openbsd
10 years agopass correct sizes to free()
tedu [Sun, 13 Jul 2014 23:33:26 +0000 (23:33 +0000)]
pass correct sizes to free()

10 years agoadd missing whitespace between .Fa macro argument and trailing punctuation
schwarze [Sun, 13 Jul 2014 23:25:09 +0000 (23:25 +0000)]
add missing whitespace between .Fa macro argument and trailing punctuation

10 years agouse mallocarray()
deraadt [Sun, 13 Jul 2014 23:24:47 +0000 (23:24 +0000)]
use mallocarray()

10 years agopciide.c, pciide_rdc_reg.h(new): ported rdcide(4) from NetBSD.
sasano [Sun, 13 Jul 2014 23:19:51 +0000 (23:19 +0000)]
pciide.c, pciide_rdc_reg.h(new): ported rdcide(4) from NetBSD.
it supports RDC's R1012 IDE controller.

tested on 86duino EduCake (DM&P Vortex86EX SoC)

ok by deraadt@

10 years agoTabs, not spaces.
jsing [Sun, 13 Jul 2014 23:19:02 +0000 (23:19 +0000)]
Tabs, not spaces.

10 years agouse mallocarray
tedu [Sun, 13 Jul 2014 23:18:01 +0000 (23:18 +0000)]
use mallocarray

10 years agoRename the context allocation from ressl_new to ressl_client, which makes
jsing [Sun, 13 Jul 2014 23:17:29 +0000 (23:17 +0000)]
Rename the context allocation from ressl_new to ressl_client, which makes
it completely obvious what the context is for. Ensure client functions are
used on client contexts.

10 years agoone .Fn argument per function argument
schwarze [Sun, 13 Jul 2014 23:12:02 +0000 (23:12 +0000)]
one .Fn argument per function argument

10 years agoSome reallocarray() use; review Jean-Philippe Ouellet, patrick keshishian
deraadt [Sun, 13 Jul 2014 23:10:23 +0000 (23:10 +0000)]
Some reallocarray() use; review Jean-Philippe Ouellet, patrick keshishian
ok tedu

10 years agoSplit the context allocation out from the configuration. This will allow
jsing [Sun, 13 Jul 2014 23:06:18 +0000 (23:06 +0000)]
Split the context allocation out from the configuration. This will allow
us to properly report errors that occur during configuration processing.

Discussed with tedu@

10 years agoDo not fold multiple function arguments into the same .Fn argument:
schwarze [Sun, 13 Jul 2014 23:03:03 +0000 (23:03 +0000)]
Do not fold multiple function arguments into the same .Fn argument:
That may cause indexing and formatting issues.
Buggy mdoc(7) code mentioned by uebayasi@ to jmc@.

10 years agoboot(9): Cosmetic changes to improve diff'ability.
uebayasi [Sun, 13 Jul 2014 22:53:38 +0000 (22:53 +0000)]
boot(9): Cosmetic changes to improve diff'ability.

10 years agoMove the client code into a separate file.
jsing [Sun, 13 Jul 2014 22:42:01 +0000 (22:42 +0000)]
Move the client code into a separate file.

10 years agoRename various configuration handling functions.
jsing [Sun, 13 Jul 2014 22:31:42 +0000 (22:31 +0000)]
Rename various configuration handling functions.

Requested by and discussed with tedu@.

10 years agoComment out option GPT until the matching userland bits are in place.
miod [Sun, 13 Jul 2014 22:28:03 +0000 (22:28 +0000)]
Comment out option GPT until the matching userland bits are in place.

10 years agoUse a single ressl.h header file.
jsing [Sun, 13 Jul 2014 22:13:52 +0000 (22:13 +0000)]
Use a single ressl.h header file.

Discussed with beck@ and tedu@.

10 years agoCosmetic changes to reduce diffs.
uebayasi [Sun, 13 Jul 2014 22:13:06 +0000 (22:13 +0000)]
Cosmetic changes to reduce diffs.

10 years agoUpdate procflags list, add PS_SYSTEM, PS_EMBRYO, PS_ZOMBIE and
claudio [Sun, 13 Jul 2014 21:59:50 +0000 (21:59 +0000)]
Update procflags list, add PS_SYSTEM, PS_EMBRYO, PS_ZOMBIE and
PS_NOBROADCASTKILL. The resulting table is shifted so far right
that a few additional lines had to be wrapped. Not ideal but the
best we can do at the moment.

10 years agoThe correct place to call _bus_dmamap_sync() is after we copy data *to* the
kettenis [Sun, 13 Jul 2014 21:51:12 +0000 (21:51 +0000)]
The correct place to call _bus_dmamap_sync() is after we copy data *to* the
bounce buffer and before we copy data *from* the bounce buffer.  Currently
_bus_dmamap_sync() is a no-op, but keeping it #ifdef'ed out in the wrong
place makes no sense.

ok deraadt@, miod@

10 years agosync
deraadt [Sun, 13 Jul 2014 21:49:02 +0000 (21:49 +0000)]
sync

10 years agoUse lerrx instead of errx since the logging subsystem is already initialized.
claudio [Sun, 13 Jul 2014 21:46:25 +0000 (21:46 +0000)]
Use lerrx instead of errx since the logging subsystem is already initialized.
OK florian@

10 years agoFix sched_stop_secondary_cpus() to properly drain CPUs
matthew [Sun, 13 Jul 2014 21:44:58 +0000 (21:44 +0000)]
Fix sched_stop_secondary_cpus() to properly drain CPUs

TAILQ_FOREACH() isn't safe to use in sched_chooseproc() to iterate
over the run queues because within the loop body we remove the threads
from their run queues and reinsert them elsewhere.  As a result, we
end up only draining the first thread of each run queue rather than
all of them.

ok kettenis

10 years agoyes indeed, it returns void *. from Jean-Philippe Ouellet, i also had
deraadt [Sun, 13 Jul 2014 21:39:00 +0000 (21:39 +0000)]
yes indeed, it returns void *.  from Jean-Philippe Ouellet, i also had
this lurking in a tree

10 years agoExplicitly initialise slen - this was not previously done due to a missing
jsing [Sun, 13 Jul 2014 21:38:23 +0000 (21:38 +0000)]
Explicitly initialise slen - this was not previously done due to a missing
M_ASN1_D2I_begin macro.

10 years agoConvert error handling to SSLerr and ERR_asprintf_error_data.
jsing [Sun, 13 Jul 2014 21:35:27 +0000 (21:35 +0000)]
Convert error handling to SSLerr and ERR_asprintf_error_data.

10 years agoSome functions need to dup() before sending a socket by imsg and don't
yasuoka [Sun, 13 Jul 2014 21:34:35 +0000 (21:34 +0000)]
Some functions need to dup() before sending a socket by imsg and don't
need to close() after sending socket since imsg_compose() closes the
passing socket.

10 years agoOnly set machdep.allowaperture if 'vga1: aperture needed' is found
rpe [Sun, 13 Jul 2014 21:24:43 +0000 (21:24 +0000)]
Only set machdep.allowaperture if 'vga1: aperture needed' is found
in dmesg output. Use that information to decide whether or not to
ask the user if he intends to use X.

initial diff from and OK halex@
OK deraadt@

10 years agoinclude stdint.h for standard ints. from Jean-Philippe Ouellet
tedu [Sun, 13 Jul 2014 21:21:25 +0000 (21:21 +0000)]
include stdint.h for standard ints. from Jean-Philippe Ouellet

10 years agofix even more trailing tabs
jasper [Sun, 13 Jul 2014 21:18:44 +0000 (21:18 +0000)]
fix even more trailing tabs

