zrj [Mon, 14 Nov 2016 08:47:36 +0000 (10:47 +0200)]
gcc50: Add alternative variant of pthread presence checking.
There are few issues that ld.gold is having with DragonFly libgcc_pic.a.
This alternative variant is less strict when linking c++ projects.
zrj [Sun, 13 Nov 2016 17:52:15 +0000 (19:52 +0200)]
LINT64: Fix buildkernel with WANT_HAMMER2=yes.
Mostly #if DEBUG paths, also use panic() in z_error() for kernel.
zrj [Fri, 11 Nov 2016 12:51:43 +0000 (14:51 +0200)]
nvi: Convert catalog creation to use hostprog.
nvi dump utility is not part of btools, so just split vidump as hostprog.
There were few issues:
* build races like dump and say spanish get created twice (and raced)
* buildworld failures, from time to time i hit ./dump linking race (rare)
* ./dump being used as target program (problems on crosscompilations)
* missing env LANG=C for sort, minor issues in awk script and SHAREDIR.
Now even "cd usr.bin/vi && make -j9 depend all" should be race free.
zrj [Thu, 10 Nov 2016 18:43:24 +0000 (20:43 +0200)]
gcc47: Remove bits/ handling.
Not used by any active sources.
Now quickworld should not delete libstdcxx .depend on no source changes.
zrj [Thu, 10 Nov 2016 18:41:54 +0000 (20:41 +0200)]
gcc50: Fix logic in cxx hostlib.
Now quickworld should not delete libstdcxx .depend on no source changes.
Antonio Huete Jimenez [Thu, 10 Nov 2016 22:59:50 +0000 (23:59 +0100)]
installer - Fix a bug in survey_storage()
- survey_storage() issues DIOCGPART on every disk present in
kern.disks in order to retrieve information about it.
The problem was that a failure to get that information from
any disk in the list caused the whole storage survey to be
marked as failed. This happens, for example, with some card
readers that present the disk devices even though there is no
media inserted on the corresponding slot.
- Fix the logic in the installer and allow to check all disks
in the system.
Antonio Huete Jimenez [Thu, 10 Nov 2016 22:26:45 +0000 (23:26 +0100)]
installer - Allow to specify cmdnames file path
Imre Vadász [Sun, 13 Nov 2016 21:36:16 +0000 (22:36 +0100)]
if_iwm - Check sc->sc_attached flag in suspend/resume callbacks.
* There is (almost) nothing to do in suspend/resume if if_iwm has failed
during initialization (e.g. because of firmware load failure) and was
already uninitialized by iwm_detach_local().
Tomohiro Kusumi [Sat, 12 Nov 2016 15:45:04 +0000 (00:45 +0900)]
sys/vfs/hammer: Fix comments
Tomohiro Kusumi [Sat, 12 Nov 2016 15:44:51 +0000 (00:44 +0900)]
sys/vfs/hammer: Remove wrong comment
Tomohiro Kusumi [Sat, 12 Nov 2016 09:57:41 +0000 (18:57 +0900)]
sys/vfs/hammer: Use HAMMER_HEAD_DOALIGN()
Tomohiro Kusumi [Sat, 12 Nov 2016 09:48:50 +0000 (18:48 +0900)]
sbin/hammer: sbin/hammer: Fix hammer show-undo output format
Add some ondisk redo fields.
Tomohiro Kusumi [Sat, 12 Nov 2016 09:26:13 +0000 (18:26 +0900)]
sys/vfs/hammer: Change unused ondisk redo_mtime to reserved
It was never used after added in
6048b411 in 2010.
Imre Vadász [Sat, 12 Nov 2016 12:41:49 +0000 (13:41 +0100)]
if_iwm - Move iwm_read_firmware() call into iwm_attach().
* We should load the firmware exactly once before the driver really
initializes the hardware the first time, and unload it at detach time.
There is no need to retrieve the firmware during execution of
iwm_mvm_load_ucode_wait_alive(), we should make sure we already have the
firmware data at hand before that.
* The existing sc_preinit_hook code fails to deal with the case where
if_iwm is loaded by the loader (or is statically linked) and the
firmware needs to be loaded from disk. So we can just call
iwm_read_firmware() from iwm_attach() directly.
* A separate solution will have to be added to properly defer the firmware
loading during bootup, until the necessary filesystem is mounted.
