o Introduce new m_getcl() interface routine that allocates an mbuf
and a cluster in one shot.
o Introduce MBP_PERSIST and MBP_PERSISTENT control bits to mb_alloc();
MBP_PERSIST means "if you can allocate, then keep the cache lock
held on exit," and MBP_PERSISTENT means "a cache lock is alredy held
on entry, so allocate from the specified (already locked) cache."
They may be used in combination.
o m_getcl() uses the MBP_PERSIST/MBP_PERSISTENT interface so that it
doesn't drop the cache lock in between the mbuf and cluster allocations.
o m_getm(), which takes a size and allocates an mbuf + cluster "best fit"
chain, has been moved from uipc_mbuf.c to subr_mbuf.c and shown how to
use MBP_PERSIST/MBP_PERSISTENT to attempt to do a grouped allocation
without dropping the cache lock in between.
Why this is good: much less bus-locked lock acquires/drops when they're
not needed. Also, prototype for m_getcl():
struct mbuf * m_getcl(int how, short type, int flags);
"how" and "type" are self-explanatory. "flags" may be M_PKTHDR, in
which case m_getcl() will make the mbuf a pkthdr-mbuf.
While I'm in subr_mbuf.c:
o Every exported routine now has a nice comment with a description of
the expected arguments. Eventually, mbuf(9) needs to be re-vamped
but there's still more code to write/finalize before I get to that.
o internal macros have been changed a bit.
o consistently use 'short' for "type." This somehow slipped through
before (that 'type' was sometimes declared as int).
Alfred has been pushing for the MBP_PERSIST{,ENT} thing for almost a
year now. Luigi asked for m_getcl(), and will probably MFC that
part of this commit.
TODO [Related]: teach mb_free() about MBP_PERSIST{, ENT}.
John Baldwin [Mon, 15 Jul 2002 15:18:34 +0000 (15:18 +0000)]
Make NDGBPORTS an MI option since the dgb(4) driver is an MI driver.
Remove comments about NDGBPORTS from the options* files. Please document
options in NOTES, not in the options* files.
Ian Dowse [Mon, 15 Jul 2002 14:37:36 +0000 (14:37 +0000)]
Re-fetch the interface handle after setting the config number,
because the previous interface handle gets freed when the config
number is set. This fixes a problem where memory could be accessed
after it was freed when the interface was ifconfig'd up.
Mark Murray [Mon, 15 Jul 2002 13:58:35 +0000 (13:58 +0000)]
Upgrade the random device to use a "real" hash instead of building
one out of a block cipher. This has 2 advantages:
1) The code is _much_ simpler
2) We aren't committing our security to one algorithm (much as we
may think we trust AES).
While I'm here, make an explicit reseed do a slow reseed instead
of a fast; this is in line with what the original paper suggested.
Alan Cox [Sun, 14 Jul 2002 23:51:55 +0000 (23:51 +0000)]
o Lock page queue accesses by vm_page_wire() that aren't
within a critical section.
o Assert that the page queues lock is held in vm_page_wire()
unless an Alpha.
Luigi Rizzo [Sun, 14 Jul 2002 23:47:18 +0000 (23:47 +0000)]
Implement keepalives for dynamic rules, so they will not expire
just because you leave your session idle.
Also, put in a fix for 64-bit architectures (to be revised).
In detail:
ip_fw.h
* Reorder fields in struct ip_fw to avoid alignment problems on
64-bit machines. This only masks the problem, I am still not
sure whether I am doing something wrong in the code or there
is a problem elsewhere (e.g. different aligmnent of structures
between userland and kernel because of pragmas etc.)
* added fields in dyn_rule to store ack numbers, so we can
generate keepalives when the dynamic rule is about to expire
ip_fw2.c
* use a local function, send_pkt(), to generate TCP RST for Reset rules;
* save about 250 bytes by cleaning up the various snprintf()
in ipfw_log() ...
* ... and use twice as many bytes to implement keepalives
(this seems to be working, but i have not tested it extensively).
Keepalives are generated once every 5 seconds for the last 20 seconds
of the lifetime of a dynamic rule for an established TCP flow. The
packets are sent to both sides, so if at least one of the endpoints
is responding, the timeout is refreshed and the rule will not expire.
You can disable this feature with
sysctl net.inet.ip.fw.dyn_keepalive=0
(the default is 1, to have them enabled).
MFC after: 1 day
(just kidding... I will supply an updated version of ipfw2 for
RELENG_4 tomorrow).
Alfred Perlstein [Sun, 14 Jul 2002 23:14:02 +0000 (23:14 +0000)]
Silence several warnings due to functions that needed to take a void *
having a char * as an argument instead. clnt_dg_control(), clnt_raw_control(),
clnt_vc_control().
Alfred Perlstein [Sun, 14 Jul 2002 17:54:00 +0000 (17:54 +0000)]
- TI-RPC is now the default again for code generation.
- As before, inetd support support is turned of per default.
Code for inetd can be made with -I.
- Support for ``transport monitors'' and the NLSPROVIDER env
variable is still there , even if their use is not clear in
non TLI stream based systems like Free-/NetBSD. It can be activated
with -P.
- There are a few corrections in rpcgen.1 and usage function to conform
to the code. Added and documented -P
- I removed the #ifdefs checks for Free-/NetBSD since we are the only
ones who use this code. MaxOS X may have the same limitations as
we have, so this code will correctly build for them.
Thomas Moestl [Sun, 14 Jul 2002 12:09:48 +0000 (12:09 +0000)]
Miscellaneous fixes:
- always reinitialize the rx descriptors, even if the mbuf is kept.
This should fix the hangs on ifconfig that were observed
- on an rx overflow, reinitialize the descriptor so that the interface
will not hang
- correct some bus_dmamap_sync() calls
- correct some debug messages
- some minor nits
Luigi Rizzo [Sun, 14 Jul 2002 09:07:13 +0000 (09:07 +0000)]
Enable building of picobsd using CURRENT sources again.
Following a suggestion by Ruslan, the initial creation of the
includes and libraries (and build tools) is now done by
invoking "make buildworld" (with -DPICOBSD which eventually will
limit the amount of stuff built with a 2-line change in Makefile.inc1).
The correct environment is then used for subsequent builds.
Also remove write_mfs_in_kernel.c in favour of using dd
All the above is conditional on __FreeBSD_version, as the previous
method still worked for versions earlier than 500035, and I am
unsure on how the "new" method works for earlier versions.
Finally, note that the crunch.conf files need some work because
some libraries (e.g. gmd) have gone away from the base installation.
part of a greater patch set..
1/ don't need to set td_state to TDS_RUNNING in fork_return.
it's already set in choosethread().
2/ Set a child process state to "normal" as opposed to "new"
when we allow it to be put on the run queue.
Allows child to receive signals from the parent if the parent
runs first and tries to immediatly signal he child.
Submitted by: (part 2) Thomas Moestl <tmoestl@gmx.net>
Warner Losh [Sun, 14 Jul 2002 06:47:52 +0000 (06:47 +0000)]
Rearrange previous commit that passed the vendor id to the kernel in a way
that's binary compatible for -stable. While binary compatibility doesn't
matter much in -current, it is critical for -stable. This change requires
pccardd/pccardc to be recompiled.