@@ -1419,30 +1419,6 @@ Absolute Positioning</h2>
14191419 Maybe it could defined as the max (or min?) current [=running position=]
14201420 of the [=grid-axis=] tracks at that point? Or the end of the item before it?
14211421
1422- <h2 id="performance-notes">
1423- Performance Notes</h2>
1424-
1425- In general, masonry layout should have significantly better performance
1426- than the equivalent regular (2-axis) grid layout,
1427- particularly when the [=stacking axis=] is the [=block axis=]
1428- since the intrinsic sizing of grid rows is typically quite expensive.
1429- Any intrinsic track sizing in the [=grid axis=] should be cheaper too,
1430- because, typically, only a subset of items contribute to the intrinsic sizing in a masonry layout,
1431- contrary to a 2-axis grid where all items spanning an intrinsically-sized track contribute.
1432- Stretched items do a second layout with the new size (when it actually changed)
1433- so this can be costly if there are a huge amount of stretched items
1434- that each contains a lot of content.
1435- Especially nested stretched masonry layouts should be avoided
1436- unless they are small/trivial.
1437-
1438- Advisement: This can be ameliorated by the author
1439- by opting out from the stretching on most items though,
1440- e.g. specifying ''justify/align-items:start''
1441- and then opting in for just a few items with ''justify/align-self:stretch''
1442- to let those items fill the [=stacking axis=] .
1443- (This performance analysis is from a Gecko perspective,
1444- but I suspect there's some truth to it for other layout engines as well.)
1445-
14461422<h2 id="graceful-degradation">
14471423Graceful Degradation</h2>
14481424
0 commit comments