AM3505 MAC Address

R. Keith Montgomery

Other Parts Discussed in Thread: AM3505

Hello,

Can someone tell me what range of values are programmed into the EFUSE for CONTROL_EFUSE_EMAC_LSB and MSB? Are these values guaranteed to be unique based on TI assigned IEEE Organizationally Unique Identifier (OUI)?

How are the EFUSE_EMAC_LSB/MSB values chosen then programmed?

Thanks,

Keith

over 12 years ago

0 Cvetolin Shulev-XID over 12 years ago

TI__Guru 65405 points

Hi Keith,

The CONTROL_EFUSE_EMAC_LSB register value (fist 3 bytes in the MAC address) are typically reserved for a specific company. The 00:24:BA:xx:xx:xx addresses are reserved for TI for example. The CONTROL_FUSE_EMAC_MSB should be unique for each one TI's device. Both registers are read only.

Tsvetolin Shulev

0 R. Keith Montgomery over 12 years ago in reply to Cvetolin Shulev-XID

Intellectual 270 points

Hi Tsvetolin,

The AM3505 that we have purchased and inspected CONTROL_EFUSE_EMAC_LSB are not equal to TI's OUI 00:24:BA. Should these values in these parts be equal to TI's OUI?

Thanks,

Keith

0 Cvetolin Shulev-XID over 12 years ago in reply to R. Keith Montgomery

TI__Guru 65405 points

Hi Keith,

The 00:24:BA is one possible MAC address part on AM35x but you could check whether some MAC address is correct TI's MAC address using various tools. There are some such web based tools:

http://www.base64online.com/mac_address.php?ven=TexasIns+++++++++++++++++Texas+Instruments

http://www.techzoom.net/tools/check-mac.en

http://www.adminsub.net/mac-address-finder/

The MAC address could be changed be setting bootargs in u-boot and I recommend you to check and this.

Tsvetolin Shulev

0 R. Keith Montgomery over 12 years ago in reply to Cvetolin Shulev-XID

Intellectual 270 points

Hi Tsvetolin,

Thanks for this information. Can you clarify what bytes/bits of CONTROL_FUSE_EMAC_LSB correspond to XX:YY:ZZ for the first 24 bits of the MAC?

CONTROL_FUSE_EMAC_LSB is a 32 bit register location. I assume Byte0[7:0] = XX? Byte1[15:8]=YY and Byte2[23:16]=ZZ?

Thanks,

Keith

0 Cvetolin Shulev-XID over 12 years ago in reply to R. Keith Montgomery

TI__Guru 65405 points

There is a short description of the CONTROL_EFUSE_EMAC_LSB register fields: bites [31:24] - reserved, Byte0[23:16]=XX, Byte1[15:8]=YY, Byte{7:0]=ZZ. The CONTROL_EFUSE_EMAC_MSB register fields should be interpreted similarly.

Look an example of the CONTROL_EFUSE_EMAC_LSB register reading and processing (function davinci_eth_set_mac_addr):

http://repository.timesys.com/buildsources/u/u-boot/u-boot-2009.11/u-boot-2009.11-dm3730logic-20110822.patch

Tsvetolin Shulev

0 R. Keith Montgomery over 12 years ago in reply to Cvetolin Shulev-XID

Intellectual 270 points

Hi Tsvetolin,

The uboot patch uses the EMAC_MSB for the OUI. You previously stated that the EMAC_LSB is the OUI bytes for TI. Can you confirm which is correct?

Thanks,

Keith

0 Cvetolin Shulev-XID over 12 years ago in reply to R. Keith Montgomery

TI__Guru 65405 points

Keith, I sorry for the mistake. There is the whole description of the MAC address in both registers if the MAC address is AA:BB:CC:XX:YY:ZZ:

CONTROL_EFUSE_EMAC_MSB - bites [31:24] - reserved, Byte0[23:16]=AA, Byte1[15:8]=BB, Byte[7:0]=CC

CONTROL_EFUSE_EMAC_LSB - bites [31:24] - reserved, Byte0[23:16]=XX, Byte1[15:8]=YY, Byte[7:0]=ZZ

The MAC address vendor is determinate by the most significant bytes AA:BB:CC.

Tsvetolin Shulev

0 R. Keith Montgomery over 12 years ago in reply to Cvetolin Shulev-XID

Intellectual 270 points

Hi Tsvetolin,

This how I'm determining the MAC address for my AM3505 product. However, the EMAC_MSB does not match any OUI when searched in websites you provided. Can you confirm these two OUIs that came from AM3505 chips?

5C:6B:32

04:E4:51

Thanks,

Keith

0 Cvetolin Shulev-XID over 12 years ago in reply to R. Keith Montgomery

TI__Guru 65405 points

Hi Keith,

The MAC address finder database seems the most detailed and gives full information. It recognize your MAC address as a valid Texas Instruments's MAC address.

Tsvetolin Shulev

0 R. Keith Montgomery over 12 years ago in reply to Cvetolin Shulev-XID

Intellectual 270 points

Hi Tsvetolin,

Thank you for your help. This clarifies and answers my questions and observation.

Thanks,

Keith

Processors

Processors forum

AM3505 MAC Address