On 5/8/18 1:48 PM, 吉藤英明 wrote:
> Hi,
>
> 2018-05-08 15:41 GMT+09:00 Andre Naujoks :
>> On 08.05.2018 08:31, 吉藤英明 wrote:
>>> Hi,
>>>
>>> 2018-05-08 15:03 GMT+09:00 Andre Naujoks :
On 11.04.2018 13:02, Andre Naujoks wrote:
> Hi.
Hi again.
Since it has been a month now,
From: Mark Bloch
For now, only add L2_TUNNEL_TO_L2 option. This will allow to perform
generic decap operation if the encapsulating protocol is L2 based, and the
inner packet is also L2 based. For example this can be used to decap VXLAN
packets.
Signed-off-by: Mark Bloch
Signed-off-by: Leon
From: Mark Bloch
There are RX and TX flow steering namespaces with different number of
actions. Initialize them accordingly.
Signed-off-by: Mark Bloch
Reviewed-by: Saeed Mahameed
Reviewed-by: Or Gerlitz
Signed-off-by: Leon Romanovsky
---
drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c | 5
From: Björn Töpel
This commit adds proper MEM_TYPE_ZERO_COPY support for
convert_to_xdp_frame. Converting a MEM_TYPE_ZERO_COPY xdp_buff to an
xdp_frame is done by transforming the MEM_TYPE_ZERO_COPY buffer into a
MEM_TYPE_PAGE_ORDER0 frame. This is costly, and in the future it might
make sense
From: Björn Töpel
Export __xdp_rxq_info_unreg_mem_model as xdp_rxq_info_unreg_mem_model,
so it can be used from netdev drivers. Also, add additional checks for
the memory type.
Signed-off-by: Björn Töpel
---
include/net/xdp.h | 1 +
net/core/xdp.c| 15 +--
2 files changed, 14
Hello dear.
It is wonderful to contact you, I want us to have correspondence. I
wish you will have the desire so that we can get acquainted to each
other. Life itself is a mystery, you never know where it might lead
you.
I'm Aisha.gaddafi, the only biological douther of Qi,muamar gaddafi of
Commit 6d0bfe226116 ("net: ipv6: Add IPv6 support to the ping socket.")
contains an error in the cleanup path of inet6_init(): when
proto_register(_prot, 1) fails, we try to unregister
_prot. When rawv6_init() fails, we skip unregistering
_prot.
Example of panic (triggered by faking a failure of
rtnl_unregister_all(PF_INET6) gets called from inet6_init in cases when
no handler has been registered for PF_INET6 yet, for example if
ip6_mr_init() fails. Abort and avoid a NULL pointer deref in that case.
Example of panic (triggered by faking a failure of
register_pernet_subsys):
general
Commit 15e668070a64 ("ipv6: reorder icmpv6_init() and ip6_mr_init()")
moved the cleanup label for ipmr_fail, but should have changed the
contents of the cleanup labels as well. Now we can end up cleaning up
icmpv6 even though it hasn't been initialized (jump to icmp_fail or
ipmr_fail).
Simply
Den tis 28 aug. 2018 kl 14:47 skrev Björn Töpel :
>
> From: Björn Töpel
>
> This patch set introduces zero-copy AF_XDP support for Intel's i40e
> driver. In the first preparatory patch we also add support for
> XDP_REDIRECT for zero-copy allocated frames so that XDP programs can
> redirect them.
From: Magnus Karlsson
This patch prepares for the upcoming zero-copy Tx functionality, by
moving common functions and refactor chunks of code into re-usable
functions, used both by the regular path and zero-copy path.
Signed-off-by: Magnus Karlsson
---
From: Björn Töpel
The -c/--copy -z/--zero-copy flags enforces either copy or zero-copy
mode.
Signed-off-by: Björn Töpel
---
samples/bpf/xdpsock_user.c | 12 +++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git a/samples/bpf/xdpsock_user.c b/samples/bpf/xdpsock_user.c
index
From: Björn Töpel
This patch adds zero-copy Rx support for AF_XDP sockets. Instead of
allocating buffers of type MEM_TYPE_PAGE_SHARED, the Rx frames are
allocated as MEM_TYPE_ZERO_COPY when AF_XDP is enabled for a certain
queue.
