Create a new header lwp_private.h to contain _lwp_getprivate_fast,
_lwp_gettcb_fast, _lwp_settcb and remove them from mcontext.h, so that:
1. we don't need special hacks to hide them
2. we can include <lwp.h> where needed to get the necessary prototypes
without redefining them locally.

Undo previous lwp.h change.

Split __lwp_getprivate_fast and __lwp_*tcb from mcontext.h into a separate
lwp.h file.

branches: 1.36.26;
Install sparc/intr.h

Remove the kernel PMC code. Sent yesterday on tech-kern@.

This change:

* Removes "options PERFCTRS", the associated includes, and the associated
ifdefs. In doing so, it removes several XXXSMPs in the MI code, which is
good.

* Removes the PMC code of ARM XSCALE.

* Removes all the pmc.h files. They were all empty, except for ARM XSCALE.

* Reorders the x86 PMC code not to rely on the legacy pmc.h file. The
definitions are put in sysarch.h.

* Removes the kern/sys_pmc.c file, and along with it, the sys_pmc_control
and sys_pmc_get_info syscalls. They are marked as OBSOL in kern,
netbsd32 and rump.

* Removes the pmc_evid_t and pmc_ctr_t types.

* Removes all the associated man pages. The sets are marked as obsolete.

branches: 1.34.2;
Move the complex logic for dynamically writing branches from ld.elf_so
into a header for reuse in crt0.o for static ifunc support. Change the
existing logic for sparc64 to use the Bicc variant of ba,a as it allows
+-8MB displacement compared to the BPcc variant's +-1MB. Teach the sparc
variant the same trick for using ba,a and not sethi+jmp when possible.

branches: 1.33.26;
Rename sljitarch.h to sljit_machdep.h.

branches: 1.32.10;
Add sljitarch.h on sparc.

branches: 1.31.2; 1.31.12;
Retire varargs.h support. Move machine/stdarg.h logic into MI
sys/stdarg.h and expect compiler to provide proper builtins, defaulting
to the GCC interface. lint still has a special fallback.
Reduce abuse of _BSD_VA_LIST_ by defining __va_list by default and
derive va_list as required by standards.

Add fenv support for sparc. Mostly copied from sparc64 and share with it.

branches: 1.29.70; 1.29.76;
Merge newlock2 to head.

branches: 1.28.20;
merge ktrace-lwp.

branches: 1.27.12;
Factor out W{CHAR,INT}_{MAX,MIN} into their own header file.

branches: 1.26.2;
Install (new) cpuconf.h.

From Christopher SEKIYA in PR port-sparc/21093.

Merge the nathanw_sa branch.

Remove KDIR=, since SYS_INCLUDE=symlinks and KDIR are not supported any more.

Merge the gehenna-devsw branch into the trunk.

This merge changes the device switch tables from static array to
dynamically generated by config(8).

- All device switches is defined as a constant structure in device drivers.

- The new grammer ``device-major'' is introduced to ``files''.

device-major <prefix> char <num> [block <num>] [<rules>]

- All device major numbers must be listed up in port dependent majors.<arch>
by using this grammer.

- Added the new naming convention.
The name of the device switch must be <prefix>_[bc]devsw for auto-generation
of device switch tables.

- The backward compatibility of loading block/character device
switch by LKM framework is broken. This is necessary to convert
from block/character device major to device name in runtime and vice versa.

- The restriction to assign device major by LKM is completely removed.
We don't need to reserve LKM entries for dynamic loading of device switch.

- In compile time, device major numbers list is packed into the kernel and
the LKM framework will refer it to assign device major number dynamically.

Implement pmc(9) -- An interface to hardware performance monitoring
counters. These counters do not exist on all CPUs, but where they
do exist, can be used for counting events such as dcache misses that
would otherwise be difficult or impossible to instrument by code
inspection or hardware simulation.

pmc(9) is meant to be a general interface. Initially, the Intel XScale
counters are the only ones supported.

branches: 1.20.4; 1.20.8; 1.20.12;
Don't do the symlink hack for fbio.h. Just use <dev/sun/fbio.h>.

branches: 1.19.2; 1.19.4;
Add definitions of C99 integer format conversion macros.
XXX Fastest minimum-width integer types haven't been decided upon yet.

Add definitions of C99 specified-width integer type limits.
XXX Fastest minimum-width integer types haven't been decided upon yet.

Add definitions of C99 integer constant macros.
Tidy Makefiles up a little.

Add definitions of C99 minimum-width and greatest-width integer types.
XXX Fastest minimum-width integer types haven't been decided upon yet.

branches: 1.15.2;
fbio.h moded to sys/dev/sun.
Retain symlink from /usr/include/machine.

Add <machine/int_types.h>, which provides namespace-pure definitions
of exact-width integer types.

Oops, install <machine/lock.h> on these systems, too.

Install "machineendian_machdep.h".

Cleanup the list of exported headers a bit.

C99: Define a NAN macro in <math.h> which evaulates to a constant expression of
a single-precision quiet NaN; only to be defined on platforms that do support
this value.

Added tctrl.h

branches: 1.8.2; 1.8.8;
Add <machine/apmvar.h> for sparc so that apmd can be built. Eventually
this allow it work on Tadpole SPARCbook 3 laptops (and/or other sparc
with power management).

Add bootinfo.h

Support for installboot

branches: 1.5.4;
Move the bswap functions from libutil to libc (this bups the
minor of libc and the major of libutil). For little-endian architectures
merge the bnswap() assembly versions with nto* and hton* using symbols
aliasing. Use symbol renaming for the bswap function in this case to avoid
namespace pollution.
Declare bswap* in machine/bswap.h, not machine/endian.h. For little-endian
machines, common code for inline macros go in machine/byte_swap.h
Sync libkern with libc.
Adjust #include in kernel sources for machine/bswap.h.

bsd_audioio.h is not used.

Remove mbppio.h from the list.
Also stop installing db_machdep.h, fbvar.h, z8530.h.

moved <machine/sun_disklabel.h> to <dev/sun/disklabel.h>

Rework the way kernel include files are installed. In the new method,
as with user-land programs, include files are installed by each directory
in the tree that has includes to install. (This allows more flexibility
as to what gets installed, makes 'partial installs' easier, and gives us
more options as to which machines' includes get installed at any given
time.) The old SYS_INCLUDES={symlinks,copies} behaviours are _both_
still supported, though at least one bug in the 'symlinks' case is
fixed by this change. Include files can't be build before installation,
so directories that have includes as targets (e.g. dev/pci) have to move
those targets into a different Makefile.

Improve am7930 family drivers to share more code.
audioamd(4) on sparc, vsaudio(4) on vax, and bba(4) are.
- Remove complex and useless callbacks: onopen, onclose, and
indirect_{read,write}. This makes audioamd and vsaudio almost the same.
- Remove (already disabled) assembly fast interrupt path from audioamd(4).
cf. http://mail-index.netbsd.org/source-changes/2009/12/19/msg004585.html
- Use trigger_* method rather than start_* method. It's more suitable.
vsaudio(4) was tested by naru@, bba(4) was tested by tsutsui@.

Use bus_space tags and handles, embedded in MD hooks AM7930_{READ_WRITE}_REG
to handle MD delay and bus padding. Tested on sparc2 by David Brownlee.

Should use regmaps, but that breaks sparc pDMA assembly code.

Switch all users (except ia64) of custom machine/ansi.h to common_ansi.h

Deduplicate the code among ports and poll definitions of types
directly from a compiler.

This fixes miscompilation of certain programs that instruct compilers
to generate code for different types. This bug has been detected with
-fshort-wchar in EFI firmware.

Proposed and discussed on a mailing list (twice).

Itanium uses custom !ELF fallback switch, temporarily leave it as it is.

branches: 1.24.18;
Make clock_t unsigned int everywhere.
Ok: matt@, mrg@

branches: 1.23.12; 1.23.28; 1.23.30;
Retire varargs.h support. Move machine/stdarg.h logic into MI
sys/stdarg.h and expect compiler to provide proper builtins, defaulting
to the GCC interface. lint still has a special fallback.
Reduce abuse of _BSD_VA_LIST_ by defining __va_list by default and
derive va_list as required by standards.

1. {wctype,wctrans,mbstate}_t: switch MD to MI like other
libc implementation (such as *BSD and glibc2).

2. don't typedef void * wc{type,trans}_t, suggested by soda@-san.
it may pass through compiler type check, it's harmful.
so i introduce dummy struct __tag_wc{type,trans}_t(iconv_t already does).

no ABI change was made.

branches: 1.21.4; 1.21.6;
merge christos-time_t

branches: 1.20.16; 1.20.18; 1.20.22; 1.20.30;
Merge the ppcoea-renovation branch to HEAD.

This branch was a major cleanup and rototill of many of the various OEA
cpu based PPC ports that focused on sharing as much code as possible
between the various ports to eliminate near-identical copies of files in
every tree. Additionally there is a new PIC system that unifies the
interface to interrupt code for all different OEA ppc arches. The work
for this branch was done by a variety of people, too long to list here.

TODO:
bebox still needs work to complete the transition to -renovation.
ofppc still needs a bunch of work, which I will be looking at.
ev64260 still needs to be renovated
amigappc was not attempted.

NOTES:
pmppc was removed as an arch, and moved to a evbppc target.

clean up some definitions around rune_t which are not needed anymore

branches: 1.18.8; 1.18.16; 1.18.22; 1.18.26; 1.18.28;
fix gcc -Werror -Wmissing-braces problem
mbstate_t(this is opaque object)'s initializer should be ``{ 0 }'',
so changed 1st field of union from character array to integer.

branches: 1.17.20; 1.17.22;
merge ktrace-lwp.

branches: 1.16.2;
For GCC3 and later, use the __builtin_va* constructs.

branches: 1.15.8; 1.15.14; 1.15.18;
Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.14.2;
add some ISO C 1995 I18N functions and types:
btowc, wctrans, towctrans, wcscoll, wcsxfrm, wctype_t and wctrans_t.

replace 'long long' with int64_t to compile stand alone program with
compiler other than GCC.

make mbstate_t bigger (32 -> 128 bytes).
XXX if you have libc after citrus locale import, please recompile libc,
and your applications that use mbstate_t (rather rare). really sorry
for the mess.

populate _BSD_MBSTATE_T_. add warning regarding to rune_t.

G/c _BSD_INTPTR_T_ and _BSD_UINTPTR_T_.

branches: 1.9.2; 1.9.10;
64-bit safety to help sharing compilation tools with sparc64.

branches: 1.8.10;
Provide definitions for intptr_t and uintptr_t, signed resp. unsigned integral
types large enough to hold any pointer.

Add _BSD_SUSECONDS_T_ and _BSD_USECONDS_T_; do some space vs. tab formatting
cleanup

Define _BSD_CLOCKID_T_ and _BSD_TIMER_T_

Various cleanup; mostly trailing spaces/tabs.

Add _BSD_WINT_T_ definition so we can handle wint_t type added in NA1.

copyright/Id cleanup

liten

need these

mark reloc bitfield linted

branches: 1.9.146; 1.9.150;
Rename __LDPGSZ to AOUT_LDPGSZ, to accurately reflect what it is.

branches: 1.8.40;
Remove trailing comma in enum definition.

copyright/Id cleanup

Fix spelling error in Copyright notice

fix a typo

Put definition of __LDPGSZ and `struct relocation_info' where they belong:
in <machine/exec.h>.

actually put something in a previously empty file.
__LDPGSZ should now be defined here!

stdarg.h, varargs.h, psl.h, autoconf.h: posted patches from torek
cpu.h: cpu_adjstack() won't work well in the NetBSD execve - need a
different model, stub enablertclock() for now..
exec.h: a netbsd-type exec.h file (ie. empty)
param.h: add MID_MACHINE
pmap.h: delete kernel_pmap hack - it doesn't work for us, + two proto's
vmparam.h: VM_MIN_ADDRESS = 0!

Chris Torek's sparc port. Missing lots of things.

Remove clause 3 and 4 from TNF licenses

branches: 1.2.108; 1.2.110; 1.2.112;
Moved MI APM definitions into dev/apm.

branches: 1.1.20;
Add <machine/apmvar.h> for sparc so that apmd can be built. Eventually
this allow it work on Tadpole SPARCbook 3 laptops (and/or other sparc
with power management).

PR 58960: fix misunderstanding in semantic and provide both the original
id string as well as _NETBSD_REVISIONID.
Do not rely on string concatenation in the inline assembler, use .ascii
and .asciz for individual string parts instead.

PR 58960: sparc/asm.h: Respect NETBSD_REVISIONID.

branches: 1.23.26;
Mark the .ident section as mergable string section to avoid redundant
entries.

branches: 1.22.18; 1.22.28;
Use .word for data that is known to be aligned.

Remove .proc 1, it has been ignored by gas for ages.

branches: 1.20.4;
Add END() macro like most other archs have.

branches: 1.19.2;
Pass PICFLAGS down to cc-as-as and use __PIC__ to decide if it is small
vs big PIC mode. Retire -DPIC and -DBIGPIC.

branches: 1.18.8; 1.18.18; 1.18.22;
Consistently use .gnu.warning with .pushsectio and .popsection on all
architectures instead of obsolete STABS frames for linker warnings.

Move counting of faults, traps, intrs, soft[intr]s, syscalls, and nswtch
from uvmexp to per-cpu cpu_data and move them to 64bits. Remove unneeded
includes of <uvm/uvm_extern.h> and/or <uvm/uvm.h>.

branches: 1.16.98;
Add a STRONG_ALIAS macro

branches: 1.15.22; 1.15.34;
RCSID should go to .ident section.
While here add __KERNEL_RCSID.

sync these with the sparc64 versions.

Add ENTRY_NOPROFILE().

branches: 1.12.8;
Changes from Martin Husemann and myself:
- Make the 32-bit SPARC profile support work with the GCC 2.95.3
SPARC ELF compiler, which uses a different name for mcount.
- Make the 64-bit SPARC profile support header look more like the 32-bit
SPARC header (no functional change -- 64-bit SPARC already used the
correct mcount name).

branches: 1.11.4; 1.11.6; 1.11.8;
Add a WEAK_ALIAS() macro.

branches: 1.10.8; 1.10.18;
add RODATA(), which is like ENTRY() but without the _PROF_PROLOGUE.

Argh, fix the ANSI CPP version of the a.out WARN_REFERENCES().

Implement WARN_REFERENCES().

branches: 1.7.4;
_GLOBAL_OFFSET_TABLE_ needs a _C_LABEL() wrapper, too.

Change _C_LABEL() to use proper underscoring depending on __ELF__ or not.

Add RCSID() macro.

PROF -> GPROF

copyright/Id cleanup

Profile asm entries also.

Define a few things here, in line with other ports.

fix various small typos, mainly in comments.

avoid using 'extern <func|data>;' inside a .c file, but instead
use header files and ensure definitions are not duplicated or
are technically (if not in codegen) wrong.

set machine_model[] earlier, so that bootstrap can use it.

also, avoid "extern <variable>" in .c files, and add
idempotency to autoconf.h.

branches: 1.48.50;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.47.2; 1.47.12;
#include <sys/bus.h> instead of <machine/bus.h>.

Remove clause 3 and 4 from TNF licenses

branches: 1.45.6; 1.45.8; 1.45.10;
Introduce device_find_by_xname and device_find_by_driver_unit to replace
alldevs iterations all over src.

Patch discussed with and improved on suggestioned from cube@.

branches: 1.44.44; 1.44.50;
De-__P.

s/u_intN_t/uintN_t/

branches: 1.42.4; 1.42.6;
merge ktrace-lwp.

branches: 1.41.2;
Sprinkle some const

- sprinkle const
- remove parameter names from function declarations

Rename bus_translate_address_generic => bus_space_translate_address_generic()
and move prototype to <machine/bus.h>.

Always use `full bus space'. In order to implement this more efficiently,
require that all bus tags have pointers to bus_space_read/write functions,
i.e. no run-time hunting for the first "upstream" implementation.
Since this changes the way bus tags should be constructed it makes sense
to do the same thing for the rest of the bus space methods.
So, now bus space tags are generally constructed by copying the parent's bus
tag and then overriding the methods that the bus driver needs to handle,
instead of starting with an empty bus tag and fiiling in only the fields needed.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.36.2;
Move makememarr() into promlib.c.

Add a generic function to translate a device address using a
parent's "ranges" property, and use it.

branches: 1.34.4;
`bus_type_t' is gone. Retain bus_space_map2() for MD drivers that need
to map registers to a fixed virtual address.

branches: 1.33.8; 1.33.12;
findzs() is now used only by the kgbd initialision code.

Remove old-style boot device recognition.

branches: 1.31.8;
Adjust to functionality and changes introduced with <machine/promlib.h>

configure() prototype is in <sys/device.h>

Move PROM interface prototypes into openprom.h
Use the `openprom_addr' structure for passing around physical addresses.

Nuke old getprop() function and put its successor in place.

Convert to [pv]addr_t & [pv]size_t.

<sparc/dev/sbusvar.h> => <dev/sbus/sbusvar.h>

Remove old mapdev() function prototype.

Make mainbus attach phys addr field a bus_addr_t.

Enter `obio_find_rom_map()' prototype.
Make `oba_paddr' field a `bus_addr_t'.
Define temporary compat `obio_bus_map()'.

Remove old `confargs' structure which is no longer used; extract the
structures representing the ROM properties.
Also, define the `mainbus' and `obio' attach arguments here.

bus_map() & bus_untmp() are only used for sun4/obio devices.

Always use `rr_iospace' to track the MMU page type bits.
This obsoletes the `bustype' argument in map[io]dev().

Remove bus_tmp().

Move "range" property description out of generic `romaux'.

Comply with recent autoconfiguration changes. Diffs graciously supplied
by Chris Demetriou.

Make room for another `value' in struct bootpath. Drop the global `bootdv'
in favour of storing the device pointer in the bootpath[] element corresponding
to the boot device. This also allows the opportunity to get to "intermediate"
devices (e.g. busses) should the need for this ever arise.

Add some v3 PROM glue.
Move autoconf subroutine prototypes here (from cpu.h).

Add missing prototypes and fix the fp struct for svr4.

Add a "mountroot hook" mechanism so that devices can take care of
special needs before being used in (*mountroot)().

Move bus_map() prototype here.

Change mapdev() and mapiodev() to take a `struct rom_reg', enabling the use
of IO spaces to construct device mappings.

declarations

Allow for multiple virtual PROM addresses.

Integrate changes from Chuck Cranor for the Sun 4/100.
Support for multiple register banks on SBUS devices (based on patches
from Francis Dupont).
Highlights:
romaux defines an array of register spaces.
pay attention to `vactype'.
quirks handling 4/100 idiosyncracies.

XXmatch change (and a few __P() as well)

copyright/Id cleanup

mapiodev takes a bustype argument now

"struct confargs" is used for all device match/attach functions. It
tells you what bus (sbus, vme, ...) you are on.

liten

stdarg.h, varargs.h, psl.h, autoconf.h: posted patches from torek
cpu.h: cpu_adjstack() won't work well in the NetBSD execve - need a
different model, stub enablertclock() for now..
exec.h: a netbsd-type exec.h file (ie. empty)
param.h: add MID_MACHINE
pmap.h: delete kernel_pmap hack - it doesn't work for us, + two proto's
vmparam.h: VM_MIN_ADDRESS = 0!

Chris Torek's sparc port. Missing lots of things.

Pass the "boothowto" from the bootloader to the kernel.
If (an old) bootloader did not pass the flags, check explicitly if we
missed RB_USERCONF and invoke userconf_prompt() in MD code, as MI code
already missed it.
This fixes PR 46466.

branches: 1.5.106; 1.5.110; 1.5.112;
De-__P.

branches: 1.4.4; 1.4.6;
merge ktrace-lwp.

branches: 1.3.16;
remove license clauses 3 and 4 from my cpoyright notices

branches: 1.2.10;
Define bootinfo record to pass the kernel file path from the boot program.

branches: 1.1.2; 1.1.18; 1.1.22;
Add bootinfo.h

bsd_audioio.h is not used.

Sync with sys/audioio.h for now, so SunOS compat works.

copyright/Id cleanup

Chris Torek's sparc port. Missing lots of things.

Remove __attribute__((__noreturn__)) from things already marked __dead
Found by the department of redundancy department.

branches: 1.24.20; 1.24.28; 1.24.32;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.23.26;
merge ktrace-lwp.

branches: 1.22.2;
- sprinkle const
- remove parameter names from function declarations

Encapsulate retrieval of properties from the global `options' prom node.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.19.2;
Definitions of some OBP generated CPU mailbox messages.

* Refine the comment describing openprom_addr.
* Add openprom_range and openprom_intr structures.

branches: 1.17.4; 1.17.8; 1.17.16; 1.17.18;
Adjust setprop() prototype to match that in dev/ofw/openfirm.h

Fix doubled 'the's in comments.

branches: 1.15.8;
Adjust to functionality and changes introduced with <machine/promlib.h>

Add wrappers for PROM's CPU start/stop functions.

Move PROM interface prototypes into openprom.h
Use the `openprom_addr' structure for passing around physical addresses.

Expose `v3' routines to everyone.

put promdev definition into <machine/bsd_openprom.h>.

Copyright police (s/Harvard University/Harvard College/).

remove some RCS id's we don't need.

Various cleanup; mostly trailing spaces/tabs.

add rom routines declarations

Add missing prototypes and fix the fp struct for svr4.

Correct some prototypes.

Change some caddr_t's to void *'s.

Rename a `xxx' function.

copyright/Id cleanup

Chris Torek's sparc port. Missing lots of things.

We can use __builtin_bswapX on sparc when using clang, so do so.

PR port-sparc64/59789 bswap is slow

sparc: Avoid using GCC builtins for byte-swapping and endian encoding.

gcc on most netbsd architectures (mostly excluding sparc and vax) can
generate fast MD byte-swapping code. On sparc it generates a function
call, which is very slow. We have existing inline macros for byte
swapping, so use those instead.

bswap on sparc with gcc is still non-ideal, so also reactivate the
portable endian encoding functions we apparently haven't been using
since the gcc2 days, rather than using bswap for that.

These changes improve the speed of disk encryption on ultrasparc ii
by almost 40% (the endian.h change being more significant in making
that number go high).

clang does the right thing on sparc (or at least introduces different
bottlenecks...), so avoid this there.

Thanks joerg for analysis of the problem, uwe for reviewing the patch.

PR port-sparc64/59789 bswap is slow
PR kern/59774 bearssl 32-bit AES is too slow

branches: 1.2.264;
Include <sys/bswap.h> for function prototypes. i386, pc532 and vax
still include <machine/byte_swap.h> and define macros for some of
the bswap*() functions.

Move the bswap functions from libutil to libc (this bups the
minor of libc and the major of libutil). For little-endian architectures
merge the bnswap() assembly versions with nto* and hton* using symbols
aliasing. Use symbol renaming for the bswap function in this case to avoid
namespace pollution.
Declare bswap* in machine/bswap.h, not machine/endian.h. For little-endian
machines, common code for inline macros go in machine/byte_swap.h
Sync libkern with libc.
Adjust #include in kernel sources for machine/bswap.h.

Switch sparc and sparc64 to new-style <sys/bus.h>.

do as phone suggested - remove sparc_bus_map_large() again and use a flag
instead ( BUS_SPACE_MAP_LARGE )

Add sparc_bus_map_large() to map large bus ranges like framebuffers and
apertures outside the IODEV range. ZX for example needs 8MB for the framebuffer
alone while the IODEV range provides only 16MB.

branches: 1.54.8; 1.54.10; 1.54.14; 1.54.18;
Remove clause 3 and 4 from TNF licenses

branches: 1.53.40; 1.53.42; 1.53.44;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

add a pair of new bus_dma(9) functions:
int _bus_dmatag_subregion(bus_dma_tag_t tag,
bus_addr_t min_addr,
bus_addr_t max_addr,
bus_dma_tag_t *newtag,
int flags)
void _bus_dmatag_destroy(bus_dma_tag_t tag)

that allow a (normally broken/limited) device to restrict the bus address
range it can talk to. this is used by bce(4) to limit DMA addresses to
1GB range, the maximum the chip can address.

all this is from Yorick Hardy <yhardy@uj.ac.za> with input from several
people on tech-kern.

XXX: bus_dma(9) needs an update still.

branches: 1.51.20;
De-__P.

s/u_intN_t/uintN_t/

Change "inline" back to "__inline" in .h files -- C99 is still too
new, and some apps compile things in C89 mode. C89 keywords stay.

As per core@.

branches: 1.48.2; 1.48.4; 1.48.6;
__asm__ -> __asm
__const__ -> const
__inline__ -> inline
__volatile__ -> volatile

Remove leading __ from __(const|inline|signed|volatile) -- it is obsolete.

merge ktrace-lwp.

branches: 1.45.4;
Add a dm_maxsegsz public member to bus_dmamap_t. This allows a user of the API
to select the maximum segment size for each bus_dmamap_load (up to the maxsegsz
supplied to bus_dmamap_create). dm_maxsegsz is reset to the value supplied to
bus_dmamap_create when the dmamap is unloaded.

branches: 1.44.4; 1.44.6;
Add bus_space_region_stream* functions (verbatim copy from sparc64).
Fixes PR port-sparc/28302. OK'd by pk.

Introduce bus_space_tag_alloc() for the common parts of bus tag allocation.

Rename bus_translate_address_generic => bus_space_translate_address_generic()
and move prototype to <machine/bus.h>.

Always use `full bus space'. In order to implement this more efficiently,
require that all bus tags have pointers to bus_space_read/write functions,
i.e. no run-time hunting for the first "upstream" implementation.
Since this changes the way bus tags should be constructed it makes sense
to do the same thing for the rest of the bus space methods.
So, now bus space tags are generally constructed by copying the parent's bus
tag and then overriding the methods that the bus driver needs to handle,
instead of starting with an empty bus tag and fiiling in only the fields needed.

Use __insn_barrier() in all bus_space read/write operations.

branches: 1.39.2;
Handle 64bit DMA addresses on PCI for platforms that can (currently only
enabled on amd64). Add a dmat64 field to various PCI attach structures,
and pass it down where needed. Implement a simple new function called
pci_dma64_available(pa) to test if 64bit DMA addresses may be used.
This returns 1 iff _PCI_HAVE_DMA64 is defined in <machine/pci_machdep.h>,
and there is more than 4G of memory.

Introduce BUS_DMA_NOCACHE, and bus_dmamem_map() of i386 supports it.

Make the *_stream_* methods always use the *_real accessors.
Define __BUS_SPACE_HAS_STREAM_METHODS.

part one of bus_space(9) fixes to enable bus spaces to override the
bus_space_{read,write}_[1248]() functions, which will allow 16-bit
PCMCIA support to work without additional hacks in MI drivers.
this option is not enabled yet.

Remove the `flags' argument from bus_intr_establish().

bus_intr_establish() now takes an optional `fast trap' handler argument.
BUS_INTR_ESTABLISH_FASTTRAP and BUS_INTR_ESTABLISH_SOFTINTR are no longer used.

Put a bus's ranges in the bus space tag, and make sparc_bus_map()
perform the translation and recursion if t->ranges != NULL. Make
sbus, cpuunit, and bootbus inherit the parent's map/mmap routines,
and delete the now-unused mapping functions. Update all places where
bus space tags are statically allocated.

branches: 1.32.2;
Delint a bit.

bus space methods should be "static __inline__", not "extern __inline__".
Otherwise an unoptimized kernel cannot be linked. Reviewed by pk.

BUS_ADDR(): avoid signed conversions.

Add bus_space_vaddr().

`bus_type_t' is gone. Retain bus_space_map2() for MD drivers that need
to map registers to a fixed virtual address.

Use regular read/writes for stream variant calls.

branches: 1.26.4;
Change bus_space_mmap() signature to the official one.

branches: 1.25.2;
Add BUS_DMA_READ and BUS_DMA_WRITE flags, that hint the back-end
at dmamap load time that the mapping will be used for a unidirectional
transfer of the specified direction.

branches: 1.24.2;
Add the BUS_DMA_STREAMING flag.

branches: 1.23.2;
Add a `device class' interrupt level argument (from machine/intr.h)
to bus_interrupt_establish().

It's currently only used in sparc64/dev/psycho.c to assign a CPU interrupt
level to devices in PCI slots.

Change the kernel mmap interface so that the offset to map is an
"off_t" and the return value is a "paddr_t" to allow mappings
at offsets past 2^31 bytes. Somewhat inspired by FreeBSD, which
only changed the offset to a "vm_offset_t".

Includes updates for the i386, pc532 and sh3 mmmmap from Jason Thorpe.

Maintain the DVMA map extent subregion boundaries in the bus_dma map
structure.

Fix pastos in bus_space_read_region_*() functions.

branches: 1.19.4;
Simplify the bus_dma(9) implementation:

- the `alignment' and `boundary' arguments in bus_dmamem_alloc() only
serve as hints on how to break up a DMA transaction into chunks
when necessary; it is not necessary on sparc DVMA hardware.

- exclusively use the VM page size (i.e. PAGE_SIZE et.al.) when computing
resource allocations.

- remember the size of DVMA resource allocations in the private `_ds_sgsize'
field in a dma segment descriptor.

The bus_dma(9) implementation now allocates DVMA addresses in
bus_dmamap_load_*() routines only. Note that DMA memory buffers
allocated with bus_dmamem_alloc() _must_ be loaded by bus_dmamap_load_raw().

define a "BUS_SPACE_MAP_PREFETCHABLE" flag which basically means that
device accesses are idempotent (but should not be cached by the CPU)

branches: 1.16.8;
Add a macro to check for sufficient pointer alignment in bus.h context,
"BUS_SPACE_ALIGNED_POINTER()".
Equal to the param.h "ALIGNED_POINTER()" normally, but obeys additional
requirements of the bus_space_xxx_n() macros. (BUS_SPACE_DEBUG)

fix address incrementing in the read_region_N, write_region_N,
set_region_N and copy_region_N methods

Turn a bunch of bus_space_*() macros into in-line functions mostly to avoid
variable aliasing.

Turn the passed tag into "used" arguments.

Convert to [pv]addr_t & [pv]size_t.

Define a BUS_DMA flag for devices that have a 24-bit address space.
Update _bus_dma* prototypes.

branches: 1.10.2;
* Pass the bus tag instead of just the cookie to all bus space methods.
* In the bus tag structure, keep a reference to the parent bus.
* Express all bus space methods as inline functions in stead of macros
and use the parent bus tag reference to find the nearest "upstream"
bus that implements the method.

`sparc_bus_mmap()' declaration moved to machdep.c

bus_space_probe() prototype.

Add `interrupt establish' and ` device mmap' methods and macros to
match to the bus tag structure.

Add offset and length parameters to bus_dmamap_sync(), used for specifiying
partial syncs of a DMA mapping.

Change the last argument of bus_dmamap_sync() from bus_dmasync_op_t to int,
and allow more than one synchronization operation to be specified in
a single call. Dissallow mixing of PRE and POST operations.

Add dm_mapsize to bus_dmamap_t and rename BUS_DMAMEM_NOSYNC to
BUS_DMA_COHERENT.

Fix parenthesis inbalance.

Re-do the `busdma' interface.

The sparc's bus.h.

Provide PRIxBUSADDR, PRIxBUSSIZE, PRIuBUSSIZE, and PRIxBSH for all arches
to follow arm and (generic) mips.

Reviewed by christos.

branches: 1.1.54;
Per discussion at
<http://mail-index.netbsd.org/tech-kern/2010/04/02/msg007941.html>,
divide each machine's bus.h into bus_defs.h (constants & data types)
and bus_funcs.h (macro implementations of bus_space(9) routines and MD
prototypes).

Note that some bus_space(9) routines' implementation will move to .c
files from inline subroutines or macros in .h files.

I've only made the split for machine architectures where there is PCI.
All of the non-PCI-having architectures will require a similar split.

These #include files are not referenced by any (committed) Makefiles or
header files, yet. Changes to Makefiles, to <sys/bus.h>, and to some
more machine-dependent files will dribble in before I throw the switch.

Per discussion at
<http://mail-index.netbsd.org/tech-kern/2010/04/02/msg007941.html>,
divide each machine's bus.h into bus_defs.h (constants & data types)
and bus_funcs.h (macro implementations of bus_space(9) routines and MD
prototypes).

Note that some bus_space(9) routines' implementation will move to .c
files from inline subroutines or macros in .h files.

I've only made the split for machine architectures where there is PCI.
All of the non-PCI-having architectures will require a similar split.

These #include files are not referenced by any (committed) Makefiles or
header files, yet. Changes to Makefiles, to <sys/bus.h>, and to some
more machine-dependent files will dribble in before I throw the switch.

Make all __ALIGNBYTES macros return the same type (size_t)

branches: 1.12.6; 1.12.10;
Change CMSG_SPACE and CMSG_LEN to provide Integer Constant Expressions
again. This was changed in sys/socket.h r1.51 to work around fallout
from the IPv6 aux data migration. It broke the historic ABI on some
platforms. This commit restores compatibility for netbsd32 code on such
platforms and provides a template for future changes to the CMSG_*
alignment. Revert PCC/Clang workarounds in postfix and tmux.

branches: 1.11.112; 1.11.116;
merge ktrace-lwp.

A __insn_barrier() is now available for all archs, remove the MD version
here. Suggested by Jason Thorpe.

Introduce __insn_barrier(), designed to prevent instruction re-ordering
with minimal side-effects across its location in the source code.

branches: 1.8.42;
Garbage-collect.

On ELF, make __indr_reference undefined and __warn_references do nothing,
as on Alpha. This should be fixed to do something in both cases.

Change _C_LABEL() to use proper underscoring depending on __ELF__ or not.

Bug fixes and cleanup from Chris Demetriou <cgd@pa.dec.com>:
- fix _C_LABEL so that it actually works.
- make __RENAME use _C_LABEL.
- fix __RENAME so that it expects an unquoted argument.
- fix __indr_reference and __warn_references so that they
supply their own final semicolon.
- define __warn_references to nothing if not GNU C (required
by the way it's used).

The __warn_references semicolon change has to be made
so that __warn_references can be defined into nothing.
(A ; all by itself isn't a great idea.) The __indr_reference
change was made for consistency.

Implement __RENAME() in <machine/cdefs.h>

branches: 1.3.14;
For now, define `__indr_reference' and `__warn_references' as empty
macros if __GNUC__ not defined.

Changed name of __weak_reference() to __indr_reference(). They really
are indirect references, and I want to add a real __weak_reference()
macro to <machine/cdefs.h> soon.

This file, which will be included by <sys/cdefs.h>, will contain macros
such as __warn_references() and __weak_reference() which are actually
machine dependant. This will make it easier for ports that are being
bootstraped with ELF and ECOFF based toolchains.

This change also introduces a new macro, _C_LABEL(x). _C_LABEL expands
its argument, an identifier, to a character string of the identifier
name as it is represented in an object file.

For most ports, _C_LABEL(x) will expand to "_x", for ELF based ports
_C_LABEL(x) will expand to "x".

s/interrput/interrupt/ and s/accesss/access/ in comments.

Correct invalid type.

DMA, not dma nor Dma.

Add a few constants used for mapping bit and pieces in the kernel driver.

copyright/Id cleanup

branches: 1.1.2;
needed for X

Merge the gehenna-devsw branch into the trunk.

This merge changes the device switch tables from static array to
dynamically generated by config(8).

- All device switches is defined as a constant structure in device drivers.

- The new grammer ``device-major'' is introduced to ``files''.

device-major <prefix> char <num> [block <num>] [<rules>]

- All device major numbers must be listed up in port dependent majors.<arch>
by using this grammer.

- Added the new naming convention.
The name of the device switch must be <prefix>_[bc]devsw for auto-generation
of device switch tables.

- The backward compatibility of loading block/character device
switch by LKM framework is broken. This is necessary to convert
from block/character device major to device name in runtime and vice versa.

- The restriction to assign device major by LKM is completely removed.
We don't need to reserve LKM entries for dynamic loading of device switch.

- In compile time, device major numbers list is packed into the kernel and
the LKM framework will refer it to assign device major number dynamically.

branches: 1.19.8;
- define other DEV_ constants that the local port uses.
- delete cdev_decl(mm) since <sys/conf.h> does it.

branches: 1.18.2; 1.18.8;
- add dev_noimpl(xxx,yyy) macro to replace "(dev_type_xxx((*))) yyy",
and use appropriately

- create more helper macros:
. cdev__xyz_init(c,n), such as cdev__ocri_init() for
/* open, close, read, ioctl */, etc.
. cdev__xRy_init(c,n), where nullop is used instead of enodev to dummy out
method `R' and the comments now read /* xxx (read) yyy */ instead
. cdev__xyz_t_init(c,n,t) - as per cdev__xyz_init, but sets d_type = t
as well

- use seltrue instead of dev_noimpl(poll,*), as (IIRC) cdevsw.d_poll should
always DTRT WRT returning a valid result. (a few devices previously
incorrectly returned ENODEV)

- use dev_noimpl(stop,enodev) instead of dev_noimpl(stop,nullop) if tty
== 0, because it doesn't matter if dev_type_stop isn't implemented in that
case, and it allows the use of the cdev__xyz_init macros. certain ports
(sparc,sparc64,x68k) used the nullop method for dev_type_stop in a few
drivers, whereas everything else uses enodev

- ensure that the comments are accurate WRT the behaviour of a given entry

branches: 1.17.6;
Add new spl level for tadpole devices splts102.
Device entrypoints for tctrl device (tadpole micro controller)
add header for direct-access ioctl to the tadpole microcontroller.

branches: 1.16.2; 1.16.8;
Initial stab at a P9100 console FB driver for the Tadpole 3GX.
It supports RASTERCONSOLE only and the colormap support is
broken (the hardware doesn't seem to be doing what the tech.
docs. say it should be doing). But it is usable as a console
(but with blue on white as the only color choice). I figured
I'd check it in since it is somewhat usable and someone else
might figure what I screwed up in the color map support.

Add preliminary support for the built-in modem on Tadpole 3GX/S
laptops. There is currently something wrong with the interrupt
code but it does attach and sense a 16550a. (anyways the built
in modem is only a 14.4Kb and you'd really want to use a PCMCIA
modem to get faster dialup rates).

branches: 1.14.6;
Updating of bdev's and cdev's to support RAIDframe.

Add scsibus entry points to the cdevsw[].

Fill the `bpp' slot.

Assign copyright to TNF.

Add `mtty' and `mbpp' entries for the magma driver by Iain Hibbert.

Use common zs code.

oops; rd -> md paul left behind...

Declare `cgtwo' and `rd' character devices.

Add cgfourteen to conf.h.
Add new framebuffer types for recent-model graphics devices to fbio.h.

Implement poll(2).

ramdisk stuff

Remove a "notyet" in last commit.

tcx declaration

New md conf.h per Jason's plan.

Break cycle by using `struct kmutex *' instead of `kmutex_t *'.

sys/sched.h included sys/mutex.h
which includes sys/intr.h
which includes machine/intr.h
which on cats includes arm/footbridge/footbridge_intr.h
which includes arm/cpu.h
which includes sys/cpu_data.h
which includes sys/sched.h

But there was never any real need for sys/mutex.h in sys/sched.h,
because it only uses pointers to the opaque struct kmutex. Cycle
broken by using `struct kmutex *' instead of pulling in sys/mutex.h
for the definition of kmutex_t.

Side effect: This revealed that sys/cpu_data.h needed sys/intr.h
(which was pulled in accidentally by sys/mutex.h via sys/sched.h) for
SOFTINT_COUNT. Also revealed some other machine/cpu.h header files
were missing includes of sys/mutex.h for kmutex_t.

branches: 1.110.4;
Improved the performance of kernel profiling on MULTIPROCESSOR, and possible to get profiling data for each CPU.

In the current implementation, locks are acquired at the entrance of the mcount
internal function, so the higher the number of cores, the more lock conflict
occurs, making profiling performance in a MULTIPROCESSOR environment unusable
and slow. Profiling buffers has been changed to be reserved for each CPU,
improving profiling performance in MP by several to several dozen times.

- Eliminated cpu_simple_lock in mcount internal function, using per-CPU buffers.
- Add ci_gmon member to struct cpu_info of each MP arch.
- Add kern.profiling.percpu node in sysctl tree.
- Add new -c <cpuid> option to kgmon(8) to specify the cpuid, like openbsd.
For compatibility, if the -c option is not specified, the entire system can be
operated as before, and the -p option will get the total profiling data for
all CPUs.

avoid using 'extern <func|data>;' inside a .c file, but instead
use header files and ensure definitions are not duplicated or
are technically (if not in codegen) wrong.

branches: 1.108.2;
PR port-sparc/55261: sparc still panics running ATF tests

Reinstate the cpu_info change and remove __HAVE_CPU_DATA_FIRST to fix
build failure.

Back out previous - it doesn't work.

PR port-sparc/55261: sparc still panics running ATF tests

Put cpu_data at the end of cpu_info to see how that modifies the symptom.

tuck curproc/curlwp under _KERNEL only (no _KMEMUSER)

PR port-sparc/54789: Sparc boot fails with "Trap 0x21 while interrupts disabled"

Remove assumption that cpu_info is smaller than 1kB.

Fix false sharing problems with cpu_info. Identified with tprof(8).
This was a very nice win in my tests on a 48 CPU box.

- Reorganise cpu_data slightly according to usage.
- Put cpu_onproc into struct cpu_info alongside ci_curlwp (now is ci_onproc).
- On x86, put some items in their own cache lines according to usage, like
the IPI bitmask and ci_want_resched.

cpu_need_resched():

- Remove all code that should be MI, leaving the bare minimum under arch/.
- Make the required actions very explicit.
- Pass in LWP pointer for convenience.
- When a trap is required on another CPU, have the IPI set it locally.
- Expunge cpu_did_resched().

- Cleanup for dynamic sysctl:
- Remove unused *_NAMES macros for sysctl.
- Remove unused *_MAXID for sysctls.
- Move CTL_MACHDEP sysctl definitions for m68k into m68k/include/cpu.h and
use them on all m68k machines.

branches: 1.100.2; 1.100.4;
implement cpuctl identify for sparc and sparc64.

sparc:
- move enum vactype and struct cacheinfo into cpu.h
- move the cache flags from cpuinfo.flags into CACHEINFO.c_flags
(this allows the new cache_printf_backend() to see them.)
remove unused CPUFLG_CACHEIOMMUTABLES and CPUFLG_CACHEDVMA.
- align xmpsg to 64 bytes
- move cache_print() into cache_print.h so it can be shared with
cpuctl. it only depends upon a working printf().
- if found, store the CPU node's "name" into cpu_longname. this
changes the default output to show the local CPU not the
generic CPU family. eg:
cpu0 at mainbus0: mid 8: Ross,RT625 @ 90 MHz, on-chip FPU
vs the generic "RT620/625" previously shown.
- for each CPU export these things:
- name
- fpuname
- mid
- cloc
- freq
- psr impl and version
- mmu impl, version, and number of contexts
- cacheinfo structure (which changed for the first time ever
with this commit.)

sparc64:
- add a minimal "cacheinfo" structure to export the i/d/e-cache
size and linesize.
- store %ver, cpu node "name" and cacheinfo in cpu_info.
- set cpu_info ver, name and cacheinfo in cpu_attach(), and
export them via sysctl, as well as CPU ID and clock freq

cpuctl:
- add identifycpu_bind() that returns false on !x86 as their
identify routines do not need to run on a particular CPU to
obtain its information, and use it to avoid trying to set
affinity when not needed.
- add sparc and sparc64 cpu identify support using the newly
exported values.

- return early in xcall() if the function is sparc_noop() instead of triggering
the IPI and then ignoring responses ( or lack thereof )
- write the .tag field last to avoid a race when polling for an incoming
IPI
- add event counters for IPIs being caught with the mutex not held, and for
messages that are already marked as completed
With this my SS20 made it through 48 hours of pkgsrc with MAKE_JOBS=3 and a
pair of SM81s.
Hypersparcs still crash but instead of craziness we get actual error messages,
apparently one CPU will occasionally do a watchdog reset, which according to
the manual is caused by catching a trap with traps disabled. Now to figure
out how that can even happen...

dumpconf(void) long doesn't exist, remove the prototype

PR kern/39714 by Henning Petersen

branches: 1.97.8;
rearrange slightly to enable crash(8) to build.

move struct cpu_info and xpmsg into <machine/cpu.h> from
<sparc/sparc/cpuvar.h>. this enables them for crash(8)
and this seems more natural to me anyway.

branches: 1.95.6; 1.95.10;
make cpu_need_resched() macros consistent; __USE flags

branches: 1.94.2; 1.94.4;
On unaligned access, provide the data address to the signal handler, instead
of the faulting instructions address.

Remove COMPAT_SA / KERN_SA. Welcome to 6.99.3!
Approved by core@.

branches: 1.92.2; 1.92.6;
Get rid of #ifdef __sparc__ in uvm code - as noted by cgd back 1996,
now that we have __HAVE_CPU_VMSPACE_EXEC/cpu_vmspace_exec().

Provide LWP_PC

branches: 1.90.2; 1.90.4;
introduce a hack to workaround the savefpstate IPI get's NULL
struct fpstate *.

add a new entry point ipi_savefpstate() that is used for this IPI
instead of the plain ipi_savefpstate(). check for %o0 being NULL,
and if so, just return.

add event counters for the savefpstate IPI, and another one for
when it detects the NULL problem.

make the cpu_info struct xmpsg be aligned to a single cache line.

with this change applied (and another minor change in testing) my
SS20 with dual SM75's has survived for 17 hours on build.sh -j5,
and has reported 7 NULL savefps. (the minor change seems to have
a significant effect at reducing this number, but more testing is
needed for it.)


XXX: this is horrible and we really need to find the real problem
XXX: but this should let people use sparc smp again, and fixes
XXX: for the real problem can be tested by seeing if any of the
XXX: savefp IPI null counter becomes non-zero.

branches: 1.89.4;
add an explicit "bool mpsafe" to intr_establish(). only set it to true
for the sun4m timer/counter interrupts. this gets lev10/14 working again
on SMP systems, where those interrupts started taking the kernel lock and
hanging...

add two new members to struct intrhand: ih_realfun and ih_realarg, and
use them to take the kernel lock around non-IPL_VM interrupts, using
a intr_biglock_wrapper() function ike x86 does.

Remove uarea swap-out functionality:

- Addresses the issue described in PR/38828.
- Some simplification in threading and sleepq subsystems.
- Eliminates pmap_collect() and, as a side note, allows pmap optimisations.
- Eliminates XS_CTL_DATA_ONSTACK in scsipi code.
- Avoids few scans on LWP list and thus potentially long holds of proc_lock.
- Cuts ~1.5k lines of code. Reduces amd64 kernel size by ~4k.
- Removes __SWAP_BROKEN cases.

Tested on x86, mips, acorn32 (thanks <mpumford>) and partly tested on
acorn26 (thanks to <bjh21>).

Discussed on <tech-kern>, reviewed by <ad>.

fix up cpu_setfunc() as noted by uwe:

- don't call lwp_startup for cpu_setfunc() users
- introduce lwp_setfunc_trampoline instead
- no need to set the "new" lwp for setfunc

avoid a local variable in a do {} while (0) macro.

branches: 1.84.4; 1.84.14; 1.84.22;
Remove CTL_MACHDEP_NAMES, it's not used anywhere.

Ok by martin@.

branches: 1.83.10; 1.83.14;
Interrupt handling changes, in discussion since February:

- Reduce available SPL levels for hardware devices to none, vm, sched, high.
- Acquire kernel_lock only for interrupts at IPL_VM.
- Implement threaded soft interrupts.

branches: 1.82.2;
Merge the ppcoea-renovation branch to HEAD.

This branch was a major cleanup and rototill of many of the various OEA
cpu based PPC ports that focused on sharing as much code as possible
between the various ports to eliminate near-identical copies of files in
every tree. Additionally there is a new PIC system that unifies the
interface to interrupt code for all different OEA ppc arches. The work
for this branch was done by a variety of people, too long to list here.

TODO:
bebox still needs work to complete the transition to -renovation.
ofppc still needs a bunch of work, which I will be looking at.
ev64260 still needs to be renovated
amigappc was not attempted.

NOTES:
pmppc was removed as an arch, and moved to a evbppc target.

branches: 1.81.8; 1.81.10;
merge yamt-idlelwp branch. asked by core@. some ports still needs work.

from doc/BRANCHES:

idle lwp, and some changes depending on it.

1. separate context switching and thread scheduling.
(cf. gmcgarry_ctxsw)
2. implement idle lwp.
3. clean up related MD/MI interfaces.
4. make scheduler(s) modular.

branches: 1.80.2; 1.80.4; 1.80.10;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.79.2;
Remove spllowersoftclock() and CLKF_BASEPRI(), and always dispatch callouts
via a soft interrupt. In the near future, softclock will be run from process
context.

Merge newlock2 to head.

branches: 1.77.20;
merge ktrace-lwp.

We need to init p_md.md_flags on fork, so implement cpu_proc_fork().

Add missing extern to make the definition of softnet_cookie into a declaration.

Cosmetic. Kill __P. Remove few remaining paramter names in
prototypes. CONSTCOND in while (0). Move wcopy/wzero declarations
with the rest of machdep.c. Add some blank lines for readability.

branches: 1.73.2;
G/c msiiep_swap_endian, it's no longer used now that we run without
byteswapping in the PCIC.

branches: 1.72.2;
Rename 'ncpus' to 'sparc_ncpus' to avoid shadow warnings in m.i. code.
Also sprinkle an __UNVOLATILE() for sparc.
n.b. sparc64 'cpus' should probably be renamed to 'cpu_info_list' to
match i386 et al.

Rename ncpu to ncpus (as other ports call it), to avoid shadow warnings.
Sprinkle some const.

Rename ncpu to ncpus (as other ports call it), to avoid shadow warnings.
Sprinkle some const.

The cpu_exit() detour is no longer necessary.

Remove the cache_flush() trampoline; it's no longer directly cross-called.
Instead, implement the MP parts in terms of cross-callable vcache_flush_range()
function.

branches: 1.66.2;
Rearrange process exit path to avoid need to free resources from different
process context ('reaper').

From within the exiting process context:
* deactivate pmap and free vmspace while we can still block
* introduce MD cpu_lwp_free() - this cleans all MD-specific context (such
as FPU state), and is the last potentially blocking operation;
all of cpu_wait(), and most of cpu_exit(), is now folded into cpu_lwp_free()
* process is now immediatelly marked as zombie and made available for pickup
by parent; the remaining last lwp continues the exit as fully detached
* MI (rather than MD) code bumps uvmexp.swtch, cpu_exit() is now same
for both 'process' and 'lwp' exit

uvm_lwp_exit() is modified to never block; the u-area memory is now
always just linked to the list of available u-areas. Introduce (blocking)
uvm_uarea_drain(), which is called to release the excessive u-area memory;
this is called by parent within wait4(), or by pagedaemon on memory shortage.
uvm_uarea_free() is now private function within uvm_glue.c.

MD process/lwp exit code now always calls lwp_exit2() immediatelly after
switching away from the exiting lwp.

g/c now unneeded routines and variables, including the reaper kernel thread

#if defined(_KERNEL_OPT) is enough, #if !defined(_LKM) isn't needed
in that case. Pointed out by mrg.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.63.2;
Add functions to set & get the interrupt target CPU.

Merge the nathanw_sa branch.

Rename the current mp_{pause,resume}_cpus() => mp_{pause,resume}_cpus_ddb().
Implement mp_pause_cpus/mp_resume_cpus on top of the PROM services.

Add CLKF_LOPRI() macro that allows a (timer) interrupt handler to determine
whether it is interrupting code running at a given IPL level.

schedcpu() has been fixed; now we can notify another CPU about a pending
reschedule request.

Remove needless indirection from the curproc() macro.

Replace `want_resched' and `want_ast' globals by per-CPU variables.

Move schedintr() to clock.c and initialise schedhz in initclocks(), so
these are available to all timer implementations.

remove dead extern fpproc/foundfpu declarations.

* intr_establish() now takes an optional `fast trap' routine argument.
* also remove __P().

setsoftint() is no longer used.

Pass the `device class interrupt level' on to intr_establish() and use to
raise the ipl in the interrupt handlers to the appropriate level. This avoids
interrupt handler interference if one of the devices actually interrupts at
a lower hardware level than the maximum level assined to a device class.

Based on code from Art Grabowski in openbsd.

Not all sun4m platforms have version 8 sparc CPUs. So go out to the PROM
and get the CPU architecture version from the PROM cpu node `sparc-version'
property.

Add `machdep.cpu_arch' sysctl to determine the CPU architecture version.
Currrently, cpu types `sun4' and `sun4c' produce version 7, all others
version 8.

* add machdep.booted_device and machdep.boot_args (both fetched from
appropriate PROM variables).
* return "netbsd" for machdep.booted_kernel if the PROM doesn't return us a
kernel name.

Export msiiep_swap_endian(onoff) from msiiep.c to turn on/off
automagic endian swapping performed by PCI controller. Need it in
several places where we want to enter PROM repl.

defopt "options SUN4*", put #define for each of SUN4, SUN4C, SUN4M and SUN4U
into opt_arch_sparc.h and include this wherever they are used.

Remove redundant declarations.

branches: 1.45.2; 1.45.4;
implement generic soft interrupts as described in softintr(9). convert
the network, clock & serial softint's to use them.

branches: 1.44.2; 1.44.4;
Don't need to prototype child_return() here, it's in <sys/proc.h>.

sprinkle KERNEL_LOCK() and KERNEL_PROC_LOCK() where they're missing. add
SYSCALL_DEBUG support to trap.c.

branches: 1.42.2;
fix bit rot so that MP kernel compiles s/cpu_no/ci_cpuid/

Make need_resched() take a "struct cpu_info *" argument. This
causes gives a primitive form of processor affinity. Its use in
roundrobin() still needs some work.

Fix typo. kern/10291.

Provide a CPU_IS_PRIMARY() macro in the MULTIPROCESSOR case.

Move switchexit() declaration in here.

- In the MULTIPROCESSOR case, initialize p_cpu before a process is
marked SONPROC.
- Fix a bug where all cpu_info structures except for the boot CPUs
would exist at both a CPU-local VA (CPUINFO_VA) and a gloal VA;
The boot CPU's existed only a CPUINFO_VA.
- Add a self-reference pointer to the cpu_info that references the
global address in the MULTIPROCESSOR case. Just allow it to reference
the `local' VA in the single-processor case, as CPUINFO_VA is global
enough when there's only one processor to care about. Change curcpu()
to return the global address.

branches: 1.36.2;
First sweep at scheduler state cleanup. Collect MI scheduler
state into global and per-CPU scheduler state:

- Global state: sched_qs (run queues), sched_whichqs (bitmap
of non-empty run queues), sched_slpque (sleep queues).
NOTE: These may collectively move into a struct schedstate
at some point in the future.

- Per-CPU state, struct schedstate_percpu: spc_runtime
(time process on this CPU started running), spc_flags
(replaces struct proc's p_schedflags), and
spc_curpriority (usrpri of processes on this CPU).

- Every platform must now supply a struct cpu_info and
a curcpu() macro. Simplify existing cpu_info declarations
where appropriate.

- All references to per-CPU scheduler state now made through
curcpu(). NOTE: this will likely be adjusted in the future
after further changes to struct proc are made.

Tested on i386 and Alpha. Changes are mostly mechanical, but apologies
in advance if it doesn't compile on a particular platform.

Provide cpu_boot_secondary_cpus().

branches: 1.34.2; 1.34.8;
Implement sysctl `machdep.booted_kernel'.

Define cpu_number() as discussed on tech-smp.

Add SIR_SERIAL soft interrupt source.

Changes to support fork_kthread():
- cpu_set_kpc() now takes void *arg third argument, passed to the
entry point.
- cpu_fork() allows parent to be non-curproc iff parent is proc0.
When forking non-curproc, assume its state has already been saved.
- Adjust various pieces of machine-dependent code to account of all of this.

CLKF_INTR() needs to have a MULTIPROCESSOR version.

cpu_set_kpc() prototype is already in <sys/systm.h>.

mp_{pause,resume,halt}_cpu() prototypes.

Nuke obsolete prototype.

iommu_{enter,remove} moved to iommuvar.h

eliminate declarations of global data in <sparc/cpu.h> and declare in
source files, reducing global symblol replication in .o kernel files

Add fix_align() and emulinstr() prototypes.

Fix an inconsistency that came in with Lite: setrq() was renamed to
setrunqueue(), but remrq() was never renamed. Rename remrq() to
remrunqueue().

Make the SPARC kernel compile again after recent tty cdevsw changes.

Move autoconf subroutine prototypes to autoconf.h
Deal with sun4m `software interrupts'.

Fix prototypes for the latest kpc changes.

Add missing prototypes and fix the fp struct for svr4.

Remove cpu_set_init_frame().

remove unused cpu_exec() definitions. moved "broken swap" markers, for
ports that still need it, to types.h.

cpu_swapout() not needed after all.

Start swapping.

define BROKEN_SWAP and/or cpu_swapout as appropriate.

- added sunos_machdep.c for sun3, atari, amiga and mac68k.
- changed machdep.c and trap.c to use struct emul.
- remove ep_setup references.
- added struct emul to all emulations.

KERNEL -> _KERNEL

prototype wcopy/wzero

copyright/Id cleanup

vme interrupt support

USPAGE

liten

profiling and clock stuff

flags foo

move defn of trapvec to cpu.h

sparc runs in current today

stdarg.h, varargs.h, psl.h, autoconf.h: posted patches from torek
cpu.h: cpu_adjstack() won't work well in the NetBSD execve - need a
different model, stub enablertclock() for now..
exec.h: a netbsd-type exec.h file (ie. empty)
param.h: add MID_MACHINE
pmap.h: delete kernel_pmap hack - it doesn't work for us, + two proto's
vmparam.h: VM_MIN_ADDRESS = 0!

Chris Torek's sparc port. Missing lots of things.

Macros expanding to defined() expression is not actually correct ISO C
behavior. Since it is not clear whether someone might decide to define
the macros to random non-1 values, introduce a level of indirection and
count the same of the helper macros.

branches: 1.4.122; 1.4.142; 1.4.146;
merge ktrace-lwp.

implement pmap_wired_count(). use more macros for determining MMU type.

branches: 1.2.8;
Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.1.2;
Move CPU configuration predicates to <machine/cpuconf.h>.

sparc: fix snprintb formats for SFSR_BITS

fix various typos in comments.

Add missing NUL to prevent buffer overrun.

branches: 1.29.22; 1.29.30; 1.29.34;
Clarify comment about SER_SZERR.

branches: 1.28.112; 1.28.122; 1.28.128;
merge ktrace-lwp.

Add definitions for microSPARC-IIep memory fault registers.

branches: 1.26.12;
Bits 0 (MMU Enable) and 1 (Fault inhibit) are common among the implementations
of the SRMMU control register. Reflect that fact in the definitions here.

Also add the swift `store allocate' bit.

Uppercase CPU, plural is CPUs.

branches: 1.24.2;
Add some more definitions: SRMMU and MXCC reset register.

Add Sun4d cpu-unit ASIs.

branches: 1.22.8; 1.22.12; 1.22.20;
Define some more bits in the MXCC control register.

Add definitions of HyperSPARC block fill/copy ASIs.

Rename MXCC control register.
Add MXCC stream register definitions.

branches: 1.19.8;
Define Hypersparc ICCR bits.

Name fault registers more like they're referred to in various docs.

Use the new "%b" format.

Define hypersparc instruction-cache flush ASI.

Remove a couple of "ms2 only" from cache-flush ASIs.

Define the turbosparc's processor configuration register.

For each major "module", define a separate set of MMU control register bits.

Copyright police (s/Harvard University/Harvard College/).

remove some RCS id's we don't need.

Add SRMMU/sun4m definitions.
Cleanup (i.e. mostly delete) `#if defined (SUN4*)' in here.

Define VIDEO enable bit (per Jason Thorpe; PR#1672).

add a couple more sun4m bits.

Add `region' ASI.

copyright/Id cleanup

vme, and cleanup some messy #ifdef stuff

add sun4 control space areas

add srmmu flush/prope ASI sub-codes

add sun4m ASI definitions

Chris Torek's sparc port. Missing lots of things.

Build db_machdep.c when KGDB option is enabled.
Do not cast (regs)->db_tf.tf_pc to db_addr_t (thus ifdef block is redundant).
Adjust ifdef conditions in db_interface.c to make it build with KGDB option.
While here, add #endif comments for longer blocks.

These changes should make sparc build with KGDB option, once ddb/db_access.c
code will be fixed to build with it.

avoid using 'extern <func|data>;' inside a .c file, but instead
use header files and ensure definitions are not duplicated or
are technically (if not in codegen) wrong.

branches: 1.28.16;
Cleanup and clarify the ELFSIZE mess:

We now have 2 variables automatically set in elf_machdep.h:

ARCH_ELFSIZE: the size for userland binaries
KERN_ELFSIZE: the size for the kernel binaries

DB_ELFSIZE has been deleted and KERN_ELFSIZE should have always the
same values DB_ELFSIZE used to have.

In sys/exec_elf.h, if ELFSIZE is not set, it is set to KERN_ELFSIZE
for the kernel and ARCH_ELFSIZE for userland. These defaults should
eliminate the need for most manual ELFSIZE setting.

rearrange slightly to enable crash(8) to build.

branches: 1.26.14; 1.26.32; 1.26.36;
Introduce DDB_EXPR_FMT and replace the logic around DB_EXPR_T_IS_QUAD.

obsolete DB_AOUT_SYMBOLS. however, we need to leave most of the code
in db_sym.[ch] as it is used by the elf version of crash(8).

i will be cleaning up the db_sym.c code in a follow up commit to avoid
having dead code compiled.

branches: 1.24.2;
Retire struct user, remove sys/user.h inclusions. Note sys/user.h header
as obsolete. Remove USER_TO_UAREA/UAREA_TO_USER macros.

Various #include fixes and review by matt@.

branches: 1.23.68;
Replace the Mach-derived boolean_t type with the C99 bool type. A
future commit will replace use of TRUE and FALSE with true and false.

branches: 1.22.14;
closes: PR kern/32359

modifies machine/db_machdep.h: BKPT_SET(inst) to BKPT_SET(inst, addr) for all archs ie; passess the
breakpoint address as well.

Patch from cherry@mahiti.org

branches: 1.21.2; 1.21.4; 1.21.6;
De-__P.

branches: 1.20.18; 1.20.32; 1.20.34;
Add a BKPT_ADDR() macro which gives MD code a chance to munge a
breakpoint address before it's used. Currently a no-op on all but sh5.

This is useful on sh5, for example, to mask off the instruction
type encoding in the bottom two address bits, and makes it possible
to do "db> break $rXX" instead of manually munging the address.

Simplify ddb register storage setup: remove MULTIPROCESSOR special cases
and keep the ddb register copies on the current stack always.

* Maintain a pointer to the cpu_info structure of the CPU being examined.
* Force cpu_Debugger() to have a stack frame, so tracing can at least
start off matching arguments and function calls correctly.

Remove unneeded declarations of the db_machine_init() function. The
ARM ports are the only ones that actually have one, and it is about
to change.

branches: 1.16.2; 1.16.6; 1.16.8;
for now, KGDB kernels do not have access to multiple cpu's trapframe.

if MULTIPROCESSOR, make the registers virtual, so that we can access other
processors registers. rework enter/exit ddb code to look more like the i386
code. make "mach cpu" and "mach cpu N" work.

branches: 1.14.2;
remove include of <vm/vm.h> and <machine/pmap.h>. <vm/vm.h> -> <uvm/uvm_extern.h>

branches: 1.13.2;
Compile in support for ELF symbols.

branches: 1.12.8;
Convert to [pv]addr_t & [pv]size_t.

Code for software-assisted DDB single-stepping from Chuck Silvers.

Switch to MI kgdb code.

branches: 1.9.4;
#define DB_AOUT_SYMBOLS

db_expr_t is now a long.

Various cleanup; mostly trailing spaces/tabs.

Add missing prototypes and fix the fp struct for svr4.

A few prototype fixes

Include vm.h, rather than guessing which files we need.

We have more registers.

copyright/Id cleanup

DDB support

Add getlabelusesmbr(), as proposed in
http://mail-index.netbsd.org/tech-userlevel/2011/08/25/msg005404.html
This is used by disk tools such as disklabel(8) to dynamically decide is
the undelyling platform uses a disklabel-in-mbr-partition or not
(instead of using a compile-time list of ports).
getlabelusesmbr() reads the sysctl kern.labelusesmbr, takes its value from the
machdep #define LABELUSESMBR.
For evbmips, make LABELUSESMBR 1 if the platform uses pmon
as bootloader, and 0 (the previous value) otherwise.

Put native NetBSD disk label in a piece of wasteland within a SunOS label.

Enable writing of disklabels (Many thanks to Jason Thorpe, Gordon Ross
and Matthew Green).

Zap duplicate RCS id.

copyright/Id cleanup

move arch dependence to archs. clean up a bit. deal with weird MAXPARTITONS.

src is too big these days to tolerate superfluous apostrophes. It's
"its", people!

branches: 1.9.44;
Remove clause 3 and 4 from TNF licenses

branches: 1.8.68; 1.8.70; 1.8.72;
De-__P.

branches: 1.7.4; 1.7.6;
merge ktrace-lwp.

branches: 1.6.2;
Move common Sun3/Sun4 EEPROM definitions in to <dev/sun/eeprom.h>

Fix some whitespace.

branches: 1.4.6;
sync these with the sparc64 versions.

branches: 1.3.6;
Provide a definition for DEV_EEPROM so mem.c will compile if SUN4 is defined.

branches: 1.2.46; 1.2.50;
Export `eeprom_va' so drivers can peek at the configuration (from Jason Thorpe).

eeprom device for Sun4 from Jason Thorpe.

sparc/elf_machdep.h: organize symbols into psABI & local groups.

Allow this file to be used with <sys/elfdefinitions.h>.

PR lib/59564

Cleanup and clarify the ELFSIZE mess:

We now have 2 variables automatically set in elf_machdep.h:

ARCH_ELFSIZE: the size for userland binaries
KERN_ELFSIZE: the size for the kernel binaries

DB_ELFSIZE has been deleted and KERN_ELFSIZE should have always the
same values DB_ELFSIZE used to have.

In sys/exec_elf.h, if ELFSIZE is not set, it is set to KERN_ELFSIZE
for the kernel and ARCH_ELFSIZE for userland. These defaults should
eliminate the need for most manual ELFSIZE setting.

Add sparc/sparc64 support for irelative relocations.

PR port-sparc64/51925: cosmetic cleanup, no functional change

branches: 1.7.22; 1.7.40; 1.7.44; 1.7.48;
Add TLS relocation definitions.

branches: 1.6.122;
Add support for dumping ELF-cormat core files.

branches: 1.5.8; 1.5.12;
Move dl* function definitions to libc on ELF.
Based on the patch supplied by Takuya Shiozaki <tshiozak@astec.co.jp>.
See http://mail-index.netbsd.org/tech-userlevel/2000/02/23/0000.html.

Update to match new SVR4-style definition names in <sys/exec_elf.h>.

branches: 1.3.2; 1.3.4; 1.3.6;
Let NetBSD/sparc know about 64-bit formats even if it can't execute them.

branches: 1.2.10;
Add relocation definitions used by ld.elf_so.

add and use a machine-dependent header, which currently defines some
macros to use to remove #ifdefs from the machine ID case check.
Eventually, these headers will contain other information, e.g.
machine-dependent relocation information, etc.

branches: 1.1.2;
Move the complex logic for dynamically writing branches from ld.elf_so
into a header for reuse in crt0.o for static ifunc support. Change the
existing logic for sparc64 to use the Bicc variant of ba,a as it allows
+-8MB displacement compared to the BPcc variant's +-1MB. Teach the sparc
variant the same trick for using ba,a and not sethi+jmp when possible.

In the `MY THAT'S GROSS' department...
Eliminate the recursive include of machine/endian.h from sys/endian.h.

Foolish consistency. Mainly, always use underscores and sys/endian.h.

branches: 1.11.2;
Include <sys/endian.h> after defining whether where are little- or
big-endian. i386, pc532 and vax still include <machine/byte_swap.h>
and define macros for the {n,h}to{h,n}*() functions. mips also
defines some endian-independent assembly-code aliases for unaligned
memory accesses.

Mark [hn]to[nh][ls]() with __const__, so they are subject to CSE.

Move the bswap functions from libutil to libc (this bups the
minor of libc and the major of libutil). For little-endian architectures
merge the bnswap() assembly versions with nto* and hton* using symbols
aliasing. Use symbol renaming for the bswap function in this case to avoid
namespace pollution.
Declare bswap* in machine/bswap.h, not machine/endian.h. For little-endian
machines, common code for inline macros go in machine/byte_swap.h
Sync libkern with libc.
Adjust #include in kernel sources for machine/bswap.h.

(Always) (practice) (safe) (macro expansion).

branches: 1.7.4;
Add byte-swapping functions (bswap16, bswap32, bswap64) to libkern.
Only assembly version for i386 bswap16 and bswap32 for now (bswap64 uses
bswap32). Contribution of assembly versions of these are welcome.
Add byte-swapping of ext2fs metadata for big-endian systems.
Tested on i386 and sparc.

branches: 1.6.10;
add in_addr_t and in_port_t

Back out previous; this is a bigger problem than I expected.

The htonl(), htons(), etc. functions return u_long and u_short. Add
casts to the big-endian versions of these calls so that the compiler
won't complain when they're passed as arguments to printf().

- Fix setcxsegmap prototype.
- Cast ?TO? macros to NULL to avoid compiler warnings

copyright/Id cleanup

need these

branches: 1.1.6;
In the `MY THAT'S GROSS' department...
Eliminate the recursive include of machine/endian.h from sys/endian.h.

moved to aout_machdep.h (via repository copy)

copyright/Id cleanup

Fix spelling error in Copyright notice

fix a typo

Put definition of __LDPGSZ and `struct relocation_info' where they belong:
in <machine/exec.h>.

actually put something in a previously empty file.
__LDPGSZ should now be defined here!

stdarg.h, varargs.h, psl.h, autoconf.h: posted patches from torek
cpu.h: cpu_adjstack() won't work well in the NetBSD execve - need a
different model, stub enablertclock() for now..
exec.h: a netbsd-type exec.h file (ie. empty)
param.h: add MID_MACHINE
pmap.h: delete kernel_pmap hack - it doesn't work for us, + two proto's
vmparam.h: VM_MIN_ADDRESS = 0!

Chris Torek's sparc port. Missing lots of things.

fbio.h moded to sys/dev/sun.
Retain symlink from /usr/include/machine.

branches: 1.6.2; 1.6.12;
Initial stab at a P9100 console FB driver for the Tadpole 3GX.
It supports RASTERCONSOLE only and the colormap support is
broken (the hardware doesn't seem to be doing what the tech.
docs. say it should be doing). But it is usable as a console
(but with blue on white as the only color choice). I figured
I'd check it in since it is somewhat usable and someone else
might figure what I screwed up in the color map support.

branches: 1.5.24;
Add cgfourteen to conf.h.
Add new framebuffer types for recent-model graphics devices to fbio.h.

Various cleanup; mostly trailing spaces/tabs.

copyright/Id cleanup

add things needed by the cgsix

fbio.h, mailed from Torek. This file is from Sprite, and clearly an
extensively modified SunOS file. *sigh*.
(ie. Feel free to rewrite it from scratch...)

Framebuffer code common to sparc and sparc64 was merged and moved to
dev/sun 3 years ago. Remove these redundant and out-of-date copies.
All code now uses dev/sun/fbvar.h

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.13.24;
Change the kernel mmap interface so that the offset to map is an
"off_t" and the return value is a "paddr_t" to allow mappings
at offsets past 2^31 bytes. Somewhat inspired by FreeBSD, which
only changed the offset to a "vm_offset_t".

Includes updates for the i386, pc532 and sh3 mmmmap from Jason Thorpe.

branches: 1.12.4;
Add `fb_is_console()', which is a helper function for frame buffer
device drivers used to decide whether or not to act as console output.

branches: 1.11.2;
Bring sparc/sparc64 fb code and kernel configuration into line with
rcons changes.

branches: 1.10.10;
Split the fb_setsize() utility in two separate functions: one to be
used by OBP machines and another one for sun4/eeprom machines.

fbrcons_{rows,cols} helper functions.

Returning ENODEV is not correct in a poll entry point. Instead, since
read/write returns an error, use seltrue().

First-cut support for P4 framebuffers. Currently supports P4 variants
of bwtwo, cgfour, cgsix, and cgeight. Includes support for attaching
the overlay plane of the cgfour and cgeight as a bwtwo instance (a'la
SunOS).

P4 register descriptions, cgfour driver, and cgeight driver partially
derived from OpenBSD.

This code may still need polishing.

Code from Jason to better configure /dev/fb.

branches: 1.5.2;
Patch from Jason to factor out a common framebuffer size determination
routine from the individual drivers.

RCONSOLE => RASTERCONSOLE.

Remove all rcons related stuff from `fbdevice'; this is now defined in
dev/rcons/rcons.h.

copyright/Id cleanup

Chris Torek's sparc port. Missing lots of things.

Make the copies of the FE_* macros match verbatim to avoid a gcc warning
when both ieeefp.h and fenv.h are included.

branches: 1.1.2; 1.1.6; 1.1.18; 1.1.36; 1.1.40;
Add fenv support for sparc. Mostly copied from sparc64 and share with it.

Sprinkle <sys/featuretest.h> where _*_SOURCE macros are consulted.

Otherwise, the feature tests might come out wrong. For example,
header files that check for _NETBSD_SOURCE won't get the default when
no other _*_SOURCE macros are defined; header files that check for
_POSIX_C_SOURCE might miss _XOPEN_SOURCE, which is supposed to imply
a corresponding _POSIX_C_SOURCE.

PR lib/58752: various header files test _*_SOURCE macros but don't
include sys/featuretest.h

branches: 1.12.100;
Remove double include that prevented the right machdep LDBL_* constants
to be used.

Noticed by Masao Uebayashi.

move the float definitions all into <sparc/float.h> and make that
version work in either 32 or 64 bit mode.

branches: 1.10.104; 1.10.108; 1.10.114;
Rename <sys/float_ieee.h> to <sys/float_ieee754.h>, following libc's
convention for these.

PR/3012: Greg A. Woods: Write all float.h files [except the vax of course]
in terms of float_ieee.h

Minor changes to make all the float.h files match.

Prototype __flt_rounds() consistently.