Explicitly set -z noseparate-code because the default is about to change
and the boot blocks might not be able to load binaries with more than two
PT_LOAD sections.

Fix GCC flags for Spectre mitigation to really only apply to GCC.

Merge the bouyer-xenpvh branch, bringing in Xen PV drivers support under HVM
guests in GENERIC.
Xen support can be disabled at runtime with
boot -c
disable hypervisor

branches: 1.194.10;
- Determine KERN_AS automatically depending on whether OPT_MODULAR is
set or not, in the same way as libcompat.

- Specify OPT_MODULAR in the port Makefile instead of KERN_AS.

Now, KERN_AS=library is used for kernels without module(7) for all ports.

OK christos

branches: 1.193.2;
No need for NORELRO here.

Disable RELRO/SANITIZER

Make SPECTRE_V2_GCC_MITIGATION overridable at the make command line.

add an SPECTRE_V2_GCC_MITIGATION option to x86 kernels, that turns
on the GCC spectre v2 mitigation options.

XXX: pullup-8.
XXX: turn on in all kernels.

branches: 1.189.2;
Revert previous, we'll need something better (and compatible with Clang).

Add -fno-shrink-wrap, to force GCC to push the frames at the very beginning
of the functions. Otherwise DDB is unable to display a correct stack trace
if a fault occurred in a function before the frame was pushed.

Discussed on tech-kern@, flag suggested by Krister Walfridsson. Should fix
PR/52560.

- Allow multiple .BEGIN targets
- Make their protection consistent

Remove several dead entries from the x86 makefiles. Looks like people (me
included) regularly forget to take care of this.

branches: 1.185.10;
Define ${LINKSCRIPT} in one place.

${KERN_LDSCRIPT} -> ${KERNLDSCRIPT}

branches: 1.183.2;
Use LINKSCRIPT.

Now that GCC 4.8 on i486 has stolen the frame pointer, do like amd64 and
add -fno-omit-frame-pointer to kernel DEFCOPTS and modules CFLAGS.

branches: 1.181.2;
Like amd64, use -msoft-float.

Change the compiler options to explicitly specify:
-mno-mmx -mno-sse -mno-avx -mno-80387 -mno-fp-ret-in-387
Since no-sse implies no-sse2 that should ensure that the compiler really
doesn't emit any instructions that might trap trying to use the FPU.
On amd64 at least some of those are needed to stop the compiler
saving the registers to stack on every varargs function.
It might be that -mno-sse did that before.

Reduce amount of -no-integrated-as on x86 as .code16 is now supported by
LLVM.

branches: 1.178.2;
Remove all references to KVM86.
It was only ever used by APMBIOS - and then only if an option was selected.
Probably didn't work well at all!

apmcall.o no longer exists.

branches: 1.176.6;
Move Clang-warning flags for ah_regdomain.c into sys/conf, they are
platform independent. Use the selective -Wno-* flags instead of
-Wno-error.

branches: 1.175.2; 1.175.6;
include wakecode after so that the depend rules are defined!

Try to ensure no SSE instructions are generated for kernel code,
as is already done on amd64.

Use integrated assembler with clang for Via PadLock.

Add a few more cases that need GNU as. One of the padlock instructions
is missing in LLVM. spl.S and lock_stubs.S break with patchfunc
otherwise.

branches: 1.171.2;
Disable -Werror for ah_regdomain.c if building with clang as workaround
for http://llvm.org/bugs/show_bug.cgi?id=10030.

Use -x assembler-with-cpp from sys.mk.

