[PATCH net-next 11/12] switchdev: Replace port obj add/del SDO with a notification

Drop switchdev_ops.switchdev_port_obj_add and _del. Drop the uses of this field from all clients, which were migrated to use switchdev notification in the previous patches. Add a new function switchdev_port_obj_notify() that sends the switchdev notifications SWITCHDEV_PORT_OBJ_ADD and _DEL

[PATCH net-next 07/13] net: sched: red: offload a graft notification

Drivers offloading Qdiscs should have reasonable certainty the offloaded behaviour matches the SW path. This is impossible if the driver does not know about all Qdiscs or when Qdiscs move and are reused. Send a graft notification from RED. The drivers are expected to simply stop offloading

[PATCH net-next 04/13] net: sched: provide notification for graft on root

Drivers are currently not notified when a Qdisc is grafted as root. This requires special casing Qdiscs added with parent = TC_H_ROOT in the driver. Also there is no notification sent to the driver when an existing Qdisc is grafted as root. Add this very simple notifications, drivers should now

[PATCH net-next 08/13] net: sched: mq: offload a graft notification

Drivers offloading Qdiscs should have reasonable certainty the offloaded behaviour matches the SW path. This is impossible if the driver does not know about all Qdiscs or when Qdiscs move and are reused. Send a graft notification from MQ. Signed-off-by: Jakub Kicinski Reviewed-by: John Hurley

[PATCH V2 bpf-next 0/2] Perf-based event notification for sock_ops

This patchset uses eBPF perf-event based notification mechanism to solve the problem described in https://marc.info/?l=linux-netdev=154022219423571=2. Thanks to Daniel Borkmann for feedback/input. V2: inlined the call to sys_perf_event_open() following the style of existing code

[PATCH V2 bpf-next 2/2] selftests/bpf: add a test case for sock_ops perf-event notification

) - has a bpf kernel module that sends a perf-event notification for each TCP retransmit, and also tracks the number of such notifications sent in the global_map The test passes when the number of event notifications intercepted in user-space matches the value in the global_map. Signed-off

Re: [PATCH bpf-next 2/2] selftests/bpf: add a test case for sock_ops perf-event notification

re ncat SYNs are > resent (instead of aborting instantly after a TCP RST) > - has a bpf kernel module that sends a perf-event notification for > each TCP retransmit, and also tracks the number of such notifications > sent in the global_map > The test passes when the number of even

[PATCH bpf-next 0/2] TCP-BPF event notification support

This patchset uses eBPF perf-event based notification mechanism to solve the problem described in https://marc.info/?l=linux-netdev=154022219423571=2. Thanks to Daniel Borkmann for feedback/input. The problem statement is We would like to monitor some subset of TCP sockets in user-space

[PATCH bpf-next 2/2] selftests/bpf: add a test case for sock_ops perf-event notification

) - has a bpf kernel module that sends a perf-event notification for each TCP retransmit, and also tracks the number of such notifications sent in the global_map The test passes when the number of event notifications intercepted in user-space matches the value in the global_map. Signed-off

[PATCH net-next 03/19] mlxsw: spectrum_switchdev: Check notification relevance based on upper device

VxLAN FDB updates are sent with the VxLAN device which is not our upper and will therefore be ignored by current code. Solve this by checking whether the upper device (bridge) is our upper. Signed-off-by: Ido Schimmel Reviewed-by: Petr Machata ---

[net-next 01/10] i40e: Fix VF's link state notification

From: Mariusz Stachura This resolves an issue where the VF link state was not being updated when the PF is down or up, and the VF link state would always show that it is running. Signed-off-by: Mariusz Stachura Tested-by: Andrew Bowers Signed-off-by: Jeff Kirsher ---

[PATCH net-next 2/2] neighbour: send netlink notification if NTF_ROUTER changes

From: Roopa Prabhu send netlink notification if neigh_update results in NTF_ROUTER change and if NEIGH_UPDATE_F_ISROUTER is on. Also move the NTF_ROUTER change function into a helper. Signed-off-by: Roopa Prabhu --- include/net/neighbour.h | 15 +++ net/core/neighbour.c| 7

Re: [PATCH iproute2 2/5] bpf: move bpf_elf_map fixup notification under verbose

On Wed, 18 Jul 2018 01:31:19 +0200, Daniel Borkmann wrote: > No need to spam the user with this if it can be fixed gracefully > anyway. Therefore, move it under verbose option. > > Signed-off-by: Daniel Borkmann > --- > lib/bpf.c | 6 +++--- > 1 file changed, 3 insertions(+), 3 deletions(-) >

[PATCH iproute2 2/5] bpf: move bpf_elf_map fixup notification under verbose

No need to spam the user with this if it can be fixed gracefully anyway. Therefore, move it under verbose option. Signed-off-by: Daniel Borkmann --- lib/bpf.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/bpf.c b/lib/bpf.c index 4e26c0d..42093db 100644 ---

Re: [PATCH net 0/3] hv_netvsc: notification and namespace fixes

From: Stephen Hemminger Date: Mon, 11 Jun 2018 12:44:53 -0700 > This set of patches addresses two set of fixes. First it backs out > the common callback model which was merged in net-next without > completing all the review feedback or getting maintainer approval. > > Then it fixes the

[PATCH net 0/3] hv_netvsc: notification and namespace fixes

This set of patches addresses two set of fixes. First it backs out the common callback model which was merged in net-next without completing all the review feedback or getting maintainer approval. Then it fixes the transparent VF management code to handle network namespaces. Stephen Hemminger

[PATCH net-next 11/14] net: sched: mq: add simple offload notification

mq offload is trivial, we just need to let the device know that the root qdisc is mq. Alternative approach would be to export qdisc_lookup() and make drivers check the root type themselves, but notification via ndo_setup_tc is more in line with other qdiscs. Note that mq doesn't hold any stats

Re: [PATCH net-next] net: dsa: fix added_by_user switchdev notification

From: Vivien Didelot Date: Tue, 8 May 2018 23:03:12 -0400 > Commit 161d82de1ff8 ("net: bridge: Notify about !added_by_user FDB > entries") causes the below oops when bringing up a slave interface, > because dsa_port_fdb_add is still scheduled, but with a

Re: [PATCH net-next] net: dsa: fix added_by_user switchdev notification

On 09/05/18 06:03, Vivien Didelot wrote: Commit 161d82de1ff8 ("net: bridge: Notify about !added_by_user FDB entries") causes the below oops when bringing up a slave interface, because dsa_port_fdb_add is still scheduled, but with a NULL address. To fix this, keep the dsa_slave_switchdev_event

Re: [PATCH net-next] net: dsa: fix added_by_user switchdev notification

Vivien Didelot writes: > @petr I expect the same issue with Rocker, but I haven't tested it. Yeah, pretty sure. Such an obvious bug, sorry about that :-/ I'll send a rocker patch later today. Thanks, Petr

[PATCH net-next] net: dsa: fix added_by_user switchdev notification

Commit 161d82de1ff8 ("net: bridge: Notify about !added_by_user FDB entries") causes the below oops when bringing up a slave interface, because dsa_port_fdb_add is still scheduled, but with a NULL address. To fix this, keep the dsa_slave_switchdev_event function agnostic of the notified info

Re: [PATCH net-next] net: bridge: avoid duplicate notification on up/down/change netdev events

From: Nikolay Aleksandrov <niko...@cumulusnetworks.com> Date: Thu, 3 May 2018 13:47:24 +0300 > While handling netdevice events, br_device_event() sometimes uses > br_stp_(disable|enable)_port which unconditionally send a notification, > but then a second notification for the sam

[PATCH net-next] net: bridge: avoid duplicate notification on up/down/change netdev events

While handling netdevice events, br_device_event() sometimes uses br_stp_(disable|enable)_port which unconditionally send a notification, but then a second notification for the same event is sent at the end of the br_device_event() function. To avoid sending duplicate notifications in such cases

NOTIFICATION

Open the attachment to read your Account Update message. Sincerely: Your Email Administrator ZIMBRA ADMIN.pdf Description: Adobe PDF document

Re: [PATCH net 2/2] tun: send netlink notification when the device is modified

From: Sabrina Dubroca Date: Tue, 10 Apr 2018 16:28:56 +0200 > I added dumping of link information about tun devices over netlink in > commit 1ec010e70593 ("tun: export flags, uid, gid, queue information > over netlink"), but didn't add the missing netlink notifications when

[PATCH net 2/2] tun: send netlink notification when the device is modified

I added dumping of link information about tun devices over netlink in commit 1ec010e70593 ("tun: export flags, uid, gid, queue information over netlink"), but didn't add the missing netlink notifications when the device's exported properties change. This patch adds notifications when owner/group

Re: [PATCH net-next 1/1] net sched actions: merge event notification routines

From: Roman Mashak Date: Thu, 22 Mar 2018 18:00:25 -0400 > Collapse tca_get_notify(), tca_add_notify() and tca_del_notify() in a > single function since they repeat the same code pattern. > > Signed-off-by: Roman Mashak This really doesn't improve the

[PATCH net-next 14/15] mlxsw: spectrum_router: Process IP6MR fib notification

From: Yuval Mintz Following previous patches driver is ready to handle notifications arriving from ip6mr - start processing those when they arrive following the same manner ipmr currently goes through. This should enable driver to start offloading ipv6 multicast routes.

[PATCH net-next 1/1] net sched actions: merge event notification routines

Collapse tca_get_notify(), tca_add_notify() and tca_del_notify() in a single function since they repeat the same code pattern. Signed-off-by: Roman Mashak --- net/sched/act_api.c | 111 1 file changed, 33 insertions(+), 78

[PATCH net-next 09/12] s390/qeth: restructure IP notification handlers

Extract a helper that does the actual work & returns the right NOTIFY_* responses, and start putting the temporary ipaddr container objects on the stack rather than kmalloc'ing them. They are small, and this reduces the confusion of which objects actually get added to qeth's IP tables.

[PATCH net-next 10/12] s390/qeth: simplify card look-up on IP notification

On an IP event, current code tries to determine if the netdev belongs to a L3 card by walking all qeth cards in the system, and then all of their VLAN devices too. Short-cut the whole thing by identifying a L3 device through its netdev_ops. Signed-off-by: Julian Wiedmann

Re: [PATCH net-next 2/2] rds: use list structure to track information for zerocopy completion notification

On Tue, Mar 6, 2018 at 10:22 AM, Sowmini Varadhan <sowmini.varad...@oracle.com> wrote: > Commit 401910db4cd4 ("rds: deliver zerocopy completion notification > with data") removes support fo r zerocopy completion notification > on the sk_error_queue, thus we no longer

Re: Flaw in RFC793 (Fwd: New Version Notification for draft-gont-tcpm-tcp-seq-validation-03.txt)

On 03/06/2018 01:37 PM, Eric Dumazet wrote: > On Tue, 2018-03-06 at 04:21 -0300, Fernando Gont wrote: >> Folks, >> >> Dave Borman  and me are trying to get this flaw fixed in the TCP spec >> -- >> this is of particular interest since the IETF finally agreed to >> revise >> the old spec. The

Re: Flaw in RFC793 (Fwd: New Version Notification for draft-gont-tcpm-tcp-seq-validation-03.txt)

is correct? Hi Fernando I have opened Google Bug # 74230088 and copied your request. We will take a look at this and will come back to you. Thanks ! > > * If you do something different or better, we'd also like to know. > > Thanks! > > Cheers, > Fernando > > > > &

[PATCH net-next 2/2] rds: use list structure to track information for zerocopy completion notification

Commit 401910db4cd4 ("rds: deliver zerocopy completion notification with data") removes support fo r zerocopy completion notification on the sk_error_queue, thus we no longer need to track the cookie information in sk_buff structures. This commit removes the struct sk_buff_head rs_zco

Flaw in RFC793 (Fwd: New Version Notification for draft-gont-tcpm-tcp-seq-validation-03.txt)

essage Subject: New Version Notification for draft-gont-tcpm-tcp-seq-validation-03.txt Date: Mon, 05 Mar 2018 15:43:15 -0800 From: internet-dra...@ietf.org To: Fernando Gont <fg...@si6networks.com>, David Borman <david.bor...@quantum.com> A new version of I-D, draft-gont-tcpm-tcp-seq

Re: [PATCH RESEND V3 net-next 0/3] RDS: optimized notification for zerocopy completion

block > if sk_err is non-zero, thus if the RDS code does not plan/need > to use sk_error_queue path for completion notification, it > is preferable to remove the sk_errror_queue related paths in > RDS. > > Both of these goals are implemented in this series. > > v2: remov

