From patchwork Thu Jan 30 18:14:47 2025 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Milan Zamazal X-Patchwork-Id: 22707 Return-Path: X-Original-To: parsemail@patchwork.libcamera.org Delivered-To: parsemail@patchwork.libcamera.org Received: from lancelot.ideasonboard.com (lancelot.ideasonboard.com [92.243.16.209]) by patchwork.libcamera.org (Postfix) with ESMTPS id F2BADBD808 for ; Thu, 30 Jan 2025 18:15:38 +0000 (UTC) Received: from lancelot.ideasonboard.com (localhost [IPv6:::1]) by lancelot.ideasonboard.com (Postfix) with ESMTP id 99C916857D; Thu, 30 Jan 2025 19:15:38 +0100 (CET) Authentication-Results: lancelot.ideasonboard.com; dkim=pass (1024-bit key; unprotected) header.d=redhat.com header.i=@redhat.com header.b="NwLMytV8"; dkim-atps=neutral Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lancelot.ideasonboard.com (Postfix) with ESMTPS id E73C368575 for ; Thu, 30 Jan 2025 19:15:35 +0100 (CET) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1738260935; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=ru5hpnF/sIFEjQqjEhzEwB1nzy8iIeBwdeGB5YPZpuE=; b=NwLMytV8V6/dH9yTmIHnQmn2T+Pz5jCQEV7Lqp7DYrphKPoQ8IaCejidrLuLzJgwEPgI2Y V1M8ubbzl8MFkuwtcq0tVtmGxrHemRIArchxZi9XkcO1AHkJrfGbvGCNnxfS1nWXM4PSi5 gv8Go1/T2UotREuwrPcaqkcGnqv6Qrw= Received: from mx-prod-mc-03.mail-002.prod.us-west-2.aws.redhat.com (ec2-54-186-198-63.us-west-2.compute.amazonaws.com [54.186.198.63]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-253-oml83FU4NxuCp6XRrGj5uA-1; Thu, 30 Jan 2025 13:15:33 -0500 X-MC-Unique: oml83FU4NxuCp6XRrGj5uA-1 X-Mimecast-MFC-AGG-ID: oml83FU4NxuCp6XRrGj5uA Received: from mx-prod-int-01.mail-002.prod.us-west-2.aws.redhat.com (mx-prod-int-01.mail-002.prod.us-west-2.aws.redhat.com [10.30.177.4]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mx-prod-mc-03.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTPS id 774BD19560AF; Thu, 30 Jan 2025 18:15:32 +0000 (UTC) Received: from mzamazal-thinkpadp1gen3.tpbc.com (unknown [10.45.224.122]) by mx-prod-int-01.mail-002.prod.us-west-2.aws.redhat.com (Postfix) with ESMTP id 724EE30001BE; Thu, 30 Jan 2025 18:15:29 +0000 (UTC) From: Milan Zamazal To: libcamera-devel@lists.libcamera.org Cc: Milan Zamazal , Robert Mader , Hans de Goede , Laurent Pinchart , Kieran Bingham Subject: [PATCH v5 10/10] libcamera: software_isp: Apply CCM in debayering Date: Thu, 30 Jan 2025 19:14:47 +0100 Message-ID: <20250130181449.130492-11-mzamazal@redhat.com> In-Reply-To: <20250130181449.130492-1-mzamazal@redhat.com> References: <20250130181449.130492-1-mzamazal@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 3.4.1 on 10.30.177.4 X-Mimecast-Spam-Score: 0 X-Mimecast-MFC-PROC-ID: vwFNrjlQXwox_dCSYlrFxlVkXzej6Rs_w_xfzU02yUo_1738260932 X-Mimecast-Originator: redhat.com content-type: text/plain; charset="US-ASCII"; x-default=true X-BeenThere: libcamera-devel@lists.libcamera.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: libcamera-devel-bounces@lists.libcamera.org Sender: "libcamera-devel" This patch applies color correction matrix (CCM) in debayering if the CCM is specified. Not using CCM must still be supported for performance reasons. The CCM is applied as follows: [r1 r2 r3] [r g b] * [g1 g2 g3] [b1 b2 b3] The CCM matrix (the right side of the multiplication) is constant during single frame processing, while the input pixel (the left side) changes. Because each of the color channels is only 8-bit in software ISP, we can make 9 lookup tables with 256 input values for multiplications of each of the r_i, g_i, b_i values. This way we don't have to multiply each pixel, we can use table lookups and additions instead. Gamma (which is non-linear and thus cannot be a part of the 9 lookup tables values) is applied on the final values rounded to integers using another lookup table. We use int16_t to store the precomputed multiplications. This seems to be noticeably (>10%) faster than `float' for the price of slightly less accuracy and it covers the range of values that sane CCMs produce. The selection and structure of data is performance critical, for example using bytes would add significant (>10%) speedup but would be too short to cover the value range. The color lookup tables can be represented either as unions, accommodating tables for both the CCM and non-CCM cases, or as separate tables for each of the cases, leaving the tables for the other case unused. The latter is selected as a matter of preference. The tables are copied (as before), which is not elegant but also not a big problem. There are patches posted that use shared buffers for parameters passing in software ISP (see software ISP TODO #5) and they can be adjusted for the new parameter format. Color gains from white balance are supposed not to be a part of the specified CCM. They are applied on it using matrix multiplication, which is simple and in correspondence with future additions in the form of matrix multiplication, like saturation adjustment. With this patch, the reported per-frame slowdown when applying CCM is about 45% on Debix Model A and about 75% on TI AM69 SK. Using std::clamp in debayering adds some performance penalty (a few percent). The clamping is necessary to eliminate out of range values possibly produced by the CCM. If it could be avoided by adjusting the precomputed tables some way then performance could be improved a bit. Signed-off-by: Milan Zamazal Reviewed-by: Laurent Pinchart --- .../internal/software_isp/debayer_params.h | 15 ++++- src/ipa/simple/algorithms/lut.cpp | 65 ++++++++++++++----- src/ipa/simple/algorithms/lut.h | 1 + src/libcamera/software_isp/debayer.cpp | 52 ++++++++++++++- src/libcamera/software_isp/debayer_cpu.cpp | 35 ++++++++-- src/libcamera/software_isp/debayer_cpu.h | 4 ++ 6 files changed, 145 insertions(+), 27 deletions(-) diff --git a/include/libcamera/internal/software_isp/debayer_params.h b/include/libcamera/internal/software_isp/debayer_params.h index 7d8fdd48..9a7dbe93 100644 --- a/include/libcamera/internal/software_isp/debayer_params.h +++ b/include/libcamera/internal/software_isp/debayer_params.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: LGPL-2.1-or-later */ /* - * Copyright (C) 2023, 2024 Red Hat Inc. + * Copyright (C) 2023-2025 Red Hat Inc. * * Authors: * Hans de Goede @@ -18,11 +18,24 @@ namespace libcamera { struct DebayerParams { static constexpr unsigned int kRGBLookupSize = 256; + struct CcmColumn { + int16_t r; + int16_t g; + int16_t b; + }; + using ColorLookupTable = std::array; + using CcmLookupTable = std::array; + using GammaLookupTable = std::array; ColorLookupTable red; ColorLookupTable green; ColorLookupTable blue; + + CcmLookupTable redCcm; + CcmLookupTable greenCcm; + CcmLookupTable blueCcm; + GammaLookupTable gammaLut; }; } /* namespace libcamera */ diff --git a/src/ipa/simple/algorithms/lut.cpp b/src/ipa/simple/algorithms/lut.cpp index e5e995c7..7719aecc 100644 --- a/src/ipa/simple/algorithms/lut.cpp +++ b/src/ipa/simple/algorithms/lut.cpp @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: LGPL-2.1-or-later */ /* - * Copyright (C) 2024, Red Hat Inc. + * Copyright (C) 2024-2025, Red Hat Inc. * * Color lookup tables construction */ @@ -80,6 +80,11 @@ void Lut::updateGammaTable(IPAContext &context) context.activeState.gamma.contrast = contrast; } +int16_t Lut::ccmValue(unsigned int i, float ccm) const +{ + return std::round(i * ccm); +} + void Lut::prepare(IPAContext &context, [[maybe_unused]] const uint32_t frame, [[maybe_unused]] IPAFrameContext &frameContext, @@ -91,28 +96,52 @@ void Lut::prepare(IPAContext &context, * observed, it's not permanently prone to minor fluctuations or * rounding errors. */ - if (context.activeState.gamma.blackLevel != context.activeState.blc.level || - context.activeState.gamma.contrast != context.activeState.knobs.contrast) + const bool gammaUpdateNeeded = + context.activeState.gamma.blackLevel != context.activeState.blc.level || + context.activeState.gamma.contrast != context.activeState.knobs.contrast; + if (gammaUpdateNeeded) updateGammaTable(context); auto &gains = context.activeState.awb.gains; auto &gammaTable = context.activeState.gamma.gammaTable; const unsigned int gammaTableSize = gammaTable.size(); - - for (unsigned int i = 0; i < DebayerParams::kRGBLookupSize; i++) { - const double div = static_cast(DebayerParams::kRGBLookupSize) / - gammaTableSize; - /* Apply gamma after gain! */ - unsigned int idx; - idx = std::min({ static_cast(i * gains.r() / div), - gammaTableSize - 1 }); - params->red[i] = gammaTable[idx]; - idx = std::min({ static_cast(i * gains.g() / div), - gammaTableSize - 1 }); - params->green[i] = gammaTable[idx]; - idx = std::min({ static_cast(i * gains.b() / div), - gammaTableSize - 1 }); - params->blue[i] = gammaTable[idx]; + const double div = static_cast(DebayerParams::kRGBLookupSize) / + gammaTableSize; + + if (!context.ccmEnabled) { + for (unsigned int i = 0; i < DebayerParams::kRGBLookupSize; i++) { + /* Apply gamma after gain! */ + unsigned int idx; + idx = std::min({ static_cast(i * gains.r() / div), + gammaTableSize - 1 }); + params->red[i] = gammaTable[idx]; + idx = std::min({ static_cast(i * gains.g() / div), + gammaTableSize - 1 }); + params->green[i] = gammaTable[idx]; + idx = std::min({ static_cast(i * gains.b() / div), + gammaTableSize - 1 }); + params->blue[i] = gammaTable[idx]; + } + } else if (context.activeState.ccm.changed || gammaUpdateNeeded) { + Matrix gainCcm = { { gains.r(), 0, 0, + 0, gains.g(), 0, + 0, 0, gains.b() } }; + auto ccm = gainCcm * context.activeState.ccm.ccm; + auto &red = params->redCcm; + auto &green = params->greenCcm; + auto &blue = params->blueCcm; + for (unsigned int i = 0; i < DebayerParams::kRGBLookupSize; i++) { + red[i].r = ccmValue(i, ccm[0][0]); + red[i].g = ccmValue(i, ccm[1][0]); + red[i].b = ccmValue(i, ccm[2][0]); + green[i].r = ccmValue(i, ccm[0][1]); + green[i].g = ccmValue(i, ccm[1][1]); + green[i].b = ccmValue(i, ccm[2][1]); + blue[i].r = ccmValue(i, ccm[0][2]); + blue[i].g = ccmValue(i, ccm[1][2]); + blue[i].b = ccmValue(i, ccm[2][2]); + params->gammaLut[i] = gammaTable[i / div]; + } } } diff --git a/src/ipa/simple/algorithms/lut.h b/src/ipa/simple/algorithms/lut.h index 889f864b..77324800 100644 --- a/src/ipa/simple/algorithms/lut.h +++ b/src/ipa/simple/algorithms/lut.h @@ -33,6 +33,7 @@ public: private: void updateGammaTable(IPAContext &context); + int16_t ccmValue(unsigned int i, float ccm) const; }; } /* namespace ipa::soft::algorithms */ diff --git a/src/libcamera/software_isp/debayer.cpp b/src/libcamera/software_isp/debayer.cpp index 45fe6960..4c84a556 100644 --- a/src/libcamera/software_isp/debayer.cpp +++ b/src/libcamera/software_isp/debayer.cpp @@ -23,9 +23,39 @@ namespace libcamera { * \brief Size of a color lookup table */ +/** + * \struct DebayerParams::CcmColumn + * \brief Type of a single column of a color correction matrix (CCM) + */ + +/** + * \var DebayerParams::CcmColumn::r + * \brief Red (first) component of a CCM column + */ + +/** + * \var DebayerParams::CcmColumn::g + * \brief Green (second) component of a CCM column + */ + +/** + * \var DebayerParams::CcmColumn::b + * \brief Blue (third) component of a CCM column + */ + /** * \typedef DebayerParams::ColorLookupTable - * \brief Type of the lookup tables for red, green, blue values + * \brief Type of the simple lookup tables for red, green, blue values + */ + +/** + * \typedef DebayerParams::CcmLookupTable + * \brief Type of the CCM lookup tables for red, green, blue values + */ + +/** + * \typedef DebayerParams::GammaLookupTable + * \brief Type of the gamma lookup tables for CCM */ /** @@ -43,6 +73,26 @@ namespace libcamera { * \brief Lookup table for blue color, mapping input values to output values */ +/** + * \var DebayerParams::redCcm + * \brief CCM lookup table for red color, mapping input values to output values + */ + +/** + * \var DebayerParams::greenCcm + * \brief CCM lookup table for green color, mapping input values to output values + */ + +/** + * \var DebayerParams::blueCcm + * \brief CCM lookup table for blue color, mapping input values to output values + */ + +/** + * \var DebayerParams::gammaLut + * \brief Gamma lookup table used with color correction matrix + */ + /** * \class Debayer * \brief Base debayering class diff --git a/src/libcamera/software_isp/debayer_cpu.cpp b/src/libcamera/software_isp/debayer_cpu.cpp index 0cd03a8f..5ddfdcf6 100644 --- a/src/libcamera/software_isp/debayer_cpu.cpp +++ b/src/libcamera/software_isp/debayer_cpu.cpp @@ -11,6 +11,7 @@ #include "debayer_cpu.h" +#include #include #include #include @@ -51,8 +52,12 @@ DebayerCpu::DebayerCpu(std::unique_ptr stats) enableInputMemcpy_ = true; /* Initialize color lookup tables */ - for (unsigned int i = 0; i < DebayerParams::kRGBLookupSize; i++) + for (unsigned int i = 0; i < DebayerParams::kRGBLookupSize; i++) { red_[i] = green_[i] = blue_[i] = i; + redCcm_[i] = { 1, 0, 0 }; + greenCcm_[i] = { 0, 1, 0 }; + blueCcm_[i] = { 0, 0, 1 }; + } } DebayerCpu::~DebayerCpu() = default; @@ -62,12 +67,24 @@ DebayerCpu::~DebayerCpu() = default; const pixel_t *curr = (const pixel_t *)src[1] + xShift_; \ const pixel_t *next = (const pixel_t *)src[2] + xShift_; -#define STORE_PIXEL(b, g, r) \ - *dst++ = blue_[b]; \ - *dst++ = green_[g]; \ - *dst++ = red_[r]; \ - if constexpr (addAlphaByte) \ - *dst++ = 255; \ +#define GAMMA(value) \ + *dst++ = gammaLut_[std::clamp(value, 0, static_cast(gammaLut_.size()) - 1)] + +#define STORE_PIXEL(b_, g_, r_) \ + if constexpr (ccmEnabled) { \ + const DebayerParams::CcmColumn &blue = blueCcm_[b_]; \ + const DebayerParams::CcmColumn &green = greenCcm_[g_]; \ + const DebayerParams::CcmColumn &red = redCcm_[r_]; \ + GAMMA(blue.r + blue.g + blue.b); \ + GAMMA(green.r + green.g + green.b); \ + GAMMA(red.r + red.g + red.b); \ + } else { \ + *dst++ = blue_[b_]; \ + *dst++ = green_[g_]; \ + *dst++ = red_[r_]; \ + } \ + if constexpr (addAlphaByte) \ + *dst++ = 255; \ x++; /* @@ -757,6 +774,10 @@ void DebayerCpu::process(uint32_t frame, FrameBuffer *input, FrameBuffer *output green_ = params.green; red_ = swapRedBlueGains_ ? params.blue : params.red; blue_ = swapRedBlueGains_ ? params.red : params.blue; + greenCcm_ = params.greenCcm; + redCcm_ = swapRedBlueGains_ ? params.blueCcm : params.redCcm; + blueCcm_ = swapRedBlueGains_ ? params.redCcm : params.blueCcm; + gammaLut_ = params.gammaLut; /* Copy metadata from the input buffer */ FrameMetadata &metadata = output->_d()->metadata(); diff --git a/src/libcamera/software_isp/debayer_cpu.h b/src/libcamera/software_isp/debayer_cpu.h index 21c08a2d..9f73a2fd 100644 --- a/src/libcamera/software_isp/debayer_cpu.h +++ b/src/libcamera/software_isp/debayer_cpu.h @@ -141,6 +141,10 @@ private: DebayerParams::ColorLookupTable red_; DebayerParams::ColorLookupTable green_; DebayerParams::ColorLookupTable blue_; + DebayerParams::CcmLookupTable redCcm_; + DebayerParams::CcmLookupTable greenCcm_; + DebayerParams::CcmLookupTable blueCcm_; + DebayerParams::GammaLookupTable gammaLut_; debayerFn debayer0_; debayerFn debayer1_; debayerFn debayer2_;