linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: LKML <linux-kernel@vger.kernel.org>
Cc: Frederic Weisbecker <frederic@kernel.org>,
	Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
	"Ahmed S. Darwish" <a.darwish@linutronix.de>,
	Peter Zijlstra <peterz@infradead.org>,
	Denis Kirjanov <kda@linux-powerpc.org>,
	"David S. Miller" <davem@davemloft.net>,
	Jakub Kicinski <kuba@kernel.org>,
	netdev@vger.kernel.org, ath9k-devel@qca.qualcomm.com,
	Kalle Valo <kvalo@codeaurora.org>,
	linux-wireless@vger.kernel.org, Chas Williams <3chas3@gmail.com>,
	linux-atm-general@lists.sourceforge.net,
	"K. Y. Srinivasan" <kys@microsoft.com>,
	Haiyang Zhang <haiyangz@microsoft.com>,
	Stephen Hemminger <sthemmin@microsoft.com>,
	Wei Liu <wei.liu@kernel.org>,
	Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>,
	Rob Herring <robh@kernel.org>,
	Bjorn Helgaas <bhelgaas@google.com>,
	linux-hyperv@vger.kernel.org, linux-pci@vger.kernel.org,
	Stefan Richter <stefanr@s5r6.in-berlin.de>,
	linux1394-devel@lists.sourceforge.net
Subject: [patch 05/14] tasklets: Replace spin wait in tasklet_unlock_wait()
Date: Tue, 09 Mar 2021 09:42:08 +0100	[thread overview]
Message-ID: <20210309084241.783936921@linutronix.de> (raw)
In-Reply-To: 20210309084203.995862150@linutronix.de

From: Peter Zijlstra <peterz@infradead.org>

tasklet_unlock_wait() spin waits for TASKLET_STATE_RUN to be cleared. This
is wasting CPU cycles in a tight loop which is especially painful in a
guest when the CPU running the tasklet is scheduled out.

tasklet_unlock_wait() is invoked from tasklet_kill() which is used in
teardown paths and not performance critical at all. Replace the spin wait
with wait_var_event().

There are no users of tasklet_unlock_wait() which are invoked from atomic
contexts. The usage in tasklet_disable() has been replaced temporarily with
the spin waiting variant until the atomic users are fixed up and will be
converted to the sleep wait variant later.

Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 include/linux/interrupt.h |   13 ++-----------
 kernel/softirq.c          |   18 ++++++++++++++++++
 2 files changed, 20 insertions(+), 11 deletions(-)

--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -664,17 +664,8 @@ static inline int tasklet_trylock(struct
 	return !test_and_set_bit(TASKLET_STATE_RUN, &(t)->state);
 }
 
-static inline void tasklet_unlock(struct tasklet_struct *t)
-{
-	smp_mb__before_atomic();
-	clear_bit(TASKLET_STATE_RUN, &(t)->state);
-}
-
-static inline void tasklet_unlock_wait(struct tasklet_struct *t)
-{
-	while (test_bit(TASKLET_STATE_RUN, &t->state))
-		cpu_relax();
-}
+void tasklet_unlock(struct tasklet_struct *t);
+void tasklet_unlock_wait(struct tasklet_struct *t);
 
 /*
  * Do not use in new code. Waiting for tasklets from atomic contexts is
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -25,6 +25,7 @@
 #include <linux/smpboot.h>
 #include <linux/tick.h>
 #include <linux/irq.h>
+#include <linux/wait_bit.h>
 
 #include <asm/softirq_stack.h>
 
@@ -621,6 +622,23 @@ void tasklet_kill(struct tasklet_struct
 }
 EXPORT_SYMBOL(tasklet_kill);
 
