From mboxrd@z Thu Jan 1 00:00:00 1970 From: kishore kadiyala Subject: [PATCH v4] OMAP: Fix for bus width which improves SD card's peformance. Date: Mon, 19 Apr 2010 21:21:53 +0530 Message-ID: References: <003b01cad0f0$6ea78040$544ff780@am.dhcp.ti.com> <20100406163211.GA29117@nokia.com> <4BBB6767.7010202@ti.com> <20100406165720.GA17916@nokia.com> <00b401cad5e0$1d1868d0$544ff780@am.dhcp.ti.com> <4BBBC628.9030207@ti.com> <00c901cad5e7$9de54350$544ff780@am.dhcp.ti.com> <4BBBD87C.1040307@ti.com> <003601cad73c$98c3aa50$544ff780@am.dhcp.ti.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Return-path: Received: from fg-out-1718.google.com ([72.14.220.156]:47578 "EHLO fg-out-1718.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751729Ab0DSPwP (ORCPT ); Mon, 19 Apr 2010 11:52:15 -0400 In-Reply-To: Sender: linux-omap-owner@vger.kernel.org List-Id: linux-omap@vger.kernel.org To: Madhusudhan Cc: Nishanth Menon , felipe.balbi@nokia.com, me@felipebalbi.com, Vimal Singh , tony@atomide.com, "S, Venkatraman" , linux-omap@vger.kernel.org, linux-kernel@vger.kernel.org, "Lavinen Jarkko (Nokia-D/Helsinki)" The previous patch was Line wrapped , resending correct patch. NM, Sorry I miss spelled your name correcting this time. Regards, Kishore 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 Signed-off-by: Nishanth Menon Acked-by: Madhusudhan Chikkature Tested-by: Jarkko Nikula --- In V4 : Updated with Nishant's comments and appened his Signed-off In V3 : Updated with Madhu's comments and appended Tested by Nikula In V2 : Appended Signed-off by Venkat and Ack by Madhu Here are my experiment numbers, on a Class 6 SDHC card: Read peformance is increased by 220% Write Performance is increased by 52% drivers/mmc/host/omap_hsmmc.c | 19 ++++++++++++++++--- 1 files changed, 16 insertions(+), 3 deletions(-) diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c index 8c97c22..9c1a60e 100644 --- a/drivers/mmc/host/omap_hsmmc.c +++ b/drivers/mmc/host/omap_hsmmc.c @@ -2091,10 +2091,23 @@ static int __init omap_hsmmc_probe(struct mmc->caps |= MMC_CAP_MMC_HIGHSPEED | MMC_CAP_SD_HIGHSPEED | MMC_CAP_WAIT_WHILE_BUSY; - if (mmc_slot(host).wires >= 8) - mmc->caps |= (MMC_CAP_8_BIT_DATA | MMC_CAP_4_BIT_DATA); - else if (mmc_slot(host).wires >= 4) + switch (mmc_slot(host).wires) { + case 8: + mmc->caps |= MMC_CAP_8_BIT_DATA; + /* Fall through */ + case 4: mmc->caps |= MMC_CAP_4_BIT_DATA; + break; + case 1: + /* Nothing to crib here */ + case 0: + /* Assuming nothing was given by board, Core use's 1-Bit */ + break; + default: + /* Completely unexpected.. Core goes with 1-Bit Width */ + dev_crit(mmc_dev(host->mmc), "Invalid width %d\n used!" + "using 1 instead\n", mmc_slot(host).wires); + } if (mmc_slot(host).nonremovable) mmc->caps |= MMC_CAP_NONREMOVABLE; -- 1.6.3.3