[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20140703073148.GT19379@twins.programming.kicks-ass.net>
Date: Thu, 3 Jul 2014 09:31:48 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Jason Low <jason.low2@...com>
Cc: torvalds@...ux-foundation.org, paulmck@...ux.vnet.ibm.com,
mingo@...nel.org, Waiman.Long@...com, davidlohr@...com,
linux-kernel@...r.kernel.org, riel@...hat.com,
akpm@...ux-foundation.org, hpa@...or.com, andi@...stfloor.org,
James.Bottomley@...senpartnership.com, rostedt@...dmis.org,
tim.c.chen@...ux.intel.com, aswin@...com, scott.norton@...com,
chegu_vinod@...com
Subject: Re: [RFC] Cancellable MCS spinlock rework
On Wed, Jul 02, 2014 at 09:39:18PM -0700, Jason Low wrote:
> And due to padding, the additional modification below reduces the
> size of struct rw_semaphore by 64 bits on my machine :)
>
>
> struct rw_semaphore {
> long count;
> - raw_spinlock_t wait_lock;
> struct list_head wait_list;
> + raw_spinlock_t wait_lock;
> #ifdef CONFIG_SMP
> + struct optimistic_spin_tail osq; /* spinner MCS lock */
> /*
> * Write owner. Used as a speculative check to see
> * if the owner is running on the cpu.
> */
> struct task_struct *owner;
> - struct optimistic_spin_tail osq; /* spinner MCS lock */
> #endif
> #ifdef CONFIG_DEBUG_LOCK_ALLOC
> struct lockdep_map dep_map;
>
Right, that might make sense.
Content of type "application/pgp-signature" skipped
Powered by blists - more mailing lists