From: Michel Lespinasse <walken@google.com>
To: Linus Torvalds <torvalds@linux-foundation.org>,
David Howells <dhowells@redhat.com>, Ingo Molnar <mingo@elte.hu>,
Thomas Gleixner <tglx@linutronix.de>
Cc: LKML <linux-kernel@vger.kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Mike Waychison <mikew@google.com>,
Suleiman Souhlal <suleiman@google.com>,
Ying Han <yinghan@google.com>,
David Howells <dhowells@redhat.com>
Subject: [PATCH 09/11] rwsem: down_read_critical infrastructure support
Date: Mon, 24 May 2010 13:31:19 -0700 [thread overview]
Message-ID: <1274733081-4623-10-git-send-email-walken@google.com> (raw)
In-Reply-To: <1274733081-4623-1-git-send-email-walken@google.com>
Add rwsem_down_read_unfair_failed() function in the non-generic rwsem
library code, similar to rwsem_down_read_failed() except that blocked
threads are placed at the head of the queue.
Make the waiter type an enumeration rather than a bitmask.
Signed-off-by: Michel Lespinasse <walken@google.com>
Signed-off-by: David Howells <dhowells@redhat.com>
---
lib/rwsem.c | 40 +++++++++++++++++++++++++++++++---------
1 files changed, 31 insertions(+), 9 deletions(-)
diff --git a/lib/rwsem.c b/lib/rwsem.c
index f236d7c..72454f4 100644
--- a/lib/rwsem.c
+++ b/lib/rwsem.c
@@ -28,12 +28,16 @@ void __init_rwsem(struct rw_semaphore *sem, const char *name,
EXPORT_SYMBOL(__init_rwsem);
+enum rwsem_waiter_type {
+ RWSEM_WAITING_FOR_WRITE,
+ RWSEM_WAITING_FOR_READ,
+ RWSEM_WAITING_FOR_UNFAIR_READ
+};
+
struct rwsem_waiter {
struct list_head list;
struct task_struct *task;
- unsigned int flags;
-#define RWSEM_WAITING_FOR_READ 0x00000001
-#define RWSEM_WAITING_FOR_WRITE 0x00000002
+ enum rwsem_waiter_type type;
};
/* Wake types for __rwsem_do_wake(). Note that RWSEM_WAKE_NO_ACTIVE and
@@ -63,7 +67,7 @@ __rwsem_do_wake(struct rw_semaphore *sem, int wake_type)
signed long oldcount, woken, loop, adjustment;
waiter = list_entry(sem->wait_list.next, struct rwsem_waiter, list);
- if (!(waiter->flags & RWSEM_WAITING_FOR_WRITE))
+ if (waiter->type != RWSEM_WAITING_FOR_WRITE)
goto readers_only;
if (wake_type == RWSEM_WAKE_READ_OWNED)
@@ -132,10 +136,10 @@ __rwsem_do_wake(struct rw_semaphore *sem, int wake_type)
waiter = list_entry(waiter->list.next,
struct rwsem_waiter, list);
- } while (waiter->flags & RWSEM_WAITING_FOR_READ);
+ } while (waiter->type != RWSEM_WAITING_FOR_WRITE);
adjustment = woken * RWSEM_ACTIVE_READ_BIAS;
- if (waiter->flags & RWSEM_WAITING_FOR_READ)
+ if (waiter->type != RWSEM_WAITING_FOR_WRITE)
/* hit end of list above */
adjustment -= RWSEM_WAITING_BIAS;
@@ -171,7 +175,7 @@ __rwsem_do_wake(struct rw_semaphore *sem, int wake_type)
*/
static struct rw_semaphore __sched *
rwsem_down_failed_common(struct rw_semaphore *sem,
- unsigned int flags, signed long adjustment)
+ enum rwsem_waiter_type type, signed long adjustment)
{
struct rwsem_waiter waiter;
struct task_struct *tsk = current;
@@ -182,12 +186,16 @@ rwsem_down_failed_common(struct rw_semaphore *sem,
/* set up my own style of waitqueue */
spin_lock_irq(&sem->wait_lock);
waiter.task = tsk;
- waiter.flags = flags;
+ waiter.type = type;
get_task_struct(tsk);
if (list_empty(&sem->wait_list))
adjustment += RWSEM_WAITING_BIAS;
- list_add_tail(&waiter.list, &sem->wait_list);
+
+ if (type == RWSEM_WAITING_FOR_UNFAIR_READ)
+ list_add(&waiter.list, &sem->wait_list);
