Use lci.

(skrll)

Add lci to get coherence index for an address.

(skrll)

Fix print message.

(kiyohara)

KNF

(skrll)

Change print messages.
  1. Use aprint instead of DPRINTF in swap mode.
  2. Locates and vendor/LSI names 1-line print.
  3. Also print (and maybe support) LAN92xx.

(kiyohara)

Set mii_media_active in lan9118_ifm_change().  This is like ukphy_status().
Also LAN9118_TXS_NC is valid when if FDX.

(kiyohara)

Fix register access to MII in lan9118_mii_readreg()/lan9118_mii_writereg().

(kiyohara)

Add a bunch of useful typical examples. Also use appropriate markup there.

(tsutsui)

Replace l_addr with uvm_lwp_getuarea() in various MD code, mostly cpu_lwp_fork().

(rmind)

cpu_lwp_fork (MIPS): replace l_addr with uvm_lwp_getuarea(), clean up a little.

(rmind)

Pass pcb_fpcpu_slock correctly, since pcb0 is a pointer.
Fixes ia64 build (hi matt@).

(rmind)

- calculate end PA correctly.
- wrap more spitfire specific (flushing I$) code.
- fix inverted #if condition for flushing I$ (and restore ASI_ICACHE_TAG).

(nakayama)

Add no/op routines for m68000 for the sticky, mask, and round settings.

(christos)

When truncating a request in bounds_check_with_mediasize() multiply
by the provided sector size instead of 512.
Fixes last bit of PR/31565

(dsl)

It seems that LDADD.file gets added to the linker invocation after
the LDADD variable expansion.  To support static linking, ensure
that the common libraries gets tacked on at the end as well (order
matters for static linking), by using a convenience variable and
add it to the end of LDADD.t_modlinkset, as well as when doing the
general LDADD+= setting.

(he)

Only do the nss module if MKPIC != no.

(he)

Use ${.CURDIR} when referring to a relative source directory.

(he)

Add LINUX_SIOCGIFMTU and LINUX_IP_HDRINCL support.
Fixes part of PR/31358
The other parts are rather too intrusive to be fixed as in the PR.

(dsl)

add the kernel config i've been mostly using for USIII/schizo support

(mrg)

implement enough code to get SB2000/SB2500 booting to multiuser.  does
not last long if you use it, but an idle machine has stayed up 10 hours.

- for the !SPITFIRE case, use ASI_DCACHE_INVALIDATE instead of
  ASI_DCACHE_TAG when clearing tlb entries.  from openbsd.

- convert some comment "! Not yet" into "WRITEME", so that attempting
  to build a USIII MULTIPROCESSOR kernel fails to build until the code
  is written.

- add some more WRITEME's for the above.

- implement a !SPITFIRE version of sp_tlb_flush_all().  since there are
  no spare regs, for now just always increase %tl around this call.  (we
  could maybe store a token in %o3 along with the saved %pstate info.)

(mrg)

Ticket 1367

(sborrill)

Cast a vaddr_t (integral type) to void* before passing to memset.

(he)

Pull up the following revisions(s) (requested by darcy in ticket #1367):
usr.bin/comm/comm.c: revision 1.18

Don't include newlines when comparing to prevent errors when lines have
characters that sort lower such as tabs.

(sborrill)

give more debug output about mapping interrupts.

(mrg)

add some ultrasparcIII defines, from openbsd.

(mrg)

Add a note telling people to use the shell getopts builtin.

(dsl)

Use shell builtin getopts instead of getopt(1).
Fixes PR/30425

(dsl)

Set irm->bufmem = uvmexp.filepages;
as suggested in PR/30408

(dsl)

Ticket 1171

(sborrill)

Pull up the following revisions(s) (requested by bouyer in ticket #1171):
sys/kern/vfs_subr.c: revision 1.392

Previous caused a deadlock with layered FS: the vrele thread can sleep on
the vnode lock, while vget is sleeping on the VI_INACTNOW flag (or the vget
caller is looping on vget returning failure because of the VI_INACTNOW
flag). With layered FSes, the upper and lower vnodes share the same lock, so
the vget() caller above can be already holding the vnode lock.

Fix by dropping VI_INACTNOW before sleeping on the vnode lock in
vrelel(), and check the ref count again once we have the lock. If the
vnode has more than one reference, don't VOP_INACTIVE it.
Fix PR kern/42318 and PR kern/42377

(sborrill)

Forgotten commit. Noticed by snj, thanks!

(mbalmer)

Update -u usage to follow changes in fdisk.c rev 1.125:
>> Allow MB, GB and CYL (not just M, G and C) and lower case.
>> Don't output a splurious 'd' before "cyl".
>> Fixes PR/37414.

XXX "NNcy" is also allowed?

(tsutsui)

initialise -> initialize
unabled -> unable
wrap long line

(cegger)

u_int32_t -> uint32_t
struct device * -> device_t

(cegger)

Add the xf86-input-elographics driver to i386 builds.

(mbalmer)

Clarify description of -f and -u options.
(-u means updating partition data, not interactive mode)

(tsutsui)

Tickets 1143 1145 1147 1148 1151 1153 1154 1155 1157 1159

(bouyer)

Pull up following revision(s) (requested by tnozaki in ticket #1159):
dist/nvi/ex/ex_cscope.c: revision 1.5
dist/nvi/ex/ex_tag.c: revision 1.7
don't use pathname directly as msgq()'s first argument of format string.

(bouyer)

Pull up following revision(s) (requested by dogcow in ticket #1157):
sbin/savecore/savecore.c: revision 1.81
Instead of exiting with an obscure error message if -N /kernelname isn't
specified, blithely assume the kernel will consume around 20 megs.

(bouyer)

Pull up following revision(s) (requested by pooka in ticket #1155):
lib/libpuffs/creds.c: revision 1.15
In getgroups(), copy only up to the number of supplementary groups
that actually exist.

(bouyer)

Pull up following revision(s) (requested by pooka in ticket #1154):
sys/fs/puffs/puffs_vnops.c: revision 1.140
Send VOP_ABORTOP() in case attempting cross-dev rename, part of
PR kern/42210.  Also, fix a memory management error in said case.

(bouyer)

Pull up following revision(s) (requested by pooka in ticket #1153):
sys/fs/puffs/puffs_vnops.c: revision 1.139
Send VOP_ABORTOP() as a FAF -- we don't care about the return value.

(bouyer)

Pull up following revision(s) (requested by nakayama in ticket #1151):
share/man/man4/man4.sparc64/lom.4: revision 1.3
sys/arch/sparc64/dev/lom.c: revision 1.2, 1.3
Merge changes between revision 1.16 and 1.19 of OpenBSD with
shutdownhook_establish(9) to pmf(9) conversion:
- LOMlite seems to get wedged from time to time; add some code to unwedge it.
- Make sure we don't insert and entry into the list of pending commends twice.
- Establish a shutdown hook to disable the watchdog timer to prevent watchdog
  triggers after the kernel has been halted.
- Handle LOMlite2 in an interrupt-driven way; avoids using delay(9) once the
  machine is up and running.
Add support for monitoring Fault LED and Alarms status.

(bouyer)

Pull up following revision(s) (requested by joerg in ticket #1148):
sys/net/if.c: revision 1.241
Simplify ifreq_setaddr:
- Drop the INET6 block. The commands are never given to this function
  and truncating the sockaddr is arguably not the desired result anyway.
- Clear the address before copying. This fixes SIOCGIFNETMASK and possible
  other ioctls for users that don't check sa_len. This includes
  COMPAT_43 and Linux emulation.
OK dyoung@

(bouyer)

Pull up following revision(s) (requested by joerg in ticket #1147):
sys/compat/linux32/common/linux32_socket.c: revision 1.11, 1.12
sys/compat/linux/common/linux_socket.c: revision 1.105, 1.106
sys/compat/linux/common/linux_sockio.h: revision 1.17
sys/compat/linux32/common/linux32_sockio.h: revision 1.3
Provide SIOCGIFNAME.
Return the result of copyout. Reminded by Niolas Joly.

