use new bsd.syscall.mk

branches: 1.7.2; 1.7.4;
Remove spurious bsd.kinc.mk include, which seems to not be needed here
and does not work at all (missing bsd.own.mk for ACTIVE_CC) anyway.

branches: 1.6.112; 1.6.116;
merge ktrace-lwp.

make this rebuild generated files by default

branches: 1.4.8; 1.4.10;
We have CVS; there is no reason to make .bak files when generating the
syscall tables.

Use ${HOST_SH} instead of `sh'.

If necessary, pull in <bsd.sys.mk> to get the definition of HOST_SH;
Makefiles that pull in one of (most of) <bsd.*.mk> will get this anyway.

branches: 1.2.42;
move sparc32 to netbsd32; split out MD part (signal handling) (mostly via repository copy, rename and perl -pi :-)

add a 32-bit compatibility module for the sparc64 port, so it can run NetBSD/sparc binaries with a LP64 kernel.

version dup3

COMPAT_NETBSD32: Add support for epoll(2).

Create compat_100 module infrastructure now that we have branched
for NetBSD-10

Compile in compat32 syscall names all the time, as we do for native
syscall names. Needed for dtrace.

branches: 1.52.4;
Add ACL support for FFS. From FreeBSD.

Add a NetBSD native futex implementation, mostly written by riastradh@.
Map the COMPAT_LINUX futex calls to the native ones.

Split out the quota code from the rest of compat_netbsd32 module. This
allows loading of compat_netbsd32 on kernels that don't have ``options
QUOTA'' enabled.

Add a new member to struct vfsstat and grow the unused members
The new member is caled f_mntfromlabel and it is the dkw_wname
of the corresponding wedge. This is now used by df -W to display
the mountpoint name as NAME=

make things compile again.

Disable netbsd32_drm.c until it receives proper review.

compat32 drm ioctl support from Surya Shankar at GSoC 2019

use option quota

remove dup

Merge the [pgoyette-compat] branch

quotactl is always included, regardless of options QUOTA.

move mmap(), compat 40 mount(), pipe() and getfh() into their own files
as the implementation is non trivial or is compat. mark old mount()
properly instead of manually #if'ing ito to ENOSYS.

move rlimit and quota code into their own modules.

(netbsd32_netbsd.c should have only simple shims. anything more
than copying arguments from one args struct to the other should
not be placed in this file. still a couple more to move out.)

branches: 1.39.14; 1.39.16;
PR kern/51514: ptrace(2) fails for 32-bit process on 64-bit kernel

Updated from the original patch in the PR by me.

branches: 1.38.2;
Missed these.

Only include the compat_14 sys calls if this being compiled as part of
the compat_netbsd32_sysvipc module.

Properly mark netbsd32_mqueue.c to depend on 'options MQUEUE'

Yes, MQUEUE is included in all kernels by default, but it can be removed.
If it is removed, the compat_netbsd32 module should not include it, either.

add netbsd32 support for nfssvc(2). we do this by defining 5 copyin/out
functions that do all the ugly work, are just plain copyin/out for the
native system calls, and do the necessary translations for netbsd32.

with this i'm able to run 32 bit nfsd and mountd on 64 bit kernel and
mount the file systems remotely.

Implement modctl, sigqueinfo and mq_*

branches: 1.34.2;
Define compat modules (but without dependencies yet).

branches: 1.33.10;
compat glue for new lwp_park

branches: 1.32.2;
P1003_1B_SEMAPHORE is no longer optional.

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

branches: 1.30.8; 1.30.12;
Make the SA support as optional as is possible.

branches: 1.29.4;
Move saemul_netbsd32 to netbsd32_sa.c and allow the KERN_SA bits to be omitted.

merge christos-time_t

branches: 1.27.4;
Make the emulations, exec formats, coredump, NFS, and the NFS server
into modules. By and large this commit:

- shuffles header files and ifdefs
- splits code out where necessary to be modular
- adds module glue for each of the components
- adds/replaces hooks for things that can be installed at runtime

branches: 1.26.2;
Merge wrstuden-revivesa into HEAD.

branches: 1.25.38; 1.25.42; 1.25.44; 1.25.48;
Add missing lwp syscalls. Put them all together into a new file,
netbsd32_lwp.c, and remove remaining traces of SA.

This still needs some MD (and possibly MI, depending on the chosen
solution) changes to actually work.

branches: 1.24.4;
- Make better use of COMPAT_XX type in syscalls.master
- Remove useless (thanks to COMPAT_XX behaviour) #ifdefs in
syscalls.master
- Make netbsd32_compat_43.c compiled per COMPAT_LINUX32 because the latter
needs stuff from it.

Fixes Perry's PR#34951.

branches: 1.23.2; 1.23.4;
netbsd32_core.c is need only under options COREDUMP.

Implement the _lwp* and sa_* families of syscalls, using the newly
introduced hooks in the native code.

branches: 1.21.2; 1.21.4; 1.21.6;
Implement the ksem_* family of syscalls.

branches: 1.20.4; 1.20.6;
merge ktrace-lwp.

- Split sys_kevent into kevent1 so that it can be used by COMPAT_NETBSD32
code.

- To achieve COMPAT_NETBSD32 compatibility, introduce a parameter to
kevent1 that points to functions that do the actual copyin/copyout
operations. This is similar to what was done in FreeBSD by Paul Saab.

- Add the COMPAT_NETBSD32 definitions and hooks.

branches: 1.18.2;
64 bit inode changes.

branches: 1.17.12;
o Add support for the statvfs family of syscalls (statvfs1, fstatvfs1,
fhstatvfs1 and getvfsstat)
o Move the statfs family out of netbsd32_fs.c and netbsd32_netbsd.c to
netbsd_compat_20.c, compiled with COMPAT_20

Reviewed by christos@.

clean up a little:
- delete ktrsyscall32()
- add a check #ifdef _LP64 to do the conversion if P_32 is set to the
standard ktrsyscall()
- add a couple of similar _LP64/P_32 checks to the systrace code.

this should get systrace working for 32 bit apps as well as complete
ktrace support for "trace_enter/trace_exit" using platforms such as amd64.

XXX: systrace isn't supported on sparc64 currently... (it doesn't use
trace_enter/trace_exit, or have it's own calls to systrace_xxx()...)

add a copy of the ktrsyscall() entry point for 32 bit emulated calls.
the main purpose of this function is to adjust the "argsize" value of
the ktrace syscall record, otherwise userland will see N/2 (rounded
down) arguments instead of N.

allow COMPAT_SUNOS without COMPAT_43.

Don't put the netbsd32_compat*.c files in the 'compat library'. Unconditionally
compiling these causes a ripple effect making it hard to conditionalize
anything on COMPAT_* in the kernel.

branches: 1.12.10;
build the netbsd32_compat* objects into libcompat.a, so they are available
to emulations not just netbsd itself. rename the compat_10 functions to
match everything else. fixes a problem reported by julian coleman.

Make the coredump routine exec-format/emulation specific. Split
out traditional NetBSD coredump routines into core_netbsd.c and
netbsd32_core.c (for COMPAT_NETBSD32).

branches: 1.10.2;
add netbsd32_uvm_unix.c and netbsd32_kern_sig.c

branches: 1.9.2;
split up netbsd32_netbsd.c into 9 new files, leaving only those syscalls that
have no special interpretations besides simple syscall args conversion.

compat_sunos needs netbsd32_compat_09.c

put a.out/elf32 specific things in netbsd32_exec_{aout|elf32}.c
emul_netbsd32 was moved to netbsd32_netbsd.c
g/c netbsd32_exec.c now

branches: 1.6.2;
Add netbsd32_compat_14.c

Rename everything from compat_netbsd32 -> netbsd32 so we don't have silly
names like compat_14_compat_netbsd32_sys_shmctl().

branches: 1.5.2;
Don't defopt COMPAT_NETBSD32 here; it's done in sys/conf/files.

branches: 1.4.4;
clean and up make compile

move sparc32 to netbsd32; split out MD part (signal handling) (mostly via repository copy, rename and perl -pi :-)

Overhaul. Now will run simple things like /bin/sh and /bin/echo.

add a 32-bit compatibility module for the sparc64 port, so it can run NetBSD/sparc binaries with a LP64 kernel.

COMPAT_NETBSD32: Add support for epoll(2).

COMPAT_NETBSD32: Catch up with sys___kevent100() addition.

XXX
Add epoll_* and memfd_create.

s/strucure/structure/ and s/structues/structures/ in comments.

Implement support for mounting UDF in compat32
Fixes PR#56801

Adapt to recent posix_spawn extensions.

COMPAT_NETBSD32 is all about running the 32-bit flavor of native
binaries on a 64-bit platform[*], as such:
- Make the logic about which "sendsig" flavor to call MI (as it is in the
native 64-bit environment) and follow the same rules as the native 32-bit
environment.
- Make COMPAT_NETBSD32 x COMPAT_16 work the same as it would in the
native 32-bit environment by providing a netbsd32_sendsig_sigcontext_16_hook,
rather than overriding the entire sendsig logic with a netbsd32_sendsig_hook.
- In netbsd32___sigaction_sigtramp(), make sure the compat_netbsd32_16
module is loaded if the trampoline version specifies a sigcontext style
handler, otherwise return EINVAL so that libc can try again with siginfo
style.

[*] ...except for arm32, which uses it to mean "run 32-bit OABI binaries
from the 32-bit EABI environment". Doing it this way was arguably a mistake,
but we are stuck with it for now, so support it by providing a machine-
dependent override for netbsd32_sendsig() that also disables the corresponding
logic in netbsd32___sigaction_sigtramp().

Fixes PR kern/56487.

The read/write/send/recv system calls return ssize_t because -1 is
returned on error. Therefore we must restrict the lengths of any
buffers to NETBSD32_SSIZE_MAX with compat32 to avoid garbage return
values.

Fixes ATF lib/libc/sys/t_write:write_err.

s/u_intN_t/uintN_t/

Handle FSSIOCSET and FSSIOCGET; vndconfig(8) works with compat32 now.
XXX: FSSIOCSET50 and FSSIOCGET50 are not (yet) handled.

Separate the compat_netbsd32_coredump from the compat_netbsd32 and
coredump modules, into its own module.

Welcome to 7.99.75 !!!

branches: 1.133.2;
Add ACL support for FFS. From FreeBSD.

Introduce PT_LWPSTATUS + PT_LWPNEXT, obsolete PT_LWPINFO

PT_LWPINFO is a legacy ptrace(2) operation that was originally intended
to retrieve the thread (LWP) information inside a traced process.

It has a number of flaws and is confused with PT_LWPINFO from FreeBSD.

PT_LWPSTATUS and PT_LWPNEXT address the problems (shortly by: rename,
removal of pl_event) and introduces new features: signal context
(pl_sigpend, pl_sigmask), LWP name (pl_name), LWP TLS base address
(pl_private). The private pointer was so far missing information for
a debugger.

PT_LWPSTATUS@nnn is now shipped with core(5) files and contain LWP specific
information, so far missed in the core(5) files.

PT_LWPSTATUS retrieves LWP information for the prompted thread.
PT_LWPNEXT retrieves LWP information for the next thread, borrowing the
semantics from NetBSD specific PT_LWPINFO.

