David S. Miller [Thu, 30 May 2019 21:47:14 +0000 (14:47 -0700)]
Merge branch 'Fixes-for-DSA-tagging-using-802-1Q'
Vladimir Oltean says:
====================
Fixes for DSA tagging using 802.1Q
During the prototyping for the "Decoupling PHYLINK from struct
net_device" patchset, the CPU port of the sja1105 driver was moved to a
different spot. This uncovered an issue in the tag_8021q DSA code,
which used to work by mistake - the CPU port was the last hardware port
numerically, and this was masking an ordering issue which is very likely
to be seen in other drivers that make use of 802.1Q tags.
A question was also raised whether the VID numbers bear any meaning, and
the conclusion was that they don't, at least not in an absolute sense.
The second patch defines bit fields inside the DSA 802.1Q VID so that
tcpdump can decode it unambiguously (although the meaning is now clear
even by visual inspection).
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Vladimir Oltean [Wed, 29 May 2019 21:42:31 +0000 (00:42 +0300)]
net: dsa: tag_8021q: Create a stable binary format
Tools like tcpdump need to be able to decode the significance of fake
VLAN headers that DSA uses to separate switch ports.
But currently these have no global significance - they are simply an
ordered list of DSA_MAX_SWITCHES x DSA_MAX_PORTS numbers ending at 4095.
The reason why this is submitted as a fix is that the existing mapping
of VIDs should not enter into a stable kernel, so we can pretend that
only the new format exists. This way tcpdump won't need to try to make
something out of the VLAN tags on 5.2 kernels.
Fixes: f9bbe4477c30 ("net: dsa: Optional VLAN-based port separation for switches without tagging") Signed-off-by: Vladimir Oltean <olteanv@gmail.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ioana Ciornei [Wed, 29 May 2019 21:42:30 +0000 (00:42 +0300)]
net: dsa: tag_8021q: Change order of rx_vid setup
The 802.1Q tagging performs an unbalanced setup in terms of RX VIDs on
the CPU port. For the ingress path of a 802.1Q switch to work, the RX
VID of a port needs to be seen as tagged egress on the CPU port.
While configuring the other front-panel ports to be part of this VID,
for bridge scenarios, the untagged flag is applied even on the CPU port
in dsa_switch_vlan_add. This happens because DSA applies the same flags
on the CPU port as on the (bridge-controlled) slave ports, and the
effect in this case is that the CPU port tagged settings get deleted.
Instead of fixing DSA by introducing a way to control VLAN flags on the
CPU port (and hence stop inheriting from the slave ports) - a hard,
perhaps intractable problem - avoid this situation by moving the setup
part of the RX VID on the CPU port after all the other front-panel ports
have been added to the VID.
Fixes: f9bbe4477c30 ("net: dsa: Optional VLAN-based port separation for switches without tagging") Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com> Signed-off-by: Vladimir Oltean <olteanv@gmail.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Antoine Tenart [Wed, 29 May 2019 13:59:48 +0000 (15:59 +0200)]
net: mvpp2: fix bad MVPP2_TXQ_SCHED_TOKEN_CNTR_REG queue value
MVPP2_TXQ_SCHED_TOKEN_CNTR_REG() expects the logical queue id but
the current code is passing the global tx queue offset, so it ends
up writing to unknown registers (between 0x8280 and 0x82fc, which
seemed to be unused by the hardware). This fixes the issue by using
the logical queue id instead.
Fixes: 3f518509dedc ("ethernet: Add new driver for Marvell Armada 375 network unit") Signed-off-by: Antoine Tenart <antoine.tenart@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 30 May 2019 19:30:47 +0000 (12:30 -0700)]
Merge branch 'mlxsw-Two-small-fixes'
Ido Schimmel says:
====================
mlxsw: Two small fixes
Patch #1 from Jiri fixes an issue specific to Spectrum-2 where the
insertion of two identical flower filters with different priorities
would trigger a warning.
Patch #2 from Amit prevents the driver from trying to configure a port
with a speed of 56Gb/s and autoneg off as this is not supported and
results in error messages from firmware.
Please consider patch #1 for stable.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Amit Cohen [Wed, 29 May 2019 07:59:45 +0000 (10:59 +0300)]
mlxsw: spectrum: Prevent force of 56G
Force of 56G is not supported by hardware in Ethernet devices. This
configuration fails with a bad parameter error from firmware.
Add check of this case. Instead of trying to set 56G with autoneg off,
return a meaningful error.
Fixes: 56ade8fe3fe1 ("mlxsw: spectrum: Add initial support for Spectrum ASIC") Signed-off-by: Amit Cohen <amitc@mellanox.com> Acked-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Ido Schimmel <idosch@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Wed, 29 May 2019 07:59:44 +0000 (10:59 +0300)]
mlxsw: spectrum_acl: Avoid warning after identical rules insertion
When identical rules are inserted, the latter one goes to C-TCAM. For
that, a second eRP with the same mask is created. These 2 eRPs by the
nature cannot be merged and also one cannot be parent of another.
Teach mlxsw_sp_acl_erp_delta_fill() about this possibility and handle it
gracefully.
Reported-by: Alex Kushnarov <alexanderk@mellanox.com> Fixes: c22291f7cf45 ("mlxsw: spectrum: acl: Implement delta for ERP") Signed-off-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: Ido Schimmel <idosch@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Heiner Kallweit [Wed, 29 May 2019 05:44:01 +0000 (07:44 +0200)]
r8169: fix MAC address being lost in PCI D3
(At least) RTL8168e forgets its MAC address in PCI D3. To fix this set
the MAC address when resuming. For resuming from runtime-suspend we
had this in place already, for resuming from S3/S5 it was missing.
The commit referenced as being fixed isn't wrong, it's just the first
one where the patch applies cleanly.
Fixes: 0f07bd850d36 ("r8169: use dev_get_drvdata where possible") Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> Reported-by: Albert Astals Cid <aacid@kde.org> Tested-by: Albert Astals Cid <aacid@kde.org> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 30 May 2019 18:12:21 +0000 (11:12 -0700)]
Merge branch 'XDP-generic-fixes'
Stephen Hemminger says:
====================
XDP generic fixes
This set of patches came about while investigating XDP
generic on Azure. The split brain nature of the accelerated
networking exposed issues with the stack device model.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
net: core: support XDP generic on stacked devices.
When a device is stacked like (team, bonding, failsafe or netvsc) the
XDP generic program for the parent device was not called.
Move the call to XDP generic inside __netif_receive_skb_core where
it can be done multiple times for stacked case.
Fixes: d445516966dc ("net: xdp: support xdp generic on virtual devices") Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
The netvsc VF skb handler should make sure that skb is not
shared. Similar logic already exists in bonding and team device
drivers.
This is not an issue in practice because the VF devicex
does not send up shared skb's. But the netvsc driver
should do the right thing if it did.
Fixes: 0c195567a8f6 ("netvsc: transparent VF management") Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Sean Tranchetti [Tue, 28 May 2019 18:22:54 +0000 (12:22 -0600)]
udp: Avoid post-GRO UDP checksum recalculation
Currently, when resegmenting an unexpected UDP GRO packet, the full UDP
checksum will be calculated for every new SKB created by skb_segment()
because the netdev features passed in by udp_rcv_segment() lack any
information about checksum offload capabilities.
Usually, we have no need to perform this calculation again, as
1) The GRO implementation guarantees that any packets making it to the
udp_rcv_segment() function had correct checksums, and, more
importantly,
2) Upon the successful return of udp_rcv_segment(), we immediately pull
the UDP header off and either queue the segment to the socket or
hand it off to a new protocol handler.
Unless userspace has set the IP_CHECKSUM sockopt to indicate that they
want the final checksum values, we can pass the needed netdev feature
flags to __skb_gso_segment() to avoid checksumming each segment in
skb_segment().
Fixes: cf329aa42b66 ("udp: cope with UDP GRO packet misdirection") Cc: Paolo Abeni <pabeni@redhat.com> Cc: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org> Signed-off-by: Sean Tranchetti <stranche@codeaurora.org> Acked-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Max Uvarov [Tue, 28 May 2019 10:00:51 +0000 (13:00 +0300)]
net: phy: dp83867: do not call config_init twice
Phy state machine calls _config_init just after
reset.
Signed-off-by: Max Uvarov <muvarov@gmail.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
After reset SGMII Autoneg timer is set to 2us (bits 6 and 5 are 01).
That is not enough to finalize autonegatiation on some devices.
Increase this timer duration to maximum supported 16ms.
Signed-off-by: Max Uvarov <muvarov@gmail.com> Cc: Heiner Kallweit <hkallweit1@gmail.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Max Uvarov [Tue, 28 May 2019 10:00:49 +0000 (13:00 +0300)]
net: phy: dp83867: fix speed 10 in sgmii mode
For supporting 10Mps speed in SGMII mode DP83867_10M_SGMII_RATE_ADAPT bit
of DP83867_10M_SGMII_CFG register has to be cleared by software.
That does not affect speeds 100 and 1000 so can be done on init.
Signed-off-by: Max Uvarov <muvarov@gmail.com> Cc: Heiner Kallweit <hkallweit1@gmail.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Russell King [Tue, 28 May 2019 09:34:42 +0000 (10:34 +0100)]
net: phy: marvell10g: report if the PHY fails to boot firmware
Some boards do not have the PHY firmware programmed in the 3310's flash,
which leads to the PHY not working as expected. Warn the user when the
PHY fails to boot the firmware and refuse to initialise.
Fixes: 20b2af32ff3f ("net: phy: add Marvell Alaska X 88X3310 10Gigabit PHY support") Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk> Tested-by: Maxime Chevallier <maxime.chevallier@bootlin.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
Ensure that we supply the same phy interface mode to mac_link_down() as
we did for the corresponding mac_link_up() call. This ensures that MAC
drivers that use the phy interface mode in these methods can depend on
mac_link_down() always corresponding to a mac_link_up() call for the
same interface mode.
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk> Signed-off-by: David S. Miller <davem@davemloft.net>
net: sh_eth: fix mdio access in sh_eth_close() for R-Car Gen2 and RZ/A1 SoCs
The sh_eth_close() resets the MAC and then calls phy_stop()
so that mdio read access result is incorrect without any error
according to kernel trace like below:
According to the hardware manual, the RMII mode should be set to 1
before operation the Ethernet MAC. However, the previous code was not
set to 1 after the driver issued the soft_reset in sh_eth_dev_exit()
so that the mdio read access result seemed incorrect. To fix the issue,
this patch adds a condition and set the RMII mode register in
sh_eth_dev_exit() for R-Car Gen2 and RZ/A1 SoCs.
Note that when I have tried to move the sh_eth_dev_exit() calling
after phy_stop() on sh_eth_close(), but it gets worse (kernel panic
happened and it seems that a register is accessed while the clock is
off).
Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Saeed Mahameed [Thu, 23 May 2019 19:55:10 +0000 (12:55 -0700)]
net/mlx5e: Disable rxhash when CQE compress is enabled
When CQE compression is enabled (Multi-host systems), compressed CQEs
might arrive to the driver rx, compressed CQEs don't have a valid hash
offload and the driver already reports a hash value of 0 and invalid hash
type on the skb for compressed CQEs, but this is not good enough.
On a congested PCIe, where CQE compression will kick in aggressively,
gro will deliver lots of out of order packets due to the invalid hash
and this might cause a serious performance drop.
The only valid solution, is to disable rxhash offload at all when CQE
compression is favorable (Multi-host systems).
wenxu [Wed, 15 May 2019 09:25:45 +0000 (17:25 +0800)]
net/mlx5e: restrict the real_dev of vlan device is the same as uplink device
When register indr block for vlan device, it should check the real_dev
of vlan device is same as uplink device. Or it will set offload rule
to mlx5e which will never hit.
Parav Pandit [Fri, 10 May 2019 15:40:08 +0000 (10:40 -0500)]
net/mlx5: Allocate root ns memory using kzalloc to match kfree
root ns is yet another fs core node which is freed using kfree() by
tree_put_node().
Rest of the other fs core objects are also allocated using kmalloc
variants.
However, root ns memory is allocated using kvzalloc().
Hence allocate root ns memory using kzalloc().
Fixes: 2530236303d9e ("net/mlx5_core: Flow steering tree initialization") Signed-off-by: Parav Pandit <parav@mellanox.com> Reviewed-by: Daniel Jurgens <danielj@mellanox.com> Reviewed-by: Mark Bloch <markb@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Parav Pandit [Fri, 10 May 2019 15:26:23 +0000 (10:26 -0500)]
net/mlx5: Avoid double free in fs init error unwinding path
In below code flow, for ingress acl table root ns memory leads
to double free.
mlx5_init_fs
init_ingress_acls_root_ns()
init_ingress_acl_root_ns
kfree(steering->esw_ingress_root_ns);
/* steering->esw_ingress_root_ns is not marked NULL */
mlx5_cleanup_fs
cleanup_ingress_acls_root_ns
steering->esw_ingress_root_ns non NULL check passes.
kfree(steering->esw_ingress_root_ns);
/* double free */
Similar issue exist for other tables.
Hence zero out the pointers to not process the table again.
