all platforms, kernel: remove __HAVE_CLOCKINTR symbol

Every platform made the clockintr switch at least six months ago.
The __HAVE_CLOCKINTR symbol is now redundant. Remove it.

Prompted by claudio@.

Link: https://marc.info/?l=openbsd-tech&m=168826181015032&w=2

"makes sense" mlarkin@

alpha: switch to clockintr

- Add missing tick_nsec initialization to cpu_initclocks().
- Set stathz = hz, profhz = stathz; we don't have any control over the
interrupt clock on alpha so everything has the same frequency.
- Set schedhz = 16 to imitate current schedclock() dispatch frequency.

Bringup help from claudio@. Tested by miod@ (2-CPU DS25).

Link: https://marc.info/?l=openbsd-tech&m=166776333303245&w=2

ok mlarkin@

#define _MAX_PAGE_SHIFT in MD _types.h as the maximum pagesize an arch
needs (looking at you sgi, but others required this before). This is for
the circumstances we need pagesize known at compile time, not getpagesize()
runtime. Use it for malloc storage sizes, for shm, and to set pthread stack
default sizes. The stack sizes were a mess, and pushing them towards
page-aligned is healthy move (which will also be needed by the coming
stack register checker)
ok guenther kettenis, discussion with stefan

lint is dead and C99 may be old enough to drive a car: delete LONGLONG
comments

ok millert@

It's safe to assumed 'signed' exists

Provide MI symbol _STACKALIGNBYTES

Take II, this time without flubbing off_t: : move several internal
type definitions that are the same across all archs from
<machine/_types.h> to <sys/_types.h>

ok deraadt@ kettenis@ otto@

backout previous; off_t must be signed and there might be other C++ API
breakage lurking; ok kettenis@

To ease future changes, move several internal type definitions that are
the same across all archs from <machine/_types.h> to <sys/_types.h>

ok deraadt@

Hide members of __va_list in the implementation namespace

ok miod@

unifdef -D __HAVE_TIMECOUNTER

Provide namespace-safe alignment macros in <machine/_types.h>, with
compat names kept in <machine/param.h>. In <sys/socket.h>, pull
in <sys/_types.h> instead of the namespace polluting <machine/param.h>
and completely eliminate __CMSG_ALIGN, replaced by _ALIGN

ok deraadt@

Make the INT_FAST*_{MIN,MAX} macros match the types they're defined to.
Since the underlying types of the int_fast types are set by machine/_types.h,
put internal macros in that same file and define the exposed
INT_FAST*_{MIN,MAX} macros from those.

ok millert@, kettenis@

Make alpha consistent with our other architectures by using "long
long" for __off_t. This is a C++ ABI bump, but martynas@ already
bumped libstdc++.

Discussed and requested by many on icb.

Normalize sentinel. Use _MACHINE_*_H_ and _<ARCH>_*_H_ properly and consitently.

Discussed and okay drahn@. Okay deraadt@.

Get rid of __HAVE_GENERIC_SOFT_INTERRUPTS now that all our platforms support it.

ok jsing@, miod@

- add proper double_t and float_t definitions for each arch
- math.h shouldn't define FLT_EVAL_METHOD, but float.h should (per
C99). remove from math.h, and add proper definitions in float.h
ok millert@

The world of __HAVEs and __HAVE_NOTs is reducing. All architectures
have cpu_info now, so kill the option.

eyeballed by jsg@ and grange@

Remove the MI implementation of mutexes and remove the __HAVE_MUTEX
option. Every architecture implements mutexes now.

oops; martin go see what you did here..

kill __HAVE_DEVICE_REGISTER by requiring all architectures to have a
device_register() function -- even if it does nothing. reduces the
cpp-based blather different between architectures
idea ok'd by miod; tested on all architectures (except a few miod will
need to cleanup because he has them)

simple single-processor only mutex implementation

ok miod@

Switch alpha to timecounters.

We use rpcc for the clock. We don't try to get the frequency from the
hwrpb, since from experience we know that it's very unreliable and
can be several % off from the real frequency. Instead we calibrate
the timecounter frequency by using the clock device and count the
number of cycles between two second ticks of the clocks.

This is not perfect, but still much better than what we had before.
We'll have to add an i8254 clock for MP though.

miod@ ok

fix generic soft interrupts; improve establishing and disestablishing of
soft interrupts

from NetBSD, 'reads good art@', also tested by robert@

__HAVE_CPUINFO for alpha, with bits from netbsd

ok miod@

Merge machine/ansi.h and machine/types.h into machine/_types.h and
rename the types using the __ protected namespace (e.g. __size_t).
Idea from FreeBSD.

alpha: switch to clockintr

- Add missing tick_nsec initialization to cpu_initclocks().
- Set stathz = hz, profhz = stathz; we don't have any control over the
interrupt clock on alpha so everything has the same frequency.
- Set schedhz = 16 to imitate current schedclock() dispatch frequency.

Bringup help from claudio@. Tested by miod@ (2-CPU DS25).

Link: https://marc.info/?l=openbsd-tech&m=166776333303245&w=2

ok mlarkin@

#define _MAX_PAGE_SHIFT in MD _types.h as the maximum pagesize an arch
needs (looking at you sgi, but others required this before). This is for
the circumstances we need pagesize known at compile time, not getpagesize()
runtime. Use it for malloc storage sizes, for shm, and to set pthread stack
default sizes. The stack sizes were a mess, and pushing them towards
page-aligned is healthy move (which will also be needed by the coming
stack register checker)
ok guenther kettenis, discussion with stefan

lint is dead and C99 may be old enough to drive a car: delete LONGLONG
comments

ok millert@

It's safe to assumed 'signed' exists

Provide MI symbol _STACKALIGNBYTES

Take II, this time without flubbing off_t: : move several internal
type definitions that are the same across all archs from
<machine/_types.h> to <sys/_types.h>

ok deraadt@ kettenis@ otto@

backout previous; off_t must be signed and there might be other C++ API
breakage lurking; ok kettenis@

To ease future changes, move several internal type definitions that are
the same across all archs from <machine/_types.h> to <sys/_types.h>

ok deraadt@

Hide members of __va_list in the implementation namespace

ok miod@

unifdef -D __HAVE_TIMECOUNTER

Provide namespace-safe alignment macros in <machine/_types.h>, with
compat names kept in <machine/param.h>. In <sys/socket.h>, pull
in <sys/_types.h> instead of the namespace polluting <machine/param.h>
and completely eliminate __CMSG_ALIGN, replaced by _ALIGN

ok deraadt@

Make the INT_FAST*_{MIN,MAX} macros match the types they're defined to.
Since the underlying types of the int_fast types are set by machine/_types.h,
put internal macros in that same file and define the exposed
INT_FAST*_{MIN,MAX} macros from those.

ok millert@, kettenis@

Make alpha consistent with our other architectures by using "long
long" for __off_t. This is a C++ ABI bump, but martynas@ already
bumped libstdc++.

Discussed and requested by many on icb.

Normalize sentinel. Use _MACHINE_*_H_ and _<ARCH>_*_H_ properly and consitently.

Discussed and okay drahn@. Okay deraadt@.

Get rid of __HAVE_GENERIC_SOFT_INTERRUPTS now that all our platforms support it.

ok jsing@, miod@

- add proper double_t and float_t definitions for each arch
- math.h shouldn't define FLT_EVAL_METHOD, but float.h should (per
C99). remove from math.h, and add proper definitions in float.h
ok millert@

The world of __HAVEs and __HAVE_NOTs is reducing. All architectures
have cpu_info now, so kill the option.

eyeballed by jsg@ and grange@

Remove the MI implementation of mutexes and remove the __HAVE_MUTEX
option. Every architecture implements mutexes now.

oops; martin go see what you did here..

kill __HAVE_DEVICE_REGISTER by requiring all architectures to have a
device_register() function -- even if it does nothing. reduces the
cpp-based blather different between architectures
idea ok'd by miod; tested on all architectures (except a few miod will
need to cleanup because he has them)

simple single-processor only mutex implementation

ok miod@

Switch alpha to timecounters.

We use rpcc for the clock. We don't try to get the frequency from the
hwrpb, since from experience we know that it's very unreliable and
can be several % off from the real frequency. Instead we calibrate
the timecounter frequency by using the clock device and count the
number of cycles between two second ticks of the clocks.

This is not perfect, but still much better than what we had before.
We'll have to add an i8254 clock for MP though.

miod@ ok

fix generic soft interrupts; improve establishing and disestablishing of
soft interrupts

from NetBSD, 'reads good art@', also tested by robert@

__HAVE_CPUINFO for alpha, with bits from netbsd

ok miod@

Merge machine/ansi.h and machine/types.h into machine/_types.h and
rename the types using the __ protected namespace (e.g. __size_t).
Idea from FreeBSD.

#define _MAX_PAGE_SHIFT in MD _types.h as the maximum pagesize an arch
needs (looking at you sgi, but others required this before). This is for
the circumstances we need pagesize known at compile time, not getpagesize()
runtime. Use it for malloc storage sizes, for shm, and to set pthread stack
default sizes. The stack sizes were a mess, and pushing them towards
page-aligned is healthy move (which will also be needed by the coming
stack register checker)
ok guenther kettenis, discussion with stefan

lint is dead and C99 may be old enough to drive a car: delete LONGLONG
comments

ok millert@

It's safe to assumed 'signed' exists

Provide MI symbol _STACKALIGNBYTES

Take II, this time without flubbing off_t: : move several internal
type definitions that are the same across all archs from
<machine/_types.h> to <sys/_types.h>

ok deraadt@ kettenis@ otto@

backout previous; off_t must be signed and there might be other C++ API
breakage lurking; ok kettenis@

To ease future changes, move several internal type definitions that are
the same across all archs from <machine/_types.h> to <sys/_types.h>

ok deraadt@

Hide members of __va_list in the implementation namespace

ok miod@

unifdef -D __HAVE_TIMECOUNTER

Provide namespace-safe alignment macros in <machine/_types.h>, with
compat names kept in <machine/param.h>. In <sys/socket.h>, pull
in <sys/_types.h> instead of the namespace polluting <machine/param.h>
and completely eliminate __CMSG_ALIGN, replaced by _ALIGN

ok deraadt@

Make the INT_FAST*_{MIN,MAX} macros match the types they're defined to.
Since the underlying types of the int_fast types are set by machine/_types.h,
put internal macros in that same file and define the exposed
INT_FAST*_{MIN,MAX} macros from those.

ok millert@, kettenis@

Make alpha consistent with our other architectures by using "long
long" for __off_t. This is a C++ ABI bump, but martynas@ already
bumped libstdc++.

Discussed and requested by many on icb.

Normalize sentinel. Use _MACHINE_*_H_ and _<ARCH>_*_H_ properly and consitently.

Discussed and okay drahn@. Okay deraadt@.

Get rid of __HAVE_GENERIC_SOFT_INTERRUPTS now that all our platforms support it.

ok jsing@, miod@

- add proper double_t and float_t definitions for each arch
- math.h shouldn't define FLT_EVAL_METHOD, but float.h should (per
C99). remove from math.h, and add proper definitions in float.h
ok millert@

The world of __HAVEs and __HAVE_NOTs is reducing. All architectures
have cpu_info now, so kill the option.

eyeballed by jsg@ and grange@

Remove the MI implementation of mutexes and remove the __HAVE_MUTEX
option. Every architecture implements mutexes now.

oops; martin go see what you did here..

kill __HAVE_DEVICE_REGISTER by requiring all architectures to have a
device_register() function -- even if it does nothing. reduces the
cpp-based blather different between architectures
idea ok'd by miod; tested on all architectures (except a few miod will
need to cleanup because he has them)

simple single-processor only mutex implementation

ok miod@

Switch alpha to timecounters.

We use rpcc for the clock. We don't try to get the frequency from the
hwrpb, since from experience we know that it's very unreliable and
can be several % off from the real frequency. Instead we calibrate
the timecounter frequency by using the clock device and count the
number of cycles between two second ticks of the clocks.

This is not perfect, but still much better than what we had before.
We'll have to add an i8254 clock for MP though.

miod@ ok

fix generic soft interrupts; improve establishing and disestablishing of
soft interrupts

from NetBSD, 'reads good art@', also tested by robert@

__HAVE_CPUINFO for alpha, with bits from netbsd

ok miod@

Merge machine/ansi.h and machine/types.h into machine/_types.h and
rename the types using the __ protected namespace (e.g. __size_t).
Idea from FreeBSD.

lint is dead and C99 may be old enough to drive a car: delete LONGLONG
comments

ok millert@

It's safe to assumed 'signed' exists

Provide MI symbol _STACKALIGNBYTES

Take II, this time without flubbing off_t: : move several internal
type definitions that are the same across all archs from
<machine/_types.h> to <sys/_types.h>

ok deraadt@ kettenis@ otto@

backout previous; off_t must be signed and there might be other C++ API
breakage lurking; ok kettenis@

To ease future changes, move several internal type definitions that are
the same across all archs from <machine/_types.h> to <sys/_types.h>

ok deraadt@

Hide members of __va_list in the implementation namespace

ok miod@

unifdef -D __HAVE_TIMECOUNTER

Provide namespace-safe alignment macros in <machine/_types.h>, with
compat names kept in <machine/param.h>. In <sys/socket.h>, pull
in <sys/_types.h> instead of the namespace polluting <machine/param.h>
and completely eliminate __CMSG_ALIGN, replaced by _ALIGN

ok deraadt@

Make the INT_FAST*_{MIN,MAX} macros match the types they're defined to.
Since the underlying types of the int_fast types are set by machine/_types.h,
put internal macros in that same file and define the exposed
INT_FAST*_{MIN,MAX} macros from those.

ok millert@, kettenis@

Make alpha consistent with our other architectures by using "long
long" for __off_t. This is a C++ ABI bump, but martynas@ already
bumped libstdc++.

Discussed and requested by many on icb.

Normalize sentinel. Use _MACHINE_*_H_ and _<ARCH>_*_H_ properly and consitently.

Discussed and okay drahn@. Okay deraadt@.

Get rid of __HAVE_GENERIC_SOFT_INTERRUPTS now that all our platforms support it.

ok jsing@, miod@

- add proper double_t and float_t definitions for each arch
- math.h shouldn't define FLT_EVAL_METHOD, but float.h should (per
C99). remove from math.h, and add proper definitions in float.h
ok millert@

The world of __HAVEs and __HAVE_NOTs is reducing. All architectures
have cpu_info now, so kill the option.

eyeballed by jsg@ and grange@

Remove the MI implementation of mutexes and remove the __HAVE_MUTEX
option. Every architecture implements mutexes now.

oops; martin go see what you did here..

kill __HAVE_DEVICE_REGISTER by requiring all architectures to have a
device_register() function -- even if it does nothing. reduces the
cpp-based blather different between architectures
idea ok'd by miod; tested on all architectures (except a few miod will
need to cleanup because he has them)

simple single-processor only mutex implementation

ok miod@

Switch alpha to timecounters.

We use rpcc for the clock. We don't try to get the frequency from the
hwrpb, since from experience we know that it's very unreliable and
can be several % off from the real frequency. Instead we calibrate
the timecounter frequency by using the clock device and count the
number of cycles between two second ticks of the clocks.

This is not perfect, but still much better than what we had before.
We'll have to add an i8254 clock for MP though.

miod@ ok

fix generic soft interrupts; improve establishing and disestablishing of
soft interrupts

from NetBSD, 'reads good art@', also tested by robert@

__HAVE_CPUINFO for alpha, with bits from netbsd

ok miod@

Merge machine/ansi.h and machine/types.h into machine/_types.h and
rename the types using the __ protected namespace (e.g. __size_t).
Idea from FreeBSD.