Commit Graph

49 Commits

Author SHA1 Message Date
ad
0eaaa024ea Move proc_lock into the data segment. It was dynamically allocated because
at the time we had mutex_obj_alloc() but not __cacheline_aligned.
2020-05-23 23:42:41 +00:00
thorpej
6925a27fb3 Sanitize the pid and lid arguments passed to do_sched_getparam()
and sys__sched_getaffinity() now that -1 as the pid argument to
lwp_find2() means "wildcard proc".
2020-04-29 01:53:48 +00:00
ad
f2686a49e2 Remove comment that is out of date and I think hinting at something other
than what it says (preemption case for SCHED_FIFO).
2020-01-27 22:05:10 +00:00
christos
b265873d52 Fix reversed test. 2016-07-30 15:38:17 +00:00
msaitoh
8bc54e5be6 KNF. Remove extra spaces. No functional change. 2016-07-07 06:55:38 +00:00
christos
7cf7644fc7 GSoC 2016 Charles Cui: Implement thread priority protection based on work
by Andy Doran. Also document the get/set pshared thread calls as not
implemented, and add a skeleton implementation that is disabled.
XXX: document _sched_protect(2).
2016-07-03 14:24:58 +00:00
pooka
4f6fb3bf35 Ensure that the top level sysctl nodes (kern, vfs, net, ...) exist before
the sysctl link sets are processed, and remove redundancy.

Shaves >13kB off of an amd64 GENERIC, not to mention >1k duplicate
lines of code.
2014-02-25 18:30:08 +00:00
rmind
0c79472223 - Convert x86 MD code, mainly pmap(9) e.g. TLB shootdown code, to use
kcpuset(9) and thus replace hardcoded CPU bitmasks.  This removes the
  limitation of maximum CPUs.

- Support up to 256 CPUs on amd64 architecture by default.

Bug fixes, improvements, completion of Xen part and testing on 64-core
AMD Opteron(tm) Processor 6282 SE (also, as Xen HVM domU with 128 CPUs)
by Manuel Bouyer.
2012-04-20 22:23:24 +00:00
yamt
6e7d55c554 - do_sched_getparam: release locks earlier.
- add comments
2012-04-13 15:27:13 +00:00
rmind
ad12c77015 Remove COMPAT_SA / KERN_SA. Welcome to 6.99.3!
Approved by core@.
2012-02-19 21:05:51 +00:00
rmind
f76667381c - Add mi_cpu_init() and initialise cpu_lock and kcpuset_attached/running there.
- Add kcpuset_running which gets set in idle_loop().
- Use kcpuset_running in pserialize_perform().
2012-01-29 22:55:40 +00:00
rmind
e71c0035e7 - Add an argument to kcpuset_create() for zeroing.
- Add kcpuset_atomic_set(), kcpuset_atomic_clear() and kcpuset_merge().
2011-08-07 21:38:32 +00:00
rmind
501dd321fb Remove LW_AFFINITY flag and fix some bugs affinity mask handling. 2011-08-07 21:13:05 +00:00
rmind
52b220e91d Add kcpuset(9) - a reworked dynamic CPU set implementation for kernel.
Suitable for use during the early boot.  MD and other implementations
should be replaced with this interface.

Discussed on: tech-kern@
2011-08-07 13:33:01 +00:00
rmind
3c507045e2 Remove pfind() and pgfind(), fix locking in various broken uses of these.
Rename real routines to proc_find() and pgrp_find(), remove PFIND_* flags
and have consistent behaviour.  Provide proc_find_raw() for special cases.
Fix memory leak in sysctl_proc_corename().

COMPAT_LINUX: rework ptrace() locking, minimise differences between
different versions per-arch.

Note: while this change adds some formal cosmetics for COMPAT_DARWIN and
COMPAT_IRIX - locking there is utterly broken (for ages).

Fixes PR/43176.
2010-07-01 02:38:26 +00:00
elad
b2f3768346 - Move sched_listener and co. from kern_synch.c to sys_sched.c, where it
really belongs (suggested by rmind@),

- Rename sched_init() to synch_init(), and introduce a new sched_init()
  in sys_sched.c where we (a) initialize the sysctl node (no more
  link-set) and (b) listen on the process scope with sched_listener.

