Sascha Wildner [Sat, 24 Nov 2012 22:14:45 +0000 (23:14 +0100)]
Fix some typos (unknow -> unknown).
Sascha Wildner [Sat, 24 Nov 2012 20:35:21 +0000 (21:35 +0100)]
usb4bsd: Initial round of build fixes for serial drivers.
Not hooked into the build because some things are still missing,
most importantly ucom(4) porting.
Sascha Wildner [Sat, 24 Nov 2012 20:11:30 +0000 (21:11 +0100)]
ndis(4): Bring in some fixes from FreeBSD.
Sascha Wildner [Sat, 24 Nov 2012 20:22:55 +0000 (21:22 +0100)]
ndis(4): Make NDIS compile with usb4bsd too.
However, testing results are pretty much as modest as with the old USB
stack. Maybe someone else has more luck.
New files taken from FreeBSD.
Sascha Wildner [Fri, 23 Nov 2012 18:46:19 +0000 (19:46 +0100)]
usb4bsd: Fix buildkernel with device usb4bsd.
Sascha Wildner [Fri, 23 Nov 2012 18:11:29 +0000 (19:11 +0100)]
wpa_supplicant.8: Document all options.
Reported-by: Max Herrgard <herrgard@gmail.com>
Taken-from: FreeBSD
Sascha Wildner [Fri, 23 Nov 2012 18:02:53 +0000 (19:02 +0100)]
wpa_supplicant.conf.5: Fix section order.
François Tigeot [Fri, 23 Nov 2012 06:57:26 +0000 (07:57 +0100)]
VFS quota: hpfs is now read-only, remove support
François Tigeot [Mon, 19 Nov 2012 09:30:34 +0000 (10:30 +0100)]
Documentation: unrhdr to idr conversion
Sascha Wildner [Thu, 22 Nov 2012 19:44:23 +0000 (20:44 +0100)]
vkernel: Make "options IPX" compilable into the vkernel.
Usage of "errno" in userspace as a variable name results in sad code.
Rename them to "error" instead.
Reported-by: tuxillo
Sascha Wildner [Thu, 22 Nov 2012 17:27:14 +0000 (18:27 +0100)]
kernel: Remove some NULL checks after kmalloc(..., M_WAITOK).
Sascha Wildner [Thu, 22 Nov 2012 17:24:47 +0000 (18:24 +0100)]
kernel: Use M_ZERO instead of bzero()ing.
Sascha Wildner [Thu, 22 Nov 2012 16:41:11 +0000 (17:41 +0100)]
ah[cd](4): Clean up some #ifdefs so it's clearer what we are doing.
Sepherosa Ziehau [Thu, 22 Nov 2012 12:53:36 +0000 (20:53 +0800)]
emx: Don't update TDT upon very packets transmitted
The TDT is updated only after certain amount of TX descriptors are
added to the hardware TX ring. The default value of the amount of
TX descriptors are 8. This value could be further tuned by per-device
sysctl node hw.emxX.tx_wreg_nsegs.
CPU usage when transmitting 18bytes tiny UDP datagrams @1.48Mpps with
82571 on AMD970 (3500Mhz) are reduced from 50% to 30% with the default
value.
Inspired-by: FreeBSD netmap's paper
Sepherosa Ziehau [Thu, 22 Nov 2012 12:02:31 +0000 (20:02 +0800)]
igb: Don't update TDT upon very packets transmitted
The TDT is updated only after certain amount of TX descriptors are
added to the hardware TX ring. The default value of the amount of
TX descriptors are 8. This value could be further tuned by per-device
sysctl node hw.igbX.tx_wreg_nsegs.
CPU usage when transmitting 18bytes tiny UDP datagrams @1.48Mpps with
82575 on AMD970 (3500Mhz) are reduced from 55% to 38% with the default
value.
Inspired-by: FreeBSD netmap's paper
Sepherosa Ziehau [Sat, 17 Nov 2012 10:07:21 +0000 (18:07 +0800)]
pktgen: Rework to improve performance using low CPU frequency
The performance improvement is measured on AMD970 operating at 800Mhz
w/ Intel 82575 NIC using 18bytes UDP datagrams.
- Don't refill and allocate new packets for each transmission, instead
a set of pregenerated packets are used.
- Use NIC driver's "txeof" to tick out more packets instead of bindly
trying to inject more packets into NIC's TX ring.
These above two modifications give me ~200Kpps (710Kpps -> 910Kpps)
transmission rate improvement.
- Don't use hardware TX checksum offloading feature. Since the set of
packets to be used are pregenerated, this will not hurt performance,
instead, this could improve performance on certain hardwares, e.g.
Intel's NIC chipsets which require additional TX descriptor for setting
up hardware TX checksum offload context.
The above modification give me additional ~40Kpps transmission rate
improvement.
Antonio Huete Jimenez [Thu, 22 Nov 2012 10:47:58 +0000 (11:47 +0100)]
hpfs - Fix a couple panics and a little cleanup.
* Fix compilation with HPFS_DEBUG.
* Fix a panic due CNP_PDIRUNLOCK flag not being cleared.
* Fix a panic where returned vnode after a lookup is not
NULL in the ENOENT case.
* Disable write support completely. It was pretty minimal
and operations like create or rename were not supported.
It has been tested with a filesystem created by OS/2 Warp 2.1.
Copying data out of it worked fine, but there is still an
outstanding issue with overlapping buffers.
Matthew Dillon [Wed, 21 Nov 2012 17:59:55 +0000 (09:59 -0800)]
kernel - Fix NFS panic when competing clients collide on hardlink
* Fix recursive execlusive lock on vnode.
* For example, if several boxes are nfs booting and have the same writable
/var/log, simultanious attempts to rotate logs can cause a panic due to
higher layers of the OS's hardlink code not being able to detect
duplicate vnodes which are created by some other client racing the same
operation.
* Adds CNP_NOTVP and cn_notvp flag which vop_compat_nlink() uses to notify
NFS's VOP_LOOKUP function to check for a vnode lookup collision.
-Matt
Sascha Wildner [Wed, 21 Nov 2012 17:13:43 +0000 (18:13 +0100)]
Remove some duplicated semicolons (mostly in the kernel).
Sascha Wildner [Wed, 21 Nov 2012 03:26:47 +0000 (04:26 +0100)]
hptiop(4): Add support for RocketRAID 4520 and 4522.
This adds support for HighPoint's recent 6Gb/s SAS/SATA RAID cards.
It's not tested in DragonFly, unfortunately, due to lack of hardware,
but I'm confident. :)
Taken-from: FreeBSD
Sepherosa Ziehau [Tue, 20 Nov 2012 12:26:52 +0000 (20:26 +0800)]
igb: Dispatch callout to the CPU that is checking status
This reduces sporadic contentions on the main serializer.
Sascha Wildner [Tue, 20 Nov 2012 05:23:49 +0000 (06:23 +0100)]
aac(4): Add a missing unlock.
Sascha Wildner [Tue, 20 Nov 2012 04:32:38 +0000 (05:32 +0100)]
iir(4): Clean up some unused code.
Antonio Huete Jimenez [Mon, 19 Nov 2012 22:32:16 +0000 (23:32 +0100)]
ext2fs - Clear CNP_PDIRUNLOCK flag after a lookup.
vop_compat_nresolve() unlocks the vnode when flag CNP_PDIRUNLOCK
is set, so either we clear the flag or avoid unlocking the vnode
after an ext2_lookup().
DragonFly-bug: http://bugs.dragonflybsd.org/issues/2454
Reported by: Skylar Hawk
Sascha Wildner [Mon, 19 Nov 2012 17:56:42 +0000 (18:56 +0100)]
Unbreak buildworld.
Reported-by: marino
John Marino [Mon, 19 Nov 2012 17:43:52 +0000 (18:43 +0100)]
/usr/Makefile: Add pkgsrc-create-shallow option
Sascha Wildner [Mon, 19 Nov 2012 16:59:05 +0000 (17:59 +0100)]
make distribution: Remove some symlinks before (re-)creating them.
Reported-by: marino
Sepherosa Ziehau [Mon, 19 Nov 2012 09:51:24 +0000 (17:51 +0800)]
netstat: Make netstat -rn more resilient to having the routing table changing
Obtained-from: FreeBSD 176289 by jhb
"Make netstat -rn more resilient to having the routing table change out from
under it while running. Note that this is still not perfect:
- Try to do something intelligent if kvm_read() fails to read a routing
table structure such as an rtentry, radix_node, or ifnet.
- Don't follow left and right node pointers in radix_nodes unless
RNF_ACTIVE is set in rn_flags. This avoids walking through freed
radix_nodes."
Sepherosa Ziehau [Mon, 19 Nov 2012 09:38:44 +0000 (17:38 +0800)]
mchian: Sync w/ FreeBSD a little bit
subr_mchain.c CVS 1.{6, 8, 9, 10, 16, 18}
Submitted-by: Alexey Slynko w/ modification by me
DragonFly-bug: http://bugs.dragonflybsd.org/issues/80
François Tigeot [Mon, 19 Nov 2012 08:48:47 +0000 (09:48 +0100)]
Porting: Document a callout_drain() replacement
Sascha Wildner [Sun, 18 Nov 2012 21:35:11 +0000 (22:35 +0100)]
kernel: Move DEVICE_POLLING to the bottom of 'options'.
Sascha Wildner [Sun, 18 Nov 2012 21:34:36 +0000 (22:34 +0100)]
kernel: Remove unused KTR_POLLING option.
Sascha Wildner [Sun, 18 Nov 2012 19:46:15 +0000 (20:46 +0100)]
mpt(4): Fix a panic in RAID operation.
I ported it wrongly. FreeBSD uses callout_init_mtx() here.
Sascha Wildner [Sat, 17 Nov 2012 21:58:30 +0000 (22:58 +0100)]
dehumanize_number(3): Add T, P and E suffix support.
Reported-by: ftigeot
Sascha Wildner [Sat, 17 Nov 2012 12:47:14 +0000 (13:47 +0100)]
mtree: Do no longer create /usr/include/objc.
Sascha Wildner [Sat, 17 Nov 2012 01:59:23 +0000 (02:59 +0100)]
Remove an unused header.
Sascha Wildner [Fri, 16 Nov 2012 23:49:42 +0000 (00:49 +0100)]
Clean up references to some obsolete functions in manual pages.
sigpause(2) -> sigsuspend(2)
sigsetmask(2)/sigblock(2) -> sigprocmask(2)
sigvec(2) -> sigaction(2)
alarm(3) -> setitimer(2)
pause(3) -> sigsuspend(2)
Based on similar changes in FreeBSD.
Sascha Wildner [Fri, 16 Nov 2012 17:19:02 +0000 (18:19 +0100)]
vfork.2: Document that vfork(2) returns pid_t.
Also adjust syscalls.master but it actually doesn't matter. These
adjustments are merely of a cosmetic nature.
Sascha Wildner [Fri, 16 Nov 2012 13:51:11 +0000 (14:51 +0100)]
nrelease/Makefile.usr: Relocate the git gc for pkgsrc to nrelease.
Since it was only meant for keeping the LiveDVD smaller, put it into
its build and don't force everyone to sit through it. The time it
takes seems to increase significantly on virtual machines with just
1G of RAM, for example. It seems to be quite memory hungry.
Reported-by: many
Sascha Wildner [Fri, 16 Nov 2012 12:19:38 +0000 (13:19 +0100)]
cp(1)/mv(1): Bring back a number of cleanups we did previously.
People (including myself) have put their time and thought into
this in the past, so we don't want this simply papered over with
the previous state.
Sascha Wildner [Fri, 16 Nov 2012 12:13:45 +0000 (13:13 +0100)]
Fix buildworld.
mv(1) is a bootstrap tool, hence it needs to build with host includes.
Sascha Wildner [Fri, 16 Nov 2012 10:42:05 +0000 (11:42 +0100)]
Update the pciconf(8) database.
November 14, 2012 snapshot from http://pciids.sourceforge.net/
John Marino [Fri, 16 Nov 2012 09:21:09 +0000 (10:21 +0100)]
/bin/mv: Sync with FreeBSD
1. Utilize central <paths.h> header
2. Meet posix spec for moving directory to existing non-empty directory
3. Meet posix spec for not moving directory over a file
4. Don't chop IO into pieces, use MAXPHYS
5. Add -h flag (similar to ln -h flag)
6. Test for input_is_terminal
7. Regression test added.
DETAILS
==========================================================================
1. remove pathnames.h file and add _PATH_RM to paths.h
2. Moving a directory to an existing non-empty directory will now fail as
required.
3. Prior to change this test past, but after change it failed on
cross-link devices. Original code added to check for case and return
-1 for the cross-link case.
4. Read files in one block and improve warn messages.
5. Using -h flag will force mv to treat a symbolic link to a directory
for the target as a symbolic link instead of a directory. This makes
it possible to atomically update a symbolic link using rename().
6. POSIX requires this behavior:
if (exists and (NOT f_options) AND
((not_writable AND input_is_terminal) OR i_option)) then
prompt
This behavior was implemented.
7. Passes all tests in suite. Prior to change it failed 2 of 32 tests.
Sepherosa Ziehau [Fri, 16 Nov 2012 08:59:53 +0000 (16:59 +0800)]
UPDATING: Mention DEVICE_POLLING is replaced by IFPOLL_ENABLE
Sepherosa Ziehau [Fri, 16 Nov 2012 08:56:19 +0000 (16:56 +0800)]
polling: Remove device_polling remainder
DEVICE_POLLING has no effect since this commit and will be nuked
after next release.
Sepherosa Ziehau [Fri, 16 Nov 2012 08:36:05 +0000 (16:36 +0800)]
polling: Remove device_polling support
Sepherosa Ziehau [Fri, 16 Nov 2012 08:35:32 +0000 (16:35 +0800)]
firewire: Mention IFPOLL_ENABLE instead of DEVICE_POLLING
Sepherosa Ziehau [Fri, 16 Nov 2012 07:50:46 +0000 (15:50 +0800)]
config: Remove device_polling from stock configures
Sepherosa Ziehau [Fri, 16 Nov 2012 07:46:01 +0000 (15:46 +0800)]
polling: Update manpage for device_polling to ifpoll switching
Sepherosa Ziehau [Fri, 16 Nov 2012 06:05:59 +0000 (14:05 +0800)]
wpi: This driver never supported polling
Sepherosa Ziehau [Fri, 16 Nov 2012 06:05:33 +0000 (14:05 +0800)]
iwn: This driver never supported polling
John Marino [Thu, 15 Nov 2012 23:31:35 +0000 (00:31 +0100)]
/bin/rm: fix removing symlinks with uchg/uappnd set
One error case described in FreeBSD PR bin/111226:
/bin/rm when running as super user and failing to unlink a
UF_APPEND|UF_IMMUTABLE protected symbolic link will reset the
UF_APPEND and UF_IMMUTABLE flags on the symbolic link's target (if
that target exists) - an object that /bin/rm should not touch! (Quote
from SUSv3: "The rm utility removes symbolic links themselves, not
the files they refer to, as a consequence of the dependence on the
unlink() functionality").
There was a failure case for /usr/bin/find as well, but swildner
recently synchronized find with FreeBSD and fixed that one.
Taken-from: FreeBSD SVN 193087 (30 MAY 2009)
John Marino [Thu, 15 Nov 2012 21:46:29 +0000 (22:46 +0100)]
libc/fts: Use O_CLOEXEC for internal file descriptors
Because fts keeps internal file descriptors open across calls, making such
descriptors close-on-exec helps not only multi-threaded applications but
also single-threaded applications.
In particular, this prevents passing a temporary file descriptor for saving
the current directory to processes created via find -exec.
Taken-from: FreeBSD SVN 241010 (27 SEP 2012)
John Marino [Thu, 15 Nov 2012 20:57:01 +0000 (21:57 +0100)]
bug 2443: bin/cp: sync with FreeBSD (bug fixes + new options)
* Fix FTS_NOCHDIR behavior on empty directory.
* Add -a option (archive mode, equivalent to -RpP)
* Add -l option (create hard links instead of copying)
* Add -x option (FS mount points are not traversed)
* Preserve file flags on symlinks on cp -Rp
* Don't copy socket anyway after "xxx is a socket (not copied)"
* Avoid division-by-zero situations
* make -r synonym for -R but maintain -r behavior with -L
Taken-from: FreeBSD (multiple commits)
https://bugs.dragonflybsd.org/issues/2443
Matthew Dillon [Thu, 15 Nov 2012 20:30:51 +0000 (12:30 -0800)]
kernel - Fix off-by-1 error in semexit
* Fix an off-by-1 error that was preventing semexit from properly undoing
all the semaphores controlled by the exiting process.
