[OpenWrt-Devel] [PATCH v3 2/5] ath79: WNR612v2: improve device support

Michal Cieslakiewicz michal.cieslakiewicz at wp.pl
Mon Aug 12 10:24:22 EDT 2019


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

On Mon, 12 Aug 2019 14:16:18 +0200
"Adrian Schmutzler" <mail at adrianschmutzler.de> wrote:

> Hi,
> 
> just two annotations below.
> 
> > -			uboot: partition at 0 {
> > +			partition at 0 {  
> 
> You could also just keep it. But that's just a matter of taste.
> 

Hello Adrian,

This 'uboot' label was used only for MAC address extraction from
u-boot partition (kinda strange, I couldn't find a clue why it was
expected there), so I decided to remove it.

> >  &pcie {
> > @@ -116,6 +122,8 @@
> >  	ath9k: wifi at 0,0 {
> >  		compatible = "pci168c,002b";
> >  		reg = <0x0000 0 0 0 0>;
> > +		mtd-mac-address = <&art 0x0>;
> > +		mtd-mac-address-increment = <1>;  
> 
> Calibration data is read here:
> https://github.com/openwrt/openwrt/blob/master/target/linux/ath79/base-files/etc/hotplug.d/firmware/10-ath9k-eeprom#L184
> 
> Does this device get a MAC address from calibration data? If yes, I
> would prefer that one to a calculated address and remove the entries
> in DTS.

Nope, it should be calculated. In theory, according to Netgear uboot
sources, it should be in ART region at 0x108c but in all devices I
could inspect that area was empty (0xff). ar71xx target also derives
wlan0 MAC from ethX MACs.

> If not, I wonder whether your setup here will persist or
> being overwritten by some "wrong" address during calibration anyway.
> (Then one would have to use ath9k_patch_fw_mac in the linked file.
> 

MAC addresses for all interfaces are correctly assigned with this
patchset in place (tested), so apparently nothing interferes here.

Cheers
Michal
-----BEGIN PGP SIGNATURE-----
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=DopH
-----END PGP SIGNATURE-----
_______________________________________________
openwrt-devel mailing list
openwrt-devel at lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel



More information about the openwrt-devel mailing list