mdocml-1.10.8 out.

(wiz)

remove some ancient, unused debug code

(macallan)

Little bit more debug information.

(blymn)

Ignore request to change the values of colour pair 0
Explicitly set foreground and/or background colours when -1 is passed
to assume_default_colors()

(blymn)

Allow attributes to be applied to the default line drawing characters
by just passing attributes as arguments to the wborder call.

(blymn)

Begin cleanup of the make parser: separate out the logic for reading files
from the parsing goo. This will now use mmap for reading if it works and
fall back to read only for pipes and such. Step 1 of a cleanup program
proposed without objection on tech-toolchain.

(dholland)

there is no R128_LITTLE_ENDIAN_EN flag, just lack of R128_HOST_BIG_ENDIAN_EN
Someone(tm) needs to test this on little endian hw

(macallan)

Correctly added the conditional part, but now need to remove the original
unconditional part.

(pgoyette)

Don't include the siginfo tests unless we actually have siginfo ability!

(pgoyette)

Don't descend into siginfo - it's not there any more

(pgoyette)

Use printf format macros for long longs.

(jakllsch)

Correct recently exposed printf format problem.

(jakllsch)

The siginfo tests have migrated to atf in src/tests/lib/libc/gen/

(pgoyette)

Move the siginfo tests from regress to atf.  While here, add a new test
for PR/43655.

XXX The sigchild_dump test currently fails when execute under atf-run.
XXX It does not fail when executed directly from the shell, so there's
XXX something in atf that prevents the child process from dumping.

(pgoyette)

We still need to update ouvmexp to output the differene from
previous iteration.

(enami)

remove typo.

(christos)

update for new yacc.

(christos)

panic takes a printf format too.

(christos)

VIDIOC_QUERYCAP changes:
- use driver name instead of device instance name in 'driver' field
- add 'get_businfo' callback to fill in 'bus_info' field instead of
  hard-coding "USB"
- use the kernel version for the 'version' field instead of 1
- adapt pseye and uvideo drivers to changes in struct video_hw_if

(jmcneill)

fill audio_info.config with the string "usb:<cookie>" instead of "uaudio",
where <cookie> comes from the usbd device handle:

  $ audioctl -d /dev/sound1 config
  config=usb:00000009

this can be used by software to find the audio device(s) associated with a
USB video capture device

(jmcneill)

Sprinkle daddr_t.

(jakllsch)

Sprinkle daddr_t.
Adjust DISK_DEBUG printf formats to match.

(jakllsch)

regen

(jmcneill)

add Hauppauge WinTV USB2 FM

(jmcneill)

dm(4): Make LVM use non 512-byte sector devices.

(mlelstv)

Establishing an UMTS connection with ubsa(4) works fine with my Pismo,
so add it to our config.

(phx)

libzfs_fru_clear was not ported to NetBSD disable it foor now.

(haad)

#include "opt_lockdebug.h"

(skrll)

LOCKDEBUG is not referenced in this file anymore.

(skrll)

restore lost overlow check fix.

(christos)

Unwrap/reformat long lines.

(skrll)

Regen amd64 files for 32bit binaries support.

(njoly)

Add support for 32bit (i386) binaries on amd64.

(njoly)

For update mounts the root vnode is already in use and we must not
free it. Since the mount persists even when the update fails,
this is not a problem either.

(mlelstv)

Adjust LDFLAGS handling to fix curses detection for mknative-gdb.

(njoly)

Fix xref and pasto.

(wiz)

Bump date.

(skrll)

Add support for DF_1_BIND_NOW, DF_1_NODELETE and DF_1_NOOPEN marked
objects, and the RTLD_NODELETE and RTLD_NOLOAD flags to dlopen(3).

Mark libpthread as DF_1_NOOPEN and use it to test the functionality.

Somewhat taken from FreeBSD.

Fixes PR 42029.

OK from christos and joerg.

(skrll)

Bump ramdisk size by 64K. This allows release builds to finish on macppc.

(aymeric)

mips_vector_init needs to be called sooner so that the spl routine are
instantiated.

(matt)

Define bus_space_mmap

(matt)

Adopt the new spl/interrupt framework.

(matt)

Adapt to the new spl/interrupt framework.
Include <mips/locore.h> when needed.

(matt)

Bring locore_mips1.S into the new world order.  You can diff locore_mips1.S
against mipsX_subr.S and have reasonable output.  A lot of comments have
been changed to be common between the two.  The ordering of vectors/functions
have also changed to improve diff output.

(matt)

MIPS_LOCK_RAS_SIZE needs to be 256 since each RAS need 64 bytes and we can
have 4 of them.

(matt)

MIPS1 needs load delay nops.  Fix a problem in the RAS mips_spin_enter
where we weren't actually decrementing ci_mtx_count.

(matt)

If compiling for MIPS1, deal with the presence of load delay slots.

(matt)

Just use the generic tlb_read_indexed instead of the mips1 specific version.

(matt)

need to include <mips/locore.h> explicitly now.
Ansify cpu_exec.c

(matt)

Deal with MIPS_NUM_TLB_PIDS not being constant.

(matt)

make this thread-safe. We could use localtime_tz and make it even better,
but not now.

(christos)

update yacc to the most recent.

(christos)

mention new yacc.

(christos)

import latest byacc

Status:

Vendor Tag: DICKEY
Release Tags: byacc-20101127

(christos)

Fix speling ("deferred," not "defered") as reported in #netbsd-code on
freenode IRC.  While I'm here, clean up the wording later in the comment.

(riz)

Change way how we get device size from device. Do not use LSEEK but
DIOCGWEDGEINFO and DIOCGDINFO. This should fix problem with raidframe + lvm
reported by

Toby Karyadi on current-users@.

Thanks for reporting this issue.

(haad)

read 32bits instead of 64bits and truncate the upper 32bits

(cegger)

build fix: add missing initializers

(cegger)

Fix sub-section reference.

(njoly)

Now that we have allowed operator to access the control node, make sure
that he cannot cause damage, by only allowing the superuser to do ioctls
that can cause damage.

(christos)

Remove double-quotes from .Sx macros.

(njoly)

Deduce raw device correctly by prepending "r" to basename.

(mlelstv)

PR/44267: Michael van Elst: LVM devices have wrong permissions
- Centralize CPPFLAGS for DM_DEVICE_{UID,GID,MODE}
- Make DM_DEVICE_GID operator DM_DEVICE_MODE 0640 to be more NetBSD like
- make all the code use DM_DEVICE_MODE instead of hard-coding.
- make sure that all mknod calls are followed by a chown call.

(christos)

false needs stdbool.h

(christos)

Remove double-quotes from .Sx macro.

(njoly)

Remove a couple tests - they've moved to atf

(pgoyette)

Migrate a couple more tests to aft

(pgoyette)

make dm aware of physical sector sizes.

For aggregates of multiple disks we use the largest sector size from
all disks. For standard power-of-2 sizes this is the same as the least
common multiple. We still require proper alignment of the targets in
the mapping table.

ok by haad@

(mlelstv)

Get sector size from device properties using DIOCGDISKINFO and
use the disklabel (DIOCGDINFO) only as a fallback. This works
with wedges.

(mlelstv)

mount(2) doesn't remove vnodes from the freelist in the error path,
so that they get reused with a invalid pointer to a mount structure.

As a workaround, free the vnodes used to create the in-filesystem journal
immediately.

(mlelstv)

First release that shipped wdogctl(8) was 1.6, not 1.5.1.

(hauke)

Make wedges aware of underlying physical block size.

(mlelstv)

Fix the $NetBSD$ tag (missed the trailing $)

(pgoyette)

Make saving/restoring video modes work on my PowerBook.
The problem was, that the driver blindly assumed that everyone uses PLL3 for
everything which isn't true at least on my PowerBook, which uses PLL0. So now
we actually restore which PLL to use and save PLL0 as well.

(macallan)

Remove these tests from regress - they've been converted to atf

(pgoyette)

Migrate a couple more tests from the old regress structure to atf

(pgoyette)

use pthread__stacksize since size has not been initialized yet.

(christos)

Regen.

(jruoho)

Add IPI0001, an identifier for ipmi(4). From Dell M710.

(jruoho)

only mprotect base if we moved it.

(christos)

__HAVE_CPU_INFO_FIRST -> __HAVE_CPU_DATA_FIRST.

(njoly)

Fix Fn arguments.

(wiz)

It just so happens we don't need -Wno-attributes if we
place __packed in the right place.

(jakllsch)

Fix URL of the paper.

(jruoho)

Use __constfunc(3) instead of __attribute__((const)) directly.

(jruoho)

Use __dead instead of __attribute__((__noreturn__)) directly.

(jruoho)

Document the __builtin_prefetch() GNU extension.

(jruoho)

Add punctuation.

(wiz)

add vsnprintf_ss

(christos)

document vsnprintf_ss

(christos)

Fix nestio's behavior on error.

The mbp->b_resid is used to track if all the nested buffers have been issued
and reported back. When the last buffer calls in, mbp->b_resid becomes zero
and biodone(mbp) is called. This is fine as long as there are no errors.

If a read-error does occure in one of the nested buffers, the mbp->b_error is
set and on its call to biodone(mbp), with mbp->b_resid is zero, physio()
panics since it asserts that IF an error is set on a buffer, there should be a
residual amount of data left to transfered.

The patch fixes this case by setting mbp->b_resid back to mbp->b_bcount on
mbp->b_error just before biodone(mbp).

This behaviour is consistent with normal buffer issueing. It either succeeds
or doesn't succeed.

(reinoud)

Running with DEBUG again revealed locking errors previously undetected. The
VAT writeout was done locked while marked locked as were the readin and
writeout of the metadata partition space table.

While here, also protect the (vp) argument of the UDF_SET_SYSTEMFILE() macro.

Tested on UDF 1.50 sequential, UDF 2.01 RW and UDF 2.50 metadata RW meda.

(reinoud)

Growing the metadata partition is not yet implemented. Be so kind to report
this as unimplemented when the debug flag asks for it!

(reinoud)

Add iseek and oseek option as aliases for skip and seek respectively.
These options exist in dd of solaris or svr3.  From FreeBSD.

(enami)

Copy argument before modifying it so that ps shows entire argument.
From OpenBSD via FreeBSD.

(enami)

Remove trailing whitespace.

(wiz)

Improve wording.

(wiz)

Remove superfluous Pp.

(wiz)

Talk more about NetBSD and less about FreeBSD. Sort sections.

(wiz)

Sort option descriptions, bump date for -a.

(wiz)

we need uvm_extern.h now

(macallan)

Rework how fixups are processed.  Inside of generating a table, we just
scan kernel text for jumps to locations between (__stub_start, __stub_end]
and if found, we actually decode the instructions in the stub to find out
where the stub would eventually jump to and then patch the original jump
to jump directly to it bypassing the stub.  This is slightly slower than
the previous method but it's a simplier and new stubs get automagically
handled.

(matt)

Cleanup definition of __HAVE_PREEMPTION

(matt)

If we are MULTIPROCESSOR (which means preemption is enabled), we need to
reload curcpu() after disabling interrupts since we have been preempted
to a different CPU before interrupt got disabled.

(matt)

Add a check to more the ASID is valid on a user fault.

(matt)

Add a pmap_asid_check which verifies the current ASID is in COP0 ENTRY_HI

(matt)

Just call splcheck directly.  The stub fixer will make it right.

(matt)

Don't signal a preemption if we aren't enabling it.

(matt)

Fix a panic message.

(matt)

If the first thing is a conditional item, and it doesn't actually get
defines, genassym get very unhappy.  So move an item up so that never
happens.

(matt)

Allocate page from anywhere in KSEG0, not just the first 256MB.

(matt)

print the error code for writing parity.

(christos)

Say that -a is non-standard and -P is the default.

(christos)

Make __HAVE_CPU_DATA_FIRST true

(christos)

PROC_PC() should have been changed to LWP_PC().

(nisimura)

Emit DWARF relocation in its expected form

(matt)

Add CTASSERT to verify __HAVE_CPU_DATA_FIRST is correct defined or undefined.

(matt)

Collect cpu_info and friends and move to m68k/include/cpu.h
Add a define __HAVE_CPU_DATA_FIRST which means that cpu_data is the first
member in struct cpu_info.

(matt)

Add a define __HAVE_CPU_DATA_FIRST which means that cpu_data is the first
member in struct cpu_info.

(matt)

don't disable interrpts in ohci_intr1 - now USB works again on my Blade 2500
( hi gimpy )

(macallan)

- make sure cpu_switchto() not to touch MIPS_CURLWP register at newlwp
  switchframe restoration stage.
- discard MIPS_CURLWP assignments exposed in cpu_lwp_fork() and
  cpu_setfunc().
- use plain 'jal' instruction to call lwp_startup().

(nisimura)

Add a define __HAVE_CPU_DATA_FIRST which means that cpu_data is the first
member in struct cpu_info.

(matt)

Add a define __HAVE_CPU_DATA_FIRST which means that cpu_data is the first
member in struct cpu_info.

(matt)

Add a define __HAVE_CPU_DATA_FIRST which means that cpu_data is the first
member in struct cpu_info.

(matt)

Add a define __HAVE_CPU_DATA_FIRST which means that cpu_data is the first
member in struct cpu_info.

(christos)

Tickets 1495-1498.

(riz)

Pull up following revision(s) (requested by tsutsui in ticket #1498):
sys/dev/audio.c: revision 1.252
Fix botches in rev 1.185 (kent-audio1 merge) which cause buffer overrun
and "tlb_exception: no copyin/out fault hander" panic in
audio_fill_silence() after audio playback on dreamcast.
The problem is tracked and fix is provieded by Ryo SHIMIZU,
who is the auther of aica(4) audio driver for dreamcast.

