Commit 1cbe06c3 authored by Linus Torvalds's avatar Linus Torvalds
Browse files

Merge tag 'for-linus' of git://

Pull more rdma updates from Doug Ledford:
 "This is the second group of code for the 4.7 merge window.  It looks
  large, but only in one sense.  I'll get to that in a minute.  The list
  of changes here breaks down as follows:

   - Dynamic counter infrastructure in the IB drivers

     This is a sysfs based code to allow free form access to the
     hardware counters RDMA devices might support so drivers don't need
     to code this up repeatedly themselves

   - SendOnlyFullMember multicast support

   - IB router support

   - A couple misc fixes

   - The big item on the list: hfi1 driver updates, plus moving the hfi1
     driver out of staging

  There was a group of 15 patches in the hfi1 list that I thought I had
  in the first pull request but they weren't.  So that added to the
  length of the hfi1 section here.

  As far as these go, everything but the hfi1 is pretty straight

  The hfi1 is, if you recall, the driver that Al had complaints about
  how it used the write/writev interfaces in an overloaded fashion.  The
  write portion of their interface behaved like the write handler in the
  IB stack proper and did bi-directional communications.  The writev
  interface, on the other hand, only accepts SDMA request structures.
  The completions for those structures are sent back via an entirely
  different event mechanism.

  With the security patch, we put security checks on the write
  interface, however, we also knew they would be going away soon.  Now,
  we've converted the write handler in the hfi1 driver to use ioctls
  from the IB reserved magic area for its bidirectional communications.
  With that change, Intel has addressed all of the items originally on
  their TODO when they went into staging (as well as many items added to
  the list later).

  As such, I moved them out, and since they were the last item in the
  staging/rdma directory, and I don't have immediate plans to use the
  staging area again, I removed the staging/rdma area.

  Because of the move out of staging, as well as a series of 5 patches
  in the hfi1 driver that removed code people thought should be done in
  a different way and was optional to begin with (a snoop debug
  interface, an eeprom driver for an eeprom connected directory to their
  hfi1 chip and not via an i2c bus, and a few other things like that),
  the line count, especially the removal count, is high"

* tag 'for-linus' of git:// (56 commits)
  staging/rdma: Remove the entire rdma subdirectory of staging
  IB/core: Make device counter infrastructure dynamic
  IB/hfi1: Fix pio map initialization
  IB/hfi1: Correct 8051 link parameter settings
  IB/hfi1: Update pkey table properly after link down or FM start
  IB/rdamvt: Fix rdmavt s_ack_queue sizing
  IB/rdmavt: Max atomic value should be a u8
  IB/hfi1: Fix hard lockup due to not using save/restore spin lock
  IB/hfi1: Add tracing support for send with invalidate opcode
  IB/hfi1, qib: Add ieth to the packet header definitions
  IB/hfi1: Move driver out of staging
  IB/hfi1: Do not free hfi1 cdev parent structure early
  IB/hfi1: Add trace message in user IOCTL handling
  IB/hfi1: Remove write(), use ioctl() for user cmds
  IB/hfi1: Add ioctl() interface for user commands
  IB/hfi1: Remove unused user command
  IB/hfi1: Remove snoop/diag interface
  IB/hfi1: Remove EPROM functionality from data device
  IB/hfi1: Remove UI char device
  IB/hfi1: Remove multiple device cdev
parents ed2608fa 7a226f9c
......@@ -56,6 +56,18 @@ SYSFS FILES
ports/1/pkeys/10 contains the value at index 10 in port 1's P_Key
There is an optional "hw_counters" subdirectory that may be under either
the parent device or the port subdirectories or both. If present,
there are a list of counters provided by the hardware. They may match
some of the counters in the counters directory, but they often include
many other counters. In addition to the various counters, there will
be a file named "lifespan" that configures how frequently the core
should update the counters when they are being accessed (counters are
not updated if they are not being accessed). The lifespan is in milli-
seconds and defaults to 10 unless set to something else by the driver.
Users may echo a value between 0 - 10000 to the lifespan file to set
the length of time between updates in milliseconds.
The Mellanox HCA driver also creates the files:
......@@ -5308,6 +5308,13 @@ F: drivers/block/cciss*
F: include/linux/cciss_ioctl.h
F: include/uapi/linux/cciss_ioctl.h
M: Mike Marciniszyn <>
M: Dennis Dalessandro <>
S: Supported
F: drivers/infiniband/hw/hfi1
S: Orphan
......@@ -5837,7 +5844,6 @@ T: git git://
S: Supported
F: Documentation/infiniband/
F: drivers/infiniband/
F: drivers/staging/rdma/
F: include/uapi/linux/if_infiniband.h
F: include/uapi/rdma/
F: include/rdma/
......@@ -10920,12 +10926,6 @@ M: Arnaud Patard <>
S: Odd Fixes
F: drivers/staging/xgifb/
M: Mike Marciniszyn <>
S: Supported
F: drivers/staging/rdma/hfi1
M: Ion Badulescu <>
S: Odd Fixes
......@@ -85,4 +85,6 @@ source "drivers/infiniband/ulp/isert/Kconfig"
source "drivers/infiniband/sw/rdmavt/Kconfig"
source "drivers/infiniband/hw/hfi1/Kconfig"
infiniband-$(CONFIG_INFINIBAND_ADDR_TRANS) := rdma_cm.o
user_access-$(CONFIG_INFINIBAND_ADDR_TRANS) := rdma_ucm.o
obj-$(CONFIG_INFINIBAND) += ib_core.o ib_mad.o ib_sa.o \
ib_cm.o iw_cm.o ib_addr.o \
obj-$(CONFIG_INFINIBAND) += ib_core.o ib_cm.o iw_cm.o \
obj-$(CONFIG_INFINIBAND_USER_MAD) += ib_umad.o
obj-$(CONFIG_INFINIBAND_USER_ACCESS) += ib_uverbs.o ib_ucm.o \
......@@ -10,14 +9,11 @@ obj-$(CONFIG_INFINIBAND_USER_ACCESS) += ib_uverbs.o ib_ucm.o \
ib_core-y := packer.o ud_header.o verbs.o cq.o rw.o sysfs.o \
device.o fmr_pool.o cache.o netlink.o \
roce_gid_mgmt.o mr_pool.o
roce_gid_mgmt.o mr_pool.o addr.o sa_query.o \
multicast.o mad.o smi.o agent.o mad_rmpp.o
ib_core-$(CONFIG_INFINIBAND_USER_MEM) += umem.o
ib_core-$(CONFIG_INFINIBAND_ON_DEMAND_PAGING) += umem_odp.o umem_rbtree.o
ib_mad-y := mad.o smi.o agent.o mad_rmpp.o
ib_sa-y := sa_query.o multicast.o
ib_cm-y := cm.o
iw_cm-y := iwcm.o iwpm_util.o iwpm_msg.o
......@@ -28,8 +24,6 @@ rdma_cm-$(CONFIG_INFINIBAND_ADDR_TRANS_CONFIGFS) += cma_configfs.o
rdma_ucm-y := ucma.o
ib_addr-y := addr.o
ib_umad-y := user_mad.o
ib_ucm-y := ucm.o
......@@ -46,10 +46,10 @@
#include <net/ip6_route.h>
#include <rdma/ib_addr.h>
#include <rdma/ib.h>
#include <rdma/rdma_netlink.h>
#include <net/netlink.h>
MODULE_AUTHOR("Sean Hefty");
MODULE_DESCRIPTION("IB Address Translation");
#include "core_priv.h"
struct addr_req {
struct list_head list;
......@@ -62,8 +62,11 @@ struct addr_req {
struct rdma_dev_addr *addr, void *context);
unsigned long timeout;
int status;
u32 seq;
static atomic_t ib_nl_addr_request_seq = ATOMIC_INIT(0);
static void process_req(struct work_struct *work);
static DEFINE_MUTEX(lock);
......@@ -71,6 +74,126 @@ static LIST_HEAD(req_list);
static DECLARE_DELAYED_WORK(work, process_req);
static struct workqueue_struct *addr_wq;
static const struct nla_policy ib_nl_addr_policy[LS_NLA_TYPE_MAX] = {
.len = sizeof(struct rdma_nla_ls_gid)},
static inline bool ib_nl_is_good_ip_resp(const struct nlmsghdr *nlh)
struct nlattr *tb[LS_NLA_TYPE_MAX] = {};
int ret;
if (nlh->nlmsg_flags & RDMA_NL_LS_F_ERR)
return false;
ret = nla_parse(tb, LS_NLA_TYPE_MAX - 1, nlmsg_data(nlh),
nlmsg_len(nlh), ib_nl_addr_policy);
if (ret)
return false;
return true;
static void ib_nl_process_good_ip_rsep(const struct nlmsghdr *nlh)
const struct nlattr *head, *curr;
union ib_gid gid;
struct addr_req *req;
int len, rem;
int found = 0;
head = (const struct nlattr *)nlmsg_data(nlh);
len = nlmsg_len(nlh);
nla_for_each_attr(curr, head, len, rem) {
if (curr->nla_type == LS_NLA_TYPE_DGID)
memcpy(&gid, nla_data(curr), nla_len(curr));
list_for_each_entry(req, &req_list, list) {
if (nlh->nlmsg_seq != req->seq)
/* We set the DGID part, the rest was set earlier */
rdma_addr_set_dgid(req->addr, &gid);
req->status = 0;
found = 1;
if (!found)
pr_info("Couldn't find request waiting for DGID: %pI6\n",
int ib_nl_handle_ip_res_resp(struct sk_buff *skb,
struct netlink_callback *cb)
const struct nlmsghdr *nlh = (struct nlmsghdr *)cb->nlh;
if ((nlh->nlmsg_flags & NLM_F_REQUEST) ||
!(NETLINK_CB(skb).sk) ||
!netlink_capable(skb, CAP_NET_ADMIN))
return -EPERM;
if (ib_nl_is_good_ip_resp(nlh))
return skb->len;
static int ib_nl_ip_send_msg(struct rdma_dev_addr *dev_addr,
const void *daddr,
u32 seq, u16 family)
struct sk_buff *skb = NULL;
struct nlmsghdr *nlh;
struct rdma_ls_ip_resolve_header *header;
void *data;
size_t size;
int attrtype;
int len;
if (family == AF_INET) {
size = sizeof(struct in_addr);
} else {
size = sizeof(struct in6_addr);
len = nla_total_size(sizeof(size));
len += NLMSG_ALIGN(sizeof(*header));
skb = nlmsg_new(len, GFP_KERNEL);
if (!skb)
return -ENOMEM;
data = ibnl_put_msg(skb, &nlh, seq, 0, RDMA_NL_LS,
if (!data) {
return -ENODATA;
/* Construct the family header first */
header = (struct rdma_ls_ip_resolve_header *)
skb_put(skb, NLMSG_ALIGN(sizeof(*header)));
header->ifindex = dev_addr->bound_dev_if;
nla_put(skb, attrtype, size, daddr);
/* Repair the nlmsg header length */
nlmsg_end(skb, nlh);
ibnl_multicast(skb, nlh, RDMA_NL_GROUP_LS, GFP_KERNEL);
/* Make the request retry, so when we get the response from userspace
* we will have something.
return -ENODATA;
int rdma_addr_size(struct sockaddr *addr)
switch (addr->sa_family) {
......@@ -199,6 +322,17 @@ static void queue_req(struct addr_req *req)
static int ib_nl_fetch_ha(struct dst_entry *dst, struct rdma_dev_addr *dev_addr,
const void *daddr, u32 seq, u16 family)
if (ibnl_chk_listeners(RDMA_NL_GROUP_LS))
/* We fill in what we can, the response will fill the rest */
rdma_copy_addr(dev_addr, dst->dev, NULL);
return ib_nl_ip_send_msg(dev_addr, daddr, seq, family);
static int dst_fetch_ha(struct dst_entry *dst, struct rdma_dev_addr *dev_addr,
const void *daddr)
......@@ -223,6 +357,39 @@ static int dst_fetch_ha(struct dst_entry *dst, struct rdma_dev_addr *dev_addr,
return ret;
static bool has_gateway(struct dst_entry *dst, sa_family_t family)
struct rtable *rt;
struct rt6_info *rt6;
if (family == AF_INET) {
rt = container_of(dst, struct rtable, dst);
return rt->rt_uses_gateway;
rt6 = container_of(dst, struct rt6_info, dst);
return rt6->rt6i_flags & RTF_GATEWAY;
