]> CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
FreeBSD/FreeBSD.git
10 years agoImport Hyper-V paravirtualized drivers from projects/hyperv
Peter Grehan [Fri, 13 Sep 2013 18:47:58 +0000 (18:47 +0000)]
Import Hyper-V paravirtualized drivers from projects/hyperv
branch into head.

Approved by: re@ (hrs)
Obtained from: Microsoft, NetApp, and Citrix.

10 years agoUnregister inet/inet6 pfil hooks on vnet destroy.
Mikolaj Golub [Fri, 13 Sep 2013 18:45:10 +0000 (18:45 +0000)]
Unregister inet/inet6 pfil hooks on vnet destroy.

Discussed with: andre
Approved by: re (rodrigc)

10 years agoImprove readelf notes output for Linux ELF files
Ed Maste [Fri, 13 Sep 2013 18:21:31 +0000 (18:21 +0000)]
Improve readelf notes output for Linux ELF files

Add four ELF note constants:
- NT_FILE and NT_SIGINFO (core file notes output by recent Linux kernels)
- NT_GNU_ABI_TAG (was incorrectly reported as NT_VERSION)
- NT_GNU_BUILD_ID (used for locating standalone debug files)

Approved by: re (kib)

10 years agoadd support for writing the pid of the daemon program to a pid file so
John-Mark Gurney [Fri, 13 Sep 2013 16:57:28 +0000 (16:57 +0000)]
add support for writing the pid of the daemon program to a pid file so
that daemon can be used w/ rc.subr and ports can use the additional
functionality, such as keeping the ldap daemon up and running, and have
the proper program to signal to exit..

PR: bin/181341
Submitted by: feld
Approved by: re (glebius)

10 years agoNote about the OpenSSH change.
Dag-Erling Smørgrav [Fri, 13 Sep 2013 10:37:24 +0000 (10:37 +0000)]
Note about the OpenSSH change.

Approved by: re (gjb)

10 years agoWhen opening or closing fifo, ensure that the vnode is locked
Konstantin Belousov [Fri, 13 Sep 2013 06:52:23 +0000 (06:52 +0000)]
When opening or closing fifo, ensure that the vnode is locked
exclusively.  Filesystems are assumed to disable shared locking for
the fifo vnode locks, but some do not.

Reported and tested by: olgeni
Discussed with: avg
Sponsored by:   The FreeBSD Foundation
MFC after: 1 week
Approved by: re (glebius)

10 years agoReduce the scope of the proctree_lock. If several processes cause
Konstantin Belousov [Fri, 13 Sep 2013 06:39:10 +0000 (06:39 +0000)]
Reduce the scope of the proctree_lock. If several processes cause
continuous calls to the uprintf(9), the proctree_lock could be
shared-locked for indefinite amount of time, starving exclusive
requests. Since proctree_lock is needed for fork() and exit(), this
effectively stops the machine.

While there, do the similar reduction for tprintf(9).

Reported and tested by: pho
Reviewed by:    ed
Sponsored by:   The FreeBSD Foundation
MFC after: 1 week
Approved by: re (glebius)

10 years agoDo not install freebsd-update.conf.5 manual if WITHOUT_FREEBSD_UPDATE is
Glen Barber [Fri, 13 Sep 2013 03:39:19 +0000 (03:39 +0000)]
Do not install freebsd-update.conf.5 manual if WITHOUT_FREEBSD_UPDATE is
set.

MFC after: 3 days
Approved by: re (delphij)
Sponsored by: The FreeBSD Foundation

10 years agoChange VM object lock assertion to match locking higher in the call
Nathan Whitehorn [Fri, 13 Sep 2013 01:12:45 +0000 (01:12 +0000)]
Change VM object lock assertion to match locking higher in the call
chain. This repairs a panic observed during pageout on some 64-bit
PowerPC systems.

Submitted by: grehan
Approved by: re (kib)
MFC after: 2 weeks
Revisit after: 10.0

10 years agoFix an issue that caused Integrated RAID volumes on LSI mps(4) controllers
Kenneth D. Merry [Thu, 12 Sep 2013 22:06:12 +0000 (22:06 +0000)]
Fix an issue that caused Integrated RAID volumes on LSI mps(4) controllers
to not get scanned on boot.

The problem originated in change 253549.  With the change to the mps(4)
driver to scan only targets that it knows it has (as opposed to scanning
the entire bus), scanning RAID volumes on boot was omitted.

So, for versions of FreeBSD that have the scanning changes
(__FreeBSD_version 1000039 and higher), scan RAID volumes that are added
whether or not we're booting.

PR: kern/181784
Reported by: Xiguang Wang <kurapica@gmail.com>
Tested by: Dennis Glatting <dg@pki2.com>
Sponsored by: Spectra Logic
Approved by: re (delphij)
MFC After: 3 days

10 years agoAfter r255321, clang uses libc++ by default. This leads to a lot of
Dimitry Andric [Thu, 12 Sep 2013 21:24:59 +0000 (21:24 +0000)]
After r255321, clang uses libc++ by default.  This leads to a lot of
errors when you enable WITH_GNUCXX to build libstdc++, since it will
include C++ headers from the libc++ installation under ${WORLDTMP}, and
those are not compatible with libstdc++ at all.

To fix this, add -stdlib=libstdc++ to CXXFLAGS when building libstdc++
(and its companion libsupc++) with clang.

Approved by: re (delphij)

10 years agoAfter r255294, building lib/msun's symbol map (using clang as the
Dimitry Andric [Thu, 12 Sep 2013 20:51:48 +0000 (20:51 +0000)]
After r255294, building lib/msun's symbol map (using clang as the
preprocessor) gives the following error:

--- Version.map ---
<stdin>:287:4: error: invalid preprocessing directive
        # Implemented as weak aliases for imprecise versions
          ^
1 error generated.

Change the comment to a C-style one, to prevent this error.

Approved by: re (hrs)

10 years ago- Document the UQUAD sysctl variants.
John Baldwin [Thu, 12 Sep 2013 20:49:20 +0000 (20:49 +0000)]
- Document the UQUAD sysctl variants.
- Clarify that exactly one of the "access" flags is required and
  list the optional flags in a separate list.  Prefer bundling
  CTLFLAG_TUN into the access flag by not documenting it as an
  optional flag to set.

Approved by: re (glebius)
MFC after: 1 week

10 years agoFix an off-by-one error when populating mincore(2) entries for
John Baldwin [Thu, 12 Sep 2013 20:46:32 +0000 (20:46 +0000)]
Fix an off-by-one error when populating mincore(2) entries for
skipped entries.  lastvecindex references the last valid byte,
so the new bytes should come after it.

Approved by: re (kib)
MFC after: 1 week

10 years agoFix a typo.
John Baldwin [Thu, 12 Sep 2013 19:52:23 +0000 (19:52 +0000)]
Fix a typo.

Approved by: re (gjb)

10 years agoDocument several recent changes to vm_map_find(9):
John Baldwin [Thu, 12 Sep 2013 19:51:40 +0000 (19:51 +0000)]
Document several recent changes to vm_map_find(9):
- Document the max_addr parameter that restricts mappings to a subset of
  the map's address space.
- Document VMFS_ALIGNED_SPACE and update for the rename of VMFS_SUPER_SPACE.
  In addition, use a table that describes the different find space
  strategies in more detail.

Reviewed by: alc
Approved by: re (kib)

10 years ago- Decode the idtype argument passed to wait6() in kdump and truss.
John Baldwin [Thu, 12 Sep 2013 18:08:25 +0000 (18:08 +0000)]
- Decode the idtype argument passed to wait6() in kdump and truss.
- Don't treat an options argument of 0 to wait4() as an error in
  kdump.
- Decode the wait options passed to wait4() and wait6() in truss
  and decode the returned rusage and exit status.

Approved by: re (kib)
MFC after: 1 week

10 years agoRegen.
John Baldwin [Thu, 12 Sep 2013 18:03:51 +0000 (18:03 +0000)]
Regen.

Approved by: re (kib)

10 years agoFix the type of the idtype argument to wait6() in syscalls.master.
John Baldwin [Thu, 12 Sep 2013 18:01:13 +0000 (18:01 +0000)]
Fix the type of the idtype argument to wait6() in syscalls.master.

(Accidentally missed this in the previous commit)

Approved by: re (kib)
MFC after: 1 week

10 years agoFix the type of the idtype argument to wait6() in syscalls.master.
John Baldwin [Thu, 12 Sep 2013 17:52:18 +0000 (17:52 +0000)]
Fix the type of the idtype argument to wait6() in syscalls.master.

Approved by: re (kib)
MFC after: 1 week

10 years agoUpdate head/ to -ALPHA1 status, as part of the 10.0-RELEASE
Glen Barber [Thu, 12 Sep 2013 17:51:18 +0000 (17:51 +0000)]
Update head/ to -ALPHA1 status, as part of the 10.0-RELEASE
cycle.

Approved by: re (implicit)
Sponsored by: The FreeBSD Foundation

10 years agoDon't issue USB resume signalling in USB device mode, if the USB power
Hans Petter Selasky [Thu, 12 Sep 2013 10:39:38 +0000 (10:39 +0000)]
Don't issue USB resume signalling in USB device mode, if the USB power
mode is ON and suspend is detected. This confuses iPads running in USB
host mode at least.

MFC after: 1 week
Approved by: re (hrs)

10 years agoWe are exceeding default limit (256) of bracket nesting and clang does not like
Hiren Panchasara [Thu, 12 Sep 2013 02:31:32 +0000 (02:31 +0000)]
We are exceeding default limit (256) of bracket nesting and clang does not like
it. We should probably fix the code but appeasing clang with this fix for now.
gcc does not have such limit.

