iMX8QM Monochrome Grey RAW 8 Sensor support

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

iMX8QM Monochrome Grey RAW 8 Sensor support

876 Views
Dharanitharan
Contributor II

Hi,

We are evaluating AR1335 monochrome sensor with iMX8QM and added GRAY8 format support in imx-isi-cap.c of platform driver side. While trying to stream gray8 format using gstreamer or v4l2-ctl, we are receiving corrupted data. While checking the single image with RAW viewer, the LSB 4 bites of a byte is always padded with zero.

For your reference, the platform driver changes and GRAY8 raw image are also added below.

REQUEST:

On our side we have added necessary changes in dtbo and platform driver side. Don’t know whether we are missing anything related to GREY8 format or anything that need to be added for GREY8 format. Waiting for your quick and valuable response.

We use the below setup for development

Toradex Apalis iMX8QM: NXP iMX8QM

Linux kernel :5.4-2.3

Tags (3)
0 Kudos
Reply
2 Replies

368 Views
rumozhizheng
Contributor I

We are also facing same kind of issue with GRAY8(Y8),How did you solve the problem in the end?

Looking forward to your answer!!!

 

0 Kudos
Reply

853 Views
Dharanitharan
Contributor II

Hi,

We are also facing same kind of issue with GRAY10(Y10). We are trying to get GRAY10 format as V4L2_PIX_FMT_Y16 format through kernel changes. By checking the single frame of Y10 with RAW viewer, The LSB 4 Bites and MSB 2 Bites are padded with zero, As it is 16 bites the intermediate 10 bites contains Y10 data.

QUERY:

While taking Y10 data using Y16 format, either MSB 6 bites or LSB 6 bites must be padded with zero. But the padding which is occurred here is bit confusing.

REQUEST:

The issue in GRAY8 and GRAY10 looks similar. The padding of LSB 4 bites seems to be issue. would you please suggest required changes in kernel to resolve this issue?

GRAY10 raw image is added below for your reference.

0 Kudos
Reply