I imagine (tm) that as the number of per-queue configuration
options grows some of them may conflict for certain drivers.
While the drivers can obviously do all the validation locally
doing so is fairly inconvenient as the config is fed to drivers
piecemeal via different ops (for different params and NIC-wide
vs per-queue).
Add a centralized callback for validating the queue config
in queue ops. The callback gets invoked before memory provider
is installed, and in the future should also be called when ring
params are modified.
The validation is done after each layer of configuration.
Since we can't fail MP un-binding we must make sure that
the config is valid both before and after MP overrides are
applied. This is moot for now since the set of MP and device
configs are disjoint. It will matter significantly in the future,
so adding it now so that we don't forget..
Link: https://patch.msgid.link/20260122005113.2476634-6-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
We should follow the prepare/commit approach for queue configuration.
The qcfg struct should be added to dev->cfg rather than directly to
queue objects so that we can clone and discard the pending config
easily.
Remove the qcfg in struct netdev_rx_queue, and switch remaining callers
to netdev_queue_config(). netdev_queue_config() will construct the qcfg
on the fly based on device defaults and state of the queue.
ndo_default_qcfg becomes optional because having the callback itself
does not have any meaningful semantics to us.
Reviewed-by: Dragos Tatulea <dtatulea@nvidia.com>
Link: https://patch.msgid.link/20260122005113.2476634-5-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
We may choose to extend or reimplement the logic which renders
the per-queue config. The drivers should not poke directly into
the queue state. Add a helper for drivers to use when they want
to query the config for a specific queue.
Link: https://patch.msgid.link/20260122005113.2476634-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Pavel Begunkov says:
====================
Add support for providers with large rx buffer
Many modern NICs support configurable receive buffer lengths, and zcrx and
memory providers can use buffers larger than 4K to improve performance.
When paired with hw-gro larger rx buffer sizes can drastically reduce
the number of buffers traversing the stack and save a lot of processing
time. It also allows to give to users larger contiguous chunks of data.
Single stream benchmarks showed up to ~30% CPU util improvement.
E.g. comparison for 4K vs 32K buffers using a 200Gbit NIC:
packets=23987040 (MB=2745098), rps=199559 (MB/s=22837)
CPU %usr %nice %sys %iowait %irq %soft %idle
0 1.53 0.00 27.78 2.72 1.31 66.45 0.22
packets=24078368 (MB=2755550), rps=200319 (MB/s=22924)
CPU %usr %nice %sys %iowait %irq %soft %idle
0 0.69 0.00 8.26 31.65 1.83 57.00 0.57
This series adds net infrastructure for memory providers configuring
the size and implements it for bnxt. It's an opt-in feature for drivers,
they should advertise support for the parameter in the qops and must check
if the hardware supports the given size. It's limited to memory providers
as it drastically simplifies implementation. It doesn't affect the fast
path zcrx uAPI, and the user exposed parameter is defined in zcrx terms,
which allows it to be flexible and adjusted in the future.
A liburing example can be found at [2]
full branch:
[1] https://github.com/isilence/linux.git zcrx/large-buffers-v8
Liburing example:
[2] https://github.com/isilence/liburing.git zcrx/rx-buf-len
* tag 'net-queue-rx-buf-len-v9' of https://github.com/isilence/linux:
io_uring/zcrx: document area chunking parameter
selftests: iou-zcrx: test large chunk sizes
eth: bnxt: support qcfg provided rx page size
eth: bnxt: adjust the fill level of agg queues with larger buffers
eth: bnxt: store rx buffer size per queue
net: pass queue rx page size from memory provider
net: add bare bone queue configs
net: reduce indent of struct netdev_queue_mgmt_ops members
net: memzero mp params when closing a queue
====================
Link: https://patch.msgid.link/
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This reverts commit 77b9c4a438, reversing
changes made to 4515ec4ad5:
931420a2fc ("selftests/net: Add netkit container tests")
ab771c938d ("selftests/net: Make NetDrvContEnv support queue leasing")
6be87fbb27 ("selftests/net: Add env for container based tests")
61d99ce3df ("selftests/net: Add bpf skb forwarding program")
920da36341 ("netkit: Add xsk support for af_xdp applications")
eef51113f8 ("netkit: Add netkit notifier to check for unregistering devices")
b5ef109d22 ("netkit: Implement rtnl_link_ops->alloc and ndo_queue_create")
b5c3fa4a0b ("netkit: Add single device mode for netkit")
0073d2fd67 ("xsk: Proxy pool management for leased queues")
1ecea95dd3 ("xsk: Extend xsk_rcv_check validation")
804bf334d0 ("net: Proxy netdev_queue_get_dma_dev for leased queues")
0caa9a8dde ("net: Proxy net_mp_{open,close}_rxq for leased queues")
ff8889ff91 ("net, ethtool: Disallow leased real rxqs to be resized")
9e2103f361 ("net: Add lease info to queue-get response")
31127dedde ("net: Implement netdev_nl_queue_create_doit")
a5546e18f7 ("net: Add queue-create operation")
The series will conflict with io_uring work, and the code needs more
polish.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Implement netdev_nl_queue_create_doit which creates a new rx queue in a
virtual netdev and then leases it to a rx queue in a physical netdev.
Example with ynl client:
# ./pyynl/cli.py \
--spec ~/netlink/specs/netdev.yaml \
--do queue-create \
--json '{"ifindex": 8, "type": "rx", "lease": {"ifindex": 4, "queue": {"type": "rx", "id": 15}}}'
{'id': 1}
Note that the netdevice locking order is always from the virtual to
the physical device.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Co-developed-by: David Wei <dw@davidwei.uk>
Signed-off-by: David Wei <dw@davidwei.uk>
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Link: https://patch.msgid.link/20260115082603.219152-3-daniel@iogearbox.net
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Allow memory providers to configure rx queues with a custom receive
page size. It's passed in struct pp_memory_provider_params, which is
copied into the queue, so it's preserved across queue restarts. Then,
it's propagated to the driver in a new queue config parameter.
Drivers should explicitly opt into using it by setting
QCFG_RX_PAGE_SIZE, in which case they should implement ndo_default_qcfg,
validate the size on queue restart and honour the current config in case
of a reset.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
We'll need to pass extra parameters when allocating a queue for memory
providers. Define a new structure for queue configurations, and pass it
to qapi callbacks. It's empty for now, actual parameters will be added
in following patches.
Configurations should persist across resets, and for that they're
default-initialised on device registration and stored in struct
netdev_rx_queue. We also add a new qapi callback for defaulting a given
config. It must be implemented if a driver wants to use queue configs
and is optional otherwise.
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Trivial change, reduce the indent. I think the original is copied
from real NDOs. It's unnecessarily deep, makes passing struct args
problematic.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Mina Almasry <almasrymina@google.com>
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
For zerocopy (io_uring, devmem), there is an assumption that the
parent device can do DMA. However that is not always the case:
- Scalable Function netdevs [1] have the DMA device in the grandparent.
- For Multi-PF netdevs [2] queues can be associated to different DMA
devices.
This patch introduces the a queue based interface for allowing drivers
to expose a different DMA device for zerocopy.
[1] Documentation/networking/device_drivers/ethernet/mellanox/mlx5/switchdev.rst
[2] Documentation/networking/multi-pf-netdev.rst
Signed-off-by: Dragos Tatulea <dtatulea@nvidia.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com>
Reviewed-by: Mina Almasry <almasrymina@google.com>
Link: https://patch.msgid.link/20250827144017.1529208-3-dtatulea@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add a new function, netif_subqueue_sent, which is a wrapper for
netdev_tx_sent_queue.
Drivers that use the subqueue variant macros, netif_subqueue_xxx,
identify queue by index and are not required to obtain
struct netdev_queue explicitly.
Such drivers still need to call netdev_tx_sent_queue which is a
counterpart of netif_subqueue_completed_wake. Allowing drivers to use a
subqueue variant for this purpose improves their code consistency by
always referring to queue by its index.
Signed-off-by: Gur Stavi <gur.stavi@huawei.com>
Link: https://patch.msgid.link/909a5c92db49cad39f0954d6cb86775e6480ef4c.1749038081.git.gur.stavi@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Older drivers and drivers with lower queue counts often have a static
array of queues, rather than allocating structs for each queue on demand.
Add a helper for adding up qstats from a queue range. Expectation is
that driver will pass a queue range [netdev->real_num_*x_queues, MAX).
It was tempting to always use num_*x_queues as the end, but virtio
seems to clamp its queue count after allocating the netdev. And this
way we can trivaly reuse the helper for [0, real_..).
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Link: https://patch.msgid.link/20250507003221.823267-2-kuba@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
We mostly needed rtnl_lock in qstat to make sure the queue count
is stable while we work. For "ops locked" drivers the instance
lock protects the queue count, so we don't have to take rtnl_lock.
For currently ops-locked drivers: netdevsim and bnxt need
the protection from netdev going down while we dump, which
instance lock provides. gve doesn't care.
Reviewed-by: Joe Damato <jdamato@fastly.com>
Acked-by: Stanislav Fomichev <sdf@fomichev.me>
Link: https://patch.msgid.link/20250408195956.412733-9-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
We seem to be missing a netif_running() check from the devmem
installation path. Starting a queue on a stopped device makes
no sense. We still want to be able to allocate the memory, just
to test that the device is indeed setting up the page pools
in a memory provider compatible way.
This is not a bug fix, because existing drivers check if
the interface is down as part of the ops. But new drivers
shouldn't have to do this, as long as they can correctly
alloc/free while down.
Reviewed-by: Mina Almasry <almasrymina@google.com>
Link: https://patch.msgid.link/20250206225638.1387810-3-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The ethtool-nl family does a good job exposing various protocol
related and IEEE/IETF statistics which used to get dumped under
ethtool -S, with creative names. Queue stats don't have a netlink
API, yet, and remain a lion's share of ethtool -S output for new
drivers. Not only is that bad because the names differ driver to
driver but it's also bug-prone. Intuitively drivers try to report
only the stats for active queues, but querying ethtool stats
involves multiple system calls, and the number of stats is
read separately from the stats themselves. Worse still when user
space asks for values of the stats, it doesn't inform the kernel
how big the buffer is. If number of stats increases in the meantime
kernel will overflow user buffer.
Add a netlink API for dumping queue stats. Queue information is
exposed via the netdev-genl family, so add the stats there.
Support per-queue and sum-for-device dumps. Latter will be useful
when subsequent patches add more interesting common stats than
just bytes and packets.
The API does not currently distinguish between HW and SW stats.
The expectation is that the source of the stats will either not
matter much (good packets) or be obvious (skb alloc errors).
Acked-by: Stanislav Fomichev <sdf@google.com>
Reviewed-by: Amritha Nambiar <amritha.nambiar@intel.com>
Reviewed-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Link: https://lore.kernel.org/r/20240306195509.1502746-2-kuba@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
netif_txq_try_stop() uses "get_desc >= start_thrs" as the check for
the call to netif_tx_start_queue().
Use ">=" i netdev_txq_completed_mb(), too.
Fixes: c91c46de6b ("net: provide macros for commonly copied lockless queue stop/wake code")
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Drivers call netdev_tx_completed_queue() right before
netif_txq_maybe_wake(). If BQL is enabled netdev_tx_completed_queue()
should issue a memory barrier, so we can depend on that separating
the stop check from the consumer index update, instead of adding
another barrier in netif_txq_maybe_wake().
This matters more than the barriers on the xmit path, because
the wake condition is almost always true. So we issue the
consumer side barrier often.
Wrap netdev_tx_completed_queue() in a local helper to issue
the barrier even if BQL is disabled. Keep the same semantics
as netdev_tx_completed_queue() (barrier only if bytes != 0)
to make it clear that the barrier is conditional.
Plus since macro gets pkt/byte counts as arguments now -
we can skip waking if there were no packets completed.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
A lot of drivers follow the same scheme to stop / start queues
without introducing locks between xmit and NAPI tx completions.
I'm guessing they all copy'n'paste each other's code.
The original code dates back all the way to e1000 and Linux 2.6.19.
Smaller drivers shy away from the scheme and introduce a lock
which may cause deadlocks in netpoll.
Provide macros which encapsulate the necessary logic.
The macros do not prevent false wake ups, the extra barrier
required to close that race is not worth it. See discussion in:
https://lore.kernel.org/all/c39312a2-4537-14b4-270c-9fe1fbb91e89@gmail.com/
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