LLVM's assembler parser doesn't support .code16 yet, so disable it as
needed.

branches: 1.168.2;
Allow use of traditional CPP to be set on a per platform base in sys.mk.
Honour this for dependency processing in bsd.dep.mk. Switch i386 and
amd64 assembly to use ISO C90 preprocessor concat and drop the
-traditional-cpp on this platform.

remove almost all the ability to build netbsd with an a.out target.
we're ELF now, and there are many missing checks against OBJECT_FMT.
if we ever consider switching, the we can figure out what new ones
we need but for now it's just clutter.

this doesn't remove any of the support for exec_aout or any actually
required-for-boot a.out support, only the ability to build a netbsd
release in a.out format. ie, most of this code has been dead for
over a decade.

i've tested builds on vax, amd64, i386, mac68k, macppc, sparc, atari,
amiga, shark, cats, dreamcast, landisk, mmeye and x68k. this covers
the 5 MACHINE_ARCH's affected, and all the other arch code touched.
it also includes some actual run-time testing of sparc, i386 and
shark, and i performed binary comparison upon amiga and x68k as well.


some minor details relevant:
- move shlib.[ch] from ld.aout_so into ldconfig proper, and cut them
down to only the parts ldconfig needs
- remove various unused source files
- switch amiga bootblocks to using elf2bb.h instead of aout2bb.h

branches: 1.166.2; 1.166.4;
Enable SSP (Stack Smash Protection) in x86 kernels by default (except
in i386 *TINY kernels). The NetBSD/i386 "ALL" kernel is unconditionally
compiled with SSP enabled.

Change approved by the core team.

Make sure to create i386 symlink on more generic targets.

Convert the recently introduced awk command to retrieve the value of the
option MODULAR to using %MODULAR%. While it is now possible to only
request the new version in the affected Makefiles, it is made mandatory for
everybody because I just fixed a bug in config(1) that would not make it
fail in the case of a syntax error in the Makefile template.

Now that "options MODULAR" is handled via defflag, the Makefile
generated by config(1) in the kernel compile directory no longer
contains IDENT=-DMODULAR. Instead, there's an opt_modular.h file that
might or might not contain "#define MODULAR 1".

Adapt to this by adding an OPT_MODULAR variable to relevant Makefiles,
set via an awk script that parses opt_modular.h.

branches: 1.162.2;
Clean-up makefile stub used to include in the build the binary HAL object

Another assym.h dependency.

LKM -> MODULAR

branches: 1.159.2; 1.159.4; 1.159.8;
provide an i386/ symlink.

branches: 1.158.6; 1.158.10; 1.158.16;
Refactor in_cksum/in4_cksum/in6_cksum implementations:
- All three functions are included in the kernel by default.
They call a backend function cpu_in_cksum after possibly
computing the checksum of the pseudo header.
- cpu_in_cksum is the core to implement the one-complement sum.
The default implementation is moderate fast on most platforms
and provides a 32bit accumulator with 16bit addends for L32 platforms
and a 64bit accumulator with 32bit addends for L64 platforms.
It handles edge cases like very large mbuf chains (could happen with
native IPv6 in the future) and provides a good base for new native
implementations.
- Modify i386 and amd64 assembly to use the new interface.

This disables the MD implementations on !x86 until the conversion is
done. For Alpha, the portable version is faster.

Remove unnecessary references to config_time.h.

Missing assym.h dependency.

i386func.S needs assym.h.

branches: 1.154.2;
Merge jmcneill-pm branch.

branches: 1.153.2; 1.153.4;
busfunc needs assym.h too.

branches: 1.152.2;
cpufunc.S depends on assym.h.

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: 1.150.8; 1.150.10; 1.150.14;
lock_stubs.S depends on assym.h

branches: 1.149.12; 1.149.14; 1.149.20;
There is no need to compile machdep.c with -O2 anymore.
Fixed by changing the asm for lidt() in cpufunc.h

branches: 1.148.2;
mention the correct PR/33725

workaround for PR 33918

Another missed dependency.

branches: 1.145.2;
convert to timecounters (from branch simonb-timecounters)

branches: 1.144.2;
Convert existing ath users to new Makefile, except for amd64, which needs
the rules due to needing to conditionally postprocess the HAL object file.

Macppc needs a a non-ELF HAL (EABI) object, so take care of that by default
in the atheros include file.

Reorganize ath layout as requested by sam@ and suggested by dyoung@ in
http://mail-index.netbsd.org/tech-net/2006/03/15/0000.html.

The new layout almost precisely matches FreeBSD, and should make
future imports much easier.

At the same time, import the current 0.9.16.16 HAL from FreeBSD. According
to sam@, this is the proper version we should be using.

branches: 1.142.4; 1.142.6; 1.142.8; 1.142.10; 1.142.12;
merge ktrace-lwp.

some assym cleanup.
- move copyin and friends from locore.S to their own file, copy.S.
share it between i386 and xen.
- defparam KERNBASE and kill KERNBASE_LOCORE hack.
- add more symbols to assym.h and use it where appropriate.

No need to create opt_ah.h in these makefiles, config(8) takes care
of it.

branches: 1.139.2; 1.139.4;
Resolve conflicts in importation of 18-May-2005 ath(4) / net80211(9)
from FreeBSD. Introduce compatibility shims (sys/dev/ic/ath_netbsd.[ch],
sys/net80211/ieee80211_netbsd.[ch]). Update drivers (an, atu, atw,
awi, ipw, iwi, rtw, wi) for the new net80211(9) API.

s/GENASSYM/GENASSYM_CONF/ so we can use "GENASSYM" as the program name.

Use an archive library for libkern if we don't support LKMs.
Enough to get INSTALL_TINY small enough again.
Someone else can fix it next time...

branches: 1.136.4; 1.136.6;
Make uudecode into a tool that can be used to build a kernel and redo
the athhal-elf.o handling to take advantage of it. This means that
bsd.files.mk is no longer included by Makefile.i386 so CFLAGS won't be
duplicated.

Add the %MAKEOPTIONSAPPEND token at the end of the file, after the
common Makefile.kern.inc has been included.

We need to also clean out athhal-elf-o, so that "make clean; make"
after an initial "make" will also (re)build athhal-elf.o.

branches: 1.133.2;
Convert athhal-elf.o to a .uue and add appropriate make hooks to use it

Make mptrap.o depend on assym.h; fixes GENERIC.MP -j N kernel build.

branches: 1.131.2;
add spl.S, vector.S and microtime.S to MD_SFILES so that
'make depend' takes care of them.

bioscall.S needs assym.h now.

microtime.S depends on assym.h

add missing dependency.

The traditional C preprocessor is fixed in GCC 3.3 now, so no need
to skip -traditional-cpp for HAVE_GCC3.

add mach_sigcode.o to the files that depend on assym.h

New interrupt code. The basic idea behind it is to hide the differences
in interrupt controllers in struct pic, and try to keep as much
common code as possible. At the lowest (asm) level, this is done
with CPP macros.

The main structure is now struct intrsource, describing an established
interrupt line, of any kind (soft/hard local apic/legacy apic/IO apic).
For quick masking, there may be a maximum of 32 sources per CPU.
Sources can be assigned to any CPU in the MP case, though currently they
all go to the boot CPU.

We don't need our on KERNLINTFLAGS any longer.

branches: 1.123.2;
Don't use -traditional-cpp if HAVE_GCC3.

branches: 1.122.10;
Roll the rest of the ports over to the new MI kernel build machinery.
Any problems reported by testers have been fixed, and massive
cross-compiling of kernels has shown that any problems that remain
with actually building kernels are not related to this.

Instead of checking for previous usage, just append to some make
variables. Tweak some variables to provide for more uniform defaults.
Minor adaption to makefiles for already converted ports.

cleanup:
options SPACE TAB
makeoptions TAB
psuedo-device TAB
remove trailing whitespace
replace multiple spaces -> tabs
options "FOO" -> options FOO
options "FOO=bar" -> options FOO=bar
options "FOO=\"bar\"" -> options FOO="\"bar\""

Abstract out a large chunk of the kernel build machinery so that new
build features (such as ross's DEBUGLIST) can easily be applied to all
ports. This should reduce the complexity of each port's kernel
Makefile considerably. Line counts:

227 arch/i386/conf/Makefile.i386.orig
98 arch/i386/conf/Makefile.i386
227 arch/alpha/conf/Makefile.alpha.orig
99 arch/alpha/conf/Makefile.alpha
219 arch/sparc/conf/Makefile.sparc.orig
102 arch/sparc/conf/Makefile.sparc
215 arch/vax/conf/Makefile.vax.orig
102 arch/vax/conf/Makefile.vax
253 conf/Makefile.kern.inc

Roll i386, alpha, sparc, and vax over to the new build machinery.

Change defaults for kernel compiles. Default all to USETOOLS?=no and have
the etc Makefile override that by putting USETOOLS into $.MAKEOVERRIDES
This way the default for kernel compiles is still to use the installed
toolchain instead of depending on $TOOLDIR. $TOOLDIR can be used by
simply adding USETOOLS=yes to the command line as usual.

Adjust each ports template to set the default no setting and also pull in
bsd.own.mk if they weren't already to ensure they'll build correctly
with the new toolchain setup.

branches: 1.117.2;
Set MACHINE_ARCH explicitly in Makefiles for which it is constant.
Also, since config(8) now explcitly sets MACHINE, there is no need
to do it here in the Makefile.

branches: 1.116.2;
Allow the text address to be set with makeoptions.

define _KERNEL_OPT as well as _KERNEL. we will use this in the future to
get kernel "opt_foo.h" headers, rather than _KERNEL && !_LKM.

changes to lint invocation:
- sort lint flags
- add flags -F (full pathname), -g (gcc extensions; XXX for now)
- lint CFILES before LINTSTUBS
- don't -Dvolatile=
- don't grep -v 'static function.*unused'

1) Add infrastructure to generate lint stub files from stylized
comments in .S files.
2) Update "make lint" target to use generated lint stubs.

branches: 1.112.2;
a few things depend upon assym.h, so add appropriate dependencies.

In the ELF case, use the kern.ldscript unless overridden by the kernel
config file.

Revert apparently accidental commit.

Add isdn devices to all archs that seem to support at least one of the
necessary busses. Sync comments with reality for archs that already had
the devices.

delete obsolete comment

If KERN_LDSCRIPT is specified as a makeoption, then add it to the
linker flags.

Add "install" target, so "make install" after building the kernel
does something useful. The target can be redefined by putting a
install-kernel-${MACHINE_NAME} target that fits your needs into
/etc/mk.conf.

HAVE_GCC28 -> HAVE_EGCS, and make it match gcc 2.9 as well. Also,
make the use of -Wno-uninitialized explicit, don't rely on a compiler
hack to do it for us.

Move KERNBASE to 0xc0000000, giving the kernel a more-or-less full 1G
of virtual address space, leaving userland with 3G, and update comments
to match the new reality.

We knew we were going to have to bite this bullet eventually, and there
are a couple of outstanding PRs related to this issue (9389 and 9313).
Complete solution to those PRs is going to involve some sort of run-time
decision on how large kmem_map should be, as well as changing some data
structure allocation strategies in UVM. However, this change will at
least allow the PR submitter to simply throw resources at the problem.

Revert STRIPPROG -> STRIP

Add "dependall" target for comfort.

STRIP -> STRIPPROG

Change STRIPFLAGS=-g for cross-toolchains.

branches: 1.99.2;
be more efficient when doing the SYSTEM_LD_TAIL for -g kernels: don't
copy them just to strip them, use strip -o.

clean up a bit after jason :)

- fix emitrules() like emitfiles() to deal with the prefix (otherwise it
would attempt to find the file in the normal base for the NORMAL_C rule).
- add emitincludes() which adds include directives for each prefix to the
$INCLUDES variable in the makefile.
- add %INCLUDES to each Makefile.arch to deal with the above.

this makes "prefix" actually work in a usable manner, and now i can move
on to fixing compiler warnings (errors) in the ESP code. :)

branches: 1.97.2; 1.97.4;
Add ELF bits. -N does not work, because ld does not compute the correct
memsz.

Some minor, mostly costmetic, changes to CPPFLAGS/CFLAGS.

Add -Wpointer-arith warning since `void *' arithmetic is not ANSI C.

change handling of libkern:
-sys/lib/libkern builds as library per default (as it was documented all
the time)
-ports able to LKM set "KERN_AS=obj" explicitely in their Makefiles
(for now; should depend on actual "option LKM" or -better- functions
included for LKM use should be pulled in by a stub)
-always link libcompat before libkern - libkern stuff can be referred to
by libcompat, but not the other way

distclean is a synonym for cleandir

Also, remove space at the end of line.

Specify a non-profiling C rule.

Add -Wno-main conditional on compiler being gcc 2.8 or egcs. (This adds
a HAVE_GCC28 check-variable that can now be used to add other gcc-2.8
flags in cases where they may be useful, or to remove gcc 2.7.2 "bug
workaround" flags.)

add the i386 MD portions for UVM.

Make sure CPP, AR, AS, and RANLIB are defined.

Define LORDER, NM, and TSORT here, like we do the rest of the tools.

branches: 1.86.2;
Move kernel entry address down to f0100000

define SIZE?=size, and use ${SIZE} instead of size. makes cross
compilation easier

PR/4162: Chris Jones: make cleandir does not work properly; it does not
remove the object files and the .depend file in the kernel build directory.

branches: 1.83.2;
Generate assym.h dependencies when making 'depend'.

Put the genassym.sh output in a temp file, and mv -f it, in case
it fails spectacularly and doesn't clean up. Also, remove dependencies for
genassym.o (since it no longer exists).

Use genassym.sh, to support cross-compilation.

branches: 1.80.4;
Nuke some options GENERIC residue.

Nuke some GENERIC residue.

update for new kernel library Makefile.inc expectations

branches: 1.77.2;
* Replace explicit -O2 in CFLAGS with COPTS macro. Default COPTS?= -O2.
Lets users over-ride with makeoptions COPTS="..." in kernel config files.

Leave `mandatory' flags (like -msoft-float which on m68k enforces no
FP in kernel) in CFLAGS.

add -Wall -Wstrict-prototypes -Wmissing-prototypes.
We are compiling cleanly now!

BIOS trampoline is a static beast; generate it in the sources like
syscalls.h & friends, and include it when needed from the source directory.

Move strip(1) flags into a separate variable, so that $STRIP can
be passed to subordinate make(1)s. Remove $TOUCH. Add HOSTED_CC,
HOSTED_CPPFLAGS, and HOSTED_CFLAGS, and use them when depending genassym.

Wrap the default definition of `S' in `.ifndef'.

add Advanced Power Management driver
Includes bootstrap support from FreeBSD (Tatsumi HOSOKAWA and <ukai>)

Put into . Remove hooks for `config-dependent' and
`device-driver' flags.

Theres no microcode/Makefile.inc now.

* Add a HOSTED_C_C variable, which strips `-p', `-pg', and
`-nostdinc', and use it when building genassym.
* Use `-nostdinc' just to be sure we're self-contained.

* Define CWARNFLAGS and MKDEP in some moderately consistent fashion.
* Make S expand to an absolute path at compile time.
* Use `-S' rather than `-x' to remove debugging symbols.
* Garbage collect unused variables.
* Reverse a handful of port-specific changes that do not correspond to
the common build model and are not needed.

branches: 1.67.4;
Make `make depend' work for libraries.

Do not build vnode_if.[ch] for each kernel. Build them once, like the
various syscall sources/headers, and just compile them. From PR 2142, OK'd
by mycroft.

Define build programs with `?=', so they can be overridden with `makeoptions'.

Fix mkdep problems due to missing flags.

Use `-traditional-cpp' when building .s and .S files.

Clone these, and fix many bugs.

assym.s -> assym.h
(Some ports did this already.)

LOCORE -> _LOCORE

Pull in microcode dependencies. Remove -fno-builtin.

Compile the kernel with -fno-builtin.

Add -Werror.

GENERIC: Add COMPAT_IBCS2 in too by default.
Makefile.`arch`: - Add -lcompat
- Change clean: target to be a double dep one,
so libraries can use the clean target too
- Add genassym to the clean targets

Removing -DKERNEL, transition to _KERNEL has been completed

clean up several ISA device interfaces: autoconfiguration, header
inclusion, and interrupt configuration. more work still needs to be done,
but it's getting better...

allow most program names to be overridden. kill a bogus } in a comment, and
an unnecessary blank line.

Added -D_KERNEL to CFLAGS alongside each -DKERNEL.

Don't ignore the exit status of ld(1).

Add `newvers' target.

vector.h should have been nuked ages ago.

isa.h -> isareg.h

Add vnode_if.[ch] to the clean target.

new RCS ID format.

Allow overriding the name of the awk program. This is a bit different from
how Gordon did it.

build vnode_if.* early in a kernel compile

Build vnode_if.o.

Clone from hp300 version.

kill symbols.{raw,sort}; no longer necessary, with kvm dbs.

kill refs to dbsym, and note that we have a working 'ex' now.

LOAD_ADDRESS is now fixed at 0xf8100000.

compile genassym -static as suggested by Jim McKim <mckim@lerc.nasa.gov>

Clean bogus arguments to newvers, and don't pass KERN_IDENT any more.

correct docs

Fixed typo.

Explicit entry point specified in kernel link command.
(so far, we've been lucky only because vers.o doesn't contain text)

use the new location for libkern, auto-depend on it, and include
the right one, for profiling or non-profiling.

genassym needs a dependency...

get rid of excess target line for machdep.o

Default LOAD_ADDRESS at 0xfe100000 (1M)

Link with libkern.

branches: 1.28.2;
Changes for boot loading of symbols.

Need LOAD_ADDRESS for depend pass.

Add RCS indentifiers.

branches: 1.25.2;
Add ${DEBUG} to CFLAGS and -f to dbsym.

Phooey. Need a better mechanism for dealing with multiple sets of local
changes.

CC should be cc. cc is gcc2...

Change to work with new config stuff for specifying load address.

Oops. Back out a change that shouldn't be committed...

Use new -T option to dbsym.

gcc-2.4.5's ffs() has been rendered safe for kernel compiling...

add -fno-builtin to CFLAGS until gcc bug fixed

don't ignore errors from dbsym... it might say that, e.g. there's
not enough symbol space!

Use NNPX instead of NPX from npx.h for activating npx.

note gprof.ex is in csu/i386 not csu.i386 like it used to be...

make conf.o actually depend on conf.c...

incorporate Bruce Evans' interrupt changes, as hacked by Rod grimes,
for patchkit patch 158. it appears to work well.

do the right thing for vers.o. i.e. don't make newvers if nothing
has changed.

changes for 'arch'

make config use a /sys/${machine}/compile directory, rather than
/sys/compile

add rcsid, and kernel description where necessary

use ed instead of ex. the script to use is identical, and we might
want to switch back to using ex when our ex supports -.

bsd->netbsd

new kernel name is bsd

now use absolute path for dbsym

automatically run dbsym on kernel

added support for microtime routines by Steve McCanne (mccanne@ee.lbl.gov)

after 0.2.2 "stable" patches applied

branches: 1.1.1;
Initial revision