Reviewed by: jmg
Approved by: re (hrs), sbruno (mentor, implicit)

10 years agoConsistently reference file descriptors as "fd". 55 other manpages
Bryan Drewery [Thu, 12 Sep 2013 00:53:38 +0000 (00:53 +0000)]
Consistently reference file descriptors as "fd". 55 other manpages
used "fd", while these used "d" and "filedes".

MFC after: 1 week
Approved by: gjb
Approved by: re (delphij)

10 years agoFix make bootstrapping with WITH_DEBUG_FILES=
Ed Maste [Thu, 12 Sep 2013 00:23:09 +0000 (00:23 +0000)]
Fix make bootstrapping with WITH_DEBUG_FILES=

A HEAD buildworld on 9.x first bootstraps bmake, but this failed when
building with standalone debug.  Pass in the PROGNAME override to the
'make all' stage as well as 'make install' so that the .debug file is
created with the correct name.

Reviewed by: sjg
Sponsored by: The FreeBSD Foundation
Approved by: re (rodrigc)

10 years agoDo not emit size for non-regular files. There is nothing that
Xin LI [Thu, 12 Sep 2013 00:14:25 +0000 (00:14 +0000)]
Do not emit size for non-regular files.  There is nothing that
mtree(1) can do in this situation and would cause confusion.

MFC candidate.

Approved by: re (hrs)

10 years agoProvide pr_ctloutput method for AF_LOCAL/SOCK_SEQPACKET sockets.
Gleb Smirnoff [Wed, 11 Sep 2013 18:22:30 +0000 (18:22 +0000)]
Provide pr_ctloutput method for AF_LOCAL/SOCK_SEQPACKET sockets.
This makes setsockopt() on them working.

Reported by: Yuri <yuri rawbw.com>
Approved by: re (kib)

10 years agoFix build with gcc.
Konstantin Belousov [Wed, 11 Sep 2013 17:31:22 +0000 (17:31 +0000)]
Fix build with gcc.

Build-tested by: gjb
Approved by: re (glebius)

10 years agoPrior to r254304, we only began scanning the active page queue when the
Alan Cox [Wed, 11 Sep 2013 17:23:42 +0000 (17:23 +0000)]
Prior to r254304, we only began scanning the active page queue when the
amount of free memory was close to the point at which we would begin
reclaiming pages.  Now, we continuously scan the active page queue,
regardless of the amount of free memory.  Consequently, we are continuously
calling pmap_ts_referenced() on active pages.

Prior to this change, pmap_ts_referenced() would always demote superpage
mappings in order to obtain finer-grained reference information.  This made
sense because we were coming under memory pressure and would soon have to
begin reclaiming pages.  Now, however, with continuous scanning of the
active page queue, these demotions are taking a toll on performance.  To
address this problem, I have replaced the demotion with a heuristic for
periodically clearing the reference flag on superpage mappings.

Approved by: re (kib)
Sponsored by: EMC / Isilon Storage Division

10 years agoClear correct data structure.
Hans Petter Selasky [Wed, 11 Sep 2013 10:18:36 +0000 (10:18 +0000)]
Clear correct data structure.

MFC after: 1 week
Approved by: re (hrs)

10 years ago Clean up SIOCSIFDSTADDR usage from ifnet drivers. The ioctl itself is
Gleb Smirnoff [Wed, 11 Sep 2013 09:19:44 +0000 (09:19 +0000)]
  Clean up SIOCSIFDSTADDR usage from ifnet drivers. The ioctl itself is
extremely outdated, and I doubt that it was ever used for ifnet drivers.
It was used for AF_INET sockets in pre-FreeBSD time.

Approved by: re (hrs)
Sponsored by: Nginx, Inc.

10 years agoFix a limitation in bhyve that would limit the number of virtual machines to
Neel Natu [Wed, 11 Sep 2013 07:11:14 +0000 (07:11 +0000)]
Fix a limitation in bhyve that would limit the number of virtual machines to
the maximum number of VT-d domains (256 on a Sandybridge). We now allocate a
VT-d domain for a guest only if the administrator has explicitly configured
one or more PCI passthru device(s).

If there are no PCI passthru devices configured (the common case) then the
number of virtual machines is no longer limited by the maximum number of
VT-d domains.

Reviewed by: grehan@
Approved by: re@

10 years agoCleanup elf macros
Baptiste Daroussin [Wed, 11 Sep 2013 06:42:55 +0000 (06:42 +0000)]
Cleanup elf macros

Only define EF_MIPS_ABI when not already supplied
Remove old now unused ARM macros

Reported by: imp
Approved by: re (kib)

10 years agoImplement sendfile(2) for the posix shared memory segment file descriptor,
Konstantin Belousov [Wed, 11 Sep 2013 06:41:15 +0000 (06:41 +0000)]
Implement sendfile(2) for the posix shared memory segment file descriptor,
in addition to the regular files.

