Open
Bug 2003217
Opened 8 months ago
Updated 1 month ago
Testcase inserting N shadow-DOM slots twice-in-a-row is superlinear in the usual areas (relocating of elements in nsTarray)
Categories
(Core :: DOM: Core & HTML, defect, P3)
Core
DOM: Core & HTML
Tracking
()
NEW
People
(Reporter: mayankleoboy1, Unassigned)
References
(Blocks 1 open bug)
Details
Attachments
(1 file)
|
3.25 KB,
text/html
|
Details |
STR:
- Open testcase
- Select "same slot name for all"
- Click run.
- Wait for it to complete
- Click run again
Step#5 is superlinear in firefox: https://share.firefox.dev/48EZK2Z
Chrome is instant
|
Reporter | |
Comment 1•8 months ago
|
||
Even with "same slot name for all" DISABLED, it is still quadratic, but to a lesser extent.
Same slot for all: Disabled
N=50000
https://share.firefox.dev/3Xr3gI4 (680ms for run1, 38s for run2)
Summary: Testcaase inserting N shadow-DOM slots twice-in-a-row is superlinear in the usual areas (memcpy in nsTarray) → Testcase inserting N shadow-DOM slots twice-in-a-row is superlinear in the usual areas (memcpy in nsTarray)
|
||
Updated•8 months ago
|
Severity: -- → S4
Priority: -- → P3
|
|
Reporter | |
Updated•4 months ago
|
Summary: Testcase inserting N shadow-DOM slots twice-in-a-row is superlinear in the usual areas (memcpy in nsTarray) → Testcase inserting N shadow-DOM slots twice-in-a-row is superlinear in the usual areas (relocating of elements in nsTarray)
|
|
Reporter | |
Comment 2•2 months ago
•
|
||
This is now 7x faster than before: https://share.firefox.dev/4utdoPr
Though still quadratic.
Here is a profile where i repeatedly click on the insert button: https://share.firefox.dev/49GJzSL
The memory increases quite a bit during insertion but immediately drops after insertion.
Improvement due to bug 2006613
Depends on: 2006613
You need to log in
before you can comment on or make changes to this bug.
Description
•