Send an RDMA-CM private message on connect, and look for one during
a connection-established event.

Both sides can communicate their various implementation limits.
Implementations that don't support this sideband protocol ignore it.

Once the client knows the server's inline threshold maxima, it can
adjust the use of Reply chunks, and eliminate most use of Position
Zero Read chunks. Moderately-sized I/O can be done using a pure
inline RDMA Send instead of RDMA operations that require memory
registration.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Clean up: The fields in the recv_wr do not vary. There is no need to
initialize them before each ib_post_recv(). This removes a large-ish
data structure from the stack.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Clean up: Most of the fields in each send_wr do not vary. There is
no need to initialize them before each ib_post_send(). This removes
a large-ish data structure from the stack.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Clean up.

Since commit fc664485 ("xprtrdma: Split the completion queue"),
rpcrdma_ep_post_recv() no longer uses the "ep" argument.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Clean up. The "ia" argument is no longer used.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Currently, each regbuf is allocated and DMA mapped at the same time.
This is done during transport creation.

When a device driver is unloaded, every DMA-mapped buffer in use by
a transport has to be unmapped, and then remapped to the new
device if the driver is loaded again. Remapping will have to be done
_after_ the connect worker has set up the new device.

But there's an ordering problem:

call_allocate, which invokes xprt_rdma_allocate which calls
rpcrdma_alloc_regbuf to allocate Send buffers, happens _before_
the connect worker can run to set up the new device.

Instead, at transport creation, allocate each buffer, but leave it
unmapped. Once the RPC carries these buffers into ->send_request, by
which time a transport connection should have been established,
check to see that the RPC's buffers have been DMA mapped. If not,
map them there.

When device driver unplug support is added, it will simply unmap all
the transport's regbufs, but it doesn't have to deallocate the
underlying memory.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

The use of DMA_BIDIRECTIONAL is discouraged by DMA-API.txt.
Fortunately, xprtrdma now knows which direction I/O is going as
soon as it allocates each regbuf.

The RPC Call and Reply buffers are no longer the same regbuf. They
can each be labeled correctly now. The RPC Reply buffer is never
part of either a Send or Receive WR, but it can be part of Reply
chunk, which is mapped and registered via ->ro_map . So it is not
DMA mapped when it is allocated (DMA_NONE), to avoid a double-
mapping.

Since Receive buffers are no longer DMA_BIDIRECTIONAL and their
contents are never modified by the host CPU, DMA-API-HOWTO.txt
suggests that a DMA sync before posting each buffer should be
unnecessary. (See my_card_interrupt_handler).
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Commit 94931746 ("xprtrdma: Limit number of RDMA segments in
RPC-over-RDMA headers") capped the number of chunks that may appear
in RPC-over-RDMA headers. The maximum header size can be estimated
and fixed to avoid allocating buffer space that is never used.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

RPC-over-RDMA needs to separate its RPC call and reply buffers.

 o When an RPC Call is sent, rq_snd_buf is DMA mapped for an RDMA
   Send operation using DMA_TO_DEVICE

 o If the client expects a large RPC reply, it DMA maps rq_rcv_buf
   as part of a Reply chunk using DMA_FROM_DEVICE

The two mappings are for data movement in opposite directions.

DMA-API.txt suggests that if these mappings share a DMA cacheline,
bad things can happen. This could occur in the final bytes of
rq_snd_buf and the first bytes of rq_rcv_buf if the two buffers
happen to share a DMA cacheline.

On x86_64 the cacheline size is typically 8 bytes, and RPC call
messages are usually much smaller than the send buffer, so this
hasn't been a noticeable problem. But the DMA cacheline size can be
larger on other platforms.

Also, often rq_rcv_buf starts most of the way into a page, thus
an additional RDMA segment is needed to map and register the end of
that buffer. Try to avoid that scenario to reduce the cost of
registering and invalidating Reply chunks.

Instead of carrying a single regbuf that covers both rq_snd_buf and
rq_rcv_buf, each struct rpcrdma_req now carries one regbuf for
rq_snd_buf and one regbuf for rq_rcv_buf.

Some incidental changes worth noting:

- To clear out some spaghetti, refactor xprt_rdma_allocate.
- The value stored in rg_size is the same as the value stored in
  the iov.length field, so eliminate rg_size
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Currently there's a hidden and indirect mechanism for finding the
rpcrdma_req that goes with an rpc_rqst. It depends on getting from
the rq_buffer pointer in struct rpc_rqst to the struct
rpcrdma_regbuf that controls that buffer, and then to the struct
rpcrdma_req it goes with.

