Re: [Patch net-next v2] net: add a generic tracepoint for TX queue timeout

Thu, May 02, 2019 at 04:56:59AM CEST, xiyou.wangc...@gmail.com wrote: >Although devlink health report does a nice job on reporting TX >timeout and other NIC errors, unfortunately it requires drivers >to support it but currently only mlx5 has implemented it. >Before other drivers could catch up, it

Re: [PATCH net] udp: fix GRO packet of death

On Wed, 2019-05-01 at 18:56 -0700, Eric Dumazet wrote: > syzbot was able to crash host by sending UDP packets with a 0 payload. > > TCP does not have this issue since we do not aggregate packets without > payload. > > Since dev_gro_receive() sets gso_size based on skb_gro_len(skb) > it seems not

[RFC bpf-next 0/7] busy poll support for AF_XDP sockets

This RFC proposes to add busy-poll support to AF_XDP sockets. With busy-poll, the driver is executed in process context by calling the poll() syscall. The main advantage with this is that all processing occurs on a single core. This eliminates the core-to-core cache transfers that occur between the

[RFC bpf-next 1/7] net: fs: make busy poll budget configurable in napi_busy_loop

This patch adds the possibility to set the busy poll budget to something else than 8 in napi_busy_loop. All the current users of napi_busy_loop will still have a budget of 8, but the for the XDP socket busy poll support, we need to have a configurable budget that is usually larger since each packet

[RFC bpf-next 4/7] netdevice: introduce busy-poll setsockopt for AF_XDP

This patch introduces a new setsockopt that enables busy-poll for XDP sockets. It is called XDP_BUSY_POLL_BATCH_SIZE and takes batch size as an argument. A value between 1 and NAPI_WEIGHT (64) will turn it on, 0 will turn it off and any other value will return an error. There is also a correspondin

[RFC bpf-next 7/7] samples/bpf: add busy-poll support to xdpsock sample

This patch adds busy-poll support to the xdpsock sample application. It is enabled by the "-b" or the "--busy-poll" command line options. Signed-off-by: Magnus Karlsson --- samples/bpf/xdpsock_user.c | 203 - 1 file changed, 125 insertions(+), 78 delet

[RFC bpf-next 2/7] net: i40e: ixgbe: tun: veth: virtio-net: centralize xdp_rxq_info and add napi id

This patch centralizes the xdp_rxq_info struct to only reside in a single place and adds napi id to the information contained in it. The reason to add napi id is that it is needed for the AF_XDP busy poll support. The xsk code needs to know what napi id to call when it gets a poll request on a soc

[RFC bpf-next 6/7] libbpf: add busy-poll support to XDP sockets

This patch adds busy-poll support for XDP sockets to libbpf. A new option is provided in the xsk_socket_config struct called busy_poll. The value of it is the desired batch size. A value between 1 and NAPI_WEIGHT (64) will turn it on, 0 will turn it off and any other value will return an error. Si

Re: [Patch net-next v2] net: add a generic tracepoint for TX queue timeout

On 5/2/2019 5:56 AM, Cong Wang wrote: > Although devlink health report does a nice job on reporting TX > timeout and other NIC errors, unfortunately it requires drivers > to support it but currently only mlx5 has implemented it. > Before other drivers could catch up, it is useful to have a > gene

[RFC bpf-next 5/7] net: add busy-poll support for XDP sockets

This patch adds busy-poll support for XDP sockets (AF_XDP). With busy-poll, the driver is executed in process context by calling the poll() syscall. The main advantage with this is that all processing occurs on a single core. This eliminates the core-to-core cache transfers that occur between the a

[net-next 07/15] ice: Add 52 byte RSS hash key support

From: Paul Greenwalt Add support to set 52 byte RSS hash key. Signed-off-by: Paul Greenwalt Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/ice_adminq_cmd.h | 3 +++ drivers/net/ethernet/intel/ice/ice_lib.c

[net-next 05/15] ice: Validate ring existence and its q_vector per VSI

From: Maciej Fijalkowski When stopping Tx rings, we use 'i' as an ring array index for looking up whether the ice_ring exists and have assigned a q_vector. This checks rings only within a given TC and we need to go through every ring in VSI. Use 'q_idx' instead. Signed-off-by: Maciej Fijalkowski