Sascha Wildner [Sat, 12 Nov 2016 11:45:01 +0000 (12:45 +0100)]
Do
3b5eca529baf58ea74a485139963552cc4c1d977 a bit better.
FWIW, keep the NO_CRYPT dependency in lib/libcrypt for various sources.
Imre Vadász [Fri, 11 Nov 2016 20:09:05 +0000 (21:09 +0100)]
if_iwm - Attach to new AC3168 and AC8265 chipsets (untested).
Imre Vadász [Fri, 11 Nov 2016 20:08:29 +0000 (21:08 +0100)]
iwmfw - Add version 22 firmwares for AC3168 and AC8265 chipsets.
Tomohiro Kusumi [Sat, 12 Nov 2016 06:25:41 +0000 (15:25 +0900)]
sbin/hammer: Fix hammer show-undo output format
The whole size should be in %d since data size is in %d as well.
Tomohiro Kusumi [Sat, 12 Nov 2016 05:37:26 +0000 (14:37 +0900)]
sys/vfs/hammer: Add hammer_xlate_to_undo()
Tomohiro Kusumi [Sat, 12 Nov 2016 04:11:10 +0000 (13:11 +0900)]
sbin/hammer: Refactor left/right boundary elms in hammer show
hammer show doesn't need to pass around pointers to left/right
boundary elms in every function while walking the B-Tree.
Left/right elms are there to test whether a B-Tree node in
question fits in boundaries defined by its parent, which means
that hammer show always requires the right elm be (left+1), but
not some random elm after the left elm. Because of this, it
only needs to pass around the left elm.
The code is more efficient plus look better without bunch of
left/right stuff.
Tomohiro Kusumi [Fri, 11 Nov 2016 18:23:52 +0000 (03:23 +0900)]
sbin/hammer: Cleanup zone statistics
Tomohiro Kusumi [Fri, 11 Nov 2016 18:14:49 +0000 (03:14 +0900)]
sbin/hammer: Cleanup hammer show-undo
Tomohiro Kusumi [Fri, 11 Nov 2016 18:09:39 +0000 (03:09 +0900)]
sbin/hammer: Make hammer show-undo support zone statistics
Tomohiro Kusumi [Wed, 9 Nov 2016 15:56:25 +0000 (00:56 +0900)]
sys/vfs/hammer: Add HAMMER_UNDO_DOALIGN()
The only reason it's "DOALIGN" instead of just "ALIGN" is
because that's how macros have been named in HAMMER1/2.
"ALIGN" macros represent alignment size.
Tomohiro Kusumi [Wed, 9 Nov 2016 15:41:31 +0000 (00:41 +0900)]
sbin/hammer: Cleanup blockmap
François Tigeot [Fri, 11 Nov 2016 11:25:55 +0000 (12:25 +0100)]
drm/linux: Add clamp_val()
Obtained-from: FreeBSD
François Tigeot [Fri, 11 Nov 2016 11:09:29 +0000 (12:09 +0100)]
drm/linux: Add ktime_get_boottime and ktime_ms_delta()
zrj [Fri, 11 Nov 2016 07:17:44 +0000 (09:17 +0200)]
gcc50: Install liblto_plugin.so as +x.
Sad but this limitation in gcc driver requires that utility be executable.
This matches how it is done on standard installations, so follow along.
Some ports rely on gcc -print-prog-name=liblto_plugin.so working.
zrj [Fri, 11 Nov 2016 07:02:07 +0000 (09:02 +0200)]
build: Suppress exessive echoing.
zrj [Wed, 9 Nov 2016 17:14:52 +0000 (19:14 +0200)]
LIBPRIV: Add common PRIVATELIB_LDFLAGS definitions.
Simply being able to change it in two places is easier than everywhere.
Also add PRIVATELIB_BLDFLAGS that is intended for bootsrapping only.
While there, perform some cleanup for dup entries.
zrj [Wed, 9 Nov 2016 15:36:05 +0000 (17:36 +0200)]
LIBPRIV: Introduce PRIVATELIB knob (just like INTERNALLIB).
Collect common private libraries path handling bits into share/mk/bsd.lib.mk.
It is cleaner this way, prevents issues like manual paths and/or ignoring stale
.depend for /usr/obj/usr/src/world_x86_64/usr/lib/priv/priv/libprivate_ncurses.a
zrj [Wed, 9 Nov 2016 10:25:23 +0000 (12:25 +0200)]
flex: Break the hardlinks to fix quickworld issues.
Upon repeated quickworld run when install -C installs the libfl.a it breaks
the previous hardlink and retrigers relink for anything that depends on -ll.
Solve this by changing logic to install compat libs for install target only.
During buildworld only main ${LIBL} should be used where needed.
Fix a single case where -lfl was used in binutils. Add missing DPADD too.
zrj [Wed, 9 Nov 2016 10:12:41 +0000 (12:12 +0200)]
flex: Rename and install the main libl.a as primary one.
libfl and libln are only used as compat ones for external sources linkage.
Whole base makefile system uses the ${LIBL}.
Tomohiro Kusumi [Wed, 9 Nov 2016 13:06:18 +0000 (22:06 +0900)]
sys/vfs/hammer: Add HAMMER_ZONE_LAYER2_NEXT_OFFSET()
Note that this is different from HAMMER_BLOCKMAP_LAYER2_OFFSET().
HAMMER_BLOCKMAP_LAYER2_OFFSET() is an offset of layer2 entry within
a big-block for layer2, but not zone offset itself.
Tomohiro Kusumi [Wed, 9 Nov 2016 12:52:21 +0000 (21:52 +0900)]
sys/vfs/hammer: Add HAMMER_ZONE_LAYER1_NEXT_OFFSET()
Note that this is different from HAMMER_BLOCKMAP_LAYER1_OFFSET().
HAMMER_BLOCKMAP_LAYER1_OFFSET() is an offset of layer1 entry within
a big-block for layer1, but not zone offset itself.
zrj [Mon, 7 Nov 2016 12:33:52 +0000 (14:33 +0200)]
bsd.crunchgen.mk: Sync with FreeBSD.
This finally allows not to relink rescue image if there are no changes and
have clean stage of "4d: building everything.."!
Helps a lot on LTO quick rebuilds.
While there, remove hack from crunchgen(1), it no longer should be needed.
Please report build races if any.
Taken-from: FreeBSD
zrj [Mon, 7 Nov 2016 11:32:11 +0000 (13:32 +0200)]
bsd.crunchgen.mk: Split few monolitic targets.
For easier debug.
Taken-from: FreeBSD
zrj [Mon, 7 Nov 2016 11:12:54 +0000 (13:12 +0200)]
bsd.crunchgen.mk: Simplify mk logic a bit.
Just provide defaults if needed.
zrj [Mon, 7 Nov 2016 10:40:29 +0000 (12:40 +0200)]
bsd.crunchgen.mk: Remove dub handling of CANONICALOBJDIR.
It is already handled in bsd.obj.mk.
While there, drop non functional NO_DEBUG_FILES and update bsd.README.
zrj [Mon, 7 Nov 2016 08:48:16 +0000 (10:48 +0200)]
bsd.crunchgen.mk: No longer exclude -P option.
It was used only when fmake.
While there, do not hardcode crunchgen name and add env handling.
Taken-from: FreeBSD
zrj [Mon, 7 Nov 2016 08:40:22 +0000 (10:40 +0200)]
bsd.crunchgen.mk: Use single style for variables.
No functional change.
Imre Vadász [Mon, 7 Nov 2016 21:02:31 +0000 (22:02 +0100)]
if_iwm - Advertise full-offload scan support, fixes scan while associated.
* This avoids the firmware panics previously experienced when starting a
wlan scan (e.g. with "ifconfig wlan0 scan" or "scan" command in a
wpa_supplicant user-interface), while associated to an AP.
Imre Vadász [Mon, 7 Nov 2016 20:45:45 +0000 (21:45 +0100)]
wlan - Import initial full-offload scan support from FreeBSD.
This is a very simple addition to the net80211 scan support.
It doesn't implement a replacement scan interface - it just disables
the pieces that we should disable to make this lifecycle a bit
more managable.
There's more work to come before full scan offload support is available
but it should be good enough for driver work.
* add a flag to say "full offload"
* don't do probe requests when scanning full-offload - firmware can do that
* don't do powersave transitions and buffering - firmware can do that
* don't abort a background scan upon reception of a single packet.
Full offload drivers don't need this behaviour - they do it in firmware.
Taken-From: FreeBSD (svn r307602, r308007 and r308008)
Tomohiro Kusumi [Mon, 7 Nov 2016 15:12:54 +0000 (00:12 +0900)]
sys/vfs/hammer: Use HAMMER_BUFSIZE_DOALIGN() and variants
This should have been in
bbc5e78b (failed to catch these).
