deploy: db56fdf

servo · Nov 11, 2024 · f5b736c · f5b736c
1 parent 410644a
commit f5b736c
Show file tree

Hide file tree

Showing 3 changed files with 5 additions and 3 deletions.
diff --git a/blog-october-2024/.stamp b/blog-october-2024/.stamp
@@ -1 +1 @@
-Mon Nov 11 04:40:17 UTC 2024
+Mon Nov 11 04:54:41 UTC 2024
diff --git a/blog-october-2024/blog/2024/11/08/this-month-in-servo/index.html b/blog-october-2024/blog/2024/11/08/this-month-in-servo/index.html
@@ -221,7 +221,8 @@ <h2 id="performance-improvements" tabindex="-1">Performance improvements <a clas
       </a></h2>
 <p>Our font system is faster now, with <strong>reduced latency</strong> when loading system fonts (<a href="https://github.com/mrobinson">@mrobinson</a>, <a href="https://github.com/servo/servo/pull/33638">#33638</a>), layout <strong>no longer blocking on sending font data</strong> to WebRender (<a href="https://github.com/mrobinson">@mrobinson</a>, <a href="https://github.com/servo/servo/pull/33600">#33600</a>), and <strong>memory mapped system fonts</strong> on macOS and FreeType platforms like Linux (<a href="https://github.com/mrobinson">@mrobinson</a>, <a href="https://github.com/mukilan">@mukilan</a>, <a href="https://github.com/servo/servo/pull/33747">#33747</a>).</p>
 <p>Servo now has a <strong>dedicated fetch thread</strong> (<a href="https://github.com/mrobinson">@mrobinson</a>, <a href="https://github.com/servo/servo/pull/33863">#33863</a>).
-This greatly reduces the number of IPC channels we create for individual requests, and should fix crashes related to file descriptor exhaustion on some platforms.</p>
+This greatly reduces the number of IPC channels we create for individual requests, and should fix crashes related to file descriptor exhaustion on some platforms.
+<strong>Brotli-compressed responses</strong> are also handled more efficiently, such that we run the parser with up to 8 KiB of decompressed data at a time, rather than only 10 bytes of compressed data at a time (<a href="https://github.com/crbrz">@crbrz</a>, <a href="https://github.com/servo/servo/pull/33611">#33611</a>).</p>
 <p><strong>Flexbox layout now uses caching</strong> to avoid doing unnecessary work (<a href="https://github.com/mrobinson">@mrobinson</a>, <a href="https://github.com/Loirooriol">@Loirooriol</a>, <a href="https://github.com/servo/servo/pull/33964">#33964</a>, <a href="https://github.com/servo/servo/pull/33967">#33967</a>), and now has experimental <strong><a href="https://docs.rs/tracing/0.1.40/tracing/">tracing</a>-based profiling support</strong> (<a href="https://github.com/mrobinson">@mrobinson</a>, <a href="https://github.com/servo/servo/pull/33647">#33647</a>), which in turn no longer spams RUST_LOG=info when not enabled (<a href="https://github.com/delan">@delan</a>, <a href="https://github.com/servo/servo/pull/33845">#33845</a>).
 We’ve also landed optimisations in table layout (<a href="https://github.com/Loirooriol">@Loirooriol</a>, <a href="https://github.com/servo/servo/pull/33575">#33575</a>) and in our layout engine as a whole (<a href="https://github.com/Loirooriol">@Loirooriol</a>, <a href="https://github.com/servo/servo/pull/33806">#33806</a>).</p>
 <p>Work continues on making our massive <code>script</code> crate build faster, with <strong>improved incremental builds</strong> (<a href="https://github.com/sagudev">@sagudev</a>, <a href="https://github.com/mrobinson">@mrobinson</a>, <a href="https://github.com/servo/servo/pull/33502">#33502</a>) and further patches towards <strong>splitting <code>script</code> into smaller crates</strong> (<a href="https://github.com/sagudev">@sagudev</a>, <a href="https://github.com/jdm">@jdm</a>, <a href="https://github.com/servo/servo/pull/33627">#33627</a>, <a href="https://github.com/servo/servo/pull/33665">#33665</a>).</p>

diff --git a/blog-october-2024/blog/feed.xml b/blog-october-2024/blog/feed.xml
@@ -35,7 +35,8 @@ We’ve also reworked WebGPU canvas presentation to ensure that we never use old
       &lt;/a&gt;&lt;/h2&gt;
 &lt;p&gt;Our font system is faster now, with &lt;strong&gt;reduced latency&lt;/strong&gt; when loading system fonts (&lt;a href=&quot;https://github.com/mrobinson&quot;&gt;@mrobinson&lt;/a&gt;, &lt;a href=&quot;https://github.com/servo/servo/pull/33638&quot;&gt;#33638&lt;/a&gt;), layout &lt;strong&gt;no longer blocking on sending font data&lt;/strong&gt; to WebRender (&lt;a href=&quot;https://github.com/mrobinson&quot;&gt;@mrobinson&lt;/a&gt;, &lt;a href=&quot;https://github.com/servo/servo/pull/33600&quot;&gt;#33600&lt;/a&gt;), and &lt;strong&gt;memory mapped system fonts&lt;/strong&gt; on macOS and FreeType platforms like Linux (&lt;a href=&quot;https://github.com/mrobinson&quot;&gt;@mrobinson&lt;/a&gt;, &lt;a href=&quot;https://github.com/mukilan&quot;&gt;@mukilan&lt;/a&gt;, &lt;a href=&quot;https://github.com/servo/servo/pull/33747&quot;&gt;#33747&lt;/a&gt;).&lt;/p&gt;
 &lt;p&gt;Servo now has a &lt;strong&gt;dedicated fetch thread&lt;/strong&gt; (&lt;a href=&quot;https://github.com/mrobinson&quot;&gt;@mrobinson&lt;/a&gt;, &lt;a href=&quot;https://github.com/servo/servo/pull/33863&quot;&gt;#33863&lt;/a&gt;).
-This greatly reduces the number of IPC channels we create for individual requests, and should fix crashes related to file descriptor exhaustion on some platforms.&lt;/p&gt;
+This greatly reduces the number of IPC channels we create for individual requests, and should fix crashes related to file descriptor exhaustion on some platforms.
+&lt;strong&gt;Brotli-compressed responses&lt;/strong&gt; are also handled more efficiently, such that we run the parser with up to 8 KiB of decompressed data at a time, rather than only 10 bytes of compressed data at a time (&lt;a href=&quot;https://github.com/crbrz&quot;&gt;@crbrz&lt;/a&gt;, &lt;a href=&quot;https://github.com/servo/servo/pull/33611&quot;&gt;#33611&lt;/a&gt;).&lt;/p&gt;
 &lt;p&gt;&lt;strong&gt;Flexbox layout now uses caching&lt;/strong&gt; to avoid doing unnecessary work (&lt;a href=&quot;https://github.com/mrobinson&quot;&gt;@mrobinson&lt;/a&gt;, &lt;a href=&quot;https://github.com/Loirooriol&quot;&gt;@Loirooriol&lt;/a&gt;, &lt;a href=&quot;https://github.com/servo/servo/pull/33964&quot;&gt;#33964&lt;/a&gt;, &lt;a href=&quot;https://github.com/servo/servo/pull/33967&quot;&gt;#33967&lt;/a&gt;), and now has experimental &lt;strong&gt;&lt;a href=&quot;https://docs.rs/tracing/0.1.40/tracing/&quot;&gt;tracing&lt;/a&gt;-based profiling support&lt;/strong&gt; (&lt;a href=&quot;https://github.com/mrobinson&quot;&gt;@mrobinson&lt;/a&gt;, &lt;a href=&quot;https://github.com/servo/servo/pull/33647&quot;&gt;#33647&lt;/a&gt;), which in turn no longer spams RUST_LOG=info when not enabled (&lt;a href=&quot;https://github.com/delan&quot;&gt;@delan&lt;/a&gt;, &lt;a href=&quot;https://github.com/servo/servo/pull/33845&quot;&gt;#33845&lt;/a&gt;).
 We’ve also landed optimisations in table layout (&lt;a href=&quot;https://github.com/Loirooriol&quot;&gt;@Loirooriol&lt;/a&gt;, &lt;a href=&quot;https://github.com/servo/servo/pull/33575&quot;&gt;#33575&lt;/a&gt;) and in our layout engine as a whole (&lt;a href=&quot;https://github.com/Loirooriol&quot;&gt;@Loirooriol&lt;/a&gt;, &lt;a href=&quot;https://github.com/servo/servo/pull/33806&quot;&gt;#33806&lt;/a&gt;).&lt;/p&gt;
 &lt;p&gt;Work continues on making our massive &lt;code&gt;script&lt;/code&gt; crate build faster, with &lt;strong&gt;improved incremental builds&lt;/strong&gt; (&lt;a href=&quot;https://github.com/sagudev&quot;&gt;@sagudev&lt;/a&gt;, &lt;a href=&quot;https://github.com/mrobinson&quot;&gt;@mrobinson&lt;/a&gt;, &lt;a href=&quot;https://github.com/servo/servo/pull/33502&quot;&gt;#33502&lt;/a&gt;) and further patches towards &lt;strong&gt;splitting &lt;code&gt;script&lt;/code&gt; into smaller crates&lt;/strong&gt; (&lt;a href=&quot;https://github.com/sagudev&quot;&gt;@sagudev&lt;/a&gt;, &lt;a href=&quot;https://github.com/jdm&quot;&gt;@jdm&lt;/a&gt;, &lt;a href=&quot;https://github.com/servo/servo/pull/33627&quot;&gt;#33627&lt;/a&gt;, &lt;a href=&quot;https://github.com/servo/servo/pull/33665&quot;&gt;#33665&lt;/a&gt;).&lt;/p&gt;