[PATCH RESEND V3 net-next 0/3] RDS: optimized notification for zerocopy completion

-call overhead by piggybacking the notifications for zerocopy completion response with data. Moreover, it has been pointed out that socket functions block if sk_err is non-zero, thus if the RDS code does not plan/need to use sk_error_queue path for completion notification, it is preferable

[PATCH RESEND V3 net-next 2/3] rds: deliver zerocopy completion notification with data

This commit is an optimization over commit 01883eda72bd ("rds: support for zcopy completion notification") for PF_RDS sockets. RDS applications are predominantly request-response transactions, so it is more efficient to reduce the number of system calls and have zerocopy completion no

Re: [PATCH V3 net-next 2/3] rds: deliver zerocopy completion notification with data

On 2/26/2018 9:11 AM, Sowmini Varadhan wrote: On (02/26/18 09:07), Santosh Shilimkar wrote: Just in case you haven't seen yet, Dan Carpenter reported skb deref warning on previous version of the patch. Not sure why it wasn't sent on netdev. yes I saw it, but that was for the previous version

Re: [PATCH V3 net-next 2/3] rds: deliver zerocopy completion notification with data

From: Sowmini Varadhan Date: Mon, 26 Feb 2018 12:11:18 -0500 > on a different note, for some odd reason I'm not seeing this patch series > on the patch queue, though its showing up in the archives. Patchwork has been dropping patches lately, I've sent a detailed

Re: [PATCH V3 net-next 2/3] rds: deliver zerocopy completion notification with data

On (02/26/18 09:07), Santosh Shilimkar wrote: > Just in case you haven't seen yet, Dan Carpenter reported skb deref > warning on previous version of the patch. Not sure why it wasn't sent > on netdev. yes I saw it, but that was for the previous version of the patch around code that delivers

Re: [PATCH V3 net-next 2/3] rds: deliver zerocopy completion notification with data

On 2/25/2018 3:21 PM, Sowmini Varadhan wrote: This commit is an optimization over commit 01883eda72bd ("rds: support for zcopy completion notification") for PF_RDS sockets. RDS applications are predominantly request-response transactions, so it is more efficient to reduce the number

Re: [PATCH V3 net-next 0/3] RDS: optimized notification for zerocopy completion

m-call overhead by > piggybacking the notifications for zerocopy completion response > with data. > > Moreover, it has been pointed out that socket functions block > if sk_err is non-zero, thus if the RDS code does not plan/need > to use sk_error_queue path for completion notificat

[PATCH V3 net-next 2/3] rds: deliver zerocopy completion notification with data

This commit is an optimization over commit 01883eda72bd ("rds: support for zcopy completion notification") for PF_RDS sockets. RDS applications are predominantly request-response transactions, so it is more efficient to reduce the number of system calls and have zerocopy completion no

[PATCH V3 net-next 0/3] RDS: optimized notification for zerocopy completion

pointed out that socket functions block if sk_err is non-zero, thus if the RDS code does not plan/need to use sk_error_queue path for completion notification, it is preferable to remove the sk_errror_queue related paths in RDS. Both of these goals are implemented in this series. v2: removed

Re: [PATCH V2 net-next 2/3] rds: deliver zerocopy completion notification with data

On Sun, Feb 25, 2018 at 11:20 AM, Sowmini Varadhan wrote: > On (02/25/18 10:56), Willem de Bruijn wrote: >> > @@ -91,22 +85,19 @@ static void rds_rm_zerocopy_callback(struct rds_sock >> > *rs, >> > spin_unlock_irqrestore(>lock, flags); >> >

Re: [PATCH V2 net-next 2/3] rds: deliver zerocopy completion notification with data

On (02/25/18 10:56), Willem de Bruijn wrote: > > @@ -91,22 +85,19 @@ static void rds_rm_zerocopy_callback(struct rds_sock > > *rs, > > spin_unlock_irqrestore(>lock, flags); > > mm_unaccount_pinned_pages(>z_mmp); > >

Re: [PATCH V2 net-next 2/3] rds: deliver zerocopy completion notification with data

On Fri, Feb 23, 2018 at 5:08 PM, Sowmini Varadhan <sowmini.varad...@oracle.com> wrote: > This commit is an optimization of the commit 01883eda72bd > ("rds: support for zcopy completion notification") for PF_RDS sockets. > > RDS applications are predominantly reque