Requested by: alc
Discussed with: emaste
Tested by: pho (previous version)
Sponsored by: The FreeBSD Foundation
Approved by: re (hrs)

10 years agoIFC @ r255459
Peter Grehan [Wed, 11 Sep 2013 00:19:16 +0000 (00:19 +0000)]
IFC @ r255459

10 years agoStick to traditional DEFAULT_SYS_PATH (/usr/share/mk)
Simon J. Gerraty [Tue, 10 Sep 2013 23:31:38 +0000 (23:31 +0000)]
Stick to traditional DEFAULT_SYS_PATH (/usr/share/mk)

Reviewed by: obrien
Approved by: re

10 years agoMissed in r255386: libssh is more than just the .so file itself.
Dag-Erling Smørgrav [Tue, 10 Sep 2013 22:40:38 +0000 (22:40 +0000)]
Missed in r255386: libssh is more than just the .so file itself.

Approved by: re (blanket)

10 years agoChange the default value of VerifyHostKeyDNS to "yes" if compiled with
Dag-Erling Smørgrav [Tue, 10 Sep 2013 22:30:22 +0000 (22:30 +0000)]
Change the default value of VerifyHostKeyDNS to "yes" if compiled with
LDNS.  With that setting, OpenSSH will silently accept host keys that
match verified SSHFP records.  If an SSHFP record exists but could not
be verified, OpenSSH will print a message and prompt the user as usual.

Approved by: re (blanket)

10 years agoClean up the OpenSSH build. It is now possible to build most components
Dag-Erling Smørgrav [Tue, 10 Sep 2013 22:26:11 +0000 (22:26 +0000)]
Clean up the OpenSSH build.  It is now possible to build most components
as static binaries, if desired.  The one exception is sshd, which runs
into trouble due to libpam.a's includion of pam_ssh.

Make OpenSSH use LDNS if available.  This allows it to verify signed
SSHFP records.

Approved by: re (blanket)

10 years agoRemove documentation describing functionality which geom(4) does not,
Colin Percival [Tue, 10 Sep 2013 21:16:18 +0000 (21:16 +0000)]
Remove documentation describing functionality which geom(4) does not,
in fact, provide.

Reviewed by: phk
MFC after: 3 days
Approved by: re (gjb)

10 years agoAdd support to detect arm vs armv6
Baptiste Daroussin [Tue, 10 Sep 2013 20:56:01 +0000 (20:56 +0000)]
Add support to detect arm vs armv6

There are two different versions of the ARM ABI depending on the
TARGET_ARCH. As these are sligntly different a package built for
one may not work on another. We need to detect which one we are on
by parsing the .ARM.attributes section.

This will only work on the ARM EABI as this section is part of the
ABI definition. As armv6 only supports the ARM EABI this is not a
problem for the oabi.

Older versions of libelf in FreeBSD fail to read the
.ARM.attributes section needed. As armv6 is unsupported on these
versions we can assume we are running on arm.

Submitted by: andrew
Approved by: re (delphij)
Obtained from: pkgng git

10 years agoClean up the Kerberos build by turning libheimipcc and libheimipcs into
Dag-Erling Smørgrav [Tue, 10 Sep 2013 18:40:43 +0000 (18:40 +0000)]
Clean up the Kerberos build by turning libheimipcc and libheimipcs into
private shared libraries, instead of hacked-together archives of PIC
objects.  This makes it possible to build a static libkrb5 that works.

Reviewed by: stas
Approved by: re (gjb)

10 years agoMake this more deterministic by sorting the libraries before processing
Dag-Erling Smørgrav [Tue, 10 Sep 2013 18:34:38 +0000 (18:34 +0000)]
Make this more deterministic by sorting the libraries before processing
them and ensuring that we always use the samme collation rules.

Approved by: re (gjb)

10 years agoadd note about gcc and aesni...
John-Mark Gurney [Tue, 10 Sep 2013 17:26:09 +0000 (17:26 +0000)]
add note about gcc and aesni...

Approved by: re (gjb, kib)

10 years agoBack out r255440. /usr/bin/gcc @r255185 (2013-09-03) can build this.
David E. O'Brien [Tue, 10 Sep 2013 16:50:13 +0000 (16:50 +0000)]
Back out r255440.  /usr/bin/gcc @r255185 (2013-09-03) can build this.

Approved by: re (kib)

10 years agoAdd a sendfile regression test for transmit length > file size.
Ed Maste [Tue, 10 Sep 2013 13:51:19 +0000 (13:51 +0000)]
Add a sendfile regression test for transmit length > file size.

This test identified the issue fixed in FreeBSD-SA-13:11.sendfile.

Sponsored by: The FreeBSD Foundation
Approved by: re (glebius)

10 years agoipfilter 5.1.2 no longer supports sysctl. Use ipf -V to determine if
Cy Schubert [Tue, 10 Sep 2013 13:48:33 +0000 (13:48 +0000)]
ipfilter 5.1.2 no longer supports sysctl. Use ipf -V to determine if
available (the kernel module is loaded or compiled into the kernel).

Approved by: glebius (mentor)
Approved by: re (blanket)

10 years agoMake a bump for r255426.
Gleb Smirnoff [Tue, 10 Sep 2013 10:38:15 +0000 (10:38 +0000)]
Make a bump for r255426.

