Re: [PATCH 3/4] thunderbolt: Add helpers to check if CL states are enabled on port

[Mika Westerberg <mika.westerberg@linux.intel.com>]

Hi Lukas,

On Tue, Aug 23, 2022 at 04:02:16PM +0200, Lukas Wunner wrote:
> On Tue, Aug 23, 2022 at 01:53:51PM +0300, Mika Westerberg wrote:
> > +/**
> > + * tb_port_is_clx_enabled() - Is given CL state enabled
> > + * @port: USB4 port to check
> > + * @clx: CL state to check
> > + *
> > + * Returns true if given CL state is enabled for @port.
> > + */
> > +bool tb_port_is_clx_enabled(struct tb_port *port, enum tb_clx clx)
> > +{
> > +	u32 phy, mask = LANE_ADP_CS_1_CL0S_ENABLE | LANE_ADP_CS_1_CL1_ENABLE;
> > +	int ret;
> > +
> > +	if (!tb_port_clx_supported(port, clx))
> > +		return false;
> > +
> > +	ret = tb_port_read(port, &phy, TB_CFG_PORT,
> > +			   port->cap_phy + LANE_ADP_CS_1, 1);
> > +	if (ret)
> > +		return false;
> > +
> > +	return (phy & mask) == mask;
> > +}
> > +
> [...]
> > +static inline bool tb_port_are_clx_enabled(struct tb_port *port)
> > +{
> > +	return tb_port_is_clx_enabled(port, TB_CL1) ||
> > +	       tb_port_is_clx_enabled(port, TB_CL2);
> > +}
> 
> If you change enum tb_clx to use "power of two" values (0 1 2 4 8 ...)
> then you could just pass a bitmask to tb_port_is_clx_enabled()
> and thus need only a single invocation in tb_port_are_clx_enabled().
> Just a thought.

Sure good point. I'll do that in v2 thanks!