[net-next 00/15][pull request] 100GbE Intel Wired LAN Driver Updates 2019-05-02

This series contains updates to the ice driver only. Anirudh introduces the framework to store queue specific information in the VSI queue contexts. This will allow future changes to update the structure to hold queue specific information. Akeem adds additional check so that if there is no queue

[net-next 10/15] ice: Fix issue when adding more than allowed VLANs

From: Akeem G Abodunrin This patch fixes issue with non trusted VFs being able to add more than permitted number of VLANs by adding a check in ice_vc_process_vlan_msg. Also don't return an error in this case as the VF does not need to know that it is not trusted. Also rework ice_vsi_kill_vlan to

[net-next 02/15] ice: Return configuration error without queue to disable

From: Akeem G Abodunrin If there is no queue to disable, return appropriate configuration error earlier without acquiring the lock. Signed-off-by: Akeem G Abodunrin Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/

[net-next 15/15] ice: Use dev_err when ice_cfg_vsi_lan fails

From: Brett Creeley dev_err makes more sense than dev_info when this call fails. Signed-off-by: Brett Creeley Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/ice_lib.c | 9 ++--- 1 file changed, 6 insertions(+

[net-next 09/15] ice: Remove unnecessary wait when disabling/enabling Rx queues

From: Brett Creeley In ice_vsi_ctrl_rx_rings() we are unnecessarily waiting for QRX_CTRL_QENA_REQ and QRX_CTRL_QENA_STAT to be the same value prior to disabling each Rx queue. There is no reason to do this so remove this wait loop as we already have a wait loop after disabling/enabling the Rx que

[net-next 14/15] ice: Refactor link event flow

From: Brett Creeley Currently the link event flow works, but can be much better. Refactor the link event flow to make it cleaner and more clear on what is going on. Signed-off-by: Brett Creeley Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher --- dr

[net-next 12/15] ice: Add reg_idx variable in ice_q_vector structure

From: Brett Creeley Every time we want to re-enable interrupts and/or write to a register that requires an interrupt vector's hardware index we do the following: vsi->hw_base_vector + q_vector->v_idx This is a wasteful operation, especially in the hot path. Fix this by adding a u16 reg_idx memb

[net-next 03/15] ice: Resolve static analysis reported issue

From: Bruce Allan Static analysis points out the default case in the switch statement in ice_get_itr_intrl_gran() is an infeasible condition causing the default case statement to be unreachable. Remove it and since the function no longer returns anything but success, change it to just return voi

[net-next 13/15] ice: Add missing PHY type to link settings

From: Tony Nguyen The PHY type ICE_PHY_TYPE_LOW_25G_AUI_C2C is missing from ice_get_settings_link_up() which is causing a warning message for unrecognized PHY. Add the PHY type to correctly set the settings and avoid the warning message. Signed-off-by: Tony Nguyen Signed-off-by: Anirudh Venkat

[net-next 04/15] ice: Reduce scope of variable in ice_vsi_cfg_rxqs

From: Brett Creeley Reduce scope of the variable 'err' to inside the for loop instead of using it as a second looping conditional. Also while here, improve the debug message if we fail to configure a Rx queue. Signed-off-by: Brett Creeley Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew

[net-next 06/15] ice: Use ice_for_each_q_vector macro where possible

From: Brett Creeley There are many places in the code where we do the following: for (i = 0; i < vsi->num_q_vectors; i++) Instead use the macro mentioned in the commit title: ice_for_each_q_vector(vsi, i) Signed-off-by: Brett Creeley Signed-off-by: Anirudh Venkataramanan Tested-by: Andrew B

[net-next 01/15] ice: Create framework for VSI queue context

From: Anirudh Venkataramanan This patch introduces a framework to store queue specific information in VSI queue contexts. Currently VSI queue context (represented by struct ice_q_ctx) only has q_handle as a member. In future patches, this structure will be updated to hold queue specific informati

[net-next 11/15] ice: Remove runtime change of PFINT_OICR_ENA register

From: Md Fahad Iqbal Polash Runtime change of PFINT_OICR_ENA register is unnecessary. The handlers should always clear the atomic bit for each task as they start, because it will make sure that any late interrupt will either 1) re-set the bit, or 2) be handled directly in the "already running" ta

[net-next 08/15] ice: Add ability to update rx-usecs-high

From: Brett Creeley Currently the driver allows rx-usecs-high values to be set, but when querying the device for rx-usecs-high the value does not stick. This is because it was not yet implemented. Add code to allow the user to change rx-usecs-high and use this to set the q_vector's intrl value.

Re: [net-next 12/12] i40e: Introduce recovery mode support

On Mon, 2019-04-29 at 21:07 -0400, Jakub Kicinski wrote: > On Mon, 29 Apr 2019 12:16:28 -0700, Jeff Kirsher wrote: > > From: Alice Michael > > > > This patch introduces "recovery mode" to the i40e driver. It is > > part of a new Any2Any idea of upgrading the firmware. In this > > approach, it is

Re: [PATCH v2 6/7] net: Rename skb_frag_t size to bv_len

://github.com/0day-ci/linux/commits/Matthew-Wilcox/Convert-skb_frag_t-to-bio_vec/20190502-161948 config: xtensa-allyesconfig (attached as .config) compiler: xtensa-linux-gcc (GCC) 8.1.0 reproduce: wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross

Re: [PATCH v2 7/7] net: Convert skb_frag_t to bio_vec

://github.com/0day-ci/linux/commits/Matthew-Wilcox/Convert-skb_frag_t-to-bio_vec/20190502-161948 config: riscv-allyesconfig (attached as .config) compiler: riscv64-linux-gcc (GCC) 8.1.0 reproduce: wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross

Re: [PATCH net] neighbor: Reset gc_entries counter if new entry is released before insert

On Wed, 1 May 2019, David Ahern wrote: > From: David Ahern > > Ian and Alan both reported seeing overflows after upgrades to 5.x kernels: > neighbour: arp_cache: neighbor table overflow! > > Alan's mpls script helped get to the bottom of this bug. When a new entry > is created the gc_entries

Re: [PATCH net] neighbor: Call __ipv4_neigh_lookup_noref in neigh_xmit

On Wed, 1 May 2019, David Ahern wrote: > From: David Ahern > > Commit cd9ff4de0107 changed the key for IFF_POINTOPOINT devices to > INADDR_ANY but neigh_xmit which is used for MPLS encapsulations was not > updated to use the altered key. The result is that every packet Tx does > a lookup on the

Re: MPLS encapsulation and arp table overflow

On Wed, 1 May 2019, David Ahern wrote: > On 5/1/19 10:03 AM, Alan Maguire wrote: > > I'm seeing the following repeated error > > > > [ 130.821362] neighbour: arp_cache: neighbor table overflow! > > > > when using MPLSoverGRE or MPLSoverUDP tunnels on bits synced > > with bpf-next as of this mor

Re: [PATCH v2 6/7] net: Rename skb_frag_t size to bv_len

://github.com/0day-ci/linux/commits/Matthew-Wilcox/Convert-skb_frag_t-to-bio_vec/20190502-161948 config: riscv-allyesconfig (attached as .config) compiler: riscv64-linux-gcc (GCC) 8.1.0 reproduce: wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross

Re: [PATCH v2 7/7] net: Convert skb_frag_t to bio_vec

://github.com/0day-ci/linux/commits/Matthew-Wilcox/Convert-skb_frag_t-to-bio_vec/20190502-161948 config: xtensa-allyesconfig (attached as .config) compiler: xtensa-linux-gcc (GCC) 8.1.0 reproduce: wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross

Re: [PATCH 4/5] net: Use skb accessors for skb->page

://github.com/0day-ci/linux/commits/Matthew-Wilcox/net-Increase-the-size-of-skb_frag_t/20190502-125302 config: arm-allmodconfig (attached as .config) compiler: arm-linux-gnueabi-gcc (Debian 7.2.0-11) 7.2.0 reproduce: wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O

[PATCH net-next 0/3] mlxsw: Firmware version update

From: Ido Schimmel This patchset updates mlxsw to use a new firmware version and adds support for split into two ports on Spectrum-2 based systems. Patch #1 updates the firmware version to 13.2000.1122 Patch #2 queries new resources from the firmware. Patch #3 makes use of these resources in o

[PATCH net-next 2/3] mlxsw: resources: Add local_ports_in_{1x, 2x}

From: Shalom Toledo Since the number of local ports in 4x changed between SPC and SPC-2, firmware expose new resources that the driver can query. Signed-off-by: Shalom Toledo Signed-off-by: Ido Schimmel --- drivers/net/ethernet/mellanox/mlxsw/resources.h | 4 1 file changed, 4 insertions

[PATCH net-next 3/3] mlxsw: spectrum: split base on local_ports_in_{1x, 2x} resources

From: Shalom Toledo When splitting a port, different local ports need to be mapped on different systems. For example: SN3700 (local_ports_in_2x=2): * Without split: front panel 1 --> local port 1 front panel 2 --> local port 5 * Split to 2: front panel 1s0 --> local por

[PATCH net-next 1/3] mlxsw: Bump firmware version to 13.2000.1122

From: Ido Schimmel The new version supports two features that are required by upcoming changes in the driver: * Querying of new resources allowing port split into two ports on Spectrum-2 systems * Querying of number of gearboxes on supported systems such as SN3800 Signed-off-by: Ido Schimmel

Re: [PATCH v2 6/7] net: Rename skb_frag_t size to bv_len

: https://github.com/0day-ci/linux/commits/Matthew-Wilcox/Convert-skb_frag_t-to-bio_vec/20190502-161948 reproduce: # apt-get install sparse make ARCH=x86_64 allmodconfig make C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' If you fix the issue, kindly add following tag R

Re: [bisected regression] e1000e: "Detected Hardware Unit Hang"

Hi All. While updating to kernel 4.19, we realised that a problem reported in 2015 for kernel 3.7 is still around. Please see this link for more details: https:// marc.info/?l=linux-netdev&m=142124954120315 Basically, when using the e1000e driver, each few minutes the following messages appear

ndisc_cache garbage collection issue

I recently upgraded a machine from a 4.20.13 kernel to 5.0.9 and am finding that after a few days I start getting a lot of these messages: neighbour: ndisc_cache: neighbor table overflow! and IPv6 networking starts to fail intermittently as a result. The neighbour table doesn't appear to have

Re: [BUG][iproute2][5.0] ip route show table default: "Error: ipv4: FIB table does not exist."

Mar 24, 2019, 10:09 PM by dsah...@gmail.com: > On 3/24/19 12:36 PM, Michal Kubecek wrote: > >> On Sun, Mar 24, 2019 at 07:29:08PM +0100, Michal Kubecek wrote: >> >>> On Sun, Mar 24, 2019 at 11:20:33AM -0600, David Ahern wrote: >>> On 3/24/19 11:02 AM, emersonbern...@tutanota.com

Re: ndisc_cache garbage collection issue

On 5/2/19 5:42 AM, Tom Hughes wrote: > I recently upgraded a machine from a 4.20.13 kernel to 5.0.9 and am > finding that after a few days I start getting a lot of these messages: > >   neighbour: ndisc_cache: neighbor table overflow! > > and IPv6 networking starts to fail intermittently as a

Re: [PATCH] net: ll_temac: Fix typo bug for 32-bit

From: Esben Haabendal Date: Thu, 2 May 2019 08:43:43 +0200 > Fixes: d84aec42151b ("net: ll_temac: Fix support for 64-bit platforms") > > Signed-off-by: Esben Haabendal Applied.

Re: kTLS broken somewhere between 4.18 and 5.0

On 14.04.2019 22:40, John Fastabend wrote: On 4/13/19 6:56 PM, Andre Tomt wrote: On 13.04.2019 17:34, Steinar H. Gunderson wrote: Hi, I've been using kTLS for a while, with my video reflector Cubemap (https://git.sesse.net/?p=cubemap). After I upgraded my server from 4.18.11 to 5.0.6, seemingl

Re: [net-next 01/12] i40e: replace switch-statement to speed-up retpoline-enabled builds

On 04/29/2019 09:16 PM, Jeff Kirsher wrote: > From: Björn Töpel > > GCC will generate jump tables for switch-statements with more than 5 > case statements. An entry into the jump table is an indirect call, > which means that for CONFIG_RETPOLINE builds, this is rather > expensive. > > This commi

[PATCH bpf] libbpf: always NULL out pobj in bpf_prog_load_xattr

Currently, code like the following segfaults if bpf_prog_load_xattr returns an error: struct bpf_object *obj; err = bpf_prog_load_xattr(&attr, &obj, &prog_fd); bpf_object__close(obj); if (err) ... Unconditionally reset pobj to NULL at the start of the function to fix this

Hyperv netvsc - regression for 32-PAE kernel

Hi all. I have a custom linux OS vm running kernel 3.14 (32b with or without PAE) in windows 2012 R2. The vm has one Network Adapter and is generation 1. With this setup everything runs fine. The problem started when I tried to update to kernel 4.19. The Synthetic network adapter driver does n

Re: [PATCH bpf] libbpf: always NULL out pobj in bpf_prog_load_xattr

On Thu, May 02, 2019 at 11:49:32AM -0400, Lorenz Bauer wrote: > Currently, code like the following segfaults if bpf_prog_load_xattr > returns an error: > > struct bpf_object *obj; > > err = bpf_prog_load_xattr(&attr, &obj, &prog_fd); > bpf_object__close(obj); This is a bug. err shoul

[RFC] folding socket->wq into struct socket

I'm not sure what's the right way to handle that. Background: new inode method (->free_inode()) allows to do RCU-delayed parts of ->destroy_inode() conveniently, killing a lot of boilerplate code in process. It's optional, so sockfs doesn't have to be converted; however, looking at

RE

My name is Ms Ella Golan, I'm the Chief Executive Officer (C.P.A) of the First International Bank of Israel (FIBI). I'm getting in touch with you in regards to a very important and urgent matter. Kindly respond back at your earliest convenience so I can provide you the details. Faithfully, Ms Ell

Re: [net-next][PATCH v2 1/2] rds: handle unsupported rdma request to fs dax memory

On 5/1/2019 11:21 PM, Leon Romanovsky wrote: On Wed, May 01, 2019 at 10:54:00AM -0700, Santosh Shilimkar wrote: On 5/1/2019 12:44 AM, Leon Romanovsky wrote: On Mon, Apr 29, 2019 at 04:37:19PM -0700, Santosh Shilimkar wrote: From: Hans Westgaard Ry RDS doesn't support RDMA on memory apertures

Re: Question re. skb_orphan for TPROXY

On 4/16/19 8:00 AM, Florian Westphal wrote: > Lorenz Bauer wrote: >> Apologies for contacting you out of the blue. I'm currently trying to >> understand how TPROXY works under the hood. As part of this endeavour, >> I've stumbled upon the commit attached to this email. >> >> From the commit mes

Re: [net-next][PATCH v2 2/2] rds: add sysctl for rds support of On-Demand-Paging

On 5/1/2019 11:18 PM, Leon Romanovsky wrote: On Wed, May 01, 2019 at 10:54:50AM -0700, Santosh Shilimkar wrote: On 5/1/2019 12:45 AM, Leon Romanovsky wrote: On Mon, Apr 29, 2019 at 04:37:20PM -0700, Santosh Shilimkar wrote: RDS doesn't support RDMA on memory apertures that require On Demand

Re: ndisc_cache garbage collection issue

On 02/05/2019 14:19, Eric Dumazet wrote: On 5/2/19 5:42 AM, Tom Hughes wrote: I recently upgraded a machine from a 4.20.13 kernel to 5.0.9 and am finding that after a few days I start getting a lot of these messages:   neighbour: ndisc_cache: neighbor table overflow! and IPv6 networking sta

[Patch net-next] sch_htb: redefine htb qdisc overlimits

In commit 3c75f6ee139d ("net_sched: sch_htb: add per class overlimits counter") we added an overlimits counter for each HTB class which could properly reflect how many times we use up all the bandwidth on each class. However, the overlimits counter in HTB qdisc does not, it is way bigger than the s

[PATCH bpf] libbpf: add libbpf_util.h to header install.

The libbpf_util.h is used by xsk.h, so add it to the install headers. Reported-by: Ben Pfaff Signed-off-by: William Tu --- tools/lib/bpf/Makefile | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/lib/bpf/Makefile b/tools/lib/bpf/Makefile index c6c06bc6683c..f91639bf5650 100644 --- a/too

[PATCH net-next] r8169: remove rtl_write_exgmac_batch

rtl_write_exgmac_batch is used in only one place, so we can remove it. Signed-off-by: Heiner Kallweit --- drivers/net/ethernet/realtek/r8169.c | 26 -- 1 file changed, 4 insertions(+), 22 deletions(-) diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ether

Re: [bpf-next PATCH v3 0/4] sockmap/ktls fixes

On 4/30/19 7:06 PM, John Fastabend wrote: > Series of fixes for sockmap and ktls, see patches for descriptions. > > v2: fix build issue for CONFIG_TLS_DEVICE and fixup couple comments > from Jakub > > v3: fix issue where release could call unhash resulting in a use after > free. Now we de

Re: Bug#927825: arm: mvneta driver used on Armada XP GP boards does not receive packets (regression from 4.9)

Hello, On Tue, Apr 30, 2019 at 10:12:27AM +0200, Uwe Kleine-König wrote: > On Thu, Apr 25, 2019 at 09:17:32PM +0200, Aurelien Jarno wrote: > > On 2019-04-25 14:50, Aurelien Jarno wrote: > > > On 2019-04-23 22:16, Aurelien Jarno wrote: > > > > Source: linux > > > > Version: 4.19.28-2 > > > > Severi

[PATCH net-next] net: openvswitch: return an error instead of doing BUG_ON()

For all other error cases in queue_userspace_packet() the error is returned, so it makes sense to do the same for these two error cases. Reported-by: Davide Caratti Signed-off-by: Eelco Chaudron --- net/openvswitch/datapath.c |7 +-- 1 file changed, 5 insertions(+), 2 deletions(-) diff

Re: [PATCH bpf] libbpf: add libbpf_util.h to header install.

On Thu, May 2, 2019 at 11:34 AM William Tu wrote: > > The libbpf_util.h is used by xsk.h, so add it to > the install headers. Can we try to change code a little bit to avoid exposing libbpf_util.h? Originally libbpf_util.h is considered as libbpf internal. I am not strongly against this patch. Bu

Re: [ovs-dev] [PATCH net-next] net: openvswitch: return an error instead of doing BUG_ON()

On Thu, May 02, 2019 at 04:12:38PM -0400, Eelco Chaudron wrote: > For all other error cases in queue_userspace_packet() the error is > returned, so it makes sense to do the same for these two error cases. > > Reported-by: Davide Caratti > Signed-off-by: Eelco Chaudron > --- LGTM Acked-by: Flavi

Re: [net-next 01/12] i40e: replace switch-statement to speed-up retpoline-enabled builds

On 2019-05-02 16:47, Daniel Borkmann wrote: On 04/29/2019 09:16 PM, Jeff Kirsher wrote: From: Björn Töpel GCC will generate jump tables for switch-statements with more than 5 case statements. An entry into the jump table is an indirect call, which means that for CONFIG_RETPOLINE builds, this i

Re: [net-next 01/12] i40e: replace switch-statement to speed-up retpoline-enabled builds

On Thu, 2019-05-02 at 22:29 +0200, Björn Töpel wrote: > On 2019-05-02 16:47, Daniel Borkmann wrote: > > On 04/29/2019 09:16 PM, Jeff Kirsher wrote: > > > From: Björn Töpel > > > > > > GCC will generate jump tables for switch-statements with more than 5 > > > case statements. An entry into the jum

Re: [net-next 01/12] i40e: replace switch-statement to speed-up retpoline-enabled builds

On Thu, 2019-05-02 at 22:56 +0200, Björn Töpel wrote: > On 2019-05-02 22:40, Jeff Kirsher wrote: > > On Thu, 2019-05-02 at 22:29 +0200, Björn Töpel wrote: > > > On 2019-05-02 16:47, Daniel Borkmann wrote: > > > > On 04/29/2019 09:16 PM, Jeff Kirsher wrote: > > > > > From: Björn Töpel > > > > > >

Re: [net-next 01/12] i40e: replace switch-statement to speed-up retpoline-enabled builds

On 2019-05-02 22:40, Jeff Kirsher wrote: On Thu, 2019-05-02 at 22:29 +0200, Björn Töpel wrote: On 2019-05-02 16:47, Daniel Borkmann wrote: On 04/29/2019 09:16 PM, Jeff Kirsher wrote: From: Björn Töpel GCC will generate jump tables for switch-statements with more than 5 case statements. An en

[PATCH net] ipmr_base: Do not reset index in mr_table_dump

From: David Ahern e is the counter used to save the location of a dump when an skb is filled. Once the walk of the table is complete, mr_table_dump needs to return without resetting that index to 0. Dump of a specific table is looping because of the reset because there is no way to indicate the w

Re: [PATCH 2/5] net: dsa: lantiq: Add VLAN unaware bridge offloading

On 5/2/19 12:34 AM, Andrew Lunn wrote: > On Wed, May 01, 2019 at 10:45:03PM +0200, Hauke Mehrtens wrote: > > Hi Hauke > > /* Add the LAN port into a bridge with the CPU port by >> + * default. This prevents automatic forwarding of >> + * packages between the LAN ports when no explicit >> + * brid

[PATCH net-next] ipmr: Do not define MAXVIFS twice

From: David Ahern b70432f7319eb refactored mroute code to make it common between ipv4 and ipv6. In the process, MAXVIFS got defined a second time: the first is in the uapi file linux/mroute.h. A second one was created presumably for IPv6 but it is not needed. Remove it and have mroute_base.h incl

RE: Hyperv netvsc - regression for 32-PAE kernel

From: Juliana Rodrigueiro Sent: Thursday, May 2, 2019 9:14 AM > > So I got to the following commit: > > commit 6ba34171bcbd10321c6cf554e0c1144d170f9d1a > Author: Michael Kelley > Date: Thu Aug 2 03:08:24 2018 + > > Drivers: hv: vmbus: Remove use of slow_virt_to_phys() > > slow_

Re: [PATCH v2] ss: add option to print socket information on one line

On 4/30/19 7:38 PM, Josh Hunt wrote: > Multi-line output in ss makes it difficult to search for things with > grep. This new option will make it easier to find sockets matching > certain criteria with simple grep commands. > ... > > Signed-off-by: Josh Hunt > --- > > v1 -> v2 > * Update long o

[RFC PATCH bpf-next v2 1/3] bpf: implement bpf_send_signal() helper

This patch tries to solve the following specific use case. Currently, bpf program can already collect stack traces when certain events happens (e.g., cache miss counter or cpu clock counter overflows). These stack traces can be used for performance analysis. For jitted programs, e.g., hhvm (jited

[RFC PATCH bpf-next v2 3/3] tools/bpf: add a selftest for bpf_send_signal() helper

The test covered both nmi and tracepoint perf events. $ ./test_send_signal_user test_send_signal (tracepoint): OK test_send_signal (perf_event): OK Signed-off-by: Yonghong Song --- tools/testing/selftests/bpf/Makefile | 5 +- tools/testing/selftests/bpf/bpf_helpers.h | 2 +

[RFC PATCH bpf-next v2 2/3] tools/bpf: sync bpf uapi header bpf.h

sync bpf uapi header bpf.h to tools directory. Signed-off-by: Yonghong Song --- tools/include/uapi/linux/bpf.h | 15 ++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h index 72336bac7573..e3e824848335 1006

[RFC PATCH bpf-next v2 0/3] bpf: implement bpf_send_signal() helper

Currently, bpf program can already collect stack traces when certain events happens (e.g., cache miss counter or cpu clock counter overflows). These stack traces can be used for performance analysis. For jitted programs, e.g., hhvm (jited php), it is very hard to get the true stack trace in the

