kevers@chromium.org chrome-a11y-core@chromium.org Tracks when combinations of mode flags are set. Records when the accessibility mode flag changes and aligns with one of the common (named) flag bundles. Bundles that do not match a named set fall into the unnamed bucket. kevers@chromium.org chrome-a11y-core@chromium.org Tracks whether individual accessibility mode flags are set. Recorded when each mode flag is flipped from false to true. Only tracks usage within web content. lucasradaelli@google.com chrome-a11y-core@chromium.org Records how many tree updates were queued at the time updates are being processed after layout. If we are trying to process too many updates at once, we should investigate how we might be able to merge similar updates without losing important information. ikilpatrick@chromium.org layout-dev@chromium.org Time spent updating accessibility in the Blink document lifecycle. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} adithyas@chromium.org chrome-instant-ecosystem@google.com Records the vertical distance of a clicked anchor from the second last pointerdown (the last pointerdown would be the one that initiated the click). The value is normalized and recorded as a percentage of screen size, and is scaled to accomodate negative distances. A value below 50 indicates that the click happened above the previous pointerdown by ~(100 - 2X)%. A value above 50 indicates that the click happened below the previous pointerdown by ~(2X - 100)%. This is only recorded when the screen is in the "{orientation}" orientation, and is recorded after an anchor element click. It is only recorded if a pointerdown was seen prior to the click. adithyas@chromium.org chrome-instant-ecosystem@google.com Records the vertical distance from the top of the screen (y-coordinate) of a clicked anchor. The value is normalized and recorded as a percentage of the screen size. This is only recorded when the screen is in the "{orientation}" orientation, and is recorded after every anchor element click. pdr@chromium.org paint-dev@chromium.org Time spent computing IntersectionObserver observations in the Blink document lifecycle where the observation is for gathering information about anchor elements within the visible viewport. Recorded when an intersection observation attributed to anchor element metrics is updated. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} pdr@chromium.org paint-dev@chromium.org Time spent processing main frame animations during a main frame update. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note that the histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} kevers@chromium.org dmangal@microsoft.com animations-dev@chromium.org Records the type of animation (all types, SVG, composited, non-composited) when an animation starts. This helps track usage patterns of different animation types across the web. This is recorded when an animation attempts to start, regardless of whether it successfully runs on the compositor. Note: This does not include SVG SMIL animations. smcgruer@chromium.org animations-dev@chromium.org Reasons an Animation is not suitable for running on the compositor thread. Note that a given Animation may have more than one reason, so this histogram cannot be used to determine the overall rate of 'composited' vs 'noncomposited' animations. Also note that "failure" to run on the compositor thread is actually good when the animation can be skipped entirely. Prior to M124, this case was restricted to composited properties that had no visual effect. This failure reason (bucket 18) has been generalized to capture cases where the animation can be optimized out even if having non-compositable properties. These animations will not run on the compositor and only tick on main at the boundaries of the active phase. kevers@chromium.org animations-dev@chromium.org Records whether a script-provided animation range offset contains a relative or element-dependent length and would be rejected if the experimental behavior were enabled. A false sample represents an absolute length or percentage. Recorded each time Blink converts a valid animation range offset supplied through the Web Animations API, before the runtime feature controls whether a TypeError is thrown. This measures the compatibility risk of rejecting relative and element-dependent offsets. kevers@chromium.org animations-dev@chromium.org Unexpected misalignment of a phase boundary when aligned to a scroll- boundary measured in 16ths of a pixel. Recorded each time phase calculation determines the animation is in the before or after phase when a scroll- driven animation has determined that the before-active or active-after boundary is aligned to the bounds of the scroll range. fdoray@chromium.org catan-team@chromium.org Duration of a synchronous load by blink::URLLoader. Recorded every time blink::URLLoader::LoadSynchronously performs a load. Synchronous load are discouraged as they block the main thread. This histogram will be used to estimate to which extent they affect Interaction to Next Paint (INP). lizeb@chromium.org jpgravel@chromium.org When a compressed hibernated 2D canvas snapshot is decompressed, duration in ms. Reported each time a canvas snapshot is decompressed. lizeb@chromium.org jpgravel@chromium.org When a hibernated 2D canvas snapshot is compressed, the compression ratio achieved. Reported each time a canvas snapshot is compressed. lizeb@chromium.org jpgravel@chromium.org When a hibernated 2D canvas snapshot is compressed, the size before compression. Reported each time a canvas snapshot is compressed. lizeb@chromium.org jpgravel@chromium.org When a hibernated 2D canvas snapshot is compressed, the CPU time it takes to compress it. Reported each time a canvas snapshot is compressed. aaronhk@chromium.org jpgravel@chromium.org Records the state of the willReadFrequently flag upon creation of a Canvas2D layer bridge, as this is the moment when it determines whether the canvas will be accelerated. fserb@chromium.org aaronhk@chromium.org Records if a Canvas is being printed as a vector or as an image. aaronhk@chromium.org fserb@chromium.org Records whether the contents in the canvas buffer needed to be preserved via a copy when replacing the target buffer for drawing a new frame. This metric is recorded each time the target buffer needs to be replaced. It's recorded once per frame. fserb@chromium.org aaronhk@chromium.org The source from which an ImageBitmap is created by a createImageBitmap call. aaronhk@chromium.org junov@chromium.org Records the state of the willReadFrequently flag during Canvas2D readback. aaronhk@chromium.org fserb@chromium.org Logs the occurrences of canvas fallback from GPU to CPU. aaronhk@chromium.org fserb@chromium.org Records the occurrence of events related to 2D canvas GPU resource hibernation. Replaced Blink.Canvas.HibernationEvents because some hibernation events were incorrectly classified. aaronhk@chromium.org fserb@chromium.org Reports the maximum number or canvas resources used at any one time by a canvas resource provider. Logged in the destructor of CanvasResourceProvider. junov@chromium.org fserb@chromium.org Reports the maximum depth of the state stack over the lifetime of a 2d rendering context. Metric is recorded once per instance of CanvasRenderingContext2D, OffscreenRenderingContext2D, and PaintRenderingContext2D; at context destruction time. aaronhk@chromium.org fserb@chromium.org When a Document is shutdown, reports the number of canvases on that page. aaronhk@chromium.org fserb@chromium.org Records the CPU, GPU and combined raster time for a random sampling of canvas frames for accelerated and deferred canvases. Records wall time for a random sampling of unaccelerated canvases. Warning: This metric may include reports from clients with low-resolution clocks (i.e. on Windows, ref. |TimeTicks::IsHighResolution()|). Such reports will cause this metric to have an abnormal distribution. When considering revising this histogram, see UMA_HISTOGRAM_CUSTOM_MICROSECONDS_TIMES for the solution. {BlinkCanvasRasterDurationType} aaronhk@chromium.org fserb@chromium.org Logs if the canvas resource provider is accelerated or not. aaronhk@chromium.org fserb@chromium.org Records the type of resource provider used for a canvas. aaronhk@chromium.org fserb@chromium.org Records the requested resource usage for a canvas resource provider. aaronhk@chromium.org fserb@chromium.org Stores the square root of the number of pixels in a new or resized offscreen canvas. Emitted from CanvasRenderingContextHost::RecordCanvasSizeToUMA. fserb@chromium.org aaronhk@chromium.org Records the status of the idle task when finishing a toBlob call. In addition, metric values from OffscreenCanvas.convertToBlob API call are also gathered into this histogram, because the logic flow is exactly the same as canvas.toBlob. It's worth to note that the values can come from idle tasks on either main or worker thread. aaronhk@chromium.org fserb@chromium.org A new version of "Blink.Canvas.ToBlob.InitiateEncodingDelay" (which will be allowed to expire) which allows for times >10ms. This metric measures the time spent from initiating image encoding (jpeg or png) on idle task to the actual execution time of initiation, as part of a canvas.toBlob API call. This metric is useful in helping us adjust the IdleTaskStartTimeoutDelay in canvas.toBlob. When the initialization idle task is delayed for longer than IdleTaskStartTimeoutDelay, the browser will switch to a non-idle task to force initialization and encoding to occur on the main thread. In addition, metric values from OffscreenCanvas.convertToBlob API call are also gathered into this histogram, because the logic flow is exactly the same as canvas.toBlob. It's worth to note that the values can come from idle tasks on either main or worker thread. Note: This metric drops reports on clients with low-resolution clocks, which means these reports will be biased against a portion of the population on Windows. See Windows.HasHighResolutionTimeTicks for the affected sample. fserb@chromium.org aaronhk@chromium.org Time in microseconds spent on the 2D canvas toBlob API call, divided by the square root of the total number of pixels of the canvas. In addition, metric values from OffscreenCanvas.convertToBlob API call are also gathered into this histogram, because the logic flow is exactly the same as canvas.toBlob. It's worth to note that the values can come from idle tasks on either main or worker thread. Note: This metric drops reports on clients with low-resolution clocks, which means these reports will be biased against a portion of the population on Windows. See Windows.HasHighResolutionTimeTicks for the affected sample. {BlinkCanvasToBlobType} aaronhk@chromium.org fserb@chromium.org A new version of "Blink.Canvas.ToBlob.InitiateEncodingDelay" (which will be allowed to expire) which allows for times >10ms. This metric measures the total time spent on completing encoding all the rows of an image (jpeg or png), as part of a canvas.toBlob API call. Encoding occurs during one or more idle periods on the main thread and the waiting time of the next idle period is included in the measurement. If the code has swtiched to force encoding path, we will not measure the delay in this metric. This metric is useful in helping us adjust the IdleTaskCompleteTimeoutDelay in canvas.toBlob. When the encoding idle task is delayed for longer than IdleTaskCompleteTimeoutDelay, the browser will switch to a non-idle task to force encoding to happen on the main thread. In addition, metric values from OffscreenCanvas.convertToBlob API call are also gathered into this histogram, because the logic flow is exactly the same as canvas.toBlob. It's worth to note that the values can come from idle tasks on either main or worker thread. Note: This metric drops reports on clients with low-resolution clocks, which means these reports will be biased against a portion of the population on Windows. See Windows.HasHighResolutionTimeTicks for the affected sample. fserb@chromium.org aaronhk@chromium.org Time spent on 2D canvas toDataURL API call divided by the square root of the total number of pixels of the image. Smaller is faster. The idea behind this unit is that the bigger the area of a canvas the longer the operation ToDataURL takes, so we want a way to measure the overall performance regardless of the size of the canvas. Warning: This metric may include reports from clients with low-resolution clocks (i.e. on Windows, ref. |TimeTicks::IsHighResolution()|). Such reports will cause this metric to have an abnormal distribution. When considering revising this histogram, see UMA_HISTOGRAM_CUSTOM_MICROSECONDS_TIMES for the solution. {BlinkCanvasToDataURLMimeType} zmo@chromium.org webgl-team@google.com Distribution of WebGL1 and WebGL2 contexts and their ANGLE implementation. Recorded at first draw call time for each context. ravjit@chromium.org src/components/permissions/PERMISSIONS_OWNERS Records whether the getUserMedia (gUM) API call by the {CapabilityType} element failed specifically due to an OverconstrainedError (true) or some other error (false). Recorded when the gUM request callback returns with an error. ravjit@chromium.org src/components/permissions/PERMISSIONS_OWNERS Records the reason why the {CapabilityType} element's style is invalid. This metric is emitted when the element's style is calculated/recalculated. Note: This metric was previously recorded as Blink.CapabilityElement.{CapabilityType}.InvalidStyleReason in M147 ravjit@chromium.org src/components/permissions/PERMISSIONS_OWNERS Recorded every time a user clicks on a {CapabilityType} element. True indicates that the interaction was accepted, false indicates that the interaction was denied. Check Blink.CapabilityElement.{CapabilityType}.UserInteraction.DeniedReason for the reason of denial. ravjit@chromium.org src/components/permissions/PERMISSIONS_OWNERS Records the reason why the user interaction with the {CapabilityType} element was denied. Recorded when the element is clicked by the user. sisidovski@chromium.org chrome-loading@google.com Time spent in ChildURLLoaderFactoryBundle::Clone() in microseconds. Recorded when ChildURLLoaderFactoryBundle::Clone() is called to clone a loader factory bundle (e.g., when initializing workers, subresource loaders, or frame navigations). Note: This metric is only reported for clients with high-resolution clocks. rbyers@chromium.org shwetabindal@microsoft.com Records the total size in KB of all blobs read during a single navigator.clipboard.read() call in eager (non-lazy) mode. All MIME type blobs are read upfront before the clipboard.read() promise resolves. Recorded once per clipboard.read() call when the ClipboardItem is created. rbyers@chromium.org shwetabindal@microsoft.com Records the number of MIME types advertised by a lazy ClipboardItem (from navigator.clipboard.read()) that were never fetched via getType() before the ClipboardItem was destroyed. Recorded once per lazy ClipboardItem when it is destroyed (frame detachment or navigation) or garbage collected. rbyers@chromium.org shwetabindal@microsoft.com Records whether a lazy-read getType() call was rejected because the clipboard content changed after clipboard.read(). True means rejected; false means the clipboard had not changed. Recorded once per getType() call on a lazy-mode ClipboardItem. rbyers@chromium.org shwetabindal@microsoft.com Records whether a lazy-read getType() call resolved with a null blob from the OS clipboard. True means the blob was null; false means valid data was returned. Recorded once per getType() resolution on a lazy-mode ClipboardItem. rbyers@chromium.org shwetabindal@microsoft.com Records the total size in KB of all blobs resolved via getType() calls for a single lazy ClipboardItem during its lifetime. Recorded once per lazy ClipboardItem when it is destroyed (frame detachment or navigation) or garbage collected. rbyers@chromium.org shwetabindal@microsoft.com Records true when a clipboard reader (PNG, text, HTML, or SVG) produces null data after processing. For PNG this means ReadPng() returned empty data; for text it means ReadPlainText() returned empty; for HTML/SVG it means sanitization produced an empty string. Recorded once per individual reader invocation. nidhijaju@chromium.org chrome-loading@google.com Measures the time it takes to commit a new document in Blink. Collected once per navigation when a frame is loaded. pdr@chromium.org paint-dev@chromium.org Time spent updating and pushing layers at the end of the paint step in the Blink document lifecycle. This is implemented by PaintArtifactCompositor. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} pdr@chromium.org paint-dev@chromium.org Time spent updating compositing inputs in the Blink document lifecycle. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} wangxianzhu@chromium.org paint-dev@chromium.org Number of fixed-position composited layers under a local frame root. Reported at the end of a full compositing update, with a 0.01 sample rate. wangxianzhu@chromium.org paint-dev@chromium.org Total number of fixed-position pending layers that are merged into composited layers, under a local frame root. This is non-zero only if MergeFixedLayers is enabled. Reported at the end of a full compositing update, with a 0.01 sample rate. wangxianzhu@chromium.org paint-dev@chromium.org Total number of sticky-position pending layers that are merged into composited layers, under a local frame root. This is non-zero only if MergeStickyLayers is enabled. Reported at the end of a full compositing update, with a 0.01 sample rate. wangxianzhu@chromium.org paint-dev@chromium.org Number of sticky-position composited layers under a local frame root. Reported at the end of a full compositing update, with a 0.01 sample rate. ricea@chromium.org yhirano@chromium.org Compression format passed to the constructor of the JavaScript API "CompressionStream", which is used for lossless compression of arbitrary binary data. Recorded once per call to the constructor. This measures which compression formats are most commonly used in the wild. ricea@chromium.org yhirano@chromium.org Compression format passed to the constructor of the JavaScript API "DecompressionStream", which is used for lossless decompression of arbitrary binary data. Recorded once per call to the constructor. This measures which compression formats are most commonly used in the wild. pdr@chromium.org paint-dev@chromium.org Time spent preparing the document for a canvas draw, SVG image draw or plugin during a main frame update. Includes time spent in style update and layout. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} flackr@chromium.org blink-interactions-team@google.com The depth of the discovered image node in the hit test list when performing a penetrating image selection search. Only recorded when a user long presses or right clicks a document (and an image is found). flackr@chromium.org blink-interactions-team@google.com The amount of time it took to iterate through the dom looking for an image node. Only recorded when a user long presses or right clicks a document. Note: This metric drops reports on clients with low-resolution clocks, which means these reports will be biased against a portion of the population on Windows. See Windows.HasHighResolutionTimeTicks for the affected sample. flackr@chromium.org blink-interactions-team@google.com The outcome of an image selection search that may or may not trigger the context menu. Only recorded when a user long presses or right clicks a document (and an image is found). flackr@chromium.org blink-interactions-team@google.com The outcome of a follow up retrieval of an image node after a selection has occurred, which occurs asynchronously and thus could be unsuccessful due to changes that may occur between the initial hit test and the retrieval. Only recorded when a user long presses or right clicks a document (and an image is found). elkurin@chromium.org chrome-owp-storage@google.com Time (in ms) added to Cookies operation with IPC call from the CookieJar used for ablation study purpose. This is incremented to 2 because the Blink.CookiesTime.AblationDelay metrics also includes the metrics for the operation without IPC call. elkurin@chromium.org chrome-owp-storage@google.com Sleep duration (in microseconds) added to Cookies operation with IPC call from the CookieJar used for ablation study. This metrics is recorded only when CookieJar ablation study is enabled, and during getting cookie with IPC. While AblationDelay2 records the wanted delay, this metrics records the actual delay. Note: This metric is only reported for clients with high-resolution clocks. elkurin@chromium.org chrome-owp-storage@google.com Measures the time it takes to perform a Cookies operation with or without IPC call from the CookieJar. IPC call can be skipped in case where the cookie string is taken successfully and has not changed. In such case, the metrics is recorded for the time to check shared memory and other tasks. The metrics should be separated by whether it needs IPC as it extremely affects the time. This is incremented to 2 because Blink.CookiesTime.{IpcNeeded} was ms units which is not precise enough to capture IpcNotNeeded metrics. Note: This metric is reported for all clients, regardless of clock resolution. mmenke@chromium.org net-dev@chromium.org Measures the time it takes to perform a Cookies operation, given the state of the restricted cookie manager pipe owned by CookieJar when the call was made. This is intended to be a short-lived histogram to specifically investigate occasional hangs when fetching cookies. gjc@chromium.org loading-dev@chromium.org net-dev@chromium.org Time spent for parsing the stylesheet. Recorded when the stylesheet finishes loading and does not have cached stylesheet contents. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. wtc@google.com jzern@google.com src/third_party/blink/renderer/platform/image-decoders/avif/OWNERS The number of images per image density measured in 0.01 bits per pixel for image size in {ImageArea} megapixels. This is logged once per image load after the whole image is loaded and only for 8-bit AVIF color still images without alpha. wtc@google.com jzern@google.com src/third_party/blink/renderer/platform/image-decoders/avif/OWNERS The compressed image size in KiB per image density measured in 0.01 bits per pixel. This is logged once per image load after the whole image is loaded and only for 8-bit AVIF still images with at least 100 pixels on the smallest dimension (width or height). The reported count for a sample represents the image size rounded to the nearest KiB. wtc@google.com jzern@google.com src/third_party/blink/renderer/platform/image-decoders/OWNERS The number of images per image density measured in 0.01 bits per pixel for image size in {ImageArea} megapixels. This is logged once per image load after the whole image is loaded and only for JPEG non-grayscale images. jyrki@google.com compression-dev@google.com The compressed image size in KiB per image density measured in 0.01 bits per pixel. This is logged once per image load after the whole image is loaded and only for JPEGs with at least 100 pixels on the smallest dimension (width or height). The reported count for a sample represents the image size rounded to the nearest KiB. veluca@chromium.org helmut@januschka.com src/third_party/blink/renderer/platform/image-decoders/jxl/OWNERS The number of images per image density measured in 0.01 bits per pixel for image size in {ImageArea} megapixels. This is logged once per image load after the whole image is loaded and only for 8-bit JXL color still images without alpha. wtc@google.com jzern@google.com src/third_party/blink/renderer/platform/image-decoders/OWNERS The number of images per image density measured in 0.01 bits per pixel for image size in {ImageArea} megapixels. This is logged once per image load after the whole image is loaded and only for WebP lossy still images without alpha. wtc@google.com jzern@google.com src/third_party/blink/renderer/platform/image-decoders/OWNERS The compressed image size in KiB per image density measured in 0.01 bits per pixel. This is logged once per image load after the whole image is loaded and only for WebP still images with at least 100 pixels on the smallest dimension (width or height). The reported count for a sample represents the image size rounded to the nearest KiB. mcasas@chromium.org andrescj@chromium.org WebP file format parsed during Blink's header parsing phase (which should happen only once per image). urvang@chromium.org Image codec inferred during decode. The histogram is incremented when enough of the image has been downloaded to infer its width and height. Animated images are counted only once, regardless of how many frames there are. mamir@chromium.org web-identity-eng@google.com Records whether a create() request to the digital credentials API had transient user activation at the time of the call. mamir@chromium.org web-identity-eng@google.com Records whether a create() request to the digital credentials API successfully utilized a delegated token. Recorded when the API is invoked. mamir@chromium.org web-identity-eng@google.com Records whether a get() request to the digital credentials API had transient user activation at the time of the call. mamir@chromium.org web-identity-eng@google.com Records whether a get() request to the digital credentials API successfully utilized a delegated token. Recorded when the API is invoked. mamir@chromium.org web-identity-eng@google.com Records the reason that the digital identity interstitial dialog was dismissed. Recorded when the dialog is dismissed. mamir@chromium.org web-identity-eng@google.com Records the response mode of OpenID4VP requests to the digital credentials get API. Recorded when a get request is received. mamir@chromium.org web-identity-eng@google.com Records the result of a request to the digital identity API. Recorded when the request completes. pdr@chromium.org paint-dev@chromium.org Time spent computing IntersectionObserver observations in the Blink document lifecycle where the observation is for Display Locking. Recorded when an intersection observation is made for the Display Locking reason. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} nidhijaju@chromium.org chrome-loading@google.com Time taken to resolve a non-null URL in Document::CompleteURLWithOverride. This is recorded each time a URL is resolved, and can happen multiple times per navigation. This histogram only records metrics on machines with high-resolution clocks. chikamune@chromium.org chrome-loading@google.com Time between CommitNavigation to StartLoadingResponse in the renderer process. This is recorded on each new navigation for http or https URLs in OutermostMainFrame. chikamune@chromium.org chrome-loading@google.com Time spent processing DocumentLoader::CreateParserPostCommit() in the renderer process. This is recorded on each new navigation for http or https URLs in OutermostMainFrame. sisidovski@chromium.org chrome-loading@google.com Total time spent processing DocumentLoader::UpdateSubresourceLoadMetrics() until the document parsing is finished in the renderer process. This is recorded on each new navigation for http or https URLs in OutermostMainFrame. This histogram only records metrics on machines with high-resolution clocks. yigu@chromium.org web-identity-eng@google.com Records the outcomes and failure reasons of the high-level Email Verification Protocol (EVP) flow from autofill side. Recorded once per autofill attempt when a comformant form contains an email verification token field. yigu@chromium.org web-identity-eng@google.com Records when a form containing a verified email verification token is submitted. Recorded once when the form submission occurs after successful email verification. yigu@chromium.org web-identity-eng@google.com Records the outcome and dismissal reasons of the Email Verification Permission UI. Recorded once per permission UI showing when it is dismissed or accepted. yigu@chromium.org web-identity-eng@google.com Records the status of the IsVerifiable stage of an Email Verification Protocol (EVP) request. Recorded once per request when the IsVerifiable check completes. yigu@chromium.org web-identity-eng@google.com Records the status of the Verify stage of an Email Verification Protocol (EVP) request. Recorded once per request when the Verify operation completes. yigu@chromium.org web-identity-eng@google.com Records the time taken from the start of checking if an email is verifiable until the pre-prompt background verification check completes. Recorded once per pre-prompt check. yigu@chromium.org web-identity-eng@google.com Records the time taken from the start of the token verification call until the post-prompt token verification and fetch completes. Recorded once per verification request execution. olivierli@chromium.org woa-performance@google Whether a GetCookiesString request for a document returned the same result as the previous one. This is an experiment to determine the usefulness of a cookie cache in the Renderer. A sample is recorded for every GetCookiesString request sent from the Document owned CookieJar. aattar@google.com chrome-catan@google Whether the first cookie request is a SET, GET, or a check to determine if the cookies are enabled. This is an experiment to determine if we should return the version and/or the cookie string on SET. A sample is recorded on the first GetCookiesString, SetCookiesString, or CookiesEnabled request sent from the Document owned CookieJar. lizeb@google.com clank-performance-team@google When reading document.cookie, whether an IPC to the network service was required. Recorded at each cookie read access. npm@chromium.org web-identity-eng@google.com Records the 0-based position of the account that was selected by the user in a FedCM dialog. Recorded whenever the user clicks on a clickable account in the FedCM dialog. Currently only recorded on desktop. yigu@chromium.org web-identity-eng@google.com Records what kind of account fields we received from the server. Recorded once per accounts endpoint response, when we receive it. Considers the union of fields over all accounts that were received. For example, if one account has only an email address and one account only has a name, we record kOnlyNameAndEmail. yigu@chromium.org web-identity-eng@google.com Records the number of accounts that match an account label. Records at the time the accounts received by the FedCM API are filtered. yigu@chromium.org web-identity-eng@google.com Records a 1 each time an accounts dialog is shown. yigu@chromium.org web-identity-eng@google.com Records a 1 each time a request is sent to the accounts endpoint. yigu@chromium.org web-identity-eng@google.com Records the size of the accounts received from the IdP before applying any filters. Records one sample per accounts fetch if there's at least one account returned. yigu@chromium.org web-identity-eng@google.com Records the size of the accounts received from the IdP after applying the filters. Records one sample per accounts fetch if there's at least one account left after filtering. goto@chromium.org web-identity-eng@google.com Records the click source of the FedCM ambient UI. Recorded when the user clicks on the omnibox icon, chip or the anchored message's button or close button (but not when the user clicks on the avatar or strings). Multiple clicks can happen per API call. goto@chromium.org web-identity-eng@google.com Records when the FedCM ambient UI is shown. Recorded when the omnibox icon, chip and anchored message is displayed to the user (can happen more than once per API call). yigu@chromium.org web-identity-eng@google.com Records a 1 each time the parent activity is destroyed while the IDP login/continuation custom tab is shown. Recorded at most once per IDP login attempt and at most once per IDP continuation popup. yigu@chromium.org web-identity-eng@google.com Records whether an IDP returns an approved clients list in the response. Records at most one sample per IDP when the accounts response is parsed. yigu@chromium.org web-identity-eng@google.com Records the size of the approved clients list returned by IDP if applicable. Records at most one sample per IDP when the IDP includes `approved_clients` in the accounts response. npm@chromium.org yigu@chromium.org web-identity-eng@google.com Records whether the FedCM auto re-authn call is blocked because the auto re-authn content settings permission is disabled. Records at most one sample per FedCM API with auto re-authn enabled: some failures could occur before this metric is recorded. npm@chromium.org yigu@chromium.org web-identity-eng@google.com Records whether the FedCM auto re-authn call is blocked because the auto re-authn is under embargo, i.e. due to cooldown. Records at most one sample per FedCM API with auto re-authn enabled: some failures could occur before this metric is recorded. npm@chromium.org yigu@chromium.org web-identity-eng@google.com Records whether auto-reauthn was disabled by an embedder (e.g. due to an active actor task). Records at most one sample per FedCM API with auto re-authn enabled: some failures could occur before this metric is recorded. npm@chromium.org web-identity-eng@google.com Records whether the FedCM auto re-authn call is blocked because preventSilentAccess() has been called on this site. Records at most one sample per FedCM API with auto re-authn enabled: some failures could occur before this metric is recorded. npm@chromium.org yigu@chromium.org web-identity-eng@google.com Records whether there was zero, one, or multiple returning accounts when a site calls the FedCM API with autoReauthn: true. Records at most one sample per failed API call: some failures could occur before this metric is recorded. npm@chromium.org yigu@chromium.org web-identity-eng@google.com Records whether auto re-authn checks succeeded or not. The count would provide roughly the number of FedCM re-authn attempts, the true count would provide the number of times the re-authn UI is shown to the user, and the false count would provide the number of attempts where auto re-authn was blocked for some non-network related reason. Records at most one sample per FedCM API with auto re-authn enabled: some failures could occur before this metric is recorded. yigu@chromium.org web-identity-eng@google.com Records the outcome of the last account chooser the user sees in a button flow. Recorded at most once per button flow. For example, if the user selects an account on the account chooser and then clicks on the back button in the next dialog where an account chooser is shown again, leaving would only record one sample. yigu@chromium.org web-identity-eng@google.com Records the outcome of the last disclosure dialog the user sees in a button flow. Recorded at most once per button flow. For example, if the user clicks on the back button in the disclosure dialog and then clicks on an account where the disclosure dialog is shown again, leaving would only record one sample. yigu@chromium.org web-identity-eng@google.com Records the outcome of the loading dialog in button flow. Button flows always start with a loading dialog. Recorded once per button flow dialog destroyed. For example, when the dialog gets updated into an account chooser, the user can then select an account and subsequently cause the dialog to be destroyed once they are signed in. Upon destroyal, this metric is recorded. yigu@chromium.org web-identity-eng@google.com Records whether the user lost focus of the pop-up and whether it was the IDP or the user that closed the pop-up. Recorded once per loading state pop-up window closed in button mode. yigu@chromium.org web-identity-eng@google.com Records whether the user lost focus of the pop-up and whether it was the IDP or the user that closed the pop-up. Recorded once per use other account pop-up window closed in button mode. yigu@chromium.org web-identity-eng@google.com Records the reason that the FedCM dialog was closed (ex user clicked the dialog close button). Recorded when the FedCM dialog is closed without the user selecting an account. This includes instances where the close action is less intentional such as the virtual keyboard showing trigerring the dialog close. npm@chromium.org web-identity-eng@google.com Records the type of FedCM Android UI sheet that is being displayed when the user manually closes it. Records when the user closes this dialog, so at most once per API call that results in UI being shown. npm@chromium.org web-identity-eng@google.com Records the type of FedCM desktop UI sheet that is being displayed when the user manually closes it. Records when the user closes this dialog, so at most once per API call that results in UI being shown. yigu@chromium.org web-identity-eng@google.com Records whether the UI that a user is closing is the verify sheet or not on Android. Records at most one sample per API call. yigu@chromium.org web-identity-eng@google.com Records whether the UI that a user is closing is the verify sheet or not on desktop. Records at most one sample per API call. yigu@chromium.org web-identity-eng@google.com Records whether the continue_on popup was closed by the user or by the page finishing the flow. Recorded once per continue_on flow when the window gets closed (at most once per FedCM API call). If the RP tab/window gets closed before the popup, this will not get recorded. yigu@chromium.org web-identity-eng@google.com Records whether we could open the continue_on popup or if not, why we couldn't. Recorded once per continue_on popup request when we receive the request, at most once per FedCM API call. yigu@chromium.org web-identity-eng@google.com Records whether the client metadata endpoint response contained a value for `client_is_third_party_to_top_frame_origin`, and if so, what value. Only recorded if the iframe is cross-site. Recorded when we receive the client metadata endpoint response. npm@chromium.org web-identity-eng@google.com Records whether the FedCM UI is shown or not. Records a sample when the FedCM call is completed, i.e. at the same time as Blink.FedCm.Status.RequestIdToken. npm@chromium.org web-identity-eng@google.com Records the type of frame that invokes disconnect(). Records once for each disconnect() call. Possible values are: main frame, same-site iframe, and cross-site iframe, where the site of the caller frame is compared with the site of the main frame. npm@chromium.org web-identity-eng@google.com Records the number of accounts that match a domain hint. Records at the time the accounts received by the FedCM API are filtered. yigu@chromium.org web-identity-eng@google.com Records the outcome of the error dialog. Recorded at most once per API call when the error dialog is displayed. yigu@chromium.org web-identity-eng@google.com Records the type of error dialog shown. Recorded at most once per API call yigu@chromium.org web-identity-eng@google.com Records whether the error URL is same-origin, cross-origin but same-site or cross-site with the config URL. Recorded at most once per API call when a valid error URL is received. yigu@chromium.org web-identity-eng@google.com Records the type of token response received. Recorded at most once per API call when a token response received. yigu@chromium.org web-identity-eng@google.com Records the type of frame that invokes the FedCM API call. Records once per API call. Possible values are: main frame, same-site iframe, and cross-site iframe, where the site of the caller frame is compared with the site of the main frame. Same-origin iframes are recorded under the same-site category. npm@chromium.org web-identity-eng@google.com Records a 1 whenever a FedCM API call includes the nonce parameter. Records towards the beginning of the call, except for some cases where the call is immediately rejected. npm@chromium.org web-identity-eng@google.com Records a 1 whenever a FedCM API call includes the nonce parameter, but does not include it inside the params JSON blob. Records towards the beginning of the call, except for some cases where the call is immediately rejected. yigu@chromium.org web-identity-eng@google.com Records whether a user is signing up with the API. Records once per successful account fetch from the IdP server after the accounts filtering. If there are multiple accounts, record the user as sign-in if IdP claims at least one of the accounts is sign-in. If an IdP does not share the user's sign-in state, record the state stored in the browser. npm@chromium.org web-identity-eng@google.com Records the count of identity providers right before fetching endpoints with multi IDP flag enabled. It counts distint providers including the ones a user is not signed in to. Recorded at most once per API call before the browser starts fetching from the IdPs. yigu@chromium.org web-identity-eng@google.com Records the reason why the pop-up window is closed. Recorded once per opening of a pop-up window. yigu@chromium.org web-identity-eng@google.com Records the duration from when IdentityProvider.close() is called to when an updated accounts dialog is shown. Recorded once per IdentityProvider.close() being called given that an accounts dialog is shown. It is not recorded when the accounts dialog is never shown. yigu@chromium.org web-identity-eng@google.com Records the outcome of the mismatch dialog. Recorded once per the mismatch dialog being displayed. yigu@chromium.org web-identity-eng@google.com Records the outcome of the IDP sign-in pop-up window. Recorded once per pop-up window triggered by the user. yigu@chromium.org web-identity-eng@google.com Records whether the user selected account is for sign-in or not. Records at most one sample per API call. yigu@chromium.org web-identity-eng@google.com The Java object is tied to the bottomsheet availability so it may not be creatable when we need it. Records one sample each time we request creating the object in active mode. Recorded whenever native is talking with the Android side which could be multiple times per API call. yigu@chromium.org web-identity-eng@google.com The Java object is tied to the bottomsheet availability so it may not be creatable when we need it. Records one sample each time we request creating the object in passive mode. Recorded whenever native is talking with the Android side which could be multiple times per API call. yigu@chromium.org web-identity-eng@google.com When we fail a FedCM request because the `Page` is not primary, this records the lifecycle state the render frame host is in. npm@chromium.org web-identity-eng@google.com Records the number of accounts that match a login hint. Records at the time the accounts received by the FedCM API are filtered. yigu@chromium.org web-identity-eng@google.com Records a 1 each time a mismatch dialog is shown. npm@chromium.org web-identity-eng@google.com Records the type of mismatch dialog shown: whether it is the first mismatch shown for the current FedCM request or not, and whether there were hints requested in the current FedCM request or not. Records every time a mismatch dialog is shown. yigu@chromium.org web-identity-eng@google.com Records whether an API call is rejected due to there's an existing in-flight API call initiated by other IdPs. Recorded once per promise rejection because of an existing API call. yigu@chromium.org web-identity-eng@google.com Records the RP mode of two consecutive token requests. Recorded one sample every time when a new request is invoked while there is another one pending. yigu@chromium.org web-identity-eng@google.com Records the number of navigator.credentials.get() requests made in a document. Requests made when FedCM is disabled or when there is a pending FedCM request are not counted. Records at most one sample for every document where at least one navigator.credentials.get() request is made. No samples are recorded if the count is 0. The bucketing method used is base::UmaHistogramCounts100. yigu@chromium.org web-identity-eng@google.com Records the dialog type whenever we open a popup/custom tab. Recorded once per popup/custom tab opened. Values 0-4 of the enum will not be recorded for this histogram. npm@chromium.org web-identity-eng@google.com Records the RP Context used in a FedCM call. Records only when the RP Context flag is enabled. npm@chromium.org web-identity-eng@google.com Records the RP mode of the FedCM API. Recorded once per API call together with Blink.FedCm.Status.RequestIdToken when the request is fulfilled (either resolved or rejected). yigu@chromium.org web-identity-eng@google.com Records whether parameters or a non-default scope were specified in the FedCM call by the RP for any IDP. Records once per FedCM call when the request is initiated. Only recorded if parameters or non-default scopes were specified and FedCmAuthz is enabled. yigu@chromium.org web-identity-eng@google.com The action taken by the user. Recorded when the FedCM UI is destroyed, if the UI shown uses segmentation platform's UI volume recommendation. npm@chromium.org web-identity-eng@google.com Records the NetError from the accounts list endpoint. This histogram is only recorded when the IDP Sign-In Status of an IDP is marked as signout as a result of the accounts list endpoint not returning any accounts. yigu@chromium.org web-identity-eng@google.com Recorded each time that the privacy-policy link in the FedCM prompt is clicked. yigu@chromium.org web-identity-eng@google.com Recorded each time that the terms-of-service link in the FedCM prompt is clicked. yigu@chromium.org web-identity-eng@google.com Records the reason why there is no valid account in the response. Recorded at most once per accounts fetch when the response is received. yigu@chromium.org web-identity-eng@google.com Records the result of CSP checks in the FedCM API. npm@chromium.org web-identity-eng@google.com Records the status of a disconnect call to the FedCM API. Recorded once per IdentityCredential.disconnect() call, once the browser knows the result. yigu@chromium.org web-identity-eng@google.com Records whether the browser's knowledge of whether the user is signed-in to the IDP based on observing signin/signout HTTP headers matches the information returned by the accounts endpoint. Recorded at most once per navigator.credentials.get() call after sending the request to the accounts endpoint. npm@chromium.org web-identity-eng@google.com Records the mediation requirement of a request token call to the FedCM API. Records a sample when the FedCM call is completed, i.e. at the same time as Blink.FedCm.Status.RequestIdToken. yigu@chromium.org web-identity-eng@google.com Records the status of a request token call to the FedCM API. yigu@chromium.org web-identity-eng@google.com Records whether user sign-in states between IDP and browser match. Records a sample after browser receives an account list from IDP prior to showing UI. yigu@chromium.org web-identity-eng@google.com Records the duration from when an accounts dialog is shown to when dialog is destroyed e.g. user selects an account, user closes the dialog, or the developer aborts the request. This duration includes the time when the accounts dialog is inactive e.g. user switches to a different tab. Samples are at most 10 minutes. Records a sample when an accounts dialog which is not the verifying UI for auto re-authentication is shown. In a previous definition, Blink.FedCm.Timing.AccountsDialogShownDuration, a sample was recorded when an accounts dialog (including the verifying UI) is shown. yigu@chromium.org web-identity-eng@google.com Records the time from when the accounts dialog is shown to when the user explicitly closes the dialog without selecting any account. Only records a sample when the user sees and *explicitly* closes the dialog. For instance, the accounts dialog being closed as a result of the user focusing a different part of the browser UI does not count as explicitly closing the dialog. yigu@chromium.org web-identity-eng@google.com Records the time from when an account is selected to when the continue_on response is received. Only records a sample if and when IdentityProvider.resolve is called from the continue_on popup. Recorded at most once per FedCM call. yigu@chromium.org web-identity-eng@google.com Records the overall time from when the API is called to when the token is received through IdentityProvider.resolve. Only records a sample if and when IdentityProvider.resolve is called from the continue_on popup. Recorded at most once per FedCM call. yigu@chromium.org web-identity-eng@google.com Records the time from when the accounts dialog is shown to when the user presses the Continue button. Only records a sample when the user sees and taps the "Continue" button on the dialog. npm@chromium.org web-identity-eng@google.com Records the time from when the disconnect() request is received by the browser process to the time in which it is completed, e.g. right before sending the response back to the renderer process. Records one sample for every disconnect call which sends a disconnect request. yigu@chromium.org web-identity-eng@google.com Records the time from when the user presses the Continue button to when the token response is received. Only records a sample when the token response is received. yigu@chromium.org web-identity-eng@google.com Records the duration from when a mismatch dialog is shown to when user triggers IDP sign-in pop-up window or when dialog is destroyed e.g. user closes the dialog. This duration includes the time when the mismatch dialog is inactive e.g. user switches to a different tab. Samples are at most 10 minutes. Records a sample when a mismatch dialog is shown. yigu@chromium.org web-identity-eng@google.com Records the time from when a call to the API was made to when the accounts dialog is ready to be shown. Only records a sample when the dialog is shown. yigu@chromium.org web-identity-eng@google.com Records the time from when an accounts request is made to when the accounts fetch response is received. Records one sample whenever it's ready to show the accounts UI which could happen multiple times per API call. yigu@chromium.org web-identity-eng@google.com Records the time from when a client metadata request is made to when the fetch response is received. Records one sample whenever it's ready to show the accounts UI which could happen multiple times per API call. If the client metadata request is skipped, e.g. for returning users, records a zero instead. yigu@chromium.org web-identity-eng@google.com Records the time from when a call to the API was made to when the well-known and config files are fetched. Records one sample whenever it's ready to show the accounts UI which could happen multiple times per API call. yigu@chromium.org web-identity-eng@google.com Records the overall time from when the API is called to when the token response is received. Only records a sample when the token response is received. yigu@chromium.org web-identity-eng@google.com Records the time from when a call to the API was made to when the well-known and config files fetch response are received. Records a sample when the response is received. Recorded at most once per API call. The histogram Blink.FedCm.Timing.ShowAccountsDialogBreakdown.WellKnownAndConfigFetch is different because it only records the fetch time when the accounts dialog is ready to show. i.e. there could be much fewer samples recorded in that histogram because many well-known and config file fetches take too much time such that the API call would fail without fetching the rest endpoints. i.e. the distribution of the two histograms could be statically different. npm@chromium.org web-identity-eng@google.com Records the outcome of the user clicking on the "use other account" button in FedCM. Recorded once per API call when the user attempts to login with a new account via the "use other account" button, when the request is fulfilled (either resolved or rejected). Only considers the last "use other account" button clicked if there are multiple attempts. npm@chromium.org web-identity-eng@google.com Records the number of accounts returned by a getUserInfo() call. Records once per request with a successful accounts fetch. Records when the request is completed. npm@chromium.org web-identity-eng@google.com Records the status of a getUserInfo() call. Records once per request when the request is completed, right before running the callback. npm@chromium.org web-identity-eng@google.com Records the time from when a getUserInfo() call is received on the browser process to the time it is completed. Records once per request with a successful accounts fetch. Records when the request is completed. yigu@chromium.org web-identity-eng@google.com Records the outcome of the verifying dialog. Recorded once per verifying dialog shown when the FedCM call is completed, i.e. at the same time as Blink.FedCm.Status.RequestIdToken. yigu@chromium.org web-identity-eng@google.com Records whether the WebContents is active when the browser is ready to show the accounts dialog to the user. Records at most one sample per API call. yigu@chromium.org web-identity-eng@google.com Records whether the WebContents is visible when the browser is ready to show the accounts dialog to the user. Records at most one sample per API call. npm@chromium.org web-identity-eng@google.com sureshpotti@microsoft.com Recorded once per FedCM well-known discovery attempt when the FedCmWebIdentitySubdomain feature is enabled. True if the https://web-identity.well-known.<eTLD+1>/.well-known/web-identity fetch returned a usable response (success ParseStatus and provider_urls.size() <= 1). False if it fell back to the apex https://<eTLD+1>/.well-known/web-identity URL due to fetch failure (network error, malformed JSON, or non-dictionary response) or more than one provider_url. yigu@chromium.org web-identity-eng@google.com Records a 1 whenever a FedCM API call includes a well-known file without accounts or login_url but where the config file has client_metadata. Records when the config and well-known files have been fetched. shivanisha@chromium.org dom@chromium.org averge@chromium.org Records creation and navigation outcomes for a fenced frame. This is written either when a fenced frame navigates successfully, or when some issue causes either the creation or navigation of a fenced frame to stop. shivanisha@chromium.org dom@chromium.org lbrady@google.com If a fenced frame failed to load due to incorrectly set sandbox flags, records whether or not the fenced frame was attempting to be created in a top-level frame. This is written whenever a fenced frame creation fails due to sandbox flags not being permissive enough. shivanisha@chromium.org dom@chromium.org lbrady@google.com Records if a fenced frame's size is set after the size had been frozen. The size of a fenced frame is frozen once it navigates to a page, and results in the inner contents being scaled to fit the new dimensions without telling it the frame's new size. This is only logged up to one time throughout a fenced frame's life, and will not be logged if the size is never set after being frozen. shivanisha@chromium.org dom@chromium.org lbrady@google.com Records whether an opaque-ads fenced frame's size was changed because the provided size was not in the list of allowed sizes. This will not log anything for a default mode fenced frame. This will log when an opaque-ads fenced frame has its size frozen (which is triggered when it navigates to its first page). shivanisha@chromium.org dom@chromium.org lbrady@google.com If a fenced frame failed to load due to incorrectly set sandbox flags, records which sandbox flags that should be unsandboxed ended up being sandboxed, causing the load to fail. This is written whenever a fenced frame creation fails due to sandbox flags not being permissive enough. If multiple sandbox flags are incorrectly set, it will log every sandbox flag that was not set correctly for each failed load. cduvall@chromium.org csharrison@chromium.org chrome-loading@google.com The total microseconds spent in ResourceFetcher::requestResource for {SchemeOrPreload}. This histogram only records metrics on machines with high-resolution clocks. Warning: this histogram was expired from 2023-05-02 to 2023-12-12; data may be missing. cduvall@chromium.org jam@chromium.org chrome-loading@google.com Measures the time it takes to fetch queued subresource preloads for a document in a {FrameType}. Logged every time preloads need to be fetched after the {IsInitial} preload scan. Warning: this histogram was expired from 2023-05-29 to 2023-12-04; data may be missing. horo@chromium.org net-dev@chromium.org Time spent for decoding the fonts on the background thread. Recorded when a font resource creates its decoded data. This histogram is only logged in platforms with high-resolution clocks. gjc@chromium.org net-dev@chromium.org Time spent for decoding the fonts. Recorded when a font resource creates its decoded data. This histogram is only logged in platforms with high-resolution clocks. nidhijaju@chromium.org chrome-loading@google.com Time spent identifying the best match for a font face to use given a family name and style, including the mojo calls. This is recorded right after the best match is found. This histogram is only logged in platforms with high-resolution clocks. nidhijaju@chromium.org chrome-loading@google.com Time spent opening a font file, including the mojo calls. This is recorded right after the font file is opened. This histogram is only logged in platforms with high-resolution clocks. andresrperez@chromium.org anthonyvd@chromium.org chrome-catan@google.com Obsolete: rebucketed and renamed to LastResortAttemptsUntilStaticMatch2 in M144. Records how many attempts it took to find a last resort fallback font. This is attempted when no other specified font could be instantiated to render a character. The last resort font from the FontDescription is tried first, and then hardcoded system fonts if needed. This histogram counts those hardcoded fonts attempts. If no fallback is found, the overflow bucket is recorded. andresrperez@chromium.org anthonyvd@chromium.org chrome-catan@google.com Records how many attempts it took to find a last resort fallback font. This is attempted when no other specified font could be instantiated to render a character. The last resort font from the FontDescription is tried first, and then hardcoded system fonts if needed. This histogram counts those hardcoded fonts attempts. If no fallback is found, the overflow bucket is recorded. Replaced Blink.Fonts.LastResortAttemptsUntilStaticMatch in M144 and is a re-bucketed version of that histogram. andresrperez@chromium.org anthonyvd@chromium.org chrome-catan@google.com Records whether a last resort fallback font was found. This is attempted when no other specified font could be instantiated to render a character. The last resort font from the FontDescription is tried first, and then hardcoded system fonts if needed. suzukikeita@chromium.org chrome-loading@google.com Records the duration of SkFontMgr::matchFamilyStyle calls on the main thread. Recorded when a typeface is looked up by family name during font creation. This measures the latency of querying the operating system's font manager. For "cold" lookups (first time a font is loaded), this includes disk I/O to read and parse font files, and potential sandbox/IPC overhead on platforms that use a font service (such as Windows, macOS, or Linux). For "warm" lookups, this measures the fast Skia-side in-memory cache-resolution latency. Note that this metric is only recorded on clients with high-resolution clocks. suzukikeita@chromium.org chrome-loading@google.com Records the duration of SkFontMgr::matchFamilyStyleCharacter calls on the main thread. Recorded when a fallback typeface is looked up for a specific character or locale. This measures the latency of system font fallback resolution. For "cold" lookups, this includes the expensive OS-level scanning of multiple system font character maps (cmaps) to find a supporting font, disk I/O, and potential sandbox/IPC overhead on platforms that use a font service (such as Windows, macOS, or Linux). For "warm" lookups, it measures the fast Skia-side in-memory cache-resolution latency. Note that this metric is only recorded on clients with high-resolution clocks. drott@chromium.org layout-dev@chromium.org Tracks adoption ratio of variable fonts compared to conventional (in the sense of non-variable) web fonts. Recorded at the time of instantiating a Skia SkTypeface from the successfully decoded web font blob. pdr@chromium.org paint-dev@chromium.org Time spent computing layouts due to Javascript value requests. Specifically, time spent in Blink Document::UpdateStyleAndLayout. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} carlosk@chromium.org harringtond@chromium.org feed@chromium.org Android only: The time spent by the renderer main thread to collect and send back to the browser process a web page's canonical link. This is recorded every time the renderer fulfills a request for this information. This histogram is only logged in platforms with high-resolution clocks. pdr@chromium.org paint-dev@chromium.org Time spent processing rAF-aligned input during a main frame update. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} kyraseevers@chromium.org tiffanysong@google.com Each time we determine an anchor element's VisitedLinkState, we record whether the link will be styled as visited. :visited is true and :unvisited is false. kyraseevers@chromium.org tiffanysong@google.com Each time we add an origin, salt pair to VisitedLinkReader, this histogram records whether the salt came from the VisitedLinkNavigationThrottle or the VisitedLinkNotificationSink IPC. The VisitedLinkNavigationThrottle salts are true and the VisitedLinkNotificationSink salts are false. pdr@chromium.org paint-dev@chromium.org Time spent preparing the document for hit testing during a main frame update. It does not compute the actual hit test time, rather the style, layout, compositing, etc time to prepare for the test. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} chikamune@chromium.org dom-dev@google.com chrome-loading@google.com Number of bytes after which the html fast-path parser aborted. This does not count the bytes the fast-path parser did not look at. We measure this since aborting the fast-path and falling back to the normal parser is wasted effort. This is recorded after every invocation of Element::setInnerHTML. WARNING: This histogram was expired from 2024-04-28 to 2026-04-09; data may be missing. chikamune@chromium.org dom-dev@google.com chrome-loading@google.com The amount of time the innerHTML fast-path parser took before reaching input it could not handle. This is recorded when Element::setInnerHTML is called, and at the moment when the input it can't handle is reached. This histogram does not record metrics on machines with low-resolution clocks. WARNING: This histogram was expired from 2024-04-28 to 2026-04-09; data may be missing. chikamune@chromium.org dom-dev@google.com chrome-loading@google.com Captures the result of the innerHTML fast-path parser. This is recorded every time Element::setInnerHTML is called and only if the innerHTML fast-path parser is enabled. WARNING: This histogram was expired from 2025-03-30 to 2026-04-09; data may be missing. chikamune@chromium.org dom-dev@google.com chrome-loading@google.com Number of bytes successfully parsed by the innerHTML fast-path parser. This is recorded if the innerHTML fast-path parser completes successfully and is logged at the end of parsing the content supplied to Element::setInnerHTML. WARNING: This histogram was expired from 2024-04-28 to 2026-04-09; data may be missing. chikamune@chromium.org dom-dev@google.com chrome-loading@google.com The time it takes for the innerHTML fast-path parser to complete processing text from Element::setInnerHTML. This is recorded if fast-path innerHTML is enabled, and completes successfully. This histogram does not record metrics on machines with low-resolution clocks. WARNING: This histogram was expired from 2024-04-28 to 2026-04-09; data may be missing. chikamune@chromium.org dom-dev@google.com chrome-loading@google.com The total time it takes to complete parsing of Element::setInnerHTML. This includes the time taken by the fast-path parser (even if it failed). This is recorded after every invocation of Element::setInnerHTML. This histogram does not record metrics on machines with low-resolution clocks. chikamune@chromium.org dom-dev@google.com chrome-loading@google.com This is logged if the innerHTML fast-path parser fails because an {Type} tag is encountered. This value is an aggregate (bitmask) that identifies which {Type} groupings were encountered. Ideally a single bitmask would be logged, but the dashboard doesn't deal well with large bitmasks. The individual groupings are recorded via one of Blink.HTMLFastPathParser.*.Mask0-3 (* is one of UnsupportedContextTag or UnsupportedTag). WARNING: This histogram was expired from 2023-12-07 to 2026-04-09; data may be missing. chikamune@chromium.org dom-dev@google.com chrome-loading@google.com See description of Blink.HTMLFastPathParser.*.CompositeMask for an overview of this including when it is logged. The value is a bitmask that identifies which tags were encountered: {MaskBit} WARNING: This histogram was expired from 2023-12-07 to 2026-04-09; data may be missing. masonf@chromium.org dom-dev@google.com The number of chunks parsed while loading a page. Each chunk represents a piece of the HTML content that can be loaded without blocking or yielding. This is only reported for http/https urls. This histogram does not record metrics on machines with low-resolution clocks. masonf@chromium.org dom-dev@google.com The number of characters processed while parsing a document during page load. This is computed as the sum of all the characters appended to the input string sequence for the parser. This is only reported for http/https urls. This histogram does not record metrics on machines with low-resolution clocks. sisidovski@chromium.org chrome-loading@google.com Records whether pausing the parser is attempted after resuming the parser. This can happen when the site sends a signal to pause the parser after sending another signal to resume the parser. This is recorded when the signal to pause the parser is dispatched via the user timing. This is measured for the experiment in crbug.com/416543903. sisidovski@chromium.org chrome-loading@google.com Records whether the parser is resumed before it's actually paused. This can happen when the site sends a signal to resume the parser without sending another signal to pause the parser. This is measured for the experiment in crbug.com/416543903. masonf@chromium.org dom-dev@google.com While parsing a page, the maximum time used to parse a single chunk before being blocked, yielding or completing, in microseconds. This does not include time spent executing script. This is only reported for http/https urls. This histogram does not record metrics on machines with low-resolution clocks. masonf@chromium.org dom-dev@google.com While parsing a page, the minimum time used to parse a single chunk before being blocked, yielding or completing, in microseconds. This does not include time spent executing script. This is only reported for http/https urls. This histogram does not record metrics on machines with low-resolution clocks. masonf@chromium.org dom-dev@google.com While parsing a page, the total amount of time spent in the HTMLDocumentParser performing parsing, in microseconds. This does not include time spent executing script. This is only reported for http/https urls. This histogram does not record metrics on machines with low-resolution clocks. sisidovski@chromium.org chrome-loading@google.com The total number of queued preload requests created by the preload scanner when HTMLDocumentParser performing parsing. This is only reported for http/https urls. This is reported at the parse end. This histogram does not record metrics on machines with low-resolution clocks. sisidovski@chromium.org chrome-loading@google.com Records whether the parser yield is resumed by the user timing signal via performance.mark(), or resumed by the scheduled timeout. This is measured for the experiment in crbug.com/416543903. masonf@chromium.org dom-dev@google.com The time between when the first content has been received by the HTMLDocumentParser and when parsing begins. This is effectively measuring the time between when SchedulePumpTokenizer() is called, and when the task runs that does the parsing. This is only reported for http/https urls. This histogram does not record metrics on machines with low-resolution clocks. WARNING: This histogram was expired from 2024-12-24 to 2025-09-12; data may be missing. masonf@chromium.org dom-dev@google.com While parsing a page, the average number of tokens parsed across all chunks. This is only reported for http/https urls. This histogram does not record metrics on machines with low-resolution clocks. masonf@chromium.org dom-dev@google.com While parsing a page, the maximum number of tokens parsed as a chunk. This is only reported for http/https urls. This histogram does not record metrics on machines with low-resolution clocks. WARNING: This histogram was expired from 2025-01-31 to 2025-09-12; data may be missing. masonf@chromium.org dom-dev@google.com While parsing a page, the minimum number of tokens parsed as a chunk. This is only reported for http/https urls. This histogram does not record metrics on machines with low-resolution clocks. WARNING: This histogram was expired from 2025-01-31 to 2025-09-12; data may be missing. masonf@chromium.org dom-dev@google.com While parsing a page, the total number of tokens parsed across all chunks. This is only reported for http/https urls. This histogram does not record metrics on machines with low-resolution clocks. sisidovski@chromium.org chrome-loading@google.com The count how many times the parser is yielded by the signal from the site. This is triggered via performance.mark(). This is measured for the experiment in crbug.com/416543903. masonf@chromium.org dom-dev@google.com While parsing a page, the average time between parsing of two chunks, in microseconds. Will not be recorded if only one chunk was parsed. This is only reported for http/https urls. This histogram does not record metrics on machines with low-resolution clocks. WARNING: This histogram was expired from 2025-01-31 to 2025-09-12; data may be missing. sisidovski@chromium.org chrome-loading@google.com The time between when the parser has been paused by the signal from the site and when it's resumed. Both pause and resume signals are triggered via performance.mark(). This is measured for the experiment in crbug.com/416543903. WARNING: This histogram was expired from 2025-02-01 to 2025-09-12; data may be missing. masonf@chromium.org dom-dev@google.com While parsing a page, the maximum time between parsing of two chunks, in microseconds. Will not be recorded if only one chunk was parsed. This is only reported for http/https urls. This histogram does not record metrics on machines with low-resolution clocks. WARNING: This histogram was expired from 2025-01-31 to 2025-09-12; data may be missing. masonf@chromium.org dom-dev@google.com While parsing a page, the minimum time between parsing of two chunks, in microseconds. Will not be recorded if only one chunk was parsed. This is only reported for http/https urls. This histogram does not record metrics on machines with low-resolution clocks. WARNING: This histogram was expired from 2025-01-31 to 2025-09-12; data may be missing. vmpstr@chromium.org khushalsagar@chromium.org This histogram records whether an image has been decoded by a single client or by multiple clients during decoding. For an image to be decoded, it is always requested by a client with a client_id. If an image has been requested by multiple clients, it would be decoded more than once. The histogram value is recorded at most two times per image (once per bucket). There will be overlap for two buckets, but we can get the the number of an image decodeded by only one client with the difference between two buckets. This helps us to know how often images get duplicated decoding in real life. pdr@chromium.org paint-dev@chromium.org Time spent updating layers in the compositor thread during a ProxyMain::BeginMainFrame. It is the actual work time, excluding time spent waiting for the compositor thread to begin. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} pdr@chromium.org paint-dev@chromium.org Time spent computing IntersectionObserver observations in the Blink document lifecycle. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} szager@chromium.org paint-dev@chromium.org Number of IntersectionObserver computations for blink-internal consumers that were performed during the course of a single main frame. {BlinkUpdateTimeSuffixes} szager@chromium.org paint-dev@chromium.org Number of IntersectionObserver computations for javascript consumers that were performed during the course of a single main frame. {BlinkUpdateTimeSuffixes} pdr@chromium.org paint-dev@chromium.org Time spent preparing the document to answer a Javascript API query or request. Examples include scrool offset requests, queries of layout dimensions, SVG path queries, etc. Includes time spent in style update and layout, or via EnsurePaintLocationDataValidForNode. Some of these calls may be servicing internal needs, but we have no way to disambiguate those from JS driven activity. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} pdr@chromium.org paint-dev@chromium.org Time spent computing IntersectionObserver observations in the Blink document lifecycle where the observation is due to a Javascript request. Recorded when an intersection observation is made for the Javascript reason. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} veluca@chromium.org helmut@januschka.com src/third_party/blink/renderer/platform/image-decoders/jxl/OWNERS JXL pixel decode throughput. A sample is recorded when each JXL frame finishes decoding, and the decode time includes only time spent inside jxl-rs. joedow@chromium.org garykac@chromium.org Records each call to the navigator.keyboard{Lock|Unlock} methods. pdr@chromium.org layout-dev@chromium.org Time spent updating layout in the Blink document lifecycle. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} pdr@chromium.org paint-dev@chromium.org Time spent computing IntersectionObserver observations in the Blink document lifecycle where the observation is due to a Lazy Image Loading request. Recorded when an intersection observation is made to determine when a lazy loaded image should be loaded. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} chikamune@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Records the reason for the image boost when image fetch priority was increased in ResourceFetcher. Currently there are two major ways to increase the image fetch priority. This UMA helps us to understand which way is more common. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Counts the number of the reported font URLs in the cross site. This metric is emitted when predicted font URLs are appended to the PreconnectPrediction structure for prefetching. sisidovski@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Records the total count of preloads, which are deferred to load based on the potentially unused preload signals from the LCP Critical Path Predictor (LCPP) database. This metric is recorded at the same time when "Renderer.Preload.UnusedResourceCount" is recorded. That is scheduled to a few seconds later after the preload request is created. sisidovski@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Counts the number of deferred preloads per ResourceType. {Trigger} are recorded. This is recorded when deferred preloads starts loading. sisidovski@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Counts the number of deferred preloads per ResourceType. {Trigger} are recorded only when those were predicted as not used in the page, but actually used. This is recorded when deferred preloads starts loading. sisidovski@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Indicates the prediction result. If true, the deferred preload is still not used at least at the timing when the scheduled loading task is executed (the actual timing is defined based on the kLCPPDeferUnusedPreloadTiming feature param), which means the prediction succeeded. If false, that means the prediction failed. This is recorded when deferred preloads starts loading. {Trigger} are the target. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Record a connection type when there is fonts to be prefetched or preloaded according to the LCPP prediction. There is a configuration on the bandwidth limit. If the bandwidth is below the limit, an empty fonts list is returned. It is used for disabling the prefetch/preload. kouhei@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Measures the size of the identified LCP element locator in bytes. The metric is emitted before the element locator is serialized to the LCP Critical Path Predictor (LCPP) database, so it can inform us if the current ElementLocator encoding scheme is reasonable in terms of its output size. chikamune@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Measures the number of resources identified to be influencing the load of the LCP element. The metric is emitted before the influencer list is serialized to the LCP Critical Path Predictor (LCPP) database. chikamune@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Measures the size of the longest URL influencing the load of the LCP element. The metric is emitted before the influencer list is serialized to the LCP Critical Path Predictor (LCPP) database. chikamune@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Counts the number of URLs matched during a subsequent page load, from the set of URLs identified to be influencing the load of the LCP element. The metric is emitted as LCP Critical Path Predictor calculates dependencies of the current LCP candidate. chikamune@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Measures the percent of matched LCP influencer URLs against the predicted set of URLs. The metric is emitted as LCP Critical Path Predictor calculates dependencies of the current LCP candidate. chikamune@chromium.org yyanagisawa@chromium.org kouhei@chromium.org Record when the page's main frame starts preloading {SubresourceType}. Measures the duration between when the navigation starts, to when preloading subresource is started. kouhei@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Measures the number of resources that got priority boost from LCPP signals for page loads with non-empty LCPP hint data. This metric is emitted for every page load when it is about to emit "onload" event. The metric would tell us how much LCP Critical Path Predictor (LCPP), is affecting page load in the wild. This metric is not emitted 1) when there are no LCPP signals, 2) when the document is not in an OutermostMainFrame, or 3) when the document URL is not a HTTP family. yoichio@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Counts the number of actual preconncted origins for image element. The metric is emitted as LCP Critical Path Predictor finalizes LCP candidate dependencies. chikamune@chromium.org yoichio@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Counts the number of origins preconnected using the LCPP prediction. This metric is emitted when predicted origins are appended to the PreconnectPrediction structure during PrepareForPageLoad. chikamune@chromium.org yoichio@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Counts the number of predicted preconnect origins that matched against the LCP image. The metric is emitted as LCP Critical Path Predictor finalizes LCP candidate dependencies. This metric should indicate "1" in the ideal case, i.e., exactly one prediction matched the actual LCP origin. Higher than one would indicate duplicate predictions and should not be normal. chikamune@chromium.org yoichio@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Records the percent of LCP origins matched against the LCPP predicted preconnect origins. The metric is emitted as LCP Critical Path Predictor finalizes LCP candidate dependencies. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Counts the number of font URLs to be prefetched using the LCPP prediction. This metric is emitted when predicted font URLs are appended to the PreconnectPrediction structure for prefetching. yoichio@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Ratio of the same site subresource URLs in prefetching subresources using LCPP prewaming HttpDiskCache database. This metric is emitted when predicted subresources URLs are appended to the PreconnectPrediction structure for prefetching. yoichio@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Counts the number of {Site} subresource URLs in prefetching subresources using LCPP prewaming HttpDiskCache database. This metric is emitted when predicted subresources URLs are appended to the PreconnectPrediction structure for prefetching. yoichio@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Records whether LCP DB misses the type of subresource to be prefeteched. We expect this rarely happens except experiment starting, where the DB scheme changes. This metric is emitted when predicted subresources URLs are appended to the PreconnectPrediction structure for prefetching. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Counts the number of fetched font URLs to be preloaded in the renderer using the LCPP prediction. This metric is emitted just after the fonts are preloaded. chikamune@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Records whether prewarming HttpDiskCache successfully warmed up the disk or not. If the cache entry does not exist, false will be recorded. This is recorded on each HttpDiskCache prewarming attempt and only if the {PrewarmMethod} prewarming method was used. chikamune@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Records the predicted subresource URL count that is predicted by the HttpDiskCachePrewarming feature. This is recorded on each navigation. yoichio@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Ratio of the same site subresource URLs in the all reported subresource URLs in prewarming HttpDiskCache. This is recorded on each navigation. yoichio@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Counts the number of {Site} subresource URLs in prewarming HttpDiskCache learnt to local LCP database. This is recorded on each navigation. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Measures the number of web fonts that was fetched in the page and recorded to the LCPP database. This metric is emitted for every page load when it is about to emit "onload" event. Unlike Blink.LCPP.ReportedFontCount, this metric counts the number of fonts after filitering with the max fonts per site. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Count the number of font URLs that hits the prediction using LCPP. This metric is emitted for every page load when it is about to emit "onload" event. Unlike Blink.LCPP.ReportedFontCount, this metric counts the number of fonts after filitering with the max fonts per site. If the page does not have web fonts, they are excluded from this metrics. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Counts the number of fetched font URLs that are already recorded in the LCPP database. This metric is emitted for every page load when it is about to emit "onload" event. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Count the number of font URLs that has already recorded in the LCPP database. This metric is emitted for every page load when it is about to emit "onload" event. Unlike Blink.LCPP.ReportedFontCount, this metric counts the number of fonts after filitering with the max fonts per site. If the page does not have web fonts, they are excluded from this metrics. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Ratio of the font URLs that hits the prediction using LCPP out of all fetched fonts. This metric is emitted for every page load when it is about to emit "onload" event. If the page does not have web fonts, they are excluded from this metrics. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Ratio of the font URLs that hits the prediction using LCPP out of fetched fonts to be recorded. This metric is emitted for every page load when it is about to emit "onload" event. Unlike Blink.LCPP.ReportedFontCount, this metric counts the number of fonts after filitering with the max fonts per site. If the page does not have web fonts, they are excluded from this metrics. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Measures the percent of font URLs in the LCPP database against all fetched font URLs. This metric is emitted for every page load when it is about to emit "onload" event. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS The number of the font URLs that reentered to LCPP out of fetched fonts to be recorded. This metric is emitted for every page load when it is about to emit "onload" event. If the page does not have web fonts, they are excluded from this metrics. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Ratio of the font URLs that reentered to LCPP out of fetched fonts to be recorded. This metric is emitted for every page load when it is about to emit "onload" event. Unlike Blink.LCPP.ReportedFontCount, this metric counts the number of fonts after filitering with the max fonts per site. If the page does not have web fonts, they are excluded from this metrics. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Measures the longest font URL that was fetched in the page and recorded to the LCPP database. This metric is emitted for every page load when it is about to emit "onload" event. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Measures the number of web fonts that was fetched in the page. This metric is emitted for every page load when it is about to emit "onload" event. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Ratio of the same site font URLs in the all reported font URLs. This metric is emitted when predicted font URLs are appended to the PreconnectPrediction structure for prefetching. yyanagisawa@chromium.org src/third_party/blink/renderer/core/lcp_critical_path_predictor/OWNERS Counts the number of the reported font URLs in the same site. This metric is emitted when predicted font URLs are appended to the PreconnectPrediction structure for prefetching. tnak@chromium.org chrome-loading@google.com Counts usage of Link header in HTTP responces to subresource fetches. This histogram are recorded when a HTTP responce with Link headers is received. For details, see PreloadHelper::LoadLinksFromHeader(). pdr@chromium.org paint-dev@chromium.org Time between a BeginMainFrame and paint results commit in Blink. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} pdr@chromium.org paint-dev@chromium.org Time spent computing IntersectionObserver observations in the Blink document lifecycle where the observation is for a media element. Recorded when an intersection observation is made attributed to a media element (such as video auto-play). Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} gjc@chromium.org blink-network-dev@chromium.org Type of cached non-static resources reused across Documents when loading a new document in the same renderer. Recorded by the ResourceFetcher when a resource is requested for a new document and is not static. If the resource is available from the memory cache and previously loaded by a different document, the resource will be considered as reused. Repetitive requests to the same url in a document will be counted only once. gjc@chromium.org blink-network-stack@chromium.org Size of the resource strong references saved to the memory cache. Recorded when the user navigates to a newpage under the same origin. Recorded only with feature flag MemoryCacheStrongReference. hiroshige@chromium.org lizeb@chromium.org gjc@chromium.org loading-dev@chromium.org RevalidationPolicy used for non-preloading requests of async and defer scripts. https://crbug.com/1043679 https://crbug.com/1376866 Warning: this histogram was expired from 2021/10/25 to 2022/12/09; data may be missing. hiroshige@chromium.org gjc@chromium.org loading-dev@chromium.org RevalidationPolicy used for requests that hit Resource only referenced from MemoryCache for each resource type ({ResourceType}). https://crbug.com/579496 Requests counted by this are also counted by Blink.MemoryCache.RevalidationPolicy2 or Blink.MemoryCache.RevalidationPolicy2.Preload. https://crbug.com/1376866 hiroshige@chromium.org gjc@chromium.org loading-dev@chromium.org RevalidationPolicy used for preloading requests for each resource type ({ResourceType}). https://crbug.com/579496 https://crbug.com/1376866 hiroshige@chromium.org gjc@chromium.org loading-dev@chromium.org RevalidationPolicy used for non-preloading requests for each resource type. https://crbug.com/579496 https://crbug.com/1376866 sisidovski@chromium.org chrome-loading@google.com Total time spent processing `ResourceFetcher::DidLoadResourceFromMemoryCache()` until the document parsing is finished in the renderer process. This is recorded on each navigation for http or https URLs in OutermostMainFrame. The UMA is reported when the document finishes parsing. This histogram only records metrics on machines with high-resolution clocks. kumagi@google.com gaul@google.com paint-dev@google.com Whether element extraction phase succeed without consuming entire 4ms time budget in MobileFriendlinessChecker::ComputeBadTapTargetsRatio() evaluation. This metric recorded at 5 seconds after first MainFrame's commit and periodically every 1 minutes after that. mfoltz@chromium.org agentic-web-platform-team@google.com Records the total number of tools registered with document.modelContext after some delay after the first registration. This is recorded once per top level document. It provides an approximate number of tool registrations, assuming that declarative tool registrations happen on the main document load, followed by some imperative tool registrations in script. nidhijaju@chromium.org chrome-loading@google.com The total length of {Type} data URLs. This is recorded per resource request when the parsing is completed. Warning: this histogram was expired from 2025-03-28 to 2025-06-25; data may be missing. nidhijaju@chromium.org chrome-loading@google.com The total microseconds spent in network_utils::ParseDataURL and is recorded per resource request for {CharacterCount} of {Type} when the parsing is completed. This histogram only records metrics on machines with high-resolution clocks. cduvall@chromium.org jam@chromium.org The time taken to query the notification permission status. This is recorded each time a sync call is made. aaronhk@chromium.org fserb@chromium.org Records the creation of a new offscreen canvas in OffscreenCanvas::Create. aaronhk@chromium.org fserb@chromium.org Stores the square root of the number of pixels in a new or resized offscreen canvas. Emitted from CanvasRenderingContextHost::RecordCanvasSizeToUMA. aaronhk@chromium.org fserb@chromium.org Records a call to transfer a canvas offscreen. aaronhk@chromium.org fserb@chromium.org Records when an offscreen canvas has been transferred from the main thread to a worker. khushalsagar@chromium.org vmpstr@chromium.org Computes the area of img, video or canvas elements which is painted outside the content rect if 'overflow' property is respected on these elements. pdr@chromium.org paint-dev@chromium.org Time spent updating paint in the Blink document lifecycle. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} cduvall@chromium.org jam@chromium.org Time spent parsing style sheets for a document. Only recorded when the CSSParserContext has a document available. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} tungnh@chromium.org src/components/permissions/PERMISSIONS_OWNERS Time spent computing IntersectionObserver observations in the Blink document lifecycle where the observation is for gathering information about permission elements within the viewport. Recorded when an intersection observation attributed to permission element is updated. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} vollick@chromium.org flackr@chromium.org Counts whether a possible sync-scroll update is attempted during an animation update of the outermost main frame of a page (see sync_scroll_attempt_heuristic.h for details). We will only record a 1 when the heuristic detects a possible sync scroll (and we will also subsample further). This means PreFCP and PostFCP will only report ones, and only when subsampled. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. {BlinkUpdateTimeSuffixes} sisidovski@chromium.org chrome-loading@google.com Measures the time from the time when PreloadRequest::Start is called to when PreloadRequest::Start is finished. This histogram only records metrics on machines with high-resolution clocks. cduvall@chromium.org swarm-team@google.com chrome-loading@google.com Measures the time from the creation of a PreloadRequest to when PreloadRequest::Start is called. Logged for each PreloadRequest::Start call. Warning: this histogram was expired from 2023-10-22 to 2023-12-12; data may be missing. pdr@chromium.org paint-dev@chromium.org Time spent updating paint properties and paint invalidation in the Blink document lifecycle. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} nidhijaju@chromium.org chrome-loading@google.com Measures the time it takes to tokenize the HTML document, if necessary, and perform other tasks around parsing completion. Logged every time there is an update to the HTML document. nidhijaju@chromium.org chrome-loading@google.com Measures the time it takes to tokenize an HTML document in Blink. Collected once per update to HTML document. horo@chromium.org net-dev@chromium.org Records whether BackgroundURLLoader can handle the resource request, and the reason why it can't handle the request. Logged for every resource request from the page which is not controlled by a service worker. olivierli@chromium.org leszeks@chromium.org Measures the time from when a code fetch is initiated to when the renderer starts processesing the result. Logged for every code fetch when the response is received. horo@chromium.org net-dev@chromium.org Measures the time from when a resource request completes to when the renderer is notified. Logged for every resource request if the request has never been frozen. Warning: this histogram was expired from 2025-11-16 to 2026-04-02; data may be missing. suzukikeita@chromium.org net-dev@chromium.org Whether the completion time of a resource request is out of range. Emitted for every resource request when TimeTicks is consistent across processes. This histogram has been expired between 2025-12-06 and 2026-07-01, thus does not contain data for this period. olivierli@chromium.org leszeks@chromium.org Measured on resource request completion for requests that queried the code cache. True if one of these operations was delayed waiting on a response from the code cache: 1) Receiving a response 2) Receiving a redirect 3) Completing the request. horo@chromium.org net-dev@chromium.org Measures the time from when a redirect from a resource request happens to when the renderer is notified. Logged for every resource request if the request has never been frozen. Warning: this histogram was expired from 2025-09-14 to 2026-04-02; data may be missing. horo@chromium.org net-dev@chromium.org Measures the time from when a response from a resource request is started to when the renderer is notified. Logged for every resource request if the request has never been frozen. horo@chromium.org net-dev@chromium.org Measures the time from when the renderer starts a resource request to when it is actually started on the network. Logged for every resource request if the request has never been frozen. Warning: this histogram was expired from 2026-01-18 to 2026-04-02; data may be missing. tluk@chromium.org webui-everywhere@google.com Whether the cached metadata managed by a WebUIBundledCachedMetadataHandler was successfully consumed and not rejected by V8 during script compilation. Emitted immediately following WebUI script compilation started with the kConsumeCodeCache compile option. The value is expected to always be true. This histogram and associated code can be removed once this behavior has been confirmed. tluk@chromium.org webui-everywhere@google.com Whether the the fetch request for WebUI bundled cached metadata completed successfully. Emitted immediately after the request to fetch the metadata from the resource bundle has completed. sullivan@chromium.org mmocny@chromium.org speed-metrics-dev@chromium.org chrome-analysis-team@google.com Blink.Responsiveness.PerAnimationFrame.* is a family of metrics representing the sub-parts (aka breakdowns) of Interactions. Reported per Animation Frame which includes at least one Event Timing which is a part of an interaction (as in INP). All subparts are based on Event Timing API measurements (and so, do not include continuous events, like scrolling). This subpart represents the duration from the time when commit stage of rendering completes to when the browser presents the next frame with the changes caused by the last event. The metric is reported for {PerAnimationFrameInteractionType}. sullivan@chromium.org mmocny@chromium.org speed-metrics-dev@chromium.org chrome-analysis-team@google.com Blink.Responsiveness.PerAnimationFrame.* is a family of metrics representing the sub-parts (aka breakdowns) of Interactions. Reported per Animation Frame which includes at least one Event Timing which is a part of an interaction (as in INP). All subparts are based on Event Timing API measurements (and so, do not include continuous events, like scrolling). This subpart represents the duration from the first event creation time (timeStamp) to when it was queued on the main thread. The metric is reported for {PerAnimationFrameInteractionType}. sullivan@chromium.org mmocny@chromium.org speed-metrics-dev@chromium.org chrome-analysis-team@google.com Blink.Responsiveness.PerAnimationFrame.* is a family of metrics representing the sub-parts (aka breakdowns) of Interactions. Reported per Animation Frame which includes at least one Event Timing which is a part of an interaction (as in INP). All subparts are based on Event Timing API measurements (and so, do not include continuous events, like scrolling). This subpart represents the duration from the first event creation time (timeStamp) to when it was processed on the main thread. Traditionally, also referred to as Input Delay. The metric is reported for {PerAnimationFrameInteractionType}. sullivan@chromium.org mmocny@chromium.org speed-metrics-dev@chromium.org chrome-analysis-team@google.com Blink.Responsiveness.PerAnimationFrame.* is a family of metrics representing the sub-parts (aka breakdowns) of Interactions. Reported per Animation Frame which includes at least one Event Timing which is a part of an interaction (as in INP). All subparts are based on Event Timing API measurements (and so, do not include continuous events, like scrolling). This subpart represents the total duration when no frame is presented, which is from the time when the first event is created to when the last event finishes processing. To get the total duration when a frame is presented, use Blink.Responsiveness.PerAnimationFrame.EventCreationToPresentationTime. The metric is reported for {PerAnimationFrameInteractionType}. sullivan@chromium.org mmocny@chromium.org speed-metrics-dev@chromium.org chrome-analysis-team@google.com Blink.Responsiveness.PerAnimationFrame.* is a family of metrics representing the sub-parts (aka breakdowns) of Interactions. Reported per Animation Frame which includes at least one Event Timing which is a part of an interaction (as in INP). All subparts are based on Event Timing API measurements (and so, do not include continuous events, like scrolling). This subpart represents the total duration between when the event is created until it is presented, which is the entire interaction duration. The metric is reported for {PerAnimationFrameInteractionType}. sullivan@chromium.org mmocny@chromium.org speed-metrics-dev@chromium.org chrome-analysis-team@google.com Blink.Responsiveness.PerAnimationFrame.* is a family of metrics representing the sub-parts (aka breakdowns) of Interactions. Reported per Animation Frame which includes at least one Event Timing which is a part of an interaction (as in INP). All subparts are based on Event Timing API measurements (and so, do not include continuous events, like scrolling). This subpart represents the total duration between when the event is queued on the main thread until commit, which is the duration on the main thread. The metric is reported for {PerAnimationFrameInteractionType}. sullivan@chromium.org mmocny@chromium.org speed-metrics-dev@chromium.org chrome-analysis-team@google.com Blink.Responsiveness.PerAnimationFrame.* is a family of metrics representing the sub-parts (aka breakdowns) of Interactions. Reported per Animation Frame which includes at least one Event Timing which is a part of an interaction (as in INP). All subparts are based on Event Timing API measurements (and so, do not include continuous events, like scrolling). This subpart represents the duration from the time when the first event was queued on the main thread to when it was processed. The metric is reported for {PerAnimationFrameInteractionType}. sullivan@chromium.org mmocny@chromium.org speed-metrics-dev@chromium.org chrome-analysis-team@google.com Blink.Responsiveness.PerAnimationFrame.* is a family of metrics representing the sub-parts (aka breakdowns) of Interactions. Reported per Animation Frame which includes at least one Event Timing which is a part of an interaction (as in INP). All subparts are based on Event Timing API measurements (and so, do not include continuous events, like scrolling). This subpart represents the duration from the time when the first event starts processing on the main thread to when the last event finishes processing. The metric is reported for {PerAnimationFrameInteractionType}. sullivan@chromium.org mmocny@chromium.org speed-metrics-dev@chromium.org chrome-analysis-team@google.com Blink.Responsiveness.PerAnimationFrame.* is a family of metrics representing the sub-parts (aka breakdowns) of Interactions. Reported per Animation Frame which includes at least one Event Timing which is a part of an interaction (as in INP). All subparts are based on Event Timing API measurements (and so, do not include continuous events, like scrolling). This subpart represents the duration from the time when the last event finishes processing on the main thread to when the browser presents the next frame with the changes caused by this event. Traditionally, also referred to as Presentation Delay. The metric is reported for {PerAnimationFrameInteractionType}. sullivan@chromium.org mmocny@chromium.org speed-metrics-dev@chromium.org chrome-analysis-team@google.com Blink.Responsiveness.PerAnimationFrame.* is a family of metrics representing the sub-parts (aka breakdowns) of Interactions. Reported per Animation Frame which includes at least one Event Timing which is a part of an interaction (as in INP). All subparts are based on Event Timing API measurements (and so, do not include continuous events, like scrolling). This subpart represents the duration from the time when the last event finishes processing on the main thread to when the rendering starts. The metric is reported for {PerAnimationFrameInteractionType}. sullivan@chromium.org mmocny@chromium.org speed-metrics-dev@chromium.org chrome-analysis-team@google.com Blink.Responsiveness.PerAnimationFrame.* is a family of metrics representing the sub-parts (aka breakdowns) of Interactions. Reported per Animation Frame which includes at least one Event Timing which is a part of an interaction (as in INP). All subparts are based on Event Timing API measurements (and so, do not include continuous events, like scrolling). This subpart represents the duration from the time when the rendering starts to when commit stage of rendering completes. The metric is reported for {PerAnimationFrameInteractionType}. sullivan@chromium.org mmocny@chromium.org speed-metrics-dev@chromium.org chrome-analysis-team@google.com Blink.Responsiveness.PerAnimationFrame.* is a family of metrics representing the sub-parts (aka breakdowns) of Interactions. Reported per Animation Frame which includes at least one Event Timing which is a part of an interaction (as in INP). All subparts are based on Event Timing API measurements (and so, do not include continuous events, like scrolling). This subpart represents the total unaccounted duration (idle time) within the processing time range which is from the time when the first event starts processing on the main thread to when the last event finishes processing. The metric is reported for {PerAnimationFrameInteractionType}. sullivan@chromium.org mmocny@chromium.org speed-metrics-dev@chromium.org The duration of a single user interaction: https://web.dev/better-responsiveness-metric/#group-events-into-interactions This metric is emitted once for every interaction in the web contents {InteractionType} cduvall@chromium.org jam@chromium.org chrome-loading@google.com Blink.ScanAndPreloadTime2 is the newer one of this histogram, which records the time with microseconds. Measures the time it takes to scan and preload subresources for a document in a {FrameType}. Logged every time the {IsInitial} preload scan is performed. Warning: this histogram was expired from 2023-05-29 to 2023-12-04; data may be missing. sisidovski@chromium.org chrome-loading@google.com Measures the time it takes to {Task} for a document in a {FrameType}. Logged every time the {IsInitial} preload scan is performed. This histogram only records metrics on machines with high-resolution clocks. chikamune@chromium.org yyanagisawa@chromium.org kouhei@chromium.org Record when async scripts or dynamically inserted scripts run. Measures the duration between when a script is ready to run, to when the script is about to run. tnak@chromium.org chrome-loading@chromium.org The result of inline script cache fetch attempts. This is recorded for each inline script only if the inline script caching is enabled and cache fetch is not skipped (e.g., for small scripts). tnak@chromium.org chrome-loading@chromium.org Time spent fetching an inline script cache from a worker thread. This is recorded for each inline script only if the inline script caching is enabled. chikamune@chromium.org yyanagisawa@chromium.org kouhei@chromium.org Record which task priority's task was run in `PostTaskWithLowPriorityUntilTimeout()` that uses LowerPriority task queue and NormalPriority task queue. This is recorded when the LowPriorityAsyncScriptExecution behavior is applied to async scripts. chikamune@chromium.org yyanagisawa@chromium.org kouhei@chromium.org Record when the LowPriorityAsyncScriptExecution behavior is applied to async scripts. Measures the duration between when a script is ready to run, to when the script is about to run. chikamune@chromium.org yyanagisawa@chromium.org kouhei@chromium.org Record when the LowPriorityAsyncScriptExecution behavior is applied to async scripts, and {RaceTaskPriority} task was eventually run. Measures the duration between when a script is ready to run, to when the script is about to run. kouhei@chromium.org hiroshige@chromium.org chikamune@chromium.org chrome-loading@google.com Number of script elements for each scheduling type, recorded when each script loading reaches the #prepare-a-script state defined in the spec, with http or https request. Warning: this histogram was expired from 2018-12-31 to 2022-05-25; data may be missing. cduvall@chromium.org jam@google.com The time spent generating the source text for a script resource. Logged when the source text is accessed for the first time. Emitted at most once per script resource. Note that clients without high-resolution clocks will report 0 for very short times. pdr@chromium.org paint-dev@chromium.org Time spent preparing the document to service browser driven behaviors. Examples include accessibility, intersection observer, display locking and spell checking. Includes time spent in style update and layout. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} elkurin@chromium.org chrome-owp-storage@google.com Time (in ms) added to SetCookie operation from the CookieJar used for ablation study purpose. wbjacksonjr@chromium.org blink-network-stack@google.com Records how often a shared dictionary hint is discovered by the renderer, either via an HTML "link" tag or an HTTP Link header. The UMA is recorded before fetching the compression dictionary. yigu@chromium.org web-identity-eng@google.com Records how often the verification backend is unavailable for the WebOTP API and the corresponding reason. This is recorded once per OTP request. yigu@chromium.org web-identity-eng@google.com Records the reason why using the WebOTP API on desktop failed on the sharing path. It records one sample per invocation of the API. yigu@chromium.org goto@chromium.org web-identity-eng@google.com Records how many times the infobar was called and how many times the infobar replaced a keyboard. yigu@chromium.org goto@chromium.org web-identity-eng@google.com Records the result of a call to the WebOTP API. yigu@chromium.org web-identity-eng@google.com Records the status of parsing an incoming SMS when using the WebOTP API. It records one sample per incoming SMS. Note: This metric does not cover cross-devices WebOTP. yigu@chromium.org goto@chromium.org web-identity-eng@google.com Records the duration from when the API is called to when the request is cancelled by the service due to duplicated requests or lack of delegate. yigu@chromium.org goto@chromium.org web-identity-eng@google.com Records the duration from when the keyboard was replaced with an infobar until when the user clicked the cancel button. yigu@chromium.org goto@chromium.org web-identity-eng@google.com Records the time from when a successful SMS was retrieved to when the user presses the Cancel button. yigu@chromium.org goto@chromium.org web-identity-eng@google.com Records the time from when a successful SMS was retrieved to when the user presses the Continue button. yigu@chromium.org goto@chromium.org web-identity-eng@google.com Records the duration from when the API is called to when an SMS has been successfully received. yigu@chromium.org goto@chromium.org web-identity-eng@google.com Records the duration from when the API is called to when the user successfully receives the SMS and presses continue to pass the incoming SMS to the site and proceed with the SMS verification flow. yigu@chromium.org goto@chromium.org web-identity-eng@google.com Records the duration from when the API is called to when the user successfully receives the SMS and presses "Deny" to terminate the SMS verification flow. yigu@chromium.org web-identity-eng@google.com Records whether the web contents that receives the OTP is visible or not for the UserConsent backend. Recorded at most once per API call when Chrome sees the OTP from the expected origin. jbroman@chromium.org src/third_party/blink/renderer/core/speculation_rules/OWNERS The amount of time between when a fetch of speculation rules is started and when it is finished. Recorded each time such a fetch occurs. jbroman@chromium.org src/third_party/blink/renderer/core/speculation_rules/OWNERS Outcome of an attempt to load a speculation rule set. Recorded once per attempt, whether triggered by a response header, by the parser or by script. etiennep@chromium.org src/third_party/blink/renderer/core/css/OWNERS Size of the bucket (CachedMatchedProperties::entries) in which an item landed in the MatchedPropertiesCache (MPC). Recorded when adding an entry to the cache (subsampled to 1%). pdr@chromium.org layout-dev@chromium.org Time spent updating style in the Blink document lifecycle. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} pdr@chromium.org paint-dev@chromium.org chikamune@chromium.org chrome-loading@google.com Time spent updating a SVG image once all data is received. This histogram does not record metrics on machines with low-resolution clocks. Warning: this histogram was expired from 2024-02-25 to 2024-04-17; data may be missing. lizeb@chromium.org paint-dev@chromium.org When propagating intersection information to remote frames, whether the update can be skipped or not, resulting in no IPC being sent. pdr@chromium.org paint-dev@chromium.org Time spent computing geometry information about iframes from the perspective of their embedding document. This time is also included in the Blink.IntersectionObservation.UpdateTime metric. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} wjmaclean@chromium.org feature-control@chromium.org Count of how many about:blank and about:srcdoc pages use various features across all frames in the page. This histogram has the same semantics as "Blink.UseCounter.Features", but applies only to `about:blank` and `about:srcdoc` URLs (which that histogram does not include). Metrics are emitted either at the time a document commits, or via UseCounterImpl::ReportMeasurement(), depending on which feature is involved. For example, SandboxedSrcdocFrameResolvesRelativeURL reports at most once per PageVisit, and only when the feature is activated. Refer to "Blink.UseCounter.Features" for more details. chasej@chromium.org feature-control@chromium.org Records usage of animated CSS properties used on a page, either statically or dynamically, from the time the page is initialised to when it is closed or navigated away from. Each property is counted at most once per page per view via PageLoadMetricsObserver. This histogram counts CSS properties only when they are animated by a CSS Animation. Refer to Blink.UseCounter.CSSProperties for more details. As of M69, Blink.UseCounter.AnimatedCSSProperties is moved from the blink side to the browser side. Not recorded if the page is prerendered and cancelled. chasej@chromium.org feature-control@chromium.org Records usage of CSS properties used on a page, either statically or dynamically, from the time the page is initialised to when it is closed or navigated away from. Each property is counted at most once per page per view via PageLoadMetricsObserver. The first time a CSS property is parsed on a page, the histogram is updated to increment the counter. Each histogram bucket corresponds to a CSS property (eg. width, border-radius). The exception is the 'Total pages measured' bucket - this counts the number of pages that CSS properties were counted on. When a page navigation occurs the page count bucket is incremented and tracking of the set of which properties have been seen is reset. These numbers give the percentage of pages that use a CSS property. For example, if the 'border-radius' histogram bucket has a count of 250, and the page count bucket (i.e. bucket number 1) has a count of 1000 - this means that 1000 pages were recorded, and border-radius was used on 25% of those pages. Note that CSS properties used inside of SVG images are tracked separately in the Blink.UseCounter.SVGImage.CSSProperties histogram. As of M57 this histogram supercedes WebCore.FeatureObserver.CSSProperties. As of M69, Blink.UseCounter.AnimatedCSSProperties is moved from the blink side to the browser side. Not recorded if the page is prerendered and cancelled. iclelland@chromium.org feature-control@chromium.org Counts enforced DocumentPolicyViolationReport generated in a document. iclelland@chromium.org feature-control@chromium.org Counts the use of a specific document policy via the "policy" attribute. Each policy is counted only once per page load. Not recorded if the page is prerendered and cancelled. iclelland@chromium.org feature-control@chromium.org Counts report only DocumentPolicyViolationReport generated in a document. chasej@chromium.org feature-control@chromium.org Count of how many page loads use various features for pages with a chrome-extension:// URL only. The PageVisits bucket is incremented for each page load, and the other buckets incremented at most once per PageVisit via the WebCore::UserCounter class. Warning: This histogram represents pre-renderer metrics and so is flawed under OOPIF. Warning: this histogram was expired from 2021-10-10 to 2022-07-12; data may be missing. iclelland@chromium.org feature-control@chromium.org Counts the use of a specific feature policy via the "Feature-Policy" HTTP response header. chasej@chromium.org feature-control@chromium.org Count of how many page loads use various features across all frames in a page. The PageVisits bucket is incremented for each page load, and the other buckets incremented at most once per PageVisit via blink::UseCounter and UseCounterPageLoadMetricsObserver. Note that this histogram only counts page with HTTP/HTTPS URL scheme. Feature used in pages with other URL schemes might be counted in other histograms: "extension://" : Blink.UseCounter.Extensions.Features "file://" : Blink.UseCounter.File.Features Not recorded if the page is prerendered and cancelled. yhirano@chromium.org mkwst@chromium.org Count of how many `file:` URL page loads use various features across all frames in the page. This histogram has the same semantics as "Blink.UseCounter.Features", but applies only to `file:` URLs (which that histogram does not include). Refer to "Blink.UseCounter.Features" for more details. This counter expired on 2021-01-28 and was re-enabled on 2022-12-19. chasej@chromium.org feature-control@chromium.org Count of how many page loads use various features across in the main frame of a page. The PageVisits bucket is incremented at the beginning of each page load and the other buckets incremented at most once per PageVisit via blink::UseCounter and UseCounterPageLoadMetricsObserver throughout the lifetime of a page load metrics observer. This histogram counts usage of web features in main frame only. Refer to Blink.UseCounter.Features for more details. Not recorded if the page is prerendered and cancelled. iclelland@chromium.org feature-control@chromium.org Counts the use of a specific permissions policy feature via the "allow" attribute on iframes. Each policy is counted only once per page load. Version 2 fixes the mismatched histogram name with data collected. Not recorded if the page is prerendered and cancelled. iclelland@chromium.org feature-control@chromium.org Counts the use of a specific permissions policy feature via both "Feature-Policy" and "Permissions-Policy" HTTP response header. Each policy is counted only once per page load. Version 2 fixes the mismatched histogram name with data collected. Not recorded if the page is prerendered and cancelled. Note that this histogram was expired from 2026-03-09 to 2026-04-16, and data may be missing. jkarlin@chromium.org chrome-ads-team@chromium.org Counts the use of a specific permissions policy that is enabled by the API but is marked as sensitive and advertising script is in the stack when the feature policy is checked. Each policy is counted only once per page load. Not recorded if the page is prerendered and cancelled. iclelland@chromium.org feature-control@chromium.org Counts the actual enforced permissions policy violation. The metrics is triggered when ExecutionContext::IsFeatureEnabled with ReportOptions::ReportOnFailure as param returns false, at most once per page visit. Not recorded if the page is prerendered and cancelled. Note that this histogram was expired from 2026-03-09 to 2026-04-16, and data may be missing. sisidovski@chromium.org chrome-loading@google.com Measures the accumulated time taken by recording the feautre usage in the renderer, at the timing of {Timing}. This metrics is recorded only when the frame is outer-most main frame, the document is not the initial empty document, and the URL is HTTP or HTTPS. This is triggered at the {Timing} phase. This histogram only records metrics on machines with high-resolution clocks. jstenback@google.com feature-control@chromium.org Count of how many page loads use various Web Platform features (as defined by https://github.com/web-platform-dx/web-features/) across all frames in a page. The PageVisits bucket is incremented for each page load, and the other buckets incremented at most once per PageVisit via blink::UseCounter and UseCounterPageLoadMetricsObserver. Note that this histogram only counts page with HTTP/HTTPS URL scheme. Web DX Feature used in pages with other URL schemes might be counted in other histograms. Not recorded if the page is prerendered and cancelled. pdr@chromium.org paint-dev@chromium.org Time spent preparing the document to service user driven behaviors. Examples include inspector, selection, editing, find-in-page and focus. Includes time spent in style update and layout. Note: As of 04/2023, scroll updates are included in user driven behaviors. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} vmpstr@chromium.org rendering-core-dev@chromium.org Elapsed time between sending the animate request and the first animating frame. This captures the delay waiting for a scheduled frame. Recorded for script-based (SPA) transitions only. This histogram only records metrics on machines with high-resolution clocks. vmpstr@chromium.org rendering-core-dev@chromium.org Elapsed time between sending a capture request and invoking the DOM change callback. Recorded for script-based (SPA) transitions only. This histogram only records metrics on machines with high-resolution clocks. vmpstr@chromium.org rendering-core-dev@chromium.org Elapsed time spent in capture tag discovery (adding transition elements from CSS). Recorded for script-based (SPA) transitions only. This histogram only records metrics on machines with high-resolution clocks. vmpstr@chromium.org rendering-core-dev@chromium.org Elapsed time between the DOM callback finishing and the animation being requested. Recorded for script-based (SPA) transitions only. This histogram only records metrics on machines with high-resolution clocks. vmpstr@chromium.org rendering-core-dev@chromium.org Elapsed time spent running the DOM change callback. Recorded for script-based (SPA) transitions only. This histogram only records metrics on machines with high-resolution clocks. vmpstr@chromium.org rendering-core-dev@chromium.org Elapsed time between a start of a view transition and the first non-initial step process, which happens at the earliest rendering opportunity. This is recorded for cross-document transitions only on the outgoing page. This histogram only records metrics on machines with high-resolution clocks. vmpstr@chromium.org rendering-core-dev@chromium.org Recorded when a view transition is skipped or aborted, logging the reason for the skip. szager@chromium.org paint-dev@chromium.org Elapsed time between when a widget first registers a state invalidation that requires a visual refresh; and when the main thread begins a main frame that will send the updated visual state to the compositor. This histogram only records metrics on machines with high-resolution clocks. {BlinkUpdateTimeSuffixes} pdr@chromium.org paint-dev@chromium.org Time spent waiting for the compositor thread to begin processing a layer update inside ProxyMain::BeginMainFrame. Note: PreFCP metrics will be recorded even if FCP is not reached, whereas AggregatedPreFCP metrics are only recorded if FCP is reached. Note: Not recorded on ~5% of Windows machines with low-resolution clocks. Note: The histograms with PreFCP/PostFCP/AggregatedPreFCP suffixes before 105.0.5175.0 were impacted by https://crbug.com/1330675. Be careful when aggregating or comparing these histograms across the version. {BlinkUpdateTimeSuffixes} dalecurtis@chromium.org media-dev-uma@chromium.org Whether a given JavaScript ImageDecoder instance completed operation successfully or not. Logged upon destruction of the instance. dalecurtis@chromium.org media-dev-uma@chromium.org Image codec provided to JavaScript ImageDecoder constructor. Logged once per successful construction upon completion of construction. dalecurtis@chromium.org media-dev-uma@chromium.org The codec type used with {AudioCodecApi}. Recorded for every successful configure() call. dalecurtis@chromium.org media-dev-uma@chromium.org The final media::StatusCode for each {CodecApi} instance. The final status code is either StatusCode::kOk or the first fatal error encountered. Recorded once upon destruction for each codec type. dalecurtis@chromium.org media-dev-uma@chromium.org The codec type used with {VideoCodecApi}. Recorded for every successful configure() call. drott@chromium.org dom-dev@google.com Measures the time to parse an XML Document using the {Backend}-backend in non-Frame situations, i.e. in situations where XSLT is not allowed, such as using the DOMParser API or accessing the XML response document of XMLHttpRequests. Recorded once per created and parsed XML document. The counter starts when the XML parser is created, and stops when parsing finishes. cduvall@chromium.org jam@chromium.org Measures the time it takes to perform a {CookieOperation} operation from the CookieJar. Logged on every operation from the CookieJar. junov@chromium.org fserb@chromium.org aaronhk@chromium.org Time spent executing a script task that draws content to a {Context} context of a {Host}. Note: This metric drops reports on clients with low-resolution clocks, which means these reports will be biased against a portion of the population on Windows. See Windows.HasHighResolutionTimeTicks for the affected sample. junov@chromium.org fserb@chromium.org aaronhk@chromium.org Memory usage on the {Heap} heap in KB, sampled after the execution of script tasks that render content to a {Host}, using a {Context} context.