We are *definitely* counting cycles as closely as DaveM, so
ensure hwcache alignment for our recv ring control structs.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
rds_ib_incoming_slab = kmem_cache_create("rds_ib_incoming",
sizeof(struct rds_ib_incoming),
- 0, 0, NULL);
+ 0, SLAB_HWCACHE_ALIGN, NULL);
if (!rds_ib_incoming_slab)
goto out;
rds_ib_frag_slab = kmem_cache_create("rds_ib_frag",
sizeof(struct rds_page_frag),
- 0, 0, NULL);
+ 0, SLAB_HWCACHE_ALIGN, NULL);
if (!rds_ib_frag_slab)
kmem_cache_destroy(rds_ib_incoming_slab);
else