support DDC2, so far only for WSDISPLAYIO_GET_EDID

(macallan)

Print out the address in the ktlbmod panic.

(matt)

Add workaround for RMI cpus (not really needed but good for completeness)

(matt)

put back IPL_AUDIO and splaudio(), at the request of rmind.  they are
aliases and now we can easily revert audio to IPL_VM if necessary,
without having to revert the whole branch.

(mrg)

add MONITOR_DATA register

(macallan)

add midi_pcppi.c to the list.  sort.  midi @ pcppi now works.

(mrg)

set hwif_softc around midi_attach()s call to ->getinfo(), like the old
code, so that midi_register_hw_if_ext() works.  add a mutex for this
to avoid multiple midi_attach()s fighting over it.

XXX: this is an ugly interface

(mrg)

Documentation fix

Evgeniy Ivanov <lolkaantimat@gmail.com>

(manu)

Remove files specified in pcn.

When remove files using name from pnode, another link on this file
can be unlinked. E.g. "touch 1; ln 1 2; rm 2" will remove file named
"1". Thus puffs_null_node_remove should remove directory entry which
name is provided by pcn (as said in puffs_ops.3). Caller should
provide appropriately initialized pcn.

From Evgeniy Ivanov <lolkaantimat@gmail.com>

(manu)

Set eofflag in puffs_fuse_node_readdir() from librefuse.

From Evgeniy Ivanov <lolkaantimat@gmail.com>

(manu)

Remove pnode of overwritten file.

When puffs_null_node_rename() overwrites existing file, its pnode
must�be removed, because src pnode already represents this file.

From Evgeniy Ivanov <lolkaantimat@gmail.com>

(manu)

get rid of a variable which is no longer used after the demise of splaudio

(agc)

add t_subr_prf

(christos)

Deep rework of the xbdback(4) driver; it now uses a thread per instance
instead of continuations directly from shm callbacks or interrupt
handlers. The whole CPS design remains but is adapted to cope with
a thread model.

This patch allows scheduling away I/O requests of domains that behave
abnormally, or even destroy them if there is a need to (without thrashing
dom0 with lots of error messages at IPL_BIO).

I took this opportunity to make the driver MPSAFE, so multiple instances
can run concurrently. Moved from home-grown pool(9) queues to
pool_cache(9), and rework the callback mechanism so that it delegates
I/O processing to thread instead of handling it itself through the
continuation trampoline.

This one fixes the potential DoS many have seen in a dom0 when trying to
suspend a NetBSD domU with a corrupted I/O ring.

Benchmarks (build.sh release runs and bonnie++) do not show any
performance regression, the "new" driver is on-par with the "old" one.

ok bouyer@.

(jym)

Add a test for the kernel snprintf.

(christos)

- fixed signed/unsigned comparison
- don't write/increment sbuf if NULL

(christos)

- don't let arguments in macros have side effects
- nul terminate properly

(christos)

splaudio is no more

(jmcneill)

audio: Audio drivers are now MP-safe.

(jmcneill)

import openresolv-3.4.5

(roy)

sync

(roy)

Import openresolv-3.4.5 with the following changes since the last version:
* More printf portabitiy fixes.
* Use read -r to avoid backslash problems.
* If we have a valid domain, put that in resolv.conf as well as search.
  This does not fix a technical problem, just stops me getting bug reports.
* Update metric and privacy even if resolv.conf didn't change.
* sortlist is now supported.
* Ensure subscriber config directories exist before writing the configs
* Don't create pdnsd.conf if it doesn't exist or is not writeable.

Status:

Vendor Tag: roy
Release Tags: openresolv-3-4-5

(roy)

note umidi as done

(mrg)

complete the port to audiomp:  take kernel lock in a few places for
USB (like uaudio), kill most of the spl* uses.

(mrg)

jmcneill-audiomp3 is done

(jmcneill)

update audio(9) docs for audiomp changes

(jmcneill)

Merge jmcneill-audiomp3 branch, which is derived from ad-audiomp2. From
the original ad-audiomp branch notes:

  Add MP locking to the audio drivers.

  Making the audio drivers MP safe is necessary before efforts
  can be made to make the VM system MP safe.

  The are two locks per device instance, an ISR lock and
  a character device lock. The ISR lock replaces calls to
  splaudio()/splx(), and will be held across calls to device
  methods which were called at splaudio() before (e.g.
  trigger_output). The character device lock is held across
  calls to nearly all of the methods, excluding some only
  used for initialization, e.g. get_locks.

Welcome to 5.99.57.

(jmcneill)

terminate ad-audiomp2, add jmcneill-audiomp3

(jmcneill)

Make it compile without DEBUG option.

(phx)

If ufs_balloc_range() fails, make sure to call ?fs_truncate() to
reset v_writesize to the right value.
If v_writesize is left larger than the allocated blocks, we may have
the same issue as the one described in
http://mail-index.netbsd.org/tech-kern/2010/02/02/msg007156.html

(bouyer)

Add mips64*-netbsd* clauses

(matt)

According to "Virtio PCI Card Specification v0.9.2 DRAFT" there is no
feature named VIRTIO_BLK_F_SECTOR_MAX so remove it.  Linux seems to use
this feature bit as VIRTIO_BLK_F_TOPOLOGY.

Use VIRTIO_BLK_F_BLK_SIZE * VIRTIO_BLK_F_SEG_MAX as the drivers maxxfer
and reorder so sc_secsize gets set before use.

As maxxfer may not be a multiple of page size add one more segment to
the dma maps.

Tested on Linux 3.1.1 host by Guillaume Lasmayous.

(hannken)

Use hashinit() / hashdone() to create the union node hash list.

Cleanup the hash lookup in union_allocvp().

Needs more work as there is still a possible deadlock between
union_allocvp() and vclean().

(hannken)

use ARG_CHAR_T instead of CHAR_T for integer promotion.

(tnozaki)

use e_key_t instead of u_int.

(tnozaki)

wss(4) and clcs(4) tested successfully

(jakllsch)

audiocs at ebus works on sparc64

(macallan)

initialize locks
now this works on my U60

(macallan)

recompile.

(nonaka)

modify for PSION NETBOOK PRO.
- ignore "Network" directory when finding the directory for configuration files.
- as the device name "SMC" (SD-MMC) to allow.
- add "SMC1:/" entry to kernel directory.

(nonaka)

Hold the intr lock for the duration of ad1848_isa_intr().

(jakllsch)

don't use WEOF directly, for --disable-widechar.

(tnozaki)

don't use L prefix directly, for --disable-widechar.

(tnozaki)

Sort option descriptions, bump date for previous.

(wiz)

Load entropy at system boot (only works at securelevel < 1); save
at system shutdown.  Disable with random_seed=NO in rc.conf if desired.

Goes to some trouble to never load or save to network filesystems.

Entropy should really be loaded by the boot loader but I am still
sorting out how to pass it to the kernel.

(tls)

minor formatting changes - no functional change

(agc)

No more users of xpmap_update(). Use pmap_pte_*() functions now.

(jym)

Kill dependency to xpmap_update(), and use setbits/clearbits to update
kernel PTE rights.

(jym)

When allocating a page for a kernel stack and PMAP_ALLOC_POOLPAGE is
defined, use it.  (allows a MIPS N32 kernel to boot when there is memory
outside of KSEG0).

(matt)

When allocating pages for kernel map entries and PMAP_ALLOC_POOLPAGE is
defined, use it.  (allows a MIPS N32 kernel to boot when there is memory
outside of KSEG0).

(matt)

Move Xen-specific functions to Xen pmap. Requested by cherry@.

Un'ifdef XEN in xen_pmap.c, it is always defined there.

(jym)

get rid of unused 'powerstate' callback in audio_hw_if

(jmcneill)

- make decimal conversions use the maximum width integers available on the
  architecture.
- make signed and unsigned code consistent.

(christos)

The rnd pseudo-device is not really optional, because it is in the same
source file as the entropy-pool code itself.  Move it to std.  This
will be cleaned up more when I split the sources up as they should be.

This fixes build breaks on several ports.  Thanks to Havard Eidnes for
pointing them out.

(tls)

Increment the source buffer when we overflow so that we don't get stuck in
an infinite loop.

(christos)

Regression tests for awk(1) (PR 44063)

(cheusov)

Set "mode" to 2 at the correct time.
Fixes recording on some isapnp(4) wss(4) chips.

(jakllsch)

Add support to recognize the 8168E model of this chip.  Taken from FreeBSD

(garbled)

audiocs at sbus works on sparc and sparc64

(macallan)

Use :Q to deal with the case that CLEANFILES or CLEANDIRFILES
contains quoted substrings (such as file names with spaces).
Problem reported by Joseph Koshy, who also provided the
important part of the fix.

(apb)

we need to initialize the locks before using them
Now this works on my Ultra 1

(macallan)

Print "Model:" information with oea_startup().

(phx)

ARM and M68K TLS support

(joerg)

Add TLS support for m68k.

(joerg)

Remove more now unnecessary IIOV() conversion.
XXX: we should have proper PA to VA macro for TT mappings for readability

(tsutsui)

Bump SYMTAB_SPACE so that the symbol table fits again.

