No error message is currently printed if the pin syscall
itself fails. It got lost in the loadall refactoring.

Fixes: 77380998 ("bpftool: add loadall command")
Reported-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Acked-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>

Change an error message to work for any object being
pinned not just programs.

Fixes: 71bb428f ("tools: bpf: add bpftool")
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>

Add 2 tests that check JIT code generation to jumps to 1st insn.
1st test is similar to syzbot reproducer.
The backwards branch is never taken at runtime.
2nd test has branch to 1st insn that executes.
The test is written as two bpf functions, since it's not possible
to construct valid single bpf program that jumps to 1st insn.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Song Liu <songliubraving@fb.com>

Introduction of bounded loops exposed old bug in x64 JIT.
JIT maintains the array of offsets to the end of all instructions to
compute jmp offsets.
addrs[0] - offset of the end of the 1st insn (that includes prologue).
addrs[1] - offset of the end of the 2nd insn.
JIT didn't keep the offset of the beginning of the 1st insn,
since classic BPF didn't have backward jumps and valid extended BPF
couldn't have a branch to 1st insn, because it didn't allow loops.
With bounded loops it's possible to construct a valid program that
jumps backwards to the 1st insn.
Fix JIT by computing:
addrs[0] - offset of the end of prologue == start of the 1st insn.
addrs[1] - offset of the end of 1st insn.

v1->v2:
- Yonghong noticed a bug in jit linfo.
  Fix it by passing 'addrs + 1' to bpf_prog_fill_jited_linfo(),
  since it expects insn_to_jit_off array to be offsets to last byte.

Reported-by: syzbot+35101610ff3e83119b1b@syzkaller.appspotmail.com
Fixes: 2589726d ("bpf: introduce bounded loops")
Fixes: 0a14842f ("net: filter: Just In Time compiler for x86-64")
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Song Liu <songliubraving@fb.com>

5d01ab7b ("libbpf: fix erroneous multi-closing of BTF FD")
introduced backwards-compatibility issue, manifesting itself as -E2BIG
error returned on program load due to unknown non-zero btf_fd attribute
value for BPF_PROG_LOAD sys_bpf() sub-command.

This patch fixes bug by ensuring that we only ever associate BTF FD with
program if there is a BTF.ext data that was successfully loaded into
kernel, which automatically means kernel supports func_info/line_info
and associated BTF FD for progs (checked and ensured also by BTF
sanitization code).

Fixes: 5d01ab7b ("libbpf: fix erroneous multi-closing of BTF FD")
Reported-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Having static variable `cpus` in libbpf_num_possible_cpus function
without guarding it with mutex makes this function thread-unsafe.

If multiple threads accessing this function, in the current form; it
leads to incrementing the static variable value `cpus` in the multiple
of total available CPUs.

Used local stack variable to calculate the number of possible CPUs and
then updated the static variable using WRITE_ONCE().

Changes since v1:
 * added stack variable to calculate cpus
 * serialized static variable update using WRITE_ONCE()
 * fixed Fixes tag

Fixes: 6446b315 ("bpf: add a new API libbpf_num_possible_cpus()")
Signed-off-by: Takshak Chahande <ctakshak@fb.com>
Acked-by: Andrey Ignatov <rdna@fb.com>
Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Libbpf stores associated BTF FD per each instance of bpf_program. When
program is unloaded, that FD is closed. This is wrong, because leads to
a race and possibly closing of unrelated files, if application
simultaneously opens new files while bpf_programs are unloaded.

It's also unnecessary, because struct btf "owns" that FD, and
btf__free(), called from bpf_object__close() will close it. Thus the fix
is to never have per-program BTF FD and fetch it from obj->btf, when
necessary.

Fixes: 2993e051 ("tools/bpf: add support to read .BTF.ext sections")
Reported-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

Saeed Mahameed says:

====================
Mellanox, mlx5 fixes 2019-07-25

This series introduces some fixes to mlx5 driver.

1) Ariel is addressing an issue with enacp flow counter race condition
2) Aya fixes ethtool speed handling
3) Edward fixes modify_cq hw bits alignment
4) Maor fixes RDMA_RX capabilities handling
5) Mark reverses unregister devices order to address an issue with LAG
6) From Tariq,
  - wrong max num channels indication regression
  - TLS counters naming and documentation as suggested by Jakub
  - kTLS, Call WARN_ONCE on netdev mismatch

There is one patch in this series that touches nfp driver to align
TLS statistics names with latest documentation, Jakub is CC'ed.

Please pull and let me know if there is any problem.

For -stable v4.9:
  ('net/mlx5: Use reversed order when unregister devices')

For -stable v4.20
  ('net/mlx5e: Prevent encap flow counter update async to user query')
  ('net/mlx5: Fix modify_cq_in alignment')

For -stable v5.1
  ('net/mlx5e: Fix matching of speed to PRM link modes')

For -stable v5.2
  ('net/mlx5: Add missing RDMA_RX capabilities')
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

The udp_ip4_ind bit is set only for IPv4 UDP non-fragmented packets
so that the hardware can flip the checksum to 0xFFFF if the computed
checksum is 0 per RFC768.

However, this bit had to be set for IPv6 UDP non fragmented packets
as well per hardware requirements. Otherwise, IPv6 UDP packets
with computed checksum as 0 were transmitted by hardware and were
dropped in the network.

In addition to setting this bit for IPv6 UDP, the field is also
appropriately renamed to udp_ind as part of this change.

Fixes: 5eb5f860 ("net: qualcomm: rmnet: Add support for TX checksum offload")
Cc: Sean Tranchetti <stranche@codeaurora.org>
Signed-off-by: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

ip4ip6/ip6ip6 tunnels run iptunnel_handle_offloads on xmit which
can cause a possible use-after-free accessing iph/ipv6h pointer
since the packet will be 'uncloned' running pskb_expand_head if
it is a cloned gso skb.

Fixes: 0e9a7095 ("ip6_tunnel, ip6_gre: fix setting of DSCP on encapsulated packets")
Signed-off-by: Haishuang Yan <yanhaishuang@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Make sure the delayed work for stats update is not pending before
wq destruction.
This fixes the module unload path.
The issue is there since day 1.

Fixes: a556c76a ("net: mscc: Add initial Ocelot switch support")
Signed-off-by: Claudiu Manoil <claudiu.manoil@nxp.com>
Reviewed-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The hw_ver field was initialized to zero. Return the chip revision.
This is relevant for rdma driver.
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The BroadMobi BM818 M.2 card uses the QMI protocol
Signed-off-by: Bob Ham <bob.ham@puri.sm>
Signed-off-by: Angus Ainslie (Purism) <angus@akkea.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>

Alexei Starovoitov says:

====================
pull-request: bpf 2019-07-25

The following pull-request contains BPF updates for your *net* tree.

The main changes are:

1) fix segfault in libbpf, from Andrii.

2) fix gso_segs access, from Eric.

3) tls/sockmap fixes, from Jakub and John.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

We need the same checks introduced by commit cb9f1b78
("ip: validate header length on virtual device xmit") for
ipip tunnel.

Fixes: cb9f1b78 ("ip: validate header length on virtual device xmit")
Signed-off-by: Haishuang Yan <yanhaishuang@cmss.chinamobile.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

ipv6_flowlabel and ipv6_flowlabel_mgr are missing from
gitignore.  Quentin points out that the original
commit 3fb321fd ("selftests/net: ipv6 flowlabel")
did add ignore entries, they are just missing the "ipv6_"
prefix.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Commit 3968d389 ("bnx2x: Fix Multi-Cos.") which enabled multi-cos
feature after prolonged time in driver added some regression causing
numerous issues (sudden reboots, tx timeout etc.) reported by customers.
We plan to backout this commit and submit proper fix once we have root
cause of issues reported with this feature enabled.

Fixes: 3968d389 ("bnx2x: Fix Multi-Cos.")
Signed-off-by: Sudarsana Reddy Kalluru <skalluru@marvell.com>
Signed-off-by: Manish Chopra <manishc@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Add missing description of counters.
Split tx_tls_encrypted counter into two, to give packets
and bytes indications.

Fixes: f42c104f ("Documentation: add TLS offload documentation")
Suggested-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Align to the naming convention in TLS documentation.

Fixes: 51a5e563 ("nfp: tls: add basic statistics")
Suggested-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

A netdev mismatch in the processed TLS SKB should not occur,
and indicates a kernel bug.
Add WARN_ONCE to spot such cases.

