Date: Fri, 12 Jul 2002 22:50:07 +0200
From: Dipankar Sarma <dipan...@in.ibm.com>
Subject: [PATCH] Read-Copy Update [new & shiny]
Message-ID: <20020713022002.A24484@in.ibm.com>
Reply-To: dipan...@in.ibm.com
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="RnlQjJ0d97Da+TV1"
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
Sender: robo...@news.nic.it
X-Mailing-List: linux-kernel@vger.kernel.org
Approved: robo...@news.nic.it (1.20)
NNTP-Posting-Host: a.804.anti-phl.bofh.it
Newsgroups: linux.kernel
Organization: linux.*_mail_to_news_unidirectional_gateway
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!syros.belnet.be!news.belnet.be!newsfeed00.sul.t-online.de!t-online.de!bofh.it!robomod
X-Original-Cc: lse-t...@lists.sourceforge.net
X-Original-Date: Sat, 13 Jul 2002 02:20:02 +0530
X-Original-Sender: linux-kernel-ow...@vger.kernel.org
X-Original-To: linux-ker...@vger.kernel.org
Lines: 696
At OLS, we presented some data on latency and overhead for
different approaches to RCU. After some discussions with Andrea and Rusty,
it seemed that it is a good idea to use this and come out with
a patch. So here is a new patch. The main features are -
1. Based on rcu_poll which is already in production kernels (SuSE)
and has low latency of grace period.
2. In order to reduce overheads, I added per-CPU queue support to
it.
3. Works in preemptive kernel environment by either requiring
the use of rcu_read_lock/unlock or using call_rcu_preempt().
4. synchronize_kernel() uses call_rcu_preempt() which requires
no disabling of preemption on read-side making it useful for
things like CPU hotplug.
5. Added docbook tags in headers.
Future plans are -
1. More overhead and latency measurements and tweaking around if
necessary. But for infrequent things like CPU hotplug, this
is not important.
2. Cleanups/bugfixes etc.
Comments, suggestions and even flames are welcome.
Thanks
--
Dipankar Sarma <dipan...@in.ibm.com> http://lse.sourceforge.net
Linux Technology Center, IBM Software Lab, Bangalore, India.
Patch
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!logbridge.uoregon.edu!news.maxwell.syr.edu!uio.no!nntp.uio.no!ifi.uio.no!internet-mailinglist
Newsgroups: fa.linux.kernel
Return-Path: <linux-kernel-owner+fa.linux.kernel=40ifi.uio...@vger.kernel.org>
Original-Date: Thu, 18 Jul 2002 22:33:53 +0530
From: Dipankar Sarma <dipan...@in.ibm.com>
To: Linus Torvalds <torva...@transmeta.com>
Cc: Andrea Arcangeli <and...@suse.de>, Rusty Russell <ru...@rustcorp.com.au>,
Paul McKenney <paul.mcken...@us.ibm.com>, linux-ker...@vger.kernel.org,
Dave Miller <da...@redhat.com>
Subject: [PATCH] Read-Copy Update 2.5.26
Original-Message-ID: <20020718223352.A5765@in.ibm.com>
Reply-To: dipan...@in.ibm.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
Sender: linux-kernel-ow...@vger.kernel.org
Precedence: bulk
X-Mailing-List: linux-kernel@vger.kernel.org
Organization: Internet mailing list
Date: Thu, 18 Jul 2002 17:01:09 GMT
Message-ID: <fa.dsdtdnv.qla5ga@ifi.uio.no>
Lines: 682
Hi Linus,
Here is the 2.5.26 RCU patch with some minor changes from the
earlier version published a few days ago. Since 2.5.25 it has -
1. Cleaned up a compile warning with preemptible kernel
2. Named initializers - [wli nitpick]
3. ISO style named initializer - [ my nitpick ]
Currently, it seems useful for -
1. CPU hotplug (synchronize_kernel())
2. ipv4 route cache lookup
3. dentry cache lookup [ more analysis and code cleanup required here]
If the time has come, please consider it for inclusion.
Thanks
--
Dipankar Sarma <dipan...@in.ibm.com> http://lse.sourceforge.net
Linux Technology Center, IBM Software Lab, Bangalore, India.
Patch
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!cyclone.bc.net!news.maxwell.syr.edu!uio.no!nntp.uio.no!ifi.uio.no!internet-mailinglist
Newsgroups: fa.linux.kernel
Return-Path: <linux-kernel-owner+fa.linux.kernel=40ifi.uio...@vger.kernel.org>
Original-Date: Thu, 25 Jul 2002 22:21:46 +0530
From: Dipankar Sarma <dipan...@in.ibm.com>
To: Linus Torvalds <torva...@transmeta.com>, da...@suse.de
Cc: linux-ker...@vger.kernel.org, Rusty Russell <ru...@rustcorp.com.au>,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>
Subject: [PATCH] Read-Copy Update 2.5.28
Original-Message-ID: <20020725222146.A12780@in.ibm.com>
Reply-To: dipan...@in.ibm.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
Sender: linux-kernel-ow...@vger.kernel.org
Precedence: bulk
X-Mailing-List: linux-kernel@vger.kernel.org
Organization: Internet mailing list
Date: Thu, 25 Jul 2002 16:49:16 GMT
Message-ID: <fa.bvnnlev.1138j3j@ifi.uio.no>
Lines: 696
Here is a 2.5.28 RCU patch with some minor changes from the
earlier version published a few days ago. Please consider it
for inclusion.
The current rcu_poll is essentially a per-CPU version of the
rcu_poll patch in -aa series of kernel. Additionally it also supports
call_rcu_preempt() which allows RCU to work in preemptive kernels
transparently.
2.5.26
------
1. Cleaned up a compile warning with preemptible kernel
2. Named initializers - [wli nitpick]
3. ISO style named initializer - [ my nitpick ]
2.5.28
------
1. Workaround for compiler breakage for uninitialized __per_cpu_data
with fake initialization.
Currently, it seems useful for -
1. CPU hotplug (synchronize_kernel())
2. ipv4 route cache lookup
3. dentry cache lookup [ more results and new patch on the way ]
4. Potential uses in XFS (Christoph ?)
It has been tested with LTP in a 4CPU PIII xeon system.
Thanks
--
Dipankar Sarma <dipan...@in.ibm.com> http://lse.sourceforge.net
Linux Technology Center, IBM Software Lab, Bangalore, India.
Patch
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!cyclone.bc.net!news.maxwell.syr.edu!uio.no!nntp.uio.no!ifi.uio.no!internet-mailinglist
Newsgroups: fa.linux.kernel
Return-Path: <linux-kernel-owner+fa.linux.kernel=40ifi.uio...@vger.kernel.org>
Original-Date: Fri, 26 Jul 2002 13:21:07 +0530
From: Dipankar Sarma <dipan...@in.ibm.com>
To: Linus Torvalds <torva...@transmeta.com>, da...@suse.de
Cc: linux-ker...@vger.kernel.org, Rusty Russell <ru...@rustcorp.com.au>,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>
Subject: [PATCH] Read-Copy Update 2.5.28 [resent]
Original-Message-ID: <20020726132107.D16440@in.ibm.com>
Reply-To: dipan...@in.ibm.com
Original-References: <20020725222146.A12...@in.ibm.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20020725222146.A12780@in.ibm.com>; from dipankar@in.ibm.com on Thu, Jul 25, 2002 at 10:21:46PM +0530
Sender: linux-kernel-ow...@vger.kernel.org
Precedence: bulk
X-Mailing-List: linux-kernel@vger.kernel.org
Organization: Internet mailing list
Date: Fri, 26 Jul 2002 07:48:20 GMT
Message-ID: <fa.bt83kmv.12j0jrp@ifi.uio.no>
References: <fa.bvnnlev.1138j3j@ifi.uio.no>
Lines: 699
Hi Linus,
I submitted a wrong version of the patch yesterday instead of
submitting the compiled and tested version of the rcu_poll
patch from my lab machine. This one is correct. Sorry about that.
Here is a 2.5.28 RCU patch with some minor changes from the
earlier version published a few days ago. Please consider it
for inclusion.
The current rcu_poll is essentially a per-CPU version of the
rcu_poll patch in -aa series of kernel. Additionally it also supports
call_rcu_preempt() which allows RCU to work in preemptive kernels
transparently.
2.5.26
------
1. Cleaned up a compile warning with preemptible kernel
2. Named initializers - [wli nitpick]
3. ISO style named initializer - [ my nitpick ]
2.5.28
------
1. Workaround for compiler breakage for uninitialized __per_cpu_data
with fake initialization.
Currently, it seems useful for -
1. CPU hotplug (synchronize_kernel())
2. ipv4 route cache lookup
3. dentry cache lookup [ more results and new patch on the way ]
4. Potential uses in XFS (Christoph ?)
It has been tested with LTP in a 4CPU PIII xeon system.
Thanks
Dipankar
Patch
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Path: archiver1.google.com!news1.google.com!sn-xit-02!sn-xit-06!sn-xit-04!supernews.com!news.maxwell.syr.edu!news-hog.berkeley.edu!ucberkeley!enews.sgi.com!news-out.spamkiller.net!propagator2-la!propagator-la!news-in.superfeed.net!news.exit.com!gehenna.pell.portland.or.us!nntp-server.caltech.edu!nntp-server.caltech.edu!mail2news96
Newsgroups: mlist.linux.kernel
Date: Sun, 28 Jul 2002 10:57:48 +0530
From: Dipankar Sarma <dipan...@in.ibm.com>
X-To: Linus Torvalds <torva...@transmeta.com>, da...@suse.de
X-Cc: Rusty Russell <ru...@rustcorp.com.au>, Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>, linux-ker...@vger.kernel.org
Subject: [PATCH] Read-Copy Update 2.5.29
Message-ID: <linux.kernel.20020728105748.A5387@in.ibm.com>
Reply-To: dipan...@in.ibm.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Approved: n...@nntp-server.caltech.edu
Lines: 698
Hi Linus,
Here is RCU 2.5.29.
The current rcu_poll is essentially a per-CPU version of the
rcu_poll patch in -aa series of kernel. Additionally it also supports
call_rcu_preempt() which allows RCU to work in preemptive kernels
transparently.
2.5.26
------
1. Cleaned up a compile warning with preemptible kernel
2. Named initializers - [wli nitpick]
3. ISO style named initializer - [ my nitpick ]
2.5.28
------
1. Workaround for compiler breakage for uninitialized __per_cpu_data
with fake initialization.
2.5.29
------
1. Just rediffed and tested.
Currently, it seems useful for -
1. CPU hotplug (synchronize_kernel())
2. ipv4 route cache lookup
3. dentry cache lookup [ more results and new patch on the way ]
4. Potential uses in XFS (Christoph ?)
It has been tested with LTP in a 4CPU PIII xeon system.
Thanks
--
Dipankar Sarma <dipan...@in.ibm.com> http://lse.sourceforge.net
Linux Technology Center, IBM Software Lab, Bangalore, India.
Patch
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!news.tele.dk!small.news.tele.dk!195.54.122.107!newsfeed1.bredband.com!bredband!uio.no!nntp.uio.no!ifi.uio.no!internet-mailinglist
Newsgroups: fa.linux.kernel
Return-Path: <linux-kernel-owner+fa.linux.kernel=40ifi.uio...@vger.kernel.org>
Original-Date: Mon, 29 Jul 2002 10:30:08 +0200
From: Pavel Machek <pa...@suse.cz>
To: Dipankar Sarma <dipan...@in.ibm.com>
Cc: Linus Torvalds <torva...@transmeta.com>, da...@suse.de,
linux-ker...@vger.kernel.org, Rusty Russell <ru...@rustcorp.com.au>,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>
Subject: Re: [PATCH] Read-Copy Update 2.5.28 [resent]
Original-Message-ID: <20020729083007.GA115@elf.ucw.cz>
Original-References: <20020725222146.A12...@in.ibm.com> <20020726132107.D16...@in.ibm.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20020726132107.D16440@in.ibm.com>
User-Agent: Mutt/1.3.28i
X-Warning: Reading this can be dangerous to your mental health.
Sender: linux-kernel-ow...@vger.kernel.org
Precedence: bulk
X-Mailing-List: linux-kernel@vger.kernel.org
Organization: Internet mailing list
Date: Mon, 29 Jul 2002 14:18:24 GMT
Message-ID: <fa.c6udd3v.164m28v@ifi.uio.no>
References: <fa.bt83kmv.12j0jrp@ifi.uio.no>
Lines: 19
Hi!
> Currently, it seems useful for -
>
> 1. CPU hotplug (synchronize_kernel())
> 2. ipv4 route cache lookup
> 3. dentry cache lookup [ more results and new patch on the way ]
> 4. Potential uses in XFS (Christoph ?)
Can it be used to make module unloading less painfull?
Pavel
--
Worst form of spam? Adding advertisment signatures ala sourceforge.net.
What goes next? Inserting advertisment *into* email?
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!uio.no!news-feed.ifi.uio.no!ifi.uio.no!internet-mailinglist
Newsgroups: fa.linux.kernel
Return-Path: <linux-kernel-owner+fa.linux.kernel=40ifi.uio...@vger.kernel.org>
X-Authentication-Warning: dualathlon.random: andrea set sender to and...@suse.de using -f
Original-Date: Mon, 29 Jul 2002 18:11:35 +0200
From: Andrea Arcangeli <and...@suse.de>
To: Pavel Machek <pa...@suse.cz>
Cc: Dipankar Sarma <dipan...@in.ibm.com>,
Linus Torvalds <torva...@transmeta.com>, da...@suse.de,
linux-ker...@vger.kernel.org, Rusty Russell <ru...@rustcorp.com.au>,
Paul McKenney <paul.mcken...@us.ibm.com>
Subject: Re: [PATCH] Read-Copy Update 2.5.28 [resent]
Original-Message-ID: <20020729161135.GT1201@dualathlon.random>
Original-References: <20020725222146.A12...@in.ibm.com> <20020726132107.D16...@in.ibm.com> <20020729083007.GA...@elf.ucw.cz>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20020729083007.GA115@elf.ucw.cz>
User-Agent: Mutt/1.3.27i
X-GnuPG-Key-URL: http://e-mind.com/~andrea/aa.gnupg.asc
X-PGP-Key-URL: http://e-mind.com/~andrea/aa.asc
Sender: linux-kernel-ow...@vger.kernel.org
Precedence: bulk
X-Mailing-List: linux-kernel@vger.kernel.org
Organization: Internet mailing list
Date: Mon, 29 Jul 2002 16:12:30 GMT
Message-ID: <fa.gapurav.1kmi6ob@ifi.uio.no>
References: <fa.c6udd3v.164m28v@ifi.uio.no>
Lines: 22
On Mon, Jul 29, 2002 at 10:30:08AM +0200, Pavel Machek wrote:
> Hi!
>
> > Currently, it seems useful for -
> >
> > 1. CPU hotplug (synchronize_kernel())
> > 2. ipv4 route cache lookup
> > 3. dentry cache lookup [ more results and new patch on the way ]
> > 4. Potential uses in XFS (Christoph ?)
>
> Can it be used to make module unloading less painfull?
yes of course, we just do this in 2.4 too. It doesn't cover all the
possible cases though, but it gives additional guarantees that you can
rely on and that's enough for most cases.
Andrea
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Date: Mon, 05 Aug 2002 13:10:08 +0200
From: Dipankar Sarma <dipan...@in.ibm.com>
Subject: [PATCH] Read-Copy Update 2.5.30
Message-ID: <20020805163922.A5009@in.ibm.com>
Reply-To: dipan...@in.ibm.com
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
Sender: robo...@news.nic.it
X-Mailing-List: linux-kernel@vger.kernel.org
Approved: robo...@news.nic.it (1.20)
NNTP-Posting-Host: a.189.anti-phl.bofh.it
Newsgroups: linux.kernel
Organization: linux.*_mail_to_news_unidirectional_gateway
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!logbridge.uoregon.edu!fr.usenet-edu.net!usenet-edu.net!newsfeed00.sul.t-online.de!t-online.de!bofh.it!robomod
X-Original-Cc: Rusty Russell <ru...@rustcorp.com.au>,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>,
linux-ker...@vger.kernel.org
X-Original-Date: Mon, 5 Aug 2002 16:39:22 +0530
X-Original-Sender: linux-kernel-ow...@vger.kernel.org
X-Original-To: Linus Torvalds <torva...@transmeta.com>, da...@suse.de
Lines: 696
Hi Linus,
Here is RCU for 2.5.30.
The current rcu_poll is essentially a per-CPU version of the
rcu_poll patch in -aa series of kernel. Additionally it also supports
call_rcu_preempt() which allows RCU to work in preemptive kernels
transparently.
2.5.29
------
1. Just rediffed and tested.
2.5.30
------
1. Just rediffed and sanity tested.
Currently, it seems useful for -
1. CPU hotplug (synchronize_kernel())
2. ipv4 route cache lookup
3. dentry cache lookup [ more results and new patch on the way ]
[webserver benchmarks show more simultaneous connections
compared to vanila 2.5.25 - detailed results to be published
soon]
4. Potential uses in XFS (Christoph ?)
It has been sanity tested with in a 4-CPU PIII xeon system.
Thanks
--
Dipankar Sarma <dipan...@in.ibm.com> http://lse.sourceforge.net
Linux Technology Center, IBM Software Lab, Bangalore, India.
Patch
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Date: Mon, 26 Aug 2002 22:50:07 +0200
From: Dipankar Sarma <dipan...@in.ibm.com>
Subject: [BKPATCH] Read-Copy Update 2.5
Message-ID: <20020827022239.C31269@in.ibm.com>
Reply-To: dipan...@in.ibm.com
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
Sender: robo...@news.nic.it
X-Mailing-List: linux-kernel@vger.kernel.org
Approved: robo...@news.nic.it (1.20)
NNTP-Posting-Host: a.523.anti-phl.bofh.it
Newsgroups: linux.kernel
Organization: linux.*_mail_to_news_unidirectional_gateway
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!cyclone.bc.net!news.mailgate.org!bofh.it!robomod
X-Original-Cc: Rusty Russell <ru...@rustcorp.com.au>,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>,
linux-ker...@vger.kernel.org
X-Original-Date: Tue, 27 Aug 2002 02:22:39 +0530
X-Original-Sender: linux-kernel-ow...@vger.kernel.org
X-Original-To: Linus Torvalds <torva...@transmeta.com>, da...@suse.de
Lines: 709
Hi,
Read-Copy Update is now available in bk for pulling -
http://lse.bkbits.net/linux-2.5-rcu
The current rcu_poll is essentially a per-CPU version of the
rcu_poll patch in -aa series of kernel. Additionally it also supports
call_rcu_preempt() which allows RCU to work in preemptive kernels
transparently.
ChangeSet:
---------
Change...@1.498, 2002-08-27 00:27:13+05:30, dipan...@in.ibm.com
Add RCU infrastructure based on rcu_poll in -aa kernels with support
for preemption and per-CPU queues.
Thanks
--
Dipankar Sarma <dipan...@in.ibm.com> http://lse.sourceforge.net
Linux Technology Center, IBM Software Lab, Bangalore, India.
Patch
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Rusty Russell <ru...@rustcorp.com.au>
Subject: Re: [BKPATCH] Read-Copy Update 2.5
In-Reply-To: Your message of "Tue, 27 Aug 2002 02:22:39 +0530."
<20020827022239.C31269@in.ibm.com>
Date: Tue, 27 Aug 2002 02:40:06 +0200
Message-ID: <20020826193708.0C64C2C07B@lists.samba.org>
Sender: robo...@news.nic.it
X-Mailing-List: linux-kernel@vger.kernel.org
Approved: robo...@news.nic.it (1.20)
NNTP-Posting-Host: a.730.anti-phl.bofh.it
Newsgroups: linux.kernel
Organization: linux.*_mail_to_news_unidirectional_gateway
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!cyclone.bc.net!news.mailgate.org!bofh.it!robomod
References: <20020827022239.C31269@in.ibm.com>
X-Original-Cc: linux-ker...@vger.kernel.org,
Linus Torvalds <torva...@transmeta.com>, da...@suse.de,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>
X-Original-Date: Tue, 27 Aug 2002 10:24:30 +1000
X-Original-Sender: linux-kernel-ow...@vger.kernel.org
X-Original-To: dipan...@in.ibm.com
Lines: 32
In message <20020827022239.C31...@in.ibm.com> you write:
> +static struct rcu_data rcu_data[NR_CPUS] __cacheline_aligned;
Not "static DEFINE_PER_CPU(struct rcu_data, rcu_data)"?
> +/* Fake initialization to work around compiler breakage */
> +DEFINE_PER_CPU(long, cpu_quiescent) = 0L;
static? And I assume you're talking about the tendency for gcc 2.95
to put uninitialized static vars in the bss, even if they are marked
as having a section attribute? If so, you should say so.
> +#ifdef CONFIG_PREEMPT
> +/* Fake initialization to work around compiler breakage */
> +DEFINE_PER_CPU(atomic_t[2], rcu_preempt_cntr) =
> + {ATOMIC_INIT(0), ATOMIC_INIT(0)};
> +DEFINE_PER_CPU(atomic_t, *curr_preempt_cntr) = NULL;
> +DEFINE_PER_CPU(atomic_t, *next_preempt_cntr) = NULL;
Also static I assume?
Other than that, it looks good. You should probably cc: Ingo Molnar
as it touches the scheduler...
Rusty.
--
Anyone who quotes me in their sig is an idiot. -- Rusty Russell.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Date: Tue, 27 Aug 2002 08:10:04 +0200
From: Dipankar Sarma <dipan...@in.ibm.com>
Subject: Re: [BKPATCH] Read-Copy Update 2.5
Message-ID: <20020827114152.A2072@in.ibm.com>
Reply-To: dipan...@in.ibm.com
References: <20020827022239.C31269@in.ibm.com> <20020826193708.0C64C2C07B@lists.samba.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <20020826193708.0C64C2C07B@lists.samba.org>; from rusty@rustcorp.com.au on Tue, Aug 27, 2002 at 10:24:30AM +1000
Sender: robo...@news.nic.it
X-Mailing-List: linux-kernel@vger.kernel.org
Approved: robo...@news.nic.it (1.20)
NNTP-Posting-Host: a.21.anti-phl.bofh.it
Newsgroups: linux.kernel
Organization: linux.*_mail_to_news_unidirectional_gateway
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!logbridge.uoregon.edu!news.maxwell.syr.edu!nntp.infostrada.it!bofh.it!robomod
X-Original-Cc: linux-ker...@vger.kernel.org,
Linus Torvalds <torva...@transmeta.com>, da...@suse.de,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>
X-Original-Date: Tue, 27 Aug 2002 11:41:52 +0530
X-Original-Sender: linux-kernel-ow...@vger.kernel.org
X-Original-To: Rusty Russell <ru...@rustcorp.com.au>
Lines: 48
Hi Rusty,
On Tue, Aug 27, 2002 at 10:24:30AM +1000, Rusty Russell wrote:
> In message <20020827022239.C31...@in.ibm.com> you write:
> > +static struct rcu_data rcu_data[NR_CPUS] __cacheline_aligned;
>
> Not "static DEFINE_PER_CPU(struct rcu_data, rcu_data)"?
Yes, I can use per-cpu data areas here. Done.
>
> > +/* Fake initialization to work around compiler breakage */
> > +DEFINE_PER_CPU(long, cpu_quiescent) = 0L;
>
> static? And I assume you're talking about the tendency for gcc 2.95
> to put uninitialized static vars in the bss, even if they are marked
> as having a section attribute? If so, you should say so.
>
> > +#ifdef CONFIG_PREEMPT
> > +/* Fake initialization to work around compiler breakage */
> > +DEFINE_PER_CPU(atomic_t[2], rcu_preempt_cntr) =
> > + {ATOMIC_INIT(0), ATOMIC_INIT(0)};
> > +DEFINE_PER_CPU(atomic_t, *curr_preempt_cntr) = NULL;
> > +DEFINE_PER_CPU(atomic_t, *next_preempt_cntr) = NULL;
>
> Also static I assume?
So, only statics are broken by gcc 2.95, right ? If so, then
the fake initializers aren't required. I got bitten by this in
another piece of RCU code where there was a static per-cpu
tasklet and I got paranoic after that.
>
> Other than that, it looks good. You should probably cc: Ingo Molnar
> as it touches the scheduler...
Ok, I will do that from now on. Thanks for the review.
I will have the new bits up as soon as I finish testing.
Thanks
--
Dipankar Sarma <dipan...@in.ibm.com> http://lse.sourceforge.net
Linux Technology Center, IBM Software Lab, Bangalore, India.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Date: Wed, 11 Sep 2002 13:20:06 +0200
From: Dipankar Sarma <dipan...@in.ibm.com>
Subject: [PATCH] Read-Copy Update 2.5.34
Message-ID: <20020911164940.C28198@in.ibm.com>
Reply-To: dipan...@in.ibm.com
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
Sender: robo...@news.nic.it
X-Mailing-List: linux-kernel@vger.kernel.org
Approved: robo...@news.nic.it (1.20)
NNTP-Posting-Host: a.814.anti-phl.bofh.it
Newsgroups: linux.kernel
Organization: linux.*_mail_to_news_unidirectional_gateway
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!newsmi-us.news.garr.it!newsmi-eu.news.garr.it!NewsITBone-GARR!elvis.franken.de!chico.franken.de!news.franken.de!uni-erlangen.de!news.uni-stuttgart.de!news.belwue.de!newsfeed01.sul.t-online.de!t-online.de!nntp.infostrada.it!bofh.it!robomod
X-Original-Cc: Ingo Molnar <mi...@elte.hu>,
Rusty Russell <ru...@rustcorp.com.au>,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>,
linux-ker...@vger.kernel.org
X-Original-Date: Wed, 11 Sep 2002 16:49:40 +0530
X-Original-Sender: linux-kernel-ow...@vger.kernel.org
X-Original-To: Linus Torvalds <torva...@transmeta.com>
Lines: 686
Hi,
Here is RCU for 2.5.34. It is just a rediff from earlier BK version.
The current rcu_poll is essentially a per-CPU version of the
rcu_poll patch in -aa series of kernel. Additionally it also supports
call_rcu_preempt() which allows RCU to work in preemptive kernels
transparently.
Ingo, it will be nice to have your comments on this, specially since
it touches the scheduler.
Read-Copy Update is also available in bk for pulling -
http://lse.bkbits.net/linux-2.5-rcu
Thanks
--
Dipankar Sarma <dipan...@in.ibm.com> http://lse.sourceforge.net
Linux Technology Center, IBM Software Lab, Bangalore, India.
Patch
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Date: Wed, 11 Sep 2002 13:30:06 +0200
From: Ingo Molnar <mi...@elte.hu>
Reply-To: Ingo Molnar <mi...@elte.hu>
Subject: Re: [PATCH] Read-Copy Update 2.5.34
In-Reply-To: <20020911164940.C28198@in.ibm.com>
Message-ID: <Pine.LNX.4.44.0209111323360.12332-100000@localhost.localdomain>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: robo...@news.nic.it
X-Mailing-List: linux-kernel@vger.kernel.org
Approved: robo...@news.nic.it (1.20)
NNTP-Posting-Host: a.290.anti-phl.bofh.it
Newsgroups: linux.kernel
Organization: linux.*_mail_to_news_unidirectional_gateway
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!canoe.uoregon.edu!logbridge.uoregon.edu!snoopy.risq.qc.ca!newsfeed.news2me.com!bofh.it!robomod
References: <20020911164940.C28198@in.ibm.com>
X-Original-Cc: Linus Torvalds <torva...@transmeta.com>,
Rusty Russell <ru...@rustcorp.com.au>,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>,
<linux-ker...@vger.kernel.org>
X-Original-Date: Wed, 11 Sep 2002 13:24:21 +0200 (CEST)
X-Original-Sender: linux-kernel-ow...@vger.kernel.org
X-Original-To: Dipankar Sarma <dipan...@in.ibm.com>
Lines: 16
On Wed, 11 Sep 2002, Dipankar Sarma wrote:
> Ingo, it will be nice to have your comments on this, specially since it
> touches the scheduler.
i dont really understand why it has to change the scheduler. You want a
facility to force a reschedule on any given CPU, correct?
Ingo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Date: Wed, 11 Sep 2002 14:20:05 +0200
From: Dipankar Sarma <dipan...@in.ibm.com>
Subject: Re: [PATCH] Read-Copy Update 2.5.34
Message-ID: <20020911175011.D28198@in.ibm.com>
Reply-To: dipan...@in.ibm.com
References: <20020911164940.C28198@in.ibm.com> <Pine.LNX.4.44.0209111323360.12332-100000@localhost.localdomain>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <Pine.LNX.4.44.0209111323360.12332-100000@localhost.localdomain>; from mingo@elte.hu on Wed, Sep 11, 2002 at 01:24:21PM +0200
Sender: robo...@news.nic.it
X-Mailing-List: linux-kernel@vger.kernel.org
Approved: robo...@news.nic.it (1.20)
NNTP-Posting-Host: a.391.anti-phl.bofh.it
Newsgroups: linux.kernel
Organization: linux.*_mail_to_news_unidirectional_gateway
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!cyclone.bc.net!news.mailgate.org!bofh.it!robomod
X-Original-Cc: Linus Torvalds <torva...@transmeta.com>,
Rusty Russell <ru...@rustcorp.com.au>,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>,
linux-ker...@vger.kernel.org
X-Original-Date: Wed, 11 Sep 2002 17:50:11 +0530
X-Original-Sender: linux-kernel-ow...@vger.kernel.org
X-Original-To: Ingo Molnar <mi...@elte.hu>
Lines: 53
On Wed, Sep 11, 2002 at 01:24:21PM +0200, Ingo Molnar wrote:
>
> i dont really understand why it has to change the scheduler. You want a
> facility to force a reschedule on any given CPU, correct?
>
Hi Ingo,
Yes, like force_cpu_reschedule().
Apart from that RCU adds some minor things in the scheduler code -
1. Per-CPU context switch counter increment in the fast path
2. For preemptive kernels, a conditional branch in the voluntary
context switch code path checking whether the current task may
have had an involuntary context switch earlier [rcu_preempt_put()].
3. Adding a field to the task structure - cpu_preempt_cntr.
4. RCU checking hook into the scheduler_tick(), but this is not
in the fast path.
#2 and #3 are necessary to support call_rcu_preempt() which
allows preemption-safe reads of data protected by RCU. A preempted
task may still contain references to RCU protected data and
the RCU grace period needs to be prolonged until all such tasks
before the update do voluntary context switches.
I did some reflex benchmarking to make sure that I didn't introduce
any false sharing by mistake in scheduler fast path and the results
look comparable -
(4CPU P3 Xeon with 1MB L2 + 1GB RAM)
vanilla-2.5.34 rcu_poll-2.5.34
-------------- ---------------
80 , 40 , 1.593 1.569
112 , 40 , 1.544 1.554
144 , 40 , 1.595 1.552
176 , 40 , 1.568 1.605
198 , 40 , 1.562 1.577
230 , 40 , 1.563 1.583
244 , 40 , 1.671 1.638
Not sure how reliable these numbers are.
Thanks
--
Dipankar Sarma <dipan...@in.ibm.com> http://lse.sourceforge.net
Linux Technology Center, IBM Software Lab, Bangalore, India.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Subject: Re: [PATCH] Read-Copy Update 2.5.34
From: Robert Love <r...@tech9.net>
In-Reply-To: <20020911175011.D28198@in.ibm.com>
References: <20020911164940.C28198@in.ibm.com>
<Pine.LNX.4.44.0209111323360.12332-100000@localhost.localdomain>
<20020911175011.D28198@in.ibm.com>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
X-Mailer: Ximian Evolution 1.0.8
Date: Wed, 11 Sep 2002 16:10:06 +0200
Message-ID: <1031753011.950.106.camel@phantasy>
MIME-Version: 1.0
Sender: robo...@news.nic.it
X-Mailing-List: linux-kernel@vger.kernel.org
Approved: robo...@news.nic.it (1.20)
NNTP-Posting-Host: a.993.anti-phl.bofh.it
Newsgroups: linux.kernel
Organization: linux.*_mail_to_news_unidirectional_gateway
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!nntp.infostrada.it!bofh.it!robomod
X-Original-Cc: Ingo Molnar <mi...@elte.hu>,
Linus Torvalds <torva...@transmeta.com>,
Rusty Russell <ru...@rustcorp.com.au>,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>,
linux-ker...@vger.kernel.org
X-Original-Date: 11 Sep 2002 10:03:28 -0400
X-Original-Sender: linux-kernel-ow...@vger.kernel.org
X-Original-To: dipan...@in.ibm.com
Lines: 27
On Wed, 2002-09-11 at 08:20, Dipankar Sarma wrote:
> vanilla-2.5.34 rcu_poll-2.5.34
> -------------- ---------------
> 80 , 40 , 1.593 1.569
> 112 , 40 , 1.544 1.554
> 144 , 40 , 1.595 1.552
> 176 , 40 , 1.568 1.605
> 198 , 40 , 1.562 1.577
> 230 , 40 , 1.563 1.583
> 244 , 40 , 1.671 1.638
>
> Not sure how reliable these numbers are.
And how bad is the performance drop from 2.5.34-preempt to
2.5.34-preempt-rcu?
I am glad you guys support kernel preemption (not that you have a chance
at this point) but I hope it was not an afterthought.
Robert Love
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Date: Wed, 11 Sep 2002 16:30:12 +0200
From: Dipankar Sarma <dipan...@in.ibm.com>
Subject: Re: [PATCH] Read-Copy Update 2.5.34
Message-ID: <20020911195939.E28198@in.ibm.com>
Reply-To: dipan...@in.ibm.com
References: <20020911164940.C28198@in.ibm.com> <Pine.LNX.4.44.0209111323360.12332-100000@localhost.localdomain> <20020911175011.D28198@in.ibm.com> <1031753011.950.106.camel@phantasy>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <1031753011.950.106.camel@phantasy>; from rml@tech9.net on Wed, Sep 11, 2002 at 10:03:28AM -0400
Sender: robo...@news.nic.it
X-Mailing-List: linux-kernel@vger.kernel.org
Approved: robo...@news.nic.it (1.20)
NNTP-Posting-Host: a.635.anti-phl.bofh.it
Newsgroups: linux.kernel
Organization: linux.*_mail_to_news_unidirectional_gateway
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed.icl.net!newsfeed.fjserv.net!news.mailgate.org!bofh.it!robomod
X-Original-Cc: Ingo Molnar <mi...@elte.hu>,
Linus Torvalds <torva...@transmeta.com>,
Rusty Russell <ru...@rustcorp.com.au>,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>,
linux-ker...@vger.kernel.org
X-Original-Date: Wed, 11 Sep 2002 19:59:39 +0530
X-Original-Sender: linux-kernel-ow...@vger.kernel.org
X-Original-To: Robert Love <r...@tech9.net>
Lines: 48
On Wed, Sep 11, 2002 at 10:03:28AM -0400, Robert Love wrote:
> On Wed, 2002-09-11 at 08:20, Dipankar Sarma wrote:
>
> > vanilla-2.5.34 rcu_poll-2.5.34
> > -------------- ---------------
> > 80 , 40 , 1.593 1.569
> > 112 , 40 , 1.544 1.554
> > 144 , 40 , 1.595 1.552
> > 176 , 40 , 1.568 1.605
> > 198 , 40 , 1.562 1.577
> > 230 , 40 , 1.563 1.583
> > 244 , 40 , 1.671 1.638
> >
> > Not sure how reliable these numbers are.
>
> And how bad is the performance drop from 2.5.34-preempt to
> 2.5.34-preempt-rcu?
>
> I am glad you guys support kernel preemption (not that you have a chance
> at this point) but I hope it was not an afterthought.
Hi Robert,
Sorry, I should have been more careful labelling them - those are 2.5.34-preempt
vs 2.5.34-preempt-rcu numbers. I did them first because rcu-poll-preempt
kernel has a conditinal branch in fast path and hence more interesting.
I will publish the vanilla vs rcu_poll reflex numbers in a few minutes
from now.
The preemption support have been in RCU for a very long time. IIRC, I added
it in around 2.5.14.
See http://marc.theaimsgroup.com/?l=linux-kernel&m=102084967517192&w=2
Our OLS paper and presentation too deals with preemption -
http://www.rdrop.com/users/paulmck/rclock/rcu.2002.07.08.pdf
http://www.rdrop.com/users/paulmck/rclock/rclock.OLS.2002.07.08a.pdf
Thanks
--
Dipankar Sarma <dipan...@in.ibm.com> http://lse.sourceforge.net
Linux Technology Center, IBM Software Lab, Bangalore, India.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Date: Wed, 11 Sep 2002 18:40:05 +0200
From: Dipankar Sarma <dipan...@in.ibm.com>
Subject: Re: [PATCH] Read-Copy Update 2.5.34
Message-ID: <20020911220514.G28198@in.ibm.com>
Reply-To: dipan...@in.ibm.com
References: <20020911164940.C28198@in.ibm.com> <Pine.LNX.4.44.0209111323360.12332-100000@localhost.localdomain> <20020911175011.D28198@in.ibm.com> <1031753011.950.106.camel@phantasy>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <1031753011.950.106.camel@phantasy>; from rml@tech9.net on Wed, Sep 11, 2002 at 10:03:28AM -0400
Sender: robo...@news.nic.it
X-Mailing-List: linux-kernel@vger.kernel.org
Approved: robo...@news.nic.it (1.20)
NNTP-Posting-Host: a.355.anti-phl.bofh.it
Newsgroups: linux.kernel
Organization: linux.*_mail_to_news_unidirectional_gateway
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!canoe.uoregon.edu!logbridge.uoregon.edu!news.maxwell.syr.edu!nntp.infostrada.it!bofh.it!robomod
X-Original-Cc: Ingo Molnar <mi...@elte.hu>,
Linus Torvalds <torva...@transmeta.com>,
Rusty Russell <ru...@rustcorp.com.au>,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>,
linux-ker...@vger.kernel.org
X-Original-Date: Wed, 11 Sep 2002 22:05:14 +0530
X-Original-Sender: linux-kernel-ow...@vger.kernel.org
X-Original-To: Robert Love <r...@tech9.net>
Lines: 42
On Wed, Sep 11, 2002 at 10:03:28AM -0400, Robert Love wrote:
> On Wed, 2002-09-11 at 08:20, Dipankar Sarma wrote:
>
> > Not sure how reliable these numbers are.
>
> And how bad is the performance drop from 2.5.34-preempt to
> 2.5.34-preempt-rcu?