PT_LWPINFO is namespaced with __LEGACY_PT_LWPINFO and still available for
the foreseeable future, without plans of removing it.

Add ATF tests for PT_LWPSTATUS + PT_LWPNEXT.

Keep ATF tests for PT_LWPINFO.

Switch GDB to new API.

Proposed on tech-kern@.

8-byte objects on i386 or arm-oabi are aligned in 4-byte boundary.
Therefore, we must use __attribute__((__aligned__(4))) for them.
netbsd32_{,u}int64 are provided for this purpose. However, we
cannot use it in <compat/sys/siginfo.h> due to circular dependency
b/w <machine/netbsd32_machdep.h>.

In order to distangle it, we choose here to have a duplicate type,
netbsd32_siginfo_uint64, in <compat/sys/siginfo.h>. The equivalence
with netbsd32_uint64 is asserted in <compat/netbsd32/netbsd32.h>.

Now, gdb for i386 works again on amd64 kernel.

Based on patch provided by kamil. Thanks!

XXX
pullup to netbsd-9

Sorry, revert unintentional parts of the previous commit:
http://mail-index.netbsd.org/source-changes/2019/11/18/msg110946.html

I was going to commit only netbsd32_signal.c.

Belatedly catch up with kern_sig.c rev 1.358:
http://cvsweb.netbsd.org/bsdweb.cgi/src/sys/kern/kern_sig.c#rev1.358

Provide syscall information with SIGTRAP TRAP_SCE/TRAP_SCX so that
picotrace/truss, for example, works fine on COMPAT_NETBSD32.

With some minor changes:

- Centralize netbsd32_si{,32}_si{32,}() into netbsd32_ksi{,32}_ksi{32,}().
- Provide si_status with SIGCHLD.
- Remove the remaining of SA.

XXX
pullup to netbsd-9

For netbsd32_statvfs, f_spare should be netbsd32_uint64, not uint64_t.

Fix syscalls using struct statvfs on COMPAT_NETBSD32 on amd64, where
NETBSD32_INT64_ALIGN is __attribute__((__aligned__(4))).

Switch kevent udata from intptr_t void*

Synchromize the struct field format with other BSDs and Darwin.

No ABI change between older and newer struct form on the ports.

The change will require no changes to most C users during the transition
period as the header keeps a caller cast.

Discussed with core@ and there were no objections for this move.

accrightslen should be u_int for compat too

avoid signed-compare warning; use uintmax_t to print.

Add a new member to struct vfsstat and grow the unused members
The new member is caled f_mntfromlabel and it is the dkw_wname
of the corresponding wedge. This is now used by df -W to display
the mountpoint name as NAME=

branches: 1.123.4;
use KASSERTMSG() in netbsd32_ptr32i().

- expand the comment about what netbsd32_machdep.h should provide.
- add missing NETBSD32_POINTER_TYPE
- add static inline functions to convert pointers to/from 32 bit,
inclusing an overflow check for NETBSD32PTR32PLUS().

Merge the [pgoyette-compat] branch

Restore netbsd32 compat code for new NPF ABI.

support nullfs in netbsd32.

branches: 1.118.2;
add {send,recv}mmsg

redo the previous -- compat/sys/siginfo.h is already included
later, as is another header. move them up near the top.

branches: 1.116.2;
include <compat/sys/siginfo.h> for siginfo32_t.

handle siginfo requests for ptrace32

branches: 1.114.8;
tv_usec in netbsd32_timeval is suseconds_t (aka int32_t) rather than
netbsd32_long (considered harmless)

approved by martin

PR kern/51514: ptrace(2) fails for 32-bit process on 64-bit kernel

Updated from the original patch in the PR by me.

Trailing whitespace

Types for wait6 and clock_getcpuclockid2

deal with not having KTRACE

branches: 1.109.2;
We never exec(2) with a kernel vmspace, so do not test for that, but instead
KASSERT() that we don't.
When calculating the load address for the interpreter (e.g. ld.elf_so),
we need to take into account wether the exec'd process will run with
topdown memory or bottom up. We can not use the current vmspace's flags
to test for that, as this happens too early. Luckily the execpack already
knows what the new state will be later, so instead of testing the current
vmspace, pass the info as additional argument to struct emul
e_vm_default_addr.
Fix all such functions and adopt all callers.

move some header-specific pointer-typedefs into a header-specific area.

add netbsd32 support for nfssvc(2). we do this by defining 5 copyin/out
functions that do all the ugly work, are just plain copyin/out for the
native system calls, and do the necessary translations for netbsd32.

with this i'm able to run 32 bit nfsd and mountd on 64 bit kernel and
mount the file systems remotely.

s/sparc64/native 64 bit/ in a comment...

Implement trivial conversion for pset_*

Implement modctl, sigqueinfo and mq_*

branches: 1.103.2;
add tmpfs.

branches: 1.102.2;
Revert the following changes:

src/sys/sys/quotactl.h 1.37
src/sys/compat/netbsd32/netbsd32.h 1.101
src/sys/compat/netbsd32/netbsd32_netbsd.c 1.188, 1.189
src/sys/kern/vfs_quotactl.c 1.39
src/sys/kern/vfs_syscalls.c 1.483
src/sys/ufs/lfs/ulfs_quota.c 1.11
src/sys/ufs/ufs/ufs_quota.c 1.116
src/lib/libquota/quota_kernel.c 1.5

and do them correctly.

If you're going to change the name of something, you need to change
the name of *all* the things with the same name, not just a handful,
and you should change it to something similar so it still matches the
rest of the system rather than just picking an arbitrarily different
name.

Hi, Joerg.

To wit, rename the quotactl "delete" operation to "del", because
"delete" is a reserved word in C++ and for some reason Joerg wants to
run internal interfaces used only by C code through his C++ compiler.
Do not rename it to "remove" instead, because this doesn't match
libquota or the rest of the usage throughout the system; and rename
all the related identifiers, not just the ones that blew the mind of
Joerg's C++ compiler.

Because this is not a user-facing API (the only userland consumer
sys/quotactl.h is libquota) it is sort of ok to make arbitrary
source-incompatible changes; however, by the same token it's completely
unnecessary. If it *were* a user-facing API that someone might have a
semi-rational reason to want to run a C++ compiler on, it would be
incorrect to change it at this point.

delete -> remove

Do not use native off_t type under compat netbsd32, but a new
netbsd32_off_t that provide the expected alignment for 64bit types.

branches: 1.99.2;
Properly translate struct swapent for COMPAT_NETBSD32 (missing commit)

sigaction until 1.4 had an int sigmask, don't trash the stack.

Change the type of the 'cookie' that holds the state of the core dump file
from 'void *' to the actual type 'struct coredump_iostate *'.
In most of the code the contents of the structure are still unknown.
This just stops the wrong type of pointer being passed to the 'void *'
parameter.
I hope I've found everything, amd64 GENERIC and i386 GENERIC & ALL compile.

branches: 1.96.2;
fix pre-5 stat system call structure which should have used the pre-64-time_t
timestamps!

branches: 1.95.2;
Rework posix_spawn locking and memory management:
- always provide a vmspace for the new proc, initially borrowing from proc0
(this part fixes PR 46286)
- increase parallelism between parent and child if arguments allow this,
avoiding a potential deadlock on exec_lock
- add a new flag for userland to request old (lockstepped) behaviour for
better error reporting
- adapt test cases to the previous two and add a new variant to test the
diagnostics flag
- fix a few memory (and lock) leaks
- provide netbsd32 compat

allow mounting ext2fs and msdosfs
while there also enable lfs but that's untested

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

branches: 1.92.2;
Be consistent about whether idtype and objtype codes are signed or
unsigned. They are signed. (While unsigned might have been a better
choice, it doesn't really matter and the majority of preexisting uses
were signed. And consistency is good.)

Improve the names of some members of struct quotactl_args. These are
effectively function parameter names, but since they need to be
described with the same names in the man page the choices do matter.
Some.

Update compat_netbsd32 for new quotactl.

Add netbsd32_socklenp_t

branches: 1.88.2; 1.88.6;
As a precaution, add sys/param.h explicitly to some headers that will
no longer get it via sys/ucred.h and don't already include it
explicitly. These should in turn be removed when it can be confirmed
that it's safe to do so. (Because sys/param.h is full of things that
are often tested with #if, it's not in general safe to remove it
without checking, as such tests fail silently if the symbol goes
missing.)

dependant -> dependent

merge the bouyer-quota2 branch. This adds a new on-disk format
to store disk quota usage and limits, integrated with ffs
metadata. Usage is checked by fsck_ffs (no more quotacheck)
and is covered by the WAPBL journal. Enabled with kernel
option QUOTA2 (added where QUOTA was enabled in kernel config files),
turned on with tunefs(8) on a per-filesystem
basis. mount_mfs(8) can also turn quotas on.

See http://mail-index.netbsd.org/tech-kern/2011/02/19/msg010025.html
for details.

Add the ability to mount NFS filesystems in COMPAT_NETBSD32
If in the kernel and NFS_ARGS_ONLY, just export struct nfs_args and its flags.

branches: 1.84.4; 1.84.6; 1.84.8;
Merge from matt-nb5-mips64

Remove semicolon from NETBSD32TOx_UAP macro definitions.

Fix dev_t alignment, by not using the new 64bit value directly but
rather a 32bit aligned version (netbsd32_dev_t = netbsd32_uint64).

ok by cube@.

branches: 1.81.2;
Revert the previous packed attribute fix. Instead make netbsd32_time_t
of type netbsd32_int64 which provides a 64-bit value with a 32-bit
alignment.

Suggested by dsl.

Make netbsd32_timespec/netbsd32_timeval structures packed. This fix an
alignment problem between 64-bit kernel and 32-bit userland versions,
which ended in different structure sizes.

merge christos-time_t

branches: 1.78.4;
implement netbsd32__sched_setparam(), netbsd32__sched_getparam(),
netbsd32__sched_setaffinity() and netbsd32__sched_getaffinity().

branches: 1.77.2; 1.77.4;
Merge wrstuden-revivesa into HEAD.

branches: 1.76.4;
remove clause #3 from my license where there are no other
copyright holders involved.

Fix netbsd32_ipc_perm and netbsd32_shmid_ds structures to use correct
types.

branches: 1.74.6; 1.74.8; 1.74.10; 1.74.12;
Fix build

Use the netbsd32_[u]int64 type for all 64bit items to ensure we have the
correct alignment/padding rules.
Removed the now-unnecessary #pragma packed for amd64.
Should fix the alignment of 'struct netbsd32_kevent' on non-amd64.

Change u_intxx_t to uintxx_t and uint to unsigined int etc.
Fix indentation.

