Ticket 1912, 1921 and 1922.

(msaitoh)

Ticket 1912, 1919, 1921 and 1922.

(msaitoh)

Pull up following revision(s) (requested by riastradh in ticket #1922):
sys/dev/pci/pci_usrreq.c 1.26 via patch

Fix to make pci(4) reject unaligned configuration register reads and writes
before feeding them to a kassert in pci_conf_read/write or to a trap in the
hardware itself.

(msaitoh)

Pull up following revision(s) (requested by riastradh in ticket #1922):
sys/dev/pci/pci_usrreq.c 1.26 via patch

Fix to make pci(4) reject unaligned configuration register reads and writes
before feeding them to a kassert in pci_conf_read/write or to a trap in the
hardware itself.

(msaitoh)

Pull up following revision(s) (requested by riastradh in ticket #1922):
sys/dev/pci/pci_usrreq.c 1.26 via patch

Fix to make pci(4) reject unaligned configuration register reads and writes
before feeding them to a kassert in pci_conf_read/write or to a trap in the
hardware itself.

(msaitoh)

Pull up following revision(s) (requested by maxv in ticket #1921):
sys/miscfs/umapfs/umap_vfsops.c: revision 1.94
1) 'error' is returned while it does not even hold an error code. Which
  means that zero is returned, and the kernel keeps mounting (and it
  probably ends up in a deadlock/memory corruption somewhere).
2) 'nentries' and 'gnentries' are int and user-controlled, and there's no
  check to ensure they are greater than zero. Since they are used to
  compute the size of two copyin's, a user can control the copied size
  by giving a negative value (like 128-2^29), and thus overwrite kernel
  memory.
Both triggerable from root only.

(msaitoh)

Pull up following revision(s) (requested by maxv in ticket #1921):
sys/miscfs/umapfs/umap_vfsops.c: revision 1.94
1) 'error' is returned while it does not even hold an error code. Which
  means that zero is returned, and the kernel keeps mounting (and it
  probably ends up in a deadlock/memory corruption somewhere).
2) 'nentries' and 'gnentries' are int and user-controlled, and there's no
  check to ensure they are greater than zero. Since they are used to
  compute the size of two copyin's, a user can control the copied size
  by giving a negative value (like 128-2^29), and thus overwrite kernel
  memory.
Both triggerable from root only.

(msaitoh)

Pull up following revision(s) (requested by maxv in ticket #1921):
sys/miscfs/umapfs/umap_vfsops.c: revision 1.94
1) 'error' is returned while it does not even hold an error code. Which
  means that zero is returned, and the kernel keeps mounting (and it
  probably ends up in a deadlock/memory corruption somewhere).
2) 'nentries' and 'gnentries' are int and user-controlled, and there's no
  check to ensure they are greater than zero. Since they are used to
  compute the size of two copyin's, a user can control the copied size
  by giving a negative value (like 128-2^29), and thus overwrite kernel
  memory.
Both triggerable from root only.

(msaitoh)

Pull up following revision(s) (requested by sborrill in ticket #1919):
sbin/ccdconfig/ccdconfig.c 1.54 via patch
sys/dev/ccd.c 1.152 via patch
sys/dev/ccdvar.h 1.34 via patch

Switch size_t to uint64_t in appropriate places to ensure that ccd(4)
works with component and total sizes of > 2TB.
Make kernel print device information when a ccd configured.
Fix some typos in comments.

(msaitoh)

Pull up following revision(s) (requested by taca in ticket #1912):
etc/namedb/root.cache patch
doc/3RDPARTY patch

Sync root.cache with the latest -current (rev. 1.18).

(msaitoh)

Pull up following revision(s) (requested by taca in ticket #1912):
etc/namedb/root.cache patch
doc/3RDPARTY patch

Sync root.cache with the latest -current (rev. 1.18).

(msaitoh)

Pull up following revision(s) (requested by taca in ticket #1912):
etc/namedb/root.cache patch
doc/3RDPARTY patch

Sync root.cache with the latest -current (rev. 1.18).

(msaitoh)

mark fstab .PHONY to prevent it being re-made from fstab.c in suffix rules.

(christos)

remove unused define

(christos)

Make the initial svcstack is doubleword aligned if EABI.

