openbsd
10 years agoimprove logging messages and style; requested by and ok reyk
mikeb [Fri, 16 May 2014 08:46:23 +0000 (08:46 +0000)]
improve logging messages and style; requested by and ok reyk

10 years agoDo not put any link-layer address on the per-ifp lists or on the RB-
mpi [Fri, 16 May 2014 08:21:54 +0000 (08:21 +0000)]
Do not put any link-layer address on the per-ifp lists or on the RB-
Tree.

Since interfaces only support one link-layer address accessible via the
if_sadl member, there's no need to have it elsewhere.  This improves
various address lookups because the first element of the list, the link-
layer address, won't necessarily be discarded.

Finally remove the empty netmask associated to every link-layer address.
This hack was needed to (ab)use the address & netmask comparison code to
do a strcmp() on the interface name embedded in the sdl_data field.

ok henning@, claudio@

10 years agoFix memory leaks upon failure.
miod [Thu, 15 May 2014 21:07:10 +0000 (21:07 +0000)]
Fix memory leaks upon failure.
ok beck@

10 years agoReplace ASN1_GENERALIZEDTIME_adj(), ASN1_UTCTIME_adj() and
miod [Thu, 15 May 2014 21:06:10 +0000 (21:06 +0000)]
Replace ASN1_GENERALIZEDTIME_adj(), ASN1_UTCTIME_adj() and
ASN1_TIME_to_generalizedtime() with wrappers around their former
implementations, making sure memory allocated is freed in all failure cases.

help and ok from beck@ and Brendan MacDonell.

10 years agomore for STANDARDS: unsupported lists (noted by guenther) and env vars;
jmc [Thu, 15 May 2014 19:42:56 +0000 (19:42 +0000)]
more for STANDARDS: unsupported lists (noted by guenther) and env vars;

10 years agoremove unused variable
chl [Thu, 15 May 2014 19:40:42 +0000 (19:40 +0000)]
remove unused variable

ok espie@

10 years agoNo longer make pqueue.h a public (and installed) header file. Replace its
miod [Thu, 15 May 2014 19:39:46 +0000 (19:39 +0000)]
No longer make pqueue.h a public (and installed) header file. Replace its
inclusion in <ssl/dtls1.h> by the benign
`struct _pqueue; typedef struct _pqueue *pqueue;'.

Note that said dtls1.h should probably be considered a private header as well...

ok beck@

10 years agoflesh out STANDARDS; ok guenther
jmc [Thu, 15 May 2014 19:38:06 +0000 (19:38 +0000)]
flesh out STANDARDS; ok guenther

10 years agouse <> for tables;
jmc [Thu, 15 May 2014 19:36:45 +0000 (19:36 +0000)]
use <> for tables;
original report from creamy;
diff from Frank Brodbeck, tweaked

10 years agoGet pqueue.h from the libssl sources rather than relying upon it being
miod [Thu, 15 May 2014 19:23:14 +0000 (19:23 +0000)]
Get pqueue.h from the libssl sources rather than relying upon it being
installed.

10 years ago- typo in comment
chl [Thu, 15 May 2014 19:18:23 +0000 (19:18 +0000)]
- typo in comment
- memset() is not needed after a calloc()
- add missing DBG_ENTER;
- use SEEK_SET instead of 0 in an lseek() call

ok krw@

10 years agoMore KNF.
jsing [Thu, 15 May 2014 15:55:55 +0000 (15:55 +0000)]
More KNF.

10 years agoAdd additional chacha20-poly1305 test vectors.
jsing [Thu, 15 May 2014 15:50:25 +0000 (15:50 +0000)]
Add additional chacha20-poly1305 test vectors.

Thanks to Piotr Sikora for providing a reference to the source.

10 years agoUse C99 initialisers for the AEAD struct.
jsing [Thu, 15 May 2014 15:44:19 +0000 (15:44 +0000)]
Use C99 initialisers for the AEAD struct.

10 years agoAdd chacha20-poly1305 test vector to regress.
jsing [Thu, 15 May 2014 15:41:44 +0000 (15:41 +0000)]
Add chacha20-poly1305 test vector to regress.

10 years agoKNF.
jsing [Thu, 15 May 2014 15:13:56 +0000 (15:13 +0000)]
KNF.

