From: Jacopo Mondi <jacopo.mondi@ideasonboard.com>
To: "Niklas Söderlund" <niklas.soderlund+renesas@ragnatech.se>
Cc: Jacopo Mondi <jacopo.mondi@ideasonboard.com>,
Jai Luthra <jai.luthra+renesas@ideasonboard.com>,
Mauro Carvalho Chehab <mchehab@kernel.org>,
Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>,
Laurent Pinchart <laurent.pinchart@ideasonboard.com>,
linux-media@vger.kernel.org, linux-renesas-soc@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [v8 02/14] media: rppx1: Add framework to support Dreamchip RPPX1 ISP
Date: Wed, 3 Jun 2026 10:27:46 +0200 [thread overview]
Message-ID: <ah_kud7fRe_DN7Mx@zed> (raw)
In-Reply-To: <20260516151718.GV332351@ragnatech.se>
Hi Niklas
On Sat, May 16, 2026 at 05:17:18PM +0200, Niklas Söderlund wrote:
> Hello Jacopo,
>
[snip]
> > > diff --git a/drivers/media/platform/dreamchip/rppx1/rppx1_shrp.c b/drivers/media/platform/dreamchip/rppx1/rppx1_shrp.c
> > > new file mode 100644
> > > index 000000000000..5bec022e8f05
> > > --- /dev/null
> > > +++ b/drivers/media/platform/dreamchip/rppx1/rppx1_shrp.c
> > > @@ -0,0 +1,64 @@
> > > +// SPDX-License-Identifier: GPL-2.0
> > > +/*
> > > + * Copyright 2025 Renesas Electronics Corp.
> > > + * Copyright 2025 Niklas Söderlund <niklas.soderlund@ragnatech.se>
> > > + */
> > > +
> > > +#include "rpp_module.h"
> > > +
> > > +#define SHRPCNR_VERSION_REG 0x0000
> > > +
> > > +#define SHRPCNR_CTRL_REG 0x0004
> > > +#define SHRPCNR_CTRL_CAD_EN BIT(3)
> > > +#define SHRPCNR_CTRL_DESAT_EN BIT(2)
> > > +#define SHRPCNR_CTRL_CNR_EN BIT(1)
> > > +#define SHRPCNR_CTRL_SHARPEN_EN BIT(0)
> > > +
> > > +#define SHRPCNR_PARAM_REG 0x0008
> > > +#define SHRPCNR_PARAM_SHARP_FACTOR_MASK GENMASK(19, 12)
> > > +#define SHRPCNR_PARAM_CORING_THR_MASK GENMASK(11, 0)
> > > +
> > > +#define SHRPCNR_MAT_1_REG 0x000c
> > > +#define SHRPCNR_MAT_2_REG 0x0010
> > > +#define SHRPCNR_CLB_LINESIZE_REG 0x0014
> > > +#define SHRPCNR_YUV2RGB_CCOR_COEFF_0_REG 0x0018
> > > +#define SHRPCNR_YUV2RGB_CCOR_COEFF_1_REG 0x001c
> > > +#define SHRPCNR_YUV2RGB_CCOR_COEFF_2_REG 0x0020
> > > +#define SHRPCNR_YUV2RGB_CCOR_COEFF_3_REG 0x0024
> > > +#define SHRPCNR_YUV2RGB_CCOR_COEFF_4_REG 0x0028
> > > +#define SHRPCNR_YUV2RGB_CCOR_COEFF_5_REG 0x002c
> > > +#define SHRPCNR_YUV2RGB_CCOR_COEFF_6_REG 0x0030
> > > +#define SHRPCNR_YUV2RGB_CCOR_COEFF_7_REG 0x0034
> > > +#define SHRPCNR_YUV2RGB_CCOR_COEFF_8_REG 0x0038
> > > +#define SHRPCNR_YUV2RGB_CCOR_OFFSET_R_REG 0x003c
> > > +#define SHRPCNR_YUV2RGB_CCOR_OFFSET_G_REG 0x0040
> > > +#define SHRPCNR_YUV2RGB_CCOR_OFFSET_B_REG 0x0044
> > > +
> > > +#define SHRPCNR_CNR_THRES_REG 0x0048
> > > +#define SHRPCNR_CNR_THRES_CNR_THRES_CR_MASK GENMASK(27, 16)
> > > +#define SHRPCNR_CNR_THRES_CNR_THRES_CB_MASK GENMASK(11, 0)
> > > +
> > > +#define SHRPCNR_CRED_THRES_REG 0x004c
> > > +#define SHRPCNR_CRED_SLOPE_REG 0x0050
> > > +#define SHRPCNR_CAD_RESTORE_LVL_REG 0x0054
> > > +#define SHRPCNR_CAD_THRESH_V_UNEG_REG 0x0058
> > > +#define SHRPCNR_CAD_THRESH_V_UPOS_REG 0x005c
> > > +#define SHRPCNR_CAD_THRESH_U_REG 0x0060
> > > +
> > > +static int rppx1_shrp_probe(struct rpp_module *mod)
> > > +{
> > > + /* Version check. */
> > > + switch (rpp_module_read(mod, SHRPCNR_VERSION_REG)) {
> > > + case 2:
> > > + mod->info.shrp.colorbits = 12;
> > > + break;
> > > + default:
> > > + return -EINVAL;
> > > + }
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +const struct rpp_module_ops rppx1_shrp_ops = {
> > > + .probe = rppx1_shrp_probe,
> > > +};
> >
> > There are quite some modules here that have no corresponding user in
> > libcamera and which are not exercized.
> >
> > As long as we don't have a userspace user, I would refrein from adding
> > them to the driver to avoid committing to a uAPI before anyone
> > actually uses it.
>
> I do have user-space that exercise each uAPI exposed by this series in
> the form of unit-tests. But I know you feel strongly about libcamera
> support so I will drop the uAPI from the ones not used by libcamera from
> this series.
>
More than libcamera itself, I think it's safer to make sure we can
exercize the user API before committing to them. The best way to do
that is to have an algorithm in libcamera indeed
> Before the switch to the RPPX1 dedicated format the RkISP1 IPA in
> libcamera did use more of the enabled blocks together with some sensors.
yeah, but thinking about denoise in example, I wouldn't fully commit
to that yet as we know that algorithm is just very basic
> So for my core use-cases dropping them will not lose me much
> functionality. And as this series now structures uAPI together with the
> code moving modules in out is easy :-)
>
> I will drop the following blocks:
>
> - BLS - statistics reporting.
> - DB
> - BD
>
> >
> > If you want to keep the kernel module, fine, but no ABI for blocks
> > without a userspace user.
>
> The probe and static init config (that instructs the ISP pipeline to
> bypass the module) I will keep. I will drop all logic and uAPU
> structures.
Thank you!
>
> >
> > > diff --git a/drivers/media/platform/dreamchip/rppx1/rppx1_wbmeas.c b/drivers/media/platform/dreamchip/rppx1/rppx1_wbmeas.c
> > > new file mode 100644
> > > index 000000000000..3d197d914d07
> > > --- /dev/null
> > > +++ b/drivers/media/platform/dreamchip/rppx1/rppx1_wbmeas.c
> > > @@ -0,0 +1,61 @@
> > > +// SPDX-License-Identifier: GPL-2.0
> > > +/*
> > > + * Copyright 2025 Renesas Electronics Corp.
> > > + * Copyright 2025 Niklas Söderlund <niklas.soderlund@ragnatech.se>
> > > + */
> > > +
> > > +#include "rpp_module.h"
> > > +
> > > +#define AWB_MEAS_VERSION_REG 0x0000
> > > +
> > > +#define AWB_MEAS_PROP_REG 0x0004
> > > +#define AWB_MEAS_PROP_MEAS_MODE_RGB BIT(16) /* 0: YCbCr 1: RGB */
> > > +#define AWB_MEAS_PROP_YMAX BIT(2)
> > > +#define AWB_MEAS_PROP_AWB_MODE_ON BIT(1)
> > > +
> > > +#define AWB_MEAS_H_OFFS_REG 0x0008
> > > +#define AWB_MEAS_V_OFFS_REG 0x000c
> > > +#define AWB_MEAS_H_SIZE_REG 0x0010
> > > +#define AWB_MEAS_V_SIZE_REG 0x0014
> > > +#define AWB_MEAS_FRAMES_REG 0x0018
> > > +#define AWB_MEAS_REF_CB_MAX_B_REG 0x001c
> > > +#define AWB_MEAS_REF_CR_MAX_R_REG 0x0020
> > > +#define AWB_MEAS_MAX_Y_REG 0x0024
> > > +#define AWB_MEAS_MIN_Y_MAX_G_REG 0x0028
> > > +#define AWB_MEAS_MAX_CSUM_REG 0x002c
> > > +#define AWB_MEAS_MIN_C_REG 0x0030
> > > +#define AWB_MEAS_WHITE_CNT_REG 0x0034
> > > +#define AWB_MEAS_MEAN_Y_G_REG 0x0038
> > > +#define AWB_MEAS_MEAN_CB_B_REG 0x003c
> > > +#define AWB_MEAS_MEAN_CR_R_REG 0x0040
> > > +
> > > +#define AWB_MEAS_CCOR_COEFF_NUM 9
> > > +#define AWB_MEAS_CCOR_COEFF_REG(n) (0x0044 + (4 * (n)))
> > > +
> > > +#define AWB_MEAS_CCOR_OFFSET_R_REG 0x0068
> > > +#define AWB_MEAS_CCOR_OFFSET_G_REG 0x006c
> > > +#define AWB_MEAS_CCOR_OFFSET_B_REG 0x0070
> > > +
> > > +static int rppx1_wbmeas_probe(struct rpp_module *mod)
> > > +{
> > > + /* Version check. */
> > > + switch (rpp_module_read(mod, AWB_MEAS_VERSION_REG)) {
> > > + case 1:
> > > + mod->info.wbmeas.colorbits = 8;
> > > + break;
> > > + case 2:
> > > + mod->info.wbmeas.colorbits = 20;
> > > + break;
> > > + case 3:
> > > + mod->info.wbmeas.colorbits = 24;
> > > + break;
> > > + default:
> > > + return -EINVAL;
> > > + }
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +const struct rpp_module_ops rppx1_wbmeas_ops = {
> > > + .probe = rppx1_wbmeas_probe,
> > > +};
> > > diff --git a/drivers/media/platform/dreamchip/rppx1/rppx1_xyz2luv.c b/drivers/media/platform/dreamchip/rppx1/rppx1_xyz2luv.c
> > > new file mode 100644
> > > index 000000000000..73789c48c057
> > > --- /dev/null
> > > +++ b/drivers/media/platform/dreamchip/rppx1/rppx1_xyz2luv.c
> > > @@ -0,0 +1,26 @@
> > > +// SPDX-License-Identifier: GPL-2.0
> > > +/*
> > > + * Copyright 2025 Renesas Electronics Corp.
> > > + * Copyright 2025 Niklas Söderlund <niklas.soderlund@ragnatech.se>
> > > + */
> > > +
> > > +#include "rpp_module.h"
> > > +
> > > +#define XYZ2LUV_VERSION_REG 0x0000
> > > +#define XYZ2LUV_U_REF_REG 0x0004
> > > +#define XYZ2LUV_V_REF_REG 0x0008
> > > +#define XYZ2LUV_LUMA_OUT_FAC_REG 0x000c
> > > +#define XYZ2LUV_CHROMA_OUT_FAC_REG 0x0010
> > > +
> > > +static int rppx1_xyz2luv_probe(struct rpp_module *mod)
> > > +{
> > > + /* Version check. */
> > > + if (rpp_module_read(mod, XYZ2LUV_VERSION_REG) != 4)
> > > + return -EINVAL;
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +const struct rpp_module_ops rppx1_xyz2luv_ops = {
> > > + .probe = rppx1_xyz2luv_probe,
> > > +};
> > > diff --git a/include/media/rppx1.h b/include/media/rppx1.h
> > > new file mode 100644
> > > index 000000000000..cb3470c27ceb
> > > --- /dev/null
> > > +++ b/include/media/rppx1.h
> > > @@ -0,0 +1,34 @@
> > > +/* SPDX-License-Identifier: GPL-2.0 */
> > > +/*
> > > + * Copyright 2025 Renesas Electronics Corp.
> > > + * Copyright 2025 Niklas Söderlund <niklas.soderlund@ragnatech.se>
> > > + */
> > > +#ifndef __MEDIA_DCT_RPPX1_H__
> > > +#define __MEDIA_DCT_RPPX1_H__
> > > +
> > > +#include <linux/v4l2-mediabus.h>
> > > +#include <linux/media/dreamchip/rppx1-config.h>
> > > +
> > > +#include <media/videobuf2-core.h>
> > > +
> > > +struct rppx1;
> > > +
> > > +struct rppx1 *rppx1_create(void __iomem *base, struct device *dev);
> > > +
> > > +void rppx1_destroy(struct rppx1 *rpp);
> > > +
> > > +int rppx1_start(struct rppx1 *rpp, const struct v4l2_mbus_framefmt *input,
> > > + const struct v4l2_mbus_framefmt *hv,
> > > + const struct v4l2_mbus_framefmt *mv);
> > > +
> > > +int rppx1_stop(struct rppx1 *rpp);
> > > +
> > > +bool rppx1_interrupt(struct rppx1 *rpp, u32 *isc);
> > > +
> > > +typedef int (*rppx1_reg_write)(void *priv, u32 offset, u32 value);
> > > +int rppx1_params(struct rppx1 *rpp, struct vb2_buffer *vb, size_t max_size,
> > > + rppx1_reg_write write, void *priv);
> > > +
> > > +void rppx1_stats_fill_isr(struct rppx1 *rpp, u32 isc, void *buf);
> > > +
> > > +#endif /* __MEDIA_DCT_RPPX1_H__ */
> > > diff --git a/include/uapi/linux/media/dreamchip/rppx1-config.h b/include/uapi/linux/media/dreamchip/rppx1-config.h
> > > new file mode 100644
> > > index 000000000000..26627be6f483
> > > --- /dev/null
> > > +++ b/include/uapi/linux/media/dreamchip/rppx1-config.h
> >
> > As said, I don't think it makes much sense to add an almost empty uapi
> > file.
> >
> > I would add it in one commit.
> >
> > Please retain Jai's authorship and add my Co-developed-by tag as I
> > rewrote most of the documentation and reworked most blocks.
>
> As discussed above, I really like adding the uAPI together with the
> logic. I will retain the scaffolding as a separate commit from Jai and
> add the Co-developed tags.
Fine with me then!
Thanks
j
>
> >
> > > @@ -0,0 +1,66 @@
> > > +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
> > > +/*
> > > + * Dreamchip RPP-X1 ISP Driver - Userspace API
> > > + *
> > > + * Copyright (C) 2026 Renesas Electronics Corp.
> > > + * Copyright (C) 2026 Ideas on Board Oy
> > > + * Copyright (C) 2026 Ragnatech AB
> > > + */
> > > +
> > > +#ifndef __UAPI_RPP_X1_CONFIG_H
> > > +#define __UAPI_RPP_X1_CONFIG_H
> > > +
> > > +#include <linux/types.h>
> > > +#include <linux/media/v4l2-isp.h>
> > > +
> > > +/**
> > > + * struct rppx1_window - Measurement window
> > > + *
> > > + * RPP-X1 measurement window. Different blocks use a window or multiple
> > > + * windows for measurement purposes. This defines a common type for all of
> > > + * them. The number of relevant bits depends on the block where the window is
> > > + * used and is specified in the per-block description
> > > + *
> > > + * @h_offs: horizontal offset from the left of the frame in pixels
> > > + * @v_offs: vertical offset from the top of the frame in pixels
> > > + * @h_size: horizontal size of the window in pixels
> > > + * @v_size: vertical size of the window in pixels
> > > + */
> > > +struct rppx1_window {
> > > + __u16 h_offs;
> > > + __u16 v_offs;
> > > + __u16 h_size;
> > > + __u16 v_size;
> > > +};
> > > +
> > > +/* ---------------------------------------------------------------------------
> > > + * Parameter Structures
> > > + *
> > > + * Native RPP-X1 precision. Fields use __u32 where the hardware provides
> > > + * wider-than-8-bit results.
> >
> > I think you could drop the first part: of course the RPP-X1 uAPI
> > header uses the RPP-X1 precision. You could add:
> >
> > The same ISP block might be instantiated in multiple pipeliness
> > and operate on a different bitdepth/precision. For fields of
> > varying length among different instances of the same block, use
> > a data type that can accommodate the larger bitdepth/precision.
> >
> > Or something similar
>
> Thanks, this should of course be updated now that we have a native RPPX1
> format.
>
> >
> > > + */
> > > +
> > > +/**
> > > + * RPPX1_PARAMS_MAX_SIZE - Maximum size of all RPP-X1 parameter blocks
> > > + *
> > > + * Some types are reported twice as the same block might be instantiated in
> > > + * multiple pipes.
> > > + */
> > > +#define RPPX1_PARAMS_MAX_SIZE 0
> > > +
> > > +/* ---------------------------------------------------------------------------
> > > + * Statistics Structures
> > > + *
> > > + * Native RPP-X1 precision. Fields use __u32 where the hardware provides
> > > + * wider-than-8-bit results.
> >
> > Same here
>
> Ditto.
