[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20090114202044.GJ6734@linux.vnet.ibm.com>
Date: Wed, 14 Jan 2009 12:20:45 -0800
From: "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
To: Ingo Molnar <mingo@...e.hu>
Cc: linux-kernel@...r.kernel.org, akpm@...ux-foundation.org,
niv@...ibm.com, dvhltc@...ibm.com
Subject: Re: [PATCH fyi] RCU: the bloatwatch edition
On Wed, Jan 14, 2009 at 08:10:35PM +0100, Ingo Molnar wrote:
>
> * Ingo Molnar <mingo@...e.hu> wrote:
>
> >
> > * Paul E. McKenney <paulmck@...ux.vnet.ibm.com> wrote:
> >
> > > Just FYI, not for inclusion.
> > >
> > > This patch is a version of RCU designed for !CONFIG_SMP provided as a
> > > proof of concept of a small-footprint RCU implementation. In
> > > particular, the implementation of synchronize_rcu() is extremely
> > > lightweight and high performance. It passes moderate rcutorture testing
> > > in each of the four relevant configurations (combinations of NO_HZ and
> > > PREEMPT) on x86.
> > >
> > > Signed-off-by: Paul E. McKenney <paulmck@...ux.vnet.ibm.com>
> >
> > maybe it could be included - dependent on PREEMPT, just like we have SLOB?
>
> s/PREEMPT/EMBEDDED
Ah! That works better, as I suspect that people do test with EMBEDDED,
which addresses my bitrot objection. I don't immediately see how to
make the default for the "RCU implementation" choice depend on what other
settings there are, but I would guess that the embedded guys would grep
the Kconfig files for settings that depend on EMBEDDED, which would lead
them to TINY_RCU.
Anyway, here is the patch updated as you suggest. Binary sizes are
as follows, all collected with !SMP and EMBEDDED, which interestingly
enough shrinks rcutree significantly but has little effect on rcuclassic
and rcutiny:
CONFIG_CLASSIC_RCU:
text data bss dec hex filename
363 12 24 399 18f kernel/rcupdate.o
1237 64 124 1425 591 kernel/rcuclassic.o
1824 Total
CONFIG_TREE_RCU:
text data bss dec hex filename
363 12 24 399 18f kernel/rcupdate.o
2344 240 184 2768 ad0 kernel/rcutree.o
3167 Total
CONFIG_TINY_RCU:
text data bss dec hex filename
294 12 24 330 14a kernel/rcupdate.o
563 36 0 599 257 kernel/rcutiny.o
929 Total
So rcutiny saves about 900 bytes compared to rcuclassic.
Signed-off-by: Paul E. McKenney <paulmck@...ux.vnet.ibm.com>
---
include/linux/rcupdate.h | 2
include/linux/rcutiny.h | 68 +++++++++++
init/Kconfig | 7 +
kernel/Makefile | 1
kernel/rcupdate.c | 4
kernel/rcutiny.c | 290 +++++++++++++++++++++++++++++++++++++++++++++++
6 files changed, 372 insertions(+)
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 921340a..06e3909 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -58,6 +58,8 @@ struct rcu_head {
#include <linux/rcutree.h>
#elif defined(CONFIG_PREEMPT_RCU)
#include <linux/rcupreempt.h>
+#elif CONFIG_TINY_RCU
+#include <linux/rcutiny.h>
#else
#error "Unknown RCU implementation specified to kernel configuration"
#endif /* #else #if defined(CONFIG_CLASSIC_RCU) */
diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h
new file mode 100644
index 0000000..f007cfa
--- /dev/null
+++ b/include/linux/rcutiny.h
@@ -0,0 +1,68 @@
+/*
+ * Read-Copy Update mechanism for mutual exclusion, the Bloatwatch edition.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright IBM Corporation, 2008
+ *
+ * Author: Paul E. McKenney <paulmck@...ux.vnet.ibm.com>
+ *
+ * For detailed explanation of Read-Copy Update mechanism see -
+ * Documentation/RCU
+ */
+
+#ifndef __LINUX_TINY_H
+#define __LINUX_TINY_H
+
+#include <linux/cache.h>
+
+/* Global control variables for rcupdate callback mechanism. */
+struct rcu_ctrlblk {
+ long completed; /* Number of last completed batch. */
+ struct rcu_head *rcucblist; /* List of pending callbacks (CBs). */
+ struct rcu_head **donetail; /* ->next pointer of last "done" CB. */
+ struct rcu_head **curtail; /* ->next pointer of last CB. */
+};
+
+void rcu_qsctr_inc(int cpu);
+void rcu_bh_qsctr_inc(int cpu);
+extern int rcu_needs_cpu(int cpu);
+
+#define __rcu_read_lock() preempt_disable()
+#define __rcu_read_unlock() preempt_enable()
+#define __rcu_read_lock_bh() local_bh_disable()
+#define __rcu_read_unlock_bh() local_bh_enable()
+#define __synchronize_sched synchronize_rcu
+#define call_rcu_sched call_rcu
+
+extern void __rcu_init(void);
+#define rcu_init_sched() do { } while (0)
+extern void rcu_check_callbacks(int cpu, int user);
+extern void rcu_restart_cpu(int cpu);
+
+extern long rcu_batches_completed(void);
+extern long rcu_batches_completed_bh(void);
+
+#define rcu_pending(cpu) 1
+
+#ifdef CONFIG_NO_HZ
+void rcu_enter_nohz(void);
+void rcu_exit_nohz(void);
+#else /* #ifdef CONFIG_NO_HZ */
+#define rcu_enter_nohz() do { } while (0)
+#define rcu_exit_nohz() do { } while (0)
+#endif /* #else #ifdef CONFIG_NO_HZ */
+
+#endif /* __LINUX_RCUTINY_H */
diff --git a/init/Kconfig b/init/Kconfig
index a724a14..5188f27 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -1004,6 +1004,13 @@ config PREEMPT_RCU
now-naive assumptions about each RCU read-side critical section
remaining on a given CPU through its execution.
+config TINY_RCU
+ bool "Tiny-Memory RCU"
+ depends on !SMP && EMBEDDED
+ help
+ This option greatly reduces the memory footprint of RCU,
+ but is usable only on UP systems.
+
endchoice
config RCU_TRACE
diff --git a/kernel/Makefile b/kernel/Makefile
index 2921d90..8c1390b 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -76,6 +76,7 @@ obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o
obj-$(CONFIG_CLASSIC_RCU) += rcuclassic.o
obj-$(CONFIG_TREE_RCU) += rcutree.o
obj-$(CONFIG_PREEMPT_RCU) += rcupreempt.o
+obj-$(CONFIG_TINY_RCU) += rcutiny.o
obj-$(CONFIG_TREE_RCU_TRACE) += rcutree_trace.o
obj-$(CONFIG_PREEMPT_RCU_TRACE) += rcupreempt_trace.o
obj-$(CONFIG_RELAY) += relay.o
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index d92a76a..4bb7b0a 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -68,6 +68,8 @@ void wakeme_after_rcu(struct rcu_head *head)
complete(&rcu->completion);
}
+#ifndef CONFIG_TINY_RCU
+
/**
* synchronize_rcu - wait until a grace period has elapsed.
