Audio problem on custom board

[andrewparlane]

Hi all,

I'm trying to get audio working on our new custom board. Starting off testing with aplay and a 16 bit .wav.

Unfortunately it fails with:

playback write error (DMA or IRQ trouble?)

aplay: pcm_write:1939: write error: Input/output error

The board is modelled after the reference designs and uses the SGTL5000 connected via I2S from the output of AUDMUX3.

When I scope the I2S clock line (CSI0_DAT4) it does not wiggle.

Looking at the pinmux, I can confirm that those pins are configured to alt 4. (AUDMX3). When I configure them to be GPIO I can wiggle the clock line through sysfs gpio150 - gpio5_22.

This worked fine on the previous rev of the board which was a solo core board.

My device tree is attached.

I've seen a few people posting about this same error message, but they either have no solutions or are due to a different issue.

Using linux 3.10.17

I also checked that the SDMA firmware had been loaded correctly.

imx-sdma 20ec000.sdma: loaded firmware 1.1

imx-sdma 20ec000.sdma: initialized

Any suggestions?

Thanks,

Andrew Parlane

Original Attachment has been moved to: device_tree_rev3_quad.dts.zip

[andrewparlane]

It looks like the SGTL5000 is "crashing" where the I2C clock line is held low.

See SGTL5000 hangs I2C bus after reset

That issues seems to be the same as I'm having but instead of requiring a reboot, I require a few I2C writes. For now I'm assuming this is due to a fault we fonud with the board earlier, meaning the I2C lines are powered from 2.5V logic in the IMX and pulled up to 3.3V externally. Somehow (not exactly sure how) we get 2.6V on the actual net, while the VDDIO pin is connected to 3.3V. Since the board is about to be respun anyway, I'm going to ignore this issue until the next rev.

Thanks,

Andy

[igorpadykov]

Hi Andrew

you are right, if I2C lines are powered from 2.5V logic in the IMX, then pulling

them up to 3.3V externally may cause problems. Regarding SDTL5000,

recommended to use external supply in the system or a separate regulator

to supply VDDD according to SGTL5000ER SGTL5000 Silicon Errata - Errata

Best regards

igor

-----------------------------------------------------------------------------------------------------------------------

Note: If this post answers your question, please click the Correct Answer button. Thank you!

-----------------------------------------------------------------------------------------------------------------------

[andrewparlane]

Hi Igor,

Thanks for your reply.

We do have an external regulator on VDDD separate to the one on VDDIO, so that shound be fine.

I'm assuming it's to do with the I2C lines, and hoping the next rev will fix this.

Andy