summaryrefslogtreecommitdiff
path: root/net/9p
diff options
context:
space:
mode:
Diffstat (limited to 'net/9p')
-rw-r--r--net/9p/Kconfig36
-rw-r--r--net/9p/Makefile18
-rw-r--r--net/9p/client.c2263
-rw-r--r--net/9p/error.c247
-rw-r--r--net/9p/mod.c201
-rw-r--r--net/9p/protocol.c628
-rw-r--r--net/9p/protocol.h34
-rw-r--r--net/9p/trans_common.c29
-rw-r--r--net/9p/trans_common.h15
-rw-r--r--net/9p/trans_fd.c1117
-rw-r--r--net/9p/trans_rdma.c801
-rw-r--r--net/9p/trans_virtio.c776
-rw-r--r--net/9p/util.c141
13 files changed, 6306 insertions, 0 deletions
diff --git a/net/9p/Kconfig b/net/9p/Kconfig
new file mode 100644
index 000000000..a75174a33
--- /dev/null
+++ b/net/9p/Kconfig
@@ -0,0 +1,36 @@
+#
+# 9P protocol configuration
+#
+
+menuconfig NET_9P
+ depends on NET
+ tristate "Plan 9 Resource Sharing Support (9P2000)"
+ help
+ If you say Y here, you will get experimental support for
+ Plan 9 resource sharing via the 9P2000 protocol.
+
+ See <http://v9fs.sf.net> for more information.
+
+ If unsure, say N.
+
+if NET_9P
+
+config NET_9P_VIRTIO
+ depends on VIRTIO
+ tristate "9P Virtio Transport"
+ help
+ This builds support for a transports between
+ guest partitions and a host partition.
+
+config NET_9P_RDMA
+ depends on INET && INFINIBAND && INFINIBAND_ADDR_TRANS
+ tristate "9P RDMA Transport (Experimental)"
+ help
+ This builds support for an RDMA transport.
+
+config NET_9P_DEBUG
+ bool "Debug information"
+ help
+ Say Y if you want the 9P subsystem to log debug information.
+
+endif
diff --git a/net/9p/Makefile b/net/9p/Makefile
new file mode 100644
index 000000000..a0874cc1f
--- /dev/null
+++ b/net/9p/Makefile
@@ -0,0 +1,18 @@
+obj-$(CONFIG_NET_9P) := 9pnet.o
+obj-$(CONFIG_NET_9P_VIRTIO) += 9pnet_virtio.o
+obj-$(CONFIG_NET_9P_RDMA) += 9pnet_rdma.o
+
+9pnet-objs := \
+ mod.o \
+ client.o \
+ error.o \
+ util.o \
+ protocol.o \
+ trans_fd.o \
+ trans_common.o \
+
+9pnet_virtio-objs := \
+ trans_virtio.o \
+
+9pnet_rdma-objs := \
+ trans_rdma.o \
diff --git a/net/9p/client.c b/net/9p/client.c
new file mode 100644
index 000000000..81925b923
--- /dev/null
+++ b/net/9p/client.c
@@ -0,0 +1,2263 @@
+/*
+ * net/9p/clnt.c
+ *
+ * 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
+ * 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
+ *
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#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/slab.h>
+#include <linux/sched.h>
+#include <linux/uaccess.h>
+#include <linux/uio.h>
+#include <net/9p/9p.h>
+#include <linux/parser.h>
+#include <net/9p/client.h>
+#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
+ */
+
+enum {
+ Opt_msize,
+ Opt_trans,
+ Opt_legacy,
+ Opt_version,
+ Opt_err,
+};
+
+static const match_table_t tokens = {
+ {Opt_msize, "msize=%u"},
+ {Opt_legacy, "noextend"},
+ {Opt_trans, "trans=%s"},
+ {Opt_version, "version=%s"},
+ {Opt_err, NULL},
+};
+
+inline int p9_is_proto_dotl(struct p9_client *clnt)
+{
+ return clnt->proto_version == p9_proto_2000L;
+}
+EXPORT_SYMBOL(p9_is_proto_dotl);
+
+inline int p9_is_proto_dotu(struct p9_client *clnt)
+{
+ return clnt->proto_version == p9_proto_2000u;
+}
+EXPORT_SYMBOL(p9_is_proto_dotu);
+
+/*
+ * Some error codes are taken directly from the server replies,
+ * make sure they are valid.
+ */
+static int safe_errno(int err)
+{
+ if ((err > 0) || (err < -MAX_ERRNO)) {
+ p9_debug(P9_DEBUG_ERROR, "Invalid error code %d\n", err);
+ return -EPROTO;
+ }
+ return err;
+}
+
+
+/* Interpret mount option for protocol version */
+static int get_protocol_version(char *s)
+{
+ int version = -EINVAL;
+
+ if (!strcmp(s, "9p2000")) {
+ version = p9_proto_legacy;
+ p9_debug(P9_DEBUG_9P, "Protocol version: Legacy\n");
+ } else if (!strcmp(s, "9p2000.u")) {
+ version = p9_proto_2000u;
+ p9_debug(P9_DEBUG_9P, "Protocol version: 9P2000.u\n");
+ } else if (!strcmp(s, "9p2000.L")) {
+ version = p9_proto_2000L;
+ p9_debug(P9_DEBUG_9P, "Protocol version: 9P2000.L\n");
+ } else
+ pr_info("Unknown protocol version %s\n", s);
+
+ return version;
+}
+
+/**
+ * parse_options - parse mount options into client structure
+ * @opts: options string passed from mount
+ * @clnt: existing v9fs client information
+ *
+ * Return 0 upon success, -ERRNO upon failure
+ */
+
+static int parse_opts(char *opts, struct p9_client *clnt)
+{
+ char *options, *tmp_options;
+ char *p;
+ substring_t args[MAX_OPT_ARGS];
+ int option;
+ char *s;
+ int ret = 0;
+
+ clnt->proto_version = p9_proto_2000L;
+ clnt->msize = 8192;
+
+ if (!opts)
+ return 0;
+
+ tmp_options = kstrdup(opts, GFP_KERNEL);
+ if (!tmp_options) {
+ p9_debug(P9_DEBUG_ERROR,
+ "failed to allocate copy of option string\n");
+ return -ENOMEM;
+ }
+ options = tmp_options;
+
+ while ((p = strsep(&options, ",")) != NULL) {
+ int token, r;
+ if (!*p)
+ continue;
+ token = match_token(p, tokens, args);
+ switch (token) {
+ case Opt_msize:
+ r = match_int(&args[0], &option);
+ if (r < 0) {
+ p9_debug(P9_DEBUG_ERROR,
+ "integer field, but no integer?\n");
+ ret = r;
+ continue;
+ }
+ clnt->msize = option;
+ break;
+ case Opt_trans:
+ s = match_strdup(&args[0]);
+ if (!s) {
+ ret = -ENOMEM;
+ p9_debug(P9_DEBUG_ERROR,
+ "problem allocating copy of trans arg\n");
+ goto free_and_return;
+ }
+ clnt->trans_mod = v9fs_get_trans_by_name(s);
+ if (clnt->trans_mod == NULL) {
+ pr_info("Could not find request transport: %s\n",
+ s);
+ ret = -EINVAL;
+ kfree(s);
+ goto free_and_return;
+ }
+ kfree(s);
+ break;
+ case Opt_legacy:
+ clnt->proto_version = p9_proto_legacy;
+ break;
+ case Opt_version:
+ s = match_strdup(&args[0]);
+ if (!s) {
+ ret = -ENOMEM;
+ p9_debug(P9_DEBUG_ERROR,
+ "problem allocating copy of version arg\n");
+ goto free_and_return;
+ }
+ ret = get_protocol_version(s);
+ if (ret == -EINVAL) {
+ kfree(s);
+ goto free_and_return;
+ }
+ kfree(s);
+ clnt->proto_version = ret;
+ break;
+ default:
+ continue;
+ }
+ }
+
+free_and_return:
+ kfree(tmp_options);
+ return ret;
+}
+
+static struct p9_fcall *p9_fcall_alloc(int alloc_msize)
+{
+ struct p9_fcall *fc;
+ fc = kmalloc(sizeof(struct p9_fcall) + alloc_msize, GFP_NOFS);
+ if (!fc)
+ return NULL;
+ fc->capacity = alloc_msize;
+ fc->sdata = (char *) fc + sizeof(struct p9_fcall);
+ return fc;
+}
+
+/**
+ * p9_tag_alloc - lookup/allocate a request by tag
+ * @c: client session to lookup tag within
+ * @tag: numeric id for transaction
+ *
+ * this is a simple array lookup, but will grow the
+ * request_slots as necessary to accommodate transaction
+ * ids which did not previously have a slot.
+ *
+ * this code relies on the client spinlock to manage locks, its
+ * possible we should switch to something else, but I'd rather
+ * stick with something low-overhead for the common case.
+ *
+ */
+
+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 */
+ tag++;
+
+ if (tag >= c->max_tag) {
+ spin_lock_irqsave(&c->lock, flags);
+ /* check again since original check was outside of lock */
+ while (tag >= c->max_tag) {
+ row = (tag / P9_ROW_MAXTAG);
+ c->reqs[row] = kcalloc(P9_ROW_MAXTAG,
+ sizeof(struct p9_req_t), GFP_ATOMIC);
+
+ if (!c->reqs[row]) {
+ pr_err("Couldn't grow tag array\n");
+ spin_unlock_irqrestore(&c->lock, flags);
+ return ERR_PTR(-ENOMEM);
+ }
+ for (col = 0; col < P9_ROW_MAXTAG; col++) {
+ c->reqs[row][col].status = REQ_STATUS_IDLE;
+ c->reqs[row][col].tc = NULL;
+ }
+ c->max_tag += P9_ROW_MAXTAG;
+ }
+ spin_unlock_irqrestore(&c->lock, flags);
+ }
+ row = tag / P9_ROW_MAXTAG;
+ col = tag % P9_ROW_MAXTAG;
+
+ req = &c->reqs[row][col];
+ if (!req->wq) {
+ req->wq = kmalloc(sizeof(wait_queue_head_t), GFP_NOFS);
+ if (!req->wq)
+ goto grow_failed;
+ init_waitqueue_head(req->wq);
+ }
+
+ if (!req->tc)
+ req->tc = p9_fcall_alloc(alloc_msize);
+ if (!req->rc)
+ req->rc = p9_fcall_alloc(alloc_msize);
+ if (!req->tc || !req->rc)
+ goto grow_failed;
+
+ p9pdu_reset(req->tc);
+ p9pdu_reset(req->rc);
+
+ req->tc->tag = tag-1;
+ req->status = REQ_STATUS_ALLOC;
+
+ return req;
+
+grow_failed:
+ pr_err("Couldn't grow tag array\n");
+ kfree(req->tc);
+ kfree(req->rc);
+ kfree(req->wq);
+ req->tc = req->rc = NULL;
+ req->wq = NULL;
+ return ERR_PTR(-ENOMEM);
+}
+
+/**
+ * p9_tag_lookup - lookup a request by tag
+ * @c: client session to lookup tag within
+ * @tag: numeric id for transaction
+ *
+ */
+
+struct p9_req_t *p9_tag_lookup(struct p9_client *c, u16 tag)
+{
+ int row, col;
+
+ /* This looks up the original request by tag so we know which
+ * buffer to read the data into */
+ tag++;
+
+ if(tag >= c->max_tag)
+ return NULL;
+
+ row = tag / P9_ROW_MAXTAG;
+ col = tag % P9_ROW_MAXTAG;
+
+ return &c->reqs[row][col];
+}
+EXPORT_SYMBOL(p9_tag_lookup);
+
+/**
+ * p9_tag_init - setup tags structure and contents
+ * @c: v9fs client struct
+ *
+ * This initializes the tags structure for each client instance.
+ *
+ */
+
+static int p9_tag_init(struct p9_client *c)
+{
+ int err = 0;
+
+ c->tagpool = p9_idpool_create();
+ if (IS_ERR(c->tagpool)) {
+ err = PTR_ERR(c->tagpool);
+ goto error;
+ }
+ err = p9_idpool_get(c->tagpool); /* reserve tag 0 */
+ if (err < 0) {
+ p9_idpool_destroy(c->tagpool);
+ goto error;
+ }
+ c->max_tag = 0;
+error:
+ return err;
+}
+
+/**
+ * p9_tag_cleanup - cleans up tags structure and reclaims resources
+ * @c: v9fs client struct
+ *
+ * This frees resources associated with the tags structure
+ *
+ */
+static void p9_tag_cleanup(struct p9_client *c)
+{
+ int row, col;
+
+ /* check to insure all requests are idle */
+ for (row = 0; row < (c->max_tag/P9_ROW_MAXTAG); row++) {
+ for (col = 0; col < P9_ROW_MAXTAG; col++) {
+ if (c->reqs[row][col].status != REQ_STATUS_IDLE) {
+ p9_debug(P9_DEBUG_MUX,
+ "Attempting to cleanup non-free tag %d,%d\n",
+ row, col);
+ /* TODO: delay execution of cleanup */
+ return;
+ }
+ }
+ }
+
+ if (c->tagpool) {
+ p9_idpool_put(0, c->tagpool); /* free reserved tag 0 */
+ p9_idpool_destroy(c->tagpool);
+ }
+
+ /* free requests associated with tags */
+ for (row = 0; row < (c->max_tag/P9_ROW_MAXTAG); row++) {
+ for (col = 0; col < P9_ROW_MAXTAG; col++) {
+ kfree(c->reqs[row][col].wq);
+ kfree(c->reqs[row][col].tc);
+ kfree(c->reqs[row][col].rc);
+ }
+ kfree(c->reqs[row]);
+ }
+ c->max_tag = 0;
+}
+
+/**
+ * p9_free_req - free a request and clean-up as necessary
+ * c: client state
+ * r: request to release
+ *
+ */
+
+static void p9_free_req(struct p9_client *c, struct p9_req_t *r)
+{
+ int tag = r->tc->tag;
+ p9_debug(P9_DEBUG_MUX, "clnt %p req %p tag: %d\n", c, r, tag);
+
+ r->status = REQ_STATUS_IDLE;
+ if (tag != P9_NOTAG && p9_idpool_check(tag, c->tagpool))
+ p9_idpool_put(tag, c->tagpool);
+}
+
+/**
+ * p9_client_cb - call back from transport to client
+ * c: client state
+ * req: request received
+ *
+ */
+void p9_client_cb(struct p9_client *c, struct p9_req_t *req, int status)
+{
+ p9_debug(P9_DEBUG_MUX, " tag %d\n", req->tc->tag);
+
+ /*
+ * This barrier is needed to make sure any change made to req before
+ * the other thread wakes up will indeed be seen by the waiting side.
+ */
+ smp_wmb();
+ req->status = status;
+
+ wake_up(req->wq);
+ p9_debug(P9_DEBUG_MUX, "wakeup: %d\n", req->tc->tag);
+}
+EXPORT_SYMBOL(p9_client_cb);
+
+/**
+ * p9_parse_header - parse header arguments out of a packet
+ * @pdu: packet to parse
+ * @size: size of packet
+ * @type: type of request
+ * @tag: tag of packet
+ * @rewind: set if we need to rewind offset afterwards
+ */
+
+int
+p9_parse_header(struct p9_fcall *pdu, int32_t *size, int8_t *type, int16_t *tag,
+ int rewind)
+{
+ int8_t r_type;
+ int16_t r_tag;
+ int32_t r_size;
+ int offset = pdu->offset;
+ int err;
+
+ pdu->offset = 0;
+ if (pdu->size == 0)
+ pdu->size = 7;
+
+ err = p9pdu_readf(pdu, 0, "dbw", &r_size, &r_type, &r_tag);
+ if (err)
+ goto rewind_and_exit;
+
+ pdu->size = r_size;
+ pdu->id = r_type;
+ pdu->tag = r_tag;
+
+ p9_debug(P9_DEBUG_9P, "<<< size=%d type: %d tag: %d\n",
+ pdu->size, pdu->id, pdu->tag);
+
+ if (type)
+ *type = r_type;
+ if (tag)
+ *tag = r_tag;
+ if (size)
+ *size = r_size;
+
+
+rewind_and_exit:
+ if (rewind)
+ pdu->offset = offset;
+ return err;
+}
+EXPORT_SYMBOL(p9_parse_header);
+
+/**
+ * p9_check_errors - check 9p packet for error return and process it
+ * @c: current client instance
+ * @req: request to parse and check for error conditions
+ *
+ * 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_errors(struct p9_client *c, struct p9_req_t *req)
+{
+ int8_t type;
+ int err;
+ 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_debug(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;
+ err = p9pdu_readf(req->rc, c->proto_version, "s?d",
+ &ename, &ecode);
+ if (err)
+ goto out_err;
+
+ if (p9_is_proto_dotu(c))
+ err = -ecode;
+
+ if (!err || !IS_ERR_VALUE(err)) {
+ err = p9_errstr2errno(ename, strlen(ename));
+
+ p9_debug(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_debug(P9_DEBUG_9P, "<<< RLERROR (%d)\n", -ecode);
+ }
+
+ return err;
+
+out_err:
+ p9_debug(P9_DEBUG_ERROR, "couldn't parse error%d\n", 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,
+ struct iov_iter *uidata, int in_hdrlen)
+{
+ 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_debug(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 */
+ int len;
+ /* 7 = header size for RERROR; */
+ int inline_len = in_hdrlen - 7;
+
+ len = req->rc->size - req->rc->offset;
+ if (len > (P9_ZC_HDR_SZ - 7)) {
+ err = -EFAULT;
+ goto out_err;
+ }
+
+ ename = &req->rc->sdata[req->rc->offset];
+ if (len > inline_len) {
+ /* We have error in external buffer */
+ err = copy_from_iter(ename + inline_len,
+ len - inline_len, uidata);
+ if (err != len - inline_len) {
+ err = -EFAULT;
+ goto out_err;
+ }
+ }
+ ename = NULL;
+ err = p9pdu_readf(req->rc, c->proto_version, "s?d",
+ &ename, &ecode);
+ if (err)
+ goto out_err;
+
+ if (p9_is_proto_dotu(c))
+ err = -ecode;
+
+ if (!err || !IS_ERR_VALUE(err)) {
+ err = p9_errstr2errno(ename, strlen(ename));
+
+ p9_debug(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_debug(P9_DEBUG_9P, "<<< RLERROR (%d)\n", -ecode);
+ }
+ return err;
+
+out_err:
+ p9_debug(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, ...);
+
+/**
+ * p9_client_flush - flush (cancel) a request
+ * @c: client state
+ * @oldreq: request to cancel
+ *
+ * This sents a flush for a particular request and links
+ * the flush request to the original request. The current
+ * code only supports a single flush request although the protocol
+ * allows for multiple flush requests to be sent for a single request.
+ *
+ */
+
+static int p9_client_flush(struct p9_client *c, struct p9_req_t *oldreq)
+{
+ struct p9_req_t *req;
+ int16_t oldtag;
+ int err;
+
+ err = p9_parse_header(oldreq->tc, NULL, NULL, &oldtag, 1);
+ if (err)
+ return err;
+
+ p9_debug(P9_DEBUG_9P, ">>> TFLUSH tag %d\n", oldtag);
+
+ req = p9_client_rpc(c, P9_TFLUSH, "w", oldtag);
+ if (IS_ERR(req))
+ return PTR_ERR(req);
+
+ /*
+ * if we haven't received a response for oldreq,
+ * remove it from the list
+ */
+ if (oldreq->status == REQ_STATUS_SENT)
+ if (c->trans_mod->cancelled)
+ c->trans_mod->cancelled(c, oldreq);
+
+ p9_free_req(c, req);
+ return 0;
+}
+
+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)
+{
+ int tag, err;
+ struct p9_req_t *req;
+
+ p9_debug(P9_DEBUG_MUX, "client %p op %d\n", c, type);
+
+ /* we allow for any status other than disconnected */
+ if (c->status == Disconnected)
+ return ERR_PTR(-EIO);
+
+ /* if status is begin_disconnected we allow only clunk request */
+ if ((c->status == BeginDisconnect) && (type != P9_TCLUNK))
+ return ERR_PTR(-EIO);
+
+ tag = P9_NOTAG;
+ if (type != P9_TVERSION) {
+ tag = p9_idpool_get(c->tagpool);
+ if (tag < 0)
+ return ERR_PTR(-ENOMEM);
+ }
+
+ req = p9_tag_alloc(c, tag, req_size);
+ if (IS_ERR(req))
+ return req;
+
+ /* marshall the data */
+ p9pdu_prepare(req->tc, tag, type);
+ err = p9pdu_vwritef(req->tc, c->proto_version, fmt, ap);
+ if (err)
+ goto reterr;
+ 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) {
+ if (err != -ERESTARTSYS && err != -EFAULT)
+ c->status = Disconnected;
+ goto reterr;
+ }
+again:
+ /* Wait for the response */
+ err = wait_event_interruptible(*req->wq,
+ req->status >= REQ_STATUS_RCVD);
+
+ /*
+ * Make sure our req is coherent with regard to updates in other
+ * threads - echoes to wmb() in the callback
+ */
+ smp_rmb();
+
+ if ((err == -ERESTARTSYS) && (c->status == Connected)
+ && (type == P9_TFLUSH)) {
+ sigpending = 1;
+ clear_thread_flag(TIF_SIGPENDING);
+ goto again;
+ }
+
+ if (req->status == REQ_STATUS_ERROR) {
+ p9_debug(P9_DEBUG_ERROR, "req_status error %d\n", req->t_err);
+ err = req->t_err;
+ }
+ if ((err == -ERESTARTSYS) && (c->status == Connected)) {
+ p9_debug(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_errors(c, req);
+ trace_9p_client_res(c, type, req->rc->tag, err);
+ if (!err)
+ return req;
+reterr:
+ p9_free_req(c, req);
+ return ERR_PTR(safe_errno(err));
+}
+
+/**
+ * p9_client_zc_rpc - issue a request and wait for a response
+ * @c: client session
+ * @type: type of request
+ * @uidata: destination for zero copy read
+ * @uodata: source 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,
+ struct iov_iter *uidata,
+ struct iov_iter *uodata,
+ int inlen, int olen, int in_hdrlen,
+ 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;
+
+ err = c->trans_mod->zc_request(c, req, uidata, uodata,
+ inlen, olen, in_hdrlen);
+ if (err < 0) {
+ if (err == -EIO)
+ c->status = Disconnected;
+ if (err != -ERESTARTSYS)
+ goto reterr;
+ }
+ if (req->status == REQ_STATUS_ERROR) {
+ p9_debug(P9_DEBUG_ERROR, "req_status error %d\n", req->t_err);
+ err = req->t_err;
+ }
+ if ((err == -ERESTARTSYS) && (c->status == Connected)) {
+ p9_debug(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);
+ trace_9p_client_res(c, type, req->rc->tag, err);
+ if (!err)
+ return req;
+reterr:
+ p9_free_req(c, req);
+ return ERR_PTR(safe_errno(err));
+}
+
+static struct p9_fid *p9_fid_create(struct p9_client *clnt)
+{
+ int ret;
+ struct p9_fid *fid;
+ unsigned long flags;
+
+ p9_debug(P9_DEBUG_FID, "clnt %p\n", clnt);
+ fid = kmalloc(sizeof(struct p9_fid), GFP_KERNEL);
+ if (!fid)
+ return ERR_PTR(-ENOMEM);
+
+ ret = p9_idpool_get(clnt->fidpool);
+ if (ret < 0) {
+ ret = -ENOSPC;
+ goto error;
+ }
+ fid->fid = ret;
+
+ memset(&fid->qid, 0, sizeof(struct p9_qid));
+ fid->mode = -1;
+ fid->uid = current_fsuid();
+ fid->clnt = clnt;
+ fid->rdir = NULL;
+ spin_lock_irqsave(&clnt->lock, flags);
+ list_add(&fid->flist, &clnt->fidlist);
+ spin_unlock_irqrestore(&clnt->lock, flags);
+
+ return fid;
+
+error:
+ kfree(fid);
+ return ERR_PTR(ret);
+}
+
+static void p9_fid_destroy(struct p9_fid *fid)
+{
+ struct p9_client *clnt;
+ unsigned long flags;
+
+ p9_debug(P9_DEBUG_FID, "fid %d\n", fid->fid);
+ clnt = fid->clnt;
+ p9_idpool_put(fid->fid, clnt->fidpool);
+ spin_lock_irqsave(&clnt->lock, flags);
+ list_del(&fid->flist);
+ spin_unlock_irqrestore(&clnt->lock, flags);
+ kfree(fid->rdir);
+ kfree(fid);
+}
+
+static int p9_client_version(struct p9_client *c)
+{
+ int err = 0;
+ struct p9_req_t *req;
+ char *version;
+ int msize;
+
+ p9_debug(P9_DEBUG_9P, ">>> TVERSION msize %d protocol %d\n",
+ c->msize, c->proto_version);
+
+ switch (c->proto_version) {
+ case p9_proto_2000L:
+ req = p9_client_rpc(c, P9_TVERSION, "ds",
+ c->msize, "9P2000.L");
+ break;
+ case p9_proto_2000u:
+ req = p9_client_rpc(c, P9_TVERSION, "ds",
+ c->msize, "9P2000.u");
+ break;
+ case p9_proto_legacy:
+ req = p9_client_rpc(c, P9_TVERSION, "ds",
+ c->msize, "9P2000");
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ if (IS_ERR(req))
+ return PTR_ERR(req);
+
+ err = p9pdu_readf(req->rc, c->proto_version, "ds", &msize, &version);
+ if (err) {
+ p9_debug(P9_DEBUG_9P, "version error %d\n", err);
+ trace_9p_protocol_dump(c, req->rc);
+ goto error;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RVERSION msize %d %s\n", msize, version);
+ if (!strncmp(version, "9P2000.L", 8))
+ c->proto_version = p9_proto_2000L;
+ else if (!strncmp(version, "9P2000.u", 8))
+ c->proto_version = p9_proto_2000u;
+ else if (!strncmp(version, "9P2000", 6))
+ c->proto_version = p9_proto_legacy;
+ else {
+ err = -EREMOTEIO;
+ goto error;
+ }
+
+ if (msize < c->msize)
+ c->msize = msize;
+
+error:
+ kfree(version);
+ p9_free_req(c, req);
+
+ return err;
+}
+
+struct p9_client *p9_client_create(const char *dev_name, char *options)
+{
+ int err;
+ struct p9_client *clnt;
+ char *client_id;
+
+ err = 0;
+ clnt = kmalloc(sizeof(struct p9_client), GFP_KERNEL);
+ if (!clnt)
+ return ERR_PTR(-ENOMEM);
+
+ clnt->trans_mod = NULL;
+ clnt->trans = NULL;
+
+ client_id = utsname()->nodename;
+ memcpy(clnt->name, client_id, strlen(client_id) + 1);
+
+ spin_lock_init(&clnt->lock);
+ INIT_LIST_HEAD(&clnt->fidlist);
+
+ err = p9_tag_init(clnt);
+ if (err < 0)
+ goto free_client;
+
+ err = parse_opts(options, clnt);
+ if (err < 0)
+ goto destroy_tagpool;
+
+ if (!clnt->trans_mod)
+ clnt->trans_mod = v9fs_get_default_trans();
+
+ if (clnt->trans_mod == NULL) {
+ err = -EPROTONOSUPPORT;
+ p9_debug(P9_DEBUG_ERROR,
+ "No transport defined or default transport\n");
+ goto destroy_tagpool;
+ }
+
+ clnt->fidpool = p9_idpool_create();
+ if (IS_ERR(clnt->fidpool)) {
+ err = PTR_ERR(clnt->fidpool);
+ goto put_trans;
+ }
+
+ p9_debug(P9_DEBUG_MUX, "clnt %p trans %p msize %d protocol %d\n",
+ clnt, clnt->trans_mod, clnt->msize, clnt->proto_version);
+
+ err = clnt->trans_mod->create(clnt, dev_name, options);
+ if (err)
+ goto destroy_fidpool;
+
+ if (clnt->msize > clnt->trans_mod->maxsize)
+ clnt->msize = clnt->trans_mod->maxsize;
+
+ err = p9_client_version(clnt);
+ if (err)
+ goto close_trans;
+
+ return clnt;
+
+close_trans:
+ clnt->trans_mod->close(clnt);
+destroy_fidpool:
+ p9_idpool_destroy(clnt->fidpool);
+put_trans:
+ v9fs_put_trans(clnt->trans_mod);
+destroy_tagpool:
+ p9_idpool_destroy(clnt->tagpool);
+free_client:
+ kfree(clnt);
+ return ERR_PTR(err);
+}
+EXPORT_SYMBOL(p9_client_create);
+
+void p9_client_destroy(struct p9_client *clnt)
+{
+ struct p9_fid *fid, *fidptr;
+
+ p9_debug(P9_DEBUG_MUX, "clnt %p\n", clnt);
+
+ if (clnt->trans_mod)
+ clnt->trans_mod->close(clnt);
+
+ v9fs_put_trans(clnt->trans_mod);
+
+ list_for_each_entry_safe(fid, fidptr, &clnt->fidlist, flist) {
+ pr_info("Found fid %d not clunked\n", fid->fid);
+ p9_fid_destroy(fid);
+ }
+
+ if (clnt->fidpool)
+ p9_idpool_destroy(clnt->fidpool);
+
+ p9_tag_cleanup(clnt);
+
+ kfree(clnt);
+}
+EXPORT_SYMBOL(p9_client_destroy);
+
+void p9_client_disconnect(struct p9_client *clnt)
+{
+ p9_debug(P9_DEBUG_9P, "clnt %p\n", clnt);
+ clnt->status = Disconnected;
+}
+EXPORT_SYMBOL(p9_client_disconnect);
+
+void p9_client_begin_disconnect(struct p9_client *clnt)
+{
+ p9_debug(P9_DEBUG_9P, "clnt %p\n", clnt);
+ clnt->status = BeginDisconnect;
+}
+EXPORT_SYMBOL(p9_client_begin_disconnect);
+
+struct p9_fid *p9_client_attach(struct p9_client *clnt, struct p9_fid *afid,
+ char *uname, kuid_t n_uname, char *aname)
+{
+ int err = 0;
+ struct p9_req_t *req;
+ struct p9_fid *fid;
+ struct p9_qid qid;
+
+
+ p9_debug(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);
+ fid = NULL;
+ goto error;
+ }
+ fid->uid = n_uname;
+
+ req = p9_client_rpc(clnt, P9_TATTACH, "ddss?u", fid->fid,
+ afid ? afid->fid : P9_NOFID, uname, aname, n_uname);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "Q", &qid);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ p9_free_req(clnt, req);
+ goto error;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RATTACH qid %x.%llx.%x\n",
+ qid.type, (unsigned long long)qid.path, qid.version);
+
+ memmove(&fid->qid, &qid, sizeof(struct p9_qid));
+
+ p9_free_req(clnt, req);
+ return fid;
+
+error:
+ if (fid)
+ p9_fid_destroy(fid);
+ return ERR_PTR(err);
+}
+EXPORT_SYMBOL(p9_client_attach);
+
+struct p9_fid *p9_client_walk(struct p9_fid *oldfid, uint16_t nwname,
+ char **wnames, int clone)
+{
+ int err;
+ struct p9_client *clnt;
+ struct p9_fid *fid;
+ struct p9_qid *wqids;
+ struct p9_req_t *req;
+ uint16_t nwqids, count;
+
+ err = 0;
+ wqids = NULL;
+ clnt = oldfid->clnt;
+ if (clone) {
+ fid = p9_fid_create(clnt);
+ if (IS_ERR(fid)) {
+ err = PTR_ERR(fid);
+ fid = NULL;
+ goto error;
+ }
+
+ fid->uid = oldfid->uid;
+ } else
+ fid = oldfid;
+
+
+ p9_debug(P9_DEBUG_9P, ">>> TWALK fids %d,%d nwname %ud wname[0] %s\n",
+ oldfid->fid, fid->fid, nwname, wnames ? wnames[0] : NULL);
+
+ req = p9_client_rpc(clnt, P9_TWALK, "ddT", oldfid->fid, fid->fid,
+ nwname, wnames);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "R", &nwqids, &wqids);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ p9_free_req(clnt, req);
+ goto clunk_fid;
+ }
+ p9_free_req(clnt, req);
+
+ p9_debug(P9_DEBUG_9P, "<<< RWALK nwqid %d:\n", nwqids);
+
+ if (nwqids != nwname) {
+ err = -ENOENT;
+ goto clunk_fid;
+ }
+
+ for (count = 0; count < nwqids; count++)
+ p9_debug(P9_DEBUG_9P, "<<< [%d] %x.%llx.%x\n",
+ count, wqids[count].type,
+ (unsigned long long)wqids[count].path,
+ wqids[count].version);
+
+ if (nwname)
+ memmove(&fid->qid, &wqids[nwqids - 1], sizeof(struct p9_qid));
+ else
+ fid->qid = oldfid->qid;
+
+ kfree(wqids);
+ return fid;
+
+clunk_fid:
+ kfree(wqids);
+ p9_client_clunk(fid);
+ fid = NULL;
+
+error:
+ if (fid && (fid != oldfid))
+ p9_fid_destroy(fid);
+
+ return ERR_PTR(err);
+}
+EXPORT_SYMBOL(p9_client_walk);
+
+int p9_client_open(struct p9_fid *fid, int mode)
+{
+ int err;
+ struct p9_client *clnt;
+ struct p9_req_t *req;
+ struct p9_qid qid;
+ int iounit;
+
+ clnt = fid->clnt;
+ p9_debug(P9_DEBUG_9P, ">>> %s fid %d mode %d\n",
+ p9_is_proto_dotl(clnt) ? "TLOPEN" : "TOPEN", fid->fid, mode);
+ err = 0;
+
+ if (fid->mode != -1)
+ return -EINVAL;
+
+ if (p9_is_proto_dotl(clnt))
+ req = p9_client_rpc(clnt, P9_TLOPEN, "dd", fid->fid, mode);
+ else
+ req = p9_client_rpc(clnt, P9_TOPEN, "db", fid->fid, mode);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "Qd", &qid, &iounit);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ goto free_and_error;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< %s qid %x.%llx.%x iounit %x\n",
+ p9_is_proto_dotl(clnt) ? "RLOPEN" : "ROPEN", qid.type,
+ (unsigned long long)qid.path, qid.version, iounit);
+
+ fid->mode = mode;
+ fid->iounit = iounit;
+
+free_and_error:
+ p9_free_req(clnt, req);
+error:
+ return err;
+}
+EXPORT_SYMBOL(p9_client_open);
+
+int p9_client_create_dotl(struct p9_fid *ofid, char *name, u32 flags, u32 mode,
+ kgid_t gid, struct p9_qid *qid)
+{
+ int err = 0;
+ struct p9_client *clnt;
+ struct p9_req_t *req;
+ int iounit;
+
+ p9_debug(P9_DEBUG_9P,
+ ">>> TLCREATE fid %d name %s flags %d mode %d gid %d\n",
+ ofid->fid, name, flags, mode,
+ from_kgid(&init_user_ns, gid));
+ clnt = ofid->clnt;
+
+ if (ofid->mode != -1)
+ return -EINVAL;
+
+ req = p9_client_rpc(clnt, P9_TLCREATE, "dsddg", ofid->fid, name, flags,
+ mode, gid);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "Qd", qid, &iounit);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ goto free_and_error;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RLCREATE qid %x.%llx.%x iounit %x\n",
+ qid->type,
+ (unsigned long long)qid->path,
+ qid->version, iounit);
+
+ ofid->mode = mode;
+ ofid->iounit = iounit;
+
+free_and_error:
+ p9_free_req(clnt, req);
+error:
+ return err;
+}
+EXPORT_SYMBOL(p9_client_create_dotl);
+
+int p9_client_fcreate(struct p9_fid *fid, char *name, u32 perm, int mode,
+ char *extension)
+{
+ int err;
+ struct p9_client *clnt;
+ struct p9_req_t *req;
+ struct p9_qid qid;
+ int iounit;
+
+ p9_debug(P9_DEBUG_9P, ">>> TCREATE fid %d name %s perm %d mode %d\n",
+ fid->fid, name, perm, mode);
+ err = 0;
+ clnt = fid->clnt;
+
+ if (fid->mode != -1)
+ return -EINVAL;
+
+ req = p9_client_rpc(clnt, P9_TCREATE, "dsdb?s", fid->fid, name, perm,
+ mode, extension);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "Qd", &qid, &iounit);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ goto free_and_error;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RCREATE qid %x.%llx.%x iounit %x\n",
+ qid.type,
+ (unsigned long long)qid.path,
+ qid.version, iounit);
+
+ fid->mode = mode;
+ fid->iounit = iounit;
+
+free_and_error:
+ p9_free_req(clnt, req);
+error:
+ return err;
+}
+EXPORT_SYMBOL(p9_client_fcreate);
+
+int p9_client_symlink(struct p9_fid *dfid, char *name, char *symtgt, kgid_t gid,
+ struct p9_qid *qid)
+{
+ int err = 0;
+ struct p9_client *clnt;
+ struct p9_req_t *req;
+
+ p9_debug(P9_DEBUG_9P, ">>> TSYMLINK dfid %d name %s symtgt %s\n",
+ dfid->fid, name, symtgt);
+ clnt = dfid->clnt;
+
+ req = p9_client_rpc(clnt, P9_TSYMLINK, "dssg", dfid->fid, name, symtgt,
+ gid);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "Q", qid);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ goto free_and_error;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RSYMLINK qid %x.%llx.%x\n",
+ qid->type, (unsigned long long)qid->path, qid->version);
+
+free_and_error:
+ p9_free_req(clnt, req);
+error:
+ return err;
+}
+EXPORT_SYMBOL(p9_client_symlink);
+
+int p9_client_link(struct p9_fid *dfid, struct p9_fid *oldfid, char *newname)
+{
+ struct p9_client *clnt;
+ struct p9_req_t *req;
+
+ p9_debug(P9_DEBUG_9P, ">>> TLINK dfid %d oldfid %d newname %s\n",
+ dfid->fid, oldfid->fid, newname);
+ clnt = dfid->clnt;
+ req = p9_client_rpc(clnt, P9_TLINK, "dds", dfid->fid, oldfid->fid,
+ newname);
+ if (IS_ERR(req))
+ return PTR_ERR(req);
+
+ p9_debug(P9_DEBUG_9P, "<<< RLINK\n");
+ p9_free_req(clnt, req);
+ return 0;
+}
+EXPORT_SYMBOL(p9_client_link);
+
+int p9_client_fsync(struct p9_fid *fid, int datasync)
+{
+ int err;
+ struct p9_client *clnt;
+ struct p9_req_t *req;
+
+ p9_debug(P9_DEBUG_9P, ">>> TFSYNC fid %d datasync:%d\n",
+ fid->fid, datasync);
+ err = 0;
+ clnt = fid->clnt;
+
+ req = p9_client_rpc(clnt, P9_TFSYNC, "dd", fid->fid, datasync);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RFSYNC fid %d\n", fid->fid);
+
+ p9_free_req(clnt, req);
+
+error:
+ return err;
+}
+EXPORT_SYMBOL(p9_client_fsync);
+
+int p9_client_clunk(struct p9_fid *fid)
+{
+ int err;
+ struct p9_client *clnt;
+ struct p9_req_t *req;
+ int retries = 0;
+
+ if (!fid) {
+ pr_warn("%s (%d): Trying to clunk with NULL fid\n",
+ __func__, task_pid_nr(current));
+ dump_stack();
+ return 0;
+ }
+
+again:
+ p9_debug(P9_DEBUG_9P, ">>> TCLUNK fid %d (try %d)\n", fid->fid,
+ retries);
+ err = 0;
+ clnt = fid->clnt;
+
+ req = p9_client_rpc(clnt, P9_TCLUNK, "d", fid->fid);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RCLUNK fid %d\n", fid->fid);
+
+ p9_free_req(clnt, req);
+error:
+ /*
+ * Fid is not valid even after a failed clunk
+ * If interrupted, retry once then give up and
+ * leak fid until umount.
+ */
+ if (err == -ERESTARTSYS) {
+ if (retries++ == 0)
+ goto again;
+ } else
+ p9_fid_destroy(fid);
+ return err;
+}
+EXPORT_SYMBOL(p9_client_clunk);
+
+int p9_client_remove(struct p9_fid *fid)
+{
+ int err;
+ struct p9_client *clnt;
+ struct p9_req_t *req;
+
+ p9_debug(P9_DEBUG_9P, ">>> TREMOVE fid %d\n", fid->fid);
+ err = 0;
+ clnt = fid->clnt;
+
+ req = p9_client_rpc(clnt, P9_TREMOVE, "d", fid->fid);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RREMOVE fid %d\n", fid->fid);
+
+ p9_free_req(clnt, req);
+error:
+ if (err == -ERESTARTSYS)
+ p9_client_clunk(fid);
+ else
+ p9_fid_destroy(fid);
+ return err;
+}
+EXPORT_SYMBOL(p9_client_remove);
+
+int p9_client_unlinkat(struct p9_fid *dfid, const char *name, int flags)
+{
+ int err = 0;
+ struct p9_req_t *req;
+ struct p9_client *clnt;
+
+ p9_debug(P9_DEBUG_9P, ">>> TUNLINKAT fid %d %s %d\n",
+ dfid->fid, name, flags);
+
+ clnt = dfid->clnt;
+ req = p9_client_rpc(clnt, P9_TUNLINKAT, "dsd", dfid->fid, name, flags);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+ p9_debug(P9_DEBUG_9P, "<<< RUNLINKAT fid %d %s\n", dfid->fid, name);
+
+ p9_free_req(clnt, req);
+error:
+ return err;
+}
+EXPORT_SYMBOL(p9_client_unlinkat);
+
+int
+p9_client_read(struct p9_fid *fid, u64 offset, struct iov_iter *to, int *err)
+{
+ struct p9_client *clnt = fid->clnt;
+ struct p9_req_t *req;
+ int total = 0;
+
+ p9_debug(P9_DEBUG_9P, ">>> TREAD fid %d offset %llu %d\n",
+ fid->fid, (unsigned long long) offset, (int)iov_iter_count(to));
+
+ while (iov_iter_count(to)) {
+ int count = iov_iter_count(to);
+ int rsize, non_zc = 0;
+ char *dataptr;
+
+ rsize = fid->iounit;
+ if (!rsize || rsize > clnt->msize-P9_IOHDRSZ)
+ rsize = clnt->msize - P9_IOHDRSZ;
+
+ if (count < rsize)
+ rsize = count;
+
+ /* 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_TREAD, to, NULL, rsize,
+ 0, 11, "dqd", fid->fid,
+ offset, rsize);
+ } else {
+ non_zc = 1;
+ req = p9_client_rpc(clnt, P9_TREAD, "dqd", fid->fid, offset,
+ rsize);
+ }
+ if (IS_ERR(req)) {
+ *err = PTR_ERR(req);
+ break;
+ }
+
+ *err = p9pdu_readf(req->rc, clnt->proto_version,
+ "D", &count, &dataptr);
+ if (*err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ p9_free_req(clnt, req);
+ break;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RREAD count %d\n", count);
+ if (!count) {
+ p9_free_req(clnt, req);
+ break;
+ }
+
+ if (non_zc) {
+ int n = copy_to_iter(dataptr, count, to);
+ total += n;
+ offset += n;
+ if (n != count) {
+ *err = -EFAULT;
+ p9_free_req(clnt, req);
+ break;
+ }
+ } else {
+ iov_iter_advance(to, count);
+ total += count;
+ offset += count;
+ }
+ p9_free_req(clnt, req);
+ }
+ return total;
+}
+EXPORT_SYMBOL(p9_client_read);
+
+int
+p9_client_write(struct p9_fid *fid, u64 offset, struct iov_iter *from, int *err)
+{
+ struct p9_client *clnt = fid->clnt;
+ struct p9_req_t *req;
+ int total = 0;
+
+ p9_debug(P9_DEBUG_9P, ">>> TWRITE fid %d offset %llu count %zd\n",
+ fid->fid, (unsigned long long) offset,
+ iov_iter_count(from));
+
+ while (iov_iter_count(from)) {
+ int count = iov_iter_count(from);
+ int rsize = fid->iounit;
+ if (!rsize || rsize > clnt->msize-P9_IOHDRSZ)
+ rsize = clnt->msize - P9_IOHDRSZ;
+
+ if (count < rsize)
+ rsize = count;
+
+ /* Don't bother zerocopy for small IO (< 1024) */
+ if (clnt->trans_mod->zc_request && rsize > 1024) {
+ req = p9_client_zc_rpc(clnt, P9_TWRITE, NULL, from, 0,
+ rsize, P9_ZC_HDR_SZ, "dqd",
+ fid->fid, offset, rsize);
+ } else {
+ req = p9_client_rpc(clnt, P9_TWRITE, "dqV", fid->fid,
+ offset, rsize, from);
+ }
+ if (IS_ERR(req)) {
+ *err = PTR_ERR(req);
+ break;
+ }
+
+ *err = p9pdu_readf(req->rc, clnt->proto_version, "d", &count);
+ if (*err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ p9_free_req(clnt, req);
+ break;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RWRITE count %d\n", count);
+
+ p9_free_req(clnt, req);
+ iov_iter_advance(from, count);
+ total += count;
+ offset += count;
+ }
+ return total;
+}
+EXPORT_SYMBOL(p9_client_write);
+
+struct p9_wstat *p9_client_stat(struct p9_fid *fid)
+{
+ int err;
+ struct p9_client *clnt;
+ struct p9_wstat *ret = kmalloc(sizeof(struct p9_wstat), GFP_KERNEL);
+ struct p9_req_t *req;
+ u16 ignored;
+
+ p9_debug(P9_DEBUG_9P, ">>> TSTAT fid %d\n", fid->fid);
+
+ if (!ret)
+ return ERR_PTR(-ENOMEM);
+
+ err = 0;
+ clnt = fid->clnt;
+
+ req = p9_client_rpc(clnt, P9_TSTAT, "d", fid->fid);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "wS", &ignored, ret);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ p9_free_req(clnt, req);
+ goto error;
+ }
+
+ p9_debug(P9_DEBUG_9P,
+ "<<< RSTAT sz=%x type=%x dev=%x qid=%x.%llx.%x\n"
+ "<<< mode=%8.8x atime=%8.8x mtime=%8.8x length=%llx\n"
+ "<<< name=%s uid=%s gid=%s muid=%s extension=(%s)\n"
+ "<<< uid=%d gid=%d n_muid=%d\n",
+ ret->size, ret->type, ret->dev, ret->qid.type,
+ (unsigned long long)ret->qid.path, ret->qid.version, ret->mode,
+ ret->atime, ret->mtime, (unsigned long long)ret->length,
+ ret->name, ret->uid, ret->gid, ret->muid, ret->extension,
+ from_kuid(&init_user_ns, ret->n_uid),
+ from_kgid(&init_user_ns, ret->n_gid),
+ from_kuid(&init_user_ns, ret->n_muid));
+
+ p9_free_req(clnt, req);
+ return ret;
+
+error:
+ kfree(ret);
+ return ERR_PTR(err);
+}
+EXPORT_SYMBOL(p9_client_stat);
+
+struct p9_stat_dotl *p9_client_getattr_dotl(struct p9_fid *fid,
+ u64 request_mask)
+{
+ int err;
+ struct p9_client *clnt;
+ struct p9_stat_dotl *ret = kmalloc(sizeof(struct p9_stat_dotl),
+ GFP_KERNEL);
+ struct p9_req_t *req;
+
+ p9_debug(P9_DEBUG_9P, ">>> TGETATTR fid %d, request_mask %lld\n",
+ fid->fid, request_mask);
+
+ if (!ret)
+ return ERR_PTR(-ENOMEM);
+
+ err = 0;
+ clnt = fid->clnt;
+
+ req = p9_client_rpc(clnt, P9_TGETATTR, "dq", fid->fid, request_mask);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "A", ret);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ p9_free_req(clnt, req);
+ goto error;
+ }
+
+ p9_debug(P9_DEBUG_9P,
+ "<<< RGETATTR st_result_mask=%lld\n"
+ "<<< qid=%x.%llx.%x\n"
+ "<<< st_mode=%8.8x st_nlink=%llu\n"
+ "<<< st_uid=%d st_gid=%d\n"
+ "<<< st_rdev=%llx st_size=%llx st_blksize=%llu st_blocks=%llu\n"
+ "<<< st_atime_sec=%lld st_atime_nsec=%lld\n"
+ "<<< st_mtime_sec=%lld st_mtime_nsec=%lld\n"
+ "<<< st_ctime_sec=%lld st_ctime_nsec=%lld\n"
+ "<<< st_btime_sec=%lld st_btime_nsec=%lld\n"
+ "<<< st_gen=%lld st_data_version=%lld",
+ ret->st_result_mask, ret->qid.type, ret->qid.path,
+ ret->qid.version, ret->st_mode, ret->st_nlink,
+ from_kuid(&init_user_ns, ret->st_uid),
+ from_kgid(&init_user_ns, ret->st_gid),
+ ret->st_rdev, ret->st_size, ret->st_blksize,
+ ret->st_blocks, ret->st_atime_sec, ret->st_atime_nsec,
+ ret->st_mtime_sec, ret->st_mtime_nsec, ret->st_ctime_sec,
+ ret->st_ctime_nsec, ret->st_btime_sec, ret->st_btime_nsec,
+ ret->st_gen, ret->st_data_version);
+
+ p9_free_req(clnt, req);
+ return ret;
+
+error:
+ kfree(ret);
+ return ERR_PTR(err);
+}
+EXPORT_SYMBOL(p9_client_getattr_dotl);
+
+static int p9_client_statsize(struct p9_wstat *wst, int proto_version)
+{
+ int ret;
+
+ /* NOTE: size shouldn't include its own length */
+ /* size[2] type[2] dev[4] qid[13] */
+ /* mode[4] atime[4] mtime[4] length[8]*/
+ /* name[s] uid[s] gid[s] muid[s] */
+ ret = 2+4+13+4+4+4+8+2+2+2+2;
+
+ if (wst->name)
+ ret += strlen(wst->name);
+ if (wst->uid)
+ ret += strlen(wst->uid);
+ if (wst->gid)
+ ret += strlen(wst->gid);
+ if (wst->muid)
+ ret += strlen(wst->muid);
+
+ if ((proto_version == p9_proto_2000u) ||
+ (proto_version == p9_proto_2000L)) {
+ ret += 2+4+4+4; /* extension[s] n_uid[4] n_gid[4] n_muid[4] */
+ if (wst->extension)
+ ret += strlen(wst->extension);
+ }
+
+ return ret;
+}
+
+int p9_client_wstat(struct p9_fid *fid, struct p9_wstat *wst)
+{
+ int err;
+ struct p9_req_t *req;
+ struct p9_client *clnt;
+
+ err = 0;
+ clnt = fid->clnt;
+ wst->size = p9_client_statsize(wst, clnt->proto_version);
+ p9_debug(P9_DEBUG_9P, ">>> TWSTAT fid %d\n", fid->fid);
+ p9_debug(P9_DEBUG_9P,
+ " sz=%x type=%x dev=%x qid=%x.%llx.%x\n"
+ " mode=%8.8x atime=%8.8x mtime=%8.8x length=%llx\n"
+ " name=%s uid=%s gid=%s muid=%s extension=(%s)\n"
+ " uid=%d gid=%d n_muid=%d\n",
+ wst->size, wst->type, wst->dev, wst->qid.type,
+ (unsigned long long)wst->qid.path, wst->qid.version, wst->mode,
+ wst->atime, wst->mtime, (unsigned long long)wst->length,
+ wst->name, wst->uid, wst->gid, wst->muid, wst->extension,
+ from_kuid(&init_user_ns, wst->n_uid),
+ from_kgid(&init_user_ns, wst->n_gid),
+ from_kuid(&init_user_ns, wst->n_muid));
+
+ req = p9_client_rpc(clnt, P9_TWSTAT, "dwS", fid->fid, wst->size+2, wst);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RWSTAT fid %d\n", fid->fid);
+
+ p9_free_req(clnt, req);
+error:
+ return err;
+}
+EXPORT_SYMBOL(p9_client_wstat);
+
+int p9_client_setattr(struct p9_fid *fid, struct p9_iattr_dotl *p9attr)
+{
+ int err;
+ struct p9_req_t *req;
+ struct p9_client *clnt;
+
+ err = 0;
+ clnt = fid->clnt;
+ p9_debug(P9_DEBUG_9P, ">>> TSETATTR fid %d\n", fid->fid);
+ p9_debug(P9_DEBUG_9P,
+ " valid=%x mode=%x uid=%d gid=%d size=%lld\n"
+ " atime_sec=%lld atime_nsec=%lld\n"
+ " mtime_sec=%lld mtime_nsec=%lld\n",
+ p9attr->valid, p9attr->mode,
+ from_kuid(&init_user_ns, p9attr->uid),
+ from_kgid(&init_user_ns, p9attr->gid),
+ p9attr->size, p9attr->atime_sec, p9attr->atime_nsec,
+ p9attr->mtime_sec, p9attr->mtime_nsec);
+
+ req = p9_client_rpc(clnt, P9_TSETATTR, "dI", fid->fid, p9attr);
+
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+ p9_debug(P9_DEBUG_9P, "<<< RSETATTR fid %d\n", fid->fid);
+ p9_free_req(clnt, req);
+error:
+ return err;
+}
+EXPORT_SYMBOL(p9_client_setattr);
+
+int p9_client_statfs(struct p9_fid *fid, struct p9_rstatfs *sb)
+{
+ int err;
+ struct p9_req_t *req;
+ struct p9_client *clnt;
+
+ err = 0;
+ clnt = fid->clnt;
+
+ p9_debug(P9_DEBUG_9P, ">>> TSTATFS fid %d\n", fid->fid);
+
+ req = p9_client_rpc(clnt, P9_TSTATFS, "d", fid->fid);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "ddqqqqqqd", &sb->type,
+ &sb->bsize, &sb->blocks, &sb->bfree, &sb->bavail,
+ &sb->files, &sb->ffree, &sb->fsid, &sb->namelen);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ p9_free_req(clnt, req);
+ goto error;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RSTATFS fid %d type 0x%lx bsize %ld "
+ "blocks %llu bfree %llu bavail %llu files %llu ffree %llu "
+ "fsid %llu namelen %ld\n",
+ fid->fid, (long unsigned int)sb->type, (long int)sb->bsize,
+ sb->blocks, sb->bfree, sb->bavail, sb->files, sb->ffree,
+ sb->fsid, (long int)sb->namelen);
+
+ p9_free_req(clnt, req);
+error:
+ return err;
+}
+EXPORT_SYMBOL(p9_client_statfs);
+
+int p9_client_rename(struct p9_fid *fid,
+ struct p9_fid *newdirfid, const char *name)
+{
+ int err;
+ struct p9_req_t *req;
+ struct p9_client *clnt;
+
+ err = 0;
+ clnt = fid->clnt;
+
+ p9_debug(P9_DEBUG_9P, ">>> TRENAME fid %d newdirfid %d name %s\n",
+ fid->fid, newdirfid->fid, name);
+
+ req = p9_client_rpc(clnt, P9_TRENAME, "dds", fid->fid,
+ newdirfid->fid, name);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RRENAME fid %d\n", fid->fid);
+
+ p9_free_req(clnt, req);
+error:
+ return err;
+}
+EXPORT_SYMBOL(p9_client_rename);
+
+int p9_client_renameat(struct p9_fid *olddirfid, const char *old_name,
+ struct p9_fid *newdirfid, const char *new_name)
+{
+ int err;
+ struct p9_req_t *req;
+ struct p9_client *clnt;
+
+ err = 0;
+ clnt = olddirfid->clnt;
+
+ p9_debug(P9_DEBUG_9P, ">>> TRENAMEAT olddirfid %d old name %s"
+ " newdirfid %d new name %s\n", olddirfid->fid, old_name,
+ newdirfid->fid, new_name);
+
+ req = p9_client_rpc(clnt, P9_TRENAMEAT, "dsds", olddirfid->fid,
+ old_name, newdirfid->fid, new_name);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RRENAMEAT newdirfid %d new name %s\n",
+ newdirfid->fid, new_name);
+
+ p9_free_req(clnt, req);
+error:
+ return err;
+}
+EXPORT_SYMBOL(p9_client_renameat);
+
+/*
+ * An xattrwalk without @attr_name gives the fid for the lisxattr namespace
+ */
+struct p9_fid *p9_client_xattrwalk(struct p9_fid *file_fid,
+ const char *attr_name, u64 *attr_size)
+{
+ int err;
+ struct p9_req_t *req;
+ struct p9_client *clnt;
+ struct p9_fid *attr_fid;
+
+ err = 0;
+ clnt = file_fid->clnt;
+ attr_fid = p9_fid_create(clnt);
+ if (IS_ERR(attr_fid)) {
+ err = PTR_ERR(attr_fid);
+ attr_fid = NULL;
+ goto error;
+ }
+ p9_debug(P9_DEBUG_9P,
+ ">>> TXATTRWALK file_fid %d, attr_fid %d name %s\n",
+ file_fid->fid, attr_fid->fid, attr_name);
+
+ req = p9_client_rpc(clnt, P9_TXATTRWALK, "dds",
+ file_fid->fid, attr_fid->fid, attr_name);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+ err = p9pdu_readf(req->rc, clnt->proto_version, "q", attr_size);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ p9_free_req(clnt, req);
+ goto clunk_fid;
+ }
+ p9_free_req(clnt, req);
+ p9_debug(P9_DEBUG_9P, "<<< RXATTRWALK fid %d size %llu\n",
+ attr_fid->fid, *attr_size);
+ return attr_fid;
+clunk_fid:
+ p9_client_clunk(attr_fid);
+ attr_fid = NULL;
+error:
+ if (attr_fid && (attr_fid != file_fid))
+ p9_fid_destroy(attr_fid);
+
+ return ERR_PTR(err);
+}
+EXPORT_SYMBOL_GPL(p9_client_xattrwalk);
+
+int p9_client_xattrcreate(struct p9_fid *fid, const char *name,
+ u64 attr_size, int flags)
+{
+ int err;
+ struct p9_req_t *req;
+ struct p9_client *clnt;
+
+ p9_debug(P9_DEBUG_9P,
+ ">>> TXATTRCREATE fid %d name %s size %lld flag %d\n",
+ fid->fid, name, (long long)attr_size, flags);
+ err = 0;
+ clnt = fid->clnt;
+ req = p9_client_rpc(clnt, P9_TXATTRCREATE, "dsqd",
+ fid->fid, name, attr_size, flags);
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+ p9_debug(P9_DEBUG_9P, "<<< RXATTRCREATE fid %d\n", fid->fid);
+ p9_free_req(clnt, req);
+error:
+ return err;
+}
+EXPORT_SYMBOL_GPL(p9_client_xattrcreate);
+
+int p9_client_readdir(struct p9_fid *fid, char *data, u32 count, u64 offset)
+{
+ int err, rsize, non_zc = 0;
+ struct p9_client *clnt;
+ struct p9_req_t *req;
+ char *dataptr;
+ struct kvec kv = {.iov_base = data, .iov_len = count};
+ struct iov_iter to;
+
+ iov_iter_kvec(&to, READ | ITER_KVEC, &kv, 1, count);
+
+ p9_debug(P9_DEBUG_9P, ">>> TREADDIR fid %d offset %llu count %d\n",
+ fid->fid, (unsigned long long) offset, count);
+
+ err = 0;
+ clnt = fid->clnt;
+
+ rsize = fid->iounit;
+ if (!rsize || rsize > clnt->msize-P9_READDIRHDRSZ)
+ rsize = clnt->msize - P9_READDIRHDRSZ;
+
+ if (count < rsize)
+ rsize = count;
+
+ /* 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, &to, NULL, rsize, 0,
+ 11, "dqd", fid->fid, offset, rsize);
+ } else {
+ non_zc = 1;
+ req = p9_client_rpc(clnt, P9_TREADDIR, "dqd", fid->fid,
+ offset, rsize);
+ }
+ if (IS_ERR(req)) {
+ err = PTR_ERR(req);
+ goto error;
+ }
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "D", &count, &dataptr);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ goto free_and_error;
+ }
+
+ p9_debug(P9_DEBUG_9P, "<<< RREADDIR count %d\n", count);
+
+ if (non_zc)
+ memmove(data, dataptr, count);
+
+ p9_free_req(clnt, req);
+ return count;
+
+free_and_error:
+ p9_free_req(clnt, req);
+error:
+ return err;
+}
+EXPORT_SYMBOL(p9_client_readdir);
+
+int p9_client_mknod_dotl(struct p9_fid *fid, char *name, int mode,
+ dev_t rdev, kgid_t gid, struct p9_qid *qid)
+{
+ int err;
+ struct p9_client *clnt;
+ struct p9_req_t *req;
+
+ err = 0;
+ clnt = fid->clnt;
+ p9_debug(P9_DEBUG_9P, ">>> TMKNOD fid %d name %s mode %d major %d "
+ "minor %d\n", fid->fid, name, mode, MAJOR(rdev), MINOR(rdev));
+ req = p9_client_rpc(clnt, P9_TMKNOD, "dsdddg", fid->fid, name, mode,
+ MAJOR(rdev), MINOR(rdev), gid);
+ if (IS_ERR(req))
+ return PTR_ERR(req);
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "Q", qid);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ goto error;
+ }
+ p9_debug(P9_DEBUG_9P, "<<< RMKNOD qid %x.%llx.%x\n", qid->type,
+ (unsigned long long)qid->path, qid->version);
+
+error:
+ p9_free_req(clnt, req);
+ return err;
+
+}
+EXPORT_SYMBOL(p9_client_mknod_dotl);
+
+int p9_client_mkdir_dotl(struct p9_fid *fid, char *name, int mode,
+ kgid_t gid, struct p9_qid *qid)
+{
+ int err;
+ struct p9_client *clnt;
+ struct p9_req_t *req;
+
+ err = 0;
+ clnt = fid->clnt;
+ p9_debug(P9_DEBUG_9P, ">>> TMKDIR fid %d name %s mode %d gid %d\n",
+ fid->fid, name, mode, from_kgid(&init_user_ns, gid));
+ req = p9_client_rpc(clnt, P9_TMKDIR, "dsdg", fid->fid, name, mode,
+ gid);
+ if (IS_ERR(req))
+ return PTR_ERR(req);
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "Q", qid);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ goto error;
+ }
+ p9_debug(P9_DEBUG_9P, "<<< RMKDIR qid %x.%llx.%x\n", qid->type,
+ (unsigned long long)qid->path, qid->version);
+
+error:
+ p9_free_req(clnt, req);
+ return err;
+
+}
+EXPORT_SYMBOL(p9_client_mkdir_dotl);
+
+int p9_client_lock_dotl(struct p9_fid *fid, struct p9_flock *flock, u8 *status)
+{
+ int err;
+ struct p9_client *clnt;
+ struct p9_req_t *req;
+
+ err = 0;
+ clnt = fid->clnt;
+ p9_debug(P9_DEBUG_9P, ">>> TLOCK fid %d type %i flags %d "
+ "start %lld length %lld proc_id %d client_id %s\n",
+ fid->fid, flock->type, flock->flags, flock->start,
+ flock->length, flock->proc_id, flock->client_id);
+
+ req = p9_client_rpc(clnt, P9_TLOCK, "dbdqqds", fid->fid, flock->type,
+ flock->flags, flock->start, flock->length,
+ flock->proc_id, flock->client_id);
+
+ if (IS_ERR(req))
+ return PTR_ERR(req);
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "b", status);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ goto error;
+ }
+ p9_debug(P9_DEBUG_9P, "<<< RLOCK status %i\n", *status);
+error:
+ p9_free_req(clnt, req);
+ return err;
+
+}
+EXPORT_SYMBOL(p9_client_lock_dotl);
+
+int p9_client_getlock_dotl(struct p9_fid *fid, struct p9_getlock *glock)
+{
+ int err;
+ struct p9_client *clnt;
+ struct p9_req_t *req;
+
+ err = 0;
+ clnt = fid->clnt;
+ p9_debug(P9_DEBUG_9P, ">>> TGETLOCK fid %d, type %i start %lld "
+ "length %lld proc_id %d client_id %s\n", fid->fid, glock->type,
+ glock->start, glock->length, glock->proc_id, glock->client_id);
+
+ req = p9_client_rpc(clnt, P9_TGETLOCK, "dbqqds", fid->fid, glock->type,
+ glock->start, glock->length, glock->proc_id, glock->client_id);
+
+ if (IS_ERR(req))
+ return PTR_ERR(req);
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "bqqds", &glock->type,
+ &glock->start, &glock->length, &glock->proc_id,
+ &glock->client_id);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ goto error;
+ }
+ p9_debug(P9_DEBUG_9P, "<<< RGETLOCK type %i start %lld length %lld "
+ "proc_id %d client_id %s\n", glock->type, glock->start,
+ glock->length, glock->proc_id, glock->client_id);
+error:
+ p9_free_req(clnt, req);
+ return err;
+}
+EXPORT_SYMBOL(p9_client_getlock_dotl);
+
+int p9_client_readlink(struct p9_fid *fid, char **target)
+{
+ int err;
+ struct p9_client *clnt;
+ struct p9_req_t *req;
+
+ err = 0;
+ clnt = fid->clnt;
+ p9_debug(P9_DEBUG_9P, ">>> TREADLINK fid %d\n", fid->fid);
+
+ req = p9_client_rpc(clnt, P9_TREADLINK, "d", fid->fid);
+ if (IS_ERR(req))
+ return PTR_ERR(req);
+
+ err = p9pdu_readf(req->rc, clnt->proto_version, "s", target);
+ if (err) {
+ trace_9p_protocol_dump(clnt, req->rc);
+ goto error;
+ }
+ p9_debug(P9_DEBUG_9P, "<<< RREADLINK target %s\n", *target);
+error:
+ p9_free_req(clnt, req);
+ return err;
+}
+EXPORT_SYMBOL(p9_client_readlink);
diff --git a/net/9p/error.c b/net/9p/error.c
new file mode 100644
index 000000000..126fd0dce
--- /dev/null
+++ b/net/9p/error.c
@@ -0,0 +1,247 @@
+/*
+ * linux/fs/9p/error.c
+ *
+ * Error string handling
+ *
+ * Plan 9 uses error strings, Unix uses error numbers. These functions
+ * try to help manage that and provide for dynamically adding error
+ * mappings.
+ *
+ * Copyright (C) 2004 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
+ * 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
+ *
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/module.h>
+#include <linux/list.h>
+#include <linux/jhash.h>
+#include <linux/errno.h>
+#include <net/9p/9p.h>
+
+/**
+ * struct errormap - map string errors from Plan 9 to Linux numeric ids
+ * @name: string sent over 9P
+ * @val: numeric id most closely representing @name
+ * @namelen: length of string
+ * @list: hash-table list for string lookup
+ */
+struct errormap {
+ char *name;
+ int val;
+
+ int namelen;
+ struct hlist_node list;
+};
+
+#define ERRHASHSZ 32
+static struct hlist_head hash_errmap[ERRHASHSZ];
+
+/* FixMe - reduce to a reasonable size */
+static struct errormap errmap[] = {
+ {"Operation not permitted", EPERM},
+ {"wstat prohibited", EPERM},
+ {"No such file or directory", ENOENT},
+ {"directory entry not found", ENOENT},
+ {"file not found", ENOENT},
+ {"Interrupted system call", EINTR},
+ {"Input/output error", EIO},
+ {"No such device or address", ENXIO},
+ {"Argument list too long", E2BIG},
+ {"Bad file descriptor", EBADF},
+ {"Resource temporarily unavailable", EAGAIN},
+ {"Cannot allocate memory", ENOMEM},
+ {"Permission denied", EACCES},
+ {"Bad address", EFAULT},
+ {"Block device required", ENOTBLK},
+ {"Device or resource busy", EBUSY},
+ {"File exists", EEXIST},
+ {"Invalid cross-device link", EXDEV},
+ {"No such device", ENODEV},
+ {"Not a directory", ENOTDIR},
+ {"Is a directory", EISDIR},
+ {"Invalid argument", EINVAL},
+ {"Too many open files in system", ENFILE},
+ {"Too many open files", EMFILE},
+ {"Text file busy", ETXTBSY},
+ {"File too large", EFBIG},
+ {"No space left on device", ENOSPC},
+ {"Illegal seek", ESPIPE},
+ {"Read-only file system", EROFS},
+ {"Too many links", EMLINK},
+ {"Broken pipe", EPIPE},
+ {"Numerical argument out of domain", EDOM},
+ {"Numerical result out of range", ERANGE},
+ {"Resource deadlock avoided", EDEADLK},
+ {"File name too long", ENAMETOOLONG},
+ {"No locks available", ENOLCK},
+ {"Function not implemented", ENOSYS},
+ {"Directory not empty", ENOTEMPTY},
+ {"Too many levels of symbolic links", ELOOP},
+ {"No message of desired type", ENOMSG},
+ {"Identifier removed", EIDRM},
+ {"No data available", ENODATA},
+ {"Machine is not on the network", ENONET},
+ {"Package not installed", ENOPKG},
+ {"Object is remote", EREMOTE},
+ {"Link has been severed", ENOLINK},
+ {"Communication error on send", ECOMM},
+ {"Protocol error", EPROTO},
+ {"Bad message", EBADMSG},
+ {"File descriptor in bad state", EBADFD},
+ {"Streams pipe error", ESTRPIPE},
+ {"Too many users", EUSERS},
+ {"Socket operation on non-socket", ENOTSOCK},
+ {"Message too long", EMSGSIZE},
+ {"Protocol not available", ENOPROTOOPT},
+ {"Protocol not supported", EPROTONOSUPPORT},
+ {"Socket type not supported", ESOCKTNOSUPPORT},
+ {"Operation not supported", EOPNOTSUPP},
+ {"Protocol family not supported", EPFNOSUPPORT},
+ {"Network is down", ENETDOWN},
+ {"Network is unreachable", ENETUNREACH},
+ {"Network dropped connection on reset", ENETRESET},
+ {"Software caused connection abort", ECONNABORTED},
+ {"Connection reset by peer", ECONNRESET},
+ {"No buffer space available", ENOBUFS},
+ {"Transport endpoint is already connected", EISCONN},
+ {"Transport endpoint is not connected", ENOTCONN},
+ {"Cannot send after transport endpoint shutdown", ESHUTDOWN},
+ {"Connection timed out", ETIMEDOUT},
+ {"Connection refused", ECONNREFUSED},
+ {"Host is down", EHOSTDOWN},
+ {"No route to host", EHOSTUNREACH},
+ {"Operation already in progress", EALREADY},
+ {"Operation now in progress", EINPROGRESS},
+ {"Is a named type file", EISNAM},
+ {"Remote I/O error", EREMOTEIO},
+ {"Disk quota exceeded", EDQUOT},
+/* errors from fossil, vacfs, and u9fs */
+ {"fid unknown or out of range", EBADF},
+ {"permission denied", EACCES},
+ {"file does not exist", ENOENT},
+ {"authentication failed", ECONNREFUSED},
+ {"bad offset in directory read", ESPIPE},
+ {"bad use of fid", EBADF},
+ {"wstat can't convert between files and directories", EPERM},
+ {"directory is not empty", ENOTEMPTY},
+ {"file exists", EEXIST},
+ {"file already exists", EEXIST},
+ {"file or directory already exists", EEXIST},
+ {"fid already in use", EBADF},
+ {"file in use", ETXTBSY},
+ {"i/o error", EIO},
+ {"file already open for I/O", ETXTBSY},
+ {"illegal mode", EINVAL},
+ {"illegal name", ENAMETOOLONG},
+ {"not a directory", ENOTDIR},
+ {"not a member of proposed group", EPERM},
+ {"not owner", EACCES},
+ {"only owner can change group in wstat", EACCES},
+ {"read only file system", EROFS},
+ {"no access to special file", EPERM},
+ {"i/o count too large", EIO},
+ {"unknown group", EINVAL},
+ {"unknown user", EINVAL},
+ {"bogus wstat buffer", EPROTO},
+ {"exclusive use file already open", EAGAIN},
+ {"corrupted directory entry", EIO},
+ {"corrupted file entry", EIO},
+ {"corrupted block label", EIO},
+ {"corrupted meta data", EIO},
+ {"illegal offset", EINVAL},
+ {"illegal path element", ENOENT},
+ {"root of file system is corrupted", EIO},
+ {"corrupted super block", EIO},
+ {"protocol botch", EPROTO},
+ {"file system is full", ENOSPC},
+ {"file is in use", EAGAIN},
+ {"directory entry is not allocated", ENOENT},
+ {"file is read only", EROFS},
+ {"file has been removed", EIDRM},
+ {"only support truncation to zero length", EPERM},
+ {"cannot remove root", EPERM},
+ {"file too big", EFBIG},
+ {"venti i/o error", EIO},
+ /* these are not errors */
+ {"u9fs rhostsauth: no authentication required", 0},
+ {"u9fs authnone: no authentication required", 0},
+ {NULL, -1}
+};
+
+/**
+ * p9_error_init - preload mappings into hash list
+ *
+ */
+
+int p9_error_init(void)
+{
+ struct errormap *c;
+ int bucket;
+
+ /* initialize hash table */
+ for (bucket = 0; bucket < ERRHASHSZ; bucket++)
+ INIT_HLIST_HEAD(&hash_errmap[bucket]);
+
+ /* load initial error map into hash table */
+ for (c = errmap; c->name != NULL; c++) {
+ c->namelen = strlen(c->name);
+ bucket = jhash(c->name, c->namelen, 0) % ERRHASHSZ;
+ INIT_HLIST_NODE(&c->list);
+ hlist_add_head(&c->list, &hash_errmap[bucket]);
+ }
+
+ return 1;
+}
+EXPORT_SYMBOL(p9_error_init);
+
+/**
+ * errstr2errno - convert error string to error number
+ * @errstr: error string
+ * @len: length of error string
+ *
+ */
+
+int p9_errstr2errno(char *errstr, int len)
+{
+ int errno;
+ struct errormap *c;
+ int bucket;
+
+ errno = 0;
+ c = NULL;
+ bucket = jhash(errstr, len, 0) % ERRHASHSZ;
+ hlist_for_each_entry(c, &hash_errmap[bucket], list) {
+ if (c->namelen == len && !memcmp(c->name, errstr, len)) {
+ errno = c->val;
+ break;
+ }
+ }
+
+ if (errno == 0) {
+ /* TODO: if error isn't found, add it dynamically */
+ errstr[len] = 0;
+ pr_err("%s: server reported unknown error %s\n",
+ __func__, errstr);
+ errno = ESERVERFAULT;
+ }
+
+ return -errno;
+}
+EXPORT_SYMBOL(p9_errstr2errno);
diff --git a/net/9p/mod.c b/net/9p/mod.c
new file mode 100644
index 000000000..6ab36aea7
--- /dev/null
+++ b/net/9p/mod.c
@@ -0,0 +1,201 @@
+/*
+ * net/9p/9p.c
+ *
+ * 9P entry point
+ *
+ * Copyright (C) 2007 by Latchesar Ionkov <lucho@ionkov.net>
+ * Copyright (C) 2004 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
+ * 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
+ *
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/sched.h>
+#include <linux/moduleparam.h>
+#include <net/9p/9p.h>
+#include <linux/fs.h>
+#include <linux/parser.h>
+#include <net/9p/client.h>
+#include <net/9p/transport.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+
+#ifdef CONFIG_NET_9P_DEBUG
+unsigned int p9_debug_level = 0; /* feature-rific global debug level */
+EXPORT_SYMBOL(p9_debug_level);
+module_param_named(debug, p9_debug_level, uint, 0);
+MODULE_PARM_DESC(debug, "9P debugging level");
+
+void _p9_debug(enum p9_debug_flags level, const char *func,
+ const char *fmt, ...)
+{
+ struct va_format vaf;
+ va_list args;
+
+ if ((p9_debug_level & level) != level)
+ return;
+
+ va_start(args, fmt);
+
+ vaf.fmt = fmt;
+ vaf.va = &args;
+
+ if (level == P9_DEBUG_9P)
+ pr_notice("(%8.8d) %pV", task_pid_nr(current), &vaf);
+ else
+ pr_notice("-- %s (%d): %pV", func, task_pid_nr(current), &vaf);
+
+ va_end(args);
+}
+EXPORT_SYMBOL(_p9_debug);
+#endif
+
+/*
+ * Dynamic Transport Registration Routines
+ *
+ */
+
+static DEFINE_SPINLOCK(v9fs_trans_lock);
+static LIST_HEAD(v9fs_trans_list);
+
+/**
+ * v9fs_register_trans - register a new transport with 9p
+ * @m: structure describing the transport module and entry points
+ *
+ */
+void v9fs_register_trans(struct p9_trans_module *m)
+{
+ spin_lock(&v9fs_trans_lock);
+ list_add_tail(&m->list, &v9fs_trans_list);
+ spin_unlock(&v9fs_trans_lock);
+}
+EXPORT_SYMBOL(v9fs_register_trans);
+
+/**
+ * v9fs_unregister_trans - unregister a 9p transport
+ * @m: the transport to remove
+ *
+ */
+void v9fs_unregister_trans(struct p9_trans_module *m)
+{
+ spin_lock(&v9fs_trans_lock);
+ list_del_init(&m->list);
+ spin_unlock(&v9fs_trans_lock);
+}
+EXPORT_SYMBOL(v9fs_unregister_trans);
+
+/**
+ * v9fs_get_trans_by_name - get transport with the matching name
+ * @name: string identifying transport
+ *
+ */
+struct p9_trans_module *v9fs_get_trans_by_name(char *s)
+{
+ struct p9_trans_module *t, *found = NULL;
+
+ spin_lock(&v9fs_trans_lock);
+
+ list_for_each_entry(t, &v9fs_trans_list, list)
+ if (strcmp(t->name, s) == 0 &&
+ try_module_get(t->owner)) {
+ found = t;
+ break;
+ }
+
+ spin_unlock(&v9fs_trans_lock);
+ return found;
+}
+EXPORT_SYMBOL(v9fs_get_trans_by_name);
+
+/**
+ * v9fs_get_default_trans - get the default transport
+ *
+ */
+
+struct p9_trans_module *v9fs_get_default_trans(void)
+{
+ struct p9_trans_module *t, *found = NULL;
+
+ spin_lock(&v9fs_trans_lock);
+
+ list_for_each_entry(t, &v9fs_trans_list, list)
+ if (t->def && try_module_get(t->owner)) {
+ found = t;
+ break;
+ }
+
+ if (!found)
+ list_for_each_entry(t, &v9fs_trans_list, list)
+ if (try_module_get(t->owner)) {
+ found = t;
+ break;
+ }
+
+ spin_unlock(&v9fs_trans_lock);
+ return found;
+}
+EXPORT_SYMBOL(v9fs_get_default_trans);
+
+/**
+ * v9fs_put_trans - put trans
+ * @m: transport to put
+ *
+ */
+void v9fs_put_trans(struct p9_trans_module *m)
+{
+ if (m)
+ module_put(m->owner);
+}
+
+/**
+ * init_p9 - Initialize module
+ *
+ */
+static int __init init_p9(void)
+{
+ int ret = 0;
+
+ p9_error_init();
+ pr_info("Installing 9P2000 support\n");
+ p9_trans_fd_init();
+
+ return ret;
+}
+
+/**
+ * exit_p9 - shutdown module
+ *
+ */
+
+static void __exit exit_p9(void)
+{
+ pr_info("Unloading 9P2000 support\n");
+
+ p9_trans_fd_exit();
+}
+
+module_init(init_p9)
+module_exit(exit_p9)
+
+MODULE_AUTHOR("Latchesar Ionkov <lucho@ionkov.net>");
+MODULE_AUTHOR("Eric Van Hensbergen <ericvh@gmail.com>");
+MODULE_AUTHOR("Ron Minnich <rminnich@lanl.gov>");
+MODULE_LICENSE("GPL");
diff --git a/net/9p/protocol.c b/net/9p/protocol.c
new file mode 100644
index 000000000..16d287565
--- /dev/null
+++ b/net/9p/protocol.c
@@ -0,0 +1,628 @@
+/*
+ * net/9p/protocol.c
+ *
+ * 9P Protocol Support Code
+ *
+ * Copyright (C) 2008 by Eric Van Hensbergen <ericvh@gmail.com>
+ *
+ * Base on code from Anthony Liguori <aliguori@us.ibm.com>
+ * Copyright (C) 2008 by IBM, Corp.
+ *
+ * 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/kernel.h>
+#include <linux/uaccess.h>
+#include <linux/slab.h>
+#include <linux/sched.h>
+#include <linux/stddef.h>
+#include <linux/types.h>
+#include <linux/uio.h>
+#include <net/9p/9p.h>
+#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, ...);
+
+void p9stat_free(struct p9_wstat *stbuf)
+{
+ kfree(stbuf->name);
+ kfree(stbuf->uid);
+ kfree(stbuf->gid);
+ kfree(stbuf->muid);
+ kfree(stbuf->extension);
+}
+EXPORT_SYMBOL(p9stat_free);
+
+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);
+ pdu->offset += len;
+ return size - len;
+}
+
+static size_t pdu_write(struct p9_fcall *pdu, const void *data, size_t size)
+{
+ size_t len = min(pdu->capacity - pdu->size, size);
+ memcpy(&pdu->sdata[pdu->size], data, len);
+ pdu->size += len;
+ return size - len;
+}
+
+static size_t
+pdu_write_u(struct p9_fcall *pdu, struct iov_iter *from, size_t size)
+{
+ size_t len = min(pdu->capacity - pdu->size, size);
+ struct iov_iter i = *from;
+ if (copy_from_iter(&pdu->sdata[pdu->size], len, &i) != len)
+ len = 0;
+
+ pdu->size += len;
+ return size - len;
+}
+
+/*
+ b - int8_t
+ w - int16_t
+ d - int32_t
+ q - int64_t
+ s - string
+ u - numeric uid
+ g - numeric gid
+ S - stat
+ Q - qid
+ D - data blob (int32_t size followed by void *, results are not freed)
+ T - array of strings (int16_t count, followed by strings)
+ R - array of qids (int16_t count, followed by qids)
+ A - stat for 9p2000.L (p9_stat_dotl)
+ ? - if optional = 1, continue parsing
+*/
+
+static int
+p9pdu_vreadf(struct p9_fcall *pdu, int proto_version, const char *fmt,
+ va_list ap)
+{
+ const char *ptr;
+ int errcode = 0;
+
+ for (ptr = fmt; *ptr; ptr++) {
+ switch (*ptr) {
+ case 'b':{
+ int8_t *val = va_arg(ap, int8_t *);
+ if (pdu_read(pdu, val, sizeof(*val))) {
+ errcode = -EFAULT;
+ break;
+ }
+ }
+ break;
+ case 'w':{
+ int16_t *val = va_arg(ap, int16_t *);
+ __le16 le_val;
+ if (pdu_read(pdu, &le_val, sizeof(le_val))) {
+ errcode = -EFAULT;
+ break;
+ }
+ *val = le16_to_cpu(le_val);
+ }
+ break;
+ case 'd':{
+ int32_t *val = va_arg(ap, int32_t *);
+ __le32 le_val;
+ if (pdu_read(pdu, &le_val, sizeof(le_val))) {
+ errcode = -EFAULT;
+ break;
+ }
+ *val = le32_to_cpu(le_val);
+ }
+ break;
+ case 'q':{
+ int64_t *val = va_arg(ap, int64_t *);
+ __le64 le_val;
+ if (pdu_read(pdu, &le_val, sizeof(le_val))) {
+ errcode = -EFAULT;
+ break;
+ }
+ *val = le64_to_cpu(le_val);
+ }
+ break;
+ case 's':{
+ char **sptr = va_arg(ap, char **);
+ uint16_t len;
+
+ errcode = p9pdu_readf(pdu, proto_version,
+ "w", &len);
+ if (errcode)
+ break;
+
+ *sptr = kmalloc(len + 1, GFP_NOFS);
+ if (*sptr == NULL) {
+ errcode = -EFAULT;
+ break;
+ }
+ if (pdu_read(pdu, *sptr, len)) {
+ errcode = -EFAULT;
+ kfree(*sptr);
+ *sptr = NULL;
+ } else
+ (*sptr)[len] = 0;
+ }
+ break;
+ case 'u': {
+ kuid_t *uid = va_arg(ap, kuid_t *);
+ __le32 le_val;
+ if (pdu_read(pdu, &le_val, sizeof(le_val))) {
+ errcode = -EFAULT;
+ break;
+ }
+ *uid = make_kuid(&init_user_ns,
+ le32_to_cpu(le_val));
+ } break;
+ case 'g': {
+ kgid_t *gid = va_arg(ap, kgid_t *);
+ __le32 le_val;
+ if (pdu_read(pdu, &le_val, sizeof(le_val))) {
+ errcode = -EFAULT;
+ break;
+ }
+ *gid = make_kgid(&init_user_ns,
+ le32_to_cpu(le_val));
+ } break;
+ case 'Q':{
+ struct p9_qid *qid =
+ va_arg(ap, struct p9_qid *);
+
+ errcode = p9pdu_readf(pdu, proto_version, "bdq",
+ &qid->type, &qid->version,
+ &qid->path);
+ }
+ break;
+ case 'S':{
+ struct p9_wstat *stbuf =
+ va_arg(ap, struct p9_wstat *);
+
+ memset(stbuf, 0, sizeof(struct p9_wstat));
+ stbuf->n_uid = stbuf->n_muid = INVALID_UID;
+ stbuf->n_gid = INVALID_GID;
+
+ errcode =
+ p9pdu_readf(pdu, proto_version,
+ "wwdQdddqssss?sugu",
+ &stbuf->size, &stbuf->type,
+ &stbuf->dev, &stbuf->qid,
+ &stbuf->mode, &stbuf->atime,
+ &stbuf->mtime, &stbuf->length,
+ &stbuf->name, &stbuf->uid,
+ &stbuf->gid, &stbuf->muid,
+ &stbuf->extension,
+ &stbuf->n_uid, &stbuf->n_gid,
+ &stbuf->n_muid);
+ if (errcode)
+ p9stat_free(stbuf);
+ }
+ break;
+ case 'D':{
+ uint32_t *count = va_arg(ap, uint32_t *);
+ void **data = va_arg(ap, void **);
+
+ errcode =
+ p9pdu_readf(pdu, proto_version, "d", count);
+ if (!errcode) {
+ *count =
+ min_t(uint32_t, *count,
+ pdu->size - pdu->offset);
+ *data = &pdu->sdata[pdu->offset];
+ }
+ }
+ break;
+ case 'T':{
+ uint16_t *nwname = va_arg(ap, uint16_t *);
+ char ***wnames = va_arg(ap, char ***);
+
+ errcode = p9pdu_readf(pdu, proto_version,
+ "w", nwname);
+ if (!errcode) {
+ *wnames =
+ kmalloc(sizeof(char *) * *nwname,
+ GFP_NOFS);
+ if (!*wnames)
+ errcode = -ENOMEM;
+ }
+
+ if (!errcode) {
+ int i;
+
+ for (i = 0; i < *nwname; i++) {
+ errcode =
+ p9pdu_readf(pdu,
+ proto_version,
+ "s",
+ &(*wnames)[i]);
+ if (errcode)
+ break;
+ }
+ }
+
+ if (errcode) {
+ if (*wnames) {
+ int i;
+
+ for (i = 0; i < *nwname; i++)
+ kfree((*wnames)[i]);
+ }
+ kfree(*wnames);
+ *wnames = NULL;
+ }
+ }
+ break;
+ case 'R':{
+ uint16_t *nwqid = va_arg(ap, uint16_t *);
+ struct p9_qid **wqids =
+ va_arg(ap, struct p9_qid **);
+
+ *wqids = NULL;
+
+ errcode =
+ p9pdu_readf(pdu, proto_version, "w", nwqid);
+ if (!errcode) {
+ *wqids =
+ kmalloc(*nwqid *
+ sizeof(struct p9_qid),
+ GFP_NOFS);
+ if (*wqids == NULL)
+ errcode = -ENOMEM;
+ }
+
+ if (!errcode) {
+ int i;
+
+ for (i = 0; i < *nwqid; i++) {
+ errcode =
+ p9pdu_readf(pdu,
+ proto_version,
+ "Q",
+ &(*wqids)[i]);
+ if (errcode)
+ break;
+ }
+ }
+
+ if (errcode) {
+ kfree(*wqids);
+ *wqids = NULL;
+ }
+ }
+ break;
+ case 'A': {
+ struct p9_stat_dotl *stbuf =
+ va_arg(ap, struct p9_stat_dotl *);
+
+ memset(stbuf, 0, sizeof(struct p9_stat_dotl));
+ errcode =
+ p9pdu_readf(pdu, proto_version,
+ "qQdugqqqqqqqqqqqqqqq",
+ &stbuf->st_result_mask,
+ &stbuf->qid,
+ &stbuf->st_mode,
+ &stbuf->st_uid, &stbuf->st_gid,
+ &stbuf->st_nlink,
+ &stbuf->st_rdev, &stbuf->st_size,
+ &stbuf->st_blksize, &stbuf->st_blocks,
+ &stbuf->st_atime_sec,
+ &stbuf->st_atime_nsec,
+ &stbuf->st_mtime_sec,
+ &stbuf->st_mtime_nsec,
+ &stbuf->st_ctime_sec,
+ &stbuf->st_ctime_nsec,
+ &stbuf->st_btime_sec,
+ &stbuf->st_btime_nsec,
+ &stbuf->st_gen,
+ &stbuf->st_data_version);
+ }
+ break;
+ case '?':
+ if ((proto_version != p9_proto_2000u) &&
+ (proto_version != p9_proto_2000L))
+ return 0;
+ break;
+ default:
+ BUG();
+ break;
+ }
+
+ if (errcode)
+ break;
+ }
+
+ return errcode;
+}
+
+int
+p9pdu_vwritef(struct p9_fcall *pdu, int proto_version, const char *fmt,
+ va_list ap)
+{
+ const char *ptr;
+ int errcode = 0;
+
+ for (ptr = fmt; *ptr; ptr++) {
+ switch (*ptr) {
+ case 'b':{
+ int8_t val = va_arg(ap, int);
+ if (pdu_write(pdu, &val, sizeof(val)))
+ errcode = -EFAULT;
+ }
+ break;
+ case 'w':{
+ __le16 val = cpu_to_le16(va_arg(ap, int));
+ if (pdu_write(pdu, &val, sizeof(val)))
+ errcode = -EFAULT;
+ }
+ break;
+ case 'd':{
+ __le32 val = cpu_to_le32(va_arg(ap, int32_t));
+ if (pdu_write(pdu, &val, sizeof(val)))
+ errcode = -EFAULT;
+ }
+ break;
+ case 'q':{
+ __le64 val = cpu_to_le64(va_arg(ap, int64_t));
+ if (pdu_write(pdu, &val, sizeof(val)))
+ errcode = -EFAULT;
+ }
+ break;
+ case 's':{
+ const char *sptr = va_arg(ap, const char *);
+ uint16_t len = 0;
+ if (sptr)
+ len = min_t(size_t, strlen(sptr),
+ USHRT_MAX);
+
+ errcode = p9pdu_writef(pdu, proto_version,
+ "w", len);
+ if (!errcode && pdu_write(pdu, sptr, len))
+ errcode = -EFAULT;
+ }
+ break;
+ case 'u': {
+ kuid_t uid = va_arg(ap, kuid_t);
+ __le32 val = cpu_to_le32(
+ from_kuid(&init_user_ns, uid));
+ if (pdu_write(pdu, &val, sizeof(val)))
+ errcode = -EFAULT;
+ } break;
+ case 'g': {
+ kgid_t gid = va_arg(ap, kgid_t);
+ __le32 val = cpu_to_le32(
+ from_kgid(&init_user_ns, gid));
+ if (pdu_write(pdu, &val, sizeof(val)))
+ errcode = -EFAULT;
+ } break;
+ case 'Q':{
+ const struct p9_qid *qid =
+ va_arg(ap, const struct p9_qid *);
+ errcode =
+ p9pdu_writef(pdu, proto_version, "bdq",
+ qid->type, qid->version,
+ qid->path);
+ } break;
+ case 'S':{
+ const struct p9_wstat *stbuf =
+ va_arg(ap, const struct p9_wstat *);
+ errcode =
+ p9pdu_writef(pdu, proto_version,
+ "wwdQdddqssss?sugu",
+ stbuf->size, stbuf->type,
+ stbuf->dev, &stbuf->qid,
+ stbuf->mode, stbuf->atime,
+ stbuf->mtime, stbuf->length,
+ stbuf->name, stbuf->uid,
+ stbuf->gid, stbuf->muid,
+ stbuf->extension, stbuf->n_uid,
+ stbuf->n_gid, stbuf->n_muid);
+ } break;
+ case 'V':{
+ uint32_t count = va_arg(ap, uint32_t);
+ struct iov_iter *from =
+ va_arg(ap, struct iov_iter *);
+ errcode = p9pdu_writef(pdu, proto_version, "d",
+ count);
+ if (!errcode && pdu_write_u(pdu, from, count))
+ errcode = -EFAULT;
+ }
+ break;
+ case 'T':{
+ uint16_t nwname = va_arg(ap, int);
+ const char **wnames = va_arg(ap, const char **);
+
+ errcode = p9pdu_writef(pdu, proto_version, "w",
+ nwname);
+ if (!errcode) {
+ int i;
+
+ for (i = 0; i < nwname; i++) {
+ errcode =
+ p9pdu_writef(pdu,
+ proto_version,
+ "s",
+ wnames[i]);
+ if (errcode)
+ break;
+ }
+ }
+ }
+ break;
+ case 'R':{
+ uint16_t nwqid = va_arg(ap, int);
+ struct p9_qid *wqids =
+ va_arg(ap, struct p9_qid *);
+
+ errcode = p9pdu_writef(pdu, proto_version, "w",
+ nwqid);
+ if (!errcode) {
+ int i;
+
+ for (i = 0; i < nwqid; i++) {
+ errcode =
+ p9pdu_writef(pdu,
+ proto_version,
+ "Q",
+ &wqids[i]);
+ if (errcode)
+ break;
+ }
+ }
+ }
+ break;
+ case 'I':{
+ struct p9_iattr_dotl *p9attr = va_arg(ap,
+ struct p9_iattr_dotl *);
+
+ errcode = p9pdu_writef(pdu, proto_version,
+ "ddugqqqqq",
+ p9attr->valid,
+ p9attr->mode,
+ p9attr->uid,
+ p9attr->gid,
+ p9attr->size,
+ p9attr->atime_sec,
+ p9attr->atime_nsec,
+ p9attr->mtime_sec,
+ p9attr->mtime_nsec);
+ }
+ break;
+ case '?':
+ if ((proto_version != p9_proto_2000u) &&
+ (proto_version != p9_proto_2000L))
+ return 0;
+ break;
+ default:
+ BUG();
+ break;
+ }
+
+ if (errcode)
+ break;
+ }
+
+ return errcode;
+}
+
+int p9pdu_readf(struct p9_fcall *pdu, int proto_version, const char *fmt, ...)
+{
+ va_list ap;
+ int ret;
+
+ va_start(ap, fmt);
+ ret = p9pdu_vreadf(pdu, proto_version, fmt, ap);
+ va_end(ap);
+
+ return ret;
+}
+
+static int
+p9pdu_writef(struct p9_fcall *pdu, int proto_version, const char *fmt, ...)
+{
+ va_list ap;
+ int ret;
+
+ va_start(ap, fmt);
+ ret = p9pdu_vwritef(pdu, proto_version, fmt, ap);
+ va_end(ap);
+
+ return ret;
+}
+
+int p9stat_read(struct p9_client *clnt, char *buf, int len, struct p9_wstat *st)
+{
+ struct p9_fcall fake_pdu;
+ int ret;
+
+ fake_pdu.size = len;
+ fake_pdu.capacity = len;
+ fake_pdu.sdata = buf;
+ fake_pdu.offset = 0;
+
+ ret = p9pdu_readf(&fake_pdu, clnt->proto_version, "S", st);
+ if (ret) {
+ p9_debug(P9_DEBUG_9P, "<<< p9stat_read failed: %d\n", ret);
+ trace_9p_protocol_dump(clnt, &fake_pdu);
+ }
+
+ return ret;
+}
+EXPORT_SYMBOL(p9stat_read);
+
+int p9pdu_prepare(struct p9_fcall *pdu, int16_t tag, int8_t type)
+{
+ pdu->id = type;
+ return p9pdu_writef(pdu, 0, "dbw", 0, type, tag);
+}
+
+int p9pdu_finalize(struct p9_client *clnt, struct p9_fcall *pdu)
+{
+ int size = pdu->size;
+ int err;
+
+ pdu->size = 0;
+ err = p9pdu_writef(pdu, 0, "d", size);
+ pdu->size = size;
+
+ trace_9p_protocol_dump(clnt, pdu);
+ p9_debug(P9_DEBUG_9P, ">>> size=%d type: %d tag: %d\n",
+ pdu->size, pdu->id, pdu->tag);
+
+ return err;
+}
+
+void p9pdu_reset(struct p9_fcall *pdu)
+{
+ pdu->offset = 0;
+ pdu->size = 0;
+}
+
+int p9dirent_read(struct p9_client *clnt, char *buf, int len,
+ struct p9_dirent *dirent)
+{
+ struct p9_fcall fake_pdu;
+ int ret;
+ char *nameptr;
+
+ fake_pdu.size = len;
+ fake_pdu.capacity = len;
+ fake_pdu.sdata = buf;
+ fake_pdu.offset = 0;
+
+ ret = p9pdu_readf(&fake_pdu, clnt->proto_version, "Qqbs", &dirent->qid,
+ &dirent->d_off, &dirent->d_type, &nameptr);
+ if (ret) {
+ p9_debug(P9_DEBUG_9P, "<<< p9dirent_read failed: %d\n", ret);
+ trace_9p_protocol_dump(clnt, &fake_pdu);
+ goto out;
+ }
+
+ strcpy(dirent->d_name, nameptr);
+ kfree(nameptr);
+
+out:
+ return fake_pdu.offset;
+}
+EXPORT_SYMBOL(p9dirent_read);
diff --git a/net/9p/protocol.h b/net/9p/protocol.h
new file mode 100644
index 000000000..2cc525fa4
--- /dev/null
+++ b/net/9p/protocol.h
@@ -0,0 +1,34 @@
+/*
+ * net/9p/protocol.h
+ *
+ * 9P Protocol Support Code
+ *
+ * Copyright (C) 2008 by Eric Van Hensbergen <ericvh@gmail.com>
+ *
+ * Base on code from Anthony Liguori <aliguori@us.ibm.com>
+ * Copyright (C) 2008 by IBM, Corp.
+ *
+ * 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
+ *
+ */
+
+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_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
new file mode 100644
index 000000000..38aa6345b
--- /dev/null
+++ b/net/9p/trans_common.c
@@ -0,0 +1,29 @@
+/*
+ * Copyright IBM Corporation, 2010
+ * Author Venkateswararao Jujjuri <jvrao@linux.vnet.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+ *
+ */
+
+#include <linux/mm.h>
+#include <linux/module.h>
+
+/**
+ * p9_release_req_pages - Release pages after the transaction.
+ */
+void p9_release_pages(struct page **pages, int nr_pages)
+{
+ int i;
+
+ for (i = 0; i < nr_pages; i++)
+ if (pages[i])
+ put_page(pages[i]);
+}
+EXPORT_SYMBOL(p9_release_pages);
diff --git a/net/9p/trans_common.h b/net/9p/trans_common.h
new file mode 100644
index 000000000..c43babb3f
--- /dev/null
+++ b/net/9p/trans_common.h
@@ -0,0 +1,15 @@
+/*
+ * Copyright IBM Corporation, 2010
+ * Author Venkateswararao Jujjuri <jvrao@linux.vnet.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+ *
+ */
+
+void p9_release_pages(struct page **, int);
diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c
new file mode 100644
index 000000000..bced8c074
--- /dev/null
+++ b/net/9p/trans_fd.c
@@ -0,0 +1,1117 @@
+/*
+ * linux/fs/9p/trans_fd.c
+ *
+ * Fd transport layer. Includes deprecated socket layer.
+ *
+ * Copyright (C) 2006 by Russ Cox <rsc@swtch.com>
+ * Copyright (C) 2004-2005 by Latchesar Ionkov <lucho@ionkov.net>
+ * 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
+ * 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
+ *
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/in.h>
+#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>
+#include <linux/uaccess.h>
+#include <linux/inet.h>
+#include <linux/idr.h>
+#include <linux/file.h>
+#include <linux/parser.h>
+#include <linux/slab.h>
+#include <net/9p/9p.h>
+#include <net/9p/client.h>
+#include <net/9p/transport.h>
+
+#include <linux/syscalls.h> /* killme */
+
+#define P9_PORT 564
+#define MAX_SOCK_BUF (64*1024)
+#define MAXPOLLWADDR 2
+
+/**
+ * struct p9_fd_opts - per-transport options
+ * @rfd: file descriptor for reading (trans=fd)
+ * @wfd: file descriptor for writing (trans=fd)
+ * @port: port to connect to (trans=tcp)
+ *
+ */
+
+struct p9_fd_opts {
+ int rfd;
+ int wfd;
+ u16 port;
+ int privport;
+};
+
+/*
+ * Option Parsing (code inspired by NFS code)
+ * - a little lazy - parse all fd-transport options
+ */
+
+enum {
+ /* Options that take integer arguments */
+ Opt_port, Opt_rfdno, Opt_wfdno, Opt_err,
+ /* Options that take no arguments */
+ Opt_privport,
+};
+
+static const match_table_t tokens = {
+ {Opt_port, "port=%u"},
+ {Opt_rfdno, "rfdno=%u"},
+ {Opt_wfdno, "wfdno=%u"},
+ {Opt_privport, "privport"},
+ {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 */
+};
+
+struct p9_poll_wait {
+ struct p9_conn *conn;
+ wait_queue_t wait;
+ wait_queue_head_t *wait_addr;
+};
+
+/**
+ * struct p9_conn - fd mux connection state information
+ * @mux_list: list link for mux to manage multiple connections (?)
+ * @client: reference to client instance for this connection
+ * @err: error state
+ * @req_list: accounting for requests which have been sent
+ * @unsent_req_list: accounting for requests that haven't been sent
+ * @req: current request being processed (if any)
+ * @tmp_buf: temporary buffer to read in header
+ * @rsize: amount to read for current frame
+ * @rpos: read position in current frame
+ * @rbuf: current read buffer
+ * @wpos: write position for current frame
+ * @wsize: amount of data to write for current frame
+ * @wbuf: current write buffer
+ * @poll_pending_link: pending links to be polled per conn
+ * @poll_wait: array of wait_q's for various worker threads
+ * @pt: poll state
+ * @rq: current read work
+ * @wq: current write work
+ * @wsched: ????
+ *
+ */
+
+struct p9_conn {
+ struct list_head mux_list;
+ struct p9_client *client;
+ int err;
+ struct list_head req_list;
+ struct list_head unsent_req_list;
+ struct p9_req_t *req;
+ char tmp_buf[7];
+ int rsize;
+ int rpos;
+ char *rbuf;
+ int wpos;
+ int wsize;
+ char *wbuf;
+ struct list_head poll_pending_link;
+ struct p9_poll_wait poll_wait[MAXPOLLWADDR];
+ poll_table pt;
+ struct work_struct rq;
+ struct work_struct wq;
+ unsigned long wsched;
+};
+
+/**
+ * struct p9_trans_fd - transport state
+ * @rd: reference to file to read from
+ * @wr: reference of file to write to
+ * @conn: connection state reference
+ *
+ */
+
+struct p9_trans_fd {
+ struct file *rd;
+ struct file *wr;
+ struct p9_conn conn;
+};
+
+static void p9_poll_workfn(struct work_struct *work);
+
+static DEFINE_SPINLOCK(p9_poll_lock);
+static LIST_HEAD(p9_poll_pending_list);
+static DECLARE_WORK(p9_poll_work, p9_poll_workfn);
+
+static unsigned int p9_ipport_resv_min = P9_DEF_MIN_RESVPORT;
+static unsigned int p9_ipport_resv_max = P9_DEF_MAX_RESVPORT;
+
+static void p9_mux_poll_stop(struct p9_conn *m)
+{
+ unsigned long flags;
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(m->poll_wait); i++) {
+ struct p9_poll_wait *pwait = &m->poll_wait[i];
+
+ if (pwait->wait_addr) {
+ remove_wait_queue(pwait->wait_addr, &pwait->wait);
+ pwait->wait_addr = NULL;
+ }
+ }
+
+ spin_lock_irqsave(&p9_poll_lock, flags);
+ list_del_init(&m->poll_pending_link);
+ spin_unlock_irqrestore(&p9_poll_lock, flags);
+}
+
+/**
+ * p9_conn_cancel - cancel all pending requests with error
+ * @m: mux data
+ * @err: error code
+ *
+ */
+
+static void p9_conn_cancel(struct p9_conn *m, int err)
+{
+ struct p9_req_t *req, *rtmp;
+ unsigned long flags;
+ LIST_HEAD(cancel_list);
+
+ p9_debug(P9_DEBUG_ERROR, "mux %p err %d\n", m, err);
+
+ spin_lock_irqsave(&m->client->lock, flags);
+
+ if (m->err) {
+ spin_unlock_irqrestore(&m->client->lock, flags);
+ return;
+ }
+
+ m->err = err;
+
+ 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_irqrestore(&m->client->lock, flags);
+
+ list_for_each_entry_safe(req, rtmp, &cancel_list, req_list) {
+ p9_debug(P9_DEBUG_ERROR, "call back req %p\n", req);
+ list_del(&req->req_list);
+ if (!req->t_err)
+ req->t_err = err;
+ p9_client_cb(m->client, req, REQ_STATUS_ERROR);
+ }
+}
+
+static int
+p9_fd_poll(struct p9_client *client, struct poll_table_struct *pt)
+{
+ int ret, n;
+ struct p9_trans_fd *ts = NULL;
+
+ if (client && client->status == Connected)
+ ts = client->trans;
+
+ if (!ts)
+ return -EREMOTEIO;
+
+ if (!ts->rd->f_op->poll)
+ return -EIO;
+
+ if (!ts->wr->f_op->poll)
+ return -EIO;
+
+ ret = ts->rd->f_op->poll(ts->rd, pt);
+ if (ret < 0)
+ return ret;
+
+ if (ts->rd != ts->wr) {
+ n = ts->wr->f_op->poll(ts->wr, pt);
+ if (n < 0)
+ return n;
+ ret = (ret & ~POLLOUT) | (n & ~POLLIN);
+ }
+
+ return ret;
+}
+
+/**
+ * p9_fd_read- read from a fd
+ * @client: client instance
+ * @v: buffer to receive data into
+ * @len: size of receive buffer
+ *
+ */
+
+static int p9_fd_read(struct p9_client *client, void *v, int len)
+{
+ int ret;
+ struct p9_trans_fd *ts = NULL;
+
+ if (client && client->status != Disconnected)
+ ts = client->trans;
+
+ if (!ts)
+ return -EREMOTEIO;
+
+ if (!(ts->rd->f_flags & O_NONBLOCK))
+ p9_debug(P9_DEBUG_ERROR, "blocking read ...\n");
+
+ ret = kernel_read(ts->rd, ts->rd->f_pos, v, len);
+ if (ret <= 0 && ret != -ERESTARTSYS && ret != -EAGAIN)
+ client->status = Disconnected;
+ return ret;
+}
+
+/**
+ * p9_read_work - called when there is some data to be read from a transport
+ * @work: container of work to be done
+ *
+ */
+
+static void p9_read_work(struct work_struct *work)
+{
+ int n, err;
+ struct p9_conn *m;
+ int status = REQ_STATUS_ERROR;
+
+ m = container_of(work, struct p9_conn, rq);
+
+ if (m->err < 0)
+ return;
+
+ p9_debug(P9_DEBUG_TRANS, "start mux %p pos %d\n", m, m->rpos);
+
+ if (!m->rbuf) {
+ m->rbuf = m->tmp_buf;
+ m->rpos = 0;
+ m->rsize = 7; /* start by reading header */
+ }
+
+ clear_bit(Rpending, &m->wsched);
+ p9_debug(P9_DEBUG_TRANS, "read mux %p pos %d size: %d = %d\n",
+ m, m->rpos, m->rsize, m->rsize-m->rpos);
+ err = p9_fd_read(m->client, m->rbuf + m->rpos,
+ m->rsize - m->rpos);
+ p9_debug(P9_DEBUG_TRANS, "mux %p got %d bytes\n", m, err);
+ if (err == -EAGAIN) {
+ goto end_clear;
+ }
+
+ if (err <= 0)
+ goto error;
+
+ m->rpos += err;
+
+ if ((!m->req) && (m->rpos == m->rsize)) { /* header read in */
+ u16 tag;
+ p9_debug(P9_DEBUG_TRANS, "got new header\n");
+
+ n = le32_to_cpu(*(__le32 *) m->rbuf); /* read packet size */
+ if (n >= m->client->msize) {
+ p9_debug(P9_DEBUG_ERROR,
+ "requested packet size too big: %d\n", n);
+ err = -EIO;
+ goto error;
+ }
+
+ tag = le16_to_cpu(*(__le16 *) (m->rbuf+5)); /* read tag */
+ p9_debug(P9_DEBUG_TRANS,
+ "mux %p pkt: size: %d bytes tag: %d\n", m, n, tag);
+
+ m->req = p9_tag_lookup(m->client, tag);
+ if (!m->req || (m->req->status != REQ_STATUS_SENT)) {
+ p9_debug(P9_DEBUG_ERROR, "Unexpected packet tag %d\n",
+ tag);
+ err = -EIO;
+ goto error;
+ }
+
+ if (m->req->rc == NULL) {
+ m->req->rc = kmalloc(sizeof(struct p9_fcall) +
+ m->client->msize, GFP_NOFS);
+ if (!m->req->rc) {
+ m->req = NULL;
+ err = -ENOMEM;
+ goto error;
+ }
+ }
+ m->rbuf = (char *)m->req->rc + sizeof(struct p9_fcall);
+ memcpy(m->rbuf, m->tmp_buf, m->rsize);
+ m->rsize = n;
+ }
+
+ /* not an else because some packets (like clunk) have no payload */
+ if ((m->req) && (m->rpos == m->rsize)) { /* packet is read in */
+ p9_debug(P9_DEBUG_TRANS, "got new packet\n");
+ spin_lock(&m->client->lock);
+ if (m->req->status != REQ_STATUS_ERROR)
+ status = REQ_STATUS_RCVD;
+ list_del(&m->req->req_list);
+ spin_unlock(&m->client->lock);
+ p9_client_cb(m->client, m->req, status);
+ m->rbuf = NULL;
+ m->rpos = 0;
+ m->rsize = 0;
+ m->req = NULL;
+ }
+
+end_clear:
+ clear_bit(Rworksched, &m->wsched);
+
+ if (!list_empty(&m->req_list)) {
+ if (test_and_clear_bit(Rpending, &m->wsched))
+ n = POLLIN;
+ else
+ n = p9_fd_poll(m->client, NULL);
+
+ if ((n & POLLIN) && !test_and_set_bit(Rworksched, &m->wsched)) {
+ p9_debug(P9_DEBUG_TRANS, "sched read work %p\n", m);
+ schedule_work(&m->rq);
+ }
+ }
+
+ return;
+error:
+ p9_conn_cancel(m, err);
+ clear_bit(Rworksched, &m->wsched);
+}
+
+/**
+ * p9_fd_write - write to a socket
+ * @client: client instance
+ * @v: buffer to send data from
+ * @len: size of send buffer
+ *
+ */
+
+static int p9_fd_write(struct p9_client *client, void *v, int len)
+{
+ int ret;
+ mm_segment_t oldfs;
+ struct p9_trans_fd *ts = NULL;
+
+ if (client && client->status != Disconnected)
+ ts = client->trans;
+
+ if (!ts)
+ return -EREMOTEIO;
+
+ if (!(ts->wr->f_flags & O_NONBLOCK))
+ p9_debug(P9_DEBUG_ERROR, "blocking write ...\n");
+
+ oldfs = get_fs();
+ set_fs(get_ds());
+ /* The cast to a user pointer is valid due to the set_fs() */
+ ret = vfs_write(ts->wr, (__force void __user *)v, len, &ts->wr->f_pos);
+ set_fs(oldfs);
+
+ if (ret <= 0 && ret != -ERESTARTSYS && ret != -EAGAIN)
+ client->status = Disconnected;
+ return ret;
+}
+
+/**
+ * p9_write_work - called when a transport can send some data
+ * @work: container for work to be done
+ *
+ */
+
+static void p9_write_work(struct work_struct *work)
+{
+ int n, err;
+ struct p9_conn *m;
+ struct p9_req_t *req;
+
+ m = container_of(work, struct p9_conn, wq);
+
+ if (m->err < 0) {
+ clear_bit(Wworksched, &m->wsched);
+ return;
+ }
+
+ if (!m->wsize) {
+ spin_lock(&m->client->lock);
+ if (list_empty(&m->unsent_req_list)) {
+ clear_bit(Wworksched, &m->wsched);
+ spin_unlock(&m->client->lock);
+ return;
+ }
+
+ req = list_entry(m->unsent_req_list.next, struct p9_req_t,
+ req_list);
+ req->status = REQ_STATUS_SENT;
+ p9_debug(P9_DEBUG_TRANS, "move req %p\n", req);
+ list_move_tail(&req->req_list, &m->req_list);
+
+ m->wbuf = req->tc->sdata;
+ m->wsize = req->tc->size;
+ m->wpos = 0;
+ spin_unlock(&m->client->lock);
+ }
+
+ p9_debug(P9_DEBUG_TRANS, "mux %p pos %d size %d\n",
+ m, m->wpos, m->wsize);
+ clear_bit(Wpending, &m->wsched);
+ err = p9_fd_write(m->client, m->wbuf + m->wpos, m->wsize - m->wpos);
+ p9_debug(P9_DEBUG_TRANS, "mux %p sent %d bytes\n", m, err);
+ if (err == -EAGAIN)
+ goto end_clear;
+
+
+ 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;
+
+end_clear:
+ clear_bit(Wworksched, &m->wsched);
+
+ if (m->wsize || !list_empty(&m->unsent_req_list)) {
+ if (test_and_clear_bit(Wpending, &m->wsched))
+ n = POLLOUT;
+ else
+ n = p9_fd_poll(m->client, NULL);
+
+ if ((n & POLLOUT) &&
+ !test_and_set_bit(Wworksched, &m->wsched)) {
+ p9_debug(P9_DEBUG_TRANS, "sched write work %p\n", m);
+ schedule_work(&m->wq);
+ }
+ }
+
+ return;
+
+error:
+ p9_conn_cancel(m, err);
+ clear_bit(Wworksched, &m->wsched);
+}
+
+static int p9_pollwake(wait_queue_t *wait, unsigned int mode, int sync, void *key)
+{
+ struct p9_poll_wait *pwait =
+ container_of(wait, struct p9_poll_wait, wait);
+ struct p9_conn *m = pwait->conn;
+ unsigned long flags;
+
+ spin_lock_irqsave(&p9_poll_lock, flags);
+ if (list_empty(&m->poll_pending_link))
+ list_add_tail(&m->poll_pending_link, &p9_poll_pending_list);
+ spin_unlock_irqrestore(&p9_poll_lock, flags);
+
+ schedule_work(&p9_poll_work);
+ return 1;
+}
+
+/**
+ * p9_pollwait - add poll task to the wait queue
+ * @filp: file pointer being polled
+ * @wait_address: wait_q to block on
+ * @p: poll state
+ *
+ * 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)
+{
+ struct p9_conn *m = container_of(p, struct p9_conn, pt);
+ struct p9_poll_wait *pwait = NULL;
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(m->poll_wait); i++) {
+ if (m->poll_wait[i].wait_addr == NULL) {
+ pwait = &m->poll_wait[i];
+ break;
+ }
+ }
+
+ if (!pwait) {
+ p9_debug(P9_DEBUG_ERROR, "not enough wait_address slots\n");
+ return;
+ }
+
+ pwait->conn = m;
+ pwait->wait_addr = wait_address;
+ init_waitqueue_func_entry(&pwait->wait, p9_pollwake);
+ add_wait_queue(wait_address, &pwait->wait);
+}
+
+/**
+ * p9_conn_create - initialize the per-session mux data
+ * @client: client instance
+ *
+ * Note: Creates the polling task if this is the first session.
