Some of the newer silicon versions in CN10K series supports a feature
where in the current PTP timestamp in HW can be updated atomically
without losing any cpu cycles unlike read/modify/write register.
This patch uses this feature so that PTP accuracy can be improved
while adjusting the master offset in HW. There is no need for SW
timecounter when using this feature. So removed references to SW
timecounter wherever appropriate.
Signed-off-by: Sai Krishna <saikrishnag@marvell.com>
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
Signed-off-by: Sunil Kovvuri Goutham <sgoutham@marvell.com>
Reviewed-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Westphal says:
====================
netfilter updates for net-next
First patch resolves a fortify warning by wrapping the to-be-copied
members via struct_group.
Second patch replaces array[0] with array[] in ebtables uapi.
Both changes from GONG Ruiqi.
The largest chunk is replacement of strncpy with strscpy_pad()
in netfilter, from Justin Stitt.
Last patch, from myself, aborts ruleset validation if a fatal
signal is pending, this speeds up process exit.
* tag 'nf-next-23-08-22' of https://git.kernel.org/pub/scm/linux/kernel/git/netfilter/nf-next:
netfilter: nf_tables: allow loop termination for pending fatal signal
netfilter: xtables: refactor deprecated strncpy
netfilter: x_tables: refactor deprecated strncpy
netfilter: nft_meta: refactor deprecated strncpy
netfilter: nft_osf: refactor deprecated strncpy
netfilter: nf_tables: refactor deprecated strncpy
netfilter: nf_tables: refactor deprecated strncpy
netfilter: ipset: refactor deprecated strncpy
netfilter: ebtables: replace zero-length array members
netfilter: ebtables: fix fortify warnings in size_entry_mwt()
====================
Link: https://lore.kernel.org/r/20230822154336.12888-1-fw@strlen.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Mat Martineau says:
====================
mptcp: Prepare MPTCP packet scheduler for BPF extension
The kernel's MPTCP packet scheduler has, to date, been a one-size-fits
all algorithm that is hard-coded. It attempts to balance latency and
throughput when transmitting data across multiple TCP subflows, and has
some limited tunability through sysctls. It has been a long-term goal of
the Linux MPTCP community to support customizable packet schedulers for
use cases that need to make different trade-offs regarding latency,
throughput, redundancy, and other metrics. BPF is well-suited for
configuring customized, per-packet scheduling decisions without having
to modify the kernel or manage out-of-tree kernel modules.
The first steps toward implementing BPF packet schedulers are to update
the existing MPTCP transmit loops to allow more flexible scheduling
decisions, and to add infrastructure for swappable packet schedulers.
The existing scheduling algorithm remains the default. BPF-related
changes will be in a future patch series.
This code has been in the MPTCP development tree for quite a while,
undergoing testing in our CI and community.
Patches 1 and 2 refactor the transmit code and do some related cleanup.
Patches 3-9 add infrastructure for registering and calling multiple
schedulers.
Patch 10 connects the in-kernel default scheduler to the new
infrastructure.
====================
Link: https://lore.kernel.org/r/20230821-upstream-net-next-20230818-v1-0-0c860fb256a8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This patch defines the default packet scheduler mptcp_sched_default.
Register it in mptcp_sched_init(), which is invoked in mptcp_proto_init().
Skip deleting this default scheduler in mptcp_unregister_scheduler().
Set msk->sched to the default scheduler when the input parameter of
mptcp_init_sched() is NULL.
Invoke mptcp_sched_default_get_subflow in get_send() and get_retrans()
if the defaut scheduler is set or msk->sched is NULL.
Acked-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Mat Martineau <martineau@kernel.org>
Link: https://lore.kernel.org/r/20230821-upstream-net-next-20230818-v1-10-0c860fb256a8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This patch adds the multiple subflows support for __mptcp_push_pending
and __mptcp_subflow_push_pending. Use get_send() wrapper instead of
mptcp_subflow_get_send() in them.
Check the subflow scheduled flags to test which subflow or subflows are
picked by the scheduler, use them to send data.
Move msk_owned_by_me() and fallback checks into get_send() wrapper from
mptcp_subflow_get_send().
This commit allows the scheduler to set the subflow->scheduled bit in
multiple subflows, but it does not allow for sending redundant data.
Multiple scheduled subflows will send sequential data on each subflow.
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Mat Martineau <martineau@kernel.org>
Link: https://lore.kernel.org/r/20230821-upstream-net-next-20230818-v1-8-0c860fb256a8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This patch defines struct mptcp_sched_ops, which has three struct members,
name, owner and list, and four function pointers: init(), release() and
get_subflow().
The scheduler function get_subflow() have a struct mptcp_sched_data
parameter, which contains a reinject flag for retrans or not, a subflows
number and a mptcp_subflow_context array.
