o Add devel/subversion and devel/subversion-static so we provide
a package for source-based users to check out the various trees,
both with and without extra dependencies.
This is a direct commit to stable/9.
Approved by: kib (mentor)
Approved by: re (glebius)
MFC: r253506
The NFSv4 server incorrectly assumed that the high order words of
the attribute bitmap argument would be non-zero. This caused an
interoperability problem for a recent patch to the Linux NFSv4 client.
The Linux folks have changed their patch to avoid this, but this
patch fixes the problem on the server.
Reported and tested by: a.heider@gmail.com (Andre Heider)
Approved by: re (Xin Li)
1) POSIX requires rand(3) return values to be in the [0, RAND_MAX]
range, but ACM formula we use have internal state (and return value) in
the [1, 0x7ffffffe] range, so our RAND_MAX (0x7fffffff) is never reached
because it is off by one, zero is not reached too.
Correct both RAND_MAX and rand(3) return value, shifting last one
to the 0 by 1 subtracted, resulting POSIXed [0, 0x7ffffffd(=new
RAND_MAX)] range.
2) Add a checks for not overflowing on too big seeds. It may happens on
the machines, where sizeof(unsigned int) > 32 bits.
This change is binary compatible because range is reduced, not expanded,
so no bump is needed.
In this GRN, Marcel Moolenaar overhauled the logic for mounting
the root file system on bootup:
|------------------------------------------------------------------------
|r214006 | marcel | 2010-10-17 22:01:53 -0700 (Sun, 17 Oct 2010) | 20 lines
|
| Re-implement the root mount logic using a recursive approach, whereby each
|root file system (starting with devfs and a synthesized configuration) can
|contain directives for mounting another file system as root.
|------------------------------------------------------------------------
This commit adds a mount.conf(8) man page which documents
the root mount logic. mount.conf(8) also provides some examples
for the /.mount.conf file, which can be used to change the root mount behavior.
MFC r253280:
Only copy as much bytes as there in superblock, instead of the full
block copy, when copying the superblock into the snapshot. UFS1 does
not align superblock on the block boundary, and bcopy runs off the end
of the buffer.
MFC 252576:
Don't perform the acpi_DeviceIsPresent() check for PCI-PCI bridges. If
we are probing a PCI-PCI bridge it is because we found one by enumerating
the devices on a PCI bus, so the bridge is definitely present. A few
BIOSes report incorrect status (_STA) for some bridges that claimed they
were not present when in fact they were.
While here, move this check earlier for Host-PCI bridges so attach fails
before doing any work that needs to be torn down.
- Configurations in ipv6_prefix_IF should be recognized even if there is no
ifconfig_IF_ipv6.
- DAD wait should be performed at once, not on a per-interface basis, if
possible. This fixes an issue that a system with a lot of IPv6-capable
interfaces takes too long for booting.
Correct the Intel network driver module builds. They were not
defining INET or INET6, and in the case of ixgbe this will cause
a panic in the TSO setup code, but in all cases the ioctl behavior
is different, this change makes the module and static consistent.
When fiddling with options of which registers to copy out for
a mailbox command and which registers to copy back in when
the command completes, the bits being set need to not only
specify what bits you want to add from the default from the
table but also what bits you want *subtract* (mask) from the
default from the table.
Fix a problem with READ ELEMENT STATUS that occurs on some
changers that don't support the DVCID and CURDATA bits that were
introduced in the SMC spec.
These changers will return an Illegal Request type error if the
bits are set. This causes "chio status" to fail.
The fix is two-fold. First, for changers that claim to be SCSI-2
or older, don't set the DVCID and CURDATA bits for READ ELEMENT
STATUS. For newer changers (SCSI-3 and newer), we default to
setting the new bits, but back off and try the READ ELEMENT STATUS
without the bits if we get an Illegal Request type error.
This has been tested on a Qualstar TLS-8211, which is a SCSI-2
changer that does not support the new bits, and a Spectra T-380,
which is a SCSI-3 changer that does support the new bits. In the
absence of a SCSI-3 changer that does not support the bits, I
tested that with some error injection code. (The SMC spec says
that support for CURDATA is mandatory, and DVCID is optional.)
scsi_ch.c: Add a new quirk, CH_Q_NO_DVCID that gets set for
SCSI-2 and older libraries, or newer libraries that
report errors when the DVCID/CURDATA bits are set.
In chgetelemstatus(), use the new quirk to
determine whether or not to set DVCID and CURDATA.
If we get an error with the bits set, back off and
try without the bits. Set the quirk flag if the
read element status succeeds without the bits set.
Increase the READ ELEMENT STATUS timeout to 60
seconds after testing with a Spectra T-380. The
previous value was 10 seconds, and too short for
the T-380. This may be decreased later after
some additional testing and investigation.
Tested by: Andre Albsmeier <Andre.Albsmeier@siemens.com>
Sponsored by: Spectra Logic
MFC r253224:
Fix bug in deleting files: If two ports had the same tarball and one of
them changed (or was removed from the tree) then portsnap would delete
that file. This happened earlier today when one of two empty port
directories was removed. Uniquifying the lists of needed files fixes
this.
