Vasundhara Volam [Tue, 11 Mar 2014 13:23:08 +0000 (18:53 +0530)]
be2net: Fix vlans_added counter
When a VLAN is added by user, adapter->vlans_added is incremented.
But if the VLAN is already programmed in HW, driver ends up
incrementing the counter wrongly.
Increment the counter only if VLAN is not already programmed in the HW.
Signed-off-by: Vasundhara Volam <vasundhara.volam@emulex.com> Signed-off-by: Sathya Perla <sathya.perla@emulex.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Vasundhara Volam [Tue, 11 Mar 2014 13:23:07 +0000 (18:53 +0530)]
be2net: Create multiple TXQs on RSS capable multi-channel BE3-R interfaces
Currently the driver creates only a single TXQ on any BE3-R multi-channel
interface.
This patch changes this and creates multiple TXQs on RSS-capable multi-channel
BE3-R interfaces. This change helps improve the TX pps performance on the
affected interface.
Signed-off-by: Vasundhara Volam <vasundhara.volam@emulex.com> Signed-off-by: Sathya Perla <sathya.perla@emulex.com> Signed-off-by: David S. Miller <davem@davemloft.net>
The allocation size must be be_max_uc() and not "be_max_uc() + 1" Signed-off-by: Ravikumar Nelavelli <ravikumar.nelavelli@emulex.com> Signed-off-by: Sathya Perla <sathya.perla@emulex.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Ravikumar Nelavelli <ravikumar.nelavelli@emulex.com> Signed-off-by: Sathya Perla <sathya.perla@emulex.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Suresh Reddy [Tue, 11 Mar 2014 13:23:04 +0000 (18:53 +0530)]
be2net: Add link state control for VFs
Add support to control VF's link state by implementing the
ndo_set_vf_link_state() hook.
Signed-off-by: Suresh Reddy <suresh.reddy@emulex.com> Signed-off-by: Sathya Perla <sathya.perla@emulex.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Suresh Reddy [Tue, 11 Mar 2014 13:23:03 +0000 (18:53 +0530)]
be2net: Use GET_PROFILE_CONFIG cmd for BE3-R to query max-vfs
Use GET_PROFILE_CONFIG_V1 cmd even for BE3-R (it's already used for
Lancer-R and Skyhawk-R), to query max-vfs value supported by the FW.
This is needed as on some configs, the value exported in the PCI-config
space is not accurate.
Signed-off-by: Suresh Reddy <suresh.reddy@emulex.com> Signed-off-by: Sathya Perla <sathya.perla@emulex.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Wed, 12 Mar 2014 19:57:26 +0000 (15:57 -0400)]
Merge branch 'mlx4-next'
Or Gerlitz says:
====================
mlx4: Add SRIOV support for RoCE
This series adds SRIOV support for RoCE (RDMA over Ethernet) to the mlx4 driver.
The patches are against net-next, as of commit 2d8d40a "pkt_sched: fq:
do not hold qdisc lock while allocating memory"
changes from V1:
- addressed feedback from Dave on patch #3 and changed get_real_sgid_index()
to be called fill_in_real_sgid_index() and be a void function.
- removed some checkpatch warnings on long lines
changes from V0:
- always check the return code of mlx4_get_roce_gid_from_slave().
The call we fixed is introduced in patch #1 and later removed by
patch #3 that allows guests to have multiple GIDS. The 1..3
separation was done for proper division of patches to logical changes.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jack Morgenstein [Wed, 12 Mar 2014 10:00:43 +0000 (12:00 +0200)]
mlx4: Activate RoCE/SRIOV
To activate RoCE/SRIOV, need to remove the following:
1. In mlx4_ib_add, need to remove the error return preventing
initialization of a RoCE port under SRIOV.
2. In update_vport_qp_params (in resource_tracker.c) need to remove
the error return when a RoCE RC or UD qp is detected.
This error return causes the INIT-to-RTR qp transition to fail
in the wrapper function under RoCE/SRIOV.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Shani Michaelli [Wed, 12 Mar 2014 10:00:42 +0000 (12:00 +0200)]
mlx4_ib: Fix SIDR support of for UD QPs under SRIOV/RoCE
* Handle CM_SIDR_REQ_ATTR_ID and CM_SIDR_REP_ATTR_ID
in multiplex_cm_handler and demux_cm_handler.
* Handle Service ID Resolution messages and REQ messages
separately, for their formats are different.
Signed-off-by: Shani Michaeli <shanim@mellanox.com> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jack Morgenstein [Wed, 12 Mar 2014 10:00:41 +0000 (12:00 +0200)]
mlx4: Implement IP based gids support for RoCE/SRIOV
Since there is no connection between the MAC/VLAN and the GID
when using IP-based addressing, the proxy QP1 (running on the
slave) must pass the source-mac, destination-mac, and vlan_id
information separately from the GID. Additionally, the Host
must pass the remote source-mac and vlan_id back to the slave,
This is achieved as follows:
Outgoing MADs:
1. Source MAC: obtained from the CQ completion structure
(struct ib_wc, smac field).
2. Destination MAC: obtained from the tunnel header
3. vlan_id: obtained from the tunnel header.
Incoming MADs
1. The source (i.e., remote) MAC and vlan_id are passed in
the tunnel header to the proxy QP1.
VST mode support:
For outgoing MADs, the vlan_id obtained from the header is
discarded, and the vlan_id specified by the Hypervisor is used
instead.
For incoming MADs, the incoming vlan_id (in the wc) is discarded, and the
"invalid" vlan (0xffff) is substituted when forwarding to the slave.
Signed-off-by: Moni Shoua <monis@mellanox.co.il> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jack Morgenstein [Wed, 12 Mar 2014 10:00:40 +0000 (12:00 +0200)]
mlx4: Add ref counting to port MAC table for RoCE
The IB side of RoCE requires the MAC table index of the
MAC address used by its QPs.
To obtain the real MAC index, the IB side registers the
MAC (increasing its ref count, and also returning the
real MAC index) during the modify-qp sequence.
