adrian [Sun, 5 Jun 2016 07:55:21 +0000 (07:55 +0000)]
[bwn] Enable PHY-LP 5GHz support for the one NIC I have tested it on so far.
After perusing the PHY-LP code (don't ask why; honest) I discovered that
it /has/ 5GHz support - but it's not ever used. I found one NIC - a
BCM4312 w/ pci id 0x4315 - which advertised dual-band PHY-LP support.
Turns out it works.
Whilst here, move up the support bit logging code so I can use it
to debug this.
arybchik [Sun, 5 Jun 2016 06:37:54 +0000 (06:37 +0000)]
sfxge(4): allow firmware to auto-configure event queues on Medford
On Medford, licenses are required to enable RX and event cut through and to
disable RX batching. To avoid the need for the driver to make decisions based on
the licensing state, the MC_CMD_INIT_EVQ has been extended to allow us to leave
the decision to the firmware. If the adapter is licensed for low-latency use,
the firmware will choose the optimal settings for latency, otherwise it will use
the best settings for throughput.
For Huntington we still need to choose the settings ourselves.
Submitted by: Mark Spender <mspender at solarflare.com>
Sponsored by: Solarflare Communications, Inc.
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D6717
adrian [Sun, 5 Jun 2016 05:55:27 +0000 (05:55 +0000)]
[ath_hal] add AR9462 (jupiter) RX gain / XLNA programming.
This seems to make 5G work better.
It doesn't fix powersave handling though, that still sees the PHY get
stuck during initial calibration and everything goes pear shaped.
I'll look into that later.
adrian [Sun, 5 Jun 2016 02:17:51 +0000 (02:17 +0000)]
[ath_hal] Add Jupiter 2.1 (AR9462 mac 640.3) support.
Turns out I wasn't even initialising or programming a lot of stuff
for the AR9462 2.1 chip. Oops.
This mostly gets it working. powersave scan results in some pretty
hilarious NFcal hangs and I don't see beacons reliably.
There are still some xlna gain tables missing that ath9k has; I'll
follow up with some fixes and then see if the QCAFN222 NIC I have
tests this path.
lidl [Sun, 5 Jun 2016 01:40:08 +0000 (01:40 +0000)]
Correct path for blacklistd.conf in OptionalObsoleteFiles.inc
Fix typo made when adding the blacklistd.conf file to OptionalObsoleteFiles.inc
Submitted by: Herbert J. Skuhra ( herbert at mailbox.org )
Reviewed by: rpaulo
Approved by: rpaulo
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D6715
landonf [Sat, 4 Jun 2016 19:58:01 +0000 (19:58 +0000)]
bhnd(4): support IPX OTP NVRAM/SPROM data source
Add support for fetching SPROM data from OTP on chipsets with an
IPX OTP controller (including the BCM43225).
This integrates the NVRAM data source into the chipc_caps capability
structure, and adds a sprom_offset field that can be used with OTP
to locate the SPROM image data (found within the general use
region, H/W subregion).
This also removes one of two duplicate parse error messages reported by
both the bhnd_sprom driver and the underlying SPROM parsing API.
Approved by: adrian (mentor)
Differential Revision: https://reviews.freebsd.org/D6729
landonf [Sat, 4 Jun 2016 19:53:47 +0000 (19:53 +0000)]
Switch mips/sentry5 to bhnd(4), and unify with mips/broadcom
Now that bhnd(4) provides feature parity with the previous siba/mips
implementation, we can switch sentry5 over and begin lifting common
support code out for use across bhnd(4) embedded targets.
Changes:
- Fixed enumeration of siba(4) per-core address maps, required for
discovery of memory mapped chipc flash region on siba(4) devices.
- Simplified bhnd kernel configuration (dropped 'bhndbus' option).
- Replaced files.broadcom's direct file references with their
corresponding standard kernel options.
- Lifted out common bcma/siba nexus support, inheriting from the new
generic bhnd_nexus driver.
- Dropped now-unused sentry5 siba code.
- Re-integrated BCM into the universe build now that it actually compiles.
Approved by: adrian (mentor)
Differential Revision: https://reviews.freebsd.org/D6712
landonf [Sat, 4 Jun 2016 19:31:06 +0000 (19:31 +0000)]
Fix several MIPS/BCM-blocking bugs in bhnd(4) chipc
- Correct IRQ lines for UART (to fix IRQ lookup in future)
- Check device unit in resource assignment during chipc_add_child
- If chipc hint->size is RM_MAX_END, resource end should be same as window end
- Clear reference from resource list entry to resource in case of resource release
- Add CHIPC_GET_CAPS implementation
- Correct chipc flash constants (to be unshifted)
- Default implementation of get_attach_type should iterate over device tree
- Add default implementation for BHND_CHIPC_GET_CAPS usable by chipc grandchildren
Submitted by: Michael Zhilin <mizhka@gmail.com>
Approved by: adrian (mentor)
Differential Revision: https://reviews.freebsd.org/D6584
Don't test for INKERNEL to check whether we're in kernel space.
It turns out that <machine/param.h> actually defines a macro under this
name, even when we're not in kernelspace. This causes us to suppress
some macro definitions that are used by userspace apps.
bdrewery [Sat, 4 Jun 2016 17:17:10 +0000 (17:17 +0000)]
Follow-up r301287: Pass external compiler metadata when used.
This fixes WITH_SYSTEM_COMPILER, WITHOUT_CROSS_COMPILER, external compiler, to
pass the external compiler metadata rather than the ${CC} metadata. On
a build host that has clang as CC it was passing the clang metadata rather than
GCC metadata during the build.
arybchik [Sat, 4 Jun 2016 09:20:46 +0000 (09:20 +0000)]
sfxge(4): always be ready to receive batched events
When the low-latency firmware variant is running, it is reported as not
being capable of batching RX events, but it can still do so if the
FORCE_EV_MERGING flag is set on an RXQ. Therefore we need to handle
batched RX events even if the capability isn't set.
If this bug is fixed in the firmware such that the capability is set
even when running the low-latency firmware variant, it will almost
always be reported so I don't think we lose much by removing the check.
Submitted by: Mark Spender <mspender at solarflare.com>
Sponsored by: Solarflare Communications, Inc.
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D6705
arybchik [Sat, 4 Jun 2016 09:17:45 +0000 (09:17 +0000)]
sfxge(4): add helper to compute timer quantum
This also adjusts the timer values used to match the Linux net
driver implementation:
a) non-zero time intervals should result in at least one quantum
b) timer load/reload values are only zero biased for Falcon/Siena
Submitted by: Andy Moreton <amoreton at solarflare.com>
Sponsored by: Solarflare Communications, Inc.
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D6704
andrew [Sat, 4 Jun 2016 08:47:45 +0000 (08:47 +0000)]
Use the UEFI event timer to update the time on arm and arm64. The current
code uses the GetTime function from the Runtime Service, however this has
been shown to not return a useable time on many arm64 UEFI implementations.
mav [Sat, 4 Jun 2016 00:18:59 +0000 (00:18 +0000)]
When negotiating NTB_SB01BASE_LOCKUP workaround, don't try to limit the
BAR size to 1MB. According to Xeon v3 specifications and my tests, that
size register is write-once and so not writeable after BIOS written it.
Instead of that, make the code work with BAR of any sufficient size,
properly calculating offset within its base. It also simplifies the code.
Discussed with: cem
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
bdrewery [Fri, 3 Jun 2016 20:49:49 +0000 (20:49 +0000)]
WITHOUT_CROSS_COMPILER: Fix installworld.
Since no WORLDTMP/usr/bin/cc is created, cc cannot be found
during installworld time since /usr/bin is not in the PATH.
Pass along the known compiler metadata to allow installworld
to work. The same fix was used for WITH_SYSTEM_COMPILER.
A better route would be to store a cookie in buildworld
containing this compiler metadata and then using that
at install time, rather than rerunning cc.
Ccache will not affect the output of the objects, so just ignore it for
meta mode handling. This avoids having everything rebuild if ccache is
updated.
bdrewery [Fri, 3 Jun 2016 19:25:41 +0000 (19:25 +0000)]
WITH_META_MODE: Don't expect meta files for side-effect generated files.
The first file in these lists will generate everything else so only
it should be getting a .meta file. With bmake's missing=yes meta
feature these would otherwise cause a rebuild without the
.NOMETA hint.
zbb [Fri, 3 Jun 2016 18:54:16 +0000 (18:54 +0000)]
Use proper interface for FDT parsing and memory mapping in CESA
Improvements after r301220.
Bus space methods are not called so simple pmap_mapdev will suffice.
Use OF_getencprop to get buffer with already converted endianess.
Pointed out by: ian
Submitted by: Michal Stanek <mst@semihalf.com>
Obtained from: Semihalf
garga [Fri, 3 Jun 2016 18:37:56 +0000 (18:37 +0000)]
One of the already implemented options in release/Makefile is NOSRC. When
it's defined, installation image is shipped without source distribution
(src.txz)
Add the hability of defining NOSRC in release.conf and pass it to
'make release' argument
kib [Fri, 3 Jun 2016 18:23:45 +0000 (18:23 +0000)]
Trim some spaces to record correct commit message for the r301278.
Reduce number of iterations used for calibrating ICR read loop. The
new number of iteration still gives the same ICR latency as before,
tested on Intel SandyBridge and Haswell machines, and on AMD. But it
significantly reduces the unneeded pause on boot in some VMs, from ~10
secs to less then 1 sec. It was reported to occur in bhyve on AMD
host.
Reported and tested by: avg
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
#ifdef SMP
-#define LOOPS 1000000
+#define LOOPS 100000
/*
* Calibrate the busy loop waiting for IPI ack in xAPIC mode.
* lapic_ipi_wait_mult contains the number of iterations which
dim [Fri, 3 Jun 2016 16:17:36 +0000 (16:17 +0000)]
For clang, move the definition of FREEBSD_CC_VERSION into its own header
file, lib/clang/freebsd_cc_version.h, instead of reusing Version.inc.
The header is only included from one .cpp file in the clang tree.
This minimizes the number of .cpp files that need to be rebuilt if the
version is bumped.
avg [Fri, 3 Jun 2016 14:37:18 +0000 (14:37 +0000)]
zfs: set VROOT / VV_ROOT consistently and in a single place
This is a followup to r300131.
A filesystem's root vnode can be reached not only through VSF_ROOT, but
by other means as well. For example, via a dot-dot lookup.
Also, a root vnode can get reclaimed and then re-created. For these
reasons it was insufficient to clear VV_ROOT flag from a root vnode of a
snapshot mounted under .zfs in zfsctl_snapdir_lookup().
So, now we set the flag in zfs_znode_sa_init() only if a vnode
represent a root of a filesystem or a standalone snapshot.
That is, the flag is not set for snapshots mounted under .zfs.
vangyzen [Fri, 3 Jun 2016 14:30:32 +0000 (14:30 +0000)]
Improve errno documentation in pthread_create(3) and thr_new(2)
Add some missing errno values to thr_new(2) and pthread_create(3).
In particular, EDEADLK was not documented in the latter.
While I'm here, improve some English and cross-references.
Reviewed by: kib
Sponsored by: Dell Inc.
Differential Revision: https://reviews.freebsd.org/D6663
Introduce a per-VNET flag to enable/disable netisr prcessing on that VNET.
Add accessor functions to toggle the state per VNET.
The base system (vnet0) will always enable itself with the normal
registration. We will share the registered protocol handlers in all
VNETs minimising duplication and management.
Upon disabling netisr processing for a VNET drain the netisr queue from
packets for that VNET.
Update netisr consumers to (de)register on a per-VNET start/teardown using
VNET_SYS(UN)INIT functionality.
The change should be transparent for non-VIMAGE kernels.
Reviewed by: gnn (, hiren)
Obtained from: projects/vnet
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D6691
royger [Fri, 3 Jun 2016 11:39:35 +0000 (11:39 +0000)]
xen-blkback: fix error path on failed attach
The current error path in case of failure during attach/initialization is
not correct and leaves blkback in a stuck state. This is due to blkback
waiting for blkfront to switch to state XenbusStateClosed, but if blkfront
never attached (because the guest is not even started) it cannot possibly
make it to that state.
Instead just wait for the frontend to be in a state different than
XenbusStateConnected in order to proceed with the shutdown. Also, it is
wrong to call xbb_detach directly because it destroys the lock which can
still be used by xbb_frontend_changed.
royger [Fri, 3 Jun 2016 11:38:52 +0000 (11:38 +0000)]
blkback: add support for hotplug scripts
Hotplug scripts are needed in order to use fancy disk configurations in xl,
like iSCSI disks. The job of hotplug scripts is to locally attach the disk
and present it to blkback as a block device or a regular file.
This change introduces a new xenstore node in the blkback hierarchy, called
"physical-device-path". This is a straigh replacement for the "params" node,
which was used before.
Hotplug scripts will need to read the "params" node, perform whatever
actions are necessary and then write the "physical-device-path" node. The
hotplug script is also in charge of detaching the disk once the domain has
been shutdown.
skra [Fri, 3 Jun 2016 11:05:55 +0000 (11:05 +0000)]
Postpone allocation of IRQ resource to the time when interrupt
controller devices are attached. This has already been done for
bus_setup_intr().
There was no doubt that if someone wants to setup an interrupt,
corresponding interrupt controller device must already be attached.
However, the same must be valid for allocation of an interrupt resource
unless the allocation is done blindly, without any information that
such interrupt even exists. While it was done this blind way before,
it won't be possible after next INTRNG change.
andrew [Fri, 3 Jun 2016 10:28:06 +0000 (10:28 +0000)]
Add the GICv3 ITS intrng driver. As the interface to the interrupt
framework has significantly changed the driver has moved to a new file.
While it shares some code with the existing driver this has been modified
to work better with the intrng framework.
This has been tested on the ThunderX servers in the netperf cluster and has
been used to boot them for other testing, including DTrace and hwpmc.
With this we can use intrng on all supported arm64 platforms I was able to
test on. It is expected we will move to intrng soon, and disable the old
arm64 interrupt framework.
Obtained from: ABT Systems Ltd
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D6437
andrew [Fri, 3 Jun 2016 10:13:18 +0000 (10:13 +0000)]
Add an interface to handle interrupt controllers that have a contiguous
range of interrupts they pass to a second controller driver to handle.
The parent driver is expected to detect when one of these interrupts has
been triggered and call intr_child_irq_handler to pass the interrupt to
a child. The children controllers are then expected to manage the range
by allocating interrupts as needed.
This will initially be used by the ARM GICv3 driver, but is is expected to
be useful for other driver where this type of allocation applies.
Obtained from: ABT Systems Ltd
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D6436
hselasky [Fri, 3 Jun 2016 08:35:07 +0000 (08:35 +0000)]
Use insertion sort instead of bubble sort in TCP LRO.
Replacing the bubble sort with insertion sort gives an 80% reduction
in runtime on average, with randomized keys, for small partitions.
If the keys are pre-sorted, insertion sort runs in linear time, and
even if the keys are reversed, insertion sort is faster than bubble
sort, although not by much.
Update comment describing "tcp_lro_sort()" while at it.
lidl [Fri, 3 Jun 2016 06:15:52 +0000 (06:15 +0000)]
Add blacklistd.conf manpage
Install the blacklistd.conf man page, missed in the original commit.
