summaryrefslogtreecommitdiff
path: root/net/sunrpc
diff options
context:
space:
mode:
Diffstat (limited to 'net/sunrpc')
-rw-r--r--net/sunrpc/auth_gss/auth_gss.c59
-rw-r--r--net/sunrpc/cache.c17
-rw-r--r--net/sunrpc/clnt.c9
-rw-r--r--net/sunrpc/rpc_pipe.c12
-rw-r--r--net/sunrpc/svc.c80
-rw-r--r--net/sunrpc/svc_xprt.c11
-rw-r--r--net/sunrpc/svcsock.c17
-rw-r--r--net/sunrpc/xdr.c1
-rw-r--r--net/sunrpc/xprt.c3
-rw-r--r--net/sunrpc/xprtrdma/svc_rdma_recvfrom.c8
-rw-r--r--net/sunrpc/xprtrdma/svc_rdma_rw.c2
-rw-r--r--net/sunrpc/xprtrdma/svc_rdma_sendto.c16
-rw-r--r--net/sunrpc/xprtrdma/svc_rdma_transport.c15
-rw-r--r--net/sunrpc/xprtsock.c16
14 files changed, 130 insertions, 136 deletions
diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index 369310909fc9..0fa244f16876 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -1545,6 +1545,7 @@ static int gss_marshal(struct rpc_task *task, struct xdr_stream *xdr)
struct kvec iov;
struct xdr_buf verf_buf;
int status;
+ u32 seqno;
/* Credential */
@@ -1556,15 +1557,16 @@ static int gss_marshal(struct rpc_task *task, struct xdr_stream *xdr)
cred_len = p++;
spin_lock(&ctx->gc_seq_lock);
- req->rq_seqno = (ctx->gc_seq < MAXSEQ) ? ctx->gc_seq++ : MAXSEQ;
+ seqno = (ctx->gc_seq < MAXSEQ) ? ctx->gc_seq++ : MAXSEQ;
+ xprt_rqst_add_seqno(req, seqno);
spin_unlock(&ctx->gc_seq_lock);
- if (req->rq_seqno == MAXSEQ)
+ if (*req->rq_seqnos == MAXSEQ)
goto expired;
trace_rpcgss_seqno(task);
*p++ = cpu_to_be32(RPC_GSS_VERSION);
*p++ = cpu_to_be32(ctx->gc_proc);
- *p++ = cpu_to_be32(req->rq_seqno);
+ *p++ = cpu_to_be32(*req->rq_seqnos);
*p++ = cpu_to_be32(gss_cred->gc_service);
p = xdr_encode_netobj(p, &ctx->gc_wire_ctx);
*cred_len = cpu_to_be32((p - (cred_len + 1)) << 2);
@@ -1678,17 +1680,31 @@ gss_refresh_null(struct rpc_task *task)
return 0;
}
+static u32
+gss_validate_seqno_mic(struct gss_cl_ctx *ctx, u32 seqno, __be32 *seq, __be32 *p, u32 len)
+{
+ struct kvec iov;
+ struct xdr_buf verf_buf;
+ struct xdr_netobj mic;
+
+ *seq = cpu_to_be32(seqno);
+ iov.iov_base = seq;
+ iov.iov_len = 4;
+ xdr_buf_from_iov(&iov, &verf_buf);
+ mic.data = (u8 *)p;
+ mic.len = len;
+ return gss_verify_mic(ctx->gc_gss_ctx, &verf_buf, &mic);
+}
+
static int
gss_validate(struct rpc_task *task, struct xdr_stream *xdr)
{
struct rpc_cred *cred = task->tk_rqstp->rq_cred;
struct gss_cl_ctx *ctx = gss_cred_get_ctx(cred);
__be32 *p, *seq = NULL;
- struct kvec iov;
- struct xdr_buf verf_buf;
- struct xdr_netobj mic;
u32 len, maj_stat;
int status;
+ int i = 1; /* don't recheck the first item */
p = xdr_inline_decode(xdr, 2 * sizeof(*p));
if (!p)
@@ -1705,13 +1721,10 @@ gss_validate(struct rpc_task *task, struct xdr_stream *xdr)
seq = kmalloc(4, GFP_KERNEL);
if (!seq)
goto validate_failed;
- *seq = cpu_to_be32(task->tk_rqstp->rq_seqno);
- iov.iov_base = seq;
- iov.iov_len = 4;
- xdr_buf_from_iov(&iov, &verf_buf);
- mic.data = (u8 *)p;
- mic.len = len;
- maj_stat = gss_verify_mic(ctx->gc_gss_ctx, &verf_buf, &mic);
+ maj_stat = gss_validate_seqno_mic(ctx, task->tk_rqstp->rq_seqnos[0], seq, p, len);
+ /* RFC 2203 5.3.3.1 - compute the checksum of each sequence number in the cache */
+ while (unlikely(maj_stat == GSS_S_BAD_SIG && i < task->tk_rqstp->rq_seqno_count))
+ maj_stat = gss_validate_seqno_mic(ctx, task->tk_rqstp->rq_seqnos[i], seq, p, len);
if (maj_stat == GSS_S_CONTEXT_EXPIRED)
clear_bit(RPCAUTH_CRED_UPTODATE, &cred->cr_flags);
if (maj_stat)
@@ -1750,7 +1763,7 @@ gss_wrap_req_integ(struct rpc_cred *cred, struct gss_cl_ctx *ctx,
if (!p)
goto wrap_failed;
integ_len = p++;
- *p = cpu_to_be32(rqstp->rq_seqno);
+ *p = cpu_to_be32(*rqstp->rq_seqnos);
if (rpcauth_wrap_req_encode(task, xdr))
goto wrap_failed;
@@ -1847,7 +1860,7 @@ gss_wrap_req_priv(struct rpc_cred *cred, struct gss_cl_ctx *ctx,
if (!p)
goto wrap_failed;
opaque_len = p++;
- *p = cpu_to_be32(rqstp->rq_seqno);
+ *p = cpu_to_be32(*rqstp->rq_seqnos);
if (rpcauth_wrap_req_encode(task, xdr))
goto wrap_failed;
@@ -2001,7 +2014,7 @@ gss_unwrap_resp_integ(struct rpc_task *task, struct rpc_cred *cred,
offset = rcv_buf->len - xdr_stream_remaining(xdr);
if (xdr_stream_decode_u32(xdr, &seqno))
goto unwrap_failed;
- if (seqno != rqstp->rq_seqno)
+ if (seqno != *rqstp->rq_seqnos)
goto bad_seqno;
if (xdr_buf_subsegment(rcv_buf, &gss_data, offset, len))
goto unwrap_failed;
@@ -2045,7 +2058,7 @@ unwrap_failed:
trace_rpcgss_unwrap_failed(task);
goto out;
bad_seqno:
- trace_rpcgss_bad_seqno(task, rqstp->rq_seqno, seqno);
+ trace_rpcgss_bad_seqno(task, *rqstp->rq_seqnos, seqno);
goto out;
bad_mic:
trace_rpcgss_verify_mic(task, maj_stat);
@@ -2077,7 +2090,7 @@ gss_unwrap_resp_priv(struct rpc_task *task, struct rpc_cred *cred,
if (maj_stat != GSS_S_COMPLETE)
goto bad_unwrap;
/* gss_unwrap decrypted the sequence number */
- if (be32_to_cpup(p++) != rqstp->rq_seqno)
+ if (be32_to_cpup(p++) != *rqstp->rq_seqnos)
goto bad_seqno;
/* gss_unwrap redacts the opaque blob from the head iovec.
@@ -2093,7 +2106,7 @@ unwrap_failed:
trace_rpcgss_unwrap_failed(task);
return -EIO;
bad_seqno:
- trace_rpcgss_bad_seqno(task, rqstp->rq_seqno, be32_to_cpup(--p));
+ trace_rpcgss_bad_seqno(task, *rqstp->rq_seqnos, be32_to_cpup(--p));
return -EIO;
bad_unwrap:
trace_rpcgss_unwrap(task, maj_stat);
@@ -2118,14 +2131,14 @@ gss_xmit_need_reencode(struct rpc_task *task)
if (!ctx)
goto out;
- if (gss_seq_is_newer(req->rq_seqno, READ_ONCE(ctx->gc_seq)))
+ if (gss_seq_is_newer(*req->rq_seqnos, READ_ONCE(ctx->gc_seq)))
goto out_ctx;
seq_xmit = READ_ONCE(ctx->gc_seq_xmit);
- while (gss_seq_is_newer(req->rq_seqno, seq_xmit)) {
+ while (gss_seq_is_newer(*req->rq_seqnos, seq_xmit)) {
u32 tmp = seq_xmit;
- seq_xmit = cmpxchg(&ctx->gc_seq_xmit, tmp, req->rq_seqno);
+ seq_xmit = cmpxchg(&ctx->gc_seq_xmit, tmp, *req->rq_seqnos);
if (seq_xmit == tmp) {
ret = false;
goto out_ctx;
@@ -2134,7 +2147,7 @@ gss_xmit_need_reencode(struct rpc_task *task)
win = ctx->gc_win;
if (win > 0)
- ret = !gss_seq_is_newer(req->rq_seqno, seq_xmit - win);
+ ret = !gss_seq_is_newer(*req->rq_seqnos, seq_xmit - win);
out_ctx:
gss_put_ctx(ctx);
diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c
index 7ce5e28a6c03..131090f31e6a 100644
--- a/net/sunrpc/cache.c
+++ b/net/sunrpc/cache.c
@@ -135,6 +135,8 @@ static struct cache_head *sunrpc_cache_add_entry(struct cache_detail *detail,
hlist_add_head_rcu(&new->cache_list, head);
detail->entries++;
+ if (detail->nextcheck > new->expiry_time)
+ detail->nextcheck = new->expiry_time + 1;
cache_get(new);
spin_unlock(&detail->hash_lock);
@@ -462,24 +464,21 @@ static int cache_clean(void)
}
}
+ spin_lock(&current_detail->hash_lock);
+
/* find a non-empty bucket in the table */
- while (current_detail &&
- current_index < current_detail->hash_size &&
+ while (current_index < current_detail->hash_size &&
hlist_empty(&current_detail->hash_table[current_index]))
current_index++;
/* find a cleanable entry in the bucket and clean it, or set to next bucket */
-
- if (current_detail && current_index < current_detail->hash_size) {
+ if (current_index < current_detail->hash_size) {
struct cache_head *ch = NULL;
struct cache_detail *d;
struct hlist_head *head;
struct hlist_node *tmp;
- spin_lock(&current_detail->hash_lock);
-
/* Ok, now to clean this strand */
-
head = &current_detail->hash_table[current_index];
hlist_for_each_entry_safe(ch, tmp, head, cache_list) {
if (current_detail->nextcheck > ch->expiry_time)
@@ -500,8 +499,10 @@ static int cache_clean(void)
spin_unlock(&cache_list_lock);
if (ch)
sunrpc_end_cache_remove_entry(ch, d);
- } else
+ } else {
+ spin_unlock(&current_detail->hash_lock);
spin_unlock(&cache_list_lock);
+ }
return rv;
}
diff --git a/net/sunrpc/clnt.c b/net/sunrpc/clnt.c
index 6f75862d9782..21426c3049d3 100644
--- a/net/sunrpc/clnt.c
+++ b/net/sunrpc/clnt.c
@@ -2771,8 +2771,13 @@ out_verifier:
case -EPROTONOSUPPORT:
goto out_err;
case -EACCES:
- /* Re-encode with a fresh cred */
- fallthrough;
+ /* possible RPCSEC_GSS out-of-sequence event (RFC2203),
+ * reset recv state and keep waiting, don't retransmit
+ */
+ task->tk_rqstp->rq_reply_bytes_recvd = 0;
+ task->tk_status = xprt_request_enqueue_receive(task);
+ task->tk_action = call_transmit_status;
+ return -EBADMSG;
default:
goto out_garbage;
}
diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c
index eadc00410ebc..98f78cd55905 100644
--- a/net/sunrpc/rpc_pipe.c
+++ b/net/sunrpc/rpc_pipe.c
@@ -631,7 +631,7 @@ static struct dentry *__rpc_lookup_create_exclusive(struct dentry *parent,
const char *name)
{
struct qstr q = QSTR(name);
- struct dentry *dentry = d_hash_and_lookup(parent, &q);
+ struct dentry *dentry = try_lookup_noperm(&q, parent);
if (!dentry) {
dentry = d_alloc(parent, &q);
if (!dentry)
@@ -658,7 +658,7 @@ static void __rpc_depopulate(struct dentry *parent,
for (i = start; i < eof; i++) {
name.name = files[i].name;
name.len = strlen(files[i].name);
- dentry = d_hash_and_lookup(parent, &name);
+ dentry = try_lookup_noperm(&name, parent);
if (dentry == NULL)
continue;
@@ -1190,7 +1190,7 @@ static const struct rpc_filelist files[] = {
struct dentry *rpc_d_lookup_sb(const struct super_block *sb,
const unsigned char *dir_name)
{
- return d_hash_and_lookup(sb->s_root, &QSTR(dir_name));
+ return try_lookup_noperm(&QSTR(dir_name), sb->s_root);
}
EXPORT_SYMBOL_GPL(rpc_d_lookup_sb);
@@ -1301,7 +1301,7 @@ rpc_gssd_dummy_populate(struct dentry *root, struct rpc_pipe *pipe_data)
struct dentry *pipe_dentry = NULL;
/* We should never get this far if "gssd" doesn't exist */
- gssd_dentry = d_hash_and_lookup(root, &QSTR(files[RPCAUTH_gssd].name));
+ gssd_dentry = try_lookup_noperm(&QSTR(files[RPCAUTH_gssd].name), root);
if (!gssd_dentry)
return ERR_PTR(-ENOENT);
@@ -1311,8 +1311,8 @@ rpc_gssd_dummy_populate(struct dentry *root, struct rpc_pipe *pipe_data)
goto out;
}
- clnt_dentry = d_hash_and_lookup(gssd_dentry,
- &QSTR(gssd_dummy_clnt_dir[0].name));
+ clnt_dentry = try_lookup_noperm(&QSTR(gssd_dummy_clnt_dir[0].name),
+ gssd_dentry);
if (!clnt_dentry) {
__rpc_depopulate(gssd_dentry, gssd_dummy_clnt_dir, 0, 1);
pipe_dentry = ERR_PTR(-ENOENT);
diff --git a/net/sunrpc/svc.c b/net/sunrpc/svc.c
index e7f9c295d13c..939b6239df8a 100644
--- a/net/sunrpc/svc.c
+++ b/net/sunrpc/svc.c
@@ -636,24 +636,22 @@ svc_destroy(struct svc_serv **servp)
EXPORT_SYMBOL_GPL(svc_destroy);
static bool
-svc_init_buffer(struct svc_rqst *rqstp, unsigned int size, int node)
+svc_init_buffer(struct svc_rqst *rqstp, const struct svc_serv *serv, int node)
{
- unsigned long pages, ret;
+ unsigned long ret;
- /* bc_xprt uses fore channel allocated buffers */
- if (svc_is_backchannel(rqstp))
- return true;
+ rqstp->rq_maxpages = svc_serv_maxpages(serv);
- pages = size / PAGE_SIZE + 1; /* extra page as we hold both request and reply.
- * We assume one is at most one page
- */
- WARN_ON_ONCE(pages > RPCSVC_MAXPAGES);
- if (pages > RPCSVC_MAXPAGES)
- pages = RPCSVC_MAXPAGES;
+ /* rq_pages' last entry is NULL for historical reasons. */
+ rqstp->rq_pages = kcalloc_node(rqstp->rq_maxpages + 1,
+ sizeof(struct page *),
+ GFP_KERNEL, node);
+ if (!rqstp->rq_pages)
+ return false;
- ret = alloc_pages_bulk_node(GFP_KERNEL, node, pages,
+ ret = alloc_pages_bulk_node(GFP_KERNEL, node, rqstp->rq_maxpages,
rqstp->rq_pages);
- return ret == pages;
+ return ret == rqstp->rq_maxpages;
}
/*
@@ -662,17 +660,19 @@ svc_init_buffer(struct svc_rqst *rqstp, unsigned int size, int node)
static void
svc_release_buffer(struct svc_rqst *rqstp)
{
- unsigned int i;
+ unsigned long i;
- for (i = 0; i < ARRAY_SIZE(rqstp->rq_pages); i++)
+ for (i = 0; i < rqstp->rq_maxpages; i++)
if (rqstp->rq_pages[i])
put_page(rqstp->rq_pages[i]);
+ kfree(rqstp->rq_pages);
}
static void
svc_rqst_free(struct svc_rqst *rqstp)
{
folio_batch_release(&rqstp->rq_fbatch);
+ kfree(rqstp->rq_bvec);
svc_release_buffer(rqstp);
if (rqstp->rq_scratch_page)
put_page(rqstp->rq_scratch_page);
@@ -708,7 +708,13 @@ svc_prepare_thread(struct svc_serv *serv, struct svc_pool *pool, int node)
if (!rqstp->rq_resp)
goto out_enomem;
- if (!svc_init_buffer(rqstp, serv->sv_max_mesg, node))
+ if (!svc_init_buffer(rqstp, serv, node))
+ goto out_enomem;
+
+ rqstp->rq_bvec = kcalloc_node(rqstp->rq_maxpages,
+ sizeof(struct bio_vec),
+ GFP_KERNEL, node);
+ if (!rqstp->rq_bvec)
goto out_enomem;
rqstp->rq_err = -EAGAIN; /* No error yet */
@@ -900,7 +906,7 @@ EXPORT_SYMBOL_GPL(svc_set_num_threads);
bool svc_rqst_replace_page(struct svc_rqst *rqstp, struct page *page)
{
struct page **begin = rqstp->rq_pages;
- struct page **end = &rqstp->rq_pages[RPCSVC_MAXPAGES];
+ struct page **end = &rqstp->rq_pages[rqstp->rq_maxpages];
if (unlikely(rqstp->rq_next_page < begin || rqstp->rq_next_page > end)) {
trace_svc_replace_page_err(rqstp);
@@ -1714,46 +1720,6 @@ int svc_encode_result_payload(struct svc_rqst *rqstp, unsigned int offset,
EXPORT_SYMBOL_GPL(svc_encode_result_payload);
/**
- * svc_fill_write_vector - Construct data argument for VFS write call
- * @rqstp: svc_rqst to operate on
- * @payload: xdr_buf containing only the write data payload
- *
- * Fills in rqstp::rq_vec, and returns the number of elements.
- */
-unsigned int svc_fill_write_vector(struct svc_rqst *rqstp,
- struct xdr_buf *payload)
-{
- struct page **pages = payload->pages;
- struct kvec *first = payload->head;
- struct kvec *vec = rqstp->rq_vec;
- size_t total = payload->len;
- unsigned int i;
-
- /* Some types of transport can present the write payload
- * entirely in rq_arg.pages. In this case, @first is empty.
- */
- i = 0;
- if (first->iov_len) {
- vec[i].iov_base = first->iov_base;
- vec[i].iov_len = min_t(size_t, total, first->iov_len);
- total -= vec[i].iov_len;
- ++i;
- }
-
- while (total) {
- vec[i].iov_base = page_address(*pages);
- vec[i].iov_len = min_t(size_t, total, PAGE_SIZE);
- total -= vec[i].iov_len;
- ++i;
- ++pages;
- }
-
- WARN_ON_ONCE(i > ARRAY_SIZE(rqstp->rq_vec));
- return i;
-}
-EXPORT_SYMBOL_GPL(svc_fill_write_vector);
-
-/**
* svc_fill_symlink_pathname - Construct pathname argument for VFS symlink call
* @rqstp: svc_rqst to operate on
* @first: buffer containing first section of pathname
diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c
index ae25405d8bd2..cb14d6ddac6c 100644
--- a/net/sunrpc/svc_xprt.c
+++ b/net/sunrpc/svc_xprt.c
@@ -488,6 +488,7 @@ void svc_xprt_enqueue(struct svc_xprt *xprt)
pool = svc_pool_for_cpu(xprt->xpt_server);
percpu_counter_inc(&pool->sp_sockets_queued);
+ xprt->xpt_qtime = ktime_get();
lwq_enqueue(&xprt->xpt_ready, &pool->sp_xprts);
svc_pool_wake_idle_thread(pool);
@@ -651,18 +652,10 @@ static void svc_check_conn_limits(struct svc_serv *serv)
static bool svc_alloc_arg(struct svc_rqst *rqstp)
{
- struct svc_serv *serv = rqstp->rq_server;
struct xdr_buf *arg = &rqstp->rq_arg;
unsigned long pages, filled, ret;
- pages = (serv->sv_max_mesg + 2 * PAGE_SIZE) >> PAGE_SHIFT;
- if (pages > RPCSVC_MAXPAGES) {
- pr_warn_once("svc: warning: pages=%lu > RPCSVC_MAXPAGES=%lu\n",
- pages, RPCSVC_MAXPAGES);
- /* use as many pages as possible */
- pages = RPCSVC_MAXPAGES;
- }
-
+ pages = rqstp->rq_maxpages;
for (filled = 0; filled < pages; filled = ret) {
ret = alloc_pages_bulk(GFP_KERNEL, pages, rqstp->rq_pages);
if (ret > filled)
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 72e5a01df3d3..e1c85123b445 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -713,8 +713,7 @@ static int svc_udp_sendto(struct svc_rqst *rqstp)
if (svc_xprt_is_dead(xprt))
goto out_notconn;
- count = xdr_buf_to_bvec(rqstp->rq_bvec,
- ARRAY_SIZE(rqstp->rq_bvec), xdr);
+ count = xdr_buf_to_bvec(rqstp->rq_bvec, rqstp->rq_maxpages, xdr);
iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, rqstp->rq_bvec,
count, rqstp->rq_res.len);
@@ -1219,8 +1218,8 @@ static int svc_tcp_sendmsg(struct svc_sock *svsk, struct svc_rqst *rqstp,
memcpy(buf, &marker, sizeof(marker));
bvec_set_virt(rqstp->rq_bvec, buf, sizeof(marker));
- count = xdr_buf_to_bvec(rqstp->rq_bvec + 1,
- ARRAY_SIZE(rqstp->rq_bvec) - 1, &rqstp->rq_res);
+ count = xdr_buf_to_bvec(rqstp->rq_bvec + 1, rqstp->rq_maxpages,
+ &rqstp->rq_res);
iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, rqstp->rq_bvec,
1 + count, sizeof(marker) + rqstp->rq_res.len);
@@ -1340,7 +1339,8 @@ static void svc_tcp_init(struct svc_sock *svsk, struct svc_serv *serv)
svsk->sk_marker = xdr_zero;
svsk->sk_tcplen = 0;
svsk->sk_datalen = 0;
- memset(&svsk->sk_pages[0], 0, sizeof(svsk->sk_pages));
+ memset(&svsk->sk_pages[0], 0,
+ svsk->sk_maxpages * sizeof(struct page *));
tcp_sock_set_nodelay(sk);
@@ -1379,10 +1379,13 @@ 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);
+ unsigned long pages;
- svsk = kzalloc(sizeof(*svsk), GFP_KERNEL);
+ pages = svc_serv_maxpages(serv);
+ svsk = kzalloc(struct_size(svsk, sk_pages, pages), GFP_KERNEL);
if (!svsk)
return ERR_PTR(-ENOMEM);
+ svsk->sk_maxpages = pages;
inet = sock->sk;
@@ -1542,7 +1545,7 @@ static struct svc_xprt *svc_create_socket(struct svc_serv *serv,
if (protocol == IPPROTO_TCP) {
sk_net_refcnt_upgrade(sock->sk);
- if ((error = kernel_listen(sock, 64)) < 0)
+ if ((error = kernel_listen(sock, SOMAXCONN)) < 0)
goto bummer;
}
diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c
index 4e003cb516fe..2ea00e354ba6 100644
--- a/net/sunrpc/xdr.c
+++ b/net/sunrpc/xdr.c
@@ -213,6 +213,7 @@ bvec_overflow:
pr_warn_once("%s: bio_vec array overflow\n", __func__);
return count - 1;
}
+EXPORT_SYMBOL_GPL(xdr_buf_to_bvec);
/**
* xdr_inline_pages - Prepare receive buffer for a large reply
diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
index 0eab15465511..d5e0cdcad9e0 100644
--- a/net/sunrpc/xprt.c
+++ b/net/sunrpc/xprt.c
@@ -1365,7 +1365,7 @@ xprt_request_enqueue_transmit(struct rpc_task *task)
INIT_LIST_HEAD(&req->rq_xmit2);
goto out;
}
- } else if (!req->rq_seqno) {
+ } else if (req->rq_seqno_count == 0) {
list_for_each_entry(pos, &xprt->xmit_queue, rq_xmit) {
if (pos->rq_task->tk_owner != task->tk_owner)
continue;
@@ -1898,6 +1898,7 @@ xprt_request_init(struct rpc_task *task)
req->rq_snd_buf.bvec = NULL;
req->rq_rcv_buf.bvec = NULL;
req->rq_release_snd_buf = NULL;
+ req->rq_seqno_count = 0;
xprt_init_majortimeo(task, req, task->tk_client->cl_timeout);
trace_xprt_reserve(req);
diff --git a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
index 292022f0976e..e7e4a39ca6c6 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
@@ -120,12 +120,16 @@ svc_rdma_recv_ctxt_alloc(struct svcxprt_rdma *rdma)
{
int node = ibdev_to_node(rdma->sc_cm_id->device);
struct svc_rdma_recv_ctxt *ctxt;
+ unsigned long pages;
dma_addr_t addr;
void *buffer;
- ctxt = kzalloc_node(sizeof(*ctxt), GFP_KERNEL, node);
+ pages = svc_serv_maxpages(rdma->sc_xprt.xpt_server);
+ ctxt = kzalloc_node(struct_size(ctxt, rc_pages, pages),
+ GFP_KERNEL, node);
if (!ctxt)
goto fail0;
+ ctxt->rc_maxpages = pages;
buffer = kmalloc_node(rdma->sc_max_req_size, GFP_KERNEL, node);
if (!buffer)
goto fail1;
@@ -497,7 +501,7 @@ static bool xdr_check_write_chunk(struct svc_rdma_recv_ctxt *rctxt)
* a computation, perform a simple range check. This is an
* arbitrary but sensible limit (ie, not architectural).
*/
- if (unlikely(segcount > RPCSVC_MAXPAGES))
+ if (unlikely(segcount > rctxt->rc_maxpages))
return false;
p = xdr_inline_decode(&rctxt->rc_stream,
diff --git a/net/sunrpc/xprtrdma/svc_rdma_rw.c b/net/sunrpc/xprtrdma/svc_rdma_rw.c
index 40797114d50a..661b3fe2779f 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_rw.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_rw.c
@@ -765,7 +765,7 @@ static int svc_rdma_build_read_segment(struct svc_rqst *rqstp,
}
len -= seg_len;
- if (len && ((head->rc_curpage + 1) > ARRAY_SIZE(rqstp->rq_pages)))
+ if (len && ((head->rc_curpage + 1) > rqstp->rq_maxpages))
goto out_overrun;
}
diff --git a/net/sunrpc/xprtrdma/svc_rdma_sendto.c b/net/sunrpc/xprtrdma/svc_rdma_sendto.c
index 96154a2367a1..914cd263c2f1 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_sendto.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_sendto.c
@@ -118,6 +118,7 @@ svc_rdma_send_ctxt_alloc(struct svcxprt_rdma *rdma)
{
int node = ibdev_to_node(rdma->sc_cm_id->device);
struct svc_rdma_send_ctxt *ctxt;
+ unsigned long pages;
dma_addr_t addr;
void *buffer;
int i;
@@ -126,13 +127,19 @@ svc_rdma_send_ctxt_alloc(struct svcxprt_rdma *rdma)
GFP_KERNEL, node);
if (!ctxt)
goto fail0;
+ pages = svc_serv_maxpages(rdma->sc_xprt.xpt_server);
+ ctxt->sc_pages = kcalloc_node(pages, sizeof(struct page *),
+ GFP_KERNEL, node);
+ if (!ctxt->sc_pages)
+ goto fail1;
+ ctxt->sc_maxpages = pages;
buffer = kmalloc_node(rdma->sc_max_req_size, GFP_KERNEL, node);
if (!buffer)
- goto fail1;
+ goto fail2;
addr = ib_dma_map_single(rdma->sc_pd->device, buffer,
rdma->sc_max_req_size, DMA_TO_DEVICE);
if (ib_dma_mapping_error(rdma->sc_pd->device, addr))
- goto fail2;
+ goto fail3;
svc_rdma_send_cid_init(rdma, &ctxt->sc_cid);
@@ -151,8 +158,10 @@ svc_rdma_send_ctxt_alloc(struct svcxprt_rdma *rdma)
ctxt->sc_sges[i].lkey = rdma->sc_pd->local_dma_lkey;
return ctxt;
-fail2:
+fail3:
kfree(buffer);
+fail2:
+ kfree(ctxt->sc_pages);
fail1:
kfree(ctxt);
fail0:
@@ -176,6 +185,7 @@ void svc_rdma_send_ctxts_destroy(struct svcxprt_rdma *rdma)
rdma->sc_max_req_size,
DMA_TO_DEVICE);
kfree(ctxt->sc_xprt_buf);
+ kfree(ctxt->sc_pages);
kfree(ctxt);
}
}
diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c
index aca8bdf65d72..3d7f1413df02 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_transport.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c
@@ -406,12 +406,12 @@ static void svc_rdma_xprt_done(struct rpcrdma_notification *rn)
*/
static struct svc_xprt *svc_rdma_accept(struct svc_xprt *xprt)
{
+ unsigned int ctxts, rq_depth, maxpayload;
struct svcxprt_rdma *listen_rdma;
struct svcxprt_rdma *newxprt = NULL;
struct rdma_conn_param conn_param;
struct rpcrdma_connect_private pmsg;
struct ib_qp_init_attr qp_attr;
- unsigned int ctxts, rq_depth;
struct ib_device *dev;
int ret = 0;
RPC_IFDEBUG(struct sockaddr *sap);
@@ -462,12 +462,14 @@ static struct svc_xprt *svc_rdma_accept(struct svc_xprt *xprt)
newxprt->sc_max_bc_requests = 2;
}
- /* Arbitrarily estimate the number of rw_ctxs needed for
- * this transport. This is enough rw_ctxs to make forward
- * progress even if the client is using one rkey per page
- * in each Read chunk.
+ /* Arbitrary estimate of the needed number of rdma_rw contexts.
*/
- ctxts = 3 * RPCSVC_MAXPAGES;
+ maxpayload = min(xprt->xpt_server->sv_max_payload,
+ RPCSVC_MAXPAYLOAD_RDMA);
+ ctxts = newxprt->sc_max_requests * 3 *
+ rdma_rw_mr_factor(dev, newxprt->sc_port_num,
+ maxpayload >> PAGE_SHIFT);
+
newxprt->sc_sq_depth = rq_depth + ctxts;
if (newxprt->sc_sq_depth > dev->attrs.max_qp_wr)
newxprt->sc_sq_depth = dev->attrs.max_qp_wr;
@@ -575,6 +577,7 @@ static struct svc_xprt *svc_rdma_accept(struct svc_xprt *xprt)
if (newxprt->sc_qp && !IS_ERR(newxprt->sc_qp))
ib_destroy_qp(newxprt->sc_qp);
rdma_destroy_id(newxprt->sc_cm_id);
+ rpcrdma_rn_unregister(dev, &newxprt->sc_rn);
/* This call to put will destroy the transport */
svc_xprt_put(&newxprt->sc_xprt);
return NULL;
diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c
index 83cc095846d3..04ff66758fc3 100644
--- a/net/sunrpc/xprtsock.c
+++ b/net/sunrpc/xprtsock.c
@@ -2726,20 +2726,14 @@ static void xs_tcp_tls_setup_socket(struct work_struct *work)
if (status)
goto out_close;
xprt_release_write(lower_xprt, NULL);
-
trace_rpc_socket_connect(upper_xprt, upper_transport->sock, 0);
- if (!xprt_test_and_set_connected(upper_xprt)) {
- upper_xprt->connect_cookie++;
- clear_bit(XPRT_SOCK_CONNECTING, &upper_transport->sock_state);
- xprt_clear_connecting(upper_xprt);
-
- upper_xprt->stat.connect_count++;
- upper_xprt->stat.connect_time += (long)jiffies -
- upper_xprt->stat.connect_start;
- xs_run_error_worker(upper_transport, XPRT_SOCK_WAKE_PENDING);
- }
rpc_shutdown_client(lower_clnt);
+ /* Check for ingress data that arrived before the socket's
+ * ->data_ready callback was set up.
+ */
+ xs_poll_check_readable(upper_transport);
+
out_unlock:
current_restore_flags(pflags, PF_MEMALLOC);
upper_transport->clnt = NULL;
generated by cgit (git 2.34.1) at 2025-06-05 12:46:13 +0000