Peter Avalos [Mon, 16 Jun 2008 08:12:40 +0000 (08:12 +0000)]
Merge from vendor branch LIBARCHIVE:
Import libarchive-2.5.4b.
Requested-by: hasso
Peter Avalos [Mon, 16 Jun 2008 08:12:40 +0000 (08:12 +0000)]
Import libarchive-2.5.4b.
Requested-by: hasso
Matthew Dillon [Mon, 16 Jun 2008 02:00:05 +0000 (02:00 +0000)]
We must process incoming IPI messages when spinning in the thread
cpu acquisition code. This avoids a deadlock which can occur between
the cpu whos thread is trying to exit and the cpu receiving the thread.
Reported-by: Gary Allan <dragonfly@gallan.co.uk>
Sascha Wildner [Sun, 15 Jun 2008 19:20:05 +0000 (19:20 +0000)]
The default value of net.inet6.ip6.v6only is 'on'.
Noticed-by: Jordan Gordeev <jgordeev@dir.bg>
Sepherosa Ziehau [Sun, 15 Jun 2008 12:15:08 +0000 (12:15 +0000)]
- In if_up(), purge ifnet.if_snd before turning on IFF_UP, this is used to
clean up packets that sneaked into ifnet.if_snd after if_down()
- Strip blank line
Sepherosa Ziehau [Sun, 15 Jun 2008 12:01:32 +0000 (12:01 +0000)]
- Don't turn on IFF_OACTIVE when dequeuing ifnet.if_snd; it no longer
makes sense
- If parent interface is NULL, purge ifnet.if_snd
Sepherosa Ziehau [Sun, 15 Jun 2008 11:41:40 +0000 (11:41 +0000)]
- Fix a possible race between getting vlan media and changing vlan parent
- White space cleanup
Sepherosa Ziehau [Sun, 15 Jun 2008 11:19:15 +0000 (11:19 +0000)]
Add missing return
Sepherosa Ziehau [Sun, 15 Jun 2008 10:41:00 +0000 (10:41 +0000)]
- Add FXP_USABLE_TXCB, which equals (FXP_NTXCB - 1)
- Turn off IFF_OACTIVE if queued TCBs is less than FXP_USABLE_TXCB
- Call if_devstart() in fxp_tick(), if ifnet.if_snd is not empty
- Remove unnecessary comment about ifnet.if_timer in fxp_init()
Sepherosa Ziehau [Sun, 15 Jun 2008 05:14:41 +0000 (05:14 +0000)]
Don't assume RV2P page size is always 4Kbytes; at least 8Kbytes page size
works too.
Sepherosa Ziehau [Sun, 15 Jun 2008 02:32:38 +0000 (02:32 +0000)]
Break long line
Sascha Wildner [Sat, 14 Jun 2008 20:09:18 +0000 (20:09 +0000)]
Remove the old FreeBSD logo screen saver.
It has not been built for >4 years now.
Matthias Schmidt [Sat, 14 Jun 2008 15:30:19 +0000 (15:30 +0000)]
Add missing ;
Reported-by: trevorjk
Sepherosa Ziehau [Sat, 14 Jun 2008 08:40:16 +0000 (08:40 +0000)]
In the final code block of in_control_internal, remove ia from the hash table
only if cmd is SIOCDIFADDR. Add assertions and comment about it.
Sepherosa Ziehau [Sat, 14 Jun 2008 07:58:46 +0000 (07:58 +0000)]
- Check cached ifnet.if_bridge again in bridge_{input,output}()
- Pass cached ifnet.if_bridge to bstp_input()
- Pass the located bif to bstp_input(), so we won't need to locate bif again
These changes make it possible to use following sequence
ifp->if_bridge = NULL;
netmsg_service_sync();
to protect ifnet.if_bridge
Sepherosa Ziehau [Sat, 14 Jun 2008 06:34:37 +0000 (06:34 +0000)]
Minor style change
Matthew Dillon [Sat, 14 Jun 2008 05:34:06 +0000 (05:34 +0000)]
Change the namecache lock warning delay from 1 to 5 seconds.
Sepherosa Ziehau [Sat, 14 Jun 2008 04:35:32 +0000 (04:35 +0000)]
Member iface's serializer does not need to be held when calling bridge_output()
Sepherosa Ziehau [Sat, 14 Jun 2008 04:00:51 +0000 (04:00 +0000)]
Remove unused parameters to bridge_output
Matthew Dillon [Sat, 14 Jun 2008 01:44:13 +0000 (01:44 +0000)]
HAMMER Utilities: Sync with commit 55 - MEDIA STRUCTURES CHANGED!
* newfs_hammer now creates a new blockmap zone for meta-data
* The hammer utility has been enhanced and brought up-to-date. One
can now reblock directories separately from other data.
* The hammer show command now properly understands the B-Tree element
localization field (fixed improper reporting of 'B'ad B-Tree elements).
Matthew Dillon [Sat, 14 Jun 2008 01:42:13 +0000 (01:42 +0000)]
HAMMER 55: Performance tuning and bug fixes - MEDIA STRUCTURES CHANGED!
* BUG-FIX: Fix a race in hammer_rel_mem_record() which could result in a
machine lockup. The code could block at an inappropriate time with both
the record and a dependancy inode pointer left unprotected.
* BUG-FIX: The direct-write code could assert on (*error != 0) due to an
incorrect conditional in the in-memory record scanning code.
* Inode data and directory entry data has been given its own zone as a
stop-gap until the low level allocator can be rewritten.
* Increase the directory object-id cache from 128 entries to 1024 entries.
* General cleanup.
* Introduce a separate reblocking domain for directories: 'hammer reblock-dirs'.
Sepherosa Ziehau [Fri, 13 Jun 2008 12:30:19 +0000 (12:30 +0000)]
Strip "serialized" suffix
Matthew Dillon [Fri, 13 Jun 2008 00:25:33 +0000 (00:25 +0000)]
HAMMER 54D/Many: Performance tuning.
* Remove major barriers to write performance and fix hicups revealed by
blogbench.
Change the HAMMER reclaim-delay algorithm to operate like a FIFO instead
of as a free-for-all. The idea of introducing a dynamic delay helped some,
but the addition of the wakeup FIFO allows burst completions by the flusher
to immediately wakeup processes that were waiting for the reclaim count to
drain. The result is far, far smoother operation.
* Remove a major blocking conflict between the buffer cache daemon and
HAMMER. The buffer cache was getting stuck on trying to overwrite dirty
records that had already been queued to the flusher. The flusher might
not act on the record(s) for a long period of time, causing the buffer
cache daemon to stall.
Fix the problem by properly using the HAMMER_RECF_INTERLOCK_BE flag,
which stays on only for a very short period of time, instead of testing
the record's flush state (record->flush_state), which can stay in
the HAMMER_FST_FLUSH state for a very long time.
* The parent B-Tree node does not need to be locked when inserting
into the child.
* Use the new B_AGE semantics to keep meta-data intact longer. This results
in another big improvement in random read and write performance.
Matthew Dillon [Fri, 13 Jun 2008 00:25:01 +0000 (00:25 +0000)]
LWPHOLD/LWPRELE must be atomic ops because an IPI can call LWPRELE.
Matthew Dillon [Thu, 12 Jun 2008 23:26:37 +0000 (23:26 +0000)]
Reimplement B_AGE. Have it cycle the buffer in the queue twice instead of
placing buffers at the head of the queue (which causes them to be run-down
backwards). Leave B_AGE set through the write cycle and have the bufdaemon
set the flag when flushing dirty buffers. B_AGE no longer effects the
ordering of the actual write and is allowed to slide through to the clean
queue when the write completes.
Matthew Dillon [Thu, 12 Jun 2008 23:25:02 +0000 (23:25 +0000)]
Add missing LWPHOLD/LWPRELE in kinfo code.
Hasso Tepper [Thu, 12 Jun 2008 07:09:31 +0000 (07:09 +0000)]
Fix probable cut-n-paste error.
Obtained-from: NetBSD
Matthew Dillon [Thu, 12 Jun 2008 01:55:58 +0000 (01:55 +0000)]
HAMMER 54C/Many: Performing tuning, bug fixes
* Adjust the reclaims delay code
* Fix a direct-read bug. Truncations were not being properly handled.
Matthew Dillon [Thu, 12 Jun 2008 00:16:10 +0000 (00:16 +0000)]
HAMMER 54B/Many: Performance tuning.
* Fix a major performance issue in the UNDO generation code. The code
was calling hammer_bread() instead of hammer_bnew() for 'new' undo buffers,
meaning it was doing a read-modify-write on the disk instead of just a
write.
This fix results in a MAJOR improvement in performance across the board.
* Replace the only lockmgr lock in the module with a hammer_lock.
* Tweak hammer_inode_waitreclaims(). This will probably need even more
tweaking as time passes.
Matthew Dillon [Wed, 11 Jun 2008 22:33:21 +0000 (22:33 +0000)]
HAMMER 54/Many: Performance tuning
* Implement hammer_vop_bmap().
* Implement cluster_read() support. This should jump up linear read
performance almost to the platter speed. I get 100 MB/sec testing
vs 35 MB/sec previously.
* Do a better job kicking an inode into the flusher when writing
sequentially. This hops up write rate at least +50%. It isn't
quite able to run at the platter speed due to B-Tree overheads
which will be addressed in a later patch.
* Do not create data fragments at the ends of files greater then 16K, use
a full 16K block. The reason is that fragments in HAMMER are allocated
out of a wholely different zone and we do not want to lose the chance of
making the tail end of the file contiguous.
Files less then 16K still use data fragments.
* Fix a machine lockup related to an interrupt race with biodone() and
insertions and deltions from hmp->lose_list.
* Fix a memory exhaustion issue.
Reported-by: Francois Tigeot <ftigeot@wolfpond.org> (machine lockup)
Credit-also: Jonathan Stuart on the 0 byte sized file bug fix.
Sascha Wildner [Wed, 11 Jun 2008 20:01:31 +0000 (20:01 +0000)]
Fix a path in the comment.
Matthew Dillon [Tue, 10 Jun 2008 22:30:21 +0000 (22:30 +0000)]
HAMMER 53H/Many: Performance tuning, bug fixes
* CHANGE THE ON-MEDIA B-TREE STRUCTURE. The number of elements per node has
been increased from 16 to 64. The intent is to reduce the number of seeks
required in a heavy random-access loading situation.
* Add a shortcut to the B-Tree node scanning code (requires more testing).
Instead of scanning linearly we do a power-of-2 narrowing search.
* Only do clustered reads for DATA types. Do not cluster meta-data (aka
B-Tree) I/O. Note that the inode data structure is considered to be
a DATA type. Reduce the cluster read side from 256K to 64K to avoid
blowing out the buffer cache.
* Augment hammer locks so one can discern between a normal lock blockage
and one that is recovering from a deadlock.
* Change the slave work threads for the flusher to pull their work off a
single queue. This fixes an issue where one slave work thread would
sometimes get a disproportionate percentage of the work and the
master thread then had to wait for it to finish while the other work
threads were twiddling their thumbs.
* Adjust the wait reclaims code to solve a long standing performance issue.
The flusher could get so far behind that the system's buffer cache buffers
would no longer have any locality of reference to what was being flushed,
causing a massive drop in performance.
* Do not queue a dirty inode to the flusher unconditionally in the strategy
write code. Only do it if system resources appear to be stressed.
The inode will get flushed when the filesystem syncs.
* Code cleanup.
* Fix a bug reported by Antonio Huete Jimenez related to 0-length writes
not working properly.
Matthew Dillon [Tue, 10 Jun 2008 17:20:52 +0000 (17:20 +0000)]
Switch from bioq_insert_tail() to bioqdisksort(). When the kernel is
juggling hundreds of I/O requests the on-disk controllers, which have limited
queue sizes, are unable to completely optimize the accesses.
This seems to make a pretty big difference on the twe device I tested with
when running blogbench on HAMMER, which can approach 1000 simultaniously
queued I/O's.
Hasso Tepper [Tue, 10 Jun 2008 10:08:11 +0000 (10:08 +0000)]
Make sure host controller interrupts are not enabled until all
initialisation is finished.
Obtained-from: NetBSD
Hasso Tepper [Tue, 10 Jun 2008 10:04:06 +0000 (10:04 +0000)]
Remove the code which disables port status change interrupts for 1 second
when one occured. It meant that event get lost or delayed until the next
change leaving the port dead under some circumstances.
Obtained-from: NetBSD
Matthew Dillon [Tue, 10 Jun 2008 08:51:02 +0000 (08:51 +0000)]
HAMMER 53G/Many: Performance tuning.
* Implement multiple flusher threads.
* Move the call to hammer_inode_waitreclaims() yet again. Call it from
hammer_vop_reclaim().
Matthew Dillon [Tue, 10 Jun 2008 08:06:28 +0000 (08:06 +0000)]
HAMMER 53F/Many: Fix deadlock.
* Do not sleep waiting for reclaimed inodes to drain while holding
an inode lock.
Hasso Tepper [Tue, 10 Jun 2008 06:35:59 +0000 (06:35 +0000)]
Remove superfluous recursive lock. This little change makes possible (safe)
to use spinlock(9) instead of lockmgr(9).
Nicolas Thery [Tue, 10 Jun 2008 05:20:10 +0000 (05:20 +0000)]
Fix ordering mismatch between KERN_* values and CTL_KERN_NAMES array.
Matthew Dillon [Tue, 10 Jun 2008 05:06:20 +0000 (05:06 +0000)]
HAMMER 53E/Many: Performance tuning
* Change the code which waits for reclaims to drain to be more inline with
the new bwillwrite(). Impose a dynamic delay instead of blocking outright.
* Move the hammer_inode_waitreclaims() call from hammer_vop_open() to
hammer_get_inode(), and only call it when we would otherwise have to
create a new inode.
* Sort HAMMER's file list in conf/files.
Matthew Dillon [Tue, 10 Jun 2008 05:02:09 +0000 (05:02 +0000)]
Change bwillwrite() to smooth out performance under heavy loads. Blocking
based on strict hystersis was being used to try to gang flushes together
but filesystems can still blow out the buffer cache and cause processes
to block for long periods of time waiting for the dirty count to drop
significantly.
Instead, as the number of dirty buffers exceeds the desired maximum
bwillwrite() imposes a dynamic delay which increases as the number of
dirty buffers increase. This improves the stall behavior under heavy loads
and keeps the system responsive.
TODO: The algorithm needs to have a per-LWP heuristic to penalize heavy
writers more then light ones.
Matthew Dillon [Tue, 10 Jun 2008 00:40:31 +0000 (00:40 +0000)]
HAMMER 53D/Many: Stabilization
* Fix an overwrite bug with direct write which could result in file
corruption.
* Reserve just-freed big blocks for two flush cycles to prevent HAMMER from
overwriting destroyed data so it does not become corrupt if the system
crashes. This is needed because the recover code does not record UNDOs
for data (nor do we want it to).
* More I/O subsystem work. There may still be an ellusive panic related
to calls to regetblk().
Matthew Dillon [Mon, 9 Jun 2008 16:54:45 +0000 (16:54 +0000)]
Add an extern for hidirtybuffers.
Reported-by: Michael Neumann <mneumann@ntecs.de>
Sepherosa Ziehau [Mon, 9 Jun 2008 11:24:24 +0000 (11:24 +0000)]
Parallelize in_ifaddrhashtbl
Sepherosa Ziehau [Mon, 9 Jun 2008 04:59:22 +0000 (04:59 +0000)]
Move stack variable near its usage
Matthew Dillon [Mon, 9 Jun 2008 04:33:08 +0000 (04:33 +0000)]
Fix a SMP race in signotify_remote(). LWPHOLD() the lwp being
passed from one cpu to another over an IPI to prevent it from exiting
in the mean time and causing the IPI to panic.
Matthew Dillon [Mon, 9 Jun 2008 04:19:10 +0000 (04:19 +0000)]
HAMMER 53C/Many: Stabilization
* HAMMER queues dirty inodes reclaimed by the kernel to the backend for
their final sync. Programs like blogbench can overload the backend and
generate more new inodes then the backend can dispose of, running
M_HAMMER out of memory.
Add code to stall on vop_open() when this condition is detected to
give the backend a chance to catch-up. (see NOTE 1 below).
* HAMMER could build up too many meta-data buffers and cause the system
to deadlock in newbuf. Recode the flusher to allow a block of UNDOs,
the volume header, and all related meta-data buffers to be flushed
piecemeal, and then continue the flush loop without closing out the
transaction. If a crash occurs the recovery code will undo the partial
flushes.
* Fix an issue located by FSX under load. The in-memory/on-disk record
merging code was not dealing with in-memory data records properly
The key field for data records is (base_offset + data_len), not just
(base_off), so a 'match' between an in-memory data record and an on-disk
data records requires a special case test. This is the case where the
in-memory record is intended to overwrite the on-disk record, so the
in-memory record must be chosen and the on-disk record discarded for
the purposes of read().
* Fix a bug in hammer_io.c related to the handling of B_LOCKED buffers
that resulted in an assertion at umount time. Buffer cache buffers
were not being properly disassociated from their hammer_buffer countparts
in the direct-write case.
* The frontend's direct-write capability for truncated buffers (such as
used with small files) was causing an assertion to occur on the backend.
