domenico/openwrt-R7800-nss
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

generic: 6.6: replace (broken) downstream patch with upstream solution

Browse Source 
Our downstream patch "net/core: add optional threading for backlog processing"
has been broken with the switch to Linux 6.6.
Replace it by backporting the now available upstream solution.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Link: https://github.com/openwrt/openwrt/pull/15592
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
This commit is contained in:
Daniel Golle
2024-05-30 12:33:37 +01:00
committed by Christian Marangi
parent f3080677f5
commit a5c095c453
7 changed files with 696 additions and 233 deletions
Download Patch File Download Diff File Expand all files Collapse all files

75
target/linux/generic/backport-6.6/600-v6.10-net-Remove-conditional-threaded-NAPI-wakeup-based-on.patch Normal file
View File

@@ -0,0 +1,75 @@
From 56364c910691f6d10ba88c964c9041b9ab777bd6 Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Mon, 25 Mar 2024 08:40:28 +0100
Subject: [PATCH 1/4] net: Remove conditional threaded-NAPI wakeup based on
task state.
A NAPI thread is scheduled by first setting NAPI_STATE_SCHED bit. If
successful (the bit was not yet set) then the NAPI_STATE_SCHED_THREADED
is set but only if thread's state is not TASK_INTERRUPTIBLE (is
TASK_RUNNING) followed by task wakeup.
If the task is idle (TASK_INTERRUPTIBLE) then the
NAPI_STATE_SCHED_THREADED bit is not set. The thread is no relying on
the bit but always leaving the wait-loop after returning from schedule()
because there must have been a wakeup.
The smpboot-threads implementation for per-CPU threads requires an
explicit condition and does not support "if we get out of schedule()
then there must be something to do".
Removing this optimisation simplifies the following integration.
Set NAPI_STATE_SCHED_THREADED unconditionally on wakeup and rely on it
in the wait path by removing the `woken' condition.
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
net/core/dev.c | 14 ++------------
1 file changed, 2 insertions(+), 12 deletions(-)
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4473,13 +4473,7 @@ static inline void ____napi_schedule(str
*/
thread = READ_ONCE(napi->thread);
if (thread) {
- /* Avoid doing set_bit() if the thread is in
- * INTERRUPTIBLE state, cause napi_thread_wait()
- * makes sure to proceed with napi polling
- * if the thread is explicitly woken from here.
- */
- if (READ_ONCE(thread->__state) != TASK_INTERRUPTIBLE)
- set_bit(NAPI_STATE_SCHED_THREADED, &napi->state);
+ set_bit(NAPI_STATE_SCHED_THREADED, &napi->state);
wake_up_process(thread);
return;
}
@@ -6635,8 +6629,6 @@ static int napi_poll(struct napi_struct
static int napi_thread_wait(struct napi_struct *napi)
{
- bool woken = false;
-
set_current_state(TASK_INTERRUPTIBLE);
while (!kthread_should_stop()) {
@@ -6645,15 +6637,13 @@ static int napi_thread_wait(struct napi_
* Testing SCHED bit is not enough because SCHED bit might be
* set by some other busy poll thread or by napi_disable().
*/
- if (test_bit(NAPI_STATE_SCHED_THREADED, &napi->state) || woken) {
+ if (test_bit(NAPI_STATE_SCHED_THREADED, &napi->state)) {
WARN_ON(!list_empty(&napi->poll_list));
__set_current_state(TASK_RUNNING);
return 0;
}
schedule();
- /* woken being true indicates this thread owns this napi. */
- woken = true;
set_current_state(TASK_INTERRUPTIBLE);
}
__set_current_state(TASK_RUNNING);

330
target/linux/generic/backport-6.6/601-v6.10-net-Allow-to-use-SMP-threads-for-backlog-NAPI.patch Normal file
View File

