On 06/05, Al Viro wrote:
>
> On Fri, Jun 05, 2015 at 11:08:57PM +0200, Oleg Nesterov wrote:
> > On 06/05, Al Viro wrote:
> > >
> > > FWIW, I hadn't really looked into stop_machine uses, but fs/locks.c one
> > > is really not all that great - there we have a large trashcan of a list
> > > (every fil
On Fri, Jun 05, 2015 at 11:08:57PM +0200, Oleg Nesterov wrote:
> On 06/05, Al Viro wrote:
> >
> > FWIW, I hadn't really looked into stop_machine uses, but fs/locks.c one
> > is really not all that great - there we have a large trashcan of a list
> > (every file_lock on the system) and the only use
On 06/05, Al Viro wrote:
>
> FWIW, I hadn't really looked into stop_machine uses, but fs/locks.c one
> is really not all that great - there we have a large trashcan of a list
> (every file_lock on the system) and the only use of that list is /proc/locks
> output generation. Sure, additions take th
On Tue, May 26, 2015 at 02:57:53PM -0700, Linus Torvalds wrote:
> Because that is another example of a complete failure of a locking
> primitive that was just too specialized to be worth it.
FWIW, I hadn't really looked into stop_machine uses, but fs/locks.c one
is really not all that great - t
On Tue, 26 May 2015, Linus Torvalds wrote:
> On Tue, May 26, 2015 at 11:42 AM, Davidlohr Bueso wrote:
> >
> > Instead of dropping percpu-rwsem, I was thinking we could instead look
> > for opportunities to convert new users, for instance shinkers, where the
> > write lock is also taken just for r
On Tue, May 26, 2015 at 11:42:11AM -0700, Davidlohr Bueso wrote:
> Instead of dropping percpu-rwsem, I was thinking we could instead look
> for opportunities to convert new users, for instance shinkers, where the
> write lock is also taken just for register and unregister purposes,
> similar to up
On Tue, May 26, 2015 at 11:42 AM, Davidlohr Bueso wrote:
>
> Instead of dropping percpu-rwsem, I was thinking we could instead look
> for opportunities to convert new users, for instance shinkers, where the
> write lock is also taken just for register and unregister purposes,
> similar to uprobes.
While at it, I'll go ahead and add this triviality.
From: Davidlohr Bueso
Date: Tue, 26 May 2015 12:51:15 -0700
Subject: [PATCH] doc,locking: Move percpu-rwsem doc into proper subdir
Signed-off-by: Davidlohr Bueso
---
Documentation/locking/percpu-rw-semaphore.txt | 27 +
Sorry for noise, forgot to mention...
On 05/26, Oleg Nesterov wrote:
>
> And it can have more users.
I can be wrong and I am obviously biased... but it seems to me that
even rcu_sync alone can have more users in future. This interface
looks natural and simple.
Oleg.
--
To unsubscribe from this
On 05/26, Oleg Nesterov wrote:
>
> On 05/26, Linus Torvalds wrote:
> >
> >
> > We literally have one single percpu-rwsem IN THE WHOLE KERNEL TREE.
> >
> > One.
>
> Well. IIRC Tejun is going to turn signal_struct->group_rwsem into
> percpu-rwsem.
>
> And it can have more users. Say, __sb_start_write
On 05/26, Linus Torvalds wrote:
>
> On Tue, May 26, 2015 at 4:43 AM, Peter Zijlstra wrote:
> >
> > This is a derived work of the cpu hotplug lock rework I did in 2013 which
> > never
> > really went anywhere because Linus didn't like it.
> >
> > This applies those same optimizations to the percpu
On Tue, 2015-05-26 at 11:12 -0700, Linus Torvalds wrote:
> On Tue, May 26, 2015 at 4:43 AM, Peter Zijlstra wrote:
> >
> > This is a derived work of the cpu hotplug lock rework I did in 2013 which
> > never
> > really went anywhere because Linus didn't like it.
> >
> > This applies those same opti
On Tue, May 26, 2015 at 11:12:04AM -0700, Linus Torvalds wrote:
> Is there some new use that I don't know about?
TJ wants to use it in cgroups.
lkml.kernel.org/r/1431549318-16756-3-git-send-email...@kernel.org
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the bod
Hello, Linus.
On Tue, May 26, 2015 at 11:12:04AM -0700, Linus Torvalds wrote:
...
> Is there some new use that I don't know about? Have people really
> looked at that uprobes code deeply? OF COURSE global locks will have
> problems, I'm not at all convinced that "let's make that global lock
> real
On Tue, May 26, 2015 at 4:43 AM, Peter Zijlstra wrote:
>
> This is a derived work of the cpu hotplug lock rework I did in 2013 which
> never
> really went anywhere because Linus didn't like it.
>
> This applies those same optimizations to the percpu-rwsem. Seeing how we did
> all the work it seem
Hi all,
This is a derived work of the cpu hotplug lock rework I did in 2013 which never
really went anywhere because Linus didn't like it.
This applies those same optimizations to the percpu-rwsem. Seeing how we did
all the work it seemed a waste to not use it at all.
--
To unsubscribe from th
16 matches
Mail list logo
