Re: [PATCH 1/2] rwsem: add rwsem_is_contended V2

From: Josef Bacik <jbacik@fusionio.com>
To: Peter Hurley <peter@hurleysoftware.com>
Cc: Josef Bacik <jbacik@fusionio.com>, <linux-btrfs@vger.kernel.org>,
	<walken@google.com>, <linux-kernel@vger.kernel.org>,
	<mingo@elte.hu>, <akpm@linux-foundation.org>
Subject: Re: [PATCH 1/2] rwsem: add rwsem_is_contended V2
Date: Thu, 19 Sep 2013 19:27:32 -0400	[thread overview]
Message-ID: <20130919232732.GC1615@localhost.localdomain> (raw)
In-Reply-To: <523B8157.4050508@hurleysoftware.com>

On Thu, Sep 19, 2013 at 06:57:27PM -0400, Peter Hurley wrote:
> On 09/19/2013 11:48 AM, Josef Bacik wrote:
> >Btrfs needs a simple way to know if it needs to let go of it's read lock on a
> >rwsem.  Introduce rwsem_is_contended to check to see if there are any waiters on
> >this rwsem currently.  This is just a hueristic, it is meant to be light and not
> >100% accurate and called by somebody already holding on to the rwsem in either
> >read or write.  Thanks,
> >
> >Signed-off-by: Josef Bacik <jbacik@fusionio.com>
> >---
> >V1->V2: took everybodys suggestions and simplified it to just one function in
> >rwsem.h so it works for both the spinlock case and non-spinlock case.
> >
> >  include/linux/rwsem.h | 13 +++++++++++++
> >  1 file changed, 13 insertions(+)
> >
> >diff --git a/include/linux/rwsem.h b/include/linux/rwsem.h
> >index 0616ffe..c340493 100644
> >--- a/include/linux/rwsem.h
> >+++ b/include/linux/rwsem.h
> >@@ -75,6 +75,19 @@ do {								\
> >  } while (0)
> >
> >  /*
> >+ * This is the same regardless of which rwsem implementation that is being used.
> >+ * It is just a heuristic meant to be called by somebody alreadying holding the
> >+ * rwsem to see if somebody from  the opposite type is wanting access to the
>                                         ^^^^^^^^^^^^^
> 
> Readers can infer that at least one writer is waiting if the wait_list is
> !empty; however, writers cannot infer anything other than some other
> thread is waiting -- it could be a reader or a writer or multiples of either.
> 

Right duh, I'll fix that up.

> 
> >+ * lock.
> >+ */
> >+static inline int rwsem_is_contended(struct rw_semaphore *sem)
> >+{
> >+	if (!list_empty(&sem->wait_list))
> >+		return 1;
> >+	return 0;
> 
> How about
> 
>         return !list_empty(&sem->wait_list);
> 
> ?
> 

Another duh, thanks I'll wait for any other input and then fix this up and
resend.  Thanks,

Josef