@@ -0,0 +1,330 @@
From dad6b97702639fba27a2bd3e986982ad6f0db3a7 Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Mon, 25 Mar 2024 08:40:29 +0100
Subject: [PATCH 2/4] net: Allow to use SMP threads for backlog NAPI.
Backlog NAPI is a per-CPU NAPI struct only (with no device behind it)
used by drivers which don't do NAPI them self, RPS and parts of the
stack which need to avoid recursive deadlocks while processing a packet.
The non-NAPI driver use the CPU local backlog NAPI. If RPS is enabled
then a flow for the skb is computed and based on the flow the skb can be
enqueued on a remote CPU. Scheduling/ raising the softirq (for backlog's
NAPI) on the remote CPU isn't trivial because the softirq is only
scheduled on the local CPU and performed after the hardirq is done.
In order to schedule a softirq on the remote CPU, an IPI is sent to the
remote CPU which schedules the backlog-NAPI on the then local CPU.
On PREEMPT_RT interrupts are force-threaded. The soft interrupts are
raised within the interrupt thread and processed after the interrupt
handler completed still within the context of the interrupt thread. The
softirq is handled in the context where it originated.
With force-threaded interrupts enabled, ksoftirqd is woken up if a
softirq is raised from hardirq context. This is the case if it is raised
from an IPI. Additionally there is a warning on PREEMPT_RT if the
softirq is raised from the idle thread.
This was done for two reasons:
- With threaded interrupts the processing should happen in thread
context (where it originated) and ksoftirqd is the only thread for
this context if raised from hardirq. Using the currently running task
instead would "punish" a random task.
- Once ksoftirqd is active it consumes all further softirqs until it
stops running. This changed recently and is no longer the case.
Instead of keeping the backlog NAPI in ksoftirqd (in force-threaded/
PREEMPT_RT setups) I am proposing NAPI-threads for backlog.
The "proper" setup with threaded-NAPI is not doable because the threads
are not pinned to an individual CPU and can be modified by the user.
Additionally a dummy network device would have to be assigned. Also
CPU-hotplug has to be considered if additional CPUs show up.
All this can be probably done/ solved but the smpboot-threads already
provide this infrastructure.
Sending UDP packets over loopback expects that the packet is processed
within the call. Delaying it by handing it over to the thread hurts
performance. It is not beneficial to the outcome if the context switch
happens immediately after enqueue or after a while to process a few
packets in a batch.
There is no need to always use the thread if the backlog NAPI is
requested on the local CPU. This restores the loopback throuput. The
performance drops mostly to the same value after enabling RPS on the
loopback comparing the IPI and the tread result.
Create NAPI-threads for backlog if request during boot. The thread runs
the inner loop from napi_threaded_poll(), the wait part is different. It
checks for NAPI_STATE_SCHED (the backlog NAPI can not be disabled).
The NAPI threads for backlog are optional, it has to be enabled via the boot
argument "thread_backlog_napi". It is mandatory for PREEMPT_RT to avoid the
wakeup of ksoftirqd from the IPI.
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
net/core/dev.c | 148 +++++++++++++++++++++++++++++++++++++------------
1 file changed, 113 insertions(+), 35 deletions(-)
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -78,6 +78,7 @@
#include <linux/slab.h>
#include <linux/sched.h>
#include <linux/sched/mm.h>
+#include <linux/smpboot.h>
#include <linux/mutex.h>
#include <linux/rwsem.h>
#include <linux/string.h>
@@ -217,6 +218,31 @@ static inline struct hlist_head *dev_ind
return &net->dev_index_head[ifindex & (NETDEV_HASHENTRIES - 1)];
}
+#ifndef CONFIG_PREEMPT_RT
+
+static DEFINE_STATIC_KEY_FALSE(use_backlog_threads_key);
+
+static int __init setup_backlog_napi_threads(char *arg)
+{
+ static_branch_enable(&use_backlog_threads_key);
+ return 0;
+}
+early_param("thread_backlog_napi", setup_backlog_napi_threads);
+
+static bool use_backlog_threads(void)
+{
+ return static_branch_unlikely(&use_backlog_threads_key);
+}
+
+#else
+
+static bool use_backlog_threads(void)
+{
+ return true;
+}
+
+#endif
+
static inline void rps_lock_irqsave(struct softnet_data *sd,
unsigned long *flags)
{
@@ -4441,6 +4467,7 @@ EXPORT_SYMBOL(__dev_direct_xmit);
/*************************************************************************
* Receiver routines
*************************************************************************/
+static DEFINE_PER_CPU(struct task_struct *, backlog_napi);
int netdev_max_backlog __read_mostly = 1000;
EXPORT_SYMBOL(netdev_max_backlog);
@@ -4473,12 +4500,16 @@ static inline void ____napi_schedule(str
*/
thread = READ_ONCE(napi->thread);
if (thread) {
+ if (use_backlog_threads() && thread == raw_cpu_read(backlog_napi))
+ goto use_local_napi;
