Merge branch 'for-linus' of git://github.com/ericvh/linux

* 'for-linus' of git://github.com/ericvh/linux:
  9p: fix 9p.txt to advertise msize instead of maxdata
  net/9p: Convert net/9p protocol dumps to tracepoints
  fs/9p: change an int to unsigned int
  fs/9p: Cleanup option parsing in 9p
  9p: move dereference after NULL check
  fs/9p: inode file operation is properly initialized init_special_inode
  fs/9p: Update zero-copy implementation in 9p

6 files changed, 554 insertions, 411 deletions

diff --git a/net/9p/client.c b/net/9p/client.c
index 0505a03..854ca7a 100644
--- a/net/9p/client.c
+++ b/net/9p/client.c
@@ -38,6 +38,9 @@
 #include <net/9p/transport.h>
 #include "protocol.h"
 
+#define CREATE_TRACE_POINTS
+#include <trace/events/9p.h>
+
 /*
   * Client Option Parsing (code inspired by NFS code)
   *  - a little lazy - parse all client options
@@ -123,21 +126,19 @@ static int parse_opts(char *opts, struct p9_client *clnt)
 	options = tmp_options;
 
 	while ((p = strsep(&options, ",")) != NULL) {
-		int token;
+		int token, r;
 		if (!*p)
 			continue;
 		token = match_token(p, tokens, args);
-		if (token < Opt_trans) {
-			int r = match_int(&args[0], &option);
+		switch (token) {
+		case Opt_msize:
+			r = match_int(&args[0], &option);
 			if (r < 0) {
 				P9_DPRINTK(P9_DEBUG_ERROR,
-					"integer field, but no integer?\n");
+					   "integer field, but no integer?\n");
 				ret = r;
 				continue;
 			}
-		}
-		switch (token) {
-		case Opt_msize:
 			clnt->msize = option;
 			break;
 		case Opt_trans:
@@ -203,11 +204,13 @@ free_and_return:
  *
  */
 
-static struct p9_req_t *p9_tag_alloc(struct p9_client *c, u16 tag)
+static struct p9_req_t *
+p9_tag_alloc(struct p9_client *c, u16 tag, unsigned int max_size)
 {
 	unsigned long flags;
 	int row, col;
 	struct p9_req_t *req;
+	int alloc_msize = min(c->msize, max_size);
 
 	/* This looks up the original request by tag so we know which
 	 * buffer to read the data into */
@@ -245,23 +248,10 @@ static struct p9_req_t *p9_tag_alloc(struct p9_client *c, u16 tag)
 			return ERR_PTR(-ENOMEM);
 		}
 		init_waitqueue_head(req->wq);
-		if ((c->trans_mod->pref & P9_TRANS_PREF_PAYLOAD_MASK) ==
-				P9_TRANS_PREF_PAYLOAD_SEP) {
-			int alloc_msize = min(c->msize, 4096);
-			req->tc = kmalloc(sizeof(struct p9_fcall)+alloc_msize,
-					  GFP_NOFS);
-			req->tc->capacity = alloc_msize;
-			req->rc = kmalloc(sizeof(struct p9_fcall)+alloc_msize,
-					  GFP_NOFS);
-			req->rc->capacity = alloc_msize;
-		} else {
-			req->tc = kmalloc(sizeof(struct p9_fcall)+c->msize,
-					  GFP_NOFS);
-			req->tc->capacity = c->msize;
-			req->rc = kmalloc(sizeof(struct p9_fcall)+c->msize,
-					  GFP_NOFS);
-			req->rc->capacity = c->msize;
-		}
+		req->tc = kmalloc(sizeof(struct p9_fcall) + alloc_msize,
+				  GFP_NOFS);
+		req->rc = kmalloc(sizeof(struct p9_fcall) + alloc_msize,
+				  GFP_NOFS);
 		if ((!req->tc) || (!req->rc)) {
 			printk(KERN_ERR "Couldn't grow tag array\n");
 			kfree(req->tc);
@@ -271,6 +261,8 @@ static struct p9_req_t *p9_tag_alloc(struct p9_client *c, u16 tag)
 			req->wq = NULL;
 			return ERR_PTR(-ENOMEM);
 		}
+		req->tc->capacity = alloc_msize;
+		req->rc->capacity = alloc_msize;
 		req->tc->sdata = (char *) req->tc + sizeof(struct p9_fcall);
 		req->rc->sdata = (char *) req->rc + sizeof(struct p9_fcall);
 	}
@@ -475,37 +467,22 @@ static int p9_check_errors(struct p9_client *c, struct p9_req_t *req)
 	int ecode;
 
 	err = p9_parse_header(req->rc, NULL, &type, NULL, 0);
+	/*
+	 * dump the response from server
+	 * This should be after check errors which poplulate pdu_fcall.
+	 */
+	trace_9p_protocol_dump(c, req->rc);
 	if (err) {
 		P9_DPRINTK(P9_DEBUG_ERROR, "couldn't parse header %d\n", err);
 		return err;
 	}
-
 	if (type != P9_RERROR && type != P9_RLERROR)
 		return 0;
 
 	if (!p9_is_proto_dotl(c)) {
 		char *ename;
-
-		if (req->tc->pbuf_size) {
-			/* Handle user buffers */
-			size_t len = req->rc->size - req->rc->offset;
-			if (req->tc->pubuf) {
-				/* User Buffer */
-				err = copy_from_user(
-					&req->rc->sdata[req->rc->offset],
-					req->tc->pubuf, len);
-				if (err) {
-					err = -EFAULT;
-					goto out_err;
-				}
-			} else {
-				/* Kernel Buffer */
-				memmove(&req->rc->sdata[req->rc->offset],
-						req->tc->pkbuf, len);
-			}
-		}
 		err = p9pdu_readf(req->rc, c->proto_version, "s?d",
-				&ename, &ecode);
+				  &ename, &ecode);
 		if (err)
 			goto out_err;
 
@@ -515,11 +492,10 @@ static int p9_check_errors(struct p9_client *c, struct p9_req_t *req)
 		if (!err || !IS_ERR_VALUE(err)) {
 			err = p9_errstr2errno(ename, strlen(ename));
 
-			P9_DPRINTK(P9_DEBUG_9P, "<<< RERROR (%d) %s\n", -ecode,
-					ename);
-
-			kfree(ename);
+			P9_DPRINTK(P9_DEBUG_9P, "<<< RERROR (%d) %s\n",
+				   -ecode, ename);
 		}
+		kfree(ename);
 	} else {
 		err = p9pdu_readf(req->rc, c->proto_version, "d", &ecode);
 		err = -ecode;
@@ -527,7 +503,6 @@ static int p9_check_errors(struct p9_client *c, struct p9_req_t *req)
 		P9_DPRINTK(P9_DEBUG_9P, "<<< RLERROR (%d)\n", -ecode);
 	}
 
-
 	return err;
 
 out_err:
@@ -536,6 +511,115 @@ out_err:
 	return err;
 }
 
+/**
+ * p9_check_zc_errors - check 9p packet for error return and process it
+ * @c: current client instance
+ * @req: request to parse and check for error conditions
+ * @in_hdrlen: Size of response protocol buffer.
+ *
+ * returns error code if one is discovered, otherwise returns 0
+ *
+ * this will have to be more complicated if we have multiple
+ * error packet types
+ */
+
+static int p9_check_zc_errors(struct p9_client *c, struct p9_req_t *req,
+			      char *uidata, int in_hdrlen, int kern_buf)
+{
+	int err;
+	int ecode;
+	int8_t type;
+	char *ename = NULL;
+
+	err = p9_parse_header(req->rc, NULL, &type, NULL, 0);
+	/*
+	 * dump the response from server
+	 * This should be after parse_header which poplulate pdu_fcall.
+	 */
+	trace_9p_protocol_dump(c, req->rc);
+	if (err) {
+		P9_DPRINTK(P9_DEBUG_ERROR, "couldn't parse header %d\n", err);
+		return err;
+	}
+
+	if (type != P9_RERROR && type != P9_RLERROR)
+		return 0;
+
+	if (!p9_is_proto_dotl(c)) {
+		/* Error is reported in string format */
+		uint16_t len;
+		/* 7 = header size for RERROR, 2 is the size of string len; */
+		int inline_len = in_hdrlen - (7 + 2);
+
+		/* Read the size of error string */
+		err = p9pdu_readf(req->rc, c->proto_version, "w", &len);
+		if (err)
+			goto out_err;
+
+		ename = kmalloc(len + 1, GFP_NOFS);
+		if (!ename) {
+			err = -ENOMEM;
+			goto out_err;
+		}
+		if (len <= inline_len) {
+			/* We have error in protocol buffer itself */
+			if (pdu_read(req->rc, ename, len)) {
+				err = -EFAULT;
+				goto out_free;
+
+			}
+		} else {
+			/*
+			 *  Part of the data is in user space buffer.
+			 */
+			if (pdu_read(req->rc, ename, inline_len)) {
+				err = -EFAULT;
+				goto out_free;
+
+			}
+			if (kern_buf) {
+				memcpy(ename + inline_len, uidata,
+				       len - inline_len);
+			} else {
+				err = copy_from_user(ename + inline_len,
+						     uidata, len - inline_len);
+				if (err) {
+					err = -EFAULT;
+					goto out_free;
+				}
+			}
+		}
+		ename[len] = 0;
+		if (p9_is_proto_dotu(c)) {
+			/* For dotu we also have error code */
+			err = p9pdu_readf(req->rc,
+					  c->proto_version, "d", &ecode);
+			if (err)
+				goto out_free;
+			err = -ecode;
+		}
+		if (!err || !IS_ERR_VALUE(err)) {
+			err = p9_errstr2errno(ename, strlen(ename));
+
+			P9_DPRINTK(P9_DEBUG_9P, "<<< RERROR (%d) %s\n",
+				   -ecode, ename);
+		}
+		kfree(ename);
+	} else {
+		err = p9pdu_readf(req->rc, c->proto_version, "d", &ecode);
+		err = -ecode;
+
+		P9_DPRINTK(P9_DEBUG_9P, "<<< RLERROR (%d)\n", -ecode);
+	}
+	return err;
+
+out_free:
+	kfree(ename);
+out_err:
+	P9_DPRINTK(P9_DEBUG_ERROR, "couldn't parse error%d\n", err);
+	return err;
+}
+
 static struct p9_req_t *
 p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...);
 
@@ -579,23 +663,12 @@ static int p9_client_flush(struct p9_client *c, struct p9_req_t *oldreq)
 	return 0;
 }
 
-/**
- * p9_client_rpc - issue a request and wait for a response
- * @c: client session
- * @type: type of request
- * @fmt: protocol format string (see protocol.c)
- *
- * Returns request structure (which client must free using p9_free_req)
- */
-
-static struct p9_req_t *
-p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...)
+static struct p9_req_t *p9_client_prepare_req(struct p9_client *c,
+					      int8_t type, int req_size,
+					      const char *fmt, va_list ap)
 {
-	va_list ap;
 	int tag, err;
 	struct p9_req_t *req;
-	unsigned long flags;
-	int sigpending;
 
 	P9_DPRINTK(P9_DEBUG_MUX, "client %p op %d\n", c, type);
 
@@ -607,12 +680,6 @@ p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...)
 	if ((c->status == BeginDisconnect) && (type != P9_TCLUNK))
 		return ERR_PTR(-EIO);
 
-	if (signal_pending(current)) {
-		sigpending = 1;
-		clear_thread_flag(TIF_SIGPENDING);
-	} else
-		sigpending = 0;
-
 	tag = P9_NOTAG;
 	if (type != P9_TVERSION) {
 		tag = p9_idpool_get(c->tagpool);
@@ -620,18 +687,51 @@ p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...)
 			return ERR_PTR(-ENOMEM);
 	}
 
-	req = p9_tag_alloc(c, tag);
+	req = p9_tag_alloc(c, tag, req_size);
 	if (IS_ERR(req))
 		return req;
 
 	/* marshall the data */
 	p9pdu_prepare(req->tc, tag, type);
-	va_start(ap, fmt);
 	err = p9pdu_vwritef(req->tc, c->proto_version, fmt, ap);
-	va_end(ap);
 	if (err)
 		goto reterr;
-	p9pdu_finalize(req->tc);
+	p9pdu_finalize(c, req->tc);
+	trace_9p_client_req(c, type, tag);
+	return req;
+reterr:
+	p9_free_req(c, req);
+	return ERR_PTR(err);
+}
+
+/**
+ * p9_client_rpc - issue a request and wait for a response
+ * @c: client session
+ * @type: type of request
+ * @fmt: protocol format string (see protocol.c)
+ *
+ * Returns request structure (which client must free using p9_free_req)
+ */
+
+static struct p9_req_t *
+p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...)
+{
+	va_list ap;
+	int sigpending, err;
+	unsigned long flags;
+	struct p9_req_t *req;
+
+	va_start(ap, fmt);
+	req = p9_client_prepare_req(c, type, c->msize, fmt, ap);
+	va_end(ap);
+	if (IS_ERR(req))
+		return req;
+
+	if (signal_pending(current)) {
+		sigpending = 1;
+		clear_thread_flag(TIF_SIGPENDING);
+	} else
+		sigpending = 0;
 
 	err = c->trans_mod->request(c, req);
 	if (err < 0) {
@@ -639,18 +739,14 @@ p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...)
 			c->status = Disconnected;
 		goto reterr;
 	}
-
-	P9_DPRINTK(P9_DEBUG_MUX, "wait %p tag: %d\n", req->wq, tag);
+	/* Wait for the response */
 	err = wait_event_interruptible(*req->wq,
-						req->status >= REQ_STATUS_RCVD);
-	P9_DPRINTK(P9_DEBUG_MUX, "wait %p tag: %d returned %d\n",
-						req->wq, tag, err);
+				       req->status >= REQ_STATUS_RCVD);
 
 	if (req->status == REQ_STATUS_ERROR) {
 		P9_DPRINTK(P9_DEBUG_ERROR, "req_status error %d\n", req->t_err);
 		err = req->t_err;
 	}
-
 	if ((err == -ERESTARTSYS) && (c->status == Connected)) {
 		P9_DPRINTK(P9_DEBUG_MUX, "flushing\n");
 		sigpending = 1;
@@ -663,25 +759,102 @@ p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...)
 		if (req->status == REQ_STATUS_RCVD)
 			err = 0;
 	}
-
 	if (sigpending) {
 		spin_lock_irqsave(&current->sighand->siglock, flags);
 		recalc_sigpending();
 		spin_unlock_irqrestore(&current->sighand->siglock, flags);
 	}
-
 	if (err < 0)
 		goto reterr;
 
 	err = p9_check_errors(c, req);
-	if (!err) {
-		P9_DPRINTK(P9_DEBUG_MUX, "exit: client %p op %d\n", c, type);
+	trace_9p_client_res(c, type, req->rc->tag, err);
+	if (!err)
+		return req;
+reterr:
+	p9_free_req(c, req);
+	return ERR_PTR(err);
+}
+
+/**
+ * p9_client_zc_rpc - issue a request and wait for a response
+ * @c: client session
+ * @type: type of request
+ * @uidata: user bffer that should be ued for zero copy read
+ * @uodata: user buffer that shoud be user for zero copy write
+ * @inlen: read buffer size
+ * @olen: write buffer size
+ * @hdrlen: reader header size, This is the size of response protocol data
+ * @fmt: protocol format string (see protocol.c)
+ *
+ * Returns request structure (which client must free using p9_free_req)
+ */
+static struct p9_req_t *p9_client_zc_rpc(struct p9_client *c, int8_t type,
+					 char *uidata, char *uodata,
+					 int inlen, int olen, int in_hdrlen,
+					 int kern_buf, const char *fmt, ...)
+{
+	va_list ap;
+	int sigpending, err;
+	unsigned long flags;
+	struct p9_req_t *req;
+
+	va_start(ap, fmt);
+	/*
+	 * We allocate a inline protocol data of only 4k bytes.
+	 * The actual content is passed in zero-copy fashion.
+	 */
+	req = p9_client_prepare_req(c, type, P9_ZC_HDR_SZ, fmt, ap);
+	va_end(ap);
+	if (IS_ERR(req))
 		return req;
+
+	if (signal_pending(current)) {
+		sigpending = 1;
+		clear_thread_flag(TIF_SIGPENDING);
+	} else
+		sigpending = 0;
+
+	/* If we are called with KERNEL_DS force kern_buf */
+	if (segment_eq(get_fs(), KERNEL_DS))
+		kern_buf = 1;
+
+	err = c->trans_mod->zc_request(c, req, uidata, uodata,
+				       inlen, olen, in_hdrlen, kern_buf);
+	if (err < 0) {
+		if (err == -EIO)
+			c->status = Disconnected;
+		goto reterr;
+	}
+	if (req->status == REQ_STATUS_ERROR) {
+		P9_DPRINTK(P9_DEBUG_ERROR, "req_status error %d\n", req->t_err);
+		err = req->t_err;
+	}
+	if ((err == -ERESTARTSYS) && (c->status == Connected)) {
+		P9_DPRINTK(P9_DEBUG_MUX, "flushing\n");
+		sigpending = 1;
+		clear_thread_flag(TIF_SIGPENDING);
+
+		if (c->trans_mod->cancel(c, req))
+			p9_client_flush(c, req);
+
+		/* if we received the response anyway, don't signal error */
+		if (req->status == REQ_STATUS_RCVD)
+			err = 0;
+	}
+	if (sigpending) {
+		spin_lock_irqsave(&current->sighand->siglock, flags);
+		recalc_sigpending();
+		spin_unlock_irqrestore(&current->sighand->siglock, flags);
 	}
+	if (err < 0)
+		goto reterr;
 
+	err = p9_check_zc_errors(c, req, uidata, in_hdrlen, kern_buf);
+	trace_9p_client_res(c, type, req->rc->tag, err);
+	if (!err)
+		return req;
 reterr:
-	P9_DPRINTK(P9_DEBUG_MUX, "exit: client %p op %d error: %d\n", c, type,
-									err);
 	p9_free_req(c, req);
 	return ERR_PTR(err);
 }
@@ -769,7 +942,7 @@ static int p9_client_version(struct p9_client *c)
 	err = p9pdu_readf(req->rc, c->proto_version, "ds", &msize, &version);
 	if (err) {
 		P9_DPRINTK(P9_DEBUG_9P, "version error %d\n", err);
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(c, req->rc);
 		goto error;
 	}
 
@@ -906,15 +1079,14 @@ EXPORT_SYMBOL(p9_client_begin_disconnect);
 struct p9_fid *p9_client_attach(struct p9_client *clnt, struct p9_fid *afid,
 	char *uname, u32 n_uname, char *aname)
 {
-	int err;
+	int err = 0;
 	struct p9_req_t *req;
 	struct p9_fid *fid;
 	struct p9_qid qid;
 
-	P9_DPRINTK(P9_DEBUG_9P, ">>> TATTACH afid %d uname %s aname %s\n",
-					afid ? afid->fid : -1, uname, aname);
-	err = 0;
 
+	P9_DPRINTK(P9_DEBUG_9P, ">>> TATTACH afid %d uname %s aname %s\n",
+		   afid ? afid->fid : -1, uname, aname);
 	fid = p9_fid_create(clnt);
 	if (IS_ERR(fid)) {
 		err = PTR_ERR(fid);
@@ -931,7 +1103,7 @@ struct p9_fid *p9_client_attach(struct p9_client *clnt, struct p9_fid *afid,
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "Q", &qid);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		p9_free_req(clnt, req);
 		goto error;
 	}
@@ -991,7 +1163,7 @@ struct p9_fid *p9_client_walk(struct p9_fid *oldfid, uint16_t nwname,
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "R", &nwqids, &wqids);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		p9_free_req(clnt, req);
 		goto clunk_fid;
 	}
@@ -1058,7 +1230,7 @@ int p9_client_open(struct p9_fid *fid, int mode)
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "Qd", &qid, &iounit);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		goto free_and_error;
 	}
 
@@ -1101,7 +1273,7 @@ int p9_client_create_dotl(struct p9_fid *ofid, char *name, u32 flags, u32 mode,
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "Qd", qid, &iounit);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		goto free_and_error;
 	}
 
@@ -1146,7 +1318,7 @@ int p9_client_fcreate(struct p9_fid *fid, char *name, u32 perm, int mode,
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "Qd", &qid, &iounit);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		goto free_and_error;
 	}
 
@@ -1185,7 +1357,7 @@ int p9_client_symlink(struct p9_fid *dfid, char *name, char *symtgt, gid_t gid,
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "Q", qid);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		goto free_and_error;
 	}
 
@@ -1330,13 +1502,15 @@ int
 p9_client_read(struct p9_fid *fid, char *data, char __user *udata, u64 offset,
 								u32 count)
 {
-	int err, rsize;
-	struct p9_client *clnt;
-	struct p9_req_t *req;
 	char *dataptr;
+	int kernel_buf = 0;
+	struct p9_req_t *req;
+	struct p9_client *clnt;
+	int err, rsize, non_zc = 0;
+
 
-	P9_DPRINTK(P9_DEBUG_9P, ">>> TREAD fid %d offset %llu %d\n", fid->fid,
-					(long long unsigned) offset, count);
+	P9_DPRINTK(P9_DEBUG_9P, ">>> TREAD fid %d offset %llu %d\n",
+		   fid->fid, (long long unsigned) offset, count);
 	err = 0;
 	clnt = fid->clnt;
 
@@ -1348,13 +1522,24 @@ p9_client_read(struct p9_fid *fid, char *data, char __user *udata, u64 offset,
 		rsize = count;
 
 	/* Don't bother zerocopy for small IO (< 1024) */
-	if (((clnt->trans_mod->pref & P9_TRANS_PREF_PAYLOAD_MASK) ==
-			P9_TRANS_PREF_PAYLOAD_SEP) && (rsize > 1024)) {
-		req = p9_client_rpc(clnt, P9_TREAD, "dqE", fid->fid, offset,
-				rsize, data, udata);
+	if (clnt->trans_mod->zc_request && rsize > 1024) {
+		char *indata;
+		if (data) {
+			kernel_buf = 1;
+			indata = data;
+		} else
+			indata = (char *)udata;
+		/*
+		 * response header len is 11
+		 * PDU Header(7) + IO Size (4)
+		 */
+		req = p9_client_zc_rpc(clnt, P9_TREAD, indata, NULL, rsize, 0,
+				       11, kernel_buf, "dqd", fid->fid,
+				       offset, rsize);
 	} else {
+		non_zc = 1;
 		req = p9_client_rpc(clnt, P9_TREAD, "dqd", fid->fid, offset,
-				rsize);
+				    rsize);
 	}
 	if (IS_ERR(req)) {
 		err = PTR_ERR(req);
@@ -1363,14 +1548,13 @@ p9_client_read(struct p9_fid *fid, char *data, char __user *udata, u64 offset,
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "D", &count, &dataptr);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		goto free_and_error;
 	}
 
 	P9_DPRINTK(P9_DEBUG_9P, "<<< RREAD count %d\n", count);
-	P9_DUMP_PKT(1, req->rc);
 
-	if (!req->tc->pbuf_size) {
+	if (non_zc) {
 		if (data) {
 			memmove(data, dataptr, count);
 		} else {
@@ -1396,6 +1580,7 @@ p9_client_write(struct p9_fid *fid, char *data, const char __user *udata,
 							u64 offset, u32 count)
 {
 	int err, rsize;
+	int kernel_buf = 0;
 	struct p9_client *clnt;
 	struct p9_req_t *req;
 
@@ -1411,19 +1596,24 @@ p9_client_write(struct p9_fid *fid, char *data, const char __user *udata,
 	if (count < rsize)
 		rsize = count;
 
-	/* Don't bother zerocopy form small IO (< 1024) */
-	if (((clnt->trans_mod->pref & P9_TRANS_PREF_PAYLOAD_MASK) ==
-				P9_TRANS_PREF_PAYLOAD_SEP) && (rsize > 1024)) {
-		req = p9_client_rpc(clnt, P9_TWRITE, "dqE", fid->fid, offset,
-				rsize, data, udata);
+	/* Don't bother zerocopy for small IO (< 1024) */
+	if (clnt->trans_mod->zc_request && rsize > 1024) {
+		char *odata;
+		if (data) {
+			kernel_buf = 1;
+			odata = data;
+		} else
+			odata = (char *)udata;
+		req = p9_client_zc_rpc(clnt, P9_TWRITE, NULL, odata, 0, rsize,
+				       P9_ZC_HDR_SZ, kernel_buf, "dqd",
+				       fid->fid, offset, rsize);
 	} else {
-
 		if (data)
 			req = p9_client_rpc(clnt, P9_TWRITE, "dqD", fid->fid,
-					offset, rsize, data);
+					    offset, rsize, data);
 		else
 			req = p9_client_rpc(clnt, P9_TWRITE, "dqU", fid->fid,
-					offset, rsize, udata);
+					    offset, rsize, udata);
 	}
 	if (IS_ERR(req)) {
 		err = PTR_ERR(req);
@@ -1432,7 +1622,7 @@ p9_client_write(struct p9_fid *fid, char *data, const char __user *udata,
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "d", &count);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		goto free_and_error;
 	}
 
@@ -1472,7 +1662,7 @@ struct p9_wstat *p9_client_stat(struct p9_fid *fid)
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "wS", &ignored, ret);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		p9_free_req(clnt, req);
 		goto error;
 	}
@@ -1523,7 +1713,7 @@ struct p9_stat_dotl *p9_client_getattr_dotl(struct p9_fid *fid,
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "A", ret);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		p9_free_req(clnt, req);
 		goto error;
 	}
@@ -1671,7 +1861,7 @@ int p9_client_statfs(struct p9_fid *fid, struct p9_rstatfs *sb)
 		&sb->bsize, &sb->blocks, &sb->bfree, &sb->bavail,
 		&sb->files, &sb->ffree, &sb->fsid, &sb->namelen);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		p9_free_req(clnt, req);
 		goto error;
 	}
@@ -1778,7 +1968,7 @@ struct p9_fid *p9_client_xattrwalk(struct p9_fid *file_fid,
 	}
 	err = p9pdu_readf(req->rc, clnt->proto_version, "q", attr_size);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		p9_free_req(clnt, req);
 		goto clunk_fid;
 	}
@@ -1824,7 +2014,7 @@ EXPORT_SYMBOL_GPL(p9_client_xattrcreate);
 
 int p9_client_readdir(struct p9_fid *fid, char *data, u32 count, u64 offset)
 {
-	int err, rsize;
+	int err, rsize, non_zc = 0;
 	struct p9_client *clnt;
 	struct p9_req_t *req;
 	char *dataptr;
@@ -1842,13 +2032,18 @@ int p9_client_readdir(struct p9_fid *fid, char *data, u32 count, u64 offset)
 	if (count < rsize)
 		rsize = count;
 
-	if ((clnt->trans_mod->pref & P9_TRANS_PREF_PAYLOAD_MASK) ==
-			P9_TRANS_PREF_PAYLOAD_SEP) {
-		req = p9_client_rpc(clnt, P9_TREADDIR, "dqF", fid->fid,
-				offset, rsize, data);
+	/* Don't bother zerocopy for small IO (< 1024) */
+	if (clnt->trans_mod->zc_request && rsize > 1024) {
+		/*
+		 * response header len is 11
+		 * PDU Header(7) + IO Size (4)
+		 */
+		req = p9_client_zc_rpc(clnt, P9_TREADDIR, data, NULL, rsize, 0,
+				       11, 1, "dqd", fid->fid, offset, rsize);
 	} else {
+		non_zc = 1;
 		req = p9_client_rpc(clnt, P9_TREADDIR, "dqd", fid->fid,
-				offset, rsize);
+				    offset, rsize);
 	}
 	if (IS_ERR(req)) {
 		err = PTR_ERR(req);
@@ -1857,13 +2052,13 @@ int p9_client_readdir(struct p9_fid *fid, char *data, u32 count, u64 offset)
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "D", &count, &dataptr);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		goto free_and_error;
 	}
 
 	P9_DPRINTK(P9_DEBUG_9P, "<<< RREADDIR count %d\n", count);
 
-	if (!req->tc->pbuf_size && data)
+	if (non_zc)
 		memmove(data, dataptr, count);
 
 	p9_free_req(clnt, req);
@@ -1894,7 +2089,7 @@ int p9_client_mknod_dotl(struct p9_fid *fid, char *name, int mode,
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "Q", qid);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		goto error;
 	}
 	P9_DPRINTK(P9_DEBUG_9P, "<<< RMKNOD qid %x.%llx.%x\n", qid->type,
@@ -1925,7 +2120,7 @@ int p9_client_mkdir_dotl(struct p9_fid *fid, char *name, int mode,
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "Q", qid);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		goto error;
 	}
 	P9_DPRINTK(P9_DEBUG_9P, "<<< RMKDIR qid %x.%llx.%x\n", qid->type,
@@ -1960,7 +2155,7 @@ int p9_client_lock_dotl(struct p9_fid *fid, struct p9_flock *flock, u8 *status)
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "b", status);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		goto error;
 	}
 	P9_DPRINTK(P9_DEBUG_9P, "<<< RLOCK status %i\n", *status);
@@ -1993,7 +2188,7 @@ int p9_client_getlock_dotl(struct p9_fid *fid, struct p9_getlock *glock)
 			&glock->start, &glock->length, &glock->proc_id,
 			&glock->client_id);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		goto error;
 	}
 	P9_DPRINTK(P9_DEBUG_9P, "<<< RGETLOCK type %i start %lld length %lld "
@@ -2021,7 +2216,7 @@ int p9_client_readlink(struct p9_fid *fid, char **target)
 
 	err = p9pdu_readf(req->rc, clnt->proto_version, "s", target);
 	if (err) {
-		P9_DUMP_PKT(1, req->rc);
+		trace_9p_protocol_dump(clnt, req->rc);
 		goto error;
 	}
 	P9_DPRINTK(P9_DEBUG_9P, "<<< RREADLINK target %s\n", *target);
diff --git a/net/9p/protocol.c b/net/9p/protocol.c
index df58375..55e10a9 100644
--- a/net/9p/protocol.c
+++ b/net/9p/protocol.c
@@ -37,40 +37,11 @@
 #include <net/9p/client.h>
 #include "protocol.h"
 
+#include <trace/events/9p.h>
+
 static int
 p9pdu_writef(struct p9_fcall *pdu, int proto_version, const char *fmt, ...);
 
-#ifdef CONFIG_NET_9P_DEBUG
-void
-p9pdu_dump(int way, struct p9_fcall *pdu)
-{
-	int len = pdu->size;
-
-	if ((p9_debug_level & P9_DEBUG_VPKT) != P9_DEBUG_VPKT) {
-		if ((p9_debug_level & P9_DEBUG_PKT) == P9_DEBUG_PKT) {
-			if (len > 32)
-				len = 32;
-		} else {
-			/* shouldn't happen */
-			return;
-		}
-	}
-
-	if (way)
-		print_hex_dump_bytes("[9P] ", DUMP_PREFIX_OFFSET, pdu->sdata,
-									len);
-	else
-		print_hex_dump_bytes("]9P[ ", DUMP_PREFIX_OFFSET, pdu->sdata,
-									len);
-}
-#else
-void
-p9pdu_dump(int way, struct p9_fcall *pdu)
-{
-}
-#endif
-EXPORT_SYMBOL(p9pdu_dump);
-
 void p9stat_free(struct p9_wstat *stbuf)
 {
 	kfree(stbuf->name);
@@ -81,7 +52,7 @@ void p9stat_free(struct p9_wstat *stbuf)
 }
 EXPORT_SYMBOL(p9stat_free);
 
-static size_t pdu_read(struct p9_fcall *pdu, void *data, size_t size)
+size_t pdu_read(struct p9_fcall *pdu, void *data, size_t size)
 {
 	size_t len = min(pdu->size - pdu->offset, size);
 	memcpy(data, &pdu->sdata[pdu->offset], len);
@@ -108,26 +79,6 @@ pdu_write_u(struct p9_fcall *pdu, const char __user *udata, size_t size)
 	return size - len;
 }
 
-static size_t
-pdu_write_urw(struct p9_fcall *pdu, const char *kdata, const char __user *udata,
-		size_t size)
-{
-	BUG_ON(pdu->size > P9_IOHDRSZ);
-	pdu->pubuf = (char __user *)udata;
-	pdu->pkbuf = (char *)kdata;
-	pdu->pbuf_size = size;
-	return 0;
-}
-
-static size_t
-pdu_write_readdir(struct p9_fcall *pdu, const char *kdata, size_t size)
-{
-	BUG_ON(pdu->size > P9_READDIRHDRSZ);
-	pdu->pkbuf = (char *)kdata;
-	pdu->pbuf_size = size;
-	return 0;
-}
-
 /*
 	b - int8_t
 	w - int16_t
@@ -459,26 +410,6 @@ p9pdu_vwritef(struct p9_fcall *pdu, int proto_version, const char *fmt,
 					errcode = -EFAULT;
 			}
 			break;
-		case 'E':{
-				 int32_t cnt = va_arg(ap, int32_t);
-				 const char *k = va_arg(ap, const void *);
-				 const char __user *u = va_arg(ap,
-							const void __user *);
-				 errcode = p9pdu_writef(pdu, proto_version, "d",
-						 cnt);
-				 if (!errcode && pdu_write_urw(pdu, k, u, cnt))
-					errcode = -EFAULT;
-			 }
-			 break;
-		case 'F':{
-				 int32_t cnt = va_arg(ap, int32_t);
-				 const char *k = va_arg(ap, const void *);
-				 errcode = p9pdu_writef(pdu, proto_version, "d",
-						 cnt);
-				 if (!errcode && pdu_write_readdir(pdu, k, cnt))
-					errcode = -EFAULT;
-			 }
-			 break;
 		case 'U':{
 				int32_t count = va_arg(ap, int32_t);
 				const char __user *udata =
@@ -591,7 +522,7 @@ p9pdu_writef(struct p9_fcall *pdu, int proto_version, const char *fmt, ...)
 	return ret;
 }
 
-int p9stat_read(char *buf, int len, struct p9_wstat *st, int proto_version)
+int p9stat_read(struct p9_client *clnt, char *buf, int len, struct p9_wstat *st)
 {
 	struct p9_fcall fake_pdu;
 	int ret;
@@ -601,10 +532,10 @@ int p9stat_read(char *buf, int len, struct p9_wstat *st, int proto_version)
 	fake_pdu.sdata = buf;
 	fake_pdu.offset = 0;
 
-	ret = p9pdu_readf(&fake_pdu, proto_version, "S", st);
+	ret = p9pdu_readf(&fake_pdu, clnt->proto_version, "S", st);
 	if (ret) {
 		P9_DPRINTK(P9_DEBUG_9P, "<<< p9stat_read failed: %d\n", ret);
-		P9_DUMP_PKT(0, &fake_pdu);
+		trace_9p_protocol_dump(clnt, &fake_pdu);
 	}
 
 	return ret;
@@ -617,7 +548,7 @@ int p9pdu_prepare(struct p9_fcall *pdu, int16_t tag, int8_t type)
 	return p9pdu_writef(pdu, 0, "dbw", 0, type, tag);
 }
 
-int p9pdu_finalize(struct p9_fcall *pdu)
+int p9pdu_finalize(struct p9_client *clnt, struct p9_fcall *pdu)
 {
 	int size = pdu->size;
 	int err;
@@ -626,7 +557,7 @@ int p9pdu_finalize(struct p9_fcall *pdu)
 	err = p9pdu_writef(pdu, 0, "d", size);
 	pdu->size = size;
 
-	P9_DUMP_PKT(0, pdu);
+	trace_9p_protocol_dump(clnt, pdu);
 	P9_DPRINTK(P9_DEBUG_9P, ">>> size=%d type: %d tag: %d\n", pdu->size,
 							pdu->id, pdu->tag);
 
@@ -637,14 +568,10 @@ void p9pdu_reset(struct p9_fcall *pdu)
 {
 	pdu->offset = 0;
 	pdu->size = 0;
-	pdu->private = NULL;
-	pdu->pubuf = NULL;
-	pdu->pkbuf = NULL;
-	pdu->pbuf_size = 0;
 }
 
-int p9dirent_read(char *buf, int len, struct p9_dirent *dirent,
-						int proto_version)
+int p9dirent_read(struct p9_client *clnt, char *buf, int len,
+		  struct p9_dirent *dirent)
 {
 	struct p9_fcall fake_pdu;
 	int ret;
@@ -655,11 +582,11 @@ int p9dirent_read(char *buf, int len, struct p9_dirent *dirent,
 	fake_pdu.sdata = buf;
 	fake_pdu.offset = 0;
 
-	ret = p9pdu_readf(&fake_pdu, proto_version, "Qqbs", &dirent->qid,
-			&dirent->d_off, &dirent->d_type, &nameptr);
+	ret = p9pdu_readf(&fake_pdu, clnt->proto_version, "Qqbs", &dirent->qid,
+			  &dirent->d_off, &dirent->d_type, &nameptr);
 	if (ret) {
 		P9_DPRINTK(P9_DEBUG_9P, "<<< p9dirent_read failed: %d\n", ret);
-		P9_DUMP_PKT(1, &fake_pdu);
+		trace_9p_protocol_dump(clnt, &fake_pdu);
 		goto out;
 	}
 
diff --git a/net/9p/protocol.h b/net/9p/protocol.h
index 2431c0f..2cc525fa 100644
--- a/net/9p/protocol.h
+++ b/net/9p/protocol.h
@@ -29,6 +29,6 @@ int p9pdu_vwritef(struct p9_fcall *pdu, int proto_version, const char *fmt,
 								va_list ap);
 int p9pdu_readf(struct p9_fcall *pdu, int proto_version, const char *fmt, ...);
 int p9pdu_prepare(struct p9_fcall *pdu, int16_t tag, int8_t type);
-int p9pdu_finalize(struct p9_fcall *pdu);
-void p9pdu_dump(int, struct p9_fcall *);
+int p9pdu_finalize(struct p9_client *clnt, struct p9_fcall *pdu);
 void p9pdu_reset(struct p9_fcall *pdu);
+size_t pdu_read(struct p9_fcall *pdu, void *data, size_t size);
diff --git a/net/9p/trans_common.c b/net/9p/trans_common.c
index 9a70ebd..de8df95 100644
--- a/net/9p/trans_common.c
+++ b/net/9p/trans_common.c
@@ -21,30 +21,25 @@
 
 /**
  *  p9_release_req_pages - Release pages after the transaction.
- *  @*private: PDU's private page of struct trans_rpage_info
  */
-void
-p9_release_req_pages(struct trans_rpage_info *rpinfo)
+void p9_release_pages(struct page **pages, int nr_pages)
 {
 	int i = 0;
-
-	while (rpinfo->rp_data[i] && rpinfo->rp_nr_pages--) {
-		put_page(rpinfo->rp_data[i]);
+	while (pages[i] && nr_pages--) {
+		put_page(pages[i]);
 		i++;
 	}
 }
-EXPORT_SYMBOL(p9_release_req_pages);
+EXPORT_SYMBOL(p9_release_pages);
 
 /**
  * p9_nr_pages - Return number of pages needed to accommodate the payload.
  */
-int
-p9_nr_pages(struct p9_req_t *req)
+int p9_nr_pages(char *data, int len)
 {
 	unsigned long start_page, end_page;
-	start_page =  (unsigned long)req->tc->pubuf >> PAGE_SHIFT;
-	end_page = ((unsigned long)req->tc->pubuf + req->tc->pbuf_size +
-			PAGE_SIZE - 1) >> PAGE_SHIFT;
+	start_page =  (unsigned long)data >> PAGE_SHIFT;
+	end_page = ((unsigned long)data + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
 	return end_page - start_page;
 }
 EXPORT_SYMBOL(p9_nr_pages);
@@ -58,35 +53,17 @@ EXPORT_SYMBOL(p9_nr_pages);
  * @nr_pages: number of pages to accommodate the payload
  * @rw: Indicates if the pages are for read or write.
  */
-int
-p9_payload_gup(struct p9_req_t *req, size_t *pdata_off, int *pdata_len,
-		int nr_pages, u8 rw)
-{
-	uint32_t first_page_bytes = 0;
-	int32_t pdata_mapped_pages;
-	struct trans_rpage_info  *rpinfo;
-
-	*pdata_off = (__force size_t)req->tc->pubuf & (PAGE_SIZE-1);
 
-	if (*pdata_off)
-		first_page_bytes = min(((size_t)PAGE_SIZE - *pdata_off),
-				       req->tc->pbuf_size);
+int p9_payload_gup(char *data, int *nr_pages, struct page **pages, int write)
+{
+	int nr_mapped_pages;
 
-	rpinfo = req->tc->private;
-	pdata_mapped_pages = get_user_pages_fast((unsigned long)req->tc->pubuf,
-			nr_pages, rw, &rpinfo->rp_data[0]);
-	if (pdata_mapped_pages <= 0)
-		return pdata_mapped_pages;
+	nr_mapped_pages = get_user_pages_fast((unsigned long)data,
+					      *nr_pages, write, pages);
+	if (nr_mapped_pages <= 0)
+		return nr_mapped_pages;
 
-	rpinfo->rp_nr_pages = pdata_mapped_pages;
-	if (*pdata_off) {
-		*pdata_len = first_page_bytes;
-		*pdata_len += min((req->tc->pbuf_size - *pdata_len),
-				((size_t)pdata_mapped_pages - 1) << PAGE_SHIFT);
-	} else {
-		*pdata_len = min(req->tc->pbuf_size,
-				(size_t)pdata_mapped_pages << PAGE_SHIFT);
-	}
+	*nr_pages = nr_mapped_pages;
 	return 0;
 }
 EXPORT_SYMBOL(p9_payload_gup);
diff --git a/net/9p/trans_common.h b/net/9p/trans_common.h
index 7630922..173bb55 100644
--- a/net/9p/trans_common.h
+++ b/net/9p/trans_common.h
@@ -12,21 +12,6 @@
  *
  */
 
-/* TRUE if it is user context */
-#define P9_IS_USER_CONTEXT (!segment_eq(get_fs(), KERNEL_DS))
-
-/**
- * struct trans_rpage_info - To store mapped page information in PDU.
- * @rp_alloc:Set if this structure is allocd, not a reuse unused space in pdu.
- * @rp_nr_pages: Number of mapped pages
- * @rp_data: Array of page pointers
- */
-struct trans_rpage_info {
-	u8 rp_alloc;
-	int rp_nr_pages;
-	struct page *rp_data[0];
-};
-
-void p9_release_req_pages(struct trans_rpage_info *);
-int p9_payload_gup(struct p9_req_t *, size_t *, int *, int, u8);
-int p9_nr_pages(struct p9_req_t *);
+void p9_release_pages(struct page **, int);
+int p9_payload_gup(char *, int *, struct page **, int);
+int p9_nr_pages(char *, int);
diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c
index e317583..32aa983 100644
--- a/net/9p/trans_virtio.c
+++ b/net/9p/trans_virtio.c
@@ -150,12 +150,10 @@ static void req_done(struct virtqueue *vq)
 	while (1) {
 		spin_lock_irqsave(&chan->lock, flags);
 		rc = virtqueue_get_buf(chan->vq, &len);
-
 		if (rc == NULL) {
 			spin_unlock_irqrestore(&chan->lock, flags);
 			break;
 		}
-
 		chan->ring_bufs_avail = 1;
 		spin_unlock_irqrestore(&chan->lock, flags);
 		/* Wakeup if anyone waiting for VirtIO ring space. */
@@ -163,17 +161,6 @@ static void req_done(struct virtqueue *vq)
 		P9_DPRINTK(P9_DEBUG_TRANS, ": rc %p\n", rc);
 		P9_DPRINTK(P9_DEBUG_TRANS, ": lookup tag %d\n", rc->tag);
 		req = p9_tag_lookup(chan->client, rc->tag);
-		if (req->tc->private) {
-			struct trans_rpage_info *rp = req->tc->private;
-			int p = rp->rp_nr_pages;
-			/*Release pages */
-			p9_release_req_pages(rp);
-			atomic_sub(p, &vp_pinned);
-			wake_up(&vp_wq);
-			if (rp->rp_alloc)
-				kfree(rp);
-			req->tc->private = NULL;
-		}
 		req->status = REQ_STATUS_RCVD;
 		p9_client_cb(chan->client, req);
 	}
@@ -193,9 +180,8 @@ static void req_done(struct virtqueue *vq)
  *
  */
 
-static int
-pack_sg_list(struct scatterlist *sg, int start, int limit, char *data,
-								int count)
+static int pack_sg_list(struct scatterlist *sg, int start,
+			int limit, char *data, int count)
 {
 	int s;
 	int index = start;
@@ -224,31 +210,36 @@ static int p9_virtio_cancel(struct p9_client *client, struct p9_req_t *req)
  * this takes a list of pages.
  * @sg: scatter/gather list to pack into
  * @start: which segment of the sg_list to start at
- * @pdata_off: Offset into the first page
  * @**pdata: a list of pages to add into sg.
+ * @nr_pages: number of pages to pack into the scatter/gather list
+ * @data: data to pack into scatter/gather list
  * @count: amount of data to pack into the scatter/gather list
  */
 static int
-pack_sg_list_p(struct scatterlist *sg, int start, int limit, size_t pdata_off,
-		struct page **pdata, int count)
+pack_sg_list_p(struct scatterlist *sg, int start, int limit,
+	       struct page **pdata, int nr_pages, char *data, int count)
 {
-	int s;
-	int i = 0;
+	int i = 0, s;
+	int data_off;
 	int index = start;
 
-	if (pdata_off) {
-		s = min((int)(PAGE_SIZE - pdata_off), count);
-		sg_set_page(&sg[index++], pdata[i++], s, pdata_off);
-		count -= s;
-	}
-
-	while (count) {
-		BUG_ON(index > limit);
-		s = min((int)PAGE_SIZE, count);
-		sg_set_page(&sg[index++], pdata[i++], s, 0);
+	BUG_ON(nr_pages > (limit - start));
+	/*
+	 * if the first page doesn't start at
+	 * page boundary find the offset
+	 */
+	data_off = offset_in_page(data);
+	while (nr_pages) {
+		s = rest_of_page(data);
+		if (s > count)
+			s = count;
+		sg_set_page(&sg[index++], pdata[i++], s, data_off);
+		data_off = 0;
+		data += s;
 		count -= s;
+		nr_pages--;
 	}
-	return index-start;
+	return index - start;
 }
 
 /**
@@ -261,114 +252,166 @@ pack_sg_list_p(struct scatterlist *sg, int start, int limit, size_t pdata_off,
 static int
 p9_virtio_request(struct p9_client *client, struct p9_req_t *req)
 {
-	int in, out, inp, outp;
-	struct virtio_chan *chan = client->trans;
+	int err;
+	int in, out;
 	unsigned long flags;
-	size_t pdata_off = 0;
-	struct trans_rpage_info *rpinfo = NULL;
-	int err, pdata_len = 0;
+	struct virtio_chan *chan = client->trans;
 
 	P9_DPRINTK(P9_DEBUG_TRANS, "9p debug: virtio request\n");
 
 	req->status = REQ_STATUS_SENT;
+req_retry:
+	spin_lock_irqsave(&chan->lock, flags);
+
+	/* Handle out VirtIO ring buffers */
+	out = pack_sg_list(chan->sg, 0,
+			   VIRTQUEUE_NUM, req->tc->sdata, req->tc->size);
 
-	if (req->tc->pbuf_size && (req->tc->pubuf && P9_IS_USER_CONTEXT)) {
-		int nr_pages = p9_nr_pages(req);
-		int rpinfo_size = sizeof(struct trans_rpage_info) +
-			sizeof(struct page *) * nr_pages;
+	in = pack_sg_list(chan->sg, out,
+			  VIRTQUEUE_NUM, req->rc->sdata, req->rc->capacity);
 
-		if (atomic_read(&vp_pinned) >= chan->p9_max_pages) {
-			err = wait_event_interruptible(vp_wq,
-				atomic_read(&vp_pinned) < chan->p9_max_pages);
+	err = virtqueue_add_buf(chan->vq, chan->sg, out, in, req->tc);
+	if (err < 0) {
+		if (err == -ENOSPC) {
+			chan->ring_bufs_avail = 0;
+			spin_unlock_irqrestore(&chan->lock, flags);
+			err = wait_event_interruptible(*chan->vc_wq,
+							chan->ring_bufs_avail);
 			if (err  == -ERESTARTSYS)
 				return err;
-			P9_DPRINTK(P9_DEBUG_TRANS, "9p: May gup pages now.\n");
-		}
 
-		if (rpinfo_size <= (req->tc->capacity - req->tc->size)) {
-			/* We can use sdata */
-			req->tc->private = req->tc->sdata + req->tc->size;
-			rpinfo = (struct trans_rpage_info *)req->tc->private;
-			rpinfo->rp_alloc = 0;
+			P9_DPRINTK(P9_DEBUG_TRANS, "9p:Retry virtio request\n");
+			goto req_retry;
 		} else {
-			req->tc->private = kmalloc(rpinfo_size, GFP_NOFS);
-			if (!req->tc->private) {
-				P9_DPRINTK(P9_DEBUG_TRANS, "9p debug: "
-					"private kmalloc returned NULL");
-				return -ENOMEM;
-			}
-			rpinfo = (struct trans_rpage_info *)req->tc->private;
-			rpinfo->rp_alloc = 1;
+			spin_unlock_irqrestore(&chan->lock, flags);
+			P9_DPRINTK(P9_DEBUG_TRANS,
+					"9p debug: "
+					"virtio rpc add_buf returned failure");
+			return -EIO;
 		}
+	}
+	virtqueue_kick(chan->vq);
+	spin_unlock_irqrestore(&chan->lock, flags);
 
-		err = p9_payload_gup(req, &pdata_off, &pdata_len, nr_pages,
-				req->tc->id == P9_TREAD ? 1 : 0);
-		if (err < 0) {
-			if (rpinfo->rp_alloc)
-				kfree(rpinfo);
+	P9_DPRINTK(P9_DEBUG_TRANS, "9p debug: virtio request kicked\n");
+	return 0;
+}
+
+static int p9_get_mapped_pages(struct virtio_chan *chan,
+			       struct page **pages, char *data,
+			       int nr_pages, int write, int kern_buf)
+{
+	int err;
+	if (!kern_buf) {
+		/*
+		 * We allow only p9_max_pages pinned. We wait for the
+		 * Other zc request to finish here
+		 */
+		if (atomic_read(&vp_pinned) >= chan->p9_max_pages) {
+			err = wait_event_interruptible(vp_wq,
+			      (atomic_read(&vp_pinned) < chan->p9_max_pages));
+			if (err == -ERESTARTSYS)
+				return err;
+		}
+		err = p9_payload_gup(data, &nr_pages, pages, write);
+		if (err < 0)
 			return err;
-		} else {
-			atomic_add(rpinfo->rp_nr_pages, &vp_pinned);
+		atomic_add(nr_pages, &vp_pinned);
+	} else {
+		/* kernel buffer, no need to pin pages */
+		int s, index = 0;
+		int count = nr_pages;
+		while (nr_pages) {
+			s = rest_of_page(data);
+			pages[index++] = virt_to_page(data);
+			data += s;
+			nr_pages--;
 		}
+		nr_pages = count;
 	}
+	return nr_pages;
+}
 
-req_retry_pinned:
-	spin_lock_irqsave(&chan->lock, flags);
+/**
+ * p9_virtio_zc_request - issue a zero copy request
+ * @client: client instance issuing the request
+ * @req: request to be issued
+ * @uidata: user bffer that should be ued for zero copy read
+ * @uodata: user buffer that shoud be user for zero copy write
+ * @inlen: read buffer size
+ * @olen: write buffer size
+ * @hdrlen: reader header size, This is the size of response protocol data
+ *
+ */
+static int
+p9_virtio_zc_request(struct p9_client *client, struct p9_req_t *req,
+		     char *uidata, char *uodata, int inlen,
+		     int outlen, int in_hdr_len, int kern_buf)
+{
+	int in, out, err;
+	unsigned long flags;
+	int in_nr_pages = 0, out_nr_pages = 0;
+	struct page **in_pages = NULL, **out_pages = NULL;
+	struct virtio_chan *chan = client->trans;
 