(matt)

Don't use ${MACHINE} but zaurus instead.

(matt)

Use .S instead of .s

(matt)

Use .S instead of .s

(matt)

Pulled up additional changes required for ticket #52.

(riz)

Pull up following revision(s) (requested by manu in ticket #52):
sys/fs/puffs/puffs_msgif.h: revision 1.81
sys/fs/puffs/puffs_sys.h: revision 1.85
sys/fs/puffs/puffs_vnops.c: revision 1.183
Add a oflags input field to open requests so that the filesystem can pass
back information about the file. Implement PUFFS_OPEN_IO_DIRECT, which
will force direct IO (bypassing page cache) for the file.

(riz)

Restore parenthesis to cast in the right order.

(joerg)

Make it compile without DIAGNOSTIC

(reinoud)

Replace casts in the inline version of _rtld_call_function_void of
_rtld_call_function_addr with ifdef on RTLD_LOADER.

(joerg)

Partial viadrm2 snapshot.

To do:

- autoconf attachment (shouldn't be hard)
- viafb (maybe steal unichromefb and adapt attachment structure)
- actually run it (no hardware here)

(riastradh)

Add Linux timer_pending and del_timer.

(riastradh)

Really revert to the original version.

(joerg)

Allow STT_GNU_IFUNC to be dynamic symbols

(matt)

Allow STT_GNU_IFUNC to be a dynamic symbol.

(matt)

Revert. There is no such thing as __HAVE_IFUNC and there should not be
one in first place.

(joerg)

Use more appropriate macro for command line command argument.

(wiz)

Make sure we default to ELFv1.

(matt)

missed setting TAG_IS_LINKED in a special case in my previous commit.
Bug reported by Brad Harder, thanks!

(aymeric)

fix set but not used.

(christos)

Modify WMP_F_1000{X,T} to WMP_F_{FIBER,COPPER}. No binary change.

(msaitoh)

Go back to skipping init/fini handling for ldd, makes problems for 32bit
compat on LP64.

(joerg)

Make it compile without DIAGNOSTIC by using __diagused

(reinoud)

Make it compile without DIAGNOSTIC

(reinoud)

Adding extra dependencies does not affect the inference rule search anymore.
XXX[1]: should that be the case?
Simplify the code anyway via selective path search.
XXX[2]: Split arch portions to arch specific files, this is a mess

(christos)

Make it compile without DIAGNOSTIC

(reinoud)

effectively disable the test to fix the build and explain why.

(christos)

Implement pci_intr_setattr()

(well, kinda.  it just ignores the MPSAFE param for now, but that's
easy to fix later)

(pooka)

fix sets

(christos)

Fix build with TFTPROOT option enabled

(manu)

alpha (and others) don't like .set name, expr. Use name = expr like elsewhere.

(christos)

Introduce a cast to fix the build for Elf32 on _LP64.

(christos)

Return 0, not uninitialized, if the condition is already true.

(riastradh)

Add basic support for indirect functions. It allows providing a public
function symbol with an implementation choosen at run time.
Refactor calls to functions by address in ld.elf_so to create temporary
function descriptors on the stack, if the address is not leaked outside.

Limitations:
- no support for initialising static storage with function pointers
- no support for unnamed resolver functions

Inspired by FreeBSD's r228435 by kib@freebsd.org.

(joerg)

Allow creation of indirection functions on all platforms, not just GNU
and FreeBSD.

(joerg)

Remove and obsolete rumpkern_time as a separate component.

(pooka)

Enable Xuhost gpio pins; they are the same on Exynos5 series.

(reinoud)

Relax the timing around the usb lan9730 powercycle.

(reinoud)

Go back to the initial context (as tests/lib/libc/sys/t_swapcontext.c does)
after checking pthread_self() didn't change. Otherwise the process exits
outside of atf context.
Should fix "Test case exited normally but failed to create the results file: Results file is empty" reports from atf-run.

(bouyer)

use the same dynamic sources trick to make sure that the first dependency
is the c source file so that IMPSRC gets set correctly.

(christos)

clang appeasement

(macallan)

No longer create a special process context to fork remote clients off
of, simply always rfork off of proc1 closing all descriptors, and have
the rump kernel open 0/1/2 if the parent process is "1".

