AM6421: OSPI Communication Issue

Ryan LIU

Prodigy 140 points

Part Number: AM6421

Tool/software:

Using the interface OSPI_writeIndirect.

Waveform diagram as follows, yellow is CS, green is clk.

Under 10M state, normal, waveform is as follows.

Under 50M state, CS and clk will be abnormal in the second half, waveform is as follows.

4 months ago

+1 Vaibhav Kumar 4 months ago

TI__Mastermind 34254 points

Hi Ryan,

Ryan LIU said:
Under 50M state, CS and clk will be abnormal in the second half, waveform is as follows.

Do this lead to write operations getting aborted? Can you see if the write operation happened correctly?

Regards,

Vaibhav

0 Ryan LIU 4 months ago in reply to Vaibhav Kumar

Prodigy 140 points

Hi Vaibhav,

The operation did not terminate, it seems that the frames were segmented, and the data is correct.

Ryan

0 Ryan LIU 4 months ago in reply to Vaibhav Kumar

Prodigy 140 points

Hi Vaibhav,

At high speeds, when reading data, the waveform appears to be split into two on the display. Is this a normal phenomenon?

As shown in the figure above, this is the clock waveform of one write operation, and we can see that there are two write enable and two write data writes.

Ryan

0 Vaibhav Kumar 4 months ago

TI__Mastermind 34254 points

Ryan LIU said:
Under 50M state, CS and clk will be abnormal in the second half, waveform is as follows.

In this case if the Phy is enabled then the clock frequency is expected to be clock frequency/divider during OSPI initial setting.

But after this, when reading values with Phy enabled the clock frequency will be just clock frequency and the divider will be bypassed.

Processors

Processors forum

AM6421: OSPI Communication Issue