10 years ago#$%#@$# CONSPEED
miod [Sun, 13 Jul 2014 21:18:26 +0000 (21:18 +0000)]
#$%#@$# CONSPEED

10 years agoDisplay zero page hit and miss counters in vmstat -s.
kettenis [Sun, 13 Jul 2014 21:13:51 +0000 (21:13 +0000)]
Display zero page hit and miss counters in vmstat -s.

ok deraadt@

10 years agomore hanging tabs
jasper [Sun, 13 Jul 2014 21:13:30 +0000 (21:13 +0000)]
more hanging tabs

10 years agougly hanging tabs; ok jasper
deraadt [Sun, 13 Jul 2014 21:11:40 +0000 (21:11 +0000)]
ugly hanging tabs; ok jasper

10 years agoShow an error if cmd_find_session can't find the current session, like
nicm [Sun, 13 Jul 2014 20:57:46 +0000 (20:57 +0000)]
Show an error if cmd_find_session can't find the current session, like
the other functions.

10 years agoNo libsa putchar wanted here actually
miod [Sun, 13 Jul 2014 20:53:35 +0000 (20:53 +0000)]
No libsa putchar wanted here actually

10 years agoAn EOF is a good reason to close a connection.
krw [Sun, 13 Jul 2014 20:51:08 +0000 (20:51 +0000)]
An EOF is a good reason to close a connection.

ok nicm@

10 years agoobvious conversion to mallocarray()
deraadt [Sun, 13 Jul 2014 20:49:42 +0000 (20:49 +0000)]
obvious conversion to mallocarray()

10 years agotweak previous;
jmc [Sun, 13 Jul 2014 20:42:33 +0000 (20:42 +0000)]
tweak previous;

10 years agomissing bracket in previous;
jmc [Sun, 13 Jul 2014 20:40:34 +0000 (20:40 +0000)]
missing bracket in previous;

10 years agoStill needs <sys/reboot.h>
miod [Sun, 13 Jul 2014 20:29:05 +0000 (20:29 +0000)]
Still needs <sys/reboot.h>

10 years agoIf a client is killed while suspended with ^Z so has gone through the
nicm [Sun, 13 Jul 2014 20:23:10 +0000 (20:23 +0000)]
If a client is killed while suspended with ^Z so has gone through the
MSG_EXITED dance, don't try to resume it since a) it's pointless and b)
the tty structures have been cleaned up and tmux will crash.

10 years agoremove all crypt choices other than bcrypt. ok afresh1 deraadt
tedu [Sun, 13 Jul 2014 20:09:38 +0000 (20:09 +0000)]
remove all crypt choices other than bcrypt. ok afresh1 deraadt

10 years agocalling this "encryption" makes me cringe. "slightly obfuscated" is better.
tedu [Sun, 13 Jul 2014 19:40:57 +0000 (19:40 +0000)]
calling this "encryption" makes me cringe. "slightly obfuscated" is better.

10 years agocompare snprintf return value with -1. this isn't really necessary because
tedu [Sun, 13 Jul 2014 18:59:40 +0000 (18:59 +0000)]
compare snprintf return value with -1. this isn't really necessary because
int promotion rules guarantee the correct result when compared with sizeof,
but it is perhaps easier for some people to understand it this way.
from Doug Hogan.

10 years agosync
deraadt [Sun, 13 Jul 2014 18:22:12 +0000 (18:22 +0000)]
sync

10 years agosync with kernel, root= -> rootdev=
jasper [Sun, 13 Jul 2014 18:08:16 +0000 (18:08 +0000)]
sync with kernel, root= -> rootdev=

10 years agoconsistency with upcoming bootloader device parsing of the uboot arguments,
jasper [Sun, 13 Jul 2014 18:07:38 +0000 (18:07 +0000)]
consistency with upcoming bootloader device parsing of the uboot arguments,
root= becomes rootdev=

10 years agoConvert d2i_SSL_SESSION to ASN1 primitives, instead of the horrific
jsing [Sun, 13 Jul 2014 17:56:56 +0000 (17:56 +0000)]
Convert d2i_SSL_SESSION to ASN1 primitives, instead of the horrific
asn1_mac.h macros. This still needs a lot of improvement, but immediately
becomes readable.

ok miod@ (sight unseen!)

10 years agoUse errx() after getpwnam() failure since errno may not be set.
claudio [Sun, 13 Jul 2014 17:53:41 +0000 (17:53 +0000)]
Use errx() after getpwnam() failure since errno may not be set.
All other privsep / privdrop daemons do this the same way.
OK florian@ some time ago

10 years agoDocument KERN_PROC_NOBROADKILL
claudio [Sun, 13 Jul 2014 17:47:03 +0000 (17:47 +0000)]
Document KERN_PROC_NOBROADKILL

10 years agoWhen reassembled IPv6 fragments are NATed or RDRed by pf, the
bluhm [Sun, 13 Jul 2014 17:41:04 +0000 (17:41 +0000)]
When reassembled IPv6 fragments are NATed or RDRed by pf, the
checksum has to be recalculated before the packet is fragmented
again.  Put a missing in6_proto_cksum_out() into pf_refragment6().
This makes run-regress-frag6 and run-regress-frag6-ext pass again.
From Matthias Pitzl; OK henning@

10 years agotweak previous;
jmc [Sun, 13 Jul 2014 17:39:57 +0000 (17:39 +0000)]
tweak previous;

10 years agodo not need malloc.h
deraadt [Sun, 13 Jul 2014 17:39:56 +0000 (17:39 +0000)]
do not need malloc.h

10 years agouse mallocarray where arguments are multiplied; ok tedu
deraadt [Sun, 13 Jul 2014 17:34:25 +0000 (17:34 +0000)]
use mallocarray where arguments are multiplied; ok tedu

10 years ago-N for SYNOPSIS and usage(), and some small changes to previous;
jmc [Sun, 13 Jul 2014 17:28:13 +0000 (17:28 +0000)]
-N for SYNOPSIS and usage(), and some small changes to previous;

10 years agobye bye src url. Never amounted to much, we went another road a while ago
espie [Sun, 13 Jul 2014 17:24:06 +0000 (17:24 +0000)]
bye bye src url. Never amounted to much, we went another road a while ago

10 years agoKeep track of port connections and port resets.
pirofti [Sun, 13 Jul 2014 17:19:17 +0000 (17:19 +0000)]
Keep track of port connections and port resets.

Also notify upstream when a port finished reseting and when the
connection status changes.

Gets things further along to the point where pipe device transfer
and control methods are required.

10 years agobundling base + packages was slow, as it needed to find lots of files.
espie [Sun, 13 Jul 2014 17:17:21 +0000 (17:17 +0000)]
bundling base + packages was slow, as it needed to find lots of files.
different approach: mark the base system in one go, then hand pick the
files from the pkglocatedb.

quite a few unaccounted files yet, this will get better

10 years agoSet KERN_PROC_NOBROADCASTKILL so that iscsid is not killed by init
claudio [Sun, 13 Jul 2014 17:07:00 +0000 (17:07 +0000)]
Set KERN_PROC_NOBROADCASTKILL so that iscsid is not killed by init
when terminating multi-user so that we have a chance to sync the
file systems mounted via iscsi.

10 years agosync
deraadt [Sun, 13 Jul 2014 17:02:48 +0000 (17:02 +0000)]
sync

10 years agokill fs2hXX/h2fsXX macros with letohXX/htoleXX
pelikan [Sun, 13 Jul 2014 16:59:35 +0000 (16:59 +0000)]
kill fs2hXX/h2fsXX macros with letohXX/htoleXX

The reason being that ext2 structures are little-endian but JBD2 journal
is big-endian.  Don't confuse readers by talking about "file system endian".

Some KNF while there.

ok guenther

10 years agoPut back a m_copyback() that got lost in rev 1.837 of pf.c. This
bluhm [Sun, 13 Jul 2014 16:58:43 +0000 (16:58 +0000)]
Put back a m_copyback() that got lost in rev 1.837 of pf.c.  This
fixes the rewrite of an IPv6 header of an ICMP6 packet in the payload
of an ICMP6 error packet.  Path MTU discovery with ping6 over pf
nat or rdr works again.
Found by src/regress/sys/net/pf_fragment make run-regress-fragping6
OK henning@

10 years agoGPTSIGNATURE must be LL, for operation on non-punk machines
deraadt [Sun, 13 Jul 2014 16:50:25 +0000 (16:50 +0000)]
GPTSIGNATURE must be LL, for operation on non-punk machines
ok punk

10 years agoAdd iscsid_flags to rc.conf so we do not try to start iscsid all the time.
claudio [Sun, 13 Jul 2014 16:49:47 +0000 (16:49 +0000)]
Add iscsid_flags to rc.conf so we do not try to start iscsid all the time.
Noticed by naddy@

10 years agokill fs2hXX/h2fsXX macros with letohXX/htoleXX
pelikan [Sun, 13 Jul 2014 16:44:20 +0000 (16:44 +0000)]
kill fs2hXX/h2fsXX macros with letohXX/htoleXX

The reason being that ext2 structures are little-endian but JBD2 journal
is big-endian.  Don't confuse readers by talking about "file system endian".

While here, nuke the copy of bswap.c and link it against the kernel's like
fsck_ext2fs has done, for better maintainability.

ok guenther

10 years agoRemove license introduced with the PSK code, which has since been removed.
jsing [Sun, 13 Jul 2014 16:43:37 +0000 (16:43 +0000)]
Remove license introduced with the PSK code, which has since been removed.

ok deraadt@

10 years agoWhen the three possible return values are -1, 0, and 1, != 1 is the same
krw [Sun, 13 Jul 2014 16:43:25 +0000 (16:43 +0000)]
When the three possible return values are -1, 0, and 1, != 1 is the same
as <= 0. And the latter is the normal idiom so use that.

ok claudio@ henning@

10 years agoIntroduce PS_NOBROADCASTKILL a process flag that excludes processes from
claudio [Sun, 13 Jul 2014 16:41:21 +0000 (16:41 +0000)]
Introduce PS_NOBROADCASTKILL a process flag that excludes processes from
receiving broadcast signals (kill -1). The flag can be set via a new
sysctl KERN_PROC_NOBROADCASTKILL. This will be used by iscsid to survive
the mass killing by init(8) when terminating multi-user operations.
With and OK guenther@

10 years agoAnother compression remnant.
jsing [Sun, 13 Jul 2014 16:39:19 +0000 (16:39 +0000)]
Another compression remnant.

10 years agoExpand the tlsext_sigalg macros. The end result is about the same number
jsing [Sun, 13 Jul 2014 16:33:01 +0000 (16:33 +0000)]
Expand the tlsext_sigalg macros. The end result is about the same number
of lines and much more readable.

ok miod@

10 years agoRewrite i2d_SSL_SESSION to use the ASN1 primitives, rather than using the
jsing [Sun, 13 Jul 2014 16:30:50 +0000 (16:30 +0000)]
Rewrite i2d_SSL_SESSION to use the ASN1 primitives, rather than using the
horrific macros from asn1_mac.h.

