| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Fri, 8 Nov 2013 13:21:07 -0800
From: "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
To: Waiman Long <Waiman.Long@...com>
Cc: Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>,
"H. Peter Anvin" <hpa@...or.com>, Arnd Bergmann <arnd@...db.de>,
linux-arch@...r.kernel.org, x86@...nel.org,
linux-kernel@...r.kernel.org,
Peter Zijlstra <peterz@...radead.org>,
Steven Rostedt <rostedt@...dmis.org>,
Andrew Morton <akpm@...ux-foundation.org>,
Michel Lespinasse <walken@...gle.com>,
Andi Kleen <andi@...stfloor.org>,
Rik van Riel <riel@...hat.com>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Raghavendra K T <raghavendra.kt@...ux.vnet.ibm.com>,
George Spelvin <linux@...izon.com>,
Tim Chen <tim.c.chen@...ux.intel.com>,
"Aswin Chandramouleeswaran\"" <aswin@...com>,
Scott J Norton <scott.norton@...com>
Subject: Re: [PATCH v5 4/4] qrwlock: Use the mcs_spinlock helper functions
for MCS queuing
On Mon, Nov 04, 2013 at 12:17:20PM -0500, Waiman Long wrote:
> There is a pending patch in the rwsem patch series that adds a generic
> MCS locking helper functions to do MCS-style locking. This patch
> will enable the queue rwlock to use that generic MCS lock/unlock
> primitives for internal queuing. This patch should only be merged
> after the merging of that generic MCS locking patch.
>
> Signed-off-by: Waiman Long <Waiman.Long@...com>
This one does might address at least some of the earlier memory-barrier
issues, at least assuming that the MCS lock is properly memory-barriered.
Then again, maybe not. Please see below.
Thanx, Paul
> ---
> include/asm-generic/qrwlock.h | 7 +--
> lib/qrwlock.c | 140 +++++++++++++----------------------------
> 2 files changed, 45 insertions(+), 102 deletions(-)
>
> diff --git a/include/asm-generic/qrwlock.h b/include/asm-generic/qrwlock.h
> index 78ad4a5..014e6e9 100644
> --- a/include/asm-generic/qrwlock.h
> +++ b/include/asm-generic/qrwlock.h
> @@ -54,10 +54,7 @@ typedef u64 __nrcpupair_t;
> * QRW_READER_BIAS to the rw field to increment the reader count won't
> * disturb the writer and the fair fields.
> */
> -struct qrwnode {
> - struct qrwnode *next;
> - bool wait; /* Waiting flag */
> -};
> +struct mcs_spinlock;
>
> typedef struct qrwlock {
> union qrwcnts {
> @@ -74,7 +71,7 @@ typedef struct qrwlock {
> };
> __nrcpupair_t rw; /* Reader/writer number pair */
> } cnts;
> - struct qrwnode *waitq; /* Tail of waiting queue */
> + struct mcs_spinlock *waitq; /* Tail of waiting queue */
> } arch_rwlock_t;
>
> /*
> diff --git a/lib/qrwlock.c b/lib/qrwlock.c
> index a85b9e1..6817853 100644
> --- a/lib/qrwlock.c
> +++ b/lib/qrwlock.c
> @@ -20,6 +20,7 @@
> #include <linux/cpumask.h>
> #include <linux/percpu.h>
> #include <linux/hardirq.h>
> +#include <linux/mcs_spinlock.h>
> #include <asm-generic/qrwlock.h>
>
> /*
> @@ -46,87 +47,16 @@
> */
>
> /**
> - * wait_in_queue - Add to queue and wait until it is at the head
> - * @lock: Pointer to queue rwlock structure
> - * @node: Node pointer to be added to the queue
> - *
> - * The use of smp_wmb() is to make sure that the other CPUs see the change
> - * ASAP.
> - */
> -static __always_inline void
> -wait_in_queue(struct qrwlock *lock, struct qrwnode *node)
> -{
> - struct qrwnode *prev;
> -
> - node->next = NULL;
> - node->wait = true;
> - prev = xchg(&lock->waitq, node);
> - if (prev) {
> - prev->next = node;
> - smp_wmb();
> - /*
> - * Wait until the waiting flag is off
> - */
> - while (ACCESS_ONCE(node->wait))
> - cpu_relax();
> - }
> -}
> -
> -/**
> - * signal_next - Signal the next one in queue to be at the head
> - * @lock: Pointer to queue rwlock structure
> - * @node: Node pointer to the current head of queue
> - */
> -static __always_inline void
> -signal_next(struct qrwlock *lock, struct qrwnode *node)
> -{
> - struct qrwnode *next;
> -
> - /*
> - * Try to notify the next node first without disturbing the cacheline
> - * of the lock. If that fails, check to see if it is the last node
> - * and so should clear the wait queue.
> - */
> - next = ACCESS_ONCE(node->next);
> - if (likely(next))
> - goto notify_next;
> -
> - /*
> - * Clear the wait queue if it is the last node
> - */
> - if ((ACCESS_ONCE(lock->waitq) == node) &&
> - (cmpxchg(&lock->waitq, node, NULL) == node))
> - return;
> - /*
> - * Wait until the next one in queue set up the next field
> - */
> - while (likely(!(next = ACCESS_ONCE(node->next))))
> - cpu_relax();
> - /*
> - * The next one in queue is now at the head
> - */
> -notify_next:
> - barrier();
> - ACCESS_ONCE(next->wait) = false;
> - smp_wmb();
> -}
> -
> -/**
> * rspin_until_writer_unlock - inc reader count & spin until writer is gone
> * @lock: Pointer to queue rwlock structure
> + * @cnts: Queue read/write lock counts structure
> *
> * In interrupt context or at the head of the queue, the reader will just
> * increment the reader count & wait until the writer releases the lock.
> */
> static __always_inline void
> -rspin_until_writer_unlock(struct qrwlock *lock, int inc)
> +rspin_until_writer_unlock(struct qrwlock *lock, union qrwcnts cnts)
> {
> - union qrwcnts cnts;
> -
> - if (inc)
> - cnts.rw = xadd(&lock->cnts.rw, QRW_READER_BIAS);
> - else
> - cnts.rw = ACCESS_ONCE(lock->cnts.rw);
> while (cnts.writer == QW_LOCKED) {
> cpu_relax();
> cnts.rw = ACCESS_ONCE(lock->cnts.rw);
> @@ -139,7 +69,7 @@ rspin_until_writer_unlock(struct qrwlock *lock, int inc)
> */
> void queue_read_lock_slowpath(struct qrwlock *lock)
> {
> - struct qrwnode node;
> + struct mcs_spinlock node;
> union qrwcnts cnts;
>
> /*
> @@ -150,7 +80,8 @@ void queue_read_lock_slowpath(struct qrwlock *lock)
> * Readers in interrupt context will spin until the lock is
> * available without waiting in the queue.
> */
> - rspin_until_writer_unlock(lock, 0);
> + cnts.rw = ACCESS_ONCE(lock->cnts.rw);
> + rspin_until_writer_unlock(lock, cnts);
> return;
> }
> cnts.rw = xadd(&lock->cnts.rw, -QRW_READER_BIAS);
> @@ -158,7 +89,7 @@ void queue_read_lock_slowpath(struct qrwlock *lock)
> /*
> * Put the reader into the wait queue
> */
> - wait_in_queue(lock, &node);
> + mcs_spin_lock(&lock->waitq, &node);
>
> /*
> * At the head of the wait queue now, try to increment the reader
> @@ -172,12 +103,36 @@ void queue_read_lock_slowpath(struct qrwlock *lock)
> while (ACCESS_ONCE(lock->cnts.writer))
> cpu_relax();
> }
> - rspin_until_writer_unlock(lock, 1);
> - signal_next(lock, &node);
> + /*
> + * Increment reader count & wait until writer unlock
> + */
> + cnts.rw = xadd(&lock->cnts.rw, QRW_READER_BIAS);
> + rspin_until_writer_unlock(lock, cnts);
> + mcs_spin_unlock(&lock->waitq, &node);
But mcs_spin_unlock() is only required to do a RELEASE barrier, which
could still allow critical-section leakage.
> }
> EXPORT_SYMBOL(queue_read_lock_slowpath);
>
> /**
> + * _write_trylock - try to acquire a write lock
> + * @lock : Pointer to queue rwlock structure
> + * @old : Old value of the qrwcnts
> + * @new : New value of the qrwcnts
> + * Return: 1 if lock acquired, 0 otherwise
> + *
> + * Put old & new as function arguments can force the compiler to generate
> + * better code with less stack memory access.
> + */
> +static __always_inline int _write_trylock(struct qrwlock *lock,
> + union qrwcnts old, union qrwcnts new)
> +{
> + new.rw = old.rw;
> + new.writer = QW_LOCKED;
> + if (likely(cmpxchg(&lock->cnts.rw, old.rw, new.rw) == old.rw))
> + return 1;
> + return 0;
> +}
> +
> +/**
> * queue_write_3step_lock - acquire write lock in 3 steps
> * @lock : Pointer to queue rwlock structure
> * Return: 1 if lock acquired, 0 otherwise
> @@ -194,33 +149,24 @@ static __always_inline int queue_write_3step_lock(struct qrwlock *lock)
> union qrwcnts old, new;
>
> old.rw = ACCESS_ONCE(lock->cnts.rw);
> + new.rw = 0;
>
> /* Step 1 */
> - if (!old.writer & !old.readers) {
> - new.rw = old.rw;
> - new.writer = QW_LOCKED;
> - if (likely(cmpxchg(&lock->cnts.rw, old.rw, new.rw) == old.rw))
> - return 1;
> - }
> + if (!old.writer && !old.readers && _write_trylock(lock, old, new))
> + return 1;
>
> /* Step 2 */
> if (old.writer || (cmpxchg(&lock->cnts.writer, 0, QW_WAITING) != 0))
> return 0;
>
> /* Step 3 */
> - while (true) {
> + cpu_relax();
> + old.rw = ACCESS_ONCE(lock->cnts.rw);
> + while (old.readers || !_write_trylock(lock, old, new)) {
> cpu_relax();
> old.rw = ACCESS_ONCE(lock->cnts.rw);
> - if (!old.readers) {
> - new.rw = old.rw;
> - new.writer = QW_LOCKED;
> - if (likely(cmpxchg(&lock->cnts.rw, old.rw, new.rw)
> - == old.rw))
> - return 1;
> - }
> }
> - /* Should never reach here */
> - return 0;
> + return 1;
This one still seems properly barriered, good!
> }
>
> /**
> @@ -229,12 +175,12 @@ static __always_inline int queue_write_3step_lock(struct qrwlock *lock)
> */
> void queue_write_lock_slowpath(struct qrwlock *lock)
> {
> - struct qrwnode node;
> + struct mcs_spinlock node;
>
> /*
> * Put the writer into the wait queue
> */
> - wait_in_queue(lock, &node);
> + mcs_spin_lock(&lock->waitq, &node);
>
> /*
> * At the head of the wait queue now, call queue_write_3step_lock()
> @@ -242,6 +188,6 @@ void queue_write_lock_slowpath(struct qrwlock *lock)
> */
> while (!queue_write_3step_lock(lock))
> cpu_relax();
> - signal_next(lock, &node);
> + mcs_spin_unlock(&lock->waitq, &node);
> }
> EXPORT_SYMBOL(queue_write_lock_slowpath);
> --
> 1.7.1
>
--
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