Arrange to update cached LWP credentials in userret() rather than during
syscall/trap entry, eliminating a test+branch on every syscall/trap.

This wasn't possible in the 3.99.x timeframe when l->l_cred came about
because there wasn't a reliable/timely way to force an ONPROC LWP running on
a remote CPU into the kernel (which is just about the only new thing in
this scheme).

PR/57261: hppa should be converted to __HAVE_SYSCALL_INTERN

Provide syscall_intern and use the md_syscall in syscall trap handling.

Remove unnecessary inclusion of <net/netisr.h>.

Only report the SSBREAKPOINT break instruction as SIGTRAP/TRAP_TRACE. All
other break instructions will be reported as SIGTRAP/TRAP_BRKPT

This fixes a mistake I made back in 2008.

PR/56866: hppa: kernel gets confused between actual breakpoints and single-step breakpoints

Handle 'NA' (non-access) traps for the lpa and probe instructions. The
change is inspired by OpenBSD with a bunch of my own, mainly stylistic,
changes.

Thanks to Tom Lane for the analysis.

PR/56118: sporadic app crashes in HPPA -current

Whitespace

fix various typos in comments, mainly origional->original,
extened->extended, incomming->incoming.

s/ be be / be /

Fix KASSERT that has been incorrect since revision 1.85 from 10 years
ago.

branches: 1.114.6;
Simplify userret function signature. From ad@

PR port-sparc/54718 (sparc install hangs since recent scheduler changes)

- userret() must be called every time we return to user, it's not optional.
- If clearing the AST with interrupts off, you must loop over userret().

mi_userret(): take care of calling preempt(), set spc_curpriority directly,
and remove MD code that does the same.

branches: 1.111.4;
Trailing whitespace

Centralized shared part of child_return() into MI part

Add a new function md_child_return() for MD specific bits only.

New child_return() is now part of MI and central code that handles
uniformly tracing code (KTR and ptrace(2)).

Synchronize value passed to ktrsysret() among ports to SYS_fork. This is
a traditional value and accessing p_lflag to check for PL_PPWAIT shall
use locking against proc_lock. Returning SYS_fork vs SYS_vfork still isn't
correct enough as there are more entry points to forking code. Instead of
making it too good, just settle with plain SYS_fork for all ports.

Overhaul the API used to fetch and store individual memory cells in
userspace. The old fetch(9) and store(9) APIs (fubyte(), fuword(),
subyte(), suword(), etc.) are retired and replaced with new ufetch(9)
and ustore(9) APIs that can return proper error codes, etc. and are
implemented consistently across all platforms. The interrupt-safe
variants are no longer supported (and several of the existing attempts
at fuswintr(), etc. were buggy and not actually interrupt-safe).

Also augmement the ucas(9) API, making it consistently available on
all plaforms, supporting uniprocessor and multiprocessor systems, even
those that do not have CAS or LL/SC primitives.

Welcome to NetBSD 8.99.37.

Rework the fork(2)/vfork(2) event signalling under ptrace(2)

Remove the constraint of SIGTRAP event being maskable by a tracee.

Now all SIGTRAP TRAP_CHLD events are delivered to debugger.

This code touches MD specific logic and the child_return routine.
It's an intermediate step with a room for refactoring in future and
right now the least invasive approach. This allows to assert expected
behavior in already existing ATF tests and make the code prettier
in future keeping the same semantics. Probably there is a need for a MI
wrapper of child_return for shared functionality between ports.

branches: 1.107.10; 1.107.18;
Send SIGBUS when accessing mmap() past end of file (handle EINVAL in fault
path). While there, handle ENOMEM as well.

branches: 1.106.6;
Move to a flat space register convention. %sr[4-7] are all now the space
number allocated to the process. gcc produces (slightly) better code
with this convention.

Retain backwards compatiblity.

Welcome to 6.99.36

Simplify cpu_spawn_return - setregs has done all the work already.

Change a few __unused to __diagused/__debugused

Use __unused on a variable used in a KASSERT

Use sy_invoke

branches: 1.101.2; 1.101.4;
Pass the trap address in ksi_addr without privilege bits.

Previously GDB would get confused by the trap address given when hitting
the _rtld_debug_state breakpoint.

unifdef FPEMUL. All our supported systems have an FPU.

Deal with setting of space registers and protection ids for posix_spawn.

Thanks to martin for the help.

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

branches: 1.97.2;
Add missing MD code for posix_spawn

branches: 1.96.4; 1.96.8;
Make some more information (psw/fpu_state) per cpu.

branches: 1.95.2; 1.95.4;
Move counting of faults, traps, intrs, soft[intr]s, syscalls, and nswtch
from uvmexp to per-cpu cpu_data and move them to 64bits. Remove unneeded
includes of <uvm/uvm_extern.h> and/or <uvm/uvm.h>.

#if 0 out the trap checks around tf_iioq_{head,tail}. The sti(4) code
can/will trigger them.

fix hppa ucas_*():
- "ret" is a kernel-space pointer.
- move the ucas RAS check from the device interrupt path
to the page-fault path.

Fix previous.

Pull change across from OpenBSD:

Make single stepping a system call work. Instead of single stepping
through the syscall gateway page, which doesn't work since that page is
shared between process, this makes us step over that bit by setting a
breakpoint on the instruction where the system call returns.

SANITY output tweak.

KNF

Remove lwp_uc_pool, replace it with kmem(9), plus add some consistency.
As discussed, a while ago, with ad@.

Include SSBREAKPOINT in the PTRACE #ifdef/#endif.

Add md_astpending to struct mdlwp and use it.

Make cpl, ipending, and intr_depth per CPU values.

Correct the list of traps that we need to get the va and space from
ii[os]q.

From OpenBSD.

Wrap long line.

Whitespace / comment formatting.

Improve the output from frame_sanity_check.

fix copy{in,out}{,str}() to return the error returned by uvm_fault().
fixes PR 41813.

The FP regs are accessed in both physical and virtual modes. Make sure
they're allocated in directly mapped memory to avoid aliasing problems
and unnecessary cache flushing.

Fix various names and references. If we ever directly map the uarea and
pcb we can change things at that point.

Significantly helps my J210 (PA7200).

Mostly from OpenBSD.

branches: 1.78.2;
Start at the first, not the second, exception register.

Comment formating.

Oops. Whitespace in previous.

Remove a cast by fixing a printf format.

Reformat a comment.

Minor tweaks.

Use __arraycount

branches: 1.71.2;
Use pmap_extract instead of kvtop.

Tweak astpending handling.

Check cpl level has returned to entry value when leaving trap/syscall.

From OpenBSD.

Add a comment.

Rename type_raw to trapnum and use in a couple more places.

Reduces diff to OpenBSD.

Count traps in uvmexp. Reduces diff to OpenBSD.

Update 64bit arg handling to use the flags in the syscall table.

Only do trace_enter/trace_exit if we're tracing.

curlwp is never NULL now. Sprinkle KASSERT just in case.

Fix DEBUG build. Hi rmind.

- Use uvm_lwp_setuarea() instead of directly setting address to lwp_t::l_addr.
- Replace most remaining uses of l_addr with uvm_lwp_getuarea() or lwp_getpcb().
- Amend assembly in ports where it accesses PCB via struct user.
- Rename L_ADDR to L_PCB in few places. Reduce sys/user.h inclusions.

Use lwp_getpcb() on hppa and ia64, clean from struct user usage.

Follow upstream license changes for files with Michael Shalayeff's
copyright. In most cases, this means dropping the 3rd and 4th clauses.

Merge nick-hppapmap.

This is a port of the OpenBSD pmap and trap handling code to get us

- Performance boost on some/all machines.
- Well on the way to PA2.0 (in 32bit mode) support. Several
machines probe hardware, but fail sometime after interrupts
are enabled.

Other things changed / fixed on the branch are

- update autoconf to use the OpenBSD code.
- com @ dino is very close to being supported.
- HPPA_REDZONE has been replaced with a working redzone which
is enabled with DIAGNOSTIC.
- UPAGES has been halved in size.
- power(4) from OpenBSD to fix a few bugs.
- updated list of modules from OpenBSD.
- initial ports of uturn(4), astro(4) and elroy(4).
- update some copyrights (remove advertising clause, etc.)

Thanks to mjf for some of the above, testing and listening.

Fixup two more syscalls with 64-bit args.

Remove reference to script that is supposed to do this automatically - I
don't have it.

Fixes PR/41185.

branches: 1.57.2; 1.57.8;
Provide a small inline wrapper for sysent::sy_call() and use it to store
the active syscall description in struct lwp. To be used at some future
point to prevent unloading of modules that provide syscalls, while the
syscalls are in use.

Merge wrstuden-revivesa into HEAD.

branches: 1.55.2; 1.55.6;
Remove clause 3 and 4 from TNF licenses

branches: 1.54.6; 1.54.8; 1.54.10;
Remove the 'args' parameter to 'trace_exit()' it is no longer used.
Instead of passing the (un)real system call code and syscall table pointer,
just pass the number of arguments - which is what ktrace really wants.
Ride forthcoming 4.99.53

KNF

Correct a comment.

Various improvements to the hppa kernel for gdb.

- add the useful space and control registers to struct reg
- update process_{read,write}_regs accordingly
- use MID_HPPA in core dumps.
- dump the fp regs into cores.
- add single stepping support from OpenBSD.

Don't pass 'curlwp' into trace_enter() and trace_exit().

branches: 1.49.2; 1.49.8;
Don't set l_usrpri / spc_curpriority here. mi_userret() does it.

Merge the ppcoea-renovation branch to HEAD.

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

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

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

branches: 1.47.2;
Make want_resched a member of cpu_info.

branches: 1.46.2; 1.46.4;
Changes to make ktrace LKM friendly and reduce ifdef KTRACE. Proposed
on tech-kern.

branches: 1.45.4; 1.45.8;
Fix up the frame sanity checking to not bitch when taking interrupts in
the sti ROM code.

Fixes PR/34715.

Thanks to David H. Gutteridge.

Minor KNF and make some output a bit easier to read.

merge yamt-idlelwp branch. asked by core@. some ports still needs work.

from doc/BRANCHES:

idle lwp, and some changes depending on it.

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

branches: 1.42.2; 1.42.4; 1.42.10;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

Remove hppa_{round,trunc}_page.

branches: 1.40.2;
Merge newlock2 to head.

Update uvm_grow to support stacks that grow upwards.

Use on hppa and fix a bug in the hppa trap handler.

branches: 1.38.2;
Make this compile without KTRACE.

From Sergey Svishchev in a private email.

branches: 1.37.2; 1.37.4;
Fix a spello in a debug message.

- Hold a reference to the process credentials in each struct lwp.
- Update the reference on syscall and user trap if p_cred has changed.
- Collect accounting flags in the LWP, and collate on LWP exit.

Fix use of user_backtrace. It takes a lwp pointer now.

branches: 1.34.4;
P_PROFIL is held in p_flag of struct proc so test for it there.

branches: 1.33.2;
adapt to uvm_fault() interface cleanup: kill the useless 3rd argument

branches: 1.32.2;
Syscall debug tracing is handled by trace_enter() / trace_exit(). Change
trace_is_enabled() to return TRUE if SYSCALL_DEBUG is defined, and g/c
all of the SYSCALL_DEBUG handling from individual system call dispatch
routines.

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: 1.30.4; 1.30.6; 1.30.8;
Remove leading __ from __(const|inline|signed|volatile) -- it is obsolete.

merge ktrace-lwp.

branches: 1.28.2;
Remove "bad" label no longer in use.

PR/29607: Christian Biere systrace doesn't handle interrupted syscalls properly
Instead of jumping to the default "bad" case, jump to the error handling
switch, so that we can deal with ERESTART/EJUSTRETURN properly.

add additional args to frame_sanity_check() so we can log which
context it was called from and the trap type (if called from trap()).

check pcb_onfault for kernel T_DATALIGN traps (for ddb).

user_backtrace() is used if DEBUG or USERTRACE is defined, but not DIAGNOSTIC.

branches: 1.23.4; 1.23.6;
handle T_CONDITION traps (by sending SIGFPE).
fix T_EXCEPTION code to handle the other FP major opcodes.

real siginfo support.

FPU fixes and improvements, adapted from openbsd:
- handle exceptions correctly
- emulation of unimplemented instructions (eg. on PA-7300LC)

use mi_userret().

clear pcb_onfault before calling uvm_fault() and restore it after.
remove some debug code that actually makes debugging harder.

redo rev. 1.13, handle curlwp being NULL correctly.

for instruction TLB miss traps, process the fault with VM_PROT_EXECUTE
instead of VM_PROT_READ (to support non-executable mappings).

