d/freebsd-dev
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
89685e2269
freebsd-dev/sys/kern/kern_thr.c

306 lines
7.3 KiB
C
Raw Normal View History

/* -> /*- for copyright notices, minor format tweaks as necessary
2005-01-06 23:35:40 +00:00
/*-
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
* Copyright (c) 2003, Jeffrey Roberson <jeff@freebsd.org>
* 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 unmodified, 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.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 THE AUTHOR 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.
*/
Use __FBSDID().
2003-06-11 00:56:59 +00:00
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
#include <sys/param.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include <sys/proc.h>
#include <sys/resourcevar.h>
- Call sched_exit_{kse,thread} and sched_fork{kse,thread} so that thr works with ULE. This was not strictly required by sched_4bsd.
2003-04-11 19:24:37 +00:00
#include <sys/sched.h>
Give libthr a choice (per system) of scope_system or scope_thread scheduling. MFC after: 4 days
2004-09-07 06:33:39 +00:00
#include <sys/sysctl.h>
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
#include <sys/smp.h>
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
#include <sys/sysent.h>
#include <sys/systm.h>
#include <sys/sysproto.h>
#include <sys/signalvar.h>
#include <sys/ucontext.h>
#include <sys/thr.h>
#include <machine/frame.h>
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
extern int max_threads_per_proc;
extern int max_groups_per_proc;
Give libthr a choice (per system) of scope_system or scope_thread scheduling. MFC after: 4 days
2004-09-07 06:33:39 +00:00
SYSCTL_DECL(_kern_threads);
static int thr_scope_sys = 0;
SYSCTL_INT(_kern_threads, OID_AUTO, thr_scope_sys, CTLFLAG_RW,
&thr_scope_sys, 0, "sys or proc scope scheduling");
fix typo MFC after: 2 days
2004-09-07 07:04:47 +00:00
static int thr_concurrency = 0;
Give libthr a choice (per system) of scope_system or scope_thread scheduling. MFC after: 4 days
2004-09-07 06:33:39 +00:00
SYSCTL_INT(_kern_threads, OID_AUTO, thr_concurrency, CTLFLAG_RW,
&thr_concurrency, 0, "a concurrency value if not default");
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
/*
* System call interface.
*/
int
thr_create(struct thread *td, struct thr_create_args *uap)
Change the thread ID (thr_id_t) used for 1:1 threading from being a pointer to the corresponding struct thread to the thread ID (lwpid_t) assigned to that thread. The primary reason for this change is that libthr now internally uses the same ID as the debugger and the kernel when referencing to a kernel thread. This allows us to implement the support for debugging without additional translations and/or mappings. To preserve the ABI, the 1:1 threading syscalls, including the umtx locking API have not been changed to work on a lwpid_t. Instead the 1:1 threading syscalls operate on long and the umtx locking API has not been changed except for the contested bit. Previously this was the least significant bit. Now it's the most significant bit. Since the contested bit should not be tested by userland, this change is not expected to be visible. Just to be sure, UMTX_CONTESTED has been removed from <sys/umtx.h>. Reviewed by: mtm@ ABI preservation tested on: i386, ia64
2004-07-02 00:40:07 +00:00
/* ucontext_t *ctx, long *id, int flags */
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
{
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
struct thread *newtd;
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
ucontext_t ctx;
Change the thread ID (thr_id_t) used for 1:1 threading from being a pointer to the corresponding struct thread to the thread ID (lwpid_t) assigned to that thread. The primary reason for this change is that libthr now internally uses the same ID as the debugger and the kernel when referencing to a kernel thread. This allows us to implement the support for debugging without additional translations and/or mappings. To preserve the ABI, the 1:1 threading syscalls, including the umtx locking API have not been changed to work on a lwpid_t. Instead the 1:1 threading syscalls operate on long and the umtx locking API has not been changed except for the contested bit. Previously this was the least significant bit. Now it's the most significant bit. Since the contested bit should not be tested by userland, this change is not expected to be visible. Just to be sure, UMTX_CONTESTED has been removed from <sys/umtx.h>. Reviewed by: mtm@ ABI preservation tested on: i386, ia64
2004-07-02 00:40:07 +00:00
long id;
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
int error;
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
struct ksegrp *kg, *newkg;
struct proc *p;
Close a race between thr_create and sysctl -w, the thr_scope_sys could be changed when thr_create is running, and we tested it for several times.
