t; mutex_unlock(&mynode->mutex);
> rcu_read_unlock()
> call_rcu(free, mynode);
>
> To perform this efficiently, we need an API function to return whether
> the node previously looked-up has been deleted since then.
My first thought was that the user can do this with a flag
On Thu, Apr 19, 2012 at 08:02:57PM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Wed, Apr 18, 2012 at 04:10:17PM -0400, Mathieu Desnoyers wrote:
> > > Hi,
> > >
> > > FYI, I pushed extra documentation of the RCU
On Mon, Apr 23, 2012 at 03:17:45PM -0400, Mathieu Desnoyers wrote:
> Hi Paul,
>
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Thu, Apr 19, 2012 at 08:02:57PM -0400, Mathieu Desnoyers wrote:
> > > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
>
On Mon, Apr 23, 2012 at 03:45:32PM -0400, Mathieu Desnoyers wrote:
> Hi Paul,
>
> Here is the updated text I plan for the next update. Comments are
> welcome, thanks !
Looks much improved! The inevitable questions and comments interspersed.
On Mon, Apr 23, 2012 at 10:54:54PM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Mon, Apr 23, 2012 at 03:45:32PM -0400, Mathieu Desnoyers wrote:
> > > Hi Paul,
> > >
> > > Here is the updated text I pl
On Tue, May 01, 2012 at 10:16:09AM -0400, Mathieu Desnoyers wrote:
> Hi!
>
> After 1 year of development, with the last 6-7 months spent polishing
> the API and testing the implementation, I think it is getting about time
> to release the RCU lock-free hash table in a new Userspace RCU version
> (
On Tue, May 01, 2012 at 11:12:15AM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Tue, May 01, 2012 at 10:16:09AM -0400, Mathieu Desnoyers wrote:
> > > Hi!
> > >
> > > After 1 year of development, with the last
On Tue, May 01, 2012 at 11:21:44AM -0400, Mathieu Desnoyers wrote:
> * Mathieu Desnoyers (mathieu.desnoy...@efficios.com) wrote:
> > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > > On Tue, May 01, 2012 at 10:16:09AM -0400, Mathieu Desnoyers wrote:
> > > >
On Tue, May 01, 2012 at 01:41:36PM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Tue, May 01, 2012 at 11:21:44AM -0400, Mathieu Desnoyers wrote:
> > > * Mathieu Desnoyers (mathieu.desnoy...@efficios.com) wrote:
> > >
On Tue, May 01, 2012 at 05:02:06PM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Tue, May 01, 2012 at 01:41:36PM -0400, Mathieu Desnoyers wrote:
> > > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > > > On T
On Wed, May 02, 2012 at 12:16:39AM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Tue, May 01, 2012 at 05:02:06PM -0400, Mathieu Desnoyers wrote:
> > > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > > > On T
On Thu, May 03, 2012 at 01:13:30PM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Wed, May 02, 2012 at 12:16:39AM -0400, Mathieu Desnoyers wrote:
> > > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > > > On T
On Fri, May 04, 2012 at 12:53:12PM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Thu, May 03, 2012 at 01:13:30PM -0400, Mathieu Desnoyers wrote:
> > > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> [...]
> > &
On Sat, May 05, 2012 at 02:22:12PM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Fri, May 04, 2012 at 12:53:12PM -0400, Mathieu Desnoyers wrote:
> > > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > > > On T
On Mon, May 07, 2012 at 12:10:55PM -0400, Mathieu Desnoyers wrote:
>
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Fri, May 04, 2012 at 12:53:12PM -0400, Mathieu Desnoyers wrote:
> [...]
> > Just to make sure I understand -- the reason that the "del&quo
On Tue, May 08, 2012 at 02:48:27PM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Mon, May 07, 2012 at 12:10:55PM -0400, Mathieu Desnoyers wrote:
> > >
> > > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> &g
gt; diff --git a/doc/uatomic-api.txt b/doc/uatomic-api.txt
> new file mode 100644
> index 000..3605acf
> --- /dev/null
> +++ b/doc/uatomic-api.txt
> @@ -0,0 +1,80 @@
> +Userspace RCU Atomic Operations API
> +by Mathieu Desnoyers and Paul E. McKenney
> +
> +
> +Th
On Mon, May 14, 2012 at 11:36:04PM -0400, Mathieu Desnoyers wrote:
> Document the concurrent data structures provided by the userspace RCU
> library.