Send per-namespace logpage commands to the controller devnode, so they
are processed as admin commands, not I/O commands.
As part of this change, pull out the code for parsing a namespace node
string into a separate function, since it is used for both identify and
logpage commands.
Try to read firmware image before prompting the user to confirm
firmware download. This correctly prints an error and exits for
an incorrect firmware image name before prompting the user to
confirm the download.
r253109:
Incorporate feedback from bde@ based on r252672 changes:
* Use 0/1 instead of sysexits. Man pages are confusing on this topic,
but 0/1 is sufficient for nvmecontrol.
* Use err function family where possible instead of fprintf/exit.
* Fix some typing errors.
* Clean up some error message inconsistencies.
r253279:
%d should be used for printing int32_t instead of %zd.
clang does not complain about this - only gcc.
Incorporated r253279 prior to MFC timeout because it is required for gcc
builds.
In r227207, to fix the issue with possible NULL inp_socket pointer
dereferencing, when checking for SO_REUSEPORT option (and SO_REUSEADDR
for multicast), INP_REUSEPORT flag was introduced to cache the socket
option. It was decided then that one flag would be enough to cache
both SO_REUSEPORT and SO_REUSEADDR: when processing SO_REUSEADDR
setsockopt(2), it was checked if it was called for a multicast address
and INP_REUSEPORT was set accordingly.
Unfortunately that approach does not work when setsockopt(2) is called
before binding to a multicast address: the multicast check fails and
INP_REUSEPORT is not set.
Fix this by adding INP_REUSEADDR flag to unconditionally cache
SO_REUSEADDR.
PR: 179901
Submitted by: Michael Gmelin freebsd grem.de (initial version)
Reviewed by: rwatson
Approved by: re (kib)
marius [Fri, 12 Jul 2013 18:02:10 +0000 (18:02 +0000)]
MFC: r240981, r240990, r240992, r244695
Add 32-bit ABI compat shims. Those are necessary for i386 binary-only
tools like sysutils/hpacucli (HP P4xx RAID controller management
suite) working on amd64 systems.
marius [Fri, 12 Jul 2013 16:41:58 +0000 (16:41 +0000)]
MFC: r253120
- As it turns out, not only MSI-X is broken for devices passed through by
VMware up to at least ESXi 5.1. Actually, using INTx in that case instead
may still result in interrupt storms, with MSI being the only working
option in some configurations. So introduce a PCI_QUIRK_DISABLE_MSIX quirk
which only blacklists MSI-X but not also MSI and use it for the VMware
PCI-PCI-bridges. Note that, currently, we still assume that if MSI doesn't
work, MSI-X won't work either - but that's part of the internal logic and
not guaranteed as part of the API contract. While at it, add and employ
a pci_has_quirk() helper.
Reported and tested by: Paul Bucher
- Use NULL instead of 0 for pointers.
Submitted by: jhb (mostly)
Approved by: re (hrs), jhb
MFC r251282:
When auto-sizing the buffer cache, limit the amount of physical memory
used as the estimation of size, to 16GB. This provides around 100K of
buffer headers and corresponding KVA for buffer map at the peak.
Sizing the cache larger is not useful, also resulting in the wasting
and exhausting of KVA for large machines.
MFC note: the commit message was adjusted to match the code change, the
sizing cap is for 16GB, as noted by delphij.
MFC r245066
-------------------------------------------------------------------------
Teach the kernel to recognize that it is executing inside a bhyve virtual
machine.
-------------------------------------------------------------------------
This will help a 9.2 guest to run more effectively as a bhyve guest.
- Allow to configure net.inet6.ip6.{accept_rtadv,no_radr} by the
loader tunables as well because they have to be configured before
interface initialization for AF_INET6.
- Allow ND6_IFF_AUTO_LINKLOCAL for IFT_BRIDGE. An interface with IFT_BRIDGE
is initialized with !ND6_IFF_AUTO_LINKLOCAL && !ND6_IFF_ACCEPT_RTADV
regardless of net.inet6.ip6.accept_rtadv and net.inet6.ip6.auto_linklocal.
To configure an autoconfigured link-local address (RFC 4862), the
following rc.conf(5) configuration can be used:
ifconfig_bridge0_ipv6="inet6 auto_linklocal"
- if_bridge(4) now removes IPv6 addresses on a member interface to be
added when the parent interface or one of the existing member
interfaces has an IPv6 address. if_bridge(4) merges each link-local
scope zone which the member interfaces form respectively, so it causes
address scope violation. Removal of the IPv6 addresses prevents it.
- if_lagg(4) now removes IPv6 addresses on a member interfaces
unconditionally.
- Set reasonable flags to non-IPv6-capable interfaces.
At boot time, all of the static routes are installed as before.
The differences are:
- "/etc/rc.d/netif start/stop <if>" now configures static routes
with :<if> if any.
