[patch net 0/3] mlxsw: Couple of fixes

From: Jiri Pirko Couple of simple fixes from Arkadi and Elad. Please queue these up for stable. Thanks. Arkadi Sharshevsky (2): mlxsw: spectrum: Fix memory leak at skb reallocation mlxsw: switchx2: Fix memory leak at skb reallocation Elad Raz (1): mlxsw: pci: Fix EQE

[patch net 1/3] mlxsw: spectrum: Fix memory leak at skb reallocation

From: Arkadi Sharshevsky During transmission the skb is checked for headroom in order to add vendor specific header. In case the skb needs to be re-allocated, skb_realloc_headroom() is called to make a private copy of the original, but doesn't release it. Current code

[patch net 2/3] mlxsw: switchx2: Fix memory leak at skb reallocation

From: Arkadi Sharshevsky During transmission the skb is checked for headroom in order to add vendor specific header. In case the skb needs to be re-allocated, skb_realloc_headroom() is called to make a private copy of the original, but doesn't release it. Current code

[patch net 3/3] mlxsw: pci: Fix EQE structure definition

From: Elad Raz The event_data starts from address 0x00-0x0C and not from 0x08-0x014. This leads to duplication with other fields in the Event Queue Element such as sub-type, cqn and owner. Fixes: eda6500a987a0 ("mlxsw: Add PCI bus implementation") Signed-off-by: Elad Raz

[PATCH iproute2/net-next 2/2] tc: flower: Support matching ARP

Support matching on ARP operation, and hardware and protocol addresses for Ethernet hardware and IPv4 protocol addresses. Example usage: tc qdisc add dev eth0 ingress tc filter add dev eth0 protocol arp parent : flower indev eth0 \ arp_op request arp_sip 10.0.0.1 action

[PATCH iproute2/net-next 1/2] tc: flower: update headers for TCA_FLOWER_KEY_ARP*

Present in net-next. Signed-off-by: Simon Horman --- include/linux/pkt_cls.h | 11 +++ 1 file changed, 11 insertions(+) diff --git a/include/linux/pkt_cls.h b/include/linux/pkt_cls.h index a081efbd61a2..1e5e1ddfdaca 100644 --- a/include/linux/pkt_cls.h +++

[PATCH iproute2/net-next 0/2] net/sched: cls_flower: Support matching ARP

Add support for support matching on ARP operation, and hardware and protocol addresses for Ethernet hardware and IPv4 protocol addresses. Changes since RFC: * Drop RFC designation; kernel patches are present in net-next Simon Horman (2): tc: flower: update headers for TCA_FLOWER_KEY_ARP* tc:

Re: [PATCH] i40e: Invoke softirqs after napi_reschedule

On Thu, 2017-01-12 at 17:04 -0800, Benjamin Poirier wrote: > The following message is logged from time to time when using i40e: > NOHZ: local_softirq_pending 08 > > i40e may schedule napi from a workqueue. Afterwards, softirqs are not run > in a deterministic time frame. The problem is the same

Re: [PATCH v5 01/13] net: ethernet: aquantia: Make and configuration files.

On 01/12/2017 09:06 PM, Joe Perches wrote: On Thu, 2017-01-12 at 21:02 -0800, Alexander Loktionov wrote: From: David VomLehn Patches to create the make and configuration files. This patch should _really_ be the last in the series not the first. Could you explain the basis

Re: [PATCH v5 01/13] net: ethernet: aquantia: Make and configuration files.

On Thu, 2017-01-12 at 21:24 -0800, David VomLehn wrote: > On 01/12/2017 09:06 PM, Joe Perches wrote: > > On Thu, 2017-01-12 at 21:02 -0800, Alexander Loktionov wrote: > > > From: David VomLehn > > > > > > Patches to create the make and configuration files. > > > > This patch

[PATCH net-next v2 12/13] tcp: remove thin_dupack feature

Thin stream DUPACK is to start fast recovery on only one DUPACK provided the connection is a thin stream (i.e., low inflight). But this older feature is now subsumed with RACK. If a connection receives only a single DUPACK, RACK would arm a reordering timer and soon starts fast recovery instead

[PATCH net-next v2 06/13] tcp: check undo conditions before detecting losses

Currently RACK would mark loss before the undo operations in TCP loss recovery. This could incorrectly identify real losses as spurious. For example a sender first experiences a delay spike and then eventually some packets were lost due to buffer overrun. In this case, the sender should perform

[PATCH net-next v2 05/13] tcp: use sequence to break TS ties for RACK loss detection

The packets inside a jumbo skb (e.g., TSO) share the same skb timestamp, even though they are sent sequentially on the wire. Since RACK is based on time, it can not detect some packets inside the same skb are lost. However, we can leverage the packet sequence numbers as extended timestamps to

[PATCH net-next v2 09/13] tcp: remove forward retransmit feature

Forward retransmit is an esoteric feature in RFC3517 (condition(3) in the NextSeg()). Basically if a packet is not considered lost by the current criteria (# of dupacks etc), but the congestion window has room for more packets, then retransmit this packet. However it actually conflicts with the

[PATCH net-next v2 10/13] tcp: remove early retransmit

This patch removes the support of RFC5827 early retransmit (i.e., fast recovery on small inflight with <3 dupacks) because it is subsumed by the new RACK loss detection. More specifically when RACK receives DUPACKs, it'll arm a reordering timer to start fast recovery after a quarter of (min)RTT,

[PATCH net-next v2 11/13] tcp: remove RFC4653 NCR

This patch removes the (partial) implementation of the aggressive limited transmit in RFC4653 TCP Non-Congestion Robustness (NCR). NCR is a mitigation to the problem created by the dynamic DUPACK threshold. With the current adaptive DUPACK threshold (tp->reordering) could cause timeouts by

[PATCH net-next v2 00/13] tcp: RACK fast recovery

The patch set enables RACK loss detection (draft-ietf-tcpm-rack-01) to trigger fast recovery with a reordering timer. Previously RACK has been running in auxiliary mode where it is used to detect packet losses once the recovery has triggered by other algorithms (e.g., FACK). By inspecting packet

[PATCH net-next v2 01/13] tcp: new helper function for RACK loss detection

Create a new helper tcp_rack_mark_skb_lost to prepare the upcoming RACK reordering timer support. Signed-off-by: Yuchung Cheng Signed-off-by: Neal Cardwell Acked-by: Eric Dumazet --- net/ipv4/tcp_recovery.c | 21

[PATCH net-next v2 08/13] tcp: extend F-RTO to catch more spurious timeouts

Current F-RTO reverts cwnd reset whenever a never-retransmitted packet was (s)acked. The timeout can be declared spurious because the packets acknoledged with this ACK was transmitted before the timeout, so clearly not all the packets are lost to reset the cwnd. This nice detection does not

[PATCH net-next v2 13/13] tcp: disable fack by default

This patch disables FACK by default as RACK is the successor of FACK (inspired by the insights behind FACK). FACK[1] in Linux works as follows: a packet P is deemed lost, if packet Q of higher sequence is s/acked and P and Q are distant by at least dupthresh number of packets in sequence space.

[PATCH net] openvswitch: maintain correct checksum state in conntrack actions

When executing conntrack actions on skbuffs with checksum mode CHECKSUM_COMPLETE, the checksum must be updated to account for header pushes and pulls. Otherwise we get "hw csum failure" logs similar to this (ICMP packet received on geneve tunnel via ixgbe NIC): [ 405.740065] genev_sys_6081: hw

Re: [PATCH iproute2 v4 3/4] ifstat: Add 64 bits based stats to extended statistics

On Thu, 12 Jan 2017 15:49:50 +0200 Nogah Frankel wrote: > The default stats for ifstat are 32 bits based. > The kernel supports 64 bits based stats. (They are returned in struct > rtnl_link_stats64 which is an exact copy of struct rtnl_link_stats, in > which the "normal"

Re: [net PATCH v3 1/5] virtio_net: use dev_kfree_skb for small buffer XDP receive

On 2017年01月13日 10:50, John Fastabend wrote: In the small buffer case during driver unload we currently use put_page instead of dev_kfree_skb. Resolve this by adding a check for virtnet mode when checking XDP queue type. Also name the function so that the code reads correctly to match the

[PATCH v5 10/13] net: ethernet: aquantia: Hardware interface and utility functions

From: David VomLehn Add functions to interface with the hardware and some utility functions. Signed-off-by: Alexander Loktionov Signed-off-by: Dmitrii Tarakanov Signed-off-by: Pavel Belous

[PATCH v5 09/13] net: ethernet: aquantia: Atlantic hardware abstraction layer

From: David VomLehn Add common functions for Atlantic hardware abstraction layer. Signed-off-by: Alexander Loktionov Signed-off-by: Dmitrii Tarakanov Signed-off-by: Pavel Belous

[PATCH v5 08/13] net: ethernet: aquantia: PCI operations

From: David VomLehn Add functions that handle the PCI bus interface. Signed-off-by: Alexander Loktionov Signed-off-by: Dmitrii Tarakanov Signed-off-by: Pavel Belous Signed-off-by:

[PATCH v5 07/13] net: ethernet: aquantia: Vector operations

From: David VomLehn Add functions to manululate the vector of receive and transmit rings. Signed-off-by: Alexander Loktionov Signed-off-by: Dmitrii Tarakanov Signed-off-by: Pavel.Belous

[PATCH v5 03/13] net: ethernet: aquantia: Add ring support code

From: David VomLehn Add code to support the transmit and receive ring buffers. Signed-off-by: Alexander Loktionov Signed-off-by: Dmitrii Tarakanov Signed-off-by: Pavel Belous

[PATCH v5 02/13] net: ethernet: aquantia: Common functions and definitions

From: David VomLehn Add files containing the functions and definitions used in common in different functional areas. Signed-off-by: Alexander Loktionov Signed-off-by: Dmitrii Tarakanov Signed-off-by: Pavel

[PATCH v5 00/13] net: ethernet: aquantia: Add AQtion 2.5/5 GB NIC driver

From: David VomLehn v1: Initial version v2: o Make necessary drivers/net/ethernet changes to integrate software o Drop intermediate atlantic directory o Remove Makefile things only appropriate to out of tree module building v3: o Move changes to

Re: [net-next 01/13] tcp: new helper function for RACK loss detection

On Thu, Jan 12, 2017 at 10:03 PM, Yuchung Cheng wrote: > > Create a new helper tcp_rack_mark_skb_lost to prepare the > upcoming RACK reordering timer support. > > Signed-off-by: Yuchung Cheng > Signed-off-by: Neal Cardwell > Acked-by:

[net PATCH v2 2/5] net: virtio: wrap rtnl_lock in test for calling with lock already held

For XDP use case and to allow ethtool reset tests it is useful to be able to use reset routines from contexts where rtnl lock is already held. Signed-off-by: John Fastabend --- drivers/net/virtio_net.c | 16 +--- 1 file changed, 9 insertions(+), 7

Re: [PATCH iproute2] Add support for rt_protos.d

On Mon, 9 Jan 2017 15:43:09 -0800 David Ahern wrote: > Add support for reading proto id/name mappings from rt_protos.d > directory. Allows users to have custom protocol values converted > to human friendly names. > > Each file under rt_protos.d has the 'id name'

Re: [PATCH iproute2] ip6tunnel: Align ipv6 tunnel key display with ipv4

On Tue, 10 Jan 2017 10:45:54 + David Forster wrote: > Show ipv6 tunnel keys on presence of GRE_KEY flag for tunnel types > other than GRE. Aligns ipv6 behaviour with ipv4. > > Signed-off-by: dfors...@brocade.com Applied thanks.

Re: [PATCH iproute2] rttable: Fix invalid range checking when table id is converted to u32

On Tue, 10 Jan 2017 15:33:55 -0800 David Ahern wrote: > Frank reported that table ids for very large numbers are not properly > detected: > $ ip li add foobar type vrf table 98765432100123456789 > > command succeeds and resulting table id is actually: > > 21: foobar:

Re: [PATCH iproute2/net-next 0/2] net/sched: cls_flower: Support matching ARP

On Thu, 12 Jan 2017 09:11:57 +0100 Simon Horman wrote: > Add support for support matching on ARP operation, and hardware and > protocol addresses for Ethernet hardware and IPv4 protocol addresses. > > Changes since RFC: > * Drop RFC designation; kernel patches are

Re: [net PATCH v2 0/5] virtio_net XDP fixes and adjust_header support

On 17-01-12 04:34 PM, John Fastabend wrote: > This has a fix to handle small buffer free logic correctly and then > also adds adjust head support. > > I pushed adjust head at net (even though its rc3) to avoid having > to push another exception case into virtio_net to catch if the > program uses

Re: [RFC] [PATCH] audit: log 32-bit socketcalls

On 2017-01-12 16:32, Paul Moore wrote: > On Thu, Jan 12, 2017 at 7:36 AM, Richard Guy Briggs wrote: > > 32-bit socketcalls were not being logged by audit on x86_64 systems. > > Log them. > > > > See: https://github.com/linux-audit/audit-kernel/issues/14 > > > > Signed-off-by:

[PATCH v5 05/13] net: ethernet: aquantia: Support for NIC-specific code

From: David VomLehn Add support for code specific to the Atlantic NIC. Signed-off-by: Alexander Loktionov Signed-off-by: Dmitrii Tarakanov Signed-off-by: Pavel Belous

[PATCH v5 12/13] net: ethernet: aquantia: Receive side scaling

From: David VomLehn Add definitions that support receive side scaling. Signed-off-by: Alexander Loktionov Signed-off-by: Dmitrii Tarakanov Signed-off-by: Pavel Belous

[PATCH v5 01/13] net: ethernet: aquantia: Make and configuration files.

From: David VomLehn Patches to create the make and configuration files. Signed-off-by: Alexander Loktionov Signed-off-by: Dmitrii Tarakanov Signed-off-by: Pavel Belous

[PATCH v5 06/13] net: ethernet: aquantia: Atlantic A0 and B0 specific functions.

From: David VomLehn Add Atlantic A0 and B0 specific functions. Signed-off-by: Alexander Loktionov Signed-off-by: Dmitrii Tarakanov Signed-off-by: Pavel Belous Signed-off-by: Dmitry

[PATCH v5 11/13] net: ethernet: aquantia: Ethtool support

From: David VomLehn Add the driver interfaces required for support by the ethtool utility. Signed-off-by: Alexander Loktionov Signed-off-by: Dmitrii Tarakanov Signed-off-by: Pavel Belous

[PATCH v5 13/13] net: ethernet: aquantia: Integrate AQtion 2.5/5 GB NIC driver

From: David VomLehn Modify the drivers/net/ethernet/{Makefile,Kconfig} file to make them a part of the network drivers build. Signed-off-by: Alexander Loktionov Signed-off-by: Dmitrii Tarakanov Signed-off-by:

Re: [PATCH] xen-netfront: Fix Rx stall during network stress and OOM

From: Vineeth Remanan Pillai Date: Wed, 11 Jan 2017 23:17:17 + > @@ -1054,7 +1059,11 @@ static int xennet_poll(struct napi_struct *napi, int > budget) > napi_complete(napi); > > RING_FINAL_CHECK_FOR_RESPONSES(>rx, more_to_do); > -

Re: To netlink or not to netlink, that is the question

From: "Jason A. Donenfeld" Date: Thu, 12 Jan 2017 21:07:43 +0100 > David - you concur with Stephen that ioctl is really not okay and I > should absolutely do netlink? Yes.

Re: [PATCH net v2] ipv6: sr: fix several BUGs when preemption is enabled

On 01/12/2017 09:26 PM, David Lebrun wrote: > From: Vivien Didelot No idea where does that come from, resending patch. Sorry about that Vivien. David signature.asc Description: OpenPGP digital signature

Re: [PATCH] [net] net/mlx5e: fix another -Wmaybe-uninitialized warning

On Thu, Jan 12, 2017 at 6:04 PM, Arnd Bergmann wrote: > On Thursday, January 12, 2017 5:21:49 PM CET Or Gerlitz wrote: >> When I build here without CONFIG_INET in my system, the build goes fine >> with this approach. However, we're pretty sure that in the past we got >> 0-day

Re: [PATCH] tcp: fix tcp_fastopen unaligned access complaints on sparc

From: Shannon Nelson Date: Thu, 12 Jan 2017 12:56:08 -0800 > > > On 1/12/2017 12:41 PM, David Miller wrote: >> From: Shannon Nelson >> Date: Thu, 12 Jan 2017 12:30:38 -0800 >> >>> On 1/12/2017 12:25 PM, Eric Dumazet wrote: On Thu,

[net PATCH 5/5] virtio_net: XDP support for adjust_head

Add support for XDP adjust head by allocating a 256B header region that XDP programs can grow into. This is only enabled when a XDP program is loaded. In order to ensure that we do not have to unwind queue headroom push queue setup below bpf_prog_add. It reads better to do a prog ref unwind vs

Re: [PATCH] tcp: fix tcp_fastopen unaligned access complaints on sparc

From: Shannon Nelson Date: Thu, 12 Jan 2017 13:58:10 -0800 > > > On 1/12/2017 1:47 PM, David Miller wrote: >> From: Eric Dumazet >> Date: Thu, 12 Jan 2017 13:36:30 -0800 >> >>> val should be an union, so that proper alignment is enforced by

Re: [net-next PATCH 1/3] Revert "icmp: avoid allocating large struct on stack"

On Tue, Jan 10, 2017 at 1:48 PM, Eric Dumazet wrote: > On Tue, 2017-01-10 at 21:08 +0100, Jesper Dangaard Brouer wrote: >> On Tue, 10 Jan 2017 10:44:59 -0800 Cong Wang >> wrote: >> >> > On Tue, Jan 10, 2017 at 10:12 AM, David Miller

RE: Marvell Phy (1510) issue since v4.7 kernel

> -Original Message- > From: Russell King - ARM Linux [mailto:li...@armlinux.org.uk] > Sent: Thursday, January 12, 2017 4:42 PM > To: Kwok, WingMan > Cc: Andrew Lunn; Karicheri, Muralidharan; netdev@vger.kernel.org > Subject: Re: Marvell Phy (1510) issue since v4.7 kernel > > On Thu,

Re: [net-next PATCH 1/3] Revert "icmp: avoid allocating large struct on stack"

On Tue, Jan 10, 2017 at 10:54 AM, David Miller wrote: > From: Cong Wang > Date: Tue, 10 Jan 2017 10:44:59 -0800 >> The only countries you hold netdev are Canada, Japan and Spain >> (to my knowledge). If you check: >> >>

Re: [PATCH] xen-netfront: Fix Rx stall during network stress and OOM

On 01/12/2017 12:17 PM, David Miller wrote: From: Vineeth Remanan Pillai Date: Wed, 11 Jan 2017 23:17:17 + @@ -1054,7 +1059,11 @@ static int xennet_poll(struct napi_struct *napi, int budget) napi_complete(napi);

Re: [PATCH] can: Fix kernel panic at security_sock_rcv_skb

On Thu, 2017-01-12 at 14:40 -0800, william.c.robe...@intel.com wrote: > From: Zhang Yanmin > > The patch is for fix the below kernel panic: > BUG: unable to handle kernel NULL pointer dereference at (null) > IP: [] selinux_socket_sock_rcv_skb+0x65/0x2a0 Same patch was

Re: [Patch net] atm: remove an unnecessary loop

On Thu, Jan 12, 2017 at 4:07 PM, Francois Romieu wrote: > Cong Wang : > [...] >> diff --git a/net/atm/common.c b/net/atm/common.c >> index a3ca922..7ec3bbc 100644 >> --- a/net/atm/common.c >> +++ b/net/atm/common.c >> @@ -72,10 +72,11 @@ static

[PATCH] i40e: Invoke softirqs after napi_reschedule

The following message is logged from time to time when using i40e: NOHZ: local_softirq_pending 08 i40e may schedule napi from a workqueue. Afterwards, softirqs are not run in a deterministic time frame. The problem is the same as what was described in commit ec13ee80145c ("virtio_net: invoke

Re: [PATCH 5/6] treewide: use kv[mz]alloc* rather than opencoded variants

> On Jan 12, 2017, at 08:37, Michal Hocko wrote: > > From: Michal Hocko > > There are many code paths opencoding kvmalloc. Let's use the helper > instead. The main difference to kvmalloc is that those users are usually > not considering all the aspects of

Re: [PATCH v5 01/13] net: ethernet: aquantia: Make and configuration files.

On Thu, 2017-01-12 at 21:02 -0800, Alexander Loktionov wrote: > From: David VomLehn > > Patches to create the make and configuration files. This patch should _really_ be the last in the series not the first.

[net-next 11/13] tcp: remove RFC4653 NCR

This patch removes the (partial) implementation of the aggressive limited transmit in RFC4653 TCP Non-Congestion Robustness (NCR). NCR is a mitigation to the problem created by the dynamic DUPACK threshold. With the current adaptive DUPACK threshold (tp->reordering) could cause timeouts by

[net-next 08/13] tcp: extend F-RTO to catch more spurious timeouts

Current F-RTO reverts cwnd reset whenever a never-retransmitted packet was (s)acked. The timeout can be declared spurious because the packets acknoledged with this ACK was transmitted before the timeout, so clearly not all the packets are lost to reset the cwnd. This nice detection does not

[net-next 05/13] tcp: use sequence to break TS ties for RACK loss detection

The packets inside a jumbo skb (e.g., TSO) share the same skb timestamp, even though they are sent sequentially on the wire. Since RACK is based on time, it can not detect some packets inside the same skb are lost. However, we can leverage the packet sequence numbers as extended timestamps to

[net-next 06/13] tcp: check undo conditions before detecting losses

Currently RACK would mark loss before the undo operations in TCP loss recovery. This could incorrectly identify real losses as spurious. For example a sender first experiences a delay spike and then eventually some packets were lost due to buffer overrun. In this case, the sender should perform

[net-next 12/13] tcp: remove thin_dupack feature

Thin stream DUPACK is to start fast recovery on only one DUPACK provided the connection is a thin stream (i.e., low inflight). But this older feature is now subsumed with RACK. If a connection receives only a single DUPACK, RACK would arm a reordering timer and soon starts fast recovery instead

[net-next 02/13] tcp: new helper for RACK to detect loss

Create a new helper tcp_rack_detect_loss to prepare the upcoming RACK reordering timer patch. Signed-off-by: Yuchung Cheng Signed-off-by: Neal Cardwell Acked-by: Eric Dumazet --- include/net/tcp.h | 3 +--

[net-next 01/13] tcp: new helper function for RACK loss detection

Create a new helper tcp_rack_mark_skb_lost to prepare the upcoming RACK reordering timer support. Signed-off-by: Yuchung Cheng Signed-off-by: Neal Cardwell Acked-by: Eric Dumazet --- net/ipv4/tcp_recovery.c | 21

[net-next 10/13] tcp: remove early retransmit

This patch removes the support of RFC5827 early retransmit (i.e., fast recovery on small inflight with <3 dupacks) because it is subsumed by the new RACK loss detection. More specifically when RACK receives DUPACKs, it'll arm a reordering timer to start fast recovery after a quarter of (min)RTT,

[net-next 09/13] tcp: remove forward retransmit feature

Forward retransmit is an esoteric feature in RFC3517 (condition(3) in the NextSeg()). Basically if a packet is not considered lost by the current criteria (# of dupacks etc), but the congestion window has room for more packets, then retransmit this packet. However it actually conflicts with the

[net-next 04/13] tcp: add reordering timer in RACK loss detection

This patch makes RACK install a reordering timer when it suspects some packets might be lost, but wants to delay the decision a little bit to accomodate reordering. It does not create a new timer but instead repurposes the existing RTO timer, because both are meant to retransmit packets.

[net-next 13/13] tcp: disable fack by default

This patch disables FACK by default as RACK is the successor of FACK (inspired by the insights behind FACK). FACK[1] in Linux works as follows: a packet P is deemed lost, if packet Q of higher sequence is s/acked and P and Q are distant by at least dupthresh number of packets in sequence space.

[net-next 07/13] tcp: enable RACK loss detection to trigger recovery

This patch changes two things: 1. Start fast recovery with RACK in addition to other heuristics (e.g., DUPACK threshold, FACK). Prior to this change RACK is enabled to detect losses only after the recovery has started by other algorithms. 2. Disable TCP early retransmit. RACK subsumes

Re: [PATCH] [v2] net: qcom/emac: add ethtool support

Timur Tabi wrote: +static void emac_get_pauseparam(struct net_device *netdev, + struct ethtool_pauseparam *pause) +{ + struct phy_device *phydev = netdev->phydev; + + if (phydev) { + if (phydev->autoneg) +

Re: [PATCH net-next v2 00/10] net: dsa: Support for pdata in dsa2

Hi Florian, Florian Fainelli writes: > Hi all, > > This is not exactly new, and was sent before, although back then, I did not > have an user of the pre-declared MDIO board information, but now we do. Note > that I have additional changes queued up to have b53 register

[net PATCH v3 0/5] virtio_net XDP fixes and adjust_header support

This has a fix to handle small buffer free logic correctly and then also adds adjust head support. I pushed adjust head at net (even though its rc3) to avoid having to push another exception case into virtio_net to catch if the program uses adjust_head and then block it. If there are any strong

[net PATCH v3 1/5] virtio_net: use dev_kfree_skb for small buffer XDP receive

In the small buffer case during driver unload we currently use put_page instead of dev_kfree_skb. Resolve this by adding a check for virtnet mode when checking XDP queue type. Also name the function so that the code reads correctly to match the additional check. Signed-off-by: John Fastabend

[net PATCH v3 4/5] virtio_net: remove duplicate queue pair binding in XDP

Factor out qp assignment. Signed-off-by: John Fastabend --- drivers/net/virtio_net.c | 18 +++--- 1 file changed, 7 insertions(+), 11 deletions(-) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index ec54644..6041828 100644 ---

[net PATCH v3 2/5] net: virtio: wrap rtnl_lock in test for calling with lock already held

For XDP use case and to allow ethtool reset tests it is useful to be able to use reset routines from contexts where rtnl lock is already held. Signed-off-by: John Fastabend --- drivers/net/virtio_net.c | 16 +--- 1 file changed, 9 insertions(+), 7

[net PATCH v2 1/5] virtio_net: use dev_kfree_skb for small buffer XDP receive

In the small buffer case during driver unload we currently use put_page instead of dev_kfree_skb. Resolve this by adding a check for virtnet mode when checking XDP queue type. Also name the function so that the code reads correctly to match the additional check. Signed-off-by: John Fastabend

[net PATCH v2 0/5] virtio_net XDP fixes and adjust_header support

This has a fix to handle small buffer free logic correctly and then also adds adjust head support. I pushed adjust head at net (even though its rc3) to avoid having to push another exception case into virtio_net to catch if the program uses adjust_head and then block it. If there are any strong

Re: [PATCH v2 7/7] uapi: export all headers under uapi directories

On Thu, Jan 12, 2017 at 05:32:09PM +0100, Nicolas Dichtel wrote: > What I was trying to say is that I export those directories like other are. > Removing those files is not related to that series. Perhaps the correct solution is to only copy files matching "*.h" to reduce the risk of copying

Re: [PATCH iproute2] bridge: fdb: add state filter support

On Thu, 12 Jan 2017 17:47:39 +0100 Nikolay Aleksandrov wrote: > This patch adds a new argument to the bridge fdb show command that allows > to filter by entry state. > Also update the man page to include all available show arguments. > > Signed-off-by: Nikolay

Re: [PATCH] i40e: Invoke softirqs after napi_reschedule

On 2017/01/12 17:15, Eric Dumazet wrote: > On Thu, 2017-01-12 at 17:04 -0800, Benjamin Poirier wrote: > > The following message is logged from time to time when using i40e: > > NOHZ: local_softirq_pending 08 > > > > i40e may schedule napi from a workqueue. Afterwards, softirqs are not run > > in

Re: [PATCH iproute2] Add support for rt_protos.d

On 1/12/17 6:27 PM, Stephen Hemminger wrote: >> diff --git a/etc/iproute2/rt_protos.d/README >> b/etc/iproute2/rt_protos.d/README >> new file mode 100644 >> index ..723509ce56b6 >> --- /dev/null >> +++ b/etc/iproute2/rt_protos.d/README >> @@ -0,0 +1,3 @@ >> +Each file in this

[net PATCH v3 5/5] virtio_net: XDP support for adjust_head

Add support for XDP adjust head by allocating a 256B header region that XDP programs can grow into. This is only enabled when a XDP program is loaded. In order to ensure that we do not have to unwind queue headroom push queue setup below bpf_prog_add. It reads better to do a prog ref unwind vs

[net PATCH v3 3/5] virtio_net: factor out xdp handler for readability

At this point the do_xdp_prog is mostly if/else branches handling the different modes of virtio_net. So remove it and handle running the program in the per mode handlers. Signed-off-by: John Fastabend --- drivers/net/virtio_net.c | 76

Re: [PATCH net-next v2 05/10] drivers: base: Add device_find_class()

From: Florian Fainelli Date: Thu, 12 Jan 2017 14:50:39 -0800 > Well, this is really so that we don't need to cast the arguments passed > to device_find_child(), which takes a void *data as well. Aha, I didn't catch that, my bad.

[net-next 03/13] tcp: record most recent RTT in RACK loss detection

Record the most recent RTT in RACK. It is often identical to the "ca_rtt_us" values in tcp_clean_rtx_queue. But when the packet has been retransmitted, RACK choses to believe the ACK is for the (latest) retransmitted packet if the RTT is over minimum RTT. This requires passing the arrival time of

[net-next 00/13] RACK fast recovery

The patch set enables RACK loss detection (draft-ietf-tcpm-rack-01) to trigger fast recovery with a reordering timer. Previously RACK has been running in auxiliary mode where it is used to detect packet losses once the recovery has triggered by other algorithms (e.g., FACK). By inspecting packet

[PATCH net-next v2 03/13] tcp: record most recent RTT in RACK loss detection

Record the most recent RTT in RACK. It is often identical to the "ca_rtt_us" values in tcp_clean_rtx_queue. But when the packet has been retransmitted, RACK choses to believe the ACK is for the (latest) retransmitted packet if the RTT is over minimum RTT. This requires passing the arrival time of

[PATCH net-next v2 04/13] tcp: add reordering timer in RACK loss detection

This patch makes RACK install a reordering timer when it suspects some packets might be lost, but wants to delay the decision a little bit to accomodate reordering. It does not create a new timer but instead repurposes the existing RTO timer, because both are meant to retransmit packets.

[PATCH net-next v2 07/13] tcp: enable RACK loss detection to trigger recovery

This patch changes two things: 1. Start fast recovery with RACK in addition to other heuristics (e.g., DUPACK threshold, FACK). Prior to this change RACK is enabled to detect losses only after the recovery has started by other algorithms. 2. Disable TCP early retransmit. RACK subsumes

[PATCH net-next v2 02/13] tcp: new helper for RACK to detect loss

Create a new helper tcp_rack_detect_loss to prepare the upcoming RACK reordering timer patch. Signed-off-by: Yuchung Cheng Signed-off-by: Neal Cardwell Acked-by: Eric Dumazet --- include/net/tcp.h | 3 +--

Re: [PATCH 5/6] treewide: use kv[mz]alloc* rather than opencoded variants

> diff --git a/drivers/xen/evtchn.c b/drivers/xen/evtchn.c > index 6890897a6f30..10f1ef582659 100644 > --- a/drivers/xen/evtchn.c > +++ b/drivers/xen/evtchn.c > @@ -87,18 +87,6 @@ struct user_evtchn { > bool enabled; > }; > > -static evtchn_port_t *evtchn_alloc_ring(unsigned int size) >

Re: [PATCH] tcp: fix tcp_fastopen unaligned access complaints on sparc

On Thu, 2017-01-12 at 13:15 -0700, Rob Gardner wrote: > > I suspect that someplace, somebody is casting val to an int * or > something like that. Then that would be the bug. Can we root cause this please ?

[PATCH net v3] ipv6: sr: fix several BUGs when preemption is enabled

When CONFIG_PREEMPT=y, CONFIG_IPV6=m and CONFIG_SEG6_HMAC=y, seg6_hmac_init() is called during the initialization of the ipv6 module. This causes a subsequent call to smp_processor_id() with preemption enabled, resulting in the following trace. [ 20.451460] BUG: using smp_processor_id() in

Re: [Patch] Fixing order of trinary operator when assigning the ifindex in XFRM Decode Session

Hi Carlos: patches sent to netdev should be plain text only. Using git send-email would be best. Also, the Subject line should start with the area affected (e.g., net: xfrm:) and really should be a bit shorter. e.g., something like this: [PATCH] net: xfrm: fix flow oif in decode_session On

Re: [PATCH] tcp: fix tcp_fastopen unaligned access complaints on sparc

From: Shannon Nelson Date: Thu, 12 Jan 2017 12:30:38 -0800 > On 1/12/2017 12:25 PM, Eric Dumazet wrote: >> On Thu, 2017-01-12 at 13:15 -0700, Rob Gardner wrote: >> >>> >>> I suspect that someplace, somebody is casting val to an int * or >>> something like that. >> >>

Re: [PATCH] tcp: fix tcp_fastopen unaligned access complaints on sparc

From: Eric Dumazet Date: Thu, 12 Jan 2017 13:36:30 -0800 > val should be an union, so that proper alignment is enforced by one > member. Sure, annotating the type so that it is aligned correctly makes sense.

[�PATCH net-next] ARM: dts: vf610-zii-dev: add EEPROM entry to Rev B

The ZII Dev Rev B board has EEPROMs hanging the 88E6352 Ethernet switch chips. Add an "eeprom-length" property to allow access from ethtool. Signed-off-by: Vivien Didelot --- arch/arm/boot/dts/vf610-zii-dev-rev-b.dts | 2 ++ 1 file changed, 2 insertions(+)

  1   2   3   4   >