To the rest of the kernel, pretend that hppa supports SIGINFO.
There are still only "sigcontext" stackframes passed to the user,
but at least we can clean up the MI sources now.

add kernel part of concurrency support for SA on MP systems
- move per VP data into struct sadata_vp referenced from l->l_savp
* VP id
* lock on VP data
* LWP on VP
* recently blocked LWP on VP
* queue of LWPs woken which ran on this VP before sleep
* faultaddr
* LWP cache for upcalls
* upcall queue
- add current concurrency and requested concurrency variables
- make process exit run LWP on all VPs
- make signal delivery consider all VPs
- make timer events consider all VPs
- add sa_newsavp to allocate new sadata_vp structure
- add sa_increaseconcurrency to prepare new VP
- make sys_sa_setconcurrency request new VP or wakeup idle VP
- make sa_yield lower current concurrency
- set sa_cpu = VP id in upcalls
- maintain cached LWPs per VP

de-__P, remove register, ansify.

In trap(), curlwp could be NULL so don't refer it implicitly.

Don't pass the (unused) return value args to the
trace_enter()/systrace_enter() functions.

NULL -> 0 for preempt() argument.

add MD part of SA/pthread pagefault handling on hppa

update for LWPs, and some lite cleanup.

__KERNEL_RCSID()

branches: 1.7.2;
Use PAGE_SIZE rather than NBPG.

Pass the system call table to trace_enter() and ktrsys() so that it is
possible to use alternate system call tables. This is usefull for
displaying correctly the arguments in Mach binaries traces.

If NULL is given, then the regular systam call table for the process is used.

Add a realcode argument to trace_enter and ktrsyscall. realcode is the
original system call number, which can be negative for a Mach trap.
We cannot just replace code by realcode, because ktrsyscall uses it as
an index in the system call table, thus crashing the kernel when the
value is negative.

Since trap_kdebug() may be a macro, be sure to cast to void when not
using its result.

branches: 1.3.2;
Now include the HPPA trap number in a user backtrace, and display
backtraces for more traps.

branches: 1.2.2;
MD systrace gluons.

Added files to support generic HP PA-RISC based machines. hp700-specific
files to follow.

PR/57261: hppa should be converted to __HAVE_SYSCALL_INTERN

Provide syscall_intern and use the md_syscall in syscall trap handling.

Remove unnecessary inclusion of <net/netisr.h>.

Only report the SSBREAKPOINT break instruction as SIGTRAP/TRAP_TRACE. All
other break instructions will be reported as SIGTRAP/TRAP_BRKPT

This fixes a mistake I made back in 2008.

PR/56866: hppa: kernel gets confused between actual breakpoints and single-step breakpoints

Handle 'NA' (non-access) traps for the lpa and probe instructions. The
change is inspired by OpenBSD with a bunch of my own, mainly stylistic,
changes.

Thanks to Tom Lane for the analysis.

PR/56118: sporadic app crashes in HPPA -current

Whitespace

fix various typos in comments, mainly origional->original,
extened->extended, incomming->incoming.

s/ be be / be /

Fix KASSERT that has been incorrect since revision 1.85 from 10 years
ago.

branches: 1.114.6;
Simplify userret function signature. From ad@

PR port-sparc/54718 (sparc install hangs since recent scheduler changes)

- userret() must be called every time we return to user, it's not optional.
- If clearing the AST with interrupts off, you must loop over userret().

mi_userret(): take care of calling preempt(), set spc_curpriority directly,
and remove MD code that does the same.

branches: 1.111.4;
Trailing whitespace

Centralized shared part of child_return() into MI part

Add a new function md_child_return() for MD specific bits only.

New child_return() is now part of MI and central code that handles
uniformly tracing code (KTR and ptrace(2)).

Synchronize value passed to ktrsysret() among ports to SYS_fork. This is
a traditional value and accessing p_lflag to check for PL_PPWAIT shall
use locking against proc_lock. Returning SYS_fork vs SYS_vfork still isn't
correct enough as there are more entry points to forking code. Instead of
making it too good, just settle with plain SYS_fork for all ports.

Overhaul the API used to fetch and store individual memory cells in
userspace. The old fetch(9) and store(9) APIs (fubyte(), fuword(),
subyte(), suword(), etc.) are retired and replaced with new ufetch(9)
and ustore(9) APIs that can return proper error codes, etc. and are
implemented consistently across all platforms. The interrupt-safe
variants are no longer supported (and several of the existing attempts
at fuswintr(), etc. were buggy and not actually interrupt-safe).

Also augmement the ucas(9) API, making it consistently available on
all plaforms, supporting uniprocessor and multiprocessor systems, even
those that do not have CAS or LL/SC primitives.

Welcome to NetBSD 8.99.37.

Rework the fork(2)/vfork(2) event signalling under ptrace(2)

Remove the constraint of SIGTRAP event being maskable by a tracee.

Now all SIGTRAP TRAP_CHLD events are delivered to debugger.

This code touches MD specific logic and the child_return routine.
It's an intermediate step with a room for refactoring in future and
right now the least invasive approach. This allows to assert expected
behavior in already existing ATF tests and make the code prettier
in future keeping the same semantics. Probably there is a need for a MI
wrapper of child_return for shared functionality between ports.

branches: 1.107.10; 1.107.18;
Send SIGBUS when accessing mmap() past end of file (handle EINVAL in fault
path). While there, handle ENOMEM as well.

branches: 1.106.6;
Move to a flat space register convention. %sr[4-7] are all now the space
number allocated to the process. gcc produces (slightly) better code
with this convention.

Retain backwards compatiblity.

Welcome to 6.99.36

Simplify cpu_spawn_return - setregs has done all the work already.

Change a few __unused to __diagused/__debugused

Use __unused on a variable used in a KASSERT

Use sy_invoke

branches: 1.101.2; 1.101.4;
Pass the trap address in ksi_addr without privilege bits.

Previously GDB would get confused by the trap address given when hitting
the _rtld_debug_state breakpoint.

unifdef FPEMUL. All our supported systems have an FPU.

Deal with setting of space registers and protection ids for posix_spawn.

Thanks to martin for the help.

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

branches: 1.97.2;
Add missing MD code for posix_spawn

branches: 1.96.4; 1.96.8;
Make some more information (psw/fpu_state) per cpu.

branches: 1.95.2; 1.95.4;
Move counting of faults, traps, intrs, soft[intr]s, syscalls, and nswtch
from uvmexp to per-cpu cpu_data and move them to 64bits. Remove unneeded
includes of <uvm/uvm_extern.h> and/or <uvm/uvm.h>.

#if 0 out the trap checks around tf_iioq_{head,tail}. The sti(4) code
can/will trigger them.

fix hppa ucas_*():
- "ret" is a kernel-space pointer.
- move the ucas RAS check from the device interrupt path
to the page-fault path.

Fix previous.

Pull change across from OpenBSD:

Make single stepping a system call work. Instead of single stepping
through the syscall gateway page, which doesn't work since that page is
shared between process, this makes us step over that bit by setting a
breakpoint on the instruction where the system call returns.

SANITY output tweak.

KNF

Remove lwp_uc_pool, replace it with kmem(9), plus add some consistency.
As discussed, a while ago, with ad@.

Include SSBREAKPOINT in the PTRACE #ifdef/#endif.

Add md_astpending to struct mdlwp and use it.

Make cpl, ipending, and intr_depth per CPU values.

Correct the list of traps that we need to get the va and space from
ii[os]q.

From OpenBSD.

Wrap long line.

Whitespace / comment formatting.

Improve the output from frame_sanity_check.

fix copy{in,out}{,str}() to return the error returned by uvm_fault().
fixes PR 41813.

The FP regs are accessed in both physical and virtual modes. Make sure
they're allocated in directly mapped memory to avoid aliasing problems
and unnecessary cache flushing.

Fix various names and references. If we ever directly map the uarea and
pcb we can change things at that point.

Significantly helps my J210 (PA7200).

Mostly from OpenBSD.

branches: 1.78.2;
Start at the first, not the second, exception register.

Comment formating.

Oops. Whitespace in previous.

Remove a cast by fixing a printf format.

Reformat a comment.

Minor tweaks.

Use __arraycount

branches: 1.71.2;
Use pmap_extract instead of kvtop.

Tweak astpending handling.

Check cpl level has returned to entry value when leaving trap/syscall.

From OpenBSD.

Add a comment.

Rename type_raw to trapnum and use in a couple more places.

Reduces diff to OpenBSD.

Count traps in uvmexp. Reduces diff to OpenBSD.

Update 64bit arg handling to use the flags in the syscall table.

Only do trace_enter/trace_exit if we're tracing.

curlwp is never NULL now. Sprinkle KASSERT just in case.

Fix DEBUG build. Hi rmind.

- Use uvm_lwp_setuarea() instead of directly setting address to lwp_t::l_addr.
- Replace most remaining uses of l_addr with uvm_lwp_getuarea() or lwp_getpcb().
- Amend assembly in ports where it accesses PCB via struct user.
- Rename L_ADDR to L_PCB in few places. Reduce sys/user.h inclusions.

Use lwp_getpcb() on hppa and ia64, clean from struct user usage.

Follow upstream license changes for files with Michael Shalayeff's
copyright. In most cases, this means dropping the 3rd and 4th clauses.

Merge nick-hppapmap.

This is a port of the OpenBSD pmap and trap handling code to get us

- Performance boost on some/all machines.
- Well on the way to PA2.0 (in 32bit mode) support. Several
machines probe hardware, but fail sometime after interrupts
are enabled.

Other things changed / fixed on the branch are

- update autoconf to use the OpenBSD code.
- com @ dino is very close to being supported.
- HPPA_REDZONE has been replaced with a working redzone which
is enabled with DIAGNOSTIC.
- UPAGES has been halved in size.
- power(4) from OpenBSD to fix a few bugs.
- updated list of modules from OpenBSD.
- initial ports of uturn(4), astro(4) and elroy(4).
- update some copyrights (remove advertising clause, etc.)

Thanks to mjf for some of the above, testing and listening.

Fixup two more syscalls with 64-bit args.

Remove reference to script that is supposed to do this automatically - I
don't have it.

Fixes PR/41185.

branches: 1.57.2; 1.57.8;
Provide a small inline wrapper for sysent::sy_call() and use it to store
the active syscall description in struct lwp. To be used at some future
point to prevent unloading of modules that provide syscalls, while the
syscalls are in use.

Merge wrstuden-revivesa into HEAD.

branches: 1.55.2; 1.55.6;
Remove clause 3 and 4 from TNF licenses

branches: 1.54.6; 1.54.8; 1.54.10;
Remove the 'args' parameter to 'trace_exit()' it is no longer used.
Instead of passing the (un)real system call code and syscall table pointer,
just pass the number of arguments - which is what ktrace really wants.
Ride forthcoming 4.99.53

KNF

Correct a comment.

Various improvements to the hppa kernel for gdb.

- add the useful space and control registers to struct reg
- update process_{read,write}_regs accordingly
- use MID_HPPA in core dumps.
- dump the fp regs into cores.
- add single stepping support from OpenBSD.

Don't pass 'curlwp' into trace_enter() and trace_exit().

branches: 1.49.2; 1.49.8;
Don't set l_usrpri / spc_curpriority here. mi_userret() does it.

Merge the ppcoea-renovation branch to HEAD.

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

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

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

branches: 1.47.2;
Make want_resched a member of cpu_info.

branches: 1.46.2; 1.46.4;
Changes to make ktrace LKM friendly and reduce ifdef KTRACE. Proposed
on tech-kern.

branches: 1.45.4; 1.45.8;
Fix up the frame sanity checking to not bitch when taking interrupts in
the sti ROM code.

Fixes PR/34715.

Thanks to David H. Gutteridge.

Minor KNF and make some output a bit easier to read.

merge yamt-idlelwp branch. asked by core@. some ports still needs work.

from doc/BRANCHES:

idle lwp, and some changes depending on it.

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

branches: 1.42.2; 1.42.4; 1.42.10;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

Remove hppa_{round,trunc}_page.

branches: 1.40.2;
Merge newlock2 to head.

Update uvm_grow to support stacks that grow upwards.

Use on hppa and fix a bug in the hppa trap handler.

branches: 1.38.2;
Make this compile without KTRACE.

From Sergey Svishchev in a private email.

branches: 1.37.2; 1.37.4;
Fix a spello in a debug message.

- Hold a reference to the process credentials in each struct lwp.
- Update the reference on syscall and user trap if p_cred has changed.
- Collect accounting flags in the LWP, and collate on LWP exit.

Fix use of user_backtrace. It takes a lwp pointer now.

branches: 1.34.4;
P_PROFIL is held in p_flag of struct proc so test for it there.

branches: 1.33.2;
adapt to uvm_fault() interface cleanup: kill the useless 3rd argument

branches: 1.32.2;
Syscall debug tracing is handled by trace_enter() / trace_exit(). Change
trace_is_enabled() to return TRUE if SYSCALL_DEBUG is defined, and g/c
all of the SYSCALL_DEBUG handling from individual system call dispatch
routines.

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: 1.30.4; 1.30.6; 1.30.8;
Remove leading __ from __(const|inline|signed|volatile) -- it is obsolete.

merge ktrace-lwp.

branches: 1.28.2;
Remove "bad" label no longer in use.

PR/29607: Christian Biere systrace doesn't handle interrupted syscalls properly
Instead of jumping to the default "bad" case, jump to the error handling
switch, so that we can deal with ERESTART/EJUSTRETURN properly.

add additional args to frame_sanity_check() so we can log which
context it was called from and the trap type (if called from trap()).

check pcb_onfault for kernel T_DATALIGN traps (for ddb).

user_backtrace() is used if DEBUG or USERTRACE is defined, but not DIAGNOSTIC.

branches: 1.23.4; 1.23.6;
handle T_CONDITION traps (by sending SIGFPE).
fix T_EXCEPTION code to handle the other FP major opcodes.

real siginfo support.

FPU fixes and improvements, adapted from openbsd:
- handle exceptions correctly
- emulation of unimplemented instructions (eg. on PA-7300LC)

use mi_userret().

clear pcb_onfault before calling uvm_fault() and restore it after.
remove some debug code that actually makes debugging harder.

redo rev. 1.13, handle curlwp being NULL correctly.

for instruction TLB miss traps, process the fault with VM_PROT_EXECUTE
instead of VM_PROT_READ (to support non-executable mappings).

To the rest of the kernel, pretend that hppa supports SIGINFO.
There are still only "sigcontext" stackframes passed to the user,
but at least we can clean up the MI sources now.

add kernel part of concurrency support for SA on MP systems
- move per VP data into struct sadata_vp referenced from l->l_savp
* VP id
* lock on VP data
* LWP on VP
* recently blocked LWP on VP
* queue of LWPs woken which ran on this VP before sleep
* faultaddr
* LWP cache for upcalls
* upcall queue
- add current concurrency and requested concurrency variables
- make process exit run LWP on all VPs
- make signal delivery consider all VPs
- make timer events consider all VPs
- add sa_newsavp to allocate new sadata_vp structure
- add sa_increaseconcurrency to prepare new VP
- make sys_sa_setconcurrency request new VP or wakeup idle VP
- make sa_yield lower current concurrency
- set sa_cpu = VP id in upcalls
- maintain cached LWPs per VP

de-__P, remove register, ansify.

In trap(), curlwp could be NULL so don't refer it implicitly.

Don't pass the (unused) return value args to the
trace_enter()/systrace_enter() functions.

NULL -> 0 for preempt() argument.

add MD part of SA/pthread pagefault handling on hppa

update for LWPs, and some lite cleanup.

__KERNEL_RCSID()

branches: 1.7.2;
Use PAGE_SIZE rather than NBPG.

Pass the system call table to trace_enter() and ktrsys() so that it is
possible to use alternate system call tables. This is usefull for
displaying correctly the arguments in Mach binaries traces.

If NULL is given, then the regular systam call table for the process is used.

Add a realcode argument to trace_enter and ktrsyscall. realcode is the
original system call number, which can be negative for a Mach trap.
We cannot just replace code by realcode, because ktrsyscall uses it as
an index in the system call table, thus crashing the kernel when the
value is negative.

Since trap_kdebug() may be a macro, be sure to cast to void when not
using its result.

branches: 1.3.2;
Now include the HPPA trap number in a user backtrace, and display
backtraces for more traps.

branches: 1.2.2;
MD systrace gluons.

Added files to support generic HP PA-RISC based machines. hp700-specific
files to follow.

Remove unnecessary inclusion of <net/netisr.h>.

Only report the SSBREAKPOINT break instruction as SIGTRAP/TRAP_TRACE. All
other break instructions will be reported as SIGTRAP/TRAP_BRKPT

This fixes a mistake I made back in 2008.

PR/56866: hppa: kernel gets confused between actual breakpoints and single-step breakpoints

Handle 'NA' (non-access) traps for the lpa and probe instructions. The
change is inspired by OpenBSD with a bunch of my own, mainly stylistic,
changes.

Thanks to Tom Lane for the analysis.

PR/56118: sporadic app crashes in HPPA -current

Whitespace

fix various typos in comments, mainly origional->original,
extened->extended, incomming->incoming.

s/ be be / be /

Fix KASSERT that has been incorrect since revision 1.85 from 10 years
ago.

branches: 1.114.6;
Simplify userret function signature. From ad@

PR port-sparc/54718 (sparc install hangs since recent scheduler changes)

- userret() must be called every time we return to user, it's not optional.
- If clearing the AST with interrupts off, you must loop over userret().

mi_userret(): take care of calling preempt(), set spc_curpriority directly,
and remove MD code that does the same.

branches: 1.111.4;
Trailing whitespace

Centralized shared part of child_return() into MI part

Add a new function md_child_return() for MD specific bits only.

New child_return() is now part of MI and central code that handles
uniformly tracing code (KTR and ptrace(2)).

Synchronize value passed to ktrsysret() among ports to SYS_fork. This is
a traditional value and accessing p_lflag to check for PL_PPWAIT shall
use locking against proc_lock. Returning SYS_fork vs SYS_vfork still isn't
correct enough as there are more entry points to forking code. Instead of
making it too good, just settle with plain SYS_fork for all ports.

Overhaul the API used to fetch and store individual memory cells in
userspace. The old fetch(9) and store(9) APIs (fubyte(), fuword(),
subyte(), suword(), etc.) are retired and replaced with new ufetch(9)
and ustore(9) APIs that can return proper error codes, etc. and are
implemented consistently across all platforms. The interrupt-safe
variants are no longer supported (and several of the existing attempts
at fuswintr(), etc. were buggy and not actually interrupt-safe).

