- Fix compatibility strings
- Properly decode upstream's two-cell interrupt specs. Our home-made dts
does not have two-cell interrupts so no need to preserve backward
compatibility
waiters. Otherwise, owners of the shared-busy state are left blocked
and might get into a deadlock.
Note that the vm_page_busy_downgrade() function is not used in the
tree right now.
Reported and tested by: pho (previous version)
Reviewed by: alc, markj
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D8195
Use isrc in attached MSI data structure instead of using map's
isrc directly. map's isrc is set to NULL on IRQ deactivation
which happens prior to pci_release_msi so MSI_RELEASE_MSI
receives array of NULLs
Reviewed by: mmel
Differential Revision: https://reviews.freebsd.org/D8206
Extend the mapping table for external port numbering to support port modes
which output to the second external port only. Where supported, map from
the current port mode rather than inferring from all the available modes.
Updated comments for clarity.
Submitted by: Richard Houldsworth <rhouldsworth at solarflare.com>
Sponsored by: Solarflare Communications, Inc.
MFC after: 2 days
Differential Revision: https://reviews.freebsd.org/D8210
This is the preamble for network device SR-IOV and
NDIS_STATUS_NETWORK_CHANGE handling.
MFC after: 1 week
Sponsored by: Microsoft
Differential Revision: https://reviews.freebsd.org/D8209
It fixes driver attach issue to a new firmware which reports a new
port-modes.
Reviewed by: gnn
Submitted by: Tom Millington <tmillington at solarflare.com>
Sponsored by: Solarflare Communications, Inc.
MFC after: 2 days
Differential Revision: https://reviews.freebsd.org/D8203
The _correct_ way to identify the supported checksum offloading and
TSO parameters is to query OID_TCP_OFFLOAD_HARDWARE_CAPABILITIES.
MFC after: 1 week
Sponsored by: Microsoft
Differential Revision: https://reviews.freebsd.org/D8088
- Fix range check
- Due to checking found value in for(;;) condition irq after loop was
always + 1 from actually found slot and wrong entry was marked as
used which lead to returning slot 0 for all requests.
ieee80211_tx_complete()
This change allows to pass packet length to rate control modules and
fixes IFCOUNTER_OBYTES calculation.
Tested with Intel 3945BG, STA mode.
- Add multitouch support (protocol B)
- Report physical size of the screen
- Switch from using busy loop to callbacks
- Enable callbacks only when there is active listener on /dev/input/eventX
Submitted by: Vladimir Kondratiev <wulf@cicgroup.ru>
Automaticaly release (send ABS_MT_TRACKING_ID = -1) MT-slots
that has not been listed in current MT protocol type B report.
Slot is counted as listed if corresponding ABS_MT_SLOT event
has been sent regardless of other MT events.
Events are sent on SYN_REPORT event.
Submitted by: Vladimir Kondratiev <wulf@cicgroup.ru>
Add new API call: evdev_register_mtx which takes lock argument that
should be used instead of internal one for evdev locking. Useful for
cases if evdev_push_event() is always called with driver's lock taken
and reduces amount of lock aquisitions. This allows to avoid LOR
between ev_open/ev_close invocations and evdev_push_event() Such LOR
can happen when ev_open/ev_close methods acquire driver lock and
evdev_push_event() is called with this lock taken.
Submitted by: Vladimir Kondratiev <wulf@cicgroup.ru>
Don't call pmap_kextract() multiple times, it wastes CPU cycles, which in a high
performance environment can be very expensive.
Inline XX_FindTracker() to allow more optimizations as well.
There are a variety of more interesting RX statistics that we should
keep track of but we don't. This is a starting point for adding more
information.
Specifically:
* now the RX rate information and some of the packet status is
passed up;
* The 32 bit or 64 bit TSF is passed up;
* the PHY mode is passed up;
* the "I'm decap'ed AMSDU!" state is passed up;
* number of RX chains is bumped to 4.
This is all mostly a placeholder for getting the data into the RX status
before we pass it up to net80211 - unfortunately we don't yet enforce
that drivers provide it, nor do we pass the provided info back up the
stack so anyone can use the data.
We're going to need to use some of this data moving forward.
Notably, now that some hardware can do AMSDU decap for us (the intel iwm
driver can do it when we flip it on; the ath10k port I'm doing does
it for us) then we need to pass it up through the stack so the duplicate
RX sequence numbers and crypto/IV details don't cause the packet to
be dropped and/or counted against a replay counter.
It's also the beginning of being able to do more interesting node
accounting in net80211. Specifically, once drivers start populating
per-packet rate information, AMPDU information, timestamps, etc,
we can start providing histograms of rate-versus-RSSI, account
for receive time spent per node and other such interesting things.
(Note: I'm also hoping to include ranging and RTT information for
future chipset support; and it's likely going to include it in
this kind of fashion.)
This change extends the nd6 lock to protect the ND prefix list as well
as the list of advertising routers associated with each prefix. To handle
cases where the nd6 lock must be dropped while iterating over either the
prefix or default router lists, a generation counter is used to track
modifications to the lists. Additionally, a new mutex is used to serialize
prefix on-link/off-link transitions. This mutex must be acquired before
the nd6 lock and is held while updating the routing table in
nd6_prefix_onlink() and nd6_prefix_offlink().
Reviewed by: ae, tuexen (SCTP bits)
Tested by: Jason Wolfe <jason@llnw.com>,
Larry Rosenman <ler@lerctr.org>
MFC after: 2 months
Differential Revision: https://reviews.freebsd.org/D8125
R_AARCH64_RELATIVE relocation found on arm64. It would try to add the
contents of the memory location being relocated to the base address and
the relocation addend. This worked when the contents was zero, however
this now seems to be set to the value of the addend so we add this twice.
Fix this by just setting the memory to the computed value.
Reviewed by: emaste
MFC after: 1 week
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D8177
that can be compiled on various OSes (including on older versions
of FreeBSD), make it possible to have it include the partitioning
scheme definitions without pulling in FreeBSD specifics.
In particular this means:
o move the scheme definitions iand related defines to header files
under sys/disk,
o make them (more) portable by using uint#_t (where applicable)
and renaming defines so that they at least have a good prefix,
o make the new headers stand-alone so that they don't need FreeBSD
definitions, like struct uuid(*)
o keep the original headers for compatibility, but rewrite them to
get the scheme definitions from <sys/disk/$scheme.h>.
(*) since UUID/GUID type definitions are non-portable and the GPT
scheme uses them, make it possible to have the scheme definitions
use an external type by allowing consumers of the header to set
GPT_UUID_TYPE. When GPT_UUID_TYPE has not been defined, the header
will use it's own type definition, which is the same as struct uuid.
The gpt_uuid_t typedef is created to abstract the details and allows
consumers to refer to a single type.
There is not conflict between the partitioning scheme headers and
what is defined in them. All headers can be included in the same
source files.
Note: consumers of the old headers have not been changed yet. Such
will be done if and when needed/beneficial.
Reviewed by: imp, jhb
MFC after: 1 month
Sponsored by: Bracket Computing
Other uses of cache_purgevfs() do rely on the cache purge for correct
operations, when paths are invalidated without unmount.
Reported and tested by: jkim
Discussed with: mjg
Sponsored by: The FreeBSD Foundation
This should allow vn_fullpath() to work even when vfs name cache is
disabled for zfs, which is the case when zfs properties like
casesensitivity and normalization are set non-default values.
The new code should be 100% reliable for directories and "mostly"
reliable for files, that is, when hardlinks across directories are
not used.
Reported by: Frederic Chardon <chardon.frederic@gmail.com>
Reviewed by: kib (vfs contract)
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D8146
netmap_kern.h currently requires all drivers including it to include
selinfo.h.
Submitted by: mmacy@nextbsd.org
Reviewed by: gnn
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D5334
In the persist case, take the SYN and FIN flags into account when updating
the sequence space sent.
Reviewed by: gnn
MFC after: 2 weeks
Sponsored by: Juniper Networks, Netflix
Differential Revision: https://reviews.freebsd.org/D7075
Tested by: Limelight, Netflix
Introduce internal counter to track opens. Using provider's counters is
not very successfull after calling g_wither_provider().
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
When tsoome@ added skein support to the ZFS boot code and zfsloader, it
resulted in an explosion in code size, running close to a number of
limits.
The default for the C version of skein is to unroll all loops for
skein-256 and 512
Disabling the loop unrolling saves 20-28kb from each binary
boot1.efi
gptzfsboot
loader.efi
userboot.so
zfsloader
Reviewed by: emaste, tsoome
Sponsored by: ScaleEngine Inc.
Differential Revision: https://reviews.freebsd.org/D7826
When a syncid bump is pending, any write to the mirror results in the
updated syncid being written to each component's metadata block. However,
the update was only being performed after the writes to the mirror
componenents were queued. Instead, synchronously update the metadata block
first.
MFC after: 3 weeks
Sponsored by: Dell EMC Isilon
Consider a mirror with two components, m1 and m2. Suppose a hardware error
results in the removal of m2, with m1's genid bumped. Suppose further that
a replacement mirror component m3 is created and synchronized, after which
the system is shut down uncleanly. During a subsequent bootup, if gmirror
tastes m1 and m2 first, m2 will be removed from the mirror because it is
broken, but the mirror will be started without bumping the syncid on m1
because all elements of the mirror are accounted for. Then m3 will be
added to the already-running mirror with the same syncid as m1, so the
components will not be synchronized despite the unclean shutdown.
Handle this scenario by bumping the syncid of healthy components if any
broken mirrors are discovered during mirror startup.
MFC after: 3 weeks
Sponsored by: Dell EMC Isilon
- Fix RX and TX teardown:
. TX teardown would not reclaim the abandoned descriptors;
. Interrupt storms in RX teardown;
. Fixed the acknowledge of the teardown completion interrupt.
- Remove temporary lists for the descriptors;
- Simplified the descriptor handling (less writes and reads from
descriptors where possible);
- Better debug;
- Add support for the RX threshold interrupts:
With interrupt moderation only, an RX overrun is likely to happen. The
RX threshold is set to trigger a non paced interrupt everytime your RX
free buffers are under the minimum threshold, helping to prevent the rx
overrun.
The NIC now survive when pushed over its limits (where previously it would
lock up in a few seconds).
uFW (600MHz SoC) can now forward up to 560Mb/s of UDP traffic (netmap
pkt-gen as source and sink). TCP forwarding rate is over 350Mb/s.
No difference (other than CPU use) was seen on Beaglebone black (1GHz SoC)
for his fast ethernet.
Tested on: uFW, BBB
Sponsored by: Rubicon Communications, LLC (Netgate)
by vm_pageout_scan() local to vm_pageout_worker(). There is no reason
to store the pass in the NUMA domain structure.
Reviewed by: kib
MFC after: 3 weeks
KASSERT in cam_ccbq_insert_ccb that only XPT_FC_QUEUED ops are queued,
and XPT_FC_USER_CCB ops are not. Otherwise cam_ccbq_ccb_done may be
skipped.
Bounds check the index used for camq_remove in order to panic instead
of scribble on removal of an out-of-bounds index (e.g. consider the
effect of camq_remove of CAM_UNQUEUED_INDEX).
KASSERT in cam_ccbq_remove_ccb that the ccb removed by index was the
one sought.
Submitted by: Ryan Libby <rlibby@gmail.com>
Reviewed by: imp, mav
MFC after: 2 weeks
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D8151
target was met.
Previously, vm_pageout_worker() itself checked the length of the free page
queues to determine whether vm_pageout_scan(pass >= 1)'s inactive queue scan
freed enough pages to meet the free page target. Specifically,
vm_pageout_worker() used vm_paging_needed(). The trouble with
vm_paging_needed() is that it compares the length of the free page queues to
the wakeup threshold for the page daemon, which is much lower than the free
page target. Consequently, vm_pageout_worker() could conclude that the
inactive queue scan succeeded in meeting its free page target when in fact
it did not; and rather than immediately triggering an all-out laundering
pass over the inactive queue, vm_pageout_worker() would go back to sleep
waiting for the free page count to fall below the page daemon wakeup
threshold again, at which point it will perform another limited (pass == 1)
scan over the inactive queue.
Changing vm_pageout_worker() to use vm_page_count_target() instead of
vm_paging_needed() won't work because any page allocations that happen
concurrently with the inactive queue scan will result in the free page count
being below the target at the end of a successful scan. Instead, having
vm_pageout_scan() return a value indicating success or failure is the most
straightforward fix.
Reviewed by: kib, markj
MFC after: 3 weeks
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D8111
Setting the doze registers wasn't actually working, and was being masked by a
bad #ifdef. Since the #ifdef was fixed, now e500mc-based SoCs hang at idle.
Fix this by using the intended wait.
MFC after: 1 week
The tag fastroute came from ipf and was removed in OpenBSD in 2011. The code
allows to skip the in pfil hooks and completely removes the out pfil invoke,
albeit looking up a route that the IP stack will likely find on its own.
The code between IPv4 and IPv6 is also inconsistent and marked as "XXX"
for years.
Submitted by: Franco Fichtner <franco@opnsense.org>
Differential Revision: https://reviews.freebsd.org/D8058
Reduce the cost of TLB invalidation on x86 by using per-CPU completion flags
Reduce contention during TLB invalidation operations by using a per-CPU
completion flag, rather than a single atomically-updated variable.
On a Westmere system (2 sockets x 4 cores x 1 threads), dtrace measurements
show that smp_tlb_shootdown is about 50% faster with this patch; observations
with VTune show that the percentage of time spent in invlrng_single_page on an
interrupt (actually doing invalidation, rather than synchronization) increases
from 31% with the old mechanism to 71% with the new one. (Running a basic file
server workload.)
Submitted by: Anton Rang <rang at acm.org>
Reviewed by: cem (earlier version)
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D8041
Usually there is some slack after the last partition due to 4k alignment
In the 10.3 EC2 images, there was not. EC2 seems to hang if you try to
read past the end of the disk in the loader, resulting in an unbootable
instance after upgrading to 11.0
PR: 213196
Reported by: Peter Ankerstal <peter@pean.org>
Tested by: cperciva
Reviewed by: tsoome
MFC after: 3 days
Sponsored by: ScaleEngine Inc.
Differential Revision: https://reviews.freebsd.org/D8144
These are older MIPS4kc parts from Atheros. They typically ran at
sub-200MHz and have 11bg, 11a, or 11abg wifi MAC/PHYs integrated.
This port is the initial non-wifi pieces required to bring up the
chip. I'll commit the redboot and other pieces later, and then
hopefully(!) wifi support will follow.
Submitted by: Mori Hiroki <yamori813@yahoo.co.jp>
Differential Revision: https://reviews.freebsd.org/D7237
for later Cortex-A CPUs that support the Multiprocessor Extensions. This
will be needed to support both in a single GENERIC kernel while still
being able to only build for a single SoC.
Reviewed by: mmel
Relnotes: yes
Sponsored by: ABT Systems Ltd
Differential Revision: https://reviews.freebsd.org/D8138
For the extended attributes the order between z_teardown_lock and the
vnode lock is different.
The bug was triggered only with DIAGNOSTIC turned on.
This fix is developed in cooperation with avos.
PR: 213112
Reported by: avos
Tested by: avos
MFC after: 1 week
This list is incomplete, however we don't have the ID values for the
missing Cortex-A32 or A35.
Submitted by: loos (Cortex-A53)
Sponsored by: ABT Systems Ltd
descriptors are reclaimed as soon as possible.
Without this the free buffers are reclaimed only on watchdog runs or after
trying to enqueue more packets.
Sponsored by: Rubicon Communications, LLC (Netgte)