+ else
+ list_add_tail(&waiter.list, &sem->wait_list);
/* we're now waiting on the lock, but no longer actively locking */
count = rwsem_atomic_update(adjustment, sem);
@@ -229,6 +237,20 @@ rwsem_down_read_failed(struct rw_semaphore *sem)
-RWSEM_ACTIVE_READ_BIAS);
}
+#ifdef __HAVE_DOWN_READ_UNFAIR
+
+/*
+ * wait for the read lock to be granted - skip waiting threads
+ */
+asmregparm struct rw_semaphore __sched *
+rwsem_down_read_unfair_failed(struct rw_semaphore *sem)
+{
+ return rwsem_down_failed_common(sem, RWSEM_WAITING_FOR_UNFAIR_READ,
+ -RWSEM_ACTIVE_READ_BIAS);
+}
+
+#endif
+
/*
* wait for the write lock to be granted
*/
--
1.7.0.1
next prev parent reply other threads:[~2010-05-24 20:33 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-05-24 20:31 [PATCH 00/11] V4: rwsem changes + down_read_critical() proposal Michel Lespinasse
2010-05-24 20:31 ` [PATCH 01/11] x86 rwsem: stay on fast path when count>0 in __up_write() Michel Lespinasse
2010-05-24 20:31 ` [PATCH 02/11] x86 rwsem: minor cleanups Michel Lespinasse
2010-05-24 20:31 ` [PATCH 03/11] rwsem: fully separate code pathes to wake writers vs readers Michel Lespinasse
2010-05-24 20:31 ` [PATCH 04/11] rwsem: lighter active count checks when waking up readers Michel Lespinasse
2010-05-24 20:49 ` Daniel Walker
2010-05-24 20:31 ` [PATCH 05/11] rwsem: let RWSEM_WAITING_BIAS represent any number of waiting threads Michel Lespinasse
2010-05-24 20:31 ` [PATCH 06/11] rwsem: wake queued readers when writer blocks on active read lock Michel Lespinasse
2010-08-12 1:24 ` Tony Luck
2010-08-12 5:02 ` Michel Lespinasse
2010-08-12 5:09 ` Michel Lespinasse
2010-08-12 15:57 ` Linus Torvalds
2010-08-12 16:24 ` Tony Luck
2010-08-12 22:23 ` Tony Luck
2010-08-13 16:09 ` Tony Luck
2010-08-13 21:19 ` Tony Luck
2010-08-13 23:38 ` Tony Luck
2010-08-13 23:48 ` Michel Lespinasse
2010-08-12 16:22 ` Tony Luck
2010-05-24 20:31 ` [PATCH 07/11] rwsem: smaller wrappers around rwsem_down_failed_common Michel Lespinasse
2010-05-24 20:31 ` [PATCH 08/11] generic rwsem: implement down_read_critical() / up_read_critical() Michel Lespinasse
2010-05-24 20:31 ` Michel Lespinasse [this message]
2010-05-24 20:31 ` [PATCH 10/11] x86 rwsem: down_read_critical implementation Michel Lespinasse
2010-05-24 20:31 ` [PATCH 11/11] Use down_read_critical() for /proc/<pid>/exe and /proc/<pid>/maps files Michel Lespinasse
2010-05-24 21:12 ` Daniel Walker
2010-05-27 14:12 ` Arjan van de Ven
2010-05-25 8:47 ` [PATCH 00/11] V4: rwsem changes + down_read_critical() proposal Peter Zijlstra
2010-05-25 9:12 ` Michel Lespinasse
2010-05-25 9:27 ` Peter Zijlstra
2010-05-26 1:23 ` KAMEZAWA Hiroyuki
2010-05-27 11:02 ` Michel Lespinasse
2010-05-27 11:45 ` Peter Zijlstra
2010-05-27 10:59 ` Michel Lespinasse
2010-05-27 11:18 ` Peter Zijlstra
2010-05-27 11:47 ` Michel Lespinasse
2010-05-25 15:16 ` Linus Torvalds
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=1274733081-4623-10-git-send-email-walken@google.com \
--to=walken@google.com \
--cc=akpm@linux-foundation.org \
--cc=dhowells@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mikew@google.com \
--cc=mingo@elte.hu \
--cc=suleiman@google.com \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=yinghan@google.com \
/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
Powered by JetHome