Re: [PATCH] spi: dw: Fix dynamic speed change

[Mark Brown <broonie-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>]

[-- Attachment #1: Type: text/plain, Size: 1107 bytes --]

On Tue, Nov 04, 2014 at 04:36:56PM -0600, tthayer-yzvPICuk2ABMcg4IHK0kFoH6Mc4MB0Vx@public.gmane.org wrote:

> When speed_hz is not declared in a SPI transfer, the transfer speed is
> not updated for the next read/write on /dev/spidevX.Y. The element

Why is the behaviour of spidev relevant here, if there is a problem with
spidev why is it being addressed in a specific driver?

> spi_transfer->speed_hz is filled with spi->max_speed_hz. The test of
> if (transfer->speed_hz != speed) doesn't work because the chip->speed_hz
> matches transfer->speed_hz and the clock divider is not updated.

In what way does the test "not work"?  What should the test be doing and
what is it actually doing?

> +	/* Always calculate the desired clock divider */
> +	speed = transfer->speed_hz ? transfer->speed_hz : chip->speed_hz;

Please avoid using the ternery operator standalone, write a normal if to
help people read the code.  Though in this case the core should always
ensure that there is a speed set on each transfer so I'm not clear when
this test would ever use anything other than transfer->speed_hz anyway.

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 473 bytes --]