]>
CyberLeo.Net >> Repos - FreeBSD/FreeBSD.git/log
Enji Cooper [Sun, 30 Nov 2014 00:55:50 +0000 (00:55 +0000)]
- Fill in MK_FILE, MK_PMC, MK_SQLITE3, and MK_ZONEINFO
- Fill in etc/pam.d entries per r275291
- Fill in lib/libaudit entries
- Fill in more MK_NTP entries
Enji Cooper [Sun, 30 Nov 2014 00:51:00 +0000 (00:51 +0000)]
Honor MK_AT, MK_FTP, MK_MAIL, and MK_OPENSSH
Enji Cooper [Sun, 30 Nov 2014 00:49:57 +0000 (00:49 +0000)]
Install bsm headers if MK_AUDIT != no
Enji Cooper [Sun, 30 Nov 2014 00:49:24 +0000 (00:49 +0000)]
Honor MK_FILE
Enji Cooper [Sun, 30 Nov 2014 00:49:06 +0000 (00:49 +0000)]
Honor MK_AUDIT, MK_FILE, MK_GPIO, and MK_SQLITE3
lib/libgpio is a standalone library, but it seems logical to conditionalize
it with MK_GPIO
Enji Cooper [Sun, 30 Nov 2014 00:45:33 +0000 (00:45 +0000)]
Add MK_FILE and MK_SQLITE3
- MK_FILE will conditionalize building lib/libmagic and usr.bin/file
- MK_SQLITE3 will conditionalize building lib/libsqlite3, and will disable
MK_SVN/MK_SVNLITE by proxy
Enji Cooper [Sun, 30 Nov 2014 00:26:50 +0000 (00:26 +0000)]
Add MK_NTP block to etc/rc.d/Makefile
Enji Cooper [Sun, 30 Nov 2014 00:25:47 +0000 (00:25 +0000)]
Honor MK_GSSAPI and MK_PKGBOOTSTRAP a bit better in etc/Makefile
Enji Cooper [Sun, 30 Nov 2014 00:23:34 +0000 (00:23 +0000)]
Better honor the following MK_ flags in etc/Makefile and etc/rc.d/Makefile:
- MK_AMD
- MK_ATM
- MK_AUDIT
- MK_BOOT
- MK_GSSAPI
- MK_JAIL
- MK_LPR
- MK_SENDMAIL
- MK_VI
- MK_WIRELESS
- MK_ZFS
Enji Cooper [Sun, 30 Nov 2014 00:21:06 +0000 (00:21 +0000)]
Rework rc.d requirements so zfs and zvol come before relevant services so they
can be decoupled from the build
Enji Cooper [Sun, 30 Nov 2014 00:03:27 +0000 (00:03 +0000)]
Make kdc run before SERVERS so kdc can be decoupled from the build
Enji Cooper [Sat, 29 Nov 2014 23:59:51 +0000 (23:59 +0000)]
Make ntpdate run before rpcbind to allow ntpdate to be decoupled from the build
Enji Cooper [Sat, 29 Nov 2014 23:57:08 +0000 (23:57 +0000)]
Make gssd come before nfsd so gssd can be decoupled from etc/rc.d/Makefile
Enji Cooper [Sat, 29 Nov 2014 23:49:00 +0000 (23:49 +0000)]
Make local_unbound come before NETWORKING instead of having NETWORKING depend
on local_unbound
This should be reworked with a dummy DNS provider
Enji Cooper [Sat, 29 Nov 2014 22:42:53 +0000 (22:42 +0000)]
atm1 already sets "BEFORE: netif", so remove "REQUIRE: atm1" from netif to
allow atm* to be decoupled from rc.d
Enji Cooper [Sat, 29 Nov 2014 09:09:54 +0000 (09:09 +0000)]
Skip usr/share/misc
Enji Cooper [Sat, 29 Nov 2014 07:41:02 +0000 (07:41 +0000)]
Honor MK_BLUETOOTH when installing bluetooth.device.conf
Enji Cooper [Sat, 29 Nov 2014 05:28:40 +0000 (05:28 +0000)]
MFHead @ r275232
Enji Cooper [Sat, 29 Nov 2014 05:14:26 +0000 (05:14 +0000)]
Adapt logic from src.opts.mk to kern.opts.mk to unbreak kernel builds when
MODULES_OVERRIDE is not defined
- Add MACHINE/MACHINE_ARCH logic from src.opts.mk
- Add AUTOFS, BHYVE, HYPERV, and ISCSI
Enji Cooper [Sat, 29 Nov 2014 01:58:52 +0000 (01:58 +0000)]
Make ee(1) optional via MK_EE
Enji Cooper [Sat, 29 Nov 2014 01:55:59 +0000 (01:55 +0000)]
Implement mdoc cross references
Enji Cooper [Sat, 29 Nov 2014 01:53:55 +0000 (01:53 +0000)]
Fill in MK_VI == no
Enji Cooper [Sat, 29 Nov 2014 01:53:27 +0000 (01:53 +0000)]
share/termcap requires ex from usr.bin/vi in order to compile, and since usr.bin
comes after share in SUBDIR in Makefile.inc1, the build will fail when vi is not
installed on the build host
Run build-tools for usr.bin/vi and install ex, etc to WORLDTMP to enable building
share/termcap on hosts that don't have nvi installed on them
Baptiste Daroussin [Sat, 29 Nov 2014 00:45:09 +0000 (00:45 +0000)]
Update libucl to latest version
While here correctly link libucl to libm and register the dependency on libm
for static building
Baptiste Daroussin [Sat, 29 Nov 2014 00:33:10 +0000 (00:33 +0000)]
Update libucl to latest version
Enji Cooper [Fri, 28 Nov 2014 22:08:50 +0000 (22:08 +0000)]
Remove the conditional
Enji Cooper [Fri, 28 Nov 2014 22:07:16 +0000 (22:07 +0000)]
Install skeleton files based on the values of MK_MAIL and MK_TCSH
Customize according to MK_GAMES and MK_VI, in particular comment out lines
that would change behavior
Change dot.cshrc and dot.profile to not check for /usr/games/fortune based
on the new behavior in customizing the files
Baptiste Daroussin [Fri, 28 Nov 2014 22:03:35 +0000 (22:03 +0000)]
Readd Makefile.inc that were used to include upper level Makefile.inc and
set (among other things) the destination where the binary should be installed
Enji Cooper [Fri, 28 Nov 2014 20:48:53 +0000 (20:48 +0000)]
Ignore usr/share/nls
Bjoern A. Zeeb [Fri, 28 Nov 2014 14:51:49 +0000 (14:51 +0000)]
After r275196 unbreak NOIP and NOINET kernels by hiding an otherwise
unused varibale under the proper #ifdef.
Eygene Ryabinkin [Fri, 28 Nov 2014 12:14:59 +0000 (12:14 +0000)]
DRM2: fix off-by-one overflow in ioctl processing
Call to the driver-specific ioctl used to process ioctl number
that will lead to the out-of-bounds access to the ioctl handler
array.
PR: 193367
Approved by: kib
MFC after: 1 week
Andrew Turner [Fri, 28 Nov 2014 11:49:26 +0000 (11:49 +0000)]
Some device tree configurations place the generic timer under the root
of the tree and not under simplebus. Update the driver to handle this.
Submitted by: Julien Grall <julien.grall AT linaro.org>
MFC after: 1 week
Andrew Turner [Fri, 28 Nov 2014 11:45:53 +0000 (11:45 +0000)]
We don't use the hypervisor interrupt, make it optional in the device tree.
Submitted by: Julien Grall <julien.grall AT linaro.org>
MFC after: 1 week
Konstantin Belousov [Fri, 28 Nov 2014 10:20:00 +0000 (10:20 +0000)]
Assert the state of the process lock and sigact mutex in
kern_sigprocmask() and reschedule_signals().
Discussed with: rea
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Hans Petter Selasky [Fri, 28 Nov 2014 09:32:07 +0000 (09:32 +0000)]
Style changes:
- Move two IOCTL related defines to the top of the C-file
- Add more comments describing the recently added IOCTL small size and
small align macros
Cy Schubert [Fri, 28 Nov 2014 04:07:06 +0000 (04:07 +0000)]
Correctly define constants.
MFC after: 1 week
Enji Cooper [Thu, 27 Nov 2014 23:42:32 +0000 (23:42 +0000)]
Use ${.TARGET} instead of hardcoding the name in the dump build rule
Alexander V. Chernikov [Thu, 27 Nov 2014 23:10:03 +0000 (23:10 +0000)]
Fix build broken by r275195.
Alexander V. Chernikov [Thu, 27 Nov 2014 23:06:25 +0000 (23:06 +0000)]
Do not return unlocked/unreferenced lle in arpresolve/nd6_storelladdr -
return lle flags IFF needed.
Do not pass rte to arpresolve - pass is_gateway flag instead.
Alexander V. Chernikov [Thu, 27 Nov 2014 21:29:19 +0000 (21:29 +0000)]
Do not try to copy header to @dst and than back to ethernet in case of
pseudo_AF_HDRCMPLT:
we copy media header from mbuf to 'struct sockaddr' @dst in bpf_movein, so
mbuf already contains valid info.
Baptiste Daroussin [Thu, 27 Nov 2014 20:24:58 +0000 (20:24 +0000)]
Remove leftovers readline being changed to an internallib
Ed Maste [Thu, 27 Nov 2014 20:22:46 +0000 (20:22 +0000)]
Put each SUBDIR on a separate line
This makes it easier to review or merge changes that modify some subset
of SUBDIRs.
Ed Maste [Thu, 27 Nov 2014 20:12:13 +0000 (20:12 +0000)]
Copy elftoolchain binutils replacements from vendor branch
Sponsored by: The FreeBSD Foundation
Roman Divacky [Thu, 27 Nov 2014 18:43:44 +0000 (18:43 +0000)]
Revert part of r275059. Comparing unsigned 8 bit value
against -'0' is always false so the conditional block is
optimized away.
Justin Hibbits [Thu, 27 Nov 2014 18:41:14 +0000 (18:41 +0000)]
Fix hwpmc sampling for ppc970 (G5-class) processors.
With this, hwpmc sampling now works on these processors.
MFC after: 3 weeks
Relnotes: yes
Maxim Konovalov [Thu, 27 Nov 2014 14:54:17 +0000 (14:54 +0000)]
o NetBSD 6.1.5, OpenBSD 5.6, FreeBSD 10.1, DragonFly 4.0.1 added.
Enji Cooper [Thu, 27 Nov 2014 09:53:17 +0000 (09:53 +0000)]
Populate MK_AUTOFS and MK_BSDCONFIG; add more dtrace bits to MK_CDDL; add minimal bits to MK_LIB32
Enji Cooper [Thu, 27 Nov 2014 09:50:12 +0000 (09:50 +0000)]
Honor MK_AUTOFS
Enji Cooper [Thu, 27 Nov 2014 09:42:40 +0000 (09:42 +0000)]
Add WITHOUT_AUTOFS and WITHOUT_BSDCONFIG
Enji Cooper [Thu, 27 Nov 2014 09:40:36 +0000 (09:40 +0000)]
Filter out /lib, /usr/lib32, and fold the /usr/lib* checks into one expression
Nick Hibma [Thu, 27 Nov 2014 09:17:42 +0000 (09:17 +0000)]
clean_build is not always done (*), so move the mkdir to the first step
that needs it and is always done: make.conf build.
(*) In our build environment we actually never do a clean build unless
instructed to do so, to make sure we do not accidentally delete the
built world when on a tight schedule.
Enji Cooper [Thu, 27 Nov 2014 08:31:20 +0000 (08:31 +0000)]
Honor MK_EXAMPLES
Enji Cooper [Thu, 27 Nov 2014 07:56:28 +0000 (07:56 +0000)]
Handle MK_RADIUS_SUPPORT in usr.sbin/ppp
Enji Cooper [Thu, 27 Nov 2014 07:49:25 +0000 (07:49 +0000)]
Only install libmap32.conf if MK_LIB32 != no
Enji Cooper [Thu, 27 Nov 2014 07:33:50 +0000 (07:33 +0000)]
Until I teach bsd.*.mk to accept NO_ROOT, make this script only accept root users
Enji Cooper [Thu, 27 Nov 2014 07:26:20 +0000 (07:26 +0000)]
Remove /etc/X11 entry from BSD.root.dist
- FreeBSD hasn't packaged X11 with the distro since 8.x
- X.org hasn't required xorg.conf [in /etc/X11] for some time now
- X.org should work with ${LOCALBASE}/etc/X11
- This path should be included in the X.org package manifest
Enji Cooper [Thu, 27 Nov 2014 07:19:36 +0000 (07:19 +0000)]
Ignore boot and remove from MK_BOOT additions done in previous commit
Enji Cooper [Thu, 27 Nov 2014 07:17:32 +0000 (07:17 +0000)]
Fill in MK_BOOT for amd64
Justin Hibbits [Thu, 27 Nov 2014 06:42:34 +0000 (06:42 +0000)]
Fix hwpmc sampling for MPC74xxx (G4) processors.
With this, hwpmc sampling now works correctly on these processors.
MFC after: 3 weeks
Relnotes: yes
Enji Cooper [Thu, 27 Nov 2014 06:04:00 +0000 (06:04 +0000)]
Convert sbin/mdconfig/tests from prove format tests to ATF format tests
As a side effect...
1. The tests now checks for the root user before continuing with kyua, which is
more visible than the test being skipped with the TAP protocol
2. The tests work with devices that aren't /dev/md0 by caching the device
attached during the test to a file, and later use the cached information to
detach the device in the cleanup routine
3. The tests no longer require perl to run
MFC after: 1 week
PR: 191191
Sponsored by: EMC / Isilon Storage Division
Enji Cooper [Thu, 27 Nov 2014 04:23:00 +0000 (04:23 +0000)]
Fix typos: MK_RADIUS -> MK_RADIUS_SUPPORT
Enji Cooper [Thu, 27 Nov 2014 04:19:01 +0000 (04:19 +0000)]
Populate MK_CUSE, MK_NETGRAPH, MK_NETGRAPH_SUPPORT, and MK_RADIUS_SUPPORT
Enji Cooper [Thu, 27 Nov 2014 04:16:47 +0000 (04:16 +0000)]
Implement MK_RADIUS_SUPPORT to make libradius and assorted support optional
Enji Cooper [Thu, 27 Nov 2014 04:01:39 +0000 (04:01 +0000)]
Honor MK_ATM, MK_CUSE, and MK_NETGRAPH_SUPPORT
Shuffle around conditionals so they're better sorted
Enji Cooper [Thu, 27 Nov 2014 03:59:57 +0000 (03:59 +0000)]
Implement knobs for MK_NETGRAPH_SUPPORT
Ed Maste [Thu, 27 Nov 2014 02:15:35 +0000 (02:15 +0000)]
ANSIfy b64_pton
Kevin Lo [Thu, 27 Nov 2014 01:37:01 +0000 (01:37 +0000)]
Init array field in the proper place.
Obtained from: OpenBSD
Andrey V. Elsukov [Thu, 27 Nov 2014 00:27:39 +0000 (00:27 +0000)]
Remove ip4_input() declaration. It was removed in r275133.
