openbsd
9 years agoUse proper casts for ctype functions. This is already fixed in nsd
millert [Mon, 12 Jan 2015 14:36:48 +0000 (14:36 +0000)]
Use proper casts for ctype functions.  This is already fixed in nsd
trunk.  OK doug@

9 years agofree->sshkey_free; ok djm@
markus [Mon, 12 Jan 2015 14:05:19 +0000 (14:05 +0000)]
free->sshkey_free; ok djm@

9 years agoKill the global list of IPv4 addresses.
mpi [Mon, 12 Jan 2015 13:51:45 +0000 (13:51 +0000)]
Kill the global list of IPv4 addresses.

ok claudio@, mikeb@, bluhm@

9 years agoMake it possible to reset "timeout" to the default value.
ajacoutot [Mon, 12 Jan 2015 13:51:20 +0000 (13:51 +0000)]
Make it possible to reset "timeout" to the default value.

9 years ago"timeout" must be a positive integer.
ajacoutot [Mon, 12 Jan 2015 13:40:12 +0000 (13:40 +0000)]
"timeout" must be a positive integer.

9 years agoallow WITH_OPENSSL w/o WITH_SSH1; ok djm@
markus [Mon, 12 Jan 2015 13:29:27 +0000 (13:29 +0000)]
allow WITH_OPENSSL w/o WITH_SSH1; ok djm@

9 years agoregen
jsg [Mon, 12 Jan 2015 12:41:30 +0000 (12:41 +0000)]
regen

9 years agoadd some E5 v3 PCIE and Bay Trail ids
jsg [Mon, 12 Jan 2015 12:40:41 +0000 (12:40 +0000)]
add some E5 v3 PCIE and Bay Trail ids

9 years agofatal if soft-PKCS11 library is missing rather (rather than continue
djm [Mon, 12 Jan 2015 11:46:32 +0000 (11:46 +0000)]
fatal if soft-PKCS11 library is missing rather (rather than continue
and fail with a more cryptic error)

9 years agoDon't forget about protocol specification when configuring flows.
mikeb [Mon, 12 Jan 2015 11:24:58 +0000 (11:24 +0000)]
Don't forget about protocol specification when configuring flows.

Tested by and OK claudio.

9 years agolet this test all supporte key types; pointed out/ok markus@
djm [Mon, 12 Jan 2015 11:12:38 +0000 (11:12 +0000)]
let this test all supporte key types; pointed out/ok markus@

9 years agoWhen setting up advanced TX descriptor use m_getptr to locate the IP
mikeb [Mon, 12 Jan 2015 10:40:51 +0000 (10:40 +0000)]
When setting up advanced TX descriptor use m_getptr to locate the IP
or IPv6 header instead of assuming contiguousness of the target buffer
across Ethernet and IP/IPv6 headers.

Tested by Kapetanakis Giannis <bilias at edu ! physics ! uoc ! gr>,
thanks!  Problem analysis and initial diff by dlg@.

9 years agoAdd a note that HOST_NAME_MAX does not include a terminating \0, unlike
nicm [Mon, 12 Jan 2015 10:11:07 +0000 (10:11 +0000)]
Add a note that HOST_NAME_MAX does not include a terminating \0, unlike
MAXHOSTNAMELEN.

ok deraadt guenther

9 years agopretty up some malloc() calls
deraadt [Mon, 12 Jan 2015 07:12:58 +0000 (07:12 +0000)]
pretty up some malloc() calls

9 years agohibernate_suspend() should not pmap_kremove by itself; hibernate_free()
deraadt [Mon, 12 Jan 2015 07:11:41 +0000 (07:11 +0000)]
hibernate_suspend() should not pmap_kremove by itself; hibernate_free()
must do that.  otherwise, pmap_kremove is called twice.  i386 in particular
does not tolerate that, found by sebastia
ok mlarkin kettenis

9 years agoSome fixes and tidying up of the receive filter handling bits.
brad [Mon, 12 Jan 2015 04:49:41 +0000 (04:49 +0000)]
Some fixes and tidying up of the receive filter handling bits.

9 years agorename blocks to words. bcrypt "blocks" are unrelated to blowfish blocks,
tedu [Mon, 12 Jan 2015 03:20:04 +0000 (03:20 +0000)]
rename blocks to words. bcrypt "blocks" are unrelated to blowfish blocks,
nor are they the same size.

9 years agoFix a memory leak in bss_dgram.
doug [Mon, 12 Jan 2015 02:15:23 +0000 (02:15 +0000)]
Fix a memory leak in bss_dgram.

Free data->saved_message.data.  Based on OpenSSL commit:
41cd41c4416f545a18ead37e09e437c75fa07c95 except this version sets a->ptr
to NULL to avoid accidental reuse and handles malloc failing.

ok beck@, input + ok miod@

9 years agostash the softc in the memory allocated for setting up a task for
dlg [Mon, 12 Jan 2015 00:07:55 +0000 (00:07 +0000)]
stash the softc in the memory allocated for setting up a task for
creating sensors.

9 years agoreplace fw_update (shell script) with a specialized version of pkg_add.
espie [Sun, 11 Jan 2015 22:34:08 +0000 (22:34 +0000)]
replace fw_update (shell script) with a specialized version of pkg_add.
This has numerous advantages, in terms of speed and control.
Initial version should be (mostly) identical to the old one.

thanks to rpe@ and sthen@ for tests.

9 years agoRemove orphaned defines.
miod [Sun, 11 Jan 2015 21:24:32 +0000 (21:24 +0000)]
Remove orphaned defines.

9 years agoNo more etc set, also sparc nowadays needs more than 16MB to boot multiuser.
miod [Sun, 11 Jan 2015 21:02:55 +0000 (21:02 +0000)]
No more etc set, also sparc nowadays needs more than 16MB to boot multiuser.

9 years agoTolerate checksum errors in ACPI tables (but complain about them). Apparently
kettenis [Sun, 11 Jan 2015 19:59:56 +0000 (19:59 +0000)]
Tolerate checksum errors in ACPI tables (but complain about them).  Apparently
there are machines out there with broken BIOSen with such checksum errors in
thier RSDT tables.  Windows runs fine on these machines, so it probably does
not check.  Linux ignores checksum errors as well (and prints a warning).

ok deraadt@, krw@

9 years agoLOCKDEBUG is dead; perform the funeral rites
guenther [Sun, 11 Jan 2015 19:34:52 +0000 (19:34 +0000)]
LOCKDEBUG is dead; perform the funeral rites

pointed out by Helg (xx404 (at) msn.com)
ok deraadt@ miod@

9 years agoswitch prototype warnings to implicit-declaration warnings.
tedu [Sun, 11 Jan 2015 19:25:13 +0000 (19:25 +0000)]
switch prototype warnings to implicit-declaration warnings.
This should catch all the same bad cases, but be a little less aggravating
in circumstances where a prototype isn't necessary
ok deraadt

9 years agorevert two latest commits until a crash is fixed
gilles [Sun, 11 Jan 2015 18:25:54 +0000 (18:25 +0000)]
revert two latest commits until a crash is fixed

9 years agoWhen using sun4 PROM I/O routines, make sure we never attempt to transfer more
miod [Sun, 11 Jan 2015 18:10:33 +0000 (18:10 +0000)]
When using sun4 PROM I/O routines, make sure we never attempt to transfer more
than the maximum DMA transfer size the PROM reports.

Crank version to 2.9.

9 years agoThere's no simple way so check for DMA buffer overrun with isochronous
mpi [Sun, 11 Jan 2015 15:41:16 +0000 (15:41 +0000)]
There's no simple way so check for DMA buffer overrun with isochronous
transfers, so skip the check for such xfers for the moment.

Issue reported by Ingo Feinerer and fix confirmed by Jan Stary.

9 years agoUse the softc-specific function pointers for bcopy() and bzero() in iestart(),
miod [Sun, 11 Jan 2015 15:35:38 +0000 (15:35 +0000)]
Use the softc-specific function pointers for bcopy() and bzero() in iestart(),
instead of, well, bcopy and bzero. This lets ie@vme transmit packets instead of
panicing due to memset() trying to use 32-bit stores on a device address.

9 years agoMake sure to initialize nbpg to a sensible value in bootxx, when running on
miod [Sun, 11 Jan 2015 15:29:03 +0000 (15:29 +0000)]
Make sure to initialize nbpg to a sensible value in bootxx, when running on
a sun4 system - the prom i/o routines depend on it.

9 years agoUnbreak sun4
miod [Sun, 11 Jan 2015 14:56:23 +0000 (14:56 +0000)]
Unbreak sun4

9 years agoglobal "list" is not the nicest variable name
deraadt [Sun, 11 Jan 2015 13:00:05 +0000 (13:00 +0000)]
global "list" is not the nicest variable name
seen my mlarkin, ok miod

9 years agoadd/fix/extend entries of exUSSR airports I've been to.
kirby [Sun, 11 Jan 2015 12:17:44 +0000 (12:17 +0000)]
add/fix/extend entries of exUSSR airports I've been to.
ok sthen@

9 years agounbreak the delivery_filename backend, now that it no longer prepends the
gilles [Sun, 11 Jan 2015 11:49:36 +0000 (11:49 +0000)]
unbreak the delivery_filename backend, now that it no longer prepends the
From separator line but gets it injected in its iobuf by the mda process,
we need to make sure not to escape the first 'From ' we receive.

spotted by deraadt@

9 years agoarmv6 introduced opcodes for reversing words in registers. we can
dlg [Sun, 11 Jan 2015 11:18:36 +0000 (11:18 +0000)]
armv6 introduced opcodes for reversing words in registers. we can
use these on armv7 as a backend for byteswapping things that endian.h
provide.

i dunno if its faster, but it makes smaller code. saves 30k on
GENERIC-OMAP.

ok jsing@ bmercer@ jsg@

9 years agocorrectly use HOST_NAME_MAX.
deraadt [Sun, 11 Jan 2015 04:14:40 +0000 (04:14 +0000)]
correctly use HOST_NAME_MAX.

Some notes:
POSIX HOST_NAME_MAX doesn't include the NUL.
POSIX LOGIN_NAME_MAX and TTY_NAME_MAX do include the NUL.

BSD MAXHOSTNAMELEN includes the NUL.  Actually, most of the historical
BSD MAX* defines did include the NUL, except for the historical
mistake of utmp fields without NULs in the string, which directly led
to strncpy..  just showing how error prone this kind of accounting is.
CSRG did right.  Somehow POSIX missed the memo on the concepts of
carefulness and consistancy, and we are still paying the price when
people trip over this.  Of course, glibc is even more amazing (that is
a hint to blackhats)

ok guenther

9 years agouse mallocarray(), then set the size afterwards. Prefer if people learn
deraadt [Sun, 11 Jan 2015 03:08:38 +0000 (03:08 +0000)]
use mallocarray(), then set the size afterwards.  Prefer if people learn
and use this idiom everywhere, it is always safest.

9 years agouse mallocarray()
deraadt [Sun, 11 Jan 2015 03:07:44 +0000 (03:07 +0000)]
use mallocarray()

9 years agomallocarray() for the rx_buffer memory
deraadt [Sun, 11 Jan 2015 03:06:19 +0000 (03:06 +0000)]
mallocarray() for the rx_buffer memory

9 years agoCasting a gid_t to gid_t for a gid_t argument is overkill
guenther [Sun, 11 Jan 2015 03:05:03 +0000 (03:05 +0000)]
Casting a gid_t to gid_t for a gid_t argument is overkill

9 years agouse mallocarray() when reallocating the mux table
deraadt [Sun, 11 Jan 2015 03:04:53 +0000 (03:04 +0000)]
use mallocarray() when reallocating the mux table

9 years agoKUE_MCFILTCNT() is non-constant, so use mallocarray()
deraadt [Sun, 11 Jan 2015 02:59:59 +0000 (02:59 +0000)]
KUE_MCFILTCNT() is non-constant, so use mallocarray()

9 years agounnecessary cast
deraadt [Sun, 11 Jan 2015 02:57:09 +0000 (02:57 +0000)]
unnecessary cast

9 years agosnc
deraadt [Sun, 11 Jan 2015 02:35:44 +0000 (02:35 +0000)]
snc

9 years agoGet the MAC address from the PCI ROM on systems without Open Firmware. Code
kettenis [Sat, 10 Jan 2015 22:14:30 +0000 (22:14 +0000)]
Get the MAC address from the PCI ROM on systems without Open Firmware.  Code
largely stolen from cas(4), but with a few tweaks.

9 years agoBye, bye simple_lock.
kettenis [Sat, 10 Jan 2015 19:52:37 +0000 (19:52 +0000)]
Bye, bye simple_lock.

9 years agoDisable (most) interrupts when the interface goes down.
kettenis [Sat, 10 Jan 2015 18:56:55 +0000 (18:56 +0000)]
Disable (most) interrupts when the interface goes down.

9 years agoEnable nep(4).
kettenis [Sat, 10 Jan 2015 17:19:02 +0000 (17:19 +0000)]
Enable nep(4).

9 years agoHook up nep(4).
kettenis [Sat, 10 Jan 2015 17:17:48 +0000 (17:17 +0000)]
Hook up nep(4).

9 years agoA few more loose ends.
kettenis [Sat, 10 Jan 2015 17:02:17 +0000 (17:02 +0000)]
A few more loose ends.

9 years agoRemove debug printfs.
kettenis [Sat, 10 Jan 2015 16:41:04 +0000 (16:41 +0000)]
Remove debug printfs.

9 years agonep(4)
kettenis [Sat, 10 Jan 2015 16:39:12 +0000 (16:39 +0000)]
nep(4)

9 years agoHook up nep(4).
kettenis [Sat, 10 Jan 2015 16:26:17 +0000 (16:26 +0000)]
Hook up nep(4).

9 years agoAdd support for promiscuous mode and multicase.
kettenis [Sat, 10 Jan 2015 16:25:33 +0000 (16:25 +0000)]
Add support for promiscuous mode and multicase.

9 years agotell the truth about DES.
sobrado [Sat, 10 Jan 2015 15:03:50 +0000 (15:03 +0000)]
tell the truth about DES.

joint work with djm@ and jsing@, who suggested stronger words warning people
away from single-DES.

ok djm@

9 years agoProperly stop DMA and tear down the rings when taking the interface down.
kettenis [Sat, 10 Jan 2015 14:55:29 +0000 (14:55 +0000)]
Properly stop DMA and tear down the rings when taking the interface down.

9 years ago1. They're flags, not counters. Set to one instead of incrementing.
tedu [Sat, 10 Jan 2015 13:48:02 +0000 (13:48 +0000)]
1. They're flags, not counters. Set to one instead of incrementing.
2. The G flag is useless and never checked. Remove it.
ok millert

9 years agodon't check for a return value that host() doesn't return, so future
tedu [Sat, 10 Jan 2015 13:47:05 +0000 (13:47 +0000)]
don't check for a return value that host() doesn't return, so future
generations don't try to change any of the values and break the code.
ok deraadt

9 years agoCorrect some comments and merge in6_if_up() into in6_ifattach() to
mpi [Sat, 10 Jan 2015 11:43:37 +0000 (11:43 +0000)]
Correct some comments and merge in6_if_up() into in6_ifattach() to
reflect that IPv6 link-local addresses are no longer automagically
configured the first time an interface is brought up.

ok henning@, stsp@

9 years agoRemove pointless empty 64bit support code stubs from bwi(4).
stsp [Sat, 10 Jan 2015 10:45:46 +0000 (10:45 +0000)]
Remove pointless empty 64bit support code stubs from bwi(4).
It seems no 64bit bus space support is forthcoming. Nothing
has happened in DragonflyBSD for several years and FreeBSD
has a different driver for 64bit chips called bwn(4).
ok mpi@ some time ago

9 years agoMerge two bwi(4) fixes from FreeBSD:
stsp [Sat, 10 Jan 2015 10:38:12 +0000 (10:38 +0000)]
Merge two bwi(4) fixes from FreeBSD:

------------------------------------------------------------------------
r192306 | imp | 2009-05-18 17:31:26 +0200 (Mon, 18 May 2009) | 10 lines

Fix a typo from the original driver.  We need to write ctrl2 into RF
register 0x52, not ctrl1.  This appears to be a mistake in the bcm
reverse engineering page, and has been corrected there.  Tracing
through the code, this is more in keeping with the "documented"
register. Sephe thinks it looks interesting and may be worth
fixing. :)

Submitted by: ddkprog at yahoo com
Reviewed by: Sepherosa Ziehau

------------------------------------------------------------------------
r192042 | nwhitehorn | 2009-05-13 16:25:55 +0200 (Wed, 13 May 2009) | 4 lines

Add a short delay after programming PHY registers to give some time for
the engine to catch up. This prevents a machine check exception from
illegal memory requests with a BCM4318.

------------------------------------------------------------------------

No regressions observed by myself, miod@, and Vesa Norrman (vesbula at gmail).

9 years agoCorrect reversed test of setmountent() return
guenther [Sat, 10 Jan 2015 07:56:16 +0000 (07:56 +0000)]
Correct reversed test of setmountent() return

Fix from Dennis Lindroos (denafcm (at) gmail.com)

9 years agorevert host() back to correct behavior.
bcook [Sat, 10 Jan 2015 01:56:52 +0000 (01:56 +0000)]
revert host() back to correct behavior.

unbreak config file address parsing

9 years agoreturn -1 on host() address parsing failure, not 1.
bcook [Fri, 9 Jan 2015 23:48:21 +0000 (23:48 +0000)]
return -1 on host() address parsing failure, not 1.

Match what parse.y expects it to return.

ok millert@

9 years agoreturn -1 on host() address parsing failure, not 1.
bcook [Fri, 9 Jan 2015 23:44:07 +0000 (23:44 +0000)]
return -1 on host() address parsing failure, not 1.

Match what parse.y expects it to return.

ok millert@

9 years agoabsurd optimization. reduce initial space by one to avoid leaving one byte
tedu [Fri, 9 Jan 2015 21:49:59 +0000 (21:49 +0000)]
absurd optimization. reduce initial space by one to avoid leaving one byte
hanging off on a page by itself.

9 years agoThe "SunMicro Virtual Eth Device" found on Oracle's more recent SPARC systems
kettenis [Fri, 9 Jan 2015 20:45:40 +0000 (20:45 +0000)]
The "SunMicro Virtual Eth Device" found on Oracle's more recent SPARC systems
claims to support both RNDIS and CDC Ethernet.  However, RNDIS doesn't seem
to work, at least not with our driver.  So blacklist it here such that cdce(4)
takes over.

ok miod@, armani@ (both a while back)

9 years agoregen
kettenis [Fri, 9 Jan 2015 20:42:40 +0000 (20:42 +0000)]
regen

9 years agoSunMicro Virtual Eth Device
kettenis [Fri, 9 Jan 2015 20:41:48 +0000 (20:41 +0000)]
SunMicro Virtual Eth Device

9 years agoChange rssi passed to ifconfig(8) to a signed value to fix printing signal
sthen [Fri, 9 Jan 2015 20:34:21 +0000 (20:34 +0000)]
Change rssi passed to ifconfig(8) to a signed value to fix printing signal
strengths on 802.11 interfaces.  ok stsp@

Power on wireless interfaces is usually scaled in dBm. rssi (received signal
strength) is expected to be a -ve dBm values (i.e. [much] <1mW). Some (though
not all!) drivers store this as negative values, so it needs to be passed to
ifconfig that way for printing, not cast to an unsigned value. Valid range is
something like -40 to -90dBm, so the range available with a signed char is
reasonable whether it's stored as a +ve or -ve number.

9 years agoProperly unwind from a failure in usbd_dma_contig_alloc(). Calling
kettenis [Fri, 9 Jan 2015 20:17:05 +0000 (20:17 +0000)]
Properly unwind from a failure in usbd_dma_contig_alloc().  Calling
bus_dmamap_unload(9) on a map that failed to load is a bad idea and causes
panics on some architectures (such as sparc64).

ok mpi@

9 years agoFix loading memory allocated with bus_dmamem_alloc(9). The old could would
kettenis [Fri, 9 Jan 2015 14:23:25 +0000 (14:23 +0000)]
Fix loading memory allocated with bus_dmamem_alloc(9).  The old could would
always load all allocated pages instead of the size specified in the
bus_dmamap_load_raw(9) call.  Also fixes the corner case where a specified
boundary is less than the page size, which would always create multiple
segments, even if the specified size was smaller than the boundary.

Fixes xhci(4) on sparc64.

9 years agoAlways check for DMA buffer overrun, not just for some specific xfers.
mpi [Fri, 9 Jan 2015 12:15:48 +0000 (12:15 +0000)]
Always check for DMA buffer overrun, not just for some specific xfers.

9 years agoCorrectly report memory allocation failure in uhidev_set_report_async().
mpi [Fri, 9 Jan 2015 12:09:51 +0000 (12:09 +0000)]
Correctly report memory allocation failure in uhidev_set_report_async().

9 years agoAdd an asynchronous version of uhidev_get_report(), from David Higgs
mpi [Fri, 9 Jan 2015 12:07:50 +0000 (12:07 +0000)]
Add an asynchronous version of uhidev_get_report(), from David Higgs
with some tweaks.

Nothing use it for the moment, but upcoming upd(4) improvements need
it.

9 years agodocument the chips we support. lots of raid controllers use them
dlg [Fri, 9 Jan 2015 11:25:05 +0000 (11:25 +0000)]
document the chips we support. lots of raid controllers use them
and im too lazy to find out what they are or list them all.

9 years agoDocument that ftp-proxy breaks explicit RFC4217 FTP TLS. While here,
stsp [Fri, 9 Jan 2015 11:19:12 +0000 (11:19 +0000)]
Document that ftp-proxy breaks explicit RFC4217 FTP TLS. While here,
chroot and privdrop is a feature, not a bug, so move it out of CAVEATS.
ok sthen

9 years agoAdd support DIOC{G,S}CACHE to mfi(4). Cache of disk on mfi(4) will
yasuoka [Fri, 9 Jan 2015 11:17:29 +0000 (11:17 +0000)]
Add support DIOC{G,S}CACHE to mfi(4).  Cache of disk on mfi(4) will
be enabled by default.  Also if the controller doesn't have any RAM,
the cache on the physical disk will be enabled.

discuss with jsg tsubai
ok dlg

9 years agoregen
jsg [Fri, 9 Jan 2015 09:35:20 +0000 (09:35 +0000)]
regen

9 years agoadd rate matching hubs seen on a C610 system
jsg [Fri, 9 Jan 2015 09:34:07 +0000 (09:34 +0000)]
add rate matching hubs seen on a C610 system

9 years agorename a variable to avoid a warning
gilles [Fri, 9 Jan 2015 08:28:02 +0000 (08:28 +0000)]
rename a variable to avoid a warning

spotted by deraadt@

9 years agomove the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
henning [Fri, 9 Jan 2015 08:09:39 +0000 (08:09 +0000)]
move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

9 years agoremove excessive/wrong use of sys/param.h
deraadt [Fri, 9 Jan 2015 07:35:37 +0000 (07:35 +0000)]
remove excessive/wrong use of sys/param.h
peanuts -- but all work has to start somewhere.

9 years agomatch on C610
jsg [Fri, 9 Jan 2015 07:29:45 +0000 (07:29 +0000)]
match on C610

9 years agoregen
jsg [Fri, 9 Jan 2015 07:28:14 +0000 (07:28 +0000)]
regen

9 years agoAdd some Intel Xeon E5 v3 and C610/X99 ids.
jsg [Fri, 9 Jan 2015 07:27:25 +0000 (07:27 +0000)]
Add some Intel Xeon E5 v3 and C610/X99 ids.

9 years agosave the bufq pointer from the buf before we turn it loose so it won't
tedu [Fri, 9 Jan 2015 05:04:22 +0000 (05:04 +0000)]
save the bufq pointer from the buf before we turn it loose so it won't
change on us. also, remove unused second arg to bufq_wait.
from pedro at bitrig via david hill.
ok beck kettenis

9 years agorename desiredvnodes to initialvnodes. less of a lie. ok beck deraadt
tedu [Fri, 9 Jan 2015 05:01:56 +0000 (05:01 +0000)]
rename desiredvnodes to initialvnodes. less of a lie. ok beck deraadt

9 years agocorrectly handle no timeouts and make timeout handling in general better.
tedu [Fri, 9 Jan 2015 04:59:54 +0000 (04:59 +0000)]
correctly handle no timeouts and make timeout handling in general better.
problem reported by Mages Simon
ok guenther

9 years agoCleanup some macros and #defines in i386 pmap. Previously committed and
mlarkin [Fri, 9 Jan 2015 03:43:52 +0000 (03:43 +0000)]
Cleanup some macros and #defines in i386 pmap. Previously committed and
backed out because of libkvm breakage, recommitting now with libkvm fix.

ok deraadt@

9 years agoimplement mfii_scsi_cmd_io for handling actual io. previously i got
dlg [Fri, 9 Jan 2015 03:34:40 +0000 (03:34 +0000)]
implement mfii_scsi_cmd_io for handling actual io. previously i got
away with being lazy and just passing everything to the vanilla
scsi cdb path.

sending io via the cdb path with chained sgls seems to trigger a
firmware fault on the new invader boards. sending the same io via
the ldio path works fine though.

tested on invader and thunderbolt boards:

mfii0 at pci1 dev 0 function 0 "Symbios Logic MegaRAID SAS3108" rev 0x02: msi
mfii0: "PERC H730 Mini", firmware 25.2.1.0037, 1024MB cache

mfii0 at pci1 dev 0 function 0 "Symbios Logic MegaRAID SAS2208" rev 0x05: msi
mfii0: "PERC H710 Mini", firmware 21.3.0-0009, 512MB cache

ok jmatthew@

9 years agoAdd test suites for SHA-224 and SHA-384.
lteo [Fri, 9 Jan 2015 03:32:58 +0000 (03:32 +0000)]
Add test suites for SHA-224 and SHA-384.

Relevant lines from testsuite.sha224 were verified against the test vectors in
RFC 3874, while testsuite.384 was verified against
https://www.cosic.esat.kuleuven.be/nessie/testvectors/hash/sha/Sha-2-384.unverified.test-vectors

ok millert@

9 years agoAvoid a crash (free of bogus pointer) when there is a syntax error
millert [Thu, 8 Jan 2015 22:10:08 +0000 (22:10 +0000)]
Avoid a crash (free of bogus pointer) when there is a syntax error
reading the config socket by clearing the entire blacklist struct
instead of just the tag string.  Also avoid holes in the blacklist
array on error since the code can't cope with them.  OK beck@

9 years agoUse the field f_file also for the forwarding file descriptor. Remove
bluhm [Thu, 8 Jan 2015 20:22:47 +0000 (20:22 +0000)]
Use the field f_file also for the forwarding file descriptor.  Remove
the special f_fd field.  Also remove an useless F_FORWUDP case.
test and OK jasper@

9 years agoincrease namecache size to follow maxvnodes. seems better than desiredvnodes.
tedu [Thu, 8 Jan 2015 18:07:35 +0000 (18:07 +0000)]
increase namecache size to follow maxvnodes. seems better than desiredvnodes.
(accuracy of variable names, aside)
ok kettenis

9 years agoClear autoconf6 flag after autoconf6 cleanup happened, not before.
florian [Thu, 8 Jan 2015 17:21:01 +0000 (17:21 +0000)]
Clear autoconf6 flag after autoconf6 cleanup happened, not before.
Otherwise clean up code will never run.
OK mpi@, benno@, henning@

9 years agoInclude the Routing Table ID in "struct rtentry" and reorder the fields
mpi [Thu, 8 Jan 2015 15:05:44 +0000 (15:05 +0000)]
Include the Routing Table ID in "struct rtentry" and reorder the fields
such that there's no size change on, at least, amd64.

This will allow us to simplify some APIs and deprecate the remaining
flavors of "struct route".

ok claudio@, phessler@

9 years agoFactorize various duplicated chunks of (old and horrible) code, checking
mpi [Thu, 8 Jan 2015 14:29:18 +0000 (14:29 +0000)]
Factorize various duplicated chunks of (old and horrible) code, checking
for the validity of a given outgoing route entry into a single function.

This change is inspired from FreeBSD r111767.  The function introduced
here, rt_checkgate(), should hopefully die in a near future.  Why should
it die?  Well, it is way too late to do such validity checks: if your
kernel can ends up in ether_output() with an invalid route, please do
not let it try to find a new one that might do the job.

Go read this function if you're wondering why you're getting messages
like:

"arpresolve: X.X.X.X: route without link local address"

Since this horrible logic has survived 20 years of copy & past and small
modifications for workarounds without a single clear commit message, let's
assume it is full of dragons and try to play safe.  This factorization is
not intended to change any behavior.

With much inputs from bluhm@, tested by weerd@ and florian@ on setups
with p2p IPv6 interfaces.

ok bluhm@, benno@, florian@

9 years agofix ssh_config FingerprintHash evaluation order; from Petr Lautrbach
djm [Thu, 8 Jan 2015 13:44:36 +0000 (13:44 +0000)]
fix ssh_config FingerprintHash evaluation order; from Petr Lautrbach

9 years ago"default" is gone.
ajacoutot [Thu, 8 Jan 2015 13:29:34 +0000 (13:29 +0000)]
"default" is gone.