This was done back in the day to avoid the need to add a per-rqst
pointer or to alter the buf_free API when support for RPC-over-RDMA
was introduced.

I'm about to change the way regbuf's work to support larger inline
thresholds. Now is a good time to replace this indirect mechanism
with something that is more straightforward. I guess this should be
considered a clean up.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

For xprtrdma, the RPC Call and Reply buffers are involved in real
I/O operations.

To start with, the DMA direction of the I/O for a Call is opposite
that of a Reply.

In the current arrangement, the Reply buffer address is on a
four-byte alignment just past the call buffer. Would be friendlier
on some platforms if that was at a DMA cache alignment instead.

Because the current arrangement allocates a single memory region
which contains both buffers, the RPC Reply buffer often contains a
page boundary in it when the Call buffer is large enough (which is
frequent).

It would be a little nicer for setting up DMA operations (and
possible registration of the Reply buffer) if the two buffers were
separated, well-aligned, and contained as few page boundaries as
possible.

Now, I could just pad out the single memory region used for the pair
of buffers. But frequently that would mean a lot of unused space to
ensure the Reply buffer did not have a page boundary.

Add a separate pointer to rpc_rqst that points right to the RPC
Reply buffer. This makes no difference to xprtsock, but it will help
xprtrdma in subsequent patches.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

xprtrdma needs to allocate the Call and Reply buffers separately.
TBH, the reliance on using a single buffer for the pair of XDR
buffers is transport implementation-specific.

Instead of passing just the rq_buffer into the buf_free method, pass
the task structure and let buf_free take care of freeing both
XDR buffers at once.

There's a micro-optimization here. In the common case, both
xprt_release and the transport's buf_free method were checking if
rq_buffer was NULL. Now the check is done only once per RPC.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

xprtrdma needs to allocate the Call and Reply buffers separately.
TBH, the reliance on using a single buffer for the pair of XDR
buffers is transport implementation-specific.

Transports that want to allocate separate Call and Reply buffers
will ignore the "size" argument anyway.  Don't bother passing it.

The buf_alloc method can't return two pointers. Instead, make the
method's return value an error code, and set the rq_buffer pointer
in the method itself.

This gives call_allocate an opportunity to terminate an RPC instead
of looping forever when a permanent problem occurs. If a request is
just bogus, or the transport is in a state where it can't allocate
resources for any request, there needs to be a way to kill the RPC
right there and not loop.

This immediately fixes a rare problem in the backchannel send path,
which loops if the server happens to send a CB request whose
call+reply size is larger than a page (which it shouldn't do yet).

One more issue: looks like xprt_inject_disconnect was incorrectly
placed in the failure path in call_allocate. It needs to be in the
success path, as it is for other call-sites.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Clean up: there is some XDR initialization logic that is common
to the forward channel and backchannel. Move it to an XDR header
so it can be shared.

rpc_rqst::rq_buffer points to a buffer containing big-endian data.
Update its annotation as part of the clean up.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Clean up: r_xprt is already available everywhere these macros are
invoked, so just dereference that directly.

RPCRDMA_INLINE_PAD_VALUE is no longer used, so it can simply be
removed.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Use a setup function to call into the NFS layer to test an rpc_xprt
for session trunking so as to not leak the rpc_xprt_switch into
the nfs layer.

Search for the address in the rpc_xprt_switch first so as not to
put an unnecessary EXCHANGE_ID on the wire.
Signed-off-by: NAndy Adamson <andros@netapp.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Signed-off-by: NAndy Adamson <andros@netapp.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Give the NFS layer access to the rpc_xprt_switch_add_xprt function
Signed-off-by: NAndy Adamson <andros@netapp.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Give the NFS layer access to the xprt_switch_put function
Signed-off-by: NAndy Adamson <andros@netapp.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

rpc_task_set_client is only called from rpc_run_task after
rpc_new_task and rpc_task_release_client is not needed as the
task is new.