Hi Robert,
Here are the detailed results from reflex benchmark all with
2.5.34 kernel and 4CPU p3 xeon with 1MB L2 cache and 1GB RAM.
vanilla-preempt rcu_poll-preempt vanilla rcu_poll
-------------- ---------------- ------- --------
80 , 40 , 1.593 1.569 1.545 1.536
112 , 40 , 1.544 1.554 1.544 1.535
144 , 40 , 1.595 1.552 1.545 1.586
176 , 40 , 1.568 1.605 1.615 1.536
198 , 40 , 1.562 1.577 1.582 1.651
230 , 40 , 1.563 1.583 1.581 1.554
244 , 40 , 1.671 1.638 1.631 1.571
>
> I am glad you guys support kernel preemption (not that you have a chance
> at this point) but I hope it was not an afterthought.
>
Forgot to mention, earlier implementations of RCU like in K42 had complete
preemption support, so the linux implementation is not really new.
We added it after preemption support went into Linus' tree.
Thanks
--
Dipankar Sarma <dipan...@in.ibm.com> http://lse.sourceforge.net
Linux Technology Center, IBM Software Lab, Bangalore, India.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Subject: Re: [PATCH] Read-Copy Update 2.5.34
From: Robert Love <r...@tech9.net>
In-Reply-To: <20020911195939.E28198@in.ibm.com>
References: <20020911164940.C28198@in.ibm.com>
<Pine.LNX.4.44.0209111323360.12332-100000@localhost.localdomain>
<20020911175011.D28198@in.ibm.com> <1031753011.950.106.camel@phantasy>
<20020911195939.E28198@in.ibm.com>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
X-Mailer: Ximian Evolution 1.0.8
Date: Wed, 11 Sep 2002 21:00:14 +0200
Message-ID: <1031770371.4693.113.camel@phantasy>
MIME-Version: 1.0
Sender: robo...@news.nic.it
X-Mailing-List: linux-kernel@vger.kernel.org
Approved: robo...@news.nic.it (1.20)
NNTP-Posting-Host: a.140.anti-phl.bofh.it
Newsgroups: linux.kernel
Organization: linux.*_mail_to_news_unidirectional_gateway
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!logbridge.uoregon.edu!fr.usenet-edu.net!usenet-edu.net!news.newsland.it!area.cu.mi.it!bofh.it!robomod
X-Original-Cc: Ingo Molnar <mi...@elte.hu>,
Linus Torvalds <torva...@transmeta.com>,
Rusty Russell <ru...@rustcorp.com.au>,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>,
linux-ker...@vger.kernel.org
X-Original-Date: 11 Sep 2002 14:52:46 -0400
X-Original-Sender: linux-kernel-ow...@vger.kernel.org
X-Original-To: dipan...@in.ibm.com
Lines: 25
On Wed, 2002-09-11 at 10:29, Dipankar Sarma wrote:
> Sorry, I should have been more careful labelling them - those are 2.5.34-preempt
> vs 2.5.34-preempt-rcu numbers. I did them first because rcu-poll-preempt
> kernel has a conditinal branch in fast path and hence more interesting.
> I will publish the vanilla vs rcu_poll reflex numbers in a few minutes
> from now.
Ahh OK. Results are OK, then - thanks :-)
> Our OLS paper and presentation too deals with preemption -
>
> http://www.rdrop.com/users/paulmck/rclock/rcu.2002.07.08.pdf
> http://www.rdrop.com/users/paulmck/rclock/rclock.OLS.2002.07.08a.pdf
Yep. Was there, saw the talk, got the t-shirt.
Robert Love
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Date: Wed, 18 Sep 2002 14:40:09 +0200
From: Dipankar Sarma <dipan...@in.ibm.com>
Subject: [PATCH] Read-Copy Update 2.5.36
Message-ID: <20020918181216.B20865@in.ibm.com>
Reply-To: dipan...@in.ibm.com
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
Sender: robo...@news.nic.it
X-Mailing-List: linux-kernel@vger.kernel.org
Approved: robo...@news.nic.it (1.20)
NNTP-Posting-Host: a.513.anti-phl.bofh.it
Newsgroups: linux.kernel
Organization: linux.*_mail_to_news_unidirectional_gateway
Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!logbridge.uoregon.edu!news.maxwell.syr.edu!nntp.infostrada.it!bofh.it!robomod
X-Original-Cc: Rusty Russell <ru...@rustcorp.com.au>,
Andrea Arcangeli <and...@suse.de>,
Paul McKenney <paul.mcken...@us.ibm.com>,
Ingo Molnar <mi...@elte.hu>, linux-ker...@vger.kernel.org
X-Original-Date: Wed, 18 Sep 2002 18:12:16 +0530
X-Original-Sender: linux-kernel-ow...@vger.kernel.org
X-Original-To: Linus Torvalds <torva...@transmeta.com>
Lines: 682
Hi,
Here is RCU for 2.5.36. It is just a rediff from earlier version.
The current rcu_poll is essentially a per-CPU version of the
rcu_poll patch in -aa series of kernel. Additionally it also supports
call_rcu_preempt() which allows RCU to work in preemptive kernels
transparently. The synchronize_kernel() interface uses call_rcu_preempt()
since it is not concerned about latencies. synchronize_kernel() is
equivalent of wait_for_later() in the "later" patch published
by Rusty earlier as a part of the new in-kernel module loader.
The module-loader needs only a part of rcu_poll funcationality.
Thanks
--
Dipankar Sarma <dipan...@in.ibm.com> http://lse.sourceforge.net
Linux Technology Center, IBM Software Lab, Bangalore, India.
Patch
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
|