+ */
+
+static void p9_conn_create(struct p9_client *client)
+{
+ int n;
+ struct p9_trans_fd *ts = client->trans;
+ struct p9_conn *m = &ts->conn;
+
+ p9_debug(P9_DEBUG_TRANS, "client %p msize %d\n", client, client->msize);
+
+ INIT_LIST_HEAD(&m->mux_list);
+ m->client = client;
+
+ INIT_LIST_HEAD(&m->req_list);
+ INIT_LIST_HEAD(&m->unsent_req_list);
+ INIT_WORK(&m->rq, p9_read_work);
+ INIT_WORK(&m->wq, p9_write_work);
+ INIT_LIST_HEAD(&m->poll_pending_link);
+ init_poll_funcptr(&m->pt, p9_pollwait);
+
+ n = p9_fd_poll(client, &m->pt);
+ if (n & POLLIN) {
+ p9_debug(P9_DEBUG_TRANS, "mux %p can read\n", m);
+ set_bit(Rpending, &m->wsched);
+ }
+
+ if (n & POLLOUT) {
+ p9_debug(P9_DEBUG_TRANS, "mux %p can write\n", m);
+ set_bit(Wpending, &m->wsched);
+ }
+}
+
+/**
+ * p9_poll_mux - polls a mux and schedules read or write works if necessary
+ * @m: connection to poll
+ *
+ */
+
+static void p9_poll_mux(struct p9_conn *m)
+{
+ int n;
+
+ if (m->err < 0)
+ return;
+
+ n = p9_fd_poll(m->client, NULL);
+ if (n < 0 || n & (POLLERR | POLLHUP | POLLNVAL)) {
+ p9_debug(P9_DEBUG_TRANS, "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_debug(P9_DEBUG_TRANS, "mux %p can read\n", m);
+ if (!test_and_set_bit(Rworksched, &m->wsched)) {
+ p9_debug(P9_DEBUG_TRANS, "sched read work %p\n", m);
+ schedule_work(&m->rq);
+ }
+ }
+
+ if (n & POLLOUT) {
+ set_bit(Wpending, &m->wsched);
+ p9_debug(P9_DEBUG_TRANS, "mux %p can write\n", m);
+ if ((m->wsize || !list_empty(&m->unsent_req_list)) &&
+ !test_and_set_bit(Wworksched, &m->wsched)) {
+ p9_debug(P9_DEBUG_TRANS, "sched write work %p\n", m);
+ schedule_work(&m->wq);
+ }
+ }
+}
+
+/**
+ * p9_fd_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.
+ *
+ * @client: client instance
+ * @req: request to be sent
+ *
+ */
+
+static int p9_fd_request(struct p9_client *client, struct p9_req_t *req)
+{
+ int n;
+ struct p9_trans_fd *ts = client->trans;
+ struct p9_conn *m = &ts->conn;
+
+ p9_debug(P9_DEBUG_TRANS, "mux %p task %p tcall %p id %d\n",
+ m, current, req->tc, req->tc->id);
+ if (m->err < 0)
+ return m->err;
+
+ spin_lock(&client->lock);
+ req->status = REQ_STATUS_UNSENT;
+ list_add_tail(&req->req_list, &m->unsent_req_list);
+ spin_unlock(&client->lock);
+
+ if (test_and_clear_bit(Wpending, &m->wsched))
+ n = POLLOUT;
+ else
+ n = p9_fd_poll(m->client, NULL);
+
+ if (n & POLLOUT && !test_and_set_bit(Wworksched, &m->wsched))
+ schedule_work(&m->wq);
+
+ return 0;
+}
+
+static int p9_fd_cancel(struct p9_client *client, struct p9_req_t *req)
+{
+ int ret = 1;
+
+ p9_debug(P9_DEBUG_TRANS, "client %p req %p\n", client, req);
+
+ spin_lock(&client->lock);
+
+ if (req->status == REQ_STATUS_UNSENT) {
+ list_del(&req->req_list);
+ req->status = REQ_STATUS_FLSHD;
+ ret = 0;
+ }
+ spin_unlock(&client->lock);
+
+ return ret;
+}
+
+static int p9_fd_cancelled(struct p9_client *client, struct p9_req_t *req)
+{
+ p9_debug(P9_DEBUG_TRANS, "client %p req %p\n", client, req);
+
+ /* we haven't received a response for oldreq,
+ * remove it from the list.
+ */
+ spin_lock(&client->lock);
+ list_del(&req->req_list);
+ spin_unlock(&client->lock);
+
+ return 0;
+}
+
+/**
+ * parse_opts - parse mount options into p9_fd_opts structure
+ * @params: options string passed from mount
+ * @opts: fd transport-specific structure to parse options into
+ *
+ * Returns 0 upon success, -ERRNO upon failure
+ */
+
+static int parse_opts(char *params, struct p9_fd_opts *opts)
+{
+ char *p;
+ substring_t args[MAX_OPT_ARGS];
+ int option;
+ char *options, *tmp_options;
+
+ opts->port = P9_PORT;
+ opts->rfd = ~0;
+ opts->wfd = ~0;
+ opts->privport = 0;
+
+ if (!params)
+ return 0;
+
+ tmp_options = kstrdup(params, GFP_KERNEL);
+ if (!tmp_options) {
+ p9_debug(P9_DEBUG_ERROR,
+ "failed to allocate copy of option string\n");
+ return -ENOMEM;
+ }
+ options = tmp_options;
+
+ while ((p = strsep(&options, ",")) != NULL) {
+ int token;
+ int r;
+ if (!*p)
+ continue;
+ token = match_token(p, tokens, args);
+ if ((token != Opt_err) && (token != Opt_privport)) {
+ r = match_int(&args[0], &option);
+ if (r < 0) {
+ p9_debug(P9_DEBUG_ERROR,
+ "integer field, but no integer?\n");
+ continue;
+ }
+ }
+ switch (token) {
+ case Opt_port:
+ opts->port = option;
+ break;
+ case Opt_rfdno:
+ opts->rfd = option;
+ break;
+ case Opt_wfdno:
+ opts->wfd = option;
+ break;
+ case Opt_privport:
+ opts->privport = 1;
+ break;
+ default:
+ continue;
+ }
+ }
+
+ kfree(tmp_options);
+ return 0;
+}
+
+static int p9_fd_open(struct p9_client *client, int rfd, int wfd)
+{
+ struct p9_trans_fd *ts = kzalloc(sizeof(struct p9_trans_fd),
+ GFP_KERNEL);
+ if (!ts)
+ return -ENOMEM;
+
+ ts->rd = fget(rfd);
+ ts->wr = fget(wfd);
+ if (!ts->rd || !ts->wr) {
+ if (ts->rd)
+ fput(ts->rd);
+ if (ts->wr)
+ fput(ts->wr);
+ kfree(ts);
+ return -EIO;
+ }
+
+ client->trans = ts;
+ client->status = Connected;
+
+ return 0;
+}
+
+static int p9_socket_open(struct p9_client *client, struct socket *csocket)
+{
+ struct p9_trans_fd *p;
+ struct file *file;
+
+ p = kzalloc(sizeof(struct p9_trans_fd), GFP_KERNEL);
+ if (!p)
+ return -ENOMEM;
+
+ csocket->sk->sk_allocation = GFP_NOIO;
+ file = sock_alloc_file(csocket, 0, NULL);
+ if (IS_ERR(file)) {
+ pr_err("%s (%d): failed to map fd\n",
+ __func__, task_pid_nr(current));
+ sock_release(csocket);
+ kfree(p);
+ return PTR_ERR(file);
+ }
+
+ get_file(file);
+ p->wr = p->rd = file;
+ client->trans = p;
+ client->status = Connected;
+
+ p->rd->f_flags |= O_NONBLOCK;
+
+ p9_conn_create(client);
+ return 0;
+}
+
+/**
+ * p9_mux_destroy - cancels all pending requests of mux
+ * @m: mux to destroy
+ *
+ */
+
+static void p9_conn_destroy(struct p9_conn *m)
+{
+ p9_debug(P9_DEBUG_TRANS, "mux %p prev %p next %p\n",
+ m, m->mux_list.prev, m->mux_list.next);
+
+ p9_mux_poll_stop(m);
+ cancel_work_sync(&m->rq);
+ cancel_work_sync(&m->wq);
+
+ p9_conn_cancel(m, -ECONNRESET);
+
+ m->client = NULL;
+}
+
+/**
+ * p9_fd_close - shutdown file descriptor transport
+ * @client: client instance
+ *
+ */
+
+static void p9_fd_close(struct p9_client *client)
+{
+ struct p9_trans_fd *ts;
+
+ if (!client)
+ return;
+
+ ts = client->trans;
+ if (!ts)
+ return;
+
+ client->status = Disconnected;
+
+ p9_conn_destroy(&ts->conn);
+
+ if (ts->rd)
+ fput(ts->rd);
+ if (ts->wr)
+ fput(ts->wr);
+
+ kfree(ts);
+}
+
+/*
+ * stolen from NFS - maybe should be made a generic function?
+ */
+static inline int valid_ipaddr4(const char *buf)
+{
+ int rc, count, in[4];
+
+ rc = sscanf(buf, "%d.%d.%d.%d", &in[0], &in[1], &in[2], &in[3]);
+ if (rc != 4)
+ return -EINVAL;
+ for (count = 0; count < 4; count++) {
+ if (in[count] > 255)
+ return -EINVAL;
+ }
+ return 0;
+}
+
+static int p9_bind_privport(struct socket *sock)
+{
+ struct sockaddr_in cl;
+ int port, err = -EINVAL;
+
+ memset(&cl, 0, sizeof(cl));
+ cl.sin_family = AF_INET;
+ cl.sin_addr.s_addr = INADDR_ANY;
+ for (port = p9_ipport_resv_max; port >= p9_ipport_resv_min; port--) {
+ cl.sin_port = htons((ushort)port);
+ err = kernel_bind(sock, (struct sockaddr *)&cl, sizeof(cl));
+ if (err != -EADDRINUSE)
+ break;
+ }
+ return err;
+}
+
+
+static int
+p9_fd_create_tcp(struct p9_client *client, const char *addr, char *args)
+{
+ int err;
+ struct socket *csocket;
+ struct sockaddr_in sin_server;
+ struct p9_fd_opts opts;
+
+ err = parse_opts(args, &opts);
+ if (err < 0)
+ return err;
+
+ if (valid_ipaddr4(addr) < 0)
+ return -EINVAL;
+
+ csocket = NULL;
+
+ sin_server.sin_family = AF_INET;
+ sin_server.sin_addr.s_addr = in_aton(addr);
+ sin_server.sin_port = htons(opts.port);
+ err = __sock_create(current->nsproxy->net_ns, PF_INET,
+ SOCK_STREAM, IPPROTO_TCP, &csocket, 1);
+ if (err) {
+ pr_err("%s (%d): problem creating socket\n",
+ __func__, task_pid_nr(current));
+ return err;
+ }
+
+ if (opts.privport) {
+ err = p9_bind_privport(csocket);
+ if (err < 0) {
+ pr_err("%s (%d): problem binding to privport\n",
+ __func__, task_pid_nr(current));
+ sock_release(csocket);
+ return err;
+ }
+ }
+
+ err = csocket->ops->connect(csocket,
+ (struct sockaddr *)&sin_server,
+ sizeof(struct sockaddr_in), 0);
+ if (err < 0) {
+ pr_err("%s (%d): problem connecting socket to %s\n",
+ __func__, task_pid_nr(current), addr);
+ sock_release(csocket);
+ return err;
+ }
+
+ return p9_socket_open(client, csocket);
+}
+
+static int
+p9_fd_create_unix(struct p9_client *client, const char *addr, char *args)
+{
+ int err;
+ struct socket *csocket;
+ struct sockaddr_un sun_server;
+
+ csocket = NULL;
+
+ if (strlen(addr) >= UNIX_PATH_MAX) {
+ pr_err("%s (%d): address too long: %s\n",
+ __func__, task_pid_nr(current), addr);
+ return -ENAMETOOLONG;
+ }
+
+ sun_server.sun_family = PF_UNIX;
+ strcpy(sun_server.sun_path, addr);
+ err = __sock_create(current->nsproxy->net_ns, PF_UNIX,
+ SOCK_STREAM, 0, &csocket, 1);
+ if (err < 0) {
+ pr_err("%s (%d): problem creating socket\n",
+ __func__, task_pid_nr(current));
+
+ return err;
+ }
+ err = csocket->ops->connect(csocket, (struct sockaddr *)&sun_server,
+ sizeof(struct sockaddr_un) - 1, 0);
+ if (err < 0) {
+ pr_err("%s (%d): problem connecting socket: %s: %d\n",
+ __func__, task_pid_nr(current), addr, err);
+ sock_release(csocket);
+ return err;
+ }
+
+ return p9_socket_open(client, csocket);
+}
+
+static int
+p9_fd_create(struct p9_client *client, const char *addr, char *args)
+{
+ int err;
+ struct p9_fd_opts opts;
+
+ parse_opts(args, &opts);
+
+ if (opts.rfd == ~0 || opts.wfd == ~0) {
+ pr_err("Insufficient options for proto=fd\n");
+ return -ENOPROTOOPT;
+ }
+
+ err = p9_fd_open(client, opts.rfd, opts.wfd);
+ if (err < 0)
+ return err;
+
+ p9_conn_create(client);
+
+ return 0;
+}
+
+static struct p9_trans_module p9_tcp_trans = {
+ .name = "tcp",
+ .maxsize = MAX_SOCK_BUF,
+ .def = 0,
+ .create = p9_fd_create_tcp,
+ .close = p9_fd_close,
+ .request = p9_fd_request,
+ .cancel = p9_fd_cancel,
+ .cancelled = p9_fd_cancelled,
+ .owner = THIS_MODULE,
+};
+
+static struct p9_trans_module p9_unix_trans = {
+ .name = "unix",
+ .maxsize = MAX_SOCK_BUF,
+ .def = 0,
+ .create = p9_fd_create_unix,
+ .close = p9_fd_close,
+ .request = p9_fd_request,
+ .cancel = p9_fd_cancel,
+ .cancelled = p9_fd_cancelled,
+ .owner = THIS_MODULE,
+};
+
+static struct p9_trans_module p9_fd_trans = {
+ .name = "fd",
+ .maxsize = MAX_SOCK_BUF,
+ .def = 0,
+ .create = p9_fd_create,
+ .close = p9_fd_close,
+ .request = p9_fd_request,
+ .cancel = p9_fd_cancel,
+ .cancelled = p9_fd_cancelled,
+ .owner = THIS_MODULE,
+};
+
+/**
+ * p9_poll_proc - poll worker thread
+ * @a: thread state and arguments
+ *
+ * polls all v9fs transports for new events and queues the appropriate
+ * work to the work queue
+ *
+ */
+
+static void p9_poll_workfn(struct work_struct *work)
+{
+ unsigned long flags;
+
+ p9_debug(P9_DEBUG_TRANS, "start %p\n", current);
+
+ spin_lock_irqsave(&p9_poll_lock, flags);
+ while (!list_empty(&p9_poll_pending_list)) {
+ struct p9_conn *conn = list_first_entry(&p9_poll_pending_list,
+ struct p9_conn,
+ poll_pending_link);
+ list_del_init(&conn->poll_pending_link);
+ spin_unlock_irqrestore(&p9_poll_lock, flags);
+
+ p9_poll_mux(conn);
+
+ spin_lock_irqsave(&p9_poll_lock, flags);
+ }
+ spin_unlock_irqrestore(&p9_poll_lock, flags);
+
+ p9_debug(P9_DEBUG_TRANS, "finish\n");
+}
+
+int p9_trans_fd_init(void)
+{
+ v9fs_register_trans(&p9_tcp_trans);
+ v9fs_register_trans(&p9_unix_trans);
+ v9fs_register_trans(&p9_fd_trans);
+
+ return 0;
+}
+
+void p9_trans_fd_exit(void)
+{
+ flush_work(&p9_poll_work);
+ v9fs_unregister_trans(&p9_tcp_trans);
+ v9fs_unregister_trans(&p9_unix_trans);
+ v9fs_unregister_trans(&p9_fd_trans);
+}
diff --git a/net/9p/trans_rdma.c b/net/9p/trans_rdma.c
new file mode 100644
index 000000000..3533d2a53
--- /dev/null
+++ b/net/9p/trans_rdma.c
@@ -0,0 +1,801 @@
+/*
+ * linux/fs/9p/trans_rdma.c
+ *
+ * RDMA transport layer based on the trans_fd.c implementation.
+ *
+ * Copyright (C) 2008 by Tom Tucker <tom@opengridcomputing.com>
+ * Copyright (C) 2006 by Russ Cox <rsc@swtch.com>
+ * Copyright (C) 2004-2005 by Latchesar Ionkov <lucho@ionkov.net>
+ * 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
+ * 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
+ *
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/in.h>
+#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>
+#include <linux/uaccess.h>
+#include <linux/inet.h>
+#include <linux/idr.h>
+#include <linux/file.h>
+#include <linux/parser.h>
+#include <linux/semaphore.h>
+#include <linux/slab.h>
+#include <net/9p/9p.h>
+#include <net/9p/client.h>
+#include <net/9p/transport.h>
+#include <rdma/ib_verbs.h>
+#include <rdma/rdma_cm.h>
+
+#define P9_PORT 5640
+#define P9_RDMA_SQ_DEPTH 32
+#define P9_RDMA_RQ_DEPTH 32
+#define P9_RDMA_SEND_SGE 4
+#define P9_RDMA_RECV_SGE 4
+#define P9_RDMA_IRD 0
+#define P9_RDMA_ORD 0
+#define P9_RDMA_TIMEOUT 30000 /* 30 seconds */
+#define P9_RDMA_MAXSIZE (1024*1024) /* 1MB */
+
+/**
+ * struct p9_trans_rdma - RDMA transport instance
+ *
+ * @state: tracks the transport state machine for connection setup and tear down
+ * @cm_id: The RDMA CM ID
+ * @pd: Protection Domain pointer
+ * @qp: Queue Pair pointer
+ * @cq: Completion Queue pointer
+ * @dm_mr: DMA Memory Region pointer
+ * @lkey: The local access only memory region key
+ * @timeout: Number of uSecs to wait for connection management events
+ * @sq_depth: The depth of the Send Queue
+ * @sq_sem: Semaphore for the SQ
+ * @rq_depth: The depth of the Receive Queue.
+ * @rq_sem: Semaphore for the RQ
+ * @excess_rc : Amount of posted Receive Contexts without a pending request.
+ * See rdma_request()
+ * @addr: The remote peer's address
+ * @req_lock: Protects the active request list
+ * @cm_done: Completion event for connection management tracking
+ */
+struct p9_trans_rdma {
+ enum {
+ P9_RDMA_INIT,
+ P9_RDMA_ADDR_RESOLVED,
+ P9_RDMA_ROUTE_RESOLVED,
+ P9_RDMA_CONNECTED,
+ P9_RDMA_FLUSHING,
+ P9_RDMA_CLOSING,
+ P9_RDMA_CLOSED,
+ } state;
+ struct rdma_cm_id *cm_id;
+ struct ib_pd *pd;
+ struct ib_qp *qp;
+ struct ib_cq *cq;
+ struct ib_mr *dma_mr;
+ u32 lkey;
+ long timeout;
+ int sq_depth;
+ struct semaphore sq_sem;
+ int rq_depth;
+ struct semaphore rq_sem;
+ atomic_t excess_rc;
+ struct sockaddr_in addr;
+ spinlock_t req_lock;
+
+ struct completion cm_done;
+};
+
+/**
+ * p9_rdma_context - Keeps track of in-process WR
+ *
+ * @wc_op: The original WR op for when the CQE completes in error.