Tomohiro Kusumi [Mon, 7 Nov 2016 15:06:43 +0000 (00:06 +0900)]
sys/vfs/hammer: Cleanups
Tomohiro Kusumi [Mon, 7 Nov 2016 14:53:08 +0000 (23:53 +0900)]
sys/vfs/hammer: Add HAMMER_BIGBLOCK_DOALIGN()
The only reason it's "DOALIGN" instead of just "ALIGN" is
because that's how macros have been named in HAMMER1/2.
"ALIGN" macros represent alignment size.
Matthew Dillon [Tue, 8 Nov 2016 02:56:34 +0000 (18:56 -0800)]
kernel - Attempt to fix cluster pbuf deadlock on recursive filesystems
* Change global pbuf count limits (used primarily for clustered I/O) to
per-mount and per-device limits. The per-mount / per-device limit
is set to nswbuf_kva / 10, allowing 10 different entities to obtain
pbufs concurrently without interference.
* This change goes a long way towards fixing deadlocks that could occur
with the old global system (a global limit of nswbuf_kva / 2) when
the I/O system recurses through a virtual block device or filesystem.
Two examples of virtual block devices are the 'vn' device and the crypto
layer.
* We also note that even normal filesystem read and write I/O strategy calls
will recurse at least once to dive the underlying block device. DFly also
had issues with pbuf hogging by one mount causing unnecessary stalls
in other mounts. This fix also prevents pbuf hogging.
* Remove unused internal O_MAPONREAD flag.
Reported-by: htse, multiple
Testing-by: htse, dillon
Matthew Dillon [Sun, 6 Nov 2016 19:55:45 +0000 (11:55 -0800)]
kernel - Add nvme to the default kernel build
* Add the nvme disk driver to the default kernel build, it no longer
needs to be loaded as a module.
Tomohiro Kusumi [Mon, 7 Nov 2016 12:04:41 +0000 (21:04 +0900)]
sbin/hammer: Cleanup alloc_blockmap()
Layer1 could be assertion.
It's essentially the same as what
0d1f6085 in 2015 checks.
Tomohiro Kusumi [Sun, 6 Nov 2016 14:14:06 +0000 (23:14 +0900)]
sbin/hammer: Add another condition to assert_volume_offset()
which is kinda obvious and expected to be false on newfs time.
Tomohiro Kusumi [Sun, 6 Nov 2016 14:29:54 +0000 (23:29 +0900)]
sbin/hammer: Split alloc_bigblock() into two functions [2/2]
This function is currently only used by zone-3, but it's actually
only usable by zone-3 from the way it's implemented, so rename it
to alloc_undo_bigblock() and get rid of zone argument.
The way blockmap offsets (in root volume) get updated for zone-3
differs from other zones like 8/9/10/11, and this function which
doesn't update blockmap offsets on allocation only works against
zone-3.
Tomohiro Kusumi [Sun, 6 Nov 2016 13:51:09 +0000 (22:51 +0900)]
sbin/hammer: Split alloc_bigblock() into two functions [1/2]
For zone-4, all you can (need to) do is increment offset by 8MB,
so split alloc_bigblock() into bootstrap_bigblock() for zone-4 and
alloc_bigblock() for zone-3. It's easier to understand this way.
Sascha Wildner [Sun, 6 Nov 2016 23:08:49 +0000 (00:08 +0100)]
Adjust various things after the removal of secure/.
* Fix etc/Makefile: This brings back installation of /etc/ssh_config
and /etc/sshd_config to our IMG/ISO.
* Add back DES and Blowfish functions to libcrypt.
* Remove installation of CA.pl/CA.sh. They don't seem to be part of
LibreSSL.
* Remove obsolete paths from whereis(1).
* Adjust hier(7) manual page.
* Adjust openssh upgrade instructions.
* While here, fix a typo in Makefile.ssh.common.
Imre Vadász [Sun, 6 Nov 2016 21:21:22 +0000 (22:21 +0100)]
if_iwm - The iwm_prepare_card_hw() in iwm_attach() is only needed on 8K hw.
* Doing the iwm_prepare_card_hw() call in iwm_attach() only on Family 8000
hardware matches the code in Linux iwlwifi.
* While there remove DEFAULT_MAX_TX_POWER definition which is unused, and
has a value different from IWL_DEFAULT_MAX_TX_POWER in iwlwifi.
Imre Vadász [Sun, 6 Nov 2016 16:44:56 +0000 (17:44 +0100)]
if_iwm - Use rates given in struct ieee80211_txparam for non-data transfers.
* Make iwm_tx_fill_cmd() just return the uint8_t rate, to avoid looking up
the firmware rate-index for DATA transfers.
Taken-From: FreeBSD (svn r306286)
Imre Vadász [Sun, 6 Nov 2016 15:25:54 +0000 (16:25 +0100)]
if_iwm - Update firmware rs table, instead of indexing the table in tx cmds.
* Rather than providing a non-zero index into the firmware RS table,
we should always use index 0 and update the firmware RS table whenever
our chosen tx rate for data-frames changes.
* Send IWM_LQ_CMD updates when the tx rate gets updated by the net80211
rate control (which is after we tell the tx status to the net80211
rate-control in iwm_mvm_rx_tx_cmd_single()).
* Disregard frames transferred with a different tx rate than the currently
selected rate for the rate-control calculations. This way we avoid
counting management frames (which are sent at a slow, and fixed rate),
as well as frames we added to the tx queue just before a new IWM_LQ_CMD
update took effect.
Inspired-By: Linux iwlwifi
Matthew Dillon [Sun, 6 Nov 2016 19:24:00 +0000 (11:24 -0800)]
build - Fix modifying operation to /usr/obj during installworld (2)
* efi/boot1 needs the same treatment as the last commit.
Reported-by: ivadasz
Matthew Dillon [Sun, 6 Nov 2016 18:56:35 +0000 (10:56 -0800)]
build - Fix modifying operation to /usr/obj during installworld.
* The build for /usr/src/sys/boot/efi/loader is incomplete. The
installworld step tries to build the last piece. Add the piece to
all: to properly build it during the build phase and not the
install phase.
Tomohiro Kusumi [Sun, 6 Nov 2016 06:17:37 +0000 (15:17 +0900)]
sys/vfs/hammer: Add HAMMER_DATA_DOALIGN() and variants
The only reason it's "DOALIGN" instead of just "ALIGN" is
because that's how macros have been named in HAMMER1/2.
"ALIGN" macros represent alignment size.
Tomohiro Kusumi [Sun, 6 Nov 2016 06:35:28 +0000 (15:35 +0900)]
sys/vfs/hammer: Add HAMMER_BUFSIZE_DOALIGN() and variants
The only reason it's "DOALIGN" instead of just "ALIGN" is
because that's how macros have been named in HAMMER1/2.
"ALIGN" macros represent alignment size.
Tomohiro Kusumi [Sun, 6 Nov 2016 05:47:26 +0000 (14:47 +0900)]
sys/vfs/hammer: Add HAMMER_BLOCKMAP_LAYER2_DOALIGN()
The only reason it's "DOALIGN" instead of just "ALIGN" is
because that's how macros have been named in HAMMER1/2.
"ALIGN" macros represent alignment size.
Tomohiro Kusumi [Sun, 6 Nov 2016 05:20:12 +0000 (14:20 +0900)]
sbin/hammer: Add a TODO comment for hammer recover
The way hammer recover simply walks through zone-2 offset from
0 to the end of the volume is inefficient.
Since we know most big-blocks are not for zone-8 (B-Tree zone),
most of the time and I/O spent for testing B-Tree nodes is wasted.
If zone-2 offset can be reverse mapped to zone-X, it no longer
needs to always call recover_top() assuming offsets are for
B-Tree nodes when we know it's mostly wrong.
Reverse mapping can also avoid possible misinterpretation of
zone-2 buffers (i.e. wrong cast to hammer_node_ondisk_t).
zrj [Sun, 6 Nov 2016 08:25:19 +0000 (10:25 +0200)]
tic(1): Add missing beforedepend for termsort.c.
Noticed-by: swildner
Antonio Huete Jimenez [Sat, 5 Nov 2016 17:56:27 +0000 (18:56 +0100)]
installer - Enhance test_storage program
- Warn when run as non-root.
- Actually print detected disks after execution.
Antonio Huete Jimenez [Sat, 5 Nov 2016 14:22:07 +0000 (15:22 +0100)]
installer - Reorganize test directory a bit
Matthew Dillon [Sat, 5 Nov 2016 19:36:31 +0000 (12:36 -0700)]
doc - Document 'crypt' option in fstab(5)
* The 'crypt' was documented in swapon(8) but most people would probably
look for it in fstab(5), so put it there too.
