dragonfly.git
5 years agoether_input: Defer promiscous packet discarding until vlan is processed
Sepherosa Ziehau [Sun, 5 Apr 2009 03:38:23 +0000 (11:38 +0800)]
ether_input: Defer promiscous packet discarding until vlan is processed

This is the one of the precondition to make vlan/bridge work: preventing
packets from being dropped prematurely by vlan's parent interface.

5 years agoUse ETHER_IS_MULTICAST, which is more expressive.
Sepherosa Ziehau [Sun, 5 Apr 2009 02:57:27 +0000 (10:57 +0800)]
Use ETHER_IS_MULTICAST, which is more expressive.

5 years agoserializer manpage: Remove IS_SERIALIZED(), which is available only if
Sepherosa Ziehau [Sun, 5 Apr 2009 02:03:26 +0000 (10:03 +0800)]
serializer manpage: Remove IS_SERIALIZED(), which is available only if
INVARIANTS is defined and probably should not be used in any places
other than the serializer implementation itself.

5 years agombuf: Don't discount the objcache limits required by non-pkthdr mbuf
Sepherosa Ziehau [Sun, 5 Apr 2009 02:00:04 +0000 (10:00 +0800)]
mbuf: Don't discount the objcache limits required by non-pkthdr mbuf

5 years agoaltq: ifq needs protection here; not ifnet
Sepherosa Ziehau [Sat, 4 Apr 2009 08:51:40 +0000 (16:51 +0800)]
altq: ifq needs protection here; not ifnet

5 years agoumount(8): Raise WARNS to 6 and fix all warnings.
Sascha Wildner [Sat, 4 Apr 2009 20:50:59 +0000 (22:50 +0200)]
umount(8): Raise WARNS to 6 and fix all warnings.

5 years agosh(1): Fix warnings when compiling with -DNO_HISTORY.
Sascha Wildner [Sat, 4 Apr 2009 20:38:31 +0000 (22:38 +0200)]
sh(1): Fix warnings when compiling with -DNO_HISTORY.

5 years agovkernel: Comment out unused variable.
Sascha Wildner [Sat, 4 Apr 2009 20:31:06 +0000 (22:31 +0200)]
vkernel: Comment out unused variable.

5 years agoRename some log() -> dolog() to not conflict with built-in log().
Sascha Wildner [Sat, 4 Apr 2009 20:26:27 +0000 (22:26 +0200)]
Rename some log() -> dolog() to not conflict with built-in log().

While here, remove trailing whitespace.

5 years agosystat(1): Correct TCP statistics reporting
Sepherosa Ziehau [Sat, 4 Apr 2009 02:43:27 +0000 (10:43 +0800)]
systat(1): Correct TCP statistics reporting

Reported-by: ruse39 on EFnet #dragonflybsd
5 years agobridge(4): if_snd.ifq_maxlen does not need explicit setting
Sepherosa Ziehau [Sat, 4 Apr 2009 02:33:04 +0000 (10:33 +0800)]
bridge(4): if_snd.ifq_maxlen does not need explicit setting

5 years agoHAMMER - Fix extra TAILQ_REMOVE in HAMMER shutdown code.
Matthew Dillon [Sat, 4 Apr 2009 01:35:07 +0000 (18:35 -0700)]
HAMMER - Fix extra TAILQ_REMOVE in HAMMER shutdown code.

Caught-by: New assertions in sys/queue.h imported by pavalos.
5 years agoAMD64 Test Environment - add SDL support to the qemu build
Matthew Dillon [Sat, 4 Apr 2009 01:12:12 +0000 (18:12 -0700)]
AMD64 Test Environment - add SDL support to the qemu build

Qemu may now be built with SDL (X support) using the make directive
WITHSDL=YES.  A new target 'xrun' runs qemu in X mode vs curses mode.

Submitted-by: Antonio Huete
5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Fri, 3 Apr 2009 22:12:02 +0000 (15:12 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agoAMD64 - Enhance help target for src/test/amd64 build
Matthew Dillon [Fri, 3 Apr 2009 22:10:56 +0000 (15:10 -0700)]
AMD64 - Enhance help target for src/test/amd64 build

Document all targets.  Also add a new target 'sysloader' which will
install the native 32 bit system loader in root.img in case the
64 bit one doesn't work.

5 years agoUse .Va for tunables in manual pages.
Sascha Wildner [Fri, 3 Apr 2009 18:13:15 +0000 (20:13 +0200)]
Use .Va for tunables in manual pages.

5 years agoRedo last commit. Do not require that SRCDIR be specified. Set it
Matthew Dillon [Fri, 3 Apr 2009 18:00:39 +0000 (11:00 -0700)]
Redo last commit.  Do not require that SRCDIR be specified.  Set it
to ${.CURDIR}/../.. by default (originally it was hardwired to /usr/src).

5 years agoRequire that SRCDIR be specified or exist in the environment instead of
Matthew Dillon [Fri, 3 Apr 2009 17:56:41 +0000 (10:56 -0700)]
Require that SRCDIR be specified or exist in the environment instead of
defaulting it to /usr/src.

5 years agoRevoke reimplementation - primary commit part 2/2.
Matthew Dillon [Fri, 3 Apr 2009 16:46:57 +0000 (09:46 -0700)]
Revoke reimplementation - primary commit part 2/2.

This finishes out the primary revoke changes.  Adjust the revoke code
to do a primary pass on the file pointer table and mark fps with FREVOKED,
then do a pass on the process descriptor tables to clean them out.

Add checks to dup() to silently use a dummy fp when the source fp has
been revoked.  This removes revoke/dup races.

Add unix domain in-transit tests.  If a revoked fp is in-transit then
unix-domain sockets are scanned and revoked file pointers replaced with
dummy file pointers.

5 years agoAdd programs to test revocation of open files in the presence of
Matthew Dillon [Fri, 3 Apr 2009 16:24:13 +0000 (09:24 -0700)]
Add programs to test revocation of open files in the presence of
unix domain sockets with in-transit descriptors.

5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Fri, 3 Apr 2009 06:48:28 +0000 (23:48 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agoRevoke reimplementation - primary commit part 1/2.
Matthew Dillon [Fri, 3 Apr 2009 06:38:31 +0000 (23:38 -0700)]
Revoke reimplementation - primary commit part 1/2.

Rip out and replace the old revoke code.  The old code basically ripped
vnodes out from under the related VFSs, faked close operations, hacked
up ref counts, and generally wreaked havoc on the system.

The new code is not as efficient, but plays a lot nicer with the kernel.
It locates and closes matching file descriptors (replacing the descriptor
with a dummy that just returns an error).  No ref count hacks are needed
and vnodes are no longer unceremoniously canned.

The new revoke code is also capable of revoking any vnode in the system,
device or file.  root, jail root, and chrooted directories cannot be
revoked.

This also fixes an issue with HAMMER getting confused when one of its
vnodes got ripped out from under it by the old revoke.

5 years agousched_dummy: Silence warning
Sascha Wildner [Fri, 3 Apr 2009 06:34:24 +0000 (08:34 +0200)]
usched_dummy: Silence warning

5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Thu, 2 Apr 2009 20:03:06 +0000 (13:03 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agoAMD64 - Fixes to make dynamic binaries run
Matthew Dillon [Thu, 2 Apr 2009 19:59:29 +0000 (12:59 -0700)]
AMD64 - Fixes to make dynamic binaries run

Fix the Elf64_Auxargs structure.

Misc stuff - Supply the original stack pointer as arg3 to rtld_bind().
Set addr = 0 if no interpreter (static binary).  Enable KTRACE.
Remove unused COPY_SIGCODE.

Taken-from: Loosely taken from FreeBSD, with help from Jordan Gordeev

5 years agovnconfig(8): remove code with no effect
Stathis Kamperis [Thu, 2 Apr 2009 17:37:57 +0000 (17:37 +0000)]
vnconfig(8): remove code with no effect

There is no point in calling anything after an err() call,
since err() itself exits and never fails.

Reviewed-by: dillon@
5 years agopppstats(8): Convert manual page to mdoc.
Sascha Wildner [Thu, 2 Apr 2009 09:17:11 +0000 (11:17 +0200)]
pppstats(8): Convert manual page to mdoc.

5 years agoAMD64 - More signal handler work. Signals now appear to work.
Matthew Dillon [Thu, 2 Apr 2009 06:23:50 +0000 (23:23 -0700)]
AMD64 - More signal handler work.  Signals now appear to work.

Finish implementing signal handler argument construction.  For AMD64 the
arguments are passed in registers rather than on the stack.

Align the stack to 16 bytes.

Misc cleanup, comments, and enable PROCFS in kernel config JG64.

Taken-from: FreeBSD code used as a loose template.

5 years agoFix permissions check for utimes() - owner can call even if u-w.
Matt Dillon [Thu, 2 Apr 2009 03:56:49 +0000 (03:56 +0000)]
Fix permissions check for utimes() - owner can call even if u-w.

The owner of a file can call utimes() to modify the file's access and
modified times even if the file is not user-writable.  Add a VOWN test
for VOP_ACCESS() / naccess() which performs this check.

As far as I know only kern_utimes() needs to use this particular type
of test.

This also fixes issues with cp -p and a few other programs.

Reported-by: Johannes Hofmann <johannes.hofmann@gmx.de>
5 years agoAMD64 - Remove some debugging.
Matthew Dillon [Thu, 2 Apr 2009 01:07:11 +0000 (18:07 -0700)]
AMD64 - Remove some debugging.

5 years agoAMD64 - Make signals operational, fix reg mappings, fix %fs management, trace
Matthew Dillon [Thu, 2 Apr 2009 00:57:01 +0000 (17:57 -0700)]
AMD64 - Make signals operational, fix reg mappings, fix %fs management, trace

Adjust sigframe, trapframe, mcontext, ucontext, and regs.  Add tf_xflags
too all structures.  Reorder struct regs to match the register layout
in the other structures.

Implement the commented out signaling code.  Signals now work, or at least
do not crash programs.  Theoretically the FP state is also saved and restored.

The exec() code failed to adjust gd_user_fs and gd_user_gs when setting
the msr registers for the user %fs and %gs, causing %fs to unexpectedly
change in running user programs.

Implement trace/debug support functions to set %rip and to single-step.

Define the missing vkernel flag FP_SOFTFP.

5 years agoAMD64 - Minor rtld syncs from FreeBSD
Matthew Dillon [Thu, 2 Apr 2009 00:52:31 +0000 (17:52 -0700)]
AMD64 - Minor rtld syncs from FreeBSD

Mainly change Elf_Word to Elf_Size in _rtld_bind(), plus some other
minor updates.

Taken-from: FreeBSD

5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Wed, 1 Apr 2009 21:29:40 +0000 (14:29 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agoCorrect elf hash element types for 64 bit ELF.
Matthew Dillon [Wed, 1 Apr 2009 21:27:01 +0000 (14:27 -0700)]
Correct elf hash element types for 64 bit ELF.

Suggested-by: smtms
Taken-from: FreeBSD

5 years agoFix quickworld for crossbuilding - use ln -hsf to replace 'cpu' softlink
Matthew Dillon [Wed, 1 Apr 2009 20:07:07 +0000 (13:07 -0700)]
Fix quickworld for crossbuilding - use ln -hsf to replace 'cpu' softlink

5 years agoamd64 test builds - add additional targets
Matthew Dillon [Wed, 1 Apr 2009 20:04:21 +0000 (13:04 -0700)]
amd64 test builds - add additional targets

Add numerous additional targets including an incremental buildworld
feature and gdb64 build to examine 64 bit binaries on 32 bit platforms.

5 years agoRemove CVS ID from the loader help file.
Sascha Wildner [Wed, 1 Apr 2009 13:08:03 +0000 (15:08 +0200)]
Remove CVS ID from the loader help file.

5 years agousched_dummy: just proceed if the current thread is trying to acquire curproc
Simon Schubert [Wed, 1 Apr 2009 09:26:51 +0000 (11:26 +0200)]
usched_dummy:  just proceed if the current thread is trying to acquire curproc

If the current thread is exiting to userland and there is no
rescheduling required, uschedcp will be lp, not NULL, so accept this
situation as well.

5 years agoSimplify the radeon microcode loading.
Hasso Tepper [Wed, 1 Apr 2009 05:50:56 +0000 (08:50 +0300)]
Simplify the radeon microcode loading.

Obtained-from: FreeBSD

5 years agoAdd a comment for reminder about nocache.
Hasso Tepper [Tue, 31 Mar 2009 17:07:06 +0000 (20:07 +0300)]
Add a comment for reminder about nocache.

5 years agoLoad the right microcode for RS780.
Hasso Tepper [Tue, 31 Mar 2009 17:03:03 +0000 (20:03 +0300)]
Load the right microcode for RS780.

Obtained-from: FreeBSD

5 years agoFix typo in cpdup.1.
Peter Avalos [Wed, 1 Apr 2009 03:55:41 +0000 (03:55 +0000)]
Fix typo in cpdup.1.

5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Wed, 1 Apr 2009 03:34:54 +0000 (20:34 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agovkernel: Fix startup race with I/O cothread support.
Matthew Dillon [Wed, 1 Apr 2009 03:32:22 +0000 (20:32 -0700)]
vkernel: Fix startup race with I/O cothread support.

The cothread interlocks using pthread calls but there is a small window
of opportunity where a cpu thread can signal the cothread before the
cothread has managed to interlock its first wait.  Moving the wait from the
top of the loop to the bottom fixes the problem.

Reported-by: Joe Talbott <josepht@cstone.net>
5 years agocpdup: '#' in cpignore files is no longer interpreted as a comment.
Matthew Dillon [Wed, 1 Apr 2009 00:16:31 +0000 (17:16 -0700)]
cpdup: '#' in cpignore files is no longer interpreted as a comment.

A '#' as the first character on a line in a cpignore file used to be
interpreted as a comment.  It is no just considered another valid
character in a filename.  Comments may no longer be represented in
cpignore files.

5 years agoUse atomic_fetchadd_int() in fdrop() instead of the spin-lock.
Matthew Dillon [Tue, 31 Mar 2009 23:56:14 +0000 (16:56 -0700)]
Use atomic_fetchadd_int() in fdrop() instead of the spin-lock.

5 years agoAdjust vfinddev() to return a referenced vnode in *vpp.
Matthew Dillon [Tue, 31 Mar 2009 22:17:33 +0000 (15:17 -0700)]
Adjust vfinddev() to return a referenced vnode in *vpp.

Fix a small race condition in sys_linux_ustat() related to accessing
a vnode which can get ripped out from under the procedure.

5 years agoMarkup fixes in several manual pages.
Sascha Wildner [Tue, 31 Mar 2009 22:14:12 +0000 (00:14 +0200)]
Markup fixes in several manual pages.

5 years agovnconfig(8): document -S default size modifier
Stathis Kamperis [Tue, 31 Mar 2009 07:26:17 +0000 (07:26 +0000)]
vnconfig(8): document -S default size modifier

5 years agoSort SEE ALSO in some manual pages.
Sascha Wildner [Mon, 30 Mar 2009 10:50:26 +0000 (12:50 +0200)]
Sort SEE ALSO in some manual pages.

