das [Thu, 7 Aug 2008 14:39:56 +0000 (14:39 +0000)]
Use cpack() and the gcc extension __imag__ to implement cimag() and
conj() instead of using expressions like z * I. The latter is bad for
several reasons:
1. It is implemented using arithmetic, which is unnecessary, and can
generate floating point exceptions, contrary to the requirements on
these functions.
2. gcc implements complex multiplication using a formula that breaks
down for infinities, e.g., it gives INFINITY * I == nan + inf I.
edwin [Thu, 7 Aug 2008 00:14:56 +0000 (00:14 +0000)]
Update the units library file.
- Merge changes from NetBSD and OpenBSD.
- Add the Euro as a primitive unit, add old converted currency and
pegged currency (Obtained from Wikipedia)
- Rename "dollar" to "usdollar" as primitive unit, remove non-pegged
currency and add pegged currency (Obtained from Wikipedia)
- Updated the accuracy of a lot of constants (Obtained from Wikipedia)
imp [Wed, 6 Aug 2008 22:22:27 +0000 (22:22 +0000)]
When the miibus for the AX88x90 or TC5299J cards fails to attach, we
would call ed_release_resources() when we should have called
ed_detach() to properly undo the effects of prior calls to
ed_attach(). This would leave a stray ed interface ifnet alive in the
system, which was, well, bad, since we called if_free() on the
underlying memory... Fix the ed_detach routine to cope being called
in this context now.
This should never come up because the miibus is always there. Except
for now when it seems to be failing for reasons unknown... That's a
different bug that hits at least ed, xl, dc and fxp...
kib [Wed, 6 Aug 2008 14:47:19 +0000 (14:47 +0000)]
User may do "mount -o snapshot ...", that causes new FFS mount to be
performed with snapshot option, while the mp->mnt_opt is NULL.
Protect against NULL pointer dereference.
dfr [Wed, 6 Aug 2008 14:02:05 +0000 (14:02 +0000)]
Add an implementation of the RPCSEC_GSS authentication protocol for RPC. This
is based on an old implementation from the University of Michigan with lots of
changes and fixes by me and the addition of a Solaris-compatible API.
imp [Wed, 6 Aug 2008 07:34:35 +0000 (07:34 +0000)]
Unify the initial card probe/attach procedure with the kldload
procedure. There were some subtle differences before that could lead
to a variety of bugs, including resources being lost (in one case
forever). pccard_probe_and_attach_card does this now, and includes
comments about what's going on and why, since it isn't obvious from
the code. Please let me know if I've missed anything...
Provide a new function called pccard_select_cfe that allows drivers to
select which configuration entry to use. This is needed for some
older pre-MFC standard cards with many functions that want to activate
all their functions by selecting alternative entries, or to work
around broken ones. pccard_select_cfe will migrate into the
pccard_if.m interface as its interface stabilizes to keep all the
pccard drivers from referencing any symbols in the pccard.ko module
directly.
jhb [Tue, 5 Aug 2008 21:04:00 +0000 (21:04 +0000)]
Remove the second check for a 64-bit BAR value on a 32-bit system in
pci_add_map(). First, this condition is already handled earlier in
the function. Second, as written the check would never fire as the
'start' value was overwritten with a long value (rman_get_start() returns
long) before the comparison was done.
jhb [Tue, 5 Aug 2008 20:41:46 +0000 (20:41 +0000)]
Add a script to perform simple analysis of a crash dump (either a full
dump or minidump). When the script is run, it generates a text file
containing the output of several commands run againt the core dump such
as kgdb (stack trace), ps, netstat, vmstat, iostat, dmesg, and fstat.
jhb [Tue, 5 Aug 2008 20:02:31 +0000 (20:02 +0000)]
If a thread that is swapped out is made runnable, then the setrunnable()
routine wakes up proc0 so that proc0 can swap the thread back in.
Historically, this has been done by waking up proc0 directly from
setrunnable() itself via a wakeup(). When waking up a sleeping thread
that was swapped out (the usual case when waking proc0 since only sleeping
threads are eligible to be swapped out), this resulted in a bit of
recursion (e.g. wakeup() -> setrunnable() -> wakeup()).
With sleep queues having separate locks in 6.x and later, this caused a
spin lock LOR (sleepq lock -> sched_lock/thread lock -> sleepq lock).
An attempt was made to fix this in 7.0 by making the proc0 wakeup use
the ithread mechanism for doing the wakeup. However, this required
grabbing proc0's thread lock to perform the wakeup. If proc0 was asleep
elsewhere in the kernel (e.g. waiting for disk I/O), then this degenerated
into the same LOR since the thread lock would be some other sleepq lock.
Fix this by deferring the wakeup of the swapper until after the sleepq
lock held by the upper layer has been locked. The setrunnable() routine
now returns a boolean value to indicate whether or not proc0 needs to be
woken up. The end result is that consumers of the sleepq API such as
*sleep/wakeup, condition variables, sx locks, and lockmgr, have to wakeup
proc0 if they get a non-zero return value from sleepq_abort(),
sleepq_broadcast(), or sleepq_signal().
Discussed with: jeff
Glanced at by: sam
Tested by: Jurgen Weber jurgen - ish com au
MFC after: 2 weeks
jhb [Tue, 5 Aug 2008 18:24:41 +0000 (18:24 +0000)]
If the kernel fails to allocate resources for the initial value of a BAR
for a PCI device during the boot-time probe of the parent PCI bus, then
zero the BAR and clear the resource list entry for that BAR. This forces
the PCI bus driver to request a valid resource range from the parent bridge
driver when the device driver tries to allocate the BAR. Similarly, if the
initial value of a BAR is a valid range but it is > 4GB and the current OS
only has 32-bit longs, then do a full teardown of the initial value of the
BAR to force a reallocation.
jhb [Tue, 5 Aug 2008 17:39:37 +0000 (17:39 +0000)]
- Consolidate module version for the pcf module into just pcf.c instead
of having duplicate versions in each bus attachment.
- Add a DRIVER_MODULE() instance so that the iicbus(4) driver will
actually attach to pcf(4) driver instances.
- Fix compile of envctrl.c.
des [Tue, 5 Aug 2008 15:24:07 +0000 (15:24 +0000)]
ufsmount.h uses "struct\tfoo *bar;", except where it doesn't.
quota.h uses "struct foo\t*bar;", except where it doesn't.
Try to make them both agree with themselves (though not with eachother)
jhb [Mon, 4 Aug 2008 21:14:24 +0000 (21:14 +0000)]
Lock the consumers of the iicbus(4) infrastructure:
- ad7418(4) uses an sx lock instead of a mtx since the iicbus(4) stuff it
calls can sleep (request_bus()). Also, I expanded the locking slightly
to serialize writes to data stored in the softc.
- Similarly, the icee(4) driver now uses an sx lock instead of a mutex.
I also removed the pointless OPENED flag and flags field from the softc.
- The locking for the ic(4) driver was a bit trickier:
- Add a mutex to the softc to protect softc data.
- The driver uses malloc'd buffers that are the size of the interface
MTU to send and receive packets. Previously, these were allocated
every time the interface was brought up and anytime the MTU was
changed, with various races that could result in memory leaks. I
changed this to be a bit simpler and more like other NIC drivers in
that we allocate buffers during attach for the default MTU size and
only reallocate them on MTU changes. The reallocation procedure
goes to some lengths with various flags to not replace either the
the receive or transmit buffers while the driver is busy receiving
or transmitting a packet.
- Store the device_t of the driver in the softc instead of detours into
new-bus using if_dunit from the ifnet and an even more bizarre detour
to get the softc instead of using if_softc.
- Drop the driver mutex when invoking netisr_dispatch() to pass the
packet up to IP.
- Use if_printf().
jhb [Mon, 4 Aug 2008 21:03:06 +0000 (21:03 +0000)]
Add locking to the core iicbus(4) drivers:
- Add an sx lock to the iic(4) driver to serialize open(), close(), read(),
and write and to protect sc_addr and sc_count in the softc.
- Use cdev->si_drv1 instead of using the minor number of the cdev to
lookup the softc via newbus in iic(4).
- Store the device_t in the softc to avoid a similar detour via minor
numbers in iic(4).