*
@@ -88,6 +90,8 @@ void synchronize_rcu(void)
}
EXPORT_SYMBOL_GPL(synchronize_rcu);
+#endif /* #ifndef CONFIG_TINY_RCU */
+
static void rcu_barrier_callback(struct rcu_head *notused)
{
if (atomic_dec_and_test(&rcu_barrier_cpu_count))
diff --git a/kernel/rcutiny.c b/kernel/rcutiny.c
new file mode 100644
index 0000000..f7f8ca0
--- /dev/null
+++ b/kernel/rcutiny.c
@@ -0,0 +1,290 @@
+/*
+ * Read-Copy Update mechanism for mutual exclusion, the Bloatwatch edition.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright IBM Corporation, 2008
+ *
+ * Author: Paul E. McKenney <paulmck@...ux.vnet.ibm.com>
+ *
+ * For detailed explanation of Read-Copy Update mechanism see -
+ * Documentation/RCU
+ */
+
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/rcupdate.h>
+#include <linux/interrupt.h>
+#include <linux/sched.h>
+#include <linux/module.h>
+#include <linux/completion.h>
+#include <linux/moduleparam.h>
+#include <linux/notifier.h>
+#include <linux/cpu.h>
+#include <linux/mutex.h>
+#include <linux/time.h>
+
+/* Definition for rcupdate control block. */
+static struct rcu_ctrlblk rcu_ctrlblk = {
+ .completed = -300,
+ .rcucblist = NULL,
+ .donetail = &rcu_ctrlblk.rcucblist,
+ .curtail = &rcu_ctrlblk.rcucblist,
+};
+static struct rcu_ctrlblk rcu_bh_ctrlblk = {
+ .completed = -300,
+ .rcucblist = NULL,
+ .donetail = &rcu_bh_ctrlblk.rcucblist,
+ .curtail = &rcu_bh_ctrlblk.rcucblist,
+};
+
+#ifdef CONFIG_NO_HZ
+static long dynticks_nesting = 1;
+#endif /* #ifdef CONFIG_NO_HZ */
+
+/*
+ * Helper function for rcu_qsctr_inc() and rcu_bh_qsctr_inc().
+ */
+static int rcu_qsctr_help(struct rcu_ctrlblk *rcp)
+{
+ if (rcp->rcucblist != NULL &&
+ rcp->donetail != rcp->curtail) {
+ rcp->donetail = rcp->curtail;
+ return 1;
+ }
+ return 0;
+}
+
+/*
+ * Record an rcu quiescent state. And an rcu_bh quiescent state
+ * while we are at it, given that any rcu quiescent state is also
+ * an rcu_bh quiescent state.
+ */
+void rcu_qsctr_inc(int cpu)
+{
+ if (rcu_qsctr_help(&rcu_ctrlblk) ||
+ rcu_qsctr_help(&rcu_bh_ctrlblk))
+ raise_softirq(RCU_SOFTIRQ);
+}
+
+/*
+ * Record an rcu_bh quiescent state.
+ */
+void rcu_bh_qsctr_inc(int cpu)
+{
+ if (rcu_qsctr_help(&rcu_bh_ctrlblk))
+ raise_softirq(RCU_SOFTIRQ);
+}
+
+/*
+ * Return non-zero if there is RCU work remaining to be done.
+ */
+int rcu_needs_cpu(int cpu)
+{
+ return rcu_ctrlblk.rcucblist != NULL ||
+ rcu_bh_ctrlblk.rcucblist != NULL;
+}
+
+/*
+ * Check to see if the scheduling-clock interrupt came from an extended
+ * quiescent state, and, if so, tell RCU about it.
+ */
+void rcu_check_callbacks(int cpu, int user)
+{
+ if (!rcu_needs_cpu(0))
+ return; /* RCU doesn't need anything to be done. */
+ if (user ||
+ (idle_cpu(cpu) &&
+ !in_softirq() &&
+ hardirq_count() <= (1 << HARDIRQ_SHIFT)))
+ rcu_qsctr_inc(cpu);
+ else if (!in_softirq())
+ rcu_bh_qsctr_inc(cpu);
+}
+
+#ifdef CONFIG_NO_HZ
+
+/*
+ * Enter dynticks-idle mode, which is an extended quiescent state
+ * if we have fully entered that mode (i.e., if the new value of
+ * dynticks_nesting is zero).
+ */
+void rcu_enter_nohz(void)
+{
+ if (--dynticks_nesting == 0)
+ rcu_qsctr_inc(0); /* implies rcu_bh_qsctr_inc(0) */
+}
+
+/*
+ * Exit dynticks-idle mode, so that we are no longer in an extended
+ * quiescent state.
+ */
+void rcu_exit_nohz(void)
+{
+ dynticks_nesting++;
+}
+
+/*
+ * Entering an interrupt handler exits nohz mode.
+ */
+void rcu_irq_enter(void)
+{
+ rcu_exit_nohz();
+}
+
+/*
+ * Exiting an interrupt handler enters nohz mode.
