[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <5543CFE5.1030509@redhat.com>
Date: Fri, 01 May 2015 15:11:33 -0400
From: Rik van Riel <riel@...hat.com>
To: Ingo Molnar <mingo@...nel.org>
CC: Andy Lutomirski <luto@...capital.net>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
X86 ML <x86@...nel.org>, williams@...hat.com,
Andrew Lutomirski <luto@...nel.org>, fweisbec@...hat.com,
Peter Zijlstra <peterz@...radead.org>,
Heiko Carstens <heiko.carstens@...ibm.com>,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>,
Paolo Bonzini <pbonzini@...hat.com>,
"Paul E. McKenney" <paulmck@...ibm.com>,
Linus Torvalds <torvalds@...ux-foundation.org>
Subject: Re: [PATCH 3/3] context_tracking,x86: remove extraneous irq disable
& enable from context tracking on syscall entry
On 05/01/2015 02:40 PM, Ingo Molnar wrote:
> Or we could do that in the syscall path with a single store of a
> constant flag to a location in the task struct. We have a number of
> natural flags that get written on syscall entry, such as:
>
> pushq_cfi $__USER_DS /* pt_regs->ss */
>
> That goes to a constant location on the kernel stack. On return from
> system calls we could write 0 to that location.
>
> So the remote CPU would have to do a read of this location. There are
> two cases:
>
> - If it's 0, then it has observed quiescent state on that CPU. (It
> does not have to be atomics anymore, as we'd only observe the value
> and MESI coherency takes care of it.)
That should do the trick.
> - If it's not 0 then the remote CPU is not executing user-space code
> and we can install (remotely) a TIF_NOHZ flag in it and expect it
> to process it either on return to user-space or on a context
> switch.
I may have to think about this a little more, but
it seems like it should work.
Can we use a separate byte in the flags word for
flags that can get set remotely, so we can do
stores and clearing of local-only flags without
atomic instructions?
> This way, unless I'm missing something, reduces the overhead to a
> single store to a hot cacheline on return-to-userspace - which
> instruction if we place it well might as well be close to zero cost.
> No syscall entry cost. Slow-return cost only in the (rare) case of
> someone using synchronize_rcu().
I think that should take care of the RCU aspect of
nohz_full.
--
All rights reversed
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists