diff --git a/doc/todo/--debug-level___40__and__47__or_-c_annex.debug-level__41___+_more_debug_information_from_git-annex.mdwn b/doc/todo/--debug-level___40__and__47__or_-c_annex.debug-level__41___+_more_debug_information_from_git-annex.mdwn
index 796b4272b6..c25098f60e 100644
--- a/doc/todo/--debug-level___40__and__47__or_-c_annex.debug-level__41___+_more_debug_information_from_git-annex.mdwn
+++ b/doc/todo/--debug-level___40__and__47__or_-c_annex.debug-level__41___+_more_debug_information_from_git-annex.mdwn
@@ -3,3 +3,5 @@ I hope that I would never need to use that, but it might end up being "days-savi
 
 [[!meta author=yoh]]
 [[!tag projects/datalad]]
+
+> [[done]] --[[Joey]] 
diff --git a/doc/todo/--debug-level___40__and__47__or_-c_annex.debug-level__41___+_more_debug_information_from_git-annex/comment_7_5ae9913b8b9a9fe443c211976adce002._comment b/doc/todo/--debug-level___40__and__47__or_-c_annex.debug-level__41___+_more_debug_information_from_git-annex/comment_7_5ae9913b8b9a9fe443c211976adce002._comment
new file mode 100644
index 0000000000..c7476f97b5
--- /dev/null
+++ b/doc/todo/--debug-level___40__and__47__or_-c_annex.debug-level__41___+_more_debug_information_from_git-annex/comment_7_5ae9913b8b9a9fe443c211976adce002._comment
@@ -0,0 +1,29 @@
+[[!comment format=mdwn
+ username="joey"
+ subject="""comment 7"""
+ date="2021-04-06T19:42:06Z"
+ content="""
+fastDebug is implemented.. Benchmarking, I get:
+
+- without debugging in Annex.Branch: 4.75s
+- debug in Annex.Branch: 4.88s, 0.027% slower
+- fastDebug in Annex.Branch: 4.83s, 0.016% slower
+
+So I was right that fastDebug would be faster, but I guess there's still
+some small overhead to check the data structure to see if debugging is
+enabled, and it may not be possible to avoid that. (Maybe unboxing would
+help?)
+
+I'm not bothered by such a small slowdown;
+git-annex has been sped up a lot in recent times and it's ok to pay that.
+(Although I actually finessed it by not debugging cache hits.)
+But cumulative slowdowns from adding debugging lots of places that end up
+all being used in the same tight loop is still something to keep an eye on.
+
+With that said, it should be easy to add
+additional debug info to most parts of git-annex now, so get in touch if
+you have something specific you want debug info for. I'm closing this todo
+here.
+
+(See also, [[todo/move_readonly_values_to_AnnexRead]])
+"""]]
diff --git a/doc/todo/move_readonly_values_to_AnnexRead.mdwn b/doc/todo/move_readonly_values_to_AnnexRead.mdwn
new file mode 100644
index 0000000000..691e660cd2
--- /dev/null
+++ b/doc/todo/move_readonly_values_to_AnnexRead.mdwn
@@ -0,0 +1,5 @@
+AnnexRead has recently been split out of AnnexState,
+and reading values from it is faster (no MVar). So,
+anything that never needs to be modified while git-annex is running can be
+moved to AnnexRead for a performance win and also to make clean how it's
+used. --[[Joey]]