public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Nick Piggin <nickpiggin@yahoo.com.au>
To: Tejun Heo <htejun@gmail.com>
Cc: zwane@linuxpower.ca, viro@zeniv.linux.org.uk,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH linux-2.6 01/04] brsem: implement big reader semaphore
Date: Sun, 25 Sep 2005 18:03:45 +1000	[thread overview]
Message-ID: <433659E1.8070905@yahoo.com.au> (raw)
In-Reply-To: <43364F70.7010705@yahoo.com.au>

[-- Attachment #1: Type: text/plain, Size: 630 bytes --]

Nick Piggin wrote:

> What would be wrong with an array of NR_CPUS rwsems? The only
> tiny trick you would have to do AFAIKS is have up_read remember
> what rwsem down_read took, but that could be returned from
> down_read as a token.

Here's something to start with. Add initialisers / cacheline
alignment / percpu_alloc to taste.

For bonus points I guess you could change the implementation
to just a single waitqueue mechanism with NR_CPUs counters.

In fact, it probably needs quite a bit of work before it is
mainline-worthy, but as a proof of concept - do you see
anything wrong with it?

Nick

-- 
SUSE Labs, Novell Inc.


[-- Attachment #2: brsem.patch --]
[-- Type: text/plain, Size: 1987 bytes --]

Index: linux-2.6/include/linux/brsem.h
===================================================================
--- /dev/null
+++ linux-2.6/include/linux/brsem.h
@@ -0,0 +1,18 @@
+#ifndef __BRSEM_H
+#define __BRSEM_H
+
+#include <linux/rwsem.h>
+struct brsem {
+	struct rw_semaphore cpu_sem[NR_CPUS];
+};
+
+#define BRSEM_READ_TRYLOCK_FAILED	-1
+typedef int brsem_read_t;
+
+brsem_read_t brsem_down_read(struct brsem *);
+brsem_read_t brsem_down_read_trylock(struct brsem *);
+void brsem_up_read(struct brsem *, brsem_read_t);
+void brsem_down_write(struct brsem *);
+void brsem_up_write(struct brsem *);
+
+#endif
Index: linux-2.6/lib/Makefile
===================================================================
--- linux-2.6.orig/lib/Makefile
+++ linux-2.6/lib/Makefile
@@ -5,7 +5,7 @@
 lib-y := errno.o ctype.o string.o vsprintf.o cmdline.o \
 	 bust_spinlocks.o rbtree.o radix-tree.o dump_stack.o \
 	 idr.o div64.o int_sqrt.o bitmap.o extable.o prio_tree.o \
-	 sha1.o
+	 sha1.o brsem.o
 
 lib-y	+= kobject.o kref.o kobject_uevent.o klist.o
 
Index: linux-2.6/lib/brsem.c
===================================================================
--- /dev/null
+++ linux-2.6/lib/brsem.c
@@ -0,0 +1,38 @@
+#include <linux/brsem.h>
+#include <linux/rwsem.h>
+#include <linux/smp.h>
+
+brsem_read_t brsem_down_read(struct brsem *brsem)
+{
+	brsem_read_t ret = smp_processor_id();
+	down_read(&brsem->cpu_sem[ret]);
+	return ret;
+}
+
+brsem_read_t brsem_down_read_trylock(struct brsem *brsem)
+{
+	brsem_read_t ret = smp_processor_id();
+	if (!down_read_trylock(&brsem->cpu_sem[ret]))
+		return BRSEM_READ_TRYLOCK_FAILED;
+	return ret;
+}
+
+void brsem_up_read(struct brsem *brsem, brsem_read_t token)
+{
+	up_read(&brsem->cpu_sem[token]);
+}
+
+void brsem_down_write(struct brsem *brsem)
+{
+	int i;
+	for (i = 0; i < NR_CPUS; i++)
+		down_write(&brsem->cpu_sem[i]);
+}
+
+void brsem_up_write(struct brsem *brsem)
+{
+	int i;
+	for (i = 0; i < NR_CPUS; i++)
+		up_write(&brsem->cpu_sem[i]);
+}
+

  reply	other threads:[~2005-09-25  8:03 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-09-25  6:43 [PATCH linux-2.6 00/04] brsem: [RFC] big reader semaphore Tejun Heo
2005-09-25  6:43 ` [PATCH linux-2.6 01/04] brsem: implement " Tejun Heo
2005-09-25  7:19   ` Nick Piggin
2005-09-25  8:03     ` Nick Piggin [this message]
2005-09-25  8:11     ` Tejun Heo
2005-09-25  8:27       ` Nick Piggin
2005-09-25  8:53         ` Tejun Heo
2005-09-25  9:24           ` Nick Piggin
2005-09-25 10:05             ` Tejun Heo
2005-09-25 11:22               ` Nick Piggin
2005-09-25  6:43 ` [PATCH linux-2.6 02/04] brsem: convert super_block->s_umount to brsem Tejun Heo
2005-09-25  6:43 ` [PATCH linux-2.6 03/04] brsem: fix ro-remount <-> open race condition Tejun Heo
2005-09-25  6:43 ` [PATCH linux-2.6 04/04] brsem: convert cpucontrol to brsem Tejun Heo
2005-09-25  7:39   ` Nick Piggin
2005-09-25  8:03     ` Tejun Heo
2005-09-25 23:46     ` Nathan Lynch
2005-09-26  1:11       ` Nick Piggin
2005-09-26  4:05         ` Tejun Heo

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=433659E1.8070905@yahoo.com.au \
    --to=nickpiggin@yahoo.com.au \
    --cc=htejun@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=viro@zeniv.linux.org.uk \
    --cc=zwane@linuxpower.ca \
    /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