convert program in PR kern/44054 to an atf test case

(pooka)

support compat ioctl's (OOOOOlalaSIOC stuff)

(pooka)

If someone wants to define COMPAT_OIFDATA / COMPAT_OIFREQ outside
of config, let them.

(pooka)

Pull in post-release fix 3d5597b0076ade841abf03fc274da72d17cb3ad6 to resolve
issues with the default NetBSD settings.  Tests were producing invalid results
because they were unexpectedly reading the system-wide settings.

(jmmv)

Note import of atf 0.12.

(jmmv)

Adjusts tests after import of atf-0.12:

- The use.fs property is gone.
- Mark the tests/fs/t_create:attrs test as broken when using the default
  unprivileged-user:_atf setting.  This probably deserves a fix somehow
  but I'm not sure at this point.

(jmmv)

Document configuration changes in tests(7); unprivileged-user is now set
out of the box and is set in 'common.conf', not 'NetBSD.conf'.

(jmmv)

Adjust file lists after atf-0.12 import

(jmmv)

Add the _atf user and group to be able to run unprivileged tests automatically
without having to manually tweak the 'unprivileged-user' setting.  Suggested
by pooka@.

(jmmv)

Adjust reachover build files after import of atf-0.12.

(jmmv)

Help merge of atf-0.12.

(jmmv)

Import atf 0.12:

Experimental version released on November 7th, 2010.

* Added the ATF_REQUIRE_THROW_RE to atf-c++, which is the same as
  ATF_REQUIRE_THROW but allows checking for the validity of the exception's
  error message by means of a regular expression.

* Added the ATF_REQUIRE_MATCH to atf-c++, which allows checking for a
  regular expression match in a string.

* Changed the default timeout for test cases from 5 minutes to 30 seconds.
  30 seconds is long enough for virtually all tests to complete, and 5
  minutes is a way too long pause in a test suite where a single test case
  stalls.

* Deprecated the use.fs property.  While this seemed like a good idea in
  the first place to impose more control on what test cases can do, it
  turns out to be bad.  First, use.fs=false prevents bogus test cases
  from dumping core so after-the-fact debugging is harder.  Second,
  supporting use.fs adds a lot of unnecessary complexity.  atf-run will
  now ignore any value provided to use.fs and will allow test cases to
  freely access the file system if they wish to.

* Added the atf_tc_get_config_var_as_{bool,long}{,_wd} functions to the atf-c
  library.  The 'text' module became private in 0.11 but was being used
  externally to simplify the parsing of configuration variables.

* Made atf-run recognize the 'unprivileged-user' configuration variable
  and automatically drop root privileges when a test case sets
  require.user=unprivileged.  Note that this is, by no means, done for
  security purposes; this is just for user convenience; tests should, in
  general, not be blindly run as root in the first place.

Status:

Vendor Tag: TNF
Release Tags: atf-0-12

(jmmv)

Replace ACPI_UINT64 by ACPICA's ACPI_INTEGER.  Remove superfluous casts.
No functional change.

(gsutre)

Sort options.

(uebayasi)

Use existing wscons USB keymap for WSKBD_TYPE_MAPLE on dreamcast.

(tsutsui)

Recognize WSKBD_TYPE_MAPLE for dreamcast.

(tsutsui)

Note fujitsu(4).

(gsutre)

Minor cleanup.

(wiz)

Bump date for schide(4).

(wiz)

Remove superfluous Pp.

(wiz)

Sort options.

(wiz)

Sort options in SYNOPSIS.

(wiz)

Take the internal functions and definitions back out of the implementation
namespace:

:g/\<__ops/s//pgp/g
:g/\<__OPS/s//__PGP/g
:g/\<OPS/s//PGP/g

No functional change, regression tests complete successfully.

(agc)

Elgamal encryption and decryption has been done - take it off the list of
tasks to do.

(agc)

Add Elgamal decryption to netpgp.  Inspired by (BSD-licensed) the
Elgamal decryption code from Postgresql by Marko Kreen.

% cp config.h f
% netpgp -e f
netpgp: default key set to "d4a643c5"
% netpgp -d < f.gpg > f.netpgp
netpgp: default key set to "d4a643c5"
signature  1024/DSA 8222c3ecd4a643c5 2010-05-19 [EXPIRES 2013-05-18]
Key fingerprint: 3e4a 5df4 033b 2333 219b 1afd 8222 c3ec d4a6 43c5
uid              Alistair Crooks (DSA TEST KEY - DO NOT USE) <agc@netbsd.org>
encryption 2048/Elgamal (Encrypt-Only) a97a7db6d727bc1e 2010-05-19 [EXPIRES 2013-05-18]
netpgp passphrase:
% ls -al f*
-rw-r--r--  1 agc  agc  5730 Nov  6 23:53 f
-rw-------  1 agc  agc  1727 Nov  6 23:53 f.gpg
-rw-r--r--  1 agc  agc  5730 Nov  6 23:54 f.netpgp
% diff f f.netpgp
%

This makes DSA keys into first class citizens, since encryption and
decryption using DSA/Elgamal is now supported.

(agc)

Add the ability to perform Elgamal encryption to netpgp.  Some of this
code is inspired by the (BSD-licensed) Elgamal crypto code in
Postgresql by Marko Kreen, but netpgp uses BIGNUM numbers instead of
MPIs, and its keys have a completely different structure, so much has
changed.

% cp config.h f
% netpgp -e f
netpgp: default key set to "d4a643c5"
% gpg -d f.gpg > f2

You need a passphrase to unlock the secret key for
user: "Alistair Crooks (DSA TEST KEY - DO NOT USE) <agc@netbsd.org>"
2048-bit ELG-E key, ID D727BC1E, created 2010-05-19 (main key ID D4A643C5)

gpg: encrypted with 2048-bit ELG-E key, ID D727BC1E, created 2010-05-19
      "Alistair Crooks (DSA TEST KEY - DO NOT USE) <agc@netbsd.org>"
% diff f f2
% ls -al f*
-rw-r--r--  1 agc  agc  5730 Nov  6 05:40 f
-rw-------  1 agc  agc  1727 Nov  6 05:40 f.gpg
-rw-r--r--  1 agc  agc  5730 Nov  6 05:41 f2
%

(agc)

Add schide(4) man page.

(jakllsch)

- similar to "bootconf", document that pxeboot_ia32.bin has the "module
loading" option turned off by default, and explain how to turn it on.
- remove XXX comment: yes, even when boot.cfg is loaded, DHCP is still
used to get file names (checked with a quick PXE boot + DHCP/TFTP platform)
- document in BUGS section that all DOM0 kernels are loaded as modules,
so you may have to explicitely set the boot device for root fs on network,
depending on your setup

Bump date. Blessed by agc :)

(jym)

remove unused variables.

(christos)

- bring some command definitions on par with their code
- describe the arguments that can be passed to NetBSD DOM0 kernel (bootdev,
console, ip, nfsroot, pciback.hide)
- document the "modules" command to boot(8)
- Xen is a trademark, hence .Tn
- add a reference to pcictl(8) for getting PCI IDs (used by pciback.hide)

Bump date.

(jym)

Bring the help command output from boot and pxeboot on par with their
code.

(jym)

call the sysctl destroy functions upon failure and unload, which
should fix PR#40029.  while there, fix the radeon unload, which
was improperly matching mutex init/destroy operations and leading
to trying to destroy lock that was never initialised.

(mrg)

pull across a few changes from the freebsd folks:

http://svn.freebsd.org/changeset/base/213044
- fixes gunzip issues
http://svn.freebsd.org/changeset/base/213927
- fixes various typos and comments

and also an older change to add support for bzip2's "-k" option:
don't delete the input file

thanks!

(mrg)

remove an XXX comment.

(mrg)

add a "-T" option to disable the twiddler, which can help booting
with serial output slowness i've seen.

(mrg)

fix NAMED_USE_PTHREADS != yes usage.

(mrg)

note raidframe(4) supports >2TB devices now.

(mrg)

PR/44052: Takahiro HAYASHI: chrooted ntpd refers freed memory
As mentioned in the pr, process_routing_msgs can delete the current reader,
so cache the next element before calling it.

(christos)

Welcome to 5.1!

(snj)

can't map the old and the new SIO calls the way we did before because the
numbers have changed. Instead provide a switch. Keep the old code there,
to handle cases we did not handle in the first switch, but this is a hack
and should be removed.

(christos)

PR/44054: Onno van der Linden: Stacksmashing in handling of ioctl OOSIO*
parameter.

(christos)

Merge from uebayasi-xip:
----------------------------
revision 1.60.2.5
date: 2010/08/14 02:09:57;  author: uebayasi;  state: Exp;  lines: +2 -1
Teach TLB miss handler (pmap_tlbmiss()) to map "Expansion ROM" area as
PA == VA.  Now we don't need to reserve a TLB entry for it.
----------------------------

(uebayasi)

More registers to support IBM405GPr "External Bus".

(uebayasi)

Add IBM405GPr "External Bus" driver.

(uebayasi)

Merge from uebayasi-xip:

Support files other than kernel executables.

For filesystem images, embed (not prepend) OpenBIOS image headers to
the images, so that alignment is kept in the resulting files.

(uebayasi)

Typo.

(uebayasi)

Handle md(4)'s embedded root disk image and size symbol names other
than the default ones ("md_root_image" / "md_root_size").

(References to xmd(4) are deliberately removed.)

(uebayasi)

Provide a forward declaration of "struct vm_page", whose internal
is opaque to uvm_pglist.h users.  Users don't need to pull in
uvm_page.h.

(uebayasi)

Remove incomplete, never worked dynamic run-time memory registration
(uvm_page_physload(9)).  This functionality will be re-added later.

(uebayasi)

Note addition of schide(4).

(jakllsch)

Put the standard TNF copyright.

(uebayasi)

fwhrng(4) is in i386 GENERIC already, add it to ALL too.

(jakllsch)

Enable siisata(4) in Xen dom0 kernels.

(jakllsch)

Enable schide(4), Intel SCH IDE controller driver from PR#42310.

(jakllsch)

Add Intel SCH IDE controller driver.
From OpenBSD via Tomokazu HARADA in PR#42310.

(jakllsch)

regen.

(jakllsch)

Add Intel SCH IDE Controller, from OpenBSD via Tomokazu HARADA in PR#42310.

(jakllsch)

Include uvm/uvm_pglist.h for struct pglist.

(uebayasi)

Include uvm/uvm.h because this is part of UVM.

(uebayasi)

Machine dependent code is considered as part of UVM.  Include
internal API header.

(uebayasi)

These don't use UVM physical page API, don't need uvm/uvm_page.h.

(uebayasi)

Include uvm/uvm.h to use UVM internal type (struct vm_page).

(uebayasi)

Include uvm/uvm_external.h to access UVM external API (uvm_km_*(9)).

(uebayasi)

Include uvm/uvm.h to access UVM internal.

(uebayasi)

fix up compat libraries for mips64
(liblua & liblzma partly missing, libmj wrong version)

(plunky)

uebayasi-xip: update base.

(uebayasi)

Sync with HEAD.

(uebayasi)

Sync with HEAD.

(uebayasi)

Sync with HEAD.

(uebayasi)

Fix build.

(uebayasi)

Unbreak Xen build, while not actually fixing the real problem.
NetBSD/xen doesn't implement disestablishing interrupts yet.

(jakllsch)

specify the libmj.la archive as a pre-req for libnetpgp.la in the
correct way - tested on RHEL 6 and Mac OS X.

(agc)

ftpd/sftp glob stuff is covered by SA2010-008.  No need to duplicate it.

(snj)

Enable detachment.

Tested on SiI 0680 (native mapped).

(jakllsch)

Enable detachment.

Tested on nForce 4 IDE (compatiblity mapped) and SATA (native mapped).

(jakllsch)

Implement pciide_machdep_compat_intr_disestablish() to help enable
detachment of compatibility-mapped pciide(4)-family controllers.

(jakllsch)

Rework pciide(4) detachment to take the legacy interrupt mapping into
consideration and avoid future code duplication.

Ports wanting to enable detachment of controllers with compatibility-mapped
channels will need to supply a pciide_machdep_compat_intr_disestablish()
function.

(jakllsch)

re-run automake to pick up changes to Makefile.am files.

(agc)

Avoid specifying the manual page twice in the Makefile.am files, which
can lead to problems at installation time on some platforms (RHEL 6,
for example), whereas Mac OS X seems to install things fine.

(agc)

Apply the patch provided in PR 44047 by Peter Pentchev to fix a problem
in the GNU autoconf infrastructure with ltmain.sh script - change all
occurrences of "$echo" to "$ECHO".

(Incidentally, this does not show up under pkgsrc, since pkgsrc uses
the platform's libtool instead of the script which comes with the
distribution)

(agc)

pciide(4): Sprinkle static, add detachment support.

(jakllsch)

Correct copy/paste error in as-of-yet-unused pciide_common_detach().

(jakllsch)

Pave the way for detachment of pciide(4)-family ATA controllers.

(jakllsch)

Need getifaddrs() for the rump client since it executes sysctl(2)

(pooka)

make sysctl(8) work as a rump client

(pooka)

Mark a buffer with EIO in wdstrategy() if the device is not "enabled"
(more like "present"), not if it is merely asleep.

If the device is not awake, get out of wdstart() right away.

Brett Lymn and KAMADA Ken'ichi report that this stops the kernel from
crashing after a suspend/resume cycle.

(dyoung)

Remove .Pp before .Bl and bump date.

(gsutre)

Implement support for unix domain sockets (important especially
for testing since we don't want to depend on global resources such
as tcp ports).

(pooka)

adapt to new rumpclient_init() signature

(pooka)

use -1/errno for failure: it's much more convenient for the users

(pooka)

support rump client mode

(pooka)

Don't echo key symbols.

(blymn)

+tp

(pooka)

test fss with underlying fs mounted with MNT_LOG

(pooka)

remove newline from error

(pooka)

Mention fujitsu(4).

(gsutre)

Make the list pretty.

(jruoho)

Serial comma.

(wiz)

To allow static linking, specify -lterminfo as well, and don't rely
on shared library inter-dependencies to pull everything together.
Fixes the build for sun2.

(he)

Hm, the warning sign was not big enough.

(gsutre)

Merge ACPI Fujitsu Driver.  Provides support for hotkeys and other
built-in components through vendor-specific ACPI devices.

ok jruoho@

(gsutre)

File system snapshots work with wapbl.

(hannken)

File system snapshots are no longer experimental.

(hannken)

Rename netbsd-compat50-5.1.tgz to compat50-5.1.tgz for consistency.
URL approved by spz@

(sborrill)

New pkg_install requires -llzma, so to support static linking, be
explicit about what libraries are required, and not only depend on
shared library inter-dependencies to pull that in.
Fixes the build for sun2.

(he)

Document DIOCGPHYSSEG.

(uebayasi)

Brief introduction of xip(4).

(uebayasi)

Document HOWTO.

(uebayasi)

Only read from HID1 when a G3 CPU was detected in cpu_get_dfs(), otherwise
the access might cause a PGM trap.

(phx)

Only include link to binary compatibility packages on relevant ports

(sborrill)

Explicitly link netpgpverify with libmj, rather than letting libnetpgp
bring it in.

Fixes a build issue on RHEL 6.

(agc)

Explicitly link netpgp with libmj, rather than letting libnetpgp bring it
in.

Fixes a build issue on RHEL 6.

