diff options
Diffstat (limited to 'net')
118 files changed, 6186 insertions, 4870 deletions
diff --git a/net/802/fc.c b/net/802/fc.c index cb3475ea6fda..34cf1ee014b8 100644 --- a/net/802/fc.c +++ b/net/802/fc.c @@ -82,13 +82,13 @@ static int fc_header(struct sk_buff *skb, struct net_device *dev, static int fc_rebuild_header(struct sk_buff *skb) { +#ifdef CONFIG_INET struct fch_hdr *fch=(struct fch_hdr *)skb->data; struct fcllc *fcllc=(struct fcllc *)(skb->data+sizeof(struct fch_hdr)); if(fcllc->ethertype != htons(ETH_P_IP)) { printk("fc_rebuild_header: Don't know how to resolve type %04X addresses ?\n", ntohs(fcllc->ethertype)); return 0; } -#ifdef CONFIG_INET return arp_find(fch->daddr, skb); #else return 0; diff --git a/net/802/psnap.c b/net/802/psnap.c index b3cfe5a14fca..70980baeb682 100644 --- a/net/802/psnap.c +++ b/net/802/psnap.c @@ -1,7 +1,7 @@ /* * SNAP data link layer. Derived from 802.2 * - * Alan Cox <Alan.Cox@linux.org>, + * Alan Cox <alan@lxorguk.ukuu.org.uk>, * from the 802.2 layer by Greg Page. * Merged in additions from Greg Page's psnap.c. * diff --git a/net/9p/Makefile b/net/9p/Makefile index 519219480db1..1041b7bd12e2 100644 --- a/net/9p/Makefile +++ b/net/9p/Makefile @@ -4,10 +4,9 @@ obj-$(CONFIG_NET_9P_VIRTIO) += 9pnet_virtio.o 9pnet-objs := \ mod.o \ client.o \ - conv.o \ error.o \ - fcprint.o \ util.o \ + protocol.o \ trans_fd.o \ 9pnet_virtio-objs := \ diff --git a/net/9p/client.c b/net/9p/client.c index 10e320307ec0..bbac2f72b4d2 100644 --- a/net/9p/client.c +++ b/net/9p/client.c @@ -33,12 +33,9 @@ #include <linux/uaccess.h> #include <net/9p/9p.h> #include <linux/parser.h> -#include <net/9p/transport.h> #include <net/9p/client.h> - -static struct p9_fid *p9_fid_create(struct p9_client *clnt); -static void p9_fid_destroy(struct p9_fid *fid); -static struct p9_stat *p9_clone_stat(struct p9_stat *st, int dotu); +#include <net/9p/transport.h> +#include "protocol.h" /* * Client Option Parsing (code inspired by NFS code) @@ -52,13 +49,16 @@ enum { Opt_err, }; -static match_table_t tokens = { +static const match_table_t tokens = { {Opt_msize, "msize=%u"}, {Opt_legacy, "noextend"}, {Opt_trans, "trans=%s"}, {Opt_err, NULL}, }; +static struct p9_req_t * +p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...); + /** * v9fs_parse_options - parse mount options into session structure * @options: options string passed from mount @@ -124,31 +124,585 @@ static int parse_opts(char *opts, struct p9_client *clnt) return ret; } +/** + * 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 + * ids which did not previously have a slot. + * + * this code relies on the client spinlock to manage locks, its + * possible we should switch to something else, but I'd rather + * stick with something low-overhead for the common case. + * + */ + +static struct p9_req_t *p9_tag_alloc(struct p9_client *c, u16 tag) +{ + unsigned long flags; + int row, col; + struct p9_req_t *req; + + /* This looks up the original request by tag so we know which + * buffer to read the data into */ + tag++; + + if (tag >= c->max_tag) { + spin_lock_irqsave(&c->lock, flags); + /* check again since original check was outside of lock */ + while (tag >= c->max_tag) { + row = (tag / P9_ROW_MAXTAG); + c->reqs[row] = kcalloc(P9_ROW_MAXTAG, + sizeof(struct p9_req_t), GFP_ATOMIC); + + if (!c->reqs[row]) { + printk(KERN_ERR "Couldn't grow tag array\n"); + return ERR_PTR(-ENOMEM); + } + for (col = 0; col < P9_ROW_MAXTAG; col++) { + c->reqs[row][col].status = REQ_STATUS_IDLE; + c->reqs[row][col].tc = NULL; + } + c->max_tag += P9_ROW_MAXTAG; + } + spin_unlock_irqrestore(&c->lock, flags); + } + row = tag / P9_ROW_MAXTAG; + col = tag % P9_ROW_MAXTAG; + + req = &c->reqs[row][col]; + if (!req->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); + } + 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); + 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; + } + + p9pdu_reset(req->tc); + p9pdu_reset(req->rc); + + req->flush_tag = 0; + req->tc->tag = tag-1; + req->status = REQ_STATUS_ALLOC; + + return &c->reqs[row][col]; +} + +/** + * p9_tag_lookup - lookup a request by tag + * @c: client session to lookup tag within + * @tag: numeric id for transaction + * + */ + +struct p9_req_t *p9_tag_lookup(struct p9_client *c, u16 tag) +{ + int row, col; + + /* This looks up the original request by tag so we know which + * buffer to read the data into */ + tag++; + + BUG_ON(tag >= c->max_tag); + + row = tag / P9_ROW_MAXTAG; + col = tag % P9_ROW_MAXTAG; + + return &c->reqs[row][col]; +} +EXPORT_SYMBOL(p9_tag_lookup); + +/** + * p9_tag_init - setup tags structure and contents + * @tags: tags structure from the client struct + * + * This initializes the tags structure for each client instance. + * + */ + +static int p9_tag_init(struct p9_client *c) +{ + int err = 0; + + c->tagpool = p9_idpool_create(); + if (IS_ERR(c->tagpool)) { + err = PTR_ERR(c->tagpool); + c->tagpool = NULL; + goto error; + } + + p9_idpool_get(c->tagpool); /* reserve tag 0 */ + + c->max_tag = 0; +error: + return err; +} /** - * p9_client_rpc - sends 9P request and waits until a response is available. - * The function can be interrupted. - * @c: client data - * @tc: request to be sent - * @rc: pointer where a pointer to the response is stored + * p9_tag_cleanup - cleans up tags structure and reclaims resources + * @tags: tags structure from the client struct + * + * This frees resources associated with the tags structure + * */ +static void p9_tag_cleanup(struct p9_client *c) +{ + int row, col; + + /* check to insure all requests are idle */ + for (row = 0; row < (c->max_tag/P9_ROW_MAXTAG); row++) { + for (col = 0; col < P9_ROW_MAXTAG; col++) { + if (c->reqs[row][col].status != REQ_STATUS_IDLE) { + P9_DPRINTK(P9_DEBUG_MUX, + "Attempting to cleanup non-free tag %d,%d\n", + row, col); + /* TODO: delay execution of cleanup */ + return; + } + } + } + + if (c->tagpool) + p9_idpool_destroy(c->tagpool); + + /* free requests associated with tags */ + for (row = 0; row < (c->max_tag/P9_ROW_MAXTAG); row++) { + for (col = 0; col < P9_ROW_MAXTAG; col++) { + kfree(c->reqs[row][col].wq); + kfree(c->reqs[row][col].tc); + kfree(c->reqs[row][col].rc); + } + kfree(c->reqs[row]); + } + c->max_tag = 0; +} + +/** + * p9_free_req - free a request and clean-up as necessary + * c: client state + * r: request to release + * + */ + +static void p9_free_req(struct p9_client *c, struct p9_req_t *r) +{ + int tag = r->tc->tag; + P9_DPRINTK(P9_DEBUG_MUX, "clnt %p req %p tag: %d\n", c, r, tag); + + r->status = REQ_STATUS_IDLE; + if (tag != P9_NOTAG && p9_idpool_check(tag, c->tagpool)) + p9_idpool_put(tag, c->tagpool); + + /* if this was a flush request we have to free response fcall */ + if (r->rc->id == P9_RFLUSH) { + kfree(r->tc); + kfree(r->rc); + } +} + +/** + * p9_client_cb - call back from transport to client + * c: client state + * req: request received + * + */ +void p9_client_cb(struct p9_client *c, struct p9_req_t *req) +{ + struct p9_req_t *other_req; + unsigned long flags; + + P9_DPRINTK(P9_DEBUG_MUX, " tag %d\n", req->tc->tag); + + if (req->status == REQ_STATUS_ERROR) + wake_up(req->wq); + + if (req->flush_tag) { /* flush receive path */ + P9_DPRINTK(P9_DEBUG_9P, "<<< RFLUSH %d\n", req->tc->tag); + spin_lock_irqsave(&c->lock, flags); + other_req = p9_tag_lookup(c, req->flush_tag); + if (other_req->status != REQ_STATUS_FLSH) /* stale flush */ + spin_unlock_irqrestore(&c->lock, flags); + else { + other_req->status = REQ_STATUS_FLSHD; + spin_unlock_irqrestore(&c->lock, flags); + wake_up(other_req->wq); + } + p9_free_req(c, req); + } else { /* normal receive path */ + P9_DPRINTK(P9_DEBUG_MUX, "normal: tag %d\n", req->tc->tag); + spin_lock_irqsave(&c->lock, flags); + if (req->status != REQ_STATUS_FLSHD) + req->status = REQ_STATUS_RCVD; + spin_unlock_irqrestore(&c->lock, flags); + wake_up(req->wq); + P9_DPRINTK(P9_DEBUG_MUX, "wakeup: %d\n", req->tc->tag); + } +} +EXPORT_SYMBOL(p9_client_cb); + +/** + * p9_parse_header - parse header arguments out of a packet + * @pdu: packet to parse + * @size: size of packet + * @type: type of request + * @tag: tag of packet + * @rewind: set if we need to rewind offset afterwards + */ + int -p9_client_rpc(struct p9_client *c, struct p9_fcall *tc, - struct p9_fcall **rc) +p9_parse_header(struct p9_fcall *pdu, int32_t *size, int8_t *type, int16_t *tag, + int rewind) { - return c->trans->rpc(c->trans, tc, rc); + int8_t r_type; + int16_t r_tag; + int32_t r_size; + int offset = pdu->offset; + int err; + + pdu->offset = 0; + if (pdu->size == 0) + pdu->size = 7; + + err = p9pdu_readf(pdu, 0, "dbw", &r_size, &r_type, &r_tag); + if (err) + goto rewind_and_exit; + + pdu->size = r_size; + pdu->id = r_type; + pdu->tag = r_tag; + + P9_DPRINTK(P9_DEBUG_9P, "<<< size=%d type: %d tag: %d\n", pdu->size, + pdu->id, pdu->tag); + + if (type) + *type = r_type; + if (tag) + *tag = r_tag; + if (size) + *size = r_size; + + +rewind_and_exit: + if (rewind) + pdu->offset = offset; + return err; } +EXPORT_SYMBOL(p9_parse_header); + +/** + * p9_check_errors - check 9p packet for error return and process it + * @c: current client instance + * @req: request to parse and check for error conditions + * + * returns error code if one is discovered, otherwise returns 0 + * + * this will have to be more complicated if we have multiple + * error packet types + */ + +static int p9_check_errors(struct p9_client *c, struct p9_req_t *req) +{ + int8_t type; + int err; + + err = p9_parse_header(req->rc, NULL, &type, NULL, 0); + if (err) { + P9_DPRINTK(P9_DEBUG_ERROR, "couldn't parse header %d\n", err); + return err; + } + + if (type == P9_RERROR) { + int ecode; + char *ename; + + err = p9pdu_readf(req->rc, c->dotu, "s?d", &ename, &ecode); + if (err) { + P9_DPRINTK(P9_DEBUG_ERROR, "couldn't parse error%d\n", + err); + return err; + } + + if (c->dotu) + err = -ecode; + + if (!err) { + err = p9_errstr2errno(ename, strlen(ename)); + + /* string match failed */ + if (!err) + err = -ESERVERFAULT; + } + + P9_DPRINTK(P9_DEBUG_9P, "<<< RERROR (%d) %s\n", -ecode, ename); + + kfree(ename); + } else + err = 0; + + return err; +} + +/** + * p9_client_flush - flush (cancel) a request + * c: client state + * req: request to cancel + * + * This sents a flush for a particular requests and links + * the flush request to the original request. The current + * code only supports a single flush request although the protocol + * allows for multiple flush requests to be sent for a single request. + * + */ + +static int p9_client_flush(struct p9_client *c, struct p9_req_t *oldreq) +{ + struct p9_req_t *req; + int16_t oldtag; + int err; + + err = p9_parse_header(oldreq->tc, NULL, NULL, &oldtag, 1); + if (err) + return err; + + P9_DPRINTK(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); + + req->flush_tag = oldtag; + + /* we don't free anything here because RPC isn't complete */ + 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, ...) +{ + va_list ap; + int tag, err; + struct p9_req_t *req; + unsigned long flags; + int sigpending; + int flushed = 0; + + P9_DPRINTK(P9_DEBUG_MUX, "client %p op %d\n", c, type); + + if (c->status != Connected) + 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); + if (tag < 0) + return ERR_PTR(-ENOMEM); + } + + req = p9_tag_alloc(c, tag); + 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->dotu, fmt, ap); + va_end(ap); + p9pdu_finalize(req->tc); + + err = c->trans_mod->request(c, req); + if (err < 0) { + c->status = Disconnected; + goto reterr; + } + + /* if it was a flush we just transmitted, return our tag */ + if (type == P9_TFLUSH) + return req; +again: + P9_DPRINTK(P9_DEBUG_MUX, "wait %p tag: %d\n", req->wq, tag); + err = wait_event_interruptible(*req->wq, + req->status >= REQ_STATUS_RCVD); + P9_DPRINTK(P9_DEBUG_MUX, "wait %p tag: %d returned %d (flushed=%d)\n", + req->wq, tag, err, flushed); + + if (req->status == REQ_STATUS_ERROR) { + P9_DPRINTK(P9_DEBUG_ERROR, "req_status error %d\n", req->t_err); + err = req->t_err; + } else if (err == -ERESTARTSYS && flushed) { + P9_DPRINTK(P9_DEBUG_MUX, "flushed - going again\n"); + goto again; + } else if (req->status == REQ_STATUS_FLSHD) { + P9_DPRINTK(P9_DEBUG_MUX, "flushed - erestartsys\n"); + err = -ERESTARTSYS; + } + + if ((err == -ERESTARTSYS) && (c->status == Connected) && (!flushed)) { + P9_DPRINTK(P9_DEBUG_MUX, "flushing\n"); + spin_lock_irqsave(&c->lock, flags); + if (req->status == REQ_STATUS_SENT) + req->status = REQ_STATUS_FLSH; + spin_unlock_irqrestore(&c->lock, flags); + sigpending = 1; + flushed = 1; + clear_thread_flag(TIF_SIGPENDING); + + if (c->trans_mod->cancel(c, req)) { + err = p9_client_flush(c, req); + if (err == 0) + goto again; + } + } + + if (sigpending) { + spin_lock_irqsave(¤t->sighand->siglock, flags); + recalc_sigpending(); + spin_unlock_irqrestore(¤t->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); + 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); +} + +static struct p9_fid *p9_fid_create(struct p9_client *clnt) +{ + int err; + struct p9_fid *fid; + + P9_DPRINTK(P9_DEBUG_FID, "clnt %p\n", clnt); + fid = kmalloc(sizeof(struct p9_fid), GFP_KERNEL); + if (!fid) + return ERR_PTR(-ENOMEM); + + fid->fid = p9_idpool_get(clnt->fidpool); + if (fid->fid < 0) { + err = -ENOSPC; + goto error; + } + + memset(&fid->qid, 0, sizeof(struct p9_qid)); + fid->mode = -1; + fid->rdir_fpos = 0; + fid->uid = current->fsuid; + fid->clnt = clnt; + fid->aux = NULL; + + spin_lock(&clnt->lock); + list_add(&fid->flist, &clnt->fidlist); + spin_unlock(&clnt->lock); + + return fid; + +error: + kfree(fid); + return ERR_PTR(err); +} + +static void p9_fid_destroy(struct p9_fid *fid) +{ + struct p9_client *clnt; + + P9_DPRINTK(P9_DEBUG_FID, "fid %d\n", fid->fid); + clnt = fid->clnt; + p9_idpool_put(fid->fid, clnt->fidpool); + spin_lock(&clnt->lock); + list_del(&fid->flist); + spin_unlock(&clnt->lock); + kfree(fid); +} + +int p9_client_version(struct p9_client *c) +{ + int err = 0; + struct p9_req_t *req; + char *version; + int msize; + + P9_DPRINTK(P9_DEBUG_9P, ">>> TVERSION msize %d extended %d\n", + c->msize, c->dotu); + req = p9_client_rpc(c, P9_TVERSION, "ds", c->msize, + c->dotu ? "9P2000.u" : "9P2000"); + if (IS_ERR(req)) + return PTR_ERR(req); + + err = p9pdu_readf(req->rc, c->dotu, "ds", &msize, &version); + if (err) { + P9_DPRINTK(P9_DEBUG_9P, "version error %d\n", err); + p9pdu_dump(1, req->rc); + goto error; + } + + P9_DPRINTK(P9_DEBUG_9P, "<<< RVERSION msize %d %s\n", msize, version); + if (!memcmp(version, "9P2000.u", 8)) + c->dotu = 1; + else if (!memcmp(version, "9P2000", 6)) + c->dotu = 0; + else { + err = -EREMOTEIO; + goto error; + } + + if (msize < c->msize) + c->msize = msize; + +error: + kfree(version); + p9_free_req(c, req); + + return err; +} +EXPORT_SYMBOL(p9_client_version); struct p9_client *p9_client_create(const char *dev_name, char *options) { - int err, n; + int err; struct p9_client *clnt; - struct p9_fcall *tc, *rc; - struct p9_str *version; err = 0; - tc = NULL; - rc = NULL; clnt = kmalloc(sizeof(struct p9_client), GFP_KERNEL); if (!clnt) return ERR_PTR(-ENOMEM); @@ -164,6 +718,8 @@ struct p9_client *p9_client_create(const char *dev_name, char *options) goto error; } + p9_tag_init(clnt); + err = parse_opts(options, clnt); if (err < 0) goto error; @@ -175,53 +731,23 @@ struct p9_client *p9_client_create(const char *dev_name, char *options) goto error; } - P9_DPRINTK(P9_DEBUG_9P, "clnt %p trans %p msize %d dotu %d\n", + P9_DPRINTK(P9_DEBUG_MUX, "clnt %p trans %p msize %d dotu %d\n", clnt, clnt->trans_mod, clnt->msize, clnt->dotu); - - clnt->trans = clnt->trans_mod->create(dev_name, options, clnt->msize, - clnt->dotu); - if (IS_ERR(clnt->trans)) { - err = PTR_ERR(clnt->trans); - clnt->trans = NULL; + err = clnt->trans_mod->create(clnt, dev_name, options); + if (err) goto error; - } if ((clnt->msize+P9_IOHDRSZ) > clnt->trans_mod->maxsize) clnt->msize = clnt->trans_mod->maxsize-P9_IOHDRSZ; - tc = p9_create_tversion(clnt->msize, clnt->dotu?"9P2000.u":"9P2000"); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; - goto error; - } - - err = p9_client_rpc(clnt, tc, &rc); + err = p9_client_version(clnt); if (err) goto error; - version = &rc->params.rversion.version; - if (version->len == 8 && !memcmp(version->str, "9P2000.u", 8)) - clnt->dotu = 1; - else if (version->len == 6 && !memcmp(version->str, "9P2000", 6)) - clnt->dotu = 0; - else { - err = -EREMOTEIO; - goto error; - } - - n = rc->params.rversion.msize; - if (n < clnt->msize) - clnt->msize = n; - - kfree(tc); - kfree(rc); return clnt; error: - kfree(tc); - kfree(rc); p9_client_destroy(clnt); return ERR_PTR(err); } @@ -231,13 +757,10 @@ void p9_client_destroy(struct p9_client *clnt) { struct p9_fid *fid, *fidptr; - P9_DPRINTK(P9_DEBUG_9P, "clnt %p\n", clnt); + P9_DPRINTK(P9_DEBUG_MUX, "clnt %p\n", clnt); - if (clnt->trans) { - clnt->trans->close(clnt->trans); - kfree(clnt->trans); - clnt->trans = NULL; - } + if (clnt->trans_mod) + clnt->trans_mod->close(clnt); v9fs_put_trans(clnt->trans_mod); @@ -247,6 +770,8 @@ void p9_client_destroy(struct p9_client *clnt) if (clnt->fidpool) p9_idpool_destroy(clnt->fidpool); + p9_tag_cleanup(clnt); + kfree(clnt); } EXPORT_SYMBOL(p9_client_destroy); @@ -254,7 +779,7 @@ EXPORT_SYMBOL(p9_client_destroy); void p9_client_disconnect(struct p9_client *clnt) { P9_DPRINTK(P9_DEBUG_9P, "clnt %p\n", clnt); - clnt->trans->status = Disconnected; + clnt->status = Disconnected; } EXPORT_SYMBOL(p9_client_disconnect); @@ -262,14 +787,13 @@ struct p9_fid *p9_client_attach(struct p9_client *clnt, struct p9_fid *afid, char *uname, u32 n_uname, char *aname) { int err; - struct p9_fcall *tc, *rc; + struct p9_req_t *req; struct p9_fid *fid; + struct p9_qid qid; - P9_DPRINTK(P9_DEBUG_9P, "clnt %p afid %d uname %s aname %s\n", - clnt, afid?afid->fid:-1, uname, aname); + P9_DPRINTK(P9_DEBUG_9P, ">>> TATTACH afid %d uname %s aname %s\n", + afid ? afid->fid : -1, uname, aname); err = 0; - tc = NULL; - rc = NULL; fid = p9_fid_create(clnt); if (IS_ERR(fid)) { @@ -278,73 +802,77 @@ struct p9_fid *p9_client_attach(struct p9_client *clnt, struct p9_fid *afid, goto error; } - tc = p9_create_tattach(fid->fid, afid?afid->fid:P9_NOFID, uname, aname, - n_uname, clnt->dotu); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; + req = p9_client_rpc(clnt, P9_TATTACH, "ddss?d", fid->fid, + afid ? afid->fid : P9_NOFID, uname, aname, n_uname); + if (IS_ERR(req)) { + err = PTR_ERR(req); goto error; } - err = p9_client_rpc(clnt, tc, &rc); - if (err) + err = p9pdu_readf(req->rc, clnt->dotu, "Q", &qid); + if (err) { + p9pdu_dump(1, req->rc); + p9_free_req(clnt, req); goto error; + } + + P9_DPRINTK(P9_DEBUG_9P, "<<< RATTACH qid %x.%llx.%x\n", + qid.type, qid.path, qid.version); - memmove(&fid->qid, &rc->params.rattach.qid, sizeof(struct p9_qid)); - kfree(tc); - kfree(rc); + memmove(&fid->qid, &qid, sizeof(struct p9_qid)); + + p9_free_req(clnt, req); return fid; error: - kfree(tc); - kfree(rc); if (fid) p9_fid_destroy(fid); return ERR_PTR(err); } EXPORT_SYMBOL(p9_client_attach); -struct p9_fid *p9_client_auth(struct p9_client *clnt, char *uname, - u32 n_uname, char *aname) +struct p9_fid * +p9_client_auth(struct p9_client *clnt, char *uname, u32 n_uname, char *aname) { int err; - struct p9_fcall *tc, *rc; - struct p9_fid *fid; + struct p9_req_t *req; + struct p9_qid qid; + struct p9_fid *afid; - P9_DPRINTK(P9_DEBUG_9P, "clnt %p uname %s aname %s\n", clnt, uname, - aname); + P9_DPRINTK(P9_DEBUG_9P, ">>> TAUTH uname %s aname %s\n", uname, aname); err = 0; - tc = NULL; - rc = NULL; - fid = p9_fid_create(clnt); - if (IS_ERR(fid)) { - err = PTR_ERR(fid); - fid = NULL; + afid = p9_fid_create(clnt); + if (IS_ERR(afid)) { + err = PTR_ERR(afid); + afid = NULL; goto error; } - tc = p9_create_tauth(fid->fid, uname, aname, n_uname, clnt->dotu); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; + req = p9_client_rpc(clnt, P9_TAUTH, "dss?d", + afid ? afid->fid : P9_NOFID, uname, aname, n_uname); + if (IS_ERR(req)) { + err = PTR_ERR(req); goto error; } - err = p9_client_rpc(clnt, tc, &rc); - if (err) + err = p9pdu_readf(req->rc, clnt->dotu, "Q", &qid); + if (err) { + p9pdu_dump(1, req->rc); + p9_free_req(clnt, req); goto error; + } - memmove(&fid->qid, &rc->params.rauth.qid, sizeof(struct p9_qid)); - kfree(tc); - kfree(rc); - return fid; + P9_DPRINTK(P9_DEBUG_9P, "<<< RAUTH qid %x.%llx.%x\n", + qid.type, qid.path, qid.version); + + memmove(&afid->qid, &qid, sizeof(struct p9_qid)); + p9_free_req(clnt, req); + return afid; error: - kfree(tc); - kfree(rc); - if (fid) - p9_fid_destroy(fid); + if (afid) + p9_fid_destroy(afid); return ERR_PTR(err); } EXPORT_SYMBOL(p9_client_auth); @@ -353,15 +881,13 @@ struct p9_fid *p9_client_walk(struct p9_fid *oldfid, int nwname, char **wnames, int clone) { int err; - struct p9_fcall *tc, *rc; struct p9_client *clnt; struct p9_fid *fid; + struct p9_qid *wqids; + struct p9_req_t *req; + int16_t nwqids, count; - P9_DPRINTK(P9_DEBUG_9P, "fid %d nwname %d wname[0] %s\n", - oldfid->fid, nwname, wnames?wnames[0]:NULL); err = 0; - tc = NULL; - rc = NULL; clnt = oldfid->clnt; if (clone) { fid = p9_fid_create(clnt); @@ -375,53 +901,49 @@ struct p9_fid *p9_client_walk(struct p9_fid *oldfid, int nwname, char **wnames, } else fid = oldfid; - tc = p9_create_twalk(oldfid->fid, fid->fid, nwname, wnames); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; + + P9_DPRINTK(P9_DEBUG_9P, ">>> TWALK fids %d,%d nwname %d wname[0] %s\n", + oldfid->fid, fid->fid, nwname, wnames ? wnames[0] : NULL); + + req = p9_client_rpc(clnt, P9_TWALK, "ddT", oldfid->fid, fid->fid, + nwname, wnames); + if (IS_ERR(req)) { + err = PTR_ERR(req); goto error; } - err = p9_client_rpc(clnt, tc, &rc); + err = p9pdu_readf(req->rc, clnt->dotu, "R", &nwqids, &wqids); if (err) { - if (rc && rc->id == P9_RWALK) - goto clunk_fid; - else - goto error; + p9pdu_dump(1, req->rc); + p9_free_req(clnt, req); + goto clunk_fid; } + p9_free_req(clnt, req); - if (rc->params.rwalk.nwqid != nwname) { + P9_DPRINTK(P9_DEBUG_9P, "<<< RWALK nwqid %d:\n", nwqids); + + if (nwqids != nwname) { err = -ENOENT; goto clunk_fid; } + for (count = 0; count < nwqids; count++) + P9_DPRINTK(P9_DEBUG_9P, "<<< [%d] %x.%llx.%x\n", + count, wqids[count].type, wqids[count].path, + wqids[count].version); + if (nwname) - memmove(&fid->qid, - &rc->params.rwalk.wqids[rc->params.rwalk.nwqid - 1], - sizeof(struct p9_qid)); + memmove(&fid->qid, &wqids[nwqids - 1], sizeof(struct p9_qid)); else fid->qid = oldfid->qid; - kfree(tc); - kfree(rc); return fid; clunk_fid: - kfree(tc); - kfree(rc); - rc = NULL; - tc = p9_create_tclunk(fid->fid); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; - goto error; - } - - p9_client_rpc(clnt, tc, &rc); + p9_client_clunk(fid); + fid = NULL; error: - kfree(tc); - kfree(rc); if (fid && (fid != oldfid)) p9_fid_destroy(fid); @@ -432,35 +954,39 @@ EXPORT_SYMBOL(p9_client_walk); int p9_client_open(struct p9_fid *fid, int mode) { int err; - struct p9_fcall *tc, *rc; struct p9_client *clnt; + struct p9_req_t *req; + struct p9_qid qid; + int iounit; - P9_DPRINTK(P9_DEBUG_9P, "fid %d mode %d\n", fid->fid, mode); + P9_DPRINTK(P9_DEBUG_9P, ">>> TOPEN fid %d mode %d\n", fid->fid, mode); err = 0; - tc = NULL; - rc = NULL; clnt = fid->clnt; if (fid->mode != -1) return -EINVAL; - tc = p9_create_topen(fid->fid, mode); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; - goto done; + req = p9_client_rpc(clnt, P9_TOPEN, "db", fid->fid, mode); + if (IS_ERR(req)) { + err = PTR_ERR(req); + goto error; } - err = p9_client_rpc(clnt, tc, &rc); - if (err) - goto done; + err = p9pdu_readf(req->rc, clnt->dotu, "Qd", &qid, &iounit); + if (err) { + p9pdu_dump(1, req->rc); + goto free_and_error; + } + + P9_DPRINTK(P9_DEBUG_9P, "<<< ROPEN qid %x.%llx.%x iounit %x\n", + qid.type, qid.path, qid.version, iounit); fid->mode = mode; - fid->iounit = rc->params.ropen.iounit; + fid->iounit = iounit; -done: - kfree(tc); - kfree(rc); +free_and_error: + p9_free_req(clnt, req); +error: return err; } EXPORT_SYMBOL(p9_client_open); @@ -469,37 +995,41 @@ int p9_client_fcreate(struct p9_fid *fid, char *name, u32 perm, int mode, char *extension) { int err; - struct p9_fcall *tc, *rc; struct p9_client *clnt; + struct p9_req_t *req; + struct p9_qid qid; + int iounit; - P9_DPRINTK(P9_DEBUG_9P, "fid %d name %s perm %d mode %d\n", fid->fid, - name, perm, mode); + P9_DPRINTK(P9_DEBUG_9P, ">>> TCREATE fid %d name %s perm %d mode %d\n", + fid->fid, name, perm, mode); err = 0; - tc = NULL; - rc = NULL; clnt = fid->clnt; if (fid->mode != -1) return -EINVAL; - tc = p9_create_tcreate(fid->fid, name, perm, mode, extension, - clnt->dotu); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; - goto done; + req = p9_client_rpc(clnt, P9_TCREATE, "dsdb?s", fid->fid, name, perm, + mode, extension); + if (IS_ERR(req)) { + err = PTR_ERR(req); + goto error; } - err = p9_client_rpc(clnt, tc, &rc); - if (err) - goto done; + err = p9pdu_readf(req->rc, clnt->dotu, "Qd", &qid, &iounit); + if (err) { + p9pdu_dump(1, req->rc); + goto free_and_error; + } + + P9_DPRINTK(P9_DEBUG_9P, "<<< RCREATE qid %x.%llx.%x iounit %x\n", + qid.type, qid.path, qid.version, iounit); fid->mode = mode; - fid->iounit = rc->params.ropen.iounit; + fid->iounit = iounit; -done: - kfree(tc); - kfree(rc); +free_and_error: + p9_free_req(clnt, req); +error: return err; } EXPORT_SYMBOL(p9_client_fcreate); @@ -507,31 +1037,25 @@ EXPORT_SYMBOL(p9_client_fcreate); int p9_client_clunk(struct p9_fid *fid) { int err; - struct p9_fcall *tc, *rc; struct p9_client *clnt; + struct p9_req_t *req; - P9_DPRINTK(P9_DEBUG_9P, "fid %d\n", fid->fid); + P9_DPRINTK(P9_DEBUG_9P, ">>> TCLUNK fid %d\n", fid->fid); err = 0; - tc = NULL; - rc = NULL; clnt = fid->clnt; - tc = p9_create_tclunk(fid->fid); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; - goto done; + req = p9_client_rpc(clnt, P9_TCLUNK, "d", fid->fid); + if (IS_ERR(req)) { + err = PTR_ERR(req); + goto error; } - err = p9_client_rpc(clnt, tc, &rc); - if (err) - goto done; + P9_DPRINTK(P9_DEBUG_9P, "<<< RCLUNK fid %d\n", fid->fid); + p9_free_req(clnt, req); p9_fid_destroy(fid); -done: - kfree(tc); - kfree(rc); +error: return err; } EXPORT_SYMBOL(p9_client_clunk); @@ -539,157 +1063,41 @@ EXPORT_SYMBOL(p9_client_clunk); int p9_client_remove(struct p9_fid *fid) { int err; - struct p9_fcall *tc, *rc; struct p9_client *clnt; + struct p9_req_t *req; - P9_DPRINTK(P9_DEBUG_9P, "fid %d\n", fid->fid); + P9_DPRINTK(P9_DEBUG_9P, ">>> TREMOVE fid %d\n", fid->fid); err = 0; - tc = NULL; - rc = NULL; clnt = fid->clnt; - tc = p9_create_tremove(fid->fid); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; - goto done; + req = p9_client_rpc(clnt, P9_TREMOVE, "d", fid->fid); + if (IS_ERR(req)) { + err = PTR_ERR(req); + goto error; } - err = p9_client_rpc(clnt, tc, &rc); - if (err) - goto done; + P9_DPRINTK(P9_DEBUG_9P, "<<< RREMOVE fid %d\n", fid->fid); + p9_free_req(clnt, req); p9_fid_destroy(fid); -done: - kfree(tc); - kfree(rc); - return err; -} -EXPORT_SYMBOL(p9_client_remove); - -int p9_client_read(struct p9_fid *fid, char *data, u64 offset, u32 count) -{ - int err, n, rsize, total; - struct p9_fcall *tc, *rc; - struct p9_client *clnt; - - P9_DPRINTK(P9_DEBUG_9P, "fid %d offset %llu %d\n", fid->fid, - (long long unsigned) offset, count); - err = 0; - tc = NULL; - rc = NULL; - clnt = fid->clnt; - total = 0; - - rsize = fid->iounit; - if (!rsize || rsize > clnt->msize-P9_IOHDRSZ) - rsize = clnt->msize - P9_IOHDRSZ; - - do { - if (count < rsize) - rsize = count; - - tc = p9_create_tread(fid->fid, offset, rsize); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; - goto error; - } - - err = p9_client_rpc(clnt, tc, &rc); - if (err) - goto error; - - n = rc->params.rread.count; - if (n > count) - n = count; - - memmove(data, rc->params.rread.data, n); - count -= n; - data += n; - offset += n; - total += n; - kfree(tc); - tc = NULL; - kfree(rc); - rc = NULL; - } while (count > 0 && n == rsize); - - return total; - error: - kfree(tc); - kfree(rc); return err; } -EXPORT_SYMBOL(p9_client_read); - -int p9_client_write(struct p9_fid *fid, char *data, u64 offset, u32 count) -{ - int err, n, rsize, total; - struct p9_fcall *tc, *rc; - struct p9_client *clnt; - - P9_DPRINTK(P9_DEBUG_9P, "fid %d offset %llu count %d\n", fid->fid, - (long long unsigned) offset, count); - err = 0; - tc = NULL; - rc = NULL; - clnt = fid->clnt; - total = 0; - - rsize = fid->iounit; - if (!rsize || rsize > clnt->msize-P9_IOHDRSZ) - rsize = clnt->msize - P9_IOHDRSZ; - - do { - if (count < rsize) - rsize = count; - - tc = p9_create_twrite(fid->fid, offset, rsize, data); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; - goto error; - } - - err = p9_client_rpc(clnt, tc, &rc); - if (err) - goto error; - - n = rc->params.rread.count; - count -= n; - data += n; - offset += n; - total += n; - kfree(tc); - tc = NULL; - kfree(rc); - rc = NULL; - } while (count > 0); - - return total; - -error: - kfree(tc); - kfree(rc); - return err; -} -EXPORT_SYMBOL(p9_client_write); +EXPORT_SYMBOL(p9_client_remove); int -p9_client_uread(struct p9_fid *fid, char __user *data, u64 offset, u32 count) +p9_client_read(struct p9_fid *fid, char *data, char __user *udata, u64 offset, + u32 count) { - int err, n, rsize, total; - struct p9_fcall *tc, *rc; + int err, rsize, total; struct p9_client *clnt; + struct p9_req_t *req; + char *dataptr; - P9_DPRINTK(P9_DEBUG_9P, "fid %d offset %llu count %d\n", fid->fid, + P9_DPRINTK(P9_DEBUG_9P, ">>> TREAD fid %d offset %llu %d\n", fid->fid, (long long unsigned) offset, count); err = 0; - tc = NULL; - rc = NULL; clnt = fid->clnt; total = 0; @@ -697,63 +1105,57 @@ p9_client_uread(struct p9_fid *fid, char __user *data, u64 offset, u32 count) if (!rsize || rsize > clnt->msize-P9_IOHDRSZ) rsize = clnt->msize - P9_IOHDRSZ; - do { - if (count < rsize) - rsize = count; + if (count < rsize) + rsize = count; - tc = p9_create_tread(fid->fid, offset, rsize); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; - goto error; - } + req = p9_client_rpc(clnt, P9_TREAD, "dqd", fid->fid, offset, rsize); + if (IS_ERR(req)) { + err = PTR_ERR(req); + goto error; + } - err = p9_client_rpc(clnt, tc, &rc); - if (err) - goto error; + err = p9pdu_readf(req->rc, clnt->dotu, "D", &count, &dataptr); + if (err) { + p9pdu_dump(1, req->rc); + goto free_and_error; + } + + P9_DPRINTK(P9_DEBUG_9P, "<<< RREAD count %d\n", count); - n = rc->params.rread.count; - if (n > count) - n = count; + if (data) { + memmove(data, dataptr, count); + data += count; + } - err = copy_to_user(data, rc->params.rread.data, n); + if (udata) { + err = copy_to_user(udata, dataptr, count); if (err) { err = -EFAULT; - goto error; + goto free_and_error; } + } - count -= n; - data += n; - offset += n; - total += n; - kfree(tc); - tc = NULL; - kfree(rc); - rc = NULL; - } while (count > 0 && n == rsize); - - return total; + p9_free_req(clnt, req); + return count; +free_and_error: + p9_free_req(clnt, req); error: - kfree(tc); - kfree(rc); return err; } -EXPORT_SYMBOL(p9_client_uread); +EXPORT_SYMBOL(p9_client_read); int -p9_client_uwrite(struct p9_fid *fid, const char __user *data, u64 offset, - u32 count) +p9_client_write(struct p9_fid *fid, char *data, const char __user *udata, + u64 offset, u32 count) { - int err, n, rsize, total; - struct p9_fcall *tc, *rc; + int err, rsize, total; struct p9_client *clnt; + struct p9_req_t *req; - P9_DPRINTK(P9_DEBUG_9P, "fid %d offset %llu count %d\n", fid->fid, - (long long unsigned) offset, count); + P9_DPRINTK(P9_DEBUG_9P, ">>> TWRITE fid %d offset %llu count %d\n", + fid->fid, (long long unsigned) offset, count); err = 0; - tc = NULL; - rc = NULL; clnt = fid->clnt; total = 0; @@ -761,325 +1163,114 @@ p9_client_uwrite(struct p9_fid *fid, const char __user *data, u64 offset, if (!rsize || rsize > clnt->msize-P9_IOHDRSZ) rsize = clnt->msize - P9_IOHDRSZ; - do { - if (count < rsize) - rsize = count; - - tc = p9_create_twrite_u(fid->fid, offset, rsize, data); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; - goto error; - } + 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); + if (IS_ERR(req)) { + err = PTR_ERR(req); + goto error; + } - err = p9_client_rpc(clnt, tc, &rc); - if (err) - goto error; + err = p9pdu_readf(req->rc, clnt->dotu, "d", &count); + if (err) { + p9pdu_dump(1, req->rc); + goto free_and_error; + } - n = rc->params.rread.count; - count -= n; - data += n; - offset += n; - total += n; - kfree(tc); - tc = NULL; - kfree(rc); - rc = NULL; - } while (count > 0); + P9_DPRINTK(P9_DEBUG_9P, "<<< RWRITE count %d\n", count); - return total; + p9_free_req(clnt, req); + return count; +free_and_error: + p9_free_req(clnt, req); error: - kfree(tc); - kfree(rc); return err; } -EXPORT_SYMBOL(p9_client_uwrite); - -int p9_client_readn(struct p9_fid *fid, char *data, u64 offset, u32 count) -{ - int n, total; - - P9_DPRINTK(P9_DEBUG_9P, "fid %d offset %llu count %d\n", fid->fid, - (long long unsigned) offset, count); - n = 0; - total = 0; - while (count) { - n = p9_client_read(fid, data, offset, count); - if (n <= 0) - break; - - data += n; - offset += n; - count -= n; - total += n; - } - - if (n < 0) - total = n; - - return total; -} -EXPORT_SYMBOL(p9_client_readn); +EXPORT_SYMBOL(p9_client_write); -struct p9_stat *p9_client_stat(struct p9_fid *fid) +struct p9_wstat *p9_client_stat(struct p9_fid *fid) { int err; - struct p9_fcall *tc, *rc; struct p9_client *clnt; - struct p9_stat *ret; + struct p9_wstat *ret = kmalloc(sizeof(struct p9_wstat), GFP_KERNEL); + struct p9_req_t *req; + u16 ignored; + + P9_DPRINTK(P9_DEBUG_9P, ">>> TSTAT fid %d\n", fid->fid); + + if (!ret) + return ERR_PTR(-ENOMEM); - P9_DPRINTK(P9_DEBUG_9P, "fid %d\n", fid->fid); err = 0; - tc = NULL; - rc = NULL; - ret = NULL; clnt = fid->clnt; - tc = p9_create_tstat(fid->fid); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; + req = p9_client_rpc(clnt, P9_TSTAT, "d", fid->fid); + if (IS_ERR(req)) { + err = PTR_ERR(req); goto error; } - err = p9_client_rpc(clnt, tc, &rc); - if (err) - goto error; - - ret = p9_clone_stat(&rc->params.rstat.stat, clnt->dotu); - if (IS_ERR(ret)) { - err = PTR_ERR(ret); - ret = NULL; - goto error; + err = p9pdu_readf(req->rc, clnt->dotu, "wS", &ignored, ret); + if (err) { + ret = ERR_PTR(err); + p9pdu_dump(1, req->rc); + goto free_and_error; } - kfree(tc); - kfree(rc); - return ret; - + P9_DPRINTK(P9_DEBUG_9P, + "<<< RSTAT sz=%x type=%x dev=%x qid=%x.%llx.%x\n" + "<<< mode=%8.8x atime=%8.8x mtime=%8.8x length=%llx\n" + "<<< name=%s uid=%s gid=%s muid=%s extension=(%s)\n" + "<<< uid=%d gid=%d n_muid=%d\n", + ret->size, ret->type, ret->dev, ret->qid.type, + ret->qid.path, ret->qid.version, ret->mode, + ret->atime, ret->mtime, ret->length, ret->name, + ret->uid, ret->gid, ret->muid, ret->extension, + ret->n_uid, ret->n_gid, ret->n_muid); + +free_and_error: + p9_free_req(clnt, req); error: - kfree(tc); - kfree(rc); - kfree(ret); - return ERR_PTR(err); + return ret; } EXPORT_SYMBOL(p9_client_stat); int p9_client_wstat(struct p9_fid *fid, struct p9_wstat *wst) { int err; - struct p9_fcall *tc, *rc; + struct p9_req_t *req; struct p9_client *clnt; - P9_DPRINTK(P9_DEBUG_9P, "fid %d\n", fid->fid); + P9_DPRINTK(P9_DEBUG_9P, ">>> TWSTAT fid %d\n", fid->fid); + P9_DPRINTK(P9_DEBUG_9P, + " sz=%x type=%x dev=%x qid=%x.%llx.%x\n" + " mode=%8.8x atime=%8.8x mtime=%8.8x length=%llx\n" + " name=%s uid=%s gid=%s muid=%s extension=(%s)\n" + " uid=%d gid=%d n_muid=%d\n", + wst->size, wst->type, wst->dev, wst->qid.type, + wst->qid.path, wst->qid.version, wst->mode, + wst->atime, wst->mtime, wst->length, wst->name, + wst->uid, wst->gid, wst->muid, wst->extension, + wst->n_uid, wst->n_gid, wst->n_muid); err = 0; - tc = NULL; - rc = NULL; clnt = fid->clnt; - tc = p9_create_twstat(fid->fid, wst, clnt->dotu); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; - goto done; - } - - err = p9_client_rpc(clnt, tc, &rc); - -done: - kfree(tc); - kfree(rc); - return err; -} -EXPORT_SYMBOL(p9_client_wstat); - -struct p9_stat *p9_client_dirread(struct p9_fid *fid, u64 offset) -{ - int err, n, m; - struct p9_fcall *tc, *rc; - struct p9_client *clnt; - struct p9_stat st, *ret; - - P9_DPRINTK(P9_DEBUG_9P, "fid %d offset %llu\n", fid->fid, - (long long unsigned) offset); - err = 0; - tc = NULL; - rc = NULL; - ret = NULL; - clnt = fid->clnt; - - /* if the offset is below or above the current response, free it */ - if (offset < fid->rdir_fpos || (fid->rdir_fcall && - offset >= fid->rdir_fpos+fid->rdir_fcall->params.rread.count)) { - fid->rdir_pos = 0; - if (fid->rdir_fcall) - fid->rdir_fpos += fid->rdir_fcall->params.rread.count; - - kfree(fid->rdir_fcall); - fid->rdir_fcall = NULL; - if (offset < fid->rdir_fpos) - fid->rdir_fpos = 0; - } - - if (!fid->rdir_fcall) { - n = fid->iounit; - if (!n || n > clnt->msize-P9_IOHDRSZ) - n = clnt->msize - P9_IOHDRSZ; - - while (1) { - if (fid->rdir_fcall) { - fid->rdir_fpos += - fid->rdir_fcall->params.rread.count; - kfree(fid->rdir_fcall); - fid->rdir_fcall = NULL; - } - - tc = p9_create_tread(fid->fid, fid->rdir_fpos, n); - if (IS_ERR(tc)) { - err = PTR_ERR(tc); - tc = NULL; - goto error; - } - - err = p9_client_rpc(clnt, tc, &rc); - if (err) - goto error; - - n = rc->params.rread.count; - if (n == 0) - goto done; - - fid->rdir_fcall = rc; - rc = NULL; - if (offset >= fid->rdir_fpos && - offset < fid->rdir_fpos+n) - break; - } - - fid->rdir_pos = 0; - } - - m = offset - fid->rdir_fpos; - if (m < 0) - goto done; - - n = p9_deserialize_stat(fid->rdir_fcall->params.rread.data + m, - fid->rdir_fcall->params.rread.count - m, &st, clnt->dotu); - - if (!n) { - err = -EIO; - goto error; - } - - fid->rdir_pos += n; - st.size = n; - ret = p9_clone_stat(&st, clnt->dotu); - if (IS_ERR(ret)) { - err = PTR_ERR(ret); - ret = NULL; - goto error; - } - -done: - kfree(tc); - kfree(rc); - return ret; - -error: - kfree(tc); - kfree(rc); - kfree(ret); - return ERR_PTR(err); -} -EXPORT_SYMBOL(p9_client_dirread); - -static struct p9_stat *p9_clone_stat(struct p9_stat *st, int dotu) -{ - int n; - char *p; - struct p9_stat *ret; - - n = sizeof(struct p9_stat) + st->name.len + st->uid.len + st->gid.len + - st->muid.len; - - if (dotu) - n += st->extension.len; - - ret = kmalloc(n, GFP_KERNEL); - if (!ret) - return ERR_PTR(-ENOMEM); - - memmove(ret, st, sizeof(struct p9_stat)); - p = ((char *) ret) + sizeof(struct p9_stat); - memmove(p, st->name.str, st->name.len); - ret->name.str = p; - p += st->name.len; - memmove(p, st->uid.str, st->uid.len); - ret->uid.str = p; - p += st->uid.len; - memmove(p, st->gid.str, st->gid.len); - ret->gid.str = p; - p += st->gid.len; - memmove(p, st->muid.str, st->muid.len); - ret->muid.str = p; - p += st->muid.len; - - if (dotu) { - memmove(p, st->extension.str, st->extension.len); - ret->extension.str = p; - p += st->extension.len; - } - - return ret; -} - -static struct p9_fid *p9_fid_create(struct p9_client *clnt) -{ - int err; - struct p9_fid *fid; - - P9_DPRINTK(P9_DEBUG_9P, "clnt %p\n", clnt); - fid = kmalloc(sizeof(struct p9_fid), GFP_KERNEL); - if (!fid) - return ERR_PTR(-ENOMEM); - - fid->fid = p9_idpool_get(clnt->fidpool); - if (fid->fid < 0) { - err = -ENOSPC; + req = p9_client_rpc(clnt, P9_TWSTAT, "dwS", fid->fid, 0, wst); + if (IS_ERR(req)) { + err = PTR_ERR(req); goto error; } - memset(&fid->qid, 0, sizeof(struct p9_qid)); - fid->mode = -1; - fid->rdir_fpos = 0; - fid->rdir_pos = 0; - fid->rdir_fcall = NULL; - fid->uid = current->fsuid; - fid->clnt = clnt; - fid->aux = NULL; - - spin_lock(&clnt->lock); - list_add(&fid->flist, &clnt->fidlist); - spin_unlock(&clnt->lock); - - return fid; + P9_DPRINTK(P9_DEBUG_9P, "<<< RWSTAT fid %d\n", fid->fid); + p9_free_req(clnt, req); error: - kfree(fid); - return ERR_PTR(err); -} - -static void p9_fid_destroy(struct p9_fid *fid) -{ - struct p9_client *clnt; - - P9_DPRINTK(P9_DEBUG_9P, "fid %d\n", fid->fid); - clnt = fid->clnt; - p9_idpool_put(fid->fid, clnt->fidpool); - spin_lock(&clnt->lock); - list_del(&fid->flist); - spin_unlock(&clnt->lock); - kfree(fid->rdir_fcall); - kfree(fid); + return err; } +EXPORT_SYMBOL(p9_client_wstat); diff --git a/net/9p/conv.c b/net/9p/conv.c deleted file mode 100644 index 5ad3a3bd73b2..000000000000 --- a/net/9p/conv.c +++ /dev/null @@ -1,1054 +0,0 @@ -/* - * net/9p/conv.c - * - * 9P protocol conversion functions - * - * Copyright (C) 2004, 2005 by Latchesar Ionkov <lucho@ionkov.net> - * Copyright (C) 2004 by Eric Van Hensbergen <ericvh@gmail.com> - * Copyright (C) 2002 by Ron Minnich <rminnich@lanl.gov> - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 - * as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to: - * Free Software Foundation - * 51 Franklin Street, Fifth Floor - * Boston, MA 02111-1301 USA - * - */ - -#include <linux/module.h> -#include <linux/errno.h> -#include <linux/fs.h> -#include <linux/sched.h> -#include <linux/idr.h> -#include <linux/uaccess.h> -#include <net/9p/9p.h> - -/* - * Buffer to help with string parsing - */ -struct cbuf { - unsigned char *sp; - unsigned char *p; - unsigned char *ep; -}; - -static inline void buf_init(struct cbuf *buf, void *data, int datalen) -{ - buf->sp = buf->p = data; - buf->ep = data + datalen; -} - -static inline int buf_check_overflow(struct cbuf *buf) -{ - return buf->p > buf->ep; -} - -static int buf_check_size(struct cbuf *buf, int len) -{ - if (buf->p + len > buf->ep) { - if (buf->p < buf->ep) { - P9_EPRINTK(KERN_ERR, - "buffer overflow: want %d has %d\n", len, - (int)(buf->ep - buf->p)); - dump_stack(); - buf->p = buf->ep + 1; - } - - return 0; - } - - return 1; -} - -static void *buf_alloc(struct cbuf *buf, int len) -{ - void *ret = NULL; - - if (buf_check_size(buf, len)) { - ret = buf->p; - buf->p += len; - } - - return ret; -} - -static void buf_put_int8(struct cbuf *buf, u8 val) -{ - if (buf_check_size(buf, 1)) { - buf->p[0] = val; - buf->p++; - } -} - -static void buf_put_int16(struct cbuf *buf, u16 val) -{ - if (buf_check_size(buf, 2)) { - *(__le16 *) buf->p = cpu_to_le16(val); - buf->p += 2; - } -} - -static void buf_put_int32(struct cbuf *buf, u32 val) -{ - if (buf_check_size(buf, 4)) { - *(__le32 *)buf->p = cpu_to_le32(val); - buf->p += 4; - } -} - -static void buf_put_int64(struct cbuf *buf, u64 val) -{ - if (buf_check_size(buf, 8)) { - *(__le64 *)buf->p = cpu_to_le64(val); - buf->p += 8; - } -} - -static char *buf_put_stringn(struct cbuf *buf, const char *s, u16 slen) -{ - char *ret; - - ret = NULL; - if (buf_check_size(buf, slen + 2)) { - buf_put_int16(buf, slen); - ret = buf->p; - memcpy(buf->p, s, slen); - buf->p += slen; - } - - return ret; -} - -static u8 buf_get_int8(struct cbuf *buf) -{ - u8 ret = 0; - - if (buf_check_size(buf, 1)) { - ret = buf->p[0]; - buf->p++; - } - - return ret; -} - -static u16 buf_get_int16(struct cbuf *buf) -{ - u16 ret = 0; - - if (buf_check_size(buf, 2)) { - ret = le16_to_cpu(*(__le16 *)buf->p); - buf->p += 2; - } - - return ret; -} - -static u32 buf_get_int32(struct cbuf *buf) -{ - u32 ret = 0; - - if (buf_check_size(buf, 4)) { - ret = le32_to_cpu(*(__le32 *)buf->p); - buf->p += 4; - } - - return ret; -} - -static u64 buf_get_int64(struct cbuf *buf) -{ - u64 ret = 0; - - if (buf_check_size(buf, 8)) { - ret = le64_to_cpu(*(__le64 *)buf->p); - buf->p += 8; - } - - return ret; -} - -static void buf_get_str(struct cbuf *buf, struct p9_str *vstr) -{ - vstr->len = buf_get_int16(buf); - if (!buf_check_overflow(buf) && buf_check_size(buf, vstr->len)) { - vstr->str = buf->p; - buf->p += vstr->len; - } else { - vstr->len = 0; - vstr->str = NULL; - } -} - -static void buf_get_qid(struct cbuf *bufp, struct p9_qid *qid) -{ - qid->type = buf_get_int8(bufp); - qid->version = buf_get_int32(bufp); - qid->path = buf_get_int64(bufp); -} - -/** - * p9_size_wstat - calculate the size of a variable length stat struct - * @wstat: metadata (stat) structure - * @dotu: non-zero if 9P2000.u - * - */ - -static int p9_size_wstat(struct p9_wstat *wstat, int dotu) -{ - int size = 0; - - if (wstat == NULL) { - P9_EPRINTK(KERN_ERR, "p9_size_stat: got a NULL stat pointer\n"); - return 0; - } - - size = /* 2 + *//* size[2] */ - 2 + /* type[2] */ - 4 + /* dev[4] */ - 1 + /* qid.type[1] */ - 4 + /* qid.vers[4] */ - 8 + /* qid.path[8] */ - 4 + /* mode[4] */ - 4 + /* atime[4] */ - 4 + /* mtime[4] */ - 8 + /* length[8] */ - 8; /* minimum sum of string lengths */ - - if (wstat->name) - size += strlen(wstat->name); - if (wstat->uid) - size += strlen(wstat->uid); - if (wstat->gid) - size += strlen(wstat->gid); - if (wstat->muid) - size += strlen(wstat->muid); - - if (dotu) { - size += 4 + /* n_uid[4] */ - 4 + /* n_gid[4] */ - 4 + /* n_muid[4] */ - 2; /* string length of extension[4] */ - if (wstat->extension) - size += strlen(wstat->extension); - } - - return size; -} - -/** - * buf_get_stat - safely decode a recieved metadata (stat) structure - * @bufp: buffer to deserialize - * @stat: metadata (stat) structure - * @dotu: non-zero if 9P2000.u - * - */ - -static void -buf_get_stat(struct cbuf *bufp, struct p9_stat *stat, int dotu) -{ - stat->size = buf_get_int16(bufp); - stat->type = buf_get_int16(bufp); - stat->dev = buf_get_int32(bufp); - stat->qid.type = buf_get_int8(bufp); - stat->qid.version = buf_get_int32(bufp); - stat->qid.path = buf_get_int64(bufp); - stat->mode = buf_get_int32(bufp); - stat->atime = buf_get_int32(bufp); - stat->mtime = buf_get_int32(bufp); - stat->length = buf_get_int64(bufp); - buf_get_str(bufp, &stat->name); - buf_get_str(bufp, &stat->uid); - buf_get_str(bufp, &stat->gid); - buf_get_str(bufp, &stat->muid); - - if (dotu) { - buf_get_str(bufp, &stat->extension); - stat->n_uid = buf_get_int32(bufp); - stat->n_gid = buf_get_int32(bufp); - stat->n_muid = buf_get_int32(bufp); - } -} - -/** - * p9_deserialize_stat - decode a received metadata structure - * @buf: buffer to deserialize - * @buflen: length of received buffer - * @stat: metadata structure to decode into - * @dotu: non-zero if 9P2000.u - * - * Note: stat will point to the buf region. - */ - -int -p9_deserialize_stat(void *buf, u32 buflen, struct p9_stat *stat, - int dotu) -{ - struct cbuf buffer; - struct cbuf *bufp = &buffer; - unsigned char *p; - - buf_init(bufp, buf, buflen); - p = bufp->p; - buf_get_stat(bufp, stat, dotu); - - if (buf_check_overflow(bufp)) - return 0; - else - return bufp->p - p; -} -EXPORT_SYMBOL(p9_deserialize_stat); - -/** - * deserialize_fcall - unmarshal a response - * @buf: recieved buffer - * @buflen: length of received buffer - * @rcall: fcall structure to populate - * @rcalllen: length of fcall structure to populate - * @dotu: non-zero if 9P2000.u - * - */ - -int -p9_deserialize_fcall(void *buf, u32 buflen, struct p9_fcall *rcall, - int dotu) -{ - - struct cbuf buffer; - struct cbuf *bufp = &buffer; - int i = 0; - - buf_init(bufp, buf, buflen); - - rcall->size = buf_get_int32(bufp); - rcall->id = buf_get_int8(bufp); - rcall->tag = buf_get_int16(bufp); - - P9_DPRINTK(P9_DEBUG_CONV, "size %d id %d tag %d\n", rcall->size, - rcall->id, rcall->tag); - - switch (rcall->id) { - default: - P9_EPRINTK(KERN_ERR, "unknown message type: %d\n", rcall->id); - return -EPROTO; - case P9_RVERSION: - rcall->params.rversion.msize = buf_get_int32(bufp); - buf_get_str(bufp, &rcall->params.rversion.version); - break; - case P9_RFLUSH: - break; - case P9_RATTACH: - rcall->params.rattach.qid.type = buf_get_int8(bufp); - rcall->params.rattach.qid.version = buf_get_int32(bufp); - rcall->params.rattach.qid.path = buf_get_int64(bufp); - break; - case P9_RWALK: - rcall->params.rwalk.nwqid = buf_get_int16(bufp); - if (rcall->params.rwalk.nwqid > P9_MAXWELEM) { - P9_EPRINTK(KERN_ERR, - "Rwalk with more than %d qids: %d\n", - P9_MAXWELEM, rcall->params.rwalk.nwqid); - return -EPROTO; - } - - for (i = 0; i < rcall->params.rwalk.nwqid; i++) - buf_get_qid(bufp, &rcall->params.rwalk.wqids[i]); - break; - case P9_ROPEN: - buf_get_qid(bufp, &rcall->params.ropen.qid); - rcall->params.ropen.iounit = buf_get_int32(bufp); - break; - case P9_RCREATE: - buf_get_qid(bufp, &rcall->params.rcreate.qid); - rcall->params.rcreate.iounit = buf_get_int32(bufp); - break; - case P9_RREAD: - rcall->params.rread.count = buf_get_int32(bufp); - rcall->params.rread.data = bufp->p; - buf_check_size(bufp, rcall->params.rread.count); - break; - case P9_RWRITE: - rcall->params.rwrite.count = buf_get_int32(bufp); - break; - case P9_RCLUNK: - break; - case P9_RREMOVE: - break; - case P9_RSTAT: - buf_get_int16(bufp); - buf_get_stat(bufp, &rcall->params.rstat.stat, dotu); - break; - case P9_RWSTAT: - break; - case P9_RERROR: - buf_get_str(bufp, &rcall->params.rerror.error); - if (dotu) - rcall->params.rerror.errno = buf_get_int16(bufp); - break; - } - - if (buf_check_overflow(bufp)) { - P9_DPRINTK(P9_DEBUG_ERROR, "buffer overflow\n"); - return -EIO; - } - - return bufp->p - bufp->sp; -} -EXPORT_SYMBOL(p9_deserialize_fcall); - -static inline void p9_put_int8(struct cbuf *bufp, u8 val, u8 * p) -{ - *p = val; - buf_put_int8(bufp, val); -} - -static inline void p9_put_int16(struct cbuf *bufp, u16 val, u16 * p) -{ - *p = val; - buf_put_int16(bufp, val); -} - -static inline void p9_put_int32(struct cbuf *bufp, u32 val, u32 * p) -{ - *p = val; - buf_put_int32(bufp, val); -} - -static inline void p9_put_int64(struct cbuf *bufp, u64 val, u64 * p) -{ - *p = val; - buf_put_int64(bufp, val); -} - -static void -p9_put_str(struct cbuf *bufp, char *data, struct p9_str *str) -{ - int len; - char *s; - - if (data) - len = strlen(data); - else - len = 0; - - s = buf_put_stringn(bufp, data, len); - if (str) { - str->len = len; - str->str = s; - } -} - -static int -p9_put_data(struct cbuf *bufp, const char *data, int count, - unsigned char **pdata) -{ - *pdata = buf_alloc(bufp, count); - if (*pdata == NULL) - return -ENOMEM; - memmove(*pdata, data, count); - return 0; -} - -static int -p9_put_user_data(struct cbuf *bufp, const char __user *data, int count, - unsigned char **pdata) -{ - *pdata = buf_alloc(bufp, count); - if (*pdata == NULL) - return -ENOMEM; - return copy_from_user(*pdata, data, count); -} - -static void -p9_put_wstat(struct cbuf *bufp, struct p9_wstat *wstat, - struct p9_stat *stat, int statsz, int dotu) -{ - p9_put_int16(bufp, statsz, &stat->size); - p9_put_int16(bufp, wstat->type, &stat->type); - p9_put_int32(bufp, wstat->dev, &stat->dev); - p9_put_int8(bufp, wstat->qid.type, &stat->qid.type); - p9_put_int32(bufp, wstat->qid.version, &stat->qid.version); - p9_put_int64(bufp, wstat->qid.path, &stat->qid.path); - p9_put_int32(bufp, wstat->mode, &stat->mode); - p9_put_int32(bufp, wstat->atime, &stat->atime); - p9_put_int32(bufp, wstat->mtime, &stat->mtime); - p9_put_int64(bufp, wstat->length, &stat->length); - - p9_put_str(bufp, wstat->name, &stat->name); - p9_put_str(bufp, wstat->uid, &stat->uid); - p9_put_str(bufp, wstat->gid, &stat->gid); - p9_put_str(bufp, wstat->muid, &stat->muid); - - if (dotu) { - p9_put_str(bufp, wstat->extension, &stat->extension); - p9_put_int32(bufp, wstat->n_uid, &stat->n_uid); - p9_put_int32(bufp, wstat->n_gid, &stat->n_gid); - p9_put_int32(bufp, wstat->n_muid, &stat->n_muid); - } -} - -static struct p9_fcall * -p9_create_common(struct cbuf *bufp, u32 size, u8 id) -{ - struct p9_fcall *fc; - - size += 4 + 1 + 2; /* size[4] id[1] tag[2] */ - fc = kmalloc(sizeof(struct p9_fcall) + size, GFP_KERNEL); - if (!fc) - return ERR_PTR(-ENOMEM); - - fc->sdata = (char *)fc + sizeof(*fc); - - buf_init(bufp, (char *)fc->sdata, size); - p9_put_int32(bufp, size, &fc->size); - p9_put_int8(bufp, id, &fc->id); - p9_put_int16(bufp, P9_NOTAG, &fc->tag); - - return fc; -} - -/** - * p9_set_tag - set the tag field of an &p9_fcall structure - * @fc: fcall structure to set tag within - * @tag: tag id to set - */ - -void p9_set_tag(struct p9_fcall *fc, u16 tag) -{ - fc->tag = tag; - *(__le16 *) (fc->sdata + 5) = cpu_to_le16(tag); -} -EXPORT_SYMBOL(p9_set_tag); - -/** - * p9_create_tversion - allocates and creates a T_VERSION request - * @msize: requested maximum data size - * @version: version string to negotiate - * - */ -struct p9_fcall *p9_create_tversion(u32 msize, char *version) -{ - int size; - struct p9_fcall *fc; - struct cbuf buffer; - struct cbuf *bufp = &buffer; - - size = 4 + 2 + strlen(version); /* msize[4] version[s] */ - fc = p9_create_common(bufp, size, P9_TVERSION); - if (IS_ERR(fc)) - goto error; - - p9_put_int32(bufp, msize, &fc->params.tversion.msize); - p9_put_str(bufp, version, &fc->params.tversion.version); - - if (buf_check_overflow(bufp)) { - kfree(fc); - fc = ERR_PTR(-ENOMEM); - } -error: - return fc; -} -EXPORT_SYMBOL(p9_create_tversion); - -/** - * p9_create_tauth - allocates and creates a T_AUTH request - * @afid: handle to use for authentication protocol - * @uname: user name attempting to authenticate - * @aname: mount specifier for remote server - * @n_uname: numeric id for user attempting to authneticate - * @dotu: 9P2000.u extension flag - * - */ - -struct p9_fcall *p9_create_tauth(u32 afid, char *uname, char *aname, - u32 n_uname, int dotu) -{ - int size; - struct p9_fcall *fc; - struct cbuf buffer; - struct cbuf *bufp = &buffer; - - /* afid[4] uname[s] aname[s] */ - size = 4 + 2 + 2; - if (uname) - size += strlen(uname); - - if (aname) - size += strlen(aname); - - if (dotu) - size += 4; /* n_uname */ - - fc = p9_create_common(bufp, size, P9_TAUTH); - if (IS_ERR(fc)) - goto error; - - p9_put_int32(bufp, afid, &fc->params.tauth.afid); - p9_put_str(bufp, uname, &fc->params.tauth.uname); - p9_put_str(bufp, aname, &fc->params.tauth.aname); - if (dotu) - p9_put_int32(bufp, n_uname, &fc->params.tauth.n_uname); - - if (buf_check_overflow(bufp)) { - kfree(fc); - fc = ERR_PTR(-ENOMEM); - } -error: - return fc; -} -EXPORT_SYMBOL(p9_create_tauth); - -/** - * p9_create_tattach - allocates and creates a T_ATTACH request - * @fid: handle to use for the new mount point - * @afid: handle to use for authentication protocol - * @uname: user name attempting to attach - * @aname: mount specifier for remote server - * @n_uname: numeric id for user attempting to attach - * @n_uname: numeric id for user attempting to attach - * @dotu: 9P2000.u extension flag - * - */ - -struct p9_fcall * -p9_create_tattach(u32 fid, u32 afid, char *uname, char *aname, - u32 n_uname, int dotu) -{ - int size; - struct p9_fcall *fc; - struct cbuf buffer; - struct cbuf *bufp = &buffer; - - /* fid[4] afid[4] uname[s] aname[s] */ - size = 4 + 4 + 2 + 2; - if (uname) - size += strlen(uname); - - if (aname) - size += strlen(aname); - - if (dotu) - size += 4; /* n_uname */ - - fc = p9_create_common(bufp, size, P9_TATTACH); - if (IS_ERR(fc)) - goto error; - - p9_put_int32(bufp, fid, &fc->params.tattach.fid); - p9_put_int32(bufp, afid, &fc->params.tattach.afid); - p9_put_str(bufp, uname, &fc->params.tattach.uname); - p9_put_str(bufp, aname, &fc->params.tattach.aname); - if (dotu) - p9_put_int32(bufp, n_uname, &fc->params.tattach.n_uname); - -error: - return fc; -} -EXPORT_SYMBOL(p9_create_tattach); - -/** - * p9_create_tflush - allocates and creates a T_FLUSH request - * @oldtag: tag id for the transaction we are attempting to cancel - * - */ - -struct p9_fcall *p9_create_tflush(u16 oldtag) -{ - int size; - struct p9_fcall *fc; - struct cbuf buffer; - struct cbuf *bufp = &buffer; - - size = 2; /* oldtag[2] */ - fc = p9_create_common(bufp, size, P9_TFLUSH); - if (IS_ERR(fc)) - goto error; - - p9_put_int16(bufp, oldtag, &fc->params.tflush.oldtag); - - if (buf_check_overflow(bufp)) { - kfree(fc); - fc = ERR_PTR(-ENOMEM); - } -error: - return fc; -} -EXPORT_SYMBOL(p9_create_tflush); - -/** - * p9_create_twalk - allocates and creates a T_FLUSH request - * @fid: handle we are traversing from - * @newfid: a new handle for this transaction - * @nwname: number of path elements to traverse - * @wnames: array of path elements - * - */ - -struct p9_fcall *p9_create_twalk(u32 fid, u32 newfid, u16 nwname, - char **wnames) -{ - int i, size; - struct p9_fcall *fc; - struct cbuf buffer; - struct cbuf *bufp = &buffer; - - if (nwname > P9_MAXWELEM) { - P9_DPRINTK(P9_DEBUG_ERROR, "nwname > %d\n", P9_MAXWELEM); - return NULL; - } - - size = 4 + 4 + 2; /* fid[4] newfid[4] nwname[2] ... */ - for (i = 0; i < nwname; i++) { - size += 2 + strlen(wnames[i]); /* wname[s] */ - } - - fc = p9_create_common(bufp, size, P9_TWALK); - if (IS_ERR(fc)) - goto error; - - p9_put_int32(bufp, fid, &fc->params.twalk.fid); - p9_put_int32(bufp, newfid, &fc->params.twalk.newfid); - p9_put_int16(bufp, nwname, &fc->params.twalk.nwname); - for (i = 0; i < nwname; i++) { - p9_put_str(bufp, wnames[i], &fc->params.twalk.wnames[i]); - } - - if (buf_check_overflow(bufp)) { - kfree(fc); - fc = ERR_PTR(-ENOMEM); - } -error: - return fc; -} -EXPORT_SYMBOL(p9_create_twalk); - -/** - * p9_create_topen - allocates and creates a T_OPEN request - * @fid: handle we are trying to open - * @mode: what mode we are trying to open the file in - * - */ - -struct p9_fcall *p9_create_topen(u32 fid, u8 mode) -{ - int size; - struct p9_fcall *fc; - struct cbuf buffer; - struct cbuf *bufp = &buffer; - - size = 4 + 1; /* fid[4] mode[1] */ - fc = p9_create_common(bufp, size, P9_TOPEN); - if (IS_ERR(fc)) - goto error; - - p9_put_int32(bufp, fid, &fc->params.topen.fid); - p9_put_int8(bufp, mode, &fc->params.topen.mode); - - if (buf_check_overflow(bufp)) { - kfree(fc); - fc = ERR_PTR(-ENOMEM); - } -error: - return fc; -} -EXPORT_SYMBOL(p9_create_topen); - -/** - * p9_create_tcreate - allocates and creates a T_CREATE request - * @fid: handle of directory we are trying to create in - * @name: name of the file we are trying to create - * @perm: permissions for the file we are trying to create - * @mode: what mode we are trying to open the file in - * @extension: 9p2000.u extension string (for special files) - * @dotu: 9p2000.u enabled flag - * - * Note: Plan 9 create semantics include opening the resulting file - * which is why mode is included. - */ - -struct p9_fcall *p9_create_tcreate(u32 fid, char *name, u32 perm, u8 mode, - char *extension, int dotu) -{ - int size; - struct p9_fcall *fc; - struct cbuf buffer; - struct cbuf *bufp = &buffer; - - /* fid[4] name[s] perm[4] mode[1] */ - size = 4 + 2 + strlen(name) + 4 + 1; - if (dotu) { - size += 2 + /* extension[s] */ - (extension == NULL ? 0 : strlen(extension)); - } - - fc = p9_create_common(bufp, size, P9_TCREATE); - if (IS_ERR(fc)) - goto error; - - p9_put_int32(bufp, fid, &fc->params.tcreate.fid); - p9_put_str(bufp, name, &fc->params.tcreate.name); - p9_put_int32(bufp, perm, &fc->params.tcreate.perm); - p9_put_int8(bufp, mode, &fc->params.tcreate.mode); - if (dotu) - p9_put_str(bufp, extension, &fc->params.tcreate.extension); - - if (buf_check_overflow(bufp)) { - kfree(fc); - fc = ERR_PTR(-ENOMEM); - } -error: - return fc; -} -EXPORT_SYMBOL(p9_create_tcreate); - -/** - * p9_create_tread - allocates and creates a T_READ request - * @fid: handle of the file we are trying to read - * @offset: offset to start reading from - * @count: how many bytes to read - */ - -struct p9_fcall *p9_create_tread(u32 fid, u64 offset, u32 count) -{ - int size; - struct p9_fcall *fc; - struct cbuf buffer; - struct cbuf *bufp = &buffer; - - size = 4 + 8 + 4; /* fid[4] offset[8] count[4] */ - fc = p9_create_common(bufp, size, P9_TREAD); - if (IS_ERR(fc)) - goto error; - - p9_put_int32(bufp, fid, &fc->params.tread.fid); - p9_put_int64(bufp, offset, &fc->params.tread.offset); - p9_put_int32(bufp, count, &fc->params.tread.count); - - if (buf_check_overflow(bufp)) { - kfree(fc); - fc = ERR_PTR(-ENOMEM); - } -error: - return fc; -} -EXPORT_SYMBOL(p9_create_tread); - -/** - * p9_create_twrite - allocates and creates a T_WRITE request from the kernel - * @fid: handle of the file we are trying to write - * @offset: offset to start writing at - * @count: how many bytes to write - * @data: data to write - * - * This function will create a requst with data buffers from the kernel - * such as the page cache. - */ - -struct p9_fcall *p9_create_twrite(u32 fid, u64 offset, u32 count, - const char *data) -{ - int size, err; - struct p9_fcall *fc; - struct cbuf buffer; - struct cbuf *bufp = &buffer; - - /* fid[4] offset[8] count[4] data[count] */ - size = 4 + 8 + 4 + count; - fc = p9_create_common(bufp, size, P9_TWRITE); - if (IS_ERR(fc)) - goto error; - - p9_put_int32(bufp, fid, &fc->params.twrite.fid); - p9_put_int64(bufp, offset, &fc->params.twrite.offset); - p9_put_int32(bufp, count, &fc->params.twrite.count); - err = p9_put_data(bufp, data, count, &fc->params.twrite.data); - if (err) { - kfree(fc); - fc = ERR_PTR(err); - goto error; - } - - if (buf_check_overflow(bufp)) { - kfree(fc); - fc = ERR_PTR(-ENOMEM); - } -error: - return fc; -} -EXPORT_SYMBOL(p9_create_twrite); - -/** - * p9_create_twrite_u - allocates and creates a T_WRITE request from userspace - * @fid: handle of the file we are trying to write - * @offset: offset to start writing at - * @count: how many bytes to write - * @data: data to write - * - * This function will create a request with data buffers from userspace - */ - -struct p9_fcall *p9_create_twrite_u(u32 fid, u64 offset, u32 count, - const char __user *data) -{ - int size, err; - struct p9_fcall *fc; - struct cbuf buffer; - struct cbuf *bufp = &buffer; - - /* fid[4] offset[8] count[4] data[count] */ - size = 4 + 8 + 4 + count; - fc = p9_create_common(bufp, size, P9_TWRITE); - if (IS_ERR(fc)) - goto error; - - p9_put_int32(bufp, fid, &fc->params.twrite.fid); - p9_put_int64(bufp, offset, &fc->params.twrite.offset); - p9_put_int32(bufp, count, &fc->params.twrite.count); - err = p9_put_user_data(bufp, data, count, &fc->params.twrite.data); - if (err) { - kfree(fc); - fc = ERR_PTR(err); - goto error; - } - - if (buf_check_overflow(bufp)) { - kfree(fc); - fc = ERR_PTR(-ENOMEM); - } -error: - return fc; -} -EXPORT_SYMBOL(p9_create_twrite_u); - -/** - * p9_create_tclunk - allocate a request to forget about a file handle - * @fid: handle of the file we closing or forgetting about - * - * clunk is used both to close open files and to discard transient handles - * which may be created during meta-data operations and hierarchy traversal. - */ - -struct p9_fcall *p9_create_tclunk(u32 fid) -{ - int size; - struct p9_fcall *fc; - struct cbuf buffer; - struct cbuf *bufp = &buffer; - - size = 4; /* fid[4] */ - fc = p9_create_common(bufp, size, P9_TCLUNK); - if (IS_ERR(fc)) - goto error; - - p9_put_int32(bufp, fid, &fc->params.tclunk.fid); - - if (buf_check_overflow(bufp)) { - kfree(fc); - fc = ERR_PTR(-ENOMEM); - } -error: - return fc; -} -EXPORT_SYMBOL(p9_create_tclunk); - -/** - * p9_create_tremove - allocate and create a request to remove a file - * @fid: handle of the file or directory we are removing - * - */ - -struct p9_fcall *p9_create_tremove(u32 fid) -{ - int size; - struct p9_fcall *fc; - struct cbuf buffer; - struct cbuf *bufp = &buffer; - - size = 4; /* fid[4] */ - fc = p9_create_common(bufp, size, P9_TREMOVE); - if (IS_ERR(fc)) - goto error; - - p9_put_int32(bufp, fid, &fc->params.tremove.fid); - - if (buf_check_overflow(bufp)) { - kfree(fc); - fc = ERR_PTR(-ENOMEM); - } -error: - return fc; -} -EXPORT_SYMBOL(p9_create_tremove); - -/** - * p9_create_tstat - allocate and populate a request for attributes - * @fid: handle of the file or directory we are trying to get the attributes of - * - */ - -struct p9_fcall *p9_create_tstat(u32 fid) -{ - int size; - struct p9_fcall *fc; - struct cbuf buffer; - struct cbuf *bufp = &buffer; - - size = 4; /* fid[4] */ - fc = p9_create_common(bufp, size, P9_TSTAT); - if (IS_ERR(fc)) - goto error; - - p9_put_int32(bufp, fid, &fc->params.tstat.fid); - - if (buf_check_overflow(bufp)) { - kfree(fc); - fc = ERR_PTR(-ENOMEM); - } -error: - return fc; -} -EXPORT_SYMBOL(p9_create_tstat); - -/** - * p9_create_tstat - allocate and populate a request to change attributes - * @fid: handle of the file or directory we are trying to change - * @wstat: &p9_stat structure with attributes we wish to set - * @dotu: 9p2000.u enabled flag - * - */ - -struct p9_fcall *p9_create_twstat(u32 fid, struct p9_wstat *wstat, - int dotu) -{ - int size, statsz; - struct p9_fcall *fc; - struct cbuf buffer; - struct cbuf *bufp = &buffer; - - statsz = p9_size_wstat(wstat, dotu); - size = 4 + 2 + 2 + statsz; /* fid[4] stat[n] */ - fc = p9_create_common(bufp, size, P9_TWSTAT); - if (IS_ERR(fc)) - goto error; - - p9_put_int32(bufp, fid, &fc->params.twstat.fid); - buf_put_int16(bufp, statsz + 2); - p9_put_wstat(bufp, wstat, &fc->params.twstat.stat, statsz, dotu); - - if (buf_check_overflow(bufp)) { - kfree(fc); - fc = ERR_PTR(-ENOMEM); - } -error: - return fc; -} -EXPORT_SYMBOL(p9_create_twstat); - diff --git a/net/9p/fcprint.c b/net/9p/fcprint.c deleted file mode 100644 index 53dd8e28dd8a..000000000000 --- a/net/9p/fcprint.c +++ /dev/null @@ -1,366 +0,0 @@ -/* - * net/9p/fcprint.c - * - * Print 9P call. - * - * Copyright (C) 2005 by Latchesar Ionkov <lucho@ionkov.net> - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 - * as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to: - * Free Software Foundation - * 51 Franklin Street, Fifth Floor - * Boston, MA 02111-1301 USA - * - */ -#include <linux/module.h> -#include <linux/errno.h> -#include <linux/fs.h> -#include <linux/idr.h> -#include <net/9p/9p.h> - -#ifdef CONFIG_NET_9P_DEBUG - -static int -p9_printqid(char *buf, int buflen, struct p9_qid *q) -{ - int n; - char b[10]; - - n = 0; - if (q->type & P9_QTDIR) - b[n++] = 'd'; - if (q->type & P9_QTAPPEND) - b[n++] = 'a'; - if (q->type & P9_QTAUTH) - b[n++] = 'A'; - if (q->type & P9_QTEXCL) - b[n++] = 'l'; - if (q->type & P9_QTTMP) - b[n++] = 't'; - if (q->type & P9_QTSYMLINK) - b[n++] = 'L'; - b[n] = '\0'; - - return scnprintf(buf, buflen, "(%.16llx %x %s)", - (long long int) q->path, q->version, b); -} - -static int -p9_printperm(char *buf, int buflen, int perm) -{ - int n; - char b[15]; - - n = 0; - if (perm & P9_DMDIR) - b[n++] = 'd'; - if (perm & P9_DMAPPEND) - b[n++] = 'a'; - if (perm & P9_DMAUTH) - b[n++] = 'A'; - if (perm & P9_DMEXCL) - b[n++] = 'l'; - if (perm & P9_DMTMP) - b[n++] = 't'; - if (perm & P9_DMDEVICE) - b[n++] = 'D'; - if (perm & P9_DMSOCKET) - b[n++] = 'S'; - if (perm & P9_DMNAMEDPIPE) - b[n++] = 'P'; - if (perm & P9_DMSYMLINK) - b[n++] = 'L'; - b[n] = '\0'; - - return scnprintf(buf, buflen, "%s%03o", b, perm&077); -} - -static int -p9_printstat(char *buf, int buflen, struct p9_stat *st, int extended) -{ - int n; - - n = scnprintf(buf, buflen, "'%.*s' '%.*s'", st->name.len, - st->name.str, st->uid.len, st->uid.str); - if (extended) - n += scnprintf(buf+n, buflen-n, "(%d)", st->n_uid); - - n += scnprintf(buf+n, buflen-n, " '%.*s'", st->gid.len, st->gid.str); - if (extended) - n += scnprintf(buf+n, buflen-n, "(%d)", st->n_gid); - - n += scnprintf(buf+n, buflen-n, " '%.*s'", st->muid.len, st->muid.str); - if (extended) - n += scnprintf(buf+n, buflen-n, "(%d)", st->n_muid); - - n += scnprintf(buf+n, buflen-n, " q "); - n += p9_printqid(buf+n, buflen-n, &st->qid); - n += scnprintf(buf+n, buflen-n, " m "); - n += p9_printperm(buf+n, buflen-n, st->mode); - n += scnprintf(buf+n, buflen-n, " at %d mt %d l %lld", - st->atime, st->mtime, (long long int) st->length); - - if (extended) - n += scnprintf(buf+n, buflen-n, " ext '%.*s'", - st->extension.len, st->extension.str); - - return n; -} - -static int -p9_dumpdata(char *buf, int buflen, u8 *data, int datalen) -{ - int i, n; - - i = n = 0; - 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"); - - return n; -} - -static int -p9_printdata(char *buf, int buflen, u8 *data, int datalen) -{ - return p9_dumpdata(buf, buflen, data, datalen < 16?datalen:16); -} - -/** - * p9_printfcall - decode and print a protocol structure into a buffer - * @buf: buffer to deposit decoded structure into - * @buflen: available space in buffer - * @fc: protocol rpc structure of type &p9_fcall - * @extended: whether or not session is operating with extended protocol - */ - -int -p9_printfcall(char *buf, int buflen, struct p9_fcall *fc, int extended) -{ - int i, ret, type, tag; - - if (!fc) - return scnprintf(buf, buflen, "<NULL>"); - - type = fc->id; - tag = fc->tag; - - ret = 0; - switch (type) { - case P9_TVERSION: - ret += scnprintf(buf+ret, buflen-ret, - "Tversion tag %u msize %u version '%.*s'", tag, - fc->params.tversion.msize, - fc->params.tversion.version.len, - fc->params.tversion.version.str); - break; - - case P9_RVERSION: - ret += scnprintf(buf+ret, buflen-ret, - "Rversion tag %u msize %u version '%.*s'", tag, - fc->params.rversion.msize, - fc->params.rversion.version.len, - fc->params.rversion.version.str); - break; - - case P9_TAUTH: - ret += scnprintf(buf+ret, buflen-ret, - "Tauth tag %u afid %d uname '%.*s' aname '%.*s'", tag, - fc->params.tauth.afid, fc->params.tauth.uname.len, - fc->params.tauth.uname.str, fc->params.tauth.aname.len, - fc->params.tauth.aname.str); - break; - - case P9_RAUTH: - ret += scnprintf(buf+ret, buflen-ret, "Rauth tag %u qid ", tag); - p9_printqid(buf+ret, buflen-ret, &fc->params.rauth.qid); - break; - - case P9_TATTACH: - ret += scnprintf(buf+ret, buflen-ret, - "Tattach tag %u fid %d afid %d uname '%.*s' aname '%.*s'", tag, - fc->params.tattach.fid, fc->params.tattach.afid, - fc->params.tattach.uname.len, fc->params.tattach.uname.str, - fc->params.tattach.aname.len, fc->params.tattach.aname.str); - break; - - case P9_RATTACH: - ret += scnprintf(buf+ret, buflen-ret, "Rattach tag %u qid ", - tag); - p9_printqid(buf+ret, buflen-ret, &fc->params.rattach.qid); - break; - - case P9_RERROR: - ret += scnprintf(buf+ret, buflen-ret, - "Rerror tag %u ename '%.*s'", tag, - fc->params.rerror.error.len, - fc->params.rerror.error.str); - if (extended) - ret += scnprintf(buf+ret, buflen-ret, " ecode %d\n", - fc->params.rerror.errno); - break; - - case P9_TFLUSH: - ret += scnprintf(buf+ret, buflen-ret, "Tflush tag %u oldtag %u", - tag, fc->params.tflush.oldtag); - break; - - case P9_RFLUSH: - ret += scnprintf(buf+ret, buflen-ret, "Rflush tag %u", tag); - break; - - case P9_TWALK: - ret += scnprintf(buf+ret, buflen-ret, - "Twalk tag %u fid %d newfid %d nwname %d", tag, - fc->params.twalk.fid, fc->params.twalk.newfid, - fc->params.twalk.nwname); - for (i = 0; i < fc->params.twalk.nwname; i++) - ret += scnprintf(buf+ret, buflen-ret, " '%.*s'", - fc->params.twalk.wnames[i].len, - fc->params.twalk.wnames[i].str); - break; - - case P9_RWALK: - ret += scnprintf(buf+ret, buflen-ret, "Rwalk tag %u nwqid %d", - tag, fc->params.rwalk.nwqid); - for (i = 0; i < fc->params.rwalk.nwqid; i++) - ret += p9_printqid(buf+ret, buflen-ret, - &fc->params.rwalk.wqids[i]); - break; - - case P9_TOPEN: - ret += scnprintf(buf+ret, buflen-ret, - "Topen tag %u fid %d mode %d", tag, - fc->params.topen.fid, fc->params.topen.mode); - break; - - case P9_ROPEN: - ret += scnprintf(buf+ret, buflen-ret, "Ropen tag %u", tag); - ret += p9_printqid(buf+ret, buflen-ret, &fc->params.ropen.qid); - ret += scnprintf(buf+ret, buflen-ret, " iounit %d", - fc->params.ropen.iounit); - break; - - case P9_TCREATE: - ret += scnprintf(buf+ret, buflen-ret, - "Tcreate tag %u fid %d name '%.*s' perm ", tag, - fc->params.tcreate.fid, fc->params.tcreate.name.len, - fc->params.tcreate.name.str); - - ret += p9_printperm(buf+ret, buflen-ret, - fc->params.tcreate.perm); - ret += scnprintf(buf+ret, buflen-ret, " mode %d", - fc->params.tcreate.mode); - break; - - case P9_RCREATE: - ret += scnprintf(buf+ret, buflen-ret, "Rcreate tag %u", tag); - ret += p9_printqid(buf+ret, buflen-ret, - &fc->params.rcreate.qid); - ret += scnprintf(buf+ret, buflen-ret, " iounit %d", - fc->params.rcreate.iounit); - break; - - case P9_TREAD: - ret += scnprintf(buf+ret, buflen-ret, - "Tread tag %u fid %d offset %lld count %u", tag, - fc->params.tread.fid, - (long long int) fc->params.tread.offset, - fc->params.tread.count); - break; - - case P9_RREAD: - ret += scnprintf(buf+ret, buflen-ret, - "Rread tag %u count %u data ", tag, - fc->params.rread.count); - ret += p9_printdata(buf+ret, buflen-ret, fc->params.rread.data, - fc->params.rread.count); - break; - - case P9_TWRITE: - ret += scnprintf(buf+ret, buflen-ret, - "Twrite tag %u fid %d offset %lld count %u data ", - tag, fc->params.twrite.fid, - (long long int) fc->params.twrite.offset, - fc->params.twrite.count); - ret += p9_printdata(buf+ret, buflen-ret, fc->params.twrite.data, - fc->params.twrite.count); - break; - - case P9_RWRITE: - ret += scnprintf(buf+ret, buflen-ret, "Rwrite tag %u count %u", - tag, fc->params.rwrite.count); - break; - - case P9_TCLUNK: - ret += scnprintf(buf+ret, buflen-ret, "Tclunk tag %u fid %d", - tag, fc->params.tclunk.fid); - break; - - case P9_RCLUNK: - ret += scnprintf(buf+ret, buflen-ret, "Rclunk tag %u", tag); - break; - - case P9_TREMOVE: - ret += scnprintf(buf+ret, buflen-ret, "Tremove tag %u fid %d", - tag, fc->params.tremove.fid); - break; - - case P9_RREMOVE: - ret += scnprintf(buf+ret, buflen-ret, "Rremove tag %u", tag); - break; - - case P9_TSTAT: - ret += scnprintf(buf+ret, buflen-ret, "Tstat tag %u fid %d", - tag, fc->params.tstat.fid); - break; - - case P9_RSTAT: - ret += scnprintf(buf+ret, buflen-ret, "Rstat tag %u ", tag); - ret += p9_printstat(buf+ret, buflen-ret, &fc->params.rstat.stat, - extended); - break; - - case P9_TWSTAT: - ret += scnprintf(buf+ret, buflen-ret, "Twstat tag %u fid %d ", - tag, fc->params.twstat.fid); - ret += p9_printstat(buf+ret, buflen-ret, - &fc->params.twstat.stat, extended); - break; - - case P9_RWSTAT: - ret += scnprintf(buf+ret, buflen-ret, "Rwstat tag %u", tag); - break; - - default: - ret += scnprintf(buf+ret, buflen-ret, "unknown type %d", type); - break; - } - - return ret; -} -#else -int -p9_printfcall(char *buf, int buflen, struct p9_fcall *fc, int extended) -{ - return 0; -} -#endif /* CONFIG_NET_9P_DEBUG */ -EXPORT_SYMBOL(p9_printfcall); - diff --git a/net/9p/mod.c b/net/9p/mod.c index 1084feb24cb0..cf8a4128cd5c 100644 --- a/net/9p/mod.c +++ b/net/9p/mod.c @@ -29,6 +29,7 @@ #include <net/9p/9p.h> #include <linux/fs.h> #include <linux/parser.h> +#include <net/9p/client.h> #include <net/9p/transport.h> #include <linux/list.h> #include <linux/spinlock.h> diff --git a/net/9p/protocol.c b/net/9p/protocol.c new file mode 100644 index 000000000000..29be52439086 --- /dev/null +++ b/net/9p/protocol.c @@ -0,0 +1,558 @@ +/* + * net/9p/protocol.c + * + * 9P Protocol Support Code + * + * Copyright (C) 2008 by Eric Van Hensbergen <ericvh@gmail.com> + * + * Base on code from Anthony Liguori <aliguori@us.ibm.com> + * Copyright (C) 2008 by IBM, Corp. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to: + * Free Software Foundation + * 51 Franklin Street, Fifth Floor + * Boston, MA 02111-1301 USA + * + */ + +#include <linux/module.h> +#include <linux/errno.h> +#include <linux/uaccess.h> +#include <linux/sched.h> +#include <net/9p/9p.h> +#include <net/9p/client.h> +#include "protocol.h" + +#ifndef MIN +#define MIN(a, b) (((a) < (b)) ? (a) : (b)) +#endif + +#ifndef MAX +#define MAX(a, b) (((a) > (b)) ? (a) : (b)) +#endif + +#ifndef offset_of +#define offset_of(type, memb) \ + ((unsigned long)(&((type *)0)->memb)) +#endif +#ifndef container_of +#define container_of(obj, type, memb) \ + ((type *)(((char *)obj) - offset_of(type, memb))) +#endif + +static int +p9pdu_writef(struct p9_fcall *pdu, int optional, const char *fmt, ...); + +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); +} +EXPORT_SYMBOL(p9pdu_dump); + +void p9stat_free(struct p9_wstat *stbuf) +{ + kfree(stbuf->name); + kfree(stbuf->uid); + kfree(stbuf->gid); + kfree(stbuf->muid); + kfree(stbuf->extension); +} +EXPORT_SYMBOL(p9stat_free); + +static size_t pdu_read(struct p9_fcall *pdu, void *data, size_t size) +{ + size_t len = MIN(pdu->size - pdu->offset, size); + memcpy(data, &pdu->sdata[pdu->offset], len); + pdu->offset += len; + return size - len; +} + +static size_t pdu_write(struct p9_fcall *pdu, const void *data, size_t size) +{ + size_t len = MIN(pdu->capacity - pdu->size, size); + memcpy(&pdu->sdata[pdu->size], data, len); + pdu->size += len; + return size - len; +} + +static size_t +pdu_write_u(struct p9_fcall *pdu, const char __user *udata, size_t size) +{ + size_t len = MIN(pdu->capacity - pdu->size, size); + int err = copy_from_user(&pdu->sdata[pdu->size], udata, len); + if (err) + printk(KERN_WARNING "pdu_write_u returning: %d\n", err); + + pdu->size += len; + return size - len; +} + +/* + b - int8_t + w - int16_t + d - int32_t + q - int64_t + s - string + S - stat + Q - qid + D - data blob (int32_t size followed by void *, results are not freed) + T - array of strings (int16_t count, followed by strings) + R - array of qids (int16_t count, followed by qids) + ? - if optional = 1, continue parsing +*/ + +static int +p9pdu_vreadf(struct p9_fcall *pdu, int optional, const char *fmt, va_list ap) +{ + const char *ptr; + int errcode = 0; + + for (ptr = fmt; *ptr; ptr++) { + switch (*ptr) { + case 'b':{ + int8_t *val = va_arg(ap, int8_t *); + if (pdu_read(pdu, val, sizeof(*val))) { + errcode = -EFAULT; + break; + } + } + break; + case 'w':{ + int16_t *val = va_arg(ap, int16_t *); + if (pdu_read(pdu, val, sizeof(*val))) { + errcode = -EFAULT; + break; + } + *val = cpu_to_le16(*val); + } + break; + case 'd':{ + int32_t *val = va_arg(ap, int32_t *); + if (pdu_read(pdu, val, sizeof(*val))) { + errcode = -EFAULT; + break; + } + *val = cpu_to_le32(*val); + } + break; + case 'q':{ + int64_t *val = va_arg(ap, int64_t *); + if (pdu_read(pdu, val, sizeof(*val))) { + errcode = -EFAULT; + break; + } + *val = cpu_to_le64(*val); + } + break; + case 's':{ + char **ptr = va_arg(ap, char **); + int16_t len; + int size; + + errcode = p9pdu_readf(pdu, optional, "w", &len); + if (errcode) + break; + + size = MAX(len, 0); + + *ptr = kmalloc(size + 1, GFP_KERNEL); + if (*ptr == NULL) { + errcode = -EFAULT; + break; + } + if (pdu_read(pdu, *ptr, size)) { + errcode = -EFAULT; + kfree(*ptr); + *ptr = NULL; + } else + (*ptr)[size] = 0; + } + break; + case 'Q':{ + struct p9_qid *qid = + va_arg(ap, struct p9_qid *); + + errcode = p9pdu_readf(pdu, optional, "bdq", + &qid->type, &qid->version, + &qid->path); + } + break; + case 'S':{ + struct p9_wstat *stbuf = + va_arg(ap, struct p9_wstat *); + + memset(stbuf, 0, sizeof(struct p9_wstat)); + stbuf->n_uid = stbuf->n_gid = stbuf->n_muid = + -1; + errcode = + p9pdu_readf(pdu, optional, + "wwdQdddqssss?sddd", + &stbuf->size, &stbuf->type, + &stbuf->dev, &stbuf->qid, + &stbuf->mode, &stbuf->atime, + &stbuf->mtime, &stbuf->length, + &stbuf->name, &stbuf->uid, + &stbuf->gid, &stbuf->muid, + &stbuf->extension, + &stbuf->n_uid, &stbuf->n_gid, + &stbuf->n_muid); + if (errcode) + p9stat_free(stbuf); + } + break; + case 'D':{ + int32_t *count = va_arg(ap, int32_t *); + void **data = va_arg(ap, void **); + + errcode = + p9pdu_readf(pdu, optional, "d", count); + if (!errcode) { + *count = + MIN(*count, + pdu->size - pdu->offset); + *data = &pdu->sdata[pdu->offset]; + } + } + break; + case 'T':{ + int16_t *nwname = va_arg(ap, int16_t *); + char ***wnames = va_arg(ap, char ***); + + errcode = + p9pdu_readf(pdu, optional, "w", nwname); + if (!errcode) { + *wnames = + kmalloc(sizeof(char *) * *nwname, + GFP_KERNEL); + if (!*wnames) + errcode = -ENOMEM; + } + + if (!errcode) { + int i; + + for (i = 0; i < *nwname; i++) { + errcode = + p9pdu_readf(pdu, optional, + "s", + &(*wnames)[i]); + if (errcode) + break; + } + } + + if (errcode) { + if (*wnames) { + int i; + + for (i = 0; i < *nwname; i++) + kfree((*wnames)[i]); + } + kfree(*wnames); + *wnames = NULL; + } + } + break; + case 'R':{ + int16_t *nwqid = va_arg(ap, int16_t *); + struct p9_qid **wqids = + va_arg(ap, struct p9_qid **); + + *wqids = NULL; + + errcode = + p9pdu_readf(pdu, optional, "w", nwqid); + if (!errcode) { + *wqids = + kmalloc(*nwqid * + sizeof(struct p9_qid), + GFP_KERNEL); + if (*wqids == NULL) + errcode = -ENOMEM; + } + + if (!errcode) { + int i; + + for (i = 0; i < *nwqid; i++) { + errcode = + p9pdu_readf(pdu, optional, + "Q", + &(*wqids)[i]); + if (errcode) + break; + } + } + + if (errcode) { + kfree(*wqids); + *wqids = NULL; + } + } + break; + case '?': + if (!optional) + return 0; + break; + default: + BUG(); + break; + } + + if (errcode) + break; + } + + return errcode; +} + +int +p9pdu_vwritef(struct p9_fcall *pdu, int optional, const char *fmt, va_list ap) +{ + const char *ptr; + int errcode = 0; + + for (ptr = fmt; *ptr; ptr++) { + switch (*ptr) { + case 'b':{ + int8_t val = va_arg(ap, int); + if (pdu_write(pdu, &val, sizeof(val))) + errcode = -EFAULT; + } + break; + case 'w':{ + int16_t val = va_arg(ap, int); + if (pdu_write(pdu, &val, sizeof(val))) + errcode = -EFAULT; + } + break; + case 'd':{ + int32_t val = va_arg(ap, int32_t); + if (pdu_write(pdu, &val, sizeof(val))) + errcode = -EFAULT; + } + break; + case 'q':{ + int64_t val = va_arg(ap, int64_t); + if (pdu_write(pdu, &val, sizeof(val))) + errcode = -EFAULT; + } + break; + case 's':{ + const char *ptr = va_arg(ap, const char *); + int16_t len = 0; + if (ptr) + len = MIN(strlen(ptr), USHORT_MAX); + + errcode = p9pdu_writef(pdu, optional, "w", len); + if (!errcode && pdu_write(pdu, ptr, len)) + errcode = -EFAULT; + } + break; + case 'Q':{ + const struct p9_qid *qid = + va_arg(ap, const struct p9_qid *); + errcode = + p9pdu_writef(pdu, optional, "bdq", + qid->type, qid->version, + qid->path); + } break; + case 'S':{ + const struct p9_wstat *stbuf = + va_arg(ap, const struct p9_wstat *); + errcode = + p9pdu_writef(pdu, optional, + "wwdQdddqssss?sddd", + stbuf->size, stbuf->type, + stbuf->dev, &stbuf->qid, + stbuf->mode, stbuf->atime, + stbuf->mtime, stbuf->length, + stbuf->name, stbuf->uid, + stbuf->gid, stbuf->muid, + stbuf->extension, stbuf->n_uid, + stbuf->n_gid, stbuf->n_muid); + } break; + case 'D':{ + int32_t count = va_arg(ap, int32_t); + const void *data = va_arg(ap, const void *); + + errcode = + p9pdu_writef(pdu, optional, "d", count); + if (!errcode && pdu_write(pdu, data, count)) + errcode = -EFAULT; + } + break; + case 'U':{ + int32_t count = va_arg(ap, int32_t); + const char __user *udata = + va_arg(ap, const void *); + errcode = + p9pdu_writef(pdu, optional, "d", count); + if (!errcode && pdu_write_u(pdu, udata, count)) + errcode = -EFAULT; + } + break; + case 'T':{ + int16_t nwname = va_arg(ap, int); + const char **wnames = va_arg(ap, const char **); + + errcode = + p9pdu_writef(pdu, optional, "w", nwname); + if (!errcode) { + int i; + + for (i = 0; i < nwname; i++) { + errcode = + p9pdu_writef(pdu, optional, + "s", + wnames[i]); + if (errcode) + break; + } + } + } + break; + case 'R':{ + int16_t nwqid = va_arg(ap, int); + struct p9_qid *wqids = + va_arg(ap, struct p9_qid *); + + errcode = + p9pdu_writef(pdu, optional, "w", nwqid); + if (!errcode) { + int i; + + for (i = 0; i < nwqid; i++) { + errcode = + p9pdu_writef(pdu, optional, + "Q", + &wqids[i]); + if (errcode) + break; + } + } + } + break; + case '?': + if (!optional) + return 0; + break; + default: + BUG(); + break; + } + + if (errcode) + break; + } + + return errcode; +} + +int p9pdu_readf(struct p9_fcall *pdu, int optional, const char *fmt, ...) +{ + va_list ap; + int ret; + + va_start(ap, fmt); + ret = p9pdu_vreadf(pdu, optional, fmt, ap); + va_end(ap); + + return ret; +} + +static int +p9pdu_writef(struct p9_fcall *pdu, int optional, const char *fmt, ...) +{ + va_list ap; + int ret; + + va_start(ap, fmt); + ret = p9pdu_vwritef(pdu, optional, fmt, ap); + va_end(ap); + + return ret; +} + +int p9stat_read(char *buf, int len, struct p9_wstat *st, int dotu) +{ + struct p9_fcall fake_pdu; + int ret; + + fake_pdu.size = len; + fake_pdu.capacity = len; + fake_pdu.sdata = buf; + fake_pdu.offset = 0; + + ret = p9pdu_readf(&fake_pdu, dotu, "S", st); + if (ret) { + P9_DPRINTK(P9_DEBUG_9P, "<<< p9stat_read failed: %d\n", ret); + p9pdu_dump(1, &fake_pdu); + } + + return ret; +} +EXPORT_SYMBOL(p9stat_read); + +int p9pdu_prepare(struct p9_fcall *pdu, int16_t tag, int8_t type) +{ + return p9pdu_writef(pdu, 0, "dbw", 0, type, tag); +} + +int p9pdu_finalize(struct p9_fcall *pdu) +{ + int size = pdu->size; + int err; + + pdu->size = 0; + err = p9pdu_writef(pdu, 0, "d", size); + pdu->size = size; + + if ((p9_debug_level & P9_DEBUG_PKT) == P9_DEBUG_PKT) + p9pdu_dump(0, pdu); + + P9_DPRINTK(P9_DEBUG_9P, ">>> size=%d type: %d tag: %d\n", pdu->size, + pdu->id, pdu->tag); + + return err; +} + +void p9pdu_reset(struct p9_fcall *pdu) +{ + pdu->offset = 0; + pdu->size = 0; +} diff --git a/net/9p/protocol.h b/net/9p/protocol.h new file mode 100644 index 000000000000..ccde462e7ac5 --- /dev/null +++ b/net/9p/protocol.h @@ -0,0 +1,34 @@ +/* + * net/9p/protocol.h + * + * 9P Protocol Support Code + * + * Copyright (C) 2008 by Eric Van Hensbergen <ericvh@gmail.com> + * + * Base on code from Anthony Liguori <aliguori@us.ibm.com> + * Copyright (C) 2008 by IBM, Corp. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to: + * Free Software Foundation + * 51 Franklin Street, Fifth Floor + * Boston, MA 02111-1301 USA + * + */ + +int +p9pdu_vwritef(struct p9_fcall *pdu, int optional, const char *fmt, va_list ap); +int p9pdu_readf(struct p9_fcall *pdu, int optional, 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 *); +void p9pdu_reset(struct p9_fcall *pdu); diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c index d652baf5ff91..be65d8242fd2 100644 --- a/net/9p/trans_fd.c +++ b/net/9p/trans_fd.c @@ -39,12 +39,11 @@ #include <linux/file.h> #include <linux/parser.h> #include <net/9p/9p.h> +#include <net/9p/client.h> #include <net/9p/transport.h> #define P9_PORT 564 #define MAX_SOCK_BUF (64*1024) -#define ERREQFLUSH 1 -#define SCHED_TIMEOUT 10 #define MAXPOLLWADDR 2 /** @@ -61,7 +60,6 @@ struct p9_fd_opts { u16 port; }; - /** * struct p9_trans_fd - transport state * @rd: reference to file to read from @@ -86,7 +84,7 @@ enum { Opt_port, Opt_rfdno, Opt_wfdno, Opt_err, }; -static match_table_t tokens = { +static const match_table_t tokens = { {Opt_port, "port=%u"}, {Opt_rfdno, "rfdno=%u"}, {Opt_wfdno, "wfdno=%u"}, @@ -100,60 +98,22 @@ enum { Wpending = 8, /* can write */ }; -enum { - None, - Flushing, - Flushed, -}; - -struct p9_req; -typedef void (*p9_conn_req_callback)(struct p9_req *req, void *a); - -/** - * struct p9_req - fd mux encoding of an rpc transaction - * @lock: protects req_list - * @tag: numeric tag for rpc transaction - * @tcall: request &p9_fcall structure - * @rcall: response &p9_fcall structure - * @err: error state - * @cb: callback for when response is received - * @cba: argument to pass to callback - * @flush: flag to indicate RPC has been flushed - * @req_list: list link for higher level objects to chain requests - * - */ - -struct p9_req { - spinlock_t lock; - int tag; - struct p9_fcall *tcall; - struct p9_fcall *rcall; - int err; - p9_conn_req_callback cb; - void *cba; - int flush; - struct list_head req_list; -}; - -struct p9_mux_poll_task { - struct task_struct *task; - struct list_head mux_list; - int muxnum; +struct p9_poll_wait { + struct p9_conn *conn; + wait_queue_t wait; + wait_queue_head_t *wait_addr; }; /** * struct p9_conn - fd mux connection state information - * @lock: protects mux_list (?) * @mux_list: list link for mux to manage multiple connections (?) - * @poll_task: task polling on this connection - * @msize: maximum size for connection (dup) - * @extended: 9p2000.u flag (dup) - * @trans: reference to transport instance for this connection - * @tagpool: id accounting for transactions + * @client: reference to client instance for this connection * @err: error state * @req_list: accounting for requests which have been sent * @unsent_req_list: accounting for requests that haven't been sent - * @rcall: current response &p9_fcall structure + * @req: current request being processed (if any) + * @tmp_buf: temporary buffer to read in header + * @rsize: amount to read for current frame * @rpos: read position in current frame * @rbuf: current read buffer * @wpos: write position for current frame @@ -169,409 +129,300 @@ struct p9_mux_poll_task { */ struct p9_conn { - spinlock_t lock; /* protect lock structure */ struct list_head mux_list; - struct p9_mux_poll_task *poll_task; - int msize; - unsigned char extended; - struct p9_trans *trans; - struct p9_idpool *tagpool; + struct p9_client *client; int err; struct list_head req_list; struct list_head unsent_req_list; - struct p9_fcall *rcall; + struct p9_req_t *req; + char tmp_buf[7]; + int rsize; int rpos; char *rbuf; int wpos; int wsize; char *wbuf; - wait_queue_t poll_wait[MAXPOLLWADDR]; - wait_queue_head_t *poll_waddr[MAXPOLLWADDR]; + struct list_head poll_pending_link; + struct p9_poll_wait poll_wait[MAXPOLLWADDR]; poll_table pt; struct work_struct rq; struct work_struct wq; unsigned long wsched; }; -/** - * struct p9_mux_rpc - fd mux rpc accounting structure - * @m: connection this request was issued on - * @err: error state - * @tcall: request &p9_fcall - * @rcall: response &p9_fcall - * @wqueue: wait queue that client is blocked on for this rpc - * - * Bug: isn't this information duplicated elsewhere like &p9_req - */ - -struct p9_mux_rpc { - struct p9_conn *m; - int err; - struct p9_fcall *tcall; - struct p9_fcall *rcall; - wait_queue_head_t wqueue; -}; - -static int p9_poll_proc(void *); -static void p9_read_work(struct work_struct *work); -static void p9_write_work(struct work_struct *work); -static void p9_pollwait(struct file *filp, wait_queue_head_t *wait_address, - poll_table *p); -static int p9_fd_write(struct p9_trans *trans, void *v, int len); -static int p9_fd_read(struct p9_trans *trans, void *v, int len); - -static DEFINE_MUTEX(p9_mux_task_lock); +static 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 int p9_mux_num; -static int p9_mux_poll_task_num; -static struct p9_mux_poll_task p9_mux_poll_tasks[100]; - -static void p9_conn_destroy(struct p9_conn *); -static unsigned int p9_fd_poll(struct p9_trans *trans, - struct poll_table_struct *pt); - -#ifdef P9_NONBLOCK -static int p9_conn_rpcnb(struct p9_conn *m, struct p9_fcall *tc, - p9_conn_req_callback cb, void *a); -#endif /* P9_NONBLOCK */ - -static void p9_conn_cancel(struct p9_conn *m, int err); - -static u16 p9_mux_get_tag(struct p9_conn *m) +static void p9_mux_poll_stop(struct p9_conn *m) { - int tag; + unsigned long flags; + int i; - tag = p9_idpool_get(m->tagpool); - if (tag < 0) - return P9_NOTAG; - else - return (u16) tag; -} + for (i = 0; i < ARRAY_SIZE(m->poll_wait); i++) { + struct p9_poll_wait *pwait = &m->poll_wait[i]; -static void p9_mux_put_tag(struct p9_conn *m, u16 tag) -{ - if (tag != P9_NOTAG && p9_idpool_check(tag, m->tagpool)) - p9_idpool_put(tag, m->tagpool); + if (pwait->wait_addr) { + remove_wait_queue(pwait->wait_addr, &pwait->wait); + pwait->wait_addr = NULL; + } + } + + spin_lock_irqsave(&p9_poll_lock, flags); + list_del_init(&m->poll_pending_link); + spin_unlock_irqrestore(&p9_poll_lock, flags); } /** - * p9_mux_calc_poll_procs - calculates the number of polling procs - * @muxnum: number of mounts + * p9_conn_cancel - cancel all pending requests with error + * @m: mux data + * @err: error code * - * Calculation is based on the number of mounted v9fs filesystems. - * The current implementation returns sqrt of the number of mounts. */ -static int p9_mux_calc_poll_procs(int muxnum) +static void p9_conn_cancel(struct p9_conn *m, int err) { - int n; - - if (p9_mux_poll_task_num) - n = muxnum / p9_mux_poll_task_num + - (muxnum % p9_mux_poll_task_num ? 1 : 0); - else - n = 1; - - if (n > ARRAY_SIZE(p9_mux_poll_tasks)) - n = ARRAY_SIZE(p9_mux_poll_tasks); - - return n; -} + struct p9_req_t *req, *rtmp; + unsigned long flags; + LIST_HEAD(cancel_list); -static int p9_mux_poll_start(struct p9_conn *m) -{ - int i, n; - struct p9_mux_poll_task *vpt, *vptlast; - struct task_struct *pproc; - - P9_DPRINTK(P9_DEBUG_MUX, "mux %p muxnum %d procnum %d\n", m, p9_mux_num, - p9_mux_poll_task_num); - mutex_lock(&p9_mux_task_lock); - - n = p9_mux_calc_poll_procs(p9_mux_num + 1); - if (n > p9_mux_poll_task_num) { - for (i = 0; i < ARRAY_SIZE(p9_mux_poll_tasks); i++) { - if (p9_mux_poll_tasks[i].task == NULL) { - vpt = &p9_mux_poll_tasks[i]; - P9_DPRINTK(P9_DEBUG_MUX, "create proc %p\n", - vpt); - pproc = kthread_create(p9_poll_proc, vpt, - "v9fs-poll"); - - if (!IS_ERR(pproc)) { - vpt->task = pproc; - INIT_LIST_HEAD(&vpt->mux_list); - vpt->muxnum = 0; - p9_mux_poll_task_num++; - wake_up_process(vpt->task); - } - break; - } - } + P9_DPRINTK(P9_DEBUG_ERROR, "mux %p err %d\n", m, err); - if (i >= ARRAY_SIZE(p9_mux_poll_tasks)) - P9_DPRINTK(P9_DEBUG_ERROR, - "warning: no free poll slots\n"); - } + spin_lock_irqsave(&m->client->lock, flags); - n = (p9_mux_num + 1) / p9_mux_poll_task_num + - ((p9_mux_num + 1) % p9_mux_poll_task_num ? 1 : 0); - - vptlast = NULL; - for (i = 0; i < ARRAY_SIZE(p9_mux_poll_tasks); i++) { - vpt = &p9_mux_poll_tasks[i]; - if (vpt->task != NULL) { - vptlast = vpt; - if (vpt->muxnum < n) { - P9_DPRINTK(P9_DEBUG_MUX, "put in proc %d\n", i); - list_add(&m->mux_list, &vpt->mux_list); - vpt->muxnum++; - m->poll_task = vpt; - memset(&m->poll_waddr, 0, - sizeof(m->poll_waddr)); - init_poll_funcptr(&m->pt, p9_pollwait); - break; - } - } + if (m->err) { + spin_unlock_irqrestore(&m->client->lock, flags); + return; } - if (i >= ARRAY_SIZE(p9_mux_poll_tasks)) { - if (vptlast == NULL) { - mutex_unlock(&p9_mux_task_lock); - return -ENOMEM; - } + m->err = err; - P9_DPRINTK(P9_DEBUG_MUX, "put in proc %d\n", i); - list_add(&m->mux_list, &vptlast->mux_list); - vptlast->muxnum++; - m->poll_task = vptlast; - memset(&m->poll_waddr, 0, sizeof(m->poll_waddr)); - init_poll_funcptr(&m->pt, p9_pollwait); + 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); } - - p9_mux_num++; - mutex_unlock(&p9_mux_task_lock); - - return 0; -} - -static void p9_mux_poll_stop(struct p9_conn *m) -{ - int i; - struct p9_mux_poll_task *vpt; - - mutex_lock(&p9_mux_task_lock); - vpt = m->poll_task; - list_del(&m->mux_list); - for (i = 0; i < ARRAY_SIZE(m->poll_waddr); i++) { - if (m->poll_waddr[i] != NULL) { - remove_wait_queue(m->poll_waddr[i], &m->poll_wait[i]); - m->poll_waddr[i] = NULL; - } + 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); } - vpt->muxnum--; - if (!vpt->muxnum) { - P9_DPRINTK(P9_DEBUG_MUX, "destroy proc %p\n", vpt); - kthread_stop(vpt->task); - vpt->task = NULL; - p9_mux_poll_task_num--; + spin_unlock_irqrestore(&m->client->lock, flags); + + list_for_each_entry_safe(req, rtmp, &cancel_list, req_list) { + list_del(&req->req_list); + P9_DPRINTK(P9_DEBUG_ERROR, "call back req %p\n", req); + p9_client_cb(m->client, req); } - p9_mux_num--; - mutex_unlock(&p9_mux_task_lock); } -/** - * p9_conn_create - allocate and initialize the per-session mux data - * @trans: transport structure - * - * Note: Creates the polling task if this is the first session. - */ - -static struct p9_conn *p9_conn_create(struct p9_trans *trans) +static unsigned int +p9_fd_poll(struct p9_client *client, struct poll_table_struct *pt) { - int i, n; - struct p9_conn *m; + int ret, n; + struct p9_trans_fd *ts = NULL; - P9_DPRINTK(P9_DEBUG_MUX, "transport %p msize %d\n", trans, - trans->msize); - m = kzalloc(sizeof(struct p9_conn), GFP_KERNEL); - if (!m) - return ERR_PTR(-ENOMEM); + if (client && client->status == Connected) + ts = client->trans; - spin_lock_init(&m->lock); - INIT_LIST_HEAD(&m->mux_list); - m->msize = trans->msize; - m->extended = trans->extended; - m->trans = trans; - m->tagpool = p9_idpool_create(); - if (IS_ERR(m->tagpool)) { - kfree(m); - return ERR_PTR(-ENOMEM); - } + if (!ts) + return -EREMOTEIO; - INIT_LIST_HEAD(&m->req_list); - INIT_LIST_HEAD(&m->unsent_req_list); - INIT_WORK(&m->rq, p9_read_work); - INIT_WORK(&m->wq, p9_write_work); - n = p9_mux_poll_start(m); - if (n) { - kfree(m); - return ERR_PTR(n); - } + if (!ts->rd->f_op || !ts->rd->f_op->poll) + return -EIO; - n = p9_fd_poll(trans, &m->pt); - if (n & POLLIN) { - P9_DPRINTK(P9_DEBUG_MUX, "mux %p can read\n", m); - set_bit(Rpending, &m->wsched); - } + if (!ts->wr->f_op || !ts->wr->f_op->poll) + return -EIO; - if (n & POLLOUT) { - P9_DPRINTK(P9_DEBUG_MUX, "mux %p can write\n", m); - set_bit(Wpending, &m->wsched); - } + ret = ts->rd->f_op->poll(ts->rd, pt); + if (ret < 0) + return ret; - for (i = 0; i < ARRAY_SIZE(m->poll_waddr); i++) { - if (IS_ERR(m->poll_waddr[i])) { - p9_mux_poll_stop(m); - kfree(m); - return (void *)m->poll_waddr; /* the error code */ - } + if (ts->rd != ts->wr) { + n = ts->wr->f_op->poll(ts->wr, pt); + if (n < 0) + return n; + ret = (ret & ~POLLOUT) | (n & ~POLLIN); } - return m; + return ret; } /** - * p9_mux_destroy - cancels all pending requests and frees mux resources - * @m: mux to destroy + * p9_fd_read- read from a fd + * @client: client instance + * @v: buffer to receive data into + * @len: size of receive buffer * */ -static void p9_conn_destroy(struct p9_conn *m) +static int p9_fd_read(struct p9_client *client, void *v, int len) { - P9_DPRINTK(P9_DEBUG_MUX, "mux %p prev %p next %p\n", m, - m->mux_list.prev, m->mux_list.next); + int ret; + struct p9_trans_fd *ts = NULL; - p9_mux_poll_stop(m); - cancel_work_sync(&m->rq); - cancel_work_sync(&m->wq); + if (client && client->status != Disconnected) + ts = client->trans; - p9_conn_cancel(m, -ECONNRESET); + if (!ts) + return -EREMOTEIO; - m->trans = NULL; - p9_idpool_destroy(m->tagpool); - kfree(m); + if (!(ts->rd->f_flags & O_NONBLOCK)) + P9_DPRINTK(P9_DEBUG_ERROR, "blocking read ...\n"); + + ret = kernel_read(ts->rd, ts->rd->f_pos, v, len); + if (ret <= 0 && ret != -ERESTARTSYS && ret != -EAGAIN) + client->status = Disconnected; + return ret; } /** - * p9_pollwait - add poll task to the wait queue - * @filp: file pointer being polled - * @wait_address: wait_q to block on - * @p: poll state + * p9_read_work - called when there is some data to be read from a transport + * @work: container of work to be done * - * called by files poll operation to add v9fs-poll task to files wait queue */ -static void -p9_pollwait(struct file *filp, wait_queue_head_t *wait_address, poll_table *p) +static void p9_read_work(struct work_struct *work) { - int i; + int n, err; struct p9_conn *m; - m = container_of(p, struct p9_conn, pt); - for (i = 0; i < ARRAY_SIZE(m->poll_waddr); i++) - if (m->poll_waddr[i] == NULL) - break; + m = container_of(work, struct p9_conn, rq); - if (i >= ARRAY_SIZE(m->poll_waddr)) { - P9_DPRINTK(P9_DEBUG_ERROR, "not enough wait_address slots\n"); + if (m->err < 0) return; - } - m->poll_waddr[i] = wait_address; + P9_DPRINTK(P9_DEBUG_TRANS, "start mux %p pos %d\n", m, m->rpos); - if (!wait_address) { - P9_DPRINTK(P9_DEBUG_ERROR, "no wait_address\n"); - m->poll_waddr[i] = ERR_PTR(-EIO); + if (!m->rbuf) { + m->rbuf = m->tmp_buf; + m->rpos = 0; + m->rsize = 7; /* start by reading header */ + } + + 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); + 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); + if (err == -EAGAIN) { + clear_bit(Rworksched, &m->wsched); return; } - init_waitqueue_entry(&m->poll_wait[i], m->poll_task->task); - add_wait_queue(wait_address, &m->poll_wait[i]); -} + if (err <= 0) + goto error; -/** - * p9_poll_mux - polls a mux and schedules read or write works if necessary - * @m: connection to poll - * - */ + m->rpos += err; -static void p9_poll_mux(struct p9_conn *m) -{ - int n; + if ((!m->req) && (m->rpos == m->rsize)) { /* header read in */ + u16 tag; + P9_DPRINTK(P9_DEBUG_TRANS, "got new header\n"); - if (m->err < 0) - return; + 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); + err = -EIO; + goto error; + } - n = p9_fd_poll(m->trans, NULL); - if (n < 0 || n & (POLLERR | POLLHUP | POLLNVAL)) { - P9_DPRINTK(P9_DEBUG_MUX, "error mux %p err %d\n", m, n); - if (n >= 0) - n = -ECONNRESET; - p9_conn_cancel(m, n); - } + 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); - if (n & POLLIN) { - set_bit(Rpending, &m->wsched); - P9_DPRINTK(P9_DEBUG_MUX, "mux %p can read\n", m); - if (!test_and_set_bit(Rworksched, &m->wsched)) { - P9_DPRINTK(P9_DEBUG_MUX, "schedule read work %p\n", m); - queue_work(p9_mux_wq, &m->rq); + m->req = p9_tag_lookup(m->client, tag); + if (!m->req) { + P9_DPRINTK(P9_DEBUG_ERROR, "Unexpected packet tag %d\n", + tag); + err = -EIO; + goto error; } - } - if (n & POLLOUT) { - set_bit(Wpending, &m->wsched); - P9_DPRINTK(P9_DEBUG_MUX, "mux %p can write\n", m); - if ((m->wsize || !list_empty(&m->unsent_req_list)) - && !test_and_set_bit(Wworksched, &m->wsched)) { - P9_DPRINTK(P9_DEBUG_MUX, "schedule write work %p\n", m); - queue_work(p9_mux_wq, &m->wq); + if (m->req->rc == NULL) { + m->req->rc = kmalloc(sizeof(struct p9_fcall) + + m->client->msize, GFP_KERNEL); + if (!m->req->rc) { + m->req = NULL; + err = -ENOMEM; + goto error; + } } + m->rbuf = (char *)m->req->rc + sizeof(struct p9_fcall); + memcpy(m->rbuf, m->tmp_buf, m->rsize); + m->rsize = n; } + + /* not an else because some packets (like clunk) have no payload */ + if ((m->req) && (m->rpos == m->rsize)) { /* packet is read in */ + P9_DPRINTK(P9_DEBUG_TRANS, "got new packet\n"); + spin_lock(&m->client->lock); + list_del(&m->req->req_list); + spin_unlock(&m->client->lock); + p9_client_cb(m->client, m->req); + + m->rbuf = NULL; + m->rpos = 0; + m->rsize = 0; + m->req = NULL; + } + + 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); + + return; +error: + p9_conn_cancel(m, err); + clear_bit(Rworksched, &m->wsched); } /** - * p9_poll_proc - poll worker thread - * @a: thread state and arguments - * - * polls all v9fs transports for new events and queues the appropriate - * work to the work queue + * p9_fd_write - write to a socket + * @client: client instance + * @v: buffer to send data from + * @len: size of send buffer * */ -static int p9_poll_proc(void *a) +static int p9_fd_write(struct p9_client *client, void *v, int len) { - struct p9_conn *m, *mtmp; - struct p9_mux_poll_task *vpt; + int ret; + mm_segment_t oldfs; + struct p9_trans_fd *ts = NULL; - vpt = a; - P9_DPRINTK(P9_DEBUG_MUX, "start %p %p\n", current, vpt); - while (!kthread_should_stop()) { - set_current_state(TASK_INTERRUPTIBLE); + if (client && client->status != Disconnected) + ts = client->trans; - list_for_each_entry_safe(m, mtmp, &vpt->mux_list, mux_list) { - p9_poll_mux(m); - } + if (!ts) + return -EREMOTEIO; - P9_DPRINTK(P9_DEBUG_MUX, "sleeping...\n"); - schedule_timeout(SCHED_TIMEOUT * HZ); - } + if (!(ts->wr->f_flags & O_NONBLOCK)) + P9_DPRINTK(P9_DEBUG_ERROR, "blocking write ...\n"); - __set_current_state(TASK_RUNNING); - P9_DPRINTK(P9_DEBUG_MUX, "finish\n"); - return 0; + oldfs = get_fs(); + set_fs(get_ds()); + /* The cast to a user pointer is valid due to the set_fs() */ + ret = vfs_write(ts->wr, (void __user *)v, len, &ts->wr->f_pos); + set_fs(oldfs); + + if (ret <= 0 && ret != -ERESTARTSYS && ret != -EAGAIN) + client->status = Disconnected; + return ret; } /** @@ -584,7 +435,7 @@ static void p9_write_work(struct work_struct *work) { int n, err; struct p9_conn *m; - struct p9_req *req; + struct p9_req_t *req; m = container_of(work, struct p9_conn, wq); @@ -599,25 +450,23 @@ static void p9_write_work(struct work_struct *work) return; } - spin_lock(&m->lock); -again: - req = list_entry(m->unsent_req_list.next, struct p9_req, + spin_lock(&m->client->lock); + req = list_entry(m->unsent_req_list.next, struct p9_req_t, req_list); + req->status = REQ_STATUS_SENT; list_move_tail(&req->req_list, &m->req_list); - if (req->err == ERREQFLUSH) - goto again; - m->wbuf = req->tcall->sdata; - m->wsize = req->tcall->size; + m->wbuf = req->tc->sdata; + m->wsize = req->tc->size; m->wpos = 0; - spin_unlock(&m->lock); + spin_unlock(&m->client->lock); } - P9_DPRINTK(P9_DEBUG_MUX, "mux %p pos %d size %d\n", m, m->wpos, + P9_DPRINTK(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->trans, m->wbuf + m->wpos, m->wsize - m->wpos); - P9_DPRINTK(P9_DEBUG_MUX, "mux %p sent %d bytes\n", m, err); + 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; @@ -638,10 +487,10 @@ again: if (test_and_clear_bit(Wpending, &m->wsched)) n = POLLOUT; else - n = p9_fd_poll(m->trans, NULL); + n = p9_fd_poll(m->client, NULL); if (n & POLLOUT) { - P9_DPRINTK(P9_DEBUG_MUX, "schedule write work %p\n", m); + P9_DPRINTK(P9_DEBUG_TRANS, "sched write work %p\n", m); queue_work(p9_mux_wq, &m->wq); } else clear_bit(Wworksched, &m->wsched); @@ -655,504 +504,197 @@ error: clear_bit(Wworksched, &m->wsched); } -static void process_request(struct p9_conn *m, struct p9_req *req) +static int p9_pollwake(wait_queue_t *wait, unsigned mode, int sync, void *key) { - int ecode; - struct p9_str *ename; - - if (!req->err && req->rcall->id == P9_RERROR) { - ecode = req->rcall->params.rerror.errno; - ename = &req->rcall->params.rerror.error; - - P9_DPRINTK(P9_DEBUG_MUX, "Rerror %.*s\n", ename->len, - ename->str); - - if (m->extended) - req->err = -ecode; + 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); - if (!req->err) { - req->err = p9_errstr2errno(ename->str, ename->len); + 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); - /* string match failed */ - if (!req->err) { - PRINT_FCALL_ERROR("unknown error", req->rcall); - req->err = -ESERVERFAULT; - } - } - } else if (req->tcall && req->rcall->id != req->tcall->id + 1) { - P9_DPRINTK(P9_DEBUG_ERROR, - "fcall mismatch: expected %d, got %d\n", - req->tcall->id + 1, req->rcall->id); - if (!req->err) - req->err = -EIO; - } + /* perform the default wake up operation */ + return default_wake_function(&dummy_wait, mode, sync, key); } /** - * p9_read_work - called when there is some data to be read from a transport - * @work: container of work to be done + * p9_pollwait - add poll task to the wait queue + * @filp: file pointer being polled + * @wait_address: wait_q to block on + * @p: poll state * + * called by files poll operation to add v9fs-poll task to files wait queue */ -static void p9_read_work(struct work_struct *work) +static void +p9_pollwait(struct file *filp, wait_queue_head_t *wait_address, poll_table *p) { - int n, err; - struct p9_conn *m; - struct p9_req *req, *rptr, *rreq; - struct p9_fcall *rcall; - char *rbuf; - - m = container_of(work, struct p9_conn, rq); - - if (m->err < 0) - return; - - rcall = NULL; - P9_DPRINTK(P9_DEBUG_MUX, "start mux %p pos %d\n", m, m->rpos); + struct p9_conn *m = container_of(p, struct p9_conn, pt); + struct p9_poll_wait *pwait = NULL; + int i; - if (!m->rcall) { - m->rcall = - kmalloc(sizeof(struct p9_fcall) + m->msize, GFP_KERNEL); - if (!m->rcall) { - err = -ENOMEM; - goto error; + for (i = 0; i < ARRAY_SIZE(m->poll_wait); i++) { + if (m->poll_wait[i].wait_addr == NULL) { + pwait = &m->poll_wait[i]; + break; } - - m->rbuf = (char *)m->rcall + sizeof(struct p9_fcall); - m->rpos = 0; } - clear_bit(Rpending, &m->wsched); - err = p9_fd_read(m->trans, m->rbuf + m->rpos, m->msize - m->rpos); - P9_DPRINTK(P9_DEBUG_MUX, "mux %p got %d bytes\n", m, err); - if (err == -EAGAIN) { - clear_bit(Rworksched, &m->wsched); + if (!pwait) { + P9_DPRINTK(P9_DEBUG_ERROR, "not enough wait_address slots\n"); return; } - if (err <= 0) - goto error; - - m->rpos += err; - while (m->rpos > 4) { - n = le32_to_cpu(*(__le32 *) m->rbuf); - if (n >= m->msize) { - P9_DPRINTK(P9_DEBUG_ERROR, - "requested packet size too big: %d\n", n); - err = -EIO; - goto error; - } - - if (m->rpos < n) - break; - - err = - p9_deserialize_fcall(m->rbuf, n, m->rcall, m->extended); - if (err < 0) - goto error; - -#ifdef CONFIG_NET_9P_DEBUG - if ((p9_debug_level&P9_DEBUG_FCALL) == P9_DEBUG_FCALL) { - char buf[150]; - - p9_printfcall(buf, sizeof(buf), m->rcall, - m->extended); - printk(KERN_NOTICE ">>> %p %s\n", m, buf); - } -#endif - - rcall = m->rcall; - rbuf = m->rbuf; - if (m->rpos > n) { - m->rcall = kmalloc(sizeof(struct p9_fcall) + m->msize, - GFP_KERNEL); - if (!m->rcall) { - err = -ENOMEM; - goto error; - } - - m->rbuf = (char *)m->rcall + sizeof(struct p9_fcall); - memmove(m->rbuf, rbuf + n, m->rpos - n); - m->rpos -= n; - } else { - m->rcall = NULL; - m->rbuf = NULL; - m->rpos = 0; - } - - P9_DPRINTK(P9_DEBUG_MUX, "mux %p fcall id %d tag %d\n", m, - rcall->id, rcall->tag); - - req = NULL; - spin_lock(&m->lock); - list_for_each_entry_safe(rreq, rptr, &m->req_list, req_list) { - if (rreq->tag == rcall->tag) { - req = rreq; - if (req->flush != Flushing) - list_del(&req->req_list); - break; - } - } - spin_unlock(&m->lock); - - if (req) { - req->rcall = rcall; - process_request(m, req); - - if (req->flush != Flushing) { - if (req->cb) - (*req->cb) (req, req->cba); - else - kfree(req->rcall); - } - } else { - if (err >= 0 && rcall->id != P9_RFLUSH) - P9_DPRINTK(P9_DEBUG_ERROR, - "unexpected response mux %p id %d tag %d\n", - m, rcall->id, rcall->tag); - kfree(rcall); - } - } - - if (!list_empty(&m->req_list)) { - if (test_and_clear_bit(Rpending, &m->wsched)) - n = POLLIN; - else - n = p9_fd_poll(m->trans, NULL); - - if (n & POLLIN) { - P9_DPRINTK(P9_DEBUG_MUX, "schedule read work %p\n", m); - queue_work(p9_mux_wq, &m->rq); - } else - clear_bit(Rworksched, &m->wsched); - } else - clear_bit(Rworksched, &m->wsched); - - return; - -error: - p9_conn_cancel(m, err); - clear_bit(Rworksched, &m->wsched); + pwait->conn = m; + pwait->wait_addr = wait_address; + init_waitqueue_func_entry(&pwait->wait, p9_pollwake); + add_wait_queue(wait_address, &pwait->wait); } /** - * p9_send_request - send 9P request - * The function can sleep until the request is scheduled for sending. - * The function can be interrupted. Return from the function is not - * a guarantee that the request is sent successfully. Can return errors - * that can be retrieved by PTR_ERR macros. - * - * @m: mux data - * @tc: request to be sent - * @cb: callback function to call when response is received - * @cba: parameter to pass to the callback function + * p9_conn_create - allocate and initialize the per-session mux data + * @client: client instance * + * Note: Creates the polling task if this is the first session. */ -static struct p9_req *p9_send_request(struct p9_conn *m, - struct p9_fcall *tc, - p9_conn_req_callback cb, void *cba) +static struct p9_conn *p9_conn_create(struct p9_client *client) { int n; - struct p9_req *req; - - P9_DPRINTK(P9_DEBUG_MUX, "mux %p task %p tcall %p id %d\n", m, current, - tc, tc->id); - if (m->err < 0) - return ERR_PTR(m->err); - - req = kmalloc(sizeof(struct p9_req), GFP_KERNEL); - if (!req) - return ERR_PTR(-ENOMEM); - - if (tc->id == P9_TVERSION) - n = P9_NOTAG; - else - n = p9_mux_get_tag(m); + struct p9_conn *m; - if (n < 0) { - kfree(req); + 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_set_tag(tc, n); + INIT_LIST_HEAD(&m->mux_list); + m->client = client; -#ifdef CONFIG_NET_9P_DEBUG - if ((p9_debug_level&P9_DEBUG_FCALL) == P9_DEBUG_FCALL) { - char buf[150]; + INIT_LIST_HEAD(&m->req_list); + INIT_LIST_HEAD(&m->unsent_req_list); + INIT_WORK(&m->rq, p9_read_work); + INIT_WORK(&m->wq, p9_write_work); + INIT_LIST_HEAD(&m->poll_pending_link); + init_poll_funcptr(&m->pt, p9_pollwait); - p9_printfcall(buf, sizeof(buf), tc, m->extended); - printk(KERN_NOTICE "<<< %p %s\n", m, buf); + n = p9_fd_poll(client, &m->pt); + if (n & POLLIN) { + P9_DPRINTK(P9_DEBUG_TRANS, "mux %p can read\n", m); + set_bit(Rpending, &m->wsched); } -#endif - - spin_lock_init(&req->lock); - req->tag = n; - req->tcall = tc; - req->rcall = NULL; - req->err = 0; - req->cb = cb; - req->cba = cba; - req->flush = None; - - spin_lock(&m->lock); - list_add_tail(&req->req_list, &m->unsent_req_list); - spin_unlock(&m->lock); - - if (test_and_clear_bit(Wpending, &m->wsched)) - n = POLLOUT; - else - n = p9_fd_poll(m->trans, NULL); - if (n & POLLOUT && !test_and_set_bit(Wworksched, &m->wsched)) - queue_work(p9_mux_wq, &m->wq); + if (n & POLLOUT) { + P9_DPRINTK(P9_DEBUG_TRANS, "mux %p can write\n", m); + set_bit(Wpending, &m->wsched); + } - return req; + return m; } -static void p9_mux_free_request(struct p9_conn *m, struct p9_req *req) -{ - p9_mux_put_tag(m, req->tag); - kfree(req); -} +/** + * p9_poll_mux - polls a mux and schedules read or write works if necessary + * @m: connection to poll + * + */ -static void p9_mux_flush_cb(struct p9_req *freq, void *a) +static void p9_poll_mux(struct p9_conn *m) { - int tag; - struct p9_conn *m; - struct p9_req *req, *rreq, *rptr; - - m = a; - P9_DPRINTK(P9_DEBUG_MUX, "mux %p tc %p rc %p err %d oldtag %d\n", m, - freq->tcall, freq->rcall, freq->err, - freq->tcall->params.tflush.oldtag); - - spin_lock(&m->lock); - tag = freq->tcall->params.tflush.oldtag; - req = NULL; - list_for_each_entry_safe(rreq, rptr, &m->req_list, req_list) { - if (rreq->tag == tag) { - req = rreq; - list_del(&req->req_list); - break; - } - } - spin_unlock(&m->lock); + int n; - if (req) { - spin_lock(&req->lock); - req->flush = Flushed; - spin_unlock(&req->lock); + if (m->err < 0) + return; - if (req->cb) - (*req->cb) (req, req->cba); - else - kfree(req->rcall); + 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); + if (n >= 0) + n = -ECONNRESET; + p9_conn_cancel(m, n); } - kfree(freq->tcall); - kfree(freq->rcall); - p9_mux_free_request(m, freq); -} - -static int -p9_mux_flush_request(struct p9_conn *m, struct p9_req *req) -{ - struct p9_fcall *fc; - struct p9_req *rreq, *rptr; - - P9_DPRINTK(P9_DEBUG_MUX, "mux %p req %p tag %d\n", m, req, req->tag); - - /* if a response was received for a request, do nothing */ - spin_lock(&req->lock); - if (req->rcall || req->err) { - spin_unlock(&req->lock); - P9_DPRINTK(P9_DEBUG_MUX, - "mux %p req %p response already received\n", m, req); - return 0; + if (n & POLLIN) { + set_bit(Rpending, &m->wsched); + P9_DPRINTK(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); + } } - req->flush = Flushing; - spin_unlock(&req->lock); - - spin_lock(&m->lock); - /* if the request is not sent yet, just remove it from the list */ - list_for_each_entry_safe(rreq, rptr, &m->unsent_req_list, req_list) { - if (rreq->tag == req->tag) { - P9_DPRINTK(P9_DEBUG_MUX, - "mux %p req %p request is not sent yet\n", m, req); - list_del(&rreq->req_list); - req->flush = Flushed; - spin_unlock(&m->lock); - if (req->cb) - (*req->cb) (req, req->cba); - return 0; + if (n & POLLOUT) { + set_bit(Wpending, &m->wsched); + P9_DPRINTK(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); } } - spin_unlock(&m->lock); - - clear_thread_flag(TIF_SIGPENDING); - fc = p9_create_tflush(req->tag); - p9_send_request(m, fc, p9_mux_flush_cb, m); - return 1; -} - -static void -p9_conn_rpc_cb(struct p9_req *req, void *a) -{ - struct p9_mux_rpc *r; - - P9_DPRINTK(P9_DEBUG_MUX, "req %p r %p\n", req, a); - r = a; - r->rcall = req->rcall; - r->err = req->err; - - if (req->flush != None && !req->err) - r->err = -ERESTARTSYS; - - wake_up(&r->wqueue); } /** - * p9_fd_rpc- sends 9P request and waits until a response is available. - * The function can be interrupted. - * @t: transport data - * @tc: request to be sent - * @rc: pointer where a pointer to the response is stored + * p9_fd_request - send 9P request + * The function can sleep until the request is scheduled for sending. + * The function can be interrupted. Return from the function is not + * a guarantee that the request is sent successfully. + * + * @client: client instance + * @req: request to be sent * */ -int -p9_fd_rpc(struct p9_trans *t, struct p9_fcall *tc, struct p9_fcall **rc) +static int p9_fd_request(struct p9_client *client, struct p9_req_t *req) { - struct p9_trans_fd *p = t->priv; - struct p9_conn *m = p->conn; - int err, sigpending; - unsigned long flags; - struct p9_req *req; - struct p9_mux_rpc r; - - r.err = 0; - r.tcall = tc; - r.rcall = NULL; - r.m = m; - init_waitqueue_head(&r.wqueue); - - if (rc) - *rc = NULL; - - sigpending = 0; - if (signal_pending(current)) { - sigpending = 1; - clear_thread_flag(TIF_SIGPENDING); - } - - req = p9_send_request(m, tc, p9_conn_rpc_cb, &r); - if (IS_ERR(req)) { - err = PTR_ERR(req); - P9_DPRINTK(P9_DEBUG_MUX, "error %d\n", err); - return err; - } + int n; + struct p9_trans_fd *ts = client->trans; + struct p9_conn *m = ts->conn; - err = wait_event_interruptible(r.wqueue, r.rcall != NULL || r.err < 0); - if (r.err < 0) - err = r.err; - - if (err == -ERESTARTSYS && m->trans->status == Connected - && m->err == 0) { - if (p9_mux_flush_request(m, req)) { - /* wait until we get response of the flush message */ - do { - clear_thread_flag(TIF_SIGPENDING); - err = wait_event_interruptible(r.wqueue, - r.rcall || r.err); - } while (!r.rcall && !r.err && err == -ERESTARTSYS && - m->trans->status == Connected && !m->err); - - err = -ERESTARTSYS; - } - sigpending = 1; - } + P9_DPRINTK(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; - if (sigpending) { - spin_lock_irqsave(¤t->sighand->siglock, flags); - recalc_sigpending(); - spin_unlock_irqrestore(¤t->sighand->siglock, flags); - } + spin_lock(&client->lock); + req->status = REQ_STATUS_UNSENT; + list_add_tail(&req->req_list, &m->unsent_req_list); + spin_unlock(&client->lock); - if (rc) - *rc = r.rcall; + if (test_and_clear_bit(Wpending, &m->wsched)) + n = POLLOUT; else - kfree(r.rcall); + n = p9_fd_poll(m->client, NULL); - p9_mux_free_request(m, req); - if (err > 0) - err = -EIO; + if (n & POLLOUT && !test_and_set_bit(Wworksched, &m->wsched)) + queue_work(p9_mux_wq, &m->wq); - return err; + return 0; } -#ifdef P9_NONBLOCK -/** - * p9_conn_rpcnb - sends 9P request without waiting for response. - * @m: mux data - * @tc: request to be sent - * @cb: callback function to be called when response arrives - * @a: value to pass to the callback function - * - */ - -int p9_conn_rpcnb(struct p9_conn *m, struct p9_fcall *tc, - p9_conn_req_callback cb, void *a) +static int p9_fd_cancel(struct p9_client *client, struct p9_req_t *req) { - int err; - struct p9_req *req; + struct p9_trans_fd *ts = client->trans; + struct p9_conn *m = ts->conn; + int ret = 1; - req = p9_send_request(m, tc, cb, a); - if (IS_ERR(req)) { - err = PTR_ERR(req); - P9_DPRINTK(P9_DEBUG_MUX, "error %d\n", err); - return PTR_ERR(req); - } + P9_DPRINTK(P9_DEBUG_TRANS, "mux %p req %p\n", m, req); - P9_DPRINTK(P9_DEBUG_MUX, "mux %p tc %p tag %d\n", m, tc, req->tag); - return 0; -} -#endif /* P9_NONBLOCK */ + spin_lock(&client->lock); + list_del(&req->req_list); -/** - * p9_conn_cancel - cancel all pending requests with error - * @m: mux data - * @err: error code - * - */ - -void p9_conn_cancel(struct p9_conn *m, int err) -{ - struct p9_req *req, *rtmp; - LIST_HEAD(cancel_list); - - P9_DPRINTK(P9_DEBUG_ERROR, "mux %p err %d\n", m, err); - m->err = err; - spin_lock(&m->lock); - list_for_each_entry_safe(req, rtmp, &m->req_list, req_list) { - list_move(&req->req_list, &cancel_list); + if (req->status == REQ_STATUS_UNSENT) { + req->status = REQ_STATUS_FLSHD; + ret = 0; } - list_for_each_entry_safe(req, rtmp, &m->unsent_req_list, req_list) { - list_move(&req->req_list, &cancel_list); - } - spin_unlock(&m->lock); - list_for_each_entry_safe(req, rtmp, &cancel_list, req_list) { - list_del(&req->req_list); - if (!req->err) - req->err = err; + spin_unlock(&client->lock); - if (req->cb) - (*req->cb) (req, req->cba); - else - kfree(req->rcall); - } + return ret; } /** @@ -1216,7 +758,7 @@ static int parse_opts(char *params, struct p9_fd_opts *opts) return 0; } -static int p9_fd_open(struct p9_trans *trans, int rfd, int wfd) +static int p9_fd_open(struct p9_client *client, int rfd, int wfd) { struct p9_trans_fd *ts = kmalloc(sizeof(struct p9_trans_fd), GFP_KERNEL); @@ -1234,13 +776,13 @@ static int p9_fd_open(struct p9_trans *trans, int rfd, int wfd) return -EIO; } - trans->priv = ts; - trans->status = Connected; + client->trans = ts; + client->status = Connected; return 0; } -static int p9_socket_open(struct p9_trans *trans, struct socket *csocket) +static int p9_socket_open(struct p9_client *client, struct socket *csocket) { int fd, ret; @@ -1251,137 +793,65 @@ static int p9_socket_open(struct p9_trans *trans, struct socket *csocket) return fd; } - ret = p9_fd_open(trans, fd, fd); + ret = p9_fd_open(client, fd, fd); if (ret < 0) { P9_EPRINTK(KERN_ERR, "p9_socket_open: failed to open fd\n"); sockfd_put(csocket); return ret; } - ((struct p9_trans_fd *)trans->priv)->rd->f_flags |= O_NONBLOCK; + ((struct p9_trans_fd *)client->trans)->rd->f_flags |= O_NONBLOCK; return 0; } /** - * p9_fd_read- read from a fd - * @trans: transport instance state - * @v: buffer to receive data into - * @len: size of receive buffer - * - */ - -static int p9_fd_read(struct p9_trans *trans, void *v, int len) -{ - int ret; - struct p9_trans_fd *ts = NULL; - - if (trans && trans->status != Disconnected) - ts = trans->priv; - - if (!ts) - return -EREMOTEIO; - - if (!(ts->rd->f_flags & O_NONBLOCK)) - P9_DPRINTK(P9_DEBUG_ERROR, "blocking read ...\n"); - - ret = kernel_read(ts->rd, ts->rd->f_pos, v, len); - if (ret <= 0 && ret != -ERESTARTSYS && ret != -EAGAIN) - trans->status = Disconnected; - return ret; -} - -/** - * p9_fd_write - write to a socket - * @trans: transport instance state - * @v: buffer to send data from - * @len: size of send buffer + * p9_mux_destroy - cancels all pending requests and frees mux resources + * @m: mux to destroy * */ -static int p9_fd_write(struct p9_trans *trans, void *v, int len) -{ - int ret; - mm_segment_t oldfs; - struct p9_trans_fd *ts = NULL; - - if (trans && trans->status != Disconnected) - ts = trans->priv; - - if (!ts) - return -EREMOTEIO; - - if (!(ts->wr->f_flags & O_NONBLOCK)) - P9_DPRINTK(P9_DEBUG_ERROR, "blocking write ...\n"); - - oldfs = get_fs(); - set_fs(get_ds()); - /* The cast to a user pointer is valid due to the set_fs() */ - ret = vfs_write(ts->wr, (void __user *)v, len, &ts->wr->f_pos); - set_fs(oldfs); - - if (ret <= 0 && ret != -ERESTARTSYS && ret != -EAGAIN) - trans->status = Disconnected; - return ret; -} - -static unsigned int -p9_fd_poll(struct p9_trans *trans, struct poll_table_struct *pt) +static void p9_conn_destroy(struct p9_conn *m) { - int ret, n; - struct p9_trans_fd *ts = NULL; - - if (trans && trans->status == Connected) - ts = trans->priv; - - if (!ts) - return -EREMOTEIO; - - if (!ts->rd->f_op || !ts->rd->f_op->poll) - return -EIO; - - if (!ts->wr->f_op || !ts->wr->f_op->poll) - return -EIO; + P9_DPRINTK(P9_DEBUG_TRANS, "mux %p prev %p next %p\n", m, + m->mux_list.prev, m->mux_list.next); - ret = ts->rd->f_op->poll(ts->rd, pt); - if (ret < 0) - return ret; + p9_mux_poll_stop(m); + cancel_work_sync(&m->rq); + cancel_work_sync(&m->wq); - if (ts->rd != ts->wr) { - n = ts->wr->f_op->poll(ts->wr, pt); - if (n < 0) - return n; - ret = (ret & ~POLLOUT) | (n & ~POLLIN); - } + p9_conn_cancel(m, -ECONNRESET); - return ret; + m->client = NULL; + kfree(m); } /** - * p9_fd_close - shutdown socket - * @trans: private socket structure + * p9_fd_close - shutdown file descriptor transport + * @client: client instance * */ -static void p9_fd_close(struct p9_trans *trans) +static void p9_fd_close(struct p9_client *client) { struct p9_trans_fd *ts; - if (!trans) + if (!client) return; - ts = xchg(&trans->priv, NULL); - + ts = client->trans; if (!ts) return; + client->status = Disconnected; + p9_conn_destroy(ts->conn); - trans->status = Disconnected; if (ts->rd) fput(ts->rd); if (ts->wr) fput(ts->wr); + kfree(ts); } @@ -1402,31 +872,23 @@ static inline int valid_ipaddr4(const char *buf) return 0; } -static struct p9_trans * -p9_trans_create_tcp(const char *addr, char *args, int msize, unsigned char dotu) +static int +p9_fd_create_tcp(struct p9_client *client, const char *addr, char *args) { int err; - struct p9_trans *trans; struct socket *csocket; struct sockaddr_in sin_server; struct p9_fd_opts opts; - struct p9_trans_fd *p; + struct p9_trans_fd *p = NULL; /* this gets allocated in p9_fd_open */ err = parse_opts(args, &opts); if (err < 0) - return ERR_PTR(err); + return err; if (valid_ipaddr4(addr) < 0) - return ERR_PTR(-EINVAL); + return -EINVAL; csocket = NULL; - trans = kmalloc(sizeof(struct p9_trans), GFP_KERNEL); - if (!trans) - return ERR_PTR(-ENOMEM); - trans->msize = msize; - trans->extended = dotu; - trans->rpc = p9_fd_rpc; - trans->close = p9_fd_close; sin_server.sin_family = AF_INET; sin_server.sin_addr.s_addr = in_aton(addr); @@ -1449,45 +911,38 @@ p9_trans_create_tcp(const char *addr, char *args, int msize, unsigned char dotu) goto error; } - err = p9_socket_open(trans, csocket); + err = p9_socket_open(client, csocket); if (err < 0) goto error; - p = (struct p9_trans_fd *) trans->priv; - p->conn = p9_conn_create(trans); + 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; goto error; } - return trans; + return 0; error: if (csocket) sock_release(csocket); - kfree(trans); - return ERR_PTR(err); + kfree(p); + + return err; } -static struct p9_trans * -p9_trans_create_unix(const char *addr, char *args, int msize, - unsigned char dotu) +static int +p9_fd_create_unix(struct p9_client *client, const char *addr, char *args) { int err; struct socket *csocket; struct sockaddr_un sun_server; - struct p9_trans *trans; - struct p9_trans_fd *p; + struct p9_trans_fd *p = NULL; /* this gets allocated in p9_fd_open */ csocket = NULL; - trans = kmalloc(sizeof(struct p9_trans), GFP_KERNEL); - if (!trans) - return ERR_PTR(-ENOMEM); - - trans->rpc = p9_fd_rpc; - trans->close = p9_fd_close; if (strlen(addr) > UNIX_PATH_MAX) { P9_EPRINTK(KERN_ERR, "p9_trans_unix: address too long: %s\n", @@ -1508,79 +963,69 @@ p9_trans_create_unix(const char *addr, char *args, int msize, goto error; } - err = p9_socket_open(trans, csocket); + err = p9_socket_open(client, csocket); if (err < 0) goto error; - trans->msize = msize; - trans->extended = dotu; - p = (struct p9_trans_fd *) trans->priv; - p->conn = p9_conn_create(trans); + 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; goto error; } - return trans; + return 0; error: if (csocket) sock_release(csocket); - kfree(trans); - return ERR_PTR(err); + kfree(p); + return err; } -static struct p9_trans * -p9_trans_create_fd(const char *name, char *args, int msize, - unsigned char extended) +static int +p9_fd_create(struct p9_client *client, const char *addr, char *args) { int err; - struct p9_trans *trans; struct p9_fd_opts opts; - struct p9_trans_fd *p; + struct p9_trans_fd *p = NULL; /* this get allocated in p9_fd_open */ parse_opts(args, &opts); if (opts.rfd == ~0 || opts.wfd == ~0) { printk(KERN_ERR "v9fs: Insufficient options for proto=fd\n"); - return ERR_PTR(-ENOPROTOOPT); + return -ENOPROTOOPT; } - trans = kmalloc(sizeof(struct p9_trans), GFP_KERNEL); - if (!trans) - return ERR_PTR(-ENOMEM); - - trans->rpc = p9_fd_rpc; - trans->close = p9_fd_close; - - err = p9_fd_open(trans, opts.rfd, opts.wfd); + err = p9_fd_open(client, opts.rfd, opts.wfd); if (err < 0) goto error; - trans->msize = msize; - trans->extended = extended; - p = (struct p9_trans_fd *) trans->priv; - p->conn = p9_conn_create(trans); + 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; goto error; } - return trans; + return 0; error: - kfree(trans); - return ERR_PTR(err); + kfree(p); + return err; } static struct p9_trans_module p9_tcp_trans = { .name = "tcp", .maxsize = MAX_SOCK_BUF, .def = 1, - .create = p9_trans_create_tcp, + .create = p9_fd_create_tcp, + .close = p9_fd_close, + .request = p9_fd_request, + .cancel = p9_fd_cancel, .owner = THIS_MODULE, }; @@ -1588,7 +1033,10 @@ static struct p9_trans_module p9_unix_trans = { .name = "unix", .maxsize = MAX_SOCK_BUF, .def = 0, - .create = p9_trans_create_unix, + .create = p9_fd_create_unix, + .close = p9_fd_close, + .request = p9_fd_request, + .cancel = p9_fd_cancel, .owner = THIS_MODULE, }; @@ -1596,23 +1044,71 @@ static struct p9_trans_module p9_fd_trans = { .name = "fd", .maxsize = MAX_SOCK_BUF, .def = 0, - .create = p9_trans_create_fd, + .create = p9_fd_create, + .close = p9_fd_close, + .request = p9_fd_request, + .cancel = p9_fd_cancel, .owner = THIS_MODULE, }; -int p9_trans_fd_init(void) +/** + * p9_poll_proc - poll worker thread + * @a: thread state and arguments + * + * polls all v9fs transports for new events and queues the appropriate + * work to the work queue + * + */ + +static int p9_poll_proc(void *a) { - int i; + unsigned long flags; + + P9_DPRINTK(P9_DEBUG_TRANS, "start %p\n", current); + repeat: + spin_lock_irqsave(&p9_poll_lock, flags); + while (!list_empty(&p9_poll_pending_list)) { + struct p9_conn *conn = list_first_entry(&p9_poll_pending_list, + struct p9_conn, + poll_pending_link); + list_del_init(&conn->poll_pending_link); + spin_unlock_irqrestore(&p9_poll_lock, flags); + + p9_poll_mux(conn); + + spin_lock_irqsave(&p9_poll_lock, flags); + } + spin_unlock_irqrestore(&p9_poll_lock, flags); + + 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; - for (i = 0; i < ARRAY_SIZE(p9_mux_poll_tasks); i++) - p9_mux_poll_tasks[i].task = NULL; + P9_DPRINTK(P9_DEBUG_TRANS, "finish\n"); + return 0; +} +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); @@ -1622,6 +1118,7 @@ int p9_trans_fd_init(void) void p9_trans_fd_exit(void) { + kthread_stop(p9_poll_task); v9fs_unregister_trans(&p9_tcp_trans); v9fs_unregister_trans(&p9_unix_trans); v9fs_unregister_trans(&p9_fd_trans); diff --git a/net/9p/trans_virtio.c b/net/9p/trans_virtio.c index 94912e077a55..2d7781ec663b 100644 --- a/net/9p/trans_virtio.c +++ b/net/9p/trans_virtio.c @@ -1,12 +1,10 @@ /* - * The Guest 9p transport driver + * The Virtio 9p transport driver * * This is a block based transport driver based on the lguest block driver * code. * - */ -/* - * Copyright (C) 2007 Eric Van Hensbergen, IBM Corporation + * Copyright (C) 2007, 2008 Eric Van Hensbergen, IBM Corporation * * Based on virtio console driver * Copyright (C) 2006, 2007 Rusty Russell, IBM Corporation @@ -41,6 +39,7 @@ #include <linux/file.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/virtio.h> @@ -53,50 +52,6 @@ static DEFINE_MUTEX(virtio_9p_lock); /* global which tracks highest initialized channel */ static int chan_index; -#define P9_INIT_MAXTAG 16 - - -/** - * enum p9_req_status_t - virtio request status - * @REQ_STATUS_IDLE: request slot unused - * @REQ_STATUS_SENT: request sent to server - * @REQ_STATUS_RCVD: response received from server - * @REQ_STATUS_FLSH: request has been flushed - * - * The @REQ_STATUS_IDLE state is used to mark a request slot as unused - * but use is actually tracked by the idpool structure which handles tag - * id allocation. - * - */ - -enum p9_req_status_t { - REQ_STATUS_IDLE, - REQ_STATUS_SENT, - REQ_STATUS_RCVD, - REQ_STATUS_FLSH, -}; - -/** - * struct p9_req_t - virtio request slots - * @status: status of this request slot - * @wq: wait_queue for the client to block on for this request - * - * The virtio transport uses an array to track outstanding requests - * instead of a list. While this may incurr overhead during initial - * allocation or expansion, it makes request lookup much easier as the - * tag id is a index into an array. (We use tag+1 so that we can accomodate - * the -1 tag for the T_VERSION request). - * This also has the nice effect of only having to allocate wait_queues - * once, instead of constantly allocating and freeing them. Its possible - * other resources could benefit from this scheme as well. - * - */ - -struct p9_req_t { - int status; - wait_queue_head_t *wq; -}; - /** * struct virtio_chan - per-instance transport information * @initialized: whether the channel is initialized @@ -121,67 +76,14 @@ static struct virtio_chan { spinlock_t lock; + struct p9_client *client; struct virtio_device *vdev; struct virtqueue *vq; - struct p9_idpool *tagpool; - struct p9_req_t *reqs; - int max_tag; - /* Scatterlist: can be too big for stack. */ struct scatterlist sg[VIRTQUEUE_NUM]; } channels[MAX_9P_CHAN]; -/** - * p9_lookup_tag - Lookup requests by tag - * @c: virtio channel 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 - * ids which did not previously have a slot. - * - * Bugs: there is currently no upper limit on request slots set - * here, but that should be constrained by the id accounting. - */ - -static struct p9_req_t *p9_lookup_tag(struct virtio_chan *c, u16 tag) -{ - /* This looks up the original request by tag so we know which - * buffer to read the data into */ - tag++; - - while (tag >= c->max_tag) { - int old_max = c->max_tag; - int count; - - if (c->max_tag) - c->max_tag *= 2; - else - c->max_tag = P9_INIT_MAXTAG; - - c->reqs = krealloc(c->reqs, sizeof(struct p9_req_t)*c->max_tag, - GFP_ATOMIC); - if (!c->reqs) { - printk(KERN_ERR "Couldn't grow tag array\n"); - BUG(); - } - for (count = old_max; count < c->max_tag; count++) { - c->reqs[count].status = REQ_STATUS_IDLE; - c->reqs[count].wq = kmalloc(sizeof(wait_queue_head_t), - GFP_ATOMIC); - if (!c->reqs[count].wq) { - printk(KERN_ERR "Couldn't grow tag array\n"); - BUG(); - } - init_waitqueue_head(c->reqs[count].wq); - } - } - - return &c->reqs[tag]; -} - - /* How many bytes left in this page. */ static unsigned int rest_of_page(void *data) { @@ -197,25 +99,13 @@ static unsigned int rest_of_page(void *data) * */ -static void p9_virtio_close(struct p9_trans *trans) +static void p9_virtio_close(struct p9_client *client) { - struct virtio_chan *chan = trans->priv; - int count; - unsigned long flags; - - spin_lock_irqsave(&chan->lock, flags); - p9_idpool_destroy(chan->tagpool); - for (count = 0; count < chan->max_tag; count++) - kfree(chan->reqs[count].wq); - kfree(chan->reqs); - chan->max_tag = 0; - spin_unlock_irqrestore(&chan->lock, flags); + struct virtio_chan *chan = client->trans; mutex_lock(&virtio_9p_lock); chan->inuse = false; mutex_unlock(&virtio_9p_lock); - - kfree(trans); } /** @@ -236,17 +126,16 @@ static void req_done(struct virtqueue *vq) struct virtio_chan *chan = vq->vdev->priv; struct p9_fcall *rc; unsigned int len; - unsigned long flags; struct p9_req_t *req; - spin_lock_irqsave(&chan->lock, flags); + P9_DPRINTK(P9_DEBUG_TRANS, ": request done\n"); + while ((rc = chan->vq->vq_ops->get_buf(chan->vq, &len)) != NULL) { - req = p9_lookup_tag(chan, rc->tag); - req->status = REQ_STATUS_RCVD; - wake_up(req->wq); + 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); + p9_client_cb(chan->client, req); } - /* In case queue is stopped waiting for more buffers. */ - spin_unlock_irqrestore(&chan->lock, flags); } /** @@ -283,8 +172,14 @@ pack_sg_list(struct scatterlist *sg, int start, int limit, char *data, return index-start; } +/* We don't currently allow canceling of virtio requests */ +static int p9_virtio_cancel(struct p9_client *client, struct p9_req_t *req) +{ + return 1; +} + /** - * p9_virtio_rpc - issue a request and wait for a response + * p9_virtio_request - issue a request * @t: transport state * @tc: &p9_fcall request to transmit * @rc: &p9_fcall to put reponse into @@ -292,44 +187,22 @@ pack_sg_list(struct scatterlist *sg, int start, int limit, char *data, */ static int -p9_virtio_rpc(struct p9_trans *t, struct p9_fcall *tc, struct p9_fcall **rc) +p9_virtio_request(struct p9_client *client, struct p9_req_t *req) { int in, out; - int n, err, size; - struct virtio_chan *chan = t->priv; - char *rdata; - struct p9_req_t *req; - unsigned long flags; - - if (*rc == NULL) { - *rc = kmalloc(sizeof(struct p9_fcall) + t->msize, GFP_KERNEL); - if (!*rc) - return -ENOMEM; - } - - rdata = (char *)*rc+sizeof(struct p9_fcall); - - n = P9_NOTAG; - if (tc->id != P9_TVERSION) { - n = p9_idpool_get(chan->tagpool); - if (n < 0) - return -ENOMEM; - } - - spin_lock_irqsave(&chan->lock, flags); - req = p9_lookup_tag(chan, n); - spin_unlock_irqrestore(&chan->lock, flags); + struct virtio_chan *chan = client->trans; + char *rdata = (char *)req->rc+sizeof(struct p9_fcall); - p9_set_tag(tc, n); + P9_DPRINTK(P9_DEBUG_TRANS, "9p debug: virtio request\n"); - P9_DPRINTK(P9_DEBUG_TRANS, "9p debug: virtio rpc tag %d\n", n); - - out = pack_sg_list(chan->sg, 0, VIRTQUEUE_NUM, tc->sdata, tc->size); - in = pack_sg_list(chan->sg, out, VIRTQUEUE_NUM-out, rdata, t->msize); + 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); req->status = REQ_STATUS_SENT; - if (chan->vq->vq_ops->add_buf(chan->vq, chan->sg, out, in, tc)) { + if (chan->vq->vq_ops->add_buf(chan->vq, chan->sg, out, in, req->tc)) { P9_DPRINTK(P9_DEBUG_TRANS, "9p debug: virtio rpc add_buf returned failure"); return -EIO; @@ -337,31 +210,7 @@ p9_virtio_rpc(struct p9_trans *t, struct p9_fcall *tc, struct p9_fcall **rc) chan->vq->vq_ops->kick(chan->vq); - wait_event(*req->wq, req->status == REQ_STATUS_RCVD); - - size = le32_to_cpu(*(__le32 *) rdata); - - err = p9_deserialize_fcall(rdata, size, *rc, t->extended); - if (err < 0) { - P9_DPRINTK(P9_DEBUG_TRANS, - "9p debug: virtio rpc deserialize returned %d\n", err); - return err; - } - -#ifdef CONFIG_NET_9P_DEBUG - if ((p9_debug_level&P9_DEBUG_FCALL) == P9_DEBUG_FCALL) { - char buf[150]; - - p9_printfcall(buf, sizeof(buf), *rc, t->extended); - printk(KERN_NOTICE ">>> %p %s\n", t, buf); - } -#endif - - if (n != P9_NOTAG && p9_idpool_check(n, chan->tagpool)) - p9_idpool_put(n, chan->tagpool); - - req->status = REQ_STATUS_IDLE; - + P9_DPRINTK(P9_DEBUG_TRANS, "9p debug: virtio request kicked\n"); return 0; } @@ -422,10 +271,9 @@ fail: /** * p9_virtio_create - allocate a new virtio channel + * @client: client instance invoking this transport * @devname: string identifying the channel to connect to (unused) * @args: args passed from sys_mount() for per-transport options (unused) - * @msize: requested maximum packet size - * @extended: 9p2000.u enabled flag * * This sets up a transport channel for 9p communication. Right now * we only match the first available channel, but eventually we couldlook up @@ -441,11 +289,9 @@ fail: * */ -static struct p9_trans * -p9_virtio_create(const char *devname, char *args, int msize, - unsigned char extended) +static int +p9_virtio_create(struct p9_client *client, const char *devname, char *args) { - struct p9_trans *trans; struct virtio_chan *chan = channels; int index = 0; @@ -463,30 +309,13 @@ p9_virtio_create(const char *devname, char *args, int msize, if (index >= MAX_9P_CHAN) { printk(KERN_ERR "9p: no channels available\n"); - return ERR_PTR(-ENODEV); + return -ENODEV; } - chan->tagpool = p9_idpool_create(); - if (IS_ERR(chan->tagpool)) { - printk(KERN_ERR "9p: couldn't allocate tagpool\n"); - return ERR_PTR(-ENOMEM); - } - p9_idpool_get(chan->tagpool); /* reserve tag 0 */ - chan->max_tag = 0; - chan->reqs = NULL; - - trans = kmalloc(sizeof(struct p9_trans), GFP_KERNEL); - if (!trans) { - printk(KERN_ERR "9p: couldn't allocate transport\n"); - return ERR_PTR(-ENOMEM); - } - trans->extended = extended; - trans->msize = msize; - trans->close = p9_virtio_close; - trans->rpc = p9_virtio_rpc; - trans->priv = chan; + client->trans = (void *)chan; + chan->client = client; - return trans; + return 0; } /** @@ -526,6 +355,9 @@ static struct virtio_driver p9_virtio_drv = { static struct p9_trans_module p9_virtio_trans = { .name = "virtio", .create = p9_virtio_create, + .close = p9_virtio_close, + .request = p9_virtio_request, + .cancel = p9_virtio_cancel, .maxsize = PAGE_SIZE*16, .def = 0, .owner = THIS_MODULE, diff --git a/net/9p/util.c b/net/9p/util.c index 958fc58cd1ff..dc4ec05ad93d 100644 --- a/net/9p/util.c +++ b/net/9p/util.c @@ -105,6 +105,7 @@ retry: else if (error) return -1; + P9_DPRINTK(P9_DEBUG_MUX, " id %d pool %p\n", i, p); return i; } EXPORT_SYMBOL(p9_idpool_get); @@ -121,6 +122,9 @@ EXPORT_SYMBOL(p9_idpool_get); 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); + spin_lock_irqsave(&p->lock, flags); idr_remove(&p->pool, id); spin_unlock_irqrestore(&p->lock, flags); diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c index 0c850427a85b..d3134e7e6ee8 100644 --- a/net/appletalk/ddp.c +++ b/net/appletalk/ddp.c @@ -2,7 +2,7 @@ * DDP: An implementation of the AppleTalk DDP protocol for * Ethernet 'ELAP'. * - * Alan Cox <Alan.Cox@linux.org> + * Alan Cox <alan@lxorguk.ukuu.org.uk> * * With more than a little assistance from * @@ -1934,6 +1934,6 @@ static void __exit atalk_exit(void) module_exit(atalk_exit); MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Alan Cox <Alan.Cox@linux.org>"); +MODULE_AUTHOR("Alan Cox <alan@lxorguk.ukuu.org.uk>"); MODULE_DESCRIPTION("AppleTalk 0.20\n"); MODULE_ALIAS_NETPROTO(PF_APPLETALK); diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c index f6348e078aa4..8f9431a12c6f 100644 --- a/net/bluetooth/af_bluetooth.c +++ b/net/bluetooth/af_bluetooth.c @@ -37,10 +37,7 @@ #include <linux/poll.h> #include <net/sock.h> #include <asm/ioctls.h> - -#if defined(CONFIG_KMOD) #include <linux/kmod.h> -#endif #include <net/bluetooth/bluetooth.h> @@ -145,11 +142,8 @@ static int bt_sock_create(struct net *net, struct socket *sock, int proto) if (proto < 0 || proto >= BT_MAX_PROTO) return -EINVAL; -#if defined(CONFIG_KMOD) - if (!bt_proto[proto]) { + if (!bt_proto[proto]) request_module("bt-proto-%d", proto); - } -#endif err = -EPROTONOSUPPORT; diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c index 96434d774c84..acdeab3d9807 100644 --- a/net/bluetooth/hidp/core.c +++ b/net/bluetooth/hidp/core.c @@ -578,7 +578,7 @@ static int hidp_session(void *arg) if (session->hid) { if (session->hid->claimed & HID_CLAIMED_INPUT) hidinput_disconnect(session->hid); - hid_free_device(session->hid); + hid_destroy_device(session->hid); } /* Wakeup user-space polling for socket errors */ @@ -623,9 +623,15 @@ static struct device *hidp_get_device(struct hidp_session *session) static int hidp_setup_input(struct hidp_session *session, struct hidp_connadd_req *req) { - struct input_dev *input = session->input; + struct input_dev *input; int i; + input = input_allocate_device(); + if (!input) + return -ENOMEM; + + session->input = input; + input_set_drvdata(input, session); input->name = "Bluetooth HID Boot Protocol Device"; @@ -677,67 +683,114 @@ static void hidp_close(struct hid_device *hid) { } -static const struct { - __u16 idVendor; - __u16 idProduct; - unsigned quirks; -} hidp_blacklist[] = { - /* Apple wireless Mighty Mouse */ - { 0x05ac, 0x030c, HID_QUIRK_MIGHTYMOUSE | HID_QUIRK_INVERT_HWHEEL }, +static int hidp_parse(struct hid_device *hid) +{ + struct hidp_session *session = hid->driver_data; + struct hidp_connadd_req *req = session->req; + unsigned char *buf; + int ret; - { } /* Terminating entry */ -}; + buf = kmalloc(req->rd_size, GFP_KERNEL); + if (!buf) + return -ENOMEM; + + if (copy_from_user(buf, req->rd_data, req->rd_size)) { + kfree(buf); + return -EFAULT; + } + + ret = hid_parse_report(session->hid, buf, req->rd_size); + + kfree(buf); + + if (ret) + return ret; + + session->req = NULL; + + return 0; +} + +static int hidp_start(struct hid_device *hid) +{ + struct hidp_session *session = hid->driver_data; + struct hid_report *report; -static void hidp_setup_quirks(struct hid_device *hid) + list_for_each_entry(report, &hid->report_enum[HID_INPUT_REPORT]. + report_list, list) + hidp_send_report(session, report); + + list_for_each_entry(report, &hid->report_enum[HID_FEATURE_REPORT]. + report_list, list) + hidp_send_report(session, report); + + return 0; +} + +static void hidp_stop(struct hid_device *hid) { - unsigned int n; + struct hidp_session *session = hid->driver_data; + + skb_queue_purge(&session->ctrl_transmit); + skb_queue_purge(&session->intr_transmit); - for (n = 0; hidp_blacklist[n].idVendor; n++) - if (hidp_blacklist[n].idVendor == le16_to_cpu(hid->vendor) && - hidp_blacklist[n].idProduct == le16_to_cpu(hid->product)) - hid->quirks = hidp_blacklist[n].quirks; + if (hid->claimed & HID_CLAIMED_INPUT) + hidinput_disconnect(hid); + hid->claimed = 0; } -static void hidp_setup_hid(struct hidp_session *session, +static struct hid_ll_driver hidp_hid_driver = { + .parse = hidp_parse, + .start = hidp_start, + .stop = hidp_stop, + .open = hidp_open, + .close = hidp_close, + .hidinput_input_event = hidp_hidinput_event, +}; + +static int hidp_setup_hid(struct hidp_session *session, struct hidp_connadd_req *req) { - struct hid_device *hid = session->hid; - struct hid_report *report; + struct hid_device *hid; bdaddr_t src, dst; + int ret; - baswap(&src, &bt_sk(session->ctrl_sock->sk)->src); - baswap(&dst, &bt_sk(session->ctrl_sock->sk)->dst); + hid = hid_allocate_device(); + if (IS_ERR(hid)) { + ret = PTR_ERR(session->hid); + goto err; + } + session->hid = hid; + session->req = req; hid->driver_data = session; - hid->country = req->country; + baswap(&src, &bt_sk(session->ctrl_sock->sk)->src); + baswap(&dst, &bt_sk(session->ctrl_sock->sk)->dst); hid->bus = BUS_BLUETOOTH; hid->vendor = req->vendor; hid->product = req->product; hid->version = req->version; + hid->country = req->country; strncpy(hid->name, req->name, 128); strncpy(hid->phys, batostr(&src), 64); strncpy(hid->uniq, batostr(&dst), 64); - hid->dev = hidp_get_device(session); - - hid->hid_open = hidp_open; - hid->hid_close = hidp_close; - - hid->hidinput_input_event = hidp_hidinput_event; + hid->dev.parent = hidp_get_device(session); + hid->ll_driver = &hidp_hid_driver; - hidp_setup_quirks(hid); + ret = hid_add_device(hid); + if (ret) + goto err_hid; - list_for_each_entry(report, &hid->report_enum[HID_INPUT_REPORT].report_list, list) - hidp_send_report(session, report); - - list_for_each_entry(report, &hid->report_enum[HID_FEATURE_REPORT].report_list, list) - hidp_send_report(session, report); - - if (hidinput_connect(hid) == 0) - hid->claimed |= HID_CLAIMED_INPUT; + return 0; +err_hid: + hid_destroy_device(hid); + session->hid = NULL; +err: + return ret; } int hidp_add_connection(struct hidp_connadd_req *req, struct socket *ctrl_sock, struct socket *intr_sock) @@ -757,38 +810,6 @@ int hidp_add_connection(struct hidp_connadd_req *req, struct socket *ctrl_sock, BT_DBG("rd_data %p rd_size %d", req->rd_data, req->rd_size); - if (req->rd_size > 0) { - unsigned char *buf = kmalloc(req->rd_size, GFP_KERNEL); - - if (!buf) { - kfree(session); - return -ENOMEM; - } - - if (copy_from_user(buf, req->rd_data, req->rd_size)) { - kfree(buf); - kfree(session); - return -EFAULT; - } - - session->hid = hid_parse_report(buf, req->rd_size); - - kfree(buf); - - if (!session->hid) { - kfree(session); - return -EINVAL; - } - } - - if (!session->hid) { - session->input = input_allocate_device(); - if (!session->input) { - kfree(session); - return -ENOMEM; - } - } - down_write(&hidp_session_sem); s = __hidp_get_session(&bt_sk(ctrl_sock->sk)->dst); @@ -816,15 +837,18 @@ int hidp_add_connection(struct hidp_connadd_req *req, struct socket *ctrl_sock, session->flags = req->flags & (1 << HIDP_BLUETOOTH_VENDOR_ID); session->idle_to = req->idle_to; - if (session->input) { + if (req->rd_size > 0) { + err = hidp_setup_hid(session, req); + if (err && err != -ENODEV) + goto err_skb; + } + + if (!session->hid) { err = hidp_setup_input(session, req); if (err < 0) - goto failed; + goto err_skb; } - if (session->hid) - hidp_setup_hid(session, req); - __hidp_link_session(session); hidp_set_timer(session); @@ -850,17 +874,16 @@ unlink: __hidp_unlink_session(session); - if (session->input) { + if (session->input) input_unregister_device(session->input); - session->input = NULL; /* don't try to free it here */ - } - + if (session->hid) + hid_destroy_device(session->hid); +err_skb: + skb_queue_purge(&session->ctrl_transmit); + skb_queue_purge(&session->intr_transmit); failed: up_write(&hidp_session_sem); - if (session->hid) - hid_free_device(session->hid); - input_free_device(session->input); kfree(session); return err; @@ -950,18 +973,43 @@ int hidp_get_conninfo(struct hidp_conninfo *ci) return err; } +static const struct hid_device_id hidp_table[] = { + { HID_BLUETOOTH_DEVICE(HID_ANY_ID, HID_ANY_ID) }, + { } +}; + +static struct hid_driver hidp_driver = { + .name = "generic-bluetooth", + .id_table = hidp_table, +}; + static int __init hidp_init(void) { + int ret; + l2cap_load(); BT_INFO("HIDP (Human Interface Emulation) ver %s", VERSION); - return hidp_init_sockets(); + ret = hid_register_driver(&hidp_driver); + if (ret) + goto err; + + ret = hidp_init_sockets(); + if (ret) + goto err_drv; + + return 0; +err_drv: + hid_unregister_driver(&hidp_driver); +err: + return ret; } static void __exit hidp_exit(void) { hidp_cleanup_sockets(); + hid_unregister_driver(&hidp_driver); } module_init(hidp_init); diff --git a/net/bluetooth/hidp/hidp.h b/net/bluetooth/hidp/hidp.h index 343fb0566b3e..e503c89057ad 100644 --- a/net/bluetooth/hidp/hidp.h +++ b/net/bluetooth/hidp/hidp.h @@ -151,6 +151,8 @@ struct hidp_session { struct sk_buff_head ctrl_transmit; struct sk_buff_head intr_transmit; + + struct hidp_connadd_req *req; }; static inline void hidp_schedule(struct hidp_session *session) diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index a4abed5b4c44..fa5cda4e552a 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -719,7 +719,7 @@ static unsigned int br_nf_forward_arp(unsigned int hook, struct sk_buff *skb, return NF_ACCEPT; } *d = (struct net_device *)in; - NF_HOOK(NF_ARP, NF_ARP_FORWARD, skb, (struct net_device *)in, + NF_HOOK(NFPROTO_ARP, NF_ARP_FORWARD, skb, (struct net_device *)in, (struct net_device *)out, br_nf_forward_finish); return NF_STOLEN; diff --git a/net/bridge/netfilter/Kconfig b/net/bridge/netfilter/Kconfig index 366d3e9d51f8..ba6f73eb06c6 100644 --- a/net/bridge/netfilter/Kconfig +++ b/net/bridge/netfilter/Kconfig @@ -4,6 +4,7 @@ menuconfig BRIDGE_NF_EBTABLES tristate "Ethernet Bridge tables (ebtables) support" + depends on BRIDGE && BRIDGE_NETFILTER select NETFILTER_XTABLES help ebtables is a general, extensible frame/packet identification diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index 5bb88eb0aad4..0fa208e86405 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -305,23 +305,14 @@ find_inlist_lock_noload(struct list_head *head, const char *name, int *error, return NULL; } -#ifndef CONFIG_KMOD -#define find_inlist_lock(h,n,p,e,m) find_inlist_lock_noload((h),(n),(e),(m)) -#else static void * find_inlist_lock(struct list_head *head, const char *name, const char *prefix, int *error, struct mutex *mutex) { - void *ret; - - ret = find_inlist_lock_noload(head, name, error, mutex); - if (!ret) { - request_module("%s%s", prefix, name); - ret = find_inlist_lock_noload(head, name, error, mutex); - } - return ret; + return try_then_request_module( + find_inlist_lock_noload(head, name, error, mutex), + "%s%s", prefix, name); } -#endif static inline struct ebt_table * find_table_lock(const char *name, int *error, struct mutex *mutex) diff --git a/net/can/af_can.c b/net/can/af_can.c index 8035fbf526ae..7d4d2b3c137e 100644 --- a/net/can/af_can.c +++ b/net/can/af_can.c @@ -128,8 +128,8 @@ static int can_create(struct net *net, struct socket *sock, int protocol) if (net != &init_net) return -EAFNOSUPPORT; -#ifdef CONFIG_KMOD - /* try to load protocol module, when CONFIG_KMOD is defined */ +#ifdef CONFIG_MODULES + /* try to load protocol module kernel is modular */ if (!proto_tab[protocol]) { err = request_module("can-proto-%d", protocol); diff --git a/net/core/datagram.c b/net/core/datagram.c index 52f577a0f544..ee631843c2f5 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -9,7 +9,7 @@ * identical recvmsg() code. So we share it here. The poll was * shared before but buried in udp.c so I moved it. * - * Authors: Alan Cox <alan@redhat.com>. (datagram_poll() from old + * Authors: Alan Cox <alan@lxorguk.ukuu.org.uk>. (datagram_poll() from old * udp.c code) * * Fixes: diff --git a/net/core/dev.c b/net/core/dev.c index 1408a083fe4e..b8a4fd0806af 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -924,10 +924,10 @@ int dev_change_name(struct net_device *dev, const char *newname) strlcpy(dev->name, newname, IFNAMSIZ); rollback: - err = device_rename(&dev->dev, dev->name); - if (err) { + ret = device_rename(&dev->dev, dev->name); + if (ret) { memcpy(dev->name, oldname, IFNAMSIZ); - return err; + return ret; } write_lock_bh(&dev_base_lock); @@ -4956,8 +4956,6 @@ EXPORT_SYMBOL(br_fdb_get_hook); EXPORT_SYMBOL(br_fdb_put_hook); #endif -#ifdef CONFIG_KMOD EXPORT_SYMBOL(dev_load); -#endif EXPORT_PER_CPU_SYMBOL(softnet_data); diff --git a/net/core/dev_mcast.c b/net/core/dev_mcast.c index 5402b3b38e0d..9e2fa39f22a3 100644 --- a/net/core/dev_mcast.c +++ b/net/core/dev_mcast.c @@ -6,7 +6,7 @@ * Richard Underwood <richard@wuzz.demon.co.uk> * * Stir fried together from the IP multicast and CAP patches above - * Alan Cox <Alan.Cox@linux.org> + * Alan Cox <alan@lxorguk.ukuu.org.uk> * * Fixes: * Alan Cox : Update the device on a real delete diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index b0dc818a91d7..f1d07b5c1e17 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -96,7 +96,7 @@ static void net_free(struct net *net) return; } #endif - + kfree(net->gen); kmem_cache_free(net_cachep, net); } diff --git a/net/core/pktgen.c b/net/core/pktgen.c index a756847e3814..99f656d35b4f 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -2474,7 +2474,7 @@ static inline int process_ipsec(struct pktgen_dev *pkt_dev, if (ret < 0) { printk(KERN_ERR "Error expanding " "ipsec packet %d\n",ret); - return 0; + goto err; } } @@ -2484,8 +2484,7 @@ static inline int process_ipsec(struct pktgen_dev *pkt_dev, if (ret) { printk(KERN_ERR "Error creating ipsec " "packet %d\n",ret); - kfree_skb(skb); - return 0; + goto err; } /* restore ll */ eth = (__u8 *) skb_push(skb, ETH_HLEN); @@ -2494,6 +2493,9 @@ static inline int process_ipsec(struct pktgen_dev *pkt_dev, } } return 1; +err: + kfree_skb(skb); + return 0; } #endif diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 3630131fa1fa..31f29d2989fd 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1040,7 +1040,7 @@ static int rtnl_newlink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) struct nlattr *linkinfo[IFLA_INFO_MAX+1]; int err; -#ifdef CONFIG_KMOD +#ifdef CONFIG_MODULES replay: #endif err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy); @@ -1129,7 +1129,7 @@ replay: return -EOPNOTSUPP; if (!ops) { -#ifdef CONFIG_KMOD +#ifdef CONFIG_MODULES if (kind[0]) { __rtnl_unlock(); request_module("rtnl-link-%s", kind); diff --git a/net/core/skb_dma_map.c b/net/core/skb_dma_map.c index 1f49afcd8e86..86234923a3b7 100644 --- a/net/core/skb_dma_map.c +++ b/net/core/skb_dma_map.c @@ -35,7 +35,7 @@ int skb_dma_map(struct device *dev, struct sk_buff *skb, return 0; unwind: - while (i-- >= 0) { + while (--i >= 0) { skb_frag_t *fp = &sp->frags[i]; dma_unmap_page(dev, sp->dma_maps[i + 1], diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 7f7bb1a636d9..4e22e3a35359 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -1,7 +1,7 @@ /* * Routines having to do with the 'struct sk_buff' memory handlers. * - * Authors: Alan Cox <iiitac@pyr.swan.ac.uk> + * Authors: Alan Cox <alan@lxorguk.ukuu.org.uk> * Florian La Roche <rzsfl@rz.uni-sb.de> * * Fixes: diff --git a/net/core/stream.c b/net/core/stream.c index a6b3437ff082..8727cead64ad 100644 --- a/net/core/stream.c +++ b/net/core/stream.c @@ -9,7 +9,7 @@ * * Authors: Arnaldo Carvalho de Melo <acme@conectiva.com.br> * (from old tcp.c code) - * Alan Cox <alan@redhat.com> (Borrowed comments 8-)) + * Alan Cox <alan@lxorguk.ukuu.org.uk> (Borrowed comments 8-)) */ #include <linux/module.h> diff --git a/net/dccp/ccid.c b/net/dccp/ccid.c index 4809753d12ae..8fe931a3d7a1 100644 --- a/net/dccp/ccid.c +++ b/net/dccp/ccid.c @@ -154,7 +154,7 @@ struct ccid *ccid_new(unsigned char id, struct sock *sk, int rx, gfp_t gfp) struct ccid *ccid = NULL; ccids_read_lock(); -#ifdef CONFIG_KMOD +#ifdef CONFIG_MODULES if (ccids[id] == NULL) { /* We only try to load if in process context */ ccids_read_unlock(); diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c index 11062780bb02..d4ce1224e008 100644 --- a/net/dccp/ipv6.c +++ b/net/dccp/ipv6.c @@ -259,7 +259,7 @@ static int dccp_v6_send_response(struct sock *sk, struct request_sock *req) fl.fl6_flowlabel = 0; fl.oif = ireq6->iif; fl.fl_ip_dport = inet_rsk(req)->rmt_port; - fl.fl_ip_sport = inet_sk(sk)->sport; + fl.fl_ip_sport = inet_rsk(req)->loc_port; security_req_classify_flow(req, &fl); opt = np->opt; @@ -558,7 +558,7 @@ static struct sock *dccp_v6_request_recv_sock(struct sock *sk, ipv6_addr_copy(&fl.fl6_src, &ireq6->loc_addr); fl.oif = sk->sk_bound_dev_if; fl.fl_ip_dport = inet_rsk(req)->rmt_port; - fl.fl_ip_sport = inet_sk(sk)->sport; + fl.fl_ip_sport = inet_rsk(req)->loc_port; security_sk_classify_flow(sk, &fl); if (ip6_dst_lookup(sk, &dst, &fl)) diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index b2804e2d1b8c..e6bf99e3e41a 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -309,6 +309,7 @@ void dccp_reqsk_init(struct request_sock *req, struct sk_buff *skb) struct dccp_request_sock *dreq = dccp_rsk(req); inet_rsk(req)->rmt_port = dccp_hdr(skb)->dccph_sport; + inet_rsk(req)->loc_port = dccp_hdr(skb)->dccph_dport; inet_rsk(req)->acked = 0; req->rcv_wnd = sysctl_dccp_feat_sequence_window; dreq->dreq_timestamp_echo = 0; diff --git a/net/dccp/output.c b/net/dccp/output.c index d06945c7d3df..809d803d5006 100644 --- a/net/dccp/output.c +++ b/net/dccp/output.c @@ -347,7 +347,7 @@ struct sk_buff *dccp_make_response(struct sock *sk, struct dst_entry *dst, /* Build and checksum header */ dh = dccp_zeroed_hdr(skb, dccp_header_size); - dh->dccph_sport = inet_sk(sk)->sport; + dh->dccph_sport = inet_rsk(req)->loc_port; dh->dccph_dport = inet_rsk(req)->rmt_port; dh->dccph_doff = (dccp_header_size + DCCP_SKB_CB(skb)->dccpd_opt_len) / 4; diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index 2f0ac3c3eb71..28e26bd08e24 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -152,7 +152,7 @@ static struct dn_dev_parms dn_dev_list[] = { #define DN_DEV_LIST_SIZE ARRAY_SIZE(dn_dev_list) -#define DN_DEV_PARMS_OFFSET(x) ((int) ((char *) &((struct dn_dev_parms *)0)->x)) +#define DN_DEV_PARMS_OFFSET(x) offsetof(struct dn_dev_parms, x) #ifdef CONFIG_SYSCTL @@ -166,7 +166,7 @@ static int max_priority[] = { 127 }; /* From DECnet spec */ static int dn_forwarding_proc(ctl_table *, int, struct file *, void __user *, size_t *, loff_t *); -static int dn_forwarding_sysctl(ctl_table *table, int __user *name, int nlen, +static int dn_forwarding_sysctl(ctl_table *table, void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen); @@ -318,7 +318,7 @@ static int dn_forwarding_proc(ctl_table *table, int write, #endif } -static int dn_forwarding_sysctl(ctl_table *table, int __user *name, int nlen, +static int dn_forwarding_sysctl(ctl_table *table, void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen) { @@ -490,9 +490,7 @@ int dn_dev_ioctl(unsigned int cmd, void __user *arg) return -EFAULT; ifr->ifr_name[IFNAMSIZ-1] = 0; -#ifdef CONFIG_KMOD dev_load(&init_net, ifr->ifr_name); -#endif switch(cmd) { case SIOCGIFADDR: diff --git a/net/decnet/sysctl_net_decnet.c b/net/decnet/sysctl_net_decnet.c index 228067c571ba..36400b266896 100644 --- a/net/decnet/sysctl_net_decnet.c +++ b/net/decnet/sysctl_net_decnet.c @@ -132,7 +132,7 @@ static int parse_addr(__le16 *addr, char *str) } -static int dn_node_address_strategy(ctl_table *table, int __user *name, int nlen, +static int dn_node_address_strategy(ctl_table *table, void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen) { @@ -217,7 +217,7 @@ static int dn_node_address_handler(ctl_table *table, int write, } -static int dn_def_dev_strategy(ctl_table *table, int __user *name, int nlen, +static int dn_def_dev_strategy(ctl_table *table, void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen) { diff --git a/net/dsa/Kconfig b/net/dsa/Kconfig index cdce4c6c672a..49211b35725b 100644 --- a/net/dsa/Kconfig +++ b/net/dsa/Kconfig @@ -1,7 +1,7 @@ menuconfig NET_DSA bool "Distributed Switch Architecture support" default n - depends on EXPERIMENTAL + depends on EXPERIMENTAL && !S390 select PHYLIB ---help--- This allows you to use hardware switch chips that use diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index b043eda60b04..1a9dd66511fc 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -663,7 +663,7 @@ out: void arp_xmit(struct sk_buff *skb) { /* Send it off, maybe filter it using firewalling first. */ - NF_HOOK(NF_ARP, NF_ARP_OUT, skb, NULL, skb->dev, dev_queue_xmit); + NF_HOOK(NFPROTO_ARP, NF_ARP_OUT, skb, NULL, skb->dev, dev_queue_xmit); } /* @@ -928,7 +928,7 @@ static int arp_rcv(struct sk_buff *skb, struct net_device *dev, memset(NEIGH_CB(skb), 0, sizeof(struct neighbour_cb)); - return NF_HOOK(NF_ARP, NF_ARP_IN, skb, dev, NULL, arp_process); + return NF_HOOK(NFPROTO_ARP, NF_ARP_IN, skb, dev, NULL, arp_process); freeskb: kfree_skb(skb); diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c index 2c0e4572cc90..490e035c6d90 100644 --- a/net/ipv4/cipso_ipv4.c +++ b/net/ipv4/cipso_ipv4.c @@ -13,7 +13,7 @@ */ /* - * (c) Copyright Hewlett-Packard Development Company, L.P., 2006 + * (c) Copyright Hewlett-Packard Development Company, L.P., 2006, 2008 * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -47,17 +47,7 @@ #include <asm/bug.h> #include <asm/unaligned.h> -struct cipso_v4_domhsh_entry { - char *domain; - u32 valid; - struct list_head list; - struct rcu_head rcu; -}; - /* List of available DOI definitions */ -/* XXX - Updates should be minimal so having a single lock for the - * cipso_v4_doi_list and the cipso_v4_doi_list->dom_list should be - * okay. */ /* XXX - This currently assumes a minimal number of different DOIs in use, * if in practice there are a lot of different DOIs this list should * probably be turned into a hash table or something similar so we @@ -119,6 +109,19 @@ int cipso_v4_rbm_strictvalid = 1; * be omitted. */ #define CIPSO_V4_TAG_RNG_CAT_MAX 8 +/* Base length of the local tag (non-standard tag). + * Tag definition (may change between kernel versions) + * + * 0 8 16 24 32 + * +----------+----------+----------+----------+ + * | 10000000 | 00000110 | 32-bit secid value | + * +----------+----------+----------+----------+ + * | in (host byte order)| + * +----------+----------+ + * + */ +#define CIPSO_V4_TAG_LOC_BLEN 6 + /* * Helper Functions */ @@ -194,25 +197,6 @@ static void cipso_v4_bitmap_setbit(unsigned char *bitmap, } /** - * cipso_v4_doi_domhsh_free - Frees a domain list entry - * @entry: the entry's RCU field - * - * Description: - * This function is designed to be used as a callback to the call_rcu() - * function so that the memory allocated to a domain list entry can be released - * safely. - * - */ -static void cipso_v4_doi_domhsh_free(struct rcu_head *entry) -{ - struct cipso_v4_domhsh_entry *ptr; - - ptr = container_of(entry, struct cipso_v4_domhsh_entry, rcu); - kfree(ptr->domain); - kfree(ptr); -} - -/** * cipso_v4_cache_entry_free - Frees a cache entry * @entry: the entry to free * @@ -457,7 +441,7 @@ static struct cipso_v4_doi *cipso_v4_doi_search(u32 doi) struct cipso_v4_doi *iter; list_for_each_entry_rcu(iter, &cipso_v4_doi_list, list) - if (iter->doi == doi && iter->valid) + if (iter->doi == doi && atomic_read(&iter->refcount)) return iter; return NULL; } @@ -496,14 +480,17 @@ int cipso_v4_doi_add(struct cipso_v4_doi *doi_def) if (doi_def->type != CIPSO_V4_MAP_PASS) return -EINVAL; break; + case CIPSO_V4_TAG_LOCAL: + if (doi_def->type != CIPSO_V4_MAP_LOCAL) + return -EINVAL; + break; default: return -EINVAL; } } - doi_def->valid = 1; + atomic_set(&doi_def->refcount, 1); INIT_RCU_HEAD(&doi_def->rcu); - INIT_LIST_HEAD(&doi_def->dom_list); spin_lock(&cipso_v4_doi_list_lock); if (cipso_v4_doi_search(doi_def->doi) != NULL) @@ -519,59 +506,129 @@ doi_add_failure: } /** + * cipso_v4_doi_free - Frees a DOI definition + * @entry: the entry's RCU field + * + * Description: + * This function frees all of the memory associated with a DOI definition. + * + */ +void cipso_v4_doi_free(struct cipso_v4_doi *doi_def) +{ + if (doi_def == NULL) + return; + + switch (doi_def->type) { + case CIPSO_V4_MAP_TRANS: + kfree(doi_def->map.std->lvl.cipso); + kfree(doi_def->map.std->lvl.local); + kfree(doi_def->map.std->cat.cipso); + kfree(doi_def->map.std->cat.local); + break; + } + kfree(doi_def); +} + +/** + * cipso_v4_doi_free_rcu - Frees a DOI definition via the RCU pointer + * @entry: the entry's RCU field + * + * Description: + * This function is designed to be used as a callback to the call_rcu() + * function so that the memory allocated to the DOI definition can be released + * safely. + * + */ +static void cipso_v4_doi_free_rcu(struct rcu_head *entry) +{ + struct cipso_v4_doi *doi_def; + + doi_def = container_of(entry, struct cipso_v4_doi, rcu); + cipso_v4_doi_free(doi_def); +} + +/** * cipso_v4_doi_remove - Remove an existing DOI from the CIPSO protocol engine * @doi: the DOI value * @audit_secid: the LSM secid to use in the audit message - * @callback: the DOI cleanup/free callback * * Description: - * Removes a DOI definition from the CIPSO engine, @callback is called to - * free any memory. The NetLabel routines will be called to release their own - * LSM domain mappings as well as our own domain list. Returns zero on - * success and negative values on failure. + * Removes a DOI definition from the CIPSO engine. The NetLabel routines will + * be called to release their own LSM domain mappings as well as our own + * domain list. Returns zero on success and negative values on failure. * */ -int cipso_v4_doi_remove(u32 doi, - struct netlbl_audit *audit_info, - void (*callback) (struct rcu_head * head)) +int cipso_v4_doi_remove(u32 doi, struct netlbl_audit *audit_info) { struct cipso_v4_doi *doi_def; - struct cipso_v4_domhsh_entry *dom_iter; spin_lock(&cipso_v4_doi_list_lock); doi_def = cipso_v4_doi_search(doi); - if (doi_def != NULL) { - doi_def->valid = 0; - list_del_rcu(&doi_def->list); + if (doi_def == NULL) { spin_unlock(&cipso_v4_doi_list_lock); - rcu_read_lock(); - list_for_each_entry_rcu(dom_iter, &doi_def->dom_list, list) - if (dom_iter->valid) - netlbl_cfg_map_del(dom_iter->domain, - audit_info); - rcu_read_unlock(); - cipso_v4_cache_invalidate(); - call_rcu(&doi_def->rcu, callback); - return 0; + return -ENOENT; + } + if (!atomic_dec_and_test(&doi_def->refcount)) { + spin_unlock(&cipso_v4_doi_list_lock); + return -EBUSY; } + list_del_rcu(&doi_def->list); spin_unlock(&cipso_v4_doi_list_lock); - return -ENOENT; + cipso_v4_cache_invalidate(); + call_rcu(&doi_def->rcu, cipso_v4_doi_free_rcu); + + return 0; } /** - * cipso_v4_doi_getdef - Returns a pointer to a valid DOI definition + * cipso_v4_doi_getdef - Returns a reference to a valid DOI definition * @doi: the DOI value * * Description: * Searches for a valid DOI definition and if one is found it is returned to * the caller. Otherwise NULL is returned. The caller must ensure that - * rcu_read_lock() is held while accessing the returned definition. + * rcu_read_lock() is held while accessing the returned definition and the DOI + * definition reference count is decremented when the caller is done. * */ struct cipso_v4_doi *cipso_v4_doi_getdef(u32 doi) { - return cipso_v4_doi_search(doi); + struct cipso_v4_doi *doi_def; + + rcu_read_lock(); + doi_def = cipso_v4_doi_search(doi); + if (doi_def == NULL) + goto doi_getdef_return; + if (!atomic_inc_not_zero(&doi_def->refcount)) + doi_def = NULL; + +doi_getdef_return: + rcu_read_unlock(); + return doi_def; +} + +/** + * cipso_v4_doi_putdef - Releases a reference for the given DOI definition + * @doi_def: the DOI definition + * + * Description: + * Releases a DOI definition reference obtained from cipso_v4_doi_getdef(). + * + */ +void cipso_v4_doi_putdef(struct cipso_v4_doi *doi_def) +{ + if (doi_def == NULL) + return; + + if (!atomic_dec_and_test(&doi_def->refcount)) + return; + spin_lock(&cipso_v4_doi_list_lock); + list_del_rcu(&doi_def->list); + spin_unlock(&cipso_v4_doi_list_lock); + + cipso_v4_cache_invalidate(); + call_rcu(&doi_def->rcu, cipso_v4_doi_free_rcu); } /** @@ -597,7 +654,7 @@ int cipso_v4_doi_walk(u32 *skip_cnt, rcu_read_lock(); list_for_each_entry_rcu(iter_doi, &cipso_v4_doi_list, list) - if (iter_doi->valid) { + if (atomic_read(&iter_doi->refcount) > 0) { if (doi_cnt++ < *skip_cnt) continue; ret_val = callback(iter_doi, cb_arg); @@ -613,85 +670,6 @@ doi_walk_return: return ret_val; } -/** - * cipso_v4_doi_domhsh_add - Adds a domain entry to a DOI definition - * @doi_def: the DOI definition - * @domain: the domain to add - * - * Description: - * Adds the @domain to the DOI specified by @doi_def, this function - * should only be called by external functions (i.e. NetLabel). This function - * does allocate memory. Returns zero on success, negative values on failure. - * - */ -int cipso_v4_doi_domhsh_add(struct cipso_v4_doi *doi_def, const char *domain) -{ - struct cipso_v4_domhsh_entry *iter; - struct cipso_v4_domhsh_entry *new_dom; - - new_dom = kzalloc(sizeof(*new_dom), GFP_KERNEL); - if (new_dom == NULL) - return -ENOMEM; - if (domain) { - new_dom->domain = kstrdup(domain, GFP_KERNEL); - if (new_dom->domain == NULL) { - kfree(new_dom); - return -ENOMEM; - } - } - new_dom->valid = 1; - INIT_RCU_HEAD(&new_dom->rcu); - - spin_lock(&cipso_v4_doi_list_lock); - list_for_each_entry(iter, &doi_def->dom_list, list) - if (iter->valid && - ((domain != NULL && iter->domain != NULL && - strcmp(iter->domain, domain) == 0) || - (domain == NULL && iter->domain == NULL))) { - spin_unlock(&cipso_v4_doi_list_lock); - kfree(new_dom->domain); - kfree(new_dom); - return -EEXIST; - } - list_add_tail_rcu(&new_dom->list, &doi_def->dom_list); - spin_unlock(&cipso_v4_doi_list_lock); - - return 0; -} - -/** - * cipso_v4_doi_domhsh_remove - Removes a domain entry from a DOI definition - * @doi_def: the DOI definition - * @domain: the domain to remove - * - * Description: - * Removes the @domain from the DOI specified by @doi_def, this function - * should only be called by external functions (i.e. NetLabel). Returns zero - * on success and negative values on error. - * - */ -int cipso_v4_doi_domhsh_remove(struct cipso_v4_doi *doi_def, - const char *domain) -{ - struct cipso_v4_domhsh_entry *iter; - - spin_lock(&cipso_v4_doi_list_lock); - list_for_each_entry(iter, &doi_def->dom_list, list) - if (iter->valid && - ((domain != NULL && iter->domain != NULL && - strcmp(iter->domain, domain) == 0) || - (domain == NULL && iter->domain == NULL))) { - iter->valid = 0; - list_del_rcu(&iter->list); - spin_unlock(&cipso_v4_doi_list_lock); - call_rcu(&iter->rcu, cipso_v4_doi_domhsh_free); - return 0; - } - spin_unlock(&cipso_v4_doi_list_lock); - - return -ENOENT; -} - /* * Label Mapping Functions */ @@ -712,7 +690,7 @@ static int cipso_v4_map_lvl_valid(const struct cipso_v4_doi *doi_def, u8 level) switch (doi_def->type) { case CIPSO_V4_MAP_PASS: return 0; - case CIPSO_V4_MAP_STD: + case CIPSO_V4_MAP_TRANS: if (doi_def->map.std->lvl.cipso[level] < CIPSO_V4_INV_LVL) return 0; break; @@ -741,7 +719,7 @@ static int cipso_v4_map_lvl_hton(const struct cipso_v4_doi *doi_def, case CIPSO_V4_MAP_PASS: *net_lvl = host_lvl; return 0; - case CIPSO_V4_MAP_STD: + case CIPSO_V4_MAP_TRANS: if (host_lvl < doi_def->map.std->lvl.local_size && doi_def->map.std->lvl.local[host_lvl] < CIPSO_V4_INV_LVL) { *net_lvl = doi_def->map.std->lvl.local[host_lvl]; @@ -775,7 +753,7 @@ static int cipso_v4_map_lvl_ntoh(const struct cipso_v4_doi *doi_def, case CIPSO_V4_MAP_PASS: *host_lvl = net_lvl; return 0; - case CIPSO_V4_MAP_STD: + case CIPSO_V4_MAP_TRANS: map_tbl = doi_def->map.std; if (net_lvl < map_tbl->lvl.cipso_size && map_tbl->lvl.cipso[net_lvl] < CIPSO_V4_INV_LVL) { @@ -812,7 +790,7 @@ static int cipso_v4_map_cat_rbm_valid(const struct cipso_v4_doi *doi_def, switch (doi_def->type) { case CIPSO_V4_MAP_PASS: return 0; - case CIPSO_V4_MAP_STD: + case CIPSO_V4_MAP_TRANS: cipso_cat_size = doi_def->map.std->cat.cipso_size; cipso_array = doi_def->map.std->cat.cipso; for (;;) { @@ -860,7 +838,7 @@ static int cipso_v4_map_cat_rbm_hton(const struct cipso_v4_doi *doi_def, u32 host_cat_size = 0; u32 *host_cat_array = NULL; - if (doi_def->type == CIPSO_V4_MAP_STD) { + if (doi_def->type == CIPSO_V4_MAP_TRANS) { host_cat_size = doi_def->map.std->cat.local_size; host_cat_array = doi_def->map.std->cat.local; } @@ -875,7 +853,7 @@ static int cipso_v4_map_cat_rbm_hton(const struct cipso_v4_doi *doi_def, case CIPSO_V4_MAP_PASS: net_spot = host_spot; break; - case CIPSO_V4_MAP_STD: + case CIPSO_V4_MAP_TRANS: if (host_spot >= host_cat_size) return -EPERM; net_spot = host_cat_array[host_spot]; @@ -921,7 +899,7 @@ static int cipso_v4_map_cat_rbm_ntoh(const struct cipso_v4_doi *doi_def, u32 net_cat_size = 0; u32 *net_cat_array = NULL; - if (doi_def->type == CIPSO_V4_MAP_STD) { + if (doi_def->type == CIPSO_V4_MAP_TRANS) { net_cat_size = doi_def->map.std->cat.cipso_size; net_cat_array = doi_def->map.std->cat.cipso; } @@ -941,7 +919,7 @@ static int cipso_v4_map_cat_rbm_ntoh(const struct cipso_v4_doi *doi_def, case CIPSO_V4_MAP_PASS: host_spot = net_spot; break; - case CIPSO_V4_MAP_STD: + case CIPSO_V4_MAP_TRANS: if (net_spot >= net_cat_size) return -EPERM; host_spot = net_cat_array[net_spot]; @@ -1277,7 +1255,7 @@ static int cipso_v4_gentag_rbm(const struct cipso_v4_doi *doi_def, } else tag_len = 4; - buffer[0] = 0x01; + buffer[0] = CIPSO_V4_TAG_RBITMAP; buffer[1] = tag_len; buffer[3] = level; @@ -1373,7 +1351,7 @@ static int cipso_v4_gentag_enum(const struct cipso_v4_doi *doi_def, } else tag_len = 4; - buffer[0] = 0x02; + buffer[0] = CIPSO_V4_TAG_ENUM; buffer[1] = tag_len; buffer[3] = level; @@ -1469,7 +1447,7 @@ static int cipso_v4_gentag_rng(const struct cipso_v4_doi *doi_def, } else tag_len = 4; - buffer[0] = 0x05; + buffer[0] = CIPSO_V4_TAG_RANGE; buffer[1] = tag_len; buffer[3] = level; @@ -1523,6 +1501,54 @@ static int cipso_v4_parsetag_rng(const struct cipso_v4_doi *doi_def, } /** + * cipso_v4_gentag_loc - Generate a CIPSO local tag (non-standard) + * @doi_def: the DOI definition + * @secattr: the security attributes + * @buffer: the option buffer + * @buffer_len: length of buffer in bytes + * + * Description: + * Generate a CIPSO option using the local tag. Returns the size of the tag + * on success, negative values on failure. + * + */ +static int cipso_v4_gentag_loc(const struct cipso_v4_doi *doi_def, + const struct netlbl_lsm_secattr *secattr, + unsigned char *buffer, + u32 buffer_len) +{ + if (!(secattr->flags & NETLBL_SECATTR_SECID)) + return -EPERM; + + buffer[0] = CIPSO_V4_TAG_LOCAL; + buffer[1] = CIPSO_V4_TAG_LOC_BLEN; + *(u32 *)&buffer[2] = secattr->attr.secid; + + return CIPSO_V4_TAG_LOC_BLEN; +} + +/** + * cipso_v4_parsetag_loc - Parse a CIPSO local tag + * @doi_def: the DOI definition + * @tag: the CIPSO tag + * @secattr: the security attributes + * + * Description: + * Parse a CIPSO local tag and return the security attributes in @secattr. + * Return zero on success, negatives values on failure. + * + */ +static int cipso_v4_parsetag_loc(const struct cipso_v4_doi *doi_def, + const unsigned char *tag, + struct netlbl_lsm_secattr *secattr) +{ + secattr->attr.secid = *(u32 *)&tag[2]; + secattr->flags |= NETLBL_SECATTR_SECID; + + return 0; +} + +/** * cipso_v4_validate - Validate a CIPSO option * @option: the start of the option, on error it is set to point to the error * @@ -1541,7 +1567,7 @@ static int cipso_v4_parsetag_rng(const struct cipso_v4_doi *doi_def, * that is unrecognized." * */ -int cipso_v4_validate(unsigned char **option) +int cipso_v4_validate(const struct sk_buff *skb, unsigned char **option) { unsigned char *opt = *option; unsigned char *tag; @@ -1566,7 +1592,7 @@ int cipso_v4_validate(unsigned char **option) goto validate_return_locked; } - opt_iter = 6; + opt_iter = CIPSO_V4_HDR_LEN; tag = opt + opt_iter; while (opt_iter < opt_len) { for (tag_iter = 0; doi_def->tags[tag_iter] != tag[0];) @@ -1584,7 +1610,7 @@ int cipso_v4_validate(unsigned char **option) switch (tag[0]) { case CIPSO_V4_TAG_RBITMAP: - if (tag_len < 4) { + if (tag_len < CIPSO_V4_TAG_RBM_BLEN) { err_offset = opt_iter + 1; goto validate_return_locked; } @@ -1602,7 +1628,7 @@ int cipso_v4_validate(unsigned char **option) err_offset = opt_iter + 3; goto validate_return_locked; } - if (tag_len > 4 && + if (tag_len > CIPSO_V4_TAG_RBM_BLEN && cipso_v4_map_cat_rbm_valid(doi_def, &tag[4], tag_len - 4) < 0) { @@ -1612,7 +1638,7 @@ int cipso_v4_validate(unsigned char **option) } break; case CIPSO_V4_TAG_ENUM: - if (tag_len < 4) { + if (tag_len < CIPSO_V4_TAG_ENUM_BLEN) { err_offset = opt_iter + 1; goto validate_return_locked; } @@ -1622,7 +1648,7 @@ int cipso_v4_validate(unsigned char **option) err_offset = opt_iter + 3; goto validate_return_locked; } - if (tag_len > 4 && + if (tag_len > CIPSO_V4_TAG_ENUM_BLEN && cipso_v4_map_cat_enum_valid(doi_def, &tag[4], tag_len - 4) < 0) { @@ -1631,7 +1657,7 @@ int cipso_v4_validate(unsigned char **option) } break; case CIPSO_V4_TAG_RANGE: - if (tag_len < 4) { + if (tag_len < CIPSO_V4_TAG_RNG_BLEN) { err_offset = opt_iter + 1; goto validate_return_locked; } @@ -1641,7 +1667,7 @@ int cipso_v4_validate(unsigned char **option) err_offset = opt_iter + 3; goto validate_return_locked; } - if (tag_len > 4 && + if (tag_len > CIPSO_V4_TAG_RNG_BLEN && cipso_v4_map_cat_rng_valid(doi_def, &tag[4], tag_len - 4) < 0) { @@ -1649,6 +1675,19 @@ int cipso_v4_validate(unsigned char **option) goto validate_return_locked; } break; + case CIPSO_V4_TAG_LOCAL: + /* This is a non-standard tag that we only allow for + * local connections, so if the incoming interface is + * not the loopback device drop the packet. */ + if (!(skb->dev->flags & IFF_LOOPBACK)) { + err_offset = opt_iter; + goto validate_return_locked; + } + if (tag_len != CIPSO_V4_TAG_LOC_BLEN) { + err_offset = opt_iter + 1; + goto validate_return_locked; + } + break; default: err_offset = opt_iter; goto validate_return_locked; @@ -1704,48 +1743,27 @@ void cipso_v4_error(struct sk_buff *skb, int error, u32 gateway) } /** - * cipso_v4_sock_setattr - Add a CIPSO option to a socket - * @sk: the socket + * cipso_v4_genopt - Generate a CIPSO option + * @buf: the option buffer + * @buf_len: the size of opt_buf * @doi_def: the CIPSO DOI to use - * @secattr: the specific security attributes of the socket + * @secattr: the security attributes * * Description: - * Set the CIPSO option on the given socket using the DOI definition and - * security attributes passed to the function. This function requires - * exclusive access to @sk, which means it either needs to be in the - * process of being created or locked. Returns zero on success and negative - * values on failure. + * Generate a CIPSO option using the DOI definition and security attributes + * passed to the function. Returns the length of the option on success and + * negative values on failure. * */ -int cipso_v4_sock_setattr(struct sock *sk, - const struct cipso_v4_doi *doi_def, - const struct netlbl_lsm_secattr *secattr) +static int cipso_v4_genopt(unsigned char *buf, u32 buf_len, + const struct cipso_v4_doi *doi_def, + const struct netlbl_lsm_secattr *secattr) { - int ret_val = -EPERM; + int ret_val; u32 iter; - unsigned char *buf; - u32 buf_len = 0; - u32 opt_len; - struct ip_options *opt = NULL; - struct inet_sock *sk_inet; - struct inet_connection_sock *sk_conn; - /* In the case of sock_create_lite(), the sock->sk field is not - * defined yet but it is not a problem as the only users of these - * "lite" PF_INET sockets are functions which do an accept() call - * afterwards so we will label the socket as part of the accept(). */ - if (sk == NULL) - return 0; - - /* We allocate the maximum CIPSO option size here so we are probably - * being a little wasteful, but it makes our life _much_ easier later - * on and after all we are only talking about 40 bytes. */ - buf_len = CIPSO_V4_OPT_LEN_MAX; - buf = kmalloc(buf_len, GFP_ATOMIC); - if (buf == NULL) { - ret_val = -ENOMEM; - goto socket_setattr_failure; - } + if (buf_len <= CIPSO_V4_HDR_LEN) + return -ENOSPC; /* XXX - This code assumes only one tag per CIPSO option which isn't * really a good assumption to make but since we only support the MAC @@ -1772,9 +1790,14 @@ int cipso_v4_sock_setattr(struct sock *sk, &buf[CIPSO_V4_HDR_LEN], buf_len - CIPSO_V4_HDR_LEN); break; + case CIPSO_V4_TAG_LOCAL: + ret_val = cipso_v4_gentag_loc(doi_def, + secattr, + &buf[CIPSO_V4_HDR_LEN], + buf_len - CIPSO_V4_HDR_LEN); + break; default: - ret_val = -EPERM; - goto socket_setattr_failure; + return -EPERM; } iter++; @@ -1782,9 +1805,58 @@ int cipso_v4_sock_setattr(struct sock *sk, iter < CIPSO_V4_TAG_MAXCNT && doi_def->tags[iter] != CIPSO_V4_TAG_INVALID); if (ret_val < 0) - goto socket_setattr_failure; + return ret_val; cipso_v4_gentag_hdr(doi_def, buf, ret_val); - buf_len = CIPSO_V4_HDR_LEN + ret_val; + return CIPSO_V4_HDR_LEN + ret_val; +} + +/** + * cipso_v4_sock_setattr - Add a CIPSO option to a socket + * @sk: the socket + * @doi_def: the CIPSO DOI to use + * @secattr: the specific security attributes of the socket + * + * Description: + * Set the CIPSO option on the given socket using the DOI definition and + * security attributes passed to the function. This function requires + * exclusive access to @sk, which means it either needs to be in the + * process of being created or locked. Returns zero on success and negative + * values on failure. + * + */ +int cipso_v4_sock_setattr(struct sock *sk, + const struct cipso_v4_doi *doi_def, + const struct netlbl_lsm_secattr *secattr) +{ + int ret_val = -EPERM; + unsigned char *buf = NULL; + u32 buf_len; + u32 opt_len; + struct ip_options *opt = NULL; + struct inet_sock *sk_inet; + struct inet_connection_sock *sk_conn; + + /* In the case of sock_create_lite(), the sock->sk field is not + * defined yet but it is not a problem as the only users of these + * "lite" PF_INET sockets are functions which do an accept() call + * afterwards so we will label the socket as part of the accept(). */ + if (sk == NULL) + return 0; + + /* We allocate the maximum CIPSO option size here so we are probably + * being a little wasteful, but it makes our life _much_ easier later + * on and after all we are only talking about 40 bytes. */ + buf_len = CIPSO_V4_OPT_LEN_MAX; + buf = kmalloc(buf_len, GFP_ATOMIC); + if (buf == NULL) { + ret_val = -ENOMEM; + goto socket_setattr_failure; + } + + ret_val = cipso_v4_genopt(buf, buf_len, doi_def, secattr); + if (ret_val < 0) + goto socket_setattr_failure; + buf_len = ret_val; /* We can't use ip_options_get() directly because it makes a call to * ip_options_get_alloc() which allocates memory with GFP_KERNEL and @@ -1822,6 +1894,80 @@ socket_setattr_failure: } /** + * cipso_v4_sock_delattr - Delete the CIPSO option from a socket + * @sk: the socket + * + * Description: + * Removes the CIPSO option from a socket, if present. + * + */ +void cipso_v4_sock_delattr(struct sock *sk) +{ + u8 hdr_delta; + struct ip_options *opt; + struct inet_sock *sk_inet; + + sk_inet = inet_sk(sk); + opt = sk_inet->opt; + if (opt == NULL || opt->cipso == 0) + return; + + if (opt->srr || opt->rr || opt->ts || opt->router_alert) { + u8 cipso_len; + u8 cipso_off; + unsigned char *cipso_ptr; + int iter; + int optlen_new; + + cipso_off = opt->cipso - sizeof(struct iphdr); + cipso_ptr = &opt->__data[cipso_off]; + cipso_len = cipso_ptr[1]; + + if (opt->srr > opt->cipso) + opt->srr -= cipso_len; + if (opt->rr > opt->cipso) + opt->rr -= cipso_len; + if (opt->ts > opt->cipso) + opt->ts -= cipso_len; + if (opt->router_alert > opt->cipso) + opt->router_alert -= cipso_len; + opt->cipso = 0; + + memmove(cipso_ptr, cipso_ptr + cipso_len, + opt->optlen - cipso_off - cipso_len); + + /* determining the new total option length is tricky because of + * the padding necessary, the only thing i can think to do at + * this point is walk the options one-by-one, skipping the + * padding at the end to determine the actual option size and + * from there we can determine the new total option length */ + iter = 0; + optlen_new = 0; + while (iter < opt->optlen) + if (opt->__data[iter] != IPOPT_NOP) { + iter += opt->__data[iter + 1]; + optlen_new = iter; + } else + iter++; + hdr_delta = opt->optlen; + opt->optlen = (optlen_new + 3) & ~3; + hdr_delta -= opt->optlen; + } else { + /* only the cipso option was present on the socket so we can + * remove the entire option struct */ + sk_inet->opt = NULL; + hdr_delta = opt->optlen; + kfree(opt); + } + + if (sk_inet->is_icsk && hdr_delta > 0) { + struct inet_connection_sock *sk_conn = inet_csk(sk); + sk_conn->icsk_ext_hdr_len -= hdr_delta; + sk_conn->icsk_sync_mss(sk, sk_conn->icsk_pmtu_cookie); + } +} + +/** * cipso_v4_getattr - Helper function for the cipso_v4_*_getattr functions * @cipso: the CIPSO v4 option * @secattr: the security attributes @@ -1859,6 +2005,9 @@ static int cipso_v4_getattr(const unsigned char *cipso, case CIPSO_V4_TAG_RANGE: ret_val = cipso_v4_parsetag_rng(doi_def, &cipso[6], secattr); break; + case CIPSO_V4_TAG_LOCAL: + ret_val = cipso_v4_parsetag_loc(doi_def, &cipso[6], secattr); + break; } if (ret_val == 0) secattr->type = NETLBL_NLTYPE_CIPSOV4; @@ -1893,6 +2042,123 @@ int cipso_v4_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr) } /** + * cipso_v4_skbuff_setattr - Set the CIPSO option on a packet + * @skb: the packet + * @secattr: the security attributes + * + * Description: + * Set the CIPSO option on the given packet based on the security attributes. + * Returns a pointer to the IP header on success and NULL on failure. + * + */ +int cipso_v4_skbuff_setattr(struct sk_buff *skb, + const struct cipso_v4_doi *doi_def, + const struct netlbl_lsm_secattr *secattr) +{ + int ret_val; + struct iphdr *iph; + struct ip_options *opt = &IPCB(skb)->opt; + unsigned char buf[CIPSO_V4_OPT_LEN_MAX]; + u32 buf_len = CIPSO_V4_OPT_LEN_MAX; + u32 opt_len; + int len_delta; + + buf_len = cipso_v4_genopt(buf, buf_len, doi_def, secattr); + if (buf_len < 0) + return buf_len; + opt_len = (buf_len + 3) & ~3; + + /* we overwrite any existing options to ensure that we have enough + * room for the CIPSO option, the reason is that we _need_ to guarantee + * that the security label is applied to the packet - we do the same + * thing when using the socket options and it hasn't caused a problem, + * if we need to we can always revisit this choice later */ + + len_delta = opt_len - opt->optlen; + /* if we don't ensure enough headroom we could panic on the skb_push() + * call below so make sure we have enough, we are also "mangling" the + * packet so we should probably do a copy-on-write call anyway */ + ret_val = skb_cow(skb, skb_headroom(skb) + len_delta); + if (ret_val < 0) + return ret_val; + + if (len_delta > 0) { + /* we assume that the header + opt->optlen have already been + * "pushed" in ip_options_build() or similar */ + iph = ip_hdr(skb); + skb_push(skb, len_delta); + memmove((char *)iph - len_delta, iph, iph->ihl << 2); + skb_reset_network_header(skb); + iph = ip_hdr(skb); + } else if (len_delta < 0) { + iph = ip_hdr(skb); + memset(iph + 1, IPOPT_NOP, opt->optlen); + } else + iph = ip_hdr(skb); + + if (opt->optlen > 0) + memset(opt, 0, sizeof(*opt)); + opt->optlen = opt_len; + opt->cipso = sizeof(struct iphdr); + opt->is_changed = 1; + + /* we have to do the following because we are being called from a + * netfilter hook which means the packet already has had the header + * fields populated and the checksum calculated - yes this means we + * are doing more work than needed but we do it to keep the core + * stack clean and tidy */ + memcpy(iph + 1, buf, buf_len); + if (opt_len > buf_len) + memset((char *)(iph + 1) + buf_len, 0, opt_len - buf_len); + if (len_delta != 0) { + iph->ihl = 5 + (opt_len >> 2); + iph->tot_len = htons(skb->len); + } + ip_send_check(iph); + + return 0; +} + +/** + * cipso_v4_skbuff_delattr - Delete any CIPSO options from a packet + * @skb: the packet + * + * Description: + * Removes any and all CIPSO options from the given packet. Returns zero on + * success, negative values on failure. + * + */ +int cipso_v4_skbuff_delattr(struct sk_buff *skb) +{ + int ret_val; + struct iphdr *iph; + struct ip_options *opt = &IPCB(skb)->opt; + unsigned char *cipso_ptr; + + if (opt->cipso == 0) + return 0; + + /* since we are changing the packet we should make a copy */ + ret_val = skb_cow(skb, skb_headroom(skb)); + if (ret_val < 0) + return ret_val; + + /* the easiest thing to do is just replace the cipso option with noop + * options since we don't change the size of the packet, although we + * still need to recalculate the checksum */ + + iph = ip_hdr(skb); + cipso_ptr = (unsigned char *)iph + opt->cipso; + memset(cipso_ptr, IPOPT_NOOP, cipso_ptr[1]); + opt->cipso = 0; + opt->is_changed = 1; + + ip_send_check(iph); + + return 0; +} + +/** * cipso_v4_skbuff_getattr - Get the security attributes from the CIPSO option * @skb: the packet * @secattr: the security attributes diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index b12dae2b0b2d..56fce3ab6c55 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -613,9 +613,7 @@ int devinet_ioctl(struct net *net, unsigned int cmd, void __user *arg) if (colon) *colon = 0; -#ifdef CONFIG_KMOD dev_load(net, ifr.ifr_name); -#endif switch (cmd) { case SIOCGIFADDR: /* Get interface address */ @@ -1283,7 +1281,7 @@ static int devinet_conf_proc(ctl_table *ctl, int write, return ret; } -static int devinet_conf_sysctl(ctl_table *table, int __user *name, int nlen, +static int devinet_conf_sysctl(ctl_table *table, void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen) { @@ -1379,12 +1377,11 @@ int ipv4_doint_and_flush(ctl_table *ctl, int write, return ret; } -int ipv4_doint_and_flush_strategy(ctl_table *table, int __user *name, int nlen, +int ipv4_doint_and_flush_strategy(ctl_table *table, void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen) { - int ret = devinet_conf_sysctl(table, name, nlen, oldval, oldlenp, - newval, newlen); + int ret = devinet_conf_sysctl(table, oldval, oldlenp, newval, newlen); struct net *net = table->extra2; if (ret == 1) diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index 55c355e63234..72b2de76f1cd 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -1,7 +1,7 @@ /* * NET3: Implementation of the ICMP protocol layer. * - * Alan Cox, <alan@redhat.com> + * Alan Cox, <alan@lxorguk.ukuu.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c index 7f9e337e3908..a0d86455c53e 100644 --- a/net/ipv4/igmp.c +++ b/net/ipv4/igmp.c @@ -9,7 +9,7 @@ * seems to fall out with gcc 2.6.2. * * Authors: - * Alan Cox <Alan.Cox@linux.org> + * Alan Cox <alan@lxorguk.ukuu.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index 89cb047ab314..564230dabcb8 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -53,11 +53,9 @@ static DEFINE_MUTEX(inet_diag_table_mutex); static const struct inet_diag_handler *inet_diag_lock_handler(int type) { -#ifdef CONFIG_KMOD if (!inet_diag_table[type]) request_module("net-pf-%d-proto-%d-type-%d", PF_NETLINK, NETLINK_INET_DIAG, type); -#endif mutex_lock(&inet_diag_table_mutex); if (!inet_diag_table[type]) diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c index 2152d222b954..e4f81f54befe 100644 --- a/net/ipv4/ip_fragment.c +++ b/net/ipv4/ip_fragment.c @@ -6,7 +6,7 @@ * The IP fragmentation functionality. * * Authors: Fred N. van Kempen <waltje@uWalt.NL.Mugnet.ORG> - * Alan Cox <Alan.Cox@linux.org> + * Alan Cox <alan@lxorguk.ukuu.org.uk> * * Fixes: * Alan Cox : Split from ip.c , see ip_input.c for history. diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c index e0bed56c51f1..861978a4f1a8 100644 --- a/net/ipv4/ip_input.c +++ b/net/ipv4/ip_input.c @@ -8,7 +8,7 @@ * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Donald Becker, <becker@super.org> - * Alan Cox, <Alan.Cox@linux.org> + * Alan Cox, <alan@lxorguk.ukuu.org.uk> * Richard Underwood * Stefan Becker, <stefanb@yello.ping.de> * Jorge Cwik, <jorge@laser.satlink.net> diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index be3f18a7a40e..2c88da6e7862 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -438,7 +438,7 @@ int ip_options_compile(struct net *net, goto error; } opt->cipso = optptr - iph; - if (cipso_v4_validate(&optptr)) { + if (cipso_v4_validate(skb, &optptr)) { pp_ptr = optptr; goto error; } diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index 4c6d2caf9203..29609d29df76 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -41,7 +41,7 @@ Made the tunnels use dev->name not tunnel: when error reporting. Added tx_dropped stat - -Alan Cox (Alan.Cox@linux.org) 21 March 95 + -Alan Cox (alan@lxorguk.ukuu.org.uk) 21 March 95 Reworked: Changed to tunnel to destination gateway in addition to the diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index c519b8d30eee..b42e082cc170 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1,7 +1,7 @@ /* * IP multicast routing support for mrouted 3.6/3.8 * - * (c) 1995 Alan Cox, <alan@redhat.com> + * (c) 1995 Alan Cox, <alan@lxorguk.ukuu.org.uk> * Linux Consultancy and Custom Driver Development * * This program is free software; you can redistribute it and/or diff --git a/net/ipv4/netfilter/nf_defrag_ipv4.c b/net/ipv4/netfilter/nf_defrag_ipv4.c index aa2c50a180f7..fa2d6b6fc3e5 100644 --- a/net/ipv4/netfilter/nf_defrag_ipv4.c +++ b/net/ipv4/netfilter/nf_defrag_ipv4.c @@ -41,12 +41,13 @@ static unsigned int ipv4_conntrack_defrag(unsigned int hooknum, int (*okfn)(struct sk_buff *)) { #if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE) +#if !defined(CONFIG_NF_NAT) && !defined(CONFIG_NF_NAT_MODULE) /* Previously seen (loopback)? Ignore. Do this before fragment check. */ if (skb->nfct) return NF_ACCEPT; #endif - +#endif /* Gather fragments. */ if (ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)) { if (nf_ct_ipv4_gather_frags(skb, diff --git a/net/ipv4/netfilter/nf_nat_core.c b/net/ipv4/netfilter/nf_nat_core.c index 2ac9eaf1a8c9..a65cf692359f 100644 --- a/net/ipv4/netfilter/nf_nat_core.c +++ b/net/ipv4/netfilter/nf_nat_core.c @@ -584,6 +584,98 @@ static struct nf_ct_ext_type nat_extend __read_mostly = { .flags = NF_CT_EXT_F_PREALLOC, }; +#if defined(CONFIG_NF_CT_NETLINK) || defined(CONFIG_NF_CT_NETLINK_MODULE) + +#include <linux/netfilter/nfnetlink.h> +#include <linux/netfilter/nfnetlink_conntrack.h> + +static const struct nla_policy protonat_nla_policy[CTA_PROTONAT_MAX+1] = { + [CTA_PROTONAT_PORT_MIN] = { .type = NLA_U16 }, + [CTA_PROTONAT_PORT_MAX] = { .type = NLA_U16 }, +}; + +static int nfnetlink_parse_nat_proto(struct nlattr *attr, + const struct nf_conn *ct, + struct nf_nat_range *range) +{ + struct nlattr *tb[CTA_PROTONAT_MAX+1]; + const struct nf_nat_protocol *npt; + int err; + + err = nla_parse_nested(tb, CTA_PROTONAT_MAX, attr, protonat_nla_policy); + if (err < 0) + return err; + + npt = nf_nat_proto_find_get(nf_ct_protonum(ct)); + if (npt->nlattr_to_range) + err = npt->nlattr_to_range(tb, range); + nf_nat_proto_put(npt); + return err; +} + +static const struct nla_policy nat_nla_policy[CTA_NAT_MAX+1] = { + [CTA_NAT_MINIP] = { .type = NLA_U32 }, + [CTA_NAT_MAXIP] = { .type = NLA_U32 }, +}; + +static int +nfnetlink_parse_nat(struct nlattr *nat, + const struct nf_conn *ct, struct nf_nat_range *range) +{ + struct nlattr *tb[CTA_NAT_MAX+1]; + int err; + + memset(range, 0, sizeof(*range)); + + err = nla_parse_nested(tb, CTA_NAT_MAX, nat, nat_nla_policy); + if (err < 0) + return err; + + if (tb[CTA_NAT_MINIP]) + range->min_ip = nla_get_be32(tb[CTA_NAT_MINIP]); + + if (!tb[CTA_NAT_MAXIP]) + range->max_ip = range->min_ip; + else + range->max_ip = nla_get_be32(tb[CTA_NAT_MAXIP]); + + if (range->min_ip) + range->flags |= IP_NAT_RANGE_MAP_IPS; + + if (!tb[CTA_NAT_PROTO]) + return 0; + + err = nfnetlink_parse_nat_proto(tb[CTA_NAT_PROTO], ct, range); + if (err < 0) + return err; + + return 0; +} + +static int +nfnetlink_parse_nat_setup(struct nf_conn *ct, + enum nf_nat_manip_type manip, + struct nlattr *attr) +{ + struct nf_nat_range range; + + if (nfnetlink_parse_nat(attr, ct, &range) < 0) + return -EINVAL; + if (nf_nat_initialized(ct, manip)) + return -EEXIST; + + return nf_nat_setup_info(ct, &range, manip); +} +#else +static int +nfnetlink_parse_nat_setup(struct nf_conn *ct, + enum nf_nat_manip_type manip, + struct nlattr *attr) +{ + return -EOPNOTSUPP; +} +#endif + static int __net_init nf_nat_net_init(struct net *net) { net->ipv4.nat_bysource = nf_ct_alloc_hashtable(&nf_nat_htable_size, @@ -654,6 +746,9 @@ static int __init nf_nat_init(void) BUG_ON(nf_nat_seq_adjust_hook != NULL); rcu_assign_pointer(nf_nat_seq_adjust_hook, nf_nat_seq_adjust); + BUG_ON(nfnetlink_parse_nat_setup_hook != NULL); + rcu_assign_pointer(nfnetlink_parse_nat_setup_hook, + nfnetlink_parse_nat_setup); return 0; cleanup_extend: @@ -667,10 +762,12 @@ static void __exit nf_nat_cleanup(void) nf_ct_l3proto_put(l3proto); nf_ct_extend_unregister(&nat_extend); rcu_assign_pointer(nf_nat_seq_adjust_hook, NULL); + rcu_assign_pointer(nfnetlink_parse_nat_setup_hook, NULL); synchronize_net(); } MODULE_LICENSE("GPL"); +MODULE_ALIAS("nf-nat-ipv4"); module_init(nf_nat_init); module_exit(nf_nat_cleanup); diff --git a/net/ipv4/netfilter/nf_nat_snmp_basic.c b/net/ipv4/netfilter/nf_nat_snmp_basic.c index ffeaffc3fffe..8303e4b406c0 100644 --- a/net/ipv4/netfilter/nf_nat_snmp_basic.c +++ b/net/ipv4/netfilter/nf_nat_snmp_basic.c @@ -742,6 +742,7 @@ static unsigned char snmp_object_decode(struct asn1_ctx *ctx, *obj = kmalloc(sizeof(struct snmp_object) + len, GFP_ATOMIC); if (*obj == NULL) { + kfree(p); kfree(id); if (net_ratelimit()) printk("OOM in bsalg (%d)\n", __LINE__); diff --git a/net/ipv4/route.c b/net/ipv4/route.c index a6d7c584f53b..2ea6dcc3e2cc 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1109,7 +1109,12 @@ restart: printk("\n"); } #endif - rt_hash_table[hash].chain = rt; + /* + * Since lookup is lockfree, we must make sure + * previous writes to rt are comitted to memory + * before making rt visible to other CPUS. + */ + rcu_assign_pointer(rt_hash_table[hash].chain, rt); spin_unlock_bh(rt_hash_lock_addr(hash)); *rp = rt; return 0; @@ -2908,8 +2913,6 @@ static int ipv4_sysctl_rtcache_flush(ctl_table *__ctl, int write, } static int ipv4_sysctl_rtcache_flush_strategy(ctl_table *table, - int __user *name, - int nlen, void __user *oldval, size_t __user *oldlenp, void __user *newval, @@ -2972,16 +2975,13 @@ static int ipv4_sysctl_rt_secret_interval(ctl_table *ctl, int write, } static int ipv4_sysctl_rt_secret_interval_strategy(ctl_table *table, - int __user *name, - int nlen, void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen) { int old = ip_rt_secret_interval; - int ret = sysctl_jiffies(table, name, nlen, oldval, oldlenp, newval, - newlen); + int ret = sysctl_jiffies(table, oldval, oldlenp, newval, newlen); rt_secret_reschedule(old); diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index 276d047fb85a..1bb10df8ce7d 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -64,8 +64,8 @@ static int ipv4_local_port_range(ctl_table *table, int write, struct file *filp, } /* Validate changes from sysctl interface. */ -static int ipv4_sysctl_local_port_range(ctl_table *table, int __user *name, - int nlen, void __user *oldval, +static int ipv4_sysctl_local_port_range(ctl_table *table, + void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen) { @@ -80,7 +80,7 @@ static int ipv4_sysctl_local_port_range(ctl_table *table, int __user *name, }; inet_get_local_port_range(range, range + 1); - ret = sysctl_intvec(&tmp, name, nlen, oldval, oldlenp, newval, newlen); + ret = sysctl_intvec(&tmp, oldval, oldlenp, newval, newlen); if (ret == 0 && newval && newlen) { if (range[1] < range[0]) ret = -EINVAL; @@ -109,8 +109,8 @@ static int proc_tcp_congestion_control(ctl_table *ctl, int write, struct file * return ret; } -static int sysctl_tcp_congestion_control(ctl_table *table, int __user *name, - int nlen, void __user *oldval, +static int sysctl_tcp_congestion_control(ctl_table *table, + void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen) { @@ -122,7 +122,7 @@ static int sysctl_tcp_congestion_control(ctl_table *table, int __user *name, int ret; tcp_get_default_congestion_control(val); - ret = sysctl_string(&tbl, name, nlen, oldval, oldlenp, newval, newlen); + ret = sysctl_string(&tbl, oldval, oldlenp, newval, newlen); if (ret == 1 && newval && newlen) ret = tcp_set_default_congestion_control(val); return ret; @@ -165,8 +165,8 @@ static int proc_allowed_congestion_control(ctl_table *ctl, return ret; } -static int strategy_allowed_congestion_control(ctl_table *table, int __user *name, - int nlen, void __user *oldval, +static int strategy_allowed_congestion_control(ctl_table *table, + void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen) @@ -179,7 +179,7 @@ static int strategy_allowed_congestion_control(ctl_table *table, int __user *nam return -ENOMEM; tcp_get_available_congestion_control(tbl.data, tbl.maxlen); - ret = sysctl_string(&tbl, name, nlen, oldval, oldlenp, newval, newlen); + ret = sysctl_string(&tbl, oldval, oldlenp, newval, newlen); if (ret == 1 && newval && newlen) ret = tcp_set_allowed_congestion_control(tbl.data); kfree(tbl.data); diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c index 6a250828b767..4ec5b4e97c4e 100644 --- a/net/ipv4/tcp_cong.c +++ b/net/ipv4/tcp_cong.c @@ -115,7 +115,7 @@ int tcp_set_default_congestion_control(const char *name) spin_lock(&tcp_cong_list_lock); ca = tcp_ca_find(name); -#ifdef CONFIG_KMOD +#ifdef CONFIG_MODULES if (!ca && capable(CAP_SYS_MODULE)) { spin_unlock(&tcp_cong_list_lock); @@ -244,7 +244,7 @@ int tcp_set_congestion_control(struct sock *sk, const char *name) if (ca == icsk->icsk_ca_ops) goto out; -#ifdef CONFIG_KMOD +#ifdef CONFIG_MODULES /* not found attempt to autoload module */ if (!ca && capable(CAP_SYS_MODULE)) { rcu_read_unlock(); diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index eacf4cfef146..2095abc3caba 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -8,7 +8,7 @@ * Authors: Ross Biro * Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG> * Arnt Gulbrandsen, <agulbra@nvg.unit.no> - * Alan Cox, <Alan.Cox@linux.org> + * Alan Cox, <alan@lxorguk.ukuu.org.uk> * Hirokazu Takahashi, <taka@valinux.co.jp> * * Fixes: diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 7b6a584b62dd..eea9542728ca 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -3982,7 +3982,6 @@ int addrconf_sysctl_forward(ctl_table *ctl, int write, struct file * filp, } static int addrconf_sysctl_forward_strategy(ctl_table *table, - int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen) diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 050e14b7f701..01edac888510 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -834,7 +834,7 @@ static void __net_exit ipv6_cleanup_mibs(struct net *net) snmp_mib_free((void **)net->mib.icmpv6msg_statistics); } -static int inet6_net_init(struct net *net) +static int __net_init inet6_net_init(struct net *net) { int err = 0; diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 840b15780a36..172438320eec 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1199,7 +1199,7 @@ static void ndisc_router_discovery(struct sk_buff *skb) } neigh->flags |= NTF_ROUTER; } else if (rt) { - rt->rt6i_flags |= (rt->rt6i_flags & ~RTF_PREF_MASK) | RTF_PREF(pref); + rt->rt6i_flags = (rt->rt6i_flags & ~RTF_PREF_MASK) | RTF_PREF(pref); } if (rt) @@ -1730,9 +1730,8 @@ int ndisc_ifinfo_sysctl_change(struct ctl_table *ctl, int write, struct file * f return ret; } -int ndisc_ifinfo_sysctl_strategy(ctl_table *ctl, int __user *name, - int nlen, void __user *oldval, - size_t __user *oldlenp, +int ndisc_ifinfo_sysctl_strategy(ctl_table *ctl, + void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen) { struct net_device *dev = ctl->extra1; @@ -1745,13 +1744,11 @@ int ndisc_ifinfo_sysctl_strategy(ctl_table *ctl, int __user *name, switch (ctl->ctl_name) { case NET_NEIGH_REACHABLE_TIME: - ret = sysctl_jiffies(ctl, name, nlen, - oldval, oldlenp, newval, newlen); + ret = sysctl_jiffies(ctl, oldval, oldlenp, newval, newlen); break; case NET_NEIGH_RETRANS_TIME_MS: case NET_NEIGH_REACHABLE_TIME_MS: - ret = sysctl_ms_jiffies(ctl, name, nlen, - oldval, oldlenp, newval, newlen); + ret = sysctl_ms_jiffies(ctl, oldval, oldlenp, newval, newlen); break; default: ret = 0; diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c index 6b29b03925f1..fd5b3a4e3329 100644 --- a/net/ipv6/netfilter.c +++ b/net/ipv6/netfilter.c @@ -12,6 +12,7 @@ int ip6_route_me_harder(struct sk_buff *skb) { + struct net *net = dev_net(skb->dst->dev); struct ipv6hdr *iph = ipv6_hdr(skb); struct dst_entry *dst; struct flowi fl = { @@ -23,7 +24,7 @@ int ip6_route_me_harder(struct sk_buff *skb) .saddr = iph->saddr, } }, }; - dst = ip6_route_output(dev_net(skb->dst->dev), skb->sk, &fl); + dst = ip6_route_output(net, skb->sk, &fl); #ifdef CONFIG_XFRM if (!(IP6CB(skb)->flags & IP6SKB_XFRM_TRANSFORMED) && @@ -33,8 +34,7 @@ int ip6_route_me_harder(struct sk_buff *skb) #endif if (dst->error) { - IP6_INC_STATS(&init_net, ip6_dst_idev(dst), - IPSTATS_MIB_OUTNOROUTES); + IP6_INC_STATS(net, ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES); LIMIT_NETDEBUG(KERN_DEBUG "ip6_route_me_harder: No more route.\n"); dst_release(dst); return -EINVAL; diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c index ec394cf5a19b..676c80b5b14b 100644 --- a/net/ipv6/syncookies.c +++ b/net/ipv6/syncookies.c @@ -204,6 +204,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) req->mss = mss; ireq->rmt_port = th->source; + ireq->loc_port = th->dest; ipv6_addr_copy(&ireq6->rmt_addr, &ipv6_hdr(skb)->saddr); ipv6_addr_copy(&ireq6->loc_addr, &ipv6_hdr(skb)->daddr); if (ipv6_opt_accepted(sk, skb) || diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index e5310c9b84dc..b6b356b7912a 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -476,7 +476,7 @@ static int tcp_v6_send_synack(struct sock *sk, struct request_sock *req) fl.fl6_flowlabel = 0; fl.oif = treq->iif; fl.fl_ip_dport = inet_rsk(req)->rmt_port; - fl.fl_ip_sport = inet_sk(sk)->sport; + fl.fl_ip_sport = inet_rsk(req)->loc_port; security_req_classify_flow(req, &fl); opt = np->opt; @@ -1309,7 +1309,7 @@ static struct sock * tcp_v6_syn_recv_sock(struct sock *sk, struct sk_buff *skb, ipv6_addr_copy(&fl.fl6_src, &treq->loc_addr); fl.oif = sk->sk_bound_dev_if; fl.fl_ip_dport = inet_rsk(req)->rmt_port; - fl.fl_ip_sport = inet_sk(sk)->sport; + fl.fl_ip_sport = inet_rsk(req)->loc_port; security_req_classify_flow(req, &fl); if (ip6_dst_lookup(sk, &dst, &fl)) @@ -1865,7 +1865,7 @@ static void get_openreq6(struct seq_file *seq, i, src->s6_addr32[0], src->s6_addr32[1], src->s6_addr32[2], src->s6_addr32[3], - ntohs(inet_sk(sk)->sport), + ntohs(inet_rsk(req)->loc_port), dest->s6_addr32[0], dest->s6_addr32[1], dest->s6_addr32[2], dest->s6_addr32[3], ntohs(inet_rsk(req)->rmt_port), diff --git a/net/mac80211/debugfs_netdev.c b/net/mac80211/debugfs_netdev.c index 2a4515623776..2ad504fc3414 100644 --- a/net/mac80211/debugfs_netdev.c +++ b/net/mac80211/debugfs_netdev.c @@ -545,8 +545,12 @@ static int netdev_notify(struct notifier_block *nb, sdata = IEEE80211_DEV_TO_SUB_IF(dev); - sprintf(buf, "netdev:%s", dev->name); dir = sdata->debugfsdir; + + if (!dir) + return 0; + + sprintf(buf, "netdev:%s", dev->name); if (!debugfs_rename(dir->d_parent, dir, dir->d_parent, buf)) printk(KERN_ERR "mac80211: debugfs: failed to rename debugfs " "dir to %s\n", buf); diff --git a/net/mac80211/debugfs_sta.c b/net/mac80211/debugfs_sta.c index b9902e425f09..189d0bafa91a 100644 --- a/net/mac80211/debugfs_sta.c +++ b/net/mac80211/debugfs_sta.c @@ -249,11 +249,22 @@ void ieee80211_sta_debugfs_add(struct sta_info *sta) DECLARE_MAC_BUF(mbuf); u8 *mac; + sta->debugfs.add_has_run = true; + if (!stations_dir) return; mac = print_mac(mbuf, sta->sta.addr); + /* + * This might fail due to a race condition: + * When mac80211 unlinks a station, the debugfs entries + * remain, but it is already possible to link a new + * station with the same address which triggers adding + * it to debugfs; therefore, if the old station isn't + * destroyed quickly enough the old station's debugfs + * dir might still be around. + */ sta->debugfs.dir = debugfs_create_dir(mac, stations_dir); if (!sta->debugfs.dir) return; diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h index 8025b294588b..156e42a003ae 100644 --- a/net/mac80211/ieee80211_i.h +++ b/net/mac80211/ieee80211_i.h @@ -816,8 +816,8 @@ struct ieee802_11_elems { u8 *ext_supp_rates; u8 *wmm_info; u8 *wmm_param; - u8 *ht_cap_elem; - u8 *ht_info_elem; + struct ieee80211_ht_cap *ht_cap_elem; + struct ieee80211_ht_addt_info *ht_info_elem; u8 *mesh_config; u8 *mesh_id; u8 *peer_link; @@ -844,8 +844,6 @@ struct ieee802_11_elems { u8 ext_supp_rates_len; u8 wmm_info_len; u8 wmm_param_len; - u8 ht_cap_elem_len; - u8 ht_info_elem_len; u8 mesh_config_len; u8 mesh_id_len; u8 peer_link_len; diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 49f86fa56bff..87665d7bb4f9 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -1348,10 +1348,8 @@ static void ieee80211_rx_mgmt_assoc_resp(struct ieee80211_sub_if_data *sdata, (ifsta->flags & IEEE80211_STA_WMM_ENABLED)) { struct ieee80211_ht_bss_info bss_info; ieee80211_ht_cap_ie_to_ht_info( - (struct ieee80211_ht_cap *) elems.ht_cap_elem, &sta->sta.ht_info); ieee80211_ht_addt_info_ie_to_ht_bss_info( - (struct ieee80211_ht_addt_info *) elems.ht_info_elem, &bss_info); ieee80211_handle_ht(local, 1, &sta->sta.ht_info, &bss_info); } @@ -1709,7 +1707,6 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_sub_if_data *sdata, struct ieee80211_ht_bss_info bss_info; ieee80211_ht_addt_info_ie_to_ht_bss_info( - (struct ieee80211_ht_addt_info *) elems.ht_info_elem, &bss_info); changed |= ieee80211_handle_ht(local, 1, &conf->ht_conf, &bss_info); diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 77e7b014872b..cf6b121e1bbf 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -1379,6 +1379,7 @@ ieee80211_rx_h_amsdu(struct ieee80211_rx_data *rx) return RX_QUEUED; } +#ifdef CONFIG_MAC80211_MESH static ieee80211_rx_result ieee80211_rx_h_mesh_fwding(struct ieee80211_rx_data *rx) { @@ -1453,7 +1454,7 @@ ieee80211_rx_h_mesh_fwding(struct ieee80211_rx_data *rx) else return RX_DROP_MONITOR; } - +#endif static ieee80211_rx_result debug_noinline ieee80211_rx_h_data(struct ieee80211_rx_data *rx) @@ -1780,8 +1781,10 @@ static void ieee80211_invoke_rx_handlers(struct ieee80211_sub_if_data *sdata, /* must be after MMIC verify so header is counted in MPDU mic */ CALL_RXH(ieee80211_rx_h_remove_qos_control) CALL_RXH(ieee80211_rx_h_amsdu) +#ifdef CONFIG_MAC80211_MESH if (ieee80211_vif_is_mesh(&sdata->vif)) CALL_RXH(ieee80211_rx_h_mesh_fwding); +#endif CALL_RXH(ieee80211_rx_h_data) CALL_RXH(ieee80211_rx_h_ctrl) CALL_RXH(ieee80211_rx_h_action) diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index 8e6685e7ae85..416bb41099f3 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -388,7 +388,8 @@ ieee80211_scan_rx(struct ieee80211_sub_if_data *sdata, struct sk_buff *skb, bss = ieee80211_bss_info_update(sdata->local, rx_status, mgmt, skb->len, &elems, freq, beacon); - ieee80211_rx_bss_put(sdata->local, bss); + if (bss) + ieee80211_rx_bss_put(sdata->local, bss); dev_kfree_skb(skb); return RX_QUEUED; diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c index 9b72d15bc8dc..7fef8ea1f5ec 100644 --- a/net/mac80211/sta_info.c +++ b/net/mac80211/sta_info.c @@ -635,7 +635,12 @@ static void sta_info_debugfs_add_work(struct work_struct *work) spin_lock_irqsave(&local->sta_lock, flags); list_for_each_entry(tmp, &local->sta_list, list) { - if (!tmp->debugfs.dir) { + /* + * debugfs.add_has_run will be set by + * ieee80211_sta_debugfs_add regardless + * of what else it does. + */ + if (!tmp->debugfs.add_has_run) { sta = tmp; __sta_info_pin(sta); break; diff --git a/net/mac80211/sta_info.h b/net/mac80211/sta_info.h index a6b51862a89d..168a39a298bd 100644 --- a/net/mac80211/sta_info.h +++ b/net/mac80211/sta_info.h @@ -300,6 +300,7 @@ struct sta_info { struct dentry *inactive_ms; struct dentry *last_seq_ctrl; struct dentry *agg_status; + bool add_has_run; } debugfs; #endif diff --git a/net/mac80211/util.c b/net/mac80211/util.c index f32561ec224c..cee4884b9d06 100644 --- a/net/mac80211/util.c +++ b/net/mac80211/util.c @@ -529,12 +529,12 @@ void ieee802_11_parse_elems(u8 *start, size_t len, elems->ext_supp_rates_len = elen; break; case WLAN_EID_HT_CAPABILITY: - elems->ht_cap_elem = pos; - elems->ht_cap_elem_len = elen; + if (elen >= sizeof(struct ieee80211_ht_cap)) + elems->ht_cap_elem = (void *)pos; break; case WLAN_EID_HT_EXTRA_INFO: - elems->ht_info_elem = pos; - elems->ht_info_elem_len = elen; + if (elen >= sizeof(struct ieee80211_ht_addt_info)) + elems->ht_info_elem = (void *)pos; break; case WLAN_EID_MESH_ID: elems->mesh_id = pos; diff --git a/net/mac80211/wext.c b/net/mac80211/wext.c index 7e0d53abde24..742f811ca416 100644 --- a/net/mac80211/wext.c +++ b/net/mac80211/wext.c @@ -775,7 +775,7 @@ static int ieee80211_ioctl_siwfrag(struct net_device *dev, * configure it here */ if (local->ops->set_frag_threshold) - local->ops->set_frag_threshold( + return local->ops->set_frag_threshold( local_to_hw(local), local->fragmentation_threshold); diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 78892cf2b021..25dcef9f2194 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -271,7 +271,6 @@ config NF_CONNTRACK_TFTP config NF_CT_NETLINK tristate 'Connection tracking netlink interface' select NETFILTER_NETLINK - depends on NF_NAT=n || NF_NAT default m if NETFILTER_ADVANCED=n help This option enables support for a netlink-based userspace interface diff --git a/net/netfilter/ipvs/Kconfig b/net/netfilter/ipvs/Kconfig index 05048e403266..79a698052218 100644 --- a/net/netfilter/ipvs/Kconfig +++ b/net/netfilter/ipvs/Kconfig @@ -25,11 +25,13 @@ menuconfig IP_VS if IP_VS config IP_VS_IPV6 - bool "IPv6 support for IPVS (DANGEROUS)" + bool "IPv6 support for IPVS" depends on EXPERIMENTAL && (IPV6 = y || IP_VS = IPV6) ---help--- Add IPv6 support to IPVS. This is incomplete and might be dangerous. + See http://www.mindbasket.com/ipvs for more information. + Say N if unsure. config IP_VS_DEBUG diff --git a/net/netfilter/nf_conntrack_acct.c b/net/netfilter/nf_conntrack_acct.c index 03591d37b9cc..b92df5c1dfcf 100644 --- a/net/netfilter/nf_conntrack_acct.c +++ b/net/netfilter/nf_conntrack_acct.c @@ -115,7 +115,7 @@ int nf_conntrack_acct_init(struct net *net) if (net_eq(net, &init_net)) { #ifdef CONFIG_NF_CT_ACCT - printk(KERN_WARNING "CONFIG_NF_CT_ACCT is deprecated and will be removed soon. Plase use\n"); + printk(KERN_WARNING "CONFIG_NF_CT_ACCT is deprecated and will be removed soon. Please use\n"); printk(KERN_WARNING "nf_conntrack.acct=1 kernel paramater, acct=1 nf_conntrack module option or\n"); printk(KERN_WARNING "sysctl net.netfilter.nf_conntrack_acct=1 to enable it.\n"); #endif diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 27de3c7b006e..622d7c671cb7 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -38,9 +38,16 @@ #include <net/netfilter/nf_conntrack_core.h> #include <net/netfilter/nf_conntrack_extend.h> #include <net/netfilter/nf_conntrack_acct.h> +#include <net/netfilter/nf_nat.h> #define NF_CONNTRACK_VERSION "0.5.0" +unsigned int +(*nfnetlink_parse_nat_setup_hook)(struct nf_conn *ct, + enum nf_nat_manip_type manip, + struct nlattr *attr) __read_mostly; +EXPORT_SYMBOL_GPL(nfnetlink_parse_nat_setup_hook); + DEFINE_SPINLOCK(nf_conntrack_lock); EXPORT_SYMBOL_GPL(nf_conntrack_lock); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index cadfd15b44f6..a040d46f85d6 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -689,71 +689,6 @@ ctnetlink_parse_tuple(struct nlattr *cda[], struct nf_conntrack_tuple *tuple, return 0; } -#ifdef CONFIG_NF_NAT_NEEDED -static const struct nla_policy protonat_nla_policy[CTA_PROTONAT_MAX+1] = { - [CTA_PROTONAT_PORT_MIN] = { .type = NLA_U16 }, - [CTA_PROTONAT_PORT_MAX] = { .type = NLA_U16 }, -}; - -static int nfnetlink_parse_nat_proto(struct nlattr *attr, - const struct nf_conn *ct, - struct nf_nat_range *range) -{ - struct nlattr *tb[CTA_PROTONAT_MAX+1]; - const struct nf_nat_protocol *npt; - int err; - - err = nla_parse_nested(tb, CTA_PROTONAT_MAX, attr, protonat_nla_policy); - if (err < 0) - return err; - - npt = nf_nat_proto_find_get(nf_ct_protonum(ct)); - if (npt->nlattr_to_range) - err = npt->nlattr_to_range(tb, range); - nf_nat_proto_put(npt); - return err; -} - -static const struct nla_policy nat_nla_policy[CTA_NAT_MAX+1] = { - [CTA_NAT_MINIP] = { .type = NLA_U32 }, - [CTA_NAT_MAXIP] = { .type = NLA_U32 }, -}; - -static inline int -nfnetlink_parse_nat(struct nlattr *nat, - const struct nf_conn *ct, struct nf_nat_range *range) -{ - struct nlattr *tb[CTA_NAT_MAX+1]; - int err; - - memset(range, 0, sizeof(*range)); - - err = nla_parse_nested(tb, CTA_NAT_MAX, nat, nat_nla_policy); - if (err < 0) - return err; - - if (tb[CTA_NAT_MINIP]) - range->min_ip = nla_get_be32(tb[CTA_NAT_MINIP]); - - if (!tb[CTA_NAT_MAXIP]) - range->max_ip = range->min_ip; - else - range->max_ip = nla_get_be32(tb[CTA_NAT_MAXIP]); - - if (range->min_ip) - range->flags |= IP_NAT_RANGE_MAP_IPS; - - if (!tb[CTA_NAT_PROTO]) - return 0; - - err = nfnetlink_parse_nat_proto(tb[CTA_NAT_PROTO], ct, range); - if (err < 0) - return err; - - return 0; -} -#endif - static inline int ctnetlink_parse_help(struct nlattr *attr, char **helper_name) { @@ -878,6 +813,36 @@ out: return err; } +#ifdef CONFIG_NF_NAT_NEEDED +static int +ctnetlink_parse_nat_setup(struct nf_conn *ct, + enum nf_nat_manip_type manip, + struct nlattr *attr) +{ + typeof(nfnetlink_parse_nat_setup_hook) parse_nat_setup; + + parse_nat_setup = rcu_dereference(nfnetlink_parse_nat_setup_hook); + if (!parse_nat_setup) { +#ifdef CONFIG_MODULES + rcu_read_unlock(); + nfnl_unlock(); + if (request_module("nf-nat-ipv4") < 0) { + nfnl_lock(); + rcu_read_lock(); + return -EOPNOTSUPP; + } + nfnl_lock(); + rcu_read_lock(); + if (nfnetlink_parse_nat_setup_hook) + return -EAGAIN; +#endif + return -EOPNOTSUPP; + } + + return parse_nat_setup(ct, manip, attr); +} +#endif + static int ctnetlink_change_status(struct nf_conn *ct, struct nlattr *cda[]) { @@ -897,31 +862,6 @@ ctnetlink_change_status(struct nf_conn *ct, struct nlattr *cda[]) /* ASSURED bit can only be set */ return -EBUSY; - if (cda[CTA_NAT_SRC] || cda[CTA_NAT_DST]) { -#ifndef CONFIG_NF_NAT_NEEDED - return -EOPNOTSUPP; -#else - struct nf_nat_range range; - - if (cda[CTA_NAT_DST]) { - if (nfnetlink_parse_nat(cda[CTA_NAT_DST], ct, - &range) < 0) - return -EINVAL; - if (nf_nat_initialized(ct, IP_NAT_MANIP_DST)) - return -EEXIST; - nf_nat_setup_info(ct, &range, IP_NAT_MANIP_DST); - } - if (cda[CTA_NAT_SRC]) { - if (nfnetlink_parse_nat(cda[CTA_NAT_SRC], ct, - &range) < 0) - return -EINVAL; - if (nf_nat_initialized(ct, IP_NAT_MANIP_SRC)) - return -EEXIST; - nf_nat_setup_info(ct, &range, IP_NAT_MANIP_SRC); - } -#endif - } - /* Be careful here, modifying NAT bits can screw up things, * so don't let users modify them directly if they don't pass * nf_nat_range. */ @@ -929,6 +869,31 @@ ctnetlink_change_status(struct nf_conn *ct, struct nlattr *cda[]) return 0; } +static int +ctnetlink_change_nat(struct nf_conn *ct, struct nlattr *cda[]) +{ +#ifdef CONFIG_NF_NAT_NEEDED + int ret; + + if (cda[CTA_NAT_DST]) { + ret = ctnetlink_parse_nat_setup(ct, + IP_NAT_MANIP_DST, + cda[CTA_NAT_DST]); + if (ret < 0) + return ret; + } + if (cda[CTA_NAT_SRC]) { + ret = ctnetlink_parse_nat_setup(ct, + IP_NAT_MANIP_SRC, + cda[CTA_NAT_SRC]); + if (ret < 0) + return ret; + } + return 0; +#else + return -EOPNOTSUPP; +#endif +} static inline int ctnetlink_change_helper(struct nf_conn *ct, struct nlattr *cda[]) @@ -1157,6 +1122,14 @@ ctnetlink_create_conntrack(struct nlattr *cda[], } } + if (cda[CTA_NAT_SRC] || cda[CTA_NAT_DST]) { + err = ctnetlink_change_nat(ct, cda); + if (err < 0) { + rcu_read_unlock(); + goto err; + } + } + if (cda[CTA_PROTOINFO]) { err = ctnetlink_change_protoinfo(ct, cda); if (err < 0) { diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c index 373e51e91ce5..1bc3001d1827 100644 --- a/net/netfilter/nf_conntrack_pptp.c +++ b/net/netfilter/nf_conntrack_pptp.c @@ -65,7 +65,7 @@ void struct nf_conntrack_expect *exp) __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_pptp_hook_expectfn); -#ifdef DEBUG +#if defined(DEBUG) || defined(CONFIG_DYNAMIC_PRINTK_DEBUG) /* PptpControlMessageType names */ const char *const pptp_msg_name[] = { "UNKNOWN_MESSAGE", diff --git a/net/netfilter/nf_tproxy_core.c b/net/netfilter/nf_tproxy_core.c index fe34f4bf74cc..cdc97f3105a3 100644 --- a/net/netfilter/nf_tproxy_core.c +++ b/net/netfilter/nf_tproxy_core.c @@ -10,7 +10,6 @@ * */ -#include <linux/version.h> #include <linux/module.h> #include <linux/net.h> diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c index b75c9c4a995d..9c0ba17a1ddb 100644 --- a/net/netfilter/nfnetlink.c +++ b/net/netfilter/nfnetlink.c @@ -44,15 +44,17 @@ static struct sock *nfnl = NULL; static const struct nfnetlink_subsystem *subsys_table[NFNL_SUBSYS_COUNT]; static DEFINE_MUTEX(nfnl_mutex); -static inline void nfnl_lock(void) +void nfnl_lock(void) { mutex_lock(&nfnl_mutex); } +EXPORT_SYMBOL_GPL(nfnl_lock); -static inline void nfnl_unlock(void) +void nfnl_unlock(void) { mutex_unlock(&nfnl_mutex); } +EXPORT_SYMBOL_GPL(nfnl_unlock); int nfnetlink_subsys_register(const struct nfnetlink_subsystem *n) { @@ -132,9 +134,10 @@ static int nfnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) return 0; type = nlh->nlmsg_type; +replay: ss = nfnetlink_get_subsys(type); if (!ss) { -#ifdef CONFIG_KMOD +#ifdef CONFIG_MODULES nfnl_unlock(); request_module("nfnetlink-subsys-%d", NFNL_SUBSYS_ID(type)); nfnl_lock(); @@ -165,7 +168,10 @@ static int nfnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) } else return -EINVAL; - return nc->call(nfnl, skb, nlh, cda); + err = nc->call(nfnl, skb, nlh, cda); + if (err == -EAGAIN) + goto replay; + return err; } } diff --git a/net/netfilter/xt_NFQUEUE.c b/net/netfilter/xt_NFQUEUE.c index 2cc1fff49307..f9977b3311f7 100644 --- a/net/netfilter/xt_NFQUEUE.c +++ b/net/netfilter/xt_NFQUEUE.c @@ -48,7 +48,7 @@ static struct xt_target nfqueue_tg_reg[] __read_mostly = { }, { .name = "NFQUEUE", - .family = NF_ARP, + .family = NFPROTO_ARP, .target = nfqueue_tg, .targetsize = sizeof(struct xt_NFQ_info), .me = THIS_MODULE, diff --git a/net/netfilter/xt_iprange.c b/net/netfilter/xt_iprange.c index 6f62c36948d9..7ac54eab0b00 100644 --- a/net/netfilter/xt_iprange.c +++ b/net/netfilter/xt_iprange.c @@ -61,7 +61,7 @@ iprange_mt4(const struct sk_buff *skb, const struct xt_match_param *par) if (info->flags & IPRANGE_SRC) { m = ntohl(iph->saddr) < ntohl(info->src_min.ip); m |= ntohl(iph->saddr) > ntohl(info->src_max.ip); - m ^= info->flags & IPRANGE_SRC_INV; + m ^= !!(info->flags & IPRANGE_SRC_INV); if (m) { pr_debug("src IP " NIPQUAD_FMT " NOT in range %s" NIPQUAD_FMT "-" NIPQUAD_FMT "\n", @@ -75,7 +75,7 @@ iprange_mt4(const struct sk_buff *skb, const struct xt_match_param *par) if (info->flags & IPRANGE_DST) { m = ntohl(iph->daddr) < ntohl(info->dst_min.ip); m |= ntohl(iph->daddr) > ntohl(info->dst_max.ip); - m ^= info->flags & IPRANGE_DST_INV; + m ^= !!(info->flags & IPRANGE_DST_INV); if (m) { pr_debug("dst IP " NIPQUAD_FMT " NOT in range %s" NIPQUAD_FMT "-" NIPQUAD_FMT "\n", @@ -114,14 +114,14 @@ iprange_mt6(const struct sk_buff *skb, const struct xt_match_param *par) if (info->flags & IPRANGE_SRC) { m = iprange_ipv6_sub(&iph->saddr, &info->src_min.in6) < 0; m |= iprange_ipv6_sub(&iph->saddr, &info->src_max.in6) > 0; - m ^= info->flags & IPRANGE_SRC_INV; + m ^= !!(info->flags & IPRANGE_SRC_INV); if (m) return false; } if (info->flags & IPRANGE_DST) { m = iprange_ipv6_sub(&iph->daddr, &info->dst_min.in6) < 0; m |= iprange_ipv6_sub(&iph->daddr, &info->dst_max.in6) > 0; - m ^= info->flags & IPRANGE_DST_INV; + m ^= !!(info->flags & IPRANGE_DST_INV); if (m) return false; } diff --git a/net/netfilter/xt_recent.c b/net/netfilter/xt_recent.c index 4ebd4ca9a991..280c471bcdf4 100644 --- a/net/netfilter/xt_recent.c +++ b/net/netfilter/xt_recent.c @@ -318,15 +318,15 @@ static bool recent_mt_check(const struct xt_mtchk_param *par) for (i = 0; i < ip_list_hash_size; i++) INIT_LIST_HEAD(&t->iphash[i]); #ifdef CONFIG_PROC_FS - t->proc = proc_create(t->name, ip_list_perms, recent_proc_dir, - &recent_mt_fops); + t->proc = proc_create_data(t->name, ip_list_perms, recent_proc_dir, + &recent_mt_fops, t); if (t->proc == NULL) { kfree(t); goto out; } #ifdef CONFIG_NETFILTER_XT_MATCH_RECENT_PROC_COMPAT - t->proc_old = proc_create(t->name, ip_list_perms, proc_old_dir, - &recent_old_fops); + t->proc_old = proc_create_data(t->name, ip_list_perms, proc_old_dir, + &recent_old_fops, t); if (t->proc_old == NULL) { remove_proc_entry(t->name, proc_old_dir); kfree(t); @@ -334,11 +334,9 @@ static bool recent_mt_check(const struct xt_mtchk_param *par) } t->proc_old->uid = ip_list_uid; t->proc_old->gid = ip_list_gid; - t->proc_old->data = t; #endif t->proc->uid = ip_list_uid; t->proc->gid = ip_list_gid; - t->proc->data = t; #endif spin_lock_bh(&recent_lock); list_add_tail(&t->list, &tables); diff --git a/net/netlabel/Makefile b/net/netlabel/Makefile index 8af18c0a47d9..ea750e9df65f 100644 --- a/net/netlabel/Makefile +++ b/net/netlabel/Makefile @@ -5,7 +5,8 @@ # # base objects -obj-y := netlabel_user.o netlabel_kapi.o netlabel_domainhash.o +obj-y := netlabel_user.o netlabel_kapi.o +obj-y += netlabel_domainhash.o netlabel_addrlist.o # management objects obj-y += netlabel_mgmt.o diff --git a/net/netlabel/netlabel_addrlist.c b/net/netlabel/netlabel_addrlist.c new file mode 100644 index 000000000000..b0925a303353 --- /dev/null +++ b/net/netlabel/netlabel_addrlist.c @@ -0,0 +1,388 @@ +/* + * NetLabel Network Address Lists + * + * This file contains network address list functions used to manage ordered + * lists of network addresses for use by the NetLabel subsystem. The NetLabel + * system manages static and dynamic label mappings for network protocols such + * as CIPSO and RIPSO. + * + * Author: Paul Moore <paul.moore@hp.com> + * + */ + +/* + * (c) Copyright Hewlett-Packard Development Company, L.P., 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See + * the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#include <linux/types.h> +#include <linux/rcupdate.h> +#include <linux/list.h> +#include <linux/spinlock.h> +#include <linux/in.h> +#include <linux/in6.h> +#include <linux/ip.h> +#include <linux/ipv6.h> +#include <net/ip.h> +#include <net/ipv6.h> +#include <linux/audit.h> + +#include "netlabel_addrlist.h" + +/* + * Address List Functions + */ + +/** + * netlbl_af4list_search - Search for a matching IPv4 address entry + * @addr: IPv4 address + * @head: the list head + * + * Description: + * Searches the IPv4 address list given by @head. If a matching address entry + * is found it is returned, otherwise NULL is returned. The caller is + * responsible for calling the rcu_read_[un]lock() functions. + * + */ +struct netlbl_af4list *netlbl_af4list_search(__be32 addr, + struct list_head *head) +{ + struct netlbl_af4list *iter; + + list_for_each_entry_rcu(iter, head, list) + if (iter->valid && (addr & iter->mask) == iter->addr) + return iter; + + return NULL; +} + +/** + * netlbl_af4list_search_exact - Search for an exact IPv4 address entry + * @addr: IPv4 address + * @mask: IPv4 address mask + * @head: the list head + * + * Description: + * Searches the IPv4 address list given by @head. If an exact match if found + * it is returned, otherwise NULL is returned. The caller is responsible for + * calling the rcu_read_[un]lock() functions. + * + */ +struct netlbl_af4list *netlbl_af4list_search_exact(__be32 addr, + __be32 mask, + struct list_head *head) +{ + struct netlbl_af4list *iter; + + list_for_each_entry_rcu(iter, head, list) + if (iter->valid && iter->addr == addr && iter->mask == mask) + return iter; + + return NULL; +} + + +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) +/** + * netlbl_af6list_search - Search for a matching IPv6 address entry + * @addr: IPv6 address + * @head: the list head + * + * Description: + * Searches the IPv6 address list given by @head. If a matching address entry + * is found it is returned, otherwise NULL is returned. The caller is + * responsible for calling the rcu_read_[un]lock() functions. + * + */ +struct netlbl_af6list *netlbl_af6list_search(const struct in6_addr *addr, + struct list_head *head) +{ + struct netlbl_af6list *iter; + + list_for_each_entry_rcu(iter, head, list) + if (iter->valid && + ipv6_masked_addr_cmp(&iter->addr, &iter->mask, addr) == 0) + return iter; + + return NULL; +} + +/** + * netlbl_af6list_search_exact - Search for an exact IPv6 address entry + * @addr: IPv6 address + * @mask: IPv6 address mask + * @head: the list head + * + * Description: + * Searches the IPv6 address list given by @head. If an exact match if found + * it is returned, otherwise NULL is returned. The caller is responsible for + * calling the rcu_read_[un]lock() functions. + * + */ +struct netlbl_af6list *netlbl_af6list_search_exact(const struct in6_addr *addr, + const struct in6_addr *mask, + struct list_head *head) +{ + struct netlbl_af6list *iter; + + list_for_each_entry_rcu(iter, head, list) + if (iter->valid && + ipv6_addr_equal(&iter->addr, addr) && + ipv6_addr_equal(&iter->mask, mask)) + return iter; + + return NULL; +} +#endif /* IPv6 */ + +/** + * netlbl_af4list_add - Add a new IPv4 address entry to a list + * @entry: address entry + * @head: the list head + * + * Description: + * Add a new address entry to the list pointed to by @head. On success zero is + * returned, otherwise a negative value is returned. The caller is responsible + * for calling the necessary locking functions. + * + */ +int netlbl_af4list_add(struct netlbl_af4list *entry, struct list_head *head) +{ + struct netlbl_af4list *iter; + + iter = netlbl_af4list_search(entry->addr, head); + if (iter != NULL && + iter->addr == entry->addr && iter->mask == entry->mask) + return -EEXIST; + + /* in order to speed up address searches through the list (the common + * case) we need to keep the list in order based on the size of the + * address mask such that the entry with the widest mask (smallest + * numerical value) appears first in the list */ + list_for_each_entry_rcu(iter, head, list) + if (iter->valid && + ntohl(entry->mask) > ntohl(iter->mask)) { + __list_add_rcu(&entry->list, + iter->list.prev, + &iter->list); + return 0; + } + list_add_tail_rcu(&entry->list, head); + return 0; +} + +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) +/** + * netlbl_af6list_add - Add a new IPv6 address entry to a list + * @entry: address entry + * @head: the list head + * + * Description: + * Add a new address entry to the list pointed to by @head. On success zero is + * returned, otherwise a negative value is returned. The caller is responsible + * for calling the necessary locking functions. + * + */ +int netlbl_af6list_add(struct netlbl_af6list *entry, struct list_head *head) +{ + struct netlbl_af6list *iter; + + iter = netlbl_af6list_search(&entry->addr, head); + if (iter != NULL && + ipv6_addr_equal(&iter->addr, &entry->addr) && + ipv6_addr_equal(&iter->mask, &entry->mask)) + return -EEXIST; + + /* in order to speed up address searches through the list (the common + * case) we need to keep the list in order based on the size of the + * address mask such that the entry with the widest mask (smallest + * numerical value) appears first in the list */ + list_for_each_entry_rcu(iter, head, list) + if (iter->valid && + ipv6_addr_cmp(&entry->mask, &iter->mask) > 0) { + __list_add_rcu(&entry->list, + iter->list.prev, + &iter->list); + return 0; + } + list_add_tail_rcu(&entry->list, head); + return 0; +} +#endif /* IPv6 */ + +/** + * netlbl_af4list_remove_entry - Remove an IPv4 address entry + * @entry: address entry + * + * Description: + * Remove the specified IP address entry. The caller is responsible for + * calling the necessary locking functions. + * + */ +void netlbl_af4list_remove_entry(struct netlbl_af4list *entry) +{ + entry->valid = 0; + list_del_rcu(&entry->list); +} + +/** + * netlbl_af4list_remove - Remove an IPv4 address entry + * @addr: IP address + * @mask: IP address mask + * @head: the list head + * + * Description: + * Remove an IP address entry from the list pointed to by @head. Returns the + * entry on success, NULL on failure. The caller is responsible for calling + * the necessary locking functions. + * + */ +struct netlbl_af4list *netlbl_af4list_remove(__be32 addr, __be32 mask, + struct list_head *head) +{ + struct netlbl_af4list *entry; + + entry = netlbl_af4list_search(addr, head); + if (entry != NULL && entry->addr == addr && entry->mask == mask) { + netlbl_af4list_remove_entry(entry); + return entry; + } + + return NULL; +} + +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) +/** + * netlbl_af6list_remove_entry - Remove an IPv6 address entry + * @entry: address entry + * + * Description: + * Remove the specified IP address entry. The caller is responsible for + * calling the necessary locking functions. + * + */ +void netlbl_af6list_remove_entry(struct netlbl_af6list *entry) +{ + entry->valid = 0; + list_del_rcu(&entry->list); +} + +/** + * netlbl_af6list_remove - Remove an IPv6 address entry + * @addr: IP address + * @mask: IP address mask + * @head: the list head + * + * Description: + * Remove an IP address entry from the list pointed to by @head. Returns the + * entry on success, NULL on failure. The caller is responsible for calling + * the necessary locking functions. + * + */ +struct netlbl_af6list *netlbl_af6list_remove(const struct in6_addr *addr, + const struct in6_addr *mask, + struct list_head *head) +{ + struct netlbl_af6list *entry; + + entry = netlbl_af6list_search(addr, head); + if (entry != NULL && + ipv6_addr_equal(&entry->addr, addr) && + ipv6_addr_equal(&entry->mask, mask)) { + netlbl_af6list_remove_entry(entry); + return entry; + } + + return NULL; +} +#endif /* IPv6 */ + +/* + * Audit Helper Functions + */ + +/** + * netlbl_af4list_audit_addr - Audit an IPv4 address + * @audit_buf: audit buffer + * @src: true if source address, false if destination + * @dev: network interface + * @addr: IP address + * @mask: IP address mask + * + * Description: + * Write the IPv4 address and address mask, if necessary, to @audit_buf. + * + */ +void netlbl_af4list_audit_addr(struct audit_buffer *audit_buf, + int src, const char *dev, + __be32 addr, __be32 mask) +{ + u32 mask_val = ntohl(mask); + char *dir = (src ? "src" : "dst"); + + if (dev != NULL) + audit_log_format(audit_buf, " netif=%s", dev); + audit_log_format(audit_buf, " %s=" NIPQUAD_FMT, dir, NIPQUAD(addr)); + if (mask_val != 0xffffffff) { + u32 mask_len = 0; + while (mask_val > 0) { + mask_val <<= 1; + mask_len++; + } + audit_log_format(audit_buf, " %s_prefixlen=%d", dir, mask_len); + } +} + +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) +/** + * netlbl_af6list_audit_addr - Audit an IPv6 address + * @audit_buf: audit buffer + * @src: true if source address, false if destination + * @dev: network interface + * @addr: IP address + * @mask: IP address mask + * + * Description: + * Write the IPv6 address and address mask, if necessary, to @audit_buf. + * + */ +void netlbl_af6list_audit_addr(struct audit_buffer *audit_buf, + int src, + const char *dev, + const struct in6_addr *addr, + const struct in6_addr *mask) +{ + char *dir = (src ? "src" : "dst"); + + if (dev != NULL) + audit_log_format(audit_buf, " netif=%s", dev); + audit_log_format(audit_buf, " %s=" NIP6_FMT, dir, NIP6(*addr)); + if (ntohl(mask->s6_addr32[3]) != 0xffffffff) { + u32 mask_len = 0; + u32 mask_val; + int iter = -1; + while (ntohl(mask->s6_addr32[++iter]) == 0xffffffff) + mask_len += 32; + mask_val = ntohl(mask->s6_addr32[iter]); + while (mask_val > 0) { + mask_val <<= 1; + mask_len++; + } + audit_log_format(audit_buf, " %s_prefixlen=%d", dir, mask_len); + } +} +#endif /* IPv6 */ diff --git a/net/netlabel/netlabel_addrlist.h b/net/netlabel/netlabel_addrlist.h new file mode 100644 index 000000000000..0242bead405f --- /dev/null +++ b/net/netlabel/netlabel_addrlist.h @@ -0,0 +1,189 @@ +/* + * NetLabel Network Address Lists + * + * This file contains network address list functions used to manage ordered + * lists of network addresses for use by the NetLabel subsystem. The NetLabel + * system manages static and dynamic label mappings for network protocols such + * as CIPSO and RIPSO. + * + * Author: Paul Moore <paul.moore@hp.com> + * + */ + +/* + * (c) Copyright Hewlett-Packard Development Company, L.P., 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See + * the GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + */ + +#ifndef _NETLABEL_ADDRLIST_H +#define _NETLABEL_ADDRLIST_H + +#include <linux/types.h> +#include <linux/rcupdate.h> +#include <linux/list.h> +#include <linux/in6.h> +#include <linux/audit.h> + +/** + * struct netlbl_af4list - NetLabel IPv4 address list + * @addr: IPv4 address + * @mask: IPv4 address mask + * @valid: valid flag + * @list: list structure, used internally + */ +struct netlbl_af4list { + __be32 addr; + __be32 mask; + + u32 valid; + struct list_head list; +}; + +/** + * struct netlbl_af6list - NetLabel IPv6 address list + * @addr: IPv6 address + * @mask: IPv6 address mask + * @valid: valid flag + * @list: list structure, used internally + */ +struct netlbl_af6list { + struct in6_addr addr; + struct in6_addr mask; + + u32 valid; + struct list_head list; +}; + +#define __af4list_entry(ptr) container_of(ptr, struct netlbl_af4list, list) + +static inline struct netlbl_af4list *__af4list_valid(struct list_head *s, + struct list_head *h) +{ + struct list_head *i = s; + struct netlbl_af4list *n = __af4list_entry(s); + while (i != h && !n->valid) { + i = i->next; + n = __af4list_entry(i); + } + return n; +} + +static inline struct netlbl_af4list *__af4list_valid_rcu(struct list_head *s, + struct list_head *h) +{ + struct list_head *i = s; + struct netlbl_af4list *n = __af4list_entry(s); + while (i != h && !n->valid) { + i = rcu_dereference(i->next); + n = __af4list_entry(i); + } + return n; +} + +#define netlbl_af4list_foreach(iter, head) \ + for (iter = __af4list_valid((head)->next, head); \ + prefetch(iter->list.next), &iter->list != (head); \ + iter = __af4list_valid(iter->list.next, head)) + +#define netlbl_af4list_foreach_rcu(iter, head) \ + for (iter = __af4list_valid_rcu((head)->next, head); \ + prefetch(iter->list.next), &iter->list != (head); \ + iter = __af4list_valid_rcu(iter->list.next, head)) + +#define netlbl_af4list_foreach_safe(iter, tmp, head) \ + for (iter = __af4list_valid((head)->next, head), \ + tmp = __af4list_valid(iter->list.next, head); \ + &iter->list != (head); \ + iter = tmp, tmp = __af4list_valid(iter->list.next, head)) + +int netlbl_af4list_add(struct netlbl_af4list *entry, + struct list_head *head); +struct netlbl_af4list *netlbl_af4list_remove(__be32 addr, __be32 mask, + struct list_head *head); +void netlbl_af4list_remove_entry(struct netlbl_af4list *entry); +struct netlbl_af4list *netlbl_af4list_search(__be32 addr, + struct list_head *head); +struct netlbl_af4list *netlbl_af4list_search_exact(__be32 addr, + __be32 mask, + struct list_head *head); +void netlbl_af4list_audit_addr(struct audit_buffer *audit_buf, + int src, const char *dev, + __be32 addr, __be32 mask); + +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + +#define __af6list_entry(ptr) container_of(ptr, struct netlbl_af6list, list) + +static inline struct netlbl_af6list *__af6list_valid(struct list_head *s, + struct list_head *h) +{ + struct list_head *i = s; + struct netlbl_af6list *n = __af6list_entry(s); + while (i != h && !n->valid) { + i = i->next; + n = __af6list_entry(i); + } + return n; +} + +static inline struct netlbl_af6list *__af6list_valid_rcu(struct list_head *s, + struct list_head *h) +{ + struct list_head *i = s; + struct netlbl_af6list *n = __af6list_entry(s); + while (i != h && !n->valid) { + i = rcu_dereference(i->next); + n = __af6list_entry(i); + } + return n; +} + +#define netlbl_af6list_foreach(iter, head) \ + for (iter = __af6list_valid((head)->next, head); \ + prefetch(iter->list.next), &iter->list != (head); \ + iter = __af6list_valid(iter->list.next, head)) + +#define netlbl_af6list_foreach_rcu(iter, head) \ + for (iter = __af6list_valid_rcu((head)->next, head); \ + prefetch(iter->list.next), &iter->list != (head); \ + iter = __af6list_valid_rcu(iter->list.next, head)) + +#define netlbl_af6list_foreach_safe(iter, tmp, head) \ + for (iter = __af6list_valid((head)->next, head), \ + tmp = __af6list_valid(iter->list.next, head); \ + &iter->list != (head); \ + iter = tmp, tmp = __af6list_valid(iter->list.next, head)) + +int netlbl_af6list_add(struct netlbl_af6list *entry, + struct list_head *head); +struct netlbl_af6list *netlbl_af6list_remove(const struct in6_addr *addr, + const struct in6_addr *mask, + struct list_head *head); +void netlbl_af6list_remove_entry(struct netlbl_af6list *entry); +struct netlbl_af6list *netlbl_af6list_search(const struct in6_addr *addr, + struct list_head *head); +struct netlbl_af6list *netlbl_af6list_search_exact(const struct in6_addr *addr, + const struct in6_addr *mask, + struct list_head *head); +void netlbl_af6list_audit_addr(struct audit_buffer *audit_buf, + int src, + const char *dev, + const struct in6_addr *addr, + const struct in6_addr *mask); +#endif /* IPV6 */ + +#endif diff --git a/net/netlabel/netlabel_cipso_v4.c b/net/netlabel/netlabel_cipso_v4.c index 0aec318bf0ef..fff32b70efa9 100644 --- a/net/netlabel/netlabel_cipso_v4.c +++ b/net/netlabel/netlabel_cipso_v4.c @@ -43,6 +43,7 @@ #include "netlabel_user.h" #include "netlabel_cipso_v4.h" #include "netlabel_mgmt.h" +#include "netlabel_domainhash.h" /* Argument struct for cipso_v4_doi_walk() */ struct netlbl_cipsov4_doiwalk_arg { @@ -51,6 +52,12 @@ struct netlbl_cipsov4_doiwalk_arg { u32 seq; }; +/* Argument struct for netlbl_domhsh_walk() */ +struct netlbl_domhsh_walk_arg { + struct netlbl_audit *audit_info; + u32 doi; +}; + /* NetLabel Generic NETLINK CIPSOv4 family */ static struct genl_family netlbl_cipsov4_gnl_family = { .id = GENL_ID_GENERATE, @@ -81,32 +88,6 @@ static const struct nla_policy netlbl_cipsov4_genl_policy[NLBL_CIPSOV4_A_MAX + 1 */ /** - * netlbl_cipsov4_doi_free - Frees a CIPSO V4 DOI definition - * @entry: the entry's RCU field - * - * Description: - * This function is designed to be used as a callback to the call_rcu() - * function so that the memory allocated to the DOI definition can be released - * safely. - * - */ -void netlbl_cipsov4_doi_free(struct rcu_head *entry) -{ - struct cipso_v4_doi *ptr; - - ptr = container_of(entry, struct cipso_v4_doi, rcu); - switch (ptr->type) { - case CIPSO_V4_MAP_STD: - kfree(ptr->map.std->lvl.cipso); - kfree(ptr->map.std->lvl.local); - kfree(ptr->map.std->cat.cipso); - kfree(ptr->map.std->cat.local); - break; - } - kfree(ptr); -} - -/** * netlbl_cipsov4_add_common - Parse the common sections of a ADD message * @info: the Generic NETLINK info block * @doi_def: the CIPSO V4 DOI definition @@ -151,9 +132,9 @@ static int netlbl_cipsov4_add_common(struct genl_info *info, * @info: the Generic NETLINK info block * * Description: - * Create a new CIPSO_V4_MAP_STD DOI definition based on the given ADD message - * and add it to the CIPSO V4 engine. Return zero on success and non-zero on - * error. + * Create a new CIPSO_V4_MAP_TRANS DOI definition based on the given ADD + * message and add it to the CIPSO V4 engine. Return zero on success and + * non-zero on error. * */ static int netlbl_cipsov4_add_std(struct genl_info *info) @@ -183,7 +164,7 @@ static int netlbl_cipsov4_add_std(struct genl_info *info) ret_val = -ENOMEM; goto add_std_failure; } - doi_def->type = CIPSO_V4_MAP_STD; + doi_def->type = CIPSO_V4_MAP_TRANS; ret_val = netlbl_cipsov4_add_common(info, doi_def); if (ret_val != 0) @@ -342,7 +323,7 @@ static int netlbl_cipsov4_add_std(struct genl_info *info) add_std_failure: if (doi_def) - netlbl_cipsov4_doi_free(&doi_def->rcu); + cipso_v4_doi_free(doi_def); return ret_val; } @@ -379,7 +360,44 @@ static int netlbl_cipsov4_add_pass(struct genl_info *info) return 0; add_pass_failure: - netlbl_cipsov4_doi_free(&doi_def->rcu); + cipso_v4_doi_free(doi_def); + return ret_val; +} + +/** + * netlbl_cipsov4_add_local - Adds a CIPSO V4 DOI definition + * @info: the Generic NETLINK info block + * + * Description: + * Create a new CIPSO_V4_MAP_LOCAL DOI definition based on the given ADD + * message and add it to the CIPSO V4 engine. Return zero on success and + * non-zero on error. + * + */ +static int netlbl_cipsov4_add_local(struct genl_info *info) +{ + int ret_val; + struct cipso_v4_doi *doi_def = NULL; + + if (!info->attrs[NLBL_CIPSOV4_A_TAGLST]) + return -EINVAL; + + doi_def = kmalloc(sizeof(*doi_def), GFP_KERNEL); + if (doi_def == NULL) + return -ENOMEM; + doi_def->type = CIPSO_V4_MAP_LOCAL; + + ret_val = netlbl_cipsov4_add_common(info, doi_def); + if (ret_val != 0) + goto add_local_failure; + + ret_val = cipso_v4_doi_add(doi_def); + if (ret_val != 0) + goto add_local_failure; + return 0; + +add_local_failure: + cipso_v4_doi_free(doi_def); return ret_val; } @@ -412,14 +430,18 @@ static int netlbl_cipsov4_add(struct sk_buff *skb, struct genl_info *info) type = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_MTYPE]); switch (type) { - case CIPSO_V4_MAP_STD: - type_str = "std"; + case CIPSO_V4_MAP_TRANS: + type_str = "trans"; ret_val = netlbl_cipsov4_add_std(info); break; case CIPSO_V4_MAP_PASS: type_str = "pass"; ret_val = netlbl_cipsov4_add_pass(info); break; + case CIPSO_V4_MAP_LOCAL: + type_str = "local"; + ret_val = netlbl_cipsov4_add_local(info); + break; } if (ret_val == 0) atomic_inc(&netlabel_mgmt_protocount); @@ -491,7 +513,7 @@ list_start: doi_def = cipso_v4_doi_getdef(doi); if (doi_def == NULL) { ret_val = -EINVAL; - goto list_failure; + goto list_failure_lock; } ret_val = nla_put_u32(ans_skb, NLBL_CIPSOV4_A_MTYPE, doi_def->type); @@ -516,7 +538,7 @@ list_start: nla_nest_end(ans_skb, nla_a); switch (doi_def->type) { - case CIPSO_V4_MAP_STD: + case CIPSO_V4_MAP_TRANS: nla_a = nla_nest_start(ans_skb, NLBL_CIPSOV4_A_MLSLVLLST); if (nla_a == NULL) { ret_val = -ENOMEM; @@ -655,7 +677,7 @@ static int netlbl_cipsov4_listall(struct sk_buff *skb, struct netlink_callback *cb) { struct netlbl_cipsov4_doiwalk_arg cb_arg; - int doi_skip = cb->args[0]; + u32 doi_skip = cb->args[0]; cb_arg.nl_cb = cb; cb_arg.skb = skb; @@ -668,6 +690,29 @@ static int netlbl_cipsov4_listall(struct sk_buff *skb, } /** + * netlbl_cipsov4_remove_cb - netlbl_cipsov4_remove() callback for REMOVE + * @entry: LSM domain mapping entry + * @arg: the netlbl_domhsh_walk_arg structure + * + * Description: + * This function is intended for use by netlbl_cipsov4_remove() as the callback + * for the netlbl_domhsh_walk() function; it removes LSM domain map entries + * which are associated with the CIPSO DOI specified in @arg. Returns zero on + * success, negative values on failure. + * + */ +static int netlbl_cipsov4_remove_cb(struct netlbl_dom_map *entry, void *arg) +{ + struct netlbl_domhsh_walk_arg *cb_arg = arg; + + if (entry->type == NETLBL_NLTYPE_CIPSOV4 && + entry->type_def.cipsov4->doi == cb_arg->doi) + return netlbl_domhsh_remove_entry(entry, cb_arg->audit_info); + + return 0; +} + +/** * netlbl_cipsov4_remove - Handle a REMOVE message * @skb: the NETLINK buffer * @info: the Generic NETLINK info block @@ -681,8 +726,11 @@ static int netlbl_cipsov4_remove(struct sk_buff *skb, struct genl_info *info) { int ret_val = -EINVAL; u32 doi = 0; + struct netlbl_domhsh_walk_arg cb_arg; struct audit_buffer *audit_buf; struct netlbl_audit audit_info; + u32 skip_bkt = 0; + u32 skip_chain = 0; if (!info->attrs[NLBL_CIPSOV4_A_DOI]) return -EINVAL; @@ -690,11 +738,15 @@ static int netlbl_cipsov4_remove(struct sk_buff *skb, struct genl_info *info) doi = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_DOI]); netlbl_netlink_auditinfo(skb, &audit_info); - ret_val = cipso_v4_doi_remove(doi, - &audit_info, - netlbl_cipsov4_doi_free); - if (ret_val == 0) - atomic_dec(&netlabel_mgmt_protocount); + cb_arg.doi = doi; + cb_arg.audit_info = &audit_info; + ret_val = netlbl_domhsh_walk(&skip_bkt, &skip_chain, + netlbl_cipsov4_remove_cb, &cb_arg); + if (ret_val == 0 || ret_val == -ENOENT) { + ret_val = cipso_v4_doi_remove(doi, &audit_info); + if (ret_val == 0) + atomic_dec(&netlabel_mgmt_protocount); + } audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_DEL, &audit_info); diff --git a/net/netlabel/netlabel_cipso_v4.h b/net/netlabel/netlabel_cipso_v4.h index 220cb9d06b49..c8a4079261f0 100644 --- a/net/netlabel/netlabel_cipso_v4.h +++ b/net/netlabel/netlabel_cipso_v4.h @@ -45,12 +45,13 @@ * NLBL_CIPSOV4_A_MTYPE * NLBL_CIPSOV4_A_TAGLST * - * If using CIPSO_V4_MAP_STD the following attributes are required: + * If using CIPSO_V4_MAP_TRANS the following attributes are required: * * NLBL_CIPSOV4_A_MLSLVLLST * NLBL_CIPSOV4_A_MLSCATLST * - * If using CIPSO_V4_MAP_PASS no additional attributes are required. + * If using CIPSO_V4_MAP_PASS or CIPSO_V4_MAP_LOCAL no additional attributes + * are required. * * o REMOVE: * Sent by an application to remove a specific DOI mapping table from the @@ -76,12 +77,13 @@ * NLBL_CIPSOV4_A_MTYPE * NLBL_CIPSOV4_A_TAGLST * - * If using CIPSO_V4_MAP_STD the following attributes are required: + * If using CIPSO_V4_MAP_TRANS the following attributes are required: * * NLBL_CIPSOV4_A_MLSLVLLST * NLBL_CIPSOV4_A_MLSCATLST * - * If using CIPSO_V4_MAP_PASS no additional attributes are required. + * If using CIPSO_V4_MAP_PASS or CIPSO_V4_MAP_LOCAL no additional attributes + * are required. * * o LISTALL: * This message is sent by an application to list the valid DOIs on the diff --git a/net/netlabel/netlabel_domainhash.c b/net/netlabel/netlabel_domainhash.c index 643c032a3a57..5fadf10e5ddf 100644 --- a/net/netlabel/netlabel_domainhash.c +++ b/net/netlabel/netlabel_domainhash.c @@ -11,7 +11,7 @@ */ /* - * (c) Copyright Hewlett-Packard Development Company, L.P., 2006 + * (c) Copyright Hewlett-Packard Development Company, L.P., 2006, 2008 * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -40,6 +40,7 @@ #include <asm/bug.h> #include "netlabel_mgmt.h" +#include "netlabel_addrlist.h" #include "netlabel_domainhash.h" #include "netlabel_user.h" @@ -72,8 +73,28 @@ static struct netlbl_dom_map *netlbl_domhsh_def = NULL; static void netlbl_domhsh_free_entry(struct rcu_head *entry) { struct netlbl_dom_map *ptr; + struct netlbl_af4list *iter4; + struct netlbl_af4list *tmp4; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + struct netlbl_af6list *iter6; + struct netlbl_af6list *tmp6; +#endif /* IPv6 */ ptr = container_of(entry, struct netlbl_dom_map, rcu); + if (ptr->type == NETLBL_NLTYPE_ADDRSELECT) { + netlbl_af4list_foreach_safe(iter4, tmp4, + &ptr->type_def.addrsel->list4) { + netlbl_af4list_remove_entry(iter4); + kfree(netlbl_domhsh_addr4_entry(iter4)); + } +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + netlbl_af6list_foreach_safe(iter6, tmp6, + &ptr->type_def.addrsel->list6) { + netlbl_af6list_remove_entry(iter6); + kfree(netlbl_domhsh_addr6_entry(iter6)); + } +#endif /* IPv6 */ + } kfree(ptr->domain); kfree(ptr); } @@ -115,13 +136,13 @@ static u32 netlbl_domhsh_hash(const char *key) static struct netlbl_dom_map *netlbl_domhsh_search(const char *domain) { u32 bkt; + struct list_head *bkt_list; struct netlbl_dom_map *iter; if (domain != NULL) { bkt = netlbl_domhsh_hash(domain); - list_for_each_entry_rcu(iter, - &rcu_dereference(netlbl_domhsh)->tbl[bkt], - list) + bkt_list = &rcu_dereference(netlbl_domhsh)->tbl[bkt]; + list_for_each_entry_rcu(iter, bkt_list, list) if (iter->valid && strcmp(iter->domain, domain) == 0) return iter; } @@ -156,6 +177,69 @@ static struct netlbl_dom_map *netlbl_domhsh_search_def(const char *domain) return entry; } +/** + * netlbl_domhsh_audit_add - Generate an audit entry for an add event + * @entry: the entry being added + * @addr4: the IPv4 address information + * @addr6: the IPv6 address information + * @result: the result code + * @audit_info: NetLabel audit information + * + * Description: + * Generate an audit record for adding a new NetLabel/LSM mapping entry with + * the given information. Caller is responsibile for holding the necessary + * locks. + * + */ +static void netlbl_domhsh_audit_add(struct netlbl_dom_map *entry, + struct netlbl_af4list *addr4, + struct netlbl_af6list *addr6, + int result, + struct netlbl_audit *audit_info) +{ + struct audit_buffer *audit_buf; + struct cipso_v4_doi *cipsov4 = NULL; + u32 type; + + audit_buf = netlbl_audit_start_common(AUDIT_MAC_MAP_ADD, audit_info); + if (audit_buf != NULL) { + audit_log_format(audit_buf, " nlbl_domain=%s", + entry->domain ? entry->domain : "(default)"); + if (addr4 != NULL) { + struct netlbl_domaddr4_map *map4; + map4 = netlbl_domhsh_addr4_entry(addr4); + type = map4->type; + cipsov4 = map4->type_def.cipsov4; + netlbl_af4list_audit_addr(audit_buf, 0, NULL, + addr4->addr, addr4->mask); +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + } else if (addr6 != NULL) { + struct netlbl_domaddr6_map *map6; + map6 = netlbl_domhsh_addr6_entry(addr6); + type = map6->type; + netlbl_af6list_audit_addr(audit_buf, 0, NULL, + &addr6->addr, &addr6->mask); +#endif /* IPv6 */ + } else { + type = entry->type; + cipsov4 = entry->type_def.cipsov4; + } + switch (type) { + case NETLBL_NLTYPE_UNLABELED: + audit_log_format(audit_buf, " nlbl_protocol=unlbl"); + break; + case NETLBL_NLTYPE_CIPSOV4: + BUG_ON(cipsov4 == NULL); + audit_log_format(audit_buf, + " nlbl_protocol=cipsov4 cipso_doi=%u", + cipsov4->doi); + break; + } + audit_log_format(audit_buf, " res=%u", result == 0 ? 1 : 0); + audit_log_end(audit_buf); + } +} + /* * Domain Hash Table Functions */ @@ -213,74 +297,106 @@ int __init netlbl_domhsh_init(u32 size) int netlbl_domhsh_add(struct netlbl_dom_map *entry, struct netlbl_audit *audit_info) { - int ret_val; - u32 bkt; - struct audit_buffer *audit_buf; - - switch (entry->type) { - case NETLBL_NLTYPE_UNLABELED: - ret_val = 0; - break; - case NETLBL_NLTYPE_CIPSOV4: - ret_val = cipso_v4_doi_domhsh_add(entry->type_def.cipsov4, - entry->domain); - break; - default: - return -EINVAL; - } - if (ret_val != 0) - return ret_val; - - entry->valid = 1; - INIT_RCU_HEAD(&entry->rcu); + int ret_val = 0; + struct netlbl_dom_map *entry_old; + struct netlbl_af4list *iter4; + struct netlbl_af4list *tmp4; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + struct netlbl_af6list *iter6; + struct netlbl_af6list *tmp6; +#endif /* IPv6 */ rcu_read_lock(); + spin_lock(&netlbl_domhsh_lock); - if (entry->domain != NULL) { - bkt = netlbl_domhsh_hash(entry->domain); - if (netlbl_domhsh_search(entry->domain) == NULL) + if (entry->domain != NULL) + entry_old = netlbl_domhsh_search(entry->domain); + else + entry_old = netlbl_domhsh_search_def(entry->domain); + if (entry_old == NULL) { + entry->valid = 1; + INIT_RCU_HEAD(&entry->rcu); + + if (entry->domain != NULL) { + u32 bkt = netlbl_domhsh_hash(entry->domain); list_add_tail_rcu(&entry->list, &rcu_dereference(netlbl_domhsh)->tbl[bkt]); - else - ret_val = -EEXIST; - } else { - INIT_LIST_HEAD(&entry->list); - if (rcu_dereference(netlbl_domhsh_def) == NULL) + } else { + INIT_LIST_HEAD(&entry->list); rcu_assign_pointer(netlbl_domhsh_def, entry); - else - ret_val = -EEXIST; - } - spin_unlock(&netlbl_domhsh_lock); - audit_buf = netlbl_audit_start_common(AUDIT_MAC_MAP_ADD, audit_info); - if (audit_buf != NULL) { - audit_log_format(audit_buf, - " nlbl_domain=%s", - entry->domain ? entry->domain : "(default)"); - switch (entry->type) { - case NETLBL_NLTYPE_UNLABELED: - audit_log_format(audit_buf, " nlbl_protocol=unlbl"); - break; - case NETLBL_NLTYPE_CIPSOV4: - audit_log_format(audit_buf, - " nlbl_protocol=cipsov4 cipso_doi=%u", - entry->type_def.cipsov4->doi); - break; } - audit_log_format(audit_buf, " res=%u", ret_val == 0 ? 1 : 0); - audit_log_end(audit_buf); - } - rcu_read_unlock(); - if (ret_val != 0) { - switch (entry->type) { - case NETLBL_NLTYPE_CIPSOV4: - if (cipso_v4_doi_domhsh_remove(entry->type_def.cipsov4, - entry->domain) != 0) - BUG(); - break; + if (entry->type == NETLBL_NLTYPE_ADDRSELECT) { + netlbl_af4list_foreach_rcu(iter4, + &entry->type_def.addrsel->list4) + netlbl_domhsh_audit_add(entry, iter4, NULL, + ret_val, audit_info); +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + netlbl_af6list_foreach_rcu(iter6, + &entry->type_def.addrsel->list6) + netlbl_domhsh_audit_add(entry, NULL, iter6, + ret_val, audit_info); +#endif /* IPv6 */ + } else + netlbl_domhsh_audit_add(entry, NULL, NULL, + ret_val, audit_info); + } else if (entry_old->type == NETLBL_NLTYPE_ADDRSELECT && + entry->type == NETLBL_NLTYPE_ADDRSELECT) { + struct list_head *old_list4; + struct list_head *old_list6; + + old_list4 = &entry_old->type_def.addrsel->list4; + old_list6 = &entry_old->type_def.addrsel->list6; + + /* we only allow the addition of address selectors if all of + * the selectors do not exist in the existing domain map */ + netlbl_af4list_foreach_rcu(iter4, + &entry->type_def.addrsel->list4) + if (netlbl_af4list_search_exact(iter4->addr, + iter4->mask, + old_list4)) { + ret_val = -EEXIST; + goto add_return; + } +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + netlbl_af6list_foreach_rcu(iter6, + &entry->type_def.addrsel->list6) + if (netlbl_af6list_search_exact(&iter6->addr, + &iter6->mask, + old_list6)) { + ret_val = -EEXIST; + goto add_return; + } +#endif /* IPv6 */ + + netlbl_af4list_foreach_safe(iter4, tmp4, + &entry->type_def.addrsel->list4) { + netlbl_af4list_remove_entry(iter4); + iter4->valid = 1; + ret_val = netlbl_af4list_add(iter4, old_list4); + netlbl_domhsh_audit_add(entry_old, iter4, NULL, + ret_val, audit_info); + if (ret_val != 0) + goto add_return; } - } +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + netlbl_af6list_foreach_safe(iter6, tmp6, + &entry->type_def.addrsel->list6) { + netlbl_af6list_remove_entry(iter6); + iter6->valid = 1; + ret_val = netlbl_af6list_add(iter6, old_list6); + netlbl_domhsh_audit_add(entry_old, NULL, iter6, + ret_val, audit_info); + if (ret_val != 0) + goto add_return; + } +#endif /* IPv6 */ + } else + ret_val = -EINVAL; +add_return: + spin_unlock(&netlbl_domhsh_lock); + rcu_read_unlock(); return ret_val; } @@ -302,35 +418,26 @@ int netlbl_domhsh_add_default(struct netlbl_dom_map *entry, } /** - * netlbl_domhsh_remove - Removes an entry from the domain hash table - * @domain: the domain to remove + * netlbl_domhsh_remove_entry - Removes a given entry from the domain table + * @entry: the entry to remove * @audit_info: NetLabel audit information * * Description: * Removes an entry from the domain hash table and handles any updates to the - * lower level protocol handler (i.e. CIPSO). Returns zero on success, - * negative on failure. + * lower level protocol handler (i.e. CIPSO). Caller is responsible for + * ensuring that the RCU read lock is held. Returns zero on success, negative + * on failure. * */ -int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info) +int netlbl_domhsh_remove_entry(struct netlbl_dom_map *entry, + struct netlbl_audit *audit_info) { - int ret_val = -ENOENT; - struct netlbl_dom_map *entry; + int ret_val = 0; struct audit_buffer *audit_buf; - rcu_read_lock(); - if (domain) - entry = netlbl_domhsh_search(domain); - else - entry = netlbl_domhsh_search_def(domain); if (entry == NULL) - goto remove_return; - switch (entry->type) { - case NETLBL_NLTYPE_CIPSOV4: - cipso_v4_doi_domhsh_remove(entry->type_def.cipsov4, - entry->domain); - break; - } + return -ENOENT; + spin_lock(&netlbl_domhsh_lock); if (entry->valid) { entry->valid = 0; @@ -338,8 +445,8 @@ int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info) list_del_rcu(&entry->list); else rcu_assign_pointer(netlbl_domhsh_def, NULL); - ret_val = 0; - } + } else + ret_val = -ENOENT; spin_unlock(&netlbl_domhsh_lock); audit_buf = netlbl_audit_start_common(AUDIT_MAC_MAP_DEL, audit_info); @@ -351,10 +458,54 @@ int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info) audit_log_end(audit_buf); } -remove_return: - rcu_read_unlock(); - if (ret_val == 0) + if (ret_val == 0) { + struct netlbl_af4list *iter4; + struct netlbl_domaddr4_map *map4; + + switch (entry->type) { + case NETLBL_NLTYPE_ADDRSELECT: + netlbl_af4list_foreach_rcu(iter4, + &entry->type_def.addrsel->list4) { + map4 = netlbl_domhsh_addr4_entry(iter4); + cipso_v4_doi_putdef(map4->type_def.cipsov4); + } + /* no need to check the IPv6 list since we currently + * support only unlabeled protocols for IPv6 */ + break; + case NETLBL_NLTYPE_CIPSOV4: + cipso_v4_doi_putdef(entry->type_def.cipsov4); + break; + } call_rcu(&entry->rcu, netlbl_domhsh_free_entry); + } + + return ret_val; +} + +/** + * netlbl_domhsh_remove - Removes an entry from the domain hash table + * @domain: the domain to remove + * @audit_info: NetLabel audit information + * + * Description: + * Removes an entry from the domain hash table and handles any updates to the + * lower level protocol handler (i.e. CIPSO). Returns zero on success, + * negative on failure. + * + */ +int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info) +{ + int ret_val; + struct netlbl_dom_map *entry; + + rcu_read_lock(); + if (domain) + entry = netlbl_domhsh_search(domain); + else + entry = netlbl_domhsh_search_def(domain); + ret_val = netlbl_domhsh_remove_entry(entry, audit_info); + rcu_read_unlock(); + return ret_val; } @@ -389,6 +540,70 @@ struct netlbl_dom_map *netlbl_domhsh_getentry(const char *domain) } /** + * netlbl_domhsh_getentry_af4 - Get an entry from the domain hash table + * @domain: the domain name to search for + * @addr: the IP address to search for + * + * Description: + * Look through the domain hash table searching for an entry to match @domain + * and @addr, return a pointer to a copy of the entry or NULL. The caller is + * responsible for ensuring that rcu_read_[un]lock() is called. + * + */ +struct netlbl_domaddr4_map *netlbl_domhsh_getentry_af4(const char *domain, + __be32 addr) +{ + struct netlbl_dom_map *dom_iter; + struct netlbl_af4list *addr_iter; + + dom_iter = netlbl_domhsh_search_def(domain); + if (dom_iter == NULL) + return NULL; + if (dom_iter->type != NETLBL_NLTYPE_ADDRSELECT) + return NULL; + + addr_iter = netlbl_af4list_search(addr, + &dom_iter->type_def.addrsel->list4); + if (addr_iter == NULL) + return NULL; + + return netlbl_domhsh_addr4_entry(addr_iter); +} + +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) +/** + * netlbl_domhsh_getentry_af6 - Get an entry from the domain hash table + * @domain: the domain name to search for + * @addr: the IP address to search for + * + * Description: + * Look through the domain hash table searching for an entry to match @domain + * and @addr, return a pointer to a copy of the entry or NULL. The caller is + * responsible for ensuring that rcu_read_[un]lock() is called. + * + */ +struct netlbl_domaddr6_map *netlbl_domhsh_getentry_af6(const char *domain, + const struct in6_addr *addr) +{ + struct netlbl_dom_map *dom_iter; + struct netlbl_af6list *addr_iter; + + dom_iter = netlbl_domhsh_search_def(domain); + if (dom_iter == NULL) + return NULL; + if (dom_iter->type != NETLBL_NLTYPE_ADDRSELECT) + return NULL; + + addr_iter = netlbl_af6list_search(addr, + &dom_iter->type_def.addrsel->list6); + if (addr_iter == NULL) + return NULL; + + return netlbl_domhsh_addr6_entry(addr_iter); +} +#endif /* IPv6 */ + +/** * netlbl_domhsh_walk - Iterate through the domain mapping hash table * @skip_bkt: the number of buckets to skip at the start * @skip_chain: the number of entries to skip in the first iterated bucket @@ -410,6 +625,7 @@ int netlbl_domhsh_walk(u32 *skip_bkt, { int ret_val = -ENOENT; u32 iter_bkt; + struct list_head *iter_list; struct netlbl_dom_map *iter_entry; u32 chain_cnt = 0; @@ -417,9 +633,8 @@ int netlbl_domhsh_walk(u32 *skip_bkt, for (iter_bkt = *skip_bkt; iter_bkt < rcu_dereference(netlbl_domhsh)->size; iter_bkt++, chain_cnt = 0) { - list_for_each_entry_rcu(iter_entry, - &rcu_dereference(netlbl_domhsh)->tbl[iter_bkt], - list) + iter_list = &rcu_dereference(netlbl_domhsh)->tbl[iter_bkt]; + list_for_each_entry_rcu(iter_entry, iter_list, list) if (iter_entry->valid) { if (chain_cnt++ < *skip_chain) continue; diff --git a/net/netlabel/netlabel_domainhash.h b/net/netlabel/netlabel_domainhash.h index 8220990ceb96..bfcb6763a1a1 100644 --- a/net/netlabel/netlabel_domainhash.h +++ b/net/netlabel/netlabel_domainhash.h @@ -11,7 +11,7 @@ */ /* - * (c) Copyright Hewlett-Packard Development Company, L.P., 2006 + * (c) Copyright Hewlett-Packard Development Company, L.P., 2006, 2008 * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -36,16 +36,43 @@ #include <linux/rcupdate.h> #include <linux/list.h> +#include "netlabel_addrlist.h" + /* Domain hash table size */ /* XXX - currently this number is an uneducated guess */ #define NETLBL_DOMHSH_BITSIZE 7 -/* Domain mapping definition struct */ +/* Domain mapping definition structures */ +#define netlbl_domhsh_addr4_entry(iter) \ + container_of(iter, struct netlbl_domaddr4_map, list) +struct netlbl_domaddr4_map { + u32 type; + union { + struct cipso_v4_doi *cipsov4; + } type_def; + + struct netlbl_af4list list; +}; +#define netlbl_domhsh_addr6_entry(iter) \ + container_of(iter, struct netlbl_domaddr6_map, list) +struct netlbl_domaddr6_map { + u32 type; + + /* NOTE: no 'type_def' union needed at present since we don't currently + * support any IPv6 labeling protocols */ + + struct netlbl_af6list list; +}; +struct netlbl_domaddr_map { + struct list_head list4; + struct list_head list6; +}; struct netlbl_dom_map { char *domain; u32 type; union { struct cipso_v4_doi *cipsov4; + struct netlbl_domaddr_map *addrsel; } type_def; u32 valid; @@ -61,12 +88,21 @@ int netlbl_domhsh_add(struct netlbl_dom_map *entry, struct netlbl_audit *audit_info); int netlbl_domhsh_add_default(struct netlbl_dom_map *entry, struct netlbl_audit *audit_info); +int netlbl_domhsh_remove_entry(struct netlbl_dom_map *entry, + struct netlbl_audit *audit_info); int netlbl_domhsh_remove(const char *domain, struct netlbl_audit *audit_info); int netlbl_domhsh_remove_default(struct netlbl_audit *audit_info); struct netlbl_dom_map *netlbl_domhsh_getentry(const char *domain); +struct netlbl_domaddr4_map *netlbl_domhsh_getentry_af4(const char *domain, + __be32 addr); int netlbl_domhsh_walk(u32 *skip_bkt, u32 *skip_chain, int (*callback) (struct netlbl_dom_map *entry, void *arg), void *cb_arg); +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) +struct netlbl_domaddr6_map *netlbl_domhsh_getentry_af6(const char *domain, + const struct in6_addr *addr); +#endif /* IPv6 */ + #endif diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c index 39793a1a93aa..b32eceb3ab0d 100644 --- a/net/netlabel/netlabel_kapi.c +++ b/net/netlabel/netlabel_kapi.c @@ -10,7 +10,7 @@ */ /* - * (c) Copyright Hewlett-Packard Development Company, L.P., 2006 + * (c) Copyright Hewlett-Packard Development Company, L.P., 2006, 2008 * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -82,7 +82,7 @@ int netlbl_cfg_unlbl_add_map(const char *domain, entry = kzalloc(sizeof(*entry), GFP_ATOMIC); if (entry == NULL) - goto cfg_unlbl_add_map_failure; + return -ENOMEM; if (domain != NULL) { entry->domain = kstrdup(domain, GFP_ATOMIC); if (entry->domain == NULL) @@ -104,49 +104,6 @@ cfg_unlbl_add_map_failure: } /** - * netlbl_cfg_cipsov4_add - Add a new CIPSOv4 DOI definition - * @doi_def: the DOI definition - * @audit_info: NetLabel audit information - * - * Description: - * Add a new CIPSOv4 DOI definition to the NetLabel subsystem. Returns zero on - * success, negative values on failure. - * - */ -int netlbl_cfg_cipsov4_add(struct cipso_v4_doi *doi_def, - struct netlbl_audit *audit_info) -{ - int ret_val; - const char *type_str; - struct audit_buffer *audit_buf; - - ret_val = cipso_v4_doi_add(doi_def); - - audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_ADD, - audit_info); - if (audit_buf != NULL) { - switch (doi_def->type) { - case CIPSO_V4_MAP_STD: - type_str = "std"; - break; - case CIPSO_V4_MAP_PASS: - type_str = "pass"; - break; - default: - type_str = "(unknown)"; - } - audit_log_format(audit_buf, - " cipso_doi=%u cipso_type=%s res=%u", - doi_def->doi, - type_str, - ret_val == 0 ? 1 : 0); - audit_log_end(audit_buf); - } - - return ret_val; -} - -/** * netlbl_cfg_cipsov4_add_map - Add a new CIPSOv4 DOI definition and mapping * @doi_def: the DOI definition * @domain: the domain mapping to add @@ -164,58 +121,71 @@ int netlbl_cfg_cipsov4_add_map(struct cipso_v4_doi *doi_def, struct netlbl_audit *audit_info) { int ret_val = -ENOMEM; + u32 doi; + u32 doi_type; struct netlbl_dom_map *entry; + const char *type_str; + struct audit_buffer *audit_buf; + + doi = doi_def->doi; + doi_type = doi_def->type; entry = kzalloc(sizeof(*entry), GFP_ATOMIC); if (entry == NULL) - goto cfg_cipsov4_add_map_failure; + return -ENOMEM; if (domain != NULL) { entry->domain = kstrdup(domain, GFP_ATOMIC); if (entry->domain == NULL) goto cfg_cipsov4_add_map_failure; } - entry->type = NETLBL_NLTYPE_CIPSOV4; - entry->type_def.cipsov4 = doi_def; - - /* Grab a RCU read lock here so nothing happens to the doi_def variable - * between adding it to the CIPSOv4 protocol engine and adding a - * domain mapping for it. */ - rcu_read_lock(); - ret_val = netlbl_cfg_cipsov4_add(doi_def, audit_info); + ret_val = cipso_v4_doi_add(doi_def); if (ret_val != 0) - goto cfg_cipsov4_add_map_failure_unlock; + goto cfg_cipsov4_add_map_failure_remove_doi; + entry->type = NETLBL_NLTYPE_CIPSOV4; + entry->type_def.cipsov4 = cipso_v4_doi_getdef(doi); + if (entry->type_def.cipsov4 == NULL) { + ret_val = -ENOENT; + goto cfg_cipsov4_add_map_failure_remove_doi; + } ret_val = netlbl_domhsh_add(entry, audit_info); if (ret_val != 0) - goto cfg_cipsov4_add_map_failure_remove_doi; - rcu_read_unlock(); + goto cfg_cipsov4_add_map_failure_release_doi; - return 0; +cfg_cipsov4_add_map_return: + audit_buf = netlbl_audit_start_common(AUDIT_MAC_CIPSOV4_ADD, + audit_info); + if (audit_buf != NULL) { + switch (doi_type) { + case CIPSO_V4_MAP_TRANS: + type_str = "trans"; + break; + case CIPSO_V4_MAP_PASS: + type_str = "pass"; + break; + case CIPSO_V4_MAP_LOCAL: + type_str = "local"; + break; + default: + type_str = "(unknown)"; + } + audit_log_format(audit_buf, + " cipso_doi=%u cipso_type=%s res=%u", + doi, type_str, ret_val == 0 ? 1 : 0); + audit_log_end(audit_buf); + } + return ret_val; + +cfg_cipsov4_add_map_failure_release_doi: + cipso_v4_doi_putdef(doi_def); cfg_cipsov4_add_map_failure_remove_doi: - cipso_v4_doi_remove(doi_def->doi, audit_info, netlbl_cipsov4_doi_free); -cfg_cipsov4_add_map_failure_unlock: - rcu_read_unlock(); + cipso_v4_doi_remove(doi, audit_info); cfg_cipsov4_add_map_failure: if (entry != NULL) kfree(entry->domain); kfree(entry); - return ret_val; -} - -/** - * netlbl_cfg_cipsov4_del - Removean existing CIPSOv4 DOI definition - * @doi: the CIPSO DOI value - * @audit_info: NetLabel audit information - * - * Description: - * Removes an existing CIPSOv4 DOI definition from the NetLabel subsystem. - * Returns zero on success, negative values on failure. - * - */ -int netlbl_cfg_cipsov4_del(u32 doi, struct netlbl_audit *audit_info) -{ - return cipso_v4_doi_remove(doi, audit_info, netlbl_cipsov4_doi_free); + goto cfg_cipsov4_add_map_return; } /* @@ -452,7 +422,9 @@ int netlbl_enabled(void) * Attach the correct label to the given socket using the security attributes * specified in @secattr. This function requires exclusive access to @sk, * which means it either needs to be in the process of being created or locked. - * Returns zero on success, negative values on failure. + * Returns zero on success, -EDESTADDRREQ if the domain is configured to use + * network address selectors (can't blindly label the socket), and negative + * values on all other failures. * */ int netlbl_sock_setattr(struct sock *sk, @@ -466,6 +438,9 @@ int netlbl_sock_setattr(struct sock *sk, if (dom_entry == NULL) goto socket_setattr_return; switch (dom_entry->type) { + case NETLBL_NLTYPE_ADDRSELECT: + ret_val = -EDESTADDRREQ; + break; case NETLBL_NLTYPE_CIPSOV4: ret_val = cipso_v4_sock_setattr(sk, dom_entry->type_def.cipsov4, @@ -484,6 +459,20 @@ socket_setattr_return: } /** + * netlbl_sock_delattr - Delete all the NetLabel labels on a socket + * @sk: the socket + * + * Description: + * Remove all the NetLabel labeling from @sk. The caller is responsible for + * ensuring that @sk is locked. + * + */ +void netlbl_sock_delattr(struct sock *sk) +{ + cipso_v4_sock_delattr(sk); +} + +/** * netlbl_sock_getattr - Determine the security attributes of a sock * @sk: the sock * @secattr: the security attributes @@ -501,6 +490,128 @@ int netlbl_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr) } /** + * netlbl_conn_setattr - Label a connected socket using the correct protocol + * @sk: the socket to label + * @addr: the destination address + * @secattr: the security attributes + * + * Description: + * Attach the correct label to the given connected socket using the security + * attributes specified in @secattr. The caller is responsible for ensuring + * that @sk is locked. Returns zero on success, negative values on failure. + * + */ +int netlbl_conn_setattr(struct sock *sk, + struct sockaddr *addr, + const struct netlbl_lsm_secattr *secattr) +{ + int ret_val; + struct sockaddr_in *addr4; + struct netlbl_domaddr4_map *af4_entry; + + rcu_read_lock(); + switch (addr->sa_family) { + case AF_INET: + addr4 = (struct sockaddr_in *)addr; + af4_entry = netlbl_domhsh_getentry_af4(secattr->domain, + addr4->sin_addr.s_addr); + if (af4_entry == NULL) { + ret_val = -ENOENT; + goto conn_setattr_return; + } + switch (af4_entry->type) { + case NETLBL_NLTYPE_CIPSOV4: + ret_val = cipso_v4_sock_setattr(sk, + af4_entry->type_def.cipsov4, + secattr); + break; + case NETLBL_NLTYPE_UNLABELED: + /* just delete the protocols we support for right now + * but we could remove other protocols if needed */ + cipso_v4_sock_delattr(sk); + ret_val = 0; + break; + default: + ret_val = -ENOENT; + } + break; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + case AF_INET6: + /* since we don't support any IPv6 labeling protocols right + * now we can optimize everything away until we do */ + ret_val = 0; + break; +#endif /* IPv6 */ + default: + ret_val = 0; + } + +conn_setattr_return: + rcu_read_unlock(); + return ret_val; +} + +/** + * netlbl_skbuff_setattr - Label a packet using the correct protocol + * @skb: the packet + * @family: protocol family + * @secattr: the security attributes + * + * Description: + * Attach the correct label to the given packet using the security attributes + * specified in @secattr. Returns zero on success, negative values on failure. + * + */ +int netlbl_skbuff_setattr(struct sk_buff *skb, + u16 family, + const struct netlbl_lsm_secattr *secattr) +{ + int ret_val; + struct iphdr *hdr4; + struct netlbl_domaddr4_map *af4_entry; + + rcu_read_lock(); + switch (family) { + case AF_INET: + hdr4 = ip_hdr(skb); + af4_entry = netlbl_domhsh_getentry_af4(secattr->domain, + hdr4->daddr); + if (af4_entry == NULL) { + ret_val = -ENOENT; + goto skbuff_setattr_return; + } + switch (af4_entry->type) { + case NETLBL_NLTYPE_CIPSOV4: + ret_val = cipso_v4_skbuff_setattr(skb, + af4_entry->type_def.cipsov4, + secattr); + break; + case NETLBL_NLTYPE_UNLABELED: + /* just delete the protocols we support for right now + * but we could remove other protocols if needed */ + ret_val = cipso_v4_skbuff_delattr(skb); + break; + default: + ret_val = -ENOENT; + } + break; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + case AF_INET6: + /* since we don't support any IPv6 labeling protocols right + * now we can optimize everything away until we do */ + ret_val = 0; + break; +#endif /* IPv6 */ + default: + ret_val = 0; + } + +skbuff_setattr_return: + rcu_read_unlock(); + return ret_val; +} + +/** * netlbl_skbuff_getattr - Determine the security attributes of a packet * @skb: the packet * @family: protocol family @@ -528,6 +639,7 @@ int netlbl_skbuff_getattr(const struct sk_buff *skb, * netlbl_skbuff_err - Handle a LSM error on a sk_buff * @skb: the packet * @error: the error code + * @gateway: true if host is acting as a gateway, false otherwise * * Description: * Deal with a LSM problem when handling the packet in @skb, typically this is @@ -535,10 +647,10 @@ int netlbl_skbuff_getattr(const struct sk_buff *skb, * according to the packet's labeling protocol. * */ -void netlbl_skbuff_err(struct sk_buff *skb, int error) +void netlbl_skbuff_err(struct sk_buff *skb, int error, int gateway) { if (CIPSO_V4_OPTEXIST(skb)) - cipso_v4_error(skb, error, 0); + cipso_v4_error(skb, error, gateway); } /** diff --git a/net/netlabel/netlabel_mgmt.c b/net/netlabel/netlabel_mgmt.c index 44be5d5261f4..ee769ecaa13c 100644 --- a/net/netlabel/netlabel_mgmt.c +++ b/net/netlabel/netlabel_mgmt.c @@ -10,7 +10,7 @@ */ /* - * (c) Copyright Hewlett-Packard Development Company, L.P., 2006 + * (c) Copyright Hewlett-Packard Development Company, L.P., 2006, 2008 * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -32,9 +32,13 @@ #include <linux/socket.h> #include <linux/string.h> #include <linux/skbuff.h> +#include <linux/in.h> +#include <linux/in6.h> #include <net/sock.h> #include <net/netlink.h> #include <net/genetlink.h> +#include <net/ip.h> +#include <net/ipv6.h> #include <net/netlabel.h> #include <net/cipso_ipv4.h> #include <asm/atomic.h> @@ -71,86 +75,337 @@ static const struct nla_policy netlbl_mgmt_genl_policy[NLBL_MGMT_A_MAX + 1] = { }; /* - * NetLabel Command Handlers + * Helper Functions */ /** * netlbl_mgmt_add - Handle an ADD message - * @skb: the NETLINK buffer * @info: the Generic NETLINK info block + * @audit_info: NetLabel audit information * * Description: - * Process a user generated ADD message and add the domains from the message - * to the hash table. See netlabel.h for a description of the message format. - * Returns zero on success, negative values on failure. + * Helper function for the ADD and ADDDEF messages to add the domain mappings + * from the message to the hash table. See netlabel.h for a description of the + * message format. Returns zero on success, negative values on failure. * */ -static int netlbl_mgmt_add(struct sk_buff *skb, struct genl_info *info) +static int netlbl_mgmt_add_common(struct genl_info *info, + struct netlbl_audit *audit_info) { int ret_val = -EINVAL; struct netlbl_dom_map *entry = NULL; - size_t tmp_size; + struct netlbl_domaddr_map *addrmap = NULL; + struct cipso_v4_doi *cipsov4 = NULL; u32 tmp_val; - struct netlbl_audit audit_info; - - if (!info->attrs[NLBL_MGMT_A_DOMAIN] || - !info->attrs[NLBL_MGMT_A_PROTOCOL]) - goto add_failure; - - netlbl_netlink_auditinfo(skb, &audit_info); entry = kzalloc(sizeof(*entry), GFP_KERNEL); if (entry == NULL) { ret_val = -ENOMEM; goto add_failure; } - tmp_size = nla_len(info->attrs[NLBL_MGMT_A_DOMAIN]); - entry->domain = kmalloc(tmp_size, GFP_KERNEL); - if (entry->domain == NULL) { - ret_val = -ENOMEM; - goto add_failure; - } entry->type = nla_get_u32(info->attrs[NLBL_MGMT_A_PROTOCOL]); - nla_strlcpy(entry->domain, info->attrs[NLBL_MGMT_A_DOMAIN], tmp_size); + if (info->attrs[NLBL_MGMT_A_DOMAIN]) { + size_t tmp_size = nla_len(info->attrs[NLBL_MGMT_A_DOMAIN]); + entry->domain = kmalloc(tmp_size, GFP_KERNEL); + if (entry->domain == NULL) { + ret_val = -ENOMEM; + goto add_failure; + } + nla_strlcpy(entry->domain, + info->attrs[NLBL_MGMT_A_DOMAIN], tmp_size); + } + + /* NOTE: internally we allow/use a entry->type value of + * NETLBL_NLTYPE_ADDRSELECT but we don't currently allow users + * to pass that as a protocol value because we need to know the + * "real" protocol */ switch (entry->type) { case NETLBL_NLTYPE_UNLABELED: - ret_val = netlbl_domhsh_add(entry, &audit_info); break; case NETLBL_NLTYPE_CIPSOV4: if (!info->attrs[NLBL_MGMT_A_CV4DOI]) goto add_failure; tmp_val = nla_get_u32(info->attrs[NLBL_MGMT_A_CV4DOI]); - /* We should be holding a rcu_read_lock() here while we hold - * the result but since the entry will always be deleted when - * the CIPSO DOI is deleted we aren't going to keep the - * lock. */ - rcu_read_lock(); - entry->type_def.cipsov4 = cipso_v4_doi_getdef(tmp_val); - if (entry->type_def.cipsov4 == NULL) { - rcu_read_unlock(); + cipsov4 = cipso_v4_doi_getdef(tmp_val); + if (cipsov4 == NULL) goto add_failure; - } - ret_val = netlbl_domhsh_add(entry, &audit_info); - rcu_read_unlock(); + entry->type_def.cipsov4 = cipsov4; break; default: goto add_failure; } + + if (info->attrs[NLBL_MGMT_A_IPV4ADDR]) { + struct in_addr *addr; + struct in_addr *mask; + struct netlbl_domaddr4_map *map; + + addrmap = kzalloc(sizeof(*addrmap), GFP_KERNEL); + if (addrmap == NULL) { + ret_val = -ENOMEM; + goto add_failure; + } + INIT_LIST_HEAD(&addrmap->list4); + INIT_LIST_HEAD(&addrmap->list6); + + if (nla_len(info->attrs[NLBL_MGMT_A_IPV4ADDR]) != + sizeof(struct in_addr)) { + ret_val = -EINVAL; + goto add_failure; + } + if (nla_len(info->attrs[NLBL_MGMT_A_IPV4MASK]) != + sizeof(struct in_addr)) { + ret_val = -EINVAL; + goto add_failure; + } + addr = nla_data(info->attrs[NLBL_MGMT_A_IPV4ADDR]); + mask = nla_data(info->attrs[NLBL_MGMT_A_IPV4MASK]); + + map = kzalloc(sizeof(*map), GFP_KERNEL); + if (map == NULL) { + ret_val = -ENOMEM; + goto add_failure; + } + map->list.addr = addr->s_addr & mask->s_addr; + map->list.mask = mask->s_addr; + map->list.valid = 1; + map->type = entry->type; + if (cipsov4) + map->type_def.cipsov4 = cipsov4; + + ret_val = netlbl_af4list_add(&map->list, &addrmap->list4); + if (ret_val != 0) { + kfree(map); + goto add_failure; + } + + entry->type = NETLBL_NLTYPE_ADDRSELECT; + entry->type_def.addrsel = addrmap; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + } else if (info->attrs[NLBL_MGMT_A_IPV6ADDR]) { + struct in6_addr *addr; + struct in6_addr *mask; + struct netlbl_domaddr6_map *map; + + addrmap = kzalloc(sizeof(*addrmap), GFP_KERNEL); + if (addrmap == NULL) { + ret_val = -ENOMEM; + goto add_failure; + } + INIT_LIST_HEAD(&addrmap->list4); + INIT_LIST_HEAD(&addrmap->list6); + + if (nla_len(info->attrs[NLBL_MGMT_A_IPV6ADDR]) != + sizeof(struct in6_addr)) { + ret_val = -EINVAL; + goto add_failure; + } + if (nla_len(info->attrs[NLBL_MGMT_A_IPV6MASK]) != + sizeof(struct in6_addr)) { + ret_val = -EINVAL; + goto add_failure; + } + addr = nla_data(info->attrs[NLBL_MGMT_A_IPV6ADDR]); + mask = nla_data(info->attrs[NLBL_MGMT_A_IPV6MASK]); + + map = kzalloc(sizeof(*map), GFP_KERNEL); + if (map == NULL) { + ret_val = -ENOMEM; + goto add_failure; + } + ipv6_addr_copy(&map->list.addr, addr); + map->list.addr.s6_addr32[0] &= mask->s6_addr32[0]; + map->list.addr.s6_addr32[1] &= mask->s6_addr32[1]; + map->list.addr.s6_addr32[2] &= mask->s6_addr32[2]; + map->list.addr.s6_addr32[3] &= mask->s6_addr32[3]; + ipv6_addr_copy(&map->list.mask, mask); + map->list.valid = 1; + map->type = entry->type; + + ret_val = netlbl_af6list_add(&map->list, &addrmap->list6); + if (ret_val != 0) { + kfree(map); + goto add_failure; + } + + entry->type = NETLBL_NLTYPE_ADDRSELECT; + entry->type_def.addrsel = addrmap; +#endif /* IPv6 */ + } + + ret_val = netlbl_domhsh_add(entry, audit_info); if (ret_val != 0) goto add_failure; return 0; add_failure: + if (cipsov4) + cipso_v4_doi_putdef(cipsov4); if (entry) kfree(entry->domain); + kfree(addrmap); kfree(entry); return ret_val; } /** + * netlbl_mgmt_listentry - List a NetLabel/LSM domain map entry + * @skb: the NETLINK buffer + * @entry: the map entry + * + * Description: + * This function is a helper function used by the LISTALL and LISTDEF command + * handlers. The caller is responsibile for ensuring that the RCU read lock + * is held. Returns zero on success, negative values on failure. + * + */ +static int netlbl_mgmt_listentry(struct sk_buff *skb, + struct netlbl_dom_map *entry) +{ + int ret_val; + struct nlattr *nla_a; + struct nlattr *nla_b; + struct netlbl_af4list *iter4; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + struct netlbl_af6list *iter6; +#endif + + if (entry->domain != NULL) { + ret_val = nla_put_string(skb, + NLBL_MGMT_A_DOMAIN, entry->domain); + if (ret_val != 0) + return ret_val; + } + + switch (entry->type) { + case NETLBL_NLTYPE_ADDRSELECT: + nla_a = nla_nest_start(skb, NLBL_MGMT_A_SELECTORLIST); + if (nla_a == NULL) + return -ENOMEM; + + netlbl_af4list_foreach_rcu(iter4, + &entry->type_def.addrsel->list4) { + struct netlbl_domaddr4_map *map4; + struct in_addr addr_struct; + + nla_b = nla_nest_start(skb, NLBL_MGMT_A_ADDRSELECTOR); + if (nla_b == NULL) + return -ENOMEM; + + addr_struct.s_addr = iter4->addr; + ret_val = nla_put(skb, NLBL_MGMT_A_IPV4ADDR, + sizeof(struct in_addr), + &addr_struct); + if (ret_val != 0) + return ret_val; + addr_struct.s_addr = iter4->mask; + ret_val = nla_put(skb, NLBL_MGMT_A_IPV4MASK, + sizeof(struct in_addr), + &addr_struct); + if (ret_val != 0) + return ret_val; + map4 = netlbl_domhsh_addr4_entry(iter4); + ret_val = nla_put_u32(skb, NLBL_MGMT_A_PROTOCOL, + map4->type); + if (ret_val != 0) + return ret_val; + switch (map4->type) { + case NETLBL_NLTYPE_CIPSOV4: + ret_val = nla_put_u32(skb, NLBL_MGMT_A_CV4DOI, + map4->type_def.cipsov4->doi); + if (ret_val != 0) + return ret_val; + break; + } + + nla_nest_end(skb, nla_b); + } +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + netlbl_af6list_foreach_rcu(iter6, + &entry->type_def.addrsel->list6) { + struct netlbl_domaddr6_map *map6; + + nla_b = nla_nest_start(skb, NLBL_MGMT_A_ADDRSELECTOR); + if (nla_b == NULL) + return -ENOMEM; + + ret_val = nla_put(skb, NLBL_MGMT_A_IPV6ADDR, + sizeof(struct in6_addr), + &iter6->addr); + if (ret_val != 0) + return ret_val; + ret_val = nla_put(skb, NLBL_MGMT_A_IPV6MASK, + sizeof(struct in6_addr), + &iter6->mask); + if (ret_val != 0) + return ret_val; + map6 = netlbl_domhsh_addr6_entry(iter6); + ret_val = nla_put_u32(skb, NLBL_MGMT_A_PROTOCOL, + map6->type); + if (ret_val != 0) + return ret_val; + + nla_nest_end(skb, nla_b); + } +#endif /* IPv6 */ + + nla_nest_end(skb, nla_a); + break; + case NETLBL_NLTYPE_UNLABELED: + ret_val = nla_put_u32(skb, NLBL_MGMT_A_PROTOCOL, entry->type); + break; + case NETLBL_NLTYPE_CIPSOV4: + ret_val = nla_put_u32(skb, NLBL_MGMT_A_PROTOCOL, entry->type); + if (ret_val != 0) + return ret_val; + ret_val = nla_put_u32(skb, NLBL_MGMT_A_CV4DOI, + entry->type_def.cipsov4->doi); + break; + } + + return ret_val; +} + +/* + * NetLabel Command Handlers + */ + +/** + * netlbl_mgmt_add - Handle an ADD message + * @skb: the NETLINK buffer + * @info: the Generic NETLINK info block + * + * Description: + * Process a user generated ADD message and add the domains from the message + * to the hash table. See netlabel.h for a description of the message format. + * Returns zero on success, negative values on failure. + * + */ +static int netlbl_mgmt_add(struct sk_buff *skb, struct genl_info *info) +{ + struct netlbl_audit audit_info; + + if ((!info->attrs[NLBL_MGMT_A_DOMAIN]) || + (!info->attrs[NLBL_MGMT_A_PROTOCOL]) || + (info->attrs[NLBL_MGMT_A_IPV4ADDR] && + info->attrs[NLBL_MGMT_A_IPV6ADDR]) || + (info->attrs[NLBL_MGMT_A_IPV4MASK] && + info->attrs[NLBL_MGMT_A_IPV6MASK]) || + ((info->attrs[NLBL_MGMT_A_IPV4ADDR] != NULL) ^ + (info->attrs[NLBL_MGMT_A_IPV4MASK] != NULL)) || + ((info->attrs[NLBL_MGMT_A_IPV6ADDR] != NULL) ^ + (info->attrs[NLBL_MGMT_A_IPV6MASK] != NULL))) + return -EINVAL; + + netlbl_netlink_auditinfo(skb, &audit_info); + + return netlbl_mgmt_add_common(info, &audit_info); +} + +/** * netlbl_mgmt_remove - Handle a REMOVE message * @skb: the NETLINK buffer * @info: the Generic NETLINK info block @@ -198,23 +453,9 @@ static int netlbl_mgmt_listall_cb(struct netlbl_dom_map *entry, void *arg) if (data == NULL) goto listall_cb_failure; - ret_val = nla_put_string(cb_arg->skb, - NLBL_MGMT_A_DOMAIN, - entry->domain); + ret_val = netlbl_mgmt_listentry(cb_arg->skb, entry); if (ret_val != 0) goto listall_cb_failure; - ret_val = nla_put_u32(cb_arg->skb, NLBL_MGMT_A_PROTOCOL, entry->type); - if (ret_val != 0) - goto listall_cb_failure; - switch (entry->type) { - case NETLBL_NLTYPE_CIPSOV4: - ret_val = nla_put_u32(cb_arg->skb, - NLBL_MGMT_A_CV4DOI, - entry->type_def.cipsov4->doi); - if (ret_val != 0) - goto listall_cb_failure; - break; - } cb_arg->seq++; return genlmsg_end(cb_arg->skb, data); @@ -268,56 +509,22 @@ static int netlbl_mgmt_listall(struct sk_buff *skb, */ static int netlbl_mgmt_adddef(struct sk_buff *skb, struct genl_info *info) { - int ret_val = -EINVAL; - struct netlbl_dom_map *entry = NULL; - u32 tmp_val; struct netlbl_audit audit_info; - if (!info->attrs[NLBL_MGMT_A_PROTOCOL]) - goto adddef_failure; + if ((!info->attrs[NLBL_MGMT_A_PROTOCOL]) || + (info->attrs[NLBL_MGMT_A_IPV4ADDR] && + info->attrs[NLBL_MGMT_A_IPV6ADDR]) || + (info->attrs[NLBL_MGMT_A_IPV4MASK] && + info->attrs[NLBL_MGMT_A_IPV6MASK]) || + ((info->attrs[NLBL_MGMT_A_IPV4ADDR] != NULL) ^ + (info->attrs[NLBL_MGMT_A_IPV4MASK] != NULL)) || + ((info->attrs[NLBL_MGMT_A_IPV6ADDR] != NULL) ^ + (info->attrs[NLBL_MGMT_A_IPV6MASK] != NULL))) + return -EINVAL; netlbl_netlink_auditinfo(skb, &audit_info); - entry = kzalloc(sizeof(*entry), GFP_KERNEL); - if (entry == NULL) { - ret_val = -ENOMEM; - goto adddef_failure; - } - entry->type = nla_get_u32(info->attrs[NLBL_MGMT_A_PROTOCOL]); - - switch (entry->type) { - case NETLBL_NLTYPE_UNLABELED: - ret_val = netlbl_domhsh_add_default(entry, &audit_info); - break; - case NETLBL_NLTYPE_CIPSOV4: - if (!info->attrs[NLBL_MGMT_A_CV4DOI]) - goto adddef_failure; - - tmp_val = nla_get_u32(info->attrs[NLBL_MGMT_A_CV4DOI]); - /* We should be holding a rcu_read_lock() here while we hold - * the result but since the entry will always be deleted when - * the CIPSO DOI is deleted we aren't going to keep the - * lock. */ - rcu_read_lock(); - entry->type_def.cipsov4 = cipso_v4_doi_getdef(tmp_val); - if (entry->type_def.cipsov4 == NULL) { - rcu_read_unlock(); - goto adddef_failure; - } - ret_val = netlbl_domhsh_add_default(entry, &audit_info); - rcu_read_unlock(); - break; - default: - goto adddef_failure; - } - if (ret_val != 0) - goto adddef_failure; - - return 0; - -adddef_failure: - kfree(entry); - return ret_val; + return netlbl_mgmt_add_common(info, &audit_info); } /** @@ -371,19 +578,10 @@ static int netlbl_mgmt_listdef(struct sk_buff *skb, struct genl_info *info) ret_val = -ENOENT; goto listdef_failure_lock; } - ret_val = nla_put_u32(ans_skb, NLBL_MGMT_A_PROTOCOL, entry->type); - if (ret_val != 0) - goto listdef_failure_lock; - switch (entry->type) { - case NETLBL_NLTYPE_CIPSOV4: - ret_val = nla_put_u32(ans_skb, - NLBL_MGMT_A_CV4DOI, - entry->type_def.cipsov4->doi); - if (ret_val != 0) - goto listdef_failure_lock; - break; - } + ret_val = netlbl_mgmt_listentry(ans_skb, entry); rcu_read_unlock(); + if (ret_val != 0) + goto listdef_failure; genlmsg_end(ans_skb, data); return genlmsg_reply(ans_skb, info); diff --git a/net/netlabel/netlabel_mgmt.h b/net/netlabel/netlabel_mgmt.h index a43bff169d6b..05d96431f819 100644 --- a/net/netlabel/netlabel_mgmt.h +++ b/net/netlabel/netlabel_mgmt.h @@ -45,6 +45,16 @@ * NLBL_MGMT_A_DOMAIN * NLBL_MGMT_A_PROTOCOL * + * If IPv4 is specified the following attributes are required: + * + * NLBL_MGMT_A_IPV4ADDR + * NLBL_MGMT_A_IPV4MASK + * + * If IPv6 is specified the following attributes are required: + * + * NLBL_MGMT_A_IPV6ADDR + * NLBL_MGMT_A_IPV6MASK + * * If using NETLBL_NLTYPE_CIPSOV4 the following attributes are required: * * NLBL_MGMT_A_CV4DOI @@ -68,13 +78,24 @@ * Required attributes: * * NLBL_MGMT_A_DOMAIN + * + * If the IP address selectors are not used the following attribute is + * required: + * * NLBL_MGMT_A_PROTOCOL * - * If using NETLBL_NLTYPE_CIPSOV4 the following attributes are required: + * If the IP address selectors are used then the following attritbute is + * required: + * + * NLBL_MGMT_A_SELECTORLIST + * + * If the mapping is using the NETLBL_NLTYPE_CIPSOV4 type then the following + * attributes are required: * * NLBL_MGMT_A_CV4DOI * - * If using NETLBL_NLTYPE_UNLABELED no other attributes are required. + * If the mapping is using the NETLBL_NLTYPE_UNLABELED type no other + * attributes are required. * * o ADDDEF: * Sent by an application to set the default domain mapping for the NetLabel @@ -100,15 +121,23 @@ * application there is no payload. On success the kernel should send a * response using the following format. * - * Required attributes: + * If the IP address selectors are not used the following attribute is + * required: * * NLBL_MGMT_A_PROTOCOL * - * If using NETLBL_NLTYPE_CIPSOV4 the following attributes are required: + * If the IP address selectors are used then the following attritbute is + * required: + * + * NLBL_MGMT_A_SELECTORLIST + * + * If the mapping is using the NETLBL_NLTYPE_CIPSOV4 type then the following + * attributes are required: * * NLBL_MGMT_A_CV4DOI * - * If using NETLBL_NLTYPE_UNLABELED no other attributes are required. + * If the mapping is using the NETLBL_NLTYPE_UNLABELED type no other + * attributes are required. * * o PROTOCOLS: * Sent by an application to request a list of configured NetLabel protocols @@ -162,6 +191,26 @@ enum { NLBL_MGMT_A_CV4DOI, /* (NLA_U32) * the CIPSOv4 DOI value */ + NLBL_MGMT_A_IPV6ADDR, + /* (NLA_BINARY, struct in6_addr) + * an IPv6 address */ + NLBL_MGMT_A_IPV6MASK, + /* (NLA_BINARY, struct in6_addr) + * an IPv6 address mask */ + NLBL_MGMT_A_IPV4ADDR, + /* (NLA_BINARY, struct in_addr) + * an IPv4 address */ + NLBL_MGMT_A_IPV4MASK, + /* (NLA_BINARY, struct in_addr) + * and IPv4 address mask */ + NLBL_MGMT_A_ADDRSELECTOR, + /* (NLA_NESTED) + * an IP address selector, must contain an address, mask, and protocol + * attribute plus any protocol specific attributes */ + NLBL_MGMT_A_SELECTORLIST, + /* (NLA_NESTED) + * the selector list, there must be at least one + * NLBL_MGMT_A_ADDRSELECTOR attribute */ __NLBL_MGMT_A_MAX, }; #define NLBL_MGMT_A_MAX (__NLBL_MGMT_A_MAX - 1) diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c index 921c118ead89..e8a5c32b0f10 100644 --- a/net/netlabel/netlabel_unlabeled.c +++ b/net/netlabel/netlabel_unlabeled.c @@ -10,7 +10,7 @@ */ /* - * (c) Copyright Hewlett-Packard Development Company, L.P., 2006 - 2007 + * (c) Copyright Hewlett-Packard Development Company, L.P., 2006 - 2008 * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -54,6 +54,7 @@ #include <asm/atomic.h> #include "netlabel_user.h" +#include "netlabel_addrlist.h" #include "netlabel_domainhash.h" #include "netlabel_unlabeled.h" #include "netlabel_mgmt.h" @@ -76,22 +77,20 @@ struct netlbl_unlhsh_tbl { struct list_head *tbl; u32 size; }; +#define netlbl_unlhsh_addr4_entry(iter) \ + container_of(iter, struct netlbl_unlhsh_addr4, list) struct netlbl_unlhsh_addr4 { - __be32 addr; - __be32 mask; u32 secid; - u32 valid; - struct list_head list; + struct netlbl_af4list list; struct rcu_head rcu; }; +#define netlbl_unlhsh_addr6_entry(iter) \ + container_of(iter, struct netlbl_unlhsh_addr6, list) struct netlbl_unlhsh_addr6 { - struct in6_addr addr; - struct in6_addr mask; u32 secid; - u32 valid; - struct list_head list; + struct netlbl_af6list list; struct rcu_head rcu; }; struct netlbl_unlhsh_iface { @@ -147,76 +146,6 @@ static const struct nla_policy netlbl_unlabel_genl_policy[NLBL_UNLABEL_A_MAX + 1 }; /* - * Audit Helper Functions - */ - -/** - * netlbl_unlabel_audit_addr4 - Audit an IPv4 address - * @audit_buf: audit buffer - * @dev: network interface - * @addr: IP address - * @mask: IP address mask - * - * Description: - * Write the IPv4 address and address mask, if necessary, to @audit_buf. - * - */ -static void netlbl_unlabel_audit_addr4(struct audit_buffer *audit_buf, - const char *dev, - __be32 addr, __be32 mask) -{ - u32 mask_val = ntohl(mask); - - if (dev != NULL) - audit_log_format(audit_buf, " netif=%s", dev); - audit_log_format(audit_buf, " src=" NIPQUAD_FMT, NIPQUAD(addr)); - if (mask_val != 0xffffffff) { - u32 mask_len = 0; - while (mask_val > 0) { - mask_val <<= 1; - mask_len++; - } - audit_log_format(audit_buf, " src_prefixlen=%d", mask_len); - } -} - -#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) -/** - * netlbl_unlabel_audit_addr6 - Audit an IPv6 address - * @audit_buf: audit buffer - * @dev: network interface - * @addr: IP address - * @mask: IP address mask - * - * Description: - * Write the IPv6 address and address mask, if necessary, to @audit_buf. - * - */ -static void netlbl_unlabel_audit_addr6(struct audit_buffer *audit_buf, - const char *dev, - const struct in6_addr *addr, - const struct in6_addr *mask) -{ - if (dev != NULL) - audit_log_format(audit_buf, " netif=%s", dev); - audit_log_format(audit_buf, " src=" NIP6_FMT, NIP6(*addr)); - if (ntohl(mask->s6_addr32[3]) != 0xffffffff) { - u32 mask_len = 0; - u32 mask_val; - int iter = -1; - while (ntohl(mask->s6_addr32[++iter]) == 0xffffffff) - mask_len += 32; - mask_val = ntohl(mask->s6_addr32[iter]); - while (mask_val > 0) { - mask_val <<= 1; - mask_len++; - } - audit_log_format(audit_buf, " src_prefixlen=%d", mask_len); - } -} -#endif /* IPv6 */ - -/* * Unlabeled Connection Hash Table Functions */ @@ -274,26 +203,28 @@ static void netlbl_unlhsh_free_addr6(struct rcu_head *entry) static void netlbl_unlhsh_free_iface(struct rcu_head *entry) { struct netlbl_unlhsh_iface *iface; - struct netlbl_unlhsh_addr4 *iter4; - struct netlbl_unlhsh_addr4 *tmp4; - struct netlbl_unlhsh_addr6 *iter6; - struct netlbl_unlhsh_addr6 *tmp6; + struct netlbl_af4list *iter4; + struct netlbl_af4list *tmp4; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + struct netlbl_af6list *iter6; + struct netlbl_af6list *tmp6; +#endif /* IPv6 */ iface = container_of(entry, struct netlbl_unlhsh_iface, rcu); /* no need for locks here since we are the only one with access to this * structure */ - list_for_each_entry_safe(iter4, tmp4, &iface->addr4_list, list) - if (iter4->valid) { - list_del_rcu(&iter4->list); - kfree(iter4); - } - list_for_each_entry_safe(iter6, tmp6, &iface->addr6_list, list) - if (iter6->valid) { - list_del_rcu(&iter6->list); - kfree(iter6); - } + netlbl_af4list_foreach_safe(iter4, tmp4, &iface->addr4_list) { + netlbl_af4list_remove_entry(iter4); + kfree(netlbl_unlhsh_addr4_entry(iter4)); + } +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + netlbl_af6list_foreach_safe(iter6, tmp6, &iface->addr6_list) { + netlbl_af6list_remove_entry(iter6); + kfree(netlbl_unlhsh_addr6_entry(iter6)); + } +#endif /* IPv6 */ kfree(iface); } @@ -316,59 +247,6 @@ static u32 netlbl_unlhsh_hash(int ifindex) } /** - * netlbl_unlhsh_search_addr4 - Search for a matching IPv4 address entry - * @addr: IPv4 address - * @iface: the network interface entry - * - * Description: - * Searches the IPv4 address list of the network interface specified by @iface. - * If a matching address entry is found it is returned, otherwise NULL is - * returned. The caller is responsible for calling the rcu_read_[un]lock() - * functions. - * - */ -static struct netlbl_unlhsh_addr4 *netlbl_unlhsh_search_addr4( - __be32 addr, - const struct netlbl_unlhsh_iface *iface) -{ - struct netlbl_unlhsh_addr4 *iter; - - list_for_each_entry_rcu(iter, &iface->addr4_list, list) - if (iter->valid && (addr & iter->mask) == iter->addr) - return iter; - - return NULL; -} - -#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) -/** - * netlbl_unlhsh_search_addr6 - Search for a matching IPv6 address entry - * @addr: IPv6 address - * @iface: the network interface entry - * - * Description: - * Searches the IPv6 address list of the network interface specified by @iface. - * If a matching address entry is found it is returned, otherwise NULL is - * returned. The caller is responsible for calling the rcu_read_[un]lock() - * functions. - * - */ -static struct netlbl_unlhsh_addr6 *netlbl_unlhsh_search_addr6( - const struct in6_addr *addr, - const struct netlbl_unlhsh_iface *iface) -{ - struct netlbl_unlhsh_addr6 *iter; - - list_for_each_entry_rcu(iter, &iface->addr6_list, list) - if (iter->valid && - ipv6_masked_addr_cmp(&iter->addr, &iter->mask, addr) == 0) - return iter; - - return NULL; -} -#endif /* IPv6 */ - -/** * netlbl_unlhsh_search_iface - Search for a matching interface entry * @ifindex: the network interface * @@ -381,12 +259,12 @@ static struct netlbl_unlhsh_addr6 *netlbl_unlhsh_search_addr6( static struct netlbl_unlhsh_iface *netlbl_unlhsh_search_iface(int ifindex) { u32 bkt; + struct list_head *bkt_list; struct netlbl_unlhsh_iface *iter; bkt = netlbl_unlhsh_hash(ifindex); - list_for_each_entry_rcu(iter, - &rcu_dereference(netlbl_unlhsh)->tbl[bkt], - list) + bkt_list = &rcu_dereference(netlbl_unlhsh)->tbl[bkt]; + list_for_each_entry_rcu(iter, bkt_list, list) if (iter->valid && iter->ifindex == ifindex) return iter; @@ -439,43 +317,26 @@ static int netlbl_unlhsh_add_addr4(struct netlbl_unlhsh_iface *iface, const struct in_addr *mask, u32 secid) { + int ret_val; struct netlbl_unlhsh_addr4 *entry; - struct netlbl_unlhsh_addr4 *iter; entry = kzalloc(sizeof(*entry), GFP_ATOMIC); if (entry == NULL) return -ENOMEM; - entry->addr = addr->s_addr & mask->s_addr; - entry->mask = mask->s_addr; - entry->secid = secid; - entry->valid = 1; + entry->list.addr = addr->s_addr & mask->s_addr; + entry->list.mask = mask->s_addr; + entry->list.valid = 1; INIT_RCU_HEAD(&entry->rcu); + entry->secid = secid; spin_lock(&netlbl_unlhsh_lock); - iter = netlbl_unlhsh_search_addr4(entry->addr, iface); - if (iter != NULL && - iter->addr == addr->s_addr && iter->mask == mask->s_addr) { - spin_unlock(&netlbl_unlhsh_lock); - kfree(entry); - return -EEXIST; - } - /* in order to speed up address searches through the list (the common - * case) we need to keep the list in order based on the size of the - * address mask such that the entry with the widest mask (smallest - * numerical value) appears first in the list */ - list_for_each_entry_rcu(iter, &iface->addr4_list, list) - if (iter->valid && - ntohl(entry->mask) > ntohl(iter->mask)) { - __list_add_rcu(&entry->list, - iter->list.prev, - &iter->list); - spin_unlock(&netlbl_unlhsh_lock); - return 0; - } - list_add_tail_rcu(&entry->list, &iface->addr4_list); + ret_val = netlbl_af4list_add(&entry->list, &iface->addr4_list); spin_unlock(&netlbl_unlhsh_lock); - return 0; + + if (ret_val != 0) + kfree(entry); + return ret_val; } #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) @@ -498,47 +359,29 @@ static int netlbl_unlhsh_add_addr6(struct netlbl_unlhsh_iface *iface, const struct in6_addr *mask, u32 secid) { + int ret_val; struct netlbl_unlhsh_addr6 *entry; - struct netlbl_unlhsh_addr6 *iter; entry = kzalloc(sizeof(*entry), GFP_ATOMIC); if (entry == NULL) return -ENOMEM; - ipv6_addr_copy(&entry->addr, addr); - entry->addr.s6_addr32[0] &= mask->s6_addr32[0]; - entry->addr.s6_addr32[1] &= mask->s6_addr32[1]; - entry->addr.s6_addr32[2] &= mask->s6_addr32[2]; - entry->addr.s6_addr32[3] &= mask->s6_addr32[3]; - ipv6_addr_copy(&entry->mask, mask); - entry->secid = secid; - entry->valid = 1; + ipv6_addr_copy(&entry->list.addr, addr); + entry->list.addr.s6_addr32[0] &= mask->s6_addr32[0]; + entry->list.addr.s6_addr32[1] &= mask->s6_addr32[1]; + entry->list.addr.s6_addr32[2] &= mask->s6_addr32[2]; + entry->list.addr.s6_addr32[3] &= mask->s6_addr32[3]; + ipv6_addr_copy(&entry->list.mask, mask); + entry->list.valid = 1; INIT_RCU_HEAD(&entry->rcu); + entry->secid = secid; spin_lock(&netlbl_unlhsh_lock); - iter = netlbl_unlhsh_search_addr6(&entry->addr, iface); - if (iter != NULL && - (ipv6_addr_equal(&iter->addr, addr) && - ipv6_addr_equal(&iter->mask, mask))) { - spin_unlock(&netlbl_unlhsh_lock); - kfree(entry); - return -EEXIST; - } - /* in order to speed up address searches through the list (the common - * case) we need to keep the list in order based on the size of the - * address mask such that the entry with the widest mask (smallest - * numerical value) appears first in the list */ - list_for_each_entry_rcu(iter, &iface->addr6_list, list) - if (iter->valid && - ipv6_addr_cmp(&entry->mask, &iter->mask) > 0) { - __list_add_rcu(&entry->list, - iter->list.prev, - &iter->list); - spin_unlock(&netlbl_unlhsh_lock); - return 0; - } - list_add_tail_rcu(&entry->list, &iface->addr6_list); + ret_val = netlbl_af6list_add(&entry->list, &iface->addr6_list); spin_unlock(&netlbl_unlhsh_lock); + + if (ret_val != 0) + kfree(entry); return 0; } #endif /* IPv6 */ @@ -658,10 +501,10 @@ static int netlbl_unlhsh_add(struct net *net, mask4 = (struct in_addr *)mask; ret_val = netlbl_unlhsh_add_addr4(iface, addr4, mask4, secid); if (audit_buf != NULL) - netlbl_unlabel_audit_addr4(audit_buf, - dev_name, - addr4->s_addr, - mask4->s_addr); + netlbl_af4list_audit_addr(audit_buf, 1, + dev_name, + addr4->s_addr, + mask4->s_addr); break; } #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) @@ -672,9 +515,9 @@ static int netlbl_unlhsh_add(struct net *net, mask6 = (struct in6_addr *)mask; ret_val = netlbl_unlhsh_add_addr6(iface, addr6, mask6, secid); if (audit_buf != NULL) - netlbl_unlabel_audit_addr6(audit_buf, - dev_name, - addr6, mask6); + netlbl_af6list_audit_addr(audit_buf, 1, + dev_name, + addr6, mask6); break; } #endif /* IPv6 */ @@ -719,35 +562,34 @@ static int netlbl_unlhsh_remove_addr4(struct net *net, const struct in_addr *mask, struct netlbl_audit *audit_info) { - int ret_val = -ENOENT; + int ret_val = 0; + struct netlbl_af4list *list_entry; struct netlbl_unlhsh_addr4 *entry; - struct audit_buffer *audit_buf = NULL; + struct audit_buffer *audit_buf; struct net_device *dev; - char *secctx = NULL; + char *secctx; u32 secctx_len; spin_lock(&netlbl_unlhsh_lock); - entry = netlbl_unlhsh_search_addr4(addr->s_addr, iface); - if (entry != NULL && - entry->addr == addr->s_addr && entry->mask == mask->s_addr) { - entry->valid = 0; - list_del_rcu(&entry->list); - ret_val = 0; - } + list_entry = netlbl_af4list_remove(addr->s_addr, mask->s_addr, + &iface->addr4_list); spin_unlock(&netlbl_unlhsh_lock); + if (list_entry == NULL) + ret_val = -ENOENT; + entry = netlbl_unlhsh_addr4_entry(list_entry); audit_buf = netlbl_audit_start_common(AUDIT_MAC_UNLBL_STCDEL, audit_info); if (audit_buf != NULL) { dev = dev_get_by_index(net, iface->ifindex); - netlbl_unlabel_audit_addr4(audit_buf, - (dev != NULL ? dev->name : NULL), - entry->addr, entry->mask); + netlbl_af4list_audit_addr(audit_buf, 1, + (dev != NULL ? dev->name : NULL), + addr->s_addr, mask->s_addr); if (dev != NULL) dev_put(dev); - if (security_secid_to_secctx(entry->secid, - &secctx, - &secctx_len) == 0) { + if (entry && security_secid_to_secctx(entry->secid, + &secctx, + &secctx_len) == 0) { audit_log_format(audit_buf, " sec_obj=%s", secctx); security_release_secctx(secctx, secctx_len); } @@ -781,36 +623,33 @@ static int netlbl_unlhsh_remove_addr6(struct net *net, const struct in6_addr *mask, struct netlbl_audit *audit_info) { - int ret_val = -ENOENT; + int ret_val = 0; + struct netlbl_af6list *list_entry; struct netlbl_unlhsh_addr6 *entry; - struct audit_buffer *audit_buf = NULL; + struct audit_buffer *audit_buf; struct net_device *dev; - char *secctx = NULL; + char *secctx; u32 secctx_len; spin_lock(&netlbl_unlhsh_lock); - entry = netlbl_unlhsh_search_addr6(addr, iface); - if (entry != NULL && - (ipv6_addr_equal(&entry->addr, addr) && - ipv6_addr_equal(&entry->mask, mask))) { - entry->valid = 0; - list_del_rcu(&entry->list); - ret_val = 0; - } + list_entry = netlbl_af6list_remove(addr, mask, &iface->addr6_list); spin_unlock(&netlbl_unlhsh_lock); + if (list_entry == NULL) + ret_val = -ENOENT; + entry = netlbl_unlhsh_addr6_entry(list_entry); audit_buf = netlbl_audit_start_common(AUDIT_MAC_UNLBL_STCDEL, audit_info); if (audit_buf != NULL) { dev = dev_get_by_index(net, iface->ifindex); - netlbl_unlabel_audit_addr6(audit_buf, - (dev != NULL ? dev->name : NULL), - addr, mask); + netlbl_af6list_audit_addr(audit_buf, 1, + (dev != NULL ? dev->name : NULL), + addr, mask); if (dev != NULL) dev_put(dev); - if (security_secid_to_secctx(entry->secid, - &secctx, - &secctx_len) == 0) { + if (entry && security_secid_to_secctx(entry->secid, + &secctx, + &secctx_len) == 0) { audit_log_format(audit_buf, " sec_obj=%s", secctx); security_release_secctx(secctx, secctx_len); } @@ -836,16 +675,18 @@ static int netlbl_unlhsh_remove_addr6(struct net *net, */ static void netlbl_unlhsh_condremove_iface(struct netlbl_unlhsh_iface *iface) { - struct netlbl_unlhsh_addr4 *iter4; - struct netlbl_unlhsh_addr6 *iter6; + struct netlbl_af4list *iter4; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + struct netlbl_af6list *iter6; +#endif /* IPv6 */ spin_lock(&netlbl_unlhsh_lock); - list_for_each_entry_rcu(iter4, &iface->addr4_list, list) - if (iter4->valid) - goto unlhsh_condremove_failure; - list_for_each_entry_rcu(iter6, &iface->addr6_list, list) - if (iter6->valid) - goto unlhsh_condremove_failure; + netlbl_af4list_foreach_rcu(iter4, &iface->addr4_list) + goto unlhsh_condremove_failure; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + netlbl_af6list_foreach_rcu(iter6, &iface->addr6_list) + goto unlhsh_condremove_failure; +#endif /* IPv6 */ iface->valid = 0; if (iface->ifindex > 0) list_del_rcu(&iface->list); @@ -1349,7 +1190,7 @@ static int netlbl_unlabel_staticlist_gen(u32 cmd, if (addr4) { struct in_addr addr_struct; - addr_struct.s_addr = addr4->addr; + addr_struct.s_addr = addr4->list.addr; ret_val = nla_put(cb_arg->skb, NLBL_UNLABEL_A_IPV4ADDR, sizeof(struct in_addr), @@ -1357,7 +1198,7 @@ static int netlbl_unlabel_staticlist_gen(u32 cmd, if (ret_val != 0) goto list_cb_failure; - addr_struct.s_addr = addr4->mask; + addr_struct.s_addr = addr4->list.mask; ret_val = nla_put(cb_arg->skb, NLBL_UNLABEL_A_IPV4MASK, sizeof(struct in_addr), @@ -1370,14 +1211,14 @@ static int netlbl_unlabel_staticlist_gen(u32 cmd, ret_val = nla_put(cb_arg->skb, NLBL_UNLABEL_A_IPV6ADDR, sizeof(struct in6_addr), - &addr6->addr); + &addr6->list.addr); if (ret_val != 0) goto list_cb_failure; ret_val = nla_put(cb_arg->skb, NLBL_UNLABEL_A_IPV6MASK, sizeof(struct in6_addr), - &addr6->mask); + &addr6->list.mask); if (ret_val != 0) goto list_cb_failure; @@ -1425,8 +1266,11 @@ static int netlbl_unlabel_staticlist(struct sk_buff *skb, u32 iter_bkt; u32 iter_chain = 0, iter_addr4 = 0, iter_addr6 = 0; struct netlbl_unlhsh_iface *iface; - struct netlbl_unlhsh_addr4 *addr4; - struct netlbl_unlhsh_addr6 *addr6; + struct list_head *iter_list; + struct netlbl_af4list *addr4; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + struct netlbl_af6list *addr6; +#endif cb_arg.nl_cb = cb; cb_arg.skb = skb; @@ -1436,44 +1280,43 @@ static int netlbl_unlabel_staticlist(struct sk_buff *skb, for (iter_bkt = skip_bkt; iter_bkt < rcu_dereference(netlbl_unlhsh)->size; iter_bkt++, iter_chain = 0, iter_addr4 = 0, iter_addr6 = 0) { - list_for_each_entry_rcu(iface, - &rcu_dereference(netlbl_unlhsh)->tbl[iter_bkt], - list) { + iter_list = &rcu_dereference(netlbl_unlhsh)->tbl[iter_bkt]; + list_for_each_entry_rcu(iface, iter_list, list) { if (!iface->valid || iter_chain++ < skip_chain) continue; - list_for_each_entry_rcu(addr4, - &iface->addr4_list, - list) { - if (!addr4->valid || iter_addr4++ < skip_addr4) + netlbl_af4list_foreach_rcu(addr4, + &iface->addr4_list) { + if (iter_addr4++ < skip_addr4) continue; if (netlbl_unlabel_staticlist_gen( - NLBL_UNLABEL_C_STATICLIST, - iface, - addr4, - NULL, - &cb_arg) < 0) { + NLBL_UNLABEL_C_STATICLIST, + iface, + netlbl_unlhsh_addr4_entry(addr4), + NULL, + &cb_arg) < 0) { iter_addr4--; iter_chain--; goto unlabel_staticlist_return; } } - list_for_each_entry_rcu(addr6, - &iface->addr6_list, - list) { - if (!addr6->valid || iter_addr6++ < skip_addr6) +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + netlbl_af6list_foreach_rcu(addr6, + &iface->addr6_list) { + if (iter_addr6++ < skip_addr6) continue; if (netlbl_unlabel_staticlist_gen( - NLBL_UNLABEL_C_STATICLIST, - iface, - NULL, - addr6, - &cb_arg) < 0) { + NLBL_UNLABEL_C_STATICLIST, + iface, + NULL, + netlbl_unlhsh_addr6_entry(addr6), + &cb_arg) < 0) { iter_addr6--; iter_chain--; goto unlabel_staticlist_return; } } +#endif /* IPv6 */ } } @@ -1504,9 +1347,12 @@ static int netlbl_unlabel_staticlistdef(struct sk_buff *skb, struct netlbl_unlhsh_iface *iface; u32 skip_addr4 = cb->args[0]; u32 skip_addr6 = cb->args[1]; - u32 iter_addr4 = 0, iter_addr6 = 0; - struct netlbl_unlhsh_addr4 *addr4; - struct netlbl_unlhsh_addr6 *addr6; + u32 iter_addr4 = 0; + struct netlbl_af4list *addr4; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + u32 iter_addr6 = 0; + struct netlbl_af6list *addr6; +#endif cb_arg.nl_cb = cb; cb_arg.skb = skb; @@ -1517,30 +1363,32 @@ static int netlbl_unlabel_staticlistdef(struct sk_buff *skb, if (iface == NULL || !iface->valid) goto unlabel_staticlistdef_return; - list_for_each_entry_rcu(addr4, &iface->addr4_list, list) { - if (!addr4->valid || iter_addr4++ < skip_addr4) + netlbl_af4list_foreach_rcu(addr4, &iface->addr4_list) { + if (iter_addr4++ < skip_addr4) continue; if (netlbl_unlabel_staticlist_gen(NLBL_UNLABEL_C_STATICLISTDEF, - iface, - addr4, - NULL, - &cb_arg) < 0) { + iface, + netlbl_unlhsh_addr4_entry(addr4), + NULL, + &cb_arg) < 0) { iter_addr4--; goto unlabel_staticlistdef_return; } } - list_for_each_entry_rcu(addr6, &iface->addr6_list, list) { - if (!addr6->valid || iter_addr6++ < skip_addr6) +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + netlbl_af6list_foreach_rcu(addr6, &iface->addr6_list) { + if (iter_addr6++ < skip_addr6) continue; if (netlbl_unlabel_staticlist_gen(NLBL_UNLABEL_C_STATICLISTDEF, - iface, - NULL, - addr6, - &cb_arg) < 0) { + iface, + NULL, + netlbl_unlhsh_addr6_entry(addr6), + &cb_arg) < 0) { iter_addr6--; goto unlabel_staticlistdef_return; } } +#endif /* IPv6 */ unlabel_staticlistdef_return: rcu_read_unlock(); @@ -1718,25 +1566,27 @@ int netlbl_unlabel_getattr(const struct sk_buff *skb, switch (family) { case PF_INET: { struct iphdr *hdr4; - struct netlbl_unlhsh_addr4 *addr4; + struct netlbl_af4list *addr4; hdr4 = ip_hdr(skb); - addr4 = netlbl_unlhsh_search_addr4(hdr4->saddr, iface); + addr4 = netlbl_af4list_search(hdr4->saddr, + &iface->addr4_list); if (addr4 == NULL) goto unlabel_getattr_nolabel; - secattr->attr.secid = addr4->secid; + secattr->attr.secid = netlbl_unlhsh_addr4_entry(addr4)->secid; break; } #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) case PF_INET6: { struct ipv6hdr *hdr6; - struct netlbl_unlhsh_addr6 *addr6; + struct netlbl_af6list *addr6; hdr6 = ipv6_hdr(skb); - addr6 = netlbl_unlhsh_search_addr6(&hdr6->saddr, iface); + addr6 = netlbl_af6list_search(&hdr6->saddr, + &iface->addr6_list); if (addr6 == NULL) goto unlabel_getattr_nolabel; - secattr->attr.secid = addr6->secid; + secattr->attr.secid = netlbl_unlhsh_addr6_entry(addr6)->secid; break; } #endif /* IPv6 */ diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index b0eacc0007cc..480184a857d2 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1,7 +1,7 @@ /* * NETLINK Kernel-user communication protocol. * - * Authors: Alan Cox <alan@redhat.com> + * Authors: Alan Cox <alan@lxorguk.ukuu.org.uk> * Alexey Kuznetsov <kuznet@ms2.inr.ac.ru> * * This program is free software; you can redistribute it and/or @@ -435,7 +435,7 @@ static int netlink_create(struct net *net, struct socket *sock, int protocol) return -EPROTONOSUPPORT; netlink_lock_table(); -#ifdef CONFIG_KMOD +#ifdef CONFIG_MODULES if (!nl_table[protocol].registered) { netlink_unlock_table(); request_module("net-pf-%d-proto-%d", PF_NETLINK, protocol); diff --git a/net/phonet/af_phonet.c b/net/phonet/af_phonet.c index 9e9c6fce11aa..b9d97effebe3 100644 --- a/net/phonet/af_phonet.c +++ b/net/phonet/af_phonet.c @@ -67,11 +67,10 @@ static int pn_socket_create(struct net *net, struct socket *sock, int protocol) } pnp = phonet_proto_get(protocol); -#ifdef CONFIG_KMOD if (pnp == NULL && request_module("net-pf-%d-proto-%d", PF_PHONET, protocol) == 0) pnp = phonet_proto_get(protocol); -#endif + if (pnp == NULL) return -EPROTONOSUPPORT; if (sock->type != pnp->sock_type) { diff --git a/net/rfkill/rfkill-input.c b/net/rfkill/rfkill-input.c index e5b69556bb5b..21124ec0a73d 100644 --- a/net/rfkill/rfkill-input.c +++ b/net/rfkill/rfkill-input.c @@ -16,6 +16,7 @@ #include <linux/workqueue.h> #include <linux/init.h> #include <linux/rfkill.h> +#include <linux/sched.h> #include "rfkill-input.h" diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 9974b3f04f05..8f457f1e0acf 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -494,7 +494,7 @@ struct tc_action *tcf_action_init_1(struct nlattr *nla, struct nlattr *est, a_o = tc_lookup_action_n(act_name); if (a_o == NULL) { -#ifdef CONFIG_KMOD +#ifdef CONFIG_MODULES rtnl_unlock(); request_module("act_%s", act_name); rtnl_lock(); diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c index 8eb79e92e94c..16e7ac9774e5 100644 --- a/net/sched/cls_api.c +++ b/net/sched/cls_api.c @@ -227,7 +227,7 @@ replay: err = -ENOENT; tp_ops = tcf_proto_lookup_ops(tca[TCA_KIND]); if (tp_ops == NULL) { -#ifdef CONFIG_KMOD +#ifdef CONFIG_MODULES struct nlattr *kind = tca[TCA_KIND]; char name[IFNAMSIZ]; diff --git a/net/sched/ematch.c b/net/sched/ematch.c index 5e6f82e0e6f3..e82519e548d7 100644 --- a/net/sched/ematch.c +++ b/net/sched/ematch.c @@ -224,7 +224,7 @@ static int tcf_em_validate(struct tcf_proto *tp, if (em->ops == NULL) { err = -ENOENT; -#ifdef CONFIG_KMOD +#ifdef CONFIG_MODULES __rtnl_unlock(); request_module("ematch-kind-%u", em_hdr->kind); rtnl_lock(); diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 1122c952aa99..b16ad2972c6b 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -764,7 +764,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue, struct qdisc_size_table *stab; ops = qdisc_lookup_ops(kind); -#ifdef CONFIG_KMOD +#ifdef CONFIG_MODULES if (ops == NULL && kind != NULL) { char name[IFNAMSIZ]; if (nla_strlcpy(name, kind, IFNAMSIZ) < IFNAMSIZ) { diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 7b5572d6beb5..93cd30ce6501 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -326,6 +326,7 @@ struct Qdisc_ops noop_qdisc_ops __read_mostly = { static struct netdev_queue noop_netdev_queue = { .qdisc = &noop_qdisc, + .qdisc_sleeping = &noop_qdisc, }; struct Qdisc noop_qdisc = { @@ -352,6 +353,7 @@ static struct Qdisc_ops noqueue_qdisc_ops __read_mostly = { static struct Qdisc noqueue_qdisc; static struct netdev_queue noqueue_netdev_queue = { .qdisc = &noqueue_qdisc, + .qdisc_sleeping = &noqueue_qdisc, }; static struct Qdisc noqueue_qdisc = { diff --git a/net/socket.c b/net/socket.c index 3e8d4e35c08f..2b7a4b5c9b72 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1142,7 +1142,7 @@ static int __sock_create(struct net *net, int family, int type, int protocol, sock->type = type; -#if defined(CONFIG_KMOD) +#ifdef CONFIG_MODULES /* Attempt to load a protocol module if the find failed. * * 12/09/1996 Marcin: But! this makes REALLY only sense, if the user diff --git a/net/sunrpc/auth.c b/net/sunrpc/auth.c index 6bfea9ed6869..436bf1b4b76c 100644 --- a/net/sunrpc/auth.c +++ b/net/sunrpc/auth.c @@ -83,10 +83,8 @@ rpcauth_create(rpc_authflavor_t pseudoflavor, struct rpc_clnt *clnt) if (flavor >= RPC_AUTH_MAXFLAVOR) goto out; -#ifdef CONFIG_KMOD if ((ops = auth_flavors[flavor]) == NULL) request_module("rpc-auth-%u", flavor); -#endif spin_lock(&rpc_authflavor_lock); ops = auth_flavors[flavor]; if (ops == NULL || !try_module_get(ops->owner)) { diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c index 76739e928d0d..4895c341e46d 100644 --- a/net/sunrpc/clnt.c +++ b/net/sunrpc/clnt.c @@ -174,7 +174,7 @@ static struct rpc_clnt * rpc_new_client(const struct rpc_create_args *args, stru clnt->cl_procinfo = version->procs; clnt->cl_maxproc = version->nrprocs; clnt->cl_protname = program->name; - clnt->cl_prog = program->number; + clnt->cl_prog = args->prognumber ? : program->number; clnt->cl_vers = version->number; clnt->cl_stats = program->stats; clnt->cl_metrics = rpc_alloc_iostats(clnt); @@ -213,10 +213,10 @@ static struct rpc_clnt * rpc_new_client(const struct rpc_create_args *args, stru } /* save the nodename */ - clnt->cl_nodelen = strlen(utsname()->nodename); + clnt->cl_nodelen = strlen(init_utsname()->nodename); if (clnt->cl_nodelen > UNX_MAXNODENAME) clnt->cl_nodelen = UNX_MAXNODENAME; - memcpy(clnt->cl_nodename, utsname()->nodename, clnt->cl_nodelen); + memcpy(clnt->cl_nodename, init_utsname()->nodename, clnt->cl_nodelen); rpc_register_client(clnt); return clnt; diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c index 24db2b4d12d3..41013dd66ac3 100644 --- a/net/sunrpc/rpcb_clnt.c +++ b/net/sunrpc/rpcb_clnt.c @@ -20,6 +20,7 @@ #include <linux/in6.h> #include <linux/kernel.h> #include <linux/errno.h> +#include <net/ipv6.h> #include <linux/sunrpc/clnt.h> #include <linux/sunrpc/sched.h> @@ -176,13 +177,12 @@ static struct rpc_clnt *rpcb_create(char *hostname, struct sockaddr *srvaddr, } static int rpcb_register_call(struct sockaddr *addr, size_t addrlen, - u32 version, struct rpc_message *msg, - int *result) + u32 version, struct rpc_message *msg) { struct rpc_clnt *rpcb_clnt; - int error = 0; + int result, error = 0; - *result = 0; + msg->rpc_resp = &result; rpcb_clnt = rpcb_create_local(addr, addrlen, version); if (!IS_ERR(rpcb_clnt)) { @@ -191,12 +191,15 @@ static int rpcb_register_call(struct sockaddr *addr, size_t addrlen, } else error = PTR_ERR(rpcb_clnt); - if (error < 0) + if (error < 0) { printk(KERN_WARNING "RPC: failed to contact local rpcbind " "server (errno %d).