| Message ID | 20171016183253.2291-1-chris.brandt@renesas.com (mailing list archive) |
|---|---|
| State | Superseded |
| Delegated to: | Geert Uytterhoeven |
| Headers | show |
Hi Wolfram, On Monday, October 16, 2017, Chris Brandt wrote: > Most systems with this i2c are going to have a clock of either 33.33MHz or > 32MHz. That 4% difference is not reason enough to warrant that the driver > to completely fail. > > Signed-off-by: Chris Brandt <chris.brandt@renesas.com> > --- > v3: > * now check a range of safe frequencies > v2: > * simplified error message. > --- Actually, if you want to hold off on this patch, I'm almost done with a real fix which is to calculate any frequency based on current parent clock and rise/fall timings specified in the DT as you suggested. Chris
> Actually, if you want to hold off on this patch, I'm almost done with a > real fix which is to calculate any frequency based on current parent > clock and rise/fall timings specified in the DT as you suggested. Sounds awesome! So, I'll wait a little how the formula goes. We can always fall back to this patch if all fails... Thanks!
diff --git a/drivers/i2c/busses/i2c-riic.c b/drivers/i2c/busses/i2c-riic.c index c811af4c8d81..4c001d09aac2 100644 --- a/drivers/i2c/busses/i2c-riic.c +++ b/drivers/i2c/busses/i2c-riic.c @@ -299,12 +299,14 @@ static int riic_init_hw(struct riic_dev *riic, u32 spd) /* * TODO: Implement formula to calculate the timing values depending on - * variable parent clock rate and arbitrary bus speed + * variable parent clock rate and arbitrary bus speed. + * For now, just use calculations based on a 33.33MHz clock. */ rate = clk_get_rate(riic->clk); - if (rate != 33325000) { + if ((rate < 32000000) || (rate > 33400000)) { dev_err(&riic->adapter.dev, - "invalid parent clk (%lu). Must be 33325000Hz\n", rate); + "invalid parent clk (%lu). Must be 32.0MHz-33.4MHz.\n", + rate); clk_disable_unprepare(riic->clk); return -EINVAL; }
Most systems with this i2c are going to have a clock of either 33.33MHz or 32MHz. That 4% difference is not reason enough to warrant that the driver to completely fail. Signed-off-by: Chris Brandt <chris.brandt@renesas.com> --- v3: * now check a range of safe frequencies v2: * simplified error message. --- --- drivers/i2c/busses/i2c-riic.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)