Looks good to me!
Reviewed-by: Paul E. McKenney
> Signed-off-by: Mathieu Desnoyers
> ---
> diff --git a/doc/cds-api.txt b/doc/cd
On Tue, May 15, 2012 at 08:10:03AM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Mon, May 14, 2012 at 10:39:01PM -0400, Mathieu Desnoyers wrote:
> > > Document each atomic operation provided by urcu/uatomic.h, along with
> >
On Wed, May 16, 2012 at 09:56:58AM -0400, Mathieu Desnoyers wrote:
> Suggested-by: Marek Vavruša
> Signed-off-by: Mathieu Desnoyers
Interesting! Jeffrey Yasskin of Google was suggesting use of
pthread_get_specific() over __thread even where __thread was available.
His justification is that pthr
On Wed, May 16, 2012 at 02:17:42PM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Tue, May 15, 2012 at 08:10:03AM -0400, Mathieu Desnoyers wrote:
> > > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > > > On M
On Wed, May 16, 2012 at 11:45:39AM -0700, Josh Triplett wrote:
> On Wed, May 16, 2012 at 11:32:38AM -0700, Paul E. McKenney wrote:
> > On Wed, May 16, 2012 at 02:17:42PM -0400, Mathieu Desnoyers wrote:
> > > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > > &
On Thu, May 17, 2012 at 01:59:43PM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Wed, May 16, 2012 at 11:45:39AM -0700, Josh Triplett wrote:
> > > On Wed, May 16, 2012 at 11:32:38AM -0700, Paul E. McKenney wrote:
> > > &
On Thu, May 17, 2012 at 06:04:13PM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Thu, May 17, 2012 at 01:59:43PM -0400, Mathieu Desnoyers wrote:
> > > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > > > On W
On Sun, May 20, 2012 at 11:28:29AM -0400, Mathieu Desnoyers wrote:
> * Christoph Hellwig (h...@infradead.org) wrote:
> > We could have a local include/uatomic.h that includes either based on
> > which one is available, still erroring out if none is present.
> >
> > I've cced the LTTng list to se i
On Thu, Jun 21, 2012 at 12:41:13PM -0400, Mathieu Desnoyers wrote:
> Hi,
>
> Currently, liburcu calls "exit(-1)" upon internal consistency error.
> This is not pretty, and usually frowned upon in libraries.
>
> One example of failure path where we use this is if pthread_mutex_lock()
> would happe
; --- a/urcu/static/wfqueue.h
> +++ b/urcu/static/wfqueue.h
> @@ -10,6 +10,7 @@
> * dynamically with the userspace rcu library.
> *
> * Copyright 2010 - Mathieu Desnoyers
> + * Copyright 2011-2012 - Lai Jiangshan
> *
> * This library is free software; you can redistribute it
On Fri, Aug 10, 2012 at 10:16:15AM -0400, Mathieu Desnoyers wrote:
> (for the records: this discussion is about userspace RCU tls-compat.h,
> which uses TLS on systems supporting it, and fall back to
> pthread_key_create/getspecific/setspecific if not. The culprit of the
> issue is that we want to
This commit ensures that all read-side functions meet the 10-line LGPL
criterion that permits them to be expanded directly into non-LGPL code,
without function-call instructions. It also documents this as the intent.
Signed-off-by: Paul E. McKenney
diff --git a/urcu/static/urcu-bp.h b/urcu
On Sat, Sep 01, 2012 at 10:13:55PM -0700, Josh Triplett wrote:
> On Sat, Sep 01, 2012 at 05:59:11PM -0700, Paul E. McKenney wrote:
> > This commit ensures that all read-side functions meet the 10-line LGPL
> > criterion that permits them to be expanded directly into non-LGPL co
On Mon, Sep 03, 2012 at 02:03:00PM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > This commit ensures that all read-side functions meet the 10-line LGPL
> > criterion that permits them to be expanded directly into non-LGPL code,
>
On Tue, Oct 02, 2012 at 10:13:07AM -0400, Mathieu Desnoyers wrote:
> Implement wait-free concurrent queues, with a new API different from
> wfqueue.h, which is already provided by Userspace RCU. The advantage of
> splitting the head and tail objects of the queue into different
> arguments is to all
On Wed, Oct 03, 2012 at 05:04:36PM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Tue, Oct 02, 2012 at 10:13:07AM -0400, Mathieu Desnoyers wrote:
> > > Implement wait-free concurrent queues, with a new API different from
> &g
On Mon, Oct 08, 2012 at 10:49:16AM -0400, Mathieu Desnoyers wrote:
> * Lai Jiangshan (la...@cn.fujitsu.com) wrote:
> > On 10/02/2012 10:16 PM, Mathieu Desnoyers wrote:
> > > Eliminate false-sharing between call_rcu (enqueuer) and worker threads
> > > on the queue head and tail.
> > >
> > > Signed-
I expect that commit
>
> commit 5161f31e09ce33dd79afad8d08a2372fbf1c4fbe
> Author: Mathieu Desnoyers
> Date: Tue Sep 25 10:50:49 2012 -0500
>
> call_rcu: use wfcqueue, eliminate false-sharing
>
> Eliminate false-sharing between call_rcu (enqueuer)
On Wed, Oct 10, 2012 at 03:52:08PM +0800, Lai Jiangshan wrote:
> rculfstack is not really require RCU-only.
>
> 1) cds_lfs_push_rcu() don't need any lock, don't need RCU nor other locks.
>
> 2) cds_lfs_pop_rcu() don't only one of the following synchronization(not only
> RCU):
> A) use rcu_
On Wed, Oct 10, 2012 at 01:53:04PM -0400, Mathieu Desnoyers wrote:
> * Mathieu Desnoyers (mathieu.desnoy...@efficios.com) wrote:
> > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > > On Wed, Oct 10, 2012 at 07:42:15AM -0400, Mathieu Desnoyers wrote:
> >
On Thu, Oct 11, 2012 at 09:31:01AM +0800, Lai Jiangshan wrote:
> On 10/11/2012 03:50 AM, Paul E. McKenney wrote:
> > On Wed, Oct 10, 2012 at 01:53:04PM -0400, Mathieu Desnoyers wrote:
> >> * Mathieu Desnoyers (mathieu.desnoy...@efficios.com) wrote:
> >>
On Sun, Oct 14, 2012 at 01:53:32PM -0400, Mathieu Desnoyers wrote:
> Hi Paul!
>
> I know you are currently looking at documentation of urcu data
> structures. I did quite a bit of work in that area these past days. Here
> is my plan:
Actually, I diverted to the atomic operations, given that the s
On Wed, Oct 17, 2012 at 11:19:46AM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Sun, Oct 14, 2012 at 01:53:32PM -0400, Mathieu Desnoyers wrote:
> > > Hi Paul!
> > >
> > > I know you are currently l
FYI, userspace RCU proposed to solve an issue with epoll.
Thanx, Paul
- Forwarded message from Matt Helsley -
Date: Fri, 26 Oct 2012 14:52:42 -0700
From: Matt Helsley
To: "Michael Kerrisk (man-pages)"
Cc: "Paton J. Lewis" , Alexa
The current implementation of cds_wfcq_dequeue_lock() and
cds_wfcq_dequeue_unlock() entails mutually assured recursion.
Redirect to _cds_wfcq_dequeue_lock() and _cds_wfcq_dequeue_unlock(),
respectively.
Signed-off-by: Paul E. McKenney
diff --git a/wfcqueue.c b/wfcqueue.c
index 3474ee0..90b810e
On Mon, Nov 19, 2012 at 03:52:18PM +0800, Lai Jiangshan wrote:
> On 11/18/2012 12:16 AM, Mathieu Desnoyers wrote:
> > Here are a couple of improvements for all userspace RCU flavors. Many
> > thanks to Alan Stern for his suggestions.
>
> It makes urcu like SRCU. (sync_rcu = check zero + flip + che
r.
> >>> With 4 readers, single updater:
> >>>
> >>> * Serialized grace periods :
> >>>
> >>> ./test_urcu_qsbr 4 1 20
> >>> SUMMARY ./test_urcu_qsbr testdur 20 nr_readers 4
> >>> rdur 0 wdur 0 n
On Fri, Dec 07, 2012 at 12:22:52PM -0500, Mathieu Desnoyers wrote:
> * Lai Jiangshan (eag0...@gmail.com) wrote:
> > On Saturday, December 8, 2012, Mathieu Desnoyers wrote:
> >
> > > * Lai Jiangshan (eag0...@gmail.com ) wrote:
> > > > we can define rcu_gp_ctr and registry with aligned attribute, bu
On Mon, Dec 10, 2012 at 01:01:12PM -0500, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Fri, Dec 07, 2012 at 12:22:52PM -0500, Mathieu Desnoyers wrote:
> > > * Lai Jiangshan (eag0...@gmail.com) wrote:
> > > > On Satur
On Thu, Dec 13, 2012 at 06:44:56AM -0500, Mathieu Desnoyers wrote:
> I noticed that in addition to having:
>
> - push/enqueue returning whether the stack/queue was empty prior to the
> operation,
> - pop_all/splice, by nature, emptying the stack/queue,
>
> it can be interesting to make pop/dequ
tore be volatile keeps gcc from splitting.
This commit therefore applies ACCESS_ONCE() to CMM_STORE_SHARED(),
which is the underlying primitive used by rcu_assign_pointer().
Signed-off-by: Paul E. McKenney
diff --git a/urcu/system.h b/urcu/system.h
index 2a45f22..7a1887e 100644
--- a/urcu/syst
[Sorry for the delay, finally getting back to this.]
On Mon, Dec 17, 2012 at 09:40:09AM -0500, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Thu, Dec 13, 2012 at 06:44:56AM -0500, Mathieu Desnoyers wrote:
> > > I noticed that in
On Wed, Jan 16, 2013 at 07:50:54AM -0500, Mathieu Desnoyers wrote:
> * Mathieu Desnoyers (mathieu.desnoy...@efficios.com) wrote:
> > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > > As noted by Konstantin Khlebnikov, gcc can split assignment of
> > > constan
On Sun, Jan 20, 2013 at 03:51:31PM -0500, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Wed, Jan 16, 2013 at 07:50:54AM -0500, Mathieu Desnoyers wrote:
> > > * Mathieu Desnoyers (mathieu.desnoy...@efficios.com) wrote:
> > >
On Thu, Feb 14, 2013 at 11:19:33AM -0500, Mathieu Desnoyers wrote:
> I've had a report of someone running into issues with the RCU lock-free
> hash table by embedding the struct cds_lfht_node into a packed structure
> by mistake, thus not respecting alignment requirements stated in
> urcu/rculfhash
snoyers
Reviewed-by: Paul E. McKenney
> diff --git a/urcu/list.h b/urcu/list.h
> index 5d04394..1d1c7b5 100644
> --- a/urcu/list.h
> +++ b/urcu/list.h
> @@ -140,12 +140,17 @@ cds_list_splice (struct cds_list_head *add, struct
> cds_list_head *head)
> #define cds_list_f
On Tue, May 07, 2013 at 07:59:14AM -0400, Mathieu Desnoyers wrote:
> Hi Richard,
>
> * Richard Braun (rbr...@sceen.net) wrote:
> > Hello,
> >
> > I'm currently studying RCU/URCU, and I have a few questions that I wasn't
> > sure where to ask.
> >
> > 1/ Why use poll instead of sched_yield in e.g
On Tue, May 07, 2013 at 04:51:12PM +0200, Richard Braun wrote:
> On Tue, May 07, 2013 at 07:23:52AM -0700, Paul E. McKenney wrote:
> > Yes, LGPLv2+ allows you to migrate to LGPLv3+, which is compatible with
> > GPLv3+. So as long as you derive the code from lttng's userspace-
On Fri, May 31, 2013 at 11:35:17AM -0400, Mathieu Desnoyers wrote:
> Awaits for all in-flight call_rcu handlers to complete execution before
> returning.
>
> Signed-off-by: Mathieu Desnoyers
One suggestion below, looks good in general.
On Wed, Jun 05, 2013 at 09:32:10AM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Fri, May 31, 2013 at 11:35:17AM -0400, Mathieu Desnoyers wrote:
> [...]
> > > +
> > > + if (work_count != count)
> > > +
lization or cleanup, so it is OK for it to update
pointers in an unsafe-for-RCU-readers manner. This commit therefore
creates an INIT_LIST_HEAD_RCU() that uses ACCESS_ONCE() to make the updates
reader-safe. The reason that we can use ACCESS_ONCE() instead of the more
typical rcu_assign_pointer(
On Fri, Aug 23, 2013 at 12:09:56PM -0700, Stephen Hemminger wrote:
> On Fri, 23 Aug 2013 13:16:53 -0400
> Mathieu Desnoyers wrote:
>
> > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > > On Thu, Aug 22, 2013 at 09:33:18PM -0700, Stephen Hemminger wrote:
>
On Fri, Aug 23, 2013 at 01:16:53PM -0400, Mathieu Desnoyers wrote:
> * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > On Thu, Aug 22, 2013 at 09:33:18PM -0700, Stephen Hemminger wrote:
> > > I needed to add into the middle of an RCU list, do
On Fri, Aug 23, 2013 at 02:08:22PM -0700, Paul E. McKenney wrote:
> On Fri, Aug 23, 2013 at 01:16:53PM -0400, Mathieu Desnoyers wrote:
> > * Paul E. McKenney (paul...@linux.vnet.ibm.com) wrote:
> > > On Thu, Aug 22, 2013 at 09:33:18PM -0700, Steph
On Thu, Aug 29, 2013 at 07:16:37PM -0700, Josh Triplett wrote:
> On Thu, Aug 29, 2013 at 05:57:33PM -0700, Paul E. McKenney wrote:
> > On Fri, Aug 23, 2013 at 02:08:22PM -0700, Paul E. McKenney wrote:
> > > On Fri, Aug 23, 2013 at 01:16:53PM -0400, Mathieu Desnoyers wrot
On Sun, Sep 01, 2013 at 01:42:10PM -0700, Josh Triplett wrote:
> On Sat, Aug 31, 2013 at 02:32:28PM -0700, Paul E. McKenney wrote:
> > On Thu, Aug 29, 2013 at 07:16:37PM -0700, Josh Triplett wrote:
> > > On Thu, Aug 29, 2013 at 05:57:33PM -0700, Paul E. McKenney wrote:
> >
On Sun, Sep 01, 2013 at 04:42:52PM -0700, Josh Triplett wrote:
> rcu_assign_pointer needs to use ACCESS_ONCE to make the assignment to
> the destination pointer volatile, to protect against compilers too
> clever for their own good.
>
> In addition, since rcu_assign_pointer force-casts the source
On Wed, Sep 11, 2013 at 02:54:41PM -0400, Mathieu Desnoyers wrote:
> * John Stultz (john.stu...@linaro.org) wrote:
> > On 09/11/2013 08:08 AM, Mathieu Desnoyers wrote:
> [...]
