dim [Tue, 24 Jun 2014 06:12:11 +0000 (06:12 +0000)]
MFC r267704:
Pull in r211435 from upstream llvm trunk (by Benjamin Kramer):
Legalizer: Add support for splitting insert_subvectors.
We handle this by spilling the whole thing to the stack and doing the
insertion as a store.
PR19492. This happens in real code because the vectorizer creates
v2i128 when AVX is enabled.
This fixes a "fatal error: error in backend: Do not know how to split
the result of this operator!" message encountered during compilation of
the net-p2p/libtorrent-rasterbar port.
rodrigc [Mon, 23 Jun 2014 22:37:49 +0000 (22:37 +0000)]
MFC r263795:
Strict value checking will cause problem.
Bay trail DN2820FYKH is supported on Linux but does not work on FreeBSD.
This behaviour is bug-compatible with Linux-3.13.5.
rodrigc [Mon, 23 Jun 2014 22:31:28 +0000 (22:31 +0000)]
MFC r263795:
Strict value checking will cause problem.
Bay trail DN2820FYKH is supported on Linux but does not work on FreeBSD.
This behaviour is bug-compatible with Linux-3.13.5.
marius [Mon, 23 Jun 2014 08:32:36 +0000 (08:32 +0000)]
MFC: r267638
Don't denounce peripherals on system shutdown. Together with r267321
(MFCed to stable/10 in r267775), we're now back to the pre-r228483
level of default verbosity. This in turn again typically allows for
reading information that userland might have printed on the screen
before initiating a halt, but still permits to debug potential device
shutdown problems on system shutdown via CAM_DEBUG etc.
Reviewed by: mav
Sponsored by: Bally Wulff Games & Entertainment GmbH
marius [Mon, 23 Jun 2014 08:27:15 +0000 (08:27 +0000)]
MFC: r267321
Avoid the USB device disconnected and controller shutdown clutter on system
shutdown by putting the former under !rebooting and turning the latter into
debug messages.
Reviewed by: hps
Sponsored by: Bally Wulff Games & Entertainment GmbH
tuexen [Mon, 23 Jun 2014 07:54:12 +0000 (07:54 +0000)]
MFC r267674:
Honor jails for unbound SCTP sockets when selecting source addresses,
reporting IP-addresses to the peer during the handshake, adding
addresses to the host, reporting the addresses via the sysctl
interface (used by netstat, for example) and reporting the
addresses to the application via socket options.
This issue was reported by Bernd Walter.
np [Mon, 23 Jun 2014 05:39:10 +0000 (05:39 +0000)]
MFC r267689:
Consider the total number of descriptors available (and not just those
that are ready to be reclaimed) when deciding whether to resume tx after
a stall.
mav [Sun, 22 Jun 2014 21:18:24 +0000 (21:18 +0000)]
MFC r267387:
Allocating new bucket for bucket zone, never take it from the zone itself,
since it will almost certanly fail. Take next bigger zone instead.
This situation should not happen with original bucket zones configuration:
"32 Bucket" zone uses "64 Bucket" and vice versa. But if "64 Bucket" zone
lock is congested, zone may grow its bucket size and start biting itself.
mav [Sun, 22 Jun 2014 21:16:18 +0000 (21:16 +0000)]
MFC r267392:
Implement simple direct-mapped cache for popular filesystem identifiers to
avoid congestion on global mountlist_mtx mutex in vfs_busyfs(), while
traversing through the list of mount points.
This change significantly improves NFS server scalability, since it had
to do this translation for every request, and the global lock becomes quite
congested.
This code is more optimized for relatively small number of mount points.
On systems with hundreds of active mount points this simple cache may have
many collisions. But the original traversal code in that case should also
behave much worse, so we are not loosing much.
pfg [Sun, 22 Jun 2014 20:24:17 +0000 (20:24 +0000)]
MFC r267490:
patch: unsign the line length to avoid overflows.
Patch(1) uses a short int for the line length, which is usually
sufficient for regular diffs, but makes no effort to signal
when there is an overflow.
Change the line length to an unsigned short int to better use
the fact that a length is never negative. The change is loosely
inspired on a related change in DragonFly, but we avoid spending
more memory than necessary.
While here adjust the messages to be clearer on what is happening.
mav [Sun, 22 Jun 2014 18:06:11 +0000 (18:06 +0000)]
MFC r267232, r267239:
Use atomics to modify numvnodes variable.
This allows to mostly avoid lock usage in getnewvnode_[drop_]reserve(),
that reduces number of global vnode_free_list_mtx mutex acquisitions
from 4 to 2 per NFS request on ZFS, improving SMP scalability.
mav [Sun, 22 Jun 2014 18:03:53 +0000 (18:03 +0000)]
MFC r267228:
Split RPC pool threads into number of smaller semi-isolated groups.
Old design with unified thread pool was good from the point of thread
utilization. But single pool-wide mutex became huge congestion point
for systems with many CPUs. To reduce the congestion create several
thread groups within a pool (one group for every 6 CPUs and 12 threads),
each group with own mutex. Each connection during its registration is
assigned to one of the groups in round-robin fashion. File affinify
code may still move requests between the groups, but otherwise groups
are self-contained.
tuexen [Sun, 22 Jun 2014 16:54:59 +0000 (16:54 +0000)]
MFC r265691:
For some UDP packets (for example with 200 byte payload) and IP options,
the IP header and the UDP header are not in the same mbuf.
Add code to in_delayed_cksum() to deal with this case.
tuexen [Sun, 22 Jun 2014 16:50:46 +0000 (16:50 +0000)]
MFC r265455:
Remove unused code. This is triggered by the bugreport of Sylvestre Ledru
which deal with useless code in the user land stack:
https://bugzilla.mozilla.org/show_bug.cgi?id=1003929
tuexen [Sun, 22 Jun 2014 16:45:42 +0000 (16:45 +0000)]
MFC r264838:
Don't free an mbuf twice. This only happens in very rare error
cases where the peer sends illegal sequencing information in
DATA chunks for an existing association.
tuexen [Sun, 22 Jun 2014 16:32:41 +0000 (16:32 +0000)]
MFC r263922:
Handle an edge case of address management similar to TCP.
This needs to be reconsidered when the address handling
will be reimplemented.
The patch is from rrs@.
tuexen [Sun, 22 Jun 2014 16:28:52 +0000 (16:28 +0000)]
MFC r263237:
* Provide information in error causes in ASCII instead of
proprietary binary format.
* Add support for a diagnostic information error cause.
The code is sysctlable and the default is 0, which
means it is not sent.
tuexen [Sun, 22 Jun 2014 16:26:16 +0000 (16:26 +0000)]
MFC r263096:
Put the offset of the CRC32C in csum_data instead of 0.
The virtio driver needs the offset to be stored in csum_data,
like in the case for UDP and TCP.
The virtio problem was reported by
Niu Zhixiong <kaiaixi@gmail.com>, who helped in debugging
and testing the patch.
pfg [Sat, 21 Jun 2014 01:48:45 +0000 (01:48 +0000)]
MFC r267426, r267464:
Avoid zeroing during allocation.
This change reverts a change from OpenBSD which made use of
calloc, and therefore wasted time initializing arrays that
will later be realloc'ed.
Consistently use FreeBSD's reallocf():
- Drop some bogus casts to size_t.
- The new_p_foo variables are not needed anymore.
Also merge the changes from OpenBSD's manpage
patch.1 Rev 1.27:
"patch was moved from user portability (UP) to base in issue 7
and is no longer optional"
pfg [Fri, 20 Jun 2014 15:43:58 +0000 (15:43 +0000)]
MFC r267544:
Update license to strptime(3) implementation.
Our strptime(3) implementation was the base for the illumos
implementation and after contacting the author, Kevin Rudy
stated the code is under a 2-Clause BSD License [1]
After reviewing our local changes to the file in question,
the FreeBSD Foundation has agreed that their contributions
to this file are not required to carry clause 3 or 4 so
the file can be relicensed as in Illumos [2].
tijl [Fri, 20 Jun 2014 07:32:03 +0000 (07:32 +0000)]
MFC r267436-267439:
- Replace malloc+memset with calloc.
- iconv_open(3): initialise ci_ilseq_invalid field of _citrus_iconv_shared
struct after allocation with malloc.
- iconvlist(3): reduce a memory leak by copying strings only once.
- iconv(1):
- Make invalids variable local to do_conv such that it prints the number
of invalid characters of the current file instead of an accumulated
value.
- Make do_conv return an error when invalid characters have been found.
Return EXIT_FAILURE from main if any file contained invalid characters.
This matches the behaviour of GNU iconv.
- Mark usage with __dead2 attribute.
- Make the long_options array const.
neel [Tue, 17 Jun 2014 21:49:03 +0000 (21:49 +0000)]
MFC r266901
Allocate a zeroed LDT.
Failing to do this might result in the LDT appearing to run out of free
descriptors because of random junk in the descriptor's 'sd_type' field.
mav [Tue, 17 Jun 2014 08:09:40 +0000 (08:09 +0000)]
MFC r267029, r267038:
Replace gethrtime() with cpu_ticks(), as source of random for the taskqueue
selection. gethrtime() in our port updated with HZ rate, so unusable for
this specific purpose, completely draining benefit of multiple taskqueues.
ray [Mon, 16 Jun 2014 11:51:00 +0000 (11:51 +0000)]
MFC: 266836, 266839
vt(4) support for vidcontrol(1).
o Teach vidcontrol(1) how to load vt(4) font.
o Teach vidcontrol(1) to distinct which virtual terminal system is running now.
o Load vt(4) fonts from different location.
o Add $FreeBSD$ tag for path.h.
vt(4) support for kbdcontrol(1).
Enable kbdcontrol(1) to use maps from vt(4) keymaps dir /usr/share/vt/keymaps
if vt(4) is present.
265391
Define a new method for probing vt(4) driver before attach it at early stage.
265392
Create dataset for vt(4) drivers.
265395
Set of updates to vt(4) core part.
o Declare vt(4) drivers dataset.
o Create single static structures for all early drivers.
o Add vt(4) to be by default in the kernel consoles list.
o Create one more sysinit point, to be able to initialize memory and lock
requirement of early drivers.
o Implement early drivers select. (Only best available will be selected).
o Fix one missed "return (0)" for VTYLOCK.
o Improve locking for cases when one driver replace another.
o Make driver replacement notification less debug-look-like.
o Minor spell fixes.
265397
Switch fb and efifb drivers to use names and new vt(4) driver probe method.
265398
Add vt(4) driver name for ofwfb driver.
265402
Revert r264997 and r265026. It is not required anymore.
265403
Switch vga drivers to use names and new vt(4) driver probe method.
265442
Implement KDMKTONE ioctl.
265546
Fix possible divide by zero.
265680
No need to assign fields required and checked on probe.
265681
Fix scrollback.
265719
Hide debug messages under VT_DEBUG.
265927
Update terminal sizes in any case when new vt(4) driver arrive.
(Plus remove one unused newline)
266010
Remove extra newlines.
No functional changes.
266495
Fix tty locking.
o Correct expected values for VT_LOCKSWITCH ioctl.
o Check current window for locked state.
266540
Proper fix of VT_LOCKSWITCH ioctl.
266835
Remove driver as unused.
267007
Fix case when vt(4) started w/o driver assigned.
o Always init locks and cv ASAP.
o Initialize driver-independent parts even if driver probing fail.
o Allow to call vt_upgrade anytime, for later loaded drivers.
o New window flag VWF_READY, to track if window already initialized.
Other updates:
o Pass vd as a cookie for kbd_allocate.
o Do not blank window on driver replacement.
kib [Sun, 15 Jun 2014 05:15:38 +0000 (05:15 +0000)]
MFC r267226:
Initialize the pbuf counter for directio using SYSINIT.
Mark ffs_rawread.c as requiring both ffs and directio options to be
compiled into the kernel. Add ffs_rawread.c to the list of ufs.ko
module' sources.
dim [Sat, 14 Jun 2014 11:20:41 +0000 (11:20 +0000)]
MFC r267335:
In some scenarios, when generating llvm/clang .inc.h files, multiple
source files could be passed to tblgen or clang-tblgen, leading to a
"Too many positional arguments specified" error message. Fix this by
replacing the too-generic ${.ALLSRC} sources with explicit paths.
pfg [Sat, 14 Jun 2014 02:57:40 +0000 (02:57 +0000)]
MFC r266865;
Fix strcasecmp_l() and strncasecmp_l() POSIX 2008 compliance.
POSIX.1-2008 specifies that those two functions should be declared by
including <strings.h>, not <string.h> (the latter only has strcoll_l()
and strxfrm_l()):
delphij [Sat, 14 Jun 2014 00:44:57 +0000 (00:44 +0000)]
MFC r267368:
Apply vendor fixes to the High Point drivers:
- Don't call xpt_free_path() in os_query_remove_device() and
always return TRUE.
- Update os_buildsgl() to support build logical SG table which
will be used by lower RAID module.
- Return CAM_SEL_TIMEOUTstatus for SCSIcommand failed as target
missing.
Many thanks to HighPoint for providing this driver update.
jhb [Fri, 13 Jun 2014 19:10:40 +0000 (19:10 +0000)]
MFC 262139,262140,262236,262281,262532:
Various x2APIC fixes and enhancements:
- Use spinlocks for the vioapic.
- Handle the SELF_IPI MSR.
- Simplify the APIC mode switching between MMIO and x2APIC. The guest is
no longer allowed to switch modes at runtime. Instead, the desired mode
is set when the virtual machine is created.
- Disallow MMIO access in x2APIC mode and MSR access in xAPIC mode.
- Add support for x2APIC virtualization assist in Intel VT-x.
jhb [Thu, 12 Jun 2014 21:36:17 +0000 (21:36 +0000)]
MFC 262615,262624:
Workaround an apparent bug in VMWare Fusion's nested VT support where it
triggers a VM exit with the exit reason of an external interrupt but
without a valid interrupt set in the exit interrupt information.
jhb [Thu, 12 Jun 2014 19:58:12 +0000 (19:58 +0000)]
MFC 261638,262144,262506,266765:
Add virtualized XSAVE support to bhyve which permits guests to use XSAVE and
XSAVE-enabled features like AVX.
- Store a per-cpu guest xcr0 register and handle xsetbv VM exits by emulating
the instruction.
- Only expose XSAVE to guests if XSAVE is enabled in the host. Only expose
a subset of XSAVE features currently supported by the guest and for which
the proper emulation of xsetbv is known. Currently this includes X87, SSE,
AVX, AVX-512, and Intel MPX.
- Add support for injecting hardware exceptions into the guest and use this
to trigger exceptions in the guest for invalid xsetbv operations instead
of potentially faulting in the host.
- Queue pending exceptions in the 'struct vcpu' instead of directly updating
the processor-specific VMCS or VMCB. The pending exception will be delivered
right before entering the guest.
- Rename the unused ioctl VM_INJECT_EVENT to VM_INJECT_EXCEPTION and restrict
it to only deliver x86 hardware exceptions. This new ioctl is now used to
inject a protection fault when the guest accesses an unimplemented MSR.
- Expose a subset of known-safe features from leaf 0 of the structured
extended features to guests if they are supported on the host including
RDFSBASE/RDGSBASE, BMI1/2, AVX2, AVX-512, HLE, ERMS, and RTM. Aside
from AVX-512, these features are all new instructions available for use
in ring 3 with no additional hypervisor changes needed.
jhb [Thu, 12 Jun 2014 17:15:56 +0000 (17:15 +0000)]
MFC 266263,266551,266552:
- Add definitions for more structured extended features as well as
XSAVE Extended Features for AVX512 and MPX (Memory Protection Extensions).
- Don't permit users to request a subset of the AVX512 or MPX xsave masks.
jhb [Thu, 12 Jun 2014 13:13:15 +0000 (13:13 +0000)]
MFC 260239,261268,265058:
Expand the support for PCI INTx interrupts including providing interrupt
routing information for INTx interrupts to I/O APIC pins and enabling
INTx interrupts in the virtio and AHCI backends.