introduce a modified allocation mechanism for mbufs and mbuf clusters; one
which can scale under SMP and which offers the possibility of resource
reclamation to be implemented in the future. Notable advantages:
o Reduce contention for SMP by offering per-CPU pools and locks.
o Better use of data cache due to per-CPU pools.
o Much less code cache pollution due to excessively large allocation macros.
o Framework for `grouping' objects from same page together so as to be able
to possibly free wired-down pages back to the system if they are no longer
needed by the network stacks.
Additional things changed with this addition:
- Moved some mbuf specific declarations and initializations from
sys/conf/param.c into mbuf-specific code where they belong.
- m_getclr() has been renamed to m_get_clrd() because the old name is really
confusing. m_getclr() HAS been preserved though and is defined to the new
name. No tree sweep has been done "to change the interface," as the old
name will continue to be supported and is not depracated. The change was
merely done because m_getclr() sounds too much like "m_get a cluster."
- TEMPORARILY disabled mbtypes statistics displaying in netstat(1) and
systat(1) (see TODO below).
- Fixed systat(1) to display number of "free mbufs" based on new per-CPU
stat structures.
- Fixed netstat(1) to display new per-CPU stats based on sysctl-exported
per-CPU stat structures. All infos are fetched via sysctl.
TODO (in order of priority):
- Re-enable mbtypes statistics in both netstat(1) and systat(1) after
introducing an SMP friendly way to collect the mbtypes stats under the
already introduced per-CPU locks (i.e. hopefully don't use atomic() - it
seems too costly for a mere stat update, especially when other locks are
already present).
- Optionally have systat(1) display not only "total free mbufs" but also
"total free mbufs per CPU pool."
- Fix minor length-fetching issues in netstat(1) related to recently
re-enabled option to read mbuf stats from a core file.
- Move reference counters at least for mbuf clusters into an unused portion
of the cluster itself, to save space and need to allocate a counter.
- Look into introducing resource freeing possibly from a kproc.
Reviewed by (in parts): jlemon, jake, silby, terry
Tested by: jlemon (Intel & Alpha), mjacob (Intel & Alpha)
Preliminary performance measurements: jlemon (and me, obviously)
URL: http://people.freebsd.org/~bmilekic/mb_alloc/
# Note, I managed to fat finger some mail commands and lost who submitted some
# of these entries. If it was you, email me and I'll for a commit to
# give proper credit.
NQNFS code is ancient, bug-ridden, and should probably be removed).
The wording here was very confusing; it was easy to get the impression
that NQNFS is an extension to NFSv3 when in fact it just uses some
NFSv3-like extensions on top of NFSv2. As witnessed by the mailing
lists and PRs, some people were reading the description and deciding
that NQNFS was what they wanted to use.
MFC after: 1 week
notes build. Instead of having doc.relnotes.mk make a guess, hardcode
quite a bit (but as little as possible) in Makefile.inc's sprinkled
strategicly throughout the tree. This has the advantage of actually
working properly (which is a Good Thing(tm)), and the disadvantages of
more files in the repository and more hardcoded paths (which are both
Bad Things(tm)).
non-printable characters to sneak into /var/log/messages (e.g.
someone aims a Solaris/Linux RCP exploit at your FreeBSD box and
you end up with his shellcode as part of a log entry). You might
get something like,
host.mydom.org login failures:
Binary file (standard input) matches
In the daily security script as a result. Allowing attackers to
mess with your security script's ability to accurately report
is a Bad Thing. Tell grep(1) to treat /var/log/messages like a
text file even if it has non-printable characters.
Submitted by: Tim Zingelman <zingelman@fnal.gov> on freebsd-security
Approved by: ru
MFC after: 1 week
the default section and into the papchap section.
It's really irritating when you run ppp with no arguments and end up
blowing away your default route !
I get a link error on in6addr_<something> and i cannot find the
symbol in any of the libraries. It might be my mistake, but in any
case the crunched binary would overflow the floppy, so...
This applies to -current only.
Make sure hints are statically compiled into the kernel,
because the bootloader is not available in picobsd and so the
hints file cannot be found at run time.
(This is kind of inconvenient if you have to handle non PnP devices,
but fortunately these days non-PnP ISA cards are disappearing...)
This must have to do with the use of devfs in -CURRENT, but i
have no idea when the devfs is actually mounted (is it a
side effect of mount -t nonfs or what ?) and when /dev/fd0c becomes
available.
For the time being, let's use this hack. Once I understand how devfs
works, this can be reverted back to the previous value, and also the
part of the build script which creates device entries can be nuked.
This is for -current only.
better place to handle dependencies.
Make another step at helping cross-compiling: when the user specifies
an alternate source tree, the script takes care of creating include
files and libraries for the new tree.
Furthermore, build and use a version of the "config" program which
matches the new sources.
It takes a long time to create libraries, and it might even not do
the right thing at once, there might be some dependencies that i
have forgotten. At any rate, with this code i have been able to
build a working picobsd image using -CURRENT sources on -STABLE
MFC after: 3 days
are duplicated by newly defined types/options in RFC3121
- We have no backward compatibility issue. There is no apps in our
distribution which use the above types/options.
Obtained from: KAME
MFC after: 2 weeks
