deraadt [Tue, 18 Apr 2017 02:20:50 +0000 (02:20 +0000)]
use freezero() instead of explicit_bzero + free
deraadt [Tue, 18 Apr 2017 02:16:48 +0000 (02:16 +0000)]
put clang binaries into clang set
deraadt [Tue, 18 Apr 2017 02:15:50 +0000 (02:15 +0000)]
use freezero() instead of 4-line conditional explicit_bzero + free
deraadt [Tue, 18 Apr 2017 01:45:33 +0000 (01:45 +0000)]
free(NULL) is ok so use it; from Michael W. Bombardieri
yasuoka [Tue, 18 Apr 2017 01:24:47 +0000 (01:24 +0000)]
Fix pipex(4) not to cause panics on strict alignment architectures.
ok claudio
jmatthew [Tue, 18 Apr 2017 00:18:51 +0000 (00:18 +0000)]
Rearrange ure_iff to resemble other drivers a bit more.
from Brad
jmc [Mon, 17 Apr 2017 22:42:57 +0000 (22:42 +0000)]
/dev/rmt* is no longer relevant; from miod
deraadt [Mon, 17 Apr 2017 21:58:27 +0000 (21:58 +0000)]
some freezero() calls
deraadt [Mon, 17 Apr 2017 21:55:20 +0000 (21:55 +0000)]
use freezero()
deraadt [Mon, 17 Apr 2017 21:49:01 +0000 (21:49 +0000)]
memset() of password field should be explicit_bzero().
[Not using freezero() in yacc files yet]
deraadt [Mon, 17 Apr 2017 21:48:26 +0000 (21:48 +0000)]
freezero() can be used here
deraadt [Mon, 17 Apr 2017 21:45:34 +0000 (21:45 +0000)]
skip mtree creation of two include dirs. gcc3 may still use them, but
if so gcc3 should create them itself.
bluhm [Mon, 17 Apr 2017 21:10:03 +0000 (21:10 +0000)]
The raw ip input functions are called from several places. Use the
address family passed down with pr_input to check that the correct
one is used.
OK florian@
bluhm [Mon, 17 Apr 2017 20:59:35 +0000 (20:59 +0000)]
Use the address family passed down with pr_input to simplify
tcp_input().
OK florian@
bluhm [Mon, 17 Apr 2017 20:48:21 +0000 (20:48 +0000)]
Use the address family passed down with pr_input to simplify
udp_input().
OK florian@
deraadt [Mon, 17 Apr 2017 20:22:14 +0000 (20:22 +0000)]
matthieu@ observes % pax (without any arguments) hits pledge violation,
because it tries MTIOCTOP against stdin, the tty. It is very inconvenient
to use isatty to distinguish this difference in userland, so return ENOTTY
for tty devices.
ok natano
schwarze [Mon, 17 Apr 2017 20:04:26 +0000 (20:04 +0000)]
Restore -kO Nd, -kO Nm, -kO sec, and -kO arch to working order.
They got broken in the SQLite removal.
As opposed to the rest of -kO, they are no longer very useful,
but they are certainly not supposed to fail assertions.
Issue reported by Gonzalo Tornaria <tornaria at cmat dot edu dot uy>.
pascal [Mon, 17 Apr 2017 19:58:53 +0000 (19:58 +0000)]
Take a more radical approach to disabling colours: pretend we don't
HAVE_TERMINFO in our local config.h. Some of the lower levels of LLVM
don't seem to respect our default setting.
As a result, delete the -ltermlib dependency.
Note that this breaks -fcolor-diagnostics=auto. However, fixing the whole
codebase is out of scope for now ...
Prompted by an instance found in the AsmParser by deraadt@
ok kettenis@ patrick@
deraadt [Mon, 17 Apr 2017 19:44:59 +0000 (19:44 +0000)]
create a gross mechanism that can handle gcc-only, clang-only,
and gcc+clang architectures
with kettenis
krw [Mon, 17 Apr 2017 18:31:08 +0000 (18:31 +0000)]
Use better types for checksum() parameters and local var from
dhclient. Eliminate extra prototypes for checksum() and wrapsum().
otto [Mon, 17 Apr 2017 18:07:15 +0000 (18:07 +0000)]
consictently use .Dv NULL and a few other tweaks; ok schwarze@
otto [Mon, 17 Apr 2017 16:37:39 +0000 (16:37 +0000)]
don't manipulate hdr.len, it's used internally by libutil now; ok florian@
otto [Mon, 17 Apr 2017 16:36:35 +0000 (16:36 +0000)]
whitespace fixes
otto [Mon, 17 Apr 2017 16:03:15 +0000 (16:03 +0000)]
don't manipulate hdr.len, it's used internally by libutil now; ok florian@
kettenis [Mon, 17 Apr 2017 15:53:21 +0000 (15:53 +0000)]
Change build infrastructure to allow building both gcc and clang. This
doesn't actually flip the switch yet, so aarch64 continues to be the only
architecture for which we build clang.
ok jsg@, deraadt@
schwarze [Mon, 17 Apr 2017 14:31:23 +0000 (14:31 +0000)]
Recognize nl_langinfo(CODESET) return values "646" and "" as aliases
for "US-ASCII", useful for different versions of NetBSD and Solaris.
Found by dtucker@ and by Tom G. Christensen <tgc at jupiterrise dot com>.
OK dtucker@ deraadt@
bluhm [Mon, 17 Apr 2017 14:27:46 +0000 (14:27 +0000)]
Test syslogd(8) "last message repeated" feature and -r option.
bluhm [Mon, 17 Apr 2017 14:18:44 +0000 (14:18 +0000)]
Add syslogd(8) option -r to suppress the summary line for pipe and
remote loghost as they are most commonly used for automated log
processing. With -rr the "last message repeated" feature can be
disabled completely.
OK sthen@ deraadt@ jmc@
schwarze [Mon, 17 Apr 2017 13:29:20 +0000 (13:29 +0000)]
test display of long URIs
schwarze [Mon, 17 Apr 2017 13:26:30 +0000 (13:26 +0000)]
implement display of long .Lk URIs to match -Tascii
kettenis [Mon, 17 Apr 2017 12:54:48 +0000 (12:54 +0000)]
Skip objc tests with clang.
schwarze [Mon, 17 Apr 2017 12:52:00 +0000 (12:52 +0000)]
Fix handling of trailing punctuation in .Lk.
This macro is unusual in so far as trailing punction needs to remain
inside the scope because it must be inside, not after the display
of long URIs in terminal output mode.
Improves formatting of fw_update(1), help(1), less(1), sendbug(1),
acx(4), inet6(4), ipsec(4), oce(4), isakmpd.conf(5), afterboot(8),
release(8), traceroute(8).
jsg [Mon, 17 Apr 2017 11:02:31 +0000 (11:02 +0000)]
Change COMPILER_VERSION tests which limited additional warnings to gcc4
to instead skip them on gcc3 as clang can handle -Wpointer-sign and
-Wold-style-definition.
florian [Mon, 17 Apr 2017 10:29:01 +0000 (10:29 +0000)]
We need to update the ia6_updatetime when changing the vltime/pltime
vio ioctl(2). Otherwise ifconfig shows the wrong times since
vltime/pltime are calculated from this value on export.
The handling of router advertisements does the right thing.
OK bluhm
nicm [Mon, 17 Apr 2017 08:10:44 +0000 (08:10 +0000)]
Don't bother moving the cursor for empty lines.
nicm [Mon, 17 Apr 2017 06:40:32 +0000 (06:40 +0000)]
Remove a couple of redraw flags that no longer have any effect.
nicm [Sun, 16 Apr 2017 20:33:46 +0000 (20:33 +0000)]
Memory leak, from David CARLIER.
nicm [Sun, 16 Apr 2017 20:32:14 +0000 (20:32 +0000)]
Use EL1 to clear lines when redrawing the leftmost pane, rather than
spaces.
jmc [Sun, 16 Apr 2017 20:26:34 +0000 (20:26 +0000)]
from brad: document rgephy
patrick [Sun, 16 Apr 2017 19:28:25 +0000 (19:28 +0000)]
Disable colored diagnostics in LLVM's lld.
Requested by deraadt@
tb [Sun, 16 Apr 2017 18:04:02 +0000 (18:04 +0000)]
According to termcap(3), char PC, *BC, *UP need to be extern. Fixes an
ld(1) error found by mestre a while ago.
fix suggested and ok naddy
bluhm [Sun, 16 Apr 2017 17:16:21 +0000 (17:16 +0000)]
Always evaluate expression in BUG_ON() macro to avoid unused variable
warnings. Makes non diagnostic kernel compile.
OK kettenis@
kettenis [Sun, 16 Apr 2017 16:59:53 +0000 (16:59 +0000)]
Make sure these tests print "SKIPPED" if the necessary variables aren't set.
The current code doesn't work since the magic .BEGIN target runs before
the regress target that prints "SKIPPED" and the .BEGIN target fails when
the variables aren't set.
ok bluhm@
deraadt [Sun, 16 Apr 2017 16:17:23 +0000 (16:17 +0000)]
sync
kettenis [Sun, 16 Apr 2017 16:14:37 +0000 (16:14 +0000)]
Define EXT_IMPLICIT_NBIT like we do on sparc64 and mips64. Makes vaious
long double math stuff (including printf) actually work.
While there remove 'struct ieee_ldouble', which isn't defined on other
architectures.
ok deraadt@
kettenis [Sun, 16 Apr 2017 16:04:48 +0000 (16:04 +0000)]
Use INT_MAX instead of SIZE_MAX as the maximum file size we can handle.
Fixes this test on 64-bit architectures.
ok visa@
kettenis [Sun, 16 Apr 2017 16:01:05 +0000 (16:01 +0000)]
Define DEF_WEAK like we do for ld.so to fix building this with clang.
ok millert@, deraadt@
deraadt [Sun, 16 Apr 2017 15:11:01 +0000 (15:11 +0000)]
Move comments into a block and uses {} to unconfuse reading.
kettenis [Sun, 16 Apr 2017 14:40:47 +0000 (14:40 +0000)]
Use %zu to print a size_t.
visa [Sun, 16 Apr 2017 14:28:07 +0000 (14:28 +0000)]
Replace fetch_and_add() with atomic_inc_int_nv() from <sys/atomic.h>
to make the code more similar to sparc64's.
OK mpi@, guenther@, kettenis@
beck [Sun, 16 Apr 2017 14:25:42 +0000 (14:25 +0000)]
Flip previously warm pages in the buffer cache to memory above the DMA
region if uvm tells us it is available.
nits from deraadt@
ok krw@ guenther@ kettenis@
beck [Sun, 16 Apr 2017 14:24:03 +0000 (14:24 +0000)]
Statistics for high memory flips in the buffer cache
nits from deraadt@
ok krw@ guenther@ kettenis@
kettenis [Sun, 16 Apr 2017 14:20:49 +0000 (14:20 +0000)]
Remove some defines marked #ifdef notyet, which really is #ifdef notever.
Reduces the diffs with other architectures.
kettenis [Sun, 16 Apr 2017 13:35:14 +0000 (13:35 +0000)]
Always return nonzero from _longjmp too.
ok jsg@
kettenis [Sun, 16 Apr 2017 11:07:03 +0000 (11:07 +0000)]
Remove unused rnd_seed variable. Upstream made the same change 4 days ago.
Fixes compiling this test using clang on arm64.
jsg [Sun, 16 Apr 2017 10:19:54 +0000 (10:19 +0000)]
remove -g from CFLAGS at florian's request
jsg [Sun, 16 Apr 2017 10:16:35 +0000 (10:16 +0000)]
Different compilers and versions of compilers have different warnings.
Remove -Werror to give code a greater chance of building.
ok deraadt@ florian@
ajacoutot [Sun, 16 Apr 2017 08:50:49 +0000 (08:50 +0000)]
Remove /etc/ssl/acme/. We don't need it now that we have a default acme-conf(5)
that direclty uses /etc/ssl/{,private} by default. Adapt the httpd.conf example
accordingly.
ok florian@ benno@ millert@
otto [Sun, 16 Apr 2017 06:26:55 +0000 (06:26 +0000)]
backout previous, data_len is not always initialized
aoyama [Sun, 16 Apr 2017 00:52:54 +0000 (00:52 +0000)]
Remove -Werror to make dhcrelay6 on gcc3 architecture successfully.
ok reyk
kettenis [Sat, 15 Apr 2017 18:58:45 +0000 (18:58 +0000)]
Build sparc64 kernels with -ffreestanding.
guenther [Sat, 15 Apr 2017 18:55:27 +0000 (18:55 +0000)]
Document RB_TIMEBAD; delete RB_* that are obsolete/unimplemented
ok visa@ deraadt@
jmc [Sat, 15 Apr 2017 18:45:15 +0000 (18:45 +0000)]
more streamlining, with help from bluhm;
bluhm [Sat, 15 Apr 2017 13:56:43 +0000 (13:56 +0000)]
After forced unmount of a file system that has other mount points
in it, dangling mounts could remain. When unmounting check the
hierarcy and unmount recursively. Also prevent that a new mount
appears during the scan.
Joint work with natano@; testing and OK krw@
schwarze [Sat, 15 Apr 2017 13:12:08 +0000 (13:12 +0000)]
Build full mandoc.db(5) databases by default using makewhatis(8)
without -Q during the build and in weekly(8). According to tests
by many developers, makewhatis(8) takes a few minutes at most even
on slower hardware like octeon, loongson, ALIX, RPI3, Soekris,
cubox, softiron etc., and security(8) is often worse than makewhatis(8).
In case this causes excessive weekly(8) run times on even slower
(~50 MHz-class) CPUs, consider adding "MAKEWHATISARGS=-Q" to
/etc/weekly.local on machines that feel unhappy.
OK sthen@ kettenis@ millert@ deraadt@
bluhm [Sat, 15 Apr 2017 11:58:51 +0000 (11:58 +0000)]
Take implementation for getsocket() from arp(8). This brings
routing table filter and pledge(2) to ndp(8) modes -s and -d.
OK florian@
bluhm [Sat, 15 Apr 2017 11:50:24 +0000 (11:50 +0000)]
Name global variable for routing socket rtsock like in ndp(8).
OK florian@
bluhm [Sat, 15 Apr 2017 11:42:09 +0000 (11:42 +0000)]
Let ndp(8) use the process's current routing table by default.
from Pierre Emeriaud; OK jca@ claudio@
Sync code and man page with arp(8). OK florian@
kettenis [Sat, 15 Apr 2017 11:15:02 +0000 (11:15 +0000)]
The tlb flushes in pmap_set_{l1,l2,l3} use ranges that don't fully make sense.
But those tlb flushes shouldn't be necessary anyway, so simply remove them.
Simplify the tlb flushing code now that we no longer flush ranges, and
revive the branch that doesn't flush a specific ASID for the kernel pmap
since its mappings are global.
ok patrick@, visa@
kettenis [Sat, 15 Apr 2017 10:23:35 +0000 (10:23 +0000)]
Handle R_SPARC_WPLT30 relocations against a local symbol. Fixes building the
audio/openal port with clang.
ok guenther@
florian [Sat, 15 Apr 2017 09:16:05 +0000 (09:16 +0000)]
regen
florian [Sat, 15 Apr 2017 09:15:45 +0000 (09:15 +0000)]
update to 4.1.16rc1
tests & OK sthen
(if there are more changes coming for 4.1.16 release we will just
commit them on top)
guenther [Sat, 15 Apr 2017 04:54:04 +0000 (04:54 +0000)]
Fix format string in ACPI_MEMDEBUG block
From Anton Lindqvist (anton.lindqvist(at)gmail.com)
visa [Sat, 15 Apr 2017 04:38:27 +0000 (04:38 +0000)]
Sync bus_dmamap_load_raw() with amd64 for better constraint checking.
Needed by xhci(4).
visa [Sat, 15 Apr 2017 04:18:40 +0000 (04:18 +0000)]
No GMX on CN73xx.
espie [Fri, 14 Apr 2017 21:37:15 +0000 (21:37 +0000)]
mark newish phony targets as phony.
okay tb@
bluhm [Fri, 14 Apr 2017 20:50:35 +0000 (20:50 +0000)]
Remove unused define satoin6().
bluhm [Fri, 14 Apr 2017 20:46:31 +0000 (20:46 +0000)]
Pass down the address family through the pr_input calls. This
allows to simplify code used for both IPv4 and IPv6.
OK mikeb@ deraadt@
schwarze [Fri, 14 Apr 2017 19:50:54 +0000 (19:50 +0000)]
adjust .Lk tests after groff and mandoc .Lk improvements
jcs [Fri, 14 Apr 2017 19:39:59 +0000 (19:39 +0000)]
once cvs's flow-control mechanism has kicked in, stop reading from
its local reader until memory usage goes back down below its
watermark.
during a checkout of a tree with big files (like www) from a fast
reader (disk) and a slow writer (net), the amount of data being
buffered can back up enough that cvs tries to allocate too many
buffers and hits its resource limit, causing death.
cvs's flow-control mechanism properly detects this early on, but the
message sent to the reader process to stop sending data takes too
long to process.
take more aggressive action and just stop reading from the reader
until the writer has ejected enough data that it can start re-using
its already-allocated buffers instead of allocating new ones.
ok deraadt
schwarze [Fri, 14 Apr 2017 19:34:54 +0000 (19:34 +0000)]
Show long .Lk URIs in like an indented display, similar to groff.
Suggested by bentley@.
bluhm [Fri, 14 Apr 2017 19:03:50 +0000 (19:03 +0000)]
Add test for IP payload compression. It is sending small and big
ping packets as only the latter get actually compressed.
schwarze [Fri, 14 Apr 2017 18:24:15 +0000 (18:24 +0000)]
Do not make the colon after the .Lk link text italic.
I just pushed the same change to GNU troff.
schwarze [Fri, 14 Apr 2017 18:23:29 +0000 (18:23 +0000)]
Do not make the colon after the .Lk link text italic.
I just pushed the same change to GNU troff.
bluhm [Fri, 14 Apr 2017 18:14:33 +0000 (18:14 +0000)]
Add tests with the ipsec.conf SA bundle keyword.
bluhm [Fri, 14 Apr 2017 18:06:28 +0000 (18:06 +0000)]
Up to now ipsecctl(8) grouped SAs with identical src and dst to the
flow which the first SA matched by the flow type. This behaviour
was mostly undocumented and unexpected. Make SA bundles explicit
in ipsec.conf(5). Only group SAs that have the same src and dst
and also the same bundle identifier.
OK hshoexer@
jsing [Fri, 14 Apr 2017 15:32:41 +0000 (15:32 +0000)]
Use freezero(3) when cleaning up session tickets - not only does it require
less code, but there is also a potential performance gain since they can be
larger allocations.
jsing [Fri, 14 Apr 2017 15:26:53 +0000 (15:26 +0000)]
Use freezero(3) to clean up the X25519 keys - simpler, cleaner code.
jsing [Fri, 14 Apr 2017 15:20:55 +0000 (15:20 +0000)]
Use freezero(3) in the CBB clean up path, since this could hold sensitive
information (such as master keys).
jsing [Fri, 14 Apr 2017 15:20:16 +0000 (15:20 +0000)]
Switch i2d_SSL_SESSION() back to freezero(3) now that the size constraints
have been relaxed.
jsing [Fri, 14 Apr 2017 15:19:39 +0000 (15:19 +0000)]
Clean up server key exchange EC point handling. Encode the point directly
into the CBB memory, rather than mallocing and memcpying, which also makes
makes the code more consistent with the client. Add a missing check for the
first EC_POINT_point2oct() call.
ok beck@
bluhm [Fri, 14 Apr 2017 15:11:31 +0000 (15:11 +0000)]
Avoid some false positives with cppcheck. No binary change.
OK jsg@
deraadt [Fri, 14 Apr 2017 15:02:51 +0000 (15:02 +0000)]
whitespace
kettenis [Fri, 14 Apr 2017 07:22:02 +0000 (07:22 +0000)]
Define __dead and __pure even if __STRICT_ANSI__ for compilers that
support GCC's __attribute__ syntax. Fixes several warnings when compiling
code with clang with -std=c++11 or some other strict ISO standard.
ok millert@, guenther@
deraadt [Fri, 14 Apr 2017 01:23:02 +0000 (01:23 +0000)]
correct path; from Klemens Nanni
mlarkin [Fri, 14 Apr 2017 01:02:28 +0000 (01:02 +0000)]
SVM: calculate max ASID value and save for later use. This will be used in
an upcoming diff to handle ASID/VPID reuse/rollover.
mlarkin [Fri, 14 Apr 2017 00:53:28 +0000 (00:53 +0000)]
Update a man page example to reflect recent changes.
Spotted by Ilya Kaliman, thanks.
kettenis [Thu, 13 Apr 2017 23:29:02 +0000 (23:29 +0000)]
A little bit more trivial cleanup.
millert [Thu, 13 Apr 2017 21:49:11 +0000 (21:49 +0000)]
Xr sigprocmask(2) not the obsolete sigsetmask(3)
kettenis [Thu, 13 Apr 2017 20:48:29 +0000 (20:48 +0000)]
Use the non-interrupt-safe pool allocator for the vp pool to avoid runninng
out of kva in the kmem_map. Avoids a hang when spawning a lot of processes.