Other Parts Discussed in Thread: OMAP-L137
I'm using OMAP-L137, Revision 2.0, and spiflash_writer_dsp.out.
I'm trying to spiflash a Winbond W25Q128BV. I've modified spiflash_writer to recognize the part with the correct number of sectors, etc. Yesterday I successfully flashed both dspais, armubl, and u-boot using spiflash_writer_dsp.out, but remembered (too late) that I would need to modify u-boot to recognize the new Winbond part as well. I've now rebuilt u-boot, but find that I can no longer flash it. In fact, although I can flash dspais, I can't successfully flash armubl either. The errors I'm seeing occur when I try to verify that the programming was successful. The received bytes have low bits that should be high for about 1/4th of the received bytes. The bit errors are "stubborn". That is, they occur in the same pattern and at the same locations for all programming attempts. (The positions changed from yesterday to today, but not from run-to-run today or from run-to-run yesterday.)
I've looked at power, ground, clock, timing of clock/data (MOSI/MISO). All look fine for the samples that fit on the scope. The same failures I'm seeing in spiflash_writer appear on the Beagle display. That is, I've added printfs to spiflash_writer and the transmit data agrees with Beagle and the receive data agrees with Beagle. Example:
transmit (write) data: 0d 00 a0 e1 b4 01 00 eb 00 00...
receive (read) data: 08 00 a0 41 b4 01 00 e9 00 00...
If any of you has seen something similar and know what might be causing the failure, please advise me. Thanks.
