From: Andrew Morton Subject: Re: [PATCH 1/4] quota: Quota core changes for quotaon on remount Date: Thu, 13 Mar 2008 14:48:02 -0700 Message-ID: <20080313144802.7397a048.akpm@linux-foundation.org> References: <20080313125944.GF12523@duck.suse.cz> <20080313130412.GG12523@duck.suse.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: linux-kernel@vger.kernel.org, linux-ext4@vger.kernel.org, linux-fsdevel@vger.kernel.org, jengelh@computergmbh.de, mjt@tls.msk.ru To: Jan Kara Return-path: In-Reply-To: <20080313130412.GG12523@duck.suse.cz> Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-ext4.vger.kernel.org On Thu, 13 Mar 2008 14:04:12 +0100 Jan Kara wrote: > +static inline int DQUOT_ON_REMOUNT(struct super_block *sb) > +{ > + int cnt; > + int ret = 0, err; > + > + if (!sb->s_qcop || !sb->s_qcop->quota_on) > + return -ENOSYS; > + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { > + err = sb->s_qcop->quota_on(sb, cnt, 0, NULL, 1); > + if (err < 0 && !ret) > + ret = err; > + } > + return ret; > +} > + That's about ten times too large to be inlined. Oh well. Sometime could you please take a look at uninlining some of the porkier functions in quotaops.h, and maybe convert them to lower case? For the latter we can do #define DQUOT_ON_REMOUNT(sb) dquot_on_remount(sb) etc, then remove those macros when everything has been converted and we've given any out-of-tree filesytem developers time to update, etc. Thanks.