Convert many of the uses of __attribute__ to equivalent
__packed, __unused and __dead macros from cdefs.h

Convert all the system call entry points from:
int foo(struct lwp *l, void *v, register_t *retval)
to:
int foo(struct lwp *l, const struct foo_args *uap, register_t *retval)
Fixup compat code to not write into 'uap' and (in some cases) to actually
pass a correctly formatted 'uap' structure with the right name to the
next routine.
A few 'compat' routines that just call standard ones have been deleted.
All the 'compat' code compiles (along with the kernels required to test
build it).
98% done by automated scripts.

branches: 1.69.4;
Add: _lwp_setname, _lwp_getname, _lwp_ctl.

branches: 1.68.4; 1.68.6;
Use netbsd32_uint64 instead of fsblkcnt_t and fsfilcnt_t in order to get
the correct alignment (4 bytes for i386) for the whole structure.

branches: 1.67.2;
Define netbsd32_uint64 for 64bit integers with the alignment requirement
of the corresponding 32bit architecture.
Use it for the 64bit items in netbsd32_statvfs so that the structure
doesn't collect 8byte alignment (and 4 bytes of trailing padding).
This replaces the 'packed' attribute which wasn't architecture specific
and would cause massive overheads accessing every member of sparc64.
Should allow the MIPS64 port do DTRT.

branches: 1.66.6; 1.66.8;
include <compat/sys/mount.h> MFSNAMELEN

Changes to sompat socket function to avoid the dreaded stackgap.

Split netbsd32___semctl14() so that it is callable from
compat_10_netbsd32_sys_semsys() (where the one parameter is already in
kernel space).
Note that the code in compat_10_netbsd32_sys_semsys() has always been wrong,
since it called compat_14_sys___semctl() - which would read 64bit values!

Rework compat stat() and statvfs() code so that it no longer uses the stackgap.

Change all the NETBSD32PTR64(SCARG(uap, xxx))) to SCARG_P32(uap, xxx).

remove all the double (and triple) casts used to convert 32bit userspace
pointers to and from 64bit kernel pointers. Instead use the defines
NETBSD32PTR64(p32) to read a 32bit pointer and (the new) NETBSD32PTR32(p32,p64)
to write a 32bit pointer throughout.
The 32bit pointer is now a struct to enforce the above.
amd64 (with linux emul) and sparc64 will both compile (when the arch stuff
goes in soon), and amd64 still runs some i386 binaries.

branches: 1.60.2; 1.60.4; 1.60.6;
fix fallout from caddr_t changes.

Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

Introduce a new member to struct emul, e_startlwp, to be used by
sys__lwp_create. It allows using the said syscall under COMPAT_NETBSD32.

The libpthread regression tests now pass on amd64 and sparc64.

branches: 1.57.2;
Merge newlock2 to head.

adjust limits takes struct proc again.

From Nicolas Joly:

> It seems that 32bits programs, running under compat_netbsd32, using
> setrlimit force all other programs to have their maximum data size
> fixed at 3GB, where native 64bits apps used 8GB previously.

I tracked this one to the `netbsd32_adjust_limits()' function (called
when creating a new process under compat_netbsd32), where data and
stack limits are set without checking for shared `p_limit' structure
(p_limit->p_refcnt > 1). This explain the side effect where processes
have their limits changed when a compat_netbsd32 (or compat_linux32)
program is run.

The fix is to use `dosetrlimit()' to ensure the needed copy-on-write
behaviour for shared structure.

branches: 1.54.4;
Fix alignement problems for fhandle_t, exposed by gcc4.1.

While touching all vptofh/fhtovp functions, get rid of VFS_MAXFIDSIZ,
version the getfh(2) syscall and explicitly pass the size available in
the filehandle from userland.

Discussed on tech-kern, with lots of help from yamt (thanks!).

branches: 1.53.2;
support ntp_gettime again
compile tested by Havard Eidnes

branches: 1.52.2;
add an extra (uintptr_t) cast to NETBSD32TOP.

Implement the _lwp* and sa_* families of syscalls, using the newly
introduced hooks in the native code.

branches: 1.50.2; 1.50.4;
Implement the SysV IPC family of syscalls.

branches: 1.49.2;
Implement the ksem_* family of syscalls.

branches: 1.48.4; 1.48.6;
merge ktrace-lwp.

Move prototypes for siginfo32 to/from siginfo converters at the right
place.

- Split sys_kevent into kevent1 so that it can be used by COMPAT_NETBSD32
code.

- To achieve COMPAT_NETBSD32 compatibility, introduce a parameter to
kevent1 that points to functions that do the actual copyin/copyout
operations. This is similar to what was done in FreeBSD by Paul Saab.

- Add the COMPAT_NETBSD32 definitions and hooks.

branches: 1.45.2;
make the previous apply only to x86_64.

add __attribute__((packed)) to struct netbsd32_stat13. the alignment of
64-bit values is annoying in the x86 world: they require 64-bit alignment
in 64-bit mode but only 32-bit alignment in 32-bit mode.

this needs stuff from compat/sys/siginfo.h now, just include it here.

make this compile again.

make the siginfo converters static and comment out the one that is not used.

fix a compilation problem on NetBSD/amd64

Fix lossage I created with the 64 bit ino_t change.

64 bit inode changes.

Regen (__sigtimedwait(2)).

Implement the timer_create(2) family of syscalls.

netbsd32_execve2() doesn't exist anymore.

branches: 1.34.2;
Implement uuidgen(2).

Rework the coredump code to have no explicit knownledge of how coredump
i/o is done. Instead, pass an opaque cookie which is then passed to a
new routine, coredump_write, which does the actual i/o. This allows the
method of doing i/o to change without affecting any future MD code.
Also, make netbsd32_core.c [re]use core_netbsd.c (in a similar manner that
core_elf64.c uses core_elf32.c) and eliminate that code duplication.
cpu_coredump{,32} is now called twice, first with a NULL iocookie to fill
the core structure and a second to actually write md parts of the coredump.
All i/o is nolonger random access and is suitable for shipping over a stream.

in siginfo conversion functions, declare the source operand as "const"

Fix some things regarding COMPAT_NETBSD32 and limits/VM addresses.

* For sparc64 and amd64, define *SIZ32 VM constants.
* Add a new function pointer to struct emul, pointing at a function
that will return the default VM map address. The default function
is uvm_map_defaultaddr, which just uses the VM_DEFAULT_ADDRESS
macro. This gives emulations control over the default map address,
and allows things to be mapped at the right address (in 32bit range)
for COMPAT_NETBSD32.
* Add code to adjust the data and stack limits when a COMPAT_NETBSD32
or COMPAT_SVR4_32 binary is executed.
* Don't use USRSTACK in kern_resource.c, use p_vmspace->vm_minsaddr
instead (emulations might have set it differently)
* Since this changes struct emul, bump kernel version to 3.99.2

Tested on amd64, compile-tested on sparc64.

branches: 1.30.2;
nuke trailing whitespace

branches: 1.29.4; 1.29.6;
o Add support for the statvfs family of syscalls (statvfs1, fstatvfs1,
fhstatvfs1 and getvfsstat)
o Move the statfs family out of netbsd32_fs.c and netbsd32_netbsd.c to
netbsd_compat_20.c, compiled with COMPAT_20

Reviewed by christos@.

Tweak sysctl setup functions (the macros, actually) for use in lkms,
and tweak lkminit_*.c (where applicable) to call them, and to call
sysctl_teardown() when being unloaded.

This consists of (1) making setup functions not be static when being
compiled as lkms (change to sys/sysctl.h), (2) making prototypes
visible for the various setup functions in header files (changes to
various header files), and (3) making simple "load" and "unload"
functions in the actual lkminit stuff.

linux_sysctl.c also needs its root exposed (ie, made not static) for
this (when built as an lkm).

branches: 1.27.2;
delete ktrsyscall32() prototype

add a copy of the ktrsyscall() entry point for 32 bit emulated calls.
the main purpose of this function is to adjust the "argsize" value of
the ktrace syscall record, otherwise userland will see N/2 (rounded
down) arguments instead of N.

Implement 32bit get/setcontext entry points.

branches: 1.24.2;
Merge the nathanw_sa branch.

Complete the rename: s/sa_/<compat>_sa_/g

Also tweak the darwin siginfo stuff to avoid the same type of collision.

In preparation for COMPAT_NETBSD32 on SH-5:

- The MD netbsd32_machdep.h header now defines the 32-bit pointer type
instead of using u_int32_t everywhere,
- The MD netbsd32_machdep.h header now defines a macro (at least on
current implementations) which converts a 32-bit pointer to its 64-bit
equivalent,
- Change the MI code to utilise the above two items in all the right places,
- Implement netbsd32___sigaction_sigtramp().

Tested on Sparc64 by Matt Green.

Make the coredump routine exec-format/emulation specific. Split
out traditional NetBSD coredump routines into core_netbsd.c and
netbsd32_core.c (for COMPAT_NETBSD32).

update copyright notices.

branches: 1.19.2;
Some of this was sparc-specific, so ifdef __sparc__ it (XXX). Also,
change the alignment of one structure with an MD ifdef. Should
be moved into the netbsd32_machdep parts.

proto for coredump32

branches: 1.17.2;
make sysctl vm.loadavg work.

de-static netbsd32_exec_aout_prep_[zno]magic().
de-static netbsd32_from_stat43().
move the guts of netbsd32_execve() into netbsd32_execve2().

all of are for the forthcoming sunos32 compat mode (for sparc64).

Add systm.h to pull in copyout* prototypes.

add e_path (emulation path) to struct emul, which replaces emulation-specific
*_emul_path variables

change macros CHECK_ALT_{CREAT|EXIST} to use that, 'root' doesn't need
to be passed explicitly any more and *_CHECK_ALT_{CREAT|EXIST} are removed
change explicit emul_find() calls in probe functions to get the emulation
path from the checked exec switch entry's emulation

remove no longer needed header files

add e_flags and e_syscall to struct emul; these are unsed and empty for now

kill gcc warnings.

declare struct firm_event32

- massive warning fix fest.
- fix pread/pwrite return values (plus some other syscalls that looked
similarly broken).
- prototypes and clean up for netbsd32_ioctl.c

now getpw*() works under compat32!

branches: 1.10.4;
Update to 32-bit compatibility routines.

branches: 1.9.2;
Add netbsd32_compat_14.c

Rename everything from compat_netbsd32 -> netbsd32 so we don't have silly
names like compat_14_compat_netbsd32_sys_shmctl().

branches: 1.8.2;
XSH5: change function signature to `void *sbrk(intptr_t)'.

branches: 1.7.4;
clean and up make compile

SPARC32->NETBSD32

include the MD header.

move sparc32 to netbsd32; split out MD part (signal handling) (mostly via repository copy, rename and perl -pi :-)

Overhaul. Now will run simple things like /bin/sh and /bin/echo.

Need to make sure high bits of addresses are zeroed out.

add a 32-bit compatibility module for the sparc64 port, so it can run NetBSD/sparc binaries with a LP64 kernel.

All mips64 has compat_netbsd32

branches: 1.2.8;
According to mrg@ we need to include aarch64 in the list of those
that need to support COMPAT_NETBSD32

branches: 1.1.4;
don't assume _LP64 == supports COMPAT_NETBSD32.

this is not true for alpha, ia64 and arm32 ports, and the first two
were not building because of it, and the latter would be missing
the oabi support (likely not a big deal, but still wrong.)

add a makefile fragment that tells you if it is supported and include
it where needed to define COMPAT_NETBSD32 when building the normal
kernel (ie, modules & rump.)


fixes alpha build, probably fixes ia64 build.


XXX: still leaves some netbsd32 code in rf_netbsdkintf.c, that should
be moved into some hooks, but first the configuration setup
needs to be moved into a common function the netbsd32 code can
call into, vs living in the switch case itself.

Add ACL support for FFS. From FreeBSD.

KNF consistency: No parentheses are needed around the return value.

branches: 1.19.12;
Merge the [pgoyette-compat] branch

branches: 1.18.12; 1.18.86; 1.18.88;
remove clause #3 from my license where there are no other
copyright holders involved.

branches: 1.17.6; 1.17.8; 1.17.10; 1.17.12;
Convert all the system call entry points from:
int foo(struct lwp *l, void *v, register_t *retval)
to:
int foo(struct lwp *l, const struct foo_args *uap, register_t *retval)
Fixup compat code to not write into 'uap' and (in some cases) to actually
pass a correctly formatted 'uap' structure with the right name to the
next routine.
A few 'compat' routines that just call standard ones have been deleted.
All the 'compat' code compiles (along with the kernels required to test
build it).
98% done by automated scripts.

branches: 1.16.4;
ANSIfy most of the function definitions in sys/compat (but not ndis).
All by the magic of sed ...

branches: 1.15.8; 1.15.10; 1.15.18;
Change all the NETBSD32PTR64(SCARG(uap, xxx))) to SCARG_P32(uap, xxx).

branches: 1.14.2; 1.14.6; 1.14.8; 1.14.10;
Merge newlock2 to head.

branches: 1.13.20;
merge ktrace-lwp.

branches: 1.12.16;
Dynamic sysctl.

Gone are the old kern_sysctl(), cpu_sysctl(), hw_sysctl(),
vfs_sysctl(), etc, routines, along with sysctl_int() et al. Now all
nodes are registered with the tree, and nodes can be added (or
removed) easily, and I/O to and from the tree is handled generically.

Since the nodes are registered with the tree, the mapping from name to
number (and back again) can now be discovered, instead of having to be
hard coded. Adding new nodes to the tree is likewise much simpler --
the new infrastructure handles almost all the work for simple types,
and just about anything else can be done with a small helper function.

All existing nodes are where they were before (numerically speaking),
so all existing consumers of sysctl information should notice no
difference.

PS - I'm sorry, but there's a distinct lack of documentation at the
moment. I'm working on sysctl(3/8/9) right now, and I promise to
watch out for buses.

branches: 1.11.2;
Back out the lwp/ktrace changes. They contained a lot of colateral damage,
and need to be examined and discussed more.

struct proc * -> struct lwp *

Merge the nathanw_sa branch.

In preparation for COMPAT_NETBSD32 on SH-5:

- The MD netbsd32_machdep.h header now defines the 32-bit pointer type
instead of using u_int32_t everywhere,
- The MD netbsd32_machdep.h header now defines a macro (at least on
current implementations) which converts a 32-bit pointer to its 64-bit
equivalent,
- Change the MI code to utilise the above two items in all the right places,
- Implement netbsd32___sigaction_sigtramp().

Tested on Sparc64 by Matt Green.

add RCSIDs (including regeneration of files as appropriate)

branches: 1.6.2; 1.6.4;
fix typo - it should be compat_09_netbsd32_uname, not compat_09_netbsd32_ouname

remove include of <vm/vm.h>. <vm/vm.h> -> <uvm/uvm_extern.h>

branches: 1.4.2;
Add netbsd32_compat_14.c

Rename everything from compat_netbsd32 -> netbsd32 so we don't have silly
names like compat_14_compat_netbsd32_sys_shmctl().

branches: 1.3.8;
SPARC32->NETBSD32

move sparc32 to netbsd32; split out MD part (signal handling) (mostly via repository copy, rename and perl -pi :-)

add a 32-bit compatibility module for the sparc64 port, so it can run NetBSD/sparc binaries with a LP64 kernel.

Merge the [pgoyette-compat] branch

branches: 1.25.16; 1.25.18;
Split out the SYSV IPC stuff into its own compat_netbsd32_sysvipc
module. Adjust dependencies as needed.

branches: 1.24.18; 1.24.36;
Replace M_IOV and some malloc(9)s with kmem(9), and while there:
- Fix invalid free (M_TEMP vs M_IOV) in do_sys_recvmsg(), spotted by jakllsch@.
Also, same fix in osf1_sys_sendmsg_xopen().
- Fix attempt to free non-allocated memory in error path in netbsd32___getfh30().
- Plug a memory leak in compat_43_netbsd32_orecvmsg().

branches: 1.23.4; 1.23.6;
merge christos-time_t

branches: 1.22.6; 1.22.10; 1.22.12; 1.22.18; 1.22.24;
Convert all the system call entry points from:
int foo(struct lwp *l, void *v, register_t *retval)
to:
int foo(struct lwp *l, const struct foo_args *uap, register_t *retval)
Fixup compat code to not write into 'uap' and (in some cases) to actually
pass a correctly formatted 'uap' structure with the right name to the
next routine.
A few 'compat' routines that just call standard ones have been deleted.
All the 'compat' code compiles (along with the kernels required to test
build it).
98% done by automated scripts.

branches: 1.21.4;
ANSIfy most of the function definitions in sys/compat (but not ndis).
All by the magic of sed ...

branches: 1.20.6; 1.20.8; 1.20.16;
Split netbsd32___semctl14() so that it is callable from
compat_10_netbsd32_sys_semsys() (where the one parameter is already in
kernel space).
Note that the code in compat_10_netbsd32_sys_semsys() has always been wrong,
since it called compat_14_sys___semctl() - which would read 64bit values!

remove all the double (and triple) casts used to convert 32bit userspace
pointers to and from 64bit kernel pointers. Instead use the defines
NETBSD32PTR64(p32) to read a 32bit pointer and (the new) NETBSD32PTR32(p32,p64)
to write a 32bit pointer throughout.
The 32bit pointer is now a struct to enforce the above.
amd64 (with linux emul) and sparc64 will both compile (when the arch stuff
goes in soon), and amd64 still runs some i386 binaries.

branches: 1.18.2; 1.18.4; 1.18.6;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.17.2;
Merge newlock2 to head.

branches: 1.16.22;
need compat/sys/shm.h

branches: 1.15.2; 1.15.18;
Merge the nathanw_sa branch.

In preparation for COMPAT_NETBSD32 on SH-5:

- The MD netbsd32_machdep.h header now defines the 32-bit pointer type
instead of using u_int32_t everywhere,
- The MD netbsd32_machdep.h header now defines a macro (at least on
current implementations) which converts a 32-bit pointer to its 64-bit
equivalent,
- Change the MI code to utilise the above two items in all the right places,
- Implement netbsd32___sigaction_sigtramp().

Tested on Sparc64 by Matt Green.

build the netbsd32_compat* objects into libcompat.a, so they are available
to emulations not just netbsd itself. rename the compat_10 functions to
match everything else. fixes a problem reported by julian coleman.

make the stackgap_{init,alloc} functions MP friendly (i.e. pass struct proc *
in, instead of using curproc). While there add an optional size argument to
stackgap_init.

add RCSIDs (including regeneration of files as appropriate)

branches: 1.10.2;
use _KERNEL_OPT.

branches: 1.9.2;
LKMify

get rid of remaining warnings.

kill gcc warnings.

remove include of <vm/vm.h>. <vm/vm.h> -> <uvm/uvm_extern.h>

remove/move more mach vm header files:

<vm/pglist.h> -> <uvm/uvm_pglist.h>
<vm/vm_inherit.h> -> <uvm/uvm_inherit.h>
<vm/vm_kern.h> -> into <uvm/uvm_extern.h>
<vm/vm_object.h> -> nothing
<vm/vm_pager.h> -> into <uvm/uvm_pager.h>

also includes a bunch of <vm/vm_page.h> include removals (due to redudancy
with <vm/vm.h>), and a scattering of other similar headers.

branches: 1.4.2;
Add netbsd32_compat_14.c

Rename everything from compat_netbsd32 -> netbsd32 so we don't have silly
names like compat_14_compat_netbsd32_sys_shmctl().

branches: 1.3.8;
move sparc32 to netbsd32; split out MD part (signal handling) (mostly via repository copy, rename and perl -pi :-)

Defopt SYSVMSG, SYSVSEM and SYSVSHM.

Overhaul. Now will run simple things like /bin/sh and /bin/echo.

Rename dup3110 to dup3100 to match historical practice for NetBSD-11,
because we already have kevent100. Fix compat_dup3 to belong in compat_100
not compat_110.

COMPAT_NETBSD32: No need to have compat_100_netbsd32_kevent_fetch_timeout.
NFCI.

COMPAT_NETBSD32: Catch up with sys___kevent100() addition.

XXX
Add epoll_* and memfd_create.

Create compat_100 module infrastructure now that we have branched
for NetBSD-10

KNF consistency: No parentheses are needed around the return value.

branches: 1.2.4; 1.2.14;
Merge the [pgoyette-compat] branch

branches: 1.1.2;
file netbsd32_compat_10_sysv.c was initially added on branch pgoyette-compat.

Rename dup3110 to dup3100 to match historical practice for NetBSD-11,
because we already have kevent100. Fix compat_dup3 to belong in compat_100
not compat_110.

version dup3

sys/compat: Memset zero before copyout.

Just in case of uninitialized padding which would lead to kernel
stack disclosure. If the compiler can prove the memset redundant
then it can optimize it away; otherwise better safe than sorry.

KNF consistency: No parentheses are needed around the return value.

branches: 1.34.4; 1.34.12;
Merge the [pgoyette-compat] branch

branches: 1.33.30; 1.33.36; 1.33.38;
fix these *stat routines: don't pass a kernel stack buffer
to a function that will try to copyout() to it.
just do both layers of compat translation here.

branches: 1.32.14; 1.32.24;
Cleanup. Kill a few netbsd32_caddr_t in syscalls argument types.

branches: 1.31.6; 1.31.12;
remove clause #3 from my license where there are no other
copyright holders involved.

branches: 1.30.6; 1.30.8; 1.30.10; 1.30.12;
Convert all the system call entry points from:
int foo(struct lwp *l, void *v, register_t *retval)
to:
int foo(struct lwp *l, const struct foo_args *uap, register_t *retval)
Fixup compat code to not write into 'uap' and (in some cases) to actually
pass a correctly formatted 'uap' structure with the right name to the
next routine.
A few 'compat' routines that just call standard ones have been deleted.
All the 'compat' code compiles (along with the kernels required to test
build it).
98% done by automated scripts.

branches: 1.29.4;
ANSIfy most of the function definitions in sys/compat (but not ndis).
All by the magic of sed ...

Remove all the __P

branches: 1.27.6; 1.27.8; 1.27.14; 1.27.16;
Remove some unused calls to stackgap_init()

Change the way that emulations locate files within the emulation root to
avoid having to allocate space in the 'stackgap'
- which is very LWP unfriendly.
The additional code for non-emulation namei() is trivial, the reduction for
the emulations is massive.
The vnode for a processes emulation root is saved in the cwdi structure
during process exec.
If the emulation root the TRYEMULROOT flag are set, namei() will do an initial
search for absolute pathnames in the emulation root, if that fails it will
retry from the normal root.
".." at the emulation root will always go to the real root, even in the middle
of paths and when expanding symlinks.
Absolute symlinks found using absolute paths in the emulation root will be
relative to the emulation root (so /usr/lib/xxx.so -> /lib/xxx.so links
inside the emulation root don't need changing).
If the root of the emulation would be returned (for an emulation lookup), then
the real root is returned instead (matching the behaviour of emul_lookup,
but being a cheap comparison here) so that programs that scan "../.."
looking for the root dircetory don't loop forever.
The target for symbolic links is no longer mangled (it used to get the
CHECK_ALT_xxx() treatment, so could get /emul/xxx prepended).
CHECK_ALT_xxx() are no more. Most of the change is deleting them, and adding
TRYEMULROOT to the flags to NDINIT().
A lot of the emulation system call stubs could now be deleted.

