match the macro name to the order of the arguments. NFC.

(ryo)

fixed mask width of DBGWVR_MASK, and added definition of DBGBVR_MASK

(ryo)

fix build error without options DDB.

kvtopte() is referenced from arm/acpi/acpi_machdep.c

(ryo)

regen

(ryo)

Add ASUSTEK AURALED

(ryo)

Add CoolerMaster, and AMD SR4 lamplight Control device (Wraith Prism RGB CPU Cooler)

(ryo)

Tickets #1227 - #1230

(martin)

Pull up following revision(s) (requested by sborrill in ticket #1230):

etc/rc.d/mountall: revision 1.15

Need to explicitly load value of zfs variable as zfs=YES may be set in
/etc/rc.conf.d/zfs, not /etc/rc.conf.

(martin)

Ticket #1662

(martin)

Pull up following revision(s) (requested by christos in ticket #1662):

sys/netinet/tcp_subr.c: revision 1.286
sys/netinet/tcp_timer.c: revision 1.96
sys/netinet/in_var.h: revision 1.102
sys/netinet/in_var.h: revision 1.99

Don't increment the iss sequence on each connection because it exposes
information (Amit Klein)

Add some randomness to the iss offset

Use a random IPv4 ID because the shuffling algorithm used before could expose
information (Amit Klein)

mv <sys/cprng.h> include to the kernel portion

(martin)

Pull up following revision(s) (requested by christos in ticket #1229):

sys/netinet/tcp_subr.c: revision 1.286
sys/netinet/tcp_timer.c: revision 1.96
sys/netinet/in_var.h: revision 1.102
sys/netinet/in_var.h: revision 1.99

Don't increment the iss sequence on each connection because it exposes
information (Amit Klein)

Add some randomness to the iss offset

Use a random IPv4 ID because the shuffling algorithm used before could expose
information (Amit Klein)

mv <sys/cprng.h> include to the kernel portion

(martin)

Pull up following revision(s) (requested by tsutsui in ticket #1228):

sys/dev/ic/mb89352.c: revision 1.58

Fix a possible race condition in spc_msgin() in NO_MANUAL_XFER case.

To avoid the race, check SSTS and INTS after XFR command as
spc_pio_datain() does.

Reported from isaki@, observed on nono emulator.

(martin)

Pull up following revision(s) (requested by 1227 in ticket #nia):

external/mit/ctwm/etc/system.ctwmrc: revision 1.2
external/mit/ctwm/etc/system.ctwmrc: revision 1.3
external/mit/ctwm/etc/system.ctwmrc: revision 1.4
external/mit/ctwm/etc/system.ctwmrc: revision 1.5
external/mit/ctwm/etc/system.ctwmrc: revision 1.7
external/mit/ctwm/etc/system.ctwmrc: revision 1.9
external/mit/ctwm/etc/system.ctwmrc: revision 1.10
external/mit/ctwm/etc/system.ctwmrc: revision 1.11
external/mit/ctwm/etc/system.ctwmrc: revision 1.12

Make it easier to identify the current active window (red border highlight)

Bump up the dark contrast a bit

xeyes should be AlwaysOnTop - requested by a user

mute binding is annoying when there's hardware mute

slightly lower workspace manager size
this works nicely on a 12.5" FHD screen

xclock also gets AlwaysOnTop

system.ctwmrc: Set Cursors to avoid strange default fallbacks

ctwmrc: remove SloppyFocus keyword, which breaks xev
the default still seems to be sloppy focus.
is this a bug? does ctwm use a different definition of sloppy focus to
the rest of the world? we may never know

ctwmrc: disable NoTitleFocus to prevent multiple windows having focus
some investigation help from Rhialto

(martin)

PR/50933: Uwe Toenjes: Document hardware limitations

(christos)

Move the offset addition in one place and mask the random generated value
to make sure that the isn is monotonic.

(christos)

Need to explicitly load value of zfs variable as zfs=YES may be set in
/etc/rc.conf.d/zfs, not /etc/rc.conf.

(sborrill)

Improve wording, use more macros.

(wiz)

Modify some parameters to reduce packet dropping.

- Background: ixgbe doesn't use common MCLGET() interface and use the
  driver specific cluster allocation mechanism (jcl). The cluster is
  pre-allocated with a fixed number and the current number per queue
  is num_rx_desc * 2 (2048*2=4096). It's too small. It also has a problem
  that the max length of the pcq which is used in the TX path is big
  (4096). Example:

    100M <----- [ixg0                                      ixg1] <----- 1G
                2048 TX descs <--- 4096 pcqs <---- 2048 RX descs

  If a machine forwards a traffic from 1G interface to 100M interface,
  It would require 2048+4096+2048=8192 descriptors, but the current number
  is 2048*2=4096. It's too small. Even if the both interface's link speed
  is the same and only small number of packet is queued in the pcq, 4096
  jcl is small because 2048(RX)+TX(2048)=4096. If jcl is exhausted, not only
  forwarding from ixg1 to ixg0 is dropped, but also another forwarding path
  from ixg1 to another interface(e.g. wm0) is also dropped. Sockets also
  queue packets, so if a lot of sockets are used and/or a socket buffer
  size is changed to bigger one, it'll also become a problem. If the jcl
  is exhausted, evcnt(9) counter "ixgX qY Rx no jumbo mbuf" is incremented.
  Example:
    vmstat -ev | grep ixg1 | grep "no jumbo"
    ixg1 q0 Rx no jumbo mbuf                          0    0 misc
    ixg1 q1 Rx no jumbo mbuf                          0    0 misc
    ixg1 q2 Rx no jumbo mbuf                      141326    0 misc
    ixg1 q3 Rx no jumbo mbuf                          0    0 misc

- To solve this problem:
  - Add new config parameter IXGBE_JCLNUM_MULTI and set the default to 3
    (2048 * 3). The minimum number is 2. The total number of jcl per queue
    is available with hw.ixgN.num_jcl_per_queue sysctl.
  - Reduce the max length of the pcq() which is used in the TX path from
    4096 to 2048.

- Reviewed by knakahara@ and ozaki-r@.

- TODO: Use MCLGET().

(msaitoh)

add support for the CPU temperature sensor found in iMac G5s

(macallan)

We don't need libwrap anymore; libhack replaces it.

(christos)

Add a stubbed version of libwrap (tcpwrappers)

(christos)

Adjust for fewer args in calling functions

(christos)

indent: remove redundant initializer in dump_line

No functional change.

(rillig)

indent: move comment about dump_line to column 1

It looked misplaced on the right side since that area is usually
reserved for small remarks, not long explanations.

No functional change.

(rillig)

indent: always use braces in do-while loops

Having a 'while' at the beginning of a line looks as if it would start a
loop.  It's confusing when it _ends_ a loop instead.

(rillig)

tests/indent: demonstrate wrong removed empty line before '//'

(rillig)

mv <sys/cprng.h> include to the kernel portion

(christos)

remove extra line

(christos)

indent: split bsearch comparison function

It may have been a clever trick to use the same memory layout for struct
templ and a string pointer, but it's not worth the extra comment and
difficulty in understanding the code.

No functional change.

(rillig)

tests/indent: demonstrate another bug in nested blocks

(rillig)

Update libelf's manual pages to upstream revision [r3928].

(jkoshy)

remove ip{,6}_id.c

(christos)

indent: inline macro for backslash

No functional change.

(rillig)

indent: convert big macros to functions

Each of these buffers is only modified in a single file.  This makes it
unnecessary to declare the macros in the global header.

(rillig)

tests/indent: add test for output line counting

(rillig)

welcome to 9.99.81, ip_randomid() signature change.

(christos)

reinstate a simple version of ip_randomid()

(christos)

tests/indent: allow golden stderr file, improve filenames, clean up code

When indent runs in filter mode, it may output messages to stderr.
Allow tests with non-empty expected stderr.

In the ATF output, the filename 'output_file.parsed' was not helpful for
casual readers of diff output since they expect the filenames to be
meaningful.  Embed the name of the test case in that filename.

Fix quoting of the shell variables.

Remove the repetition of the regular expression to clean up the test
files.

(rillig)

Appease new lint enum errors

(christos)

indent: make it easy to compile indent in debug mode

(rillig)

indent: fix printing of uninitialized 'token' in debug output

(rillig)

no need for ip6_id.c...

(christos)

Remove the unused "addin" argument (it was always 0) and go back using
a random iss by default (instead of rfc1948)

(christos)

remove now unused pseudo-random ip id code.

(christos)

Add some randomness to the iss offset

(christos)

Use a random IPv4 ID because the shuffling algorithm used before could expose
information (Amit Klein)

(christos)

Don't increment the iss sequence on each connection because it exposes
information (Amit Klein)

(christos)

give names to the enums so we can cast by name for lint

(christos)

cast to the proper enums for lint

(christos)

Remove incorrect byte and word conversions from fpu_explode.
The correct operation here is arithmetic right shift, but nobody calls it.

(isaki)

Add comments for 2711 local peripherals

(mlelstv)

Move interrupt register definitions to driver.

(mlelstv)

Undocumented register to multiplex emmc2 pins to legacy sdhc

(mlelstv)

Add some firmware support functions and define more properties
- framebuffer
- power management
- vchiq init
- RPI4 USB firmware support

(mlelstv)

RPI4 PCIe driver, based on pcihost_fdt.c

(mlelstv)

Interrupt no longer needs kernel lock.

(mlelstv)

Compute CRC for all segments of a multi-buffer packet.
Add interrupt mitigation for transmit and receive.
Use separate transmit lock.
Fix some error paths.

(mlelstv)

Use no_jmbuf evcnt for the failure case of ixgbe_getjcl().

(msaitoh)

Add bluetooth also to 32bit kernel.

(mlelstv)

early console for pico-pi-imx7

(mlelstv)

always re-read status register in the loop.

(mlelstv)

indent: fix handling of '//' end-of-line comments

(rillig)

fix missing chars

(christos)

indent: sprinkle a few const

No functional change.

(rillig)

indent: remove redundant parentheses around return value

No functional change.

(rillig)

lint: move keyword 'continue' over to the other control flow keywords

No functional change since neither rw_jump nor rw_inline_or_restrict is
mentioned in any switch statement, and lint didn't find any other
suspicious enum operations.

(rillig)

indent: use named constants for the different types of keywords

This reduces the magic numbers in the code.  Most of these had their
designated constant name written in a nearby comment anyway.

The one instance where arithmetic was performed on this new enum type
(in indent.c) was a bit tricky to understand.

The combination rw_continue_or_inline_or_restrict looks strange, the
'continue' should intuitively belong to the other control flow keywords
in rw_break_or_goto_or_return.

No functional change.

(rillig)

lint: do not allow assignment-expression for constant-expression

It's a funny idea to do something like 'case a = 13:', but since any
compiler will reject this code, there is no point in lint supporting it.

No functional change since everywhere the grammar allows a constant
expression, there is no ambiguity where an assignment could be
interpreted differently.

(rillig)

lint: rename grammar rule 'constant'

C99 6.4.4 already defines a grammar rule named 'constant' for an number
literal or an enum constant, so don't use that name for something else.

No functional change.

(rillig)

lint: align names of C grammar rules with C99

No functional change.

(rillig)

lint: in strict C mode, warn about initialization with '[a ... b]'

https://gcc.gnu.org/onlinedocs/gcc/Case-Ranges.html

(rillig)

Ticket #1661

(martin)

Pull up following revision(s) (requested by christos in ticket #1661):

sys/netinet6/ip6_id.c: revision 1.19-1.21
sys/netinet6/ip6_var.h: revision 1.88
sys/netinet/ip_input.c: revision 1.400
sys/netinet/tcp_subr.c: revision 1.285
sys/netinet/ip6.h: revision 1.30

netinet: Enable random IP fragment ids by default (from riastradh)

netinet: Enable RFC 1948 pseudorandom TCP ISS selection by default.
(from riastradh)

netinet6: Mark randomid unused.

Will make merging and bisection easier if anything goes wrong with
flow label or fragment id randomization changes.
(from riastradh)

netinet/netinet6: Add necessary includes to make these standalone.
(from riastradh)

Replace randomid() by cprng_fast32()

(martin)

Tickets #1225 and #1226

(martin)

Pull up following revision(s) (requested by christos in ticket #1226):

sys/netinet6/ip6_id.c: revision 1.19-1.21
sys/netinet6/ip6_var.h: revision 1.88
sys/netinet/ip_input.c: revision 1.400
sys/netinet/tcp_subr.c: revision 1.285
sys/netinet/ip6.h: revision 1.30

netinet: Enable random IP fragment ids by default (from riastradh)

netinet: Enable RFC 1948 pseudorandom TCP ISS selection by default.
(from riastradh)

netinet6: Mark randomid unused.

Will make merging and bisection easier if anything goes wrong with
flow label or fragment id randomization changes.
(from riastradh)

netinet/netinet6: Add necessary includes to make these standalone.
(from riastradh)

Replace randomid() by cprng_fast32()

(martin)

Pull up following revision(s) (requested by isaki in ticket #1225):

sys/dev/usb/uaudio.c: revision 1.169

Fix my copy-and-paste bug in rev1.160.

This fixes recording sample dropout.

(martin)

lint: fix off-by-one error in 'case 3...5'

According to the GCC documentation[1], the high end of the range is
inclusive as well, which makes sense since otherwise there would be no
way of specifying a range that includes the maximum representable
number.

Since the range is not used at all in the code, none of the tests could
possibly fail.

[1] https://gcc.gnu.org/onlinedocs/gcc/Case-Ranges.html

No functional change.

(rillig)

lint: inline 'case 3...5' in grammar

There was no point in having a separate grammar rule for the '3...' part
of a range expression, it just made the code more complicated than
necessary.

No functional change.

(rillig)

lint: in the name of an array type, list the dimension first

before: array of unsigned int[4]
now:    array[4] of unsigned int

Listing the array dimension first keeps it in contact with the keyword
'array'.  This reduces confusion, especially for nested arrays.

(rillig)

tests/lint: demonstrate wrong warning for array subscript

Seen in usr.bin/indent/lexi.c.

(rillig)

G12

(nia)

add TRAP_SIGDEBUG support to mips.

(christos)

Add blocklist support to libwrap which enables all programs using libwrap
to block access from hosts we deny. (libwrap support from Greg A. Woods)

(christos)

Amend missed messages:

netinet6: Pick IPv6 fragment ids uniformly at random.

Expected number of packets before collision is ~2^16, about the same
as we get for IPv4 with alternating disjoint random cycles.  Keep it
simple unless we determine we really need something much better for
IPv6 than what IPv4 can achieve anyway.

netinet6: Rip out now-unused IPv6 fragment id logic.

(from riastradh)

(christos)

netinet/netinet6: Add necessary includes to make these standalone.
(from riastradh)

(christos)

netinet/netinet6: Add necessary includes to make these standalone.
(from riastradh)

(christos)

netinet6: Mark randomid unused.

Will make merging and bisection easier if anything goes wrong with
flow label or fragment id randomization changes.

(from riastradh)

(christos)

netinet: Enable RFC 1948 pseudorandom TCP ISS selection by default.
(from riastradh)

(christos)

netinet: Enable random IP fragment ids by default (from riastradh)

(christos)

Sync with Elftoolchain [r3916]: update the list of known EM_* values.

(jkoshy)

For LP64, remove members of struct cpu_info that exist just for
compatible with booke or ibm4xx. Even if MODULAR || _MODULE,
these members are useless for powerpc64.

(rin)

For _LP64, disable members of struct mdproc just for compatible with
booke and ibm4xx, even if MODULAR or _MODULE is defined.

Fix build failure for evbppc64 due to CTASSERT in COND_SET_STRUCT,
which restricts sizeof(struct mdproc) < 31 for unknown reason...

(rin)

Remove an '#ifndef hp300' and make STI_{EN,DIS}ABLE_ROM simply check the
sc_{en,disable}_rom methdos provide by the bus attachment glue.

(skrll)

Intoduce "-n" switch.

Running btat2dpd with "-n" especially for use with pad(4) will stop bta2dpd
from exiting upon a read error from pad(4) devices.

This is to compenstate for the behaviour in NetBSD current that pad(4) will
no longer output 0's when its corresponding audio(4) device is not active.

I believe that this new pad(4) behaviour is not present in -9 and -8.

While here I've improved its function as a daemon and it should reconnect to
bluetooth speakers when they are disconnected/reconnected.

Reported and tested by reinoud@ and maya@

(nat)

Stop advertising more openings after limit is reached.

(mlelstv)

indent: in debug mode, output detailed token information

The main ingredient for understanding how indent works is the tokenizer
and the 4 buffers in which the text is collected.

Inspecting this debug log for the test comment-line-end makes it obvious
why indent messes up code that contains '//' comments.  The cause is
that indent interprets '//' as an operator, just like '&&' or '||'.  The
sequence '/////' is interpreted as a single operator as well, by the
way.

Since '//' is interpreted as an ordinary operator, any words following
it are plain identifiers, usually several of them in a row, which is a
syntax error.  Depending on the context, the operator '//' is either a
unary operator (no space around) or a binary operator (space around).
This explains why the word 'line-end' is expanded to 'line - end'.

No functional change outside of debug mode.

(rillig)

indent: for the token types, use enum instead of #define

This makes it easier to step through the code in a debugger.

No functional change.

(rillig)

indent: use all headers in all files

This is a prerequisite for converting the token types to an enum instead
of a preprocessor define, since the return type of lexi will become
token_type.  Having the enum will make debugging easier.

There was a single naming collision, which forced the variable in
scan_profile to be renamed.  All other token names are used nowhere
else.

No change to the resulting binary.

(rillig)

Build Xorg server for evbppc with minimum wsfb(4) support for Explora.

(rin)

Add initial support for Xorg server for wsfb(4):

- Fix WSDISPLAYIO_GTYPE ioctl.

- Support WSDISPLAYIO_LINEBYTES, _{GET,PUT}CMAP, and _SMODE ioctl.

As well as tiny improvements related to it:

- Use actual size of framebuffer for bus_space_mmap(9), as well as upper
  bound for mmap(2).

- Use base address obtained from autoconf(9) layer for mmap(2), instead of
  hard-coded one.

- Use rasops_cmap as default color map, instead of our own ANSI color map.

- Show monitor resolution and color depth instead of size as tty.

I've confirmed that mlterm-fb from pkgsrc/x11/mlterm also works fine.

(rin)

tests/indent: add tests for the remaining command line options

(rillig)

Use bus_space_mmap in sti_mmap.  This makes X on sti for my B160L nearly
work (the colour map is somewhat wrong), but at least the machine doesn't
wedge.

(skrll)

Provide a bus_space_mmap that always fails.  Nothing uses it yet, but it
will be very soon.

OK tsutsui@

(skrll)

Misc cleanup for clarity. No functional changes.

- Group function declarations, and reorder function bodies accordingly.
- Misc KNF.

(rin)

No need to clear framebuffer here; rasops(4) takes care of it.

(rin)

Reduce ifdef. No functional changes.

(rin)

Revert this commit:

http://www.nerv.org/netbsd/?q=id:20210305T064016Z.410b0e33273c3a62110d09e8bb27939f6a8bab38

> Map framebuffer to virtual space when fb(4) is present,
> even if it is not console.

Direct mapping for framebuffer is required only if fb(4) is console.

XXX
It would be nice if we can remove reserved TLB entries after
fb(4) is properly attached.

(rin)

tests/indent: allow for comments in either the input or output files

(rillig)

BER

(nia)

<tab> consistency

(skrll)

Build modules for PPC_OEA64 on evbppc64.

(rin)

whitespace --> tab. No functional changes.

(rin)

distrib/amd64/cdroms: honor CDBOOTOPTIONS when installing EFI bootloader

(dbj)

Remove unused/unreferenced legacy intrcnt/intrnames.

(simonb)

TODO and ipfilter improvements from Greg A. Woods

(christos)

- Enable WARMSTART
- Pass fd to libwrap
- Add blocklist
>From Greg A. Woods

- Tidy up debugging

(christos)

tests/indent: add tests for negative options

Since most of these options are of the form "don't do anything", the
input and output of these tests is very similar.

(rillig)

tests/indent: add more tests, discover more bugs

(rillig)

tests/indent: add 17 more tests, found only a single new bug

Discovering one bug in 17 command line options is acceptable.  This
compensates the bad first impression I got in the previous batch of
tests, which consisted of 11 tests and found 8 bugs.

(rillig)

indent.1: sort options alphabetically

(rillig)

indent: fix space-tab alignment in indent's own code

These parts are not fixed automatically by indent since they are in box
comments.

No functional change.

(rillig)

Don't or the got object in, just assign it. Fixes cc -mabi=64 hello.c -lcrypto
Now all 64 bit binaries work, and I tested that we did not break the 32 bit
ones
XXX: That 0x80000000 seems wrong for 64 bit stuff.

(christos)

pretend we know about %# and %j...

(christos)

tests/indent: fix unintended trailing whitespace and space-tab

(rillig)

tests/indent: add 11 test cases, demonstrate 8 bugs

(rillig)

return early from identify_hypervisor() if we already know we're running
Xen PV or PVH, as this was before 1.119.
Trying to read the BIOS faults (as expected, as there's no BIOS in this case).
Problem pointed out and fix tested by Brian Marcotte

(bouyer)

tests/indent: add templates for options tests

Given that indent "has even more switches than ls(1)", there are far too
few tests.  To make it easier to add meaningful tests for each of the
options, add the templates for the tests right now, ready to be filled
in.

(rillig)

tests/indent: extend test for '//' comments with more examples

(rillig)

Fix trivial typo. No need to bump date.

(rin)

Minimal (but hackish) change to make a DEBUG kernel compilable.

(martin)

Improve fdt_isprint so that it returns false if there are consecutive
zeroes.

(skrll)

For ibm4xx and booke, restore saved PSL_EE bit as done for oea,
instead of forcibly enabling interrupts.

(rin)

Convert to KASSERTMSG(9) to display wrong IPL.

(rin)

Reorder things in cpu_startup() for clarity.

No functional changes intended.

(rin)

For booke and ibm4xx, switch to software-based single-stepping for PT_STEP
ptrace(2) command from broken hardware-based implementation.

As described in proposal on port-powerpc@,

http://mail-index.netbsd.org/port-powerpc/2021/02/26/msg003597.html

hardware debug facilities of booke and 4xx use critical interrupts, that
are difficult to handle for this purpose; they are not automatically masked
when entering kernel mode via system call trap or hardware interrupt.
See my proposal above for more details.

Now, hardware debug facilities are exclusively usable by kernel itself.
They are much more functional than PSL_SE MSR bit of oea, and should be
useful to, e.g., support byte-granular watchpoint for DDB in the future.

(rin)

Change pic_name from "openpic" to "u3_ht" so that it can be
distinguishable with generic OpenPIC driver.

(rin)

Include PIC name to interrupt source, instead of just "irq", so that
it appears in "interrupt id" field of intrctl(8).

Should be useful when multiple PICs are simultaneously available as in
405EX (where uic[12] are cascaded to uic0).

(rin)