remove unused variable

(christos)

remove unused code

(christos)

mark variables __diagused

(christos)

mark a variable __diagused

(christos)

remove unused code

(christos)

use __USE for debug variable

(christos)

use __USE

(christos)

add __USE() for unused volatile variables, so that we generate the same code.

(christos)

fix incorrect sizes to memset

(christos)

Add missing braces.
Fix logic error - check the data read from the chip against 0xff.

(jdc)

PR/48366 Always zero the allocated memory in xhci_alloc, not just in
DIAGNOSTIC.

(skrll)

Use the term "Lua modules" instead of "Lua libraries", list the modules
available.

(mbalmer)

- remove set but unused variables
- move some variables inside their relevant use #ifdef

(mrg)

- remove set but unused variables

(mrg)

gcc 4.8.1 wants:
- remove some set but unused variables
- move some variables inside their usage #ifdef's
- cast to (void) instead of using a dummy return variable

(there are more to come, but this one is now tested with gcc 4.5)

(mrg)

- apply some __diagused
- remove unused variables
- move some variables inside their relevant use #ifdef

(mrg)

let progressmeter deal with the timeout once we've started transferring.

(christos)

cleanvnode():
- VC_XLOCK/VC_MASK are not used anymore, remove.
- If we get a reference while cleaning, there is no need to retry as
  these reference and this vnode will disappear soon.
- Make sure we run inside a fstrans transaction to prevent deadlocks
  against vget().

vrecycle():
- don't even try to recycle a vnode currently cleaning.

(hannken)

use __USE() in the iterator variable

(christos)

move unused variable inside if 0

(christos)

more unused inside 0

(christos)

move unused code inside if 0

(christos)

sprinkle vax

(christos)

make a gcc unitialized variable

(christos)

mark some variables as unused, but keep them around as comments

(christos)

fix old style decls

(christos)

one more debuglib

(christos)

Fix debuglib entries

(christos)

_g libraries don't belong here.

(christos)

PR/48098: Brian Marcotte: panic: kernel diagnostic assertion "cred != NULL":
Fix from Michael van Elst, tcpdrop crashes kernel on ebryonic connections.

(christos)

Tag liblto_plugin.so.0.0.debug with gcc=48.

(jmmv)

PR/34796: Hauke Fath: ftp does not timeout on http fetches.

(christos)

mark unused function as used.

(christos)

these can never be null.

(christos)

Turn the "esiop_intr: I shouldn't be there !" panic into a printf
followed by a return from the interrupt handler.  The condition is
triggered on some KVM virtual hosts, apparently harmlessly, and not
panicing makes it possible to boot a NetBSD GENERIC kernel on those
hosts.  Also make the same change to siop to minimize divergence
between siop and esiop.  Fixes PR kern/48277.

(gson)

Stop using v_mount of an unreferenced vnode -- save the mount while
the vnode has a reference.

(hannken)

put attributes in the right spot

(christos)

more sets fixes

(christos)

more sets fixes (libgomp, libobjc)

(christos)

more gcc-4.8 files

(christos)

sprinkle __diagused

(christos)

sprinke __diagused

(christos)

Workaround a gcc -Wuninitilized whine

github/buildrump.sh issue #44

(pooka)

go back to using file:line: from uwe.

(christos)

Remove the bogus CPPFLAGS+="-I$S/../../../include"

Prompted by the build problem reported by John Klos

(skrll)

fix broken volatile assignment that gcc-4.8.1 does not like

(christos)

make the code compilable with gcc-4.8.1; this loop has an array overflow,
but it is only used for sparc register windows; ifdef it sparc for now,
it's the easiest fix.

(christos)

pick up omp.h from the libgomp directory not from /usr/include/gcc-4.5. Makes
the code compilable with gcc-4.8.1

(christos)

make config errors look more like other programs: <file>,<line>:
instead of <file>:<line>:

(christos)

fix type in comparison

(christos)

ignore all -std=gnu patterns not just 99

(christos)

PR/48312: Dieter Roelands: According to TOG, unset should not return an error
for functions are variables that were not previously set:
http://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html

(christos)

tmpfs: fix the zero-length symlink target case as NetBSD supports them.

(rmind)

Style.

(skrll)

Don't always define DWC2_DEBUG

(skrll)

DOTG is dead. Long live DWC2.

(skrll)

Clean up (mostly remove unneeded) #includes

(skrll)

PR/48342. Add aurateconv as a dependency for vcaudio.

(skrll)

HOST_SH is defined in bsd.sys.mk, so we need to include that
as well as bsd.own.mk.  Revert previous use of ${HOST_SH:Ush}.

(apb)

If HOST_SH is not defined, use sh to parse osrelease.sh

(martin)

rfc3720 Appendix D. SendTargets Operation requires that requesting
information for all relevant targets is done with the value 'All'
(upper case A) and some implementations require it.

(mlelstv)

Document that patterns are specified using extended regular expressions.

(apb)

add support for GCC 4.8 sets.

(mrg)

add some NetBSD_DISABLED_* rules.

(mrg)

move variable definition/set inside the same #ifdef of the usage.

(mrg)

apply __diagused where appropriate.

(mrg)

sprinkle some __diagused where appropriate.

(mrg)

Reinstate .include <bsd.own.mk>.  Without this, many variables are undefined.

Define DISTRIBVER in the same way that src/etc/Makefile defines it.
Without this, "make params" in src/Makefile uses an undefined value.

Remove a spurious ">&3".

(apb)

provide an ldscript for booters that strips the unwind sections.
XXX: untested

(christos)

provide an ldscript to strip the unwind sections
XXX: untested

(christos)

Handle new objc for gcc-4.8.1. Someone with more objc foo can do this better.

(christos)

avoid pointer aliasing problems.

(christos)

Handle more than gcc 4.5

(christos)

fix build on arm

(mbalmer)

Add a reference to secmodel_extensions(9), to make it a little easier to
figure out how to enable mounts for unprivileged users.  (Why don't we
just explain vfs.generic.usermount in the mount(8) man page?)

Also add another example of different ways in which the "special"
argument is interpreted by different file systems: mount_tmpfs(8)
ignores it.

(apb)

Add 40bit DMA bug workaround(BGEF_40BIT_BUG) from FreeBSD (part of r199670).
This workaround is for 5714/5715 controllers and is not actually a MAC
controller bug but an issue with the embedded PCIe to PCI-X bridge in the
device.
This change uses bus_dmatag_subregion(), so this workaround won't work on some
archs which doesn't support bus_dmatag_subregion().

(msaitoh)

Cleanup flag setting. No functional change.
- Split flags into bge_flags and bge_phy_flags.
- Rename flags. PHY related flags are prefixed with BGEPHYF_*.
  Other flags are prefixed with BGEF_*.

(msaitoh)

create mkdep output with a temporary file so that failures in
complation during .d creation don't leave an empty (stale) .d
file around to break future builds.

(hmm, probably could use .PRECIOUS here.)

(mrg)

tmpfs_alloc_node: it is less error-prone to store the link path with
the NIL terminator included.  Adjust tmpfs_readlink() to exclude NIL.
Also, remove the check for zero-length and add some asserts.

(rmind)

Try improving the description.

(wiz)

Preserve NetBSD version from the ABI tag for compat use.

(joerg)

Update LLVM/Clang snapshot to r193633. Fixes an old and somewhat amusing
bug involving a source file foo.c, a function foo and a static
function-local variable c in the direct object writer.

(joerg)

play the addref/delref game on suspend, prevents crash if the disk/CF Card
is eg. in a PCMCIA adapter and not mounted

(drochner)

-recognize CF cards by the magic value in inquiry data
-kill CFG_ATAPI_MASK, didn't see anything in the specs supporting
that it exists

(drochner)

In "make sortlists", if the file was already sorted then don't
change its timestamp.

(apb)

Instead of stupid postprocessor tricks, use redirect tricks
to ensure that the output from ${PRINT_PARAMS} contains only
the desired information without any debugging noise.

(apb)

Ignore SIGPIPE; this is silly, we are getting EOF from a client, then while
processing the error SIGPIPE comes because the client closed the connection,
and then we end up calling error() while processing the error which prints
silly messages on the console because it logs with LOG_EMERG.

(christos)

Oh, stdint.h isn't written with a comma ...

(pooka)

need to include stdint.h before rump/rumpuser.h

(pooka)

glibc on MIPS uses the same DYN d_ptr variant as NetBSD and musl.

from Justin Cormack

(pooka)

mark rump's CPU_INFO_ITERATOR as being __unused.

(mrg)

avoid uninitialised variable use.

