On Thu, Apr 30, 2015 at 06:05:37AM +0200, Patrick McHardy wrote:
> 06:05:37AM
do you ever sleep? ;)
> For the sake of completeness - on ingress, this alternativ means simply
> ditching TC.
I'm sure 'ditching TC' doesn't mean to kill ingress qdisc
completely. Compatibility needs to be preserved
From: Eric Dumazet
Date: Wed, 29 Apr 2015 16:20:58 -0700
> From: Eric Dumazet
>
> I forgot to update tcp_westwood when changing get_info() behavior,
> this patch should fix this.
>
> Fixes: 64f40ff5bbdb ("tcp: prepare CC get_info() access from getsockopt()")
> Reported-by: kbuild test robot
>
Hi all,
This patch series is the long promised API rework to allow registering DSA
switch as PCI, platform, SPI, or PHY drivers, and this is pretty much the
grand plan described here:
http://permalink.gmane.org/gmane.linux.network/329848
Note that unless appropriate Device Tree or platform_data c
On 30.04, Daniel Borkmann wrote:
> On 04/30/2015 02:37 AM, Patrick McHardy wrote:
> >On 30.04, Pablo Neira Ayuso wrote:
> >>On the bugfix front, the illegal mangling of shared skb from actions
> >>like stateless nat and bpf look also important to be addressed to me.
> >>David already suggested to p
On 30.04, Daniel Borkmann wrote:
> >
> >I can also see there were also intentions to support userspace
> >queueing at some point since TC_ACT_QUEUED has been there since the
> >beginning. That should be possible at some point using this
> >infrastructure (once there are no further concerns on the
On 04/30/2015 01:32 AM, Pablo Neira Ayuso wrote:
...
Actually, the extra cost is roughly (getting inlined stuff away and
other non-relevant stuff):
`-> nf_hook_slow()
`-> [for each entry in hook list]
`-> nf_iterate()
`-> (*elemp)->hook()
Yep, agreed.
as part of the
On 29.04, Cong Wang wrote:
> On Wed, Apr 29, 2015 at 11:53 AM, Pablo Neira Ayuso
> wrote:
> > diff --git a/net/sched/Kconfig b/net/sched/Kconfig
> > index 2274e72..23b57da 100644
> > --- a/net/sched/Kconfig
> > +++ b/net/sched/Kconfig
> > @@ -312,6 +312,7 @@ config NET_SCH_PIE
> > config NET_SCH
On Wed, Apr 29, 2015 at 7:20 PM, Eric Dumazet wrote:
> From: Eric Dumazet
>
> I forgot to update tcp_westwood when changing get_info() behavior,
> this patch should fix this.
>
> Fixes: 64f40ff5bbdb ("tcp: prepare CC get_info() access from getsockopt()")
> Reported-by: kbuild test robot
> Signed
On Wed, Apr 29, 2015 at 10:27:05PM +0200, Daniel Borkmann wrote:
> On 04/29/2015 08:53 PM, Pablo Neira Ayuso wrote:
> >Port qdisc ingress on top of the Netfilter ingress allows us to detach the
> >qdisc ingress filtering code from the core, so now it resides where it really
> >belongs.
>
> Hm, but
On Wed, Apr 29, 2015 at 4:07 PM, Anuradha Karuppiah
wrote:
> On Wed, Apr 29, 2015 at 3:13 PM, Stephen Hemminger
> wrote:
>> On Mon, 27 Apr 2015 10:38:21 -0700
>> anurad...@cumulusnetworks.com wrote:
>>
>>> From: Anuradha Karuppiah
>>>
>>> This patch introduces an IFF_PROTO_DOWN flag that can be
On 4/28/15 10:23 PM, Eric Dumazet wrote:
On Tue, 2015-04-28 at 19:11 -0700, Alexei Starovoitov wrote:
Hi,
there were many requests for performance numbers in the past, but not
everyone has access to 10/40G nics and we need a common way to talk
about RX path performance without overhead of drive
From: Yuchung Cheng
Date: Wed, 29 Apr 2015 11:28:30 -0700
> tcp_mark_lost_retrans is not used when FACK is disabled. Since
> tcp_update_reordering may disable FACK, it should be called first
> before tcp_mark_lost_retrans.
>
> Signed-off-by: Yuchung Cheng
> Signed-off-by: Nandita Dukkipati
> S
From: Eric Dumazet
Date: Tue, 28 Apr 2015 16:23:49 -0700
> Some Congestion Control modules can provide per flow information,
> but current way to get this information is to use netlink.