Sascha Wildner [Sat, 5 Nov 2016 07:28:51 +0000 (08:28 +0100)]
boot: Use snprintf() when filling command_errbuf[] w/ dynamic content.
Taken-from: FreeBSD
Sascha Wildner [Fri, 4 Nov 2016 18:49:54 +0000 (19:49 +0100)]
Sync zoneinfo database with tzdata2016i from ftp://ftp.iana.org/tz/releases
* Pacific/Tongatapu begins DST on 2016-11-06 at 02:00, ending on
2017-01-15 at 03:00. Assume future observances in Tonga will be
from the first Sunday in November through the third Sunday in
January, like Fiji. (Thanks to Pulu ʻAnau.) Switch to numeric
time zone abbreviations for this zone.
* Northern Cyprus is now +03 year round, causing a split in Cyprus
time zones starting 2016-10-30 at 04:00. This creates a zone
Asia/Famagusta. (Thanks to Even Scharning and Matt Johnson.)
* Antarctica/Casey switched from +08 to +11 on 2016-10-22.
(Thanks to Steffen Thorsen.)
* Several corrections were made for pre-1975 time stamps in Italy.
These affect Europe/Malta, Europe/Rome, Europe/San_Marino, and
Europe/Vatican.
First, the 1893-11-01 00:00 transition in Italy used the new UT
offset (+01), not the old (+00:49:56). (Thanks to Michael
Deckers.)
Second, rules for daylight saving in Italy were changed to agree
with Italy's National Institute of Metrological Research (INRiM)
except for 1944, as follows (thanks to Pierpaolo Bernardi, Brian
Inglis, and Michael Deckers):
The 1916-06-03 transition was at 24:00, not 00:00.
The 1916-10-01, 1919-10-05, and 1920-09-19 transitions were at
00:00, not 01:00.
The 1917-09-30 and 1918-10-06 transitions were at 24:00, not
01:00.
The 1944-09-17 transition was at 03:00, not 01:00. This
particular change is taken from Italian law as INRiM's table,
(which says 02:00) appears to have a typo here. Also, keep the
1944-04-03 transition for Europe/Rome, as Rome was controlled by
Germany then.
The 1967-1970 and 1972-1974 fallback transitions were at 01:00,
not 00:00.
Sascha Wildner [Fri, 4 Nov 2016 17:39:27 +0000 (18:39 +0100)]
<sys/cdefs.h>/<machine/stdint.h>: Fix __STDC_VERSION__ macro.
Noticed-by: zrj
zrj [Fri, 4 Nov 2016 07:24:32 +0000 (09:24 +0200)]
<netinet/ip.h>: Few style(9) fixes.
Looks like IPTOS_CE and IPTOS_ECT were deprecated and should not be used.
Leave for now, still used by ip_ecn.c and ipfw2.c.
Same applies to _IP_VHL version.
No functional change.
zrj [Fri, 4 Nov 2016 07:16:53 +0000 (09:16 +0200)]
<netinet/ip.h>: Add ToS defs for RFC 2474 and 5865.
While there, convert IPTOS_PREC macros to use IPTOS_DSCP definitions.
Taken-from: OpenBSD, FreeBSD(IPTOS_DSCP_VA)
zrj [Thu, 3 Nov 2016 17:19:09 +0000 (19:19 +0200)]
nc(1): Adjust makefile and hook into buildworld.
Given that nc(1) in dports is only installed when dports/libressl is used,
but not available when dports/openssl, it is useful to have base nc(1).
Also it makes a very good test case for LTO toolchain testing:
LTO + NOSHARED + libressl + openbsd compat layer + certain parts of libc.
Since base does not provide libprivate_tls lib, compile tls bits just for nc.
zrj [Thu, 3 Nov 2016 17:15:43 +0000 (19:15 +0200)]
libressl: Few local modifications for netcat.
* Exclude pledge(), has no use for us.
* Make strtoport() static to prevent missing prototype warning.
No functional change for libprivate_crypto or libprivate_ssl.
zrj [Fri, 4 Nov 2016 08:57:12 +0000 (10:57 +0200)]
gcc50: Hook missing source into the libgcc.a.
Missing hidden symbol __sfp_handle_exception.
