jsg [Thu, 14 May 2015 11:52:43 +0000 (11:52 +0000)]
rev 1.3 introduced a check to an if statement without adding braces.
Claudio points out the size is checked by an earlier test so just
remove it to restore the original handling of the partial octet case.
Discussed with claudio and gilles.
rpe [Thu, 14 May 2015 10:58:55 +0000 (10:58 +0000)]
Extend autoinstall(8) to allow for <hostname>-<mode>.conf response files
and to put response files in a subdir of the webserver's document root.
Based on diffs from Nathanael Rensen, thanks!
While here fix a buglet introduced by the $_server -> $AI_SERVER change.
OK krw, halex
mpi [Thu, 14 May 2015 10:55:28 +0000 (10:55 +0000)]
Allocate the input packet handler as part of the trunk_port structure
since they have the same lifetime.
Requested by and ok dlg@
deraadt [Thu, 14 May 2015 10:30:56 +0000 (10:30 +0000)]
sync
jsg [Thu, 14 May 2015 03:13:20 +0000 (03:13 +0000)]
Remove BOARD_ID* tests that do the same thing for every imx board.
Tested by and ok djm@
jsg [Thu, 14 May 2015 02:56:01 +0000 (02:56 +0000)]
Stop including CFLAGS in CXXFLAGS it creates problems when mixing
C and C++ when wanting to pass flags to only the C compiler.
Tested in a ports bulk build by naddy and the handful of problems
found were fixed by naddy, jca and pascal.
djm [Thu, 14 May 2015 02:10:29 +0000 (02:10 +0000)]
rework imxenet hardware address setup: if COTP doesn't provide an
address, check whether the firmware/bootloader has already programmed
one. If all else fails, use ether_fakeaddr(); feedback and ok jsg@
(committed from Novena)
deraadt [Thu, 14 May 2015 00:15:30 +0000 (00:15 +0000)]
sync
bluhm [Wed, 13 May 2015 21:01:54 +0000 (21:01 +0000)]
If crypt(3) is called with an unknown setting, return NULL instead
of some undefined value.
OK tedu@
jsg [Wed, 13 May 2015 10:42:46 +0000 (10:42 +0000)]
test mbuf pointers against NULL not 0
ok krw@ miod@
mpi [Wed, 13 May 2015 08:16:01 +0000 (08:16 +0000)]
Get rid of the last "#if NTRUNK" by overwriting trunk ports' output
function.
ok claudio@, reyk@
mpi [Wed, 13 May 2015 08:01:21 +0000 (08:01 +0000)]
Remove useless "bridge.h" include.
ok miod@
gilles [Wed, 13 May 2015 07:34:49 +0000 (07:34 +0000)]
The enqueuer should never encouter the "From " separator in its headers, as
it is added by the mda. If it sees one, it means that the enqueuer was used
as the mda and it needs to strip it otherwise the message will end with two
delimiters later down the road. Same applies to "Return-Path".
bug experienced by James Turner, confirmed by Giovanni.
fix suggested by Todd Miller, diff ok eric@
uebayasi [Wed, 13 May 2015 05:29:57 +0000 (05:29 +0000)]
Define END() to set ELF symbol size.
OK miod@
jsg [Wed, 13 May 2015 02:39:28 +0000 (02:39 +0000)]
Having the same settings on the same phy across multiple boards is just
asking too much in the arm world.
Translate the fec parameters from the novena dtb to set a different clock
skew to the same micrel phy used on sabre lite. The novena dtb sets
txd[0-3]-skew-ps to 3000, the sabre lite sets them to 0. When run through
the shifting/oring process from the micrel phy driver in FreeBSD this ends
up being a write of 0xffff on the novena and 0x0000 on sabre lite when
writing to TX_DATA_PAD_SKEW.
