From: Florian Westphal
Date: Thu, 9 Jun 2016 00:27:38 +0200
> iproute2 does not implement any options that result in the
> TCA_CBQ_OVL_STRATEGY/TCA_CBQ_POLICE attributes being set/used.
>
> This series removes these two attributes from cbq and makes kernel reject
> them via EOPNOTSUPP in case
On Wed, Jun 01, 2016 at 11:37:50PM +0800, Lijun Ou wrote:
> This patch mainly added icm support for RoCE. It initializes icm
> which managers the relative memory blocks for RoCE. The data
> structures of RoCE will be located in it. For example, CQ table,
> QP table and MTPT table so on.
I wonder i
On Wed, Jun 01, 2016 at 11:37:45PM +0800, Lijun Ou wrote:
> This patch mainly added the initial bare main driver. It
> could get the relative configure information of net node.
>
> Signed-off-by: Wei Hu
> Signed-off-by: Nenglong Zhao
> Signed-off-by: Lijun Ou
> ---
> drivers/infiniband/hw/hns/
On Wed, Jun 01, 2016 at 11:37:53PM +0800, Lijun Ou wrote:
> This patch registered IB device when loaded, and unregistered
> IB device when removed.
>
> Signed-off-by: Wei Hu
> Signed-off-by: Nenglong Zhao
> Signed-off-by: Lijun Ou
> ---
> drivers/infiniband/hw/hns/hns_roce_main.c | 46
> +
Hello Tien Hock
On 6/9/2016 7:48 AM, Tien Hock Loh wrote:
[snip]
.../devicetree/bindings/net/socfpga-dwmac.txt | 4 +
drivers/net/ethernet/stmicro/stmmac/Makefile | 2 +-
.../net/ethernet/stmicro/stmmac/dwmac-socfpga.c| 140 +--
drivers/net/ethernet/stmicro/stmmac/t
On 09/06/2016 08:06, Alexander Stein wrote:
> On Wednesday 08 June 2016 14:30:08, Rob Herring wrote:
>>> diff --git a/Documentation/devicetree/bindings/phy/phy-leds.txt
>>> b/Documentation/devicetree/bindings/phy/phy-leds.txt new file mode 100644
>>> index 000..1a35e3d
>>> --- /dev/null
>>> +
On 2016/06/08 10:19PM, Nilay Vaish wrote:
> Naveen, can you point out where in the patch you update the variable:
> idx, a member of codegen_contex structure? Somehow I am unable to
> figure it out. I can only see that we set it to 0 in the
> bpf_int_jit_compile function. Since all your test cas
On Wednesday 08 June 2016 14:30:08, Rob Herring wrote:
> > diff --git a/Documentation/devicetree/bindings/phy/phy-leds.txt
> > b/Documentation/devicetree/bindings/phy/phy-leds.txt new file mode 100644
> > index 000..1a35e3d
> > --- /dev/null
> > +++ b/Documentation/devicetree/bindings/phy/phy-l
From: Florian Westphal
Date: Wed, 8 Jun 2016 23:23:01 +0200
> When we need to create a new aggregate to enqueue the skb we call kzalloc.
> If that fails we returned ENOBUFS without freeing the skb.
>
> Spotted during code review.
>
> Signed-off-by: Florian Westphal
Applied.
From: Shweta Choudaha
Date: Wed, 8 Jun 2016 20:15:43 +0100
> The ip6 GRE tap device should not be forced to down state to change
> the mac address and should allow live address change for tap device
> similar to ipv4 gre.
>
> Signed-off-by: Shweta Choudaha
Applied to net-next, thanks.
On 6/8/2016 8:21 PM, Ben Dooks wrote:
The dwmac4_set_umac_addr() takes a struct mac_device_info as
the first parameter, but is being passed a ioaddr instead from
dwmac4_set_filter(). Fix the warning/bug by changing the first
parameter.
drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c:159:46: wa
The series is aimed at getting rid of CURRENT_TIME and CURRENT_TIME_SEC macros.
The macros are not y2038 safe. There is no plan to transition them into being
y2038 safe.
ktime_get_* api's can be used in their place. And, these are y2038 safe.
All filesystem timestamps use current_fs_time() for the
(Cc'ing netdev...)
On Mon, Jun 6, 2016 at 1:24 PM, Steven Caron wrote:
> Back in 2011, Bill Sommerfeld submitted an update to prevent ip_append_data
> to create malformed packets:
> Commit: d9be4f7a6f5a8da3133b832eca41c3591420b1ca
>
> Now we're finding that we need to apply the same logic to ip
On Wed, Jun 8, 2016 at 9:35 PM, David Ahern wrote:
>
>> On Jun 8, 2016, at 10:01 PM, Eric Dumazet wrote:
>>
>>> On Wed, Jun 8, 2016 at 8:41 PM, David Ahern
>>> wrote:
On 6/8/16 9:29 PM, Eric Dumazet wrote:
In case a qdisc is used on a vrf device, we need to use different
loc
From: Jakub Kicinski
Date: Wed, 8 Jun 2016 20:11:02 +0100
> These are incremental changes from v1 of cls_u32 fixes.
> First patch is reposted in its entirety, patch 2 is an
> incremental change from patch 2 of the original series.
Series applied, thanks.
> On Jun 8, 2016, at 10:01 PM, Eric Dumazet wrote:
>
>> On Wed, Jun 8, 2016 at 8:41 PM, David Ahern wrote:
>>> On 6/8/16 9:29 PM, Eric Dumazet wrote:
>>>
>>> In case a qdisc is used on a vrf device, we need to use different
>>> lockdep classes to avoid false positives.
>>>
>>> Fixes: f9eb8aea
On Wed, Jun 8, 2016 at 6:18 PM, William Tu wrote:
> The patch adds a new OVS action, OVS_ACTION_ATTR_TRUNC, in order to
> truncate packets. A 'max_len' is added for setting up the maximum
> packet size, and a 'cutlen' field is to record the number of bytes
> to trim the packet when the packet is o
Updates for arm64 eBPF JIT.
The main addition here is implementation of bpf_tail_call.
#1: Fix missing header inclusion in linux/bpf.h.
#2: Add bpf_tail_call for arm64.
#3,4: Optimizations to reduce instruction count for jitted code.
Changes since v2:
- None. Resubmit per David Miller.
Changes
Remove superfluous stack frame, saving us 3 instructions for every
LD_ABS or LD_IND.
Signed-off-by: Zi Shen Lim
---
arch/arm64/net/bpf_jit_comp.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c
index 7ae304e..b2fc97a 100644
---
Remove superfluous stack frame, saving us 3 instructions for
every JMP_CALL.
Signed-off-by: Zi Shen Lim
---
arch/arm64/net/bpf_jit_comp.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c
index 51abc97..7ae304e 100644
--- a/arch/a
Commit 0fc174dea545 ("ebpf: make internal bpf API independent of
CONFIG_BPF_SYSCALL ifdefs") introduced usage of ERR_PTR() in
bpf_prog_get(), however did not include linux/err.h.
Without this patch, when compiling arm64 BPF without CONFIG_BPF_SYSCALL:
...
In file included from arch/arm64/net/bpf_j
Add support for JMP_CALL_X (tail call) introduced by commit 04fd61ab36ec
("bpf: allow bpf programs to tail-call other bpf programs").
bpf_tail_call() arguments:
ctx - context pointer passed to next program
array - pointer to map which type is BPF_MAP_TYPE_PROG_ARRAY
index - index inside ar
Removed most of the module parameters
Tested in a rack using between 1 and 380 active TCP-NV flows.
Consists of the following patches:
[PATCH net-next v2 1/2] tcp: add in_flight to tcp_skb_cb
[PATCH net-next v2 2/2] tcp: add NV congestion control
Signed-off-by: Lawrence Brakmo
include/net/tcp
TCP-NV (New Vegas) is a major update to TCP-Vegas.
An earlier version of NV was presented at 2010's LPC.
It is a delayed based congestion avoidance for the
data center. This version has been tested within a
10G rack where the HW RTTs are 20-50us and with
1 to 400 flows.
