Revert "net: replace dev_addr_sem with netdev instance lock"

This reverts commit df43d8bf10316a7c3b1e47e3cc0057a54df4a5b8.

Cc: Kohei Enju <[email protected]>
Reviewed-by: Kuniyuki Iwashima <kuniyu@a

Revert "net: replace dev_addr_sem with netdev instance lock"

This reverts commit df43d8bf10316a7c3b1e47e3cc0057a54df4a5b8.

Cc: Kohei Enju <[email protected]>
Reviewed-by: Kuniyuki Iwashima <[email protected]>
Fixes: df43d8bf1031 ("net: replace dev_addr_sem with netdev instance lock")
Signed-off-by: Stanislav Fomichev <[email protected]>
Link: https://patch.msgid.link/[email protected]
Tested-by: Lei Yang <[email protected]>
Signed-off-by: Paolo Abeni <[email protected]>

show more ...

net: move misc netdev_lock flavors to a separate header

Move the more esoteric helpers for netdev instance lock to
a dedicated header. This avoids growing netdevice.h to infinity
and makes rebuildin

net: move misc netdev_lock flavors to a separate header

Move the more esoteric helpers for netdev instance lock to
a dedicated header. This avoids growing netdevice.h to infinity
and makes rebuilding the kernel much faster (after touching
the header with the helpers).

The main netdev_lock() / netdev_unlock() functions are used
in static inlines in netdevice.h and will probably be used
most commonly, so keep them in netdevice.h.

Acked-by: Stanislav Fomichev <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: replace dev_addr_sem with netdev instance lock

Lockdep reports possible circular dependency in [0]. Instead of
fixing the ordering, replace global dev_addr_sem with netdev
instance lock. Most o

net: replace dev_addr_sem with netdev instance lock

Lockdep reports possible circular dependency in [0]. Instead of
fixing the ordering, replace global dev_addr_sem with netdev
instance lock. Most of the paths that set/get mac are RTNL
protected. Two places where it's not, convert to explicit
locking:
- sysfs address_show
- dev_get_mac_address via dev_ioctl

0: https://netdev-3.bots.linux.dev/vmksft-forwarding-dbg/results/993321/24-router-bridge-1d-lag-sh/stderr

Signed-off-by: Stanislav Fomichev <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: hold netdev instance lock during rtnetlink operations

To preserve the atomicity, hold the lock while applying multiple
attributes. The major issue with a full conversion to the instance
lock ar

net: hold netdev instance lock during rtnetlink operations

To preserve the atomicity, hold the lock while applying multiple
attributes. The major issue with a full conversion to the instance
lock are software nesting devices (bonding/team/vrf/etc). Those
devices call into the core stack for their lower (potentially
real hw) devices. To avoid explicitly wrapping all those places
into instance lock/unlock, introduce new API boundaries:

- (some) existing dev_xxx calls are now considered "external"
(to drivers) APIs and they transparently grab the instance
lock if needed (dev_api.c)
- new netif_xxx calls are internal core stack API (naming is
sketchy, I've tried netdev_xxx_locked per Jakub's suggestion,
but it feels a bit verbose; but happy to get back to this
naming scheme if this is the preference)

This avoids touching most of the existing ioctl/sysfs/drivers paths.

Note the special handling of ndo_xxx_slave operations: I exploit
the fact that none of the drivers that call these functions
need/use instance lock. At the same time, they use dev_xxx
APIs, so the lower device has to be unlocked.

Changes in unregister_netdevice_many_notify (to protect dev->state
with instance lock) trigger lockdep - the loop over close_list
(mostly from cleanup_net) introduces spurious ordering issues.
netdev_lock_cmp_fn has a justification on why it's ok to suppress
for now.

