qspi_flash example broken (at least on ESP32-S2 - most probably for all chips) since b5f0246

[bjoernQ]

Bug description

Since b5f0246

Output from the example is

[21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21]
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

While it should be

[48, 65, 6c, 6c, 6f, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff]
Hello!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!................................................................

To Reproduce

Run the qspi_flash example with an SPI flash chip attached

Environment

Tested on ESP32-S2 but most probably broken for all chips.

(We should also make sure to improve our HIL-tests to catch this)

[bugadani]

Thanks, I will take care of this.

[bugadani]

I'm somewhat confused here about what's happening. Are you sure you were checking newest main? You switched the buffer sizes back to what they are supposed to be in #2139 and since then, dma_rx_buf should be 320 bytes long. However, in the OP you pasted a 256 byte long output. (And while I don't yet have a flash chip attached to my S2, I get 320 bytes of 0's).

[bjoernQ]

I'm somewhat confused here about what's happening. Are you sure you were checking newest main? You switched the buffer sizes back to what they are supposed to be in #2139 and since then, dma_rx_buf should be 320 bytes long. However, in the OP you pasted a 256 byte long output. (And while I don't yet have a flash chip attached to my S2, I get 320 bytes of 0's).

I did git bisect and the "good" output was the last good output, the "bad" was the last bad output I have seen. I can repeat the tests

[bugadani]

If it isn't a huge time sink, please paste the output of the example from the main branch. It shouldn't be 256 bytes, that's for sure.

[bjoernQ]

[88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, 88, 98, 88, 89, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff, ff]
................................................................................................................................................................................................................................................................................................................................

commit 3bb49b0