- Only add at most one instance of iic(4) and iicsmb(4) to each iicbus(4)
instance, and do it in the child driver.
- Add a mutex to the iicbus(4) softc to synchronize the request/release bus
stuff.
- Use __BUS_ACCESSOR() for IICBUS_ACCESSOR() instead of rolling our own.
- Add a mutex to the iicsmb(4) softc to protect softc state updated in the
interrupt handler.
- Remove Giant from all the smbus methods in iicsmb(4) now that all the
iicbus(4) backend is locked.
jhb [Mon, 4 Aug 2008 20:46:15 +0000 (20:46 +0000)]
Add locking to the various iicbus(4) bridge drivers:
- Just grab Giant in the ixp425_iic(4) driver since this driver uses
a shared address/data register window pair to access the actual
I2C registers. None of the other ixp425 drivers lock access to these
shared address/data registers yet and that would need to be done before
this could use any meaningful locking.
- Add locking to the interrupt handler and 'iicbus_reset' methods of the
at91_twi(4) driver.
- Add locking to the pcf(4) driver. Other pcf(4) fixes include:
- Don't needlessly zero the softc.
- Use bus_foo rather than bus_space_foo and remove bus space tag and
handle from softc.
- The lpbb(4) driver just grabs Giant for now. This will be refined later
when ppbus(4) is locked.
- As was done with smbus earlier, move the DRIVER_MODULE() lines to match
the bus driver (either iicbus or iicbb) to the bridge driver into the
bridge drivers.
jhb [Mon, 4 Aug 2008 19:49:05 +0000 (19:49 +0000)]
- Close a race with concurrent open's of a pts master device which could
result in leaked tty structures.
- When constructing a new pty, allocate it's tty structure before adding
it to the list.
jhb [Mon, 4 Aug 2008 19:19:18 +0000 (19:19 +0000)]
Add locking to snc(4) so it is MPSAFE:
- Add a mutex to the softc to protect the softc and device hardware.
- Use a private timer routine to drive the transmit watchdog timer instead
of using if_watchdog/if_timer.
- If if_alloc() fails during attach, fail the attach with an error rather
than panic'ing.
- Clear RUNNING and OACTIVE only in sncstop().
- Don't mess with IFF_UP.
- Don't leak 'struct ifnet' on detach.
- Setup interrupt handler after ether_ifattach().
- Call ether_ifdetach() rather than if_detach() in the pccard detach
routine.
kib [Mon, 4 Aug 2008 12:22:33 +0000 (12:22 +0000)]
Lower the priority of the sleep in the syscons for "waitvt" wchan to
PZERO + 1. The sleeping process at the priority <= PZERO is counted as
blocked, or, as comment states, 'disk wait'. PZERO + 1 works as well,
and does not cause user confusion.
Reported by: sam <samflanker at gmail com>
MFC after: 1 week
cperciva [Mon, 4 Aug 2008 07:01:42 +0000 (07:01 +0000)]
Dereferencing uninitialized pointers considered harmful. Prior to this
commit, calling i386_parsedev(..., X, ...) where X is "ad", "bge", or
any other disk or network device name without a unit number, would
result in dereferencing whatever happened to be on the stack where the
variable "cp" is stored.
cperciva [Mon, 4 Aug 2008 06:55:42 +0000 (06:55 +0000)]
Initialize "nconv" to a reasonable value in all code paths. Prior to
this commit, sprintf("%s", "") could fail depending on what happened
to be on the stack.
cperciva [Mon, 4 Aug 2008 06:53:13 +0000 (06:53 +0000)]
Set "max" to a reasonable value if BLOCKSIZE has a bogus unit. Prior
to this commit, "env BLOCKSIZE=4X df" prints not only "4X: unknown
blocksize" as expected, but sometimes also "maximum blocksize is 1G"
and "minimum blocksize is 512" depending on what happened to be on
the stack.
yongari [Mon, 4 Aug 2008 02:34:40 +0000 (02:34 +0000)]
Don't enable TSO by default. Users of RTL8169/8110 reported
watchdog timeout issues and the root cause seems to stem from
silicon bug of controller. Personally I couldn't reproduce it on
RTL8169 controller but it seems it's dependent on usage pattern.
