ian [Mon, 23 Mar 2015 22:42:42 +0000 (22:42 +0000)]
Do not save/restore the TLS pointer on context switch for armv6. The
pointer cannot be changed directly by userland code on armv6 (it can be on
armv4), so there's no need to save/restore.
dim [Mon, 23 Mar 2015 21:13:29 +0000 (21:13 +0000)]
Pull in r230348 from upstream llvm trunk (by Tim Northover):
ARM: treat [N x i32] and [N x i64] as AAPCS composite types
The logic is almost there already, with our special homogeneous
aggregate handling. Tweaking it like this allows front-ends to emit
AAPCS compliant code without ever having to count registers or add
discarded padding arguments.
Only arrays of i32 and i64 are needed to model AAPCS rules, but I
decided to apply the logic to all integer arrays for more consistency.
This fixes a possible "Unexpected member type for HA" error when
compiling lib/msun/bsdsrc/b_tgamma.c for armv6.
benno [Mon, 23 Mar 2015 18:45:37 +0000 (18:45 +0000)]
Be consistent with M_ZERO when allocating ccbs.
There are four places, all in cam_xpt.c, where ccbs are malloc'ed. Two of
these use M_ZERO, two don't. The two that don't meant that allocated ccbs
had trash in them making it hard to debug errors where they showed up. Due
to this, use M_ZERO all the time when allocating ccbs.
dteske [Mon, 23 Mar 2015 17:31:22 +0000 (17:31 +0000)]
Add missing variables password/bootlock_password.
NB: Using NULL for default values in-case someone
or something uncomments it and reboots. See
check-password.4th(8) for additional details.
dteske [Mon, 23 Mar 2015 16:43:46 +0000 (16:43 +0000)]
Increase max input for password/bootlock_password from 16 to 255.
When taking user input, don't show asterisks as the user types
but instead spin a twiddle. Implement Ctrl-U to clear user input.
If the buffer is empty, either because the user has yet to type
anything, presses Ctrl-U at any time, or presses backspace enough
to end in an empty buffer, the twiddle is erased to provide feed-
back to the user.
dteske [Mon, 23 Mar 2015 16:31:27 +0000 (16:31 +0000)]
Prevent password/bootlock_password features of loader.conf(5) from
locking out everyone in the case of setting a password longer than
the maximum (currently 16 characters). Now the required password is
truncated to the maximum input that can be read from the user.
PR: kern/198760
MFC after: 3 days
MFH: stable/10 stable/9
arybchik [Mon, 23 Mar 2015 15:45:39 +0000 (15:45 +0000)]
sfxge: add barriers to BAR write macros
In theory the barriers are required to cope with write combining and
reordering. Two barriers are added (sometimes merged to one):
1. Before the first write to guarantee that previous writes to the region
have been done
2. Before the last write to guarantee that write to the last dword/qword is
done after previous writes
Barriers are inserted before in the assumption that it is better to
postpone barriers as much as it is possible (more chances that the
operation has already been already done and barrier does not stall CPU).
On x86 and amd64 bus space write barriers are just compiler memory barriers
which are definitely required.
Sponsored by: Solarflare Communications, Inc.
Differential Revision: https://reviews.freebsd.org/D2077
arybchik [Mon, 23 Mar 2015 15:44:17 +0000 (15:44 +0000)]
sfxge: assert either kernel or internal copy of interface flags
ioctl to put interface down sets ifp->if_flags which holds the intended
administratively defined state and calls driver callback to apply it.
When everything is done, driver updates internal copy of
interface flags sc->if_flags which holds the operational state.
So, transmit from Rx path is possible when interface is intended to be
administratively down in accordance with ifp->if_flags, but not applied
yet and the operational state is up in accordance with sc->if_flags.
Sponsored by: Solarflare Communications, Inc.
