Messages in this thread | | | Subject | Re: [PATCH] cpsw: Fix interrupt storm among other things | From | Pantelis Antoniou <> | Date | Wed, 30 Jan 2013 10:34:12 +0200 |
| |
Hi Peter,
On Jan 28, 2013, at 11:14 PM, Peter Korsgaard wrote:
>>>>>> "Pantelis" == Pantelis Antoniou <panto@antoniou-consulting.com> writes: > > 'among other things' is not a very descriptive commit message. > > Pantelis> Fix interrupt storm on bone A4 cause by non-by-the-book > Pantelis> interrupt handling. While at it, added a non-NAPI mode > Pantelis> (which is easier to debug), plus some general fixes. > > 'bone A4' is also not very descriptive. There also was an A4 revision of > the "old" beaglebone. I guess you're instead referring to a new die > revision? > > What is the impact of this change on earlier devices? >
Same driver works perfectly fine with the PG1.0 bone version I have. If you take a look at the TRM it should be pretty obvious why the change was needed.
> Pantelis> +++ b/Documentation/devicetree/bindings/net/cpsw.txt > Pantelis> @@ -20,6 +20,7 @@ Required properties: > Pantelis> - cpts_clock_shift : Denominator to convert input clock ticks into nanoseconds > Pantelis> - phy_id : Specifies slave phy id > Pantelis> - mac-address : Specifies slave MAC address > Pantelis> +- disable-napi : Disables driver NAPI > > napi is not a hardware feature, so it doesn't belong here (if anything, > it should be linux,disable-napi). >
Point taken.
> -- > Bye, Peter Korsgaard
Regards
-- Pantelis
| |