aboutsummaryrefslogtreecommitdiff
path: root/net/9p
diff options
context:
space:
mode:
Diffstat (limited to 'net/9p')
-rw-r--r--net/9p/Kconfig10
-rw-r--r--net/9p/Makefile1
-rw-r--r--net/9p/client.c977
-rw-r--r--net/9p/error.c10
-rw-r--r--net/9p/mod.c39
-rw-r--r--net/9p/protocol.c136
-rw-r--r--net/9p/protocol.h4
-rw-r--r--net/9p/trans_common.c69
-rw-r--r--net/9p/trans_common.h17
-rw-r--r--net/9p/trans_fd.c386
-rw-r--r--net/9p/trans_rdma.c182
-rw-r--r--net/9p/trans_virtio.c354
-rw-r--r--net/9p/util.c23
13 files changed, 1498 insertions, 710 deletions
diff --git a/net/9p/Kconfig b/net/9p/Kconfig
index 7ed75c7bd5d..a75174a3372 100644
--- a/net/9p/Kconfig
+++ b/net/9p/Kconfig
@@ -3,8 +3,8 @@
#
menuconfig NET_9P
- depends on NET && EXPERIMENTAL
- tristate "Plan 9 Resource Sharing Support (9P2000) (Experimental)"
+ 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.
@@ -16,14 +16,14 @@ menuconfig NET_9P
if NET_9P
config NET_9P_VIRTIO
- depends on EXPERIMENTAL && VIRTIO
- tristate "9P Virtio Transport (Experimental)"
+ 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 && EXPERIMENTAL
+ depends on INET && INFINIBAND && INFINIBAND_ADDR_TRANS
tristate "9P RDMA Transport (Experimental)"
help
This builds support for an RDMA transport.
diff --git a/net/9p/Makefile b/net/9p/Makefile
index 198a640d53a..a0874cc1f71 100644
--- a/net/9p/Makefile
+++ b/net/9p/Makefile
@@ -9,6 +9,7 @@ obj-$(CONFIG_NET_9P_RDMA) += 9pnet_rdma.o
util.o \
protocol.o \
trans_fd.o \
+ trans_common.o \
9pnet_virtio-objs := \
trans_virtio.o \
diff --git a/net/9p/client.c b/net/9p/client.c
index a848bca9fbf..0004cbaac4a 100644
--- a/net/9p/client.c
+++ b/net/9p/client.c
@@ -23,6 +23,8 @@
*
*/
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
#include <linux/module.h>
#include <linux/errno.h>
#include <linux/fs.h>
@@ -38,6 +40,9 @@
#include <net/9p/transport.h>
#include "protocol.h"
+#define CREATE_TRACE_POINTS
+#include <trace/events/9p.h>
+
/*
* Client Option Parsing (code inspired by NFS code)
* - a little lazy - parse all client options
@@ -71,30 +76,40 @@ inline int p9_is_proto_dotu(struct p9_client *clnt)
}
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(const substring_t *name)
+static int get_protocol_version(char *s)
{
int version = -EINVAL;
- if (!strncmp("9p2000", name->from, name->to-name->from)) {
+ if (!strcmp(s, "9p2000")) {
version = p9_proto_legacy;
- P9_DPRINTK(P9_DEBUG_9P, "Protocol version: Legacy\n");
- } else if (!strncmp("9p2000.u", name->from, name->to-name->from)) {
+ p9_debug(P9_DEBUG_9P, "Protocol version: Legacy\n");
+ } else if (!strcmp(s, "9p2000.u")) {
version = p9_proto_2000u;
- P9_DPRINTK(P9_DEBUG_9P, "Protocol version: 9P2000.u\n");
- } else if (!strncmp("9p2000.L", name->from, name->to-name->from)) {
+ p9_debug(P9_DEBUG_9P, "Protocol version: 9P2000.u\n");
+ } else if (!strcmp(s, "9p2000.L")) {
version = p9_proto_2000L;
- P9_DPRINTK(P9_DEBUG_9P, "Protocol version: 9P2000.L\n");
- } else {
- P9_DPRINTK(P9_DEBUG_ERROR, "Unknown protocol version %s. ",
- name->from);
- }
+ p9_debug(P9_DEBUG_9P, "Protocol version: 9P2000.L\n");
+ } else
+ pr_info("Unknown protocol version %s\n", s);
+
return version;
}
-static struct p9_req_t *
-p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...);
-
/**
* parse_options - parse mount options into client structure
* @opts: options string passed from mount
@@ -109,9 +124,10 @@ static int parse_opts(char *opts, struct p9_client *clnt)
char *p;
substring_t args[MAX_OPT_ARGS];
int option;
+ char *s;
int ret = 0;
- clnt->proto_version = p9_proto_2000u;
+ clnt->proto_version = p9_proto_2000L;
clnt->msize = 8192;
if (!opts)
@@ -119,47 +135,63 @@ static int parse_opts(char *opts, struct p9_client *clnt)
tmp_options = kstrdup(opts, GFP_KERNEL);
if (!tmp_options) {
- P9_DPRINTK(P9_DEBUG_ERROR,
- "failed to allocate copy of option string\n");
+ 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 token, r;
if (!*p)
continue;
token = match_token(p, tokens, args);
- if (token < Opt_trans) {
- int r = match_int(&args[0], &option);
+ switch (token) {
+ case Opt_msize:
+ r = match_int(&args[0], &option);
if (r < 0) {
- P9_DPRINTK(P9_DEBUG_ERROR,
- "integer field, but no integer?\n");
+ p9_debug(P9_DEBUG_ERROR,
+ "integer field, but no integer?\n");
ret = r;
continue;
}
- }
- switch (token) {
- case Opt_msize:
clnt->msize = option;
break;
case Opt_trans:
- clnt->trans_mod = v9fs_get_trans_by_name(&args[0]);
- if(clnt->trans_mod == NULL) {
- P9_DPRINTK(P9_DEBUG_ERROR,
- "Could not find request transport: %s\n",
- (char *) &args[0]);
+ 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:
- ret = get_protocol_version(&args[0]);
- if (ret == -EINVAL)
+ 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:
@@ -172,13 +204,24 @@ free_and_return:
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 accomodate transaction
+ * 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
@@ -187,11 +230,13 @@ free_and_return:
*
*/
-static struct p9_req_t *p9_tag_alloc(struct p9_client *c, u16 tag)
+static struct p9_req_t *
+p9_tag_alloc(struct p9_client *c, u16 tag, unsigned int max_size)
{
unsigned long flags;
int row, col;
struct p9_req_t *req;
+ int alloc_msize = min(c->msize, max_size);
/* This looks up the original request by tag so we know which
* buffer to read the data into */
@@ -206,7 +251,7 @@ static struct p9_req_t *p9_tag_alloc(struct p9_client *c, u16 tag)
sizeof(struct p9_req_t), GFP_ATOMIC);
if (!c->reqs[row]) {
- printk(KERN_ERR "Couldn't grow tag array\n");
+ pr_err("Couldn't grow tag array\n");
spin_unlock_irqrestore(&c->lock, flags);
return ERR_PTR(-ENOMEM);
}
@@ -222,39 +267,36 @@ static struct p9_req_t *p9_tag_alloc(struct p9_client *c, u16 tag)
col = tag % P9_ROW_MAXTAG;
req = &c->reqs[row][col];
- if (!req->tc) {
- req->wq = kmalloc(sizeof(wait_queue_head_t), GFP_KERNEL);
- if (!req->wq) {
- printk(KERN_ERR "Couldn't grow tag array\n");
- return ERR_PTR(-ENOMEM);
- }
+ if (!req->wq) {
+ req->wq = kmalloc(sizeof(wait_queue_head_t), GFP_NOFS);
+ if (!req->wq)
+ goto grow_failed;
init_waitqueue_head(req->wq);
- req->tc = kmalloc(sizeof(struct p9_fcall)+c->msize,
- GFP_KERNEL);
- req->rc = kmalloc(sizeof(struct p9_fcall)+c->msize,
- GFP_KERNEL);
- if ((!req->tc) || (!req->rc)) {
- printk(KERN_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);
- }
- req->tc->sdata = (char *) req->tc + sizeof(struct p9_fcall);
- req->tc->capacity = c->msize;
- req->rc->sdata = (char *) req->rc + sizeof(struct p9_fcall);
- req->rc->capacity = c->msize;
}
+ 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 &c->reqs[row][col];
+ 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);
}
/**
@@ -272,7 +314,8 @@ struct p9_req_t *p9_tag_lookup(struct p9_client *c, u16 tag)
* buffer to read the data into */
tag++;
- BUG_ON(tag >= c->max_tag);
+ if(tag >= c->max_tag)
+ return NULL;
row = tag / P9_ROW_MAXTAG;
col = tag % P9_ROW_MAXTAG;
@@ -296,12 +339,13 @@ static int p9_tag_init(struct p9_client *c)
c->tagpool = p9_idpool_create();
if (IS_ERR(c->tagpool)) {
err = PTR_ERR(c->tagpool);
- c->tagpool = NULL;
goto error;
}
-
- p9_idpool_get(c->tagpool); /* reserve tag 0 */
-
+ 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;
@@ -322,9 +366,9 @@ static void p9_tag_cleanup(struct p9_client *c)
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_DPRINTK(P9_DEBUG_MUX,
- "Attempting to cleanup non-free tag %d,%d\n",
- row, col);
+ p9_debug(P9_DEBUG_MUX,
+ "Attempting to cleanup non-free tag %d,%d\n",
+ row, col);
/* TODO: delay execution of cleanup */
return;
}
@@ -358,7 +402,7 @@ static void p9_tag_cleanup(struct p9_client *c)
static void p9_free_req(struct p9_client *c, struct p9_req_t *r)
{
int tag = r->tc->tag;
- P9_DPRINTK(P9_DEBUG_MUX, "clnt %p req %p tag: %d\n", c, r, 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))
@@ -371,11 +415,19 @@ static void p9_free_req(struct p9_client *c, struct p9_req_t *r)
* req: request received
*
*/
-void p9_client_cb(struct p9_client *c, struct p9_req_t *req)
+void p9_client_cb(struct p9_client *c, struct p9_req_t *req, int status)
{
- P9_DPRINTK(P9_DEBUG_MUX, " tag %d\n", req->tc->tag);
+ 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_DPRINTK(P9_DEBUG_MUX, "wakeup: %d\n", req->tc->tag);
+ p9_debug(P9_DEBUG_MUX, "wakeup: %d\n", req->tc->tag);
}
EXPORT_SYMBOL(p9_client_cb);
@@ -410,8 +462,8 @@ p9_parse_header(struct p9_fcall *pdu, int32_t *size, int8_t *type, int16_t *tag,
pdu->id = r_type;
pdu->tag = r_tag;
- P9_DPRINTK(P9_DEBUG_9P, "<<< size=%d type: %d tag: %d\n", pdu->size,
- pdu->id, pdu->tag);
+ p9_debug(P9_DEBUG_9P, "<<< size=%d type: %d tag: %d\n",
+ pdu->size, pdu->id, pdu->tag);
if (type)
*type = r_type;
@@ -443,58 +495,152 @@ 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_DPRINTK(P9_DEBUG_ERROR, "couldn't parse header %d\n", 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 (type == P9_RERROR || type == P9_RLERROR) {
- int ecode;
+ 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
+ */
- if (!p9_is_proto_dotl(c)) {
- char *ename;
+static int p9_check_zc_errors(struct p9_client *c, struct p9_req_t *req,
+ char *uidata, int in_hdrlen, int kern_buf)
+{
+ int err;
+ int ecode;
+ int8_t type;
+ char *ename = NULL;
- err = p9pdu_readf(req->rc, c->proto_version, "s?d",
- &ename, &ecode);
- if (err)
- goto out_err;
+ 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 (p9_is_proto_dotu(c))
- err = -ecode;
+ if (type != P9_RERROR && type != P9_RLERROR)
+ return 0;
- if (!err || !IS_ERR_VALUE(err)) {
- err = p9_errstr2errno(ename, strlen(ename));
+ 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;
- P9_DPRINTK(P9_DEBUG_9P, "<<< RERROR (%d) %s\n", -ecode, ename);
+ len = req->rc->size - req->rc->offset;
+ if (len > (P9_ZC_HDR_SZ - 7)) {
+ err = -EFAULT;
+ goto out_err;
+ }
- kfree(ename);
+ ename = &req->rc->sdata[req->rc->offset];
+ if (len > inline_len) {
+ /* We have error in external buffer */
+ if (kern_buf) {
+ memcpy(ename + inline_len, uidata,
+ len - inline_len);
+ } else {
+ err = copy_from_user(ename + inline_len,
+ uidata, len - inline_len);
+ if (err) {
+ err = -EFAULT;
+ goto out_err;
+ }
}
- } else {
- err = p9pdu_readf(req->rc, c->proto_version, "d", &ecode);
+ }
+ 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;
- P9_DPRINTK(P9_DEBUG_9P, "<<< RLERROR (%d)\n", -ecode);
- }
+ if (!err || !IS_ERR_VALUE(err)) {
+ err = p9_errstr2errno(ename, strlen(ename));
- } else
- err = 0;
+ 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_DPRINTK(P9_DEBUG_ERROR, "couldn't parse error%d\n", 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 requests and links
+ * 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.
@@ -511,43 +657,32 @@ static int p9_client_flush(struct p9_client *c, struct p9_req_t *oldreq)
if (err)
return err;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TFLUSH tag %d\n", oldtag);
+ 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. */
- spin_lock(&c->lock);
- if (oldreq->status == REQ_STATUS_FLSH)
- list_del(&oldreq->req_list);
- spin_unlock(&c->lock);
+ /*
+ * 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;
}
-/**
- * p9_client_rpc - issue a request and wait for a response
- * @c: client session
- * @type: type of request
- * @fmt: protocol format string (see protocol.c)
- *
- * Returns request structure (which client must free using p9_free_req)
- */
-
-static struct p9_req_t *
-p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...)
+static struct p9_req_t *p9_client_prepare_req(struct p9_client *c,
+ int8_t type, int req_size,
+ const char *fmt, va_list ap)
{
- va_list ap;
int tag, err;
struct p9_req_t *req;
- unsigned long flags;
- int sigpending;
- P9_DPRINTK(P9_DEBUG_MUX, "client %p op %d\n", c, type);
+ p9_debug(P9_DEBUG_MUX, "client %p op %d\n", c, type);
/* we allow for any status other than disconnected */
if (c->status == Disconnected)
@@ -557,12 +692,6 @@ p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...)
if ((c->status == BeginDisconnect) && (type != P9_TCLUNK))
return ERR_PTR(-EIO);
- if (signal_pending(current)) {
- sigpending = 1;
- clear_thread_flag(TIF_SIGPENDING);
- } else
- sigpending = 0;
-
tag = P9_NOTAG;
if (type != P9_TVERSION) {
tag = p9_idpool_get(c->tagpool);
@@ -570,39 +699,82 @@ p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...)
return ERR_PTR(-ENOMEM);
}
- req = p9_tag_alloc(c, tag);
+ req = p9_tag_alloc(c, tag, req_size);
if (IS_ERR(req))
return req;
/* marshall the data */
p9pdu_prepare(req->tc, tag, type);
- va_start(ap, fmt);
err = p9pdu_vwritef(req->tc, c->proto_version, fmt, ap);
- va_end(ap);
if (err)
goto reterr;
- p9pdu_finalize(req->tc);
+ p9pdu_finalize(c, req->tc);
+ trace_9p_client_req(c, type, tag);
+ return req;
+reterr:
+ p9_free_req(c, req);
+ return ERR_PTR(err);
+}
+
+/**
+ * p9_client_rpc - issue a request and wait for a response
+ * @c: client session
+ * @type: type of request
+ * @fmt: protocol format string (see protocol.c)
+ *
+ * Returns request structure (which client must free using p9_free_req)
+ */
+
+static struct p9_req_t *
+p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...)
+{
+ va_list ap;
+ int sigpending, err;
+ unsigned long flags;
+ struct p9_req_t *req;
+
+ va_start(ap, fmt);
+ req = p9_client_prepare_req(c, type, c->msize, fmt, ap);
+ va_end(ap);
+ if (IS_ERR(req))
+ return req;
+
+ if (signal_pending(current)) {
+ sigpending = 1;
+ clear_thread_flag(TIF_SIGPENDING);
+ } else
+ sigpending = 0;
err = c->trans_mod->request(c, req);
if (err < 0) {
- if (err != -ERESTARTSYS)
+ if (err != -ERESTARTSYS && err != -EFAULT)
c->status = Disconnected;
goto reterr;
}
-
- P9_DPRINTK(P9_DEBUG_MUX, "wait %p tag: %d\n", req->wq, tag);
+again:
+ /* Wait for the response */
err = wait_event_interruptible(*req->wq,
- req->status >= REQ_STATUS_RCVD);
- P9_DPRINTK(P9_DEBUG_MUX, "wait %p tag: %d returned %d\n",
- req->wq, tag, err);
+ req->status >= REQ_STATUS_RCVD);
+
+ /*
+ * 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_DPRINTK(P9_DEBUG_ERROR, "req_status error %d\n", req->t_err);
+ p9_debug(P9_DEBUG_ERROR, "req_status error %d\n", req->t_err);
err = req->t_err;
}
-
if ((err == -ERESTARTSYS) && (c->status == Connected)) {
- P9_DPRINTK(P9_DEBUG_MUX, "flushing\n");
+ p9_debug(P9_DEBUG_MUX, "flushing\n");
sigpending = 1;
clear_thread_flag(TIF_SIGPENDING);
@@ -613,27 +785,104 @@ p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...)
if (req->status == REQ_STATUS_RCVD)
err = 0;
}
-
if (sigpending) {
spin_lock_irqsave(&current->sighand->siglock, flags);
recalc_sigpending();
spin_unlock_irqrestore(&current->sighand->siglock, flags);
}
-
if (err < 0)
goto reterr;
err = p9_check_errors(c, req);
- if (!err) {
- P9_DPRINTK(P9_DEBUG_MUX, "exit: client %p op %d\n", c, type);
+ trace_9p_client_res(c, type, req->rc->tag, err);
+ if (!err)
+ return req;
+reterr:
+ p9_free_req(c, req);
+ return ERR_PTR(safe_errno(err));
+}
+
+/**
+ * p9_client_zc_rpc - issue a request and wait for a response
+ * @c: client session
+ * @type: type of request
+ * @uidata: user bffer that should be ued for zero copy read
+ * @uodata: user buffer that shoud be user for zero copy write
+ * @inlen: read buffer size
+ * @olen: write buffer size
+ * @hdrlen: reader header size, This is the size of response protocol data
+ * @fmt: protocol format string (see protocol.c)
+ *
+ * Returns request structure (which client must free using p9_free_req)
+ */
+static struct p9_req_t *p9_client_zc_rpc(struct p9_client *c, int8_t type,
+ char *uidata, char *uodata,
+ int inlen, int olen, int in_hdrlen,
+ int kern_buf, const char *fmt, ...)
+{
+ va_list ap;
+ int sigpending, err;
+ unsigned long flags;
+ struct p9_req_t *req;
+
+ va_start(ap, fmt);
+ /*
+ * We allocate a inline protocol data of only 4k bytes.
+ * The actual content is passed in zero-copy fashion.
+ */
+ req = p9_client_prepare_req(c, type, P9_ZC_HDR_SZ, fmt, ap);
+ va_end(ap);
+ if (IS_ERR(req))
return req;
+
+ if (signal_pending(current)) {
+ sigpending = 1;
+ clear_thread_flag(TIF_SIGPENDING);
+ } else
+ sigpending = 0;
+
+ /* If we are called with KERNEL_DS force kern_buf */
+ if (segment_eq(get_fs(), KERNEL_DS))
+ kern_buf = 1;
+
+ err = c->trans_mod->zc_request(c, req, uidata, uodata,
+ inlen, olen, in_hdrlen, kern_buf);
+ if (err < 0) {
+ if (err == -EIO)
+ c->status = Disconnected;
+ goto reterr;
}
+ if (req->status == REQ_STATUS_ERROR) {
+ p9_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, kern_buf);
+ trace_9p_client_res(c, type, req->rc->tag, err);
+ if (!err)
+ return req;
reterr:
- P9_DPRINTK(P9_DEBUG_MUX, "exit: client %p op %d error: %d\n", c, type,
- err);
p9_free_req(c, req);
- return ERR_PTR(err);
+ return ERR_PTR(safe_errno(err));
}
static struct p9_fid *p9_fid_create(struct p9_client *clnt)
@@ -642,7 +891,7 @@ static struct p9_fid *p9_fid_create(struct p9_client *clnt)
struct p9_fid *fid;
unsigned long flags;
- P9_DPRINTK(P9_DEBUG_FID, "clnt %p\n", clnt);
+ p9_debug(P9_DEBUG_FID, "clnt %p\n", clnt);
fid = kmalloc(sizeof(struct p9_fid), GFP_KERNEL);
if (!fid)
return ERR_PTR(-ENOMEM);
@@ -675,7 +924,7 @@ static void p9_fid_destroy(struct p9_fid *fid)
struct p9_client *clnt;
unsigned long flags;
- P9_DPRINTK(P9_DEBUG_FID, "fid %d\n", fid->fid);
+ 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);
@@ -692,8 +941,8 @@ static int p9_client_version(struct p9_client *c)
char *version;
int msize;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TVERSION msize %d protocol %d\n",
- c->msize, c->proto_version);
+ p9_debug(P9_DEBUG_9P, ">>> TVERSION msize %d protocol %d\n",
+ c->msize, c->proto_version);
switch (c->proto_version) {
case p9_proto_2000L:
@@ -718,12 +967,12 @@ static int p9_client_version(struct p9_client *c)
err = p9pdu_readf(req->rc, c->proto_version, "ds", &msize, &version);
if (err) {
- P9_DPRINTK(P9_DEBUG_9P, "version error %d\n", err);
- p9pdu_dump(1, req->rc);
+ p9_debug(P9_DEBUG_9P, "version error %d\n", err);
+ trace_9p_protocol_dump(c, req->rc);
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RVERSION msize %d %s\n", msize, version);
+ 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))
@@ -749,6 +998,7 @@ 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);
@@ -757,41 +1007,46 @@ struct p9_client *p9_client_create(const char *dev_name, char *options)
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);
- p9_tag_init(clnt);
+ err = p9_tag_init(clnt);
+ if (err < 0)
+ goto free_client;
err = parse_opts(options, clnt);
if (err < 0)
- goto free_client;
+ goto destroy_tagpool;
if (!clnt->trans_mod)
clnt->trans_mod = v9fs_get_default_trans();
if (clnt->trans_mod == NULL) {
err = -EPROTONOSUPPORT;
- P9_DPRINTK(P9_DEBUG_ERROR,
- "No transport defined or default transport\n");
- goto free_client;
+ 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);
- clnt->fidpool = NULL;
goto put_trans;
}
- P9_DPRINTK(P9_DEBUG_MUX, "clnt %p trans %p msize %d protocol %d\n",
- clnt, clnt->trans_mod, clnt->msize, clnt->proto_version);
+ 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+P9_IOHDRSZ) > clnt->trans_mod->maxsize)
- clnt->msize = clnt->trans_mod->maxsize-P9_IOHDRSZ;
+ if (clnt->msize > clnt->trans_mod->maxsize)
+ clnt->msize = clnt->trans_mod->maxsize;
err = p9_client_version(clnt);
if (err)
@@ -805,6 +1060,8 @@ 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);
@@ -815,7 +1072,7 @@ void p9_client_destroy(struct p9_client *clnt)
{
struct p9_fid *fid, *fidptr;
- P9_DPRINTK(P9_DEBUG_MUX, "clnt %p\n", clnt);
+ p9_debug(P9_DEBUG_MUX, "clnt %p\n", clnt);
if (clnt->trans_mod)
clnt->trans_mod->close(clnt);
@@ -823,7 +1080,7 @@ void p9_client_destroy(struct p9_client *clnt)
v9fs_put_trans(clnt->trans_mod);
list_for_each_entry_safe(fid, fidptr, &clnt->fidlist, flist) {
- printk(KERN_INFO "Found fid %d not clunked\n", fid->fid);
+ pr_info("Found fid %d not clunked\n", fid->fid);
p9_fid_destroy(fid);
}
@@ -838,30 +1095,29 @@ EXPORT_SYMBOL(p9_client_destroy);
void p9_client_disconnect(struct p9_client *clnt)
{
- P9_DPRINTK(P9_DEBUG_9P, "clnt %p\n", 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_DPRINTK(P9_DEBUG_9P, "clnt %p\n", 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, u32 n_uname, char *aname)
+ char *uname, kuid_t n_uname, char *aname)
{
- int err;
+ int err = 0;
struct p9_req_t *req;
struct p9_fid *fid;
struct p9_qid qid;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TATTACH afid %d uname %s aname %s\n",
- afid ? afid->fid : -1, uname, aname);
- err = 0;
+ p9_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);
@@ -869,7 +1125,7 @@ struct p9_fid *p9_client_attach(struct p9_client *clnt, struct p9_fid *afid,
goto error;
}
- req = p9_client_rpc(clnt, P9_TATTACH, "ddss?d", fid->fid,
+ 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);
@@ -878,15 +1134,13 @@ struct p9_fid *p9_client_attach(struct p9_client *clnt, struct p9_fid *afid,
err = p9pdu_readf(req->rc, clnt->proto_version, "Q", &qid);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
p9_free_req(clnt, req);
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RATTACH qid %x.%llx.%x\n",
- qid.type,
- (unsigned long long)qid.path,
- qid.version);
+ 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));
@@ -900,15 +1154,15 @@ error:
}
EXPORT_SYMBOL(p9_client_attach);
-struct p9_fid *p9_client_walk(struct p9_fid *oldfid, int nwname, char **wnames,
- int clone)
+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;
- int16_t nwqids, count;
+ uint16_t nwqids, count;
err = 0;
wqids = NULL;
@@ -926,8 +1180,8 @@ struct p9_fid *p9_client_walk(struct p9_fid *oldfid, int nwname, char **wnames,
fid = oldfid;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TWALK fids %d,%d nwname %d wname[0] %s\n",
- oldfid->fid, fid->fid, nwname, wnames ? wnames[0] : NULL);
+ 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);
@@ -938,13 +1192,13 @@ struct p9_fid *p9_client_walk(struct p9_fid *oldfid, int nwname, char **wnames,
err = p9pdu_readf(req->rc, clnt->proto_version, "R", &nwqids, &wqids);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
p9_free_req(clnt, req);
goto clunk_fid;
}
p9_free_req(clnt, req);
- P9_DPRINTK(P9_DEBUG_9P, "<<< RWALK nwqid %d:\n", nwqids);
+ p9_debug(P9_DEBUG_9P, "<<< RWALK nwqid %d:\n", nwqids);
if (nwqids != nwname) {
err = -ENOENT;
@@ -952,7 +1206,7 @@ struct p9_fid *p9_client_walk(struct p9_fid *oldfid, int nwname, char **wnames,
}
for (count = 0; count < nwqids; count++)
- P9_DPRINTK(P9_DEBUG_9P, "<<< [%d] %x.%llx.%x\n",
+ p9_debug(P9_DEBUG_9P, "<<< [%d] %x.%llx.%x\n",
count, wqids[count].type,
(unsigned long long)wqids[count].path,
wqids[count].version);
@@ -987,7 +1241,7 @@ int p9_client_open(struct p9_fid *fid, int mode)
int iounit;
clnt = fid->clnt;
- P9_DPRINTK(P9_DEBUG_9P, ">>> %s fid %d mode %d\n",
+ p9_debug(P9_DEBUG_9P, ">>> %s fid %d mode %d\n",
p9_is_proto_dotl(clnt) ? "TLOPEN" : "TOPEN", fid->fid, mode);
err = 0;
@@ -1005,11 +1259,11 @@ int p9_client_open(struct p9_fid *fid, int mode)
err = p9pdu_readf(req->rc, clnt->proto_version, "Qd", &qid, &iounit);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
goto free_and_error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< %s qid %x.%llx.%x iounit %x\n",
+ 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);
@@ -1024,22 +1278,23 @@ error:
EXPORT_SYMBOL(p9_client_open);
int p9_client_create_dotl(struct p9_fid *ofid, char *name, u32 flags, u32 mode,
- gid_t gid, struct p9_qid *qid)
+ kgid_t gid, struct p9_qid *qid)
{
int err = 0;
struct p9_client *clnt;
struct p9_req_t *req;
int iounit;
- P9_DPRINTK(P9_DEBUG_9P,
+ p9_debug(P9_DEBUG_9P,
">>> TLCREATE fid %d name %s flags %d mode %d gid %d\n",
- ofid->fid, name, flags, mode, gid);
+ 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, "dsddd", ofid->fid, name, flags,
+ req = p9_client_rpc(clnt, P9_TLCREATE, "dsddg", ofid->fid, name, flags,
mode, gid);
if (IS_ERR(req)) {
err = PTR_ERR(req);
@@ -1048,11 +1303,11 @@ int p9_client_create_dotl(struct p9_fid *ofid, char *name, u32 flags, u32 mode,
err = p9pdu_readf(req->rc, clnt->proto_version, "Qd", qid, &iounit);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
goto free_and_error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RLCREATE qid %x.%llx.%x iounit %x\n",
+ p9_debug(P9_DEBUG_9P, "<<< RLCREATE qid %x.%llx.%x iounit %x\n",
qid->type,
(unsigned long long)qid->path,
qid->version, iounit);
@@ -1076,7 +1331,7 @@ int p9_client_fcreate(struct p9_fid *fid, char *name, u32 perm, int mode,
struct p9_qid qid;
int iounit;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TCREATE fid %d name %s perm %d mode %d\n",
+ 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;
@@ -1093,11 +1348,11 @@ int p9_client_fcreate(struct p9_fid *fid, char *name, u32 perm, int mode,
err = p9pdu_readf(req->rc, clnt->proto_version, "Qd", &qid, &iounit);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
goto free_and_error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RCREATE qid %x.%llx.%x iounit %x\n",
+ p9_debug(P9_DEBUG_9P, "<<< RCREATE qid %x.%llx.%x iounit %x\n",
qid.type,
(unsigned long long)qid.path,
qid.version, iounit);
@@ -1112,18 +1367,18 @@ error:
}
EXPORT_SYMBOL(p9_client_fcreate);
-int p9_client_symlink(struct p9_fid *dfid, char *name, char *symtgt, gid_t gid,
+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_DPRINTK(P9_DEBUG_9P, ">>> TSYMLINK dfid %d name %s symtgt %s\n",
+ 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, "dssd", dfid->fid, name, symtgt,
+ req = p9_client_rpc(clnt, P9_TSYMLINK, "dssg", dfid->fid, name, symtgt,
gid);
if (IS_ERR(req)) {
err = PTR_ERR(req);
@@ -1132,11 +1387,11 @@ int p9_client_symlink(struct p9_fid *dfid, char *name, char *symtgt, gid_t gid,
err = p9pdu_readf(req->rc, clnt->proto_version, "Q", qid);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
goto free_and_error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RSYMLINK qid %x.%llx.%x\n",
+ p9_debug(P9_DEBUG_9P, "<<< RSYMLINK qid %x.%llx.%x\n",
qid->type, (unsigned long long)qid->path, qid->version);
free_and_error:
@@ -1151,7 +1406,7 @@ int p9_client_link(struct p9_fid *dfid, struct p9_fid *oldfid, char *newname)
struct p9_client *clnt;
struct p9_req_t *req;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TLINK dfid %d oldfid %d newname %s\n",
+ 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,
@@ -1159,7 +1414,7 @@ int p9_client_link(struct p9_fid *dfid, struct p9_fid *oldfid, char *newname)
if (IS_ERR(req))
return PTR_ERR(req);
- P9_DPRINTK(P9_DEBUG_9P, "<<< RLINK\n");
+ p9_debug(P9_DEBUG_9P, "<<< RLINK\n");
p9_free_req(clnt, req);
return 0;
}
@@ -1171,7 +1426,7 @@ int p9_client_fsync(struct p9_fid *fid, int datasync)
struct p9_client *clnt;
struct p9_req_t *req;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TFSYNC fid %d datasync:%d\n",
+ p9_debug(P9_DEBUG_9P, ">>> TFSYNC fid %d datasync:%d\n",
fid->fid, datasync);
err = 0;
clnt = fid->clnt;
@@ -1182,7 +1437,7 @@ int p9_client_fsync(struct p9_fid *fid, int datasync)
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RFSYNC fid %d\n", fid->fid);
+ p9_debug(P9_DEBUG_9P, "<<< RFSYNC fid %d\n", fid->fid);
p9_free_req(clnt, req);
@@ -1196,14 +1451,18 @@ int p9_client_clunk(struct p9_fid *fid)
int err;
struct p9_client *clnt;
struct p9_req_t *req;
+ int retries = 0;
if (!fid) {
- P9_EPRINTK(KERN_WARNING, "Trying to clunk with NULL fid\n");
+ pr_warn("%s (%d): Trying to clunk with NULL fid\n",
+ __func__, task_pid_nr(current));
dump_stack();
return 0;
}
- P9_DPRINTK(P9_DEBUG_9P, ">>> TCLUNK fid %d\n", fid->fid);
+again:
+ p9_debug(P9_DEBUG_9P, ">>> TCLUNK fid %d (try %d)\n", fid->fid,
+ retries);
err = 0;
clnt = fid->clnt;
@@ -1213,12 +1472,20 @@ int p9_client_clunk(struct p9_fid *fid)
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RCLUNK fid %d\n", fid->fid);
+ p9_debug(P9_DEBUG_9P, "<<< RCLUNK fid %d\n", fid->fid);
p9_free_req(clnt, req);
- p9_fid_destroy(fid);
-
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);
@@ -1229,7 +1496,7 @@ int p9_client_remove(struct p9_fid *fid)
struct p9_client *clnt;
struct p9_req_t *req;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TREMOVE fid %d\n", fid->fid);
+ p9_debug(P9_DEBUG_9P, ">>> TREMOVE fid %d\n", fid->fid);
err = 0;
clnt = fid->clnt;
@@ -1239,29 +1506,56 @@ int p9_client_remove(struct p9_fid *fid)
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RREMOVE fid %d\n", fid->fid);
+ p9_debug(P9_DEBUG_9P, "<<< RREMOVE fid %d\n", fid->fid);
p9_free_req(clnt, req);
error:
- p9_fid_destroy(fid);
+ 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, char *data, char __user *udata, u64 offset,
u32 count)
{
- int err, rsize, total;
- struct p9_client *clnt;
- struct p9_req_t *req;
char *dataptr;
+ int kernel_buf = 0;
+ struct p9_req_t *req;
+ struct p9_client *clnt;
+ int err, rsize, non_zc = 0;
+
- P9_DPRINTK(P9_DEBUG_9P, ">>> TREAD fid %d offset %llu %d\n", fid->fid,
- (long long unsigned) offset, count);
+ p9_debug(P9_DEBUG_9P, ">>> TREAD fid %d offset %llu %d\n",
+ fid->fid, (unsigned long long) offset, count);
err = 0;
clnt = fid->clnt;
- total = 0;
rsize = fid->iounit;
if (!rsize || rsize > clnt->msize-P9_IOHDRSZ)
@@ -1270,7 +1564,26 @@ p9_client_read(struct p9_fid *fid, char *data, char __user *udata, u64 offset,
if (count < rsize)
rsize = count;
- req = p9_client_rpc(clnt, P9_TREAD, "dqd", fid->fid, offset, rsize);
+ /* Don't bother zerocopy for small IO (< 1024) */
+ if (clnt->trans_mod->zc_request && rsize > 1024) {
+ char *indata;
+ if (data) {
+ kernel_buf = 1;
+ indata = data;
+ } else
+ indata = (__force char *)udata;
+ /*
+ * response header len is 11
+ * PDU Header(7) + IO Size (4)
+ */
+ req = p9_client_zc_rpc(clnt, P9_TREAD, indata, NULL, rsize, 0,
+ 11, kernel_buf, "dqd", fid->fid,
+ offset, rsize);
+ } else {
+ non_zc = 1;
+ req = p9_client_rpc(clnt, P9_TREAD, "dqd", fid->fid, offset,
+ rsize);
+ }
if (IS_ERR(req)) {
err = PTR_ERR(req);
goto error;
@@ -1278,19 +1591,21 @@ p9_client_read(struct p9_fid *fid, char *data, char __user *udata, u64 offset,
err = p9pdu_readf(req->rc, clnt->proto_version, "D", &count, &dataptr);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
goto free_and_error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RREAD count %d\n", count);
+ p9_debug(P9_DEBUG_9P, "<<< RREAD count %d\n", count);
- if (data) {
- memmove(data, dataptr, count);
- } else {
- err = copy_to_user(udata, dataptr, count);
- if (err) {
- err = -EFAULT;
- goto free_and_error;
+ if (non_zc) {
+ if (data) {
+ memmove(data, dataptr, count);
+ } else {
+ err = copy_to_user(udata, dataptr, count);
+ if (err) {
+ err = -EFAULT;
+ goto free_and_error;
+ }
}
}
p9_free_req(clnt, req);
@@ -1307,15 +1622,15 @@ int
p9_client_write(struct p9_fid *fid, char *data, const char __user *udata,
u64 offset, u32 count)
{
- int err, rsize, total;
+ int err, rsize;
+ int kernel_buf = 0;
struct p9_client *clnt;
struct p9_req_t *req;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TWRITE fid %d offset %llu count %d\n",
- fid->fid, (long long unsigned) offset, count);
+ p9_debug(P9_DEBUG_9P, ">>> TWRITE fid %d offset %llu count %d\n",
+ fid->fid, (unsigned long long) offset, count);
err = 0;
clnt = fid->clnt;
- total = 0;
rsize = fid->iounit;
if (!rsize || rsize > clnt->msize-P9_IOHDRSZ)
@@ -1323,12 +1638,26 @@ p9_client_write(struct p9_fid *fid, char *data, const char __user *udata,
if (count < rsize)
rsize = count;
- if (data)
- req = p9_client_rpc(clnt, P9_TWRITE, "dqD", fid->fid, offset,
- rsize, data);
- else
- req = p9_client_rpc(clnt, P9_TWRITE, "dqU", fid->fid, offset,
- rsize, udata);
+
+ /* Don't bother zerocopy for small IO (< 1024) */
+ if (clnt->trans_mod->zc_request && rsize > 1024) {
+ char *odata;
+ if (data) {
+ kernel_buf = 1;
+ odata = data;
+ } else
+ odata = (char *)udata;
+ req = p9_client_zc_rpc(clnt, P9_TWRITE, NULL, odata, 0, rsize,
+ P9_ZC_HDR_SZ, kernel_buf, "dqd",
+ fid->fid, offset, rsize);
+ } else {
+ if (data)
+ req = p9_client_rpc(clnt, P9_TWRITE, "dqD", fid->fid,
+ offset, rsize, data);
+ else
+ req = p9_client_rpc(clnt, P9_TWRITE, "dqU", fid->fid,
+ offset, rsize, udata);
+ }
if (IS_ERR(req)) {
err = PTR_ERR(req);
goto error;
@@ -1336,11 +1665,11 @@ p9_client_write(struct p9_fid *fid, char *data, const char __user *udata,
err = p9pdu_readf(req->rc, clnt->proto_version, "d", &count);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
goto free_and_error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RWRITE count %d\n", count);
+ p9_debug(P9_DEBUG_9P, "<<< RWRITE count %d\n", count);
p9_free_req(clnt, req);
return count;
@@ -1360,7 +1689,7 @@ struct p9_wstat *p9_client_stat(struct p9_fid *fid)
struct p9_req_t *req;
u16 ignored;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TSTAT fid %d\n", fid->fid);
+ p9_debug(P9_DEBUG_9P, ">>> TSTAT fid %d\n", fid->fid);
if (!ret)
return ERR_PTR(-ENOMEM);
@@ -1376,12 +1705,12 @@ struct p9_wstat *p9_client_stat(struct p9_fid *fid)
err = p9pdu_readf(req->rc, clnt->proto_version, "wS", &ignored, ret);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
p9_free_req(clnt, req);
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P,
+ 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"
@@ -1390,7 +1719,9 @@ struct p9_wstat *p9_client_stat(struct p9_fid *fid)
(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,
- ret->n_uid, ret->n_gid, ret->n_muid);
+ 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;
@@ -1410,7 +1741,7 @@ struct p9_stat_dotl *p9_client_getattr_dotl(struct p9_fid *fid,
GFP_KERNEL);
struct p9_req_t *req;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TGETATTR fid %d, request_mask %lld\n",
+ p9_debug(P9_DEBUG_9P, ">>> TGETATTR fid %d, request_mask %lld\n",
fid->fid, request_mask);
if (!ret)
@@ -1427,12 +1758,12 @@ struct p9_stat_dotl *p9_client_getattr_dotl(struct p9_fid *fid,
err = p9pdu_readf(req->rc, clnt->proto_version, "A", ret);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
p9_free_req(clnt, req);
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P,
+ p9_debug(P9_DEBUG_9P,
"<<< RGETATTR st_result_mask=%lld\n"
"<<< qid=%x.%llx.%x\n"
"<<< st_mode=%8.8x st_nlink=%llu\n"
@@ -1444,8 +1775,10 @@ struct p9_stat_dotl *p9_client_getattr_dotl(struct p9_fid *fid,
"<<< 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, ret->st_uid,
- ret->st_gid, ret->st_rdev, ret->st_size, ret->st_blksize,
+ 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,
@@ -1498,8 +1831,8 @@ int p9_client_wstat(struct p9_fid *fid, struct p9_wstat *wst)
err = 0;
clnt = fid->clnt;
wst->size = p9_client_statsize(wst, clnt->proto_version);
- P9_DPRINTK(P9_DEBUG_9P, ">>> TWSTAT fid %d\n", fid->fid);
- P9_DPRINTK(P9_DEBUG_9P,
+ 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"
@@ -1508,7 +1841,9 @@ int p9_client_wstat(struct p9_fid *fid, struct p9_wstat *wst)
(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,
- wst->n_uid, wst->n_gid, wst->n_muid);
+ 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)) {
@@ -1516,7 +1851,7 @@ int p9_client_wstat(struct p9_fid *fid, struct p9_wstat *wst)
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RWSTAT fid %d\n", fid->fid);
+ p9_debug(P9_DEBUG_9P, "<<< RWSTAT fid %d\n", fid->fid);
p9_free_req(clnt, req);
error:
@@ -1532,12 +1867,14 @@ int p9_client_setattr(struct p9_fid *fid, struct p9_iattr_dotl *p9attr)
err = 0;
clnt = fid->clnt;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TSETATTR fid %d\n", fid->fid);
- P9_DPRINTK(P9_DEBUG_9P,
+ 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, p9attr->uid, p9attr->gid,
+ 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);
@@ -1547,7 +1884,7 @@ int p9_client_setattr(struct p9_fid *fid, struct p9_iattr_dotl *p9attr)
err = PTR_ERR(req);
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RSETATTR fid %d\n", fid->fid);
+ p9_debug(P9_DEBUG_9P, "<<< RSETATTR fid %d\n", fid->fid);
p9_free_req(clnt, req);
error:
return err;
@@ -1563,7 +1900,7 @@ int p9_client_statfs(struct p9_fid *fid, struct p9_rstatfs *sb)
err = 0;
clnt = fid->clnt;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TSTATFS fid %d\n", fid->fid);
+ 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)) {
@@ -1575,12 +1912,12 @@ int p9_client_statfs(struct p9_fid *fid, struct p9_rstatfs *sb)
&sb->bsize, &sb->blocks, &sb->bfree, &sb->bavail,
&sb->files, &sb->ffree, &sb->fsid, &sb->namelen);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
p9_free_req(clnt, req);
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RSTATFS fid %d type 0x%lx bsize %ld "
+ 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,
@@ -1593,7 +1930,8 @@ error:
}
EXPORT_SYMBOL(p9_client_statfs);
-int p9_client_rename(struct p9_fid *fid, struct p9_fid *newdirfid, char *name)
+int p9_client_rename(struct p9_fid *fid,
+ struct p9_fid *newdirfid, const char *name)
{
int err;
struct p9_req_t *req;
@@ -1602,7 +1940,7 @@ int p9_client_rename(struct p9_fid *fid, struct p9_fid *newdirfid, char *name)
err = 0;
clnt = fid->clnt;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TRENAME fid %d newdirfid %d name %s\n",
+ 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,
@@ -1612,7 +1950,7 @@ int p9_client_rename(struct p9_fid *fid, struct p9_fid *newdirfid, char *name)
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RRENAME fid %d\n", fid->fid);
+ p9_debug(P9_DEBUG_9P, "<<< RRENAME fid %d\n", fid->fid);
p9_free_req(clnt, req);
error:
@@ -1620,6 +1958,36 @@ error:
}
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
*/
@@ -1639,7 +2007,7 @@ struct p9_fid *p9_client_xattrwalk(struct p9_fid *file_fid,
attr_fid = NULL;
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P,
+ p9_debug(P9_DEBUG_9P,
">>> TXATTRWALK file_fid %d, attr_fid %d name %s\n",
file_fid->fid, attr_fid->fid, attr_name);
@@ -1651,12 +2019,12 @@ struct p9_fid *p9_client_xattrwalk(struct p9_fid *file_fid,
}
err = p9pdu_readf(req->rc, clnt->proto_version, "q", attr_size);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
p9_free_req(clnt, req);
goto clunk_fid;
}
p9_free_req(clnt, req);
- P9_DPRINTK(P9_DEBUG_9P, "<<< RXATTRWALK fid %d size %llu\n",
+ p9_debug(P9_DEBUG_9P, "<<< RXATTRWALK fid %d size %llu\n",
attr_fid->fid, *attr_size);
return attr_fid;
clunk_fid:
@@ -1677,7 +2045,7 @@ int p9_client_xattrcreate(struct p9_fid *fid, const char *name,
struct p9_req_t *req;
struct p9_client *clnt;
- P9_DPRINTK(P9_DEBUG_9P,
+ 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;
@@ -1688,7 +2056,7 @@ int p9_client_xattrcreate(struct p9_fid *fid, const char *name,
err = PTR_ERR(req);
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RXATTRCREATE fid %d\n", fid->fid);
+ p9_debug(P9_DEBUG_9P, "<<< RXATTRCREATE fid %d\n", fid->fid);
p9_free_req(clnt, req);
error:
return err;
@@ -1697,17 +2065,16 @@ EXPORT_SYMBOL_GPL(p9_client_xattrcreate);
int p9_client_readdir(struct p9_fid *fid, char *data, u32 count, u64 offset)
{
- int err, rsize, total;
+ int err, rsize, non_zc = 0;
struct p9_client *clnt;
struct p9_req_t *req;
char *dataptr;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TREADDIR fid %d offset %llu count %d\n",
- fid->fid, (long long unsigned) offset, 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;
- total = 0;
rsize = fid->iounit;
if (!rsize || rsize > clnt->msize-P9_READDIRHDRSZ)
@@ -1716,7 +2083,19 @@ int p9_client_readdir(struct p9_fid *fid, char *data, u32 count, u64 offset)
if (count < rsize)
rsize = count;
- req = p9_client_rpc(clnt, P9_TREADDIR, "dqd", fid->fid, offset, rsize);
+ /* Don't bother zerocopy for small IO (< 1024) */
+ if (clnt->trans_mod->zc_request && rsize > 1024) {
+ /*
+ * response header len is 11
+ * PDU Header(7) + IO Size (4)
+ */
+ req = p9_client_zc_rpc(clnt, P9_TREADDIR, data, NULL, rsize, 0,
+ 11, 1, "dqd", fid->fid, offset, rsize);
+ } else {
+ non_zc = 1;
+ req = p9_client_rpc(clnt, P9_TREADDIR, "dqd", fid->fid,
+ offset, rsize);
+ }
if (IS_ERR(req)) {
err = PTR_ERR(req);
goto error;
@@ -1724,13 +2103,13 @@ int p9_client_readdir(struct p9_fid *fid, char *data, u32 count, u64 offset)
err = p9pdu_readf(req->rc, clnt->proto_version, "D", &count, &dataptr);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
goto free_and_error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RREADDIR count %d\n", count);
+ p9_debug(P9_DEBUG_9P, "<<< RREADDIR count %d\n", count);
- if (data)
+ if (non_zc)
memmove(data, dataptr, count);
p9_free_req(clnt, req);
@@ -1744,7 +2123,7 @@ error:
EXPORT_SYMBOL(p9_client_readdir);
int p9_client_mknod_dotl(struct p9_fid *fid, char *name, int mode,
- dev_t rdev, gid_t gid, struct p9_qid *qid)
+ dev_t rdev, kgid_t gid, struct p9_qid *qid)
{
int err;
struct p9_client *clnt;
@@ -1752,19 +2131,19 @@ int p9_client_mknod_dotl(struct p9_fid *fid, char *name, int mode,
err = 0;
clnt = fid->clnt;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TMKNOD fid %d name %s mode %d major %d "
+ 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, "dsdddd", fid->fid, name, mode,
+ 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) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RMKNOD qid %x.%llx.%x\n", qid->type,
+ p9_debug(P9_DEBUG_9P, "<<< RMKNOD qid %x.%llx.%x\n", qid->type,
(unsigned long long)qid->path, qid->version);
error:
@@ -1775,7 +2154,7 @@ error:
EXPORT_SYMBOL(p9_client_mknod_dotl);
int p9_client_mkdir_dotl(struct p9_fid *fid, char *name, int mode,
- gid_t gid, struct p9_qid *qid)
+ kgid_t gid, struct p9_qid *qid)
{
int err;
struct p9_client *clnt;
@@ -1783,19 +2162,19 @@ int p9_client_mkdir_dotl(struct p9_fid *fid, char *name, int mode,
err = 0;
clnt = fid->clnt;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TMKDIR fid %d name %s mode %d gid %d\n",
- fid->fid, name, mode, gid);
- req = p9_client_rpc(clnt, P9_TMKDIR, "dsdd", fid->fid, name, mode,
+ 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) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RMKDIR qid %x.%llx.%x\n", qid->type,
+ p9_debug(P9_DEBUG_9P, "<<< RMKDIR qid %x.%llx.%x\n", qid->type,
(unsigned long long)qid->path, qid->version);
error:
@@ -1813,7 +2192,7 @@ int p9_client_lock_dotl(struct p9_fid *fid, struct p9_flock *flock, u8 *status)
err = 0;
clnt = fid->clnt;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TLOCK fid %d type %i flags %d "
+ 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);
@@ -1827,10 +2206,10 @@ int p9_client_lock_dotl(struct p9_fid *fid, struct p9_flock *flock, u8 *status)
err = p9pdu_readf(req->rc, clnt->proto_version, "b", status);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RLOCK status %i\n", *status);
+ p9_debug(P9_DEBUG_9P, "<<< RLOCK status %i\n", *status);
error:
p9_free_req(clnt, req);
return err;
@@ -1846,7 +2225,7 @@ int p9_client_getlock_dotl(struct p9_fid *fid, struct p9_getlock *glock)
err = 0;
clnt = fid->clnt;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TGETLOCK fid %d, type %i start %lld "
+ 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);
@@ -1860,10 +2239,10 @@ int p9_client_getlock_dotl(struct p9_fid *fid, struct p9_getlock *glock)
&glock->start, &glock->length, &glock->proc_id,
&glock->client_id);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RGETLOCK type %i start %lld length %lld "
+ 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:
@@ -1880,7 +2259,7 @@ int p9_client_readlink(struct p9_fid *fid, char **target)
err = 0;
clnt = fid->clnt;
- P9_DPRINTK(P9_DEBUG_9P, ">>> TREADLINK fid %d\n", fid->fid);
+ 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))
@@ -1888,10 +2267,10 @@ int p9_client_readlink(struct p9_fid *fid, char **target)
err = p9pdu_readf(req->rc, clnt->proto_version, "s", target);
if (err) {
- p9pdu_dump(1, req->rc);
+ trace_9p_protocol_dump(clnt, req->rc);
goto error;
}
- P9_DPRINTK(P9_DEBUG_9P, "<<< RREADLINK target %s\n", *target);
+ p9_debug(P9_DEBUG_9P, "<<< RREADLINK target %s\n", *target);
error:
p9_free_req(clnt, req);
return err;
diff --git a/net/9p/error.c b/net/9p/error.c
index 52518512a93..126fd0dceea 100644
--- a/net/9p/error.c
+++ b/net/9p/error.c
@@ -27,6 +27,8 @@
*
*/
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
#include <linux/module.h>