Add the scheduler registering, unregistering and finding functions to add,
delete and find a packet scheduler on the global list mptcp_sched_list.
Acked-by: Paolo Abeni <pabeni@redhat.com>
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Mat Martineau <martineau@kernel.org>
Link: https://lore.kernel.org/r/20230821-upstream-net-next-20230818-v1-3-0c860fb256a8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
To support redundant package schedulers more easily, this patch refactors
__mptcp_push_pending() logic from:
For each dfrag:
While sends succeed:
Call the scheduler (selects subflow and msk->snd_burst)
Update subflow locks (push/release/acquire as needed)
Send the dfrag data with mptcp_sendmsg_frag()
Update already_sent, snd_nxt, snd_burst
Update msk->first_pending
Push/release on final subflow
->
While first_pending isn't empty:
Call the scheduler (selects subflow and msk->snd_burst)
Update subflow locks (push/release/acquire as needed)
For each pending dfrag:
While sends succeed:
Send the dfrag data with mptcp_sendmsg_frag()
Update already_sent, snd_nxt, snd_burst
Update msk->first_pending
Break if required by msk->snd_burst / etc
Push/release on final subflow
Refactors __mptcp_subflow_push_pending logic from:
For each dfrag:
While sends succeed:
Call the scheduler (selects subflow and msk->snd_burst)
Send the dfrag data with mptcp_subflow_delegate(), break
Send the dfrag data with mptcp_sendmsg_frag()
Update dfrag->already_sent, msk->snd_nxt, msk->snd_burst
Update msk->first_pending
->
While first_pending isn't empty:
Call the scheduler (selects subflow and msk->snd_burst)
Send the dfrag data with mptcp_subflow_delegate(), break
Send the dfrag data with mptcp_sendmsg_frag()
For each pending dfrag:
While sends succeed:
Send the dfrag data with mptcp_sendmsg_frag()
Update already_sent, snd_nxt, snd_burst
Update msk->first_pending
Break if required by msk->snd_burst / etc
Move the duplicate code from __mptcp_push_pending() and
__mptcp_subflow_push_pending() into a new helper function, named
__subflow_push_pending(). Simplify __mptcp_push_pending() and
__mptcp_subflow_push_pending() by invoking this helper.
Also move the burst check conditions out of the function
mptcp_subflow_get_send(), check them in __subflow_push_pending() in
the inner "for each pending dfrag" loop.
Reviewed-by: Mat Martineau <martineau@kernel.org>
Signed-off-by: Geliang Tang <geliang.tang@suse.com>
Signed-off-by: Mat Martineau <martineau@kernel.org>
Link: https://lore.kernel.org/r/20230821-upstream-net-next-20230818-v1-1-0c860fb256a8@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Saeed Mahameed says:
====================
mlx5-updates-2023-08-16
1) aRFS ethtool stats
Improve aRFS observability by adding new set of counters. Each Rx
ring will have this set of counters listed below.
These counters are exposed through ethtool -S.
1.1) arfs_add: number of times a new rule has been created.
1.2) arfs_request_in: number of times a rule was requested to move from
its current Rx ring to a new Rx ring (incremented on the destination
Rx ring).
1.3) arfs_request_out: number of times a rule was requested to move out
from its current Rx ring (incremented on source/current Rx ring).
1.4) arfs_expired: number of times a rule has been expired by the
kernel and removed from HW.
1.5) arfs_err: number of times a rule creation or modification has
failed.
2) Supporting inline WQE when possible in SW steering
3) Misc cleanups and fixups to net-next branch
* tag 'mlx5-updates-2023-08-16' of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux:
net/mlx5: Devcom, only use devcom after NULL check in mlx5_devcom_send_event()
net/mlx5: DR, Supporting inline WQE when possible
net/mlx5: Rename devlink port ops struct for PFs/VFs
net/mlx5: Remove VPORT_UPLINK handling from devlink_port.c
net/mlx5: Call mlx5_esw_offloads_rep_load/unload() for uplink port directly
net/mlx5: Update dead links in Kconfig documentation
net/mlx5: Remove health syndrome enum duplication
net/mlx5: DR, Remove unneeded local variable
net/mlx5: DR, Fix code indentation
net/mlx5: IRQ, consolidate irq and affinity mask allocation
net/mlx5e: Fix spelling mistake "Faided" -> "Failed"
net/mlx5e: aRFS, Introduce ethtool stats
net/mlx5e: aRFS, Warn if aRFS table does not exist for aRFS rule
net/mlx5e: aRFS, Prevent repeated kernel rule migrations requests
====================
Link: https://lore.kernel.org/r/20230821175739.81188-1-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
dev_queue_xmit_nit() already uses rcu_read_lock() / rcu_read_unlock()
and nothing suggests that softIRQs should be disabled around it.
Therefore, remove the rcu_read_lock_bh() / rcu_read_unlock_bh()
surrounding it.
Tested using [1] with lockdep enabled.
[1]
#!/bin/bash
ip link add name vrf1 up type vrf table 100
ip link add name veth0 type veth peer name veth1
ip link set dev veth1 master vrf1
ip link set dev veth0 up
ip link set dev veth1 up
ip address add 192.0.2.1/24 dev veth0
ip address add 192.0.2.2/24 dev veth1
ip rule add pref 32765 table local
ip rule del pref 0
tcpdump -i vrf1 -c 20 -w /dev/null &
sleep 10
ping -i 0.1 -c 10 -q 192.0.2.2
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20230821142339.1889961-1-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Commit 264a9c5c9d ("net: sparx5: Remove unused GLAG handling in PGID")
removed sparx5_pgid_alloc_glag() but not its declaration.
Commit 27d293ccee ("net: microchip: sparx5: Add support for rule count by cookie")
removed vcap_rule_iter() but not its declaration.
Commit 8beef08f46 ("net: microchip: sparx5: Adding initial VCAP API support")
declared but never implemented vcap_api_set_client().
Signed-off-by: Yue Haibing <yuehaibing@huawei.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20230821135556.43224-1-yuehaibing@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
abort early so task can exit faster if a fatal signal is pending,
no need to continue validation in that case.
Signed-off-by: Florian Westphal <fw@strlen.de>
Prefer `strscpy_pad` as it's a more robust interface whilst maintaing
zero-padding behavior.
There may have existed a bug here due to both `tbl->repl.name` and
`info->name` having a size of 32 as defined below:
| #define XT_TABLE_MAXNAMELEN 32
This may lead to buffer overreads in some situations -- `strscpy` solves
this by guaranteeing NUL-termination of the dest buffer.
Signed-off-by: Justin Stitt <justinstitt@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Use `strscpy_pad` over `strncpy` for NUL-terminated strings.
We can also drop the + 1 from `NFT_OSF_MAXGENRELEN + 1` since `strscpy`
will guarantee NUL-termination.
Signed-off-by: Justin Stitt <justinstitt@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
When compiling with gcc 13 and CONFIG_FORTIFY_SOURCE=y, the following
warning appears:
In function ‘fortify_memcpy_chk’,
inlined from ‘size_entry_mwt’ at net/bridge/netfilter/ebtables.c:2118:2:
./include/linux/fortify-string.h:592:25: error: call to ‘__read_overflow2_field’
declared with attribute warning: detected read beyond size of field (2nd parameter);
maybe use struct_group()? [-Werror=attribute-warning]
592 | __read_overflow2_field(q_size_field, size);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The compiler is complaining:
memcpy(&offsets[1], &entry->watchers_offset,
sizeof(offsets) - sizeof(offsets[0]));
where memcpy reads beyong &entry->watchers_offset to copy
{watchers,target,next}_offset altogether into offsets[]. Silence the
warning by wrapping these three up via struct_group().
Signed-off-by: GONG, Ruiqi <gongruiqi1@huawei.com>
Reviewed-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Florian Westphal <fw@strlen.de>
When the ifdown function in the dst_ops structure is referenced, the input
parameter 'how' is always true. In the current implementation of the
ifdown interface, ip6_dst_ifdown does not use the input parameter 'how',
xfrm6_dst_ifdown and xfrm4_dst_ifdown functions use the input parameter
'unregister'. But false judgment on 'unregister' in xfrm6_dst_ifdown and
xfrm4_dst_ifdown is false, so remove the input parameter 'how' in ifdown
function.
Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20230821084104.3812233-1-shaozhengchao@huawei.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
The following message shows up when compiling with W=1:
In function ‘fortify_memcpy_chk’,
inlined from ‘alx_get_ethtool_stats’ at drivers/net/ethernet/atheros/alx/ethtool.c:297:2:
./include/linux/fortify-string.h:592:4: error: call to ‘__read_overflow2_field’
declared with attribute warning: detected read beyond size of field (2nd parameter);
maybe use struct_group()? [-Werror=attribute-warning]
592 | __read_overflow2_field(q_size_field, size);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In order to get alx stats altogether, alx_get_ethtool_stats() reads
beyond hw->stats.rx_ok. Fix this warning by directly copying hw->stats,
and refactor the unnecessarily complicated BUILD_BUG_ON btw.
