From: johannes.goede@oss.qualcomm.com
To: Loic Poulain <loic.poulain@oss.qualcomm.com>,
Bryan O'Donoghue <bryan.odonoghue@linaro.org>,
Vladimir Zapolskiy <vladimir.zapolskiy@linaro.org>,
Mauro Carvalho Chehab <mchehab@kernel.org>,
Kees Cook <kees@kernel.org>,
"Gustavo A. R. Silva" <gustavoars@kernel.org>,
Rob Herring <robh@kernel.org>,
Krzysztof Kozlowski <krzk+dt@kernel.org>,
Conor Dooley <conor+dt@kernel.org>,
Bjorn Andersson <andersson@kernel.org>,
Konrad Dybcio <konradybcio@kernel.org>
Cc: linux-kernel@vger.kernel.org, linux-media@vger.kernel.org,
linux-arm-msm@vger.kernel.org, linux-hardening@vger.kernel.org,
devicetree@vger.kernel.org
Subject: Re: [PATCH v5 3/5] media: uapi: Add CAMSS OPE ISP configuration definition
Date: Wed, 29 Jul 2026 15:27:49 +0200 [thread overview]
Message-ID: <77f79a27-6a19-4879-b302-e7b42faaaced@oss.qualcomm.com> (raw)
In-Reply-To: <20260724-camss-isp-ope-v5-3-e70ad4fa39ce@oss.qualcomm.com>
Hi Loic, et.al,
On 24-Jul-26 14:42, Loic Poulain wrote:
> Add the uapi header camss-ope-config.h defining the ISP parameter
> structures used by the CAMSS Offline Processing Engine (OPE) driver.
> This includes structures for white balance, chroma enhancement and
> color correction configuration.
>
> Signed-off-by: Loic Poulain <loic.poulain@oss.qualcomm.com>
> ---
> include/uapi/linux/camss-ope-config.h | 160 ++++++++++++++++++++++++++++++++++
> 1 file changed, 160 insertions(+)
>
> diff --git a/include/uapi/linux/camss-ope-config.h b/include/uapi/linux/camss-ope-config.h
> new file mode 100644
> index 0000000000000000000000000000000000000000..800e6ef1dba218ca33b565e1e51d6647f02f8753
> --- /dev/null
> +++ b/include/uapi/linux/camss-ope-config.h
> @@ -0,0 +1,160 @@
> +/* SPDX-License-Identifier: GPL-2.0-only WITH Linux-syscall-note */
> +/*
> + * Qualcomm CAMSS Offline Processing Engine (OPE) ISP parameters UAPI
> + *
> + * Uses the generic V4L2 extensible ISP parameters buffer format defined in
> + * <uapi/linux/media/v4l2-isp.h>.
> + *
> + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries.
> + */
> +
> +#ifndef _UAPI_LINUX_CAMSS_OPE_CONFIG_H
> +#define _UAPI_LINUX_CAMSS_OPE_CONFIG_H
> +
> +#include <linux/types.h>
> +#include <linux/media/v4l2-isp.h>
> +
> +/**
> + * enum camss_ope_params_block_type - CAMSS ISP parameter block identifiers
> + *
> + * Each value identifies one ISP processing block. The value is placed in
> + * the @type field of &struct v4l2_isp_params_block_header.
> + */
> +enum camss_ope_params_block_type {
> + CAMSS_OPE_PARAMS_WB_GAIN = 1,
> + CAMSS_OPE_PARAMS_CHROMA_ENHAN = 2,
> + CAMSS_OPE_PARAMS_COLOR_CORRECT = 3,
> +};
I don't think we want this enum to be OPE specific. The whole idea behind
the extensible ISP parameters is that there is a single set of parameter
definitions spanning multiple ISP generations.
There is going to be a mirroring patch for libcamera defining some C++
magic around this header:
https://patchwork.libcamera.org/patch/27430/
and this will be used in the camss libcamera IPA and we don't want to
have one version of the libcamera wrapper per ISP "generation".
So I think this should all be moved back to a generic camss-config.h
(or maybe camss-parameters.h ?) and then if for the planned Hamoa /
Monoca inline ISP work of we need say a new version of CAMSS_PARAMS_WB_GAIN
just add a CAMSS_OPE_PARAMS_WB_GAIN_V2 for that.
That is the whole idea behind the extensible params, otherwise libcamera
will effectively need to have 1 IPA per ISP generation / type which is
not what we want.
Regards,
Hans
> +
> +/**
> + * struct camss_ope_params_wb_gain - White Balance gains
> + *
> + * Implements the CLC_WB pipeline module. The pipeline applies three
> + * sequential operations per channel:
> + * 1. Subtract sub-offset (black-level subtraction)
> + * 2. Multiply by gain (colour balance)
> + * 3. Add add-offset (output pedestal)
> + *
> + * Gains are 15uQ10 (15-bit unsigned, 10 fractional bits). Offsets
> + * are 16-bit unsigned, normalised to full input scale (65535 = 1.0)
> + *
> + * @header: block header; @header.type = CAMSS_OPE_PARAMS_WB_GAIN
> + * @g_gain: green channel gain (15uQ10, 1024 = 1.0)
> + * @b_gain: blue channel gain (15uQ10, 1024 = 1.0)
> + * @r_gain: red channel gain (15uQ10, 1024 = 1.0)
> + * @g_sub: green sub-offset, subtracted before gain (16u)
> + * @b_sub: blue sub-offset, subtracted before gain (16u)
> + * @r_sub: red sub-offset, subtracted before gain (16u)
> + * @g_add: green add-offset, added after gain (16u)
> + * @b_add: blue add-offset, added after gain (16u)
> + * @r_add: red add-offset, added after gain (16u)
> + */
> +struct camss_ope_params_wb_gain {
> + struct v4l2_isp_params_block_header header;
> + __u16 g_gain;
> + __u16 b_gain;
> + __u16 r_gain;
> + __u16 g_sub;
> + __u16 b_sub;
> + __u16 r_sub;
> + __u16 g_add;
> + __u16 b_add;
> + __u16 r_add;
> + __u16 _pad[3];
> +} __attribute__((aligned(8)));
> +
> +/**
> + * struct camss_ope_params_chroma_enhan - RGB to YUV colour transfer matrix
> + *
> + * Implements the CLC_CHROMA_ENHAN pipeline module. All coefficients are
> + * signed 12-bit fixed-point Q3.8 (range roughly -8.0 to +7.996).
> + *
> + * RGB2Y - Luma (Y) coefficients
> + * Y = v0 * R + v1 * G + v2 * B
> + *
> + * @luma_v0: R-to-Y coefficient (12sQ8)
> + * @luma_v1: G-to-Y coefficient (12sQ8)
> + * @luma_v2: B-to-Y coefficient (12sQ8)
> + * @luma_k: Y output offset (9s, 0 = no offset)
> + *
> + * RGB2Cb - Chroma (Cb) coefficients
> + * Cb = a x ((B - G) + b(R - G)) + KCb
> + * with:
> + * a = ap, when (B-G) + b(R-G) > 0; a = am, when (B-G) + b(R-G) ≤ 0;
> + * b = bp when (R-G) > 0; b = bm when (R-G) ≤ 0
> + *
> + * @coeff_ap: Cb positive coefficient (12sQ8)
> + * @coeff_am: Cb negative coefficient (12sQ8)
> + * @coeff_bp: Cb positive coefficient (12sQ8)
> + * @coeff_bm: Cb negative coefficient (12sQ8)
> + * @kcb: Cb output offset (11s)
> + *
> + * RGB2Cr - Chroma (Cr) coefficients:
> + * Cr = c x ((R - G) + d(B - G)) + KCr
> + * with:
> + * c = cp, when (R-G) + d(B-G) > 0; c = cm, when (R-G) + d(B-G) ≤ 0
> + * d = dp when (B-G) > 0; d = dm when (B-G) ≤ 0
> + *
> + * @coeff_cp: Cr positive coefficient (12sQ8)
> + * @coeff_cm: Cr negative coefficient (12sQ8)
> + * @coeff_dp: Cr positive coefficient (12sQ8)
> + * @coeff_dm: Cr negative coefficient (12sQ8)
> + * @kcr: Cr output offset (11s)
> + *
> + * @header: generic block header; @header.type = CAMSS_OPE_PARAMS_CHROMA_ENHAN
> + */
> +struct camss_ope_params_chroma_enhan {
> + struct v4l2_isp_params_block_header header;
> + __u16 luma_v0;
> + __u16 luma_v1;
> + __u16 luma_v2;
> + __u16 luma_k;
> + __u16 coeff_ap;
> + __u16 coeff_am;
> + __u16 coeff_bp;
> + __u16 coeff_bm;
> + __u16 coeff_cp;
> + __u16 coeff_cm;
> + __u16 coeff_dp;
> + __u16 coeff_dm;
