np [Tue, 26 Feb 2013 00:27:27 +0000 (00:27 +0000)]
cxgbe(4): Ask the card's firmware to pad up tiny CPLs by encapsulating
them in a firmware message if it is able to do so. This works out
better for one of the FIFOs in the chip.
adrian [Mon, 25 Feb 2013 22:45:02 +0000 (22:45 +0000)]
Part #2 of the TX chainmask changes:
* Remove ar5416UpdateChainmasks();
* Remove the TX chainmask override code from the ar5416 TX descriptor
setup routines;
* Write a driver method to calculate the current chainmask based on the
operating mode and update the driver state;
* Call the HAL chainmask method before calling ath_hal_reset();
* Use the currently configured chainmask in the TX descriptors rather than
the hardware TX chainmasks.
adrian [Mon, 25 Feb 2013 22:42:43 +0000 (22:42 +0000)]
Begin adding support to explicitly set the current chainmask.
Right now the only way to set the chainmask is to set the hardware
configured chainmask through capabilities. This is fine for forcing
the chainmask to be something other than what the hardware is capable
of (eg to reduce TX/RX to one connected antenna) but it does change what
the HAL hardware chainmask configuration is.
For operational mode changes, it (may?) make sense to separately control
the TX/RX chainmask.
Right now it's done as part of ar5416_reset.c - ar5416UpdateChainMasks()
calculates which TX/RX chainmasks to enable based on the operating mode.
(1 for legacy and whatever is supported for 11n operation.) But doing
this in the HAL is suboptimal - the driver needs to know the currently
configured chainmask in order to correctly enable things for each
TX descriptor. This is currently done by overriding the chainmask
config in the ar5416 TX routines but this has to disappear - the AR9300
HAL support requires the driver to dynamically set the TX chainmask based
on the TX power and TX rate in order to meet mini-PCIe slot power
requirements.
So:
* Introduce a new HAL method to set the operational chainmask variables;
* Introduce null methods for the previous generation chipsets;
* Add new driver state to record the current chainmask separate from
the hardware configured chainmask.
Part #2 of this will involve disabling ar5416UpdateChainMasks() and moving
it into the driver; as well as properly programming the TX chainmask
based on the currently configured HAL chainmask.
Tested:
* AR5416, STA mode - both legacy (11a/11bg) and 11n rates - verified
that AR_SELFGEN_MASK (the chainmask used for self-generated frames like
ACKs and RTSes) is correct, as well as the TX descriptor contents is
correct.
des [Mon, 25 Feb 2013 22:25:56 +0000 (22:25 +0000)]
In all situations where we need to bootstrap yacc, we also need to
bootstrap liby. This was not readily apparent, because a normal 'make
buildworld' or 'make toolchain' builds liby before building anything
that uses yacc. However, 'make kernel-toolchain' does not build
headers or libraries, so it was not possible to build a kernel from
head on, say, stable/9 without first building the complete toolchain.
dteske [Mon, 25 Feb 2013 19:55:32 +0000 (19:55 +0000)]
Import media selection/preparation framework (sysinstall inspired). Makes
accessing files from various types of media nice and abstracted away from
the wet-work involved in preparing, validating, and initializing those
types of media. This will be used for the package management system module
and other modules that need access to files and want to allow the user to
decide where those files come from (either in a scripted fashion, prompted
fashion, or any combination thereof).
Heavily inspired by sysinstall and even uses the same reserved words so
that scripts are portable. Coded over months, tested continuously through-
out, and reviewed several times.
Some notes about the changes:
- Move network-setting acquisition/validation routines to media/tcpip.subr
- The options screen from sysinstall has been converted to a dialog menu
- The "UFS" media choice is renamed to "Directory" to reflect how sysinstall
treats the choice and a new [true] "UFS" media choice has been added that
acts on real UFS partitions (such as external disks with disklabels).
- Many more help files have been resurrected from sysinstall (I noticed that
some of the content seems a bit dated; I gave them a once-over but they
could really use an update).
- A total of 10 media choices are presented (via mediaGetType) including:
CD/DVD, FTP, FTP Passive, HTTP Proxy, Directory, NFS, DOS, UFS, Floppy, USB
- Novel struct/device management layer for managing the issue of passing
more information than can comfortably fit in an argument list.
sbruno [Mon, 25 Feb 2013 19:22:56 +0000 (19:22 +0000)]
The 5300 series ciss(4) board does not work in performant mode with our
currnet initialization sequence. Set it to simple mode only so that
systems can be updated from stable/7 to newer installations.
At some point, we should figure out why we cannot initialize performant
mode on this board.
peterj [Mon, 25 Feb 2013 19:05:40 +0000 (19:05 +0000)]
Enhance test(1) by adding provision to compare any combination of the
access, birth, change and modify times of two files, instead of only
being able to compare modify times. The builtin test in sh(1) will
automagically acquire the same expansion.
des [Mon, 25 Feb 2013 16:47:09 +0000 (16:47 +0000)]
- Implement and use usage().
- Document the -C option.
- Allow rtadvd to be invoked without any interfaces on the command line,
and document that possibility.
mm [Mon, 25 Feb 2013 12:33:31 +0000 (12:33 +0000)]
MFV v242732:
Merge the ZFS I/O deadman thread from vendor (illumos).
This feature panics the system on hanging ZFS I/O, helps debugging
and resumes failed service.
The panic behavior can be controlled with the loader-only tunables:
vfs.zfs.deadman_enabled (enable or disable panic on stalled ZFS I/O)
vfs.zfs.deadman_synctime (expiration time for stalled ZFS I/O)
By default, ZFS I/O deadman is enabled by default on amd64 and i386
excluding virtual guest machines.
mjacob [Mon, 25 Feb 2013 11:22:54 +0000 (11:22 +0000)]
Turn off fast posting for the ISP2100- I'd forgotten that it actually
might have been enabled for them- now that we use all 32 bits of handle.
Fast Posting doesn't pass the full 32 bits.
Noticed by: Bugs in NetBSD. Only a NetBSD user might actually still use such old hardware.
MFC after: 1 week
gonzo [Mon, 25 Feb 2013 08:04:47 +0000 (08:04 +0000)]
- Fix off-by-one error when returning max pin number
- Fix GPIOGET for output pins. Requesting state for
output pin is valid operation, get the state from
TI_GPIO_DATAOUTX register
marcel [Mon, 25 Feb 2013 02:41:38 +0000 (02:41 +0000)]
kernacc() expects all KVAs to be covered in the kernel map. With the
introduction of the PBVM, this stopped being the case. Redefine the
VM parameters so that the PBVM is included in the kernel map. In
particular this introduces VM_INIT_KERNEL_ADDRESS to point to the base
of region 5 now that VM_MIN_KERNEL_ADDRESS points to the base of
region 4 to include the PBVM.
While here define KERNBASE to the actual link address of the kernel as
is intended.
kientzle [Mon, 25 Feb 2013 01:50:04 +0000 (01:50 +0000)]
Fix the bug I introduced in r247045.
After digging through more carefully, it looks like there's
no real need to have the DTB in the module directory.
So we can simplify a lot: Just copy DTB into local heap
for "fdt addr" and U-Boot integration, drop all the extra
COPYIN() calls.
I've left one final COPYIN() to update the in-kernel DTB
for consistency with how this code used to work, but I'm
no longer convinced it's appropriate here.
I've also remove the mem_load_raw() utility that I added
to boot/common/module.c with r247045 since it's no longer
necessary.
jilles [Sun, 24 Feb 2013 20:53:32 +0000 (20:53 +0000)]
libc/opendir: Improve behaviour of union uniquifier:
* Reopen the directory using openat(fd, ".", ...) instead of opening the
pathname again. This fixes a race condition where the meaning of the
pathname changes and allows a reopen with fdopendir().
* Always reopen the directory for union stacks, not only when DTF_REWIND
is passed. Applications should be able to fchdir(dirfd(dir)) and
*at(dirfd(dir), ...). DTF_REWIND now does nothing.
markj [Sun, 24 Feb 2013 15:15:50 +0000 (15:15 +0000)]
Merge part of r1712 from elftoolchain, making it possible to resize ELF
sections and indirectly change the layout of an ELF file when
ELF_F_LAYOUT is not set.
dim [Sat, 23 Feb 2013 22:48:12 +0000 (22:48 +0000)]
Pull in r175962 from upstream llvm trunk:
X86: Disable cmov-memory patterns on subtargets without cmov.
Fixes PR15115.
For the i386 arch, this should enable cmov instructions only on
-march=pentiumpro and higher. Since our default CPU is i486, cmov
instructions will now be disabled by default.
kientzle [Sat, 23 Feb 2013 20:34:47 +0000 (20:34 +0000)]
"fdt addr" gets run from loader.rc before the kernel is loaded.
This was broken by r247045 which tried to copy the FDT into the
module directory immediately.
Instead, store the address and arrange for the FDT to get
copied into the module directory later when the usual
FDT initialization runs.
mav [Sat, 23 Feb 2013 18:32:42 +0000 (18:32 +0000)]
Add basic and not very reliable protection against going to sleep with
thread scheduled by interrupt fired after we entered critical section.
None of cpu_sleep() implementations on ARM check sched_runnable() now, so
put the first line of defence here. This mostly fixes unexpectedly long
sleeps in synthetic tests of calloutng code and probably other situations.
keramida [Sat, 23 Feb 2013 12:31:52 +0000 (12:31 +0000)]
Now that qsort(3) has a sample comparison function, point to that
example from bsearch(3) too, so that we don't have to duplicate
the example code in both places.
dim [Fri, 22 Feb 2013 22:40:10 +0000 (22:40 +0000)]
Pull in r172354 from upstream clang trunk:
Refactor the x86 CPU name logic in the driver and pass -march and -mcpu
flag information down from the Clang driver into the Gold linker plugin
for LTO. This allows specifying -march on the linker commandline and
should hopefully have it pass all the way through to the LTO optimizer.
Fixes PR14697.
Pull in r175919 from upstream clang trunk:
Driver: Pass down the -march setting down to -cc1as on x86 too.
The assembler historically didn't make use of any target features, but this has
changed when support for old CPUs that don't support long nops was added.
This should fix the long nops that still occurred in crt*.o, and
possibly other object files, if the system was compiled for a CPU that
does not support those, such as Geode.
Note that gcc on i386 also does not pass through any -march, -mcpu or
-mtune setting to gas, but this has not caused any trouble yet, because
gas defaults to i386.
mav [Fri, 22 Feb 2013 21:43:21 +0000 (21:43 +0000)]
Fix command timeout caused by data underrun during fetching ATAPI sense
data, introduced by r246713. There are two places where ata_request is
filled in ATA_CAM: ata_cam_begin_transaction() and ata_cam_request_sense().
In the first case DMA should be done for addresses from the CCB. In second
case, DMA should be done to the different address, the address of the sense
buffer inside the CCB structure itself.
mav [Fri, 22 Feb 2013 19:53:12 +0000 (19:53 +0000)]
Hide SEMB port of the SiI3826 Port Multiplier by default to avoid extra
errors while it tries to talk via I2C to usually missing external SEP.
There is tunable to enable it back when needed.
gallatin [Fri, 22 Feb 2013 19:21:29 +0000 (19:21 +0000)]
Improvements for newer mxge nics:
- Some mxge nics may store the serial number in the SN2 field of the
EEPROM. These will also have an SN=0 field, so parse the SN2 field,
and give it precedence.
- Skip MXGEFW_CMD_UNALIGNED_TEST on mxge nics which do not require it.
This saves roughly 10ms per port at device attach time.
mav [Fri, 22 Feb 2013 17:45:32 +0000 (17:45 +0000)]
Add DA_Q_NO_PREVENT quirk for Kingston DataTraveler G3 1.00 USB flash.
PREVENT ALLOW MEDIUM REMOVAL commands return errors on these devices
without returning sense data. In some cases unrelated following commands
start to return errors too, that makes device to be dropped by CAM.
gallatin [Fri, 22 Feb 2013 16:46:28 +0000 (16:46 +0000)]
Try harder to make mxge safe for all combinations of INET and INET6
- Re-fix build by restoring local removed in r247151, but protected
by #if defined(INET) || defined(INET6) so that the compile
succeeds in the !(INET||INET6) case.
- Protect call to in_pseudo() with an #ifdef INET, to allow
a kernel to link with mxge when INET is not compiled in.
- Also remove an errant (improperly commented) obsolete debugging printf
Thanks to Glebius for pointing out the !(INET||INET6) build issue.
adrian [Fri, 22 Feb 2013 07:07:11 +0000 (07:07 +0000)]
Add a workaround for AR5416, AR9130 and AR9160 chipsets - work around
an incorrectly calculated RTS duration value when transmitting aggregates.
These earlier 802.11n NICs incorrectly used the ACK duration time when
calculating what to put in the RTS of an aggregate frame. Instead it
should have used the block-ack time. The result is that other stations
may not reserve enough time and start transmitting _over_ the top of
the in-progress blockack field. Tsk.
This workaround is to popuate the burst duration field with the delta
between the ACK duration the hardware is using and the required duration
for the block-ack. The result is that the RTS field should now contain
the correct duration for the subsequent block-ack.
grehan [Fri, 22 Feb 2013 00:46:32 +0000 (00:46 +0000)]
Add the ability to have a 'fallback' search for memory ranges.
These set of ranges will be looked at if a standard memory
range isn't found, and won't be installed in the cache.
Use this to implement the memory behaviour of the PCI hole on
x86 systems, where writes are ignored and reads always return -1.
This allows breakpoints to be set when issuing a 'boot -d', which
has the side effect of accessing the PCI hole when changing the
PTE protection on kernel code, since the pmap layer hasn't been
initialized (a bug, but present in existing FreeBSD releases so
has to be handled).
attilio [Thu, 21 Feb 2013 21:54:53 +0000 (21:54 +0000)]
Hide the details for the assertion for VM_OBJECT_LOCK operations.
Rename current VM_OBJECT_LOCK_ASSERT(foo, RA_WLOCKED) into
VM_OBJECT_ASSERT_WLOCKED(foo)
jhb [Thu, 21 Feb 2013 19:02:50 +0000 (19:02 +0000)]
Further refine the handling of stop signals in the NFS client. The
changes in r246417 were incomplete as they did not add explicit calls to
sigdeferstop() around all the places that previously passed SBDRY to
_sleep(). In addition, nfs_getcacheblk() could trigger a write RPC from
getblk() resulting in sigdeferstop() recursing. Rather than manually
deferring stop signals in specific places, change the VFS_*() and VOP_*()
methods to defer stop signals for filesystems which request this behavior
via a new VFCF_SBDRY flag. Note that this has to be a VFC flag rather than
a MNTK flag so that it works properly with VFS_MOUNT() when the mount is
not yet fully constructed. For now, only the NFS clients are set this new
flag in VFS_SET().
A few other related changes:
- Add an assertion to ensure that TDF_SBDRY doesn't leak to userland.
- When a lookup request uses VOP_READLINK() to follow a symlink, mark
the request as being on behalf of the thread performing the lookup
(cnp_thread) rather than using a NULL thread pointer. This causes
NFS to properly handle signals during this VOP on an interruptible
mount.
gallatin [Thu, 21 Feb 2013 17:00:35 +0000 (17:00 +0000)]
Fix tcp_lro_rx_ipv4() for drivers that do not set CSUM_IP_CHECKED.
Specifcially, in_cksum_hdr() returns 0 (not 0xffff) when the IPv4
checksum is correct. Without this fix, the tcp_lro code will reject
good IPv4 traffic from drivers that do not implement IPv4 header
harder csum offload.
imp [Thu, 21 Feb 2013 15:35:48 +0000 (15:35 +0000)]
Use critical_enter/critical_exit around the time sensitive part of
this code to depessimize the worst case we've lived with silently and
uneventfully for the past 12 years. Add a comment about a refinement
for those needing more assurance of accuracy.
Fix ddb's show rtc command deadlock potential when debugging rtc code
by not taking the lock if we're in the debugger. If you need a thumb
to count the number of people that have encountered this, I'd be
surprised.
imp [Thu, 21 Feb 2013 07:16:40 +0000 (07:16 +0000)]
Locking for todr got pushed down into inittodr and the client
libraries it calls (although some might not be doing it right). We are
serialized right now by giant as well. This means the splsoftclock are
now an anachronism that has no benefit, even marking where locking
needs to happen. Remove them.
adrian [Thu, 21 Feb 2013 06:18:40 +0000 (06:18 +0000)]
Add a new option to limit the maximum size of aggregates.
The default is to limit them to what the hardware is capable of.
Add sysctl twiddles for both the non-RTS and RTS protected aggregate
generation.
Whilst here, add some comments about stuff that I've discovered during
my exploration of the TX aggregate / delimiter setup path from the
reference driver.