Signed-off-by: GONG, Ruiqi <gongruiqi1@huawei.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://lore.kernel.org/r/20230821013218.1614265-1-gongruiqi@huaweicloud.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
When switching from 10GBase-R/5GBase-R/USXGMII to one of the interface
modes provided by mtk-pcs-lynxi we need to make sure to always perform
a full configuration of the PHYA.
Implement pcs_disable op which resets the stored interface mode to
PHY_INTERFACE_MODE_NA to trigger a full reconfiguration once the LynxI
PCS driver had previously been deselected in favor of another PCS
driver such as the to-be-added driver for the USXGMII PCS found in
MT7988.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Link: https://lore.kernel.org/r/f23d1a60d2c9d2fb72e32dcb0eaa5f7e867a3d68.1692327891.git.daniel@makrotopia.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
There is a warning reported by kernel test robot:
smatch warnings:
drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c:264
mlx5_devcom_send_event() warn: variable dereferenced before
IS_ERR check devcom (see line 259)
The reason for the warning is that the pointer is used before check, put
the assignment to comp after devcom check to silence the warning.
Fixes: 88d162b479 ("net/mlx5: Devcom, Infrastructure changes")
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <error27@gmail.com>
Closes: https://lore.kernel.org/r/202308041028.AkXYDwJ6-lkp@intel.com/
Signed-off-by: Li Zetao <lizetao1@huawei.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
In WQE (Work Queue Entry), the two types of data segments memories are
pointers and inline data, where inline data is passed directly as
part of the WQE.
For software steering, the maximal inline size should be less than
2*MLX5_SEND_WQE_BB, i.e., the potential data must fit with the required
inline WQE headers.
Two consecutive blocks (MLX5_SEND_WQE_BB) are not guaranteed to reside
on the same memory page. Hence, writes to MLX5_SEND_WQE_BB should be
done separately, i.e., each MLX5_SEND_WQE_BB should be obtained using
the mlx5_wq_cyc_get_wqe macro.
Signed-off-by: Itamar Gozlan <igozlan@nvidia.com>
Reviewed-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
As this struct is only used for devlink ports created for PF/VF,
add it to the name of the variable to distinguish from the SF related
ops struct.
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
It is not possible that the functions in devlink_port.c are called for
uplink port. Remove this leftover code.
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
For uplink port, mlx5_esw_offloads_load/unload_rep() are currently
called. There are 2 check inside, which effectively make the
functions a simple wrappers of mlx5_esw_offloads_rep_load/unload()
for uplink port. So avoid one check and indirection and call
mlx5_esw_offloads_rep_load/unload() for uplink port directly.
Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Point to NVIDIA documentation for device specific information now that the
Mellanox documentation site is deprecated. Refer to kernel documentation
sources for generic information not specific to mlx5 devices.
Signed-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Reviewed-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Health syndrome enum values were duplicated in mlx5_ifc and health.c,
the correct place for them is mlx5_ifc.
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Remove local variable that is already defined outside of
the scope of this block.
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Consolidate the mlx5_irq and mlx5_irq->mask allocation, to simplify
error flows and to match the dealloctation sequence @irq_release for
symmetry.
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Improve aRFS observability by adding new set of counters. Each Rx
ring will have this set of counters listed below.
These counters are exposed through ethtool -S.
1) arfs_add: number of times a new rule has been created.
2) arfs_request_in: number of times a rule was requested to move from
its current Rx ring to a new Rx ring (incremented on the destination
Rx ring).
3) arfs_request_out: number of times a rule was requested to move out
from its current Rx ring (incremented on source/current Rx ring).
4) arfs_expired: number of times a rule has been expired by the
kernel and removed from HW.
5) arfs_err: number of times a rule creation or modification has
failed.
This patch removes rx[i]_xsk_arfs_err counter and its documentation in
mlx5/counters.rst since aRFS activity does not occur in XSK RQ's.
Signed-off-by: Adham Faris <afaris@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
aRFS tables should be allocated and exist in advance. Driver shouldn't
reach a point where it tries to add aRFS rule to table that does not
exist.
Add warning if driver encounters such situation.
Signed-off-by: Adham Faris <afaris@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
aRFS rule movement requests from one Rx ring to other Rx ring arrive
from the kernel to ensure that packets are steered to the right Rx ring.
In the time interval until satisfying such a request, several more
requests might follow, for the same flow.
This patch detects and prevents repeated aRFS rules movement requests.
In mlx5e_rx_flow_steer() ndo, after finding the aRFS rule that have been
requested to move by the kernel, check if it's already requested to move
by calling work_busy(&arfs_rule->arfs_work) handler. IOW, if this
request is pending to be executed (in the work queue) or it's executing
now but hasn't finished yet, return current filter ID and don't issue a
new transition work.
Signed-off-by: Adham Faris <afaris@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