(agc)

Explicitly link netpgp with libmj, rather than letting libnetpgp bring it
in.

Fixes a build issue on RHEL 6.

(agc)

1465

(snj)

Pull up following revision(s) (requested by martin in ticket #1465):
sys/arch/i386/stand/misc/rawr32.exe.uue: revision 1.1
sys/arch/i386/stand/misc/Makefile: revision 1.7
sys/arch/i386/stand/misc/rawr32src.zip.uue: file removal
sys/arch/i386/stand/misc/rawr32.zip.uue: file removal
Update Rawrite32 to the latest version

(snj)

Pull up following revision(s) (requested by martin in ticket #1465):
distrib/notes/common/main: revision 1.467
Adapt to newer Rawrite32 version

(snj)

Pull up following revision(s) (requested by martin in ticket #1465):
distrib/notes/macppc/prep.OPENFIRMWARE: revision 1.14
Update rawrite32 url

(snj)

Miscellaneous changes to bring the GNU autoconf framework into a state
where a package can be build and install properly.

(agc)

avoid calling a debug function that's not exported in libnetpgp

(agc)

Find the libmj header file in the correct directory

(agc)

run autoconf so we can generate a Makefile in src/libmj

(agc)

generate the Makefile in src/libmj too

(agc)

ip_randomid: make mechanism MP-safe and more modular.

OK matt@

(rmind)

Add pfsync to ALL kernel.

(rmind)

ip_reass_packet: finish abstraction; some clean-up.
Discussed some time ago with matt@.

(rmind)

#ifdef variable correctly

(pooka)

be consistent in what to leave outside of SMALL

(pooka)

Support RUMP_ACTION to compile a rump kernel client.

(pooka)

Update RUMP_ACTION to use the newstyle rump client.

(pooka)

make that 48

(pooka)

Replace the copyright with a new TNF copyright since nothing of the old
ip_id.c remains.  Remove old comments which have no relevance anymore.

(matt)

scrub the whole array, not just the first n malloced entries.

(christos)

Add rumpclient to setlists.

You know, I used to think setlists were ridiculous back when ~20
entries per lib were required.  It seems we're up to 47 and I've
long since run out of superlatives to describe the situation.

(pooka)

Add library for rump syscall client stubs.

(pooka)

bump rumpuser version

(pooka)

Refactor the sysproxy code so that rumpuser contains only the server side.

(pooka)

regen: standalone client support for rump syscalls

(pooka)

Create rump_syscalls.c in two flavours: one which can be compiled
standalone as the clientside and one which expects to be compiled
into the rump kernel.

(pooka)

Initialize pointers to NULL to avoid accidentally free(9)'ing garbage.
Should address the panic reported by Alan Bueno in PR # 44042. The root
cause (an empty _DOD) is yet to be determined.

(jruoho)

Extend discussion of libpixman and libXfont compat issue and compatibility
packages (URL is not yet 100% decided, but is currently correct).

Rename discussion of pthread and SA support to fix duplicate subsection names
(previously two sections about upgrading from 4.x).

Change some text from 5.0 to 5.x.

(sborrill)

regen

(pooka)

apply "shut up, lint" for syscallargs.h

(pooka)

fill in bits for 5.1:
* SA list
* dedication
* statically linked pthreaders have been fixed

(pooka)

Apply the patch from Peter Pentchev in PR 44041

``When netpgp thinks (even for a little while, even if it
should change its mind almost immediately) that it's
processing a detached signature, it sends a message to that
effect to io->outs.
This kind of interferes with "netpgp --cat" :)''

Use io->errs instead of io->outs for the informational message.

(agc)

Apply patch from Peter Pentchev in PR 44040

The patch fixes two problems when verifying a clearsigned message:
- a copy/paste error - "litdata" should be "cleartext"
- a use of an uninitialized variable, resulting in freeing
  an uninitialized pointer on the stack... resulting in a segfault

(agc)

Note the user-specified cipher has been implemented

(agc)

add an additional test for the user-specified cipher

(agc)