(riz)

Pull up following revision(s) (requested by tsutsui in ticket #1497):
sys/arch/dreamcast/dev/g2/aica.c: revision 1.19
Make aica_round_blocksize() always return a fixed buffer size.
I'm not sure if this is correct, but audio playback doesn't work
on dreamcast without this workaround. (since kent-audio1 merge?)

(riz)

Pull up following revision(s) (requested by skrll in ticket #1496):
sys/kern/init_main.c: revision 1.418
fix LOCKDEBUG panic 'uninitialized lock'.
seminit() calls exithook_establish(). exithook_establish() uses the exec_lock.
exec_lock is initialzed by exec_init(1).
Call exec_init(1) before seminit().

(riz)

Pull up following revision(s) (requested by skrll in ticket #1495):
sys/arch/acorn32/stand/lib/riscospart.c: revision 1.3
Use partition_cyl_high and partition_cyl_low to find the "RiscBSD"
partition.
Makes nbfs somewhat work on my RiscPC.

(riz)

Add -a archive flag. from Aleksey Cheusov

(christos)

All ports have to expose 'struct cpu_info' when _KMEMUSER is defined.
We will just expose cpu_info.ci_data for the moment, which is needed for
vmstat(1) to compile. Doing so with the whole cpu_info structure is
problematic, because of imask_t.

(phx)

PR/44259: Aleksey Cheusov: xargs: -r option is supported but not documented
Synced with the FreeBSD doc.

(christos)

Add ttyE1-ttyE3 for wscons.
Use serial devices tty00, tty01, tty10 and tty11, which are configured
by MAKEDEV.

(phx)

Correct entry [jmcneill 20090804] (the hw.wake was moved to hw.acpi.wake).

(jruoho)

ACPICA 20101209 is out.

Major rework of the GPE handling. A lot of API breakages. Perhaps not worth
updating before 6.0, given that the current version works relatively well.

(jruoho)

Clean-up some of my entries.

(jruoho)

fix incorrect overflow check from Nicholas Joly

(christos)

+t_libm.debug.

(njoly)

Fix xref: rb(3) -> rbtree(3).

(njoly)

Document the __builtin_types_compatible_p() GNU/GCC extension.

(jruoho)

These tests have been converted to atf

(pgoyette)

Migrate a few more tests from old regress style to atf

(pgoyette)

Note vm.user_va0_disable.

(jruoho)

Note the 'vm.user_va0_disable', labelled as "protections against NULL
pointer dereferences".

(jruoho)

Remark acpicpu(4).

(jruoho)

PR/44213 - B ICT A.P. de Brouwer Jr. -- add some Motorola phones

(jnemeth)

regen

(jnemeth)

PR/44213 - B ICT A.P. de Brouwer Jr. -- clarify Motorola USBLAN

(jnemeth)

We don't expect tests to fail anymore.

(christos)

- document limitations
- add extra number name parsing

(christos)

- document limitations
- add synonyms for the numbers

(christos)

Remove round test from old regress structure

(pgoyette)

Move the only regress/lib/libm test to the new atf format

(pgoyette)

Now, get the return error too, in case that could help with EC2
troubleshooting...

(jym)

Consistently use .gnu.warning with .pushsectio and .popsection on all
architectures instead of obsolete STABS frames for linker warnings.

(joerg)

add myself

(ryo)

The driver didn't work, because it expects the whole data packet in a
single interrupt. Reality shows that with current kernels a 81 bytes packet
is split into 3 interrupts with 32 + 32 + 17 bytes. I have added a workaround
to deal with it. The Geyser2 devices with 64 byte packets are unsupported
at the moment. Somebody needs to test it.

New: Added support for the iBook 12-inch trackpad.

(phx)

A small clarification in the example.

(jruoho)

usbd_dump_*() functions are also needed when just EHCI_DEBUG is defined,
without USB_DEBUG.

(phx)

Xref module(7) and iic(4).

(jruoho)

Note options I2C_SCAN.

(jruoho)

Fix section, atf(1) -> atf(7).

(njoly)

Add testcase from PR/44255, that exercize parsedate(3) with sample
strings taken from the manpage.

(njoly)

Use ${HOST_SH}

(joerg)

Remove unneeded .debug entry for t_swsensor script.

(njoly)

Fix typos.

(wiz)

Remove C1E from CAVEATS, as it is now handled by acpicpu(4).

(jruoho)

Note that all relatively new x86 systems should prefer acpicpu(4).

Also mention that the legacy ENHANCED_SPEEDSTEP and POWERNOW_K8 are known to
be problematic or entirely dysfunctional with new processors.

(jruoho)

Use branch annotations in couple of places. Add two comments.

(jruoho)

Add a rump-based set of tests for sysmon_envsys(8) + powerd(8) + swsensor(4)

(pgoyette)

To use x86/cpu.h struct cpu_info from userland, we need via_padlock.h installed.

(christos)