| History log of /src/sys/kern/subr_pserialize.c |
| Revision | | Date | Author | Comments |
| 1.24 |
| 04-Oct-2023 |
ad | Eliminate l->l_ncsw and l->l_nivcsw. From memory think they were added
before we had per-LWP struct rusage; the same is now tracked there.
|
| 1.23 |
| 16-Apr-2023 |
riastradh | pserialize(9): Fix buggy assertion inside assertion.
If there has been a context switch, then we can't be in a pserialize
read section -- _and_ whatever we read out of curcpu() is stale, so
we can't assert anything based on it. So just delete the assertion.
Reported-by: syzbot+d7065ab16d7cd9cc09af@syzkaller.appspotmail.com
https://syzkaller.appspot.com/bug?id=0c63db890575d29c6a51fb24b003d3bab1f7f04b
|
| 1.22 |
| 14-Apr-2023 |
riastradh | pserialize(9): Fix bug in recent micro-optimization.
curcpu() is always a valid struct cpu_info pointer, but curlwp->l_cpu
is not necessarily -- early at boot on some architectures (like x86,
apparently) it may be null.
Not clear that l->l_cpu is always a cost win over curcpu(): on some
architectures curcpu() is basically a macro for curlwp->l_cpu, while
on others it is a fixed VA. So let's use the one that always works
instead of the one that's sometimes cheaper and sometimes crashier.
|
| 1.21 |
| 09-Apr-2023 |
riastradh | pserialize(9): Micro-optimize pserialize_read_exit.
Most of the time we're not cold, so let's convince gcc to make that a
statically predicted-not-taken branch.
|
| 1.20 |
| 09-Apr-2023 |
riastradh | pserialize(9): Micro-optimize pserialize_not_in_read_section_p.
Load l_ncsw to test whether we have been preempted, rather than
loading and storing l_nopreempt (via function call) to prevent it.
|
| 1.19 |
| 15-Nov-2022 |
macallan | don't KASSERT(kpreempt_disabled()) while cold
pserialize_read_*() can be called *very* early in kernel startup these days
|
| 1.18 |
| 10-Oct-2021 |
riastradh | pserialize(9): Lift rule that pserialize_perform be serialized.
There may have been a technical reason for this back when we were
following the expired passive serialization patent to the letter, but
no more -- and this is a real burden for some applications.
|
| 1.17 |
| 05-Dec-2019 |
riastradh | Switch psz_ev_excl to static evcnt.
|
| 1.16 |
| 05-Dec-2019 |
riastradh | Restore psz_lock just for the event count.
Cost of mutex_enter/exit is negligible compared to the xcall we just
did, so this is not going to meaningfully affect performance.
|
| 1.15 |
| 03-Dec-2019 |
martin | Stopgap hack to unbreak the build: #ifdef __HAVE_ATOMIC64_LOADSTORE
the event counter update. From rmind@
|
| 1.14 |
| 03-Dec-2019 |
riastradh | Rip out pserialize(9) logic now that the RCU patent has expired.
pserialize_perform() is now basically just xc_barrier(XC_HIGHPRI).
No more tentacles throughout the scheduler. Simplify the psz read
count for diagnostic assertions by putting it unconditionally into
cpu_info.
From rmind@, tidied up by me.
|
| 1.13 |
| 06-Oct-2019 |
uwe | xc_barrier - convenience function to xc_broadcast() a nop.
Make the intent more clear and also avoid a bunch of (xcfunc_t)nullop
casts that gcc 8 -Wcast-function-type is not happy about.
|
| 1.12 |
| 14-Aug-2018 |
ozaki-r | Change the place to check if a context switch doesn't happen within a pserialize read section
The previous place (pserialize_switchpoint) was not a good place because at that
point a suspect thread is already switched so that a backtrace gotten on
a KASSERT failure doesn't point out where a context switch happens.
|
| 1.11 |
| 16-Apr-2018 |
hannken | branches: 1.11.2;
Function pserialize_perform() usually succeeds after two cross calls
so defer kpause() to iterations three and above.
Speeds up VOP_REVOKE() on /proc/XXX/status by a factor of ~12.
Ok: core@
|
| 1.10 |
| 28-Dec-2017 |
msaitoh | branches: 1.10.2;
Prevent panic or hangup in softint_disestablish(), pserialize_perform() or
psref_target_destroy() while mp_online == false.
See http://mail-index.netbsd.org/tech-kern/2017/12/25/msg022829.html
|
| 1.9 |
| 21-Nov-2017 |
ozaki-r | Implement debugging feature for pserialize(9)
The debugging feature detects violations of pserialize constraints.
It causes a panic:
- if a context switch happens in a read section, or
- if a sleepable function is called in a read section.
The feature is enabled only if LOCKDEBUG is on.
Discussed on tech-kern@
|
| 1.8 |
| 12-Jun-2015 |
dholland | branches: 1.8.10;
Typo in comment.
|
| 1.7 |
| 07-Feb-2013 |
rmind | branches: 1.7.14;
- pserialize_switchpoint: check for passing twice, not more than needed.
- pserialize_perform: avoid a possible race with softint handler.
Reported by hannken@.
|
| 1.6 |
| 07-Jan-2013 |
rmind | - softint_dispatch: perform pserialize(9) switchpoint when softintr processing
finishes (without blocking). Problem reported by hannken@, thanks!
- pserialize_read_enter: use splsoftserial(), not splsoftclock().
- pserialize_perform: add xcall(9) barrier as interrupts may be coalesced.
|
| 1.5 |
| 29-Jan-2012 |
rmind | branches: 1.5.2; 1.5.6; 1.5.8;
- 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().
|
| 1.4 |
| 07-Aug-2011 |
rmind | branches: 1.4.2; 1.4.6;
- Add an argument to kcpuset_create() for zeroing.
- Add kcpuset_atomic_set(), kcpuset_atomic_clear() and kcpuset_merge().
|
| 1.3 |
| 07-Aug-2011 |
rmind | 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@
|
| 1.2 |
| 01-Aug-2011 |
he | Need to include <sys/evcnt.h> explicitly to build this on at least amiga.
|
| 1.1 |
| 30-Jul-2011 |
christos | Add an implementation of passive serialization as described in expired
US patent 4809168. This is a reader / writer synchronization mechanism,
designed for lock-less read operations.
|
| 1.4.6.1 |
| 18-Feb-2012 |
mrg | merge to -current.
|
| 1.4.2.3 |
| 22-May-2014 |
yamt | sync with head.
for a reference, the tree before this commit was tagged
as yamt-pagecache-tag8.
this commit was splitted into small chunks to avoid
a limitation of cvs. ("Protocol error: too many arguments")
|
| 1.4.2.2 |
| 23-Jan-2013 |
yamt | sync with head
|
| 1.4.2.1 |
| 17-Apr-2012 |
yamt | sync with head
|
| 1.5.8.2 |
| 11-Feb-2013 |
riz | Pull up following revision(s) (requested by rmind in ticket #811):
sys/kern/subr_pserialize.c: revision 1.7
- pserialize_switchpoint: check for passing twice, not more than needed.
- pserialize_perform: avoid a possible race with softint handler.
Reported by hannken@.
|
| 1.5.8.1 |
| 08-Feb-2013 |
riz | Pull up following revision(s) (requested by rmind in ticket #782):
sys/rump/include/machine/intr.h: revision 1.19
sys/kern/subr_pserialize.c: revision 1.6
sys/kern/kern_softint.c: revision 1.39
- softint_dispatch: perform pserialize(9) switchpoint when softintr processing
finishes (without blocking). Problem reported by hannken@, thanks!
- pserialize_read_enter: use splsoftserial(), not splsoftclock().
- pserialize_perform: add xcall(9) barrier as interrupts may be coalesced.
Provide splsoftserial.
GRRR RUMP
|
| 1.5.6.2 |
| 03-Dec-2017 |
jdolecek | update from HEAD
|
| 1.5.6.1 |
| 25-Feb-2013 |
tls | resync with head
|
| 1.5.2.2 |
| 11-Feb-2013 |
riz | Pull up following revision(s) (requested by rmind in ticket #811):
sys/kern/subr_pserialize.c: revision 1.7
- pserialize_switchpoint: check for passing twice, not more than needed.
- pserialize_perform: avoid a possible race with softint handler.
Reported by hannken@.
|
| 1.5.2.1 |
| 08-Feb-2013 |
riz | Pull up following revision(s) (requested by rmind in ticket #782):
sys/rump/include/machine/intr.h: revision 1.19
sys/kern/subr_pserialize.c: revision 1.6
sys/kern/kern_softint.c: revision 1.39
- softint_dispatch: perform pserialize(9) switchpoint when softintr processing
finishes (without blocking). Problem reported by hannken@, thanks!
- pserialize_read_enter: use splsoftserial(), not splsoftclock().
- pserialize_perform: add xcall(9) barrier as interrupts may be coalesced.
Provide splsoftserial.
GRRR RUMP
|
| 1.7.14.1 |
| 22-Sep-2015 |
skrll | Sync with HEAD
|
| 1.8.10.3 |
| 17-Apr-2018 |
martin | Pull up following revision(s) (requested by hannken in ticket #771):
sys/kern/subr_pserialize.c: revision 1.11
Function pserialize_perform() usually succeeds after two cross calls
so defer kpause() to iterations three and above.
Speeds up VOP_REVOKE() on /proc/XXX/status by a factor of ~12.
Ok: core@
|
| 1.8.10.2 |
| 22-Jan-2018 |
martin | Pull up following revision(s) (requested by jdolecek in ticket #506):
sys/kern/kern_softint.c: revision 1.45
sys/rump/librump/rumpkern/rump.c: revision 1.331
sys/kern/subr_pserialize.c: revision 1.10
sys/kern/subr_psref.c: revision 1.10
Prevent panic or hangup in softint_disestablish(), pserialize_perform() or
psref_target_destroy() while mp_online == false.
See http://mail-index.netbsd.org/tech-kern/2017/12/25/msg022829.html
Set mp_online = true. This change might fix PR#52886.
|
| 1.8.10.1 |
| 30-Nov-2017 |
martin | Pull up following revision(s) (requested by ozaki-r in ticket #405):
sys/sys/pserialize.h: revision 1.2
sys/kern/kern_lock.c: revision 1.160
sys/kern/subr_pserialize.c: revision 1.9
sys/rump/librump/rumpkern/emul.c: revision 1.184
sys/rump/librump/rumpkern/emul.c: revision 1.185
sys/rump/librump/rumpkern/rump.c: revision 1.330
Implement debugging feature for pserialize(9)
The debugging feature detects violations of pserialize constraints.
It causes a panic:
- if a context switch happens in a read section, or
- if a sleepable function is called in a read section.
The feature is enabled only if LOCKDEBUG is on.
Discussed on tech-kern@
Add missing inclusion of pserialize.h (fix build)
|
| 1.10.2.2 |
| 06-Sep-2018 |
pgoyette | Sync with HEAD
Resolve a couple of conflicts (result of the uimin/uimax changes)
|
| 1.10.2.1 |
| 22-Apr-2018 |
pgoyette | Sync with HEAD
|
| 1.11.2.2 |
| 13-Apr-2020 |
martin | Mostly merge changes from HEAD upto 20200411
|
| 1.11.2.1 |
| 10-Jun-2019 |
christos | Sync with HEAD
|
