freebsd-skq/sys/kern/ksched.c

270 lines
6.1 KiB
C
Raw Normal View History

/*
* Copyright (c) 1996, 1997
* HD Associates, Inc. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by HD Associates, Inc
* 4. Neither the name of the author nor the names of any co-contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY HD ASSOCIATES AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL HD ASSOCIATES OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* $FreeBSD$
*/
/* ksched: Soft real time scheduling based on "rtprio".
*/
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/proc.h>
#include <sys/resource.h>
#include <machine/cpu.h> /* For need_resched */
Commit major SMP cleanups and move the BGL (big giant lock) in the syscall path inward. A system call may select whether it needs the MP lock or not (the default being that it does need it). A great deal of conditional SMP code for various deadended experiments has been removed. 'cil' and 'cml' have been removed entirely, and the locking around the cpl has been removed. The conditional separately-locked fast-interrupt code has been removed, meaning that interrupts must hold the CPL now (but they pretty much had to anyway). Another reason for doing this is that the original separate-lock for interrupts just doesn't apply to the interrupt thread mechanism being contemplated. Modifications to the cpl may now ONLY occur while holding the MP lock. For example, if an otherwise MP safe syscall needs to mess with the cpl, it must hold the MP lock for the duration and must (as usual) save/restore the cpl in a nested fashion. This is precursor work for the real meat coming later: avoiding having to hold the MP lock for common syscalls and I/O's and interrupt threads. It is expected that the spl mechanisms and new interrupt threading mechanisms will be able to run in tandem, allowing a slow piecemeal transition to occur. This patch should result in a moderate performance improvement due to the considerable amount of code that has been removed from the critical path, especially the simplification of the spl*() calls. The real performance gains will come later. Approved by: jkh Reviewed by: current, bde (exception.s) Some work taken from: luoqi's patch
2000-03-28 07:16:37 +00:00
#include <machine/ipl.h> /* For need_resched */
#include <posix4/posix4.h>
/* ksched: Real-time extension to support POSIX priority scheduling.
*/
struct ksched {
struct timespec rr_interval;
};
int ksched_attach(struct ksched **p)
{
struct ksched *ksched= p31b_malloc(sizeof(*ksched));
ksched->rr_interval.tv_sec = 0;
ksched->rr_interval.tv_nsec = 1000000000L / roundrobin_interval();
*p = ksched;
return 0;
}
int ksched_detach(struct ksched *p)
{
p31b_free(p);
return 0;
}
/*
* XXX About priorities
*
* POSIX 1003.1b requires that numerically higher priorities be of
* higher priority. It also permits sched_setparam to be
* implementation defined for SCHED_OTHER. I don't like
* the notion of inverted priorites for normal processes when
* you can use "setpriority" for that.
*
* I'm rejecting sched_setparam for SCHED_OTHER with EINVAL.
*/
/* Macros to convert between the unix (lower numerically is higher priority)
* and POSIX 1003.1b (higher numerically is higher priority)
*/
#define p4prio_to_rtpprio(P) (RTP_PRIO_MAX - (P))
#define rtpprio_to_p4prio(P) (RTP_PRIO_MAX - (P))
/* These improve readability a bit for me:
*/
#define P1B_PRIO_MIN rtpprio_to_p4prio(RTP_PRIO_MAX)
#define P1B_PRIO_MAX rtpprio_to_p4prio(RTP_PRIO_MIN)
1998-04-15 17:47:40 +00:00
static __inline int
getscheduler(register_t *ret, struct ksched *ksched, struct proc *p)
{
Implement a unified run queue and adjust priority levels accordingly. - All processes go into the same array of queues, with different scheduling classes using different portions of the array. This allows user processes to have their priorities propogated up into interrupt thread range if need be. - I chose 64 run queues as an arbitrary number that is greater than 32. We used to have 4 separate arrays of 32 queues each, so this may not be optimal. The new run queue code was written with this in mind; changing the number of run queues only requires changing constants in runq.h and adjusting the priority levels. - The new run queue code takes the run queue as a parameter. This is intended to be used to create per-cpu run queues. Implement wrappers for compatibility with the old interface which pass in the global run queue structure. - Group the priority level, user priority, native priority (before propogation) and the scheduling class into a struct priority. - Change any hard coded priority levels that I found to use symbolic constants (TTIPRI and TTOPRI). - Remove the curpriority global variable and use that of curproc. This was used to detect when a process' priority had lowered and it should yield. We now effectively yield on every interrupt. - Activate propogate_priority(). It should now have the desired effect without needing to also propogate the scheduling class. - Temporarily comment out the call to vm_page_zero_idle() in the idle loop. It interfered with propogate_priority() because the idle process needed to do a non-blocking acquire of Giant and then other processes would try to propogate their priority onto it. The idle process should not do anything except idle. vm_page_zero_idle() will return in the form of an idle priority kernel thread which is woken up at apprioriate times by the vm system. - Update struct kinfo_proc to the new priority interface. Deliberately change its size by adjusting the spare fields. It remained the same size, but the layout has changed, so userland processes that use it would parse the data incorrectly. The size constraint should really be changed to an arbitrary version number. Also add a debug.sizeof sysctl node for struct kinfo_proc.
2001-02-12 00:20:08 +00:00
struct rtprio rtp;
int e = 0;
Implement a unified run queue and adjust priority levels accordingly. - All processes go into the same array of queues, with different scheduling classes using different portions of the array. This allows user processes to have their priorities propogated up into interrupt thread range if need be. - I chose 64 run queues as an arbitrary number that is greater than 32. We used to have 4 separate arrays of 32 queues each, so this may not be optimal. The new run queue code was written with this in mind; changing the number of run queues only requires changing constants in runq.h and adjusting the priority levels. - The new run queue code takes the run queue as a parameter. This is intended to be used to create per-cpu run queues. Implement wrappers for compatibility with the old interface which pass in the global run queue structure. - Group the priority level, user priority, native priority (before propogation) and the scheduling class into a struct priority. - Change any hard coded priority levels that I found to use symbolic constants (TTIPRI and TTOPRI). - Remove the curpriority global variable and use that of curproc. This was used to detect when a process' priority had lowered and it should yield. We now effectively yield on every interrupt. - Activate propogate_priority(). It should now have the desired effect without needing to also propogate the scheduling class. - Temporarily comment out the call to vm_page_zero_idle() in the idle loop. It interfered with propogate_priority() because the idle process needed to do a non-blocking acquire of Giant and then other processes would try to propogate their priority onto it. The idle process should not do anything except idle. vm_page_zero_idle() will return in the form of an idle priority kernel thread which is woken up at apprioriate times by the vm system. - Update struct kinfo_proc to the new priority interface. Deliberately change its size by adjusting the spare fields. It remained the same size, but the layout has changed, so userland processes that use it would parse the data incorrectly. The size constraint should really be changed to an arbitrary version number. Also add a debug.sizeof sysctl node for struct kinfo_proc.
2001-02-12 00:20:08 +00:00
pri_to_rtp(&p->p_pri, &rtp);
switch (rtp.type)
{
case RTP_PRIO_FIFO:
*ret = SCHED_FIFO;
break;
case RTP_PRIO_REALTIME:
*ret = SCHED_RR;
break;
default:
*ret = SCHED_OTHER;
break;
}
return e;
}
int ksched_setparam(register_t *ret, struct ksched *ksched,
struct proc *p, const struct sched_param *param)
{
register_t policy;
int e;
e = getscheduler(&policy, ksched, p);
if (e == 0)
{
if (policy == SCHED_OTHER)
e = EINVAL;
else
e = ksched_setscheduler(ret, ksched, p, policy, param);
}
return e;
}
int ksched_getparam(register_t *ret, struct ksched *ksched,
struct proc *p, struct sched_param *param)
{
Implement a unified run queue and adjust priority levels accordingly. - All processes go into the same array of queues, with different scheduling classes using different portions of the array. This allows user processes to have their priorities propogated up into interrupt thread range if need be. - I chose 64 run queues as an arbitrary number that is greater than 32. We used to have 4 separate arrays of 32 queues each, so this may not be optimal. The new run queue code was written with this in mind; changing the number of run queues only requires changing constants in runq.h and adjusting the priority levels. - The new run queue code takes the run queue as a parameter. This is intended to be used to create per-cpu run queues. Implement wrappers for compatibility with the old interface which pass in the global run queue structure. - Group the priority level, user priority, native priority (before propogation) and the scheduling class into a struct priority. - Change any hard coded priority levels that I found to use symbolic constants (TTIPRI and TTOPRI). - Remove the curpriority global variable and use that of curproc. This was used to detect when a process' priority had lowered and it should yield. We now effectively yield on every interrupt. - Activate propogate_priority(). It should now have the desired effect without needing to also propogate the scheduling class. - Temporarily comment out the call to vm_page_zero_idle() in the idle loop. It interfered with propogate_priority() because the idle process needed to do a non-blocking acquire of Giant and then other processes would try to propogate their priority onto it. The idle process should not do anything except idle. vm_page_zero_idle() will return in the form of an idle priority kernel thread which is woken up at apprioriate times by the vm system. - Update struct kinfo_proc to the new priority interface. Deliberately change its size by adjusting the spare fields. It remained the same size, but the layout has changed, so userland processes that use it would parse the data incorrectly. The size constraint should really be changed to an arbitrary version number. Also add a debug.sizeof sysctl node for struct kinfo_proc.
2001-02-12 00:20:08 +00:00
struct rtprio rtp;
pri_to_rtp(&p->p_pri, &rtp);
if (RTP_PRIO_IS_REALTIME(rtp.type))
param->sched_priority = rtpprio_to_p4prio(rtp.prio);
return 0;
}
/*
* XXX The priority and scheduler modifications should
* be moved into published interfaces in kern/kern_sync.
*
* The permissions to modify process p were checked in "p31b_proc()".
*
*/
int ksched_setscheduler(register_t *ret, struct ksched *ksched,
struct proc *p, int policy, const struct sched_param *param)
{
int e = 0;
struct rtprio rtp;
switch(policy)
{
case SCHED_RR:
case SCHED_FIFO:
if (param->sched_priority >= P1B_PRIO_MIN &&
param->sched_priority <= P1B_PRIO_MAX)
{
rtp.prio = p4prio_to_rtpprio(param->sched_priority);
rtp.type = (policy == SCHED_FIFO)
? RTP_PRIO_FIFO : RTP_PRIO_REALTIME;
Implement a unified run queue and adjust priority levels accordingly. - All processes go into the same array of queues, with different scheduling classes using different portions of the array. This allows user processes to have their priorities propogated up into interrupt thread range if need be. - I chose 64 run queues as an arbitrary number that is greater than 32. We used to have 4 separate arrays of 32 queues each, so this may not be optimal. The new run queue code was written with this in mind; changing the number of run queues only requires changing constants in runq.h and adjusting the priority levels. - The new run queue code takes the run queue as a parameter. This is intended to be used to create per-cpu run queues. Implement wrappers for compatibility with the old interface which pass in the global run queue structure. - Group the priority level, user priority, native priority (before propogation) and the scheduling class into a struct priority. - Change any hard coded priority levels that I found to use symbolic constants (TTIPRI and TTOPRI). - Remove the curpriority global variable and use that of curproc. This was used to detect when a process' priority had lowered and it should yield. We now effectively yield on every interrupt. - Activate propogate_priority(). It should now have the desired effect without needing to also propogate the scheduling class. - Temporarily comment out the call to vm_page_zero_idle() in the idle loop. It interfered with propogate_priority() because the idle process needed to do a non-blocking acquire of Giant and then other processes would try to propogate their priority onto it. The idle process should not do anything except idle. vm_page_zero_idle() will return in the form of an idle priority kernel thread which is woken up at apprioriate times by the vm system. - Update struct kinfo_proc to the new priority interface. Deliberately change its size by adjusting the spare fields. It remained the same size, but the layout has changed, so userland processes that use it would parse the data incorrectly. The size constraint should really be changed to an arbitrary version number. Also add a debug.sizeof sysctl node for struct kinfo_proc.
2001-02-12 00:20:08 +00:00
rtp_to_pri(&rtp, &p->p_pri);
need_resched();
}
else
e = EPERM;
break;
case SCHED_OTHER:
{
rtp.type = RTP_PRIO_NORMAL;
rtp.prio = p4prio_to_rtpprio(param->sched_priority);
Implement a unified run queue and adjust priority levels accordingly. - All processes go into the same array of queues, with different scheduling classes using different portions of the array. This allows user processes to have their priorities propogated up into interrupt thread range if need be. - I chose 64 run queues as an arbitrary number that is greater than 32. We used to have 4 separate arrays of 32 queues each, so this may not be optimal. The new run queue code was written with this in mind; changing the number of run queues only requires changing constants in runq.h and adjusting the priority levels. - The new run queue code takes the run queue as a parameter. This is intended to be used to create per-cpu run queues. Implement wrappers for compatibility with the old interface which pass in the global run queue structure. - Group the priority level, user priority, native priority (before propogation) and the scheduling class into a struct priority. - Change any hard coded priority levels that I found to use symbolic constants (TTIPRI and TTOPRI). - Remove the curpriority global variable and use that of curproc. This was used to detect when a process' priority had lowered and it should yield. We now effectively yield on every interrupt. - Activate propogate_priority(). It should now have the desired effect without needing to also propogate the scheduling class. - Temporarily comment out the call to vm_page_zero_idle() in the idle loop. It interfered with propogate_priority() because the idle process needed to do a non-blocking acquire of Giant and then other processes would try to propogate their priority onto it. The idle process should not do anything except idle. vm_page_zero_idle() will return in the form of an idle priority kernel thread which is woken up at apprioriate times by the vm system. - Update struct kinfo_proc to the new priority interface. Deliberately change its size by adjusting the spare fields. It remained the same size, but the layout has changed, so userland processes that use it would parse the data incorrectly. The size constraint should really be changed to an arbitrary version number. Also add a debug.sizeof sysctl node for struct kinfo_proc.
2001-02-12 00:20:08 +00:00
rtp_to_pri(&rtp, &p->p_pri);
/* XXX Simply revert to whatever we had for last
* normal scheduler priorities.
* This puts a requirement
* on the scheduling code: You must leave the
* scheduling info alone.
*/
need_resched();
}
break;
}
return e;
}
int ksched_getscheduler(register_t *ret, struct ksched *ksched, struct proc *p)
{
return getscheduler(ret, ksched, p);
}
/* ksched_yield: Yield the CPU.
*/
int ksched_yield(register_t *ret, struct ksched *ksched)
{
need_resched();
return 0;
}
int ksched_get_priority_max(register_t*ret, struct ksched *ksched, int policy)
{
int e = 0;
switch (policy)
{
case SCHED_FIFO:
case SCHED_RR:
*ret = RTP_PRIO_MAX;
break;
case SCHED_OTHER:
*ret = PRIO_MAX;
break;
default:
e = EINVAL;
}
return e;
}
int ksched_get_priority_min(register_t *ret, struct ksched *ksched, int policy)
{
int e = 0;
switch (policy)
{
case SCHED_FIFO:
case SCHED_RR:
*ret = P1B_PRIO_MIN;
break;
case SCHED_OTHER:
*ret = PRIO_MIN;
break;
default:
e = EINVAL;
}
return e;
}
int ksched_rr_get_interval(register_t *ret, struct ksched *ksched,
struct proc *p, struct timespec *timespec)
{
*timespec = ksched->rr_interval;
return 0;
}