Signed-off-by: Hariprasad Shenai
---
drivers/net/ethernet/chelsio/cxgb4/cxgb4.h | 10 +++
drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 77 ++-
drivers/net/ethernet/chelsio/cxgb4/t4_hw.c |2 +-
3 files changed, 86 insertions(+), 3 deletions(-)
diff --git
Hi,
This patch series adds support for ndo_set_vf_vlan and ndo_get_vf_config
for cxgb4 driver.
This patch series has been created against net-next tree.
We have included all the maintainers of respective drivers. Kindly review
the change and let us know in case of any review comments.
Thanks
H
Signed-off-by: Hariprasad Shenai
---
drivers/net/ethernet/chelsio/cxgb4/cxgb4.h |1 +
drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c| 21 +++
drivers/net/ethernet/chelsio/cxgb4/t4_hw.c | 25 +
drivers/net/ethernet/chelsio/cxgb4vf/adapter.h
This allows a privileged process to filter by socket mark when
dumping sockets via INET_DIAG_BY_FAMILY. This is useful on
systems that use mark-based routing such as Android.
The ability to filter socket marks requires CAP_NET_ADMIN, which
is consistent with other privileged operations allowed by
This simplifies the code a bit and also allows inet_diag_bc_audit
to send to userspace an error that isn't EINVAL.
Signed-off-by: Lorenzo Colitti
---
net/ipv4/inet_diag.c | 28 +---
1 file changed, 17 insertions(+), 11 deletions(-)
diff --git a/net/ipv4/inet_diag.c b/net
From: Eric Dumazet
Date: Tue, 23 Aug 2016 11:39:25 -0700
> Since IPv6 socket lookups no longer dereference pinet6 pointer
> and UDP lost SLAB_DESTROY_BY_RCU special rules, we no longer
> need special clear_sk() methods.
Series applied, thanks Eric.
There were some merge conflicts because I appl
From: Wei Yongjun
Date: Tue, 23 Aug 2016 23:01:02 +
> From: Wei Yongjun
>
> Use kfree_skb() instead of kfree() to free sk_buff.
>
> Fixes: 0d051bf93c06 ("tipc: make bearer packet filtering generic")
> Signed-off-by: Wei Yongjun
Applied, thanks.
From: David Ahern
Date: Tue, 23 Aug 2016 21:05:27 -0700
> inet_diag_find_one_icsk takes a reference to a socket that is not
> released if sock_diag_destroy returns an error. Fix by changing
> tcp_diag_destroy to manage the refcnt for all cases and remove
> the sock_put calls from tcp_abort.
>
>
From: David Ahern
Date: Tue, 23 Aug 2016 21:06:33 -0700
> This implements SOCK_DESTROY for UDP sockets similar to what was done
> for TCP with commit c1e64e298b8ca ("net: diag: Support destroying TCP
> sockets.") A process with a UDP socket targeted for destroy is awakened
> and recvmsg fails wit
From: Soheil Hassas Yeganeh
Date: Tue, 23 Aug 2016 18:22:33 -0400
> From: Soheil Hassas Yeganeh
>
> Instead of using sock_tx_timestamp, use skb_tx_timestamp to record
> software transmit timestamp of a packet.
>
> sock_tx_timestamp resets and overrides the tx_flags of the skb.
> The function i
From: "Xue, Ying"
Date: Wed, 24 Aug 2016 02:47:32 +
> Acked-by: Ying Xue
Please do not ACK patches like this.
If you quote the patch with no demarcation in the initial column for
the quoted text, like so:
> From: Wei Yongjun [mailto:weiyj...@gmail.com]
> Sent: Wednesday, August 24, 2016
>> > Ah, I see what you're doing. Ok, this makes some sense, at least on the
>> > receive
>> > side, when you get a cookie unpacked and modify the remote peers address
>> > list,
>> > it makes sense to check for duplicates. On the local side however, I
>> > would,
>> > instead of checking it w
> This patch series provides following support
> a) Reorganized fields based out of SFF-8024 fields i.e. Identifier/
>Encoding/Connector types which are common across SFP/SFP+ (SFF-8472)
>and QSFP+/QSFP28 (SFF-8436/SFF-8636) modules into sff-common files.
