Skip to content

C
clone-Linux

gsplhtlxg / clone-Linux

通知 2
Star 0
Fork 1
C
clone-Linux

体验新版 GitCode，发现更多精彩内容 >>

提交 a50243b1 编写于 作者: L Linus Torvalds
浏览文件
操作

Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma 

Pull rdma updates from Jason Gunthorpe:
 "This has been a slightly more active cycle than normal with ongoing
  core changes and quite a lot of collected driver updates.

   - Various driver fixes for bnxt_re, cxgb4, hns, mlx5, pvrdma, rxe

   - A new data transfer mode for HFI1 giving higher performance

   - Significant functional and bug fix update to the mlx5
     On-Demand-Paging MR feature

   - A chip hang reset recovery system for hns

   - Change mm->pinned_vm to an atomic64

   - Update bnxt_re to support a new 57500 chip

   - A sane netlink 'rdma link add' method for creating rxe devices and
     fixing the various unregistration race conditions in rxe's
     unregister flow

   - Allow lookup up objects by an ID over netlink

   - Various reworking of the core to driver interface:
       - drivers should not assume umem SGLs are in PAGE_SIZE chunks
       - ucontext is accessed via udata not other means
       - start to make the core code responsible for object memory
         allocation
       - drivers should convert struct device to struct ib_device via a
         helper
       - drivers have more tools to avoid use after unregister problems"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (280 commits)
  net/mlx5: ODP support for XRC transport is not enabled by default in FW
  IB/hfi1: Close race condition on user context disable and close
  RDMA/umem: Revert broken 'off by one' fix
  RDMA/umem: minor bug fix in error handling path
  RDMA/hns: Use GFP_ATOMIC in hns_roce_v2_modify_qp
  cxgb4: kfree mhp after the debug print
  IB/rdmavt: Fix concurrency panics in QP post_send and modify to error
  IB/rdmavt: Fix loopback send with invalidate ordering
  IB/iser: Fix dma_nents type definition
  IB/mlx5: Set correct write permissions for implicit ODP MR
  bnxt_re: Clean cq for kernel consumers only
  RDMA/uverbs: Don't do double free of allocated PD
  RDMA: Handle ucontext allocations by IB/core
  RDMA/core: Fix a WARN() message
  bnxt_re: fix the regression due to changes in alloc_pbl
  IB/mlx4: Increase the timeout for CM cache
  IB/core: Abort page fault handler silently during owning process exit
  IB/mlx5: Validate correct PD before prefetch MR
  IB/mlx5: Protect against prefetch of invalid MR
  RDMA/uverbs: Store PR pointer before it is overwritten
  ...
上级 2901752c fca22e7e
展开全部 隐藏空白更改
内联 并排
Showing with 16710 addition and 5014 deletion
.clang-format
浏览文件 @ a50243b1
......@@ -240,6 +240,7 @@ ForEachMacros:
- 'for_each_set_bit'
- 'for_each_set_bit_from'
- 'for_each_sg'
- 'for_each_sg_dma_page'
- 'for_each_sg_page'
- 'for_each_sibling_event'
- '__for_each_thread'
......@@ -360,6 +361,7 @@ ForEachMacros:
- 'radix_tree_for_each_slot'
- 'radix_tree_for_each_tagged'
- 'rbtree_postorder_for_each_entry_safe'
- 'rdma_for_each_port'
- 'resource_list_for_each_entry'
- 'resource_list_for_each_entry_safe'
- 'rhl_for_each_entry_rcu'
......
Documentation/infiniband/user_verbs.txt
浏览文件 @ a50243b1
......@@ -46,11 +46,11 @@ Memory pinning
I/O targets be kept resident at the same physical address. The
ib_uverbs module manages pinning and unpinning memory regions via
get_user_pages() and put_page() calls. It also accounts for the
amount of memory pinned in the process's locked_vm, and checks that
amount of memory pinned in the process's pinned_vm, and checks that
unprivileged processes do not exceed their RLIMIT_MEMLOCK limit.
Pages that are pinned multiple times are counted each time they are
pinned, so the value of locked_vm may be an overestimate of the
pinned, so the value of pinned_vm may be an overestimate of the
number of pages pinned by a process.
/dev files
......
drivers/gpu/drm/vmwgfx/vmwgfx_ttm_buffer.c
浏览文件 @ a50243b1
......@@ -311,7 +311,13 @@ static dma_addr_t __vmw_piter_dma_addr(struct vmw_piter *viter)
static dma_addr_t __vmw_piter_sg_addr(struct vmw_piter *viter)
{
return sg_page_iter_dma_address(&viter->iter);
/*
* FIXME: This driver wrongly mixes DMA and CPU SG list iteration and
* needs revision. See
* https://lore.kernel.org/lkml/20190104223531.GA1705@ziepe.ca/
*/
return sg_page_iter_dma_address(
container_of(&viter->iter, struct sg_dma_page_iter, base));
}
......
drivers/infiniband/Kconfig
浏览文件 @ a50243b1
......@@ -89,6 +89,7 @@ config INFINIBAND_ADDR_TRANS_CONFIGFS
This allows the user to config the default GID type that the CM
uses for each device, when initiaing new connections.
if INFINIBAND_USER_ACCESS || !INFINIBAND_USER_ACCESS
source "drivers/infiniband/hw/mthca/Kconfig"
source "drivers/infiniband/hw/qib/Kconfig"
source "drivers/infiniband/hw/cxgb3/Kconfig"
......@@ -101,6 +102,12 @@ source "drivers/infiniband/hw/ocrdma/Kconfig"
source "drivers/infiniband/hw/vmw_pvrdma/Kconfig"
source "drivers/infiniband/hw/usnic/Kconfig"
source "drivers/infiniband/hw/hns/Kconfig"
source "drivers/infiniband/hw/bnxt_re/Kconfig"
source "drivers/infiniband/hw/hfi1/Kconfig"
source "drivers/infiniband/hw/qedr/Kconfig"
source "drivers/infiniband/sw/rdmavt/Kconfig"
source "drivers/infiniband/sw/rxe/Kconfig"
endif
source "drivers/infiniband/ulp/ipoib/Kconfig"
......@@ -111,13 +118,5 @@ source "drivers/infiniband/ulp/iser/Kconfig"
source "drivers/infiniband/ulp/isert/Kconfig"
source "drivers/infiniband/ulp/opa_vnic/Kconfig"
source "drivers/infiniband/sw/rdmavt/Kconfig"
source "drivers/infiniband/sw/rxe/Kconfig"
source "drivers/infiniband/hw/hfi1/Kconfig"
source "drivers/infiniband/hw/qedr/Kconfig"
source "drivers/infiniband/hw/bnxt_re/Kconfig"
endif # INFINIBAND
drivers/infiniband/core/Makefile
浏览文件 @ a50243b1
......@@ -15,8 +15,6 @@ ib_core-y := packer.o ud_header.o verbs.o cq.o rw.o sysfs.o \
nldev.o restrack.o
ib_core-$(CONFIG_SECURITY_INFINIBAND) += security.o
ib_core-$(CONFIG_INFINIBAND_USER_MEM) += umem.o
ib_core-$(CONFIG_INFINIBAND_ON_DEMAND_PAGING) += umem_odp.o
ib_core-$(CONFIG_CGROUP_RDMA) += cgroup.o
ib_cm-y := cm.o
......@@ -39,3 +37,5 @@ ib_uverbs-y := uverbs_main.o uverbs_cmd.o uverbs_marshall.o \
uverbs_std_types_flow_action.o uverbs_std_types_dm.o \
uverbs_std_types_mr.o uverbs_std_types_counters.o \
uverbs_uapi.o uverbs_std_types_device.o
ib_uverbs-$(CONFIG_INFINIBAND_USER_MEM) += umem.o
ib_uverbs-$(CONFIG_INFINIBAND_ON_DEMAND_PAGING) += umem_odp.o
drivers/infiniband/core/cache.c
浏览文件 @ a50243b1
......@@ -185,7 +185,7 @@ EXPORT_SYMBOL(ib_cache_gid_parse_type_str);
static struct ib_gid_table *rdma_gid_table(struct ib_device *device, u8 port)
{
return device->cache.ports[port - rdma_start_port(device)].gid;
return device->port_data[port].cache.gid;
}
static bool is_gid_entry_free(const struct ib_gid_table_entry *entry)
......@@ -547,21 +547,19 @@ int ib_cache_gid_add(struct ib_device *ib_dev, u8 port,
unsigned long mask;
int ret;
if (ib_dev->ops.get_netdev) {
idev = ib_dev->ops.get_netdev(ib_dev, port);
if (idev && attr->ndev != idev) {
union ib_gid default_gid;
idev = ib_device_get_netdev(ib_dev, port);
if (idev && attr->ndev != idev) {
union ib_gid default_gid;
/* Adding default GIDs in not permitted */
make_default_gid(idev, &default_gid);
if (!memcmp(gid, &default_gid, sizeof(*gid))) {
dev_put(idev);
return -EPERM;
}
}
if (idev)
/* Adding default GIDs is not permitted */
make_default_gid(idev, &default_gid);
if (!memcmp(gid, &default_gid, sizeof(*gid))) {
dev_put(idev);
return -EPERM;
}
}
if (idev)
dev_put(idev);
mask = GID_ATTR_FIND_MASK_GID |
GID_ATTR_FIND_MASK_GID_TYPE |
......@@ -765,7 +763,7 @@ static struct ib_gid_table *alloc_gid_table(int sz)
return NULL;
}
static void release_gid_table(struct ib_device *device, u8 port,
static void release_gid_table(struct ib_device *device,
struct ib_gid_table *table)
{
bool leak = false;
......@@ -863,31 +861,27 @@ static void gid_table_reserve_default(struct ib_device *ib_dev, u8 port,
static void gid_table_release_one(struct ib_device *ib_dev)
{
struct ib_gid_table *table;
u8 port;
unsigned int p;
for (port = 0; port < ib_dev->phys_port_cnt; port++) {
table = ib_dev->cache.ports[port].gid;
release_gid_table(ib_dev, port, table);
ib_dev->cache.ports[port].gid = NULL;
rdma_for_each_port (ib_dev, p) {
release_gid_table(ib_dev, ib_dev->port_data[p].cache.gid);
ib_dev->port_data[p].cache.gid = NULL;
}
}
static int _gid_table_setup_one(struct ib_device *ib_dev)
{
u8 port;
struct ib_gid_table *table;
unsigned int rdma_port;
for (port = 0; port < ib_dev->phys_port_cnt; port++) {
u8 rdma_port = port + rdma_start_port(ib_dev);
table = alloc_gid_table(
ib_dev->port_immutable[rdma_port].gid_tbl_len);
rdma_for_each_port (ib_dev, rdma_port) {
table = alloc_gid_table(
ib_dev->port_data[rdma_port].immutable.gid_tbl_len);
if (!table)
goto rollback_table_setup;
gid_table_reserve_default(ib_dev, rdma_port, table);
ib_dev->cache.ports[port].gid = table;
ib_dev->port_data[rdma_port].cache.gid = table;
}
return 0;
......@@ -898,14 +892,11 @@ static int _gid_table_setup_one(struct ib_device *ib_dev)
static void gid_table_cleanup_one(struct ib_device *ib_dev)
{
struct ib_gid_table *table;
u8 port;
unsigned int p;
for (port = 0; port < ib_dev->phys_port_cnt; port++) {
table = ib_dev->cache.ports[port].gid;
cleanup_gid_table_port(ib_dev, port + rdma_start_port(ib_dev),
table);
}
rdma_for_each_port (ib_dev, p)
cleanup_gid_table_port(ib_dev, p,
ib_dev->port_data[p].cache.gid);
}
static int gid_table_setup_one(struct ib_device *ib_dev)
......@@ -983,17 +974,17 @@ const struct ib_gid_attr *rdma_find_gid(struct ib_device *device,
unsigned long mask = GID_ATTR_FIND_MASK_GID |
GID_ATTR_FIND_MASK_GID_TYPE;
struct ib_gid_attr gid_attr_val = {.ndev = ndev, .gid_type = gid_type};
u8 p;
unsigned int p;
if (ndev)
mask |= GID_ATTR_FIND_MASK_NETDEV;
for (p = 0; p < device->phys_port_cnt; p++) {
rdma_for_each_port(device, p) {
struct ib_gid_table *table;
unsigned long flags;
int index;
table = device->cache.ports[p].gid;
table = device->port_data[p].cache.gid;
read_lock_irqsave(&table->rwlock, flags);
index = find_gid(table, gid, &gid_attr_val, false, mask, NULL);
if (index >= 0) {
......@@ -1025,7 +1016,7 @@ int ib_get_cached_pkey(struct ib_device *device,
read_lock_irqsave(&device->cache.lock, flags);
cache = device->cache.ports[port_num - rdma_start_port(device)].pkey;
cache = device->port_data[port_num].cache.pkey;
if (index < 0 || index >= cache->table_len)
ret = -EINVAL;
......@@ -1043,14 +1034,12 @@ int ib_get_cached_subnet_prefix(struct ib_device *device,
u64 *sn_pfx)
{
unsigned long flags;
int p;
if (!rdma_is_port_valid(device, port_num))
return -EINVAL;
p = port_num - rdma_start_port(device);
read_lock_irqsave(&device->cache.lock, flags);
*sn_pfx = device->cache.ports[p].subnet_prefix;
*sn_pfx = device->port_data[port_num].cache.subnet_prefix;
read_unlock_irqrestore(&device->cache.lock, flags);
return 0;
......@@ -1073,7 +1062,7 @@ int ib_find_cached_pkey(struct ib_device *device,
read_lock_irqsave(&device->cache.lock, flags);
cache = device->cache.ports[port_num - rdma_start_port(device)].pkey;
cache = device->port_data[port_num].cache.pkey;
*index = -1;
......@@ -1113,7 +1102,7 @@ int ib_find_exact_cached_pkey(struct ib_device *device,
read_lock_irqsave(&device->cache.lock, flags);
cache = device->cache.ports[port_num - rdma_start_port(device)].pkey;
