{"id":19836,"url":"https://patchwork.libcamera.org/api/patches/19836/?format=json","web_url":"https://patchwork.libcamera.org/patch/19836/","project":{"id":1,"url":"https://patchwork.libcamera.org/api/projects/1/?format=json","name":"libcamera","link_name":"libcamera","list_id":"libcamera_core","list_email":"libcamera-devel@lists.libcamera.org","web_url":"","scm_url":"","webscm_url":""},"msgid":"<20240404084657.353464-18-mzamazal@redhat.com>","date":"2024-04-04T08:46:54","name":"[v7,17/18] libcamera: Add \"Software ISP benchmarking\" documentation","commit_ref":null,"pull_url":null,"state":"accepted","archived":false,"hash":"c0cea652b6b96d23c1cf776223c68af3276069c1","submitter":{"id":177,"url":"https://patchwork.libcamera.org/api/people/177/?format=json","name":"Milan Zamazal","email":"mzamazal@redhat.com"},"delegate":null,"mbox":"https://patchwork.libcamera.org/patch/19836/mbox/","series":[{"id":4246,"url":"https://patchwork.libcamera.org/api/series/4246/?format=json","web_url":"https://patchwork.libcamera.org/project/libcamera/list/?series=4246","date":"2024-04-04T08:46:37","name":"libcamera: introduce Software ISP and Software IPA","version":7,"mbox":"https://patchwork.libcamera.org/series/4246/mbox/"}],"comments":"https://patchwork.libcamera.org/api/patches/19836/comments/","check":"pending","checks":"https://patchwork.libcamera.org/api/patches/19836/checks/","tags":{},"headers":{"Return-Path":"<libcamera-devel-bounces@lists.libcamera.org>","X-Original-To":"parsemail@patchwork.libcamera.org","Delivered-To":"parsemail@patchwork.libcamera.org","Received":["from lancelot.ideasonboard.com (lancelot.ideasonboard.com\n\t[92.243.16.209])\n\tby patchwork.libcamera.org (Postfix) with ESMTPS id 71463C32D1\n\tfor <parsemail@patchwork.libcamera.org>;\n\tThu,  4 Apr 2024 08:47:58 +0000 (UTC)","from lancelot.ideasonboard.com (localhost [IPv6:::1])\n\tby lancelot.ideasonboard.com (Postfix) with ESMTP id 18D9F633AC;\n\tThu,  4 Apr 2024 10:47:58 +0200 (CEST)","from us-smtp-delivery-124.mimecast.com\n\t(us-smtp-delivery-124.mimecast.com [170.10.133.124])\n\tby lancelot.ideasonboard.com (Postfix) with ESMTPS id 482F76338F\n\tfor <libcamera-devel@lists.libcamera.org>;\n\tThu,  4 Apr 2024 10:47:56 +0200 (CEST)","from mimecast-mx02.redhat.com (mx-ext.redhat.com [66.187.233.73])\n\tby relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3,\n\tcipher=TLS_AES_256_GCM_SHA384) id us-mta-584-GW6euUnbMXq29v91d69fnQ-1;\n\tThu, 04 Apr 2024 04:47:53 -0400","from smtp.corp.redhat.com\n\t(int-mx05.intmail.prod.int.rdu2.redhat.com [10.11.54.5])\n\t(using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits)\n\tkey-exchange X25519 server-signature RSA-PSS (2048 bits)\n\tserver-digest SHA256) (No client certificate requested)\n\tby mimecast-mx02.redhat.com (Postfix) with ESMTPS id 803B21C07F47;\n\tThu,  4 Apr 2024 08:47:53 +0000 (UTC)","from nuthatch.brq.redhat.com (unknown [10.43.17.39])\n\tby smtp.corp.redhat.com (Postfix) with ESMTP id 298068173;\n\tThu,  4 Apr 2024 08:47:52 +0000 (UTC)"],"Authentication-Results":"lancelot.ideasonboard.com; dkim=pass (1024-bit key;\n\tunprotected) header.d=redhat.com header.i=@redhat.com\n\theader.b=\"cPZYfSFL\"; dkim-atps=neutral","DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com;\n\ts=mimecast20190719; t=1712220475;\n\th=from:from:reply-to:subject:subject:date:date:message-id:message-id:\n\tto:to:cc:cc:mime-version:mime-version:content-type:content-type:\n\tcontent-transfer-encoding:content-transfer-encoding:\n\tin-reply-to:in-reply-to:references:references;\n\tbh=1WrIk2YyjdzCk85xMGw0sKBidVUpOqFk1DAvxeZCVRI=;\n\tb=cPZYfSFLCZt+idjGXT6QROUw9iss0PwjSxOjjPby5EDjBkbc9uCVyFPEk6LqHTkHlVLAQJ\n\tReyu03TrdC2UDdH0OMcQJ2wandbXzCG+zMBnfW+c9oxKOwvZ0DpWJpnbMik7t2rlL5rwVb\n\tkGh1fypwXSFhBOivzet2bPeTcyTHZDI=","X-MC-Unique":"GW6euUnbMXq29v91d69fnQ-1","From":"Milan Zamazal <mzamazal@redhat.com>","To":"libcamera-devel@lists.libcamera.org","Cc":"Hans de Goede <hdegoede@redhat.com>,\n\tAndrei Konovalov <andrey.konovalov.ynk@gmail.com>,\n\tBryan O'Donoghue <bryan.odonoghue@linaro.org>,\n\tMaxime Ripard <mripard@redhat.com>, Milan Zamazal <mzamazal@redhat.com>, \n\tPavel Machek <pavel@ucw.cz>,\n\tKieran Bingham <kieran.bingham@ideasonboard.com>,\n\tLaurent Pinchart <laurent.pinchart@ideasonboard.com>,\n\tStefan Klug <stefan.klug@ideasonboard.com>","Subject":"[PATCH v7 17/18] libcamera: Add \"Software ISP benchmarking\"\n\tdocumentation","Date":"Thu,  4 Apr 2024 10:46:54 +0200","Message-ID":"<20240404084657.353464-18-mzamazal@redhat.com>","In-Reply-To":"<20240404084657.353464-1-mzamazal@redhat.com>","References":"<20240404084657.353464-1-mzamazal@redhat.com>","MIME-Version":"1.0","X-Scanned-By":"MIMEDefang 3.4.1 on 10.11.54.5","X-Mimecast-Spam-Score":"0","X-Mimecast-Originator":"redhat.com","Content-Type":"text/plain; charset=UTF-8","Content-Transfer-Encoding":"8bit","X-BeenThere":"libcamera-devel@lists.libcamera.org","X-Mailman-Version":"2.1.29","Precedence":"list","List-Id":"<libcamera-devel.lists.libcamera.org>","List-Unsubscribe":"<https://lists.libcamera.org/options/libcamera-devel>,\n\t<mailto:libcamera-devel-request@lists.libcamera.org?subject=unsubscribe>","List-Archive":"<https://lists.libcamera.org/pipermail/libcamera-devel/>","List-Post":"<mailto:libcamera-devel@lists.libcamera.org>","List-Help":"<mailto:libcamera-devel-request@lists.libcamera.org?subject=help>","List-Subscribe":"<https://lists.libcamera.org/listinfo/libcamera-devel>,\n\t<mailto:libcamera-devel-request@lists.libcamera.org?subject=subscribe>","Errors-To":"libcamera-devel-bounces@lists.libcamera.org","Sender":"\"libcamera-devel\" <libcamera-devel-bounces@lists.libcamera.org>"},"content":"From: Hans de Goede <hdegoede@redhat.com>\n\nAdd a \"Software ISP benchmarking\" documentation section which describes the\nperformance/power consumption measurements used during the Software ISP's\ndevelopment.\n\nReviewed-by: Milan Zamazal <mzamazal@redhat.com>\nReviewed-by: Stefan Klug <stefan.klug@ideasonboard.com>\nReviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>\nSigned-off-by: Hans de Goede <hdegoede@redhat.com>\n---\n Documentation/index.rst                     |  1 +\n Documentation/meson.build                   |  1 +\n Documentation/software-isp-benchmarking.rst | 77 +++++++++++++++++++++\n 3 files changed, 79 insertions(+)\n create mode 100644 Documentation/software-isp-benchmarking.rst","diff":"diff --git a/Documentation/index.rst b/Documentation/index.rst\nindex 63fac72d..5442ae75 100644\n--- a/Documentation/index.rst\n+++ b/Documentation/index.rst\n@@ -24,3 +24,4 @@\n    Lens driver requirements <lens_driver_requirements>\n    Python Bindings <python-bindings>\n    Camera Sensor Model <camera-sensor-model>\n+   SoftwareISP Benchmarking <software-isp-benchmarking>\ndiff --git a/Documentation/meson.build b/Documentation/meson.build\nindex 7a58fec8..3872e0a8 100644\n--- a/Documentation/meson.build\n+++ b/Documentation/meson.build\n@@ -80,6 +80,7 @@ if sphinx.found()\n         'lens_driver_requirements.rst',\n         'python-bindings.rst',\n         'sensor_driver_requirements.rst',\n+        'software-isp-benchmarking.rst',\n        '../README.rst',\n     ]\n \ndiff --git a/Documentation/software-isp-benchmarking.rst b/Documentation/software-isp-benchmarking.rst\nnew file mode 100644\nindex 00000000..b3033132\n--- /dev/null\n+++ b/Documentation/software-isp-benchmarking.rst\n@@ -0,0 +1,77 @@\n+.. SPDX-License-Identifier: CC-BY-SA-4.0\n+\n+.. _software-isp-benchmarking:\n+\n+Software ISP benchmarking\n+=========================\n+\n+The Software ISP is particularly sensitive to performance regressions therefore\n+it is a good idea to always benchmark the Software ISP before and after making\n+changes to it and ensure that there are no performance regressions.\n+\n+DebayerCpu class builtin benchmark\n+----------------------------------\n+\n+The DebayerCpu class has a builtin benchmark. This benchmark measures the time\n+spent on processing (collecting statistics and debayering) only, it does not\n+measure the time spent on capturing or outputting the frames.\n+\n+The builtin benchmark always runs. So this can be used by simply running \"cam\"\n+or \"qcam\" with a pipeline using the Software ISP.\n+\n+When it runs it will skip measuring the first 30 frames to allow the caches and\n+the CPU temperature (turbo-ing) to warm-up and then it measures 30 fps and shows\n+the total and per frame processing time using an info level log message:\n+\n+.. code-block:: text\n+\n+   INFO Debayer debayer_cpu.cpp:907 Processed 30 frames in 244317us, 8143 us/frame\n+\n+To get stable measurements it is advised to disable any other processes which\n+may cause significant CPU usage (e.g. disable wifi, bluetooth and browsers).\n+When possible it is also advisable to disable CPU turbo-ing and\n+frequency-scaling.\n+\n+For example when benchmarking on a Lenovo ThinkPad X1 Yoga Gen 8, with the\n+charger plugged in, the CPU can be fixed to run at 2 GHz using:\n+\n+.. code-block:: shell\n+\n+   sudo x86_energy_perf_policy --turbo-enable 0\n+   sudo cpupower frequency-set -d 2GHz -u 2GHz\n+\n+with these settings the builtin bench reports a processing time of ~7.8ms/frame\n+on this laptop for FHD SGRBG10 (unpacked) bayer data.\n+\n+Measuring power consumption\n+---------------------------\n+\n+Since the Software ISP is often used on mobile devices it is also important to\n+measure power consumption and ensure that that does not regress.\n+\n+For example to measure power consumption on a Lenovo ThinkPad X1 Yoga Gen 8 it\n+needs to be running on battery and it should be configured with its\n+platform-profile (/sys/firmware/acpi/platform_profile) set to balanced and with\n+its default turbo and frequency-scaling behavior to match real world usage.\n+\n+Then start qcam to capture a FHD picture at 30 fps and position the qcam window\n+so that it is fully visible. After this run the following command to monitor the\n+power consumption:\n+\n+.. code-block:: shell\n+\n+   watch -n 10 cat /sys/class/power_supply/BAT0/power_now /sys/class/hwmon/hwmon6/fan?_input\n+\n+Note this not only measures the power consumption in µW it also monitors the\n+speed of this laptop's 2 fans. This is important because depending on the\n+ambient temperature the 2 fans may spin up while testing and this will cause an\n+additional power consumption of approx. 0.5 W messing up the measurement.\n+\n+After starting qcam + the watch command let the laptop sit without using it for\n+2 minutes for the readings to stabilize. Then check that the fans have not\n+turned on and manually take a couple of consecutive power readings and average\n+these.\n+\n+On the example Lenovo ThinkPad X1 Yoga Gen 8 laptop this results in a measured\n+power consumption of approx. 13 W while running qcam versus approx. 4-5 W while\n+setting idle with its OLED panel on.\n","prefixes":["v7","17/18"]}