When called from rpc_new_task, rpc_task_set_client also removed the
assigned rpc_xprt which is not desired.
Signed-off-by: NAndy Adamson <andros@netapp.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

Clean up.
Signed-off-by: NTrond Myklebust <trond.myklebust@primarydata.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

The variable `err` is not used anywhere and just returns the
predefined value `0` at the end of the function. Hence, remove the
variable and return 0 explicitly.
Signed-off-by: NAmitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>

rsc_lookup steals the passed-in memory to avoid doing an allocation of
its own, so we can't just pass in a pointer to memory that someone else
is using.

If we really want to avoid allocation there then maybe we should
preallocate somwhere, or reference count these handles.

For now we should revert.

On occasion I see this on my server:

kernel: kernel BUG at /home/cel/src/linux/linux-2.6/mm/slub.c:3851!
kernel: invalid opcode: 0000 [#1] SMP
kernel: Modules linked in: cts rpcsec_gss_krb5 sb_edac edac_core x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm irqbypass crct10dif_pclmul crc32_pclmul ghash_clmulni_intel aesni_intel lrw gf128mul glue_helper ablk_helper cryptd btrfs xor iTCO_wdt iTCO_vendor_support raid6_pq pcspkr i2c_i801 i2c_smbus lpc_ich mfd_core mei_me sg mei shpchp wmi ioatdma ipmi_si ipmi_msghandler acpi_pad acpi_power_meter rpcrdma ib_ipoib rdma_ucm ib_ucm ib_uverbs ib_umad rdma_cm ib_cm iw_cm nfsd nfs_acl lockd grace auth_rpcgss sunrpc ip_tables xfs libcrc32c mlx4_ib mlx4_en ib_core sr_mod cdrom sd_mod ast drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ttm drm crc32c_intel igb mlx4_core ahci libahci libata ptp pps_core dca i2c_algo_bit i2c_core dm_mirror dm_region_hash dm_log dm_mod
kernel: CPU: 7 PID: 145 Comm: kworker/7:2 Not tainted 4.8.0-rc4-00006-g9d06b0b #15
kernel: Hardware name: Supermicro Super Server/X10SRL-F, BIOS 1.0c 09/09/2015
kernel: Workqueue: events do_cache_clean [sunrpc]
kernel: task: ffff8808541d8000 task.stack: ffff880854344000
kernel: RIP: 0010:[<ffffffff811e7075>]  [<ffffffff811e7075>] kfree+0x155/0x180
kernel: RSP: 0018:ffff880854347d70  EFLAGS: 00010246
kernel: RAX: ffffea0020fe7660 RBX: ffff88083f9db064 RCX: 146ff0f9d5ec5600
kernel: RDX: 000077ff80000000 RSI: ffff880853f01500 RDI: ffff88083f9db064
kernel: RBP: ffff880854347d88 R08: ffff8808594ee000 R09: ffff88087fdd8780
kernel: R10: 0000000000000000 R11: ffffea0020fe76c0 R12: ffff880853f01500
kernel: R13: ffffffffa013cf76 R14: ffffffffa013cff0 R15: ffffffffa04253a0
kernel: FS:  0000000000000000(0000) GS:ffff88087fdc0000(0000) knlGS:0000000000000000
kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
kernel: CR2: 00007fed60b020c3 CR3: 0000000001c06000 CR4: 00000000001406e0
kernel: Stack:
kernel: ffff8808589f2f00 ffff880853f01500 0000000000000001 ffff880854347da0
kernel: ffffffffa013cf76 ffff8808589f2f00 ffff880854347db8 ffffffffa013d006
kernel: ffff8808589f2f20 ffff880854347e00 ffffffffa0406f60 0000000057c7044f
kernel: Call Trace:
kernel: [<ffffffffa013cf76>] rsc_free+0x16/0x90 [auth_rpcgss]
kernel: [<ffffffffa013d006>] rsc_put+0x16/0x30 [auth_rpcgss]
kernel: [<ffffffffa0406f60>] cache_clean+0x2e0/0x300 [sunrpc]
kernel: [<ffffffffa04073ee>] do_cache_clean+0xe/0x70 [sunrpc]
kernel: [<ffffffff8109a70f>] process_one_work+0x1ff/0x3b0
kernel: [<ffffffff8109b15c>] worker_thread+0x2bc/0x4a0
kernel: [<ffffffff8109aea0>] ? rescuer_thread+0x3a0/0x3a0
kernel: [<ffffffff810a0ba4>] kthread+0xe4/0xf0
kernel: [<ffffffff8169c47f>] ret_from_fork+0x1f/0x40
kernel: [<ffffffff810a0ac0>] ? kthread_stop+0x110/0x110
kernel: Code: f7 ff ff eb 3b 65 8b 05 da 30 e2 7e 89 c0 48 0f a3 05 a0 38 b8 00 0f 92 c0 84 c0 0f 85 d1 fe ff ff 0f 1f 44 00 00 e9 f5 fe ff ff <0f> 0b 49 8b 03 31 f6 f6 c4 40 0f 85 62 ff ff ff e9 61 ff ff ff
kernel: RIP  [<ffffffff811e7075>] kfree+0x155/0x180
kernel: RSP <ffff880854347d70>
kernel: ---[ end trace 3fdec044969def26 ]---

It seems to be most common after a server reboot where a client has been
using a Kerberos mount, and reconnects to continue its workload.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Cc: stable@vger.kernel.org
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>

Previously, without GSO, it was easy to identify it: if the chunk didn't
fit and there was no data chunk in the packet yet, we could fragment at
IP level. So if there was an auth chunk and we were bundling a big data
chunk, it would fragment regardless of the size of the auth chunk. This
also works for the context of PMTU reductions.

But with GSO, we cannot distinguish such PMTU events anymore, as the
packet is allowed to exceed PMTU.

So we need another check: to ensure that the chunk that we are adding,
actually fits the current PMTU. If it doesn't, trigger a flush and let
it be fragmented at IP level in the next round.
Signed-off-by: NMarcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit 76174004
(tcp: do not slow start when cwnd equals ssthresh )
introduced regression in TCP YeAH. Using 100ms delay 1% loss virtual
ethernet link kernel 4.2 shows bandwidth ~500KB/s for single TCP
connection and kernel 4.3 and above (including 4.8-rc4) shows bandwidth
~100KB/s.
   That is caused by stalled cwnd when cwnd equals ssthresh. This patch
fixes it by proper increasing cwnd in this case.
Signed-off-by: NArtem Germanov <agermanov@anchorfree.com>
Acked-by: NDmitry Adamushko <d.adamushko@anchorfree.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When DATA and/or FIN are carried in a SYN/ACK message or SYN message,
we append an skb in socket receive queue, but we forget to call
sk_forced_mem_schedule().

Effect is that the socket has a negative sk->sk_forward_alloc as long as
the message is not read by the application.

Josh Hunt fixed a similar issue in commit d22e1537 ("tcp: fix tcp
fin memory accounting")

Fixes: 168a8f58 ("tcp: TCP Fast Open Server - main code path")
Signed-off-by: NEric Dumazet <edumazet@google.com>
Reviewed-by: NJosh Hunt <johunt@akamai.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In general, when DAD detected IPv6 duplicate address, ifp->state
will be set to INET6_IFADDR_STATE_ERRDAD and DAD is stopped by a
delayed work, the call tree should be like this:

ndisc_recv_ns
  -> addrconf_dad_failure        <- missing ifp put
     -> addrconf_mod_dad_work
       -> schedule addrconf_dad_work()
         -> addrconf_dad_stop()  <- missing ifp hold before call it

addrconf_dad_failure() called with ifp refcont holding but not put.
addrconf_dad_work() call addrconf_dad_stop() without extra holding
refcount. This will not cause any issue normally.

But the race between addrconf_dad_failure() and addrconf_dad_work()
may cause ifp refcount leak and netdevice can not be unregister,
dmesg show the following messages:

IPv6: eth0: IPv6 duplicate address fe80::XX:XXXX:XXXX:XX detected!
...
unregister_netdevice: waiting for eth0 to become free. Usage count = 1

Cc: stable@vger.kernel.org
Fixes: c15b1cca ("ipv6: move DAD and addrconf_verify processing
to workqueue")
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When deleting an IP address from an interface, there is a clean-up of
routes which refer to this local address. However, there was no check to
see that the VRF matched. This meant that deletion wasn't confined to
the VRF it should have been.

To solve this, a new field has been added to fib_info to hold a table
id. When removing fib entries corresponding to a local ip address, this
table id is also used in the comparison.

The table id is populated when the fib_info is created. This was already
done in some places, but not in ip_rt_ioctl(). This has now been fixed.

Fixes: 021dd3b8 ("net: Add routes to the table associated with the device")
Acked-by: NDavid Ahern <dsa@cumulusnetworks.com>
Tested-by: NDavid Ahern <dsa@cumulusnetworks.com>
Signed-off-by: NMark Tomlinson <mark.tomlinson@alliedtelesis.co.nz>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

An RPC can terminate before its reply arrives, if a credential
problem or a soft timeout occurs. After this happens, xprtrdma
reports it is out of Receive buffers.

A Receive buffer is posted before each RPC is sent, and returned to
the buffer pool when a reply is received. If no reply is received
for an RPC, that Receive buffer remains posted. But xprtrdma tries
to post another when the next RPC is sent.

If this happens a few dozen times, there are no receive buffers left
to be posted at send time. I don't see a way for a transport
connection to recover at that point, and it will spit warnings and
unnecessarily delay RPCs on occasion for its remaining lifetime.

Commit 1e465fd4 ("xprtrdma: Replace send and receive arrays")
removed a little bit of logic to detect this case and not provide
a Receive buffer so no more buffers are posted, and then transport
operation continues correctly. We didn't understand what that logic
did, and it wasn't commented, so it was removed as part of the
overhaul to support backchannel requests.

Restore it, but be wary of the need to keep extra Receives posted
to deal with backchannel requests.

Fixes: 1e465fd4 ("xprtrdma: Replace send and receive arrays")
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: NTrond Myklebust <trond.myklebust@primarydata.com>

Receive buffer exhaustion, if it were to actually occur, would be
catastrophic. However, when there are no reply buffers to post, that
means all of them have already been posted and are waiting for
incoming replies. By design, there can never be more RPCs in flight
than there are available receive buffers.

A receive buffer can be left posted after an RPC exits without a
received reply; say, due to a credential problem or a soft timeout.
This does not result in fewer posted receive buffers than there are
pending RPCs, and there is already logic in xprtrdma to deal
appropriately with this case.

It also looks like the "+ 2" that was removed was accidentally
accommodating the number of extra receive buffers needed for
receiving backchannel requests. That will need to be addressed by
another patch.

Fixes: 3d4cf35b ("xprtrdma: Reply buffer exhaustion can be...")
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Reviewed-by: NAnna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: NTrond Myklebust <trond.myklebust@primarydata.com>

Neither the failure or success paths of ping_v6_sendmsg release
the dst it acquires.  This leads to a flood of warnings from
"net/core/dst.c:288 dst_release" on older kernels that
don't have 8bf4ada2 backported.

That patch optimistically hoped this had been fixed post 3.10, but
it seems at least one case wasn't, where I've seen this triggered
a lot from machines doing unprivileged icmp sockets.

Cc: Martin Lau <kafai@fb.com>
Signed-off-by: NDave Jones <davej@codemonkey.org.uk>
Acked-by: NMartin KaFai Lau <kafai@fb.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Right now we use the 'readlock' both for protecting some of the af_unix
IO path and for making the bind be single-threaded.

The two are independent, but using the same lock makes for a nasty
deadlock due to ordering with regards to filesystem locking.  The bind
locking would want to nest outside the VSF pathname locking, but the IO
locking wants to nest inside some of those same locks.

We tried to fix this earlier with commit c845acb3 ("af_unix: Fix
splice-bind deadlock") which moved the readlock inside the vfs locks,
but that caused problems with overlayfs that will then call back into
filesystem routines that take the lock in the wrong order anyway.

Splitting the locks means that we can go back to having the bind lock be
the outermost lock, and we don't have any deadlocks with lock ordering.
Acked-by: NRainer Weikusat <rweikusat@cyberadapt.com>
Acked-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
Acked-by: NHannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This reverts commit c845acb3.

It turns out that it just replaces one deadlock with another one: we can
still get the wrong lock ordering with the readlock due to overlayfs
calling back into the filesystem layer and still taking the vfs locks
after the readlock.

The proper solution ends up being to just split the readlock into two
pieces: the bind lock (taken *outside* the vfs locks) and the IO lock
(taken *inside* the filesystem locks).  The two locks are independent
anyway.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
Reviewed-by: NShmulik Ladkani <shmulik.ladkani@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Following few steps will crash kernel -

  (a) Create bonding master
      > modprobe bonding miimon=50
  (b) Create macvlan bridge on eth2
      > ip link add link eth2 dev mvl0 address aa:0:0:0:0:01 \
	   type macvlan
  (c) Now try adding eth2 into the bond
      > echo +eth2 > /sys/class/net/bond0/bonding/slaves
      <crash>

Bonding does lots of things before checking if the device enslaved is
busy or not.

In this case when the notifier call-chain sends notifications, the
bond_netdev_event() assumes that the rx_handler /rx_handler_data is
registered while the bond_enslave() hasn't progressed far enough to
register rx_handler for the new slave.

This patch adds a rx_handler check that can be performed right at the
beginning of the enslave code to avoid getting into this situation.
Signed-off-by: NMahesh Bandewar <maheshb@google.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The commit f9b2ee71 ("SUNRPC: Move UDP receive data path
into a workqueue context"), as a side effect, moved the
skb_free_datagram() call outside the scope of the related socket
lock, but UDP sockets require such lock to be held for proper
memory accounting.
Fix it by replacing skb_free_datagram() with
skb_free_datagram_locked().

Fixes: f9b2ee71 ("SUNRPC: Move UDP receive data path into a workqueue context")
Reported-and-tested-by: NJan Stancek <jstancek@redhat.com>
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Cc: stable@vger.kernel.org # 4.4+
Signed-off-by: NTrond Myklebust <trond.myklebust@primarydata.com>

Tunnel deletion is delayed by both a workqueue (l2tp_tunnel_delete -> wq
 -> l2tp_tunnel_del_work) and RCU (sk_destruct -> RCU ->
l2tp_tunnel_destruct).

By the time l2tp_tunnel_destruct() runs to destroy the tunnel and finish
destroying the socket, the private data reserved via the net_generic
mechanism has already been freed, but l2tp_tunnel_destruct() actually
uses this data.

Make sure tunnel deletion for the netns has completed before returning
from l2tp_exit_net() by first flushing the tunnel removal workqueue, and
then waiting for RCU callbacks to complete.

Fixes: 167eb17e ("l2tp: create tunnel sockets in the right namespace")
Signed-off-by: NSabrina Dubroca <sd@queasysnail.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit 8eb30be0 ("ipv6: Create ip6_tnl_xmit") unsets
flowi6_proto in ip4ip6_tnl_xmit() and ip6ip6_tnl_xmit().
Since xfrm_selector_match() relies on this info, IPv6 packets
sent by an ip6tunnel cannot be properly selected by their
protocols after removing it. This patch puts flowi6_proto back.

Cc: stable@vger.kernel.org
Fixes: 8eb30be0 ("ipv6: Create ip6_tnl_xmit")
Signed-off-by: NEli Cooper <elicooper@gmx.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The original codes depend on that the function parameters are evaluated from
left to right. But the parameter's evaluation order is not defined in C
standard actually.

When flow_keys_have_l4(&keys) is invoked before ___skb_get_hash(skb, &keys,
hashrnd) with some compilers or environment, the keys passed to
flow_keys_have_l4 is not initialized.

Fixes: 6db61d79 ("flow_dissector: Ignore flow dissector return value from ___skb_get_hash")
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NGao Feng <fgao@ikuai8.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Yuchung noticed that on the first TFO server data packet sent after
the (TFO) handshake, the server echoed the TCP timestamp value in the
SYN/data instead of the timestamp value in the final ACK of the
handshake. This problem did not happen on regular opens.

The tcp_replace_ts_recent() logic that decides whether to remember an
incoming TS value needs tp->rcv_wup to hold the latest receive
sequence number that we have ACKed (latest tp->rcv_nxt we have
ACKed). This commit fixes this issue by ensuring that a TFO server
properly updates tp->rcv_wup to match tp->rcv_nxt at the time it sends
a SYN/ACK for the SYN/data.
Reported-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NNeal Cardwell <ncardwell@google.com>
Signed-off-by: NYuchung Cheng <ycheng@google.com>
Signed-off-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NSoheil Hassas Yeganeh <soheil@google.com>
Fixes: 168a8f58 ("tcp: TCP Fast Open Server - main code path")
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

pskb_may_pull may fail due to various reasons (e.g. alloc failure), but the
skb isn't changed/dropped and processing continues so we shouldn't
increment tx_dropped.

CC: Kyeyoon Park <kyeyoonp@codeaurora.org>
CC: Roopa Prabhu <roopa@cumulusnetworks.com>
CC: Stephen Hemminger <stephen@networkplumber.org>
CC: bridge@lists.linux-foundation.org
Fixes: 95850116 ("bridge: Add support for IEEE 802.11 Proxy ARP")
Signed-off-by: NNikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>