On 12/1/20 12:52 PM, Andrew Payne via USRP-users wrote: > That worked as planned, no more duplicate MAC addresses now. > > As you can see from the following 3 boot message blocks (1. before, 2. > after applying said env settings, 3. thereafter) it knew the environment > MAC didn't match the ROM MAC. But it's just a warning. > > Thanks!
Thanks for testing. I'm sure this will come up with work I have, nice to already understand the issue. This definitely should be fixable in u-boot. I suspect it can be done with config options, otherwise the code paths need reviewing. mac addresses follow hardware not removable sd cards. Philip > > --------------------------------------------------------------------------------------- > 1. before > --------------------------------------------------------------------------------------- > U-Boot 2018.07 (Dec 16 2019 - 20:52:43 +0000) > > Model: NI Ettus Research E31x SDR > DRAM: ECC disabled 1 GiB > MMC: sdhci@e0100000: 0 > Loading Environment from MMC... OK > In: serial@e0000000 > Out: serial@e0000000 > Err: serial@e0000000 > NI Ettus Research E31x SG3 SDR Rev H s/n 31370F8 > Net: ZYNQ GEM: e000b000, phyaddr 0, interface rgmii-id > > Warning: ethernet@e000b000 MAC addresses don't match: > Address in ROM is 00:80:2f:18:24:ef > Address in environment is 00:80:2f:19:4c:37 > eth0: ethernet@e000b000 > Automatic boot in 3s... > Enter 'noautoboot' to enter prompt without timeout > ni-e31x-uboot> > > --------------------------------------------------------------------------------------- > 2. after applying said env settings > --------------------------------------------------------------------------------------- > U-Boot 2018.07 (Dec 16 2019 - 20:52:43 +0000) > > Model: NI Ettus Research E31x SDR > DRAM: ECC disabled 1 GiB > MMC: sdhci@e0100000: 0 > Loading Environment from MMC... OK > In: serial@e0000000 > Out: serial@e0000000 > Err: serial@e0000000 > NI Ettus Research E31x SG3 SDR Rev H s/n 31370F8 > Net: ZYNQ GEM: e000b000, phyaddr 0, interface rgmii-id > > Warning: ethernet@e000b000 using MAC address from ROM > eth0: ethernet@e000b000 > Automatic boot in 3s... > Enter 'noautoboot' to enter prompt without timeout > Saving Environment to MMC... Writing to redundant MMC(0)... OK > Copying FIT from SD to RAM... > 5866988 bytes read in 336 ms (16.7 MiB/s) > ## Loading kernel from FIT Image at 02000000 ... > > --------------------------------------------------------------------------------------- > 3. thereafter > --------------------------------------------------------------------------------------- > U-Boot 2018.07 (Dec 16 2019 - 20:52:43 +0000) > > Model: NI Ettus Research E31x SDR > DRAM: ECC disabled 1 GiB > MMC: sdhci@e0100000: 0 > Loading Environment from MMC... OK > In: serial@e0000000 > Out: serial@e0000000 > Err: serial@e0000000 > NI Ettus Research E31x SG3 SDR Rev H s/n 31370F8 > Net: ZYNQ GEM: e000b000, phyaddr 0, interface rgmii-id > eth0: ethernet@e000b000 > Automatic boot in 3s... > Enter 'noautoboot' to enter prompt without timeout > Copying FIT from SD to RAM... > 5866988 bytes read in 336 ms (16.7 MiB/s) > ## Loading kernel from FIT Image at 02000000 ... > > On Fri, Nov 27, 2020 at 7:39 PM Philip Balister <phi...@balister.org> wrote: > >> OK try this at the u-boot prompt: >> >> env default -a >> env save >> >> and then >> >> reset >> >> This should reset the u-boot env to the default values and I think this >> resets the ethaddr variable. Then you write it to the mmc (99% >> certain). On the next hard reset hopefully it reads the address from the >> eeprom. >> >> Philip >> >> On 11/27/20 2:20 PM, Andrew Payne wrote: >>> Thanks Aneesh but the decompiled dts file has no mention of the exact MAC >>> address for eth0, but just to read from the eeprom from what I can >> gather. >>> Plus an md5sum of the dtb file on the sdimg from Ettus is the same >> checksum >>> as an e310 that has booted. >>> >>> On Fri, Nov 27, 2020 at 1:40 PM Philip Balister <phi...@balister.org> >> wrote: >>> >>>> On 11/27/20 1:34 PM, aneesh patel via USRP-users wrote: >>>>> Hi Andrew, >>>>> The MAC may be in the devicetree blobs in the boot area of the SD >> image. >>>>> You will need dtcedit to decompile, edit, and recompile as needed. >>>> >>>> I don't think so, since the first time the card boots it does read from >>>> the i2c eeprom. >>>> >>>> I'm guessing u-boot sets the ethaddr env var and saves it in the >>>> environment, but I forget where that might be saved. Need to review the >>>> u-boot configuration for setting about the u-boot env (maybe getting >>>> saved to the sd card). >>>> >>>> Anyone from Ettus paying attention? This is a pretty serious problem for >>>> people copying cards and using them in different units. The correct >>>> behavior is follow the MAC address programmed into the i2c eeprom. >>>> >>>> Philip >>>> >>>>> Amp >>>>> >>>>> Sent from Yahoo Mail on Android >>>>> >>>>> On Fri, Nov 27, 2020 at 1:23 PM, Andrew Payne via USRP-users< >>>> usrp-users@lists.ettus.com> wrote: >>>> _______________________________________________ >>>>> USRP-users mailing list >>>>> USRP-users@lists.ettus.com >>>>> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com >>>>> >>>>> >>>>> >>>>> _______________________________________________ >>>>> USRP-users mailing list >>>>> USRP-users@lists.ettus.com >>>>> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com >>>>> >>>> >>> >> > > > _______________________________________________ > USRP-users mailing list > USRP-users@lists.ettus.com > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com > _______________________________________________ USRP-users mailing list USRP-users@lists.ettus.com http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
