summaryrefslogtreecommitdiff
path: root/src/common/mempool.c
diff options
context:
space:
mode:
authorNick Mathewson <nickm@torproject.org>2008-02-05 19:40:26 +0000
committerNick Mathewson <nickm@torproject.org>2008-02-05 19:40:26 +0000
commita51deb9a9c7bf33ba353e008724c150bddde144b (patch)
tree5797a9e4129ba027269bb48f4e3d4571f98fe473 /src/common/mempool.c
parent2866c53eec8cd592955f15cec1a317e04aaafc7f (diff)
downloadtor-a51deb9a9c7bf33ba353e008724c150bddde144b.tar.gz
tor-a51deb9a9c7bf33ba353e008724c150bddde144b.zip
r17903@catbus: nickm | 2008-02-05 14:40:03 -0500
Remove some dead code; fix some XXX020s; turn some XXX020s into XXXX_IP6s (i.e., "needs to be fixed when we add ipv6 support"). svn:r13382
Diffstat (limited to 'src/common/mempool.c')
-rw-r--r--src/common/mempool.c5
1 files changed, 0 insertions, 5 deletions
diff --git a/src/common/mempool.c b/src/common/mempool.c
index 36eb0e545f..e4e81b759d 100644
--- a/src/common/mempool.c
+++ b/src/common/mempool.c
@@ -49,11 +49,6 @@
* - We keep a list of full chunks (so we can have a "nuke everything"
* function). Obmalloc's pools leave full chunks to float unanchored.
*
- * [XXXX020 Another way to support 'nuke everything' would be to keep
- * _all_ the chunks in a doubly-linked-list. This would have more
- * space overhead per chunk, but less pointer manipulation overhead
- * than the current approach.]
- *
* LIMITATIONS:
* - Not even slightly threadsafe.
* - Likes to have lots of items per chunks.