linux.git/drivers/infiniband/core/addr.c, branch for-next

Revert "RDMA/core: Fix ENODEV error for iWARP test over vlan"

2024-11-12T14:53:11+00:00

The citied commit in Fixes line caused to regression for udaddy [1]
application. It doesn't work over VLANs anymore.

Client:
  ifconfig eth2 1.1.1.1
  ip link add link eth2 name p0.3597 type vlan protocol 802.1Q id 3597
  ip link set dev p0.3597 up
  ip addr add 2.2.2.2/16 dev p0.3597
  udaddy -S 847 -C 220 -c 2 -t 0 -s 2.2.2.3 -b 2.2.2.2

Server:
  ifconfig eth2 1.1.1.3
  ip link add link eth2 name p0.3597 type vlan protocol 802.1Q id 3597
  ip link set dev p0.3597 up
  ip addr add 2.2.2.3/16 dev p0.3597
  udaddy -S 847 -C 220 -c 2 -t 0 -b 2.2.2.3

[1] https://github.com/linux-rdma/rdma-core/blob/master/librdmacm/examples/udaddy.c

Fixes: 5069d7e202f6 ("RDMA/core: Fix ENODEV error for iWARP test over vlan")
Reported-by: Leon Romanovsky 
Closes: https://lore.kernel.org/all/20241110130746.GA48891@unreal
Link: https://patch.msgid.link/bb9d403419b2b9566da5b8bf0761fa8377927e49.1731401658.git.leon@kernel.org
Signed-off-by: Leon Romanovsky

RDMA/core: Fix ENODEV error for iWARP test over vlan

2024-10-08T12:07:41+00:00

If traffic is over vlan, cma_validate_port() fails to match vlan
net_device ifindex with bound_if_index and results in ENODEV error.
It is because rdma_copy_src_l2_addr() always assigns bound_if_index with
real net_device ifindex.
This patch fixes the issue by assigning bound_if_index with vlan
net_device index if traffic is over vlan.

Fixes: f8ef1be816bf ("RDMA/cma: Avoid GID lookups on iWARP devices")
Signed-off-by: Anumula Murali Mohan Reddy 
Signed-off-by: Potnuri Bharat Teja 
Link: https://patch.msgid.link/20241008114334.146702-1-anumula@chelsio.com
Signed-off-by: Leon Romanovsky

inet: introduce dst_rtable() helper

2024-05-01T01:32:38+00:00

I added dst_rt6_info() in commit
e8dfd42c17fa ("ipv6: introduce dst_rt6_info() helper")

This patch does a similar change for IPv4.

Instead of (struct rtable *)dst casts, we can use :

 #define dst_rtable(_ptr) \
             container_of_const(_ptr, struct rtable, dst)

Patch is smaller than IPv6 one, because IPv4 has skb_rtable() helper.

Signed-off-by: Eric Dumazet 
Reviewed-by: David Ahern 
Reviewed-by: Sabrina Dubroca 
Link: https://lore.kernel.org/r/20240429133009.1227754-1-edumazet@google.com
Signed-off-by: Jakub Kicinski

ipv6: introduce dst_rt6_info() helper

2024-04-29T12:32:01+00:00

Instead of (struct rt6_info *)dst casts, we can use :

 #define dst_rt6_info(_ptr) \
         container_of_const(_ptr, struct rt6_info, dst)

Some places needed missing const qualifiers :

ip6_confirm_neigh(), ipv6_anycast_destination(),
ipv6_unicast_destination(), has_gateway()

v2: added missing parts (David Ahern)

Signed-off-by: Eric Dumazet 
Reviewed-by: David Ahern 
Signed-off-by: David S. Miller

RDMA/core: Delete useless module.h include

2022-01-28T17:03:12+00:00

There is no need in include of module.h in the following files.

Link: https://lore.kernel.org/r/e412c83b45b6ebdd937886cc9c2cc7c8abcc34fa.1642960861.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky 
Signed-off-by: Jason Gunthorpe

RDMA/addr: Be strict with gid size

2021-04-08T19:14:56+00:00

The nla_len() is less than or equal to 16.  If it's less than 16 then end
of the "gid" buffer is uninitialized.