Cc: Saeed Mahameed <[email protected]>
Signed-off-by: Stanislav Fomichev <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: hold netdev instance lock during ndo_open/ndo_stop

For the drivers that use shaper API, switch to the mode where
core stack holds the netdev lock. This affects two drivers:

* iavf - already gr

net: hold netdev instance lock during ndo_open/ndo_stop

For the drivers that use shaper API, switch to the mode where
core stack holds the netdev lock. This affects two drivers:

* iavf - already grabs netdev lock in ndo_open/ndo_stop, so mostly
remove these
* netdevsim - switch to _locked APIs to avoid deadlock

iavf_close diff is a bit confusing, the existing call looks like this:
iavf_close() {
netdev_lock()
..
netdev_unlock()
wait_event_timeout(down_waitqueue)
}

I change it to the following:
netdev_lock()
iavf_close() {
..
netdev_unlock()
wait_event_timeout(down_waitqueue)
netdev_lock() // reusing this lock call
}
netdev_unlock()

Since I'm reusing existing netdev_lock call, so it looks like I only
add netdev_unlock.

Cc: Saeed Mahameed <[email protected]>
Signed-off-by: Stanislav Fomichev <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: page_pool: avoid false positive warning if NAPI was never added

We expect NAPI to be in disabled state when page pool is torn down.
But it is also legal if the NAPI is completely uninitialized.

net: page_pool: avoid false positive warning if NAPI was never added

We expect NAPI to be in disabled state when page pool is torn down.
But it is also legal if the NAPI is completely uninitialized.

Reviewed-by: Mina Almasry <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: make netdev netlink ops hold netdev_lock()

In prep for dropping rtnl_lock, start locking netdev->lock in netlink
genl ops. We need to be using netdev->up instead of flags & IFF_UP.

We can remo

net: make netdev netlink ops hold netdev_lock()

In prep for dropping rtnl_lock, start locking netdev->lock in netlink
genl ops. We need to be using netdev->up instead of flags & IFF_UP.

We can remove the RCU lock protection for the NAPI since NAPI list
is protected by netdev->lock already.

Reviewed-by: Eric Dumazet <[email protected]>
Reviewed-by: Kuniyuki Iwashima <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: add netdev->up protected by netdev_lock()

Some uAPI (netdev netlink) hide net_device's sub-objects while
the interface is down to ensure uniform behavior across drivers.
To remove the rtnl_lock

net: add netdev->up protected by netdev_lock()

Some uAPI (netdev netlink) hide net_device's sub-objects while
the interface is down to ensure uniform behavior across drivers.
To remove the rtnl_lock dependency from those uAPIs we need a way
to safely tell if the device is down or up.

Add an indication of whether device is open or closed, protected
by netdev->lock. The semantics are the same as IFF_UP, but taking
netdev_lock around every write to ->flags would be a lot of code
churn.

We don't want to blanket the entire open / close path by netdev_lock,
because it will prevent us from applying it to specific structures -
core helpers won't be able to take that lock from any function
called by the drivers on open/close paths.

So the state of the flag is "pessimistic", as in it may report false
negatives, but never false positives.

Reviewed-by: Joe Damato <[email protected]>
Reviewed-by: Eric Dumazet <[email protected]>
Reviewed-by: Kuniyuki Iwashima <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: add helpers for lookup and walking netdevs under netdev_lock()

Add helpers for accessing netdevs under netdev_lock().
There's some careful handling needed to find the device and lock it
safely,

net: add helpers for lookup and walking netdevs under netdev_lock()

Add helpers for accessing netdevs under netdev_lock().
There's some careful handling needed to find the device and lock it
safely, without it getting unregistered, and without taking rtnl_lock
(the latter being the whole point of the new locking, after all).

Reviewed-by: Eric Dumazet <[email protected]>
Reviewed-by: Kuniyuki Iwashima <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: hide the definition of dev_get_by_napi_id()

There are no module callers of dev_get_by_napi_id(),
and commit d1cacd747768 ("netdev: prevent accessing NAPI instances
from another namespace") prov

net: hide the definition of dev_get_by_napi_id()

There are no module callers of dev_get_by_napi_id(),
and commit d1cacd747768 ("netdev: prevent accessing NAPI instances
from another namespace") proves that getting NAPI by id
needs to be done with care. So hide dev_get_by_napi_id().

Reviewed-by: Jacob Keller <[email protected]>
Reviewed-by: Kalesh AP <[email protected]>
Reviewed-by: Joe Damato <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

netdev: prevent accessing NAPI instances from another namespace

The NAPI IDs were not fully exposed to user space prior to the netlink
API, so they were never namespaced. The netlink API must ensure

netdev: prevent accessing NAPI instances from another namespace

The NAPI IDs were not fully exposed to user space prior to the netlink
API, so they were never namespaced. The netlink API must ensure that
at the very least NAPI instance belongs to the same netns as the owner
of the genl sock.

napi_by_id() can become static now, but it needs to move because of
dev_get_by_napi_id().

Cc: [email protected]
Fixes: 1287c1ae0fc2 ("netdev-genl: Support setting per-NAPI config values")
Fixes: 27f91aaf49b3 ("netdev-genl: Add netlink framework functions for napi")
Reviewed-by: Sridhar Samudrala <[email protected]>
Reviewed-by: Joe Damato <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: Make net_hwtstamp_validate accessible

Make the net_hwtstamp_validate function accessible in prevision to use
it from ethtool to validate the hwtstamp configuration before setting it.

Reviewed-

net: Make net_hwtstamp_validate accessible

Make the net_hwtstamp_validate function accessible in prevision to use
it from ethtool to validate the hwtstamp configuration before setting it.

Reviewed-by: Florian Fainelli <[email protected]>
Reviewed-by: Jacob Keller <[email protected]>
Signed-off-by: Kory Maincent <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

show more ...

net: Make dev_get_hwtstamp_phylib accessible

Make the dev_get_hwtstamp_phylib function accessible in prevision to use
it from ethtool to read the hwtstamp current configuration.

Reviewed-by: Floria

net: Make dev_get_hwtstamp_phylib accessible

Make the dev_get_hwtstamp_phylib function accessible in prevision to use
it from ethtool to read the hwtstamp current configuration.

Reviewed-by: Florian Fainelli <[email protected]>
Reviewed-by: Jacob Keller <[email protected]>
Signed-off-by: Kory Maincent <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

show more ...

net: Add napi_struct parameter irq_suspend_timeout

Add a per-NAPI IRQ suspension parameter, which can be get/set with
netdev-genl.

This patch doesn't change any behavior but prepares the code for o

net: Add napi_struct parameter irq_suspend_timeout

Add a per-NAPI IRQ suspension parameter, which can be get/set with
netdev-genl.

This patch doesn't change any behavior but prepares the code for other
changes in the following commits which use irq_suspend_timeout as a
timeout for IRQ suspension.

Signed-off-by: Martin Karsten <[email protected]>
Co-developed-by: Joe Damato <[email protected]>
Signed-off-by: Joe Damato <[email protected]>
Tested-by: Joe Damato <[email protected]>
Tested-by: Martin Karsten <[email protected]>
Acked-by: Stanislav Fomichev <[email protected]>
Reviewed-by: Sridhar Samudrala <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: napi: Add napi_config

Add a persistent NAPI config area for NAPI configuration to the core.
Drivers opt-in to setting the persistent config for a NAPI by passing an
index when calling netif_nap

net: napi: Add napi_config

Add a persistent NAPI config area for NAPI configuration to the core.
Drivers opt-in to setting the persistent config for a NAPI by passing an
index when calling netif_napi_add_config.

napi_config is allocated in alloc_netdev_mqs, freed in free_netdev
(after the NAPIs are deleted).

Drivers which call netif_napi_add_config will have persistent per-NAPI
settings: NAPI IDs, gro_flush_timeout, and defer_hard_irq settings.

Per-NAPI settings are saved in napi_disable and restored in napi_enable.

Co-developed-by: Martin Karsten <[email protected]>
Signed-off-by: Martin Karsten <[email protected]>
Signed-off-by: Joe Damato <[email protected]>
Reviewed-by: Jakub Kicinski <[email protected]>
Reviewed-by: Eric Dumazet <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: napi: Make gro_flush_timeout per-NAPI

Allow per-NAPI gro_flush_timeout setting.

The existing sysfs parameter is respected; writes to sysfs will write to
all NAPI structs for the device and the

net: napi: Make gro_flush_timeout per-NAPI

Allow per-NAPI gro_flush_timeout setting.

The existing sysfs parameter is respected; writes to sysfs will write to
all NAPI structs for the device and the net_device gro_flush_timeout
field. Reads from sysfs will read from the net_device field.

The ability to set gro_flush_timeout on specific NAPI instances will be
added in a later commit, via netdev-genl.

Signed-off-by: Joe Damato <[email protected]>
Reviewed-by: Eric Dumazet <[email protected]>
Reviewed-by: Jakub Kicinski <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: napi: Make napi_defer_hard_irqs per-NAPI

Add defer_hard_irqs to napi_struct in preparation for per-NAPI
settings.

The existing sysfs parameter is respected; writes to sysfs will write to
all N

net: napi: Make napi_defer_hard_irqs per-NAPI

Add defer_hard_irqs to napi_struct in preparation for per-NAPI
settings.

The existing sysfs parameter is respected; writes to sysfs will write to
all NAPI structs for the device and the net_device defer_hard_irq field.
Reads from sysfs show the net_device field.

The ability to set defer_hard_irqs on specific NAPI instances will be
added in a later commit, via netdev-genl.

Signed-off-by: Joe Damato <[email protected]>
Reviewed-by: Eric Dumazet <[email protected]>
Reviewed-by: Jakub Kicinski <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net-shapers: implement shaper cleanup on queue deletion

hook into netif_set_real_num_tx_queues() to cleanup any shaper
configured on top of the to-be-destroyed TX queues.

Reviewed-by: Jiri Pirko <j

net-shapers: implement shaper cleanup on queue deletion

hook into netif_set_real_num_tx_queues() to cleanup any shaper
configured on top of the to-be-destroyed TX queues.

Reviewed-by: Jiri Pirko <[email protected]>
Reviewed-by: Jakub Kicinski <[email protected]>
Signed-off-by: Paolo Abeni <[email protected]>
Link: https://patch.msgid.link/6da4ee03cae2b2a757d7b59e88baf09cc94c5ef1.1728460186.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net-shapers: implement NL get operation

Introduce the basic infrastructure to implement the net-shaper
core functionality. Each network devices carries a net-shaper cache,
the NL get() operation fet

net-shapers: implement NL get operation

Introduce the basic infrastructure to implement the net-shaper
core functionality. Each network devices carries a net-shaper cache,
the NL get() operation fetches the data from such cache.

The cache is initially empty, will be fill by the set()/group()
operation implemented later and is destroyed at device cleanup time.

The net_shaper_fill_handle(), net_shaper_ctx_init(), and
net_shaper_generic_pre() implementations handle generic index type
attributes, despite the current caller always pass a constant value
to avoid more noise in later patches using them with different
attributes.

Reviewed-by: Jakub Kicinski <[email protected]>
Reviewed-by: Jiri Pirko <[email protected]>
Signed-off-by: Paolo Abeni <[email protected]>
Link: https://patch.msgid.link/ddd10fd645a9367803ad02fca4a5664ea5ace170.1728460186.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: softnet_data: Make xmit per task.

Softirq is preemptible on PREEMPT_RT. Without a per-CPU lock in
local_bh_disable() there is no guarantee that only one device is
transmitting at a time.
With p

net: softnet_data: Make xmit per task.

