tree: https://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next.git
master
head: 547b9ca879eafeb3507c00dc77002953c9c04b0b
commit: e5d3a51cefbb08c8b8c521f627aef1f998834530 [367/387] bnx2x: extend DCBx
support
config: powerpc-defconfig (attached as .config)
reproduce:
wget
https:/
Hello,
Garbled output for "ethtool -m ethX", in igb-driven NICs with module /
plugin EEPROM (i.e. SFP information). Each output data byte appears
duplicated.
In igb_ethtool.c, igb_get_module_eeprom() is reading the EEPROM via i2c;
the eeprom offset for each word that's read via igb_read_phy_reg_i
Wed, Feb 17, 2016 at 06:18:03AM CET, john.fastab...@gmail.com wrote:
>This is a helper function drivers can use to learn if the
>action type is a drop action.
>
>Signed-off-by: John Fastabend
Acked-by: Jiri Pirko
Wed, Feb 17, 2016 at 06:17:09AM CET, john.fastab...@gmail.com wrote:
>This patch allows netdev drivers to consume cls_u32 offloads via
>the ndo_setup_tc ndo op.
>
>This works aligns with how network drivers have been doing qdisc
>offloads for mqprio.
>
>Signed-off-by: John Fastabend
Acked-by: Jir
It needs very similar workarounds to the one on BCM4707. It was tested
on D-Link DIR-885L home router.
Signed-off-by: Rafał Miłecki
---
drivers/net/ethernet/broadcom/bgmac.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/broadcom/bgmac.c
b/drivers
Hello,
The Intel i211 LOM pcie ethernet controllers' iNVM operates as an OTP
and has no externel EEPROM interface [1]. The following allows the
driver to pickup the MAC address from a device tree blob when CONFIG_OF
has been enabled.
[1]http://www.intel.com/content/www/us/en/embedded/products/ne
On February 16, 2016 3:06:55 AM PST, Sergei Shtylyov
wrote:
>Hello.
>
>On 2/16/2016 10:42 AM, Geert Uytterhoeven wrote:
>
>>> I have observed what appears to be a regression in the ravb ethernet
>driver
>>> caused by d5c3d84657db ("net: phy: Avoid polling PHY with
>>> PHY_IGNORE_INTERRUPTS").
>>>
On February 15, 2016 2:46:45 PM PST, Clemens Gruber
wrote:
>For the Marvell 88E1510, marvell_of_reg_init was called too late, in
>the
>config_aneg function.
>Since commit 113c74d83eef ("net: phy: turn carrier off on phy attach"),
>this lead to the link not coming up at boot anymore, due to the ph
From: Shannon Nelson
Add the new Cisco VXLAN-GPE cloud tunnel type for the Add Cloud Filter
and UDP tunnel AQ commands.
Change-ID: I2c093c7d79726c7fca08a36e5c63581a905da3d2
Signed-off-by: Shannon Nelson
Acked-by: Kevin Scott
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsher
---
drivers/n
From: Shannon Nelson
Add the new Set Switch Config AdminQ command, and mark the L2 Filter
bit as deprecated in the Add VEB command.
Change-ID: I5b24790f14c56f0ddf3f70df1e486844146b039f
Signed-off-by: Shannon Nelson
Acked-by: Kevin Scott
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsher
--
From: Pandi Maharajan
lan_vsi_idx and lan_vsi_id are assigned to u16 data sized variables but
declared in u8. This patch fixes the width of the datatype.
Change-ID: If4bcbcc7d32f2b287c51cb33d17879691258dce2
Signed-off-by: Pandi Kumar Maharajan
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsh
From: Mitch Williams
When calculating TX bandwidth for VFs, we need to know the link speed to
make sure we don't allocate more bandwidth than is available. Add 20G
link speed to the switch statement so we can support devices that link
at that speed.
Change-ID: I5409f6139d549e5832777db9c22ca0664e
From: Shannon Nelson
Add the new AQ command and struct for managing a thermal sensor.