This protects against the ETH side deleting or modifying
that MAC table entry while the QP is active.
Note that until the modify-qp command returns success,
the MAC and VLAN information only has "candidate" status.
If the modify-qp succeeds, the "candidate" info is promoted
to the operational MAC/VLAN info for the qp. If the modify fails,
the candidate MAC/VLAN is unregistered, and the old qp info
is preserved.
The patch is a bit complex, because there are multiple qp
transitions where the primary-path information may be
modified: INIT-to-RTR, and SQD-to-SQD.
Similarly for the alternate path information.
Therefore the code must handle cases where path information
has already been entered into the QP context by previous
qp transitions.
For the MAC address, the success logic is as follows:
1. If there was no previous MAC, simply move the candidate
MAC information to the operational information, and reset
the candidate MAC info.
2. If there was a previous MAC, unregister it. Then move
the MAC information from candidate to operational, and
reset the candidate info (as in 1. above).
The MAC address failure logic is the same for all cases:
- Unregister the candidate MAC, and reset the candidate MAC info.
For Vlan registration, the logic is similar.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jack Morgenstein [Wed, 12 Mar 2014 10:00:39 +0000 (12:00 +0200)]
mlx4: In RoCE allow guests to have multiple GIDS
The GIDs are statically distributed, as follows:
PF: gets 16 GIDs
VFs: Remaining GIDS are divided evenly between VFs activated by the driver.
If the division is not even, lower-numbered VFs get an extra GID.
For an IB interface, the number of gids per guest remains as before: one gid per guest.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jack Morgenstein [Wed, 12 Mar 2014 10:00:38 +0000 (12:00 +0200)]
mlx4_core: For RoCE, allow slaves to set the GID entry at that slave's index
For IB transport, the host determines the slave GIDs. For ETH (RoCE),
however, the slave's GID is determined by the IP address that the slave
itself assigns to the ETH device used by RoCE.
In this case, the slave must be able to write its GIDs to the HCA gid table
(at the GID indices that slave "owns").
This commit adds processing for the SET_PORT_GID_TABLE opcode modifier
for the SET_PORT command wrapper (so that slaves may modify their GIDS
for RoCE).
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jack Morgenstein [Wed, 12 Mar 2014 10:00:37 +0000 (12:00 +0200)]
mlx4: Adjust QP1 multiplexing for RoCE/SRIOV
This requires the following modifications:
1. Fix build_mlx4_header to properly fill in the ETH fields
2. Adjust mux and demux QP1 flow to support RoCE.
This commit still assumes only one GID per slave for RoCE.
The commit enabling multiple GIDs is a subsequent commit, and
is done separately because of its complexity.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Wed, 12 Mar 2014 19:53:54 +0000 (15:53 -0400)]
Merge branch 'tipc'
Jon Maloy says:
====================
tipc: simplifications in socket and port layer
After the removal of the tipc native API the relation between
tipc_port and its API types is strictly one-to-one, i.e, the latter
can now only be a socket API. This change opens up for
simplifications both in the code, data and locking structure.
We start with this series, where we ensure that port and socket
structures are co-allocated. Note that the first commit in the
series is unrelated to the above.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jon Paul Maloy [Wed, 12 Mar 2014 15:31:13 +0000 (11:31 -0400)]
tipc: eliminate redundant lookups in registry
As an artefact from the native interface, the message sending functions
in the port takes a port ref as first parameter, and then looks up in
the registry to find the corresponding port pointer. This despite the
fact that the only currently existing caller, tipc_sock, already knows
this pointer.
We change the signature of these functions to take a struct tipc_port*
argument, and remove the redundant lookups.
We also remove an unmotivated extra lookup in the function
socket.c:auto_connect(), and, as the lookup functions tipc_port_deref()
and ref_deref() now become unused, we remove these two functions.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Reviewed-by: Ying Xue <ying.xue@windriver.com> Reviewed-by: Erik Hugne <erik.hugne@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jon Paul Maloy [Wed, 12 Mar 2014 15:31:12 +0000 (11:31 -0400)]
tipc: align usage of variable names and macros in socket
The practice of naming variables in TIPC is inconistent, sometimes
even within the same file.
In this commit we align variable names and declarations within
socket.c, and function and macro names within socket.h. We also
reduce the number of conversion macros to two, in order to make
usage less obsure.
These changes are purely cosmetic.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Reviewed-by: Ying Xue <ying.xue@windriver.com> Reviewed-by: Erik Hugne <erik.hugne@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jon Paul Maloy [Wed, 12 Mar 2014 15:31:11 +0000 (11:31 -0400)]
tipc: eliminate redundant locking
The three functions tipc_portimportance(), tipc_portunreliable() and
tipc_portunreturnable() and their corresponding tipc_set* functions,
are all grabbing port_lock when accessing the targeted port. This is
unnecessary in the current code, since these calls only are made from
within socket downcalls, already protected by sock_lock.
We remove the redundant locking. Also, since the functions now become
trivial one-liners, we move them to port.h and make them inline.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Reviewed-by: Ying Xue <ying.xue@windriver.com> Reviewed-by: Erik Hugne <erik.hugne@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jon Paul Maloy [Wed, 12 Mar 2014 15:31:10 +0000 (11:31 -0400)]
tipc: eliminate upcall function pointers between port and socket
Due to the original one-to-many relation between port and user API
layers, upcalls to the API have been performed via function pointers,
installed in struct tipc_port at creation. Since this relation now
always is one-to-one, we can instead use ordinary function calls.
We remove the function pointers 'dispatcher' and ´wakeup' from
struct tipc_port, and replace them with calls to the renamed
functions tipc_sk_rcv() and tipc_sk_wakeup().
At the same time we change the name and signature of the functions
tipc_createport() and tipc_deleteport() to reflect their new role
as mere initialization/destruction functions.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Reviewed-by: Ying Xue <ying.xue@windriver.com> Reviewed-by: Erik Hugne <erik.hugne@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jon Paul Maloy [Wed, 12 Mar 2014 15:31:09 +0000 (11:31 -0400)]
tipc: aggregate port structure into socket structure
After the removal of the tipc native API the relation between
a tipc_port and its API types is strictly one-to-one, i.e, the
latter can now only be a socket API. There is therefore no need
to allocate struct tipc_port and struct sock independently.
In this commit, we aggregate struct tipc_port into struct tipc_sock,
hence saving both CPU cycles and structure complexity.
There are no functional changes in this commit, except for the
elimination of the separate allocation/freeing of tipc_port.
All other changes are just adaptatons to the new data structure.
This commit also opens up for further code simplifications and
code volume reduction, something we will do in later commits.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Reviewed-by: Ying Xue <ying.xue@windriver.com> Reviewed-by: Erik Hugne <erik.hugne@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jon Paul Maloy [Wed, 12 Mar 2014 15:31:08 +0000 (11:31 -0400)]
tipc: remove redundant 'peer_name' field in struct tipc_sock
The field 'peer_name' in struct tipc_sock is redundant, since
this information already is available from tipc_port, to which
tipc_sock has a reference.
We remove the field, and ensure that peer node and peer port
info instead is fetched via the functions that already exist
for this purpose.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Reviewed-by: Ying Xue <ying.xue@windriver.com> Reviewed-by: Erik Hugne <erik.hugne@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jon Paul Maloy [Wed, 12 Mar 2014 15:31:07 +0000 (11:31 -0400)]
tipc: replace reference table rwlock with spinlock
The lock for protecting the reference table is declared as an
RWLOCK, although it is only used in write mode, never in read
mode.
We redefine it to become a spinlock.
Signed-off-by: Jon Maloy <jon.maloy@ericsson.com> Reviewed-by: Ying Xue <ying.xue@windriver.com> Reviewed-by: Erik Hugne <erik.hugne@ericsson.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Steffen Klassert [Wed, 12 Mar 2014 08:43:17 +0000 (09:43 +0100)]
flowcache: Fix resource leaks on namespace exit.
We leak an active timer, the hotcpu notifier and all allocated
resources when we exit a namespace. Fix this by introducing a
flow_cache_fini() function where we release the resources before
we exit.
Fixes: ca925cf1534e ("flowcache: Make flow cache name space aware") Reported-by: Jakub Kicinski <moorray3@wp.pl> Tested-by: Jakub Kicinski <moorray3@wp.pl> Cc: Eric Dumazet <eric.dumazet@gmail.com> Cc: Fan Du <fan.du@windriver.com> Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com> Acked-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Claudiu Manoil [Tue, 11 Mar 2014 16:01:24 +0000 (18:01 +0200)]
gianfar: Fix multi-queue support checks @probe()
priv is not instantiated at gfar_of_init() time, when
parsing the DT for info on supported HW queues. Before
the netdev can be allocated, the number of supported
queues must be known. Because the number of supported
queues depends on device type, move the compatibility
checks before netdev allocation. Local vars are used
to hold the operation mode info before netdev allocation.
This fixes the null accesses for priv->.., in gfar_of_init.
Signed-off-by: Claudiu Manoil <claudiu.manoil@freescale.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Tue, 11 Mar 2014 00:11:43 +0000 (17:11 -0700)]
pkt_sched: add cond_resched() to class and qdisc dump
We have seen delays of more than 50ms in class or qdisc dumps, in case
device is under high TX stress, even with the prior 4KB per skb limit.
Add cond_resched() to give a chance to higher prio tasks to get cpu.
Signed-off-by; Eric Dumazet <edumazet@google.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
netdev: set __percpu attribute on netdev_alloc_pcpu_stats
This patch fixes sparse warnings in vlan driver.
It propagates the sparse __percpu attribute from alloc_percpu
into netdev_alloc_pcpu_stats. I expect it may trigger additional
sparse warnings from other drivers that are missing the __percpu
attribute.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Cong Wang <cwang@twopensource.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Tobias Klauser [Mon, 10 Mar 2014 12:12:23 +0000 (13:12 +0100)]
net: eth: cpsw: Use net_device_stats from struct net_device
Instead of using an own copy of struct net_device_stats in struct
cpsw_priv, use stats from struct net_device. Also remove the thus
unnecessary .ndo_get_stats function, as it just returns dev->stats,
which is the default.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch> Acked-by: Mugunthan V N <mugunthanvnm@ti.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Mon, 10 Mar 2014 14:09:07 +0000 (07:09 -0700)]
flowcache: restore a single flow_cache kmem_cache
It is not legal to create multiple kmem_cache having the same name.
flowcache can use a single kmem_cache, no need for a per netns
one.
Fixes: ca925cf1534e ("flowcache: Make flow cache name space aware") Reported-by: Jakub Kicinski <moorray3@wp.pl> Tested-by: Jakub Kicinski <moorray3@wp.pl> Tested-by: Fan Du <fan.du@windriver.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Gu Zheng [Mon, 10 Mar 2014 01:57:34 +0000 (09:57 +0800)]
net: add a pre-check of net_ns in sk_change_net()
We do not need to switch the net_ns if the target net_ns the same
as the current one, so here we add a pre-check of net_ns to avoid
this as David suggested.