Reported-by: lentferj, vsrinivas, others
Antonio Huete Jimenez [Thu, 15 Nov 2012 19:13:03 +0000 (20:13 +0100)]
share/misc/gdbinit - New function lstdtok
- List tokens for a specified thread.
Sascha Wildner [Wed, 14 Nov 2012 22:57:04 +0000 (23:57 +0100)]
init(8): Resurrect kern.shutdown_timeout as kern.init_shutdown_timeout.
Taken from FreeBSD. Quoting their commit msg:
Fix system shutdown timeout handling by again supporting longer running
shutdown procedures (which have a duration of more than 120 seconds).
We have two user-space affecting shutdown timeouts: a "soft" one in
/etc/rc.shutdown and a "hard" one in init(8). The first one can be
configured via /etc/rc.conf variable "rcshutdown_timeout" and defaults
to 30 seconds. The second one was originally (in 1998) intended to be
configured via sysctl(8) variable "kern.shutdown_timeout" and defaults
to 120 seconds.
Unfortunately, the "kern.shutdown_timeout" was declared "unused" in 1999
(as it obviously is actually not used within the kernel itself) and
hence was intentionally but misleadingly removed in revision 1.107 from
init_main.c. Kernel sysctl(8) variables are certainly a wrong way to
control user-space processes in general, but in this particular case the
sysctl(8) variable should have remained as it supports init(8), which
isn't passed command line flags (which in turn could have been set via
/etc/rc.conf), etc.
As there is already a similar "kern.init_path" sysctl(8) variable which
directly affects init(8), resurrect the init(8) shutdown timeout under
sysctl(8) variable "kern.init_shutdown_timeout". But this time document
it as being intentionally unused within the kernel and used by init(8).
Also document it in the manpages init(8) and rc.conf(5).
Sascha Wildner [Wed, 14 Nov 2012 22:14:32 +0000 (23:14 +0100)]
libk{core,info}: The kern.ccpu sysctl is gone, so remove *get_sched_ccpu().
Interestingly, kcore_get_sched_ccpu() seems to never have returned ccpu,
but instead tk_nin, which is not related.
Sascha Wildner [Wed, 14 Nov 2012 21:24:23 +0000 (22:24 +0100)]
top(1): Remove some code that is unused after the UP removal.
Sascha Wildner [Wed, 24 Oct 2012 19:20:36 +0000 (21:20 +0200)]
<sys/proc.h>: Adjust the names of two unused bits to use bit numbering.
Sascha Wildner [Wed, 14 Nov 2012 10:06:26 +0000 (11:06 +0100)]
mount{,_ufs}: Fix wrong mountctl(2) error check.
mountctl(2) returns the number of copied bytes when successful, and -1
if there was an error. This prevented ", local" from being printed in
mount output.
Reported-by: YONETANI Tomokazu
Sascha Wildner [Tue, 13 Nov 2012 18:41:51 +0000 (19:41 +0100)]
Sync zoneinfo database with tzdata2012j from ftp://ftp.iana.org/tz/releases
* africa: Lybia set their time one hour back at 02:00 on November 10
(last Saturday).
* Additional miscellaneous fixes in various comments.
Sascha Wildner [Tue, 13 Nov 2012 18:39:40 +0000 (19:39 +0100)]
tzfile.5: Little wording fix (from tzcode2012j).
Sascha Wildner [Tue, 13 Nov 2012 17:35:17 +0000 (18:35 +0100)]
ipcs(1): Fix the "ipcs: kvm_nlist: No such file or directory" error.
The "semu" variable was removed in
79757ed082, causing this breakage.
Strangely, even before its removal it wasn't used for anything in
ipcs(1).
Reported-by: beket
Sascha Wildner [Tue, 13 Nov 2012 17:13:30 +0000 (18:13 +0100)]
ipcs(8): Raise WARNS to 6.
Sascha Wildner [Mon, 12 Nov 2012 20:59:30 +0000 (21:59 +0100)]
Really fix systat(1).
Sepherosa Ziehau [Mon, 12 Nov 2012 12:34:39 +0000 (20:34 +0800)]
ifconfig: Deprecate "pollcpu" and make "polling" same as "polling"
Sepherosa Ziehau [Mon, 12 Nov 2012 12:21:31 +0000 (20:21 +0800)]
fwe: Switch from device_polling to ifpoll
Antonio Huete Jimenez [Sun, 11 Nov 2012 23:18:01 +0000 (00:18 +0100)]
systat - Do not use ccpu for %cpu calculation in pigs view.
- ccpu went away after the new scheduler work so sync with
the current status of the kernel.
- Also remove all the unused kvm* stuff and get fscale via sysctl.
Reported-by: tpchin <tpchin@telstra.com>
Dragonfly-bug: <http://bugs.dragonflybsd.org/issues/2245>
Sascha Wildner [Sun, 11 Nov 2012 20:51:34 +0000 (21:51 +0100)]
Remove an ancient file that is no longer used.
Sascha Wildner [Sun, 11 Nov 2012 11:30:20 +0000 (12:30 +0100)]
loader: Detect SMBIOS and export its information into the environment.
Useful to check motherboard vendor/version, BIOS version and so forth.
Use kenv(1) for viewing the information.
This information can be accessed from the kernel but should only be
as a last resort.
Might also help in the future to better detect virtual machine types.
Taken-from: FreeBSD
Sepherosa Ziehau [Sun, 11 Nov 2012 11:10:50 +0000 (19:10 +0800)]
vge: Switch from device_polling to ifpoll
Sepherosa Ziehau [Sun, 11 Nov 2012 10:20:17 +0000 (18:20 +0800)]
nge: Switch from device_polling to ifpoll
Sepherosa Ziehau [Sun, 11 Nov 2012 09:44:04 +0000 (17:44 +0800)]
dc: Add missing ifpoll_compat_setup
Sepherosa Ziehau [Sun, 11 Nov 2012 09:29:24 +0000 (17:29 +0800)]
dc: Switch from device_polling to ifpoll
Sascha Wildner [Sun, 11 Nov 2012 09:26:17 +0000 (10:26 +0100)]
hammer.8: Add back an .El that was removed by mistake.
Sascha Wildner [Sat, 10 Nov 2012 10:03:17 +0000 (11:03 +0100)]
rpc.statd(8): Remove a duplicate function.
By now, xdr_stat_chge() is in librpcsvc, unlike what the comments suggest.
Adam Sakareassen [Thu, 8 Nov 2012 15:46:00 +0000 (02:46 +1100)]
Fix INVARIANT build issue
François Tigeot [Fri, 9 Nov 2012 23:07:51 +0000 (00:07 +0100)]
vquota(8): size information is expressed in bytes
Sascha Wildner [Fri, 9 Nov 2012 23:19:30 +0000 (00:19 +0100)]
Mark some functions in userland __printflike.
Sascha Wildner [Fri, 9 Nov 2012 23:19:05 +0000 (00:19 +0100)]
ndis_events(8): Remove a bogus argument.
John Marino [Thu, 8 Nov 2012 10:46:55 +0000 (11:46 +0100)]
bsd.init.mk: Establish compatibility with FreeBSD makefiles
FreeBSD changed NOMAN to NO_MAN and NOSHARED to NO_SHARED. Rather than
follow suit, establish NO_MAN and NO_SHARED as aliases for the DragonFly
versions. If both are defined, the DragonFly version takes precedence.
These were put in bsd.init.mk because they have to be loaded after the
makefile being processed but before bsd.prog.mk, bsd.lib.mk, etc.
Antonio Huete Jimenez [Thu, 8 Nov 2012 15:17:32 +0000 (16:17 +0100)]
debug - Add a basic Makefile
- Sync wildcardinfo with current state of affairs.
Stathis Kamperis [Thu, 8 Nov 2012 16:18:29 +0000 (18:18 +0200)]
objcache.9: cluster_limit is no longer changed
Sascha Wildner [Thu, 8 Nov 2012 08:50:27 +0000 (09:50 +0100)]
kernel: Mark two more functions __printflike.
Sascha Wildner [Thu, 8 Nov 2012 08:28:04 +0000 (09:28 +0100)]
mpt(4): Mark two functions __printflike.
John Marino [Wed, 7 Nov 2012 20:15:35 +0000 (21:15 +0100)]
getline(3): Fix segfault caused by NULL pointer
This fixes a segfault seen with pkg's audit function, and effectively syncs
getdelim with FreeBSD.
Taken-from: FreeBSD SVN 197752 (04 OCT 2009)
- Tolerate applications that pass a NULL pointer for the buffer and
claim that the capacity of the buffer is nonzero.
- If an application passes in a non-NULL buffer pointer and claims the
buffer has zero capacity, we should free (well, realloc) it anyway.
It could have been obtained from malloc(0), so failing to free it
would be a small memory leak.
Sascha Wildner [Wed, 7 Nov 2012 21:10:51 +0000 (22:10 +0100)]
mfi(4): Don't assert lock holding when dumping or during a panic.
FreeBSD's mtx_assert() does it the same way, but we usually translate
mtx_assert() into a KKASSERT(lockstatus(&lock, curthread) != 0).
For now, replace all of these assertions in mfi(4) with a function that
mimics the mtx_assert() behavior of FreeBSD, since some of the asserting
functions are used by the dumping code.
We might think about a real lockassert() function in the future that
has this exception too.
This commit fixes dumping on most supported adapters. Thunderbolt series
adapters still have an (unrelated) command timeout issue which I'm
looking at.
Reported and tested (on a PERC 5/i) by Tim Howe <tim@130collective.org>.
Sepherosa Ziehau [Wed, 7 Nov 2012 13:47:59 +0000 (21:47 +0800)]
igb: Configure TX head write back before turn enable bit in TXDCTL
The original configure sequence works on 82575 and 82576. However,
it does not work on i350 and probably will not work on 82580. For
these two chips TX head write back _must_ be configured before
turning on enable bit in TXDCTL according to their datasheet.
Reported-by: ftigeot@
Tested-by: ftigeot@ (i350), me w/ (82575, 82576)
Sepherosa Ziehau [Tue, 6 Nov 2012 14:19:10 +0000 (22:19 +0800)]
vr: Switch from device_polling to ifpoll
Sepherosa Ziehau [Tue, 6 Nov 2012 11:58:19 +0000 (19:58 +0800)]
fxp: Switch device_polling to ifpoll
Sepherosa Ziehau [Mon, 5 Nov 2012 13:31:43 +0000 (21:31 +0800)]
rl: Switch from device_polling to ifpoll
Optimize the RX path a little bit by reading the RL_CURRXADDR and
RL_CURRXBUF only if necessary; reading these two registers is quite
time consuming
Sepherosa Ziehau [Mon, 5 Nov 2012 13:30:19 +0000 (21:30 +0800)]
ifpoll: Closely follow the old status polling fraction in the compat code
Sascha Wildner [Tue, 6 Nov 2012 06:06:41 +0000 (07:06 +0100)]
ndiscvt(8): Add a missing space.
Submitted-by: Eitan Adler <lists@eitanadler.com>
Sascha Wildner [Tue, 6 Nov 2012 06:05:14 +0000 (07:05 +0100)]
hptrr(4): Add missing intrhook description.
Sascha Wildner [Tue, 6 Nov 2012 06:04:59 +0000 (07:04 +0100)]
hpt27xx(4): Add missing bootverbose and intrhook description.
Sascha Wildner [Tue, 6 Nov 2012 02:44:40 +0000 (03:44 +0100)]
Fix some typos in user visible messages, etc.
Sascha Wildner [Mon, 5 Nov 2012 08:27:54 +0000 (09:27 +0100)]
hexdump.3: Add missing .Sh LIBRARY and bump .Dd to when it was imported.
Sascha Wildner [Mon, 5 Nov 2012 05:28:03 +0000 (06:28 +0100)]
bsd-family-tree: Sync with FreeBSD (adds DragonFly 3.2 and OpenBSD 5.2).
Sascha Wildner [Mon, 5 Nov 2012 04:35:26 +0000 (05:35 +0100)]
sysctl.9: Fix another typo. frequence -> frequency
Found-with: igor