+ */
+void rcu_irq_exit(void)
+{
+ rcu_enter_nohz();
+}
+
+void rcu_nmi_enter(void)
+{
+}
+
+void rcu_nmi_exit(void)
+{
+}
+
+#endif /* #ifdef CONFIG_NO_HZ */
+
+/*
+ * Helper function for rcu_process_callbacks() that operates on the
+ * specified rcu_ctrlkblk structure.
+ */
+static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp)
+{
+ unsigned long flags;
+ struct rcu_head *next, *list;
+
+ /* If no RCU callbacks ready to invoke, just return. */
+ if (&rcp->rcucblist == rcp->donetail)
+ return;
+
+ /* Move the ready-to-invoke callbacks to a local list. */
+ local_irq_save(flags);
+ rcp->completed++;
+ list = rcp->rcucblist;
+ rcp->rcucblist = *rcp->donetail;
+ *rcp->donetail = NULL;
+ if (rcp->curtail == rcp->donetail)
+ rcp->curtail = &rcp->rcucblist;
+ rcp->donetail = &rcp->rcucblist;
+ local_irq_restore(flags);
+
+ /* Invoke the callbacks on the local list. */
+ while (list) {
+ next = list->next;
+ prefetch(next);
+ list->func(list);
+ list = next;
+ }
+}
+
+/*
+ * Invoke any callbacks whose grace period has completed.
+ */
+static void rcu_process_callbacks(struct softirq_action *unused)
+{
+ __rcu_process_callbacks(&rcu_ctrlblk);
+ __rcu_process_callbacks(&rcu_bh_ctrlblk);
+}
+
+/*
+ * Wait for a grace period to elapse. But it is illegal to invoke
+ * synchronize_rcu() from within an RCU read-side critical section.
+ * Therefore, any legal call to synchronize_rcu() is a quiescent
+ * state, and so on a UP system, synchronize_rcu() need do nothing.
+ *
+ * Cool, huh? (Due to Josh Triplett.)
+ *
+ * However, we do update the grace-period counter to prevent rcutorture
+ * from hammering us.
+ */
+void synchronize_rcu(void)
+{
+ unsigned long flags;
+
+ local_irq_save(flags);
+ rcu_ctrlblk.completed++;
+ local_irq_restore(flags);
+}
+EXPORT_SYMBOL(synchronize_rcu);
+
+/*
+ * Helper function for call_rcu() and call_rcu_bh().
+ */
+static void __call_rcu(struct rcu_head *head,
+ void (*func)(struct rcu_head *rcu),
+ struct rcu_ctrlblk *rcp)
+{
+ unsigned long flags;
+
+ head->func = func;
+ head->next = NULL;
+ local_irq_save(flags);
+ *rcp->curtail = head;
+ rcp->curtail = &head->next;
+ local_irq_restore(flags);
+}
+
+/*
+ * Post an RCU callback to be invoked after the end of an RCU grace
+ * period. But since we have but one CPU, that would be after any
+ * quiescent state.
+ */
+void call_rcu(struct rcu_head *head,
+ void (*func)(struct rcu_head *rcu))
+{
+ __call_rcu(head, func, &rcu_ctrlblk);
+}
+EXPORT_SYMBOL(call_rcu);
+
+/*
+ * Post an RCU bottom-half callback to be invoked after any subsequent
+ * quiescent state.
+ */
+void call_rcu_bh(struct rcu_head *head,
+ void (*func)(struct rcu_head *rcu))
+{
+ __call_rcu(head, func, &rcu_bh_ctrlblk);
+}
+EXPORT_SYMBOL(call_rcu_bh);
+
+/*
+ * Return the number of grace periods.
+ */
+long rcu_batches_completed(void)
+{
+ return rcu_ctrlblk.completed;
+}
+EXPORT_SYMBOL(rcu_batches_completed);
+
+/*
+ * Return the number of bottom-half grace periods.
+ */
+long rcu_batches_completed_bh(void)
+{
+ return rcu_bh_ctrlblk.completed;
+}
+EXPORT_SYMBOL(rcu_batches_completed_bh);
+
+void __init __rcu_init(void)
+{
+ open_softirq(RCU_SOFTIRQ, rcu_process_callbacks);
+}
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists