should not be using .include <> but .include "" here.

Do not include bsd.kinc.mk in each arch Makefile, it's already pulled
by bsd.syscall.mk (from Makefile.inc).

branches: 1.2.184; 1.2.194; 1.2.200; 1.2.210;
Attempt to fix the mess.

New makefiles for includes, syscall gen.

Switch Linux futex emulation to native ucas_int(), remove unecessary
futex-specific assembler routines.

Fixes PR/40490. Reviewed by <ad>.

branches: 1.9.2; 1.9.4; 1.9.8;
futex support fixes from FreeBSD via cube+pengo.

branches: 1.8.52; 1.8.62;
Update to deal with options PTRACE

branches: 1.7.4; 1.7.8;
merge ktrace-lwp.

branches: 1.6.12;
Split out the setup stack function for the benefit of lkms.
XXX: Our lkm system sucks. We really need to be exporting symbols, but
that leads to symbol conflicts.

branches: 1.5.10;
bring the reverse signal mapping array into linux_signo.c

branches: 1.4.6; 1.4.8;
Incomplete but already working emulation of Linux's ptrace(2) for i386.

branches: 1.3.8; 1.3.14;
Speed up compilation by #include-ing C files from common into one file.

Attempt to fix the mess.

New list of files for the i386.

fix various typos, mainly in comments.

Always clear FUTEX_PRIVATE_FLAG for Linux processes. NetBSD-native futexes
exist in different namespace depending on FUTEX_PRIVATE_FLAG. This appears
not to be the case in Linux, and some futex users will mix private and non-
private ops on the same futex object. Provide a convenience wrapper that
puts this logic in one place witn a comment explaining why.

While here, move the Linux futex wrapper out of its own file and plop
it in linux_misc.c, which is where it lives in the linux32 module.

Fix path for fadvise64 files.

Add fadvise64 and fadvise64_64 syscalls to compat_linux and compat_linux32.

branches: 1.14.2; 1.14.4;
Nuke fs_lfs.h included for unknown reasons (if there was a reason
other than "accident", i hereby declare that reason invalid)

branches: 1.13.6;
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: 1.12.6; 1.12.10; 1.12.12; 1.12.16; 1.12.18;
Extract the fcntl64() suppoort from linux_file64.c into linux_fcntl64.c
Unravel some of the knots that caused linux_file64.c to be compiled twice
for an amd64 kernel (once for linux and once for linux32) with different
parts being skipped each time.

compat linux (and linux32) uid16 functions cleanup and fixes.

- Move uid16 functions to their own file linux_uid16.c, included by
needed archs (arm, i386 and m68k).
- Add new MI types linux_{u,g}id16_t.
- Add macros to handle linux_uid16_t and uid_t conversions.
- Add linux_sys_getres{uid,gid}16 syscalls, to fix an overflow with
bad sizes given to copyout when linux_sys_getres{uid,gid} are used.
- Update arm syscall table to use more uid16 functions.

branches: 1.10.2; 1.10.8; 1.10.14;
Changes to make ktrace LKM friendly and reduce ifdef KTRACE. Proposed
on tech-kern.

branches: 1.9.2; 1.9.6;
Add futex.

branches: 1.8.6; 1.8.8;
Merge newlock2 to head.

branches: 1.7.18; 1.7.48;
Merge the nathanw_sa branch.

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

add RCSIDs (including regeneration of files as appropriate)

branches: 1.4.2;
use _KERNEL_OPT.

branches: 1.3.2;
make LKM safe

add emulation of Linux truncate64(2), ftruncate64(2), *stat64(2)

branches: 1.1.8;
Speed up compilation by #include-ing C files from common into one file.

Cleanup compat linux errnos by adding generic errnos in a single file.
Use it for all arches except alpha and mips.
Add a few missing native to linux errno conversions.

branches: 1.10.34; 1.10.44; 1.10.50;
Remove clause 3 and 4 from TNF licenses

branches: 1.9.70; 1.9.72; 1.9.74;
merge ktrace-lwp.

nuke trailing whitespace

branches: 1.7.24; 1.7.32; 1.7.34;
Clarified comments about the LINUX_SCERR_SIGN macro

Moved the fix for negative errno returned to userland on the PowerPC to a
more machine independent fashion

kill local declaration of native_to_linux_errno[], one in common/linux_errno.h suffices

Introduce 2 new flags in types.h:
* __HAVE_SYSCALL_INTERN. If this is defined, e_syscall is replaced by
e_syscall_intern, which is called at key places in the kernel. This can be
used to set a MD syscall handler pointer. This obsoletes and replaces the
*_HAS_SEPARATED_SYSCALL flags.
* __HAVE_MINIMAL_EMUL. If this is defined, certain (deprecated) elements in
struct emul are omitted.

branches: 1.3.12;
Assign my copyright to TNF. Merge with others were appropriate.
Regen syscall files after script change.

Split compat/linux/linux_errno.h into i386 and common parts.

Added Linux compat code.

- LINUX_GO_RT0_SIGNATURE for compat_linux is no longer enabled by default.
it is now defined in each arch same as the other *_SIGNATURE definitions.
- add new LINUX32_GO_RT0_SIGNATURE for compat_linux32, and is defined in amd64 and aarch64.

es_arglen is now in units of bytes, update the emulations accordingly.

branches: 1.12.6; 1.12.10; 1.12.14;
account for the 16 bytes of AT_RANDOM data in the stack setup. fixes PR 48518.
use cprng_strong32() instead of random(). add AT_RANDOM support for linux32.

branches: 1.11.2; 1.11.4;
define LINUX_DEBUGLINK_SIGNATURE for i386 too.
this allows linux ldd to work.

branches: 1.10.16; 1.10.20; 1.10.22;
LINUX_GCC_SIGNATURE also matches NetBSD/i386 ELF binaries, so disable it.

branches: 1.9.6; 1.9.8;
Remove clause 3 and 4 from TNF licenses

branches: 1.8.70; 1.8.72; 1.8.74;
merge ktrace-lwp.

branches: 1.7.16;
Hide prototype of linux_exec_setup_stack() inside #ifdef _KERNEL so
that sbin/sysctl will compile.

- GC all the setup_stack functions
- add one for linux/i386

branches: 1.5.6;
add generic linux compat ELF copyargs function

this gives:
* linux sysconf(_SC_CLK_TCK) gives correct value for linux binaries (hz)
even if hz != 100
* glibc gets proper information on real/effective uid and enables
secure mode for suid binaries

g/c LINUX_COPYARGS_FUNCTION, replaced by linux ELF copyargs function
g/c alpha-specific linux ELF copyargs function and linux ELF defines

branches: 1.4.2;
restructure struct emul and execsw, in preparation to make emulations LKMable:
* move all exec-type specific information from struct emul to execsw[] and
provide single struct emul per emulation
* elf:
- kern/exec_elf32.c:probe_funcs[] is gone, execsw[] how has one entry
per emulation and contains pointer to respective probe function
- interp is allocated via MALLOC() rather than on stack
- elf_args structure is allocated via MALLOC() rather than malloc()
* ecoff: the per-emulation hooks moved from alpha and mips specific code
to OSF1 and Ultrix compat code as appropriate, execsw[] has one entry per
emulation supporting ecoff with appropriate probe function
* the makecmds/probe functions don't set emulation, pointer to emulation is
part of appropriate execsw[] entry
* constify couple of structures

Make the linux compatibilty code work on the alpha. (horay!) (at last!)
Two main changes:
Create a linux_elf64_copyargs that uses the linux specific LinuxAuxInfo
structure. This is only used on the alpha. i386 and m68k use the
standard elf copyargs function.

Since linux's approach to binary compatibilty is to look as much
like osf1 as possible, add all the osf1 syscalls that we have
implemented to the linux syscall table. This includes get/setsysinfo,
ported from FreeBSD.

In order for linux compat to work you must have COMPAT_OSF1, COMPAT_43,
COMPAT_09, COMPAT_12 and COMPAT_13 on also.

branches: 1.2.12;
Define LINUX_GCC_SIGNATURE.

Add linux i386 machine type.

Round of uvm.h cleanup.

The poorly named uvm.h is generally supposed to be for uvm-internal
users only.

- Narrow it to files that actually need it -- mostly files that need
to query whether curlwp is the pagedaemon, which should maybe be
exposed by an external header.

- Use uvm_extern.h where feasible and uvm_*.h for things not exposed
by it. We should split up uvm_extern.h but this will serve for now
to reduce the uvm.h dependencies.

- Use uvm_stat.h and #ifdef UVMHIST uvm.h for files that use
UVMHIST(ubchist), since ubchist is declared in uvm.h but the
reference evaporates if UVMHIST is not defined, so we reduce header
file dependencies.

- Make uvm_device.h and uvm_swap.h independently includable while
here.

ok chs@

don't include opt_user_ldt.h when it is not needed

Remove vm86.

Pass 3.

branches: 1.17.14; 1.17.32;
tidy up include pathname to exclude undefined behaviour

"If the characters ', \, ", //, or /* occur in the sequence between
the < and > delimiters, the behavior is undefined."

branches: 1.16.2;
many changes for COMPAT_LINUX:
- update the linux syscall table for each platform.
- support new-style (NPTL) linux pthreads on all platforms.
clone() with CLONE_THREAD uses 1 process with many LWPs
instead of separate processes.
- move the contents of sys__lwp_setprivate() into a new
lwp_setprivate() and use that everywhere.
- update linux_release[] and linux32_release[] to "2.6.18".
- adjust placement of emul fork/exec/exit hooks as needed
and adjust other emul code to match.
- convert all struct emul definitions to use named initializers.
- change the pid allocator to allow multiple pids to refer to the same proc.
- remove a few fields from struct proc that are no longer needed.
- disable the non-functional "vdso" code in linux32/amd64,
glibc works fine without it.
- fix a race in the futex code where we could miss a wakeup after
a requeue operation.
- redo futex locking to be a little more efficient.

Use struct segment_descriptor for pcb_fsd and pcb_gsd instead of int[2].

branches: 1.14.2; 1.14.4;
Use lwp_getpcb() in compat code, clean from struct user.

Trying to fix build problem. Thanks to joerg's advise.

- add new RLIMIT_AS (aka RLIMIT_VMEM) resource that limits the total
address space available to processes. this limit exists in most other
modern unix variants, and like most of them, our defaults are unlimited.
remove the old mmap / rlimit.datasize hack.

- adds the VMCMD_STACK flag to all the stack-creation vmcmd callers.
it is currently unused, but was added a few years ago.

- add a pair of new process size values to kinfo_proc2{}. one is the
total size of the process memory map, and the other is the total size
adjusted for unused stack space (since most processes have a lot of
this...)

- patch sh, and csh to notice RLIMIT_AS. (in some cases, the alias
RLIMIT_VMEM was already present and used if availble.)

- patch ps, top and systat to notice the new k_vm_vsize member of
kinfo_proc2{}.

- update irix, svr4, svr4_32, linux and osf1 emulations to support
this information. (freebsd could be done, but that it's best left
as part of the full-update of compat/freebsd.)


this addresses PR 7897. it also gives correct memory usage values,
which have never been entirely correct (since mmap), and have been
very incorrect since jemalloc() was enabled.

tested on i386 and sparc64, build tested on several other platforms.

thanks to many folks for feedback and testing but most espcially
chuq and yamt for critical suggestions that lead to this patch not
having a special ugliness i wasn't happy with anyway :-)

branches: 1.11.2; 1.11.4; 1.11.8;
changes from pengo that make initializing skype more reliable.

use a symbolic constant instead of 6.

use GUGS_SEL instead of 3.

Assert that the segment descriptors we generate for TLS will not allow
privilege elevation.

don't forget the percpu stuff and load gs immediately.

Add preliminary tls support for i386. Does not work yet. From FreeBSD.

branches: 1.5.6;
Remove clause 3 and 4 from TNF licenses

branches: 1.4.16; 1.4.18; 1.4.20;
machine/{bus,cpu,intr}.h -> sys/{bus,cpu,intr}.h

branches: 1.3.30; 1.3.44; 1.3.46; 1.3.50;
merge ktrace-lwp.

branches: 1.2.4;
nuke trailing whitespace

branches: 1.1.2; 1.1.6; 1.1.8;
Split out the setup stack function for the benefit of lkms.
XXX: Our lkm system sucks. We really need to be exporting symbols, but
that leads to symbol conflicts.

Add a few missing open(2) flags (LINUX_O_*). Fix alpha wrong values.

branches: 1.7.18; 1.7.32;
define LINUX_O_CLOEXEC.
sort definitions by value.
consistently use hex instead of octal for the values.

implement O_DIRECTORY as standardized in POSIX-2008,
for both native and linux emulations.
this fixes the rest of PR 43695.

branches: 1.5.20; 1.5.22;
Remove clause 3 and 4 from TNF licenses

branches: 1.4.116; 1.4.118; 1.4.120;
Added initial support for fcntl64 system call (file locking is not
implemented)

branches: 1.3.24; 1.3.26; 1.3.28;
Assign my copyright to TNF. Merge with others were appropriate.
Regen syscall files after script change.

Split linux_fcntl.h into i386 and common parts.

Added Linux compat code.

Remove clause 3 and 4 from TNF licenses

branches: 1.10.122; 1.10.124; 1.10.126;
The gist of the Linux _IOX definitions was the same for all platforms;
move it to the MI linux_ioctl.h and only leave the MD bits in the MD
header files.

branches: 1.9.12; 1.9.22;
Assign my copyright to TNF. Merge with others were appropriate.
Regen syscall files after script change.

Split compat/linux/linux_ioctl.h into i386 and common parts.

PR/4794: Urban Boquist: Linux cdrom ioctls are missing

* Move the Linux audio emulation into its own directory and rename it
OSS (aka VoxWare) audio emulation.
* Use the OSS audio emulation for Linux and FreeBSD.
* Add mixer emulation to the OSS emulator.

Emulate the SOUND_PCM_WRITE_CHANNELS ioctl. From PR 3356, by Lennart
Augustsson. Also add hooks for mixer ioctls.

Prototyping changes.

Separate ioctl emulation by group. Add (minimal) audio emulation.

Convert 230400 baud.

Added Linux compat code.

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.

fix typos in asymmetry, asymmetric(al), symmetrical.

branches: 1.166.2;
Rename

fpu_save_area_clear -> fpu_clear
fpu_save_area_reset -> fpu_sigreset

Clearer, and reduces a future diff. No real functional change.

branches: 1.165.4;
Remove the second argument from USERMODE and KERNELMODE, it is unused
now that we don't have vm86 anymore.

Remove vm86. Simplifies a number of critical places.

Pass 2.

branches: 1.163.4; 1.163.6;
Rename ldt->ldtstore and gdt->gdtstore on i386. It reduces the diff with
amd64, and makes it easier to track down these variables on nxr - 'ldt'
and 'gdt' being common keywords.

branches: 1.162.2; 1.162.4;
x86_alldisks can be NULL, so don't dereference it. Not tested, but obvious
enough.

Fix DIOCGPARTINFO

Replace DIOCGPART -> DIOCGPARTINFO which returns the data needed instead of
pointers.

branches: 1.159.2;
Do not uselessly include <sys/malloc.h>.

branches: 1.158.4;
Add explicit #include <x86/fpu.h> instead of relying on pcb.h including it.

Remove all references to MDL_USEDFPU and deferred fpu initialisation.
The cost of zeroing the save area on exec is minimal.
This stops the FP registers of a random process being used the first
time an lwp uses the fpu.
sendsig_siginfo() and get_mcontext() now unconditionally copy the FP
registers.
I'll remove the double-copy for signal handlers soon.
get_mcontext() might have been leaking kernel memory to userspace - and
may still do so if i386_use_fxsave is false (short copies).