All AF_XDP specific functions are added to a new file, i40e_xsk.c.
From: Magnus Karlsson
This patch adds zero-copy Tx support for AF_XDP sockets. It implements
the ndo_xsk_async_xmit netdev ndo and performs all the Tx logic from a
NAPI context. This means pulling egress packets from the Tx ring,
placing the frames on the NIC HW descriptor ring and completing
From: Björn Töpel
Add functions for queue pair enable/disable. Instead of resetting the
whole device, only the affected queue pair is disabled or enabled.
This plumbing is used in a later commit, when zero-copy AF_XDP support
is introduced.
Signed-off-by: Björn Töpel
---
From: Björn Töpel
Move the xdp_umem_get_{data,dma} functions to include/net/xdp_sock.h,
so that the upcoming zero-copy implementation in the Ethernet drivers
can utilize them.
Also, supply some dummy function implementations for
CONFIG_XDP_SOCKETS=n configs.
Signed-off-by: Björn Töpel
---
From: Björn Töpel
This patch prepares for the upcoming zero-copy Rx functionality, by
moving/changing linkage of common functions, used both by the regular
path and zero-copy path.
Signed-off-by: Björn Töpel
---
drivers/net/ethernet/intel/i40e/i40e_txrx.c | 33 ---
From: Magnus Karlsson
The function napi_if_scheduled_mark_missed is used to check if the
NAPI context is scheduled, if so set NAPIF_STATE_MISSED and return
true. Used by the AF_XDP zero-copy i40e Tx code implementation in
order to make sure that irq affinity is honored by the napi context.
From: Björn Töpel
This patch set introduces zero-copy AF_XDP support for Intel's i40e
driver. In the first preparatory patch we also add support for
XDP_REDIRECT for zero-copy allocated frames so that XDP programs can
redirect them. This was a ToDo from the first AF_XDP zero-copy patch
set from
From: Björn Töpel
In this commit, the Rx path is refactored some, as a step torwards the
introduction AF_XDP Rx zero-copy.
The page re-use counter is moved into the i40e_reuse_rx_page, instead
of bumping the counter in many places. The Rx buffer page clearing is
moved for better readability.
DEAR ALL
TAKE NOTE
20183516 document set.xlsx
Description: Binary data
The error path of inet6_init() can trigger multiple kernel panics,
mostly due to wrong ordering of cleanups. This series fixes those
issues.
Sabrina Dubroca (3):
ipv6: fix cleanup ordering for ip6_mr failure
ipv6: fix cleanup ordering for pingv6 registration
net: rtnl: return early from
From: Mark Bloch
This makes it clear and safe to access constants passed in from user
space. We define a consistent ABI of u64 for all constants, and verify
that the data passed in can be represented by the type the user supplies.
The expectation is this will always be used with an enum
From: Mark Bloch
We expose new actions:
L2_TO_L2_TUNNEL - A generic encap from L2 to L2, the data passed should
be the encapsulating headers.
L3_TUNNEL_TO_L2 - Will do decap where the inner packet starts from L3,
the data should be mac or mac + vlan (14 or
From: Mark Bloch
Expose new abilities when creating a packet reformat context.
The new types which can be created are:
MLX5_REFORMAT_TYPE_L2_TO_L2_TUNNEL: Ability to create generic encap
operation to be done by the HW.
MLX5_REFORMAT_TYPE_L3_TUNNEL_TO_L2: Ability to create generic decap
From: Mark Bloch
Expose the ability to create a flow action which changes packet
headers. The data passed from userspace should be modify header actions
as defined by HW specification.
Signed-off-by: Mark Bloch
Signed-off-by: Leon Romanovsky
---
drivers/infiniband/hw/mlx5/flow.c |
From: Mark Bloch
Refactor the initialization of a flow action object to a common function.
Signed-off-by: Mark Bloch
Signed-off-by: Leon Romanovsky
---
drivers/infiniband/core/uverbs_std_types_flow_action.c | 7 ++-
drivers/infiniband/hw/mlx5/flow.c | 8 +++-
From: Mark Bloch
Currently we attach packet reformat actions only to the FDB namespace.
In preparation to be able to use that for NIC steering, pass the actual
namespace as a parameter.
