imp [Mon, 9 Oct 2017 22:12:46 +0000 (22:12 +0000)]
Define LIBSA* and use them instead of overloaded LIBSTAND
LIBSA is the current stand alone library. LIBSA32 is the 32-bit
version of the library. LIBSAU is the userboot version of libsa. Use
the proper define instead of the more generic define.
imp [Mon, 9 Oct 2017 22:12:32 +0000 (22:12 +0000)]
Define SASRC and use it
Define SASRC to point to the current libstand sources. Include
../Makefile.inc early enough in a few places so we can .include
"${SASRC}/Makefile" and have it work. Create a new pass-up
Makefile.inc in sys/boot/userboot to allow this pattern to work.
glebius [Mon, 9 Oct 2017 21:06:16 +0000 (21:06 +0000)]
Improvements to sendfile(2) mbuf free routine.
o Fall back to default m_ext free mech, using function pointer in
m_ext_free, and remove sf_ext_free() called directly from mbuf code.
Testing on modern CPUs showed no regression.
o Provide internally used flag EXT_FLAG_SYNC, to mark that I/O uses
SF_SYNC flag. Lack of the flag allows us not to dereference
ext_arg2, saving from a cache line miss.
o Create function sendfile_free_page() that later will be used, for
multi-page mbufs. For now compiler will inline it into
sendfile_free_mext().
In collaboration with: gallatin
Differential Revision: https://reviews.freebsd.org/D12615
glebius [Mon, 9 Oct 2017 20:51:58 +0000 (20:51 +0000)]
In mb_dupcl() don't copy full m_ext, to avoid cache miss. Respectively,
in mb_free_ext() always use fields from the original refcount holding
mbuf (see. r296242) mbuf. Cuts another cache miss from mb_free_ext().
However, treat EXT_EXTREF mbufs differently, since they are different -
they don't have a refcount holding mbuf.
Provide longer comments in m_ext declaration to explain this change
and change from r296242.
In collaboration with: gallatin
Differential Revision: https://reviews.freebsd.org/D12615
glebius [Mon, 9 Oct 2017 20:35:31 +0000 (20:35 +0000)]
Shorten list of arguments to mbuf external storage freeing function.
All of these arguments are stored in m_ext, so there is no reason
to pass them in the argument list. Not all functions need the second
argument, some don't even need the first one. The second argument
lives in next cache line, so not dereferencing it is a performance
gain. This was discovered in sendfile(2), which will be covered by
next commits.
The second goal of this commit is to bring even more flexibility
to m_ext mbufs, allowing to create more fields in m_ext, opaque to
the generic mbuf code, and potentially set and dereferenced by
subsystems.
hselasky [Mon, 9 Oct 2017 18:33:29 +0000 (18:33 +0000)]
When showing the sleepqueues from the in-kernel debugger,
properly dump all the sendqueues and not just the first one
History:
It appears that in the commit which introduced the code,
r165272, the array indexes of "sq_blocked[0]" and "td_name[i]"
were interchanged. In r180927 "td_name[i]" was corrected to
"td_name[0]", but "sq_blocked[0]" was left unchanged.
alc [Mon, 9 Oct 2017 18:19:06 +0000 (18:19 +0000)]
The recent change to initialization of blists (r324420) relied on '-1'
appearing only where the code explicitly set it, but since much of the
data was not initialized, '-1' appeared other places too, and led to
panics. Clear the allocated data before initializing nonzero values by
allocating with M_ZERO.
Submitted by: Doug Moore <dougm@rice.edu>
Reported by: Oleg V. Nauman <oleg@theweb.org.ua>, cy
Tested by: Oleg V. Nauman <oleg@theweb.org.ua>
MFC after: 1 week
X-MFC with: r324420
Differential Revision: https://reviews.freebsd.org/D12627
kib [Mon, 9 Oct 2017 16:20:39 +0000 (16:20 +0000)]
Change amd64_get_ldt() to return 'EOF' when the LDT is not yet
allocated, when requested range of descriptors does not fit into
currently allocated LDT, or trim the return if the range fits
partially. Before, the function returned EINVAL.
Reviewed by: bde
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
kib [Mon, 9 Oct 2017 16:19:26 +0000 (16:19 +0000)]
Change i386_get_ldt() to return 'EOF' when the requested range of
descriptors does not fit into currently allocated LDT, or trim the
return if the range fits partially. Before, the function returned
EINVAL.
Fix two bugs in r324366: use capped num counter for malloc size, and
do not leak allocated buffer on EINVAL (by handling EINVAL case as
normal, see above).
Reviewed by: bde
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
kib [Mon, 9 Oct 2017 16:07:27 +0000 (16:07 +0000)]
Improvements to set_user_ldt().
Remove mtx_owned() checks from set_user_ldt(). Split the function
into _locked() version which requires the dt_lock spinlock owned, and
make set_user_ldt() a wrapper. Add a comment in swtch.s noting that
the call to the new set_user_ldt() cannot recurse on dt_lock.
Remove #ifdef SMP block, the addend is always zero on UP.
Fix type of set_user_ldt_rv(), making it match the type used for
smb_rendezvous() callback, and remove the cast. Use curproc.
Reviewed by: bde
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
kib [Mon, 9 Oct 2017 15:39:43 +0000 (15:39 +0000)]
Reset the fs and gs bases on exec(2).
The values from the old address space do not make sense for the new
program. In particular, gsbase might be the TLS base for the old
program but the new program has no TLS now.
amd64 already handles this correctly.
Reported and reviewed by: bde
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
kevans [Mon, 9 Oct 2017 14:50:02 +0000 (14:50 +0000)]
patch(1): Don't overrun line buffer in some cases
Patches like file.txt attached to PR 190195 with a final line formed
like ">(EOL)" could cause a copy past the end of the current line buffer. In the
case of PR 191641, this caused a duplicate line to be copied into the resulting
file.
Instead of running past the end, treat it as if it were a blank line.
alc [Sun, 8 Oct 2017 22:17:39 +0000 (22:17 +0000)]
The blst_radix_init function has two purposes - to compute the number of
nodes to allocate for the blist, and to initialize them. The computation
can be done much more quickly by identifying the terminating node, if any,
at every level of the tree and then summing the number of nodes at each
level that precedes the topmost terminator. The initialization can also be
done quickly, since settings at the root mark the tree as all-allocated, and
only a few terminator nodes need to be marked in the rest of the tree.
Eliminate blst_radix_init, and perform its two functions more simply in
blist_create.
The allocation of the blist takes places in two pieces, but there's no good
reason to do so, when a single allocation is sufficient, and simpler.
Allocate the blist struct, and the array of nodes associated with it, with a
single allocation.
ian [Sun, 8 Oct 2017 18:38:22 +0000 (18:38 +0000)]
Fix imx6 hdmi init after r323553, which used a config_intrhook to defer the
attachment of i2c devices needed by hdmi.
The hdmi init also uses an intrhook callback to defer initialization, and if
the hdmi callback runs first, the i2c devices will not yet have registered
their device_t in association with their FDT phandle, which allows cross-
device references on FDT systems.
Now the hdmi deferred init checks for the i2c device registration, and if
it's not complete yet, it registers as an eventhandler watching for newbus
attach events. When the i2c device eventually attaches, the hdmi driver
unregisters from watching further events, and continues with the hdmi init.
Because the function signatures for an intrhook callback and an event
handler callback are the same, a single function is used for both callbacks.
ian [Sun, 8 Oct 2017 17:33:49 +0000 (17:33 +0000)]
Add eventhandler notifications for newbus device attach/detach.
The detach case is slightly complicated by the fact that some in-kernel
consumers may want to know before a device detaches (so they can release
related resources, stop using the device, etc), but the detach can fail. So
there are pre- and post-detach notifications for those consumers who need to
handle all cases.
A couple salient comments from the review, they amount to some helpful
documentation about these events, but there's currently no good place for
such documentation...
Note that in the current newbus locking model, DETACH_BEGIN and
DETACH_COMPLETE/FAILED sequence of event handler invocation might interweave
with other attach/detach events arbitrarily. The handlers should be prepared
for such situations.
Also should note that detach may be called after the parent bus knows the
hardware has left the building. In-kernel consumers have to be prepared to
cope with this race.
ian [Sun, 8 Oct 2017 17:21:16 +0000 (17:21 +0000)]
Restore the ability to deregister an eventhandler from within the callback.
When the EVENTHANDLER(9) subsystem was created, it was a documented feature
that an eventhandler callback function could safely deregister itself. In
r200652 that feature was inadvertantly broken by adding drain-wait logic to
eventhandler_deregister(), so that it would be safe to unload a module upon
return from deregistering its event handlers.
There are now 145 callers of EVENTHANDLER_DEREGISTER(), and it's likely many
of them are depending on the drain-wait logic that has been in place for 8
years. So instead of creating a separate eventhandler_drain() and adding it
to some or all of those 145 call sites, this creates a separate
eventhandler_drain_nowait() function for the specific purpose of
deregistering a callback from within the running callback.
alc [Sun, 8 Oct 2017 16:54:42 +0000 (16:54 +0000)]
Replace an unnecessary call to vm_page_activate() by an assertion that
the page is already wired or queued. Prior to the elimination of PG_CACHED
pages, vm_page_grab() might have returned a valid, previously PG_CACHED
page, in which case enqueueing the page was necessary. Now, that can't
happen. Moreover, activating the page is a dubious choice, since the page
is not being accessed.
sbruno [Sat, 7 Oct 2017 23:33:14 +0000 (23:33 +0000)]
Fix symlink if_igb.ko in -current such that its relative and doesn't
end up with non-standard DESTDIR information in its symlink. This
can happen very trivially if the release scripts are used.
sbruno [Sat, 7 Oct 2017 23:30:57 +0000 (23:30 +0000)]
Check so_error early in sendfile() call. Prior to this patch, if a
connection was reset by the remote end, sendfile() would just report
ENOTCONN instead of ECONNRESET.
cxgbe(4): Update T6, T5, and T4 firmwares to 1.16.63.0.
Changes since 1.16.26.0 for all three firmwares are listed below. This
list was obtained from the Release Notes of the Chelsio Unified Wire
v3.5.05 release for Linux.
T6 Firmware
++++++++++++
================================================================================
Version : 1.16.63.0
Date : 09/29/2017
================================================================================
Fixes
-----
BASE:
- Fixed a fw crash when configured traffic rate limit is less than 10kbps.
- Fixed traffic rate limiting for smaller traffic rate value.
ETH:
- Fixed 40G link failure when interface is toggled.
- Fixed adapter crash when interface is toggled during traffic.
- Fixed 25G link failure when PEER only supports consortium mode autoneg
for 25G.
- Fixed 100G optics link failure when cable is plugged in after bringing up
the interface.
- Enable RS FEC as default if speed is 100G.
- Fixed DCBX configuration refresh failure.
================================================================================
Version : 1.16.59.0
Date : 09/05/2017
================================================================================
FIXES
-----
BASE:
- Fixed fw crash caused by MC parity error in SO adapters.
- Generate Timer0Int interrupt if fw crashes due to unaligned access error. Host
driver must look into PCIE_FW register to see if any fw fatal error has
encountered. If PCIE_FW doesn't indicate any error then driver must ignore this
interrupt.
- Fixed receive buffer threshold settings which was resulting in error frames on
receive side.
ETH:
- Fixed an issue in connection traffic shaping when
FLOWC_WR->FW_FLOWC_MNEM_SCHEDCLASS is not received in first WR on the connection.
- Fixed link failure when speed is changed from 10G-1G-10G due to incorrect flag
check.
- Fixed improper LED behaviour for blink test and when traffic is running.
- Removed storage of previous fec settings from fw. Driver needs to pass the user
settings whenever a new module is plugged in as fw resets these when a module is
unplugged.
OFLD
- OVS offload: TP cache is flushed periodically to get the accuate filters stats
(hit count).
ENHANCEMENTS
------------
BASE:
- Ring backbone feature added. New FW_PARAMS_PARAM_DEV_RING_BACKBONE param type
added to query and enable ring backbone support.
- VNI support added for filtering. New entry_type FW_VI_MAC_TYPE_EXACTMAC_VNI
added to FW_VI_MAC_CMD.
- Added new API FW_PARAM_PARAM_DEV_MPSBGMAP to read the priority to buffer group
mapping for the ports.
- FW_PARAMS_PARAM_DEV_TPCHMAP API added to read the port to channel mapping.
- HMA (Host memory access) support added. New FW_HMA_CMD and
FW_PARAMS_PARAM_DEV_HMA_SIZE added to query and configure the HMA. It
enables the memfree support (256 connections) for iwarp.
- PTP support enabled.
ETH:
- Added consortium mode 50G support.
- Added the ability to allow only selected speeds to be advertised during auto
negotiation.
- Increased port capability from 16 to 32 bits to support more speeds.
FW_PARAMS_PARAM_PFVF_PORT_CAPS32 added to query whether fw supports 16 or 32
bit port capability.
OFLD:
- RDMA Write with immediate support added (iwarp 2.0 feature)
- FW_TLS_KEYCTX_TX_WR removed and security key management moved to driver.
- 256 offloaded connections support for iwarp on SO adapters.
iSCSI:
- New param FW_PARAMS_PARAM_DEV_PPOD_EDRAM added for iscsi ppod configuration
in EDRAM (performance improvement).
FOiSCSI:
- iSCSI Command offload target support added.
FOFCoE:
- FCoE support enabled.
================================================================================
Version : 1.16.43.0
Date : 05/05/2017
================================================================================
FIXES
-----
BASE:
- Fixed default DCB mode to AUTO.
- Fixed DCBX bugs when AUTO mode is configured in config file.
- Fixed an issue where even after removing PFC from switch, PFC wasn't getting
reset.
- Fixed DDR3/DDR4 ECC errors.
- Fixed an FLR issue where FLR completion was going to host before FLR
processing is finished in fw.
ETH:
- Fixed bug in writing multi-bytes using i2c interface.
- Fixed the link failure when optical cable is inserted into the QSA module
after loading the driver.
- Fixed false link up when peer interface was brought down.
- Enabling RS FEC by default for 100Gbase-SR4 according to 802.3BJ standard.
- Fixed bugs related to negotiated fec based local/peer fec ability and request.
- Fixed auto-neg failure with few switches.
- T6 Performance improvement fixes.
OFLD
- Fixed an extra credit issue if FW_RI_TYPE_FINI is delayed in fw due to
backpressure.
- Added a new queue type FW_IQ_TYPE_VF_CQ to handle the FW_PARAMS_PARAM_DMAQ*
commands. queue type will be part of the FW_PARAMS_PARAM_DMAQ_IQ_INTIDX
value. Used in guest RDMA (RDMA from VM/VF) usecase.
- T6 Crypto Coprocessor mode bug fixes.
- T6 Crypto TLS-inline mode bug fixes.
ENHANCEMENTS
------------
BASE:
- Added new API FW_PARAM_PARAM_DEV_MPSBGMAP to read the priority to buffer
group mapping for the ports.
ETH:
- Added broadcom consortium next page support for 25G CR.
This can be enabled using flags=an_brcm option in the t6-config.txt file.
- Added spider mode support.
- Added support for 10G-BaseT converter sfp+ module.
- Added support for additional 25G/100G cables.
- Added support to enable/disable auto-neg using ethtool.
================================================================================
Version : 1.16.33.0
Date : 02/24/2017
================================================================================
Fixes
-----
BASE:
- Fixed DDR4 uncorrectable errors.
ETH:
- Enabled link auto negotiation (AN) by default in config file.
- Added AN and FEC control api. Host driver and application can enable/disable
AN and FEC.
ETH:
- Added new workrequest ETH_TX_PKTS2_WR (see fw api document for more details).
================================================================================
Version : 1.16.29.0
Date : 01/27/2017
================================================================================
FIXES
-----
BASE:
- Set multiple fec values only if AN is enabled in config file and when module
is connected.
- Fixed intermittent DDR3/4 ECC errors.
- max number of ethctrl queue in VF set to 2 (reverted the last change
because it causes problem in VF drivers).
ETH:
- Made devlog more verbose by printing cable information in redable form.
- Updated AN settings to work with more 25G/100G switches.
- Added support for more SFP28/QSFP28 cables.
- Fixed an issue of link going down after few hours of idle time.
OFLD:
- Fixed an issue in TLS which was causing fw crash on running TLS traffic.
FOiSCSI:
- Fixed the failure of PXE boot OS install on an iscsi lun.
ENHANCEMENTS
------------
OFLD:
- Added filtering support for NAT. New WR FW_FILTER2_WR and
FW_PARAMS_PARAM_DEV_FILTER2_WR added for the same.
- Added RDMA guest mode (mode 3 or RDMA from VF) support.
T5 Firmware
++++++++++++
================================================================================
Version : 1.16.63.0
Date : 09/29/2017
================================================================================
Fixes
-----
BASE:
- Fixed offload memory overcommit in case of SO adapter.
ETH:
- Fixed DCBX configuration refresh failure.
OFLD
- Fixed 0B iWARP ingress read failure.
FOiSCSI:
- Fixed vlan interface ping failure.
================================================================================
Version : 1.16.59.0
Date : 09/05/2017
================================================================================
FIXES
-----
BASE:
- Fixed an FLR issue which was causing error when VF attached VM was powered on.
ETH:
- Fixed an issue in connection traffic shaping when
FLOWC_WR->FW_FLOWC_MNEM_SCHEDCLASS is not received in first WR on the connection.
- Fixed link failure when speed is changed from 10G-1G-10G due to incorrect flag
check.
- Fixed T580 link failure with few switches which take more time for
establishing link.
ENHANCEMENTS
------------
BASE:
- Ring backbone feature added. New FW_PARAMS_PARAM_DEV_RING_BACKBONE param type
added to query and enable ring backbone support.
- Added new API FW_PARAM_PARAM_DEV_MPSBGMAP to read the priority to buffer group
mapping for the ports.
- FW_PARAMS_PARAM_DEV_TPCHMAP API added to read the port to channel mapping.
FOiSCSI:
- iSCSI Command offload target support added.
================================================================================
Version : 1.16.43.0
Date : 05/05/2017
================================================================================
FIXES
-----
BASE:
- Fixed default DCB mode to AUTO.
- Fixed DCBX bugs when AUTO mode is configured in config file.
- Fixed an issue where even after removing PFC from switch, PFC wasn't getting
reset.
ETH:
- Fixed bug in writing multi-bytes using i2c interface.
- Fixed the link failure when optical cable is inserted into the QSA module
after loading the driver.
OFLD
- Fixed an extra credit issue if FW_RI_TYPE_FINI is delayed in fw due to
backpressure.
- Added a new queue type FW_IQ_TYPE_VF_CQ to handle the FW_PARAMS_PARAM_DMAQ*
commands. queue type will be part of the FW_PARAMS_PARAM_DMAQ_IQ_INTIDX
value. Used in guest RDMA (RDMA from VM/VF) usecase.
ENHANCEMENTS
------------
BASE:
- Added new API FW_PARAM_PARAM_DEV_MPSBGMAP to read the priority to buffer
group mapping for the ports.
================================================================================
Version : 1.16.33.0
Date : 02/24/2017
================================================================================
ENHANCEMENTS
------------
ETH:
- Added new workrequest ETH_TX_PKTS2_WR (see fw api document for more details).
================================================================================
Version : 1.16.29.0
Date : 01/27/2017
================================================================================
FIXES
-----
BASE:
- max number of ethctrl queue in VF set to 2 (reverted the last change
because it causes problem in VF drivers).
FOiSCSI:
- Fixed the failure of PXE boot OS install on an iscsi lun.
ENHANCEMENTS
------------
OFLD:
- Added filtering support for NAT. New WR FW_FILTER2_WR and
FW_PARAMS_PARAM_DEV_FILTER2_WR added for the same.
- Added RDMA guest mode (mode 3 or RDMA from VF) support.
T4 Firmware
+++++++++++
================================================================================
Version : 1.16.63.0
Date : 09/29/2017
================================================================================
Fixes
-----
ETH:
- Fixed DCBX configuration refresh failure.
FOiSCSI:
- Fixed vlan interface ping failure.
================================================================================
Version : 1.16.59.0
Date : 09/05/2017
================================================================================
FIXES
-----
ETH:
- Fixed an issue in connection traffic shaping when
FLOWC_WR->FW_FLOWC_MNEM_SCHEDCLASS is not received in first WR on the connection.
ENHANCEMENTS
------------
BASE:
- FW_PARAMS_PARAM_DEV_TPCHMAP API added to read the port to channel mapping.
================================================================================
Version : 1.16.43.0
Date : 05/05/2017
================================================================================
FIXES
-----
BASE:
- Fixed default DCB mode to AUTO.
- Fixed DCBX bugs when AUTO mode is configured in config file.
- Fixed an issue where even after removing PFC from switch, PFC wasn't getting
reset.
ETH:
- Fixed bug in writing multi-bytes using i2c interface.
OFLD
- Fixed an extra credit issue if FW_RI_TYPE_FINI is delayed in fw due to
backpressure.
- Added a new queue type FW_IQ_TYPE_VF_CQ to handle the FW_PARAMS_PARAM_DMAQ*
commands. queue type will be part of the FW_PARAMS_PARAM_DMAQ_IQ_INTIDX
value. Used in guest RDMA (RDMA from VM/VF) usecase.
ENHANCEMENTS
------------
BASE:
- Added new API FW_PARAM_PARAM_DEV_MPSBGMAP to read the priority to buffer
group mapping for the ports.
manu [Sat, 7 Oct 2017 16:48:42 +0000 (16:48 +0000)]
Allwinner: Add clock driver for ccu_sun8i_r
SUN8I and SUN50I (H3, H5, A83T and A64) have a second clock controller
unit. It controls the clocks for the second gpio controller, the IR
controller etc ...
Support for A83T is not supported.
mjg [Fri, 6 Oct 2017 23:05:55 +0000 (23:05 +0000)]
namecache: factor out ~MAKEENTRY lookups from the common path
Lookups of the sort are rare compared to regular ones and succesfull ones
result in removing entries from the cache.
In the current code buckets are rlocked and a trylock dance is performed,
which can fail and cause a restart. Fixing it will require a little bit
of surgery and in order to keep the code maintaineable the 2 cases have
to split.
If VIMAGE is present we can start jails with their own pf instance. This
makes it fairly easy to run tests.
For example, this basic test verifies that drop/pass and icmp
classification works. It's a basic sanity test for pf, and hopefully an
example on how to write more pf tests.
The tests are skipped if VIMAGE is not enabled.
This work is inspired by the GSoC work of Panagiotes Mousikides.
cem [Fri, 6 Oct 2017 18:27:55 +0000 (18:27 +0000)]
random(4): Discard low entropy inputs
The later fields of the harvest_event structure are predictable and provide
little value to the entropy pool. Only feed in the relatively high entropy
counter and explicit entropy buffer to increase measured input entropy.
See also:
https://people.freebsd.org/~jmg/vbsdcon_2017_ddfreebsdrng_slides.pdf
PR: 222807
Submitted by: W. Dean Freeman <badfilemagic AT gmail.com>
Reviewed by: jmg (earlier version), delphij
Approved by: secteam (delphij)
Obtained from: HBSD 8d809124d563937edd84c9c9d5494406e359c55c
Security: no -- low entropy marginal input has no known negative affect on pool quality
Differential Revision: https://reviews.freebsd.org/D12610
marcel [Fri, 6 Oct 2017 16:38:00 +0000 (16:38 +0000)]
Fix alignment of 'last' in autofill.
'last' is the sector number of the last usable sector. Sector
numbers start with 0. As such, 'last' is always 1 less than
the count of sectors and aligning 'last' down as-is means that
the number of free sectors is pessimized by 'alignment - 1' if
the number of usable sectors was already a multiple of the
alignment. Consequently, gpart(8) failed to create a partition
when the alignment and size were such that it would extend to
the end of the disk.
andrew [Fri, 6 Oct 2017 15:46:11 +0000 (15:46 +0000)]
Also handle psci 1.0. This can be seen as a bug fix update for the 0.2
specification we already support, with the only changes in functions we
don't currently use.
trasz [Fri, 6 Oct 2017 15:09:28 +0000 (15:09 +0000)]
Fix kvm_getprocs(3) error reporting in ps(1).
Previously it just didn't work at all - kvm_getprocs(3) doesn't update
the &nentries when it returns NULL. The end result was that ps(1) showed
garbage data instead of reporting kinfo_proc size mismatch.
kib [Fri, 6 Oct 2017 14:29:53 +0000 (14:29 +0000)]
Improve i386_get_ldt().
Provide consistent snapshot of the requested descriptors by preventing
other threads from modifying LDT while we fetch the data, lock dt_lock
around the read. Copy the data into intermediate buffer, which is
copied out after the lock is dropped.
Comparing with the amd64 version, the read is done byte by byte, since
there is no atomic 64bit read (cmpxchg8b method is too heavy comparing
with the avoided issues).
Improve overflow checking for the descriptors range calculations and
remove unneeded casts. Use unsigned types for sizes.
Allow zero num argument to i386_get_ldt() and i386_set_ldt(). This
case is handled naturally by the code flow.
Reviewed by: bde
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
eugen [Fri, 6 Oct 2017 13:46:05 +0000 (13:46 +0000)]
ftpd(8): fix user context handling.
Apply authenticated user context after update of wtmp(5) at start of session,
so that ftpd process is not killed by kernel with SIGXFSZ when user has
"filesize" limit lower than size of system wtmp file. Same applies
to session finalization: revert to super-user context before update of wtmp.
If ftpd hits limit while writing a file at user request,
do not get killed with SIGXFSZ instantly but apparently ignore the signal,
process error and report it to the user, and continue with the session.
imp [Fri, 6 Oct 2017 13:43:16 +0000 (13:43 +0000)]
Oversight on armv7 bulk commit: Make MACHINE_ARCH be armv7 for new
enough processors. This isn't ideal, because one could still compile
MACHINE_ARCH armv6, but with armv7 options enabled. We don't normally
do that, and it's a bit of an edge case so accept the less than ideal
solution here in the absence of something better.
jlh [Fri, 6 Oct 2017 08:49:15 +0000 (08:49 +0000)]
Bump src.conf.5's Dd.
This file shouldn't be modified manually but well, I did it in my previous
commit. So go down further the rabbit hole so as to at least keep some
consistency.
jlh [Fri, 6 Oct 2017 08:43:14 +0000 (08:43 +0000)]
Remove rcmds.
If they are still needed, you can find them in the net/bsdrcmds port.
This was proposed June, 20th and approved by various committers [1].
They have been marked as deprecated on CURRENT in r320644 [2] on July, 4th.
Both stable/11 and release/11.1 contain the deprecation notice (thanks to
allanjude@).
Note that ruptime(1)/rwho(1)/rwhod(8) were initially thought to be part of
rcmds but this was a mistake and those are therefore NOT removed.
FreeBSD note: the manual page is to be updated separately.
https://www.illumos.org/issues/8067
Add an option to zdb to print a literal embedded block pointer supplied on the
command line:
zdb -E [-A] word0:word1:...:word15
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Alex Reece <alex@delphix.com>
Reviewed by: Yuri Pankov <yuri.pankov@gmail.com>
Approved by: Robert Mustacchi <rm@joyent.com>
Author: Matthew Ahrens <mahrens@delphix.com>
https://www.illumos.org/issues/7340
When -o origin=<snapshot> is specified as part of a ZFS receive, that origin
should override the automatic detection in libzfs.
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Approved by: Robert Mustacchi <rm@joyent.com>
Author: Paul Dagnelie <pcd@delphix.com>
https://www.illumos.org/issues/5142
the current libzfs only allows simple disk and mirror setup for boot pool, as
loader does support booting from raidz, this feature will remove raidz
restriction from boot pool setup.
FreeBSD note: we have long supported this feature, this commit only
removes a small difference in libzfs.
Reviewed by: George Wilson <george.wilson@delphix.com>
Reviewed by: Yuri Pankov <yuri.pankov@nexenta.com>
Reviewed by: Andrew Stormont <andyjstormont@gmail.com>
Reviewed by: Albert Lee <trisk@omniti.com>
Approved by: Robert Mustacchi <rm@joyent.com>
Author: Toomas Soome <tsoome@me.com>
https://www.illumos.org/issues/6268
The zfs diff command presents a description of the changes that have occurred
to files within a filesystem between two snapshots. If a file is renamed, the
tool is capable of reporting this, e.g.:
cd /some/zfs/dataset/subdir
mv file0 file1
Will result in a diff record like:
R /some/zfs/dataset/subdir/file0 -> /some/zfs/dataset/subdir/file1
Unfortunately, it seems that rename detection only uses the base filename to
determine if a file has been renamed or simply modified. This leads to
misreporting only the original filename, omitting the more relevant destination
filename entirely. For example:
cd /some/zfs/dataset/subdir
mv file0 ../otherdir/file0
Will result in a diff entry:
M /some/zfs/dataset/subdir/file0
But it should really emit:
R /some/zfs/dataset/subdir/file0 -> /some/zfs/dataset/otherdir/file0
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Justin Gibbs <gibbs@scsiguy.com>
Approved by: Dan McDonald <danmcd@omniti.com>
Author: Joshua M. Clulow <josh@sysmgr.org>
https://www.illumos.org/issues/7571
ZFS displays the default value for non-present readonly numeric (and index)
properties. However, these properties default values are not meaningful.
Instead, we should display a "-", indicating that they are not present. For
example, on a version-12 pool, the usedby* properties are not available, but
they show up as the incorrect value "0":
1. zfs get all test12
...
test12 usedbysnapshots 0 -
test12 usedbydataset 0 -
test12 usedbychildren 0 -
test12 usedbyrefreservation 0 -
We will be introducing more sometimes-present numeric readonly properties, so
it would be nice to fix this.
Reviewed by: Dan Kimmel <dan.kimmel@delphix.com>
Reviewed by: George Wilson <george.wilson@delphix.com>
Approved by: Robert Mustacchi <rm@joyent.com>
Author: Matthew Ahrens <mahrens@delphix.com>
FreeBSD note: the manual page is to be updated separately.
https://www.illumos.org/issues/6392
When given a pool name via -e, zdb would attempt an import. If it
failed, then it would attempt a verbatim import. This behavior is
not always desirable so a -V switch is added to zdb to control the
behavior. When specified, a verbatim import is done. Otherwise,
the behavior is as it was previously, except no verbatim import
is done on failure.
https://github.com/zfsonlinux/zfs/commit/a5778ea2427bd340e3b4f697d9b6e1452bd71909
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Yuri Pankov <yuri.pankov@gmail.com>
Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed by: Matt Ahrens <mahrens@delphix.com>
Reviewed by: Pavel Zakharov <pavel.zakharov@delphix.com>
Approved by: Dan McDonald <danmcd@omniti.com>
Author: Richard Yao <ryao@gentoo.org>
FreeBSD note: this commit does not update the manual page.
The original change includes conversion of the manual page from *roff format
to mandoc format. So, it is hard to extract the content change from
that. I am going to replace our zdb manual page, which is an earlier
independent conversion, with a slighly modified version of the upstream page.
https://www.illumos.org/issues/6410
This is primarily intended to ease debugging & testing ZFS when one is only
interested in things like the on-disk location of a specific object's blocks,
but doesn't know their object id. This allows doing things like the following
(FreeBSD-based example):
# zpool create -f foo da0
# dd if=/dev/zero of=/foo/1 bs=1M count=4 >/dev/null 2>&1
# zpool export foo
# zdb -vvvvv -o "ZFS plain file" foo /1
Object lvl iblk dblk dsize lsize %full type
8 2 16K 128K 3.99M 4M 100.00 ZFS plain file (K=inherit) (Z=inherit)
168 bonus System attributes
dnode flags: USED_BYTES USERUSED_ACCOUNTED
dnode maxblkid: 31
path /1
uid 0
gid 0
atime Thu Apr 23 22:45:32 2015
mtime Thu Apr 23 22:45:32 2015
ctime Thu Apr 23 22:45:32 2015
crtime Thu Apr 23 22:45:32 2015
gen 7
mode 100644
size 4194304
parent 4
links 1
pflags 40800000004
Indirect blocks:
0 L1 DVA[0]=<0:c19200:600> DVA[1]=<0:10800019200:600> [L1 ZFS
plain file] fletcher4 lz4 LE contiguous unique double size=4000L/200P birth=7L/
Reviewed by: Pavel Zakharov <pavel.zakharov@delphix.com>
Reviewed by: Matthew Ahrens <mahrens@delphix.com>
Reviewed by: Will Andrews <will@freebsd.org>
Approved by: Dan McDonald <danmcd@omniti.com>
Author: Yuri Pankov <yuri.pankov@nexenta.com>
imp [Thu, 5 Oct 2017 23:01:50 +0000 (23:01 +0000)]
Tag all armv7 kernels as such in their machine config line.
Transition all boards that support arm cortex CPUs to armv7. This
leaves two armv6 kernels in the tree. RPI-B, which uses the BCM2835
which has a ARM1176 core, and VERSATILEPB, which is a qemu board setup
around the time RPI-B went in. Copy std.armv6 to std.armv7, even
though that duplicates a lot of stuff. More work needs to be done to
sort out the duplication.
imp [Thu, 5 Oct 2017 23:01:33 +0000 (23:01 +0000)]
Support armv7 builds for userland
Make armv7 as a new MACHINE_ARCH.
Copy all the places we do armv6 and add armv7 as basically an
alias. clang appears to generate code for armv7 by default. armv7 hard
float isn't supported by the the in-tree gcc, so it hasn't been
updated to have a new default.
Support armv7 as a new valid MACHINE_ARCH (and by extension
TARGET_ARCH).
rmacklem [Thu, 5 Oct 2017 20:10:40 +0000 (20:10 +0000)]
Add Flex File Layout support to the NFSv4.1 pNFS client.
This patch adds support for the Flexible File Layout to the pNFS client.
Although the patch is rather large, it should only affect NFS mounts
using the "pnfs" option against pNFS servers that do not support File
Layout.
There are still a couple of things missing from the Flexible File Layout
client implementation:
- The code does not yet do a LayoutReturn with I/O error stats when
I/O error(s) occur when attempting to do I/O on a DS.
This will be fixed in a future commit, since it is important for the
MDS to know that I/O on a DS is failing.
- The current code does writes and commits to mirror DSs serially.
Making them happen concurrently will be done in a future commit,
after discussion on freebsd-current@ on the best way to do this.
- The code does not handle NFSv4.0 DSs. Since there is no extant pNFS
server that implements NFSv4.0 DSs and NFSv4.1 DSs makes more sense
now, I don't intend to implement this until there is a need for it.
There is support for NFSv4.1 and NFSv3 DSs.
mjg [Thu, 5 Oct 2017 19:18:02 +0000 (19:18 +0000)]
locks: take the number of readers into account when waiting
Previous code would always spin once before checking the lock. But a lock
with e.g. 6 readers is not going to become free in the duration of once spin
even if they start draining immediately.
Conservatively perform one for each reader.
Note that the total number of allowed spins is still extremely small and is
subject to change later.
shurd [Thu, 5 Oct 2017 17:52:38 +0000 (17:52 +0000)]
bnxt: fix intermittent VLAN issues
bnxt_init() is invokes Function Reset (bnxt_hwrm_func_reset) and thus FW
configuration for VLANs get erased. To fix this, in bnxt_init(), Invoking
HWRM to configure VLANs (bnxt_hwrm_cfa_l2_set_rx_mask).
imp [Thu, 5 Oct 2017 16:48:05 +0000 (16:48 +0000)]
This README file was quite relevant for FreeBSD 3 or 4. However, the
information in this file is now somewhat dated, or is present mostly
correct in the man pages. Retire this file rather than fix it.
shurd [Thu, 5 Oct 2017 14:43:30 +0000 (14:43 +0000)]
Fix "taskqgroup_attach: setaffinity failed: 3" with iflib drivers
Improved logging added in r323879 exposed an error during
attach. We need the irq, not the rid to work correctly. em uses
shared irqs, so it will use the same irq for TX as RX. bnxt does
not use shared irqs, or TX irqs at all, so there's no need to set
the TX irq affinity.
kib [Thu, 5 Oct 2017 13:12:59 +0000 (13:12 +0000)]
Ensure that after sucessfull i386_set_ldt() call, other threads can
use LDT segments immediately.
If the i386_set_ldt() call created a first LDT descriptor (and
consequently created the LDT) for our address space, LDTR is currently
loaded only on the CPU executing the syscall. Other CPUs executing
threads sharing the address space, would only load LDTR after context
switch.
Uncomment set_user_ldt_rv() and call it on all CPUs. Remove critical
section inside set_user_ldt(), it is not needed in the context of call
from smp_rendezvous().
Set md_ldt after md_ldt_sd is initialized using the same code sequence
as in user_ldt_free(). Do the whole initialization in a critical
section, to not race with the context switching while we set LDT.
Discussed with: bde
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
kib [Thu, 5 Oct 2017 12:50:03 +0000 (12:50 +0000)]
Avoid a race betweem freeing LDT and context switches.
cpu_switch.S uses curproc->p_md.md_ldt value as the flag indicating
presence of the process LDT. The flag is checked and then ldt segment
descriptor is copied into the CPU' GDT slot.
Disallow context switches around clearing of the curproc LDT state by
performing the cleanup in critical section. Ensure that the md_ldt
flag is cleared before md_ldt_sd descriptor content is destroyed by
inserting fence between the operations.
We depend on the x86 memory model strong ordering guarantees, in
particular, that cpu_switch.S observes the writes to md_ldt and
md_ldt_sd in the expected order.
Discussed with: bde
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
avg [Thu, 5 Oct 2017 12:38:26 +0000 (12:38 +0000)]
fix the misleading log facility used in devd/zfs.conf
In general, the "kern" facility is reserved for the kernel use only.
If a program specifies that facility, then it is silently converted
to "user" facility.
So, using logger -p kern.xxx was both misleading and non-specific.
Thus, change the facility to local7, so that users can create
more adequate syslogd configurations.
While local0..local7 are documented as being for local use we already
have several examples in the tree where they are used because none of
the named facilities really fits.
avg [Thu, 5 Oct 2017 12:32:14 +0000 (12:32 +0000)]
sysctl-s in a module should be accessible only when the module is initialized
A sysctl can have a custom handler that may access data that is initialized
via SYSINIT(9) or via a module event handler (also invoked via SYSINIT).
Thus, it is not safe to allow access to the module's sysctl-s until
the initialization is performed. Likewise, we should not allow access
to teh sysctl-s after the module is uninitialized.
The latter is easy to achieve by properly ordering linker_file_unregister_sysctls
and linker_file_sysuninit.
The former is not as easy for two reasons:
- the initialization may depend on tunables which get set when sysctl-s are
registered, so we need to set the tunables before running sysinit-s
- the initialization may try to dynamically add more sysctl-s under statically
defined sysctl nodes
So, this change splits the sysctl setup into two phases. In the first phase
the sysctl-s are registered as before but they are disabled and hidden from
consumers. In the second phase, done after sysinit-s, normal access to the
sysctl-s is enabled.
The change should affect only dynamic module loading and unloading after
the system boot-up. Nothing changes for sysctl-s compiled into the kernel
and sysctl-s in preloaded modules.
kib [Thu, 5 Oct 2017 12:29:34 +0000 (12:29 +0000)]
Improve amd64_get_ldt().
Provide consistent snapshot of the requested descriptors by preventing
other threads from modifying LDT while we fetch the data, lock dt_lock
around the read. Copy the data into intermediate buffer, which is
copied out after the lock is dropped.
Use guaranteed atomic (aligned volatile) reads of the descriptors to
use same-size atomic as CPU update to set A bit in the descriptor type
field.
Improve overflow checking for the descriptors range calculations and
remove unneeded casts.
Reviewed by: bde
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
avg [Thu, 5 Oct 2017 12:25:18 +0000 (12:25 +0000)]
remove heuristic error detection from ddi_strto*()
Zero, <TYPE>_MIN and <TYPE>_MAX values can result from valid conversions.
They don't necessarily imply any error.
Since we do not have any reliable error signaling from libkern's strto*(),
it's better to always assume success rather than to report an error when
there is none.