Re: [PATCH] mmc: dt: Add 'broken-cd' DT binding

Hi Chris,

On Tue, Aug 21, 2012 at 10:48:43AM -0400, Chris Ball wrote:

Aside: the bindings do not match the code.  The bindings document says
to use "fsl,cd-internal", and imx51-babbage.dts does so -- but the code
doesn't check for "fsl,cd-internal", it checks for "fsl,cd-controller":

        if (of_get_property(np, "fsl,cd-controller", NULL))
                boarddata->cd_type = ESDHC_CD_CONTROLLER;

The same confusion is present for fsl,wp-{controller,internal}.

I'm trying to fix the mismatch between dts and code, but seeing that
the "controller" mode has been broken for quite a while.  I just ran
a bisect and was told as below.

30832ab56c80d96cfaf5a786524f0d8c57fadfa1 is the first bad commit
commit 30832ab56c80d96cfaf5a786524f0d8c57fadfa1
Author: Todd Poynor [off-list ref]
Date:   Tue Dec 27 15:48:46 2011 +0200

    mmc: sdhci: Always pass clock request value zero to set_clock host op

    To allow the set_clock host op to disable the SDCLK source when not
    needed, always call the host op when the requested clock speed is
    zero.  Do this even if host->clock already equals zero, because
    the SDHCI driver may set that value (without calling the host op)
    to force an update at the next (non-zero-speed) call.

    Signed-off-by: Todd Poynor [off-list ref]
    Acked-by: Adrian Hunter [off-list ref]
    Signed-off-by: Chris Ball [off-list ref]

Reverting the commit will get "controller" mode back to work.

-- 
Regards,
Shawn