From mboxrd@z Thu Jan 1 00:00:00 1970 From: bfields-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org (J. Bruce Fields) Subject: Re: [PATCH] sunrpc: eliminate RPC_DEBUG Date: Mon, 6 Apr 2015 14:18:58 -0400 Message-ID: <20150406181858.GB3363@fieldses.org> References: <1428327960-22213-1-git-send-email-msalter@redhat.com> <20150406100452.402754a6@tlielax.poochiereds.net> <4110847.o7VvYRDU5Z@wuerfel> <20150406104536.63b01b09@synchrony.poochiereds.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20150406104536.63b01b09-08S845evdOaAjSkqwZiSMmfYqLom42DlXqFh9Ls21Oc@public.gmane.org> Sender: linux-api-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Jeff Layton Cc: Arnd Bergmann , Mark Salter , Trond Myklebust , Anna Schumaker , linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-api@vger.kernel.org On Mon, Apr 06, 2015 at 10:45:36AM -0400, Jeff Layton wrote: > On Mon, 06 Apr 2015 16:38:09 +0200 > Arnd Bergmann wrote: > > > On Monday 06 April 2015 10:04:52 Jeff Layton wrote: > > > > diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c > > > > index 665ef5a..a563ddb 100644 > > > > --- a/fs/lockd/svcsubs.c > > > > +++ b/fs/lockd/svcsubs.c > > > > @@ -31,7 +31,7 @@ > > > > static struct hlist_head nlm_files[FILE_NRHASH]; > > > > static DEFINE_MUTEX(nlm_file_mutex); > > > > > > > > -#ifdef NFSD_DEBUG > > > > +#ifdef CONFIG_SUNRPC_DEBUG > > > > > > Technically, you should use #if IS_ENABLED(CONFIG_SUNRPC_DEBUG). That's > > > supposed to help the compiler do checking of the code inside the block > > > even when it's not defined. > > > > You probably meant the right thing and wrote something else: you should use > > 'if (IS_ENABLED(CONFIG_SUNRPC_DEBUG)) { ... }' to get that effect, but that > > only works within a function. '#if IS_ENABLED(CONFIG_SUNRPC_DEBUG)' is just > > a nicer way to write '#if defined(CONFIG_SUNRPC_DEBUG) || > > defined(CONFIG_SUNRPC_DEBUG_MODULE)', which is harmless but pointless here > > as the symbol would never be set to 'm'. > > > > Arnd > > Ahh ok, good to know. Then this patch is fine as-is then. Thanks, queueing up for 4.1 and stable. --b.