platform/kernel/linux-rpi.git
13 months agosfc: Add devlink dev info support for EF10
Martin Habets [Fri, 9 Jun 2023 07:57:36 +0000 (08:57 +0100)]
sfc: Add devlink dev info support for EF10

Reuse the work done for EF100 to add devlink support for EF10.
There is no devlink port support for EF10.

Signed-off-by: Martin Habets <habetsm.xilinx@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agonet/sched: act_pedit: Use kmemdup() to replace kmalloc + memcpy
Jiapeng Chong [Fri, 9 Jun 2023 07:01:17 +0000 (15:01 +0800)]
net/sched: act_pedit: Use kmemdup() to replace kmalloc + memcpy

./net/sched/act_pedit.c:245:21-28: WARNING opportunity for kmemdup.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5478
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Reviewed-by: Pedro Tammela <pctammela@mojatatu.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agoionic: add support for ethtool extended stat link_down_count
Nitya Sunkad [Fri, 9 Jun 2023 05:50:16 +0000 (22:50 -0700)]
ionic: add support for ethtool extended stat link_down_count

Following the example of 'commit 9a0f830f8026 ("ethtool: linkstate:
add a statistic for PHY down events")', added support for link down
events.

Add callback ionic_get_link_ext_stats to ionic_ethtool.c to support
link_down_count, a property of netdev that gets reported exclusively
on physical link down events.

Run ethtool -I <devname> to display the device link down count.

Signed-off-by: Nitya Sunkad <nitya.sunkad@amd.com>
Signed-off-by: Shannon Nelson <shannon.nelson@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agoMerge branch '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next...
David S. Miller [Mon, 12 Jun 2023 07:52:09 +0000 (08:52 +0100)]
Merge branch '100GbE' of git://git./linux/kernel/git/tnguy/next-queue

Tony Nguyen says:

====================
ice: Improve miscellaneous interrupt code

Jacob Keller says:

This series improves the driver's use of the threaded IRQ and the
communication between ice_misc_intr() and the ice_misc_intr_thread_fn()
which was previously introduced by commit 1229b33973c7 ("ice: Add low
latency Tx timestamp read").

First, a new custom enumerated return value is used instead of a boolean for
ice_ptp_process_ts(). This significantly reduces the cognitive burden when
reviewing the logic for this function, as the expected action is clear from
the return value name.

Second, the unconditional loop in ice_misc_intr_thread_fn() is removed,
replacing it with a write to the Other Interrupt Cause register. This causes
the MAC to trigger the Tx timestamp interrupt again. This makes it possible
to safely use the ice_misc_intr_thread_fn() to handle other tasks beyond
just the Tx timestamps. It is also easier to reason about since the thread
function will exit cleanly if we do something like disable the interrupt and
call synchronize_irq().

Third, refactor the handling for external timestamp events to use the
miscellaneous thread function. This resolves an issue with the external
time stamps getting blocked while processing the periodic work function
task.

Fourth, a simplification of the ice_misc_intr() function to always return
IRQ_WAKE_THREAD, and schedule the ice service task in the
ice_misc_intr_thread_fn() instead.

Finally, the Other Interrupt Cause is kept disabled over the thread function
processing, rather than immediately re-enabled.

Special thanks to Michal Schmidt for the careful review of the series and
pointing out my misunderstandings of the kernel IRQ code. It has been
determined that the race outlined as being fixed in previous series was
actually introduced by this series itself, which I've since corrected.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agonet: wwan: iosm: enable runtime pm support for 7560
M Chetan Kumar [Thu, 8 Jun 2023 10:08:03 +0000 (15:38 +0530)]
net: wwan: iosm: enable runtime pm support for 7560

Adds runtime pm support for 7560.

As part of probe procedure auto suspend is enabled and auto suspend
delay is set to 5000 ms for runtime pm use. Later auto flag is set
to power manage the device at run time.

On successful communication establishment between host and device the
device usage counter is dropped and request to put the device into
sleep state (suspend).

In TX path, the device usage counter is raised and device is moved out
of sleep(resume) for data transmission. In RX path, if the device has
some data to be sent it request host platform to change the power state
by giving PCI PME message.

Signed-off-by: M Chetan Kumar <m.chetan.kumar@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agodt-bindings: net: xlnx,axi-ethernet: convert bindings document to yaml
Radhey Shyam Pandey [Thu, 8 Jun 2023 08:24:58 +0000 (13:54 +0530)]
dt-bindings: net: xlnx,axi-ethernet: convert bindings document to yaml

Convert the bindings document for Xilinx AXI Ethernet Subsystem
from txt to yaml. No changes to existing binding description.

Signed-off-by: Radhey Shyam Pandey <radhey.shyam.pandey@xilinx.com>
Signed-off-by: Sarath Babu Naidu Gaddam <sarath.babu.naidu.gaddam@amd.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agoselftests: net: vxlan: Fix selftest regression after changes in iproute2.
Vladimir Nikishkin [Thu, 8 Jun 2023 06:44:48 +0000 (14:44 +0800)]
selftests: net: vxlan: Fix selftest regression after changes in iproute2.

The iproute2 output that eventually landed upstream is different than
the one used in this test, resulting in failures. Fix by adjusting the
test to use iproute2's JSON output, which is more stable than regular
output.

Fixes: 305c04189997 ("selftests: net: vxlan: Add tests for vxlan nolocalbypass option.")
Signed-off-by: Vladimir Nikishkin <vladimir@nikishkin.pw>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Tested-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agoMerge branch 'renesas-rswitch-perf'
David S. Miller [Sat, 10 Jun 2023 18:49:34 +0000 (19:49 +0100)]
Merge branch 'renesas-rswitch-perf'

Yoshihiro Shimoda says:

====================
net: renesas: rswitch: Improve perfromance of TX/RX

This patch series is based on net-next.git / main branch [1]. This patch
series can improve perfromance of TX in a specific condition. The previous code
used "global rate limiter" feature so that this is possible to cause
performance down if we use multiple ports at the same time. To resolve this
issue, use "hardware pause" features of GWCA and COMA. Note that this is not
related to the ethernet PAUSE frames.

< UDP TX by iperf3 >
 before: about 450Mbps on both tsn0 and tsn1
 after:  about 950Mbps on both tsn0 and tsn1

Also, this patch series can improve performance of RX by using
napi_gro_receive().

< TCP RX by iperf >
 before: about 670Mbps on tsn0
 after:  about 840Mbps on tsn0

[1]
The commit e06bd5e3adae ("Merge branch 'followup-fixes-for-the-dwmac-and-altera-lynx-conversion'")

Changes from v3:
https://lore.kernel.org/all/20230607015641.1724057-1-yoshihiro.shimoda.uh@renesas.com/
 - Rebased on the latest net-next.git / main branch.
 - Added Reviewed-by in the patch 2/2. (Maciej, thanks!)
 - Fix typos in the commit description in the patch 2/2.

Changes from v2:
https://lore.kernel.org/all/20230606085558.1708766-1-yoshihiro.shimoda.uh@renesas.com/
 - Rebased on the latest net-next.git / main branch.
 - Added Reviewed-by in the patch 1/2. (Maciej, thanks!)
 - Revise the commit description in the patch 2/2.
 - Add definition to remove magic hardcoded numbers in the patch 2/2.

Changes from v1:
https://lore.kernel.org/all/20230529080840.1156458-1-yoshihiro.shimoda.uh@renesas.com/
 - Rebased on the latest net-next.git / main branch.
 - Use "hardware pause" feature instead of "per-queue limiter" feature.
 - Drop refactaring for "per-queue limiter".
 - Drop dt-bindings update because "hardware pause" doesn't need additional
   clock information.
 - Use napi_gro_receive() to improve RX performance.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agonet: renesas: rswitch: Use hardware pause features
Yoshihiro Shimoda [Thu, 8 Jun 2023 02:20:07 +0000 (11:20 +0900)]
net: renesas: rswitch: Use hardware pause features

Since this driver used the "global rate limiter" feature of GWCA,
the TX performance of each port was reduced when multiple ports
transmitted frames simultaneously. To improve performance, remove
the use of the "global rate limiter" feature and use "hardware pause"
features of the following:
 - "per priority pause" of GWCA
 - "global pause" of COMA

Note that these features are not related to the ethernet PAUSE frame.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agonet: renesas: rswitch: Use napi_gro_receive() in RX
Yoshihiro Shimoda [Thu, 8 Jun 2023 02:20:06 +0000 (11:20 +0900)]
net: renesas: rswitch: Use napi_gro_receive() in RX

This hardware can receive multiple frames so that using
napi_gro_receive() instead of netif_receive_skb() gets good
performance of RX.

Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Reviewed-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agoMerge branch 'sfc-tc-encap-actions-offload'
Jakub Kicinski [Sat, 10 Jun 2023 07:11:51 +0000 (00:11 -0700)]
Merge branch 'sfc-tc-encap-actions-offload'

Edward Cree says:

====================
sfc: TC encap actions offload

This series adds support for offloading TC tunnel_key set actions to the
 EF100 driver, supporting VxLAN and GENEVE tunnels over IPv4 or IPv6.
====================

Link: https://lore.kernel.org/r/cover.1686240142.git.ecree.xilinx@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agosfc: generate encap headers for TC offload
Edward Cree [Thu, 8 Jun 2023 16:42:35 +0000 (17:42 +0100)]
sfc: generate encap headers for TC offload

Support constructing VxLAN and GENEVE headers, on either IPv4 or IPv6,
 using the neighbouring information obtained in encap->neigh to
 populate the Ethernet header.
Note that the ef100 hardware does not insert UDP checksums when
 performing encap, so for IPv6 the remote endpoint will need to be
 configured with udp6zerocsumrx or equivalent.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Pieter Jansen van Vuuren <pieter.jansen-van-vuuren@amd.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agosfc: neighbour lookup for TC encap action offload
Edward Cree [Thu, 8 Jun 2023 16:42:34 +0000 (17:42 +0100)]
sfc: neighbour lookup for TC encap action offload

For each neighbour we're interested in, create a struct efx_neigh_binder
 object which has a list of all the encap_actions using it.  When we
 receive a neighbouring update (through the netevent notifier), find the
 corresponding efx_neigh_binder and update all its users.
Since the actual generation of encap headers is still only a stub, the
 resulting rules still get left on fallback actions.

Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Pieter Jansen van Vuuren <pieter.jansen-van-vuuren@amd.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agosfc: MAE functions to create/update/delete encap headers
Edward Cree [Thu, 8 Jun 2023 16:42:33 +0000 (17:42 +0100)]
sfc: MAE functions to create/update/delete encap headers

Besides the raw header data, also pass the tunnel type, so that the
 hardware knows it needs to update the IP Total Length and UDP Length
 fields (and corresponding checksums) for each packet.
Also, populate the ENCAP_HEADER_ID field in efx_mae_alloc_action_set()
 with the fw_id returned from efx_mae_allocate_encap_md().

Reviewed-by: Pieter Jansen van Vuuren <pieter.jansen-van-vuuren@amd.com>
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agosfc: add function to atomically update a rule in the MAE
Edward Cree [Thu, 8 Jun 2023 16:42:32 +0000 (17:42 +0100)]
sfc: add function to atomically update a rule in the MAE

efx_mae_update_rule() changes the action-set-list attached to an MAE
 flow rule in the Action Rule Table.
We will use this when neighbouring updates change encap actions.

Reviewed-by: Pieter Jansen van Vuuren <pieter.jansen-van-vuuren@amd.com>
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agosfc: some plumbing towards TC encap action offload
Edward Cree [Thu, 8 Jun 2023 16:42:31 +0000 (17:42 +0100)]
sfc: some plumbing towards TC encap action offload

Create software objects to manage the metadata for encap actions that
 can be attached to TC rules.  However, since we don't yet have the
 neighbouring information (needed to generate the Ethernet header),
 all rules with encap actions are marked as "unready" and thus insert
 the fallback action into hardware rather than actually offloading the
 encapsulation action.

Reviewed-by: Pieter Jansen van Vuuren <pieter.jansen-van-vuuren@amd.com>
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agosfc: add fallback action-set-lists for TC offload
Edward Cree [Thu, 8 Jun 2023 16:42:30 +0000 (17:42 +0100)]
sfc: add fallback action-set-lists for TC offload

When offloading a TC encap action, the action information for the
 hardware might not be "ready": if there's currently no neighbour entry
 available for the destination address, we can't construct the Ethernet
 header to prepend to the packet.  In this case, we still offload the
 flow rule, but with its action-set-list ID pointing at a "fallback"
 action which simply delivers the packet to its default destination (as
 though no flow rule had matched), thus allowing software TC to handle
 it.  Later, when we receive a neighbouring update that allows us to
 construct the encap header, the rule will become "ready" and we will
 update its action-set-list ID in hardware to point at the actual
 offloaded actions.
This patch sets up these fallback ASLs, but does not yet use them.

Reviewed-by: Pieter Jansen van Vuuren <pieter.jansen-van-vuuren@amd.com>
Signed-off-by: Edward Cree <ecree.xilinx@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonet: move gso declarations and functions to their own files
Eric Dumazet [Thu, 8 Jun 2023 19:17:37 +0000 (19:17 +0000)]
net: move gso declarations and functions to their own files

Move declarations into include/net/gso.h and code into net/core/gso.c

Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Stanislav Fomichev <sdf@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20230608191738.3947077-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agoMerge branch 'mptcp-unify-pm-interfaces'
Jakub Kicinski [Sat, 10 Jun 2023 07:06:01 +0000 (00:06 -0700)]
Merge branch 'mptcp-unify-pm-interfaces'

Matthieu Baerts says:

====================
mptcp: unify PM interfaces

These patches from Geliang better isolate the two MPTCP path-managers by
avoiding calling userspace PM functions from the in-kernel PM. Instead,
new functions declared in pm.c directly dispatch to the right PM.

In addition to have a clearer code, this also avoids a bit of duplicated
checks.

This is a refactoring, there is no behaviour change intended here.
====================

Link: https://lore.kernel.org/r/20230608-upstream-net-next-20230608-mptcp-unify-pm-interfaces-v1-0-b301717c9ff5@tessares.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agomptcp: unify pm set_flags interfaces
Geliang Tang [Thu, 8 Jun 2023 13:20:52 +0000 (15:20 +0200)]
mptcp: unify pm set_flags interfaces

This patch unifies the three PM set_flags() interfaces:

mptcp_pm_nl_set_flags() in mptcp/pm_netlink.c for the in-kernel PM and
mptcp_userspace_pm_set_flags() in mptcp/pm_userspace.c for the
userspace PM.

They'll be switched in the common PM infterface mptcp_pm_set_flags() in
mptcp/pm.c based on whether token is NULL or not.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Reviewed-by: Larysa Zaremba <larysa.zaremba@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agomptcp: unify pm get_flags_and_ifindex_by_id
Geliang Tang [Thu, 8 Jun 2023 13:20:51 +0000 (15:20 +0200)]
mptcp: unify pm get_flags_and_ifindex_by_id

This patch unifies the three PM get_flags_and_ifindex_by_id() interfaces:

mptcp_pm_nl_get_flags_and_ifindex_by_id() in mptcp/pm_netlink.c for the
in-kernel PM and mptcp_userspace_pm_get_flags_and_ifindex_by_id() in
mptcp/pm_userspace.c for the userspace PM.

They'll be switched in the common PM infterface
mptcp_pm_get_flags_and_ifindex_by_id() in mptcp/pm.c based on whether
mptcp_pm_is_userspace() or not.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Reviewed-by: Larysa Zaremba <larysa.zaremba@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agomptcp: unify pm get_local_id interfaces
Geliang Tang [Thu, 8 Jun 2023 13:20:50 +0000 (15:20 +0200)]
mptcp: unify pm get_local_id interfaces

This patch unifies the three PM get_local_id() interfaces:

mptcp_pm_nl_get_local_id() in mptcp/pm_netlink.c for the in-kernel PM and
mptcp_userspace_pm_get_local_id() in mptcp/pm_userspace.c for the
userspace PM.

They'll be switched in the common PM infterface mptcp_pm_get_local_id()
in mptcp/pm.c based on whether mptcp_pm_is_userspace() or not.

Also put together the declarations of these three functions in protocol.h.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Reviewed-by: Larysa Zaremba <larysa.zaremba@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agomptcp: export local_address
Geliang Tang [Thu, 8 Jun 2023 13:20:49 +0000 (15:20 +0200)]
mptcp: export local_address

Rename local_address() with "mptcp_" prefix and export it in protocol.h.

This function will be re-used in the common PM code (pm.c) in the
following commit.

Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Reviewed-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Reviewed-by: Larysa Zaremba <larysa.zaremba@intel.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agoMerge tag 'wireless-next-2023-06-09' of git://git.kernel.org/pub/scm/linux/kernel...
Jakub Kicinski [Sat, 10 Jun 2023 06:26:55 +0000 (23:26 -0700)]
Merge tag 'wireless-next-2023-06-09' of git://git./linux/kernel/git/wireless/wireless-next

Kalle Valo says:

====================
wireless-next patches for v6.5

The second pull request for v6.5. We have support for three new
Realtek chipsets, all from different generations. Shows how active
Realtek development is right now, even older generations are being
worked on.

Note: We merged wireless into wireless-next to avoid complex conflicts
between the trees.