>
> >
> > > + */
> > > +
> > > +/**
> > > + * RPPX1_STATS_MAX_SIZE - Maximum size of all RPP-X1 statistics
> > > + *
> > > + * Some types are reported twice as the same block might be instantiated in
> > > + * multiple pipes.
> > > + */
> > > +#define RPPX1_STATS_MAX_SIZE 0
> > > +
> > > +#endif /* __UAPI_RPP_X1_CONFIG_H */
> > > --
> > > 2.54.0
> > >
>
> --
> Kind Regards,
> Niklas Söderlund
>
next prev parent reply other threads:[~2026-06-03 8:27 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-05-04 1:05 [v8 00/14] media: Add support for R-Car ISP using Dreamchip RPPX1 ISP Niklas Söderlund
2026-05-04 1:05 ` [v8 01/14] media: Add RPP_X1_PARAMS and RPP_X1_STATS meta formats Niklas Söderlund
2026-05-06 7:45 ` Jacopo Mondi
2026-05-04 1:05 ` [v8 02/14] media: rppx1: Add framework to support Dreamchip RPPX1 ISP Niklas Söderlund
2026-05-06 8:53 ` Jacopo Mondi
2026-05-16 15:17 ` Niklas Söderlund
2026-06-03 8:27 ` Jacopo Mondi [this message]
2026-05-04 1:05 ` [v8 03/14] media: rcar-isp: Add support for ISPCORE Niklas Söderlund
2026-05-06 15:02 ` Jacopo Mondi
2026-05-16 20:16 ` Niklas Söderlund
2026-06-03 9:11 ` Jacopo Mondi
2026-06-03 12:45 ` Niklas Söderlund
2026-05-04 1:05 ` [v8 04/14] media: rppx1: wbmeas: Add support for white balance measurement Niklas Söderlund
2026-05-06 13:58 ` Antoine Bouyer
2026-05-06 14:22 ` Jacopo Mondi
2026-05-16 20:27 ` Niklas Söderlund
2026-05-04 1:05 ` [v8 05/14] media: rppx1: awbg: Add support for white balance gain settings Niklas Söderlund
2026-05-04 1:05 ` [v8 06/14] media: rppx1: exm: Add support for exposure measurement Niklas Söderlund
2026-05-04 1:05 ` [v8 07/14] media: rppx1: hist: Add support histogram measurement Niklas Söderlund
2026-05-06 15:51 ` Jacopo Mondi
2026-05-16 20:31 ` Niklas Söderlund
2026-05-04 1:05 ` [v8 08/14] media: rppx1: bls: Add support for black level compensation Niklas Söderlund
2026-05-06 15:25 ` Jacopo Mondi
2026-05-16 20:44 ` Niklas Söderlund
2026-06-03 9:35 ` Jacopo Mondi
2026-05-04 1:05 ` [v8 09/14] media: rppx1: ccor: Add support for color correction matrix Niklas Söderlund
2026-05-04 1:05 ` [v8 10/14] media: rppx1: lsc: Add support for lens shade correction Niklas Söderlund
2026-05-04 1:05 ` [v8 11/14] media: rppx1: ga: Add support for gamma out correction Niklas Söderlund
2026-05-04 1:05 ` [v8 12/14] media: rppx1: db: Add support for debayering filters Niklas Söderlund
2026-05-06 15:54 ` Jacopo Mondi
2026-05-16 20:47 ` Niklas Söderlund
2026-05-04 1:05 ` [v8 13/14] media: rppx1: bd: Add support for bilateral denoising Niklas Söderlund
2026-05-04 1:05 ` [v8 14/14] media: rppx1: lin: Add support for gamma sensor linearization Niklas Söderlund
2026-05-06 15:57 ` Jacopo Mondi
2026-05-16 20:56 ` Niklas Söderlund
2026-05-06 12:19 ` [v8 00/14] media: Add support for R-Car ISP using Dreamchip RPPX1 ISP Geert Uytterhoeven
2026-05-06 12:29 ` Niklas Söderlund
2026-05-06 12:49 ` Jacopo Mondi
2026-05-06 12:56 ` Geert Uytterhoeven
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ah_kud7fRe_DN7Mx@zed \
--to=jacopo.mondi@ideasonboard.com \
--cc=jai.luthra+renesas@ideasonboard.com \
--cc=kuninori.morimoto.gx@renesas.com \
--cc=laurent.pinchart@ideasonboard.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=linux-renesas-soc@vger.kernel.org \
--cc=mchehab@kernel.org \
--cc=niklas.soderlund+renesas@ragnatech.se \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox