jilles [Sun, 12 Dec 2010 22:59:34 +0000 (22:59 +0000)]
sh: Various simplifications to jobs.c:
* Prefer kill(-X) to killpg(X).
* Remove some dead code.
* No additional SIGINT is needed if int_pending() is already true.
joel [Sun, 12 Dec 2010 21:26:12 +0000 (21:26 +0000)]
Remove the advertising clause from UCB copyrighted files in sbin. This
is in accordance with the information at
ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change
kib [Sun, 12 Dec 2010 16:16:39 +0000 (16:16 +0000)]
In fpudna()/npxdna(), mark FPU context initialized and optionally
mark user FPU context initialized, if current context is user context.
It was reversed in r215865, by inadequate change of this code fragment
to a call to fpuuserinited()/npxuserinited().
The issue is only relevant for in-kernel users of FPU.
Reported by: Jan Henrik Sylvester <me janh de>, Mike Tancsa <mike sentex net>
Tested by: Mike Tancsa
MFC after: 3 days
trasz [Sun, 12 Dec 2010 13:04:30 +0000 (13:04 +0000)]
Add regression test for new NFSv4 ACL semantics, verified with ZFSv28.
Note that to run it, you need not only ZFSv28, but also a modified setfacl(1),
which is not in the tree yet.
jh [Sun, 12 Dec 2010 08:52:13 +0000 (08:52 +0000)]
Handle the special ruleset 0 in devfs_ruleset_use(). An attempt set the
current ruleset to 0 with command "devfs ruleset 0" triggered a KASSERT
in devfs_ruleset_create().
jchandra [Sun, 12 Dec 2010 06:00:26 +0000 (06:00 +0000)]
I2C drivers for XLR/XLS processors.
- Major update to xlr_i2c.c: do multi-byte ops correctly, remove unnecessary
code, add mutex to protect bus operations, style(9) fixes.
- Drivers for I2C devices on XLR/XLS engineering boards, ds1374u RTC, max6657
temparature sensor and at24co2n EEPROM.
Submitted by: Sreekanth M. S. (kanthms at netlogicmicro com)
jilles [Sun, 12 Dec 2010 00:07:27 +0000 (00:07 +0000)]
sh: Remove the herefd hack.
The herefd hack wrote out partial here documents while expanding them. It
seems unnecessary complication given that other expansions just allocate
memory. It causes bugs because the stack is also used for intermediate
results such as arithmetic expressions. Such places should disable herefd
for the duration but not all of them do, and I prefer removing the need for
disabling herefd to disabling it everywhere needed.
Here documents larger than 1024 bytes will use a bit more CPU time and
memory.
Additionally this allows a later change to expand here documents in the
current shell environment. (This is faster for small here documents but also
changes behaviour.)
cperciva [Sat, 11 Dec 2010 22:33:33 +0000 (22:33 +0000)]
Reduce the Xen timecounter from 1GHz to 2^-9 GHz, thereby increasing the
timecounter period from 2^32 ns (~4.3s) to 2^41 ns (~36m39s). Some time
sharing systems can skip clock interrupts for a few seconds when under
load (e.g., if we've recently used more than our fair share of CPU and
someone else wants a burst of CPU) and we were losing time in quanta of
2^32 ns due to timecounter wrapping.
Increasing the timecounter period up to 2^41 ns is definitely overkill,
but we still have microsecond timecounter precision, and anyone using
paravirtualized hardware when they need submicrosecond timing is crazy.
jilles [Sat, 11 Dec 2010 17:47:27 +0000 (17:47 +0000)]
sh: Use vsnprintf() rather than crafting our own in fmtstr().
Add INTOFF/INTON as longjmp out of vsnprintf may cause memory leaks or
undefined behaviour.
joel [Sat, 11 Dec 2010 09:38:12 +0000 (09:38 +0000)]
Remove the advertising clause from UCB copyrighted files in usr.sbin. This
is in accordance with the information provided at
ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change
joel [Sat, 11 Dec 2010 08:32:16 +0000 (08:32 +0000)]
Remove the advertising clause from UCB copyrighted files in usr.bin. This
is in accordance with the information provided at
ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change
Also add $FreeBSD$ to a few files to keep svn happy.
rwatson [Fri, 10 Dec 2010 22:22:01 +0000 (22:22 +0000)]
Derive the XENHVM kernel from GENERIC, adding only the options required
to support PV drivers (such as xenpci), and non-adptive locking (along
with a comment about why).
This change eliminates the synchronisation problem between GENERIC and
XENHVM, which had become severely rotted in HEAD, and in 8-STABLE
included non-production kernel debugging features such as WITNESS.
However, it comes at the cost of enabling devices and options that may
not be present under Xen (such as random ethernet cards). For now, opt
for a simpler kernel configuration file rather than using nooptions/
nodevice to enumerate and eliminate them. This leads to a somewhat
larger XENHVM kernel.
This is an MFC candidate for 8-STABLE before 8.2, in order to provide
a production-worthy XENHVM kernel configuration for amd64.
Discussed with: gibbs, cperciva
Reported by: Piete Brooks <Piete.Brooks at cl.cam.ac.uk>
Sponsored by: DARPA, AFRL
MFC after: 3 days
ken [Fri, 10 Dec 2010 21:45:10 +0000 (21:45 +0000)]
Fix an event handling bug with the mps(4) driver.
This bug manifested itself after repeated device arrivals and
departures. The root of the problem was that the last entry in the
reply array wasn't initialized/allocated. So every time we got
around to that event, we had a bogus address.
There were a couple more problems with the code that are also fixed:
- The reply mechanism was being treated as sequential (indexed by
sc->replycurindex) even though the spec says that the driver
should use the ReplyFrameAddress field of the post queue
descriptor to figure out where the reply is. There is no
guarantee that the reply descriptors will be used in sequential
order.
- The second word of the reply post queue descriptor wasn't being
checked in mps_intr_locked() to make sure that it wasn't
0xffffffff. So the driver could potentially come across a
partially DMAed descriptor.
- The number of replies allocated was one less than the actual
size of the queue. Instead, it was the size of the number of
replies that can be used at one time. (Which is one less than
the size of the queue.)
mps.c: When initializing the entries in the reply free
queue, make sure we initialize the full number that
we tell the chip we have (sc->fqdepth), not the
number that can be used at any one time (sc->num_replies).
When allocating replies, make sure we allocate the
number of replies that we've told the chip exist,
not just the number that can be used simultaneously.
Use the ReplyFrameAddress field of the post queue
descriptor to figure out which reply is being
referenced. This is what the spec says to do, and
the spec doesn't guarantee that the replies will be
used in order.
Put a check in to verify that the reply address passed
back from the card is valid. (Panic if it isn't, we'll
panic when we try to deference the reply pointer in any
case.)
In mps_intr_locked(), verify that the second word of the
post queue descriptor is not 0xffffffff in addition to
verifying that the unused flag is not set, so we can
make sure we didn't get a partially DMAed descriptor.
Remove references to sc->replycurindex, it isn't needed
now.
mpsvar.h: Remove replycurindex from the softc, it isn't needed now.
ken [Fri, 10 Dec 2010 21:38:51 +0000 (21:38 +0000)]
Fix a few issues related to the XPT_GDEV_ADVINFO CCB.
camcontrol.c: In buildbusdevlist(), don't attempt to get call
getdevid() for an unconfigured device, even when the
verbose flag is set. The cam_open_btl() call will almost
certainly fail.
Probe for the buffer size when issuing the XPT_GDEV_ADVINFO
CCB. Probing for the buffer size first helps us avoid
allocating the maximum buffer size when it really may not
be necessary. This also helps avoid errors from
cam_periph_mapmem() if we attempt to map more than MAXPHYS.
cam_periph.c: In cam_periph_mapmem(), if the XPT_GDEV_ADVINFO CCB
shows a bufsiz of 0, we don't have anything to map,
so just return.
Also, set the maximum mapping size to MAXPHYS
instead of DFLTPHYS for XPT_GDEV_ADVINFO CCBs,
since they don't actually go down to the hardware.
scsi_pass.c: Don't bother mapping the buffer in XPT_GDEV_ADVINFO
CCBs if bufsiz is 0.
andreast [Fri, 10 Dec 2010 20:27:50 +0000 (20:27 +0000)]
On the Xserve G5 we find the LM75 instead of the DS1775. The core
functionality is the same, a difference is that the DS1775 has a better
precision than the LM75. But we do not use it in our setup. Make the
LM75 work the same as the DS1775.
Fix a typo in device_set_desc.
Tested by: Paul Mather <paul at gromit dlib vt edu>
joerg [Fri, 10 Dec 2010 14:04:41 +0000 (14:04 +0000)]
Fix __retval vs. retval confusion: retval is meant to store the (userland)
pointer where data is to be returned by ibask() (currently unimplemented),
while __retval holds the value returned by the libgpib ibfoo() functions.
The confusion resulted in the ibfoo() functions returning an uninitialized
value except in situations where the GPIB activity has been terminated
abnormally.
tijl [Fri, 10 Dec 2010 08:53:06 +0000 (08:53 +0000)]
Add a .note.ABI-tag section to ia64 startup files by linking crtbrand.c
in crt1.o. On other architectures crtbrand.c is included from crt1.c,
but that's not a C source code file on ia64. Instead it is compiled
separately and included in crt1.o using incremental linking.
Tested by: dim (previous version)
Approved by: kib (mentor)
dougb [Thu, 9 Dec 2010 23:57:23 +0000 (23:57 +0000)]
Bring in the change from OpenBSD's 1.14:
"synchronize synopsis and usage; "-l", "-r", "-s" and "-x" are mutually
exclusive; while here, slightly improve spacing in the source code
so it fits on a 80-column display again.
dim [Thu, 9 Dec 2010 21:31:21 +0000 (21:31 +0000)]
Let all .c and .S files under lib/csu consistently use the __FBSDID()
macro for identification, instead of several different hand-rolled
variants (plain .ident, .ascii, etc).
jkim [Thu, 9 Dec 2010 21:29:36 +0000 (21:29 +0000)]
Remove AMD Family 0Fh, Model 6Bh, Stepping 2 from the list of P-state
invariant CPUs. I do not believe this model is P-state invariant any more.
Maybe cpufreq(4) was broken at the time of commit. :-(
mlaier [Thu, 9 Dec 2010 21:02:22 +0000 (21:02 +0000)]
Fix a long standing (from the original 4.4BSD lite sources) race between
vmspace_fork and vm_map_wire that would lead to "vm_fault_copy_wired: page
missing" panics. While faulting in pages for a map entry that is being
wired down, mark the containing map as busy. In vmspace_fork wait until the
map is unbusy, before we try to copy the entries.
Reviewed by: kib
MFC after: 5 days
Sponsored by: Isilon Systems, Inc.
jhb [Thu, 9 Dec 2010 20:28:30 +0000 (20:28 +0000)]
When reopening a stream backed by an open file descriptor, do not close
the existing file descriptor. Instead, let dup2() atomically close the
old file descriptor when assigning the newly opened file to the same
descriptor. This closes a race in a multithreaded application where a
concurrent open() could allocate the existing file descriptor in between
the calls to close() and dup2().
alc [Thu, 9 Dec 2010 20:16:00 +0000 (20:16 +0000)]
When r207410 eliminated the acquisition and release of the page queues
lock from pmap_extract_and_hold(), it didn't take into account that
pmap_pte_quick() sometimes requires the page queues lock to be held.
This change reimplements pmap_extract_and_hold() such that it no
longer uses pmap_pte_quick(), and thus never requires the page queues
lock.
For consistency, adopt the same idiom as used by the new
implementation of pmap_extract_and_hold() in pmap_extract() and
pmap_mincore(). It also happens to make these functions shorter.
rmacklem [Thu, 9 Dec 2010 19:02:23 +0000 (19:02 +0000)]
Disable attempts to establish a callback connection from the
experimental NFSv4 server to a NFSv4 client when delegations are not
being issued, even if the client advertises a callback path.
This avoids a problem where a Linux client advertises a
callback path that doesn't work, due to a firewall, and then
times out an Open attempt before the FreeBSD server gives up
its callback connection attempt. (Suggested by
drb at karlov.mff.cuni.cz to fix the Linux client problem that
he reported on the fs-stable mailing list.)
The server should probably have
a 1sec timeout on callback connection attempts when there are
no delegations issued to the client, but that patch will require
changes to the krpc and this serves as a work around until then.
Tested by: drb at karlov.mff.cuni.cz
MFC after: 5 days
jchandra [Thu, 9 Dec 2010 06:34:28 +0000 (06:34 +0000)]
UMA_MD_SMALL_ALLOC for mips.
Implement uma_small_alloc() and uma_small_free() for mips that allocates
pages from direct mapped memory. Uses the same mechanism as the page table
page allocator, so that we allocate from KSEG0 in 32 bit, and from XKPHYS
on 64 bit.
davidxu [Thu, 9 Dec 2010 02:42:02 +0000 (02:42 +0000)]
MFp4:
It is possible a lower priority thread lending priority to higher priority
thread, in old code, it is ignored, however the lending should always be
recorded, add field td_lend_user_pri to fix the problem, if a thread does
not have borrowed priority, its value is PRI_MAX.
cperciva [Wed, 8 Dec 2010 20:20:10 +0000 (20:20 +0000)]
On amd64, we have (since r1.72, in December 2005) MAX_BPAGES=8192,
while on i386 we have MAX_BPAGES=512. Implement this difference via
'#ifdef __i386__'.
With this commit, the i386 and amd64 busdma_machdep.c files become
identical; they will soon be replaced by a single file under sys/x86.
cperciva [Wed, 8 Dec 2010 19:52:04 +0000 (19:52 +0000)]
MFi386 r1.94: If XEN, make pmap_kextract = pmap_kextract_ma. This is a
no-op currently, since FreeBSD/amd64 doesn't have (paravirtualized) Xen
support, but if/when that support is ever added we'll want this, and
until then it's harmless.
syrinx [Wed, 8 Dec 2010 17:27:59 +0000 (17:27 +0000)]
Add (disabled) sample configurations needed to enable the snmp_usm and
snmp_vacm modules and minimal user/view configurations needed to for
the modules to work properly.
Sponsored by: The FreeBSD Foundation
Reviewed by: philip@ (mostly)
Approved by: philip@
syrinx [Wed, 8 Dec 2010 14:30:25 +0000 (14:30 +0000)]
Add bsnmpd(1)'s SNMP client tools (including SNMPv3 support) to the base system.
Sponsored by: The FreeBSD Foundation (the SNMPv3 bits), Google Summer of Code 2005
Reviewed by: philip@ (mostly), bz@ (earlier version based on p4 ch124545)
Approved by: philip@
syrinx [Wed, 8 Dec 2010 13:51:38 +0000 (13:51 +0000)]
In bsnmpd(1) add support for SNMPv3 message processing model, including message authentication, packet encryption & view-based access control (RFC 3412, 3414, 3415).
Sponsored by: The FreeBSD Foundation
Reviewed by: philip@ (mostly)
Approved by: philip@
yongari [Wed, 8 Dec 2010 01:24:05 +0000 (01:24 +0000)]
r184610 changed the way how TX frames are handled on AX88178 and
AX88772 controllers. ASIX added a new feature for AX88178/AX88772
controllers which allows combining multiple TX frames into a single
big frame. This was to overcome one of USB limitation where it
can't generate more than 8k interrupts/sec which in turn means USB
ethernet controllers can not send more than 8k packets per second.
Using ASIX's feature greatly enhanced TX performance(more than 3~4
times) compared to 7.x driver. However it seems r184610 removed
boundary checking for buffered frames which in turn caused
instability issues under certain conditions. In addition, using
ASIX's feature triggered another issue which made USB controller
hang under certain conditions. Restarting ethernet controller
didn't help under this hang condition and unplugging and replugging
the controller was the only solution. I believe there is a silicon
bug in TX frame combining feature on AX88178/AX88772 controllers.
To address these issues, reintroduce the boundary checking for both
AX88178 and AX88772 after copying a frame to USB buffer and do not
use ASIX's multiple frame combining feature. Instead, use USB
controller's multi-frame transmit capability to enhance TX
performance as suggested by Hans[1].
This should fix a long standing axe(4) instability issues reported
on AX88772 and AX88178 controllers. While I'm here remove
unnecessary TX frame length check since upper stack always
guarantee the size of a frame to be less than MCLBYTES.
Special thanks to Derrick Brashear who tried numerous patches
during last 4 months and waited real fix with patience. Without
this enthusiastic support, patience and H/W donation I couldn't fix
it since I was not able to trigger the issue on my box.
cperciva [Tue, 7 Dec 2010 23:33:20 +0000 (23:33 +0000)]
Postpone the unmasking of event channels (aka. interrupts) until after
the interrupt handlers have been registered. NULL isn't a very good
interrupt handler.
jkim [Tue, 7 Dec 2010 22:23:26 +0000 (22:23 +0000)]
Now the P-state invariant TSC is probed early enough, do not register event
handlers for CPU freqency changes when it is found P-state invariant.
Adjust a comment about non-existent tsc_freq_max() while I am here.
brucec [Tue, 7 Dec 2010 20:46:11 +0000 (20:46 +0000)]
Don't warn if a partition appears not to be aligned on a track boundary.
Modern disks use LBA and create a fake CHS geometry that doesn't have any
relation to the on-disk layout of data.
weongyo [Tue, 7 Dec 2010 20:23:47 +0000 (20:23 +0000)]
Introduces IFF_CANTCONFIG interface flag to point that the interface
isn't configurable in a meaningful way. This is for ifconfig(8) or
other tools not to change code whenever IFT_USB-like interfaces are
registered at the interface list.
jhb [Tue, 7 Dec 2010 18:49:11 +0000 (18:49 +0000)]
Use proper resource ID's for HPET IRQ resources. This mostly consists of
looking to see if there is an existing IRQ resource for a given IRQ
provided by the BIOS and using that RID if so. Otherwise, allocate a new
RID for the new IRQ.
ivoras [Tue, 7 Dec 2010 15:24:08 +0000 (15:24 +0000)]
Undo r216230: the interaction between saved ashift in metadata and
detected ashift does not support this. With this change, pools
created while stripesize=512 could not be imported when stripesize
becomes larger (on the same drive).
kib [Tue, 7 Dec 2010 12:44:33 +0000 (12:44 +0000)]
Update some comments related to use of amd64 full context switch.
In exec_linux_setregs(), use locally cached pointer to pcb to set
pcb_full_iret.
In set_regs(), note that full return is needed when code that sets
segment registers is enabled.
avg [Tue, 7 Dec 2010 12:25:26 +0000 (12:25 +0000)]
opensolaris cyclic: fix deadlock and make a little bit closer to upstream
The dealock was caused in the following way:
- thread T1 on CPU C1 holds a spin mutex, IPIs CPU C2 and waits for the
IPI to be handled
- C2 executes timer interrupt filter, thus has interrupts disabled, and
gets blocked on the spin mutex held by T1
The problem seems to have been introduced by simplifications made to
OpenSolaris code during porting.
The problem is fixed by reorganizing the code to more closely resemble
the upstream version. Interrupt filter (cyclic_fire) now doesn't
acquire any locks, all per-CPU data accesses are performed on a
target CPU with preemption and interrupts disabled thus precluding
concurrent access to the data.
cyp_mtx spin mutex is used to disable preemtion and interrupts; it's not
used for classical mutual exclusion, because xcall already serializes
calls to a CPU. It's an emulation of OpenSolaris
cyb_set_level(CY_HIGH_LEVEL) call, the spin mutexes could probably be
reduced to just a spinlock_enter()/_exit() pair.
Diff with upstream version is now reduced by ~500 lines, however it still
remains quite large - many things that are not needed (at the moment) or
are irrelevant on FreeBSD were simply ripped out during porting.
Examples of such things:
- support for CPU onlining/offlining
- support for suspend/resume
- support for running callouts at soft interrupt levels
- support for callout rebinding from CPU to CPU
- support for CPU partitions