Skip to content

L
linux
Switch branch/tag
  1. 22 Oct, 2020 5 commits
    • Michael S. Tsirkin's avatar
      Revert "virtio-net: ethtool configurable RXCSUM" · cf8691cb
      Michael S. Tsirkin authored 
      This reverts commit 3618ad2a.

When control vq is not negotiated, that commit causes a crash:

[   72.229171] kernel BUG at drivers/net/virtio_net.c:1667!
[   72.230266] invalid opcode: 0000 [#1] PREEMPT SMP
[   72.231172] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.9.0-rc8-02934-g3618ad2a #1
[   72.231172] EIP: virtnet_send_command+0x120/0x140
[   72.231172] Code: 00 0f 94 c0 8b 7d f0 65 33 3d 14 00 00 00 75 1c 8d 65 f4 5b 5e 5f 5d c3 66 90 be 01 00 00 00 e9 6e ff ff ff 8d b6 00
+00 00 00 <0f> 0b e8 d9 bb 82 00 eb 17 8d b4 26 00 00 00 00 8d b4 26 00 00 00
[   72.231172] EAX: 0000000d EBX: f72895c0 ECX: 00000017 EDX: 00000011
[   72.231172] ESI: f7197800 EDI: ed69bd00 EBP: ed69bcf4 ESP: ed69bc98
[   72.231172] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 EFLAGS: 00010246
[   72.231172] CR0: 80050033 CR2: 00000000 CR3: 02c84000 CR4: 000406f0
[   72.231172] Call Trace:
[   72.231172]  ? __virt_addr_valid+0x45/0x60
[   72.231172]  ? ___cache_free+0x51f/0x760
[   72.231172]  ? kobject_uevent_env+0xf4/0x560
[   72.231172]  virtnet_set_guest_offloads+0x4d/0x80
[   72.231172]  virtnet_set_features+0x85/0x120
[   72.231172]  ? virtnet_set_guest_offloads+0x80/0x80
[   72.231172]  __netdev_update_features+0x27a/0x8e0
[   72.231172]  ? kobject_uevent+0xa/0x20
[   72.231172]  ? netdev_register_kobject+0x12c/0x160
[   72.231172]  register_netdevice+0x4fe/0x740
[   72.231172]  register_netdev+0x1c/0x40
[   72.231172]  virtnet_probe+0x728/0xb60
[   72.231172]  ? _raw_spin_unlock+0x1d/0x40
[   72.231172]  ? virtio_vdpa_get_status+0x1c/0x20
[   72.231172]  virtio_dev_probe+0x1c6/0x271
[   72.231172]  really_probe+0x195/0x2e0
[   72.231172]  driver_probe_device+0x26/0x60
[   72.231172]  device_driver_attach+0x49/0x60
[   72.231172]  __driver_attach+0x46/0xc0
[   72.231172]  ? device_driver_attach+0x60/0x60
[   72.231172]  bus_add_driver+0x197/0x1c0
[   72.231172]  driver_register+0x66/0xc0
[   72.231172]  register_virtio_driver+0x1b/0x40
[   72.231172]  virtio_net_driver_init+0x61/0x86
[   72.231172]  ? veth_init+0x14/0x14
[   72.231172]  do_one_initcall+0x76/0x2e4
[   72.231172]  ? rdinit_setup+0x2a/0x2a
[   72.231172]  do_initcalls+0xb2/0xd5
[   72.231172]  kernel_init_freeable+0x14f/0x179
[   72.231172]  ? rest_init+0x100/0x100
[   72.231172]  kernel_init+0xd/0xe0
[   72.231172]  ret_from_fork+0x1c/0x30
[   72.231172] Modules linked in:
[   72.269563] ---[ end trace a6ebc4afea0e6cb1 ]---

The reason is that virtnet_set_features now calls virtnet_set_guest_offloads
unconditionally, it used to only call it when there is something
to configure.

If device does not have a control vq, everything breaks.

Revert the original commit for now.

Cc: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Fixes: 3618ad2a ("virtio-net: ethtool configurable RXCSUM")
Reported-by: default avatarkernel test robot <lkp@intel.com>
Signed-off-by: default avatarMichael S. Tsirkin <mst@redhat.com>
Acked-by: default avatarWillem de Bruijn <willemb@google.com>
Acked-by: default avatarJason Wang <jasowang@redhat.com>
Link: https://lore.kernel.org/r/20201021142944.13615-1-mst@redhat.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      cf8691cb
    • Di Zhu's avatar
      rtnetlink: fix data overflow in rtnl_calcit() · ebfe3c51
      Di Zhu authored 
      "ip addr show" command execute error when we have a physical
network card with a large number of VFs

The return value of if_nlmsg_size() in rtnl_calcit() will exceed
range of u16 data type when any network cards has a larger number of
VFs. rtnl_vfinfo_size() will significant increase needed dump size when
the value of num_vfs is larger.

Eventually we get a wrong value of min_ifinfo_dump_size because of overflow
which decides the memory size needed by netlink dump and netlink_dump()
will return -EMSGSIZE because of not enough memory was allocated.

So fix it by promoting  min_dump_alloc data type to u32 to
avoid whole netlink message size overflow and it's also align
with the data type of struct netlink_callback{}.min_dump_alloc
which is assigned by return value of rtnl_calcit()
Signed-off-by: default avatarDi Zhu <zhudi21@huawei.com>
Link: https://lore.kernel.org/r/20201021020053.1401-1-zhudi21@huawei.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      ebfe3c51
    • Bartosz Golaszewski's avatar
      net: ethernet: mtk-star-emac: select REGMAP_MMIO · 5403caf2
      Bartosz Golaszewski authored 
      The driver depends on mmio regmap API but doesn't select the appropriate
Kconfig option. This fixes it.

Fixes: 8c7bd5a4 ("net: ethernet: mtk-star-emac: new driver")
Signed-off-by: default avatarBartosz Golaszewski <bgolaszewski@baylibre.com>
Link: https://lore.kernel.org/r/20201020073515.22769-1-brgl@bgdev.plSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      5403caf2
    • Xie He's avatar
      net: hdlc_raw_eth: Clear the IFF_TX_SKB_SHARING flag after calling ether_setup · 5fce1e43
      Xie He authored 
      This driver calls ether_setup to set up the network device.
The ether_setup function would add the IFF_TX_SKB_SHARING flag to the
device. This flag indicates that it is safe to transmit shared skbs to
the device.

However, this is not true. This driver may pad the frame (in eth_tx)
before transmission, so the skb may be modified.

Fixes: 550fd08c ("net: Audit drivers to identify those needing IFF_TX_SKB_SHARING cleared")
Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: Krzysztof Halasa <khc@pm.waw.pl>
Signed-off-by: default avatarXie He <xie.he.0141@gmail.com>
Link: https://lore.kernel.org/r/20201020063420.187497-1-xie.he.0141@gmail.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      5fce1e43
    • Xie He's avatar
      net: hdlc: In hdlc_rcv, check to make sure dev is an HDLC device · 01c4ceae
      Xie He authored 
      The hdlc_rcv function is used as hdlc_packet_type.func to process any
skb received in the kernel with skb->protocol == htons(ETH_P_HDLC).
The purpose of this function is to provide second-stage processing for
skbs not assigned a "real" L3 skb->protocol value in the first stage.

This function assumes the device from which the skb is received is an
HDLC device (a device created by this module). It assumes that
netdev_priv(dev) returns a pointer to "struct hdlc_device".

However, it is possible that some driver in the kernel (not necessarily
in our control) submits a received skb with skb->protocol ==
htons(ETH_P_HDLC), from a non-HDLC device. In this case, the skb would
still be received by hdlc_rcv. This will cause problems.

hdlc_rcv should be able to recognize and drop invalid skbs. It should
first make sure "dev" is actually an HDLC device, before starting its
processing. This patch adds this check to hdlc_rcv.

Fixes: 1da177e4 ("Linux-2.6.12-rc2")
Cc: Krzysztof Halasa <khc@pm.waw.pl>
Signed-off-by: default avatarXie He <xie.he.0141@gmail.com>
Link: https://lore.kernel.org/r/20201020013152.89259-1-xie.he.0141@gmail.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      01c4ceae
  3. 21 Oct, 2020 16 commits
  5. 20 Oct, 2020 13 commits
  7. 19 Oct, 2020 1 commit
  9. 18 Oct, 2020 5 commits
    • Taehee Yoo's avatar
      net: core: use list_del_init() instead of list_del() in netdev_run_todo() · 0e8b8d6a
      Taehee Yoo authored 
      dev->unlink_list is reused unless dev is deleted.
So, list_del() should not be used.
Due to using list_del(), dev->unlink_list can't be reused so that
dev->nested_level update logic doesn't work.
In order to fix this bug, list_del_init() should be used instead
of list_del().

