[soft navs] Expected changes to page_load_metrics.mojom [chromium/src : main]

[Michal Mocny (Gerrit)]

Attention needed from Johannes Henkel

Michal Mocny added 3 comments

Commit Message

Line 15, Patchset 1 (Latest):3) Add soft_navigation_count to LargestContentfulPaintTiming,

Michal Mocny . unresolved

why add soft_navigation_count to LCP? That feels out of place.

If you need an ID, you can call it navigationId or interactionId, even if you use "count" as the id in early prototype?

File components/page_load_metrics/common/page_load_metrics.mojom

Line 550, Patchset 1 (Latest): mojo_base.mojom.TimeTicks layout_shift_time;

Johannes Henkel . unresolved

This will be compared with SoftNavigationMetrics.commit_time, when we interpret the render_data's new_layout_shifts, which arrive as a parameter of the UpdateTiming call.

Michal Mocny

The only risk with this is that the timestamp of layout shifts is sort of unique and different relative to paint timings.

If you had:

• URL change, for a soft-nav that already have a paint in the past (commit time)
• DOM re-render
• Layout and pre-paint (layout shifts)
• Paint (LCP etc)

... then Layout Shifts get attributed to this URL.

If you had:

• URL change, for a soft-nav that did not already have a paint in the past\
• DOM re-render
• Layout and pre-paint (layout shifts)
• Paint (LCP etc, commit time)

... then Layout Shifts get attributed to the previous URL.

---

We cannot easily move the timestamps around for spec / interop reasons, though we could consider making Shifts a paint timing and giving them a paintTime instead of their currently "random" middle-of-rendering time.

But we could consider forcing their navigationId to match "the current animation frame" navigationId, since the user sees this shift together with the other stuff, intra-frame ordering doesn't matter.

(Alternatively, we could expose a paintTime to LayoutShift just for internal metrics, if thats the direction)

Line 575, Patchset 1 (Latest): mojo_base.mojom.TimeTicks start_time;

Johannes Henkel . unresolved

This will be compared with SoftNavigationMetrics.commit_time, when we interpret the event_timings parameter of the UpdateTiming call.

Michal Mocny

Similar issue to Layout Shift, the start_time is the wrong unit of time for comparison (its async from the main thread).

But for INP we already plan to add paintTime very soon.

Open in Gerrit

Related details

Attention is currently required from:

• Johannes Henkel

Submit Requirements:

• Code-Coverage
• Code-Owners
• Code-Review
• No-Unresolved-Comments
• Review-Enforcement

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: comment

Gerrit-Project: chromium/src

Gerrit-Branch: main

Gerrit-Change-Id: I533d1216b9b14dbded9a06e0ce70eae540baf56a

Gerrit-Change-Number: 7595809

Gerrit-PatchSet: 1

Gerrit-Owner: Johannes Henkel <joha...@chromium.org>

Gerrit-CC: Michal Mocny <mmo...@chromium.org>

Gerrit-CC: chromium...@chromium.org

Gerrit-Attention: Johannes Henkel <joha...@chromium.org>

Gerrit-Comment-Date: Mon, 23 Feb 2026 13:06:30 +0000

Gerrit-HasComments: Yes

Gerrit-Has-Labels: No

Comment-In-Reply-To: Johannes Henkel <joha...@chromium.org>

[Johannes Henkel (Gerrit)]

Johannes Henkel abandoned this change.

View Change

Abandoned this is done

Johannes Henkel abandoned this change

Related details

Attention set is empty

Submit Requirements:

• Code-Coverage

• Code-Review
• No-Unresolved-Comments
• Review-Enforcement

Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings.

Gerrit

Gerrit-MessageType: abandon