This change resolves the stability problems djm was seeing with imxenet
on novena.
ok djm@
nicm [Tue, 12 May 2015 22:40:38 +0000 (22:40 +0000)]
To replace c0-*, add a high watermark to the pty event, and also backoff
when the any of the ttys the pane is going to write to has buffered
enough data.
kettenis [Tue, 12 May 2015 21:05:42 +0000 (21:05 +0000)]
Revert rev 1.3. While that change appears to make the behaviour similar to
binutils 2.15, it introduces bogus failures when inter-library dependencies
are present.
With the binutils 2.17 behaviour restored, --as-needed will cause linker
failures when necessary inter-library dependencies are not present. This is
actually a good thing as binutils 2.15 will silently produce executables
that will fail at runtime in those cases.
ok miod@
kettenis [Tue, 12 May 2015 20:20:18 +0000 (20:20 +0000)]
Make sure the rx ring lwm is set to at least 4. As far as we know, all
hardware variants need at least 4 descriptors on the rx ring to be able to
receive packets. Should fix the issue reported by Christian Schulte on
bugs@.
ok mikeb@, sthen@
guenther [Tue, 12 May 2015 20:14:09 +0000 (20:14 +0000)]
Add dlclose(3) to SEE ALSO
ok millert@ jmc@ schwarze@
guenther [Tue, 12 May 2015 20:13:15 +0000 (20:13 +0000)]
Document pthread_atfork(3)'s interaction with dlclose(3)
Use Xr instead of Fn for functions documented on other manpages
ok millert@ jmc@ schwarze@
millert [Tue, 12 May 2015 19:38:13 +0000 (19:38 +0000)]
Race less.
nicm [Tue, 12 May 2015 19:36:08 +0000 (19:36 +0000)]
Tidy blank lines when outputting server info.
ratchov [Tue, 12 May 2015 18:39:30 +0000 (18:39 +0000)]
remove the sc->isopen flag, as sc->flags already hold the
same information.
ratchov [Tue, 12 May 2015 18:32:49 +0000 (18:32 +0000)]
store byte count in a size_t and switch from uiomovei() to uiomove()
ratchov [Tue, 12 May 2015 18:23:38 +0000 (18:23 +0000)]
Remove unused code & definitions: midi_attach(), midi_unit_count(),
midi_getinfo(), midi_writebytes(). They were "used" by the sequencer
interface which we removed years ago.
ratchov [Tue, 12 May 2015 18:05:43 +0000 (18:05 +0000)]
Don't hold the audio mutex when calling uiomove(), as uiomove()
may sleep in case of a page fault
ratchov [Tue, 12 May 2015 16:35:23 +0000 (16:35 +0000)]
Remove references to (deleted) auconv.h and mulaw.h files. Fixes
i386 build. Found by Mark Patruck <mark at wrapped.cx>, thanks!
nicm [Tue, 12 May 2015 15:29:29 +0000 (15:29 +0000)]
Add a session_alerts format which is a list of all the alerts in the
current session in symbolic form (something like "0!,4~,5!"). Use this
in the default set-titles-string. Prompted by a request from Jan ONDREJ.
nicm [Tue, 12 May 2015 15:27:46 +0000 (15:27 +0000)]
Add bell-action "other" to pass through bells in all windows except the
current, suggested by Jan ONDREJ.
mikeb [Tue, 12 May 2015 12:56:47 +0000 (12:56 +0000)]
Fixup potential use after free and a memory leak.
Found by Maxime Villard <max at m00nbsd ! net> with the Brainy Code Scanner,
thanks!
mpi [Tue, 12 May 2015 12:35:10 +0000 (12:35 +0000)]
This file is only compiled when "pseudo-device bridge" is included in
your kernel configuration, no need for a #if NBRIDGE > 0 dance.
mpi [Tue, 12 May 2015 12:27:17 +0000 (12:27 +0000)]
MPLS also needs a definition for etherip_output(), fix build without
bridge.
stsp [Tue, 12 May 2015 11:53:13 +0000 (11:53 +0000)]
Add newly matched devices to urtwn(4) hardware list.
from Mikhail mp39590@gmail on tech@
Grammar fixes by sthen@. Made mandoc -T lint clean by me (must escape 'D1').
stsp [Tue, 12 May 2015 11:46:15 +0000 (11:46 +0000)]
Merge various urtwn(4) fixes from FreeBSD:
Match additional devices (RTL8188CU_3, DWA123D1, DWA125D1).
Fix efuse reading (fixes potential error in MAC address read from efuse).
Replace some magic numbers with macros.
Based on FreeBSD r270191, r273589, r281918, r281592, r282119, and r282623
via Mikhail mp39590@gmail on tech@
ok sthen@
stsp [Tue, 12 May 2015 11:19:35 +0000 (11:19 +0000)]
Fix definition of R92C_USTIME_TSF.
from FreeBSD r282623 via Mikhail mp39590@gmail on tech@
stsp [Tue, 12 May 2015 11:15:42 +0000 (11:15 +0000)]
regen
stsp [Tue, 12 May 2015 11:15:17 +0000 (11:15 +0000)]
usbdevs_data.h
stsp [Tue, 12 May 2015 11:14:06 +0000 (11:14 +0000)]
Add 4 additional urtwn(4) device IDs.
from FreeBSD via Mikhail mp39590@gmail on tech@
mikeb [Tue, 12 May 2015 09:30:35 +0000 (09:30 +0000)]
Drop and reacquire the kernel lock in the vfs_shutdown and "cold"
portions of msleep and tsleep to give interrupts a chance to run
on other CPUs.
Tweak and OK kettenis
yasuoka [Tue, 12 May 2015 05:30:24 +0000 (05:30 +0000)]
Fix an uninitialized variable access when pptp.hostname is configured.
jsg [Tue, 12 May 2015 04:40:31 +0000 (04:40 +0000)]
imxiic is known to be broken don't try attaching it on utilite
jsg [Tue, 12 May 2015 04:31:10 +0000 (04:31 +0000)]
Raise VM_PHYSSEG_MAX to two and load an additional physical memory
segment if u-boot reports it. Needed for the utilite where u-boot
reports two 1GB segments of physical memory.
From Patrick Wildt in bitrig with some additional sanity checks added.
miod [Tue, 12 May 2015 04:22:01 +0000 (04:22 +0000)]
Put the right number of leading underscores in ___cerror and __cerror, for
the sake of libpthread. Probably missed during the a.out->ELF switch, I
can't believe this went unnoticed for so long.
jsg [Tue, 12 May 2015 03:23:15 +0000 (03:23 +0000)]
Move the logit() prototype so parse.y will see it.
ok yasuoka@
jsg [Tue, 12 May 2015 02:49:48 +0000 (02:49 +0000)]
Add missing braces.
Tested by and ok djm@
guenther [Tue, 12 May 2015 02:44:06 +0000 (02:44 +0000)]
chflagsat() is also async-signal-safe
jsg [Tue, 12 May 2015 02:33:39 +0000 (02:33 +0000)]
The i211 does not support an external EEPROM only a OTP
Internal Non-Volatile Memory (iNVM). Add support for reading
words out of it instead of an EEPROM.
From Patrick Wildt with some more offsets added.
jasper [Mon, 11 May 2015 20:20:33 +0000 (20:20 +0000)]
regen
jasper [Mon, 11 May 2015 20:20:03 +0000 (20:20 +0000)]
add id for virtio rng
ok sf@
pelikan [Mon, 11 May 2015 12:24:06 +0000 (12:24 +0000)]
Make softraid(4) compile with SR_DEBUG by fixing __kprintf__ specifiers.
ok jsing krw
henning [Mon, 11 May 2015 12:22:14 +0000 (12:22 +0000)]
fix a potential use-after-free in pf_state_rm_src_node
found by jsg; ok jsg mikeb
pelikan [Mon, 11 May 2015 12:14:22 +0000 (12:14 +0000)]
explicit_bzero(3) on private data, in case of core dumps (and being a good example)
ok florian jsing
nicm [Mon, 11 May 2015 10:58:22 +0000 (10:58 +0000)]
Left the c0-* options behind in the table.
nicm [Mon, 11 May 2015 10:10:16 +0000 (10:10 +0000)]
Remove the c0-* options which never really worked satisfactorily. Going
to try something else...
mpi [Mon, 11 May 2015 08:51:31 +0000 (08:51 +0000)]
Refresh sensor values asynchronously.
From David Higgs.
mpi [Mon, 11 May 2015 08:41:43 +0000 (08:41 +0000)]
Take trunk(4) out of ether_input().
Each physical interface of a trunk now gets a specific ifih (interface
input handler) that runs before ether_input().
Tested by sthen@, dlg@, Theo Buehler and <mxb AT alumni.chalmers DOT se>
ok sthen@, dlg@
ratchov [Mon, 11 May 2015 06:52:35 +0000 (06:52 +0000)]
Remove support for ADPCM encoding which isn't used nowadays and
not available to programs anyway.
ratchov [Mon, 11 May 2015 06:46:21 +0000 (06:46 +0000)]
Remove all audio format conversion code from the kernel (btw holding
the kernel_lock), as we already do better conversions in
user-mode. Yet, no need for every single driver to fiddle with the
conversion code as they are done transparently by common MI code. With
help from armani and miod, support from mpi
ok armani@
guenther [Mon, 11 May 2015 06:31:17 +0000 (06:31 +0000)]
Whoops, need to pass through O_ACCMODE flags to the underlying __*_open()
problem pointed out by Mark Patruck (mark (at) wrapped.cx)
guenther [Mon, 11 May 2015 02:01:01 +0000 (02:01 +0000)]
Convert from uiomovei() to uiomove() to prevent short tranfers
diff from natano@bitrig
guenther [Mon, 11 May 2015 01:56:26 +0000 (01:56 +0000)]
Convert from uiomovei() to uiomove() to prevent short tranfers
diff from natano@bitrig, with some additional format and cast tweaks
guenther [Mon, 11 May 2015 00:42:54 +0000 (00:42 +0000)]
When checking flags that will be passed to open(), test the O_ACCMODE portion
separately to avoid false negatives.
ok miod@ millert@
millert [Sun, 10 May 2015 22:35:38 +0000 (22:35 +0000)]
Set POLLHUP even if no valid events were specified as per POSIX.
Since we use the poll backend for select(2), care must be taken not
to set the fd's bit in writefds in this case. A kernel-only flag,
POLLNOHUP, is used by selscan() to tell the poll backend not to
return POLLHUP on EOF. This is currently only used by fifo_poll().
The fifofs regress now passes. OK guenther@
guenther [Sun, 10 May 2015 20:41:19 +0000 (20:41 +0000)]
Add _bfd_ar_spacepadll() for formatting long long values in archive headers.
ok kettenis@
stsp [Sun, 10 May 2015 19:40:56 +0000 (19:40 +0000)]
Fix urtwn(4) 11b performance for RTL8188EU devices.
With 11b APs the firmware retries excessively unless we force it into
11b-only mode instead of mixed b/g. Perhaps a firmware bug, perhaps
a bug in our code elsewhere. For now, this makes it work.
ok kettenis
guenther [Sun, 10 May 2015 18:33:15 +0000 (18:33 +0000)]
In the child after fork, the dl lock has to be forced as its inner spinlock
may have been grabbed by another thread in the parent before the fork
problem report from dcoppa@, ok kettenis@
miod [Sun, 10 May 2015 17:51:27 +0000 (17:51 +0000)]
sync
jsg [Sun, 10 May 2015 15:56:28 +0000 (15:56 +0000)]
VM_MAX_KERNEL_ADDRESS is the end of the kva range not the end of
addressable virtual memory space but arm ports were using a value of
0xffffffff for this.
Instead of using a shared VM_MAX_KERNEL_ADDRESS define add md
VM_KERNEL_SPACE_SIZE defines based on the KERNEL_VM_SIZE values
from the respective machdep.c files.
djm's novena was hitting "panic: bufinit: can't reserve VM for buffers"
without a similiar change.
ok miod@
stsp [Sun, 10 May 2015 15:10:46 +0000 (15:10 +0000)]
My commit adding RTL8188EU did introduce a regression after all.
Repair firmware reset on RTL8188CUS devices (and perhaps others) by
restoring the proper register write order in the non-8188EU code path.
Fixes "urtwn0: timeout waiting for firmware readiness" after device reset.
jsg [Sun, 10 May 2015 05:42:46 +0000 (05:42 +0000)]
limit physical memory to (paddr_t)-PAGE_SIZE (0xfffff000)
novena has 4GB of physical memory and it's u-boot tells us
memstart: 0x10000000
memsize: 0xf0000000
which would previously cause an overflow leading to
"panic: initarm: out of memory"
tweak from and ok miod@
millert [Sun, 10 May 2015 01:45:12 +0000 (01:45 +0000)]
A new version from the original author of math.sed that does exponents
and lots of other neat things. Via FreeBSD
schwarze [Sat, 9 May 2015 20:36:18 +0000 (20:36 +0000)]
For half and reverse line feeds, recognize SUSv2-style escape-digit
sequences in the input stream in addition to traditional BSD-style
escape-control-char sequences because traditional tools, for example
Heirloom roff, produce SUSv2-style sequences. Switch the encoding
of forward half line feeds in the output of -f back to SUSv2 style
because that's likely to work with more tools than the non-standard
traditional BSD-style sequence. Fully document these choices.
Issues originally reported by bapt at FreeBSD.
OK millert@ jmc@, and bapt@ also more or less agrees with the direction.
krw [Sat, 9 May 2015 17:11:26 +0000 (17:11 +0000)]
Cleanup/rejig GPT code to be more readable, support different endian
archs and different sized disk sectors. Make MBR have higher priority
than GPT. Add many paranoia checks and associated DPRINTF's to make
further development easier. Keep everything hidden behind #ifdef
GPT.
Tested and ok doug@ mpi@. Nothing bad seen by millert@.
jsg [Sat, 9 May 2015 10:27:32 +0000 (10:27 +0000)]
Fix an occurance of mtx_enter() that should have been mtx_leave().
This line was originally an splx() before being converted to a mutex.
ok ratchov@
miod [Fri, 8 May 2015 22:56:53 +0000 (22:56 +0000)]
Need explicit <sys/kcore.h> include now.
miod [Fri, 8 May 2015 21:30:37 +0000 (21:30 +0000)]
Make this run on strict alignment architectures.
miod [Fri, 8 May 2015 19:17:20 +0000 (19:17 +0000)]
When changing screen saver parameters with wsconsctl (or any other way to
perform ioctls), check the flags of the currently displayed screen in order
to decide whether the screen saver needs to be retriggered, rather than the
flags of the device we are issuing the ioctl on.
wsconsctl will always use ttyC0, while the X server may run on another
virtual screen (such as ttyC4), and the kernel screen saver would then be
reenabled behind the X server's back.
While there, apply this `should the screen burner get reenabled or disabled?'
logic at the end of every virtual screen switch. The screen burner will now
get reenabled when switching from X11 to a virtual text console, and disabled
when switching back to X.
miod [Fri, 8 May 2015 19:12:51 +0000 (19:12 +0000)]
Add a new `don't read back' flag for variables, to prevent reading their value
after modifying them. Give this flag to `display.focus', since screen
switching is asynchronous, and reading back will return the screen we are
switching from if the switch has not completed yet.
Also, disallow -= and += syntax for display.focus, as it doesn't make any sense.
nicm [Fri, 8 May 2015 16:48:12 +0000 (16:48 +0000)]
Put the tty structs together, and tabify.
nicm [Fri, 8 May 2015 16:44:03 +0000 (16:44 +0000)]
grid_marker_cell is no longer used.
nicm [Fri, 8 May 2015 16:33:29 +0000 (16:33 +0000)]
mode_key_entry can go into mode-key.c; also a few spaces->tabs.
schwarze [Fri, 8 May 2015 16:30:07 +0000 (16:30 +0000)]
Avoid scattering octal escapes all over the place,
use the definitions given at the top of the file.
No functional change.
From bapt at FreeBSD.
nicm [Fri, 8 May 2015 16:23:34 +0000 (16:23 +0000)]
window_choose_mode_item can move into window-choose.c.
nicm [Fri, 8 May 2015 16:18:04 +0000 (16:18 +0000)]
Move input parser structs into input.c (removing fairly useless
saved_cursor_[xy] formats as a side-effect).
nicm [Fri, 8 May 2015 15:56:49 +0000 (15:56 +0000)]
Remove some stuff that accidentally ended up here from portable, and
remove a little-used debug function.
bcallah [Fri, 8 May 2015 12:35:08 +0000 (12:35 +0000)]
Remove ereply prototype since it is never used. Tweak a comment to reflect
ereply removal.
ok jasper@ florian@
schwarze [Fri, 8 May 2015 12:30:27 +0000 (12:30 +0000)]
Fix about ten integer overflows and underflows and a handful of logic
errors in line number handling. Detailed explanations were sent to tech@
on October 18, 2014.
OK doug@, and bapt at FreeBSD says he likes the direction
sthen [Fri, 8 May 2015 12:15:50 +0000 (12:15 +0000)]
Separately track the number of items in alloc_table in a variable. With
the changes for handling template files in r1.293, alloc_table became
a pointer to an array of structs (rather than the array itself), so
nitems(alloc_table) no longer returns the number of elements. As found
by sebastia@, autosize would only try the first allocation scheme,
so installation would fail on small disks. ok henning@
kettenis [Fri, 8 May 2015 10:28:34 +0000 (10:28 +0000)]
Switch i386 and sparc64 to binutils 2.17. More architectures to follow if
this doesn't cause any serious regressions.
ok espie@, pascal@, deraadt@
djm [Fri, 8 May 2015 07:29:00 +0000 (07:29 +0000)]
whitespace
djm [Fri, 8 May 2015 07:26:13 +0000 (07:26 +0000)]
whitespace at EOL
jsg [Fri, 8 May 2015 07:25:36 +0000 (07:25 +0000)]
Fix the indentation. While originally I thought this was an if statement
with missing braces ratchov checked the specs to see the current
behaviour is correct, just poorly formatted.
ok ratchov@
djm [Fri, 8 May 2015 06:45:13 +0000 (06:45 +0000)]
moar whitespace at eol
djm [Fri, 8 May 2015 06:41:56 +0000 (06:41 +0000)]
whitespace at EOL
jsg [Fri, 8 May 2015 04:47:27 +0000 (04:47 +0000)]
Add missing sabresd case statements. As sabresd_devs[] includes imxgpio
an attempt to boot on sabresd would have tried to deref NULL function
pointers.
djm [Fri, 8 May 2015 03:56:51 +0000 (03:56 +0000)]
whitespace at EOL
jsg [Fri, 8 May 2015 03:38:26 +0000 (03:38 +0000)]
Add initial board specific parts of novena support, tested by djm@
While novena has it's own board id it's u-boot does not set it.
"setenv machid 10ad" in u-boot will set the novena board id (0x10ad/4269).
dtucker [Fri, 8 May 2015 03:25:07 +0000 (03:25 +0000)]
Use xcalloc for permitted_adm_opens instead of xmalloc to ensure it's zeroed.
Fixes post-auth crash with permitopen=none. bz#2355, ok djm@
djm [Fri, 8 May 2015 03:17:49 +0000 (03:17 +0000)]
don't choke on new-format private keys encrypted with an AEAD cipher;
bz#2366, patch from Ron Frederick; ok markus@
mikeb [Thu, 7 May 2015 18:30:27 +0000 (18:30 +0000)]
msleep(9) must prevent kernel from attempting a context switch
during autoconf and after panics.
Tweak and OK guenther, OK miod