summaryrefslogtreecommitdiff
path: root/Todo.mdwn
diff options
context:
space:
mode:
authorKillian De Volder <killian.de.volderc@megasoft.be>2017-04-01 12:17:21 +0200
committerKillian De Volder <killian.de.volderc@megasoft.be>2017-04-01 12:17:21 +0200
commit7c21019184846d04d92b35bb6b0739bc009dcadf (patch)
tree466d1d69d3dc10be359a65c1289da8933e5fa03a /Todo.mdwn
parent306cbe00b913bb7ce5ab85bf7b970251751fe254 (diff)
parenta0a1f415b0d7b3f059cedb0fcde0b4db6f8528d1 (diff)
Merge branch 'master' of ssh://evilpiepirate.org:22/home/bcachefs/bcachefs
Diffstat (limited to 'Todo.mdwn')
-rw-r--r--Todo.mdwn14
1 files changed, 14 insertions, 0 deletions
diff --git a/Todo.mdwn b/Todo.mdwn
index 036fe3f..7c132c8 100644
--- a/Todo.mdwn
+++ b/Todo.mdwn
@@ -15,6 +15,18 @@
is implementing a new kind of reservation of journal space for the new, long
running transactions.
+ * Allocation information (currently just bucket generation numbers & priority
+ numbers, for LRU caching) needs to be moved into a btree, and we need to
+ start persisting actual allocation information so we don't have to walk all
+ extents at mount time.
+
+ Just moving the existing prios/gens to a btree will be a significant
+ improvement - besides getting us incrementally closer to persisting full
+ allocation information, the existing code is a rather hacky mechanism dating
+ from the early days of bcache and has recently been the source of an annoying
+ bug due to the design being a bit fragile, and it'll be a performance
+ improvement since it'll get rid of the last source of forced journal flushes.
+
### Other
* When we're using compression, we end up wasting a fair amount of space on
@@ -56,6 +68,8 @@
## Developers
+ * End user documentation needs a lot of work - complete man pages, etc.
+
* bcachefs-tools needs some fleshing out in the --help department
* Write a tool to benchmark tail-latency.