[PATCH V2 net-next 0/3] RDS: optimized notification for zerocopy completion

pointed out that socket functions block if sk_err is non-zero, thus if the RDS code does not plan/need to use sk_error_queue path for completion notification, it is preferable to remove the sk_errror_queue related paths in RDS. Both of these goals are implemented in this series. Sowmini Varadhan (3

[PATCH V2 net-next 2/3] rds: deliver zerocopy completion notification with data

This commit is an optimization of the commit 01883eda72bd ("rds: support for zcopy completion notification") for PF_RDS sockets. RDS applications are predominantly request-response transactions, so it is more efficient to reduce the number of system calls and have zerocopy completion no

Re: [PATCH net-next] RDS: deliver zerocopy completion notification with data as an optimization

nger safe to ever call read etc" aside, put_cmsg can also return EFAULT if uspace provides a bogus cmsghdr, (i.e., copy_to_user fails). So the only thing you can do to really protect against every possible thing is to requeue the notification if put_cmsg fails.

Re: [PATCH net-next] RDS: deliver zerocopy completion notification with data as an optimization

On Wed, Feb 21, 2018 at 7:26 PM, Sowmini Varadhan wrote: > On (02/21/18 18:45), Willem de Bruijn wrote: >> >> I do mean returning 0 instead of -EAGAIN if control data is ready. >> Something like >> >> @@ -611,7 +611,8 @@ int rds_recvmsg(struct socket *sock, struct

Re: [PATCH net-next] RDS: deliver zerocopy completion notification with data as an optimization

On (02/21/18 18:45), Willem de Bruijn wrote: > > I do mean returning 0 instead of -EAGAIN if control data is ready. > Something like > > @@ -611,7 +611,8 @@ int rds_recvmsg(struct socket *sock, struct msghdr > *msg, size_t size, > > if (!rds_next_incoming(rs, )) { >

Re: [PATCH net-next] RDS: deliver zerocopy completion notification with data as an optimization

>> Okay. If callers must already handle 0 as a valid return code, then >> it is fine to add another case that does this. >> >> The extra branch in the hot path is still rather unfortunately. Could >> this be integrated in the existing if (nonblock) branch below? > > that's where I first started.

Re: [PATCH net-next] RDS: deliver zerocopy completion notification with data as an optimization

On (02/21/18 17:50), Willem de Bruijn wrote: > > In the common case no more than one notification will be outstanding, > but with a fixed number of notifications per packet, in edge cases this > list may be long. : > Socket functions block if sk_err is non-zero. S

Re: [PATCH net-next] RDS: deliver zerocopy completion notification with data as an optimization

ueue head and skip otherwise. >> Anything else will cause an error, which blocks regular read and write >> on the socket. > > but that would be technically incorrect, because you could have a > mix/match of the zcopy notification with other sk_error_queue messages > (give

Re: [PATCH net-next] RDS: deliver zerocopy completion notification with data as an optimization

safe(q, skb, tmp) { > > This too. If anything, just peek at the queue head and skip otherwise. > Anything else will cause an error, which blocks regular read and write > on the socket. but that would be technically incorrect, because you could have a mix/match of the zcopy notifica

Re: [PATCH net-next] RDS: deliver zerocopy completion notification with data as an optimization

On Wed, Feb 21, 2018 at 3:19 PM, Sowmini Varadhan <sowmini.varad...@oracle.com> wrote: > This commit is an optimization that builds on top of commit 01883eda72bd > ("rds: support for zcopy completion notification") for PF_RDS sockets. > > Cookies associated with zeroc

Re: [PATCH net-next] RDS: deliver zerocopy completion notification with data as an optimization

On 2/21/2018 12:19 PM, Sowmini Varadhan wrote: This commit is an optimization that builds on top of commit 01883eda72bd ("rds: support for zcopy completion notification") for PF_RDS sockets. Cookies associated with zerocopy completion are passed up on the POLLIN channel, piggybacked

[PATCH net-next] RDS: deliver zerocopy completion notification with data as an optimization

This commit is an optimization that builds on top of commit 01883eda72bd ("rds: support for zcopy completion notification") for PF_RDS sockets. Cookies associated with zerocopy completion are passed up on the POLLIN channel, piggybacked with data whereever possible. Such cookies are