(bouyer)

Pull up following revision(s) (requested by tnozaki in ticket #1145):
dist/nvi/common/exf.c: revision 1.4
fix format string bug, filename may contain % character, don't use it
as format string.

(bouyer)

Pull up following revision(s) (requested by mbalmer in ticket #1143):
sys/arch/i386/stand/lib/pcio.c: revision 1.26
Have the boot countdown on i386 display "starting in N seconds." instead
of "starting in N" and eliminate a use of sprintf.  Note that on
some rare machines it can be that the BIOS does not provide the delay
function.  On such machines the countdown will almost immediately count down
to zero display "starting in 0 seconds."; apparently the net4801 is such a
machine.
Feedback, ideas, and inspiration from tron, ok tron/tonnerre

(bouyer)

elographics configuration

(mbalmer)

Initial import of xf86-input-elographics-1.2.3.  From X.Org via OpenBSD.

Status:

Vendor Tag: mbalmer
Release Tags: xf86-input-elographics-1-2-3

(mbalmer)

Flush only as much as is required. Pointed out by rmind.

(skrll)

Comment out interrupt for LAN9118_INT_PHY_INT and LAN9118_INT_PME_INT.  These
not support yet...

(kiyohara)

Check mbuf size and buffers.

(kiyohara)

Don't use #ifdef __${MACHINE_ARCH}__ to enable machine dependent features.
Instead, use proper macro defined in Makefile per ${MACHINE_ARCH}.

__${MACHINE_ARCH}__ doesn't represent an architecture of tool's target
but an architecture of binaries being compiled, so required features
are not prolery enabled or unintentionally enabled on certain host
and target combinations during src/tools build.

(tsutsui)

Mention xf86-input-ws

(mbalmer)

Link the xf86-input-ws X.Org input driver to the build and add it to the lists
that also have the wsmouse driver.

xf86-input-ws adds support for touchpanels in X11 that report absolute position
data through wscons(4), e.g. the extended ums(4) driver.  xf86-input-ws was
written by Matthieu Herrb for OpenBSD.

Please remember to also update xsrc before the next system build.

(mbalmer)

Apply some band-aid to pmap_activate() for PR kern/41058:

There's a corner case here which can leave turds in the cache as
reported in kern/41058. They're probably left over during tear-down and
switching away from an exiting process. Until the root cause is identified
and fixed, zap the cache when switching pmaps. This will result in a few
unnecessary cache flushes, but that's better than silently corrupting data.

Also remove an extraneous return statement in pmap_page_protect() which
crept in during the matt-armv6 merge.

(scw)

Make dreamcast, evbsh3, and mmeye use -DUSE_MBR.
All of these ports use src/sys/arch/sh3/sh3/disksubr.c
which is MBR aware.

(tsutsui)

- Display "Enter '?' for help" when starting interactive mode
- Use %.40g rather than %g when printing sectors and MB for existing
  partition size/offset.
  Changes [1.93802e+06c, 1953525105s, 953870M]:
  to:    [1938021c, 1953525105s, 953869.6875M]:

(abs)

Shuffle the order of examples to put the interactive option (-i) ahead
of $EDITOR and "disklabel -w -r /dev/rsd0c sd2212 foo"

(abs)

Previous did cause a deadlock with layered FS: the vrele thread
can sleep on the vnode lock, while vget is sleeping on the
VI_INACTNOW flag (or the vget caller is looping on vget returning failure
because of the VI_INACTNOW flag). With layered FSes, the upper and lower
vnodes share the same lock, so the vget() caller above can be already
holding the vnode lock.

Fix by dropping VI_INACTNOW before sleeping on the vnode lock in
vrelel(), and check the ref count again once we have the lock. If the
vnode has more than one reference, donc VOP_INACTIVE it.
Fix PR kern/42318 and PR kern/42377
patch tested by Hisashi T Fujinaka, Joachim K�nig, Stephen Borrill and
Matthias Scheler.

