Module name: src
Changes by: doug@cvs.openbsd.org 2014/08/17 15:01:24

Modified files:
regress/usr.sbin/relayd: args-http-headerlength.pl

Log message:
Sync regress/usr.sbin/relayd with recent http.h changes

Module name: src
Changes by: schwarze@cvs.openbsd.org 2014/08/17 15:02:06

Modified files:
usr.bin/mandoc : makewhatis.8

Log message:
typo; Steven Honeyman <stevenhoneyman at gmail dot com>

Module name: src
Changes by: schwarze@cvs.openbsd.org 2014/08/17 16:06:49

Modified files:
usr.bin/mandoc : term_ascii.c

Log message:
While all current callers pass valid data to ascii_hspan() only,
it's safer to assume incoming enum data might be invalid
and catch it instead of happily returning an unitialized int.
No functional change right now.

Module name: src
Changes by: schwarze@cvs.openbsd.org 2014/08/17 16:08:53

Modified files:
usr.bin/mandoc : term_ascii.c

Log message:
typo, sorry

Module name: src
Changes by: guenther@cvs.openbsd.org 2014/08/17 16:25:53

Modified files:
usr.bin/kdump : kdump.c

Log message:
Display symbolicly the mode argument of mkdir, mkfifo, mknod, and umask

Module name: src
Changes by: guenther@cvs.openbsd.org 2014/08/17 16:43:07

Modified files:
usr.bin/kdump : mksubr

Log message:
Use %#o instead of %#x for mode_t

Module name: src
Changes by: dlg@cvs.openbsd.org 2014/08/17 19:28:44

Modified files:
sys/kern : subr_pool.c

Log message:
external page headers use an RB tree to find the page header
containing an item when its returned to the pool. this means you
need to do an inexact comparison between an items address and the
page address, cos a pool page can contain many items.

previously this used RB_FIND with a compare function that would do math
on every node comparison to see if one node (the key) was within the other
node (the tree element).

this cuts it over to using RB_NFIND to find the closest tree node
instead of the exact tree node. the node compares turns into simple
< and > operations, which inline very nicely with the RB_NFIND. the
constraint (an item must be within a page) is then checked only
once after the NFIND call.

feedback from matthew@ and tedu@

Module name: src
Changes by: guenther@cvs.openbsd.org 2014/08/17 21:29:53

Modified files:
usr.bin/kdump : kdump.c

Log message:
Add fancy printing of ktrace()'s ops argument
mquery() has the exact same argument layout as mmap(), so share the case
Fix a couple brace placement glitches

Module name: src
Changes by: dlg@cvs.openbsd.org 2014/08/17 22:06:16

Modified files:
share/man/man9 : Makefile mbuf.9
sys/kern : uipc_mbuf.c
sys/sys : mbuf.h

Log message:
introduce mbuf list and queue apis. both manage fifo lists of mbufs
and a count of the mbufs.

struct mbuf_list and the ml_foo() apis can be used to build lists of
mbufs where you dont need locking (eg, on the stack).

struct mbuf_queue and mq_foo() wrap mbuf_lists with a mutex, and
limits the number of mbufs that can be queued. they can be useful
for moving mbufs between contexts/subsystems.

with help from jmc@ for the manpage bits
mpi@ is keen

Module name: src
Changes by: dlg@cvs.openbsd.org 2014/08/17 22:23:11

Modified files:
sys/dev/pcmcia : pcmcia_cis_quirks.c

Log message:
its a bit silly to include mbuf.h to get to malloc.h. just include
malloc.h instead.

Module name: src
Changes by: guenther@cvs.openbsd.org 2014/08/17 22:48:22

Modified files:
lib/libc/gen : exec.3

Log message:
Add mention of execvpe() setting errno on failure.

from Justus Baumgartner (meddev50 (at) gmail.com)

Module name: src
Changes by: dlg@cvs.openbsd.org 2014/08/17 23:11:03

Modified files:
sys/arch/amd64/amd64: aesni.c
sys/arch/hppa/hppa: machdep.c
sys/arch/hppa64/hppa64: machdep.c
sys/crypto : cryptodev.c
sys/dev/pci : if_myx.c
sys/kern : kern_sysctl.c uipc_mbuf2.c
sys/sys : mbuf.h

Log message:
dont rely on mbuf.h to provide pool.h.

ok miod@, who has offerred to help with any MD fallout
ok guenther@

Module name: src
Changes by: reyk@cvs.openbsd.org 2014/08/18 03:43:02

Modified files:
sbin/iked : iked.c iked.h proc.c

Log message:
Sync proc.c with httpd. httpd needs SIGUSR1 but iked will ignore it
now instead of terminating the process.

ok mikeb@

Module name: src
Changes by: reyk@cvs.openbsd.org 2014/08/18 06:59:00

Modified files:
usr.sbin/relayd: proc.c relayd.c relayd.h

Log message:
Sync proc.c with httpd. httpd needs SIGUSR1 but relayd will ignore it
now instead of terminating the process.

ok florian@

Module name: src
Changes by: reyk@cvs.openbsd.org 2014/08/18 07:13:42

Modified files:
usr.sbin/snmpd : proc.c snmpd.c snmpd.h

Log message:
Sync proc.c with httpd. httpd needs SIGUSR1 but snmpd will ignore it
now instead of terminating the process.

ok blambert@

Module name: src
Changes by: schwarze@cvs.openbsd.org 2014/08/18 07:25:54

Modified files:
usr.bin/mandoc : mandoc.c

Log message:
kristaps@ found this with valgrind, merge his patch from bsd.lv:
Fix a corner case where \H<nil> (where <nil> is the \0 character) would
cause mandoc_escape() to read past the end of an allocated string.
Found when a script scanning of all Mac OSX manuals accidentally also
scanned binary (gzip'd) files, discussed with schwarze@ on tech@mdocml.

Module name: src
Changes by: dlg@cvs.openbsd.org 2014/08/18 07:29:13

Modified files:
sys/arch/sparc64/dev: vnet.c

Log message:
this uses pools, but relied on mbuf.h to provide them.

found by benoit@

Module name: xenocara
Changes by: okan@cvs.openbsd.org 2014/08/18 07:57:57

Modified files:
app/cwm : calmwm.h group.c

Log message:
Get rid of nhidden in group_ctx; it actually never reported correctly
since nhidden wasn't incremented nor decremeted in all the right places,
thus confusing matters. We don't need to carry a count around, so just
use a local variable in the one place we need one to supply
XRestackWindows().

Module name: src
Changes by: tedu@cvs.openbsd.org 2014/08/18 08:34:58

Modified files:
lib/libc/stdlib: malloc.c

Log message:
a small tweak to improve malloc in multithreaded programs. we don't need
to hold the malloc lock across mmap syscalls in all cases. dropping it
allows another thread to access the existing chunk cache if necessary.
could be improved to be a bit more aggressive, but i've been testing this
simple diff for some time now with good results.

Module name: src
Changes by: tedu@cvs.openbsd.org 2014/08/18 08:35:37

Modified files:
sys/conf : GENERIC

Log message:
pool debug back on

Module name: src
Changes by: schwarze@cvs.openbsd.org 2014/08/18 10:26:13

Modified files:
usr.bin/mandoc : man_macro.c
regress/usr.bin/mandoc/man/RS: Makefile
regress/usr.bin/mandoc/man/SH: Makefile
Added files:
regress/usr.bin/mandoc/man/RS: broken.in broken.out_ascii
broken.out_lint
regress/usr.bin/mandoc/man/SH: empty_before.in
empty_before.out_ascii
empty_before.out_lint

Log message:
When the first child of the node being validated gets deleted during
validation, man_node_unlink() switches to MAN_NEXT_CHILD. After
that, we have to switch back to MAN_NEXT_SIBLING after completing
validation, or subsequent parsing would add content into an already
closed node, clobbering potentially existing children, causing
information loss and a memory leak. Bug found by kristaps@ with
valgrind in groff(7) on Mac OS X.

Note that the switch back must be conditional, for if the node being
validated itself gets deleted, we must *not* go to MAN_NEXT_SIBLING,
which would not only yield wrong results in general but also crash
in malformed manuals having an empty paragraph before the first .SH,
for example OpenBSD c++filt(1).

While here, add the missing <sys/types.h> as required before mandoc.h.

Module name: src
Changes by: miod@cvs.openbsd.org 2014/08/18 11:23:06

Modified files:
sys/arch/mips64/mips64: trap.c

Log message:
Sigh, ignoring instruction fetch bus errors for the kernel code should not
depend upon the address being at the beginning of a cache line, for we may
arrive in the middle of a line thanks to a branch. Noticed the hard way...

Module name: src
Changes by: miod@cvs.openbsd.org 2014/08/18 11:56:45

Modified files:
sys/arch/armv7/omap: if_cpsw.c
sys/arch/macppc/macppc: machdep.c

Log message:
Explicitely need <sys/pool.h> now.

Module name: src
Changes by: bcook@cvs.openbsd.org 2014/08/18 13:11:48

Modified files:
lib/libssl/src/crypto: md32_common.h
lib/libssl/src/crypto/md4: md4_dgst.c md4_locl.h
lib/libssl/src/crypto/md5: md5_locl.h
lib/libssl/src/crypto/ripemd: rmd_dgst.c rmd_locl.h
lib/libssl/src/crypto/sha: sha256.c sha_locl.h

Log message:
remove return value from HOST_c2l/l2c macros

These macros and asm inlines simulate a function returning a value, but
nothing ever uses this return value. Remove the pseudo-returns and
(void) casts discarding the unused values.

This, maybe unsurprisingly, speeds things up a bit. It also removes the
GCC 4.9 warnings about unused values.

ok miod@ deraadt@

Module name: src
Changes by: bcook@cvs.openbsd.org 2014/08/18 13:15:34

Modified files:
lib/libssl/src/crypto/des: des_locl.h
lib/libssl/src/crypto/rc5: rc5_locl.h

Log message:
replace more ROTATE macros with plain-old C code.

Let the compiler optimize these. Even older versions of gcc generate
equal or better quality code than the inline asm.

ok miod@

Module name: src
Changes by: deraadt@cvs.openbsd.org 2014/08/18 13:41:18

Modified files:
distrib/sets/lists/comp: mi

Log message:
sync

Module name: src
Changes by: tobias@cvs.openbsd.org 2014/08/18 14:58:56

Modified files:
usr.sbin/dhcpd : packet.c

Log message:
Variables to count seen packets should be unsigned.

ok krw@

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/18 15:43:45

Added files:
regress/usr.sbin/relayd: args-http-multi.pl
args-http-put-multi.pl
args-https-multi.pl
args-https-put-multi.pl

Log message:
Add test cases for multiple small HTTP requests. Many PUT requests
with ony byte content length triggered relayd to hang. This has
been fixed already.

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/18 15:49:17

Modified files:
regress/sys/kern/sosplice/udp: args-multi.pl

Log message:
Make the multiple UDP packet test more reliable. The receive buffer
limit also counts the mbufs in it, so packets could be dropped.
Increase both send and receive socket buffer size.

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/18 15:51:45

Modified files:
regress/sys/kern/sosplice: funcs.pl

Log message:
Make the forking tests more reliable. During copy the read(2)
system call might fail with EAGAIN when the other forked process
has already read the data.

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/18 15:53:40

Modified files:
regress/sys/kern/sosplice/tcp: args-oobinline-nonblock.pl
args-oobinline-reverse-nonblock.pl
args-oobinline-reverse.pl
args-oobinline.pl

Log message:
Make out of band data tests more reliable. As multiple tcp urgent
bytes cannot be handled correctly, allow another variation.

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/18 16:05:08

Modified files:
regress/usr.sbin/relayd: README
Added files:
regress/sys/kern/sosplice: README

Log message:
Add a description of what the relayd and sosplice regression tests
do. Also explain the optional environment variables.

Module name: src
Changes by: schwarze@cvs.openbsd.org 2014/08/18 16:21:52

Modified files:
usr.bin/mandoc : term.c

Log message:
Fix read access to uninitialized memory found by kristaps@ with valgrind,
patch from kristaps@.
Theoretically, this could lead to a buffer overrun and segfault,
but only for very long output lines (about 1000 charecters) of
exactly the right length or if by ill chance, the complete unused,
uninitialized tail of the line output buffer (at least 1000 characters
total length) would contain all blank characters. Anyway, the
uninitialized data read wasn't used for anything.

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/18 16:58:19

Modified files:
regress/usr.sbin/relayd: Proc.pm Relayd.pm Remote.pm direct.pl
funcs.pl relayd.pl remote.pl
regress/usr.sbin/ospfd: Client.pm Ospfd.pm Proc.pm ospfd.pl
regress/sys/kern/sosplice: LICENSE Proc.pm Remote.pm
regress/sys/kern/sosplice/error: error.pl remote.pl
regress/sys/kern/sosplice/tcp: direct.pl relay.pl remote.pl
regress/sys/kern/sosplice/udp: direct.pl relay.pl remote.pl
regress/sys/net/pf_divert: LICENSE Makefile Proc.pm Remote.pm

Log message:
Make the perl modules consistent for the multiple regression tests.
This includes coding style, better error messages and variable
naming.

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/18 18:24:00

Modified files:
usr.sbin/syslogd: syslogd.c

Log message:
Fix trailing white spaces.
OK sthen@

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/18 18:28:48

Modified files:
usr.sbin/syslogd: privsep.c

Log message:
Replace manually written function names with __func__.
OK sthen@

Module name: src
Changes by: bluhm@cvs.openbsd.org 2014/08/18 18:53:01

Modified files:
usr.sbin/syslogd: privsep.c

Log message:
The _exit codes in syslogd privsep.c were the wrong way around.
Fatal errors should exit with 1, and regular shutdown should result
in exit with 0.
OK deraadt@

Module name: src
Changes by: dlg@cvs.openbsd.org 2014/08/18 21:28:53

Modified files:
usr.bin/tcpbench: tcpbench.1 tcpbench.c

Log message:
add -4 and -6 flags for saying if you want to force ipv4 or ipv6
respectively.

tweaks from jmc@
ok djm@

Module name: src
Changes by: daniel@cvs.openbsd.org 2014/08/18 23:18:16

Modified files:
usr.bin/mandoc : roff.c

Log message:
libroff depends on libmandoc and so libmandoc should be included first.

this change makes the code slightly more portable by removing a gcc
extension (incomplete enum types) which is not part of ISO C.

ok schwarze@

Module name: src
Changes by: uebayasi@cvs.openbsd.org 2014/08/19 02:04:51

Modified files:
sys/net : if_pppx.c

Log message:
pppx(4): Correct pppx_dev_lookup() and pppx_dev2pxd() local declarations

Those two functions take one dev_t argument, not int. Match declarations
with reality. No functional changes.

Module name: www
Changes by: sthen@cvs.openbsd.org 2014/08/19 03:38:39

Modified files:
build : mirrors.dat

Log message:
update email address and add fingerprints for anoncvs.jp.openbsd.org

Module name: www
Changes by: sthen@cvs.openbsd.org 2014/08/19 03:38:55

Modified files:
. : anoncvs.html

Log message:
sync

Module name: src
Changes by: uebayasi@cvs.openbsd.org 2014/08/19 04:01:50

Modified files:
libexec/ld.so : boot.c

Log message:
ld.so(1): Missing stack var initialization

Explicitly zero-clear stack array to read dynamic segments. This code has
been working because stack is mapped as anon and zero'ed right after execve(2).

Found by work-in-progress Valgrind port for OpenBSD/amd64.

OK millert@ guenther@

Module name: src
Changes by: dlg@cvs.openbsd.org 2014/08/19 05:13:16

Modified files:
sys/dev/pci : if_myx.c

Log message:
in myx_start, replace

while (space) {
IFQ_POLL;
myx_dequeue(free descr);
IFQ_DEQUEUE;
etc;
}

with

while (space && myx_dequeue(free descr)) {
IFQ_DEQUEUE;
etc;
}

Module name: src
Changes by: mpi@cvs.openbsd.org 2014/08/19 06:24:04

Modified files:
sys/dev/pckbc : pms.c

Log message:
Make sure state changes are properly serialized.

When pms(4) is attached to a touchpad it generally presents two different
wsmouse(4) devices: one for the touchpad itself and one for the clitpad
and/or some interleaved packets. But since both devices are writing to
the same pckbc slot, a race can occur if they try to change the state at
the same time.

So prevent two process opening the two /dev/wsmouse* node at the same time
to corrupt the magic sequences needed to enable/disable the touchpad.

ok schadchin@

Module name: src
Changes by: mpi@cvs.openbsd.org 2014/08/19 06:28:03

Modified files:
sys/netinet6 : in6_proto.c ip6_mroute.c

Log message:
Keep pim code under #ifdef PIM to be coherent with what's in netinet/.

Module name: src
Changes by: mpi@cvs.openbsd.org 2014/08/19 06:39:36

Modified files:
sbin/route : route.c
usr.sbin/arp : arp.c
usr.sbin/ndp : ndp.c

Log message:
Entries for broadcast addresses should also be ignored, just like local
entries.

ok florian@, mikeb@, henning@

Module name: xenocara
Changes by: okan@cvs.openbsd.org 2014/08/19 06:47:51

Modified files:
app/cwm : calmwm.h group.c

Log message:
Pull highstack from group_ctx (and useless calculations of); in the one
place that we use highstack, replace that usage with a local variable
(for now until stacking is done properly).

Module name: src
Changes by: mpi@cvs.openbsd.org 2014/08/19 06:49:41

Modified files:
sys/net : route.c
sys/netinet : if_ether.c

Log message:
When a local route entry is added for an ifa having a broadcast address,
also adds a broadcast entry flagged with RTF_BROADCAST.

Prior to this change broadcast entries were simple clonned ARP entries,
that would be deleted once their timer expired since they would always
be incomplete.

With this change they are now persistant and identifiable with a new flag.

Committing early to be able to deal with any potential fallout before we
start relying on this.

ok florian@, mikeb@, henning@

Module name: src
Changes by: ajacoutot@cvs.openbsd.org 2014/08/19 08:06:38

src/usr.sbin/rcctl

Update of /cvs/src/usr.sbin/rcctl
In directory cvs.openbsd.org:/tmp/cvs-serv19263/rcctl

Log Message:
Directory /cvs/src/usr.sbin/rcctl added to the repository