Remove support for 'external' floating point units and the MS-DOS
compatible method of handling floating point exceptions.
Make kernel support for teh fpu non-optional (486SX should still work).
Only 386 cpus support external fpu, and i386 support was removed years ago.
This means that the npx code no longer uses port 0xf0 or interupt 13.
All the "npx at isa" lines go from the configs, arch/i386/isa/npx.c
is now mandatory for all i386 kernels.
I've renamed npxinit() to fpuinit() and npxinit_cpu() to fpuinit_cpu()
to match the very similar amd64 functions.
The fpu of the boot cpu is now initialised by a direct call from
cpu_configure(), this enables FP emulation for a 486SX.
(for amd64 the cr0 values are set in locore.S and similar).
This fixes a long-standing bug in linux_setregs() - which did not
save the fpu regsiters if they were active.
I've test booted a single cpu i386 kernel (using anita).
amd64 builds - none of teh changes should affect it.
The i386 XEN kernels build, but I'm not sure where they set cr0, and
it might have got lost!

Flatten the floating point register structures bu moving the
'environment' registers into the main strcuture.
There are x87 instructions that only affect the environment, but that
is no real excuse for all the sub-structures.
Rename fields so that there are far fewer collisions between the two
sets of field names.

Missing change to the fxsave structure rename

revert fpu/pcu changes until we figure out what's wrong; they cause random
freezes

Use the MI "pcu" framework for bookkeeping of npx/fpu states on x86.
This reduces the amount of MD code enormously, and makes it easier
to implement support for newer CPU features which require more fpu
state, or for fpu usage by the kernel.
For access to FPU state across CPUs, an xcall kthread is used now
rather than a dedicated IPI.
No user visible changes intended.

branches: 1.151.10; 1.151.14;
- add sigtimedwait support.
- merge the siginfo population code.

branches: 1.150.4;
Refactor ps_strings access. Based on PK_32, write either the normal
version or the 32bit compat layout in execve1. Introduce a new function
copyin_psstrings for reading it back from userland and converting it to
the native layout. Refactor procfs to share most of the code with the
kern.proc_args sysctl handler.

This material is based upon work partially supported by
The NetBSD Foundation under a contract with Joerg Sonnenberger.

branches: 1.149.2; 1.149.4;
many changes for COMPAT_LINUX:
- update the linux syscall table for each platform.
- support new-style (NPTL) linux pthreads on all platforms.
clone() with CLONE_THREAD uses 1 process with many LWPs
instead of separate processes.
- move the contents of sys__lwp_setprivate() into a new
lwp_setprivate() and use that everywhere.
- update linux_release[] and linux32_release[] to "2.6.18".
- adjust placement of emul fork/exec/exit hooks as needed
and adjust other emul code to match.
- convert all struct emul definitions to use named initializers.
- change the pid allocator to allow multiple pids to refer to the same proc.
- remove a few fields from struct proc that are no longer needed.
- disable the non-functional "vdso" code in linux32/amd64,
glibc works fine without it.
- fix a race in the futex code where we could miss a wakeup after
a requeue operation.
- redo futex locking to be a little more efficient.

branches: 1.148.2; 1.148.4;
Remove extra parenthesis. Found by cppcheck, reported by
Henning Petersen in PR 42732.

Remove extra semicolon.

Use lwp_getpcb() in compat code, clean from struct user.

Add native to linux siginfo si_status translation, used on i386 and
amd64.

Convert si_code value from native to linux.

PR port-i386/40143 Viewing an mpeg transport stream with mplayer causes crash

Fix numerous problems:

1. LDT updates are not atomic.

2. Number of processes running with private LDTs and/or I/O bitmaps
is not capped. System with high maxprocs can be paniced.

3. LDTR can be leaked over context switch.

4. GDT slot allocations can race, giving the same LDT slot to two procs.

5. Incomplete interrupt/trap frames can be stacked.

6. In some rare cases segment faults are not handled correctly.

branches: 1.142.2;
merge christos-time_t

branches: 1.141.2; 1.141.4; 1.141.6;
Add preliminary tls support for i386. Does not work yet. From FreeBSD.

Do not use native_to_linux_si_code() outside LINUX_NPTL; until i
figure a better way to handle it.

Add native to linux siginfo si_code translation, mostly for negative
values. Adjust amd64 and i386 accordingly, not sure about aother archs
i can't test.

Define a PSL_CLEARSIG macro for the psl flags to be cleared on signal delivery
and use it everywhere.

branches: 1.137.4;
Fix free() argument in linux_read_ldt.

branches: 1.136.2;
Remove clause 3 and 4 from TNF licenses

branches: 1.135.2;
Merge proc::p_mutex and proc::p_smutex into a single adaptive mutex, since
we no longer need to guard against access from hardware interrupt handlers.

Additionally, if cloning a process with CLONE_SIGHAND, arrange to have the
child process share the parent's lock so that signal state may be kept in
sync. Partially addresses PR kern/37437.

branches: 1.134.2;
Catch up with descriptor handling changes. See kern_descrip.c revision
1.173 for details.

branches: 1.133.6;
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: 1.132.4;
ANSIfy most of the function definitions in sys/compat (but not ndis).
All by the magic of sed ...

Remove all the __P

branches: 1.130.2; 1.130.4;
machine/{bus,cpu,intr}.h -> sys/{bus,cpu,intr}.h

branches: 1.129.6; 1.129.8; 1.129.12;
Call fp->f_ops->fo_ioctl() directly (not by sys_ioctl() for LINUX_VT_GETMODE
and LINUX_VT_SETMODE. One fewer use of the stackgap.

Read and write the ldt without using the stackgap.

Get ready for NPTL but don't turn it on as of yet.

rename si_sigval -> si_value to match POSIX RTS.

- linux uname -m on x86 returns i{3,4,5,6}86 make it so.
- delete the machine sysctl that never worked, and gc linux_machine.

Share the sysarch stuff between the x86 ports. PR kern/36046.

branches: 1.123.2; 1.123.4;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.122.2;
Merge newlock2 to head.

Remove more duplicate headers.
Patch by Slava Semushin <slava.semushin@gmail.com>

Again, this was tested by comparing obj files from a pristine and a patched
source tree against an i386/ALL kernel, and also for src/sbin/fsck_ffs,
src/sbin/fsdb and src/usr.sbin/makefs. Only changes in assert() line numbers
were detected in 'objdump -d' output.

Adapt to recent machdep scope changes.

Pointed out by Kouichirou Hiratsuka, thanks!

branches: 1.119.2;
__unused removal on arguments; approved by core.

Clean some KAUTH_GENERIC_ISSUSER usage in compat code.

sprinkle __unused.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.115.2;
avoid empty if

branches: 1.114.4;
Use the LWP cached credentials where sane.

integrate kauth.

branches: 1.112.4; 1.112.6; 1.112.8; 1.112.10; 1.112.12;
merge ktrace-lwp.

linux_machdepioctl: remove duplicated FILE_USE/UNUSE.

branches: 1.110.2;
- include <sys/wait.h> so that this compile
- use the code field directly, instead of redoing the logic.
- XXX: the status field must be wrong. I think that the _WSTATUS()
should not be used directly.

More accurate SIGCHLD code and status for siginfo

report the right process status

Define linux_usertrap function, and set it in struct emul. For all
but amd64, it just returns 0, doing nothing.

For amd64, it implements vsyscalls through cheating: if the faulting
address is in the vsyscall area (which is statically known on Linux/amd64),
and the intruction pointer is too, it must have been a vsyscall. In that
case, retrieve the return address from the user stack, fix up %rip and
%rsp, and just execute the normal system call. It will return as if
the vsyscall has been executed.

branches: 1.106.2;
nuke trailing whitespace

branches: 1.105.4; 1.105.6;
Remove setup stack function. It moved to linux_exec_machdep.c.

Add ptyfs pty device number translation.

Move boot device detection code from i386 and amd64 ports to x86_autoconf.c.
Rename i386_alldisks and x86_64_alldisks to x86_alldisks, adjust other
references to compensate.

rename i386 mdlwp flags from MDP_ to MDL_
to avoid confusion with mdproc flags which also have MDP_ prefixes.

remove error(1) comment.

Catch up with chuck's stack changes. Don't map the linux stuff executable
and avoid mapping segments that have 0 size.

constify sendsig/trapsignal

make compile after last change (!)

Now that we have siginfo, use it, instead of pretending to have one.

