From mboxrd@z Thu Jan 1 00:00:00 1970 From: Michal Simek Date: Wed, 07 May 2014 13:45:58 +0200 Subject: [U-Boot] ARM: zynq: sdhci clock frequency init question In-Reply-To: <99075D6EE6F8604C91868421B2682FFF1AFC392C8A@IAD2MBX09.mex02.mlsrvr.com> References: <99075D6EE6F8604C91868421B2682FFF1AFC392C8A@IAD2MBX09.mex02.mlsrvr.com> Message-ID: <536A1CF6.8060100@monstr.eu> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de Hi Krunal, sorry for delay. On 04/25/2014 06:19 PM, Krunal Desai wrote: > Hi all - > > I noticed that in zynq_sdhci.c, responsible for initializing PS SD controller(s), host controller max clock frequency is always set to 52MHz (http://git.denx.de/?p=u-boot.git;a=blob;f=drivers/mmc/zynq_sdhci.c;h=fdce2c2c10ec85c4a291532f927eae4a0b5627c9;hb=master#l34). In cases where user is using EMIO connectivity, max clock speed is limited to 25MHz. This results in out-of-spec operation as the divider calculation logic trusts the input to add_sdhci() and does not check the SLCR itself to confirm what the IO clock to SD controller actually is. > > I think I saw OF/device-tree support patched in recently (I am on the 2013.4 tag); I think the "right" way to solve this is add capability in zynq_sdhci_init to read device-tree for 'clock-frequency' property, and use that to populate the arguments to add_sdhci() with that information, defaulting to a safe minimum (25MHz?) if no entry is found. Otherwise, I suppose a config could be added akin to 'ZYNQ_SD0_MIO'/'ZYNQ_SD1_MIO' to populate the correct minimum value. > > Does this sound sane? I am thinking of implementing that as a patch for ourselves internally, but I think it will be of value to the greater community as well. we didn't test this configuration that's why 52MHz is there as default case. I think that should be easily possible to detect MIO setting as we are doing for qspi/nand/usb. We have this code in our xilinx repository and I have sent patches for mainline review 2 weeks ago or something like that. I haven't got any NACK and I am going to send pull request to ARM custodian when I fix fpga patches. It means detection via MIO setting is reasonable way how to do it. When you know that this go through EMIO 25MHz should be used. Thanks, Michal -- Michal Simek, Ing. (M.Eng), OpenPGP -> KeyID: FE3D1F91 w: www.monstr.eu p: +42-0-721842854 Maintainer of Linux kernel - Microblaze cpu - http://www.monstr.eu/fdt/ Maintainer of Linux kernel - Xilinx Zynq ARM architecture Microblaze U-BOOT custodian and responsible for u-boot arm zynq platform -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 263 bytes Desc: OpenPGP digital signature URL: