Add a new sysctl, vfs.zfs.vol.unmap_enabled, which allows the system
administrator to toggle whether ZFS should ignore UNMAP requests.
Illumos issue:
5149 zvols need a way to ignore DKIOCFREE
Add tunable for number of metaslabs per vdev
(vfs.zfs.vdev.metaslabs_per_vdev). The default remains
at 200.
Illumos issue:
5161 add tunable for number of metaslabs per vdev
Make space_map_truncate() always do space_map_reallocate(). Without
this, setting space_map_max_blksz would cause panic for existing pool,
as dmu_objset_set_blocksize would fail if the object have multiple blocks.
Illumos issues:
5164 space_map_max_blksz causes panic, does not work
5165 zdb fails assertion when run on pool with recently-enabled
spacemap_histogram feature
Make this subcommand less FC-specific, reporting target and port addresses
in more generic way. Also make it report list of connected initiators in
unified way, working for both FC and iSCSI, and potentially others.
Queued async events handling in CAM opened race, that may lead to duplicate
AC_PATH_REGISTERED events delivery during boot. That was not happening
before r272935 because the driver was initialized later. After that change
it started create duplicate ports in CTL.
Target mode operation does not depend on the initiator mode scan process.
This change allows the target driver to attach earlier and receive some
async events (like AC_CONTRACT) that could be lost otherwise.
Make ctld start even if some LUNs are unable to open backing storage.
Such LUNs will be visible to initiators, but return "not ready" status
on media access commands. If backing storage become available later,
`ctladm modify ...` or `service ctld reload` can trigger its reopen.
Store persistent reservation keys as uint64_t instead of uint8_t[8].
This allows to simplify the code and save 512KB of RAM per LUN (8%)
by removing no longer needed "registered" keys flags.
It allows to push out some final data from the send queue to the socket
before its close. In particular, it increases chances for logout response
to be delivered to the initiator.
Add support for WRITE ATOMIC (16) command and report SBC-4 compliance.
Atomic writes are only supported for ZVOLs in "dev" mode. In other cases
atomicity can not be guarantied and so the command is blocked.
Add support for MaxBurstLength and Expected Data transfer Length parameters.
Before this change target could send R2T request for write transfer of any
size, that could violate iSCSI RFC, which allows initiator to limit maximum
R2T size by negotiating MaxBurstLength connection parameter.
Also report an error in case of write underflow, when initiator provides
less data than initiator expects. Previously in such case our target
sent R2T request for non-existing data, violating the RFC, and confusing
some initiators. SCSI specs don't explicitly define how write underflows
should be handled and there are different oppinions, but reporting error
is hopefully better then violating iSCSI RFC with unpredictable results.
Do not blindly announce 1000baseT half-duplex capability in
autonegotiation. Some controllers like cgem(4) do not support
half-duplex at gigabit speeds.
Improvements to asmc(4):
1. changed the code so that 2**16 keys are supported
2. changed the number of possible fans in a system from 2 to 6
3. added write support for some fan sysctls
4. added a new sysctl which shows the ID of the fan
5. added four more apple models with their temperature keys
6. changed the maxnumber of temperature keys from 36 to 80
7. replaced several fixed buf sizes to sizeof buf
Obtained from: Denis Ahrens denis at h3q.com
vt(4) is the default console for UEFI boot [1], and the bitmapped
kern.vt.spclkeys sysctl has been replaced with individual kern.vt.kbd_*
enable sysctls.
PR: 193710
1) Fix the case we have less arguments for format string than we expected.
2) Return error on unsupported format specs.
(both according to POSIX)
3) For %Z format, understand "UTC" name too.
PR: 93197
Fix the reported streams in a SCTP_STREAM_RESET_EVENT, if a
sent incoming stream reset request was responded with failed
or denied.
Thanks to Peter Bostroem from Google for reporting the issue.
Update the OFED Linux compatibility layer and
Mellanox hardware driver(s):
- Properly name an inclusion guard
- Fix compile warnings regarding unsigned enums
- Add two new sysctl nodes
- Remove all empty linux header files
- Make an error printout more verbose
- Use "mod_delayed_work()" instead of
cancelling and starting a timeout.
- Implement more Linux scatterlist
functions.
Sponsored by: Mellanox Technologies
He noticed issues setting this bit in SRRCTL after the queue was up,
so doing it from the sysctl handler isn't enough and may not actually
work correctly.
This commit doesn't remove the sysctl path or try to change its
behaviour. I'll talk with others about how to finish fixing that
before I tackle that.
PR: kern/194311
Submitted by: luigi
MFC after: 3 days
Sponsored by: Norse Corp, Inc
Rewrite timeout(9) to be callout(9)-centric instead. Move the description
of timeout(9) to the end and mark it prominently as deprecated. Document
somewhat how times are specified for the 'sbt' variants. Better explain
how using callout_init_*() to associate a lock with a callout resolves
common races.
Properly set the timeout in a query_state. The global query_timeout
configuration value is an integer count of seconds, it is not a timeval.
Using memcpy() to copy a timeval from it put garbage into the tv_usec
field.
PR: 194025
Don't panic if a resource is allocated twice. Instead, print a warning and
fail the allocation request. Allocations of "reserved" resources such as
PCI BARs already fail the request instead of panic'ing in this case.
r273093:
Merge the following from ^/projects/release-vmimage:
r272436, r272437, r272792:
r272436:
Remove the first argument to panic(), which was initially
intended to be the exit code, however when a non-zero exit
code was returned to release/Makefile, this would prevent
any remaining (and possibly successful) stages from being
attempted.
r272437:
If the vm-base target fails, prevent the vm-image target
from being run since it cannot possibly succeed.
r272792:
Add /usr/local/bin and /usr/local/sbin to PATH, needed
if third-party software needs to use utilities outside
of the base system during post-install stages (indexinfo
is one culprit).
r273096:
Merge the following from ^/projects/release-vmimage:
r273076, r273077, r273079, r273095:
r273076:
Add a separate make(1) target to release/Makefile to
build FreeBSD virtual machine disk images for use on
the Microsoft Azure service.
For now, this target is not directly connected to the
build, however can be manually invoked.
The 'vm-azure' target invokes {amd64,i386}/mk-azure.sh,
which does the heavy lifting to produce proper VHDs.
mk-azure.sh uses a configuration file, defaulting to
tools/azure.conf if otherwise unset.
r273077:
Clear VM_RC_LIST.
r273079:
Fix signal list to trigger umount(8).
r273095:
Output an informational message when mkimg(1) runs, so it
does not appear that the process has stopped while waiting
for a 'y/n' response when waagent is deprovisioned.
Relnotes: yes
Sponsored by: The FreeBSD Foundation
Don't inherit flags other than DS_FLAG_CI_DATASET and DS_FLAG_INCONSISTENT
when cloning. This prevents DS_FLAG_DEFER_DESTROY being inherited from a
clone that is marked for deferred destroy, which causes snapshots of the
clone being destroyed when getting a hold or clone.
Illumos issue:
5150 zfs clone of a defer_destroy snapshot causes strangeness
Use write_psize instead of write_asize when doing vdev_space_update.
Without this change the accounting of L2ARC usage would be wrong and
give 16EB free space because the number became negative and overflows.
Obtained from: FreeNAS (issue #6239)