SA_SIGINFO changes

add support for non-executable mappings (where the hardware allows this)
and make the stack and heap non-executable by default. the changes
fall into two basic catagories:

- pmap and trap-handler changes. these are all MD:
= alpha: we already track per-page execute permission with the (software)
PG_EXEC bit, so just have the trap handler pay attention to it.
= i386: use a new GDT segment for %cs for processes that have no
executable mappings above a certain threshold (currently the
bottom of the stack). track per-page execute permission with
the last unused PTE bit.
= powerpc/ibm4xx: just use the hardware exec bit.
= powerpc/oea: we already track per-page exec bits, but the hardware only
implements non-exec mappings at the segment level. so track the
number of executable mappings in each segment and turn on the no-exec
segment bit iff the count is 0. adjust the trap handler to deal.
= sparc (sun4m): fix our use of the hardware protection bits.
fix the trap handler to recognize text faults.
= sparc64: split the existing unified TSB into data and instruction TSBs,
and only load TTEs into the appropriate TSB(s) for the permissions.
fix the trap handler to check for execute permission.
= not yet implemented: amd64, hppa, sh5

- changes in all the emulations that put a signal trampoline on the stack.
instead, we now put the trampoline into a uvm_aobj and map that into
the process separately.

originally from openbsd, adapted for netbsd by me.

Fix syscall_vm86 argument:
struct trapframe -> struct trapframe *

Explain a diagnostic message a bit better, and make it a debug only message.

- GC all the setup_stack functions
- add one for linux/i386

use copyout() instead of subyte()

Implement linux_rt_sigreturn for real. Add a hack for more plausible siginfo.
The java segv handler still segv's again inside the signal handler, but our
sample test program now works [gets the proper ucontext].

branches: 1.89.2;
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

Change 'data' argument to fo_ioctl and fo_fcntl from 'caddr_t' to 'void *'.
Avoids a lot of casting and removes the need for some line breaks.
Removed a load of (caddr_t) casts from calls to copyin/copyout as well.
(approved by christos - he has a plan to remove caddr_t...)

make compat_linux struct file interlock friendly.

Merge the nathanw_sa branch.

Fix compile error.

add a linux_rt_sigframe to support SA_SIGINFO, and use it when SA_SIGINFO
is requested. This appears to be how linuxthreads decides which signal handler
to use.

rename sa_ -> linux_sa_
first attempt at providing siginfo_t to the signal handler.

Get rid of pcb_flags which is no longer existent.

Don't include <sys/map.h>.

Merge the gehenna-devsw branch into the trunk.

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

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

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

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

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

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

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

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

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

Add kernel support for having userland provide the signal trampoline:

* struct sigacts gets a new sigact_sigdesc structure, which has the
sigaction and the trampoline/version. Version 0 means "legacy kernel
provided trampoline". Other versions are coordinated with machine-
dependent code in libc.
* sigaction1() grows two more arguments -- the trampoline pointer and
the trampoline version.
* A new __sigaction_sigtramp() system call is provided to register a
trampoline along with a signal handler.
* The handler is no longer passed to sensig() functions. Instead,
sendsig() looks up the handler by peeking in the sigacts for the
process getting the signal (since it has to look in there for the
trampoline anyway).
* Native sendsig() functions now select the appropriate trampoline and
its arguments based on the trampoline version in the sigacts.

Changes to libc to use the new facility will be checked in later. Kernel
version not bumped; we will ride the 1.6C bump made recently.

Fix bug introduced in rev. 1.69 - linux_fakedev() returned always
zero for nontranslated raw devices.
This fixes recently discovered jdb /dev/dsp stdin problem, reported
on current-users by Mark Davies.

branches: 1.76.2;
Remove an unreachable "return 0;".

save/restore fs and gs; this should be pulled up.

make signal array handling uniform, and put signal arrays in a separate
file. This is simular to errno array handling.

1. fill in the ldt.sd_xx field, for total compat with linux.
2. in setregs, use gdt's instead of ldt's because linux wants to play with
all the ldts. Apparently linuxthreads uses ldt's to allocate space for
pthread_t's.

sync with current reality and fix debugging.

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.

1. load cr2 now that we have it.
2. ldt_info.base_addr == 0 && ldt_info.limit == 0 means zero the ldt
From Robert V. Barron's wabi patch.

- add a real linux_sys___sysctl because the mib numbers are different.
implement only the kernel version getting entries.
- make the signal conversion functions consistent; dest is arg 1.
- make linux_fakedev deal with block and char devices differently.
- add linux_sys_ugetrlimit, and flesh out the regular get/setrlimit so
that they work properly.
- add linux_mmap2 [untested].
- bump kernel to 2.4.18, and make the date be valentine's day :-)
- linux_sys_*stat64 family was totally busted. Fix it.

tested only on i386.

Status:
gdb, telnet, work
jdk-1.4.0 extractor works, jdk still coredumps.

christos

add RCSIDs (including regeneration of files as appropriate)

Deal with the SSE FPU save format.

execute signal handlers on separate signal stack, if it's been provided

branches: 1.65.2;
Add %fs/%gs to trap frame and save/restore them on
trap/interrupt/syscall entry from userspace.

Remove special-case "by hand" validation of fs/gs register values as
well as special handling of them in various signal handling paths.

Now, like %ds and %es, they are validated by the hardware on return to
userland.

This paves the way for the use of %fs for per-cpu data on
multiprocessor systems, and fixes an otherwise difficult-to-fix
interaction between threads/clone(2) and USER_LDT.

Discussed in advance with Frank van der Linden.

Fix a partial construction problem that can cause race conditions
between creation of a file descriptor and close(2) when using kernel
assisted threads. What we do is stick descriptors in the table, but
mark them as "larval". This causes essentially everything to treat
it as a non-existent descriptor, except for fdalloc(), which sees a
filled slot so that it won't (incorrectly) allocate it again. When
a descriptor is fully constructed, the code that has constructed it
marks it as "mature" (which actually clears the "larval" flag), and
things continue to work as normal.

While here, gather all the code that gets a descriptor from the table
into a fd_getfile() function, and call it, rather than having the
same (sometimes incorrect) code copied all over the place.

use _KERNEL_OPT.

branches: 1.62.2;
Moved include <sys/ioctl.h> so that we can build without WSCONS enabled.

constify

EINVAL -> ENOTTY

Add emulation of KIOCSOUND.

Implement KDGETMODE ioctl.

Under Linux it's possible to return values through the ioctl system
call directly (i.e. not a reference argument, but a direct return
value). So far I have only seen this with binaries that talk
to their own device modules, using non-standard ioctl calls.
Hence, allow for a pass-through ioctl call using PTIOCLINUX to
return EJUSTRETURN, signalling that the return value is in
the 'data' field of the structure, which can then be put in 'retval'.

split off thread specific stuff from struct sigacts to struct sigctx, leaving
only signal handler array sharable between threads
move other random signal stuff from struct proc to struct sigctx

This addresses kern/10981 by Matthew Orgass.

For unknown ioctls, pass them through to a device using PTIOCLINUX,
so that it may deal with them if it knows them (i.e. LKMs which
deal with Linux binaries).

Allow for passing through of certain ranges of ioctl command values
directly to a device. The device has to be aware of how to handle
with these special ioctls. Useful for LKMs that deal with Linux
binaries.

Fix previous:
- EBADF, not NULL when a file descriptor can't be found.
- Check FIF_WANTCLOSE properly.

Add i386-specific code for a few ioctl calls.

LKMify

branches: 1.50.2;
Remove the pcvt support.

branches: 1.49.2;
Remove conditionization upon XSERVER -- all the code
so conditionalized is already conditionalized on (NWSDISPLAY > 0),
so is wscons-specific. wscons has no conditionalization on XSERVER;
neither should the linux compatibility code.

Removing this is necessary to keep linux X server emulation functionining
after GENERIC rev 1.337 removing "options XSERVER" (relegating it
to the pccons/pcvt-specific, and therefore depricated).

Kill register declarations.

branches: 1.47.2;
Handle the newer, larger sigset_t Linux now uses.

Rearrange the emulation-specific setregs() entry to set up the FPU control
word *after* calling the native setregs(), which would clobber it again.

Assign my copyright to TNF. Merge with others were appropriate.
Regen syscall files after script change.

Attempt to fix the mess.

Moved here from arch/i386/i386/linux_machdep.c.