A description of TCP-NV, in
Add in_flight (bytes in flight when packet was sent) field
to tx component of tcp_skb_cb and make it available to
congestion modules' pkts_acked() function through the
ack_sample function argument.
Signed-off-by: Lawrence Brakmo
Acked-by: Yuchung Cheng
---
include/net/tcp.h | 2 ++
net/ipv4
On Wed, Jun 8, 2016 at 8:41 PM, David Ahern wrote:
> On 6/8/16 9:29 PM, Eric Dumazet wrote:
>>
>> In case a qdisc is used on a vrf device, we need to use different
>> lockdep classes to avoid false positives.
>>
>> Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a
>> seqcount")
>
From: David Ahern
Date: Wed, 8 Jun 2016 19:31:51 -0600
> On 6/8/16 1:08 PM, David Ahern wrote:
>> Getting the following splat with ping in a VRF:
>
> ...
>
>> Fix by adding the IFF_NO_QUEUE flag for VRF devices.
>>
>> Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a
>> seqcou
On 6/8/16 9:29 PM, Eric Dumazet wrote:
In case a qdisc is used on a vrf device, we need to use different
lockdep classes to avoid false positives.
Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a seqcount")
Reported-by: David Ahern
Signed-off-by: Eric Dumazet
same result w
In case a qdisc is used on a vrf device, we need to use different
lockdep classes to avoid false positives.
Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a seqcount")
Reported-by: David Ahern
Signed-off-by: Eric Dumazet
---
drivers/net/vrf.c | 2 +-
1 file changed, 1 inserti
In case a qdisc is used on a macvlan device, we need to use different
lockdep classes to avoid false positives.
Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a seqcount")
Fixes: 79640a4ca695 ("net: add additional lock to qdisc to increase throughput")
Signed-off-by: Eric Dumaze
In case a qdisc is used on a ipvlan device, we need to use different
lockdep classes to avoid false positives.
Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a seqcount")
Fixes: 79640a4ca695 ("net: add additional lock to qdisc to increase throughput")
Signed-off-by: Eric Dumazet
It is time to add netdev_lockdep_set_classes() helper
so that lockdep annotations per device type are easier to manage.
Signed-off-by: Eric Dumazet
---
drivers/net/bonding/bond_main.c | 24 +---
drivers/net/ppp/ppp_generic.c | 6 +-
drivers/net/team/team.c | 21
Naveen, can you point out where in the patch you update the variable:
idx, a member of codegen_contex structure? Somehow I am unable to
figure it out. I can only see that we set it to 0 in the
bpf_int_jit_compile function. Since all your test cases pass, I am
clearly overlooking something.
Than
On Thu, 2016-06-09 at 09:43 +0800, Su Xuemin wrote:
> On Wed 2016-06-08 at 08:43 -0700, Eric Dumazet wrote:
> > I am not convinced this is the right way to fix the issue.
> >
> > Fact that you did not change udp4_lib_lookup2() is telling me something.
> >
> >
> > 1) If host has 100 different IP,
On Wed, Jun 08, 2016 at 08:44:53PM -0400, Vivien Didelot wrote:
> Extract the allocation and registration code related to the dsa_switch
> structure in a mv88e6xxx_register_switch helper function.
>
> For symmetry in the code, add a mv88e6xxx_unregister_switch function.
You say what you are doing
On Wed, Jun 08, 2016 at 08:44:52PM -0400, Vivien Didelot wrote:
> The MDIO device probe and remove functions are respectively incrementing
> and decrementing the bus refcount themselves. Since these bus level
> actions are out of the device scope, remove them.
I agree with the patch. But have you
On Wed, Jun 08, 2016 at 08:44:51PM -0400, Vivien Didelot wrote:
> In the MDIO probing function, dev is already assigned to &mdiodev->dev
> and np is already assigned to mdiodev->dev.of_node, so use them.
>
> Signed-off-by: Vivien Didelot
Reviewed-by: Andrew Lunn
Andrew
On Wed, Jun 08, 2016 at 08:44:50PM -0400, Vivien Didelot wrote:
> The chip->ds and ds->slave_mii_bus assignments are common to both legacy
> and new MDIO probing and are already done in the later setup code.
>
> Remove the duplicated assignments from the MDIO probing code.
>
> Signed-off-by: Vivi
On Wed, Jun 08, 2016 at 08:44:49PM -0400, Vivien Didelot wrote:
> This patch fixes 5 style problems reported by checkpatch:
>
> WARNING: suspect code indent for conditional statements (8, 24)
> #492: FILE: drivers/net/dsa/mv88e6xxx.c:492:
> + if (phydev->link)
> + r
On Wed, Jun 08, 2016 at 08:44:56PM -0400, Vivien Didelot wrote:
> Now that we have access at probe time to the chip info described in the
> device tree, check if the probed device matches the device node,
> otherwise warn the user and fail.
What good is this? So what if the device tree says a diff
On Wed, Jun 08, 2016 at 08:44:55PM -0400, Vivien Didelot wrote:
> Thanks to the new device probing, we can explicit the exact switch model
> in the device tree.
>
> Name the driver "mv88e6xxx" and list all its compatible supported chips.
No. This goes against the usual way of doing device tree co
On Wed, Jun 8, 2016 at 6:49 PM, David Ahern wrote:
> On 6/8/16 7:43 PM, Eric Dumazet wrote:
>>
>> For this one, it looks vrf misses the _xmit_lock lockdep support.
>>
>> We might need to factorize the code found for example in
>> bond_set_lockdep_class_one()
>>
>> Have you run lockdep before ? Str
On 6/8/16 7:43 PM, Eric Dumazet wrote:
For this one, it looks vrf misses the _xmit_lock lockdep support.
We might need to factorize the code found for example in
bond_set_lockdep_class_one()
Have you run lockdep before ? Strange that these issues were not
spotted earlier.
Standard config for
On Wed 2016-06-08 at 08:43 -0700, Eric Dumazet wrote:
> I am not convinced this is the right way to fix the issue.
>
> Fact that you did not change udp4_lib_lookup2() is telling me something.
>
>
> 1) If host has 100 different IP, and 10 sockets bound to 0.0.0.0:53
> 2) If we receive datagrams
On Wed, Jun 8, 2016 at 6:26 PM, David Ahern wrote:
> On 6/8/16 5:55 PM, Eric Dumazet wrote:
>>
>> In case a qdisc is used on a vrf device, we need to use different
>> lockdep classes to avoid false positives.
>>
>> Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a
>> seqcount")
>
On 6/8/16 1:08 PM, David Ahern wrote:
Getting the following splat with ping in a VRF:
...
Fix by adding the IFF_NO_QUEUE flag for VRF devices.
Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a seqcount")
Signed-off-by: David Ahern
---
drivers/net/vrf.c | 2 ++
1 file chan
On 6/8/16 5:55 PM, Eric Dumazet wrote:
In case a qdisc is used on a vrf device, we need to use different
lockdep classes to avoid false positives.
Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a seqcount")
Reported-by: David Ahern
Signed-off-by: Eric Dumazet
---
drivers/ne
The patch adds a new OVS action, OVS_ACTION_ATTR_TRUNC, in order to
truncate packets. A 'max_len' is added for setting up the maximum
packet size, and a 'cutlen' field is to record the number of bytes
to trim the packet when the packet is outputting to a port, or when
the packet is sent to userspac
This patchset cleans a bit both legacy and the new MDIO probing functions by
extracting common code and adding helpers.
Then, thanks to the new probing model allowing switch chips to be true Linux
devices, a device node can explicit the switch model in the tree and thus
allows a more robust probin
The MDIO device probe and remove functions are respectively incrementing
and decrementing the bus refcount themselves. Since these bus level
actions are out of the device scope, remove them.
Signed-off-by: Vivien Didelot
---
drivers/net/dsa/mv88e6xxx.c | 3 ---
1 file changed, 3 deletions(-)
di
Extract the common code to read the switch ID and allocate the private
chip data, found in both legacy and new MDIO probe functions, into its
own mv88e6xxx_detect helper to reduce boiler plate.
Signed-off-by: Vivien Didelot
---
drivers/net/dsa/mv88e6xxx.c | 68 +++
Extract the allocation and registration code related to the dsa_switch
structure in a mv88e6xxx_register_switch helper function.
For symmetry in the code, add a mv88e6xxx_unregister_switch function.
Signed-off-by: Vivien Didelot
---
drivers/net/dsa/mv88e6xxx.c | 41 -
This patch fixes 5 style problems reported by checkpatch:
WARNING: suspect code indent for conditional statements (8, 24)
#492: FILE: drivers/net/dsa/mv88e6xxx.c:492:
+ if (phydev->link)
+ reg |= PORT_PCS_CTRL_LINK_UP;
CHECK: Logical continuations should be
Now that we have access at probe time to the chip info described in the
device tree, check if the probed device matches the device node,
otherwise warn the user and fail.
Signed-off-by: Vivien Didelot
---
drivers/net/dsa/mv88e6xxx.c | 26 ++
1 file changed, 26 insertions(
The chip->ds and ds->slave_mii_bus assignments are common to both legacy
and new MDIO probing and are already done in the later setup code.
Remove the duplicated assignments from the MDIO probing code.
Signed-off-by: Vivien Didelot
---
drivers/net/dsa/mv88e6xxx.c | 3 ---
1 file changed, 3 dele
Thanks to the new device probing, we can explicit the exact switch model
in the device tree.
Name the driver "mv88e6xxx" and list all its compatible supported chips.
In the meantime, rename the of_device_id table to avoid confusion with a
later introduce matching function.
Signed-off-by: Vivien
In the MDIO probing function, dev is already assigned to &mdiodev->dev
and np is already assigned to mdiodev->dev.of_node, so use them.
Signed-off-by: Vivien Didelot
---
drivers/net/dsa/mv88e6xxx.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/net/dsa/mv88e6xx
On Wed, Jun 08, 2016 at 03:25:38PM -0700, Vincent Palatin wrote:
> On Tue, Jun 7, 2016 at 12:23 AM, Giuseppe CAVALLARO
> wrote:
> > Hello
> >
> > On 6/3/2016 7:29 PM, Vincent Palatin wrote:
> >>
> >> Do not shutdown the PHY if Wake-on-Lan is enabled, else it cannot wake
> >> us up.
> >>
> >
> > I
On 09.06.16 02:11, Schuyler Patton wrote:
On 06/08/2016 09:06 AM, Ivan Khoronzhuk wrote:
On 08.06.16 17:01, Ivan Khoronzhuk wrote:
Hi Schuyer,
On 07.06.16 18:26, Schuyler Patton wrote:
Hi,
On 06/07/2016 08:59 AM, Ivan Khoronzhuk wrote:
There is no reason in rx_descs property because d
In case a qdisc is used on a ipvlan device, we need to use different
lockdep classes to avoid false positives.
Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a seqcount")
Fixes: 79640a4ca695 ("net: add additional lock to qdisc to increase throughput")
Signed-off-by: Eric Dumazet
In case a qdisc is used on a macvlan device, we need to use different
lockdep classes to avoid false positives.
Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a seqcount")
Fixes: 79640a4ca695 ("net: add additional lock to qdisc to increase throughput")
Signed-off-by: Eric Dumaze
It is time to add netdev_lockdep_set_classes() helper
so that lockdep annotations per device type are easier to manage.
Signed-off-by: Eric Dumazet
---
drivers/net/bonding/bond_main.c | 5 +
drivers/net/ppp/ppp_generic.c | 6 +-
drivers/net/team/team.c | 5 +
include/linux/
In case a qdisc is used on a vrf device, we need to use different
lockdep classes to avoid false positives.
Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a seqcount")
Reported-by: David Ahern
Signed-off-by: Eric Dumazet
---
drivers/net/vrf.c | 2 +-
1 file changed, 1 inserti
On Wed, 8 Jun 2016 16:16:01 -0700
Sridhar Samudrala wrote:
> On devices that support TC U32 offloads, these flags enable a filter to be
> added only to HW or only to SW. skip_sw and skip_hw are mutually exclusive
> flags. By default without any flags, the filter is added to both HW and SW,
> but
On devices that support TC U32 offloads, these flags enable a filter to be
added only to HW or only to SW. skip_sw and skip_hw are mutually exclusive
flags. By default without any flags, the filter is added to both HW and SW,
but no error checks are done in case of failure to add to HW.
With skip-s
On 06/08/2016 09:06 AM, Ivan Khoronzhuk wrote:
On 08.06.16 17:01, Ivan Khoronzhuk wrote:
Hi Schuyer,
On 07.06.16 18:26, Schuyler Patton wrote:
Hi,
On 06/07/2016 08:59 AM, Ivan Khoronzhuk wrote:
There is no reason in rx_descs property because davinici_cpdma
driver splits pool of descripto
On Wed, 2016-06-08 at 09:28 +0200, Johannes Thumshirn wrote:
> On Tue, Jun 07, 2016 at 09:44:00AM +0200, Johannes Thumshirn wrote:
> > The first patch in this series introduces the following 4 helper
> functions to
> > the PCI core:
> >
> > * pci_request_mem_regions()
> > * pci_request_io_regions(
On Wed, 2016-06-08 at 08:33 -0700, Stephen Hemminger wrote:
> Applied but I had to fix a couple of conflicts because recent checkpatch
> cleanup
> of tc files.
Thanks Stephen, this looks good to me.
Earlier commits removed two members from struct Qdisc which places
next_sched/gso_skb into a different cacheline than ->state.
This restores the struct layout to what it was before the removal.
Move the two members, then add an annotation so they all reside in the
same cacheline.
This adds a 16 b
after removal of TCA_CBQ_OVL_STRATEGY from cbq scheduler, there are no
more callers of ->drop() outside of other ->drop functions, i.e.
nothing calls them.
Signed-off-by: Florian Westphal
---
include/net/sch_generic.h | 36
net/sched/sch_atm.c | 15
iproute2 doesn't implement any cbq option that results in this attribute
being sent to kernel.
To make use of it, user would have to
- patch iproute2
- add a class
- attach a qdisc to the class (default pfifo doesn't work as
q->handle is 0 and cbq_set_police() is a no-op in this case)
- re-'add
After the removal of TCA_CBQ_POLICE in cbq scheduler qdisc->reshape_fail
is always NULL, i.e. qdisc_rehape_fail is now the same as qdisc_drop.
Signed-off-by: Florian Westphal
---
include/net/sch_generic.h | 19 ---
net/sched/sch_fifo.c | 4 ++--
net/sched/sch_netem.c |
since initial revision of cbq in 2004 iproute 2 has never implemented
support for TCA_CBQ_OVL_STRATEGY, which is what needs to be set to
activate the class->drop() call (TC_CBQ_OVL_DROP strategy must be
set by userspace value must be set by userspace).
David Miller says:
It seems really safe to
iproute2 does not implement any options that result in the
TCA_CBQ_OVL_STRATEGY/TCA_CBQ_POLICE attributes being set/used.
This series removes these two attributes from cbq and makes kernel reject
them via EOPNOTSUPP in case they are present.
The two followup changes then remove several features
On Tue, Jun 7, 2016 at 12:23 AM, Giuseppe CAVALLARO
wrote:
> Hello
>
> On 6/3/2016 7:29 PM, Vincent Palatin wrote:
>>
>> Do not shutdown the PHY if Wake-on-Lan is enabled, else it cannot wake
>> us up.
>>
>
> I do not understand why you need that.
> This is done inside the PHY layer and it is test
On Wed, 2016-06-08 at 23:01 +0200, Florian Westphal wrote:
>
> Would you mind an annotation rather than covering the hole?
>
> --- a/include/net/sch_generic.h
> +++ b/include/net/sch_generic.h
> @@ -71,11 +71,11 @@ struct Qdisc {
> struct gnet_stats_basic_cpu __percpu *cpu_bstats;
>
All devices that are missing busylock were already buggy ;)
On Wed, Jun 8, 2016 at 1:59 PM, David Ahern wrote:
> On 6/8/16 2:32 PM, Eric Dumazet wrote:
>>
>> On Wed, Jun 8, 2016 at 1:31 PM, David Ahern
>> wrote:
>>>
>>> On 6/8/16 2:30 PM, Eric Dumazet wrote:
It looks VRF was
On 06/06/2016 09:52 PM, Cong Wang wrote:
On Mon, Jun 6, 2016 at 12:25 PM, Daniel Borkmann wrote:
[...]
This is fundamental for libnl to update caches.
I don't understand why it should be separated, since notification is
not a feature, we already have notifications in other paths.
Looking in
When we need to create a new aggregate to enqueue the skb we call kzalloc.
If that fails we returned ENOBUFS without freeing the skb.
Spotted during code review.
Signed-off-by: Florian Westphal
---
V2: also increment cl->qstats.drops
net/sched/sch_qfq.c | 6 --
1 file changed, 4 insertion
Eric Dumazet wrote:
> On Wed, 2016-06-08 at 14:31 +0200, Florian Westphal wrote:
> > When we need to create a new aggregate to enqueue the skb we need
> > to kzalloc the new struct.
> >
> > If that fails we returned ENOBUFS without freeing the skb.
> >
> > Spotted during code review.
> >
> > Si
Eric Dumazet wrote:
> On Wed, 2016-06-08 at 17:35 +0200, Florian Westphal wrote:
> > After the removal of TCA_CBQ_POLICE in cbq scheduler qdisc->reshape_fail
> > is always NULL, i.e. qdisc_rehape_fail is now the same as qdisc_drop.
> >
> > Signed-off-by: Florian Westphal
> > ---
> > include/net
On 6/8/16 2:32 PM, Eric Dumazet wrote:
On Wed, Jun 8, 2016 at 1:31 PM, David Ahern wrote:
On 6/8/16 2:30 PM, Eric Dumazet wrote:
It looks VRF was already lacking annotation for busylock then ?
I sent a patch adding IFF_NO_QUEUE flag
OK, but we still need the fix, in case a qdisc is ins
poport
> > Signed-off-by: Julia Lawall
> > Signed-off-by: Fengguang Wu
>
> This doesn't apply cleanly to any of my trees.
Sorry not to have included the source information. It seems to be from
here:
https://github.com/0day-ci/linux/commits/Mike-Rapoport/virtio_net-add-_UAPI-prefix-to-virtio_net-header-guards/20160608-211558
julia
Since msleep is based on jiffies the PHY reset could take longer
than expected. So use msleep for values greater than 20 msec otherwise
usleep_range.
Signed-off-by: Stefan Wahren
---
drivers/net/ethernet/freescale/fec_main.c |7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --
On Wed, Jun 8, 2016 at 1:31 PM, David Ahern wrote:
> On 6/8/16 2:30 PM, Eric Dumazet wrote:
>>
>>
>> It looks VRF was already lacking annotation for busylock then ?
>
>
> I sent a patch adding IFF_NO_QUEUE flag
>
OK, but we still need the fix, in case a qdisc is installed on vrf
(who knows...)
I
On 6/8/16 2:30 PM, Eric Dumazet wrote:
It looks VRF was already lacking annotation for busylock then ?
I sent a patch adding IFF_NO_QUEUE flag
- Original Message -
> From: "Nicolas Dichtel"
> To: "Vincent Bernat"
> Cc: "David S. Miller" , "Vijay Pandurangan"
> , "Paolo Abeni"
> , netdev@vger.kernel.org
> Sent: Tuesday, May 31, 2016 5:17:20 AM
> Subject: Re: [net v3] veth: advertise peer link once both links are tied
> together
On Wed, Jun 8, 2016 at 11:17 AM, David Ahern wrote:
> Getting the below splat with ping in a VRF. git bisect points to:
>
> dsa@kenny:~/kernel-2.git$ git bisect good
> f9eb8aea2a1e12fc2f584d1627deeb957435a801 is the first bad commit
> commit f9eb8aea2a1e12fc2f584d1627deeb957435a801
> Author: Eric
On Tue, Jun 07, 2016 at 04:59:36PM +0300, Ivan Khoronzhuk wrote:
> There is no reason to hold s/w dependent parameter in device tree.