Change all the NETBSD32PTR64(SCARG(uap, xxx))) to SCARG_P32(uap, xxx).

remove all the double (and triple) casts used to convert 32bit userspace
pointers to and from 64bit kernel pointers. Instead use the defines
NETBSD32PTR64(p32) to read a 32bit pointer and (the new) NETBSD32PTR32(p32,p64)
to write a 32bit pointer throughout.
The 32bit pointer is now a struct to enforce the above.
amd64 (with linux emul) and sparc64 will both compile (when the arch stuff
goes in soon), and amd64 still runs some i386 binaries.

branches: 1.23.2; 1.23.4; 1.23.6;
fix fallout from caddr_t changes.

Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.21.2;
Merge newlock2 to head.

branches: 1.20.20;
merge ktrace-lwp.

make this compile again.

branches: 1.18.2; 1.18.18;
Back out the lwp/ktrace changes. They contained a lot of colateral damage,
and need to be examined and discussed more.

Pass lwp pointers throughtout the kernel, as required, so that the lwpid can
be inserted into ktrace records. The general change has been to replace
"struct proc *" with "struct lwp *" in various function prototypes, pass
the lwp through and use l_proc to get the process pointer when needed.

Bump the kernel rev up to 1.6V

Merge the nathanw_sa branch.

In preparation for COMPAT_NETBSD32 on SH-5:

- The MD netbsd32_machdep.h header now defines the 32-bit pointer type
instead of using u_int32_t everywhere,
- The MD netbsd32_machdep.h header now defines a macro (at least on
current implementations) which converts a 32-bit pointer to its 64-bit
equivalent,
- Change the MI code to utilise the above two items in all the right places,
- Implement netbsd32___sigaction_sigtramp().

Tested on Sparc64 by Matt Green.

make the stackgap_{init,alloc} functions MP friendly (i.e. pass struct proc *
in, instead of using curproc). While there add an optional size argument to
stackgap_init.

add RCSIDs (including regeneration of files as appropriate)

update copyright notices.

branches: 1.11.2;
use _KERNEL_OPT.

branches: 1.10.2;
compat_12_netbsd32_{,fl}stat12() were not copyout()ing the converted result.

Include opt_* files to make sure prototypes are included.

prototype netbsd32_stat12_to_netbsd32()

add e_path (emulation path) to struct emul, which replaces emulation-specific
*_emul_path variables

change macros CHECK_ALT_{CREAT|EXIST} to use that, 'root' doesn't need
to be passed explicitly any more and *_CHECK_ALT_{CREAT|EXIST} are removed
change explicit emul_find() calls in probe functions to get the emulation
path from the checked exec switch entry's emulation

remove no longer needed header files

add e_flags and e_syscall to struct emul; these are unsed and empty for now

branches: 1.6.2;
Add netbsd32_compat_14.c

Rename everything from compat_netbsd32 -> netbsd32 so we don't have silly
names like compat_14_compat_netbsd32_sys_shmctl().

branches: 1.5.8;
SPARC32->NETBSD32

move sparc32 to netbsd32; split out MD part (signal handling) (mostly via repository copy, rename and perl -pi :-)

Overhaul. Now will run simple things like /bin/sh and /bin/echo.

move <vm/vm_swap.h> to <sys/swap.h>. <vm/vm_swap.h> still works for now (goes away later)

add a 32-bit compatibility module for the sparc64 port, so it can run NetBSD/sparc binaries with a LP64 kernel.

KNF consistency: No parentheses are needed around the return value.

branches: 1.27.12;
Merge the [pgoyette-compat] branch

branches: 1.26.28; 1.26.30;
need to lock the process.

branches: 1.25.12; 1.25.34; 1.25.44; 1.25.50;
remove clause #3 from my license where there are no other
copyright holders involved.

branches: 1.24.6; 1.24.8; 1.24.10; 1.24.12;
Convert all the system call entry points from:
int foo(struct lwp *l, void *v, register_t *retval)
to:
int foo(struct lwp *l, const struct foo_args *uap, register_t *retval)
Fixup compat code to not write into 'uap' and (in some cases) to actually
pass a correctly formatted 'uap' structure with the right name to the
next routine.
A few 'compat' routines that just call standard ones have been deleted.
All the 'compat' code compiles (along with the kernels required to test
build it).
98% done by automated scripts.

branches: 1.23.4;
ANSIfy most of the function definitions in sys/compat (but not ndis).
All by the magic of sed ...

branches: 1.22.6; 1.22.8; 1.22.16;
Modify the (recently added) compat_sigaltstack() #define to take the
values of the SS_ONSTACK and SS_DISABLE constants.
Use it to shorten the source files when this action is replicated.
Actually, given the monstrous complexity of sigaltstack1() there is
probably a much better way to do this...

Add a #define that expands to the compat wrapper for sigaltstack1().
Use it for the netbsd32_compat_13.c version of the code.

Calling sigaltstack1() directly is so much easier than going via
compat_13_sys_sigaltstack().

Change all the NETBSD32PTR64(SCARG(uap, xxx))) to SCARG_P32(uap, xxx).

remove all the double (and triple) casts used to convert 32bit userspace
pointers to and from 64bit kernel pointers. Instead use the defines
NETBSD32PTR64(p32) to read a 32bit pointer and (the new) NETBSD32PTR32(p32,p64)
to write a 32bit pointer throughout.
The 32bit pointer is now a struct to enforce the above.
amd64 (with linux emul) and sparc64 will both compile (when the arch stuff
goes in soon), and amd64 still runs some i386 binaries.

branches: 1.17.2; 1.17.4; 1.17.6;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.16.2;
Merge newlock2 to head.

branches: 1.15.20;
merge ktrace-lwp.

make this compile again.

branches: 1.13.2; 1.13.18;
Merge the nathanw_sa branch.

In preparation for COMPAT_NETBSD32 on SH-5:

- The MD netbsd32_machdep.h header now defines the 32-bit pointer type
instead of using u_int32_t everywhere,
- The MD netbsd32_machdep.h header now defines a macro (at least on
current implementations) which converts a 32-bit pointer to its 64-bit
equivalent,
- Change the MI code to utilise the above two items in all the right places,
- Implement netbsd32___sigaction_sigtramp().

Tested on Sparc64 by Matt Green.

make the stackgap_{init,alloc} functions MP friendly (i.e. pass struct proc *
in, instead of using curproc). While there add an optional size argument to
stackgap_init.

add RCSIDs (including regeneration of files as appropriate)

update copyright notices.

branches: 1.8.2; 1.8.4;
fix the glaring errors in compat_13_netbsd32_sigaltstack13

- massive warning fix fest.
- fix pread/pwrite return values (plus some other syscalls that looked
similarly broken).
- prototypes and clean up for netbsd32_ioctl.c

now getpw*() works under compat32!

Kill register declarations.

Update to 32-bit compatibility routines.

branches: 1.4.2;
Add netbsd32_compat_14.c

Rename everything from compat_netbsd32 -> netbsd32 so we don't have silly
names like compat_14_compat_netbsd32_sys_shmctl().

branches: 1.3.8;
move sparc32 to netbsd32; split out MD part (signal handling) (mostly via repository copy, rename and perl -pi :-)

use right syscalls

add a 32-bit compatibility module for the sparc64 port, so it can run NetBSD/sparc binaries with a LP64 kernel.

this file was moved to netbsd32_compat_14_sysv.c.

fix naming errors in previous. (this file is no longer compiled, but
this fix makes the pull up more obvious, before deleting this file.)

for sysv ipc stat operations, explicitly copy the exported parts
instead of the whole ds structure.

besides triggering a recently added assert in netbsd32, this stops
exposing kernel addresses.

copy the mode clamping to 0777 from sem to shm and msg.


while here, make sure that the compat callers to sysv_ipc clear
the contents of the compat structure before setting the result
members to ensure padding bytes are cleared.

don't set/copy _sem_base, _msg_first, _msg_last or _shm_internal.
even if used, which seems very dodgy, they leak KVAs as well.
possibly this may affect linux binaries, in particular, the
comments around _shm_internal ("XXX Oh well.") may mean apps
rely upon these but hopefully not -- the comments date back to
rev 1.1 in 1995.

the _key, _seq and _msg_cbytes members are exported as before as
i found multiple consumers of these (no less than ipcs(1), and
they appear to be useful for debugging and more.


XXX: the naming of compat functions have too many styles. there
are at least 3 different ones changed here.

Fix kernel info leaks.

+ Possible info leak: [len=80, leaked=10]
| #0 0xffffffff80bad7a7 in kleak_copyout
| #1 0xffffffff8048e71b in netbsd32___msgctl50
| #2 0xffffffff8022fb5b in netbsd32_syscall
| #3 0xffffffff802096dd in handle_syscall

branches: 1.26.10; 1.26.16; 1.26.18;
Avoid defining inline functions that are never used when certain options
are missing.

Strange patch(1) behavior got me three copies of the code. One is
sufficient so remove the excess.

If compiling as part of a built-in kernel module, make sure we include
the relevant opt_sysv.h file

Missed these.

Only include the compat_14 sys calls if this being compiled as part of
the compat_netbsd32_sysvipc module.

Split out the SYSV IPC stuff into its own compat_netbsd32_sysvipc
module. Adjust dependencies as needed.

branches: 1.21.12; 1.21.56; 1.21.74; 1.21.76; 1.21.78; 1.21.82;
Convert all the system call entry points from:
int foo(struct lwp *l, void *v, register_t *retval)
to:
int foo(struct lwp *l, const struct foo_args *uap, register_t *retval)
Fixup compat code to not write into 'uap' and (in some cases) to actually
pass a correctly formatted 'uap' structure with the right name to the
next routine.
A few 'compat' routines that just call standard ones have been deleted.
All the 'compat' code compiles (along with the kernels required to test
build it).
98% done by automated scripts.

branches: 1.20.4;
ANSIfy most of the function definitions in sys/compat (but not ndis).
All by the magic of sed ...

branches: 1.19.8; 1.19.10; 1.19.18;
More SCARG_P32 fallout.

Fix typo. Reported by martin.

Change all the NETBSD32PTR64(SCARG(uap, xxx))) to SCARG_P32(uap, xxx).

remove all the double (and triple) casts used to convert 32bit userspace
pointers to and from 64bit kernel pointers. Instead use the defines
NETBSD32PTR64(p32) to read a 32bit pointer and (the new) NETBSD32PTR32(p32,p64)
to write a 32bit pointer throughout.
The 32bit pointer is now a struct to enforce the above.
amd64 (with linux emul) and sparc64 will both compile (when the arch stuff
goes in soon), and amd64 still runs some i386 binaries.

branches: 1.15.2; 1.15.4; 1.15.6;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.14.2;
Merge newlock2 to head.

branches: 1.13.4;
Use the LWP cached credentials where sane.

branches: 1.12.4; 1.12.8;
Remove leading __ from __(const|inline|signed|volatile) -- it is obsolete.

need compat/sys/shm.h

branches: 1.10.4;
nuke trailing whitespace

branches: 1.9.2; 1.9.10; 1.9.12;
Merge the nathanw_sa branch.