This is a classic example of using macros to obfuscate code, in an attempt
to reduce the line count. The end result is so ridiculously convoluted that
it is completely unreadable and it takes hours to deconstruct the macros
and figure out what is actually going on behind the scenes.

ok miod@

10 years agokill fs2hXX/h2fsXX macros with letohXX/htoleXX
pelikan [Sun, 13 Jul 2014 16:08:53 +0000 (16:08 +0000)]
kill fs2hXX/h2fsXX macros with letohXX/htoleXX

The reason being that ext2 structures are little-endian but JBD2 journal
is big-endian.  Don't confuse readers by talking about "file system endian".

ok guenther

10 years agoAdd a regress test for the ASN1 handling of SSL session tickets.
jsing [Sun, 13 Jul 2014 16:03:54 +0000 (16:03 +0000)]
Add a regress test for the ASN1 handling of SSL session tickets.

10 years agoThe bell tolls for BUF_strdup - Start the migration to using
beck [Sun, 13 Jul 2014 16:03:09 +0000 (16:03 +0000)]
The bell tolls for BUF_strdup - Start the migration to using
intrinsics. This is the easy ones, a few left to check one at
a time.
ok miod@ deraadt@

10 years agotedu setlist support, superseded by the new locate dbs.
espie [Sun, 13 Jul 2014 15:58:40 +0000 (15:58 +0000)]
tedu setlist support, superseded by the new locate dbs.

10 years agostuff no longer needed, now that base/x release install its own locate dbs
espie [Sun, 13 Jul 2014 15:54:28 +0000 (15:54 +0000)]
stuff no longer needed, now that base/x release install its own locate dbs

10 years agoNo need for <netinet/in_systm.h> here.
mpi [Sun, 13 Jul 2014 15:52:49 +0000 (15:52 +0000)]
No need for <netinet/in_systm.h> here.

10 years agobzero -> memset. for the speeds.
tedu [Sun, 13 Jul 2014 15:52:38 +0000 (15:52 +0000)]
bzero -> memset. for the speeds.

10 years agotrim some casts
tedu [Sun, 13 Jul 2014 15:48:41 +0000 (15:48 +0000)]
trim some casts

10 years agoRemove stale comment which used to explain why we had a special 31-bit freelist
miod [Sun, 13 Jul 2014 15:48:32 +0000 (15:48 +0000)]
Remove stale comment which used to explain why we had a special 31-bit freelist
for DMA, back when this was applicable.

10 years agoFix memory leak.
logan [Sun, 13 Jul 2014 15:47:51 +0000 (15:47 +0000)]
Fix memory leak.

OK from beck@ and miod@

10 years agoKERNEL_ASSERT_LOCKED(9): Assertion for kernel lock (Rev. 3)
uebayasi [Sun, 13 Jul 2014 15:46:21 +0000 (15:46 +0000)]
KERNEL_ASSERT_LOCKED(9): Assertion for kernel lock (Rev. 3)

This adds a new assertion macro, KERNEL_ASSERT_LOCKED(), to assert that
kernel_lock is held.  In the long process of removing kernel_lock, there will
be a lot (hundreds or thousands) of use of this; virtually almost all functions
in !MP-safe subsystems should have this assertion.  Thus this assertion should
have a short, good name.

Not only that "KERNEL_ASSERT_LOCKED" is consistent with other KERNEL_* and
SCHED_ASSERT_LOCKED() macros.

Input from dlg@ guenther@ kettenis@.

OK dlg@ guenther@

10 years agoOPENSSL_{malloc,free} -> {malloc,free}
miod [Sun, 13 Jul 2014 15:42:42 +0000 (15:42 +0000)]
OPENSSL_{malloc,free} -> {malloc,free}

10 years agoWarn about the use of BUF_strdup.
miod [Sun, 13 Jul 2014 15:41:33 +0000 (15:41 +0000)]
Warn about the use of BUF_strdup.

10 years agoRemove a debug message
reyk [Sun, 13 Jul 2014 15:39:01 +0000 (15:39 +0000)]
Remove a debug message

10 years agoSince the event(s) passed to a callback can be a mask of all events
krw [Sun, 13 Jul 2014 15:38:09 +0000 (15:38 +0000)]
Since the event(s) passed to a callback can be a mask of all events
of interest and EV_WRITE is or'ed into the interesting events, it
is more correct to check both events each time. Pointed out by
Claudio.

ok henning@ claudio@

10 years agoCompatibility hack for the old "manpath=OpenBSD<blank>" query parameter format;
schwarze [Sun, 13 Jul 2014 15:38:06 +0000 (15:38 +0000)]
Compatibility hack for the old "manpath=OpenBSD<blank>" query parameter format;
unfortunate, more than 400 links needing this are scattered all around
the www.openbsd.org website, and CVSweb needs this as well.

10 years agoFall back to smaller spacing on the mips64 machines.
pirofti [Sun, 13 Jul 2014 15:33:28 +0000 (15:33 +0000)]
Fall back to smaller spacing on the mips64 machines.

Please spare some change for the mips64 memory-challenged machines..
Some change, Sir?

Fixes at least the octeon platform. Found the hardway on my DSR500.

Found by Boss tedu@ and Boss deraadt@
Okay Boss miod@

10 years agoInitial support to read GPT partition tables in the kernel, if option GPT.
miod [Sun, 13 Jul 2014 15:32:28 +0000 (15:32 +0000)]
Initial support to read GPT partition tables in the kernel, if option GPT.
Contributed by Markus Mueller; code based upon Bitrig's GPT support, with
stricter GPT structures validation and support for alternate header places.

ok deraadt@ jsing@ krw@

10 years agoStop using old n_long and n_short types.
mpi [Sun, 13 Jul 2014 15:31:20 +0000 (15:31 +0000)]
Stop using old n_long and n_short types.

10 years agouse mallocarray where arguments are multipled. ok deraadt
tedu [Sun, 13 Jul 2014 15:29:04 +0000 (15:29 +0000)]
use mallocarray where arguments are multipled. ok deraadt

10 years agoSync file to be identical in relayd(8) and httpd(8).
reyk [Sun, 13 Jul 2014 15:11:23 +0000 (15:11 +0000)]
Sync file to be identical in relayd(8) and httpd(8).

10 years agoFinish writing the output before closing the connection
reyk [Sun, 13 Jul 2014 15:07:50 +0000 (15:07 +0000)]
Finish writing the output before closing the connection
(adopted from relayd).

10 years agofill in proper sizes in free(9)
pelikan [Sun, 13 Jul 2014 15:07:01 +0000 (15:07 +0000)]
fill in proper sizes in free(9)

10 years agoUpdate BUGS section.
miod [Sun, 13 Jul 2014 15:06:02 +0000 (15:06 +0000)]
Update BUGS section.

10 years agopass the size to free in some of the obvious cases
tedu [Sun, 13 Jul 2014 15:00:40 +0000 (15:00 +0000)]
pass the size to free in some of the obvious cases

10 years agoif the freedsize isn't zero, check that's reasonable. ok beck
tedu [Sun, 13 Jul 2014 14:59:28 +0000 (14:59 +0000)]
if the freedsize isn't zero, check that's reasonable. ok beck

10 years agoIf the only process in the pgrp is in the middle of exiting, it might
guenther [Sun, 13 Jul 2014 14:56:56 +0000 (14:56 +0000)]
If the only process in the pgrp is in the middle of exiting, it might
not have any threads left.  Treat that the same as an empty pgrp.

encountered by and ok deraadt@

10 years agoWhen renewing a lease, update client->active or you end up in a
krw [Sun, 13 Jul 2014 14:50:03 +0000 (14:50 +0000)]
When renewing a lease, update client->active or you end up in a
tight renew loop. Caused by recent reshuffling of code in
bind_lease().

10 years agoClose the connection after the response is completed (no Keepalive yet).
reyk [Sun, 13 Jul 2014 14:46:52 +0000 (14:46 +0000)]
Close the connection after the response is completed (no Keepalive yet).