>
> Now focusing on features (the fix discussion is in a separate
> sub-thread):
>
> >
> > > LTTng uses ktime to have th
On Fri, Nov 01, 2013 at 01:55:14PM +, Mathieu Desnoyers wrote:
> - Original Message -
> > From: "Mathieu Desnoyers"
> > To: "Vladimir Nikulichev"
> > Cc: "Paul E. McKenney"
> > Sent: Friday, November 1, 2013 9:42:16 AM
>
On Fri, Nov 01, 2013 at 08:18:59PM +, Mathieu Desnoyers wrote:
> - Original Message -
> > From: "Mathieu Desnoyers"
> > To: "Vladimir Nikulichev"
> > Cc: lttng-dev@lists.lttng.org, "Paul E. McKenney"
> >
> > Sent: Frida
On Sun, Nov 03, 2013 at 02:13:52PM +, Mathieu Desnoyers wrote:
> - Original Message -
> > From: "Paul E. McKenney"
> > To: "Mathieu Desnoyers"
> > Cc: "Vladimir Nikulichev" , lttng-dev@lists.lttng.org
> > Sent: Sunday, Novemb
s
> >
> > Hello there,
>
> Hi Simon,
>
> While reading this reply, please keep in mind that I'm in a
> mindset where I've been in a full week of meeting, and it's late on
> Friday evening here. So YMMV ;-) I'm CCing Paul E. McKenney, so he can
> debu
On Sat, Dec 07, 2013 at 05:58:54AM +, Mathieu Desnoyers wrote:
> - Original Message -
> > From: "Paul E. McKenney"
> > To: "Mathieu Desnoyers"
> > Cc: "Simon Marchi" , lttng-dev@lists.lttng.org
> > Sent: Friday, December
On Thu, Mar 12, 2015 at 08:56:00PM +, Mathieu Desnoyers wrote:
> (sorry for re-send, my mail client tricked me into posting HTML
> to lkml)
>
> Hi,
>
> Michael Sullivan proposed a clever hack abusing mprotect() to
> perform the same effect as sys_membarrier() I submitted a few
> years ago
On Fri, Mar 13, 2015 at 09:07:43AM +0100, Ingo Molnar wrote:
>
> * Mathieu Desnoyers wrote:
>
> > - Original Message -
> > > From: "Linus Torvalds"
> > > To: "Mathieu Desnoyers"
> > > Cc: "Michael Sullivan" , lttn
On Fri, Apr 17, 2015 at 12:23:46PM +0300, Eugene Ivanov wrote:
> Hi Mathieu,
>
> On 04/10/2015 11:26 PM, Mathieu Desnoyers wrote:
> >- Original Message -
> >>Hi,
> >>
> >>I use rcu-bp (0.8.6) and get deadlock between call_rcu thread and
> >>threads willing to do rcu_read_lock():
> >>1. Som
On Fri, Apr 17, 2015 at 07:18:18AM -0700, Paul E. McKenney wrote:
> On Fri, Apr 17, 2015 at 12:23:46PM +0300, Eugene Ivanov wrote:
> > Hi Mathieu,
> >
> > On 04/10/2015 11:26 PM, Mathieu Desnoyers wrote:
> > >- Original Message -
> > >>Hi,
egistration/unregistration to complete even though synchronize_rcu is
> awaiting for RCU read-side critical sections to complete.
>
> Changes since v1:
> - Hold both rcu_gp_lock and rcu_registry_lock across fork in urcu-bp.
>
> Signed-off-by: Mathieu Desnoyers
Reviewed-by:
.
>
> Reported-by: Michael Jeanson
> Suggested-by: Paul E. McKenney
> Signed-off-by: Mathieu Desnoyers
A couple of issues, but otherwise good. (They might even be issues
with your code rather than my eyes, you never know!)
Tha
On Mon, Jun 29, 2015 at 11:06:15PM +, Mathieu Desnoyers wrote:
> - On Jun 29, 2015, at 7:01 PM, Paul E. McKenney
> paul...@linux.vnet.ibm.com wrote:
>
> > On Mon, Jun 29, 2015 at 06:56:34PM -0400, Mathieu Desnoyers wrote:
> >> Make call_rcu_thread() affine
ust one.
Looks like a reasonable approach to me.
Acked-by: Paul E. McKenney
> Signed-off-by: Mathieu Desnoyers
> CC: Paul E. McKenney
> CC: Michael Jeanson
> CC: Jon Bernard
> ---
> Makefile.am | 2 --
> urcu/futex.h | 70
> +
On Thu, Jan 21, 2016 at 05:06:09PM +, Mathieu Desnoyers wrote:
> - On Jan 21, 2016, at 11:59 AM, Josh Triplett j...@joshtriplett.org wrote:
>
> > On Thu, Jan 21, 2016 at 04:45:20PM +, Mathieu Desnoyers wrote:
> >> - On Jan 19, 2016, at 3:57 PM, Mathieu Desnoyers
> >> mathieu.desnoy
On Fri, Jan 22, 2016 at 12:22:02PM -0500, Jérémie Galarneau wrote:
> Following a prior discussion on this mailing list [1], I am proposing
> the introduction of rcu_barrier_finalize() in liburcu.
>
> *** Use Case ***
>
> As pointed out in the e-mail thread, some applications are nesting
> liburcu
Hello!
Still working with the C++ standards committee to add RCU to C++, and
of course one step on that path is to make C++ programs build against
liburcu. The following hack-patch makes this work by disabling C++ access
to wfcqueue and by making C++ see the rcu_head ->next field as a void *.
Th
fcq_head and struct __cds_wfcq_head pointers to
> cds_wfcq_head_ptr_t.
>
> Signed-off-by: Mathieu Desnoyers
Good stuff!!!
Reviewed-by: Paul E. McKenney
Tested-by: Paul E. McKenney
> ---
> urcu/static/wfcqueue.h | 16 +---
> urcu/wfcqueue.h| 40 +++
On Mon, Mar 07, 2016 at 07:33:43PM -0500, Mathieu Desnoyers wrote:
> Signed-off-by: Mathieu Desnoyers
Reviewed-by: Paul E. McKenney
Tested-by: Paul E. McKenney
> ---
> urcu/wfcqueue.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/urcu/wfcqueue.h
On Tue, Mar 08, 2016 at 05:20:25PM +, Mathieu Desnoyers wrote:
> - On Mar 8, 2016, at 12:11 PM, Paul E. McKenney
> paul...@linux.vnet.ibm.com wrote:
>
> > On Mon, Mar 07, 2016 at 07:33:44PM -0500, Mathieu Desnoyers wrote:
> >> Introduce __cds_wfcq_head_cast an
know real time system focus on predictability on both timing and
> > memory consumption.
> > So how does real time urcu support predictability?
> > Could you provide me some papers, documents or any materials about any
> > aspect of real time urcu?
>
> Adding Paul E. McKen
this work on -rt Linux.
Thanx, Paul
> Thanks again.
>
> On Fri, Mar 11, 2016 at 10:00 PM, Mathieu Desnoyers
> wrote:
> > ----- On Mar 11, 2016, at 6:45 AM, Paul E. McKenney
> > paul...@linux.vnet.ibm.com wrote:
> >
> >> On Thu, Mar 10, 2016 at 08
tation deal with this problem?
> And if so, how does it solve this?
>
> Thanks
> Yuxin
>
> On Thu, Mar 31, 2016 at 7:22 AM, Paul E. McKenney
> wrote:
> > On Thu, Mar 31, 2016 at 09:20:07AM +0800, Yuxin Ren wrote:
> >> Thank you all!!
> >>
>
On Wed, Apr 27, 2016 at 09:34:16PM -0400, Yuxin Ren wrote:
> Hi,
>
> I am learning the URCU code.
>
> Why do we need rcu_gp_lock in synchronize_rcu?
> https://github.com/urcu/userspace-rcu/blob/master/urcu.c#L401
>
> In the comment, it says this lock ensures mutual exclusion between
> threads ca
n't currently perform grace period, why do we use the rcu_gp_lock?
>
> Thank you.
> Yuxin
>
> On Wed, Apr 27, 2016 at 10:08 PM, Paul E. McKenney
> wrote:
> > On Wed, Apr 27, 2016 at 09:34:16PM -0400, Yuxin Ren wrote:
> >> Hi,
> >>
> >> I am l
On Thu, Apr 28, 2016 at 08:44:01PM +0800, Boqun Feng wrote:
> Hi Paul and Yuxin,
>
> On Wed, Apr 27, 2016 at 09:23:27PM -0700, Paul E. McKenney wrote:
> > Try building without it and see what happens when you run the tests.