Change-ID: I6f5631839a0f3dca352a6c222f1269a960e2310a
Signed-off-by: Shannon Nelson
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsher
---
drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h | 17 +
From: Anjali Singhai Jain
We were not doing write-back on interrupt throttle for Legacy case in X722.
This patch fixes that, so we do WB_ON_ITR for Legacy as well. Plus the issue
that we should still be setting NO_ITR if we are touching the DYN_CTLN register
since we do not want to change ITR set
From: Mitch Williams
Sometimes, ARQ overflows are a big deal and tell us that the
firmware/hardware/driver/something is having problems. But normally
they're no big deal. To assist in assessing this, add a counter to
our Ethtool stats. A handful of ARQ overflows during VF init is no
problem. A la
From: Jesse Brandeburg
This is a simple refactor suggested by the community
to change a multi-level if statement into a switch.
Change-ID: I831cf3c4042600aa9b43990022d22dfd50db
Signed-off-by: Jesse Brandeburg
Acked-by: Shannon Nelson
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsher
-
From: Shannon Nelson
Add flags to MAC allocation requests to signify that the MAC VLAN filters
should come from the shared resource pool rather than the dedicated PF
resource pools.
Change-ID: I4c2da64c01856edcb0982bc4aab75c5a91047a7a
Signed-off-by: Shannon Nelson
Acked-by: Kevin Scott
Tested-
From: Jesse Brandeburg
Probe routine already has too many locals, just convert one
used for kzalloc into a kcalloc, eliminating the local.
Change-ID: I349049872b71f858cbeb91ad7836e6767fc7b7d1
Signed-off-by: Jesse Brandeburg
Reviewed-by: Anjali Singhai
Tested-by: Andrew Bowers
Signed-off-by: J
From: Mitch Williams
If the VF is reset via VFLR, the device will be knocked out of bus
master mode, and the driver will fail to recover from the reset. Fix
this by enabling bus mastering after every reset. In a non-VFLR case,
the bus master bit will not be disabled, and this call will have no ef
From: Jesse Brandeburg
The PF doesn't need to know about the VF's device IDs, so remove them.
Change-ID: I62cf0e0fffa1ace586e58e00bc271b10ae440f05
Signed-off-by: Jesse Brandeburg
Acked-by: Shannon Nelson
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsher
---
drivers/net/ethernet/intel/i40
Appreciate your efforts!
Just trying to make sure it does not get lost.
Introduced in rc1, not fixed by ... rc4.
Anyway, I will continue to test, lots of other things still to chase
even in rc4!
Regards,
Ross Green
On Wed, Feb 17, 2016 at 2:24 AM, Kalle Valo wrote:
> Ross Green writes:
>
>>
From: Shannon Nelson
Bump AQ minor version to 1.5 for new FW features.
Change-ID: I5a790f7f519a2a8921aaa1c5663727dd1897ffec
Signed-off-by: Shannon Nelson
Acked-by: Kevin Scott
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsher
---
drivers/net/ethernet/intel/i40e/i40e_adminq_cmd.h | 2 +-
From: Catherine Sullivan
Bump.
Change-ID: I21aa520a3c8c5f4f562a98019bf8b76b3706c480
Signed-off-by: Catherine Sullivan
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsher
---
drivers/net/ethernet/intel/i40e/i40e_main.c | 2 +-
drivers/net/ethernet/intel/i40evf/i40evf_main.c | 2 +-
2 fil
From: Shannon Nelson
Add a few more bits of netdev data into the debugfs output for dump VSI.
For now, we'll add the features, hw_features, vlan_features, and flags
bitflags and the state. More could be added later if needed.
Also, tweak a couple nearby output lines for output readability.
Chan
From: Shannon Nelson
Add a little more detail to an NVM update debug message in order to
see the full ethtool request data.
