[PATCH V3 2/2] vhost_net: conditionally enable tx polling

We always poll tx for socket, this is sub optimal since: - it will be only used when we exceed the sndbuf of the socket. - since we use two independent polls for tx and vq, this will slightly increase the waitqueue traversing time and more important, vhost could not benefit from commit

[PATCH V3 0/2] vhost_net polling optimization

Hi: This series tries to optimize vhost_net polling at two points: - Stop rx polling for reduicng the unnecessary wakeups during handle_rx(). - Conditonally enable tx polling for reducing the unnecessary traversing and spinlock touching. Test shows about 17% improvement on rx pps. Please

[PATCH V3 1/2] vhost_net: stop polling socket during rx processing

We don't stop rx polling socket during rx processing, this will lead unnecessary wakeups from under layer net devices (E.g sock_def_readable() form tun). Rx will be slowed down in this way. This patch avoids this by stop polling socket during rx processing. A small drawback is that this introduces

Re: [PATCH net-next] tcp: accept RST if SEQ matches right edge of SACK block

On Tue, May 31, 2016 at 9:50 AM, Pau Espin wrote: > > Hi, sorry re-sending because I didn't have plain text mode enabled and > message didn't arrive to the mailing list. Also dropping some mail > addresses from RFC authors which probed to be unavailable anymore. > > The

Re: [v2 PATCH 1/2] macvlan: Fix potential use-after free for broadcasts

On Tue, May 31, 2016 at 09:19:37PM -0700, Cong Wang wrote: > > Hmm, why could this happen? The upper device should be linked > with the lower device, where a refcount is already held. > Also, the work is cancelled in ->uninit(). Of course it can happen. We are talking about the source macvlan

Re: [PATCH 8/8] mwifiex: use better message and error code when OF node doesn't match

Hi All, On Sat, May 28, 2016 at 12:18 AM, Javier Martinez Canillas wrote: > The Documentation/devicetree/bindings/net/wireless/marvell-sd8xxx.txt DT > binding document lists the possible compatible strings that a SDIO child > node can have, so the driver checks if the

Re: [PATCH 7/8] mwifiex: don't print an error if an optional DT property is missing

Hi All, On Sat, May 28, 2016 at 12:18 AM, Javier Martinez Canillas wrote: > The Documentation/devicetree/bindings/net/wireless/marvell-sd8xxx.txt DT > binding document say that the "interrupts" property in the child node is > optional. So the property being missed

Re: [PATCH 6/8] mwifiex: check if mwifiex_sdio_probe_of() fails and return error

Hi All, On Sat, May 28, 2016 at 12:18 AM, Javier Martinez Canillas wrote: > The function can fail so the returned value should be checked > and the error propagated to the caller in case of a failure. > > Signed-off-by: Javier Martinez Canillas

Re: [v2 PATCH 1/2] macvlan: Fix potential use-after free for broadcasts

On Tue, May 31, 2016 at 8:43 PM, Herbert Xu wrote: > When we postpone a broadcast packet we save the source port in > the skb if it is local. However, the source port can disappear > before we get a chance to process the packet. > Hmm, why could this happen? The

Re: [PATCH 4/8] mwifiex: consolidate mwifiex_sdio_probe() error paths

Hi All, On Sat, May 28, 2016 at 12:18 AM, Javier Martinez Canillas wrote: > Instead of duplicating part of the cleanups needed in case of an error > in .probe callback, have a single error path and use goto labels as is > common practice in the kernel. > > This also has

Re: [PATCH 3/8] mwifiex: propagate mwifiex_add_card() errno code in mwifiex_sdio_probe()

Hi All, On Sat, May 28, 2016 at 12:18 AM, Javier Martinez Canillas wrote: > There's only a check if mwifiex_add_card() returned a nonzero value, but > the actual error code is neither stored nor propagated to the caller. So > instead of always returning -1 (which is

Re: [PATCH 5/8] mwifiex: use dev_err() instead of pr_err() in mwifiex_sdio_probe()

Hi All, On Sat, May 28, 2016 at 12:18 AM, Javier Martinez Canillas wrote: > It's better to have the device name prefixed in the error message. > > Signed-off-by: Javier Martinez Canillas This looks right to me. Reviewed-by: Julian Calaby

Re: [PATCH 2/8] mwifiex: propagate sdio_enable_func() errno code in mwifiex_sdio_probe()

Hi All, On Sat, May 28, 2016 at 12:18 AM, Javier Martinez Canillas wrote: > If the sdio_enable_func() function fails on .probe, the -EIO errno code > is always returned but that could make more difficult to debug and find > the cause of why the function actually failed. >

Re: [PATCH 1/8] mwifiex: only call mwifiex_sdio_probe_of() if dev has an OF node

Hi All, On Sat, May 28, 2016 at 12:18 AM, Javier Martinez Canillas wrote: > SDIO is an auto enumerable bus so the SDIO devices are matched using the > sdio_device_id table and not using compatible strings from a OF id table. > > However, commit ce4f6f0c353b ("mwifiex: add

[PATCH iputils 6/6] ping: fix -i number parsing in locales

Always use #.# format for the -i flag even when the current locale uses a different separator. Locale de_DE which uses #,# normally. Simple testcase: $ make USE_IDN=1 $ LANG=de_DE.UTF8 ./ping -i 0.5 localhost Reported-by: Sergey Fionov Signed-off-by: Mike Frysinger

[PATCH iputils 4/6] fix handling of CFLAGS

This defaults CFLAGS to -O3 without clobbering settings people have set up in the environment already. Signed-off-by: Mike Frysinger --- Makefile | 13 + 1 file changed, 5 insertions(+), 8 deletions(-) diff --git a/Makefile b/Makefile index 7147f08..362f1c8

[PATCH iputils 2/6] doc: fix parallel build of html/man pages

The use of the same tempdir prevents building of these files in parallel. So build all of them in unique tempdirs so we can do them in parallel. Signed-off-by: Mike Frysinger --- doc/Makefile | 37 - 1 file changed, 24 insertions(+), 13

[PATCH iputils 3/6] ping6: allow disabling of openssl support

Signed-off-by: Mike Frysinger --- Makefile | 5 - iputils_md5dig.h | 4 +++- ping6.c | 10 ++ 3 files changed, 17 insertions(+), 2 deletions(-) diff --git a/Makefile b/Makefile index 54e5a6d..7147f08 100644 --- a/Makefile +++ b/Makefile @@ -36,7

[PATCH iputils 1/6] start gitignore files

Signed-off-by: Mike Frysinger --- .gitignore | 22 ++ doc/.gitignore | 2 ++ 2 files changed, 24 insertions(+) create mode 100644 .gitignore create mode 100644 doc/.gitignore diff --git a/.gitignore b/.gitignore new file mode 100644 index

[PATCH iputils 5/6] tftpd: fix syslog setup

Commit d81a44625b04d487c895473aa77af13420b7afdd added support for checking the set*id calls, but would call syslog() before it had called openlog(). Move the call up earlier to fix that. Signed-off-by: Mike Frysinger --- tftpd.c | 3 ++- 1 file changed, 2 insertions(+), 1

[v2 PATCH 2/2] macvlan: Avoid unnecessary multicast cloning

Currently we always queue a multicast packet for further processing, even if none of the macvlan devices are subscribed to the address. This patch optimises this by adding a global multicast filter for a macvlan_port. Note that this patch doesn't handle the broadcast addresses of the individual

[v2 PATCH 1/2] macvlan: Fix potential use-after free for broadcasts

When we postpone a broadcast packet we save the source port in the skb if it is local. However, the source port can disappear before we get a chance to process the packet. This patch fixes this by holding a ref count on the netdev. It also delays the skb->cb modification until after we allocate

[v2 PATCH 0/2] macvlan: Avoid unnecessary multicast cloning

On Tue, May 31, 2016 at 02:07:13PM -0700, David Miller wrote: > > I think you need to set the vlan->port->mc_filter to all 1's in the > PROMISC/ALLMUTI branch here. > > Otherwise packets won't properly pass your new hash test. Good point. Here's v2. This patch tries to improve macvlan

[GIT] Networking

1) Fix negative error code usage in ATM layer, from Stefan Hajnoczi. 2) If CONFIG_SYSCTL is disabled, the default TTL is not initialized properly. From Ezequiel Garcia. 3) Missing spinlock init in mvneta driver, from Gregory CLEMENT. 4) Missing unlocks in hwmb error paths, also from

Re: [PATCH v2 net] sfc: Track RPS flow IDs per channel instead of per function

From: Edward Cree Date: Tue, 31 May 2016 19:12:32 +0100 > From: Jon Cooper > > Otherwise we get confused when two flows on different channels get the > same flow ID. > > Signed-off-by: Edward Cree Applied, thanks.

[PATCH iproute2 net-next] iproute2: correct port in FOU/GRE example

This resolves what appears to be a typo. Cc: Tom Herbert Reviewed-by: Dinan Gunawardena Signed-off-by: Simon Horman --- man/man8/ip-fou.8 | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git

Re: [PATCH iproute2 net-next] iproute2: correct port in FOU/GRE example

On Wed, Jun 01, 2016 at 11:26:03AM +0900, Simon Horman wrote: > From: Simon Horman > > This resolves what appears to be a typo. > > Cc: Tom Herbert > Reviewed-by: Dinan Gunawardena > Signed-off-by: Simon Horman

[PATCH iproute2 net-next] iproute2: correct port in FOU/GRE example

From: Simon Horman This resolves what appears to be a typo. Cc: Tom Herbert Reviewed-by: Dinan Gunawardena Signed-off-by: Simon Horman --- man/man8/ip-fou.8 | 2 +- 1 file changed, 1 insertion(+),

Re: [PATCH v2 3/5] drivers: net: phy: Add MDIO driver

On Tue, May 31, 2016 at 05:10:38PM -0700, Iyappan Subramanian wrote: > +static int xgene_mdio_reset(struct xgene_mdio_pdata *pdata) > +{ > + int ret; > + > + if (pdata->mdio_id == XGENE_MDIO_RGMII) { > + if (pdata->dev->of_node) { > +

Re: r8169 regression: UDP packets dropped intermittantly

On Wed, May 18, 2016 at 04:21:37PM +0930, Jonathan Woithe wrote: > On Thu, Apr 07, 2016 at 12:14:18PM +0930, Jonathan Woithe wrote: > > On Mon, Feb 08, 2016 at 01:03:19PM +1030, Jonathan Woithe wrote: > > > On Wed, Dec 02, 2015 at 12:58:52AM +0100, Francois Romieu wrote: > > > > Jonathan Woithe

[PATCH v2 4/5] dtb: xgene: Add MDIO node

Added mdio node for mdio driver. Also added phy-handle reference to the ethernet nodes. Removed unused mdio subnode within storm menet ethernet node. Removed unused clock node from storm sgenet1. Signed-off-by: Iyappan Subramanian Tested-by: Fushen Chen

[PATCH v2 3/5] drivers: net: phy: Add MDIO driver

Currently, SGMII based 1G rely on the hardware registers for link state and sometimes it's not reliable. To get most accurate link state, this interface has to use the MDIO bus to poll the PHY. In X-Gene SoC, MDIO bus is shared across RGMII and SGMII based 1G interfaces, so adding this driver to

[PATCH v2 1/5] drivers: net: xgene: MAC and PHY configuration changes

This patch fixes MAC configuration to support 10/100GbE for SGMII and link_state call back. It also sets pdata->mdio_driver flag based on ethernet mdio subnode and prepare for MDIO driver support. In summary, following are the changes, - Added set_speed function pointer in mac_ops - Changed

[PATCH v2 0/5] drivers: net: xgene: Fix 1G hot-plug and module support

This patchset addresses the following issues, 1. hot-plug issue on the SGMII 1G interface - by adding a driver for MDIO management 2. fixes the kernel crash when the driver loaded as an kernel module - by fixing hardware cleanups and rearrange kernel API calls Signed-off-by:

[PATCH v2 5/5] drivers: net: xgene: Fix module load/unload crash

When the driver is configured as kernel module and when it gets unloaded and reloaded, kernel crash was observed, due to incomplete hardware cleanup. This patch addresses this issue with the following changes, - Reordered mac enable and disable - Added hardware prefetch buffer cleanup - Added Tx

[PATCH v2 2/5] drivers: net: xgene: Backward compatibility with older firmware

This patch looks for CONFIG_MDIO_XGENE and based on phy-handle DT/ACPI fields, sets the mdio_driver flag. The rest of the driver uses the this flag for any MDIO management, in the case of backward compatibility. Also, some code clean up done around mdio configuration/remove. Signed-off-by:

Re: [PATCH] hfsc: ensure class is added to eltree exactly once

Eric Dumazet wrote: > On Wed, 2016-06-01 at 01:00 +0200, Florian Westphal wrote: > Generally speaking, a non work conserving qdisc could return NULL if it > decides to drop all packet(s) that were sitting in the queue. > > Say we add a 'max sojourn time' on skbs, as yet

Re: [PATCH v1 5/6] dtb: xgene: Remove clock nodes

On Mon, May 30, 2016 at 12:34 AM, Matthias Brugger wrote: > > > On 27/05/16 09:22, Iyappan Subramanian wrote: >> >> Since the MDIO will be responsible for clock reset, removing the clock >> nodes from shadowcat xge0 and storm sgenet1. >> >> Signed-off-by: Iyappan