- "/etc/rc.d/routing start/stop <af> <if>" works as well. <af> cannot be
omitted when <if> is specified, but a keyword "any" or "all" can be used
for <af> and <if>.
- ipv6_enable + ipv6_gateway_enable should unset ACCEPT_RTADV by default for
backward compatibility.
- Configurations in ipv6_prefix_IF should be recognized even if there is no
ifconfig_IF_ipv6.
- DAD wait should be performed at once, not on a per-interface basis, if
possible. This fixes an issue that a system with a lot of IPv6-capable
interfaces takes too long for booting.
- Add CIDR notation support like 192.168.1-2.10-16/24 to $ifconfig_IF_aliasN.
This is an extended version of ipv4_addr_IF which supports both IPv4 and
IPv6, and multiple range specifications. To avoid to generate too many
addresses, the maximum number of the generated addresses is currently
limited to 31.
- Add $ifconfig_IF_aliases, which accepts multiple IP aliases in a variable.
- ipv6_prefix_IF now supports !/64 prefix length. In addition to the old
64-bit format (2001:db8:1:1), a full 128-bit format like 2001:db8:1:1::/64
is supported.
- Replace ifconfig command with $IFCONFIG_CMD variable to support
a dry-run mode in the future.
- Remove IP aliases before removing all of IPv4 addresses when doing
"rc.d/netif down".
- Add a DAD wait to network6_getladdr() because it is possible to fail to
configure an EUI64 address when ipv6_prefix_IF is specified.
dim [Thu, 11 Jul 2013 20:09:46 +0000 (20:09 +0000)]
MFC r240966 (by brooks)
Add a COMPILER_FEATURES variable that is set in bsd.compiler.mk.
When the compiler is clang add c++11 to COMPILER_FEATURES.
When the compiler supports c++11, default to building libc++. This will
not impact the compliation of programs, but will make it easier for
users with clang compiled systems to test libc++ by passing -stdlib=libc++
to their c++ compliations.
Enhancement when writing an entire block of a file (from UFS r231313).
This change first attempts the uiomove() to the newly allocated
(and dirty) buffer and only zeros it if the uiomove() fails. The
effect is to eliminate the gratuitous zeroing of the buffer in
the usual case where the uiomove() successfully fills it.
dim [Thu, 11 Jul 2013 06:48:53 +0000 (06:48 +0000)]
MFC r253042:
Pull in r185616 from llvm trunk:
FastISel can only append to basic blocks.
Compute the insertion point from the end of the basic block instead of
skipping labels from the front.
This caused failures in landing pads when live-in copies where inserted
before instruction selection.
I missed this change in r252720; without it, certain compilation flags
can cause exception labels to not be generated, but still referenced,
leading to link errors.
MFC: r252673
A problem with the old NFS client where large writes to large files
would sometimes result in a corrupted file was reported via email.
This problem appears to have been caused by r251719 (reverting
r251719 fixed the problem). Although I have not been able to
reproduce this problem, I suspect it is caused by another thread
increasing np->n_size after the mtx_unlock(&np->n_mtx) but before
the vnode_pager_setsize() call. Since the np->n_mtx mutex serializes
updates to np->n_size, doing the vnode_pager_setsize() with the
mutex locked appears to avoid the problem.
Unfortunately, vnode_pager_setsize() where the new size is smaller,
cannot be called with a mutex held.
This patch returns the semantics to be close to pre-r251719 such that the
call to the vnode_pager_setsize() is only delayed until after the mutex is
unlocked when np->n_size is shrinking. Since the file is growing
when being written, I believe this will fix the corruption.
MFC r253118: Fix conditional (der should match the comment above it).
MFC r253117:
Do two things: First, don't obscure the backtitle. Second, read ~/.dialogrc
if it exists to determine if use_shadow is true (ON) or false (OFF).
The purpose of determining the value of use_shadow is to know how many lines
to subtract from the maximum height value in assuring that the backtitle is
not obscured.
The detriment of obscuring the backtitle is that it provides information
that is not easily obtained elsewhere. That is the command-line shortcut
used to access the current menu. As you navigate from one dialog to the
next, invariably transparently corssing module boundaries, the backtitle
represents the command-line argument used to get there. Obscuring this
information with a widget that is too-tall and/or too-wide would see that
data go unnoticed (leaving few other ways to get that information in the
same helpful context).
So despite the fact that this change reduces the standard maximum height for
all widgets, there is a trap-door to prevent this calculation. If you want
to utilize the full screen height on the terminal (remember, this adjustment
is not made for Xdialog(1)) you can set $NO_BACKTITLE to 1 (or any non-NULL
value for that matter) and this calculation will be skipped. You will be
able to draw a widget that partially obscures the backtitle if-necessary.
Direct commit to stable branch to fix ABI breakage:
In r227839, when removing libkvm dependency on procfs(5),
kvm_uread() function, used for reading from /proc/pid/mem, was
removed too. This change broke ABI.
Bring kvm_uread() back so that the ABI is kept. In head soname is
going to be bumped instead.
Reported by: rmh
Discussed on: arch
Suggested by: jilles