Differential Revision: https://reviews.freebsd.org/D2075
andrew [Sun, 22 Mar 2015 18:17:55 +0000 (18:17 +0000)]
Stop calling x86_efi_copyin and x86_efi_getdev directly. This is to help
port loader.efi to both 32 and 64-bit ARM where we can use this file with
minimal changes.
dim [Sun, 22 Mar 2015 17:56:49 +0000 (17:56 +0000)]
Add FREEBSD-Xlist files for llvm, clang and lldb.
These are generated, and not "optimized" in any way, since I am not
entirely sure of the syntax or format of this type of file. Feel free
to suggest ways of shortening these lists.
The general idea is the same for all three files, though:
* Get rid of upstream build infrastructure (CMakeLists, Makefiles, etc)
* Delete tests, tools and utilities we don't want or use (including
samples)
* Remove various bits of upstream metadata files that we don't want or
use (.arcconfig, .gitignore, etc)
adrian [Sun, 22 Mar 2015 17:54:00 +0000 (17:54 +0000)]
Initialise the pps / packet tracking timestamp so 11n aggregation works again.
There's a bug in the ticks handling where when initialised at '0', once
the ticks counter wrapped the comparison math would never trigger.
The pps calculation would never happen, and thus aggregation was never
enabled.
It manifests itself as "oh you only get 11n transmit aggregation for the
first 10 minutes of uptime."
I'm sure there are other ticks related issues lurking in net80211.
Tested:
* ath / iwn, both with 'wlandebug +11n' and a little bit of iperf to
kick off the transmit A-MPDU negotiation once the pps gets high enough.
scottl [Sun, 22 Mar 2015 17:29:14 +0000 (17:29 +0000)]
Convert savecore to libxo. The 'verbose' paths still use printf since they're
meant for interactivity on the command line and contain a lot of redundant
information.
hselasky [Sun, 22 Mar 2015 13:11:56 +0000 (13:11 +0000)]
Fix for out of order device destruction notifications when using the
delist_dev() function. In addition to this change:
- add a proper description of this function
- add a proper witness assert inside this function
- switch a nearby line to use the "cdp" pointer instead of cdev2priv()
kib [Sun, 22 Mar 2015 09:09:26 +0000 (09:09 +0000)]
msdosfs: mark unused compat-mount fields
The magic number MSDOSFS_ARGSMAGIC, which used to distinguish
"old" vs "new" msdosfs mount arguments, has not been used since
2005; it should just go away now.
Likewise, the local-to-Unicode table that changed at the same
time is unused.
Leave the space reserved in the old style mount arguments, though,
since we still support the old mount call (via the cmount entry
point).
Submitted by: Chris Torek <chris.torek@gmail.com>
MFC after: 2 weeks
adrian [Sat, 21 Mar 2015 23:39:34 +0000 (23:39 +0000)]
re-enable building modules for the AR933x
* add ipfw
* delete ath / ath_ahb for now, until I can have Warner beat me
with the clue stick about putting in conditional build things into
the ath Makefile so the module builds can just have the HAL bits
that are relevant for a particular target.
mjg [Sat, 21 Mar 2015 20:24:03 +0000 (20:24 +0000)]
fork: assign refed credentials earlier
Prior to this change the kernel would take p1's credentials and assign
them tempororarily to p2. But p1 could change credentials at that time
and in effect give us a use-after-free.
dim [Sat, 21 Mar 2015 19:13:13 +0000 (19:13 +0000)]
Correctly pass the -mllvm -enable-gvn=false flag in CLANG_OPT_SMALL
(this has to be passed as a combination of two flags). Should fix the
case where the clang version is before 3.5.0.