Test commands:
    ip link add bond0 type bond
    ip link add bond1 type bond
    ip link set bond0 master bond1
    ip link set bond0 nomaster
    ip link set bond1 master bond0
    ip link set bond1 nomaster

Splat looks like:
[  255.750458][ T1030] ============================================
[  255.751967][ T1030] WARNING: possible recursive locking detected
[  255.753435][ T1030] 5.9.0-rc8+ #772 Not tainted
[  255.754553][ T1030] --------------------------------------------
[  255.756047][ T1030] ip/1030 is trying to acquire lock:
[  255.757304][ T1030] ffff88811782a280 (&dev_addr_list_lock_key/1){+...}-{2:2}, at: dev_mc_sync_multiple+0xc2/0x150
[  255.760056][ T1030]
[  255.760056][ T1030] but task is already holding lock:
[  255.761862][ T1030] ffff88811130a280 (&dev_addr_list_lock_key/1){+...}-{2:2}, at: bond_enslave+0x3d4d/0x43e0 [bonding]
[  255.764581][ T1030]
[  255.764581][ T1030] other info that might help us debug this:
[  255.766645][ T1030]  Possible unsafe locking scenario:
[  255.766645][ T1030]
[  255.768566][ T1030]        CPU0
[  255.769415][ T1030]        ----
[  255.770259][ T1030]   lock(&dev_addr_list_lock_key/1);
[  255.771629][ T1030]   lock(&dev_addr_list_lock_key/1);
[  255.772994][ T1030]
[  255.772994][ T1030]  *** DEADLOCK ***
[  255.772994][ T1030]
[  255.775091][ T1030]  May be due to missing lock nesting notation
[  255.775091][ T1030]
[  255.777182][ T1030] 2 locks held by ip/1030:
[  255.778299][ T1030]  #0: ffffffffb1f63250 (rtnl_mutex){+.+.}-{3:3}, at: rtnetlink_rcv_msg+0x2e4/0x8b0
[  255.780600][ T1030]  #1: ffff88811130a280 (&dev_addr_list_lock_key/1){+...}-{2:2}, at: bond_enslave+0x3d4d/0x43e0 [bonding]
[  255.783411][ T1030]
[  255.783411][ T1030] stack backtrace:
[  255.784874][ T1030] CPU: 7 PID: 1030 Comm: ip Not tainted 5.9.0-rc8+ #772
[  255.786595][ T1030] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1ubuntu1 04/01/2014
[  255.789030][ T1030] Call Trace:
[  255.789850][ T1030]  dump_stack+0x99/0xd0
[  255.790882][ T1030]  __lock_acquire.cold.71+0x166/0x3cc
[  255.792285][ T1030]  ? register_lock_class+0x1a30/0x1a30
[  255.793619][ T1030]  ? rcu_read_lock_sched_held+0x91/0xc0
[  255.794963][ T1030]  ? rcu_read_lock_bh_held+0xa0/0xa0
[  255.796246][ T1030]  lock_acquire+0x1b8/0x850
[  255.797332][ T1030]  ? dev_mc_sync_multiple+0xc2/0x150
[  255.798624][ T1030]  ? bond_enslave+0x3d4d/0x43e0 [bonding]
[  255.800039][ T1030]  ? check_flags+0x50/0x50
[  255.801143][ T1030]  ? lock_contended+0xd80/0xd80
[  255.802341][ T1030]  _raw_spin_lock_nested+0x2e/0x70
[  255.803592][ T1030]  ? dev_mc_sync_multiple+0xc2/0x150
[  255.804897][ T1030]  dev_mc_sync_multiple+0xc2/0x150
[  255.806168][ T1030]  bond_enslave+0x3d58/0x43e0 [bonding]
[  255.807542][ T1030]  ? __lock_acquire+0xe53/0x51b0
[  255.808824][ T1030]  ? bond_update_slave_arr+0xdc0/0xdc0 [bonding]
[  255.810451][ T1030]  ? check_chain_key+0x236/0x5e0
[  255.811742][ T1030]  ? mutex_is_locked+0x13/0x50
[  255.812910][ T1030]  ? rtnl_is_locked+0x11/0x20
[  255.814061][ T1030]  ? netdev_master_upper_dev_get+0xf/0x120
[  255.815553][ T1030]  do_setlink+0x94c/0x3040
[ ... ]

Reported-by: syzbot+4a0f7bc34e3997a6c7df@syzkaller.appspotmail.com
Fixes: 1fc70edb ("net: core: add nested_level variable in net_device")
Signed-off-by: default avatarTaehee Yoo <ap420073@gmail.com>
Link: https://lore.kernel.org/r/20201015162606.9377-1-ap420073@gmail.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      0e8b8d6a
    • Jakub Kicinski's avatar
      ixgbe: fix probing of multi-port devices with one MDIO · bd7f14df
      Jakub Kicinski authored 
      Ian reports that after upgrade from v5.8.14 to v5.9 only one
of his 4 ixgbe netdevs appear in the system.

Quoting the comment on ixgbe_x550em_a_has_mii():
 * Returns true if hw points to lowest numbered PCI B:D.F x550_em_a device in
 * the SoC.  There are up to 4 MACs sharing a single MDIO bus on the x550em_a,
 * but we only want to register one MDIO bus.

This matches the symptoms, since the return value from
ixgbe_mii_bus_init() is no longer ignored we need to handle
the higher ports of x550em without an error.

Fixes: 09ef193f ("net: ethernet: ixgbe: check the return value of ixgbe_mii_bus_init()")
Reported-by: default avatarIan Kumlien <ian.kumlien@gmail.com>
Tested-by: default avatarIan Kumlien <ian.kumlien@gmail.com>
Acked-by: default avatarJesse Brandeburg <jesse.brandeburg@intel.com>
Link: https://lore.kernel.org/r/20201016232006.3352947-1-kuba@kernel.orgSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      bd7f14df
    • Anant Thazhemadam's avatar
      net: usb: rtl8150: don't incorrectly assign random MAC addresses · 60f1626f
      Anant Thazhemadam authored 
      In set_ethernet_addr(), if get_registers() succeeds, the ethernet address
that was read must be copied over. Otherwise, a random ethernet address
must be assigned.

get_registers() returns 0 if successful, and negative error number
otherwise. However, in set_ethernet_addr(), this return value is
incorrectly checked.

Since this return value will never be equal to sizeof(node_id), a
random MAC address will always be generated and assigned to the
device; even in cases when get_registers() is successful.

Correctly modifying the condition that checks if get_registers() was
successful or not fixes this problem, and copies the ethernet address
appropriately.

Fixes: b2a0f274 ("net: rtl8150: Use the new usb control message API.")
Signed-off-by: default avatarAnant Thazhemadam <anant.thazhemadam@gmail.com>
Link: https://lore.kernel.org/r/20201011173030.141582-1-anant.thazhemadam@gmail.comSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      60f1626f
    • Ido Schimmel's avatar
      selftests: forwarding: Add missing 'rp_filter' configuration · 71a0e29e
      Ido Schimmel authored 
      When 'rp_filter' is configured in strict mode (1) the tests fail because
packets received from the macvlan netdevs would not be forwarded through
them on the reverse path.

Fix this by disabling the 'rp_filter', meaning no source validation is
performed.

Fixes: 1538812e ("selftests: forwarding: Add a test for VXLAN asymmetric routing")
Fixes: 438a4f56 ("selftests: forwarding: Add a test for VXLAN symmetric routing")
Signed-off-by: default avatarIdo Schimmel <idosch@nvidia.com>
Reported-by: default avatarHangbin Liu <liuhangbin@gmail.com>
Tested-by: default avatarHangbin Liu <liuhangbin@gmail.com>
Link: https://lore.kernel.org/r/20201015084525.135121-1-idosch@idosch.orgSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      71a0e29e
    • Eelco Chaudron's avatar
      net: openvswitch: fix to make sure flow_lookup() is not preempted · f981fc3d
      Eelco Chaudron authored 
      The flow_lookup() function uses per CPU variables, which must be called
with BH disabled. However, this is fine in the general NAPI use case
where the local BH is disabled. But, it's also called from the netlink
context. The below patch makes sure that even in the netlink path, the
BH is disabled.

In addition, u64_stats_update_begin() requires a lock to ensure one writer
which is not ensured here. Making it per-CPU and disabling NAPI (softirq)
ensures that there is always only one writer.

Fixes: eac87c41 ("net: openvswitch: reorder masks array based on usage")
Reported-by: default avatarJuri Lelli <jlelli@redhat.com>
Signed-off-by: default avatarEelco Chaudron <echaudro@redhat.com>
Link: https://lore.kernel.org/r/160295903253.7789.826736662555102345.stgit@ebuildSigned-off-by: default avatarJakub Kicinski <kuba@kernel.org>
      f981fc3d
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7