Micron eMMC and OMAP L138

Gunter,

OMAP-L138 does not support eMMC standard right now OMAP-L138 Supports the following standards

The MMC/SD card controller supports the following industry standards

MMC (Multimedia Card) Specification v4.0

SD (Secure Digital) Physical Layer Specification v1.1

SDIO (Secure Digital Input Output) Specification v2.0

Regards

Antony  

• --------------------------------------------------------------------------------------------------------
  Please click the Verify Answer button on this post if it answers your question.
  --------------------------------------------------------------------------------------------------------

Hi Antony, thanks - but I want to keep digging.

Do you mean - can't use eMMC at all, or can't use eMMC to boot?

 From: http://www.ti.com/lit/an/sprab41d/sprab41d.pdf

The below says:"does not support eMMC to boot" and gives an example of how to get around this using eMMC.

6.5 MMC/SD Boot

The MMC/SD boot mode is compliant with version 2.0 of the SD specification and version 4.2 of the eMMC specification. The boot loader and MMC/SD peripheral does not support features of the eMMC 4.3 or 4.4 specifications related to boot (no boot operation mode and no support for boot partitions). The AIS boot image is expected to be in the user data area of the memory device, written to address 0 of that region. The bootloader detects the AIS image by checking for the magic word (0x41504954). If the magic word is not found, the bootloader increments the starting address by 0x200 and tries again. The bootloader fails if the magic word is not found within the first 2 MB of the memory card.

Typically, the ROM boot loader will first try to detect an SD card. If that fails (i.e., a timeout occurs), the boot loader will then attempt to find an MMC or eMMC device. In cases where it is known that an SD card will never be present (for example, a memory card slot is not used and an eMMC device is placed on the PCB), the BOOT[5] pin of the SoC device should be pulled high to force the bootloader to skip the SD detection. This will guarantee the quickest boot in these circumstances. For further details, see Section 9.

Ping!

Thanks Antony,

Gunter

Gunter,

Sorry for the delay

I'm discussing with the Internal team regarding you Query. I willl get back to you soon.

In the bootloader application note says it support eMMC Version 4.2 .But in the TRM and the datasheet no where it's mentioned.

Regards

Antony  

Great, thank you Mukul, that clears things up.

If  we have any questions once we have an eMMC hooked up to the OMAP we'll let you know.

Thanks again,

Gunter

Hey Gunter,

Just read the thread and I wanted to know how it worked out for you ?

Thanks,

Maxime

Hey,

I've been trying to make my Micron eMMC work ... All the initialization works, but read and write don't.

Have you been experiencing the same Gunter ?

Thanks,

Maxime

I have a similar question.  Is the MMC/SD boot features of eMMC 4.3 or 4.4 for a eMMC card supported in the C6748 yet (boot operation mode, boot partitions)?  SPRAAT2E (Using the TMS... Bootloader) says no, but that document is dated August 2011.

Thanks,

Tobyn VanVeghten

Hi Maxime,

just wanted to hear if you got it working, and what you did to accomplish it. We seem to have the same problem, but with another TI ARM processor. Initialization works but we can't read or write properly.

Thomas

Hi Thimas !

I did not get the time to try again.

I got some support from Micron some time ago and they told me the chips I was testing were in end of life, since I didn't know what else to try and I didn't have much more time to waste, I was waiting for the new chips.

What chip have you been testing ?
The ones that were not working for me have the reference 'N2M400GDB321A3CE' and were to be replaced by 'MTFC8GLDDQ-4M IT'.

I hope I can try the new chips very soon.

Can you keep me updated with any progress of yours ?

Thanks.

Hey Thomas,

I'm coming back to you about this eMMC chip, have you managed to resolve your problem ?

Regards,

--

Maxime

Hi Maxime,

thank you for returning. Unfortunately we haven't solved our problem yet. We have some very strange behaviour which we can't figure out.

Our problem is not booting from eMMC, we haven't gotten to that yet (right now we boot via Ethernet). It's with our eMMC driver when trying to read/write from the eMMC.

The first time we try to read a block of data we get 4 times the amount requested and data is rubbish. If we then repeat the exact same read the sequence we get the right amount of data and data is correct. Repeating it again starts the process all over with first to many and false data and then the correct data.

Was your problem with the eMMC device, processor or driver?

Thomas

Hi,

Well, we don't want to boot from it, at least for now, we have a spi memory.

The problem for us was that we could read all the registers properly from the card and make the initialization process.

The write and read commands never work (single or multiple blocks). The card never answers. Either the chip is not working or something is wrong with my driver. I wrote the driver from a SD card driver that, according to the documentation, is very similar to an eMMC.

Would you be interested in sharing your driver so I coul take a look and compare it with mine ? I would share what I found afterwards.

Have you been able to take a look at the references of the chips you were using ?

Thanks.

I'm sorry to insist Thomas, would you be interested in what I suggested above ?

That would be great if I could get this eMMC chip over with !

Maxime

Sorry for not getting back to you, but we have been a bit busy. We have however solved our problem which related to at least 1 missing pull-up resistor on D0. It seems that there is some intermediate signalling while not transferring data to or from the eMMC. I still haven't found a description of this anywhere, but I suppose it must be described somewhere.

But by just enabling the internal pull-up resistors on the data bus everything suddenly worked (we actually thought we had done that). This may be a note worth in TI's documentation.