p9_mux_poll_start(m);
+       if (n) {
+               kfree(m);
+               return ERR_PTR(n);
+       }
+
+       n = p9_fd_poll(trans, &m->pt);
+       if (n & POLLIN) {
+               P9_DPRINTK(P9_DEBUG_MUX, "mux %p can read\n", m);
+               set_bit(Rpending, &m->wsched);
+       }
+
+       if (n & POLLOUT) {
+               P9_DPRINTK(P9_DEBUG_MUX, "mux %p can write\n", m);
+               set_bit(Wpending, &m->wsched);
+       }
+
+       for (i = 0; i < ARRAY_SIZE(m->poll_waddr); i++) {
+               if (IS_ERR(m->poll_waddr[i])) {
+                       p9_mux_poll_stop(m);
+                       mtmp = (void *)m->poll_waddr;   /* the error code */
+                       kfree(m);
+                       m = mtmp;
+                       break;
+               }
+       }
+
+       return m;
+}
+
+/**
+ * p9_mux_destroy - cancels all pending requests and frees mux resources
+ */
+static void p9_conn_destroy(struct p9_conn *m)
+{
+       P9_DPRINTK(P9_DEBUG_MUX, "mux %p prev %p next %p\n", m,
+               m->mux_list.prev, m->mux_list.next);
+       p9_conn_cancel(m, -ECONNRESET);
+
+       if (!list_empty(&m->req_list)) {
+               /* wait until all processes waiting on this session exit */
+               P9_DPRINTK(P9_DEBUG_MUX,
+                       "mux %p waiting for empty request queue\n", m);
+               wait_event_timeout(m->equeue, (list_empty(&m->req_list)), 5000);
+               P9_DPRINTK(P9_DEBUG_MUX, "mux %p request queue empty: %d\n", m,
+                       list_empty(&m->req_list));
+       }
+
+       p9_mux_poll_stop(m);
+       m->trans = NULL;
+       p9_idpool_destroy(m->tagpool);
+       kfree(m);
+}
+
+/**
+ * p9_pollwait - called by files poll operation to add v9fs-poll task
+ *     to files wait queue
+ */
+static void
+p9_pollwait(struct file *filp, wait_queue_head_t *wait_address, poll_table *p)
+{
+       int i;
+       struct p9_conn *m;
+
+       m = container_of(p, struct p9_conn, pt);
+       for (i = 0; i < ARRAY_SIZE(m->poll_waddr); i++)
+               if (m->poll_waddr[i] == NULL)
+                       break;
+
+       if (i >= ARRAY_SIZE(m->poll_waddr)) {
+               P9_DPRINTK(P9_DEBUG_ERROR, "not enough wait_address slots\n");
+               return;
+       }
+
+       m->poll_waddr[i] = wait_address;
+
+       if (!wait_address) {
+               P9_DPRINTK(P9_DEBUG_ERROR, "no wait_address\n");
+               m->poll_waddr[i] = ERR_PTR(-EIO);
+               return;
+       }
+
+       init_waitqueue_entry(&m->poll_wait[i], m->poll_task->task);
+       add_wait_queue(wait_address, &m->poll_wait[i]);
+}
+
+/**
+ * p9_poll_mux - polls a mux and schedules read or write works if necessary
+ */
+static void p9_poll_mux(struct p9_conn *m)
+{
+       int n;
+
+       if (m->err < 0)
+               return;
+
+       n = p9_fd_poll(m->trans, NULL);
+       if (n < 0 || n & (POLLERR | POLLHUP | POLLNVAL)) {
+               P9_DPRINTK(P9_DEBUG_MUX, "error mux %p err %d\n", m, n);
+               if (n >= 0)
+                       n = -ECONNRESET;
+               p9_conn_cancel(m, n);
+       }
+
+       if (n & POLLIN) {
+               set_bit(Rpending, &m->wsched);
+               P9_DPRINTK(P9_DEBUG_MUX, "mux %p can read\n", m);
+               if (!test_and_set_bit(Rworksched, &m->wsched)) {
+                       P9_DPRINTK(P9_DEBUG_MUX, "schedule read work %p\n", m);
+                       queue_work(p9_mux_wq, &m->rq);
+               }
+       }
+
+       if (n & POLLOUT) {
+               set_bit(Wpending, &m->wsched);
+               P9_DPRINTK(P9_DEBUG_MUX, "mux %p can write\n", m);
+               if ((m->wsize || !list_empty(&m->unsent_req_list))
+                   && !test_and_set_bit(Wworksched, &m->wsched)) {
+                       P9_DPRINTK(P9_DEBUG_MUX, "schedule write work %p\n", m);
+                       queue_work(p9_mux_wq, &m->wq);
+               }
+       }
+}
+
+/**
+ * p9_poll_proc - polls all v9fs transports for new events and queues
+ *     the appropriate work to the work queue
+ */
+static int p9_poll_proc(void *a)
+{
+       struct p9_conn *m, *mtmp;
+       struct p9_mux_poll_task *vpt;
+
+       vpt = a;
+       P9_DPRINTK(P9_DEBUG_MUX, "start %p %p\n", current, vpt);
+       while (!kthread_should_stop()) {
+               set_current_state(TASK_INTERRUPTIBLE);
+
+               list_for_each_entry_safe(m, mtmp, &vpt->mux_list, mux_list) {
+                       p9_poll_mux(m);
+               }
+
+               P9_DPRINTK(P9_DEBUG_MUX, "sleeping...\n");
+               schedule_timeout(SCHED_TIMEOUT * HZ);