Fixes: 9b93ab981e3bf ("net/mlx5: Separate ingress/egress namespaces for each vport") Fixes: 40c3eebb49e51 ("net/mlx5: Add support in RDMA RX steering") Signed-off-by: Parav Pandit <parav@mellanox.com> Reviewed-by: Mark Bloch <markb@mellanox.com> Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Parav Pandit [Thu, 9 May 2019 22:23:24 +0000 (17:23 -0500)]
net/mlx5: Avoid double free of root ns in the error flow path
When root ns setup for rdma, sniffer tx and sniffer rx fails,
such root ns cleanup is done by the error unwinding path of
mlx5_cleanup_fs().
Below call graph shows an example for sniffer_rx_root_ns.
Both sysfs-bus-mdio and sysfs-class-net-phydev contain the same
duplication information. There is not currently any MDIO bus specific
attribute, but there are PHY device (struct phy_device) specific
attributes. Use the more precise description from sysfs-bus-mdio and
carry that over to sysfs-class-net-phydev.
Fixes: 86f22d04dfb5 ("net: sysfs: Document PHY device sysfs attributes") Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: syzbot <syzkaller@googlegroups.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Mon, 27 May 2019 17:42:23 +0000 (19:42 +0200)]
selftests: pmtu: Fix encapsulating device in pmtu_vti6_link_change_mtu
In the pmtu_vti6_link_change_mtu test, both local and remote addresses
for the vti6 tunnel are assigned to the same address given to the dummy
interface that we use as encapsulating device with a known MTU.
This works as long as the dummy interface is actually selected, via
rt6_lookup(), as encapsulating device. But if the remote address of the
tunnel is a local address too, the loopback interface could also be
selected, and there's nothing wrong with it.
This is what some older -stable kernels do (3.18.z, at least), and
nothing prevents us from subtly changing FIB implementation to revert
back to that behaviour in the future.
Define an IPv6 prefix instead, and use two separate addresses as local
and remote for vti6, so that the encapsulating device can't be a
loopback interface.
Reported-by: Xiumei Mu <xmu@redhat.com> Fixes: 1fad59ea1c34 ("selftests: pmtu: Add pmtu_vti6_link_change_mtu test") Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Madalin Bucur [Mon, 27 May 2019 12:24:05 +0000 (15:24 +0300)]
dpaa_eth: use only online CPU portals
Make sure only the portals for the online CPUs are used.
Without this change, there are issues when someone boots with
maxcpus=n, with n < actual number of cores available as frames
either received or corresponding to the transmit confirmation
path would be offered for dequeue to the offline CPU portals,
getting lost.
Signed-off-by: Madalin Bucur <madalin.bucur@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jisheng Zhang [Mon, 27 May 2019 11:04:17 +0000 (11:04 +0000)]
net: mvneta: Fix err code path of probe
Fix below issues in err code path of probe:
1. we don't need to unregister_netdev() because the netdev isn't
registered.
2. when register_netdev() fails, we also need to destroy bm pool for
HWBM case.
Fixes: dc35a10f68d3 ("net: mvneta: bm: add support for hardware buffer management") Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Thierry Reding [Mon, 27 May 2019 10:52:51 +0000 (12:52 +0200)]
net: stmmac: Do not output error on deferred probe
If the subdriver defers probe, do not show an error message. It's
perfectly fine for this error to occur since the driver will get another
chance to probe after some time and will usually succeed after all of
the resources that it requires have been registered.
Signed-off-by: Thierry Reding <treding@nvidia.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
Nikita Danilov [Sat, 25 May 2019 09:58:05 +0000 (09:58 +0000)]
net: aquantia: tcp checksum 0xffff being handled incorrectly
Thats a known quirk in windows tcp stack it can produce 0xffff checksum.
Thats incorrect but it is.
Atlantic HW with LRO enabled handles that incorrectly and changes csum to
0xfffe - but indicates that csum is invalid. This causes driver to pass
packet to linux networking stack with CSUM_NONE, stack eventually drops
the packet.
There is a quirk in atlantic HW to enable correct processing of
0xffff incorrect csum. Enable it.
The visible bug is that windows link partner with software generated csums
caused TCP connection to be unstable since all packets that csum value
are dropped.
Reported-by: Dmitry Bezrukov <dmitry.bezrukov@aquantia.com> Signed-off-by: Nikita Danilov <ndanilov@aquantia.com> Signed-off-by: Igor Russkikh <igor.russkikh@aquantia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Dmitry Bogdanov [Sat, 25 May 2019 09:58:03 +0000 (09:58 +0000)]
net: aquantia: fix LRO with FCS error
Driver stops producing skbs on ring if a packet with FCS error
was coalesced into LRO session. Ring gets hang forever.
Thats a logical error in driver processing descriptors:
When rx_stat indicates MAC Error, next pointer and eop flags
are not filled. This confuses driver so it waits for descriptor 0
to be filled by HW.
Solution is fill next pointer and eop flag even for packets with FCS error.
Fixes: bab6de8fd180b ("net: ethernet: aquantia: Atlantic A0 and B0 specific functions.") Signed-off-by: Igor Russkikh <igor.russkikh@aquantia.com> Signed-off-by: Dmitry Bogdanov <dmitry.bogdanov@aquantia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Dmitry Bogdanov [Sat, 25 May 2019 09:58:01 +0000 (09:58 +0000)]
net: aquantia: check rx csum for all packets in LRO session
Atlantic hardware does not aggregate nor breaks LRO sessions
with bad csum packets. This means driver should take care of that.
If in LRO session there is a non-first descriptor with invalid
checksum (L2/L3/L4), the driver must account this information
in csum application logic.
Fixes: 018423e90bee8 ("net: ethernet: aquantia: Add ring support code") Signed-off-by: Igor Russkikh <igor.russkikh@aquantia.com> Signed-off-by: Dmitry Bogdanov <dmitry.bogdanov@aquantia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Igor Russkikh [Sat, 25 May 2019 09:57:59 +0000 (09:57 +0000)]
net: aquantia: tx clean budget logic error
In case no other traffic happening on the ring, full tx cleanup
may not be completed. That may cause socket buffer to overflow
and tx traffic to stuck until next activity on the ring happens.
This is due to logic error in budget variable decrementor.
Variable is compared with zero, and then post decremented,
causing it to become MAX_INT. Solution is remove decrementor
from the `for` statement and rewrite it in a clear way.
