Remove unnecessary include of <sys/malloc.h>.

Trailing whitespace

mips/rmi: Hack to get XLSATX64.MP kernel building again.

Using <mips/asm.h> in a .c file is kinda grody but CALLFRAME_SIZ
doesn't seem to be defined anywhere else. Not sure how this was ever
supposed to work...

Don't include mips/asm.h from a C file

Rip out pserialize(9) logic now that the RCU patent has expired.

pserialize_perform() is now basically just xc_barrier(XC_HIGHPRI).
No more tentacles throughout the scheduler. Simplify the psz read
count for diagnostic assertions by putting it unconditionally into
cpu_info.

From rmind@, tidied up by me.

Fix false sharing problems with cpu_info. Identified with tprof(8).
This was a very nice win in my tests on a 48 CPU box.

- Reorganise cpu_data slightly according to usage.
- Put cpu_onproc into struct cpu_info alongside ci_curlwp (now is ci_onproc).
- On x86, put some items in their own cache lines according to usage, like
the IPI bitmask and ci_want_resched.

branches: 1.9.18;
Print both user and kernel segtabs

Transition from __cpuset_t to kcpuset_t *. This brings the local pmap one
step closer to uvm/pmap, its eventual replacement. Tested on ERLITE MP kernel.

Rework cavium support in preparation for MULTIPROCESSOR support

branches: 1.6.6;
nobody uses qchain and it is about to be GC'ed.

branches: 1.5.2;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.4.4; 1.4.14;
minor cleanups.

- add cpu_rmixl_run(), and set in mips_locoresw.lsw_cpu_run
to be called from cpu_hatch() once cpus are running,
so we can determine what threads are configured
and running, and can finish initialization of per-core registers
depending on that.
- in cpu_rmixl_db_watch_init() clear IEU_DEFEATURE[DBE],
and init all COP0 watchpoint regs
- option MIPS_DDB_WATCH is deprecated, removed; use of cpu watchpoints
is longer depends on that or DDB

Merge forward from matt-nb5-mips64.

branches: 1.1.2; 1.1.6; 1.1.8; 1.1.10;
file rmixl_cpu.c was initially added on branch matt-nb5-mips64.

mips/rmi: Hack to get XLSATX64.MP kernel building again.

Using <mips/asm.h> in a .c file is kinda grody but CALLFRAME_SIZ
doesn't seem to be defined anywhere else. Not sure how this was ever
supposed to work...

Don't include mips/asm.h from a C file

Rip out pserialize(9) logic now that the RCU patent has expired.

pserialize_perform() is now basically just xc_barrier(XC_HIGHPRI).
No more tentacles throughout the scheduler. Simplify the psz read
count for diagnostic assertions by putting it unconditionally into
cpu_info.

From rmind@, tidied up by me.

Fix false sharing problems with cpu_info. Identified with tprof(8).
This was a very nice win in my tests on a 48 CPU box.

- Reorganise cpu_data slightly according to usage.
- Put cpu_onproc into struct cpu_info alongside ci_curlwp (now is ci_onproc).
- On x86, put some items in their own cache lines according to usage, like
the IPI bitmask and ci_want_resched.

branches: 1.9.18;
Print both user and kernel segtabs

Transition from __cpuset_t to kcpuset_t *. This brings the local pmap one
step closer to uvm/pmap, its eventual replacement. Tested on ERLITE MP kernel.

Rework cavium support in preparation for MULTIPROCESSOR support

branches: 1.6.6;
nobody uses qchain and it is about to be GC'ed.

branches: 1.5.2;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.4.4; 1.4.14;
minor cleanups.

- add cpu_rmixl_run(), and set in mips_locoresw.lsw_cpu_run
to be called from cpu_hatch() once cpus are running,
so we can determine what threads are configured
and running, and can finish initialization of per-core registers
depending on that.
- in cpu_rmixl_db_watch_init() clear IEU_DEFEATURE[DBE],
and init all COP0 watchpoint regs
- option MIPS_DDB_WATCH is deprecated, removed; use of cpu watchpoints
is longer depends on that or DDB

Merge forward from matt-nb5-mips64.

branches: 1.1.2; 1.1.6; 1.1.8; 1.1.10;
file rmixl_cpu.c was initially added on branch matt-nb5-mips64.

Don't include mips/asm.h from a C file

Rip out pserialize(9) logic now that the RCU patent has expired.

pserialize_perform() is now basically just xc_barrier(XC_HIGHPRI).
No more tentacles throughout the scheduler. Simplify the psz read
count for diagnostic assertions by putting it unconditionally into
cpu_info.

From rmind@, tidied up by me.

Fix false sharing problems with cpu_info. Identified with tprof(8).
This was a very nice win in my tests on a 48 CPU box.

- Reorganise cpu_data slightly according to usage.
- Put cpu_onproc into struct cpu_info alongside ci_curlwp (now is ci_onproc).
- On x86, put some items in their own cache lines according to usage, like
the IPI bitmask and ci_want_resched.

branches: 1.9.18;
Print both user and kernel segtabs

Transition from __cpuset_t to kcpuset_t *. This brings the local pmap one
step closer to uvm/pmap, its eventual replacement. Tested on ERLITE MP kernel.

Rework cavium support in preparation for MULTIPROCESSOR support

branches: 1.6.6;
nobody uses qchain and it is about to be GC'ed.

branches: 1.5.2;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.4.4; 1.4.14;
minor cleanups.

- add cpu_rmixl_run(), and set in mips_locoresw.lsw_cpu_run
to be called from cpu_hatch() once cpus are running,
so we can determine what threads are configured
and running, and can finish initialization of per-core registers
depending on that.
- in cpu_rmixl_db_watch_init() clear IEU_DEFEATURE[DBE],
and init all COP0 watchpoint regs
- option MIPS_DDB_WATCH is deprecated, removed; use of cpu watchpoints
is longer depends on that or DDB

Merge forward from matt-nb5-mips64.

branches: 1.1.2; 1.1.6; 1.1.8; 1.1.10;
file rmixl_cpu.c was initially added on branch matt-nb5-mips64.

Rip out pserialize(9) logic now that the RCU patent has expired.

pserialize_perform() is now basically just xc_barrier(XC_HIGHPRI).
No more tentacles throughout the scheduler. Simplify the psz read
count for diagnostic assertions by putting it unconditionally into
cpu_info.

From rmind@, tidied up by me.

Fix false sharing problems with cpu_info. Identified with tprof(8).
This was a very nice win in my tests on a 48 CPU box.

- Reorganise cpu_data slightly according to usage.
- Put cpu_onproc into struct cpu_info alongside ci_curlwp (now is ci_onproc).
- On x86, put some items in their own cache lines according to usage, like
the IPI bitmask and ci_want_resched.

Print both user and kernel segtabs

Transition from __cpuset_t to kcpuset_t *. This brings the local pmap one
step closer to uvm/pmap, its eventual replacement. Tested on ERLITE MP kernel.

Rework cavium support in preparation for MULTIPROCESSOR support

branches: 1.6.6;
nobody uses qchain and it is about to be GC'ed.

branches: 1.5.2;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.4.4; 1.4.14;
minor cleanups.

- add cpu_rmixl_run(), and set in mips_locoresw.lsw_cpu_run
to be called from cpu_hatch() once cpus are running,
so we can determine what threads are configured
and running, and can finish initialization of per-core registers
depending on that.
- in cpu_rmixl_db_watch_init() clear IEU_DEFEATURE[DBE],
and init all COP0 watchpoint regs
- option MIPS_DDB_WATCH is deprecated, removed; use of cpu watchpoints
is longer depends on that or DDB

Merge forward from matt-nb5-mips64.

branches: 1.1.2; 1.1.6; 1.1.8; 1.1.10;
file rmixl_cpu.c was initially added on branch matt-nb5-mips64.

Fix false sharing problems with cpu_info. Identified with tprof(8).
This was a very nice win in my tests on a 48 CPU box.

- Reorganise cpu_data slightly according to usage.
- Put cpu_onproc into struct cpu_info alongside ci_curlwp (now is ci_onproc).
- On x86, put some items in their own cache lines according to usage, like
the IPI bitmask and ci_want_resched.

Print both user and kernel segtabs

Transition from __cpuset_t to kcpuset_t *. This brings the local pmap one
step closer to uvm/pmap, its eventual replacement. Tested on ERLITE MP kernel.

Rework cavium support in preparation for MULTIPROCESSOR support

branches: 1.6.6;
nobody uses qchain and it is about to be GC'ed.

branches: 1.5.2;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.4.4; 1.4.14;
minor cleanups.

- add cpu_rmixl_run(), and set in mips_locoresw.lsw_cpu_run
to be called from cpu_hatch() once cpus are running,
so we can determine what threads are configured
and running, and can finish initialization of per-core registers
depending on that.
- in cpu_rmixl_db_watch_init() clear IEU_DEFEATURE[DBE],
and init all COP0 watchpoint regs
- option MIPS_DDB_WATCH is deprecated, removed; use of cpu watchpoints
is longer depends on that or DDB

Merge forward from matt-nb5-mips64.

branches: 1.1.2; 1.1.6; 1.1.8; 1.1.10;
file rmixl_cpu.c was initially added on branch matt-nb5-mips64.