Commit Graph

469 Commits

Author SHA1 Message Date
Luigi Rizzo
5007b59f26 implement listing of a subset of pipes/queues/schedulers.
The filtering of the output is done in the kernel instead of userland
to reduce the amount of data transfered.
2010-03-11 22:42:33 +00:00
Luigi Rizzo
a63e8ce9e7 add back DPADD (removed by mistake in a previous commit) 2010-03-08 14:43:55 +00:00
Luigi Rizzo
34ae843479 more documentation on new dummynet features. 2010-03-05 14:13:58 +00:00
Luigi Rizzo
f10f583fec make the listing of queues/pipes/schedulers handle the case of
data size increasing while we fetch the info.
2010-03-04 16:56:36 +00:00
Luigi Rizzo
c751221af4 fix handling of sets 2010-03-04 16:55:32 +00:00
Luigi Rizzo
6429067819 reduce diffs with the cross-platform version (windows needs
some extra initialization)
2010-03-04 16:54:56 +00:00
Luigi Rizzo
1009f2c9dd remove stale comment 2010-03-04 16:08:51 +00:00
Luigi Rizzo
cc4d3c30ea Bring in the most recent version of ipfw and dummynet, developed
and tested over the past two months in the ipfw3-head branch.  This
also happens to be the same code available in the Linux and Windows
ports of ipfw and dummynet.

The major enhancement is a completely restructured version of
dummynet, with support for different packet scheduling algorithms
(loadable at runtime), faster queue/pipe lookup, and a much cleaner
internal architecture and kernel/userland ABI which simplifies
future extensions.

In addition to the existing schedulers (FIFO and WF2Q+), we include
a Deficit Round Robin (DRR or RR for brevity) scheduler, and a new,
very fast version of WF2Q+ called QFQ.

Some test code is also present (in sys/netinet/ipfw/test) that
lets you build and test schedulers in userland.

Also, we have added a compatibility layer that understands requests
from the RELENG_7 and RELENG_8 versions of the /sbin/ipfw binaries,
and replies correctly (at least, it does its best; sometimes you
just cannot tell who sent the request and how to answer).
The compatibility layer should make it possible to MFC this code in a
relatively short time.

Some minor glitches (e.g. handling of ipfw set enable/disable,
and a workaround for a bug in RELENG_7's /sbin/ipfw) will be
fixed with separate commits.

CREDITS:
This work has been partly supported by the ONELAB2 project, and
mostly developed by Riccardo Panicucci and myself.
The code for the qfq scheduler is mostly from Fabio Checconi,
and Marta Carbone and Francesco Magno have helped with testing,
debugging and some bug fixes.
2010-03-02 17:40:48 +00:00
Ruslan Ermilov
bee10047c0 Fixed dependencies (make checkdpadd). 2010-02-25 20:24:19 +00:00
Luigi Rizzo
472099c4b0 implement a new match option,
lookup {dst-ip|src-ip|dst-port|src-port|uid|jail} N

which searches the specified field in table N and sets tablearg
accordingly.
With dst-ip or src-ip the option replicates two existing options.
When used with other arguments, the option can be useful to
quickly dispatch traffic based on other fields.

Work supported by the Onelab project.

MFC after:	1 week
2009-12-15 09:46:27 +00:00
Luigi Rizzo
2f12516b1f fix the indentation for addr: values
MFC after:	3 days
2009-12-15 09:32:35 +00:00
Luigi Rizzo
6e47ca1bc5 restore setting of sin_len (was removed in 1.146 last february) as
it seems that now it is necessary for 'forward' to work outside lo0.
The bug (and fix) was reported on 8.0. This patch probably applies
to RELENG_7 as well.
It seems that 'pf' has a similar bug.

Submitted by:	Lytochkin Boris
MFC after:	3 days
2009-12-06 18:04:26 +00:00
Luigi Rizzo
7139c5c496 fix argument type in the call to expand_number
Submitted by:	gcc 4.3
MFC after:	3 days
2009-12-04 14:18:30 +00:00
Luigi Rizzo
01ab76323b use qsort_r instead of heapsort;
staticize two functions.

