From patchwork Wed Apr 24 23:42:24 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Laurent Pinchart X-Patchwork-Id: 19951 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 99C9AC3200 for ; Wed, 24 Apr 2024 23:42:50 +0000 (UTC) Received: from lancelot.ideasonboard.com (localhost [IPv6:::1]) by lancelot.ideasonboard.com (Postfix) with ESMTP id 60C0D63413; Thu, 25 Apr 2024 01:42:49 +0200 (CEST) Authentication-Results: lancelot.ideasonboard.com; dkim=pass (1024-bit key; unprotected) header.d=ideasonboard.com header.i=@ideasonboard.com header.b="WL28wOUH"; dkim-atps=neutral Received: from perceval.ideasonboard.com (perceval.ideasonboard.com [213.167.242.64]) by lancelot.ideasonboard.com (Postfix) with ESMTPS id 4471F633F3 for ; Thu, 25 Apr 2024 01:42:36 +0200 (CEST) Received: from pendragon.ideasonboard.com (117.145-247-81.adsl-dyn.isp.belgacom.be [81.247.145.117]) by perceval.ideasonboard.com (Postfix) with ESMTPSA id F0080B1 for ; Thu, 25 Apr 2024 01:41:43 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=ideasonboard.com; s=mail; t=1714002104; bh=hIGqR6Kt4tDC1fVeCf/uWAtzTPzODPcv5qnjoaKSsoo=; h=From:To:Subject:Date:In-Reply-To:References:From; b=WL28wOUHhGjQvzTjE2FGsBKx+9a6Gxwi6l26tM2u2SgMxF8UWIy1KThksNLmeJPnC 6Uf8KB30/+V1Sx229bQovB9GJEDMA78XArchlyLUDP+Hygr0DUcNmHoT8ej0H7qy/N 88Zuxadw5JshTQpCMdZif1V6Qe1mg7POJML553R0= From: Laurent Pinchart To: libcamera-devel@lists.libcamera.org Subject: [PATCH v1 6/6] pipeline: vimc: Don't hardcode scaling factor with recent kernels Date: Thu, 25 Apr 2024 02:42:24 +0300 Message-ID: <20240424234224.9658-7-laurent.pinchart@ideasonboard.com> X-Mailer: git-send-email 2.43.2 In-Reply-To: <20240424234224.9658-1-laurent.pinchart@ideasonboard.com> References: <20240424234224.9658-1-laurent.pinchart@ideasonboard.com> MIME-Version: 1.0 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" Starting in kernel v5.16, the vimc driver stopped hardcoding the scaler factor. Use this to lift constraints on the camera configuration, and in particular on the exotic output size alignment to a multiple of 6. As a result, vimc-based cameras can more easily match common display resolutions. Signed-off-by: Laurent Pinchart Reviewed-by: Kieran Bingham --- src/libcamera/pipeline/vimc/vimc.cpp | 47 +++++++++++++++++++--------- 1 file changed, 33 insertions(+), 14 deletions(-) diff --git a/src/libcamera/pipeline/vimc/vimc.cpp b/src/libcamera/pipeline/vimc/vimc.cpp index 5e66ee1d26c1..be7a6733472b 100644 --- a/src/libcamera/pipeline/vimc/vimc.cpp +++ b/src/libcamera/pipeline/vimc/vimc.cpp @@ -114,6 +114,9 @@ static const std::map pixelformats{ { formats::BGR888, MEDIA_BUS_FMT_RGB888_1X24 }, }; +static constexpr Size kMinSize{ 16, 16 }; +static constexpr Size kMaxSize{ 4096, 2160 }; + } /* namespace */ VimcCameraConfiguration::VimcCameraConfiguration(VimcCameraData *data) @@ -153,14 +156,20 @@ CameraConfiguration::Status VimcCameraConfiguration::validate() const Size size = cfg.size; /* - * The scaler hardcodes a x3 scale-up ratio, and the sensor output size - * is aligned to two pixels in both directions. The output width and - * height thus have to be multiples of 6. + * The sensor output size is aligned to two pixels in both directions. + * Additionally, prior to v5.16, the scaler hardcodes a x3 scale-up + * ratio, requiring the output width and height to be multiples of 6. */ - cfg.size.width = std::max(48U, std::min(4096U, cfg.size.width)); - cfg.size.height = std::max(48U, std::min(2160U, cfg.size.height)); - cfg.size.width -= cfg.size.width % 6; - cfg.size.height -= cfg.size.height % 6; + Size minSize{ kMinSize }; + unsigned int alignment = 2; + + if (data_->media_->version() < KERNEL_VERSION(5, 16, 0)) { + minSize *= 3; + alignment *= 3; + } + + cfg.size.expandTo(minSize).boundTo(kMaxSize) + .alignDownTo(alignment, alignment); if (cfg.size != size) { LOG(VIMC, Debug) @@ -216,10 +225,12 @@ PipelineHandlerVimc::generateConfiguration(Camera *camera, } } - /* The scaler hardcodes a x3 scale-up ratio. */ - std::vector sizes{ - SizeRange{ { 48, 48 }, { 4096, 2160 } } - }; + /* Prior to v5.16, the scaler hardcodes a x3 scale-up ratio. */ + Size minSize{ kMinSize }; + if (data->media_->version() < KERNEL_VERSION(5, 16, 0)) + minSize *= 3; + + std::vector sizes{ { minSize, kMaxSize } }; formats[pixelformat.first] = sizes; } @@ -242,10 +253,18 @@ int PipelineHandlerVimc::configure(Camera *camera, CameraConfiguration *config) StreamConfiguration &cfg = config->at(0); int ret; - /* The scaler hardcodes a x3 scale-up ratio. */ + /* + * Prior to v5.16, the scaler hardcodes a x3 scale-up ratio. For newer + * kernels, use a sensor resolution of 1920x1080 and let the scaler + * produce the requested stream size. + */ + Size sensorSize{ 1920, 1080 }; + if (data->media_->version() < KERNEL_VERSION(5, 16, 0)) + sensorSize = { cfg.size.width / 3, cfg.size.height / 3 }; + V4L2SubdeviceFormat subformat = {}; subformat.code = MEDIA_BUS_FMT_SGRBG8_1X8; - subformat.size = { cfg.size.width / 3, cfg.size.height / 3 }; + subformat.size = sensorSize; ret = data->sensor_->setFormat(&subformat); if (ret) @@ -293,7 +312,7 @@ int PipelineHandlerVimc::configure(Camera *camera, CameraConfiguration *config) * vimc driver will fail pipeline validation. */ format.fourcc = V4L2PixelFormat(V4L2_PIX_FMT_SGRBG8); - format.size = { cfg.size.width / 3, cfg.size.height / 3 }; + format.size = sensorSize; ret = data->raw_->setFormat(&format); if (ret)