Signed-off-by: Mark Bloch
Reviewed-by: Or Gerlitz
Reviewed-by: Saeed Mahameed
Signed-off-by: Leon
From: Mark Bloch
This will allow for the RDMA side to allocate packet reformat context.
Signed-off-by: Mark Bloch
Reviewed-by: Saeed Mahameed
Signed-off-by: Leon Romanovsky
---
drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c| 2 ++
drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h |
Rob Herring writes:
> In preparation to remove the node name pointer from struct device_node,
> convert printf users to use the %pOFn format specifier.
>
> Cc: Kalle Valo
> Cc: "David S. Miller"
> Cc: linux-wirel...@vger.kernel.org
> Cc: netdev@vger.kernel.org
> Signed-off-by: Rob Herring
>
From: Mark Bloch
Refactor the switch logic so it's simpler to follow and understand.
Signed-off-by: Mark Bloch
Reviewed-by: Saeed Mahameed
Signed-off-by: Leon Romanovsky
---
drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 24 ++-
1 file changed, 6 insertions(+), 18
From: Mark Bloch
Renames all encap mlx5_{core,ib} code to use the new naming of packet
reformat. This change doesn't introduce any function change and is
needed to properly reflect the operation being done by this action.
For example not only can we encapsulate a packet, but also decapsulate it.
From: Leon Romanovsky
>From Mark,
This series exposes the ability to create flow actions which can
mutate packet headers. We do that by exposing two new verbs:
* modify header - can change existing packet headers. packet
* reformat - can encapsulate or decapsulate a packet.
Once
From: Mark Bloch
Those functions will be used by the RDMA side to create modify header
actions to be attached to flow steering rules via verbs.
Signed-off-by: Mark Bloch
Reviewed-by: Saeed Mahameed
Signed-off-by: Leon Romanovsky
---
drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c| 2 ++
From: Mark Bloch
Extend the ability to add steering rules to NIC TX flow tables.
For now, we are only adding TX bypass (egress) which is used by the RDMA
side. This will allow to shape outgoing traffic and tweak it if needed, for
example performing encapsulation or rewriting headers.
From: Mark Bloch
Today we are able to attach encap and decap actions only to the FDB. In
preparation to enable those actions on the NIC flow tables, break the
single flag into two. Those flags control whatever a decap or encap
operations can be attached to the flow table created. For FDB, if
From: Mark Bloch
Those bits are hardware specification and should be defined in the
IFC header file.
Signed-off-by: Mark Bloch
Reviewed-by: Or Gerlitz
Reviewed-by: Saeed Mahameed
Signed-off-by: Leon Romanovsky
---
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 5 -
On Mon, 27 Aug 2018 20:35:02 -0700
Alexei Starovoitov wrote:
> I'm also running Arch Linux in my VM, but I'm not able to reproduce
> umount issue. I'm guessing it's somehow related to non-static build
> and libc.so being busy with old systemd.
Oh, I mentioned it in a previous draft of my
When adding support for JSON output the new code for printing
the destination prefix adds a stray blank character before
the bitmask. This causes some user-space parsing to fail.
Current output:
...: from x.x.x.x/l to y.y.y.y /l
Previous output:
...: from x.x.x.x/l to y.y.y.y/l
Fixes:
After the commit 802bfb19152c ("net/sched: user-space can't set
unknown tcfa_action values"), unknown tcfa_action values are
converted to TC_ACT_UNSPEC, but the common agreement is instead
rejecting such configurations.
This change also introduce an helper to simplify the destruction
of a single
As agreed some time ago, this changeset reject unknown tcfa_action values,
instead of changing such values under the hood.
A tdc test is included to verify the new behavior.
Paolo Abeni (2):
net_sched: reject unknown tcfa_action values
tc-testing: add test-cases for numeric and invalid
Only the police action allows us to specify an arbitrary numeric value
for the control action. This change introduces an explicit test case
for the above feature and then leverage it for testing the kernel behavior
for invalid control actions (reject).
Signed-off-by: Paolo Abeni
---
Can anyone help me understand why a busy database server gets these
kernel warnings? It comes from this WARN_ON macro, apparently because
the socket is still owned by the user process. Why would that happen?
WARN_ON(sock_owned_by_user(sk));
[726780.788201] WARNING: CPU: 15 PID: 52245 at
While recently going over bpf_msg_pull_data(), I noticed three
issues which are fixed in here:
1) When we attempt to find the first scatterlist element (sge)
for the start offset, we add len to the offset before we check
for start < offset + len, whereas it should come after when
we
You know the drill...
http://vger.kernel.org/~davem/net-next.html
On Mon, Aug 27, 2018 at 9:08 PM Neil Horman wrote:
>
> On Mon, Aug 27, 2018 at 06:38:31PM +0800, Xin Long wrote:
> > As Marcelo noticed, in sctp_transport_get_next, it is iterating over
> > transports but then also accessing the association directly, without
> > checking any refcnts before that,
Hello Andreas, Nick,
On 28 August 2018 at 06:06, Nicholas Piggin wrote:
> On Mon, 27 Aug 2018 19:11:01 +0200
> Andreas Schwab wrote:
>
>> I'm getting this Oops when running iptables -F OUTPUT:
>>
>> [ 91.139409] Unable to handle kernel paging request for data at address
>> 0xd001fff12f34
- Original Message -
> The error path of inet6_init() can trigger multiple kernel panics,
> mostly due to wrong ordering of cleanups. This series fixes those
> issues.
>
> Sabrina Dubroca (3):
> ipv6: fix cleanup ordering for ip6_mr failure
> ipv6: fix cleanup ordering for pingv6
On Tue, 2018-08-28 at 16:27 +0200, Stefan Bader wrote:
> When adding support for JSON output the new code for printing
> the destination prefix adds a stray blank character before
> the bitmask. This causes some user-space parsing to fail.
>
> Current output:
> ...: from x.x.x.x/l to y.y.y.y /l
On Tue, 28 Aug 2018 14:44:25 +0200
Björn Töpel wrote:
> From: Björn Töpel
>
> This commit adds proper MEM_TYPE_ZERO_COPY support for
> convert_to_xdp_frame. Converting a MEM_TYPE_ZERO_COPY xdp_buff to an
> xdp_frame is done by transforming the MEM_TYPE_ZERO_COPY buffer into a
>
On 28 August 2018 at 15:56, Ard Biesheuvel wrote:
> Hello Andreas, Nick,
>
> On 28 August 2018 at 06:06, Nicholas Piggin wrote:
>> On Mon, 27 Aug 2018 19:11:01 +0200
>> Andreas Schwab wrote:
>>
>>> I'm getting this Oops when running iptables -F OUTPUT:
>>>
>>> [ 91.139409] Unable to handle
Currently, we do a shutdown(sk, SHUT_RDWR) on both peer sockets and
a shutdown on the sender as well. However, this is incorrect and can
occasionally cause issues if you happen to have bad timing. First
peer1 or peer2 may still be in use depending on the test and timing.
Second we really should
Two small test sockmap updates for bpf-next. These help me run some
additional tests with test_sockmap.
---
John Fastabend (2):
bpf: sockmap test remove shutdown() calls
bpf: use --cgroup in test_suite if supplied
tools/testing/selftests/bpf/test_sockmap.c | 56
If the user supplies a --cgroup value in the arguments when running
the test_suite go ahaead and run the self tests there. I use this
to test with multiple cgroup users.
Signed-off-by: John Fastabend
---
tools/testing/selftests/bpf/test_sockmap.c | 53
1 file
On Mon, Aug 27, 2018 at 1:56 PM Davide Caratti wrote:
>
> in the (rare) case of failure in nla_nest_start(), missing NULL checks in
> tcf_pedit_key_ex_dump() can make the following command
>
> # tc action add action pedit ex munge ip ttl set 64
>
> dereference a NULL pointer:
>
> BUG: unable to
Den tis 28 aug. 2018 kl 16:11 skrev Jesper Dangaard Brouer :
>
> On Tue, 28 Aug 2018 14:44:25 +0200
> Björn Töpel wrote:
>
> > From: Björn Töpel
> >
> > This commit adds proper MEM_TYPE_ZERO_COPY support for
> > convert_to_xdp_frame. Converting a MEM_TYPE_ZERO_COPY xdp_buff to an
> > xdp_frame
This test creates a raw IPv4 socket, fragments a largish UDP
datagram and sends the fragments out of order.
Then repeats in a loop with different message and fragment lengths.
Then does the same with overlapping fragments (with overlapping
fragments the expectation is that the recv times out).
The current behavior of IP defragmentation is inconsistent:
- some overlapping/wrong length fragments are dropped without
affecting the queue;
- most overlapping fragments cause the whole frag queue to be dropped.
This patch brings consistency: if a bad fragment is detected,
the whole frag
Ugh, CC: netdev..
On Tue, 28 Aug 2018 20:05:39 +0200, Jakub Kicinski wrote:
> Hi!
>
> I wonder if we can use phys_port_id in switchdev to group together
> interfaces of a single PCI PF? Here is the problem:
>
> With a mix of PF and VF interfaces it gets increasingly difficult to
> figure out
On Tue, Aug 28, 2018 at 7:25 AM Paolo Abeni wrote:
>
> +int tcf_action_destroy_one(struct tc_action *a, int bind)
> +{
> + struct tc_action *actions[] = { a, NULL };
> +
> + return tcf_action_destroy(actions, bind);
> +}
Make it static.
> +
> static int tcf_action_put(struct
This gets rid of the licence boilerblate in favor of SPDX identifier
which only takes a single line comment.
Signed-off-by: Mika Westerberg
---
drivers/net/thunderbolt.c | 5 +
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/drivers/net/thunderbolt.c b/drivers/net/thunderbolt.c
On Aug 28 2018, Ard Biesheuvel wrote:
> diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
> index 6a501b25dd85..57d09d5ceb1a 100644
> --- a/arch/powerpc/kernel/setup_64.c
> +++ b/arch/powerpc/kernel/setup_64.c
> @@ -779,7 +779,6 @@ EXPORT_SYMBOL(__per_cpu_offset);
>
>
From: Rick Farrington
Signed-off-by: Rick Farrington
Signed-off-by: Felix Manlunas
---
drivers/net/ethernet/cavium/liquidio/request_manager.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c
This series contains new features and implementation updates for the
ice driver.
Anirudh reworks the current flex programming logic to add support for
a second flex descriptor profile. Updated the transmit scheduler
code to handle changes to the spec, specifically the firmware expects
a 4KB
From: Brett Creeley
VLAN pruning is not valid when VLAN 0 is not active. If VLAN
pruning is enabled and VLAN 0 is not active (8021q driver not loaded)
then normal, non-VLAN, traffic will not pass.
TX/RX VLAN pruning is enabled when the VLAN 0 is added to the
active_vlan bitmap and it is
From: Dan Nowlin
Align Request Resource Ownership AQ command (0x0008) to the latest
specification. This includes:
- Correcting the resource IDs for the Global Cfg and Change locks.
- new enum ICE_CHANGE_LOCK_RES_ID
- new enum ICE_GLOBAL_CFG_LOCK_RES_ID
- Altering the flow for Global Config Lock
From: Anirudh Venkataramanan
This patch is an adaptation of the work originally done by Grishma
Kotecha that in summary refactors the
switch filtering logic in the driver. More specifically,
- Update the recipe structure to also store list of rules
- Update the existing code for recipes like
From: Sudheer Mogilappagari
When a malicious operation is detected, the firmware triggers an
interrupt, which is then picked up by the service task (specifically by
ice_handle_mdd_event). A reset is scheduled if required.
Tx hang detection works in a similar way, except the logic here monitors
From: Anirudh Venkataramanan
Remove the "ice" prefix for the driver version string and bump version
to 0.7.1-k.
Signed-off-by: Anirudh Venkataramanan
Tested-by: Tony Brelinski
Signed-off-by: Jeff Kirsher
---
drivers/net/ethernet/intel/ice/ice_main.c | 2 +-
1 file changed, 1 insertion(+), 1
From: Akeem G Abodunrin
This patch introduces SERVICE_DIS flag to use for stopping service task.
This flag will be checked before scheduling new tasks. Also add new
functions ice_service_task_stop to stop service task.
Signed-off-by: Akeem G Abodunrin
Signed-off-by: Anirudh Venkataramanan
From: Hieu Tran
To enable FW logging, the "cq_en" and "uart_en" enable bits of the
"fw_log" element in struct ice_hw need to set accordingly based on
some user-provided parameters during driver loading. To select which
FW log events to be emitted, the "cfg" elements of corresponding FW
modules
From: Anirudh Venkataramanan
The driver can support two flex descriptor profiles, ICE_RXDID_FLEX_NIC
and ICE_RXDID_FLEX_NIC_2. This patch reworks the current flex programming
logic to add support for the latter profile.
Signed-off-by: Anirudh Venkataramanan
Tested-by: Tony Brelinski
Dear Björn,
On 08/24/18 16:00, Jesper Dangaard Brouer wrote:
> On Fri, 24 Aug 2018 13:21:59 +0200
> Björn Töpel wrote:
>
>> When XDP is enabled, the driver will report incorrect
>> statistics. Received frames will reported as transmitted frames.
>>
>> This commits fixes the i40e implementation
From: Anirudh Venkataramanan
This patch refactors aspects of the VSI allocation, deletion and rebuild
flow. Some of the more noteworthy changes are described below.
1) On reset, all switch filters applied in the hardware are lost. In
the rebuild flow, only MAC and broadcast filters are being
From: Anirudh Venkataramanan
This patch cleans up the existing register definitions.
1) Several instances of long defines names used in the BIT() macro
were replaced to use the actual values they represent. As a
result some defines for shifts (ending with _S) that were used
only to
From: Md Fahad Iqbal Polash
ice_bridge_getlink returns the current bridge mode using
ndo_dflt_bridge_getlink and the mode parameter available in
first_switch->bridge_mode.
ice_bridge_setlink is invoked when the bridge mode needs to
changed. The value to be changed to is available as a netlink
From: Anirudh Venkataramanan
1) The maximum device nodes is a global value and shared by the whole
device. Add element AQ command would fail if there is no space to
add new nodes so the check for max nodes isn't required. So remove
ice_sched_get_num_nodes_per_layer and
From: Chinh Cao
This patch implements handlers for ethtool get_link_ksettings and
set_link_ksettings. Helper functions use by these handlers are also
introduced in this patch.
Signed-off-by: Chinh Cao
Signed-off-by: Anirudh Venkataramanan
Tested-by: Tony Brelinski
Signed-off-by: Jeff Kirsher
From: Zhenning Xiao
Use the buffer in the s_rule structure directly instead of using
a local array eth_hdr[DUMMY_ETH_HDR_LEN]
Signed-off-by: Zhenning Xiao
Signed-off-by: Anirudh Venkataramanan
Tested-by: Tony Brelinski
Signed-off-by: Jeff Kirsher
---
From: Anirudh Venkataramanan
Once reset is issued, the driver loses all control queue interfaces.
Exercising control queue operations during reset is incorrect and
may result in long timeouts.
This patch introduces a new field 'reset_ongoing' in the hw structure.
This is set to 1 by the core
On 2018-08-28 19:00, Paul Menzel wrote:
Dear Björn,
On 08/24/18 16:00, Jesper Dangaard Brouer wrote:
On Fri, 24 Aug 2018 13:21:59 +0200
Björn Töpel wrote:
When XDP is enabled, the driver will report incorrect
statistics. Received frames will reported as transmitted frames.
This commits
From: Rick Farrington
In lio_enable_irq, the pkt_in_done count register was being cleared to
zero. However, there could be some completed instructions which were not
yet processed due to budget and limit constraints.
So, only write this register with the number of actual completions
that were
On Mon, Jul 30, 2018 at 2:06 AM Jason Wang wrote:
>
>
>
> On 2018年07月25日 08:17, Jon Olson wrote:
> > On Tue, Jul 24, 2018 at 3:46 PM Michael S. Tsirkin wrote:
> >> On Tue, Jul 24, 2018 at 06:31:54PM -0400, Willem de Bruijn wrote:
> >>> On Tue, Jul 24, 2018 at 6:23 PM Michael S. Tsirkin
> >>>
The name value from SET_NETDEV_DEVTYPE only ended up in the uevent sysfs
file as DEVTYPE= information. To avoid any kind of race conditions
between netlink messages and reading from sysfs, it is useful to add the
same string as new IFLA_DEVTYPE attribute included in the RTM_NEWLINK
messages.
For
From: Shannon Nelson
Add functions to translate VF IPsec offload add and delete requests
into something the existing code can work with.
Signed-off-by: Shannon Nelson
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsher
---
.../net/ethernet/intel/ixgbe/ixgbe_ipsec.c| 256
From: Shannon Nelson
Restore the IPsec hardware IP table after reloading the SA tables.
This doesn't make much difference now, but will matter when we add
support for VF IPsec offloads.
Signed-off-by: Shannon Nelson
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsher
---
From: Shannon Nelson
When failing the request because we can't support that offload,
reporting EOPNOTSUPP makes much more sense than ENXIO.
Signed-off-by: Shannon Nelson
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsher
---
drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c | 2 +-
1 file
From: Shannon Nelson
Add the IPsec offload support code. This is based off of the similar
code in ixgbe, but instead of writing the SA registers, the VF asks
the PF to setup the offload by sending the offload information to the
PF via the standard mailbox.
Signed-off-by: Shannon Nelson
From: Shannon Nelson
Add the IPsec initialization into the driver startup and
add the Rx and Tx processing hooks.
Signed-off-by: Shannon Nelson
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsher
---
drivers/net/ethernet/intel/ixgbevf/defines.h | 2 +-
From: Shannon Nelson
The software SA record counters should not be cleared when clearing
the hardware tables. This causes the counters to be out of sync
after a driver reset.
Fixes: 63a67fe229ea ("ixgbe: add ipsec offload add and remove SA")
Signed-off-by: Shannon Nelson
Tested-by: Andrew
From: Shannon Nelson
There seems to be a problem in the x540's internal switch wherein if SR-IOV
mode is enabled and an offloaded IPsec packet is sent to a local VF,
the packet is silently dropped. This might never be a problem as it is
somewhat a corner case, but if someone happens to be using
This series contains updates to ixgbe and ixgbevf only.
Sebastian adds support for firmware NVM recovery mode, which logs a
message when errors are detected and un-registers the device. Also
fixed RSS type recognition with VF to VF communication.
Shannon Nelson implements IPsec hardware offload
From: Shannon Nelson
Pull out a couple of values from a function so they can be used
later elsewhere.
Signed-off-by: Shannon Nelson
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsher
---
drivers/net/ethernet/intel/ixgbe/ixgbe_ipsec.c | 8 +---
1 file changed, 5 insertions(+), 3
To enable easier FW distribution NFP can now automatically
select between FW stored on the flash and loaded from the
kernel.
If FW loading policy is set to auto it will compare the
versions of FW from the host and from the flash and load
the newer one. If FW type doesn't match (e.g. one advanced
The sysfs field was introduced 4 years ago along with fixes to various
drivers that erroneously used `dev_id' for that purpose, but it was not
properly documented anywhere.
See commit v3.14-rc3-739-g3f85944fe207.
Signed-off-by: Arseny Maslennikov
---
Documentation/ABI/testing/sysfs-class-net |
Some InfiniBand network devices have multiple ports on the same PCI
function. This initializes the `dev_port' sysfs field of those
network interfaces with their port number.
The use of `dev_id' was considered correct until Linux 3.15,
when another field, `dev_port', was defined for this
Some InfiniBand network devices have multiple ports on the same PCI
function. Prior to this the kernel erroneously used the `dev_id' sysfs
field of those network interfaces to convey the port number to userspace.
`dev_id' is currently reserved for distinguishing stacked ifaces
(e.g: VLANs) with
Pre-3.15 userspace had trouble distinguishing different ports
of a NIC on a single PCI bus/device/function. To solve this,
a sysfs field `dev_port' was introduced quite a while ago
(commit v3.14-rc3-739-g3f85944fe207), and some relevant device
drivers were fixed to use it, but not in case of
From: Sebastian Basierski
While VF2VF with RSS communication, RSS Type were wrongly recognized
and RSS hash was not calculated as it should be. Packets was
distributed on various queues by accident.
This commit fixes that behaviour and causes proper RSS Type recognition.
Signed-off-by:
From: Shannon Nelson
The software SA record counters should not be cleared when clearing
the hardware tables. This causes the counters to be out of sync
after a driver reset.
Fixes: 63a67fe229ea ("ixgbe: add ipsec offload add and remove SA")
Signed-off-by: Shannon Nelson
Tested-by: Andrew
1 - 100 of 145 matches
Mail list logo