Re: [PATCH] ipv4: do not cache local route when using SO_BINDTODEVICE

From: Kouya Shimura Date: Tue, 23 Feb 2016 14:27:32 +0900 > After v3.6, output routes are cached, however, the 'rt_iif' field of > struct rtable can not be shared by various traffics using SO_BINDTODEVICE. > It causes that traffic can not reach the local receiver which also

Re: [PATCH][net-next][v2] bridge: allow the maximum mtu to 64k

From: Li RongQing Date: Thu, 25 Feb 2016 09:50:41 +0800 > On Thu, Feb 25, 2016 at 5:44 AM, Stephen Hemminger > wrote: >>> This is especially annoying for the virtualization case because the >>> KVM's tap driver will by default adopt the

for stable: xfrm: Let the flowcache handle its size by default.

Hi David, can you queue commit c386578f1cdb4dac230395a951f88027f64346e3 for stable? Without it, any system with > 16 cpus and a lot of ipsec use, that leaves xfrmN_gc_thresh set to its default, will (eventually) see failures to create new ipsec connections. It fixes commit

Re: [PATCH net 1/5] ppp: lock ppp structure before modifying mru in ppp_ioctl()

On Wed, Feb 24, 2016 at 03:32:02PM -0500, David Miller wrote: > From: Guillaume Nault > Date: Mon, 22 Feb 2016 20:47:13 +0100 > > > PPP's Tx and Rx paths read ppp->mru under protection of ppp_xmit_lock() > > and ppp_recv_lock() respectively. > > Therefore ppp_ioctl() must

Re: [Patch net-next v2 0/2] net_sched: add network namespace support for tc actions

From: Cong Wang Date: Mon, 22 Feb 2016 15:57:51 -0800 > This patchset adds network namespace support for tc actions. > > v2: > * pull the first patch into net-next > * reduce code duplication by introducing more helper functions Series applied, thanks.

Re: [PATCH net-next 01/10] bnxt_en: Improve bnxt_vf_update_mac().

From: Michael Chan Date: Thu, 25 Feb 2016 09:15:39 -0800 > I agree it is complicated. The default, if nobody does anything, is random > MAC. But random MAC has disadvantages, so we allow some options > for the PF or VF users to change it. Ok, thanks for explaining.

Re: [PATCH RFC 0/3] intermediate representation for jit and cls_u32 conversion

On 16-02-25 09:37 AM, Pablo Neira Ayuso wrote: > Hi, > > This patchset contains the core infrastructure for the generic > intermediate representation that I presented during NetDev 1.1's > nftables switchdev talk. This includes the basic infrastructure to > convert the tc cls_u32 based on John's

[PATCH] 3c59x: mask LAST_FRAG bit from length field in ring

Recently, I fixed a bug in 3c59x: commit 6e144419e4da11a9a4977c8d899d7247d94ca338 Author: Neil Horman Date: Wed Jan 13 12:43:54 2016 -0500 3c59x: fix another page map/single unmap imbalance Which correctly rebalanced dma mapping and unmapping types. Unfortunately

[PATCH RFC 2/3] net: intermediate representation for jit translation

This patch adds the generic intermediate representation that consists of abstract syntax tree (ast) that is expressed via a list of statements. Currently there are two type of statements: * Expression, that is used to match packet selector and meta information. Expressions are basically small

[PATCH RFC 3/3] net: convert tc_u32 to use the intermediate representation

This patch moves the u32 parser from the ixgbe that John has made to the core u32. This parser has been adapted to build the intermediate representation. To store the parsing information, this patch introduces a parse table object, one per device, so we don't need to store the parsing states in

[PATCH RFC 0/3] intermediate representation for jit and cls_u32 conversion

Hi, This patchset contains the core infrastructure for the generic intermediate representation that I presented during NetDev 1.1's nftables switchdev talk. This includes the basic infrastructure to convert the tc cls_u32 based on John's parser. The main goals of this patchset are: * Provide an

[PATCH RFC 1/3] net: ixgbe: add struct igxbe_filter

Just a container structure, this change is required by follow up patches that rework the u32 classifier offload on top of the generic intermediate representation. Signed-off-by: Pablo Neira Ayuso --- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 51

Re: [net-next PATCH 4/4] net: sched: create hardware only classifier filter

On 16-02-25 05:14 AM, Jamal Hadi Salim wrote: > On 16-02-24 03:47 AM, Jiri Pirko wrote: >> Tue, Feb 23, 2016 at 08:03:56PM CET, john.fastab...@gmail.com wrote: >>> If users want to run filters specifically in hardware without software >>> running the classifiers we need to use a special handler