+ * @busa: Bus address to unmap when the WR completes
+ * @req: Keeps track of requests (send)
+ * @rc: Keepts track of replies (receive)
+ */
+struct p9_rdma_req;
+struct p9_rdma_context {
+ enum ib_wc_opcode wc_op;
+ dma_addr_t busa;
+ union {
+ struct p9_req_t *req;
+ struct p9_fcall *rc;
+ };
+};
+
+/**
+ * p9_rdma_opts - Collection of mount options
+ * @port: port of connection
+ * @sq_depth: The requested depth of the SQ. This really doesn't need
+ * to be any deeper than the number of threads used in the client
+ * @rq_depth: The depth of the RQ. Should be greater than or equal to SQ depth
+ * @timeout: Time to wait in msecs for CM events
+ */
+struct p9_rdma_opts {
+ short port;
+ int sq_depth;
+ int rq_depth;
+ long timeout;
+ int privport;
+};
+
+/*
+ * Option Parsing (code inspired by NFS code)
+ */
+enum {
+ /* Options that take integer arguments */
+ Opt_port, Opt_rq_depth, Opt_sq_depth, Opt_timeout,
+ /* Options that take no argument */
+ Opt_privport,
+ Opt_err,
+};
+
+static match_table_t tokens = {
+ {Opt_port, "port=%u"},
+ {Opt_sq_depth, "sq=%u"},
+ {Opt_rq_depth, "rq=%u"},
+ {Opt_timeout, "timeout=%u"},
+ {Opt_privport, "privport"},
+ {Opt_err, NULL},
+};
+
+/**
+ * parse_opts - parse mount options into rdma options structure
+ * @params: options string passed from mount
+ * @opts: rdma transport-specific structure to parse options into
+ *
+ * Returns 0 upon success, -ERRNO upon failure
+ */
+static int parse_opts(char *params, struct p9_rdma_opts *opts)
+{
+ char *p;
+ substring_t args[MAX_OPT_ARGS];
+ int option;
+ char *options, *tmp_options;
+
+ opts->port = P9_PORT;
+ opts->sq_depth = P9_RDMA_SQ_DEPTH;
+ opts->rq_depth = P9_RDMA_RQ_DEPTH;
+ opts->timeout = P9_RDMA_TIMEOUT;
+ opts->privport = 0;
+
+ if (!params)
+ return 0;
+
+ tmp_options = kstrdup(params, GFP_KERNEL);
+ if (!tmp_options) {
+ p9_debug(P9_DEBUG_ERROR,
+ "failed to allocate copy of option string\n");
+ return -ENOMEM;
+ }
+ options = tmp_options;
+
+ while ((p = strsep(&options, ",")) != NULL) {
+ int token;
+ int r;
+ if (!*p)
+ continue;
+ token = match_token(p, tokens, args);
+ if ((token != Opt_err) && (token != Opt_privport)) {
+ r = match_int(&args[0], &option);
+ if (r < 0) {
+ p9_debug(P9_DEBUG_ERROR,
+ "integer field, but no integer?\n");
+ continue;
+ }
+ }
+ switch (token) {
+ case Opt_port:
+ opts->port = option;
+ break;
+ case Opt_sq_depth:
+ opts->sq_depth = option;
+ break;
+ case Opt_rq_depth:
+ opts->rq_depth = option;
+ break;
+ case Opt_timeout:
+ opts->timeout = option;
+ break;
+ case Opt_privport:
+ opts->privport = 1;
+ break;
+ default:
+ continue;
+ }
+ }
+ /* RQ must be at least as large as the SQ */
+ opts->rq_depth = max(opts->rq_depth, opts->sq_depth);
+ kfree(tmp_options);
+ return 0;
+}
+
+static int
+p9_cm_event_handler(struct rdma_cm_id *id, struct rdma_cm_event *event)
+{
+ struct p9_client *c = id->context;
+ struct p9_trans_rdma *rdma = c->trans;
+ switch (event->event) {
+ case RDMA_CM_EVENT_ADDR_RESOLVED:
+ BUG_ON(rdma->state != P9_RDMA_INIT);
+ rdma->state = P9_RDMA_ADDR_RESOLVED;
+ break;
+
+ case RDMA_CM_EVENT_ROUTE_RESOLVED:
+ BUG_ON(rdma->state != P9_RDMA_ADDR_RESOLVED);
+ rdma->state = P9_RDMA_ROUTE_RESOLVED;
+ break;
+
+ case RDMA_CM_EVENT_ESTABLISHED:
+ BUG_ON(rdma->state != P9_RDMA_ROUTE_RESOLVED);
+ rdma->state = P9_RDMA_CONNECTED;
+ break;
+
+ case RDMA_CM_EVENT_DISCONNECTED:
+ if (rdma)
+ rdma->state = P9_RDMA_CLOSED;
+ if (c)
+ c->status = Disconnected;
+ break;
+
+ case RDMA_CM_EVENT_TIMEWAIT_EXIT:
+ break;
+
+ case RDMA_CM_EVENT_ADDR_CHANGE:
+ case RDMA_CM_EVENT_ROUTE_ERROR:
+ case RDMA_CM_EVENT_DEVICE_REMOVAL:
+ case RDMA_CM_EVENT_MULTICAST_JOIN:
+ case RDMA_CM_EVENT_MULTICAST_ERROR:
+ case RDMA_CM_EVENT_REJECTED:
+ case RDMA_CM_EVENT_CONNECT_REQUEST:
+ case RDMA_CM_EVENT_CONNECT_RESPONSE:
+ case RDMA_CM_EVENT_CONNECT_ERROR:
+ case RDMA_CM_EVENT_ADDR_ERROR:
+ case RDMA_CM_EVENT_UNREACHABLE:
+ c->status = Disconnected;
+ rdma_disconnect(rdma->cm_id);
+ break;
+ default:
+ BUG();
+ }
+ complete(&rdma->cm_done);
+ return 0;
+}
+
+static void
+handle_recv(struct p9_client *client, struct p9_trans_rdma *rdma,
+ struct p9_rdma_context *c, enum ib_wc_status status, u32 byte_len)
+{
+ struct p9_req_t *req;
+ int err = 0;
+ int16_t tag;
+
+ req = NULL;
+ ib_dma_unmap_single(rdma->cm_id->device, c->busa, client->msize,
+ DMA_FROM_DEVICE);
+
+ if (status != IB_WC_SUCCESS)
+ goto err_out;
+
+ err = p9_parse_header(c->rc, NULL, NULL, &tag, 1);
+ if (err)
+ goto err_out;
+
+ req = p9_tag_lookup(client, tag);
+ if (!req)
+ goto err_out;
+
+ /* Check that we have not yet received a reply for this request.
+ */
+ if (unlikely(req->rc)) {
+ pr_err("Duplicate reply for request %d", tag);
+ goto err_out;
+ }
+
+ req->rc = c->rc;
+ p9_client_cb(client, req, REQ_STATUS_RCVD);
+
+ return;
+
+ err_out:
+ p9_debug(P9_DEBUG_ERROR, "req %p err %d status %d\n", req, err, status);
+ rdma->state = P9_RDMA_FLUSHING;
+ client->status = Disconnected;
+}
+
+static void
+handle_send(struct p9_client *client, struct p9_trans_rdma *rdma,
+ struct p9_rdma_context *c, enum ib_wc_status status, u32 byte_len)
+{
+ ib_dma_unmap_single(rdma->cm_id->device,
+ c->busa, c->req->tc->size,
+ DMA_TO_DEVICE);
+}
+
+static void qp_event_handler(struct ib_event *event, void *context)
+{
+ p9_debug(P9_DEBUG_ERROR, "QP event %d context %p\n",
+ event->event, context);
+}
+
+static void cq_comp_handler(struct ib_cq *cq, void *cq_context)
+{
+ struct p9_client *client = cq_context;
+ struct p9_trans_rdma *rdma = client->trans;
+ int ret;
+ struct ib_wc wc;
+
+ ib_req_notify_cq(rdma->cq, IB_CQ_NEXT_COMP);
+ while ((ret = ib_poll_cq(cq, 1, &wc)) > 0) {
+ struct p9_rdma_context *c = (void *) (unsigned long) wc.wr_id;
+
+ switch (c->wc_op) {
+ case IB_WC_RECV:
+ handle_recv(client, rdma, c, wc.status, wc.byte_len);
+ up(&rdma->rq_sem);
+ break;
+
+ case IB_WC_SEND:
+ handle_send(client, rdma, c, wc.status, wc.byte_len);
+ up(&rdma->sq_sem);
+ break;
+
+ default:
+ pr_err("unexpected completion type, c->wc_op=%d, wc.opcode=%d, status=%d\n",
+ c->wc_op, wc.opcode, wc.status);
+ break;
+ }
+ kfree(c);
+ }
+}
+
+static void cq_event_handler(struct ib_event *e, void *v)
+{
+ p9_debug(P9_DEBUG_ERROR, "CQ event %d context %p\n", e->event, v);
+}
+
+static void rdma_destroy_trans(struct p9_trans_rdma *rdma)
+{
+ if (!rdma)
+ return;
+
+ if (rdma->dma_mr && !IS_ERR(rdma->dma_mr))
+ ib_dereg_mr(rdma->dma_mr);
+
+ if (rdma->qp && !IS_ERR(rdma->qp))
+ ib_destroy_qp(rdma->qp);
+
+ if (rdma->pd && !IS_ERR(rdma->pd))
+ ib_dealloc_pd(rdma->pd);
+
+ if (rdma->cq && !IS_ERR(rdma->cq))
+ ib_destroy_cq(rdma->cq);
+
+ if (rdma->cm_id && !IS_ERR(rdma->cm_id))
+ rdma_destroy_id(rdma->cm_id);
+
+ kfree(rdma);
+}
+
+static int
+post_recv(struct p9_client *client, struct p9_rdma_context *c)
+{
+ struct p9_trans_rdma *rdma = client->trans;
+ struct ib_recv_wr wr, *bad_wr;
+ struct ib_sge sge;
+
+ c->busa = ib_dma_map_single(rdma->cm_id->device,
+ c->rc->sdata, client->msize,
+ DMA_FROM_DEVICE);
+ if (ib_dma_mapping_error(rdma->cm_id->device, c->busa))
+ goto error;
+
+ sge.addr = c->busa;
+ sge.length = client->msize;
+ sge.lkey = rdma->lkey;
+
+ wr.next = NULL;
+ c->wc_op = IB_WC_RECV;
+ wr.wr_id = (unsigned long) c;
+ wr.sg_list = &sge;
+ wr.num_sge = 1;
+ return ib_post_recv(rdma->qp, &wr, &bad_wr);
+
+ error:
+ p9_debug(P9_DEBUG_ERROR, "EIO\n");
+ return -EIO;
+}
+
+static int rdma_request(struct p9_client *client, struct p9_req_t *req)
+{
+ struct p9_trans_rdma *rdma = client->trans;
+ struct ib_send_wr wr, *bad_wr;
+ struct ib_sge sge;
+ int err = 0;
+ unsigned long flags;
+ struct p9_rdma_context *c = NULL;
+ struct p9_rdma_context *rpl_context = NULL;
+
+ /* When an error occurs between posting the recv and the send,
+ * there will be a receive context posted without a pending request.
+ * Since there is no way to "un-post" it, we remember it and skip
+ * post_recv() for the next request.
+ * So here,
+ * see if we are this `next request' and need to absorb an excess rc.
+ * If yes, then drop and free our own, and do not recv_post().
+ **/
+ if (unlikely(atomic_read(&rdma->excess_rc) > 0)) {
+ if ((atomic_sub_return(1, &rdma->excess_rc) >= 0)) {
+ /* Got one ! */
+ kfree(req->rc);
+ req->rc = NULL;
+ goto dont_need_post_recv;
+ } else {
+ /* We raced and lost. */
+ atomic_inc(&rdma->excess_rc);
+ }
+ }
+
+ /* Allocate an fcall for the reply */
+ rpl_context = kmalloc(sizeof *rpl_context, GFP_NOFS);
+ if (!rpl_context) {
+ err = -ENOMEM;
+ goto recv_error;
+ }
+ rpl_context->rc = req->rc;
+
+ /*
+ * Post a receive buffer for this request. We need to ensure
+ * there is a reply buffer available for every outstanding
+ * request. A flushed request can result in no reply for an
+ * outstanding request, so we must keep a count to avoid
+ * overflowing the RQ.
+ */
+ if (down_interruptible(&rdma->rq_sem)) {
+ err = -EINTR;
+ goto recv_error;
+ }
+
+ err = post_recv(client, rpl_context);
+ if (err) {
+ p9_debug(P9_DEBUG_FCALL, "POST RECV failed\n");
+ goto recv_error;
+ }
+ /* remove posted receive buffer from request structure */
+ req->rc = NULL;
+
+dont_need_post_recv:
+ /* Post the request */
+ c = kmalloc(sizeof *c, GFP_NOFS);
+ if (!c) {
+ err = -ENOMEM;
+ goto send_error;
+ }
+ c->req = req;
+
+ c->busa = ib_dma_map_single(rdma->cm_id->device,
+ c->req->tc->sdata, c->req->tc->size,
+ DMA_TO_DEVICE);
+ if (ib_dma_mapping_error(rdma->cm_id->device, c->busa)) {
+ err = -EIO;
+ goto send_error;
+ }
+
+ sge.addr = c->busa;
+ sge.length = c->req->tc->size;
+ sge.lkey = rdma->lkey;
+
+ wr.next = NULL;
+ c->wc_op = IB_WC_SEND;
+ wr.wr_id = (unsigned long) c;
+ wr.opcode = IB_WR_SEND;
+ wr.send_flags = IB_SEND_SIGNALED;
+ wr.sg_list = &sge;
+ wr.num_sge = 1;
+
+ if (down_interruptible(&rdma->sq_sem)) {
+ err = -EINTR;
+ goto send_error;
+ }
+
+ /* Mark request as `sent' *before* we actually send it,
+ * because doing if after could erase the REQ_STATUS_RCVD
+ * status in case of a very fast reply.
+ */
+ req->status = REQ_STATUS_SENT;
+ err = ib_post_send(rdma->qp, &wr, &bad_wr);
+ if (err)
+ goto send_error;
+
+ /* Success */
+ return 0;
+
+ /* Handle errors that happened during or while preparing the send: */
+ send_error:
+ req->status = REQ_STATUS_ERROR;
+ kfree(c);
+ p9_debug(P9_DEBUG_ERROR, "Error %d in rdma_request()\n", err);
+
+ /* Ach.
+ * We did recv_post(), but not send. We have one recv_post in excess.
+ */
+ atomic_inc(&rdma->excess_rc);
+ return err;
+
+ /* Handle errors that happened during or while preparing post_recv(): */
+ recv_error:
+ kfree(rpl_context);
+ spin_lock_irqsave(&rdma->req_lock, flags);
+ if (rdma->state < P9_RDMA_CLOSING) {
+ rdma->state = P9_RDMA_CLOSING;
+ spin_unlock_irqrestore(&rdma->req_lock, flags);
+ rdma_disconnect(rdma->cm_id);
+ } else
+ spin_unlock_irqrestore(&rdma->req_lock, flags);
+ return err;
+}
+
+static void rdma_close(struct p9_client *client)
+{
+ struct p9_trans_rdma *rdma;
+
+ if (!client)
+ return;
+
+ rdma = client->trans;
+ if (!rdma)
+ return;
+
+ client->status = Disconnected;
+ rdma_disconnect(rdma->cm_id);
+ rdma_destroy_trans(rdma);
+}
+
+/**
+ * alloc_rdma - Allocate and initialize the rdma transport structure
+ * @opts: Mount options structure
+ */
+static struct p9_trans_rdma *alloc_rdma(struct p9_rdma_opts *opts)
+{
+ struct p9_trans_rdma *rdma;
+
+ rdma = kzalloc(sizeof(struct p9_trans_rdma), GFP_KERNEL);
+ if (!rdma)
+ return NULL;
+
+ rdma->sq_depth = opts->sq_depth;
+ rdma->rq_depth = opts->rq_depth;
+ rdma->timeout = opts->timeout;
+ spin_lock_init(&rdma->req_lock);
+ init_completion(&rdma->cm_done);
+ sema_init(&rdma->sq_sem, rdma->sq_depth);
+ sema_init(&rdma->rq_sem, rdma->rq_depth);
+ atomic_set(&rdma->excess_rc, 0);
+
+ return rdma;
+}
+
+static int rdma_cancel(struct p9_client *client, struct p9_req_t *req)
+{
+ /* Nothing to do here.
+ * We will take care of it (if we have to) in rdma_cancelled()
+ */
+ return 1;
+}
+
+/* A request has been fully flushed without a reply.
+ * That means we have posted one buffer in excess.
+ */
+static int rdma_cancelled(struct p9_client *client, struct p9_req_t *req)
+{
+ struct p9_trans_rdma *rdma = client->trans;
+ atomic_inc(&rdma->excess_rc);
+ return 0;
+}
+
+static int p9_rdma_bind_privport(struct p9_trans_rdma *rdma)
+{
+ struct sockaddr_in cl = {
+ .sin_family = AF_INET,
+ .sin_addr.s_addr = htonl(INADDR_ANY),
+ };
+ int port, err = -EINVAL;
+
+ for (port = P9_DEF_MAX_RESVPORT; port >= P9_DEF_MIN_RESVPORT; port--) {
+ cl.sin_port = htons((ushort)port);
+ err = rdma_bind_addr(rdma->cm_id, (struct sockaddr *)&cl);
+ if (err != -EADDRINUSE)
+ break;
+ }
+ return err;
+}
+
+/**
+ * trans_create_rdma - Transport method for creating atransport instance
+ * @client: client instance
+ * @addr: IP address string
+ * @args: Mount options string
+ */
+static int
+rdma_create_trans(struct p9_client *client, const char *addr, char *args)
+{
+ int err;
+ struct p9_rdma_opts opts;
+ struct p9_trans_rdma *rdma;
+ struct rdma_conn_param conn_param;
+ struct ib_qp_init_attr qp_attr;
+ struct ib_device_attr devattr;
+
+ /* Parse the transport specific mount options */
+ err = parse_opts(args, &opts);
+ if (err < 0)
+ return err;
+
+ /* Create and initialize the RDMA transport structure */
+ rdma = alloc_rdma(&opts);
+ if (!rdma)
+ return -ENOMEM;
+
+ /* Create the RDMA CM ID */
+ rdma->cm_id = rdma_create_id(p9_cm_event_handler, client, RDMA_PS_TCP,
+ IB_QPT_RC);
+ if (IS_ERR(rdma->cm_id))
+ goto error;
+
+ /* Associate the client with the transport */
+ client->trans = rdma;
+
+ /* Bind to a privileged port if we need to */
+ if (opts.privport) {
+ err = p9_rdma_bind_privport(rdma);
+ if (err < 0) {
+ pr_err("%s (%d): problem binding to privport: %d\n",
+ __func__, task_pid_nr(current), -err);
+ goto error;
+ }
+ }
+
+ /* Resolve the server's address */
+ rdma->addr.sin_family = AF_INET;
+ rdma->addr.sin_addr.s_addr = in_aton(addr);
+ rdma->addr.sin_port = htons(opts.port);
+ err = rdma_resolve_addr(rdma->cm_id, NULL,
+ (struct sockaddr *)&rdma->addr,
+ rdma->timeout);
+ if (err)
+ goto error;
+ err = wait_for_completion_interruptible(&rdma->cm_done);
+ if (err || (rdma->state != P9_RDMA_ADDR_RESOLVED))
+ goto error;
+
+ /* Resolve the route to the server */
+ err = rdma_resolve_route(rdma->cm_id, rdma->timeout);
+ if (err)
+ goto error;
+ err = wait_for_completion_interruptible(&rdma->cm_done);
+ if (err || (rdma->state != P9_RDMA_ROUTE_RESOLVED))
+ goto error;
+
+ /* Query the device attributes */
+ err = ib_query_device(rdma->cm_id->device, &devattr);
+ if (err)
+ goto error;
+
+ /* Create the Completion Queue */
+ rdma->cq = ib_create_cq(rdma->cm_id->device, cq_comp_handler,
+ cq_event_handler, client,
+ opts.sq_depth + opts.rq_depth + 1, 0);
+ if (IS_ERR(rdma->cq))
+ goto error;
+ ib_req_notify_cq(rdma->cq, IB_CQ_NEXT_COMP);
+
+ /* Create the Protection Domain */
+ rdma->pd = ib_alloc_pd(rdma->cm_id->device);
+ if (IS_ERR(rdma->pd))
+ goto error;
+
+ /* Cache the DMA lkey in the transport */
+ rdma->dma_mr = NULL;
+ if (devattr.device_cap_flags & IB_DEVICE_LOCAL_DMA_LKEY)
+ rdma->lkey = rdma->cm_id->device->local_dma_lkey;
+ else {
+ rdma->dma_mr = ib_get_dma_mr(rdma->pd, IB_ACCESS_LOCAL_WRITE);
+ if (IS_ERR(rdma->dma_mr))
+ goto error;
+ rdma->lkey = rdma->dma_mr->lkey;
+ }
+
+ /* Create the Queue Pair */
+ memset(&qp_attr, 0, sizeof qp_attr);
+ qp_attr.event_handler = qp_event_handler;
+ qp_attr.qp_context = client;
+ qp_attr.cap.max_send_wr = opts.sq_depth;
+ qp_attr.cap.max_recv_wr = opts.rq_depth;
+ qp_attr.cap.max_send_sge = P9_RDMA_SEND_SGE;
+ qp_attr.cap.max_recv_sge = P9_RDMA_RECV_SGE;
+ qp_attr.sq_sig_type = IB_SIGNAL_REQ_WR;
+ qp_attr.qp_type = IB_QPT_RC;
+ qp_attr.send_cq = rdma->cq;
+ qp_attr.recv_cq = rdma->cq;
+ err = rdma_create_qp(rdma->cm_id, rdma->pd, &qp_attr);
+ if (err)
+ goto error;
+ rdma->qp = rdma->cm_id->qp;
+
+ /* Request a connection */
+ memset(&conn_param, 0, sizeof(conn_param));
+ conn_param.private_data = NULL;
+ conn_param.private_data_len = 0;
+ conn_param.responder_resources = P9_RDMA_IRD;
+ conn_param.initiator_depth = P9_RDMA_ORD;
+ err = rdma_connect(rdma->cm_id, &conn_param);
+ if (err)
+ goto error;
+ err = wait_for_completion_interruptible(&rdma->cm_done);
+ if (err || (rdma->state != P9_RDMA_CONNECTED))
+ goto error;
+
+ client->status = Connected;
+
+ return 0;
+
+error:
+ rdma_destroy_trans(rdma);
+ return -ENOTCONN;
+}
+
+static struct p9_trans_module p9_rdma_trans = {
+ .name = "rdma",
+ .maxsize = P9_RDMA_MAXSIZE,
+ .def = 0,
+ .owner = THIS_MODULE,
+ .create = rdma_create_trans,
+ .close = rdma_close,
+ .request = rdma_request,
+ .cancel = rdma_cancel,
+ .cancelled = rdma_cancelled,
+};
+
+/**
+ * p9_trans_rdma_init - Register the 9P RDMA transport driver
+ */
+static int __init p9_trans_rdma_init(void)
+{
+ v9fs_register_trans(&p9_rdma_trans);
+ return 0;
+}
+
+static void __exit p9_trans_rdma_exit(void)
+{
+ v9fs_unregister_trans(&p9_rdma_trans);
+}
+
+module_init(p9_trans_rdma_init);
+module_exit(p9_trans_rdma_exit);
+
+MODULE_AUTHOR("Tom Tucker <tom@opengridcomputing.com>");
+MODULE_DESCRIPTION("RDMA Transport for 9P");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c
new file mode 100644
index 000000000..9dd49ca67
--- /dev/null
+++ b/net/9p/trans_virtio.c
@@ -0,0 +1,776 @@
+/*
+ * The Virtio 9p transport driver
+ *
+ * This is a block based transport driver based on the lguest block driver
+ * code.
+ *
+ * Copyright (C) 2007, 2008 Eric Van Hensbergen, IBM Corporation
+ *
+ * Based on virtio console driver
+ * Copyright (C) 2006, 2007 Rusty Russell, IBM Corporation
+ *
+ * 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
+ *
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/in.h>
+#include <linux/module.h>
+#include <linux/net.h>
+#include <linux/ipv6.h>
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/un.h>
+#include <linux/uaccess.h>
+#include <linux/inet.h>
+#include <linux/idr.h>
+#include <linux/file.h>
+#include <linux/highmem.h>
+#include <linux/slab.h>
+#include <net/9p/9p.h>
+#include <linux/parser.h>
+#include <net/9p/client.h>
+#include <net/9p/transport.h>
+#include <linux/scatterlist.h>
+#include <linux/swap.h>
+#include <linux/virtio.h>
+#include <linux/virtio_9p.h>
+#include "trans_common.h"
+
+#define VIRTQUEUE_NUM 128
+
+/* a single mutex to manage channel initialization and attachment */
+static DEFINE_MUTEX(virtio_9p_lock);
+static DECLARE_WAIT_QUEUE_HEAD(vp_wq);
+static atomic_t vp_pinned = ATOMIC_INIT(0);
+
+/**
+ * struct virtio_chan - per-instance transport information
+ * @initialized: whether the channel is initialized
+ * @inuse: whether the channel is in use
+ * @lock: protects multiple elements within this structure
+ * @client: client instance
+ * @vdev: virtio dev associated with this channel
+ * @vq: virtio queue associated with this channel
+ * @sg: scatter gather list which is used to pack a request (protected?)
+ *
+ * We keep all per-channel information in a structure.
+ * This structure is allocated within the devices dev->mem space.
+ * A pointer to the structure will get put in the transport private.
+ *
+ */
+
+struct virtio_chan {
+ bool inuse;
+
+ spinlock_t lock;
+
+ struct p9_client *client;
+ struct virtio_device *vdev;
+ struct virtqueue *vq;
+ int ring_bufs_avail;
+ wait_queue_head_t *vc_wq;
+ /* This is global limit. Since we don't have a global structure,
+ * will be placing it in each channel.
+ */
+ unsigned long p9_max_pages;
+ /* Scatterlist: can be too big for stack. */
+ struct scatterlist sg[VIRTQUEUE_NUM];
+
+ int tag_len;
+ /*
+ * tag name to identify a mount Non-null terminated
+ */
+ char *tag;
+
+ struct list_head chan_list;
+};
+
+static struct list_head virtio_chan_list;
+
+/* How many bytes left in this page. */
+static unsigned int rest_of_page(void *data)
+{
+ return PAGE_SIZE - ((unsigned long)data % PAGE_SIZE);
+}
+
+/**
+ * p9_virtio_close - reclaim resources of a channel
+ * @client: client instance
+ *
+ * This reclaims a channel by freeing its resources and
+ * reseting its inuse flag.
+ *
+ */
+
+static void p9_virtio_close(struct p9_client *client)
+{
+ struct virtio_chan *chan = client->trans;
+
+ mutex_lock(&virtio_9p_lock);
+ if (chan)
+ chan->inuse = false;
+ mutex_unlock(&virtio_9p_lock);
+}
+
+/**
+ * req_done - callback which signals activity from the server
+ * @vq: virtio queue activity was received on
+ *
+ * This notifies us that the server has triggered some activity
+ * on the virtio channel - most likely a response to request we
+ * sent. Figure out which requests now have responses and wake up
+ * those threads.
+ *
+ * Bugs: could do with some additional sanity checking, but appears to work.
+ *
+ */
+
+static void req_done(struct virtqueue *vq)
+{
+ struct virtio_chan *chan = vq->vdev->priv;
+ struct p9_fcall *rc;
+ unsigned int len;
+ struct p9_req_t *req;
+ unsigned long flags;
+
+ p9_debug(P9_DEBUG_TRANS, ": request done\n");
+
+ 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. */
+ wake_up(chan->vc_wq);
+ p9_debug(P9_DEBUG_TRANS, ": rc %p\n", rc);
+ p9_debug(P9_DEBUG_TRANS, ": lookup tag %d\n", rc->tag);
+ req = p9_tag_lookup(chan->client, rc->tag);
+ p9_client_cb(chan->client, req, REQ_STATUS_RCVD);
+ }
+}
+
+/**
+ * pack_sg_list - pack a scatter gather list from a linear buffer
+ * @sg: scatter/gather list to pack into
+ * @start: which segment of the sg_list to start at
+ * @limit: maximum segment to pack data to
+ * @data: data to pack into scatter/gather list
+ * @count: amount of data to pack into the scatter/gather list
+ *
+ * sg_lists have multiple segments of various sizes. This will pack
+ * arbitrary data into an existing scatter gather list, segmenting the
+ * data as necessary within constraints.
+ *
+ */
+
+static int pack_sg_list(struct scatterlist *sg, int start,
+ int limit, char *data, int count)
+{
+ int s;
+ int index = start;
+
+ while (count) {
+ s = rest_of_page(data);
+ if (s > count)
+ s = count;
+ BUG_ON(index > limit);
+ /* Make sure we don't terminate early. */
+ sg_unmark_end(&sg[index]);
+ sg_set_buf(&sg[index++], data, s);
+ count -= s;
+ data += s;
+ }
+ if (index-start)
+ sg_mark_end(&sg[index - 1]);
+ return index-start;
+}
+
+/* We don't currently allow canceling of virtio requests */
+static int p9_virtio_cancel(struct p9_client *client, struct p9_req_t *req)
+{
+ return 1;
+}
+
+/**
+ * pack_sg_list_p - Just like pack_sg_list. Instead of taking a buffer,
+ * this takes a list of pages.
+ * @sg: scatter/gather list to pack into
+ * @start: which segment of the sg_list to start at
+ * @pdata: a list of pages to add into sg.
+ * @nr_pages: number of pages to pack into the scatter/gather list
+ * @offs: amount of data in the beginning of first page _not_ to pack
+ * @count: amount of data to pack into the scatter/gather list
+ */
+static int
+pack_sg_list_p(struct scatterlist *sg, int start, int limit,
+ struct page **pdata, int nr_pages, size_t offs, int count)
+{
+ int i = 0, s;
+ int data_off = offs;
+ int index = start;
+
+ BUG_ON(nr_pages > (limit - start));
+ /*
+ * if the first page doesn't start at
+ * page boundary find the offset
+ */
+ while (nr_pages) {
+ s = PAGE_SIZE - data_off;
+ if (s > count)
+ s = count;
+ /* Make sure we don't terminate early. */
+ sg_unmark_end(&sg[index]);
+ sg_set_page(&sg[index++], pdata[i++], s, data_off);
+ data_off = 0;
+ count -= s;
+ nr_pages--;
+ }
+
+ if (index-start)
+ sg_mark_end(&sg[index - 1]);
+ return index - start;
+}
+
+/**
+ * p9_virtio_request - issue a request
+ * @client: client instance issuing the request
+ * @req: request to be issued
+ *
+ */
+
+static int
+p9_virtio_request(struct p9_client *client, struct p9_req_t *req)
+{
+ int err;
+ int in, out, out_sgs, in_sgs;
+ unsigned long flags;
+ struct virtio_chan *chan = client->trans;
+ struct scatterlist *sgs[2];
+
+ p9_debug(P9_DEBUG_TRANS, "9p debug: virtio request\n");
+
+ req->status = REQ_STATUS_SENT;
+req_retry:
+ spin_lock_irqsave(&chan->lock, flags);
+
+ out_sgs = in_sgs = 0;
+ /* Handle out VirtIO ring buffers */
+ out = pack_sg_list(chan->sg, 0,
+ VIRTQUEUE_NUM, req->tc->sdata, req->tc->size);
+ if (out)
+ sgs[out_sgs++] = chan->sg;
+
+ in = pack_sg_list(chan->sg, out,
+ VIRTQUEUE_NUM, req->rc->sdata, req->rc->capacity);
+ if (in)
+ sgs[out_sgs + in_sgs++] = chan->sg + out;
+
+ err = virtqueue_add_sgs(chan->vq, sgs, out_sgs, in_sgs, req->tc,
+ GFP_ATOMIC);
+ 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_debug(P9_DEBUG_TRANS, "Retry virtio request\n");
+ goto req_retry;
+ } else {
+ spin_unlock_irqrestore(&chan->lock, flags);
+ p9_debug(P9_DEBUG_TRANS,
+ "virtio rpc add_sgs returned failure\n");
+ return -EIO;
+ }
+ }
+ virtqueue_kick(chan->vq);
+ spin_unlock_irqrestore(&chan->lock, flags);
+
+ p9_debug(P9_DEBUG_TRANS, "virtio request kicked\n");
+ return 0;
+}
+
+static int p9_get_mapped_pages(struct virtio_chan *chan,
+ struct page ***pages,
+ struct iov_iter *data,
+ int count,
+ size_t *offs,
+ int *need_drop)
+{
+ int nr_pages;
+ int err;
+
+ if (!iov_iter_count(data))
+ return 0;
+
+ if (!(data->type & ITER_KVEC)) {
+ int n;
+ /*
+ * 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;
+ }
+ n = iov_iter_get_pages_alloc(data, pages, count, offs);
+ if (n < 0)
+ return n;
+ *need_drop = 1;
+ nr_pages = DIV_ROUND_UP(n + *offs, PAGE_SIZE);
+ atomic_add(nr_pages, &vp_pinned);
+ return n;
+ } else {
+ /* kernel buffer, no need to pin pages */
+ int index;
+ size_t len;
+ void *p;
+
+ /* we'd already checked that it's non-empty */
+ while (1) {
+ len = iov_iter_single_seg_count(data);
+ if (likely(len)) {
+ p = data->kvec->iov_base + data->iov_offset;
+ break;
+ }
+ iov_iter_advance(data, 0);
+ }
+ if (len > count)
+ len = count;
+
+ nr_pages = DIV_ROUND_UP((unsigned long)p + len, PAGE_SIZE) -
+ (unsigned long)p / PAGE_SIZE;
+
+ *pages = kmalloc(sizeof(struct page *) * nr_pages, GFP_NOFS);
+ if (!*pages)
+ return -ENOMEM;
+
+ *need_drop = 0;
+ p -= (*offs = (unsigned long)p % PAGE_SIZE);
+ for (index = 0; index < nr_pages; index++) {
+ if (is_vmalloc_addr(p))
+ (*pages)[index] = vmalloc_to_page(p);
+ else
+ (*pages)[index] = kmap_to_page(p);
+ p += PAGE_SIZE;
+ }
+ return len;
+ }
+}
+
+/**
+ * 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,
+ struct iov_iter *uidata, struct iov_iter *uodata,
+ int inlen, int outlen, int in_hdr_len)
+{
+ int in, out, err, out_sgs, in_sgs;
+ 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;
+ struct scatterlist *sgs[4];
+ size_t offs;
+ int need_drop = 0;
+
+ p9_debug(P9_DEBUG_TRANS, "virtio request\n");
+
+ if (uodata) {
+ int n = p9_get_mapped_pages(chan, &out_pages, uodata,
+ outlen, &offs, &need_drop);
+ if (n < 0)
+ return n;
+ out_nr_pages = DIV_ROUND_UP(n + offs, PAGE_SIZE);
+ if (n != outlen) {
+ __le32 v = cpu_to_le32(n);
+ memcpy(&req->tc->sdata[req->tc->size - 4], &v, 4);
+ outlen = n;
+ }
+ } else if (uidata) {
+ int n = p9_get_mapped_pages(chan, &in_pages, uidata,
+ inlen, &offs, &need_drop);
+ if (n < 0)
+ return n;
+ in_nr_pages = DIV_ROUND_UP(n + offs, PAGE_SIZE);
+ if (n != inlen) {
+ __le32 v = cpu_to_le32(n);
+ memcpy(&req->tc->sdata[req->tc->size - 4], &v, 4);
+ inlen = n;
+ }
+ }
+ req->status = REQ_STATUS_SENT;
+req_retry_pinned:
+ spin_lock_irqsave(&chan->lock, flags);
+
+ out_sgs = in_sgs = 0;
+
+ /* out data */
+ out = pack_sg_list(chan->sg, 0,
+ VIRTQUEUE_NUM, req->tc->sdata, req->tc->size);
+
+ if (out)
+ sgs[out_sgs++] = chan->sg;
+
+ if (out_pages) {
+ sgs[out_sgs++] = chan->sg + out;
+ out += pack_sg_list_p(chan->sg, out, VIRTQUEUE_NUM,
+ out_pages, out_nr_pages, offs, 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)
+ sgs[out_sgs + in_sgs++] = chan->sg + out;
+
+ if (in_pages) {
+ sgs[out_sgs + in_sgs++] = chan->sg + out + in;
+ in += pack_sg_list_p(chan->sg, out + in, VIRTQUEUE_NUM,
+ in_pages, in_nr_pages, offs, inlen);
+ }
+
+ BUG_ON(out_sgs + in_sgs > ARRAY_SIZE(sgs));
+ err = virtqueue_add_sgs(chan->vq, sgs, out_sgs, in_sgs, req->tc,
+ GFP_ATOMIC);
+ 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)
+ goto err_out;
+
+ p9_debug(P9_DEBUG_TRANS, "Retry virtio request\n");
+ goto req_retry_pinned;
+ } else {
+ spin_unlock_irqrestore(&chan->lock, flags);
+ p9_debug(P9_DEBUG_TRANS,
+ "virtio rpc add_sgs returned failure\n");
+ err = -EIO;
+ goto err_out;
+ }
+ }
+ virtqueue_kick(chan->vq);
+ spin_unlock_irqrestore(&chan->lock, flags);
+ p9_debug(P9_DEBUG_TRANS, "virtio request kicked\n");
+ err = wait_event_interruptible(*req->wq,
+ req->status >= REQ_STATUS_RCVD);
+ /*
+ * Non kernel buffers are pinned, unpin them
+ */
+err_out:
+ if (need_drop) {
+ 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,
+ struct device_attribute *attr, char *buf)
+{
+ struct virtio_chan *chan;
+ struct virtio_device *vdev;
+
+ vdev = dev_to_virtio(dev);
+ chan = vdev->priv;
+
+ memcpy(buf, chan->tag, chan->tag_len);
+ buf[chan->tag_len] = 0;
+
+ return chan->tag_len + 1;
+}
+
+static DEVICE_ATTR(mount_tag, 0444, p9_mount_tag_show, NULL);
+
+/**
+ * p9_virtio_probe - probe for existence of 9P virtio channels
+ * @vdev: virtio device to probe
+ *
+ * This probes for existing virtio channels.
+ *
+ */
+
+static int p9_virtio_probe(struct virtio_device *vdev)
+{
+ __u16 tag_len;
+ char *tag;
+ int err;
+ struct virtio_chan *chan;
+
+ if (!vdev->config->get) {
+ dev_err(&vdev->dev, "%s failure: config access disabled\n",
+ __func__);
+ return -EINVAL;
+ }
+
+ chan = kmalloc(sizeof(struct virtio_chan), GFP_KERNEL);
+ if (!chan) {
+ pr_err("Failed to allocate virtio 9P channel\n");
+ err = -ENOMEM;
+ goto fail;
+ }
+
+ chan->vdev = vdev;
+
+ /* We expect one virtqueue, for requests. */
+ chan->vq = virtio_find_single_vq(vdev, req_done, "requests");
+ if (IS_ERR(chan->vq)) {
+ err = PTR_ERR(chan->vq);
+ goto out_free_vq;
+ }
+ chan->vq->vdev->priv = chan;
+ spin_lock_init(&chan->lock);
+
+ sg_init_table(chan->sg, VIRTQUEUE_NUM);
+
+ chan->inuse = false;
+ if (virtio_has_feature(vdev, VIRTIO_9P_MOUNT_TAG)) {
+ virtio_cread(vdev, struct virtio_9p_config, tag_len, &tag_len);
+ } else {
+ err = -EINVAL;
+ goto out_free_vq;
+ }
+ tag = kmalloc(tag_len, GFP_KERNEL);
+ if (!tag) {
+ err = -ENOMEM;
+ goto out_free_vq;
+ }
+
+ virtio_cread_bytes(vdev, offsetof(struct virtio_9p_config, tag),
+ tag, tag_len);
+ chan->tag = tag;
+ chan->tag_len = tag_len;
+ err = sysfs_create_file(&(vdev->dev.kobj), &dev_attr_mount_tag.attr);
+ if (err) {
+ goto out_free_tag;
+ }
+ chan->vc_wq = kmalloc(sizeof(wait_queue_head_t), GFP_KERNEL);
+ if (!chan->vc_wq) {
+ err = -ENOMEM;
+ goto out_free_tag;
+ }
+ init_waitqueue_head(chan->vc_wq);
+ chan->ring_bufs_avail = 1;
+ /* Ceiling limit to avoid denial of service attacks */
+ chan->p9_max_pages = nr_free_buffer_pages()/4;
+
+ virtio_device_ready(vdev);
+
+ mutex_lock(&virtio_9p_lock);
+ list_add_tail(&chan->chan_list, &virtio_chan_list);
+ mutex_unlock(&virtio_9p_lock);
+
+ /* Let udev rules use the new mount_tag attribute. */
+ kobject_uevent(&(vdev->dev.kobj), KOBJ_CHANGE);
+
+ return 0;
+
+out_free_tag:
+ kfree(tag);
+out_free_vq:
+ vdev->config->del_vqs(vdev);
+ kfree(chan);
+fail:
+ return err;
+}
+
+
+/**
+ * p9_virtio_create - allocate a new virtio channel
+ * @client: client instance invoking this transport
+ * @devname: string identifying the channel to connect to (unused)
+ * @args: args passed from sys_mount() for per-transport options (unused)
+ *
+ * This sets up a transport channel for 9p communication. Right now
+ * we only match the first available channel, but eventually we couldlook up
+ * 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 int
+p9_virtio_create(struct p9_client *client, const char *devname, char *args)
+{
+ struct virtio_chan *chan;
+ int ret = -ENOENT;
+ int found = 0;
+
+ mutex_lock(&virtio_9p_lock);
+ list_for_each_entry(chan, &virtio_chan_list, chan_list) {
+ if (!strncmp(devname, chan->tag, chan->tag_len) &&
+ strlen(devname) == chan->tag_len) {
+ if (!chan->inuse) {
+ chan->inuse = true;
+ found = 1;
+ break;
+ }
+ ret = -EBUSY;
+ }
+ }
+ mutex_unlock(&virtio_9p_lock);
+
+ if (!found) {
+ pr_err("no channels available\n");
+ return ret;
+ }
+
+ client->trans = (void *)chan;
+ client->status = Connected;
+ chan->client = client;
+
+ return 0;
+}
+
+/**
+ * p9_virtio_remove - clean up resources associated with a virtio device
+ * @vdev: virtio device to remove
+ *
+ */
+
+static void p9_virtio_remove(struct virtio_device *vdev)
+{
+ struct virtio_chan *chan = vdev->priv;
+ unsigned long warning_time;
+
+ mutex_lock(&virtio_9p_lock);
+
+ /* Remove self from list so we don't get new users. */
+ list_del(&chan->chan_list);
+ warning_time = jiffies;
+
+ /* Wait for existing users to close. */
+ while (chan->inuse) {
+ mutex_unlock(&virtio_9p_lock);
+ msleep(250);
+ if (time_after(jiffies, warning_time + 10 * HZ)) {
+ dev_emerg(&vdev->dev,
+ "p9_virtio_remove: waiting for device in use.\n");
+ warning_time = jiffies;
+ }
+ mutex_lock(&virtio_9p_lock);
+ }
+
+ mutex_unlock(&virtio_9p_lock);
+
+ vdev->config->del_vqs(vdev);
+
+ sysfs_remove_file(&(vdev->dev.kobj), &dev_attr_mount_tag.attr);
+ kobject_uevent(&(vdev->dev.kobj), KOBJ_CHANGE);
+ kfree(chan->tag);
+ kfree(chan->vc_wq);
+ kfree(chan);
+
+}
+
+static struct virtio_device_id id_table[] = {
+ { VIRTIO_ID_9P, VIRTIO_DEV_ANY_ID },
+ { 0 },
+};
+
+static unsigned int features[] = {
+ VIRTIO_9P_MOUNT_TAG,
+};
+
+/* The standard "struct lguest_driver": */
+static struct virtio_driver p9_virtio_drv = {
+ .feature_table = features,
+ .feature_table_size = ARRAY_SIZE(features),
+ .driver.name = KBUILD_MODNAME,
+ .driver.owner = THIS_MODULE,
+ .id_table = id_table,
+ .probe = p9_virtio_probe,
+ .remove = p9_virtio_remove,
+};
+
+static struct p9_trans_module p9_virtio_trans = {
+ .name = "virtio",
+ .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
+ * that are not at page boundary, that can result in an extra
+ * page in zero copy.
+ */
+ .maxsize = PAGE_SIZE * (VIRTQUEUE_NUM - 3),
+ .def = 1,
+ .owner = THIS_MODULE,
+};
+
+/* The standard init function */
+static int __init p9_virtio_init(void)
+{
+ INIT_LIST_HEAD(&virtio_chan_list);
+
+ v9fs_register_trans(&p9_virtio_trans);
+ return register_virtio_driver(&p9_virtio_drv);
+}
+
+static void __exit p9_virtio_cleanup(void)
+{
+ unregister_virtio_driver(&p9_virtio_drv);
+ v9fs_unregister_trans(&p9_virtio_trans);
+}
+
+module_init(p9_virtio_init);
+module_exit(p9_virtio_cleanup);
+
+MODULE_DEVICE_TABLE(virtio, id_table);
+MODULE_AUTHOR("Eric Van Hensbergen <ericvh@gmail.com>");
+MODULE_DESCRIPTION("Virtio 9p Transport");
+MODULE_LICENSE("GPL");
diff --git a/net/9p/util.c b/net/9p/util.c
new file mode 100644
index 000000000..59f278e64
--- /dev/null
+++ b/net/9p/util.c
@@ -0,0 +1,141 @@
+/*
+ * net/9p/util.c
+ *
+ * This file contains some helper functions
+ *
+ * Copyright (C) 2007 by Latchesar Ionkov <lucho@ionkov.net>
+ * Copyright (C) 2004 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
+ * 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/sched.h>
+#include <linux/parser.h>
+#include <linux/idr.h>
+#include <linux/slab.h>
+#include <net/9p/9p.h>
+
+/**
+ * struct p9_idpool - per-connection accounting for tag idpool
+ * @lock: protects the pool
+ * @pool: idr to allocate tag id from
+ *
+ */
+
+struct p9_idpool {
+ spinlock_t lock;
+ struct idr pool;
+};
+
+/**
+ * p9_idpool_create - create a new per-connection id pool
+ *
+ */
+
+struct p9_idpool *p9_idpool_create(void)
+{
+ struct p9_idpool *p;
+
+ p = kmalloc(sizeof(struct p9_idpool), GFP_KERNEL);
+ if (!p)
+ return ERR_PTR(-ENOMEM);
+
+ spin_lock_init(&p->lock);
+ idr_init(&p->pool);
+
+ return p;
+}
+EXPORT_SYMBOL(p9_idpool_create);
+
+/**
+ * p9_idpool_destroy - create a new per-connection id pool
+ * @p: idpool to destroy
+ */
+
+void p9_idpool_destroy(struct p9_idpool *p)
+{
+ idr_destroy(&p->pool);
+ kfree(p);
+}
+EXPORT_SYMBOL(p9_idpool_destroy);
+
+/**
+ * p9_idpool_get - allocate numeric id from pool
+ * @p: pool to allocate from
+ *
+ * Bugs: This seems to be an awful generic function, should it be in idr.c with
+ * the lock included in struct idr?
+ */
+
+int p9_idpool_get(struct p9_idpool *p)
+{
+ int i;
+ unsigned long flags;
+
+ idr_preload(GFP_NOFS);
+ spin_lock_irqsave(&p->lock, flags);
+
+ /* no need to store exactly p, we just need something non-null */
+ i = idr_alloc(&p->pool, p, 0, 0, GFP_NOWAIT);
+
+ spin_unlock_irqrestore(&p->lock, flags);
+ idr_preload_end();
+ if (i < 0)
+ return -1;
+
+ p9_debug(P9_DEBUG_MUX, " id %d pool %p\n", i, p);
+ return i;
+}
+EXPORT_SYMBOL(p9_idpool_get);
+
+/**
+ * p9_idpool_put - release numeric id from pool
+ * @id: numeric id which is being released
+ * @p: pool to release id into
+ *
+ * Bugs: This seems to be an awful generic function, should it be in idr.c with
+ * the lock included in struct idr?
+ */
+
+void p9_idpool_put(int id, struct p9_idpool *p)
+{
+ unsigned long flags;
+
+ p9_debug(P9_DEBUG_MUX, " id %d pool %p\n", id, p);
+
+ spin_lock_irqsave(&p->lock, flags);
+ idr_remove(&p->pool, id);
+ spin_unlock_irqrestore(&p->lock, flags);
+}
+EXPORT_SYMBOL(p9_idpool_put);
+
+/**
+ * p9_idpool_check - check if the specified id is available
+ * @id: id to check
+ * @p: pool to check
+ */
+
+int p9_idpool_check(int id, struct p9_idpool *p)
+{
+ return idr_find(&p->pool, id) != NULL;
+}
+EXPORT_SYMBOL(p9_idpool_check);
+