Make sure the CCB xflags field is initialized to zero so that
CAM_EXTLUN_VALID is not erroneously set. Also add an XPORT_SRP
identifier to the known SCSI transports for the SCSI RDMA protocol, as
used, for example with Infiniband storage.
Rework handling of ofw_quiesce(), making it the responsibility of the
platform modules. Whether to call this function or not is highly machine
dependent: on some systems, it is required, while on others it breaks
everything. Platform modules are in a better position to figure this
out. This is required for POWER hypervisor SCSI to work correctly. There
are no functional changes on Powermac systems.
Make sure that ref and changed bits propagate back to the VM layer
whenever pages are unmapped. The old code had several races that could
allow these to become stale.
Allow Open Firmware syscons to attach to devices without an "address"
property such as those found on some real and emulated IBM systems. The
approach, which is taken from Linux, is to scan through the PCI bars
until we find one large enough to contain the linear framebuffer and
which is ideally prefetchable if no "address" property can be found.
This makes the graphical console work with the pSeries target in QEMU.
Fix bug where UART unit number was not set properly, which prevented
operation on systems with multiple serial ports. Also turn on
interrupts for the UART device, which were disabled due to a
now-fixed bug in QEMU.
Acquire a hold reference on the vnode when a knote is instantiated.
Otherwise, knote keeps a pointer to a vnode which could become invalid
any time.
Reported by: many
Tested by: Patrick Lamaiziere <patfbsd@davenulle.org>
Discussed with: jmg
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Approved by: re (marius)
Make the callout arithmetic more robust adding checks for overflow.
Without these, if the timeout value passed is "large enough", the
value of the sum of it and other factors (e.g. current time as
returned by sbinuptime() or 'precision' argument) might result in a
negative number. This negative number is then passed to
eventtimers(4), which causes et_start() routine to load et_min_period
into eventtimer, making the CPU where the thread is stuck forever in
timer interrupt handler routine. This is now avoided rounding to
INT64_MAX the timeout period in case of overflow.
andrew [Thu, 26 Sep 2013 07:53:18 +0000 (07:53 +0000)]
Add an elf note on ARM to store the MACHINE_ARCH an executable was built
for. This is useful for software needing to know which architecture a
binary is built for as arm and armv6 have slight differences meaning only
some binaries build for one will work as expected on the other. It is
expected pkgng will be able to make use of this to simplify the logic to
determine which package ABI to use.
Correct a NULL pointer deference in nslookup and nsupdate that would
cause the utility to crash in interactive mode when the user gives
an EOF on standard input.
Re-do r255853. Along with adding back the API/ABI changes from the
original, this hides the contents of cam_compat.h from ktrace/kdump/truss,
avoiding problems there. There are no user-servicable parts in there, so
no need for those tools to be groping around in there.
Re-do r255853. Along with adding back the API/ABI changes from the
original, this hides the contents of cam_compat.h from ktrace/kdump/truss,
avoiding problems there. There are no user-servicable parts in there, so
no need for those tools to be groping around in there.
Avoid memory accesses reordering which can result in fget_unlocked()
seeing a stale fd_ofiles table once fd_nfiles is already updated,
resulting in OOB accesses.
Approved by: re (kib)
Sponsored by: EMC / Isilon storage division
Reported and tested by: pho
Reviewed by: benno
fix a bug where we access a bread buffer after we have brelse'd it...
The kernel normally didn't unmap/context switch away before we accessed
the buffer most of the time, but under heavy I/O pressure and lots of
mount/unmounting this would cause a fault on nofault panic...
Reviewed by: dteske
Approved by: re (kib)
Sponsored by: Vicor
MFC after: 3 days
des [Tue, 24 Sep 2013 20:05:16 +0000 (20:05 +0000)]
Introduce a kern.geom.notaste sysctl that can be used to temporarily
disable GEOM tasting to avoid the "bouncing GEOM" problem where, when
you shut down the consumer of a provider which can be viewed in multiple
ways (typically a mirror whose members are labeled partitions), GEOM
will immediately taste that provider's alter ego and reattach the
consumer.
- Remove the timeout_ch field. It's been deprecated since FreeBSD 7.0;
MPSAFE drivers should be managing their own timeout storage. The
remaining non-MPSAFE drivers have been modified to also manage their own
storage, and should be considered for updating to MPSAFE (or removal)
during the FreeBSD 10.x lifecycle.
- Add fields related to soft timeouts and quality of service, to be used
in upcoming work.
- Add room for more flags in the CCB header and path_inq structures.
- Begin support for extended 64-bit LUNs.
- Bump the CAM version number to 0x18, but add compat shims. Tested with
camcontrol and smartctl.
In pmap_clear_modify(), initialize pvh even for fictitious managed
page, otherwise the small mappings loop would use uninitialized value.
Note that currently pmap_clear_modify() is not called for fictitious
pages.
Sponsored by: The FreeBSD Foundation
Approved by: re (glebius)
Make load average sampling asynchronous to hardclock ticks. This improves
measurement of load caused by time-related events still using hardclock.
For example, without this change dummynet, scheduling events each hardclock
tick, was always miscounted as load of 1.
There is still aliasing with events delayed by the new precision mechanism,
but it probably can't be avoided without moving this sampling from using
callout to some lower-level code or handling it in some other special way.
des [Mon, 23 Sep 2013 20:35:54 +0000 (20:35 +0000)]
Unbreak the WITHOUT_KERBEROS build and try to reduce the odds of a
repeat performance by introducing a script that runs configure with and
without Kerberos, diffs the result and generates krb5_config.h, which
contains the preprocessor macros that need to be defined in the Kerberos
case and undefined otherwise.
des [Mon, 23 Sep 2013 20:06:59 +0000 (20:06 +0000)]
Prevent resolvconf from updating /etc/resolv.conf. As Jakob Schlyter
pointed out, having additional nameservers listed in /etc/resolv.conf
can break DNSSEC verification by providing a false positive if unbound
returns SERVFAIL due to an invalid signature. The downside is that
the domain / search path won't get updated either, but we can live
with that.
Fix a bug in HTTP checking/fetching.
Fix a bug in HTTP checking/fetching. Add Main Site to HTTP menu. Add new
example script browse_packages_http.sh and move existing example script
browse_packages.sh -> browse_packages_ftp.sh
Reviewed by: gjb, brd
Approved by: re (gjb), clusteradm (brd)
MFC after: 3 days
Add installer support for CHRP/PAPR PowerPC systems that use MBR+BSD
formatting, like x86, but with an additional MBR slice containing a raw
boot partition.
Prevent "lock (iscsi) sx does not match earlier (sleep mutex) lock" panic
triggered by "kldload iscsi_initiator; kldunload iscsi_initiator; kldload
iscsi".
des [Mon, 23 Sep 2013 04:36:51 +0000 (04:36 +0000)]
Add a setup script for unbound(8) called local-unbound-setup. It
generates a configuration suitable for running unbound as a caching
forwarding resolver, and configures resolvconf(8) to update unbound's
list of forwarders in addition to /etc/resolv.conf. The initial list
is taken from the existing resolv.conf, which is rewritten to point to
localhost. Alternatively, a list of forwarders can be provided on the
command line.
To assist this script, add an rc.subr command called "enabled" which
does nothing except return 0 if the service is enabled and 1 if it is
not, without going through the usual checks. We should consider doing
the same for "status", which is currently pointless.
Add an rc script for unbound, called local_unbound. If there is no
configuration file, the rc script runs local-unbound-setup to generate
one.
Note that these scripts place the unbound configuration files in
/var/unbound rather than /etc/unbound. This is necessary so that
unbound can reload its configuration while chrooted. We should
probably provide symlinks in /etc.
ian [Mon, 23 Sep 2013 00:04:36 +0000 (00:04 +0000)]
Launch the bourne shell using "sh" rather than "${SHELL}", as the latter
may come in from the environment and reflect the user's interactive shell.
Using bare "sh" is the dominant pattern in existing makefiles.
dim [Sun, 22 Sep 2013 22:03:30 +0000 (22:03 +0000)]
Pull in r191165 from upstream llvm trunk:
ISelDAG: spot chain cycles involving MachineNodes
Previously, the DAGISel function WalkChainUsers was spotting that it
had entered already-selected territory by whether a node was a
MachineNode (amongst other things). Since it's fairly common practice
to insert MachineNodes during ISelLowering, this was not the correct
check.