Reviewed by and okay rmind@.
2009-10-03 22:32:56 +00:00
rmind
4f1720c349 lwp_create: fix the locking bugs on affinity ingerition path (mea culpa).
pset_assign: traverse the list of LWPs safely.
sched_setaffinity: free cpuset (unused path) outside the lock.

Reviewed (with feedback) by <ad>.
2009-03-03 21:55:06 +00:00
rmind
909e7f4259 - Make thread-affinity and processor-set interfaces mutually exlusive.
- pset_assign: when CPU is assigned, migrate out all LWPs from it.
2009-01-20 01:57:35 +00:00
rmind
8f1873ea3b - Avoid the race with CPU online/offline state changes, when setting the
affinity (cpu_lock protects these operations now).
- Disallow setting of state of CPU to to offline, if there are bound LWPs,
  which have no CPU to migrate.
- Disallow setting of affinity for the LWP(s), if all CPUs in the dynamic
  CPU-set are offline.
- sched_setaffinity: fix invalid check of kcpuset_isset().
- Rename cpu_setonline() to cpu_setstate().

Should fix PR/39349.
2008-10-31 00:36:22 +00:00
rmind
a8552a3a74 Obviously intention was to check for SCHED_OTHER, not SCHED_FIFO. 2008-10-18 19:24:04 +00:00
rmind
d5ea013e1d Disallow user priority adjustments for SCHED_OTHER policy, simplify
convert_pri().  Sync schedctl(8) with the change.  Closes PR/38009.
2008-10-18 03:40:18 +00:00
wrstuden
fc7511b00e Merge wrstuden-revivesa into HEAD. 2008-10-15 06:51:17 +00:00
rmind
4f91cff093 - Disallow setting of affinity for zombie LWPs.
- Fix the possible NULL dereference when LWP exiting.
- Fix the inhertance of affinity.
2008-07-14 01:19:37 +00:00
christos
1d875fc75f Adjust to separate kcpuset_t and cpuset_t. 2008-06-22 00:06:36 +00:00
rmind
481ae1556f - Add general cpuset macros.
- Use kcpuset name for kernel-only functions.
- Use cpuid_t to specify CPU ID.
- Unify all cpuset users.

API is expected to be stable now.
2008-06-16 01:41:20 +00:00
rmind
a6092d3c4a gencpuset: avoid returning of unitialized error.
sched_setaffinity: instead of E2BIG, return EINVAL in a case of invalid
CPU, which is more consistent.
2008-06-15 23:29:09 +00:00
christos
f30b5785d5 Don't expose struct cpuset, share the l_affinity flag and only allocate it
if we need to. This is not a compatible change, but the syscalls are new
enough and they don't need to be versioned. Approved by rmind.
2008-06-15 20:32:57 +00:00
ad
1cc9a3ae7e If converting a process/thread from SCHED_OTHER to a realtime thread,
ignore the existing priority. If no priority is specified, give threads
the minimum user RT priority.
2008-05-25 23:34:24 +00:00
ad
284c2b9aef Merge proc::p_mutex and proc::p_smutex into a single adaptive mutex, since
we no longer need to guard against access from hardware interrupt handlers.

Additionally, if cloning a process with CLONE_SIGHAND, arrange to have the
child process share the parent's lock so that signal state may be kept in
sync. Partially addresses PR kern/37437.
2008-04-24 18:39:20 +00:00
ad
6d70f903e6 Network protocol interrupts can now block on locks, so merge the globals
proclist_mutex and proclist_lock into a single adaptive mutex (proc_lock).
Implications:

- Inspecting process state requires thread context, so signals can no longer
  be sent from a hardware interrupt handler. Signal activity must be
  deferred to a soft interrupt or kthread.

- As the proc state locking is simplified, it's now safe to take exit()
  and wait() out from under kernel_lock.

- The system spends less time at IPL_SCHED, and there is less lock activity.
2008-04-24 15:35:27 +00:00
njoly
425fc32da6 s/Minimal/Maximal/ in kern.sched.pri_max sysctl node description. 2008-03-05 12:47:13 +00:00
elad
67470a76a1 Factor out the guts of get/setparam so it can be used from the compat code.
Make the FreeBSD and Linux compat code convert the parameters to their
native representation and call the native routines.

Remove KAUTH_PROCESS_SCHEDULER_GET/SET.

Update documentation and examples.

XXX: For now, only the Linux compat code does the priority conversion
XXX: right.

Linux priority conversion code from yamt@, thanks!

Okay yamt@.
2008-02-28 16:09:18 +00:00
ad
9a0b455f25 - Add a one line comment indicating what this file does.
- Move the check for PK_SYSTEM to a more sane place.
2008-02-22 23:10:12 +00:00
rmind
9850c0557d sys__sched_getparam and sys__sched_getaffinity: Do not assume that LWP
with LID=1 exists, use LIST_FIRST(&p->p_lwps) instead.
Fixes PR/37987 by <yamt>.

While here, adjust license.
2008-02-22 22:32:49 +00:00
drochner
cc0caabca7 fix sched_setparam() error reporting (broken by recent kauth changes) 2008-02-19 19:38:18 +00:00
yamt
aa57485fe2 sys__sched_setparam: fix a lock leak. 2008-02-19 09:44:26 +00:00
yamt
146e2f8297 sys__sched_setparam: use a correct lwp. 2008-02-18 02:00:08 +00:00
elad
0bb7f5ccb3 PR/37986: YAMAMOTO Takashi: any user can hog the all cpu with
_sched_setparam.

Pass proper context to kauth(9) for a decision to made based on the
scheduling policy and priority.
2008-02-17 19:22:35 +00:00
elad
e99760e7e4 Fold KAUTH_REQ_PROCESS_SCHEDULER_* to KAUTH_PROCESS_SCHEDULER_*. In other
words, don't pass an action and a request, and just use a single action to
indicate what is the operation in question.

This is the first step in fixing PR/37986, which calls for policy/priority
checking in the secmodel code. Right now we're lacking room for another
parameter required to make a decision, and this change makes room for such.
2008-02-16 16:39:34 +00:00
yamt
16b042cbdb from rmind@,
- revert struct sched_param changes to restore ABI.
- instead, add 'policy' arguments to _sched_{get,set}param syscalls.
  (this is an API/ABI change.)
- correct kauth_authorize_process arguments.

while i'm here,
- don't bother to kmem_alloc for 4-byte structure.
2008-02-09 16:58:01 +00:00
elad
17bc624c06 EACCES -> EPERM. 2008-01-31 01:21:17 +00:00
elad
fb37bad459 Use proper kauth(9) actions/requests for native scheduler stuff and the
recently introduced processor-sets.

Discussed with and okay rmind@, yamt@, and christos@.
2008-01-30 17:54:55 +00:00
rmind
b5e9addd22 sched_setparam: fix the case when incorrect (according to the class)
in-kernel priority is used.  Reported by <drochner>.

Minor fixes for scheduling calls to conform the POSIX:
- If pid is equal to zero, use the calling process;
- In case of permission problem, return EPERM instead of EACESS;
- sched_setscheduler() should return previously used policy;
- pthread_* calls should return the error code or zero;

Should fix the namespace problems (and builds of some packages):
- Move cpuset_t defintion from pset.h to sched.h;
- Remove the #include of pset.h in pthread.h;
2008-01-26 17:55:29 +00:00
rmind
aa633f0365 Fix the return values to conform the standard and/or be compatible with
other systems.  Reported by <drochner>.
2008-01-24 14:41:12 +00:00
rmind
5c71a4d49f Implementation of processor-sets, affinity and POSIX real-time extensions.
Add schedctl(8) - a program to control scheduling of processes and threads.

Notes:
- This is supported only by SCHED_M2;
- Migration of LWP mechanism will be revisited;

Proposed on: <tech-kern>. Reviewed by: <ad>.
2008-01-15 03:37:10 +00:00
dsl
7e2790cf6f Convert all the system call entry points from:
int foo(struct lwp *l, void *v, register_t *retval)
to:
    int foo(struct lwp *l, const struct foo_args *uap, register_t *retval)
Fixup compat code to not write into 'uap' and (in some cases) to actually
pass a correctly formatted 'uap' structure with the right name to the
next routine.
A few 'compat' routines that just call standard ones have been deleted.
All the 'compat' code compiles (along with the kernels required to test
build it).
98% done by automated scripts.
2007-12-20 23:02:38 +00:00
ad
b07ec3fc38 Merge newlock2 to head. 2007-02-09 21:55:00 +00:00
dogcow
8f871f1a8d #include <sys/sa.h> to get sys_upcall_t def. 2007-01-16 20:12:40 +00:00
ad
606e323b3e Add sched_yield(). 2007-01-16 05:28:45 +00:00