summaryrefslogtreecommitdiffstats
path: root/include/net/dst.h
diff options
context:
space:
mode:
authorEric Dumazet <eric.dumazet@gmail.com>2010-10-03 22:17:54 -0700
committerDavid S. Miller <davem@davemloft.net>2010-10-03 22:17:54 -0700
commitc7d4426a98a5f6654cd0b4b33d9dab2e77192c18 (patch)
tree0db2524e6f3f742861765dd6aa696a9271767056 /include/net/dst.h
parent9a7241c21b06c3a3f8ebcf3e347bd68556369da7 (diff)
net: introduce DST_NOCACHE flag
While doing stress tests with IP route cache disabled, and multi queue devices, I noticed a very high contention on one rwlock used in neighbour code. When many cpus are trying to send frames (possibly using a high performance multiqueue device) to the same neighbour, they fight for the neigh->lock rwlock in order to call neigh_hh_init(), and fight on hh->hh_refcnt (a pair of atomic_inc/atomic_dec_and_test()) But we dont need to call neigh_hh_init() for dst that are used only once. It costs four atomic operations at least, on two contended cache lines, plus the high contention on neigh->lock rwlock. Introduce a new dst flag, DST_NOCACHE, that is set when dst was not inserted in route cache. With the stress test bench, sending 160000000 frames on one neighbour, results are : Before patch: real 2m28.406s user 0m11.781s sys 36m17.964s After patch: real 1m26.532s user 0m12.185s sys 20m3.903s Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'include/net/dst.h')
-rw-r--r--include/net/dst.h9
1 files changed, 5 insertions, 4 deletions
diff --git a/include/net/dst.h b/include/net/dst.h
index aa53fbc34b2..a217c838ec0 100644
--- a/include/net/dst.h
+++ b/include/net/dst.h
@@ -43,10 +43,11 @@ struct dst_entry {
short error;
short obsolete;
int flags;
-#define DST_HOST 1
-#define DST_NOXFRM 2
-#define DST_NOPOLICY 4
-#define DST_NOHASH 8
+#define DST_HOST 0x0001
+#define DST_NOXFRM 0x0002
+#define DST_NOPOLICY 0x0004
+#define DST_NOHASH 0x0008
+#define DST_NOCACHE 0x0010
unsigned long expires;
unsigned short header_len; /* more space at head required */