- Add support for auditing arguments in setaudit_addr(2)
- Fix bug which which would result in in_addr_ex tokens not being
parsed properly
- Calculate the size of an ip6 address correctly
This brings extended subject state support into RELENG_6 for audit
records. This means that applications like sshd can take advantage
of extended subject for IPv6. It should be noted that sshd has not
been modified.
marius [Sat, 10 Nov 2007 12:34:28 +0000 (12:34 +0000)]
MFC: 1.7
In _pthread_key_create() ensure that libthr is initialized. This
fixes a NULL-dereference of curthread when libstdc+ initializes
the exception handling globals on archs we can't use GNU TLS due
to lack of support in binutils 2.15 (i.e. arm and sparc64), yet,
thus making threaded C++ programs compiled with GCC 4.2.1 work
on these archs.
delphij [Fri, 9 Nov 2007 09:58:08 +0000 (09:58 +0000)]
MFC: Sync the -HEAD state of ata(4) back to RELENG_6, except interrupt filtering,
CAM MPSAFE, BIO_FLUSH and alpha support removal, which are not suitable for
RELENG_6.
This makes RELENG_6 to support the same set of ATA devices as RELENG_7 and -CURRENT
did.
jhb [Thu, 8 Nov 2007 20:13:04 +0000 (20:13 +0000)]
MFC: Rework the read/write support in the bios disk driver some to cut down
on duplicated code and support 64-bit LBAs for GPT. This also changes the
code to always use EDD on devices that support it which matches the rest of
the x86 bootstrap.
alc [Thu, 8 Nov 2007 18:02:27 +0000 (18:02 +0000)]
MFC
Add comments explaining why all stores updating a non-kernel page table
must be globally performed before calling any of the TLB invalidation
functions.
With one exception, on amd64, this requirement was already met. Fix this
one case. Also, as a clarification, change an existing atomic op into a
release.
rwatson [Thu, 8 Nov 2007 16:38:49 +0000 (16:38 +0000)]
Merge audit.c:1.29, audit.h:1.12, audit_arg.c:1.12, audit_bsm.c:1.17,
audit_bsm_klib.c:1.7, audit_pipe.c:1.11, audit_private.h:1.16,
audit_worker.c:1.16 from HEAD to RELENG_6:
Clean up audit comments--formatting, spelling, etc.
rwatson [Thu, 8 Nov 2007 15:49:00 +0000 (15:49 +0000)]
Merge audit.c:1.26 from HEAD to RELENG_6:
No need to force __inline__ of currecord(), as the compiler will usefully
inline it when needed already, and the symbol is also required outside of
audit.c. This silences a new gcc warning on the topic of using __inline__
instead of __inline.
rwatson [Thu, 8 Nov 2007 15:33:10 +0000 (15:33 +0000)]
Mege audit_arg.c:1.9 from HEAD to RELENG_6:
When returning early from audit_arg_file() due to so->so_pcb being NULL
(due to an early reset or the like), remember to unlock the socket lock.
This will not occur in 7-CURRENT, but could in theory occur in 6-STABLE.
rwatson [Thu, 8 Nov 2007 15:31:48 +0000 (15:31 +0000)]
Merge audit_arg.c:1.8, audit_bsm_klib.c:1.5 from HEAD to RELENG_6:
Add a witness sleep warning to canon_path(), which invokes vput() and hence
may perform an unbounded sleep. Remove an XXX comment suggesting that one
be added.
emax [Wed, 7 Nov 2007 17:42:45 +0000 (17:42 +0000)]
MFC
Allow RFCOMM servers to bind to a ''wildcard'' RFCOMM channel
zero (0). Actual RFCOMM channel will be assigned after listen(2)
call is done on a RFCOMM socket bound to a ''wildcard'' RFCOMM
channel zero (0).
Address locking issues in ng_btsocket_rfcomm_bind()
Submitted by: Heiko Wundram (Beenic) < wundram at beenic dot net >
Approved by: re (kensmith)
jfv [Wed, 7 Nov 2007 01:33:28 +0000 (01:33 +0000)]
This delta addresses some users experiencing watchdogs under load with
the last driver version. I suspect sheduling to be at the bottom of it,
but this version fixes it by:
- Making FAST interrupt handling off by default, if you want
to use it #define EM_FAST_IRQ in the Makefile.
- Add dual locks, one for the TX subsystem, the other CORE
for things like init and interface managment.
- Add cleaner MSI/X setup code.
- If using FAST interrupts there is now a mechanism in the
watchdog routine to kick off TX cleanups if they have not
been scheduled.
I think everyone will be happier with this change.
simon [Tue, 6 Nov 2007 18:22:05 +0000 (18:22 +0000)]
MFC 1.17:
If the INDEX file cannot be opened, print the actual filename / path
pkg_version tried to open instead of just "INDEX" to make the actual
problem more clear (e.g. missing INDEX-8).
ups [Mon, 5 Nov 2007 16:18:00 +0000 (16:18 +0000)]
MFC: Remove mpte optimization from pmap_enter_quick().
There is a race with the current locking scheme and removing
it should have no measurable performance impact.
This fixes page faults leading to panics in pmap_enter_quick_locked()
on amd64/i386.
yongari [Mon, 5 Nov 2007 02:06:27 +0000 (02:06 +0000)]
MFC rgephy.c rev 1.17-1.18, rgephyreg.h rev 1.3 to RELENG_6
rgephy.c rev 1.17, rgephyreg.h rev 1.3
Add support for RealTek RTL8211B(L) PHY. It's based on the patch
made by Michael Eisele and the patch was slightly modified by me.
With this change several NVIDIA ethernet controllers(e.g. MCP61)
works.
RTL8211B(L) is RealTek's new gigabit PHY. The PHY has several
features including crossover correction, polarity correction as
well as supporting triple speed(10/100/1000bps). Data transfer
between MAC and PHY is via RGMII for 1000baseT, MII for
10baseT/100baseTX.
Unfortunately, RealTek used the same model number for RTL8211B(L)
PHY so there is no way to discriminate between RTL8211B(L) and its
predecessors. ATM RTL8211B uses revision number 2 so checking the
revision number seems to be only way to identify it.
rgephyreg.c rev 1.18
Setting sc->mii_anegticks to MII_ANEGTICKS_GIGE in rgephy_attach()
is redundant. mii_phy_add_media() already takes care of that.
yongari [Mon, 5 Nov 2007 02:03:09 +0000 (02:03 +0000)]
MFC rgephy.c rev 1.16 to RELENG_6
o RTL8169S/8110S integrated PHY and RTL8211B are gigabit PHYs so set
mii_anegticks to MII_ANEGTICKS_GIGE and use it. Previously it used
to MII_ANEGTICKS which may not be enough to wait before retrying
autonegotiation process at 1000bps.
o Reset autonegotation timer if media option is not IFM_AUTO or we
got a valid link.
o Announce link loss right after it happends.
o Autonegiation is retried every mii_anegticks seconds.
o Report link state changes right after setting autonegotiation.
keramida [Sun, 4 Nov 2007 15:07:28 +0000 (15:07 +0000)]
MFC: mdconfig.8 -r 1.37
% Add an example which shows how mdconfig(8) can be used
% to mount an ISO 9660 CD image file.
%
% PR: 112691
% Submitted by: Warren Block, wblock at wonkity.com
luigi [Sun, 4 Nov 2007 03:28:31 +0000 (03:28 +0000)]
MFC 2007-10-05 07:26:40 UTC
Add entries for Epson multifunction scanner/printer/card readers,
with all functions supported.
More details on related topics (SANE configuration, OCR, etc.)
at http://info.iet.unipi.it/~luigi/FreeBSD/dx5050.html
dwmalone [Fri, 2 Nov 2007 16:27:22 +0000 (16:27 +0000)]
MFC 1.28:
Make the type of the memory used by the BPF filter unsigned, so it
matches the BPF registers (which are the only thing that is assigned
to/from BPF memory). This is a pedantic change that shouldn't change
any behaviour.
PR: 115931
Submitted by: Matthew Luckie <mjl@luckie.org.nz>
kientzle [Fri, 2 Nov 2007 08:44:11 +0000 (08:44 +0000)]
MFC 1.18: Skip requests use a minimum of 1 byte so that the
underlying I/O framework can avoid copying and so that we
avoid a 32-bit overflow issue that was causing bsdtar to
fail when skipping uncompressed tar entries that were an
exact multiple of 4GB.
alc [Fri, 2 Nov 2007 06:33:20 +0000 (06:33 +0000)]
MFC revision 1.389
Correct an error in vm_map_sync(), nee vm_map_clean(), that has existed
since revision 1.1. Specifically, neither traversal of the vm map checks
whether the end of the vm map has been reached. Consequently, the first
traversal can wrap around and bogusly return an error.
emax [Wed, 31 Oct 2007 18:53:07 +0000 (18:53 +0000)]
MFC: etc/rc.d/ppp,v1.14 and etc/defaults/rc.conf,v1.319
Teach /etc/rc.d/ppp to start multiple instances of ppp.
ppp_profile variable can now contain multiple profiles.
Overrides for ppp mode and nat can go into ppp_$profile_mode
and ppp_$profile_nat variables respectively. If those are
not specified, defaults from ppp_mode and ppp_nat are used.
MFC: etc/rc.d/ppp v1.15
Teach /etc/rc.d/ppp how to start/stop individual instances
of ppp. This is an extension of previous commit.
Submitted by: Yuri Kurenkov < y dot kurenkov at init dot ru >
Reviewed by: mtm
Approved by: re (kensmith)
Implement RFC3442, the Classless Static Route option.
The original DHCP specification includes a route option but it
supports only class-based routes. RFC3442 adds support for
specifying the netmask width for each static route. A variable
length encoding is used to minimize the size of this option.
The minimum size of an RFC3442 destination descriptor is five bytes,
so correct test to -ge 5. Without this change an RFC3442 encoded
default route would be ignored.
rafan [Tue, 30 Oct 2007 01:30:42 +0000 (01:30 +0000)]
Restore ctype(3) ABI forward compatibility in 6.x world. This was broken
with last ctype(3) single-byte and multi-byte separation for fixing
operating on UTF-8 locale. To be specific, we introduced a new symbol in
libc. And this symbol is referred via inline functions here. So, you can
not run a binary built with this version of libc on an older system. To
restore the compatibility, make these functions built as non-inlined
form (see lib/libc/nomarco.c) and MFC rev 1.32 which drops the usage of
that symbol in question for __isctype(). As this may impact performance,
I only intend to fix in 6.x, but not 7 (not yet released) and 8 (it's HEAD).
Discussed on: cvs-src
Requested by: scottl
Reviewed by: ache, delphij
Approved by: re (kensmith)