Fixes: b647d3980948e ("net: aquantia: Add tx clean budget and valid budget handling logic") Signed-off-by: Igor Russkikh <igor.russkikh@aquantia.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Kees Cook [Fri, 24 May 2019 20:20:19 +0000 (13:20 -0700)]
net: tulip: de4x5: Drop redundant MODULE_DEVICE_TABLE()
Building with Clang reports the redundant use of MODULE_DEVICE_TABLE():
drivers/net/ethernet/dec/tulip/de4x5.c:2110:1: error: redefinition of '__mod_eisa__de4x5_eisa_ids_device_table'
MODULE_DEVICE_TABLE(eisa, de4x5_eisa_ids);
^
./include/linux/module.h:229:21: note: expanded from macro 'MODULE_DEVICE_TABLE'
extern typeof(name) __mod_##type##__##name##_device_table \
^
<scratch space>:90:1: note: expanded from here
__mod_eisa__de4x5_eisa_ids_device_table
^
drivers/net/ethernet/dec/tulip/de4x5.c:2100:1: note: previous definition is here
MODULE_DEVICE_TABLE(eisa, de4x5_eisa_ids);
^
./include/linux/module.h:229:21: note: expanded from macro 'MODULE_DEVICE_TABLE'
extern typeof(name) __mod_##type##__##name##_device_table \
^
<scratch space>:85:1: note: expanded from here
__mod_eisa__de4x5_eisa_ids_device_table
^
This drops the one further from the table definition to match the common
use of MODULE_DEVICE_TABLE().
Fixes: 07563c711fbc ("EISA bus MODALIAS attributes support") Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
net/tls: two fixes for rx_list pre-handling
tls_sw_recvmsg() had been modified to cater better to async decrypt.
Partially read records now live on the rx_list. Data is copied from
this list before the old do {} while loop, and the not included
correctly in deciding whether to sleep or not and lowat threshold
handling. These modifications, unfortunately, added some bugs.
First patch fixes lowat - we need to calculate the threshold early
and make sure all copied data is compared to the threshold, not just
the freshly decrypted data.
Third patch fixes sleep - if data is picked up from rx_list and
no flags are set, we should not put the process to sleep, but
rather return the partial read.
Patches 2 and 4 add test cases for these bugs, both will cause
a sleep and test timeout before the fix.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Fri, 24 May 2019 17:34:33 +0000 (10:34 -0700)]
selftests/tls: add test for sleeping even though there is data
Add a test which sends 15 bytes of data, and then tries
to read 10 byes twice. Previously the second read would
sleep indifinitely, since the record was already decrypted
and there is only 5 bytes left, not full 10.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Fri, 24 May 2019 17:34:32 +0000 (10:34 -0700)]
net/tls: fix no wakeup on partial reads
When tls_sw_recvmsg() partially copies a record it pops that
record from ctx->recv_pkt and places it on rx_list.
Next iteration of tls_sw_recvmsg() reads from rx_list via
process_rx_list() before it enters the decryption loop.
If there is no more records to be read tls_wait_data()
will put the process on the wait queue and got to sleep.
This is incorrect, because some data was already copied
in process_rx_list().
In case of RPC connections process may never get woken up,
because peer also simply blocks in read().
I think this may also fix a similar issue when BPF is at
play, because after __tcp_bpf_recvmsg() returns some data
we subtract it from len and use continue to restart the
loop, but len could have just reached 0, so again we'd
sleep unnecessarily. That's added by:
commit d3b18ad31f93 ("tls: add bpf support to sk_msg handling")
Fixes: 692d7b5d1f91 ("tls: Fix recvmsg() to be able to peek across multiple records") Reported-by: David Beckett <david.beckett@netronome.com> Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Tested-by: David Beckett <david.beckett@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Fri, 24 May 2019 17:34:31 +0000 (10:34 -0700)]
selftests/tls: test for lowat overshoot with multiple records
Set SO_RCVLOWAT and test it gets respected when gathering
data from multiple records.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Fri, 24 May 2019 17:34:30 +0000 (10:34 -0700)]
net/tls: fix lowat calculation if some data came from previous record
If some of the data came from the previous record, i.e. from
the rx_list it had already been decrypted, so it's not counted
towards the "decrypted" variable, but the "copied" variable.
Take that into account when checking lowat.
When calculating lowat target we need to pass the original len.
E.g. if lowat is at 80, len is 100 and we had 30 bytes on rx_list
target would currently be incorrectly calculated as 70, even though
we only need 50 more bytes to make up the 80.
Fixes: 692d7b5d1f91 ("tls: Fix recvmsg() to be able to peek across multiple records") Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Tested-by: David Beckett <david.beckett@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ioana Radulescu [Fri, 24 May 2019 15:15:17 +0000 (18:15 +0300)]
dpaa2-eth: Make constant 64-bit long
Function dpaa2_eth_cls_key_size() expects a 64bit argument,
but DPAA2_ETH_DIST_ALL is defined as UINT_MAX. Fix this.
Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com> Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ioana Radulescu [Fri, 24 May 2019 15:15:15 +0000 (18:15 +0300)]
dpaa2-eth: Fix potential spectre issue
Smatch reports a potential spectre vulnerability in the dpaa2-eth
driver, where the value of rxnfc->fs.location (which is provided
from user-space) is used as index in an array.
Add a call to array_index_nospec() to sanitize the access.
Signed-off-by: Ioana Radulescu <ruxandra.radulescu@nxp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jarod Wilson [Fri, 24 May 2019 13:49:28 +0000 (09:49 -0400)]
bonding/802.3ad: fix slave link initialization transition states
Once in a while, with just the right timing, 802.3ad slaves will fail to
properly initialize, winding up in a weird state, with a partner system
mac address of 00:00:00:00:00:00. This started happening after a fix to
properly track link_failure_count tracking, where an 802.3ad slave that
reported itself as link up in the miimon code, but wasn't able to get a
valid speed/duplex, started getting set to BOND_LINK_FAIL instead of
BOND_LINK_DOWN. That was the proper thing to do for the general "my link
went down" case, but has created a link initialization race that can put
the interface in this odd state.
The simple fix is to instead set the slave link to BOND_LINK_DOWN again,
if the link has never been up (last_link_up == 0), so the link state
doesn't bounce from BOND_LINK_DOWN to BOND_LINK_FAIL -- it hasn't failed
in this case, it simply hasn't been up yet, and this prevents the
unnecessary state change from DOWN to FAIL and getting stuck in an init
failure w/o a partner mac.
Fixes: ea53abfab960 ("bonding/802.3ad: fix link_failure_count tracking") CC: Jay Vosburgh <j.vosburgh@gmail.com> CC: Veaceslav Falico <vfalico@gmail.com> CC: Andy Gospodarek <andy@greyhouse.net> CC: "David S. Miller" <davem@davemloft.net> CC: netdev@vger.kernel.org Tested-by: Heesoon Kim <Heesoon.Kim@stratus.com> Signed-off-by: Jarod Wilson <jarod@redhat.com> Acked-by: Jay Vosburgh <jay.vosburgh@canonical.com> Signed-off-by: David S. Miller <davem@davemloft.net>
net: ethtool: Document get_rxfh_context and set_rxfh_context ethtool ops
ethtool ops get_rxfh_context and set_rxfh_context are used to create,
remove and access parameters associated to RSS contexts, in a similar
fashion to get_rxfh and set_rxfh.
Add a small descritopn of these callbacks in the struct ethtool_ops doc.
Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Sat, 25 May 2019 18:02:31 +0000 (11:02 -0700)]
Merge branch 'fix-some-bugs-in-stmmac'
Biao Huang says:
====================
fix some bugs in stmmac
changes in v4:
since MTL_OPERATION_MODE write back issue has be fixed in the latest driver,
remove original patch#3
changes in v3:
add a Fixes:tag for each patch
changes in v2:
1. update rx_tail_addr as Jose's comment
2. changes clk_csr condition as Alex's proposition
3. remove init lines in dwmac-mediatek, get clk_csr from dts instead.
v1:
This series fix some bugs in stmmac driver
3 patches are for common stmmac or dwmac4:
1. update rx tail pointer to fix rx dma hang issue.
2. change condition for mdc clock to fix csr_clk can't be zero issue.
3. write the modified value back to MTL_OPERATION_MODE.
1 patch is for dwmac-mediatek:
modify csr_clk value to fix mdio read/write fail issue for dwmac-mediatek
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Biao Huang [Fri, 24 May 2019 06:26:09 +0000 (14:26 +0800)]
net: stmmac: dwmac-mediatek: modify csr_clk value to fix mdio read/write fail
1. the frequency of csr clock is 66.5MHz, so the csr_clk value should
be 0 other than 5.
2. the csr_clk can be got from device tree, so remove initialization here.
Fixes: 9992f37e346b ("stmmac: dwmac-mediatek: add support for mt2712") Signed-off-by: Biao Huang <biao.huang@mediatek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Biao Huang [Fri, 24 May 2019 06:26:08 +0000 (14:26 +0800)]
net: stmmac: fix csr_clk can't be zero issue
The specific clk_csr value can be zero, and
stmmac_clk is necessary for MDC clock which can be set dynamically.
So, change the condition from plat->clk_csr to plat->stmmac_clk to
fix clk_csr can't be zero issue.
Fixes: cd7201f477b9 ("stmmac: MDC clock dynamically based on the csr clock input") Signed-off-by: Biao Huang <biao.huang@mediatek.com> Acked-by: Alexandre TORGUE <alexandre.torgue@st.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Biao Huang [Fri, 24 May 2019 06:26:07 +0000 (14:26 +0800)]
net: stmmac: update rx tail pointer register to fix rx dma hang issue.
Currently we will not update the receive descriptor tail pointer in
stmmac_rx_refill. Rx dma will think no available descriptors and stop
once received packets exceed DMA_RX_SIZE, so that the rx only test will fail.
Update the receive tail pointer in stmmac_rx_refill to add more descriptors
to the rx channel, so packets can be received continually
Fixes: 54139cf3bb33 ("net: stmmac: adding multiple buffers for rx") Signed-off-by: Biao Huang <biao.huang@mediatek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Gen Zhang [Fri, 24 May 2019 03:24:26 +0000 (11:24 +0800)]
ip_sockglue: Fix missing-check bug in ip_ra_control()
In function ip_ra_control(), the pointer new_ra is allocated a memory
space via kmalloc(). And it is used in the following codes. However,
when there is a memory allocation error, kmalloc() fails. Thus null
pointer dereference may happen. And it will cause the kernel to crash.
Therefore, we should check the return value and handle the error.
Signed-off-by: Gen Zhang <blackgod016574@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Gen Zhang [Fri, 24 May 2019 03:19:46 +0000 (11:19 +0800)]
ipv6_sockglue: Fix a missing-check bug in ip6_ra_control()
In function ip6_ra_control(), the pointer new_ra is allocated a memory
space via kmalloc(). And it is used in the following codes. However,
when there is a memory allocation error, kmalloc() fails. Thus null
pointer dereference may happen. And it will cause the kernel to crash.
Therefore, we should check the return value and handle the error.
Signed-off-by: Gen Zhang <blackgod016574@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Vlad Buslov [Thu, 23 May 2019 06:32:31 +0000 (09:32 +0300)]
net: sched: don't use tc_action->order during action dump
Function tcf_action_dump() relies on tc_action->order field when starting
nested nla to send action data to userspace. This approach breaks in
several cases:
- When multiple filters point to same shared action, tc_action->order field
is overwritten each time it is attached to filter. This causes filter
dump to output action with incorrect attribute for all filters that have
the action in different position (different order) from the last set
tc_action->order value.
- When action data is displayed using tc action API (RTM_GETACTION), action
order is overwritten by tca_action_gd() according to its position in
resulting array of nl attributes, which will break filter dump for all
filters attached to that shared action that expect it to have different
order value.
Don't rely on tc_action->order when dumping actions. Set nla according to
action position in resulting array of actions instead.
Signed-off-by: Vlad Buslov <vladbu@mellanox.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Vishal Kulkarni [Thu, 23 May 2019 02:37:21 +0000 (08:07 +0530)]
cxgb4: Revert "cxgb4: Remove SGE_HOST_PAGE_SIZE dependency on page size"
This reverts commit 2391b0030e241386d710df10e53e2cfc3c5d4fc1 which has
introduced regression. Now SGE's BAR2 Doorbell/GTS Page Size is
interpreted correctly in the firmware itself by using actual host
page size. Hence previous commit needs to be reverted.
Signed-off-by: Vishal Kulkarni <vishal@chelsio.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Andy Duan [Thu, 23 May 2019 01:55:28 +0000 (01:55 +0000)]
net: fec: fix the clk mismatch in failed_reset path
Fix the clk mismatch in the error path "failed_reset" because
below error path will disable clk_ahb and clk_ipg directly, it
should use pm_runtime_put_noidle() instead of pm_runtime_put()
to avoid to call runtime resume callback.
Reported-by: Baruch Siach <baruch@tkos.co.il> Signed-off-by: Fugang Duan <fugang.duan@nxp.com> Tested-by: Baruch Siach <baruch@tkos.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
Raju Rangoju [Thu, 23 May 2019 15:11:44 +0000 (20:41 +0530)]
cxgb4: offload VLAN flows regardless of VLAN ethtype
VLAN flows never get offloaded unless ivlan_vld is set in filter spec.
It's not compulsory for vlan_ethtype to be set.
So, always enable ivlan_vld bit for offloading VLAN flows regardless of
vlan_ethtype is set or not.
Fixes: ad9af3e09c (cxgb4: add tc flower match support for vlan) Signed-off-by: Raju Rangoju <rajur@chelsio.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Andreas Oetken [Thu, 23 May 2019 11:57:14 +0000 (13:57 +0200)]
hsr: fix don't prune the master node from the node_db
Don't prune the master node in the hsr_prune_nodes function.
Neither time_in[HSR_PT_SLAVE_A] nor time_in[HSR_PT_SLAVE_B]
will ever be updated by hsr_register_frame_in for the master port.
Thus, the master node will be repeatedly pruned leading to
repeated packet loss.
This bug never appeared because the hsr_prune_nodes function
was only called once. Since commit 5150b45fd355
("net: hsr: Fix node prune function for forget time expiry") this issue
is fixed unveiling the issue described above.
Fixes: 5150b45fd355 ("net: hsr: Fix node prune function for forget time expiry") Signed-off-by: Andreas Oetken <andreas.oetken@siemens.com> Tested-by: Murali Karicheri <m-karicheri2@ti.com> Signed-off-by: David S. Miller <davem@davemloft.net>
The flow_rule is only used when configuring the classification tables,
and should be free'd once we're done using it. The current code only
frees it in the error path.
Fixes: 90b509b39ac9 ("net: mvpp2: cls: Add Classification offload support") Signed-off-by: Maxime Chevallier <maxime.chevallier@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Thu, 23 May 2019 01:35:16 +0000 (18:35 -0700)]
ipv4/igmp: fix build error if !CONFIG_IP_MULTICAST
ip_sf_list_clear_all() needs to be defined even if !CONFIG_IP_MULTICAST
Fixes: 3580d04aa674 ("ipv4/igmp: fix another memory leak in igmpv3_del_delrec()") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: kbuild test robot <lkp@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Fixes: 9c8bb163ae78 ("igmp, mld: Fix memory leak in igmpv3/mld_del_delrec()") Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Hangbin Liu <liuhangbin@gmail.com> Reported-by: syzbot <syzkaller@googlegroups.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 23 May 2019 01:02:14 +0000 (18:02 -0700)]
Merge branch 'bnxt_en-Bug-fixes'
Michael Chan says:
===================
bnxt_en: Bug fixes.
There are 4 driver fixes in this series:
1. Fix RX buffer leak during OOM condition.
2. Call pci_disable_msix() under correct conditions to prevent hitting BUG.
3. Reduce unneeded mmeory allocation in kdump kernel to prevent OOM.
4. Don't read device serial number on VFs because it is not supported.
Please queue #1, #2, #3 for -stable as well. Thanks.
===================
Signed-off-by: David S. Miller <davem@davemloft.net>
Vasundhara Volam [Wed, 22 May 2019 23:12:57 +0000 (19:12 -0400)]
bnxt_en: Device serial number is supported only for PFs.
Don't read DSN on VFs that do not have the PCI capability.
Fixes: 03213a996531 ("bnxt: move bp->switch_id initialization to PF probe") Signed-off-by: Vasundhara Volam <vasundhara-v.volam@broadcom.com> Signed-off-by: Michael Chan <michael.chan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Michael Chan [Wed, 22 May 2019 23:12:56 +0000 (19:12 -0400)]
bnxt_en: Reduce memory usage when running in kdump kernel.
Skip RDMA context memory allocations, reduce to 1 ring, and disable
TPA when running in the kdump kernel. Without this patch, the driver
fails to initialize with memory allocation errors when running in a
typical kdump kernel.
Fixes: cf6daed098d1 ("bnxt_en: Increase context memory allocations on 57500 chips for RDMA.") Signed-off-by: Michael Chan <michael.chan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Michael Chan [Wed, 22 May 2019 23:12:55 +0000 (19:12 -0400)]
bnxt_en: Fix possible BUG() condition when calling pci_disable_msix().
When making configuration changes, the driver calls bnxt_close_nic()
and then bnxt_open_nic() for the changes to take effect. A parameter
irq_re_init is passed to the call sequence to indicate if IRQ
should be re-initialized. This irq_re_init parameter needs to
be included in the bnxt_reserve_rings() call. bnxt_reserve_rings()
can only call pci_disable_msix() if the irq_re_init parameter is
true, otherwise it may hit BUG() because some IRQs may not have been
freed yet.
Fixes: 41e8d7983752 ("bnxt_en: Modify the ring reservation functions for 57500 series chips.") Signed-off-by: Michael Chan <michael.chan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Michael Chan [Wed, 22 May 2019 23:12:54 +0000 (19:12 -0400)]
bnxt_en: Fix aggregation buffer leak under OOM condition.
For every RX packet, the driver replenishes all buffers used for that
packet and puts them back into the RX ring and RX aggregation ring.
In one code path where the RX packet has one RX buffer and one or more
aggregation buffers, we missed recycling the aggregation buffer(s) if
we are unable to allocate a new SKB buffer. This leads to the
aggregation ring slowly running out of buffers over time. Fix it
by properly recycling the aggregation buffers.
Fixes: c0c050c58d84 ("bnxt_en: New Broadcom ethernet driver.") Reported-by: Rakesh Hemnani <rhemnani@fb.com> Signed-off-by: Michael Chan <michael.chan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David Ahern [Wed, 22 May 2019 22:12:18 +0000 (15:12 -0700)]
ipv6: Fix redirect with VRF
IPv6 redirect is broken for VRF. __ip6_route_redirect walks the FIB
entries looking for an exact match on ifindex. With VRF the flowi6_oif
is updated by l3mdev_update_flow to the l3mdev index and the
FLOWI_FLAG_SKIP_NH_OIF set in the flags to tell the lookup to skip the
device match. For redirects the device match is requires so use that
flag to know when the oif needs to be reset to the skb device index.
Fixes: ca254490c8df ("net: Add VRF support to IPv6 stack") Signed-off-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jisheng Zhang [Wed, 22 May 2019 10:05:09 +0000 (10:05 +0000)]
net: stmmac: fix reset gpio free missing
Commit 984203ceff27 ("net: stmmac: mdio: remove reset gpio free")
removed the reset gpio free, when the driver is unbinded or rmmod,
we miss the gpio free.
This patch uses managed API to request the reset gpio, so that the
gpio could be freed properly.
Fixes: 984203ceff27 ("net: stmmac: mdio: remove reset gpio free") Signed-off-by: Jisheng Zhang <Jisheng.Zhang@synaptics.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Dan Carpenter [Wed, 22 May 2019 08:45:13 +0000 (11:45 +0300)]
mISDN: make sure device name is NUL terminated
The user can change the device_name with the IMSETDEVNAME ioctl, but we
need to ensure that the user's name is NUL terminated. Otherwise it
could result in a buffer overflow when we copy the name back to the user
with IMGETDEVINFO ioctl.
I also changed two strcpy() calls which handle the name to strscpy().
Hopefully, there aren't any other ways to create a too long name, but
it's nice to do this as a kernel hardening measure.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Claudiu Beznea [Wed, 22 May 2019 08:24:43 +0000 (08:24 +0000)]
net: macb: save/restore the remaining registers and features
SAMA5D2 SoC has a suspend mode where SoC's power is cut off. Due to this
the registers content is lost after a suspend/resume cycle. The current
suspend/resume implementation covers some of these registers. However
there are few which were not treated (e.g. SCRT2 and USRIO). Apart
from this, netdev features are not restored. Treat these issues.
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
net/tls: fix device surprise removal with offload
This series fixes two issues with device surprise removal.
First we need to take a read lock around resync, otherwise
netdev notifier handler may clear the structures from under
our feet.
Secondly we need to be careful about the interpretation
of device features. Offload has to be properly cleaned
up even if the TLS device features got cleared after
connection state was installed.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Wed, 22 May 2019 02:02:02 +0000 (19:02 -0700)]
net/tls: don't ignore netdev notifications if no TLS features
On device surprise removal path (the notifier) we can't
bail just because the features are disabled. They may
have been enabled during the lifetime of the device.
This bug leads to leaking netdev references and
use-after-frees if there are active connections while
device features are cleared.
Fixes: e8f69799810c ("net/tls: Add generic NIC offload infrastructure") Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Wed, 22 May 2019 02:02:01 +0000 (19:02 -0700)]
net/tls: fix state removal with feature flags off
TLS offload drivers shouldn't (and currently don't) block
the TLS offload feature changes based on whether there are
active offloaded connections or not.
This seems to be a good idea, because we want the admin to
be able to disable the TLS offload at any time, and there
is no clean way of disabling it for active connections
(TX side is quite problematic). So if features are cleared
existing connections will stay offloaded until they close,
and new connections will not attempt offload to a given
device.
However, the offload state removal handling is currently
broken if feature flags get cleared while there are
active TLS offloads.
RX side will completely bail from cleanup, even on normal
remove path, leaving device state dangling, potentially
causing issues when the 5-tuple is reused. It will also
fail to release the netdev reference.
Remove the RX-side warning message, in next release cycle
it should be printed when features are disabled, rather
than when connection dies, but for that we need a more
efficient method of finding connection of a given netdev
(a'la BPF offload code).
Fixes: 4799ac81e52a ("tls: Add rx inline crypto offload") Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Wed, 22 May 2019 02:02:00 +0000 (19:02 -0700)]
net/tls: avoid NULL-deref on resync during device removal
When netdev with active kTLS sockets in unregistered
notifier callback walks the offloaded sockets and
cleans up offload state. RX data may still be processed,
however, and if resync was requested prior to device
removal we would hit a NULL pointer dereference on
ctx->netdev use.
Make sure resync is under the device offload lock
and NULL-check the netdev pointer.
This should be safe, because the pointer is set to
NULL either in the netdev notifier (under said lock)
or when socket is completely dead and no resync can
happen.
The other access to ctx->netdev in tls_validate_xmit_skb()
does not dereference the pointer, it just checks it against
other device pointer, so it should be pretty safe (perhaps
we can add a READ_ONCE/WRITE_ONCE there, if paranoid).
Fixes: 4799ac81e52a ("tls: Add rx inline crypto offload") Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
This set adds documentation for TLS offload. It starts
by making the networking documentation a little easier
to navigate by hiding driver docs a little deeper.
It then RSTifys the existing Kernel TLS documentation.
Last but not least TLS offload documentation is added.
This should help vendors navigate the TLS offload, and
help ensure different implementations stay aligned from
user perspective.
v2:
- address Alexei's and Boris'es commands on patch 3.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Wed, 22 May 2019 01:57:14 +0000 (18:57 -0700)]
Documentation: add TLS offload documentation
Describe existing kernel TLS offload (added back in Linux 4.19) -
the mechanism, the expected behavior and the notable corner cases.
This documentation is mostly targeting hardware vendors who want
to implement offload, to ensure consistency between implementations.
v2:
- add emphasis around TLS_SW/TLS_HW/TLS_HW_RECORD;
- remove mentions of ongoing work (Boris);
- split the flow of data in SW vs. HW cases in TX overview
(Boris);
- call out which fields are updated by the device and which
are filled by the stack (Boris);
- move error handling into it's own section (Boris);
- add more words about fallback (Boris);
- note that checksum validation is required (Alexei);
- note that drivers shouldn't pay attention to the TLS
device features.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Acked-by: Dave Watson <davejwatson@fb.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Boris Pismenny <borisp@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Wed, 22 May 2019 01:57:13 +0000 (18:57 -0700)]
Documentation: tls: RSTify the ktls documentation
Convert the TLS doc to RST. Use C code blocks for the code
samples, and mark hyperlinks.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Acked-by: Dave Watson <davejwatson@fb.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Wed, 22 May 2019 01:57:12 +0000 (18:57 -0700)]
Documentation: net: move device drivers docs to a submenu
Some of the device drivers have really long document titles
making the networking table of contents hard to look through.
Place vendor drivers under a submenu.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Acked-by: Dave Watson <davejwatson@fb.com> Acked-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Claudiu Manoil [Tue, 21 May 2019 16:52:55 +0000 (19:52 +0300)]
ocelot: Dont allocate another multicast list, use __dev_mc_sync
Doing kmalloc in atomic context is always an issue,
more so for a list that can grow significantly.
Turns out that the driver only uses the duplicated
list of multicast mac addresses to keep track of
what addresses to delete from h/w before committing
the new list from kernel to h/w back again via set_rx_mode,
every time this list gets updated by the kernel.
Given that the h/w knows how to add and delete mac addresses
based on the mac address value alone, __dev_mc_sync should be
the much better choice of kernel API for these operations
avoiding the considerable overhead of maintaining a duplicated
list in the driver.
Signed-off-by: Claudiu Manoil <claudiu.manoil@nxp.com> Tested-by: Alexandre Belloni <alexandre.belloni@bootlin.com> Acked-by: Alexandre Belloni <alexandre.belloni@bootlin.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Validate required parameters in inet6_validate_link_af
inet6_set_link_af requires that at least one of IFLA_INET6_TOKEN or
IFLA_INET6_ADDR_GET_MODE is passed. If none of them is passed, it
returns -EINVAL, which may cause do_setlink() to fail in the middle of
processing other commands and give the following warning message:
A link change request failed with some changes committed already.
Interface eth0 may have been left with an inconsistent configuration,
please check.
Check the presence of at least one of them in inet6_validate_link_af to
detect invalid parameters at an early stage, before do_setlink does
anything. Also validate the address generation mode at an early stage.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Linus Torvalds [Wed, 22 May 2019 15:36:16 +0000 (08:36 -0700)]
Merge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Pull arm64 fixes from Will Deacon:
- Fix SPE probe failure when backing auxbuf with high-order pages
- Fix handling of DMA allocations from outside of the vmalloc area
- Fix generation of build-id ELF section for vDSO object
- Disable huge I/O mappings if kernel page table dumping is enabled
- A few other minor fixes (comments, kconfig etc)
* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
arm64: vdso: Explicitly add build-id option
arm64/mm: Inhibit huge-vmap with ptdump
arm64: Print physical address of page table base in show_pte()
arm64: don't trash config with compat symbol if COMPAT is disabled
arm64: assembler: Update comment above cond_yield_neon() macro
drivers/perf: arm_spe: Don't error on high-order pages for aux buf
arm64/iommu: handle non-remapped addresses in ->mmap and ->get_sgtable
1) Clear up some recent tipc regressions because of registration
ordering. Fix from Junwei Hu.
2) tipc's TLV_SET() can read past the end of the supplied buffer during
the copy. From Chris Packham.
3) ptp example program doesn't match the kernel, from Richard Cochran.
4) Outgoing message type fix in qrtr, from Bjorn Andersson.
5) Flow control regression in stmmac, from Tan Tee Min.
6) Fix inband autonegotiation in phylink, from Russell King.
7) Fix sk_bound_dev_if handling in rawv6_bind(), from Mike Manning.
8) Fix usbnet crash after disconnect, from Kloetzke Jan.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (21 commits)
usbnet: fix kernel crash after disconnect
selftests: fib_rule_tests: use pre-defined DEV_ADDR
net-next: net: Fix typos in ip-sysctl.txt
ipv6: Consider sk_bound_dev_if when binding a raw socket to an address
net: phylink: ensure inband AN works correctly
usbnet: ipheth: fix racing condition
net: stmmac: dma channel control register need to be init first
net: stmmac: fix ethtool flow control not able to get/set
net: qrtr: Fix message type of outgoing packets
networking: : fix typos in code comments
ptp: Fix example program to match kernel.
fddi: fix typos in code comments
selftests: fib_rule_tests: enable forwarding before ipv4 from/iif test
selftests: fib_rule_tests: fix local IPv4 address typo
tipc: Avoid copying bytes beyond the supplied data
2/2] net: xilinx_emaclite: use readx_poll_timeout() in mdio wait function
1/2] net: axienet: use readx_poll_timeout() in mdio wait function
vlan: Mark expected switch fall-through
macvlan: Mark expected switch fall-through
net/mlx4_en: ethtool, Remove unsupported SFP EEPROM high pages query
...
Linus Torvalds [Wed, 22 May 2019 15:10:35 +0000 (08:10 -0700)]
Merge tag 'for-5.2/dm-fix-1' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm
Pull device mapper fix from Mike Snitzer:
"Fix a particularly glaring oversight in a DM core commit from 5.1 that
doesn't properly trim special IOs (e.g. discards) relative to
corresponding target's max_io_len_target_boundary()"
* tag 'for-5.2/dm-fix-1' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
dm: make sure to obey max_io_len_target_boundary
Commit 4d207133e9c3 changed the types of the statistic values in struct
gfs2_lkstats from s64 to u64. Because of that, what should be a signed
value in gfs2_update_stats turned into an unsigned value. When shifted
right, we end up with a large positive value instead of a small negative
value, which results in an incorrect variance estimate.
Fixes: 4d207133e9c3 ("gfs2: Make statistics unsigned, suitable for use with do_div()") Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> Cc: stable@vger.kernel.org # v4.4+
Florian Westphal [Tue, 21 May 2019 11:24:32 +0000 (13:24 +0200)]
netfilter: nft_flow_offload: don't offload when sequence numbers need adjustment
We can't deal with tcp sequence number rewrite in flow_offload.
While at it, simplify helper check, we only need to know if the extension
is present, we don't need the helper data.
Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Florian Westphal [Tue, 21 May 2019 11:24:31 +0000 (13:24 +0200)]
netfilter: nft_flow_offload: set liberal tracking mode for tcp
Without it, whenever a packet has to be pushed up the stack (e.g. because
of mtu mismatch), then conntrack will flag packets as invalid, which in
turn breaks NAT.
Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Michael Lass [Tue, 21 May 2019 19:58:07 +0000 (21:58 +0200)]
dm: make sure to obey max_io_len_target_boundary
Commit 61697a6abd24 ("dm: eliminate 'split_discard_bios' flag from DM
target interface") incorrectly removed code from
__send_changing_extent_only() that is required to impose a per-target IO
boundary on IO that exceeds max_io_len_target_boundary(). Otherwise
"special" IO (e.g. DISCARD, WRITE SAME, WRITE ZEROES) can write beyond
where allowed.
Fix this by restoring the max_io_len_target_boundary() limit in
__send_changing_extent_only()
Fixes: 61697a6abd24 ("dm: eliminate 'split_discard_bios' flag from DM target interface") Cc: stable@vger.kernel.org # 5.1+ Signed-off-by: Michael Lass <bevan@bi-co.net> Signed-off-by: Mike Snitzer <snitzer@redhat.com>
The crash happens roughly 125..130ms after the disconnect. This
correlates with the 'delay' timer that is started on certain USB tx/rx
errors in the URB completion handler.
The problem is a race of usbnet_stop() with usbnet_start_xmit(). In
usbnet_stop() we call usbnet_terminate_urbs() to cancel all URBs in
flight. This only makes sense if no new URBs are submitted
concurrently, though. But the usbnet_start_xmit() can run at the same
time on another CPU which almost unconditionally submits an URB. The
error callback of the new URB will then schedule the timer after it was
already stopped.
The fix adds a check if the tx queue is stopped after the tx list lock
has been taken. This should reliably prevent the submission of new URBs
while usbnet_terminate_urbs() does its job. The same thing is done on
the rx side even though it might be safe due to other flags that are
checked there.
Signed-off-by: Jan Klötzke <Jan.Kloetzke@preh.de> Signed-off-by: David S. Miller <davem@davemloft.net>
Hangbin Liu [Tue, 21 May 2019 06:40:47 +0000 (14:40 +0800)]
selftests: fib_rule_tests: use pre-defined DEV_ADDR
DEV_ADDR is defined but not used. Use it in address setting.
Do the same with IPv6 for consistency.
Reported-by: David Ahern <dsahern@gmail.com> Fixes: fc82d93e57e3 ("selftests: fib_rule_tests: fix local IPv4 address typo") Signed-off-by: Hangbin Liu <liuhangbin@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
projects / kernel.git / log