cache = device->port_data[port_num].cache.pkey;
*index = -1;
......@@ -1141,7 +1130,7 @@ int ib_get_cached_lmc(struct ib_device *device,
return -EINVAL;
read_lock_irqsave(&device->cache.lock, flags);
*lmc = device->cache.ports[port_num - rdma_start_port(device)].lmc;
*lmc = device->port_data[port_num].cache.lmc;
read_unlock_irqrestore(&device->cache.lock, flags);
return ret;
......@@ -1159,8 +1148,7 @@ int ib_get_cached_port_state(struct ib_device *device,
return -EINVAL;
read_lock_irqsave(&device->cache.lock, flags);
*port_state = device->cache.ports[port_num
- rdma_start_port(device)].port_state;
*port_state = device->port_data[port_num].cache.port_state;
read_unlock_irqrestore(&device->cache.lock, flags);
return ret;
......@@ -1361,16 +1349,13 @@ static void ib_cache_update(struct ib_device *device,
write_lock_irq(&device->cache.lock);
old_pkey_cache = device->cache.ports[port -
rdma_start_port(device)].pkey;
old_pkey_cache = device->port_data[port].cache.pkey;
device->cache.ports[port - rdma_start_port(device)].pkey = pkey_cache;
device->cache.ports[port - rdma_start_port(device)].lmc = tprops->lmc;
device->cache.ports[port - rdma_start_port(device)].port_state =
tprops->state;
device->port_data[port].cache.pkey = pkey_cache;
device->port_data[port].cache.lmc = tprops->lmc;
device->port_data[port].cache.port_state = tprops->state;
device->cache.ports[port - rdma_start_port(device)].subnet_prefix =
tprops->subnet_prefix;
device->port_data[port].cache.subnet_prefix = tprops->subnet_prefix;
write_unlock_irq(&device->cache.lock);
if (enforce_security)
......@@ -1428,27 +1413,17 @@ static void ib_cache_event(struct ib_event_handler *handler,
int ib_cache_setup_one(struct ib_device *device)
{
int p;
unsigned int p;
int err;
rwlock_init(&device->cache.lock);
device->cache.ports =
kcalloc(rdma_end_port(device) - rdma_start_port(device) + 1,
sizeof(*device->cache.ports),
GFP_KERNEL);
if (!device->cache.ports)
return -ENOMEM;
err = gid_table_setup_one(device);
if (err) {
kfree(device->cache.ports);
device->cache.ports = NULL;
if (err)
return err;
}
for (p = 0; p <= rdma_end_port(device) - rdma_start_port(device); ++p)
ib_cache_update(device, p + rdma_start_port(device), true);
rdma_for_each_port (device, p)
ib_cache_update(device, p, true);
INIT_IB_EVENT_HANDLER(&device->cache.event_handler,
device, ib_cache_event);
......@@ -1458,7 +1433,7 @@ int ib_cache_setup_one(struct ib_device *device)
void ib_cache_release_one(struct ib_device *device)
{
int p;
unsigned int p;
/*
* The release function frees all the cache elements.
......@@ -1466,11 +1441,10 @@ void ib_cache_release_one(struct ib_device *device)
* all the device's resources when the cache could no
* longer be accessed.
*/
for (p = 0; p <= rdma_end_port(device) - rdma_start_port(device); ++p)
kfree(device->cache.ports[p].pkey);
rdma_for_each_port (device, p)
kfree(device->port_data[p].cache.pkey);
gid_table_release_one(device);
kfree(device->cache.ports);
}
void ib_cache_cleanup_one(struct ib_device *device)
......
drivers/infiniband/core/cgroup.c
浏览文件 @ a50243b1
......@@ -21,12 +21,11 @@
* Register with the rdma cgroup. Should be called before
* exposing rdma device to user space applications to avoid
* resource accounting leak.
* Returns 0 on success or otherwise failure code.
*/
int ib_device_register_rdmacg(struct ib_device *device)
void ib_device_register_rdmacg(struct ib_device *device)
{
device->cg_device.name = device->name;
return rdmacg_register_device(&device->cg_device);
rdmacg_register_device(&device->cg_device);
}
/**
......
drivers/infiniband/core/cm.c
浏览文件 @ a50243b1
......@@ -4052,8 +4052,7 @@ static void cm_recv_handler(struct ib_mad_agent *mad_agent,
atomic_long_inc(&port->counter_group[CM_RECV].
counter[attr_id - CM_ATTR_ID_OFFSET]);
work = kmalloc(sizeof(*work) + sizeof(struct sa_path_rec) * paths,
GFP_KERNEL);
work = kmalloc(struct_size(work, path, paths), GFP_KERNEL);
if (!work) {
ib_free_recv_mad(mad_recv_wc);
return;
......
drivers/infiniband/core/cma.c
浏览文件 @ a50243b1
......@@ -659,7 +659,7 @@ static int cma_acquire_dev_by_src_ip(struct rdma_id_private *id_priv)
struct cma_device *cma_dev;
enum ib_gid_type gid_type;
int ret = -ENODEV;
u8 port;
unsigned int port;
if (dev_addr->dev_type != ARPHRD_INFINIBAND &&
id_priv->id.ps == RDMA_PS_IPOIB)
......@@ -673,8 +673,7 @@ static int cma_acquire_dev_by_src_ip(struct rdma_id_private *id_priv)
mutex_lock(&lock);
list_for_each_entry(cma_dev, &dev_list, list) {
for (port = rdma_start_port(cma_dev->device);
port <= rdma_end_port(cma_dev->device); port++) {
rdma_for_each_port (cma_dev->device, port) {
gidp = rdma_protocol_roce(cma_dev->device, port) ?
&iboe_gid : &gid;
gid_type = cma_dev->default_gid_type[port - 1];
......@@ -888,6 +887,7 @@ struct rdma_cm_id *__rdma_create_id(struct net *net,
id_priv->id.ps = ps;
id_priv->id.qp_type = qp_type;
id_priv->tos_set = false;
id_priv->timeout_set = false;
id_priv->gid_type = IB_GID_TYPE_IB;
spin_lock_init(&id_priv->lock);
mutex_init(&id_priv->qp_mutex);
......@@ -1130,6 +1130,9 @@ int rdma_init_qp_attr(struct rdma_cm_id *id, struct ib_qp_attr *qp_attr,
} else
ret = -ENOSYS;
if ((*qp_attr_mask & IB_QP_TIMEOUT) && id_priv->timeout_set)
qp_attr->timeout = id_priv->timeout;
return ret;
}
EXPORT_SYMBOL(rdma_init_qp_attr);
......@@ -2410,6 +2413,7 @@ static int cma_iw_listen(struct rdma_id_private *id_priv, int backlog)
return PTR_ERR(id);
id->tos = id_priv->tos;
id->tos_set = id_priv->tos_set;
id_priv->cm_id.iw = id;
memcpy(&id_priv->cm_id.iw->local_addr, cma_src_addr(id_priv),
......@@ -2462,6 +2466,8 @@ static void cma_listen_on_dev(struct rdma_id_private *id_priv,
atomic_inc(&id_priv->refcount);
dev_id_priv->internal_id = 1;
dev_id_priv->afonly = id_priv->afonly;
dev_id_priv->tos_set = id_priv->tos_set;
dev_id_priv->tos = id_priv->tos;
ret = rdma_listen(id, id_priv->backlog);
if (ret)
......@@ -2490,6 +2496,34 @@ void rdma_set_service_type(struct rdma_cm_id *id, int tos)
}
EXPORT_SYMBOL(rdma_set_service_type);
/**
* rdma_set_ack_timeout() - Set the ack timeout of QP associated
* with a connection identifier.
* @id: Communication identifier to associated with service type.
* @timeout: Ack timeout to set a QP, expressed as 4.096 * 2^(timeout) usec.
*
* This function should be called before rdma_connect() on active side,
* and on passive side before rdma_accept(). It is applicable to primary
* path only. The timeout will affect the local side of the QP, it is not
* negotiated with remote side and zero disables the timer.
*
* Return: 0 for success
*/
int rdma_set_ack_timeout(struct rdma_cm_id *id, u8 timeout)
{
struct rdma_id_private *id_priv;
if (id->qp_type != IB_QPT_RC)
return -EINVAL;
id_priv = container_of(id, struct rdma_id_private, id);
id_priv->timeout = timeout;
id_priv->timeout_set = true;
return 0;
}
EXPORT_SYMBOL(rdma_set_ack_timeout);
static void cma_query_handler(int status, struct sa_path_rec *path_rec,
void *context)
{
......@@ -2966,13 +3000,22 @@ static void addr_handler(int status, struct sockaddr *src_addr,
{
struct rdma_id_private *id_priv = context;
struct rdma_cm_event event = {};
struct sockaddr *addr;
struct sockaddr_storage old_addr;
mutex_lock(&id_priv->handler_mutex);
if (!cma_comp_exch(id_priv, RDMA_CM_ADDR_QUERY,
RDMA_CM_ADDR_RESOLVED))
goto out;
memcpy(cma_src_addr(id_priv), src_addr, rdma_addr_size(src_addr));
/*
* Store the previous src address, so that if we fail to acquire
* matching rdma device, old address can be restored back, which helps
* to cancel the cma listen operation correctly.
*/
addr = cma_src_addr(id_priv);
memcpy(&old_addr, addr, rdma_addr_size(addr));
memcpy(addr, src_addr, rdma_addr_size(src_addr));
if (!status && !id_priv->cma_dev) {
status = cma_acquire_dev_by_src_ip(id_priv);
if (status)
......@@ -2983,6 +3026,8 @@ static void addr_handler(int status, struct sockaddr *src_addr,
}
if (status) {
memcpy(addr, &old_addr,
rdma_addr_size((struct sockaddr *)&old_addr));
if (!cma_comp_exch(id_priv, RDMA_CM_ADDR_RESOLVED,
RDMA_CM_ADDR_BOUND))
goto out;
......@@ -3798,6 +3843,7 @@ static int cma_connect_iw(struct rdma_id_private *id_priv,
return PTR_ERR(cm_id);
cm_id->tos = id_priv->tos;
cm_id->tos_set = id_priv->tos_set;
id_priv->cm_id.iw = cm_id;
memcpy(&cm_id->local_addr, cma_src_addr(id_priv),
......@@ -4501,7 +4547,7 @@ static void cma_add_one(struct ib_device *device)
if (!cma_dev->default_roce_tos)
goto free_gid_type;
for (i = rdma_start_port(device); i <= rdma_end_port(device); i++) {
rdma_for_each_port (device, i) {
supported_gids = roce_gid_type_mask_support(device, i);
WARN_ON(!supported_gids);
if (supported_gids & (1 << CMA_PREFERRED_ROCE_GID_TYPE))
......@@ -4605,85 +4651,6 @@ static void cma_remove_one(struct ib_device *device, void *client_data)
kfree(cma_dev);
}
static int cma_get_id_stats(struct sk_buff *skb, struct netlink_callback *cb)
{
struct nlmsghdr *nlh;
struct rdma_cm_id_stats *id_stats;
struct rdma_id_private *id_priv;
struct rdma_cm_id *id = NULL;
struct cma_device *cma_dev;
int i_dev = 0, i_id = 0;
/*
* We export all of the IDs as a sequence of messages. Each
* ID gets its own netlink message.
*/
mutex_lock(&lock);
list_for_each_entry(cma_dev, &dev_list, list) {
if (i_dev < cb->args[0]) {
i_dev++;
continue;
}
i_id = 0;
list_for_each_entry(id_priv, &cma_dev->id_list, list) {
if (i_id < cb->args[1]) {
i_id++;
continue;
}
id_stats = ibnl_put_msg(skb, &nlh, cb->nlh->nlmsg_seq,
sizeof *id_stats, RDMA_NL_RDMA_CM,
RDMA_NL_RDMA_CM_ID_STATS,
NLM_F_MULTI);
if (!id_stats)
goto out;
memset(id_stats, 0, sizeof *id_stats);
id = &id_priv->id;
id_stats->node_type = id->route.addr.dev_addr.dev_type;
id_stats->port_num = id->port_num;
id_stats->bound_dev_if =
id->route.addr.dev_addr.bound_dev_if;
if (ibnl_put_attr(skb, nlh,
rdma_addr_size(cma_src_addr(id_priv)),
cma_src_addr(id_priv),
RDMA_NL_RDMA_CM_ATTR_SRC_ADDR))
goto out;
if (ibnl_put_attr(skb, nlh,
rdma_addr_size(cma_dst_addr(id_priv)),
cma_dst_addr(id_priv),
RDMA_NL_RDMA_CM_ATTR_DST_ADDR))
goto out;
id_stats->pid = task_pid_vnr(id_priv->res.task);
id_stats->port_space = id->ps;
id_stats->cm_state = id_priv->state;
id_stats->qp_num = id_priv->qp_num;
id_stats->qp_type = id->qp_type;
i_id++;
nlmsg_end(skb, nlh);
}
cb->args[1] = 0;
i_dev++;
}
out:
mutex_unlock(&lock);
cb->args[0] = i_dev;
cb->args[1] = i_id;
return skb->len;
}
static const struct rdma_nl_cbs cma_cb_table[RDMA_NL_RDMA_CM_NUM_OPS] = {
[RDMA_NL_RDMA_CM_ID_STATS] = { .dump = cma_get_id_stats},
};
static int cma_init_net(struct net *net)
{
struct cma_pernet *pernet = cma_pernet(net);
......@@ -4732,7 +4699,6 @@ static int __init cma_init(void)
if (ret)
goto err;
rdma_nl_register(RDMA_NL_RDMA_CM, cma_cb_table);
cma_configfs_init();
return 0;
......@@ -4748,7 +4714,6 @@ static int __init cma_init(void)
static void __exit cma_cleanup(void)
{
cma_configfs_exit();
rdma_nl_unregister(RDMA_NL_RDMA_CM);
ib_unregister_client(&cma_client);
unregister_netdevice_notifier(&cma_nb);
ib_sa_unregister_client(&sa_client);
......@@ -4756,7 +4721,5 @@ static void __exit cma_cleanup(void)
destroy_workqueue(cma_wq);
}
MODULE_ALIAS_RDMA_NETLINK(RDMA_NL_RDMA_CM, 1);
module_init(cma_init);
module_exit(cma_cleanup);
drivers/infiniband/core/cma_priv.h
浏览文件 @ a50243b1
......@@ -84,9 +84,11 @@ struct rdma_id_private {
u32 options;
u8 srq;
u8 tos;
bool tos_set;
u8 tos_set:1;
u8 timeout_set:1;
u8 reuseaddr;
u8 afonly;
u8 timeout;
enum ib_gid_type gid_type;
/*
......
drivers/infiniband/core/core_priv.h
浏览文件 @ a50243b1
......@@ -54,9 +54,9 @@ struct pkey_index_qp_list {
struct list_head qp_list;
};
int ib_device_register_sysfs(struct ib_device *device,
int (*port_callback)(struct ib_device *,
u8, struct kobject *));
extern const struct attribute_group ib_dev_attr_group;
int ib_device_register_sysfs(struct ib_device *device);
void ib_device_unregister_sysfs(struct ib_device *device);
int ib_device_rename(struct ib_device *ibdev, const char *name);
......@@ -66,6 +66,9 @@ typedef void (*roce_netdev_callback)(struct ib_device *device, u8 port,
typedef bool (*roce_netdev_filter)(struct ib_device *device, u8 port,
struct net_device *idev, void *cookie);
struct net_device *ib_device_get_netdev(struct ib_device *ib_dev,
unsigned int port);
void ib_enum_roce_netdev(struct ib_device *ib_dev,
roce_netdev_filter filter,
void *filter_cookie,
......@@ -117,7 +120,7 @@ void ib_cache_cleanup_one(struct ib_device *device);
void ib_cache_release_one(struct ib_device *device);
#ifdef CONFIG_CGROUP_RDMA
int ib_device_register_rdmacg(struct ib_device *device);
void ib_device_register_rdmacg(struct ib_device *device);
void ib_device_unregister_rdmacg(struct ib_device *device);
int ib_rdmacg_try_charge(struct ib_rdmacg_object *cg_obj,
......@@ -128,21 +131,26 @@ void ib_rdmacg_uncharge(struct ib_rdmacg_object *cg_obj,
struct ib_device *device,
enum rdmacg_resource_type resource_index);
#else
static inline int ib_device_register_rdmacg(struct ib_device *device)
{ return 0; }
static inline void ib_device_register_rdmacg(struct ib_device *device)
{
}
static inline void ib_device_unregister_rdmacg(struct ib_device *device)
{ }
{
}
static inline int ib_rdmacg_try_charge(struct ib_rdmacg_object *cg_obj,
struct ib_device *device,
enum rdmacg_resource_type resource_index)
{ return 0; }
{
return 0;
}
static inline void ib_rdmacg_uncharge(struct ib_rdmacg_object *cg_obj,
struct ib_device *device,
enum rdmacg_resource_type resource_index)
{ }
{
}
#endif
static inline bool rdma_is_upper_dev_rcu(struct net_device *dev,
......@@ -178,7 +186,7 @@ int ib_get_cached_subnet_prefix(struct ib_device *device,
u64 *sn_pfx);
#ifdef CONFIG_SECURITY_INFINIBAND
void ib_security_destroy_port_pkey_list(struct ib_device *device);
void ib_security_release_port_pkey_list(struct ib_device *device);
void ib_security_cache_change(struct ib_device *device,
u8 port_num,
......@@ -199,8 +207,9 @@ int ib_mad_agent_security_setup(struct ib_mad_agent *agent,
enum ib_qp_type qp_type);
void ib_mad_agent_security_cleanup(struct ib_mad_agent *agent);
int ib_mad_enforce_security(struct ib_mad_agent_private *map, u16 pkey_index);
void ib_mad_agent_security_change(void);
#else
static inline void ib_security_destroy_port_pkey_list(struct ib_device *device)
static inline void ib_security_release_port_pkey_list(struct ib_device *device)
{
}
......@@ -264,6 +273,10 @@ static inline int ib_mad_enforce_security(struct ib_mad_agent_private *map,
{
return 0;
}
static inline void ib_mad_agent_security_change(void)
{
}
#endif
struct ib_device *ib_device_get_by_index(u32 ifindex);
......
drivers/infiniband/core/device.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/core/iwcm.c
浏览文件 @ a50243b1
......@@ -87,7 +87,8 @@ static struct rdma_nl_cbs iwcm_nl_cb_table[RDMA_NL_IWPM_NUM_OPS] = {
[RDMA_NL_IWPM_REMOTE_INFO] = {.dump = iwpm_remote_info_cb},
[RDMA_NL_IWPM_HANDLE_ERR] = {.dump = iwpm_mapping_error_cb},
[RDMA_NL_IWPM_MAPINFO] = {.dump = iwpm_mapping_info_cb},
[RDMA_NL_IWPM_MAPINFO_NUM] = {.dump = iwpm_ack_mapping_info_cb}
[RDMA_NL_IWPM_MAPINFO_NUM] = {.dump = iwpm_ack_mapping_info_cb},
[RDMA_NL_IWPM_HELLO] = {.dump = iwpm_hello_cb}
};
static struct workqueue_struct *iwcm_wq;
......@@ -504,7 +505,7 @@ static int iw_cm_map(struct iw_cm_id *cm_id, bool active)
{
const char *devname = dev_name(&cm_id->device->dev);
const char *ifname = cm_id->device->iwcm->ifname;
struct iwpm_dev_data pm_reg_msg;
struct iwpm_dev_data pm_reg_msg = {};
struct iwpm_sa_data pm_msg;
int status;
......@@ -515,8 +516,8 @@ static int iw_cm_map(struct iw_cm_id *cm_id, bool active)
cm_id->m_local_addr = cm_id->local_addr;
cm_id->m_remote_addr = cm_id->remote_addr;
strncpy(pm_reg_msg.dev_name, devname, sizeof(pm_reg_msg.dev_name));
strncpy(pm_reg_msg.if_name, ifname, sizeof(pm_reg_msg.if_name));
strcpy(pm_reg_msg.dev_name, devname);
strcpy(pm_reg_msg.if_name, ifname);
if (iwpm_register_pid(&pm_reg_msg, RDMA_NL_IWCM) ||
!iwpm_valid_pid())
......@@ -525,6 +526,8 @@ static int iw_cm_map(struct iw_cm_id *cm_id, bool active)
cm_id->mapped = true;
pm_msg.loc_addr = cm_id->local_addr;
pm_msg.rem_addr = cm_id->remote_addr;
pm_msg.flags = (cm_id->device->iwcm->driver_flags & IW_F_NO_PORT_MAP) ?
IWPM_FLAGS_NO_PORT_MAP : 0;
if (active)
status = iwpm_add_and_query_mapping(&pm_msg,
RDMA_NL_IWCM);
......@@ -543,7 +546,7 @@ static int iw_cm_map(struct iw_cm_id *cm_id, bool active)
return iwpm_create_mapinfo(&cm_id->local_addr,
&cm_id->m_local_addr,
RDMA_NL_IWCM);
RDMA_NL_IWCM, pm_msg.flags);
}
/*
......
drivers/infiniband/core/iwpm_msg.c
浏览文件 @ a50243b1
......@@ -34,18 +34,25 @@
#include "iwpm_util.h"
static const char iwpm_ulib_name[IWPM_ULIBNAME_SIZE] = "iWarpPortMapperUser";
static int iwpm_ulib_version = 3;
u16 iwpm_ulib_version = IWPM_UABI_VERSION_MIN;
static int iwpm_user_pid = IWPM_PID_UNDEFINED;
static atomic_t echo_nlmsg_seq;
/**
* iwpm_valid_pid - Check if the userspace iwarp port mapper pid is valid
*
* Returns true if the pid is greater than zero, otherwise returns false
*/
int iwpm_valid_pid(void)
{
return iwpm_user_pid > 0;
}
/*
* iwpm_register_pid - Send a netlink query to user space
* for the iwarp port mapper pid
/**
* iwpm_register_pid - Send a netlink query to userspace
* to get the iwarp port mapper pid
* @pm_msg: Contains driver info to send to the userspace port mapper
* @nl_client: The index of the netlink client
*
* nlmsg attributes:
* [IWPM_NLA_REG_PID_SEQ]
......@@ -124,12 +131,19 @@ int iwpm_register_pid(struct iwpm_dev_data *pm_msg, u8 nl_client)
return ret;
}
/*
* iwpm_add_mapping - Send a netlink add mapping message
* to the port mapper
/**
* iwpm_add_mapping - Send a netlink add mapping request to
* the userspace port mapper
* @pm_msg: Contains the local ip/tcp address info to send
* @nl_client: The index of the netlink client
*
* nlmsg attributes:
* [IWPM_NLA_MANAGE_MAPPING_SEQ]
* [IWPM_NLA_MANAGE_ADDR]
* [IWPM_NLA_MANAGE_FLAGS]
*
* If the request is successful, the pm_msg stores
* the port mapper response (mapped address info)
*/
int iwpm_add_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client)
{
......@@ -173,6 +187,18 @@ int iwpm_add_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client)
if (ret)
goto add_mapping_error;
/* If flags are required and we're not V4, then return a quiet error */
if (pm_msg->flags && iwpm_ulib_version == IWPM_UABI_VERSION_MIN) {
ret = -EINVAL;
goto add_mapping_error_nowarn;
}
if (iwpm_ulib_version > IWPM_UABI_VERSION_MIN) {
ret = ibnl_put_attr(skb, nlh, sizeof(u32), &pm_msg->flags,
IWPM_NLA_MANAGE_FLAGS);
if (ret)
goto add_mapping_error;
}
nlmsg_end(skb, nlh);
nlmsg_request->req_buffer = pm_msg;
......@@ -187,6 +213,7 @@ int iwpm_add_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client)
return ret;
add_mapping_error:
pr_info("%s: %s (client = %d)\n", __func__, err_str, nl_client);
add_mapping_error_nowarn:
if (skb)
dev_kfree_skb(skb);
if (nlmsg_request)
......@@ -194,13 +221,17 @@ int iwpm_add_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client)
return ret;
}
/*
* iwpm_add_and_query_mapping - Send a netlink add and query
* mapping message to the port mapper
/**
* iwpm_add_and_query_mapping - Process the port mapper response to
* iwpm_add_and_query_mapping request
* @pm_msg: Contains the local ip/tcp address info to send
* @nl_client: The index of the netlink client
*
* nlmsg attributes:
* [IWPM_NLA_QUERY_MAPPING_SEQ]
* [IWPM_NLA_QUERY_LOCAL_ADDR]
* [IWPM_NLA_QUERY_REMOTE_ADDR]
* [IWPM_NLA_QUERY_FLAGS]
*/
int iwpm_add_and_query_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client)
{
......@@ -251,6 +282,18 @@ int iwpm_add_and_query_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client)
if (ret)
goto query_mapping_error;
/* If flags are required and we're not V4, then return a quite error */
if (pm_msg->flags && iwpm_ulib_version == IWPM_UABI_VERSION_MIN) {
ret = -EINVAL;
goto query_mapping_error_nowarn;
}
if (iwpm_ulib_version > IWPM_UABI_VERSION_MIN) {
ret = ibnl_put_attr(skb, nlh, sizeof(u32), &pm_msg->flags,
IWPM_NLA_QUERY_FLAGS);
if (ret)
goto query_mapping_error;
}
nlmsg_end(skb, nlh);
nlmsg_request->req_buffer = pm_msg;
......@@ -264,6 +307,7 @@ int iwpm_add_and_query_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client)
return ret;
query_mapping_error:
pr_info("%s: %s (client = %d)\n", __func__, err_str, nl_client);
query_mapping_error_nowarn:
if (skb)
dev_kfree_skb(skb);
if (nlmsg_request)
......@@ -271,9 +315,13 @@ int iwpm_add_and_query_mapping(struct iwpm_sa_data *pm_msg, u8 nl_client)
return ret;
}
/*
* iwpm_remove_mapping - Send a netlink remove mapping message
* to the port mapper
/**
* iwpm_remove_mapping - Send a netlink remove mapping request
* to the userspace port mapper
*
* @local_addr: Local ip/tcp address to remove
* @nl_client: The index of the netlink client
*
* nlmsg attributes:
* [IWPM_NLA_MANAGE_MAPPING_SEQ]
* [IWPM_NLA_MANAGE_ADDR]
......@@ -344,9 +392,14 @@ static const struct nla_policy resp_reg_policy[IWPM_NLA_RREG_PID_MAX] = {
[IWPM_NLA_RREG_PID_ERR] = { .type = NLA_U16 }
};
/*
* iwpm_register_pid_cb - Process a port mapper response to
* iwpm_register_pid()
/**
* iwpm_register_pid_cb - Process the port mapper response to
* iwpm_register_pid query
* @skb:
* @cb: Contains the received message (payload and netlink header)
*
* If successful, the function receives the userspace port mapper pid
* which is used in future communication with the port mapper
*/
int iwpm_register_pid_cb(struct sk_buff *skb, struct netlink_callback *cb)
{
......@@ -379,7 +432,7 @@ int iwpm_register_pid_cb(struct sk_buff *skb, struct netlink_callback *cb)
/* check device name, ulib name and version */
if (strcmp(pm_msg->dev_name, dev_name) ||
strcmp(iwpm_ulib_name, iwpm_name) ||
iwpm_version != iwpm_ulib_version) {
iwpm_version < IWPM_UABI_VERSION_MIN) {
pr_info("%s: Incorrect info (dev = %s name = %s version = %d)\n",
__func__, dev_name, iwpm_name, iwpm_version);
......@@ -387,6 +440,10 @@ int iwpm_register_pid_cb(struct sk_buff *skb, struct netlink_callback *cb)
goto register_pid_response_exit;
}
iwpm_user_pid = cb->nlh->nlmsg_pid;
iwpm_ulib_version = iwpm_version;
if (iwpm_ulib_version < IWPM_UABI_VERSION)
pr_warn_once("%s: Down level iwpmd/pid %u. Continuing...",
__func__, iwpm_user_pid);
atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq);
pr_debug("%s: iWarp Port Mapper (pid = %d) is available!\n",
__func__, iwpm_user_pid);
......@@ -403,15 +460,19 @@ int iwpm_register_pid_cb(struct sk_buff *skb, struct netlink_callback *cb)
/* netlink attribute policy for the received response to add mapping request */
static const struct nla_policy resp_add_policy[IWPM_NLA_RMANAGE_MAPPING_MAX] = {
[IWPM_NLA_MANAGE_MAPPING_SEQ] = { .type = NLA_U32 },
[IWPM_NLA_MANAGE_ADDR] = { .len = sizeof(struct sockaddr_storage) },
[IWPM_NLA_MANAGE_MAPPED_LOC_ADDR] = { .len = sizeof(struct sockaddr_storage) },
[IWPM_NLA_RMANAGE_MAPPING_ERR] = { .type = NLA_U16 }
[IWPM_NLA_RMANAGE_MAPPING_SEQ] = { .type = NLA_U32 },
[IWPM_NLA_RMANAGE_ADDR] = {
.len = sizeof(struct sockaddr_storage) },
[IWPM_NLA_RMANAGE_MAPPED_LOC_ADDR] = {
.len = sizeof(struct sockaddr_storage) },
[IWPM_NLA_RMANAGE_MAPPING_ERR] = { .type = NLA_U16 }
};
/*
* iwpm_add_mapping_cb - Process a port mapper response to
* iwpm_add_mapping()
/**
* iwpm_add_mapping_cb - Process the port mapper response to
* iwpm_add_mapping request
* @skb:
* @cb: Contains the received message (payload and netlink header)
*/
int iwpm_add_mapping_cb(struct sk_buff *skb, struct netlink_callback *cb)
{
......@@ -430,7 +491,7 @@ int iwpm_add_mapping_cb(struct sk_buff *skb, struct netlink_callback *cb)
atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq);
msg_seq = nla_get_u32(nltb[IWPM_NLA_MANAGE_MAPPING_SEQ]);
msg_seq = nla_get_u32(nltb[IWPM_NLA_RMANAGE_MAPPING_SEQ]);
nlmsg_request = iwpm_find_nlmsg_request(msg_seq);
if (!nlmsg_request) {
pr_info("%s: Could not find a matching request (seq = %u)\n",
......@@ -439,9 +500,9 @@ int iwpm_add_mapping_cb(struct sk_buff *skb, struct netlink_callback *cb)
}
pm_msg = nlmsg_request->req_buffer;
local_sockaddr = (struct sockaddr_storage *)
nla_data(nltb[IWPM_NLA_MANAGE_ADDR]);
nla_data(nltb[IWPM_NLA_RMANAGE_ADDR]);
mapped_sockaddr = (struct sockaddr_storage *)
nla_data(nltb[IWPM_NLA_MANAGE_MAPPED_LOC_ADDR]);
nla_data(nltb[IWPM_NLA_RMANAGE_MAPPED_LOC_ADDR]);
if (iwpm_compare_sockaddr(local_sockaddr, &pm_msg->loc_addr)) {
nlmsg_request->err_code = IWPM_USER_LIB_INFO_ERR;
......@@ -472,17 +533,23 @@ int iwpm_add_mapping_cb(struct sk_buff *skb, struct netlink_callback *cb)
/* netlink attribute policy for the response to add and query mapping request
* and response with remote address info */
static const struct nla_policy resp_query_policy[IWPM_NLA_RQUERY_MAPPING_MAX] = {
[IWPM_NLA_QUERY_MAPPING_SEQ] = { .type = NLA_U32 },
[IWPM_NLA_QUERY_LOCAL_ADDR] = { .len = sizeof(struct sockaddr_storage) },
[IWPM_NLA_QUERY_REMOTE_ADDR] = { .len = sizeof(struct sockaddr_storage) },
[IWPM_NLA_RQUERY_MAPPED_LOC_ADDR] = { .len = sizeof(struct sockaddr_storage) },
[IWPM_NLA_RQUERY_MAPPED_REM_ADDR] = { .len = sizeof(struct sockaddr_storage) },
[IWPM_NLA_RQUERY_MAPPING_SEQ] = { .type = NLA_U32 },
[IWPM_NLA_RQUERY_LOCAL_ADDR] = {
.len = sizeof(struct sockaddr_storage) },
[IWPM_NLA_RQUERY_REMOTE_ADDR] = {
.len = sizeof(struct sockaddr_storage) },
[IWPM_NLA_RQUERY_MAPPED_LOC_ADDR] = {
.len = sizeof(struct sockaddr_storage) },
[IWPM_NLA_RQUERY_MAPPED_REM_ADDR] = {
.len = sizeof(struct sockaddr_storage) },
[IWPM_NLA_RQUERY_MAPPING_ERR] = { .type = NLA_U16 }
};
/*
* iwpm_add_and_query_mapping_cb - Process a port mapper response to
* iwpm_add_and_query_mapping()
/**
* iwpm_add_and_query_mapping_cb - Process the port mapper response to
* iwpm_add_and_query_mapping request
* @skb:
* @cb: Contains the received message (payload and netlink header)
*/
int iwpm_add_and_query_mapping_cb(struct sk_buff *skb,
struct netlink_callback *cb)
......@@ -502,7 +569,7 @@ int iwpm_add_and_query_mapping_cb(struct sk_buff *skb,
return -EINVAL;
atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq);
msg_seq = nla_get_u32(nltb[IWPM_NLA_QUERY_MAPPING_SEQ]);
msg_seq = nla_get_u32(nltb[IWPM_NLA_RQUERY_MAPPING_SEQ]);
nlmsg_request = iwpm_find_nlmsg_request(msg_seq);
if (!nlmsg_request) {
pr_info("%s: Could not find a matching request (seq = %u)\n",
......@@ -511,9 +578,9 @@ int iwpm_add_and_query_mapping_cb(struct sk_buff *skb,
}
pm_msg = nlmsg_request->req_buffer;
local_sockaddr = (struct sockaddr_storage *)
nla_data(nltb[IWPM_NLA_QUERY_LOCAL_ADDR]);
nla_data(nltb[IWPM_NLA_RQUERY_LOCAL_ADDR]);
remote_sockaddr = (struct sockaddr_storage *)
nla_data(nltb[IWPM_NLA_QUERY_REMOTE_ADDR]);
nla_data(nltb[IWPM_NLA_RQUERY_REMOTE_ADDR]);
mapped_loc_sockaddr = (struct sockaddr_storage *)
nla_data(nltb[IWPM_NLA_RQUERY_MAPPED_LOC_ADDR]);
mapped_rem_sockaddr = (struct sockaddr_storage *)
......@@ -560,9 +627,13 @@ int iwpm_add_and_query_mapping_cb(struct sk_buff *skb,
return 0;
}
/*
* iwpm_remote_info_cb - Process a port mapper message, containing
* the remote connecting peer address info
/**
* iwpm_remote_info_cb - Process remote connecting peer address info, which
* the port mapper has received from the connecting peer
* @skb:
* @cb: Contains the received message (payload and netlink header)
*
* Stores the IPv4/IPv6 address info in a hash table
*/
int iwpm_remote_info_cb(struct sk_buff *skb, struct netlink_callback *cb)
{
......@@ -588,9 +659,9 @@ int iwpm_remote_info_cb(struct sk_buff *skb, struct netlink_callback *cb)
atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq);
local_sockaddr = (struct sockaddr_storage *)
nla_data(nltb[IWPM_NLA_QUERY_LOCAL_ADDR]);
nla_data(nltb[IWPM_NLA_RQUERY_LOCAL_ADDR]);
remote_sockaddr = (struct sockaddr_storage *)
nla_data(nltb[IWPM_NLA_QUERY_REMOTE_ADDR]);
nla_data(nltb[IWPM_NLA_RQUERY_REMOTE_ADDR]);
mapped_loc_sockaddr = (struct sockaddr_storage *)
nla_data(nltb[IWPM_NLA_RQUERY_MAPPED_LOC_ADDR]);
mapped_rem_sockaddr = (struct sockaddr_storage *)
......@@ -635,8 +706,14 @@ static const struct nla_policy resp_mapinfo_policy[IWPM_NLA_MAPINFO_REQ_MAX] = {
[IWPM_NLA_MAPINFO_ULIB_VER] = { .type = NLA_U16 }
};
/*
* iwpm_mapping_info_cb - Process a port mapper request for mapping info
/**
* iwpm_mapping_info_cb - Process a notification that the userspace
* port mapper daemon is started
* @skb:
* @cb: Contains the received message (payload and netlink header)
*
* Using the received port mapper pid, send all the local mapping
* info records to the userspace port mapper
*/
int iwpm_mapping_info_cb(struct sk_buff *skb, struct netlink_callback *cb)
{
......@@ -655,7 +732,7 @@ int iwpm_mapping_info_cb(struct sk_buff *skb, struct netlink_callback *cb)
iwpm_name = (char *)nla_data(nltb[IWPM_NLA_MAPINFO_ULIB_NAME]);
iwpm_version = nla_get_u16(nltb[IWPM_NLA_MAPINFO_ULIB_VER]);
if (strcmp(iwpm_ulib_name, iwpm_name) ||
iwpm_version != iwpm_ulib_version) {
iwpm_version < IWPM_UABI_VERSION_MIN) {
pr_info("%s: Invalid port mapper name = %s version = %d\n",
__func__, iwpm_name, iwpm_version);
return ret;
......@@ -669,6 +746,11 @@ int iwpm_mapping_info_cb(struct sk_buff *skb, struct netlink_callback *cb)
iwpm_set_registration(nl_client, IWPM_REG_INCOMPL);
atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq);
iwpm_user_pid = cb->nlh->nlmsg_pid;
if (iwpm_ulib_version < IWPM_UABI_VERSION)
pr_warn_once("%s: Down level iwpmd/pid %u. Continuing...",
__func__, iwpm_user_pid);
if (!iwpm_mapinfo_available())
return 0;
pr_debug("%s: iWarp Port Mapper (pid = %d) is available!\n",
......@@ -684,9 +766,11 @@ static const struct nla_policy ack_mapinfo_policy[IWPM_NLA_MAPINFO_NUM_MAX] = {
[IWPM_NLA_MAPINFO_ACK_NUM] = { .type = NLA_U32 }
};
/*
* iwpm_ack_mapping_info_cb - Process a port mapper ack for
* the provided mapping info records
/**
* iwpm_ack_mapping_info_cb - Process the port mapper ack for
* the provided local mapping info records
* @skb:
* @cb: Contains the received message (payload and netlink header)
*/
int iwpm_ack_mapping_info_cb(struct sk_buff *skb, struct netlink_callback *cb)
{
......@@ -712,8 +796,11 @@ static const struct nla_policy map_error_policy[IWPM_NLA_ERR_MAX] = {
[IWPM_NLA_ERR_CODE] = { .type = NLA_U16 },
};
/*
* iwpm_mapping_error_cb - Process a port mapper error message
/**
* iwpm_mapping_error_cb - Process port mapper notification for error
*
* @skb:
* @cb: Contains the received message (payload and netlink header)
*/
int iwpm_mapping_error_cb(struct sk_buff *skb, struct netlink_callback *cb)
{
......@@ -748,3 +835,46 @@ int iwpm_mapping_error_cb(struct sk_buff *skb, struct netlink_callback *cb)
up(&nlmsg_request->sem);
return 0;
}
/* netlink attribute policy for the received hello request */
static const struct nla_policy hello_policy[IWPM_NLA_HELLO_MAX] = {
[IWPM_NLA_HELLO_ABI_VERSION] = { .type = NLA_U16 }
};
/**
* iwpm_hello_cb - Process a hello message from iwpmd
*
* @skb:
* @cb: Contains the received message (payload and netlink header)
*
* Using the received port mapper pid, send the kernel's abi_version
* after adjusting it to support the iwpmd version.
*/
int iwpm_hello_cb(struct sk_buff *skb, struct netlink_callback *cb)
{
struct nlattr *nltb[IWPM_NLA_HELLO_MAX];
const char *msg_type = "Hello request";
u8 nl_client;
u16 abi_version;
int ret = -EINVAL;
if (iwpm_parse_nlmsg(cb, IWPM_NLA_HELLO_MAX, hello_policy, nltb,
msg_type)) {
pr_info("%s: Unable to parse nlmsg\n", __func__);
return ret;
}
abi_version = nla_get_u16(nltb[IWPM_NLA_HELLO_ABI_VERSION]);
nl_client = RDMA_NL_GET_CLIENT(cb->nlh->nlmsg_type);
if (!iwpm_valid_client(nl_client)) {
pr_info("%s: Invalid port mapper client = %d\n",
__func__, nl_client);
return ret;
}
iwpm_set_registration(nl_client, IWPM_REG_INCOMPL);
atomic_set(&echo_nlmsg_seq, cb->nlh->nlmsg_seq);
iwpm_ulib_version = min_t(u16, IWPM_UABI_VERSION, abi_version);
pr_debug("Using ABI version %u\n", iwpm_ulib_version);
iwpm_user_pid = cb->nlh->nlmsg_pid;
ret = iwpm_send_hello(nl_client, iwpm_user_pid, iwpm_ulib_version);
return ret;
}
drivers/infiniband/core/iwpm_util.c
浏览文件 @ a50243b1
......@@ -51,6 +51,12 @@ static DEFINE_SPINLOCK(iwpm_reminfo_lock);
static DEFINE_MUTEX(iwpm_admin_lock);
static struct iwpm_admin_data iwpm_admin;
/**
* iwpm_init - Allocate resources for the iwarp port mapper
* @nl_client: The index of the netlink client
*
* Should be called when network interface goes up.
*/
int iwpm_init(u8 nl_client)
{
int ret = 0;
......@@ -87,6 +93,12 @@ int iwpm_init(u8 nl_client)
static void free_hash_bucket(void);
static void free_reminfo_bucket(void);
/**
* iwpm_exit - Deallocate resources for the iwarp port mapper
* @nl_client: The index of the netlink client
*
* Should be called when network interface goes down.
*/
int iwpm_exit(u8 nl_client)
{
......@@ -112,9 +124,17 @@ int iwpm_exit(u8 nl_client)
static struct hlist_head *get_mapinfo_hash_bucket(struct sockaddr_storage *,
struct sockaddr_storage *);
/**
* iwpm_create_mapinfo - Store local and mapped IPv4/IPv6 address
* info in a hash table
* @local_addr: Local ip/tcp address
* @mapped_addr: Mapped local ip/tcp address
* @nl_client: The index of the netlink client
* @map_flags: IWPM mapping flags
*/
int iwpm_create_mapinfo(struct sockaddr_storage *local_sockaddr,
struct sockaddr_storage *mapped_sockaddr,
u8 nl_client)
u8 nl_client, u32 map_flags)
{
struct hlist_head *hash_bucket_head = NULL;
struct iwpm_mapping_info *map_info;
......@@ -132,6 +152,7 @@ int iwpm_create_mapinfo(struct sockaddr_storage *local_sockaddr,
memcpy(&map_info->mapped_sockaddr, mapped_sockaddr,
sizeof(struct sockaddr_storage));
map_info->nl_client = nl_client;
map_info->map_flags = map_flags;
spin_lock_irqsave(&iwpm_mapinfo_lock, flags);
if (iwpm_hash_bucket) {
......@@ -150,6 +171,15 @@ int iwpm_create_mapinfo(struct sockaddr_storage *local_sockaddr,
return ret;
}
/**
* iwpm_remove_mapinfo - Remove local and mapped IPv4/IPv6 address
* info from the hash table
* @local_addr: Local ip/tcp address
* @mapped_local_addr: Mapped local ip/tcp address
*
* Returns err code if mapping info is not found in the hash table,
* otherwise returns 0
*/
int iwpm_remove_mapinfo(struct sockaddr_storage *local_sockaddr,
struct sockaddr_storage *mapped_local_addr)
