On Mon, Jul 06, 2015 at 02:46:49PM -0700, Stephen Hemminger wrote:
> On Mon, 29 Jun 2015 10:53:15 +0200 (CEST)
> Michal Kubecek wrote:
>
> > Copy of kernel include/uapi/linux/tipc.h is needed to build on systems
> > with pre-3.16 kernel headers.
> >
> > Signed-off-by: Michal Kubecek
>
> Ok, I
Frag needed should be sent only if the inner header asked
to not fragment. Currently fragmentation is broken if the
tunnel has df set, but df was not asked in the original
packet. The tunnel's df needs to be still checked to update
internally the pmtu cache.
Commit 23a3647bc4f93bac broke it, and t
I ran some tests and after setting the "ip route change quickack 1"
knob there were still many delayed ACKs sent. This occured
because when icsk_ack.quick=0 the !icsk_ack.pingpong value is
subsequently ignored as tcp_in_quickack_mode() checks both these
values. The condition for a quick ack to
On Mon, Jul 6, 2015 at 7:00 PM, Andrew Lunn wrote:
> On Tue, Jul 07, 2015 at 01:38:04AM +0200, Andrew Lunn wrote:
>> On Sun, Jul 05, 2015 at 10:14:50PM -0400, Vivien Didelot wrote:
>> > Hi all,
>> >
>> > This patchset brings full support for hardware VLANs in DSA, and the
>> > Marvell
>> > 88E6xx
From: Matteo Croce
Date: Mon, 6 Jul 2015 21:44:06 +0200
> 2015-07-06 12:49 GMT+02:00 :
>> On Thu, 02 Jul 2015 10:56:01 +0200, Matteo Croce said:
>>> Add option to disable any reply not related to a listening socket,
>>> like RST/ACK for TCP and ICMP Port-Unreachable for UDP.
>>> Also disables IC
> +static int _mv88e6xxx_vtu_getnext(struct dsa_switch *ds, u16 vid,
> + struct mv88e6xxx_vtu_entry *entry)
> +{
> + int ret, i;
> +
> + ret = _mv88e6xxx_vtu_wait(ds);
> + if (ret < 0)
> + return ret;
> +
> + ret = _mv88e6xxx_reg_write(ds, R
On Tue, Jul 07, 2015 at 01:38:04AM +0200, Andrew Lunn wrote:
> On Sun, Jul 05, 2015 at 10:14:50PM -0400, Vivien Didelot wrote:
> > Hi all,
> >
> > This patchset brings full support for hardware VLANs in DSA, and the Marvell
> > 88E6xxx compatible switch chips.
>
> Hi Vivien
>
> I just booted the
On 7/6/15 1:05 PM, Eric Dumazet wrote:
Add multiqueue capabilities to ifb netdevice.
This removes last bottleneck for ingress when mq qdisc can be used
to shard load from multiple RX queues on physical device.
...
Signed-off-by: Eric Dumazet
Cc: Alexei Starovoitov
Cc: Jamal Hadi Salim
Cc: John
On 7/3/15, 1:30 PM, "Neal Cardwell" wrote:
>> diff --git a/include/linux/tcp.h b/include/linux/tcp.h
>> index 48c3696..05e0da5 100644
>> --- a/include/linux/tcp.h
>> +++ b/include/linux/tcp.h
>> @@ -254,6 +254,10 @@ struct tcp_sock {
>> u32 lost_out; /* Lost packets
On Sun, Jul 05, 2015 at 10:14:50PM -0400, Vivien Didelot wrote:
> Hi all,
>
> This patchset brings full support for hardware VLANs in DSA, and the Marvell
> 88E6xxx compatible switch chips.
Hi Vivien
I just booted these patches on my board, and i'm getting WARNINGS:
[ 61.111302] WARNING: CPU:
On Mon, 6 Jul 2015 09:57:37 -0700
Tom Lendacky wrote:
> When running a kernel configured with CONFIG_DMA_API_DEBUG=y a warning
> is issued:
> DMA-API: device driver tries to sync DMA memory it has not allocated
>
> This warning is the result of mapping the full range of the Rx buffer
> pages al
On 7/3/15, 10:10 AM, "Tom Herbert" wrote:
>On Thu, Jul 2, 2015 at 6:21 PM, Lawrence Brakmo wrote:
>> This is a request for comments.
>>
>> TCP-NV (New Vegas) is a major update to TCP-Vegas. An earlier version of
>> NV was presented at 2010's LPC (slides). It is a delayed based
>> congestion av
On 7/3/15, 9:47 AM, "David Miller" wrote:
>From: Lawrence Brakmo
>Date: Thu, 2 Jul 2015 18:21:24 -0700
>
>> This is a request for comments.
>>
>> TCP-NV (New Vegas) is a major update to TCP-Vegas. An earlier version of
>> NV was presented at 2010's LPC (slides). It is a delayed based
>> conge
Jason Gunthorpe reported that since commit c02db8c6290b ("rtnetlink: make
SR-IOV VF interface symmetric"), we don't verify IFLA_VF_INFO attributes
anymore with respect to their policy, that is, ifla_vfinfo_policy[].
Before, they were part of ifla_policy[], but they have been nested since
placed un
On Mon, 29 Jun 2015 10:53:15 +0200 (CEST)
Michal Kubecek wrote:
> Copy of kernel include/uapi/linux/tipc.h is needed to build on systems
> with pre-3.16 kernel headers.
>
> Signed-off-by: Michal Kubecek
Ok, I applied (and fixed) this.
The headers you want are not directly from include/uapi/lin
On Thu, Jul 02, 2015 at 06:14:48PM -0700, Christopher Hall wrote:
> * getsynctime64()
Hello Christopher-
A couple comments below.
>
> This takes 2 arguments referring to system and device time
>
> With this callback drivers may provide both system time and device time
> to ensure precise corre
On Mon, Jul 6, 2015 at 1:35 AM, Timo Teräs wrote:
> Frag needed should be sent only if the inner header asked
> to not fragment. Currently fragmentation is broken if the
> tunnel has df set. The tunnel's df needs to be still checked
> to update internally the pmtu cache.
>
> This got broken in com
From: Eric Dumazet
Add multiqueue capabilities to ifb netdevice.
This removes last bottleneck for ingress when mq qdisc can be used
to shard load from multiple RX queues on physical device.
Tested:
# netem based setup, installed at receiver side
ETH=eth0
IFB=ifb10
EST="est 1sec 4sec" # Optiona
When packet encapsulation is in use, the MTU needs to be reduced for
headroom reservation.
The existing code takes the updated MTU value only from the host side.
But vSwitch extensions, such as Open vSwitch, require the flexibility
to change the MTU to different values from within a guest during th
> -Original Message-
> From: David Miller [mailto:da...@davemloft.net]
> Sent: Friday, July 3, 2015 12:17 PM
> To: dan.carpen...@oracle.com
> Cc: Haiyang Zhang; netdev@vger.kernel.org; o...@aepfle.de;
> jasow...@redhat.com; driverdev-de...@linuxdriverproject.org; linux-
> ker...@vger.kern
2015-07-06 12:49 GMT+02:00 :
> On Thu, 02 Jul 2015 10:56:01 +0200, Matteo Croce said:
>> Add option to disable any reply not related to a listening socket,
>> like RST/ACK for TCP and ICMP Port-Unreachable for UDP.
>> Also disables ICMP replies to echo request and timestamp.
>> The stealth mode ca
On Wed, 24 Jun 2015 11:55:38 +0900
Taku Izumi wrote:
> This patch adds buffer address regist/unregistration routine.
>
> This function is mainly invoked when network device's
> activation (open) and deactivation (close)
> in order to retist/unregist shared buffer address.
registe
On 7/6/15 7:20 AM, Michael Holzheu wrote:
The trace bpf samples do not compile on s390x because they use x86
specific fields from the "pt_regs" structure.
Fix this and access the fields via new PT_REGS macros.
Signed-off-by: Michael Holzheu
---
samples/bpf/bpf_helpers.h | 25
On 07/06/2015 01:37 PM, Vitaly Andrianov wrote:
> This commit adds a MIB entry for out data chunk discards.
> Number of outgoing SCTP DATA chunks for a SCTP association for which no
> problems were encountered to prevent their transmission but were discarded.
> Data chunks are discarded due to ungr
When a switch is attached to the mdio bus, the mdio bus can be used
while the interface is not open. If the IPG clock is not enabled, MDIO
reads/writes will simply time out.
Add support for runtime PM to control this clock. Enable/disable this
clock using runtime PM, with open()/close() and mdio r
On 7/6/15 5:18 AM, Eric Dumazet wrote:
qdisc_bstats_update_cpu() and other helpers were added to support
percpu stats for qdisc.
We want to add percpu stats for tc action, so this patch add common
helpers.
qdisc_bstats_update_cpu() is renamed to qdisc_bstats_cpu_update()
qdisc_qstats_drop_cpu()
On 7/6/15 5:18 AM, Eric Dumazet wrote:
Like act_gact, act_mirred can be lockless in packet processing
1) Use percpu stats
2) update lastuse only every clock tick to avoid false sharing
3) use rcu to protect tcfm_dev
4) Remove spinlock usage, as it is no longer needed.
Next step : add multi queu
ifb patch seems to work very well ;)
# tc -s -d qd sh dev ifb10
qdisc mq 1: root
Sent 190952 bytes 31798616 pkt (dropped 0, overlimits 0 requeues 0)
backlog 29460b 491p requeues 0
qdisc netem 8002: parent 1:1 limit 10 delay 3.0ms
Sent 238320936 bytes 3971225 pkt (dropped 0, overlimits 0
This kernel patch exports the value of the new
ignore_routes_with_linkdown via netconf.
Signed-off-by: Andy Gospodarek
Suggested-by: Nicolas Dichtel
---
include/uapi/linux/netconf.h | 1 +
net/ipv4/devinet.c | 7 +++
2 files changed, 8 insertions(+)
diff --git a/include/uapi/linu
No no problem,
Just trying to get the functional aspects worked out. the global
search replace will be easy.
Was hoping to see some more responses on the naming suggestions here
from the community. If there is not disagreement we can spin patches
with MRF as the name.
On Mon, Jul 6, 2015 at 8:4
On 07/06/2015 07:24 PM, Huan Le wrote:
Erik Hugne ericsson.com> writes:
I'm having troubles with TC policing, the ingress filters does not seem to be
applied.
Kernel: net-next/latest
iproute2: shemminger/master
//E
I observed similar behavior when configuring a filter on ingress qdisc.
On Mon, 2015-07-06 at 19:13 +0800, Bob Liu wrote:
>
> Thank you for the quick fix!
> Tested by rebooting several times and didn't hit this panic any more.
Thanks Bob, I will submit an official patch then ;)
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a m
Erik Hugne ericsson.com> writes:
>
> I'm having troubles with TC policing, the ingress filters does not seem to be
> applied.
>
>
> Kernel: net-next/latest
> iproute2: shemminger/master
>
> //E
>
I observed similar behavior when configuring a filter on ingress qdisc.
Test shows that traffi
This commit adds a MIB entry for out data chunk discards.
Number of outgoing SCTP DATA chunks for a SCTP association for which no
problems were encountered to prevent their transmission but were discarded.
Data chunks are discarded due to ungraceful closing of the SCTP
association.
Signed-off-by:
> On Jun 26, 2015, at 11:04 AM, Rustad, Mark D wrote:
>
>> Sorry, Mark, I've just been busy with other issues and haven't had a
>> chance to look at this yet.
>
> Is there any chance of this getting into this merge window?
Well, it has missed the merge window, but this really is a bug fix. Thes
2015-07-05 10:16 GMT-07:00 Rob Herring :
> On UML builds, mdio-mux-mmioreg.c fails to compile:
>
> drivers/net/phy/mdio-mux-mmioreg.c:50:3: error: implicit declaration of
> function ‘ioremap’ [-Werror=implicit-function-declaration]
> drivers/net/phy/mdio-mux-mmioreg.c:63:3: error: implicit declara
When running a kernel configured with CONFIG_DMA_API_DEBUG=y a warning
is issued:
DMA-API: device driver tries to sync DMA memory it has not allocated
This warning is the result of mapping the full range of the Rx buffer
pages allocated and then performing a dma_sync_single_for_cpu against
a cal
On 7/6/15 10:37 AM, Nikolay Aleksandrov wrote:
+static int vrf_add_slave(struct net_device *dev,
+struct net_device *port_dev)
+{
+ if (!dev || !port_dev || dev_net(dev) != dev_net(port_dev))
+ return -ENODEV;
+
+ if (!vrf_is_master(port_dev) && !
On 06.07.2015 17:25, Nicolas Dichtel wrote:
This reverts commit e1622baf54df8cc958bf29d71de5ad545ea7d93c.
The side effect of this commit is to add a '@NONE' after each virtual
interface name with a 'ip link'. It may break existing scripts.
Reported-by: Olivier Hartkopp
^^^
On 07/06/2015 05:03 PM, David Ahern wrote:
> This driver borrows heavily from IPvlan and teaming drivers.
>
> Routing domains (VRF-lite) are created by instantiating a device
> and enslaving all routed interfaces that participate in the domain.
> As part of the enslavement, all local routes pointi
Set speed only on ethtool..Driver is r8169 .
From 84ea65a6d1c2bc9bc28e8f54df5766395ea9ce05 Mon Sep 17 00:00:00 2001
From: Corcodel Marian
Date: Mon, 6 Jul 2015 16:33:18 +0300
Subject: [net-next Disable setting speed outside the ethtool] Committer:
Corcodel Marian
To: netdev@vger.kernel.org
MIME
Thanks for your reply, Tom!
> How is the tunnel/session being created on the server side?
My server is xl2tpd. If I understand correctly, session and tunnel are
being created in start_pppd() function, see [1].
Judging from xl2tpd logs (see [2]), start_pppd() function is executed,
in turn, from co
Begin forwarded message:
Date: Sun, 5 Jul 2015 14:35:59 +
From: "bugzilla-dae...@bugzilla.kernel.org"
To: "shemmin...@linux-foundation.org"
Subject: [Bug 100971] New: Crash in networking subsystem on removing PPP
interface
https://bugzilla.kernel.org/show_bug.cgi?id=100971
On Mon, 2015-07-06 at 17:11 +0200, Ursula Braun wrote:
> From: Ursula Braun
>
> The SMC-R protocol defines dynamic discovery of peers. This is done by
> implementing experimental TCP options as defined in RFC6994. The TCP code
> needs to be extended to support RFC6994.
>
> Setting the TCP experi
Le 06/07/2015 17:03, David Ahern a écrit :
In the context of internet scale routing a requirement that always
comes up is the need to partition the available routing tables into
disjoint routing planes. A specific use case is the multi-tenancy
problem where each tenant has their own unique routin
Le 06/07/2015 17:03, David Ahern a écrit :
This driver borrows heavily from IPvlan and teaming drivers.
Routing domains (VRF-lite) are created by instantiating a device
and enslaving all routed interfaces that participate in the domain.
As part of the enslavement, all local routes pointing to en
This reverts commit e1622baf54df8cc958bf29d71de5ad545ea7d93c.
The side effect of this commit is to add a '@NONE' after each virtual
interface name with a 'ip link'. It may break existing scripts.
Reported-by: Olivier Hartkopp
Signed-off-by: Nicolas Dichtel
---
net/core/dev.c | 4
1 file c
On Mon, Jul 06, 2015 at 05:34:01PM +0530, GMAIL wrote:
> On Friday 03 July 2015 11:46 PM, Jay Vosburgh wrote:
> >GMAIL wrote:
[...]
>
> Looks good, added cosmetic changes for more readability,
> it might save some instructions :)
>
>
> diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/
On 7/6/15, 6:03 AM, Thomas Graf wrote:
The integration into the routing code looks much better now. Any
chance you can rebase your tree and squash it into logical
commits?
Will do today.
I'm rebasing again onto yours and I think we should submit
all of this in a single series so everything ca
From: Eric Dumazet
User space can crash kernel with
ip link add ifb10 numtxqueues 10 type ifb
We must replace a BUG_ON() by proper test and return -EINVAL for
crazy values.
Fixes: 60877a32bce00 ("net: allow large number of tx queues")
Signed-off-by: Eric Dumazet
---
net/core/dev.c |3
From: Ursula Braun
The new socket family is assigned the next available address / protocol
family constant 41.
Signed-off-by: Ursula Braun
---
include/linux/socket.h | 4 +++-
include/net/smc.h | 13 +
net/Kconfig| 1 +
net/Makefile | 1 +
4 files chan
From: Ursula Braun
The SMC-R protocol defines dynamic discovery of peers. This is done by
implementing experimental TCP options as defined in RFC6994. The TCP code
needs to be extended to support RFC6994.
Setting the TCP experimental option for SMC-R [2] will be triggered from
kernel exploiters
From: Ursula Braun
In 2013, IBM introduced an optimized communications solution for the
IBM zEnterprise EC12 and BC12 (s390 in Linux terminology) that is
comprised of the IBM 10GbE RoCE Express feature with Shared Memory
Communications-RDMA (SMC-R) protocol [1].
SMC-R is designed for the enterpri
Add a VRF_MASTER flag for interfaces and helper functions for determining
if a device is a VRF_MASTER.
Also, add link attribute for passing VRF_TABLE id.
Both are used in the following patch that adds a VRF device driver.
Signed-off-by: Shrijeet Mukherjee
Signed-off-by: David Ahern
---
includ
Example of how to use the default bind to interface option for tasks and
correlate with VRF devices.
Signed-off-by: David Ahern
---
tools/net/Makefile | 6 +-
tools/net/chvrf.c | 225 +
2 files changed, 229 insertions(+), 2 deletions(-)
cre
This change is needed for the following VRF driver.
No active code path changes.
Signed-off-by: Shrijeet Mukherjee
Signed-off-by: David Ahern
---
net/ipv4/fib_frontend.c | 1 +
net/ipv4/fib_trie.c | 1 +
2 files changed, 2 insertions(+)
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_
Allow user to create a vrf device and specify its table binding.
Based on the iplink_vlan implementation.
Signed-off-by: Shrijeet Mukherjee
Signed-off-by: David Ahern
---
include/linux/if_link.h | 8 +
ip/Makefile | 2 +-
ip/iplink.c | 2 +-
ip/iplink_vrf.c
With the following tweaks to the IPv4 stack:
- enslaving devices to a VRF device automatically moves routes to the
VRF table; removing the VRF master moves routes back to the main table
- the following use cases work for both Rx and Tx:
+ ICMP (ping -I )
+ TCP server and client bound to VRF
Allow tasks to have a default device index for binding sockets. If set
the value is passed to all AF_INET/AF_INET6 sockets when they are created.
The task setting is passed parent to child on fork, but can be set or
changed after task creation using prctl (if task has CAP_NET_ADMIN
permissions). T
This driver borrows heavily from IPvlan and teaming drivers.
Routing domains (VRF-lite) are created by instantiating a device
and enslaving all routed interfaces that participate in the domain.
As part of the enslavement, all local routes pointing to enslaved
devices are re-pointed to the vrf devi
In the context of internet scale routing a requirement that always
comes up is the need to partition the available routing tables into
disjoint routing planes. A specific use case is the multi-tenancy
problem where each tenant has their own unique routing tables and in
the very least need different
On Mon, Jul 6, 2015 at 2:53 PM, Jamal Hadi Salim wrote:
> cant wait for the multi queue ifb.
>
Yeah, me too ;)
Do not try this on a production host :
ip link add ifb10 numtxqueues 100 type ifb
[284151.950695] kernel BUG at /build/buildd/linux-3.13.0/net/core/dev.c:5868!
[284151.950762] in
From: Carol Soto
Add extra check for total vfs for SRIOV to check if that value is
bigger than total vfs in pci SRIOV capabalities. Fix a check and
print of the number of maximum vfs that hw can handle. Fix a check
and print of the number of maximum vfs per port that driver can handle.
Signe
The trace bpf samples do not compile on s390x because they use x86
specific fields from the "pt_regs" structure.
Fix this and access the fields via new PT_REGS macros.
Signed-off-by: Michael Holzheu
---
samples/bpf/bpf_helpers.h | 25 +
samples/bpf/tracex1_kern.c | 2 +
On Mon, Jul 06, 2015 at 09:30:40PM +0800, Herbert Xu wrote:
> On Mon, Jul 06, 2015 at 02:01:42PM +0200, Phil Sutter wrote:
> > diff --git a/lib/rhashtable.c b/lib/rhashtable.c
> > index a60a6d3..e36b94b 100644
> > --- a/lib/rhashtable.c
> > +++ b/lib/rhashtable.c
> > @@ -585,6 +585,7 @@ void *rhash
On Mon, Jul 06, 2015 at 03:51:20PM +0200, Phil Sutter wrote:
> If rhashtable_walk_next detects a resize operation in progress, it jumps
> to the new table and continues walking that one. But it misses to drop
> the reference to it's current item, leading it to continue traversing
> the new table's
If rhashtable_walk_next detects a resize operation in progress, it jumps
to the new table and continues walking that one. But it misses to drop
the reference to it's current item, leading it to continue traversing
the new table's bucket in which the current item is sorted into, and
after reaching t
On Fri, Jul 03, 2015 at 09:41:41AM -0700, Joe Perches wrote:
> On Fri, 2015-07-03 at 07:51 -0400, Neil Horman wrote:
> > On Thu, Jul 02, 2015 at 02:54:56PM -0700, Joe Perches wrote:
> > > It's not clear to me that the sctp_fwdtsn_skip array is
> > > always initialized when used.
> > >
> > > It is
Hyper-V VM Sockets (hvsock) is a byte-stream based communication mechanism
between Windowsd 10 (or later) host and a guest. It's kind of TCP over
VMBus, but the transportation layer (VMBus) is much simpler than IP.
With Hyper-V VM Sockets, applications between the host and a guest can
talk with eac
On Mon, Jul 06, 2015 at 02:01:42PM +0200, Phil Sutter wrote:
> If rhashtable_walk_next detects a resize operation in progress, it jumps
> to the new table and continues walking that one. But it misses to drop
> the reference to it's current item, leading it to continue traversing
> the new table's
This will be used by the coming net/hvsock driver.
Signed-off-by: Dexuan Cui
---
include/linux/hyperv.h | 6 ++
1 file changed, 6 insertions(+)
diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h
index 307910b3..c52f4cb 100644
--- a/include/linux/hyperv.h
+++ b/include/linux/hyperv
A helper function is also added.
Signed-off-by: Dexuan Cui
---
include/linux/hyperv.h | 7 +++
1 file changed, 7 insertions(+)
diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h
index 30d3a1f..aa21814 100644
--- a/include/linux/hyperv.h
+++ b/include/linux/hyperv.h
@@ -236,6 +236,
Hyper-V VM Sockets (hvsock) is a byte-stream based communication mechanism
between Windowsd 10 (or later) host and a guest. It's kind of TCP over
VMBus, but the transportation layer (VMBus) is much simpler than IP.
With Hyper-V VM Sockets, applications between the host and a guest can
talk with eac
In the SMP guest case, when the per-channel callback hvsock_events() is
running on virtual CPU A, if the guest tries to close the connection on
virtual CPU B: we invoke vmbus_close() -> vmbus_close_internal(),
then we can have trouble: on B, vmbus_close_internal() will send IPI
reset_channel_cb() t
Hyper-V VM sockets (hvsock) supplies a byte-stream based communication
mechanism between the host and a guest. It's kind of TCP over VMBus, but
the transportation layer (VMBus) is much simpler than IP. With Hyper-V VM
Sockets, applications between the host and a guest can talk with each
other direc
With the 2 APIs supplied by the VMBus driver, the coming net/hvsock driver
can register 2 callbacks and can know when a new hvsock connection is
offered by the host, and when a hvsock connection is being closed by the
host.
Signed-off-by: Dexuan Cui
---
drivers/hv/Makefile | 4 ++-
dr
This will be used by the coming net/hvsock driver.
Signed-off-by: Dexuan Cui
---
drivers/hv/channel.c | 131 ++
drivers/hv/hyperv_vmbus.h | 4 ++
drivers/hv/ring_buffer.c | 14 +
include/linux/hyperv.h| 33
4 files change
A function to send the type of message is also added.
The coming net/hvsock driver will use this function to proactively request
the host to offer a VMBus channel for a new hvsock connection.
Signed-off-by: Dexuan Cui
---
drivers/hv/channel.c | 15 +++
drivers/hv/channel_mgmt.c
On Sun, Jul 05, 2015 at 03:00:11PM +0200, Willy Tarreau wrote:
> Hi Thomas,
>
> On Fri, Jul 03, 2015 at 04:46:24PM +0200, Thomas Petazzoni wrote:
> > Maxime,
> >
> > On Fri, 3 Jul 2015 16:25:51 +0200, Maxime Ripard wrote:
> >
> > > +static void mvneta_percpu_enable(void *arg)
> > > +{
> > > + s
Hi Willy,
On Sun, Jul 05, 2015 at 02:37:08PM +0200, Willy Tarreau wrote:
> Hi Maxime,
>
> On Fri, Jul 03, 2015 at 04:25:49PM +0200, Maxime Ripard wrote:
> > Now that our interrupt controller is allowing us to use per-CPU interrupts,
> > actually use it in the mvneta driver.
> >
> > This involves
On 07/04/15 at 11:21pm, roopa wrote:
> Thx. I have fixed this since,...did not realize it came in as part of this
> RFC series.
> >
> >Other than that I managed to rebase my changes onto yours and it
> >looks clean.
> Glad to know!. thanks Thomas. I had a few more changes (mostly cleanup/bug
> fixe
From: Satish Ashok
Start the delete timer when adding temp static entries so they can expire.
Signed-off-by: Satish Ashok
Signed-off-by: Nikolay Aleksandrov
Fixes: ccb1c31a7a87 ("bridge: add flags to distinguish permanent mdb entires")
---
net/bridge/br_mdb.c | 3 +++
1 file changed, 3 insert
On 07/06/15 08:18, Eric Dumazet wrote:
Like act_gact, act_mirred can be lockless in packet processing
1) Use percpu stats
2) update lastuse only every clock tick to avoid false sharing
3) use rcu to protect tcfm_dev
4) Remove spinlock usage, as it is no longer needed.
Next step : add multi queu
Second step for gact RCU operation :
We want to get rid of the spinlock protecting gact operations.
Stats (packets/bytes) will soon be per cpu.
gact_determ() would not work without a central packet counter,
so lets add it for this mode.
Signed-off-by: Eric Dumazet
Cc: Alexei Starovoitov
Acked-
Third step for gact RCU operation :
Following patch will get rid of spinlock protection,
so we need to read tcfg_ptype once.
Signed-off-by: Eric Dumazet
Cc: Alexei Starovoitov
Acked-by: Jamal Hadi Salim
Acked-by: John Fastabend
---
net/sched/act_gact.c | 14 +++---
1 file changed, 7
qdisc_bstats_update_cpu() and other helpers were added to support
percpu stats for qdisc.
We want to add percpu stats for tc action, so this patch add common
helpers.
qdisc_bstats_update_cpu() is renamed to qdisc_bstats_cpu_update()
qdisc_qstats_drop_cpu() is renamed to qdisc_qstats_cpu_drop()
S
Like act_gact, act_mirred can be lockless in packet processing
1) Use percpu stats
2) update lastuse only every clock tick to avoid false sharing
3) use rcu to protect tcfm_dev
4) Remove spinlock usage, as it is no longer needed.
