[Webkit-unassigned] [Bug 122815] New: Optimize RenderBlock::lowestFloatLogicalBottom by caching previously calculated values

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=122815

           Summary: Optimize RenderBlock::lowestFloatLogicalBottom by
                    caching previously calculated values
           Product: WebKit
           Version: 528+ (Nightly build)
          Platform: Unspecified
        OS/Version: Unspecified
            Status: NEW
          Keywords: BlinkMergeCandidate
          Severity: Normal
          Priority: P2
         Component: Layout and Rendering
        AssignedTo: webkit-unassigned at lists.webkit.org
        ReportedBy: rniwa at webkit.org
                CC: hyatt at apple.com, bdakin at apple.com, koivisto at iki.fi,
                    simon.fraser at apple.com, dino at apple.com,
                    barraclough at apple.com, akling at apple.com,
                    psolanki at apple.com


Consider merging https://chromium.googlesource.com/chromium/blink/+/b0ba9a1ad19d5519cb03c3d42b38c3b8976e855e

This CL moves RenderBlock::lowestFloatLogicalBottom into FloatingObjects, and introduces a small cache to track the previously calculated lowestFloatLogicalBottom for each FloatingObject::type.

On my z620, this is ~25% faster on blink_perf's float_50_100 benchmark, and ~30% faster on android-gn. On oilevent from tough_layout_cases, this is about ~10% improvement.

Linux: 

Without patch: 
Avg floats_50_100: 255.006988ms 
Sd floats_50_100: 6.590191ms 

With patch: 
Avg floats_50_100: 189.995312ms 
Sd floats_50_100: 4.420704ms 


Android: 

Without patch: 
Avg floats_50_100: 6374.133600ms 
Sd floats_50_100: 852.598792ms 

With patch: 
Avg floats_50_100: 4205.389400ms 
Sd floats_50_100: 659.635617ms

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.