+
set_bit(NAPI_STATE_SCHED_THREADED, &napi->state);
wake_up_process(thread);
return;
}
}
+use_local_napi:
list_add_tail(&napi->poll_list, &sd->poll_list);
WRITE_ONCE(napi->list_owner, smp_processor_id());
/* If not called from net_rx_action()
@@ -4724,6 +4755,11 @@ static void napi_schedule_rps(struct sof
#ifdef CONFIG_RPS
if (sd != mysd) {
+ if (use_backlog_threads()) {
+ __napi_schedule_irqoff(&sd->backlog);
+ return;
+ }
+
sd->rps_ipi_next = mysd->rps_ipi_list;
mysd->rps_ipi_list = sd;
@@ -5947,7 +5983,7 @@ static void net_rps_action_and_irq_enabl
#ifdef CONFIG_RPS
struct softnet_data *remsd = sd->rps_ipi_list;
- if (remsd) {
+ if (!use_backlog_threads() && remsd) {
sd->rps_ipi_list = NULL;
local_irq_enable();
@@ -5962,7 +5998,7 @@ static void net_rps_action_and_irq_enabl
static bool sd_has_rps_ipi_waiting(struct softnet_data *sd)
{
#ifdef CONFIG_RPS
- return sd->rps_ipi_list != NULL;
+ return !use_backlog_threads() && sd->rps_ipi_list;
#else
return false;
#endif
@@ -6006,7 +6042,7 @@ static int process_backlog(struct napi_s
* We can use a plain write instead of clear_bit(),
* and we dont need an smp_mb() memory barrier.
*/
- napi->state = 0;
+ napi->state &= NAPIF_STATE_THREADED;
again = false;
} else {
skb_queue_splice_tail_init(&sd->input_pkt_queue,
@@ -6672,43 +6708,48 @@ static void skb_defer_free_flush(struct
}
}
-static int napi_threaded_poll(void *data)
+static void napi_threaded_poll_loop(struct napi_struct *napi)
{
- struct napi_struct *napi = data;
struct softnet_data *sd;
- void *have;
+ unsigned long last_qs = jiffies;
- while (!napi_thread_wait(napi)) {
- unsigned long last_qs = jiffies;
+ for (;;) {
+ bool repoll = false;
+ void *have;
- for (;;) {
- bool repoll = false;
+ local_bh_disable();
+ sd = this_cpu_ptr(&softnet_data);
+ sd->in_napi_threaded_poll = true;
- local_bh_disable();
- sd = this_cpu_ptr(&softnet_data);
- sd->in_napi_threaded_poll = true;
-
- have = netpoll_poll_lock(napi);
- __napi_poll(napi, &repoll);
- netpoll_poll_unlock(have);
-
- sd->in_napi_threaded_poll = false;
- barrier();
-
- if (sd_has_rps_ipi_waiting(sd)) {
- local_irq_disable();
- net_rps_action_and_irq_enable(sd);
- }
- skb_defer_free_flush(sd);
- local_bh_enable();
+ have = netpoll_poll_lock(napi);
+ __napi_poll(napi, &repoll);
+ netpoll_poll_unlock(have);
+
+ sd->in_napi_threaded_poll = false;
+ barrier();
+
+ if (sd_has_rps_ipi_waiting(sd)) {
+ local_irq_disable();
+ net_rps_action_and_irq_enable(sd);
+ }
+ skb_defer_free_flush(sd);
+ local_bh_enable();
- if (!repoll)
- break;
+ if (!repoll)
+ break;
- rcu_softirq_qs_periodic(last_qs);
- cond_resched();
- }
+ rcu_softirq_qs_periodic(last_qs);
+ cond_resched();
}
+}
+
+static int napi_threaded_poll(void *data)
+{
+ struct napi_struct *napi = data;
+
+ while (!napi_thread_wait(napi))
+ napi_threaded_poll_loop(napi);
+
return 0;
}
@@ -11288,7 +11329,7 @@ static int dev_cpu_dead(unsigned int old
list_del_init(&napi->poll_list);
if (napi->poll == process_backlog)
- napi->state = 0;
+ napi->state &= NAPIF_STATE_THREADED;
else
____napi_schedule(sd, napi);
}
@@ -11296,12 +11337,14 @@ static int dev_cpu_dead(unsigned int old
raise_softirq_irqoff(NET_TX_SOFTIRQ);
local_irq_enable();
+ if (!use_backlog_threads()) {
#ifdef CONFIG_RPS
- remsd = oldsd->rps_ipi_list;
- oldsd->rps_ipi_list = NULL;
+ remsd = oldsd->rps_ipi_list;
+ oldsd->rps_ipi_list = NULL;
#endif
- /* send out pending IPI's on offline CPU */
- net_rps_send_ipi(remsd);
+ /* send out pending IPI's on offline CPU */
+ net_rps_send_ipi(remsd);
+ }
/* Process offline CPU's input_pkt_queue */
while ((skb = __skb_dequeue(&oldsd->process_queue))) {
@@ -11564,6 +11607,38 @@ static struct pernet_operations __net_in
*
*/
+static int backlog_napi_should_run(unsigned int cpu)
+{
+ struct softnet_data *sd = per_cpu_ptr(&softnet_data, cpu);
+ struct napi_struct *napi = &sd->backlog;
+
+ return test_bit(NAPI_STATE_SCHED_THREADED, &napi->state);
+}
+
+static void run_backlog_napi(unsigned int cpu)
+{
+ struct softnet_data *sd = per_cpu_ptr(&softnet_data, cpu);
+
+ napi_threaded_poll_loop(&sd->backlog);
+}
+
+static void backlog_napi_setup(unsigned int cpu)
+{
+ struct softnet_data *sd = per_cpu_ptr(&softnet_data, cpu);
+ struct napi_struct *napi = &sd->backlog;
+
+ napi->thread = this_cpu_read(backlog_napi);
+ set_bit(NAPI_STATE_THREADED, &napi->state);
+}
+
+static struct smp_hotplug_thread backlog_threads = {
+ .store = &backlog_napi,
+ .thread_should_run = backlog_napi_should_run,
+ .thread_fn = run_backlog_napi,
+ .thread_comm = "backlog_napi/%u",
+ .setup = backlog_napi_setup,
+};
+
/*
* This is called single threaded during boot, so no need
* to take the rtnl semaphore.
@@ -11614,7 +11689,10 @@ static int __init net_dev_init(void)
init_gro_hash(&sd->backlog);
sd->backlog.poll = process_backlog;
sd->backlog.weight = weight_p;
+ INIT_LIST_HEAD(&sd->backlog.poll_list);
}
+ if (use_backlog_threads())
+ smpboot_register_percpu_thread(&backlog_threads);
dev_boot_phase = 0;

121
target/linux/generic/backport-6.6/602-v6.10-net-Use-backlog-NAPI-to-clean-up-the-defer_list.patch Normal file
View File

@@ -0,0 +1,121 @@
From 80d2eefcb4c84aa9018b2a997ab3a4c567bc821a Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Mon, 25 Mar 2024 08:40:30 +0100
Subject: [PATCH 3/4] net: Use backlog-NAPI to clean up the defer_list.
The defer_list is a per-CPU list which is used to free skbs outside of
the socket lock and on the CPU on which they have been allocated.
The list is processed during NAPI callbacks so ideally the list is
cleaned up.
Should the amount of skbs on the list exceed a certain water mark then
the softirq is triggered remotely on the target CPU by invoking a remote
function call. The raise of the softirqs via a remote function call
leads to waking the ksoftirqd on PREEMPT_RT which is undesired.
The backlog-NAPI threads already provide the infrastructure which can be
utilized to perform the cleanup of the defer_list.
The NAPI state is updated with the input_pkt_queue.lock acquired. It
order not to break the state, it is needed to also wake the backlog-NAPI
thread with the lock held. This requires to acquire the use the lock in
rps_lock_irq*() if the backlog-NAPI threads are used even with RPS
disabled.
Move the logic of remotely starting softirqs to clean up the defer_list
into kick_defer_list_purge(). Make sure a lock is held in
rps_lock_irq*() if backlog-NAPI threads are used. Schedule backlog-NAPI
for defer_list cleanup if backlog-NAPI is available.
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
include/linux/netdevice.h | 1 +
net/core/dev.c | 25 +++++++++++++++++++++----
net/core/skbuff.c | 4 ++--
3 files changed, 24 insertions(+), 6 deletions(-)
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3300,6 +3300,7 @@ static inline void dev_xmit_recursion_de
__this_cpu_dec(softnet_data.xmit.recursion);
}
+void kick_defer_list_purge(struct softnet_data *sd, unsigned int cpu);
void __netif_schedule(struct Qdisc *q);
void netif_schedule_queue(struct netdev_queue *txq);
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -246,7 +246,7 @@ static bool use_backlog_threads(void)
static inline void rps_lock_irqsave(struct softnet_data *sd,
unsigned long *flags)
{
- if (IS_ENABLED(CONFIG_RPS))
+ if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads())
spin_lock_irqsave(&sd->input_pkt_queue.lock, *flags);
else if (!IS_ENABLED(CONFIG_PREEMPT_RT))
local_irq_save(*flags);
@@ -254,7 +254,7 @@ static inline void rps_lock_irqsave(stru
static inline void rps_lock_irq_disable(struct softnet_data *sd)
{
- if (IS_ENABLED(CONFIG_RPS))
+ if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads())
spin_lock_irq(&sd->input_pkt_queue.lock);
else if (!IS_ENABLED(CONFIG_PREEMPT_RT))
local_irq_disable();
@@ -263,7 +263,7 @@ static inline void rps_lock_irq_disable(
static inline void rps_unlock_irq_restore(struct softnet_data *sd,
unsigned long *flags)
{
- if (IS_ENABLED(CONFIG_RPS))
+ if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads())
spin_unlock_irqrestore(&sd->input_pkt_queue.lock, *flags);
else if (!IS_ENABLED(CONFIG_PREEMPT_RT))
local_irq_restore(*flags);
@@ -271,7 +271,7 @@ static inline void rps_unlock_irq_restor
static inline void rps_unlock_irq_enable(struct softnet_data *sd)
{
- if (IS_ENABLED(CONFIG_RPS))
+ if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads())
spin_unlock_irq(&sd->input_pkt_queue.lock);
else if (!IS_ENABLED(CONFIG_PREEMPT_RT))
local_irq_enable();
@@ -4774,6 +4774,23 @@ static void napi_schedule_rps(struct sof
__napi_schedule_irqoff(&mysd->backlog);
}
+void kick_defer_list_purge(struct softnet_data *sd, unsigned int cpu)
+{
+ unsigned long flags;
+
+ if (use_backlog_threads()) {
+ rps_lock_irqsave(sd, &flags);
+
+ if (!__test_and_set_bit(NAPI_STATE_SCHED, &sd->backlog.state))
+ __napi_schedule_irqoff(&sd->backlog);
+
+ rps_unlock_irq_restore(sd, &flags);
+
+ } else if (!cmpxchg(&sd->defer_ipi_scheduled, 0, 1)) {
+ smp_call_function_single_async(cpu, &sd->defer_csd);
+ }
+}
+
#ifdef CONFIG_NET_FLOW_LIMIT
int netdev_flow_limit_table_len __read_mostly = (1 << 12);
#endif
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -6863,8 +6863,8 @@ nodefer: __kfree_skb(skb);
/* Make sure to trigger NET_RX_SOFTIRQ on the remote CPU
* if we are unlucky enough (this seems very unlikely).
*/
- if (unlikely(kick) && !cmpxchg(&sd->defer_ipi_scheduled, 0, 1))
- smp_call_function_single_async(cpu, &sd->defer_csd);
+ if (unlikely(kick))
+ kick_defer_list_purge(sd, cpu);
}
static void skb_splice_csum_page(struct sk_buff *skb, struct page *page,

164
target/linux/generic/backport-6.6/603-v6.10-net-Rename-rps_lock-to-backlog_lock.patch Normal file
View File

@@ -0,0 +1,164 @@
From 765b11f8f4e20b7433e4ba4a3e9106a0d59501ed Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Mon, 25 Mar 2024 08:40:31 +0100
Subject: [PATCH 4/4] net: Rename rps_lock to backlog_lock.
The rps_lock.*() functions use the inner lock of a sk_buff_head for
locking. This lock is used if RPS is enabled, otherwise the list is
accessed lockless and disabling interrupts is enough for the
synchronisation because it is only accessed CPU local. Not only the list
is protected but also the NAPI state protected.
With the addition of backlog threads, the lock is also needed because of
the cross CPU access even without RPS. The clean up of the defer_list
list is also done via backlog threads (if enabled).
It has been suggested to rename the locking function since it is no
longer just RPS.
Rename the rps_lock*() functions to backlog_lock*().
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
net/core/dev.c | 34 +++++++++++++++++-----------------
1 file changed, 17 insertions(+), 17 deletions(-)
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -243,8 +243,8 @@ static bool use_backlog_threads(void)
#endif
-static inline void rps_lock_irqsave(struct softnet_data *sd,
- unsigned long *flags)
+static inline void backlog_lock_irq_save(struct softnet_data *sd,
+ unsigned long *flags)
{
if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads())
spin_lock_irqsave(&sd->input_pkt_queue.lock, *flags);
@@ -252,7 +252,7 @@ static inline void rps_lock_irqsave(stru
local_irq_save(*flags);
}
-static inline void rps_lock_irq_disable(struct softnet_data *sd)
+static inline void backlog_lock_irq_disable(struct softnet_data *sd)
{
if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads())
spin_lock_irq(&sd->input_pkt_queue.lock);
@@ -260,8 +260,8 @@ static inline void rps_lock_irq_disable(
local_irq_disable();
}
-static inline void rps_unlock_irq_restore(struct softnet_data *sd,
- unsigned long *flags)
+static inline void backlog_unlock_irq_restore(struct softnet_data *sd,
+ unsigned long *flags)
{
if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads())
spin_unlock_irqrestore(&sd->input_pkt_queue.lock, *flags);
@@ -269,7 +269,7 @@ static inline void rps_unlock_irq_restor
local_irq_restore(*flags);
}
-static inline void rps_unlock_irq_enable(struct softnet_data *sd)
+static inline void backlog_unlock_irq_enable(struct softnet_data *sd)
{
if (IS_ENABLED(CONFIG_RPS) || use_backlog_threads())
spin_unlock_irq(&sd->input_pkt_queue.lock);
@@ -4779,12 +4779,12 @@ void kick_defer_list_purge(struct softne
unsigned long flags;
if (use_backlog_threads()) {
- rps_lock_irqsave(sd, &flags);
+ backlog_lock_irq_save(sd, &flags);
if (!__test_and_set_bit(NAPI_STATE_SCHED, &sd->backlog.state))
__napi_schedule_irqoff(&sd->backlog);
- rps_unlock_irq_restore(sd, &flags);
+ backlog_unlock_irq_restore(sd, &flags);
} else if (!cmpxchg(&sd->defer_ipi_scheduled, 0, 1)) {
smp_call_function_single_async(cpu, &sd->defer_csd);
@@ -4846,7 +4846,7 @@ static int enqueue_to_backlog(struct sk_
reason = SKB_DROP_REASON_NOT_SPECIFIED;
sd = &per_cpu(softnet_data, cpu);
- rps_lock_irqsave(sd, &flags);
+ backlog_lock_irq_save(sd, &flags);
if (!netif_running(skb->dev))
goto drop;
qlen = skb_queue_len(&sd->input_pkt_queue);
@@ -4855,7 +4855,7 @@ static int enqueue_to_backlog(struct sk_
enqueue:
__skb_queue_tail(&sd->input_pkt_queue, skb);
input_queue_tail_incr_save(sd, qtail);
- rps_unlock_irq_restore(sd, &flags);
+ backlog_unlock_irq_restore(sd, &flags);
return NET_RX_SUCCESS;
}
@@ -4870,7 +4870,7 @@ enqueue:
drop:
sd->dropped++;
- rps_unlock_irq_restore(sd, &flags);
+ backlog_unlock_irq_restore(sd, &flags);
dev_core_stats_rx_dropped_inc(skb->dev);
kfree_skb_reason(skb, reason);
@@ -5901,7 +5901,7 @@ static void flush_backlog(struct work_st
local_bh_disable();
sd = this_cpu_ptr(&softnet_data);
- rps_lock_irq_disable(sd);
+ backlog_lock_irq_disable(sd);
skb_queue_walk_safe(&sd->input_pkt_queue, skb, tmp) {
if (skb->dev->reg_state == NETREG_UNREGISTERING) {
__skb_unlink(skb, &sd->input_pkt_queue);
@@ -5909,7 +5909,7 @@ static void flush_backlog(struct work_st
input_queue_head_incr(sd);
}
}
- rps_unlock_irq_enable(sd);
+ backlog_unlock_irq_enable(sd);
skb_queue_walk_safe(&sd->process_queue, skb, tmp) {
if (skb->dev->reg_state == NETREG_UNREGISTERING) {
@@ -5927,14 +5927,14 @@ static bool flush_required(int cpu)
struct softnet_data *sd = &per_cpu(softnet_data, cpu);
bool do_flush;
- rps_lock_irq_disable(sd);
+ backlog_lock_irq_disable(sd);
/* as insertion into process_queue happens with the rps lock held,
* process_queue access may race only with dequeue
*/
do_flush = !skb_queue_empty(&sd->input_pkt_queue) ||
!skb_queue_empty_lockless(&sd->process_queue);
- rps_unlock_irq_enable(sd);
+ backlog_unlock_irq_enable(sd);
return do_flush;
#endif
@@ -6049,7 +6049,7 @@ static int process_backlog(struct napi_s
}
- rps_lock_irq_disable(sd);
+ backlog_lock_irq_disable(sd);
if (skb_queue_empty(&sd->input_pkt_queue)) {
/*
* Inline a custom version of __napi_complete().
@@ -6065,7 +6065,7 @@ static int process_backlog(struct napi_s
skb_queue_splice_tail_init(&sd->input_pkt_queue,
&sd->process_queue);
}
- rps_unlock_irq_enable(sd);
+ backlog_unlock_irq_enable(sd);
}
return work;

2
target/linux/generic/backport-6.6/770-net-introduce-napi_is_scheduled-helper.patch
View File

@@ -85,7 +85,7 @@ Signed-off-by: Paolo Abeni <pabeni@redhat.com>
/** /**
--- a/net/core/dev.c --- a/net/core/dev.c
+++ b/net/core/dev.c +++ b/net/core/dev.c
@@ -6555,7 +6555,7 @@ static int __napi_poll(struct napi_struc @@ -6602,7 +6602,7 @@ static int __napi_poll(struct napi_struc
* accidentally calling ->poll() when NAPI is not scheduled. * accidentally calling ->poll() when NAPI is not scheduled.
*/ */
work = 0; work = 0;

10
target/linux/generic/hack-6.6/721-net-add-packet-mangeling.patch
View File

@@ -19,7 +19,7 @@ Signed-off-by: Felix Fietkau <nbd@nbd.name>
--- a/include/linux/netdevice.h --- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h +++ b/include/linux/netdevice.h
@@ -1759,6 +1759,7 @@ enum netdev_priv_flags { @@ -1758,6 +1758,7 @@ enum netdev_priv_flags {
IFF_TX_SKB_NO_LINEAR = BIT_ULL(31), IFF_TX_SKB_NO_LINEAR = BIT_ULL(31),
IFF_CHANGE_PROTO_DOWN = BIT_ULL(32), IFF_CHANGE_PROTO_DOWN = BIT_ULL(32),
IFF_SEE_ALL_HWTSTAMP_REQUESTS = BIT_ULL(33), IFF_SEE_ALL_HWTSTAMP_REQUESTS = BIT_ULL(33),
@@ -27,7 +27,7 @@ Signed-off-by: Felix Fietkau <nbd@nbd.name>
}; };
#define IFF_802_1Q_VLAN IFF_802_1Q_VLAN #define IFF_802_1Q_VLAN IFF_802_1Q_VLAN
@@ -1792,6 +1793,7 @@ enum netdev_priv_flags { @@ -1791,6 +1792,7 @@ enum netdev_priv_flags {
#define IFF_FAILOVER_SLAVE IFF_FAILOVER_SLAVE #define IFF_FAILOVER_SLAVE IFF_FAILOVER_SLAVE
#define IFF_L3MDEV_RX_HANDLER IFF_L3MDEV_RX_HANDLER #define IFF_L3MDEV_RX_HANDLER IFF_L3MDEV_RX_HANDLER
#define IFF_TX_SKB_NO_LINEAR IFF_TX_SKB_NO_LINEAR #define IFF_TX_SKB_NO_LINEAR IFF_TX_SKB_NO_LINEAR
@@ -35,7 +35,7 @@ Signed-off-by: Felix Fietkau <nbd@nbd.name>
/* Specifies the type of the struct net_device::ml_priv pointer */ /* Specifies the type of the struct net_device::ml_priv pointer */
enum netdev_ml_priv_type { enum netdev_ml_priv_type {
@@ -2184,6 +2186,11 @@ struct net_device { @@ -2183,6 +2185,11 @@ struct net_device {
const struct tlsdev_ops *tlsdev_ops; const struct tlsdev_ops *tlsdev_ops;
#endif #endif
@@ -47,7 +47,7 @@ Signed-off-by: Felix Fietkau <nbd@nbd.name>
const struct header_ops *header_ops; const struct header_ops *header_ops;
unsigned char operstate; unsigned char operstate;
@@ -2257,6 +2264,10 @@ struct net_device { @@ -2256,6 +2263,10 @@ struct net_device {
struct mctp_dev __rcu *mctp_ptr; struct mctp_dev __rcu *mctp_ptr;
#endif #endif
@@ -105,7 +105,7 @@ Signed-off-by: Felix Fietkau <nbd@nbd.name>
help help
--- a/net/core/dev.c --- a/net/core/dev.c
+++ b/net/core/dev.c +++ b/net/core/dev.c
@@ -3571,6 +3571,11 @@ static int xmit_one(struct sk_buff *skb, @@ -3597,6 +3597,11 @@ static int xmit_one(struct sk_buff *skb,
if (dev_nit_active(dev)) if (dev_nit_active(dev))
dev_queue_xmit_nit(skb, dev); dev_queue_xmit_nit(skb, dev);

227
target/linux/generic/pending-6.6/760-net-core-add-optional-threading-for-backlog-processi.patch
View File

@@ -1,227 +0,0 @@
From: Felix Fietkau <nbd@nbd.name>
Date: Thu, 16 Feb 2023 18:39:04 +0100
Subject: [PATCH] net/core: add optional threading for backlog processing
When dealing with few flows or an imbalance on CPU utilization, static RPS
CPU assignment can be too inflexible. Add support for enabling threaded NAPI
for backlog processing in order to allow the scheduler to better balance
processing. This helps better spread the load across idle CPUs.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -558,6 +558,7 @@ static inline bool napi_complete(struct
}
int dev_set_threaded(struct net_device *dev, bool threaded);
+int backlog_set_threaded(bool threaded);
/**
* napi_disable - prevent NAPI from scheduling
@@ -3236,6 +3237,7 @@ struct softnet_data {
/* stats */
unsigned int processed;
unsigned int time_squeeze;
+ unsigned int process_queue_empty;
#ifdef CONFIG_RPS
struct softnet_data *rps_ipi_list;
#endif
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4729,7 +4729,7 @@ static void napi_schedule_rps(struct sof
struct softnet_data *mysd = this_cpu_ptr(&softnet_data);
#ifdef CONFIG_RPS
- if (sd != mysd) {
+ if (sd != mysd && !test_bit(NAPI_STATE_THREADED, &sd->backlog.state)) {
sd->rps_ipi_next = mysd->rps_ipi_list;
mysd->rps_ipi_list = sd;
@@ -5848,6 +5848,8 @@ static DEFINE_PER_CPU(struct work_struct
/* Network device is going away, flush any packets still pending */
static void flush_backlog(struct work_struct *work)
{
+ unsigned int process_queue_empty;
+ bool threaded, flush_processq;
struct sk_buff *skb, *tmp;
struct softnet_data *sd;
@@ -5862,8 +5864,17 @@ static void flush_backlog(struct work_st
input_queue_head_incr(sd);
}
}
+
+ threaded = test_bit(NAPI_STATE_THREADED, &sd->backlog.state);
+ flush_processq = threaded &&
+ !skb_queue_empty_lockless(&sd->process_queue);
+ if (flush_processq)
+ process_queue_empty = sd->process_queue_empty;
rps_unlock_irq_enable(sd);
+ if (threaded)
+ goto out;
+
skb_queue_walk_safe(&sd->process_queue, skb, tmp) {
if (skb->dev->reg_state == NETREG_UNREGISTERING) {
__skb_unlink(skb, &sd->process_queue);
@@ -5871,7 +5882,16 @@ static void flush_backlog(struct work_st
input_queue_head_incr(sd);
}
}
+
+out:
local_bh_enable();
+
+ while (flush_processq) {
+ msleep(1);
+ rps_lock_irq_disable(sd);
+ flush_processq = process_queue_empty == sd->process_queue_empty;
+ rps_unlock_irq_enable(sd);
+ }
}
static bool flush_required(int cpu)
@@ -6003,6 +6023,7 @@ static int process_backlog(struct napi_s
}
rps_lock_irq_disable(sd);
+ sd->process_queue_empty++;
if (skb_queue_empty(&sd->input_pkt_queue)) {
/*
* Inline a custom version of __napi_complete().
@@ -6012,7 +6033,8 @@ static int process_backlog(struct napi_s
* We can use a plain write instead of clear_bit(),
* and we dont need an smp_mb() memory barrier.
*/
- napi->state = 0;
+ napi->state &= ~(NAPIF_STATE_SCHED |
+ NAPIF_STATE_SCHED_THREADED);
again = false;
} else {
skb_queue_splice_tail_init(&sd->input_pkt_queue,
@@ -6426,6 +6448,55 @@ int dev_set_threaded(struct net_device *
}
EXPORT_SYMBOL(dev_set_threaded);
