kan [Wed, 9 Oct 2013 21:33:19 +0000 (21:33 +0000)]
Unbreak zfsloader with LOADER_TFTP_SUPPORT on
Only accept 'net' and 'pxe' devices as underlying transport
in tftp.c on x86. Prior to this change tftp code would attempt
to send packets over any boot device, including zfs one with
predictably sad results.
glebius [Wed, 9 Oct 2013 19:04:40 +0000 (19:04 +0000)]
There are some high performance NICs that count statistics in hardware,
and there are ifnets, that do that via counter(9). Provide a flag that
would skip cache line trashing '+=' operation in ether_input().
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
Reviewed by: melifaro, adrian
Approved by: re (marius)
kib [Wed, 9 Oct 2013 18:45:01 +0000 (18:45 +0000)]
The device vnodes are often unlocked when bread() or bwrite() is
called. This probably should be fixed eventually, but for now it is
not needed to try to flush such vnodes from the buffer allocation
context.
Reported and tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Approved by: re (gjb)
kib [Wed, 9 Oct 2013 18:43:29 +0000 (18:43 +0000)]
Do not flush buffers when the v_object of the passed vnode does not
really belong to it. Such vnodes, with the pointers to other vnodes
v_objects, are typically instantiated by the bypass filesystems.
Invalidating mappings of other vnode pages and the pages is wrong,
since reclamation of the upper vnode does not imply that lower vnode
is reclaimed too.
One of the consequences of the improper reclamation was destruction of
the wired mappings of the lower vnode pages, triggering miscellaneous
assertions in the VM system.
Reported by: John Marshall <john.marshall@riverwillow.com.au>
Tested by: John Marshall <john.marshall@riverwillow.com.au>, pho
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Approved by: re (gjb)
kib [Wed, 9 Oct 2013 18:41:35 +0000 (18:41 +0000)]
When growing the file descriptor table, new larger memory chunk is
allocated, but the old table is kept around to handle the case of
threads still performing unlocked accesses to it.
Grow the table exponentially instead of increasing its size by
sizeof(long) * 8 chunks when overflowing. This mode significantly
reduces the total memory use for the processes consuming large numbers
of the file descriptors which open them one by one.
Reported and tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Approved by: re (marius)
jfv [Wed, 9 Oct 2013 17:32:52 +0000 (17:32 +0000)]
Update the Intel igb driver to version 2.4.0
- This version has support for the new Intel Avoton systems,
including 2.5Gb support, further it now has IPv6/TSO6 support as
well. Shared code has been updated where necessary as well. Thanks
to my new assistant Eric Joyner for doing the transmit path changes
to bring in the IPv6/TSO6 support. Thanks to Gleb for catching the
one bug and change needed in NETMAP.
mav [Wed, 9 Oct 2013 12:09:01 +0000 (12:09 +0000)]
Close the race on path ID allocation in xpt_bus_register() if two buses are
registered simultaneously. Due to topology unlock between the ID allocation
and the bus registration there is a chance that two buses may get the
same IDs. That is supposed reason of lock assertion panic in CAM during
initial bus scanning after new iscsid initiates two sessions same time.
trasz [Wed, 9 Oct 2013 12:03:04 +0000 (12:03 +0000)]
Fix NOP-In/NOP-Out payload handling. Previous way didn't work at all; fortunately
nothing seems to actually use this feature, but it's required by standard.
Approved by: re (glebius)
Sponsored by: FreeBSD Foundation
glebius [Wed, 9 Oct 2013 12:00:38 +0000 (12:00 +0000)]
When processing ACK in tcp_do_segment, use sbcut_locked() instead of
sbdrop_locked() to cut acked mbufs from the socket buffer. Free this
chain a batch manner after the socket buffer lock is dropped.
This measurably reduces contention on socket buffer.
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
Approved by: re (marius)
glebius [Wed, 9 Oct 2013 11:57:53 +0000 (11:57 +0000)]
- Substitute sbdrop_internal() with sbcut_internal(). The latter doesn't free
mbufs, but return chain of free mbufs to a caller. Caller can either reuse
them or return to allocator in a batch manner.
- Implement sbdrop()/sbdrop_locked() as a wrapper around sbcut_internal().
- Expose sbcut_locked() for outside usage.
Sponsored by: Netflix
Sponsored by: Nginx, Inc.
Approved by: re (marius)
dteske [Wed, 9 Oct 2013 08:12:26 +0000 (08:12 +0000)]
Introduce centralized [X]dialog(1) exit codes to `dialog.subr' and make-use
throughout the bsdconfig(8) code. While we're here, add an explicit argument
to lvalue-seeking invocations of "return" that previously had no argument.
Also, consolidate a single instance of double-newline and remove some
comments that are no longer required (given increased readability with new
exit codes).
adrian [Wed, 9 Oct 2013 00:21:21 +0000 (00:21 +0000)]
Add two new MIPS CPU families - mips24k and mips74k.
They're both different cores:
* mips24k is an 8-stage pipeline, mips32r1 ABI, non-superscalar core.
* mips74k is a dual-issue 15-stage superscalar design, mips32r2 ABI.
They have different sets of quirks and bugs; these #define entries
will be used to work around these.
Now, strictly speaking, we should have CPU ABI families (mips32r1, mips32r2,
etc) and CPU core types (mips4k, mips24k, mips74k, etc.) But this is the
starting point of that particular tidy-up.
dteske [Tue, 8 Oct 2013 20:39:38 +0000 (20:39 +0000)]
Remove mention of [fixed] known issue not-specific to updating.
Vendor update to [c]dialog-1.2-20130925 has fixed known problems.
Vendor update SVN revisions: 255917-255917, 255852, and 255958.
dim [Tue, 8 Oct 2013 19:39:21 +0000 (19:39 +0000)]
In usr.sbin/bhyve/pci_ahci.c, fix several gcc warnings of the form
"assignment makes pointer from integer without a cast", by changing the
cmd_lst and rbis members of struct ahci_port from integers to pointers.
Also surround a pow-of-2 test expression with parentheses to clarify it,
and avoid another gcc warning.
Approved by: re (glebius)
Reviewed by: grehan, mav
dim [Tue, 8 Oct 2013 17:26:28 +0000 (17:26 +0000)]
Now our binutils's assembler supports the Intel Random Number Generator
extensions, we can change the .byte directives in sys/dev/random/ivy.c
to plain 'rdrand' mnemonics. This already worked for clang users, but
now it will also work for gcc users.
Approved by: re (kib)
Approved by: so (des)
MFC after: 1 week
neel [Tue, 8 Oct 2013 16:36:17 +0000 (16:36 +0000)]
Change the behavior of bhyve such that the gdb listening port is opt-in
rather than opt-out.
Prior to this change if the "-g" option was not specified then a listening
socket for tunneling gdb packets would be opened at port 6466. If a second
virtual machine is fired up, also without the "-g" option, then that would
fail because there is already a listener on port 6466.
After this change if a gdb tunnel port needs to be created it needs to be
explicitly specified with a "-g <portnum>" command line option.
jimharris [Tue, 8 Oct 2013 16:00:12 +0000 (16:00 +0000)]
Log and then disable asynchronous notification of persistent events after
they occur.
This prevents repeated notifications of the same event.
Status of these events may be viewed at any time by viewing the
SMART/Health Info Page using nvmecontrol, whether or not asynchronous
events notifications for those events are enabled. This log page can
be viewed using:
nvmecontrol logpage -p 2 <ctrlr id>
Future enhancements may re-enable these notifications on a periodic basis
so that if the notified condition persists, it will continue to be logged.
Sponsored by: Intel
Reviewed by: carl
Approved by: re (hrs)
MFC after: 1 week
markj [Tue, 8 Oct 2013 12:56:46 +0000 (12:56 +0000)]
Initialize and free the DTrace taskqueue in the dtrace module load/unload
handlers rather than in the dtrace device open/close methods. The current
approach can cause a panic if the device is closed which the taskqueue
thread is active, or if a kernel module containing a provider is unloaded
while retained enablings are present and the dtrace device isn't opened.
adrian [Tue, 8 Oct 2013 11:28:59 +0000 (11:28 +0000)]
Add channel survey support to the AR5212 HAL.
The AR5212 series of MACs implement the same channel counters as the
later 11n chips - except, of course, the 11n specific counter (extension
channel busy.)
This allows users of these NICs to use 'athsurvey' to see how busy their
current channel is.
des [Mon, 7 Oct 2013 10:26:38 +0000 (10:26 +0000)]
Introduce the /libexec/freebsd-version script, which is intended to be
used by auditing tools to determine the userland patch level when it
differs from what `uname -r` reports. This can happen when the system
is kept up-to-date using freebsd-update and the last SA did not touch
the kernel, or when a new kernel has been installed but the system has
not yet rebooted.
gibbs [Sat, 5 Oct 2013 23:11:01 +0000 (23:11 +0000)]
Formalize the concept of virtual CPU ids by adding a per-cpu vcpu_id
field. Perform vcpu enumeration for Xen PV and HVM environments
and convert all Xen drivers to use vcpu_id instead of a hard coded
assumption of the mapping algorithm (acpi or apic ID) in use.
amd64/include/pcpu.h:
i386/include/pcpu.h:
Add vcpu_id to the amd64 and i386 pcpu structures.
dev/xen/timer/timer.c
x86/xen/xen_intr.c
Use new vcpu_id instead of assuming acpi_id == vcpu_id.
i386/xen/mp_machdep.c:
i386/xen/mptable.c
x86/xen/hvm.c:
Perform Xen HVM and Xen full PV vcpu_id mapping.
x86/xen/hvm.c:
x86/acpica/madt.c
Change SYSINIT ordering of acpi CPU enumeration so that it
is guaranteed to be available at the time of Xen HVM vcpu
id mapping.
neel [Sat, 5 Oct 2013 21:22:35 +0000 (21:22 +0000)]
Merge projects/bhyve_npt_pmap into head.
Make the amd64/pmap code aware of nested page table mappings used by bhyve
guests. This allows bhyve to associate each guest with its own vmspace and
deal with nested page faults in the context of that vmspace. This also
enables features like accessed/dirty bit tracking, swapping to disk and
transparent superpage promotions of guest memory.
Guest vmspace:
Each bhyve guest has a unique vmspace to represent the physical memory
allocated to the guest. Each memory segment allocated by the guest is
mapped into the guest's address space via the 'vmspace->vm_map' and is
backed by an object of type OBJT_DEFAULT.
pmap types:
The amd64/pmap now understands two types of pmaps: PT_X86 and PT_EPT.
The PT_X86 pmap type is used by the vmspace associated with the host kernel
as well as user processes executing on the host. The PT_EPT pmap is used by
the vmspace associated with a bhyve guest.
Page Table Entries:
The EPT page table entries as mostly similar in functionality to regular
page table entries although there are some differences in terms of what
bits are used to express that functionality. For e.g. the dirty bit is
represented by bit 9 in the nested PTE as opposed to bit 6 in the regular
x86 PTE. Therefore the bitmask representing the dirty bit is now computed
at runtime based on the type of the pmap. Thus PG_M that was previously a
macro now becomes a local variable that is initialized at runtime using
'pmap_modified_bit(pmap)'.
An additional wrinkle associated with EPT mappings is that older Intel
processors don't have hardware support for tracking accessed/dirty bits in
the PTE. This means that the amd64/pmap code needs to emulate these bits to
provide proper accounting to the VM subsystem. This is achieved by using
the following mapping for EPT entries that need emulation of A/D bits:
Bit Position Interpreted By
PG_V 52 software (accessed bit emulation handler)
PG_RW 53 software (dirty bit emulation handler)
PG_A 0 hardware (aka EPT_PG_RD)
PG_M 1 hardware (aka EPT_PG_WR)
The idea to use the mapping listed above for A/D bit emulation came from
Alan Cox (alc@).
The final difference with respect to x86 PTEs is that some EPT implementations
do not support superpage mappings. This is recorded in the 'pm_flags' field
of the pmap.
TLB invalidation:
The amd64/pmap code has a number of ways to do invalidation of mappings
that may be cached in the TLB: single page, multiple pages in a range or the
entire TLB. All of these funnel into a single EPT invalidation routine called
'pmap_invalidate_ept()'. This routine bumps up the EPT generation number and
sends an IPI to the host cpus that are executing the guest's vcpus. On a
subsequent entry into the guest it will detect that the EPT has changed and
invalidate the mappings from the TLB.
Guest memory access:
Since the guest memory is no longer wired we need to hold the host physical
page that backs the guest physical page before we can access it. The helper
functions 'vm_gpa_hold()/vm_gpa_release()' are available for this purpose.
PCI passthru:
Guest's with PCI passthru devices will wire the entire guest physical address
space. The MMIO BAR associated with the passthru device is backed by a
vm_object of type OBJT_SG. An IOMMU domain is created only for guest's that
have one or more PCI passthru devices attached to them.
Limitations:
There isn't a way to map a guest physical page without execute permissions.
This is because the amd64/pmap code interprets the guest physical mappings as
user mappings since they are numerically below VM_MAXUSER_ADDRESS. Since PG_U
shares the same bit position as EPT_PG_EXECUTE all guest mappings become
automatically executable.
Thanks to Alan Cox and Konstantin Belousov for their rigorous code reviews
as well as their support and encouragement.
Thanks for John Baldwin for reviewing the use of OBJT_SG as the backing
object for pci passthru mmio regions.
Special thanks to Peter Holm for testing the patch on short notice.
gibbs [Sat, 5 Oct 2013 19:51:09 +0000 (19:51 +0000)]
Correct panic caused by attaching both Xen PV and HyperV virtualization
aware drivers on Xen hypervisors that advertise support for some
HyperV features.
x86/xen/hvm.c:
When running in HVM mode on a Xen hypervisor, set vm_guest
to VM_GUEST_XEN so other virtualization aware components in
the FreeBSD kernel can detect this mode is active.
dev/hyperv/vmbus/hv_hv.c:
Use vm_guest to ignore Xen's HyperV emulation when Xen is
detected and Xen PV drivers are active.
alc [Sat, 5 Oct 2013 18:53:03 +0000 (18:53 +0000)]
Tidy up kmeminit(): Since r245575, 'nmbclusters' is calculated after
kmeminit() runs, so it contributes nothing to 'vm_kmem_size'; update a
comment to reflect that r254025 replaced the kmem submap with the kmem
arena.
grehan [Fri, 4 Oct 2013 23:29:07 +0000 (23:29 +0000)]
Remove obsolete cmd-line options and code associated with
these.
The mux-vcpus option may return at some point, given it's utility
in finding bhyve (and FreeBSD) bugs.