+       }
+
+       __set_current_state(TASK_RUNNING);
+       P9_DPRINTK(P9_DEBUG_MUX, "finish\n");
+       return 0;
+}
+
+/**
+ * p9_write_work - called when a transport can send some data
+ */
+static void p9_write_work(struct work_struct *work)
+{
+       int n, err;
+       struct p9_conn *m;
+       struct p9_req *req;
+
+       m = container_of(work, struct p9_conn, wq);
+
+       if (m->err < 0) {
+               clear_bit(Wworksched, &m->wsched);
+               return;
+       }
+
+       if (!m->wsize) {
+               if (list_empty(&m->unsent_req_list)) {
+                       clear_bit(Wworksched, &m->wsched);
+                       return;
+               }
+
+               spin_lock(&m->lock);
+again:
+               req = list_entry(m->unsent_req_list.next, struct p9_req,
+                              req_list);
+               list_move_tail(&req->req_list, &m->req_list);
+               if (req->err == ERREQFLUSH)
+                       goto again;
+
+               m->wbuf = req->tcall->sdata;
+               m->wsize = req->tcall->size;
+               m->wpos = 0;
+               spin_unlock(&m->lock);
+       }
+
+       P9_DPRINTK(P9_DEBUG_MUX, "mux %p pos %d size %d\n", m, m->wpos,
+                                                               m->wsize);
+       clear_bit(Wpending, &m->wsched);
+       err = p9_fd_write(m->trans, m->wbuf + m->wpos, m->wsize - m->wpos);
+       P9_DPRINTK(P9_DEBUG_MUX, "mux %p sent %d bytes\n", m, err);
+       if (err == -EAGAIN) {
+               clear_bit(Wworksched, &m->wsched);
+               return;
+       }
+
+       if (err < 0)
+               goto error;
+       else if (err == 0) {
+               err = -EREMOTEIO;
+               goto error;
+       }
+
+       m->wpos += err;
+       if (m->wpos == m->wsize)
+               m->wpos = m->wsize = 0;
+
+       if (m->wsize == 0 && !list_empty(&m->unsent_req_list)) {
+               if (test_and_clear_bit(Wpending, &m->wsched))
+                       n = POLLOUT;
+               else
+                       n = p9_fd_poll(m->trans, NULL);
+
+               if (n & POLLOUT) {
+                       P9_DPRINTK(P9_DEBUG_MUX, "schedule write work %p\n", m);
+                       queue_work(p9_mux_wq, &m->wq);
+               } else
+                       clear_bit(Wworksched, &m->wsched);
+       } else
+               clear_bit(Wworksched, &m->wsched);
+
+       return;
+
+error:
+       p9_conn_cancel(m, err);
+       clear_bit(Wworksched, &m->wsched);
+}
+
+static void process_request(struct p9_conn *m, struct p9_req *req)
+{
+       int ecode;
+       struct p9_str *ename;
+
+       if (!req->err && req->rcall->id == P9_RERROR) {
+               ecode = req->rcall->params.rerror.errno;
+               ename = &req->rcall->params.rerror.error;
+
+               P9_DPRINTK(P9_DEBUG_MUX, "Rerror %.*s\n", ename->len,
+                                                               ename->str);
+
+               if (m->extended)
+                       req->err = -ecode;
+
+               if (!req->err) {
+                       req->err = p9_errstr2errno(ename->str, ename->len);
+
+                       /* string match failed */
+                       if (!req->err) {
+                               PRINT_FCALL_ERROR("unknown error", req->rcall);
+                               req->err = -ESERVERFAULT;
+                       }
+               }
+       } else if (req->tcall && req->rcall->id != req->tcall->id + 1) {
+               P9_DPRINTK(P9_DEBUG_ERROR,
+                               "fcall mismatch: expected %d, got %d\n",
+                               req->tcall->id + 1, req->rcall->id);
+               if (!req->err)
+                       req->err = -EIO;
+       }
+}
+
+/**
+ * p9_read_work - called when there is some data to be read from a transport
+ */
+static void p9_read_work(struct work_struct *work)
+{
+       int n, err;
+       struct p9_conn *m;
+       struct p9_req *req, *rptr, *rreq;
+       struct p9_fcall *rcall;
+       char *rbuf;
+
+       m = container_of(work, struct p9_conn, rq);
+
+       if (m->err < 0)
+               return;
+
+       rcall = NULL;
+       P9_DPRINTK(P9_DEBUG_MUX, "start mux %p pos %d\n", m, m->rpos);
+
+       if (!m->rcall) {
+               m->rcall =
+                   kmalloc(sizeof(struct p9_fcall) + m->msize, GFP_KERNEL);
+               if (!m->rcall) {
+                       err = -ENOMEM;
+                       goto error;
+               }
+
+               m->rbuf = (char *)m->rcall + sizeof(struct p9_fcall);
