[PATCH v3 00/15] media: microchip-isc: fixes and enhancements

Balakrishnan Sambath posted 15 patches 1 month ago
There is a newer version of this series
.../userspace-api/media/drivers/index.rst     |   1 +
.../media/drivers/microchip-isc.rst           |  71 ++
MAINTAINERS                                   |   1 +
.../platform/microchip/microchip-isc-base.c   | 728 ++++++++++++++++--
.../platform/microchip/microchip-isc-regs.h   |  11 +-
.../media/platform/microchip/microchip-isc.h  |  56 +-
.../microchip/microchip-sama5d2-isc.c         |   2 +-
.../microchip/microchip-sama7g5-isc.c         | 101 ++-
include/linux/atmel-isc-media.h               |  31 +
9 files changed, 894 insertions(+), 108 deletions(-)
create mode 100644 Documentation/userspace-api/media/drivers/microchip-isc.rst
[PATCH v3 00/15] media: microchip-isc: fixes and enhancements
Posted by Balakrishnan Sambath 1 month ago
Bug fixes and feature additions for the Microchip ISC/XISC driver.

Fixes:
 - SBGGR10 Bayer pattern was mapped incorrectly (red/blue swap)
 - WB register fields corrupted by sign extension
 - Race between histogram IRQ and stream stop
 - PM runtime reference leak in AWB work handler

Features:
 - Driver documentation
 - Gamma 1.8/2.4 curves, per-channel gamma LUT
 - Hue/saturation controls for SAMA7G5
 - Color correction matrix controls
 - Grey World AWB with EMA smoothing

Split from v1 per review. Histogram stats follow as Series 2.

Tested on SAMA7G5-EK with IMX219 (RAW10 Bayer capture, AWB, color
controls verified).

Based on v6.19-rc8 (18f7fcd5e69a).

v1: https://lore.kernel.org/linux-media/20251009155251.102472-1-balamanikandan.gunasundar@microchip.com/

v3:
 - Fix bisect failures in patches 9-11 (regmap declaration, gamma LUT
   macro ordering)
 - Fix Fixes: tag in patch 2 (use mainline commit, not staging)
 - Add Co-developed-by for patches co-authored with Balamanikandan Gunasundar

v2:
 - Split series (histogram stats moved to Series 2)
 - Bug fixes first, then features
 - New fixes and features as listed above
 - Commit message cleanups
 - Rebased on v6.19-rc8

Balakrishnan Sambath (15):
  media: microchip-isc: fix SBGGR10 Bayer pattern
  media: microchip-isc: mask WB offset and gain register fields
  media: microchip-isc: fix race condition on stream stop
  media: microchip-isc: fix PM runtime leak in AWB work handler
  media: microchip-isc: add driver documentation
  media: microchip-isc: set SAM9X7 maximum resolution to 2560x1920
  media: microchip-isc: configure DPC and pipeline for SAMA7G5
  media: microchip-isc: add gamma 1.8 and 2.4 correction curves
  media: microchip-isc: add SAMA7G5 hue and saturation controls
  media: microchip-isc: expose color correction matrix as V4L2 controls
  media: microchip-isc: add per-channel gamma LUT controls
  media: microchip-isc: reset pipeline state on kernel AWB enable
  media: microchip-isc: use weighted averages for Grey World AWB
  media: microchip-isc: smooth AWB gains with EMA filter
  media: microchip-isc: scale DPC black level to sensor bit depth

 .../userspace-api/media/drivers/index.rst     |   1 +
 .../media/drivers/microchip-isc.rst           |  71 ++
 MAINTAINERS                                   |   1 +
 .../platform/microchip/microchip-isc-base.c   | 728 ++++++++++++++++--
 .../platform/microchip/microchip-isc-regs.h   |  11 +-
 .../media/platform/microchip/microchip-isc.h  |  56 +-
 .../microchip/microchip-sama5d2-isc.c         |   2 +-
 .../microchip/microchip-sama7g5-isc.c         | 101 ++-
 include/linux/atmel-isc-media.h               |  31 +
 9 files changed, 894 insertions(+), 108 deletions(-)
 create mode 100644 Documentation/userspace-api/media/drivers/microchip-isc.rst

-- 
2.34.1
Re: [PATCH v3 00/15] media: microchip-isc: fixes and enhancements
Posted by Sakari Ailus 4 weeks ago
Hi Balakhrisnan,

On Wed, May 13, 2026 at 12:47:27PM +0530, Balakrishnan Sambath wrote:
> Bug fixes and feature additions for the Microchip ISC/XISC driver.

Please avoid using git send-email --in-reply-to argument when posting new
versions. You could add a reference to the old series using the Message-ID:
header though.

-- 
Regards,

Sakari Ailus