For newer PCIe based controllers I have no TSO complaints but
turning off TSO would be more safe. Users who are sure that
their controller works with TSO can still reenable the TSO with
ifconfig(8).
Reported by: Oliver Lehmann (lehmann at ans-netz dot de), Eugene Butusov (ebutusov at gmail dot com)
yongari [Mon, 4 Aug 2008 02:05:09 +0000 (02:05 +0000)]
The number of bits reserved for MSS in RealTek controllers is
11bits. This limits the maximum interface MTU size in TSO case
as upper stack should not generate TCP segments with MSS greater
than the limit. Armed with this information, disable TSO if
interface MTU is greater than the limit.
delphij [Mon, 4 Aug 2008 00:43:49 +0000 (00:43 +0000)]
Make quota(1) to compile with WARNS=6:
- ANSI'fy showrawquotas().
- Shut up GCC by initializing bgrace and igrace. The situation
that caused the GCC warning can never happen though.
scf [Sun, 3 Aug 2008 22:47:23 +0000 (22:47 +0000)]
Restructure and use different variables in the tests that involve
environ[0] to be more obvious that environ is not NULL before environ[0]
is tested. Although I believe the previous code worked, this change
improves code maintainability.
cognet [Sun, 3 Aug 2008 20:36:40 +0000 (20:36 +0000)]
ctime() expects a time_t, but qup->dqblk.dqb_btime is an int32_t, so for
big endian platforms where time_t is 64bits (ie armeb and sparc64), it will
be a problem.
Use a temporary time_t to work around this.
Submitted by: Matthew Luckie <mjl AT luckie DOT org dot nz>
MFC after: 3 days
ache [Sun, 3 Aug 2008 20:15:22 +0000 (20:15 +0000)]
Restored from previous backing out (because that is OpenBSD way, so
assumed to be reviewd by them):
Stir directly from the kernel PRNG, without taking less random pid & time
bytes too (when it is possible).
The difference with OpenBSD code is that they have KERN_ARND sysctl for
that task, while we need to read /dev/random
rwatson [Sun, 3 Aug 2008 15:44:56 +0000 (15:44 +0000)]
Remove broken code to replace st_mode value with ACCESSPERMS when
lstat(2) is called on symlinks -- this code appears never to have
worked. The PR this addresses suggests that the intended
original behavior is the right one, but as bde points out in the
PR comments, we do actually support storing a mode on symlinks,
so returning it seems reasonable.
This is consistent with Mac OS X, which despite documentation to
the contrary does return the mode set on a symlink, but not some
other platforms. The Single Unix Spec requires only that the
returned bits be "meaningful", which seems at best unhelpful as
advice goes.
kib [Sun, 3 Aug 2008 13:33:45 +0000 (13:33 +0000)]
Calling linker_load_dependencies() while holding the module'
vnode lock may cause a LOR between kld_sx lock and vnode lock.
linker_load_dependencies() drops kld_sx, and another thread may attempt
to load the same kld.
Disconnect drivers that haven't been ported to MPSAFE TTY yet.
As clearly mentioned on the mailing lists, there is a list of drivers
that have not been ported to the MPSAFE TTY layer yet. Remove them from
the kernel configuration files. This means people can now still use
these drivers if they explicitly put them in their kernel configuration
file, which is good.
People should keep in mind that after August 10, these drivers will not
work anymore. Even though owners of the hardware are capable of getting
these drivers working again, I will see if I can at least get them to a
compilable state (if time permits).
imp [Sat, 2 Aug 2008 23:29:35 +0000 (23:29 +0000)]
Add entry for Prism-3 based Siemens SpeedStream card. It has
identical packaging and model numbers, but has had the Prism-II chips
replaced by Prism-3. The wi driver was just updated with the new
entry.
das [Sat, 2 Aug 2008 19:17:00 +0000 (19:17 +0000)]
A few minor corrections, including some from bde:
- When y/x is huge, it's faster and more accurate to return pi/2
instead of pi - pi/2.
- There's no need for 3 lines of bit fiddling to compute -z.
- Fix a comment.