Approved by: re (gjb)

10 years agoFix the length calculation for the final block of a sendfile(2)
Dag-Erling Smørgrav [Tue, 10 Sep 2013 10:05:59 +0000 (10:05 +0000)]
Fix the length calculation for the final block of a sendfile(2)
transmission which could be tricked into rounding up to the nearest
page size, leaking up to a page of kernel memory.  [13:11]

In IPv6 and NetATM, stop SIOCSIFADDR, SIOCSIFBRDADDR, SIOCSIFDSTADDR
and SIOCSIFNETMASK at the socket layer rather than pass them on to the
link layer without validation or credential checks.  [SA-13:12]

Prevent cross-mount hardlinks between different nullfs mounts of the
same underlying filesystem.  [SA-13:13]

Security: CVE-2013-5666
Security: FreeBSD-SA-13:11.sendfile
Security: CVE-2013-5691
Security: FreeBSD-SA-13:12.ifioctl
Security: CVE-2013-5710
Security: FreeBSD-SA-13:13.nullfs
Approved by: re

10 years agoOnly use a clang'ism if ${CC} is clang.
David E. O'Brien [Tue, 10 Sep 2013 05:49:31 +0000 (05:49 +0000)]
Only use a clang'ism if ${CC} is clang.

Reviewed by: sjg
Approved by: re (kib)

10 years agoCall free() on the pointer returned from malloc().
Konstantin Belousov [Tue, 10 Sep 2013 05:17:53 +0000 (05:17 +0000)]
Call free() on the pointer returned from malloc().

Reported and tested by: Oliver Pinter <oliver.pntr@gmail.com>
Sponsored by: The FreeBSD Foundation
MFC after: 3 days
Approved by: re (delphij)

10 years agoGo way past 11 and bump bhyve's max vCPUs to 16.
Peter Grehan [Tue, 10 Sep 2013 03:48:18 +0000 (03:48 +0000)]
Go way past 11 and bump bhyve's max vCPUs to 16.

This should be sufficient for 10.0 and will do
until forthcoming work to avoid limitations
in this area is complete.

Thanks to Bela Lubkin at tidalscale for the
headsup on the apic/cpu id/io apic ASL parameters
that are actually hex values and broke when
written as decimal when 11 vCPUs were configured.

Approved by: re@

10 years agoMFV r247844 (illumos-gate 13975:ef6409bc370f)
Xin LI [Tue, 10 Sep 2013 01:46:47 +0000 (01:46 +0000)]
MFV r247844 (illumos-gate 13975:ef6409bc370f)

Illumos ZFS issues:
  3582 zfs_delay() should support a variable resolution
  3584 DTrace sdt probes for ZFS txg states

Provide a compatibility shim for Solaris's cv_timedwait_hires
to help aid future porting.

Approved by: re (ZFS blanket)

10 years agoFix the aborting of association with the iterator using an empty
Michael Tuexen [Mon, 9 Sep 2013 21:40:07 +0000 (21:40 +0000)]
Fix the aborting of association with the iterator using an empty
user initiated error cause (using SCTP_ABORT|SCTP_SENDALL).

Approved by: re (delphij)
MFC after: 1 week

10 years agoEmit the correct standard library dependency line for C++ programs. In
Dag-Erling Smørgrav [Mon, 9 Sep 2013 21:18:16 +0000 (21:18 +0000)]
Emit the correct standard library dependency line for C++ programs.  In
the CLANG_IS_CC case, the default is now libc++.  Only use libstdc++ if
!CLANG_IS_CC or it was explicitly requested in CXXFLAGS.

Submitted by: theraven
Approved by: re (gjb)

10 years agoPass -n (do not emit comments) when saving mtree information for future
Xin LI [Mon, 9 Sep 2013 20:36:28 +0000 (20:36 +0000)]
Pass -n (do not emit comments) when saving mtree information for future
mergemaster(8) runs.

MFC after: 3 days
Approved by: re (kib)

10 years agoRevert the kvp code - there's still some work that
Peter Grehan [Mon, 9 Sep 2013 19:27:44 +0000 (19:27 +0000)]
Revert the kvp code - there's still some work that
needs to be done for that.

Discussed with: Microsoft hyper-v devs

10 years agoAdd a mmap flag (MAP_32BIT) on 64-bit platforms to request that a mapping use
John Baldwin [Mon, 9 Sep 2013 18:11:59 +0000 (18:11 +0000)]
Add a mmap flag (MAP_32BIT) on 64-bit platforms to request that a mapping use
an address in the first 2GB of the process's address space.  This flag should
have the same semantics as the same flag on Linux.

To facilitate this, add a new parameter to vm_map_find() that specifies an
optional maximum virtual address.  While here, fix several callers of
vm_map_find() to use a VMFS_* constant for the findspace argument instead of
TRUE and FALSE.

Reviewed by: alc
Approved by: re (kib)

10 years agoThe correct variable is apparently MACHINE_ARCH, not TARGET_ARCH.
Dag-Erling Smørgrav [Mon, 9 Sep 2013 17:38:02 +0000 (17:38 +0000)]
The correct variable is apparently MACHINE_ARCH, not TARGET_ARCH.

Approved by: re (blanket)

10 years agoMake the primary name of the OF console device /dev/ofwcons, and only
Nathan Whitehorn [Mon, 9 Sep 2013 16:51:35 +0000 (16:51 +0000)]
Make the primary name of the OF console device /dev/ofwcons, and only
alias it to the contents of the output property if it is defined. This
avoids a panic when booting machines (QEMU) where the output-device
property is not defined.

Since output-device is free-form and potentially conflicts with other
entries in /dev, I also am not sure we should be doing the aliasing at
all, but this at least makes things work again.

Approved by: re (kib)

10 years agoThese three files appeared in 6.0p1, which was imported into the vendor
Dag-Erling Smørgrav [Mon, 9 Sep 2013 13:56:58 +0000 (13:56 +0000)]
These three files appeared in 6.0p1, which was imported into the vendor
branch but never merged to head.  They were inadvertantly left out when
6.1p1 was merged to head.  It didn't make any difference at the time,
because they were unused, but one of them is required for DNS-based host
key verification.

Approved by: re (blanket)

10 years agoRevert r255420. This seems to break some Powermac systems and will be
Nathan Whitehorn [Mon, 9 Sep 2013 13:40:53 +0000 (13:40 +0000)]
Revert r255420. This seems to break some Powermac systems and will be
revisited much later.

Pointy hat to: me
Approved by: re (kib, implicit due to breakage 10 minutes ago)

10 years agoAttach only on hardware that is actually supported as opposed to hardware
Nathan Whitehorn [Mon, 9 Sep 2013 12:54:08 +0000 (12:54 +0000)]
Attach only on hardware that is actually supported as opposed to hardware
that seems like it has some of the problems we might want.

Approved by: re (kib)

10 years agoRaise artificial limits on number of CPUs and number of interrupts.
Nathan Whitehorn [Mon, 9 Sep 2013 12:52:34 +0000 (12:52 +0000)]
Raise artificial limits on number of CPUs and number of interrupts.

Approved by: re (kib)

10 years agoAdd POWER CPUs to the kernel's knowledge. This does not imply we currently
Nathan Whitehorn [Mon, 9 Sep 2013 12:51:24 +0000 (12:51 +0000)]
Add POWER CPUs to the kernel's knowledge. This does not imply we currently
actually run on any machines with POWER CPUs but avoids closing that door
unnecessarily.

Approved by: re (kib)

10 years agoAdd hook called when every new processor is brought online -- including the
Nathan Whitehorn [Mon, 9 Sep 2013 12:49:19 +0000 (12:49 +0000)]
Add hook called when every new processor is brought online -- including the
BSP -- so that platform modules have a chance to add the new CPU to any
internal bookkeeping.

Approved by: re (kib)

10 years agoUse a spin lock instead of a mutex to gate RTAS. This is required if RTAS
Nathan Whitehorn [Mon, 9 Sep 2013 12:45:41 +0000 (12:45 +0000)]
Use a spin lock instead of a mutex to gate RTAS. This is required if RTAS
calls are involved in interrupt handling.

Approved by: re (kib)

10 years agoUse the canonical bits for wired, etc. in the PTE. This is important for
Nathan Whitehorn [Mon, 9 Sep 2013 12:44:48 +0000 (12:44 +0000)]
Use the canonical bits for wired, etc. in the PTE. This is important for
interactions with certain kinds of hypervisors that look into the PTEs
more closely than they should.

Approved by: re (kib)

10 years agoLatest update from Microsoft.
Peter Grehan [Mon, 9 Sep 2013 08:07:46 +0000 (08:07 +0000)]
Latest update from Microsoft.

Obtained from: Microsoft Hyper-v dev team

10 years agoRemove unneeded mappings from libmap32.conf. Move it up one level and
Dag-Erling Smørgrav [Mon, 9 Sep 2013 06:02:30 +0000 (06:02 +0000)]
Remove unneeded mappings from libmap32.conf.  Move it up one level and
install it on powerpc64 in addition to amd64.

Reviewed by: kib
Approved by: re (blanket)

10 years agoIn r243868, the error message buffer errmsg have been changed from
Xin LI [Mon, 9 Sep 2013 05:01:18 +0000 (05:01 +0000)]
In r243868, the error message buffer errmsg have been changed from
an on-stack array to a pointer and therefore sizeof(errmsg) would
become 4 or 8 bytes depending on the architecture.

Fix this by using ERRMSGL in place of sizeof().

Submitted by: J David <j.david.lists@gmail.com>
MFC after: 3 days
Approved by: re (kib)

10 years agoRework the tx credit mechanism between the cxgbe/tom driver
Navdeep Parhar [Mon, 9 Sep 2013 04:38:57 +0000 (04:38 +0000)]
Rework the tx credit mechanism between the cxgbe/tom driver
and the card.  This helps smooth out some burstiness in the
exchange.