10 years agoAdd a ChaCha20-Poly1305 AEAD EVP implementation to libcrypto, from Adam
jsing [Thu, 15 May 2014 15:09:01 +0000 (15:09 +0000)]
Add a ChaCha20-Poly1305 AEAD EVP implementation to libcrypto, from Adam
Langley's Chromium OpenSSL patches.

ok miod@

10 years agoKNF.
jsing [Thu, 15 May 2014 14:06:13 +0000 (14:06 +0000)]
KNF.

10 years agoAdd a regress test for AEAD, based on Adam Langley's code.
jsing [Thu, 15 May 2014 13:56:14 +0000 (13:56 +0000)]
Add a regress test for AEAD, based on Adam Langley's code.

10 years agoAdd an AEAD EVP interface to libcrypto, along with AES-GCM AEAD
jsing [Thu, 15 May 2014 13:53:46 +0000 (13:53 +0000)]
Add an AEAD EVP interface to libcrypto, along with AES-GCM AEAD
implementations. This largely pulls in Adam Langley's AEAD patches from
Chromium's OpenSSL.

ok miod@

10 years agofix the band-aid error.
espie [Thu, 15 May 2014 13:14:15 +0000 (13:14 +0000)]
fix the band-aid error.
tedu was bitten by the RETARDED interface in sha2.h (who names string LENGTH
things that are actually buffer SIZES) ?
make -C mode work again.

okay beck@

10 years agoStop being a dummy... presumably these are left overs from pedantic mode
jsing [Thu, 15 May 2014 11:25:59 +0000 (11:25 +0000)]
Stop being a dummy... presumably these are left overs from pedantic mode
that were not wrapped with #if PEDANTIC.

ok miod@

10 years agoregen
mpi [Thu, 15 May 2014 09:47:55 +0000 (09:47 +0000)]
regen

10 years agoXHCI -> xHCI
mpi [Thu, 15 May 2014 09:47:26 +0000 (09:47 +0000)]
XHCI -> xHCI

10 years agoUse a more appropriate malloc(9) type than M_TEMP for the keymap memory
mpi [Thu, 15 May 2014 09:29:38 +0000 (09:29 +0000)]
Use a more appropriate malloc(9) type than M_TEMP for the keymap memory
since it is only freed when your keyboard is detached.

10 years agoDo not use nd6_rtrequest() for addresses configured on loopback
mpi [Thu, 15 May 2014 09:05:13 +0000 (09:05 +0000)]
Do not use nd6_rtrequest() for addresses configured on loopback
interfaces.

ok mikeb@

10 years agoUse log_warn() to include errno if write() fails.
reyk [Thu, 15 May 2014 07:56:26 +0000 (07:56 +0000)]
Use log_warn() to include errno if write() fails.

From thib

10 years agoDeal with - actually, ignore - route info messages on the listening side.
jca [Thu, 15 May 2014 05:03:24 +0000 (05:03 +0000)]
Deal with - actually, ignore - route info messages on the listening side.
Issue reported by consus at gmx dot com on bugs@.
ok bluhm@

10 years agoknote_processexit() needs the thread to pass down to FRELE(), so pass it
guenther [Thu, 15 May 2014 04:43:25 +0000 (04:43 +0000)]
knote_processexit() needs the thread to pass down to FRELE(), so pass it
the exiting thread instead of assuming that that's ps_mainproc.
Also, panic no matter which thread of init takes it down.

ok tedu@

10 years agoPrefer prsignal() to send process signals
guenther [Thu, 15 May 2014 04:36:33 +0000 (04:36 +0000)]
Prefer prsignal() to send process signals

10 years agoMove from struct proc to process the reference-count-holding pointers
guenther [Thu, 15 May 2014 03:52:25 +0000 (03:52 +0000)]
Move from struct proc to process the reference-count-holding pointers
to the process's vmspace and filedescs.  struct proc continues to
keep copies of the pointers, copying them on fork, clearing them
on exit, and (for vmspace) refreshing on exec.
Also, make uvm_swapout_threads() thread aware, eliminating p_swtime
in kernel.

particular testing by ajacoutot@ and sebastia@

10 years agostop flushing streams in abort(). it's hackish and unsafe, and no longer
tedu [Wed, 14 May 2014 21:54:20 +0000 (21:54 +0000)]
stop flushing streams in abort(). it's hackish and unsafe, and no longer
required. try to document this fact and some of the history.
with feedback from deraadt guenther millert

10 years agoimprove style - basically, what my fingers did at Marrakesh airport when
henning [Wed, 14 May 2014 21:48:50 +0000 (21:48 +0000)]
improve style - basically, what my fingers did at Marrakesh airport when
waiting for the flight home from m2k14. ok reyk

10 years agosync
todd [Wed, 14 May 2014 21:33:50 +0000 (21:33 +0000)]
sync

10 years agoAdjust route outputs to the new lo(4) MTU which is now the same on all archs.
claudio [Wed, 14 May 2014 19:10:21 +0000 (19:10 +0000)]
Adjust route outputs to the new lo(4) MTU which is now the same on all archs.
Makes this regress work again.

10 years agoThere is no reason why one needs to have a lo(4) MTU that is arch specific.
claudio [Wed, 14 May 2014 19:08:33 +0000 (19:08 +0000)]
There is no reason why one needs to have a lo(4) MTU that is arch specific.
Also making the size 32k + some extra is not very helpful since we don't
do something like zerocopy or sendfile. Just go back to 32768 and we can
bikeshed over the right value later on once we have features that benefit
from it. OK mpi@

10 years agoRemove second arg in SetMouseSpeed(), it is always 1200.
shadchin [Wed, 14 May 2014 18:28:22 +0000 (18:28 +0000)]
Remove second arg in SetMouseSpeed(), it is always 1200.

ok miod@

10 years agoRemove mouse.baudrate, it is always 1200.
shadchin [Wed, 14 May 2014 18:23:22 +0000 (18:23 +0000)]
Remove mouse.baudrate, it is always 1200.

ok miod@

10 years agoRemove unused code.
shadchin [Wed, 14 May 2014 18:18:01 +0000 (18:18 +0000)]
Remove unused code.

ok miod@

10 years agosync
todd [Wed, 14 May 2014 18:15:26 +0000 (18:15 +0000)]
sync

10 years agoRemove unused variables.
shadchin [Wed, 14 May 2014 18:14:09 +0000 (18:14 +0000)]
Remove unused variables.

ok miod@

10 years agowsmoused() called if cmd == WSDISPLAYIO_WSMOUSED only,
shadchin [Wed, 14 May 2014 18:11:24 +0000 (18:11 +0000)]
wsmoused() called if cmd == WSDISPLAYIO_WSMOUSED only,
don't need check cmd twice.

ok miod@

10 years agolet cpp do some of the work
tedu [Wed, 14 May 2014 16:02:34 +0000 (16:02 +0000)]
let cpp do some of the work

10 years agospell out base64 in error messages
tedu [Wed, 14 May 2014 15:56:41 +0000 (15:56 +0000)]
spell out base64 in error messages

10 years agosave some memory by reducing buffer sizes
tedu [Wed, 14 May 2014 15:55:11 +0000 (15:55 +0000)]
save some memory by reducing buffer sizes

10 years agorecode base64 hashes if necessary
tedu [Wed, 14 May 2014 15:52:24 +0000 (15:52 +0000)]
recode base64 hashes if necessary

10 years agorealloc like a boss, so checksum verification doesn't take minutes.
tedu [Wed, 14 May 2014 15:33:41 +0000 (15:33 +0000)]
realloc like a boss, so checksum verification doesn't take minutes.

10 years agoAdd poly1305 to libcrypto utilising Andrew Moon's public domain
jsing [Wed, 14 May 2014 14:46:35 +0000 (14:46 +0000)]
Add poly1305 to libcrypto utilising Andrew Moon's public domain
implementation.

ok miod@

10 years agoSync ktable code with bgpd to fetch, store and perform lookups in
mikeb [Wed, 14 May 2014 09:42:22 +0000 (09:42 +0000)]
Sync ktable code with bgpd to fetch, store and perform lookups in
multiple routing tables.  Currently it doesn't do anything useful
but it's a prerequisite for any future work in this direction.

ok sthen, reyk, blambert

10 years agoChange key-table to mode-table to allow for some future work. From Keith
nicm [Wed, 14 May 2014 06:45:35 +0000 (06:45 +0000)]
Change key-table to mode-table to allow for some future work. From Keith
Amling.

10 years agoSimplify copy lines, from Keith Amling.
nicm [Wed, 14 May 2014 06:39:58 +0000 (06:39 +0000)]
Simplify copy lines, from Keith Amling.

10 years agoNow that cmdlists are reference counted, there is no need for two-step
nicm [Wed, 14 May 2014 06:21:19 +0000 (06:21 +0000)]
Now that cmdlists are reference counted, there is no need for two-step
deletion via the dead_key_bindings tree. From Keith Amling.

10 years agoDon't allow multiple buffers with the same name, from Thomas Adam.
nicm [Tue, 13 May 2014 22:54:18 +0000 (22:54 +0000)]
Don't allow multiple buffers with the same name, from Thomas Adam.

10 years agosync some of the text of ENVIRONMENT with more(1), and cut some
jmc [Tue, 13 May 2014 22:07:03 +0000 (22:07 +0000)]
sync some of the text of ENVIRONMENT with more(1), and cut some
verbiage;

ok millert

10 years agoWe don't need no stinkin' 'extern long dev_bsize;'. At least not since
krw [Tue, 13 May 2014 22:06:06 +0000 (22:06 +0000)]
We don't need no stinkin' 'extern long dev_bsize;'. At least not since
fsck_ffs stopped providing it.

10 years agofor non-valid combinations with `#', instead of trying to list them all,
jmc [Tue, 13 May 2014 20:51:00 +0000 (20:51 +0000)]
for non-valid combinations with `#', instead of trying to list them all,
simply say all other combinations are behaviour undefined;

ok guenther, who reminded me to change these too;

10 years ago- document two escape sequences as extensions in STANDARDS
jmc [Tue, 13 May 2014 20:37:48 +0000 (20:37 +0000)]
- document two escape sequences as extensions in STANDARDS
- document undefined behaviour for unlisted formats in
combination with `#'; guenther helped me organise this into a saner
piece of text

ok guenther

10 years agoWhile Rev 1.285 fixed a RB tree corruption it caused a TAILQ corruption
claudio [Tue, 13 May 2014 14:33:25 +0000 (14:33 +0000)]
While Rev 1.285 fixed a RB tree corruption it caused a TAILQ corruption
in the case where the rdomain was not switched. Make sure ifa_add() is
only called if ifa_del() was called previously. Hopefully we got all the
corruption fixed.
With and OK mpi@

10 years agopass SA initiator not the exchange initator to sa_address(); ok mikeb@
markus [Tue, 13 May 2014 14:24:35 +0000 (14:24 +0000)]
pass SA initiator not the exchange initator to sa_address(); ok mikeb@

10 years agovarious macro fixes;
jmc [Tue, 13 May 2014 14:05:02 +0000 (14:05 +0000)]
various macro fixes;

10 years agozap trailing whitespace;
jmc [Tue, 13 May 2014 14:01:41 +0000 (14:01 +0000)]
zap trailing whitespace;

10 years agoohash stuff in libutil now;
jmc [Tue, 13 May 2014 14:00:59 +0000 (14:00 +0000)]
ohash stuff in libutil now;

10 years agoNuke another pointless dev_bsize variable always set to 512 and use
krw [Tue, 13 May 2014 12:51:40 +0000 (12:51 +0000)]
Nuke another pointless dev_bsize variable always set to 512 and use
DEV_BSIZE. David Vasek says it still works the same on 4K-byte-sector
device. i.e.  prints one value incorrectly. For which a fix should
be following soon.

10 years agoskip incomplete entries in /etc/hosts and /etc/networks
eric [Tue, 13 May 2014 11:57:35 +0000 (11:57 +0000)]
skip incomplete entries in /etc/hosts and /etc/networks
fix null deref spotted by Seth Hanford, pinpointed by dtucker@

ok florian@

10 years agoRemove some m->m_pkthdr.rcvif dereference to help for upcoming
mpi [Tue, 13 May 2014 09:30:27 +0000 (09:30 +0000)]
Remove some m->m_pkthdr.rcvif dereference to help for upcoming
receiving pointer -> index conversion.  No functional change.

ok claudio@, mikeb@, lteo@

10 years agoIf multiple arguments are given to new-session, new-window,
nicm [Tue, 13 May 2014 08:08:32 +0000 (08:08 +0000)]
If multiple arguments are given to new-session, new-window,
split-window, respawn-window or respawn-pane, pass them directly to
execvp() to help avoid quoting problems. One argument still goes to "sh
-c" like before. Requested by many over the years. Patch from J Raynor.

10 years agoAdd a copy mode key binding to copy to a named buffer. From J Raynor.
nicm [Tue, 13 May 2014 07:54:20 +0000 (07:54 +0000)]
Add a copy mode key binding to copy to a named buffer. From J Raynor.

10 years agoAdd support for named buffers. If you don't name a buffer, things work
nicm [Tue, 13 May 2014 07:34:35 +0000 (07:34 +0000)]
Add support for named buffers. If you don't name a buffer, things work
much as before - buffers are automatically named "buffer0000",
"buffer0001" and so on and ordered as a stack. Buffers can be named
explicitly when creating ("loadb -b foo" etc) or renamed ("setb -b
buffer0000 -n foo"). If buffers are named explicitly, they are not
deleted when buffer-limit is reached. Diff from J Raynor.

10 years agoSince this is limited to ffs, use ufsinfo_t instead of ino_t.
guenther [Tue, 13 May 2014 05:50:24 +0000 (05:50 +0000)]
Since this is limited to ffs, use ufsinfo_t instead of ino_t.
Grow the array of cached inodes exponentially instead of arithmetically.
Prefer sizeof(*pointer) over sizeof(type) in mallocs.
Don't leak memory in searchdir().
Fix multiple bugs in the handling of indirect blocks, including reuse of
a static buffer in a recursive function and failure to track the remaining
size of the directory blocks to process

Tested with a directory *doubly* indirect block!
ok krw@

10 years agoAll callers of helpfile() are #ifdef SMALL so only define that
millert [Tue, 13 May 2014 02:34:58 +0000 (02:34 +0000)]
All callers of helpfile() are #ifdef SMALL so only define that
function when SMALL is not defined.  OK guenther@

10 years agoRefactors "dummyrun" flag as "dryrun" with an approach similar to the one used
andre [Mon, 12 May 2014 21:30:42 +0000 (21:30 +0000)]
Refactors "dummyrun" flag as "dryrun" with an approach similar to the one used
for client/server processes checks through $self->{down} state control
variable, while here, do not instantiate a client when noclient is defined,
reflecting server object instantiation behaviour.

ok bluhm

10 years agoRemove useless dev_bsize variable and use DEV_BSIZE as needed. Divide
krw [Mon, 12 May 2014 21:10:35 +0000 (21:10 +0000)]
Remove useless dev_bsize variable and use DEV_BSIZE as needed. Divide
sblock_try[] entries by DEV_BSIZE to get proper 512-byte-block address
for bread().

Confirmed to still work on 4k-byte sector devices by David Vasek.

10 years agoFix format strings involving time_t arguments, fixes `ntpd -d' on sparc (and
miod [Mon, 12 May 2014 20:50:46 +0000 (20:50 +0000)]
Fix format strings involving time_t arguments, fixes `ntpd -d' on sparc (and
probably more 32-bit platforms).

Problem noticed by tobiasu@; ok tobiasu@ dtucker@ sthen@ benno@

10 years agoMove GTT management for Sandy Bridge and up into inteldrm(4). This makes
kettenis [Mon, 12 May 2014 19:29:16 +0000 (19:29 +0000)]
Move GTT management for Sandy Bridge and up into inteldrm(4).  This makes
it possible to use the non-mappable part of the GTT, prepares the way for
using the PPGTT and reduces the diffs with Linux.

ok jsg@

10 years agothese files have moved to libutil
espie [Mon, 12 May 2014 19:27:24 +0000 (19:27 +0000)]
these files have moved to libutil

10 years ago...better if I actually unplug it, Makefile doesn't do it
espie [Mon, 12 May 2014 19:25:16 +0000 (19:25 +0000)]
...better if I actually unplug it, Makefile doesn't do it

10 years agoRemove AES_bi_ige_encrypt() from libcrypto. This routine is supposed to use
miod [Mon, 12 May 2014 19:19:55 +0000 (19:19 +0000)]
Remove AES_bi_ige_encrypt() from libcrypto. This routine is supposed to use
two keys and four IVs to do much magic, is specified as such with test vectors,
but the implementation actually always uses the first key, and the test vectors
were computed with it, so they are wrong.

Fixing the code to match the intended specification would break
interoperability with existing code (assuming such code would exist), so it is
better to remove this interface, which is obviously too complex for mere
mortals if even its author can not implement it correctly.

Riding on the libcrypto major bump.

10 years agosync
todd [Mon, 12 May 2014 19:18:23 +0000 (19:18 +0000)]
sync

10 years agoBring back ASN1_seq_pack(), ASN1_seq_unpack(), ASN1_pack_string() and
miod [Mon, 12 May 2014 19:16:35 +0000 (19:16 +0000)]
Bring back ASN1_seq_pack(), ASN1_seq_unpack(), ASN1_pack_string() and
ASN1_unpack_string() from the NO_ASN1_OLD purge; turns out that some Ruby
extensions still use them, unfortunately, so we need to give them some more
time to catch up on ``recent'' interfaces.

Riding on the libcrypto major bump.

10 years agoMove the `pqueue' part of libcrypto, which is a glorified sorted linked list
miod [Mon, 12 May 2014 19:14:14 +0000 (19:14 +0000)]
Move the `pqueue' part of libcrypto, which is a glorified sorted linked list
of 64-bit data, and only used by DTLS, to libssl where it belongs.

Remove pqueue_print() which is a debugging interface and serves no useful
purpose, except for the regress test, which grows its own pqueue_print()
routine.

Bump libcrypto major and libssl minor.

WARNING: do not update your tree right now, more changes are coming, which
will ride the libcrypto major bump.

10 years agono md5crypt
tedu [Mon, 12 May 2014 19:13:51 +0000 (19:13 +0000)]
no md5crypt

10 years agoremove md5crypt.
tedu [Mon, 12 May 2014 19:13:14 +0000 (19:13 +0000)]
remove md5crypt.
while changing things, add a crypt_checkpass wrapper that handles most of
the edge cases. (not quite ready for production, though.)
ok deraadt

10 years agoadjust to ohash being in libutil now, and to the interface changes.
espie [Mon, 12 May 2014 19:11:19 +0000 (19:11 +0000)]
adjust to ohash being in libutil now, and to the interface changes.
fix potential integer overflows in memory allocation (mostly for pedagogical
purposes, these are unlikely to overflow in practice)
move the rest of lst.lib stuff into its own directory.

10 years agomove the ohash functions into libutil by popular demand.
espie [Mon, 12 May 2014 19:09:00 +0000 (19:09 +0000)]
move the ohash functions into libutil by popular demand.
It's not a standard interface, so it doesn't belong in libc.
I hate duplicating the code in client programs, so do beck@, kettenis@,
schwarze@, millert@, miod@... and they agree with libutil.

10 years agotweak interface to do the right thing, after useful comments from millert@
espie [Mon, 12 May 2014 19:07:37 +0000 (19:07 +0000)]
tweak interface to do the right thing, after useful comments from millert@
(mostly use a calloc function to avoid potential integer overflow)

10 years agochange to having four freelists per size, to reduce another source of
tedu [Mon, 12 May 2014 19:02:20 +0000 (19:02 +0000)]
change to having four freelists per size, to reduce another source of
deterministic behavior. four selected because it's more than three, less
than five. i.e., no particular reason.

10 years agoDon't exit on encountering RTM_NEWADDR or RTM_DELADDR routing messages
krw [Mon, 12 May 2014 18:50:02 +0000 (18:50 +0000)]
Don't exit on encountering RTM_NEWADDR or RTM_DELADDR routing messages
that lack the appropriate address info.

Simply ignore the message as is done for short or incorrectly
versioned messages

ok claudio@

10 years agoKeep libusbhid's HID parser in sync with the kernel one. This is mostly
mpi [Mon, 12 May 2014 17:03:28 +0000 (17:03 +0000)]
Keep libusbhid's HID parser in sync with the kernel one.  This is mostly
a backport from FreeBSD's r205728, r224511 and 225839 without the new
functions to set/get a report ID nor the usbhidctl(1) & usbhidaction(1)
bits.