(bouyer)

Always sector size is treated as 512 bytes.

(nonaka)

Fix function name that was changed by mistake in the previous whitespace
commit.

(mbalmer)

Support mii_tick().
  Thanks tsubai@.

(kiyohara)

pxartc(4) must set the device address.

(nonaka)

Add support for monitoring Fault LED and Alarms status.

(nakayama)

Don't include newlines when comparing to prevent errors when lines have
characters that sort lower such as tabs.

This is a temporary fix to allow pullups to existing, supported versions
of NetBSD.  I will follow up with a version for current using the new
getline function.

(darcy)

Correct return value of ne_*_match().
It's attach priority, not ne2000 model.

(isaki)

white space -> tab.

(isaki)

Modernize and clean up.

(isaki)

clean up some debug messages a little bit.

(mrg)

Add patch from PR/11805

(dsl)

Fix DEBUG build.  Hi rmind.

(skrll)

Fix compressed server accesses.
According to http://cvs.savannah.gnu.org/viewvc/ccvs/src/?root=cvs
this was broken between revs 1.29 and 1.32 for a perion of 4 months
about 4 years ago.
Not sure at all why/how we have the broken version.
Our old code, the latest gnu code and this fix all differ slightly,
but have the same effect.

(dsl)

Remove no longer necessary vfs weak symbols.

(pooka)

include subr_kobj_vfs

(pooka)

Due to the schizophrenic nature of kobj (mem + vfs source),
split the module in twain to subj_kobj.c (master + mem) and
subr_kobj_vfs.c (vfs).

(pooka)

g/c unused protos

(pooka)

Move rootfs-related init from init_main() to vfs_mountroot().
Reduces code re-written in rump.

(pooka)

Yes, you need a VOP_UNLOCK() there.

(pooka)

Fix another fine example of "if you make the kernel panic, it's
more secure".

(pooka)

Don't spam if mount fails due to invalid file system.

(pooka)

Remove fdisk/mbr for macppc which was blindly pulled from OpenBSD.
It just fakes MBR partition map which contains 1MB FAT16B partition
and ~1GB OpenBSD partition, and we can always create necessary
MBR partitions for OpenFirmware by the fdisk(8) command itself.

(tsutsui)

Fix typo and hopefully the dependency problem that manifests itself with
compile problems around lang_insert_orphan.

(skrll)

initial import of xf86-input-ws, from OpenBSD

Status:

Vendor Tag: mbalmer
Release Tags: xf86-input-ws-1-2-9

(mbalmer)

Move -mcmodel=kernel CFLAGS from bsd.klinks.mk to amd64/include/Makefile.inc
to avoid having the kernel toolchain flags split over a billion different
files.

(pooka)

Now that Makefile.rump was changed and everything gets built in
update builds too, flip the allocator define to prefer the kernel
pool/kmem instead of malloc(3).  Use malloc(3) only if
RUMP_USE_UNREAL_ALLOCATORS is defined.

(pooka)

Include arch/include/Makefile.inc for some compiler flags.

(pooka)

Make this work on some m68k ports which like putting the disklabel
in the third sector (or have copypasted disklabel.h from a port
which likes doing that ;).

(pooka)

The expression (yyvs != 0) is always true and not a necessary test
in the first place since realloc(NULL, size) is equivalent to
malloc(size).  Write code straightforward while i'm here.

(enami)

Make break(2) reject high adresses that wrap to 0 after page rounding.

(njoly)

If CPPFLAGS or CXXFLAGS already contains -nostdinc or -nostdinc++,
don't add ${CPPFLAGS_ISYSTEM} or ${CPPFLAGS_ISYSTEMXX} to refer
${DESTDIR}/usr/include or ${DESTDIR}/usr/include/g++.

This change might cause errors on some MD stand dir, but in that case
each Makefile should be fixed to search proper system include paths
by -Ipath option in own CPPFLAGS.

(tsutsui)

u_short -> uint16_t, some KNF.

(tsutsui)

Add definitions for more reserved inodes.

