iscsi-target: Bump default TCP listen backlog to 256

This patch bumps the default TCP listen backlog within iscsit_setup_np() from 5 to 256, in order to reduce the overall latency caused by a small backlog with 100's of simultaneous login attempts directed to the same single network portal. Also add a ISCSIT_TCP_BACKLOG macro in iscsi_target_core.h. Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
mariux64 · Sep 11, 2013 · 837f645 · 837f645
1 parent f730f91
commit 837f645
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 1 deletion.
diff --git a/drivers/target/iscsi/iscsi_target_core.h b/drivers/target/iscsi/iscsi_target_core.h
@@ -19,6 +19,7 @@
 #define WHITE_SPACE			" \t\v\f\n\r"
 #define ISCSIT_MIN_TAGS			16
 #define ISCSIT_EXTRA_TAGS		8
+#define ISCSIT_TCP_BACKLOG		256
 
 /* struct iscsi_node_attrib sanity values */
 #define NA_DATAOUT_TIMEOUT		3

diff --git a/drivers/target/iscsi/iscsi_target_login.c b/drivers/target/iscsi/iscsi_target_login.c
@@ -873,7 +873,7 @@ int iscsit_setup_np(
 	struct __kernel_sockaddr_storage *sockaddr)
 {
 	struct socket *sock = NULL;
-	int backlog = 5, ret, opt = 0, len;
+	int backlog = ISCSIT_TCP_BACKLOG, ret, opt = 0, len;
 
 	switch (np->np_network_transport) {
 	case ISCSI_TCP: