In the current implementation of ILA, LWT is used to perform translation on both the input and output paths. This is functional, however there is a big performance hit in the receive path. Early demux occurs before the routing lookup (a hit actually obviates the route lookup). Therefore the stack currently performs early demux before translation so that a local connection with ILA addresses is never matched. Note that this issue is not just with ILA, but pretty much any translated or encapsulated packet handled by LWT would miss the opportunity for early demux. Solving the general problem seems non trivial since we would need to move the route lookup before early demx thereby mitigating the value.
This patch set addresses the issue for ILA by adding a fast locator lookup that occurs before early demux. This is done by using a hook at NF_INET_PRE_ROUTING. For the backend we implement an rhashtable that contains identifier to locator to mappings. The table also allows more specific matches that include original locator and interface. Note that we are not adding functionality to support nfhook for output (e.g. NF_INET_POST_ROUTING). This is because netfilter is done post routing which is prolematic since we need are changing the destination address in ILA. There is an in/out parameter in table entries to allow for the future possibility of performing lookups on the output path. This patch set: - Add an rhashtable function to atomically replace and element. This is useful to implement sub-trees from a table entry without needing to use a special anchor structure as the table entry. - Add a start callback for starting a netlink dump. - Creates an ila directory under net/ipv6 and moves ila.c to it. ila.c is split into ila_common.c and ila_lwt.c. - Implement a table to do identifier->locator mapping. This is an rhashtable - Configuration for the table with netlink - Set nfhook for IPv6 NF_INET_PRE_ROUTING.do ILA lookup and translation. Testing: Running 200 netperf TCP_RR streams No ILA, baseline 85.36% CPU utilization 1917187 tps 90/157/327 50/90/99% latencies ILA before fix (LWT on both input and output) 82.86% CPU utilization 1668895 tps (-14% from baseline) 106/180/336 50/90/99% latencies ILA after fix (NF hook for input) 82.69% CPU utilization 1865113 tps (-2.7% from baseline) 93/162/331 50/90/99% latencies Tom Herbert (4): rhashtable: add function to replace an element netlink: add a start callback for starting a netlink dump ila: Create net/ipv6/ila directory ila: Add support for netfilter NF_INET_PRE_ROUTING hook include/linux/netlink.h | 2 + include/linux/rhashtable.h | 80 ++++++ include/net/genetlink.h | 2 + include/uapi/linux/ila.h | 22 ++ net/ipv6/Makefile | 2 +- net/ipv6/ila.c | 229 ---------------- net/ipv6/ila/Makefile | 7 + net/ipv6/ila/ila.h | 48 ++++ net/ipv6/ila/ila_common.c | 102 +++++++ net/ipv6/ila/ila_lwt.c | 149 ++++++++++ net/ipv6/ila/ila_xlat.c | 665 +++++++++++++++++++++++++++++++++++++++++++++ net/netlink/af_netlink.c | 4 + net/netlink/genetlink.c | 16 ++ 13 files changed, 1098 insertions(+), 230 deletions(-) delete mode 100644 net/ipv6/ila.c create mode 100644 net/ipv6/ila/Makefile create mode 100644 net/ipv6/ila/ila.h create mode 100644 net/ipv6/ila/ila_common.c create mode 100644 net/ipv6/ila/ila_lwt.c create mode 100644 net/ipv6/ila/ila_xlat.c -- 2.4.6 -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html