Change-ID: Iab10437cb32d6fddc67ee347e7c0b42511e152cd
Signed-off-by: Shannon Nelson
Acked-by: Kevin Scott
Tested-by: Andrew Bowers
Signed-off-by: Jeff Kirsher
---
drive
This series contains updates to i40e/i40evf only.
Shannon adds flags to MAC allocation requests to signify that the MAC VLAN
filters should come from the shared resource pool. Added a new "set switch
config" admin queue command and the new Cisco VXLAN-GPE cloud tunnel type
for the admin queue com
On 16-02-16 05:17 PM, David Miller wrote:
> From: John Fastabend
> Date: Mon, 15 Feb 2016 22:00:03 -0800
>
>> This extends the setup_tc framework so it can support more than just
>> the mqprio offload and push other classifiers and qdiscs into the
>> hardware.
> ...
>
> Jiri gave some minor sty
This adds initial support for offloading the u32 tc classifier. This
initial implementation only implements a few base matches and actions
to illustrate the use of the infrastructure patches.
However it is an interesting subset because it handles the u32 next
hdr logic to correctly map tcp packets
This patch ensures ixgbe will not try to offload hash tables from the
u32 module. The device class does not currently support this so until
it is enabled just abort on these tables.
Interestingly the more flexible your hardware is the less code you
need to implement to guard against these cases.
This adds an ixgbe data structure that is used to determine what
headers:fields can be matched and in what order they are supported.
For hardware devices this can be a bit tricky because typically
only pre-programmed (firmware, ucode, rtl) parse graphs will be
supported and we don't yet have an in
Its useful to turn off the qdisc offload feature at a per device
level. This gives us a big hammer to enable/disable offloading.
More fine grained control (i.e. per rule) may be supported later.
Signed-off-by: John Fastabend
Acked-by: Jiri Pirko
---
include/linux/netdev_features.h |3 +++
n
This is a helper function drivers can use to learn if the
action type is a drop action.
Signed-off-by: John Fastabend
---
include/net/tc_act/tc_gact.h | 16
1 file changed, 16 insertions(+)
diff --git a/include/net/tc_act/tc_gact.h b/include/net/tc_act/tc_gact.h
index 592a6bc
This patch allows netdev drivers to consume cls_u32 offloads via
the ndo_setup_tc ndo op.
This works aligns with how network drivers have been doing qdisc
offloads for mqprio.
Signed-off-by: John Fastabend
---
include/linux/netdevice.h |6 ++-
include/net/pkt_cls.h | 34 ++
This patch updates setup_tc so we can pass additional parameters into
the ndo op in a generic way. To do this we provide structured union
and type flag.
This lets each classifier and qdisc provide its own set of attributes
without having to add new ndo ops or grow the signature of the
callback.
S
The ndo_setup_tc() op was added to support drivers offloading tx
qdiscs however only support for mqprio was ever added. So we
only ever added support for passing the number of traffic classes
to the driver.
This patch generalizes the ndo_setup_tc op so that a handle can
be provided to indicate if
This extends the setup_tc framework so it can support more than just
the mqprio offload and push other classifiers and qdiscs into the
hardware. The series here targets the u32 classifier and ixgbe
driver. I worked out the u32 classifier because it is protocol
oblivious and aligns with multiple har
On Tue, Feb 16, 2016 at 04:35:27PM -0600, Tom Zanussi wrote:
> On Sun, 2016-02-14 at 01:02 +0100, Alexei Starovoitov wrote:
> > On Fri, Feb 12, 2016 at 10:11:18AM -0600, Tom Zanussi wrote:
> > this hist triggers belong in the kernel. BPF already can do
> > way more complex aggregation and histogra
On Tue, 16 Feb 2016 10:09:51 -0500
Vivien Didelot wrote:
> Remove the shared br_log_state function and print the info directly in
> br_set_state, where the net_bridge_port state is actually changed.
>
> Signed-off-by: Vivien Didelot
> ---
Looks good and fixes the weird case where set_state fai
On Tue, Feb 16, 2016 at 11:01:25PM +0800, Antonio Quartulli wrote:
> Hello David,
>
> this pull request is intended for net.
David,
when merging net into net-next these patches will create a conflict which git
should try to fix on its own.
However, it will still ask you to confirm something. Her
On 2/16/16 7:10 PM, YOSHIFUJI Hideaki wrote:
> Hi,
>
> David Ahern wrote:
>> On 2/16/16 1:45 AM, YOSHIFUJI Hideaki wrote:
diff --git a/Documentation/networking/ip-sysctl.txt
b/Documentation/networking/ip-sysctl.txt
index 24ce97f42d35..7ddbbb67f0db 100644
--- a/Documentation/ne
On Thu, Feb 11, 2016 at 07:57:05PM +, Brian Russell wrote:
> Support encap/decap of Network Service Header (NSH) as defined in
> https://tools.ietf.org/html/draft-ietf-sfc-nsh-01
>
> Includes support for Type 1 and Type 2 metadata and a simple registration
> for listeners to see decapsulated p
On 16-02-16 12:20 PM, Jiri Pirko wrote:
> Tue, Feb 16, 2016 at 07:01:25AM CET, john.fastab...@gmail.com wrote:
>> This patch allows netdev drivers to consume cls_u32 offloads via
>> the ndo_setup_tc ndo op.
>>
>> This works aligns with how network drivers have been doing qdisc
>> offloads for mqpri
On 16-02-16 12:26 PM, Jiri Pirko wrote:
> Tue, Feb 16, 2016 at 07:02:16AM CET, john.fastab...@gmail.com wrote:
>> This is a helper function drivers can use to learn if the
>> action type is a drop action.
>>
>> Signed-off-by: John Fastabend
>> ---
>> include/net/tc_act/tc_gact.h | 16 +++
An error response from a RTM_GETNETCONF request can return the positive
error value EINVAL in the struct nlmsgerr that can mislead userspace.
Signed-off-by: Anton Protopopov
---
net/ipv4/devinet.c | 2 +-
net/ipv6/addrconf.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a
Hi,
David Ahern wrote:
> On 2/16/16 1:45 AM, YOSHIFUJI Hideaki wrote:
>>> diff --git a/Documentation/networking/ip-sysctl.txt
>>> b/Documentation/networking/ip-sysctl.txt
>>> index 24ce97f42d35..7ddbbb67f0db 100644
>>> --- a/Documentation/networking/ip-sysctl.txt
>>> +++ b/Documentation/networkin
From: Bjørn Mork
Date: Fri, 12 Feb 2016 16:42:14 +0100
> Thomas reports:
>
> T: Bus=01 Lev=01 Prnt=01 Port=03 Cnt=01 Dev#= 4 Spd=480 MxCh= 0
> D: Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs= 1
> P: Vendor=05c6 ProdID=6001 Rev=00.00
> S: Manufacturer=USB Modem
> S: Product=USB Mod
From: Kalle Valo
Date: Fri, 12 Feb 2016 16:40:23 +0200
> here's first pull request for 4.6. We have a new kconfig option
> CONFIG_IWLWIFI_PCIE_RTPM for iwlwifi runtime power management. wl12xxx
> got device tree support for the spi bus and added a bindings document
> for it. More info in the tag
From: Sudip Mukherjee
Date: Sat, 13 Feb 2016 19:33:15 +0530
> The builds of allmodconfig of s390, m68k, tilegx, tilepro is failing
> with the error:
> drivers/net/phy/spi_ks8995.c:477:3: error: implicit declaration of function
> 'gpiod_set_value'
> drivers/net/phy/spi_ks8995.c:477:19: error: i
From: Arnd Bergmann
Date: Fri, 12 Feb 2016 11:42:34 +0100
> The ks8995 phy driver just started using gpiod_* functions, which are
> declared in linux/gpio/consumer.h, not linux/gpio.h, resulting in a
> build failure in randconfig builds that do not have CONFIG_GPIOLIB
> enabled:
>
> drivers/net/
From: Eric Dumazet
Date: Thu, 11 Feb 2016 22:02:53 -0800
> From: Eric Dumazet
>
> tcpi_min_rtt reports the minimal rtt observed by TCP stack for the flow,
> in usec unit. Might be ~0U if not yet known.
