From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-pf0-f197.google.com (mail-pf0-f197.google.com [209.85.192.197]) by kanga.kvack.org (Postfix) with ESMTP id 630EE6B0038 for ; Mon, 17 Oct 2016 00:00:13 -0400 (EDT) Received: by mail-pf0-f197.google.com with SMTP id l29so90422736pfg.7 for ; Sun, 16 Oct 2016 21:00:13 -0700 (PDT) Received: from mail-pa0-x243.google.com (mail-pa0-x243.google.com. [2607:f8b0:400e:c03::243]) by mx.google.com with ESMTPS id ro4si24056390pab.36.2016.10.16.21.00.12 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Sun, 16 Oct 2016 21:00:12 -0700 (PDT) Received: by mail-pa0-x243.google.com with SMTP id os4so4655552pac.3 for ; Sun, 16 Oct 2016 21:00:12 -0700 (PDT) Date: Mon, 17 Oct 2016 15:00:05 +1100 From: Nicholas Piggin Subject: Re: [PATCH v2] mm: vmalloc: Replace purge_lock spinlock with atomic refcount Message-ID: <20161017150005.4c8f890d@roar.ozlabs.ibm.com> In-Reply-To: <1476528162-21981-1-git-send-email-joelaf@google.com> References: <1476528162-21981-1-git-send-email-joelaf@google.com> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org List-ID: To: Joel Fernandes Cc: linux-kernel@vger.kernel.org, linux-rt-users@vger.kernel.org, Chris Wilson , Jisheng Zhang , John Dias , Andrew Morton , "open list:MEMORY MANAGEMENT" On Sat, 15 Oct 2016 03:42:42 -0700 Joel Fernandes wrote: > The purge_lock spinlock causes high latencies with non RT kernel. This has been > reported multiple times on lkml [1] [2] and affects applications like audio. > > In this patch, I replace the spinlock with an atomic refcount so that > preemption is kept turned on during purge. This Ok to do since [3] builds the > lazy free list in advance and atomically retrieves the list so any instance of > purge will have its own list it is purging. Since the individual vmap area > frees are themselves protected by a lock, this is Ok. This is a good idea, and good results, but that's not what the spinlock was for -- it was for enforcing the sync semantics. Going this route, you'll have to audit callers to expect changed behavior and change documentation of sync parameter. I suspect a better approach would be to instead use a mutex for this, and require that all sync=1 callers be able to sleep. I would say that most probably already can. Thanks, Nick -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org