Substantial signal handling changes:
* Increase the size of sigset_t to accomodate 128 signals -- adding new
versions of sys_setprocmask(), sys_sigaction(), sys_sigpending() and
sys_sigsuspend() to handle the changed arguments.
* Abstract the guts of sys_sigaltstack(), sys_setprocmask(), sys_sigaction(),
sys_sigpending() and sys_sigsuspend() into separate functions, and call them
from all the emulations rather than hard-coding everything. (Avoids uses
the stackgap crap for these system calls.)
* Add a new flag (p_checksig) to indicate that a process may have signals
pending and userret() needs to do the full (slow) check.
* Eliminate SAS_ALTSTACK; it's exactly the inverse of SS_DISABLE.
* Correct emulation bugs with restoring SS_ONSTACK.
* Make the signal mask in the sigcontext always use the emulated mask format.
* Store signals internally in sigaction structures, rather than maintaining a
bunch of little sigsets for each SA_* bit.
* Keep track of where we put the signal trampoline, rather than figuring it out
in *_sendsig().
* Issue a warning when a non-emulated sigaction bit is observed.
* Add missing emulated signals, and a native SIGPWR (currently not used).
* Implement the `not reset when caught' semantics for relevant signals.

Note: Only code touched by the i386 port has been modified. Other ports and
emulations need to be updated.

Since wscons can support VT_xxx ioctl()s, enable them if NWSDISPLAY > 0.
Implement KDGKBTYPE and KDGKBENT, this will make X servers happy.

Fix some arithmetics lossage on typeless pointers.

Store the initial NPX control word appropriate for the emulation in the saved
FPU state, and use it when initializing the FPU on demand.

Need foo_exec.h.

Add local versions of setregs(), so I can tweak some registers later.

Generate dependencies on the USER_LDT option.

Generate dependencies on the VM86 option.

branches: 1.34.2;
Pull in vmparam.h as necessary.

branches: 1.33.12;
Workaround for building linux-emulation LKM.

backout previous kprintf changes

Need <sys/exec.h> for the PS_STRINGS definition.

printf -> kprintf, sprintf -> ksprintf

Prototype and gcc -Wall cleanups

Turn off the alignment check flag when entering a signal handler.

Turn off the trace flag when delivering a signal.

Export [gs]et_vflags(), and always use them when saving/restoring registers
in VM86 mode.
Allow changing of *all* PSL bits in VM86 mode; some applications don't work
right otherwise, and all the nasty bits are virtualized anyway.
Make sure PSL_VIF, PSL_VIP, and PSL_VM don't show up in the user's view of
the PSL.

Add a message to help us debug Linux compat.

Changed name of sigaltstack's ss_base field to ss_sp to match XPG4.2 and
traditional usage.

Switch to user code and data selectors in the GDT while handling signals.

Add USERMODE() and KERNELMODE() macros, and use them.

Implement the KDGKBMODE ioctl for Linux (possible with pcvt 3.32).

Prefix names of system call implementation functions with `sys_'.

Make system calls conform to a standard prototype and bring those
prototypes into scope.

Fix PR #1445 (incorrect ifdef/else/endif), from Bill Sommerfeld via
John Kohl.

Get notion of VT number right.

SA_ONSTACK --> SS_ONSTACK

Only define some vars (one of pcvt-specific type) when pcvt is configured.
Fixes PR #1424 (from Mike Long)

Remove VT-switch disabling hack now that pcvt handles this.

Add some yucky code to emulate Linux VTs well enough to make svgalib
stuff work reasonably well.

Export bsd_to_linux_sig[], and use it.

Rearrange #includes.

Stackgap_init() now takes an p->p_emul argument.

Include file change.

Turn off a debugging printf().

Improve Linux signal emulation, and emulate the modify_ldt() call.

Add stub for linux_modify_ldt().

Fix typos.

Save and restore %fs and %gs across signals.

Update to match include files. Eliminate _ucodesel, _udatasel, and IdlePTD.

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

Changes because of proper Linux signal stackframe building, and
seperate signal trampoline code. Ugly hack in trap.c no longer
needed.

many changes for COMPAT_LINUX:
- update the linux syscall table for each platform.
- support new-style (NPTL) linux pthreads on all platforms.
clone() with CLONE_THREAD uses 1 process with many LWPs
instead of separate processes.
- move the contents of sys__lwp_setprivate() into a new
lwp_setprivate() and use that everywhere.
- update linux_release[] and linux32_release[] to "2.6.18".
- adjust placement of emul fork/exec/exit hooks as needed
and adjust other emul code to match.
- convert all struct emul definitions to use named initializers.
- change the pid allocator to allow multiple pids to refer to the same proc.
- remove a few fields from struct proc that are no longer needed.
- disable the non-functional "vdso" code in linux32/amd64,
glibc works fine without it.
- fix a race in the futex code where we could miss a wakeup after
a requeue operation.
- redo futex locking to be a little more efficient.

branches: 1.36.6; 1.36.8;
Remove LKMs and switch to the module framework, pass 1.

Proposed on tech-kern@.

branches: 1.35.2;
use a symbolic constant instead of 6.

Add preliminary tls support for i386. Does not work yet. From FreeBSD.

branches: 1.33.6;
Remove clause 3 and 4 from TNF licenses

branches: 1.32.28; 1.32.30; 1.32.32;
Get ready for NPTL but don't turn it on as of yet.

cleanup some bitrot; from Matthew Orgass

- linux uname -m on x86 returns i{3,4,5,6}86 make it so.
- delete the machine sysctl that never worked, and gc linux_machine.

branches: 1.29.26; 1.29.30; 1.29.32;
merge ktrace-lwp.

branches: 1.28.12;
Add ptyfs pty device number translation.

SA_SIGINFO changes

Implement linux_rt_sigreturn for real. Add a hack for more plausible siginfo.
The java segv handler still segv's again inside the signal handler, but our
sample test program now works [gets the proper ucontext].

branches: 1.25.6;
add a linux_rt_sigframe to support SA_SIGINFO, and use it when SA_SIGINFO
is requested. This appears to be how linuxthreads decides which signal handler
to use.

rename sa_ -> linux_sa_
first attempt at providing siginfo_t to the signal handler.

Merge the gehenna-devsw branch into the trunk.

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

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

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

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

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

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

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

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

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

Add kernel support for having userland provide the signal trampoline:

* struct sigacts gets a new sigact_sigdesc structure, which has the
sigaction and the trampoline/version. Version 0 means "legacy kernel
provided trampoline". Other versions are coordinated with machine-
dependent code in libc.
* sigaction1() grows two more arguments -- the trampoline pointer and
the trampoline version.
* A new __sigaction_sigtramp() system call is provided to register a
trampoline along with a signal handler.
* The handler is no longer passed to sensig() functions. Instead,
sendsig() looks up the handler by peeking in the sigacts for the
process getting the signal (since it has to look in there for the
trampoline anyway).
* Native sendsig() functions now select the appropriate trampoline and
its arguments based on the trampoline version in the sigacts.

Changes to libc to use the new facility will be checked in later. Kernel
version not bumped; we will ride the 1.6C bump made recently.

branches: 1.21.8;
- add a real linux_sys___sysctl because the mib numbers are different.
implement only the kernel version getting entries.
- make the signal conversion functions consistent; dest is arg 1.
- make linux_fakedev deal with block and char devices differently.
- add linux_sys_ugetrlimit, and flesh out the regular get/setrlimit so
that they work properly.
- add linux_mmap2 [untested].
- bump kernel to 2.4.18, and make the date be valentine's day :-)
- linux_sys_*stat64 family was totally busted. Fix it.

tested only on i386.

Status:
gdb, telnet, work
jdk-1.4.0 extractor works, jdk still coredumps.

christos

branches: 1.20.2; 1.20.4;
Add emulation of KIOCSOUND.

Implement KDGETMODE ioctl.

Some namespace hiding I forgot before.

Allow for passing through of certain ranges of ioctl command values
directly to a device. The device has to be aware of how to handle
with these special ioctls. Useful for LKMs that deal with Linux
binaries.

Introduce 2 new flags in types.h:
* __HAVE_SYSCALL_INTERN. If this is defined, e_syscall is replaced by
e_syscall_intern, which is called at key places in the kernel. This can be
used to set a MD syscall handler pointer. This obsoletes and replaces the
*_HAS_SEPARATED_SYSCALL flags.
* __HAVE_MINIMAL_EMUL. If this is defined, certain (deprecated) elements in
struct emul are omitted.

Add i386-specific code for a few ioctl calls.

how many different prototypes does linux_syscall() need finally? sheesh!

if machdep define LINUX_MACHDEP_HAS_SEPARATED_SYSCALL is defined, put
linux_syscall() to emul_linux
define LINUX_MACHDEP_HAS_SEPARATED_SYSCALL for i386

branches: 1.12.2;
Fix a comment.

Remove the pcvt support.

branches: 1.10.2; 1.10.10;
Handle the newer, larger sigset_t Linux now uses.

Assign my copyright to TNF. Merge with others were appropriate.
Regen syscall files after script change.

Moved here from arch/i386/include/linux_machdep.h.

Substantial signal handling changes:
* Increase the size of sigset_t to accomodate 128 signals -- adding new
versions of sys_setprocmask(), sys_sigaction(), sys_sigpending() and
sys_sigsuspend() to handle the changed arguments.
* Abstract the guts of sys_sigaltstack(), sys_setprocmask(), sys_sigaction(),
sys_sigpending() and sys_sigsuspend() into separate functions, and call them
from all the emulations rather than hard-coding everything. (Avoids uses
the stackgap crap for these system calls.)
* Add a new flag (p_checksig) to indicate that a process may have signals
pending and userret() needs to do the full (slow) check.
* Eliminate SAS_ALTSTACK; it's exactly the inverse of SS_DISABLE.
* Correct emulation bugs with restoring SS_ONSTACK.
* Make the signal mask in the sigcontext always use the emulated mask format.
* Store signals internally in sigaction structures, rather than maintaining a
bunch of little sigsets for each SA_* bit.
* Keep track of where we put the signal trampoline, rather than figuring it out
in *_sendsig().
* Issue a warning when a non-emulated sigaction bit is observed.
* Add missing emulated signals, and a native SIGPWR (currently not used).
* Implement the `not reset when caught' semantics for relevant signals.

Note: Only code touched by the i386 port has been modified. Other ports and
emulations need to be updated.

Since wscons can support VT_xxx ioctl()s, enable them if NWSDISPLAY > 0.
Implement KDGKBTYPE and KDGKBENT, this will make X servers happy.

Add missing prototypes.

A couple of definitions added for previous changes.

Improve Linux signal emulation, and emulate the modify_ldt() call.

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

Arch dependant defs for Linux. Just signal stackframe definition for now.

Support Linux MAP_LOCKED mmap flag.

branches: 1.6.44; 1.6.64;
Remove clause 3 and 4 from TNF licenses

branches: 1.5.70; 1.5.72; 1.5.74;
merge ktrace-lwp.

g/c LINUX_MAP_IGNMASK define - it's not used anywhere

branches: 1.3.46;
Assign my copyright to TNF. Merge with others were appropriate.
Regen syscall files after script change.

Split compat/linux/linux_mmap.h into i386 and common parts.

Added Linux compat code.

G/C ISSET() macro copies. It has been defined in <sys/types.h> since 2006.

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.

Move proc_lock into the data segment. It was dynamically allocated because
at the time we had mutex_obj_alloc() but not __cacheline_aligned.

branches: 1.33.4;
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: 1.32.2; 1.32.4;
Remove the filesystem tracing feature

This is a legacy interface from 4.4BSD, and it was
introduced to overcome shortcomings of ptrace(2) at that time, which are
no longer relevant (performance). Today /proc/#/ctl offers a narrow
subset of ptrace(2) commands and is not applicable for modern
applications use beyond simplistic tracing scenarios.

This removal will simplify kernel internals. Users will still be able to
use all the other /proc files.

This change won't affect other procfs files neither Linux compat
features within mount_procfs(8). /proc/#/ctl isn't available on Linux.

Remove:
- /proc/#/ctl from mount_procfs(8)
- P_FSTRACE note from the documentation of ps(1)
- /proc/#/ctl and filesystem tracing documentation from mount_procfs(8)
- KAUTH_REQ_PROCESS_PROCFS_CTL documentation from kauth(9)
- source code file miscfs/procfs/procfs_ctl.c
- PFSctl and procfs_doctl() from sys/miscfs/procfs/procfs.h
- KAUTH_REQ_PROCESS_PROCFS_CTL from sys/sys/kauth.h
- PSL_FSTRACE (0x00010000) from sys/sys/proc.h
- P_FSTRACE (0x00010000) from sys/sys/sysctl.h

Reduce code complexity after removal of this functionality.

Update TODO.ptrace accordingly: remove two entries about /proc tracing.

Do not keep legacy notes as comments in the headers about removed
PSL_FSTRACE / P_FSTRACE, as this interface had little number of users
(close or equal to zero).

Proposed on tech-kern@.

All filesystem tracing utility users are encouraged to switch to ptrace(2).

Sponsored by <The NetBSD Foundation>

branches: 1.31.10;
Don't release proc_lock until we're done looking at things that are
protected by the lock, particularly p_stat and p_waited. Found by
Robert Elz.

XXX Pullup to NetBSD-7, -6, -6-0, and -6-1

branches: 1.30.2;
Do not uselessly include <sys/malloc.h>.

fix leak

branches: 1.28.2;
There are two times the same branches.

} else if (addr == LUSR_OFF(__signal)) {
error = ENOTSUP;
} else if (addr == LUSR_OFF(__signal)) {
error = ENOTSUP;
}

Just delete one of them. Spotted by my code scanner.

ok christos@

branches: 1.27.2;
Remove __HAVE_PROCESS_XFPREGS and add the extra parameter for the size
of the fp save area to all the process_read_fpregs() and
process_write_fpregs() functions.
None of the functions have been modified to use the new parameters.
The size is set for all the writes, but some of the arch-specific reads
just pass NULL.
The amd64 (and i386) need variable sized fp register save areas in order
to support AVX and other enhanced register areas.
These functions are rarely called - so the extra argument won't matter.

branches: 1.26.8; 1.26.14; 1.26.18; 1.26.20; 1.26.22; 1.26.28;
many changes for COMPAT_LINUX:
- update the linux syscall table for each platform.
- support new-style (NPTL) linux pthreads on all platforms.
clone() with CLONE_THREAD uses 1 process with many LWPs
instead of separate processes.
- move the contents of sys__lwp_setprivate() into a new
lwp_setprivate() and use that everywhere.
- update linux_release[] and linux32_release[] to "2.6.18".
- adjust placement of emul fork/exec/exit hooks as needed
and adjust other emul code to match.
- convert all struct emul definitions to use named initializers.
- change the pid allocator to allow multiple pids to refer to the same proc.
- remove a few fields from struct proc that are no longer needed.
- disable the non-functional "vdso" code in linux32/amd64,
glibc works fine without it.
- fix a race in the futex code where we could miss a wakeup after
a requeue operation.
- redo futex locking to be a little more efficient.

Remove pfind() and pgfind(), fix locking in various broken uses of these.
Rename real routines to proc_find() and pgrp_find(), remove PFIND_* flags
and have consistent behaviour. Provide proc_find_raw() for special cases.
Fix memory leak in sysctl_proc_corename().

COMPAT_LINUX: rework ptrace() locking, minimise differences between
different versions per-arch.

Note: while this change adds some formal cosmetics for COMPAT_DARWIN and
COMPAT_IRIX - locking there is utterly broken (for ages).

Fixes PR/43176.

branches: 1.24.2; 1.24.4;
Remove uarea swap-out functionality:

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

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

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

kill MALLOC and FREE macros.

branches: 1.22.2; 1.22.8;
Remove clause 3 and 4 from TNF licenses

branches: 1.21.2;
Disable bitrotted linux ptrace interfaces, which are likely now a
security hole. Can be re-enabled at runtime. If you want to repair
them please look at the native implementation.

branches: 1.20.6; 1.20.8;
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: 1.19.4;
ANSIfy most of the function definitions in sys/compat (but not ndis).
All by the magic of sed ...

branches: 1.18.6; 1.18.8; 1.18.16;
Merge some of the less invasive changes from the vmlocking branch:

- kthread, callout, devsw API changes
- select()/poll() improvements
- miscellaneous MT safety improvements

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

branches: 1.16.2;
Merge newlock2 to head.

branches: 1.15.2;
When calling PTRACE from an LKM, use sysent[SYS_ptrace].sy_call in case
the sys_ptrace symbol isn't present.