5 years agovnconfig: remove unnecessary casts
Stathis Kamperis [Mon, 30 Mar 2009 09:57:36 +0000 (09:57 +0000)]
vnconfig: remove unnecessary casts

Approved-by: corecode@
5 years agovnconfig -l: print an error if supplied vn is unknown
Stathis Kamperis [Mon, 30 Mar 2009 07:48:36 +0000 (07:48 +0000)]
vnconfig -l: print an error if supplied vn is unknown

The code as it is has no effect, because `tmp' is never NULL.
Make it behave as initially intended. Also, since we are here
replace the printf() call with errx().

Approved-by: corecode@
5 years agorpc.rstatd(8): Silence warnings.
Sascha Wildner [Sun, 29 Mar 2009 16:00:23 +0000 (18:00 +0200)]
rpc.rstatd(8): Silence warnings.

5 years ago<sys/tree.h>: Add __unused (fixes warning in undo(1)).
Sascha Wildner [Sun, 29 Mar 2009 16:04:26 +0000 (18:04 +0200)]
<sys/tree.h>: Add __unused (fixes warning in undo(1)).

5 years agorpc.rusersd(8): Silence warnings.
Sascha Wildner [Sun, 29 Mar 2009 16:08:10 +0000 (18:08 +0200)]
rpc.rusersd(8): Silence warnings.

5 years agorpc.rwalld(8): Silence warnings.
Sascha Wildner [Sun, 29 Mar 2009 16:16:45 +0000 (18:16 +0200)]
rpc.rwalld(8): Silence warnings.

5 years agorpc.sprayd(8): Silence warnings.
Sascha Wildner [Sun, 29 Mar 2009 16:18:36 +0000 (18:18 +0200)]
rpc.sprayd(8): Silence warnings.

5 years agorpc.lockd(8): Silence warnings.
Sascha Wildner [Sun, 29 Mar 2009 17:52:21 +0000 (19:52 +0200)]
rpc.lockd(8): Silence warnings.

5 years agorpc.statd(8): Silence warnings.
Sascha Wildner [Sun, 29 Mar 2009 17:56:50 +0000 (19:56 +0200)]
rpc.statd(8): Silence warnings.

5 years agorpc.ypxfrd(8): Silence warnings.
Sascha Wildner [Sun, 29 Mar 2009 17:58:31 +0000 (19:58 +0200)]
rpc.ypxfrd(8): Silence warnings.

5 years agoyp_mkdb(8): Silence warnings.
Sascha Wildner [Sun, 29 Mar 2009 17:59:48 +0000 (19:59 +0200)]
yp_mkdb(8): Silence warnings.

5 years agolib/libc/yp: Silence warnings.
Sascha Wildner [Sun, 29 Mar 2009 19:56:27 +0000 (21:56 +0200)]
lib/libc/yp: Silence warnings.

5 years agoypset(8): Silence warnings.
Sascha Wildner [Sun, 29 Mar 2009 20:03:14 +0000 (22:03 +0200)]
ypset(8): Silence warnings.

5 years agomountd(8): Silence warnings.
Sascha Wildner [Sun, 29 Mar 2009 20:03:31 +0000 (22:03 +0200)]
mountd(8): Silence warnings.

5 years agoAdjust patch for qemu build - increase size of gen_opc_buf
Matthew Dillon [Sun, 29 Mar 2009 17:29:40 +0000 (10:29 -0700)]
Adjust patch for qemu build - increase size of gen_opc_buf

Work around a memory tromping bug in qemu 0.10.0 and 0.10.1 by
expanding the gen_opc_buf[] array.  I'm not sure where the bug
actually is, but this makes qemu work for us.

5 years agoAdd an indexing feature to the -t option.
Matthew Dillon [Sun, 29 Mar 2009 07:44:30 +0000 (00:44 -0700)]
Add an indexing feature to the -t option.

Add an indexing feature to the -t option where -t 0 indicates the
latest version and higher values indicate successively older versions.
Transaction ids are specified with 0x.

Automatically go into DIFF mode (-d) by default if two -t options are
specified, unless overridden.