Major changes:

rtl8xxxu
 - RTL8192FU support

rtw89
 - RTL8851BE support

rtw88
 - RTL8723DS support

ath11k
 - Multiple Basic Service Set Identifier (MBSSID) and Enhanced MBSSID
   Advertisement (EMA) support in AP mode

iwlwifi
 - support for segmented PNVM images and power tables
 - new vendor entries for PPAG (platform antenna gain) feature

cfg80211/mac80211
 - more Multi-Link Operation (MLO) support such as hardware restart
 - fixes for a potential work/mutex deadlock and with it beginnings of
   the previously discussed locking simplifications

* tag 'wireless-next-2023-06-09' of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless-next: (162 commits)
  wifi: rtlwifi: remove misused flag from HAL data
  wifi: rtlwifi: remove unused dualmac control leftovers
  wifi: rtlwifi: remove unused timer and related code
  wifi: rsi: Do not set MMC_PM_KEEP_POWER in shutdown
  wifi: rsi: Do not configure WoWlan in shutdown hook if not enabled
  wifi: brcmfmac: Detect corner error case earlier with log
  wifi: rtw89: 8852c: update RF radio A/B parameters to R63
  wifi: rtw89: 8852c: update TX power tables to R63 with 6 GHz power type (3 of 3)
  wifi: rtw89: 8852c: update TX power tables to R63 with 6 GHz power type (2 of 3)
  wifi: rtw89: 8852c: update TX power tables to R63 with 6 GHz power type (1 of 3)
  wifi: rtw89: process regulatory for 6 GHz power type
  wifi: rtw89: regd: update regulatory map to R64-R40
  wifi: rtw89: regd: judge 6 GHz according to chip and BIOS
  wifi: rtw89: refine clearing supported bands to check 2/5 GHz first
  wifi: rtw89: 8851b: configure CRASH_TRIGGER feature for 8851B
  wifi: rtw89: set TX power without precondition during setting channel
  wifi: rtw89: debug: txpwr table access only valid page according to chip
  wifi: rtw89: 8851b: enable hw_scan support
  wifi: cfg80211: move scan done work to wiphy work
  wifi: cfg80211: move sched scan stop to wiphy work
  ...
====================

Link: https://lore.kernel.org/r/87bkhohkbg.fsf@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agoMerge branch 'tools-ynl-gen-code-gen-improvements-before-ethtool'
Jakub Kicinski [Fri, 9 Jun 2023 21:40:33 +0000 (14:40 -0700)]
Merge branch 'tools-ynl-gen-code-gen-improvements-before-ethtool'

Jakub Kicinski says:

====================
tools: ynl-gen: code gen improvements before ethtool

I was going to post ethtool but I couldn't stand the ugliness
of the if conditions which were previously generated.
So I cleaned that up and improved a number of other things
ethtool will benefit from.
====================

Link: https://lore.kernel.org/r/20230608211200.1247213-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: support / skip pads on the way to kernel
Jakub Kicinski [Thu, 8 Jun 2023 21:12:00 +0000 (14:12 -0700)]
tools: ynl-gen: support / skip pads on the way to kernel

Kernel does not have padding requirements for 64b attrs.
We can ignore pad attrs.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: don't pass op_name to RenderInfo
Jakub Kicinski [Thu, 8 Jun 2023 21:11:59 +0000 (14:11 -0700)]
tools: ynl-gen: don't pass op_name to RenderInfo

The op_name argument is barely used and identical to op.name
in all cases.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: support code gen for events
Jakub Kicinski [Thu, 8 Jun 2023 21:11:58 +0000 (14:11 -0700)]
tools: ynl-gen: support code gen for events

Netlink specs support both events and notifications (former can
define their own message contents). Plug in missing code to
generate types, parsers and include events into notification
tables.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: sanitize notification tracking
Jakub Kicinski [Thu, 8 Jun 2023 21:11:57 +0000 (14:11 -0700)]
tools: ynl-gen: sanitize notification tracking

Don't modify the raw dicts (as loaded from YAML) to pretend
that the notify attributes also exist on the ops. This makes
the code easier to follow.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl: regen: stop generating common notification handlers
Jakub Kicinski [Thu, 8 Jun 2023 21:11:56 +0000 (14:11 -0700)]
tools: ynl: regen: stop generating common notification handlers

Remove unused notification handlers.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: stop generating common notification handlers
Jakub Kicinski [Thu, 8 Jun 2023 21:11:55 +0000 (14:11 -0700)]
tools: ynl-gen: stop generating common notification handlers

Common notification handler was supposed to be a way for the user
to parse the notifications from a socket synchronously.
I don't think we'll end up using it, ynl_ntf_check() works for
all known use cases.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl: regen: regenerate the if ladders
Jakub Kicinski [Thu, 8 Jun 2023 21:11:54 +0000 (14:11 -0700)]
tools: ynl: regen: regenerate the if ladders

Renegate the code to combine } and else and use tmp variable
to store type.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: get attr type outside of if()
Jakub Kicinski [Thu, 8 Jun 2023 21:11:53 +0000 (14:11 -0700)]
tools: ynl-gen: get attr type outside of if()

Reading attr type with mnl_attr_get_type() for each condition
leads to most conditions being longer than 80 chars.
Avoid this by reading the type to a variable on the stack.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: combine else with closing bracket
Jakub Kicinski [Thu, 8 Jun 2023 21:11:52 +0000 (14:11 -0700)]
tools: ynl-gen: combine else with closing bracket

Code gen currently prints:

  }
  else if (...

This is really ugly. Fix it by delaying printing of closing
brackets in anticipation of else coming along.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: complete the C keyword list
Jakub Kicinski [Thu, 8 Jun 2023 21:11:51 +0000 (14:11 -0700)]
tools: ynl-gen: complete the C keyword list

C keywords need to be avoided when naming things.
Complete the list (ethtool has at least one thing called "auto").

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl: regen: cleanup user space header includes
Jakub Kicinski [Thu, 8 Jun 2023 21:11:50 +0000 (14:11 -0700)]
tools: ynl: regen: cleanup user space header includes

Remove unnecessary includes.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: cleanup user space header includes
Jakub Kicinski [Thu, 8 Jun 2023 21:11:49 +0000 (14:11 -0700)]
tools: ynl-gen: cleanup user space header includes

Bots started screaming that we're including stdlib.h twice.
While at it move string.h into a common spot and drop stdio.h
which we don't need.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5464
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5466
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5467
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agoRevert "tools: ynl: Remove duplicated include in handshake-user.c"
Jakub Kicinski [Fri, 9 Jun 2023 18:00:59 +0000 (11:00 -0700)]
Revert "tools: ynl: Remove duplicated include in handshake-user.c"

This reverts commit e7c5433c5aaab52ddd5448967a9a5db94a3939cc.

Commit e7c5433c5aaa ("tools: ynl: Remove duplicated include
in handshake-user.c") was applied too hastily. It changes
an auto-generated file, and there's already a proper fix
on the list.

Link: https://lore.kernel.org/all/ZIMPLYi%2FxRih+DlC@nanopsycho/
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl: Remove duplicated include in handshake-user.c
Yang Li [Thu, 8 Jun 2023 08:31:48 +0000 (16:31 +0800)]
tools: ynl: Remove duplicated include in handshake-user.c

./tools/net/ynl/generated/handshake-user.c: stdlib.h is included more than once.

Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=5464
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agoMerge branch 'broadcom-phy-led-brightness'
David S. Miller [Fri, 9 Jun 2023 09:38:44 +0000 (10:38 +0100)]
Merge branch 'broadcom-phy-led-brightness'

Florian Fainelli says:

====================
LED brightness support for Broadcom PHYs

This patch series adds support for controlling the LED brightness on
Broadcom PHYs.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agonet: phy: broadcom: Add support for setting LED brightness
Florian Fainelli [Wed, 7 Jun 2023 18:34:53 +0000 (11:34 -0700)]
net: phy: broadcom: Add support for setting LED brightness

Broadcom PHYs have two LEDs selector registers which allow us to control
the LED assignment, including how to turn them on/off.

Signed-off-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agonet: phy: broadcom: Rename LED registers
Florian Fainelli [Wed, 7 Jun 2023 18:34:52 +0000 (11:34 -0700)]
net: phy: broadcom: Rename LED registers

These registers are common to most PHYs and are not specific to the
BCM5482, renamed the constants accordingly, no functional change.

Signed-off-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agoMerge branch 'net-ncsi-refactoring-for-GMA-cmd'
David S. Miller [Fri, 9 Jun 2023 09:32:51 +0000 (10:32 +0100)]
Merge branch 'net-ncsi-refactoring-for-GMA-cmd'

Ivan Mikhaylov says:

====================
net/ncsi: refactoring for GMA command

Make one GMA function for all manufacturers, change ndo_set_mac_address
to dev_set_mac_address for notifiying net layer about MAC change which
ndo_set_mac_address doesn't do.

Changes from v1:
1. delete ftgmac100.txt changes about mac-address-increment
2. add convert to yaml from ftgmac100.txt
3. add mac-address-increment option for ethernet-controller.yaml

Changes from v2:
1. remove DT changes from series, will be done in another one
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agonet/ncsi: change from ndo_set_mac_address to dev_set_mac_address
Ivan Mikhaylov [Wed, 7 Jun 2023 15:17:42 +0000 (18:17 +0300)]
net/ncsi: change from ndo_set_mac_address to dev_set_mac_address

Change ndo_set_mac_address to dev_set_mac_address because
dev_set_mac_address provides a way to notify network layer about MAC
change. In other case, services may not aware about MAC change and keep
using old one which set from network adapter driver.

As example, DHCP client from systemd do not update MAC address without
notification from net subsystem which leads to the problem with acquiring
the right address from DHCP server.

Fixes: cb10c7c0dfd9e ("net/ncsi: Add NCSI Broadcom OEM command")
Cc: stable@vger.kernel.org # v6.0+ 2f38e84 net/ncsi: make one oem_gma function for all mfr id
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
Signed-off-by: Ivan Mikhaylov <fr0st61te@gmail.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agonet/ncsi: make one oem_gma function for all mfr id
Ivan Mikhaylov [Wed, 7 Jun 2023 15:17:41 +0000 (18:17 +0300)]
net/ncsi: make one oem_gma function for all mfr id

Make the one Get Mac Address function for all manufacturers and change
this call in handlers accordingly.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Ivan Mikhaylov <fr0st61te@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agousbnet: ipheth: update Kconfig description
Foster Snowhill [Wed, 7 Jun 2023 13:57:02 +0000 (15:57 +0200)]
usbnet: ipheth: update Kconfig description

This module has for a long time not been limited to iPhone <= 3GS.
Update description to match the actual state of the driver.

Remove dead link from 2010, instead reference an existing userspace
iOS device pairing implementation as part of libimobiledevice.

Signed-off-by: Foster Snowhill <forst@pen.gy>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agousbnet: ipheth: add CDC NCM support
Foster Snowhill [Wed, 7 Jun 2023 13:57:01 +0000 (15:57 +0200)]
usbnet: ipheth: add CDC NCM support

Recent iOS releases support CDC NCM encapsulation on RX. This mode is
the default on macOS and Windows. In this mode, an iOS device may include
one or more Ethernet frames inside a single URB.

Freshly booted iOS devices start in legacy mode, but are put into
NCM mode by the official Apple driver. When reconnecting such a device
from a macOS/Windows machine to a Linux host, the device stays in
NCM mode, making it unusable with the legacy ipheth driver code.

To correctly support such a device, the driver has to either support
the NCM mode too, or put the device back into legacy mode.

To match the behaviour of the macOS/Windows driver, and since there
is no documented control command to revert to legacy mode, implement
NCM support. The device is attempted to be put into NCM mode by default,
and falls back to legacy mode if the attempt fails.

Signed-off-by: Foster Snowhill <forst@pen.gy>
Tested-by: Georgi Valkov <gvalkov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agousbnet: ipheth: transmit URBs without trailing padding
Foster Snowhill [Wed, 7 Jun 2023 13:57:00 +0000 (15:57 +0200)]
usbnet: ipheth: transmit URBs without trailing padding

The behaviour of the official iOS tethering driver on macOS is to not
transmit any trailing padding at the end of URBs. This is applicable
to both NCM and legacy modes, including older devices.

Adapt the driver to not include trailing padding in TX URBs, matching
the behaviour of the official macOS driver.

Signed-off-by: Foster Snowhill <forst@pen.gy>
Tested-by: Georgi Valkov <gvalkov@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agousbnet: ipheth: fix risk of NULL pointer deallocation
Georgi Valkov [Wed, 7 Jun 2023 13:56:59 +0000 (15:56 +0200)]
usbnet: ipheth: fix risk of NULL pointer deallocation

The cleanup precedure in ipheth_probe will attempt to free a
NULL pointer in dev->ctrl_buf if the memory allocation for
this buffer is not successful. While kfree ignores NULL pointers,
and the existing code is safe, it is a better design to rearrange
the goto labels and avoid this.

Signed-off-by: Georgi Valkov <gvalkov@gmail.com>
Signed-off-by: Foster Snowhill <forst@pen.gy>
Signed-off-by: David S. Miller <davem@davemloft.net>
13 months agoMerge branch 'splice-net-rewrite-splice-to-socket-fix-splice_f_more-and-handle-msg_sp...
Jakub Kicinski [Fri, 9 Jun 2023 02:40:32 +0000 (19:40 -0700)]
Merge branch 'splice-net-rewrite-splice-to-socket-fix-splice_f_more-and-handle-msg_splice_pages-in-af_tls'

David Howells says:

====================
splice, net: Rewrite splice-to-socket, fix SPLICE_F_MORE and handle MSG_SPLICE_PAGES in AF_TLS

Here are patches to do the following:

 (1) Block MSG_SENDPAGE_* flags from leaking into ->sendmsg() from
     userspace, whilst allowing splice_to_socket() to pass them in.

 (2) Allow MSG_SPLICE_PAGES to be passed into tls_*_sendmsg().  Until
     support is added, it will be ignored and a splice-driven sendmsg()
     will be treated like a normal sendmsg().  TCP, UDP, AF_UNIX and
     Chelsio-TLS already handle the flag in net-next.

 (3) Replace a chain of functions to splice-to-sendpage with a single
     function to splice via sendmsg() with MSG_SPLICE_PAGES.  This allows a
     bunch of pages to be spliced from a pipe in a single call using a
     bio_vec[] and pushes the main processing loop down into the bowels of
     the protocol driver rather than repeatedly calling in with a page at a
     time.

 (4) Provide a ->splice_eof() op[2] that allows splice to signal to its
     output that the input observed a premature EOF and that the caller
     didn't flag SPLICE_F_MORE, thereby allowing a corked socket to be
     flushed.  This attempts to maintain the current behaviour.  It is also
     not called if we didn't manage to read any data and so didn't called
     the actor function.

     This needs routing though several layers to get it down to the network
     protocol.

     [!] Note that I chose not to pass in any flags - I'm not sure it's
       particularly useful to pass in the splice flags; I also elected
       not to return any error code - though we might actually want to do
       that.

 (5) Provide tls_{device,sw}_splice_eof() to flush a pending TLS record if
     there is one.

 (6) Provide splice_eof() for UDP, TCP, Chelsio-TLS and AF_KCM.  AF_UNIX
     doesn't seem to pay attention to the MSG_MORE or MSG_SENDPAGE_NOTLAST
     flags.

 (7) Alter the behaviour of sendfile() and fix SPLICE_F_MORE/MSG_MORE
     signalling[1] such SPLICE_F_MORE is always signalled until we have
     read sufficient data to finish the request.  If we get a zero-length
     before we've managed to splice sufficient data, we now leave the
     socket expecting more data and leave it to userspace to deal with it.

 (8) Make AF_TLS handle the MSG_SPLICE_PAGES internal sendmsg flag.
     MSG_SPLICE_PAGES is an internal hint that tells the protocol that it
     should splice the pages supplied if it can.  Its sendpage
     implementations are then turned into wrappers around that.

Link: https://lore.kernel.org/r/499791.1685485603@warthog.procyon.org.uk/
Link: https://lore.kernel.org/r/CAHk-=wh=V579PDYvkpnTobCLGczbgxpMgGmmhqiTyE34Cpi5Gg@mail.gmail.com/
Link: https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next.git/commit/?id=51c78a4d532efe9543a4df019ff405f05c6157f6
Link: https://lore.kernel.org/r/20230524153311.3625329-1-dhowells@redhat.com/
====================

Link: https://lore.kernel.org/r/20230607181920.2294972-1-dhowells@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotls/device: Convert tls_device_sendpage() to use MSG_SPLICE_PAGES
David Howells [Wed, 7 Jun 2023 18:19:20 +0000 (19:19 +0100)]
tls/device: Convert tls_device_sendpage() to use MSG_SPLICE_PAGES

Convert tls_device_sendpage() to use sendmsg() with MSG_SPLICE_PAGES rather
than directly splicing in the pages itself.  With that, the tls_iter_offset
union is no longer necessary and can be replaced with an iov_iter pointer
and the zc_page argument to tls_push_data() can also be removed.

This allows ->sendpage() to be replaced by something that can handle
multiple multipage folios in a single transaction.

Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Jakub Kicinski <kuba@kernel.org>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: John Fastabend <john.fastabend@gmail.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotls/device: Support MSG_SPLICE_PAGES
David Howells [Wed, 7 Jun 2023 18:19:19 +0000 (19:19 +0100)]
tls/device: Support MSG_SPLICE_PAGES

Make TLS's device sendmsg() support MSG_SPLICE_PAGES.  This causes pages to
be spliced from the source iterator if possible.

This allows ->sendpage() to be replaced by something that can handle
multiple multipage folios in a single transaction.

Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: John Fastabend <john.fastabend@gmail.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotls/sw: Convert tls_sw_sendpage() to use MSG_SPLICE_PAGES
David Howells [Wed, 7 Jun 2023 18:19:18 +0000 (19:19 +0100)]
tls/sw: Convert tls_sw_sendpage() to use MSG_SPLICE_PAGES

Convert tls_sw_sendpage() and tls_sw_sendpage_locked() to use sendmsg()
with MSG_SPLICE_PAGES rather than directly splicing in the pages itself.

[!] Note that tls_sw_sendpage_locked() appears to have the wrong locking
    upstream.  I think the caller will only hold the socket lock, but it
    should hold tls_ctx->tx_lock too.

This allows ->sendpage() to be replaced by something that can handle
multiple multipage folios in a single transaction.

Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: John Fastabend <john.fastabend@gmail.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotls/sw: Support MSG_SPLICE_PAGES
David Howells [Wed, 7 Jun 2023 18:19:17 +0000 (19:19 +0100)]
tls/sw: Support MSG_SPLICE_PAGES

Make TLS's sendmsg() support MSG_SPLICE_PAGES.  This causes pages to be
spliced from the source iterator if possible.

This allows ->sendpage() to be replaced by something that can handle
multiple multipage folios in a single transaction.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: John Fastabend <john.fastabend@gmail.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agosplice, net: Fix SPLICE_F_MORE signalling in splice_direct_to_actor()
David Howells [Wed, 7 Jun 2023 18:19:16 +0000 (19:19 +0100)]
splice, net: Fix SPLICE_F_MORE signalling in splice_direct_to_actor()

splice_direct_to_actor() doesn't manage SPLICE_F_MORE correctly[1] - and,
as a result, it incorrectly signals/fails to signal MSG_MORE when splicing
to a socket.  The problem I'm seeing happens when a short splice occurs
because we got a short read due to hitting the EOF on a file: as the length
read (read_len) is less than the remaining size to be spliced (len),
SPLICE_F_MORE (and thus MSG_MORE) is set.

