From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dmitry Torokhov Subject: Re: [PATCH 2/9] input: goodix: fix variable length array warning Date: Fri, 5 Jun 2015 10:11:38 -0700 Message-ID: <20150605171138.GF26708@dtor-ws> References: <1432817265-23891-1-git-send-email-irina.tirdea@intel.com> <1432817265-23891-3-git-send-email-irina.tirdea@intel.com> <20150528175753.1f85723b5b880ecf02fd4a79@ao2.it> <1F3AC3675D538145B1661F571FE1805F19A78DD0@irsmsx105.ger.corp.intel.com> <20150603224953.6613b674931390f187b81e7a@ao2.it> <1F3AC3675D538145B1661F571FE1805F19A79D7B@irsmsx105.ger.corp.intel.com> <20150605164053.GA26708@dtor-ws> <1F3AC3675D538145B1661F571FE1805F19A79E7E@irsmsx105.ger.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: <1F3AC3675D538145B1661F571FE1805F19A79E7E@irsmsx105.ger.corp.intel.com> Sender: linux-input-owner@vger.kernel.org To: "Tirdea, Irina" Cc: 'Antonio Ospite' , Bastien Nocera , "linux-input@vger.kernel.org" , "devicetree@vger.kernel.org" , "linux-kernel@vger.kernel.org" List-Id: devicetree@vger.kernel.org On Fri, Jun 05, 2015 at 05:00:05PM +0000, Tirdea, Irina wrote: >=20 >=20 > > -----Original Message----- > > From: Dmitry Torokhov [mailto:dmitry.torokhov@gmail.com] > > Sent: 05 June, 2015 19:41 > > To: Tirdea, Irina > > Cc: 'Antonio Ospite'; Bastien Nocera; linux-input@vger.kernel.org; = devicetree@vger.kernel.org; linux-kernel@vger.kernel.org > > Subject: Re: [PATCH 2/9] input: goodix: fix variable length array w= arning > >=20 > > On Fri, Jun 05, 2015 at 04:34:38PM +0000, Tirdea, Irina wrote: > > > > > > > > > > -----Original Message----- > > > > From: linux-input-owner@vger.kernel.org [mailto:linux-input-own= er@vger.kernel.org] On Behalf Of Antonio Ospite > > > > Sent: 03 June, 2015 23:50 > > > > To: Tirdea, Irina > > > > Cc: Dmitry Torokhov; Bastien Nocera; linux-input@vger.kernel.or= g; devicetree@vger.kernel.org; linux-kernel@vger.kernel.org > > > > Subject: Re: [PATCH 2/9] input: goodix: fix variable length arr= ay warning > > > > > > > > On Wed, 3 Jun 2015 10:26:47 +0000 > > > > "Tirdea, Irina" wrote: > > > > > > > > > > -----Original Message----- > > > > > > From: Antonio Ospite [mailto:ao2@ao2.it] > > > > > > Sent: 28 May, 2015 18:58 > > > > > > To: Tirdea, Irina > > > > > > Cc: Dmitry Torokhov; Bastien Nocera; linux-input@vger.kerne= l.org; devicetree@vger.kernel.org; linux-kernel@vger.kernel.org > > > > > > Subject: Re: [PATCH 2/9] input: goodix: fix variable length= array warning > > > > > > > > > > > > On Thu, 28 May 2015 15:47:38 +0300 > > > > > > Irina Tirdea wrote: > > > > > > > > > > > > > Fix sparse warning: > > > > > > > drivers/input/touchscreen/goodix.c:182:26: warning: > > > > > > > Variable length array is used. > > > > > > > > > > > > > > Replace the variable length array with fixed length. > > > > > > > > > > > > > > Signed-off-by: Irina Tirdea > > > > > > > --- > > > > > > > drivers/input/touchscreen/goodix.c | 2 +- > > > > > > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > > > > > > > > > > > diff --git a/drivers/input/touchscreen/goodix.c b/drivers= /input/touchscreen/goodix.c > > > > > > > index c2e785c..dac1b3c 100644 > > > > > > > --- a/drivers/input/touchscreen/goodix.c > > > > > > > +++ b/drivers/input/touchscreen/goodix.c > > > > > > > @@ -147,7 +147,7 @@ static void goodix_ts_report_touch(st= ruct goodix_ts_data *ts, u8 *coor_data) > > > > > > > */ > > > > > > > static void goodix_process_events(struct goodix_ts_data = *ts) > > > > > > > { > > > > > > > - u8 point_data[1 + GOODIX_CONTACT_SIZE * ts->max_touch_= num]; > > > > > > > + u8 point_data[1 + GOODIX_CONTACT_SIZE * GOODIX_MAX_CON= TACTS]; > > > > > > > > > > > > Hi, > > > > > > > > > > > > > > > > Hi Antonio, > > > > > > > > > > > this fixes the warning from sparse, but also changes the se= mantics of > > > > > > the code: ts->max_touch_num is less that GOODIX_MAX_CONTACT= S for 5 > > > > > > touches devices and in this case we'll end up using more me= mory than is > > > > > > necessary. > > > > > > > > > > > > > > > > I wasn't sure if it is better to save the 5 bytes or fix the = warning, > > > > > so I sent this to get some more input. > > > > > Thanks for the feedback, I will drop this patch. > > > > > > > > > > > > > Use kmalloc() or, alternatively, add at least a comment telling= why you > > > > think that sacrificing a few bytes =E2=80=94only for some devic= es=E2=80=94 has > > > > advantages over dynamic allocation. > > > > > > > > > > You are right, kmalloc will solve both problems - the sparse warn= ing and allocating > > > more bytes than necessary. Don't know why I did not think of that= =2E > > > Will use that in v2. > >=20 > > Please leave the patch as is. We can spare 80 bytes on the stack gi= ven > > that we are running in threaded IRQ. kmallocing will use more code = and > > runtime resources and won't give any benefits. >=20 > I was actually thinking of allocating it with devm_kzalloc just once = at device init and > storing a pointer to it in the goodix_ts_data structure.=20 Still, why? Even if you do this once you will need both code and runtim= e resources to bookkeeping whereas allocating 80 bytes on stack are just = a matter of register addition. And because we are running in threaded IRQ context there is no concern of blowing up the limited stack. The story would be different if you needed, let's say 800 or 1600 bytes= =2E Thanks. --=20 Dmitry -- To unsubscribe from this list: send the line "unsubscribe linux-input" = in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html