there is no REL_DEBUG

(christos)

- Add support for "midnight" "noon", dawn etc.
- Add validation to date/time strings by checking that mktime did not change
  the fields of struct tm from the ones requested
- Allow the format "year monthname day".
From kre

(christos)

new tests from kre

(christos)

don't release mutexes you did not enter? Please develop with LOCKDEBUG on!

(christos)

regen

(christos)

name moschip devices consistently

(christos)

Bugfixes
  - Calculate cycle_time correctly. _ftb is from +127 to -128.
  - Print latency in correct order.
- Don't print the size twice.
- KNF

(msaitoh)

Modularize drvctl(4)

(pgoyette)

remove extra shebang and fix a potentially bug

(knakahara)

Handle reading of 4kbit EEPROMs in SPD devices. This is needed for DDR4 DIMMs.

(mlelstv)

Forgot this file in previous commit (modularization of clockctl(4))

(pgoyette)

CID 1341546: Fix integer handling issue (CONSTANT_EXPRESSION_RESULT)

n > INT_MAX where n is a long integer variable never be true on 32bit
architectures. Use time_t(int64_t) instead of long for the variable.

(ozaki-r)

synch usage() with the man page.

(pgoyette)

Modularize the clockctl pseudo-device and link to the build.

(pgoyette)

Improve the httpd(8) printenv.lua Lua example

Stop using Lua builtin print function and replace them with http.* ones.
httpd.print and http.write wraps SSL support when needed.

Print http headers, without them browser may interpret page as raw text.

No need to hardcode prefix path in the form.

Add comments for a user with tips how to use this script.

Patch by Travis Paul

Closes PR misc/50502

(kamil)

new openssl

(christos)

merge conflicts

(christos)

Restructure the xfer methods of ohci(4) so that minimal work in done in
softint context.  Now all memory allocation is done in thread context.

Addresses kern/48308 for ohci(4), might fix some locking bugs around the
*TD free lists, and plugs some memory leaks on error conditions.

(skrll)

Fix west australia offset, from kre

(christos)

gdb-7.10.1 out.

(wiz)

KNF.

(msaitoh)

Fix typo in comment.

(wiz)

Bump date for previous.

(wiz)

Bump date for previous.
Remove trailing whitespace.

(wiz)

Fix the definition of DDR4's tFAW (offset 0x24).

(msaitoh)

Clean up assertions and catch integer overflow.

(mlelstv)

document new pin configuration flags

(mlelstv)

pass new pin configuration flags to driver.

(mlelstv)

remove more duplicated code.

(christos)

Make options SCIFCONSOLE patchable via a variable using gdb(1) etc.

Tested on dreamcast.

(tsutsui)

Convert tsleep(9)/wakeup(9) pairs to condvar(9) with mutex(9).

Briefly tested with mkbd(4).

(tsutsui)

Add listing commands for type, set and unset.
Add help and formatting for set and unset.
Change show to print all the attribute info in one line.

(christos)

Add test for type, set, unset

(christos)

try 2

(marty)

more variables

(christos)

file shuffling

(christos)

new files

(christos)

Add debugging and more to migration

(christos)

Add a migration test

(christos)

fix div clk set/get

(jmcneill)

Add biosboot test

(christos)

Add clock controller backend for Exynos5422. Work in progress.

(jmcneill)

Add basic clock device infrastructure.

(jmcneill)

Sprinkle __attribute__((__noreturn__)) to fix build on 7.99.23 with clang.

(wiz)

ifdef out unused variables to fix build on NetBSD 7.99.23/amd64 with gcc

(wiz)

Fix a date to make this more reproducible.

(wiz)

messing with uninitialized structs is a bad thing

(jnemeth)

Use the DDR4 rows and columns fields when printing DDR4 data

(pgoyette)

Calculated CRC is for bytes from 0 to 125 on DDR4. Now we can attach spdmem
driver for DDR4.

(msaitoh)

Fix lying comment.

(dholland)

The COMPAT_10 code for SYSVxxx is now part of the sysv_ipc module, so
don't install syscall function pointers from the compat module.

(pgoyette)

When unloading the module, clear the address of the kauth_listener
after removing the listener.  This enables the module to (attempt to)
initialize again without hitting the KASSERT() is sysvipc_init().

XXX This will allow standard kernels to boot successfully.  However,
XXX module initializtion is still broken.

(pgoyette)

Use DIOCGMEDIASIZE instead of DIOCGPART so that we are not limited to 2G.
XXX: All DIOCGPART code needs to be removed...
XXX: pullup-7

(christos)

type confusion

(christos)

Oops +

(christos)

add gpt directory

(christos)

remove extra

(christos)

Add gpt tests

(christos)

Add gpt tests

(christos)

Add gpt tests

(christos)

Add remove partition data file

(christos)

Add label change tests

(christos)

Add resize data files

(christos)

Add resize test.

(christos)

Kill newline

(christos)

Fix resizedisk.

(christos)

Update comments to reflect reality

(skrll)

Use tabs instead of spaces

(skrll)

Shave another whole instruction when KERNEL_BASE_VOFFSET == 0

(skrll)

Model 0x5e is also 6th gen Core or Xeon E3-1500 v5 like model 0x4e.

(msaitoh)

gif(4): Infinite recursion calls prevention code works again now.

The prevention code haven't worked since gif(4) was changed
to use softint(9). To work this prevention, git_output uses
m_tag(9) like FreeBSD and OpenBSD.

I tested with following code.
====================
# ifconfig gif0 create
# ifconfig gif0 tunnel 10.1.1.1  10.1.1.2
# ifconfig gif0 inet 192.168.100.1 192.168.100.100

# ifconfig gif1 create
# ifconfig gif1 tunnel 192.168.100.1 192.168.100.100
# ifconfig gif1 inet 192.168.101.1 192.168.101.101

# ifconfig gif2 create
# ifconfig gif2 tunnel 192.168.101.1 192.168.101.101
# ifconfig gif2 inet 192.168.102.1 192.168.102.102

# ping -w 1 -c 1 192.168.102.102
# dmesg | tail -n 1
gif0: recursively called too many times(2)
====================

(knakahara)

put back usage from kre

(christos)

don't print dkctl message for files
fix typo

(christos)

fix label printing.

(christos)

fix the open error messages to include the right file and strerror

(christos)

Make the test work

(christos)

checking multiple things does not work! atf-bug!

(christos)

I wrote that

(christos)

Add some gpt tests

(christos)

Update dependencies - compat_linux32 requires compat_netbsd32_sysvipc

(pgoyette)

CID 1341564: Arg memory leak

(christos)

more leak

(christos)

CID 1341563: fix leak

(christos)

CID 1341561: Argument memory leak

(christos)

CID 1341560: Free arg.

(christos)

CID 1341558: Fix proplib memory leaks

(christos)

CID 1341556: Don't leak map info

(christos)

CID 1341552: Fix NULL deref

(christos)

CID 1341552: Fix NULL deref.

(christos)

CID 1341551: Don't bother formatting if ti == NULL

(christos)

CID 1341548: Check error returns

(christos)

simplify.

(christos)

PR/50496: Alex Kozlov: teach unzip to extract symlinks

Our libarchive does not handle symliks yet, but later versions do.
Set mode and times for symlinks. Finally handle hardlinks for
completion although they are not handled yet by librarchive.

(christos)

It is wrong to -I/usr/include/ssp

(christos)

Strange patch(1) behavior got me three copies of the code.  One is
sufficient so remove the excess.

(pgoyette)

If compiling as part of a built-in kernel module, make sure we include
the relevant opt_sysv.h file

(pgoyette)

Missed these.

Only include the compat_14 sys calls if this being compiled as part of
the compat_netbsd32_sysvipc module.

(pgoyette)

root.cache is 2015120100 (December 01, 2015) now.

(taca)

Update to 2015120100.

* Addresses of H.ROOT-SERVERS.NET has updated.
* Make IPv6 address lower case.

(taca)

Regen syscall definitions - the compat_netbsd32 sysv-ipc stuff is now
modular and can be autoloaded when needed

(pgoyette)

Split out the SYSV IPC stuff into its own compat_netbsd32_sysvipc
module.  Adjust dependencies as needed.

(pgoyette)

compare to stdout, thanks kre

(christos)

Bring READMEs up to date.  Update $ARFE$.

(dyoung)

PR/50492 - Nikolai Lifanov -- add MBA

(jnemeth)

LIST_REMOVE should be done before clearing members of the list element.

(knakahara)

Update sets list for merge of compat_sysv --> sysv_ipc

(pgoyette)

Regen

(pgoyette)

Merge the compat_sysv module into the sysv_ipc module - it should
never have been a separate module in the first place (my bad).

Adjust dependencies as appropriate.

(pgoyette)

remove extra encap_detach().

encap_detach() is already done in gif_delete_tunnel()->in{,6}_gif_detach().

(knakahara)

Add a space.

(christos)

Fix stray paste, noted by kre.

(christos)

WARNS=6

(christos)

check the return values from the "put" method

(agc)

add c++ compilation guard to the header file

(agc)

remove extra assignment

(christos)

eliminate static globals so that commands can be re-used.

(christos)

Add comment to #ifdef - it's a long way from its original #if

(pgoyette)

Make CVS ignore *.undo.

(dyoung)

Straggler from last: add sym.[ch].  Remove some files belonging to dead
tests.

(dyoung)

Straggler from last: delete some tests, add a test.

(dyoung)

Executive summary: ARFE now understands C-like symbols.  I'm using a
different algorithm to figure out where to subdivide the longest common
subsequence search.  I'm actually computing an edit distance instead of
the longest common sequence, but the algorithms are duals so there's not
much practical difference.  I've discarded some tests, and added at least
one new one.

Qualify %d for ptrdiff_t, %td.  Quiets compilation on 64-bit Darwin.

Change the type of the dynamic program cells from size_t to cell_t.  For
now, a cell_t is just a struct containing a size_t score.

Add algq(), a routine for finding k such that

        lcs(A[1:m/2], B[1:k]) | lcs(A[m/2+1:m], B[k+1:n]) = lcs(A, B).

algq() is based on the function Half(i, j) defined in Jeff Erickson's
(jeffe@cs.illinois.edu) lecture notes on advanced dynamic programming.
See http://jeffe.cs.illinois.edu/teaching/algorithms/.

Add to the Makefile (commented out) lines for tracking code coverage.
Use gcov <source file> to see the coverage.

Make the cleandir target remove gcov(1)-related files.

Delete a bunch of dead code and the now unnecessary argument to algc(),
expected_lcs.

We don't need backwards subslices any more, so get rid of that.  When
I got rid of the slice_t member `backward' and all of its uses, GCC
inlined clocc_ends_at() with a really bad effect on performance (>10s
on elmendorf for the t/netstat-s.[01] test, instead of <8s).  I marked
clocc_ends_at() __noinline for a net performance gain.  The gcc version
is (NetBSD nb2 20110806) 4.5.3, btw.

Disable the dbg_assert()s for more reliable performance comparisons.

Rename algq's splitn argument to splitnp since that's my convention
for arguments of that kind.

In algq(), don't get(A, i) m x n times, just get(A, i) m times.

Lightly constify.

Provide __noinline on non-NetBSD systems.

Sprinkle the $ARFE$ keyword.

Rename algc -> findlcs, algq -> findsplitn.

Make Subversion fill $ARFE$ in macaddr.h.

Extract the tags target from {dt,it,tt}/Makefile, put it
in ./Makefile.inc.

We only ever call findsplitn(..., true), so get rid of the do_clocc
argument.

Add an experimental routine, count_records(), that tries to count the
records in its second slice_t argument, using the first slice_t argument
as record template.

Change the class-occurrence (clocc_t) score, clocc_score(), to one plus
the minimum length of the class occurrences, from one plus the product
of the class occurrences' lengths.  This speeds things up a bit.

Simplify findsplitn() by pulling common statements out from if-else
branches, et cetera.

In clocc_starts_in_slice_at(), pass the wlenp argument to
clocc_starts_at().  Nothing passed a non-NULL wlenp to
clocc_starts_in_slice_at(), so this doesn't make any functional
difference.

Compute the edit distance instead of the longest common subsequence.
The one algorithm is a dual of the other.  I may find it easier to
add to the edit distance algorithm improvements like affine gap
penalties, hence the change.

Snapshot of work in progress.  These changes make things quite a bit
slower!  Add affine gap penalties.

Bring count_records() in line with findsplitn(), adding affine gap
penalties.  Update the instrumentation.  Count up the number of gaps
accumulated.

XXX This change makes 'dt netstat-s.0 netstat-s.1' more than twice as
XXX slow as it used to be, owing largely (I think) to the increase in
XXX size of a cell_t, where three ssize_t's track the number of gaps.

Exit with a message and error return code if we run out of slots for
class occurrences.  The class-occurrence array is still statically
allocated---yech.  I'm going to fix it one of these days, I promise.

Stop detecting occurrences of class "string" (KIND_STRING), which
consisted of the names 'abe', 'ada', and 'daria'.  Remove the tests
related to that.

Start detecting occurrences of class "symbol" (KIND_SYMBOL), which
resemble C symbol names: they start with a letter of the alphabet or
underscore.  Following characters are letters, numbers, or underscore.
Update tests to match: the netstat and ifconfig tests produce much more
sensible results, now.  Delete the 'quack<number>quack' tests, since
the symbol detector matches the entire string, now, and the tests don't
stand for any practical use-case.

Add test #5 to tt, which demonstrates how one can use a symbol in the
match template to match a symbol in the input for reproduction in the
transform template.

(dyoung)

- factor out into smaller separate functions
- fix signed/unsigned confusion
- do proper write checks
- fix some memory leaks

(christos)

don't hard-code the program name

(christos)

merge duplicated code and check allocations.

(christos)

Sanity checks in ohci_dump_tds

(skrll)

check allocations

(christos)

annotate map additions with non-allocated memory.

(christos)

Whitespace

(skrll)

- check errors from map allocation
- make map_resize return -1 instead of 0, and handle errors locally
  explaining what's going wrong

(christos)

PR/44915: Havard Eidnes:
- Convert NS_NOTFOUND to EAI_NONAME instead
  of EAI_NODATA which more inline with what other OS's do.
- Convert NO_DATA to NS_NOTFOUND since there is no equivalent mapping
  for nsswitch (perhaps we could add one so what we could recover the exact
  error and return EAI_NODATA?)
- If we end up following a CNAME chain that does not find any data return that
  instead of internal error.

XXX: pullup-7

(christos)

Allow backup and restore to operate on files.

(christos)

don't advertise stuff we don't support.

(christos)

Always set NOSYNC if regular file, even if we got the size from the command
line.

(christos)

hrmm...

i386devel: {245} ./gpt add -b 1000 -s 1000 disk
Segmentation fault (core dumped)

Where's my flyswatter?  .......zzzzzzzap

(jnemeth)

- dereferencing NULL is generally bad, found by visual inspection
- automatically set nosync when opening a file to avoid spurious errors

(jnemeth)

Group TD/ITD assignments.  No functional change.

(skrll)

amend ticket 1005

(snj)

Pull up following revision(s) (requested by mrg in ticket #1005):
external/gpl3/gcc/dist/gcc/doc/gcc.info: patch
external/gpl3/gcc/dist/gcc/doc/gccinstall.info: patch
external/gpl3/gcc/dist/gcc/doc/gccint.info: patch
external/gpl3/gcc/dist/libitm/libitm.info: patch
Doc updates for gcc 4.8.5.

(snj)

Make checks strict

rump.arp should fail with File exists (EEXIST).

(ozaki-r)

print the old and the new style usage.

(christos)

refactor the utf code so that it does not leak memory.

(christos)

split out into smaller functions.

(christos)

use the utility function.

(christos)

Don't bother calling log for values <= 1... XXX: should use ilog() instead.
Fixes vax.

(christos)

Skip protective MBR early too.

(christos)

- write a function to set the entity, so that we don't reset to guid.
- entry 0 means create entry.

(christos)

By default, don't display the module load address.  Add a -k option
to restore display of this field.

(pgoyette)

Fix evaluation of defined(FOO) and ${FOO} > 0
add a unit-test to catch it.

(sjg)

Ooops, remove a local change that wasn't supposed to be committed.

(pgoyette)

Add new compat_netbsd32_mqueue module to sets lists.

(pgoyette)

Create new compat_netbsd32_mqueue module, and descend into subdir to
build it.

(pgoyette)

Regen for new compat_netbsd32_mqueue module

(pgoyette)

Prepare for new compat_netbsd32_mqueue module to be split out of
compat_netbsd32.  Refactor the COMPAT_50 code related to mqueue,
and update the syscall definitions.  Remove dependency on mqueue
from the compat_netbsd32 module (the new module will gain this
dependency).

Continuation of work on PR kern/50489

(pgoyette)

Properly mark netbsd32_mqueue.c to depend on 'options MQUEUE'

Yes, MQUEUE is included in all kernels by default, but it can be removed.
If it is removed, the compat_netbsd32 module should not include it, either.

(pgoyette)

merge command line parsers and check all memory allocations.

(christos)