The issue is that, for the moment, we have no way to know *why* the short
read occurred and so can't make a good decision on whether we *should* keep
MSG_MORE set.

MSG_SENDPAGE_NOTLAST was added to work around this, but that is also set
incorrectly under some circumstances - for example if a short read fills a
single pipe_buffer, but the next read would return more (seqfile can do
this).

This was observed with the multi_chunk_sendfile tests in the tls kselftest
program.  Some of those tests would hang and time out when the last chunk
of file was less than the sendfile request size:

build/kselftest/net/tls -r tls.12_aes_gcm.multi_chunk_sendfile

This has been observed before[2] and worked around in AF_TLS[3].

Fix this by making splice_direct_to_actor() always signal SPLICE_F_MORE if
we haven't yet hit the requested operation size.  SPLICE_F_MORE remains
signalled if the user passed it in to splice() but otherwise gets cleared
when we've read sufficient data to fulfill the request.

If, however, we get a premature EOF from ->splice_read(), have sent at
least one byte and SPLICE_F_MORE was not set by the caller, ->splice_eof()
will be invoked.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Linus Torvalds <torvalds@linux-foundation.org>
cc: Jens Axboe <axboe@kernel.dk>
cc: Christoph Hellwig <hch@lst.de>
cc: Al Viro <viro@zeniv.linux.org.uk>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jan Kara <jack@suse.cz>
cc: Jeff Layton <jlayton@kernel.org>
cc: David Hildenbrand <david@redhat.com>
cc: Christian Brauner <brauner@kernel.org>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: John Fastabend <john.fastabend@gmail.com>
cc: linux-mm@kvack.org

Link: https://lore.kernel.org/r/499791.1685485603@warthog.procyon.org.uk/
Link: https://lore.kernel.org/r/1591392508-14592-1-git-send-email-pooja.trivedi@stackpath.com/
Link: https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next.git/commit/?id=d452d48b9f8b1a7f8152d33ef52cfd7fe1735b0a
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agokcm: Use splice_eof() to flush
David Howells [Wed, 7 Jun 2023 18:19:15 +0000 (19:19 +0100)]
kcm: Use splice_eof() to flush

Allow splice to undo the effects of MSG_MORE after prematurely ending a
splice/sendfile due to getting an EOF condition (->splice_read() returned
0) after splice had called sendmsg() with MSG_MORE set when the user didn't
set MSG_MORE.

Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/CAHk-=wh=V579PDYvkpnTobCLGczbgxpMgGmmhqiTyE34Cpi5Gg@mail.gmail.com/
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Tom Herbert <tom@herbertland.com>
cc: Tom Herbert <tom@quantonium.net>
cc: Cong Wang <cong.wang@bytedance.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agochelsio/chtls: Use splice_eof() to flush
David Howells [Wed, 7 Jun 2023 18:19:14 +0000 (19:19 +0100)]
chelsio/chtls: Use splice_eof() to flush

Allow splice to end a Chelsio TLS record after prematurely ending a
splice/sendfile due to getting an EOF condition (->splice_read() returned
0) after splice had called sendmsg() with MSG_MORE set when the user didn't
set MSG_MORE.

Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/CAHk-=wh=V579PDYvkpnTobCLGczbgxpMgGmmhqiTyE34Cpi5Gg@mail.gmail.com/
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Ayush Sawal <ayush.sawal@chelsio.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agoipv4, ipv6: Use splice_eof() to flush
David Howells [Wed, 7 Jun 2023 18:19:13 +0000 (19:19 +0100)]
ipv4, ipv6: Use splice_eof() to flush

Allow splice to undo the effects of MSG_MORE after prematurely ending a
splice/sendfile due to getting an EOF condition (->splice_read() returned
0) after splice had called sendmsg() with MSG_MORE set when the user didn't
set MSG_MORE.

For UDP, a pending packet will not be emitted if the socket is closed
before it is flushed; with this change, it be flushed by ->splice_eof().

For TCP, it's not clear that MSG_MORE is actually effective.

Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/CAHk-=wh=V579PDYvkpnTobCLGczbgxpMgGmmhqiTyE34Cpi5Gg@mail.gmail.com/
Signed-off-by: David Howells <dhowells@redhat.com>
cc: Kuniyuki Iwashima <kuniyu@amazon.com>
cc: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
cc: David Ahern <dsahern@kernel.org>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotls/device: Use splice_eof() to flush
David Howells [Wed, 7 Jun 2023 18:19:12 +0000 (19:19 +0100)]
tls/device: Use splice_eof() to flush

Allow splice to end a TLS record after prematurely ending a splice/sendfile
due to getting an EOF condition (->splice_read() returned 0) after splice
had called TLS with a sendmsg() with MSG_MORE set when the user didn't set
MSG_MORE.

Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/CAHk-=wh=V579PDYvkpnTobCLGczbgxpMgGmmhqiTyE34Cpi5Gg@mail.gmail.com/
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: John Fastabend <john.fastabend@gmail.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotls/sw: Use splice_eof() to flush
David Howells [Wed, 7 Jun 2023 18:19:11 +0000 (19:19 +0100)]
tls/sw: Use splice_eof() to flush

Allow splice to end a TLS record after prematurely ending a splice/sendfile
due to getting an EOF condition (->splice_read() returned 0) after splice
had called TLS with a sendmsg() with MSG_MORE set when the user didn't set
MSG_MORE.

Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/CAHk-=wh=V579PDYvkpnTobCLGczbgxpMgGmmhqiTyE34Cpi5Gg@mail.gmail.com/
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: John Fastabend <john.fastabend@gmail.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agosplice, net: Add a splice_eof op to file-ops and socket-ops
David Howells [Wed, 7 Jun 2023 18:19:10 +0000 (19:19 +0100)]
splice, net: Add a splice_eof op to file-ops and socket-ops

Add an optional method, ->splice_eof(), to allow splice to indicate the
premature termination of a splice to struct file_operations and struct
proto_ops.

This is called if sendfile() or splice() encounters all of the following
conditions inside splice_direct_to_actor():

 (1) the user did not set SPLICE_F_MORE (splice only), and

 (2) an EOF condition occurred (->splice_read() returned 0), and

 (3) we haven't read enough to fulfill the request (ie. len > 0 still), and

 (4) we have already spliced at least one byte.

A further patch will modify the behaviour of SPLICE_F_MORE to always be
passed to the actor if either the user set it or we haven't yet read
sufficient data to fulfill the request.

Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Link: https://lore.kernel.org/r/CAHk-=wh=V579PDYvkpnTobCLGczbgxpMgGmmhqiTyE34Cpi5Gg@mail.gmail.com/
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
cc: Jens Axboe <axboe@kernel.dk>
cc: Christoph Hellwig <hch@lst.de>
cc: Al Viro <viro@zeniv.linux.org.uk>
cc: Matthew Wilcox <willy@infradead.org>
cc: Jan Kara <jack@suse.cz>
cc: Jeff Layton <jlayton@kernel.org>
cc: David Hildenbrand <david@redhat.com>
cc: Christian Brauner <brauner@kernel.org>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: John Fastabend <john.fastabend@gmail.com>
cc: linux-mm@kvack.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agosplice, net: Use sendmsg(MSG_SPLICE_PAGES) rather than ->sendpage()
David Howells [Wed, 7 Jun 2023 18:19:09 +0000 (19:19 +0100)]
splice, net: Use sendmsg(MSG_SPLICE_PAGES) rather than ->sendpage()

Replace generic_splice_sendpage() + splice_from_pipe + pipe_to_sendpage()
with a net-specific handler, splice_to_socket(), that calls sendmsg() with
MSG_SPLICE_PAGES set instead of calling ->sendpage().

MSG_MORE is used to indicate if the sendmsg() is expected to be followed
with more data.

This allows multiple pipe-buffer pages to be passed in a single call in a
BVEC iterator, allowing the processing to be pushed down to a loop in the
protocol driver.  This helps pave the way for passing multipage folios down
too.

Protocols that haven't been converted to handle MSG_SPLICE_PAGES yet should
just ignore it and do a normal sendmsg() for now - although that may be a
bit slower as it may copy everything.

Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotls: Allow MSG_SPLICE_PAGES but treat it as normal sendmsg
David Howells [Wed, 7 Jun 2023 18:19:08 +0000 (19:19 +0100)]
tls: Allow MSG_SPLICE_PAGES but treat it as normal sendmsg

Allow MSG_SPLICE_PAGES to be specified to sendmsg() but treat it as normal
sendmsg for now.  This means the data will just be copied until
MSG_SPLICE_PAGES is handled.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: John Fastabend <john.fastabend@gmail.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonet: Block MSG_SENDPAGE_* from being passed to sendmsg() by userspace
David Howells [Wed, 7 Jun 2023 18:19:07 +0000 (19:19 +0100)]
net: Block MSG_SENDPAGE_* from being passed to sendmsg() by userspace

It is necessary to allow MSG_SENDPAGE_* to be passed into ->sendmsg() to
allow sendmsg(MSG_SPLICE_PAGES) to replace ->sendpage().  Unblocking them
in the network protocol, however, allows these flags to be passed in by
userspace too[1].

Fix this by marking MSG_SENDPAGE_NOPOLICY, MSG_SENDPAGE_NOTLAST and
MSG_SENDPAGE_DECRYPTED as internal flags, which causes sendmsg() to object
if they are passed to sendmsg() by userspace.  Network protocol ->sendmsg()
implementations can then allow them through.

Note that it should be possible to remove MSG_SENDPAGE_NOTLAST once
sendpage is removed as a whole slew of pages will be passed in in one go by
splice through sendmsg, with MSG_MORE being set if it has more data waiting
in the pipe.

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Chuck Lever <chuck.lever@oracle.com>
cc: Boris Pismenny <borisp@nvidia.com>
cc: John Fastabend <john.fastabend@gmail.com>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Link: https://lore.kernel.org/r/20230526181338.03a99016@kernel.org/
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotcp: let tcp_mtu_probe() build headless packets
Eric Dumazet [Wed, 7 Jun 2023 21:41:13 +0000 (21:41 +0000)]
tcp: let tcp_mtu_probe() build headless packets

tcp_mtu_probe() is still copying payload from skbs in the write queue,
using skb_copy_bits(), ignoring potential errors.

Modern TCP stack wants to only deal with payload found in page frags,
as this is a prereq for TCPDirect (host stack might not have access
to the payload)

Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20230607214113.1992947-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agoMerge tag 'mlx5-updates-2023-06-06' of git://git.kernel.org/pub/scm/linux/kernel...
Jakub Kicinski [Fri, 9 Jun 2023 02:28:20 +0000 (19:28 -0700)]
Merge tag 'mlx5-updates-2023-06-06' of git://git./linux/kernel/git/saeed/linux

Saeed Mahameed says:

====================
mlx5-updates-2023-06-06

1) Support 4 ports VF LAG, part 2/2
2) Few extra trivial cleanup patches

Shay Drory Says:
================

Support 4 ports VF LAG, part 2/2

This series continues the series[1] "Support 4 ports VF LAG, part1/2".
This series adds support for 4 ports VF LAG (single FDB E-Switch).

This series of patches refactoring LAG code that make assumptions
about VF LAG supporting only two ports and then enable 4 ports VF LAG.

Patch 1:
- Fix for ib rep code
Patches 2-5:
- Refactors LAG layer.
Patches 6-7:
- Block LAG types which doesn't support 4 ports.
Patch 8:
- Enable 4 ports VF LAG.

This series specifically allows HCAs with 4 ports to create a VF LAG
with only 4 ports. It is not possible to create a VF LAG with 2 or 3
ports using HCAs that have 4 ports.

Currently, the Merged E-Switch feature only supports HCAs with 2 ports.
However, upcoming patches will introduce support for HCAs with 4 ports.

In order to activate VF LAG a user can execute:

devlink dev eswitch set pci/0000:08:00.0 mode switchdev
devlink dev eswitch set pci/0000:08:00.1 mode switchdev
devlink dev eswitch set pci/0000:08:00.2 mode switchdev
devlink dev eswitch set pci/0000:08:00.3 mode switchdev
ip link add name bond0 type bond
ip link set dev bond0 type bond mode 802.3ad
ip link set dev eth2 master bond0
ip link set dev eth3 master bond0
ip link set dev eth4 master bond0
ip link set dev eth5 master bond0

Where eth2, eth3, eth4 and eth5 are net-interfaces of pci/0000:08:00.0
pci/0000:08:00.1 pci/0000:08:00.2 pci/0000:08:00.3 respectively.

User can verify LAG state and type via debugfs:
/sys/kernel/debug/mlx5/0000\:08\:00.0/lag/state
/sys/kernel/debug/mlx5/0000\:08\:00.0/lag/type

[1]
https://lore.kernel.org/netdev/20230601060118.154015-1-saeed@kernel.org/T/#mf1d2083780970ba277bfe721554d4925f03f36d1

================

* tag 'mlx5-updates-2023-06-06' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux:
  net/mlx5e: simplify condition after napi budget handling change
  mlx5/core: E-Switch, Allocate ECPF vport if it's an eswitch manager
  net/mlx5: Skip inline mode check after mlx5_eswitch_enable_locked() failure
  net/mlx5e: TC, refactor access to hash key
  net/mlx5e: Remove RX page cache leftovers
  net/mlx5e: Expose catastrophic steering error counters
  net/mlx5: Enable 4 ports VF LAG
  net/mlx5: LAG, block multiport eswitch LAG in case ldev have more than 2 ports
  net/mlx5: LAG, block multipath LAG in case ldev have more than 2 ports
  net/mlx5: LAG, change mlx5_shared_fdb_supported() to static
  net/mlx5: LAG, generalize handling of shared FDB
  net/mlx5: LAG, check if all eswitches are paired for shared FDB
  {net/RDMA}/mlx5: introduce lag_for_each_peer
  RDMA/mlx5: Free second uplink ib port
====================

Link: https://lore.kernel.org/r/20230607210410.88209-1-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agoethtool: ioctl: improve error checking for set_wol
Justin Chen [Wed, 7 Jun 2023 23:14:11 +0000 (16:14 -0700)]
ethtool: ioctl: improve error checking for set_wol

The netlink version of set_wol checks for not supported wolopts and avoids
setting wol when the correct wolopt is already set. If we do the same with
the ioctl version then we can remove these checks from the driver layer.

Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: Justin Chen <justin.chen@broadcom.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Link: https://lore.kernel.org/r/1686179653-29750-1-git-send-email-justin.chen@broadcom.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agoMerge branch 'complete-lynx-mdio-device-handling'
Jakub Kicinski [Fri, 9 Jun 2023 02:19:52 +0000 (19:19 -0700)]
Merge branch 'complete-lynx-mdio-device-handling'

Russell King says:

====================
complete Lynx mdio device handling

This series completes the mdio device lifetime handling for Lynx PCS
users which do not create their own mdio device, but instead fetch
it using a firmware description - namely the DPAA2 and FMAN_MEMAC
drivers.

In a previous patch set, lynx_pcs_create() was modified to increase
the mdio device refcount, and lynx_pcs_destroy() to drop that
refcount.

The first two patches change these two drivers to put the reference
which they hold immediately after lynx_pcs_create(), effectively
handing the responsibility for maintaining the refcount to the Lynx
PCS driver.

A side effect of the first two patches is that lynx_get_mdio_device()
is no longer used, so patch 3 removes it.

Patch 4 adds a new helper - lynx_pcs_create_fwnode(), which creates
a Lynx PCS instance from the fwnode.

Patch 5 and 6 convert the two drivers to make use of this new helper,
which simply has to find the mdio device, and then create the Lynx
PCS from that.

With those conversions done, lynx_pcs_create() is no longer required
outside pcs-lynx.c, so remove it from public view.

Patch 8 we changes lynx_pcs_create() to return an error-pointer rather
than NULL to bring consistency to the return style, and means that we
can remove the NULL-to-error-pointer conversion from both
lynx_pcs_create_fwnode() and lynx_pcs_create_mdiodev().

Patch 9 adds a check for the fwnode being available, and returns an
-ENODEV error pointer if unavailable.

Patch 10 removes this check from DPAA2, detecting the error pointer
value to continue printing the helpful message.

Patch 11 removes this check from fman_memac, and in doing so fixes a
bug where if the node is unavailable, the reference count is not
dropped.
====================

Link: https://lore.kernel.org/r/ZIBwuw+IuGQo5yV8@shell.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonet: fman_memac: use pcs-lynx's check for fwnode availability
Russell King (Oracle) [Wed, 7 Jun 2023 11:59:04 +0000 (12:59 +0100)]
net: fman_memac: use pcs-lynx's check for fwnode availability

Use pcs-lynx's check rather than our own when determining if the device
is available. This fixes a bug where the reference gained by
of_parse_phandle() is not dropped if the device is not available.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonet: dpaa2: use pcs-lynx's check for fwnode availability
Russell King (Oracle) [Wed, 7 Jun 2023 11:58:59 +0000 (12:58 +0100)]
net: dpaa2: use pcs-lynx's check for fwnode availability

Use pcs-lynx's check rather than our own when determining if the device
is available.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonet: pcs: lynx: check that the fwnode is available prior to use
Russell King (Oracle) [Wed, 7 Jun 2023 11:58:54 +0000 (12:58 +0100)]
net: pcs: lynx: check that the fwnode is available prior to use

Check that the fwnode is marked as available prior to trying to lookup
the PCS device, and return -ENODEV if unavailable. Document the return
codes from lynx_pcs_create_fwnode().

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonet: pcs: lynx: change lynx_pcs_create() to return error-pointers
Russell King (Oracle) [Wed, 7 Jun 2023 11:58:49 +0000 (12:58 +0100)]
net: pcs: lynx: change lynx_pcs_create() to return error-pointers

Change lynx_pcs_create() to return an error-pointer on failure to
allocate memory, rather than returning NULL. This allows the removal
of the conversion in lynx_pcs_create_fwnode() and
lynx_pcs_create_mdiodev().

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonet: pcs: lynx: make lynx_pcs_create() static
Russell King (Oracle) [Wed, 7 Jun 2023 11:58:44 +0000 (12:58 +0100)]
net: pcs: lynx: make lynx_pcs_create() static

We no longer need to export lynx_pcs_create() for drivers to use as we
now have all the functionality we need in the two new creation helpers.
Remove the export and prototype, and make it static.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonet: fman_memac: use lynx_pcs_create_fwnode()
Russell King (Oracle) [Wed, 7 Jun 2023 11:58:39 +0000 (12:58 +0100)]
net: fman_memac: use lynx_pcs_create_fwnode()

Use lynx_pcs_create_fwnode() to create a lynx PCS from a fwnode handle.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonet: dpaa2-mac: use lynx_pcs_create_fwnode()
Russell King (Oracle) [Wed, 7 Jun 2023 11:58:34 +0000 (12:58 +0100)]
net: dpaa2-mac: use lynx_pcs_create_fwnode()

Use lynx_pcs_create_fwnode() to create a lynx PCS from a fwnode handle.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonet: pcs: lynx: add lynx_pcs_create_fwnode()
Russell King (Oracle) [Wed, 7 Jun 2023 11:58:29 +0000 (12:58 +0100)]
net: pcs: lynx: add lynx_pcs_create_fwnode()

Add a helper to create a lynx PCS from a fwnode handle.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonet: pcs: lynx: remove lynx_get_mdio_device()
Russell King (Oracle) [Wed, 7 Jun 2023 11:58:24 +0000 (12:58 +0100)]
net: pcs: lynx: remove lynx_get_mdio_device()

lynx_get_mdio_device() is no longer necessary, let's remove it so the
lynx PCS code is always managing the lifetime of the mdiodev.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonet: fman_memac: allow lynx PCS to handle mdiodev lifetime
Russell King (Oracle) [Wed, 7 Jun 2023 11:58:18 +0000 (12:58 +0100)]
net: fman_memac: allow lynx PCS to handle mdiodev lifetime

Put the mdiodev after lynx_pcs_create() so that the Lynx PCS driver
can manage the lifetime of the mdiodev its using.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonet: dpaa2-mac: allow lynx PCS to manage mdiodev lifetime
Russell King (Oracle) [Wed, 7 Jun 2023 11:58:13 +0000 (12:58 +0100)]
net: dpaa2-mac: allow lynx PCS to manage mdiodev lifetime

Put the mdiodev after lynx_pcs_create() so that the Lynx PCS driver
can manage the lifetime of the mdiodev its using.

Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonet: pch_gbe: Allow build on MIPS_GENERIC kernel
Jiaxun Yang [Wed, 7 Jun 2023 05:59:53 +0000 (13:59 +0800)]
net: pch_gbe: Allow build on MIPS_GENERIC kernel

MIPS Boston board, which is using MIPS_GENERIC kernel is using
EG20T PCH and thus need this driver.

Dependency of PCH_GBE, PTP_1588_CLOCK_PCH is also fixed for
MIPS_GENERIC.

Note that CONFIG_PCH_GBE is selected in arch/mips/configs/generic/
board-boston.config for a while, some how it's never wired up
in Kconfig.

Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/20230607055953.34110-1-jiaxun.yang@flygoat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agomlxsw: spectrum_nve_vxlan: Fix unsupported flag regression
Ido Schimmel [Wed, 7 Jun 2023 12:19:26 +0000 (14:19 +0200)]
mlxsw: spectrum_nve_vxlan: Fix unsupported flag regression

The recently added 'VXLAN_F_LOCALBYPASS' flag is set by default on VXLAN
devices and denotes a behavior that is irrelevant for the hardware data
path. Add it to the lists of IPv4 and IPv6 supported flags to avoid
rejecting offload of VXLAN devices which have this flag set.

Fixes: 69474a8a5837 ("net: vxlan: Add nolocalbypass option to vxlan.")
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Link: https://lore.kernel.org/r/5533e63643bf719bbe286fef60f749c9cad35005.1686139716.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agoMerge branch 'tools-ynl-generate-code-for-the-devlink-family'
Jakub Kicinski [Thu, 8 Jun 2023 21:01:12 +0000 (14:01 -0700)]
Merge branch 'tools-ynl-generate-code-for-the-devlink-family'

Jakub Kicinski says:

====================
tools: ynl: generate code for the devlink family

Another chunk of changes to support more capabilities in the YNL
code gen. Devlink brings in deep nesting and directional messages
(requests and responses have different IDs). We need a healthy
dose of codegen changes to support those (I wasn't planning to
support code gen for "directional" families initially, but
the importance of devlink and ethtool is undeniable).
====================

Link: https://lore.kernel.org/r/20230607202403.1089925-1-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl: add sample for devlink
Jakub Kicinski [Wed, 7 Jun 2023 20:24:03 +0000 (13:24 -0700)]
tools: ynl: add sample for devlink

Add a sample to show off how to issue basic devlink requests.
For added testing issue get requests while walking a dump.

$ ./devlink
netdevsim/netdevsim1:
    driver: netdevsim
    running fw:
        fw.mgmt: 10.20.30
    ...
netdevsim/netdevsim2:
    driver: netdevsim
    running fw:
        fw.mgmt: 10.20.30
    ...

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl: generate code for the devlink family
Jakub Kicinski [Wed, 7 Jun 2023 20:24:02 +0000 (13:24 -0700)]
tools: ynl: generate code for the devlink family

Admittedly the devlink.yaml spec is fairly limitted,
it only covers basic device get and info-get ops.
That's sufficient to be useful (monitoring FW versions
in the fleet). Plus it gives us a chance to exercise
deep nesting and directional messaging in YNL.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: don't generate forward declarations for policies - regen
Jakub Kicinski [Wed, 7 Jun 2023 20:24:01 +0000 (13:24 -0700)]
tools: ynl-gen: don't generate forward declarations for policies - regen

Renegerate code after dropping forward declarations for policies.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: don't generate forward declarations for policies
Jakub Kicinski [Wed, 7 Jun 2023 20:24:00 +0000 (13:24 -0700)]
tools: ynl-gen: don't generate forward declarations for policies

Now that all nested types have structs and are sorted topologically
there should be no need to generate forward declarations for policies.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: walk nested types in depth
Jakub Kicinski [Wed, 7 Jun 2023 20:23:59 +0000 (13:23 -0700)]
tools: ynl-gen: walk nested types in depth

So far we had only created structures for nested types nested
directly in messages (second level of attrs so to speak).
Walk types in depth to support deeper nesting.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: inherit struct use info
Jakub Kicinski [Wed, 7 Jun 2023 20:23:58 +0000 (13:23 -0700)]
tools: ynl-gen: inherit struct use info

We only render parse and netlink generation helpers as needed,
to avoid generating dead code. Propagate the information from
first- and second-layer attribute sets onto all children.
Otherwise devlink won't work, it has a lot more levels of nesting.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: try to sort the types more intelligently
Jakub Kicinski [Wed, 7 Jun 2023 20:23:57 +0000 (13:23 -0700)]
tools: ynl-gen: try to sort the types more intelligently

We need to sort the structures to avoid the need for forward
declarations. While at it remove the sort of structs when
rendering, it doesn't do anything.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: enable code gen for directional specs
Jakub Kicinski [Wed, 7 Jun 2023 20:23:56 +0000 (13:23 -0700)]
tools: ynl-gen: enable code gen for directional specs

I think that user space code gen for directional specs
works after recent changes. Let them through.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: refactor strmap helper generation
Jakub Kicinski [Wed, 7 Jun 2023 20:23:55 +0000 (13:23 -0700)]
tools: ynl-gen: refactor strmap helper generation

Move generating strmap lookup function to a helper.
No functional changes.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agotools: ynl-gen: use enum names in op strmap more carefully
Jakub Kicinski [Wed, 7 Jun 2023 20:23:54 +0000 (13:23 -0700)]
tools: ynl-gen: use enum names in op strmap more carefully

In preparation for supporting families which use different msg
ids to and from the kernel - make sure the ids in op strmap
are correct. The map is expected to be used mostly for notifications,
don't generate a separate map for the "to kernel" direction.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agonetlink: specs: devlink: fill in some details important for C
Jakub Kicinski [Wed, 7 Jun 2023 20:23:53 +0000 (13:23 -0700)]
netlink: specs: devlink: fill in some details important for C

Python YNL is much more forgiving than the C code gen in terms
of the spec completeness. Fill in a handful of devlink details
to make the spec usable in C.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agoice: do not re-enable miscellaneous interrupt until thread_fn completes
Jacob Keller [Thu, 1 Jun 2023 21:15:07 +0000 (14:15 -0700)]
ice: do not re-enable miscellaneous interrupt until thread_fn completes

The ice driver uses threaded IRQ for managing Tx timestamps via the
devm_request_threaded_irq() interface. The ice_misc_intr() handler function
is responsible for processing the hard interrupt context, and can wake the
ice_misc_intr_thread_fn() by returning IRQ_WAKE_THREAD.

The request_threaded_irq() function comment says:

  @handler is still called in hard interrupt context and has to check
  whether the interrupt originates from the device. If yes, it needs to
  disable the interrupt on the device and return IRQ_WAKE_THREAD which will
  wake up the handler thread and run the @thread_fn.

We currently re-enable the Other Interrupt Cause Register (OCIR) at the end of
ice_misc_intr(). In practice, this seems to be ok, but it can make
communicating between the handler function and the thread function
difficult. This is because the interrupt can trigger again while the thread
function is still processing.

