Hi,
I'm working with a private custom board so I'm not planning on posting it. It has signal integrity issues with its interface to the SD card controller and needs to run the clock at 26MHz versus 52MHz. In other words, the SD card is not bad, the board is the problem.
I thought it might be a useful change for others.
I should update the commit comment, though. I did some more experimentation and it's not that I don't get rail-rail-swing, but that there's a lot ringing and overshoot at the faster clock speed.
Thanks, Jonathan
On Mon, Feb 20, 2012 at 11:15 AM, Tom Rini trini@ti.com wrote:
On Fri, Feb 17, 2012 at 04:33:05PM -0800, Jonathan Solnit wrote:
Adds parameters to the OMAP MMC initialization function so the board can mask host capabilities and set the maximum clock frequency. While the OMAP supports a certain set of MMC host capabilities, individual boards may be more restricted. In particular, if the clock is set faster than the board can handle there will not be rail-to-rail swing at the card.
So, this is for a not yet (but I believe the intention is to post soon) mainline board. My concern here is that a given bad SD card would / should show the problem on any board, so we should be adjusting our caps globally, yes? Or is there a board (and not SoC) specific part of the equation? Thanks!
-- Tom