Re: [PATCH] hfsc: ensure class is added to eltree exactly once

On Wed, 2016-06-01 at 01:00 +0200, Florian Westphal wrote: > Eric Dumazet wrote: > > > diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c > > > index d783d7c..0854be3 100644 > > > --- a/net/sched/sch_hfsc.c > > > +++ b/net/sched/sch_hfsc.c > > > @@ -1583,6 +1583,7 @@

Re: [PATCH] hfsc: ensure class is added to eltree exactly once

Eric Dumazet wrote: > > diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c > > index d783d7c..0854be3 100644 > > --- a/net/sched/sch_hfsc.c > > +++ b/net/sched/sch_hfsc.c > > @@ -1583,6 +1583,7 @@ hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch) > > { > >

Re: [PATCH] qed: fix qed_fill_link() error handling

On Tuesday, May 31, 2016 2:20:46 PM CEST David Miller wrote: > From: Yuval Mintz > Date: Mon, 30 May 2016 16:24:07 + > > >> +if (IS_ENABLED(CONFIG_QED_SRIOV) && !IS_PF(hwfn->cdev)) { > >> +qed_vf_get_link_params(hwfn, params); > >> +

[PATCH net-next] udp: avoid csum_partial() for validated skb

From: Eric Dumazet In commit e6afc8ace6dd5 ("udp: remove headers from UDP packets before queueing"), udp_csum_pull_header() helper was added but missed fact that CHECKSUM_UNNECESSARY packets were now converted to CHECKSUM_NONE and skb->csum_valid was set to 1 for them.

Fwd: Investment Proposal

Business Proposal view the attached letter for more details'' CONFIDENTIAL INVESTMENT PROPOSAL.doc Description: MS-Word document

Re: [PATCH net v3] virtio_net: fix virtnet_open and virtnet_probe competing for try_fill_recv

From: Yunjian Wang Date: Tue, 31 May 2016 11:52:43 +0800 > In function virtnet_open() and virtnet_probe(), func try_fill_recv() may > be executed at the same time. VQ in virtqueue_add() has not been protected > well and BUG_ON will be triggered when virito_net.ko being

Re: usbnet: smsc95xx: fix link detection for disabled autonegotiation

From: Christoph Fritz Date: Mon, 30 May 2016 22:07:42 +0200 > On Sun, 2016-05-29 at 22:30 -0700, David Miller wrote: >> From: Christoph Fritz >> Date: Thu, 26 May 2016 04:06:47 +0200 >> >> > @@ -1695,6 +1745,7 @@ static int

Update on XDP

Hello netdev, XDP (eXpress Data Path) is a high performance, programmable network data path that is being developed as a community effort (initially within IO-visor). XDP will be useful for packet processing in forwarding, load balancing, DOS mitigation, firewalls, etc. See

Re: [PATCH] qed: fix qed_fill_link() error handling

From: Yuval Mintz Date: Mon, 30 May 2016 16:24:07 + >> +if (IS_ENABLED(CONFIG_QED_SRIOV) && !IS_PF(hwfn->cdev)) { >> +qed_vf_get_link_params(hwfn, params); >> +qed_vf_get_link_state(hwfn, link); >> +qed_vf_get_link_caps(hwfn,

Re: [PATCH net] bnx2x: avoid leaking memory on bnx2x_init_one() failures

From: Vitaly Kuznetsov Date: Mon, 30 May 2016 15:00:54 +0200 > bnx2x_init_bp() allocates memory with bnx2x_alloc_mem_bp() so if we > fail later in bnx2x_init_one() we need to free this memory > with bnx2x_free_mem_bp() to avoid leakages. E.g. I'm observing memory > leaks

Re: [PATCH v2] fib_rules: Added NLM_F_EXCL support to fib_nl_newrule

From: David Ahern Date: Mon, 30 May 2016 20:21:29 -0600 > missing a namespace (fr_net) compare. Everything else is relative to > it. > > I suggest making the rule compare a helper function. I also explicitly asked the submitted to use a helper function.

Re: [PATCH] fou: fix IPv6 Kconfig options

From: Arnd Bergmann Date: Tue, 31 May 2016 22:42:11 +0200 > The Kconfig options I added to work around broken compilation ended > up screwing up things more, as I used the wrong symbol to control > compilation of the file, resulting in IPv6 fou support to never be built > into the

Re: [PATCH 2/2] macvlan: Avoid unnecessary multicast cloning

From: Herbert Xu Date: Mon, 30 May 2016 16:28:28 +0800 > @@ -725,6 +730,8 @@ static void macvlan_set_mac_lists(struct net_device *dev) > { > struct macvlan_dev *vlan = netdev_priv(dev); > > + dev_uc_sync(vlan->lowerdev, dev); > +

Re: [PATCH net v2] openvswitch: update checksum in {push,pop}_mpls

From: Simon Horman Date: Mon, 30 May 2016 14:04:25 +0900 > In the case of CHECKSUM_COMPLETE the skb checksum should be updated in > {push,pop}_mpls() as they the type in the ethernet header. > > As suggested by Pravin Shelar. > > Cc: Pravin Shelar

Re: [PATCH v2 4/7] skbuff: introduce skb_gso_validate_mtu

On Tue, May 31, 2016 at 12:54 PM, Marcelo Ricardo Leitner wrote: > On Tue, May 31, 2016 at 12:07:54PM -0700, Alexander Duyck wrote: >> On Tue, May 31, 2016 at 11:55 AM, Marcelo Ricardo Leitner >> wrote: >> > skb_gso_network_seglen is not

[PATCH] fou: fix IPv6 Kconfig options

The Kconfig options I added to work around broken compilation ended up screwing up things more, as I used the wrong symbol to control compilation of the file, resulting in IPv6 fou support to never be built into the kernel. Changing CONFIG_NET_FOU_IPV6_TUNNELS to CONFIG_IPV6_FOU fixes that

Re: [PATCH v2 4/7] skbuff: introduce skb_gso_validate_mtu

On Tue, May 31, 2016 at 12:07:54PM -0700, Alexander Duyck wrote: > On Tue, May 31, 2016 at 11:55 AM, Marcelo Ricardo Leitner > wrote: > > skb_gso_network_seglen is not enough for checking fragment sizes if > > skb is using GSO_BY_FRAGS as we have to check frag per frag.

Re: [iproute PATCH] man: ip, ip-link: Fix ip option location

On Mon, 30 May 2016 20:46:27 +0200 Phil Sutter wrote: > This patch drops the redundant description of some of ip's options in > ip-link.8's description of the 'show' subcommand, preserving the > description of -iec (but appending it to the list in ip.8 with minor > fixes). > >

Re: [PATCH v2] Make builds default to quiet mode

On Tue, 24 May 2016 15:04:49 -0700 David Ahern wrote: > Similar to the Linux kernel and perf add infrastructure to reduce the > amount of output tossed to a user during a build. Full build output > can be obtained with 'make V=1' > > Builds go from: > > make[1]:

Re: [PATCH net v2] openvswitch: update checksum in {push,pop}_mpls

On Sun, May 29, 2016 at 10:04 PM, Simon Horman wrote: > In the case of CHECKSUM_COMPLETE the skb checksum should be updated in > {push,pop}_mpls() as they the type in the ethernet header. > > As suggested by Pravin Shelar. > > Cc: Pravin Shelar >

Re: [PATCH iproute2 net-next 3/5] bridge: add json support for bridge fdb show

On Tue, May 31, 2016 at 11:59 AM, Stephen Hemminger wrote: > On Fri, 27 May 2016 21:37:14 -0700 > Roopa Prabhu wrote: > >> Sample output: >> $bridge -j fdb show >> [{ >> "mac": "44:38:39:00:69:88", >> "dev": "swp2s0", >>

Re: [PATCH v2 4/7] skbuff: introduce skb_gso_validate_mtu

On Tue, May 31, 2016 at 11:55 AM, Marcelo Ricardo Leitner wrote: > skb_gso_network_seglen is not enough for checking fragment sizes if > skb is using GSO_BY_FRAGS as we have to check frag per frag. > > This patch introduces skb_gso_validate_mtu, based on the former,

Re: [PATCH iproute2] Added support for selection of new HSR version

On Mon, 30 May 2016 15:32:07 +0200 Peter Heise wrote: > A new HSR version was added in 4.7 that can be enabled > via iproute2. Per default the old version is selected, > however, with "ip link add [..] type hsr [..] version 1" > the newer version can be enabled. > >

Re: [PATCH iproute2] Added support for selection of new HSR version

On Mon, 30 May 2016 15:32:07 +0200 Peter Heise wrote: > A new HSR version was added in 4.7 that can be enabled > via iproute2. Per default the old version is selected, > however, with "ip link add [..] type hsr [..] version 1" > the newer version can be enabled. > >

Re: [PATCH iproute2 net-next 3/5] bridge: add json support for bridge fdb show

On Fri, 27 May 2016 21:37:14 -0700 Roopa Prabhu wrote: > Sample output: > $bridge -j fdb show > [{ > "mac": "44:38:39:00:69:88", > "dev": "swp2s0", > "vlan": 2, > "master": "br0", > "state": "permanent" > },{ > "mac":