Re: [RFC bpf-next 4/7] netdevice: introduce busy-poll setsockopt for AF_XDP

On 5/2/2019 1:39 AM, Magnus Karlsson wrote: This patch introduces a new setsockopt that enables busy-poll for XDP sockets. It is called XDP_BUSY_POLL_BATCH_SIZE and takes batch size as an argument. A value between 1 and NAPI_WEIGHT (64) will turn it on, 0 will turn it off and any other value w

Re: [RFC bpf-next 5/7] net: add busy-poll support for XDP sockets

On 5/2/2019 1:39 AM, Magnus Karlsson wrote: This patch adds busy-poll support for XDP sockets (AF_XDP). With busy-poll, the driver is executed in process context by calling the poll() syscall. The main advantage with this is that all processing occurs on a single core. This eliminates the core

Re: [PATCH v5 bpf-next 13/17] s390: bpf: eliminate zero extension code-gen

Hi Jiong, Thank you for the patch! Yet something to improve: [auto build test ERROR on bpf-next/master] url: https://github.com/0day-ci/linux/commits/Jiong-Wang/bpf-eliminate-zero-extensions-for-sub-register-writes/20190503-104459 base: https://git.kernel.org/pub/scm/linux/kernel/git/bpf/b

Re: [RFC HACK] xfrm: make state refcounting percpu

On Wed, Apr 24, 2019 at 12:40:23PM +0200, Florian Westphal wrote: > I'm not sure this is a good idea to begin with, refcount > is right next to state spinlock which is taken for both tx and rx ops, > plus this complicates debugging quite a bit. Hm, what would be the usecase where this could help?

RE: [RFC HACK] xfrm: make state refcounting percpu

> -Original Message- > From: Steffen Klassert > Sent: Friday, May 3, 2019 11:38 AM > To: Florian Westphal > Cc: Vakul Garg ; netdev@vger.kernel.org > Subject: Re: [RFC HACK] xfrm: make state refcounting percpu > > On Wed, Apr 24, 2019 at 12:40:23PM +0200, Florian Westphal wrote: > > I

Re: [RFC HACK] xfrm: make state refcounting percpu

On Fri, May 03, 2019 at 06:13:22AM +, Vakul Garg wrote: > > > > -Original Message- > > From: Steffen Klassert > > Sent: Friday, May 3, 2019 11:38 AM > > To: Florian Westphal > > Cc: Vakul Garg ; netdev@vger.kernel.org > > Subject: Re: [RFC HACK] xfrm: make state refcounting percpu >

Re: [RFC bpf-next 4/7] netdevice: introduce busy-poll setsockopt for AF_XDP

On Fri, May 3, 2019 at 2:26 AM Samudrala, Sridhar wrote: > > > > On 5/2/2019 1:39 AM, Magnus Karlsson wrote: > > This patch introduces a new setsockopt that enables busy-poll for XDP > > sockets. It is called XDP_BUSY_POLL_BATCH_SIZE and takes batch size as > > an argument. A value between 1 and N

RE: [RFC HACK] xfrm: make state refcounting percpu

> -Original Message- > From: Steffen Klassert > Sent: Friday, May 3, 2019 11:52 AM > To: Vakul Garg > Cc: Florian Westphal ; netdev@vger.kernel.org > Subject: Re: [RFC HACK] xfrm: make state refcounting percpu > > On Fri, May 03, 2019 at 06:13:22AM +, Vakul Garg wrote: > > > > > >

Re: [RFC HACK] xfrm: make state refcounting percpu

On Fri, May 03, 2019 at 06:34:29AM +, Vakul Garg wrote: > > -Original Message- > > From: Steffen Klassert > > > > Also, is this a new problem or was it always like that? > > It is always like this. On 4-core, 8-core platforms as well, these atomics > consume significant cpu > (8 co

RE: [RFC HACK] xfrm: make state refcounting percpu

> -Original Message- > From: Steffen Klassert > Sent: Friday, May 3, 2019 12:16 PM > To: Vakul Garg > Cc: Florian Westphal ; netdev@vger.kernel.org > Subject: Re: [RFC HACK] xfrm: make state refcounting percpu > > On Fri, May 03, 2019 at 06:34:29AM +, Vakul Garg wrote: > > > -