+               m->rpos = 0;
+       }
+
+       clear_bit(Rpending, &m->wsched);
+       err = p9_fd_read(m->trans, m->rbuf + m->rpos, m->msize - m->rpos);
+       P9_DPRINTK(P9_DEBUG_MUX, "mux %p got %d bytes\n", m, err);
+       if (err == -EAGAIN) {
+               clear_bit(Rworksched, &m->wsched);
+               return;
+       }
+
+       if (err <= 0)
+               goto error;
+
+       m->rpos += err;
+       while (m->rpos > 4) {
+               n = le32_to_cpu(*(__le32 *) m->rbuf);
+               if (n >= m->msize) {
+                       P9_DPRINTK(P9_DEBUG_ERROR,
+                               "requested packet size too big: %d\n", n);
+                       err = -EIO;
+                       goto error;
+               }
+
+               if (m->rpos < n)
+                       break;
+
+               err =
+                   p9_deserialize_fcall(m->rbuf, n, m->rcall, m->extended);
+               if (err < 0)
+                       goto error;
+
+#ifdef CONFIG_NET_9P_DEBUG
+               if ((p9_debug_level&P9_DEBUG_FCALL) == P9_DEBUG_FCALL) {
+                       char buf[150];
+
+                       p9_printfcall(buf, sizeof(buf), m->rcall,
+                               m->extended);
+                       printk(KERN_NOTICE ">>> %p %s\n", m, buf);
+               }
+#endif
+
+               rcall = m->rcall;
+               rbuf = m->rbuf;
+               if (m->rpos > n) {
+                       m->rcall = kmalloc(sizeof(struct p9_fcall) + m->msize,
+                                          GFP_KERNEL);
+                       if (!m->rcall) {
+                               err = -ENOMEM;
+                               goto error;
+                       }
+
+                       m->rbuf = (char *)m->rcall + sizeof(struct p9_fcall);
+                       memmove(m->rbuf, rbuf + n, m->rpos - n);
+                       m->rpos -= n;
+               } else {
+                       m->rcall = NULL;
+                       m->rbuf = NULL;
+                       m->rpos = 0;
+               }
+
+               P9_DPRINTK(P9_DEBUG_MUX, "mux %p fcall id %d tag %d\n", m,
+                                                       rcall->id, rcall->tag);
+
+               req = NULL;
+               spin_lock(&m->lock);
+               list_for_each_entry_safe(rreq, rptr, &m->req_list, req_list) {
+                       if (rreq->tag == rcall->tag) {
+                               req = rreq;
+                               if (req->flush != Flushing)
+                                       list_del(&req->req_list);
+                               break;
+                       }
+               }
+               spin_unlock(&m->lock);
+
+               if (req) {
+                       req->rcall = rcall;
+                       process_request(m, req);
+
+                       if (req->flush != Flushing) {
+                               if (req->cb)
+                                       (*req->cb) (req, req->cba);
+                               else
+                                       kfree(req->rcall);
+
+                               wake_up(&m->equeue);
+                       }
+               } else {
+                       if (err >= 0 && rcall->id != P9_RFLUSH)
+                               P9_DPRINTK(P9_DEBUG_ERROR,
+                                 "unexpected response mux %p id %d tag %d\n",
+                                 m, rcall->id, rcall->tag);
+                       kfree(rcall);
+               }
+       }
+
+       if (!list_empty(&m->req_list)) {
+               if (test_and_clear_bit(Rpending, &m->wsched))
+                       n = POLLIN;
+               else
+                       n = p9_fd_poll(m->trans, NULL);
+
+               if (n & POLLIN) {
+                       P9_DPRINTK(P9_DEBUG_MUX, "schedule read work %p\n", m);
+                       queue_work(p9_mux_wq, &m->rq);
+               } else
+                       clear_bit(Rworksched, &m->wsched);
+       } else
+               clear_bit(Rworksched, &m->wsched);
+
+       return;
+
+error:
+       p9_conn_cancel(m, err);
+       clear_bit(Rworksched, &m->wsched);
+}
+
+/**
+ * p9_send_request - send 9P request
+ * The function can sleep until the request is scheduled for sending.
+ * The function can be interrupted. Return from the function is not
+ * a guarantee that the request is sent successfully. Can return errors
+ * that can be retrieved by PTR_ERR macros.
+ *
+ * @m: mux data
+ * @tc: request to be sent
+ * @cb: callback function to call when response is received