Looking around, it seems that other nodes get their NodeId set to -1
upon selection, so this makes sure the same thing happens to all
MachineNodes and uses that characteristic to determine whether we
should stop looking for a loop during selection.
This should fix PR15840.
Specifically, this fixes the long-standing assertion failure when
compiling the multimedia/gstreamer port on i386. Thanks to Tijl
Coosemans for his help in getting upstream to fix it.
Pre-acquire the filedesc sx when a possibility exists that the later
code could need to remove a kqueue from the filedesc list. Global
lock is already locked, which causes sleepable after non-sleepable
lock acquisition.
Reported and tested by: pho
Reviewed by: jmg
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Approved by: re (gjb)
Increase the chance of the buffer write from the bufdaemon helper
context to succeed. If the locked vnode which owns the buffer to be
written is shared locked, try the non-blocking upgrade of the lock to
exclusive.
PR: kern/178997
Reported and tested by: Klaus Weber <fbsd-bugs-2013-1@unix-admin.de>
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Approved by: re (marius)
ian [Sun, 22 Sep 2013 19:15:24 +0000 (19:15 +0000)]
Launch the shell, passing it the path to the mk-osreldate script, rather
than launching the script directly and relying on #! to launch the shell.
This avoids problems when the source is mounted with the noexec flag.
des [Sun, 22 Sep 2013 19:09:43 +0000 (19:09 +0000)]
Due to a missing command-line argument, yacc regenerated the parser but
not its header file, resulting in a mismatch between the lexer and parser
and strange errors when reading the configuration file.
Consistently use the same value to indicate exclusively-held and
shared-held locks for all the primitives in lc_lock/lc_unlock routines.
This fixes the problems introduced in r255747, which indeed introduced an
inversion in the logic.
Reported by: many
Tested by: bdrewery, pho, lme, Adam McDougall, O. Hartmann
Approved by: re (glebius)
- Create kern.ipc.sendfile namespace, and put the new "readhead" OID
there as "kern.ipc.sendfile.readahead".
- Push all nsfbuf related tunables into MD code. Don't move them
to new namespace in favor of POLA.
andrew [Sun, 22 Sep 2013 07:30:17 +0000 (07:30 +0000)]
Remove the armv6eb architecture as it is unused, and almost certainly
broken. None of our kernels can boot armv6eb. The little-endian kernels do
not have the required code to be able to switch endian when running a
big-endian executable.
Fix ia64 and mips kernel builds due to XENHVM=>GENERIC integration in
revision 255744.
sys/kern/subr_smp.c:
IPI_SUSPEND is only available on amd64 and i386. Protect
new uses of this constant with #ifdefs to avoid impacting
other platforms.
Give argtype struct names a different prefix than probe struct names.
Otherwise it's possible to declare SDT probes in such a way that a name
collision occurs, causing an unexpected compilation error.
ian [Sat, 21 Sep 2013 22:36:07 +0000 (22:36 +0000)]
Create a separate script to generate osreldate.h rather than sourcing
newvers.sh into a temporary subshell with inline make rules.
Using a separate script fixes a variety of problems, including establishing
the correct dependencies in the makefiles. It also eliminates a problem
with the way newvers.sh uses `realpath $0`, because $0 expands differently
within a script sourced into a rule in a makefile depending on the version
of make and of /bin/sh being used. The latter can cause build breakage in a
cross-build environment, and can also make it difficult to compile 10.0 on
older pre-10.0 systems.
PR: 160646 174422
Submitted by: Garrett Cooper <yaneurabeya@gmail.com>
Approved by: re (gjb)
MFC after: 2 weeks
Stability fixes for Intel LynxPoint XHCI controllers. Disable XHCI port
routing if we get certain errors. Poll for command completion upon
command timeouts. The XHCI error events might not generate interrupts.
MFC after: 1 week
Reported by: Daniel Gerzo <danger@rulez.sk>, Antonis Anastasiadis <anastasiadis@datalive.gr>
PR: usb/181159
Approved by: re (gjb)
des [Sat, 21 Sep 2013 21:34:22 +0000 (21:34 +0000)]
Ditch the random seeding code, which never really worked as intended.
Add config variables to enable / disable individual host key algorithms.
Clean up the host key generation code.