(mrg)

ensure variables are initialised

(mrg)

used __diagused where appropriate.

(mrg)

Fix incorrect function names in panic message strings.

(gson)

remove unused variables

(mrg)

#ifdef a variable decl/setting with it's use.

(mrg)

use __diagused where appropriate.

(mrg)

G/C struct user *proc0paddr.

(uebayasi)

Remove extra period.

(msaitoh)

add ./usr/include/gcc-4.8 directory to the build.

(mrg)

add lua(4) related modules

(mbalmer)

include sys/condvar.h

(mbalmer)

initialize variables

(mbalmer)

npf_session_setnat: fix the race condition when the old connection is still
being expired while a new/duplicate is being created.

(rmind)

Add MIPS.  Requested by Justin Cormack.

(pooka)

Maintain arch list in a separate variable.
no functional change

(pooka)

Unbreak src/tools usage of C++.

(joerg)

Mark usage as dead.

(joerg)

Remove bogus cast. The LHS type is smaller, so let the compiler decide
on the correct (unsigned type).

(joerg)

PR/48351: Dennis Ferguson: Fix incorrect parsing of flock flags.
XXX: still flock -s 0 fails with EINVAL, why?

(christos)

Small typo. From Michael W. Bombardieri.

(njoly)

Note #978 was via patch

(sborrill)

Ticket #1888

(sborrill)

Pull up the following revisions(s) (requested by bad in ticket #1888):
sys/ufs/ffs/ffs_alloc.c: revision 1.144 via patch

Pull in fix from FreeBSD ffs_alloc.c r121785:
Consider only cylinder groups with at least 75% of the average free space
per cylinder group and 75% of the average free inodes per cylinder group
as candidates for the creation of a new directory.  Avoids excessive I/O
scanning for a suitable cylinder group on relatively full file systems.

(sborrill)

Vnode API cleanup pass 1.

- Make these defines and functions private to vfs_vnode.c:

  VC_MASK, VC_LOCK, DOCLOSE, VI_IANCTREDO and VI_INACTNOW
  vclean() and vrelel()

- Remove the long time unused lwp argument from vrecycle().

- Remove vtryget(), it is responsible for ugly hacks and doesn't
  look that effective.

Presented on tech-kern.

Welcome to 6.99.25

(hannken)

Ticket #978

(sborrill)

Pull up the following revisions(s) (requested by bad in ticket #978):
sys/ufs/ffs/ffs_alloc.c: revision 1.144

Pull in fix from FreeBSD ffs_alloc.c r121785:
Consider only cylinder groups with at least 75% of the average free space
per cylinder group and 75% of the average free inodes per cylinder group
as candidates for the creation of a new directory.  Avoids excessive I/O
scanning for a suitable cylinder group on relatively full file systems.

(sborrill)

add pmf(9lua) man page

(mbalmer)

check parameters

(mbalmer)

add a core(9lua) man page

(mbalmer)

Don't endian-twiddle solid source colours in composite operations on R1xx
Now gtk2-rendered text looks right again with EXA and xrender acceleration
enabled.
Probably needs similar fixes for R2xx, also needs testing on little-endian
hardware.

(macallan)

Add AARCH64 definitions

(matt)

Add EM_AARCH64 from Elliott Hughes

(christos)

Add the if_axen kmod to md.evbppc to unbreak the evbppc build.

(riz)

Instead of guessing the suffix in the code, use the suffix list previously
loaded via man.conf(5). While there, zap unused iteration code.
(Franco Fichtner)

(christos)

Make sure certain files are always assembled with -marm

(matt)

Add support for the BCM56340 iProc based switch

(matt)

Make these compatible with thumb/thumb2

(matt)

Support for the Broadcom BCM56340 iProc based switch.

(matt)

Adding luactl.debug to lists to fix debug build.

(htodd)

add an alias for the linux name for the interface index

(christos)

Add DF_* values for DT_FLAGS

(matt)

Pull in fix from FreeBSD ffs_alloc.c r121785:
Consider only cylinder groups with at least 75% of the average free space
per cylinder group and 75% of the average free inodes per cylinder group
as candidates for the creation of a new directory.  Avoids excessive I/O
scanning for a suitable cylinder group on relatively full file systems.

Tested by sborril and me.

Pullup: netbsd-6, netbsd-5

Original commit message:

Tweak the calculation of minbfree in ffs_dirpref() so that only
those cylinder groups that have at least 75% of the average free
space per cylinder group for that file system are considered as
candidates for the creation of a new directory.  The previous formula
for minbfree would set it to zero if the file system was more than
75% full, which allowed cylinder groups with no free space at all
to be chosen as candidates for directory creation, which resulted
in an expensive search for free blocks for each file that was
subsequently created in that directory.

Modify the calculation of minifree in the same way.

Decrease maxcontigdirs as the file system fills to decrease the
likelyhood that a cluster of directories will overflow the available
space in a cylinder group.

Reviewed by: mckusick
Tested by: kmarx@vicor.com
MFC after: 2 weeks

(bad)

linke pmf(9l) to the build

(mbalmer)

Initial Lua binding for pmf(9)

(mbalmer)

link core(9l) (docs pending) to the build

(mbalmer)

link luactl(8) to the build

(mbalmer)

link the lua(4) module to the build

(mbalmer)

create a lua device node for lua(4) and luactl(8)

(mbalmer)

Add support for scratchpad buffers (required for some XHCI devices).
Fix an endian issue.

(matt)

Add an explicit initialization of .new_device = NULL

(matt)

Fix name of sysctl setup function.

(rkujawa)

Support prefetch size.

(msaitoh)

Set maximum read byte count to 2048 for PCI-X BCM5703/5704 devices.
For PCI-X BCM5704, set maximum outstanding split transactions to 0.
Same as Linux tg3 and FreeBSD (part of FreeBSD r204978).

(msaitoh)

add MKDEPCXX, that is used for C++ files depend creation.  also
add some missing ${COPTS.${.IMPSRC:T}} and ${CPUFLAGS.${.IMPSRC:T}}
to the dependancy rules.

this makes it possible to have C++ files .d files created with the
C++ compiler, and with the normal options for this file, which
means that we can build libstdc++-v3 from gcc 4.8.

(mrg)

convert to ANSI C definitions.

(mrg)

Enable holding implicit threads over explicit scheduling points (i.e.
a kernel driver calling rump_lwproc_switch()).  Also, correctly handle
curcpu()->ci_curlwp and biglock there.

(pooka)

No more walk on the wild side.  RIP Lou Reed.

(mbalmer)

regen usbdevs

(rmind)

usbdevs: add an entry for external WD disk.

(rmind)

librumpuser: add some #ifdef __APPLE__ and missing bits to build it on OS X.
OK pooka@

(rmind)

- fpu_state_save() on amd64: ci->ci_fpsaving is no longer used, simplify.
- Remove some unnecessary header inclusions.  Misc clean up while here.

(rmind)

Add NPF_MAX_RULES, an artificial limit (set it to 1M).

(rmind)

vttodt_tab: index in C99 way, makes it less error-prone.

(rmind)

Note some BPF and NPF changes.

(rmind)

Give the hypercall interface a version number.  This is for
implementations outside of the tree.

(pooka)

add a sqlite(3lua) manual page

(mbalmer)

More user friendly sqlite.open() function.
Add a few more symbols.
Register all function in the sqlite table.

(mbalmer)

Add man mage links for __debugused(3) and __diagused(3).

(apb)

missing word

(mbalmer)

Fix date and remove superfluous comma.

(wiz)

Remove trailing whitespace.

(wiz)

vfs.wapbl.flush_disk_cache is more about metadata than data, and
more about safety in the event of power loss than safety in the
event of a crash.

(apb)

Fix memory leak in gensub()

(mlelstv)

refer to __USE(3)

(apb)

Add "make sortlists", for sorting the set lists.  This may be
run manually, before committing changes to the lists files.

(apb)

sort

(apb)

grep-2.15 out.

(wiz)

lua.h is not imaginary anymore

(mbalmer)

install lua.h

(mbalmer)

Revert mi because the added line refers to an imaginary file.

(htodd)

Note that NetBSD/x68k now supports upto 1008 MB extended memory found on XM6i.

(tsutsui)

Probe extended memories before pmap_boostrap() for actual fix of PR/45915.

Also expand expected maximum extended memory regions to recognize
upto 1008 MB RAM that is supported by the XM6i emulator.

See my post on port-x68k@ for more details:
http://mail-index.NetBSD.org/port-x68k/2013/10/19/msg000039.html

Probably worth to pullup to netbsd-6.

(tsutsui)