Re: [PATCH iproute2-next] devlink: report cell size

On Sun, Feb 03, 2019 at 05:39:19PM -0800, Jakub Kicinski wrote: > diff --git a/man/man8/devlink-sb.8 b/man/man8/devlink-sb.8 > index 1882833a3fa7..767c932dd4c1 100644 > --- a/man/man8/devlink-sb.8 > +++ b/man/man8/devlink-sb.8 > @@ -128,6 +128,16 @@ Behaviour of this argument it the same for every

Re: [PATCH iproute2-next v2] devlink: add info subcommand

Mon, Feb 04, 2019 at 01:54:56AM CET, jakub.kicin...@netronome.com wrote: >Add support for reading the device serial number, driver name >and various versions. Example: > >$ devlink dev info pci/:05:00.0 >pci/:05:00.0: > driver nfp > serial_number 16240145 > versions: >fixed: >

Re: [PATCH net-next v7 0/8] devlink: Add configuration parameters support for devlink_port

On Wed, Jan 23, 2019 at 3:48 AM Jakub Kicinski wrote: > > On Fri, 18 Jan 2019 15:33:19 +0100, Michal Kubecek wrote: > > On Fri, Jan 18, 2019 at 12:39:37PM +0530, Vasundhara Volam wrote: > > > There is difference of opinion on adding WOL parameter to devlink, between > > > Jakub Kicinski and Michae

Re: [PATCH net-next V2 1/1] openvswitch: Declare ovs key structures using macros

On Sun, Feb 3, 2019 at 1:12 AM Eli Britstein wrote: > > Declare ovs key structures using macros as a pre-step towards to > enable retrieving fields information, as a work done in proposed > commit in the OVS tree https://patchwork.ozlabs.org/patch/1023406/ > ("odp-util: Do not rewrite fields with

Re: general protection fault in rose_send_frame

syzbot has found a reproducer for the following crash on: HEAD commit:9fb20801dab4 net: Fix ip_mc_{dec,inc}_group allocation con.. git tree: net-next console output: https://syzkaller.appspot.com/x/log.txt?x=17530a0f40 kernel config: https://syzkaller.appspot.com/x/.config?x=33ad02

Re: linux-next: build failure after merge of the net-next tree

From: Stephen Rothwell Date: Mon, 4 Feb 2019 15:34:46 +1100 > From: Stephen Rothwell > Date: Mon, 4 Feb 2019 15:24:11 +1100 > Subject: [PATCH] socket: fix for Add SO_TIMESTAMP[NS]_NEW > > Fixes: 887feae36aee ("socket: Add SO_TIMESTAMP[NS]_NEW") > Signed-off-by: Stephen Rothwell Thanks for fix

linux-next: build failure after merge of the net-next tree

Hi all, After merging the net-next tree, today's linux-next build (sparc64 defconfig) failed like this: /home/sfr/next/next/net/core/sock.c: In function 'sock_setsockopt': /home/sfr/next/next/net/core/sock.c:1146:2: error: duplicate case value case SO_BINDTOIFINDEX: ^~~~ /home/sfr/next/next/n

Re: linux-next: build warning after merge of the net-next tree

Hi Dave, On Sun, 03 Feb 2019 20:26:06 -0800 (PST) David Miller wrote: > > Thanks, I just pushed the following: > > > [PATCH] net: Fix fall through warning in y2038 tstamp changes. > > net/core/sock.c: In function 'sock_setsockopt': > net/core/sock.c:914:3: warning: this st

Re: [PATCH] netdevice.h: Add __cold to netdev_ logging functions

From: Joe Perches Date: Sat, 02 Feb 2019 19:47:25 -0800 > Add __cold to the netdev_ logging functions similar to > the use of __cold in the generic printk function. > > Using __cold moves all the netdev_ logging functions > out-of-line possibly improving code locality and runtime > performance.

Re: linux-next: build warning after merge of the net-next tree

From: Stephen Rothwell Date: Mon, 4 Feb 2019 10:44:27 +1100 > After merging the net-next tree, today's linux-next build (arm > multi_v7_defconfig) produced this warning: > > net/core/sock.c: In function 'sock_setsockopt': > net/core/sock.c:914:3: warning: this statement may fall through > [-Wim

Re: [PATCH net] virtio_net: Account for tx bytes and packets on sending xdp_frames

From: Toshiaki Makita Date: Thu, 31 Jan 2019 20:40:30 +0900 > Previously virtnet_xdp_xmit() did not account for device tx counters, > which caused confusions. > To be consistent with SKBs, account them on freeing xdp_frames. > > Reported-by: David Ahern > Signed-off-by: Toshiaki Makita Applie

Re: [PATCH] bpfilter: remove extra header search paths for bpfilter_umh

From: Masahiro Yamada Date: Thu, 31 Jan 2019 12:15:35 +0900 > Currently, the header search paths -Itools/include and > -Itools/include/uapi are not used. Let's drop the unused code. > > We can remove -I. too by fixing up one C file. > > Signed-off-by: Masahiro Yamada I guess I'm ok with this,

Re: [Patch net-next] net_sched: add performance counters for basic filter

On 02/03/2019 11:35 AM, Cong Wang wrote: > On Sun, Feb 3, 2019 at 10:35 AM Eric Dumazet wrote: >>> + for_each_possible_cpu(cpu) { >>> + struct tc_basic_pcnt *pf = per_cpu_ptr(f->pf, cpu); >>> + >>> + gpf.rcnt += pf->rcnt; >>> + gpf.rhit += pf->rhit; >>> +

[PATCH iproute2-next] devlink: report cell size

Print the value of DEVLINK_ATTR_SB_POOL_CELL_SIZE, if reported. Example: pci/:82:00.0: sb 1 pool 0 type egress size 40945664 thtype static cell_size 2048 sb 2 pool 0 type egress size 258867200 thtype static cell_size 10240 ... Signed-off-by: Jakub Kicinski Reviewed-by: Quentin Monnet --

[PATCH iproute2-next v2] devlink: add info subcommand

Add support for reading the device serial number, driver name and various versions. Example: $ devlink dev info pci/:05:00.0 pci/:05:00.0: driver nfp serial_number 16240145 versions: fixed: board.id AMDA0099-0001 board.rev 07 board.vendor SMA board.model

linux-next: build warning after merge of the net-next tree

Hi all, After merging the net-next tree, today's linux-next build (arm multi_v7_defconfig) produced this warning: net/core/sock.c: In function 'sock_setsockopt': net/core/sock.c:914:3: warning: this statement may fall through [-Wimplicit-fallthrough=] sock_set_flag(sk, SOCK_TSTAMP_NEW); ^~

Re: [PATCH bpf-next] bpf: support SO_DEBUG in bpf_setsockopt()

On 2/3/19, 12:15 AM, "Yafang Shao" wrote: Then we can enable/disable socket debugging without modifying user code. That is more convenient for debugging. Signed-off-by: Yafang Shao Acked-by: Lawrence Brakmo --- include/net/sock.h | 8 net/core/filter.c

Re: [PATCH bpf-next] bpf: support SO_DEBUG in bpf_setsockopt()

On Sun, Feb 3, 2019 at 12:18 AM Yafang Shao wrote: > > Then we can enable/disable socket debugging without modifying user code. > That is more convenient for debugging. > > Signed-off-by: Yafang Shao Acked-by: Yonghong Song > --- > include/net/sock.h | 8 > net/core/filter.c | 3 +++

Re: [PATCHv3 net] sctp: check and update stream->out_curr when allocating stream_out

From: Xin Long Date: Mon, 4 Feb 2019 03:27:58 +0800 > Now when using stream reconfig to add out streams, stream->out > will get re-allocated, and all old streams' information will > be copied to the new ones and the old ones will be freed. > > So without stream->out_curr updated, next time when

Re: [PATCH 0/4 v2 net-next] net: phy: aquantia: number of improvements

From: Heiner Kallweit Date: Sun, 3 Feb 2019 21:15:19 +0100 > This patch series is based on work from Andrew. I adjusted and added > certain parts. The series improves few aspects of driver, no functional > change intended. > > v2: > - add my SoB to patch 1 > - leave kernel.h in in patch 2 I thi

Re: [PATCHv3 net] sctp: check and update stream->out_curr when allocating stream_out

On Mon, Feb 04, 2019 at 03:27:58AM +0800, Xin Long wrote: > Now when using stream reconfig to add out streams, stream->out > will get re-allocated, and all old streams' information will > be copied to the new ones and the old ones will be freed. > > So without stream->out_curr updated, next time w

Linux kernel and RFC 7217 support

Hi, I'm testing IPv6 stable-secret identifiers SII under Linux RFC 7217. One important feature of RFC 7217 is to deliver stable IPv6 addresses for server addressing use that will not change in case of new network hardware (changed hardware address). Besides o

[PATCH 4/4 v2 net-next] net: phy: aquantia: replace magic numbers with constants

Replace magic numbers with proper constants. The original patch is from Andrew, I extended / adjusted certain parts: - Use decimal bit numbers. The datasheet uses hex bit numbers 0 .. F. - Order defines from highest to lowest bit numbers - correct some typos - add constant MDIO_AN_TX_VEND_INT_MASK2

[PATCH 2/4 v2 net-next] net: phy: aquantia: remove unneeded includes

Remove unneeded header includes. v2: - leave kernel.h in Signed-off-by: Heiner Kallweit --- drivers/net/phy/aquantia.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/net/phy/aquantia.c b/drivers/net/phy/aquantia.c index 880ec4bba..2b349de1d 100644 --- a/drivers/net/phy/aquantia.c

Re: [v2, 0/9] Add ENETC PTP clock driver

From: David Miller Date: Sun, 03 Feb 2019 11:39:21 -0800 (PST) > From: Yangbo Lu > Date: Sat, 2 Feb 2019 10:56:58 +0800 > >> There is same QorIQ 1588 timer IP block on the new ENETC Ethernet >> controller with eTSEC/DPAA Ethernet controllers. However it's >> different endianness (little-endian

[PATCH 3/4 v2 net-next] net: phy: aquantia: use macro PHY_ID_MATCH_MODEL

Make use of macro PHY_ID_MATCH_MODEL to simplify the code. Signed-off-by: Heiner Kallweit Reviewed-by: Andrew Lunn --- drivers/net/phy/aquantia.c | 30 -- 1 file changed, 12 insertions(+), 18 deletions(-) diff --git a/drivers/net/phy/aquantia.c b/drivers/net/phy/aqu

[PATCH 1/4 v2 net-next] net: phy: aquantia: Shorten name space prefix to aqr_

From: Andrew Lunn aquantia_ as a name space prefix is rather long, resulting in lots of lines needing wrapping, reducing readability. Use the prefix aqr_ instead, which fits with the vendor naming there devices aqr107, for example. v2: - add SoB from Heiner Signed-off-by: Andrew Lunn Signed-off

[PATCH 0/4 v2 net-next] net: phy: aquantia: number of improvements

This patch series is based on work from Andrew. I adjusted and added certain parts. The series improves few aspects of driver, no functional change intended. v2: - add my SoB to patch 1 - leave kernel.h in in patch 2 Andrew Lunn (1): net: phy: aquantia: Shorten name space prefix to aqr_ Heiner

Re: [PATCH 2/4 net-next] net: phy: aquantia: remove unneeded includes

On 03.02.2019 20:54, Andrew Lunn wrote: > On Sun, Feb 03, 2019 at 08:33:54PM +0100, Heiner Kallweit wrote: >> Remove unneeded header includes. >> >> Signed-off-by: Heiner Kallweit > > >> --- >> drivers/net/phy/aquantia.c | 4 >> 1 file changed, 4 deletions(-) >> >> diff --git a/drivers/net

Re: [PATCH 4/4 net-next] net: phy: aquantia: replace magic numbers with constants