#include <linux/list.h>
#include <linux/jhash.h>
@@ -219,15 +221,13 @@ EXPORT_SYMBOL(p9_error_init);
int p9_errstr2errno(char *errstr, int len)
{
int errno;
- struct hlist_node *p;
struct errormap *c;
int bucket;
errno = 0;
- p = NULL;
c = NULL;
bucket = jhash(errstr, len, 0) % ERRHASHSZ;
- hlist_for_each_entry(c, p, &hash_errmap[bucket], list) {
+ hlist_for_each_entry(c, &hash_errmap[bucket], list) {
if (c->namelen == len && !memcmp(c->name, errstr, len)) {
errno = c->val;
break;
@@ -237,8 +237,8 @@ int p9_errstr2errno(char *errstr, int len)
if (errno == 0) {
/* TODO: if error isn't found, add it dynamically */
errstr[len] = 0;
- printk(KERN_ERR "%s: server reported unknown error %s\n",
- __func__, errstr);
+ pr_err("%s: server reported unknown error %s\n",
+ __func__, errstr);
errno = ESERVERFAULT;
}
diff --git a/net/9p/mod.c b/net/9p/mod.c
index cf8a4128cd5..6ab36aea772 100644
--- a/net/9p/mod.c
+++ b/net/9p/mod.c
@@ -24,7 +24,11 @@
*
*/
+#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>
@@ -39,6 +43,29 @@ 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
/*
@@ -80,14 +107,14 @@ EXPORT_SYMBOL(v9fs_unregister_trans);
* @name: string identifying transport
*
*/
-struct p9_trans_module *v9fs_get_trans_by_name(const substring_t *name)
+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 (strncmp(t->name, name->from, name->to-name->from) == 0 &&
+ if (strcmp(t->name, s) == 0 &&
try_module_get(t->owner)) {
found = t;
break;
@@ -139,7 +166,7 @@ void v9fs_put_trans(struct p9_trans_module *m)
}
/**
- * v9fs_init - Initialize module
+ * init_p9 - Initialize module
*
*/
static int __init init_p9(void)
@@ -147,20 +174,20 @@ static int __init init_p9(void)
int ret = 0;
p9_error_init();
- printk(KERN_INFO "Installing 9P2000 support\n");
+ pr_info("Installing 9P2000 support\n");
p9_trans_fd_init();
return ret;
}
/**
- * v9fs_init - shutdown module
+ * exit_p9 - shutdown module
*
*/
static void __exit exit_p9(void)
{
- printk(KERN_INFO "Unloading 9P2000 support\n");
+ pr_info("Unloading 9P2000 support\n");
p9_trans_fd_exit();
}
diff --git a/net/9p/protocol.c b/net/9p/protocol.c
index 1e308f21092..ab9127ec5b7 100644
--- a/net/9p/protocol.c
+++ b/net/9p/protocol.c
@@ -37,46 +37,11 @@
#include <net/9p/client.h>
#include "protocol.h"
+#include <trace/events/9p.h>
+
static int
p9pdu_writef(struct p9_fcall *pdu, int proto_version, const char *fmt, ...);
-#ifdef CONFIG_NET_9P_DEBUG
-void
-p9pdu_dump(int way, struct p9_fcall *pdu)
-{
- int i, n;
- u8 *data = pdu->sdata;
- int datalen = pdu->size;
- char buf[255];
- int buflen = 255;
-
- i = n = 0;
- if (datalen > (buflen-16))
- datalen = buflen-16;
- while (i < datalen) {
- n += scnprintf(buf + n, buflen - n, "%02x ", data[i]);
- if (i%4 == 3)
- n += scnprintf(buf + n, buflen - n, " ");
- if (i%32 == 31)
- n += scnprintf(buf + n, buflen - n, "\n");
-
- i++;
- }
- n += scnprintf(buf + n, buflen - n, "\n");
-
- if (way)
- P9_DPRINTK(P9_DEBUG_PKT, "[[[(%d) %s\n", datalen, buf);
- else
- P9_DPRINTK(P9_DEBUG_PKT, "]]](%d) %s\n", datalen, buf);
-}
-#else
-void
-p9pdu_dump(int way, struct p9_fcall *pdu)
-{
-}
-#endif
-EXPORT_SYMBOL(p9pdu_dump);
-
void p9stat_free(struct p9_wstat *stbuf)
{
kfree(stbuf->name);
@@ -87,7 +52,7 @@ void p9stat_free(struct p9_wstat *stbuf)
}
EXPORT_SYMBOL(p9stat_free);
-static size_t pdu_read(struct p9_fcall *pdu, void *data, size_t size)
+size_t pdu_read(struct p9_fcall *pdu, void *data, size_t size)
{
size_t len = min(pdu->size - pdu->offset, size);
memcpy(data, &pdu->sdata[pdu->offset], len);
@@ -120,6 +85,8 @@ pdu_write_u(struct p9_fcall *pdu, const char __user *udata, size_t size)
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)
@@ -185,7 +152,7 @@ p9pdu_vreadf(struct p9_fcall *pdu, int proto_version, const char *fmt,
if (errcode)
break;
- *sptr = kmalloc(len + 1, GFP_KERNEL);
+ *sptr = kmalloc(len + 1, GFP_NOFS);
if (*sptr == NULL) {
errcode = -EFAULT;
break;
@@ -198,6 +165,26 @@ p9pdu_vreadf(struct p9_fcall *pdu, int proto_version, const char *fmt,
(*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 *);
@@ -212,11 +199,12 @@ p9pdu_vreadf(struct p9_fcall *pdu, int proto_version, const char *fmt,
va_arg(ap, struct p9_wstat *);
memset(stbuf, 0, sizeof(struct p9_wstat));
- stbuf->n_uid = stbuf->n_gid = stbuf->n_muid =
- -1;
+ stbuf->n_uid = stbuf->n_muid = INVALID_UID;
+ stbuf->n_gid = INVALID_GID;
+
errcode =
p9pdu_readf(pdu, proto_version,
- "wwdQdddqssss?sddd",
+ "wwdQdddqssss?sugu",
&stbuf->size, &stbuf->type,
&stbuf->dev, &stbuf->qid,
&stbuf->mode, &stbuf->atime,
@@ -245,7 +233,7 @@ p9pdu_vreadf(struct p9_fcall *pdu, int proto_version, const char *fmt,
}
break;
case 'T':{
- int16_t *nwname = va_arg(ap, int16_t *);
+ uint16_t *nwname = va_arg(ap, uint16_t *);
char ***wnames = va_arg(ap, char ***);
errcode = p9pdu_readf(pdu, proto_version,
@@ -253,7 +241,7 @@ p9pdu_vreadf(struct p9_fcall *pdu, int proto_version, const char *fmt,
if (!errcode) {
*wnames =
kmalloc(sizeof(char *) * *nwname,
- GFP_KERNEL);
+ GFP_NOFS);
if (!*wnames)
errcode = -ENOMEM;
}
@@ -297,7 +285,7 @@ p9pdu_vreadf(struct p9_fcall *pdu, int proto_version, const char *fmt,
*wqids =
kmalloc(*nwqid *
sizeof(struct p9_qid),
- GFP_KERNEL);
+ GFP_NOFS);
if (*wqids == NULL)
errcode = -ENOMEM;
}
@@ -329,7 +317,7 @@ p9pdu_vreadf(struct p9_fcall *pdu, int proto_version, const char *fmt,
memset(stbuf, 0, sizeof(struct p9_stat_dotl));
errcode =
p9pdu_readf(pdu, proto_version,
- "qQdddqqqqqqqqqqqqqqq",
+ "qQdugqqqqqqqqqqqqqqq",
&stbuf->st_result_mask,
&stbuf->qid,
&stbuf->st_mode,
@@ -403,7 +391,7 @@ p9pdu_vwritef(struct p9_fcall *pdu, int proto_version, const char *fmt,
const char *sptr = va_arg(ap, const char *);
uint16_t len = 0;
if (sptr)
- len = min_t(uint16_t, strlen(sptr),
+ len = min_t(size_t, strlen(sptr),
USHRT_MAX);
errcode = p9pdu_writef(pdu, proto_version,
@@ -412,6 +400,20 @@ p9pdu_vwritef(struct p9_fcall *pdu, int proto_version, const char *fmt,
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 *);
@@ -425,7 +427,7 @@ p9pdu_vwritef(struct p9_fcall *pdu, int proto_version, const char *fmt,
va_arg(ap, const struct p9_wstat *);
errcode =
p9pdu_writef(pdu, proto_version,
- "wwdQdddqssss?sddd",
+ "wwdQdddqssss?sugu",
stbuf->size, stbuf->type,
stbuf->dev, &stbuf->qid,
stbuf->mode, stbuf->atime,
@@ -456,7 +458,7 @@ p9pdu_vwritef(struct p9_fcall *pdu, int proto_version, const char *fmt,
}
break;
case 'T':{
- int16_t nwname = va_arg(ap, int);
+ uint16_t nwname = va_arg(ap, int);
const char **wnames = va_arg(ap, const char **);
errcode = p9pdu_writef(pdu, proto_version, "w",
@@ -503,7 +505,7 @@ p9pdu_vwritef(struct p9_fcall *pdu, int proto_version, const char *fmt,
struct p9_iattr_dotl *);
errcode = p9pdu_writef(pdu, proto_version,
- "ddddqqqqq",
+ "ddugqqqqq",
p9attr->valid,
p9attr->mode,
p9attr->uid,
@@ -557,7 +559,7 @@ p9pdu_writef(struct p9_fcall *pdu, int proto_version, const char *fmt, ...)
return ret;
}
-int p9stat_read(char *buf, int len, struct p9_wstat *st, int proto_version)
+int p9stat_read(struct p9_client *clnt, char *buf, int len, struct p9_wstat *st)
{
struct p9_fcall fake_pdu;
int ret;
@@ -567,10 +569,10 @@ int p9stat_read(char *buf, int len, struct p9_wstat *st, int proto_version)
fake_pdu.sdata = buf;
fake_pdu.offset = 0;
- ret = p9pdu_readf(&fake_pdu, proto_version, "S", st);
+ ret = p9pdu_readf(&fake_pdu, clnt->proto_version, "S", st);
if (ret) {
- P9_DPRINTK(P9_DEBUG_9P, "<<< p9stat_read failed: %d\n", ret);
- p9pdu_dump(1, &fake_pdu);
+ p9_debug(P9_DEBUG_9P, "<<< p9stat_read failed: %d\n", ret);
+ trace_9p_protocol_dump(clnt, &fake_pdu);
}
return ret;
@@ -579,10 +581,11 @@ 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_fcall *pdu)
+int p9pdu_finalize(struct p9_client *clnt, struct p9_fcall *pdu)
{
int size = pdu->size;
int err;
@@ -591,13 +594,9 @@ int p9pdu_finalize(struct p9_fcall *pdu)
err = p9pdu_writef(pdu, 0, "d", size);
pdu->size = size;
-#ifdef CONFIG_NET_9P_DEBUG
- if ((p9_debug_level & P9_DEBUG_PKT) == P9_DEBUG_PKT)
- p9pdu_dump(0, pdu);
-#endif
-
- P9_DPRINTK(P9_DEBUG_9P, ">>> size=%d type: %d tag: %d\n", pdu->size,
- pdu->id, pdu->tag);
+ 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;
}
@@ -608,8 +607,8 @@ void p9pdu_reset(struct p9_fcall *pdu)
pdu->size = 0;
}
-int p9dirent_read(char *buf, int len, struct p9_dirent *dirent,
- int proto_version)
+int p9dirent_read(struct p9_client *clnt, char *buf, int len,
+ struct p9_dirent *dirent)
{
struct p9_fcall fake_pdu;
int ret;
@@ -620,15 +619,16 @@ int p9dirent_read(char *buf, int len, struct p9_dirent *dirent,
fake_pdu.sdata = buf;
fake_pdu.offset = 0;
- ret = p9pdu_readf(&fake_pdu, proto_version, "Qqbs", &dirent->qid,
- &dirent->d_off, &dirent->d_type, &nameptr);
+ ret = p9pdu_readf(&fake_pdu, clnt->proto_version, "Qqbs", &dirent->qid,
+ &dirent->d_off, &dirent->d_type, &nameptr);
if (ret) {
- P9_DPRINTK(P9_DEBUG_9P, "<<< p9dirent_read failed: %d\n", ret);
- p9pdu_dump(1, &fake_pdu);
+ 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;
diff --git a/net/9p/protocol.h b/net/9p/protocol.h
index 2431c0f38d5..2cc525fa49f 100644
--- a/net/9p/protocol.h
+++ b/net/9p/protocol.h
@@ -29,6 +29,6 @@ int p9pdu_vwritef(struct p9_fcall *pdu, int proto_version, const char *fmt,
va_list ap);
int p9pdu_readf(struct p9_fcall *pdu, int proto_version, const char *fmt, ...);
int p9pdu_prepare(struct p9_fcall *pdu, int16_t tag, int8_t type);
-int p9pdu_finalize(struct p9_fcall *pdu);
-void p9pdu_dump(int, struct p9_fcall *);
+int p9pdu_finalize(struct p9_client *clnt, struct p9_fcall *pdu);
void p9pdu_reset(struct p9_fcall *pdu);
+size_t pdu_read(struct p9_fcall *pdu, void *data, size_t size);
diff --git a/net/9p/trans_common.c b/net/9p/trans_common.c
new file mode 100644
index 00000000000..2ee3879161b
--- /dev/null
+++ b/net/9p/trans_common.c
@@ -0,0 +1,69 @@
+/*
+ * 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/slab.h>
+#include <linux/module.h>
+#include <net/9p/9p.h>
+#include <net/9p/client.h>
+#include <linux/scatterlist.h>
+#include "trans_common.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);
+
+/**
+ * p9_nr_pages - Return number of pages needed to accommodate the payload.
+ */
+int p9_nr_pages(char *data, int len)
+{
+ unsigned long start_page, end_page;
+ start_page = (unsigned long)data >> PAGE_SHIFT;
+ end_page = ((unsigned long)data + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+ return end_page - start_page;
+}
+EXPORT_SYMBOL(p9_nr_pages);
+
+/**
+ * payload_gup - Translates user buffer into kernel pages and
+ * pins them either for read/write through get_user_pages_fast().
+ * @req: Request to be sent to server.
+ * @pdata_off: data offset into the first page after translation (gup).
+ * @pdata_len: Total length of the IO. gup may not return requested # of pages.
+ * @nr_pages: number of pages to accommodate the payload
+ * @rw: Indicates if the pages are for read or write.
+ */
+
+int p9_payload_gup(char *data, int *nr_pages, struct page **pages, int write)
+{
+ int nr_mapped_pages;
+
+ nr_mapped_pages = get_user_pages_fast((unsigned long)data,
+ *nr_pages, write, pages);
+ if (nr_mapped_pages <= 0)
+ return nr_mapped_pages;
+
+ *nr_pages = nr_mapped_pages;
+ return 0;
+}
+EXPORT_SYMBOL(p9_payload_gup);
diff --git a/net/9p/trans_common.h b/net/9p/trans_common.h
new file mode 100644
index 00000000000..173bb550a9e
--- /dev/null
+++ b/net/9p/trans_common.h
@@ -0,0 +1,17 @@
+/*
+ * 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);
+int p9_payload_gup(char *, int *, struct page **, int);
+int p9_nr_pages(char *, int);
diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c
index 078eb162d9b..80d08f6664c 100644
--- a/net/9p/trans_fd.c
+++ b/net/9p/trans_fd.c
@@ -25,6 +25,8 @@
*
*/
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
#include <linux/in.h>
#include <linux/module.h>
#include <linux/net.h>
@@ -61,20 +63,7 @@ struct p9_fd_opts {
int rfd;
int wfd;
u16 port;
-};
-
-/**
- * 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;
+ int privport;
};
/*
@@ -85,12 +74,15 @@ struct p9_trans_fd {
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},
};
@@ -153,10 +145,28 @@ struct p9_conn {
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 struct workqueue_struct *p9_mux_wq;
-static struct task_struct *p9_poll_task;
+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)
{
@@ -190,7 +200,7 @@ static void p9_conn_cancel(struct p9_conn *m, int err)
unsigned long flags;
LIST_HEAD(cancel_list);
- P9_DPRINTK(P9_DEBUG_ERROR, "mux %p err %d\n", m, err);
+ p9_debug(P9_DEBUG_ERROR, "mux %p err %d\n", m, err);
spin_lock_irqsave(&m->client->lock, flags);
@@ -202,23 +212,19 @@ static void p9_conn_cancel(struct p9_conn *m, int err)
m->err = err;
list_for_each_entry_safe(req, rtmp, &m->req_list, req_list) {
- req->status = REQ_STATUS_ERROR;
- if (!req->t_err)
- req->t_err = err;
list_move(&req->req_list, &cancel_list);
}
list_for_each_entry_safe(req, rtmp, &m->unsent_req_list, req_list) {
- req->status = REQ_STATUS_ERROR;
- if (!req->t_err)
- req->t_err = err;
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_DPRINTK(P9_DEBUG_ERROR, "call back req %p\n", req);
+ p9_debug(P9_DEBUG_ERROR, "call back req %p\n", req);
list_del(&req->req_list);
- p9_client_cb(m->client, req);
+ if (!req->t_err)
+ req->t_err = err;
+ p9_client_cb(m->client, req, REQ_STATUS_ERROR);
}
}
@@ -234,10 +240,10 @@ p9_fd_poll(struct p9_client *client, struct poll_table_struct *pt)
if (!ts)
return -EREMOTEIO;
- if (!ts->rd->f_op || !ts->rd->f_op->poll)
+ if (!ts->rd->f_op->poll)
return -EIO;
- if (!ts->wr->f_op || !ts->wr->f_op->poll)
+ if (!ts->wr->f_op->poll)
return -EIO;
ret = ts->rd->f_op->poll(ts->rd, pt);
@@ -274,7 +280,7 @@ static int p9_fd_read(struct p9_client *client, void *v, int len)
return -EREMOTEIO;
if (!(ts->rd->f_flags & O_NONBLOCK))
- P9_DPRINTK(P9_DEBUG_ERROR, "blocking read ...\n");
+ 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)
@@ -292,13 +298,14 @@ 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_DPRINTK(P9_DEBUG_TRANS, "start mux %p pos %d\n", m, m->rpos);
+ p9_debug(P9_DEBUG_TRANS, "start mux %p pos %d\n", m, m->rpos);
if (!m->rbuf) {
m->rbuf = m->tmp_buf;
@@ -307,14 +314,13 @@ static void p9_read_work(struct work_struct *work)
}
clear_bit(Rpending, &m->wsched);
- P9_DPRINTK(P9_DEBUG_TRANS, "read mux %p pos %d size: %d = %d\n", m,
- m->rpos, m->rsize, m->rsize-m->rpos);
+ 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_DPRINTK(P9_DEBUG_TRANS, "mux %p got %d bytes\n", m, err);
+ p9_debug(P9_DEBUG_TRANS, "mux %p got %d bytes\n", m, err);
if (err == -EAGAIN) {
- clear_bit(Rworksched, &m->wsched);
- return;
+ goto end_clear;
}
if (err <= 0)
@@ -324,32 +330,31 @@ static void p9_read_work(struct work_struct *work)
if ((!m->req) && (m->rpos == m->rsize)) { /* header read in */
u16 tag;
- P9_DPRINTK(P9_DEBUG_TRANS, "got new header\n");
+ 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_DPRINTK(P9_DEBUG_ERROR,
- "requested packet size too big: %d\n", n);
+ 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_DPRINTK(P9_DEBUG_TRANS,
- "mux %p pkt: size: %d bytes tag: %d\n", m, n, 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 &&
- m->req->status != REQ_STATUS_FLSH)) {
- P9_DPRINTK(P9_DEBUG_ERROR, "Unexpected packet tag %d\n",
- 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_KERNEL);
+ m->client->msize, GFP_NOFS);
if (!m->req->rc) {
m->req = NULL;
err = -ENOMEM;
@@ -363,32 +368,33 @@ static void p9_read_work(struct work_struct *work)
/* not an else because some packets (like clunk) have no payload */
if ((m->req) && (m->rpos == m->rsize)) { /* packet is read in */
- P9_DPRINTK(P9_DEBUG_TRANS, "got new packet\n");
+ p9_debug(P9_DEBUG_TRANS, "got new packet\n");
spin_lock(&m->client->lock);
if (m->req->status != REQ_STATUS_ERROR)
- m->req->status = REQ_STATUS_RCVD;
+ status = REQ_STATUS_RCVD;
list_del(&m->req->req_list);
spin_unlock(&m->client->lock);
- p9_client_cb(m->client, m->req);
+ 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) {
- P9_DPRINTK(P9_DEBUG_TRANS, "sched read work %p\n", m);
- queue_work(p9_mux_wq, &m->rq);
- } else
- clear_bit(Rworksched, &m->wsched);
- } else
- clear_bit(Rworksched, &m->wsched);
+ 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:
@@ -417,7 +423,7 @@ static int p9_fd_write(struct p9_client *client, void *v, int len)
return -EREMOTEIO;
if (!(ts->wr->f_flags & O_NONBLOCK))
- P9_DPRINTK(P9_DEBUG_ERROR, "blocking write ...\n");
+ p9_debug(P9_DEBUG_ERROR, "blocking write ...\n");
oldfs = get_fs();
set_fs(get_ds());
@@ -450,16 +456,17 @@ static void p9_write_work(struct work_struct *work)
}
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;
}
- spin_lock(&m->client->lock);
req = list_entry(m->unsent_req_list.next, struct p9_req_t,
req_list);
req->status = REQ_STATUS_SENT;
- P9_DPRINTK(P9_DEBUG_TRANS, "move req %p\n", req);
+ p9_debug(P9_DEBUG_TRANS, "move req %p\n", req);
list_move_tail(&req->req_list, &m->req_list);
m->wbuf = req->tc->sdata;
@@ -468,15 +475,14 @@ static void p9_write_work(struct work_struct *work)
spin_unlock(&m->client->lock);
}
- P9_DPRINTK(P9_DEBUG_TRANS, "mux %p pos %d size %d\n", m, m->wpos,
- m->wsize);
+ 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_DPRINTK(P9_DEBUG_TRANS, "mux %p sent %d bytes\n", m, err);
- if (err == -EAGAIN) {
- clear_bit(Wworksched, &m->wsched);
- return;
- }
+ p9_debug(P9_DEBUG_TRANS, "mux %p sent %d bytes\n", m, err);
+ if (err == -EAGAIN)
+ goto end_clear;
+
if (err < 0)
goto error;
@@ -489,19 +495,21 @@ static void p9_write_work(struct work_struct *work)
if (m->wpos == m->wsize)
m->wpos = m->wsize = 0;
- if (m->wsize == 0 && !list_empty(&m->unsent_req_list)) {
+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) {
- P9_DPRINTK(P9_DEBUG_TRANS, "sched write work %p\n", m);
- queue_work(p9_mux_wq, &m->wq);
- } else
- clear_bit(Wworksched, &m->wsched);
- } else
- clear_bit(Wworksched, &m->wsched);
+ 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;
@@ -510,21 +518,20 @@ error:
clear_bit(Wworksched, &m->wsched);
}
-static int p9_pollwake(wait_queue_t *wait, unsigned mode, int sync, void *key)
+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;
- DECLARE_WAITQUEUE(dummy_wait, p9_poll_task);
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);
- /* perform the default wake up operation */
- return default_wake_function(&dummy_wait, mode, sync, key);
+ schedule_work(&p9_poll_work);
+ return 1;
}
/**
@@ -551,7 +558,7 @@ p9_pollwait(struct file *filp, wait_queue_head_t *wait_address, poll_table *p)
}
if (!pwait) {
- P9_DPRINTK(P9_DEBUG_ERROR, "not enough wait_address slots\n");
+ p9_debug(P9_DEBUG_ERROR, "not enough wait_address slots\n");
return;
}
@@ -562,22 +569,19 @@ p9_pollwait(struct file *filp, wait_queue_head_t *wait_address, poll_table *p)
}
/**
- * p9_conn_create - allocate and initialize the per-session mux data
+ * p9_conn_create - initialize the per-session mux data
* @client: client instance
*
* Note: Creates the polling task if this is the first session.
*/
-static struct p9_conn *p9_conn_create(struct p9_client *client)
+static void p9_conn_create(struct p9_client *client)
{
int n;
- struct p9_conn *m;
+ struct p9_trans_fd *ts = client->trans;
+ struct p9_conn *m = &ts->conn;
- P9_DPRINTK(P9_DEBUG_TRANS, "client %p msize %d\n", client,
- client->msize);
- m = kzalloc(sizeof(struct p9_conn), GFP_KERNEL);
- if (!m)
- return ERR_PTR(-ENOMEM);
+ p9_debug(P9_DEBUG_TRANS, "client %p msize %d\n", client, client->msize);
INIT_LIST_HEAD(&m->mux_list);
m->client = client;
@@ -591,16 +595,14 @@ static struct p9_conn *p9_conn_create(struct p9_client *client)
n = p9_fd_poll(client, &m->pt);
if (n & POLLIN) {
- P9_DPRINTK(P9_DEBUG_TRANS, "mux %p can read\n", m);
+ p9_debug(P9_DEBUG_TRANS, "mux %p can read\n", m);
set_bit(Rpending, &m->wsched);
}
if (n & POLLOUT) {
- P9_DPRINTK(P9_DEBUG_TRANS, "mux %p can write\n", m);
+ p9_debug(P9_DEBUG_TRANS, "mux %p can write\n", m);
set_bit(Wpending, &m->wsched);
}
-
- return m;
}
/**
@@ -618,7 +620,7 @@ static void p9_poll_mux(struct p9_conn *m)
n = p9_fd_poll(m->client, NULL);
if (n < 0 || n & (POLLERR | POLLHUP | POLLNVAL)) {
- P9_DPRINTK(P9_DEBUG_TRANS, "error mux %p err %d\n", m, n);
+ p9_debug(P9_DEBUG_TRANS, "error mux %p err %d\n", m, n);
if (n >= 0)
n = -ECONNRESET;
p9_conn_cancel(m, n);
@@ -626,20 +628,20 @@ static void p9_poll_mux(struct p9_conn *m)
if (n & POLLIN) {
set_bit(Rpending, &m->wsched);
- P9_DPRINTK(P9_DEBUG_TRANS, "mux %p can read\n", m);
+ p9_debug(P9_DEBUG_TRANS, "mux %p can read\n", m);
if (!test_and_set_bit(Rworksched, &m->wsched)) {
- P9_DPRINTK(P9_DEBUG_TRANS, "sched read work %p\n", m);
- queue_work(p9_mux_wq, &m->rq);
+ p9_debug(P9_DEBUG_TRANS, "sched read work %p\n", m);
+ schedule_work(&m->rq);
}
}
if (n & POLLOUT) {
set_bit(Wpending, &m->wsched);
- P9_DPRINTK(P9_DEBUG_TRANS, "mux %p can write\n", m);
+ 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_DPRINTK(P9_DEBUG_TRANS, "sched write work %p\n", m);
- queue_work(p9_mux_wq, &m->wq);
+ p9_debug(P9_DEBUG_TRANS, "sched write work %p\n", m);
+ schedule_work(&m->wq);
}
}
}
@@ -659,10 +661,10 @@ 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;
+ struct p9_conn *m = &ts->conn;
- P9_DPRINTK(P9_DEBUG_TRANS, "mux %p task %p tcall %p id %d\n", m,
- current, req->tc, req->tc->id);
+ 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;
@@ -677,7 +679,7 @@ static int p9_fd_request(struct p9_client *client, struct p9_req_t *req)
n = p9_fd_poll(m->client, NULL);
if (n & POLLOUT && !test_and_set_bit(Wworksched, &m->wsched))
- queue_work(p9_mux_wq, &m->wq);
+ schedule_work(&m->wq);
return 0;
}
@@ -686,7 +688,7 @@ static int p9_fd_cancel(struct p9_client *client, struct p9_req_t *req)
{
int ret = 1;
- P9_DPRINTK(P9_DEBUG_TRANS, "client %p req %p\n", client, req);
+ p9_debug(P9_DEBUG_TRANS, "client %p req %p\n", client, req);
spin_lock(&client->lock);
@@ -694,14 +696,26 @@ static int p9_fd_cancel(struct p9_client *client, struct p9_req_t *req)
list_del(&req->req_list);
req->status = REQ_STATUS_FLSHD;
ret = 0;
- } else if (req->status == REQ_STATUS_SENT)
- req->status = REQ_STATUS_FLSH;
-
+ }
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
@@ -716,7 +730,6 @@ static int parse_opts(char *params, struct p9_fd_opts *opts)
substring_t args[MAX_OPT_ARGS];
int option;
char *options, *tmp_options;
- int ret;
opts->port = P9_PORT;
opts->rfd = ~0;
@@ -727,8 +740,8 @@ static int parse_opts(char *params, struct p9_fd_opts *opts)
tmp_options = kstrdup(params, GFP_KERNEL);
if (!tmp_options) {
- P9_DPRINTK(P9_DEBUG_ERROR,
- "failed to allocate copy of option string\n");
+ p9_debug(P9_DEBUG_ERROR,
+ "failed to allocate copy of option string\n");
return -ENOMEM;
}
options = tmp_options;
@@ -739,12 +752,11 @@ static int parse_opts(char *params, struct p9_fd_opts *opts)
if (!*p)
continue;