2004-10-06 02:29:19 +00:00
int scope_sys;
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
p = td->td_proc;
kg = td->td_ksegrp;
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
if ((error = copyin(uap->ctx, &ctx, sizeof(ctx))))
return (error);
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
/* Have race condition but it is cheap */
if ((p->p_numksegrps >= max_groups_per_proc) ||
(p->p_numthreads >= max_threads_per_proc)) {
return (EPROCLIM);
}
Close a race between thr_create and sysctl -w, the thr_scope_sys could be changed when thr_create is running, and we tested it for several times.
2004-10-06 02:29:19 +00:00
scope_sys = thr_scope_sys;
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
/* Initialize our td and new ksegrp.. */
newtd = thread_alloc();
Close a race between thr_create and sysctl -w, the thr_scope_sys could be changed when thr_create is running, and we tested it for several times.
2004-10-06 02:29:19 +00:00
if (scope_sys)
Give libthr a choice (per system) of scope_system or scope_thread scheduling. MFC after: 4 days
2004-09-07 06:33:39 +00:00
newkg = ksegrp_alloc();
else
newkg = kg;
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
/*
* Try the copyout as soon as we allocate the td so we don't have to
* tear things down in a failure case below.
*/
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
id = newtd->td_tid;
Change the thread ID (thr_id_t) used for 1:1 threading from being a pointer to the corresponding struct thread to the thread ID (lwpid_t) assigned to that thread. The primary reason for this change is that libthr now internally uses the same ID as the debugger and the kernel when referencing to a kernel thread. This allows us to implement the support for debugging without additional translations and/or mappings. To preserve the ABI, the 1:1 threading syscalls, including the umtx locking API have not been changed to work on a lwpid_t. Instead the 1:1 threading syscalls operate on long and the umtx locking API has not been changed except for the contested bit. Previously this was the least significant bit. Now it's the most significant bit. Since the contested bit should not be tested by userland, this change is not expected to be visible. Just to be sure, UMTX_CONTESTED has been removed from <sys/umtx.h>. Reviewed by: mtm@ ABI preservation tested on: i386, ia64
2004-07-02 00:40:07 +00:00
if ((error = copyout(&id, uap->id, sizeof(long)))) {
Close a race between thr_create and sysctl -w, the thr_scope_sys could be changed when thr_create is running, and we tested it for several times.
2004-10-06 02:29:19 +00:00
if (scope_sys)
Give libthr a choice (per system) of scope_system or scope_thread scheduling. MFC after: 4 days
2004-09-07 06:33:39 +00:00
ksegrp_free(newkg);
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
thread_free(newtd);
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
return (error);
}
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
bzero(&newtd->td_startzero,
Remove local definitions of RANGEOF() and use __rangeof() instead. Also remove a few bogus casts.
2004-11-20 23:00:59 +00:00
__rangeof(struct thread, td_startzero, td_endzero));
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
bcopy(&td->td_startcopy, &newtd->td_startcopy,
Remove local definitions of RANGEOF() and use __rangeof() instead. Also remove a few bogus casts.
2004-11-20 23:00:59 +00:00
__rangeof(struct thread, td_startcopy, td_endcopy));
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
Close a race between thr_create and sysctl -w, the thr_scope_sys could be changed when thr_create is running, and we tested it for several times.
2004-10-06 02:29:19 +00:00
if (scope_sys) {
Give libthr a choice (per system) of scope_system or scope_thread scheduling. MFC after: 4 days
2004-09-07 06:33:39 +00:00
bzero(&newkg->kg_startzero,
Remove local definitions of RANGEOF() and use __rangeof() instead. Also remove a few bogus casts.
2004-11-20 23:00:59 +00:00
__rangeof(struct ksegrp, kg_startzero, kg_endzero));
Give libthr a choice (per system) of scope_system or scope_thread scheduling. MFC after: 4 days
2004-09-07 06:33:39 +00:00
bcopy(&kg->kg_startcopy, &newkg->kg_startcopy,
Remove local definitions of RANGEOF() and use __rangeof() instead. Also remove a few bogus casts.
2004-11-20 23:00:59 +00:00
__rangeof(struct ksegrp, kg_startcopy, kg_endcopy));
Give libthr a choice (per system) of scope_system or scope_thread scheduling. MFC after: 4 days
2004-09-07 06:33:39 +00:00
}
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
newtd->td_proc = td->td_proc;
newtd->td_ucred = crhold(td->td_ucred);
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
/* Set up our machine context. */
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
cpu_set_upcall(newtd, td);
error = set_mcontext(newtd, &ctx.uc_mcontext);
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
if (error != 0) {
Close a race between thr_create and sysctl -w, the thr_scope_sys could be changed when thr_create is running, and we tested it for several times.
2004-10-06 02:29:19 +00:00
if (scope_sys)
Give libthr a choice (per system) of scope_system or scope_thread scheduling. MFC after: 4 days
2004-09-07 06:33:39 +00:00
ksegrp_free(newkg);
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
thread_free(newtd);
crfree(td->td_ucred);
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
goto out;
Change the thread ID (thr_id_t) used for 1:1 threading from being a pointer to the corresponding struct thread to the thread ID (lwpid_t) assigned to that thread. The primary reason for this change is that libthr now internally uses the same ID as the debugger and the kernel when referencing to a kernel thread. This allows us to implement the support for debugging without additional translations and/or mappings. To preserve the ABI, the 1:1 threading syscalls, including the umtx locking API have not been changed to work on a lwpid_t. Instead the 1:1 threading syscalls operate on long and the umtx locking API has not been changed except for the contested bit. Previously this was the least significant bit. Now it's the most significant bit. Since the contested bit should not be tested by userland, this change is not expected to be visible. Just to be sure, UMTX_CONTESTED has been removed from <sys/umtx.h>. Reviewed by: mtm@ ABI preservation tested on: i386, ia64
2004-07-02 00:40:07 +00:00
}
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
/* Link the thread and kse into the ksegrp and make it runnable. */
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
PROC_LOCK(td->td_proc);
Close a race between thr_create and sysctl -w, the thr_scope_sys could be changed when thr_create is running, and we tested it for several times.
2004-10-06 02:29:19 +00:00
if (scope_sys) {
Give libthr a choice (per system) of scope_system or scope_thread scheduling. MFC after: 4 days
2004-09-07 06:33:39 +00:00
sched_init_concurrency(newkg);
} else {
if ((td->td_proc->p_flag & P_HADTHREADS) == 0) {
sched_set_concurrency(kg,
thr_concurrency ? thr_concurrency : (2*mp_ncpus));
}
}
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
td->td_proc->p_flag |= P_HADTHREADS;
newtd->td_sigmask = td->td_sigmask;
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
mtx_lock_spin(&sched_lock);
Close a race between thr_create and sysctl -w, the thr_scope_sys could be changed when thr_create is running, and we tested it for several times.
2004-10-06 02:29:19 +00:00
if (scope_sys)
Give libthr a choice (per system) of scope_system or scope_thread scheduling. MFC after: 4 days
2004-09-07 06:33:39 +00:00
ksegrp_link(newkg, p);
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
thread_link(newtd, newkg);
mtx_unlock_spin(&sched_lock);
PROC_UNLOCK(p);
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
/* let the scheduler know about these things. */
mtx_lock_spin(&sched_lock);
Close a race between thr_create and sysctl -w, the thr_scope_sys could be changed when thr_create is running, and we tested it for several times.
