[U-Boot] spi_flash_probe returns zero idcodes and hangs

[Jon Loeliger]

Amit,

If you can, place a SPI protocol analyzer on the bus and trace its transactions.

I recently debugged a similar symptom and discovered that I was not driving
the Tx FIFO consistently enough to maintain the appearance of a single
transaction on the ID request to the flash.

How is your SPI device (flash, right?) being selected?  If via a GPIO, are you
sure you are driving it properly?  If it is directly using a
controller's slave-select
line, you may need to ensure a full Tx FIFO for both the write and response.

HTH,
jdl


On Sat, May 24, 2014 at 12:15 PM, Jagan Teki <jagannadh.teki at gmail.com> wrote:
> On Sat, May 24, 2014 at 10:27 PM, Amit Mahadik <amitmahadik35 at yahoo.com> wrote:
>> Hello All,
>>                 I am using a custom ARM based development board running u-boot from PNOR memory. However, it does not have support for the Micron SPI flash memory (N25Q064A). I have added support for the micron memory in drivers/mtd/spi/stmicro.c file and included the flag CONFIG_SPI_FLASH_STMICRO in the config file and recompiled the u-boot source source. Then, ran it directly through RAM memory.
>> The new u-boot ran from RAM. Now when I run sf probe 0 command I get the following output
>> Got Idcodes 000000: 00 00 00 00  and the code hangs after this.
>
> Looks like your driver itself is not returning any proper idcode.
>
> What id details you added - I think you're using old u-boot no stmicro.c
> in current u-boot, please re-base to master and try.
>
> thanks!
> --
> Jagan.
> _______________________________________________
> U-Boot mailing list
> U-Boot at lists.denx.de
> http://lists.denx.de/mailman/listinfo/u-boot