Update to deal with options PTRACE

branches: 1.13.4; 1.13.8;
merge ktrace-lwp.

branches: 1.12.4;
nuke trailing whitespace

branches: 1.11.2; 1.11.10; 1.11.12;
Merge the nathanw_sa branch.

- Rename struct linux_ctx to linux_fpctx.
- KNF.

Fix namespace collision caused by recent change to "linux_machdep.h".

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

add RCSIDs (including regeneration of files as appropriate)

branches: 1.6.2;
delint: fix comments

branches: 1.5.2;
implement PTRACE_{GET|SET}FPREGS and framework for PTRACE_{PEEK|POKE}USER

this makes it possible to attach to linux process with linux gdb and
see top-most function on traceback; lower functions are not available,
probably due to bad frame setup
it's not possible to setup breakpoints - Linux gdb uses hw breakpoints,
so support for them would need to be written

branches: 1.4.2;
move definition of LINUX_PTRACE_{SET|GET}(FP)REGS flags to
sys/compat/linux/common/linux_ptrace.h, they seem to be actually MI

branches: 1.3.2;
Add support for reading and writing non FPU register with ptrace(2).

Support a few more request types in ptrace(2) emulation.

Incomplete but already working emulation of Linux's ptrace(2) for i386.

Remove clause 3 and 4 from TNF licenses

branches: 1.24.108; 1.24.110; 1.24.112;
revert previous

make it compile (missing includes)

bring the reverse signal mapping array into linux_signo.c

make signal array handling uniform, and put signal arrays in a separate
file. This is simular to errno array handling.

- Now that RT-signals fit, fix the array to deliver them. [i386 only].
- jdk-1.4 works perfectly now. :-)

add RCSIDs (including regeneration of files as appropriate)

branches: 1.18.2; 1.18.4;
constify

branches: 1.17.2;
Handle the newer, larger sigset_t Linux now uses.

Assign my copyright to TNF. Merge with others were appropriate.
Regen syscall files after script change.

Attempt to fix the mess.

Split compat/linux/linux_signal.c into i386, common and multi-architecture parts.

Bracket stray printf()s by #ifdef DEBUG.

Substantial signal handling changes:
* Increase the size of sigset_t to accomodate 128 signals -- adding new
versions of sys_setprocmask(), sys_sigaction(), sys_sigpending() and
sys_sigsuspend() to handle the changed arguments.
* Abstract the guts of sys_sigaltstack(), sys_setprocmask(), sys_sigaction(),
sys_sigpending() and sys_sigsuspend() into separate functions, and call them
from all the emulations rather than hard-coding everything. (Avoids uses
the stackgap crap for these system calls.)
* Add a new flag (p_checksig) to indicate that a process may have signals
pending and userret() needs to do the full (slow) check.
* Eliminate SAS_ALTSTACK; it's exactly the inverse of SS_DISABLE.
* Correct emulation bugs with restoring SS_ONSTACK.
* Make the signal mask in the sigcontext always use the emulated mask format.
* Store signals internally in sigaction structures, rather than maintaining a
bunch of little sigsets for each SA_* bit.
* Keep track of where we put the signal trampoline, rather than figuring it out
in *_sendsig().
* Issue a warning when a non-emulated sigaction bit is observed.
* Add missing emulated signals, and a native SIGPWR (currently not used).
* Implement the `not reset when caught' semantics for relevant signals.

Note: Only code touched by the i386 port has been modified. Other ports and
emulations need to be updated.

bzero->memset, bcopy->memcpy, bcmp->memcmp

- Add missing prototypes
- Fix bugs with reversed arguments that prototyping discovered

Prefix names of system call implementation functions with `sys_'.

Make system calls conform to a standard prototype and bring those
prototypes into scope.

Move signal types into signal.h.

Convert SA_NOMASK and SA_ONESHOT.
Clean up this code a bit.

Use compat_util.[ch].

* Changed to use generic ELF code from kern/exec_elf.c (which was taken
from the svr4 exec code plus the linux compat mods)
* Include file change
* Add multicast sockopts (from John Brezak)

Add the Linux pause() system call. Fix the arguments to sigsuspend().

Fix typo in comment.

Added Linux compat code.

- add sigtimedwait support.
- merge the siginfo population code.

branches: 1.7.16;
Move si_code definitions from MD to MI linux_siginfo.h, all archs
(except mips for 3 values) are identical.
While here, remove unused/unneeded LINUX__SI_CODE macro. Only the
lower 16-bits are used for userland si_code.

branches: 1.6.2;
Add preliminary tls support for i386. Does not work yet. From FreeBSD.

branches: 1.5.6;
Remove clause 3 and 4 from TNF licenses

branches: 1.4.70; 1.4.72; 1.4.74;
merge ktrace-lwp.

nuke trailing whitespace

branches: 1.2.8; 1.2.10;
Implement linux_rt_sigreturn for real. Add a hack for more plausible siginfo.
The java segv handler still segv's again inside the signal handler, but our
sample test program now works [gets the proper ucontext].

branches: 1.1.2; 1.1.4; 1.1.10;
Make this match reality [not used yet]

Remove clause 3 and 4 from TNF licenses

branches: 1.15.12; 1.15.14; 1.15.16;
Remove all the __P

branches: 1.14.44; 1.14.46; 1.14.52; 1.14.56;
merge ktrace-lwp.

MINSIGSTKSZ on linux is different depending on the platform and usually
smaller than ours.

branches: 1.12.6; 1.12.22;
rename sa_ -> linux_sa_
first attempt at providing siginfo_t to the signal handler.

- Now that RT-signals fit, fix the array to deliver them. [i386 only].
- jdk-1.4 works perfectly now. :-)

- add a real linux_sys___sysctl because the mib numbers are different.
implement only the kernel version getting entries.
- make the signal conversion functions consistent; dest is arg 1.
- make linux_fakedev deal with block and char devices differently.
- add linux_sys_ugetrlimit, and flesh out the regular get/setrlimit so
that they work properly.
- add linux_mmap2 [untested].
- bump kernel to 2.4.18, and make the date be valentine's day :-)
- linux_sys_*stat64 family was totally busted. Fix it.

tested only on i386.

Status:
gdb, telnet, work
jdk-1.4.0 extractor works, jdk still coredumps.

christos

branches: 1.9.2; 1.9.4;
add support for syscall 186 [sigaltstack]

branches: 1.8.2;
Handle the newer, larger sigset_t Linux now uses.

Assign my copyright to TNF. Merge with others were appropriate.
Regen syscall files after script change.

Split compat/linux/linux_signal.h into i386 and common parts.

Substantial signal handling changes:
* Increase the size of sigset_t to accomodate 128 signals -- adding new
versions of sys_setprocmask(), sys_sigaction(), sys_sigpending() and
sys_sigsuspend() to handle the changed arguments.
* Abstract the guts of sys_sigaltstack(), sys_setprocmask(), sys_sigaction(),
sys_sigpending() and sys_sigsuspend() into separate functions, and call them
from all the emulations rather than hard-coding everything. (Avoids uses
the stackgap crap for these system calls.)
* Add a new flag (p_checksig) to indicate that a process may have signals
pending and userret() needs to do the full (slow) check.
* Eliminate SAS_ALTSTACK; it's exactly the inverse of SS_DISABLE.
* Correct emulation bugs with restoring SS_ONSTACK.
* Make the signal mask in the sigcontext always use the emulated mask format.
* Store signals internally in sigaction structures, rather than maintaining a
bunch of little sigsets for each SA_* bit.
* Keep track of where we put the signal trampoline, rather than figuring it out
in *_sendsig().
* Issue a warning when a non-emulated sigaction bit is observed.
* Add missing emulated signals, and a native SIGPWR (currently not used).
* Implement the `not reset when caught' semantics for relevant signals.

Note: Only code touched by the i386 port has been modified. Other ports and
emulations need to be updated.

Add a hook for MD ioctl calls, plus a couple of more changes to make
svgalib binaries work on the i386

Export bsd_to_linux_sig[], and use it.

Move signal types into signal.h.

Added Linux compat code.

Update linux socket socket options (SO_*). Add common version to its
own file (linux_socket_generic.h); use it for all but alpha and mips.

branches: 1.8.34; 1.8.44; 1.8.50;
Remove clause 3 and 4 from TNF licenses

branches: 1.7.70; 1.7.72; 1.7.74;
merge ktrace-lwp.

add LINUX_SO_PEERNAME, LINUX_SO_TIMESTAMP where it has been missing
fix value for LINUX_SO_DETACH_FILTER on couple places (unused anyway)

g/c LINUX_SCM_TIMESTAMP definition for some archs, it will be defined
in generic linux_socket.h

branches: 1.5.46;
Assign my copyright to TNF. Merge with others were appropriate.
Regen syscall files after script change.

Split linux_socket.h into i386 and common parts.

Convert TCP socket options.

Convert IP socket options.

Added Linux compat code.

Switch Linux futex emulation to native ucas_int(), remove unecessary
futex-specific assembler routines.

Fixes PR/40490. Reviewed by <ad>.

branches: 1.4.4; 1.4.6; 1.4.8;
Unneeded ifdef

branches: 1.3.2; 1.3.4;
Need opt_multiprocessor.h for MULTIPROCESSOR definition.

multiprocessor fixes, from rmind

futex support fixes from FreeBSD via cube+pengo.

regen

regen

regen

regen

branches: 1.127.2;
regen

regen

regen

regen

regen

regen

regen

Regen for preadv(2) / pwritev(2).

Regen for eventfd.

Regen for timerfd.

Regen for POSIX timer syscalls.

branches: 1.116.2;
Regen for native futex calls.

regen

Regen

branches: 1.113.8; 1.113.10;
regen

branches: 1.112.4;
regen all

branches: 1.111.2;
regen

Regen (something apparently went wrong in previous)

Regen

Regen for readlink/readlinkat signatures.

branches: 1.107.2;
regen

Regen

regen

branches: 1.104.2;
Regen for ppoll signature fix

branches: 1.103.2;
Regen for munmap(2) signature.

Regen for utimes(2) change.

Regen for getgroups/setgroups update.

Regen for fadvise64 offset types changes.

Regen for pread/pwrite 64bit offset argument.

Regen.

branches: 1.97.2;
regen

Regen for dup/dup2/dup3 argument types fix.

Regen for utimensat(2).

branches: 1.94.4;
regen for utimes

regen for ppoll (and, apparently, SYCALL_ARG_PTR)

branches: 1.92.10;
- add sigtimedwait support.
- merge the siginfo population code.

branches: 1.91.4;
Regenerate files to pick up correct RCS Ids. Suggested by Chris Badura.

Add fadvise64 and fadvise64_64 syscalls to compat_linux and compat_linux32.

Regen: Argh:
cat: rumphdr.types: No such file or directory
rm: rumphdr.types: No such file or directory
*** Error code 1

branches: 1.88.2;
regen

regen

branches: 1.86.2; 1.86.4;
Regen (new makesyscalls.sh)

Regen for personality(2) update.

Regen for rt_queueinfo(2) update.

branches: 1.83.2;
Regen for wait4 update.

Regen to prove I didn't screw up the conversion: purely RCSID changes.

merge christos-time_t

Regen.

Regen for futex timeout argument cleanup/fix.

Regen for nanosleep(2) fix.

branches: 1.77.2; 1.77.4;
regen

regen

branches: 1.75.2; 1.75.4; 1.75.8;
Regen.

branches: 1.74.2;
Regen for linux_sys_msync removal.

branches: 1.73.6; 1.73.8;
Regen.

Regen for linux_sys_getpgid removal.

Regen for getfsuid/setfsgid syscall exchange.

regen

branches: 1.69.2; 1.69.6;
Regen, for syscalls cleanup.

branches: 1.68.6; 1.68.8; 1.68.12; 1.68.14;
regen

regen.

branches: 1.66.2; 1.66.4;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.65.8;
Regen.

Regen

regen

branches: 1.62.4; 1.62.8; 1.62.14;
merge ktrace-lwp.

regen

Regen with correct RCS Ids.

mirror manu's amd64 changes to implement linux sched_{seg,get}affinity;
the kernel and linux lkm now compile again.

Regen.

branches: 1.57.2;
Regen.

Regen.

regen

branches: 1.54.4; 1.54.6;
regen:
implement support for Linux statfs64() syscall - 64bit variant of statfs()

regen: add exit_group(2)

regen - sync syscall lists with Linux kernel 2.6.7

regen:

connect madvise(2) and mincore(2) - apparently the newer Linux libs
don't stub it anymore, so allow the application to take advantage
of them

the Linux calls appear to be compatible in the flag values and semantics,
so a wrapper is not necessary

connect madvise(2) and mincore(2) - apparently the newer Linux libs
don't stub it anymore, so allow the application to take advantage
of them

the Linux calls appear to be compatible in the flag values and semantics,
so a wrapper is not necessary

regen - added wrapper for Linux ftruncate64()

Regen.

branches: 1.47.2;
Regen.

Regen for correct RCS IDs.

Merge the nathanw_sa branch.

branches: 1.44.4;
regen

regen

regen

- add a real linux_sys___sysctl because the mib numbers are different.
implement only the kernel version getting entries.
- make the signal conversion functions consistent; dest is arg 1.
- make linux_fakedev deal with block and char devices differently.
- add linux_sys_ugetrlimit, and flesh out the regular get/setrlimit so
that they work properly.
- add linux_mmap2 [untested].
- bump kernel to 2.4.18, and make the date be valentine's day :-)
- linux_sys_*stat64 family was totally busted. Fix it.

tested only on i386.

Status:
gdb, telnet, work
jdk-1.4.0 extractor works, jdk still coredumps.

christos

add RCSIDs (including regeneration of files as appropriate)

Fixes my last commit: only mips really need fcntl64 to work now. i386 and
powerpc could use it later if needed. m68k and alpha should never need it.

branches: 1.38.2; 1.38.4;
use _KERNEL_OPT.

Improved the accuracy of gettimeofday/settimeofday: Linux stores a srtuct
timezone in the kernel. We now keep track of it for emulated processes.
Tested for powerpc and i386. Untested on m68k, but it should be ok
Alpha has a particular implementation (osf1_sys_gettimeofday), and it has
therefore not been touched.

regen

regen

branches: 1.34.2;
Regen; getpid(2) is MP-safe.

Regen; add sy_flags.

Regen.

Regen.

Regen.

Regen.

regen

regen

Regen.

regen

Regen.

regen after addition of linux_sys_swapoff()

regen

regen

regen

update for changed makesyscalls.master

update for new makesyscalls.sh, and also for correct ids for changed
syscalls.master files. (you need to regenerate after checking in
syscalls.master to get the correct ids.)

branches: 1.17.4;
Regen.

regen

Regen.

Regen.

Regen.

branches: 1.12.2; 1.12.8;
Regen from "syscalls.master".

Regen.

Regen.

Regen.

branches: 1.8.2; 1.8.4;
Regenerate (with poll())

Regen.

Regen; pread(2) and pwrite(2).

Regen

Regen

Assign my copyright to TNF. Merge with others were appropriate.
Regen syscall files after script change.

Attempt to fix the mess.

Regen.

regen

regen

regen

regen

branches: 1.127.2;
regen

regen

regen

regen

regen

regen

regen

Regen for preadv(2) / pwritev(2).

Regen for eventfd.

Regen for timerfd.

Regen for POSIX timer syscalls.

branches: 1.116.2;
Regen for native futex calls.

regen

Regen

branches: 1.113.8; 1.113.10;
regen

branches: 1.112.4;
regen all

branches: 1.111.2;
regen

Regen (something apparently went wrong in previous)

Regen

Regen for readlink/readlinkat signatures.

branches: 1.107.2;
regen

Regen

regen

branches: 1.104.2;
Regen for ppoll signature fix

branches: 1.103.2;
Regen for munmap(2) signature.

Regen for utimes(2) change.

Regen for getgroups/setgroups update.