Also augmement the ucas(9) API, making it consistently available on
all plaforms, supporting uniprocessor and multiprocessor systems, even
those that do not have CAS or LL/SC primitives.

Welcome to NetBSD 8.99.37.

Rework the fork(2)/vfork(2) event signalling under ptrace(2)

Remove the constraint of SIGTRAP event being maskable by a tracee.

Now all SIGTRAP TRAP_CHLD events are delivered to debugger.

This code touches MD specific logic and the child_return routine.
It's an intermediate step with a room for refactoring in future and
right now the least invasive approach. This allows to assert expected
behavior in already existing ATF tests and make the code prettier
in future keeping the same semantics. Probably there is a need for a MI
wrapper of child_return for shared functionality between ports.

branches: 1.107.10; 1.107.18;
Send SIGBUS when accessing mmap() past end of file (handle EINVAL in fault
path). While there, handle ENOMEM as well.

branches: 1.106.6;
Move to a flat space register convention. %sr[4-7] are all now the space
number allocated to the process. gcc produces (slightly) better code
with this convention.

Retain backwards compatiblity.

Welcome to 6.99.36

Simplify cpu_spawn_return - setregs has done all the work already.

Change a few __unused to __diagused/__debugused

Use __unused on a variable used in a KASSERT

Use sy_invoke

branches: 1.101.2; 1.101.4;
Pass the trap address in ksi_addr without privilege bits.

Previously GDB would get confused by the trap address given when hitting
the _rtld_debug_state breakpoint.

unifdef FPEMUL. All our supported systems have an FPU.

Deal with setting of space registers and protection ids for posix_spawn.

Thanks to martin for the help.

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

branches: 1.97.2;
Add missing MD code for posix_spawn

branches: 1.96.4; 1.96.8;
Make some more information (psw/fpu_state) per cpu.

branches: 1.95.2; 1.95.4;
Move counting of faults, traps, intrs, soft[intr]s, syscalls, and nswtch
from uvmexp to per-cpu cpu_data and move them to 64bits. Remove unneeded
includes of <uvm/uvm_extern.h> and/or <uvm/uvm.h>.

#if 0 out the trap checks around tf_iioq_{head,tail}. The sti(4) code
can/will trigger them.

fix hppa ucas_*():
- "ret" is a kernel-space pointer.
- move the ucas RAS check from the device interrupt path
to the page-fault path.

Fix previous.

Pull change across from OpenBSD:

Make single stepping a system call work. Instead of single stepping
through the syscall gateway page, which doesn't work since that page is
shared between process, this makes us step over that bit by setting a
breakpoint on the instruction where the system call returns.

SANITY output tweak.

KNF

Remove lwp_uc_pool, replace it with kmem(9), plus add some consistency.
As discussed, a while ago, with ad@.

Include SSBREAKPOINT in the PTRACE #ifdef/#endif.

Add md_astpending to struct mdlwp and use it.

Make cpl, ipending, and intr_depth per CPU values.

Correct the list of traps that we need to get the va and space from
ii[os]q.

From OpenBSD.

Wrap long line.

Whitespace / comment formatting.

Improve the output from frame_sanity_check.

fix copy{in,out}{,str}() to return the error returned by uvm_fault().
fixes PR 41813.

The FP regs are accessed in both physical and virtual modes. Make sure
they're allocated in directly mapped memory to avoid aliasing problems
and unnecessary cache flushing.

Fix various names and references. If we ever directly map the uarea and
pcb we can change things at that point.

Significantly helps my J210 (PA7200).

Mostly from OpenBSD.

branches: 1.78.2;
Start at the first, not the second, exception register.

Comment formating.

Oops. Whitespace in previous.

Remove a cast by fixing a printf format.

Reformat a comment.

Minor tweaks.

Use __arraycount

branches: 1.71.2;
Use pmap_extract instead of kvtop.

Tweak astpending handling.

Check cpl level has returned to entry value when leaving trap/syscall.

From OpenBSD.

Add a comment.

Rename type_raw to trapnum and use in a couple more places.

Reduces diff to OpenBSD.

Count traps in uvmexp. Reduces diff to OpenBSD.

Update 64bit arg handling to use the flags in the syscall table.

Only do trace_enter/trace_exit if we're tracing.

curlwp is never NULL now. Sprinkle KASSERT just in case.

Fix DEBUG build. Hi rmind.

- Use uvm_lwp_setuarea() instead of directly setting address to lwp_t::l_addr.
- Replace most remaining uses of l_addr with uvm_lwp_getuarea() or lwp_getpcb().
- Amend assembly in ports where it accesses PCB via struct user.
- Rename L_ADDR to L_PCB in few places. Reduce sys/user.h inclusions.

Use lwp_getpcb() on hppa and ia64, clean from struct user usage.

Follow upstream license changes for files with Michael Shalayeff's
copyright. In most cases, this means dropping the 3rd and 4th clauses.

Merge nick-hppapmap.

This is a port of the OpenBSD pmap and trap handling code to get us

- Performance boost on some/all machines.
- Well on the way to PA2.0 (in 32bit mode) support. Several
machines probe hardware, but fail sometime after interrupts
are enabled.

Other things changed / fixed on the branch are

- update autoconf to use the OpenBSD code.
- com @ dino is very close to being supported.
- HPPA_REDZONE has been replaced with a working redzone which
is enabled with DIAGNOSTIC.
- UPAGES has been halved in size.
- power(4) from OpenBSD to fix a few bugs.
- updated list of modules from OpenBSD.
- initial ports of uturn(4), astro(4) and elroy(4).
- update some copyrights (remove advertising clause, etc.)

Thanks to mjf for some of the above, testing and listening.

Fixup two more syscalls with 64-bit args.

Remove reference to script that is supposed to do this automatically - I
don't have it.

Fixes PR/41185.

branches: 1.57.2; 1.57.8;
Provide a small inline wrapper for sysent::sy_call() and use it to store
the active syscall description in struct lwp. To be used at some future
point to prevent unloading of modules that provide syscalls, while the
syscalls are in use.

Merge wrstuden-revivesa into HEAD.

branches: 1.55.2; 1.55.6;
Remove clause 3 and 4 from TNF licenses

branches: 1.54.6; 1.54.8; 1.54.10;
Remove the 'args' parameter to 'trace_exit()' it is no longer used.
Instead of passing the (un)real system call code and syscall table pointer,
just pass the number of arguments - which is what ktrace really wants.
Ride forthcoming 4.99.53

KNF

Correct a comment.

Various improvements to the hppa kernel for gdb.

- add the useful space and control registers to struct reg
- update process_{read,write}_regs accordingly
- use MID_HPPA in core dumps.
- dump the fp regs into cores.
- add single stepping support from OpenBSD.

Don't pass 'curlwp' into trace_enter() and trace_exit().

branches: 1.49.2; 1.49.8;
Don't set l_usrpri / spc_curpriority here. mi_userret() does it.

Merge the ppcoea-renovation branch to HEAD.

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

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

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

branches: 1.47.2;
Make want_resched a member of cpu_info.

branches: 1.46.2; 1.46.4;
Changes to make ktrace LKM friendly and reduce ifdef KTRACE. Proposed
on tech-kern.

branches: 1.45.4; 1.45.8;
Fix up the frame sanity checking to not bitch when taking interrupts in
the sti ROM code.

Fixes PR/34715.

Thanks to David H. Gutteridge.

Minor KNF and make some output a bit easier to read.

merge yamt-idlelwp branch. asked by core@. some ports still needs work.

from doc/BRANCHES:

idle lwp, and some changes depending on it.

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

branches: 1.42.2; 1.42.4; 1.42.10;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

Remove hppa_{round,trunc}_page.

branches: 1.40.2;
Merge newlock2 to head.

Update uvm_grow to support stacks that grow upwards.

Use on hppa and fix a bug in the hppa trap handler.

branches: 1.38.2;
Make this compile without KTRACE.

From Sergey Svishchev in a private email.

branches: 1.37.2; 1.37.4;
Fix a spello in a debug message.

- Hold a reference to the process credentials in each struct lwp.
- Update the reference on syscall and user trap if p_cred has changed.
- Collect accounting flags in the LWP, and collate on LWP exit.

Fix use of user_backtrace. It takes a lwp pointer now.

branches: 1.34.4;
P_PROFIL is held in p_flag of struct proc so test for it there.

branches: 1.33.2;
adapt to uvm_fault() interface cleanup: kill the useless 3rd argument

branches: 1.32.2;
Syscall debug tracing is handled by trace_enter() / trace_exit(). Change
trace_is_enabled() to return TRUE if SYSCALL_DEBUG is defined, and g/c
all of the SYSCALL_DEBUG handling from individual system call dispatch
routines.

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: 1.30.4; 1.30.6; 1.30.8;
Remove leading __ from __(const|inline|signed|volatile) -- it is obsolete.

merge ktrace-lwp.

branches: 1.28.2;
Remove "bad" label no longer in use.

PR/29607: Christian Biere systrace doesn't handle interrupted syscalls properly
Instead of jumping to the default "bad" case, jump to the error handling
switch, so that we can deal with ERESTART/EJUSTRETURN properly.

add additional args to frame_sanity_check() so we can log which
context it was called from and the trap type (if called from trap()).

check pcb_onfault for kernel T_DATALIGN traps (for ddb).

user_backtrace() is used if DEBUG or USERTRACE is defined, but not DIAGNOSTIC.

branches: 1.23.4; 1.23.6;
handle T_CONDITION traps (by sending SIGFPE).
fix T_EXCEPTION code to handle the other FP major opcodes.

real siginfo support.

FPU fixes and improvements, adapted from openbsd:
- handle exceptions correctly
- emulation of unimplemented instructions (eg. on PA-7300LC)

use mi_userret().

clear pcb_onfault before calling uvm_fault() and restore it after.
remove some debug code that actually makes debugging harder.

redo rev. 1.13, handle curlwp being NULL correctly.

for instruction TLB miss traps, process the fault with VM_PROT_EXECUTE
instead of VM_PROT_READ (to support non-executable mappings).

To the rest of the kernel, pretend that hppa supports SIGINFO.
There are still only "sigcontext" stackframes passed to the user,
but at least we can clean up the MI sources now.

add kernel part of concurrency support for SA on MP systems
- move per VP data into struct sadata_vp referenced from l->l_savp
* VP id
* lock on VP data
* LWP on VP
* recently blocked LWP on VP
* queue of LWPs woken which ran on this VP before sleep
* faultaddr
* LWP cache for upcalls
* upcall queue
- add current concurrency and requested concurrency variables
- make process exit run LWP on all VPs
- make signal delivery consider all VPs
- make timer events consider all VPs
- add sa_newsavp to allocate new sadata_vp structure
- add sa_increaseconcurrency to prepare new VP
- make sys_sa_setconcurrency request new VP or wakeup idle VP
- make sa_yield lower current concurrency
- set sa_cpu = VP id in upcalls
- maintain cached LWPs per VP

de-__P, remove register, ansify.

In trap(), curlwp could be NULL so don't refer it implicitly.

Don't pass the (unused) return value args to the
trace_enter()/systrace_enter() functions.

NULL -> 0 for preempt() argument.

add MD part of SA/pthread pagefault handling on hppa

update for LWPs, and some lite cleanup.

__KERNEL_RCSID()

branches: 1.7.2;
Use PAGE_SIZE rather than NBPG.

Pass the system call table to trace_enter() and ktrsys() so that it is
possible to use alternate system call tables. This is usefull for
displaying correctly the arguments in Mach binaries traces.

If NULL is given, then the regular systam call table for the process is used.

Add a realcode argument to trace_enter and ktrsyscall. realcode is the
original system call number, which can be negative for a Mach trap.
We cannot just replace code by realcode, because ktrsyscall uses it as
an index in the system call table, thus crashing the kernel when the
value is negative.

Since trap_kdebug() may be a macro, be sure to cast to void when not
using its result.

branches: 1.3.2;
Now include the HPPA trap number in a user backtrace, and display
backtraces for more traps.

branches: 1.2.2;
MD systrace gluons.

Added files to support generic HP PA-RISC based machines. hp700-specific
files to follow.

Only report the SSBREAKPOINT break instruction as SIGTRAP/TRAP_TRACE. All
other break instructions will be reported as SIGTRAP/TRAP_BRKPT

This fixes a mistake I made back in 2008.

PR/56866: hppa: kernel gets confused between actual breakpoints and single-step breakpoints

Handle 'NA' (non-access) traps for the lpa and probe instructions. The
change is inspired by OpenBSD with a bunch of my own, mainly stylistic,
changes.

Thanks to Tom Lane for the analysis.

PR/56118: sporadic app crashes in HPPA -current

Whitespace

fix various typos in comments, mainly origional->original,
extened->extended, incomming->incoming.

s/ be be / be /

Fix KASSERT that has been incorrect since revision 1.85 from 10 years
ago.

branches: 1.114.6;
Simplify userret function signature. From ad@

PR port-sparc/54718 (sparc install hangs since recent scheduler changes)

- userret() must be called every time we return to user, it's not optional.
- If clearing the AST with interrupts off, you must loop over userret().

mi_userret(): take care of calling preempt(), set spc_curpriority directly,
and remove MD code that does the same.

branches: 1.111.4;
Trailing whitespace

Centralized shared part of child_return() into MI part

Add a new function md_child_return() for MD specific bits only.

New child_return() is now part of MI and central code that handles
uniformly tracing code (KTR and ptrace(2)).

Synchronize value passed to ktrsysret() among ports to SYS_fork. This is
a traditional value and accessing p_lflag to check for PL_PPWAIT shall
use locking against proc_lock. Returning SYS_fork vs SYS_vfork still isn't
correct enough as there are more entry points to forking code. Instead of
making it too good, just settle with plain SYS_fork for all ports.

Overhaul the API used to fetch and store individual memory cells in
userspace. The old fetch(9) and store(9) APIs (fubyte(), fuword(),
subyte(), suword(), etc.) are retired and replaced with new ufetch(9)
and ustore(9) APIs that can return proper error codes, etc. and are
implemented consistently across all platforms. The interrupt-safe
variants are no longer supported (and several of the existing attempts
at fuswintr(), etc. were buggy and not actually interrupt-safe).

Also augmement the ucas(9) API, making it consistently available on
all plaforms, supporting uniprocessor and multiprocessor systems, even
those that do not have CAS or LL/SC primitives.

Welcome to NetBSD 8.99.37.

Rework the fork(2)/vfork(2) event signalling under ptrace(2)

Remove the constraint of SIGTRAP event being maskable by a tracee.

Now all SIGTRAP TRAP_CHLD events are delivered to debugger.

This code touches MD specific logic and the child_return routine.
It's an intermediate step with a room for refactoring in future and
right now the least invasive approach. This allows to assert expected
behavior in already existing ATF tests and make the code prettier
in future keeping the same semantics. Probably there is a need for a MI
wrapper of child_return for shared functionality between ports.

branches: 1.107.10; 1.107.18;
Send SIGBUS when accessing mmap() past end of file (handle EINVAL in fault
path). While there, handle ENOMEM as well.

branches: 1.106.6;
Move to a flat space register convention. %sr[4-7] are all now the space
number allocated to the process. gcc produces (slightly) better code
with this convention.

Retain backwards compatiblity.

Welcome to 6.99.36

Simplify cpu_spawn_return - setregs has done all the work already.

Change a few __unused to __diagused/__debugused

Use __unused on a variable used in a KASSERT

Use sy_invoke

branches: 1.101.2; 1.101.4;
Pass the trap address in ksi_addr without privilege bits.

Previously GDB would get confused by the trap address given when hitting
the _rtld_debug_state breakpoint.

unifdef FPEMUL. All our supported systems have an FPU.

Deal with setting of space registers and protection ids for posix_spawn.

Thanks to martin for the help.

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

branches: 1.97.2;
Add missing MD code for posix_spawn

branches: 1.96.4; 1.96.8;
Make some more information (psw/fpu_state) per cpu.

branches: 1.95.2; 1.95.4;
Move counting of faults, traps, intrs, soft[intr]s, syscalls, and nswtch
from uvmexp to per-cpu cpu_data and move them to 64bits. Remove unneeded
includes of <uvm/uvm_extern.h> and/or <uvm/uvm.h>.

#if 0 out the trap checks around tf_iioq_{head,tail}. The sti(4) code
can/will trigger them.

fix hppa ucas_*():
- "ret" is a kernel-space pointer.
- move the ucas RAS check from the device interrupt path
to the page-fault path.

Fix previous.

Pull change across from OpenBSD:

Make single stepping a system call work. Instead of single stepping
through the syscall gateway page, which doesn't work since that page is
shared between process, this makes us step over that bit by setting a
breakpoint on the instruction where the system call returns.

SANITY output tweak.

KNF

Remove lwp_uc_pool, replace it with kmem(9), plus add some consistency.
As discussed, a while ago, with ad@.

Include SSBREAKPOINT in the PTRACE #ifdef/#endif.

Add md_astpending to struct mdlwp and use it.

Make cpl, ipending, and intr_depth per CPU values.

Correct the list of traps that we need to get the va and space from
ii[os]q.

From OpenBSD.

Wrap long line.

Whitespace / comment formatting.

Improve the output from frame_sanity_check.

fix copy{in,out}{,str}() to return the error returned by uvm_fault().
fixes PR 41813.

The FP regs are accessed in both physical and virtual modes. Make sure
they're allocated in directly mapped memory to avoid aliasing problems
and unnecessary cache flushing.

Fix various names and references. If we ever directly map the uarea and
pcb we can change things at that point.

Significantly helps my J210 (PA7200).

Mostly from OpenBSD.

branches: 1.78.2;
Start at the first, not the second, exception register.

Comment formating.

Oops. Whitespace in previous.

Remove a cast by fixing a printf format.

Reformat a comment.

Minor tweaks.

Use __arraycount

branches: 1.71.2;
Use pmap_extract instead of kvtop.

Tweak astpending handling.

Check cpl level has returned to entry value when leaving trap/syscall.

From OpenBSD.

Add a comment.

Rename type_raw to trapnum and use in a couple more places.

Reduces diff to OpenBSD.

Count traps in uvmexp. Reduces diff to OpenBSD.

Update 64bit arg handling to use the flags in the syscall table.

Only do trace_enter/trace_exit if we're tracing.

curlwp is never NULL now. Sprinkle KASSERT just in case.

Fix DEBUG build. Hi rmind.

- Use uvm_lwp_setuarea() instead of directly setting address to lwp_t::l_addr.
- Replace most remaining uses of l_addr with uvm_lwp_getuarea() or lwp_getpcb().
- Amend assembly in ports where it accesses PCB via struct user.
- Rename L_ADDR to L_PCB in few places. Reduce sys/user.h inclusions.