paging.
Shadow objects are marked with a COLLAPSING flag while they are collapsing with
their backing object. This gives us an explicit test rather than overloading
paging-in-progress. While split is on-going we mark an object with SPLIT.
These two operations will modify the swap tree so they must be serialized
and swap_pager_getpages() can now directly detect these conditions and page
more conservatively.
Callers to vm_object_collapse() now will reliably wait for a collapse to finish
so that the backing chain is as short as possible before other decisions are
made that may inflate the object chain. For example, split, coalesce, etc.
It is now safe to run fault concurrently with collapse. It is safe to increase
or decrease paging in progress with no lock so long as there is another valid
ref on increase.
This change makes collapse more reliable as a secondary benefit. The primary
benefit is making it safe to drop the object lock much earlier in fault or
never acquire it at all.
This was tested with a new shadow chain test script that uncovered long
standing bugs and will be integrated with stress2.
Reviewed by: kib, markj
Differential Revision: https://reviews.freebsd.org/D22908
Since r356672 ("vfs: rework vnode list management") there is nothing to do
apart from altering freevnodes count, but this much can be safely done based
on the result of atomic_fetchadd.
Reviewed by: kib
Tested by: pho
Differential Revision: https://reviews.freebsd.org/D23186
Compressing .txz packages can be rather slow, and speed is likely more
important than disk space during development. Allow package format to
be set via PKG_FORMAT make variable.
Reviewed by: bapt
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D23257
Due to the way u-boot for 32-bit powerpc is compiled, the interrupt code
assumes that the GOT pointer (r30) on u-boot is always intact.
When making syscalls to u-boot, ensure that we have restored r30 like we
found it before we enable interrupts to prevent u-boot from crashing if a
timer interrupt was pending.
This fixes ubldr on e500 qemu (assuming you have recompiled qemu's u-boot
with API support!)
Reviewed by: jhibbits
Sponsored by: Tag1 Consulting, Inc.
Differential Revision: https://reviews.freebsd.org/D23258
More background is available in r356876, but this new interface is more
portable across ZFS implementations and cleaner for what libbe is attempting
to achieve anyways.
MFC after: 3 days
This will be used in libbe in place of the internal zmount(); libbe only
wants to be able to mount a dataset at an arbitrary mountpoint without
altering dataset/pool properties. The natural way to do this in a portable
way is by creating a zfs_mount_at() interface that's effectively zfs_mount()
+ a mountpoint parameter. zfs_mount() is now a light wrapper around the new
method.
The interface and implementation have already been accepted into ZFS On
Linux, and the next commit to switch libbe() over to this new interface will
solve the last compatibility issue with ZoL. The next sysutils/openzfs
rebase against ZoL should be able to build libbe/bectl with only minor
adjustments to build glue.
Reviewed by: Ryan Moeller <ryan freqlabs com>
MFC after: 3 days
Differential Revision: https://reviews.freebsd.org/D23132
Rather than invoking `pkg config ABI` repeatedly.
Reviewed by: manu
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D23255
r355473 vastly improved the readability and cleanliness of these Makefiles.
Every single one of them follows the same pattern and duplicates the exact
same logic.
Now that we have GENERATED/SRCS, split SRCS up into the two parameters we'll
use for ${MAKESYSCALLS} rather than assuming a specific ordering of SRCS and
include a common sysent.mk to handle the rest. This makes it less tedious to
make sweeping changes.
Some default values are provided for GENERATED/SYSENT_*; almost all of these
just use a 'syscalls.master' and 'syscalls.conf' in cwd, and they all use
effectively the same filenames with an arbitrary prefix. Most ABIs will be
able to get away with just setting GENERATED_PREFIX and including
^/sys/conf/sysent.mk, while others only need light additions. kern/Makefile
is the notable exception, as it doesn't take a SYSENT_CONF and the generated
files are spread out between ^/sys/kern and ^/sys/sys, but it otherwise fits
the pattern enough to use the common version.
Reviewed by: brooks, imp
Nice!: emaste
Differential Revision: https://reviews.freebsd.org/D23197
Some systems, such as higher end Threadripper, may have
NUMA domains with no physical memory, Don't allocate
from these domains.
This fixes a "panic: vm_wait in early boot" on my 2990WX desktop
Reviewed by: jeff
Sponsored by: Netflix
Correction after r333476:
- write this as LOG_DEBUG again instead of LOG_INFO;
- get back function name into the message;
- error may be ESRCH if an address is removed in process (by carp f.e.),
not only ENOENT;
- expression complexity grows, so try making it more readable.
MFC after: 1 week
In rS354701, I replaced text relocations with offsets from &generictrap.
Unfortunately, the magic variable I was using doesn't actually mean the
address of &generictrap, in bridge mode it actually means &generictrap64.
So, for bridge mode to work, it is necessary to differentiate between
"where do we need to branch to to handle a trap" and "where is &generictrap
for purposes of doing relative math".
Introduce a new TRAP_ENTRY and use it instead of TRAP_GENTRAP for doing
actual calls to the generic trap handler.
Reported by: Mark Millard <marklmi@yahoo.com>
Reviewed by: jhibbits
Sponsored by: Tag1 Consulting, Inc.
Differential Revision: https://reviews.freebsd.org/D23057
%20%20%20 is ugly and doesn't really help make human-readable devfs names.
PR: 243318
Reported by: Peter Eriksson <pen AT lysator.liu.se>
Relnotes: yes
Summary:
This makes the interface described in the definition file act like a
pseudo-IFUNC service, by caching the found method locally.
Applying this to the PowerPC MMU definitions, it yields a significant
(15-20%) performance improvement, seen in both a 'make buildworld' and a
parallel build of LLVM, on a POWER9 system.
Reviewed By: imp
Differential Revision: https://reviews.freebsd.org/D23245
It gets rolled up to the global when deferred requeueing is performed.
A dedicated read routine makes sure to return a value only off by a certain
amount.
This soothes a global serialisation point for all 0<->1 hold count transitions.
Reviewed by: jeff
Differential Revision: https://reviews.freebsd.org/D23235
Summary:
Consolidate the NUMA associativity handling into a platform function.
Non-NUMA platforms will just fall back to the default (0). Currently
only implemented for powernv, which uses a lookup table to map the
device tree associativity into a system NUMA domain.
Fixes hangs on powernv after r356534, and corrects a fairly longstanding
bug in powernv's NUMA handling, which ended up using domains 1 and 2 for
devices and memory on power9, while CPUs were bound to domains 0 and 1.
Reviewed by: bdragon, luporl
Differential Revision: https://reviews.freebsd.org/D23220
It turns out the maximum TLB1 page size on e5500 is 4G, despite the format
being defined for up to 1TB.
So, we need to clamp the DMAP TLB1 entries to not attempt to create 16G or
larger entries.
Fixes boot on my X5000 in which I just installed 16G of RAM.
Reviewed by: jhibbits
Sponsored by: Tag1 Consulting, Inc.
Differential Revision: https://reviews.freebsd.org/D23244
In r291668, an instruction was added to sigcode64.S without the nop pad at
the end being taken out.
Due to alignment, this means that a dword is being wasted on the shared
page for no reason.
Take out this nop, and add some comments while I'm here.
Reviewed by: jhibbits
Sponsored by: Tag1 Consulting, Inc.
Differential Revision: https://reviews.freebsd.org/D23055
This resulted in the partitioning step failing if either of the
"Auto (UFS)" or "Manual" options were selected.
Reason: partedit was attempting to open a directory (TMPDIR) read/write,
which resulted in errno(2) 21 - EISDIR - Is a directory.
Reported by: Clay Daniels <clay.daniels.jr@gmail.com>
Reviewed by: Ryan Moeller <ryan@freqlabs.com>
Approved by: emaste, bcran
Differential Revision: https://reviews.freebsd.org/D23232
This makes amrr_node_stats() cleaner and allows the rate printing to be
reusable.
Submitted by: Neel Chauhan <neel at neelc.org>
Reviewed by: adrian
Differential Revision: https://reviews.freebsd.org/D22318
Raspberry Pi are all over the board, and the reality is that there's no harm
in including all of the definitions by default but plenty of harm in the
current situation. This change is safe because we match a definition by root
/compatible in the FDT, so there will be no false-positives because of it.
The main array of definitions grows, but it's only walked exactly once to
determine which we need to use.
Previously, this check was only in sys_sigreturn() which meant that
user applications could write invalid values to the register via
setcontext() or swapcontext().
Reviewed by: mhorne
Sponsored by: DARPA
Differential Revision: https://reviews.freebsd.org/D23219
arm64 and riscv were only saving and restoring floating point
registers for sendsig() and sys_sigreturn(), but not for getcontext(),
setcontext(), and swapcontext().
While here, remove an always-false check for uap being NULL from
sys_sigreturn().
Reviewed by: br, mhorne
Sponsored by: DARPA
Differential Revision: https://reviews.freebsd.org/D23218
dma(8) depends on OpenSSL unconditionally.
Reported by: Michael Dexter's Build Options Survey run
MFC after: 1 weeks
Sponsored by: The FreeBSD Foundation
As part of the RISC-V ABI, the gp register is expected to initialized
with the address of __global_pointer$ as early as possible. This allows
loads and stores from .sdata to be relaxed based on the value of gp. In
locore.S we do this initialization twice, once each for va and mpva.
However, in both cases the initialization is preceded by an la
instruction, which in theory could be relaxed by the linker.
Move the initialization of gp to be slightly earlier (before la
cpu_exception_handler), and add an additional gp initialization at the
very beginning of _start, before virtual memory is set up.
Reported by: jrtc27
Reviewed by: jrtc27
Differential Revision: https://reviews.freebsd.org/D23139
as these functions should do zero-extend.
Discovered by running pci_read_cap(), and by hint from James Clarke.
Reviewed by: James Clarke <jrtc27@jrtc27.com>
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D23236
Prior to introduction of this op libc's readdir would call fstatfs(2), in
effect unnecessarily copying kilobytes of data just to check fs name and a
mount flag.
Reviewed by: kib (previous version)
Differential Revision: https://reviews.freebsd.org/D23162
Filemon will add the ability to ignore the cookie if the installed file is
missing. Without filemon that's not possible though so if the cookie is present
an the command unchanged then the install wouldn't run.
Sponsored by: DellEMC
MFC after: 2 weeks
page that was previously mapped read-only it exists in pmap until pmap_enter()
returns. However, we held no reference to the original page after the copy
was complete. This allowed vm_object_scan_all_shadowed() to collapse an
object that still had pages mapped. To resolve this, add another page pointer
to the faultstate so we can keep the page xbusy until we're done with
pmap_enter(). Handle busy pages in scan_all_shadowed. This is already done
in vm_object_collapse_scan().
Reviewed by: kib, markj
Differential Revision: https://reviews.freebsd.org/D23155
'shutdown -c' is supposed to power cycle the system rather than doing a normal
reboot. However, when that fails, it halts the system. This is not quite right
since the intent isn't to halt the system but to restart. Make the default init
behavior be to restart the system. The halt(8) interface can be used if you'd
like to powercycle or halt.
MFC After: 1 week
Differential Revision: https://reviews.freebsd.org/D23129
Only send a speedup when there's a shortage. While this is a little racy, lost
races aren't a big deal for this function. If there's a shorage just popping up
after we check these values, then we'll catch it next time. If there's a
shortage that's just clearing up, we may do some work at the lower layers a
little sooner than we otherwise would have. Sicne shortages are relatively rare
events, both races are acceptable.
Reviewed by: chs
Differential Revision: https://reviews.freebsd.org/D23182
It turns out there's a problem with using g_io to send the speedup. It leads to
a race when there's a resource shortage when a disk fails.
Instead, send BIO_SPEEDUP via struct buf. This is pretty straight forward,
except we need to transfer the bio_flags from b_ioflags for BIO_SPEEDUP commands
in g_vfs_strategy.
Reviewed by: kirk, chs
Differential Revision: https://reviews.freebsd.org/D23117
While some geom layers pass unknown commands down, not all do. For the ones that
don't, pass BIO_SPEEDUP down to the providers that constittue the geom, as
applicable. No changes to vinum or virstor because I was unsure how to add this
support, and I'm also unsure how to test these. gvinum doesn't implement
BIO_FLUSH either, so it may just be poorly maintained. gvirstor is for testing
and not supportig BIO_SPEEDUP is fine.
Reviewed by: chs
Differential Revision: https://reviews.freebsd.org/D23183
When _a is empty we end up with an invalid invocation of pfctl, and no output.
We must add quotes to make it clear to pfctl that we're passing an empty anchor
name.
PR: 224415
Submitted by: sigsys AT gmail.com
MFC after: 2 weeks
The vnode list lock is only needed to reclaim free vnodes or kick the vnlru
thread (or to block and not miss a wake up (but note the sleep has a timeout so
this would not be a correctness issue)). Try to get away without the lock by
just doing an atomic increment.
The lock is contended e.g., during poudriere -j 104 where about half of all
acquires come from vnode allocation code.
Note the entire scheme needs a rewrite, the above just reduces it's SMP impact.
Reviewed by: kib
Differential Revision: https://reviews.freebsd.org/D23140
Semantics are almost identical. Some code is deduplicated and there are
fewer memory accesses.
Reviewed by: kib, jeff
Differential Revision: https://reviews.freebsd.org/D23158
After r355693, random(6) -f sometimes fail to output all the lines of the
input file. This is because the range from which random indices are chosen
is too big, so occasionally the random selection doesn't correspond to any
line and nothing gets printed.
(Ed. note: Mea culpa. Working on r355693, I was confused by the sometime
use of 1-indexing, sometimes 0-indexing in randomize_fd().)
Submitted by: Ryan Moeller <ryan AT freqlabs.com>
X-MFC-With: r355693
Sponsored by: iXsystems, Inc.
Differential Revision: https://reviews.freebsd.org/D23199