\n", -error); - dprintk("RPC: registration status %d/%d\n", error, *result); + return error; + } - return error; + if (!result) + return -EACCES; + return 0; } /** @@ -205,7 +208,11 @@ static int rpcb_register_call(struct sockaddr *addr, size_t addrlen, * @vers: RPC version number to bind * @prot: transport protocol to register * @port: port value to register - * @okay: OUT: result code + * + * Returns zero if the registration request was dispatched successfully + * and the rpcbind daemon returned success. Otherwise, returns an errno + * value that reflects the nature of the error (request could not be + * dispatched, timed out, or rpcbind returned an error). * * RPC services invoke this function to advertise their contact * information via the system's rpcbind daemon. RPC services @@ -217,15 +224,6 @@ static int rpcb_register_call(struct sockaddr *addr, size_t addrlen, * all registered transports for [program, version] from the local * rpcbind database. * - * Returns zero if the registration request was dispatched - * successfully and a reply was received. The rpcbind daemon's - * boolean result code is stored in *okay. - * - * Returns an errno value and sets *result to zero if there was - * some problem that prevented the rpcbind request from being - * dispatched, or if the rpcbind daemon did not respond within - * the timeout. - * * This function uses rpcbind protocol version 2 to contact the * local rpcbind daemon. * @@ -236,7 +234,7 @@ static int rpcb_register_call(struct sockaddr *addr, size_t addrlen, * IN6ADDR_ANY (ie available for all AF_INET and AF_INET6 * addresses). */ -int rpcb_register(u32 prog, u32 vers, int prot, unsigned short port, int *okay) +int rpcb_register(u32 prog, u32 vers, int prot, unsigned short port) { struct rpcbind_args map = { .r_prog = prog, @@ -246,7 +244,6 @@ int rpcb_register(u32 prog, u32 vers, int prot, unsigned short port, int *okay) }; struct rpc_message msg = { .rpc_argp = &map, - .rpc_resp = okay, }; dprintk("RPC: %sregistering (%u, %u, %d, %u) with local " @@ -259,7 +256,7 @@ int rpcb_register(u32 prog, u32 vers, int prot, unsigned short port, int *okay) return rpcb_register_call((struct sockaddr *)&rpcb_inaddr_loopback, sizeof(rpcb_inaddr_loopback), - RPCBVERS_2, &msg, okay); + RPCBVERS_2, &msg); } /* @@ -290,7 +287,7 @@ static int rpcb_register_netid4(struct sockaddr_in *address_to_register, return rpcb_register_call((struct sockaddr *)&rpcb_inaddr_loopback, sizeof(rpcb_inaddr_loopback), - RPCBVERS_4, msg, msg->rpc_resp); + RPCBVERS_4, msg); } /* @@ -304,10 +301,13 @@ static int rpcb_register_netid6(struct sockaddr_in6 *address_to_register, char buf[64]; /* Construct AF_INET6 universal address */ - snprintf(buf, sizeof(buf), - NIP6_FMT".%u.%u", - NIP6(address_to_register->sin6_addr), - port >> 8, port & 0xff); + if (ipv6_addr_any(&address_to_register->sin6_addr)) + snprintf(buf, sizeof(buf), "::.%u.%u", + port >> 8, port & 0xff); + else + snprintf(buf, sizeof(buf), NIP6_FMT".%u.%u", + NIP6(address_to_register->sin6_addr), + port >> 8, port & 0xff); map->r_addr = buf; dprintk("RPC: %sregistering [%u, %u, %s, '%s'] with " @@ -321,7 +321,7 @@ static int rpcb_register_netid6(struct sockaddr_in6 *address_to_register, return rpcb_register_call((struct sockaddr *)&rpcb_in6addr_loopback, sizeof(rpcb_in6addr_loopback), - RPCBVERS_4, msg, msg->rpc_resp); + RPCBVERS_4, msg); } /** @@ -330,7 +330,11 @@ static int rpcb_register_netid6(struct sockaddr_in6 *address_to_register, * @version: RPC version number of service to (un)register * @address: address family, IP address, and port to (un)register * @netid: netid of transport protocol to (un)register - * @result: result code from rpcbind RPC call + * + * Returns zero if the registration request was dispatched successfully + * and the rpcbind daemon returned success. Otherwise, returns an errno + * value that reflects the nature of the error (request could not be + * dispatched, timed out, or rpcbind returned an error). * * RPC services invoke this function to advertise their contact * information via the system's rpcbind daemon. RPC services @@ -342,15 +346,6 @@ static int rpcb_register_netid6(struct sockaddr_in6 *address_to_register, * to zero. Callers pass a netid of "" to unregister all * transport netids associated with [program, version, address]. * - * Returns zero if the registration request was dispatched - * successfully and a reply was received. The rpcbind daemon's - * result code is stored in *result. - * - * Returns an errno value and sets *result to zero if there was - * some problem that prevented the rpcbind request from being - * dispatched, or if the rpcbind daemon did not respond within - * the timeout. - * * This function uses rpcbind protocol version 4 to contact the * local rpcbind daemon. The local rpcbind daemon must support * version 4 of the rpcbind protocol in order for these functions @@ -372,8 +367,7 @@ static int rpcb_register_netid6(struct sockaddr_in6 *address_to_register, * advertises the service on all IPv4 and IPv6 addresses. */ int rpcb_v4_register(const u32 program, const u32 version, - const struct sockaddr *address, const char *netid, - int *result) + const struct sockaddr *address, const char *netid) { struct rpcbind_args map = { .r_prog = program, @@ -383,11 +377,8 @@ int rpcb_v4_register(const u32 program, const u32 version, }; struct rpc_message msg = { .rpc_argp = &map, - .rpc_resp = result, }; - *result = 0; - switch (address->sa_family) { case AF_INET: return rpcb_register_netid4((struct sockaddr_in *)address, @@ -469,6 +460,28 @@ static struct rpc_task *rpcb_call_async(struct rpc_clnt *rpcb_clnt, struct rpcbi return rpc_run_task(&task_setup_data); } +/* + * In the case where rpc clients have been cloned, we want to make + * sure that we use the program number/version etc of the actual + * owner of the xprt. To do so, we walk back up the tree of parents + * to find whoever created the transport and/or whoever has the + * autobind flag set. + */ +static struct rpc_clnt *rpcb_find_transport_owner(struct rpc_clnt *clnt) +{ + struct rpc_clnt *parent = clnt->cl_parent; + + while (parent != clnt) { + if (parent->cl_xprt != clnt->cl_xprt) + break; + if (clnt->cl_autobind) + break; + clnt = parent; + parent = parent->cl_parent; + } + return clnt; +} + /** * rpcb_getport_async - obtain the port for a given RPC service on a given host * @task: task that is waiting for portmapper request @@ -478,10 +491,10 @@ static struct rpc_task *rpcb_call_async(struct rpc_clnt *rpcb_clnt, struct rpcbi */ void rpcb_getport_async(struct rpc_task *task) { - struct rpc_clnt *clnt = task->tk_client; + struct rpc_clnt *clnt; struct rpc_procinfo *proc; u32 bind_version; - struct rpc_xprt *xprt = task->tk_xprt; + struct rpc_xprt *xprt; struct rpc_clnt *rpcb_clnt; static struct rpcbind_args *map; struct rpc_task *child; @@ -490,13 +503,13 @@ void rpcb_getport_async(struct rpc_task *task) size_t salen; int status; + clnt = rpcb_find_transport_owner(task->tk_client); + xprt = clnt->cl_xprt; + dprintk("RPC: %5u %s(%s, %u, %u, %d)\n", task->tk_pid, __func__, clnt->cl_server, clnt->cl_prog, clnt->cl_vers, xprt->prot); - /* Autobind on cloned rpc clients is discouraged */ - BUG_ON(clnt->cl_parent != clnt); - /* Put self on the wait queue to ensure we get notified if * some other task is already attempting to bind the port */ rpc_sleep_on(&xprt->binding, task, NULL); @@ -558,7 +571,7 @@ void rpcb_getport_async(struct rpc_task *task) status = -ENOMEM; dprintk("RPC: %5u %s: no memory available\n", task->tk_pid, __func__); - goto bailout_nofree; + goto bailout_release_client; } map->r_prog = clnt->cl_prog; map->r_vers = clnt->cl_vers; @@ -578,11 +591,13 @@ void rpcb_getport_async(struct rpc_task *task) task->tk_pid, __func__); return; } - rpc_put_task(child); - task->tk_xprt->stat.bind_count++; + xprt->stat.bind_count++; + rpc_put_task(child); return; +bailout_release_client: + rpc_release_client(rpcb_clnt); bailout_nofree: rpcb_wake_rpcbind_waiters(xprt, status); task->tk_status = status; @@ -633,7 +648,7 @@ static void rpcb_getport_done(struct rpc_task *child, void *data) static int rpcb_encode_mapping(struct rpc_rqst *req, __be32 *p, struct rpcbind_args *rpcb) { - dprintk("RPC: rpcb_encode_mapping(%u, %u, %d, %u)\n", + dprintk("RPC: encoding rpcb request (%u, %u, %d, %u)\n", rpcb->r_prog, rpcb->r_vers, rpcb->r_prot, rpcb->r_port); *p++ = htonl(rpcb->r_prog); *p++ = htonl(rpcb->r_vers); @@ -648,7 +663,7 @@ static int rpcb_decode_getport(struct rpc_rqst *req, __be32 *p, unsigned short *portp) { *portp = (unsigned short) ntohl(*p++); - dprintk("RPC: rpcb_decode_getport result %u\n", + dprintk("RPC: rpcb getport result: %u\n", *portp); return 0; } @@ -657,7 +672,7 @@ static int rpcb_decode_set(struct rpc_rqst *req, __be32 *p, unsigned int *boolp) { *boolp = (unsigned int) ntohl(*p++); - dprintk("RPC: rpcb_decode_set: call %s\n", + dprintk("RPC: rpcb set/unset call %s\n", (*boolp ? "succeeded" : "failed")); return 0; } @@ -665,7 +680,7 @@ static int rpcb_decode_set(struct rpc_rqst *req, __be32 *p, static int rpcb_encode_getaddr(struct rpc_rqst *req, __be32 *p, struct rpcbind_args *rpcb) { - dprintk("RPC: rpcb_encode_getaddr(%u, %u, %s)\n", + dprintk("RPC: encoding rpcb request (%u, %u, %s)\n", rpcb->r_prog, rpcb->r_vers, rpcb->r_addr); *p++ = htonl(rpcb->r_prog); *p++ = htonl(rpcb->r_vers); diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c index 5a32cb7c4bb4..54c98d876847 100644 --- a/net/sunrpc/svc.c +++ b/net/sunrpc/svc.c @@ -28,6 +28,8 @@ #define RPCDBG_FACILITY RPCDBG_SVCDSP +static void svc_unregister(const struct svc_serv *serv); + #define svc_serv_is_pooled(serv) ((serv)->sv_function) /* @@ -357,7 +359,7 @@ svc_pool_for_cpu(struct svc_serv *serv, int cpu) */ static struct svc_serv * __svc_create(struct svc_program *prog, unsigned int bufsize, int npools, - void (*shutdown)(struct svc_serv *serv)) + sa_family_t family, void (*shutdown)(struct svc_serv *serv)) { struct svc_serv *serv; unsigned int vers; @@ -366,6 +368,7 @@ __svc_create(struct svc_program *prog, unsigned int bufsize, int npools, if (!(serv = kzalloc(sizeof(*serv), GFP_KERNEL))) return NULL; + serv->sv_family = family; serv->sv_name = prog->pg_name; serv->sv_program = prog; serv->sv_nrthreads = 1; @@ -416,30 +419,29 @@ __svc_create(struct svc_program *prog, unsigned int bufsize, int npools, spin_lock_init(&pool->sp_lock); } - /* Remove any stale portmap registrations */ - svc_register(serv, 0, 0); + svc_unregister(serv); return serv; } struct svc_serv * svc_create(struct svc_program *prog, unsigned int bufsize, - void (*shutdown)(struct svc_serv *serv)) + sa_family_t family, void (*shutdown)(struct svc_serv *serv)) { - return __svc_create(prog, bufsize, /*npools*/1, shutdown); + return __svc_create(prog, bufsize, /*npools*/1, family, shutdown); } EXPORT_SYMBOL(svc_create); struct svc_serv * svc_create_pooled(struct svc_program *prog, unsigned int bufsize, - void (*shutdown)(struct svc_serv *serv), + sa_family_t family, void (*shutdown)(struct svc_serv *serv), svc_thread_fn func, struct module *mod) { struct svc_serv *serv; unsigned int npools = svc_pool_map_get(); - serv = __svc_create(prog, bufsize, npools, shutdown); + serv = __svc_create(prog, bufsize, npools, family, shutdown); if (serv != NULL) { serv->sv_function = func; @@ -486,8 +488,7 @@ svc_destroy(struct svc_serv *serv) if (svc_serv_is_pooled(serv)) svc_pool_map_put(); - /* Unregister service with the portmapper */ - svc_register(serv, 0, 0); + svc_unregister(serv); kfree(serv->sv_pools); kfree(serv); } @@ -718,55 +719,245 @@ svc_exit_thread(struct svc_rqst *rqstp) } EXPORT_SYMBOL(svc_exit_thread); +#ifdef CONFIG_SUNRPC_REGISTER_V4 + /* - * Register an RPC service with the local portmapper. - * To unregister a service, call this routine with - * proto and port == 0. + * Register an "inet" protocol family netid with the local + * rpcbind daemon via an rpcbind v4 SET request. + * + * No netconfig infrastructure is available in the kernel, so + * we map IP_ protocol numbers to netids by hand. + * + * Returns zero on success; a negative errno value is returned + * if any error occurs. */ -int -svc_register(struct svc_serv *serv, int proto, unsigned short port) +static int __svc_rpcb_register4(const u32 program, const u32 version, + const unsigned short protocol, + const unsigned short port) +{ + struct sockaddr_in sin = { + .sin_family = AF_INET, + .sin_addr.s_addr = htonl(INADDR_ANY), + .sin_port = htons(port), + }; + char *netid; + + switch (protocol) { + case IPPROTO_UDP: + netid = RPCBIND_NETID_UDP; + break; + case IPPROTO_TCP: + netid = RPCBIND_NETID_TCP; + break; + default: + return -EPROTONOSUPPORT; + } + + return rpcb_v4_register(program, version, + (struct sockaddr *)&sin, netid); +} + +/* + * Register an "inet6" protocol family netid with the local + * rpcbind daemon via an rpcbind v4 SET request. + * + * No netconfig infrastructure is available in the kernel, so + * we map IP_ protocol numbers to netids by hand. + * + * Returns zero on success; a negative errno value is returned + * if any error occurs. + */ +static int __svc_rpcb_register6(const u32 program, const u32 version, + const unsigned short protocol, + const unsigned short port) +{ + struct sockaddr_in6 sin6 = { + .sin6_family = AF_INET6, + .sin6_addr = IN6ADDR_ANY_INIT, + .sin6_port = htons(port), + }; + char *netid; + + switch (protocol) { + case IPPROTO_UDP: + netid = RPCBIND_NETID_UDP6; + break; + case IPPROTO_TCP: + netid = RPCBIND_NETID_TCP6; + break; + default: + return -EPROTONOSUPPORT; + } + + return rpcb_v4_register(program, version, + (struct sockaddr *)&sin6, netid); +} + +/* + * Register a kernel RPC service via rpcbind version 4. + * + * Returns zero on success; a negative errno value is returned + * if any error occurs. + */ +static int __svc_register(const u32 program, const u32 version, + const sa_family_t family, + const unsigned short protocol, + const unsigned short port) +{ + int error; + + switch (family) { + case AF_INET: + return __svc_rpcb_register4(program, version, + protocol, port); + case AF_INET6: + error = __svc_rpcb_register6(program, version, + protocol, port); + if (error < 0) + return error; + + /* + * Work around bug in some versions of Linux rpcbind + * which don't allow registration of both inet and + * inet6 netids. + * + * Error return ignored for now. + */ + __svc_rpcb_register4(program, version, + protocol, port); + return 0; + } + + return -EAFNOSUPPORT; +} + +#else /* CONFIG_SUNRPC_REGISTER_V4 */ + +/* + * Register a kernel RPC service via rpcbind version 2. + * + * Returns zero on success; a negative errno value is returned + * if any error occurs. + */ +static int __svc_register(const u32 program, const u32 version, + sa_family_t family, + const unsigned short protocol, + const unsigned short port) +{ + if (family != AF_INET) + return -EAFNOSUPPORT; + + return rpcb_register(program, version, protocol, port); +} + +#endif /* CONFIG_SUNRPC_REGISTER_V4 */ + +/** + * svc_register - register an RPC service with the local portmapper + * @serv: svc_serv struct for the service to register + * @proto: transport protocol number to advertise + * @port: port to advertise + * + * Service is registered for any address in serv's address family + */ +int svc_register(const struct svc_serv *serv, const unsigned short proto, + const unsigned short port) { struct svc_program *progp; - unsigned long flags; unsigned int i; - int error = 0, dummy; + int error = 0; - if (!port) - clear_thread_flag(TIF_SIGPENDING); + BUG_ON(proto == 0 && port == 0); for (progp = serv->sv_program; progp; progp = progp->pg_next) { for (i = 0; i < progp->pg_nvers; i++) { if (progp->pg_vers[i] == NULL) continue; - dprintk("svc: svc_register(%s, %s, %d, %d)%s\n", + dprintk("svc: svc_register(%sv%d, %s, %u, %u)%s\n", progp->pg_name, + i, proto == IPPROTO_UDP? "udp" : "tcp", port, - i, + serv->sv_family, progp->pg_vers[i]->vs_hidden? " (but not telling portmap)" : ""); if (progp->pg_vers[i]->vs_hidden) continue; - error = rpcb_register(progp->pg_prog, i, proto, port, &dummy); + error = __svc_register(progp->pg_prog, i, + serv->sv_family, proto, port); if (error < 0) break; - if (port && !dummy) { - error = -EACCES; - break; - } } } - if (!port) { - spin_lock_irqsave(¤t->sighand->siglock, flags); - recalc_sigpending(); - spin_unlock_irqrestore(¤t->sighand->siglock, flags); + return error; +} + +#ifdef CONFIG_SUNRPC_REGISTER_V4 + +static void __svc_unregister(const u32 program, const u32 version, + const char *progname) +{ + struct sockaddr_in6 sin6 = { + .sin6_family = AF_INET6, + .sin6_addr = IN6ADDR_ANY_INIT, + .sin6_port = 0, + }; + int error; + + error = rpcb_v4_register(program, version, + (struct sockaddr *)&sin6, ""); + dprintk("svc: %s(%sv%u), error %d\n", + __func__, progname, version, error); +} + +#else /* CONFIG_SUNRPC_REGISTER_V4 */ + +static void __svc_unregister(const u32 program, const u32 version, + const char *progname) +{ + int error; + + error = rpcb_register(program, version, 0, 0); + dprintk("svc: %s(%sv%u), error %d\n", + __func__, progname, version, error); +} + +#endif /* CONFIG_SUNRPC_REGISTER_V4 */ + +/* + * All netids, bind addresses and ports registered for [program, version] + * are removed from the local rpcbind database (if the service is not + * hidden) to make way for a new instance of the service. + * + * The result of unregistration is reported via dprintk for those who want + * verification of the result, but is otherwise not important. + */ +static void svc_unregister(const struct svc_serv *serv) +{ + struct svc_program *progp; + unsigned long flags; + unsigned int i; + + clear_thread_flag(TIF_SIGPENDING); + + for (progp = serv->sv_program; progp; progp = progp->pg_next) { + for (i = 0; i < progp->pg_nvers; i++) { + if (progp->pg_vers[i] == NULL) + continue; + if (progp->pg_vers[i]->vs_hidden) + continue; + + __svc_unregister(progp->pg_prog, i, progp->pg_name); + } } - return error; + spin_lock_irqsave(¤t->sighand->siglock, flags); + recalc_sigpending(); + spin_unlock_irqrestore(¤t->sighand->siglock, flags); } /* diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c index e46c825f4954..bf5b5cdafebf 100644 --- a/net/sunrpc/svc_xprt.c +++ b/net/sunrpc/svc_xprt.c @@ -159,15 +159,44 @@ void svc_xprt_init(struct svc_xprt_class *xcl, struct svc_xprt *xprt, } EXPORT_SYMBOL_GPL(svc_xprt_init); -int svc_create_xprt(struct svc_serv *serv, char *xprt_name, unsigned short port, - int flags) +static struct svc_xprt *__svc_xpo_create(struct svc_xprt_class *xcl, + struct svc_serv *serv, + unsigned short port, int flags) { - struct svc_xprt_class *xcl; struct sockaddr_in sin = { .sin_family = AF_INET, .sin_addr.s_addr = htonl(INADDR_ANY), .sin_port = htons(port), }; + struct sockaddr_in6 sin6 = { + .sin6_family = AF_INET6, + .sin6_addr = IN6ADDR_ANY_INIT, + .sin6_port = htons(port), + }; + struct sockaddr *sap; + size_t len; + + switch (serv->sv_family) { + case AF_INET: + sap = (struct sockaddr *)&sin; + len = sizeof(sin); + break; + case AF_INET6: + sap = (struct sockaddr *)&sin6; + len = sizeof(sin6); + break; + default: + return ERR_PTR(-EAFNOSUPPORT); + } + + return xcl->xcl_ops->xpo_create(serv, sap, len, flags); +} + +int svc_create_xprt(struct svc_serv *serv, char *xprt_name, unsigned short port, + int flags) +{ + struct svc_xprt_class *xcl; + dprintk("svc: creating transport %s[%d]\n", xprt_name, port); spin_lock(&svc_xprt_class_lock); list_for_each_entry(xcl, &svc_xprt_class_list, xcl_list) { @@ -180,9 +209,7 @@ int svc_create_xprt(struct svc_serv *serv, char *xprt_name, unsigned short port, goto err; spin_unlock(&svc_xprt_class_lock); - newxprt = xcl->xcl_ops-> - xpo_create(serv, (struct sockaddr *)&sin, sizeof(sin), - flags); + newxprt = __svc_xpo_create(xcl, serv, port, flags); if (IS_ERR(newxprt)) { module_put(xcl->xcl_owner); return PTR_ERR(newxprt); diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c index 3e65719f1ef6..95293f549e9c 100644 --- a/net/sunrpc/svcsock.c +++ b/net/sunrpc/svcsock.c @@ -1114,6 +1114,7 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, struct svc_sock *svsk; struct sock *inet; int pmap_register = !(flags & SVC_SOCK_ANONYMOUS); + int val; dprintk("svc: svc_setup_socket %p\n", sock); if (!(svsk = kzalloc(sizeof(*svsk), GFP_KERNEL))) { @@ -1146,6 +1147,18 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, else svc_tcp_init(svsk, serv); + /* + * We start one listener per sv_serv. We want AF_INET + * requests to be automatically shunted to our AF_INET6 + * listener using a mapped IPv4 address. Make sure + * no-one starts an equivalent IPv4 listener, which + * would steal our incoming connections. + */ + val = 0; + if (serv->sv_family == AF_INET6) + kernel_setsockopt(sock, SOL_IPV6, IPV6_V6ONLY, + (char *)&val, sizeof(val)); + dprintk("svc: svc_setup_socket created %p (inet %p)\n", svsk, svsk->sk_sk); @@ -1154,8 +1167,7 @@ static struct svc_sock *svc_setup_socket(struct svc_serv *serv, int svc_addsock(struct svc_serv *serv, int fd, - char *name_return, - int *proto) + char *name_return) { int err = 0; struct socket *so = sockfd_lookup(fd, &err); @@ -1190,7 +1202,6 @@ int svc_addsock(struct svc_serv *serv, sockfd_put(so); return err; } - if (proto) *proto = so->sk->sk_protocol; return one_sock_name(name_return, svsk); } EXPORT_SYMBOL_GPL(svc_addsock); diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c index 99a52aabe332..29e401bb612e 100644 --- a/net/sunrpc/xprt.c +++ b/net/sunrpc/xprt.c @@ -108,13 +108,10 @@ int xprt_register_transport(struct xprt_class *transport) goto out; } - result = -EINVAL; - if (try_module_get(THIS_MODULE)) { - list_add_tail(&transport->list, &xprt_list); - printk(KERN_INFO "RPC: Registered %s transport module.\n", - transport->name); - result = 0; - } + list_add_tail(&transport->list, &xprt_list); + printk(KERN_INFO "RPC: Registered %s transport module.\n", + transport->name); + result = 0; out: spin_unlock(&xprt_list_lock); @@ -143,7 +140,6 @@ int xprt_unregister_transport(struct xprt_class *transport) "RPC: Unregistered %s transport module.\n", transport->name); list_del_init(&transport->list); - module_put(THIS_MODULE); goto out; } } diff --git a/net/sunrpc/xprtrdma/rpc_rdma.c b/net/sunrpc/xprtrdma/rpc_rdma.c index 5c1954d28d09..14106d26bb95 100644 --- a/net/sunrpc/xprtrdma/rpc_rdma.c +++ b/net/sunrpc/xprtrdma/rpc_rdma.c @@ -118,6 +118,10 @@ rpcrdma_convert_iovs(struct xdr_buf *xdrbuf, unsigned int pos, } if (xdrbuf->tail[0].iov_len) { + /* the rpcrdma protocol allows us to omit any trailing + * xdr pad bytes, saving the server an RDMA operation. */ + if (xdrbuf->tail[0].iov_len < 4 && xprt_rdma_pad_optimize) + return n; if (n == nsegs) return 0; seg[n].mr_page = NULL; @@ -508,8 +512,8 @@ rpcrdma_marshal_req(struct rpc_rqst *rqst) if (hdrlen == 0) return -1; - dprintk("RPC: %s: %s: hdrlen %zd rpclen %zd padlen %zd\n" - " headerp 0x%p base 0x%p lkey 0x%x\n", + dprintk("RPC: %s: %s: hdrlen %zd rpclen %zd padlen %zd" + " headerp 0x%p base 0x%p lkey 0x%x\n", __func__, transfertypes[wtype], hdrlen, rpclen, padlen, headerp, base, req->rl_iov.lkey); @@ -594,7 +598,7 @@ rpcrdma_count_chunks(struct rpcrdma_rep *rep, unsigned int max, int wrchunk, __b * Scatter inline received data back into provided iov's. */ static void -rpcrdma_inline_fixup(struct rpc_rqst *rqst, char *srcp, int copy_len) +rpcrdma_inline_fixup(struct rpc_rqst *rqst, char *srcp, int copy_len, int pad) { int i, npages, curlen, olen; char *destp; @@ -660,6 +664,13 @@ rpcrdma_inline_fixup(struct rpc_rqst *rqst, char *srcp, int copy_len) } else rqst->rq_rcv_buf.tail[0].iov_len = 0; + if (pad) { + /* implicit padding on terminal chunk */ + unsigned char *p = rqst->rq_rcv_buf.tail[0].iov_base; + while (pad--) + p[rqst->rq_rcv_buf.tail[0].iov_len++] = 0; + } + if (copy_len) dprintk("RPC: %s: %d bytes in" " %d extra segments (%d lost)\n", @@ -681,12 +692,14 @@ rpcrdma_conn_func(struct rpcrdma_ep *ep) struct rpc_xprt *xprt = ep->rep_xprt; spin_lock_bh(&xprt->transport_lock); + if (++xprt->connect_cookie == 0) /* maintain a reserved value */ + ++xprt->connect_cookie; if (ep->rep_connected > 0) { if (!xprt_test_and_set_connected(xprt)) xprt_wake_pending_tasks(xprt, 0); } else { if (xprt_test_and_clear_connected(xprt)) - xprt_wake_pending_tasks(xprt, ep->rep_connected); + xprt_wake_pending_tasks(xprt, -ENOTCONN); } spin_unlock_bh(&xprt->transport_lock); } @@ -792,14 +805,20 @@ repost: ((unsigned char *)iptr - (unsigned char *)headerp); status = rep->rr_len + rdmalen; r_xprt->rx_stats.total_rdma_reply += rdmalen; + /* special case - last chunk may omit padding */ + if (rdmalen &= 3) { + rdmalen = 4 - rdmalen; + status += rdmalen; + } } else { /* else ordinary inline */ + rdmalen = 0; iptr = (__be32 *)((unsigned char *)headerp + 28); rep->rr_len -= 28; /*sizeof *headerp;*/ status = rep->rr_len; } /* Fix up the rpc results for upper layer */ - rpcrdma_inline_fixup(rqst, (char *)iptr, rep->rr_len); + rpcrdma_inline_fixup(rqst, (char *)iptr, rep->rr_len, rdmalen); break; case htonl(RDMA_NOMSG): diff --git a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c index 74de31a06616..a4756576d687 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c +++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c @@ -116,7 +116,7 @@ static void rdma_build_arg_xdr(struct svc_rqst *rqstp, * * Assumptions: * - chunk[0]->position points to pages[0] at an offset of 0 - * - pages[] is not physically or virtually contigous and consists of + * - pages[] is not physically or virtually contiguous and consists of * PAGE_SIZE elements. * * Output: @@ -125,7 +125,7 @@ static void rdma_build_arg_xdr(struct svc_rqst *rqstp, * chunk in the read list * */ -static int rdma_rcl_to_sge(struct svcxprt_rdma *xprt, +static int map_read_chunks(struct svcxprt_rdma *xprt, struct svc_rqst *rqstp, struct svc_rdma_op_ctxt *head, struct rpcrdma_msg *rmsgp, @@ -211,26 +211,128 @@ static int rdma_rcl_to_sge(struct svcxprt_rdma *xprt, return sge_no; } -static void rdma_set_ctxt_sge(struct svcxprt_rdma *xprt, - struct svc_rdma_op_ctxt *ctxt, - struct kvec *vec, - u64 *sgl_offset, - int count) +/* Map a read-chunk-list to an XDR and fast register the page-list. + * + * Assumptions: + * - chunk[0] position points to pages[0] at an offset of 0 + * - pages[] will be made physically contiguous by creating a one-off memory + * region using the fastreg verb. + * - byte_count is # of bytes in read-chunk-list + * - ch_count is # of chunks in read-chunk-list + * + * Output: + * - sge array pointing into pages[] array. + * - chunk_sge array specifying sge index and count for each + * chunk in the read list + */ +static int fast_reg_read_chunks(struct svcxprt_rdma *xprt, + struct svc_rqst *rqstp, + struct svc_rdma_op_ctxt *head, + struct rpcrdma_msg *rmsgp, + struct svc_rdma_req_map *rpl_map, + struct svc_rdma_req_map *chl_map, + int ch_count, + int byte_count) +{ + int page_no; + int ch_no; + u32 offset; + struct rpcrdma_read_chunk *ch; + struct svc_rdma_fastreg_mr *frmr; + int ret = 0; + + frmr = svc_rdma_get_frmr(xprt); + if (IS_ERR(frmr)) + return -ENOMEM; + + head->frmr = frmr; + head->arg.head[0] = rqstp->rq_arg.head[0]; + head->arg.tail[0] = rqstp->rq_arg.tail[0]; + head->arg.pages = &head->pages[head->count]; + head->hdr_count = head->count; /* save count of hdr pages */ + head->arg.page_base = 0; + head->arg.page_len = byte_count; + head->arg.len = rqstp->rq_arg.len + byte_count; + head->arg.buflen = rqstp->rq_arg.buflen + byte_count; + + /* Fast register the page list */ + frmr->kva = page_address(rqstp->rq_arg.pages[0]); + frmr->direction = DMA_FROM_DEVICE; + frmr->access_flags = (IB_ACCESS_LOCAL_WRITE|IB_ACCESS_REMOTE_WRITE); + frmr->map_len = byte_count; + frmr->page_list_len = PAGE_ALIGN(byte_count) >> PAGE_SHIFT; + for (page_no = 0; page_no < frmr->page_list_len; page_no++) { + frmr->page_list->page_list[page_no] = + ib_dma_map_single(xprt->sc_cm_id->device, + page_address(rqstp->rq_arg.pages[page_no]), + PAGE_SIZE, DMA_TO_DEVICE); + if (ib_dma_mapping_error(xprt->sc_cm_id->device, + frmr->page_list->page_list[page_no])) + goto fatal_err; + atomic_inc(&xprt->sc_dma_used); + head->arg.pages[page_no] = rqstp->rq_arg.pages[page_no]; + } + head->count += page_no; + + /* rq_respages points one past arg pages */ + rqstp->rq_respages = &rqstp->rq_arg.pages[page_no]; + + /* Create the reply and chunk maps */ + offset = 0; + ch = (struct rpcrdma_read_chunk *)&rmsgp->rm_body.rm_chunks[0]; + for (ch_no = 0; ch_no < ch_count; ch_no++) { + rpl_map->sge[ch_no].iov_base = frmr->kva + offset; + rpl_map->sge[ch_no].iov_len = ch->rc_target.rs_length; + chl_map->ch[ch_no].count = 1; + chl_map->ch[ch_no].start = ch_no; + offset += ch->rc_target.rs_length; + ch++; + } + + ret = svc_rdma_fastreg(xprt, frmr); + if (ret) + goto fatal_err; + + return ch_no; + + fatal_err: + printk("svcrdma: error fast registering xdr for xprt %p", xprt); + svc_rdma_put_frmr(xprt, frmr); + return -EIO; +} + +static int rdma_set_ctxt_sge(struct svcxprt_rdma *xprt, + struct svc_rdma_op_ctxt *ctxt, + struct svc_rdma_fastreg_mr *frmr, + struct kvec *vec, + u64 *sgl_offset, + int count) { int i; ctxt->count = count; ctxt->direction = DMA_FROM_DEVICE; for (i = 0; i < count; i++) { - atomic_inc(&xprt->sc_dma_used); - ctxt->sge[i].addr = - ib_dma_map_single(xprt->sc_cm_id->device, - vec[i].iov_base, vec[i].iov_len, - DMA_FROM_DEVICE); + ctxt->sge[i].length = 0; /* in case map fails */ + if (!frmr) { + ctxt->sge[i].addr = + ib_dma_map_single(xprt->sc_cm_id->device, + vec[i].iov_base, + vec[i].iov_len, + DMA_FROM_DEVICE); + if (ib_dma_mapping_error(xprt->sc_cm_id->device, + ctxt->sge[i].addr)) + return -EINVAL; + ctxt->sge[i].lkey = xprt->sc_dma_lkey; + atomic_inc(&xprt->sc_dma_used); + } else { + ctxt->sge[i].addr = (unsigned long)vec[i].iov_base; + ctxt->sge[i].lkey = frmr->mr->lkey; + } ctxt->sge[i].length = vec[i].iov_len; - ctxt->sge[i].lkey = xprt->sc_phys_mr->lkey; *sgl_offset = *sgl_offset + vec[i].iov_len; } + return 0; } static int rdma_read_max_sge(struct svcxprt_rdma *xprt, int sge_count) @@ -278,6 +380,7 @@ static int rdma_read_xdr(struct svcxprt_rdma *xprt, struct svc_rdma_op_ctxt *hdr_ctxt) { struct ib_send_wr read_wr; + struct ib_send_wr inv_wr; int err = 0; int ch_no; int ch_count; @@ -301,9 +404,20 @@ static int rdma_read_xdr(struct svcxprt_rdma *xprt, svc_rdma_rcl_chunk_counts(ch, &ch_count, &byte_count); if (ch_count > RPCSVC_MAXPAGES) return -EINVAL; - sge_count = rdma_rcl_to_sge(xprt, rqstp, hdr_ctxt, rmsgp, - rpl_map, chl_map, - ch_count, byte_count); + + if (!xprt->sc_frmr_pg_list_len) + sge_count = map_read_chunks(xprt, rqstp, hdr_ctxt, rmsgp, + rpl_map, chl_map, ch_count, + byte_count); + else + sge_count = fast_reg_read_chunks(xprt, rqstp, hdr_ctxt, rmsgp, + rpl_map, chl_map, ch_count, + byte_count); + if (sge_count < 0) { + err = -EIO; + goto out; + } + sgl_offset = 0; ch_no = 0; @@ -312,13 +426,16 @@ static int rdma_read_xdr(struct svcxprt_rdma *xprt, next_sge: ctxt = svc_rdma_get_context(xprt); ctxt->direction = DMA_FROM_DEVICE; + ctxt->frmr = hdr_ctxt->frmr; + ctxt->read_hdr = NULL; clear_bit(RDMACTXT_F_LAST_CTXT, &ctxt->flags); + clear_bit(RDMACTXT_F_FAST_UNREG, &ctxt->flags); /* Prepare READ WR */ memset(&read_wr, 0, sizeof read_wr); - ctxt->wr_op = IB_WR_RDMA_READ; read_wr.wr_id = (unsigned long)ctxt; read_wr.opcode = IB_WR_RDMA_READ; + ctxt->wr_op = read_wr.opcode; read_wr.send_flags = IB_SEND_SIGNALED; read_wr.wr.rdma.rkey = ch->rc_target.rs_handle; read_wr.wr.rdma.remote_addr = @@ -327,10 +444,15 @@ next_sge: read_wr.sg_list = ctxt->sge; read_wr.num_sge = rdma_read_max_sge(xprt, chl_map->ch[ch_no].count); - rdma_set_ctxt_sge(xprt, ctxt, - &rpl_map->sge[chl_map->ch[ch_no].start], - &sgl_offset, - read_wr.num_sge); + err = rdma_set_ctxt_sge(xprt, ctxt, hdr_ctxt->frmr, + &rpl_map->sge[chl_map->ch[ch_no].start], + &sgl_offset, + read_wr.num_sge); + if (err) { + svc_rdma_unmap_dma(ctxt); + svc_rdma_put_context(ctxt, 0); + goto out; + } if (((ch+1)->rc_discrim == 0) && (read_wr.num_sge == chl_map->ch[ch_no].count)) { /* @@ -339,6 +461,29 @@ next_sge: * the client and the RPC needs to be enqueued. */ set_bit(RDMACTXT_F_LAST_CTXT, &ctxt->flags); + if (hdr_ctxt->frmr) { + set_bit(RDMACTXT_F_FAST_UNREG, &ctxt->flags); + /* + * Invalidate the local MR used to map the data + * sink. + */ + if (xprt->sc_dev_caps & + SVCRDMA_DEVCAP_READ_W_INV) { + read_wr.opcode = + IB_WR_RDMA_READ_WITH_INV; + ctxt->wr_op = read_wr.opcode; + read_wr.ex.invalidate_rkey = + ctxt->frmr->mr->lkey; + } else { + /* Prepare INVALIDATE WR */ + memset(&inv_wr, 0, sizeof inv_wr); + inv_wr.opcode = IB_WR_LOCAL_INV; + inv_wr.send_flags = IB_SEND_SIGNALED; + inv_wr.ex.invalidate_rkey = + hdr_ctxt->frmr->mr->lkey; + read_wr.next = &inv_wr; + } + } ctxt->read_hdr = hdr_ctxt; } /* Post the read */ diff --git a/net/sunrpc/xprtrdma/svc_rdma_sendto.c b/net/sunrpc/xprtrdma/svc_rdma_sendto.c index 84d328329d98..9a7a8e7ae038 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_sendto.c +++ b/net/sunrpc/xprtrdma/svc_rdma_sendto.c @@ -69,9 +69,127 @@ * array is only concerned with the reply we are assured that we have * on extra page for the RPCRMDA header. */ -static void xdr_to_sge(struct svcxprt_rdma *xprt, - struct xdr_buf *xdr, - struct svc_rdma_req_map *vec) +int fast_reg_xdr(struct svcxprt_rdma *xprt, + struct xdr_buf *xdr, + struct svc_rdma_req_map *vec) +{ + int sge_no; + u32 sge_bytes; + u32 page_bytes; + u32 page_off; + int page_no = 0; + u8 *frva; + struct svc_rdma_fastreg_mr *frmr; + + frmr = svc_rdma_get_frmr(xprt); + if (IS_ERR(frmr)) + return -ENOMEM; + vec->frmr = frmr; + + /* Skip the RPCRDMA header */ + sge_no = 1; + + /* Map the head. */ + frva = (void *)((unsigned long)(xdr->head[0].iov_base) & PAGE_MASK); + vec->sge[sge_no].iov_base = xdr->head[0].iov_base; + vec->sge[sge_no].iov_len = xdr->head[0].iov_len; + vec->count = 2; + sge_no++; + + /* Build the FRMR */ + frmr->kva = frva; + frmr->direction = DMA_TO_DEVICE; + frmr->access_flags = 0; + frmr->map_len = PAGE_SIZE; + frmr->page_list_len = 1; + frmr->page_list->page_list[page_no] = + ib_dma_map_single(xprt->sc_cm_id->device, + (void *)xdr->head[0].iov_base, + PAGE_SIZE, DMA_TO_DEVICE); + if (ib_dma_mapping_error(xprt->sc_cm_id->device, + frmr->page_list->page_list[page_no])) + goto fatal_err; + atomic_inc(&xprt->sc_dma_used); + + page_off = xdr->page_base; + page_bytes = xdr->page_len + page_off; + if (!page_bytes) + goto encode_tail; + + /* Map the pages */ + vec->sge[sge_no].iov_base = frva + frmr->map_len + page_off; + vec->sge[sge_no].iov_len = page_bytes; + sge_no++; + while (page_bytes) { + struct page *page; + + page = xdr->pages[page_no++]; + sge_bytes = min_t(u32, page_bytes, (PAGE_SIZE - page_off)); + page_bytes -= sge_bytes; + + frmr->page_list->page_list[page_no] = + ib_dma_map_page(xprt->sc_cm_id->device, page, 0, + PAGE_SIZE, DMA_TO_DEVICE); + if (ib_dma_mapping_error(xprt->sc_cm_id->device, + frmr->page_list->page_list[page_no])) + goto fatal_err; + + atomic_inc(&xprt->sc_dma_used); + page_off = 0; /* reset for next time through loop */ + frmr->map_len += PAGE_SIZE; + frmr->page_list_len++; + } + vec->count++; + + encode_tail: + /* Map tail */ + if (0 == xdr->tail[0].iov_len) + goto done; + + vec->count++; + vec->sge[sge_no].iov_len = xdr->tail[0].iov_len; + + if (((unsigned long)xdr->tail[0].iov_base & PAGE_MASK) == + ((unsigned long)xdr->head[0].iov_base & PAGE_MASK)) { + /* + * If head and tail use the same page, we don't need + * to map it again. + */ + vec->sge[sge_no].iov_base = xdr->tail[0].iov_base; + } else { + void *va; + + /* Map another page for the tail */ + page_off = (unsigned long)xdr->tail[0].iov_base & ~PAGE_MASK; + va = (void *)((unsigned long)xdr->tail[0].iov_base & PAGE_MASK); + vec->sge[sge_no].iov_base = frva + frmr->map_len + page_off; + + frmr->page_list->page_list[page_no] = + ib_dma_map_single(xprt->sc_cm_id->device, va, PAGE_SIZE, + DMA_TO_DEVICE); + if (ib_dma_mapping_error(xprt->sc_cm_id->device, + frmr->page_list->page_list[page_no])) + goto fatal_err; + atomic_inc(&xprt->sc_dma_used); + frmr->map_len += PAGE_SIZE; + frmr->page_list_len++; + } + + done: + if (svc_rdma_fastreg(xprt, frmr)) + goto fatal_err; + + return 0; + + fatal_err: + printk("svcrdma: Error fast registering memory for xprt %p\n", xprt); + svc_rdma_put_frmr(xprt, frmr); + return -EIO; +} + +static int map_xdr(struct svcxprt_rdma *xprt, + struct xdr_buf *xdr, + struct svc_rdma_req_map *vec) { int sge_max = (xdr->len+PAGE_SIZE-1) / PAGE_SIZE + 3; int sge_no; @@ -83,6 +201,9 @@ static void xdr_to_sge(struct svcxprt_rdma *xprt, BUG_ON(xdr->len != (xdr->head[0].iov_len + xdr->page_len + xdr->tail[0].iov_len)); + if (xprt->sc_frmr_pg_list_len) + return fast_reg_xdr(xprt, xdr, vec); + /* Skip the first sge, this is for the RPCRDMA header */ sge_no = 1; @@ -116,9 +237,12 @@ static void xdr_to_sge(struct svcxprt_rdma *xprt, BUG_ON(sge_no > sge_max); vec->count = sge_no; + return 0; } /* Assumptions: + * - We are using FRMR + * - or - * - The specified write_len can be represented in sc_max_sge * PAGE_SIZE */ static int send_write(struct svcxprt_rdma *xprt, struct svc_rqst *rqstp, @@ -158,30 +282,35 @@ static int send_write(struct svcxprt_rdma *xprt, struct svc_rqst *rqstp, sge_no = 0; /* Copy the remaining SGE */ - while (bc != 0 && xdr_sge_no < vec->count) { - sge[sge_no].lkey = xprt->sc_phys_mr->lkey; - sge_bytes = min((size_t)bc, - (size_t)(vec->sge[xdr_sge_no].iov_len-sge_off)); + while (bc != 0) { + sge_bytes = min_t(size_t, + bc, vec->sge[xdr_sge_no].iov_len-sge_off); sge[sge_no].length = sge_bytes; - atomic_inc(&xprt->sc_dma_used); - sge[sge_no].addr = - ib_dma_map_single(xprt->sc_cm_id->device, - (void *) - vec->sge[xdr_sge_no].iov_base + sge_off, - sge_bytes, DMA_TO_DEVICE); - if (dma_mapping_error(xprt->sc_cm_id->device->dma_device, - sge[sge_no].addr)) - goto err; + if (!vec->frmr) { + sge[sge_no].addr = + ib_dma_map_single(xprt->sc_cm_id->device, + (void *) + vec->sge[xdr_sge_no].iov_base + sge_off, + sge_bytes, DMA_TO_DEVICE); + if (ib_dma_mapping_error(xprt->sc_cm_id->device, + sge[sge_no].addr)) + goto err; + atomic_inc(&xprt->sc_dma_used); + sge[sge_no].lkey = xprt->sc_dma_lkey; + } else { + sge[sge_no].addr = (unsigned long) + vec->sge[xdr_sge_no].iov_base + sge_off; + sge[sge_no].lkey = vec->frmr->mr->lkey; + } + ctxt->count++; + ctxt->frmr = vec->frmr; sge_off = 0; sge_no++; - ctxt->count++; xdr_sge_no++; + BUG_ON(xdr_sge_no > vec->count); bc -= sge_bytes; } - BUG_ON(bc != 0); - BUG_ON(xdr_sge_no > vec->count); - /* Prepare WRITE WR */ memset(&write_wr, 0, sizeof write_wr); ctxt->wr_op = IB_WR_RDMA_WRITE; @@ -226,7 +355,10 @@ static int send_write_chunks(struct svcxprt_rdma *xprt, res_ary = (struct rpcrdma_write_array *) &rdma_resp->rm_body.rm_chunks[1]; - max_write = xprt->sc_max_sge * PAGE_SIZE; + if (vec->frmr) + max_write = vec->frmr->map_len; + else + max_write = xprt->sc_max_sge * PAGE_SIZE; /* Write chunks start at the pagelist */ for (xdr_off = rqstp->rq_res.head[0].iov_len, chunk_no = 0; @@ -297,7 +429,10 @@ static int send_reply_chunks(struct svcxprt_rdma *xprt, res_ary = (struct rpcrdma_write_array *) &rdma_resp->rm_body.rm_chunks[2]; - max_write = xprt->sc_max_sge * PAGE_SIZE; + if (vec->frmr) + max_write = vec->frmr->map_len; + else + max_write = xprt->sc_max_sge * PAGE_SIZE; /* xdr offset starts at RPC message */ for (xdr_off = 0, chunk_no = 0; @@ -307,7 +442,6 @@ static int send_reply_chunks(struct svcxprt_rdma *xprt, ch = &arg_ary->wc_array[chunk_no].wc_target; write_len = min(xfer_len, ch->rs_length); - /* Prepare the reply chunk given the length actually * written */ rs_offset = get_unaligned(&(ch->rs_offset)); @@ -366,6 +500,7 @@ static int send_reply(struct svcxprt_rdma *rdma, int byte_count) { struct ib_send_wr send_wr; + struct ib_send_wr inv_wr; int sge_no; int sge_bytes; int page_no; @@ -385,27 +520,45 @@ static int send_reply(struct svcxprt_rdma *rdma, /* Prepare the context */ ctxt->pages[0] = page; ctxt->count = 1; + ctxt->frmr = vec->frmr; + if (vec->frmr) + set_bit(RDMACTXT_F_FAST_UNREG, &ctxt->flags); + else + clear_bit(RDMACTXT_F_FAST_UNREG, &ctxt->flags); /* Prepare the SGE for the RPCRDMA Header */ - atomic_inc(&rdma->sc_dma_used); ctxt->sge[0].addr = ib_dma_map_page(rdma->sc_cm_id->device, page, 0, PAGE_SIZE, DMA_TO_DEVICE); + if (ib_dma_mapping_error(rdma->sc_cm_id->device, ctxt->sge[0].addr)) + goto err; + atomic_inc(&rdma->sc_dma_used); + ctxt->direction = DMA_TO_DEVICE; + ctxt->sge[0].length = svc_rdma_xdr_get_reply_hdr_len(rdma_resp); - ctxt->sge[0].lkey = rdma->sc_phys_mr->lkey; + ctxt->sge[0].lkey = rdma->sc_dma_lkey; /* Determine how many of our SGE are to be transmitted */ for (sge_no = 1; byte_count && sge_no < vec->count; sge_no++) { sge_bytes = min_t(size_t, vec->sge[sge_no].iov_len, byte_count); byte_count -= sge_bytes; - atomic_inc(&rdma->sc_dma_used); - ctxt->sge[sge_no].addr = - ib_dma_map_single(rdma->sc_cm_id->device, - vec->sge[sge_no].iov_base, - sge_bytes, DMA_TO_DEVICE); + if (!vec->frmr) { + ctxt->sge[sge_no].addr = + ib_dma_map_single(rdma->sc_cm_id->device, + vec->sge[sge_no].iov_base, + sge_bytes, DMA_TO_DEVICE); + if (ib_dma_mapping_error(rdma->sc_cm_id->device, + ctxt->sge[sge_no].addr)) + goto err; + atomic_inc(&rdma->sc_dma_used); + ctxt->sge[sge_no].lkey = rdma->sc_dma_lkey; + } else { + ctxt->sge[sge_no].addr = (unsigned long) + vec->sge[sge_no].iov_base; + ctxt->sge[sge_no].lkey = vec->frmr->mr->lkey; + } ctxt->sge[sge_no].length = sge_bytes; - ctxt->sge[sge_no].lkey = rdma->sc_phys_mr->lkey; } BUG_ON(byte_count != 0); @@ -417,11 +570,16 @@ static int send_reply(struct svcxprt_rdma *rdma, ctxt->pages[page_no+1] = rqstp->rq_respages[page_no]; ctxt->count++; rqstp->rq_respages[page_no] = NULL; - /* If there are more pages than SGE, terminate SGE list */ + /* + * If there are more pages than SGE, terminate SGE + * list so that svc_rdma_unmap_dma doesn't attempt to + * unmap garbage. + */ if (page_no+1 >= sge_no) ctxt->sge[page_no+1].length = 0; } BUG_ON(sge_no > rdma->sc_max_sge); + BUG_ON(sge_no > ctxt->count); memset(&send_wr, 0, sizeof send_wr); ctxt->wr_op = IB_WR_SEND; send_wr.wr_id = (unsigned long)ctxt; @@ -429,12 +587,26 @@ static int send_reply(struct svcxprt_rdma *rdma, send_wr.num_sge = sge_no; send_wr.opcode = IB_WR_SEND; send_wr.send_flags = IB_SEND_SIGNALED; + if (vec->frmr) { + /* Prepare INVALIDATE WR */ + memset(&inv_wr, 0, sizeof inv_wr); + inv_wr.opcode = IB_WR_LOCAL_INV; + inv_wr.send_flags = IB_SEND_SIGNALED; + inv_wr.ex.invalidate_rkey = + vec->frmr->mr->lkey; + send_wr.next = &inv_wr; + } ret = svc_rdma_send(rdma, &send_wr); if (ret) - svc_rdma_put_context(ctxt, 1); + goto err; - return ret; + return 0; + + err: + svc_rdma_put_frmr(rdma, vec->frmr); + svc_rdma_put_context(ctxt, 1); + return -EIO; } void svc_rdma_prep_reply_hdr(struct svc_rqst *rqstp) @@ -477,8 +649,9 @@ int svc_rdma_sendto(struct svc_rqst *rqstp) ctxt = svc_rdma_get_context(rdma); ctxt->direction = DMA_TO_DEVICE; vec = svc_rdma_get_req_map(); - xdr_to_sge(rdma, &rqstp->rq_res, vec); - + ret = map_xdr(rdma, &rqstp->rq_res, vec); + if (ret) + goto err0; inline_bytes = rqstp->rq_res.len; /* Create the RDMA response header */ @@ -498,7 +671,7 @@ int svc_rdma_sendto(struct svc_rqst *rqstp) if (ret < 0) { printk(KERN_ERR "svcrdma: failed to send write chunks, rc=%d\n", ret); - goto error; + goto err1; } inline_bytes -= ret; @@ -508,7 +681,7 @@ int svc_rdma_sendto(struct svc_rqst *rqstp) if (ret < 0) { printk(KERN_ERR "svcrdma: failed to send reply chunks, rc=%d\n", ret); - goto error; + goto err1; } inline_bytes -= ret; @@ -517,9 +690,11 @@ int svc_rdma_sendto(struct svc_rqst *rqstp) svc_rdma_put_req_map(vec); dprintk("svcrdma: send_reply returns %d\n", ret); return ret; - error: + + err1: + put_page(res_page); + err0: svc_rdma_put_req_map(vec); svc_rdma_put_context(ctxt, 0); - put_page(res_page); return ret; } diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c index 900cb69728c6..6fb493cbd29f 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_transport.c +++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c @@ -100,20 +100,29 @@ struct svc_rdma_op_ctxt *svc_rdma_get_context(struct svcxprt_rdma *xprt) ctxt->xprt = xprt; INIT_LIST_HEAD(&ctxt->dto_q); ctxt->count = 0; + ctxt->frmr = NULL; atomic_inc(&xprt->sc_ctxt_used); return ctxt; } -static void svc_rdma_unmap_dma(struct svc_rdma_op_ctxt *ctxt) +void svc_rdma_unmap_dma(struct svc_rdma_op_ctxt *ctxt) { struct svcxprt_rdma *xprt = ctxt->xprt; int i; for (i = 0; i < ctxt->count && ctxt->sge[i].length; i++) { - atomic_dec(&xprt->sc_dma_used); - ib_dma_unmap_single(xprt->sc_cm_id->device, - ctxt->sge[i].addr, - ctxt->sge[i].length, - ctxt->direction); + /* + * Unmap the DMA addr in the SGE if the lkey matches + * the sc_dma_lkey, otherwise, ignore it since it is + * an FRMR lkey and will be unmapped later when the + * last WR that uses it completes. + */ + if (ctxt->sge[i].lkey == xprt->sc_dma_lkey) { + atomic_dec(&xprt->sc_dma_used); + ib_dma_unmap_single(xprt->sc_cm_id->device, + ctxt->sge[i].addr, + ctxt->sge[i].length, + ctxt->direction); + } } } @@ -150,6 +159,7 @@ struct svc_rdma_req_map *svc_rdma_get_req_map(void) schedule_timeout_uninterruptible(msecs_to_jiffies(500)); } map->count = 0; + map->frmr = NULL; return map; } @@ -316,6 +326,50 @@ static void rq_cq_reap(struct svcxprt_rdma *xprt) } /* + * Processs a completion context + */ +static void process_context(struct svcxprt_rdma *xprt, + struct svc_rdma_op_ctxt *ctxt) +{ + svc_rdma_unmap_dma(ctxt); + + switch (ctxt->wr_op) { + case IB_WR_SEND: + if (test_bit(RDMACTXT_F_FAST_UNREG, &ctxt->flags)) + svc_rdma_put_frmr(xprt, ctxt->frmr); + svc_rdma_put_context(ctxt, 1); + break; + + case IB_WR_RDMA_WRITE: + svc_rdma_put_context(ctxt, 0); + break; + + case IB_WR_RDMA_READ: + case IB_WR_RDMA_READ_WITH_INV: + if (test_bit(RDMACTXT_F_LAST_CTXT, &ctxt->flags)) { + struct svc_rdma_op_ctxt *read_hdr = ctxt->read_hdr; + BUG_ON(!read_hdr); + if (test_bit(RDMACTXT_F_FAST_UNREG, &ctxt->flags)) + svc_rdma_put_frmr(xprt, ctxt->frmr); + spin_lock_bh(&xprt->sc_rq_dto_lock); + set_bit(XPT_DATA, &xprt->sc_xprt.xpt_flags); + list_add_tail(&read_hdr->dto_q, + &xprt->sc_read_complete_q); + spin_unlock_bh(&xprt->sc_rq_dto_lock); + svc_xprt_enqueue(&xprt->sc_xprt); + } + svc_rdma_put_context(ctxt, 0); + break; + + default: + printk(KERN_ERR "svcrdma: unexpected completion type, " + "opcode=%d\n", + ctxt->wr_op); + break; + } +} + +/* * Send Queue Completion Handler - potentially called on interrupt context. * * Note that caller must hold a transport reference. @@ -327,17 +381,12 @@ static void sq_cq_reap(struct svcxprt_rdma *xprt) struct ib_cq *cq = xprt->sc_sq_cq; int ret; - if (!test_and_clear_bit(RDMAXPRT_SQ_PENDING, &xprt->sc_flags)) return; ib_req_notify_cq(xprt->sc_sq_cq, IB_CQ_NEXT_COMP); atomic_inc(&rdma_stat_sq_poll); while ((ret = ib_poll_cq(cq, 1, &wc)) > 0) { - ctxt = (struct svc_rdma_op_ctxt *)(unsigned long)wc.wr_id; - xprt = ctxt->xprt; - - svc_rdma_unmap_dma(ctxt); if (wc.status != IB_WC_SUCCESS) /* Close the transport */ set_bit(XPT_CLOSE, &xprt->sc_xprt.xpt_flags); @@ -346,35 +395,10 @@ static void sq_cq_reap(struct svcxprt_rdma *xprt) atomic_dec(&xprt->sc_sq_count); wake_up(&xprt->sc_send_wait); - switch (ctxt->wr_op) { - case IB_WR_SEND: - svc_rdma_put_context(ctxt, 1); - break; - - case IB_WR_RDMA_WRITE: - svc_rdma_put_context(ctxt, 0); - break; - - case IB_WR_RDMA_READ: - if (test_bit(RDMACTXT_F_LAST_CTXT, &ctxt->flags)) { - struct svc_rdma_op_ctxt *read_hdr = ctxt->read_hdr; - BUG_ON(!read_hdr); - spin_lock_bh(&xprt->sc_rq_dto_lock); - set_bit(XPT_DATA, &xprt->sc_xprt.xpt_flags); - list_add_tail(&read_hdr->dto_q, - &xprt->sc_read_complete_q); - spin_unlock_bh(&xprt->sc_rq_dto_lock); - svc_xprt_enqueue(&xprt->sc_xprt); - } - svc_rdma_put_context(ctxt, 0); - break; + ctxt = (struct svc_rdma_op_ctxt *)(unsigned long)wc.wr_id; + if (ctxt) + process_context(xprt, ctxt); - default: - printk(KERN_ERR "svcrdma: unexpected completion type, " - "opcode=%d, status=%d\n", - wc.opcode, wc.status); - break; - } svc_xprt_put(&xprt->sc_xprt); } @@ -425,10 +449,12 @@ static struct svcxprt_rdma *rdma_create_xprt(struct svc_serv *serv, INIT_LIST_HEAD(&cma_xprt->sc_dto_q); INIT_LIST_HEAD(&cma_xprt->sc_rq_dto_q); INIT_LIST_HEAD(&cma_xprt->sc_read_complete_q); + INIT_LIST_HEAD(&cma_xprt->sc_frmr_q); init_waitqueue_head(&cma_xprt->sc_send_wait); spin_lock_init(&cma_xprt->sc_lock); spin_lock_init(&cma_xprt->sc_rq_dto_lock); + spin_lock_init(&cma_xprt->sc_frmr_q_lock); cma_xprt->sc_ord = svcrdma_ord; @@ -462,7 +488,7 @@ int svc_rdma_post_recv(struct svcxprt_rdma *xprt) struct ib_recv_wr recv_wr, *bad_recv_wr; struct svc_rdma_op_ctxt *ctxt; struct page *page; - unsigned long pa; + dma_addr_t pa; int sge_no; int buflen; int ret; @@ -474,13 +500,15 @@ int svc_rdma_post_recv(struct svcxprt_rdma *xprt) BUG_ON(sge_no >= xprt->sc_max_sge); page = svc_rdma_get_page(); ctxt->pages[sge_no] = page; - atomic_inc(&xprt->sc_dma_used); pa = ib_dma_map_page(xprt->sc_cm_id->device, page, 0, PAGE_SIZE, DMA_FROM_DEVICE); + if (ib_dma_mapping_error(xprt->sc_cm_id->device, pa)) + goto err_put_ctxt; + atomic_inc(&xprt->sc_dma_used); ctxt->sge[sge_no].addr = pa; ctxt->sge[sge_no].length = PAGE_SIZE; - ctxt->sge[sge_no].lkey = xprt->sc_phys_mr->lkey; + ctxt->sge[sge_no].lkey = xprt->sc_dma_lkey; buflen += PAGE_SIZE; } ctxt->count = sge_no; @@ -496,6 +524,10 @@ int svc_rdma_post_recv(struct svcxprt_rdma *xprt) svc_rdma_put_context(ctxt, 1); } return ret; + + err_put_ctxt: + svc_rdma_put_context(ctxt, 1); + return -ENOMEM; } /* @@ -566,7 +598,7 @@ static int rdma_listen_handler(struct rdma_cm_id *cma_id, dprintk("svcrdma: Connect request on cma_id=%p, xprt = %p, " "event=%d\n", cma_id, cma_id->context, event->event); handle_connect_req(cma_id, - event->param.conn.responder_resources); + event->param.conn.initiator_depth); break; case RDMA_CM_EVENT_ESTABLISHED: @@ -686,6 +718,97 @@ static struct svc_xprt *svc_rdma_create(struct svc_serv *serv, return ERR_PTR(ret); } +static struct svc_rdma_fastreg_mr *rdma_alloc_frmr(struct svcxprt_rdma *xprt) +{ + struct ib_mr *mr; + struct ib_fast_reg_page_list *pl; + struct svc_rdma_fastreg_mr *frmr; + + frmr = kmalloc(sizeof(*frmr), GFP_KERNEL); + if (!frmr) + goto err; + + mr = ib_alloc_fast_reg_mr(xprt->sc_pd, RPCSVC_MAXPAGES); + if (!mr) + goto err_free_frmr; + + pl = ib_alloc_fast_reg_page_list(xprt->sc_cm_id->device, + RPCSVC_MAXPAGES); + if (!pl) + goto err_free_mr; + + frmr->mr = mr; + frmr->page_list = pl; + INIT_LIST_HEAD(&frmr->frmr_list); + return frmr; + + err_free_mr: + ib_dereg_mr(mr); + err_free_frmr: + kfree(frmr); + err: + return ERR_PTR(-ENOMEM); +} + +static void rdma_dealloc_frmr_q(struct svcxprt_rdma *xprt) +{ + struct svc_rdma_fastreg_mr *frmr; + + while (!list_empty(&xprt->sc_frmr_q)) { + frmr = list_entry(xprt->sc_frmr_q.next, + struct svc_rdma_fastreg_mr, frmr_list); + list_del_init(&frmr->frmr_list); + ib_dereg_mr(frmr->mr); + ib_free_fast_reg_page_list(frmr->page_list); + kfree(frmr); + } +} + +struct svc_rdma_fastreg_mr *svc_rdma_get_frmr(struct svcxprt_rdma *rdma) +{ + struct svc_rdma_fastreg_mr *frmr = NULL; + + spin_lock_bh(&rdma->sc_frmr_q_lock); + if (!list_empty(&rdma->sc_frmr_q)) { + frmr = list_entry(rdma->sc_frmr_q.next, + struct svc_rdma_fastreg_mr, frmr_list); + list_del_init(&frmr->frmr_list); + frmr->map_len = 0; + frmr->page_list_len = 0; + } + spin_unlock_bh(&rdma->sc_frmr_q_lock); + if (frmr) + return frmr; + + return rdma_alloc_frmr(rdma); +} + +static void frmr_unmap_dma(struct svcxprt_rdma *xprt, + struct svc_rdma_fastreg_mr *frmr) +{ + int page_no; + for (page_no = 0; page_no < frmr->page_list_len; page_no++) { + dma_addr_t addr = frmr->page_list->page_list[page_no]; + if (ib_dma_mapping_error(frmr->mr->device, addr)) + continue; + atomic_dec(&xprt->sc_dma_used); + ib_dma_unmap_single(frmr->mr->device, addr, PAGE_SIZE, + frmr->direction); + } +} + +void svc_rdma_put_frmr(struct svcxprt_rdma *rdma, + struct svc_rdma_fastreg_mr *frmr) +{ + if (frmr) { + frmr_unmap_dma(rdma, frmr); + spin_lock_bh(&rdma->sc_frmr_q_lock); + BUG_ON(!list_empty(&frmr->frmr_list)); + list_add(&frmr->frmr_list, &rdma->sc_frmr_q); + spin_unlock_bh(&rdma->sc_frmr_q_lock); + } +} + /* * This is the xpo_recvfrom function for listening endpoints. Its * purpose is to accept incoming connections. The CMA callback handler @@ -704,6 +827,8 @@ static struct svc_xprt *svc_rdma_accept(struct svc_xprt *xprt) struct rdma_conn_param conn_param; struct ib_qp_init_attr qp_attr; struct ib_device_attr devattr; + int dma_mr_acc; + int need_dma_mr; int ret; int i; @@ -819,15 +944,77 @@ static struct svc_xprt *svc_rdma_accept(struct svc_xprt *xprt) } newxprt->sc_qp = newxprt->sc_cm_id->qp; - /* Register all of physical memory */ - newxprt->sc_phys_mr = ib_get_dma_mr(newxprt->sc_pd, - IB_ACCESS_LOCAL_WRITE | - IB_ACCESS_REMOTE_WRITE); - if (IS_ERR(newxprt->sc_phys_mr)) { - dprintk("svcrdma: Failed to create DMA MR ret=%d\n", ret); + /* + * Use the most secure set of MR resources based on the + * transport type and available memory management features in + * the device. Here's the table implemented below: + * + * Fast Global DMA Remote WR + * Reg LKEY MR Access + * Sup'd Sup'd Needed Needed + * + * IWARP N N Y Y + * N Y Y Y + * Y N Y N + * Y Y N - + * + * IB N N Y N + * N Y N - + * Y N Y N + * Y Y N - + * + * NB: iWARP requires remote write access for the data sink + * of an RDMA_READ. IB does not. + */ + if (devattr.device_cap_flags & IB_DEVICE_MEM_MGT_EXTENSIONS) { + newxprt->sc_frmr_pg_list_len = + devattr.max_fast_reg_page_list_len; + newxprt->sc_dev_caps |= SVCRDMA_DEVCAP_FAST_REG; + } + + /* + * Determine if a DMA MR is required and if so, what privs are required + */ + switch (rdma_node_get_transport(newxprt->sc_cm_id->device->node_type)) { + case RDMA_TRANSPORT_IWARP: + newxprt->sc_dev_caps |= SVCRDMA_DEVCAP_READ_W_INV; + if (!(newxprt->sc_dev_caps & SVCRDMA_DEVCAP_FAST_REG)) { + need_dma_mr = 1; + dma_mr_acc = + (IB_ACCESS_LOCAL_WRITE | + IB_ACCESS_REMOTE_WRITE); + } else if (!(devattr.device_cap_flags & IB_DEVICE_LOCAL_DMA_LKEY)) { + need_dma_mr = 1; + dma_mr_acc = IB_ACCESS_LOCAL_WRITE; + } else + need_dma_mr = 0; + break; + case RDMA_TRANSPORT_IB: + if (!(devattr.device_cap_flags & IB_DEVICE_LOCAL_DMA_LKEY)) { + need_dma_mr = 1; + dma_mr_acc = IB_ACCESS_LOCAL_WRITE; + } else + need_dma_mr = 0; + break; + default: goto errout; } + /* Create the DMA MR if needed, otherwise, use the DMA LKEY */ + if (need_dma_mr) { + /* Register all of physical memory */ + newxprt->sc_phys_mr = + ib_get_dma_mr(newxprt->sc_pd, dma_mr_acc); + if (IS_ERR(newxprt->sc_phys_mr)) { + dprintk("svcrdma: Failed to create DMA MR ret=%d\n", + ret); + goto errout; + } + newxprt->sc_dma_lkey = newxprt->sc_phys_mr->lkey; + } else + newxprt->sc_dma_lkey = + newxprt->sc_cm_id->device->local_dma_lkey; + /* Post receive buffers */ for (i = 0; i < newxprt->sc_max_requests; i++) { ret = svc_rdma_post_recv(newxprt); @@ -961,6 +1148,9 @@ static void __svc_rdma_free(struct work_struct *work) WARN_ON(atomic_read(&rdma->sc_ctxt_used) != 0); WARN_ON(atomic_read(&rdma->sc_dma_used) != 0); + /* De-allocate fastreg mr */ + rdma_dealloc_frmr_q(rdma); + /* Destroy the QP if present (not a listener) */ if (rdma->sc_qp && !IS_ERR(rdma->sc_qp)) ib_destroy_qp(rdma->sc_qp); @@ -1014,21 +1204,59 @@ static int svc_rdma_has_wspace(struct svc_xprt *xprt) return 1; } +/* + * Attempt to register the kvec representing the RPC memory with the + * device. + * + * Returns: + * NULL : The device does not support fastreg or there were no more + * fastreg mr. + * frmr : The kvec register request was successfully posted. + * <0 : An error was encountered attempting to register the kvec. + */ +int svc_rdma_fastreg(struct svcxprt_rdma *xprt, + struct svc_rdma_fastreg_mr *frmr) +{ + struct ib_send_wr fastreg_wr; + u8 key; + + /* Bump the key */ + key = (u8)(frmr->mr->lkey & 0x000000FF); + ib_update_fast_reg_key(frmr->mr, ++key); + + /* Prepare FASTREG WR */ + memset(&fastreg_wr, 0, sizeof fastreg_wr); + fastreg_wr.opcode = IB_WR_FAST_REG_MR; + fastreg_wr.send_flags = IB_SEND_SIGNALED; + fastreg_wr.wr.fast_reg.iova_start = (unsigned long)frmr->kva; + fastreg_wr.wr.fast_reg.page_list = frmr->page_list; + fastreg_wr.wr.fast_reg.page_list_len = frmr->page_list_len; + fastreg_wr.wr.fast_reg.page_shift = PAGE_SHIFT; + fastreg_wr.wr.fast_reg.length = frmr->map_len; + fastreg_wr.wr.fast_reg.access_flags = frmr->access_flags; + fastreg_wr.wr.fast_reg.rkey = frmr->mr->lkey; + return svc_rdma_send(xprt, &fastreg_wr); +} + int svc_rdma_send(struct svcxprt_rdma *xprt, struct ib_send_wr *wr) { - struct ib_send_wr *bad_wr; + struct ib_send_wr *bad_wr, *n_wr; + int wr_count; + int i; int ret; if (test_bit(XPT_CLOSE, &xprt->sc_xprt.xpt_flags)) return -ENOTCONN; BUG_ON(wr->send_flags != IB_SEND_SIGNALED); - BUG_ON(((struct svc_rdma_op_ctxt *)(unsigned long)wr->wr_id)->wr_op != - wr->opcode); + wr_count = 1; + for (n_wr = wr->next; n_wr; n_wr = n_wr->next) + wr_count++; + /* If the SQ is full, wait until an SQ entry is available */ while (1) { spin_lock_bh(&xprt->sc_lock); - if (xprt->sc_sq_depth == atomic_read(&xprt->sc_sq_count)) { + if (xprt->sc_sq_depth < atomic_read(&xprt->sc_sq_count) + wr_count) { spin_unlock_bh(&xprt->sc_lock); atomic_inc(&rdma_stat_sq_starve); @@ -1043,19 +1271,26 @@ int svc_rdma_send(struct svcxprt_rdma *xprt, struct ib_send_wr *wr) return 0; continue; } - /* Bumped used SQ WR count and post */ - svc_xprt_get(&xprt->sc_xprt); + /* Take a transport ref for each WR posted */ + for (i = 0; i < wr_count; i++) + svc_xprt_get(&xprt->sc_xprt); + + /* Bump used SQ WR count and post */ + atomic_add(wr_count, &xprt->sc_sq_count); ret = ib_post_send(xprt->sc_qp, wr, &bad_wr); - if (!ret) - atomic_inc(&xprt->sc_sq_count); - else { - svc_xprt_put(&xprt->sc_xprt); + if (ret) { + set_bit(XPT_CLOSE, &xprt->sc_xprt.xpt_flags); + atomic_sub(wr_count, &xprt->sc_sq_count); + for (i = 0; i < wr_count; i ++) + svc_xprt_put(&xprt->sc_xprt); dprintk("svcrdma: failed to post SQ WR rc=%d, " "sc_sq_count=%d, sc_sq_depth=%d\n", ret, atomic_read(&xprt->sc_sq_count), xprt->sc_sq_depth); } spin_unlock_bh(&xprt->sc_lock); + if (ret) + wake_up(&xprt->sc_send_wait); break; } return ret; @@ -1079,10 +1314,14 @@ void svc_rdma_send_error(struct svcxprt_rdma *xprt, struct rpcrdma_msg *rmsgp, length = svc_rdma_xdr_encode_error(xprt, rmsgp, err, va); /* Prepare SGE for local address */ - atomic_inc(&xprt->sc_dma_used); sge.addr = ib_dma_map_page(xprt->sc_cm_id->device, p, 0, PAGE_SIZE, DMA_FROM_DEVICE); - sge.lkey = xprt->sc_phys_mr->lkey; + if (ib_dma_mapping_error(xprt->sc_cm_id->device, sge.addr)) { + put_page(p); + return; + } + atomic_inc(&xprt->sc_dma_used); + sge.lkey = xprt->sc_dma_lkey; sge.length = length; ctxt = svc_rdma_get_context(xprt); @@ -1103,6 +1342,9 @@ void svc_rdma_send_error(struct svcxprt_rdma *xprt, struct rpcrdma_msg *rmsgp, if (ret) { dprintk("svcrdma: Error %d posting send for protocol error\n", ret); + ib_dma_unmap_page(xprt->sc_cm_id->device, + sge.addr, PAGE_SIZE, + DMA_FROM_DEVICE); svc_rdma_put_context(ctxt, 1); } } diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c index a564c1a39ec5..9839c3d94145 100644 --- a/net/sunrpc/xprtrdma/transport.c +++ b/net/sunrpc/xprtrdma/transport.c @@ -70,11 +70,8 @@ static unsigned int xprt_rdma_slot_table_entries = RPCRDMA_DEF_SLOT_TABLE; static unsigned int xprt_rdma_max_inline_read = RPCRDMA_DEF_INLINE; static unsigned int xprt_rdma_max_inline_write = RPCRDMA_DEF_INLINE; static unsigned int xprt_rdma_inline_write_padding; -#if !RPCRDMA_PERSISTENT_REGISTRATION -static unsigned int xprt_rdma_memreg_strategy = RPCRDMA_REGISTER; /* FMR? */ -#else -static unsigned int xprt_rdma_memreg_strategy = RPCRDMA_ALLPHYSICAL; -#endif +static unsigned int xprt_rdma_memreg_strategy = RPCRDMA_FRMR; + int xprt_rdma_pad_optimize = 0; #ifdef RPC_DEBUG @@ -140,6 +137,14 @@ static ctl_table xr_tunables_table[] = { .extra2 = &max_memreg, }, { + .ctl_name = CTL_UNNUMBERED, + .procname = "rdma_pad_optimize", + .data = &xprt_rdma_pad_optimize, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, + { .ctl_name = 0, }, }; @@ -458,6 +463,8 @@ xprt_rdma_close(struct rpc_xprt *xprt) struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt); dprintk("RPC: %s: closing\n", __func__); + if (r_xprt->rx_ep.rep_connected > 0) + xprt->reestablish_timeout = 0; xprt_disconnect_done(xprt); (void) rpcrdma_ep_disconnect(&r_xprt->rx_ep, &r_xprt->rx_ia); } @@ -485,6 +492,11 @@ xprt_rdma_connect(struct rpc_task *task) /* Reconnect */ schedule_delayed_work(&r_xprt->rdma_connect, xprt->reestablish_timeout); + xprt->reestablish_timeout <<= 1; + if (xprt->reestablish_timeout > (30 * HZ)) + xprt->reestablish_timeout = (30 * HZ); + else if (xprt->reestablish_timeout < (5 * HZ)) + xprt->reestablish_timeout = (5 * HZ); } else { schedule_delayed_work(&r_xprt->rdma_connect, 0); if (!RPC_IS_ASYNC(task)) @@ -591,6 +603,7 @@ xprt_rdma_allocate(struct rpc_task *task, size_t size) } dprintk("RPC: %s: size %zd, request 0x%p\n", __func__, size, req); out: + req->rl_connect_cookie = 0; /* our reserved value */ return req->rl_xdr_buf; outfail: @@ -694,13 +707,21 @@ xprt_rdma_send_request(struct rpc_task *task) req->rl_reply->rr_xprt = xprt; } - if (rpcrdma_ep_post(&r_xprt->rx_ia, &r_xprt->rx_ep, req)) { - xprt_disconnect_done(xprt); - return -ENOTCONN; /* implies disconnect */ - } + /* Must suppress retransmit to maintain credits */ + if (req->rl_connect_cookie == xprt->connect_cookie) + goto drop_connection; + req->rl_connect_cookie = xprt->connect_cookie; + + if (rpcrdma_ep_post(&r_xprt->rx_ia, &r_xprt->rx_ep, req)) + goto drop_connection; + task->tk_bytes_sent += rqst->rq_snd_buf.len; rqst->rq_bytes_sent = 0; return 0; + +drop_connection: + xprt_disconnect_done(xprt); + return -ENOTCONN; /* implies disconnect */ } static void xprt_rdma_print_stats(struct rpc_xprt *xprt, struct seq_file *seq) @@ -770,7 +791,7 @@ static void __exit xprt_rdma_cleanup(void) { int rc; - dprintk("RPCRDMA Module Removed, deregister RPC RDMA transport\n"); + dprintk(KERN_INFO "RPCRDMA Module Removed, deregister RPC RDMA transport\n"); #ifdef RPC_DEBUG if (sunrpc_table_header) { unregister_sysctl_table(sunrpc_table_header); diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c index 8ea283ecc522..a5fef5e6c323 100644 --- a/net/sunrpc/xprtrdma/verbs.c +++ b/net/sunrpc/xprtrdma/verbs.c @@ -284,6 +284,7 @@ rpcrdma_conn_upcall(struct rdma_cm_id *id, struct rdma_cm_event *event) switch (event->event) { case RDMA_CM_EVENT_ADDR_RESOLVED: case RDMA_CM_EVENT_ROUTE_RESOLVED: + ia->ri_async_rc = 0; complete(&ia->ri_done); break; case RDMA_CM_EVENT_ADDR_ERROR: @@ -338,13 +339,32 @@ connected: wake_up_all(&ep->rep_connect_wait); break; default: - ia->ri_async_rc = -EINVAL; - dprintk("RPC: %s: unexpected CM event %X\n", + dprintk("RPC: %s: unexpected CM event %d\n", __func__, event->event); - complete(&ia->ri_done); break; } +#ifdef RPC_DEBUG + if (connstate == 1) { + int ird = attr.max_dest_rd_atomic; + int tird = ep->rep_remote_cma.responder_resources; + printk(KERN_INFO "rpcrdma: connection to %u.%u.%u.%u:%u " + "on %s, memreg %d slots %d ird %d%s\n", + NIPQUAD(addr->sin_addr.s_addr), + ntohs(addr->sin_port), + ia->ri_id->device->name, + ia->ri_memreg_strategy, + xprt->rx_buf.rb_max_requests, + ird, ird < 4 && ird < tird / 2 ? " (low!)" : ""); + } else if (connstate < 0) { + printk(KERN_INFO "rpcrdma: connection to %u.%u.%u.%u:%u " + "closed (%d)\n", + NIPQUAD(addr->sin_addr.s_addr), + ntohs(addr->sin_port), + connstate); + } +#endif + return 0; } @@ -355,6 +375,8 @@ rpcrdma_create_id(struct rpcrdma_xprt *xprt, struct rdma_cm_id *id; int rc; + init_completion(&ia->ri_done); + id = rdma_create_id(rpcrdma_conn_upcall, xprt, RDMA_PS_TCP); if (IS_ERR(id)) { rc = PTR_ERR(id); @@ -363,26 +385,28 @@ rpcrdma_create_id(struct rpcrdma_xprt *xprt, return id; } - ia->ri_async_rc = 0; + ia->ri_async_rc = -ETIMEDOUT; rc = rdma_resolve_addr(id, NULL, addr, RDMA_RESOLVE_TIMEOUT); if (rc) { dprintk("RPC: %s: rdma_resolve_addr() failed %i\n", __func__, rc); goto out; } - wait_for_completion(&ia->ri_done); + wait_for_completion_interruptible_timeout(&ia->ri_done, + msecs_to_jiffies(RDMA_RESOLVE_TIMEOUT) + 1); rc = ia->ri_async_rc; if (rc) goto out; - ia->ri_async_rc = 0; + ia->ri_async_rc = -ETIMEDOUT; rc = rdma_resolve_route(id, RDMA_RESOLVE_TIMEOUT); if (rc) { dprintk("RPC: %s: rdma_resolve_route() failed %i\n", __func__, rc); goto out; } - wait_for_completion(&ia->ri_done); + wait_for_completion_interruptible_timeout(&ia->ri_done, + msecs_to_jiffies(RDMA_RESOLVE_TIMEOUT) + 1); rc = ia->ri_async_rc; if (rc) goto out; @@ -423,11 +447,10 @@ rpcrdma_clean_cq(struct ib_cq *cq) int rpcrdma_ia_open(struct rpcrdma_xprt *xprt, struct sockaddr *addr, int memreg) { - int rc; + int rc, mem_priv; + struct ib_device_attr devattr; struct rpcrdma_ia *ia = &xprt->rx_ia; - init_completion(&ia->ri_done); - ia->ri_id = rpcrdma_create_id(xprt, ia, addr); if (IS_ERR(ia->ri_id)) { rc = PTR_ERR(ia->ri_id); @@ -443,6 +466,73 @@ rpcrdma_ia_open(struct rpcrdma_xprt *xprt, struct sockaddr *addr, int memreg) } /* + * Query the device to determine if the requested memory + * registration strategy is supported. If it isn't, set the + * strategy to a globally supported model. + */ + rc = ib_query_device(ia->ri_id->device, &devattr); + if (rc) { + dprintk("RPC: %s: ib_query_device failed %d\n", + __func__, rc); + goto out2; + } + + if (devattr.device_cap_flags & IB_DEVICE_LOCAL_DMA_LKEY) { + ia->ri_have_dma_lkey = 1; + ia->ri_dma_lkey = ia->ri_id->device->local_dma_lkey; + } + + switch (memreg) { + case RPCRDMA_MEMWINDOWS: + case RPCRDMA_MEMWINDOWS_ASYNC: + if (!(devattr.device_cap_flags & IB_DEVICE_MEM_WINDOW)) { + dprintk("RPC: %s: MEMWINDOWS registration " + "specified but not supported by adapter, " + "using slower RPCRDMA_REGISTER\n", + __func__); + memreg = RPCRDMA_REGISTER; + } + break; + case RPCRDMA_MTHCAFMR: + if (!ia->ri_id->device->alloc_fmr) { +#if RPCRDMA_PERSISTENT_REGISTRATION + dprintk("RPC: %s: MTHCAFMR registration " + "specified but not supported by adapter, " + "using riskier RPCRDMA_ALLPHYSICAL\n", + __func__); + memreg = RPCRDMA_ALLPHYSICAL; +#else + dprintk("RPC: %s: MTHCAFMR registration " + "specified but not supported by adapter, " + "using slower RPCRDMA_REGISTER\n", + __func__); + memreg = RPCRDMA_REGISTER; +#endif + } + break; + case RPCRDMA_FRMR: + /* Requires both frmr reg and local dma lkey */ + if ((devattr.device_cap_flags & + (IB_DEVICE_MEM_MGT_EXTENSIONS|IB_DEVICE_LOCAL_DMA_LKEY)) != + (IB_DEVICE_MEM_MGT_EXTENSIONS|IB_DEVICE_LOCAL_DMA_LKEY)) { +#if RPCRDMA_PERSISTENT_REGISTRATION + dprintk("RPC: %s: FRMR registration " + "specified but not supported by adapter, " + "using riskier RPCRDMA_ALLPHYSICAL\n", + __func__); + memreg = RPCRDMA_ALLPHYSICAL; +#else + dprintk("RPC: %s: FRMR registration " + "specified but not supported by adapter, " + "using slower RPCRDMA_REGISTER\n", + __func__); + memreg = RPCRDMA_REGISTER; +#endif + } + break; + } + + /* * Optionally obtain an underlying physical identity mapping in * order to do a memory window-based bind. This base registration * is protected from remote access - that is enabled only by binding @@ -450,22 +540,28 @@ rpcrdma_ia_open(struct rpcrdma_xprt *xprt, struct sockaddr *addr, int memreg) * revoked after the corresponding completion similar to a storage * adapter. */ - if (memreg > RPCRDMA_REGISTER) { - int mem_priv = IB_ACCESS_LOCAL_WRITE; - switch (memreg) { + switch (memreg) { + case RPCRDMA_BOUNCEBUFFERS: + case RPCRDMA_REGISTER: + case RPCRDMA_FRMR: + break; #if RPCRDMA_PERSISTENT_REGISTRATION - case RPCRDMA_ALLPHYSICAL: - mem_priv |= IB_ACCESS_REMOTE_WRITE; - mem_priv |= IB_ACCESS_REMOTE_READ; - break; + case RPCRDMA_ALLPHYSICAL: + mem_priv = IB_ACCESS_LOCAL_WRITE | + IB_ACCESS_REMOTE_WRITE | + IB_ACCESS_REMOTE_READ; + goto register_setup; #endif - case RPCRDMA_MEMWINDOWS_ASYNC: - case RPCRDMA_MEMWINDOWS: - mem_priv |= IB_ACCESS_MW_BIND; - break; - default: + case RPCRDMA_MEMWINDOWS_ASYNC: + case RPCRDMA_MEMWINDOWS: + mem_priv = IB_ACCESS_LOCAL_WRITE | + IB_ACCESS_MW_BIND; + goto register_setup; + case RPCRDMA_MTHCAFMR: + if (ia->ri_have_dma_lkey) break; - } + mem_priv = IB_ACCESS_LOCAL_WRITE; + register_setup: ia->ri_bind_mem = ib_get_dma_mr(ia->ri_pd, mem_priv); if (IS_ERR(ia->ri_bind_mem)) { printk(KERN_ALERT "%s: ib_get_dma_mr for " @@ -475,7 +571,15 @@ rpcrdma_ia_open(struct rpcrdma_xprt *xprt, struct sockaddr *addr, int memreg) memreg = RPCRDMA_REGISTER; ia->ri_bind_mem = NULL; } + break; + default: + printk(KERN_ERR "%s: invalid memory registration mode %d\n", + __func__, memreg); + rc = -EINVAL; + goto out2; } + dprintk("RPC: %s: memory registration strategy is %d\n", + __func__, memreg); /* Else will do memory reg/dereg for each chunk */ ia->ri_memreg_strategy = memreg; @@ -483,6 +587,7 @@ rpcrdma_ia_open(struct rpcrdma_xprt *xprt, struct sockaddr *addr, int memreg) return 0; out2: rdma_destroy_id(ia->ri_id); + ia->ri_id = NULL; out1: return rc; } @@ -503,15 +608,17 @@ rpcrdma_ia_close(struct rpcrdma_ia *ia) dprintk("RPC: %s: ib_dereg_mr returned %i\n", __func__, rc); } - if (ia->ri_id != NULL && !IS_ERR(ia->ri_id) && ia->ri_id->qp) - rdma_destroy_qp(ia->ri_id); + if (ia->ri_id != NULL && !IS_ERR(ia->ri_id)) { + if (ia->ri_id->qp) + rdma_destroy_qp(ia->ri_id); + rdma_destroy_id(ia->ri_id); + ia->ri_id = NULL; + } if (ia->ri_pd != NULL && !IS_ERR(ia->ri_pd)) { rc = ib_dealloc_pd(ia->ri_pd); dprintk("RPC: %s: ib_dealloc_pd returned %i\n", __func__, rc); } - if (ia->ri_id != NULL && !IS_ERR(ia->ri_id)) - rdma_destroy_id(ia->ri_id); } /* @@ -541,6 +648,12 @@ rpcrdma_ep_create(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia, ep->rep_attr.srq = NULL; ep->rep_attr.cap.max_send_wr = cdata->max_requests; switch (ia->ri_memreg_strategy) { + case RPCRDMA_FRMR: + /* Add room for frmr register and invalidate WRs */ + ep->rep_attr.cap.max_send_wr *= 3; + if (ep->rep_attr.cap.max_send_wr > devattr.max_qp_wr) + return -EINVAL; + break; case RPCRDMA_MEMWINDOWS_ASYNC: case RPCRDMA_MEMWINDOWS: /* Add room for mw_binds+unbinds - overkill! */ @@ -617,29 +730,13 @@ rpcrdma_ep_create(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia, ep->rep_remote_cma.private_data_len = 0; /* Client offers RDMA Read but does not initiate */ - switch (ia->ri_memreg_strategy) { - case RPCRDMA_BOUNCEBUFFERS: + ep->rep_remote_cma.initiator_depth = 0; + if (ia->ri_memreg_strategy == RPCRDMA_BOUNCEBUFFERS) ep->rep_remote_cma.responder_resources = 0; - break; - case RPCRDMA_MTHCAFMR: - case RPCRDMA_REGISTER: - ep->rep_remote_cma.responder_resources = cdata->max_requests * - (RPCRDMA_MAX_DATA_SEGS / 8); - break; - case RPCRDMA_MEMWINDOWS: - case RPCRDMA_MEMWINDOWS_ASYNC: -#if RPCRDMA_PERSISTENT_REGISTRATION - case RPCRDMA_ALLPHYSICAL: -#endif - ep->rep_remote_cma.responder_resources = cdata->max_requests * - (RPCRDMA_MAX_DATA_SEGS / 2); - break; - default: - break; - } - if (ep->rep_remote_cma.responder_resources > devattr.max_qp_rd_atom) + else if (devattr.max_qp_rd_atom > 32) /* arbitrary but <= 255 */ + ep->rep_remote_cma.responder_resources = 32; + else ep->rep_remote_cma.responder_resources = devattr.max_qp_rd_atom; - ep->rep_remote_cma.initiator_depth = 0; ep->rep_remote_cma.retry_count = 7; ep->rep_remote_cma.flow_control = 0; @@ -679,21 +776,16 @@ rpcrdma_ep_destroy(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia) if (rc) dprintk("RPC: %s: rpcrdma_ep_disconnect" " returned %i\n", __func__, rc); + rdma_destroy_qp(ia->ri_id); + ia->ri_id->qp = NULL; } - ep->rep_func = NULL; - /* padding - could be done in rpcrdma_buffer_destroy... */ if (ep->rep_pad_mr) { rpcrdma_deregister_internal(ia, ep->rep_pad_mr, &ep->rep_pad); ep->rep_pad_mr = NULL; } - if (ia->ri_id->qp) { - rdma_destroy_qp(ia->ri_id); - ia->ri_id->qp = NULL; - } - rpcrdma_clean_cq(ep->rep_cq); rc = ib_destroy_cq(ep->rep_cq); if (rc) @@ -712,9 +804,8 @@ rpcrdma_ep_connect(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia) struct rdma_cm_id *id; int rc = 0; int retry_count = 0; - int reconnect = (ep->rep_connected != 0); - if (reconnect) { + if (ep->rep_connected != 0) { struct rpcrdma_xprt *xprt; retry: rc = rpcrdma_ep_disconnect(ep, ia); @@ -745,6 +836,7 @@ retry: goto out; } /* END TEMP */ + rdma_destroy_qp(ia->ri_id); rdma_destroy_id(ia->ri_id); ia->ri_id = id; } @@ -769,14 +861,6 @@ if (strnicmp(ia->ri_id->device->dma_device->bus->name, "pci", 3) == 0) { } } - /* Theoretically a client initiator_depth > 0 is not needed, - * but many peers fail to complete the connection unless they - * == responder_resources! */ - if (ep->rep_remote_cma.initiator_depth != - ep->rep_remote_cma.responder_resources) - ep->rep_remote_cma.initiator_depth = - ep->rep_remote_cma.responder_resources; - ep->rep_connected = 0; rc = rdma_connect(ia->ri_id, &ep->rep_remote_cma); @@ -786,9 +870,6 @@ if (strnicmp(ia->ri_id->device->dma_device->bus->name, "pci", 3) == 0) { goto out; } - if (reconnect) - return 0; - wait_event_interruptible(ep->rep_connect_wait, ep->rep_connected != 0); /* @@ -805,14 +886,16 @@ if (strnicmp(ia->ri_id->device->dma_device->bus->name, "pci", 3) == 0) { if (ep->rep_connected <= 0) { /* Sometimes, the only way to reliably connect to remote * CMs is to use same nonzero values for ORD and IRD. */ - ep->rep_remote_cma.initiator_depth = - ep->rep_remote_cma.responder_resources; - if (ep->rep_remote_cma.initiator_depth == 0) - ++ep->rep_remote_cma.initiator_depth; - if (ep->rep_remote_cma.responder_resources == 0) - ++ep->rep_remote_cma.responder_resources; - if (retry_count++ == 0) + if (retry_count++ <= RDMA_CONNECT_RETRY_MAX + 1 && + (ep->rep_remote_cma.responder_resources == 0 || + ep->rep_remote_cma.initiator_depth != + ep->rep_remote_cma.responder_resources)) { + if (ep->rep_remote_cma.responder_resources == 0) + ep->rep_remote_cma.responder_resources = 1; + ep->rep_remote_cma.initiator_depth = + ep->rep_remote_cma.responder_resources; goto retry; + } rc = ep->rep_connected; } else { dprintk("RPC: %s: connected\n", __func__); @@ -863,6 +946,7 @@ rpcrdma_buffer_create(struct rpcrdma_buffer *buf, struct rpcrdma_ep *ep, char *p; size_t len; int i, rc; + struct rpcrdma_mw *r; buf->rb_max_requests = cdata->max_requests; spin_lock_init(&buf->rb_lock); @@ -873,7 +957,7 @@ rpcrdma_buffer_create(struct rpcrdma_buffer *buf, struct rpcrdma_ep *ep, * 2. arrays of struct rpcrdma_req to fill in pointers * 3. array of struct rpcrdma_rep for replies * 4. padding, if any - * 5. mw's, if any + * 5. mw's, fmr's or frmr's, if any * Send/recv buffers in req/rep need to be registered */ @@ -881,6 +965,10 @@ rpcrdma_buffer_create(struct rpcrdma_buffer *buf, struct rpcrdma_ep *ep, (sizeof(struct rpcrdma_req *) + sizeof(struct rpcrdma_rep *)); len += cdata->padding; switch (ia->ri_memreg_strategy) { + case RPCRDMA_FRMR: + len += buf->rb_max_requests * RPCRDMA_MAX_SEGS * + sizeof(struct rpcrdma_mw); + break; case RPCRDMA_MTHCAFMR: /* TBD we are perhaps overallocating here */ len += (buf->rb_max_requests + 1) * RPCRDMA_MAX_SEGS * @@ -927,15 +1015,37 @@ rpcrdma_buffer_create(struct rpcrdma_buffer *buf, struct rpcrdma_ep *ep, * and also reduce unbind-to-bind collision. */ INIT_LIST_HEAD(&buf->rb_mws); + r = (struct rpcrdma_mw *)p; switch (ia->ri_memreg_strategy) { + case RPCRDMA_FRMR: + for (i = buf->rb_max_requests * RPCRDMA_MAX_SEGS; i; i--) { + r->r.frmr.fr_mr = ib_alloc_fast_reg_mr(ia->ri_pd, + RPCRDMA_MAX_SEGS); + if (IS_ERR(r->r.frmr.fr_mr)) { + rc = PTR_ERR(r->r.frmr.fr_mr); + dprintk("RPC: %s: ib_alloc_fast_reg_mr" + " failed %i\n", __func__, rc); + goto out; + } + r->r.frmr.fr_pgl = + ib_alloc_fast_reg_page_list(ia->ri_id->device, + RPCRDMA_MAX_SEGS); + if (IS_ERR(r->r.frmr.fr_pgl)) { + rc = PTR_ERR(r->r.frmr.fr_pgl); + dprintk("RPC: %s: " + "ib_alloc_fast_reg_page_list " + "failed %i\n", __func__, rc); + goto out; + } + list_add(&r->mw_list, &buf->rb_mws); + ++r; + } + break; case RPCRDMA_MTHCAFMR: - { - struct rpcrdma_mw *r = (struct rpcrdma_mw *)p; - struct ib_fmr_attr fa = { - RPCRDMA_MAX_DATA_SEGS, 1, PAGE_SHIFT - }; /* TBD we are perhaps overallocating here */ for (i = (buf->rb_max_requests+1) * RPCRDMA_MAX_SEGS; i; i--) { + static struct ib_fmr_attr fa = + { RPCRDMA_MAX_DATA_SEGS, 1, PAGE_SHIFT }; r->r.fmr = ib_alloc_fmr(ia->ri_pd, IB_ACCESS_REMOTE_WRITE | IB_ACCESS_REMOTE_READ, &fa); @@ -948,12 +1058,9 @@ rpcrdma_buffer_create(struct rpcrdma_buffer *buf, struct rpcrdma_ep *ep, list_add(&r->mw_list, &buf->rb_mws); ++r; } - } break; case RPCRDMA_MEMWINDOWS_ASYNC: case RPCRDMA_MEMWINDOWS: - { - struct rpcrdma_mw *r = (struct rpcrdma_mw *)p; /* Allocate one extra request's worth, for full cycling */ for (i = (buf->rb_max_requests+1) * RPCRDMA_MAX_SEGS; i; i--) { r->r.mw = ib_alloc_mw(ia->ri_pd); @@ -966,7 +1073,6 @@ rpcrdma_buffer_create(struct rpcrdma_buffer *buf, struct rpcrdma_ep *ep, list_add(&r->mw_list, &buf->rb_mws); ++r; } - } break; default: break; @@ -1046,6 +1152,7 @@ rpcrdma_buffer_destroy(struct rpcrdma_buffer *buf) { int rc, i; struct rpcrdma_ia *ia = rdmab_to_ia(buf); + struct rpcrdma_mw *r; /* clean up in reverse order from create * 1. recv mr memory (mr free, then kfree) @@ -1065,11 +1172,19 @@ rpcrdma_buffer_destroy(struct rpcrdma_buffer *buf) } if (buf->rb_send_bufs && buf->rb_send_bufs[i]) { while (!list_empty(&buf->rb_mws)) { - struct rpcrdma_mw *r; r = list_entry(buf->rb_mws.next, struct rpcrdma_mw, mw_list); list_del(&r->mw_list); switch (ia->ri_memreg_strategy) { + case RPCRDMA_FRMR: + rc = ib_dereg_mr(r->r.frmr.fr_mr); + if (rc) + dprintk("RPC: %s:" + " ib_dereg_mr" + " failed %i\n", + __func__, rc); + ib_free_fast_reg_page_list(r->r.frmr.fr_pgl); + break; case RPCRDMA_MTHCAFMR: rc = ib_dealloc_fmr(r->r.fmr); if (rc) @@ -1115,6 +1230,8 @@ rpcrdma_buffer_get(struct rpcrdma_buffer *buffers) { struct rpcrdma_req *req; unsigned long flags; + int i; + struct rpcrdma_mw *r; spin_lock_irqsave(&buffers->rb_lock, flags); if (buffers->rb_send_index == buffers->rb_max_requests) { @@ -1135,9 +1252,8 @@ rpcrdma_buffer_get(struct rpcrdma_buffer *buffers) } buffers->rb_send_bufs[buffers->rb_send_index++] = NULL; if (!list_empty(&buffers->rb_mws)) { - int i = RPCRDMA_MAX_SEGS - 1; + i = RPCRDMA_MAX_SEGS - 1; do { - struct rpcrdma_mw *r; r = list_entry(buffers->rb_mws.next, struct rpcrdma_mw, mw_list); list_del(&r->mw_list); @@ -1171,6 +1287,7 @@ rpcrdma_buffer_put(struct rpcrdma_req *req) req->rl_reply = NULL; } switch (ia->ri_memreg_strategy) { + case RPCRDMA_FRMR: case RPCRDMA_MTHCAFMR: case RPCRDMA_MEMWINDOWS_ASYNC: case RPCRDMA_MEMWINDOWS: @@ -1252,7 +1369,11 @@ rpcrdma_register_internal(struct rpcrdma_ia *ia, void *va, int len, va, len, DMA_BIDIRECTIONAL); iov->length = len; - if (ia->ri_bind_mem != NULL) { + if (ia->ri_have_dma_lkey) { + *mrp = NULL; + iov->lkey = ia->ri_dma_lkey; + return 0; + } else if (ia->ri_bind_mem != NULL) { *mrp = NULL; iov->lkey = ia->ri_bind_mem->lkey; return 0; @@ -1329,15 +1450,292 @@ rpcrdma_unmap_one(struct rpcrdma_ia *ia, struct rpcrdma_mr_seg *seg) seg->mr_dma, seg->mr_dmalen, seg->mr_dir); } +static int +rpcrdma_register_frmr_external(struct rpcrdma_mr_seg *seg, + int *nsegs, int writing, struct rpcrdma_ia *ia, + struct rpcrdma_xprt *r_xprt) +{ + struct rpcrdma_mr_seg *seg1 = seg; + struct ib_send_wr frmr_wr, *bad_wr; + u8 key; + int len, pageoff; + int i, rc; + + pageoff = offset_in_page(seg1->mr_offset); + seg1->mr_offset -= pageoff; /* start of page */ + seg1->mr_len += pageoff; + len = -pageoff; + if (*nsegs > RPCRDMA_MAX_DATA_SEGS) + *nsegs = RPCRDMA_MAX_DATA_SEGS; + for (i = 0; i < *nsegs;) { + rpcrdma_map_one(ia, seg, writing); + seg1->mr_chunk.rl_mw->r.frmr.fr_pgl->page_list[i] = seg->mr_dma; + len += seg->mr_len; + ++seg; + ++i; + /* Check for holes */ + if ((i < *nsegs && offset_in_page(seg->mr_offset)) || + offset_in_page((seg-1)->mr_offset + (seg-1)->mr_len)) + break; + } + dprintk("RPC: %s: Using frmr %p to map %d segments\n", + __func__, seg1->mr_chunk.rl_mw, i); + + /* Bump the key */ + key = (u8)(seg1->mr_chunk.rl_mw->r.frmr.fr_mr->rkey & 0x000000FF); + ib_update_fast_reg_key(seg1->mr_chunk.rl_mw->r.frmr.fr_mr, ++key); + + /* Prepare FRMR WR */ + memset(&frmr_wr, 0, sizeof frmr_wr); + frmr_wr.opcode = IB_WR_FAST_REG_MR; + frmr_wr.send_flags = 0; /* unsignaled */ + frmr_wr.wr.fast_reg.iova_start = (unsigned long)seg1->mr_dma; + frmr_wr.wr.fast_reg.page_list = seg1->mr_chunk.rl_mw->r.frmr.fr_pgl; + frmr_wr.wr.fast_reg.page_list_len = i; + frmr_wr.wr.fast_reg.page_shift = PAGE_SHIFT; + frmr_wr.wr.fast_reg.length = i << PAGE_SHIFT; + frmr_wr.wr.fast_reg.access_flags = (writing ? + IB_ACCESS_REMOTE_WRITE : IB_ACCESS_REMOTE_READ); + frmr_wr.wr.fast_reg.rkey = seg1->mr_chunk.rl_mw->r.frmr.fr_mr->rkey; + DECR_CQCOUNT(&r_xprt->rx_ep); + + rc = ib_post_send(ia->ri_id->qp, &frmr_wr, &bad_wr); + + if (rc) { + dprintk("RPC: %s: failed ib_post_send for register," + " status %i\n", __func__, rc); + while (i--) + rpcrdma_unmap_one(ia, --seg); + } else { + seg1->mr_rkey = seg1->mr_chunk.rl_mw->r.frmr.fr_mr->rkey; + seg1->mr_base = seg1->mr_dma + pageoff; + seg1->mr_nsegs = i; + seg1->mr_len = len; + } + *nsegs = i; + return rc; +} + +static int +rpcrdma_deregister_frmr_external(struct rpcrdma_mr_seg *seg, + struct rpcrdma_ia *ia, struct rpcrdma_xprt *r_xprt) +{ + struct rpcrdma_mr_seg *seg1 = seg; + struct ib_send_wr invalidate_wr, *bad_wr; + int rc; + + while (seg1->mr_nsegs--) + rpcrdma_unmap_one(ia, seg++); + + memset(&invalidate_wr, 0, sizeof invalidate_wr); + invalidate_wr.opcode = IB_WR_LOCAL_INV; + invalidate_wr.send_flags = 0; /* unsignaled */ + invalidate_wr.ex.invalidate_rkey = seg1->mr_chunk.rl_mw->r.frmr.fr_mr->rkey; + DECR_CQCOUNT(&r_xprt->rx_ep); + + rc = ib_post_send(ia->ri_id->qp, &invalidate_wr, &bad_wr); + if (rc) + dprintk("RPC: %s: failed ib_post_send for invalidate," + " status %i\n", __func__, rc); + return rc; +} + +static int +rpcrdma_register_fmr_external(struct rpcrdma_mr_seg *seg, + int *nsegs, int writing, struct rpcrdma_ia *ia) +{ + struct rpcrdma_mr_seg *seg1 = seg; + u64 physaddrs[RPCRDMA_MAX_DATA_SEGS]; + int len, pageoff, i, rc; + + pageoff = offset_in_page(seg1->mr_offset); + seg1->mr_offset -= pageoff; /* start of page */ + seg1->mr_len += pageoff; + len = -pageoff; + if (*nsegs > RPCRDMA_MAX_DATA_SEGS) + *nsegs = RPCRDMA_MAX_DATA_SEGS; + for (i = 0; i < *nsegs;) { + rpcrdma_map_one(ia, seg, writing); + physaddrs[i] = seg->mr_dma; + len += seg->mr_len; + ++seg; + ++i; + /* Check for holes */ + if ((i < *nsegs && offset_in_page(seg->mr_offset)) || + offset_in_page((seg-1)->mr_offset + (seg-1)->mr_len)) + break; + } + rc = ib_map_phys_fmr(seg1->mr_chunk.rl_mw->r.fmr, + physaddrs, i, seg1->mr_dma); + if (rc) { + dprintk("RPC: %s: failed ib_map_phys_fmr " + "%u@0x%llx+%i (%d)... status %i\n", __func__, + len, (unsigned long long)seg1->mr_dma, + pageoff, i, rc); + while (i--) + rpcrdma_unmap_one(ia, --seg); + } else { + seg1->mr_rkey = seg1->mr_chunk.rl_mw->r.fmr->rkey; + seg1->mr_base = seg1->mr_dma + pageoff; + seg1->mr_nsegs = i; + seg1->mr_len = len; + } + *nsegs = i; + return rc; +} + +static int +rpcrdma_deregister_fmr_external(struct rpcrdma_mr_seg *seg, + struct rpcrdma_ia *ia) +{ + struct rpcrdma_mr_seg *seg1 = seg; + LIST_HEAD(l); + int rc; + + list_add(&seg1->mr_chunk.rl_mw->r.fmr->list, &l); + rc = ib_unmap_fmr(&l); + while (seg1->mr_nsegs--) + rpcrdma_unmap_one(ia, seg++); + if (rc) + dprintk("RPC: %s: failed ib_unmap_fmr," + " status %i\n", __func__, rc); + return rc; +} + +static int +rpcrdma_register_memwin_external(struct rpcrdma_mr_seg *seg, + int *nsegs, int writing, struct rpcrdma_ia *ia, + struct rpcrdma_xprt *r_xprt) +{ + int mem_priv = (writing ? IB_ACCESS_REMOTE_WRITE : + IB_ACCESS_REMOTE_READ); + struct ib_mw_bind param; + int rc; + + *nsegs = 1; + rpcrdma_map_one(ia, seg, writing); + param.mr = ia->ri_bind_mem; + param.wr_id = 0ULL; /* no send cookie */ + param.addr = seg->mr_dma; + param.length = seg->mr_len; + param.send_flags = 0; + param.mw_access_flags = mem_priv; + + DECR_CQCOUNT(&r_xprt->rx_ep); + rc = ib_bind_mw(ia->ri_id->qp, seg->mr_chunk.rl_mw->r.mw, ¶m); + if (rc) { + dprintk("RPC: %s: failed ib_bind_mw " + "%u@0x%llx status %i\n", + __func__, seg->mr_len, + (unsigned long long)seg->mr_dma, rc); + rpcrdma_unmap_one(ia, seg); + } else { + seg->mr_rkey = seg->mr_chunk.rl_mw->r.mw->rkey; + seg->mr_base = param.addr; + seg->mr_nsegs = 1; + } + return rc; +} + +static int +rpcrdma_deregister_memwin_external(struct rpcrdma_mr_seg *seg, + struct rpcrdma_ia *ia, + struct rpcrdma_xprt *r_xprt, void **r) +{ + struct ib_mw_bind param; + LIST_HEAD(l); + int rc; + + BUG_ON(seg->mr_nsegs != 1); + param.mr = ia->ri_bind_mem; + param.addr = 0ULL; /* unbind */ + param.length = 0; + param.mw_access_flags = 0; + if (*r) { + param.wr_id = (u64) (unsigned long) *r; + param.send_flags = IB_SEND_SIGNALED; + INIT_CQCOUNT(&r_xprt->rx_ep); + } else { + param.wr_id = 0ULL; + param.send_flags = 0; + DECR_CQCOUNT(&r_xprt->rx_ep); + } + rc = ib_bind_mw(ia->ri_id->qp, seg->mr_chunk.rl_mw->r.mw, ¶m); + rpcrdma_unmap_one(ia, seg); + if (rc) + dprintk("RPC: %s: failed ib_(un)bind_mw," + " status %i\n", __func__, rc); + else + *r = NULL; /* will upcall on completion */ + return rc; +} + +static int +rpcrdma_register_default_external(struct rpcrdma_mr_seg *seg, + int *nsegs, int writing, struct rpcrdma_ia *ia) +{ + int mem_priv = (writing ? IB_ACCESS_REMOTE_WRITE : + IB_ACCESS_REMOTE_READ); + struct rpcrdma_mr_seg *seg1 = seg; + struct ib_phys_buf ipb[RPCRDMA_MAX_DATA_SEGS]; + int len, i, rc = 0; + + if (*nsegs > RPCRDMA_MAX_DATA_SEGS) + *nsegs = RPCRDMA_MAX_DATA_SEGS; + for (len = 0, i = 0; i < *nsegs;) { + rpcrdma_map_one(ia, seg, writing); + ipb[i].addr = seg->mr_dma; + ipb[i].size = seg->mr_len; + len += seg->mr_len; + ++seg; + ++i; + /* Check for holes */ + if ((i < *nsegs && offset_in_page(seg->mr_offset)) || + offset_in_page((seg-1)->mr_offset+(seg-1)->mr_len)) + break; + } + seg1->mr_base = seg1->mr_dma; + seg1->mr_chunk.rl_mr = ib_reg_phys_mr(ia->ri_pd, + ipb, i, mem_priv, &seg1->mr_base); + if (IS_ERR(seg1->mr_chunk.rl_mr)) { + rc = PTR_ERR(seg1->mr_chunk.rl_mr); + dprintk("RPC: %s: failed ib_reg_phys_mr " + "%u@0x%llx (%d)... status %i\n", + __func__, len, + (unsigned long long)seg1->mr_dma, i, rc); + while (i--) + rpcrdma_unmap_one(ia, --seg); + } else { + seg1->mr_rkey = seg1->mr_chunk.rl_mr->rkey; + seg1->mr_nsegs = i; + seg1->mr_len = len; + } + *nsegs = i; + return rc; +} + +static int +rpcrdma_deregister_default_external(struct rpcrdma_mr_seg *seg, + struct rpcrdma_ia *ia) +{ + struct rpcrdma_mr_seg *seg1 = seg; + int rc; + + rc = ib_dereg_mr(seg1->mr_chunk.rl_mr); + seg1->mr_chunk.rl_mr = NULL; + while (seg1->mr_nsegs--) + rpcrdma_unmap_one(ia, seg++); + if (rc) + dprintk("RPC: %s: failed ib_dereg_mr," + " status %i\n", __func__, rc); + return rc; +} + int rpcrdma_register_external(struct rpcrdma_mr_seg *seg, int nsegs, int writing, struct rpcrdma_xprt *r_xprt) { struct rpcrdma_ia *ia = &r_xprt->rx_ia; - int mem_priv = (writing ? IB_ACCESS_REMOTE_WRITE : - IB_ACCESS_REMOTE_READ); - struct rpcrdma_mr_seg *seg1 = seg; - int i; int rc = 0; switch (ia->ri_memreg_strategy) { @@ -1352,114 +1750,25 @@ rpcrdma_register_external(struct rpcrdma_mr_seg *seg, break; #endif - /* Registration using fast memory registration */ + /* Registration using frmr registration */ + case RPCRDMA_FRMR: + rc = rpcrdma_register_frmr_external(seg, &nsegs, writing, ia, r_xprt); + break; + + /* Registration using fmr memory registration */ case RPCRDMA_MTHCAFMR: - { - u64 physaddrs[RPCRDMA_MAX_DATA_SEGS]; - int len, pageoff = offset_in_page(seg->mr_offset); - seg1->mr_offset -= pageoff; /* start of page */ - seg1->mr_len += pageoff; - len = -pageoff; - if (nsegs > RPCRDMA_MAX_DATA_SEGS) - nsegs = RPCRDMA_MAX_DATA_SEGS; - for (i = 0; i < nsegs;) { - rpcrdma_map_one(ia, seg, writing); - physaddrs[i] = seg->mr_dma; - len += seg->mr_len; - ++seg; - ++i; - /* Check for holes */ - if ((i < nsegs && offset_in_page(seg->mr_offset)) || - offset_in_page((seg-1)->mr_offset+(seg-1)->mr_len)) - break; - } - nsegs = i; - rc = ib_map_phys_fmr(seg1->mr_chunk.rl_mw->r.fmr, - physaddrs, nsegs, seg1->mr_dma); - if (rc) { - dprintk("RPC: %s: failed ib_map_phys_fmr " - "%u@0x%llx+%i (%d)... status %i\n", __func__, - len, (unsigned long long)seg1->mr_dma, - pageoff, nsegs, rc); - while (nsegs--) - rpcrdma_unmap_one(ia, --seg); - } else { - seg1->mr_rkey = seg1->mr_chunk.rl_mw->r.fmr->rkey; - seg1->mr_base = seg1->mr_dma + pageoff; - seg1->mr_nsegs = nsegs; - seg1->mr_len = len; - } - } + rc = rpcrdma_register_fmr_external(seg, &nsegs, writing, ia); break; /* Registration using memory windows */ case RPCRDMA_MEMWINDOWS_ASYNC: case RPCRDMA_MEMWINDOWS: - { - struct ib_mw_bind param; - rpcrdma_map_one(ia, seg, writing); - param.mr = ia->ri_bind_mem; - param.wr_id = 0ULL; /* no send cookie */ - param.addr = seg->mr_dma; - param.length = seg->mr_len; - param.send_flags = 0; - param.mw_access_flags = mem_priv; - - DECR_CQCOUNT(&r_xprt->rx_ep); - rc = ib_bind_mw(ia->ri_id->qp, - seg->mr_chunk.rl_mw->r.mw, ¶m); - if (rc) { - dprintk("RPC: %s: failed ib_bind_mw " - "%u@0x%llx status %i\n", - __func__, seg->mr_len, - (unsigned long long)seg->mr_dma, rc); - rpcrdma_unmap_one(ia, seg); - } else { - seg->mr_rkey = seg->mr_chunk.rl_mw->r.mw->rkey; - seg->mr_base = param.addr; - seg->mr_nsegs = 1; - nsegs = 1; - } - } + rc = rpcrdma_register_memwin_external(seg, &nsegs, writing, ia, r_xprt); break; /* Default registration each time */ default: - { - struct ib_phys_buf ipb[RPCRDMA_MAX_DATA_SEGS]; - int len = 0; - if (nsegs > RPCRDMA_MAX_DATA_SEGS) - nsegs = RPCRDMA_MAX_DATA_SEGS; - for (i = 0; i < nsegs;) { - rpcrdma_map_one(ia, seg, writing); - ipb[i].addr = seg->mr_dma; - ipb[i].size = seg->mr_len; - len += seg->mr_len; - ++seg; - ++i; - /* Check for holes */ - if ((i < nsegs && offset_in_page(seg->mr_offset)) || - offset_in_page((seg-1)->mr_offset+(seg-1)->mr_len)) - break; - } - nsegs = i; - seg1->mr_base = seg1->mr_dma; - seg1->mr_chunk.rl_mr = ib_reg_phys_mr(ia->ri_pd, - ipb, nsegs, mem_priv, &seg1->mr_base); - if (IS_ERR(seg1->mr_chunk.rl_mr)) { - rc = PTR_ERR(seg1->mr_chunk.rl_mr); - dprintk("RPC: %s: failed ib_reg_phys_mr " - "%u@0x%llx (%d)... status %i\n", - __func__, len, - (unsigned long long)seg1->mr_dma, nsegs, rc); - while (nsegs--) - rpcrdma_unmap_one(ia, --seg); - } else { - seg1->mr_rkey = seg1->mr_chunk.rl_mr->rkey; - seg1->mr_nsegs = nsegs; - seg1->mr_len = len; - } - } + rc = rpcrdma_register_default_external(seg, &nsegs, writing, ia); break; } if (rc) @@ -1473,7 +1782,6 @@ rpcrdma_deregister_external(struct rpcrdma_mr_seg *seg, struct rpcrdma_xprt *r_xprt, void *r) { struct rpcrdma_ia *ia = &r_xprt->rx_ia; - struct rpcrdma_mr_seg *seg1 = seg; int nsegs = seg->mr_nsegs, rc; switch (ia->ri_memreg_strategy) { @@ -1486,56 +1794,21 @@ rpcrdma_deregister_external(struct rpcrdma_mr_seg *seg, break; #endif + case RPCRDMA_FRMR: + rc = rpcrdma_deregister_frmr_external(seg, ia, r_xprt); + break; + case RPCRDMA_MTHCAFMR: - { - LIST_HEAD(l); - list_add(&seg->mr_chunk.rl_mw->r.fmr->list, &l); - rc = ib_unmap_fmr(&l); - while (seg1->mr_nsegs--) - rpcrdma_unmap_one(ia, seg++); - } - if (rc) - dprintk("RPC: %s: failed ib_unmap_fmr," - " status %i\n", __func__, rc); + rc = rpcrdma_deregister_fmr_external(seg, ia); break; case RPCRDMA_MEMWINDOWS_ASYNC: case RPCRDMA_MEMWINDOWS: - { - struct ib_mw_bind param; - BUG_ON(nsegs != 1); - param.mr = ia->ri_bind_mem; - param.addr = 0ULL; /* unbind */ - param.length = 0; - param.mw_access_flags = 0; - if (r) { - param.wr_id = (u64) (unsigned long) r; - param.send_flags = IB_SEND_SIGNALED; - INIT_CQCOUNT(&r_xprt->rx_ep); - } else { - param.wr_id = 0ULL; - param.send_flags = 0; - DECR_CQCOUNT(&r_xprt->rx_ep); - } - rc = ib_bind_mw(ia->ri_id->qp, - seg->mr_chunk.rl_mw->r.mw, ¶m); - rpcrdma_unmap_one(ia, seg); - } - if (rc) - dprintk("RPC: %s: failed ib_(un)bind_mw," - " status %i\n", __func__, rc); - else - r = NULL; /* will upcall on completion */ + rc = rpcrdma_deregister_memwin_external(seg, ia, r_xprt, &r); break; default: - rc = ib_dereg_mr(seg1->mr_chunk.rl_mr); - seg1->mr_chunk.rl_mr = NULL; - while (seg1->mr_nsegs--) - rpcrdma_unmap_one(ia, seg++); - if (rc) - dprintk("RPC: %s: failed ib_dereg_mr," - " status %i\n", __func__, rc); + rc = rpcrdma_deregister_default_external(seg, ia); break; } if (r) { diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h index 2427822f8bd4..c7a7eba991bc 100644 --- a/net/sunrpc/xprtrdma/xprt_rdma.h +++ b/net/sunrpc/xprtrdma/xprt_rdma.h @@ -51,6 +51,9 @@ #include <linux/sunrpc/rpc_rdma.h> /* RPC/RDMA protocol */ #include <linux/sunrpc/xprtrdma.h> /* xprt parameters */ +#define RDMA_RESOLVE_TIMEOUT (5000) /* 5 seconds */ +#define RDMA_CONNECT_RETRY_MAX (2) /* retries if no listener backlog */ + /* * Interface Adapter -- one per transport instance */ @@ -58,6 +61,8 @@ struct rpcrdma_ia { struct rdma_cm_id *ri_id; struct ib_pd *ri_pd; struct ib_mr *ri_bind_mem; + u32 ri_dma_lkey; + int ri_have_dma_lkey; struct completion ri_done; int ri_async_rc; enum rpcrdma_memreg ri_memreg_strategy; @@ -156,6 +161,10 @@ struct rpcrdma_mr_seg { /* chunk descriptors */ union { struct ib_mw *mw; struct ib_fmr *fmr; + struct { + struct ib_fast_reg_page_list *fr_pgl; + struct ib_mr *fr_mr; + } frmr; } r; struct list_head mw_list; } *rl_mw; @@ -175,6 +184,7 @@ struct rpcrdma_req { size_t rl_size; /* actual length of buffer */ unsigned int rl_niovs; /* 0, 2 or 4 */ unsigned int rl_nchunks; /* non-zero if chunks */ + unsigned int rl_connect_cookie; /* retry detection */ struct rpcrdma_buffer *rl_buffer; /* home base for this structure */ struct rpcrdma_rep *rl_reply;/* holder for reply buffer */ struct rpcrdma_mr_seg rl_segments[RPCRDMA_MAX_SEGS];/* chunk segments */ @@ -198,7 +208,7 @@ struct rpcrdma_buffer { atomic_t rb_credits; /* most recent server credits */ unsigned long rb_cwndscale; /* cached framework rpc_cwndscale */ int rb_max_requests;/* client max requests */ - struct list_head rb_mws; /* optional memory windows/fmrs */ + struct list_head rb_mws; /* optional memory windows/fmrs/frmrs */ int rb_send_index; struct rpcrdma_req **rb_send_bufs; int rb_recv_index; @@ -273,6 +283,11 @@ struct rpcrdma_xprt { #define rpcx_to_rdmax(x) container_of(x, struct rpcrdma_xprt, xprt) #define rpcx_to_rdmad(x) (rpcx_to_rdmax(x)->rx_data) +/* Setting this to 0 ensures interoperability with early servers. + * Setting this to 1 enhances certain unaligned read/write performance. + * Default is 0, see sysctl entry and rpc_rdma.c rpcrdma_convert_iovs() */ +extern int xprt_rdma_pad_optimize; + /* * Interface Adapter calls - xprtrdma/verbs.c */ diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index 4486c59c3aca..9a288d5eea64 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -3,8 +3,8 @@ * * Client-side transport implementation for sockets. * - * TCP callback races fixes (C) 1998 Red Hat Software <alan@redhat.com> - * TCP send fixes (C) 1998 Red Hat Software <alan@redhat.com> + * TCP callback races fixes (C) 1998 Red Hat + * TCP send fixes (C) 1998 Red Hat * TCP NFS related read + write fixes * (C) 1999 Dave Airlie, University of Limerick, Ireland <airlied@linux.ie> * diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 015606b54d9b..c647aab8d418 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -1,7 +1,7 @@ /* * NET4: Implementation of BSD Unix domain sockets. * - * Authors: Alan Cox, <alan.cox@linux.org> + * Authors: Alan Cox, <alan@lxorguk.ukuu.org.uk> * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License diff --git a/net/wireless/core.c b/net/wireless/core.c index 5cadbeb76a14..5031db7b275b 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -13,7 +13,6 @@ #include <linux/debugfs.h> #include <linux/notifier.h> #include <linux/device.h> -#include <linux/list.h> #include <net/genetlink.h> #include <net/cfg80211.h> #include <net/wireless.h> @@ -185,7 +184,8 @@ int cfg80211_dev_rename(struct cfg80211_registered_device *rdev, if (result) goto out_unlock; - if (!debugfs_rename(rdev->wiphy.debugfsdir->d_parent, + if (rdev->wiphy.debugfsdir && + !debugfs_rename(rdev->wiphy.debugfsdir->d_parent, rdev->wiphy.debugfsdir, rdev->wiphy.debugfsdir->d_parent, newname)) @@ -318,6 +318,8 @@ int wiphy_register(struct wiphy *wiphy) drv->wiphy.debugfsdir = debugfs_create_dir(wiphy_name(&drv->wiphy), ieee80211_debugfs_dir); + if (IS_ERR(drv->wiphy.debugfsdir)) + drv->wiphy.debugfsdir = NULL; res = 0; out_unlock: |