Fixes: d2ead1f3 ("net/mlx5e: Add kTLS TX HW offload support")
Suggested-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

This patch prevents a race between user invoked cached counters
query and a neighbor last usage updater.

The cached flow counter stats can be queried by calling
"mlx5_fc_query_cached" which provides the number of bytes and
packets that passed via this flow since the last time this counter
was queried.
It does so by reducting the last saved stats from the current, cached
stats and then updating the last saved stats with the cached stats.
It also provide the lastuse value for that flow.

Since "mlx5e_tc_update_neigh_used_value" needs to retrieve the
last usage time of encapsulation flows, it calls the flow counter
query method periodically and async to user queries of the flow counter
using cls_flower.
This call is causing the driver to update the last reported bytes and
packets from the cache and therefore, future user queries of the flow
stats will return lower than expected number for bytes and packets
since the last saved stats in the driver was updated async to the last
saved stats in cls_flower.

This causes wrong stats presentation of encapsulation flows to user.

Since the neighbor usage updater only needs the lastuse stats from the
cached counter, the fix is to use a dedicated lastuse query call that
returns the lastuse value without synching between the cached stats and
the last saved stats.

Fixes: f6dfb4c3 ("net/mlx5e: Update neighbour 'used' state using HW flow rules counters")
Signed-off-by: Ariel Levkovich <lariel@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Speed translation is performed based on legacy or extended PTYS
register. Translate speed with respect to:
1) Capability bit of extended PTYS table.
2) User request:
 a) When auto-negotiation is turned on, inspect advertisement whether it
 contains extended link modes.
 b) When auto-negotiation is turned off, speed > 100Gbps (maximal
 speed supported in legacy mode).
With both conditions fulfilled translation is done with extended PTYS
table otherwise use legacy PTYS table.
Without this patch 25/50/100 Gbps speed cannot be set, since try to
configure in extended mode but read from legacy mode.

Fixes: dd1b9e09 ("net/mlx5: ethtool, Allow legacy link-modes configuration via non-extended ptys")
Signed-off-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

No XSK support in the enhanced IPoIB driver and representors.
Add a profile property to specify this, and enhance the logic
that calculates the max number of channels to take it into
account.

Fixes: db05815b ("net/mlx5e: Add XSK zero-copy support")
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Fix modify_cq_in alignment to match the device specification.
After this fix the 'cq_umem_valid' field will be in the right offset.

Cc: <stable@vger.kernel.org> # 4.19
Fixes: bd371975 ("net/mlx5: Update mlx5_ifc with DEVX UID bits")
Signed-off-by: Edward Srouji <edwards@mellanox.com>
Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

New flow table type RDMA_RX was added but the MLX5_CAP_FLOW_TABLE_TYPE
didn't handle this new flow table type.
This means that MLX5_CAP_FLOW_TABLE_TYPE returns an empty capability to
this flow table type.

Update both the macro and the maximum supported flow table type to
RDMA_RX.

Fixes: d83eb50e ("net/mlx5: Add support in RDMA RX steering")
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

When lag is active, which is controlled by the bonded mlx5e netdev, mlx5
interface unregestering must happen in the reverse order where rdma is
unregistered (unloaded) first, to guarantee all references to the lag
context in hardware is removed, then remove mlx5e netdev interface which
will cleanup the lag context from hardware.

Without this fix during destroy of LAG interface, we observed following
errors:
 * mlx5_cmd_check:752:(pid 12556): DESTROY_LAG(0x843) op_mod(0x0) failed,
   status bad parameter(0x3), syndrome (0xe4ac33)
 * mlx5_cmd_check:752:(pid 12556): DESTROY_LAG(0x843) op_mod(0x0) failed,
   status bad parameter(0x3), syndrome (0xa5aee8).