MFC after: 1 month
Enji Cooper [Thu, 27 Nov 2014 00:10:59 +0000 (00:10 +0000)]
Honor MK_NETGRAPH for netgraph manpages
Enji Cooper [Thu, 27 Nov 2014 00:10:16 +0000 (00:10 +0000)]
Honor MK_NETGRAPH for netgraph components
Enji Cooper [Thu, 27 Nov 2014 00:02:42 +0000 (00:02 +0000)]
Only build iscsi/iscsi_initiator if MK_ISCSI != no
Enji Cooper [Wed, 26 Nov 2014 23:07:54 +0000 (23:07 +0000)]
Move hyperv/vmm under MK_HYPERV/MK_BHYVE, respectively
Enji Cooper [Wed, 26 Nov 2014 22:41:51 +0000 (22:41 +0000)]
- Fill in MK_LEGACY_CONSOLE == no and SYSCONS == no
- Don't install syscons(4)
- Don't install the moused or syscons rc.d scripts if MK_LEGACY_CONSOLE == no
Enji Cooper [Wed, 26 Nov 2014 22:27:31 +0000 (22:27 +0000)]
Followup to r275144 -- don't traverse down usr.sbin/zzz if ACPI/APM are disabled
Enji Cooper [Wed, 26 Nov 2014 22:26:22 +0000 (22:26 +0000)]
- Fix egrep -> grep typo introduced in previous commit
- Do depth-first search on OLD_DIRS so the list is logically sorted
Enji Cooper [Wed, 26 Nov 2014 21:53:53 +0000 (21:53 +0000)]
Installing zzz(8) doesn't make sense when both ACPI and APM are disabled
Enji Cooper [Wed, 26 Nov 2014 21:32:12 +0000 (21:32 +0000)]
Add missing line continuation
Enji Cooper [Wed, 26 Nov 2014 21:18:52 +0000 (21:18 +0000)]
Clean up more usb related files when MK_USB == no when dealing with
manpages, libraries, and binaries
Enji Cooper [Wed, 26 Nov 2014 21:17:32 +0000 (21:17 +0000)]
- Handle OLD_LIBS properly by looking for versioned .so's
- Filter out /usr/libdata and /usr/libdata/pkgconfig from OLD_DIRS
Ed Maste [Wed, 26 Nov 2014 20:56:08 +0000 (20:56 +0000)]
Increase default and maximum callchain depths
Bump the default from 16 to 32, to accommodate kernel flamegraphs.
Bump the maximum from 32 to 128, to accommodate deep user stacks.
Reviewed by: gnn
MFC after: 1 month
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D1203
Glen Barber [Wed, 26 Nov 2014 20:44:54 +0000 (20:44 +0000)]
Regen src.conf(5) after r275138.
Sponsored by: The FreeBSD Foundation
Glen Barber [Wed, 26 Nov 2014 20:43:09 +0000 (20:43 +0000)]
Whitespace cleanup to fix rendering nits in
src.conf(5).
Sponsored by: The FreeBSD Foundation
Adrian Chadd [Wed, 26 Nov 2014 20:34:05 +0000 (20:34 +0000)]
Add PCI ID for Intel Lynx Point LP controller.
PR: kern/195398
Submitted by: grembo
Obtained from: DragonflyBSD
MFC after: 1 week
Alfred Perlstein [Wed, 26 Nov 2014 20:19:36 +0000 (20:19 +0000)]
Make igb and ixgbe check tunables at probe time.
This allows one to make a kernel module to tune the
number of queues before the driver loads.
This is needed so that a module at SI_SUB_CPU can set
tunables for these drivers to take. Otherwise getenv
is called too early by the TUNABLE macros.
Reviewed by: smh
Phabric: https://reviews.freebsd.org/D1149
Andrey V. Elsukov [Wed, 26 Nov 2014 17:44:49 +0000 (17:44 +0000)]
Do not use xform_ipip as decapsulation fallback.
xform_ipip was used as fallback with low priority for IPIP
encapsulated packets that were decrypted. In some cases
it can decapsulate packets, that it shouldn't. This leads to situations,
when wrong configurations are magically working. Also it can propagate
wrong ingress interface and this can break security.
Now we redesigned the IPSEC code and IPIP encapsulation is called directly
from ipsec_output, and decapsulation is done in the ipsec_input with m_striphdr.
Differential Revision: https://reviews.freebsd.org/D1220
MFC after: 1 month
Sponsored by: Yandex LLC
Alexander Motin [Wed, 26 Nov 2014 16:05:01 +0000 (16:05 +0000)]
Fix WWNN/WWPN generation for virtual channels.
MFC after: 1 week
Alexander Motin [Wed, 26 Nov 2014 15:03:21 +0000 (15:03 +0000)]
Fix incorrect check, blocking MULTIID functionality.