token = match_token(p, tokens, args);
- if (token != Opt_err) {
+ if ((token != Opt_err) && (token != Opt_privport)) {
r = match_int(&args[0], &option);
if (r < 0) {
- P9_DPRINTK(P9_DEBUG_ERROR,
- "integer field, but no integer?\n");
- ret = r;
+ p9_debug(P9_DEBUG_ERROR,
+ "integer field, but no integer?\n");
continue;
}
}
@@ -758,6 +770,9 @@ static int parse_opts(char *params, struct p9_fd_opts *opts)
case Opt_wfdno:
opts->wfd = option;
break;
+ case Opt_privport:
+ opts->privport = 1;
+ break;
default:
continue;
}
@@ -769,7 +784,7 @@ static int parse_opts(char *params, struct p9_fd_opts *opts)
static int p9_fd_open(struct p9_client *client, int rfd, int wfd)
{
- struct p9_trans_fd *ts = kmalloc(sizeof(struct p9_trans_fd),
+ struct p9_trans_fd *ts = kzalloc(sizeof(struct p9_trans_fd),
GFP_KERNEL);
if (!ts)
return -ENOMEM;
@@ -794,53 +809,43 @@ static int p9_fd_open(struct p9_client *client, int rfd, int wfd)
static int p9_socket_open(struct p9_client *client, struct socket *csocket)
{
struct p9_trans_fd *p;
- int ret, fd;
+ struct file *file;
- p = kmalloc(sizeof(struct p9_trans_fd), GFP_KERNEL);
+ p = kzalloc(sizeof(struct p9_trans_fd), GFP_KERNEL);
if (!p)
return -ENOMEM;
csocket->sk->sk_allocation = GFP_NOIO;
- fd = sock_map_fd(csocket, 0);
- if (fd < 0) {
- P9_EPRINTK(KERN_ERR, "p9_socket_open: failed to map fd\n");
+ 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 fd;
+ return PTR_ERR(file);
}
- get_file(csocket->file);
- get_file(csocket->file);
- p->wr = p->rd = csocket->file;
+ get_file(file);
+ p->wr = p->rd = file;
client->trans = p;
client->status = Connected;
- sys_close(fd); /* still racy */
-
p->rd->f_flags |= O_NONBLOCK;
- p->conn = p9_conn_create(client);
- if (IS_ERR(p->conn)) {
- ret = PTR_ERR(p->conn);
- p->conn = NULL;
- kfree(p);
- sockfd_put(csocket);
- sockfd_put(csocket);
- return ret;
- }
+ p9_conn_create(client);
return 0;
}
/**
- * p9_mux_destroy - cancels all pending requests and frees mux resources
+ * p9_mux_destroy - cancels all pending requests of mux
* @m: mux to destroy
*
*/
static void p9_conn_destroy(struct p9_conn *m)
{
- P9_DPRINTK(P9_DEBUG_TRANS, "mux %p prev %p next %p\n", m,
- m->mux_list.prev, m->mux_list.next);
+ 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);
@@ -849,7 +854,6 @@ static void p9_conn_destroy(struct p9_conn *m)
p9_conn_cancel(m, -ECONNRESET);
m->client = NULL;
- kfree(m);
}
/**
@@ -871,7 +875,7 @@ static void p9_fd_close(struct p9_client *client)
client->status = Disconnected;
- p9_conn_destroy(ts->conn);
+ p9_conn_destroy(&ts->conn);
if (ts->rd)
fput(ts->rd);
@@ -898,6 +902,24 @@ static inline int valid_ipaddr4(const char *buf)
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)
{
@@ -918,20 +940,30 @@ p9_fd_create_tcp(struct p9_client *client, const char *addr, char *args)
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_kern(PF_INET, SOCK_STREAM, IPPROTO_TCP, &csocket);
-
+ err = __sock_create(read_pnet(&current->nsproxy->net_ns), PF_INET,
+ SOCK_STREAM, IPPROTO_TCP, &csocket, 1);
if (err) {
- P9_EPRINTK(KERN_ERR, "p9_trans_tcp: problem creating socket\n");
+ 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) {
- P9_EPRINTK(KERN_ERR,
- "p9_trans_tcp: problem connecting socket to %s\n",
- addr);
+ pr_err("%s (%d): problem connecting socket to %s\n",
+ __func__, task_pid_nr(current), addr);
sock_release(csocket);
return err;
}
@@ -949,24 +981,26 @@ p9_fd_create_unix(struct p9_client *client, const char *addr, char *args)
csocket = NULL;
if (strlen(addr) >= UNIX_PATH_MAX) {
- P9_EPRINTK(KERN_ERR, "p9_trans_unix: address too long: %s\n",
- addr);
+ 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_kern(PF_UNIX, SOCK_STREAM, 0, &csocket);
+ err = __sock_create(read_pnet(&current->nsproxy->net_ns), PF_UNIX,
+ SOCK_STREAM, 0, &csocket, 1);
if (err < 0) {
- P9_EPRINTK(KERN_ERR, "p9_trans_unix: problem creating socket\n");
+ 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) {
- P9_EPRINTK(KERN_ERR,
- "p9_trans_unix: problem connecting socket: %s: %d\n",
- addr, err);
+ pr_err("%s (%d): problem connecting socket: %s: %d\n",
+ __func__, task_pid_nr(current), addr, err);
sock_release(csocket);
return err;
}
@@ -984,7 +1018,7 @@ p9_fd_create(struct p9_client *client, const char *addr, char *args)
parse_opts(args, &opts);
if (opts.rfd == ~0 || opts.wfd == ~0) {
- printk(KERN_ERR "v9fs: Insufficient options for proto=fd\n");
+ pr_err("Insufficient options for proto=fd\n");
return -ENOPROTOOPT;
}
@@ -993,14 +1027,7 @@ p9_fd_create(struct p9_client *client, const char *addr, char *args)
return err;
p = (struct p9_trans_fd *) client->trans;
- p->conn = p9_conn_create(client);
- if (IS_ERR(p->conn)) {
- err = PTR_ERR(p->conn);
- p->conn = NULL;
- fput(p->rd);
- fput(p->wr);
- return err;
- }
+ p9_conn_create(client);
return 0;
}
@@ -1008,11 +1035,12 @@ p9_fd_create(struct p9_client *client, const char *addr, char *args)
static struct p9_trans_module p9_tcp_trans = {
.name = "tcp",
.maxsize = MAX_SOCK_BUF,
- .def = 1,
+ .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,
};
@@ -1024,6 +1052,7 @@ static struct p9_trans_module p9_unix_trans = {
.close = p9_fd_close,
.request = p9_fd_request,
.cancel = p9_fd_cancel,
+ .cancelled = p9_fd_cancelled,
.owner = THIS_MODULE,
};
@@ -1035,6 +1064,7 @@ static struct p9_trans_module p9_fd_trans = {
.close = p9_fd_close,
.request = p9_fd_request,
.cancel = p9_fd_cancel,
+ .cancelled = p9_fd_cancelled,
.owner = THIS_MODULE,
};
@@ -1047,12 +1077,12 @@ static struct p9_trans_module p9_fd_trans = {
*
*/
-static int p9_poll_proc(void *a)
+static void p9_poll_workfn(struct work_struct *work)
{
unsigned long flags;
- P9_DPRINTK(P9_DEBUG_TRANS, "start %p\n", current);
- repeat:
+ 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,
@@ -1067,35 +1097,11 @@ static int p9_poll_proc(void *a)
}
spin_unlock_irqrestore(&p9_poll_lock, flags);
- set_current_state(TASK_INTERRUPTIBLE);
- if (list_empty(&p9_poll_pending_list)) {
- P9_DPRINTK(P9_DEBUG_TRANS, "sleeping...\n");
- schedule();
- }
- __set_current_state(TASK_RUNNING);
-
- if (!kthread_should_stop())
- goto repeat;
-
- P9_DPRINTK(P9_DEBUG_TRANS, "finish\n");
- return 0;
+ p9_debug(P9_DEBUG_TRANS, "finish\n");
}
int p9_trans_fd_init(void)
{
- p9_mux_wq = create_workqueue("v9fs");
- if (!p9_mux_wq) {
- printk(KERN_WARNING "v9fs: mux: creating workqueue failed\n");
- return -ENOMEM;
- }
-
- p9_poll_task = kthread_run(p9_poll_proc, NULL, "v9fs-poll");
- if (IS_ERR(p9_poll_task)) {
- destroy_workqueue(p9_mux_wq);
- printk(KERN_WARNING "v9fs: mux: creating poll task failed\n");
- return PTR_ERR(p9_poll_task);
- }
-
v9fs_register_trans(&p9_tcp_trans);
v9fs_register_trans(&p9_unix_trans);
v9fs_register_trans(&p9_fd_trans);
@@ -1105,10 +1111,8 @@ int p9_trans_fd_init(void)
void p9_trans_fd_exit(void)
{
- kthread_stop(p9_poll_task);
+ flush_work(&p9_poll_work);
v9fs_unregister_trans(&p9_tcp_trans);
v9fs_unregister_trans(&p9_unix_trans);
v9fs_unregister_trans(&p9_fd_trans);
-
- destroy_workqueue(p9_mux_wq);
}
diff --git a/net/9p/trans_rdma.c b/net/9p/trans_rdma.c
index 17c5ba7551a..14ad43b5cf8 100644
--- a/net/9p/trans_rdma.c
+++ b/net/9p/trans_rdma.c
@@ -26,6 +26,8 @@
*
*/
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
#include <linux/in.h>
#include <linux/module.h>
#include <linux/net.h>
@@ -55,11 +57,8 @@
#define P9_RDMA_IRD 0
#define P9_RDMA_ORD 0
#define P9_RDMA_TIMEOUT 30000 /* 30 seconds */
-#define P9_RDMA_MAXSIZE (4*4096) /* Min SGE is 4, so we can
- * safely advertise a maxsize
- * of 64k */
+#define P9_RDMA_MAXSIZE (1024*1024) /* 1MB */
-#define P9_RDMA_MAX_SGE (P9_RDMA_MAXSIZE >> PAGE_SHIFT)
/**
* struct p9_trans_rdma - RDMA transport instance
*
@@ -74,7 +73,9 @@
* @sq_depth: The depth of the Send Queue
* @sq_sem: Semaphore for the SQ
* @rq_depth: The depth of the Receive Queue.
- * @rq_count: Count of requests in 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
@@ -99,7 +100,8 @@ struct p9_trans_rdma {
int sq_depth;
struct semaphore sq_sem;
int rq_depth;
- atomic_t rq_count;
+ struct semaphore rq_sem;
+ atomic_t excess_rc;
struct sockaddr_in addr;
spinlock_t req_lock;
@@ -168,7 +170,6 @@ static int parse_opts(char *params, struct p9_rdma_opts *opts)
substring_t args[MAX_OPT_ARGS];
int option;
char *options, *tmp_options;
- int ret;
opts->port = P9_PORT;
opts->sq_depth = P9_RDMA_SQ_DEPTH;
@@ -180,8 +181,8 @@ static int parse_opts(char *params, struct p9_rdma_opts *opts)
tmp_options = kstrdup(params, GFP_KERNEL);
if (!tmp_options) {
- P9_DPRINTK(P9_DEBUG_ERROR,
- "failed to allocate copy of option string\n");
+ p9_debug(P9_DEBUG_ERROR,
+ "failed to allocate copy of option string\n");
return -ENOMEM;
}
options = tmp_options;
@@ -192,11 +193,12 @@ static int parse_opts(char *params, struct p9_rdma_opts *opts)
if (!*p)
continue;
token = match_token(p, tokens, args);
+ if (token == Opt_err)
+ continue;
r = match_int(&args[0], &option);
if (r < 0) {
- P9_DPRINTK(P9_DEBUG_ERROR,
- "integer field, but no integer?\n");
- ret = r;
+ p9_debug(P9_DEBUG_ERROR,
+ "integer field, but no integer?\n");
continue;
}
switch (token) {
@@ -297,15 +299,20 @@ handle_recv(struct p9_client *client, struct p9_trans_rdma *rdma,
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;
- req->status = REQ_STATUS_RCVD;
- p9_client_cb(client, req);
+ p9_client_cb(client, req, REQ_STATUS_RCVD);
return;
err_out:
- P9_DPRINTK(P9_DEBUG_ERROR, "req %p err %d status %d\n",
- req, err, status);
+ p9_debug(P9_DEBUG_ERROR, "req %p err %d status %d\n", req, err, status);
rdma->state = P9_RDMA_FLUSHING;
client->status = Disconnected;
}
@@ -321,8 +328,8 @@ handle_send(struct p9_client *client, struct p9_trans_rdma *rdma,
static void qp_event_handler(struct ib_event *event, void *context)
{
- P9_DPRINTK(P9_DEBUG_ERROR, "QP event %d context %p\n", event->event,
- 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)
@@ -338,8 +345,8 @@ static void cq_comp_handler(struct ib_cq *cq, void *cq_context)
switch (c->wc_op) {
case IB_WC_RECV:
- atomic_dec(&rdma->rq_count);
handle_recv(client, rdma, c, wc.status, wc.byte_len);
+ up(&rdma->rq_sem);
break;
case IB_WC_SEND:
@@ -348,8 +355,7 @@ static void cq_comp_handler(struct ib_cq *cq, void *cq_context)
break;
default:
- printk(KERN_ERR "9prdma: unexpected completion type, "
- "c->wc_op=%d, wc.opcode=%d, status=%d\n",
+ pr_err("unexpected completion type, c->wc_op=%d, wc.opcode=%d, status=%d\n",
c->wc_op, wc.opcode, wc.status);
break;
}
@@ -359,7 +365,7 @@ static void cq_comp_handler(struct ib_cq *cq, void *cq_context)
static void cq_event_handler(struct ib_event *e, void *v)
{
- P9_DPRINTK(P9_DEBUG_ERROR, "CQ event %d context %p\n", e->event, 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)
@@ -410,7 +416,7 @@ post_recv(struct p9_client *client, struct p9_rdma_context *c)
return ib_post_recv(rdma->qp, &wr, &bad_wr);
error:
- P9_DPRINTK(P9_DEBUG_ERROR, "EIO\n");
+ p9_debug(P9_DEBUG_ERROR, "EIO\n");
return -EIO;
}
@@ -424,32 +430,33 @@ static int rdma_request(struct p9_client *client, struct p9_req_t *req)
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_KERNEL);
+ rpl_context = kmalloc(sizeof *rpl_context, GFP_NOFS);
if (!rpl_context) {
err = -ENOMEM;
- goto err_close;
- }
-
- /*
- * If the request has a buffer, steal it, otherwise
- * allocate a new one. Typically, requests should already
- * have receive buffers allocated and just swap them around
- */
- if (!req->rc) {
- req->rc = kmalloc(sizeof(struct p9_fcall)+client->msize,
- GFP_KERNEL);
- if (req->rc) {
- req->rc->sdata = (char *) req->rc +
- sizeof(struct p9_fcall);
- req->rc->capacity = client->msize;
- }
+ goto recv_error;
}
rpl_context->rc = req->rc;
- if (!rpl_context->rc) {
- err = -ENOMEM;
- goto err_free2;
- }
/*
* Post a receive buffer for this request. We need to ensure
@@ -458,29 +465,35 @@ static int rdma_request(struct p9_client *client, struct p9_req_t *req)
* outstanding request, so we must keep a count to avoid
* overflowing the RQ.