Approved by: re (glebius)

10 years agoFix a miscalculation that caused cxgbe/tom to auto-increment
Navdeep Parhar [Mon, 9 Sep 2013 00:16:59 +0000 (00:16 +0000)]
Fix a miscalculation that caused cxgbe/tom to auto-increment
a TOE socket's tx buffer size too aggressively.

Approved by: re (delphij)

10 years agoPrior to r254304, we only began scanning the active page queue when the
Alan Cox [Sun, 8 Sep 2013 21:30:53 +0000 (21:30 +0000)]
Prior to r254304, we only began scanning the active page queue when the
amount of free memory was close to the point at which we would begin
reclaiming pages.  Now, we continuously scan the active page queue,
regardless of the amount of free memory.  Consequently, we are continuously
calling pmap_ts_referenced() on active pages.

Prior to this change, pmap_ts_referenced() would always demote superpage
mappings in order to obtain finer-grained reference information.  This made
sense because we were coming under memory pressure and would soon have to
begin reclaiming pages.  Now, however, with continuous scanning of the
active page queue, these demotions are taking a toll on performance.  For
example, on one of my test machines, the running time for the HPCC Random
Access benchmark (also known as GUPS) has increased by 54%.  To address this
problem, I have replaced the demotion with a heuristic for periodically
clearing the reference flag on superpage mappings.

Reviewed by: kib
Approved by: re (glebius)
Sponsored by: EMC / Isilon Storage Division

10 years agoHook host(1) up to the build in the LDNS case.
Dag-Erling Smørgrav [Sun, 8 Sep 2013 20:48:23 +0000 (20:48 +0000)]
Hook host(1) up to the build in the LDNS case.

Approved by: re (blanket)

10 years agoImport Magerya Vitaly's ldns-host, and build it instead of the BIND version
Dag-Erling Smørgrav [Sun, 8 Sep 2013 19:40:32 +0000 (19:40 +0000)]
Import Magerya Vitaly's ldns-host, and build it instead of the BIND version
in the WITH_LDNS_UTILS case.

Approved by: re (blanket)

10 years agoLDNS needs OpenSSL. This wasn't a problem as long as it was only build
Dag-Erling Smørgrav [Sun, 8 Sep 2013 19:39:18 +0000 (19:39 +0000)]
LDNS needs OpenSSL.  This wasn't a problem as long as it was only build
statically, since any program using it would have to link with it anyway.

Approved by: re (blanket)

10 years agoUse correct type for the vmx vlan filter table
Bryan Venteicher [Sun, 8 Sep 2013 19:13:06 +0000 (19:13 +0000)]
Use correct type for the vmx vlan filter table

Approved by: re (glebius, gjb)

10 years agoRelese the interface in the last.
Mikolaj Golub [Sun, 8 Sep 2013 18:19:40 +0000 (18:19 +0000)]
Relese the interface in the last.

Reviewed by: glebius
Approved by: re (kib)

10 years agoDrain for the xbusy state for two places which potentially do
Konstantin Belousov [Sun, 8 Sep 2013 17:51:22 +0000 (17:51 +0000)]
Drain for the xbusy state for two places which potentially do
pmap_remove_all(). Not doing the drain allows the pmap_enter() to
proceed in parallel, making the pmap_remove_all() effects void.

The race results in an invalidated page mapped wired by usermode.

Reported and tested by: pho
Reviewed by: alc
Sponsored by: The FreeBSD Foundation
Approved by: re (glebius)

10 years agoMFV (r255387): undo autoprop damage
Dag-Erling Smørgrav [Sun, 8 Sep 2013 16:56:17 +0000 (16:56 +0000)]
MFV (r255387): undo autoprop damage

Approved by: re (glebius)

10 years agoFix verbose output line; needs <NL>
Mark Murray [Sun, 8 Sep 2013 16:48:03 +0000 (16:48 +0000)]
Fix verbose output line; needs <NL>

Submitted by: Sean Bruno <sean_bruno@yahoo.com>
Approved by: re (glebius)

10 years agoDamn autoprops!
Dag-Erling Smørgrav [Sun, 8 Sep 2013 11:21:08 +0000 (11:21 +0000)]
Damn autoprops!

10 years agoMake libldns and libssh private.
Dag-Erling Smørgrav [Sun, 8 Sep 2013 10:04:26 +0000 (10:04 +0000)]
Make libldns and libssh private.

Approved by: re (blanket)

10 years agoAdd a stock libmap32.conf for amd64. The first two lines have no effect
Dag-Erling Smørgrav [Sun, 8 Sep 2013 09:46:22 +0000 (09:46 +0000)]
Add a stock libmap32.conf for amd64.  The first two lines have no effect
except to document the hardcoded standard library search path for 32-bit
binaries.  The third line performs the equivalent substitution for the
private library directory.

Ironically, these entries rely on functionality which is only available
in the COMPAT_32BIT version of rtld-elf.

Approved by: re (blanket)

10 years agoCreate a private library directory (LIBPRIVATEDIR) for libraries which
Dag-Erling Smørgrav [Sun, 8 Sep 2013 09:40:23 +0000 (09:40 +0000)]
Create a private library directory (LIBPRIVATEDIR) for libraries which
we don't want to expose but which can't or shouldn't be static.

To mark a library as private, define PRIVATELIB in its Makefile.  It
will be installed in LIBPRIVATEDIR, which is normally /usr/lib/private
(or /usr/lib32/private for 32-bit libraries on 64-bit platforms).

To indicate that a program or library depends on a private library,
define USEPRIVATELIB in its Makefile.  The correct version of
LIBPRIVATEDIR will be added to its run-time library search path.

Approved by: re (blanket)

10 years agoFix the build; Certain linkable symbols need to always be present.
Mark Murray [Sat, 7 Sep 2013 22:07:36 +0000 (22:07 +0000)]
Fix the build; Certain linkable symbols need to always be present.

Pass the pointy hat please.

Also unblock the software (Yarrow) generator for now. This will be
reverted; Yarrow needs to block until secure, not this behaviour
of serving as soon as asked.

Folks with specific requiremnts will be able to (can!) unblock this
device with any write, and are encouraged to do so in /etc/rc.d/*
scripting. ("Any" in this case could be "echo '' > /dev/random" as
root).

10 years agoFix error in r252115: space for the softc needs to be allocated. This
Nathan Whitehorn [Sat, 7 Sep 2013 20:52:31 +0000 (20:52 +0000)]
Fix error in r252115: space for the softc needs to be allocated. This
seemed to be working by chance on most systems.

10 years agoTweak wording.
Dag-Erling Smørgrav [Sat, 7 Sep 2013 20:25:22 +0000 (20:25 +0000)]
Tweak wording.

10 years agoUpdate to OpenPAM Nummularia.
Dag-Erling Smørgrav [Sat, 7 Sep 2013 19:43:39 +0000 (19:43 +0000)]
Update to OpenPAM Nummularia.

10 years agoMerge upstream r743: caught_signal should be static.
Dag-Erling Smørgrav [Sat, 7 Sep 2013 19:27:58 +0000 (19:27 +0000)]
Merge upstream r743: caught_signal should be static.

10 years agoSort properly.
Pawel Jakub Dawidek [Sat, 7 Sep 2013 19:16:02 +0000 (19:16 +0000)]
Sort properly.

10 years agoAdjust optional obsolete files with new MK_GNUCXX
Antoine Brodin [Sat, 7 Sep 2013 19:04:28 +0000 (19:04 +0000)]
Adjust optional obsolete files with new MK_GNUCXX

10 years agoFix panic in cap_rights_is_valid() when invalid rights are provided -
Pawel Jakub Dawidek [Sat, 7 Sep 2013 19:03:16 +0000 (19:03 +0000)]
Fix panic in cap_rights_is_valid() when invalid rights are provided -
the right_to_index() function should assert correctness in this case.

Improve other assertions.

Reported by: pho
Tested by: pho

10 years agoThis was a good idea that never went anywhere.
Dag-Erling Smørgrav [Sat, 7 Sep 2013 18:55:52 +0000 (18:55 +0000)]
This was a good idea that never went anywhere.

10 years agoExport a function to allow BCM2835's peripheral devices to enable their
Luiz Otavio O Souza [Sat, 7 Sep 2013 18:48:15 +0000 (18:48 +0000)]
Export a function to allow BCM2835's peripheral devices to enable their
altenate pin function (from GPIO pins) as needed.

Approved by: adrian (mentor)

10 years agoMFV (r255364): move the code around in preparation for Nummularia.
Dag-Erling Smørgrav [Sat, 7 Sep 2013 18:46:35 +0000 (18:46 +0000)]
MFV (r255364): move the code around in preparation for Nummularia.

10 years agoNetlogic XLP network driver update
Jayachandran C. [Sat, 7 Sep 2013 18:26:16 +0000 (18:26 +0000)]
Netlogic XLP network driver update

Changes are to
- update board and network interface detection logic
- fix reading onboard CPLD in little-endian config
- print NAE frequency conrrectly for Bx chips
- update XAUI config to disable Rx/Tx until interface is up

Submitted by: Venkatesh J V <venkatesh.vivekanandan@broadcom.com>

10 years agoUse a better version of memcpy/bcopy for mips kernel.
Jayachandran C. [Sat, 7 Sep 2013 16:31:30 +0000 (16:31 +0000)]
Use a better version of memcpy/bcopy for mips kernel.

Use a variant of mips libc memcpy for kernel. This implementation uses
64-bit operations when compiled for 64-bit, and is significantly faster
in that case.

Submitted by: Tanmay Jagdale <tanmayj@broadcom.com>

10 years agoVendor import of OpenPAM Nummularia..
Dag-Erling Smørgrav [Sat, 7 Sep 2013 16:15:30 +0000 (16:15 +0000)]
Vendor import of OpenPAM Nummularia..