From patchwork Sun May 3 11:40:01 2026 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Robert Mader X-Patchwork-Id: 26599 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 39A74BE173 for ; Sun, 3 May 2026 11:40:50 +0000 (UTC) Received: from lancelot.ideasonboard.com (localhost [IPv6:::1]) by lancelot.ideasonboard.com (Postfix) with ESMTP id E36BF62FD3; Sun, 3 May 2026 13:40:49 +0200 (CEST) Authentication-Results: lancelot.ideasonboard.com; dkim=pass (1024-bit key; unprotected) header.d=collabora.com header.i=robert.mader@collabora.com header.b="ec0w3PpG"; dkim-atps=neutral Received: from sender4-pp-f112.zoho.com (sender4-pp-f112.zoho.com [136.143.188.112]) by lancelot.ideasonboard.com (Postfix) with ESMTPS id 9CC0962FEA for ; Sun, 3 May 2026 13:40:46 +0200 (CEST) ARC-Seal: i=1; a=rsa-sha256; t=1777808442; cv=none; d=zohomail.com; s=zohoarc; b=bDj10qVCTkDdj8JA78yKiyCW7ce+jDGgJqNEUFOxxD5Ge3/UPUVD/Qcoc2vNTwuk9K4mHEvP43uPn21FK1U+mDT30ieqsEOFhDKxbfLAanwyOGMYC866PGDBHj4y012tHttBPfZhq26TRdBt7uOGpvsJuP6absvNCss+vGeYq/M= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1777808442; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:MIME-Version:Message-ID:References:Subject:Subject:To:To:Message-Id:Reply-To; bh=i65c0BUiXvt7qykXzHQOd9vKcclacvakhvD8tLa6N/8=; b=C7iU9Qzw+ejEDCrJvvmVh892T6VvnOQKJej1c/midBOXO8kXwksqZ0QCaESTzgWZwmGmkecFwbe8oG5oCjOADNPMnjnqP7VtInaq58DhXAogRGkadJbjPj/ZGfqIuEjGI51cGthw87K+zw2eb4tu5CC4VB1osn3PfPPO8SOiWm4= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=collabora.com; spf=pass smtp.mailfrom=robert.mader@collabora.com; dmarc=pass header.from= DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; t=1777808442; s=zohomail; d=collabora.com; i=robert.mader@collabora.com; h=From:From:To:To:Cc:Cc:Subject:Subject:Date:Date:Message-ID:In-Reply-To:References:MIME-Version:Content-Transfer-Encoding:Message-Id:Reply-To; bh=i65c0BUiXvt7qykXzHQOd9vKcclacvakhvD8tLa6N/8=; b=ec0w3PpGlbRglsup8i1bqKiLVfKg1O4DhvVDRnwa4X2JIShOUuhWqM9MAFTOSqzI TUmt7W0alTLKciL9RWrdcyXqUeZgTWfABmhYJmpBYbk1TbYeTP7QbLeA+L9kAmYz4Aj rIJg7H8WIvck01o7OI6ladFwQY6fJfAojHlh+Euw= Received: by mx.zohomail.com with SMTPS id 1777808440328651.2845930465663; Sun, 3 May 2026 04:40:40 -0700 (PDT) From: Robert Mader To: libcamera-devel@lists.libcamera.org Cc: Robert Mader Subject: [PATCH 2/3] debayer_egl: Implement dmabuf import for input buffers Date: Sun, 3 May 2026 13:40:01 +0200 Message-ID: <20260503114002.139255-3-robert.mader@collabora.com> X-Mailer: git-send-email 2.54.0 In-Reply-To: <20260503114002.139255-1-robert.mader@collabora.com> References: <20260503114002.139255-1-robert.mader@collabora.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" In many cases we can import the GPU-ISP input buffers, dmabufs from v4l2, directly into EGL instead of mapping and uploading - i.e. copying - them. Doing so can have positive effects in multiple areas, including reducing memory bandwidth and CPU usage, as well as avoiding expensive dmabuf syncs and syscalls. The main reason direct imports may not work are the more demanding stride alignment requirements many GPUs have - often 128 or 256 bytes - compared to ISPs - apparently often closer to 32 bytes. Thus we first try to import buffers directly and - if that fails - fall back to the previous upload path. Failing imports should come at low cost as drivers know the limitations and can bail out early, without causing additional IO or context switches. In the future we might be able to request buffers with a matching stride from v4l2 drivers in many cases, making direct import the norm instead of a hit-or-miss. An optional kernel API for that exists, but doesn't seem to be implemented by any driver tested so far. While on it, add some minor code adjustments to make it easier to follow. Signed-off-by: Robert Mader --- src/libcamera/software_isp/debayer_egl.cpp | 28 ++++++++++++---------- src/libcamera/software_isp/debayer_egl.h | 2 +- 2 files changed, 17 insertions(+), 13 deletions(-) diff --git a/src/libcamera/software_isp/debayer_egl.cpp b/src/libcamera/software_isp/debayer_egl.cpp index 8f0c229fd..624469947 100644 --- a/src/libcamera/software_isp/debayer_egl.cpp +++ b/src/libcamera/software_isp/debayer_egl.cpp @@ -495,16 +495,26 @@ void DebayerEGL::setShaderVariableValues(const DebayerParams ¶ms) return; } -int DebayerEGL::debayerGPU(MappedFrameBuffer &in, int out_fd, const DebayerParams ¶ms) +int DebayerEGL::debayerGPU(FrameBuffer *input, FrameBuffer *output, std::vector &dmaSyncers, const DebayerParams ¶ms) { /* eGL context switch */ egl_.makeCurrent(); /* Create a standard texture input */ - egl_.createTexture2D(*eglImageBayerIn_, glFormat_, inputConfig_.stride / bytesPerPixel_, height_, in.planes()[0].data()); + if (egl_.createInputDMABufTexture2D(*eglImageBayerIn_, glFormat_, inputConfig_.stride / bytesPerPixel_, height_, input->planes()[0].fd.get()) != 0) { + LOG(Debayer, Debug) << "Importing input buffer with DMABuf import failed, falling back to upload"; + + dmaSyncBegin(dmaSyncers, input, nullptr); + MappedFrameBuffer in(input, MappedFrameBuffer::MapFlag::Read); + if (!in.isValid()) { + LOG(Debayer, Error) << "mmap-ing buffer(s) failed"; + return -ENODEV; + } + egl_.createTexture2D(*eglImageBayerIn_, glFormat_, inputConfig_.stride / bytesPerPixel_, height_, in.planes()[0].data()); + } /* Generate the output render framebuffer as render to texture */ - egl_.createOutputDMABufTexture2D(*eglImageBayerOut_, out_fd); + egl_.createOutputDMABufTexture2D(*eglImageBayerOut_, output->planes()[0].fd.get()); setShaderVariableValues(params); glViewport(0, 0, width_, height_); @@ -528,21 +538,13 @@ void DebayerEGL::process(uint32_t frame, FrameBuffer *input, FrameBuffer *output std::vector dmaSyncers; - dmaSyncBegin(dmaSyncers, input, nullptr); - /* Copy metadata from the input buffer */ FrameMetadata &metadata = output->_d()->metadata(); metadata.status = input->metadata().status; metadata.sequence = input->metadata().sequence; metadata.timestamp = input->metadata().timestamp; - MappedFrameBuffer in(input, MappedFrameBuffer::MapFlag::Read); - if (!in.isValid()) { - LOG(Debayer, Error) << "mmap-ing buffer(s) failed"; - goto error; - } - - if (debayerGPU(in, output->planes()[0].fd.get(), params)) { + if (debayerGPU(input, output, dmaSyncers, params)) { LOG(Debayer, Error) << "debayerGPU failed"; goto error; } @@ -552,6 +554,8 @@ void DebayerEGL::process(uint32_t frame, FrameBuffer *input, FrameBuffer *output metadata.planes()[0].bytesused = output->planes()[0].length; /* Calculate stats for the whole frame */ + if (dmaSyncers.empty() && (frame % SwStatsCpu::kStatPerNumFrames) == 0) + dmaSyncBegin(dmaSyncers, input, nullptr); stats_->processFrame(frame, 0, input); dmaSyncers.clear(); diff --git a/src/libcamera/software_isp/debayer_egl.h b/src/libcamera/software_isp/debayer_egl.h index fcd281f4c..62cb4f7f1 100644 --- a/src/libcamera/software_isp/debayer_egl.h +++ b/src/libcamera/software_isp/debayer_egl.h @@ -66,7 +66,7 @@ private: int initBayerShaders(PixelFormat inputFormat, PixelFormat outputFormat); int getShaderVariableLocations(); void setShaderVariableValues(const DebayerParams ¶ms); - int debayerGPU(MappedFrameBuffer &in, int out_fd, const DebayerParams ¶ms); + int debayerGPU(FrameBuffer *input, FrameBuffer *output, std::vector &dmaSyncers, const DebayerParams ¶ms); /* Shader program identifiers */ GLuint vertexShaderId_ = 0;