Messages in this thread | | | Subject | Re: [RESEND] i2c: mediatek: Get device clock-stretch time via dts | From | Qii Wang <> | Date | Wed, 14 Apr 2021 09:37:24 +0800 |
| |
On Tue, 2021-04-13 at 22:17 +0200, Wolfram Sang wrote: > On Mon, Apr 12, 2021 at 08:03:14PM +0800, Qii Wang wrote: > > I can't see the relationship between "i2c-scl-falling-time-ns" and clock > > stretching, is there a parameter related to clock stretching? > > ( you wrote "i2c-scl-falling-time-ns" above, didn't you mean > "i2c-scl-internal-delay-ns" instead? ) >
I am sorry, I have confused your comment with lkjoon's comment in the last mail. what I actually want to say is "i2c-scl-internal-delay-ns".
> Not yet, and I wonder if there can be one. In I2C (not SMBus), devices > are allowed to stretch the clock as long as they want, so what should be > specified here? > > I suggesteed "internal-delay" because AFAIU your hardware needs this > delay to be able to cope with clock stretching. >
If there is not a maximum value for clock stretching, "i2c-scl-internal-delay-ns" should be a good choice for our hardware, although it maybe not for clock stretching.
> > If you think both of them will affect the ac-timing of SCL, at this > > point, "i2c-scl-falling-time-ns" maybe a good choice. > > Do you mean "i2c-scl-falling-time-ns" or "i2c-scl-internal-delay-ns"? >
"i2c-scl-internal-delay-ns" is better.
Thanks for your review. Qii
| |