On Sun, Feb 03, 2019 at 08:35:54PM +0100, Heiner Kallweit wrote: > Replace magic numbers with proper constants. The original patch is > from Andrew, I extended / adjusted certain parts: > - Use decimal bit numbers. The datasheet uses hex bit numbers 0 .. F. > - Order defines from highest to lowest

Re: [PATCH 3/4 net-next] net: phy: aquantia: use macro PHY_ID_MATCH_MODEL

On Sun, Feb 03, 2019 at 08:34:37PM +0100, Heiner Kallweit wrote: > Make use of macro PHY_ID_MATCH_MODEL to simplify the code. > > Signed-off-by: Heiner Kallweit Reviewed-by: Andrew Lunn Andrew

Re: [PATCH 2/4 net-next] net: phy: aquantia: remove unneeded includes

On Sun, Feb 03, 2019 at 08:33:54PM +0100, Heiner Kallweit wrote: > Remove unneeded header includes. > > Signed-off-by: Heiner Kallweit > --- > drivers/net/phy/aquantia.c | 4 > 1 file changed, 4 deletions(-) > > diff --git a/drivers/net/phy/aquantia.c b/drivers/net/phy/aquantia.c > index

Re: [PATCH 1/4 net-next] net: phy: aquantia: shorten name space prefix to aqr_

On 03.02.2019 20:38, Andrew Lunn wrote: > On Sun, Feb 03, 2019 at 08:33:13PM +0100, Heiner Kallweit wrote: >> From: Andrew Lunn >> aquantia_ as a name space prefix is rather long, resulting in lots of >> lines needing wrapping, reducing readability. Use the prefix aqr_ >> instead, which fits with

Re: [PATCH net-next] net: Fix ip_mc_{dec,inc}_group allocation context

From: Florian Fainelli Date: Fri, 1 Feb 2019 20:20:52 -0800 > After 4effd28c1245 ("bridge: join all-snoopers multicast address"), I > started seeing the following sleep in atomic warnings: ... > while toggling the bridge's multicast_snooping attribute dynamically. > > Pass a gfp_t down to igmp

Re: [v2, 0/9] Add ENETC PTP clock driver

From: Yangbo Lu Date: Sat, 2 Feb 2019 10:56:58 +0800 > There is same QorIQ 1588 timer IP block on the new ENETC Ethernet > controller with eTSEC/DPAA Ethernet controllers. However it's > different endianness (little-endian) and using PCI driver. > > To support ENETC PTP driver, ptp_qoriq driver

Re: [PATCH 1/4 net-next] net: phy: aquantia: shorten name space prefix to aqr_

On Sun, Feb 03, 2019 at 08:33:13PM +0100, Heiner Kallweit wrote: > From: Andrew Lunn > aquantia_ as a name space prefix is rather long, resulting in lots of > lines needing wrapping, reducing readability. Use the prefix aqr_ > instead, which fits with the vendor naming there devices aqr107, for >

[PATCH 3/4 net-next] net: phy: aquantia: use macro PHY_ID_MATCH_MODEL

Make use of macro PHY_ID_MATCH_MODEL to simplify the code. Signed-off-by: Heiner Kallweit --- drivers/net/phy/aquantia.c | 30 -- 1 file changed, 12 insertions(+), 18 deletions(-) diff --git a/drivers/net/phy/aquantia.c b/drivers/net/phy/aquantia.c index 8a51589a3..1

[PATCH 4/4 net-next] net: phy: aquantia: replace magic numbers with constants

Replace magic numbers with proper constants. The original patch is from Andrew, I extended / adjusted certain parts: - Use decimal bit numbers. The datasheet uses hex bit numbers 0 .. F. - Order defines from highest to lowest bit numbers - correct some typos - add constant MDIO_AN_TX_VEND_INT_MASK2

[PATCH 2/4 net-next] net: phy: aquantia: remove unneeded includes

Remove unneeded header includes. Signed-off-by: Heiner Kallweit --- drivers/net/phy/aquantia.c | 4 1 file changed, 4 deletions(-) diff --git a/drivers/net/phy/aquantia.c b/drivers/net/phy/aquantia.c index 880ec4bba..8a51589a3 100644 --- a/drivers/net/phy/aquantia.c +++ b/drivers/net/phy/a

