linux-rt-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Steven Rostedt <rostedt@goodmis.org>
To: linux-kernel@vger.kernel.org,
	linux-rt-users <linux-rt-users@vger.kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>,
	Carsten Emde <C.Emde@osadl.org>,
	Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
	John Kacur <jkacur@redhat.com>,
	Paul Gortmaker <paul.gortmaker@windriver.com>,
	<stable-rt@vger.kernel.org>
Subject: [PATCH RT 03/22] net: move xmit_recursion to per-task variable on -RT
Date: Wed, 02 Mar 2016 10:09:02 -0500	[thread overview]
Message-ID: <20160302151110.746804820@goodmis.org> (raw)
In-Reply-To: 20160302150859.204542604@goodmis.org

[-- Attachment #1: 0003-net-move-xmit_recursion-to-per-task-variable-on-RT.patch --]
[-- Type: text/plain, Size: 3700 bytes --]

4.1.15-rt18-rc1 stable review patch.
If anyone has any objections, please let me know.

------------------

From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>

A softirq on -RT can be preempted. That means one task is in
__dev_queue_xmit(), gets preempted and another task may enter
__dev_queue_xmit() aw well. netperf together with a bridge device
will then trigger the `recursion alert` because each task increments
the xmit_recursion variable which is per-CPU.
A virtual device like br0 is required to trigger this warning.

This patch moves the counter to per task instead per-CPU so it counts
the recursion properly on -RT.

Cc: stable-rt@vger.kernel.org
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
 include/linux/netdevice.h |  9 +++++++++
 include/linux/sched.h     |  3 +++
 net/core/dev.c            | 41 ++++++++++++++++++++++++++++++++++++++---
 3 files changed, 50 insertions(+), 3 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 7a289e802a23..d24fe5d9980d 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2192,11 +2192,20 @@ void netdev_freemem(struct net_device *dev);
 void synchronize_net(void);
 int init_dummy_netdev(struct net_device *dev);
 
+#ifdef CONFIG_PREEMPT_RT_FULL
+static inline int dev_recursion_level(void)
+{
+	return current->xmit_recursion;
+}
+
+#else
+
 DECLARE_PER_CPU(int, xmit_recursion);
 static inline int dev_recursion_level(void)
 {
 	return this_cpu_read(xmit_recursion);
 }
+#endif
 
 struct net_device *dev_get_by_index(struct net *net, int ifindex);
 struct net_device *__dev_get_by_index(struct net *net, int ifindex);
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 1a56c0512491..4d995add9497 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1806,6 +1806,9 @@ struct task_struct {
 #ifdef CONFIG_DEBUG_ATOMIC_SLEEP
 	unsigned long	task_state_change;
 #endif
+#ifdef CONFIG_PREEMPT_RT_FULL
+	int xmit_recursion;
+#endif
 	int pagefault_disabled;
 };
 
diff --git a/net/core/dev.c b/net/core/dev.c
index 16fbef81024d..7d8ad99de55f 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2883,9 +2883,44 @@ static void skb_update_prio(struct sk_buff *skb)
 #define skb_update_prio(skb)
 #endif
 
+#ifdef CONFIG_PREEMPT_RT_FULL
+
+static inline int xmit_rec_read(void)
+{
+       return current->xmit_recursion;
+}
+
+static inline void xmit_rec_inc(void)
+{
+       current->xmit_recursion++;
+}
+
+static inline void xmit_rec_dec(void)
+{
+       current->xmit_recursion--;
+}
+
+#else
+
 DEFINE_PER_CPU(int, xmit_recursion);
 EXPORT_SYMBOL(xmit_recursion);
 
+static inline int xmit_rec_read(void)
+{
+	return __this_cpu_read(xmit_recursion);
+}
+
+static inline void xmit_rec_inc(void)
+{
+	__this_cpu_inc(xmit_recursion);
+}
+
+static inline int xmit_rec_dec(void)
+{
+	__this_cpu_dec(xmit_recursion);
+}
+#endif
+
 #define RECURSION_LIMIT 10
 
 /**
@@ -2987,7 +3022,7 @@ static int __dev_queue_xmit(struct sk_buff *skb, void *accel_priv)
 
 		if (txq->xmit_lock_owner != cpu) {
 
-			if (__this_cpu_read(xmit_recursion) > RECURSION_LIMIT)
+			if (xmit_rec_read() > RECURSION_LIMIT)
 				goto recursion_alert;
 
 			skb = validate_xmit_skb(skb, dev);
@@ -2997,9 +3032,9 @@ static int __dev_queue_xmit(struct sk_buff *skb, void *accel_priv)
 			HARD_TX_LOCK(dev, txq, cpu);
 
 			if (!netif_xmit_stopped(txq)) {
-				__this_cpu_inc(xmit_recursion);
+				xmit_rec_inc();
 				skb = dev_hard_start_xmit(skb, dev, txq, &rc);
-				__this_cpu_dec(xmit_recursion);
+				xmit_rec_dec();
 				if (dev_xmit_complete(rc)) {
 					HARD_TX_UNLOCK(dev, txq);
 					goto out;
-- 
2.7.0

  parent reply	other threads:[~2016-03-02 15:09 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-03-02 15:08 [PATCH RT 00/22] Linux 4.1.15-rt18-rc1 Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 01/22] sched: reset tasks lockless wake-queues on fork() Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 02/22] ptrace: dont open IRQs in ptrace_freeze_traced() too early Steven Rostedt
2016-03-02 15:09 ` Steven Rostedt [this message]
2016-03-02 15:09 ` [PATCH RT 04/22] kernel/softirq: use cond_resched_rcu_qs() on -RT as well (run_ksoftirqd()) Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 05/22] net/core: protect users of napi_alloc_cache against reentrance Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 06/22] preempt-lazy: Add the lazy-preemption check to preempt_schedule() Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 07/22] softirq: split timer softirqs out of ksoftirqd Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 08/22] net: provide a way to delegate processing a softirq to ksoftirqd Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 09/22] latencyhist: disable jump-labels Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 10/22] arm64: replace read_lock to rcu lock in call_step_hook Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 11/22] kernel: migrate_disable() do fastpath in atomic & irqs-off Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 12/22] kernel: softirq: unlock with irqs on Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 13/22] kernel/stop_machine: partly revert "stop_machine: Use raw spinlocks" Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 14/22] tick/broadcast: Make broadcast hrtimer irqsafe Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 15/22] sched,rt: __always_inline preemptible_lazy() Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 16/22] drm,radeon,i915: Use preempt_disable/enable_rt() where recommended Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 17/22] drm,i915: Use local_lock/unlock_irq() in intel_pipe_update_start/end() Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 18/22] trace: Use rcuidle version for preemptoff_hist trace point Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 19/22] f2fs: Mutex cant be used by down_write_nest_lock() Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 20/22] rcu/torture: Comment out rcu_bh ops on PREEMPT_RT_FULL Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 21/22] kernel: sched: Fix preempt_disable_ip recodring for preempt_disable() Steven Rostedt
2016-03-02 15:09 ` [PATCH RT 22/22] Linux 4.1.15-rt18-rc1 Steven Rostedt

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20160302151110.746804820@goodmis.org \
    --to=rostedt@goodmis.org \
    --cc=C.Emde@osadl.org \
    --cc=bigeasy@linutronix.de \
    --cc=jkacur@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rt-users@vger.kernel.org \
    --cc=paul.gortmaker@windriver.com \
    --cc=stable-rt@vger.kernel.org \
    --cc=tglx@linutronix.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).