(he)

more steps towards making this work:
- avoid using uninitialised memory to get a pointer, this fixes
  hangs in sequencerwrite() and general mayhem when it corrupts
  random other memory.
- re-take sc->lock in sequencewrite() when continuing from the
  main loop.

(mrg)

binutils-2.22 out

(wiz)

IPL_AUDIO is no more

(mrg)

- make this not crash with "midiplay -l" anymore by killing store_thread()
  with kthread_exit() instead of return

- adjust wait channels to be unique

- add some asserts that insert it not called while destroy is in progress

- note some probably dead code with KASSERT(FALSE) for now

(mrg)

adjust DPRINTF*() calls to actually build.
avoid a potential NULl pointer deref.

(mrg)

port umidi to audiomp: get_locks support, take locks in interrupts.

untested, and midi in this branch has general problems still anyway.

(mrg)

adjust some DPRINTF*() calls so that they compile now.

(mrg)

avoid one use-after-free error.  midi_store->lock is unlocked after
we cv_broadcast() to get the midi_store freed.  keep a local copy
of the lock pointer to avoid this.

(mrg)

snapper works

(macallan)

convert snapper
code from mrg, tested on my iBook G4

(macallan)

ym(4) successfully tested lightly on i386

(jakllsch)

don't attempt to obtain the intr mutex again, only KASSERT we own it already

(jakllsch)

dbri works on sparc.

(mrg)

note3 uaudio and hdafg on amd64.

(mrg)

port to audiomp.  mostly from jared, tested and completed by me.

this one is a little more tricky.  besides the usual spl and get_locks
changes, we have to deal with the USB subsystem wanting to sleep and
expecting to be called with the kernel lock held.  so we always take
the kernel lock and drop the interrupt lock before calling into usbd.

(mrg)

awacs works now

(macallan)

adapt to jmcneill-audiomp3
from mrg@, minor fixes and testing by me

(macallan)

Handle simple cases (strlen(charset) <= 1) more efficiently.

(joerg)

I tested autri

(macallan)

Add test cases for strcspn, strpbrk, strspn, wcscspn, wcspbrk and
wcsspn.

(joerg)

NIST_CTR_DRBG.V is accessed as (unsigned long *) so we need to make sure
it's aligned accordingly or we go boom on sparc64

(macallan)

Adjust setup_linux_rt_sigframe/setup_linux_sigframe prototypes.

(njoly)

Make emuxki(4) work w/ audiomp changes.

(jakllsch)

Allow pf flags to be specified in rc.conf.
Add default to defaults/rc.d as suggested by lukem@

(darcy)

- don't clear the flags twice in the loop
- reorder calculation to preserve as much precision as possible.

(christos)

auich(4) successfully tested on amd64 (nForce4)

(jakllsch)

Replace flag based union node locking with generic vnode lock, support
shared and nowait locks and protect un_uppervp and un_*sz with mutex.

Mark file system MPSAFE.

(hannken)

make this compile again. VOP_CLOSE is really vn_close() through macros.

(christos)

gcc 4.5 on vax creates PLT stubs with an entry mask of 0xffc
rather than 0 like gcc 4.1 did, so the sneaky assembly functions
that "ret" without really returning now clobber their registers.
adjust these functions to avoid this problem.

(chs)

New sentence, new line.

(wiz)

Clean up a bit in preparation for more serious changes

(joerg)

Note news68k PROM function based framebuffer console support.

(tsutsui)

Switch from .nf/.fi groff macros to .Bd -literal/.Ed blocks.
While here, convert .sp macro to .Pp.

ok wiz@.

(njoly)

Pass M4=${TOOL_M4:Q} to CONFIGURE_ENV for tools/gmp.
Fixes configure error on cygwin.

(tsutsui)

Include MD <machine/cpu_counter.h> only if defined(__HAVE_CPU_COUNTER).

XXX: Why not timecounter(9) but deprecated cpu_counter32() and microtime(9)?

(tsutsui)

No need to include MD <machine/cpu_counter.h> here.

(tsutsui)

i tested azalia

(jmcneill)

Add missing fstrans_done().

Should fix PR #45635 (KASSERT "fli->fli_trans_cnt == 0" failed)

(hannken)

one more error path that didn't free a lock.

(mlelstv)

nfsrv_lookup(): Defer the postopattr lookup on dirp until the
                child node is unlocked.

Ok: YAMAMOTO Takashi <yamt@netbsd.org>

(hannken)

Fix title and xrefs.

(njoly)

correct declaration of vn_close()
XXX: Should all those dup declarations be here?

(christos)

change printf gcc attribute to __printflike(), requested by joerg.

(christos)

- use va_copy where needed
- always nul terminate vsnprintf()
- make snprintf() call vsnprintf()

(christos)

close file in a way that it does not trigger an assertion :-)
now zpool create works again.

(christos)

add more missing printf attributes.

(christos)

snprintf/vsnprintf can accept NULL/0 buffers and work properly as expected.
Before they used to return incorrect length on short buffers. Remove unused
error path.

(christos)

simplify, no need for va_copy here. Add KASSERT.

(christos)

mark printf function

(christos)

Implement timeouts when blocking the calling process with tsleep.

Fixes PR kern/33452

(gavan)

Add Xorg modules debug entries.

(njoly)

Reshuffle decls among the quota headers so everything is in the place
it should be:
  - stuff for the proplib interface goes in <quota/quotaprop.h>
  - stuff for userlevel only goes in <quota/quota.h>
  - stuff shared between user and kernel goes in <sys/quota.h>

Note that <quota/quota.h> and <quota/quotaprop.h> are expected to be
moved or removed later on... one thing at a time.

Update include directives in other files as needed.

(dholland)

Use va_copy to avoid undefined behaviour in handling the
va_list arg.

(apb)

Add missing CLEANFILES+=x11perfcomp

(dholland)

avoid passing size 0 to kmem_alloc when the sequencer device is opened and
no midi devices are attached

(jmcneill)

improve the last change, after discussing it with rmind

(jmcneill)

Expose pmap_pdp_cache publicly to x86/xen pmap. Provide suspend/resume
callbacks for Xen pmap.

Turn static internal callbacks of pmap_pdp_cache.

XXX the implementation of pool_cache_invalidate(9) is still wrong, and
IMHO this needs fixing before -6. See
http://mail-index.netbsd.org/tech-kern/2011/11/18/msg011924.html

(jym)

i tested sb under qemu

(jmcneill)

don't kpause with intr_lock held

(jmcneill)

revert machdep.c rev. 1.168 because it's likely to have the same problem
as i386's one.
http://mail-index.NetBSD.org/source-changes-d/2011/11/19/msg004283.html

(yamt)

revert dumpsys.c rev. 1.13 because it was reported to cause a problem.
http://mail-index.NetBSD.org/source-changes-d/2011/11/19/msg004283.html

(yamt)

no need to enter intr_lock from trigger_output, it's already held

(jmcneill)

Add an entry for romcons.

(tsutsui)

Create romcons device node.

(tsutsui)

paud is done

(jmcneill)

adapt to audiomp api changes

(jmcneill)

vs is done

(jmcneill)

adapt to audiomp api changes

(jmcneill)

fix size passed to kmem_free in stream filter dtor

(jmcneill)

tms320av110 (melody) is done

(jmcneill)

adapt to audiomp api changes

(jmcneill)

Add preliminary PROM internal function based framebuffer console support,
which was demonstrated at Open Source Conference 2011 Kansai @ Kyoto
back in July:
http://www.NetBSD.org/gallery/events.html#opensourceconf2011-Kansai

- map 0xc0000000-0xffffffff PA region (which is mirror of PA 0x0-0x3fffffff)
  to the same VA via %tt0 and %tt1 registers and move KVA space accordingly
  (like luna68k does for its devices)
- save trap #0 vector for PROM function calls in early bootstrap
  and register it to trap #14 to call it from kernel for console output
- add dumb romcall based tty attachment taken from src/sys/dev/ofw/ofcons.c
- add rom function call stubs from news68k/stand/common/romcalls.S
- remove IIOV() macro for device registers where now mapped PA==VA via %tt1

XXX: romcons is not enabled yet because there is no generic interface
XXX: to attach wskbd(4) to non wsdisplay(4) devices like this romcons.

(tsutsui)

make it compile without DIAGNOSTIC defined

(para)

harmony is done

(jmcneill)

adapt to audiomp api changes

(jmcneill)

aica is done

(jmcneill)

adapt to audiomp api changes

(jmcneill)

vraiu is done

(jmcneill)

adapt to audiomp api changes

(jmcneill)

splaudio -> splvm

(jmcneill)

build.sh -m prep tools failed for me so mark paud_isa "toolchain-fails"

(jmcneill)

port zaudio to audiomp api changes for real this time

(jmcneill)

zaudio is done

(jmcneill)

adapt to audiomp api changes

(jmcneill)

note pxa2x0_ac97 now compiles and should be done

(mrg)

port to audiomp: alloc/free changes, get_locks support, no more IPL_AUDIO,
convert splaudio() -> sc_intr_lock usage, take sc_intr_lock in acu_intr().

(mrg)

Ticket 1437

(bouyer)

Ticket 1437

(bouyer)