Submitted by: Herbert J. Skuhra ( herbert at mailbox.org )
Reviewed by: rpaulo
Approved by: rpaulo
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D6702
dim [Thu, 2 Jun 2016 19:54:38 +0000 (19:54 +0000)]
Pull in r271548 from upstream llvm trunk (by me):
Only attempt to detect AVG if SSE2 is available
Summary:
In PR29973 Sanjay Patel reported an assertion failure when a certain
loop was optimized, for a target without SSE2 support. It turned out
this was because of the AVG pattern detection introduced in rL253952.
Prevent the assertion failure by bailing out early in
`detectAVGPattern()`, if the target does not support SSE2.
zbb [Thu, 2 Jun 2016 18:41:33 +0000 (18:41 +0000)]
Add support for CESA on Armada38x
Changes:
- added new SoC ID in CESA attach
- allowed crypto driver IDs other than 0
- added CESA nodes to Armada38x .dts files
- enabled required devices in kernconf
zbb [Thu, 2 Jun 2016 18:39:33 +0000 (18:39 +0000)]
Add HMAC-SHA256 support in CESA
Only HMAC-SHA256 is added as it is the only SHA-2 variant supported by
cryptodev. It is not possible to register hardware support for other
algorithms in the family including regular non-keyed SHA256.
zbb [Thu, 2 Jun 2016 18:37:50 +0000 (18:37 +0000)]
Truncate HMAC output only if requested by the client
The output of HMAC was previously truncated to 12 bytes. This was only
correct in case of one particular crypto client - the new version of IPSEC.
Fix by taking into account the cri_mlen field in cryptoini session request
filled in by the client.
zbb [Thu, 2 Jun 2016 18:35:35 +0000 (18:35 +0000)]
Split CESA memory resource into TDMA and CESA regs
TDMA and CESA registers are placed in different ranges of memory. Split
memory resource in DTS to reflect that. This change is needed to support
multiple CESA nodes as otherwise the ranges of different nodes would
overlap.
In consequence, CESA_WRITE and CESA_READ macros have been split depending
on which range of registers is accessed. Offsets for CESA registers have
been modified as the base address has changed.
zbb [Thu, 2 Jun 2016 18:31:36 +0000 (18:31 +0000)]
Map CESA SRAM memory in driver attach for Armada38x
On other platforms with CESA accelerator the SRAM memory is mapped in
early init before driver is attached. This method only works correctly
with mappings no smaller than L1 section size (1MB). There may be more
SRAM blocks and they may have smaller sizes than 1MB as is the case
for Armada38x. Instead, map SRAM memory with bus_space_map() in CESA
driver attach. Note that we can no longer assume that VA == PA for the
SRAM.
lidl [Thu, 2 Jun 2016 18:25:32 +0000 (18:25 +0000)]
Fixup path in NetBSD supplied documentation for FreeBSD
NetBSD installs the blacklist-helper script in /libexec, and
it goes into /usr/libexec on FreeBSD. Update the docs to
match FreeBSD's installation location.
Reviewed by: rpaulo
Approved by: rpaulo
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D6592
gnn [Thu, 2 Jun 2016 17:51:29 +0000 (17:51 +0000)]
This change re-adds L2 caching for TCP and UDP, as originally added in D4306
but removed due to other changes in the system. Restore the llentry pointer
to the "struct route", and use it to cache the L2 lookup (ARP or ND6) as
appropriate.
Submitted by: Mike Karels
Differential Revision: https://reviews.freebsd.org/D6262
pfg [Thu, 2 Jun 2016 17:28:39 +0000 (17:28 +0000)]
citrus: Remove redundant code in _citrus_esdb_get_list().
It appears "sorted" may have not been implemented. Sorted or not,
we always follow the same action so simplify the code.
Leave a note for future generations.
markj [Thu, 2 Jun 2016 17:17:15 +0000 (17:17 +0000)]
Always start IPv6 DAD asynchronously.
Otherwise we transmit the first neighbour solicitation in the context of the
caller of nd6_dad_start(), which can easily result in lock recursion. When
DAD is to be started after some delay, we send the first NS from the DAD
callout handler, so just change the implementation to do this in the
non-delayed case as well.