Re: [PATCHv2 net] sctp: sctp_diag should dump sctp socket type

From: Xin Long Date: Sun, 29 May 2016 17:42:13 +0800 > Now we cannot distinguish that one sk is a udp or sctp style when > we use ss to dump sctp_info. it's necessary to dump it as well. > > For sctp_diag, ss support is not officially available, thus there > are no

Re: [PATCH net v2 1/1] net: fec: update dirty_tx even if no skb

From: Troy Kisky Date: Fri, 27 May 2016 13:30:40 -0700 > If dirty_tx isn't updated, then dma_unmap_single > can be called twice. > > This fixes a > [ 58.420980] [ cut here ] > [ 58.425667] WARNING: CPU: 0 PID: 377 at >

Re: [PATCH iproute2 net-next 1/5] json_writer: allow base json data type to be array or object

On Fri, 27 May 2016 21:37:12 -0700 Roopa Prabhu wrote: > From: Anuradha Karuppiah > > This patch adds a type qualifier to json_writer. Type can be a > json object or array. This can be extended to other types like > json-string,

[PATCH v2 6/7] sctp: Add GSO support

SCTP has this pecualiarity that its packets cannot be just segmented to (P)MTU. Its chunks must be contained in IP segments, padding respected. So we can't just generate a big skb, set gso_size to the fragmentation point and deliver it to IP layer. This patch takes a different approach. SCTP will

[PATCH v2 7/7] sctp: improve debug message to also log curr pkt and new chunk size

This is useful for debugging packet sizes. Signed-off-by: Marcelo Ricardo Leitner Tested-by: Xin Long --- net/sctp/output.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/sctp/output.c b/net/sctp/output.c index

Re: [PATCH net v5] vlan: Propagate MAC address to VLANs

From: Mike Manning Date: Fri, 27 May 2016 17:45:07 +0100 > The MAC address of the physical interface is only copied to the VLAN > when it is first created, resulting in an inconsistency after MAC > address changes of only newly created VLANs having an up-to-date MAC. > >

[PATCH v2 3/7] sk_buff: allow segmenting based on frag sizes

This patch allows segmenting a skb based on its frags sizes instead of based on a fixed value. Signed-off-by: Marcelo Ricardo Leitner Tested-by: Xin Long --- include/linux/skbuff.h | 5 + net/core/skbuff.c | 10 +++--- 2 files

[PATCH v2 4/7] skbuff: introduce skb_gso_validate_mtu

skb_gso_network_seglen is not enough for checking fragment sizes if skb is using GSO_BY_FRAGS as we have to check frag per frag. This patch introduces skb_gso_validate_mtu, based on the former, which will wrap the use case inside it as all calls to skb_gso_network_seglen were to validate if it

[PATCH v2 1/7] loopback: make use of NETIF_F_GSO_SOFTWARE

NETIF_F_GSO_SOFTWARE was defined to list all GSO software types, so lets make use of it in loopback code. Note that veth/vxlan/others already uses it. Within this patch series, this patch causes lo to pick up SCTP GSO feature automatically (as it's added to NETIF_F_GSO_SOFTWARE) and thus avoiding

[PATCH v2 2/7] skbuff: export skb_gro_receive

sctp GSO requires it and sctp can be compiled as a module, so we need to export this function. Signed-off-by: Marcelo Ricardo Leitner Tested-by: Xin Long --- net/core/skbuff.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/core/skbuff.c

[PATCH v2 0/7] sctp: Add GSO support

This patchset adds sctp GSO support. Performance tests indicates that increases throughput by 10% if using bigger chunk sizes, specially if bigger than MTU. For small chunks, it doesn't help much if not using heavy firewall rules. For small chunks it will probably be of more use once we get

[PATCH v2 5/7] sctp: delay as much as possible skb_linearize

This patch is a preparation for the GSO one. In order to successfully handle GSO packets on rx path we must not call skb_linearize, otherwise it defeats any gain GSO may have had. This patch thus delays as much as possible the call to skb_linearize, leaving it to sctp_inq_pop() moment. For that

Re: [patch] atm: firestream: add more reserved strings

From: Dan Carpenter Date: Fri, 27 May 2016 13:33:50 +0300 > This bug was there when the driver was first added in back in year 2000. > It causes a Smatch warning: > > drivers/atm/firestream.c:849 process_incoming() > error: buffer overflow 'res_strings' 60 <=

Re: [patch] atm: iphase: off by one in rx_pkt()

From: Dan Carpenter Date: Fri, 27 May 2016 13:34:35 +0300 > The iadev->rx_open[] array holds "iadev->num_vc" pointers (this code > assumes that pointers are 32 bits). So the > here should be >= or else > we could end up reading a garbage pointer from one element beyond

Re: [PATCH 1/2] timer: Export destroy_hrtimer_on_stack()

From: Guenter Roeck Date: Thu, 26 May 2016 17:21:05 -0700 > hrtimer_init_on_stack() needs a matching call to > destroy_hrtimer_on_stack(), so both need to be exported. > > Signed-off-by: Guenter Roeck Applied.

Re: [PATCH 2/2] net: pktgen: Call destroy_hrtimer_on_stack()

From: Guenter Roeck Date: Thu, 26 May 2016 17:21:06 -0700 > If CONFIG_DEBUG_OBJECTS_TIMERS=y, hrtimer_init_on_stack() requires > a matching call to destroy_hrtimer_on_stack() to clean up timer > debug objects. > > Signed-off-by: Guenter Roeck Applied.

Re: [PATCH 0/2] macvlan: Avoid unnecessary multicast cloning

From: Herbert Xu Date: Tue, 31 May 2016 08:49:45 +0800 > On Mon, May 30, 2016 at 07:27:59PM +0300, Lennert Buytenhek wrote: >> >> That and stack switches to kworker threads and serialisation on >> the bc_queue queue lock. > > My patch should resolve these problems

Re: [PATCH] net: Fragment large datagrams even when IP_HDRINCL is set.

From: Alan Davey Date: Mon, 23 May 2016 15:23:45 +0100 > One of the bugs documented in the raw(7) man page is as follows: When the > IP_HDRINCL option is set, datagrams will not be fragmented and are limited to > the interface MTU. > > This patch fixes the bug by

Re: [PATCH] net: fec: dt: convert phy-reset-gpios to use GPIOD API

Hello Russell, On Tue, May 31, 2016 at 07:26:42PM +0100, Russell King - ARM Linux wrote: > On Tue, May 31, 2016 at 02:24:50PM -0400, George G. Davis wrote: > > Hello Fabio, > > > > On Tue, May 31, 2016 at 02:26:54PM -0300, Fabio Estevam wrote: > > > On Tue, May 31, 2016 at 2:17 PM, George G.

Re: [PATCH] net: fec: dt: convert phy-reset-gpios to use GPIOD API

On Tue, May 31, 2016 at 02:24:50PM -0400, George G. Davis wrote: > Hello Fabio, > > On Tue, May 31, 2016 at 02:26:54PM -0300, Fabio Estevam wrote: > > On Tue, May 31, 2016 at 2:17 PM, George G. Davis > > wrote: > > > Convert the i.MX FEC driver and DT files to use the

Re: [PATCH] net: fec: dt: convert phy-reset-gpios to use GPIOD API

Hello Fabio, On Tue, May 31, 2016 at 02:26:54PM -0300, Fabio Estevam wrote: > On Tue, May 31, 2016 at 2:17 PM, George G. Davis > wrote: > > Convert the i.MX FEC driver and DT files to use the GPIOD API > > which automatically handles GPIO polarity. > > > > Signed-off-by:

[PATCH v2 net] sfc: Track RPS flow IDs per channel instead of per function

From: Jon Cooper Otherwise we get confused when two flows on different channels get the same flow ID. Signed-off-by: Edward Cree --- drivers/net/ethernet/sfc/efx.c| 32 +++-

[PATCH net-next v2] ravb: Add ESF in RCR for enabling separation filter

From: Masaru Nagai The H/W manual recommends B'10 or B'11 in a value of the separation filtering select bits in the receive configuration register (RCR.ESF). When B'10 is set, frames from non-matching streams are discarded. When B'11 is set, frames from non-matching

Re: [PATCH] net: fec: dt: convert phy-reset-gpios to use GPIOD API

Hello George, On Tue, May 31, 2016 at 01:17:51PM -0400, George G. Davis wrote: > Convert the i.MX FEC driver and DT files to use the GPIOD API > which automatically handles GPIO polarity. additionally to what Fabio said, there is also a nicer approach in the making. The resulting dts would look

Re: [PATCH] net: fec: dt: convert phy-reset-gpios to use GPIOD API

On Tue, May 31, 2016 at 2:17 PM, George G. Davis wrote: > Convert the i.MX FEC driver and DT files to use the GPIOD API > which automatically handles GPIO polarity. > > Signed-off-by: George G. Davis In order to avoid breaking old dtb's a

[PATCH] net: fec: dt: convert phy-reset-gpios to use GPIOD API

Convert the i.MX FEC driver and DT files to use the GPIOD API which automatically handles GPIO polarity. Signed-off-by: George G. Davis --- Documentation/devicetree/bindings/net/fsl-fec.txt | 7 ++- arch/arm/boot/dts/imx25-karo-tx25.dts | 3 ++-

Re: [PATCH net-next] tcp: accept RST if SEQ matches right edge of SACK block

Hi, sorry re-sending because I didn't have plain text mode enabled and message didn't arrive to the mailing list. Also dropping some mail addresses from RFC authors which probed to be unavailable anymore. The problem is not caused by the host-wide rate limit. I analyzed the scenario with tcpdump

Re: [PATCH net-next] ravb: Add ESF in RCR for enabling separation filter

Hi Sergei, 2016-05-31 6:19 GMT+09:00 Sergei Shtylyov : > On 05/30/2016 05:30 PM, Sergei Shtylyov wrote: > >>> From: Masaru Nagai >>> >>> This patch adds enabling separation filter(ESF) is setting value of B'11. >>> This setting

Re: [PATCH net-next] tcp: accept RST if SEQ matches right edge of SACK block

On Tue, 2016-05-31 at 13:38 +0200, Pau Espin Pedrol wrote: > RFC 5961 advises to only accept RST packets containing a seq number > matching the next expected seq number instead of the whole receive > window in order to avoid spoofing attacks. > > However, this situation is not optimal in the case

[PATCH] tcp: extend window to fit all restored unacked data in a send queue

Data from a send queue is sent only if there is enough space in a window, so when we restore unacked data, we need to expand a window to fit this data. Currently we don't do this and tcp_acceptable_seq() returns tcp_wnd_end(tp) instead of tp->snd_nxt. The other side drops packets with this seq,

Re: [PATCH] hfsc: ensure class is added to eltree exactly once

On Tue, 2016-05-31 at 12:12 +0200, Florian Westphal wrote: > Intent is to insert the class into the eligible tree when first packet > is enqueued (its removed from list when class becomes empty again). > > Checking for a size of 1 is problematic: > > 1. child qdisc might have segmented the skb,

Re: [PATCH 1/1] net: ethernet: Add SGMII support to dwmac-socfpga

Hello Loh Tien On 5/31/2016 11:10 AM, Loh Tien Hock wrote: Hi Peppe, Sorry for the late reply. no pbl at all. I believe my patch's title is a little confusing. The patch is to enable Altera TSE PCS SGMII support, not to add SGMII support to stmmac (well that in a way tests SGMII for

Re: codel/fq_codel triggers heaps of WARNs in net/sched/sch_hfsc.c:1426

>> I'll submit a hfsc patch soon (it does fix a real problem). Just FYI, the patch from another thread [PATCH] hfsc: ensure class is added to eltree exactly once doesn't help the situation, I just tested it. The debugging output is exactly the same as before. -mk

[PATCH v2 3/7] binding: mdio-mux: Add DT binding doc for Broadcom MDIO bus mutiplexer

Add DT binding doc for Broadcom MDIO bus mutiplexer driver. Signed-off-by: Pramod Kumar --- .../bindings/net/brcm,mdio-mux-iproc.txt | 60 ++ 1 file changed, 60 insertions(+) create mode 100644

[PATCH v2 2/7] DT: phy.txt: Add mdio-integrated-mux property

This property is used by integrated MDIO multiplexer which has bus selection and mdio transaction generation logic, integrated inside. Signed-off-by: Pramod Kumar --- Documentation/devicetree/bindings/net/mdio-mux.txt | 9 - 1 file changed, 8 insertions(+), 1

[PATCH v2 4/7] DT:mdio-mux: Add mdio multiplexer driver node

Add integrated MDIO multiplexer driver node which contains two mux PCIe bus and one ethernet bus along with phys lying on these bus. Signed-off-by: Pramod Kumar --- arch/arm64/boot/dts/broadcom/ns2-svk.dts | 12 arch/arm64/boot/dts/broadcom/ns2.dtsi|

[PATCH v2 0/7] Add MDIO bus multiplexer support for iProc SoCs

Broadcom iProc based SoCs use a MDIO bus multiplexer where child buses could be internal as well external to SoCs. These buses could supports MDIO transaction compatible to C-22/C-45. Broadcom MDIO bus mulitplexer is an integrated multiplexer where child bus selection and mdio transaction logic

  1   2   >