Fixes: a31208b1 ("net/mlx5_core: New init and exit flow for mlx5_core")
Reviewed-by: Parav Pandit <parav@mellanox.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Mark Zhang <markz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

fm_set_max_frm() existed in the Freescale SDK as a callback for an
early_param. When this code was ported to the upstream kernel the
early_param was converted to a module_param making the reference to the
function incorrect. The rest of the comment already does a good job of
explaining the parameter so removing the reference to the non-existent
function seems like the best thing to do.
Signed-off-by: Chris Packham <chris.packham@alliedtelesis.co.nz>
Signed-off-by: David S. Miller <davem@davemloft.net>

devm_kzalloc may fail and return NULL. So the null check is needed.
Signed-off-by: Navid Emamdoost <navid.emamdoost@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

devm_kzalloc may fail and return null. So the null check is needed.
Signed-off-by: Navid Emamdoost <navid.emamdoost@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

- v1 -> v2: Move skb_set_owner_w to __tun_build_skb to reduce patch size

Small packets going out of a tap device go through an optimized code
path that uses build_skb() rather than sock_alloc_send_pskb(). The
latter calls skb_set_owner_w(), but the small packet code path does not.

The net effect is that small packets are not owned by the userland
application's socket (e.g. QEMU), while large packets are.
This can be seen with a TCP session, where packets are not owned when
the window size is small enough (around PAGE_SIZE), while they are once
the window grows (note that this requires the host to support virtio
tso for the guest to offload segmentation).
All this leads to inconsistent behaviour in the kernel, especially on
netfilter modules that uses sk->socket (e.g. xt_owner).

Fixes: 66ccbc9c ("tap: use build_skb() for small packet")
Signed-off-by: Alexis Bauvin <abauvin@scaleway.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Leon Romanovsky says:

====================
DIM fixes for 5.3

Those two fixes for recently merged DIM patches, both exposed through
RDMa DIM usage.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

DIM causes to the following warnings during kernel compilation
which indicates that tx_profile and rx_profile are supposed to
be declared in *.c and not in *.h files.

In file included from ./include/rdma/ib_verbs.h:64,
                 from ./include/linux/mlx5/device.h:37,
                 from ./include/linux/mlx5/driver.h:51,
                 from ./include/linux/mlx5/vport.h:36,
                 from drivers/infiniband/hw/mlx5/ib_virt.c:34:
./include/linux/dim.h:326:1: warning: _tx_profile_ defined but not used [-Wunused-const-variable=]
  326 | tx_profile[DIM_CQ_PERIOD_NUM_MODES][NET_DIM_PARAMS_NUM_PROFILES] = {
      | ^~~~~~~~~~
./include/linux/dim.h:320:1: warning: _rx_profile_ defined but not used [-Wunused-const-variable=]
  320 | rx_profile[DIM_CQ_PERIOD_NUM_MODES][NET_DIM_PARAMS_NUM_PROFILES] = {
      | ^~~~~~~~~~

Fixes: 4f75da36 ("linux/dim: Move implementation to .c files")
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

While using net_dim, a dim_sample was used without ever initializing the
comps value. Added use of DIV_ROUND_DOWN_ULL() to prevent potential
overflow, it should not be a problem to save the final result in an int
because after the division by epms the value should not be larger than a
few thousand.

[ 1040.127124] UBSAN: Undefined behaviour in lib/dim/dim.c:78:23
[ 1040.130118] signed integer overflow:
[ 1040.131643] 134718714 * 100 cannot be represented in type 'int'

Fixes: 398c2b05 ("linux/dim: Add completions count to dim_sample")
Signed-off-by: Yamin Friedman <yaminf@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

act_ife at least requires TCA_IFE_PARMS, so we have to bail out
when there is no attribute passed in.

Reported-by: syzbot+fbb5b288c9cb6a2eeac4@syzkaller.appspotmail.com
Fixes: ef6980b6 ("introduce IFE action")
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Despite a proper NULL-termination after strncpy(..., ..., IFNAMSIZ - 1),
GCC8 still complains about *expected* string truncation:

  xsk.c:330:2: error: 'strncpy' output may be truncated copying 15 bytes
  from a string of length 15 [-Werror=stringop-truncation]
    strncpy(ifr.ifr_name, xsk->ifname, IFNAMSIZ - 1);

This patch gets rid of the issue altogether by using memcpy instead.
There is no performance regression, as strncpy will still copy and fill
all of the bytes anyway.

v1->v2:
- rebase against bpf tree.

Cc: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
Acked-by: Magnus Karlsson <magnus.karlsson@intel.com>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

sock_efree() releases the sock refcnt, if we don't hold this refcnt
when setting skb->destructor to it, the refcnt would not be balanced.
This leads to several bug reports from syzbot.

I have checked other users of sock_efree(), all of them hold the
sock refcnt.

Fixes: c8c8218e ("netrom: fix a memory leak in nr_rx_frame()")
Reported-and-tested-by: <syzbot+622bdabb128acc33427d@syzkaller.appspotmail.com>
Reported-and-tested-by: <syzbot+6eaef7158b19e3fec3a0@syzkaller.appspotmail.com>
Reported-and-tested-by: <syzbot+9399c158fcc09b21d0d2@syzkaller.appspotmail.com>
Reported-and-tested-by: <syzbot+a34e5f3d0300163f0c87@syzkaller.appspotmail.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Some functions in the datapath code are factored out so that each
one has a stack frame smaller than 1024 bytes with gcc. However,
when compiling with clang, the functions are inlined more aggressively
and combined again so we get

net/openvswitch/datapath.c:1124:12: error: stack frame size of 1528 bytes in function 'ovs_flow_cmd_set' [-Werror,-Wframe-larger-than=]

Marking both get_flow_actions() and ovs_nla_init_match_and_action()
as 'noinline_for_stack' gives us the same behavior that we see with
gcc, and no warning. Note that this does not mean we actually use
less stack, as the functions call each other, and we still get
three copies of the large 'struct sw_flow_key' type on the stack.

The comment tells us that this was previously considered safe,
presumably since the netlink parsing functions are called with
a known backchain that does not also use a lot of stack space.

Fixes: 9cc9a5cb ("datapath: Avoid using stack larger than 1024.")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

I've been spending quite a bit of time fixing and
preventing bit rot in the core TLS code. TLS seems
to only be growing in importance, I'd like to help
ensuring the quality of our implementation.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Simon Horman <simon.horman@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The memory allocated for the stats array may contain arbitrary data.

Fixes: e4f9ba64 ("net: phy: mscc: add support for VSC8514 PHY.")
Fixes: 00d70d8e ("net: phy: mscc: add support for VSC8574 PHY")
Fixes: a5afc167 ("net: phy: mscc: add support for VSC8584 PHY")
Fixes: f76178dc ("net: phy: mscc: add ethtool statistics counters")
Signed-off-by: Andreas Schwab <schwab@suse.de>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>

SFP modules connected using the SGMII interface have their own PHYs which
are handled by the struct phylink's phydev field. On the other hand, for
the modules connected using 1000Base-X interface that field is not set.

Since commit ce0aa27f ("sfp: add sfp-bus to bridge between network
devices and sfp cages") phylink_start() ends up setting the phydev field
using the sfp-bus infrastructure, which eventually calls phy_start() on it,
and then calling phy_start() again on the same phydev from phylink_start()
itself. Similar call sequence holds for phylink_stop(), only in the reverse
order. This results in WARNs during network interface bringup and shutdown
when a copper SFP module is connected, as phy_start() and phy_stop() are
called twice in a row for the same phy_device:

  % ip link set up dev eth0
  ------------[ cut here ]------------
  called from state UP
  WARNING: CPU: 1 PID: 155 at drivers/net/phy/phy.c:895 phy_start+0x74/0xc0
  Modules linked in:
  CPU: 1 PID: 155 Comm: backend Not tainted 5.2.0+ #1
  NIP:  c0227bf0 LR: c0227bf0 CTR: c004d224
  REGS: df547720 TRAP: 0700   Not tainted  (5.2.0+)
  MSR:  00029000 <CE,EE,ME>  CR: 24002822  XER: 00000000

  GPR00: c0227bf0 df5477d8 df5d7080 00000014 df9d2370 df9d5ac4 1f4eb000 00000001
  GPR08: c061fe58 00000000 00000000 df5477d8 0000003c 100c8768 00000000 00000000
  GPR16: df486a00 c046f1c8 c046eea0 00000000 c046e904 c0239604 db68449c 00000000
  GPR24: e9083204 00000000 00000001 db684460 e9083404 00000000 db6dce00 db6dcc00
  NIP [c0227bf0] phy_start+0x74/0xc0
  LR [c0227bf0] phy_start+0x74/0xc0
  Call Trace:
  [df5477d8] [c0227bf0] phy_start+0x74/0xc0 (unreliable)
  [df5477e8] [c023cad0] startup_gfar+0x398/0x3f4
  [df547828] [c023cf08] gfar_enet_open+0x364/0x374
  [df547898] [c029d870] __dev_open+0xe4/0x140
  [df5478c8] [c029db70] __dev_change_flags+0xf0/0x188
  [df5478f8] [c029dc28] dev_change_flags+0x20/0x54
  [df547918] [c02ae304] do_setlink+0x310/0x818
  [df547a08] [c02b1eb8] __rtnl_newlink+0x384/0x6b0
  [df547c28] [c02b222c] rtnl_newlink+0x48/0x68
  [df547c48] [c02ad7c8] rtnetlink_rcv_msg+0x240/0x27c
  [df547c98] [c02cc068] netlink_rcv_skb+0x8c/0xf0
  [df547cd8] [c02cba3c] netlink_unicast+0x114/0x19c
  [df547d08] [c02cbd74] netlink_sendmsg+0x2b0/0x2c0
  [df547d58] [c027b668] sock_sendmsg_nosec+0x20/0x40
  [df547d68] [c027d080] ___sys_sendmsg+0x17c/0x1dc
  [df547e98] [c027df7c] __sys_sendmsg+0x68/0x84
  [df547ef8] [c027e430] sys_socketcall+0x1a0/0x204
  [df547f38] [c000d1d8] ret_from_syscall+0x0/0x38
  --- interrupt: c01 at 0xfd4e030
      LR = 0xfd4e010
  Instruction dump:
  813f0188 38800000 2b890005 419d0014 3d40c046 5529103a 394aa208 7c8a482e
  3c60c046 3863a1b8 4cc63182 4be009a1 <0fe00000> 48000030 3c60c046 3863a1d0
  ---[ end trace d4c095aeaf6ea998 ]---

and

  % ip link set down dev eth0
  ------------[ cut here ]------------
  called from state HALTED
  WARNING: CPU: 1 PID: 184 at drivers/net/phy/phy.c:858 phy_stop+0x3c/0x88

  <...>

  Call Trace:
  [df581788] [c0228450] phy_stop+0x3c/0x88 (unreliable)
  [df581798] [c022d548] sfp_sm_phy_detach+0x1c/0x44
  [df5817a8] [c022e8cc] sfp_sm_event+0x4b0/0x87c
  [df581848] [c022f04c] sfp_upstream_stop+0x34/0x44
  [df581858] [c0225608] phylink_stop+0x7c/0xe4
  [df581868] [c023c57c] stop_gfar+0x7c/0x94
  [df581888] [c023c5b8] gfar_close+0x24/0x94
  [df5818a8] [c0298688] __dev_close_many+0xdc/0xf8
  [df5818c8] [c029db58] __dev_change_flags+0xd8/0x188
  [df5818f8] [c029dc28] dev_change_flags+0x20/0x54
  [df581918] [c02ae304] do_setlink+0x310/0x818
  [df581a08] [c02b1eb8] __rtnl_newlink+0x384/0x6b0
  [df581c28] [c02b222c] rtnl_newlink+0x48/0x68
  [df581c48] [c02ad7c8] rtnetlink_rcv_msg+0x240/0x27c
  [df581c98] [c02cc068] netlink_rcv_skb+0x8c/0xf0
  [df581cd8] [c02cba3c] netlink_unicast+0x114/0x19c
  [df581d08] [c02cbd74] netlink_sendmsg+0x2b0/0x2c0
  [df581d58] [c027b668] sock_sendmsg_nosec+0x20/0x40
  [df581d68] [c027d080] ___sys_sendmsg+0x17c/0x1dc
  [df581e98] [c027df7c] __sys_sendmsg+0x68/0x84
  [df581ef8] [c027e430] sys_socketcall+0x1a0/0x204
  [df581f38] [c000d1d8] ret_from_syscall+0x0/0x38

  <...>

  ---[ end trace d4c095aeaf6ea999 ]---

SFP modules with the 1000Base-X interface are not affected.

Place explicit calls to phy_start() and phy_stop() before enabling or after
disabling an attached SFP module, where phydev is not yet set (or is
already unset), so they will be made only from the inside of sfp-bus, if
needed.

Fixes: 21796261 ("net: phy: warn if phy_start is called from invalid state")
Signed-off-by: Arseny Solokha <asolokha@kb.kras.ru>
Acked-by: Russell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>