summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorecashin@coraid.com <ecashin@coraid.com>2005-04-18 22:00:21 -0700
committerGreg KH <greg@press.kroah.org>2005-04-18 22:00:21 -0700
commite3b7df65e089f143b9228472b80fb96c495fb634 (patch)
treea2366a9a09aaa9063b3b4e41eab747fc0c8b29a5
parentd098840e37468fdd0143d8bcfe86bc53627bf96e (diff)
[PATCH] aoe 9/12: add note about the need for deadlock-free sk_buff allocation
add note about the need for deadlock-free sk_buff allocation Signed-off-by: Ed L. Cashin <ecashin@coraid.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
-rw-r--r--Documentation/aoe/todo.txt14
1 files changed, 14 insertions, 0 deletions
diff --git a/Documentation/aoe/todo.txt b/Documentation/aoe/todo.txt
new file mode 100644
index 000000000000..7fee1e1165bc
--- /dev/null
+++ b/Documentation/aoe/todo.txt
@@ -0,0 +1,14 @@
+There is a potential for deadlock when allocating a struct sk_buff for
+data that needs to be written out to aoe storage. If the data is
+being written from a dirty page in order to free that page, and if
+there are no other pages available, then deadlock may occur when a
+free page is needed for the sk_buff allocation. This situation has
+not been observed, but it would be nice to eliminate any potential for
+deadlock under memory pressure.
+
+Because ATA over Ethernet is not fragmented by the kernel's IP code,
+the destructore member of the struct sk_buff is available to the aoe
+driver. By using a mempool for allocating all but the first few
+sk_buffs, and by registering a destructor, we should be able to
+efficiently allocate sk_buffs without introducing any potential for
+deadlock.