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
Some external tools just do a 'ls /dev/vmm' to figure out the bhyve virtual
machines on the host. These tools break if the devmem device nodes also
appear in /dev/vmm.
Requested by: grehan
the 'user' sysctl tree, which have all been coming back 0 or empty
since r240176.
Differential Revision: https://reviews.freebsd.org/D2945
Reviewed by: sbruno
Approved by: jmallett (mentor)
MFC after: 3 days
This function is equivalent to fclose(3) function except that it
does not close the underlying file descriptor.
fdclose(3) is step forward to make FILE structure private.
Reviewed by: wblock, jilles, jhb, pjd
Approved by: pjd (mentor)
Differential Revision: https://reviews.freebsd.org/D2697
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)
Update setkey and libipsec to understand aes-gcm-16 as an
encryption method.
A partial commit of the work in review D2936.
Submitted by: eri
Reviewed by: jmg
MFC after: 2 weeks
Sponsored by: Rubicon Communications (Netgate)
temporary file is created and then a rename() call move it to official file.
This operation didn't have any check to make sure data was written to disk
and if a power cycle happens system could end up with a 0 length passwd
or group database.
There is a pfSense bug with more infor about it:
https://redmine.pfsense.org/issues/4523
The following changes were made to protect passwd and group operations:
* lib/libutil/gr_util.c:
- Replace mkstemp() by mkostemp() with O_SYNC flag to create temp file
- After rename(), fsync() call on directory for faster result
* lib/libutil/pw_util.c
- Replace mkstemp() by mkostemp() with O_SYNC flag to create temp file
* usr.sbin/pwd_mkdb/pwd_mkdb.c
- Added O_SYNC flag on dbopen() calls
- After rename(), fsync() call on directory for faster result
* lib/libutil/pw_util.3
- pw_lock() returns a file descriptor to master password file on success
Differential Revision: https://reviews.freebsd.org/D2978
Approved by: bapt
Sponsored by: Netgate
point support. The fenv-vfp.c file overrides -mfloat-abi so it can use
floating point instructions if it detects support at runtime. Make it also
override -mfpu in case the user has set -mfpu=none.
to be before the lavel, otherwise an extra word may be added between the
label and the data.
Obtained from: ABT Systems Ltd
Sponsored by: The FReeBSD Foundation
CLANG WARNINGS.
BUMP SHLIB_MAJOR version as the ABI potentially changed.
Also run an 'exp run' to double check if any external project are using
those functions. Thanks antoine@.
PR: 200807
Differential Revision: D2775
Reviewed by: kib, ngie
This symbol can sometimes be emitted by clang++, and was not yet
exported from libcxxrt. Attempt to be compatible with libsupc++ by
using the same CXXABI_1.3.6 symbol version.
Reported by: yuri@rawbw.com
PR: 200863
Reviewed by: emaste
MFC after: 3 days
Differential Revision: https://reviews.freebsd.org/D2850
When enough time has passed for users to update their userland the kernel
fix will be applied. This will change the ABI to have x0 point to the args
and sp be correctly aligned.
It is expected this compatibility code can be removed when the kernel and
qemu usermode emulation have both been updated for the new ABI.
This fixes clang failures, and most likely other crashes.
Obtained from: ABT Systems Ltd
Sponsored by: The FreeBSD Foundation
due to a change in behavior of the 'vm_map_gpa()'.
Prior to r284539 if 'vm_map_gpa()' was called to map an address range in the
guest MMIO region then it would return NULL. This was used by the "movs"
emulation to detect if the 'src' or 'dst' operand was in MMIO space.
Post r284539 'vm_map_gpa()' started returning a non-NULL pointer even when
mapping the guest MMIO region.
Fix this by returning non-NULL only if [gaddr, gaddr+len) is entirely
within the 'lowmem' or 'highmem' regions and NULL otherwise.
Pointy hat to: neel
Reviewed by: grehan
Reported by: tychon, Ben Perrault (ben.perrault@gmail.com)
MFC after: 1 week
devmem is used to represent MMIO devices like the boot ROM or a VESA framebuffer
where doing a trap-and-emulate for every access is impractical. devmem is a
hybrid of system memory (sysmem) and emulated device models.
devmem is mapped in the guest address space via nested page tables similar
to sysmem. However the address range where devmem is mapped may be changed
by the guest at runtime (e.g. by reprogramming a PCI BAR). Also devmem is
usually mapped RO or RW as compared to RWX mappings for sysmem.
Each devmem segment is named (e.g. "bootrom") and this name is used to
create a device node for the devmem segment (e.g. /dev/vmm/testvm.bootrom).
The device node supports mmap(2) and this decouples the host mapping of
devmem from its mapping in the guest address space (which can change).
Reviewed by: tychon
Discussed with: grehan
Differential Revision: https://reviews.freebsd.org/D2762
MFC after: 4 weeks
- Fix a case where NaNs were not mixed correctly and signalling NaNs were
not converted to quiet NaNs.
- Eliminate two negations from ctan(z).
In collaboration with: bde
