From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Madhusudhan" Subject: RE: [PATCH-V2] OMAP: Fix for bus width which improves SD card's peformance. Date: Thu, 1 Apr 2010 10:41:52 -0500 Message-ID: <003c01cad1b1$da2cdbf0$544ff780@am.dhcp.ti.com> References: <35697.192.168.10.89.1265898023.squirrel@dbdmail.itg.ti.com> <004701caab40$68b576e0$544ff780@am.dhcp.ti.com> <003b01cad0f0$6ea78040$544ff780@am.dhcp.ti.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from arroyo.ext.ti.com ([192.94.94.40]:39945 "EHLO arroyo.ext.ti.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756850Ab0DAPmA convert rfc822-to-8bit (ORCPT ); Thu, 1 Apr 2010 11:42:00 -0400 In-Reply-To: Sender: linux-omap-owner@vger.kernel.org List-Id: linux-omap@vger.kernel.org To: 'kishore kadiyala' Cc: 'Vimal Singh' , tony@atomide.com, svenkatr@ti.com, linux-omap@vger.kernel.org, linux-kernel@vger.kernel.org, jarkko.lavinen@nokia.com > -----Original Message----- > From: kishore kadiyala [mailto:kishorek.kadiyala@gmail.com] > Sent: Thursday, April 01, 2010 1:32 AM > To: Madhusudhan > Cc: Vimal Singh; tony@atomide.com; svenkatr@ti.com; linux- > omap@vger.kernel.org; linux-kernel@vger.kernel.org; > jarkko.lavinen@nokia.com > Subject: Re: [PATCH-V2] OMAP: Fix for bus width which improves SD car= d's > peformance. >=20 > On Wed, Mar 31, 2010 at 10:07 PM, Madhusudhan wrote= : > > > > > >> -----Original Message----- > >> From: kishore kadiyala [mailto:kishorek.kadiyala@gmail.com] > >> Sent: Wednesday, March 31, 2010 2:03 AM > >> To: Vimal Singh > >> Cc: Madhusudhan; tony@atomide.com; svenkatr@ti.com; linux- > >> omap@vger.kernel.org; linux-kernel@vger.kernel.org; > >> jarkko.lavinen@nokia.com > >> Subject: Re: [PATCH-V2] OMAP: Fix for bus width which improves SD > card's > >> peformance. > >> > >> Sorry for that and here's the Updated one. > >> > >> From: Kishore Kadiyala > >> > >> This patch improves low speeds for SD cards. > >> OMAP-MMC controller's can support maximum bus width of '8'. > >> when bus width is mentioned as "8" in controller data,the SD > >> stack will check whether bus width is "4" and if not it will > >> set bus width to "1" and there by degrading performance. > >> This patch fixes the issue and improves the performance of > >> SD cards. > >> > >> Signed-off-by: Kishore Kadiyala > >> Signed-off-by: Venkatraman S > >> Acked-by: Madhusudhan Chikkature > >> > >> --- > >> In V2 : Appended Signed-off by Venkat and Ack by Madhu > >> > >> =A0Here are my experiment numbers, on a Class 6 SDHC card: > >> =A0Read peformance is increased by 220% > >> =A0Write Performance is increased by 52% > >> > >> =A0drivers/mmc/host/omap_hsmmc.c | =A0 =A02 +- > >> =A01 files changed, 1 insertions(+), 1 deletions(-) > >> > >> diff --git a/drivers/mmc/host/omap_hsmmc.c > b/drivers/mmc/host/omap_hsmmc.c > >> index 83f0aff..8c97c22 100644 > >> --- a/drivers/mmc/host/omap_hsmmc.c > >> +++ b/drivers/mmc/host/omap_hsmmc.c > >> @@ -2092,7 +2092,7 @@ static int __init omap_hsmmc_probe(struct > >> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0MMC_CAP_WAIT_WHILE_BUSY; > >> > >> =A0 =A0 =A0 if (mmc_slot(host).wires >=3D 8) > >> - =A0 =A0 =A0 =A0 =A0 =A0 mmc->caps |=3D MMC_CAP_8_BIT_DATA; > >> + =A0 =A0 =A0 =A0 =A0 =A0 mmc->caps |=3D (MMC_CAP_8_BIT_DATA | MMC= _CAP_4_BIT_DATA); > >> =A0 =A0 =A0 else if (mmc_slot(host).wires >=3D 4) > >> =A0 =A0 =A0 =A0 =A0 =A0 =A0 mmc->caps |=3D MMC_CAP_4_BIT_DATA; > >> > > Kishore, > > > > Since this patch is not yet pushed it makes sense to fix the readab= ility > > issue. > > > > Since 8-bit is the max how about: > > > > =A0=A0 =A0 =A0 =A0if (mmc_slot(host).wires =3D=3D 8) > > =A0=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0mmc->caps |=3D MMC_CAP_8_BIT_DATA= | MMC_CAP_4_BIT_DATA; > > =A0=A0 =A0 =A0 =A0if (mmc_slot(host).wires =3D=3D 4) > > =A0=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0mmc->caps |=3D MMC_CAP_4_BIT_DATA= ; > > > Madhu, >=20 > In the above snippet, it checks whether wires are 8 or 4 and if not > neither set's capability to "1". > Does it make sense to check whether the wires are 8,4,1 and if not > any[8,4,1] throw error and come out. >=20 It is good enough to just check for 8-bit and 4-bit. The 1-bit does not= need any explicit check since it is default. Regards, Madhu > Regards, > Kishore > > This would be little easy to read the code. > > > > Can you please repost the patch?? > > > > Regards, > > Madhu > > > >> -- > >> 1.6.3.3 > > > > -- To unsubscribe from this list: send the line "unsubscribe linux-omap" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html