Fixes tests/rump/rumpkernel/t_sp, which was failing since the
abovementioned special process change due to attempting to deliver a
signal to the special process and the special process was not equipped
to handle one.

(pooka)

Use .PATH because the newmake does not apply pattern rules to sources in
other directories. Leave the comment and the old code for now so people
are aware that this might not work.

(christos)

Fix date.

(wiz)

Add a man page for arcofi(4) driver.  From OpenBSD.

(tsutsui)

don't return after exit.

(christos)

change to no-deprecated-declarations

(christos)

add missing lint library.
XXX: pullup-7

(christos)

lso translate errno in the rumpfiber version

(justin)

Fix error in previous.  The sed command really wants one long line.

(apb)

Add errno translation for bio operations

(justin)

Use the right BAR for MMIO register sizing on i810.

(riastradh)

Correct the mmc clock.  Banana Pi can now find an SD card.

(skrll)

Set the WM_F_ATTACHED flag if wm_attach() finished succesfully and check
the flag in wm_detach(). It will avoid to panic in wm_detach().
Fixes PR#49102.

(msaitoh)

npf_config_submit: finally, include the saved connections.

(rmind)

- npf_conn_import: add a missing stat counter increment.
- npf_nat_import: add a missing reference and make a comment.

(rmind)

Add note about the config rebuild.

(joerg)

Revert last, bump required config version instead.

(joerg)

Bump CONFIG_VERSION.

(joerg)

Initialize IOMMU control and status register properly - from OpenBSD - ok mrg@

(palle)

Generalize the code to allow for arbitrary interrupt wirings - with this change the serial console is usable on Sun Fire V445 systems - from OpenBSD - ok mrg@ martin@

(palle)

Avoid splitting sed programs over multiple lines with
<backslash><newline><indentation>.  Use multiple -e options instead.

(apb)

Let "make clean" delete the files and dirs created by some of the tests.

(apb)

Assume anything not vax has IEEEFP.

(matt)

In either a variable assignmentor a command, backslash-backslash-newline
does not escape the newline.  This is compatible with gmake.

(apb)

Add entry for kern/kern_ksyms_buf.c

(matt)

Depending on whitespace eating behavior is non portable (our make was
not compliant until today -- it ate all whitespace). Fix commands so
that they don't depend on that.

(christos)

Our practice is that an even number of backslashes before a newline
in a variable assignment simply stores the backslashes as part of the
value, and treats the newline as though it was not escaped.  This
is compatible with GNU make.

(apb)

Our practice, despite what POSIX might say, is that "\#"
in a variable assignment stores "#" as part of the value.
The "\" is not taken literally, and the "#" does not begin a comment.

(apb)

Convert the variable printing tests to use .USE rules,
to ensure that the value printed always comes from the expected variable.

(apb)

RCS Id police (because we can. or have you heard of git id police?)

(pooka)

* attempt to clarify what "POSIX" means
* link to Platforms page on wiki for a list of known implementation

(pooka)

Adjust tests to correct for the fact that end-of-line comments
in variable assignments should not be included in the value.

Also fix more typos.

(apb)

Fix more typos in variable names.

(apb)

Fix typos in variable names, and in a comment.

(apb)

Ensure nbytes > 0. Otherwise bad things may happen.
Compile-tested only.

ok christos@

(maxv)

pull in opt_arm_debug.h for VERBOSE_INIT_ARM

(jmcneill)

use RAW_PART to trigger read of disklabel, other units might by busy due
to discovered wedges.

(mlelstv)

Add tests for backslash escaping in make(1).

Many of these tests fail, and I have populated the escape.exp file with
the results that I expect, not with the results that make(1) actually
produces.

Also update the set lists for these tests.

(apb)

sort

(apb)

fix typo. (s/MODULE_DEFAULT_AUTOLOAD/MODULAR_DEFAULT_AUTOLOAD/)

(nonaka)

Pull up following revision(s) (requested by tsutsui in ticket #53):
sys/arch/sun3/dev/fd.c: revision 1.78
sys/arch/sparc/dev/fd.c: revision 1.155
sys/arch/sparc64/dev/fdc.c: revision 1.42
Fix panic() on opening fd(4), caused by a wrong pointer passed to memset().