2004-10-06 02:29:19 +00:00
if (scope_sys)
Give libthr a choice (per system) of scope_system or scope_thread scheduling. MFC after: 4 days
2004-09-07 06:33:39 +00:00
sched_fork_ksegrp(td, newkg);
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
sched_fork_thread(td, newtd);
- Call sched_exit_{kse,thread} and sched_fork{kse,thread} so that thr works with ULE. This was not strictly required by sched_4bsd.
2003-04-11 19:24:37 +00:00
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
TD_SET_CAN_RUN(newtd);
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
if ((uap->flags & THR_SUSPENDED) == 0)
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
setrunqueue(newtd, SRQ_BORING);
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
mtx_unlock_spin(&sched_lock);
out:
return (error);
}
int
thr_self(struct thread *td, struct thr_self_args *uap)
Change the thread ID (thr_id_t) used for 1:1 threading from being a pointer to the corresponding struct thread to the thread ID (lwpid_t) assigned to that thread. The primary reason for this change is that libthr now internally uses the same ID as the debugger and the kernel when referencing to a kernel thread. This allows us to implement the support for debugging without additional translations and/or mappings. To preserve the ABI, the 1:1 threading syscalls, including the umtx locking API have not been changed to work on a lwpid_t. Instead the 1:1 threading syscalls operate on long and the umtx locking API has not been changed except for the contested bit. Previously this was the least significant bit. Now it's the most significant bit. Since the contested bit should not be tested by userland, this change is not expected to be visible. Just to be sure, UMTX_CONTESTED has been removed from <sys/umtx.h>. Reviewed by: mtm@ ABI preservation tested on: i386, ia64
2004-07-02 00:40:07 +00:00
/* long *id */
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
{
Change the thread ID (thr_id_t) used for 1:1 threading from being a pointer to the corresponding struct thread to the thread ID (lwpid_t) assigned to that thread. The primary reason for this change is that libthr now internally uses the same ID as the debugger and the kernel when referencing to a kernel thread. This allows us to implement the support for debugging without additional translations and/or mappings. To preserve the ABI, the 1:1 threading syscalls, including the umtx locking API have not been changed to work on a lwpid_t. Instead the 1:1 threading syscalls operate on long and the umtx locking API has not been changed except for the contested bit. Previously this was the least significant bit. Now it's the most significant bit. Since the contested bit should not be tested by userland, this change is not expected to be visible. Just to be sure, UMTX_CONTESTED has been removed from <sys/umtx.h>. Reviewed by: mtm@ ABI preservation tested on: i386, ia64
2004-07-02 00:40:07 +00:00
long id;
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
int error;
Change the thread ID (thr_id_t) used for 1:1 threading from being a pointer to the corresponding struct thread to the thread ID (lwpid_t) assigned to that thread. The primary reason for this change is that libthr now internally uses the same ID as the debugger and the kernel when referencing to a kernel thread. This allows us to implement the support for debugging without additional translations and/or mappings. To preserve the ABI, the 1:1 threading syscalls, including the umtx locking API have not been changed to work on a lwpid_t. Instead the 1:1 threading syscalls operate on long and the umtx locking API has not been changed except for the contested bit. Previously this was the least significant bit. Now it's the most significant bit. Since the contested bit should not be tested by userland, this change is not expected to be visible. Just to be sure, UMTX_CONTESTED has been removed from <sys/umtx.h>. Reviewed by: mtm@ ABI preservation tested on: i386, ia64
2004-07-02 00:40:07 +00:00
id = td->td_tid;
if ((error = copyout(&id, uap->id, sizeof(long))))
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
return (error);
return (0);
}
int
thr_exit(struct thread *td, struct thr_exit_args *uap)
Close a race between a thread exiting and the freeing of it's stack. After some discussion the best option seems to be to signal the thread's death from within the kernel. This requires that thr_exit() take an argument. Discussed with: davidxu, deischen, marcel MFC after: 3 days
2004-10-06 14:23:00 +00:00
/* long *state */
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
{
struct proc *p;
p = td->td_proc;
Close a race between a thread exiting and the freeing of it's stack. After some discussion the best option seems to be to signal the thread's death from within the kernel. This requires that thr_exit() take an argument. Discussed with: davidxu, deischen, marcel MFC after: 3 days
2004-10-06 14:23:00 +00:00
/* Signal userland that it can free the stack. */
if ((void *)uap->state != NULL)
suword((void *)uap->state, 1);
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
PROC_LOCK(p);
mtx_lock_spin(&sched_lock);
/*
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
* Shutting down last thread in the proc. This will actually
* call exit() in the trampoline when it returns.
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
*/
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
if (p->p_numthreads != 1) {
thread_exit();
/* NOTREACHED */
}
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
mtx_unlock_spin(&sched_lock);
Refactor a bunch of scheduler code to give basically the same behaviour but with slightly cleaned up interfaces. The KSE structure has become the same as the "per thread scheduler private data" structure. In order to not make the diffs too great one is #defined as the other at this time. The KSE (or td_sched) structure is now allocated per thread and has no allocation code of its own. Concurrency for a KSEGRP is now kept track of via a simple pair of counters rather than using KSE structures as tokens. Since the KSE structure is different in each scheduler, kern_switch.c is now included at the end of each scheduler. Nothing outside the scheduler knows the contents of the KSE (aka td_sched) structure. The fields in the ksegrp structure that are to do with the scheduler's queueing mechanisms are now moved to the kg_sched structure. (per ksegrp scheduler private data structure). In other words how the scheduler queues and keeps track of threads is no-one's business except the scheduler's. This should allow people to write experimental schedulers with completely different internal structuring. A scheduler call sched_set_concurrency(kg, N) has been added that notifies teh scheduler that no more than N threads from that ksegrp should be allowed to be on concurrently scheduled. This is also used to enforce 'fainess' at this time so that a ksegrp with 10000 threads can not swamp a the run queue and force out a process with 1 thread, since the current code will not set the concurrency above NCPU, and both schedulers will not allow more than that many onto the system run queue at a time. Each scheduler should eventualy develop their own methods to do this now that they are effectively separated. Rejig libthr's kernel interface to follow the same code paths as linkse for scope system threads. This has slightly hurt libthr's performance but I will work to recover as much of it as I can. Thread exit code has been cleaned up greatly. exit and exec code now transitions a process back to 'standard non-threaded mode' before taking the next step. Reviewed by: scottl, peter MFC after: 1 week
2004-09-05 02:09:54 +00:00
PROC_UNLOCK(p);
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
return (0);
}
int
thr_kill(struct thread *td, struct thr_kill_args *uap)
Change the thread ID (thr_id_t) used for 1:1 threading from being a pointer to the corresponding struct thread to the thread ID (lwpid_t) assigned to that thread. The primary reason for this change is that libthr now internally uses the same ID as the debugger and the kernel when referencing to a kernel thread. This allows us to implement the support for debugging without additional translations and/or mappings. To preserve the ABI, the 1:1 threading syscalls, including the umtx locking API have not been changed to work on a lwpid_t. Instead the 1:1 threading syscalls operate on long and the umtx locking API has not been changed except for the contested bit. Previously this was the least significant bit. Now it's the most significant bit. Since the contested bit should not be tested by userland, this change is not expected to be visible. Just to be sure, UMTX_CONTESTED has been removed from <sys/umtx.h>. Reviewed by: mtm@ ABI preservation tested on: i386, ia64
2004-07-02 00:40:07 +00:00
/* long id, int sig */
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
{
struct thread *ttd;
struct proc *p;
int error;
p = td->td_proc;
error = 0;
PROC_LOCK(p);
style(9) o Remove double-spacing, and while I'm here add a couple of braces as well. Requested by: bde
2003-07-04 06:59:28 +00:00
FOREACH_THREAD_IN_PROC(p, ttd) {
Change the thread ID (thr_id_t) used for 1:1 threading from being a pointer to the corresponding struct thread to the thread ID (lwpid_t) assigned to that thread. The primary reason for this change is that libthr now internally uses the same ID as the debugger and the kernel when referencing to a kernel thread. This allows us to implement the support for debugging without additional translations and/or mappings. To preserve the ABI, the 1:1 threading syscalls, including the umtx locking API have not been changed to work on a lwpid_t. Instead the 1:1 threading syscalls operate on long and the umtx locking API has not been changed except for the contested bit. Previously this was the least significant bit. Now it's the most significant bit. Since the contested bit should not be tested by userland, this change is not expected to be visible. Just to be sure, UMTX_CONTESTED has been removed from <sys/umtx.h>. Reviewed by: mtm@ ABI preservation tested on: i386, ia64
2004-07-02 00:40:07 +00:00
if (ttd->td_tid == uap->id)
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
break;
style(9) o Remove double-spacing, and while I'm here add a couple of braces as well. Requested by: bde
2003-07-04 06:59:28 +00:00
}
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
if (ttd == NULL) {
error = ESRCH;
goto out;
}
if (uap->sig == 0)
goto out;
if (!_SIG_VALID(uap->sig)) {
error = EINVAL;
goto out;
}
Signals sent specifically to a particular thread must be delivered to that thread, regardless of whether it has it masked or not. Previously, if the targeted thread had the signal masked, it would be put on the processes' siglist. If another thread has the signal umasked or unmasks it before the target, then the thread it was intended for would never receive it. This patch attempts to solve the problem by requiring callers of tdsignal() to say whether the signal is for the thread or for the process. If it is for the process, then normal processing occurs and any thread that has it unmasked can receive it. But if it is destined for a specific thread, it is put on that thread's pending list regardless of whether it is currently masked or not. The new behaviour still needs more work, though. If the signal is reposted for some reason it is always posted back to the thread that handled it because the information regarding the target of the signal has been lost by then. Reviewed by: jdp, jeff, bde (style)
2003-07-03 19:09:59 +00:00
tdsignal(ttd, uap->sig, SIGTARGET_TD);
- Add two files to support the thr threading interface. - sys/thr.h contains the user space visible api that is intended only for use in threading library packages. - kern/kern_thr.c contains thr system calls and other thr specific code.
2003-04-01 00:30:30 +00:00
out:
PROC_UNLOCK(p);
return (error);
}
Separate thread synchronization from signals in libthr. Instead use msleep() and wakeup_one(). Discussed with: jhb, peter, tjr
2004-03-27 14:30:43 +00:00
int
thr_suspend(struct thread *td, struct thr_suspend_args *uap)
/* const struct timespec *timeout */
{
struct timespec ts;
struct timeval tv;
int error;
int hz;
hz = 0;
error = 0;
if (uap->timeout != NULL) {
error = copyin((const void *)uap->timeout, (void *)&ts,
sizeof(struct timespec));
if (error != 0)
return (error);
if (ts.tv_nsec < 0 || ts.tv_nsec > 1000000000)
return (EINVAL);
if (ts.tv_sec == 0 && ts.tv_nsec == 0)
return (ETIMEDOUT);
TIMESPEC_TO_TIMEVAL(&tv, &ts);
hz = tvtohz(&tv);
}
PROC_LOCK(td->td_proc);
writers must hold both sched_lock and the process lock; therefore, readers need only obtain the process lock.
2004-07-12 15:28:31 +00:00
if ((td->td_flags & TDF_THRWAKEUP) == 0)
Separate thread synchronization from signals in libthr. Instead use msleep() and wakeup_one(). Discussed with: jhb, peter, tjr
2004-03-27 14:30:43 +00:00
error = msleep((void *)td, &td->td_proc->p_mtx,
td->td_priority | PCATCH, "lthr", hz);
If a thread is resumed by thr_wake, it should return 0, especially it should not return ERESTART after it caught a signal, otherwise thr_wake() call will be lost, also a timeout wait should not be restarted. Final, using wakeup not wakeup_one to be safeness.
2004-12-01 13:50:04 +00:00
if (td->td_flags & TDF_THRWAKEUP) {
mtx_lock_spin(&sched_lock);
td->td_flags &= ~TDF_THRWAKEUP;
mtx_unlock_spin(&sched_lock);
PROC_UNLOCK(td->td_proc);
return (0);
}
Separate thread synchronization from signals in libthr. Instead use msleep() and wakeup_one(). Discussed with: jhb, peter, tjr
2004-03-27 14:30:43 +00:00
PROC_UNLOCK(td->td_proc);
If a thread is resumed by thr_wake, it should return 0, especially it should not return ERESTART after it caught a signal, otherwise thr_wake() call will be lost, also a timeout wait should not be restarted. Final, using wakeup not wakeup_one to be safeness.
2004-12-01 13:50:04 +00:00
if (error == EWOULDBLOCK)
error = ETIMEDOUT;
else if (error == ERESTART) {
if (hz != 0)
error = EINTR;
}
return (error);
Separate thread synchronization from signals in libthr. Instead use msleep() and wakeup_one(). Discussed with: jhb, peter, tjr
2004-03-27 14:30:43 +00:00
}
int
thr_wake(struct thread *td, struct thr_wake_args *uap)
Change the thread ID (thr_id_t) used for 1:1 threading from being a pointer to the corresponding struct thread to the thread ID (lwpid_t) assigned to that thread. The primary reason for this change is that libthr now internally uses the same ID as the debugger and the kernel when referencing to a kernel thread. This allows us to implement the support for debugging without additional translations and/or mappings. To preserve the ABI, the 1:1 threading syscalls, including the umtx locking API have not been changed to work on a lwpid_t. Instead the 1:1 threading syscalls operate on long and the umtx locking API has not been changed except for the contested bit. Previously this was the least significant bit. Now it's the most significant bit. Since the contested bit should not be tested by userland, this change is not expected to be visible. Just to be sure, UMTX_CONTESTED has been removed from <sys/umtx.h>. Reviewed by: mtm@ ABI preservation tested on: i386, ia64
2004-07-02 00:40:07 +00:00
/* long id */
Separate thread synchronization from signals in libthr. Instead use msleep() and wakeup_one(). Discussed with: jhb, peter, tjr
2004-03-27 14:30:43 +00:00
{
Change the thread ID (thr_id_t) used for 1:1 threading from being a pointer to the corresponding struct thread to the thread ID (lwpid_t) assigned to that thread. The primary reason for this change is that libthr now internally uses the same ID as the debugger and the kernel when referencing to a kernel thread. This allows us to implement the support for debugging without additional translations and/or mappings. To preserve the ABI, the 1:1 threading syscalls, including the umtx locking API have not been changed to work on a lwpid_t. Instead the 1:1 threading syscalls operate on long and the umtx locking API has not been changed except for the contested bit. Previously this was the least significant bit. Now it's the most significant bit. Since the contested bit should not be tested by userland, this change is not expected to be visible. Just to be sure, UMTX_CONTESTED has been removed from <sys/umtx.h>. Reviewed by: mtm@ ABI preservation tested on: i386, ia64
2004-07-02 00:40:07 +00:00
struct thread *ttd;
Separate thread synchronization from signals in libthr. Instead use msleep() and wakeup_one(). Discussed with: jhb, peter, tjr
2004-03-27 14:30:43 +00:00