Re: [PATCH net-next 1/3] perf: generalize perf_callchain

On 2/25/16 8:47 AM, Peter Zijlstra wrote: On Wed, Feb 17, 2016 at 07:58:57PM -0800, Alexei Starovoitov wrote: +static inline int perf_callchain_store(struct perf_callchain_entry *entry, u64 ip) { + if (entry->nr < PERF_MAX_STACK_DEPTH) { entry->ip[entry->nr++] = ip; +

Re: [PATCH net-next 01/10] bnxt_en: Improve bnxt_vf_update_mac().

On Thu, Feb 25, 2016 at 9:09 AM, David Miller wrote: > From: Michael Chan > Date: Thu, 25 Feb 2016 08:57:01 -0800 > >> We are not registering an invalid MAC address. We are just storing it in the >> driver's VF data structure. There are 2 cases:

Re: [PATCH net-next 01/10] bnxt_en: Improve bnxt_vf_update_mac().

From: Michael Chan Date: Thu, 25 Feb 2016 08:57:01 -0800 > We are not registering an invalid MAC address. We are just storing it in the > driver's VF data structure. There are 2 cases: > > 1. VF comes up and the MAC address from firmware is 0. The VF will >

Re: [PATCH v17 0/9] Compile-time stack metadata validation

On Thu, Feb 25, 2016 at 09:02:04AM +0100, Ingo Molnar wrote: > > * Josh Poimboeuf wrote: > > > This is v17 of the compile-time stack metadata validation patch set. > > > > It's based on tip:x86/debug. However, note that when run against that > > branch it will give a lot

Re: [PATCH net-next 01/10] bnxt_en: Improve bnxt_vf_update_mac().

On Thu, Feb 25, 2016 at 8:31 AM, David Miller wrote: > From: Michael Chan > Date: Thu, 25 Feb 2016 03:19:25 -0500 > >> From: Jeffrey Huang >> >> Allow the VF to setup its own MAC address if the PF has not administratively >>

Re: [PATCH net-next 2/3] bpf: introduce BPF_MAP_TYPE_STACK_TRACE

On Thu, Feb 25, 2016 at 08:42:40AM -0800, Alexei Starovoitov wrote: > On 2/25/16 6:23 AM, Peter Zijlstra wrote: > >>+ id = hash & (smap->n_buckets - 1); > >Its not at all clear where the corresponding rcu_read_lock() is at. > > > >>>+ bucket = rcu_dereference(smap->buckets[id]); > > bpf

Re: [PATCH net-next V2 00/16] net: fec: cleanup and fixes

On Thu, 25 Feb 2016 09:05:34 -0700 Troy Kisky wrote: > On 2/24/2016 7:52 PM, Joshua Clayton wrote: > > Hello Troy, > > I'm replying here instead of to a particular commit because several > > of the commit messages seem inadequate. > > > > The first line summaries

Re: [PATCH net-next 1/3] perf: generalize perf_callchain

On Thu, Feb 25, 2016 at 08:37:34AM -0800, Alexei Starovoitov wrote: > On 2/25/16 6:18 AM, Peter Zijlstra wrote: > >> arch/x86/include/asm/stacktrace.h | 2 +- > >> arch/x86/kernel/cpu/perf_event.c | 4 ++-- > >> arch/x86/kernel/dumpstack.c | 6 -- > >> arch/x86/kernel/stacktrace.c

Re: [PATCH net-next 1/3] perf: generalize perf_callchain

On Wed, Feb 17, 2016 at 07:58:57PM -0800, Alexei Starovoitov wrote: > +static inline int perf_callchain_store(struct perf_callchain_entry *entry, > u64 ip) > { > + if (entry->nr < PERF_MAX_STACK_DEPTH) { > entry->ip[entry->nr++] = ip; > + return 0; > + } else {

Re: [PATCH net-next 1/3] perf: generalize perf_callchain

On Thu, Feb 25, 2016 at 08:37:34AM -0800, Alexei Starovoitov wrote: > On 2/25/16 6:18 AM, Peter Zijlstra wrote: > >On Wed, Feb 17, 2016 at 07:58:57PM -0800, Alexei Starovoitov wrote: > >>. avoid walking the stack when there is no room left in the buffer > >>. generalize get_perf_callchain() to be

Re: [PATCH net-next 1/3] perf: generalize perf_callchain

On 2/25/16 6:18 AM, Peter Zijlstra wrote: On Wed, Feb 17, 2016 at 07:58:57PM -0800, Alexei Starovoitov wrote: . avoid walking the stack when there is no room left in the buffer . generalize get_perf_callchain() to be called from bpf helper If it does two things it should be two patches.