static int fetch_ha(struct dst_entry *dst, struct rdma_dev_addr *dev_addr,
const struct sockaddr *dst_in, u32 seq)
const struct sockaddr_in *dst_in4 =
(const struct sockaddr_in *)dst_in;
const struct sockaddr_in6 *dst_in6 =
(const struct sockaddr_in6 *)dst_in;
const void *daddr = (dst_in->sa_family == AF_INET) ?
(const void *)&dst_in4->sin_addr.s_addr :
(const void *)&dst_in6->sin6_addr;
sa_family_t family = dst_in->sa_family;
/* Gateway + ARPHRD_INFINIBAND -> IB router */
if (has_gateway(dst, family) && dst->dev->type == ARPHRD_INFINIBAND)
return ib_nl_fetch_ha(dst, dev_addr, daddr, seq, family);
return dst_fetch_ha(dst, dev_addr, daddr);
static int addr4_resolve(struct sockaddr_in *src_in,
const struct sockaddr_in *dst_in,
struct rdma_dev_addr *addr,
......@@ -246,10 +413,11 @@ static int addr4_resolve(struct sockaddr_in *src_in,
src_in->sin_family = AF_INET;
src_in->sin_addr.s_addr = fl4.saddr;
/* If there's a gateway, we're definitely in RoCE v2 (as RoCE v1 isn't
* routable) and we could set the network type accordingly.
/* If there's a gateway and type of device not ARPHRD_INFINIBAND, we're
* definitely in RoCE v2 (as RoCE v1 isn't routable) set the network
* type accordingly.
if (rt->rt_uses_gateway)
if (rt->rt_uses_gateway && rt->>type != ARPHRD_INFINIBAND)
addr->network = RDMA_NETWORK_IPV4;
addr->hoplimit = ip4_dst_hoplimit(&rt->dst);
......@@ -291,10 +459,12 @@ static int addr6_resolve(struct sockaddr_in6 *src_in,
src_in->sin6_addr = fl6.saddr;
/* If there's a gateway, we're definitely in RoCE v2 (as RoCE v1 isn't
* routable) and we could set the network type accordingly.
/* If there's a gateway and type of device not ARPHRD_INFINIBAND, we're
* definitely in RoCE v2 (as RoCE v1 isn't routable) set the network
* type accordingly.
if (rt->rt6i_flags & RTF_GATEWAY)
if (rt->rt6i_flags & RTF_GATEWAY &&
ip6_dst_idev(dst)->dev->type != ARPHRD_INFINIBAND)
addr->network = RDMA_NETWORK_IPV6;
addr->hoplimit = ip6_dst_hoplimit(dst);
......@@ -317,7 +487,8 @@ static int addr6_resolve(struct sockaddr_in6 *src_in,
static int addr_resolve_neigh(struct dst_entry *dst,
const struct sockaddr *dst_in,
struct rdma_dev_addr *addr)
struct rdma_dev_addr *addr,
u32 seq)
if (dst->dev->flags & IFF_LOOPBACK) {
int ret;
......@@ -331,17 +502,8 @@ static int addr_resolve_neigh(struct dst_entry *dst,
/* If the device doesn't do ARP internally */
if (!(dst->dev->flags & IFF_NOARP)) {
const struct sockaddr_in *dst_in4 =
(const struct sockaddr_in *)dst_in;
const struct sockaddr_in6 *dst_in6 =
(const struct sockaddr_in6 *)dst_in;
return dst_fetch_ha(dst, addr,
dst_in->sa_family == AF_INET ?
(const void *)&dst_in4->sin_addr.s_addr :
(const void *)&dst_in6->sin6_addr);
if (!(dst->dev->flags & IFF_NOARP))
return fetch_ha(dst, addr, dst_in, seq);
return rdma_copy_addr(addr, dst->dev, NULL);
......@@ -349,7 +511,8 @@ static int addr_resolve_neigh(struct dst_entry *dst,
static int addr_resolve(struct sockaddr *src_in,
const struct sockaddr *dst_in,
struct rdma_dev_addr *addr,
bool resolve_neigh)
bool resolve_neigh,
u32 seq)
struct net_device *ndev;
struct dst_entry *dst;
......@@ -366,7 +529,7 @@ static int addr_resolve(struct sockaddr *src_in,
return ret;
if (resolve_neigh)
ret = addr_resolve_neigh(&rt->dst, dst_in, addr);
ret = addr_resolve_neigh(&rt->dst, dst_in, addr, seq);
ndev = rt->;
......@@ -383,7 +546,7 @@ static int addr_resolve(struct sockaddr *src_in,
return ret;
if (resolve_neigh)
ret = addr_resolve_neigh(dst, dst_in, addr);
ret = addr_resolve_neigh(dst, dst_in, addr, seq);
ndev = dst->dev;
......@@ -412,7 +575,7 @@ static void process_req(struct work_struct *work)
src_in = (struct sockaddr *) &req->src_addr;
dst_in = (struct sockaddr *) &req->dst_addr;
req->status = addr_resolve(src_in, dst_in, req->addr,
true, req->seq);
if (req->status && time_after_eq(jiffies, req->timeout))
req->status = -ETIMEDOUT;
else if (req->status == -ENODATA)
......@@ -471,8 +634,9 @@ int rdma_resolve_ip(struct rdma_addr_client *client,
req->context = context;
req->client = client;
req->seq = (u32)atomic_inc_return(&ib_nl_addr_request_seq);
req->status = addr_resolve(src_in, dst_in, addr, true);
req->status = addr_resolve(src_in, dst_in, addr, true, req->seq);
switch (req->status) {
case 0:
req->timeout = jiffies;
......@@ -510,7 +674,7 @@ int rdma_resolve_ip_route(struct sockaddr *src_addr,
src_in->sa_family = dst_addr->sa_family;
return addr_resolve(src_in, dst_addr, addr, false);
return addr_resolve(src_in, dst_addr, addr, false, 0);
......@@ -634,7 +798,7 @@ static struct notifier_block nb = {
.notifier_call = netevent_callback
static int __init addr_init(void)
int addr_init(void)
addr_wq = create_singlethread_workqueue("ib_addr");
if (!addr_wq)
......@@ -642,15 +806,13 @@ static int __init addr_init(void)
return 0;
static void __exit addr_cleanup(void)
void addr_cleanup(void)
......@@ -137,4 +137,20 @@ static inline bool rdma_is_upper_dev_rcu(struct net_device *dev,
return _upper == upper;
int addr_init(void);
void addr_cleanup(void);
int ib_mad_init(void);
void ib_mad_cleanup(void);
int ib_sa_init(void);
void ib_sa_cleanup(void);
int ib_nl_handle_resolve_resp(struct sk_buff *skb,
struct netlink_callback *cb);
int ib_nl_handle_set_timeout(struct sk_buff *skb,
struct netlink_callback *cb);
int ib_nl_handle_ip_res_resp(struct sk_buff *skb,
struct netlink_callback *cb);
#endif /* _CORE_PRIV_H */
......@@ -955,6 +955,29 @@ struct net_device *ib_get_net_dev_by_params(struct ib_device *dev,
static struct ibnl_client_cbs ibnl_ls_cb_table[] = {
.dump = ib_nl_handle_resolve_resp,
.module = THIS_MODULE },
.dump = ib_nl_handle_set_timeout,
.module = THIS_MODULE },
.dump = ib_nl_handle_ip_res_resp,
.module = THIS_MODULE },
static int ib_add_ibnl_clients(void)
return ibnl_add_client(RDMA_NL_LS, ARRAY_SIZE(ibnl_ls_cb_table),
static void ib_remove_ibnl_clients(void)
static int __init ib_core_init(void)
int ret;
......@@ -983,10 +1006,41 @@ static int __init ib_core_init(void)
goto err_sysfs;
ret = addr_init();
if (ret) {
pr_warn("Could't init IB address resolution\n");
goto err_ibnl;
ret = ib_mad_init();
if (ret) {
pr_warn("Couldn't init IB MAD\n");
goto err_addr;
ret = ib_sa_init();
if (ret) {
pr_warn("Couldn't init SA\n");
goto err_mad;
if (ib_add_ibnl_clients()) {
pr_warn("Couldn't register ibnl clients\n");
goto err_sa;
return 0;
......@@ -999,6 +1053,10 @@ static int __init ib_core_init(void)
static void __exit ib_core_cleanup(void)
......@@ -47,11 +47,7 @@
#include "smi.h"
#include "opa_smi.h"
#include "agent.h"
MODULE_AUTHOR("Hal Rosenstock");
MODULE_AUTHOR("Sean Hefty");
#include "core_priv.h"
static int mad_sendq_size = IB_MAD_QP_SEND_SIZE;
static int mad_recvq_size = IB_MAD_QP_RECV_SIZE;
......@@ -3316,7 +3312,7 @@ static struct ib_client mad_client = {
.remove = ib_mad_remove_device
static int __init ib_mad_init_module(void)
int ib_mad_init(void)
mad_recvq_size = min(mad_recvq_size, IB_MAD_QP_MAX_SIZE);
mad_recvq_size = max(mad_recvq_size, IB_MAD_QP_MIN_SIZE);
......@@ -3334,10 +3330,7 @@ static int __init ib_mad_init_module(void)
return 0;
static void __exit ib_mad_cleanup_module(void)
void ib_mad_cleanup(void)
......@@ -93,6 +93,18 @@ enum {
struct mcast_member;
* There are 4 types of join states:
* FullMember, NonMember, SendOnlyNonMember, SendOnlyFullMember.
enum {
struct mcast_group {
struct ib_sa_mcmember_rec rec;
struct rb_node node;
......@@ -102,7 +114,7 @@ struct mcast_group {
struct list_head pending_list;
struct list_head active_list;
struct mcast_member *last_join;
int members[3];
atomic_t refcount;
enum mcast_group_state state;
struct ib_sa_query *query;
......@@ -220,8 +232,9 @@ static void queue_join(struct mcast_member *member)
* A multicast group has three types of members: full member, non member, and
* send only member. We need to keep track of the number of members of each
* A multicast group has four types of members: full member, non member,
* sendonly non member and sendonly full member.