Re: [PATCH v4 2/2] i2c: dwsignware: determine HS tHIGH and tLOW based on HW parameters

[Jarkko Nikula]

On 10/3/24 2:41 PM, Andy Shevchenko wrote:
> On Thu, Oct 03, 2024 at 07:15:24PM +0800, Michael Wu wrote:
> > In commit 35eba185fd1a ("i2c: designware: Calculate SCL timing parameter
> > for High Speed Mode") the SCL high period count and low period count for
> > high speed mode are calculated based on fixed tHIGH = 160 and tLOW = 120.
> > However, the set of two fixed values is only applicable to the combination
> > of hardware parameters IC_CAP_LOADING is 400 and IC_CLK_FREQ_OPTIMIZATION
> > is true. Outside of this combination, the SCL frequency may not reach
> > 3.4 MHz because the fixed tHIGH and tLOW are not small enough.
> >
> > If IC_CAP_LOADING is 400, it means the bus capacitance is 400pF;
> > Otherwise, 100 pF. If IC_CLK_FREQ_OPTIMIZATION is true, it means that the
> > hardware reduces its internal clock frequency by reducing the internal
> > latency required to generate the high period and low period of the SCL line.
> >
> > Section 3.15.4.5 in DesignWare DW_apb_i2b Databook v2.03 says that when
> > IC_CLK_FREQ_OPTIMIZATION = 0,
> >
> >      MIN_SCL_HIGHtime = 60 ns for 3.4 Mbps, bus loading = 100pF
> > 		     = 120 ns for 3.4 Mbps, bus loading = 400pF
> >      MIN_SCL_LOWtime = 160 ns for 3.4 Mbps, bus loading = 100pF
> > 		    = 320 ns for 3.4 Mbps, bus loading = 400pF
> >
> > and section 3.15.4.6 says that when IC_CLK_FREQ_OPTIMIZATION = 1,
> >
> >      MIN_SCL_HIGHtime = 60 ns for 3.4 Mbps, bus loading = 100pF
> > 		     = 160 ns for 3.4 Mbps, bus loading = 400pF
> >      MIN_SCL_LOWtime = 120 ns for 3.4 Mbps, bus loading = 100pF
> > 		    = 320 ns for 3.4 Mbps, bus loading = 400pF
> >
> > In order to calculate more accurate SCL high period count and low period
> > count for high speed mode, two hardware parameters IC_CAP_LOADING and
> > IC_CLK_FREQ_OPTIMIZATION must be considered together. Since there're no
> > registers controlliing these these two hardware parameters, users can
> > declare them in the device tree so that the driver can obtain them.
>
> As long as DT schema (new properties) is accepted, this LGTM now,
> Reviewed-by: Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx>
Acked-by: Jarkko Nikula <jarkko.nikula@xxxxxxxxxxxxxxx>