In preparation for COMPAT_NETBSD32 on SH-5:

- The MD netbsd32_machdep.h header now defines the 32-bit pointer type
instead of using u_int32_t everywhere,
- The MD netbsd32_machdep.h header now defines a macro (at least on
current implementations) which converts a 32-bit pointer to its 64-bit
equivalent,
- Change the MI code to utilise the above two items in all the right places,
- Implement netbsd32___sigaction_sigtramp().

Tested on Sparc64 by Matt Green.

don't need <sys/types.h> when including <sys/param.h>

add RCSIDs (including regeneration of files as appropriate)

branches: 1.5.2; 1.5.4;
only define SYSVMSG, SYSVSEM, SYSVSHM if it's not defined already

kill gcc warnings.

- massive warning fix fest.
- fix pread/pwrite return values (plus some other syscalls that looked
similarly broken).
- prototypes and clean up for netbsd32_ioctl.c

now getpw*() works under compat32!

Update to 32-bit compatibility routines.

branches: 1.1.2; 1.1.8;
Add netbsd32_compat_14.c

Rename everything from compat_netbsd32 -> netbsd32 so we don't have silly
names like compat_14_compat_netbsd32_sys_shmctl().

KNF consistency: No parentheses are needed around the return value.

branches: 1.3.4; 1.3.14;
for sysv ipc stat operations, explicitly copy the exported parts
instead of the whole ds structure.

besides triggering a recently added assert in netbsd32, this stops
exposing kernel addresses.

copy the mode clamping to 0777 from sem to shm and msg.


while here, make sure that the compat callers to sysv_ipc clear
the contents of the compat structure before setting the result
members to ensure padding bytes are cleared.

don't set/copy _sem_base, _msg_first, _msg_last or _shm_internal.
even if used, which seems very dodgy, they leak KVAs as well.
possibly this may affect linux binaries, in particular, the
comments around _shm_internal ("XXX Oh well.") may mean apps
rely upon these but hopefully not -- the comments date back to
rev 1.1 in 1995.

the _key, _seq and _msg_cbytes members are exported as before as
i found multiple consumers of these (no less than ipcs(1), and
they appear to be useful for debugging and more.


XXX: the naming of compat functions have too many styles. there
are at least 3 different ones changed here.

Merge the [pgoyette-compat] branch

branches: 1.1.2;
file netbsd32_compat_14_sysv.c was initially added on branch pgoyette-compat.

whitespace

fix previous (wrong ifdef), thanks wiz@

PR/59100: Onno van der Linden: i386 compat32 broken on amd64

Crude patch to allow building and loading of compat_16 and compat_netbsd32_16 code
on aarch64.

netbsd32_compat_16.c: Fix whitespace.

No functional change intended.

branches: 1.5.2;
Enable compat sigreturn system call.

branches: 1.4.4;
Fix anonymous memory object leak for sigcode.

- Repeating "modload compat_linux && /emul/linux/bin/ls && modunload compat_linux"
will reproduce this problem.
- It cause in exec_sigcode_map(), anon-object for sigcode was created at
first exec, but it remained even after exec_remove.
- Fixed that the anon-object for sigcode is created at exec_add(), and the
anon-object reference is removed at exec_remove().
- sigobject_lock is no longer needed since it is locked by exec_lock.
- The compat_16 module rewrites the e_sigcode entry in emul_netbsd directly and
does not use exec_add()/exec_remove(), so it needs to call
sigcode_alloc()/sigcode_free() on its own.

Remove clause 3 and 4 leftovers from TNF licenses in more sources.

Confirmed by martin@ in PR/54760.

branches: 1.2.4; 1.2.6;
Merge the [pgoyette-compat] branch

branches: 1.1.2;
file netbsd32_compat_16.c was initially added on branch pgoyette-compat.

sys/compat: Memset zero before copyout.

Just in case of uninitialized padding which would lead to kernel
stack disclosure. If the compiler can prove the memset redundant
then it can optimize it away; otherwise better safe than sorry.

Fix syntax error

KNF consistency: No parentheses are needed around the return value.

branches: 1.39.8;
Fix three stack info leaks, found by kMSan when just invoking all syscalls
with a zero page as argument.

MSan: Uninitialized Stack Memory In copyout() At Offset 0, Variable 'sb32' From compat_20_netbsd32_getfsstat()
MSan: Uninitialized Stack Memory In copyout() At Offset 12, Variable 'oss' From compat_43_sys_sigstack()
MSan: Uninitialized Stack Memory In copyout() At Offset 0, Variable 'sb' From compat_50_netbsd32___fhstat40()

branches: 1.38.4;
Merge the [pgoyette-compat] branch

branches: 1.37.2;
add {send,recv}mmsg

branches: 1.36.4; 1.36.10;
Switch compat_20_netbsd32_getfsstat() to mountlist iterator.

never pass mp->mnt_stat to dostatvfs(), always use a separate buffer.
dostatvfs() clears its buffer, but mp->mnt_stat should never be cleared.

branches: 1.34.2; 1.34.4; 1.34.6;
Try not to use f_data, use f_{vnode,socket,pipe,mqueue,kqueue,ksem} to get
a correctly typed pointer.

branches: 1.33.4; 1.33.10;
remove useless flag mangling. It is done already.

Fix locking botch, and ...
Yes, we want to be careful because the sizes are different.

Change the queue.3 *_END(&head) macros to NULL. Since we don't have CIRCLEQ
anymore, all the macros expand to NULL anyway, so this improves readability.
Requested by rmind@

fix open-coded mountlist scan

Since mountlist is now a TAILQ, convert some missed usages
so things build again.

branches: 1.28.8; 1.28.18; 1.28.22;
Replace M_IOV and some malloc(9)s with kmem(9), and while there:
- Fix invalid free (M_TEMP vs M_IOV) in do_sys_recvmsg(), spotted by jakllsch@.
Also, same fix in osf1_sys_sendmsg_xopen().
- Fix attempt to free non-allocated memory in error path in netbsd32___getfh30().
- Plug a memory leak in compat_43_netbsd32_orecvmsg().

branches: 1.27.2; 1.27.4;
Convert 67 namei call sites to use namei_simple, in these functions:

check_console, veriexecclose, veriexec_delete, veriexec_file_add,
emul_find_root, coff_load_shlib (sh3 version), coff_load_shlib,
compat_20_sys_statfs, compat_20_netbsd32_statfs,
ELFNAME2(netbsd32,probe_noteless), darwin_sys_statfs,
ibcs2_sys_statfs, ibcs2_sys_statvfs, linux_sys_uselib,
osf1_sys_statfs, sunos_sys_statfs, sunos32_sys_statfs,
ultrix_sys_statfs, do_sys_mount, fss_create_files (3 of 4),
adosfs_mount, cd9660_mount, coda_ioctl, coda_mount, ext2fs_mount,
ffs_mount, filecore_mount, hfs_mount, lfs_mount, msdosfs_mount,
ntfs_mount, sysvbfs_mount, udf_mount, union_mount, sys_chflags,
sys_lchflags, sys_chmod, sys_lchmod, sys_chown, sys_lchown,
sys___posix_chown, sys___posix_lchown, sys_link, do_sys_pstatvfs,
sys_quotactl, sys_revoke, sys_truncate, do_sys_utimes, sys_extattrctl,
sys_extattr_set_file, sys_extattr_set_link, sys_extattr_get_file,
sys_extattr_get_link, sys_extattr_delete_file,
sys_extattr_delete_link, sys_extattr_list_file, sys_extattr_list_link,
sys_setxattr, sys_lsetxattr, sys_getxattr, sys_lgetxattr,
sys_listxattr, sys_llistxattr, sys_removexattr, sys_lremovexattr

All have been scrutinized (several times, in fact) and compile-tested,
but not all have been explicitly tested in action.

XXX: While I haven't (intentionally) changed the use or nonuse of
XXX: TRYEMULROOT in any of these places, I'm not convinced all the
XXX: uses are correct; an audit might be desirable.

branches: 1.26.10; 1.26.12;
Replace references to getsock/getvnode.

branches: 1.25.2;
remove clause #3 from my license where there are no other
copyright holders involved.

branches: 1.24.2;
PR kern/38141 lookup/vfs_busy acquire rwlock recursively

Simplify the mount locking. Remove all the crud to deal with recursion on
the mount lock, and crud to deal with unmount as another weirdo lock.

Hopefully this will once and for all fix the deadlocks with this. With this
commit there are two locks on each mount:

- krwlock_t mnt_unmounting. This is used to prevent unmount across critical
sections like getnewvnode(). It's only ever read locked with rw_tryenter(),
and is only ever write locked in dounmount(). A write hold can't be taken
on this lock if the current LWP could hold a vnode lock.

- kmutex_t mnt_updating. This is taken by threads updating the mount, for
example when going r/o -> r/w, and is only present to serialize updates.
In order to take this lock, a read hold must first be taken on
mnt_unmounting, and the two need to be held across the operation.

One effect of this change: previously if an unmount failed, we would make a
half hearted attempt to back out of it gracefully, but that was unlikely to
work in a lot of cases. Now while an unmount that will be aborted is in
progress, new file operations within the mount will fail instead of being
delayed. That is unlikely to be a problem though, because if the admin
requests unmount of a file system then s(he) has made a decision to deny
access to the resource.

PR kern/38135 vfs_busy/vfs_trybusy confusion

The previous fix worked, but it opened a window where mounts could have
disappeared from mountlist while the caller was traversing it using
vfs_trybusy(). Fix that.

Fix build of these files after the vfs_trybusy change. Apply the same
modification done in compat/common/vfs_syscalls_20.c:1.27.

branches: 1.21.2; 1.21.4;
Catch up with descriptor handling changes. See kern_descrip.c revision
1.173 for details.

branches: 1.20.6;
PR kern/37706 (forced unmount of file systems is unsafe):

- Do reference counting for 'struct mount'. Each vnode associated with a
mount takes a reference, and in turn the mount takes a reference to the
vfsops.
- Now that mounts are reference counted, replace the overcomplicated mount
locking inherited from 4.4BSD with a recursable rwlock.

Convert all the system call entry points from:
int foo(struct lwp *l, void *v, register_t *retval)
to:
int foo(struct lwp *l, const struct foo_args *uap, register_t *retval)
Fixup compat code to not write into 'uap' and (in some cases) to actually
pass a correctly formatted 'uap' structure with the right name to the
next routine.
A few 'compat' routines that just call standard ones have been deleted.
All the 'compat' code compiles (along with the kernels required to test
build it).
98% done by automated scripts.

branches: 1.18.4;
Remove cn_lwp from struct componentname. curlwp should be used
from on. The NDINIT() macro no longer takes the lwp parameter and
associates the credentials of the calling thread with the namei
structure.

ANSIfy most of the function definitions in sys/compat (but not ndis).
All by the magic of sed ...

branches: 1.16.2;
more VFS_STATVFS(x,y,z) fallout; change them to VFS_STATVFS(x,y). (hi, pooka!)

branches: 1.15.4;
mountlist_slock is now a mutex.

branches: 1.14.4; 1.14.6; 1.14.8; 1.14.10;
Fix typo: mnfromname -> mntfromname.

branches: 1.13.2;
kill silly constants.

Change the way that emulations locate files within the emulation root to
avoid having to allocate space in the 'stackgap'
- which is very LWP unfriendly.
The additional code for non-emulation namei() is trivial, the reduction for
the emulations is massive.
The vnode for a processes emulation root is saved in the cwdi structure
during process exec.
If the emulation root the TRYEMULROOT flag are set, namei() will do an initial
search for absolute pathnames in the emulation root, if that fails it will
retry from the normal root.
".." at the emulation root will always go to the real root, even in the middle
of paths and when expanding symlinks.
Absolute symlinks found using absolute paths in the emulation root will be
relative to the emulation root (so /usr/lib/xxx.so -> /lib/xxx.so links
inside the emulation root don't need changing).
If the root of the emulation would be returned (for an emulation lookup), then
the real root is returned instead (matching the behaviour of emul_lookup,
but being a cheap comparison here) so that programs that scan "../.."
looking for the root dircetory don't loop forever.
The target for symbolic links is no longer mangled (it used to get the
CHECK_ALT_xxx() treatment, so could get /emul/xxx prepended).
CHECK_ALT_xxx() are no more. Most of the change is deleting them, and adding
TRYEMULROOT to the flags to NDINIT().
A lot of the emulation system call stubs could now be deleted.

I'm not sure if I want to know what MNT_LAXY means, so I'll just
assume the comment has a tyop of MNT_LAZY.

Change all the NETBSD32PTR64(SCARG(uap, xxx))) to SCARG_P32(uap, xxx).

branches: 1.9.2; 1.9.4; 1.9.6;
fix fallout from caddr_t changes.

Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.7.2;
Merge newlock2 to head.

branches: 1.6.4;
Make filehandles opaque to userland

branches: 1.5.2;
Clean up fallout proc_is_traced_p() change:
- proc_is_traced_p() -> trace_is_enabled(), to match trace_enter() and
trace_exit().
- trace_is_enabled() becomes a real function.
- Remove unnecessary include files from various files that used to care
about KTRACE and SYSTRACE, but do no more.

branches: 1.4.4; 1.4.6; 1.4.8;
Remove leading __ from __(const|inline|signed|volatile) -- it is obsolete.

merge ktrace-lwp.

Fix lossage I created with the 64 bit ino_t change.

branches: 1.1.2; 1.1.14;
o Add support for the statvfs family of syscalls (statvfs1, fstatvfs1,
fhstatvfs1 and getvfsstat)
o Move the statfs family out of netbsd32_fs.c and netbsd32_netbsd.c to
netbsd_compat_20.c, compiled with COMPAT_20

Reviewed by christos@.

KNF consistency: No parentheses are needed around the return value.

branches: 1.35.6;
Fix copyout overflows in fhstat, found by the LGTM bot. Not a big problem
since this syscall is privileged.

VFS_VGET(), VFS_ROOT(), VFS_FHTOVP(): give them a "int lktype" argument, to
allow us to get shared locks (or no lock) on the returned vnode. Matches
FreeBSD.

branches: 1.33.6;
Merge the [pgoyette-compat] branch

Rename min/max -> uimin/uimax for better honesty.

These functions are defined on unsigned int. The generic name
min/max should not silently truncate to 32 bits on 64-bit systems.
This is purely a name change -- no functional change intended.

HOWEVER! Some subsystems have

#define min(a, b) ((a) < (b) ? (a) : (b))
#define max(a, b) ((a) > (b) ? (a) : (b))

even though our standard name for that is MIN/MAX. Although these
may invite multiple evaluation bugs, these do _not_ cause integer
truncation.

To avoid `fixing' these cases, I first changed the name in libkern,
and then compile-tested every file where min/max occurred in order to
confirm that it failed -- and thus confirm that nothing shadowed
min/max -- before changing it.

I have left a handful of bootloaders that are too annoying to
compile-test, and some dead code:

cobalt ews4800mips hp300 hppa ia64 luna68k vax
acorn32/if_ie.c (not included in any kernels)
macppc/if_gm.c (superseded by gem(4))

It should be easy to fix the fallout once identified -- this way of
doing things fails safe, and the goal here, after all, is to _avoid_
silent integer truncations, not introduce them.

Maybe one day we can reintroduce min/max as type-generic things that
never silently truncate. But we should avoid doing that for a while,
so that existing code has a chance to be detected by the compiler for
conversion to uimin/uimax without changing the semantics until we can
properly audit it all. (Who knows, maybe in some cases integer
truncation is actually intended!)

branches: 1.31.16; 1.31.18;
User-triggerable kmem_alloc(0).

Ok martin@ christos@

User

branches: 1.30.14; 1.30.18; 1.30.20; 1.30.28; 1.30.34; 1.30.36;
Replace M_IOV and some malloc(9)s with kmem(9), and while there:
- Fix invalid free (M_TEMP vs M_IOV) in do_sys_recvmsg(), spotted by jakllsch@.
Also, same fix in osf1_sys_sendmsg_xopen().
- Fix attempt to free non-allocated memory in error path in netbsd32___getfh30().
- Plug a memory leak in compat_43_netbsd32_orecvmsg().

branches: 1.29.4; 1.29.6;
Make all fstat(2) compat syscalls consistently use do_sys_fstat(),
instead of fd_getfile()/fd_putfile() dance.

ok by christos.

merge christos-time_t

branches: 1.27.4;
Make the emulations, exec formats, coredump, NFS, and the NFS server
into modules. By and large this commit:

- shuffles header files and ifdefs
- splits code out where necessary to be modular
- adds module glue for each of the components
- adds/replaces hooks for things that can be installed at runtime

branches: 1.26.2; 1.26.4;
Replace references to getsock/getvnode.

branches: 1.25.2;
remove clause #3 from my license where there are no other
copyright holders involved.

branches: 1.24.2; 1.24.4; 1.24.6;
Catch up with descriptor handling changes. See kern_descrip.c revision
1.173 for details.

branches: 1.23.6;
Convert all the system call entry points from:
int foo(struct lwp *l, void *v, register_t *retval)
to:
int foo(struct lwp *l, const struct foo_args *uap, register_t *retval)
Fixup compat code to not write into 'uap' and (in some cases) to actually
pass a correctly formatted 'uap' structure with the right name to the
next routine.
A few 'compat' routines that just call standard ones have been deleted.
All the 'compat' code compiles (along with the kernels required to test
build it).
98% done by automated scripts.

branches: 1.22.4;
ANSIfy most of the function definitions in sys/compat (but not ndis).
All by the magic of sed ...

branches: 1.21.6; 1.21.8; 1.21.16;
Split the statvfs functions so that the 'work' is done to a kernel buffer
which can either be copied directly to userspace, or converted then copied.
Saves replicating a lot of code in the compat functions (esp. for
getvfsstat) at a cast of an extra function call in the non-emulated case -
which is unlikely to be measurable given the other costs of the actions
involved (even on vax).
Remove dofhstat() and dofhstatvfs() (and the last caller).
Remove some redundant stackgap_init() calls.

Change the way that emulations locate files within the emulation root to
avoid having to allocate space in the 'stackgap'
- which is very LWP unfriendly.
The additional code for non-emulation namei() is trivial, the reduction for
the emulations is massive.
The vnode for a processes emulation root is saved in the cwdi structure
during process exec.
If the emulation root the TRYEMULROOT flag are set, namei() will do an initial
search for absolute pathnames in the emulation root, if that fails it will
retry from the normal root.
".." at the emulation root will always go to the real root, even in the middle
of paths and when expanding symlinks.
Absolute symlinks found using absolute paths in the emulation root will be
relative to the emulation root (so /usr/lib/xxx.so -> /lib/xxx.so links
inside the emulation root don't need changing).
If the root of the emulation would be returned (for an emulation lookup), then
the real root is returned instead (matching the behaviour of emul_lookup,
but being a cheap comparison here) so that programs that scan "../.."
looking for the root dircetory don't loop forever.
The target for symbolic links is no longer mangled (it used to get the
CHECK_ALT_xxx() treatment, so could get /emul/xxx prepended).
CHECK_ALT_xxx() are no more. Most of the change is deleting them, and adding
TRYEMULROOT to the flags to NDINIT().
A lot of the emulation system call stubs could now be deleted.

Change all the NETBSD32PTR64(SCARG(uap, xxx))) to SCARG_P32(uap, xxx).

branches: 1.18.2; 1.18.4;
Update all the compat stuff to not use the 'stackgap' for processing
sys_stat() and friends, instead use do_sys_stat() and do_sys_fstat()
that write the answer into a kernel buffer (on stack) that can be
converted to the correct form and written the userspace.
I've test compiled a few kernels, and tested i386 netbsd1.6 ls.
Given I think I've fixed some bugs, it might be 50-50 with new ones.

branches: 1.17.2;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.16.2;
Merge newlock2 to head.

Clean some KAUTH_GENERIC_ISSUSER usage in compat code.

- Make better use of COMPAT_XX type in syscalls.master
- Remove useless (thanks to COMPAT_XX behaviour) #ifdefs in
syscalls.master
- Make netbsd32_compat_43.c compiled per COMPAT_LINUX32 because the latter
needs stuff from it.

Fixes Perry's PR#34951.

branches: 1.13.4; 1.13.6;
some filehandle syscall related changes.

- remove the support of variable-sized filehandle from compat version of
syscalls. (strictly speaking, it breaks abi. i don't think it's a problem
because this feature is short-lived and there are no affected in-tree
filesystems.)
- unify vfs_copyinfh_alloc and vfs_copyinfh_alloc_size.
- vfs_copyinfh_alloc_size: check fhsize strictly.
- reduce code duplication between compat and current syscalls.

Make filehandles opaque to userland

Use the LWP cached credentials where sane.

do not protect the getfh syscall by
#if defined(NFS) || defined(NFSSERVER).
The native syscall also isn't.

Fix alignement problems for fhandle_t, exposed by gcc4.1.

While touching all vptofh/fhtovp functions, get rid of VFS_MAXFIDSIZ,
version the getfh(2) syscall and explicitly pass the size available in
the filehandle from userland.

Discussed on tech-kern, with lots of help from yamt (thanks!).

branches: 1.8.2; 1.8.4;
Limit the size of any kernel buffers allocated by the VOP_READDIR
routines to MAXBSIZE.

branches: 1.7.2;
integrate kauth.

Move fhstat as sys___fhstat30 in netbsd32_fs.c because it doesn't belong
to netbsd32_netbsd.c (which incidentally means it had been broken for a
while I guess). Add compat_30 version.

branches: 1.5.2; 1.5.4;
Clean up fallout proc_is_traced_p() change:
- proc_is_traced_p() -> trace_is_enabled(), to match trace_enter() and
trace_exit().
- trace_is_enabled() becomes a real function.
- Remove unnecessary include files from various files that used to care
about KTRACE and SYSTRACE, but do no more.

branches: 1.4.4; 1.4.6; 1.4.8;
more proc to lwp problems.

merge ktrace-lwp.

branches: 1.2.6;
Fix lossage I created with the 64 bit ino_t change.