Next step : add multi queue capability to ifb device
Signed-off-by
As mentioned by Alexei last week in Budapest, it is a bit weird
to take a spinlock in order to drop a packet in a tc filter...
Lets add percpu infra for tc actions and use it for gact & mirred.
Before changes, my host with 8 RX queues was handling 5 Mpps with gact,
and more than 11 Mpps after.
M
Final step for gact RCU operation :
1) Use percpu stats
2) update lastuse only every clock tick to avoid false sharing
3) Remove spinlock acquisition, as it is no longer needed.
Since this is the last contended lock in packet RX when tc gact is used,
this gives impressive gain.
My host with 8 RX
Reuse existing percpu infrastructure John Fastabend added for qdisc.
This patch adds a new cpustats parameter to tcf_hash_create() and all
actions pass false, meaning this patch should have no effect yet.
Signed-off-by: Eric Dumazet
Cc: Alexei Starovoitov
Acked-by: Jamal Hadi Salim
Acked-by: J
First step for gact RCU operation :
Instead of testing if tcfg_pval is zero or not, just make it 1.
No change in behavior, but slightly faster code.
The smp_rmb()/smp_wmb() barriers, while not strictly needed at this
stage are added for upcoming spinlock removal.
Signed-off-by: Eric Dumazet
Ac
On Mon, Jul 6, 2015 at 2:12 PM, Jamal Hadi Salim wrote:
> on mirred:
> The main thing that is read-write is the stats. lastuse timestamp
> as well, but that one doesnt have to be accurate (used mainly
> to help decide when to flush out policies). so the lock around
> all that code is not necessar
On 07/06/15 02:41, Eric Dumazet wrote:
As mentioned by Alexei last week in Budapest, it is a bit weird
to take a spinlock in order to drop a packet in a tc filter...
Lets add percpu infra for tc actions and use it for gact & mirred.
Before changes, my host with 8 RX queues was handling 5 Mpps w
On Friday 03 July 2015 11:46 PM, Jay Vosburgh wrote:
GMAIL wrote:
Hi Jay,
On Friday 03 July 2015 02:12 AM, Jay Vosburgh wrote:
[ added netdev to cc ]
Mazhar Rana wrote:
When "primary_reselect" is set to "failure", primary interface should
not become active until current active s
If rhashtable_walk_next detects a resize operation in progress, it jumps
to the new table and continues walking that one. But it misses to drop
the reference to it's current item, leading it to continue traversing
the new table's bucket in which the current item is sorted into, and
after reaching t
Just to be clear - this happens on the real machine as well, but here the trace
is difficult to extract, because even with the help of someone seeing the
screen, I noticed the screen doesn't get updated. I am using vesa right now.
--
To unsubscribe from this list: send the line "unsubscribe netd
Signed-off-by: Hariprasad Shenai
---
drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h | 18 ++
1 file changed, 18 insertions(+)
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h
b/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h
index d7ca106..8353a6c 100644
--
Sure Oliver!
Here it is.
And - I tried with various approach. I tired also kzallocating the needed
memory inside the tx_fixup function using the GFP_ATOMIC flag due to the fact I
am in an interrupt handler.
At some point, the problem started manifesting in a memset call that whasn't in
my patc
we keep commands only on net_device_ops structure - struct net_device_ops.
From 0749e12c21b564d62c89ec35c5fc8f4289a7e4d9 Mon Sep 17 00:00:00 2001
From: Corcodel Marian
Date: Mon, 6 Jul 2015 13:24:59 +0300
Subject: [PATCH] Committer: Corcodel Marian
Disable exec commands outside ops structure
On 06/07/15 12:19, Florian Westphal wrote:
> Julien Grall wrote:
>> On 03/07/15 21:42, Florian Westphal wrote:
>>> I think it makes more sense to use __in6_dev_get() instead which doesn't
>>> take a reference.
>>
>> __in6_dev_get requires to hold rcu_read_lock or RTNL. My knowledge on
>> this code
1 - 100 of 113 matches
Mail list logo
