From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932623Ab0KLQ3a (ORCPT ); Fri, 12 Nov 2010 11:29:30 -0500 Received: from xenotime.net ([72.52.115.56]:58491 "HELO xenotime.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S932370Ab0KLQ33 convert rfc822-to-8bit (ORCPT ); Fri, 12 Nov 2010 11:29:29 -0500 Date: Fri, 12 Nov 2010 08:29:25 -0800 From: Randy Dunlap To: Belisko Marek Cc: Greg Kroah-Hartman , devel@driverdev.osuosl.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 1/3] staging: ft1000: Fix compilation warning. Message-Id: <20101112082925.1550229e.rdunlap@xenotime.net> In-Reply-To: References: <1289384128-19823-1-git-send-email-marek.belisko@gmail.com> <20101110083814.55e222c4.rdunlap@xenotime.net> Organization: YPO4 X-Mailer: Sylpheed 2.7.1 (GTK+ 2.16.6; x86_64-unknown-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 12 Nov 2010 10:57:19 +0100 Belisko Marek wrote: > Hi Randy, > > On Wed, Nov 10, 2010 at 5:38 PM, Randy Dunlap wrote: > > On Wed, 10 Nov 2010 11:15:26 +0100 Marek Belisko wrote: > > > >> Signed-off-by: Marek Belisko > >> --- > >>  drivers/staging/ft1000/ft1000-usb/ft1000_usb.c |    2 +- > >>  1 files changed, 1 insertions(+), 1 deletions(-) > > > > > > a.  always include the warning or error message that the patch fixes > > > > b.  (quoting from another email yesterday:) > > > > so in your analysis of this compiler warning, was the warning correct & justified, > > or was it false?  I.e., is the init to NULL needed? > After code analysis NULL assignment is necessary. Function have > multiple goto statements > and pointer is assigned after some goto are placed. In one of > statements at the end of function > there is kthread stopping which use pft1000info pointer (this part > isn't applicable if pft1000info isn't initialized). > Hopefully this is what compiler complain about (could return with some > goto statement but pft1000info isn't initialized but > used). This warning was introduced when make common return point for > some error states. Thanks for checking and reporting on that. Sorry, but I disagree. Greg can do whatever he wants to do with the patch. > > If it was false, could we just silence the warning by using: > > > >        struct ft1000_info *unitialized_var(pft1000info); > > > > plus #include ? > > > > > >> diff --git a/drivers/staging/ft1000/ft1000-usb/ft1000_usb.c b/drivers/staging/ft1000/ft1000-usb/ft1000_usb.c > >> index 99e3339..b7c4602 100644 > >> --- a/drivers/staging/ft1000/ft1000-usb/ft1000_usb.c > >> +++ b/drivers/staging/ft1000/ft1000-usb/ft1000_usb.c > >> @@ -64,7 +64,7 @@ static int ft1000_probe(struct usb_interface *interface, > >>       int i, ret = 0, size; > >> > >>       struct ft1000_device *ft1000dev; > >> -     struct ft1000_info *pft1000info; > >> +     struct ft1000_info *pft1000info = NULL; > >>       const struct firmware *dsp_fw; > >> > >>       ft1000dev = kmalloc(sizeof(struct ft1000_device), GFP_KERNEL); > >> -- > > > > > > --- > > thanks, > > marek --- ~Randy *** Remember to use Documentation/SubmitChecklist when testing your code ***