This patch defines a new ETHTOOL_GLINKSETTINGS/SLINKSETTINGS API,
handled by the new get_link_ksettings/set_link_ksettings callbacks.
This API provides support for most legacy ethtool_cmd fields, adds
support for larger link mode masks (up to 4064 bits, variable length),
and removes ethtool_cmd deprecated
fields (transceiver/maxrxpkt/maxtxpkt).
This API is deprecating the legacy ETHTOOL_GSET/SSET API and provides
the following backward compatibility properties:
- legacy ethtool with legacy drivers: no change, still using the
get_settings/set_settings callbacks.
- legacy ethtool with new get/set_link_ksettings drivers: the new
driver callbacks are used, data internally converted to legacy
ethtool_cmd. ETHTOOL_GSET will return only the 1st 32b of each link
mode mask. ETHTOOL_SSET will fail if user tries to set the
ethtool_cmd deprecated fields to
non-0 (transceiver/maxrxpkt/maxtxpkt). A kernel warning is logged if
driver sets higher bits.
- future ethtool with legacy drivers: no change, still using the
get_settings/set_settings callbacks, internally converted to new data
structure. Deprecated fields (transceiver/maxrxpkt/maxtxpkt) will be
ignored and seen as 0 from user space. Note that that "future"
ethtool tool will not allow changes to these deprecated fields.
- future ethtool with new drivers: direct call to the new callbacks.
By "future" ethtool, what is meant is:
- query: first try ETHTOOL_GLINKSETTINGS, and revert to ETHTOOL_GSET if
fails
- set: query first and remember which of ETHTOOL_GLINKSETTINGS or
ETHTOOL_GSET was successful
+ if ETHTOOL_GLINKSETTINGS was successful, then change config with
ETHTOOL_SLINKSETTINGS. A failure there is final (do not try
ETHTOOL_SSET).
+ otherwise ETHTOOL_GSET was successful, change config with
ETHTOOL_SSET. A failure there is final (do not try
ETHTOOL_SLINKSETTINGS).
The interaction user/kernel via the new API requires a small
ETHTOOL_GLINKSETTINGS handshake first to agree on the length of the link
mode bitmaps. If kernel doesn't agree with user, it returns the bitmap
length it is expecting from user as a negative length (and cmd field is
0). When kernel and user agree, kernel returns valid info in all
fields (ie. link mode length > 0 and cmd is ETHTOOL_GLINKSETTINGS).
Data structure crossing user/kernel boundary is 32/64-bit
agnostic. Converted internally to a legal kernel bitmap.
The internal __ethtool_get_settings kernel helper will gradually be
replaced by __ethtool_get_link_ksettings by the time the first
"link_settings" drivers start to appear. So this patch doesn't change
it, it will be removed before it needs to be changed.
Signed-off-by: David Decotigny <decot@googlers.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch add the SO_CNX_ADVICE socket option (setsockopt only). The
purpose is to allow an application to give feedback to the kernel about
the quality of the network path for a connected socket. The value
argument indicates the type of quality report. For this initial patch
the only supported advice is a value of 1 which indicates "bad path,
please reroute"-- the action taken by the kernel is to call
dst_negative_advice which will attempt to choose a different ECMP route,
reset the TX hash for flow label and UDP source port in encapsulation,
etc.
This facility should be useful for connected UDP sockets where only the
application can provide any feedback about path quality. It could also
be useful for TCP applications that have additional knowledge about the
path outside of the normal TCP control loop.
Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, all ipv6 addresses are flushed when the interface is configured
down, including global, static addresses:
$ ip -6 addr show dev eth1
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 state UP qlen 1000
inet6 2100:1::2/120 scope global
valid_lft forever preferred_lft forever
inet6 fe80::e0:f9ff:fe79:34bd/64 scope link
valid_lft forever preferred_lft forever
$ ip link set dev eth1 down
$ ip -6 addr show dev eth1
<< nothing; all addresses have been flushed>>
Add a new sysctl to make this behavior optional. The new setting defaults to
flush all addresses to maintain backwards compatibility. When the set global
addresses with no expire times are not flushed on an admin down. The sysctl
is per-interface or system-wide for all interfaces
$ sysctl -w net.ipv6.conf.eth1.keep_addr_on_down=1
or
$ sysctl -w net.ipv6.conf.all.keep_addr_on_down=1
Will keep addresses on eth1 on an admin down.