> + __u16 kcb;
> + __u16 kcr;
> + __u16 _pad[2];
> +} __attribute__((aligned(8)));
> +
> +/**
> + * struct camss_ope_params_color_correct - colour correction matrix
> + *
> + * Implements the CLC_CC pipeline module. The matrix computes:
> + * Out_ch0 (G) = a0*G + b0*B + c0*R + k0
> + * Out_ch1 (B) = a1*G + b1*B + c1*R + k1
> + * Out_ch2 (R) = a2*G + b2*B + c2*R + k2
> + *
> + * @header: block header; @header.type = CAMSS_OPE_PARAMS_COLOR_CORRECT
> + * @a: G-input coefficients per output channel (12s;
> + * a[0]=Out_G, a[1]=Out_B, a[2]=Out_R)
> + * @b: B-input coefficients (12s)
> + * @c: R-input coefficients (12s)
> + * @k: per-output-channel offsets (typically 9s effective)
> + * @qfactor: Q-format selector (2u):
> + * 0 = 12sQ7 (range ~-16.0 .. +15.992)
> + * 1 = 12sQ8 (range ~-8.0 .. +7.996)
> + * 2 = 12sQ9 (range ~-4.0 .. +3.998)
> + * 3 = 12sQ10 (range ~-2.0 .. +1.999)
> + */
> +struct camss_ope_params_color_correct {
> + struct v4l2_isp_params_block_header header;
> + __u16 a[3];
> + __u16 b[3];
> + __u16 c[3];
> + __u16 k[3];
> + __u16 qfactor;
> + __u16 _pad[3];
> +} __attribute__((aligned(8)));
> +
> +#define CAMSS_OPE_PARAMS_MAX_PAYLOAD \
> + (sizeof(struct camss_ope_params_wb_gain) +\
> + sizeof(struct camss_ope_params_chroma_enhan) +\
> + sizeof(struct camss_ope_params_color_correct))
> +
> +#endif /* _UAPI_LINUX_CAMSS_OPE_CONFIG_H */
>
next prev parent reply other threads:[~2026-07-29 13:27 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-24 12:42 [PATCH v5 0/5] media: qcom: camss: CAMSS Offline Processing Engine support Loic Poulain
2026-07-24 12:42 ` [PATCH v5 1/5] media: qcom: camss: Add V4L2 meta format for CAMSS ISP parameters Loic Poulain
2026-07-29 18:28 ` Gjorgji Rosikopulos (Consultant)
2026-07-30 11:44 ` Loic Poulain
2026-07-30 11:57 ` Bryan O'Donoghue
2026-07-30 12:34 ` Loic Poulain
2026-07-30 13:15 ` Gjorgji Rosikopulos (Consultant)
2026-07-24 12:42 ` [PATCH v5 2/5] dt-bindings: media: qcom: Add CAMSS Offline Processing Engine (OPE) Loic Poulain
2026-07-24 12:42 ` [PATCH v5 3/5] media: uapi: Add CAMSS OPE ISP configuration definition Loic Poulain
2026-07-29 13:27 ` johannes.goede [this message]
2026-07-29 18:31 ` Gjorgji Rosikopulos (Consultant)
2026-07-24 12:42 ` [PATCH v5 4/5] media: qcom: camss: Add CAMSS Offline Processing Engine driver Loic Poulain
2026-07-29 19:53 ` Gjorgji Rosikopulos (Consultant)
2026-08-21 6:34 ` Nihal Kumar Gupta
2026-07-24 12:42 ` [PATCH v5 5/5] arm64: dts: qcom: agatti: Add OPE node Loic Poulain
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=77f79a27-6a19-4879-b302-e7b42faaaced@oss.qualcomm.com \
--to=johannes.goede@oss.qualcomm.com \
--cc=andersson@kernel.org \
--cc=bryan.odonoghue@linaro.org \
--cc=conor+dt@kernel.org \
--cc=devicetree@vger.kernel.org \
--cc=gustavoars@kernel.org \
--cc=kees@kernel.org \
--cc=konradybcio@kernel.org \
--cc=krzk+dt@kernel.org \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-hardening@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=loic.poulain@oss.qualcomm.com \
--cc=mchehab@kernel.org \
--cc=robh@kernel.org \
--cc=vladimir.zapolskiy@linaro.org \
/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
all inboxes | Powered by JetHome®