(martin)

Ticket 53

(martin)

Tickets 51 and 52

(martin)

Pull up following revision(s) (requested by manu in ticket #52):
lib/libperfuse/perfuse_priv.h: revision 1.33
lib/libperfuse/ops.c: revision 1.67
lib/libpuffs/dispatcher.c: revision 1.47
lib/libperfuse/perfuse_priv.h: revision 1.34
lib/libperfuse/ops.c: revision 1.68
lib/libperfuse/ops.c: revision 1.69
lib/libpuffs/puffs_ops.3: revision 1.37
lib/libpuffs/puffs_ops.3: revision 1.38
lib/libperfuse/perfuse.c: revision 1.32
lib/libperfuse/perfuse.c: revision 1.33
usr.sbin/perfused/msg.c: revision 1.22
lib/libpuffs/puffs.h: revision 1.125
Implement FUSE direct I/O.
Remove useless code and warnings

(martin)

Pull up following revision(s) (requested by maxv in ticket #51):
sys/netinet6/ip6_output.c: revision 1.158
sys/rump/librump/rumpvfs/rumpfs.c: revision 1.130
Fix memory leaks in error cases

(martin)

Add new arcofi(4) audio driver for NetBSD/hp300, ported from OpenBSD.

The arcofi(4) is a driver for the HP "Audio1" device
(Siemens PSB 2160 "ARCOFI" phone quality audio chip)
found on the HP9000/425e and HP9000/{705,710,745,747} models
(but only hp300 attachment is ported for now).
The chip supports 8-bit mono 8kHz U-law, A-law and
16-bit mono slinear_be formats.

The old HP9000/425e playing tunes with this new arcofi(4) audio driver
was also demonstrated at Open Source Conference 2014 Shimane.

(tsutsui)

mention MODULAR_DEFAULT_AUTOLOAD

(jnemeth)

Create an "options MODULAR_DEFAULT_AUTOLOAD" config option and add
it to all kernel configs that contain "options MODULAR".  This
option turns on module autoloading by default (which is the current
default).  This allows people who don't want module autoloading on
by default to disable it by simply removing/commentting this line.

(jnemeth)

fix the sources list to match the upstream Makefile.am.

fixes problems reported by jdbaker.

XXX: pullup-7.

(mrg)

Fix a few logical vs. physical port number confusions to make this
driver work with two port variants.

(martin)

Add missing COMPAT_{50,60,70} lines, commented out or not as per the
COMPAT_40 line.

(dholland)

Add options COMPAT_44, which controls mostly MD code but should be in
here anyway.

(dholland)

Systematize (and in many cases, fix) the comments on options COMPAT_NN.

There are quite a few configs that are missing some COMPAT_NN options
in ways that don't make sense; this should probably get cleaned up
too, but for the time being I've not added or removed anything.

(dholland)

Outdated, kill.

(joerg)

Typo.

(joerg)

Remove commented-out debug code that was accidentally committed yesterday.

(apb)

fix output based on new sources

(christos)

not needed anymore.

(christos)

new make tests

(christos)

Instead of adding the test in 48095 I added the tests in 48096. Add the
missing test now.

(christos)

undo previous bogus merge of PR49085

(christos)

PR/46096: Jarmo Jaakkola: fix many problems with dependencies (PR 49086)

Quite extensive rewrite of the Suff module.  Some ripple effects into
Parse and Targ modules too.

Dependency searches in general were made to honor explicit rules so
implicit and explicit sources are no longer applied on targets that
do not invoke a transformation rule.

Archive member dependency search was rewritten.  Explicit rules now
work properly and $(.TARGET) is set correctly.  POSIX semantics for
lib(member.o) and .s1.a rules are supported.

.SUFFIXES list maintenance was rewritten so that scanning of existing
rules works when suffixes are added and that clearing the suffix list
removes single suffix rules too.  Transformation rule nodes are now
mixed with regular nodes so they are available as regular targets too
if needed (especially after the known suffixes are cleared).

The .NULL target was documented in the manual page, especially to
warn against using it when a single suffix rule would work.
A deprecation warning was also added to the manual and make also
warns the user if it encounters .NULL.