{
......@@ -250,6 +280,17 @@ void iwpm_add_remote_info(struct iwpm_remote_info *rem_info)
spin_unlock_irqrestore(&iwpm_reminfo_lock, flags);
}
/**
* iwpm_get_remote_info - Get the remote connecting peer address info
*
* @mapped_loc_addr: Mapped local address of the listening peer
* @mapped_rem_addr: Mapped remote address of the connecting peer
* @remote_addr: To store the remote address of the connecting peer
* @nl_client: The index of the netlink client
*
* The remote address info is retrieved and provided to the client in
* the remote_addr. After that it is removed from the hash table
*/
int iwpm_get_remote_info(struct sockaddr_storage *mapped_loc_addr,
struct sockaddr_storage *mapped_rem_addr,
struct sockaddr_storage *remote_addr,
......@@ -686,6 +727,14 @@ int iwpm_send_mapinfo(u8 nl_client, int iwpm_pid)
if (ret)
goto send_mapping_info_unlock;
if (iwpm_ulib_version > IWPM_UABI_VERSION_MIN) {
ret = ibnl_put_attr(skb, nlh, sizeof(u32),
&map_info->map_flags,
IWPM_NLA_MAPINFO_FLAGS);
if (ret)
goto send_mapping_info_unlock;
}
nlmsg_end(skb, nlh);
iwpm_print_sockaddr(&map_info->local_sockaddr,
......@@ -754,3 +803,38 @@ int iwpm_mapinfo_available(void)
spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags);
return full_bucket;
}
int iwpm_send_hello(u8 nl_client, int iwpm_pid, u16 abi_version)
{
struct sk_buff *skb = NULL;
struct nlmsghdr *nlh;
const char *err_str = "";
int ret = -EINVAL;
skb = iwpm_create_nlmsg(RDMA_NL_IWPM_HELLO, &nlh, nl_client);
if (!skb) {
err_str = "Unable to create a nlmsg";
goto hello_num_error;
}
nlh->nlmsg_seq = iwpm_get_nlmsg_seq();
err_str = "Unable to put attribute of abi_version into nlmsg";
ret = ibnl_put_attr(skb, nlh, sizeof(u16), &abi_version,
IWPM_NLA_HELLO_ABI_VERSION);
if (ret)
goto hello_num_error;
nlmsg_end(skb, nlh);
ret = rdma_nl_unicast(skb, iwpm_pid);
if (ret) {
skb = NULL;
err_str = "Unable to send a nlmsg";
goto hello_num_error;
}
pr_debug("%s: Sent hello abi_version = %u\n", __func__, abi_version);
return 0;
hello_num_error:
pr_info("%s: %s\n", __func__, err_str);
if (skb)
dev_kfree_skb(skb);
return ret;
}
drivers/infiniband/core/iwpm_util.h
浏览文件 @ a50243b1
......@@ -78,6 +78,7 @@ struct iwpm_mapping_info {
struct sockaddr_storage local_sockaddr;
struct sockaddr_storage mapped_sockaddr;
u8 nl_client;
u32 map_flags;
};
struct iwpm_remote_info {
......@@ -266,4 +267,15 @@ int iwpm_parse_nlmsg(struct netlink_callback *cb, int policy_max,
* @msg: Message to print
*/
void iwpm_print_sockaddr(struct sockaddr_storage *sockaddr, char *msg);
/**
* iwpm_send_hello - Send hello response to iwpmd
*
* @nl_client: The index of the netlink client
* @abi_version: The kernel's abi_version
*
* Returns 0 on success or a negative error code
*/
int iwpm_send_hello(u8 nl_client, int iwpm_pid, u16 abi_version);
extern u16 iwpm_ulib_version;
#endif
drivers/infiniband/core/mad.c
浏览文件 @ a50243b1
......@@ -3326,9 +3326,9 @@ static void ib_mad_init_device(struct ib_device *device)
static void ib_mad_remove_device(struct ib_device *device, void *client_data)
{
int i;
unsigned int i;
for (i = rdma_start_port(device); i <= rdma_end_port(device); i++) {
rdma_for_each_port (device, i) {
if (!rdma_cap_ib_mad(device, i))
continue;
......
drivers/infiniband/core/netlink.c
浏览文件 @ a50243b1
......@@ -56,7 +56,6 @@ EXPORT_SYMBOL(rdma_nl_chk_listeners);
static bool is_nl_msg_valid(unsigned int type, unsigned int op)
{
static const unsigned int max_num_ops[RDMA_NL_NUM_CLIENTS] = {
[RDMA_NL_RDMA_CM] = RDMA_NL_RDMA_CM_NUM_OPS,
[RDMA_NL_IWCM] = RDMA_NL_IWPM_NUM_OPS,
[RDMA_NL_LS] = RDMA_NL_LS_NUM_OPS,
[RDMA_NL_NLDEV] = RDMA_NLDEV_NUM_OPS,
......@@ -181,8 +180,7 @@ static int rdma_nl_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
return -EINVAL;
}
/* FIXME: Convert IWCM to properly handle doit callbacks */
if ((nlh->nlmsg_flags & NLM_F_DUMP) || index == RDMA_NL_RDMA_CM ||
index == RDMA_NL_IWCM) {
if ((nlh->nlmsg_flags & NLM_F_DUMP) || index == RDMA_NL_IWCM) {
struct netlink_dump_control c = {
.dump = cb_table[op].dump,
};
......
drivers/infiniband/core/nldev.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/core/rdma_core.c
浏览文件 @ a50243b1
......@@ -438,6 +438,38 @@ struct ib_uobject *rdma_lookup_get_uobject(const struct uverbs_api_object *obj,
uverbs_uobject_put(uobj);
return ERR_PTR(ret);
}
struct ib_uobject *_uobj_get_read(enum uverbs_default_objects type,
u32 object_id,
struct uverbs_attr_bundle *attrs)
{
struct ib_uobject *uobj;
uobj = rdma_lookup_get_uobject(uobj_get_type(attrs, type), attrs->ufile,
object_id, UVERBS_LOOKUP_READ);
if (IS_ERR(uobj))
return uobj;
attrs->context = uobj->context;
return uobj;
}
struct ib_uobject *_uobj_get_write(enum uverbs_default_objects type,
u32 object_id,
struct uverbs_attr_bundle *attrs)
{
struct ib_uobject *uobj;
uobj = rdma_lookup_get_uobject(uobj_get_type(attrs, type), attrs->ufile,
object_id, UVERBS_LOOKUP_WRITE);
if (IS_ERR(uobj))
return uobj;
attrs->context = uobj->context;
return uobj;
}
static struct ib_uobject *
alloc_begin_idr_uobject(const struct uverbs_api_object *obj,
......@@ -801,6 +833,7 @@ void uverbs_close_fd(struct file *f)
/* Pairs with filp->private_data in alloc_begin_fd_uobject */
uverbs_uobject_put(uobj);
}
EXPORT_SYMBOL(uverbs_close_fd);
/*
* Drop the ucontext off the ufile and completely disconnect it from the
......@@ -811,7 +844,6 @@ static void ufile_destroy_ucontext(struct ib_uverbs_file *ufile,
{
struct ib_ucontext *ucontext = ufile->ucontext;
struct ib_device *ib_dev = ucontext->device;
int ret;
/*
* If we are closing the FD then the user mmap VMAs must have
......@@ -829,12 +861,8 @@ static void ufile_destroy_ucontext(struct ib_uverbs_file *ufile,
rdma_restrack_del(&ucontext->res);
/*
* FIXME: Drivers are not permitted to fail dealloc_ucontext, remove
* the error return.
*/
ret = ib_dev->ops.dealloc_ucontext(ucontext);
WARN_ON(ret);
ib_dev->ops.dealloc_ucontext(ucontext);
kfree(ucontext);
ufile->ucontext = NULL;
}
......
drivers/infiniband/core/restrack.c
浏览文件 @ a50243b1
......@@ -11,17 +11,51 @@
#include <linux/pid_namespace.h>
#include "cma_priv.h"
#include "restrack.h"
static int fill_res_noop(struct sk_buff *msg,
struct rdma_restrack_entry *entry)
static int rt_xa_alloc_cyclic(struct xarray *xa, u32 *id, void *entry,
u32 *next)
{
return 0;
int err;
*id = *next;
if (*next == U32_MAX)
*id = 0;
xa_lock(xa);
err = __xa_alloc(xa, id, U32_MAX, entry, GFP_KERNEL);
if (err && *next != U32_MAX) {
*id = 0;
err = __xa_alloc(xa, id, *next, entry, GFP_KERNEL);
}
if (!err)
*next = *id + 1;
xa_unlock(xa);
return err;
}
void rdma_restrack_init(struct rdma_restrack_root *res)
/**
* rdma_restrack_init() - initialize and allocate resource tracking
* @dev: IB device
*
* Return: 0 on success
*/
int rdma_restrack_init(struct ib_device *dev)
{
init_rwsem(&res->rwsem);
res->fill_res_entry = fill_res_noop;
struct rdma_restrack_root *rt;
int i;
dev->res = kcalloc(RDMA_RESTRACK_MAX, sizeof(*rt), GFP_KERNEL);
if (!dev->res)
return -ENOMEM;
rt = dev->res;
for (i = 0; i < RDMA_RESTRACK_MAX; i++)
xa_init_flags(&rt[i].xa, XA_FLAGS_ALLOC);
return 0;
}
static const char *type2str(enum rdma_restrack_type type)
......@@ -38,55 +72,79 @@ static const char *type2str(enum rdma_restrack_type type)
return names[type];
};
void rdma_restrack_clean(struct rdma_restrack_root *res)
/**
* rdma_restrack_clean() - clean resource tracking
* @dev: IB device
*/
void rdma_restrack_clean(struct ib_device *dev)
{
struct rdma_restrack_root *rt = dev->res;
struct rdma_restrack_entry *e;
char buf[TASK_COMM_LEN];
struct ib_device *dev;
bool found = false;
const char *owner;
int bkt;
if (hash_empty(res->hash))
return;
dev = container_of(res, struct ib_device, res);
pr_err("restrack: %s", CUT_HERE);
dev_err(&dev->dev, "BUG: RESTRACK detected leak of resources\n");
hash_for_each(res->hash, bkt, e, node) {
if (rdma_is_kernel_res(e)) {
owner = e->kern_name;
} else {
/*
* There is no need to call get_task_struct here,
* because we can be here only if there are more
* get_task_struct() call than put_task_struct().
*/
get_task_comm(buf, e->task);
owner = buf;
int i;
for (i = 0 ; i < RDMA_RESTRACK_MAX; i++) {
struct xarray *xa = &dev->res[i].xa;
if (!xa_empty(xa)) {
unsigned long index;
if (!found) {
pr_err("restrack: %s", CUT_HERE);
dev_err(&dev->dev, "BUG: RESTRACK detected leak of resources\n");
}
xa_for_each(xa, index, e) {
if (rdma_is_kernel_res(e)) {
owner = e->kern_name;
} else {
/*
* There is no need to call get_task_struct here,
* because we can be here only if there are more
* get_task_struct() call than put_task_struct().
*/
get_task_comm(buf, e->task);
owner = buf;
}
pr_err("restrack: %s %s object allocated by %s is not freed\n",
rdma_is_kernel_res(e) ? "Kernel" :
"User",
type2str(e->type), owner);
}
found = true;
}
pr_err("restrack: %s %s object allocated by %s is not freed\n",
rdma_is_kernel_res(e) ? "Kernel" : "User",
type2str(e->type), owner);
xa_destroy(xa);
}
pr_err("restrack: %s", CUT_HERE);
if (found)
pr_err("restrack: %s", CUT_HERE);
kfree(rt);
}
int rdma_restrack_count(struct rdma_restrack_root *res,
enum rdma_restrack_type type,
/**
* rdma_restrack_count() - the current usage of specific object
* @dev: IB device
* @type: actual type of object to operate
* @ns: PID namespace
*/
int rdma_restrack_count(struct ib_device *dev, enum rdma_restrack_type type,
struct pid_namespace *ns)
{
struct rdma_restrack_root *rt = &dev->res[type];
struct rdma_restrack_entry *e;
XA_STATE(xas, &rt->xa, 0);
u32 cnt = 0;
down_read(&res->rwsem);
hash_for_each_possible(res->hash, e, node, type) {
xa_lock(&rt->xa);
xas_for_each(&xas, e, U32_MAX) {
if (ns == &init_pid_ns ||
(!rdma_is_kernel_res(e) &&
ns == task_active_pid_ns(e->task)))
cnt++;
}
up_read(&res->rwsem);
xa_unlock(&rt->xa);
return cnt;
}
EXPORT_SYMBOL(rdma_restrack_count);
......@@ -157,28 +215,28 @@ EXPORT_SYMBOL(rdma_restrack_set_task);
static void rdma_restrack_add(struct rdma_restrack_entry *res)
{
struct ib_device *dev = res_to_dev(res);
struct rdma_restrack_root *rt;
int ret;
if (!dev)
return;
if (res->type != RDMA_RESTRACK_CM_ID || rdma_is_kernel_res(res))
res->task = NULL;
if (!rdma_is_kernel_res(res)) {
if (!res->task)
rdma_restrack_set_task(res, NULL);
res->kern_name = NULL;
} else {
set_kern_name(res);
}
rt = &dev->res[res->type];
kref_init(&res->kref);
init_completion(&res->comp);
res->valid = true;
if (res->type != RDMA_RESTRACK_QP)
ret = rt_xa_alloc_cyclic(&rt->xa, &res->id, res, &rt->next_id);
else {
/* Special case to ensure that LQPN points to right QP */
struct ib_qp *qp = container_of(res, struct ib_qp, res);
ret = xa_insert(&rt->xa, qp->qp_num, res, GFP_KERNEL);
res->id = ret ? 0 : qp->qp_num;
}
down_write(&dev->res.rwsem);
hash_add(dev->res.hash, &res->node, res->type);
up_write(&dev->res.rwsem);
if (!ret)
res->valid = true;
}
/**
......@@ -187,6 +245,8 @@ static void rdma_restrack_add(struct rdma_restrack_entry *res)
*/
void rdma_restrack_kadd(struct rdma_restrack_entry *res)
{
res->task = NULL;
set_kern_name(res);
res->user = false;
rdma_restrack_add(res);
}
......@@ -198,6 +258,13 @@ EXPORT_SYMBOL(rdma_restrack_kadd);
*/
void rdma_restrack_uadd(struct rdma_restrack_entry *res)
{
if (res->type != RDMA_RESTRACK_CM_ID)
res->task = NULL;
if (!res->task)
rdma_restrack_set_task(res, NULL);
res->kern_name = NULL;
res->user = true;
rdma_restrack_add(res);
}
......@@ -209,6 +276,31 @@ int __must_check rdma_restrack_get(struct rdma_restrack_entry *res)