Re: [PATCH] ipv4: in new netns initialize sysctls in net.ipv4.conf.* with defaults

From: Nicolas Dichtel Date: Thu, 25 Feb 2016 15:20:48 +0100 > Le 24/02/2016 23:05, Eric W. Biederman a écrit : > [snip] >> In the general case the current behavior is random and not something >> applications can count on, and we would do well to fix it so it is >> less

Re: [PATCH net-next 0/3] bpf_get_stackid() and stack_trace map

From: Peter Zijlstra Date: Thu, 25 Feb 2016 15:24:54 +0100 > On Sat, Feb 20, 2016 at 12:25:05AM -0500, David Miller wrote: >> From: Alexei Starovoitov >> Date: Wed, 17 Feb 2016 19:58:56 -0800 >> >> > This patch set introduces new map type to store stack

Re: [PATCH net-next 2/3] bpf: introduce BPF_MAP_TYPE_STACK_TRACE

On 2/25/16 6:23 AM, Peter Zijlstra wrote: + id = hash & (smap->n_buckets - 1); Its not at all clear where the corresponding rcu_read_lock() is at. >+ bucket = rcu_dereference(smap->buckets[id]); bpf programs of all types are always executing under rcu_read_lock(). This is

Re: [PATCH net] bpf: fix csum setting for bpf_set_tunnel_key

From: Daniel Borkmann Date: Thu, 25 Feb 2016 11:45:47 +0100 > On 02/24/2016 10:24 PM, David Miller wrote: > [...] >> Applied to 'net', thanks Daniel. > > Thanks! > > Btw, in case net gets merged into net-next, there will be a minor > merge conflict due to stackid flags

Re: [net-next PATCH 3/4] net: sched: cls_u32 add bit to specify software only rules

On 16-02-25 05:19 AM, Jamal Hadi Salim wrote: > On 16-02-24 11:09 PM, John Fastabend wrote: >> On 16-02-24 01:29 AM, Jiri Benc wrote: >>> On Wed, 24 Feb 2016 00:55:55 -0800, John Fastabend wrote: The flags however likely stays with with TCA_U32_FLAGS until there is some better way to

[PATCH net-next 02/10] net/mlx5e: Placement changed for carrier state updates

From: Tariq Toukan More proper to declare carrier state UP only after the channels are ready for traffic. Signed-off-by: Tariq Toukan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_main.c |4 ++--

[PATCH net-next 10/10] net/mlx5: Avoid double mapping of io mapped memory

From: Moshe Lazer Device page may be mapped to non-cached(NC) or to write combining(WC). The code before this fix tries to map it both to WC and NC contrary to what stated in Intel's software developer manual. Fixes: 88a85f99e51f ('TX latency optimization to save DMA

[PATCH net-next 09/10] net/mlx5: Fix global UAR mapping

From: Moshe Lazer Remove the global WC mapping of the total UARs since UAR mapping should be decided per UAR. Fixes: 88a85f99e51f ('TX latency optimization to save DMA reads') Signed-off-by: Moshe Lazer Reviewed-by: Achiad Shochat

[PATCH net-next 03/10] net/mlx5e: Changed naming convention of tx queues in ethtool stats

From: Tariq Toukan Instead of the pair (channel, tc), we now use a single number that goes over all tx queues of a TC, for all TCs. Signed-off-by: Tariq Toukan Signed-off-by: Saeed Mahameed ---

[PATCH net-next 07/10] net/mlx5e: Don't modify CQ before it was created

From: Gal Pressman Calling mlx5e_set_coalesce while the interface is down will result in modifying CQs that don't exist. Fixes: f62b8bb8f2d3 ('net/mlx5: Extend mlx5_core to support ConnectX-4 Ethernet functionality') Signed-off-by: Gal Pressman

[PATCH net-next 06/10] net/mlx5e: Don't try to modify CQ moderation if it is not supported

From: Gal Pressman If CQ moderation is not supported by the device, print a warning on netdevice load, and return error when trying to modify/query cq moderation via ethtool. Fixes: f62b8bb8f2d3 ('net/mlx5: Extend mlx5_core to support ConnectX-4 Ethernet functionality')

