On Wed, Feb 11, 2015 at 1:28 PM, Ulf Hansson [off-list ref] wrote:

[...]

quoted

quoted

+int mmc_pwrseq_simple_alloc(struct mmc_host *host, struct device *dev)
+{
+       struct mmc_pwrseq_simple *pwrseq;
+
+       pwrseq = kzalloc(sizeof(struct mmc_pwrseq_simple), GFP_KERNEL);
+       if (!pwrseq)
+               return -ENOMEM;
+
+       pwrseq->pwrseq.ops = &mmc_pwrseq_simple_ops;
+       host->pwrseq = &pwrseq->pwrseq;

How about making this function return a struct mmc_pwrseq * so this
last line can be moved to mmc_pwrseq_alloc() instead of requiring all
power sequences to do it?

The same applies to

    host->pwrseq = NULL;

in mmc_pwrseq_simple_free(), which could be done in mmc_pwrseq_free() it seems.

Thanks for reviewing!

I like you suggestion, though $subject patch is already part of the PR for 3.20.

Feel free to send a patch, I would happily apply it.

Damn, why am I always so late to review patches...

I will send a fixup patch - better to have this done early.

Thanks!