Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1505497 > unrolled thread

Re: [PATCH] net: cpsw: fix obtaining mac address for am3517

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH] net: cpsw: fix obtaining mac address for am3517 Tony Lindgren <tony@atomide.com> - 2016-10-21 08:40 +0200
    Re: [PATCH] net: cpsw: fix obtaining mac address for am3517 Tony Lindgren <tony@atomide.com> - 2016-10-21 10:00 +0200
      Re: [PATCH] net: cpsw: fix obtaining mac address for am3517 Tony Lindgren <tony@atomide.com> - 2016-10-21 12:00 +0200

#1505497 — Re: [PATCH] net: cpsw: fix obtaining mac address for am3517

* Jeroen Hofstee <jhofstee@victronenergy.com> [161020 12:57]:
> Commit b6745f6e4e63 ("drivers: net: cpsw: davinci_emac: move reading mac
> id to common file") did not only move the code for an am3517, it also
> added the slave parameter, resulting in a invalid (all zero) mac address
> being returned. So change it back to always read from slave zero, so it
> works again.

Hmm doesn't this now break it for cpsw with two instances? We may need
am3517 specific quirk flag instead?

Regards,

Tony

[toc] | [next] | [standalone]

#1505542

* Jeroen Hofstee <jhofstee@victronenergy.com> [161021 00:37]:
> Hello Tony,
> 
> On 21-10-16 08:38, Tony Lindgren wrote:
> > * Jeroen Hofstee <jhofstee@victronenergy.com> [161020 12:57]:
> > > Commit b6745f6e4e63 ("drivers: net: cpsw: davinci_emac: move reading mac
> > > id to common file") did not only move the code for an am3517, it also
> > > added the slave parameter, resulting in a invalid (all zero) mac address
> > > being returned. So change it back to always read from slave zero, so it
> > > works again.
> > Hmm doesn't this now break it for cpsw with two instances?
> > 
> 
> Yes, well, they get the same mac address at least. But does it matter?
> This changes davinci_emac_3517_get_macid, the only way to get there
> is:
> 
>     if (of_device_is_compatible(dev->of_node, "ti,am3517-emac"))
>         return davinci_emac_3517_get_macid(dev, 0x110, slave, mac_addr)
> 
> and the only user of ti,am3517-emac is arch/arm/boot/dts/am3517.dtsi,
> which only has one emac. So the change is already am3517 specific.
> 
> > We may need am3517 specific quirk flag instead?
> 
> Given above, it is already am3517 specific. Let me know if you prefer this
> route then I will have a look at it.

Oh OK thanks for explaining it :) As it's already am3517 specific:

Acked-by: Tony Lindgren <tony@atomide.com>

[toc] | [prev] | [next] | [standalone]

#1505749

* Jeroen Hofstee <jhofstee@victronenergy.com> [161021 02:31]:
> Aaah, lets wait a sec. I just saw there is another user of this function,
> so above is simply not true....
> 
>     if (of_machine_is_compatible("ti,dra7"))
>         return davinci_emac_3517_get_macid(dev, 0x514, slave, mac_addr);

Oh OK, then this will produce duplicate macs.

> So let me check if I don't break that one..... or better, let me send a v2,
> which
> changes the caller to pass slave as 0 here?
> 
>     if (of_device_is_compatible(dev->of_node, "ti,am3517-emac"))
>         return davinci_emac_3517_get_macid(dev, 0x110, slave, mac_addr);
> 

Yeah that sounds good to me.

Regards,

Tony

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web