> b) Support for diagnostics inform
This implements SOCK_DESTROY for UDP sockets similar to what was done
for TCP with commit c1e64e298b8ca ("net: diag: Support destroying TCP
sockets.") A process with a UDP socket targeted for destroy is awakened
and recvmsg fails with ECONNABORTED.
Signed-off-by: David Ahern
---
v4
- added suppor
inet_diag_find_one_icsk takes a reference to a socket that is not
released if sock_diag_destroy returns an error. Fix by changing
tcp_diag_destroy to manage the refcnt for all cases and remove
the sock_put calls from tcp_abort.
Fixes: c1e64e298b8ca ("net: diag: Support destroying TCP sockets")
Rep
Acked-by: Ying Xue
-Original Message-
From: Wei Yongjun [mailto:weiyj...@gmail.com]
Sent: Wednesday, August 24, 2016 7:01 AM
To: Jon Maloy; Xue, Ying; David S. Miller
Cc: Wei Yongjun; netdev@vger.kernel.org; tipc-discuss...@lists.sourceforge.net
Subject: [PATCH -next] tipc: use kfree_skb
On 8/23/16, 7:52 AM, Jiri Pirko wrote:
> Tue, Aug 23, 2016 at 04:46:37PM CEST, ro...@cumulusnetworks.com wrote:
>> On 8/23/16, 12:26 AM, Jiri Pirko wrote:
>>> Tue, Aug 23, 2016 at 09:04:15AM CEST, da...@davemloft.net wrote:
From: Jiri Pirko
Date: Tue, 23 Aug 2016 08:53:18 +0200
From: Eric Dumazet
Date: Tue, 23 Aug 2016 09:57:51 -0700
> From: Eric Dumazet
>
> After commit ca065d0cf80f ("udp: no longer use SLAB_DESTROY_BY_RCU")
> we do not need this special allocation mode anymore, even if it is
> harmless.
>
> Signed-off-by: Eric Dumazet
Applied.
From: Rami Rosen
Date: Tue, 23 Aug 2016 20:20:17 +0300
> This patch changes the return type of ena_set_push_mode() to be void,
> as it always returns 0.
>
> Signed-off-by: Rami Rosen
Applied, thanks.
xrpc-rewrite-20160823-2
Also pulled, thanks David.
From: Lance Richardson
Date: Tue, 23 Aug 2016 11:40:52 -0400
> The function sctp_diag_dump_one() currently performs a memcpy()
> of 64 bytes from a 16 byte field into another 16 byte field. Fix
> by using correct size, use sizeof to obtain correct size instead
> of using a hard-coded constant.
>
From: David Howells
Date: Tue, 23 Aug 2016 16:26:18 +0100
> Here are some cleanups for the AF_RXRPC rewrite:
...
> git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs.git
> rxrpc-rewrite-20160823-1
Pulled.
From: Lars Persson
Date: Tue, 23 Aug 2016 16:31:28 +0200
> From: Rabin Vincent
>
> We currently enable interrupts before we enable NAPI. If an RX interrupt
> hits before we enabled NAPI then the NAPI callback is never called and
> we leave the hardware with RX interrupts disabled, which of cour
From: Wei Yongjun
Date: Tue, 23 Aug 2016 15:09:49 +
> From: Wei Yongjun
>
> Remove including that don't need it.
>
> Signed-off-by: Wei Yongjun
Applied.
From: Wei Yongjun
Date: Tue, 23 Aug 2016 15:06:05 +
> From: Wei Yongjun
>
> Fixes the following sparse warning:
>
> drivers/net/phy/xilinx_gmii2rgmii.c:61:5: warning:
> symbol 'xgmiitorgmii_probe' was not declared. Should it be static?
>
> Signed-off-by: Wei Yongjun
Applied.
From: Wei Yongjun
Date: Tue, 23 Aug 2016 15:11:03 +
> From: Wei Yongjun
>
> There is a error message within devm_ioremap_resource
> already, so remove the dev_err call to avoid redundant
> error message.
>
> Signed-off-by: Wei Yongjun
Applied.
From: Sudarsana Reddy Kalluru
Date: Tue, 23 Aug 2016 10:56:55 -0400
> Add provision for configuring the fastpath queues with Tx (or Rx) only
> functionality.
>
> Signed-off-by: Sudarsana Reddy Kalluru
> Signed-off-by: Yuval Mintz
Applied.
From: Fabio Estevam
Date: Tue, 23 Aug 2016 09:48:20 -0300
> clk_prepare_enable() may fail, so we should better check its return
> value and propagate it in the case of failure
>
> While at it, replace __lpc_eth_clock_enable() with a plain
> clk_prepare_enable/clk_disable_unprepare() call in orde
From: Hariprasad Shenai
Date: Tue, 23 Aug 2016 11:35:32 +0530
> When we disable SRIOV, we used to unregister the netdev but wasn't
> freed. But next time when the same netdev is registered, since the state
> was in 'NETREG_UNREGISTERED', we used to hit BUG_ON in register_netdevice,
> where it exp
From: Phil Sutter
Date: Tue, 23 Aug 2016 13:14:31 +0200
> Since the features bit field has bits for internal only use as well, it
> may happen that the kernel exports RTAX_FEATURES attribute with zero
> value which is pointless.
>
> Fix this by making sure the attribute is added only if the expo
On 16-08-23 04:25 PM, Eric Dumazet wrote:
> On Tue, 2016-08-23 at 13:24 -0700, John Fastabend wrote:
>> The per cpu qstats support was added with per cpu bstat support which
>> is currently used by the ingress qdisc. This patch adds a set of
>> helpers needed to make other qdiscs that use qstats pe
From: Yuchung Cheng
Date: Mon, 22 Aug 2016 17:17:54 -0700
> TFO_SERVER_WO_SOCKOPT2 was intended for debugging purposes during
> Fast Open development. Remove this config option and also
> update/clean-up the documentation of the Fast Open sysctl.
>
> Reported-by: Piotr Jurkiewicz
> Signed-off-b
From: Wei Yongjun
Date: Mon, 22 Aug 2016 23:01:29 +
> From: Wei Yongjun
>
> The callback function of call_rcu() just calls a kfree(), so we
> can use kfree_rcu() instead of call_rcu() + callback function.
>
> Signed-off-by: Wei Yongjun
Applied.
From: Jamie Lentin
Date: Mon, 22 Aug 2016 22:47:08 +0100
> The PORT_RATE_CONTROL register works differently on 88e6095/6095f/6131
> in comparison to 6123/61/65, and 0x0 disables. The distinction was lost
> Linux 4.1 --> 4.2
>
> Signed-off-by: Jamie Lentin
Applied.
From: Xander Huff
Date: Mon, 22 Aug 2016 15:57:16 -0500
> Like the ksz8081, the ksz9031 has the behavior where it will clear the
> interrupt enable bits when leaving power down. This takes advantage of the
> solution provided by f5aba91.
>
> Signed-off-by: Xander Huff
> Signed-off-by: Nathan Su
From: Eric Dumazet
Date: Mon, 22 Aug 2016 11:31:10 -0700
> From: Eric Dumazet
>
> When sending an ack in SYN_RECV state, we must scale the offered
> window if wscale option was negotiated and accepted.
>
> Tested:
> Following packetdrill test demonstrates the issue :
...
> Signed-off-by: Eri
From: Nicholas Mc Guire
Date: Mon, 22 Aug 2016 17:52:00 +0200
> liquidio_set_rxcsum_command is a local function only, no need to expose
> it outside of lio_main.c so declare it static and make sparse happy.
>
> Signed-off-by: Nicholas Mc Guire
Applied.
From: f...@ikuai8.com
Date: Mon, 22 Aug 2016 22:50:02 +0800
> From: Gao Feng
>
> Use PPP_ALLSTATIONS, PPP_UI, and SEND_SHUTDOWN instead of 0xff,
> 0x03, and 2 separately.
>
> Signed-off-by: Gao Feng
Applied.
From: Zefir Kurtisi
Date: Mon, 22 Aug 2016 15:58:12 +0200
> The current scatter-gather logic in gianfar is flawed, since
> it does not consider the eTSEC's RxBD 'Data Length' field is
> context depening: for the last fragment it contains the full
> frame size, while fragments contain the fragment
From: Mikko Rapeli
Date: Wed, 24 Aug 2016 00:21:45 +0300
> I could propose patches but they are bound to be incomplete so
> networking devs could do this faster and with fewer review
> iterations.
If I did it I'm seriously not going to go any further than taking the
diffstat of your patches and
From: Zefir Kurtisi
Date: Mon, 22 Aug 2016 15:56:38 +0200
> The eTSEC register MRBLR defines the maximum space in
> the RX buffers and is set to 1536 by gianfar. This
> reasonably covers the common use case where the MTU
> is kept at default 1500. In that case, the largest
> Ethernet frame size o
From: Eric Dumazet
Date: Tue, 23 Aug 2016 13:59:33 -0700
> From: Eric Dumazet
>
> Laura tracked poll() [and friends] regression caused by commit
> e6afc8ace6dd ("udp: remove headers from UDP packets before queueing")
>
> udp_poll() needs to know if there is a valid packet in receive queue,
> e
On Tue, 2016-08-23 at 13:24 -0700, John Fastabend wrote:
> The per cpu qstats support was added with per cpu bstat support which
> is currently used by the ingress qdisc. This patch adds a set of
> helpers needed to make other qdiscs that use qstats per cpu as well.
>
> Signed-off-by: John Fastabe
From: Tom Herbert
Date: Tue, 23 Aug 2016 11:55:29 -0700
> Fix locking issue in kcm and losing events when paused.
Series applied, thanks Tom.
From: Tobias Regnery
Date: Tue, 23 Aug 2016 22:21:21 +0200
> @@ -993,6 +993,18 @@ static void alx_reset(struct work_struct *work)
> rtnl_unlock();
> }
>
> +static inline int alx_tpd_req(struct sk_buff *skb)
> +{
Please don't use inline for functions in foo.c files, let the compiler
deci
On Tue, 2016-08-23 at 18:22 -0400, Soheil Hassas Yeganeh wrote:
> From: Soheil Hassas Yeganeh
>
> Instead of using sock_tx_timestamp, use skb_tx_timestamp to record
> software transmit timestamp of a packet.
>
> sock_tx_timestamp resets and overrides the tx_flags of the skb.
> The function is in
From: Wei Yongjun
Use kfree_skb() instead of kfree() to free sk_buff.
Fixes: 0d051bf93c06 ("tipc: make bearer packet filtering generic")
Signed-off-by: Wei Yongjun
---
net/tipc/bearer.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/tipc/bearer.c b/net/tipc/bearer.c
in
On 16-08-23 02:08 PM, Eric Dumazet wrote:
> On Tue, 2016-08-23 at 13:23 -0700, John Fastabend wrote:
>> This patch adds a flag for queueing disciplines to indicate the
>> stack does not need to use the qdisc lock to protect operations.
>> This can be used to build lockless scheduling algorithms and
From: Soheil Hassas Yeganeh
Instead of using sock_tx_timestamp, use skb_tx_timestamp to record
software transmit timestamp of a packet.
sock_tx_timestamp resets and overrides the tx_flags of the skb.
The function is intended to be called from within the protocol
layer when creating the skb, not
On 8/23/2016 2:03 PM, Florian Fainelli wrote:
+others,
On 08/23/2016 04:13 AM, Christophe Leroy wrote:
In ERRATA DS8700A dated 05 May 2016, Microship recommends to
not use software power down mode on KSZ8041 family.
s/Microship/Microchip/
They say they have no plan to fix this ERRATA in
On Tue, Aug 23, 2016 at 04:49:33PM -0400, Lennart Sorensen wrote:
> That would be padding after the structure elements.
>
> I think what was meant is that it won't add padding in the middle of the
> structure due to alignment, ie it isn't doing:
>
> struct ethtool_wolinfo {
> __u32
On Tue, 2016-08-23 at 13:23 -0700, John Fastabend wrote:
> This patch adds a flag for queueing disciplines to indicate the stack
> does not need to use the qdisc lock to protect operations. This can
> be used to build lockless scheduling algorithms and improving
> performance.
>
> The flag is chec
On Tue, 2016-08-23 at 13:23 -0700, John Fastabend wrote:
> sch_direct_xmit() uses qdisc_qlen as a return value but all call sites
> of the routine only check if it is zero or not. Simplify the logic so
> that we don't need to return an actual queue length value.
>
> This introduces a case now wher
On Mon, Aug 22, 2016 at 04:32:51PM -0700, David Miller wrote:
> You have to post this properly to the netdev list.
>
> If netdev is not CC:'d it doesn't get properly logged in patchwork,
> and won't be applied.
Sorry, get_maintainers.pl did not find netdev list for these patches.
Could you updat
On Tue, Aug 23, 2016 at 01:34:05PM -0700, Joe Perches wrote:
> On Tue, 2016-08-23 at 21:09 +0100, Al Viro wrote:
> > On Tue, Aug 23, 2016 at 11:24:06AM -0700, David Miller wrote:
> > ... and then we can file a bug report against the sodding compiler. Note
> > that
> > struct ethtool_wolinfo {
> >
From: Eric Dumazet
Laura tracked poll() [and friends] regression caused by commit
e6afc8ace6dd ("udp: remove headers from UDP packets before queueing")
udp_poll() needs to know if there is a valid packet in receive queue,
even if its payload length is 0.
Change first_packet_length() to return a
On Tue, 2016-08-23 at 13:53 -0700, Eric Dumazet wrote:
> From: Eric Dumazet
>
...
> diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
> index e61f7cd65d08..2a2ac9e0c985 100644
> --- a/net/ipv4/udp.c
> +++ b/net/ipv4/udp.c
> @@ -1184,11 +1184,11 @@ out:
> * Drops all bad checksum frames, until a v
Fixes these userspace compile errors and warnings when glibc
netrose/rose.h is included before linux/rose.h:
linux/rose.h:25:0: warning: "SIOCRSGCAUSE" redefined
linux/rose.h:26:0: warning: "SIOCRSSCAUSE" redefined
linux/rose.h:27:0: warning: "SIOCRSL2CALL" redefined
linux/rose.h:28:0: warning: "S
Fixes these userspace compile error when glibc netax25/ax25.h is included
before linux/ax25.h:
linux/ax25.h:28:0: warning: "SIOCAX25GETUID" redefined
linux/ax25.h:38:0: warning: "SIOCAX25GETINFO" redefined
linux/ax25.h:47:3: error: conflicting types for ‘ax25_address’
linux/ax25.h:49:8: error: red
Fixes this userspace compile error when glibc netinet/ip_icmp.h is included
before linux/icmp.h:
linux/icmp.h:68:8: error: redefinition of ‘struct icmphdr’
Signed-off-by: Mikko Rapeli
---
include/uapi/linux/icmp.h| 4 +++-
include/uapi/linux/libc-compat.h | 10 ++
2 files chang
Fixes this userspace compile error when net/route.h is included before
linux/route.h:
linux/route.h:30:8: error: redefinition of ‘struct rtentry’
Signed-off-by: Mikko Rapeli
---
include/uapi/linux/libc-compat.h | 10 ++
include/uapi/linux/route.h | 4 +++-
2 files changed, 13 ins
From: Eric Dumazet
Laura tracked poll() [and friends] regression caused by commit
e6afc8ace6dd ("udp: remove headers from UDP packets before queueing")
udp_poll() needs to know if there is a valid packet in receive queue,
even if its payload length is 0.
Change first_packet_length() to return a
From: Stephen Warren
The Synopsys DWC EQoS is a configurable IP block which supports multiple
options for bus type, clocking and reset structure, and feature list.
Extend the DT binding to define a "compatible value" for the configuration
contained in NVIDIA's Tegra186 SoC, and define some new pr
On Tue, 2016-08-23 at 13:06 -0700, Laura Abbott wrote:
>
> Fixes the test for me. You're welcome to take this as a Tested-by.
Thanks Laura, I will submit an official patch immediately.
This converts the pfifo_fast qdisc to use the skb_array data structure
and set the lockless qdisc bit.
This also removes the logic used to pick the next band to dequeue from
and instead just checks a per priority array for packets from top priority
to lowest. This might need to be a bit more cleve
The sch_mqprio qdisc creates a sub-qdisc per tx queue which are then
called independently for enqueue and dequeue operations. However
statistics are aggregated and pushed up to the "master" qdisc.
This patch adds support for any of the sub-qdiscs to be per cpu
statistic qdiscs. To handle this case
On Tue, 2016-08-23 at 13:22 -0700, John Fastabend wrote:
> Currently __qdisc_run calls qdisc_run_end() but does not call
> qdisc_run_begin(). This makes it hard to track pairs of
> qdisc_run_{begin,end} across function calls.
>
> To simplify reading these code paths and simpler code this
> patch m
On Tue, 2016-08-23 at 21:09 +0100, Al Viro wrote:
> On Tue, Aug 23, 2016 at 11:24:06AM -0700, David Miller wrote:
> > > On some versions and architectures. Can you guarantee that you will
> > > notice when an exception appears?
> > Again, show me the assembler output exhibiting the lack of
> > ini
The sch_mq qdisc creates a sub-qdisc per tx queue which are then
called independently for enqueue and dequeue operations. However
statistics are aggregated and pushed up to the "master" qdisc.
This patch adds support for any of the sub-qdiscs to be per cpu
statistic qdiscs. To handle this case add
Hello everyone.
I'm working in a small ISP company, and we're using x86 servers under debian as
core and border routers.
The pair of core routers (active-backup redundancy scheme) are the dual-socket
servers with Xeon E5-2637 v2 CPUs and one 82599EB dual-port 10G NIC. Tasks of
these routers:
Reporting qlen when qlen is per cpu requires aggregating the per
cpu counters. This adds a helper routine for this.
Signed-off-by: John Fastabend
---
include/net/sch_generic.h | 15 +++
net/sched/sch_api.c |3 ++-
2 files changed, 17 insertions(+), 1 deletion(-)
diff --g
Add qdisc qlen helper routines for lockless qdiscs to use.
The qdisc qlen is no longer used in the hotpath but it is reported
via stats query on the qdisc so it still needs to be tracked. This
adds the per cpu operations needed.
Signed-off-by: John Fastabend
---
include/net/sch_generic.h |8
netif_schedule uses a bit QDISC_STATE_SCHED to tell the qdisc layer
if a run of the qdisc has been scheduler. This is important when
tearing down qdisc instances. We can not rcu_free an instance for
example if its possible that we might have outstanding references to
it.
Perhaps more importantly i
In the Hip06 SoC, the RoCE Engine is part of the Hisilicon Network
Subsystem and is dependent upon DSAF module. Therefore, certain
functions like RESET are exposed through the common registers of
HNS DSAF module which are memory-mapped by the HNS driver and
currently can only be accessed through DT
This patch is meant to add support of ACPI to the Hisilicon RoCE
driver.
Changes done are primarily meant to detect the type and then either
use DT specific or ACPI spcific functions. Where ever possible,
this patch tries to make use of Unified Device Property Interface
APIs to support both DT and
This patch is meant to add support of ACPI to the Hisilicon RoCE driver.
Following changes have been made in the driver(s):
Patch 1/2: HNS Ethernet Driver: changes to support ACPI have been done in
the RoCE reset function part of the HNS ethernet driver. Earlier it only
supported DT/syscon.
Similar to how gso is handled skb_bad_tx needs to be per cpu to handle
lockless qdisc with multiple writer/producers.
Signed-off-by: John Fastabend
---
include/net/sch_generic.h |7 +++
net/sched/sch_api.c |6 +++
net/sched/sch_generic.c | 95 +++
The qdisc_reset operation depends on the qdisc lock at the moment
to halt any additions to gso_skb and statistics while the list is
free'd and the stats zeroed.
Without the qdisc lock we can not guarantee another cpu is not in
the process of adding a skb to one of the "cells".
To resolve the dev_
In qdisc_graft_qdisc a "new" qdisc is attached and the 'qdisc_destroy'
operation is called on the old qdisc. The destroy operation will wait
a rcu grace period and call qdisc_rcu_free(). At which point
gso_cpu_skb is free'd along with all stats so no need to zero stats
and gso_cpu_skb from the graf
The per cpu qstats support was added with per cpu bstat support which
is currently used by the ingress qdisc. This patch adds a set of
helpers needed to make other qdiscs that use qstats per cpu as well.
Signed-off-by: John Fastabend
---
include/net/sch_generic.h | 39 +
The net sched infrastructure has a gso ptr that points to skb structs
that have failed to be enqueued by the device driver.
This can happen when multiple cores try to push a skb onto the same
underlying hardware queue resulting in lock contention. This case is
handled by a cpu collision handler ha
Enable dflt qdisc support for per cpu stats before this patch a
dflt qdisc was required to use the global statistics qstats and
bstats.
Signed-off-by: John Fastabend
---
net/sched/sch_generic.c | 24
1 file changed, 20 insertions(+), 4 deletions(-)
diff --git a/net/sc
sch_direct_xmit() uses qdisc_qlen as a return value but all call sites
of the routine only check if it is zero or not. Simplify the logic so
that we don't need to return an actual queue length value.
This introduces a case now where sch_direct_xmit would have returned
a qlen of zero but now it ret
Latest round of lockless qdisc patch set with performance metric
primarily using pktgen to inject pkts into the qdisc layer
This series introduces a flag to allow qdiscs to indicate they can run
without holding the qdisc lock. In order to set this bit most qdiscs
will need to be modified to use lo
This patch adds a flag for queueing disciplines to indicate the stack
does not need to use the qdisc lock to protect operations. This can
be used to build lockless scheduling algorithms and improving
performance.
The flag is checked in the tx path and the qdisc lock is only taken
if it is not set.
Currently __qdisc_run calls qdisc_run_end() but does not call
qdisc_run_begin(). This makes it hard to track pairs of
qdisc_run_{begin,end} across function calls.
To simplify reading these code paths and simpler code this
patch moves begin/end calls into qdisc_run().
Signed-off-by: John Fastabend
Add tso/tso6 support to the alx driver.
Based on information from the downstream driver at github.com/qca/alx
Signed-off-by: Tobias Regnery
---
drivers/net/ethernet/atheros/alx/main.c | 75 +++--
1 file changed, 71 insertions(+), 4 deletions(-)
diff --git a/drivers/n
On Tue, Aug 23, 2016 at 11:24:06AM -0700, David Miller wrote:
> > On some versions and architectures. Can you guarantee that you will
> > notice when an exception appears?
>
> Again, show me the assembler output exhibiting the lack of
> initialization, for this specific structure and situation.
On 16-08-17 04:00 PM, John Fastabend wrote:
> On 16-08-17 03:58 PM, Eric Dumazet wrote:
>> On Wed, 2016-08-17 at 12:36 -0700, John Fastabend wrote:
>>> Similar to how gso is handled skb_bad_tx needs to be per cpu to handle
>>> lockless qdisc with multiple writer/producers.
>> \
>>> @@ -1021,6 +1026
On 08/23/2016 12:03 PM, Eric Dumazet wrote:
On Tue, 2016-08-23 at 11:25 -0700, David Miller wrote:
From: Laura Abbott
Date: Tue, 23 Aug 2016 10:53:26 -0700
Fedora received a report[1] of a unit test failing on Ruby when using
the
4.7 kernel. This was a test to send a zero sized UDP packet. Wi
On 8/22/16 8:51 AM, Simon Horman wrote:
>
> The scheme that OvS uses so far is that mac_len denotes the number of bytes
> from the start of the MAC header until its end. In the absence of MPLS that
> will be the beginning of the network header. And in the presence of MPLS it
> will be the beginnin
Lock the lower socket in kcm_unattach. Release during call to strp_done
since that function cancels the RX timers and work queue with sync.
Also added some status information in psock reporting.
Signed-off-by: Tom Herbert
---
net/kcm/kcmproc.c | 20 +++-
net/kcm/kcmsock.c | 13 +
Fix locking issue in kcm and losing events when paused.
Tom Herbert (2):
strparser: Queue work when being unpaused
kcm: Fix locking issue
include/net/strparser.h | 5 +
net/kcm/kcmproc.c | 20 +++-
net/kcm/kcmsock.c | 13 -
net/strparser/str
When the upper layer unpauses a stream parser connection we need to
queue rx_work to make sure no events are missed.
Signed-off-by: Tom Herbert
---
include/net/strparser.h | 5 +
net/strparser/strparser.c | 11 +++
2 files changed, 12 insertions(+), 4 deletions(-)
diff --git a/in
From: sthem...@exchange.microsoft.com
Date: Tue, 23 Aug 2016 12:29:32 -0700
> Hyper-V network driver cleanups.
>
> The only new functionality is minor extensions to ethtool.
Series applied.
+others,
On 08/23/2016 04:13 AM, Christophe Leroy wrote:
> In ERRATA DS8700A dated 05 May 2016, Microship recommends to
> not use software power down mode on KSZ8041 family.
s/Microship/Microchip/
> They say they have no plan to fix this ERRATA in future releases.
The errata applies to spec
On Tue, 2016-08-23 at 11:25 -0700, David Miller wrote:
> From: Laura Abbott
> Date: Tue, 23 Aug 2016 10:53:26 -0700
>
> > Fedora received a report[1] of a unit test failing on Ruby when using
> > the
> > 4.7 kernel. This was a test to send a zero sized UDP packet. With the
> > 4.7 kernel, the tes
David Vrabel writes:
> On 22/08/16 16:42, Vitaly Kuznetsov wrote:
>>
>> I see two ways to fix the issue:
>> - Change the 'wire' protocol between netfront and netback to start keeping
>> the original SKB structure. We'll have to add a flag indicating the fact
>> that the particular request is
On 8/23/16 12:06 PM, Lorenzo Colitti wrote:
> @@ -713,10 +728,11 @@ static bool valid_markcond(const struct inet_diag_bc_op
> *op, int len,
> return len >= *min_len;
> }
>
> -static int inet_diag_bc_audit(struct nlattr *attr)
> +static int inet_diag_bc_audit(struct nlattr *attr, const str
On 8/23/2016 2:39 PM, Leon Romanovsky wrote:
> On Tue, Aug 23, 2016 at 07:10:53PM +0300, Saeed Mahameed wrote:
>> On Tue, Aug 23, 2016 at 6:57 PM, Doug Ledford wrote:
>>> On 8/23/2016 4:49 AM, Saeed Mahameed wrote:
From Hadar and Ilya, added the needed hardware bits and infrastructure for
>>
1 - 100 of 270 matches
Mail list logo