MFC after:	3 days
2009-12-03 12:23:48 +00:00
Alexander Leidinger
ac2e492b19 Fix minor resource leak in a function.
Reviewed by:	luigi
MFC after:	1 week
2009-11-21 10:46:49 +00:00
Christian Brueffer
8fba046b09 Fix setfib(1) section number.
PR:		133765
Submitted by:	Konstantin Zolotukhin <erebus@gorodok.net>
MFC after:	3 days
2009-09-18 14:17:00 +00:00
Oleg Bulyzhin
a6f1444804 - 'burst' description rewritten.
Submitted by:	Ben Kaduk
Approved by:	re (kib)
2009-06-26 19:49:06 +00:00
Maxim Konovalov
067e91e8c2 o Kill grammar nits.
PR:		docs/136061
Submitted by:	Ben Kaduk
MFC after:	1 week
2009-06-26 05:09:00 +00:00
Oleg Bulyzhin
6882bf4d92 - fix dummynet 'fast' mode for WF2Q case.
- fix printing of pipe profile data.
- introduce new pipe parameter: 'burst' - how much data can be sent through
  pipe bypassing bandwidth limit.
2009-06-24 22:57:07 +00:00
Luigi Rizzo
7a459517da Permit the specification of bandwidth values within
"profile" files (bandwidth is mandatory when using a
profile, so it makes sense to have everything in one place).

Update the manpage accordingly.

Submitted by:	Marta Carbone
2009-06-08 14:32:29 +00:00
Luigi Rizzo
91dbeea7b6 add a missing format in a printf
Detected building with gcc 4.3.3

MFC after:	3 days
2009-06-08 10:53:18 +00:00
Luigi Rizzo
b87ce5545b Several ipfw options and actions use a 16-bit argument to indicate
pipes, queues, tags, rule numbers and so on.
These are all different namespaces, and the only thing they have in
common is the fact they use a 16-bit slot to represent the argument.

There is some confusion in the code, mostly for historical reasons,
on how the values 0 and 65535 should be used. At the moment, 0 is
forbidden almost everywhere, while 65535 is used to represent a
'tablearg' argument, i.e. the result of the most recent table() lookup.

For now, try to use explicit constants for the min and max allowed
values, and do not overload the default rule number for that.

Also, make the MTAG_IPFW declaration only visible to the kernel.

NOTE: I think the issue needs to be revisited before 8.0 is out:
the 2^16 namespace limit for rule numbers and pipe/queue is
annoying, and we can easily bump the limit to 2^32 which gives
a lot more flexibility in partitioning the namespace.

MFC after:	5 days
2009-06-05 16:16:07 +00:00
Luigi Rizzo
bbf46d80db remove a printf that was only useful for debugging.
MFC after:	3 days
2009-06-05 13:11:34 +00:00
Tom Rhodes
81bdd4cbcd Kill hard sentence break added in the previous revision. 2009-04-11 08:52:02 +00:00
Luigi Rizzo
4bb7ae9deb Add emulation of delay profiles, which lets you model various
types of MAC overheads such as preambles, link level retransmissions
and more.

Note- this commit changes the userland/kernel ABI for pipes
(but not for ordinary firewall rules) so you need to rebuild
kernel and /sbin/ipfw to use dummynet features.

Please check the manpage for details on the new feature.

The MFC would be trivial but it breaks the ABI, so it will
be postponed until after 7.2 is released.

Interested users are welcome to apply the patch manually
to their RELENG_7 tree.

Work supported by the European Commission, Projects Onelab and
Onelab2 (contract 224263).
2009-04-09 12:46:00 +00:00
Maxim Konovalov
d8ec4cdefb o Grammar. 2009-04-08 17:46:45 +00:00
Luigi Rizzo
de24303241 Various cleanup of text, moving a couple of paragraphs
above to avoid referencing undefined terms (humans are not compilers
but still care about these things).

Change some .Sh to .Ss to better reflect the structure of the text.

No new content.
2009-04-08 15:18:21 +00:00
Tom Rhodes
c4abdf1c01 Remove contractions, reword a sentence to avoid a double negative,
and bump document date for previous change.

OKed by:	piso
2009-04-07 13:51:41 +00:00
Paolo Pisati
0240be035c Improve a bit reass documentation:
-document fragment handling sysctls
-mention some caveats about fragments handling (and to deal with it)
2009-04-05 15:24:27 +00:00
Paolo Pisati
eb2e411915 Implement an ipfw action to reassemble ip packets: reass. 2009-04-01 20:23:47 +00:00
Christian Brueffer
cdd14ccabe Mdoc style, spelling, grammar and wording fixes. This manpage needs more work. 2009-03-19 10:42:07 +00:00
Luigi Rizzo
1fd3fc6988 move a variable declaration to the beginning of the block
(unfortunately, it is far away; we need to pack this code in
a better way).
2009-03-05 08:08:09 +00:00
Luigi Rizzo
d7a32e24af remove some signed/unsigned and one const/!const warning 2009-03-05 08:01:58 +00:00
Luigi Rizzo
3e9771d290 mark a function static, as it is 2009-03-05 08:01:19 +00:00
Paolo Pisati
37ce2656ec Add SCTP NAT support.
Submitted by: CAIA (http://caia.swin.edu.au)
2009-02-07 18:49:42 +00:00
Luigi Rizzo
a52e28c7dd Explain that we assume AF_INET and only use the addr and port field
from a struct sockaddr_in, so there is no need to initialize sin_len
2009-02-02 11:02:19 +00:00
Luigi Rizzo
d84d38734f remove duplicate #include 2009-02-02 10:58:05 +00:00
Luigi Rizzo
23c608c8f6 put the altq-related functions into a separate file.
Minor cleanup of the includes used by the various source files,
including annotations of why certain headers are used.
2009-02-01 16:00:49 +00:00
Luigi Rizzo
16e3606f57 Avoid the use of duplicated typedefs -- see the comment for details. 2009-01-28 11:43:12 +00:00
Luigi Rizzo
50a99912c1 fix printing of uint64_t values, so we can use WARNS=2 2009-01-27 20:26:45 +00:00
Luigi Rizzo
b361cf5805 fix wrong variable usage... 2009-01-27 12:24:53 +00:00
Luigi Rizzo
ead75a59f1 Put nat and ipv6 support in their own files.
Usual moving of code with no changes from ipfw2.c to the
newly created files, and addition of prototypes to ipfw2.h

I have added forward declarations for ipfw_insn_* in ipfw2.h
to avoid a global dependency on ip_fw.h
2009-01-27 12:01:30 +00:00
Luigi Rizzo
4e9c8ae7b5 Put dummynet-related code in a separate file.
To this purpose, add prototypes for global functions in ipfw2.h
and move there also the list of tokens used in various places in the code.
2009-01-27 11:06:59 +00:00
Luigi Rizzo
d36bf6e792 never mind, for the time being let's stick with WARNS=0 until
we sort out all proper printf formats.
2009-01-27 11:03:47 +00:00
Luigi Rizzo
3c0c871777 Start splitting the monster file in smaller blocks.
In this episode:
- introduce a common header with a minimal set of common definitions;
- bring the main() function and options parser in main.c
- rename the main functions with an ipfw_ prefix

No code changes except for the introduction of a global variable,
resvd_set_number, which stores the RESVD_SET value from ip_fw.h
and is used to remove the dependency of main.c from ip_fw.h
(and the subtree of dependencies) for just a single constant.
2009-01-27 10:18:55 +00:00
Luigi Rizzo
0e22daad2d put the usage() function inline, it was only 1 line and used once;
slightly reformat the help() text;
slightly correct the text for the 'extraneous filename' error message;
2009-01-27 09:27:13 +00:00
Luigi Rizzo
db7c522411 put all options in a single struct, and document them.
This will allow us to easily restore the original values when processing
commands from a file (where each individual line can have its own options).
2009-01-27 09:06:25 +00:00
Luigi Rizzo
2edea98d7f I believe this is safe to build with WARNS=2 now 2009-01-27 09:04:29 +00:00
Luigi Rizzo
be39e0b2b1 remove a couple of rarely used #define;
change PRINT_UINT from a macro to a function (renaming is
postponed to reduce clutter)
2009-01-27 07:40:16 +00:00
Luigi Rizzo
5f356082eb wrap all malloc/calloc/realloc calls so they exit on failure
without having to check in each place.

Remove an wrong strdup from previous commit.
2009-01-26 14:26:35 +00:00