CFLAGS_nfstrace.o += -I$(src)
nfs-y := client.o dir.o file.o getroot.o inode.o super.o \
io.o direct.o pagelist.o read.o symlink.o unlink.o \
- write.o namespace.o mount_clnt.o nfstrace.o export.o
+ write.o namespace.o mount_clnt.o nfstrace.o \
+ export.o sysfs.o
nfs-$(CONFIG_ROOT_NFS) += nfsroot.o
nfs-$(CONFIG_SYSCTL) += sysctl.o
nfs-$(CONFIG_NFS_FSCACHE) += fscache.o fscache-index.o
validate_seqid(const struct nfs4_slot_table *tbl, const struct nfs4_slot *slot,
const struct cb_sequenceargs * args)
{
+ __be32 ret;
+
+ ret = cpu_to_be32(NFS4ERR_BADSLOT);
if (args->csa_slotid > tbl->server_highest_slotid)
- return htonl(NFS4ERR_BADSLOT);
+ goto out_err;
/* Replay */
if (args->csa_sequenceid == slot->seq_nr) {
+ ret = cpu_to_be32(NFS4ERR_DELAY);
if (nfs4_test_locked_slot(tbl, slot->slot_nr))
- return htonl(NFS4ERR_DELAY);
+ goto out_err;
+
/* Signal process_op to set this error on next op */
+ ret = cpu_to_be32(NFS4ERR_RETRY_UNCACHED_REP);
if (args->csa_cachethis == 0)
- return htonl(NFS4ERR_RETRY_UNCACHED_REP);
+ goto out_err;
/* Liar! We never allowed you to set csa_cachethis != 0 */
- return htonl(NFS4ERR_SEQ_FALSE_RETRY);
+ ret = cpu_to_be32(NFS4ERR_SEQ_FALSE_RETRY);
+ goto out_err;
}
/* Note: wraparound relies on seq_nr being of type u32 */
- if (likely(args->csa_sequenceid == slot->seq_nr + 1))
- return htonl(NFS4_OK);
-
/* Misordered request */
- return htonl(NFS4ERR_SEQ_MISORDERED);
+ ret = cpu_to_be32(NFS4ERR_SEQ_MISORDERED);
+ if (args->csa_sequenceid != slot->seq_nr + 1)
+ goto out_err;
+
+ return cpu_to_be32(NFS4_OK);
+
+out_err:
+ trace_nfs4_cb_seqid_err(args, ret);
+ return ret;
}
/*
#include "pnfs.h"
#include "nfs.h"
#include "netns.h"
+#include "sysfs.h"
#define NFSDBG_FACILITY NFSDBG_CLIENT
clp->cl_rpcclient = ERR_PTR(-EINVAL);
clp->cl_proto = cl_init->proto;
+ clp->cl_nconnect = cl_init->nconnect;
clp->cl_net = get_net(cl_init->net);
clp->cl_principal = "*";
EXPORT_SYMBOL_GPL(nfs_alloc_client);
#if IS_ENABLED(CONFIG_NFS_V4)
-void nfs_cleanup_cb_ident_idr(struct net *net)
+static void nfs_cleanup_cb_ident_idr(struct net *net)
{
struct nfs_net *nn = net_generic(net, nfs_net_id);
}
#else
-void nfs_cleanup_cb_ident_idr(struct net *net)
+static void nfs_cleanup_cb_ident_idr(struct net *net)
{
}
clp = nfs_match_client(cl_init);
if (clp) {
spin_unlock(&nn->nfs_client_lock);
- if (IS_ERR(clp))
- return clp;
if (new)
new->rpc_ops->free_client(new);
+ if (IS_ERR(clp))
+ return clp;
return nfs_found_client(cl_init, clp);
}
if (new) {
struct rpc_create_args args = {
.net = clp->cl_net,
.protocol = clp->cl_proto,
+ .nconnect = clp->cl_nconnect,
.address = (struct sockaddr *)&clp->cl_addr,
.addrsize = clp->cl_addrlen,
.timeout = cl_init->timeparms,
.net = data->net,
.timeparms = &timeparms,
.cred = server->cred,
+ .nconnect = data->nfs_server.nconnect,
};
struct nfs_client *clp;
int error;
#endif
spin_lock_init(&nn->nfs_client_lock);
nn->boot_time = ktime_get_real();
+
+ nfs_netns_sysfs_setup(nn, net);
+}
+
+void nfs_clients_exit(struct net *net)
+{
+ struct nfs_net *nn = net_generic(net, nfs_net_id);
+
+ nfs_netns_sysfs_destroy(nn);
+ nfs_cleanup_cb_ident_idr(net);
+ WARN_ON_ONCE(!list_empty(&nn->nfs_client_list));
+ WARN_ON_ONCE(!list_empty(&nn->nfs_volume_list));
}
#ifdef CONFIG_PROC_FS
ctx->dup_cookie = 0;
ctx->cred = get_cred(cred);
spin_lock(&dir->i_lock);
+ if (list_empty(&nfsi->open_files) &&
+ (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER))
+ nfsi->cache_validity |= NFS_INO_INVALID_DATA |
+ NFS_INO_REVAL_FORCED;
list_add(&ctx->list, &nfsi->open_files);
spin_unlock(&dir->i_lock);
return ctx;
struct nfs_cache_array_entry array[0];
};
-struct readdirvec {
- unsigned long nr;
- unsigned long index;
- struct page *pages[NFS_MAX_READDIR_RAPAGES];
-};
-
typedef int (*decode_dirent_t)(struct xdr_stream *, struct nfs_entry *, bool);
typedef struct {
struct file *file;
struct page *page;
struct dir_context *ctx;
unsigned long page_index;
- struct readdirvec pvec;
u64 *dir_cookie;
u64 last_cookie;
loff_t current_index;
struct nfs_cache_array *array;
unsigned int count = 0;
int status;
- int max_rapages = NFS_MAX_READDIR_RAPAGES;
-
- desc->pvec.index = desc->page_index;
- desc->pvec.nr = 0;
scratch = alloc_page(GFP_KERNEL);
if (scratch == NULL)
if (desc->plus)
nfs_prime_dcache(file_dentry(desc->file), entry);
- status = nfs_readdir_add_to_array(entry, desc->pvec.pages[desc->pvec.nr]);
- if (status == -ENOSPC) {
- desc->pvec.nr++;
- if (desc->pvec.nr == max_rapages)
- break;
- status = nfs_readdir_add_to_array(entry, desc->pvec.pages[desc->pvec.nr]);
- }
+ status = nfs_readdir_add_to_array(entry, page);
if (status != 0)
break;
} while (!entry->eof);
- /*
- * page and desc->pvec.pages[0] are valid, don't need to check
- * whether or not to be NULL.
- */
- copy_highpage(page, desc->pvec.pages[0]);
-
out_nopages:
if (count == 0 || (status == -EBADCOOKIE && entry->eof != 0)) {
- array = kmap_atomic(desc->pvec.pages[desc->pvec.nr]);
+ array = kmap(page);
array->eof_index = array->size;
status = 0;
- kunmap_atomic(array);
+ kunmap(page);
}
put_page(scratch);
-
- /*
- * desc->pvec.nr > 0 means at least one page was completely filled,
- * we should return -ENOSPC. Otherwise function
- * nfs_readdir_xdr_to_array will enter infinite loop.
- */
- if (desc->pvec.nr > 0)
- return -ENOSPC;
return status;
}
return -ENOMEM;
}
-/*
- * nfs_readdir_rapages_init initialize rapages by nfs_cache_array structure.
- */
-static
-void nfs_readdir_rapages_init(nfs_readdir_descriptor_t *desc)
-{
- struct nfs_cache_array *array;
- int max_rapages = NFS_MAX_READDIR_RAPAGES;
- int index;
-
- for (index = 0; index < max_rapages; index++) {
- array = kmap_atomic(desc->pvec.pages[index]);
- memset(array, 0, sizeof(struct nfs_cache_array));
- array->eof_index = -1;
- kunmap_atomic(array);
- }
-}
-
static
int nfs_readdir_xdr_to_array(nfs_readdir_descriptor_t *desc, struct page *page, struct inode *inode)
{
int status = -ENOMEM;
unsigned int array_size = ARRAY_SIZE(pages);
- /*
- * This means we hit readdir rdpages miss, the preallocated rdpages
- * are useless, the preallocate rdpages should be reinitialized.
- */
- nfs_readdir_rapages_init(desc);
-
entry.prev_cookie = 0;
entry.cookie = desc->last_cookie;
entry.eof = 0;
struct inode *inode = file_inode(desc->file);
int ret;
- /*
- * If desc->page_index in range desc->pvec.index and
- * desc->pvec.index + desc->pvec.nr, we get readdir cache hit.
- */
- if (desc->page_index >= desc->pvec.index &&
- desc->page_index < (desc->pvec.index + desc->pvec.nr)) {
- /*
- * page and desc->pvec.pages[x] are valid, don't need to check
- * whether or not to be NULL.
- */
- copy_highpage(page, desc->pvec.pages[desc->page_index - desc->pvec.index]);
- ret = 0;
- } else {
- ret = nfs_readdir_xdr_to_array(desc, page, inode);
- if (ret < 0)
- goto error;
- }
-
+ ret = nfs_readdir_xdr_to_array(desc, page, inode);
+ if (ret < 0)
+ goto error;
SetPageUptodate(page);
if (invalidate_inode_pages2_range(inode->i_mapping, page->index + 1, -1) < 0) {
*desc = &my_desc;
struct nfs_open_dir_context *dir_ctx = file->private_data;
int res = 0;
- int max_rapages = NFS_MAX_READDIR_RAPAGES;
dfprintk(FILE, "NFS: readdir(%pD2) starting at cookie %llu\n",
file, (long long)ctx->pos);
desc->decode = NFS_PROTO(inode)->decode_dirent;
desc->plus = nfs_use_readdirplus(inode, ctx);
- res = nfs_readdir_alloc_pages(desc->pvec.pages, max_rapages);
- if (res < 0)
- return -ENOMEM;
-
- nfs_readdir_rapages_init(desc);
-
if (ctx->pos == 0 || nfs_attribute_cache_expired(inode))
res = nfs_revalidate_mapping(inode, file->f_mapping);
if (res < 0)
break;
} while (!desc->eof);
out:
- nfs_readdir_free_pages(desc->pvec.pages, max_rapages);
if (res > 0)
res = 0;
dfprintk(FILE, "NFS: readdir(%pD2) returns %d\n", file, res);
if (pgio->pg_error < 0)
return;
out_mds:
+ trace_pnfs_mds_fallback_pg_init_read(pgio->pg_inode,
+ 0, NFS4_MAX_UINT64, IOMODE_READ,
+ NFS_I(pgio->pg_inode)->layout,
+ pgio->pg_lseg);
pnfs_put_lseg(pgio->pg_lseg);
pgio->pg_lseg = NULL;
nfs_pageio_reset_read_mds(pgio);
return;
out_mds:
+ trace_pnfs_mds_fallback_pg_init_write(pgio->pg_inode,
+ 0, NFS4_MAX_UINT64, IOMODE_RW,
+ NFS_I(pgio->pg_inode)->layout,
+ pgio->pg_lseg);
pnfs_put_lseg(pgio->pg_lseg);
pgio->pg_lseg = NULL;
nfs_pageio_reset_write_mds(pgio);
if (pgio->pg_lseg)
return FF_LAYOUT_MIRROR_COUNT(pgio->pg_lseg);
+ trace_pnfs_mds_fallback_pg_get_mirror_count(pgio->pg_inode,
+ 0, NFS4_MAX_UINT64, IOMODE_RW,
+ NFS_I(pgio->pg_inode)->layout,
+ pgio->pg_lseg);
/* no lseg means that pnfs is not in use, so no mirroring here */
nfs_pageio_reset_write_mds(pgio);
out:
hdr->args.count,
(unsigned long long)hdr->args.offset);
+ trace_pnfs_mds_fallback_write_done(hdr->inode,
+ hdr->args.offset, hdr->args.count,
+ IOMODE_RW, NFS_I(hdr->inode)->layout,
+ hdr->lseg);
task->tk_status = pnfs_write_done_resend_to_mds(hdr);
}
}
hdr->args.count,
(unsigned long long)hdr->args.offset);
+ trace_pnfs_mds_fallback_read_done(hdr->inode,
+ hdr->args.offset, hdr->args.count,
+ IOMODE_READ, NFS_I(hdr->inode)->layout,
+ hdr->lseg);
task->tk_status = pnfs_read_done_resend_to_mds(hdr);
}
}
out_failed:
if (ff_layout_avoid_mds_available_ds(lseg))
return PNFS_TRY_AGAIN;
+ trace_pnfs_mds_fallback_read_pagelist(hdr->inode,
+ hdr->args.offset, hdr->args.count,
+ IOMODE_READ, NFS_I(hdr->inode)->layout, lseg);
return PNFS_NOT_ATTEMPTED;
}
out_failed:
if (ff_layout_avoid_mds_available_ds(lseg))
return PNFS_TRY_AGAIN;
+ trace_pnfs_mds_fallback_write_pagelist(hdr->inode,
+ hdr->args.offset, hdr->args.count,
+ IOMODE_RW, NFS_I(hdr->inode)->layout, lseg);
return PNFS_NOT_ATTEMPTED;
}
if (status == 0)
return 0;
- if (mirror->mirror_ds == NULL)
+ if (IS_ERR_OR_NULL(mirror->mirror_ds))
return -EINVAL;
dserr = kmalloc(sizeof(*dserr), gfp_flags);
#include "pnfs.h"
#include "nfs.h"
#include "netns.h"
+#include "sysfs.h"
#include "nfstrace.h"
}
if (inode->i_mapping->nrpages == 0)
- flags &= ~NFS_INO_INVALID_DATA;
+ flags &= ~(NFS_INO_INVALID_DATA|NFS_INO_DATA_INVAL_DEFER);
nfsi->cache_validity |= flags;
if (flags & NFS_INO_INVALID_DATA)
nfs_fscache_invalidate(inode);
i_size_write(inode, offset);
/* Optimisation */
if (offset == 0)
- NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_DATA;
+ NFS_I(inode)->cache_validity &= ~(NFS_INO_INVALID_DATA |
+ NFS_INO_DATA_INVAL_DEFER);
NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_SIZE;
spin_unlock(&inode->i_lock);
struct nfs_inode *nfsi = NFS_I(inode);
spin_lock(&inode->i_lock);
+ if (list_empty(&nfsi->open_files) &&
+ (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER))
+ nfsi->cache_validity |= NFS_INO_INVALID_DATA |
+ NFS_INO_REVAL_FORCED;
list_add_tail_rcu(&ctx->list, &nfsi->open_files);
spin_unlock(&inode->i_lock);
}
nfs_fscache_open_file(inode, filp);
return 0;
}
+EXPORT_SYMBOL_GPL(nfs_open);
/*
* This function is called whenever some part of NFS notices that
set_bit(NFS_INO_INVALIDATING, bitlock);
smp_wmb();
- nfsi->cache_validity &= ~NFS_INO_INVALID_DATA;
+ nfsi->cache_validity &= ~(NFS_INO_INVALID_DATA|
+ NFS_INO_DATA_INVAL_DEFER);
spin_unlock(&inode->i_lock);
trace_nfs_invalidate_mapping_enter(inode);
ret = nfs_invalidate_mapping(inode, mapping);
dprintk("NFS: change_attr change on server for file %s/%ld\n",
inode->i_sb->s_id,
inode->i_ino);
- }
+ } else if (!have_delegation)
+ nfsi->cache_validity |= NFS_INO_DATA_INVAL_DEFER;
inode_set_iversion_raw(inode, fattr->change_attr);
attr_changed = true;
}
static void nfs_net_exit(struct net *net)
{
- struct nfs_net *nn = net_generic(net, nfs_net_id);
-
nfs_fs_proc_net_exit(net);
- nfs_cleanup_cb_ident_idr(net);
- WARN_ON_ONCE(!list_empty(&nn->nfs_client_list));
- WARN_ON_ONCE(!list_empty(&nn->nfs_volume_list));
+ nfs_clients_exit(net);
}
static struct pernet_operations nfs_net_ops = {
{
int err;
+ err = nfs_sysfs_init();
+ if (err < 0)
+ goto out10;
+
err = register_pernet_subsys(&nfs_net_ops);
if (err < 0)
goto out9;
out8:
unregister_pernet_subsys(&nfs_net_ops);
out9:
+ nfs_sysfs_exit();
+out10:
return err;
}
unregister_nfs_fs();
nfs_fs_proc_exit();
nfsiod_stop();
+ nfs_sysfs_exit();
}
/* Not quite true; I just maintain it */
* Maximum number of pages that readdir can use for creating
* a vmapped array of pages.
*/
-#define NFS_MAX_READDIR_PAGES 64
-#define NFS_MAX_READDIR_RAPAGES 8
+#define NFS_MAX_READDIR_PAGES 8
struct nfs_client_initdata {
unsigned long init_flags;
struct nfs_subversion *nfs_mod;
int proto;
u32 minorversion;
+ unsigned int nconnect;
struct net *net;
const struct rpc_timeout *timeparms;
const struct cred *cred;
char *export_path;
int port;
unsigned short protocol;
+ unsigned short nconnect;
} nfs_server;
void *lsm_opts;
/* client.c */
extern const struct rpc_program nfs_program;
extern void nfs_clients_init(struct net *net);
+extern void nfs_clients_exit(struct net *net);
extern struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *);
int nfs_create_rpc_client(struct nfs_client *, const struct nfs_client_initdata *, rpc_authflavor_t);
struct nfs_client *nfs_get_client(const struct nfs_client_initdata *);
struct nfs_server *nfs_alloc_server(void);
void nfs_server_copy_userdata(struct nfs_server *, struct nfs_server *);
-extern void nfs_cleanup_cb_ident_idr(struct net *);
extern void nfs_put_client(struct nfs_client *);
extern void nfs_free_client(struct nfs_client *);
extern struct nfs_client *nfs4_find_client_ident(struct net *, int);
uint32_t major, minor;
};
+struct nfs_netns_client;
+
struct nfs_net {
struct cache_detail *nfs_dns_resolve;
struct rpc_pipe *bl_device_pipe;
unsigned short nfs_callback_tcpport6;
int cb_users[NFS4_MAX_MINOR_VERSION + 1];
#endif
+ struct nfs_netns_client *nfs_client;
spinlock_t nfs_client_lock;
ktime_t boot_time;
#ifdef CONFIG_PROC_FS
return 0;
out_status:
*status = be32_to_cpup(p);
- trace_nfs_xdr_status((int)*status);
+ trace_nfs_xdr_status(xdr, (int)*status);
return 0;
}
return ERR_PTR(-EINVAL);
cl_init.hostname = buf;
+ if (mds_clp->cl_nconnect > 1 && ds_proto == XPRT_TRANSPORT_TCP)
+ cl_init.nconnect = mds_clp->cl_nconnect;
+
if (mds_srv->flags & NFS_MOUNT_NORESVPORT)
set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags);
return 0;
out_status:
*status = be32_to_cpup(p);
- trace_nfs_xdr_status((int)*status);
+ trace_nfs_xdr_status(xdr, (int)*status);
return 0;
}
const struct nfs_lock_context *l_ctx,
fmode_t fmode);
+extern int nfs4_proc_get_lease_time(struct nfs_client *clp,
+ struct nfs_fsinfo *fsinfo);
#if defined(CONFIG_NFS_V4_1)
extern int nfs41_sequence_done(struct rpc_task *, struct nfs4_sequence_res *);
extern int nfs4_proc_create_session(struct nfs_client *, const struct cred *);
extern int nfs4_proc_destroy_session(struct nfs4_session *, const struct cred *);
-extern int nfs4_proc_get_lease_time(struct nfs_client *clp,
- struct nfs_fsinfo *fsinfo);
extern int nfs4_proc_layoutcommit(struct nfs4_layoutcommit_data *data,
bool sync);
extern int nfs4_detect_session_trunking(struct nfs_client *clp,
const size_t addrlen,
const char *ip_addr,
int proto, const struct rpc_timeout *timeparms,
- u32 minorversion, struct net *net)
+ u32 minorversion, unsigned int nconnect,
+ struct net *net)
{
struct nfs_client_initdata cl_init = {
.hostname = hostname,
};
struct nfs_client *clp;
+ if (minorversion > 0 && proto == XPRT_TRANSPORT_TCP)
+ cl_init.nconnect = nconnect;
if (server->flags & NFS_MOUNT_NORESVPORT)
set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags);
if (server->options & NFS_OPTION_MIGRATION)
return ERR_PTR(-EINVAL);
cl_init.hostname = buf;
+ if (mds_clp->cl_nconnect > 1 && ds_proto == XPRT_TRANSPORT_TCP)
+ cl_init.nconnect = mds_clp->cl_nconnect;
+
if (mds_srv->flags & NFS_MOUNT_NORESVPORT)
__set_bit(NFS_CS_NORESVPORT, &cl_init.init_flags);
data->nfs_server.protocol,
&timeparms,
data->minorversion,
+ data->nfs_server.nconnect,
data->net);
if (error < 0)
return error;
XPRT_TRANSPORT_RDMA,
parent_server->client->cl_timeout,
parent_client->cl_mvops->minor_version,
+ parent_client->cl_nconnect,
parent_client->cl_net);
if (!error)
goto init_server;
XPRT_TRANSPORT_TCP,
parent_server->client->cl_timeout,
parent_client->cl_mvops->minor_version,
+ parent_client->cl_nconnect,
parent_client->cl_net);
if (error < 0)
goto error;
set_bit(NFS_MIG_TSM_POSSIBLE, &server->mig_status);
error = nfs4_set_client(server, hostname, sap, salen, buf,
clp->cl_proto, clnt->cl_timeout,
- clp->cl_minorversion, net);
+ clp->cl_minorversion,
+ clp->cl_nconnect, net);
clear_bit(NFS_MIG_TSM_POSSIBLE, &server->mig_status);
if (error != 0) {
nfs_server_insert_lists(server);
return err;
if ((openflags & O_ACCMODE) == 3)
- openflags--;
+ return nfs_open(inode, filp);
/* We can't create new files here */
openflags &= ~(O_CREAT|O_EXCL);
bool same_inode = false;
int ret;
- if (remap_flags & ~(REMAP_FILE_DEDUP | REMAP_FILE_ADVISORY))
+ /* NFS does not support deduplication. */
+ if (remap_flags & REMAP_FILE_DEDUP)
+ return -EOPNOTSUPP;
+
+ if (remap_flags & ~REMAP_FILE_ADVISORY)
return -EINVAL;
/* check alignment w.r.t. clone_blksize */
return nfs4_delay_killable(timeout);
}
+static const nfs4_stateid *
+nfs4_recoverable_stateid(const nfs4_stateid *stateid)
+{
+ if (!stateid)
+ return NULL;
+ switch (stateid->type) {
+ case NFS4_OPEN_STATEID_TYPE:
+ case NFS4_LOCK_STATEID_TYPE:
+ case NFS4_DELEGATION_STATEID_TYPE:
+ return stateid;
+ default:
+ break;
+ }
+ return NULL;
+}
+
/* This is the error handling routine for processes that are allowed
* to sleep.