+#ifdef CONFIG_SMP
+void tasklet_unlock(struct tasklet_struct *t)
+{
+	smp_mb__before_atomic();
+	clear_bit(TASKLET_STATE_RUN, &t->state);
+	smp_mb__after_atomic();
+	wake_up_var(&t->state);
+}
+EXPORT_SYMBOL_GPL(tasklet_unlock);
+
+void tasklet_unlock_wait(struct tasklet_struct *t)
+{
+	wait_var_event(&t->state, !test_bit(TASKLET_STATE_RUN, &t->state));
+}
+EXPORT_SYMBOL_GPL(tasklet_unlock_wait);
+#endif
+
 void __init softirq_init(void)
 {
 	int cpu;


  parent reply	other threads:[~2021-03-09  8:46 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-09  8:42 [patch 00/14] tasklets: Replace the spin wait loops and make it RT safe Thomas Gleixner
2021-03-09  8:42 ` [patch 01/14] tasklets: Replace barrier() with cpu_relax() in tasklet_unlock_wait() Thomas Gleixner
2021-03-09  8:42 ` [patch 02/14] tasklets: Use static inlines for stub implementations Thomas Gleixner
2021-03-09  8:42 ` [patch 03/14] tasklets: Provide tasklet_disable_in_atomic() Thomas Gleixner
2021-03-09  8:42 ` [patch 04/14] tasklets: Use spin wait in tasklet_disable() temporarily Thomas Gleixner
2021-03-09  8:42 ` Thomas Gleixner [this message]
2021-03-09  8:42 ` [patch 06/14] tasklets: Replace spin wait in tasklet_kill() Thomas Gleixner
2021-03-09  8:42 ` [patch 07/14] tasklets: Prevent tasklet_unlock_spin_wait() deadlock on RT Thomas Gleixner
2021-03-09 15:00   ` Sebastian Andrzej Siewior
2021-03-09 15:21     ` Sebastian Andrzej Siewior
2021-03-10  8:35       ` Thomas Gleixner
2021-03-09  8:42 ` [patch 08/14] net: jme: Replace link-change tasklet with work Thomas Gleixner
2021-03-09  8:42 ` [patch 09/14] net: sundance: Use tasklet_disable_in_atomic() Thomas Gleixner
2021-03-09  8:42 ` [patch 10/14] ath9k: " Thomas Gleixner
2021-03-09 10:23   ` Kalle Valo
2021-03-09  8:42 ` [patch 11/14] atm: eni: Use tasklet_disable_in_atomic() in the send() callback Thomas Gleixner
2021-03-09  8:42 ` [patch 12/14] PCI: hv: Use tasklet_disable_in_atomic() Thomas Gleixner
2021-03-09 22:17   ` Bjorn Helgaas
2021-03-10 11:34   ` Wei Liu
2021-03-09  8:42 ` [patch 13/14] firewire: ohci: Use tasklet_disable_in_atomic() where required Thomas Gleixner
2021-03-09  8:42 ` [patch 14/14] tasklets: Switch tasklet_disable() to the sleep wait variant Thomas Gleixner
2021-03-09 14:04 ` [patch 00/14] tasklets: Replace the spin wait loops and make it RT safe Peter Zijlstra

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=20210309084241.783936921@linutronix.de \
    --to=tglx@linutronix.de \
    --cc=3chas3@gmail.com \
    --cc=a.darwish@linutronix.de \
    --cc=ath9k-devel@qca.qualcomm.com \
    --cc=bhelgaas@google.com \
    --cc=bigeasy@linutronix.de \
    --cc=davem@davemloft.net \
    --cc=frederic@kernel.org \
    --cc=haiyangz@microsoft.com \
    --cc=kda@linux-powerpc.org \
    --cc=kuba@kernel.org \
    --cc=kvalo@codeaurora.org \
    --cc=kys@microsoft.com \
    --cc=linux-atm-general@lists.sourceforge.net \
    --cc=linux-hyperv@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=linux-wireless@vger.kernel.org \
    --cc=linux1394-devel@lists.sourceforge.net \
    --cc=lorenzo.pieralisi@arm.com \
    --cc=netdev@vger.kernel.org \
    --cc=peterz@infradead.org \
    --cc=robh@kernel.org \
    --cc=stefanr@s5r6.in-berlin.de \
    --cc=sthemmin@microsoft.com \
    --cc=wei.liu@kernel.org \
    /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).