MFC after: 1 week
Konstantin Belousov [Wed, 26 Nov 2014 14:10:00 +0000 (14:10 +0000)]
The process spin lock currently has the following distinct uses:
- Threads lifetime cycle, in particular, counting of the threads in
the process, and interlocking with process mutex and thread lock.
The main reason of this is that turnstile locks are after thread
locks, so you e.g. cannot unlock blockable mutex (think process
mutex) while owning thread lock.
- Virtual and profiling itimers, since the timers activation is done
from the clock interrupt context. Replace the p_slock by p_itimmtx
and PROC_ITIMLOCK().
- Profiling code (profil(2)), for similar reason. Replace the p_slock
by p_profmtx and PROC_PROFLOCK().
- Resource usage accounting. Need for the spinlock there is subtle,
my understanding is that spinlock blocks context switching for the
current thread, which prevents td_runtime and similar fields from
changing (updates are done at the mi_switch()). Replace the p_slock
by p_statmtx and PROC_STATLOCK().
The split is done mostly for code clarity, and should not affect
scalability.
Tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Konstantin Belousov [Wed, 26 Nov 2014 14:09:04 +0000 (14:09 +0000)]
Fix SA_SIGINFO | SA_RESETHAND handling. The sysent' sv_sendsig()
method needs pre-reset state of the ps_siginfo to correctly construct
signal frame.
Move sigdflt() call after the sv_sendsig() invocation in postsig().
Simultaneously extract common code from trapsignal() and postsig()
into new helper postsig_done().
Submitted by: rea
MFC after: 1 week
Enji Cooper [Wed, 26 Nov 2014 14:01:27 +0000 (14:01 +0000)]
Fix use of OLD_FILES and OLD_LIBS for MK_BSNMP and MK_TESTS
Convention in the file is that .a files be added to OLD_FILES and versioned .so's
be added to OLD_LIBS
Alexander Motin [Wed, 26 Nov 2014 13:56:54 +0000 (13:56 +0000)]
Some microoptimizations.
MFC after: 1 month
Enji Cooper [Wed, 26 Nov 2014 13:26:12 +0000 (13:26 +0000)]
Ignore LIBPRIVATEDIR when suggesting directories to prune
Enji Cooper [Wed, 26 Nov 2014 13:25:09 +0000 (13:25 +0000)]
Add MK_TIMED option for optionally building usr.sbin/timed, et al
Enji Cooper [Wed, 26 Nov 2014 12:54:58 +0000 (12:54 +0000)]
Account for /usr/include/protocols/r{outed,whod}.h with their respective build knobs
Enji Cooper [Wed, 26 Nov 2014 12:49:37 +0000 (12:49 +0000)]
Fix some accidentally inverted logic
Alexander Motin [Wed, 26 Nov 2014 12:25:00 +0000 (12:25 +0000)]
Make isp_find_pdb_by_*() search for targets in portdb in reverse order.
Records with target_mode == 1 are allocated from the end of portdb, so it
seems logical to start search from the end not traverse whole array.
MFC after: 1 month
Hans Petter Selasky [Wed, 26 Nov 2014 10:58:08 +0000 (10:58 +0000)]
Add new USB quirk.
MFC after: 1 week
PR: 195372
Hans Petter Selasky [Wed, 26 Nov 2014 09:51:38 +0000 (09:51 +0000)]
Add support for 64-byte CQE size.
Sponsored by: Mellanox Technologies
MFC after: 3 days
Enji Cooper [Wed, 26 Nov 2014 08:44:05 +0000 (08:44 +0000)]
Conditionalize a number of components in the tree
- bootparamd
- bootpd
- finger/fingerd
- ftp/ftpd
- hastctl/hastd
- iscsid, et al
- rbootd
- talk/talkd
- tcpd, et al
- tftp/tftpd
Add src.conf entries for the various components and do a best effort
at adding components to tools/build/mk/OptionalObsoleteFiles.inc
Baptiste Daroussin [Wed, 26 Nov 2014 08:14:30 +0000 (08:14 +0000)]
Remove MINUSLPAM the LIBADD framework handles static dependencies just fine