Submitted-by: "Joel K. Pettersson" <joelkpettersson@gmail.com>
5 years agoFix an IPI messaging lockup which can occur while in DDB.
Matthew Dillon [Sun, 29 Mar 2009 05:07:42 +0000 (22:07 -0700)]
Fix an IPI messaging lockup which can occur while in DDB.

When entering DDB on a SMP system we must also enter a critical section
prior to stopping the other cpus.  If we do not interrupts may generate
calls to sched_ithd() instead of just flagging the interrupt in ipending
and returning.  This, in turn, may attempt to send an IPI to another
cpu.  These IPIs will not be processed while the target cpu is stopped.
If the IPI FIFO then fills up the lwkt_send_ipiq*() function will hard-loop
and the system will lock-up.

Resizing the window a vkernel is running in while the vkernel is in DDB
exercises the bug.  The bug is more easily demonstrated simply by sending
a stream of SIGWINCH signals via a while(1)/kill loop to the vkernel.

The bug is potentially an issue on all architectures so fix all
architectures.

Reported-by: Joe Talbott <josepht@cstone.net>
With-help-from: YONETANI Tomokazu <qhwt+dfly@les.ath.cx>

5 years agoobjcache(9): Fix wrong function name in the manual page.
Sascha Wildner [Sat, 28 Mar 2009 07:30:06 +0000 (08:30 +0100)]
objcache(9): Fix wrong function name in the manual page.

5 years agovkernel(7): Start sentences on new lines.
Sascha Wildner [Fri, 27 Mar 2009 21:27:38 +0000 (22:27 +0100)]
vkernel(7): Start sentences on new lines.

5 years agomakekey(8): Ansify and remove 'register'.
Sascha Wildner [Fri, 27 Mar 2009 21:02:01 +0000 (22:02 +0100)]
makekey(8): Ansify and remove 'register'.

5 years agomknetid(8): Raise WARNS to 6 and fix warnings (also -Wold-style-definition).
Sascha Wildner [Fri, 27 Mar 2009 20:57:11 +0000 (21:57 +0100)]
mknetid(8): Raise WARNS to 6 and fix warnings (also -Wold-style-definition).

5 years agoddb(4): mention 'show ktr' command
Stathis Kamperis [Sat, 28 Mar 2009 18:02:33 +0000 (18:02 +0000)]
ddb(4): mention 'show ktr' command

5 years agoipflow: Add macros to check ipflow refcnt, which is more expressive
Sepherosa Ziehau [Sat, 28 Mar 2009 06:20:28 +0000 (14:20 +0800)]
ipflow: Add macros to check ipflow refcnt, which is more expressive

5 years agoipflow: React to local address change quickly.
Sepherosa Ziehau [Sat, 28 Mar 2009 04:25:50 +0000 (12:25 +0800)]
ipflow: React to local address change quickly.

5 years agoipflow: Flush ipflow if a new route is successfully added.
Sepherosa Ziehau [Sat, 28 Mar 2009 03:02:45 +0000 (11:02 +0800)]
ipflow: Flush ipflow if a new route is successfully added.

ipflow suffers same problem as described in FreeBSD-PR: kern/10778
Though the time of mis-routing will not be "arbitrarily long time"
for intermittent packet flow (thanks to the ipflow timeout), it
still could be quite long for a constant packet flow.

5 years agoAdd a build system for amd64 build/test environments in /usr/src/test/amd64
Matthew Dillon [Sat, 28 Mar 2009 03:32:37 +0000 (20:32 -0700)]
Add a build system for amd64 build/test environments in /usr/src/test/amd64

This build system will fetch, compile, and install qemu, do all
necessary buildworlds and buildkernels, including the amd64 cross-builds,
and generate a fully bootable root.img.  The Makefile also includes a
run target to run qemu on the image.

It can take a few hours to build the whole mess due to the buildworlds but
it's a hit-the-start-button-and-forget build which should prove useful.

WARNING: Initial commit only.  qemu 0.10.{0,1} is seriously broken and is not
yet able to fully-boot either a 32 or 64 bit kernel without seg-faulting.

5 years agoMisc fixes for cross-building amd64.
Matthew Dillon [Sat, 28 Mar 2009 03:28:41 +0000 (20:28 -0700)]
Misc fixes for cross-building amd64.

Use the proper IMAKEENV instead of CROSSENV when installing a 64 bit
kernel via installkernel.  IMAKEENV supplies the proper cross-building
path so objcopy works properly.

Remove the -mcmodel option from the compilation of genassym.c.  This
utility is compiled for the host OS, not the target.

Add MACHINE_ARCH, MACHINE, and MACHINE_PLATFORM to the module build
called from the kernel build.

Add -fno-omit-frame-pointer to kmod.mk for amd64 builds (from FreeBSD).

5 years agoFix improper directory following when 'cpu' softlink already exists.
Matthew Dillon [Fri, 27 Mar 2009 22:24:46 +0000 (15:24 -0700)]
Fix improper directory following when 'cpu' softlink already exists.

Change ln -fs to ln -hfs for the 'cpu' directory link.  Without this
the ln command will create a cpu/cpu softlink instead of replacing the
cpu softlink.

This also fixes a build failure which will occur if the world
sources are mounted read-only.

5 years agoMake the jg_tos_ok debug check amd64-specific only and fix compiler warnings.
Matthew Dillon [Fri, 27 Mar 2009 17:08:14 +0000 (10:08 -0700)]
Make the jg_tos_ok debug check amd64-specific only and fix compiler warnings.

5 years agoAdjust the KTR code to use cache-aligned per-cpu structures.
Matthew Dillon [Fri, 27 Mar 2009 17:05:46 +0000 (10:05 -0700)]
Adjust the KTR code to use cache-aligned per-cpu structures.

* Create a cache-aligned per-cpu structure containing per-cpu information.

* Increase the __cachealign constants from 32 bytes to 64 bytes
  (512 bit cache line).

* Adjust ktrdump to handle both the old and new kernel structures.

5 years agoAdjust vnconfig recommendation when creating a new rootimg.
Matthew Dillon [Fri, 27 Mar 2009 16:03:10 +0000 (09:03 -0700)]
Adjust vnconfig recommendation when creating a new rootimg.

When using a HAMMER filesystem it makes no sense to pre-zero the rootimg
before mounting and populating it, because HAMMER will not reuse the
underlying data blocks anyway.

5 years agoMerge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly
Matthew Dillon [Fri, 27 Mar 2009 16:01:35 +0000 (09:01 -0700)]
Merge branch 'master' of ssh://crater.dragonflybsd.org/repository/git/dragonfly

5 years agoFix DDB's 'show ktr' command - improper panic and improper scan
Matthew Dillon [Fri, 27 Mar 2009 16:00:05 +0000 (09:00 -0700)]
Fix DDB's 'show ktr' command - improper panic and improper scan

DDB's 'show ktr' command was using the wrong index to check the KTR
buffers and also improperly panicing when no KTR entries could be found.

5 years agovkernel(7): mention how to autologin as root
Stathis Kamperis [Fri, 27 Mar 2009 14:43:02 +0000 (14:43 +0000)]
vkernel(7): mention how to autologin as root

Approved-by: swildner@, aggelos@
5 years agobootpd(8): Remove 'register' and P() and silence -Wold-style-definition.
Sascha Wildner [Fri, 27 Mar 2009 12:46:52 +0000 (13:46 +0100)]
bootpd(8): Remove 'register' and P() and silence -Wold-style-definition.

5 years agoFix a few typos.
Sascha Wildner [Fri, 27 Mar 2009 10:13:48 +0000 (11:13 +0100)]
Fix a few typos.

5 years agoRemove old README.amd64
Sascha Wildner [Fri, 27 Mar 2009 07:54:00 +0000 (08:54 +0100)]
Remove old README.amd64

5 years agobuild(7): Add some lines about TARGET_PLATFORM.
Sascha Wildner [Fri, 27 Mar 2009 07:51:55 +0000 (08:51 +0100)]
build(7): Add some lines about TARGET_PLATFORM.

5 years agokernconf(5): Add some words about pc64.
Sascha Wildner [Fri, 27 Mar 2009 07:50:28 +0000 (08:50 +0100)]
kernconf(5): Add some words about pc64.

5 years agoAdd support for devctl(4) into MAKEDEV.
Hasso Tepper [Thu, 26 Mar 2009 08:33:40 +0000 (10:33 +0200)]
Add support for devctl(4) into MAKEDEV.

5 years agoAdd a -O option: force one volume.
Hasso Tepper [Fri, 27 Mar 2009 08:32:47 +0000 (10:32 +0200)]
Add a -O option: force one volume.

This causes pax to not prompt for a new volume upon premature end-of-volume
(e.g. broken pipe).  This is especially useful in automated environments
where error recovery cannot be performed by a human.

Unbreaks a lot of pkgsrc packages.

Obtained-from: NeBSD

5 years agomd5(1): Raise WARNS to 6.
Sascha Wildner [Thu, 26 Mar 2009 11:01:44 +0000 (12:01 +0100)]
md5(1): Raise WARNS to 6.

5 years agoSync DRM code with FreeBSD trunk rev 190433.
Hasso Tepper [Thu, 26 Mar 2009 08:34:49 +0000 (10:34 +0200)]
Sync DRM code with FreeBSD trunk rev 190433.

5 years agoFix a recently introduced compile time error w/ IPSEC.
Matthew Dillon [Fri, 27 Mar 2009 02:52:05 +0000 (19:52 -0700)]
Fix a recently introduced compile time error w/ IPSEC.

Submitted-by: Sepherosa Ziehau <sephe@dragonflybsd.org>
5 years agocommitter(7): fix typo
Stathis Kamperis [Thu, 26 Mar 2009 15:22:55 +0000 (15:22 +0000)]
committer(7): fix typo

5 years agoip_forward: Don't cache IP forwarding route.
Sepherosa Ziehau [Thu, 26 Mar 2009 12:57:19 +0000 (20:57 +0800)]
ip_forward: Don't cache IP forwarding route.

There are two problems with this cached route:
- Any newly created route will nuke this cached route.  If any new
  route was created when the ip_output() was put to sleep due to
  contention on the output path, then the rtentry of this cached
  route would be ripped away behind our back.
  The reported panic was caused by this problem.  The call path:
  ip_forward()->ip_output()->ipflow_create(); the cached route was
  ripped before ipflow_create() was called.
- All network threads on the same CPU share this cached route:
  o  Network thread1 and network thread2 are on the same CPU
  o  Network thread1 uses this cached route to do the ip_output()
  o  Network thread1 is put to sleep due to contention on the
     output path
  o  Network thread2 uses this cached route to do the ip_output()
  o  Now network thread1 was scheduled ...

If we ever want to cached previously used route, we will have to
use per-thread route cache.

5 years agoem/emx(4): Add 82571EB Bypass to the supported device list.
Sepherosa Ziehau [Thu, 26 Mar 2009 10:46:44 +0000 (18:46 +0800)]
em/emx(4): Add 82571EB Bypass to the supported device list.

Reported/Tested-by: ruse39 on EFnet #dragonflybsd

5 years agoip_dooptions: Don't abuse ipforward_rt
Sepherosa Ziehau [Wed, 25 Mar 2009 11:59:15 +0000 (19:59 +0800)]
ip_dooptions: Don't abuse ipforward_rt

5 years agocsh(1): Raise WARNS to 2.
Sascha Wildner [Thu, 26 Mar 2009 10:54:57 +0000 (11:54 +0100)]
csh(1): Raise WARNS to 2.

5 years agodevices.conf(5): mdoc cleanup
Sascha Wildner [Thu, 26 Mar 2009 09:20:04 +0000 (10:20 +0100)]
devices.conf(5): mdoc cleanup