[PATCH net-next 08/10] net/mlx5: Make command timeout way shorter

From: Or Gerlitz The command timeout is terribly long, whole two hours. Make it 60s so if things do go wrong, the user gets feedback in relatively short time, so they can take corrective actions and/or investigate using tools and such. Fixes: e126ba97dba9 ('mlx5: Add

[PATCH net-next 04/10] net/mlx5e: Move common case counters within sq_stats struct

From: Tariq Toukan For data cache locality considerations, we moved the nop and csum_offload_inner within sq_stats struct as they are more commonly accessed in xmit path. Signed-off-by: Tariq Toukan Signed-off-by: Saeed Mahameed

[PATCH net-next 00/10] mlx5 driver updates

Hi Dave, This series includes some bug fixes and updates for the mlx5 core and ethernet driver. >From Gal, two fixes that protects the update CQ moderation flows when it is not allowed. >From Moshe, two fixes for the core and ethernet driver in non-cached(NC) and write combining(WC) buffers

[PATCH net-next 01/10] net/mlx5e: Replace async events spinlock with synchronize_irq()

From: Tariq Toukan We only need to flush the irq handler to make sure it does not queue a work into the global work queue after we start to flush it. So using synchronize_irq() is more appropriate than a spin lock. Signed-off-by: Tariq Toukan

[PATCH net-next 05/10] net/mlx5e: Set drop RQ's necessary parameters only

From: Tariq Toukan By its role, there is no need to set all the other parameters for the drop RQ. Signed-off-by: Tariq Toukan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 12 ++--

Re: [PATCH net-next 04/10] bnxt_en: Refactor bnxt_hwrm_set_coal().

From: Michael Chan Date: Thu, 25 Feb 2016 03:19:28 -0500 > Add a function to set all the coalescing parameters. The function can > be used later to set both rx and tx coalescing parameters. > > Signed-off-by: Michael Chan > --- >

Re: [PATCH net-next 01/10] bnxt_en: Improve bnxt_vf_update_mac().

From: Michael Chan Date: Thu, 25 Feb 2016 03:19:25 -0500 > From: Jeffrey Huang > > Allow the VF to setup its own MAC address if the PF has not administratively > set it for the VF. If the perm_mac_address returned by firmware is all > zeros,

Re: [PATCH net-next RFC] rtnetlink: add new RTM_GETSTATS to dump link stats

From: roopa Date: Wed, 24 Feb 2016 21:25:50 -0800 > I did go back and forth on the attribute vs mask. > cosmetic but, i guess i did not feel good about having to redefine every > attribute again > for the bitmap filter ...and i anticipate the list of stat attributes

Re: [PATCH net-next V2 00/16] net: fec: cleanup and fixes

On 2/24/2016 7:52 PM, Joshua Clayton wrote: > Hello Troy, > I'm replying here instead of to a particular commit because several of > the commit messages seem inadequate. > > The first line summaries all look good. > > The descriptions should each also include the "user visible impact" of > the

Re: [PATCH net-next V2 00/16] net: fec: cleanup and fixes

On 2/25/2016 1:39 AM, Holger Schurig wrote: > Hi Troy, > > what is the general aim of your patches? Stability? Speed? Cleanup? > 1. Stability 2. performance 3. easier to read 4. more debug info The 2nd goal is very hard to measure. It seems function alignment changes swamp most any other

Re: [PATCH v3] net: jme: fix suspend/resume on JMC260

On Thu, Feb 25, 2016 at 1:58 AM, David Miller wrote: > From: Diego Viola > Date: Tue, 23 Feb 2016 12:04:04 -0300 > >> The JMC260 network card fails to suspend/resume because the call to >> jme_start_irq() was too early, moving the call to

Re: [PATCH net-next v2 1/5] netdev: introduce ndo_set_rx_headroom

On Thu, 2016-02-25 at 17:21 +0300, Sergei Shtylyov wrote: > Hello. > > On 2/25/2016 2:09 PM, Paolo Abeni wrote: > > > This method allows the controlling device (i.e. the bridge) to specify > > additional headroom to be allocated for skb head on frame reception. > > > > Signed-off-by: Paolo Abeni

Re: RFC: blueprint for proxy PHY module

> Question is how to connect the userspace? I think that corresponding > file in /sys would do the job, but I am not sure whether it is the > proper way of doing that. (?) A historian might be able to help you. Going back to the days of 10MBps Ethernet, cards used to have a few different ports,

Re: RFC: blueprint for proxy PHY module

On Thu, Feb 25, 2016 at 02:26:00PM +0100, Tomas Hlavacek wrote: > Hello! > > We have a board with a MAC in SoC that has SGMII connected through a > signal switch to either 10/100/1000 wired PHY or to a SFP cage. The > signal switch is set to pass signal to the wired PHY by default. > When the

Re: [PATCH 2/4] net: ipv4: Use y2038 safe functions and data structures

On Wednesday 24 February 2016 23:07:09 Deepa Dinamani wrote: > ICMP timestamp messages and IP source route options require > timestamps to be in milliseconds modulo 24 hours from > midnight UT format. > > Timestamps are calculated from realtime by making a call > to getnstimeofday() which uses

Re: [PATCH 3/4] net: ipv4: tcp_probe: Replace timespec with timespec64

On Wednesday 24 February 2016 23:07:10 Deepa Dinamani wrote: > TCP probe log timestamps use struct timespec which is > not y2038 safe. Even though timespec might be good enough here > as it is used to represent delta time, the plan is to get rid > of all uses of timespec in the kernel. > Replace

Re: [Y2038] [PATCH 4/4] net: sctp: Convert log timestamps to be y2038 safe

On Wednesday 24 February 2016 23:07:11 Deepa Dinamani wrote: > SCTP probe log timestamps use struct timespec which is > not y2038 safe. > Use struct timespec64 which is 2038 safe instead. > > Use monotonic time instead of real time as only time > differences are logged. > > Signed-off-by: Deepa

Re: [Y2038] [PATCH 1/4] kernel: time: Add current_nw_timestamp() for network timestamps

On Wednesday 24 February 2016 23:07:08 Deepa Dinamani wrote: > ICMP timestamp messages and IP source route options require > timestamps to be in milliseconds modulo 24 hours from > midnight UTC format. > > Add a time function to support this. The function returns the > required timestamp in

Re: [PATCH net-next 0/3] bpf_get_stackid() and stack_trace map

On Sat, Feb 20, 2016 at 12:25:05AM -0500, David Miller wrote: > From: Alexei Starovoitov > Date: Wed, 17 Feb 2016 19:58:56 -0800 > > > This patch set introduces new map type to store stack traces and > > corresponding bpf_get_stackid() helper. > ... > > Series applied, thanks

Re: [PATCH net-next 2/3] bpf: introduce BPF_MAP_TYPE_STACK_TRACE

On Wed, Feb 17, 2016 at 07:58:58PM -0800, Alexei Starovoitov wrote: > +static u64 bpf_get_stackid(u64 r1, u64 r2, u64 flags, u64 r4, u64 r5) > +{ > + struct pt_regs *regs = (struct pt_regs *) (long) r1; > + struct bpf_map *map = (struct bpf_map *) (long) r2; > + struct bpf_stack_map

Re: [PATCH net-next v2 1/5] netdev: introduce ndo_set_rx_headroom

Hello. On 2/25/2016 2:09 PM, Paolo Abeni wrote: This method allows the controlling device (i.e. the bridge) to specify additional headroom to be allocated for skb head on frame reception. Signed-off-by: Paolo Abeni --- v1 -> v2: - fixed netdev_get_fwd_headroom() -

Re: [PATCH] ipv4: in new netns initialize sysctls in net.ipv4.conf.* with defaults

Le 24/02/2016 23:05, Eric W. Biederman a écrit : [snip] In the general case the current behavior is random and not something applications can count on, and we would do well to fix it so it is less random. In particular consider the case of an application in a non-initial network namespace

Re: [PATCH net-next 1/3] perf: generalize perf_callchain

On Wed, Feb 17, 2016 at 07:58:57PM -0800, Alexei Starovoitov wrote: > . avoid walking the stack when there is no room left in the buffer > . generalize get_perf_callchain() to be called from bpf helper If it does two things it should be two patches. > Signed-off-by: Alexei Starovoitov

Re: [PATCH 4/4] net: sctp: Convert log timestamps to be y2038 safe

On Wed, Feb 24, 2016 at 11:07:11PM -0800, Deepa Dinamani wrote: > SCTP probe log timestamps use struct timespec which is > not y2038 safe. > Use struct timespec64 which is 2038 safe instead. > > Use monotonic time instead of real time as only time > differences are logged. > > Signed-off-by:

Re: [PATCH v2 net] bonding: don't use stale speed and duplex information

zhuyj wrote: [...] >I delved into the source code and Emil's tests. I think that the problem >that this patch expects to fix occurs very unusually. > >Do you agree with me? > >If so, maybe the following patch can reduce the performance loss. >Please comment on it. Thanks a

RFC: blueprint for proxy PHY module

Hello! We have a board with a MAC in SoC that has SGMII connected through a signal switch to either 10/100/1000 wired PHY or to a SFP cage. The signal switch is set to pass signal to the wired PHY by default. When the MOD-DEF0 pin from SFP cage is low (which acts as a transceiver-present

Regression: netlink fail (triggered by iw) removes extra wlan (phy) interface

Hi, After updating kernel in OpenWrt from 4.1.6 to 4.1.10 I noticed that if "iw" command fails (which happens very rarely) my wlan0-1 interface disappears. To trigger this problem easily I'm using this trivial script: while [ 1 ] do iw phy phy0 interface add mon0 type monitor

Re: [net-next PATCH 3/4] net: sched: cls_u32 add bit to specify software only rules

On 16-02-24 11:09 PM, John Fastabend wrote: On 16-02-24 01:29 AM, Jiri Benc wrote: On Wed, 24 Feb 2016 00:55:55 -0800, John Fastabend wrote: The flags however likely stays with with TCA_U32_FLAGS until there is some better way to group common attributes in 'tc' framework. That's pretty bad,

Re: [net-next PATCH 4/4] net: sched: create hardware only classifier filter

On 16-02-24 03:47 AM, Jiri Pirko wrote: Tue, Feb 23, 2016 at 08:03:56PM CET, john.fastab...@gmail.com wrote: If users want to run filters specifically in hardware without software running the classifiers we need to use a special handler for this. By using a new classifier list we are able to

Re: [net-next PATCH 3/4] net: sched: cls_u32 add bit to specify software only rules

On 16-02-24 11:04 PM, John Fastabend wrote: On 16-02-24 05:31 AM, Jamal Hadi Salim wrote: I think this is absolutely necessary not only for performance of reporting the rules back to software but if we don't do it generically the driver will have to do it anyways because doing the inverse

Re: [net-next PATCH 0/5] net_sched: Add support for IFE action

On 16-02-24 12:58 PM, Daniel Borkmann wrote: On 02/24/2016 01:49 PM, Jamal Hadi Salim wrote: Yes, a bit of that ++. I am between two worlds: There are people who do user space packet processing that claim they do so because they can quickly prototype without compiling the kernel. My goal is

Re: [net-next PATCH 0/5] net_sched: Add support for IFE action

On 16-02-24 12:48 PM, Daniel Borkmann wrote: On 02/24/2016 01:49 PM, Jamal Hadi Salim wrote: Drivers do set the hash. My use case is slightly different. I have a NIC which has an embedded cavium processor. This thing strips off the TLV and uses the hash to select the host MSI. Only thing we

Re: [net-next PATCH v2 1/5] introduce IFE action

On 16-02-24 12:37 PM, Daniel Borkmann wrote: On 02/23/2016 01:49 PM, Jamal Hadi Salim wrote: From: Jamal Hadi Salim [...] +static const struct nla_policy ife_policy[TCA_IFE_MAX + 1] = { +[TCA_IFE_PARMS] = {.len = sizeof(struct tc_ife)}, +[TCA_IFE_DMAC] = {.type =

Re: [Patch net-next v2 2/2] net_sched: add network namespace support for tc actions

On 16-02-24 12:21 PM, Cong Wang wrote: On Wed, Feb 24, 2016 at 5:19 AM, Jamal Hadi Salim wrote: [..] Did you mean something else or am i missing the obvious? I am looking at the call sites for tcf_action_destroy(): -tc_dump_tfilter() has access to *net -tcf_exts_change() -

[iproute PATCH] iprule: Align help text with man page synopsis

The help text was misleading: One could think it is possible to list rules by selector, which would be nice but isn't. This change also clarifies that 'ip rule' defaults to 'list' if no further arguments are given. Signed-off-by: Phil Sutter --- ip/iprule.c | 5 +++-- 1 file

[PATCH net-next v2 3/5] ovs: propagate per dp max headroom to all vports

This patch implements bookkeeping support to compute the maximum headroom for all the devices in each datapath. When said value changes, the underlying devs are notified via the ndo_set_rx_headroom method. This also increases the internal vports xmit performance. Signed-off-by: Paolo Abeni

[PATCH net-next v2 4/5] net/tun: implement ndo_set_rx_headroom

ndo_set_rx_headroom controls the align value used by tun devices to allocate skbs on frame reception. When the xmit device adds a large encapsulation, this avoids an skb head reallocation on forwarding. The measured improvement when forwarding towards a vxlan dev with frame size below the egress

[PATCH net-next v2 5/5] veth: implement ndo_set_rx_headroom

The rx headroom for veth dev is the peer device needed_headroom. Avoid ping-pong updates setting the private flag IFF_PHONY_HEADROOM. This avoids skb head reallocation when forwarding from a veth dev towards a device adding some kind of encapsulation. When transmitting frames below the MTU size

[PATCH net-next v2 1/5] netdev: introduce ndo_set_rx_headroom

This method allows the controlling device (i.e. the bridge) to specify additional headroom to be allocated for skb head on frame reception. Signed-off-by: Paolo Abeni --- v1 -> v2: - fixed netdev_get_fwd_headroom() - added netdev_set_rx_headroom() and

[PATCH net-next v2 0/5] bridge/ovs: avoid skb head copy on frame forwarding

Currently, while when an OVS or Linux bridge is used to forward frames towards some tunnel device, a skb_head_copy() may occur if the ingress device do not provide enough headroom for the tx encapsulation. This patch series tries to address the issue implementing a new ndo operation to allow the

[PATCH net-next v2 2/5] bridge: notify ensabled devices of headroom changes

On bridge needed_headroom changes, the enslaved devices are notified via the ndo_set_rx_headroom method Signed-off-by: Paolo Abeni --- v1 -> v2: - use netdev_set_rx_headroom() to update the br ports headroom - reset the dev rx headroom upon dev removal from bridge ---

Re: [PATCH net-next 4/5] net/tun: implement ndo_set_rx_headroom

On Tue, 2016-02-23 at 13:53 +0100, Paolo Abeni wrote: > ndo_set_rx_headroom controls the align value used by tun devices to > allocate skbs on frame reception. > When the xmit device adds a large encapsulation, this avoids an skb > head reallocation on forwarding. > > The measured improvement

Re: [PATCH net] bpf: fix csum setting for bpf_set_tunnel_key

On 02/24/2016 10:24 PM, David Miller wrote: [...] Applied to 'net', thanks Daniel. Thanks! Btw, in case net gets merged into net-next, there will be a minor merge conflict due to stackid flags present in net-next. This can be easily resolved by keeping both chunks. For example, result after

Re: [net-next RFC 0/4] SO_BINDTOSUBNET

> On 24 Feb 2016, at 05:06, Tom Herbert wrote: > > On Tue, Feb 23, 2016 at 7:27 AM, Gilberto Bertin > wrote: >> This series introduces support for the SO_BINDTOSUBNET socket option, which >> allows a listener socket to bind to a subnet instead

Re: [PATCH 1/4] kernel: time: Add current_nw_timestamp() for network timestamps

>> include/linux/ip.h | 2 ++ >> include/linux/time64.h | 3 +++ >> kernel/time/time.c | 26 ++ >> 3 files changed, 31 insertions(+) >> >> diff --git a/include/linux/ip.h b/include/linux/ip.h >> index 492bc65..edf923e 100644 >> --- a/include/linux/ip.h >> +++

Re: custom ioctl-based interface to control LED in networking (was Re: [PATCHv2 09/10] rfkill: Userspace control for airplane mode)

On Wed 2016-02-24 14:31:33, Johannes Berg wrote: > On Wed, 2016-02-24 at 13:14 +0100, Pavel Machek wrote: > >  > > Why would it need to? It could look at default triggers for the led > > if it really wanted to. > > And then it needs to change them; if anything goes wrong error recovery > is

Re: [PATCH 1/4] kernel: time: Add current_nw_timestamp() for network timestamps

Hi, Deepa Dinamani wrote: > ICMP timestamp messages and IP source route options require > timestamps to be in milliseconds modulo 24 hours from > midnight UTC format. > > Add a time function to support this. The function returns the > required timestamp in network byte order. > > The function

Re: [PATCH v2 2/4] Documentation: Bindings: Add STM32 DWMAC glue

Hi Joachim, 2016-02-23 23:37 GMT+01:00 Joachim Eastwood : > Hi Alexandre, > > You should copy 'devicet...@vger.kernel.org' on bindings doc. Adding cc here. > > On 23 February 2016 at 16:10, Alexandre TORGUE > wrote: >> Signed-off-by: Alexandre

Re: [PATCH net-next V2 00/16] net: fec: cleanup and fixes

Hi Troy, what is the general aim of your patches? Stability? Speed? Cleanup?

Re: [PATCH v2 net] bonding: don't use stale speed and duplex information

On 02/09/2016 04:10 AM, Jay Vosburgh wrote: There is presently a race condition between the bonding periodic link monitor and the updating of a slave's speed and duplex. The former occurs on a periodic basis, and the latter in response to a driver's calling of netif_carrier_on.

[PATCH net-next 04/10] bnxt_en: Refactor bnxt_hwrm_set_coal().

Add a function to set all the coalescing parameters. The function can be used later to set both rx and tx coalescing parameters. Signed-off-by: Michael Chan --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 26 +- 1 file changed, 17 insertions(+), 9

[PATCH net-next 01/10] bnxt_en: Improve bnxt_vf_update_mac().

From: Jeffrey Huang Allow the VF to setup its own MAC address if the PF has not administratively set it for the VF. If the perm_mac_address returned by firmware is all zeros, that means the PF has not set up the MAC address for the VF and we should store it. This will

[PATCH net-next 08/10] bnxt_en: Add installed-package firmware version reporting via Ethtool GDRVINFO

From: Rob Swindell For everything to fit, we remove the PHY microcode version and replace it with the firmware package version in the fw_version string. Signed-off-by: Rob Swindell Signed-off-by: Michael Chan ---

[PATCH net-next 02/10] bnxt_en: Send PF driver unload notification to all VFs.

From: Jeffrey Huang During remove_one() when SRIOV is enabled, the PF driver should broadcast PF driver unload notification to all VFs that are attached to VMs. Upon receiving the PF driver unload notification, the VF driver should print a warning message to message log.

[PATCH net-next 10/10] bnxt_en: Add hwrm_send_message_silent().

This is used to send NVM_FIND_DIR_ENTRY messages which can return error if the entry is not found. This is normal and the error message will cause unnecessary alarm, so silence it. Signed-off-by: Michael Chan --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 11

[PATCH net-next 09/10] bnxt_en: Refactor _hwrm_send_message().

Add a new function bnxt_do_send_msg() to do essentially the same thing with an additional paramter to silence error response messages. All current callers will set silent to false. Signed-off-by: Michael Chan --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 14 +-

[PATCH net-next 07/10] bnxt_en: Fix dmesg log firmware error messages.

Use appropriate firmware request header structure to prepare the firmware messages. This avoids the unnecessary conversion of the fields to 32-bit fields. Add appropriate endian conversion when printing out the message fields in dmesg so that they appear correct in the log. Reported-by: Rob

[PATCH net-next 03/10] bnxt_en: Store irq coalescing timer values in micro seconds.

Don't convert these to internal hardware tick values before storing them. This avoids the confusion of ethtool -c returning slightly different values than the ones set using ethtool -C when we convert hardware tick values back to micro seconds. Add better comments for the hardware settings.

[PATCH net-next 06/10] bnxt_en: Use firmware provided message timeout value.

Before this patch, we used a hardcoded value of 500 msec as the default value for firmware message response timeout. For better portability with future hardware or debug platforms, use the value provided by firmware in the first response and store it for all susequent messages. Redefine the

[PATCH net-next 00/10] bnxt_en: updates for net-next.

Miscellaneous updates covering SRIOV, IRQ coalescing, firmware logging and package version for net-next. Thanks. Jeffrey Huang (2): bnxt_en: Improve bnxt_vf_update_mac(). bnxt_en: Send PF driver unload notification to all VFs. Michael Chan (8): bnxt_en: Store irq coalescing timer values

[PATCH net-next 05/10] bnxt_en: Add coalescing support for tx rings.

When tx and rx rings don't share the same completion ring, tx coalescing parameters can be set differently from the rx coalescing parameters. Otherwise, use rx coalescing parameters on shared completion rings. Adjust rx coalescing default values to lower interrupt rate. Signed-off-by: Michael

Re: [PATCH v17 0/9] Compile-time stack metadata validation

* Josh Poimboeuf wrote: > This is v17 of the compile-time stack metadata validation patch set. > > It's based on tip:x86/debug. However, note that when run against that > branch it will give a lot of warnings: > > objtool: arch/x86/ia32/sys_ia32.o: __ex_table size not

<    1   2