Submitted by: Pedro Arthur <bygrandao@gmail.com>
X-MFC-With: r279018, r279378
alc [Sat, 21 Mar 2015 17:56:55 +0000 (17:56 +0000)]
Introduce vm_object_color() and use it in mmap(2) to set the color of
named objects to zero before the virtual address is selected. Previously,
the color setting was delayed until after the virtual address was
selected. In rtld, this delay effectively prevented the mapping of a
shared library's code section using superpages. Now, for example, we see
the first 1 MB of libc's code on armv6 mapped by a superpage after we've
gotten through the initial cold misses that bring the first 1 MB of code
into memory. (With the page clustering that we perform on read faults,
this happens quickly.)
cognet [Sat, 21 Mar 2015 15:32:59 +0000 (15:32 +0000)]
When waiting on PTE allocation, another thread could free the l2_dtable while
we're not looking at it.
Fix this by increasing l2->l2_occupancy before we try to alloc (and decrease
it if the allocation failed, or if another thread did a similar allocation).
kib [Sat, 21 Mar 2015 15:01:19 +0000 (15:01 +0000)]
Somewhat modernize the SysV shm code:
- Use real locking, replace Giant with global sx protecting the
subsystem. Since the subsystem' lock is no longer dropped during
the sleepsk, remove not needed SHMSEG_WANTED segment flag, and
revert r278963.
- To do proper code simplification possible after the change of the
lock, restructure several functions into _locked body and
originally-named wrapper which calls into _locked variant. This
allows to eliminate the 'goto done2' spread over the code.
- Merge shm_find_segment_by_shmid() and shm_find_segment_by_shmidx().
- Consistently change all function prototypes to ANSI C.
Reviewed by: mjg (who has earlier version of the similar patch to
introduce real locking)
Tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
hselasky [Sat, 21 Mar 2015 09:45:45 +0000 (09:45 +0000)]
The synchronisation value returned by the so-called feedback endpoint
appears to be too inaccurate that it can be used to synchronize the
playback data stream. If there is a recording endpoint associated with
the playback endpoint, use that instead. That means if the isochronous
OUT endpoint is asynchronus the USB audio driver will automatically
start recording, if possible, to get exact information about the
needed sample rate adjustments. In no recording endpoint is present,
no rate adaption will be done.
While at it fix an issue where the hardware buffer pointers don't get
reset at the first device PCM trigger.
Make some variables 32-bit to avoid problems with multithreading.
trasz [Sat, 21 Mar 2015 09:42:37 +0000 (09:42 +0000)]
Make the autofs LDAP script cope with server returning entries with
ENTRY_ATTRIBUTE (eg cn) after the VALUE_ATTRIBUTE (eg automountInformation),
instead of before.
PR: 198557
MFC after: 1 month
Sponsored by: The FreeBSD Foundation
adrian [Sat, 21 Mar 2015 06:16:23 +0000 (06:16 +0000)]
Add support for the TP-Link TL-WDR4300 and TL-WDR3600.
These are actually almost the same units; except one is 3x3 5GHz, and
one is 2x2 5GHz.
Tested:
* TP-Link TL-WDR3600
TODO:
* The ath0/ath1 MAC addresses are ye garbage (00:02:03:04:05:06); fixing
that will take a little more time. It works fine with the ath0/ath1
MAC addresses set manually.
* Go through and yank the AR9344 on-board switch config (arswitch1);
it's not required here for this AP.
adrian [Sat, 21 Mar 2015 06:08:35 +0000 (06:08 +0000)]
Add GPIO function mux configuration for AR934x SoCs.
The AR934x (and maybe others in this family) have a more complicated
GPIO mux. The AR71xx just has a single function register for a handful
of "GPIO or X" options, however the AR934x allows for one of roughly
100 behaviours for each GPIO pin.
So, this adds a quick hints based mechanism to configure the output
functions, which is required for some of the more interesting board
configurations. Specifically, some use external LNAs to improve
RX, and without the MUX/output configured right, the 2GHz RX side
will be plain terrible.
It doesn't yet configure the "input" side yet; I'll add that if
it's required.
Tested:
* TP-Link TL-WDR3600, testing 2GHz STA/AP modes, checking some
basic RX sensitivity things (ie, "can I see the AP on the other
side of the apartment that intentionally has poor signal reception
from where I am right now.")
Whilst here, fix a silly bug in the maxpin routine; I was missing
a break.
delphij [Sat, 21 Mar 2015 01:14:11 +0000 (01:14 +0000)]
Disable timestamping on devfs read/write operations by default.
Currently we update timestamps unconditionally when doing read or
write operations. This may slow things down on hardware where
reading timestamps is expensive (e.g. HPET, because of the default
vfs.timestamp_precision setting is nanosecond now) with limited
benefit.
A new sysctl variable, vfs.devfs.dotimes is added, which can be
set to non-zero value when the old behavior is desirable.
glebius [Fri, 20 Mar 2015 20:42:58 +0000 (20:42 +0000)]
In vlan_clone_match_ethervid():
- Use ifunit() instead of going through the interface list ourselves.
- Remove unused parameter.
- Move the most important comment above the function.
cperciva [Fri, 20 Mar 2015 19:40:19 +0000 (19:40 +0000)]
When creating VM images, copy the contents of the created filesystem into
a new filesystem before packaging it into a disk image. This prevents
"remnants" of deleted files from showing up in the VM images, and reduces
their compressed size (by about 10% for the cloudware images) as a result.
jhb [Fri, 20 Mar 2015 10:27:06 +0000 (10:27 +0000)]
Expand the bitcount* API to support 64-bit integers, plain ints and longs
and create a "hidden" API that can be used in other system headers without
adding namespace pollution.
- If the POPCNT instruction is enabled at compile time, use
__builtin_popcount*() to implement __bitcount*(), otherwise fall back
to software implementations.
- Use the existing bitcount16() and bitcount32() from <sys/systm.h> to
implement the non-POPCNT __bitcount16() and __bitcount32() in
<sys/types.h>.
- For the non-POPCNT __bitcount64(), use a similar SWAR method on 64-bit
systems. For 32-bit systems, use two __bitcount32() operations on the
two halves.
- Use __bitcount32() to provide a __bitcount() that operates on plain ints.
- Use either __bitcount32() or __bitcount64() to provide a
__bitcountl() that operates on longs.
- Add public bitcount*() wrappers for __bitcount*() for use in the kernel
in <sys/libkern.h>.
- Use __builtinl() instead of __builtin_popcountl() in BIT_COUNT().
zbb [Fri, 20 Mar 2015 10:15:34 +0000 (10:15 +0000)]
Allow to override default kernel virtual address assignment on ARM
Each plaform performs virtual memory split between kernel and user space
and assigns kernel certain amount of memory space. However, is is sometimes
reasonable to change the default values. Such situation may happen on
systems where the demand for kernel buffers is high, many devices occupying
memory etc. This of course comes with the cost of decreasing user space
memory range so shall be used with care. Most embedded systems will not
suffer from this limtation but rather take advantage of this potential
since default behavior is left unchanged.
kib [Thu, 19 Mar 2015 13:57:47 +0000 (13:57 +0000)]
Use VT-d interrupt remapping block (IR) to perform FSB messages
translation. In particular, despite IO-APICs only take 8bit apic id,
IR translation structures accept 32bit APIC Id, which allows x2APIC
mode to function properly. Extend msi_cpu of struct msi_intrsrc and
io_cpu of ioapic_intsrc to full int from one byte.
KPI of IR is isolated into the x86/iommu/iommu_intrmap.h, to avoid
bringing all dmar headers into interrupt code. The non-PCI(e) devices
which generate message interrupts on FSB require special handling. The
HPET FSB interrupts are remapped, while DMAR interrupts are not.
For each msi and ioapic interrupt source, the iommu cookie is added,
which is in fact index of the IRE (interrupt remap entry) in the IR
table. Cookie is made at the source allocation time, and then used at
the map time to fill both IRE and device registers. The MSI
address/data registers and IO-APIC redirection registers are
programmed with the special values which are recognized by IR and used
to restore the IRE index, to find proper delivery mode and target.
Map all MSI interrupts in the block when msi_map() is called.
Since an interrupt source setup and dismantle code are done in the
non-sleepable context, flushing interrupt entries cache in the IR
hardware, which is done async and ideally waits for the interrupt,
requires busy-wait for queue to drain. The dmar_qi_wait_for_seq() is
modified to take a boolean argument requesting busy-wait for the
written sequence number instead of waiting for interrupt.
Some interrupts are configured before IR is initialized, e.g. ACPI
SCI. Add intr_reprogram() function to reprogram all already
configured interrupts, and call it immediately before an IR unit is
enabled. There is still a small window after the IO-APIC redirection
entry is reprogrammed with cookie but before the unit is enabled, but
to fix this properly, IR must be started much earlier.
Add workarounds for 5500 and X58 northbridges, some revisions of which
have severe flaws in handling IR. Use the same identification methods
as employed by Linux.
andrew [Thu, 19 Mar 2015 13:53:47 +0000 (13:53 +0000)]
Start to import support for the AArch64 architecture from ARM. This change
only adds support for kernel-toolchain, however it is expected further
changes to add kernel and userland support will be committed as they are
reviewed.
As our copy of binutils is too old the devel/aarch64-binutils port needs
to be installed to pull in a linker.
To build either TARGET needs to be set to arm64, or TARGET_ARCH set to
aarch64. The latter is set so uname -p will return aarch64 as existing
third party software expects this.
Differential Revision: https://reviews.freebsd.org/D2005
Relnotes: Yes
Sponsored by: The FreeBSD Foundation
andrew [Thu, 19 Mar 2015 13:00:02 +0000 (13:00 +0000)]
Fix building ixgbe with gcc, it doesn't like nested extern declarations.
The fix is to move the extern declaration ix_crcstrip out of
ixgbe_setup_hw_rsc.
adrian [Thu, 19 Mar 2015 03:58:25 +0000 (03:58 +0000)]
Fix the label search routine in geom_map to not trip up on '\0' bytes.
* Just do the buf check early and fail out
* If the offset being searched is:
00110000 00 b5 7e 45 61 e2 76 d3 c1 78 dd 15 95 cd 1f f1 |..~Ea.v..x......|
.. and the match string is '.!/bin/sh'
.. then it'll set the match string[0] to '\0', do a strncmp() against
the read buffer, find it's matching two zero-length strings, and think
that's where to start.
alc [Thu, 19 Mar 2015 01:40:43 +0000 (01:40 +0000)]
Fix the root cause of the "vm_reserv_populate: reserv <address> is already
promoted" panics. The sequence of events that leads to a panic is rather
long and circuitous. First, suppose that process P has a promoted
superpage S within vm object O that it can write to. Then, suppose that P
forks, which leads to S being write protected. Now, before P's child
exits, suppose that P writes to another virtual page within O. Since the
pages within O are copy on write, a shadow object for O is created to
house the new physical copy of the faulted on virtual page. Then, before
P can fault on S, P's child exists. Now, when P faults on S, it will
follow the "optimized" path for copy-on-write faults in vm_fault(),
wherein the underlying physical page is moved from O to its shadow object
rather than allocating a new page and copying the new page's contents from
the old page. Moreover, suppose that every 4 KB physical page making up S
is moved to the shadow object in this way. However, the optimized path
does not move the underlying superpage reservation, which is the root
cause of the panics! Ultimately, P performs vm_object_collapse() on O's
shadow object, which destroys O and in doing so breaks any reservations
still belonging to O. This leaves the reservation underlying S in an
inconsistent state: It's simultaneously not in use and promoted. Breaking
a reservation does not demote it because I never intended for a promoted
reservation to be broken. It makes little sense. Finally, this
inconsistency leads to an assertion failure the next time that the
reservation is used.
The failing assertion does not (currently) exist in FreeBSD 10.x or
earlier. There, we will quietly break the promoted reservation. While
illogical and unintended, breaking the reservation is essentially
harmless.