[css-grid] Expand https://lists.w3.org/Archives/Public/www-style/2016Jan/0128.html into an inlined issue.

Author: fantasai

css-grid/Overview.bs

@@ -3437,6 +3437,35 @@ Grid Sizing Algorithm</h3>
 			if any row with a <a>definite</a> <a>max track sizing function</a>
 			had that size and all other rows were infinite.
 
+		<div class="issue">
+			Alternatively, we could
+			assume the <a>available space</a> that it would have as the maximum of:
+
+			<ul>
+				<li>
+					the sum of all <a>definite</a> track sizes that it spans
+					(using the maximum of a track’s min and max sizing functions,
+					if both are definite,
+					the argument to ''fit-content()'' if that is definite).
+				<li>
+					the item’s ''min-content'' size,
+					if any track that it spans
+					has a ''min-content'' or ''fit-content()'' sizing function.
+				<li>
+					the item's <a>automatic minimum size</a>,
+					if any track that it spans
+					has an ''auto'' min sizing function.
+				<li>
+					infinity, if any track that it spans
+					has a ''max-content'' min sizing function
+					or a ''max-content'', ''auto'', or <<flex>> max sizing function.
+			</ul>
+
+			This is likely to reduce the amount of re-layout passes that are necessary,
+			but will it produce a different result in any cases?
+			Should we adopt it into the spec?
+		</div>
+
 		<li>
 			Next, the <a>track sizing algorithm</a> resolves the sizes of the <a>grid rows</a>,
 			using the <a>grid column</a> sizes calculated in the previous step.