As from today, numeric fields may begin with an optional
plus or minus sign, not only an optional minus sign.

(apb)

more (and less) locking fixes

(christos)

- nbuf_ensure_contig: rework to use m_ensure_contig(9), which will not free
  the mbuf chain on failure.  Fixes some corner cases.  Improve regression
  test and sprinkle some asserts.
- npf_reassembly: clear nbuf on IPv6 reassembly failure path (partial fix).
  The problem was found and fix provided by Anthony Mallet.

(rmind)

Only the first symbol (pthread_dbg) should return NOLIB, the rest indicate
internal inconsistencies, so let the code return a more vocal error.

(christos)

tsd_alloc is no more, use tsd_list, and disable the destructor walking code
since it needs work. Fixes gdb.

(christos)

Provide xc_send_ipi() routine in RUMP, which is required for high-priority
xcall(9) mechanism.  It is emulated using low-priority xcall(9).

(rmind)

Provide register_t compat on solarisa.
(would be nice to handle register_t compat internally instead, but the
interface was already exposed in a release)

(pooka)

Regen

(pgoyette)

Spell company name correctly

(pgoyette)

Remove unused bmcintr() function.

(tsutsui)

Spell the company name correctly: Intrepid Control Systems

(pgoyette)

Add line to get the ftdi driver to use the serial function of the FTDI 232H.

(reinoud)

No need to include <sys/systm.h> in standalone sources.

(tsutsui)

tickets 1835, 1841, 1842

(bouyer)

regen for ticket #1842

(bouyer)

Apply patch, requested by msaitoh in ticket #1842:
sys/dev/pci/pcidevs                    1.969 via patch
sys/dev/pci/viaide.c                    1.57

Add VT8237S Integrated SATA Controller support.
PR#47452.

(bouyer)

Pull up following revision(s) (requested by msaitoh in ticket #1841):
sys/arch/arm/arm32/cpu.c: revision 1.89
S/,/;/

(bouyer)

Pull up following revision(s) (requested by plunky in ticket #1835):
sys/dev/bluetooth/bthidev.c: revision 1.23 via patch
sys/dev/bluetooth/bthidev.c: revision 1.24 via patch
usr.sbin/btdevctl/btdevctl.8: revision 1.8 via patch
sys/dev/bluetooth/btmagic.c: revision 1.4 via patch
sys/dev/bluetooth/btmagic.c: revision 1.5 via patch
when no link-mode is specified, explicitly set a mode of (int)0
otherwise l2cap_setmode() will fail during connection setup
for bthidev.c, this fixes a problem with Microsoft Wedge Touch
mouse (which may not be able to authenticate)
for btmagic.c, include this fix in case somebody tries that
related to PR/47286
increase some diagnostics
add a comment pointing out that the link-mode may need to be
changed manually, for devices which cannot authenticate
for PR/47286

(bouyer)

Mark up occurence of printf() in text with .Fn

(uwe)

In newfmt() example one level of indentation is enough.

(uwe)

When parsing numbers, allow a leading '+'.

(apb)

Don't call internal cngetc() and cnputc() directly.
Use public getchar() and puthchar() instead.

(tsutsui)

Use hex value rather than decimal for register and device macro.

(tsutsui)

Add RCS Id.

(tsutsui)

Specify text address via Makefile rather than hardcoding in boot.ldscript.

(tsutsui)

FILESBUILD_<filename>=yes can replace both
CLEANFILES+=<filename> and realall: <filename>

(apb)

add a dependency to realall from Takeshi Nakayama

(christos)

avoid double assignment

(christos)

fix bogus argument to kmem_free()

(christos)

if the firmware gives us a bad emmc clock rate, ignore it

(jmcneill)

xbd has been able to cope with any value of MAXPHYS since
src/sys/arch/xen/xen/xbd_xenbus.c 1.56

(jakllsch)

PR kern/47419: Antony Mallet: ifconfig doesn't diplay MTU on gif(4)

There is no special treatment for SIOCGITMTU in gif(4), so just pass it to
ifioctl_common().

(degroote)

Fix typo.

(mbalmer)

Document that errno may be used to distinguish between a
non-error result of -1 and an error.

(apb)

Fix typo.

(uwe)

Distinguish between a non-error result of (time_t)-1 and an error result.
Modernise the test code inside #ifdef TEST.

(apb)

Test parsedate("@-2",...) (should return -2 and not set errno);
and parsedate("@junk",...) (should return -1 and set errno).
We were already testing "@-1".

(apb)

Pay attention to the return value from cv_wait_sig as well in
usbd_transfer.  hi christos.

(skrll)

Improve some DIAGNOSTIC stuff

(skrll)

Rename the GRXSTSP registers and don't print the "pop" version.

(skrll)

Compile fix

(skrll)

Whitespace

(skrll)

oops, disable the counters by default

(skrll)

Don't count bit0 events in gintsts. bit 0 tells us which mode we're in
0 = device, 1 = host

(skrll)

* Test that parsedate("@0", ...) returns (time_t)0 regardless of timezone.
* Test that parsedate("@-1", NULL, NULL) returns (time_t)-1
  without setting errno.

(apb)

Document the fact that ctime_rz, localtime_rz, and mktime_z,
accept a NULL timezone_t pointer as a reference to UTC,
and the fact that tzalloc accepts a NULL zone name.

(apb)

Avoid warning from cngetc byt doing the proper cnpollc dance

(skrll)

Add -d as a synonym for getting into DDB to match other ports.

(skrll)

mktime(3) works in local time, and trying to trick it by
setting tm_gmtoff doesn't work.  Instead, call mktime_z(3)
with a null timezone (so it works in UTC) and adjust the
result afterwards.  Now "date -d @0" correctly
prints the local equivalent of 1970-01-01 00:00:00 UTC.

(apb)

Use PR_NOWAIT in dwc_otg_allocx as it can be called in the USB softint via
completions.

(skrll)

meta_oodate: rename ignoreOODATE to needOODATE.
if target uses ${.OODATE} in its script;
1. we canno compare commands
2. if we decide it is out-of-date .OODATE must not be empty.
  The only option is to set it to .ALLSRC

(sjg)

Sort options in SYNOPSIS and text.
XXX: please check if SYNOPSIS are correct, I think they missing some .Op.

(wiz)

Add !defined(KMEMUSER)

(matt)

- physmap_map, physmap_map_fini: pmap_update() must be performed before
  freeing the VA; otherwise there is a window when it can be re-used while
  stale TLB entries may be present.
- physmap_fill: use MIN() instead of min(), since vsize_t is used.
- Add RCS ID comment while here and prevent physmap.h inclusion in userland.

(rmind)

Add m_ensure_contig() routine, which is equivalent to m_pullup, but does not
destroy the mbuf chain on failure (it is kept valid).

(rmind)

Remove a KASSERT that is no longer valid (no restricted to only coherent
memory).
Fix c&p error for a !RCVMAGIC bus_dmamap_sync PREREAD.

(matt)

Fix __HAVE_MM_MD_CACHE_ALIASING case to use UVM_KMF_COLORMATCH so that
uvm returns a page of the correct color.

(matt)

Switch from only UVM_PGA_STRAT_ONLY to UVM_PGA_STRAT_FALLBACK in
arm_pmap_alloc_poolpage.

(matt)

Fix typo, unless `ro raising' is something I don't know about.

(riastradh)

Simple version of strlcat for ARM.

(matt)

- Don't dump core or print random junk on corrupt utmp entries.
- Factor out duplicated code in the process.
- The actual code is now smaller and does error checking, and encoding.

(christos)

add ctf man pages

(christos)

add man pages

(christos)

Now that ctf is fixed for at least amd64, go back to an equivalent to the
original functionality, use the tool only if it exists.

(christos)

Handle ddb symbol table loaded by the native bootloader properly.
- add a function that check if the symbol table is loaded and
  return the table size (taken from sun3/sun3/locore2.c)
- calculate end of symbol table address in locore.s
- call ksym_addsyms_elf() with proper args

(tsutsui)

pc files for openssl

(christos)

Add pkgconfig gluons

(christos)

Now that GPT_ENT_TYPE_LINUX_DATA is diffrent from the Windows one, use it.

(jakllsch)

#!/bin/sh

(christos)

gdisk 0.7.2 finally introduced a Linux Basic Data partition GUID.

(jakllsch)

Handle enum forward declarations.

(christos)

Forgot to commit additional file.

(christos)

Fix KASSERT() in biodone().  Clear BO_DONE in writedisklabel().

(kiyohara)

if doing #ifdef INET, you need opt_inet.h!

(jmcneill)

Print the frame pointer for each level and don't print the registers
within the frame by default. Provide a '/f' flag to get the full trace.

Discussed with matt@

(skrll)

Contain support routines used to create and destroy lists of physical pages
from lists of pages or ranges of virtual address.  By using these physical
maps, the kernel can avoid mapping physical I/O in the kernel's address space
in most cases.

(matt)

Use structure initializers.

(matt)

PR#47463:
- Check each mbrp_type field in fdisk partition table in MBR to know whether
  a sector is really MBR or not. Same as MI subr_disk_mbr.c
- Check whether each number in a partition table is out of bound or not.
  Same as MI subr_disk_mbr.c
- Update d_packname and d_npartitions if we found any partitions.
- Currently, the exended partitions is not supported...

(msaitoh)

make this work with lockdebug. while here, set urtwn_debug=0 now that
USB_DEBUG implies URTWN_DEBUG

(jmcneill)

Helps ELF platforms if I spell __ELF__ correctly ...

(pooka)

finish bump of libdwarf

(christos)

add new file

(christos)

remove RCSID, other files don't have it.

(christos)

toolify

(christos)

fix cut-n-paste

(christos)

document libdwarf.

(christos)

fix the extern "C" stuff

(pooka)

bump libdwarf

(christos)

resolve conflicts, bump

(christos)

sync with FreeBSD

Status:

Vendor Tag: FreeBSD
Release Tags: FreeBSD-2012-01-17

(christos)

bump version for recent portability fixes

(pooka)

Include sys/sysmacros.h on solaris for MIN/MAX.

(pooka)

the set files themselves should not be conditional to anything so that
they are always generated.

(christos)

regen

(pooka)

Include rumpuser_port.h in rump_syscalls.c when compiled for librumpclient

(pooka)

For non-NetBSD, assume leading underscore with !__ELF__ (fixes PE-COFF)

(pooka)

Solaris 10 fixes

(pooka)

Apply GCC/35998 from FreeBSD http://svnweb.freebsd.org/changeset/base/245024

gcc: avoid generating negative values to DW_AT_byte_size.

There is a bug in gcc (GCC/35998) where dwarf reports
sizes of unsigned -1 (0xffffffff).

On NetBSD this generated a faulty CTF entry which then
caused a segfault in ctfmerge. The issue was worked
around in NetBSD's Dtrace but since the issue originated
in gcc, it seems reasonable to fix it here.

Thanks to Christoph Mallon for pointing out a correct fix.

(christos)

Apply GCC/35998 from FreeBSD http://svnweb.freebsd.org/changeset/base/245024

gcc: avoid generating negative values to DW_AT_byte_size.

There is a bug in gcc (GCC/35998) where dwarf reports
sizes of unsigned -1 (0xffffffff).

On NetBSD this generated a faulty CTF entry which then
caused a segfault in ctfmerge. The issue was worked
around in NetBSD's Dtrace but since the issue originated
in gcc, it seems reasonable to fix it here.

Thanks to Christoph Mallon for pointing out a correct fix.

(christos)

ut_line does not have /dev in front of it. Add it, from dholland@

(christos)

move dk_ioctl to a header file for the benefit of x-building.

(christos)

Disable CTF Conversion for now until it is fixed.
- ERROR: ctrl_iface_ap.c: die 2767: failed to get unsigned: No entry found
  [dwarf _attrval_unsigned(219)]

(christos)

An uninitialized next-free-cluster value in the file system information
block is valid; do not consider it for correction.

(jakllsch)

If the host doesn't support RTLD_NEXT, don't play the dynamic linking game.

(pooka)

add the set files themselves

(christos)

Now that we have separate debug sets, add -g if ${MKDEBUG} is set for
program builds since we don't need to install that symbol files.

(christos)

Remove bogus conditionals that were also copied to other files presumably
because they did not work from here, and explain why.

(christos)

Add a separate bool to indicate a symbol table has been loaded.
ksym_initted indicates whether the kmutex has been initted or not.
Add __cacheline_aligned to the kmutex.

(matt)

Make loadfile() accept OMAGIC a.out binary.
LUNA's ROM monitor assumes it.

(tsutsui)

Make sure a USB small memory fragment is a multiple of the cache line size.
(since CACHE_LINE_SIZE defaults to 64, for most instances this won't result
in any code changing).

(matt)

Pullup from HEAD:
Cortex needs ._dmamap_sync_post for BUS_DMASYNC_POSTREAD

(matt)

Cortex needs ._dmamap_sync_post for BUS_DMASYNC_POSTREAD

(matt)

don't draw blanks by hand, use the blitter instead

(macallan)

- Add support for xsets
- Factor out common evbarm code

(christos)

calc_timer_factors: use supplied ints_per_sec, not hz, when calculating
counter frequency.

(jmcneill)

Pullup from HEAD:
On Cortex, speculative loads can cache lines to be populated after then they've
been invalidated for a DMA read.  So after the DMA read we have to reinvalidate
them again.  We have to both invalidates since the former prevents dirty lines
overwriting just DMAed data.

(matt)

On Cortex, speculative loads can cache lines to be populated after then they've
been invalidated for a DMA read.  So after the DMA read we have to reinvalidate
them again.  We have to both invalidates since the former prevents dirty lines
overwriting just DMAed data.

(matt)

Add some comments to illustrate what is actually happening.

(matt)

- add xdebug set
- document debug better
- fix previous debug fallout.

(christos)

Do the protocol consistency check hack only when compiling ELF.

(pooka)

copy data in 32bit chunks whenever alignment permits
-> very noticeable speedup

(macallan)

fix some bit #define, now the name actually makes some sense

(macallan)

fix misplaced paren

(christos)

Add a new "debug" set that gets built when ${MKDEBUG} is set. On evbarm:
-rw-r--r--  1 root  src  35806068 Jan 16 10:47 base.tgz
-rw-r--r--  1 root  src  57374069 Jan 16 10:48 comp.tgz
-rw-r--r--  1 root  src  101286193 Jan 16 10:49 debug.tgz
...
This allows always building the debugging code and libraries but not
installing it (although now it is placed in the default sets if ${MKDEBUG}
is set

(christos)

tidy up some more by factoring out the sizing code.

(christos)

Add UFS2 support.  Also bump version.

(tsutsui)

don't wait for completion if we were interrupted.

(christos)

From:  Steffen "Daode" Nurpmeso:

Amend the previous commit to support zone-style RFC 822 dates.
RFC 5322 marks this zone-style as obsolete (obs-zone), but still supports it.
But then, even if you don't wanna support military style single-letter zones,
the UT zone exists and will not be supported by this commit.

The changes here are not strictly correct (since not all possible
single-letter zones are assigned, and only *UT* has two letters),
but which will match all thre possible *obs-zone* forms.
Compiled and tested on a single MBOX.

(christos)

Enable and add ls command for ufs.

(tsutsui)

Make sure there is enough space for sockaddr subtypes
Use u_char for IP_MULTICAST_LOOP as specified in in.h

(kefren)

Set resource limit. The rnd_process_events() function is called every tick
and process the sample queue. Without limitation, if a lot of rnd_add_*()
are called, all kernel memory may be eaten up.

(msaitoh)

Sort sections.

(wiz)

sync with (a bit old) head

(yamt)

MPU_* registers are not valid for OMAP3, so restrict access to them to
OMAP2 boards.

(jmcneill)

Perform error message reporting differently, using an error buffer in
the diff_t structure.

Don't violate abstractions by accessing the status field in the
structure directly, use an accessor function to get the exit status
from the struct.

(agc)

implement right to left and bottom to top blits, now scrolling works right
in all directions.
while there, fix the xor mask for the cursor attribute - now it's actually
inverted

(macallan)

Don't do the xfer once, wait till it is done. I did not find this, someone
else did. Fixes slow devices such as scanners.

(christos)

- simplify getinput.
- add adjust command.

(christos)

add some things to add next time we modify the function

(christos)

PR/47449: David Holland: Don't log to terminals with mesg n, unless it is a
wall message.

(christos)

Make autosizing really work:
    - Use newfs to compute the actual filesystem required size.
    - Fix the computation of set sizes to account for blocks and fragments.
This results into a 95% full filesystem for my test (Total 566MB, 24MB free)
from which 8MB is my requested overhead. Not perfect, but good enough.

(christos)

now that we compute sizes correctly, reduce the extra to how much free space
we want.

(christos)

fix the fix of the fix of the format string.

(christos)

PR/47453: Martin Branderburg: Mail (mail, mailx) does not recognize messages
which have RFC 822 format dates.
XXX: Pullup 6

(christos)

Make the default disklabel (for disks without label) have saner value.
I.e. use d_secperunit for RAW_PART rather than 0x1fffffff for part a.

(tsutsui)

netbsd32_posix_spawn_fa_alloc: use the right length for path allocation.

This error lead to memory pool corruption when freeing kmem with wrong size.

(hannken)

Fix case when searching for NUL.

(matt)

Remove support for -l option which piped output through pr(1). Not really
useful in 2013, and gets rid of a fork()/exec() in a library function.

(agc)

add kernfs and procfs, support BEAGLEBOARD.local

(jmcneill)

cpu_switchto() returns argument oldlwp.  Not &ci->ci_curlwp.

(kiyohara)

Add macro _C_LABEL() for .S file.

(kiyohara)

Fix a debug printf format.

(mbalmer)

Make this compile with TRAPDEBUG enabled.
Pointed out by Holm Tiffe.

(martin)

Remove trailing whitespace.

(wiz)

Add ARM optimized version of strrchr.

(matt)

Implement the -n (nowrap) argument to wdiff(1).  With the -n argument
specified, and when a change spans two lines, the end of region
character sequence is displayed before the new line character, and the
start of region character sequence after it.

Add a test which implements change bars on the left for any changed
text:

% netwdiff -1n f1 f2 | sed -e 's/^/  /;/{+/s/^ /|/;s/{+//g;s/+}//g'

  ...
  static int
| difffile(const char *filename1, const char *filename2)
  {
  ...

(agc)

Add missing ! on str

(matt)

fix printf formats

(christos)

make autosizing work (kind of)

(christos)

Fix off-by-one read error.

(msaitoh)

not sure how my name ended up in the list of authors here

(jmcneill)

first step at autosizing, still does not work.

(christos)

don't use kprintf() directly, it does not lock so it just garbles the output.

(christos)

set kerneldir= to the path used by build.sh

(jmcneill)

expand set name to path when checking to see if the tgz exists

(jmcneill)

make metalog files trully optional (don't print errors when they don't exist)

(christos)

Add an ARM optimized version of strchr.

(matt)

Add simple/small versions of strchr/strrchr for ARM.

(matt)

proper test for case-insensitive netcmp(1).

(agc)

No need for a flags argument to the cmp_file() function.

Check that fp isn't NULL before trying to close it in diff_fini().

(agc)

Add rudimentary support for cmp(1) in the shape of netcmp(1). More to come
on this one, too.

(agc)

Add support to netwdiff to specify the insert and delete region
character sequences from the command line via the -w, -x, -y and -z
args. The default is, as before, "[- ... -]" for deleted regions,
and "{+ ... +}" for inserted text.

With this in place, the arguments can be used to highlight word diffs
portably across shells:

        csh/tcsh:
        % set ansi_red = "`printf '\e[31m'`"
        % set ansi_end = "`printf '\e[0m'`"
        % set ansi_green = "`printf '\e[32m'`"

        sh/ksh/bash/zsh:
        $ ansi_red="$(printf '\e[31m')"
        $ ansi_end="$(printf '\e[0m')"
        $ ansi_green="$(printf '\e[32m')"

        and then in all shells:

        netwdiff -w $ansi_red -x $ansi_end -y $ansi_green -z $ansi_end f1 f2

-- this will highlight deleted words in red, and added words in green,
on standard ANSI terminals.  This gets around bash's twee $'' idiom
for tputs sequences.  Thanks to Christos Zoulas for setting me
straight on this, and thanks also to Alan Barrett for the nudge to
implement the region arguments.

Thanks to all of the people who pointed me to comm(1)'s arguments as
being the model for wdiff's inhibit.  My apologies, I should have
thought of that.  I'm still trying to work out the model for the
choice of -w/-x/-y/-z; all suggestions gratefully received.

(agc)