Add an interlock on the related hammer_buffer to prevent the frontend
from attempting to modify the buffer while the backend is trying to
write it to the media.
* Dynamically size the dirty buffer limit. This still needs some work.
(NOTE 1): On read/write performance issues. Currently HAMMER's frontend
VOPs are massively disassociated from modifying B-Tree updates. Even though
a direct-write capability now exists, it applies only to bulk data writes
to disk and NOT to B-Tree updates. Each direct write creates a record
which must be queued to the backend to do the B-Tree update on the
media. The flusher is currently single-threaded and when HAMMER gets
too far behind doing these updates the current safeties will cause
performance to degrade drastically. This is a known issue that
will be addressed.
Sascha Wildner [Sun, 8 Jun 2008 20:40:38 +0000 (20:40 +0000)]
Fix typo.
Sascha Wildner [Sun, 8 Jun 2008 20:40:15 +0000 (20:40 +0000)]
Fix typos.
Matthew Dillon [Sun, 8 Jun 2008 18:16:26 +0000 (18:16 +0000)]
HAMMER 53B/Many: Complete overhaul of strategy code, reservations, etc
* Completely overhaul the strategy code. Implement direct reads and writes
for all cases. REMOVE THE BACKEND BIO QUEUE. BIOs are no longer queued
to the flusher under any circumstances.
Remove numerous hacks that were previously emplaced to deal with BIO's
being queued to the flusher.
* Add a mechanism to invalidate buffer cache buffers that might be shadowed
by direct I/O. e.g. if a strategy write uses the vnode's bio directly
there may be a shadow hammer_buffer that will then become stale and must
be invalidated.
* Implement a reservation tracking structure (hammer_reserve) to track
storage reservations made by the frontend. The backend will not attempt
to free or reuse reserved space if it encounters it.
Use reservations to back cached holes (struct hammer_hole) for the
same reason.
* Index hammer_buffer on the zone-X offset instead of the zone-2 offset.
Base the RB tree in the hammer_mount instead of (zone-2) hammer_volume.
This removes nearly all blockmap lookup operations from the critical path.
* Do a much better job tracking cached dirty data for the purposes of
calculating whether the filesystem will become full or not.
* Fix a critical bug in the CRC generation of short data buffers.
* Fix a VM deadlock.
* Use 16-byte alignment for all on-disk data instead of 8-byte alignment.
* Major code cleanup.
As-of this commit write performance is now extremely good.
Matthew Dillon [Sun, 8 Jun 2008 17:19:09 +0000 (17:19 +0000)]
HAMMER Utilities: Critical bug in newfs_hammer
* newfs_hammer was not properly setting up the small-data zone.
Sepherosa Ziehau [Sun, 8 Jun 2008 10:06:05 +0000 (10:06 +0000)]
Add tunable to enable/disable PBCC support in acx(4) and it is enabled
by default.
Sepherosa Ziehau [Sun, 8 Jun 2008 08:38:06 +0000 (08:38 +0000)]
Parallelize in_ifaddrhead operation
Nicolas Thery [Sun, 8 Jun 2008 07:56:06 +0000 (07:56 +0000)]
Assert that move in directory entry hash table can't fail.
Sepherosa Ziehau [Sun, 8 Jun 2008 03:58:03 +0000 (03:58 +0000)]
- oia is no longer used
- Tranform for(;;) loop into while() loop
Michael Neumann [Sat, 7 Jun 2008 12:30:26 +0000 (12:30 +0000)]
Move fetching of "hw.hasbrokenint12" tunable closer to it's usage.
Michael Neumann [Sat, 7 Jun 2008 12:15:33 +0000 (12:15 +0000)]
Cosmetic changes (remove whitespace).
Michael Neumann [Sat, 7 Jun 2008 12:03:52 +0000 (12:03 +0000)]
Remove unnecessary conversion to kilobytes (divide by 1024) to then later
multiply it again by 1024 to get to bytes.
Michael Neumann [Sat, 7 Jun 2008 11:44:04 +0000 (11:44 +0000)]
Use NULL instead of 0.
Michael Neumann [Sat, 7 Jun 2008 11:37:23 +0000 (11:37 +0000)]
Correct typos.
Matthew Dillon [Sat, 7 Jun 2008 07:41:51 +0000 (07:41 +0000)]
HAMMER 53A/Many: Read and write performance enhancements, etc.
* Add hammer_io_direct_read(). For full-block reads this code allows
a high-level frontend buffer cache buffer associated with the
regular file vnode to directly access the underlying storage,
instead of loading that storage via a hammer_buffer and bcopy()ing it.
* Add a write bypass, allowing the frontend to bypass the flusher and
write full-blocks directly to the underlying storage, greatly improving
frontend write performance. Caveat: See note at bottom.
The write bypass is implemented by adding a feature whereby the frontend
can soft-reserve unused disk space on the physical media without having
to interact (much) with on-disk meta-data structures. This allows the
frontend to flush high-level buffer cache buffers directly to disk
and release the buffer for reuse by the system, resulting in very high
write performance.
To properly associate the reserved space with the filesystem so it can be
accessed in later reads, an in-memory hammer_record is created referencing
it. This record is queued to the backend flusher for final disposition.
The backend disposes of the record by inserting the appropriate B-Tree
element and marking the storage as allocated. At that point the storage
becomes official.
* Clean up numerous procedures to support the above new features. In
particular, do a major cleanup of the cached truncation offset code
(this is the code which allows HAMMER to implement wholely asynchronous
truncate()/ftruncate() support.
Also clean up the flusher triggering code, removing numerous hacks that
had been in place to deal with the lack of a direct-write mechanism.
* Start working on statistics gathering to track record and B-Tree
operations.
* CAVEAT: The backend flusher creates a significant cpu burden when flushing
a large number of in-memory data records. Even though the data itself
has already been written to disk, there is currently a great deal of
overhead involved in manipulating the B-Tree to insert the new records.
Overall write performance will only be modestly improved until these
code paths are optimized.
Sepherosa Ziehau [Sat, 7 Jun 2008 07:22:22 +0000 (07:22 +0000)]
Use ASSERT_IFAC_VALID whenever possible
Sepherosa Ziehau [Sat, 7 Jun 2008 06:34:57 +0000 (06:34 +0000)]
Add ASSERT_IFAC_VALID
Sepherosa Ziehau [Sat, 7 Jun 2008 04:59:01 +0000 (04:59 +0000)]
- Expose ifa_forwardmsg()
- Add ifa_domsg()
# They will be needed soon
Sascha Wildner [Fri, 6 Jun 2008 13:19:25 +0000 (13:19 +0000)]
Don't use NULL where 0 is meant.
Sepherosa Ziehau [Fri, 6 Jun 2008 12:35:27 +0000 (12:35 +0000)]
Make sure that ifac is still valid before unlinking it from or linking it to ifnet
Sepherosa Ziehau [Fri, 6 Jun 2008 10:47:14 +0000 (10:47 +0000)]
Add periodic rf calibration support for acx111 part. This seems to stablize
performance during long time TX stress.
Sascha Wildner [Thu, 5 Jun 2008 18:06:33 +0000 (18:06 +0000)]
* Fix some cases where NULL was used but 0 was meant (and vice versa).
* Remove some bogus casts of NULL to (void *).
Sascha Wildner [Thu, 5 Jun 2008 18:01:49 +0000 (18:01 +0000)]
Remove some unneeded definitions of NULL.
Sascha Wildner [Thu, 5 Jun 2008 17:53:10 +0000 (17:53 +0000)]
Include <sys/_null.h> for the definition of NULL.
Sascha Wildner [Thu, 5 Jun 2008 17:49:53 +0000 (17:49 +0000)]
Add <sys/_null.h> for the definition of NULL:
[...]
#ifndef __cplusplus
#define NULL ((void *)0)
#else
#define NULL 0
#endif
[...]
Sepherosa Ziehau [Thu, 5 Jun 2008 15:29:47 +0000 (15:29 +0000)]
Add rt_cpuid, which records rtentry's owning CPU id. It could ease route
entry related debugging and sanity checks.
Nicolas Thery [Wed, 4 Jun 2008 04:34:54 +0000 (04:34 +0000)]
Fix bugs in spin_trylock_wr():
- globaldata.gd_spinlock_wr was not decremented back on failure;
- incorrect comparison in loop trying to clear cached shared bits (loop
must fail if spinlock is still held for read by another cpu).
Reviewed-by: dillon@
Matthew Dillon [Tue, 3 Jun 2008 18:47:25 +0000 (18:47 +0000)]
HAMMER 52/Many: Read-only mounts and mount upgrades/downgrades.
* Finish implementing MNT_UPDATE, allowing a HAMMER mount to be upgraded
or downgraded.
* Adjust the recovery code to not flush buffers dirtied by recovery
operations (running the UNDOs) when the mount is read-only. The
buffers will be flushed when the mount is updated to read-write.
* Improve recovery performance by not flushing dirty buffers until the
end (if a read-write mount).
* A crash which occurs during recovery might cause the next recovery to
fail. Delay writing out the recovered volume header until all the other
buffers have been written out to fix the problem.
Matthew Dillon [Tue, 3 Jun 2008 18:43:34 +0000 (18:43 +0000)]
HAMMER Utilities: Enhance mount_hammer
* Allow devices to be specified as dev:dev:dev, so a multi-volume hammer
mount can be specified in /etc/fstab.
* Implement -u (mount update)
Matthew Dillon [Tue, 3 Jun 2008 16:16:40 +0000 (16:16 +0000)]
Do not update f_offset on EINVAL.
Reported-by: VOROSKOI Andras <voroskoi@gmail.com>
Sascha Wildner [Tue, 3 Jun 2008 12:40:09 +0000 (12:40 +0000)]
mdoc cleanup
Sascha Wildner [Tue, 3 Jun 2008 09:33:27 +0000 (09:33 +0000)]
Fix a crash when "Arctic Ocean" was selected.
Taken-from: FreeBSD
Matthew Dillon [Tue, 3 Jun 2008 06:20:30 +0000 (06:20 +0000)]
HAMMER Utilities: More pre-formatting, cleanup
* Fully initialize the large-data and small-data blockmaps in addition
to the B-Tree blockmap.
* Set vol0_stat_bigblocks properly so used space shows as 0, or otherwise
a fairly small number, when the volume is empty.
* Display the total amount of space pre-allocated by newfs_hammer for
the boot-area, memory-log, undo-buffer, and blockmap infrastructure.
Sascha Wildner [Mon, 2 Jun 2008 20:40:07 +0000 (20:40 +0000)]
Add 'options HAMMER' to LINT.
Noticed-by: Dionysus Blazakis <dion.blazakis@gmail.com>
Matthew Dillon [Mon, 2 Jun 2008 20:19:03 +0000 (20:19 +0000)]
HAMMER 51/Many: Filesystem full casework, nohistory flag.
* Track the amount of unsynced information and return ENOSPC if the
filesystem would become full. The idea here is to detect that the
filesystem is full and yet still give the flusher enough runway to
flush cached dirty data and inodes.
* Implement the NOHISTORY flag. Implement inheritance of NOHISTORY and
NODUMP.
The NOHISTORY flag tells HAMMER not to retain historical information on
a filesystem object. If set on a directory any objects created in that
directory will also inherit the flag. For example, it could be set
on /usr/obj.
Matthew Dillon [Mon, 2 Jun 2008 20:17:07 +0000 (20:17 +0000)]
Report the nohistory, noshistory, and nouhistory flags, and allow them
to be specified with chflags.
Set the inverted field for nosunlink, nosunlnk, nouunlink, nouulnk. This
is an inverted flag.
Matthew Dillon [Mon, 2 Jun 2008 20:13:38 +0000 (20:13 +0000)]
Add the UF_NOHISTORY and SF_NOHISTORY chflags flags. The nohistory flag
allows you to mount a HAMMER filesystem normally and still specify that
certain files and directory not retain historical information.
If set on a directory the flag will be inherited by any objects
created within that directory.
Adjust UFS to inherit the NOHISTORY and NODUMP flags from the parent
directory when creating new objects. NODUMP used to be ufs-specific
and UFS's backup/restore remembered the inheritance. As a more generic
flag it needs to be inherited within the filesystem. Note that UFS
has no historical retention capability and ignores the NOHISTORY flag,
but we might use it with the journaling audit trail later.
Matthew Dillon [Mon, 2 Jun 2008 20:06:36 +0000 (20:06 +0000)]
Disallow negative seek positions for regular files, directories, and
character-special devices to conform to OpenGroup specifications.
Reported-by: VOROSKOI Andras <voroskoi@gmail.com>
Matthew Dillon [Mon, 2 Jun 2008 20:03:22 +0000 (20:03 +0000)]
Add missing exit(1).
Reported-by: Johannes Hofmann <hofmann@blob.baaderstrasse.com>
Matthew Dillon [Mon, 2 Jun 2008 16:57:53 +0000 (16:57 +0000)]
HAMMER Utilities: Correct vol0_stat_freebigblocks.
* The root_vol->ondisk->vol0_stat_freebigblocks was not being properly
decremented when newfs_hammer allocated big blocks, causing it to report
a value that is too large.
Matthew Dillon [Mon, 2 Jun 2008 16:55:08 +0000 (16:55 +0000)]
Fix kernel compile warnings.
Matthew Dillon [Mon, 2 Jun 2008 16:54:21 +0000 (16:54 +0000)]
Even using the objcache we need a one-per-cpu free-thread cache in order
to keep an exiting thread intact throughout its exit sequence.
This fixes a double-fault which can occur on shutdown. The bug was mainly
tickled by exiting kernel threads.
Hasso Tepper [Mon, 2 Jun 2008 06:50:08 +0000 (06:50 +0000)]
According to SUSv3 including just regex.h must be enough. Fixes build of
several pkgsrc packages.
Hasso Tepper [Mon, 2 Jun 2008 06:42:45 +0000 (06:42 +0000)]
Unbreak buildworld.
Matthew Dillon [Sun, 1 Jun 2008 21:05:39 +0000 (21:05 +0000)]
HAMMER 50/Many: VFS_STATVFS() support, stabilization.
* Add support for VFS_STATVFS(), returning 64 bit quantities for available
space, etc.
* When freeing a big-block any holes cached for that block must be
cleaned out.
* Fix a conditional testing whether a layer2 big-block must be allocated in
layer1. The bug could only occur if a layer2 big-block gets freed in
layer1, and we currently never do this.
* Clean-up comments related to freeing blocks.
Matthew Dillon [Sun, 1 Jun 2008 20:59:29 +0000 (20:59 +0000)]
HAMMER Utilities: Performance adjustments, bug fixes.
* Newfs_hammer now pre-allocates the layer1 and layer2 blockmap blocks,
and pre-sizes each blockmap to 4x the initial filesystem size instead
of 100x the initial filesystem size.
The blockmap can be dynamically resized at any time, given a little code.
In addition, there is simply no need to give it a 100x initial dynamic
range. This only bloats the size of the layer-2 map unnecessarily.
* Change alloc_blockmap() to use rootmap->next_offset for allocations
instead of rootmap->alloc_offset and fix a bug where rootmap->phys_offset
was improperly being incremented (it is a fixed field once set).
The bug was in a code-path that could not by executed by current
incarnations of newfs_hammer.
Matthew Dillon [Sun, 1 Jun 2008 20:52:21 +0000 (20:52 +0000)]
Use newly available libc and system calls related to statvfs to make df
work with 64 bit statvfs fields. Tested with a 4TB VN-backed HAMMER
filesystem.
Matthew Dillon [Sun, 1 Jun 2008 20:46:45 +0000 (20:46 +0000)]
Add getmntvinfo() which uses the new getvfsstat() system call.
Matthew Dillon [Sun, 1 Jun 2008 20:44:45 +0000 (20:44 +0000)]
More header file cleanups related statvfs.
Matthew Dillon [Sun, 1 Jun 2008 20:18:03 +0000 (20:18 +0000)]
Clean up statvfs() and related prototypes. Place the prototypes in the
correct file.
Matthew Dillon [Sun, 1 Jun 2008 19:55:32 +0000 (19:55 +0000)]
Implement a new system call: getvfsstat(). This system call returns
an array of statfs and statvfs structures. Unfortunately there is no way
to just return an array of statvfs structures because the statvfs structure
does not have sufficient information in it to identify the mount point.
getvfsstat(struct statfs *buf, struct statvfs *vbuf,
long vbufsize, int flags);
Matthew Dillon [Sun, 1 Jun 2008 19:27:37 +0000 (19:27 +0000)]
* Implement new system calls in the kernel: statvfs(), fstatvfs(),
fhstatvfs().
* Implement a new VFS op, VFS_STATVFS(). Implement a default for this new
op for VFSs which do not implement VFS_STATVFS(), which calls VFS_STATFS()
and converts the structure (using Joerg's conversion procedure from libc).
* Remove statvfs(), fstatvfs(), and fhstatvfs() from libc. These functions
are now system calls.
Sascha Wildner [Sun, 1 Jun 2008 08:54:18 +0000 (08:54 +0000)]
Raise the size of the /etc MFS to 12MB (for ssh blacklists).
Sepherosa Ziehau [Sun, 1 Jun 2008 08:09:14 +0000 (08:09 +0000)]
- Rename ifa_portfn() to ifnet_portfn()
- Create inline function ifa_portfn() which simply calls ifnet_portfn()