}
EXPORT_SYMBOL(rdma_restrack_get);
/**
* rdma_restrack_get_byid() - translate from ID to restrack object
* @dev: IB device
* @type: resource track type
* @id: ID to take a look
*
* Return: Pointer to restrack entry or -ENOENT in case of error.
*/
struct rdma_restrack_entry *
rdma_restrack_get_byid(struct ib_device *dev,
enum rdma_restrack_type type, u32 id)
{
struct rdma_restrack_root *rt = &dev->res[type];
struct rdma_restrack_entry *res;
xa_lock(&rt->xa);
res = xa_load(&rt->xa, id);
if (!res || !rdma_restrack_get(res))
res = ERR_PTR(-ENOENT);
xa_unlock(&rt->xa);
return res;
}
EXPORT_SYMBOL(rdma_restrack_get_byid);
static void restrack_release(struct kref *kref)
{
struct rdma_restrack_entry *res;
......@@ -225,23 +317,25 @@ EXPORT_SYMBOL(rdma_restrack_put);
void rdma_restrack_del(struct rdma_restrack_entry *res)
{
struct rdma_restrack_entry *old;
struct rdma_restrack_root *rt;
struct ib_device *dev;
if (!res->valid)
goto out;
dev = res_to_dev(res);
if (!dev)
if (WARN_ON(!dev))
return;
rdma_restrack_put(res);
rt = &dev->res[res->type];
wait_for_completion(&res->comp);
down_write(&dev->res.rwsem);
hash_del(&res->node);
old = xa_erase(&rt->xa, res->id);
WARN_ON(old != res);
res->valid = false;
up_write(&dev->res.rwsem);
rdma_restrack_put(res);
wait_for_completion(&res->comp);
out:
if (res->task) {
......
drivers/infiniband/core/restrack.h 0 → 100644
浏览文件 @ a50243b1
/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */
/*
* Copyright (c) 2017-2019 Mellanox Technologies. All rights reserved.
*/
#ifndef _RDMA_CORE_RESTRACK_H_
#define _RDMA_CORE_RESTRACK_H_
#include <linux/mutex.h>
/**
* struct rdma_restrack_root - main resource tracking management
* entity, per-device
*/
struct rdma_restrack_root {
/**
* @xa: Array of XArray structure to hold restrack entries.
*/
struct xarray xa;
/**
* @next_id: Next ID to support cyclic allocation
*/
u32 next_id;
};
int rdma_restrack_init(struct ib_device *dev);
void rdma_restrack_clean(struct ib_device *dev);
#endif /* _RDMA_CORE_RESTRACK_H_ */
drivers/infiniband/core/rw.c
浏览文件 @ a50243b1
......@@ -179,7 +179,6 @@ static int rdma_rw_init_map_wrs(struct rdma_rw_ctx *ctx, struct ib_qp *qp,
struct scatterlist *sg, u32 sg_cnt, u32 offset,
u64 remote_addr, u32 rkey, enum dma_data_direction dir)
{
struct ib_device *dev = qp->pd->device;
u32 max_sge = dir == DMA_TO_DEVICE ? qp->max_write_sge :
qp->max_read_sge;
struct ib_sge *sge;
......@@ -209,8 +208,8 @@ static int rdma_rw_init_map_wrs(struct rdma_rw_ctx *ctx, struct ib_qp *qp,
rdma_wr->wr.sg_list = sge;
for (j = 0; j < nr_sge; j++, sg = sg_next(sg)) {
sge->addr = ib_sg_dma_address(dev, sg) + offset;
sge->length = ib_sg_dma_len(dev, sg) - offset;
sge->addr = sg_dma_address(sg) + offset;
sge->length = sg_dma_len(sg) - offset;
sge->lkey = qp->pd->local_dma_lkey;
total_len += sge->length;
......@@ -236,14 +235,13 @@ static int rdma_rw_init_single_wr(struct rdma_rw_ctx *ctx, struct ib_qp *qp,
struct scatterlist *sg, u32 offset, u64 remote_addr, u32 rkey,
enum dma_data_direction dir)
{
struct ib_device *dev = qp->pd->device;
struct ib_rdma_wr *rdma_wr = &ctx->single.wr;
ctx->nr_ops = 1;
ctx->single.sge.lkey = qp->pd->local_dma_lkey;
ctx->single.sge.addr = ib_sg_dma_address(dev, sg) + offset;
ctx->single.sge.length = ib_sg_dma_len(dev, sg) - offset;
ctx->single.sge.addr = sg_dma_address(sg) + offset;
ctx->single.sge.length = sg_dma_len(sg) - offset;
memset(rdma_wr, 0, sizeof(*rdma_wr));
if (dir == DMA_TO_DEVICE)
......@@ -294,7 +292,7 @@ int rdma_rw_ctx_init(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u8 port_num,
* Skip to the S/G entry that sg_offset falls into:
*/
for (;;) {
u32 len = ib_sg_dma_len(dev, sg);
u32 len = sg_dma_len(sg);
if (sg_offset < len)
break;
......
drivers/infiniband/core/sa_query.c
浏览文件 @ a50243b1
......@@ -2342,9 +2342,7 @@ static void ib_sa_add_one(struct ib_device *device)
s = rdma_start_port(device);
e = rdma_end_port(device);
sa_dev = kzalloc(sizeof *sa_dev +
(e - s + 1) * sizeof (struct ib_sa_port),
GFP_KERNEL);
sa_dev = kzalloc(struct_size(sa_dev, port, e - s + 1), GFP_KERNEL);
if (!sa_dev)
return;
......
drivers/infiniband/core/security.c
浏览文件 @ a50243b1
......@@ -39,22 +39,25 @@
#include "core_priv.h"
#include "mad_priv.h"
static LIST_HEAD(mad_agent_list);
/* Lock to protect mad_agent_list */
static DEFINE_SPINLOCK(mad_agent_list_lock);
static struct pkey_index_qp_list *get_pkey_idx_qp_list(struct ib_port_pkey *pp)
{
struct pkey_index_qp_list *pkey = NULL;
struct pkey_index_qp_list *tmp_pkey;
struct ib_device *dev = pp->sec->dev;
spin_lock(&dev->port_pkey_list[pp->port_num].list_lock);
list_for_each_entry(tmp_pkey,
&dev->port_pkey_list[pp->port_num].pkey_list,
pkey_index_list) {
spin_lock(&dev->port_data[pp->port_num].pkey_list_lock);
list_for_each_entry (tmp_pkey, &dev->port_data[pp->port_num].pkey_list,
pkey_index_list) {
if (tmp_pkey->pkey_index == pp->pkey_index) {
pkey = tmp_pkey;
break;
}
}
spin_unlock(&dev->port_pkey_list[pp->port_num].list_lock);
spin_unlock(&dev->port_data[pp->port_num].pkey_list_lock);
return pkey;
}
......@@ -259,12 +262,12 @@ static int port_pkey_list_insert(struct ib_port_pkey *pp)
if (!pkey)
return -ENOMEM;
spin_lock(&dev->port_pkey_list[port_num].list_lock);
spin_lock(&dev->port_data[port_num].pkey_list_lock);
/* Check for the PKey again. A racing process may
* have created it.
*/
list_for_each_entry(tmp_pkey,
&dev->port_pkey_list[port_num].pkey_list,
&dev->port_data[port_num].pkey_list,
pkey_index_list) {
if (tmp_pkey->pkey_index == pp->pkey_index) {
kfree(pkey);
......@@ -279,9 +282,9 @@ static int port_pkey_list_insert(struct ib_port_pkey *pp)
spin_lock_init(&pkey->qp_list_lock);
INIT_LIST_HEAD(&pkey->qp_list);
list_add(&pkey->pkey_index_list,
&dev->port_pkey_list[port_num].pkey_list);
&dev->port_data[port_num].pkey_list);
}
spin_unlock(&dev->port_pkey_list[port_num].list_lock);
spin_unlock(&dev->port_data[port_num].pkey_list_lock);
}
spin_lock(&pkey->qp_list_lock);
......@@ -418,12 +421,15 @@ void ib_close_shared_qp_security(struct ib_qp_security *sec)
int ib_create_qp_security(struct ib_qp *qp, struct ib_device *dev)
{
u8 i = rdma_start_port(dev);
unsigned int i;
bool is_ib = false;
int ret;
while (i <= rdma_end_port(dev) && !is_ib)
rdma_for_each_port (dev, i) {
is_ib = rdma_protocol_ib(dev, i++);
if (is_ib)
break;
}
/* If this isn't an IB device don't create the security context */
if (!is_ib)
......@@ -544,9 +550,8 @@ void ib_security_cache_change(struct ib_device *device,
{
struct pkey_index_qp_list *pkey;
list_for_each_entry(pkey,
&device->port_pkey_list[port_num].pkey_list,
pkey_index_list) {
list_for_each_entry (pkey, &device->port_data[port_num].pkey_list,
pkey_index_list) {
check_pkey_qps(pkey,
device,
port_num,
......@@ -554,21 +559,19 @@ void ib_security_cache_change(struct ib_device *device,
}
}
void ib_security_destroy_port_pkey_list(struct ib_device *device)
void ib_security_release_port_pkey_list(struct ib_device *device)
{
struct pkey_index_qp_list *pkey, *tmp_pkey;
int i;
unsigned int i;
for (i = rdma_start_port(device); i <= rdma_end_port(device); i++) {
spin_lock(&device->port_pkey_list[i].list_lock);
rdma_for_each_port (device, i) {
list_for_each_entry_safe(pkey,
tmp_pkey,
&device->port_pkey_list[i].pkey_list,
&device->port_data[i].pkey_list,
pkey_index_list) {
list_del(&pkey->pkey_index_list);
kfree(pkey);
}
spin_unlock(&device->port_pkey_list[i].list_lock);
}
}
......@@ -676,19 +679,18 @@ static int ib_security_pkey_access(struct ib_device *dev,
return security_ib_pkey_access(sec, subnet_prefix, pkey);
}
static int ib_mad_agent_security_change(struct notifier_block *nb,
unsigned long event,
void *data)
void ib_mad_agent_security_change(void)
{
struct ib_mad_agent *ag = container_of(nb, struct ib_mad_agent, lsm_nb);
if (event != LSM_POLICY_CHANGE)
return NOTIFY_DONE;
ag->smp_allowed = !security_ib_endport_manage_subnet(
ag->security, dev_name(&ag->device->dev), ag->port_num);
return NOTIFY_OK;
struct ib_mad_agent *ag;
spin_lock(&mad_agent_list_lock);
list_for_each_entry(ag,
&mad_agent_list,
mad_agent_sec_list)
WRITE_ONCE(ag->smp_allowed,
!security_ib_endport_manage_subnet(ag->security,
dev_name(&ag->device->dev), ag->port_num));
spin_unlock(&mad_agent_list_lock);
}
int ib_mad_agent_security_setup(struct ib_mad_agent *agent,
......@@ -699,6 +701,8 @@ int ib_mad_agent_security_setup(struct ib_mad_agent *agent,
if (!rdma_protocol_ib(agent->device, agent->port_num))
return 0;
INIT_LIST_HEAD(&agent->mad_agent_sec_list);
ret = security_ib_alloc_security(&agent->security);
if (ret)
return ret;
......@@ -706,20 +710,22 @@ int ib_mad_agent_security_setup(struct ib_mad_agent *agent,
if (qp_type != IB_QPT_SMI)
return 0;
spin_lock(&mad_agent_list_lock);
ret = security_ib_endport_manage_subnet(agent->security,
dev_name(&agent->device->dev),
agent->port_num);
if (ret)
return ret;
goto free_security;
agent->lsm_nb.notifier_call = ib_mad_agent_security_change;
ret = register_lsm_notifier(&agent->lsm_nb);
if (ret)
return ret;
agent->smp_allowed = true;
agent->lsm_nb_reg = true;
WRITE_ONCE(agent->smp_allowed, true);
list_add(&agent->mad_agent_sec_list, &mad_agent_list);
spin_unlock(&mad_agent_list_lock);
return 0;
free_security:
spin_unlock(&mad_agent_list_lock);
security_ib_free_security(agent->security);
return ret;
}
void ib_mad_agent_security_cleanup(struct ib_mad_agent *agent)
......@@ -727,9 +733,13 @@ void ib_mad_agent_security_cleanup(struct ib_mad_agent *agent)
if (!rdma_protocol_ib(agent->device, agent->port_num))
return;
if (agent->qp->qp_type == IB_QPT_SMI) {
spin_lock(&mad_agent_list_lock);
list_del(&agent->mad_agent_sec_list);
spin_unlock(&mad_agent_list_lock);
}
security_ib_free_security(agent->security);
if (agent->lsm_nb_reg)
unregister_lsm_notifier(&agent->lsm_nb);
}
int ib_mad_enforce_security(struct ib_mad_agent_private *map, u16 pkey_index)
......@@ -738,7 +748,7 @@ int ib_mad_enforce_security(struct ib_mad_agent_private *map, u16 pkey_index)
return 0;
if (map->agent.qp->qp_type == IB_QPT_SMI) {
if (!map->agent.smp_allowed)
if (!READ_ONCE(map->agent.smp_allowed))
return -EACCES;
return 0;
}
......
drivers/infiniband/core/sysfs.c
浏览文件 @ a50243b1
......@@ -1015,9 +1015,7 @@ static void setup_hw_stats(struct ib_device *device, struct ib_port *port,
return;
}
static int add_port(struct ib_device *device, int port_num,
int (*port_callback)(struct ib_device *,
u8, struct kobject *))
static int add_port(struct ib_device *device, int port_num)
{
struct ib_port *p;
struct ib_port_attr attr;
......@@ -1113,8 +1111,8 @@ static int add_port(struct ib_device *device, int port_num,
if (ret)
goto err_free_pkey;
if (port_callback) {
ret = port_callback(device, port_num, &p->kobj);
if (device->ops.init_port) {
ret = device->ops.init_port(device, port_num, &p->kobj);
if (ret)
goto err_remove_pkey;
}
......@@ -1189,7 +1187,7 @@ static int add_port(struct ib_device *device, int port_num,
static ssize_t node_type_show(struct device *device,
struct device_attribute *attr, char *buf)
{
struct ib_device *dev = container_of(device, struct ib_device, dev);
struct ib_device *dev = rdma_device_to_ibdev(device);
switch (dev->node_type) {
case RDMA_NODE_IB_CA: return sprintf(buf, "%d: CA\n", dev->node_type);
......@@ -1206,7 +1204,7 @@ static DEVICE_ATTR_RO(node_type);
static ssize_t sys_image_guid_show(struct device *device,
