c2e43be1c5
methods called Vestigial Time-Wait (VTW) and Maximum Segment Lifetime Truncation (MSLT). MSLT and VTW were contributed by Coyote Point Systems, Inc. Even after a TCP session enters the TIME_WAIT state, its corresponding socket and protocol control blocks (PCBs) stick around until the TCP Maximum Segment Lifetime (MSL) expires. On a host whose workload necessarily creates and closes down many TCP sockets, the sockets & PCBs for TCP sessions in TIME_WAIT state amount to many megabytes of dead weight in RAM. Maximum Segment Lifetimes Truncation (MSLT) assigns each TCP session to a class based on the nearness of the peer. Corresponding to each class is an MSL, and a session uses the MSL of its class. The classes are loopback (local host equals remote host), local (local host and remote host are on the same link/subnet), and remote (local host and remote host communicate via one or more gateways). Classes corresponding to nearer peers have lower MSLs by default: 2 seconds for loopback, 10 seconds for local, 60 seconds for remote. Loopback and local sessions expire more quickly when MSLT is used. Vestigial Time-Wait (VTW) replaces a TIME_WAIT session's PCB/socket dead weight with a compact representation of the session, called a "vestigial PCB". VTW data structures are designed to be very fast and memory-efficient: for fast insertion and lookup of vestigial PCBs, the PCBs are stored in a hash table that is designed to minimize the number of cacheline visits per lookup/insertion. The memory both for vestigial PCBs and for elements of the PCB hashtable come from fixed-size pools, and linked data structures exploit this to conserve memory by representing references with a narrow index/offset from the start of a pool instead of a pointer. When space for new vestigial PCBs runs out, VTW makes room by discarding old vestigial PCBs, oldest first. VTW cooperates with MSLT. It may help to think of VTW as a "FIN cache" by analogy to the SYN cache. A 2.8-GHz Pentium 4 running a test workload that creates TIME_WAIT sessions as fast as it can is approximately 17% idle when VTW is active versus 0% idle when VTW is inactive. It has 103 megabytes more free RAM when VTW is active (approximately 64k vestigial PCBs are created) than when it is inactive. |
||
|---|---|---|
| .. | ||
| dev | ||
| fs | ||
| include | ||
| kern | ||
| librump | ||
| net | ||
| ldscript.rump | ||
| Makefile | ||
| Makefile.rump | ||
| README.dirs | ||
| TODO | ||
$NetBSD: README.dirs,v 1.11 2010/05/11 11:58:14 pooka Exp $
The following is a quick rundown of the current directory structure.
First, components in the kernel namespace, i.e. compiled with -D_KERNEL
sys/rump/librump - kernel runtime emulation
/rumpkern - kernel core, e.g. syscall, interrupt and lock support
/rumpcrypto - kernel cryptographic routines
/rumpdev - device support, e.g. autoconf subsystem
/rumpnet - networking support and sockets layer
/rumpvfs - file system support
sys/rump/include
/machine - used for architectures where the rump ABI is not yet the
same as the kernel module ABI. will eventually disappear
completely
/rump - rump headers installed to userspace
sys/rump/dev - device components, e.g. audio, raidframe, usb drivers
sys/rump/fs - file system components
/lib/lib${fs} - kernel file system code
sys/rump/net - networking components
/lib/libnet - subroutines from sys/net, e.g. route and if_ethersubr
/lib/libnetinet - TCP/IP
/lib/libvirtif - a virtual interface which uses host tap(4) to shovel
packets. This is used by netinet and if_ethersubr.
/lib/libsockin - implements PF_INET using host kernel sockets. This is
mutually exclusive with net, netinet and virtif.
The rest are out-of-kernel components (i.e. no -D_KERNEL)
related to rump.
hypercall interface:
src/lib/librumpuser
The "rumpuser" set of interfaces is used by rump to communicate
with the host.
Users:
src/lib
/libp2k - puffs-to-vfs adaption layer, userspace namespace
/libukfs - user kernel file system, a library to access file system
images (or devices) directly in userspace without going
through a system call and puffs. It provides a slightly
higher interface than pure rump syscalls.
src/usr.sbin/puffs
rump_$fs - userspace file system daemons using the kernel fs code
src/share/examples/rump
Various examples detailing use of rump in different scenarios.
These are provided source-only.