Copyright (c) 2003, Sun Microsystems, Inc. All Rights Reserved.
The contents of this file are subject to the terms of the Common Development and Distribution License (the "License"). You may not use this file except in compliance with the License.
You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE or http://www.opensolaris.org/os/licensing. See the License for the specific language governing permissions and limitations under the License.
When distributing Covered Code, include this CDDL HEADER in each file and include the License file at usr/src/OPENSOLARIS.LICENSE. If applicable, add the following below this CDDL HEADER, with the fields enclosed by brackets "[]" replaced with your own identifying information: Portions Copyright [yyyy] [name of copyright owner]
cc [ flag... ] file... [ library... ] #include <schedctl.h> schedctl_t *schedctl_init(void);
schedctl_t *schedctl_lookup(void);
void schedctl_exit(void);
void schedctl_start(schedctl_t *ptr);
void schedctl_stop(schedctl_t *ptr);
These functions provide limited control over the scheduling of a thread (see threads(5)). They allow a running thread to give a hint to the kernel that preemptions of that thread should be avoided. The most likely use for these functions is to block preemption while holding a spinlock. Improper use of this facility, including attempts to block preemption for sustained periods of time, may result in reduced performance.
The schedctl_init() function initializes preemption control for the calling thread and returns a pointer used to refer to the data. If schedctl_init() is called more than once by the same thread, the most recently returned pointer is the only valid one.
The schedctl_lookup() function returns the currently allocated preemption control data associated with the calling thread that was previously returned by schedctl_init(). This can be useful in programs where it is difficult to maintain local state for each thread.
The schedctl_exit() function removes the preemption control data associated with the calling thread.
The schedctl_start() macro gives a hint to the kernel scheduler that preemption should be avoided on the current thread. The pointer passed to the macro must be the same as the pointer returned by the call to schedctl_init() by the current thread. The behavior of the program when other values are passed is undefined.
The schedctl_stop() macro removes the hint that was set by schedctl_start(). As with schedctl_start(), the pointer passed to the macro must be the same as the pointer returned by the call to schedctl_init() by the current thread.
The schedctl_start() and schedctl_stop() macros are intended to be used to bracket short critical sections, such as the time spent holding a spinlock. Other uses, including the failure to call schedctl_stop() soon after calling schedctl_start(), might result in poor performance.
The schedctl_init() function returns a pointer to a schedctl_t structure if the initialization was successful, or NULL otherwise. The schedctl_lookup() function returns a pointer to a schedctl_t structure if the data for that thread was found, or NULL otherwise.
No errors are returned.
See attributes(5) for descriptions of the following attributes:
| ATTRIBUTE TYPE ATTRIBUTE VALUE |
| Interface Stability Stable |
| MT-Level MT-Safe |
priocntl(1), exec(2), fork(2), priocntl(2), attributes(5), threads(5)
Preemption control is intended for use by threads belonging to the time-sharing (TS), interactive (IA), fair-share (FSS), and fixed-priority (FX) scheduling classes. If used by threads in other scheduling classes, such as real-time (RT), no errors will be returned but schedctl_start() and schedctl_stop() will not have any effect.
The data used for preemption control are not copied in the child of a fork(2). Thus, if a process containing threads using preemption control calls fork and the child does not immediately call exec(2), each thread in the child must call schedctl_init() again prior to any future uses of schedctl_start() and schedctl_stop(). Failure to do so will result in undefined behavior.