(tsutsui)

Add src/common/include to -Ipath for proplib(3) so that build works
even without installed DESTDIR.

(tsutsui)

Add src/common/include to -Ipath for proplib(3) so that build works
even without installed DESTDIR.

(tsutsui)

Move some devices where they should belong to.

(tsutsui)

Fix typo in comment.

(tsutsui)

Use %zu in printf format for size_t value.

(tsutsui)

Remove unnecessary and never used usr/mdec/mbr from install ramdisk.

(tsutsui)

Ticket 1164

(sborrill)

Pull up the following revisions(s) (requested by mbalmer in ticket #1164):
doc/CHANGES: revision 1.1320 via patch
share/man/man4/ums.4: revision 1.9
sys/dev/usb/ums.c: revision 1.77

Add support for USB HID devices that report absolute coordinates instead
of relative movement data, e.g. touchpanels.  Please note that calibration
must be done outside the wscons(4) driver for now.

(sborrill)

Ticket 1141

(sborrill)

Pull up the following revisions(s) (requested by dsl in ticket #1141):
sys/kern/tty_subr.c: revision 1.38

Fix clrbits() so that it doesn't mask no bits out of the byte after the
range (when the last bit to be cleared is the msb of a byte).
Fixes PR/42312.

(sborrill)

Ticket 1141

(sborrill)

Pull up the following revisions(s) (requested by dsl in ticket #1141):
sys/kern/tty_subr.c: revision 1.38

Fix clrbits() so that it doesn't mask no bits out of the byte after the
range (when the last bit to be cleared is the msb of a byte).
Fixes PR/42312.

(sborrill)

Ticket 1163

(sborrill)

Pull up the following revisions(s) (requested by tsutsui in ticket #1163):
sbin/newfs_msdos/newfs_msdos.c: revision 1.30

Make fs size detection get proper size rather than disk size.
Without this, newfs_msdos(8) assumes the target fs size is
whole disk size, so newfs_msdos(8) will fail or create wrong fs.

(sborrill)

Ticket 1163

(sborrill)

Pull up the following revisions(s) (requested by tsutsui in ticket #1163):
sbin/newfs_msdos/newfs_msdos.c: revision 1.30

Make fs size detection get proper size rather than disk size.
Without this, newfs_msdos(8) assumes the target fs size is
whole disk size, so newfs_msdos(8) will fail or create wrong fs.

(sborrill)

Rafal ticket 1162

(sborrill)

Regen

(sborrill)

Pull up the following revisions(s) (requested by rafal in ticket #1162):
sys/dev/usb/uhidev.c: revision 1.44
sys/dev/usb/ums.c: revision 1.75
sys/dev/usb/usbdevs: revision 1.536

Fix kern/41737. Add quirks to make MS Wireless Laser Mouse 6000 work.

(sborrill)

Add support for USB HID devices that report absolute coordinates instead
of relative movement data, i.e. touchpanels.  Please note that calibration
must be done outside the wsons(4) driver for now.

(mbalmer)

- Use uvm_lwp_setuarea() instead of directly setting address to lwp_t::l_addr.
- Replace most remaining uses of l_addr with uvm_lwp_getuarea() or lwp_getpcb().
- Amend assembly in ports where it accesses PCB via struct user.
- Rename L_ADDR to L_PCB in few places.  Reduce sys/user.h inclusions.

(rmind)

Lookup lwp0.l_addr instead of proc0paddr to locate PCB.

(enami)

Revert unrelated/unintended changes from previous commit.
Should fix MIPS builds (hi matt@).

(rmind)

Make sure rumpfs is attached on all host platforms.

(pooka)

For rumpfs, do mountroot instead of the bunch of homegrown hacks
currently there.  Still needs a little massage to get the kernel
interfaces right and avoid copypaste especially from main().

Also, move it a bit more into the direction of a real file system
(finally!) by giving it a vfsops.  Most ops are still unimplemented,
though.

(pooka)

Add DV_VIRTUAL for non-backed virtual devices and allow to mount
root from a DV_VIRTUAL device.

(pooka)

report the old -> new interrupt mapping when swizzling.  minor KNF.

(mrg)

Cast vstart to (struct user *) when assigning it to lwp0.l_addr.

(phx)

remember to call pipe_init()

(pooka)

Since rumpfs has supported file system sockets for quite a while
now, we don't need tmpfs here.  But, rumpfs doesn't support regular
files, so pass a pipe descriptor instead of an open file fd.

(pooka)

Provide some sort of cv_is_valid().  What a silly routine.

(pooka)

include sys_pipe.c

(pooka)

regen for rump_sys_pipe()

(pooka)

typo in comment (it actually breaks the script totally.  i wish
more typos in comments were as effective)

(pooka)

pipe +RUMP

(pooka)

Add rump support for the special handling required by pipe(2).

(pooka)

regen: retval -> retval[2]

(pooka)

Instead of a single register_t as the retval of rump syscalls,
use an array of two.  No functional change ... yet.

(pooka)

include prerequisite headers

(pooka)

* load symbols only up to actual size of symbol table instead of size
  we allocated for storing the table
* trimm unused end off of table before passing to the kernel

(pooka)

Cleanup interrupt establish error messages. Do not mix
aprint_error/aprint_normal/printf calls for a single line.

(njoly)

add MOUNT_RUMPFS

(pooka)

add some recent activities

(pooka)

Shuffle types around and add some random typecasts to make lint
and LP64 gcc sign comparison shut up.  I'm not sure what benefit
this brings apart from introducing some more bugs for everyone's
debugging pleasure.

(pooka)

Remove assym.h.  It's no longer required by hppa since spcopy is
not included anymore.

(pooka)

update a bit

(pooka)

Build the kernel symbol table in rumpuser bootstrap and feed it to
the rump kernel.  After this change it is possible to use the
in-kernel linker and rump_sys_modctl() to load kernel modules at
runtime.

Previously loading modules at runtime was possible only through
using the host system ld.so.  Note that it is still preferred to
use shared libs when possible, since they are PIC and n virtual
kernels will only require one copy of r/o segments.  However, when
there is no access to source code, a binary kernel module is the
only thing available ...

(pooka)

regen: rump_kernelfsym_load()

(pooka)

Add rump_kernelfsym_load(), which loads the kernel symbol table.

(pooka)

regen: rump_sys_modctl()

(pooka)

modctl +RUMP

(pooka)

Use strong alias within the kernel namespace regardless of if we're
dealing with a hard or soft kernel (kernel linker doesn't support
weak symbols).

(pooka)

include sys/param.h first, as is the convention

(pooka)

Include kern_module_vfs and the modctl() syscall.  The modctl()
syscall is here because module loading is a vfs operation.
Theoretically you shouldn't be able to have modules without vfs
support.

(pooka)

Don't build sh64 code even if BFD64 as suggested by uwe@ (privately).  We
don't need sh64 at all.  This fixes sh3 MKCROSSGDB=yes build.

(uebayasi)

cgthree works on sparc64 so let's enable it for those poor souls who can't
find at least a CG6

(macallan)

On the ARM platform, sys/param.h needs to be before sys/mutex.h, otherwise
sys/device.h will fail compilation (struct kmutex will not be defined
completely).

(kenh)

Back out unintended commit.

(matt)

Kill proc0paddr.  Use lwp0.l_addr instead.

(matt)

make this compile on sparc64

(macallan)

Fix a missed proc -> lwp change, only compiled if NWSKBD

(abs)

Fix build of kernels without PCI support like "GENERIC_TINY".

(tron)

Enable SSP (Stack Smash Protection) in x86 kernels by default (except
in i386 *TINY kernels). The NetBSD/i386 "ALL" kernel is unconditionally
compiled with SSP enabled.

Change approved by the core team.

(tron)

Don't use LDADD.progname because it duplicates all the libs.

(pooka)

aprintify.

(njoly)

* report partition size in blocks instead of bytes
* "support" DIOCCACHESYNC.

(pooka)

make WAPBL_DEBUG_PRINT compile

(pooka)