>
> Like TCP_INFO, let's add TCP_CC_INFO so that applications can
> issue a getsockopt() if they have a socket
From: Eric Dumazet
Date: Tue, 28 Apr 2015 16:23:48 -0700
> We would like that optional info provided by Congestion Control
> modules using netlink can also be read using getsockopt()
>
> This patch changes get_info() to put this information in a buffer,
> instead of skb, like tcp_get_info(), so
From: Eric Dumazet
Date: Tue, 28 Apr 2015 15:28:18 -0700
> This patch tracks total number of payload bytes received on a TCP socket.
> This is the sum of all changes done to tp->rcv_nxt
>
> RFC4898 named this : tcpEStatsAppHCThruOctetsReceived
>
> This is a 64bit field, and can be fetched both
From: Eric Dumazet
Date: Tue, 28 Apr 2015 15:28:17 -0700
> This patch tracks total number of bytes acked for a TCP socket.
> This is the sum of all changes done to tp->snd_una, and allows
> for precise tracking of delivered data.
>
> RFC4898 named this : tcpEStatsAppHCThruOctetsAcked
>
> This i
On Wed, Apr 29, 2015 at 11:53 AM, Pablo Neira Ayuso wrote:
> diff --git a/net/sched/Kconfig b/net/sched/Kconfig
> index 2274e72..23b57da 100644
> --- a/net/sched/Kconfig
> +++ b/net/sched/Kconfig
> @@ -312,6 +312,7 @@ config NET_SCH_PIE
> config NET_SCH_INGRESS
> tristate "Ingress Qdisc"
Hi,
On Wed, Apr 29, 2015 at 11:28:46AM +0300, Julian Anastasov wrote:
>
> Hello,
>
> On Tue, 28 Apr 2015, Martin KaFai Lau wrote:
>
> > -static inline struct in6_addr *rt6_nexthop(struct rt6_info *rt)
> > +static inline struct in6_addr *rt6_nexthop(struct rt6_info *rt,
> > +
The net-next tree is now open for submissions.
Fire away...
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
From: Vishwanath Pai
Date: Wed, 29 Apr 2015 16:30:44 -0400
> Since the patch has been applied already - should I send a V2 with the
> suggested changes? Or submit a new patch that applies on top of the
> first patch?
You should always send me a relative patch in this situation.
--
To unsubscribe
From: Hariprasad Shenai
Date: Wed, 29 Apr 2015 17:19:05 +0530
> Commit 6559a7e8296002b4 ("cxgb4: Cleanup macros so they follow the same
> style and look consistent") introduced a regression where reading MC1
> memory in adapters where MC0 isn't present or MC0 size is not equal to MC1
> size cause
From: Yuval Mintz
Date: Wed, 29 Apr 2015 08:09:49 +0300
> In a kdump environment interfaces might be re-loaded without a proper
> unload sequence in the previous running kernel.
> bnx2x management FW and driver maintains a `pulse' that notifies the FW
> that the driver is still up and running.
>
On 04/29/2015 08:24 PM, Pai wrote:
> This patch fixes a Kernel Panic in bonding driver debugfs file:
> rlb_hash_table.
>
> $> modprobe bonding mode=6
> $> cat /sys/kernel/debug/bonding/bond0/rlb_hash_table
>
> This will crash the kernel. The struct alb_bond_info is initialized only when
> the bo
On Wed, 2015-04-29 at 13:34 -0400, Toan Pham wrote:
> Prashant,
>
> Unfortunately, I ran the same test 3 times with the new patch and all
> of them failed.
> Attached file is the dmesg log, after the Watchdog had timed out, and
> tried to restart the NIC.
> Feel free to let me know if you would li
From: Andy Gospodarek
Date: Wed, 29 Apr 2015 14:51:07 -0400
> On Wed, Apr 29, 2015 at 02:24:23PM -0400, Pai wrote:
>> This patch fixes a Kernel Panic in bonding driver debugfs file:
>> rlb_hash_table.
>>
>> $> modprobe bonding mode=6
>> $> cat /sys/kernel/debug/bonding/bond0/rlb_hash_table
>>
From: Guenter Roeck
Date: Wed, 29 Apr 2015 10:56:15 -0700
> eeprom-length is a switch property, not a dsa property, and thus
> needs to be attached to the switch node, not to the dsa node.
>
> Reported-by: Andrew Lunn
> Fixes: 6793abb4e849 ("net: dsa: Add support for switch EEPROM access")
> Si
From: "K. Y. Srinivasan"
Date: Tue, 28 Apr 2015 17:59:48 -0700
> Commit b08cc79155fc26d0d112b1470d1ece5034651a4b eliminated memory
> allocation in the packet send path:
>
> "hv_netvsc: Eliminate memory allocation in the packet send path
>
> The network protocol used to communicate with
From: Punnaiah Choudary Kalluri
Date: Wed, 29 Apr 2015 08:34:46 +0530
> From: Punnaiah Choudary Kalluri
>
> Under heavy Rx load, observed that the Hw is updating the USED bit
> and it is not updating the received frame status to the BD control
> field. This could be lack of resources for proces
From: Vlastimil Setka
Date: Wed, 29 Apr 2015 00:17:11 +0200
> From: Vlastimil Setka
>
> Altera TSE MAC rx DMA transfer starts with the 2 additional bytes for IP
> payload alignment. This patch fixes tse_rx() function loop which reads DMA
> rx status and extracts packet length from it. Status si
From: Benjamin Poirier
Date: Tue, 28 Apr 2015 14:49:29 -0700
> By default, the number of tx queues is limited by the number of online cpus
> in mlx4_en_get_profile(). However, this limit no longer holds after the
> ethtool .set_channels method has been called. In that situation, the driver
> may
From: Jon Maloy
Date: Tue, 28 Apr 2015 16:59:04 -0400
> Currently, we try to accumulate arrived packets in the links's
> 'deferred' queue during the parallel link syncronization phase.
>
> This entails two problems:
>
> - With an unlucky combination of arriving packets the algorithm
> may go
From: Ivan Vecera
Date: Tue, 28 Apr 2015 21:39:32 +0200
> The driver unlike other drivers does not log link state changes. It's
> better for an user when asynchronous link states are logged in the
> system log.
>
> v3: Changes from v2 discarded as "not necessary"
>
> Cc: Sathya Perla
> Cc: Sub
From: Nicolas Dichtel
Date: Tue, 28 Apr 2015 18:33:47 +0200
> The first three patches fix a wrong use of this flag.
>
> Because it's not the first time that this kind of bug happens, it could be
> good to find something to help people.
>
> The only idea that I have is in the last patch. It's ju
From: Florian Westphal
Date: Tue, 28 Apr 2015 13:33:21 +0200
> This action is meant to be passive, i.e. we should not alter
> skb->nfct: If nfct is present just leave it alone.
>
> Compile tested only.
>
> Cc: Jamal Hadi Salim
> Signed-off-by: Florian Westphal
Applied.
--
To unsubscribe from
From: Antonio Ospite
Date: Tue, 28 Apr 2015 13:11:29 +0200
> Fix the 0x0x prefix in an integer constant.
>
> In this case, while at it, also fix a typo (s/unitcast/unicast/).
>
> Signed-off-by: Antonio Ospite
Applied.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the
From: Antonio Ospite
Date: Tue, 28 Apr 2015 13:11:27 +0200
> Fix the 0x0x prefix in an integer constant.
>
> Signed-off-by: Antonio Ospite
Applied.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majord...@vger.kernel.org
More majordomo info at
Port qdisc ingress on top of the Netfilter ingress allows us to detach the
qdisc ingress filtering code from the core, so now it resides where it really
belongs.
The specific qdisc ingress static key is also gone since we now rely on the
generic netfilter hook static key infrastructure.
This only
From: Yuval Mintz
Date: Tue, 28 Apr 2015 10:45:17 +
>> I noticed some simplification possibilities while looking into the bug fixed
>> by
>> "bnx2x: really disable TPA if 'disable_tpa' is set'.
>>
>> Michal Schmidt (3):
>> bnx2x: mark LRO as a fixed disabled feature if disable_tpa is set
From: Michal Schmidt
Date: Tue, 28 Apr 2015 11:34:20 +0200
> I noticed some simplification possibilities while looking into the bug
> fixed by "bnx2x: really disable TPA if 'disable_tpa' is set'.
>
> Michal Schmidt (3):
> bnx2x: mark LRO as a fixed disabled feature if disable_tpa is set
> bn
From: six...@microsoft.com
Date: Tue, 28 Apr 2015 01:05:17 -0700
> From: Simon Xiao
>
> 1. Introduce netif-msg to netvsc to control debug logging output
> and keep msg_enable in netvsc_device_context so that it is
> kept persistently.
> 2. Only call dump_rndis_message() when NETIF_MSG_RX_ERR or
From: Herbert Xu
Date: Tue, 28 Apr 2015 11:43:15 +0800
> Subject: route: Use ipv4_mtu instead of raw rt_pmtu
>
> The commit 3cdaa5be9e81a914e633a6be7b7d2ef75b528562 ("ipv4: Don't
> increase PMTU with Datagram Too Big message") broke PMTU in cases
> where the rt_pmtu value has expired but is smal
On Wed, Apr 29, 2015 at 10:56:15AM -0700, Guenter Roeck wrote:
> eeprom-length is a switch property, not a dsa property, and thus
> needs to be attached to the switch node, not to the dsa node.
>
> Reported-by: Andrew Lunn
> Fixes: 6793abb4e849 ("net: dsa: Add support for switch EEPROM access")
>
eeprom-length is a switch property, not a dsa property, and thus
needs to be attached to the switch node, not to the dsa node.
Reported-by: Andrew Lunn
Fixes: 6793abb4e849 ("net: dsa: Add support for switch EEPROM access")
Signed-off-by: Guenter Roeck
---
net/dsa/dsa.c | 2 +-
1 file changed, 1
Stephen Hemminger wrote:
> On Sat, 25 Apr 2015 22:54:11 +0200
> Felix Janda wrote:
>
> > Hello.
> >
> > Background:
> > Current iproute2 does not build when the libc is musl instead of glibc.
> > This is because of redefinition of in6_addr in netinet/in.h and
> > linux/in6.h. There are workaroun
On Wed, 2015-04-29 at 17:17 +0200, D.S. Ljungmark wrote:
> On 29/04/15 16:51, Denys Vlasenko wrote:
> > On Wed, Apr 1, 2015 at 9:48 PM, David Miller wrote:
> >> D.S. Ljungmark (1):
> >> ipv6: Don't reduce hop limit for an interface
> >
> > https://git.kernel.org/cgit/linux/kernel/git/torval
From: One Thousand Gnomes
Date: Wed, 29 Apr 2015 11:15:21 +0100
>> Unfortunately, the Cadence MACB doesn't support the enabling or
>> disabling of checksum generation per descriptor.
>
> So how does packet forwarding work ?
Yeah actually rechecksumming in that scenerio is illegal.
--
To unsubs
From: Martin Zhang
Date: Wed, 29 Apr 2015 17:09:24 +0800
> 'if' and 'elseif' do the same operation,
> so merge then into a case.
>
> Signed-off-by: Martin Zhang
> ---
> net/ipv4/route.c | 7 ++-
> 1 file changed, 2 insertions(+), 5 deletions(-)
>
> diff --git a/net/ipv4/route.c b/net/ipv4
This introduces the Freescale Data Path Acceleration Architecture
(DPAA) Ethernet driver (dpaa_eth) that builds upon the DPAA QMan,
BMan, PAMU and FMan drivers to deliver Ethernet connectivity on
the Freescale DPAA QorIQ platforms.
Signed-off-by: Madalin Bucur
---
drivers/net/ethernet/freescale/
Export Frame Queue and Buffer Pool IDs through sysfs.
Signed-off-by: Madalin Bucur
---
drivers/net/ethernet/freescale/dpaa/Makefile | 2 +-
drivers/net/ethernet/freescale/dpaa/dpaa_eth.c | 2 +
drivers/net/ethernet/freescale/dpaa/dpaa_eth.h | 3 +
.../net/ethernet/freescale/d
On Wed, Apr 1, 2015 at 9:48 PM, David Miller wrote:
> D.S. Ljungmark (1):
> ipv6: Don't reduce hop limit for an interface
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=6fd99094de2b83d1d4c8457f2c83483b2828e75a
I was testing this change and apparently it doesn't
On 04/29/2015 01:07 PM, Eric Dumazet wrote:
On Wed, 2015-04-29 at 10:17 +0200, Daniel Borkmann wrote:
On 04/29/2015 01:23 AM, Eric Dumazet wrote:
Some Congestion Control modules can provide per flow information,
but current way to get this information is to use netlink.
Like TCP_INFO, let's ad
> -Original Message-
> From: Ivan Vecera [mailto:ivec...@redhat.com]
>
> The driver unlike other drivers does not log link state changes. It's
> better for an user when asynchronous link states are logged in the
> system log.
>
> v3: Changes from v2 discarded as "not necessary"
>
> Cc: S
Hello,
On Tue, 28 Apr 2015, Martin KaFai Lau wrote:
> -static inline struct in6_addr *rt6_nexthop(struct rt6_info *rt)
> +static inline struct in6_addr *rt6_nexthop(struct rt6_info *rt,
> +struct in6_addr *daddr)
> {
> - return &rt->rt6i_gatew
On 04/29/2015 01:23 AM, Eric Dumazet wrote:
Some Congestion Control modules can provide per flow information,
but current way to get this information is to use netlink.
Like TCP_INFO, let's add TCP_CC_INFO so that applications can
issue a getsockopt() if they have a socket file descriptor,
inste
Tom Herbert wrote:
> This is used to get the skb->hash and then perturb it for a local use.
>
> Signed-off-by: Tom Herbert
> ---
> include/linux/skbuff.h | 15 +++
> 1 file changed, 15 insertions(+)
>
> diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
> index 66e374d..b
On 04/29/2015 01:23 AM, Eric Dumazet wrote:
We would like that optional info provided by Congestion Control
modules using netlink can also be read using getsockopt()
This patch changes get_info() to put this information in a buffer,
instead of skb, like tcp_get_info(), so that following patch
ca
56 matches
Mail list logo