Fixes: ae43f8286730 ("IB/core: Add IP to GID netlink offload")
Link: https://lore.kernel.org/r/20210405074434.264221-1-leon@kernel.org
Reported-by: Dan Carpenter 
Signed-off-by: Mark Bloch 
Signed-off-by: Leon Romanovsky 
Signed-off-by: Jason Gunthorpe

RDMA/addr: Fix race with netevent_callback()/rdma_addr_cancel()

2020-09-30T18:29:05+00:00

This three thread race can result in the work being run once the callback
becomes NULL:

       CPU1                 CPU2                   CPU3
 netevent_callback()
                     process_one_req()       rdma_addr_cancel()
                      [..]
     spin_lock_bh()
  	set_timeout()
     spin_unlock_bh()

						spin_lock_bh()
						list_del_init(&req->list);
						spin_unlock_bh()

		     req->callback = NULL
		     spin_lock_bh()
		       if (!list_empty(&req->list))
                         // Skipped!
		         // cancel_delayed_work(&req->work);
		     spin_unlock_bh()

		    process_one_req() // again
		     req->callback() // BOOM
						cancel_delayed_work_sync()

The solution is to always cancel the work once it is completed so any
in between set_timeout() does not result in it running again.

Cc: stable@vger.kernel.org
Fixes: 44e75052bc2a ("RDMA/rdma_cm: Make rdma_addr_cancel into a fence")
Link: https://lore.kernel.org/r/20200930072007.1009692-1-leon@kernel.org
Reported-by: Dan Aloni 
Signed-off-by: Leon Romanovsky 
Signed-off-by: Jason Gunthorpe

RDMA/addr: Mark addr_resolve as might_sleep()

2020-05-13T00:32:52+00:00

Under one path through ib_nl_fetch_ha() this calls nlmsg_new(GFP_KERNEL)
which is a sleeping call. This is a very rare path, so mark fetch_ha() and
the module external entry point that conditionally calls through to
fetch_ha() as might_sleep().

Link: https://lore.kernel.org/r/20200506074701.9775-2-leon@kernel.org
Signed-off-by: Leon Romanovsky 
Signed-off-by: Jason Gunthorpe

RDMA/netlink: Do not always generate an ACK for some netlink operations

2020-01-03T20:02:32+00:00

In rdma_nl_rcv_skb(), the local variable err is assigned the return value
of the supplied callback function, which could be one of
ib_nl_handle_resolve_resp(), ib_nl_handle_set_timeout(), or
ib_nl_handle_ip_res_resp(). These three functions all return skb->len on
success.

rdma_nl_rcv_skb() is merely a copy of netlink_rcv_skb(). The callback
functions used by the latter have the convention: "Returns 0 on success or
a negative error code".

In particular, the statement (equal for both functions):

   if (nlh->nlmsg_flags & NLM_F_ACK || err)

implies that rdma_nl_rcv_skb() always will ack a message, independent of
the NLM_F_ACK being set in nlmsg_flags or not.

The fix could be to change the above statement, but it is better to keep
the two *_rcv_skb() functions equal in this respect and instead change the
three callback functions in the rdma subsystem to the correct convention.

Fixes: 2ca546b92a02 ("IB/sa: Route SA pathrecord query through netlink")
Fixes: ae43f8286730 ("IB/core: Add IP to GID netlink offload")
Link: https://lore.kernel.org/r/20191216120436.3204814-1-haakon.bugge@oracle.com
Suggested-by: Mark Haywood 
Signed-off-by: Håkon Bugge 
Tested-by: Mark Haywood 
Reviewed-by: Leon Romanovsky 
Reviewed-by: Jason Gunthorpe 
Signed-off-by: Jason Gunthorpe

net: ipv6_stub: use ip6_dst_lookup_flow instead of ip6_dst_lookup

2019-12-04T20:27:13+00:00

ipv6_stub uses the ip6_dst_lookup function to allow other modules to
perform IPv6 lookups. However, this function skips the XFRM layer
entirely.

All users of ipv6_stub->ip6_dst_lookup use ip_route_output_flow (via the
ip_route_output_key and ip_route_output helpers) for their IPv4 lookups,
which calls xfrm_lookup_route(). This patch fixes this inconsistent
behavior by switching the stub to ip6_dst_lookup_flow, which also calls
xfrm_lookup_route().

This requires some changes in all the callers, as these two functions
take different arguments and have different return types.

Fixes: 5f81bd2e5d80 ("ipv6: export a stub for IPv6 symbols used by vxlan")
Reported-by: Xiumei Mu 
Signed-off-by: Sabrina Dubroca 
Signed-off-by: David S. Miller