+ * @cba: parameter to pass to the callback function
+ */
+static struct p9_req *p9_send_request(struct p9_conn *m,
+                                         struct p9_fcall *tc,
+                                         p9_conn_req_callback cb, void *cba)
+{
+       int n;
+       struct p9_req *req;
+
+       P9_DPRINTK(P9_DEBUG_MUX, "mux %p task %p tcall %p id %d\n", m, current,
+               tc, tc->id);
+       if (m->err < 0)
+               return ERR_PTR(m->err);
+
+       req = kmalloc(sizeof(struct p9_req), GFP_KERNEL);
+       if (!req)
+               return ERR_PTR(-ENOMEM);
+
+       if (tc->id == P9_TVERSION)
+               n = P9_NOTAG;
+       else
+               n = p9_mux_get_tag(m);
+
+       if (n < 0)
+               return ERR_PTR(-ENOMEM);
+
+       p9_set_tag(tc, n);
+
+#ifdef CONFIG_NET_9P_DEBUG
+       if ((p9_debug_level&P9_DEBUG_FCALL) == P9_DEBUG_FCALL) {
+               char buf[150];
+
+               p9_printfcall(buf, sizeof(buf), tc, m->extended);
+               printk(KERN_NOTICE "<<< %p %s\n", m, buf);
+       }
+#endif
+
+       spin_lock_init(&req->lock);
+       req->tag = n;
+       req->tcall = tc;
+       req->rcall = NULL;
+       req->err = 0;
+       req->cb = cb;
+       req->cba = cba;
+       req->flush = None;
+
+       spin_lock(&m->lock);
+       list_add_tail(&req->req_list, &m->unsent_req_list);
+       spin_unlock(&m->lock);
+
+       if (test_and_clear_bit(Wpending, &m->wsched))
+               n = POLLOUT;
+       else
+               n = p9_fd_poll(m->trans, NULL);
+
+       if (n & POLLOUT && !test_and_set_bit(Wworksched, &m->wsched))
+               queue_work(p9_mux_wq, &m->wq);
+
+       return req;
+}
+
+static void p9_mux_free_request(struct p9_conn *m, struct p9_req *req)
+{
+       p9_mux_put_tag(m, req->tag);
+       kfree(req);
+}
+
+static void p9_mux_flush_cb(struct p9_req *freq, void *a)
+{
+       p9_conn_req_callback cb;
+       int tag;
+       struct p9_conn *m;
+       struct p9_req *req, *rreq, *rptr;
+
+       m = a;
+       P9_DPRINTK(P9_DEBUG_MUX, "mux %p tc %p rc %p err %d oldtag %d\n", m,
+               freq->tcall, freq->rcall, freq->err,
+               freq->tcall->params.tflush.oldtag);
+
+       spin_lock(&m->lock);
+       cb = NULL;
+       tag = freq->tcall->params.tflush.oldtag;
+       req = NULL;
+       list_for_each_entry_safe(rreq, rptr, &m->req_list, req_list) {
+               if (rreq->tag == tag) {
+                       req = rreq;
+                       list_del(&req->req_list);
+                       break;
+               }
+       }
+       spin_unlock(&m->lock);
+
+       if (req) {
+               spin_lock(&req->lock);
+               req->flush = Flushed;
+               spin_unlock(&req->lock);
+
+               if (req->cb)
+                       (*req->cb) (req, req->cba);
+               else
+                       kfree(req->rcall);
+
+               wake_up(&m->equeue);
+       }
+
+       kfree(freq->tcall);
+       kfree(freq->rcall);
+       p9_mux_free_request(m, freq);
+}
+
+static int
+p9_mux_flush_request(struct p9_conn *m, struct p9_req *req)
+{
+       struct p9_fcall *fc;
+       struct p9_req *rreq, *rptr;
+
+       P9_DPRINTK(P9_DEBUG_MUX, "mux %p req %p tag %d\n", m, req, req->tag);
+
+       /* if a response was received for a request, do nothing */
+       spin_lock(&req->lock);
+       if (req->rcall || req->err) {
+               spin_unlock(&req->lock);
+               P9_DPRINTK(P9_DEBUG_MUX,
+                       "mux %p req %p response already received\n", m, req);
+               return 0;
+       }
+
+       req->flush = Flushing;
+       spin_unlock(&req->lock);
+
+       spin_lock(&m->lock);
+       /* if the request is not sent yet, just remove it from the list */
+       list_for_each_entry_safe(rreq, rptr, &m->unsent_req_list, req_list) {
+               if (rreq->tag == req->tag) {
+                       P9_DPRINTK(P9_DEBUG_MUX,
+                          "mux %p req %p request is not sent yet\n", m, req);
+                       list_del(&rreq->req_list);
+                       req->flush = Flushed;
+                       spin_unlock(&m->lock);
+                       if (req->cb)
+                               (*req->cb) (req, req->cba);
+                       return 0;
+               }
+       }
+       spin_unlock(&m->lock);
+
+       clear_thread_flag(TIF_SIGPENDING);
+       fc = p9_create_tflush(req->tag);
+       p9_send_request(m, fc, p9_mux_flush_cb, m);
+       return 1;
+}
+
+static void
+p9_conn_rpc_cb(struct p9_req *req, void *a)
+{
+       struct p9_mux_rpc *r;
+
+       P9_DPRINTK(P9_DEBUG_MUX, "req %p r %p\n", req, a);
+       r = a;
+       r->rcall = req->rcall;
+       r->err = req->err;
+
+       if (req->flush != None && !req->err)
+               r->err = -ERESTARTSYS;
+
+       wake_up(&r->wqueue);
+}
+
+/**
+ * p9_fd_rpc- sends 9P request and waits until a response is available.
+ *     The function can be interrupted.
+ * @m: mux data
+ * @tc: request to be sent
+ * @rc: pointer where a pointer to the response is stored
+ */
+int
+p9_fd_rpc(struct p9_trans *t, struct p9_fcall *tc, struct p9_fcall **rc)
+{
+       struct p9_trans_fd *p = t->priv;
+       struct p9_conn *m = p->conn;
+       int err, sigpending;
+       unsigned long flags;
+       struct p9_req *req;
+       struct p9_mux_rpc r;
+
+       r.err = 0;
+       r.tcall = tc;
+       r.rcall = NULL;
+       r.m = m;
+       init_waitqueue_head(&r.wqueue);
+
+       if (rc)
+               *rc = NULL;
+
+       sigpending = 0;
+       if (signal_pending(current)) {
+               sigpending = 1;
+               clear_thread_flag(TIF_SIGPENDING);
+       }
+
+       req = p9_send_request(m, tc, p9_conn_rpc_cb, &r);
+       if (IS_ERR(req)) {
+               err = PTR_ERR(req);
+               P9_DPRINTK(P9_DEBUG_MUX, "error %d\n", err);
+               return err;
+       }
+
+       err = wait_event_interruptible(r.wqueue, r.rcall != NULL || r.err < 0);
+       if (r.err < 0)
+               err = r.err;
+
+       if (err == -ERESTARTSYS && m->trans->status == Connected
+                                                       && m->err == 0) {
+               if (p9_mux_flush_request(m, req)) {
+                       /* wait until we get response of the flush message */
+                       do {
+                               clear_thread_flag(TIF_SIGPENDING);
+                               err = wait_event_interruptible(r.wqueue,
+                                       r.rcall || r.err);
+                       } while (!r.rcall && !r.err && err == -ERESTARTSYS &&
+                               m->trans->status == Connected && !m->err);
+
+                       err = -ERESTARTSYS;
+               }
+               sigpending = 1;
+       }
+
+       if (sigpending) {
+               spin_lock_irqsave(&current->sighand->siglock, flags);
+               recalc_sigpending();
+               spin_unlock_irqrestore(&current->sighand->siglock, flags);
+       }
+
+       if (rc)
+               *rc = r.rcall;
+       else
+               kfree(r.rcall);
+
+       p9_mux_free_request(m, req);
+       if (err > 0)
+               err = -EIO;
+
+       return err;
+}
+
+#ifdef P9_NONBLOCK
+/**
+ * p9_conn_rpcnb - sends 9P request without waiting for response.
+ * @m: mux data
+ * @tc: request to be sent
+ * @cb: callback function to be called when response arrives
+ * @cba: value to pass to the callback function
+ */
+int p9_conn_rpcnb(struct p9_conn *m, struct p9_fcall *tc,
+                  p9_conn_req_callback cb, void *a)
+{
+       int err;
+       struct p9_req *req;
+
+       req = p9_send_request(m, tc, cb, a);
+       if (IS_ERR(req)) {
+               err = PTR_ERR(req);
+               P9_DPRINTK(P9_DEBUG_MUX, "error %d\n", err);
+               return PTR_ERR(req);
+       }
+
+       P9_DPRINTK(P9_DEBUG_MUX, "mux %p tc %p tag %d\n", m, tc, req->tag);
+       return 0;
+}
+#endif /* P9_NONBLOCK */
+
+/**
+ * p9_conn_cancel - cancel all pending requests with error
+ * @m: mux data
+ * @err: error code
+ */
+void p9_conn_cancel(struct p9_conn *m, int err)
+{
+       struct p9_req *req, *rtmp;
+       LIST_HEAD(cancel_list);
+
+       P9_DPRINTK(P9_DEBUG_ERROR, "mux %p err %d\n", m, err);
+       m->err = err;
+       spin_lock(&m->lock);
+       list_for_each_entry_safe(req, rtmp, &m->req_list, req_list) {
+               list_move(&req->req_list, &cancel_list);
+       }
+       list_for_each_entry_safe(req, rtmp, &m->unsent_req_list, req_list) {
+               list_move(&req->req_list, &cancel_list);
+       }
+       spin_unlock(&m->lock);
+
+       list_for_each_entry_safe(req, rtmp, &cancel_list, req_list) {
+               list_del(&req->req_list);
+               if (!req->err)
+                       req->err = err;
+
+               if (req->cb)
+                       (*req->cb) (req, req->cba);
+               else
+                       kfree(req->rcall);
+       }
+
+       wake_up(&m->equeue);
+}
+
 /**
  * v9fs_parse_options - parse mount options into session structure
  * @options: options string passed from mount
@@ -268,7 +1294,7 @@ end:
 }
 
 /**
- * p9_sock_close - shutdown socket
+ * p9_fd_close - shutdown socket
  * @trans: private socket structure
  *
  */
@@ -284,6 +1310,8 @@ static void p9_fd_close(struct p9_trans *trans)
        if (!ts)
                return;
 
+       p9_conn_destroy(ts->conn);
+
        trans->status = Disconnected;
        if (ts->rd)
                fput(ts->rd);
@@ -292,13 +1320,15 @@ static void p9_fd_close(struct p9_trans *trans)
        kfree(ts);
 }
 
-static struct p9_trans *p9_trans_create_tcp(const char *addr, char *args)
+static struct p9_trans *
+p9_trans_create_tcp(const char *addr, char *args, int msize, unsigned char dotu)
 {
        int err;
        struct p9_trans *trans;
        struct socket *csocket;
        struct sockaddr_in sin_server;
        struct p9_fd_opts opts;
+       struct p9_trans_fd *p;
 
        parse_opts(args, &opts);
 
@@ -306,11 +1336,10 @@ static struct p9_trans *p9_trans_create_tcp(const char *addr, char *args)
        trans = kmalloc(sizeof(struct p9_trans), GFP_KERNEL);
        if (!trans)
                return ERR_PTR(-ENOMEM);
-
-       trans->write = p9_fd_write;
-       trans->read = p9_fd_read;
+       trans->msize = msize;
+       trans->extended = dotu;
+       trans->rpc = p9_fd_rpc;
        trans->close = p9_fd_close;
-       trans->poll = p9_fd_poll;
 
        sin_server.sin_family = AF_INET;
        sin_server.sin_addr.s_addr = in_aton(addr);
@@ -337,6 +1366,14 @@ static struct p9_trans *p9_trans_create_tcp(const char *addr, char *args)
        if (err < 0)
                goto error;
 
+       p = (struct p9_trans_fd *) trans->priv;
+       p->conn = p9_conn_create(trans);
+       if (IS_ERR(p->conn)) {
+               err = PTR_ERR(p->conn);
+               p->conn = NULL;
+               goto error;
+       }
+
        return trans;
 
 error:
@@ -347,22 +1384,23 @@ error:
        return ERR_PTR(err);
 }
 
-static struct p9_trans *p9_trans_create_unix(const char *addr, char *args)
+static struct p9_trans *
+p9_trans_create_unix(const char *addr, char *args, int msize,
+                                                       unsigned char dotu)
 {
        int err;
        struct socket *csocket;
        struct sockaddr_un sun_server;
        struct p9_trans *trans;
+       struct p9_trans_fd *p;
 
        csocket = NULL;
        trans = kmalloc(sizeof(struct p9_trans), GFP_KERNEL);
        if (!trans)
                return ERR_PTR(-ENOMEM);
 
-       trans->write = p9_fd_write;
-       trans->read = p9_fd_read;
+       trans->rpc = p9_fd_rpc;
        trans->close = p9_fd_close;
-       trans->poll = p9_fd_poll;
 
        if (strlen(addr) > UNIX_PATH_MAX) {
                P9_EPRINTK(KERN_ERR, "p9_trans_unix: address too long: %s\n",
@@ -387,6 +1425,16 @@ static struct p9_trans *p9_trans_create_unix(const char *addr, char *args)
        if (err < 0)
                goto error;
 
+       trans->msize = msize;
+       trans->extended = dotu;
+       p = (struct p9_trans_fd *) trans->priv;
+       p->conn = p9_conn_create(trans);
+       if (IS_ERR(p->conn)) {
+               err = PTR_ERR(p->conn);
+               p->conn = NULL;
+               goto error;
+       }
+
        return trans;
 
 error:
@@ -397,11 +1445,14 @@ error:
        return ERR_PTR(err);
 }
 
-static struct p9_trans *p9_trans_create_fd(const char *name, char *args)
+static struct p9_trans *
+p9_trans_create_fd(const char *name, char *args, int msize,
+                                                       unsigned char extended)
 {
        int err;
        struct p9_trans *trans;
        struct p9_fd_opts opts;
+       struct p9_trans_fd *p;
 
        parse_opts(args, &opts);
 
@@ -414,15 +1465,23 @@ static struct p9_trans *p9_trans_create_fd(const char *name, char *args)
        if (!trans)
                return ERR_PTR(-ENOMEM);
 
-       trans->write = p9_fd_write;
-       trans->read = p9_fd_read;
+       trans->rpc = p9_fd_rpc;
        trans->close = p9_fd_close;
-       trans->poll = p9_fd_poll;
 
        err = p9_fd_open(trans, opts.rfd, opts.wfd);
        if (err < 0)
                goto error;
 
+       trans->msize = msize;
+       trans->extended = extended;
+       p = (struct p9_trans_fd *) trans->priv;
+       p->conn = p9_conn_create(trans);
+       if (IS_ERR(p->conn)) {
+               err = PTR_ERR(p->conn);
+               p->conn = NULL;
+               goto error;
+       }
+
        return trans;
 
 error:
@@ -453,6 +1512,12 @@ static struct p9_trans_module p9_fd_trans = {
 
 static int __init p9_trans_fd_init(void)
 {
+       int ret = p9_mux_global_init();
+       if (ret) {
+               printk(KERN_WARNING "9p: starting mux failed\n");
+               return ret;
+       }
+
        v9fs_register_trans(&p9_tcp_trans);
        v9fs_register_trans(&p9_unix_trans);
        v9fs_register_trans(&p9_fd_trans);
@@ -460,13 +1525,7 @@ static int __init p9_trans_fd_init(void)
        return 1;
 }
 
-static void __exit p9_trans_fd_exit(void) {
-       printk(KERN_ERR "Removal of 9p transports not implemented\n");
-       BUG();
-}
-
 module_init(p9_trans_fd_init);
-module_exit(p9_trans_fd_exit);
 
 MODULE_AUTHOR("Latchesar Ionkov <lucho@ionkov.net>");
 MODULE_AUTHOR("Eric Van Hensbergen <ericvh@gmail.com>");
index c957080..0117b9f 100644 (file)
@@ -130,29 +130,6 @@ static unsigned int rest_of_page(void *data)
        return PAGE_SIZE - ((unsigned long)data % PAGE_SIZE);
 }
 
-static int p9_virtio_write(struct p9_trans *trans, void *buf, int count)
-{
-       /* Only use the rpc mechanism for now */
-       return count;
-}
-
-static int p9_virtio_read(struct p9_trans *trans, void *buf, int count)
-{
-       /* Only use the rpc mechanism for now */
-       return 0;
-}
-
-/* The poll function is used by 9p transports to determine if there
- * is there is activity available on a particular channel.  In our case
- * we use it to wait for a callback from the input routines.
- */
-static unsigned int
-p9_virtio_poll(struct p9_trans *trans, struct poll_table_struct *pt)
-{
-       /* Only use the rpc mechanism for now */
-       return 0;
-}
-
 static void p9_virtio_close(struct p9_trans *trans)
 {
        struct virtio_chan *chan = trans->priv;
@@ -214,8 +191,7 @@ pack_sg_list(struct scatterlist *sg, int start, int limit, char *data,
 }
 
 static int
-p9_virtio_rpc(struct p9_trans *t, struct p9_fcall *tc, struct p9_fcall **rc,
-               int msize, int dotu)
+p9_virtio_rpc(struct p9_trans *t, struct p9_fcall *tc, struct p9_fcall **rc)
 {
        int in, out;
        int n, err, size;
@@ -225,7 +201,7 @@ p9_virtio_rpc(struct p9_trans *t, struct p9_fcall *tc, struct p9_fcall **rc,
        unsigned long flags;
 
        if (*rc == NULL) {
-               *rc = kmalloc(sizeof(struct p9_fcall) + msize, GFP_KERNEL);
+               *rc = kmalloc(sizeof(struct p9_fcall) + t->msize, GFP_KERNEL);
                if (!*rc)
                        return -ENOMEM;
        }
@@ -248,7 +224,7 @@ p9_virtio_rpc(struct p9_trans *t, struct p9_fcall *tc, struct p9_fcall **rc,
        P9_DPRINTK(P9_DEBUG_TRANS, "9p debug: virtio rpc tag %d\n", n);
 
        out = pack_sg_list(chan->sg, 0, VIRTQUEUE_NUM, tc->sdata, tc->size);
-       in = pack_sg_list(chan->sg, out, VIRTQUEUE_NUM-out, rdata, msize);
+       in = pack_sg_list(chan->sg, out, VIRTQUEUE_NUM-out, rdata, t->msize);
 
        req->status = REQ_STATUS_SENT;
 
@@ -264,7 +240,7 @@ p9_virtio_rpc(struct p9_trans *t, struct p9_fcall *tc, struct p9_fcall **rc,
 
        size = le32_to_cpu(*(__le32 *) rdata);
 
-       err = p9_deserialize_fcall(rdata, size, *rc, dotu);
+       err = p9_deserialize_fcall(rdata, size, *rc, t->extended);
        if (err < 0) {
                P9_DPRINTK(P9_DEBUG_TRANS,
                        "9p debug: virtio rpc deserialize returned %d\n", err);
@@ -275,7 +251,7 @@ p9_virtio_rpc(struct p9_trans *t, struct p9_fcall *tc, struct p9_fcall **rc,
        if ((p9_debug_level&P9_DEBUG_FCALL) == P9_DEBUG_FCALL) {
                char buf[150];
 
-               p9_printfcall(buf, sizeof(buf), *rc, dotu);
+               p9_printfcall(buf, sizeof(buf), *rc, t->extended);
                printk(KERN_NOTICE ">>> %p %s\n", t, buf);
        }
 #endif
@@ -337,7 +313,9 @@ fail:
  * alternate channels by matching devname versus a virtio_config entry.
  * We use a simple reference count mechanism to ensure that only a single
  * mount has a channel open at a time. */
-static struct p9_trans *p9_virtio_create(const char *devname, char *args)
+static struct p9_trans *
+p9_virtio_create(const char *devname, char *args, int msize,
+                                                       unsigned char extended)
 {
        struct p9_trans *trans;
        struct virtio_chan *chan = channels;
@@ -374,11 +352,9 @@ static struct p9_trans *p9_virtio_create(const char *devname, char *args)
                printk(KERN_ERR "9p: couldn't allocate transport\n");
                return ERR_PTR(-ENOMEM);
        }
-
-       trans->write = p9_virtio_write;
-       trans->read = p9_virtio_read;
+       trans->extended = extended;
+       trans->msize = msize;
        trans->close = p9_virtio_close;
-       trans->poll = p9_virtio_poll;
        trans->rpc = p9_virtio_rpc;
        trans->priv = chan;