Previously vputx would detect the condition and clear the flag.
With this change it is invalid to have both v_usecount > 0 and the flag
set. Assert the condition is met in all revlevant places.
Reviewed by: kib
Previously several places were doing it on its own, partially
incorrectly (e.g. without the filedesc locked) or even actively harmful
by populating jdir or assigning rootvnode without vrefing it.
Reviewed by: kib
This is based on work done by jeff@ and jhb@, as well as the numa.diff
patch that has been circulating when someone asks for first-touch NUMA
on -10 or -11.
* Introduce a simple set of VM policy and iterator types.
* tie the policy types into the vm_phys path for now, mirroring how
the initial first-touch allocation work was enabled.
* add syscalls to control changing thread and process defaults.
* add a global NUMA VM domain policy.
* implement a simple cascade policy order - if a thread policy exists, use it;
if a process policy exists, use it; use the default policy.
* processes inherit policies from their parent processes, threads inherit
policies from their parent threads.
* add a simple tool (numactl) to query and modify default thread/process
policities.
* add documentation for the new syscalls, for numa and for numactl.
* re-enable first touch NUMA again by default, as now policies can be
set in a variety of methods.
This is only relevant for very specific workloads.
This doesn't pretend to be a final NUMA solution.
The previous defaults in -HEAD (with MAXMEMDOM set) can be achieved by
'sysctl vm.default_policy=rr'.
This is only relevant if MAXMEMDOM is set to something other than 1.
Ie, if you're using GENERIC or a modified kernel with non-NUMA, then
this is a glorified no-op for you.
Thank you to Norse Corp for giving me access to rather large
(for FreeBSD!) NUMA machines in order to develop and verify this.
Thank you to Dell for providing me with dual socket sandybridge
and westmere v3 hardware to do NUMA development with.
Thank you to Scott Long at Netflix for providing me with access
to the two-socket, four-domain haswell v3 hardware.
Thank you to Peter Holm for running the stress testing suite
against the NUMA branch during various stages of development!
Tested:
* MIPS (regression testing; non-NUMA)
* i386 (regression testing; non-NUMA GENERIC)
* amd64 (regression testing; non-NUMA GENERIC)
* westmere, 2 socket (thankyou norse!)
* sandy bridge, 2 socket (thankyou dell!)
* ivy bridge, 2 socket (thankyou norse!)
* westmere-EX, 4 socket / 1TB RAM (thankyou norse!)
* haswell, 2 socket (thankyou norse!)
* haswell v3, 2 socket (thankyou dell)
* haswell v3, 2x18 core (thankyou scott long / netflix!)
* Peter Holm ran a stress test suite on this work and found one
issue, but has not been able to verify it (it doesn't look NUMA
related, and he only saw it once over many testing runs.)
* I've tested bhyve instances running in fixed NUMA domains and cpusets;
all seems to work correctly.
Verified:
* intel-pcm - pcm-numa.x and pcm-memory.x, whilst selecting different
NUMA policies for processes under test.
Review:
This was reviewed through phabricator (https://reviews.freebsd.org/D2559)
as well as privately and via emails to freebsd-arch@. The git history
with specific attributes is available at https://github.com/erikarn/freebsd/
in the NUMA branch (https://github.com/erikarn/freebsd/compare/local/adrian_numa_policy).
This has been reviewed by a number of people (stas, rpaulo, kib, ngie,
wblock) but not achieved a clear consensus. My hope is that with further
exposure and testing more functionality can be implemented and evaluated.
Notes:
* The VM doesn't handle unbalanced domains very well, and if you have an overly
unbalanced memory setup whilst under high memory pressure, VM page allocation
may fail leading to a kernel panic. This was a problem in the past, but it's
much more easily triggered now with these tools.
* This work only controls the path through vm_phys; it doesn't yet strongly/predictably
affect contigmalloc, KVA placement, UMA, etc. So, driver placement of memory
isn't really guaranteed in any way. That's next on my plate.
Sponsored by: Norse Corp, Inc.; Dell
objects, i.e. for buffer objects which vnode was reclaimed. Buffer
cache cannot write such buffers. Return the error and discard the
buffer immediately on write attempt.
BO_DIRTY now always set during vnode reclamation, since it is used not
only for the INVARIANTS checks. Do allow placement of the clean
buffers on dead bufobj list, otherwise filesystems cannot use bufcache
at all after the devvp reclaim.
Reported and tested by: trasz
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
- make mode enum start from 0 so that the assertion covers all cases [1]
- rename prefix _CLOEXEC flag with _FLAG
- postpone fhold on the old file descriptor, which eliminates the need to fdrop
in error cases.
- fixup FDDUP_FCNTL check missed in the previous commit
This removes 'fp == oldfde->fde_file' assertion which had little value. kern_dup
only calls fd-related functions which cannot drop the lock or a whole lot of
races would be introduced.
Noted by: kib [1]
to more C11-ish atomic_thread_fence_seq_cst().
Note that on PowerPC, which currently uses lwsync for mb(), the change
actually fixes the missed store/load barrier, intended by r271604 [*].
Reviewed by: alc
Noted by: alc [*]
Sponsored by: The FreeBSD Foundation
MFC after: 3 weeks
We currently return EINVAL when calling listen() on a UNIX socket that
has not been bound to a pathname. If my interpretation of POSIX is
correct, we should return EDESTADDRREQ: "The socket is not bound to a
local address, and the protocol does not support listening on an unbound
socket."
Return EDESTADDRREQ instead when not bound and not connected.
Differential Revision: https://reviews.freebsd.org/D3038
Reviewed by: gnn, network
The logic is reorganised so that there is one exit point prior to the
lookup loop. This is an intermediate step to making audit logging
functions use found vnode instead of translating ni_dirfd on their own.
ni_startdir validation is removed. The only in-tree consumer is nfs
which already makes sure it is a directory.
Reviewed by: kib
All of the CloudABI system calls that operate on file descriptors of an
arbitrary type are prefixed with fd_. This change adds wrappers for
most of these system calls around their FreeBSD equivalents.
The dup2() system call present on CloudABI deviates from POSIX, in the
sense that it can only be used to replace existing file descriptor. It
cannot be used to create new ones. The reason for this is that this is
inherently thread-unsafe. Furthermore, there is no need on CloudABI to
use fixed file descriptor numbers. File descriptors 0, 1 and 2 have no
special meaning.
This change exposes the kern_dup() through <sys/syscallsubr.h> and puts
the FDDUP_* flags in <sys/filedesc.h>. It then adds a new flag,
FDDUP_MUSTREPLACE to force that file descriptors are replaced -- not
allocated.
Differential Revision: https://reviews.freebsd.org/D3035
Reviewed by: mjg
namei used to vref fd_cdir, which was immediatley vrele'd on entry to
the loop.
Check for absolute lookup and vref the right vnode the first time.
Reviewed by: kib
fd_rdir vnode was stored in ni_rootdir without refing it in any way,
after which the filedsc lock was being dropped.
The vnode could have been freed by mountcheckdirs or another thread doing
chroot.
VREF the vnode while the lock is held.
Reviewed by: kib
MFC after: 1 week
While writing tests for CloudABI, I noticed that close() on process
descriptors returns the process ID of the child process. This is
interesting, as close() is only allowed to return 0 or -1. It turns out
that we clobber td->td_retval[0] in proc_reap(), so that wait*()
properly returns the process ID.
Change proc_reap() to leave td->td_retval[0] alone. Set the return value
in kern_wait6() instead, by keeping track of the PID before we
(potentially) reap the process.
Differential Revision: https://reviews.freebsd.org/D3032
Reviewed by: kib
loop finds the selfd entry and clears its sf_si pointer, which is
handled by selfdfree() in parallel, NULL sf_si makes selfdfree() free
the memory. The result is the race and accesses to the freed memory.
Refcount the selfd ownership. One reference is for the sf_link
linkage, which is unconditionally dereferenced by selfdfree().
Another reference is for sf_threads, both selfdfree() and
doselwakeup() race to deref it, the winner unlinks and than frees the
selfd entry.
Reported by: Larry Rosenman <ler@lerctr.org>
Tested by: Larry Rosenman <ler@lerctr.org>, pho
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
CloudABI is a pure capability-based runtime environment for UNIX. It
works similar to Capsicum, except that processes already run in
capabilities mode on startup. All functionality that conflicts with this
model has been omitted, making it a compact binary interface that can be
supported by other operating systems without too much effort.
CloudABI is 'secure by default'; the idea is that it should be safe to
run arbitrary third-party binaries without requiring any explicit
hardware virtualization (Bhyve) or namespace virtualization (Jails). The
rights of an application are purely determined by the set of file
descriptors that you grant it on startup.
The datatypes and constants used by CloudABI's C library (cloudlibc) are
defined in separate files called syscalldefs_mi.h (pointer size
independent) and syscalldefs_md.h (pointer size dependent). We import
these files in sys/contrib/cloudabi and wrap around them in
cloudabi*_syscalldefs.h.
We then add stubs for all of the system calls in sys/compat/cloudabi or
sys/compat/cloudabi64, depending on whether the system call depends on
the pointer size. We only have nine system calls that depend on the
pointer size. If we ever want to support 32-bit binaries, we can simply
add sys/compat/cloudabi32 and implement these nine system calls again.
The next step is to send in code reviews for the individual system call
implementations, but also add a sysentvec, to allow CloudABI executabled
to be started through execve().
More information about CloudABI:
- GitHub: https://github.com/NuxiNL/cloudlibc
- Talk at BSDCan: https://www.youtube.com/watch?v=SVdF84x1EdA
Differential Revision: https://reviews.freebsd.org/D2848
Reviewed by: emaste, brooks
Obtained from: https://github.com/NuxiNL/freebsd
for timehands consumers, by using fences.
Ensure that the timehands->th_generation reset to zero is visible
before the data update is visible [*]. tc_setget() allowed data update
writes to become visible before generation (but not on TSO
architectures).
Remove tc_setgen(), tc_getgen() helpers, use atomics inline [**].
Noted by: alc [*]
Requested by: bde [**]
Reviewed by: alc, bde
Sponsored by: The FreeBSD Foundation
MFC after: 3 weeks
The number of available lock list entries for a thread is LOCK_CHILDCOUNT,
and each entry can record up to LOCK_NCHILDREN locks. When iterating over
the locks held by a thread, a bound on the loop index is therefore given
by LOCK_CHILDCOUNT * LOCK_NCHILDREN; WITNESS_COUNT is an unrelated
constant.
Reviewed by: jhb
MFC after: 1 week
Sponsored by: EMC / Isilon Storage Division
Differential Revision: https://reviews.freebsd.org/D2974
Place sched_random nearer to where it's first used: moving the
code nearer to where it is used makes the code easier to read
and we can reduce the initial "#ifdef SMP" island.
Reword a little the comment and clean some whitespaces
while here.
pointer is NULL, as in that case there are no userland pages that
could potentially be wired. It is common for old to be NULL and
oldlenp to be non-NULL in calls to userland_sysctl(), as this is used
to probe for the length of a variable-length sysctl entry before
retrieving a value. Note that it is typical for such calls to be made
with an uninitialized value in *oldlenp, so sysctlmemlock was
essentially being acquired at random (depending on the uninitialized
value in *oldlenp being > PAGE_SIZE or not) for these calls prior to
this patch.
Differential Revision: https://reviews.freebsd.org/D2987
Reviewed by: mjg, kib
Approved by: jmallett (mentor)
MFC after: 1 month
This obviates the need for a MNTK_SUSPENDABLE flag, since passthrough
filesystems like nullfs and unionfs no longer need to inherit this
information from their lower layer(s). This change also restores the
pre-r273336 behaviour of using the presence of a susp_clean VFS method to
request suspension support.
Reviewed by: kib, mjg
Differential Revision: https://reviews.freebsd.org/D2937
directory sys/contrib/libnv.
The goal of this operation is to NOT install header files which shouldn't
be used outside the nvlist library.
Approved by: pjd (mentor)
asserts are made. Remove them, since we might dereference freed
memory. Leaked locks are asserted by the syscall return code anyway.
Reported and tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
* GENERAL
- Update copyright.
- Make kernel options for RANDOM_YARROW and RANDOM_DUMMY. Set
neither to ON, which means we want Fortuna
- If there is no 'device random' in the kernel, there will be NO
random(4) device in the kernel, and the KERN_ARND sysctl will
return nothing. With RANDOM_DUMMY there will be a random(4) that
always blocks.
- Repair kern.arandom (KERN_ARND sysctl). The old version went
through arc4random(9) and was a bit weird.
- Adjust arc4random stirring a bit - the existing code looks a little
suspect.
- Fix the nasty pre- and post-read overloading by providing explictit
functions to do these tasks.
- Redo read_random(9) so as to duplicate random(4)'s read internals.
This makes it a first-class citizen rather than a hack.
- Move stuff out of locked regions when it does not need to be
there.
- Trim RANDOM_DEBUG printfs. Some are excess to requirement, some
behind boot verbose.
- Use SYSINIT to sequence the startup.
- Fix init/deinit sysctl stuff.
- Make relevant sysctls also tunables.
- Add different harvesting "styles" to allow for different requirements
(direct, queue, fast).
- Add harvesting of FFS atime events. This needs to be checked for
weighing down the FS code.
- Add harvesting of slab allocator events. This needs to be checked for
weighing down the allocator code.
- Fix the random(9) manpage.
- Loadable modules are not present for now. These will be re-engineered
when the dust settles.
- Use macros for locks.
- Fix comments.
* src/share/man/...
- Update the man pages.
* src/etc/...
- The startup/shutdown work is done in D2924.
* src/UPDATING
- Add UPDATING announcement.
* src/sys/dev/random/build.sh
- Add copyright.
- Add libz for unit tests.
* src/sys/dev/random/dummy.c
- Remove; no longer needed. Functionality incorporated into randomdev.*.
* live_entropy_sources.c live_entropy_sources.h
- Remove; content moved.
- move content to randomdev.[ch] and optimise.
* src/sys/dev/random/random_adaptors.c src/sys/dev/random/random_adaptors.h
- Remove; plugability is no longer used. Compile-time algorithm
selection is the way to go.
* src/sys/dev/random/random_harvestq.c src/sys/dev/random/random_harvestq.h
- Add early (re)boot-time randomness caching.
* src/sys/dev/random/randomdev_soft.c src/sys/dev/random/randomdev_soft.h
- Remove; no longer needed.
* src/sys/dev/random/uint128.h
- Provide a fake uint128_t; if a real one ever arrived, we can use
that instead. All that is needed here is N=0, N++, N==0, and some
localised trickery is used to manufacture a 128-bit 0ULLL.
* src/sys/dev/random/unit_test.c src/sys/dev/random/unit_test.h
- Improve unit tests; previously the testing human needed clairvoyance;
now the test will do a basic check of compressibility. Clairvoyant
talent is still a good idea.
- This is still a long way off a proper unit test.
* src/sys/dev/random/fortuna.c src/sys/dev/random/fortuna.h
- Improve messy union to just uint128_t.
- Remove unneeded 'static struct fortuna_start_cache'.
- Tighten up up arithmetic.
- Provide a method to allow eternal junk to be introduced; harden
it against blatant by compress/hashing.
- Assert that locks are held correctly.
- Fix the nasty pre- and post-read overloading by providing explictit
functions to do these tasks.
- Turn into self-sufficient module (no longer requires randomdev_soft.[ch])
* src/sys/dev/random/yarrow.c src/sys/dev/random/yarrow.h
- Improve messy union to just uint128_t.
- Remove unneeded 'staic struct start_cache'.
- Tighten up up arithmetic.
- Provide a method to allow eternal junk to be introduced; harden
it against blatant by compress/hashing.
- Assert that locks are held correctly.
- Fix the nasty pre- and post-read overloading by providing explictit
functions to do these tasks.
- Turn into self-sufficient module (no longer requires randomdev_soft.[ch])
- Fix some magic numbers elsewhere used as FAST and SLOW.
Differential Revision: https://reviews.freebsd.org/D2025
Reviewed by: vsevolod,delphij,rwatson,trasz,jmg
Approved by: so (delphij)
First, on the write error, bufdone() call from ffs_backgroundwrite()
panics because pbrelvp() cleared bp->b_bufobj, while brelse() would
try to re-dirty the copy of the cg buffer. Handle this by setting
B_INVAL for the case of BIO_ERROR.
Second, we must re-dirty the real buffer containing the cylinder group
block data when background write failed. Real cg buffer was already
marked clean in ffs_bufwrite(). After the BV_BKGRDINPROG flag is
cleared on the real cg buffer in ffs_backgroundwrite(), buffer scan
may reuse the buffer at any moment. The result is lost write, and if
the write error was only transient, we get corrupted bitmaps.
We cannot re-dirty the original cg buffer in the
ffs_backgroundwritedone(), since the context is not sleepable,
preventing us from sleeping for origbp' lock. Add BV_BKGDERR flag
(protected by the buffer object lock), which is converted into delayed
write by brelse(), bqrelse() and buffer scan.
In collaboration with: Conrad Meyer <cse.cem@gmail.com>
Reviewed by: mckusick
Sponsored by: The FreeBSD Foundation (kib),
EMC/Isilon storage division (Conrad)
MFC after: 2 weeks
CPU set operations in my upcoming NUMA work.
Tested/compiled:
* i386 (run)
* amd64 (run)
* mips (run)
* mips64 (run)
* armv6 (built)
Sponsored by: Norse Corp, Inc.