$ ip -6 addr show dev eth1
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 state UP qlen 1000
inet6 2100:1::2/120 scope global
valid_lft forever preferred_lft forever
inet6 fe80::e0:f9ff:fe79:34bd/64 scope link
valid_lft forever preferred_lft forever
$ ip link set dev eth1 down
$ ip -6 addr show dev eth1
3: eth1: <BROADCAST,MULTICAST> mtu 1500 state DOWN qlen 1000
inet6 2100:1::2/120 scope global tentative
valid_lft forever preferred_lft forever
inet6 fe80::e0:f9ff:fe79:34bd/64 scope link tentative
valid_lft forever preferred_lft forever
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
msg.dst_sk needs to be set up with a valid socket because some callbacks
later derive the netns from it.
Fixes: 263ea09084d172d ("Revert "genl: Add genlmsg_new_unicast() for unicast message allocation")
Reported-by: Jon Maloy <maloy@donjonn.com>
Bisected-by: Jon Maloy <maloy@donjonn.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When the TIPC module is unloaded, we have identified a race condition
that allows a node reference counter to go to zero and the node instance
being freed before the node timer is finished with accessing it. This
leads to occasional crashes, especially in multi-namespace environments.
The scenario goes as follows:
CPU0:(node_stop) CPU1:(node_timeout) // ref == 2
1: if(!mod_timer())
2: if (del_timer())
3: tipc_node_put() // ref -> 1
4: tipc_node_put() // ref -> 0
5: kfree_rcu(node);
6: tipc_node_get(node)
7: // BOOM!
We now clean up this functionality as follows:
1) We remove the node pointer from the node lookup table before we
attempt deactivating the timer. This way, we reduce the risk that
tipc_node_find() may obtain a valid pointer to an instance marked
for deletion; a harmless but undesirable situation.
2) We use del_timer_sync() instead of del_timer() to safely deactivate
the node timer without any risk that it might be reactivated by the
timeout handler. There is no risk of deadlock here, since the two
functions never touch the same spinlocks.
3: We remove a pointless tipc_node_get() + tipc_node_put() from the
timeout handler.
Reported-by: Zhijiang Hu <huzhijiang@gmail.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Although we have never seen it happen, we have identified the
following problematic scenario when nodes are stopped and deleted:
CPU0: CPU1:
tipc_node_xxx() //ref == 1
tipc_node_put() //ref -> 0
tipc_node_find() // node still in table
tipc_node_delete()
list_del_rcu(n. list)
tipc_node_get() //ref -> 1, bad
kfree_rcu()
tipc_node_put() //ref to 0 again.
kfree_rcu() // BOOM!
We fix this by introducing use of the conditional kref_get_if_not_zero()
instead of kref_get() in the function tipc_node_find(). This eliminates
any risk of post-mortem access.
Reported-by: Zhijiang Hu <huzhijiang@gmail.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Yuval Mintz says:
====================
qed*: Driver updates
Usually I try to provide a sensible description of the patch set even if
it lacks a general 'motif', but this simply contains several small,
unrelated and self-explenatory tweaks and additions.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Drop the `QL4xxx 40G/100G' and use `FastLinQ 4xxxx' instead.
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't allow driver to probe on an adapter at a failed state;
Gracefully block the probe instead.
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Module is using a binary firmware file and so should be marked as such.
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are several corner cases where driver might get a 2nd notification
about the same link change. Don't log any additional changes if the
physical carrier is already reported as it should.
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There's a corner-case in HW where an SKB queued for transmission that
contains too many frags will cause FW to assert.
This patch solves this by linearizing the SKB if necessary.
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The device ID for the 10g module has changed. Populate the pci_ids table
accordingly.
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Introduce the use of kobj_to_dev() helper function instead of open
coding it with container_of()
The Coccinelle semantic patch used to make this change is as follows:
//<smpl>
@@
expression a;
symbol kobj;
@@
- container_of(a, struct device, kobj)
+ kobj_to_dev(a)
//</smpl>
Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Convert a call to init_timer and accompanying intializations of
the timer's data and function fields to a call to setup_timer.
The Coccinelle semantic patch that fixes this problem is
as follows:
// <smpl>
@@
expression t,f,d;
@@
-init_timer(&t);
+setup_timer(&t,f,d);
...
-t.data = d;
-t.function = f;
// </smpl>
Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Convert a call to init_timer and accompanying intializations of
the timer's data and function fields to a call to setup_timer.
The Coccinelle semantic patch that fixes this problem is
as follows:
// <smpl>
@@
expression t,f,d;
@@
-init_timer(&t);
+setup_timer(&t,f,d);
-t.data = d;
-t.function = f;
// </smpl>
Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Convert a call to init_timer and accompanying intializations of
the timer's data and function fields to a call to setup_timer.
The Coccinelle semantic patch that fixes this problem is
as follows:
// <smpl>
@@
expression t,f,d;
@@
-init_timer(&t);
+setup_timer(&t,f,d);
-t.data = d;
-t.function = f;
// </smpl>
Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Yangbo Lu says:
====================
gianfar: Add PTP support for ls1021a platform
This patchset is to enable ptp support for ls1021a platform. The endianness
issue in gianfar driver and gianfar ptp driver must be fixed, and a 1588
timer node must be added into dts.
Changes for v2:
- Modified commit message
- Added more reviewers
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix endianness for the 64-bit hardware timestamp value with
be64_to_cpu to support both PowerPC platforms and ARM platforms.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace get_of_u32 with standard helper function of_property_read_u32
since the latter can process cpu endianness.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add the 1588 timer node for ls1021a platform to
support gianfar ptp driver.
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A bug was introduced in the merge commit b633353115 ("Merge
git://git.kernel.org/pub/scm/linux/kernel/git/davem/net")
The generic marvell_config_init (and therefore marvell_of_reg_init) is
not called anymore for the Marvell 88E1510 (in net-next).
This patch calls marvell_config_init and moves the specific init
function for the 88E1510 below the marvell_config_init function to avoid
adding a function predeclaration.
Signed-off-by: Clemens Gruber <clemens.gruber@pqgruber.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Vivien Didelot says:
====================
net: dsa: add port VLAN dump operation
The VLAN GetNext approach is specific to some switches and thus hard to
implement for others. This patchset replaces it with a simpler port VLAN dump
operation, similar to the corresponding FDB operation.
The mv88e6xxx driver is the only one currently affected by the change.
The documentation is updated accordingly.
Note: this patchset uses http://www.spinics.net/lists/kernel/msg2186705.html
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
The VLAN GetNext operation is specific to some switches, and thus can be
complicated to implement for some drivers.
Remove the support for the vlan_getnext/port_pvid_get approach in favor
of the generic and simpler port_vlan_dump function.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove the port_pvid_get and vlan_getnext functions in favor of a
simpler mv88e6xxx_port_vlan_dump function.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Similar to port_fdb_dump, add a port_vlan_dump function to DSA drivers
which gets passed the switchdev VLAN object and callback.
This function, if implemented, takes precedence over the soon legacy
vlan_getnext/port_pvid_get approach.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Benc says:
====================
vxlan: consolidate rx handling
Currently, vxlan_rcv is just called at the end of vxlan_udp_encap_recv,
continuing the rx processing where vxlan_udp_encap_recv left it. There's no
clear border between those two functions. This patchset moves
vxlan_udp_encap_recv and vxlan_rcv into a single function.
This also allows to do some simplification in error path.
The VXLAN-GPE implementation that will follow up this set can be seen at:
https://github.com/jbenc/linux-vxlan/commits/master
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Now when the packet is scrubbed early, the metadata_dst can be assigned to
the skb as soon as it is allocated. This simplifies the error cleanup path,
as the dst will be freed by kfree_skb. It is also not necessary to pass it
as a parameter to functions anymore.
Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now when both vxlan_udp_encap_recv and vxlan_rcv are much shorter, combine
them into a single function.
Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This code will be different for VXLAN-GPE, so move it to a separate
function. It will also make the rx path less spaghetti-like.
Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now when the packet is scrubbed early, skb->mark can be set in the GBP
handling code.
Signed-off-by: Jiri Benc <jbenc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Cong Wang says:
====================
net_sched: add network namespace support for tc actions
This patchset adds network namespace support for tc actions.
v2:
* pull the first patch into net-next
* reduce code duplication by introducing more helper functions
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently tc actions are stored in a per-module hashtable,
therefore are visible to all network namespaces. This is
probably the last part of the tc subsystem which is not
aware of netns now. This patch makes them per-netns,
several tc action API's need to be adjusted for this.
The tc action API code is ugly due to historical reasons,
we need to refactor that code in the future.
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We only release the memory of the hashtable itself, not its
entries inside. This is not a problem yet since we only call
it in module release path, and module is refcount'ed by
actions. This would be a problem after we move the per module
hinfo into per netns in the latter patch.
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
* Split big conditional statement.
* Check (data.length <= CCP_MAX_OPTION_LENGTH) only once.
* Don't read ccp_option[1] if not initialised.
Reading uninitialised ccp_option[1] was harmless, because this could
only happen when data.length was 0 or 1. So even then, we couldn't pass
the (ccp_option[1] < 2 || ccp_option[1] > data.length) test anyway.
Signed-off-by: Guillaume Nault <g.nault@alphalink.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
When CONFIG_GENEVE is built as a loadable module, and bnx2x is built-in,
we get this link error:
drivers/net/built-in.o: In function `bnx2x_open':
:(.text+0x33322): undefined reference to `geneve_get_rx_port'
drivers/net/built-in.o: In function `bnx2x_sp_rtnl_task':
:(.text+0x3e632): undefined reference to `geneve_get_rx_port'
This avoids the problem by adding a separate Kconfig symbol named
CONFIG_BNX2X_GENEVE that is only enabled when the code is
reachable from the driver.
This is the same trick that BNX2X does for VXLAN support, and
is similar to how I40E handles both.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: 883ce97d25 ("bnx2x: Add Geneve inner-RSS support")
Acked-By: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Calling dev_close() causes IFF_UP to be cleared which will remove the
interfaces routes and some addresses. That's probably not what the user
intended when running the offline selftest. Besides this does not happen
if the interface is brought down before the test, so the current
behaviour is inconsistent.
Instead call the net_device_ops ndo_stop function directly and avoid
touching IFF_UP at all.
Signed-off-by: Stefan Assmann <sassmann@kpanic.de>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Problem: When switching off VLAN offloading on an i350, the VLAN
interface gets unusable. For testing, set up a VLAN on an i350
and some remote machine, e.g.:
$ ip link add link eth0 name eth0.42 type vlan id 42
$ ip addr add 192.168.42.1/24 dev eth0.42
$ ip link set dev eth0.42 up
Offloading is switched on by default:
$ ethtool -k eth0 | grep vlan-offload
rx-vlan-offload: on
tx-vlan-offload: on
$ ping -c 3 -I eth0.42 192.168.42.2
[...works as usual...]
Now switch off VLAN offloading and try again:
$ ethtool -K eth0 rxvlan off
Actual changes:
rx-vlan-offload: off
tx-vlan-offload: off [requested on]
$ ping -c 3 -I eth0.42 192.168.42.2
PING 192.168.42.2 (192.168.42.2) from 192.168.42.1 eth0.42: 56(84) bytes of da
ta.
--- 192.168.42.2 ping statistics ---
3 packets transmitted, 0 received, 100% packet loss, time 1999ms
I can only reproduce it on an i350, the above works fine on a 82580.
While inspecting the igb source, I came across the code in igb_set_vmolr
which sets the E1000_VMOLR_STRVLAN/E1000_DVMOLR_STRVLAN flags once and
for all, and in all of the igb code there's no other place where the
STRVLAN is set or cleared. Thus, VLAN stripping is enabled in igb
unconditionally, independently of the offloading setting.
I compared that to the latest Intel igb-5.3.3.5 driver from
http://sourceforge.net/projects/e1000/ which in fact sets and clears the
STRVLAN flag independently from igb_set_vmolr in its own function
igb_set_vf_vlan_strip, depending on the vlan settings.
So I included the STRVLAN handling from the igb-5.3.3.5 driver into our
current igb driver and tested the above scenario again. This time ping
still works after switching off VLAN offloading.
Tested on i350, with and without addtional VFs, as well as on 82580
successfully.
Signed-off-by: Corinna Vinschen <vinschen@redhat.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
A similar issue was addressed a few years ago in the following thread:
http://www.spinics.net/lists/netdev/msg245877.html
At that time there were concerns that removing this statement may cause other
side effects. However the submitter addressed those concerns. But the dialogue
went cold. We have a new case where a customers application is registering and
un-registering multicast addresses every few seconds. This is leading to many
"Link is Up" messages in the logs as a result of the
"netif_carrier_off(netdev)" statement called by igbvf_msix_other(). Also on
some kernels it is interfering with the bonding driver causing it to failover
and subsequently affecting connectivity.
The Sourgeforge driver does not make this call and is therefore not affected.
If there were any side effects I would expect that driver to also be affected.
I have tested re-loading the igbvf driver and downing the adapter with the PF
entity on the host where the VM has this patch. When I bring it back up again
connectivity is restored as expected. Therefore I request that this patch gets
submitted.
Signed-off-by: Jon Maxwell <jmaxwell37@gmail.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
This patch adds support for generic Tx checksums to the igbvf driver. It
turns out this is actually pretty easy after going over the datasheet as we
were doing a number of steps we didn't need to.
In order to perform a Tx checksum for an L4 header we need to fill in the
following fields in the Tx descriptor:
MACLEN (maximum of 127), retrieved from:
skb_network_offset()
IPLEN (maximum of 511), retrieved from:
skb_checksum_start_offset() - skb_network_offset()
TUCMD.L4T indicates offset and if checksum or crc32c, based on:
skb->csum_offset
The added advantage to doing this is that we can support inner checksum
offloads for tunnels and MPLS while still being able to transparently
insert VLAN tags.
I also took the opportunity to clean-up many of the feature flag
configuration bits to make them a bit more consistent between drivers. In
the case of the VF drivers this meant adding support for SCTP CRCs, and
inner checksum offloads for MPLS and various tunnel types.
Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