64 bit inode changes.

branches: 1.4.4;
add missing includes

Add the appropriate module glue. Seems to have gotten lost during
the merge.

move mmap(), compat 40 mount(), pipe() and getfh() into their own files
as the implementation is non trivial or is compat. mark old mount()
properly instead of manually #if'ing ito to ENOSYS.

branches: 1.1.2;
file netbsd32_compat_40.c was initially added on branch pgoyette-compat.

revert previous, msg_flags is already initialized as MSG_NAMEMBUF
to free the sockname mbuf.

Don't use flags uninitialized.

branches: 1.63.10;
Use "stack_t" instead of "struct sigaltstack", as the former is the
newer standardized name. NFC.

sys/compat: Memset zero before copyout.

Just in case of uninitialized padding which would lead to kernel
stack disclosure. If the compiler can prove the memset redundant
then it can optimize it away; otherwise better safe than sorry.

KNF consistency: No parentheses are needed around the return value.

branches: 1.60.4;
Properly mark netbsd32_compat_43 syscalls that are directly implemented
by calling their compat_43 equivalents. With these changes, and with
built-in versions of COMPAT_NETBSD32, COMPAT_NOMID, and COMPAT_09, I can
now run a netbsd-0.9 statically linked i386 (32-bit) version of /bin/ls
on a 9.99.x amd64 host!

Addresses PR kern/55047 but more changes coming to handle non-built-in
modules.

XXX pullup-9

Fix sizeof mismatch in copyin. This leads to a user-triggerable stack
overflow. On my test build at least, by luck, the compiler orders the
variables in a way that the overflow hits only local structures which
haven't yet been initialized and used, so the overflow is harmless.

Very easily seeable with kASan - just invoke the syscall from a 32bit
binary.

Fix info leak.

branches: 1.57.4;
Merge the [pgoyette-compat] branch

branches: 1.56.2;
Fix COMPAT_NETBSD32 cmsg handling:

1. alignment was wrong for > 1 message
2. macros were doing incorrect pointer comparisons, fortunately ending
the iteration early after the fists cmsg instead of crashing.
3. don't output 32 bit ktrace records for cmsg. 32 bit programs running
under emulation on 64 bit systems should produce 64 bit ktrace records
so that the native ktrace can handle the records; remove extra arguments
that are now not needed (the 32 bit msghdr).
4. output the correct type for cmsg trace records.
5. output all the cmsg records in traces instead of just the first one.

Welcome to 8.99.15 because of the argument removal.

XXX: Really all the code should be changed to use the CMSG_{FIRST,NXT}HDR
macros...

PR/53103: Timo Buhrmester: linux emulation of sendto(2) broken

The sockargs refactoring broke it, because sockargs only works with a user
address. Added an argument to sockargs to indicate where the address is
coming from. Welcome to 8.99.14.

branches: 1.54.8; 1.54.14;
Make the ktrace record written by do_sys_sendmsg/do_sys_recvmsg overridable
by the caller. Use this in compat_netbsd32 to log the 32bit version, so
the 32bit userland kdump is happy.

branches: 1.53.18; 1.53.34; 1.53.36; 1.53.44;
Fix compilation.

Replace M_IOV and some malloc(9)s with kmem(9), and while there:
- Fix invalid free (M_TEMP vs M_IOV) in do_sys_recvmsg(), spotted by jakllsch@.
Also, same fix in osf1_sys_sendmsg_xopen().
- Fix attempt to free non-allocated memory in error path in netbsd32___getfh30().
- Plug a memory leak in compat_43_netbsd32_orecvmsg().

branches: 1.51.4; 1.51.6;
Cleanup. Kill a few netbsd32_caddr_t in syscalls argument types.

branches: 1.50.6; 1.50.12;
remove clause #3 from my license where there are no other
copyright holders involved.

branches: 1.49.2; 1.49.4; 1.49.6;
Catch up with descriptor handling changes. See kern_descrip.c revision
1.173 for details.

branches: 1.48.6;
Remove some dubious casting of the address of integer variables that
can only be 'not invalid' if the sizes match - when it is unnecessary.

Convert all the system call entry points from:
int foo(struct lwp *l, void *v, register_t *retval)
to:
int foo(struct lwp *l, const struct foo_args *uap, register_t *retval)
Fixup compat code to not write into 'uap' and (in some cases) to actually
pass a correctly formatted 'uap' structure with the right name to the
next routine.
A few 'compat' routines that just call standard ones have been deleted.
All the 'compat' code compiles (along with the kernels required to test
build it).
98% done by automated scripts.

branches: 1.46.4;
ANSIfy most of the function definitions in sys/compat (but not ndis).
All by the magic of sed ...

Remove all the __P

branches: 1.44.6; 1.44.8; 1.44.14; 1.44.16;
Changes to sompat socket function to avoid the dreaded stackgap.

The code to directly call sigaction1() and sigaltstack1() is much shorter
than the mess required to call the compat_43_xxx functions.

sendmsg() shouldn't modify the 'msg' parameter, and recvmsg() doesn't have
to modify the 'iov' array - it isn't updated.

Rework compat stat() and statvfs() code so that it no longer uses the stackgap.

Change the way that emulations locate files within the emulation root to
avoid having to allocate space in the 'stackgap'
- which is very LWP unfriendly.
The additional code for non-emulation namei() is trivial, the reduction for
the emulations is massive.
The vnode for a processes emulation root is saved in the cwdi structure
during process exec.
If the emulation root the TRYEMULROOT flag are set, namei() will do an initial
search for absolute pathnames in the emulation root, if that fails it will
retry from the normal root.
".." at the emulation root will always go to the real root, even in the middle
of paths and when expanding symlinks.
Absolute symlinks found using absolute paths in the emulation root will be
relative to the emulation root (so /usr/lib/xxx.so -> /lib/xxx.so links
inside the emulation root don't need changing).
If the root of the emulation would be returned (for an emulation lookup), then
the real root is returned instead (matching the behaviour of emul_lookup,
but being a cheap comparison here) so that programs that scan "../.."
looking for the root dircetory don't loop forever.
The target for symbolic links is no longer mangled (it used to get the
CHECK_ALT_xxx() treatment, so could get /emul/xxx prepended).
CHECK_ALT_xxx() are no more. Most of the change is deleting them, and adding
TRYEMULROOT to the flags to NDINIT().
A lot of the emulation system call stubs could now be deleted.

Change all the NETBSD32PTR64(SCARG(uap, xxx))) to SCARG_P32(uap, xxx).

remove all the double (and triple) casts used to convert 32bit userspace
pointers to and from 64bit kernel pointers. Instead use the defines
NETBSD32PTR64(p32) to read a 32bit pointer and (the new) NETBSD32PTR32(p32,p64)
to write a 32bit pointer throughout.
The 32bit pointer is now a struct to enforce the above.
amd64 (with linux emul) and sparc64 will both compile (when the arch stuff
goes in soon), and amd64 still runs some i386 binaries.

branches: 1.37.2; 1.37.4; 1.37.6;
fix fallout from caddr_t changes.

Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.35.2;
Merge newlock2 to head.

branches: 1.34.20;
merge ktrace-lwp.

need compat/sys/socket.h

need to include sys/socket.h here now, sys/mount.h no longer does it for us.

make this compile again.

branches: 1.30.4;
nuke trailing whitespace

branches: 1.29.8; 1.29.10;
Dynamic sysctl.

Gone are the old kern_sysctl(), cpu_sysctl(), hw_sysctl(),
vfs_sysctl(), etc, routines, along with sysctl_int() et al. Now all
nodes are registered with the tree, and nodes can be added (or
removed) easily, and I/O to and from the tree is handled generically.

Since the nodes are registered with the tree, the mapping from name to
number (and back again) can now be discovered, instead of having to be
hard coded. Adding new nodes to the tree is likewise much simpler --
the new infrastructure handles almost all the work for simple types,
and just about anything else can be done with a small helper function.

All existing nodes are where they were before (numerically speaking),
so all existing consumers of sysctl information should notice no
difference.

PS - I'm sorry, but there's a distinct lack of documentation at the
moment. I'm working on sysctl(3/8/9) right now, and I promise to
watch out for buses.

don't use extra variables where there is no need to.

branches: 1.27.2;
Back out the lwp/ktrace changes. They contained a lot of colateral damage,
and need to be examined and discussed more.

struct proc * -> struct lwp *

Pass lwp pointers throughtout the kernel, as required, so that the lwpid can
be inserted into ktrace records. The general change has been to replace
"struct proc *" with "struct lwp *" in various function prototypes, pass
the lwp through and use l_proc to get the process pointer when needed.

Bump the kernel rev up to 1.6V

Merge the nathanw_sa branch.

In preparation for COMPAT_NETBSD32 on SH-5:

- The MD netbsd32_machdep.h header now defines the 32-bit pointer type
instead of using u_int32_t everywhere,
- The MD netbsd32_machdep.h header now defines a macro (at least on
current implementations) which converts a 32-bit pointer to its 64-bit
equivalent,
- Change the MI code to utilise the above two items in all the right places,
- Implement netbsd32___sigaction_sigtramp().

Tested on Sparc64 by Matt Green.

Add missing prototype for compat_43_netbsd32_sigsetmask.

make the stackgap_{init,alloc} functions MP friendly (i.e. pass struct proc *
in, instead of using curproc). While there add an optional size argument to
stackgap_init.

add RCSIDs (including regeneration of files as appropriate)

update copyright notices.

Fix bug in netbsd32_compat_43_olseek return value.

branches: 1.17.2;
use _KERNEL_OPT.

branches: 1.16.2;
fix the glaring errors in emulations for osendmsg, orecvmsg, osigvec and osigstack.

properly implement compat_43_netbsd32_stat43, compat_43_netbsd32_lstat43
and compat_43_netbsd32_fstat43. now 32-bit ls(1) works. so does static
sunos `/sbin/sh' on sparc64.

de-static netbsd32_exec_aout_prep_[zno]magic().
de-static netbsd32_from_stat43().
move the guts of netbsd32_execve() into netbsd32_execve2().

all of are for the forthcoming sunos32 compat mode (for sparc64).

Include opt_* files to make sure prototypes are included.

add e_path (emulation path) to struct emul, which replaces emulation-specific
*_emul_path variables

change macros CHECK_ALT_{CREAT|EXIST} to use that, 'root' doesn't need
to be passed explicitly any more and *_CHECK_ALT_{CREAT|EXIST} are removed
change explicit emul_find() calls in probe functions to get the emulation
path from the checked exec switch entry's emulation

remove no longer needed header files

add e_flags and e_syscall to struct emul; these are unsed and empty for now

The ovadvise syscall is (correctly) listed as STD, so move it's implementation
to another place. This makes kernels without options COMPAT_43 link.

remove include of <vm/vm.h>. <vm/vm.h> -> <uvm/uvm_extern.h>

Update to 32-bit compatibility routines.

branches: 1.8.2;
Add netbsd32_compat_14.c

Rename everything from compat_netbsd32 -> netbsd32 so we don't have silly
names like compat_14_compat_netbsd32_sys_shmctl().

branches: 1.7.2;
fix errors in the copyright on this.

branches: 1.6.2;
SPARC32->NETBSD32

move sparc32 to netbsd32; split out MD part (signal handling) (mostly via repository copy, rename and perl -pi :-)

Overhaul. Now will run simple things like /bin/sh and /bin/echo.