From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933529AbbBIJGQ (ORCPT ); Mon, 9 Feb 2015 04:06:16 -0500 Received: from down.free-electrons.com ([37.187.137.238]:50616 "EHLO mail.free-electrons.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S932781AbbBIJGI (ORCPT ); Mon, 9 Feb 2015 04:06:08 -0500 Date: Mon, 9 Feb 2015 10:03:38 +0100 From: Maxime Ripard To: Thomas =?iso-8859-1?Q?Niederpr=FCm?= Cc: linux-fbdev@vger.kernel.org, plagnioj@jcrosoft.com, tomi.valkeinen@ti.com, linux-kernel@vger.kernel.org Subject: Re: [PATCH 6/8] fbdev: ssd1307fb: Add module parameter to set update delay of the deffered io. Message-ID: <20150209090338.GY2079@lukather> References: <1423261694-5939-1-git-send-email-niederp@physik.uni-kl.de> <1423261694-5939-7-git-send-email-niederp@physik.uni-kl.de> <20150207112627.GP2079@lukather> <20150207171211.773dc996@maestro.intranet> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="8a48p4KKsB+A/3VY" Content-Disposition: inline In-Reply-To: <20150207171211.773dc996@maestro.intranet> User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --8a48p4KKsB+A/3VY Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sat, Feb 07, 2015 at 05:12:11PM +0100, Thomas Niederpr=FCm wrote: > Am Sat, 7 Feb 2015 12:26:27 +0100 > schrieb Maxime Ripard : >=20 > > On Fri, Feb 06, 2015 at 11:28:12PM +0100, niederp@physik.uni-kl.de > > wrote: > > > From: Thomas Niederpr=FCm > > >=20 > > > This patch adds the module parameter "delaydivider" to set delay > > > for the deferred io. Effectively this is setting the refresh rate > > > for mmap access to the framebuffer. The delay for the deferred io > > > is HZ/delaydivider. > >=20 > > So this is actually a refresh rate? > >=20 > > Maybe you could expose it as such, and pass a frequency in Hz as an > > argument. >=20 > Good idea! I'll try to do it that way. >=20 > >=20 > > Exposing the divider directly has some issues, since the bootloader > > that set the parameter won't know the HZ value, you'll end up with > > different rates for different configurations, without any way to do > > something about it. > >=20 > > >=20 > > > Signed-off-by: Thomas Niederpr=FCm > > > --- > > > drivers/video/fbdev/ssd1307fb.c | 7 ++++++- > > > 1 file changed, 6 insertions(+), 1 deletion(-) > > >=20 > > > diff --git a/drivers/video/fbdev/ssd1307fb.c > > > b/drivers/video/fbdev/ssd1307fb.c index 1d81877..b38315d 100644 > > > --- a/drivers/video/fbdev/ssd1307fb.c > > > +++ b/drivers/video/fbdev/ssd1307fb.c > > > @@ -44,10 +44,14 @@ > > > #define SSD1307FB_SET_VCOMH 0xdb > > > =20 > > > #define BITSPERPIXEL 1 > > > +#define DELAYDIVIDER 20 > > > =20 > > > static u_int bitsperpixel =3D BITSPERPIXEL; > > > module_param(bitsperpixel, uint, 0); > > > =20 > > > +static u_int delaydivider =3D DELAYDIVIDER; > > > +module_param(delaydivider, uint, 0); > > > + > >=20 > > You're breaking the existing behaviour. >=20 > True! I'll try to keep the existing behaviour when rewriting this to > use the refresh rate. >=20 > >=20 > > > struct ssd1307fb_par; > > > =20 > > > struct ssd1307fb_deviceinfo { > > > @@ -312,7 +316,7 @@ static void ssd1307fb_deferred_io(struct > > > fb_info *info, } > > > =20 > > > static struct fb_deferred_io ssd1307fb_defio =3D { > > > - .delay =3D HZ, > > > + .delay =3D HZ/DELAYDIVIDER, > > > .deferred_io =3D ssd1307fb_deferred_io, > > > }; > > > =20 > > > @@ -601,6 +605,7 @@ static int ssd1307fb_probe(struct i2c_client > > > *client, info->fix =3D ssd1307fb_fix; > > > info->fix.line_length =3D par->width * bitsperpixel / 8; > > > info->fbdefio =3D &ssd1307fb_defio; > > > + info->fbdefio->delay =3D HZ/delaydivider; > >=20 > > That won't work with multiple instances of the same driver > > unfortunately. >=20 > Could you please elaborate why? I'm not seeing it... On a general basis, because the structure is shared by all the instances of the driver, so it's usually not such a good idea to mix the static declaration of the structure and the dynamic one. =46rom a more fundamental point of view, because this parameter will most likely be different from one instance to another? Maxime --=20 Maxime Ripard, Free Electrons Embedded Linux, Kernel and Android engineering http://free-electrons.com --8a48p4KKsB+A/3VY Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQIcBAEBAgAGBQJU2HfqAAoJEBx+YmzsjxAglIQP/jxmGs/rvwLaqTwr9D7gqxpJ LRgWPyFz2bOiRcWWYwgtMjQQWiSgwQcpAX5xW0GpsWrRSoChXJWOm512OL08ODv8 Q7mvhl02wZatXUm3eKOaRgD95ZlC3RTABIOnJ4hYv7Ip1DtD4q0APRpSXa7uyxJX mD/suUTCREuadUpHc42oxUT5EaHaAMTAvzg28o6+/v64FMA1nfDF54DS9YakOpce DJnt5E7FGkklNord46CV0RlsPjKj/AqOU0QSZkzia8nzfDFB5UN87Igr4UTi4nsP lVk9fK9s5yo9Uc7vlzJWbA9X/SnmwsG4dc/sw7geBsU3QxU7FGvLuBOrjCPhkSMC at+8+MeefwlYOmJ8OrFwn+mQ7DolgWX0gYkBCl+1Ts1DLd08AU97yGdOiuL3Y3zs vgXGrZcLgysdAuYgAa1b/tOGuQAM6/ho5OdVPufU9WW3PDZAbCdfLNjbMewvvaDm TRTf+0spWVV/ejs293obIQZZs1tNWtCHcOfIH3Z+vI5w+IGJpUzzjFA5oT7rpOeM 5GbB3ANS1ajL3JhxZYzIW0tzEC9dWoq9zMuVsvCHvtxYMCvcSHRPTPc8Zz0VSnbo Bgn3Dl3M/8x7xpYb9pIM60/d8a+vDf1PWZFMDbOryeSkSRNorWkviJeIDwFkWZ6S y/DN6YT5PSgubMryilsf =RBDe -----END PGP SIGNATURE----- --8a48p4KKsB+A/3VY--