*/
{
struct nfs_client *clp = server->nfs_client;
struct nfs4_state *state = exception->state;
- const nfs4_stateid *stateid = exception->stateid;
+ const nfs4_stateid *stateid;
struct inode *inode = exception->inode;
int ret = errorcode;
exception->recovering = 0;
exception->retry = 0;
+ stateid = nfs4_recoverable_stateid(exception->stateid);
if (stateid == NULL && state != NULL)
- stateid = &state->stateid;
+ stateid = nfs4_recoverable_stateid(&state->stateid);
switch(errorcode) {
case 0:
return true;
}
+static fmode_t _nfs4_ctx_to_accessmode(const struct nfs_open_context *ctx)
+{
+ return ctx->mode & (FMODE_READ|FMODE_WRITE|FMODE_EXEC);
+}
+
+static fmode_t _nfs4_ctx_to_openmode(const struct nfs_open_context *ctx)
+{
+ fmode_t ret = ctx->mode & (FMODE_READ|FMODE_WRITE);
+
+ return (ctx->mode & FMODE_EXEC) ? FMODE_READ | ret : ret;
+}
+
static u32
nfs4_map_atomic_open_share(struct nfs_server *server,
fmode_t fmode, int openflags)
}
static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata,
- fmode_t fmode,
- int flags,
- struct nfs_open_context *ctx)
+ int flags, struct nfs_open_context *ctx)
{
struct nfs4_state_owner *sp = opendata->owner;
struct nfs_server *server = sp->so_server;
struct dentry *dentry;
struct nfs4_state *state;
+ fmode_t acc_mode = _nfs4_ctx_to_accessmode(ctx);
unsigned int seq;
int ret;
/* Parse layoutget results before we check for access */
pnfs_parse_lgopen(state->inode, opendata->lgp, ctx);
- ret = nfs4_opendata_access(sp->so_cred, opendata, state, fmode, flags);
+ ret = nfs4_opendata_access(sp->so_cred, opendata, state,
+ acc_mode, flags);
if (ret != 0)
goto out;
struct dentry *dentry = ctx->dentry;
const struct cred *cred = ctx->cred;
struct nfs4_threshold **ctx_th = &ctx->mdsthreshold;
- fmode_t fmode = ctx->mode & (FMODE_READ|FMODE_WRITE|FMODE_EXEC);
+ fmode_t fmode = _nfs4_ctx_to_openmode(ctx);
enum open_claim_type4 claim = NFS4_OPEN_CLAIM_NULL;
struct iattr *sattr = c->sattr;
struct nfs4_label *label = c->label;
if (d_really_is_positive(dentry))
opendata->state = nfs4_get_open_state(d_inode(dentry), sp);
- status = _nfs4_open_and_get_state(opendata, fmode, flags, ctx);
+ status = _nfs4_open_and_get_state(opendata, flags, ctx);
if (status != 0)
goto err_free_label;
state = ctx->state;
if (ctx->state == NULL)
return;
if (is_sync)
- nfs4_close_sync(ctx->state, ctx->mode);
+ nfs4_close_sync(ctx->state, _nfs4_ctx_to_openmode(ctx));
else
- nfs4_close_state(ctx->state, ctx->mode);
+ nfs4_close_state(ctx->state, _nfs4_ctx_to_openmode(ctx));
}
#define FATTR4_WORD1_NFS40_MASK (2*FATTR4_WORD1_MOUNTED_ON_FILEID - 1UL)
.rpc_message = &msg,
.callback_ops = &nfs4_setclientid_ops,
.callback_data = &setclientid,
- .flags = RPC_TASK_TIMEOUT,
+ .flags = RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN,
};
int status;
dprintk("NFS call setclientid_confirm auth=%s, (client ID %llx)\n",
clp->cl_rpcclient->cl_auth->au_ops->au_name,
clp->cl_clientid);
- status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
+ status = rpc_call_sync(clp->cl_rpcclient, &msg,
+ RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN);
trace_nfs4_setclientid_confirm(clp, status);
dprintk("NFS reply setclientid_confirm: %d\n", status);
return status;
NFS_SP4_MACH_CRED_SECINFO, &clnt, &msg);
status = nfs4_call_sync(clnt, NFS_SERVER(dir), &msg, &args.seq_args,
- &res.seq_res, 0);
+ &res.seq_res, RPC_TASK_NO_ROUND_ROBIN);
dprintk("NFS reply secinfo: %d\n", status);
put_cred(cred);
.rpc_client = clp->cl_rpcclient,
.callback_ops = &nfs4_exchange_id_call_ops,
.rpc_message = &msg,
- .flags = RPC_TASK_TIMEOUT,
+ .flags = RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN,
};
struct nfs41_exchange_id_data *calldata;
int status;
};
int status;
- status = rpc_call_sync(clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
+ status = rpc_call_sync(clp->cl_rpcclient, &msg,
+ RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN);
trace_nfs4_destroy_clientid(clp, status);
if (status)
dprintk("NFS: Got error %d from the server %s on "
return ret;
}
+#endif /* CONFIG_NFS_V4_1 */
+
struct nfs4_get_lease_time_data {
struct nfs4_get_lease_time_args *args;
struct nfs4_get_lease_time_res *res;
(struct nfs4_get_lease_time_data *)calldata;
dprintk("--> %s\n", __func__);
- if (!nfs41_sequence_done(task, &data->res->lr_seq_res))
+ if (!nfs4_sequence_done(task, &data->res->lr_seq_res))
return;
switch (task->tk_status) {
case -NFS4ERR_DELAY:
return status;
}
+#ifdef CONFIG_NFS_V4_1
+
/*
* Initialize the values to be used by the client in CREATE_SESSION
* If nfs4_init_session set the fore channel request and response sizes,
{
unsigned int max_rqst_sz, max_resp_sz;
unsigned int max_bc_payload = rpc_max_bc_payload(clnt);
+ unsigned int max_bc_slots = rpc_num_bc_slots(clnt);
max_rqst_sz = NFS_MAX_FILE_IO_SIZE + nfs41_maxwrite_overhead;
max_resp_sz = NFS_MAX_FILE_IO_SIZE + nfs41_maxread_overhead;
args->bc_attrs.max_resp_sz_cached = 0;
args->bc_attrs.max_ops = NFS4_MAX_BACK_CHANNEL_OPS;
args->bc_attrs.max_reqs = max_t(unsigned short, max_session_cb_slots, 1);
+ if (args->bc_attrs.max_reqs > max_bc_slots)
+ args->bc_attrs.max_reqs = max_bc_slots;
dprintk("%s: Back Channel : max_rqst_sz=%u max_resp_sz=%u "
"max_resp_sz_cached=%u max_ops=%u max_reqs=%u\n",
nfs4_init_channel_attrs(&args, clp->cl_rpcclient);
args.flags = (SESSION4_PERSIST | SESSION4_BACK_CHAN);
- status = rpc_call_sync(session->clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
+ status = rpc_call_sync(session->clp->cl_rpcclient, &msg,
+ RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN);
trace_nfs4_create_session(clp, status);
switch (status) {
if (!test_and_clear_bit(NFS4_SESSION_ESTABLISHED, &session->session_state))
return 0;
- status = rpc_call_sync(session->clp->cl_rpcclient, &msg, RPC_TASK_TIMEOUT);
+ status = rpc_call_sync(session->clp->cl_rpcclient, &msg,
+ RPC_TASK_TIMEOUT | RPC_TASK_NO_ROUND_ROBIN);
trace_nfs4_destroy_session(session->clp, status);
if (status)
.rpc_client = clp->cl_rpcclient,
.rpc_message = &msg,
.callback_ops = &nfs4_reclaim_complete_call_ops,
- .flags = RPC_TASK_ASYNC,
+ .flags = RPC_TASK_ASYNC | RPC_TASK_NO_ROUND_ROBIN,
};
int status = -ENOMEM;
dprintk("--> %s\n", __func__);
status = nfs4_call_sync(clnt, server, &msg, &args.seq_args,
- &res.seq_res, 0);
+ &res.seq_res, RPC_TASK_NO_ROUND_ROBIN);
dprintk("<-- %s status=%d\n", __func__, status);
put_cred(cred);
static DEFINE_MUTEX(nfs_clid_init_mutex);
+static int nfs4_setup_state_renewal(struct nfs_client *clp)
+{
+ int status;
+ struct nfs_fsinfo fsinfo;
+ unsigned long now;
+
+ if (!test_bit(NFS_CS_CHECK_LEASE_TIME, &clp->cl_res_state)) {
+ nfs4_schedule_state_renewal(clp);
+ return 0;
+ }
+
+ now = jiffies;
+ status = nfs4_proc_get_lease_time(clp, &fsinfo);
+ if (status == 0) {
+ nfs4_set_lease_period(clp, fsinfo.lease_time * HZ, now);
+ nfs4_schedule_state_renewal(clp);
+ }
+
+ return status;
+}
+
int nfs4_init_clientid(struct nfs_client *clp, const struct cred *cred)
{
struct nfs4_setclientid_res clid = {
if (status != 0)
goto out;
clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
- nfs4_schedule_state_renewal(clp);
+ nfs4_setup_state_renewal(clp);
out:
return status;
}
#if defined(CONFIG_NFS_V4_1)
-static int nfs41_setup_state_renewal(struct nfs_client *clp)
-{
- int status;
- struct nfs_fsinfo fsinfo;
- unsigned long now;
-
- if (!test_bit(NFS_CS_CHECK_LEASE_TIME, &clp->cl_res_state)) {
- nfs4_schedule_state_renewal(clp);
- return 0;
- }
-
- now = jiffies;
- status = nfs4_proc_get_lease_time(clp, &fsinfo);
- if (status == 0) {
- nfs4_set_lease_period(clp, fsinfo.lease_time * HZ, now);
- nfs4_schedule_state_renewal(clp);
- }
-
- return status;
-}
-
static void nfs41_finish_session_reset(struct nfs_client *clp)
{
clear_bit(NFS4CLNT_LEASE_CONFIRM, &clp->cl_state);
clear_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state);
/* create_session negotiated new slot table */
clear_bit(NFS4CLNT_BIND_CONN_TO_SESSION, &clp->cl_state);
- nfs41_setup_state_renewal(clp);
+ nfs4_setup_state_renewal(clp);
}
int nfs41_init_clientid(struct nfs_client *clp, const struct cred *cred)
* choose to use.
*/
goto out;
- nfs4_copy_open_stateid(dst, state);
- ret = 0;
+ ret = nfs4_copy_open_stateid(dst, state) ? 0 : -EAGAIN;
out:
if (nfs_server_capable(state->inode, NFS_CAP_STATEID_NFSV41))
dst->seqid = 0;
EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_read);
EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_write);
EXPORT_TRACEPOINT_SYMBOL_GPL(nfs4_pnfs_commit_ds);
+
+EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_pg_init_read);
+EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_pg_init_write);
+EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_pg_get_mirror_count);
+EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_read_done);
+EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_write_done);
+EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_read_pagelist);
+EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_write_pagelist);
#endif
TRACE_DEFINE_ENUM(NFS4ERR_XDEV);
#define show_nfsv4_errors(error) \
- __print_symbolic(-(error), \
+ __print_symbolic(error, \
{ NFS4_OK, "OK" }, \
/* Mapped by nfs4_stat_to_errno() */ \
{ EPERM, "EPERM" }, \
TP_STRUCT__entry(
__string(dstaddr, clp->cl_hostname)
- __field(int, error)
+ __field(unsigned long, error)
),
TP_fast_assign(
),
TP_printk(
- "error=%d (%s) dstaddr=%s",
- __entry->error,
+ "error=%ld (%s) dstaddr=%s",
+ -__entry->error,
show_nfsv4_errors(__entry->error),
__get_str(dstaddr)
)
__field(unsigned int, highest_slotid)
__field(unsigned int, target_highest_slotid)
__field(unsigned int, status_flags)
- __field(int, error)
+ __field(unsigned long, error)
),
TP_fast_assign(
__entry->error = res->sr_status;
),
TP_printk(
- "error=%d (%s) session=0x%08x slot_nr=%u seq_nr=%u "
+ "error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u "
"highest_slotid=%u target_highest_slotid=%u "
"status_flags=%u (%s)",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
__entry->session,
__entry->slot_nr,
__field(unsigned int, seq_nr)
__field(unsigned int, highest_slotid)
__field(unsigned int, cachethis)
- __field(int, error)
+ __field(unsigned long, error)
),
TP_fast_assign(
__entry->seq_nr = args->csa_sequenceid;
__entry->highest_slotid = args->csa_highestslotid;
__entry->cachethis = args->csa_cachethis;
- __entry->error = -be32_to_cpu(status);
+ __entry->error = be32_to_cpu(status);
),
TP_printk(
- "error=%d (%s) session=0x%08x slot_nr=%u seq_nr=%u "
+ "error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u "
"highest_slotid=%u",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
__entry->session,
__entry->slot_nr,
__entry->highest_slotid
)
);
+
+TRACE_EVENT(nfs4_cb_seqid_err,
+ TP_PROTO(
+ const struct cb_sequenceargs *args,
+ __be32 status
+ ),
+ TP_ARGS(args, status),
+
+ TP_STRUCT__entry(
+ __field(unsigned int, session)
+ __field(unsigned int, slot_nr)
+ __field(unsigned int, seq_nr)
+ __field(unsigned int, highest_slotid)
+ __field(unsigned int, cachethis)
+ __field(unsigned long, error)
+ ),
+
+ TP_fast_assign(
+ __entry->session = nfs_session_id_hash(&args->csa_sessionid);
+ __entry->slot_nr = args->csa_slotid;
+ __entry->seq_nr = args->csa_sequenceid;
+ __entry->highest_slotid = args->csa_highestslotid;
+ __entry->cachethis = args->csa_cachethis;
+ __entry->error = be32_to_cpu(status);
+ ),
+
+ TP_printk(
+ "error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u "
+ "highest_slotid=%u",
+ -__entry->error,
+ show_nfsv4_errors(__entry->error),
+ __entry->session,
+ __entry->slot_nr,
+ __entry->seq_nr,
+ __entry->highest_slotid
+ )
+);
+
#endif /* CONFIG_NFS_V4_1 */
TRACE_EVENT(nfs4_setup_sequence,
TRACE_EVENT(nfs4_xdr_status,
TP_PROTO(
+ const struct xdr_stream *xdr,
u32 op,
int error
),
- TP_ARGS(op, error),
+ TP_ARGS(xdr, op, error),
TP_STRUCT__entry(
+ __field(unsigned int, task_id)
+ __field(unsigned int, client_id)
+ __field(u32, xid)
__field(u32, op)
- __field(int, error)
+ __field(unsigned long, error)
),
TP_fast_assign(
+ const struct rpc_rqst *rqstp = xdr->rqst;
+ const struct rpc_task *task = rqstp->rq_task;
+
+ __entry->task_id = task->tk_pid;
+ __entry->client_id = task->tk_client->cl_clid;
+ __entry->xid = be32_to_cpu(rqstp->rq_xid);
__entry->op = op;
- __entry->error = -error;
+ __entry->error = error;
),
TP_printk(
- "operation %d: nfs status %d (%s)",
- __entry->op,
- __entry->error, show_nfsv4_errors(__entry->error)
+ "task:%u@%d xid=0x%08x error=%ld (%s) operation=%u",
+ __entry->task_id, __entry->client_id, __entry->xid,
+ -__entry->error, show_nfsv4_errors(__entry->error),
+ __entry->op
)
);
TP_ARGS(ctx, flags, error),
TP_STRUCT__entry(
- __field(int, error)
+ __field(unsigned long, error)
__field(unsigned int, flags)
__field(unsigned int, fmode)
__field(dev_t, dev)
const struct nfs4_state *state = ctx->state;
const struct inode *inode = NULL;
- __entry->error = error;
+ __entry->error = -error;
__entry->flags = flags;
__entry->fmode = (__force unsigned int)ctx->mode;
__entry->dev = ctx->dentry->d_sb->s_dev;
),
TP_printk(
- "error=%d (%s) flags=%d (%s) fmode=%s "
+ "error=%ld (%s) flags=%d (%s) fmode=%s "
"fileid=%02x:%02x:%llu fhandle=0x%08x "
"name=%02x:%02x:%llu/%s stateid=%d:0x%08x "
"openstateid=%d:0x%08x",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
__entry->flags,
show_open_flags(__entry->flags),
__field(u32, fhandle)
__field(u64, fileid)
__field(unsigned int, fmode)
- __field(int, error)
+ __field(unsigned long, error)
__field(int, stateid_seq)
__field(u32, stateid_hash)
),
),
TP_printk(
- "error=%d (%s) fmode=%s fileid=%02x:%02x:%llu "
+ "error=%ld (%s) fmode=%s fileid=%02x:%02x:%llu "
"fhandle=0x%08x openstateid=%d:0x%08x",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
__entry->fmode ? show_fmode_flags(__entry->fmode) :
"closed",
TP_ARGS(request, state, cmd, error),
TP_STRUCT__entry(
- __field(int, error)
+ __field(unsigned long, error)
__field(int, cmd)
__field(char, type)
__field(loff_t, start)
),
TP_printk(
- "error=%d (%s) cmd=%s:%s range=%lld:%lld "
+ "error=%ld (%s) cmd=%s:%s range=%lld:%lld "
"fileid=%02x:%02x:%llu fhandle=0x%08x "
"stateid=%d:0x%08x",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
show_lock_cmd(__entry->cmd),
show_lock_type(__entry->type),
TP_ARGS(request, state, lockstateid, cmd, error),
TP_STRUCT__entry(
- __field(int, error)
+ __field(unsigned long, error)
__field(int, cmd)
__field(char, type)
__field(loff_t, start)
),
TP_printk(
- "error=%d (%s) cmd=%s:%s range=%lld:%lld "
+ "error=%ld (%s) cmd=%s:%s range=%lld:%lld "
"fileid=%02x:%02x:%llu fhandle=0x%08x "
"stateid=%d:0x%08x lockstateid=%d:0x%08x",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
show_lock_cmd(__entry->cmd),
show_lock_type(__entry->type),
TP_STRUCT__entry(
__field(dev_t, dev)
__field(u32, fhandle)
- __field(int, error)
+ __field(unsigned long, error)
__field(int, stateid_seq)
__field(u32, stateid_hash)
),
),
TP_printk(
- "error=%d (%s) dev=%02x:%02x fhandle=0x%08x "
+ "error=%ld (%s) dev=%02x:%02x fhandle=0x%08x "
"stateid=%d:0x%08x",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
__entry->fhandle,
TP_ARGS(state, lsp, error),
TP_STRUCT__entry(
- __field(int, error)
+ __field(unsigned long, error)
__field(dev_t, dev)
__field(u32, fhandle)
__field(u64, fileid)
),
TP_printk(
- "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+ "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
"stateid=%d:0x%08x",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->fileid,
TP_STRUCT__entry(
__field(dev_t, dev)
- __field(int, error)
+ __field(unsigned long, error)
__field(u64, dir)
__string(name, name->name)
),
TP_fast_assign(
__entry->dev = dir->i_sb->s_dev;
__entry->dir = NFS_FILEID(dir);
- __entry->error = error;
+ __entry->error = -error;
__assign_str(name, name->name);
),
TP_printk(
- "error=%d (%s) name=%02x:%02x:%llu/%s",
- __entry->error,
+ "error=%ld (%s) name=%02x:%02x:%llu/%s",
+ -__entry->error,
show_nfsv4_errors(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->dir,
TP_STRUCT__entry(
__field(dev_t, dev)
__field(u64, ino)
- __field(int, error)
+ __field(unsigned long, error)
),
TP_fast_assign(
),
TP_printk(
- "error=%d (%s) inode=%02x:%02x:%llu",
- __entry->error,
+ "error=%ld (%s) inode=%02x:%02x:%llu",
+ -__entry->error,
show_nfsv4_errors(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->ino
TP_STRUCT__entry(
__field(dev_t, dev)
- __field(int, error)
+ __field(unsigned long, error)
__field(u64, olddir)
__string(oldname, oldname->name)
__field(u64, newdir)
),
TP_printk(
- "error=%d (%s) oldname=%02x:%02x:%llu/%s "
+ "error=%ld (%s) oldname=%02x:%02x:%llu/%s "
"newname=%02x:%02x:%llu/%s",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->olddir,
__field(dev_t, dev)
__field(u32, fhandle)
__field(u64, fileid)
- __field(int, error)
+ __field(unsigned long, error)
),
TP_fast_assign(
__entry->dev = inode->i_sb->s_dev;
__entry->fileid = NFS_FILEID(inode);
__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
- __entry->error = error;
+ __entry->error = error < 0 ? -error : 0;
),
TP_printk(
- "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x",
- __entry->error,
+ "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x",
+ -__entry->error,
show_nfsv4_errors(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->fileid,
__field(dev_t, dev)
__field(u32, fhandle)
__field(u64, fileid)
- __field(int, error)
+ __field(unsigned long, error)
__field(int, stateid_seq)
__field(u32, stateid_hash)
),
),
TP_printk(
- "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+ "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
"stateid=%d:0x%08x",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->fileid,
__field(u32, fhandle)
__field(u64, fileid)
__field(unsigned int, valid)
- __field(int, error)
+ __field(unsigned long, error)
),
TP_fast_assign(
),
TP_printk(
- "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+ "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
"valid=%s",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->fileid,
TP_ARGS(clp, fhandle, inode, error),
TP_STRUCT__entry(
- __field(int, error)
+ __field(unsigned long, error)
__field(dev_t, dev)
__field(u32, fhandle)
__field(u64, fileid)
),
TP_printk(
- "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+ "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
"dstaddr=%s",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->fileid,
TP_ARGS(clp, fhandle, inode, stateid, error),
TP_STRUCT__entry(
- __field(int, error)
+ __field(unsigned long, error)
__field(dev_t, dev)
__field(u32, fhandle)
__field(u64, fileid)
),
TP_printk(
- "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+ "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
"stateid=%d:0x%08x dstaddr=%s",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->fileid,
TP_ARGS(name, len, id, error),
TP_STRUCT__entry(
- __field(int, error)
+ __field(unsigned long, error)
__field(u32, id)
__dynamic_array(char, name, len > 0 ? len + 1 : 1)
),
),
TP_printk(
- "error=%d id=%u name=%s",
- __entry->error,
+ "error=%ld (%s) id=%u name=%s",
+ -__entry->error, show_nfsv4_errors(__entry->error),
__entry->id,
__get_str(name)
)
__field(u64, fileid)
__field(loff_t, offset)
__field(size_t, count)
- __field(int, error)
+ __field(unsigned long, error)
__field(int, stateid_seq)
__field(u32, stateid_hash)
),
__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
__entry->offset = hdr->args.offset;
__entry->count = hdr->args.count;
- __entry->error = error;
+ __entry->error = error < 0 ? -error : 0;
__entry->stateid_seq =
be32_to_cpu(state->stateid.seqid);
__entry->stateid_hash =
),
TP_printk(
- "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+ "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
"offset=%lld count=%zu stateid=%d:0x%08x",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->fileid,
__field(u64, fileid)
__field(loff_t, offset)
__field(size_t, count)
- __field(int, error)
+ __field(unsigned long, error)
__field(int, stateid_seq)
__field(u32, stateid_hash)
),
__entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
__entry->offset = hdr->args.offset;
__entry->count = hdr->args.count;
- __entry->error = error;
+ __entry->error = error < 0 ? -error : 0;
__entry->stateid_seq =
be32_to_cpu(state->stateid.seqid);
__entry->stateid_hash =
),
TP_printk(
- "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+ "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
"offset=%lld count=%zu stateid=%d:0x%08x",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->fileid,
__field(u64, fileid)
__field(loff_t, offset)
__field(size_t, count)
- __field(int, error)
+ __field(unsigned long, error)
),
TP_fast_assign(
),
TP_printk(
- "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+ "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
"offset=%lld count=%zu",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->fileid,
__field(u32, iomode)
__field(u64, offset)
__field(u64, count)
- __field(int, error)
+ __field(unsigned long, error)
__field(int, stateid_seq)
__field(u32, stateid_hash)
__field(int, layoutstateid_seq)
),
TP_printk(
- "error=%d (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
+ "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
"iomode=%s offset=%llu count=%llu stateid=%d:0x%08x "
"layoutstateid=%d:0x%08x",
- __entry->error,
+ -__entry->error,
show_nfsv4_errors(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->fileid,
TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_INVALID_OPEN);
TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_RETRY);
TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_SEND_LAYOUTGET);
+TRACE_DEFINE_ENUM(PNFS_UPDATE_LAYOUT_EXIT);
#define show_pnfs_update_layout_reason(reason) \
__print_symbolic(reason, \
{ PNFS_UPDATE_LAYOUT_BLOCKED, "layouts blocked" }, \
{ PNFS_UPDATE_LAYOUT_INVALID_OPEN, "invalid open" }, \
{ PNFS_UPDATE_LAYOUT_RETRY, "retrying" }, \
- { PNFS_UPDATE_LAYOUT_SEND_LAYOUTGET, "sent layoutget" })
+ { PNFS_UPDATE_LAYOUT_SEND_LAYOUTGET, "sent layoutget" }, \
+ { PNFS_UPDATE_LAYOUT_EXIT, "exit" })
TRACE_EVENT(pnfs_update_layout,
TP_PROTO(struct inode *inode,
)
);
+DECLARE_EVENT_CLASS(pnfs_layout_event,
+ TP_PROTO(struct inode *inode,
+ loff_t pos,
+ u64 count,
+ enum pnfs_iomode iomode,
+ struct pnfs_layout_hdr *lo,
+ struct pnfs_layout_segment *lseg
+ ),
+ TP_ARGS(inode, pos, count, iomode, lo, lseg),
+ TP_STRUCT__entry(
+ __field(dev_t, dev)
+ __field(u64, fileid)
+ __field(u32, fhandle)
+ __field(loff_t, pos)
+ __field(u64, count)
+ __field(enum pnfs_iomode, iomode)
+ __field(int, layoutstateid_seq)
+ __field(u32, layoutstateid_hash)
+ __field(long, lseg)
+ ),
+ TP_fast_assign(
+ __entry->dev = inode->i_sb->s_dev;
+ __entry->fileid = NFS_FILEID(inode);
+ __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode));
+ __entry->pos = pos;
+ __entry->count = count;
+ __entry->iomode = iomode;
+ if (lo != NULL) {
+ __entry->layoutstateid_seq =
+ be32_to_cpu(lo->plh_stateid.seqid);
+ __entry->layoutstateid_hash =
+ nfs_stateid_hash(&lo->plh_stateid);
+ } else {
+ __entry->layoutstateid_seq = 0;
+ __entry->layoutstateid_hash = 0;
+ }
+ __entry->lseg = (long)lseg;
+ ),
+ TP_printk(
+ "fileid=%02x:%02x:%llu fhandle=0x%08x "
+ "iomode=%s pos=%llu count=%llu "
+ "layoutstateid=%d:0x%08x lseg=0x%lx",
+ MAJOR(__entry->dev), MINOR(__entry->dev),
+ (unsigned long long)__entry->fileid,
+ __entry->fhandle,
+ show_pnfs_iomode(__entry->iomode),
+ (unsigned long long)__entry->pos,
+ (unsigned long long)__entry->count,
+ __entry->layoutstateid_seq, __entry->layoutstateid_hash,
+ __entry->lseg
+ )
+);
+
+#define DEFINE_PNFS_LAYOUT_EVENT(name) \
+ DEFINE_EVENT(pnfs_layout_event, name, \
+ TP_PROTO(struct inode *inode, \
+ loff_t pos, \
+ u64 count, \
+ enum pnfs_iomode iomode, \
+ struct pnfs_layout_hdr *lo, \
+ struct pnfs_layout_segment *lseg \
+ ), \
+ TP_ARGS(inode, pos, count, iomode, lo, lseg))
+
+DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_pg_init_read);
+DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_pg_init_write);
+DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_pg_get_mirror_count);
+DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_read_done);
+DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_write_done);
+DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_read_pagelist);
+DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_write_pagelist);
+
#endif /* CONFIG_NFS_V4_1 */
#endif /* _TRACE_NFS4_H */
#define NFS4_dec_sequence_sz \
(compound_decode_hdr_maxsz + \
decode_sequence_maxsz)
+#endif
#define NFS4_enc_get_lease_time_sz (compound_encode_hdr_maxsz + \
encode_sequence_maxsz + \
encode_putrootfh_maxsz + \
decode_sequence_maxsz + \
decode_putrootfh_maxsz + \
decode_fsinfo_maxsz)
+#if defined(CONFIG_NFS_V4_1)
#define NFS4_enc_reclaim_complete_sz (compound_encode_hdr_maxsz + \
encode_sequence_maxsz + \
encode_reclaim_complete_maxsz)
encode_nops(&hdr);
}
+#endif
+
/*
* a GET_LEASE_TIME request
*/
encode_nops(&hdr);
}
+#ifdef CONFIG_NFS_V4_1
+
/*
* a RECLAIM_COMPLETE request
*/
return true;
out_status:
nfserr = be32_to_cpup(p);
- trace_nfs4_xdr_status(opnum, nfserr);
+ trace_nfs4_xdr_status(xdr, opnum, nfserr);
*nfs_retval = nfs4_stat_to_errno(nfserr);
return true;
out_bad_operation:
*res = be32_to_cpup(p);
bitmap[0] &= ~FATTR4_WORD0_LEASE_TIME;
}
- dprintk("%s: file size=%u\n", __func__, (unsigned int)*res);
+ dprintk("%s: lease time=%u\n", __func__, (unsigned int)*res);
return 0;
}
return status;
}
+#endif
+
/*
* Decode GET_LEASE_TIME response
*/
return status;
}
+#ifdef CONFIG_NFS_V4_1
+
/*
* Decode RECLAIM_COMPLETE response
*/
PROC41(CREATE_SESSION, enc_create_session, dec_create_session),
PROC41(DESTROY_SESSION, enc_destroy_session, dec_destroy_session),
PROC41(SEQUENCE, enc_sequence, dec_sequence),
- PROC41(GET_LEASE_TIME, enc_get_lease_time, dec_get_lease_time),
+ PROC(GET_LEASE_TIME, enc_get_lease_time, dec_get_lease_time),
PROC41(RECLAIM_COMPLETE,enc_reclaim_complete, dec_reclaim_complete),
PROC41(GETDEVICEINFO, enc_getdeviceinfo, dec_getdeviceinfo),
PROC41(LAYOUTGET, enc_layoutget, dec_layoutget),
#include <linux/tracepoint.h>
#include <linux/iversion.h>
+TRACE_DEFINE_ENUM(DT_UNKNOWN);
+TRACE_DEFINE_ENUM(DT_FIFO);
+TRACE_DEFINE_ENUM(DT_CHR);
+TRACE_DEFINE_ENUM(DT_DIR);
+TRACE_DEFINE_ENUM(DT_BLK);
+TRACE_DEFINE_ENUM(DT_REG);
+TRACE_DEFINE_ENUM(DT_LNK);
+TRACE_DEFINE_ENUM(DT_SOCK);
+TRACE_DEFINE_ENUM(DT_WHT);
+
#define nfs_show_file_type(ftype) \
__print_symbolic(ftype, \
{ DT_UNKNOWN, "UNKNOWN" }, \
{ DT_SOCK, "SOCK" }, \
{ DT_WHT, "WHT" })
+TRACE_DEFINE_ENUM(NFS_INO_INVALID_DATA);
+TRACE_DEFINE_ENUM(NFS_INO_INVALID_ATIME);
+TRACE_DEFINE_ENUM(NFS_INO_INVALID_ACCESS);
+TRACE_DEFINE_ENUM(NFS_INO_INVALID_ACL);
+TRACE_DEFINE_ENUM(NFS_INO_REVAL_PAGECACHE);
+TRACE_DEFINE_ENUM(NFS_INO_REVAL_FORCED);
+TRACE_DEFINE_ENUM(NFS_INO_INVALID_LABEL);
+TRACE_DEFINE_ENUM(NFS_INO_INVALID_CHANGE);
+TRACE_DEFINE_ENUM(NFS_INO_INVALID_CTIME);
+TRACE_DEFINE_ENUM(NFS_INO_INVALID_MTIME);
+TRACE_DEFINE_ENUM(NFS_INO_INVALID_SIZE);
+TRACE_DEFINE_ENUM(NFS_INO_INVALID_OTHER);
+
#define nfs_show_cache_validity(v) \
__print_flags(v, "|", \
- { NFS_INO_INVALID_ATTR, "INVALID_ATTR" }, \
{ NFS_INO_INVALID_DATA, "INVALID_DATA" }, \
{ NFS_INO_INVALID_ATIME, "INVALID_ATIME" }, \
{ NFS_INO_INVALID_ACCESS, "INVALID_ACCESS" }, \
{ NFS_INO_INVALID_ACL, "INVALID_ACL" }, \
{ NFS_INO_REVAL_PAGECACHE, "REVAL_PAGECACHE" }, \
{ NFS_INO_REVAL_FORCED, "REVAL_FORCED" }, \
- { NFS_INO_INVALID_LABEL, "INVALID_LABEL" })
+ { NFS_INO_INVALID_LABEL, "INVALID_LABEL" }, \
+ { NFS_INO_INVALID_CHANGE, "INVALID_CHANGE" }, \
+ { NFS_INO_INVALID_CTIME, "INVALID_CTIME" }, \
+ { NFS_INO_INVALID_MTIME, "INVALID_MTIME" }, \
+ { NFS_INO_INVALID_SIZE, "INVALID_SIZE" }, \
+ { NFS_INO_INVALID_OTHER, "INVALID_OTHER" })
+
+TRACE_DEFINE_ENUM(NFS_INO_ADVISE_RDPLUS);
+TRACE_DEFINE_ENUM(NFS_INO_STALE);
+TRACE_DEFINE_ENUM(NFS_INO_ACL_LRU_SET);
+TRACE_DEFINE_ENUM(NFS_INO_INVALIDATING);
+TRACE_DEFINE_ENUM(NFS_INO_FSCACHE);
+TRACE_DEFINE_ENUM(NFS_INO_FSCACHE_LOCK);
+TRACE_DEFINE_ENUM(NFS_INO_LAYOUTCOMMIT);
+TRACE_DEFINE_ENUM(NFS_INO_LAYOUTCOMMITTING);
+TRACE_DEFINE_ENUM(NFS_INO_LAYOUTSTATS);
+TRACE_DEFINE_ENUM(NFS_INO_ODIRECT);
#define nfs_show_nfsi_flags(v) \
__print_flags(v, "|", \
- { 1 << NFS_INO_ADVISE_RDPLUS, "ADVISE_RDPLUS" }, \
- { 1 << NFS_INO_STALE, "STALE" }, \
- { 1 << NFS_INO_INVALIDATING, "INVALIDATING" }, \
- { 1 << NFS_INO_FSCACHE, "FSCACHE" }, \
- { 1 << NFS_INO_LAYOUTCOMMIT, "NEED_LAYOUTCOMMIT" }, \
- { 1 << NFS_INO_LAYOUTCOMMITTING, "LAYOUTCOMMIT" })
+ { BIT(NFS_INO_ADVISE_RDPLUS), "ADVISE_RDPLUS" }, \
+ { BIT(NFS_INO_STALE), "STALE" }, \
+ { BIT(NFS_INO_ACL_LRU_SET), "ACL_LRU_SET" }, \
+ { BIT(NFS_INO_INVALIDATING), "INVALIDATING" }, \
+ { BIT(NFS_INO_FSCACHE), "FSCACHE" }, \
+ { BIT(NFS_INO_FSCACHE_LOCK), "FSCACHE_LOCK" }, \
+ { BIT(NFS_INO_LAYOUTCOMMIT), "NEED_LAYOUTCOMMIT" }, \
+ { BIT(NFS_INO_LAYOUTCOMMITTING), "LAYOUTCOMMIT" }, \
+ { BIT(NFS_INO_LAYOUTSTATS), "LAYOUTSTATS" }, \
+ { BIT(NFS_INO_ODIRECT), "ODIRECT" })
DECLARE_EVENT_CLASS(nfs_inode_event,
TP_PROTO(
TP_ARGS(inode, error),
TP_STRUCT__entry(
- __field(int, error)
+ __field(unsigned long, error)
__field(dev_t, dev)
__field(u32, fhandle)
__field(unsigned char, type)
TP_fast_assign(
const struct nfs_inode *nfsi = NFS_I(inode);
- __entry->error = error;
+ __entry->error = error < 0 ? -error : 0;
__entry->dev = inode->i_sb->s_dev;
__entry->fileid = nfsi->fileid;
__entry->fhandle = nfs_fhandle_hash(&nfsi->fh);
),
TP_printk(
- "error=%d fileid=%02x:%02x:%llu fhandle=0x%08x "
+ "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x "
"type=%u (%s) version=%llu size=%lld "
- "cache_validity=%lu (%s) nfs_flags=%ld (%s)",
- __entry->error,
+ "cache_validity=0x%lx (%s) nfs_flags=0x%lx (%s)",
+ -__entry->error, nfs_show_status(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->fileid,
__entry->fhandle,
DEFINE_NFS_INODE_EVENT(nfs_access_enter);
DEFINE_NFS_INODE_EVENT_DONE(nfs_access_exit);
+TRACE_DEFINE_ENUM(LOOKUP_FOLLOW);
+TRACE_DEFINE_ENUM(LOOKUP_DIRECTORY);
+TRACE_DEFINE_ENUM(LOOKUP_AUTOMOUNT);
+TRACE_DEFINE_ENUM(LOOKUP_PARENT);
+TRACE_DEFINE_ENUM(LOOKUP_REVAL);
+TRACE_DEFINE_ENUM(LOOKUP_RCU);
+TRACE_DEFINE_ENUM(LOOKUP_NO_REVAL);
+TRACE_DEFINE_ENUM(LOOKUP_NO_EVAL);
+TRACE_DEFINE_ENUM(LOOKUP_OPEN);
+TRACE_DEFINE_ENUM(LOOKUP_CREATE);
+TRACE_DEFINE_ENUM(LOOKUP_EXCL);
+TRACE_DEFINE_ENUM(LOOKUP_RENAME_TARGET);
+TRACE_DEFINE_ENUM(LOOKUP_JUMPED);
+TRACE_DEFINE_ENUM(LOOKUP_ROOT);
+TRACE_DEFINE_ENUM(LOOKUP_EMPTY);
+TRACE_DEFINE_ENUM(LOOKUP_DOWN);
+
#define show_lookup_flags(flags) \
- __print_flags((unsigned long)flags, "|", \
- { LOOKUP_AUTOMOUNT, "AUTOMOUNT" }, \
+ __print_flags(flags, "|", \
+ { LOOKUP_FOLLOW, "FOLLOW" }, \
{ LOOKUP_DIRECTORY, "DIRECTORY" }, \
+ { LOOKUP_AUTOMOUNT, "AUTOMOUNT" }, \
+ { LOOKUP_PARENT, "PARENT" }, \
+ { LOOKUP_REVAL, "REVAL" }, \
+ { LOOKUP_RCU, "RCU" }, \
+ { LOOKUP_NO_REVAL, "NO_REVAL" }, \
+ { LOOKUP_NO_EVAL, "NO_EVAL" }, \
{ LOOKUP_OPEN, "OPEN" }, \
{ LOOKUP_CREATE, "CREATE" }, \
- { LOOKUP_EXCL, "EXCL" })
+ { LOOKUP_EXCL, "EXCL" }, \
+ { LOOKUP_RENAME_TARGET, "RENAME_TARGET" }, \
+ { LOOKUP_JUMPED, "JUMPED" }, \
+ { LOOKUP_ROOT, "ROOT" }, \
+ { LOOKUP_EMPTY, "EMPTY" }, \
+ { LOOKUP_DOWN, "DOWN" })
DECLARE_EVENT_CLASS(nfs_lookup_event,
TP_PROTO(
TP_ARGS(dir, dentry, flags),
TP_STRUCT__entry(
- __field(unsigned int, flags)
+ __field(unsigned long, flags)
__field(dev_t, dev)
__field(u64, dir)
__string(name, dentry->d_name.name)
),
TP_printk(
- "flags=%u (%s) name=%02x:%02x:%llu/%s",
+ "flags=0x%lx (%s) name=%02x:%02x:%llu/%s",
__entry->flags,
show_lookup_flags(__entry->flags),
MAJOR(__entry->dev), MINOR(__entry->dev),
TP_ARGS(dir, dentry, flags, error),
TP_STRUCT__entry(
- __field(int, error)
- __field(unsigned int, flags)
+ __field(unsigned long, error)
+ __field(unsigned long, flags)
__field(dev_t, dev)
__field(u64, dir)
__string(name, dentry->d_name.name)
TP_fast_assign(
__entry->dev = dir->i_sb->s_dev;
__entry->dir = NFS_FILEID(dir);
- __entry->error = error;
+ __entry->error = error < 0 ? -error : 0;
__entry->flags = flags;
__assign_str(name, dentry->d_name.name);
),
TP_printk(
- "error=%d flags=%u (%s) name=%02x:%02x:%llu/%s",
- __entry->error,
+ "error=%ld (%s) flags=0x%lx (%s) name=%02x:%02x:%llu/%s",
+ -__entry->error, nfs_show_status(__entry->error),
__entry->flags,
show_lookup_flags(__entry->flags),
MAJOR(__entry->dev), MINOR(__entry->dev),
DEFINE_NFS_LOOKUP_EVENT(nfs_lookup_revalidate_enter);
DEFINE_NFS_LOOKUP_EVENT_DONE(nfs_lookup_revalidate_exit);
+TRACE_DEFINE_ENUM(O_WRONLY);
+TRACE_DEFINE_ENUM(O_RDWR);
+TRACE_DEFINE_ENUM(O_CREAT);
+TRACE_DEFINE_ENUM(O_EXCL);
+TRACE_DEFINE_ENUM(O_NOCTTY);
+TRACE_DEFINE_ENUM(O_TRUNC);
+TRACE_DEFINE_ENUM(O_APPEND);
+TRACE_DEFINE_ENUM(O_NONBLOCK);
+TRACE_DEFINE_ENUM(O_DSYNC);
+TRACE_DEFINE_ENUM(O_DIRECT);
+TRACE_DEFINE_ENUM(O_LARGEFILE);
+TRACE_DEFINE_ENUM(O_DIRECTORY);
+TRACE_DEFINE_ENUM(O_NOFOLLOW);
+TRACE_DEFINE_ENUM(O_NOATIME);
+TRACE_DEFINE_ENUM(O_CLOEXEC);
+
#define show_open_flags(flags) \
- __print_flags((unsigned long)flags, "|", \
+ __print_flags(flags, "|", \
+ { O_WRONLY, "O_WRONLY" }, \
+ { O_RDWR, "O_RDWR" }, \
{ O_CREAT, "O_CREAT" }, \
{ O_EXCL, "O_EXCL" }, \
+ { O_NOCTTY, "O_NOCTTY" }, \
{ O_TRUNC, "O_TRUNC" }, \
{ O_APPEND, "O_APPEND" }, \
+ { O_NONBLOCK, "O_NONBLOCK" }, \
{ O_DSYNC, "O_DSYNC" }, \
{ O_DIRECT, "O_DIRECT" }, \
- { O_DIRECTORY, "O_DIRECTORY" })
+ { O_LARGEFILE, "O_LARGEFILE" }, \
+ { O_DIRECTORY, "O_DIRECTORY" }, \
+ { O_NOFOLLOW, "O_NOFOLLOW" }, \
+ { O_NOATIME, "O_NOATIME" }, \
+ { O_CLOEXEC, "O_CLOEXEC" })
+
+TRACE_DEFINE_ENUM(FMODE_READ);
+TRACE_DEFINE_ENUM(FMODE_WRITE);
+TRACE_DEFINE_ENUM(FMODE_EXEC);
#define show_fmode_flags(mode) \
__print_flags(mode, "|", \
TP_ARGS(dir, ctx, flags),
TP_STRUCT__entry(
- __field(unsigned int, flags)
+ __field(unsigned long, flags)
__field(unsigned int, fmode)
__field(dev_t, dev)
__field(u64, dir)
),
TP_printk(
- "flags=%u (%s) fmode=%s name=%02x:%02x:%llu/%s",
+ "flags=0x%lx (%s) fmode=%s name=%02x:%02x:%llu/%s",
__entry->flags,
show_open_flags(__entry->flags),
show_fmode_flags(__entry->fmode),
TP_ARGS(dir, ctx, flags, error),
TP_STRUCT__entry(
- __field(int, error)
- __field(unsigned int, flags)
+ __field(unsigned long, error)
+ __field(unsigned long, flags)
__field(unsigned int, fmode)
__field(dev_t, dev)
__field(u64, dir)
),
TP_fast_assign(
- __entry->error = error;
+ __entry->error = -error;
__entry->dev = dir->i_sb->s_dev;
__entry->dir = NFS_FILEID(dir);
__entry->flags = flags;
),
TP_printk(
- "error=%d flags=%u (%s) fmode=%s "
+ "error=%ld (%s) flags=0x%lx (%s) fmode=%s "
"name=%02x:%02x:%llu/%s",
- __entry->error,
+ -__entry->error, nfs_show_status(__entry->error),
__entry->flags,
show_open_flags(__entry->flags),
show_fmode_flags(__entry->fmode),
TP_ARGS(dir, dentry, flags),
TP_STRUCT__entry(
- __field(unsigned int, flags)
+ __field(unsigned long, flags)
__field(dev_t, dev)
__field(u64, dir)
__string(name, dentry->d_name.name)
),
TP_printk(
- "flags=%u (%s) name=%02x:%02x:%llu/%s",
+ "flags=0x%lx (%s) name=%02x:%02x:%llu/%s",
__entry->flags,
show_open_flags(__entry->flags),
MAJOR(__entry->dev), MINOR(__entry->dev),
TP_ARGS(dir, dentry, flags, error),
TP_STRUCT__entry(
- __field(int, error)
- __field(unsigned int, flags)
+ __field(unsigned long, error)
+ __field(unsigned long, flags)
__field(dev_t, dev)
__field(u64, dir)
__string(name, dentry->d_name.name)
),
TP_fast_assign(
- __entry->error = error;
+ __entry->error = -error;
__entry->dev = dir->i_sb->s_dev;
__entry->dir = NFS_FILEID(dir);
__entry->flags = flags;
),
TP_printk(
- "error=%d flags=%u (%s) name=%02x:%02x:%llu/%s",
- __entry->error,
+ "error=%ld (%s) flags=0x%lx (%s) name=%02x:%02x:%llu/%s",
+ -__entry->error, nfs_show_status(__entry->error),
__entry->flags,
show_open_flags(__entry->flags),
MAJOR(__entry->dev), MINOR(__entry->dev),
TP_ARGS(dir, dentry, error),
TP_STRUCT__entry(
- __field(int, error)
+ __field(unsigned long, error)
__field(dev_t, dev)
__field(u64, dir)
__string(name, dentry->d_name.name)
TP_fast_assign(
__entry->dev = dir->i_sb->s_dev;
__entry->dir = NFS_FILEID(dir);
- __entry->error = error;
+ __entry->error = error < 0 ? -error : 0;
__assign_str(name, dentry->d_name.name);
),
TP_printk(
- "error=%d name=%02x:%02x:%llu/%s",
- __entry->error,
+ "error=%ld (%s) name=%02x:%02x:%llu/%s",
+ -__entry->error, nfs_show_status(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->dir,
__get_str(name)
TP_ARGS(inode, dir, dentry, error),
TP_STRUCT__entry(
- __field(int, error)
+ __field(unsigned long, error)
__field(dev_t, dev)
__field(u64, fileid)
__field(u64, dir)
__entry->dev = inode->i_sb->s_dev;
__entry->fileid = NFS_FILEID(inode);
__entry->dir = NFS_FILEID(dir);
- __entry->error = error;
+ __entry->error = error < 0 ? -error : 0;
__assign_str(name, dentry->d_name.name);
),
TP_printk(
- "error=%d fileid=%02x:%02x:%llu name=%02x:%02x:%llu/%s",
- __entry->error,
+ "error=%ld (%s) fileid=%02x:%02x:%llu name=%02x:%02x:%llu/%s",
+ -__entry->error, nfs_show_status(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
__entry->fileid,
MAJOR(__entry->dev), MINOR(__entry->dev),
TP_STRUCT__entry(
__field(dev_t, dev)
- __field(int, error)
+ __field(unsigned long, error)
__field(u64, old_dir)
__string(old_name, old_dentry->d_name.name)
__field(u64, new_dir)
TP_fast_assign(
__entry->dev = old_dir->i_sb->s_dev;
+ __entry->error = -error;
__entry->old_dir = NFS_FILEID(old_dir);
__entry->new_dir = NFS_FILEID(new_dir);
- __entry->error = error;
__assign_str(old_name, old_dentry->d_name.name);
__assign_str(new_name, new_dentry->d_name.name);
),
TP_printk(
- "error=%d old_name=%02x:%02x:%llu/%s "
+ "error=%ld (%s) old_name=%02x:%02x:%llu/%s "
"new_name=%02x:%02x:%llu/%s",
- __entry->error,
+ -__entry->error, nfs_show_status(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->old_dir,
__get_str(old_name),
TP_STRUCT__entry(
__field(dev_t, dev)
- __field(int, error)
+ __field(unsigned long, error)
__field(u64, dir)
__dynamic_array(char, name, data->args.name.len + 1)
),
size_t len = data->args.name.len;
__entry->dev = dir->i_sb->s_dev;
__entry->dir = NFS_FILEID(dir);
- __entry->error = error;
+ __entry->error = -error;
memcpy(__get_str(name),
data->args.name.name, len);
__get_str(name)[len] = 0;
),
TP_printk(
- "error=%d name=%02x:%02x:%llu/%s",
- __entry->error,
+ "error=%ld (%s) name=%02x:%02x:%llu/%s",
+ -__entry->error, nfs_show_status(__entry->error),
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->dir,
__get_str(name)
TRACE_DEFINE_ENUM(NFSERR_NOENT);
TRACE_DEFINE_ENUM(NFSERR_IO);
TRACE_DEFINE_ENUM(NFSERR_NXIO);
+TRACE_DEFINE_ENUM(ECHILD);
+TRACE_DEFINE_ENUM(NFSERR_EAGAIN);
TRACE_DEFINE_ENUM(NFSERR_ACCES);
TRACE_DEFINE_ENUM(NFSERR_EXIST);
TRACE_DEFINE_ENUM(NFSERR_XDEV);
TRACE_DEFINE_ENUM(NFSERR_NOSPC);
TRACE_DEFINE_ENUM(NFSERR_ROFS);
TRACE_DEFINE_ENUM(NFSERR_MLINK);
+TRACE_DEFINE_ENUM(NFSERR_OPNOTSUPP);
TRACE_DEFINE_ENUM(NFSERR_NAMETOOLONG);
TRACE_DEFINE_ENUM(NFSERR_NOTEMPTY);
TRACE_DEFINE_ENUM(NFSERR_DQUOT);
{ NFSERR_NOENT, "NOENT" }, \
{ NFSERR_IO, "IO" }, \
{ NFSERR_NXIO, "NXIO" }, \
+ { ECHILD, "CHILD" }, \
+ { NFSERR_EAGAIN, "AGAIN" }, \
{ NFSERR_ACCES, "ACCES" }, \
{ NFSERR_EXIST, "EXIST" }, \
{ NFSERR_XDEV, "XDEV" }, \
{ NFSERR_NOSPC, "NOSPC" }, \
{ NFSERR_ROFS, "ROFS" }, \
{ NFSERR_MLINK, "MLINK" }, \
+ { NFSERR_OPNOTSUPP, "OPNOTSUPP" }, \
{ NFSERR_NAMETOOLONG, "NAMETOOLONG" }, \
{ NFSERR_NOTEMPTY, "NOTEMPTY" }, \
{ NFSERR_DQUOT, "DQUOT" }, \
TRACE_EVENT(nfs_xdr_status,
TP_PROTO(
+ const struct xdr_stream *xdr,
int error
),
- TP_ARGS(error),
+ TP_ARGS(xdr, error),
TP_STRUCT__entry(
- __field(int, error)
+ __field(unsigned int, task_id)
+ __field(unsigned int, client_id)
+ __field(u32, xid)
+ __field(unsigned long, error)
),
TP_fast_assign(
+ const struct rpc_rqst *rqstp = xdr->rqst;
+ const struct rpc_task *task = rqstp->rq_task;
+
+ __entry->task_id = task->tk_pid;
+ __entry->client_id = task->tk_client->cl_clid;
+ __entry->xid = be32_to_cpu(rqstp->rq_xid);
__entry->error = error;
),
TP_printk(
- "error=%d (%s)",
- __entry->error, nfs_show_status(__entry->error)
+ "task:%u@%d xid=0x%08x error=%ld (%s)",
+ __entry->task_id, __entry->client_id, __entry->xid,
+ -__entry->error, nfs_show_status(__entry->error)
)
);
static inline struct nfs_page *
nfs_page_alloc(void)
{
- struct nfs_page *p = kmem_cache_zalloc(nfs_page_cachep, GFP_NOIO);
+ struct nfs_page *p = kmem_cache_zalloc(nfs_page_cachep, GFP_KERNEL);
if (p)
INIT_LIST_HEAD(&p->wb_list);
return p;
if (pagecount <= ARRAY_SIZE(pg_array->page_array))
pg_array->pagevec = pg_array->page_array;
else {
- if (hdr->rw_mode == FMODE_WRITE)
- gfp_flags = GFP_NOIO;
pg_array->pagevec = kcalloc(pagecount, sizeof(struct page *), gfp_flags);
if (!pg_array->pagevec) {
pg_array->npages = 0;
desc->pg_mirrors_dynamic = NULL;
if (mirror_count == 1)
return desc->pg_mirrors_static;
- ret = kmalloc_array(mirror_count, sizeof(*ret), GFP_NOFS);
+ ret = kmalloc_array(mirror_count, sizeof(*ret), GFP_KERNEL);
if (ret != NULL) {
for (i = 0; i < mirror_count; i++)
nfs_pageio_mirror_init(&ret[i], desc->pg_bsize);
spin_unlock(&ino->i_lock);
lseg = ERR_PTR(wait_var_event_killable(&lo->plh_outstanding,
!atomic_read(&lo->plh_outstanding)));
- if (IS_ERR(lseg) || !list_empty(&lo->plh_segs))
+ if (IS_ERR(lseg))
goto out_put_layout_hdr;
pnfs_put_layout_hdr(lo);
goto lookup_again;
* stateid.
