- Delete the per-entry lock, and borrow the associated vnode's v_interlock
instead. We need to acquire it during lookup anyway. We can revisit this
in the future but for now it's a stepping stone, and works within the
quite limited context of what we have (BSD namecache/lookup design).
- Implement an idea that Mateusz Guzik (mjg@FreeBSD.org) gave me. In
cache_reclaim(), we don't need to lock out all of the CPUs to garbage
collect entries. All we need to do is observe their locks unheld at least
once: then we know they are not in the critical section, and no longer
have visibility of the entries about to be garbage collected.
- The above makes it safe for sysctl to take only namecache_lock to get stats,
and we can remove all the crap dealing with per-CPU locks.
- For lockstat, make namecache_lock a static now we have __cacheline_aligned.
- Avoid false sharing - don't write back to nc_hittime unless it has changed.
Put a a comment in place explaining this. Pretty sure this was there in
2008/2009 but someone removed it (understandably, the code looks weird).
- Use a mutex to protect the garbage collection queue instead of atomics, and
adjust the low water mark up so that cache_reclaim() isn't doing so much
work at once.
- sched_tick: cpu_need_resched is no longer the correct thing to do here.
All we need to do is OR the request into the local ci_want_resched.
- sched_resched_cpu: we need to set RESCHED_UPREEMPT even on softint LWPs,
especially in the !__HAVE_FAST_SOFTINTS case, because the LWP with the
LP_INTR flag could be running via softint_overlay() - i.e. it has been
temporarily borrowed from a user process, and it needs to notice the
resched after it has stopped running softints.
- Don't do the calculation if there is a CRC error.
- If we get any kind of error during a refresh, retry up to three times.
- Add event counters to report what's going on.
- Re-do the signalling to be a little more forgiving and efficient.
- If bus reset fails during probe, try a second time.
- Spread out kernel threads for many busses to avoid thundering herd effect.
Since binutils 2.15, nm(1) cannot be used for character devices.
We worked around this by a local patch:
http://cvsweb.netbsd.org/bsdweb.cgi/src/gnu/dist/binutils/binutils/Attic/bucomm.c?r1=1.1.1.2&hideattic=0#rev1.2
With recent update of binutils, 'nm /dev/ksyms' got broken again.
This is due to a consistency check involving file size reported by
stat(2), which is always zero for character devices. So, skip this
check if file size is zero.
Explicitly clear status string to prevent dmsetup(8) from
printing binary junk to stdout. Similar change has been
applied to DragonFlyBSD since 2015.
a6cf54187f
taken-from: DragonFlyBSD
For .txt and .more drop meaningless page numbers, those formats are
not paginated (with all due respect to line printers, 80s are over).
For .ps use actual formatting with proportional font and properly
aligned page numbers. Thanks to kre@ for his help with some of the
darker corners of troff (with all due respect to my 80s self, he/I
should have paid more attention).
Guarantee no fusing and no tearing, and can optionally impose
ordering relative to other memory operations.
Unordered:
- atomic_load_relaxed
- atomic_store_relaxed
Ordered:
- atomic_load_acquire
- atomic_load_consume
- atomic_store_release
These are intended to match C11 semantics, and can be defined in
terms of the C11 atomic API when ready.
ad