Comment 31 for bug 604122

Revision history for this message
Andy Whitcroft (apw) wrote :

Ok had a quick look at this option. Seems it got enabled as part of rebasing forward for Maverick. It seems that for v2.6.36 that this version of the diable was deprecated in favour of a better mechanism, and in that commit they note that it did not work so well either:

  commit ccc92c23240cdf952ef7cc39ba563910dcbc9cbe
  Author: Maxim Levitsky <email address hidden>
  Date: Tue Aug 10 18:01:42 2010 -0700

    mmc: make sdhci work with ricoh mmc controller

    The current way of disabling it is not well tested by vendor and has all
    kinds of bugs that show up on resume from ram/disk. A very good example
    is a dead SDHCI controller.

    Old way of disabling is still supported by continuing to use
    CONFIG_MMC_RICOH_MMC.

    Based on 'http://list.drzeus.cx/pipermail/sdhci-devel/2007-December/002085.html'
    Therefore most of the credit for this goes to Andrew de Quincey

From this write up it seems appropriate for kernels on v2.6.36 and later to turn this off expecting the device to work correctly without, it may also be appropriate to disable this work around in Maverick but that is less clear. I will put together a Natty test kernel with this change for testing. If that works ok we can try older releases.