[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20240812125717.413108-3-jdamato@fastly.com>
Date: Mon, 12 Aug 2024 12:57:05 +0000
From: Joe Damato <jdamato@...tly.com>
To: netdev@...r.kernel.org
Cc: mkarsten@...terloo.ca,
amritha.nambiar@...el.com,
sridhar.samudrala@...el.com,
sdf@...ichev.me,
Joe Damato <jdamato@...tly.com>,
"David S. Miller" <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>,
Jakub Kicinski <kuba@...nel.org>,
Paolo Abeni <pabeni@...hat.com>,
Jiri Pirko <jiri@...nulli.us>,
Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
Lorenzo Bianconi <lorenzo@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>,
linux-kernel@...r.kernel.org (open list)
Subject: [RFC net-next 2/5] net: Suspend softirq when prefer_busy_poll is set
From: Martin Karsten <mkarsten@...terloo.ca>
When NAPI_F_PREFER_BUSY_POLL is set during busy_poll_stop and the
irq_suspend_timeout sysfs is nonzero, this timeout is used to defer
softirq scheduling, potentially longer than gro_flush_timeout. This can
be used to effectively suspend softirq processing during the time it
takes for an application to process data and return to the next
busy-poll.
The call to napi->poll in busy_poll_stop might lead to an invocation of
napi_complete_done, but the prefer-busy flag is still set at that time,
so the same logic is used to defer softirq scheduling for
irq_suspend_timeout.
Signed-off-by: Martin Karsten <mkarsten@...terloo.ca>
Co-developed-by: Joe Damato <jdamato@...tly.com>
Signed-off-by: Joe Damato <jdamato@...tly.com>
Tested-by: Joe Damato <jdamato@...tly.com>
Tested-by: Martin Karsten <mkarsten@...terloo.ca>
---
net/core/dev.c | 19 +++++++++++++------
1 file changed, 13 insertions(+), 6 deletions(-)
diff --git a/net/core/dev.c b/net/core/dev.c
index bf45c90aebcd..a19470253eb4 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -6214,7 +6214,6 @@ bool napi_complete_done(struct napi_struct *n, int work_done)
{
unsigned long flags, val, new, timeout = 0;
bool ret = true;
-
/*
* 1) Don't let napi dequeue from the cpu poll list
* just in case its running on a different cpu.
@@ -6230,7 +6229,12 @@ bool napi_complete_done(struct napi_struct *n, int work_done)
timeout = READ_ONCE(n->dev->gro_flush_timeout);
n->defer_hard_irqs_count = READ_ONCE(n->dev->napi_defer_hard_irqs);
}
- if (n->defer_hard_irqs_count > 0) {
+ if (napi_prefer_busy_poll(n)) {
+ timeout = READ_ONCE(n->dev->irq_suspend_timeout);
+ if (timeout)
+ ret = false;
+ }
+ if (ret && n->defer_hard_irqs_count > 0) {
n->defer_hard_irqs_count--;
timeout = READ_ONCE(n->dev->gro_flush_timeout);
if (timeout)
@@ -6349,7 +6353,6 @@ static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock,
bool skip_schedule = false;
unsigned long timeout;
int rc;
-
/* Busy polling means there is a high chance device driver hard irq
* could not grab NAPI_STATE_SCHED, and that NAPI_STATE_MISSED was
* set in napi_schedule_prep().
@@ -6366,9 +6369,13 @@ static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock,
bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
if (flags & NAPI_F_PREFER_BUSY_POLL) {
- napi->defer_hard_irqs_count = READ_ONCE(napi->dev->napi_defer_hard_irqs);
- timeout = READ_ONCE(napi->dev->gro_flush_timeout);
- if (napi->defer_hard_irqs_count && timeout) {
+ timeout = READ_ONCE(napi->dev->irq_suspend_timeout);
+ if (!timeout) {
+ napi->defer_hard_irqs_count = READ_ONCE(napi->dev->napi_defer_hard_irqs);
+ if (napi->defer_hard_irqs_count)
+ timeout = READ_ONCE(napi->dev->gro_flush_timeout);
+ }
+ if (timeout) {
hrtimer_start(&napi->timer, ns_to_ktime(timeout), HRTIMER_MODE_REL_PINNED);
skip_schedule = true;
}
--
2.25.1
Powered by blists - more mailing lists