Hi Alan,
On Mon, 8 May 2023 17:27:48 +0000, Alan Huang wrote:
> The code of resizable hash table has been updated so that it doesn't
> need to compute the hash twice since 2019. Here are some related
> commits:
> 2ea492b,
> 2a7f20d,
> ...
Can you say (as recommended in patches to Linux kernel) ?:
2ea492bbab9d ("datatruct/hash: Don't recompute hashes in hashtab_add()")
2a7f20d234e7 ("datastruct/hash: Don't recompute hashes in hashtab_del()")
(hashes of 7 chars should be sufficient for perfbook, but our eyes
expect 12 chars ...)
> So, the patch removes the outdated content.
Indeed.
However, -nq builds need additional tweaks appended below (on top).
Can you do a respin with my Rb tag.
Reviewed-by: Akira Yokosawa <akiyks@xxxxxxxxx>
See qqz.sty's header comment for the format of consecutive QQz's.
Thanks, Akira
diff --git a/datastruct/datastruct.tex b/datastruct/datastruct.tex
index 9aa04bb5db2e..0242ccc7ba4b 100644
--- a/datastruct/datastruct.tex
+++ b/datastruct/datastruct.tex
@@ -2017,10 +2017,10 @@ which could allow the compiler to inline the resulting fixed functions,
eliminating not only the overhead of the call instruction, but the
argument marshalling as well.
-\QuickQuizE{
+\QuickQuiz{
How much do these specializations really save?
Are they really worth it?
-}\QuickQuizAnswerE{
+}\QuickQuizAnswer{
The answer to the first question is left as an exercise to
the reader.
Try specializing the resizable hash table and see how much
@@ -2029,7 +2029,7 @@ argument marshalling as well.
instead be answered with respect to a specific use case.
Some use cases are extremely sensitive to performance and
scalability, while others are less so.
-}\QuickQuizEndE
+}\QuickQuizEnd
All that aside, one of the great benefits of modern hardware compared
to that available when I first started learning to program back in
