Make sure we hold alldevs_mtx for access to alldevs in deviter.

- Extend alldevs_mtx section in deviter_init.
- Assert ownership of alldevs_mtx in private functions:
  . deviter_reinit
  . deviter_next1
  . deviter_next2
- Acquire alldevs_mtx in deviter_next.

(alldevs_mtx is not relevant to the struct deviter object, which is
private to the caller who must guarantee exclusive access to it.)

(riastradh)

#if DIAGNOSTIC panic ---> KASSERT

(riastradh)

__diagused police

(riastradh)

__diagused police

(riastradh)

Fix inadvertently reversed sense of comparisons.

(riastradh)

#if DIAGNOSTIC panic ---> KASSERT

(riastradh)

PR bin/52090

Add 5 new test cases to test various ways that $* can be expanded.
Currently 3 of the 5 are marked as "expected to fail" because of the
bug in this PR.

(kre)

Fix configure script to not always provide an extern for strsep().

(jakllsch)

don't redefine EFTYPE (tools build, defined by compat_defs.h)

(christos)

Support __HAVE_MM_MD_DIRECT_MAPPED_PHYS

(skrll)

Mention about addition of luna68k's front panel LCD driver.

(tsutsui)

regen

(riastradh)

Fix locking comments.

Forgot to save my Emacs buffer before last commit.

(riastradh)

Summarize lifetime of cache entries.

(riastradh)

Omit duplicate forward declaration of cache_invalidate.

(riastradh)

Fix lock order statement.  Annotate with references to examples.

(riastradh)

Omit excessive indentation for text.

(riastradh)

Bump date.

(riastradh)

regen

(riastradh)

Build with DTrace support by default.
To activate, define the relevant modules in /etc/modules.conf.
For example (incomplete list):
solaris
dtrace
dtrace_sdt
dtrace_fbt
dtrace_lockstat
dtrace_profile
dtrace_syscall

Reviewed by christos

(sevan)

Rework namecache locking commentary.

- Annotate struct namecache declaration in namei.src/namei.h.
- Bulleted lists, not walls of texts.
- Identify lock order too.
- Note subtle exceptions.

(riastradh)

Nix trailing whitespace.

(riastradh)

Sort #includes.

(riastradh)

Omit vestigial comment.

- We have not dropped the cache entry on vget failure since 2008.
- We have not had `generation numbers' since 2001.

(riastradh)

regen

(riastradh)

Make cache_lookup return bool for clarity.

(riastradh)

Remove excessive indentation from function description text.

(riastradh)

Move out-of-place note about interrupt context and namei.

Tweak paragraph break.

(riastradh)

Use `\(em', not `--'.

Refill sentences and tweak wording where appropriate while here.

(riastradh)

Clarify wording: pathname component is not a vnode or lookup result.

(riastradh)

Organize namei flags by whose perspective they are relevant to.

- namei callers
- input to VOP_LOOKUP set by caller
- input to VOP_LOOKUP set by namei
- output from VOP_LOOKUP

(riastradh)

Rewrite intro to say positively, actively how you use namei.

(riastradh)

Rewrite operating modes section.

- Identify contract with caller.
- Identify contract with file system.
- Identify failure modes.
- Write in bulleted lists, not long rambly paragraphs.

(riastradh)

Bump image size to unbreak builds.
nbmakefs: `work' size of 21610496 is larger than the maxsize of 20480000.

ok jakllsch

(sevan)

Clarify what NOCACHE and MAKEENTRY do and who uses them.

De-editorialize.

(riastradh)

new mdocml; fix sqlite

(christos)

merge conflicts.

(christos)

Mmh, allow iret to be handled when an #SS fault (T_STKFLT) happens. Even
if the sdm is far from being clear, it appears that iret can trigger an #SS
fault if %ss points to a writable but non-present segment; in which case
the kernel would panic, thinking the fault was internal to it.

In particular, userland can create a broken segment in the ldt with
USER_LDT, update its %ss with setcontext and trigger the panic. I don't
think amd64 is affected since USER_LDT does not exist there, and the
changes on tf_ss seem correct - but I'm still adding T_STKFLT for safety.

(maxv)

Style, and remove debug code that does not work anyway.

(maxv)

Add NOLUNS quirk for SEAGATE ST39236LC disk drives.

Found on "SCSI drive conservation activity" for poor Tier-II machines.

(tsutsui)

Make VIMPL_TO_VNODE and VNODE_TO_VIMPL type-safe.

(riastradh)

1353, 1365, 1368

(snj)

Need membar_datadep_consumer here.

(riastradh)

Nix trailing whitespace.

(riastradh)

Nix trailing whitespace.

(riastradh)

#if DIAGNOSTIC panic ---> KASSERT

(riastradh)

#if DIAGNOSTIC panic ---> KASSERT

(riastradh)

#if DIAGNOSTIC panic ---> KASSERT

(riastradh)

#if DIAGNOSTIC panic ---> KASSERT

(riastradh)

#if DIAGNOSTIC panic ---> KASSERT

(riastradh)

Pull up following revision(s) (requested by uwe in ticket #1368):
sys/kern/exec_elf.c: revision 1.88 via patch
netbsd_elf_signature - look at note segments (phdrs) not note
sections.  They point to the same data in the file, but sections are
for linkers and are not necessarily present in an executable.
The original switch from phdrs to shdrs seems to be just a cop-out to
avoid parsing multiple notes per segment, which doesn't really avoid
the problem b/c sections also can contain multiple notes.

(snj)

#if DIAGNOSTIC panic ---> KASSERT

(riastradh)

Pull up following revision(s) (requested by kre in ticket #1365):
usr.bin/pkill/pkill.c: revision 1.31
PR bin/50934 -- avoid core dump if a process being examined has set
its argv[0] to NULL.

(snj)

Pull up following revision(s) (requested by nonaka in ticket #1353):
usr.bin/ftp/fetch.c: revision 1.227
Use the first name we requested the http/https URL for, not any name we ended
up with after random redirects.

(snj)

apply changes to this file as well, as part of:
allow pcu_save() and pcu_discard() to be called on other threads,
ptrace needs to use it that way.

(chs)

fixes from FreeBSD via Kurt Lidl

(christos)

Add the local route after finishing the configuration of the address.
This fixes the issue where the initial address announced had an
invalid broadcast address.

(roy)

If we're not doing DAD, don't set IN_IFF_TENTATIVE.

(roy)

Report the PID for the process changing the address.

(roy)

Modify some variables to make code more understandable.

(msaitoh)

Don't print BAR's window size on pcictl pciN dump again
because there is no way to know it's correct value from
userland now.

(msaitoh)

Bump date for previous.

(wiz)

Add missing NULL check

Fix PR kern/52083

(ozaki-r)

- Don't try to sizing a BAR not if a device is host bridge but a device's
  PCI_COMMAND_MEM_ENABLE is set.
- Fix 17 years old bug that all effective BARs are marked "not sized" on
  pcictl pciN dump.
- Print  BAR's window size on pcictl pciN dump.
- remove extra "(?)"

(msaitoh)

Fix cksum calculation for clearing the cached route.

In ping.c:r1.104, the size of echoreply packet was changed to ICMP_MINLEN,
Therefore also calculation size must be ICMP_MINLEN.

(ryo)

Back out part of previous: missed a caller of wapbl_write_inodes.

(riastradh)

Nix trailing whitespace.

(riastradh)

Sort includes.

(riastradh)

Assert write lock in wapbl_write_revocations, wapbl_write_inodes.

Only one call site, so trivial to prove correct.

(riastradh)

need to turn off async during ffs_sync(), otherwise its bwrite() calls are
themselves turned to bdwrite(), creating dirty delayed writes

fixes panic for 'mount -o log,async ...' reported by Masanobu SAITOH
on current-users; fix help by hannken@, thank you

(jdolecek)

lrint() and lrintf() are supposed to use the current rounding mode,
so use vcvtr rather than vcvt.

