On Wed, Jun 5, 2019 at 6:30 AM Eduardo Valentin [off-list ref] wrote:

Hey Andry,

Long time no seeing :-)

True!

quoted

quoted

+#define MQ_MSGBUF_SIZE             CONFIG_I2C_SLAVE_MQUEUE_MESSAGE_SIZE
+#define MQ_QUEUE_SIZE              CONFIG_I2C_SLAVE_MQUEUE_QUEUE_SIZE

quoted

+#define MQ_QUEUE_NEXT(x)   (((x) + 1) & (MQ_QUEUE_SIZE - 1))

Also possible ((x + 1) % ..._SIZE)

Right.. but I suppose the original idea is to avoid divisions on the hotpath.

So, I am actually fine with the limitation of only using power of 2.

The original code implies that anyway, so, my proposal doesn't
restrict it any farther.

quoted

quoted

+   {
+           .compatible = "i2c-slave-mqueue",
+   },

quoted

+   { },

No need for comma here.

It does not hurt to have it either :-)

It's just a protection against some weird cases of adding entries
behind the terminator.

quoted

quoted

+           .of_match_table = of_match_ptr(i2c_slave_mqueue_of_match),

Wouldn't compiler warn you due to unused data?
Perhaps drop of_match_ptr() for good...

Not sure what you meant here. I dont see any compiler warning.
Also, of_match_ptr seams to be well spread in the kernel.

If this will be compiled with CONFIG_OF=n...
Though I didn't check all dependencies to see if it even possible. In
any case of_match_ptr() is redundant in both cases here.
Either you need to protect i2c_slave_mqueue_of_match with #ifdef
CONFIG_OF, or drop the macro use.

P.S. Taking into account the last part, I would wait for v7 with that
fixed followed by fixing other nits.

-- 
With Best Regards,
Andy Shevchenko