Search for suffix rules no longer allows the explicit dependencies
to override the selected transformation rule.  A check is made in
the search that the transformation that would be tried does not
already exist in the chain.  This prevents getting stuck in an infinite
loop under specific circumstances.  Local variables are now set
before node's children are expanded so dynamic sources work in
multi-stage transformations.  Make_HandleUse() no longer expands
the added children for transformation nodes, preventing triple
expansion and allowing the Suff module to properly postpone their
expansion until proper values are set for the local variables.

Directory prefix is no longer removed from $(.PREFIX) if the target
is found via directory search.

The last rule defined is now used instead of the first one (POSIX
requirement) in case a rule is defined multiple times.  Everything
defined in the first instance is undone, but things added "globally"
are honored.  To implement this, each node tracks attribute bits
which have been set by special targets (global) instead of special
sources (local).  They also track dependencies that were added by
a rule with commands (local) instead of rule with no commands (global).

New attribute, OP_FROM_SYS_MK is introduced.  It is set on all targets
found in system makefiles so that they are not eligible to become
the main target.  We cannot just set OP_NOTMAIN because it is one of
the attributes inherited from transformation and .USE rules and would
make any eligible target that uses a built-in inference rule ineligible.

The $(.IMPSRC) local variable now works like in gmake: it is set to
the first prerequisite for explicit rules.  For implicit rules it
is still the implied source.

The manual page is improved regarding the fixed features.  Test cases
for the fixed problems are added.

Other improvements in the Suff module include:
  - better debug messages for transformation rule search (length of
    the chain is now visualized by indentation)
  - Suff structures are created, destroyed and moved around by a set
    of maintenance functions so their reference counts are easier
    to track (this also gets rid of a lot of code duplication)
  - some unreasonably long functions were split into smaller ones
  - many local variables had their names changed to describe their
    purpose instead of their type

(christos)

Bump date for previous.

(wiz)

Convert test in PR/49085

(christos)

PR/49085: Jarmo Jaakkola: fix several parsing errors

Don't exit from var.c:Var_Parse() before possible modifiers are handled
on D and F modified versions of local variables.  Properly expand $(?D)
and $(?F) too.

Make line continuations in rule's commands POSIX compliant.

Fix the syntax error caused by lib(member) as the last target before
a dependency operator.

Document the line continuation change in the manual page.  Also talk
more about the POSIX style local variables and their modifiers.

Add tests covering the fixed problems into d_posix.mk.  The test is
a known failure at the moment because of PR 49086 and PR 49092.

[XXX: unconverted tests]

(christos)

Another round of nouveau whack-a-mole.

(riastradh)

Make -mabi=elfv2 work for powerpc64

(matt)

Tickets 39, 50.

(riz)

Pull up following revision(s) (requested by nakayama in ticket #50):
sys/dev/isa/ess.c: revision 1.81
sys/dev/isa/ess.c: revision 1.82
sys/dev/isa/essvar.h: revision 1.27
sys/dev/isa/essreg.h: revision 1.17
Remove redundant colon.
Add support for Spatializer, 3D audio effects embedded in ES1869
and ES1879 to ess(4).
Tested on my old laptop, mobio NX.

(riz)

Pull up following revision(s) (requested by martin in ticket #39):
usr.sbin/sysinst/main.c: revision 1.4
usr.sbin/sysinst/disks.c: revision 1.5
usr.sbin/sysinst/main.c: revision 1.5
usr.sbin/sysinst/mbr.c: revision 1.3
usr.sbin/sysinst/net.c: revision 1.3
usr.sbin/sysinst/net.c: revision 1.4
usr.sbin/sysinst/partman.c: revision 1.5
usr.sbin/sysinst/partman.c: revision 1.6
usr.sbin/sysinst/util.c: revision 1.4
Remove duplicate fclose(), pointed out by coverity scan.
Remove (accidently) duplicated code - pointed out by coverity scan.
Make sure the interface name is 0 terminated, pointed out by coverity scan.
Check return value from getenv() before using the value. Pointed out by
coverity scan.
Make an invariant more clear to avoid fals positives from coverity scan
if chdir() fails, try chdir /
Check return values from stat before comparing contents of the result
structures. Pointed out by coverity scan.
Cosmetic changes trying to appease coverity scan.
Make sure to not dereference a NULL pointer.

(riz)

Change powerpc64 to use GCC 4.8

(matt)

#if 0 some code not in use yet

(matt)

Minimize differences between powerpc and powerpc64
Fix powerpc64 assembly to use right instructions for 64bit CPUs.
Use hidden for __curbrk and __minbrk to avoid GOT/TOC relocations.

(matt)

Fix cmpptr/cmpreq/etc to use right mnemonics

(matt)

Add
LDFLAGS+=              -Wl,-m,elf32ppc_nbsd

(matt)

Fix powerpc64 support for NetBSD.
Add __PPC_ISEL__ to indicate cpu support the isel instruction

(matt)

Accept driver names only in the two args form together with a minor number.

That's what the documentation says and also avoids the case where the driver
name converted to a major number is interpreted as a packed dev_t number.

(mlelstv)

In tagq_free(), fix the test checking whether a TAGQ should be removed from
its tailq. This required introducing a new flag in the TAGQ structure to
avoid reverting to poking under the hood of the queue.h API.

The concrete bug it solves is that using tags would make vi crash reliably
on exit.

(aymeric)

Rework to be avoid a conditional assignment which results in:
sbin/ping6/ping6.c:1879:8: error: 'cbit.skip' may be used uninitialized in this function [-Werror=maybe-uninitialized]
    cur += ntohs(cbit.skip) * 32;

(matt)

Fix what is almost certianly an off-by-one error when configuring pads on
AM335x for BPP greater than 16.

(jakllsch)

Support WSDISPLAYIO_GET_FBINFO ioctl in tifb(4).

(jakllsch)

tifb can support 32 and 24 BPP framebuffer modes; pull in corresponding raspos
support.

(jakllsch)

Ticket 49

(martin)

Pull up following revision(s) (requested by hannken in ticket #49):
sys/ufs/ext2fs/ext2fs_vfsops.c: revision 1.184
Use mount from argument "mp", "vp->v_mount" is not valid here.
PR kern/49142 (panic in ext2fs_loadvnode mounting an ext2fs filesystem)

(martin)

Remove enclosing parens on return.

(matt)

Fix failure case in rb_tree_find_node_leq/geq.

Return NULL, not `NULL - offset'.

XXX pullup to netbsd-5, netbsd-6, netbsd-7

(riastradh)

Use mount from argument "mp", "vp->v_mount" is not valid here.

PR kern/49142 (panic in ext2fs_loadvnode mounting an ext2fs filesystem)

Needs pullup to -7

(hannken)

Test make(1) by running the maintained tests, not unmaintained
copies of them.

* Remove all old tests from src/tests/usr.bin/make/d_*.  These tests
  were unmaintained old copies of the actual tests which are maintained
  under src/usr.bin/make/unit-tests.  One exception is the test in
  d_unmatchedvarparen.mk, which was new, but has nw been added to
  src/usr.bin/make/unit-tests/varmisc.mk.
* In src/tests/usr.bin/make/Makefile, copy all
  the tests from src/usr.bin/make/unit-tests to
  ${DESTDIR}/usr/tests/usr.bin/make/unit-tests.
* In src/tests/usr.bin/make/t_make.sh, run the tests installed above,
  instead of the old tests.
* In etc/mtree/NetBSD.dist.tests, create the
  usr/tests/usr.bin/make/unit-tests diectory.
* Update the set lists for all the above.

(apb)

add powerpc64 support

(matt)

Revert to mapping the whole GTTMMADR region all at once for now.

This was split up so we could map the GTT prefetchable without
interfering with the MMIO registers, but it interferes with old drm
which tries to map the whole region and which is still hooked up for
the moment.

Once we unhook old drm for good we can put this back.

(riastradh)

Add a compile-time selector for I/O space operations.  Needs more work
some day, but allows virtio drivers to work today.

(pooka)

Nuke the DOMAINADD() macro and just call domain_attach(), now that things
work correctly that way.

(pooka)

Change the meaning of the boolean parameter to domaininit().
was: attach route domain if found from domains link set
now: attach any domain which is found from domains link set

No functional change to monolithic kernel.  In a rump kernel, which
attaches networking domains "modularly", prevents domains from
accidentally leaking in via the link set in static linked scenarios.

Problem reported by cube, Tom Bousso, and the voices in my head.

(pooka)