[PATCH 0/6] mmc: sdhci-esdhci-imx: fix cd/wp regression

On Tue, Jul 14, 2015 at 07:25:09AM +0200, Soeren Moch wrote:

On 13.07.2015 13:43, Dong Aisheng wrote:

quoted

On Thu, Jul 09, 2015 at 09:27:36PM +0800, Shawn Guo wrote:

quoted

On Thu, Jul 09, 2015 at 05:29:50PM +0800, Dong Aisheng wrote:

quoted

I agree with you.
One difference is that i'd like to fix it ASAP without reverting 8d86e4f
due to more patches depends on it is already there as i mentioned above..
Revert it may need to revert a lot others.

The pain is that v4.1 is left broken.

Oh, crap!  I thought it's been there with only v4.2-rc1, and did not
know v4.1 is already broken.  In that case, reverting 8d86e4f isn't
the best option.  I suggest you rebase the dts series on top of
v4.2-rc1, and send it via mmc tree.

Yes, will go that way.

Then you may need help take care of new dts going through your tree
to make sure no broken code in again.

Dong,

I use a microSD card (mmc0) with the root filesystem on a tbs2910 board.
I cannot see any difference between linux-4.0 and linux-4.1 (vanilla
kernel with corresponding dtb), there is absolutely no problem with card
detect.

Can you please explain what exactly is broken in linux-4.1?

The CD function is broken.
If you enable CONFIG_MMC_DEBUG, you will see the MMC core keep polling the
card because MMC_CAP_NEEDS_POLL is wrongly set.
You can refer to cover-letter for more detailed explaination.

If you're using a non-removable card(having non-removable property in dts),
you may not have this issue.

Regards
Dong Aisheng

Thanks,
Soeren

quoted

Thanks

Regards
Dong Aisheng

quoted

Shawn