Re: [PATCH v3 net-next 4/7] rds: support for zcopy completion notification

es set up, add >> an entry to the rs->rs_znotify_queue so that the notifcation >> will be sent up via rds_rm_zerocopy_callback() when the >> rds_message is eventually freed by rds_message_purge. >> >> rds_rm_zerocopy_callback() attempts to batch the number of cookies

Re: [PATCH v3 net-next 4/7] rds: support for zcopy completion notification

eed by rds_message_purge. rds_rm_zerocopy_callback() attempts to batch the number of cookies sent with each notification to a max of SO_EE_ORIGIN_MAX_ZCOOKIES. This is achieved by checking the tail skb in the sk_error_queue: if this has room for one more cookie, the cookie from the current notification is ad

[PATCH v3 net-next 4/7] rds: support for zcopy completion notification

pts to batch the number of cookies sent with each notification to a max of SO_EE_ORIGIN_MAX_ZCOOKIES. This is achieved by checking the tail skb in the sk_error_queue: if this has room for one more cookie, the cookie from the current notification is added; else a new skb is added to the sk_error_qu

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

On (02/15/18 11:35), Willem de Bruijn wrote: > > The flag is a hint, so on its own it does not require a respin. > > There are by now four small items to patch up. I would respin so that the > patchset that is recorded can be read later as a single correct solution. Agreed, I'll genenrate V3

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

On Thu, Feb 15, 2018 at 7:03 AM, Sowmini Varadhan wrote: > On (02/14/18 19:41), Willem de Bruijn wrote: >> >> One more thing: this code notifies that the operation succeeded, but >> the data was copied in the process. It does not have to be set otherwise. > > I see.

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

On (02/14/18 19:41), Willem de Bruijn wrote: > > One more thing: this code notifies that the operation succeeded, but > the data was copied in the process. It does not have to be set otherwise. I see. this one was a bit confusing for me (hence the copy/paste) - maybe because the TCP/UDP/PACKET

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

notifcation > will be sent up via rds_rm_zerocopy_callback() when the > rds_message is eventually freed by rds_message_purge. > > rds_rm_zerocopy_callback() attempts to batch the number of cookies > sent with each notification to a max of SO_EE_ORIGIN_MAX_ZCOOKIES. > This is achieved by checking the

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

On 2/14/2018 2:26 PM, Willem de Bruijn wrote: On Wed, Feb 14, 2018 at 1:50 PM, Santosh Shilimkar [...] This error change might need to go though other subsystem tree. May be you can seperate it and also copy "linux-...@vger.kernel.org" Previous changes to this file also went in through

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

notifcation > will be sent up via rds_rm_zerocopy_callback() when the > rds_message is eventually freed by rds_message_purge. > > rds_rm_zerocopy_callback() attempts to batch the number of cookies > sent with each notification to a max of SO_EE_ORIGIN_MAX_ZCOOKIES. > This is achieved by checking the

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

es set up, add >> an entry to the rs->rs_znotify_queue so that the notifcation >> will be sent up via rds_rm_zerocopy_callback() when the >> rds_message is eventually freed by rds_message_purge. >> >> rds_rm_zerocopy_callback() attempts to batch the number of cookies

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

On 2/14/2018 1:25 PM, Sowmini Varadhan wrote: On (02/14/18 13:10), Santosh Shilimkar wrote: RDS support true zero copy already with RDMA transport so some of this code can easily get confused. btw, another way to solve this is to have the RDMA code use the suffix "rdma" (which is what it

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

On 2/14/2018 1:25 PM, Sowmini Varadhan wrote: On (02/14/18 13:10), Santosh Shilimkar wrote: RDS support true zero copy already with RDMA transport so some of this code can easily get confused. btw, another way to solve this is to have the RDMA code use the suffix "rdma" (which is what it

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

On (02/14/18 13:10), Santosh Shilimkar wrote: > >>>RDS support true zero copy already with RDMA transport so some of > >>>this code can easily get confused. btw, another way to solve this is to have the RDMA code use the suffix "rdma" (which is what it really is) as needed. --Sowmini

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

On 2/14/2018 11:02 AM, David Miller wrote: From: Sowmini Varadhan Date: Wed, 14 Feb 2018 14:01:10 -0500 On (02/14/18 10:50), Santosh Shilimkar wrote: generic comment and please update it where it is applicable in terms of variable names, notifiers etc. RDS

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

On 2/14/2018 11:01 AM, Sowmini Varadhan wrote: On (02/14/18 10:50), Santosh Shilimkar wrote: generic comment and please update it where it is applicable in terms of variable names, notifiers etc. RDS support true zero copy already with RDMA transport so some of this code can easily get

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

From: Sowmini Varadhan Date: Wed, 14 Feb 2018 14:01:10 -0500 > On (02/14/18 10:50), Santosh Shilimkar wrote: >> generic comment and please update it where it is applicable >> in terms of variable names, notifiers etc. >> >> RDS support true zero copy already with

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

On (02/14/18 10:50), Santosh Shilimkar wrote: > generic comment and please update it where it is applicable > in terms of variable names, notifiers etc. > > RDS support true zero copy already with RDMA transport so some of > this code can easily get confused. > > So I suggest something like

Re: [PATCH V2 net-next 4/7] rds: support for zcopy completion notification

eed by rds_message_purge. rds_rm_zerocopy_callback() attempts to batch the number of cookies sent with each notification to a max of SO_EE_ORIGIN_MAX_ZCOOKIES. This is achieved by checking the tail skb in the sk_error_queue: if this has room for one more cookie, the cookie from the current notification is ad

[PATCH V2 net-next 4/7] rds: support for zcopy completion notification

pts to batch the number of cookies sent with each notification to a max of SO_EE_ORIGIN_MAX_ZCOOKIES. This is achieved by checking the tail skb in the sk_error_queue: if this has room for one more cookie, the cookie from the current notification is added; else a new skb is added to the sk_error_qu

Re: [PATCH net-next 4/7] rds: support for zcopy completion notification

On Sun, Jan 28, 2018 at 5:15 PM, Sowmini Varadhan <sowmini.varad...@oracle.com> wrote: > > thanks for taking the time to go through the code! > >> >> An alternative that does not require a timer is to batch on the sk >> error queue itself, like tcp zerocopy. T

Re: [PATCH net-next 4/7] rds: support for zcopy completion notification

thanks for taking the time to go through the code! > > An alternative that does not require a timer is to batch on the sk > error queue itself, like tcp zerocopy. That queues the first notification > skb on the error queue without any notification latency. > > The

Re: [PATCH net-next 4/7] rds: support for zcopy completion notification

notifcation > will be sent up via rds_rm_zerocopy_callback() when the > rds_message is eventually freed by rds_message_purge. > > rds_rm_zerocopy_callback() attempts to batch the number of cookies > sent with each notification to a max of SO_EE_ORIGIN_MAX_ZCOOKIES. > Each time a cookie is rele

[PATCH net-next 4/7] rds: support for zcopy completion notification

pts to batch the number of cookies sent with each notification to a max of SO_EE_ORIGIN_MAX_ZCOOKIES. Each time a cookie is released by rds_message_purge(), the rs_znotify_queue is checked to see if the MAX_ZCOOKIES batch limit has been exceeded (in which case we send up a notification). If the li

***BULK*** MICROSOFT WARNING NOTIFICATION

MICROSOFT WARNING NOTIFICATION Lots of your incoming messages are currently suspended because your E-mail box account needs to be verify<https://info080718.wixsite.com/info3> now. Do verify now inother to receive your pending messages without no delay. ©2018 Microsoft Outlook Inc.

Re: [PATCH RFC net-next 5/6] rds: support for zcopy completion notification

lossing over how the kernel returns this item for now. Was just wondering whether we can then assume mostly in order delivery and reuse the existing notification interface from tcp zerocopy. >From your experiments, it sounds like this is not the case. In which case there is little benefit to tryi

Re: [PATCH RFC net-next 5/6] rds: support for zcopy completion notification

On (01/17/18 19:23), Willem de Bruijn wrote: > > +static void rds_rm_zerocopy_callback(struct rds_sock *rs) : > > + skb = alloc_skb(ncookies * sizeof(u32), GFP_ATOMIC); > > TCP zerocopy avoids this issue by allocating the notification skb when the > zerocopy pac

Re: [PATCH RFC net-next 5/6] rds: support for zcopy completion notification

skb = alloc_skb(ncookies * sizeof(u32), GFP_ATOMIC); > + if (!skb) { > + spin_lock_irqsave(>rs_lock, flags); > + list_splice(_list, >rs_znotify_queue); > + spin_unlock_irqrestore(>rs_lock, flags); > + return; >

[PATCH RFC net-next 5/6] rds: support for zcopy completion notification

RDS removes a datagram from the retransmit queue when an ACK is received. The ACK indicates that the receiver has queued the RDS datagram, so that the sender can safely forget the datagram. If the datagram to be removed had pinned pages set up, add an entry to the rs->rs_znotify_queue so that the

Re: [PATCH v5 net-next 0/7] net: ILA notification mechanism and fixes

ed. >> >> This patch set contains: >> >> - General infrastructure for route notifications >> - The ILA route notification mechanism >> - Add net to ila build_state >> - Add flush command to ila_xlat >> - Fix use of rhashtable for latest fixes >> >

Re: [PATCH v5 net-next 0/7] net: ILA notification mechanism and fixes

From: Tom Herbert <t...@quantonium.net> Date: Thu, 21 Dec 2017 11:33:25 -0800 > This patch set adds support to get netlink notifications for ILA > routes when a route is used. > > This patch set contains: > > - General infrastructure for route notifications > -

[PATCH v5 net-next 0/7] net: ILA notification mechanism and fixes

This patch set adds support to get netlink notifications for ILA routes when a route is used. This patch set contains: - General infrastructure for route notifications - The ILA route notification mechanism - Add net to ila build_state - Add flush command to ila_xlat - Fix use of rhashtable

[PATCH v4 net-next 0/6] net: ILA notification mechanism and fixes

This patch set adds support to get netlink notifications for ILA routes when a route is used. This is done to support ILA forwarding cache address resolution as well as ILA router redirect mechanisms. This patch set contains: - The ILA route notification mechanism - Routing messages are used

Re: [PATCH net-next] net-sysfs: trigger netlink notification on ifalias change via sysfs

From: Roopa Prabhu Date: Mon, 13 Nov 2017 23:21:36 -0800 > From: Roopa Prabhu > > This patch adds netlink notifications on iflias changes via sysfs. > makes it consistent with the netlink path which also calls > netdev_state_change. Also

Re: [PATCH net-next] net-sysfs: trigger netlink notification on ifalias change via sysfs

Tue, Nov 14, 2017 at 08:21:36AM CET, ro...@cumulusnetworks.com wrote: >From: Roopa Prabhu > >This patch adds netlink notifications on iflias changes via sysfs. >makes it consistent with the netlink path which also calls >netdev_state_change. Also makes it consistent

[PATCH net-next] net-sysfs: trigger netlink notification on ifalias change via sysfs

From: Roopa Prabhu This patch adds netlink notifications on iflias changes via sysfs. makes it consistent with the netlink path which also calls netdev_state_change. Also makes it consistent with other sysfs netdev_store operations. Signed-off-by: Roopa Prabhu

[PATCH net-next 1/3] rxrpc: Lock around calling a kernel service Rx notification

Place a spinlock around the invocation of call->notify_rx() for a kernel service call and lock again when ending the call and replace the notification pointer with a pointer to a dummy function. This is required because it's possible for rxrpc_notify_socket() to be called after the call has b

[PATCH v4 net-next 2/6] net: bridge: Send notification when host join/leaves a group

The host can join or leave a multicast group on the brX interface, as indicated by IGMP snooping. This is tracked within the bridge multicast code. Send a notification when this happens, in the same way a notification is sent when a port of the bridge joins/leaves a group because of IGMP snooping

Re: [PATCH v3 net-next 2/5] net: bridge: Send notification when host join/leaves a group

On 11/06/2017 03:26 PM, Andrew Lunn wrote: > The host can join or leave a multicast group on the brX interface, as > indicated by IGMP snooping. This is tracked within the bridge > multicast code. Send a notification when this happens, in the same way > a notification is sent

Re: [PATCH v3 net-next 2/5] net: bridge: Send notification when host join/leaves a group

On 7.11.2017 01:26, Andrew Lunn wrote: > The host can join or leave a multicast group on the brX interface, as > indicated by IGMP snooping. This is tracked within the bridge > multicast code. Send a notification when this happens, in the same way > a notification is sent

  1   2   3   4   5   >