Casueword(9) on ll/sc architectures must be prepared for userspace
constantly modifying the same cache line as containing the CAS word,
and not loop infinitely. Otherwise, rogue userspace livelocks the
kernel.
To fix the issue, change casueword(9) interface to return new value 1
indicating that either comparision or store failed, instead of relying
on the oldval == *oldvalp comparison. The primitive no longer retries
the operation if it failed spuriously. Modify callers of
casueword(9), all in kern_umtx.c, to handle retries, and react to
stops and requests to terminate between retries.
On x86, despite cmpxchg should not return spurious failures, we can
take advantage of the new interface and just return PSL.ZF.
Reviewed by: andrew (arm64, previous version), markj
Tested by: pho
Reported by: https://xenbits.xen.org/xsa/advisory-295.txt
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
Differential revision: https://reviews.freebsd.org/D20772
The hold_count and wire_count fields of struct vm_page are separate
reference counters with similar semantics. The remaining essential
differences are that holds are not counted as a reference with respect
to LRU, and holds have an implicit free-on-last unhold semantic whereas
vm_page_unwire() callers must explicitly determine whether to free the
page once the last reference to the page is released.
This change removes the KPIs which directly manipulate hold_count.
Functions such as vm_fault_quick_hold_pages() now return wired pages
instead. Since r328977 the overhead of maintaining LRU for wired pages
is lower, and in many cases vm_fault_quick_hold_pages() callers would
swap holds for wirings on the returned pages anyway, so with this change
we remove a number of page lock acquisitions.
No functional change is intended. __FreeBSD_version is bumped.
Reviewed by: alc, kib
Discussed with: jeff
Discussed with: jhb, np (cxgbe)
Tested by: pho (previous version)
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D19247
Change to use registers instead of register, as it is customary to use
plural when talking about PCI registers.
This was missed in r349150.
MFC after: 3 days
Submitted by: Ka Ho Ng <khng300 at gmail.com>
Reviewed by: mckusick
MFC after: 3 days
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D20695
Unmapped mbufs allow sendfile to carry multiple pages of data in a
single mbuf, without mapping those pages. It is a requirement for
Netflix's in-kernel TLS, and provides a 5-10% CPU savings on heavy web
serving workloads when used by sendfile, due to effectively
compressing socket buffers by an order of magnitude, and hence
reducing cache misses.
For this new external mbuf buffer type (EXT_PGS), the ext_buf pointer
now points to a struct mbuf_ext_pgs structure instead of a data
buffer. This structure contains an array of physical addresses (this
reduces cache misses compared to an earlier version that stored an
array of vm_page_t pointers). It also stores additional fields needed
for in-kernel TLS such as the TLS header and trailer data that are
currently unused. To more easily detect these mbufs, the M_NOMAP flag
is set in m_flags in addition to M_EXT.
Various functions like m_copydata() have been updated to safely access
packet contents (using uiomove_fromphys()), to make things like BPF
safe.
NIC drivers advertise support for unmapped mbufs on transmit via a new
IFCAP_NOMAP capability. This capability can be toggled via the new
'nomap' and '-nomap' ifconfig(8) commands. For NIC drivers that only
transmit packet contents via DMA and use bus_dma, adding the
capability to if_capabilities and if_capenable should be all that is
required.
If a NIC does not support unmapped mbufs, they are converted to a
chain of mapped mbufs (using sf_bufs to provide the mapping) in
ip_output or ip6_output. If an unmapped mbuf requires software
checksums, it is also converted to a chain of mapped mbufs before
computing the checksum.
Submitted by: gallatin (earlier version)
Reviewed by: gallatin, hselasky, rrs
Discussed with: ae, kp (firewalls)
Relnotes: yes
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D20616
This change was originally in D20378. Making it in a new diff since it's a
bugfix.
Submitted by: alfredo.junior_eldorado.org.br
Reviewed by: emaste, luporl
Differential Revision: https://reviews.freebsd.org/D20756
The epoch_drain_callbacks() function is used to drain all pending
callbacks which have been invoked by prior epoch_call() function calls
on the same epoch. This function is useful when there are shared
memory structure(s) referred to by the epoch callback(s) which are not
refcounted and are rarely freed. The typical place for calling this
function is right before freeing or invalidating the shared
resource(s) used by the epoch callback(s). This function can sleep and
is not optimized for performance.
Differential Revision: https://reviews.freebsd.org/D20109
MFC after: 1 week
Sponsored by: Mellanox Technologies
"pin_list" allows to specify child pins as a list of pin numbers.
Existing hint "pins" serves the same purpose but with a 32-bit wide bit
mask. One problem with that is that a controller can have more than 32
pins. One example is amdgpio. Also, a list of numbers is a little bit
more human friendly than a matching bit mask. As a side note, it seems
that in FDT pins are typically specified by their numbers as well.
This commit also adds accessors for instance variables (IVARs) that
define the child pins. My primary goal is to allow a child to be
configured programmatically rather than via hints (assuming that FDT is
not supported on a platform). Also, while a child should not care about
specific pin numbers that are allocated to it, it could be interested in
how many were actually assigned to it.
While there, I removed "flags" instance variable. It was unused.
Reviewed by: mizhka
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D20459
"at" keyword is documented in device.hints(5) for all buses, but it does
hurt to add another reference to it.
"pins" keyword is specific to gpiobus.
At least these two hints should be configured for any gpiobus device on
a hints based system.
MFC after: 10 days
NANDFS has been broken for years. Remove it. The NAND drivers that
remain are for ancient parts that are no longer relevant. They are
polled, have terrible performance and just for ancient arm
hardware. NAND parts have evolved significantly from this early work
and little to none of it would be relevant should someone need to
update to support raw nand. This code has been off by default for
years and has violated the vnode protocol leading to panics since it
was committed.
Numerous posts to arch@ and other locations have found no actual users
for this software.
Relnotes: Yes
No Objection From: arch@
Differential Revision: https://reviews.freebsd.org/D20745
Summary:
Toolchain follow-up to r349350. LLVM patches will be submitted upstream for
9.0 as well.
The bsd.cpu.mk change is required because GNU ld assumes BSS-PLT if it
cannot determine for certain that it needs Secure-PLT, and some binaries do
not compile in such a way to make it know to use Secure-PLT.
Reviewed By: nwhitehorn, bdragon, pfg
Differential Revision: https://reviews.freebsd.org/D20598
Summary:
Adds a list of valid CPUTYPE flags for arm and arm64 architectures
List taken from share/mk/bsd.cpu.mk
Submitted by: Daniel Engberg
Reviewed By: imp
Differential Revision: https://reviews.freebsd.org/D20315
'-E' appears on the swapon command line, or if "trimonce" appears as
an fstab option.
Discussed at: BSDCAN
Tested by: markj
Reviewed by: markj
Approved by: markj (mentor)
Differential Revision:https://reviews.freebsd.org/D20599
Sort methods alphabetically. Wrap long lines. Start sentences on a new
line. Remove contractions (not because it's a good idea, just to silence
igor). Add some explanation of the units for the period and duty arguments
and the convention for channel numbers.
interfaces were unified into pwmbus(9), and the PWMBUS_CHANNEL_MAX method
was renamed PWMBUS_CHANNEL_COUNT. The pwmbus_attach_bus() function just
went away completely. Also, fix a few typos such as s/is/if/.
As reported in review D20709 by brooks calling vm_map_protect to set a
new max_protection value downgrades existing mappings if necessary (as
opposed to returning an error).
Reported by: brooks
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
It's implied by the man page's RETURN VALUES section, but be explicit in
the description that vm_map_protect can not set new protection bits that
are already in each entry's max_protection.
Reviewed by: brooks
MFC After: 1 week
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D20709
wakeup_one() and underlying sleepq_signal() spend additional time trying
to be fair, waking thread with highest priority, sleeping longest time.
But in case of taskqueue there are many absolutely identical threads, and
any fairness between them is quite pointless. It makes even worse, since
round-robin wakeups not only make previous CPU affinity in scheduler quite
useless, but also hide from user chance to see CPU bottlenecks, when
sequential workload with one request at a time looks evenly distributed
between multiple threads.
This change adds new SLEEPQ_UNFAIR flag to sleepq_signal(), making it wakeup
thread that went to sleep last, but no longer in context switch (to avoid
immediate spinning on the thread lock). On top of that new wakeup_any()
function is added, equivalent to wakeup_one(), but setting the flag.
On top of that taskqueue(9) is switchied to wakeup_any() to wakeup its
threads.
As result, on 72-core Xeon v4 machine sequential ZFS write to 12 ZVOLs
with 16KB block size spend 34% less time in wakeup_any() and descendants
then it was spending in wakeup_one(), and total write throughput increased
by ~10% with the same as before CPU usage.
Reviewed by: markj, mmacy
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
Differential Revision: https://reviews.freebsd.org/D20669
This is still targeting bin/sh cyclic dependency issues. Only apply
guessed dependencies that are explicitly set for an object (which
gnu/lib/cc/cc_tools needs) and if no custom target exists with its
own dependencies.
This was manifesting as a missing yacc.h in usr.bin/mkesdb_static when
built without -j (or -B). No actual yacc.h dependency ordering was
defined but with -j it got lucky and built fine.
Before r349061 the behavior was different for META_MODE but that logic
difference isn't needed.
X-MFC-With: r349061
Sponsored by: DellEMC
NetBSD 7.0 was a separate branch, subsequent 8.x releases did not emerge from
this branch.
Clean up minor visual nits, centre OpenBSD listing on the B, DragonFly
listings on the y.
Add missing words after PCI in the description of the PCIOCWRITE and
PCIOCATTACHED ioctls.
Use singular in PCIOCREAD, we only read one register at the time.
Reviewed by: bcr, bjk, rgrimes, cem
MFC after: 2 weeks
X-MFC-with: r349133
Differential Revision: https://reviews.freebsd.org/D20671
Document the PCIOCATTACHED ioctl(2) in the pci(4) manual.
PCIOCATTACHED is used to query if a driver has attached to a PCI.
Reviewed by: bcr, imp
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D20652
Default to tracking .depend.* for OBJS rather than SRCS.
This helps cover some special case builds like gnu/lib/csu which
do more of a PROGS-like thing with bsd.prog.mk.
It is possible this causes out-of-tree Makefiles to have problems if they use
this pattern:
foo.o: foo.c
${CC} -o ${.TARGET} ${.ALLSRC}
This may cause multiple source files to be compiled due to finding the
'foo.o: foo.c' dependency both in the Makefile at the .depend file. Or
it may try compiling headers. This can be worked around by either of these:
foo.o: foo.c
${CC} -o ${.TARGET} ${.ALLSRC:N*.h:[1]}
Or
foo.o: foo.c
${CC} -o ${.TARGET} ${.CURDIR}/foo.c
In the latter case the ${.CURDIR} may need to be a different path. The
first case covers automatically using .PATH.
Sponsored by: DellEMC
If a meta mode change is triggered but then the build fails then the
next build will not retrigger meta mode. This only prevented by
removing the target on rebuild or on the failure to rebuild.
Sponsored by: DellEMC