Fixes linkage: gcc main.c -Wl,--whole-archive -Wl,-v -nostdlib -lgcc -lc
Matthew Dillon [Fri, 4 Nov 2016 06:32:24 +0000 (23:32 -0700)]
build - Fix bootstrap build for csh
* Fix the bootstrap portion of the buildworld for csh on systems which are
updated past the termcap library rename but do not yet have
libprivate_crypto.
Matthew Dillon [Fri, 4 Nov 2016 06:31:53 +0000 (23:31 -0700)]
md5 - Fix seg-fault
* Fix seg-fault due to mixed API use.
Reported-by: marino, others
Imre Vadász [Thu, 3 Nov 2016 20:49:05 +0000 (21:49 +0100)]
if_iwm - Sync rs (rate-selection) API definitions from Linux iwlwifi.
* While there clean up alignments and line wrapping in existing
definitions for rs API in if_iwmreg.h
Taken-From: Linux iwlwifi
Imre Vadász [Thu, 3 Nov 2016 19:53:11 +0000 (20:53 +0100)]
if_iwm - Add iwm_mvm_send_lq_cmd() from Linux iwlwifi to if_iwm_util.c.
Taken-From: Linux iwlwifi
Tomohiro Kusumi [Thu, 3 Nov 2016 15:30:26 +0000 (00:30 +0900)]
sbin/hammer: Fix wrong error message
It's supposed to be "does not match".
Tomohiro Kusumi [Thu, 3 Nov 2016 14:37:09 +0000 (23:37 +0900)]
sbin/mount_hammer: Fix/cleanup test_volumes()
The requirement of this function is basically the same as
__verify_volume() in sbin/hammer/ondisk.c, so make it
__verify_volume() based.
Note that this function doesn't necessarily print the real
reason mount(2) failed, as shown in a comment. This function
just takes a wild guess with sanity checks on volume header.
The real reason is likely to be in dmesg. This originally
didn't exist, but was added by
1a607e3e in 2009.
Tomohiro Kusumi [Thu, 3 Nov 2016 13:37:06 +0000 (22:37 +0900)]
sbin/mount_hammer: Simplify ondisk verification
Unlike sbin/hammer or sbin/newfs_hammer, sbin/mount_hammer
has no restriction regarding I/O size since it's readonly.
It doesn't have to be 16KB as long as I/O size meets the
requirement of read(2) over a raw block device.
Tomohiro Kusumi [Thu, 3 Nov 2016 12:06:14 +0000 (21:06 +0900)]
sys/vfs/hammer: Add HAMMER_ENCODE_UNDO()
Add and use HAMMER_ENCODE_UNDO(offset) instead of
HAMMER_ZONE_ENCODE(HAMMER_ZONE_UNDO_INDEX, offset),
just like other ENCODE macros available for other zones.
Since zone-3 doesn't take volume# by design (see L136 of
sys/vfs/hammer/hammer_disk.h), HAMMER_ENCODE_UNDO() only
takes offset, but not (volume#, offset). In other words,
UNDO/REDO FIFO is in the root volume.
sys/vfs/hammer/hammer_disk.h
* zone 1 (z,v,o): raw volume relative (offset 0 is the volume header)
* zone 2 (z,v,o): raw buffer relative (offset 0 is the first buffer)
* zone 3 (z,o): undo/redo fifo - fixed zone-2 offset array in volume header
* zone 4 (z,v,o): freemap - only real blockmap
* zone 8 (z,v,o): B-Tree - actually zone-2 address
* zone 9 (z,v,o): meta - actually zone-2 address
* zone 10 (z,v,o): large-data - actually zone-2 address
* zone 11 (z,v,o): small-data - actually zone-2 address
zrj [Fri, 28 Oct 2016 07:43:03 +0000 (10:43 +0300)]
build: Fix wrong DPADD for libcrypto.
Follow up to
0fe46dc6296951eb138485d8c6b580bac0488fd8
zrj [Fri, 28 Oct 2016 07:34:34 +0000 (10:34 +0300)]
bsd.libnames.mk: Remove outdated XXX.
No longer applies after
0fe46dc6296951eb138485d8c6b580bac0488fd8
zrj [Thu, 27 Oct 2016 16:25:09 +0000 (19:25 +0300)]
gcc47: Do not include empty header.
Including this header only poisons and triggers ctype_members.cc to be recompiled
on every repeated quickworld together with relink of libstdc++ for no good reason.
zrj [Thu, 27 Oct 2016 15:37:02 +0000 (18:37 +0300)]
tic(1): Restrict termsort.c hack to btools.
It is unclear why this was needed, for now just limit it to BOOTSTRAPPING case.
zrj [Thu, 27 Oct 2016 14:30:09 +0000 (17:30 +0300)]
bsd.sys.mk: Exclude PROG from PHONY list.
This prevents surprises when any PROG would match anything in that list.
Was causing contantly relinking of patch(1) and fetch(1)!
zrj [Thu, 27 Oct 2016 05:26:53 +0000 (08:26 +0300)]
installer: Fix typo in handling null mounts array.
This could have prevented to install on certain custom system layouts.
Tomohiro Kusumi [Thu, 3 Nov 2016 07:03:48 +0000 (16:03 +0900)]
sys/vfs/hammer: Add ascii-art for volume layout
Tomohiro Kusumi [Thu, 3 Nov 2016 05:51:32 +0000 (14:51 +0900)]
sys/vfs/hammer: Fix terminology for undo fifo
Always use "UNDO/REDO FIFO" whenever printf/manpage/etc refer to
undo in terms of HAMMER's zone-3 data structure, because that's
how it's been used originally.
But leave comments as they are now. There are tons of comments
that refer to UNDO, but not REDO when they could be UNDO/REDO,
but those mostly come from the fact that REDO was added later.
Imre Vadász [Wed, 2 Nov 2016 21:08:28 +0000 (22:08 +0100)]
if_iwm - Sync statistics API definitions with Linux iwlwifi.
Taken-From: Linux iwlwifi
John Marino [Wed, 2 Nov 2016 14:54:36 +0000 (09:54 -0500)]
Bump __DragonFly_version after md/crypt/ressl conflict resolution
Sascha Wildner [Tue, 1 Nov 2016 23:14:40 +0000 (00:14 +0100)]
<sys/cdefs.h>: Remove some duplicate definitions.
Imre Vadász [Tue, 1 Nov 2016 15:34:14 +0000 (16:34 +0100)]
if_iwm - Cleanup mbufq draining. Add iwm_xmit_queue_drain() like in iwn(4).
* iwm_xmit_queue_drain() calls ieee80211_free_node(), removing a possible
memory leak, compared to using just mbufq_drain().
* Remove duplicate mbufq_drain() from iwm_mvm_rm_sta(), this should be
handled in the caller.
Tomohiro Kusumi [Tue, 1 Nov 2016 12:44:06 +0000 (21:44 +0900)]
sys/vfs/hammer: Use HAMMER_OFF_{SHORT,LONG}_ENCODE()
HAMMER_OFF_XXX_ENCODE() macros cast result to hammer_off_t,
which is uint64_t, but makes no difference in these cases.
Tomohiro Kusumi [Tue, 1 Nov 2016 11:34:05 +0000 (20:34 +0900)]
sbin/hammer: Fix print_blockmap() for non-root volume
It should work the same regardless of root/non-root,
though vol0_blockmap[] is empty for non-root volumes.
Tomohiro Kusumi [Tue, 1 Nov 2016 11:31:29 +0000 (20:31 +0900)]
sbin/newfs_hammer: Fix print_volume() for non-root volume
It should work the same regardless of root/non-root.
Imre Vadász [Tue, 1 Nov 2016 11:08:10 +0000 (12:08 +0100)]
if_iwm - Use chan list from ieee80211_scan_state for scan, not ic_channels.
* Limiting the channel list with "ifconfig wlan0 chanlist ..." now will
actually set the list of channels scanned by iwm.
Imre Vadász [Mon, 31 Oct 2016 23:34:28 +0000 (00:34 +0100)]
if_iwm - Factor out and improve iwm_mvm_scan_rxon_flags() in if_iwm_scan.c.
* Actually look at the first channel in the list. If it's a 2.4GHz channel,
set IWM_PHY_BAND_24 flag. The IWM_PHY_BAND_5 flag is 0 anyway, so we
don't need to look further.
* While there factor out the iwm_mvm_rrm_scan_needed() tlv capability check.
Taken-From: Linux iwlwifi
Imre Vadász [Mon, 31 Oct 2016 20:44:12 +0000 (21:44 +0100)]
if_iwm - Remove support for fw older than -17 and -22
* iwm(4) didn't use any of these definitions yet, anyway.
Taken-From: Linux iwlwifi (commit
4b87e5af638b6056bd6c20b0954d09a5a58633be)