> Even more, there is no reason in this parameter because davinici_cpdma
> driver splits pool of descriptors equally between tx and rx channels
> anyway.
>
> Signed
On Mon, Jun 06, 2016 at 08:01:56PM -0700, th...@altera.com wrote:
> From: Tien Hock Loh
>
> This adds support for TSE PCS that uses SGMII adapter when the phy-mode of
> the dwmac is set to sgmii
>
> Signed-off-by: Tien Hock Loh
>
> ---
> v2:
> - Refactored the TSE PCS out from the dwmac-socfpg
On Mon, Jun 06, 2016 at 06:11:36PM +0530, Pramod Kumar wrote:
> Add DT binding doc for Broadcom MDIO bus multiplexer driver.
>
> Reviewed-by: Andrew Lunn
> Signed-off-by: Pramod Kumar
> ---
> .../bindings/net/brcm,mdio-mux-iproc.txt | 59
> ++
> 1 file changed, 59
On Tue, Jun 07, 2016 at 05:27:29PM +0530, Pramod Kumar wrote:
> Hi Florian,
>
> > -Original Message-
> > From: Florian Fainelli [mailto:f.faine...@gmail.com]
> > Sent: 07 June 2016 00:06
> > To: Pramod Kumar; Rob Herring; Pawel Moll; Mark Rutland; Ian Campbell;
> > Kumar Gala; Catalin Mari
On Mon, Jun 06, 2016 at 06:11:35PM +0530, Pramod Kumar wrote:
> Change "mdio-parent-bus" from mandatory section to optional
> as it won't be required by integrated MDIO multiplexer
> which has bus selection and mdio transaction generation logic,
> integrated inside.
>
> Signed-off-by: Pramod Kumar
On Sun, Jun 05, 2016 at 11:45:49PM +0200, Hauke Mehrtens wrote:
> This binding makes it possible to control the LEDs of an Ethernet PHY.
> These settings allow it to abstract the hardware configuration which
> tells the hardware when to switch the LED constant on or blink for
> example. This will b
On Wed, 08 Jun 2016 11:09:36 -0700 (PDT), David Miller wrote:
> From: Jakub Kicinski
> Date: Wed, 8 Jun 2016 11:18:30 +0100
>
> > On Tue, 07 Jun 2016 16:27:31 -0700 (PDT), David Miller wrote:
> >> From: Jakub Kicinski
> >> Date: Mon, 6 Jun 2016 16:16:46 +0100
> >>
> >> > This set fixes two
The ip6 GRE tap device should not be forced to down state to change
the mac address and should allow live address change for tap device
similar to ipv4 gre.
Signed-off-by: Shweta Choudaha
---
net/ipv6/ip6_gre.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/
On Mon, Jun 06, 2016 at 08:10:54PM +0200, Corentin LABBE wrote:
> Le 06/06/2016 16:14, Rob Herring a écrit :
> > On Fri, Jun 03, 2016 at 11:56:28AM +0200, LABBE Corentin wrote:
> >> This patch adds documentation for Device-Tree bindings for the
> >> Allwinner sun8i-emac driver.
> >>
> >> Signed-off
Return an error if user requested skip-sw and the underlaying
hardware cannot handle tc offloads (or offloads are disabled).
This patch fixes the knode handling.
Signed-off-by: Jakub Kicinski
---
net/sched/cls_u32.c | 37 +++--
1 file changed, 19 insertions(+), 18
Errors reported by u32_replace_hw_hnode() were not propagated.
Signed-off-by: Jakub Kicinski
Acked-by: Sridhar Samudrala
---
net/sched/cls_u32.c | 8 +++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c
index 27b99fd774d7..54ab32a8ff4c
These are incremental changes from v1 of cls_u32 fixes.
First patch is reposted in its entirety, patch 2 is an
incremental change from patch 2 of the original series.
Jakub Kicinski (2):
net: cls_u32: catch all hardware offload errors
net: cls_u32: be more strict about skip-sw flag (knode part
1 - 100 of 210 matches
Mail list logo