+int backlog_set_threaded(bool threaded)
+{
+ static bool backlog_threaded;
+ int err = 0;
+ int i;
+
+ if (backlog_threaded == threaded)
+ return 0;
+
+ for_each_possible_cpu(i) {
+ struct softnet_data *sd = &per_cpu(softnet_data, i);
+ struct napi_struct *n = &sd->backlog;
+
+ if (n->thread)
+ continue;
+ n->thread = kthread_run(napi_threaded_poll, n, "napi/backlog-%d", i);
+ if (IS_ERR(n->thread)) {
+ err = PTR_ERR(n->thread);
+ pr_err("kthread_run failed with err %d\n", err);
+ n->thread = NULL;
+ threaded = false;
+ break;
+ }
+
+ }
+
+ backlog_threaded = threaded;
+
+ /* Make sure kthread is created before THREADED bit
+ * is set.
+ */
+ smp_mb__before_atomic();
+
+ for_each_possible_cpu(i) {
+ struct softnet_data *sd = &per_cpu(softnet_data, i);
+ struct napi_struct *n = &sd->backlog;
+ unsigned long flags;
+
+ rps_lock_irqsave(sd, &flags);
+ if (threaded)
+ n->state |= NAPIF_STATE_THREADED;
+ else
+ n->state &= ~NAPIF_STATE_THREADED;
+ rps_unlock_irq_restore(sd, &flags);
+ }
+
+ return err;
+}
+
void netif_napi_add_weight(struct net_device *dev, struct napi_struct *napi,
int (*poll)(struct napi_struct *, int), int weight)
{
@@ -11307,6 +11378,9 @@ static int dev_cpu_dead(unsigned int old
raise_softirq_irqoff(NET_TX_SOFTIRQ);
local_irq_enable();
+ if (test_bit(NAPI_STATE_THREADED, &oldsd->backlog.state))
+ return 0;
+
#ifdef CONFIG_RPS
remsd = oldsd->rps_ipi_list;
oldsd->rps_ipi_list = NULL;
@@ -11622,6 +11696,7 @@ static int __init net_dev_init(void)
INIT_CSD(&sd->defer_csd, trigger_rx_softirq, sd);
spin_lock_init(&sd->defer_lock);
+ INIT_LIST_HEAD(&sd->backlog.poll_list);
init_gro_hash(&sd->backlog);
sd->backlog.poll = process_backlog;
sd->backlog.weight = weight_p;
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -30,6 +30,7 @@ static int int_3600 = 3600;
static int min_sndbuf = SOCK_MIN_SNDBUF;
static int min_rcvbuf = SOCK_MIN_RCVBUF;
static int max_skb_frags = MAX_SKB_FRAGS;
+static int backlog_threaded;
static int min_mem_pcpu_rsv = SK_MEMORY_PCPU_RESERVE;
static int net_msg_warn; /* Unused, but still a sysctl */
@@ -189,6 +190,23 @@ static int rps_sock_flow_sysctl(struct c
}
#endif /* CONFIG_RPS */
+static int backlog_threaded_sysctl(struct ctl_table *table, int write,
+ void *buffer, size_t *lenp, loff_t *ppos)
+{
+ static DEFINE_MUTEX(backlog_threaded_mutex);
+ int ret;
+
+ mutex_lock(&backlog_threaded_mutex);
+
+ ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
+ if (write && !ret)
+ ret = backlog_set_threaded(backlog_threaded);
+
+ mutex_unlock(&backlog_threaded_mutex);
+
+ return ret;
+}
+
#ifdef CONFIG_NET_FLOW_LIMIT
static DEFINE_MUTEX(flow_limit_update_mutex);
@@ -541,6 +559,15 @@ static struct ctl_table net_core_table[]
.proc_handler = rps_sock_flow_sysctl
},
#endif
+ {
+ .procname = "backlog_threaded",
+ .data = &backlog_threaded,
+ .maxlen = sizeof(unsigned int),
+ .mode = 0644,
+ .proc_handler = backlog_threaded_sysctl,
+ .extra1 = SYSCTL_ZERO,
+ .extra2 = SYSCTL_ONE
+ },
#ifdef CONFIG_NET_FLOW_LIMIT
{
.procname = "flow_limit_cpu_bitmap",