fix r1.492 bug, sorry

(knakahara)

Since it is unlikely that any of this really wants to test whether
[v]fork() produces pid==1 (init) (if it were useful to test that, an
entirely different test would be appropriate) change the tests so
they check for [v]fork() failure, as probably intended, instead.

(kre)

Improve wording and formatting.
New sentence, new line.
Fix typo.

(wiz)

add WM_RX_PROCESS_LIMIT_DEFAULT and WM_RX_INTR_PROCESS_LIMIT_DEFAULT description

pointed out by pgoyette@n.o, thanks.

(knakahara)

efiboot: fix efivar command.

(nonaka)

implement AIM(Adaptive Interrupt Moderation), but not enable yet.

Currently, it causes performance degration if enabled.
More ITR calcuration tuning is required...

(knakahara)

make kernel config option WM_RX_PROCESS_LIMIT_DEFAULT and WM_RX_INTR_PROCESS_LIMIT_DEFAULT

e.g. if
    - WM_RX_PROCESS_LIMIT_DEFAULT is set 0
    - WM_RX_INTR_PROCESS_LIMIT_DEFAULT is set -1 (means almost infinite)
that means wm(4) does not use polling mode.

(knakahara)

Provide a more robust regexp for time formats of 1day-ish

(ozaki-r)

implement polling mode for multiqueue. It can suppress performance degration at high load.

e.g. I354 IP forwarding throughput performance
    + before
      - 133Mbps forwarding performance at 200Mbps input load
      - 41Mbps forwarding performance at 800Mbps input load
    + after
      - 150Mbps forwarding performance at 200Mbps input load
      - 150Mbps forwarding performance at 800Mbps input load

(knakahara)

Pass inpcb/in6pcb instead of socket to ip_output/ip6_output

- Passing a socket to Layer 3 is layer violation and even unnecessary
- The change makes codes of callers and IPsec a bit simple

(ozaki-r)

TODO.ptrace: Remove entry about addition of TRAP_SCE and TRAP_SCX

These siginfo(2) codes have been committed.

(kamil)

Add new entry in TODO.ptrace about QPassSignals

Added:
  support QPassSignals (PT_SET_SIGPASS/PT_GET_SIGPASS) in the kernel, a way to
  stop routing a set of signals to tracer as they are uninteresting - GDB and
  LLDB expect this feature

(kamil)

Add missing opt_net_mpsafe.h.

(msaitoh)

Set PCI_INTR_MPSAFE, SOFTINT_MPSAFE, CALLOUT_MPSAFE and IFEF_START_MPSAFE
if NET_MPSAFE is defined.

(msaitoh)

IFEF_START_MPSAFE flag should be set only if WM_MPSAFE defined. some refactor.

pointed out and suggested by msaitoh@n.o, thanks.

(knakahara)

avoid using %s in UVMHIST.

(mrg)

remove debugging abort that makes tests fail.

(christos)

Add new entry in TODO.ptrace about PT_STEP with a signal

Added:
  GDB Remote Protocol expects a case with a step with a signal to be sent,
  this is currently unsupported on NetBSD

(kamil)

Add an operation to test a mount for fstrans support and use it for
_fstrans_start(), fstrans_done(), fstrans_is_owner(), vfs_suspend()
and vfs_resume().

Test for fstrans support before ASSERT_SLEEPABLE().

(hannken)

Plug a race condition on accessing i6mm_maddr

(ozaki-r)

Fix racy in6m_sol

Relook up the entry instead of reusing it, which makes locking simple.

(ozaki-r)

1374

(snj)

Pull up following revision(s) (requested by kre in ticket #1374):
doc/3RDPARTY: 1.1420 via patch
distrib/sets/lists/base/mi: 1.1155
external/public-domain/tz/dist/Makefile: up to 1.1.1.17
external/public-domain/tz/dist/NEWS: up to 1.1.1.17
external/public-domain/tz/dist/TZDATA_VERSION: up to 1.7
external/public-domain/tz/dist/Theory: up to 1.1.1.10
external/public-domain/tz/dist/africa: up to 1.1.1.10
external/public-domain/tz/dist/antarctica: up to 1.1.1.8
external/public-domain/tz/dist/asia: up to 1.1.1.16
external/public-domain/tz/dist/australasia: up to 1.1.1.11
external/public-domain/tz/dist/backward: up to 1.1.1.7
external/public-domain/tz/dist/backzone: up to 1.1.1.11
external/public-domain/tz/dist/europe: up to 1.1.1.17
external/public-domain/tz/dist/leap-seconds.list: up to 1.1.1.7
external/public-domain/tz/dist/leapseconds: up to 1.1.1.8
external/public-domain/tz/dist/northamerica: up to 1.1.1.15
external/public-domain/tz/dist/southamerica: up to 1.1.1.11
external/public-domain/tz/dist/version: up to 1.1.1.4
external/public-domain/tz/dist/zone.tab: up to 1.1.1.11
external/public-domain/tz/dist/zone1970.tab: up to 1.1.1.13
Update to tzdata2017a.

(snj)

1374

(snj)

Pull up following revision(s) (requested by kre in ticket #1374):
doc/3RDPARTY: 1.1420 via patch
distrib/sets/lists/base/mi: 1.1155
external/public-domain/tz/dist/Makefile: up to 1.1.1.17
external/public-domain/tz/dist/NEWS: up to 1.1.1.17
external/public-domain/tz/dist/TZDATA_VERSION: up to 1.7
external/public-domain/tz/dist/Theory: up to 1.1.1.10
external/public-domain/tz/dist/africa: up to 1.1.1.10
external/public-domain/tz/dist/antarctica: up to 1.1.1.8
external/public-domain/tz/dist/asia: up to 1.1.1.16
external/public-domain/tz/dist/australasia: up to 1.1.1.11
external/public-domain/tz/dist/backward: up to 1.1.1.7
external/public-domain/tz/dist/backzone: up to 1.1.1.11
external/public-domain/tz/dist/europe: up to 1.1.1.17
external/public-domain/tz/dist/leap-seconds.list: up to 1.1.1.7
external/public-domain/tz/dist/leapseconds: up to 1.1.1.8
external/public-domain/tz/dist/northamerica: up to 1.1.1.15
external/public-domain/tz/dist/southamerica: up to 1.1.1.11
external/public-domain/tz/dist/version: up to 1.1.1.4
external/public-domain/tz/dist/zone.tab: up to 1.1.1.11
external/public-domain/tz/dist/zone1970.tab: up to 1.1.1.13
Update to tzdata2017a.

(snj)

Fix a problem that m_defrag() isn't called in if_transmit path. Now both
if_start and if_transmit do m_defrag() correctly. This change improves the
performance of TSO.

(msaitoh)

Protect ia_allhosts by in_ifaddr_lock

(ozaki-r)

Make sure imo_membership is protected by inp's lock (solock)

(ozaki-r)

Protect ia6_memberships by in6_ifaddr_lock

(ozaki-r)

Make sure im6o_memberships is protected by in6p's lock (solock)

(ozaki-r)

Make usages of ifp MP-safe in some functions of IP multicast

(ozaki-r)

Fix a bug that TSO doesn't work correctly. This bug was added in
ix_txrx.c rev. 1.17.

(msaitoh)

drmkms_pci: Use MSI if available (Take 2)

tested by martin@ and yhardy@

(nonaka)

CardBus capitalization police

(snj)

Use LIST_* macros

No functional change.

(ozaki-r)

ifdef reduction

(christos)

Make compile again without "options WAPBL".

From John D. Baker via current-users@, slightly modified by me.

(hannken)

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

(skrll)

loadfile: in the mmap case ensures that the buffer ends with \n
do the same for the non-mmap case - avoids core dump.

(sjg)

Fix compile... is this right?

(skrll)

Improve appearance (a little) when read with nroff, rather than troff

(kre)

Can't profile ras_atomic_cas_noupdate

(skrll)

Protect genfs_getpages() against vnodes disappearing during a
forced mount update from read-write to read-only.

(hannken)

Bring back read-write to read-only mount update for ffs.

(hannken)

Add a diagnostic test for buffers written to a block device holding
a read-only mounted file system.

This will become a KASSERT in the near future.

(hannken)

Suspend the mounted file system while updating.

(hannken)

Change the protocol to update a mounted file system from read-write
to read-only and vice versa:

- Add an internal flag IMNT_WANTRDONLY.
- Set either IMNT_WANTRDWR or IMNT_WANTRDONLY if going from or to read-only.
- After successfull call to VFS_MOUNT() set or clear MNT_RDONLY.

Adapt tmpfs and rumpfs to the new protocol.  Other file systems will be
updated when they get the IMNT_CAN_RWTORO property.

Welcome to 7.99.64

(hannken)

Must always lock the parent -> lock the child -> unlock the parent.

(hannken)

Remove now redundant calls to fstrans_start()/fstrans_done().

(hannken)

Remove now redundant calls to fstrans_start()/fstrans_done().

(hannken)

Correct the date of PT_GETDBREGS and PT_SETDBREGS introduction

Feb 28th -> Feb 23rd

Sponsored by <The NetBSD Foundation>

(kamil)

Document PT_GETDBREGS, PT_SETDBREGS, TRAP_SCE and TRAP_SCX in CHANGES

Added entries:
      ptrace(2): Add new API replacing PT_WATCHPOINT for Debug Registers:
              PT_GETDBREGS and PT_SETDBREGS
      siginfo(2): Add new si_code values for SIGTRAP: TRAP_SCE and TRAP_SCX

Sponsored by <The NetBSD Foundation>

(kamil)

Make IPv6 multicast MP-safe partially

To complete the task, we need to make users of IPv6 multicast MP-safe, for
example socket/PCB and CARP.

(ozaki-r)

refactor: functionalize ITR/EITR writing processing.

tested I354 and 82545(VMware e1000).

(knakahara)

Provide in6_multi_group

Use it when checking if we belong to the group, instead of in6_lookup_multi.

No functional change.

(ozaki-r)

divide ITR/EITR value from wm_softc to wm_queue as EITR register is bound to each H/W queue.

(knakahara)

Mark syscall1 broken in t_ptrace_wait*

Mark this test broken with kern/52012 as the tracee does not stop on syscall
entry. Check for siginfo(2) in syscall1 and unveil two new bugs.

PR kern/52012: PT_SYSCALL does not stop on syscall entry
PR kern/52018: ptrace(2) PT_SYSCALL generates SIGTRAP without si_code
PR kern/52019: ptrace(2) PT_SYSCALL generate SIGTRAP to the parent but
              returns SIGSTOP in PT_GET_SIGINFO

Sponsored by <The NetBSD Foundation>

(kamil)

tzdata2017a: new timezone America/Punta_Arenas

(kre)

Restore/add some softnet_lock for nd6_rt_flush and defrouter_addreq

May help PR kern/52015

(ozaki-r)

Note update to tzdata2017a

(kre)

Merge tzdata2017a

(kre)

Add new SIGTRAP types: TRAP_SCE and TRAP_SCX

New entries:
- TRAP_SCE        7      /* Process syscall entry trap          */
- TRAP_SCX        8      /* Process syscall exit trap            */

Sponsored by <The NetBSD Foundation>

(kamil)

change DIOCCACHESYNC to not issue the FLUSH command at all when the controller
has no volatile write cache

(jdolecek)

implement DIOCGCACHE

(jdolecek)

in cpu_switchto() and the fast-softint context switch code,
put back the stwcx. instruction to clear the reservation.
we used to have this in the old cpu_switch() until it was
if-0'd in 2003 and removed completely in 2007.
this fixes hangs I've seen where a softint thread is
blocked waiting for a mutex that is not held.
this should also fix PR 44387.

(chs)

typo

(skrll)

Mark resume1 and syscallemu1 tests broken in t_ptrace_wait*

resume1:
    PR kern/51995 ptrace(2) PT_RESUME is not reliable

syscallemu1:
    PR kern/52012 PT_SYSCALL does not stop on syscall entry

Sponsored by <The NetBSD Foundation>

(kamil)

Add new test syscallemu1 in t_ptrace_wait*

syscallemu1:
    Verify that exit(2) can be intercepted with PT_SYSCALLEMU

This test is failing as PT_SYSCALL does not stop the process on syscall
entry. This renders PT_SYSCALL broken on NetBSD.

Sponsored by <The NetBSD Foundation>

(kamil)

Sync TODO.ptrace with reality

Remove entries:
- add new ptrace(2) calls to lock (suspend) and unlock LWP within a process
- switch PT_WATCHPOINT* to PT_*ETDBREGS and document it, add ATF tests
- add ATF tests for PT_SYSCALL and PT_SYSCALLEMU

Sponsored by <The NetBSD Foundation>

(kamil)

fix EITR setting.

    + 82574
      - add EITR setting, that is MSI-X mode interrupt interval
    + 82575
      - fix EITR value. 82575's EITR usage is the same as legacy (not NEWQUEUE)
        controllers
      - apply workaround which 82575's EITR does not have CNT_INGR bit
    + other NEWQUEUE controllers
      - fix interrupt interval field. NEWQUEUE (include 82575) controllers'
        interrupt interval field is 2:14
      - use CNT_INGR bit which avoid to overwrite counter

tested 82574, 82575 and I354.
ok by msaitoh@n.o.

(knakahara)

Add tests that destroy an interface while the mcast program is running

(ozaki-r)

remove unnecessary substitution. no functional changes.

(rin)

determine the buffer size by roundup(9)

(rin)

Remove PARANOID macro; we should always update secsize if necessary.

(rin)

Separate the code of joining multicast groups

No functional change.

(ozaki-r)

Add tests for loopback interface

(ozaki-r)

Prevent ia6 from being freed in in6_ifinit

It fixes a panic (diagnostic assertion "entry->ple_prevp != NULL" failed)
on:
  ifconfig lo1 create
  ifconfig lo1 127.0.0.2
reported by ryo@

(ozaki-r)

pi_bsize must be at least pi_secsize

Allows block device accesses to 4KiB logical sector disks to function on the
vast majority of ports with 2KiB BLKDEV_IOSIZE.

(jakllsch)

Revert "drmkms_pci: use MSI if available."

Fix PR/51997.

(nonaka)

minor clean ups:
- mark a bunch of local things static
- export an explicit set of functions for audiobell, instead of hooking
  directly into the device callbacks, and avoid non-shared headers for
  the same prototypes
- s/audioprobe/audiomatch/
- use __func__

(mrg)

remote extra pasted line

(jdolecek)

Welcome to 7.99.63 - struct ld_softc changes

(jdolecek)

refactor the ld(4) DIOCCACHESYNC hook into general ioctl hook, so that attachments
would be able to implement arbitrary other ioctls

(jdolecek)

pass also DIOCGCACHE to underlying device, so that upper layers would be able
to get the device cache properties without knowing the topology; while here also
pass down DIOCGSTRATEGY for neater dkctl(8) output

(jdolecek)

pass also DIOCGCACHE to underlying device, so that upper layers would be able
to get device cache properties without knowing the topology

(jdolecek)

Avoid __ffs64(0).

From maya@.

(riastradh)

tag the powerpc hackish_return_address() as __noinline.

(chs)

check the lower bits of the stack pointer,
not the lower bits of the upper half of the stack pointer.

(chs)

Flesh out missing functions.  Get IO space and interrupts working.

Successfully tested with RTL8029 ne(4)@pci(4).

(jakllsch)

Avoid __ffs64(0).

(riastradh)

fix printf formats (clang build)

(christos)

If lp->d_secsize is updated, we need to reallocate the buffer.
Thanks mlelstv for his comment on port-amiga.

(rin)

Decode AGP capability.

(msaitoh)

Attach the usb2 bus - missed in merge from nick-nhusb

(skrll)

Audio passes the atf test again.

(nat)

fix Tx queue assignment when the number of H/W queues is less than ncpu.

(knakahara)

Make CARP on IPv6 work

It passes ATF tests but no more, no less.

(ozaki-r)

Test handovers on interface down as well as server halt

(ozaki-r)

Add a test case for CARP on IPv6

The test case fails expectedly because the implementation of CARP on IPv6
is incomplete yet.

(ozaki-r)

add glint driver debuginfo file.

(chs)

in mips_emul_fp(), clear all pending FP exceptions rather than
just a particular one, otherwise the kernel can take another
FPU trap when it writes back the new FCSR value.
discovered via the fesetround() bug that wrote garbage to the FCSR.

(chs)

fix fesetround() to set the FPSR to the desired value rather than
a pointer to a local variable.  wrap the asm in inline functions so that
the compiler can do type checking for us.

(chs)

the second operand to cfc1/ctc1 isn't an FPU data register
so don't make it look like one.

(chs)

the FP_* rounding constants need to be different from the new FE_* constants
to preserve the ABI, so shift them as needed when using them.

(chs)

fix asm operands: "ctc1" uses the register as an input, not an output.

(chs)

update the FE0/FE1 MSR bits as needed when changing the exception mask.

(chs)

have fpsetmask() change the FE0/FE1 MSR bits to precise mode if any
FP exceptions are enabled.  fix the kernel emulation of mfmsr and mtmsr
to use the correct opcodes for these instructions.  ignore PSL_FE
(the FP enable bit) in the MSR that a user program tries to set,
since it will naturally be set for FP-using processes but
we can't let the user process manage that bit.

(chs)

the FP_* exception constants need to be different from the new FE_* constants
to preserve the ABI, so shift them as needed when using them.

(chs)

fix the build of this kernel config by adding the omapscm driver.
turn on options for dtrace, turn off USB debugging.

(chs)

the MI lrint() code assumes that rounding is done in at most double precision
but m68k (68881) uses extended precision by default, so put the FPU in
double-precision mode temporarily here.

(chs)

the FP_* constants need to be different from the new FE_* constants
to preserve the ABI, so shift them as needed when using them.

(chs)

use FPCR instead of FPSR to manipulate the exception mask.

(chs)

apply changes equivalent to freebsd r256822, r267759 and r298171:
fix the dtrace "args" variable to return correct values.
simplify the invop handlers a bit by changing the second argument
from a pointer into the stack to a pointer to the trapframe.

(chs)

Fix the default value of rltime

According to rtadvd.conf(5), the default value of rltime is 1800 seconds.

PR bin/51994

(ozaki-r)

regen.

(msaitoh)

Add some NVIDIA devices.

(msaitoh)

got 'bad mbi comp_stat 5' during DIOCGCACHE under Parallels, according
to linux driver it's possible status when controller thinks the CCB is invalid;
handle same as BHA_MBI_ABORT/BHA_MBI_UNKNOWN i.e. just clear the CCB anyway

(jdolecek)

mark local functions static, similar to other ld(4) attachments

(jdolecek)

Fix Dd argument.

(wiz)

shorten hardware address wording

(mbalmer)

Paranoia: parenthesize macro operands.

Should be no functional change, based on how these macros are used in
this file.

(riastradh)

Clarify that kauth_cred_get doesn't modify with reference count.

(riastradh)

Comment out DISKLABEL_RDB as it is too specific for general usage.
(Also add commented out that entry for GENERIC on amd64; I forgot it)

(rin)

Remove commented DISKLABEL_EI for strictly little endian architectures

It will never be useful.

(maya)

add DKWEDGE_METHOD_RDB

(rin)

Add DKWEDGE_METHOD_RDB option, which is enabled for x86, commented out for
other platforms by default.

(rin)

Add DKWEDGE_METHOD_RDB option, which is Amiga Rigid Disk Block (RDB)
partitioning detection method for dk(4).

(rin)

s/u_char/unsigned char/

(rin)

Include <sys/types.h> and replace u_long with uint32_t in preparation to be
used from LP64 environments.

(rin)

Add a DEBUG assert for the problem noted in PR kern/51731
of some drivers not sufficiently locking.

The relevant drivers might not work, but if someone gets around to them
then the problem could be visible.

sd doesn't trip the assert.

(maya)

catch up with i386 rename of gdt to gdtstore
fixes i386 ALL kernel build (pnpbios is disabled in GENERIC)

(maya)

Apply fix from natuve clock_nanosleep(2) to not copyout remaining time
struct if TIMER_ABSTIME flag is set.

(njoly)

Disks without a disk label happen in normal operation (both as USB
stick or with GPT), so do not cause an error message when opening
them.

(martin)

Trailing whitespace

(skrll)

Use PEX numbers consistently.  No functional change.

(skrll)

- Fix the base power's value in the power budget capability.
- Fix typo (s/Maximun/Maximum/)

(msaitoh)

multiply the missing factor to offset when sorting partitions
(harmless for ``normal'' disks although)

(rin)

fix off-by-one

(rin)

Switch from __ABICALLS__ to __mips_abicalls like upstream GCC does in
the generic MIPS target logic.

(joerg)

kill stray q

(christos)

fix unused.

(christos)

make bitfields unsigned

(christos)

Change reference from malloc(3) to jemalloc(3) as the MALLOC_OPTIONS variable
is actually described in jemalloc(3).

(abhinav)

pacify disklabel validation message

(mlelstv)

Garbage collect unneeded inclusion of <x86/dbregs.h> in <machine/proc.h>

This is left over after introduction of Debug Register accessors.
This interface replaced older watchpoint API.

Sponsored by <The NetBSD Foundation>

(kamil)

Use correct lock for session wakeup.

(mlelstv)

Add pci attachment for mvsata

(skrll)

EFI console is drawing faster by shadowfb.

(nonaka)

genfb(4): Enabling/Disabling shadowfb can be controlled via prop.

(nonaka)

fix printf format

(christos)

Typo in a model string

(skrll)

Remove use of sc_intr_lock from audio_read.  Fixes LOCKDEBUG panics.

(nat)

Only do DaD if the interface actually has the address.

(roy)

Remove RSS's note in BUGS section because it's supported now.

(msaitoh)

fix printf format.

(christos)

efiboot: Don't access old stack after copying a kernel.

It's possible that the old stack is overwritten by the kernel.

(nonaka)

apply 82574 RFCTL workaround the same as FreeBSD and linux.

In fact, this workaround reduces interrupt count.

(knakahara)

82574 should follow MSI-X mode IMS manner.

Howerver some legacy interrupts occur as MSI-X other interrupt yet...

(knakahara)

Simplify locking.  Hold intr lock only when dealing with sc_pr (mix ring)
or virtual channel 0 (hardware).

(nat)

Fix a locking botch in previous

(skrll)

Use macro. KNF. No functional change.

(msaitoh)

Add new test syscall1 in t_ptrace_wait*

syscall1:
    Verify that getpid(2) can be traced with PT_SYSCALL

Enforce usage of syscall(2), it should prevent failing on any possible
optimizations in future as a libc can ship this information not through
syscall (shared struct with kernel or similar).

Sponsored by <The NetBSD Foundation>

(kamil)

Print mac.type and the TrackID.

(msaitoh)

Fix 0x%u...

(msaitoh)

refresh

(christos)