>
> tcpi_notsent_bytes reports the amount of bytes in the write queue that
> were not yet se
From: Eric Dumazet
Date: Thu, 11 Feb 2016 22:50:29 -0800
> If tcp_v4_inbound_md5_hash() returns an error, we must release
> the refcount on the request socket, not on the listener.
>
> The bug was added for IPv4 only.
>
> Fixes: 079096f103fac ("tcp/dccp: install syn_recv requests into ehash tab
From: Paolo Abeni
Date: Fri, 12 Feb 2016 15:43:52 +0100
> This patch series try to unify the dst cache implementations currently
> present in the kernel, namely in ip_tunnel.c and ip6_tunnel.c, introducing a
> new generic implementation, replacing the existing ones, and then using
> the new imple
From: Tom Lendacky
Date: Fri, 12 Feb 2016 08:32:24 -0600
> +#ifndef CRCPOLY_LE
> +#define CRCPOLY_LE 0xedb88320
> +#endif
> +static u32 xgbe_vid_crc32_le(__le16 vid_le)
Please do not implement yet another copy of a CRC implementation.
Make use of the various versions the kernel provides already
From: John Fastabend
Date: Mon, 15 Feb 2016 22:00:03 -0800
> This extends the setup_tc framework so it can support more than just
> the mqprio offload and push other classifiers and qdiscs into the
> hardware.
...
Jiri gave some minor stylistic feedback, please address that and
respin, and I'll
From: Yuval Mintz
Date: Tue, 16 Feb 2016 18:07:56 +0200
> This series contains several changes - the biggest change is the
> addition of Geneve NDO support [allows device to perform RSS according
> to inner-headers of encapsulated packet, similar to what it does for
> vxlan]. It also extends dcbx
From: Ben Hutchings
Date: Wed, 17 Feb 2016 00:24:51 +
> I agree that 'if (err) goto cleanup;' is widely used and is generally
> understandable (though more creative uses of goto are often not).
>
> My objection was to 'err = -EFOO; if (cond) goto cleanup;'. That is
> definitely not clear an
From: David Lamparter
Date: Wed, 17 Feb 2016 00:41:15 +0100
> ... this won't do enough. If you look at the 4.3 patch I sent you, I
> was adding a comment:
>
> + /* For VRFs, the VRF device takes the place of the loopback device,
> +with addresses on it being preferred. Note in such cases t
On Tue, 2016-02-16 at 07:32 -0500, Kan Liang wrote:
> From: Kan Liang
>
> This patch implements sub command ETHTOOL_SCOALESCE for ioctl
> ETHTOOL_PERQUEUE. It introduces an interface set_per_queue_coalesce to
> set coalesce of each masked queue to device driver. The wanted coalesce
> information
On Tue, 2016-02-16 at 07:32 -0500, Kan Liang wrote:
> From: Kan Liang
>
> This patch implements sub command ETHTOOL_GCOALESCE for ioctl
> ETHTOOL_PERQUEUE. It introduces an interface get_per_queue_coalesce to
> get coalesce of each masked queue from device driver. Then the interrupt
> coalescing
This reverts commit 348e3435cbefa815bd56a5205c1412b5afe7b92e.
It breaks HTB classful qdiscs on the loopback interface.
It has been broken since kernel v4.2. The offending commit has
been identified by bissection of the issue with the following
test-case. It appears that the loopback interface does
Hi all,
Today's linux-next merge of the net-next tree got a conflict in:
drivers/net/phy/bcm7xxx.c
between commit:
c6dd213abe40 ("net: phy: bcm7xxx: Fix 40nm EPHY features")
815717d1473e ("net: phy: bcm7xxx: Remove wildcard entries"
from the net tree and commit:
3125c081a593 ("net: ph
> Even if your driver does not support tagging
Looking at the data sheet, it looks like the hardware does support
tagging. You add an extra "VLAN" tag, whos bits have very little to do
with VLANs. So a new tag implementation will be needed, but that is
not hard.
Andrew
On Tue, 2016-02-16 at 12:51 -0500, David Miller wrote:
> From: Rainer Weikusat
> Date: Mon, 08 Feb 2016 18:47:19 +
>
> > The present unix_stream_read_generic contains various code sequences of
> > the form
> >
> > err = -EDISASTER;
> > if ()
> > goto out;
> >
> > This has the unfortun
On 16/02/16 14:15, Andrew Lunn wrote:
>> I just spoke with my manager, and we would like to change the target
>> device from LAN9352 to LAN9250. The LAN9250 is the same as the
>> LAN9352 but without the switch. It has one mac and one phy.
>
> It is not so easy to get an overview from the website,
On Tue, Feb 16, 2016 at 1:16 PM, Jiri Benc wrote:
> ether_setup sets IFF_TX_SKB_SHARING but this is not supported by vxlan as it
> modifies the skb on xmit.
>
> Signed-off-by: Jiri Benc
Presumably this is something that should be done on all tunnels that
can carry Ethernet frames, right?
Well, unfortunately... [below]
On Tue, Feb 16, 2016 at 02:59:51PM -0800, David Ahern wrote:
> +++ b/net/ipv4/devinet.c
> @@ -1214,12 +1215,16 @@ __be32 inet_select_addr(const struct net_device *dev,
> __be32 dst, int scope)
> if (addr)
> goto out_unlock;
> no_in_dev:
> +
When selecting an address in context of a VRF, the vrf master should be
preferred for address selection. If it isn't, the user has a hard time
getting the system to select to their preference - the code will pick
the address off the first in-VRF interface it can find, which on a
router could well
On 2/16/16 4:41 PM, David Lamparter wrote:
... this won't do enough. If you look at the 4.3 patch I sent you, I
was adding a comment:
+ /* For VRFs, the VRF device takes the place of the loopback device,
+with addresses on it being preferred. Note in such cases the
+loopback device wil
As requested by Rob Herring on patch
https://patchwork.ozlabs.org/patch/580862/.
This is a new property that it's still in net-next and has never been
used in production, so we are not breaking anything with the
incompatible binding change.
Signed-off-by: Sergio Prado
---
Changes v1 -> v2: add c
David Lamparter noted a use case where the source address selection fails
to pick an address from a VRF interface - unnumbered interfaces.
Relevant commands from his script:
ip addr add 9.9.9.9/32 dev lo
ip link set lo up
ip link add name vrf0 type vrf table 101
ip rule add oif vr
On Fri, Feb 12, 2016 at 11:25 AM, Simon Horman
wrote:
> If an skb was not MPLS initially then it may be GSO and in that case if it
> became MPLS then GSO can't be performed because both MPLS and tunnels make
> use of the inner_protocol field of struct skbuff in order to allow GSO to
> be performed
> I can supply some kind of usage description/example, showing how to
> use this feature to pull in the MAC address from u-boot and route it
> to the i210. Just uncertain as how to name the respective
> file. igb.txt or i210.txt come to mind. Do you prefer one over the
> other?
Hi John
It is quit
On 16.02.2016 21:09, Rainer Weikusat wrote:
https://patchwork.ozlabs.org/patch/579654
lists this as 'superseded', among with the older versions of the patch
which changed the error handling. But at least, I couldn't find anything
superseding it. This was supposed to address the different-but-rel
On Sun, 2016-02-14 at 01:02 +0100, Alexei Starovoitov wrote:
> On Fri, Feb 12, 2016 at 10:11:18AM -0600, Tom Zanussi wrote:
> > Hi,
> >
> > As promised in previous threads, this patchset shares some common
> > functionality with the hist triggers code and enables trace events to
> > be accessed fr
> On Feb 16, 2016, at 17:58, Andrew Lunn wrote:
>
>> On Sat, Feb 13, 2016 at 11:04:45PM +0100, John Holland wrote:
>> Hello,
>>
>> The Intel i211 LOM pcie ethernet controllers' iNVM operates as an
>> OTP and has no externel EEPROM interface [1]. The following allows
>> the driver to pickup the
> I just spoke with my manager, and we would like to change the target
> device from LAN9352 to LAN9250. The LAN9250 is the same as the
> LAN9352 but without the switch. It has one mac and one phy.
It is not so easy to get an overview from the website, but it looks
like:
LAN9250 - one port MAC/PH
On 12.02.2016 20:10, bryan.whiteh...@microchip.com wrote:
> Lino,
>
> Regarding "a matching smp_rmb() in the irq handler"
> There is a smp_wmb() in the irq handler, since in both cases we are forcing a
> write operation on software_irq_signal.
>
> I suppose using atomic operations on software_ir
From:
Date: Tue, 16 Feb 2016 21:37:40 +
> Again, I'd like to change our target device to LAN9250, which is just an
> Ethernet controller with no switch.
>
> Will a driver for that device be more easily accepted?
No, please just implement your driver correctly instead of trying to side-ste
From:
Date: Tue, 16 Feb 2016 21:32:20 +
> I just spoke with my manager, and we would like to change the target
> device from LAN9352 to LAN9250. The LAN9250 is the same as the
> LAN9352 but without the switch. It has one mac and one phy.
>
> Since there is no switch in that product, will tha
On Tue, Feb 16, 2016 at 10:09 PM, Saeed Mahameed
> @@ -1602,7 +1622,7 @@ static int mlx5e_create_tis(struct mlx5e_priv
*priv, int tc)
>
> memset(in, 0, sizeof(in));
>
> - MLX5_SET(tisc, tisc, prio, tc);
> + MLX5_SET(tisc, tisc, prio, tc << 1);
point bug fix? or we could never
From: Tom Herbert
Date: Tue, 16 Feb 2016 13:31:34 -0800
> On Feb 16, 2016 12:53 PM, "David Miller" wrote:
>> And practically speaking we disappear from the internet for VXLAN tunnel
>> endpoints implementing the VXLAN spec properly.
>>
>> That's not going to help anyone at all.
>
> A lot of eff
> From: Andrew Lunn [mailto:and...@lunn.ch]
> Sent: Tuesday, February 16, 2016 3:58 PM
> > I believe I can make the physical phys accessible through ethertool.
> > Would that be reasonable?
>
> How, you have no netdev to attach them to?
>
> There was a nice presentation at netdev last week by Me
> From: David Miller [mailto:da...@davemloft.net]
> Sent: Tuesday, February 16, 2016 3:53 PM
> >>
> >> I do not think, with all the infrastructure we have, that we should
> >> accept pure ethernet drivers for such devices any more.
> >>
> >> About year ago I would have responded differently, but th
Hello.
On 02/16/2016 06:01 PM, Or Gerlitz wrote:
From: Daniel Jurgens
The PCI channel could go offline during reset due to EEH. Don't bug on in
this case, the error is recoverable.
Fixes: f6bc11e42646 ('net/mlx4_core: Enhance the catas flow to support device
reset')
Signed-off-by: Daniel J
In metadata mode, the vxlan interface is not supposed to use the fdb control
plane but an external one (openvswitch or static routes). With the current
code, packets may leak into the fdb handling code which usually causes them
to be dropped anyway but may have strange side effects.
Just drop the
ether_setup sets IFF_TX_SKB_SHARING but this is not supported by vxlan as it
modifies the skb on xmit.
Signed-off-by: Jiri Benc
---
drivers/net/vxlan.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index a31cd954b308..db96f3a16f6c 100644
--- a/driv
On Tue, Feb 16, 2016 at 10:09 PM, Saeed Mahameed wrote:
> Support the ndo_setup_tc callback.
> Added en_dcbnl.c which implements the set/get DCBNL IEEE ETS,
> set/get dcbx and registers the mlx5e dcbnl ops.
>
> So far each channel had a single TXQ.
> Now each channel has a TXQ per TC (Traffic Clas
From: Sascha Hauer
Date: Thu, 11 Feb 2016 11:44:48 +0100
> The Marvell 88E6240 has been tested successfully without further
> changes. Add entry to the table of supported devices.
>
> Signed-off-by: Sascha Hauer
Applied.
From: Sascha Hauer
Date: Thu, 11 Feb 2016 11:44:49 +0100
> The phy has not been initialized, disconnecting it in the error
> path results in a NULL pointer exception. Drop the phy_disconnect
> from the error path.
>
> Signed-off-by: Sascha Hauer
Applied.
From: Eric Dumazet
Date: Thu, 11 Feb 2016 08:58:18 -0800
> From: Eric Dumazet
>
> There are some cases where rtt_us derives from deltas of jiffies,
> instead of using usec timestamps.
>
> Since we want to track minimal rtt, better to assume a delta of 0 jiffie
> might be in fact be very close
From: Ken Kawasaki
Date: Thu, 11 Feb 2016 20:15:21 +0900
> add new id (CONTEC C-NET(PC)C-100TX2)
>
> Signed-off-by: Ken Kawasaki
Applied, thanks.
From: Richard Alpe
Date: Thu, 11 Feb 2016 10:43:15 +0100
> Refactor tipc_node_xmit() to fail fast and fail early. Fix several
> potential memory leaks in unexpected error paths.
>
> Reported-by: Dmitry Vyukov
> Reviewed-by: Jon Maloy
> Signed-off-by: Richard Alpe
Applied, thanks.
To make vxlan_udp_encap_recv shorter and more comprehensible.
Signed-off-by: Jiri Benc
---
drivers/net/vxlan.c | 33 +++--
1 file changed, 19 insertions(+), 14 deletions(-)
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 6ac9776a8eb0..45467552162a 100644
Bring the extension handling to a single place and move the actual handling
logic out of vxlan_udp_encap_recv as much as possible.
Signed-off-by: Jiri Benc
---
drivers/net/vxlan.c | 62 +++--
1 file changed, 32 insertions(+), 30 deletions(-)
diff
Part of the parameters is not needed. Simplify the caller of this function
in preparation of making vxlan rx more comprehensible.
Signed-off-by: Jiri Benc
---
drivers/net/vxlan.c | 22 --
1 file changed, 8 insertions(+), 14 deletions(-)
diff --git a/drivers/net/vxlan.c b/dri
The rx path of VXLAN turned over time into kind of spaghetti code. The rx
processing is split between vxlan_udp_encap_recv and vxlan_rcv but in an
artificial way: 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
For metadata based tunnels, VNI is ignored when doing vxlan device lookups
(because such tunnel receives all VNIs). However, this was not honored by
vxlan_xmit_one when doing encapsulation bypass. Move the check for metadata
based tunnel to the common place where it belongs.
Signed-off-by: Jiri Be
Prevent repeated conversions from and to network order in the fast path.
To achieve this, define all flag constants in big endian order and store VNI
as __be32. To prevent confusion between the actual VNI value and the VNI
field from the header (which contains additional reserved byte), strictly
d
Currently, pointer to the vxlan header is kept in a local variable. It has
to be reloaded whenever the pskb pull operations are performed which usually
happens somewhere deep in called functions.
Create a vxlan_hdr function and use it to reference the vxlan header
instead.
Signed-off-by: Jiri Ben
When there are unrecognized flags present in the vxlan header, it doesn't
make much sense to return the packet for further UDP processing, especially
considering that for other invalid flag combinations we drop the packet
because of previous checks.
This means we return positive value only at the
1 - 100 of 289 matches
Mail list logo