*/
- if (atomic_inc_return(&rdma->rq_count) <= rdma->rq_depth) {
- err = post_recv(client, rpl_context);
- if (err)
- goto err_free1;
- } else
- atomic_dec(&rdma->rq_count);
+ 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_KERNEL);
+ c = kmalloc(sizeof *c, GFP_NOFS);
if (!c) {
err = -ENOMEM;
- goto err_free1;
+ 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))
- goto error;
+ 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;
@@ -494,22 +507,38 @@ static int rdma_request(struct p9_client *client, struct p9_req_t *req)
wr.sg_list = &sge;
wr.num_sge = 1;
- if (down_interruptible(&rdma->sq_sem))
- goto error;
+ if (down_interruptible(&rdma->sq_sem)) {
+ err = -EINTR;
+ goto send_error;
+ }
- return ib_post_send(rdma->qp, &wr, &bad_wr);
+ /* 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;
- error:
+ /* Success */
+ return 0;
+
+ /* Handle errors that happened during or while preparing the send: */
+ send_error:
+ req->status = REQ_STATUS_ERROR;
kfree(c);
- kfree(rpl_context->rc);
- kfree(rpl_context);
- P9_DPRINTK(P9_DEBUG_ERROR, "EIO\n");
- return -EIO;
- err_free1:
- kfree(rpl_context->rc);
- err_free2:
+ 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);
- err_close:
spin_lock_irqsave(&rdma->req_lock, flags);
if (rdma->state < P9_RDMA_CLOSING) {
rdma->state = P9_RDMA_CLOSING;
@@ -554,17 +583,30 @@ static struct p9_trans_rdma *alloc_rdma(struct p9_rdma_opts *opts)
spin_lock_init(&rdma->req_lock);
init_completion(&rdma->cm_done);
sema_init(&rdma->sq_sem, rdma->sq_depth);
- atomic_set(&rdma->rq_count, 0);
+ sema_init(&rdma->rq_sem, rdma->rq_depth);
+ atomic_set(&rdma->excess_rc, 0);
return rdma;
}
-/* its not clear to me we can do anything after send has been posted */
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;
+}
+
/**
* trans_create_rdma - Transport method for creating atransport instance
* @client: client instance
@@ -592,7 +634,8 @@ rdma_create_trans(struct p9_client *client, const char *addr, char *args)
return -ENOMEM;
/* Create the RDMA CM ID */
- rdma->cm_id = rdma_create_id(p9_cm_event_handler, client, RDMA_PS_TCP);
+ 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;
@@ -697,6 +740,7 @@ static struct p9_trans_module p9_rdma_trans = {
.close = rdma_close,
.request = rdma_request,
.cancel = rdma_cancel,
+ .cancelled = rdma_cancelled,
};
/**
diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c
index c8f3f72ab20..6940d8fe897 100644
--- a/net/9p/trans_virtio.c
+++ b/net/9p/trans_virtio.c
@@ -26,6 +26,8 @@
*
*/
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
#include <linux/in.h>
#include <linux/module.h>
#include <linux/net.h>
@@ -37,19 +39,24 @@
#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
@@ -77,7 +84,10 @@ struct virtio_chan {
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];
@@ -138,28 +148,24 @@ static void req_done(struct virtqueue *vq)
struct p9_req_t *req;
unsigned long flags;
- P9_DPRINTK(P9_DEBUG_TRANS, ": request done\n");
+ p9_debug(P9_DEBUG_TRANS, ": request done\n");
- do {
+ while (1) {
spin_lock_irqsave(&chan->lock, flags);
rc = virtqueue_get_buf(chan->vq, &len);
-
- if (rc != NULL) {
- if (!chan->ring_bufs_avail) {
- chan->ring_bufs_avail = 1;
- wake_up(chan->vc_wq);
- }
- spin_unlock_irqrestore(&chan->lock, flags);
- P9_DPRINTK(P9_DEBUG_TRANS, ": rc %p\n", rc);
- P9_DPRINTK(P9_DEBUG_TRANS, ": lookup tag %d\n",
- rc->tag);
- req = p9_tag_lookup(chan->client, rc->tag);
- req->status = REQ_STATUS_RCVD;
- p9_client_cb(chan->client, req);
- } else {
+ if (rc == NULL) {
spin_unlock_irqrestore(&chan->lock, flags);
+ break;
}
- } while (rc != NULL);
+ 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);
+ }
}
/**
@@ -176,9 +182,8 @@ static void req_done(struct virtqueue *vq)
*
*/
-static int
-pack_sg_list(struct scatterlist *sg, int start, int limit, char *data,
- int count)
+static int pack_sg_list(struct scatterlist *sg, int start,
+ int limit, char *data, int count)
{
int s;
int index = start;
@@ -187,12 +192,15 @@ pack_sg_list(struct scatterlist *sg, int start, int limit, char *data,
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;
- BUG_ON(index > limit);
}
-
+ if (index-start)
+ sg_mark_end(&sg[index - 1]);
return index-start;
}
@@ -203,6 +211,48 @@ static int p9_virtio_cancel(struct p9_client *client, struct p9_req_t *req)
}
/**
+ * 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
+ * @data: data to pack into scatter/gather list
+ * @count: amount of data to pack into the scatter/gather list
+ */
+static int
+pack_sg_list_p(struct scatterlist *sg, int start, int limit,
+ struct page **pdata, int nr_pages, char *data, int count)
+{
+ int i = 0, s;
+ int data_off;
+ int index = start;
+
+ BUG_ON(nr_pages > (limit - start));
+ /*
+ * if the first page doesn't start at
+ * page boundary find the offset
+ */
+ data_off = offset_in_page(data);
+ while (nr_pages) {
+ s = rest_of_page(data);
+ if (s > count)
+ s = count;
+ /* 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;
+ data += s;
+ 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
@@ -212,24 +262,32 @@ static int p9_virtio_cancel(struct p9_client *client, struct p9_req_t *req)
static int
p9_virtio_request(struct p9_client *client, struct p9_req_t *req)
{
- int in, out;
- struct virtio_chan *chan = client->trans;
- char *rdata = (char *)req->rc+sizeof(struct p9_fcall);
- unsigned long flags;
int err;
+ int in, out, out_sgs, in_sgs;
+ unsigned long flags;
+ struct virtio_chan *chan = client->trans;
+ struct scatterlist *sgs[2];
- P9_DPRINTK(P9_DEBUG_TRANS, "9p debug: virtio request\n");
+ p9_debug(P9_DEBUG_TRANS, "9p debug: virtio request\n");
-req_retry:
req->status = REQ_STATUS_SENT;
-
+req_retry:
spin_lock_irqsave(&chan->lock, flags);
- out = pack_sg_list(chan->sg, 0, VIRTQUEUE_NUM, req->tc->sdata,
- req->tc->size);
- in = pack_sg_list(chan->sg, out, VIRTQUEUE_NUM-out, rdata,
- client->msize);
- err = virtqueue_add_buf(chan->vq, chan->sg, out, in, req->tc);
+ 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;
@@ -239,24 +297,204 @@ req_retry:
if (err == -ERESTARTSYS)
return err;
- P9_DPRINTK(P9_DEBUG_TRANS, "9p:Retry virtio request\n");
+ p9_debug(P9_DEBUG_TRANS, "Retry virtio request\n");
goto req_retry;
} else {
spin_unlock_irqrestore(&chan->lock, flags);
- P9_DPRINTK(P9_DEBUG_TRANS,
- "9p debug: "
- "virtio rpc add_buf returned failure");
+ 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_DPRINTK(P9_DEBUG_TRANS, "9p debug: virtio request kicked\n");
+ p9_debug(P9_DEBUG_TRANS, "virtio request kicked\n");
return 0;
}
+static int p9_get_mapped_pages(struct virtio_chan *chan,
+ struct page **pages, char *data,
+ int nr_pages, int write, int kern_buf)
+{
+ int err;
+ if (!kern_buf) {
+ /*
+ * We allow only p9_max_pages pinned. We wait for the
+ * Other zc request to finish here
+ */
+ if (atomic_read(&vp_pinned) >= chan->p9_max_pages) {
+ err = wait_event_interruptible(vp_wq,
+ (atomic_read(&vp_pinned) < chan->p9_max_pages));
+ if (err == -ERESTARTSYS)
+ return err;
+ }
+ err = p9_payload_gup(data, &nr_pages, pages, write);
+ if (err < 0)
+ return err;
+ atomic_add(nr_pages, &vp_pinned);
+ } else {
+ /* kernel buffer, no need to pin pages */
+ int s, index = 0;
+ int count = nr_pages;
+ while (nr_pages) {
+ s = rest_of_page(data);
+ if (is_vmalloc_addr(data))
+ pages[index++] = vmalloc_to_page(data);
+ else
+ pages[index++] = kmap_to_page(data);
+ data += s;
+ nr_pages--;
+ }
+ nr_pages = count;
+ }
+ return nr_pages;
+}
+
+/**
+ * p9_virtio_zc_request - issue a zero copy request
+ * @client: client instance issuing the request
+ * @req: request to be issued
+ * @uidata: user bffer that should be ued for zero copy read
+ * @uodata: user buffer that shoud be user for zero copy write
+ * @inlen: read buffer size
+ * @olen: write buffer size
+ * @hdrlen: reader header size, This is the size of response protocol data
+ *
+ */
+static int
+p9_virtio_zc_request(struct p9_client *client, struct p9_req_t *req,
+ char *uidata, char *uodata, int inlen,
+ int outlen, int in_hdr_len, int kern_buf)
+{
+ int in, out, err, 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];
+
+ p9_debug(P9_DEBUG_TRANS, "virtio request\n");
+
+ if (uodata) {
+ out_nr_pages = p9_nr_pages(uodata, outlen);
+ out_pages = kmalloc(sizeof(struct page *) * out_nr_pages,
+ GFP_NOFS);
+ if (!out_pages) {
+ err = -ENOMEM;
+ goto err_out;
+ }
+ out_nr_pages = p9_get_mapped_pages(chan, out_pages, uodata,
+ out_nr_pages, 0, kern_buf);
+ if (out_nr_pages < 0) {
+ err = out_nr_pages;
+ kfree(out_pages);
+ out_pages = NULL;
+ goto err_out;
+ }
+ }
+ if (uidata) {
+ in_nr_pages = p9_nr_pages(uidata, inlen);
+ in_pages = kmalloc(sizeof(struct page *) * in_nr_pages,
+ GFP_NOFS);
+ if (!in_pages) {
+ err = -ENOMEM;
+ goto err_out;
+ }
+ in_nr_pages = p9_get_mapped_pages(chan, in_pages, uidata,
+ in_nr_pages, 1, kern_buf);
+ if (in_nr_pages < 0) {
+ err = in_nr_pages;
+ kfree(in_pages);
+ in_pages = NULL;
+ goto err_out;
+ }
+ }
+ 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, uodata, outlen);
+ }
+
+ /*
+ * Take care of in data
+ * For example TREAD have 11.
+ * 11 is the read/write header = PDU Header(7) + IO Size (4).
+ * Arrange in such a way that server places header in the
+ * alloced memory and payload onto the user buffer.
+ */
+ in = pack_sg_list(chan->sg, out,
+ VIRTQUEUE_NUM, req->rc->sdata, in_hdr_len);
+ if (in)
+ 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, uidata, 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 (!kern_buf) {
+ if (in_pages) {
+ p9_release_pages(in_pages, in_nr_pages);
+ atomic_sub(in_nr_pages, &vp_pinned);
+ }
+ if (out_pages) {
+ p9_release_pages(out_pages, out_nr_pages);
+ atomic_sub(out_nr_pages, &vp_pinned);
+ }
+ /* wakeup anybody waiting for slots to pin pages */
+ wake_up(&vp_wq);
+ }
+ kfree(in_pages);
+ kfree(out_pages);
+ return err;
+}
+
static ssize_t p9_mount_tag_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
@@ -288,7 +526,7 @@ static int p9_virtio_probe(struct virtio_device *vdev)
chan = kmalloc(sizeof(struct virtio_chan), GFP_KERNEL);
if (!chan) {
- printk(KERN_ERR "9p: Failed to allocate virtio 9P channel\n");
+ pr_err("Failed to allocate virtio 9P channel\n");
err = -ENOMEM;
goto fail;
}
@@ -308,9 +546,7 @@ static int p9_virtio_probe(struct virtio_device *vdev)
chan->inuse = false;
if (virtio_has_feature(vdev, VIRTIO_9P_MOUNT_TAG)) {
- vdev->config->get(vdev,
- offsetof(struct virtio_9p_config, tag_len),
- &tag_len, sizeof(tag_len));
+ virtio_cread(vdev, struct virtio_9p_config, tag_len, &tag_len);
} else {
err = -EINVAL;
goto out_free_vq;
@@ -320,8 +556,9 @@ static int p9_virtio_probe(struct virtio_device *vdev)
err = -ENOMEM;
goto out_free_vq;
}
- vdev->config->get(vdev, offsetof(struct virtio_9p_config, tag),
- tag, tag_len);
+
+ 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);
@@ -335,10 +572,16 @@ static int p9_virtio_probe(struct virtio_device *vdev)
}
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;
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:
@@ -387,7 +630,7 @@ p9_virtio_create(struct p9_client *client, const char *devname, char *args)
mutex_unlock(&virtio_9p_lock);
if (!found) {
- printk(KERN_ERR "9p: no channels available\n");
+ pr_err("no channels available\n");
return ret;
}
@@ -408,13 +651,15 @@ static void p9_virtio_remove(struct virtio_device *vdev)
{
struct virtio_chan *chan = vdev->priv;
- BUG_ON(chan->inuse);
+ if (chan->inuse)
+ p9_virtio_close(chan->client);
vdev->config->del_vqs(vdev);
mutex_lock(&virtio_9p_lock);
list_del(&chan->chan_list);
mutex_unlock(&virtio_9p_lock);
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);
@@ -446,9 +691,16 @@ static struct p9_trans_module p9_virtio_trans = {
.create = p9_virtio_create,
.close = p9_virtio_close,
.request = p9_virtio_request,
+ .zc_request = p9_virtio_zc_request,
.cancel = p9_virtio_cancel,
- .maxsize = PAGE_SIZE*16,
- .def = 0,
+ /*
+ * 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,
};
diff --git a/net/9p/util.c b/net/9p/util.c
index e048701a72d..59f278e64f5 100644
--- a/net/9p/util.c
+++ b/net/9p/util.c
@@ -67,7 +67,7 @@ EXPORT_SYMBOL(p9_idpool_create);
/**
* p9_idpool_destroy - create a new per-connection id pool
- * @p: idpool to destory
+ * @p: idpool to destroy
*/
void p9_idpool_destroy(struct p9_idpool *p)
@@ -87,26 +87,21 @@ EXPORT_SYMBOL(p9_idpool_destroy);
int p9_idpool_get(struct p9_idpool *p)
{
- int i = 0;
- int error;
+ int i;
unsigned long flags;
-retry:
- if (idr_pre_get(&p->pool, GFP_KERNEL) == 0)
- return 0;
-
+ idr_preload(GFP_NOFS);
spin_lock_irqsave(&p->lock, flags);
/* no need to store exactly p, we just need something non-null */
- error = idr_get_new(&p->pool, p, &i);
- spin_unlock_irqrestore(&p->lock, flags);
+ i = idr_alloc(&p->pool, p, 0, 0, GFP_NOWAIT);
- if (error == -EAGAIN)
- goto retry;
- else if (error)
+ spin_unlock_irqrestore(&p->lock, flags);
+ idr_preload_end();
+ if (i < 0)
return -1;
- P9_DPRINTK(P9_DEBUG_MUX, " id %d pool %p\n", i, p);
+ p9_debug(P9_DEBUG_MUX, " id %d pool %p\n", i, p);
return i;
}
EXPORT_SYMBOL(p9_idpool_get);
@@ -124,7 +119,7 @@ void p9_idpool_put(int id, struct p9_idpool *p)
{
unsigned long flags;
- P9_DPRINTK(P9_DEBUG_MUX, " id %d pool %p\n", id, p);
+ p9_debug(P9_DEBUG_MUX, " id %d pool %p\n", id, p);
spin_lock_irqsave(&p->lock, flags);
idr_remove(&p->pool, id);
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2025-11-09 19:37:02 +0000