If you're trying to find out if a thread is valid and in the same process as the current thread it makes absolutely no sense to lock the parent process through the pointer in said thread. Submitted by: pho (with minor correction) Pointy Hat To: mtm
2004-04-19 14:20:01 +00:00
PROC_LOCK(td->td_proc);
FOREACH_THREAD_IN_PROC(td->td_proc, ttd) {
Change the thread ID (thr_id_t) used for 1:1 threading from being a pointer to the corresponding struct thread to the thread ID (lwpid_t) assigned to that thread. The primary reason for this change is that libthr now internally uses the same ID as the debugger and the kernel when referencing to a kernel thread. This allows us to implement the support for debugging without additional translations and/or mappings. To preserve the ABI, the 1:1 threading syscalls, including the umtx locking API have not been changed to work on a lwpid_t. Instead the 1:1 threading syscalls operate on long and the umtx locking API has not been changed except for the contested bit. Previously this was the least significant bit. Now it's the most significant bit. Since the contested bit should not be tested by userland, this change is not expected to be visible. Just to be sure, UMTX_CONTESTED has been removed from <sys/umtx.h>. Reviewed by: mtm@ ABI preservation tested on: i386, ia64
2004-07-02 00:40:07 +00:00
if (ttd->td_tid == uap->id)
Separate thread synchronization from signals in libthr. Instead use msleep() and wakeup_one(). Discussed with: jhb, peter, tjr
2004-03-27 14:30:43 +00:00
break;
}
if (ttd == NULL) {
If you're trying to find out if a thread is valid and in the same process as the current thread it makes absolutely no sense to lock the parent process through the pointer in said thread. Submitted by: pho (with minor correction) Pointy Hat To: mtm
2004-04-19 14:20:01 +00:00
PROC_UNLOCK(td->td_proc);
Separate thread synchronization from signals in libthr. Instead use msleep() and wakeup_one(). Discussed with: jhb, peter, tjr
2004-03-27 14:30:43 +00:00
return (ESRCH);
}
mtx_lock_spin(&sched_lock);
Change the thread ID (thr_id_t) used for 1:1 threading from being a pointer to the corresponding struct thread to the thread ID (lwpid_t) assigned to that thread. The primary reason for this change is that libthr now internally uses the same ID as the debugger and the kernel when referencing to a kernel thread. This allows us to implement the support for debugging without additional translations and/or mappings. To preserve the ABI, the 1:1 threading syscalls, including the umtx locking API have not been changed to work on a lwpid_t. Instead the 1:1 threading syscalls operate on long and the umtx locking API has not been changed except for the contested bit. Previously this was the least significant bit. Now it's the most significant bit. Since the contested bit should not be tested by userland, this change is not expected to be visible. Just to be sure, UMTX_CONTESTED has been removed from <sys/umtx.h>. Reviewed by: mtm@ ABI preservation tested on: i386, ia64
2004-07-02 00:40:07 +00:00
ttd->td_flags |= TDF_THRWAKEUP;
Separate thread synchronization from signals in libthr. Instead use msleep() and wakeup_one(). Discussed with: jhb, peter, tjr
2004-03-27 14:30:43 +00:00
mtx_unlock_spin(&sched_lock);
If a thread is resumed by thr_wake, it should return 0, especially it should not return ERESTART after it caught a signal, otherwise thr_wake() call will be lost, also a timeout wait should not be restarted. Final, using wakeup not wakeup_one to be safeness.
2004-12-01 13:50:04 +00:00
wakeup((void *)ttd);
If you're trying to find out if a thread is valid and in the same process as the current thread it makes absolutely no sense to lock the parent process through the pointer in said thread. Submitted by: pho (with minor correction) Pointy Hat To: mtm
2004-04-19 14:20:01 +00:00
PROC_UNLOCK(td->td_proc);
Separate thread synchronization from signals in libthr. Instead use msleep() and wakeup_one(). Discussed with: jhb, peter, tjr
2004-03-27 14:30:43 +00:00
return (0);
}
Reference in New Issue Copy Permalink