(chs)

allow pcu_save() and pcu_discard() to be called on other threads,
ptrace needs to use it that way.

(chs)

Undo local changes not intended to be committed (and certainly not
with that commit message) in the previous update.  This stuff works,
and will probably appear sometime, but not right now.

(kre)

Have "make clean" remove sh.html1 and adapt it to clean trace files
the way they have been generated the past 20 years or so...

(kre)

Add a test case of ifconfig <if_index>

I don't know if <if_index> is expected to be accepted instead of
an interface name. Please update the test case if the behavior is
just a bug and ifconfig gets fixed.

(ozaki-r)

Simplify ifunit, if_get and if_get_byindex by reusing other functions

Inspired by kre@'s comment

(ozaki-r)

Fix panic on ifconfig <number>

Pointed out by s-yamaguchi@IIJ

(ozaki-r)

fix: remove unmatched mutex_exit/enter. must be forgetting to remove at crypto.c:r1.41

(knakahara)

clarify history.

prompted by a comment on the blog from "dark"

(snj)

actually cast to unsigned long long and use %llu. certainly not use hex (oops)
suggested by dh

(maya)

Dedup functions which vary only by type using fancy macros.
mark them static. NFCI

(maya)

more trash to take out at bump time

(dholland)

print inode number in an assert I keep hitting and the adjacent one.
use PRIx64 for printing inode number elsewhere.

(maya)

PR/52078: Don't panic on 0 allocation, check more bounds.

(christos)

Add a comment to answer a question regarding privilege separation when
modifying a PTE from an active page tree. The question is from Manuel
Bouyer, and the answer is from me.

(maxv)

Change duplicate 'D' option to 'm'. From Ed Maste @ FreeBSD

(christos)

Bump date for previous.

(wiz)

Sort SEE ALSO.

(wiz)

Do not assume an all-zeroes disk label will read back as a disk with
an 'a'  partition covering the whole disk, instead use the raw partition.
Also skip this test if the fss device is not available.
With lots of help from pgoyette.

(martin)

Fix memory leak in vlan_start

(ozaki-r)

Add another 100 Series SMBus (100SERIES_LP_SMB)

(msaitoh)

7.1_STABLE

(snj)

7.1 -> 7.1.1

(snj)

add netbsd-7-1

(snj)

7.1.0_PATCH

(snj)

add 7.1 release date

(snj)

iwm(4): update firmware to maintained core release.

firmware from https://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/linux-firmware.git/tree/?id=4b9559fb0565f582a21a4070a81a3b6a7d77610b

(nonaka)

libevent is at 2.1.8, not 2.1.18

(snj)

Build support for libevent.

(roy)

use x alt format and print vendor and product in hex too

(christos)

PR/51380: Mutex error: mutex_vector_enter: locking against myself

Change the interrupt handler IPL level to IPL_VM (aka IPL_BIO aka
IPL_SDMMC) so that the sdhc_host_lock prevents edma_intr from running

(skrll)

Fix build of puffs

(ozaki-r)

Document option console=auto. Works since at least NetBSD 6. [i386,amd64]

(bad)

wpa_supplicant.conf(5) not (8)

(kre)

document the current easiest way to use WPA for wireless networking.
don't have my laptop here and going by memory so it may be slightly wrong.

(maya)

Add missing pserialize_read_exit

Pointed out by riastradh@

(ozaki-r)

Add garbage instructions at end of kern_assert after vpanic.

This avoids having the return address passed to vpanic confusingly
appear to be some other symbol.

(riastradh)

Bump kernel version for ifnet_psref_class hiding

Welcome to 7.99.66

(ozaki-r)

Use if_acquire and if_release instead of using psref API directly

- Provide if_release for consistency to if_acquire
- Use if_acquire and if_release for ifp iterations
- Make ifnet_psref_class static

(ozaki-r)

Fix use of curlwp_bind

There was an error path that returned without curlwp_bindx.

(ozaki-r)

ECN: Change Root Complex Event Collector Class Code
- Old RCEC has subclass 0x06. It's the same as IOMMU. Read the type in PCIe
  extend capability to know whether it's RCEC or IOMMU.

(msaitoh)

Pint PCI Interface name if available.

(msaitoh)

Add two new PCI classes:
- processing accelerators
- non-essential instrumentation

(msaitoh)

Fix race condition in sysctl_iflist

We need to use psref for the ifa iteration because iflist_addr can sleep.

(ozaki-r)

No functional change:
- modify variable names
- KNF

(msaitoh)

boottime is a timespec these days.
From Tomoyuki Sahara in PR 52059.

(dholland)

Use frame_dump if TRAP_SIGDEBUG is enabled

This unbreaks the build of i386 ALL kernel.

(ozaki-r)

Mark the msdos parition active for boards like the BeagleBone Black

(skrll)

Replace DIAGNOSTIC + panic with KASSERT

(ozaki-r)

Replace DIAGNOSTIC + panic with CTASSERT

(ozaki-r)

Avoid debug printf just if DIAGNOSTIC

(ozaki-r)

Remove unnecessary NULL check

(ozaki-r)

regen

(nonaka)

Added Intel Wireless 3168 entry.

(nonaka)

#if DIAGNOSTIC panic ---> KASSERT

- Omit mutex_exit before panic.  No need.
- Sprinkle some more information into a few messages.
- Prefer __diagused over #if DIAGNOSTIC for declarations,
  to reduce conditionals.

ok mrg@

(riastradh)

several small tweaks:
- use MSIX if available
- fix off-by-one for nluns/ntargets (code inspection, not actual effect)
- when virtio_enqueue_reserve() fails, do not execute vioscsi_req_put()
  and hence virtio_dequeue_commit() as there is nothing to commit, just
  unload the xs and return; also report XS_RESOURCE_SHORTAGE, it's
  semi-normal situation
- set status/resid in vioscsi_req_done() before the switch, so that
  the override for VIRTIO_SCSI_S_BAD_TARGET actually has effect
- g/c vioscsi_req_put(), call the appropriate cleanup routines directly
- stop initializing id/task_attr in vioscsi_req_get(), it's overwritten
  in vioscsi_scsipi_request() anyway

(jdolecek)

replace req queue offset with define, only set & start the req interrupt for
that queue; add some masks for the lun for visibility; use slot as tag number

no actual functional change, just small cleanup

(jdolecek)

Fill in some XXXs with the exact action described in them. match
lfs_valloc behaviour.

(maya)

#if DIAGNOSTIC panic ---> KASSERT

Replace some #if DEBUG by this too.  DEBUG is only for expensive
assertions; these are not.

(riastradh)

#if DIAGNOSTIC panic ---> KASSERTMSG

(riastradh)

Add UB (undefined behavior)

Seen in several commit messages

(leot)

Sync with netbsd-7-1-RELEASE

(skrll)

use printf format macros for fixed-sized integers.

(mlelstv)

Update zlib version in pkgconfig file.

(wiz)

Christos updated tzcode to 2017a

(kre)

Implement {pci,isa}_intr_disestablish

(martin)

Include the atheros hal files

(martin)

Remove `fix module_init()' entry.

I've been committed.

(nonaka)

efiboot: Kernel modules are loaded in pre-allocated memory.

(nonaka)

Fix for the "${unset-var#$(cmd1)}$(cmd2)" runs the wrong command bug.
... From FreeBSD

(kre)

Add new test case in to check (coming) bug fix for newly discovered
ash based shell bug
echo ${unset_var##$(echo a)}$(echo b)
should say "b" but instead says "a" ... the first "echo a" is not
evaluated because it cannot possibly match an unset variable, but is
not removed from the list of command substitutions, when the shell
needs to execute the 2nd cmdsub, "echo b" should be at the head of
the list, but isn't, "echo a" is still there...

This test should fail (for now) - should show 4 of 40 subtests failing.
It isn't marked as atf_expect_fail as the fix for this will be coming
later today (I will just wait at least 1 b5 build cycle so the failure
can be observed).

Detecting the bug, and the fix, are from FreeBSD.

(kre)

Do not consider using -C arg "as is" for .CURDIR unless it starts with '/'.

Reviewed by: christos

(sjg)

fix build failure.

(nonaka)

Add missing comma, else the strings are concatenated (and possibly the
wrong message is used). noted by coverity.

Linux iwlwifi strings seem similar.

(maya)

new sqlite3 tzcode

(christos)

merge 2017a

(christos)

bump libsqlite3

(christos)

new sqlite man pages

(christos)

merge 3.17.0

(christos)

fix fallout from ticket #1366: __nothing is nothing on netbsd-7

(snj)

Mmh, remove a debug printf I mistakenly added in my previous commit

(maxv)

memcpy of overlapping arguments is UB. use memmove.
CID 1299071

ok riastradh

(maya)

Fixup the catchall for the media instructions that are currently unhandled
so that swi, etc can match

(skrll)

Don't loop infinitely if the output pointer was changed
Untested, but the change done seems to match the comment above it

CID 980321

(maya)

Add the AMD 10h family, with additional events that I believe are useful,
the DTLB misses on large pages for example.

While here, remove a few K7 flags that do not actually exist on K7 (there
must have been a confusion between K7 and K8); and make the 'pmc list'
command a little more user-friendly.

(maxv)

Remove already fixed HACKS entry. correct path to makeinfo Makefile for
easy reference.

(maya)

7.1

(snj)

7.1

(snj)

Remove extra newline during attach.

> acpiec0 at acpi0 (EC, PNP0C09)
> : io 0x62,0x66
> acpiec0: using global ACPI lock

acpi_resource_parse() with &acpi_resource_parse_ops_default prints
resouce info and newline.  Found and tested on ThinkPad X21.

(tsutsui)

1384

(snj)

Pull up following revision(s) (requested by tsutsui in ticket #1384):
sys/conf/copyright: revision 1.14, 1.15
WELCOME 2016!
--
goodbye 2016, hello 2017!

(snj)

1366, 1384

(snj)

Pull up following revision(s) (requested by tsutsui in ticket #1384):
sys/conf/copyright: 1.14, 1.15
WELCOME 2016!
--
goodbye 2016, hello 2017!

(snj)

Pull up following revision(s) (requested by skrll in ticket #1366):
sys/arch/arm/include/arm32/pmap.h: 1.145
sys/arch/arm/arm32/pmap.c: 1.343, 1.344
sys/arch/evbarm/gumstix/gumstix_machdep.c: 1.58 via patch
Fixup the compile time decisions around PMAP_{INCLUDE,NEEDS}_PTE_SYNC and
fix the options for xscale boards which require the code in
pmap_l2ptp_ctor marked as #ifndef PMAP_INCLUDE_PTE_SYNC.
Fix the typo (pte -> opte) in this code block and consistently use opte
elsewhere.
PR/51990: Regression data_abort_handler: data_aborts fsr=0x406 far=0xbfffeff5  on copyout in init
--
fix unused.
--
Set xscale_cache_clean_addr appropriately and re-arrange default KVA
layout to allow direct map for all boards.
OVERO/DUOVERO/PEPPER aren't tested.
PR/52010: Regression: Gumstix Verdex is hanging in enabling cache + KASSERT ram_size

(snj)

Add ixv(4) to MP-safe component's list.

(msaitoh)

search SMBIOS from UEFI configuration table when boot with UEFI.

(nonaka)

AMA: ask me anything

(riastradh)

Add ixg(4)

(ozaki-r)