-	/* Handle out VirtIO ring buffers */
-	out = pack_sg_list(chan->sg, 0, VIRTQUEUE_NUM, req->tc->sdata,
-			req->tc->size);
-
-	if (req->tc->pbuf_size && (req->tc->id == P9_TWRITE)) {
-		/* We have additional write payload buffer to take care */
-		if (req->tc->pubuf && P9_IS_USER_CONTEXT) {
-			outp = pack_sg_list_p(chan->sg, out, VIRTQUEUE_NUM,
-					pdata_off, rpinfo->rp_data, pdata_len);
-		} else {
-			char *pbuf;
-			if (req->tc->pubuf)
-				pbuf = (__force char *) req->tc->pubuf;
-			else
-				pbuf = req->tc->pkbuf;
-			outp = pack_sg_list(chan->sg, out, VIRTQUEUE_NUM, pbuf,
-					req->tc->pbuf_size);
+	P9_DPRINTK(P9_DEBUG_TRANS, "9p debug: virtio request\n");
+
+	if (uodata) {
+		out_nr_pages = p9_nr_pages(uodata, outlen);
+		out_pages = kmalloc(sizeof(struct page *) * out_nr_pages,
+				    GFP_NOFS);
+		if (!out_pages) {
+			err = -ENOMEM;
+			goto err_out;
+		}
+		out_nr_pages = p9_get_mapped_pages(chan, out_pages, uodata,
+						   out_nr_pages, 0, kern_buf);
+		if (out_nr_pages < 0) {
+			err = out_nr_pages;
+			kfree(out_pages);
+			out_pages = NULL;
+			goto err_out;
 		}
-		out += outp;
 	}
-
-	/* Handle in VirtIO ring buffers */
-	if (req->tc->pbuf_size &&
-		((req->tc->id == P9_TREAD) || (req->tc->id == P9_TREADDIR))) {
-		/*
-		 * Take care of additional Read payload.
-		 * 11 is the read/write header = PDU Header(7) + IO Size (4).
-		 * Arrange in such a way that server places header in the
-		 * alloced memory and payload onto the user buffer.
-		 */
-		inp = pack_sg_list(chan->sg, out,
-				   VIRTQUEUE_NUM, req->rc->sdata, 11);
-		/*
-		 * Running executables in the filesystem may result in
-		 * a read request with kernel buffer as opposed to user buffer.
-		 */
-		if (req->tc->pubuf && P9_IS_USER_CONTEXT) {
-			in = pack_sg_list_p(chan->sg, out+inp, VIRTQUEUE_NUM,
-					pdata_off, rpinfo->rp_data, pdata_len);
-		} else {
-			char *pbuf;
-			if (req->tc->pubuf)
-				pbuf = (__force char *) req->tc->pubuf;
-			else
-				pbuf = req->tc->pkbuf;
-
-			in = pack_sg_list(chan->sg, out+inp, VIRTQUEUE_NUM,
-					pbuf, req->tc->pbuf_size);
+	if (uidata) {
+		in_nr_pages = p9_nr_pages(uidata, inlen);
+		in_pages = kmalloc(sizeof(struct page *) * in_nr_pages,
+				   GFP_NOFS);
+		if (!in_pages) {
+			err = -ENOMEM;
+			goto err_out;
+		}
+		in_nr_pages = p9_get_mapped_pages(chan, in_pages, uidata,
+						  in_nr_pages, 1, kern_buf);
+		if (in_nr_pages < 0) {
+			err = in_nr_pages;
+			kfree(in_pages);
+			in_pages = NULL;
+			goto err_out;
 		}
-		in += inp;
-	} else {
-		in = pack_sg_list(chan->sg, out, VIRTQUEUE_NUM,
-				  req->rc->sdata, req->rc->capacity);
 	}
+	req->status = REQ_STATUS_SENT;
+req_retry_pinned:
+	spin_lock_irqsave(&chan->lock, flags);
+	/* out data */
+	out = pack_sg_list(chan->sg, 0,
+			   VIRTQUEUE_NUM, req->tc->sdata, req->tc->size);
+
+	if (out_pages)
+		out += pack_sg_list_p(chan->sg, out, VIRTQUEUE_NUM,
+				      out_pages, out_nr_pages, uodata, outlen);
+	/*
+	 * Take care of in data
+	 * For example TREAD have 11.
+	 * 11 is the read/write header = PDU Header(7) + IO Size (4).
+	 * Arrange in such a way that server places header in the
+	 * alloced memory and payload onto the user buffer.
+	 */
+	in = pack_sg_list(chan->sg, out,
+			  VIRTQUEUE_NUM, req->rc->sdata, in_hdr_len);
+	if (in_pages)
+		in += pack_sg_list_p(chan->sg, out + in, VIRTQUEUE_NUM,
+				     in_pages, in_nr_pages, uidata, inlen);
 
 	err = virtqueue_add_buf(chan->vq, chan->sg, out, in, req->tc);
 	if (err < 0) {
@@ -376,28 +419,45 @@ req_retry_pinned:
 			chan->ring_bufs_avail = 0;
 			spin_unlock_irqrestore(&chan->lock, flags);
 			err = wait_event_interruptible(*chan->vc_wq,
-							chan->ring_bufs_avail);
+						       chan->ring_bufs_avail);
 			if (err  == -ERESTARTSYS)
-				return err;
+				goto err_out;
 
 			P9_DPRINTK(P9_DEBUG_TRANS, "9p:Retry virtio request\n");
 			goto req_retry_pinned;
 		} else {
 			spin_unlock_irqrestore(&chan->lock, flags);
 			P9_DPRINTK(P9_DEBUG_TRANS,
-					"9p debug: "
-					"virtio rpc add_buf returned failure");
-			if (rpinfo && rpinfo->rp_alloc)
-				kfree(rpinfo);
-			return -EIO;
+				   "9p debug: "
+				   "virtio rpc add_buf returned failure");
+			err = -EIO;
+			goto err_out;
 		}
 	}
-
 	virtqueue_kick(chan->vq);
 	spin_unlock_irqrestore(&chan->lock, flags);
-
 	P9_DPRINTK(P9_DEBUG_TRANS, "9p debug: virtio request kicked\n");
-	return 0;
+	err = wait_event_interruptible(*req->wq,
+				       req->status >= REQ_STATUS_RCVD);
+	/*
+	 * Non kernel buffers are pinned, unpin them
+	 */
+err_out:
+	if (!kern_buf) {
+		if (in_pages) {
+			p9_release_pages(in_pages, in_nr_pages);
+			atomic_sub(in_nr_pages, &vp_pinned);
+		}
+		if (out_pages) {
+			p9_release_pages(out_pages, out_nr_pages);
+			atomic_sub(out_nr_pages, &vp_pinned);
+		}
+		/* wakeup anybody waiting for slots to pin pages */
+		wake_up(&vp_wq);
+	}
+	kfree(in_pages);
+	kfree(out_pages);
+	return err;
 }
 
 static ssize_t p9_mount_tag_show(struct device *dev,
@@ -591,8 +651,8 @@ static struct p9_trans_module p9_virtio_trans = {
 	.create = p9_virtio_create,
 	.close = p9_virtio_close,
 	.request = p9_virtio_request,
+	.zc_request = p9_virtio_zc_request,
 	.cancel = p9_virtio_cancel,
-
 	/*
 	 * We leave one entry for input and one entry for response
 	 * headers. We also skip one more entry to accomodate, address
@@ -600,7 +660,6 @@ static struct p9_trans_module p9_virtio_trans = {
 	 * page in zero copy.
 	 */
 	.maxsize = PAGE_SIZE * (VIRTQUEUE_NUM - 3),
-	.pref = P9_TRANS_PREF_PAYLOAD_SEP,
 	.def = 0,
 	.owner = THIS_MODULE,
 };