From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
        id S943145AbcJ0Ozt (ORCPT <rfc822;w@1wt.eu>);
        Thu, 27 Oct 2016 10:55:49 -0400
Received: from shards.monkeyblade.net ([184.105.139.130]:53280 "EHLO
        shards.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S943067AbcJ0Ozn (ORCPT
        <rfc822;linux-kernel@vger.kernel.org>);
        Thu, 27 Oct 2016 10:55:43 -0400
Date: Thu, 27 Oct 2016 10:55:39 -0400 (EDT)
Message-Id: <20161027.105539.1624301498966110301.davem@davemloft.net>
To: nitin.m.gupta@oracle.com
Cc: sowmini.varadhan@oracle.com, aaron.young@oracle.com, bp@suse.de,
        sparclinux@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] sparc64: Make SLUB the default allocator
From: David Miller <davem@davemloft.net>
In-Reply-To: <1477520897-47950-1-git-send-email-nitin.m.gupta@oracle.com>
References: <1477520897-47950-1-git-send-email-nitin.m.gupta@oracle.com>
X-Mailer: Mew version 6.7 on Emacs 24.5 / Mule 6.0 (HANACHIRUSATO)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.5.12 (shards.monkeyblade.net [149.20.54.216]); Thu, 27 Oct 2016 07:55:42 -0700 (PDT)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

From: Nitin Gupta <nitin.m.gupta@oracle.com>
Date: Wed, 26 Oct 2016 15:27:56 -0700

> SLUB has better debugging support.
> 
> Signed-off-by: Nitin Gupta <nitin.m.gupta@oracle.com>
 ...
> @@ -7,7 +7,9 @@ CONFIG_LOG_BUF_SHIFT=18
>  CONFIG_BLK_DEV_INITRD=y
>  CONFIG_PERF_EVENTS=y
>  # CONFIG_COMPAT_BRK is not set
> -CONFIG_SLAB=y
> +CONFIG_SLUB_DEBUG=y
> +CONFIG_SLUB=y
> +CONFIG_SLUB_CPU_PARTIAL=y
>  CONFIG_PROFILING=y
>  CONFIG_OPROFILE=m
>  CONFIG_KPROBES=y

This patch made me try to find guidance on other architectures as to what
they put in their defconfig files.

And it seems to me that they leave things like this to the user, and the
defconfig only really contains things that really are required by the
architecture or are architecture specific.

So maybe instead of making arbitrary choices for these values which
one set of users want and another set of users do not (I prefer SLAB,
it is faster, and I know what the downsides wrt. NUMA balancing
compared to SLAB are and am willing to accept that), we should do
what other platforms do and not specify things like this at all.