Signed-off-by: Gu Zheng <guz.fnst@cn.fujitsu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Haiyang Zhang [Sun, 9 Mar 2014 23:10:59 +0000 (16:10 -0700)]
hyperv: Change the receive buffer size for legacy hosts
Due to a bug in the Hyper-V host verion 2008R2, we need to use a slightly smaller
receive buffer size, otherwise the buffer will not be accepted by the legacy hosts.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
KY Srinivasan [Sun, 9 Mar 2014 03:23:18 +0000 (19:23 -0800)]
Drivers: net: hyperv: Enable large send offload
Enable segmentation offload.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
KY Srinivasan [Sun, 9 Mar 2014 03:23:17 +0000 (19:23 -0800)]
Drivers: net: hyperv: Enable send side checksum offload
Enable send side checksum offload.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
KY Srinivasan [Sun, 9 Mar 2014 03:23:16 +0000 (19:23 -0800)]
Drivers: net: hyperv: Enable receive side IP checksum offload
Enable receive side checksum offload.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
KY Srinivasan [Sun, 9 Mar 2014 03:23:15 +0000 (19:23 -0800)]
Drivers: net: hyperv: Enable offloads on the host
Prior to enabling guest side offloads, enable the offloads on the host.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
KY Srinivasan [Sun, 9 Mar 2014 03:23:14 +0000 (19:23 -0800)]
Drivers: net: hyperv: Cleanup the send path
In preparation for enabling offloads, cleanup the send path.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
KY Srinivasan [Sun, 9 Mar 2014 03:23:13 +0000 (19:23 -0800)]
Drivers: net: hyperv: Enable scatter gather I/O
Cleanup the code and enable scatter gather I/O.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com> Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Tim Harvey [Sat, 8 Mar 2014 04:59:53 +0000 (20:59 -0800)]
sky2: allow mac to come from dt
The driver reads the mac address from the device registers which would
need to have been programmed by the bootloader. This patch adds
the ability to pull the mac from devicetree via the pci device dt node.
Signed-off-by: Tim Harvey <tharvey@gateworks.com> Cc: netdev@vger.kernel.org Cc: devicetree@vger.kernel.org Cc: Grant Likely <grant.likely@linaro.org> Cc: Rob Herring <robh+dt@kernel.org>
Changes since v2:
- eliminated use of stack tmpaddr per feedback
Changes since v1:
- simplified based on feedback
- fixed formatting Acked-by: Rob Herring <robh@kernel.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Fixes: 31c70d5956fc ("l2tp: keep original skb ownership") Reported-by: kbuild test robot <fengguang.wu@intel.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
IB/mlx5_core: remove unreachable function call in module init
The call to mlx5_health_cleanup() in the module init function can never
be reached. Removing it.
Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Eli Cohen <eli@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Fri, 7 Mar 2014 20:02:33 +0000 (12:02 -0800)]
netlink: autosize skb lengthes
One known problem with netlink is the fact that NLMSG_GOODSIZE is
really small on PAGE_SIZE==4096 architectures, and it is difficult
to know in advance what buffer size is used by the application.
This patch adds an automatic learning of the size.
First netlink message will still be limited to ~4K, but if user used
bigger buffers, then following messages will be able to use up to 16KB.
This speedups dump() operations by a large factor and should be safe
for legacy applications.
Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Thomas Graf <tgraf@suug.ch> Acked-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 10 Mar 2014 17:17:44 +0000 (13:17 -0400)]
Merge branch 'gianfar-next'
Claudiu Manoil says:
====================
gianfar: Tx timeout issue
There's an older Tx timeout issue showing up on etsec2 devices
with 2 CPUs. I pinned this issue down to processing overhead
incurred by supporting multiple Tx/Rx rings, as explained in
the 2nd patch below. But before this, there's also a concurency
issue leading to Rx/Tx spurrious interrupts, addressed by the
'Tx NAPI' patch below.
The Tx timeout can be triggered with multiple Tx flows,
'iperf -c -N 8' commands, on a 2 CPUs etsec2 based (P1020) board.
Claudiu Manoil [Fri, 7 Mar 2014 12:42:46 +0000 (14:42 +0200)]
gianfar: Use Single-Queue polling for "fsl,etsec2"
For the "fsl,etsec2" compatible models the driver currently
supports 8 Tx and Rx DMA rings (aka HW queues). However, there
are only 2 pairs of Rx/Tx interrupt lines, as these controllers
are integrated in low power SoCs with 2 CPUs at most. As a result,
there are at most 2 NAPI instances that have to service multiple
Tx and Rx queues for these devices. This complicates the NAPI
polling routine having to iterate over the mutiple Rx/Tx queues
hooked to the same interrupt lines. And there's also an overhead
at HW level, as the controller needs to service all the 8 Tx rings
in a round robin manner. The combined overhead shows up for multi
parallel Tx flows transmitted by the kernel stack, when the driver
usually starts returning NETDEV_TX_BUSY leading to NETDEV WATCHDOG
Tx timeout triggering if the Tx path is congested for too long.
As an alternative, this patch makes the driver support only one
Tx/Rx DMA ring per NAPI instance (per interrupt group or pair
of Tx/Rx interrupt lines) by default. The simplified single queue
polling routine (gfar_poll_sq) will be the default napi poll routine
for the etsec2 devices too. Some adjustments needed to be made to
link the Tx/Rx HW queues with each NAPI instance (2 in this case).
The gfar_poll_sq() is already successfully used by older SQ_SG_MODE
(single interrupt group) controllers.
This patch fixes Tx timeout triggering under heavy Tx traffic load
(i.e. iperf -c -P 8) for the "fsl,etsec2" (currently the only
MQ_MG_MODE devices). There's also a significant memory footprint
reduction by supporting 2 Rx/Tx DMA rings (at most), instead of 8,
for these devices.
Signed-off-by: Claudiu Manoil <claudiu.manoil@freescale.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Claudiu Manoil [Fri, 7 Mar 2014 12:42:45 +0000 (14:42 +0200)]
gianfar: Separate out the Tx interrupt handling (Tx NAPI)
There are some concurrency issues on devices w/ 2 CPUs related
to the handling of Rx and Tx interrupts. eTSEC has separate
interrupt lines for Rx and Tx but a single imask register
to mask these interrupts and a single NAPI instance to handle
both Rx and Tx work. As a result, the Rx and Tx ISRs are
identical, both are invoking gfar_schedule_cleanup(), however
both handlers can be entered at the same time when the Rx and
Tx interrupts are taken by different CPUs. In this case
spurrious interrupts (SPU) show up (in /proc/interrupts)
indicating a concurrency issue. Also, Tx overruns followed
by Tx timeout have been observed under heavy Tx traffic load.
To address these issues, the schedule cleanup ISR part has
been changed to handle the Rx and Tx interrupts independently.
The patch adds a separate NAPI poll routine for Tx cleanup to
be triggerred independently by the Tx confirmation interrupts
only. Existing poll functions are modified to handle only
the Rx path processing. The Tx poll routine does not need a
budget, since Tx processing doesn't consume NAPI budget, and
hence it is registered with minimum NAPI weight.
NAPI scheduling does not require locking since there are
different NAPI instances between the Rx and Tx confirmation
paths now.
So, the patch fixes the occurence of spurrious Rx/Tx interrupts.
Tx overruns also occur less frequently now.
Signed-off-by: Claudiu Manoil <claudiu.manoil@freescale.com> Signed-off-by: David S. Miller <davem@davemloft.net>
dingtianhong [Fri, 7 Mar 2014 10:45:31 +0000 (18:45 +0800)]
vlan: use use ether_addr_equal_64bits to instead of ether_addr_equal
Ether_addr_equal_64bits is more efficient than ether_addr_equal, and
can be used when each argument is an array within a structure that
contains at least two bytes of data beyond the array, so it is safe
to use it for vlan, and make sense for fast path.
Cc: Joe Perches <joe@perches.com> Cc: Patrick McHardy <kaber@trash.net> Cc: David S. Miller <davem@davemloft.net> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
dingtianhong [Fri, 7 Mar 2014 10:45:30 +0000 (18:45 +0800)]
vlan: slight optimization for vlan_do_receive()
According Joe's suggestion, maybe it'd be faster to add an unlikely to
the test for PCKET_OTHERHOST, so I add it and see whether the performance
could be better, although the differences is so small and negligible, but
it is hard to catch that any lower device would set the skb type to
PACKET_OTHERHOST, so most of time, I think it make sense to add unlikely
for the test.
Cc: Joe Perches <joe@perches.com> Cc: Patrick McHardy <kaber@trash.net> Cc: David S. Miller <davem@davemloft.net> Signed-off-by: Ding Tianhong <dingtianhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Fri, 7 Mar 2014 06:57:52 +0000 (22:57 -0800)]
pkt_sched: fq: do not hold qdisc lock while allocating memory
Resizing fq hash table allocates memory while holding qdisc spinlock,
with BH disabled.
This is definitely not good, as allocation might sleep.
We can drop the lock and get it when needed, we hold RTNL so no other
changes can happen at the same time.
Signed-off-by: Eric Dumazet <edumazet@google.com> Fixes: afe4fd062416 ("pkt_sched: fq: Fair Queue packet scheduler") Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Sat, 8 Mar 2014 23:49:29 +0000 (18:49 -0500)]
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-next
Jeff Kirsher says:
====================
Intel Wired LAN Driver Updates
This series contains updates to e1000e, ixgbevf and igb.
Majority of this series contains fixes and cleanups to e1000e,
most notably are:
Todd provides a fix to PTP in e1000e which adds a lock in e1000e_phc_adjfreq
to prevent concurrent changes to TIMINCA and SYSTIMH/L. Then provides an
igb fix to use ARRAY_SIZE for array size calculation.
David provides the remaining e1000e which contain:
- cleanup of pointer references that are no longer used
- fix an issue on systems with Management Engine enabled with the
ethernet cable unplugged
- fix an issue on 82579 where enabling EEE LPI sooner than one second
after link up causes link issues on some switches
- refactor the power management flows to prevent the suspend path from
being executed twice when hibernating
- refactor the runtime power management to fix interfering with the
functionality of Energy Efficient Ethernet when enabled and to fix
the device from repeatedly flip between suspend and resume with the
interface administratively downed
- enable the feature PHY Ultra Low Power Mode which is a power saving
feature that reduces the power consumption of the PHY when a cable is
not connected
- fix the ethtool offline tests for 82579 parts
- fix SHRA register access for 82579 parts which was introduced by
previous commit c3a0dce35af0 "e1000e: fix overrun of PHY RAR array"
Florian provides a fix for ixgbevf where skb->pkt_type was being checked
like a bitmask, but it is not a bitmask.
Fix an issue reported by Stephen Hemminger where there was a warning
about code defined but never used if IGB_HWMON is not defined.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jeff Kirsher [Thu, 6 Mar 2014 05:28:06 +0000 (05:28 +0000)]
igb: fix warning if !CONFIG_IGB_HWMON
Fix warning about code defined but never used if IGB_HWMON not defined.
Reported-by: Stephen Hemminger <stephen@networkplumber.org> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Florian Fainelli [Fri, 28 Feb 2014 23:46:49 +0000 (15:46 -0800)]
ixgbevf: fix skb->pkt_type checks
skb->pkt_type is not a bitmask, but contains only value at a time from
the range defined in include/uapi/linux/if_packet.h.
Checking it like if it was a bitmask of values would also cause
PACKET_OTHERHOST, PACKET_LOOPBACK and PACKET_FASTROUTE to be matched by
this check since their lower 2 bits are also set, although that does not
fix a real bug, it is still potentially confusing.
This bogus check was introduced in commit 815cccbf ("ixgbe: add setlink,
getlink support to ixgbe and ixgbevf").
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
David Ertman [Wed, 5 Mar 2014 07:50:46 +0000 (07:50 +0000)]
e1000e: Fix SHRA register access for 82579
Previous commit c3a0dce35af0 fixed an overrun for the RAR on i218 devices.
This commit also attempted to homogenize the RAR/SHRA access for all parts
accessed by the e1000e driver. This change introduced an error for
assigning MAC addresses to guest OS's for 82579 devices.
Only RAR[0] is accessible to the driver for 82579 parts, and additional
addresses must be placed into the SHRA[L|H] registers. The rar_entry_count
was changed in the previous commit to an inaccurate value that accounted
for all RAR and SHRA registers, not just the ones usable by the driver.
This patch fixes the count to the correct value and adjusts the
e1000_rar_set_pch2lan() function to user the correct index.
Cc: John Greene <jogreene@redhat.com> Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
David Ertman [Wed, 5 Mar 2014 07:54:19 +0000 (07:54 +0000)]
e1000e: Fix ethtool offline tests for 82579 parts
Changes to the rar_entry_count value require a change to the indexing
used to access the SHRA[H|L] registers when testing them with
'ethtool -t <iface> offline'
Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
David Ertman [Wed, 26 Feb 2014 00:02:24 +0000 (00:02 +0000)]
e1000e: Fix not generating an error on invalid load parameter
Valid values for InterruptThrottleRate are 10-100000, or one of
0, 1, 3, 4. '2' is not valid. This is a legacy from the branching
from the e1000 driver code that e1000e was based from.
Prior to this patch, if the e1000e driver was loaded with a forced
invalid InterruptThrottleRate of '2', then no throttle rate would be
set and no error message generated.
Now, a message will be generated that an invalid value was used and the
value for InterruptThrottleRate will be set to the default value.
Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
David Ertman [Sat, 22 Feb 2014 03:15:17 +0000 (03:15 +0000)]
e1000e: Feature Enable PHY Ultra Low Power Mode (ULP)
ULP is a power saving feature that reduces the power consumption of the
PHY when a cable is not connected.
ULP is gated on the following conditions:
1) The hardware must support ULP. Currently this is only I218
devices from Intel
2) ULP is initiated by the driver, so, no driver results in no ULP.
3) ULP's implementation utilizes Runtime Power Management to toggle its
execution. ULP is enabled/disabled based on the state of Runtime PM.
4) ULP is not active when wake-on-unicast, multicast or broadcast is active
as these features are mutually-exclusive.
Since the PHY is in an unavailable state while ULP is active, any access
of the PHY registers will fail. This is resolved by utilizing kernel
calls that cause the device to exit Runtime PM (e.g. pm_runtime_get_sync)
and then, after PHY access is complete, allow the device to resume
Runtime PM (e.g. pm_runtime_put_sync).
Under certain conditions, toggling the LANPHYPC is necessary to disable
ULP mode. Break out existing code to toggle LANPHYPC to a new function
to avoid code duplication.
Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com> Cc: Bruce Allan <bruce.w.allan@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
David Ertman [Fri, 14 Feb 2014 07:16:46 +0000 (07:16 +0000)]
e1000e Refactor of Runtime Power Management
Fix issues with:
RuntimePM causing the device to repeatedly flip between suspend and resume
with the interface administratively downed.
Having RuntimePM enabled interfering with the functionality of Energy
Efficient Ethernet.
Added checks to disallow functions that should not be executed if the
device is currently runtime suspended
Make runtime_idle callback to use same deterministic behavior as the igb
driver.
Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com> Acked-by: Bruce Allan <bruce.w.allan@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
David Ertman [Fri, 14 Feb 2014 07:16:41 +0000 (07:16 +0000)]
e1000e: Refactor PM flows
Refactor the system power management flows to prevent the suspend path from
being executed twice when hibernating since both the freeze and
poweroff callbacks were set to e1000_suspend() via SET_SYSTEM_SLEEP_PM_OPS.
There are HW workarounds that are performed during this flow and calling
them twice was causing erroneous behavior.
Re-arrange the code to take advantage of common code paths and explicitly
set the individual dev_pm_ops callbacks for suspend, resume, freeze,
thaw, poweroff and restore.
Add a boolean parameter (reset) to the e1000e_down function to allow
for cases when the HW should not be reset when downed during a PM event.
Now that all suspend/shutdown paths result in a call to __e1000_shutdown()
that checks Wake on Lan status, removing redundant check for WoL in
e1000_power_down_phy().
Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com> Acked-by: Bruce Allan <bruce.w.allan@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
David Ertman [Wed, 5 Feb 2014 01:09:54 +0000 (01:09 +0000)]
e1000e: Add missing branding strings in ich8lan.c
Branding strings from recently released and soon to be released
hardware configurations that are supported by e1000e.
Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com> Acked-by: Bruce Allan <bruce.w.allan@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
David Ertman [Tue, 4 Feb 2014 01:56:06 +0000 (01:56 +0000)]
e1000e: Cleanup - Update GPL header and Copyright
This patch is to update the GPL header by removing the portion that
refers to the Free Software Foundation address.
Change the copyright date for 2014.
Reformat the header comments to conform to kernel networking coding norms
Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
David Ertman [Fri, 24 Jan 2014 23:07:48 +0000 (23:07 +0000)]
e1000e: Fix 82579 sets LPI too early.
Enabling EEE LPI sooner than one second after link up on 82579 causes link
issues with some switches.
Remove EEE enablement for 82579 parts from the link initialization flow to
avoid initializing too early. EEE initialization for 82579 will be done
in e1000e_update_phy_task.
Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com> Acked-by: Bruce W Allan <bruce.w.allan@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
On a ME enabled system with the cable out, the driver init flow would
generate an erroneous message indicating that resets were being blocked
by an active ME session. Cause was ME clearing the semaphore bit to
block further PHY resets for up to 50 msec during power-on/cycle. After
this interval, ME would re-set the bit and allow PHY resets.
To resolve this, change the flow of e1000e_phy_hw_reset_generic() to
utilize a delay and retry method. Poll the FWSM register to minimize
any extra time added to the flow. If the delay times out at 100ms
(checked in 10msec increments), then return the value E1000_BLK_PHY_RESET,
as this is the accurate state of the PHY. Attempting to alter just the
call to e1000e_phy_hw_reset_generic() in e1000_init_phy_workarounds_pchlan()
just caused the problem to move further down the flow.
Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com> Acked-by: Bruce W. Allan <bruce.w.allan@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
David Ertman [Wed, 22 Jan 2014 00:21:41 +0000 (00:21 +0000)]
e1000e: Cleanup unecessary references
Cleaning up some pointer references that are no longer necessary
Signed-off-by: Dave Ertman <davidx.m.ertman@intel.com> Tested-by: Jeff Pieper <jeffrey.e.pieper@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
this is a pull request of twelve patches for net-next/master.
Alexander Shiyan contributes two patches for the mcp251x, one making
the driver more quiet and the other one improves the compile time
coverage by removing the #ifdef CONFIG_PM_SLEEP. Then two patches for
the flexcan driver by me, one removing the #ifdef CONFIG_PM_SLEEP, too,
the other one making use of platform_get_device_id(). Another patch by
me which converts the janz-ican3 driver to use netdev_<level>(). The
remaining 7 patches are by Oliver Hartkopp, they add CAN FD support to
the netlink configuration interface.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 7 Mar 2014 21:24:54 +0000 (16:24 -0500)]
Merge branch 'r8152'
Hayes Wang says:
====================
r8152: tx/rx improvement
- Select the suitable spin lock for each function.
- Add additional check to reduce the spin lock.
- Up the priority of the tx to avoid interrupted by rx.
- Support rx checksum, large send, and IPv6 hw checksum.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
hayeswang [Fri, 7 Mar 2014 03:04:40 +0000 (11:04 +0800)]
r8152: support IPv6
Support hw IPv6 checksum for TCP and UDP packets.
Note that the hw has the limitation of the range of the transport
offset. Besides, the TCP Pseudo Header of the IPv6 TSO of the hw
bases on the Microsoft document which excludes the packet length.
Signed-off-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
hayeswang [Fri, 7 Mar 2014 03:04:36 +0000 (11:04 +0800)]
r8152: up the priority of the transmission
move the tx_bottom() from delayed_work to tasklet. It makes the rx
and tx balanced. If the device is in runtime suspend when getting
the tx packet, wakeup the device before trasmitting.
Signed-off-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
hayeswang [Fri, 7 Mar 2014 03:04:34 +0000 (11:04 +0800)]
r8152: replace spin_lock_irqsave and spin_unlock_irqrestore
Use spin_lock and spin_unlock in interrupt context.
The ndo_start_xmit would not be called in interrupt context, so
replace the relative spin_lock_irqsave and spin_unlock_irqrestore
with spin_lock_bh and spin_unlock_bh.
Signed-off-by: Hayes Wang <hayeswang@realtek.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 7 Mar 2014 21:12:43 +0000 (16:12 -0500)]
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-next
Jeff Kirsher says:
====================
Intel Wired LAN Driver Updates
This series contains updates to i40e and i40evf.
Most notable are:
Joseph completes the implementation of the ethtool ntuple rule
management interface by adding the get, update and delete interface
reset.
Akeem provides a fix to prevent a possible overflow due to multiplication
of number and size by using kzalloc, so use kcalloc.
Jesse provides an implementation for skb_set_hash() and adds the L4 type
return when we know it is an L4 hash. He also adds a counter to
statistics for Tx timeouts to help users. Lastly he provides a change
to stay away from the cache line where the done bit may be getting
written back for the transmit ring since the hardware may be writing the
whole cache line for a partial update.
Shannon cleans up code comments.
Anjali removes a firmware workaround for newer firmware since the number
of MSIx vectors are being reported correctly.
v2:
- dropped patch 01 of the series based on feedback from the author
Joe Perches and Shannon Nelson.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 7 Mar 2014 21:08:59 +0000 (16:08 -0500)]
Merge tag 'rxrpc-devel-20140304' of git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs
David Howells says:
====================
net-next: AF_RXRPC fixes and development
Here are some AF_RXRPC fixes:
(1) Fix to remove incorrect checksum calculation made during recvmsg(). It's
unnecessary to try to do this there since we check the checksum before
reading the RxRPC header from the packet.
(2) Fix to prevent the sending of an ABORT packet in response to another
ABORT packet and inducing a storm.
(3) Fix UDP MTU calculation from parsing ICMP_FRAG_NEEDED packets where we
don't handle the ICMP packet not specifying an MTU size.
And development patches:
(4) Add sysctls for configuring RxRPC parameters, specifically various delays
pertaining to ACK generation, the time before we resend a packet for
which we don't receive an ACK, the maximum time a call is permitted to
live and the amount of time transport, connection and dead call
information is cached.
(5) Improve ACK packet production by adjusting the handling of ACK_REQUESTED
packets, ignoring the MORE_PACKETS flag, delaying the production of
otherwise immediate ACK_IDLE packets and delaying all ACK_IDLE production
(barring the call termination) to half a second.
(6) Add more sysctl parameters to expose the Rx window size, the maximum
packet size that we're willing to receive and the number of jumbo rxrpc
packets we're willing to handle in a single UDP packet.
(7) Request ACKs on alternate DATA packets so that the other side doesn't
wait till we fill up the Tx window.
(8) Use a RCU hash table to look up the rxrpc_call for an incoming packet
rather than stepping through a hierarchy involving several spinlocks.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Fri, 7 Mar 2014 20:57:47 +0000 (15:57 -0500)]
Merge branch 'xen-netback-next'
Zoltan Kiss says:
====================
xen-netback: TX grant mapping with SKBTX_DEV_ZEROCOPY instead of copy
A long known problem of the upstream netback implementation that on the TX
path (from guest to Dom0) it copies the whole packet from guest memory into
Dom0. That simply became a bottleneck with 10Gb NICs, and generally it's a
huge perfomance penalty. The classic kernel version of netback used grant
mapping, and to get notified when the page can be unmapped, it used page
destructors. Unfortunately that destructor is not an upstreamable solution.
Ian Campbell's skb fragment destructor patch series [1] tried to solve this
problem, however it seems to be very invasive on the network stack's code,
and therefore haven't progressed very well.
This patch series use SKBTX_DEV_ZEROCOPY flags to tell the stack it needs to
know when the skb is freed up. That is the way KVM solved the same problem,
and based on my initial tests it can do the same for us. Avoiding the extra
copy boosted up TX throughput from 6.8 Gbps to 7.9 (I used a slower AMD
Interlagos box, both Dom0 and guest on upstream kernel, on the same NUMA node,
running iperf 2.0.5, and the remote end was a bare metal box on the same 10Gb
switch)
Based on my investigations the packet get only copied if it is delivered to
Dom0 IP stack through deliver_skb, which is due to this [2] patch. This affects
DomU->Dom0 IP traffic and when Dom0 does routing/NAT for the guest. That's a bit
unfortunate, but luckily it doesn't cause a major regression for this usecase.
In the future we should try to eliminate that copy somehow.
There are a few spinoff tasks which will be addressed in separate patches:
- grant copy the header directly instead of map and memcpy. This should help
us avoiding TLB flushing
- use something else than ballooned pages
- fix grant map to use page->index properly
I've tried to broke it down to smaller patches, with mixed results, so I
welcome suggestions on that part as well:
1: Use skb->cb to store pending_idx
2: Some refactoring
3: Change RX path for mapped SKB fragments (moved here to keep bisectability,
review it after #4)
4: Introduce TX grant mapping
5: Remove old TX grant copy definitons and fix indentations
6: Add stat counters for zerocopy
7: Handle guests with too many frags
8: Timeout packets in RX path
9: Aggregate TX unmap operations
v2: I've fixed some smaller things, see the individual patches. I've added a
few new stat counters, and handling the important use case when an older guest
sends lots of slots. Instead of delayed copy now we timeout packets on the RX
path, based on the assumption that otherwise packets should get stucked
anywhere else. Finally some unmap batching to avoid too much TLB flush
v3: Apart from fixing a few things mentioned in responses the important change
is the use the hypercall directly for grant [un]mapping, therefore we can
avoid m2p override.
v4: Now we are using a new grant mapping API to avoid m2p_override. The RX queue
timeout logic changed also.
v5: Only minor fixes based on Wei's comments
v6: Important bugfixes for xenvif_poll exit path and zerocopy callback, see
first 2 patches. Also rework of handling packets with too many slots, and
reorder the series a bit.
v7: Small fixes in comments/log messages/error paths, and merging the frag
overflow stats patch into its parent.
Zoltan Kiss [Thu, 6 Mar 2014 21:48:31 +0000 (21:48 +0000)]
xen-netback: Aggregate TX unmap operations
Unmapping causes TLB flushing, therefore we should make it in the largest
possible batches. However we shouldn't starve the guest for too long. So if
the guest has space for at least two big packets and we don't have at least a
quarter ring to unmap, delay it for at most 1 milisec.
Signed-off-by: Zoltan Kiss <zoltan.kiss@citrix.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Zoltan Kiss [Thu, 6 Mar 2014 21:48:30 +0000 (21:48 +0000)]
xen-netback: Timeout packets in RX path
A malicious or buggy guest can leave its queue filled indefinitely, in which
case qdisc start to queue packets for that VIF. If those packets came from an
another guest, it can block its slots and prevent shutdown. To avoid that, we
make sure the queue is drained in every 10 seconds.
The QDisc queue in worst case takes 3 round to flush usually.
Signed-off-by: Zoltan Kiss <zoltan.kiss@citrix.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Zoltan Kiss [Thu, 6 Mar 2014 21:48:29 +0000 (21:48 +0000)]
xen-netback: Handle guests with too many frags
Xen network protocol had implicit dependency on MAX_SKB_FRAGS. Netback has to
handle guests sending up to XEN_NETBK_LEGACY_SLOTS_MAX slots. To achieve that:
- create a new skb
- map the leftover slots to its frags (no linear buffer here!)
- chain it to the previous through skb_shinfo(skb)->frag_list
- map them
- copy and coalesce the frags into a brand new one and send it to the stack
- unmap the 2 old skb's pages
It's also introduces new stat counters, which help determine how often the guest
sends a packet with more than MAX_SKB_FRAGS frags.
NOTE: if bisect brought you here, you should apply the series up until
"xen-netback: Timeout packets in RX path", otherwise malicious guests can block
other guests by not releasing their sent packets.
Signed-off-by: Zoltan Kiss <zoltan.kiss@citrix.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Zoltan Kiss [Thu, 6 Mar 2014 21:48:28 +0000 (21:48 +0000)]
xen-netback: Add stat counters for zerocopy
These counters help determine how often the buffers had to be copied. Also
they help find out if packets are leaked, as if "sent != success + fail",
there are probably packets never freed up properly.
NOTE: if bisect brought you here, you should apply the series up until
"xen-netback: Timeout packets in RX path", otherwise Windows guests can't work
properly and malicious guests can block other guests by not releasing their sent
packets.
Signed-off-by: Zoltan Kiss <zoltan.kiss@citrix.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Zoltan Kiss [Thu, 6 Mar 2014 21:48:27 +0000 (21:48 +0000)]
xen-netback: Remove old TX grant copy definitons and fix indentations
These became obsolete with grant mapping. I've left intentionally the
indentations in this way, to improve readability of previous patches.
NOTE: if bisect brought you here, you should apply the series up until
"xen-netback: Timeout packets in RX path", otherwise Windows guests can't work
properly and malicious guests can block other guests by not releasing their sent
packets.
Signed-off-by: Zoltan Kiss <zoltan.kiss@citrix.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Zoltan Kiss [Thu, 6 Mar 2014 21:48:26 +0000 (21:48 +0000)]
xen-netback: Introduce TX grant mapping
This patch introduces grant mapping on netback TX path. It replaces grant copy
operations, ditching grant copy coalescing along the way. Another solution for
copy coalescing is introduced in "xen-netback: Handle guests with too many
frags", older guests and Windows can broke before that patch applies.
There is a callback (xenvif_zerocopy_callback) from core stack to release the
slots back to the guests when kfree_skb or skb_orphan_frags called. It feeds a
separate dealloc thread, as scheduling NAPI instance from there is inefficient,
therefore we can't do dealloc from the instance.
Signed-off-by: Zoltan Kiss <zoltan.kiss@citrix.com> Signed-off-by: David S. Miller <davem@davemloft.net>
projects / linux-beck.git / log