Committing now to open the "Monday's Bump Parade".

10 years agofix a possible double free when tls is required but not advertised by
eric [Mon, 12 May 2014 14:37:32 +0000 (14:37 +0000)]
fix a possible double free when tls is required but not advertised by
the server.

ok gilles@

10 years agoFix format strings in debug code.
kettenis [Mon, 12 May 2014 14:35:56 +0000 (14:35 +0000)]
Fix format strings in debug code.

10 years agoFix a leak from a recent added mark/marked keywords check.
andre [Mon, 12 May 2014 14:28:22 +0000 (14:28 +0000)]
Fix a leak from a recent added mark/marked keywords check.

ok reyk

10 years agoSince all static leases are in one file, /etc/dhclient.conf, there is no
krw [Mon, 12 May 2014 13:12:41 +0000 (13:12 +0000)]
Since all static leases are in one file, /etc/dhclient.conf, there is no
'wrong interface' specification. So display that error message only for
non-static leases. i.e. when reading /var/db/dhclient.leases.<if>.

10 years agoNo need for a dev_bsize variable that is always set to 512,
krw [Mon, 12 May 2014 12:16:53 +0000 (12:16 +0000)]
No need for a dev_bsize variable that is always set to 512,
a.k.a. DEV_BSIZE. So just use DEV_BSIZE.

Confirmed to still work on 4k-byte sector disks by David Vasek.

10 years agoBackport FreeBSD's HID parser in order to properly parse `moderm'
mpi [Mon, 12 May 2014 09:50:44 +0000 (09:50 +0000)]
Backport FreeBSD's HID parser in order to properly parse `moderm'
input device descriptors, notably if they include an Item with an
Usage array (Min-Max range) and others single Usage elements.

Tested by many, thanks!  ok andre@

10 years agoIncludes a router altert option (RAO) in IGMP packets. Without this
mpi [Mon, 12 May 2014 09:15:00 +0000 (09:15 +0000)]
Includes a router altert option (RAO) in IGMP packets.   Without this
option, required by the RFC2236, some L3 switches do not examine the
packets.

Based on FreeBSD's r14622 via Florian Riehm on tech@. ok bluhm@, jca@

10 years agoMake ifconfig do something intelligent based on the required length of
beck [Mon, 12 May 2014 08:47:37 +0000 (08:47 +0000)]
Make ifconfig do something intelligent based on the required length of
WEP keys rather then being silently dumb, so when using WEP:
1) If the key is a plausible size try to use it.
2) If they key would be a plausible size with '0x' in front of it, add that.
3) If the key is not a plausible size, emit a warning and do not try to use it.
ok sthen@

10 years agoreplace realloc(p, N * M) with reallocarray(p, N, M) and remove some
halex [Sun, 11 May 2014 21:25:07 +0000 (21:25 +0000)]
replace realloc(p, N * M) with reallocarray(p, N, M) and remove some
pointless cleanup if we're obviously going to die anyway

ok guenther@

10 years agostruct direct's d_ino is no longer the same size as an ino_t, so copy it
guenther [Sun, 11 May 2014 21:14:03 +0000 (21:14 +0000)]
struct direct's d_ino is no longer the same size as an ino_t, so copy it
for the call to bsearch()

problem report and ok krw@

10 years agoThe relation between uhci(4), ohci(4) and their root hub device is
mpi [Sun, 11 May 2014 16:33:21 +0000 (16:33 +0000)]
The relation between uhci(4), ohci(4) and their root hub device is
delicate.  HC drivers do not always accept to be left alone.

I don't know if it is a love crisis, but apparently receiving a root
hub status change interrupt before having an uhub(4) attached breaks
resume.

So make sure the root hub is re-attached before interrupt get enabled.
It is safe to do it during DVACT_RESUME since attaching root hubs do
not require any USB transfer.

Based on a diff from yasuoka@, fix a regression introduced in last
commit an reported by Abel Abraham Camarillo Ojeda on tech@ and Nils
R on bugs@.

10 years agoTweak DHCPREQUEST log message to omit 'port N', since the port is not
krw [Sun, 11 May 2014 14:00:52 +0000 (14:00 +0000)]
Tweak DHCPREQUEST log message to omit 'port N', since the port is not
set at that point, and is always set to 67 in send_packet().