*/
if (test_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags)) {
+ int status;
/*
* The first layoutget for the file. Need to serialize per
}
first = true;
- if (nfs4_select_rw_stateid(ctx->state,
+ status = nfs4_select_rw_stateid(ctx->state,
iomode == IOMODE_RW ? FMODE_WRITE : FMODE_READ,
- NULL, &stateid, NULL) != 0) {
+ NULL, &stateid, NULL);
+ if (status != 0) {
trace_pnfs_update_layout(ino, pos, count,
iomode, lo, lseg,
PNFS_UPDATE_LAYOUT_INVALID_OPEN);
- goto out_unlock;
+ if (status != -EAGAIN)
+ goto out_unlock;
+ spin_unlock(&ino->i_lock);
+ nfs4_schedule_stateid_recovery(server, ctx->state);
+ pnfs_clear_first_layoutget(lo);
+ pnfs_put_layout_hdr(lo);
+ goto lookup_again;
}
} else {
nfs4_stateid_copy(&stateid, &lo->plh_stateid);
out_put_layout_hdr:
if (first)
pnfs_clear_first_layoutget(lo);
+ trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg,
+ PNFS_UPDATE_LAYOUT_EXIT);
pnfs_put_layout_hdr(lo);
out:
dprintk("%s: inode %s/%llu pNFS layout segment %s for "
wb_size,
IOMODE_RW,
false,
- GFP_NOFS);
+ GFP_KERNEL);
if (IS_ERR(pgio->pg_lseg)) {
pgio->pg_error = PTR_ERR(pgio->pg_lseg);
pgio->pg_lseg = NULL;
#define NFS_DEFAULT_VERSION 2
#endif
+#define NFS_MAX_CONNECTIONS 16
+
enum {
/* Mount options that take no arguments */
Opt_soft, Opt_softerr, Opt_hard,
Opt_nfsvers,
Opt_sec, Opt_proto, Opt_mountproto, Opt_mounthost,
Opt_addr, Opt_mountaddr, Opt_clientaddr,
+ Opt_nconnect,
Opt_lookupcache,
Opt_fscache_uniq,
Opt_local_lock,
{ Opt_mounthost, "mounthost=%s" },
{ Opt_mountaddr, "mountaddr=%s" },
+ { Opt_nconnect, "nconnect=%s" },
+
{ Opt_lookupcache, "lookupcache=%s" },
{ Opt_fscache_uniq, "fsc=%s" },
{ Opt_local_lock, "local_lock=%s" },
}
default:
if (showdefaults)
- seq_printf(m, ",mountaddr=unspecified");
+ seq_puts(m, ",mountaddr=unspecified");
}
if (nfss->mountd_version || showdefaults)
seq_printf(m, ",proto=%s",
rpc_peeraddr2str(nfss->client, RPC_DISPLAY_NETID));
rcu_read_unlock();
+ if (clp->cl_nconnect > 0)
+ seq_printf(m, ",nconnect=%u", clp->cl_nconnect);
if (version == 4) {
if (nfss->port != NFS_PORT)
seq_printf(m, ",port=%u", nfss->port);
nfs_show_nfsv4_options(m, nfss, showdefaults);
if (nfss->options & NFS_OPTION_FSCACHE)
- seq_printf(m, ",fsc");
+ seq_puts(m, ",fsc");
if (nfss->options & NFS_OPTION_MIGRATION)
- seq_printf(m, ",migration");
+ seq_puts(m, ",migration");
if (nfss->flags & NFS_MOUNT_LOOKUP_CACHE_NONEG) {
if (nfss->flags & NFS_MOUNT_LOOKUP_CACHE_NONE)
- seq_printf(m, ",lookupcache=none");
+ seq_puts(m, ",lookupcache=none");
else
- seq_printf(m, ",lookupcache=pos");
+ seq_puts(m, ",lookupcache=pos");
}
local_flock = nfss->flags & NFS_MOUNT_LOCAL_FLOCK;
local_fcntl = nfss->flags & NFS_MOUNT_LOCAL_FCNTL;
if (!local_flock && !local_fcntl)
- seq_printf(m, ",local_lock=none");
+ seq_puts(m, ",local_lock=none");
else if (local_flock && local_fcntl)
- seq_printf(m, ",local_lock=all");
+ seq_puts(m, ",local_lock=all");
else if (local_flock)
- seq_printf(m, ",local_lock=flock");
+ seq_puts(m, ",local_lock=flock");
else
- seq_printf(m, ",local_lock=posix");
+ seq_puts(m, ",local_lock=posix");
}
/*
EXPORT_SYMBOL_GPL(nfs_show_options);
#if IS_ENABLED(CONFIG_NFS_V4)
+static void show_lease(struct seq_file *m, struct nfs_server *server)
+{
+ struct nfs_client *clp = server->nfs_client;
+ unsigned long expire;
+
+ seq_printf(m, ",lease_time=%ld", clp->cl_lease_time / HZ);
+ expire = clp->cl_last_renewal + clp->cl_lease_time;
+ seq_printf(m, ",lease_expired=%ld",
+ time_after(expire, jiffies) ? 0 : (jiffies - expire) / HZ);
+}
#ifdef CONFIG_NFS_V4_1
static void show_sessions(struct seq_file *m, struct nfs_server *server)
{
if (nfs4_has_session(server->nfs_client))
- seq_printf(m, ",sessions");
+ seq_puts(m, ",sessions");
}
#else
static void show_sessions(struct seq_file *m, struct nfs_server *server) {}
/*
* Display all mount option settings
*/
- seq_printf(m, "\n\topts:\t");
+ seq_puts(m, "\n\topts:\t");
seq_puts(m, sb_rdonly(root->d_sb) ? "ro" : "rw");
seq_puts(m, root->d_sb->s_flags & SB_SYNCHRONOUS ? ",sync" : "");
seq_puts(m, root->d_sb->s_flags & SB_NOATIME ? ",noatime" : "");
show_implementation_id(m, nfss);
- seq_printf(m, "\n\tcaps:\t");
+ seq_puts(m, "\n\tcaps:\t");
seq_printf(m, "caps=0x%x", nfss->caps);
seq_printf(m, ",wtmult=%u", nfss->wtmult);
seq_printf(m, ",dtsize=%u", nfss->dtsize);
#if IS_ENABLED(CONFIG_NFS_V4)
if (nfss->nfs_client->rpc_ops->version == 4) {
- seq_printf(m, "\n\tnfsv4:\t");
+ seq_puts(m, "\n\tnfsv4:\t");
seq_printf(m, "bm0=0x%x", nfss->attr_bitmask[0]);
seq_printf(m, ",bm1=0x%x", nfss->attr_bitmask[1]);
seq_printf(m, ",bm2=0x%x", nfss->attr_bitmask[2]);
seq_printf(m, ",acl=0x%x", nfss->acl_bitmask);
show_sessions(m, nfss);
show_pnfs(m, nfss);
+ show_lease(m, nfss);
}
#endif
preempt_enable();
}
- seq_printf(m, "\n\tevents:\t");
+ seq_puts(m, "\n\tevents:\t");
for (i = 0; i < __NFSIOS_COUNTSMAX; i++)
seq_printf(m, "%lu ", totals.events[i]);
- seq_printf(m, "\n\tbytes:\t");
+ seq_puts(m, "\n\tbytes:\t");
for (i = 0; i < __NFSIOS_BYTESMAX; i++)
seq_printf(m, "%Lu ", totals.bytes[i]);
#ifdef CONFIG_NFS_FSCACHE
if (nfss->options & NFS_OPTION_FSCACHE) {
- seq_printf(m, "\n\tfsc:\t");
+ seq_puts(m, "\n\tfsc:\t");
for (i = 0; i < __NFSIOS_FSCACHEMAX; i++)
seq_printf(m, "%Lu ", totals.fscache[i]);
}
#endif
- seq_printf(m, "\n");
+ seq_putc(m, '\n');
rpc_clnt_show_stats(m, nfss->client);
if (mnt->mount_server.addrlen == 0)
goto out_invalid_address;
break;
+ case Opt_nconnect:
+ if (nfs_get_option_ul_bound(args, &option, 1, NFS_MAX_CONNECTIONS))
+ goto out_invalid_value;
+ mnt->nfs_server.nconnect = option;
+ break;
case Opt_lookupcache:
string = match_strdup(args);
if (string == NULL)
--- /dev/null
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2019 Hammerspace Inc
+ */
+
+#include <linux/module.h>
+#include <linux/kobject.h>
+#include <linux/sysfs.h>
+#include <linux/fs.h>
+#include <linux/slab.h>
+#include <linux/netdevice.h>
+#include <linux/string.h>
+#include <linux/nfs_fs.h>
+#include <linux/rcupdate.h>
+
+#include "nfs4_fs.h"
+#include "netns.h"
+#include "sysfs.h"
+
+struct kobject *nfs_client_kobj;
+static struct kset *nfs_client_kset;
+
+static void nfs_netns_object_release(struct kobject *kobj)
+{
+ kfree(kobj);
+}
+
+static const struct kobj_ns_type_operations *nfs_netns_object_child_ns_type(
+ struct kobject *kobj)
+{
+ return &net_ns_type_operations;
+}
+
+static struct kobj_type nfs_netns_object_type = {
+ .release = nfs_netns_object_release,
+ .sysfs_ops = &kobj_sysfs_ops,
+ .child_ns_type = nfs_netns_object_child_ns_type,
+};
+
+static struct kobject *nfs_netns_object_alloc(const char *name,
+ struct kset *kset, struct kobject *parent)
+{
+ struct kobject *kobj;
+
+ kobj = kzalloc(sizeof(*kobj), GFP_KERNEL);
+ if (kobj) {
+ kobj->kset = kset;
+ if (kobject_init_and_add(kobj, &nfs_netns_object_type,
+ parent, "%s", name) == 0)
+ return kobj;
+ kobject_put(kobj);
+ }
+ return NULL;
+}
+
+int nfs_sysfs_init(void)
+{
+ nfs_client_kset = kset_create_and_add("nfs", NULL, fs_kobj);
+ if (!nfs_client_kset)
+ return -ENOMEM;
+ nfs_client_kobj = nfs_netns_object_alloc("net", nfs_client_kset, NULL);
+ if (!nfs_client_kobj) {
+ kset_unregister(nfs_client_kset);
+ nfs_client_kset = NULL;
+ return -ENOMEM;
+ }
+ return 0;
+}
+
+void nfs_sysfs_exit(void)
+{
+ kobject_put(nfs_client_kobj);
+ kset_unregister(nfs_client_kset);
+}
+
+static ssize_t nfs_netns_identifier_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ struct nfs_netns_client *c = container_of(kobj,
+ struct nfs_netns_client,
+ kobject);
+ return scnprintf(buf, PAGE_SIZE, "%s\n", c->identifier);
+}
+
+/* Strip trailing '\n' */
+static size_t nfs_string_strip(const char *c, size_t len)
+{
+ while (len > 0 && c[len-1] == '\n')
+ --len;
+ return len;
+}
+
+static ssize_t nfs_netns_identifier_store(struct kobject *kobj,
+ struct kobj_attribute *attr,
+ const char *buf, size_t count)
+{
+ struct nfs_netns_client *c = container_of(kobj,
+ struct nfs_netns_client,
+ kobject);
+ const char *old;
+ char *p;
+ size_t len;
+
+ len = nfs_string_strip(buf, min_t(size_t, count, CONTAINER_ID_MAXLEN));
+ if (!len)
+ return 0;
+ p = kmemdup_nul(buf, len, GFP_KERNEL);
+ if (!p)
+ return -ENOMEM;
+ old = xchg(&c->identifier, p);
+ if (old) {
+ synchronize_rcu();
+ kfree(old);
+ }
+ return count;
+}
+
+static void nfs_netns_client_release(struct kobject *kobj)
+{
+ struct nfs_netns_client *c = container_of(kobj,
+ struct nfs_netns_client,
+ kobject);
+
+ if (c->identifier)
+ kfree(c->identifier);
+ kfree(c);
+}
+
+static const void *nfs_netns_client_namespace(struct kobject *kobj)
+{
+ return container_of(kobj, struct nfs_netns_client, kobject)->net;
+}
+
+static struct kobj_attribute nfs_netns_client_id = __ATTR(identifier,
+ 0644, nfs_netns_identifier_show, nfs_netns_identifier_store);
+
+static struct attribute *nfs_netns_client_attrs[] = {
+ &nfs_netns_client_id.attr,
+ NULL,
+};
+
+static struct kobj_type nfs_netns_client_type = {
+ .release = nfs_netns_client_release,
+ .default_attrs = nfs_netns_client_attrs,
+ .sysfs_ops = &kobj_sysfs_ops,
+ .namespace = nfs_netns_client_namespace,
+};
+
+static struct nfs_netns_client *nfs_netns_client_alloc(struct kobject *parent,
+ struct net *net)
+{
+ struct nfs_netns_client *p;
+
+ p = kzalloc(sizeof(*p), GFP_KERNEL);
+ if (p) {
+ p->net = net;
+ p->kobject.kset = nfs_client_kset;
+ if (kobject_init_and_add(&p->kobject, &nfs_netns_client_type,
+ parent, "nfs_client") == 0)
+ return p;
+ kobject_put(&p->kobject);
+ }
+ return NULL;
+}
+
+void nfs_netns_sysfs_setup(struct nfs_net *netns, struct net *net)
+{
+ struct nfs_netns_client *clp;
+
+ clp = nfs_netns_client_alloc(nfs_client_kobj, net);
+ if (clp) {
+ netns->nfs_client = clp;
+ kobject_uevent(&clp->kobject, KOBJ_ADD);
+ }
+}
+
+void nfs_netns_sysfs_destroy(struct nfs_net *netns)
+{
+ struct nfs_netns_client *clp = netns->nfs_client;
+
+ if (clp) {
+ kobject_uevent(&clp->kobject, KOBJ_REMOVE);
+ kobject_del(&clp->kobject);
+ kobject_put(&clp->kobject);
+ netns->nfs_client = NULL;
+ }
+}
--- /dev/null
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2019 Hammerspace Inc
+ */
+
+#ifndef __NFS_SYSFS_H
+#define __NFS_SYSFS_H
+
+#define CONTAINER_ID_MAXLEN (64)
+
+struct nfs_netns_client {
+ struct kobject kobject;
+ struct net *net;
+ const char *identifier;
+};
+
+extern struct kobject *nfs_client_kobj;
+
+extern int nfs_sysfs_init(void);
+extern void nfs_sysfs_exit(void);
+
+void nfs_netns_sysfs_setup(struct nfs_net *netns, struct net *net);
+void nfs_netns_sysfs_destroy(struct nfs_net *netns);
+
+#endif
static struct nfs_pgio_header *nfs_writehdr_alloc(void)
{
- struct nfs_pgio_header *p = mempool_alloc(nfs_wdata_mempool, GFP_NOIO);
+ struct nfs_pgio_header *p = mempool_alloc(nfs_wdata_mempool, GFP_KERNEL);
memset(p, 0, sizeof(*p));
p->rw_mode = FMODE_WRITE;
struct inode *inode = mapping->host;
struct nfs_pageio_descriptor pgio;
struct nfs_io_completion *ioc;
- unsigned int pflags = memalloc_nofs_save();
int err;
nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGES);
- ioc = nfs_io_completion_alloc(GFP_NOFS);
+ ioc = nfs_io_completion_alloc(GFP_KERNEL);
if (ioc)
nfs_io_completion_init(ioc, nfs_io_completion_commit, inode);
nfs_pageio_complete(&pgio);
nfs_io_completion_put(ioc);
- memalloc_nofs_restore(pflags);
-
if (err < 0)
goto out_err;
err = pgio.pg_error;
PNFS_UPDATE_LAYOUT_BLOCKED,
PNFS_UPDATE_LAYOUT_INVALID_OPEN,
PNFS_UPDATE_LAYOUT_SEND_LAYOUTGET,
+ PNFS_UPDATE_LAYOUT_EXIT,
};
#define NFS4_OP_MAP_NUM_LONGS \
#define NFS_INO_INVALID_MTIME BIT(10) /* cached mtime is invalid */
#define NFS_INO_INVALID_SIZE BIT(11) /* cached size is invalid */
#define NFS_INO_INVALID_OTHER BIT(12) /* other attrs are invalid */
+#define NFS_INO_DATA_INVAL_DEFER \
+ BIT(13) /* Deferred cache invalidation */
#define NFS_INO_INVALID_ATTR (NFS_INO_INVALID_CHANGE \
| NFS_INO_INVALID_CTIME \
struct nfs_subversion * cl_nfs_mod; /* pointer to nfs version module */
u32 cl_minorversion;/* NFSv4 minorversion */
+ unsigned int cl_nconnect; /* Number of connections */
const char * cl_principal; /* used for machine cred */
#if IS_ENABLED(CONFIG_NFS_V4)
int xprt_setup_bc(struct rpc_xprt *xprt, unsigned int min_reqs);
void xprt_destroy_bc(struct rpc_xprt *xprt, unsigned int max_reqs);
void xprt_free_bc_rqst(struct rpc_rqst *req);
+unsigned int xprt_bc_max_slots(struct rpc_xprt *xprt);
/*
* Determine if a shared backchannel is in use
u32 prognumber; /* overrides program->number */
u32 version;
rpc_authflavor_t authflavor;
+ u32 nconnect;
unsigned long flags;
char *client_name;
struct svc_xprt *bc_xprt; /* NFSv4.1 backchannel */
void rpc_release_client(struct rpc_clnt *);
void rpc_task_release_transport(struct rpc_task *);
void rpc_task_release_client(struct rpc_task *);
+struct rpc_xprt *rpc_task_get_xprt(struct rpc_clnt *clnt,
+ struct rpc_xprt *xprt);
int rpcb_create_local(struct net *);
void rpcb_put_local(struct net *);
struct net * rpc_net_ns(struct rpc_clnt *);
size_t rpc_max_payload(struct rpc_clnt *);
size_t rpc_max_bc_payload(struct rpc_clnt *);
+unsigned int rpc_num_bc_slots(struct rpc_clnt *);
void rpc_force_rebind(struct rpc_clnt *);
size_t rpc_peeraddr(struct rpc_clnt *, struct sockaddr *, size_t);
const char *rpc_peeraddr2str(struct rpc_clnt *, enum rpc_display_format_t);
#include <linux/ktime.h>
#include <linux/spinlock.h>
-#define RPC_IOSTATS_VERS "1.0"
+#define RPC_IOSTATS_VERS "1.1"
struct rpc_iostats {
spinlock_t om_lock;
ktime_t om_queue, /* queued for xmit */
om_rtt, /* RPC RTT */
om_execute; /* RPC execution */
+ /*
+ * The count of operations that complete with tk_status < 0.
+ * These statuses usually indicate error conditions.
+ */
+ unsigned long om_error_status;
} ____cacheline_aligned;
struct rpc_task;
#define RPC_CALL_MAJORSEEN 0x0020 /* major timeout seen */
#define RPC_TASK_ROOTCREDS 0x0040 /* force root creds */
#define RPC_TASK_DYNAMIC 0x0080 /* task was kmalloc'ed */
+#define RPC_TASK_NO_ROUND_ROBIN 0x0100 /* send requests on "main" xprt */
#define RPC_TASK_SOFT 0x0200 /* Use soft timeouts */
#define RPC_TASK_SOFTCONN 0x0400 /* Fail if can't connect */
#define RPC_TASK_SENT 0x0800 /* message was sent */
#define RPC_NR_PRIORITY (1 + RPC_PRIORITY_PRIVILEGED - RPC_PRIORITY_LOW)
struct rpc_timer {
- struct timer_list timer;
struct list_head list;
+ unsigned long expires;
+ struct delayed_work dwork;
};
/*
int (*bc_setup)(struct rpc_xprt *xprt,
unsigned int min_reqs);
size_t (*bc_maxpayload)(struct rpc_xprt *xprt);
+ unsigned int (*bc_num_slots)(struct rpc_xprt *xprt);
void (*bc_free_rqst)(struct rpc_rqst *rqst);
void (*bc_destroy)(struct rpc_xprt *xprt,
unsigned int max_reqs);
/*
* Send stuff
*/
+ atomic_long_t queuelen;
spinlock_t transport_lock; /* lock transport info */
spinlock_t reserve_lock; /* lock slot table */
spinlock_t queue_lock; /* send/receive queue lock */
#if defined(CONFIG_SUNRPC_BACKCHANNEL)
struct svc_serv *bc_serv; /* The RPC service which will */
/* process the callback */
- int bc_alloc_count; /* Total number of preallocs */
- atomic_t bc_free_slots;
+ unsigned int bc_alloc_max;
+ unsigned int bc_alloc_count; /* Total number of preallocs */
+ atomic_t bc_slot_count; /* Number of allocated slots */
spinlock_t bc_pa_lock; /* Protects the preallocated
* items */
struct list_head bc_pa_list; /* List of preallocated
*/
struct rpc_xprt *xprt_create_transport(struct xprt_create *args);
void xprt_connect(struct rpc_task *task);
+unsigned long xprt_reconnect_delay(const struct rpc_xprt *xprt);
+void xprt_reconnect_backoff(struct rpc_xprt *xprt,
+ unsigned long init_to);
void xprt_reserve(struct rpc_task *task);
void xprt_retry_reserve(struct rpc_task *task);
int xprt_reserve_xprt(struct rpc_xprt *xprt, struct rpc_task *task);
struct kref xps_kref;
unsigned int xps_nxprts;
+ unsigned int xps_nactive;
+ atomic_long_t xps_queuelen;
struct list_head xps_xprt_list;
struct net * xps_net;
*/
unsigned long sock_state;
struct delayed_work connect_worker;
+ struct work_struct error_worker;
struct work_struct recv_worker;
struct mutex recv_mutex;
struct sockaddr_storage srcaddr;
#define XPRT_SOCK_CONNECTING 1U
#define XPRT_SOCK_DATA_READY (2)
#define XPRT_SOCK_UPD_TIMEOUT (3)
+#define XPRT_SOCK_WAKE_ERROR (4)
+#define XPRT_SOCK_WAKE_WRITE (5)
+#define XPRT_SOCK_WAKE_PENDING (6)
+#define XPRT_SOCK_WAKE_DISCONNECT (7)
#endif /* __KERNEL__ */
), \
TP_ARGS(task, mr, nsegs))
-TRACE_DEFINE_ENUM(FRWR_IS_INVALID);
-TRACE_DEFINE_ENUM(FRWR_IS_VALID);
-TRACE_DEFINE_ENUM(FRWR_FLUSHED_FR);
-TRACE_DEFINE_ENUM(FRWR_FLUSHED_LI);
-
-#define xprtrdma_show_frwr_state(x) \
- __print_symbolic(x, \
- { FRWR_IS_INVALID, "INVALID" }, \
- { FRWR_IS_VALID, "VALID" }, \
- { FRWR_FLUSHED_FR, "FLUSHED_FR" }, \
- { FRWR_FLUSHED_LI, "FLUSHED_LI" })
-
DECLARE_EVENT_CLASS(xprtrdma_frwr_done,
TP_PROTO(
const struct ib_wc *wc,
TP_STRUCT__entry(
__field(const void *, mr)
- __field(unsigned int, state)
__field(unsigned int, status)
__field(unsigned int, vendor_err)
),
TP_fast_assign(
__entry->mr = container_of(frwr, struct rpcrdma_mr, frwr);
- __entry->state = frwr->fr_state;
__entry->status = wc->status;
__entry->vendor_err = __entry->status ? wc->vendor_err : 0;
),
TP_printk(
- "mr=%p state=%s: %s (%u/0x%x)",
- __entry->mr, xprtrdma_show_frwr_state(__entry->state),
- rdma_show_wc_status(__entry->status),
+ "mr=%p: %s (%u/0x%x)",
+ __entry->mr, rdma_show_wc_status(__entry->status),
__entry->status, __entry->vendor_err
)
);
DEFINE_RXPRT_EVENT(xprtrdma_op_close);
DEFINE_RXPRT_EVENT(xprtrdma_op_connect);
+TRACE_EVENT(xprtrdma_op_set_cto,
+ TP_PROTO(
+ const struct rpcrdma_xprt *r_xprt,
+ unsigned long connect,
+ unsigned long reconnect
+ ),
+
+ TP_ARGS(r_xprt, connect, reconnect),
+
+ TP_STRUCT__entry(
+ __field(const void *, r_xprt)
+ __field(unsigned long, connect)
+ __field(unsigned long, reconnect)
+ __string(addr, rpcrdma_addrstr(r_xprt))
+ __string(port, rpcrdma_portstr(r_xprt))
+ ),
+
+ TP_fast_assign(
+ __entry->r_xprt = r_xprt;
+ __entry->connect = connect;
+ __entry->reconnect = reconnect;
+ __assign_str(addr, rpcrdma_addrstr(r_xprt));
+ __assign_str(port, rpcrdma_portstr(r_xprt));
+ ),
+
+ TP_printk("peer=[%s]:%s r_xprt=%p: connect=%lu reconnect=%lu",
+ __get_str(addr), __get_str(port), __entry->r_xprt,
+ __entry->connect / HZ, __entry->reconnect / HZ
+ )
+);
+
TRACE_EVENT(xprtrdma_qp_event,
TP_PROTO(
const struct rpcrdma_xprt *r_xprt,
TRACE_EVENT(xprtrdma_marshal,
TP_PROTO(
- const struct rpc_rqst *rqst,
- unsigned int hdrlen,
+ const struct rpcrdma_req *req,
unsigned int rtype,
unsigned int wtype
),
- TP_ARGS(rqst, hdrlen, rtype, wtype),
+ TP_ARGS(req, rtype, wtype),
TP_STRUCT__entry(
__field(unsigned int, task_id)
),
TP_fast_assign(
+ const struct rpc_rqst *rqst = &req->rl_slot;
+
__entry->task_id = rqst->rq_task->tk_pid;
__entry->client_id = rqst->rq_task->tk_client->cl_clid;
__entry->xid = be32_to_cpu(rqst->rq_xid);
- __entry->hdrlen = hdrlen;
+ __entry->hdrlen = req->rl_hdrbuf.len;
__entry->headlen = rqst->rq_snd_buf.head[0].iov_len;
__entry->pagelen = rqst->rq_snd_buf.page_len;
__entry->taillen = rqst->rq_snd_buf.tail[0].iov_len;
)
);
+TRACE_EVENT(xprtrdma_prepsend_failed,
+ TP_PROTO(const struct rpc_rqst *rqst,
+ int ret
+ ),
+
+ TP_ARGS(rqst, ret),
+
+ TP_STRUCT__entry(
+ __field(unsigned int, task_id)
+ __field(unsigned int, client_id)
+ __field(u32, xid)
+ __field(int, ret)
+ ),
+
+ TP_fast_assign(
+ __entry->task_id = rqst->rq_task->tk_pid;
+ __entry->client_id = rqst->rq_task->tk_client->cl_clid;
+ __entry->xid = be32_to_cpu(rqst->rq_xid);
+ __entry->ret = ret;
+ ),
+
+ TP_printk("task:%u@%u xid=0x%08x: ret=%d",
+ __entry->task_id, __entry->client_id, __entry->xid,
+ __entry->ret
+ )
+);
+
TRACE_EVENT(xprtrdma_post_send,
TP_PROTO(
const struct rpcrdma_req *req,
const struct rpc_rqst *rqst = &req->rl_slot;
__entry->task_id = rqst->rq_task->tk_pid;
- __entry->client_id = rqst->rq_task->tk_client->cl_clid;
+ __entry->client_id = rqst->rq_task->tk_client ?
+ rqst->rq_task->tk_client->cl_clid : -1;
__entry->req = req;
__entry->num_sge = req->rl_sendctx->sc_wr.num_sge;
__entry->signaled = req->rl_sendctx->sc_wr.send_flags &
DEFINE_FRWR_DONE_EVENT(xprtrdma_wc_fastreg);
DEFINE_FRWR_DONE_EVENT(xprtrdma_wc_li);
DEFINE_FRWR_DONE_EVENT(xprtrdma_wc_li_wake);
+DEFINE_FRWR_DONE_EVENT(xprtrdma_wc_li_done);
TRACE_EVENT(xprtrdma_frwr_alloc,
TP_PROTO(
If unsure, say Y.
-config CONFIG_SUNRPC_DISABLE_INSECURE_ENCTYPES
+config SUNRPC_DISABLE_INSECURE_ENCTYPES
bool "Secure RPC: Disable insecure Kerberos encryption types"
depends on RPCSEC_GSS_KRB5
default n
#define RPCDBG_FACILITY RPCDBG_TRANS
#endif
+#define BC_MAX_SLOTS 64U
+
+unsigned int xprt_bc_max_slots(struct rpc_xprt *xprt)
+{
+ return BC_MAX_SLOTS;
+}
+
/*
* Helper routines that track the number of preallocation elements
* on the transport.
*/
static inline int xprt_need_to_requeue(struct rpc_xprt *xprt)
{
- return xprt->bc_alloc_count < atomic_read(&xprt->bc_free_slots);
-}
-
-static inline void xprt_inc_alloc_count(struct rpc_xprt *xprt, unsigned int n)
-{
- atomic_add(n, &xprt->bc_free_slots);
- xprt->bc_alloc_count += n;
-}
-
-static inline int xprt_dec_alloc_count(struct rpc_xprt *xprt, unsigned int n)
-{
- atomic_sub(n, &xprt->bc_free_slots);
- return xprt->bc_alloc_count -= n;
+ return xprt->bc_alloc_count < xprt->bc_alloc_max;
}
/*
dprintk("RPC: setup backchannel transport\n");
+ if (min_reqs > BC_MAX_SLOTS)
+ min_reqs = BC_MAX_SLOTS;
+
/*
* We use a temporary list to keep track of the preallocated
* buffers. Once we're done building the list we splice it
*/
spin_lock(&xprt->bc_pa_lock);
list_splice(&tmp_list, &xprt->bc_pa_list);
- xprt_inc_alloc_count(xprt, min_reqs);
+ xprt->bc_alloc_count += min_reqs;
+ xprt->bc_alloc_max += min_reqs;
+ atomic_add(min_reqs, &xprt->bc_slot_count);
spin_unlock(&xprt->bc_pa_lock);
dprintk("RPC: setup backchannel transport done\n");
goto out;
spin_lock_bh(&xprt->bc_pa_lock);
- xprt_dec_alloc_count(xprt, max_reqs);
+ xprt->bc_alloc_max -= max_reqs;
list_for_each_entry_safe(req, tmp, &xprt->bc_pa_list, rq_bc_pa_list) {
dprintk("RPC: req=%p\n", req);
list_del(&req->rq_bc_pa_list);
xprt_free_allocation(req);
+ xprt->bc_alloc_count--;
+ atomic_dec(&xprt->bc_slot_count);
if (--max_reqs == 0)
break;
}
struct rpc_rqst *req = NULL;
dprintk("RPC: allocate a backchannel request\n");
- if (atomic_read(&xprt->bc_free_slots) <= 0)
- goto not_found;
if (list_empty(&xprt->bc_pa_list)) {
if (!new)
goto not_found;
+ if (atomic_read(&xprt->bc_slot_count) >= BC_MAX_SLOTS)
+ goto not_found;
list_add_tail(&new->rq_bc_pa_list, &xprt->bc_pa_list);
xprt->bc_alloc_count++;
+ atomic_inc(&xprt->bc_slot_count);
}
req = list_first_entry(&xprt->bc_pa_list, struct rpc_rqst,
rq_bc_pa_list);
if (xprt_need_to_requeue(xprt)) {
list_add_tail(&req->rq_bc_pa_list, &xprt->bc_pa_list);
xprt->bc_alloc_count++;
+ atomic_inc(&xprt->bc_slot_count);
req = NULL;
}
spin_unlock_bh(&xprt->bc_pa_lock);
spin_lock(&xprt->bc_pa_lock);
list_del(&req->rq_bc_pa_list);
- xprt_dec_alloc_count(xprt, 1);
+ xprt->bc_alloc_count--;
spin_unlock(&xprt->bc_pa_lock);
req->rq_private_buf.len = copied;
.bc_xprt = args->bc_xprt,
};
char servername[48];
+ struct rpc_clnt *clnt;
+ int i;
if (args->bc_xprt) {
WARN_ON_ONCE(!(args->protocol & XPRT_TRANSPORT_BC));
if (args->flags & RPC_CLNT_CREATE_NONPRIVPORT)
xprt->resvport = 0;
- return rpc_create_xprt(args, xprt);
+ clnt = rpc_create_xprt(args, xprt);
+ if (IS_ERR(clnt) || args->nconnect <= 1)
+ return clnt;
+
+ for (i = 0; i < args->nconnect - 1; i++) {
+ if (rpc_clnt_add_xprt(clnt, &xprtargs, NULL, NULL) < 0)
+ break;
+ }
+ return clnt;
}
EXPORT_SYMBOL_GPL(rpc_create);
}
EXPORT_SYMBOL_GPL(rpc_bind_new_program);
+struct rpc_xprt *
+rpc_task_get_xprt(struct rpc_clnt *clnt, struct rpc_xprt *xprt)
+{
+ struct rpc_xprt_switch *xps;
+
+ if (!xprt)
+ return NULL;
+ rcu_read_lock();
+ xps = rcu_dereference(clnt->cl_xpi.xpi_xpswitch);
+ atomic_long_inc(&xps->xps_queuelen);
+ rcu_read_unlock();
+ atomic_long_inc(&xprt->queuelen);
+
+ return xprt;
+}
+
+static void
+rpc_task_release_xprt(struct rpc_clnt *clnt, struct rpc_xprt *xprt)
+{
+ struct rpc_xprt_switch *xps;
+
+ atomic_long_dec(&xprt->queuelen);
+ rcu_read_lock();
+ xps = rcu_dereference(clnt->cl_xpi.xpi_xpswitch);
+ atomic_long_dec(&xps->xps_queuelen);
+ rcu_read_unlock();
+
+ xprt_put(xprt);
+}
+
void rpc_task_release_transport(struct rpc_task *task)
{
struct rpc_xprt *xprt = task->tk_xprt;
if (xprt) {
task->tk_xprt = NULL;
- xprt_put(xprt);
+ if (task->tk_client)
+ rpc_task_release_xprt(task->tk_client, xprt);
+ else
+ xprt_put(xprt);
}
}
EXPORT_SYMBOL_GPL(rpc_task_release_transport);
{
struct rpc_clnt *clnt = task->tk_client;
+ rpc_task_release_transport(task);
if (clnt != NULL) {
/* Remove from client task list */
spin_lock(&clnt->cl_lock);
rpc_release_client(clnt);
}
- rpc_task_release_transport(task);
+}
+
+static struct rpc_xprt *
+rpc_task_get_first_xprt(struct rpc_clnt *clnt)
+{
+ struct rpc_xprt *xprt;
+
+ rcu_read_lock();
+ xprt = xprt_get(rcu_dereference(clnt->cl_xprt));
+ rcu_read_unlock();
+ return rpc_task_get_xprt(clnt, xprt);
+}
+
+static struct rpc_xprt *
+rpc_task_get_next_xprt(struct rpc_clnt *clnt)
+{
+ return rpc_task_get_xprt(clnt, xprt_iter_get_next(&clnt->cl_xpi));
}
static
void rpc_task_set_transport(struct rpc_task *task, struct rpc_clnt *clnt)
{
- if (!task->tk_xprt)
- task->tk_xprt = xprt_iter_get_next(&clnt->cl_xpi);
+ if (task->tk_xprt)
+ return;
+ if (task->tk_flags & RPC_TASK_NO_ROUND_ROBIN)
+ task->tk_xprt = rpc_task_get_first_xprt(clnt);
+ else
+ task->tk_xprt = rpc_task_get_next_xprt(clnt);
}
static
}
EXPORT_SYMBOL_GPL(rpc_max_bc_payload);
+unsigned int rpc_num_bc_slots(struct rpc_clnt *clnt)
+{
+ struct rpc_xprt *xprt;
+ unsigned int ret;
+
+ rcu_read_lock();
+ xprt = rcu_dereference(clnt->cl_xprt);
+ ret = xprt->ops->bc_num_slots(xprt);
+ rcu_read_unlock();
+ return ret;
+}
+EXPORT_SYMBOL_GPL(rpc_num_bc_slots);
+
/**
* rpc_force_rebind - force transport to check that remote port is unchanged
* @clnt: client to rebind
req->rq_snd_buf.head[0].iov_len = 0;
xdr_init_encode(&xdr, &req->rq_snd_buf,
req->rq_snd_buf.head[0].iov_base, req);
+ xdr_free_bvec(&req->rq_snd_buf);
if (rpc_encode_header(task, &xdr))
return;
rpc_call_rpcerror(task, task->tk_status);
}
return;
- } else {
- xprt_request_prepare(task->tk_rqstp);
}
/* Add task to reply queue before transmission to avoid races */
return -ENOMEM;
data->xps = xprt_switch_get(xps);
data->xprt = xprt_get(xprt);
+ if (rpc_xprt_switch_has_addr(data->xps, (struct sockaddr *)&xprt->addr)) {
+ rpc_cb_add_xprt_release(data);
+ goto success;
+ }
task = rpc_call_null_helper(clnt, xprt, NULL,
RPC_TASK_SOFT|RPC_TASK_SOFTCONN|RPC_TASK_ASYNC|RPC_TASK_NULLCREDS,
if (IS_ERR(task))
return PTR_ERR(task);
rpc_put_task(task);
+success:
return 1;
}
EXPORT_SYMBOL_GPL(rpc_clnt_test_and_add_xprt);
// SPDX-License-Identifier: GPL-2.0
-/**
+/*
* debugfs interface for sunrpc
*
* (c) 2014 Jeff Layton <jlayton@primarydata.com>
.release = tasks_release,
};
+static int do_xprt_debugfs(struct rpc_clnt *clnt, struct rpc_xprt *xprt, void *numv)
+{
+ int len;
+ char name[24]; /* enough for "../../rpc_xprt/ + 8 hex digits + NULL */
+ char link[9]; /* enough for 8 hex digits + NULL */
+ int *nump = numv;
+
+ if (IS_ERR_OR_NULL(xprt->debugfs))
+ return 0;
+ len = snprintf(name, sizeof(name), "../../rpc_xprt/%s",
+ xprt->debugfs->d_name.name);
+ if (len > sizeof(name))
+ return -1;
+ if (*nump == 0)
+ strcpy(link, "xprt");
+ else {
+ len = snprintf(link, sizeof(link), "xprt%d", *nump);
+ if (len > sizeof(link))
+ return -1;
+ }
+ debugfs_create_symlink(link, clnt->cl_debugfs, name);
+ (*nump)++;
+ return 0;
+}
+
void
rpc_clnt_debugfs_register(struct rpc_clnt *clnt)
{
int len;
- char name[24]; /* enough for "../../rpc_xprt/ + 8 hex digits + NULL */
- struct rpc_xprt *xprt;
+ char name[9]; /* enough for 8 hex digits + NULL */
+ int xprtnum = 0;
len = snprintf(name, sizeof(name), "%x", clnt->cl_clid);
if (len >= sizeof(name))
debugfs_create_file("tasks", S_IFREG | 0400, clnt->cl_debugfs, clnt,
&tasks_fops);
- rcu_read_lock();
- xprt = rcu_dereference(clnt->cl_xprt);
- /* no "debugfs" dentry? Don't bother with the symlink. */
- if (IS_ERR_OR_NULL(xprt->debugfs)) {
- rcu_read_unlock();
- return;
- }
- len = snprintf(name, sizeof(name), "../../rpc_xprt/%s",
- xprt->debugfs->d_name.name);
- rcu_read_unlock();
-
- if (len >= sizeof(name))
- goto out_err;
-
- debugfs_create_symlink("xprt", clnt->cl_debugfs, name);
-
- return;
-out_err:
- debugfs_remove_recursive(clnt->cl_debugfs);
- clnt->cl_debugfs = NULL;
+ rpc_clnt_iterate_for_each_xprt(clnt, do_xprt_debugfs, &xprtnum);
}
void
#include <linux/sched/mm.h>
#include <linux/sunrpc/clnt.h>
+#include <linux/sunrpc/metrics.h>
#include "sunrpc.h"
static void rpc_async_schedule(struct work_struct *);
static void rpc_release_task(struct rpc_task *task);
-static void __rpc_queue_timer_fn(struct timer_list *t);
+static void __rpc_queue_timer_fn(struct work_struct *);
/*
* RPC tasks sit here while waiting for conditions to improve.
*/
struct workqueue_struct *rpciod_workqueue __read_mostly;
struct workqueue_struct *xprtiod_workqueue __read_mostly;
+EXPORT_SYMBOL_GPL(xprtiod_workqueue);
unsigned long
rpc_task_timeout(const struct rpc_task *task)
task->tk_timeout = 0;
list_del(&task->u.tk_wait.timer_list);
if (list_empty(&queue->timer_list.list))
- del_timer(&queue->timer_list.timer);
+ cancel_delayed_work(&queue->timer_list.dwork);
}
static void
rpc_set_queue_timer(struct rpc_wait_queue *queue, unsigned long expires)
{
- timer_reduce(&queue->timer_list.timer, expires);
+ unsigned long now = jiffies;
+ queue->timer_list.expires = expires;
+ if (time_before_eq(expires, now))
+ expires = 0;
+ else
+ expires -= now;
+ mod_delayed_work(rpciod_workqueue, &queue->timer_list.dwork, expires);
}
/*
task->tk_pid, jiffies_to_msecs(timeout - jiffies));
task->tk_timeout = timeout;
- rpc_set_queue_timer(queue, timeout);
+ if (list_empty(&queue->timer_list.list) || time_before(timeout, queue->timer_list.expires))
+ rpc_set_queue_timer(queue, timeout);
list_add(&task->u.tk_wait.timer_list, &queue->timer_list.list);
}
queue->maxpriority = nr_queues - 1;
rpc_reset_waitqueue_priority(queue);
queue->qlen = 0;
- timer_setup(&queue->timer_list.timer, __rpc_queue_timer_fn, 0);
+ queue->timer_list.expires = 0;
+ INIT_DEFERRABLE_WORK(&queue->timer_list.dwork, __rpc_queue_timer_fn);
INIT_LIST_HEAD(&queue->timer_list.list);
rpc_assign_waitqueue_name(queue, qname);
}
void rpc_destroy_wait_queue(struct rpc_wait_queue *queue)
{
- del_timer_sync(&queue->timer_list.timer);
+ cancel_delayed_work_sync(&queue->timer_list.dwork);
}
EXPORT_SYMBOL_GPL(rpc_destroy_wait_queue);
/*
* Protect the queue operations.
*/
- spin_lock_bh(&q->lock);
+ spin_lock(&q->lock);
__rpc_sleep_on_priority_timeout(q, task, timeout, task->tk_priority);
- spin_unlock_bh(&q->lock);
+ spin_unlock(&q->lock);
}
EXPORT_SYMBOL_GPL(rpc_sleep_on_timeout);
/*
* Protect the queue operations.
*/
- spin_lock_bh(&q->lock);
+ spin_lock(&q->lock);
__rpc_sleep_on_priority(q, task, task->tk_priority);
- spin_unlock_bh(&q->lock);
+ spin_unlock(&q->lock);
}
EXPORT_SYMBOL_GPL(rpc_sleep_on);
/*
* Protect the queue operations.
*/
- spin_lock_bh(&q->lock);
+ spin_lock(&q->lock);
__rpc_sleep_on_priority_timeout(q, task, timeout, priority);
- spin_unlock_bh(&q->lock);
+ spin_unlock(&q->lock);
}
EXPORT_SYMBOL_GPL(rpc_sleep_on_priority_timeout);
/*
* Protect the queue operations.
*/
- spin_lock_bh(&q->lock);
+ spin_lock(&q->lock);
__rpc_sleep_on_priority(q, task, priority);
- spin_unlock_bh(&q->lock);
+ spin_unlock(&q->lock);
}
EXPORT_SYMBOL_GPL(rpc_sleep_on_priority);
{
if (!RPC_IS_QUEUED(task))
return;
- spin_lock_bh(&queue->lock);
+ spin_lock(&queue->lock);
rpc_wake_up_task_on_wq_queue_locked(wq, queue, task);
- spin_unlock_bh(&queue->lock);
+ spin_unlock(&queue->lock);
}
/*
{
if (!RPC_IS_QUEUED(task))
return;
- spin_lock_bh(&queue->lock);
+ spin_lock(&queue->lock);
rpc_wake_up_task_queue_locked(queue, task);
- spin_unlock_bh(&queue->lock);
+ spin_unlock(&queue->lock);
}
EXPORT_SYMBOL_GPL(rpc_wake_up_queued_task);
{
if (!RPC_IS_QUEUED(task))
return;
- spin_lock_bh(&queue->lock);
+ spin_lock(&queue->lock);
rpc_wake_up_task_queue_set_status_locked(queue, task, status);
- spin_unlock_bh(&queue->lock);
+ spin_unlock(&queue->lock);
}
/*
dprintk("RPC: wake_up_first(%p \"%s\")\n",
queue, rpc_qname(queue));
- spin_lock_bh(&queue->lock);
+ spin_lock(&queue->lock);
task = __rpc_find_next_queued(queue);
if (task != NULL)
task = rpc_wake_up_task_on_wq_queue_action_locked(wq, queue,
task, func, data);
- spin_unlock_bh(&queue->lock);
+ spin_unlock(&queue->lock);
return task;
}
{
struct list_head *head;
- spin_lock_bh(&queue->lock);
+ spin_lock(&queue->lock);
head = &queue->tasks[queue->maxpriority];
for (;;) {
while (!list_empty(head)) {
break;
head--;
}
- spin_unlock_bh(&queue->lock);
+ spin_unlock(&queue->lock);
}
EXPORT_SYMBOL_GPL(rpc_wake_up);
{
struct list_head *head;
- spin_lock_bh(&queue->lock);
+ spin_lock(&queue->lock);
head = &queue->tasks[queue->maxpriority];
for (;;) {
while (!list_empty(head)) {
break;
head--;
}
- spin_unlock_bh(&queue->lock);
+ spin_unlock(&queue->lock);
}
EXPORT_SYMBOL_GPL(rpc_wake_up_status);
-static void __rpc_queue_timer_fn(struct timer_list *t)
+static void __rpc_queue_timer_fn(struct work_struct *work)
{
- struct rpc_wait_queue *queue = from_timer(queue, t, timer_list.timer);
+ struct rpc_wait_queue *queue = container_of(work,
+ struct rpc_wait_queue,
+ timer_list.dwork.work);
struct rpc_task *task, *n;
unsigned long expires, now, timeo;
void rpc_exit_task(struct rpc_task *task)
{
task->tk_action = NULL;
+ if (task->tk_ops->rpc_count_stats)
+ task->tk_ops->rpc_count_stats(task, task->tk_calldata);
+ else if (task->tk_client)
+ rpc_count_iostats(task, task->tk_client->cl_metrics);
if (task->tk_ops->rpc_call_done != NULL) {
task->tk_ops->rpc_call_done(task, task->tk_calldata);
if (task->tk_action != NULL) {
* rpc_task pointer may still be dereferenced.
*/
queue = task->tk_waitqueue;
- spin_lock_bh(&queue->lock);
+ spin_lock(&queue->lock);
if (!RPC_IS_QUEUED(task)) {
- spin_unlock_bh(&queue->lock);
+ spin_unlock(&queue->lock);
continue;
}
rpc_clear_running(task);
- spin_unlock_bh(&queue->lock);
+ spin_unlock(&queue->lock);
if (task_is_async)
return;
/* Initialize workqueue for async tasks */
task->tk_workqueue = task_setup_data->workqueue;
- task->tk_xprt = xprt_get(task_setup_data->rpc_xprt);
+ task->tk_xprt = rpc_task_get_xprt(task_setup_data->rpc_client,
+ xprt_get(task_setup_data->rpc_xprt));
task->tk_op_cred = get_rpccred(task_setup_data->rpc_op_cred);
execute = ktime_sub(now, task->tk_start);
op_metrics->om_execute = ktime_add(op_metrics->om_execute, execute);
+ if (task->tk_status < 0)
+ op_metrics->om_error_status++;
spin_unlock(&op_metrics->om_lock);
a->om_queue = ktime_add(a->om_queue, b->om_queue);
a->om_rtt = ktime_add(a->om_rtt, b->om_rtt);
a->om_execute = ktime_add(a->om_execute, b->om_execute);
+ a->om_error_status += b->om_error_status;
}
static void _print_rpc_iostats(struct seq_file *seq, struct rpc_iostats *stats,
int op, const struct rpc_procinfo *procs)
{
_print_name(seq, op, procs);
- seq_printf(seq, "%lu %lu %lu %Lu %Lu %Lu %Lu %Lu\n",
+ seq_printf(seq, "%lu %lu %lu %llu %llu %llu %llu %llu %lu\n",
stats->om_ops,
stats->om_ntrans,
stats->om_timeouts,
stats->om_bytes_recv,
ktime_to_ms(stats->om_queue),
ktime_to_ms(stats->om_rtt),
- ktime_to_ms(stats->om_execute));
+ ktime_to_ms(stats->om_execute),
+ stats->om_error_status);
+}
+
+static int do_print_stats(struct rpc_clnt *clnt, struct rpc_xprt *xprt, void *seqv)
+{
+ struct seq_file *seq = seqv;
+
+ xprt->ops->print_stats(xprt, seq);
+ return 0;
}
void rpc_clnt_show_stats(struct seq_file *seq, struct rpc_clnt *clnt)
{
- struct rpc_xprt *xprt;
unsigned int op, maxproc = clnt->cl_maxproc;
if (!clnt->cl_metrics)
seq_printf(seq, "p/v: %u/%u (%s)\n",
clnt->cl_prog, clnt->cl_vers, clnt->cl_program->name);
- rcu_read_lock();
- xprt = rcu_dereference(clnt->cl_xprt);
- if (xprt)
- xprt->ops->print_stats(xprt, seq);
- rcu_read_unlock();
+ rpc_clnt_iterate_for_each_xprt(clnt, do_print_stats, seq);
seq_printf(seq, "\tper-op statistics\n");
for (op = 0; op < maxproc; op++) {
/* Parse and execute the bc call */
proc_error = svc_process_common(rqstp, argv, resv);
- atomic_inc(&req->rq_xprt->bc_free_slots);
+ atomic_dec(&req->rq_xprt->bc_slot_count);
if (!proc_error) {
/* Processing error: drop the request */
xprt_free_bc_request(req);
if (test_bit(XPRT_LOCKED, &xprt->state) && xprt->snd_task == task)
return 1;
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
retval = xprt->ops->reserve_xprt(xprt, task);
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
return retval;
}
{
if (xprt->snd_task != task)
return;
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
xprt->ops->release_xprt(xprt, task);
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
}
/*
if (req->rq_cong)
return true;
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
ret = __xprt_get_cong(xprt, req) != 0;
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
return ret;
}
EXPORT_SYMBOL_GPL(xprt_request_get_cong);
xprt_clear_congestion_window_wait(struct rpc_xprt *xprt)
{
if (test_and_clear_bit(XPRT_CWND_WAIT, &xprt->state)) {
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
__xprt_lock_write_next_cong(xprt);
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
}
}
if (!test_bit(XPRT_WRITE_SPACE, &xprt->state))
return false;
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
ret = xprt_clear_write_space_locked(xprt);
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
return ret;
}
EXPORT_SYMBOL_GPL(xprt_write_space);
req->rq_retries = 0;
xprt_reset_majortimeo(req);
/* Reset the RTT counters == "slow start" */
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
rpc_init_rtt(req->rq_task->tk_client->cl_rtt, to->to_initval);
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
status = -ETIMEDOUT;
}
void xprt_disconnect_done(struct rpc_xprt *xprt)
{
dprintk("RPC: disconnected transport %p\n", xprt);
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
xprt_clear_connected(xprt);
xprt_clear_write_space_locked(xprt);
xprt_wake_pending_tasks(xprt, -ENOTCONN);
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
}
EXPORT_SYMBOL_GPL(xprt_disconnect_done);
void xprt_force_disconnect(struct rpc_xprt *xprt)
{
/* Don't race with the test_bit() in xprt_clear_locked() */
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
set_bit(XPRT_CLOSE_WAIT, &xprt->state);
/* Try to schedule an autoclose RPC call */
if (test_and_set_bit(XPRT_LOCKED, &xprt->state) == 0)
else if (xprt->snd_task)
rpc_wake_up_queued_task_set_status(&xprt->pending,
xprt->snd_task, -ENOTCONN);
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
}
EXPORT_SYMBOL_GPL(xprt_force_disconnect);
void xprt_conditional_disconnect(struct rpc_xprt *xprt, unsigned int cookie)
{
/* Don't race with the test_bit() in xprt_clear_locked() */
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
if (cookie != xprt->connect_cookie)
goto out;
if (test_bit(XPRT_CLOSING, &xprt->state))
queue_work(xprtiod_workqueue, &xprt->task_cleanup);
xprt_wake_pending_tasks(xprt, -EAGAIN);
out:
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
}
static bool
xprt_schedule_autodisconnect(struct rpc_xprt *xprt)
__must_hold(&xprt->transport_lock)
{
+ xprt->last_used = jiffies;
if (RB_EMPTY_ROOT(&xprt->recv_queue) && xprt_has_timer(xprt))
mod_timer(&xprt->timer, xprt->last_used + xprt->idle_timeout);
}
{
struct rpc_xprt *xprt = from_timer(xprt, t, timer);
- spin_lock(&xprt->transport_lock);
if (!RB_EMPTY_ROOT(&xprt->recv_queue))
- goto out_abort;
+ return;
/* Reset xprt->last_used to avoid connect/autodisconnect cycling */
xprt->last_used = jiffies;
if (test_and_set_bit(XPRT_LOCKED, &xprt->state))
- goto out_abort;
- spin_unlock(&xprt->transport_lock);
+ return;
queue_work(xprtiod_workqueue, &xprt->task_cleanup);
- return;
-out_abort:
- spin_unlock(&xprt->transport_lock);
}
bool xprt_lock_connect(struct rpc_xprt *xprt,
{
bool ret = false;
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
if (!test_bit(XPRT_LOCKED, &xprt->state))
goto out;
if (xprt->snd_task != task)
xprt->snd_task = cookie;
ret = true;
out:
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
return ret;
}
void xprt_unlock_connect(struct rpc_xprt *xprt, void *cookie)
{
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
if (xprt->snd_task != cookie)
goto out;
if (!test_bit(XPRT_LOCKED, &xprt->state))
xprt->ops->release_xprt(xprt, NULL);
xprt_schedule_autodisconnect(xprt);
out:
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
wake_up_bit(&xprt->state, XPRT_LOCKED);
}
xprt_release_write(xprt, task);
}
+/**
+ * xprt_reconnect_delay - compute the wait before scheduling a connect
+ * @xprt: transport instance
+ *
+ */
+unsigned long xprt_reconnect_delay(const struct rpc_xprt *xprt)
+{
+ unsigned long start, now = jiffies;
+
+ start = xprt->stat.connect_start + xprt->reestablish_timeout;
+ if (time_after(start, now))
+ return start - now;
+ return 0;
+}
+EXPORT_SYMBOL_GPL(xprt_reconnect_delay);
+
+/**
+ * xprt_reconnect_backoff - compute the new re-establish timeout
+ * @xprt: transport instance
+ * @init_to: initial reestablish timeout
+ *
+ */
+void xprt_reconnect_backoff(struct rpc_xprt *xprt, unsigned long init_to)
+{
+ xprt->reestablish_timeout <<= 1;
+ if (xprt->reestablish_timeout > xprt->max_reconnect_timeout)
+ xprt->reestablish_timeout = xprt->max_reconnect_timeout;
+ if (xprt->reestablish_timeout < init_to)
+ xprt->reestablish_timeout = init_to;
+}
+EXPORT_SYMBOL_GPL(xprt_reconnect_backoff);
+
enum xprt_xid_rb_cmp {
XID_RB_EQUAL,
XID_RB_LEFT,
if (!xprt_request_need_enqueue_receive(task, req))
return;
+
+ xprt_request_prepare(task->tk_rqstp);
spin_lock(&xprt->queue_lock);
/* Update the softirq receive buffer */
xprt_inject_disconnect(xprt);
task->tk_flags |= RPC_TASK_SENT;
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
xprt->stat.sends++;
xprt->stat.req_u += xprt->stat.sends - xprt->stat.recvs;
xprt->stat.bklog_u += xprt->backlog.qlen;
xprt->stat.sending_u += xprt->sending.qlen;
xprt->stat.pending_u += xprt->pending.qlen;
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
req->rq_connect_cookie = connect_cookie;
out_dequeue:
}
xprt = req->rq_xprt;
- if (task->tk_ops->rpc_count_stats != NULL)
- task->tk_ops->rpc_count_stats(task, task->tk_calldata);
- else if (task->tk_client)
- rpc_count_iostats(task, task->tk_client->cl_metrics);
xprt_request_dequeue_all(task, req);
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
xprt->ops->release_xprt(xprt, task);
if (xprt->ops->release_request)
xprt->ops->release_request(task);
- xprt->last_used = jiffies;
xprt_schedule_autodisconnect(xprt);
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
if (req->rq_buffer)
xprt->ops->buf_free(task);
xprt_inject_disconnect(xprt);
#include <linux/sunrpc/addr.h>
#include <linux/sunrpc/xprtmultipath.h>
-typedef struct rpc_xprt *(*xprt_switch_find_xprt_t)(struct list_head *head,
+typedef struct rpc_xprt *(*xprt_switch_find_xprt_t)(struct rpc_xprt_switch *xps,
const struct rpc_xprt *cur);
static const struct rpc_xprt_iter_ops rpc_xprt_iter_singular;
if (xps->xps_nxprts == 0)
xps->xps_net = xprt->xprt_net;
xps->xps_nxprts++;
+ xps->xps_nactive++;
}
/**
if (xprt == NULL)
return;
spin_lock(&xps->xps_lock);
- if ((xps->xps_net == xprt->xprt_net || xps->xps_net == NULL) &&
- !rpc_xprt_switch_has_addr(xps, (struct sockaddr *)&xprt->addr))
+ if (xps->xps_net == xprt->xprt_net || xps->xps_net == NULL)
xprt_switch_add_xprt_locked(xps, xprt);
spin_unlock(&xps->xps_lock);
}
{
if (unlikely(xprt == NULL))
return;
+ xps->xps_nactive--;
xps->xps_nxprts--;
if (xps->xps_nxprts == 0)
xps->xps_net = NULL;
if (xps != NULL) {
spin_lock_init(&xps->xps_lock);
kref_init(&xps->xps_kref);
- xps->xps_nxprts = 0;
+ xps->xps_nxprts = xps->xps_nactive = 0;
+ atomic_long_set(&xps->xps_queuelen, 0);
+ xps->xps_net = NULL;
INIT_LIST_HEAD(&xps->xps_xprt_list);
xps->xps_iter_ops = &rpc_xprt_iter_singular;
xprt_switch_add_xprt_locked(xps, xprt);
WRITE_ONCE(xpi->xpi_cursor, NULL);
}
+static
+bool xprt_is_active(const struct rpc_xprt *xprt)
+{
+ return kref_read(&xprt->kref) != 0;
+}
+
static
struct rpc_xprt *xprt_switch_find_first_entry(struct list_head *head)
{
- return list_first_or_null_rcu(head, struct rpc_xprt, xprt_switch);
+ struct rpc_xprt *pos;
+
+ list_for_each_entry_rcu(pos, head, xprt_switch) {
+ if (xprt_is_active(pos))
+ return pos;
+ }
+ return NULL;
}
static
const struct rpc_xprt *cur)
{
struct rpc_xprt *pos;
+ bool found = false;
list_for_each_entry_rcu(pos, head, xprt_switch) {
if (cur == pos)
+ found = true;
+ if (found && xprt_is_active(pos))
return pos;
}
return NULL;
const struct rpc_xprt *cur)
{
struct rpc_xprt *pos, *prev = NULL;
+ bool found = false;
list_for_each_entry_rcu(pos, head, xprt_switch) {
if (cur == prev)
+ found = true;
+ if (found && xprt_is_active(pos))
return pos;
prev = pos;
}
}
static
-struct rpc_xprt *xprt_switch_set_next_cursor(struct list_head *head,
+struct rpc_xprt *xprt_switch_set_next_cursor(struct rpc_xprt_switch *xps,
struct rpc_xprt **cursor,
xprt_switch_find_xprt_t find_next)
{
- struct rpc_xprt *cur, *pos, *old;
+ struct rpc_xprt *pos, *old;
- cur = READ_ONCE(*cursor);
- for (;;) {
- old = cur;
- pos = find_next(head, old);
- if (pos == NULL)
- break;
- cur = cmpxchg_relaxed(cursor, old, pos);
- if (cur == old)
- break;
- }
+ old = smp_load_acquire(cursor);
+ pos = find_next(xps, old);
+ smp_store_release(cursor, pos);
return pos;
}
if (xps == NULL)
return NULL;
- return xprt_switch_set_next_cursor(&xps->xps_xprt_list,
- &xpi->xpi_cursor,
- find_next);
+ return xprt_switch_set_next_cursor(xps, &xpi->xpi_cursor, find_next);
}
static
-struct rpc_xprt *xprt_switch_find_next_entry_roundrobin(struct list_head *head,
+struct rpc_xprt *__xprt_switch_find_next_entry_roundrobin(struct list_head *head,
const struct rpc_xprt *cur)
{
struct rpc_xprt *ret;
return xprt_switch_find_first_entry(head);
}
+static
+struct rpc_xprt *xprt_switch_find_next_entry_roundrobin(struct rpc_xprt_switch *xps,
+ const struct rpc_xprt *cur)
+{
+ struct list_head *head = &xps->xps_xprt_list;
+ struct rpc_xprt *xprt;
+ unsigned int nactive;
+
+ for (;;) {
+ unsigned long xprt_queuelen, xps_queuelen;
+
+ xprt = __xprt_switch_find_next_entry_roundrobin(head, cur);
+ if (!xprt)
+ break;
+ xprt_queuelen = atomic_long_read(&xprt->queuelen);
+ xps_queuelen = atomic_long_read(&xps->xps_queuelen);
+ nactive = READ_ONCE(xps->xps_nactive);
+ /* Exit loop if xprt_queuelen <= average queue length */
+ if (xprt_queuelen * nactive <= xps_queuelen)
+ break;
+ cur = xprt;
+ }
+ return xprt;
+}
+
static
struct rpc_xprt *xprt_iter_next_entry_roundrobin(struct rpc_xprt_iter *xpi)
{
xprt_switch_find_next_entry_roundrobin);
}
+static
+struct rpc_xprt *xprt_switch_find_next_entry_all(struct rpc_xprt_switch *xps,
+ const struct rpc_xprt *cur)
+{
+ return xprt_switch_find_next_entry(&xps->xps_xprt_list, cur);
+}
+
static
struct rpc_xprt *xprt_iter_next_entry_all(struct rpc_xprt_iter *xpi)
{
- return xprt_iter_next_entry_multiple(xpi, xprt_switch_find_next_entry);
+ return xprt_iter_next_entry_multiple(xpi,
+ xprt_switch_find_next_entry_all);
}
/*
return maxmsg - RPCRDMA_HDRLEN_MIN;
}
+unsigned int xprt_rdma_bc_max_slots(struct rpc_xprt *xprt)
+{
+ struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
+
+ return r_xprt->rx_buf.rb_bc_srv_max_requests;
+}
+
static int rpcrdma_bc_marshal_reply(struct rpc_rqst *rqst)
{
struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(rqst->rq_xprt);
frwr_release_mr(mr);
}
+/* frwr_reset - Place MRs back on the free list
+ * @req: request to reset
+ *
+ * Used after a failed marshal. For FRWR, this means the MRs
+ * don't have to be fully released and recreated.
+ *
+ * NB: This is safe only as long as none of @req's MRs are
+ * involved with an ongoing asynchronous FAST_REG or LOCAL_INV
+ * Work Request.
+ */
+void frwr_reset(struct rpcrdma_req *req)
+{
+ while (!list_empty(&req->rl_registered)) {
+ struct rpcrdma_mr *mr;
+
+ mr = rpcrdma_mr_pop(&req->rl_registered);
+ rpcrdma_mr_unmap_and_put(mr);
+ }
+}
+
/**
* frwr_init_mr - Initialize one MR
* @ia: interface adapter
goto out_list_err;
mr->frwr.fr_mr = frmr;
- mr->frwr.fr_state = FRWR_IS_INVALID;
mr->mr_dir = DMA_NONE;
INIT_LIST_HEAD(&mr->mr_list);
INIT_WORK(&mr->mr_recycle, frwr_mr_recycle_worker);
(ia->ri_max_segs - 2) * ia->ri_max_frwr_depth);
}
-/**
- * frwr_wc_fastreg - Invoked by RDMA provider for a flushed FastReg WC
- * @cq: completion queue (ignored)
- * @wc: completed WR
- *
- */
-static void
-frwr_wc_fastreg(struct ib_cq *cq, struct ib_wc *wc)
-{
- struct ib_cqe *cqe = wc->wr_cqe;
- struct rpcrdma_frwr *frwr =
- container_of(cqe, struct rpcrdma_frwr, fr_cqe);
-
- /* WARNING: Only wr_cqe and status are reliable at this point */
- if (wc->status != IB_WC_SUCCESS)
- frwr->fr_state = FRWR_FLUSHED_FR;
- trace_xprtrdma_wc_fastreg(wc, frwr);
-}
-
-/**
- * frwr_wc_localinv - Invoked by RDMA provider for a flushed LocalInv WC
- * @cq: completion queue (ignored)
- * @wc: completed WR
- *
- */
-static void
-frwr_wc_localinv(struct ib_cq *cq, struct ib_wc *wc)
-{
- struct ib_cqe *cqe = wc->wr_cqe;
- struct rpcrdma_frwr *frwr = container_of(cqe, struct rpcrdma_frwr,
- fr_cqe);
-
- /* WARNING: Only wr_cqe and status are reliable at this point */
- if (wc->status != IB_WC_SUCCESS)
- frwr->fr_state = FRWR_FLUSHED_LI;
- trace_xprtrdma_wc_li(wc, frwr);
-}
-
-/**
- * frwr_wc_localinv_wake - Invoked by RDMA provider for a signaled LocalInv WC
- * @cq: completion queue (ignored)
- * @wc: completed WR
- *
- * Awaken anyone waiting for an MR to finish being fenced.
- */
-static void
-frwr_wc_localinv_wake(struct ib_cq *cq, struct ib_wc *wc)
-{
- struct ib_cqe *cqe = wc->wr_cqe;
- struct rpcrdma_frwr *frwr = container_of(cqe, struct rpcrdma_frwr,
- fr_cqe);
-
- /* WARNING: Only wr_cqe and status are reliable at this point */
- if (wc->status != IB_WC_SUCCESS)
- frwr->fr_state = FRWR_FLUSHED_LI;
- trace_xprtrdma_wc_li_wake(wc, frwr);
- complete(&frwr->fr_linv_done);
-}
-
/**
* frwr_map - Register a memory region
* @r_xprt: controlling transport
{
struct rpcrdma_ia *ia = &r_xprt->rx_ia;
bool holes_ok = ia->ri_mrtype == IB_MR_TYPE_SG_GAPS;
- struct rpcrdma_frwr *frwr;
struct rpcrdma_mr *mr;
struct ib_mr *ibmr;
struct ib_reg_wr *reg_wr;
int i, n;
u8 key;
- mr = NULL;
- do {
- if (mr)
- rpcrdma_mr_recycle(mr);
- mr = rpcrdma_mr_get(r_xprt);
- if (!mr)
- return ERR_PTR(-EAGAIN);
- } while (mr->frwr.fr_state != FRWR_IS_INVALID);
- frwr = &mr->frwr;
- frwr->fr_state = FRWR_IS_VALID;
+ mr = rpcrdma_mr_get(r_xprt);
+ if (!mr)
+ goto out_getmr_err;
if (nsegs > ia->ri_max_frwr_depth)
nsegs = ia->ri_max_frwr_depth;
if (!mr->mr_nents)
goto out_dmamap_err;
- ibmr = frwr->fr_mr;
+ ibmr = mr->frwr.fr_mr;
n = ib_map_mr_sg(ibmr, mr->mr_sg, mr->mr_nents, NULL, PAGE_SIZE);
if (unlikely(n != mr->mr_nents))
goto out_mapmr_err;
key = (u8)(ibmr->rkey & 0x000000FF);
ib_update_fast_reg_key(ibmr, ++key);
- reg_wr = &frwr->fr_regwr;
+ reg_wr = &mr->frwr.fr_regwr;
reg_wr->mr = ibmr;
reg_wr->key = ibmr->rkey;
reg_wr->access = writing ?
*out = mr;
return seg;
+out_getmr_err:
+ xprt_wait_for_buffer_space(&r_xprt->rx_xprt);
+ return ERR_PTR(-EAGAIN);
+
out_dmamap_err:
mr->mr_dir = DMA_NONE;
trace_xprtrdma_frwr_sgerr(mr, i);
return ERR_PTR(-EIO);
}
+/**
+ * frwr_wc_fastreg - Invoked by RDMA provider for a flushed FastReg WC
+ * @cq: completion queue (ignored)
+ * @wc: completed WR
+ *
+ */
+static void frwr_wc_fastreg(struct ib_cq *cq, struct ib_wc *wc)
+{
+ struct ib_cqe *cqe = wc->wr_cqe;
+ struct rpcrdma_frwr *frwr =
+ container_of(cqe, struct rpcrdma_frwr, fr_cqe);
+
+ /* WARNING: Only wr_cqe and status are reliable at this point */
+ trace_xprtrdma_wc_fastreg(wc, frwr);
+ /* The MR will get recycled when the associated req is retransmitted */
+}
+
/**
* frwr_send - post Send WR containing the RPC Call message
* @ia: interface adapter
if (mr->mr_handle == rep->rr_inv_rkey) {
list_del_init(&mr->mr_list);
trace_xprtrdma_mr_remoteinv(mr);
- mr->frwr.fr_state = FRWR_IS_INVALID;
rpcrdma_mr_unmap_and_put(mr);
break; /* only one invalidated MR per RPC */
}
}
+static void __frwr_release_mr(struct ib_wc *wc, struct rpcrdma_mr *mr)
+{
+ if (wc->status != IB_WC_SUCCESS)
+ rpcrdma_mr_recycle(mr);
+ else
+ rpcrdma_mr_unmap_and_put(mr);
+}
+
/**
- * frwr_unmap_sync - invalidate memory regions that were registered for @req
- * @r_xprt: controlling transport
- * @mrs: list of MRs to process
+ * frwr_wc_localinv - Invoked by RDMA provider for a LOCAL_INV WC
+ * @cq: completion queue (ignored)
+ * @wc: completed WR
+ *
+ */
+static void frwr_wc_localinv(struct ib_cq *cq, struct ib_wc *wc)
+{
+ struct ib_cqe *cqe = wc->wr_cqe;
+ struct rpcrdma_frwr *frwr =
+ container_of(cqe, struct rpcrdma_frwr, fr_cqe);
+ struct rpcrdma_mr *mr = container_of(frwr, struct rpcrdma_mr, frwr);
+
+ /* WARNING: Only wr_cqe and status are reliable at this point */
+ trace_xprtrdma_wc_li(wc, frwr);
+ __frwr_release_mr(wc, mr);
+}
+
+/**
+ * frwr_wc_localinv_wake - Invoked by RDMA provider for a LOCAL_INV WC
+ * @cq: completion queue (ignored)
+ * @wc: completed WR
*
- * Sleeps until it is safe for the host CPU to access the
- * previously mapped memory regions.
+ * Awaken anyone waiting for an MR to finish being fenced.
+ */
+static void frwr_wc_localinv_wake(struct ib_cq *cq, struct ib_wc *wc)
+{
+ struct ib_cqe *cqe = wc->wr_cqe;
+ struct rpcrdma_frwr *frwr =
+ container_of(cqe, struct rpcrdma_frwr, fr_cqe);
+ struct rpcrdma_mr *mr = container_of(frwr, struct rpcrdma_mr, frwr);
+
+ /* WARNING: Only wr_cqe and status are reliable at this point */
+ trace_xprtrdma_wc_li_wake(wc, frwr);
+ complete(&frwr->fr_linv_done);
+ __frwr_release_mr(wc, mr);
+}
+
+/**
+ * frwr_unmap_sync - invalidate memory regions that were registered for @req
+ * @r_xprt: controlling transport instance
+ * @req: rpcrdma_req with a non-empty list of MRs to process
*
- * Caller ensures that @mrs is not empty before the call. This
- * function empties the list.
+ * Sleeps until it is safe for the host CPU to access the previously mapped
+ * memory regions. This guarantees that registered MRs are properly fenced
+ * from the server before the RPC consumer accesses the data in them. It
+ * also ensures proper Send flow control: waking the next RPC waits until
+ * this RPC has relinquished all its Send Queue entries.
*/
-void frwr_unmap_sync(struct rpcrdma_xprt *r_xprt, struct list_head *mrs)
+void frwr_unmap_sync(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req)
{
struct ib_send_wr *first, **prev, *last;
const struct ib_send_wr *bad_wr;
- struct rpcrdma_ia *ia = &r_xprt->rx_ia;
struct rpcrdma_frwr *frwr;
struct rpcrdma_mr *mr;
- int count, rc;
+ int rc;
/* ORDER: Invalidate all of the MRs first
*
* a single ib_post_send() call.
*/
frwr = NULL;
- count = 0;
prev = &first;
- list_for_each_entry(mr, mrs, mr_list) {
- mr->frwr.fr_state = FRWR_IS_INVALID;
+ while (!list_empty(&req->rl_registered)) {
+ mr = rpcrdma_mr_pop(&req->rl_registered);
- frwr = &mr->frwr;
trace_xprtrdma_mr_localinv(mr);
+ r_xprt->rx_stats.local_inv_needed++;
+ frwr = &mr->frwr;
frwr->fr_cqe.done = frwr_wc_localinv;
last = &frwr->fr_invwr;
- memset(last, 0, sizeof(*last));
+ last->next = NULL;
last->wr_cqe = &frwr->fr_cqe;
+ last->sg_list = NULL;
+ last->num_sge = 0;
last->opcode = IB_WR_LOCAL_INV;
+ last->send_flags = IB_SEND_SIGNALED;
last->ex.invalidate_rkey = mr->mr_handle;
- count++;
*prev = last;
prev = &last->next;
}
- if (!frwr)
- goto unmap;
/* Strong send queue ordering guarantees that when the
* last WR in the chain completes, all WRs in the chain
* are complete.
*/
- last->send_flags = IB_SEND_SIGNALED;
frwr->fr_cqe.done = frwr_wc_localinv_wake;
reinit_completion(&frwr->fr_linv_done);
* replaces the QP. The RPC reply handler won't call us
* unless ri_id->qp is a valid pointer.
*/
- r_xprt->rx_stats.local_inv_needed++;
bad_wr = NULL;
- rc = ib_post_send(ia->ri_id->qp, first, &bad_wr);
+ rc = ib_post_send(r_xprt->rx_ia.ri_id->qp, first, &bad_wr);
+ trace_xprtrdma_post_send(req, rc);
+
+ /* The final LOCAL_INV WR in the chain is supposed to
+ * do the wake. If it was never posted, the wake will
+ * not happen, so don't wait in that case.
+ */
if (bad_wr != first)
wait_for_completion(&frwr->fr_linv_done);
- if (rc)
- goto out_release;
+ if (!rc)
+ return;
- /* ORDER: Now DMA unmap all of the MRs, and return
- * them to the free MR list.
+ /* Recycle MRs in the LOCAL_INV chain that did not get posted.
*/
-unmap:
- while (!list_empty(mrs)) {
- mr = rpcrdma_mr_pop(mrs);
- rpcrdma_mr_unmap_and_put(mr);
+ while (bad_wr) {
+ frwr = container_of(bad_wr, struct rpcrdma_frwr,
+ fr_invwr);
+ mr = container_of(frwr, struct rpcrdma_mr, frwr);
+ bad_wr = bad_wr->next;
+
+ list_del_init(&mr->mr_list);
+ rpcrdma_mr_recycle(mr);
}
- return;
+}
-out_release:
- pr_err("rpcrdma: FRWR invalidate ib_post_send returned %i\n", rc);
+/**
+ * frwr_wc_localinv_done - Invoked by RDMA provider for a signaled LOCAL_INV WC
+ * @cq: completion queue (ignored)
+ * @wc: completed WR
+ *
+ */
+static void frwr_wc_localinv_done(struct ib_cq *cq, struct ib_wc *wc)
+{
+ struct ib_cqe *cqe = wc->wr_cqe;
+ struct rpcrdma_frwr *frwr =
+ container_of(cqe, struct rpcrdma_frwr, fr_cqe);
+ struct rpcrdma_mr *mr = container_of(frwr, struct rpcrdma_mr, frwr);
- /* Unmap and release the MRs in the LOCAL_INV WRs that did not
- * get posted.
+ /* WARNING: Only wr_cqe and status are reliable at this point */
+ trace_xprtrdma_wc_li_done(wc, frwr);
+ rpcrdma_complete_rqst(frwr->fr_req->rl_reply);
+ __frwr_release_mr(wc, mr);
+}
+
+/**
+ * frwr_unmap_async - invalidate memory regions that were registered for @req
+ * @r_xprt: controlling transport instance
+ * @req: rpcrdma_req with a non-empty list of MRs to process
+ *
+ * This guarantees that registered MRs are properly fenced from the
+ * server before the RPC consumer accesses the data in them. It also
+ * ensures proper Send flow control: waking the next RPC waits until
+ * this RPC has relinquished all its Send Queue entries.
+ */
+void frwr_unmap_async(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req)
+{
+ struct ib_send_wr *first, *last, **prev;
+ const struct ib_send_wr *bad_wr;
+ struct rpcrdma_frwr *frwr;
+ struct rpcrdma_mr *mr;
+ int rc;
+
+ /* Chain the LOCAL_INV Work Requests and post them with
+ * a single ib_post_send() call.
+ */
+ frwr = NULL;
+ prev = &first;
+ while (!list_empty(&req->rl_registered)) {
+ mr = rpcrdma_mr_pop(&req->rl_registered);
+
+ trace_xprtrdma_mr_localinv(mr);
+ r_xprt->rx_stats.local_inv_needed++;
+
+ frwr = &mr->frwr;
+ frwr->fr_cqe.done = frwr_wc_localinv;
+ frwr->fr_req = req;
+ last = &frwr->fr_invwr;
+ last->next = NULL;
+ last->wr_cqe = &frwr->fr_cqe;
+ last->sg_list = NULL;
+ last->num_sge = 0;
+ last->opcode = IB_WR_LOCAL_INV;
+ last->send_flags = IB_SEND_SIGNALED;
+ last->ex.invalidate_rkey = mr->mr_handle;
+
+ *prev = last;
+ prev = &last->next;
+ }
+
+ /* Strong send queue ordering guarantees that when the
+ * last WR in the chain completes, all WRs in the chain
+ * are complete. The last completion will wake up the
+ * RPC waiter.
+ */
+ frwr->fr_cqe.done = frwr_wc_localinv_done;
+
+ /* Transport disconnect drains the receive CQ before it
+ * replaces the QP. The RPC reply handler won't call us
+ * unless ri_id->qp is a valid pointer.
+ */
+ bad_wr = NULL;
+ rc = ib_post_send(r_xprt->rx_ia.ri_id->qp, first, &bad_wr);
+ trace_xprtrdma_post_send(req, rc);
+ if (!rc)
+ return;
+
+ /* Recycle MRs in the LOCAL_INV chain that did not get posted.
*/
while (bad_wr) {
- frwr = container_of(bad_wr, struct rpcrdma_frwr,
- fr_invwr);
+ frwr = container_of(bad_wr, struct rpcrdma_frwr, fr_invwr);
mr = container_of(frwr, struct rpcrdma_mr, frwr);
bad_wr = bad_wr->next;
- list_del_init(&mr->mr_list);
rpcrdma_mr_recycle(mr);
}
+
+ /* The final LOCAL_INV WR in the chain is supposed to
+ * do the wake. If it was never posted, the wake will
+ * not happen, so wake here in that case.
+ */
+ rpcrdma_complete_rqst(req->rl_reply);
}
unsigned int pos;
int nsegs;
+ if (rtype == rpcrdma_noch)
+ goto done;
+
pos = rqst->rq_snd_buf.head[0].iov_len;
if (rtype == rpcrdma_areadch)
pos = 0;
nsegs -= mr->mr_nents;
} while (nsegs);
- return 0;
+done:
+ return encode_item_not_present(xdr);
}
/* Register and XDR encode the Write list. Supports encoding a list
int nsegs, nchunks;
__be32 *segcount;
+ if (wtype != rpcrdma_writech)
+ goto done;
+
seg = req->rl_segments;
nsegs = rpcrdma_convert_iovs(r_xprt, &rqst->rq_rcv_buf,
rqst->rq_rcv_buf.head[0].iov_len,
/* Update count of segments in this Write chunk */
*segcount = cpu_to_be32(nchunks);
- return 0;
+done:
+ return encode_item_not_present(xdr);
}
/* Register and XDR encode the Reply chunk. Supports encoding an array
int nsegs, nchunks;
__be32 *segcount;
+ if (wtype != rpcrdma_replych)
+ return encode_item_not_present(xdr);
+
seg = req->rl_segments;
nsegs = rpcrdma_convert_iovs(r_xprt, &rqst->rq_rcv_buf, 0, wtype, seg);
if (nsegs < 0)
return 0;
}
+static void rpcrdma_sendctx_done(struct kref *kref)
+{
+ struct rpcrdma_req *req =
+ container_of(kref, struct rpcrdma_req, rl_kref);
+ struct rpcrdma_rep *rep = req->rl_reply;
+
+ rpcrdma_complete_rqst(rep);
+ rep->rr_rxprt->rx_stats.reply_waits_for_send++;
+}
+
/**
* rpcrdma_sendctx_unmap - DMA-unmap Send buffer
* @sc: sendctx containing SGEs to unmap
{
struct ib_sge *sge;
+ if (!sc->sc_unmap_count)
+ return;
+
/* The first two SGEs contain the transport header and
* the inline buffer. These are always left mapped so
* they can be cheaply re-used.
ib_dma_unmap_page(sc->sc_device, sge->addr, sge->length,
DMA_TO_DEVICE);
- if (test_and_clear_bit(RPCRDMA_REQ_F_TX_RESOURCES,
- &sc->sc_req->rl_flags))
- wake_up_bit(&sc->sc_req->rl_flags, RPCRDMA_REQ_F_TX_RESOURCES);
+ kref_put(&sc->sc_req->rl_kref, rpcrdma_sendctx_done);
}
/* Prepare an SGE for the RPC-over-RDMA transport header.
out:
sc->sc_wr.num_sge += sge_no;
if (sc->sc_unmap_count)
- __set_bit(RPCRDMA_REQ_F_TX_RESOURCES, &req->rl_flags);
+ kref_get(&req->rl_kref);
return true;
out_regbuf:
struct rpcrdma_req *req, u32 hdrlen,
struct xdr_buf *xdr, enum rpcrdma_chunktype rtype)
{
+ int ret;
+
+ ret = -EAGAIN;
req->rl_sendctx = rpcrdma_sendctx_get_locked(r_xprt);
if (!req->rl_sendctx)
- return -EAGAIN;
+ goto err;
req->rl_sendctx->sc_wr.num_sge = 0;
req->rl_sendctx->sc_unmap_count = 0;
req->rl_sendctx->sc_req = req;
- __clear_bit(RPCRDMA_REQ_F_TX_RESOURCES, &req->rl_flags);
+ kref_init(&req->rl_kref);
+ ret = -EIO;
if (!rpcrdma_prepare_hdr_sge(r_xprt, req, hdrlen))
- return -EIO;
-
+ goto err;
if (rtype != rpcrdma_areadch)
if (!rpcrdma_prepare_msg_sges(r_xprt, req, xdr, rtype))
- return -EIO;
-
+ goto err;
return 0;
+
+err:
+ trace_xprtrdma_prepsend_failed(&req->rl_slot, ret);
+ return ret;
}
/**
* send a Call message with a Position Zero Read chunk and a
* regular Read chunk at the same time.
*/
- if (rtype != rpcrdma_noch) {
- ret = rpcrdma_encode_read_list(r_xprt, req, rqst, rtype);
- if (ret)
- goto out_err;
- }
- ret = encode_item_not_present(xdr);
+ ret = rpcrdma_encode_read_list(r_xprt, req, rqst, rtype);
if (ret)
goto out_err;
-
- if (wtype == rpcrdma_writech) {
- ret = rpcrdma_encode_write_list(r_xprt, req, rqst, wtype);
- if (ret)
- goto out_err;
- }
- ret = encode_item_not_present(xdr);
+ ret = rpcrdma_encode_write_list(r_xprt, req, rqst, wtype);
if (ret)
goto out_err;
-
- if (wtype != rpcrdma_replych)
- ret = encode_item_not_present(xdr);
- else
- ret = rpcrdma_encode_reply_chunk(r_xprt, req, rqst, wtype);
+ ret = rpcrdma_encode_reply_chunk(r_xprt, req, rqst, wtype);
if (ret)
goto out_err;
- trace_xprtrdma_marshal(rqst, xdr_stream_pos(xdr), rtype, wtype);
-
- ret = rpcrdma_prepare_send_sges(r_xprt, req, xdr_stream_pos(xdr),
+ ret = rpcrdma_prepare_send_sges(r_xprt, req, req->rl_hdrbuf.len,
&rqst->rq_snd_buf, rtype);
if (ret)
goto out_err;
+
+ trace_xprtrdma_marshal(req, rtype, wtype);
return 0;
out_err:
trace_xprtrdma_marshal_failed(rqst, ret);
- switch (ret) {
- case -EAGAIN:
- xprt_wait_for_buffer_space(rqst->rq_xprt);
- break;
- case -ENOBUFS:
- break;
- default:
- r_xprt->rx_stats.failed_marshal_count++;
- }
+ r_xprt->rx_stats.failed_marshal_count++;
+ frwr_reset(req);
return ret;
}
goto out;
}
-void rpcrdma_release_rqst(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req)
-{
- /* Invalidate and unmap the data payloads before waking
- * the waiting application. This guarantees the memory
- * regions are properly fenced from the server before the
- * application accesses the data. It also ensures proper
- * send flow control: waking the next RPC waits until this
- * RPC has relinquished all its Send Queue entries.
- */
- if (!list_empty(&req->rl_registered))
- frwr_unmap_sync(r_xprt, &req->rl_registered);
-
- /* Ensure that any DMA mapped pages associated with
- * the Send of the RPC Call have been unmapped before
- * allowing the RPC to complete. This protects argument
- * memory not controlled by the RPC client from being
- * re-used before we're done with it.
- */
- if (test_bit(RPCRDMA_REQ_F_TX_RESOURCES, &req->rl_flags)) {
- r_xprt->rx_stats.reply_waits_for_send++;
- out_of_line_wait_on_bit(&req->rl_flags,
- RPCRDMA_REQ_F_TX_RESOURCES,
- bit_wait,
- TASK_UNINTERRUPTIBLE);
- }
-}
-
-/* Reply handling runs in the poll worker thread. Anything that
- * might wait is deferred to a separate workqueue.
- */
-void rpcrdma_deferred_completion(struct work_struct *work)
+static void rpcrdma_reply_done(struct kref *kref)
{
- struct rpcrdma_rep *rep =
- container_of(work, struct rpcrdma_rep, rr_work);
- struct rpcrdma_req *req = rpcr_to_rdmar(rep->rr_rqst);
- struct rpcrdma_xprt *r_xprt = rep->rr_rxprt;
+ struct rpcrdma_req *req =
+ container_of(kref, struct rpcrdma_req, rl_kref);
- trace_xprtrdma_defer_cmp(rep);
- if (rep->rr_wc_flags & IB_WC_WITH_INVALIDATE)
- frwr_reminv(rep, &req->rl_registered);
- rpcrdma_release_rqst(r_xprt, req);
- rpcrdma_complete_rqst(rep);
+ rpcrdma_complete_rqst(req->rl_reply);
}
-/* Process received RPC/RDMA messages.
+/**
+ * rpcrdma_reply_handler - Process received RPC/RDMA messages
+ * @rep: Incoming rpcrdma_rep object to process
*
* Errors must result in the RPC task either being awakened, or
* allowed to timeout, to discover the errors at that time.
else if (credits > buf->rb_max_requests)
credits = buf->rb_max_requests;
if (buf->rb_credits != credits) {
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
buf->rb_credits = credits;
xprt->cwnd = credits << RPC_CWNDSHIFT;
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
}
req = rpcr_to_rdmar(rqst);
}
req->rl_reply = rep;
rep->rr_rqst = rqst;
- clear_bit(RPCRDMA_REQ_F_PENDING, &req->rl_flags);
trace_xprtrdma_reply(rqst->rq_task, rep, req, credits);
- queue_work(buf->rb_completion_wq, &rep->rr_work);
+
+ if (rep->rr_wc_flags & IB_WC_WITH_INVALIDATE)
+ frwr_reminv(rep, &req->rl_registered);
+ if (!list_empty(&req->rl_registered))
+ frwr_unmap_async(r_xprt, req);
+ /* LocalInv completion will complete the RPC */
+ else
+ kref_put(&req->rl_kref, rpcrdma_reply_done);
return;
out_badversion:
else if (credits > r_xprt->rx_buf.rb_bc_max_requests)
credits = r_xprt->rx_buf.rb_bc_max_requests;
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
xprt->cwnd = credits << RPC_CWNDSHIFT;
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
spin_lock(&xprt->queue_lock);
ret = 0;
* Enqueue the new transport on the accept queue of the listening
* transport
*/
- spin_lock_bh(&listen_xprt->sc_lock);
+ spin_lock(&listen_xprt->sc_lock);
list_add_tail(&newxprt->sc_accept_q, &listen_xprt->sc_accept_q);
- spin_unlock_bh(&listen_xprt->sc_lock);
+ spin_unlock(&listen_xprt->sc_lock);
set_bit(XPT_CONN, &listen_xprt->sc_xprt.xpt_flags);
svc_xprt_enqueue(&listen_xprt->sc_xprt);
listen_rdma = container_of(xprt, struct svcxprt_rdma, sc_xprt);
clear_bit(XPT_CONN, &xprt->xpt_flags);
/* Get the next entry off the accept list */
- spin_lock_bh(&listen_rdma->sc_lock);
+ spin_lock(&listen_rdma->sc_lock);
if (!list_empty(&listen_rdma->sc_accept_q)) {
newxprt = list_entry(listen_rdma->sc_accept_q.next,
struct svcxprt_rdma, sc_accept_q);
}
if (!list_empty(&listen_rdma->sc_accept_q))
set_bit(XPT_CONN, &listen_rdma->sc_xprt.xpt_flags);
- spin_unlock_bh(&listen_rdma->sc_lock);
+ spin_unlock(&listen_rdma->sc_lock);
if (!newxprt)
return NULL;
module_put(THIS_MODULE);
}
+/* 60 second timeout, no retries */
static const struct rpc_timeout xprt_rdma_default_timeout = {
.to_initval = 60 * HZ,
.to_maxval = 60 * HZ,
if (!xprt)
return ERR_PTR(-ENOMEM);
- /* 60 second timeout, no retries */
xprt->timeout = &xprt_rdma_default_timeout;
+ xprt->connect_timeout = xprt->timeout->to_initval;
+ xprt->max_reconnect_timeout = xprt->timeout->to_maxval;
xprt->bind_timeout = RPCRDMA_BIND_TO;
xprt->reestablish_timeout = RPCRDMA_INIT_REEST_TO;
xprt->idle_timeout = RPCRDMA_IDLE_DISC_TO;
}
/**
- * xprt_rdma_connect - try to establish a transport connection
+ * xprt_rdma_set_connect_timeout - set timeouts for establishing a connection
+ * @xprt: controlling transport instance
+ * @connect_timeout: reconnect timeout after client disconnects
+ * @reconnect_timeout: reconnect timeout after server disconnects
+ *
+ */
+static void xprt_rdma_tcp_set_connect_timeout(struct rpc_xprt *xprt,
+ unsigned long connect_timeout,
+ unsigned long reconnect_timeout)
+{
+ struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
+
+ trace_xprtrdma_op_set_cto(r_xprt, connect_timeout, reconnect_timeout);
+
+ spin_lock(&xprt->transport_lock);
+
+ if (connect_timeout < xprt->connect_timeout) {
+ struct rpc_timeout to;
+ unsigned long initval;
+
+ to = *xprt->timeout;
+ initval = connect_timeout;
+ if (initval < RPCRDMA_INIT_REEST_TO << 1)
+ initval = RPCRDMA_INIT_REEST_TO << 1;
+ to.to_initval = initval;
+ to.to_maxval = initval;
+ r_xprt->rx_timeout = to;
+ xprt->timeout = &r_xprt->rx_timeout;
+ xprt->connect_timeout = connect_timeout;
+ }
+
+ if (reconnect_timeout < xprt->max_reconnect_timeout)
+ xprt->max_reconnect_timeout = reconnect_timeout;
+
+ spin_unlock(&xprt->transport_lock);
+}
+
+/**
+ * xprt_rdma_connect - schedule an attempt to reconnect
* @xprt: transport state
- * @task: RPC scheduler context
+ * @task: RPC scheduler context (unused)
*
*/
static void
xprt_rdma_connect(struct rpc_xprt *xprt, struct rpc_task *task)
{
struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(xprt);
+ unsigned long delay;
trace_xprtrdma_op_connect(r_xprt);
+
+ delay = 0;
if (r_xprt->rx_ep.rep_connected != 0) {
- /* Reconnect */
- schedule_delayed_work(&r_xprt->rx_connect_worker,
- xprt->reestablish_timeout);
- xprt->reestablish_timeout <<= 1;
- if (xprt->reestablish_timeout > RPCRDMA_MAX_REEST_TO)
- xprt->reestablish_timeout = RPCRDMA_MAX_REEST_TO;
- else if (xprt->reestablish_timeout < RPCRDMA_INIT_REEST_TO)
- xprt->reestablish_timeout = RPCRDMA_INIT_REEST_TO;
- } else {
- schedule_delayed_work(&r_xprt->rx_connect_worker, 0);
- if (!RPC_IS_ASYNC(task))
- flush_delayed_work(&r_xprt->rx_connect_worker);
+ delay = xprt_reconnect_delay(xprt);
+ xprt_reconnect_backoff(xprt, RPCRDMA_INIT_REEST_TO);
}
+ queue_delayed_work(xprtiod_workqueue, &r_xprt->rx_connect_worker,
+ delay);
}
/**
static void
xprt_rdma_free_slot(struct rpc_xprt *xprt, struct rpc_rqst *rqst)
{
+ struct rpcrdma_xprt *r_xprt =
+ container_of(xprt, struct rpcrdma_xprt, rx_xprt);
+
memset(rqst, 0, sizeof(*rqst));
- rpcrdma_buffer_put(rpcr_to_rdmar(rqst));
+ rpcrdma_buffer_put(&r_xprt->rx_buf, rpcr_to_rdmar(rqst));
rpc_wake_up_next(&xprt->backlog);
}
struct rpcrdma_xprt *r_xprt = rpcx_to_rdmax(rqst->rq_xprt);
struct rpcrdma_req *req = rpcr_to_rdmar(rqst);
- if (test_bit(RPCRDMA_REQ_F_PENDING, &req->rl_flags))
- rpcrdma_release_rqst(r_xprt, req);
trace_xprtrdma_op_free(task, req);
+
+ if (!list_empty(&req->rl_registered))
+ frwr_unmap_sync(r_xprt, req);
+
+ /* XXX: If the RPC is completing because of a signal and
+ * not because a reply was received, we ought to ensure
+ * that the Send completion has fired, so that memory
+ * involved with the Send is not still visible to the NIC.
+ */
}
/**
goto drop_connection;
rqst->rq_xtime = ktime_get();
- __set_bit(RPCRDMA_REQ_F_PENDING, &req->rl_flags);
if (rpcrdma_ep_post(&r_xprt->rx_ia, &r_xprt->rx_ep, req))
goto drop_connection;
.send_request = xprt_rdma_send_request,
.close = xprt_rdma_close,
.destroy = xprt_rdma_destroy,
+ .set_connect_timeout = xprt_rdma_tcp_set_connect_timeout,
.print_stats = xprt_rdma_print_stats,
.enable_swap = xprt_rdma_enable_swap,
.disable_swap = xprt_rdma_disable_swap,
#if defined(CONFIG_SUNRPC_BACKCHANNEL)
.bc_setup = xprt_rdma_bc_setup,
.bc_maxpayload = xprt_rdma_bc_maxpayload,
+ .bc_num_slots = xprt_rdma_bc_max_slots,
.bc_free_rqst = xprt_rdma_bc_free_rqst,
.bc_destroy = xprt_rdma_bc_destroy,
#endif
*/
static void rpcrdma_xprt_drain(struct rpcrdma_xprt *r_xprt)
{
- struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
struct rpcrdma_ia *ia = &r_xprt->rx_ia;
/* Flush Receives, then wait for deferred Reply work
* to complete.
*/
ib_drain_rq(ia->ri_id->qp);
- drain_workqueue(buf->rb_completion_wq);
/* Deferred Reply processing might have scheduled
* local invalidations.
* completions recently. This is a sign the Send Queue is
* backing up. Cause the caller to pause and try again.
*/
- set_bit(RPCRDMA_BUF_F_EMPTY_SCQ, &buf->rb_flags);
+ xprt_wait_for_buffer_space(&r_xprt->rx_xprt);
r_xprt->rx_stats.empty_sendctx_q++;
return NULL;
}
/* Paired with READ_ONCE */
smp_store_release(&buf->rb_sc_tail, next_tail);
- if (test_and_clear_bit(RPCRDMA_BUF_F_EMPTY_SCQ, &buf->rb_flags)) {
- smp_mb__after_atomic();
- xprt_write_space(&sc->sc_xprt->rx_xprt);
- }
+ xprt_write_space(&sc->sc_xprt->rx_xprt);
}
static void
r_xprt->rx_stats.mrs_allocated += count;
spin_unlock(&buf->rb_mrlock);
trace_xprtrdma_createmrs(r_xprt, count);
-
- xprt_write_space(&r_xprt->rx_xprt);
}
static void
rx_buf);
rpcrdma_mrs_create(r_xprt);
+ xprt_write_space(&r_xprt->rx_xprt);
}
/**
if (!req->rl_recvbuf)
goto out4;
- req->rl_buffer = buffer;
INIT_LIST_HEAD(&req->rl_registered);
spin_lock(&buffer->rb_lock);
list_add(&req->rl_all, &buffer->rb_allreqs);
return NULL;
}
-static bool rpcrdma_rep_create(struct rpcrdma_xprt *r_xprt, bool temp)
+static struct rpcrdma_rep *rpcrdma_rep_create(struct rpcrdma_xprt *r_xprt,
+ bool temp)
{
- struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
struct rpcrdma_rep *rep;
rep = kzalloc(sizeof(*rep), GFP_KERNEL);
DMA_FROM_DEVICE, GFP_KERNEL);
if (!rep->rr_rdmabuf)
goto out_free;
+
xdr_buf_init(&rep->rr_hdrbuf, rdmab_data(rep->rr_rdmabuf),
rdmab_length(rep->rr_rdmabuf));
-
rep->rr_cqe.done = rpcrdma_wc_receive;
rep->rr_rxprt = r_xprt;
- INIT_WORK(&rep->rr_work, rpcrdma_deferred_completion);
rep->rr_recv_wr.next = NULL;
rep->rr_recv_wr.wr_cqe = &rep->rr_cqe;
rep->rr_recv_wr.sg_list = &rep->rr_rdmabuf->rg_iov;
rep->rr_recv_wr.num_sge = 1;
rep->rr_temp = temp;
-
- spin_lock(&buf->rb_lock);
- list_add(&rep->rr_list, &buf->rb_recv_bufs);
- spin_unlock(&buf->rb_lock);
- return true;
+ return rep;
out_free:
kfree(rep);
out:
- return false;
+ return NULL;
}
/**
struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
int i, rc;
- buf->rb_flags = 0;
buf->rb_max_requests = r_xprt->rx_ep.rep_max_requests;
buf->rb_bc_srv_max_requests = 0;
spin_lock_init(&buf->rb_mrlock);
if (rc)
goto out;
- buf->rb_completion_wq = alloc_workqueue("rpcrdma-%s",
- WQ_MEM_RECLAIM | WQ_HIGHPRI,
- 0,
- r_xprt->rx_xprt.address_strings[RPC_DISPLAY_ADDR]);
- if (!buf->rb_completion_wq) {
- rc = -ENOMEM;
- goto out;
- }
-
return 0;
out:
rpcrdma_buffer_destroy(buf);
{
cancel_delayed_work_sync(&buf->rb_refresh_worker);
- if (buf->rb_completion_wq) {
- destroy_workqueue(buf->rb_completion_wq);
- buf->rb_completion_wq = NULL;
- }
-
rpcrdma_sendctxs_destroy(buf);
while (!list_empty(&buf->rb_recv_bufs)) {
/**
* rpcrdma_buffer_put - Put request/reply buffers back into pool
+ * @buffers: buffer pool
* @req: object to return
*
*/
-void
-rpcrdma_buffer_put(struct rpcrdma_req *req)
+void rpcrdma_buffer_put(struct rpcrdma_buffer *buffers, struct rpcrdma_req *req)
{
- struct rpcrdma_buffer *buffers = req->rl_buffer;
struct rpcrdma_rep *rep = req->rl_reply;
req->rl_reply = NULL;
struct ib_send_wr *send_wr = &req->rl_sendctx->sc_wr;
int rc;
- if (!ep->rep_send_count ||
- test_bit(RPCRDMA_REQ_F_TX_RESOURCES, &req->rl_flags)) {
+ if (!ep->rep_send_count || kref_read(&req->rl_kref) > 1) {
send_wr->send_flags |= IB_SEND_SIGNALED;
ep->rep_send_count = ep->rep_send_batch;
} else {
{
struct rpcrdma_buffer *buf = &r_xprt->rx_buf;
struct rpcrdma_ep *ep = &r_xprt->rx_ep;
- struct ib_recv_wr *wr, *bad_wr;
+ struct ib_recv_wr *i, *wr, *bad_wr;
+ struct rpcrdma_rep *rep;
int needed, count, rc;
rc = 0;
count = 0;
+
needed = buf->rb_credits + (buf->rb_bc_srv_max_requests << 1);
if (ep->rep_receive_count > needed)
goto out;
if (!temp)
needed += RPCRDMA_MAX_RECV_BATCH;
- count = 0;
+ /* fast path: all needed reps can be found on the free list */
wr = NULL;
+ spin_lock(&buf->rb_lock);
while (needed) {
- struct rpcrdma_regbuf *rb;
- struct rpcrdma_rep *rep;
-
- spin_lock(&buf->rb_lock);
rep = list_first_entry_or_null(&buf->rb_recv_bufs,
struct rpcrdma_rep, rr_list);
- if (likely(rep))
- list_del(&rep->rr_list);
- spin_unlock(&buf->rb_lock);
- if (!rep) {
- if (!rpcrdma_rep_create(r_xprt, temp))
- break;
- continue;
- }
+ if (!rep)
+ break;
- rb = rep->rr_rdmabuf;
- if (!rpcrdma_regbuf_dma_map(r_xprt, rb)) {
- rpcrdma_recv_buffer_put(rep);
+ list_del(&rep->rr_list);
+ rep->rr_recv_wr.next = wr;
+ wr = &rep->rr_recv_wr;
+ --needed;
+ }
+ spin_unlock(&buf->rb_lock);
+
+ while (needed) {
+ rep = rpcrdma_rep_create(r_xprt, temp);
+ if (!rep)
break;
- }
- trace_xprtrdma_post_recv(rep->rr_recv_wr.wr_cqe);
rep->rr_recv_wr.next = wr;
wr = &rep->rr_recv_wr;
- ++count;
--needed;
}
- if (!count)
+ if (!wr)
goto out;
+ for (i = wr; i; i = i->next) {
+ rep = container_of(i, struct rpcrdma_rep, rr_recv_wr);
+
+ if (!rpcrdma_regbuf_dma_map(r_xprt, rep->rr_rdmabuf))
+ goto release_wrs;
+
+ trace_xprtrdma_post_recv(rep->rr_recv_wr.wr_cqe);
+ ++count;
+ }
+
rc = ib_post_recv(r_xprt->rx_ia.ri_id->qp, wr,
(const struct ib_recv_wr **)&bad_wr);
+out:
+ trace_xprtrdma_post_recvs(r_xprt, count, rc);
if (rc) {
- for (wr = bad_wr; wr; wr = wr->next) {
+ for (wr = bad_wr; wr;) {
struct rpcrdma_rep *rep;
rep = container_of(wr, struct rpcrdma_rep, rr_recv_wr);
+ wr = wr->next;
rpcrdma_recv_buffer_put(rep);
--count;
}
}
ep->rep_receive_count += count;
-out:
- trace_xprtrdma_post_recvs(r_xprt, count, rc);
+ return;
+
+release_wrs:
+ for (i = wr; i;) {
+ rep = container_of(i, struct rpcrdma_rep, rr_recv_wr);
+ i = i->next;
+ rpcrdma_recv_buffer_put(rep);
+ }
}
#include <linux/wait.h> /* wait_queue_head_t, etc */
#include <linux/spinlock.h> /* spinlock_t, etc */
-#include <linux/atomic.h> /* atomic_t, etc */
+#include <linux/atomic.h> /* atomic_t, etc */
+#include <linux/kref.h> /* struct kref */
#include <linux/workqueue.h> /* struct work_struct */
#include <rdma/rdma_cm.h> /* RDMA connection api */
bool rr_temp;
struct rpcrdma_regbuf *rr_rdmabuf;
struct rpcrdma_xprt *rr_rxprt;
- struct work_struct rr_work;
+ struct rpc_rqst *rr_rqst;
struct xdr_buf rr_hdrbuf;
struct xdr_stream rr_stream;
- struct rpc_rqst *rr_rqst;
struct list_head rr_list;
struct ib_recv_wr rr_recv_wr;
};
* An external memory region is any buffer or page that is registered
* on the fly (ie, not pre-registered).
*/
-enum rpcrdma_frwr_state {
- FRWR_IS_INVALID, /* ready to be used */
- FRWR_IS_VALID, /* in use */
- FRWR_FLUSHED_FR, /* flushed FASTREG WR */
- FRWR_FLUSHED_LI, /* flushed LOCALINV WR */
-};
-
+struct rpcrdma_req;
struct rpcrdma_frwr {
struct ib_mr *fr_mr;
struct ib_cqe fr_cqe;
- enum rpcrdma_frwr_state fr_state;
struct completion fr_linv_done;
+ struct rpcrdma_req *fr_req;
union {
struct ib_reg_wr fr_regwr;
struct ib_send_wr fr_invwr;
struct rpcrdma_req {
struct list_head rl_list;
struct rpc_rqst rl_slot;
- struct rpcrdma_buffer *rl_buffer;
struct rpcrdma_rep *rl_reply;
struct xdr_stream rl_stream;
struct xdr_buf rl_hdrbuf;
struct rpcrdma_regbuf *rl_recvbuf; /* rq_rcv_buf */
struct list_head rl_all;
- unsigned long rl_flags;
+ struct kref rl_kref;
struct list_head rl_registered; /* registered segments */
struct rpcrdma_mr_seg rl_segments[RPCRDMA_MAX_SEGS];
};
-/* rl_flags */
-enum {
- RPCRDMA_REQ_F_PENDING = 0,
- RPCRDMA_REQ_F_TX_RESOURCES,
-};
-
static inline struct rpcrdma_req *
rpcr_to_rdmar(const struct rpc_rqst *rqst)
{
struct list_head rb_recv_bufs;
struct list_head rb_allreqs;
- unsigned long rb_flags;
u32 rb_max_requests;
u32 rb_credits; /* most recent credit grant */
u32 rb_bc_srv_max_requests;
u32 rb_bc_max_requests;
- struct workqueue_struct *rb_completion_wq;
struct delayed_work rb_refresh_worker;
};
-/* rb_flags */
-enum {
- RPCRDMA_BUF_F_EMPTY_SCQ = 0,
-};
-
/*
* Statistics for RPCRDMA
*/
struct rpcrdma_ep rx_ep;
struct rpcrdma_buffer rx_buf;
struct delayed_work rx_connect_worker;
+ struct rpc_timeout rx_timeout;
struct rpcrdma_stats rx_stats;
};
}
struct rpcrdma_req *rpcrdma_buffer_get(struct rpcrdma_buffer *);
-void rpcrdma_buffer_put(struct rpcrdma_req *);
+void rpcrdma_buffer_put(struct rpcrdma_buffer *buffers,
+ struct rpcrdma_req *req);
void rpcrdma_recv_buffer_put(struct rpcrdma_rep *);
bool rpcrdma_regbuf_realloc(struct rpcrdma_regbuf *rb, size_t size,
/* Memory registration calls xprtrdma/frwr_ops.c
*/
bool frwr_is_supported(struct ib_device *device);
+void frwr_reset(struct rpcrdma_req *req);
int frwr_open(struct rpcrdma_ia *ia, struct rpcrdma_ep *ep);
int frwr_init_mr(struct rpcrdma_ia *ia, struct rpcrdma_mr *mr);
void frwr_release_mr(struct rpcrdma_mr *mr);
struct rpcrdma_mr **mr);
int frwr_send(struct rpcrdma_ia *ia, struct rpcrdma_req *req);
void frwr_reminv(struct rpcrdma_rep *rep, struct list_head *mrs);
-void frwr_unmap_sync(struct rpcrdma_xprt *r_xprt,
- struct list_head *mrs);
+void frwr_unmap_sync(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req);
+void frwr_unmap_async(struct rpcrdma_xprt *r_xprt, struct rpcrdma_req *req);
/*
* RPC/RDMA protocol calls - xprtrdma/rpc_rdma.c
void rpcrdma_set_max_header_sizes(struct rpcrdma_xprt *);
void rpcrdma_complete_rqst(struct rpcrdma_rep *rep);
void rpcrdma_reply_handler(struct rpcrdma_rep *rep);
-void rpcrdma_release_rqst(struct rpcrdma_xprt *r_xprt,
- struct rpcrdma_req *req);
-void rpcrdma_deferred_completion(struct work_struct *work);
static inline void rpcrdma_set_xdrlen(struct xdr_buf *xdr, size_t len)
{
#if defined(CONFIG_SUNRPC_BACKCHANNEL)
int xprt_rdma_bc_setup(struct rpc_xprt *, unsigned int);
size_t xprt_rdma_bc_maxpayload(struct rpc_xprt *);
+unsigned int xprt_rdma_bc_max_slots(struct rpc_xprt *);
int rpcrdma_bc_post_recv(struct rpcrdma_xprt *, unsigned int);
void rpcrdma_bc_receive_call(struct rpcrdma_xprt *, struct rpcrdma_rep *);
int xprt_rdma_bc_send_reply(struct rpc_rqst *rqst);
req->rq_slen);
/* Protect against races with write_space */
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
/* Don't race with disconnect */
if (xprt_connected(xprt)) {
} else
ret = -ENOTCONN;
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
/* Race breaker in case memory is freed before above code is called */
if (ret == -EAGAIN) {
static void
xs_stream_prepare_request(struct rpc_rqst *req)
{
+ xdr_free_bvec(&req->rq_rcv_buf);
req->rq_task->tk_status = xdr_alloc_bvec(&req->rq_rcv_buf, GFP_KERNEL);
}
struct sock_xprt *transport = container_of(xprt, struct sock_xprt, xprt);
clear_bit(XPRT_SOCK_DATA_READY, &transport->sock_state);
+ clear_bit(XPRT_SOCK_WAKE_ERROR, &transport->sock_state);
+ clear_bit(XPRT_SOCK_WAKE_WRITE, &transport->sock_state);
+ clear_bit(XPRT_SOCK_WAKE_DISCONNECT, &transport->sock_state);
+}
+
+static void xs_run_error_worker(struct sock_xprt *transport, unsigned int nr)
+{
+ set_bit(nr, &transport->sock_state);
+ queue_work(xprtiod_workqueue, &transport->error_worker);
}
static void xs_sock_reset_connection_flags(struct rpc_xprt *xprt)
*/
static void xs_error_report(struct sock *sk)
{
+ struct sock_xprt *transport;
struct rpc_xprt *xprt;
int err;
if (!(xprt = xprt_from_sock(sk)))
goto out;
+ transport = container_of(xprt, struct sock_xprt, xprt);
err = -sk->sk_err;
if (err == 0)
goto out;
dprintk("RPC: xs_error_report client %p, error=%d...\n",
xprt, -err);
trace_rpc_socket_error(xprt, sk->sk_socket, err);
- xprt_wake_pending_tasks(xprt, err);
+ xs_run_error_worker(transport, XPRT_SOCK_WAKE_ERROR);
out:
read_unlock_bh(&sk->sk_callback_lock);
}
cancel_delayed_work_sync(&transport->connect_worker);
xs_close(xprt);
cancel_work_sync(&transport->recv_worker);
+ cancel_work_sync(&transport->error_worker);
xs_xprt_free(xprt);
module_put(THIS_MODULE);
}
}
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
xprt_adjust_cwnd(xprt, task, copied);
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
spin_lock(&xprt->queue_lock);
xprt_complete_rqst(task, copied);
__UDPX_INC_STATS(sk, UDP_MIB_INDATAGRAMS);
trace_rpc_socket_state_change(xprt, sk->sk_socket);
switch (sk->sk_state) {
case TCP_ESTABLISHED:
- spin_lock(&xprt->transport_lock);
if (!xprt_test_and_set_connected(xprt)) {
xprt->connect_cookie++;
clear_bit(XPRT_SOCK_CONNECTING, &transport->sock_state);
xprt->stat.connect_count++;
xprt->stat.connect_time += (long)jiffies -
xprt->stat.connect_start;
- xprt_wake_pending_tasks(xprt, -EAGAIN);
+ xs_run_error_worker(transport, XPRT_SOCK_WAKE_PENDING);
}
- spin_unlock(&xprt->transport_lock);
break;
case TCP_FIN_WAIT1:
/* The client initiated a shutdown of the socket */
/* The server initiated a shutdown of the socket */
xprt->connect_cookie++;
clear_bit(XPRT_CONNECTED, &xprt->state);
- xs_tcp_force_close(xprt);
+ xs_run_error_worker(transport, XPRT_SOCK_WAKE_DISCONNECT);
/* fall through */
case TCP_CLOSING:
/*
xprt_clear_connecting(xprt);
clear_bit(XPRT_CLOSING, &xprt->state);
/* Trigger the socket release */
- xs_tcp_force_close(xprt);
+ xs_run_error_worker(transport, XPRT_SOCK_WAKE_DISCONNECT);
}
out:
read_unlock_bh(&sk->sk_callback_lock);
static void xs_write_space(struct sock *sk)
{
struct socket_wq *wq;
+ struct sock_xprt *transport;
struct rpc_xprt *xprt;
if (!sk->sk_socket)
if (unlikely(!(xprt = xprt_from_sock(sk))))
return;
+ transport = container_of(xprt, struct sock_xprt, xprt);
rcu_read_lock();
wq = rcu_dereference(sk->sk_wq);
if (!wq || test_and_clear_bit(SOCKWQ_ASYNC_NOSPACE, &wq->flags) == 0)
goto out;
- if (xprt_write_space(xprt))
- sk->sk_write_pending--;
+ xs_run_error_worker(transport, XPRT_SOCK_WAKE_WRITE);
+ sk->sk_write_pending--;
out:
rcu_read_unlock();
}
*/
static void xs_udp_timer(struct rpc_xprt *xprt, struct rpc_task *task)
{
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
xprt_adjust_cwnd(xprt, task, -ETIMEDOUT);
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
}
static int xs_get_random_port(void)
unsigned int opt_on = 1;
unsigned int timeo;
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
keepidle = DIV_ROUND_UP(xprt->timeout->to_initval, HZ);
keepcnt = xprt->timeout->to_retries + 1;
timeo = jiffies_to_msecs(xprt->timeout->to_initval) *
(xprt->timeout->to_retries + 1);
clear_bit(XPRT_SOCK_UPD_TIMEOUT, &transport->sock_state);
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
/* TCP Keepalive options */
kernel_setsockopt(sock, SOL_SOCKET, SO_KEEPALIVE,
struct rpc_timeout to;
unsigned long initval;
- spin_lock_bh(&xprt->transport_lock);
+ spin_lock(&xprt->transport_lock);
if (reconnect_timeout < xprt->max_reconnect_timeout)
xprt->max_reconnect_timeout = reconnect_timeout;
if (connect_timeout < xprt->connect_timeout) {
xprt->connect_timeout = connect_timeout;
}
set_bit(XPRT_SOCK_UPD_TIMEOUT, &transport->sock_state);
- spin_unlock_bh(&xprt->transport_lock);
+ spin_unlock(&xprt->transport_lock);
}
static int xs_tcp_finish_connecting(struct rpc_xprt *xprt, struct socket *sock)
xprt_wake_pending_tasks(xprt, status);
}
-static unsigned long xs_reconnect_delay(const struct rpc_xprt *xprt)
-{
- unsigned long start, now = jiffies;
-
- start = xprt->stat.connect_start + xprt->reestablish_timeout;
- if (time_after(start, now))
- return start - now;
- return 0;
-}
-
-static void xs_reconnect_backoff(struct rpc_xprt *xprt)
-{
- xprt->reestablish_timeout <<= 1;
- if (xprt->reestablish_timeout > xprt->max_reconnect_timeout)
- xprt->reestablish_timeout = xprt->max_reconnect_timeout;
- if (xprt->reestablish_timeout < XS_TCP_INIT_REEST_TO)
- xprt->reestablish_timeout = XS_TCP_INIT_REEST_TO;
-}
-
/**
* xs_connect - connect a socket to a remote endpoint
* @xprt: pointer to transport structure
/* Start by resetting any existing state */
xs_reset_transport(transport);
- delay = xs_reconnect_delay(xprt);
- xs_reconnect_backoff(xprt);
+ delay = xprt_reconnect_delay(xprt);
+ xprt_reconnect_backoff(xprt, XS_TCP_INIT_REEST_TO);
} else
dprintk("RPC: xs_connect scheduled xprt %p\n", xprt);
delay);
}
+static void xs_wake_disconnect(struct sock_xprt *transport)
+{
+ if (test_and_clear_bit(XPRT_SOCK_WAKE_DISCONNECT, &transport->sock_state))
+ xs_tcp_force_close(&transport->xprt);
+}
+
+static void xs_wake_write(struct sock_xprt *transport)
+{
+ if (test_and_clear_bit(XPRT_SOCK_WAKE_WRITE, &transport->sock_state))
+ xprt_write_space(&transport->xprt);
+}
+
+static void xs_wake_error(struct sock_xprt *transport)
+{
+ int sockerr;
+ int sockerr_len = sizeof(sockerr);
+
+ if (!test_bit(XPRT_SOCK_WAKE_ERROR, &transport->sock_state))
+ return;
+ mutex_lock(&transport->recv_mutex);
+ if (transport->sock == NULL)
+ goto out;
+ if (!test_and_clear_bit(XPRT_SOCK_WAKE_ERROR, &transport->sock_state))
+ goto out;
+ if (kernel_getsockopt(transport->sock, SOL_SOCKET, SO_ERROR,
+ (char *)&sockerr, &sockerr_len) != 0)
+ goto out;
+ if (sockerr < 0)
+ xprt_wake_pending_tasks(&transport->xprt, sockerr);
+out:
+ mutex_unlock(&transport->recv_mutex);
+}
+
+static void xs_wake_pending(struct sock_xprt *transport)
+{
+ if (test_and_clear_bit(XPRT_SOCK_WAKE_PENDING, &transport->sock_state))
+ xprt_wake_pending_tasks(&transport->xprt, -EAGAIN);
+}
+
+static void xs_error_handle(struct work_struct *work)
+{
+ struct sock_xprt *transport = container_of(work,
+ struct sock_xprt, error_worker);
+
+ xs_wake_disconnect(transport);
+ xs_wake_write(transport);
+ xs_wake_error(transport);
+ xs_wake_pending(transport);
+}
+
/**
* xs_local_print_stats - display AF_LOCAL socket-specifc stats
* @xprt: rpc_xprt struct containing statistics
#ifdef CONFIG_SUNRPC_BACKCHANNEL
.bc_setup = xprt_setup_bc,
.bc_maxpayload = xs_tcp_bc_maxpayload,
+ .bc_num_slots = xprt_bc_max_slots,
.bc_free_rqst = xprt_free_bc_rqst,
.bc_destroy = xprt_destroy_bc,
#endif
xprt->timeout = &xs_local_default_timeout;
INIT_WORK(&transport->recv_worker, xs_stream_data_receive_workfn);
+ INIT_WORK(&transport->error_worker, xs_error_handle);
INIT_DELAYED_WORK(&transport->connect_worker, xs_dummy_setup_socket);
switch (sun->sun_family) {
xprt->timeout = &xs_udp_default_timeout;
INIT_WORK(&transport->recv_worker, xs_udp_data_receive_workfn);
+ INIT_WORK(&transport->error_worker, xs_error_handle);
INIT_DELAYED_WORK(&transport->connect_worker, xs_udp_setup_socket);
switch (addr->sa_family) {
(xprt->timeout->to_retries + 1);
INIT_WORK(&transport->recv_worker, xs_stream_data_receive_workfn);
+ INIT_WORK(&transport->error_worker, xs_error_handle);
INIT_DELAYED_WORK(&transport->connect_worker, xs_tcp_setup_socket);
switch (addr->sa_family) {
projects / linux-2.6-block.git / commitdiff