> >
>
> I've run a 'regtest
> On Thu, Apr 28, 2016 at 8:44 AM, Boqun Feng wrote:
> > Hi Paul and Yuxin,
> >
> > On Wed, Apr 27, 2016 at 09:23:27PM -0700, Paul E. McKenney wrote:
> >> Try building without it and see what happens when you run the tests.
> >>
> >
> >
> Thanks again!!
> > Yuxin
> >
> > On Thu, Apr 28, 2016 at 8:44 AM, Boqun Feng wrote:
> >> Hi Paul and Yuxin,
> >>
> >> On Wed, Apr 27, 2016 at 09:23:27PM -0700, Paul E. McKenney wrote:
> >>> Try building without it and see what
On Wed, May 18, 2016 at 06:40:03PM +, Mathieu Desnoyers wrote:
> - On May 18, 2016, at 5:44 AM, songxin wrote:
>
> > Hi,
> > Now I get a crash because receiving signal SIGSEGV as below.
>
> > #0 arena_alloc (arena=) at
> > /usr/src/debug/liburcu/0.9.1+git5fd33b1e5003ca316bd314ec3fd1447f
Hello!
On the off chance that this is new news of interest...
https://arxiv.org/pdf/1606.01400v1.pdf
"Operational Aspects of C/C++ Concurrency", Anton Podkopaev, Ilya Sergey,
Aleksandar Nanevski.
At first glance, they seem to be using a combination of formal
verification and testing, using a si
e?
> > Not only RCU, any other synchronization technique on top of
> > non-cache-coherent multi-core
> > is also helpful.
>
> CCing Paul E. McKenney, who might know more on this topic.
>
> Back in 2009 when I started the liburcu.org project, I
> planned to eventuall
On Sat, Sep 24, 2016 at 10:42:24AM +0300, Evgeniy Ivanov wrote:
> Hi Mathieu,
>
> On Sat, Sep 24, 2016 at 12:59 AM, Mathieu Desnoyers
> wrote:
> > - On Sep 22, 2016, at 3:14 PM, Evgeniy Ivanov lolkaanti...@gmail.com
> > wrote:
> >
> >> Hi all,
> >>
> >> I'm investigating high memory usage of
On Sat, Sep 24, 2016 at 03:34:47PM +, Mathieu Desnoyers wrote:
> - On Sep 24, 2016, at 11:22 AM, Paul E. McKenney
> paul...@linux.vnet.ibm.com wrote:
>
> > On Sat, Sep 24, 2016 at 10:42:24AM +0300, Evgeniy Ivanov wrote:
> >> Hi Mathieu,
> >>
>
On Mon, Dec 05, 2016 at 02:14:47PM +, Mathieu Desnoyers wrote:
> Hi Paul,
>
> So about the liburcu rcu_xchg_pointer() barriers, here is the current
> situation:
>
> rcu_xchg_pointer is implemented as:
>
> #define _rcu_xchg_pointer(p, v) \
> __extension__
On Mon, Dec 05, 2016 at 11:01:10PM +, Mathieu Desnoyers wrote:
> - On Dec 5, 2016, at 5:35 PM, Paul E. McKenney paul...@linux.vnet.ibm.com
> wrote:
>
> > On Mon, Dec 05, 2016 at 02:14:47PM +, Mathieu Desnoyers wrote:
> >> Hi Paul,
> >>
> >
On Thu, May 11, 2017 at 04:05:45PM -0400, Yuxin Ren wrote:
> Hi,
>
> I am learning U-RCU now.
> And I read paper Concurrent Updates with RCU: Search Tree as an Example
> (
> https://pdfs.semanticscholar.org/73e4/cd29273cf9d98d35bc184330e694ba798987.pdf
> )
>
> In this paper, the authors present
1 - 100 of 133 matches
Mail list logo