struct device_attribute *dev_attr, char *buf)
{
struct ib_device *dev = container_of(device, struct ib_device, dev);
struct ib_device *dev = rdma_device_to_ibdev(device);
return sprintf(buf, "%04x:%04x:%04x:%04x\n",
be16_to_cpu(((__be16 *) &dev->attrs.sys_image_guid)[0]),
......@@ -1219,7 +1217,7 @@ static DEVICE_ATTR_RO(sys_image_guid);
static ssize_t node_guid_show(struct device *device,
struct device_attribute *attr, char *buf)
{
struct ib_device *dev = container_of(device, struct ib_device, dev);
struct ib_device *dev = rdma_device_to_ibdev(device);
return sprintf(buf, "%04x:%04x:%04x:%04x\n",
be16_to_cpu(((__be16 *) &dev->node_guid)[0]),
......@@ -1232,7 +1230,7 @@ static DEVICE_ATTR_RO(node_guid);
static ssize_t node_desc_show(struct device *device,
struct device_attribute *attr, char *buf)
{
struct ib_device *dev = container_of(device, struct ib_device, dev);
struct ib_device *dev = rdma_device_to_ibdev(device);
return sprintf(buf, "%.64s\n", dev->node_desc);
}
......@@ -1241,7 +1239,7 @@ static ssize_t node_desc_store(struct device *device,
struct device_attribute *attr,
const char *buf, size_t count)
{
struct ib_device *dev = container_of(device, struct ib_device, dev);
struct ib_device *dev = rdma_device_to_ibdev(device);
struct ib_device_modify desc = {};
int ret;
......@@ -1260,7 +1258,7 @@ static DEVICE_ATTR_RW(node_desc);
static ssize_t fw_ver_show(struct device *device, struct device_attribute *attr,
char *buf)
{
struct ib_device *dev = container_of(device, struct ib_device, dev);
struct ib_device *dev = rdma_device_to_ibdev(device);
ib_get_device_fw_str(dev, buf);
strlcat(buf, "\n", IB_FW_VERSION_NAME_MAX);
......@@ -1277,21 +1275,21 @@ static struct attribute *ib_dev_attrs[] = {
NULL,
};
static const struct attribute_group dev_attr_group = {
const struct attribute_group ib_dev_attr_group = {
.attrs = ib_dev_attrs,
};
static void free_port_list_attributes(struct ib_device *device)
static void ib_free_port_attrs(struct ib_device *device)
{
struct kobject *p, *t;
list_for_each_entry_safe(p, t, &device->port_list, entry) {
struct ib_port *port = container_of(p, struct ib_port, kobj);
list_del(&p->entry);
if (port->hw_stats) {
kfree(port->hw_stats);
if (port->hw_stats_ag)
free_hsag(&port->kobj, port->hw_stats_ag);
}
kfree(port->hw_stats);
if (port->pma_table)
sysfs_remove_group(p, port->pma_table);
......@@ -1308,62 +1306,47 @@ static void free_port_list_attributes(struct ib_device *device)
kobject_put(device->ports_kobj);
}
int ib_device_register_sysfs(struct ib_device *device,
int (*port_callback)(struct ib_device *,
u8, struct kobject *))
static int ib_setup_port_attrs(struct ib_device *device)
{
struct device *class_dev = &device->dev;
unsigned int port;
int ret;
int i;
device->groups[0] = &dev_attr_group;
class_dev->groups = device->groups;
ret = device_add(class_dev);
if (ret)
goto err;
device->ports_kobj = kobject_create_and_add("ports", &class_dev->kobj);
if (!device->ports_kobj) {
ret = -ENOMEM;
goto err_put;
}
device->ports_kobj = kobject_create_and_add("ports", &device->dev.kobj);
if (!device->ports_kobj)
return -ENOMEM;
if (rdma_cap_ib_switch(device)) {
ret = add_port(device, 0, port_callback);
rdma_for_each_port (device, port) {
ret = add_port(device, port);
if (ret)
goto err_put;
} else {
for (i = 1; i <= device->phys_port_cnt; ++i) {
ret = add_port(device, i, port_callback);
if (ret)
goto err_put;
}
}
if (device->ops.alloc_hw_stats)
setup_hw_stats(device, NULL, 0);
return 0;
err_put:
free_port_list_attributes(device);
device_del(class_dev);
err:
ib_free_port_attrs(device);
return ret;
}
void ib_device_unregister_sysfs(struct ib_device *device)
int ib_device_register_sysfs(struct ib_device *device)
{
/* Hold device until ib_dealloc_device() */
get_device(&device->dev);
int ret;
ret = ib_setup_port_attrs(device);
if (ret)
return ret;
if (device->ops.alloc_hw_stats)
setup_hw_stats(device, NULL, 0);
free_port_list_attributes(device);
return 0;
}
if (device->hw_stats) {
kfree(device->hw_stats);
void ib_device_unregister_sysfs(struct ib_device *device)
{
if (device->hw_stats_ag)
free_hsag(&device->dev.kobj, device->hw_stats_ag);
}
kfree(device->hw_stats);
device_unregister(&device->dev);
ib_free_port_attrs(device);
}
drivers/infiniband/core/ucma.c
浏览文件 @ a50243b1
......@@ -1236,6 +1236,13 @@ static int ucma_set_option_id(struct ucma_context *ctx, int optname,
}
ret = rdma_set_afonly(ctx->cm_id, *((int *) optval) ? 1 : 0);
break;
case RDMA_OPTION_ID_ACK_TIMEOUT:
if (optlen != sizeof(u8)) {
ret = -EINVAL;
break;
}
ret = rdma_set_ack_timeout(ctx->cm_id, *((u8 *)optval));
break;
default:
ret = -ENOSYS;
}
......
drivers/infiniband/core/umem.c
浏览文件 @ a50243b1
......@@ -72,15 +72,16 @@ static void __ib_umem_release(struct ib_device *dev, struct ib_umem *umem, int d
* If access flags indicate ODP memory, avoid pinning. Instead, stores
* the mm for future page fault handling in conjunction with MMU notifiers.
*
* @context: userspace context to pin memory for
* @udata: userspace context to pin memory for
* @addr: userspace virtual address to start at
* @size: length of region to pin
* @access: IB_ACCESS_xxx flags for memory being pinned
* @dmasync: flush in-flight DMA when the memory region is written
*/
struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr,
struct ib_umem *ib_umem_get(struct ib_udata *udata, unsigned long addr,
size_t size, int access, int dmasync)
{
struct ib_ucontext *context;
struct ib_umem *umem;
struct page **page_list;
struct vm_area_struct **vma_list;
......@@ -95,6 +96,14 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr,
struct scatterlist *sg, *sg_list_start;
unsigned int gup_flags = FOLL_WRITE;
if (!udata)
return ERR_PTR(-EIO);
context = container_of(udata, struct uverbs_attr_bundle, driver_udata)
->context;
if (!context)
return ERR_PTR(-EIO);
if (dmasync)
dma_attrs |= DMA_ATTR_WRITE_BARRIER;
......@@ -160,15 +169,12 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr,
lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
down_write(&mm->mmap_sem);
if (check_add_overflow(mm->pinned_vm, npages, &new_pinned) ||
(new_pinned > lock_limit && !capable(CAP_IPC_LOCK))) {
up_write(&mm->mmap_sem);
new_pinned = atomic64_add_return(npages, &mm->pinned_vm);
if (new_pinned > lock_limit && !capable(CAP_IPC_LOCK)) {
atomic64_sub(npages, &mm->pinned_vm);
ret = -ENOMEM;
goto out;
}
mm->pinned_vm = new_pinned;
up_write(&mm->mmap_sem);
cur_base = addr & PAGE_MASK;
......@@ -228,9 +234,7 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr,
umem_release:
__ib_umem_release(context->device, umem, 0);
vma:
down_write(&mm->mmap_sem);
mm->pinned_vm -= ib_umem_num_pages(umem);
up_write(&mm->mmap_sem);
atomic64_sub(ib_umem_num_pages(umem), &mm->pinned_vm);
out:
if (vma_list)
free_page((unsigned long) vma_list);
......@@ -253,25 +257,12 @@ static void __ib_umem_release_tail(struct ib_umem *umem)
kfree(umem);
}
static void ib_umem_release_defer(struct work_struct *work)
{
struct ib_umem *umem = container_of(work, struct ib_umem, work);
down_write(&umem->owning_mm->mmap_sem);
umem->owning_mm->pinned_vm -= ib_umem_num_pages(umem);
up_write(&umem->owning_mm->mmap_sem);
__ib_umem_release_tail(umem);
}
/**
* ib_umem_release - release memory pinned with ib_umem_get
* @umem: umem struct to release
*/
void ib_umem_release(struct ib_umem *umem)
{
struct ib_ucontext *context = umem->context;
if (umem->is_odp) {
ib_umem_odp_release(to_ib_umem_odp(umem));
__ib_umem_release_tail(umem);
......@@ -280,26 +271,7 @@ void ib_umem_release(struct ib_umem *umem)
__ib_umem_release(umem->context->device, umem, 1);
/*
* We may be called with the mm's mmap_sem already held. This
* can happen when a userspace munmap() is the call that drops
* the last reference to our file and calls our release
* method. If there are memory regions to destroy, we'll end
* up here and not be able to take the mmap_sem. In that case
* we defer the vm_locked accounting a workqueue.
*/
if (context->closing) {
if (!down_write_trylock(&umem->owning_mm->mmap_sem)) {
INIT_WORK(&umem->work, ib_umem_release_defer);
queue_work(ib_wq, &umem->work);
return;
}
} else {
down_write(&umem->owning_mm->mmap_sem);
}
umem->owning_mm->pinned_vm -= ib_umem_num_pages(umem);
up_write(&umem->owning_mm->mmap_sem);
atomic64_sub(ib_umem_num_pages(umem), &umem->owning_mm->pinned_vm);
__ib_umem_release_tail(umem);
}
EXPORT_SYMBOL(ib_umem_release);
......
drivers/infiniband/core/umem_odp.c
浏览文件 @ a50243b1
......@@ -40,6 +40,7 @@
#include <linux/vmalloc.h>
#include <linux/hugetlb.h>
#include <linux/interval_tree_generic.h>
#include <linux/pagemap.h>
#include <rdma/ib_verbs.h>
#include <rdma/ib_umem.h>
......@@ -299,7 +300,7 @@ static void free_per_mm(struct rcu_head *rcu)
kfree(container_of(rcu, struct ib_ucontext_per_mm, rcu));
}
void put_per_mm(struct ib_umem_odp *umem_odp)
static void put_per_mm(struct ib_umem_odp *umem_odp)
{
struct ib_ucontext_per_mm *per_mm = umem_odp->per_mm;
struct ib_ucontext *ctx = umem_odp->umem.context;
......@@ -332,9 +333,10 @@ void put_per_mm(struct ib_umem_odp *umem_odp)
mmu_notifier_call_srcu(&per_mm->rcu, free_per_mm);
}
struct ib_umem_odp *ib_alloc_odp_umem(struct ib_ucontext_per_mm *per_mm,
struct ib_umem_odp *ib_alloc_odp_umem(struct ib_umem_odp *root,
unsigned long addr, size_t size)
{
struct ib_ucontext_per_mm *per_mm = root->per_mm;
struct ib_ucontext *ctx = per_mm->context;
struct ib_umem_odp *odp_data;
struct ib_umem *umem;
......@@ -349,7 +351,7 @@ struct ib_umem_odp *ib_alloc_odp_umem(struct ib_ucontext_per_mm *per_mm,
umem->length = size;
umem->address = addr;
umem->page_shift = PAGE_SHIFT;
umem->writable = 1;
umem->writable = root->umem.writable;
umem->is_odp = 1;
odp_data->per_mm = per_mm;
umem->owning_mm = per_mm->mm;
......@@ -617,7 +619,7 @@ int ib_umem_odp_map_dma_pages(struct ib_umem_odp *umem_odp, u64 user_virt,
* mmget_not_zero will fail in this case.
*/
owning_process = get_pid_task(umem_odp->per_mm->tgid, PIDTYPE_PID);
if (WARN_ON(!mmget_not_zero(umem_odp->umem.owning_mm))) {
if (!owning_process || !mmget_not_zero(owning_mm)) {
ret = -EINVAL;
goto out_put_task;
}
......@@ -684,9 +686,14 @@ int ib_umem_odp_map_dma_pages(struct ib_umem_odp *umem_odp, u64 user_virt,
mutex_unlock(&umem_odp->umem_mutex);
if (ret < 0) {
/* Release left over pages when handling errors. */
for (++j; j < npages; ++j)
put_page(local_page_list[j]);
/*
* Release pages, remembering that the first page
* to hit an error was already released by
* ib_umem_odp_map_dma_single_page().
*/
if (npages - (j + 1) > 0)
release_pages(&local_page_list[j+1],
npages - (j + 1));
break;
}
}
......
drivers/infiniband/core/user_mad.c
浏览文件 @ a50243b1
......@@ -957,19 +957,22 @@ static int ib_umad_open(struct inode *inode, struct file *filp)
{
struct ib_umad_port *port;
struct ib_umad_file *file;
int ret = -ENXIO;
int ret = 0;
port = container_of(inode->i_cdev, struct ib_umad_port, cdev);
mutex_lock(&port->file_mutex);
if (!port->ib_dev)
if (!port->ib_dev) {
ret = -ENXIO;
goto out;
}
ret = -ENOMEM;
file = kzalloc(sizeof *file, GFP_KERNEL);
if (!file)
file = kzalloc(sizeof(*file), GFP_KERNEL);
if (!file) {
ret = -ENOMEM;
goto out;
}
mutex_init(&file->mutex);
spin_lock_init(&file->send_lock);
......@@ -982,14 +985,7 @@ static int ib_umad_open(struct inode *inode, struct file *filp)
list_add_tail(&file->port_list, &port->file_list);
ret = nonseekable_open(inode, filp);
if (ret) {
list_del(&file->port_list);
kfree(file);
goto out;
}
ib_umad_dev_get(port->umad_dev);
nonseekable_open(inode, filp);
out:
mutex_unlock(&port->file_mutex);
return ret;
......@@ -998,7 +994,6 @@ static int ib_umad_open(struct inode *inode, struct file *filp)
static int ib_umad_close(struct inode *inode, struct file *filp)
{
struct ib_umad_file *file = filp->private_data;
struct ib_umad_device *dev = file->port->umad_dev;
struct ib_umad_packet *packet, *tmp;
int already_dead;
int i;
......@@ -1027,7 +1022,6 @@ static int ib_umad_close(struct inode *inode, struct file *filp)
mutex_unlock(&file->port->file_mutex);
kfree(file);
ib_umad_dev_put(dev);
return 0;
}
......@@ -1073,17 +1067,9 @@ static int ib_umad_sm_open(struct inode *inode, struct file *filp)
filp->private_data = port;
ret = nonseekable_open(inode, filp);
if (ret)
goto err_clr_sm_cap;
ib_umad_dev_get(port->umad_dev);
nonseekable_open(inode, filp);
return 0;
err_clr_sm_cap:
swap(props.set_port_cap_mask, props.clr_port_cap_mask);
ib_modify_port(port->ib_dev, port->port_num, 0, &props);
err_up_sem:
up(&port->sm_sem);
......@@ -1106,7 +1092,6 @@ static int ib_umad_sm_close(struct inode *inode, struct file *filp)
up(&port->sm_sem);
ib_umad_dev_put(port->umad_dev);
return ret;
}
......@@ -1283,10 +1268,12 @@ static void ib_umad_kill_port(struct ib_umad_port *port)
mutex_unlock(&port->file_mutex);
cdev_device_del(&port->sm_cdev, &port->sm_dev);
put_device(&port->sm_dev);
cdev_device_del(&port->cdev, &port->dev);
put_device(&port->dev);
ida_free(&umad_ida, port->dev_num);
/* balances device_initialize() */
put_device(&port->sm_dev);
put_device(&port->dev);
}
static void ib_umad_add_one(struct ib_device *device)
......@@ -1329,21 +1316,24 @@ static void ib_umad_add_one(struct ib_device *device)
ib_umad_kill_port(&umad_dev->ports[i - s]);
}
free:
/* balances kref_init */
ib_umad_dev_put(umad_dev);
}
static void ib_umad_remove_one(struct ib_device *device, void *client_data)
{
struct ib_umad_device *umad_dev = client_data;
int i;
unsigned int i;
if (!umad_dev)
return;
for (i = 0; i <= rdma_end_port(device) - rdma_start_port(device); ++i) {
if (rdma_cap_ib_mad(device, i + rdma_start_port(device)))
ib_umad_kill_port(&umad_dev->ports[i]);
rdma_for_each_port (device, i) {
if (rdma_cap_ib_mad(device, i))
ib_umad_kill_port(
&umad_dev->ports[i - rdma_start_port(device)]);
}
/* balances kref_init() */
ib_umad_dev_put(umad_dev);
}
......
drivers/infiniband/core/uverbs_ioctl.c
浏览文件 @ a50243b1
......@@ -213,6 +213,7 @@ static int uverbs_process_idrs_array(struct bundle_priv *pbundle,
ret = PTR_ERR(attr->uobjects[i]);
break;
}
pbundle->bundle.context = attr->uobjects[i]->context;
}
attr->len = i;
......@@ -330,6 +331,7 @@ static int uverbs_process_attr(struct bundle_priv *pbundle,
uattr->data_s64);
if (IS_ERR(o_attr->uobject))
return PTR_ERR(o_attr->uobject);
pbundle->bundle.context = o_attr->uobject->context;
__set_bit(attr_bkey, pbundle->uobj_finalize);
if (spec->u.obj.access == UVERBS_ACCESS_NEW) {
......@@ -592,6 +594,7 @@ static int ib_uverbs_cmd_verbs(struct ib_uverbs_file *ufile,
pbundle->method_elm = method_elm;
pbundle->method_key = attrs_iter.index;
pbundle->bundle.ufile = ufile;
pbundle->bundle.context = NULL; /* only valid if bundle has uobject */
pbundle->radix = &uapi->radix;
pbundle->radix_slots = slot;
pbundle->radix_slots_len = radix_tree_chunk_size(&attrs_iter);
......
drivers/infiniband/core/uverbs_main.c
浏览文件 @ a50243b1
......@@ -695,6 +695,7 @@ static ssize_t ib_uverbs_write(struct file *filp, const char __user *buf,
memset(bundle.attr_present, 0, sizeof(bundle.attr_present));
bundle.ufile = file;
bundle.context = NULL; /* only valid if bundle has uobject */
if (!method_elm->is_ex) {
size_t in_len = hdr.in_words * 4 - sizeof(hdr);
size_t out_len = hdr.out_words * 4;
......@@ -1135,6 +1136,7 @@ static const struct file_operations uverbs_mmap_fops = {
static struct ib_client uverbs_client = {
.name = "uverbs",
.no_kverbs_req = true,
.add = ib_uverbs_add_one,
.remove = ib_uverbs_remove_one
};
......
drivers/infiniband/core/uverbs_std_types.c
浏览文件 @ a50243b1
......@@ -188,7 +188,7 @@ static int uverbs_free_pd(struct ib_uobject *uobject,
if (ret)
return ret;
ib_dealloc_pd((struct ib_pd *)uobject->object);
ib_dealloc_pd(pd);
return 0;
}
......
drivers/infiniband/core/verbs.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/cxgb3/iwch.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/cxgb4/cm.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/cxgb4/mem.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/cxgb4/qp.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/cxgb4/t4.h
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/Makefile
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/chip.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/chip.h
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/common.h
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/driver.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/fault.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/hfi.h
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/init.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/iowait.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/iowait.h
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/opfn.c 0 → 100644
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/opfn.h 0 → 100644
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/pio.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/qp.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/qp.h
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/rc.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/rc.h 0 → 100644
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/ruc.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/sdma.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/sysfs.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/trace.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/trace.h
浏览文件 @ a50243b1
......@@ -63,3 +63,4 @@ __print_symbolic(etype, \
#include "trace_tx.h"
#include "trace_mmu.h"
#include "trace_iowait.h"
#include "trace_tid.h"
drivers/infiniband/hw/hfi1/trace_tid.h 0 → 100644
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/uc.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/ud.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/verbs.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hfi1/verbs.h
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hns/Kconfig
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/hns/Makefile
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx4/Kconfig
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx4/cm.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx4/cq.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx4/main.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx4/mr.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx4/qp.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx4/srq.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx5/Kconfig
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx5/cong.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx5/cq.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx5/devx.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx5/ib_rep.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx5/main.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx5/mem.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx5/mr.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx5/odp.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx5/qp.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx5/srq.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/mlx5/srq.h
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/nes/Kconfig
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/qedr/main.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/qedr/verbs.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/qedr/verbs.h
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/qib/qib_rc.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/hw/qib/qib_ud.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rdmavt/mr.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rdmavt/pd.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rdmavt/pd.h
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rdmavt/qp.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rdmavt/rc.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rdmavt/srq.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rdmavt/vt.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rxe/rxe.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rxe/rxe.h
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rxe/rxe_av.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rxe/rxe_loc.h
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rxe/rxe_mr.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rxe/rxe_net.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rxe/rxe_net.h
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/sw/rxe/rxe_qp.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/infiniband/ulp/srp/ib_srp.c
浏览文件 @ a50243b1
此差异已折叠。
drivers/misc/mic/scif/scif_rma.c
浏览文件 @ a50243b1
此差异已折叠。
fs/proc/task_mmu.c
浏览文件 @ a50243b1
此差异已折叠。
include/linux/cgroup_rdma.h
浏览文件 @ a50243b1
此差异已折叠。
include/linux/mlx5/driver.h
浏览文件 @ a50243b1
此差异已折叠。
include/linux/mm_types.h
浏览文件 @ a50243b1
此差异已折叠。
include/linux/scatterlist.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/ib_hdrs.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/ib_mad.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/ib_umem.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/ib_umem_odp.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/ib_verbs.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/iw_cm.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/iw_portmap.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/rdma_cm.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/rdma_netlink.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/rdma_vt.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/rdmavt_qp.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/restrack.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/tid_rdma_defs.h 0 → 100644
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/uverbs_ioctl.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/uverbs_std_types.h
浏览文件 @ a50243b1
此差异已折叠。
include/rdma/uverbs_types.h
浏览文件 @ a50243b1
此差异已折叠。
include/uapi/rdma/bnxt_re-abi.h
浏览文件 @ a50243b1
此差异已折叠。
include/uapi/rdma/ib_user_verbs.h
浏览文件 @ a50243b1
此差异已折叠。
include/uapi/rdma/rdma_netlink.h
浏览文件 @ a50243b1
此差异已折叠。
include/uapi/rdma/rdma_user_cm.h
浏览文件 @ a50243b1
此差异已折叠。
include/uapi/rdma/rdma_user_rxe.h
浏览文件 @ a50243b1
此差异已折叠。
kernel/cgroup/rdma.c
浏览文件 @ a50243b1
此差异已折叠。
kernel/events/core.c
浏览文件 @ a50243b1
此差异已折叠。
kernel/fork.c
浏览文件 @ a50243b1
此差异已折叠。
lib/irq_poll.c
浏览文件 @ a50243b1
此差异已折叠。
lib/scatterlist.c
浏览文件 @ a50243b1
此差异已折叠。
mm/debug.c
浏览文件 @ a50243b1
此差异已折叠。
net/rds/ib.h
浏览文件 @ a50243b1
此差异已折叠。
net/rds/ib_fmr.c
浏览文件 @ a50243b1
此差异已折叠。
net/rds/ib_frmr.c
浏览文件 @ a50243b1
此差异已折叠。
net/rds/ib_recv.c
浏览文件 @ a50243b1
此差异已折叠。
net/rds/ib_send.c
浏览文件 @ a50243b1
此差异已折叠。
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或