Factor increases are not needed, +1 appears to work as well.
ok dlg

m_cluncount() needs to walk the mbuf chain to correctly uncount all clusters
but don't do that in m_free() as that will cause a double loop behaviour when
called via m_freem().
OK dlg@, deraadt@

MCLADDREFERENCE() was the last user of MBUFLOCK() inline the code and remove
MBUFLOCK()

Another unused macro that needs to go is MCHTYPE. Changing the allocation
type of a mbuf after creation is just wrong.

Let MRESETDATA follow MEXTMALLOC. That macro is way to dangerous but luckily
it is unused.

Bye bye MEXTMALLOC you're not used and you should not be used. You're evil.

add several backend pools to allocate mbufs clusters of various sizes out
of. currently limited to MCLBYTES (2048 bytes) and 4096 bytes until pools
can allocate objects of sizes greater than PAGESIZE

add several backend pools to allocate mbufs clusters of various sizes out
of. currently limited to MCLBYTES (2048 bytes) and 4096 bytes until pools
can allocate objects of sizes greater than PAGESIZE.

this allows drivers to ask for "jumbo" packets to fill rx rings with.

the second half of this change is per interface mbuf cluster allocator
statistics. drivers can use the new interface (MCLGETI), which will use
these stats to selectively fail allocations based on demand for mbufs. if
the driver isnt rapidly consuming rx mbufs, we dont allow it to allocate
many to put on its rx ring.

drivers require modifications to take advantage of both the new allocation
semantic and large clusters.

this was written and developed with deraadt@ over the last two days
ok deraadt@ claudio@

show more ...

rename struct m_ext to be struct mbuf_ext.

ok deraadt@ claudio@

use do { } while (/* CONSTCOND */ 0) loops for all multi-line macros
ok dlg otto claudio

Remove the M_ANYCAST6 mbuf flag by doing the detection all in ip6_input().
M_ANYCAST6 was only used to signal tcp6_input() that it should drop the
packet and send back icmp error. This can be done in

Remove the M_ANYCAST6 mbuf flag by doing the detection all in ip6_input().
M_ANYCAST6 was only used to signal tcp6_input() that it should drop the
packet and send back icmp error. This can be done in ip6_input() without
the need for a mbuf flag. Gives us back one slot in m_flags for possible
future need. Looked at and some input by naddy@ and henning@. OK dlg@

show more ...

reorder mbuf flags so they are in numerical order again. brad@ concurs

Introduce the infrastructure required to support hardware VLAN tag
stripping: Add a field to the mbuf pkthdr to hold the tag and an
mbuf flag that tells if the tag is valid. Inspired by FreeBSD.

S

Introduce the infrastructure required to support hardware VLAN tag
stripping: Add a field to the mbuf pkthdr to hold the tag and an
mbuf flag that tells if the tag is valid. Inspired by FreeBSD.

Struct packing suggested by kettenis@. csum_flags is now 16 bits.
Adapt to this in the drivers.

ok reyk@, henning@

show more ...

MSIZE and MCLBYTES are now defined in sys/param.h; from Jordan Gordeev
ok thib

Kill the _MCLDEREFERENCE() macro it was only used once and it should be only
used once -- in m_free(). Removed so that people don't get stupid ideas.
OK thib@

grow MINCLSIZE to MHLEN + MLEN + 1 again. we shrinked it just before the
release to be on the safe side, now we have 6 months to find remaining
offenders misusing MINCLSIZE

shrink MINCLSIZE to MHLEN + 1 again for the release, we're not confident
enough yet that all drivers handle this correctly. intended to be grown
again after release. discussed with claudio, thib and

shrink MINCLSIZE to MHLEN + 1 again for the release, we're not confident
enough yet that all drivers handle this correctly. intended to be grown
again after release. discussed with claudio, thib and theo

show more ...

undo r1.77 from 2004 - shrinking MINCLSIZE so that we always use a cluster
for data storage instead of another mbuf if it would fit. we did that
change whiledoing the release and we thought there was

undo r1.77 from 2004 - shrinking MINCLSIZE so that we always use a cluster
for data storage instead of another mbuf if it would fit. we did that
change whiledoing the release and we thought there was a problem with this,
which later turned out not to be true. surprise and ok theo

show more ...

store a pointer to the stack side state key in the mbuf packet
header inbound. on the outbound side, we take that and look for the key
that is the exact opposite, and store that mapping in the state

store a pointer to the stack side state key in the mbuf packet
header inbound. on the outbound side, we take that and look for the key
that is the exact opposite, and store that mapping in the state key. on
subsequent packets we don't have to do the lookup on outbound any more.
almost unable to get real benchmarks going here, we know for sure this
gives a more than 5% increase in forwarding performance.
many thanks to ckuethe for stress- and performance-testing.
ok ryan theo

show more ...

divert packets to local socket without modifying the ip header;
makes transparent proxies much easier; ok beck@, feedback claudio@

move protos to right place

make the deceptively complicated leading and trailing space into functions.
this reduces kernel size quite a bit. ok claudio

provide m_inithdr(), which takes an mbuf and gives an initialized M_PKTHDR
mbuf back.
for fixing PR5563 in a few, tested janjaap@stack.nl, ok claudio

several changes:

- unconditionally include sys/malloc.h
- remove empty and unused MCLBUFREF macro
- remove unused m_clalloc function declaration
- style(9)-ify macros
- remove external declarations

several changes:

- unconditionally include sys/malloc.h
- remove empty and unused MCLBUFREF macro
- remove unused m_clalloc function declaration
- style(9)-ify macros
- remove external declarations of mbpool and mclpool
- minor comment edit for clarity and accuracy
- move MCLGET definition to more logical place

ok henning@ claudio@

show more ...

Remove the MFREE() macro and replace it with a call to m_free().
Also remove the _MEXTREMOVE macro which was only used by MFREE.
This time with the uipc_mbuf.c change that I missed last time.

Revert previous commit, it probably was intended to come with matching
changes to uipc_mbuf.c, but right now all it causes is infinite recursion.