Move the OICR update to the end of the thread function, leaving the other
interrupt cause disabled in hardware until we complete one pass of the
thread function. This prevents the miscellaneous interrupt from firing
until after we finish the thread function.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Arpana Arland <arpanax.arland@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
13 months agoice: trigger PFINT_OICR_TSYN_TX interrupt instead of polling
Jacob Keller [Thu, 1 Jun 2023 21:15:06 +0000 (14:15 -0700)]
ice: trigger PFINT_OICR_TSYN_TX interrupt instead of polling

In ice_misc_intr_thread_fn(), if we do not complete all Tx timestamp work,
the thread function will poll continuously forever.

For E822 hardware, this wastes time as the return value from
ice_ptp_process_ts() is accurate and always reports correctly that the PHY
actually has new timestamp data.

In addition, if we receive enough timestamps with the right pacing, we may
never exit this polling. Should this occur, other tasks handled by the
ice_misc_intr_thread_fn() will never be processed.

Fix this by instead writing to PFINT_OICR, causing an emulated interrupt to
be triggered immediately. This does take slightly more processing than just
re-checking the timestamps. However, it allows all of the other interrupt
causes a chance to be processed first in the hard IRQ function.

Note that the OICR interrupt is configured to be throttled  to no more than
once every 124 microseconds. This gives an effective interrupt rate of
~8000 interrupts per second. This should thus not cause a significant
increase in overall CPU usage when compared to sleeping.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Arpana Arland <arpanax.arland@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
13 months agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Jakub Kicinski [Thu, 8 Jun 2023 18:34:28 +0000 (11:34 -0700)]
Merge git://git./linux/kernel/git/netdev/net

Cross-merge networking fixes after downstream PR.

Conflicts:

net/sched/sch_taprio.c
  d636fc5dd692 ("net: sched: add rcu annotations around qdisc->qdisc_sleeping")
  dced11ef84fb ("net/sched: taprio: don't overwrite "sch" variable in taprio_dump_class_stats()")

net/ipv4/sysctl_net_ipv4.c
  e209fee4118f ("net/ipv4: ping_group_range: allow GID from 2147483648 to 4294967294")
  ccce324dabfe ("tcp: make the first N SYN RTO backoffs linear")
https://lore.kernel.org/all/20230605100816.08d41a7b@canb.auug.org.au/

No adjacent changes.

Signed-off-by: Jakub Kicinski <kuba@kernel.org>
13 months agoice: introduce ICE_TX_TSTAMP_WORK enumeration
Jacob Keller [Thu, 1 Jun 2023 21:15:05 +0000 (14:15 -0700)]
ice: introduce ICE_TX_TSTAMP_WORK enumeration

The ice_ptp_process_ts() function and its various helper functions return a
boolean value indicating whether any work is remaining. This use of a
boolean has grown confusing as we have multiple helpers that pass status
between each other. Readers must be aware of what "true" and "false" mean,
and it is very easy to get their meaning inverted. The names of the
functions are not standard "yes/no" questions, which is the best practice
for boolean returns.

Replace this use of an enumeration with a custom type, enum
ice_tx_tstamp_work. This enumeration clearly indicates whether all work is
done, or if more work is pending.

To aid in readability, factor the actual list iteration and processing out
into ice_ptp_process_tx_tstamp(), making it void. Then call this in
ice_ptp_tx_tstamp() ensuring that we always check the Tracker list at the
end when determining the appropriate return value.

Now the return value is an explicit name instead of the true or false
value. This is easier to follow and makes reading the resulting callers
much simpler.

In addition, this paves the way for future work to allow E822 hardware to
process timestamps for all functions using a single interrupt on the clock
owning PF.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Arpana Arland <arpanax.arland@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
13 months agoice: always return IRQ_WAKE_THREAD in ice_misc_intr()
Karol Kolacinski [Thu, 1 Jun 2023 21:15:04 +0000 (14:15 -0700)]
ice: always return IRQ_WAKE_THREAD in ice_misc_intr()

Refactor the ice_misc_intr() function to always return IRQ_WAKE_THREAD, and
schedule the service task during the soft IRQ thread function instead of at
the end of the hard IRQ handler.

Remove the duplicate call to ice_service_task_schedule() that happened when
we got a PCI exception.

Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Arpana Arland <arpanax.arland@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
13 months agoice: handle extts in the miscellaneous interrupt thread
Karol Kolacinski [Thu, 1 Jun 2023 21:15:03 +0000 (14:15 -0700)]
ice: handle extts in the miscellaneous interrupt thread

The ice_ptp_extts_work() and ice_ptp_periodic_work() functions are both
scheduled on the same kthread worker, pf.ptp.kworker. The
ice_ptp_periodic_work() function sends to the firmware to interact with the
PHY, and must block to wait for responses.

This can cause delay in responding to the PFINT_OICR_TSYN_EVNT interrupt
cause, ultimately resulting in disruption to processing an input signal of
the frequency is high enough. In our testing, even 100 Hz signals get
disrupted.

Fix this by instead processing the signal inside the miscellaneous
interrupt thread prior to handling Tx timestamps.

Use atomic bits in a new pf->misc_thread bitmap in order to safely
communicate which tasks require processing within the
ice_misc_intr_thread_fn(). This ensures the communication of desired tasks
from the ice_misc_intr() are correctly processed without racing even in the
event that the interrupt triggers again before the thread function exits.

Fixes: 172db5f91d5f ("ice: add support for auxiliary input/output pins")
Signed-off-by: Karol Kolacinski <karol.kolacinski@intel.com>
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Arpana Arland <arpanax.arland@intel.com> (A Contingent worker at Intel)
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
13 months agoMerge tag 'net-6.4-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Linus Torvalds [Thu, 8 Jun 2023 16:27:19 +0000 (09:27 -0700)]
Merge tag 'net-6.4-rc6' of git://git./linux/kernel/git/netdev/net

Pull networking fixes from Paolo Abeni:
 "Including fixes from can, wifi, netfilter, bluetooth and ebpf.

  Current release - regressions:

   - bpf: sockmap: avoid potential NULL dereference in
     sk_psock_verdict_data_ready()

   - wifi: iwlwifi: fix -Warray-bounds bug in iwl_mvm_wait_d3_notif()

   - phylink: actually fix ksettings_set() ethtool call

   - eth: dwmac-qcom-ethqos: fix a regression on EMAC < 3

  Current release - new code bugs:

   - wifi: mt76: fix possible NULL pointer dereference in
     mt7996_mac_write_txwi()

  Previous releases - regressions:

   - netfilter: fix NULL pointer dereference in nf_confirm_cthelper

   - wifi: rtw88/rtw89: correct PS calculation for SUPPORTS_DYNAMIC_PS

   - openvswitch: fix upcall counter access before allocation

   - bluetooth:
      - fix use-after-free in hci_remove_ltk/hci_remove_irk
      - fix l2cap_disconnect_req deadlock

   - nic: bnxt_en: prevent kernel panic when receiving unexpected
     PHC_UPDATE event

  Previous releases - always broken:

   - core: annotate rfs lockless accesses

   - sched: fq_pie: ensure reasonable TCA_FQ_PIE_QUANTUM values

   - netfilter: add null check for nla_nest_start_noflag() in
     nft_dump_basechain_hook()

   - bpf: fix UAF in task local storage

   - ipv4: ping_group_range: allow GID from 2147483648 to 4294967294

   - ipv6: rpl: fix route of death.

   - tcp: gso: really support BIG TCP

   - mptcp: fixes for user-space PM address advertisement

   - smc: avoid to access invalid RMBs' MRs in SMCRv1 ADD LINK CONT

   - can: avoid possible use-after-free when j1939_can_rx_register fails

   - batman-adv: fix UaF while rescheduling delayed work

   - eth: qede: fix scheduling while atomic

   - eth: ice: make writes to /dev/gnssX synchronous"

* tag 'net-6.4-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (83 commits)
  bnxt_en: Implement .set_port / .unset_port UDP tunnel callbacks
  bnxt_en: Prevent kernel panic when receiving unexpected PHC_UPDATE event
  bnxt_en: Skip firmware fatal error recovery if chip is not accessible
  bnxt_en: Query default VLAN before VNIC setup on a VF
  bnxt_en: Don't issue AP reset during ethtool's reset operation
  bnxt_en: Fix bnxt_hwrm_update_rss_hash_cfg()
  net: bcmgenet: Fix EEE implementation
  eth: ixgbe: fix the wake condition
  eth: bnxt: fix the wake condition
  lib: cpu_rmap: Fix potential use-after-free in irq_cpu_rmap_release()
  bpf: Add extra path pointer check to d_path helper
  net: sched: fix possible refcount leak in tc_chain_tmplt_add()
  net: sched: act_police: fix sparse errors in tcf_police_dump()
  net: openvswitch: fix upcall counter access before allocation
  net: sched: move rtm_tca_policy declaration to include file
  ice: make writes to /dev/gnssX synchronous
  net: sched: add rcu annotations around qdisc->qdisc_sleeping
  rfs: annotate lockless accesses to RFS sock flow table
  rfs: annotate lockless accesses to sk->sk_rxhash
  virtio_net: use control_buf for coalesce params
  ...