Softirq is preemptible on PREEMPT_RT. Without a per-CPU lock in
local_bh_disable() there is no guarantee that only one device is
transmitting at a time.
With preemption and multiple senders it is possible that the per-CPU
`recursion' counter gets incremented by different threads and exceeds
XMIT_RECURSION_LIMIT leading to a false positive recursion alert.
The `more' member is subject to similar problems if set by one thread
for one driver and wrongly used by another driver within another thread.

Instead of adding a lock to protect the per-CPU variable it is simpler
to make xmit per-task. Sending and receiving skbs happens always
in thread context anyway.

Having a lock to protected the per-CPU counter would block/ serialize two
sending threads needlessly. It would also require a recursive lock to
ensure that the owner can increment the counter further.

Make the softnet_data.xmit a task_struct member on PREEMPT_RT. Add
needed wrapper.

Cc: Ben Segall <[email protected]>
Cc: Daniel Bristot de Oliveira <[email protected]>
Cc: Dietmar Eggemann <[email protected]>
Cc: Juri Lelli <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Steven Rostedt <[email protected]>
Cc: Valentin Schneider <[email protected]>
Cc: Vincent Guittot <[email protected]>
Signed-off-by: Sebastian Andrzej Siewior <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: Move dev_set_hwtstamp_phylib to net/core/dev.h

This declaration was added to the header to be called from ethtool.
ethtool is separated from core for code organization but it is not really
a se

net: Move dev_set_hwtstamp_phylib to net/core/dev.h

This declaration was added to the header to be called from ethtool.
ethtool is separated from core for code organization but it is not really
a separate entity, it controls very core things.
As ethtool is an internal stuff it is not wise to have it in netdevice.h.
Move the declaration to net/core/dev.h instead.

Remove the EXPORT_SYMBOL_GPL call as ethtool can not be built as a module.

Reviewed-by: Willem de Bruijn <[email protected]>
Signed-off-by: Kory Maincent <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: move sysctl_skb_defer_max to net_hotdata

sysctl_skb_defer_max is used in TCP fast path,
move it to net_hodata.

Signed-off-by: Eric Dumazet <[email protected]>
Reviewed-by: David Ahern <dsahe

net: move sysctl_skb_defer_max to net_hotdata

sysctl_skb_defer_max is used in TCP fast path,
move it to net_hodata.

Signed-off-by: Eric Dumazet <[email protected]>
Reviewed-by: David Ahern <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

net: move dev_xmit_recursion() helpers to net/core/dev.h

Move dev_xmit_recursion() and friends to net/core/dev.h

They are only used from net/core/dev.c and net/core/filter.c.

Signed-off-by: Eric D

net: move dev_xmit_recursion() helpers to net/core/dev.h

Move dev_xmit_recursion() and friends to net/core/dev.h

They are only used from net/core/dev.c and net/core/filter.c.

Signed-off-by: Eric Dumazet <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

show more ...

net: move kick_defer_list_purge() to net/core/dev.h

kick_defer_list_purge() is defined in net/core/dev.c
and used from net/core/skubff.c

Because we need softnet_data, include <linux/netdevice.h>
fr

net: move kick_defer_list_purge() to net/core/dev.h

kick_defer_list_purge() is defined in net/core/dev.c
and used from net/core/skubff.c

Because we need softnet_data, include <linux/netdevice.h>
from net/core/dev.h

Signed-off-by: Eric Dumazet <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

show more ...

net: move netdev_tstamp_prequeue into net_hotdata

netdev_tstamp_prequeue is used in rx path.

Move it to net_hotdata for better cache locality.

Signed-off-by: Eric Dumazet <[email protected]>
Ack

net: move netdev_tstamp_prequeue into net_hotdata

netdev_tstamp_prequeue is used in rx path.

Move it to net_hotdata for better cache locality.

Signed-off-by: Eric Dumazet <[email protected]>
Acked-by: Soheil Hassas Yeganeh <[email protected]>
Reviewed-by: David Ahern <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...