summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorDavid Reid <dreid@apache.org>2001-05-30 11:15:36 +0000
committerDavid Reid <dreid@apache.org>2001-05-30 11:15:36 +0000
commit7bde13763285da45dd928e4560c622d45896e411 (patch)
treed1a2a9d4e9d745a16e22083392063597c64f3463
parent0a12885165bf5c19101af49b0d073d0f1c2c1327 (diff)
downloadapr-7bde13763285da45dd928e4560c622d45896e411.tar.gz
Add some more stuff to the TODO
git-svn-id: https://svn.apache.org/repos/asf/apr/apr/trunk@61680 13f79535-47bb-0310-9956-ffa450edef68
-rw-r--r--memory/unix/TODO15
1 files changed, 15 insertions, 0 deletions
diff --git a/memory/unix/TODO b/memory/unix/TODO
index e107fa885..4b3c8cc8f 100644
--- a/memory/unix/TODO
+++ b/memory/unix/TODO
@@ -4,10 +4,20 @@ STATUS file for APR as none of these can be considered
showstoppers for the time being.
david - 28 May 2001
+- add a shared memory module.
+
+- locking needs to be addressed. The scope of the locks needs
+ to be defined and it's likely we'll need some way of
+ varying the scope when locking.
+
- we need to add dynamic loading ability for memory
systems. As to how it should be done this needs to be
looked at. Some known issues include
o differing arguments for create functions
+ o it would be very cool to use the APR dso code, but
+ as this uses pools and we're not using pools anywhere in the
+ memory code (for obvious reasons) this is a bit of a stopper.
+ Sander says he found the same thing with using the locking code
Just to clarify why this will be very cool if we can get
it working, we give the user the ability to use 3rd party modules
@@ -27,3 +37,8 @@ david - 28 May 2001
some special stuff for pools under Linux on 2.0, so we just
need some ideas
+Possible Extras
+
+- Is there any ebenfit in adding a version number to the memory
+ systems? This probably isn't an issue until we have dynamic
+ loading when binary memory systems may be distributed.