[PATCH 1/4 net-next] net: phy: aquantia: shorten name space prefix to aqr_

From: Andrew Lunn aquantia_ as a name space prefix is rather long, resulting in lots of lines needing wrapping, reducing readability. Use the prefix aqr_ instead, which fits with the vendor naming there devices aqr107, for example. Signed-off-by: Andrew Lunn --- drivers/net/phy/aquantia.c | 64

Re: [Patch net-next] net_sched: add performance counters for basic filter

On Sun, Feb 3, 2019 at 10:35 AM Eric Dumazet wrote: > > + for_each_possible_cpu(cpu) { > > + struct tc_basic_pcnt *pf = per_cpu_ptr(f->pf, cpu); > > + > > + gpf.rcnt += pf->rcnt; > > + gpf.rhit += pf->rhit; > > + } > > > This looks missing some synchroni

[PATCH 0/4 net-next] net: phy: aquantia: number of improvements

This patch series is based on work from Andrew. I adjusted and added certain parts. The series improves few aspects of driver, no functional change intended. Andrew Lunn (1): net: phy: aquantia: shorten name space prefix to aqr_ Heiner Kallweit (3): net: phy: aquantia: remove unneeded include

Re: [PATCHv2 net] sctp: check and update stream->out_curr when allocating stream_out

On Sat, Feb 2, 2019 at 2:38 AM David Miller wrote: > > From: Neil Horman > Date: Fri, 1 Feb 2019 07:31:18 -0500 > > > On Thu, Jan 31, 2019 at 10:39:41PM -0200, Marcelo Ricardo Leitner wrote: > >> On Tue, Jan 29, 2019 at 07:58:07PM +0100, Tuxdriver wrote: > >> > I was initially under the impressio

[PATCHv3 net] sctp: check and update stream->out_curr when allocating stream_out

Now when using stream reconfig to add out streams, stream->out will get re-allocated, and all old streams' information will be copied to the new ones and the old ones will be freed. So without stream->out_curr updated, next time when trying to send from stream->out_curr stream, a panic would be ca

Re: [PATCH net-next] net: devlink: report cell size of shared buffers

From: Jakub Kicinski Date: Fri, 1 Feb 2019 17:56:28 -0800 > Shared buffer allocation is usually done in cell increments. > Drivers will either round up the allocation or refuse the > configuration if it's not an exact multiple of cell size. > Drivers know exactly the cell size of shared buffer,

Re: [PATCH net] net: systemport: Fix WoL with password after deep sleep

From: Florian Fainelli Date: Fri, 1 Feb 2019 13:23:38 -0800 > Broadcom STB chips support a deep sleep mode where all register > contents are lost. Because we were stashing the MagicPacket password > into some of these registers a suspend into that deep sleep then a > resumption would not lead to

Re: [PATCH net-next] bonding: check slave set command firstly

From: xiangxia.m@gmail.com Date: Fri, 1 Feb 2019 06:07:15 -0800 > From: Tonghao Zhang > > This patch is a little improvement. If user use the > command shown as below, we should print the info [1] > instead of [2]. The eth0 exists actually, and it may > confuse user. > > $ echo "eth0" > /s

Re: [PATCH net-next v3] cxgb4/cxgb4vf: Program hash region for {t4/t4vf}_change_mac()

From: Arjun Vynipadath Date: Fri, 1 Feb 2019 17:07:04 +0530 > {t4/t4_vf}_change_mac() API's were only doing additions to MPS_TCAM. > This will fail, when the number of tcam entries is limited particularly > in vf's. > This fix programs hash region with the mac address, when TCAM > addtion fails

Re: [PATCH v3 0/2] vsock/virtio: fix issues on device hot-unplug

From: Stefano Garzarella Date: Fri, 1 Feb 2019 12:42:05 +0100 > These patches try to handle the hot-unplug of vsock virtio transport device in > a proper way. > > Maybe move the vsock_core_init()/vsock_core_exit() functions in the > module_init > and module_exit of vsock_virtio_transport modul

Re: [Patch net-next] net_sched: add performance counters for basic filter

On 01/17/2019 05:14 PM, Cong Wang wrote: > Similar to u32 filter, it is useful to know how many times > we reach each basic filter and how many times we pass the > ematch attached to it. > > Sample output: > > filter protocol arp pref 49152 basic chain 0 > filter protocol arp pref 49152 basic

Re: [PATCH net] add snmp counter document

From: yupeng Date: Thu, 31 Jan 2019 21:45:59 -0800 > add document for tcp retransmission, tcp fast open, syn cookies, > challenge ack, prune and several general counters > > Signed-off-by: yupeng Please respin this against net-next, thank you.

Re: [PATCH net-next] ipv4/igmp: Don't drop IGMP pkt with zeros src addr

From: Edward Chron Date: Thu, 31 Jan 2019 15:00:40 -0800 > Don't drop IGMP packets with a source address of all zeros which are > IGMP proxy reports. This is documented in Section 2.1.1 IGMP > Forwarding Rules of RFC 4541 IGMP and MLD Snooping Switches > Considerations. > > Signed-off-by: Edward

Re: [PATCH net-next] net: phy: realtek: add generic Realtek PHY driver

From: Heiner Kallweit Date: Sun, 3 Feb 2019 16:07:33 +0100 > The integrated PHY's of later RTL8168 network chips report the generic > PHYID 0x001cc800 (Realtek OUI, model and revision number both set to > zero) and therefore currently the genphy driver is used. > > To be able to use the paged ve

Re: [PATCH bpf-next 3/3] bpf, doc: add RISC-V to filter.txt

From: bjorn.to...@gmail.com Date: Sun, 3 Feb 2019 12:51:31 +0100 > From: Björn Töpel > > Update Documentation/networking/filter.txt to mention RISC-V. > > Signed-off-by: Björn Töpel Acked-by: David S. Miller

Re: [PATCH iproute2-next] tc: add 'kind' property to 'csum' action

On 1/31/19 10:58 AM, Davide Caratti wrote: > unlike other TC actions already supporting JSON printout, 'csum' does not > print the value of TCA_KIND in the 'kind' property: remove 'csum' word > from 'csum' property, and add a separate 'kind' property containing the > action name. The human-readable

Re: [PATCH bpf-next 2/3] MAINTAINERS: add RISC-V BPF JIT maintainer

From: bjorn.to...@gmail.com Date: Sun, 3 Feb 2019 12:51:30 +0100 > From: Björn Töpel > > Add Björn Töpel as RISC-V BPF JIT maintainer. > > Signed-off-by: Björn Töpel Acked-by: David S. Miller

Re: [PATCH bpf-next 1/3] bpf, riscv: add BPF JIT for RV64G

From: bjorn.to...@gmail.com Date: Sun, 3 Feb 2019 12:51:29 +0100 > From: Björn Töpel > > This commit adds BPF JIT for RV64G. > > The JIT is a two-pass JIT, and has a dynamic prolog/epilogue (similar > to the MIPS64 BPF JIT) instead of static ones (e.g. x86_64). > > At the moment the RISC-V Li

Re: [PATCH iproute2-next] tc: full JSON support for 'bpf' actions

On 1/31/19 10:58 AM, Davide Caratti wrote: > Add full JSON output support in the dump of 'act_bpf'. > > Example using eBPF: > > # tc actions flush action bpf > # tc action add action bpf object bpf/action.o section 'action-ok' > # tc -j action list action bpf | jq > [ >{ > "total act

Re: [PATCH] net: fix IPv6 prefix route residue

From: Zhiqiang Liu Date: Sun, 3 Feb 2019 14:10:31 +0800 > @@ -1165,7 +1165,8 @@ enum cleanup_prefix_rt_t { > list_for_each_entry(ifa, &idev->addr_list, if_list) { > if (ifa == ifp) > continue; > - if (!ipv6_prefix_equal(&ifa->addr, &ifp->addr,

[PATCH net-next] net: phy: realtek: add generic Realtek PHY driver

The integrated PHY's of later RTL8168 network chips report the generic PHYID 0x001cc800 (Realtek OUI, model and revision number both set to zero) and therefore currently the genphy driver is used. To be able to use the paged version of e.g. phy_write() we need a PHY driver with the read_page and w

[RFC/PATCH net-next 3/9] linux/dim: Rename externally exposed macros

Renamed macros in use by external drivers. Signed-off-by: Tal Gilboa --- drivers/net/ethernet/broadcom/bcmsysport.c | 4 ++-- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 2 +- drivers/net/ethernet/broadcom/bnxt/bnxt_dim.c | 2 +- drivers/net/ethernet/broadcom/genet/bcmgenet.c | 4 +

[RFC/PATCH net-next 4/9] linux/dim: Rename net_dim_sample() to net_dim_create_sample()

In order to avoid confusion between the function and the similarly named struct. In preparation for removing the 'net' prefix from dim members. Signed-off-by: Tal Gilboa --- drivers/net/ethernet/broadcom/bcmsysport.c| 4 ++-- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 8

[RFC/PATCH net-next 7/9] linux/dim: Add completions count to dim_sample

From: Yamin Friedman Added a measurement of completions per/msec to allow for completion based dim algorithms. Signed-off-by: Yamin Friedman Signed-off-by: Tal Gilboa --- drivers/net/ethernet/broadcom/bcmsysport.c| 2 +- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 1 + drive

[RFC/PATCH net-next 8/9] linux/dim: Implement blk_dim.h

From: Yamin Friedman blk_dim implements a different algorithm than net_dim that is optimized for nvmf storage applications. The algorithm optimizes for number of completions and ratio between completions and events. It also has a feature for fast reduction of moderation level when the traffic cha

[RFC/PATCH net-next 6/9] linux/dim: Move implementation to .c files

Moved all logic from dim.h and net_dim.h to dim.c and net_dim.c. Signed-off-by: Tal Gilboa --- MAINTAINERS | 2 + include/linux/dim.h | 86 +++--- include/linux/net_dim.h | 182 ++--- lib/Kconfig | 7 ++ lib/Makefile

[RFC/PATCH net-next 5/9] linux/dim: Rename externally used net_dim members

Removed 'net' prefix from functions and structs used by external drivers. Signed-off-by: Tal Gilboa --- drivers/net/ethernet/broadcom/bcmsysport.c| 16 +- drivers/net/ethernet/broadcom/bcmsysport.h| 2 +- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 10 +++ drivers/net/et

[RFC/PATCH net-next 2/9] linux/dim: Remove "net" prefix from internal DIM members

Only renaming functions and structs which aren't used by an external code. Signed-off-by: Tal Gilboa --- include/linux/dim.h | 86 - include/linux/net_dim.h | 86 - 2 files changed, 86 insertions(+), 86 deletions

[RFC/PATCH net-next 0/9] net/dim: Support for multiple implementations

net_dim.h lib exposes an implementation of the DIM algorithm for dynamically-tuned interrupt moderation for networking interfaces. We need the same behavior for any block CQ. The main motivation is two benefit from maximized completion rate and reduced interrupt overhead that DIM may provide. C

[RFC/PATCH net-next 1/9] linux/dim: Move logic to dim.h

In preparation for supporting more implementations of the DIM algorithm, I'm moving what would become common logic to a common library. Downstream DIM implementations will use the common lib for their implementation. Signed-off-by: Tal Gilboa --- include/linux/dim.h | 183 +++

[RFC/PATCH net-next 9/9] drivers/infiniband: Use blk_dim in infiniband driver

From: Yamin Friedman Added the interface in the infiniband driver that applies the blk_dim adaptive moderation. Performance improvment (ConnectX-5 100GbE, x86) running FIO benchmark over NVMf between two equal end-hosts across a Mellanox switch: Running long tests that switch between periods of

[PATCH bpf-next 1/3] bpf, riscv: add BPF JIT for RV64G

From: Björn Töpel This commit adds BPF JIT for RV64G. The JIT is a two-pass JIT, and has a dynamic prolog/epilogue (similar to the MIPS64 BPF JIT) instead of static ones (e.g. x86_64). At the moment the RISC-V Linux port does not support HAVE_KPROBES, which means that CONFIG_BPF_EVENTS is not s

[PATCH bpf-next 0/3] Add RISC-V (RV64G) BPF JIT

From: Björn Töpel Hi! This series adds an RV64G BPF JIT to the kernel. I've sent out a RFC for a couple of weeks ago, and think this code is feature-complete/stable enough (famous last words) for a proper patch. Unfortunately, kprobes isn't supported by RISC-V yet (Patrick Stählin sent out an R

[PATCH bpf-next 2/3] MAINTAINERS: add RISC-V BPF JIT maintainer

From: Björn Töpel Add Björn Töpel as RISC-V BPF JIT maintainer. Signed-off-by: Björn Töpel --- MAINTAINERS | 6 ++ 1 file changed, 6 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 019a2bcfbd09..b4491132b9ce 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2907,6 +2907,12 @@ L:

[PATCH bpf-next 3/3] bpf, doc: add RISC-V to filter.txt

From: Björn Töpel Update Documentation/networking/filter.txt to mention RISC-V. Signed-off-by: Björn Töpel --- Documentation/networking/filter.txt | 16 +--- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/Documentation/networking/filter.txt b/Documentation/networkin

Re: [PATCH iproute2-next] devlink: add info subcommand

Sat, Feb 02, 2019 at 11:06:08PM CET, jakub.kicin...@netronome.com wrote: >On Sat, 2 Feb 2019 16:59:38 +0100, Jiri Pirko wrote: >> Sat, Feb 02, 2019 at 01:03:38AM CET, jakub.kicin...@netronome.com wrote: >> >Add support for reading the device serial number and versions >> >from the kernel. >> > >> >

Re: [PATCH rdma-next 00/12] Add SRQ and XRC support for ODP MRs

On Thu, Jan 31, 2019 at 04:27:39PM -0700, Jason Gunthorpe wrote: > On Tue, Jan 22, 2019 at 08:48:39AM +0200, Leon Romanovsky wrote: > > From: Leon Romanovsky > > > > Hi, > > > > This series extend ODP to work with SRQ and XRC. Being both per-operation > > (e.g. RDMA write, RDMA read and atomic) an

[PATCH net-next V2 1/1] openvswitch: Declare ovs key structures using macros

Declare ovs key structures using macros as a pre-step towards to enable retrieving fields information, as a work done in proposed commit in the OVS tree https://patchwork.ozlabs.org/patch/1023406/ ("odp-util: Do not rewrite fields with the same values as matched"), with no functional change. Signe

Re: [PATCH mlx5-next 12/12] net/mlx5: Set ODP SRQ support in firmware

On Thu, Jan 31, 2019 at 04:28:44PM -0700, Jason Gunthorpe wrote: > On Tue, Jan 22, 2019 at 08:48:51AM +0200, Leon Romanovsky wrote: > > From: Moni Shoua > > > > To avoid compatibility issue with older kernels the firmware doesn't > > allow SRQ to work with ODP unless kernel asks for it. > > > > Si

Re: [RFC PATCH net-next 1/6 v2] net/sched: Introduce act_ct

On 01/02/2019 15:23, Marcelo Leitner wrote: > On Tue, Jan 29, 2019 at 10:02:01AM +0200, Paul Blakey wrote: > ... >> diff --git a/include/uapi/linux/tc_act/tc_ct.h >> b/include/uapi/linux/tc_act/tc_ct.h >> new file mode 100644 >> index 000..6dbd771 >> --- /dev/null >> +++ b/include/uapi/linu

[PATCH bpf-next] bpf: support SO_DEBUG in bpf_setsockopt()

Then we can enable/disable socket debugging without modifying user code. That is more convenient for debugging. Signed-off-by: Yafang Shao --- include/net/sock.h | 8 net/core/filter.c | 3 +++ net/core/sock.c| 8 3 files changed, 11 insertions(+), 8 deletions(-) diff --g