pv list lock is the main bottleneck during poudriere -j 104 and
pmap_remove_pages is the most impactful consumer. It frees chunks with the lock
held even though it plays no role in correctness. Moreover chunks are often
freed in groups, sample counts during buildkernel (0-sized frees removed):
value ------------- Distribution ------------- count
0 | 0
1 | 8
2 |@@@@@@@ 19329
4 |@@@@@@@@@@@@@@@@@@@@@@ 58517
8 | 1085
16 | 71
32 |@@@@@@@@@@ 24919
64 | 899
128 | 7
256 | 2
512 | 0
Thus:
1. batch freeing
2. move it past unlocking pv list
Reviewed by: alc (previous version), markj (previous version), kib
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D21832
Centralize calculation of signal and ucode delivered on unhandled page
fault in new function vm_fault_trap(). MD trap_pfault() now almost
always uses the signal numbers and error codes calculated in
consistent MI way.
This introduces the protection fault compatibility sysctls to all
non-x86 architectures which did not have that bug, but apparently they
were already much more wrong in selecting delivered signals on
protection violations.
Change the delivered signal for accesses to mapped area after the
backing object was truncated. According to POSIX description for
mmap(2):
The system shall always zero-fill any partial page at the end of an
object. Further, the system shall never write out any modified
portions of the last page of an object which are beyond its
end. References within the address range starting at pa and
continuing for len bytes to whole pages following the end of an
object shall result in delivery of a SIGBUS signal.
An implementation may generate SIGBUS signals when a reference
would cause an error in the mapped object, such as out-of-space
condition.
Adjust according to the description, keeping the existing
compatibility code for SIGSEGV/SIGBUS on protection failures.
For situations where kernel cannot handle page fault due to resource
limit enforcement, SIGBUS with a new error code BUS_OBJERR is
delivered. Also, provide a new error code SEGV_PKUERR for SIGSEGV on
amd64 due to protection key access violation.
vm_fault_hold() is renamed to vm_fault(). Fixed some nits in
trap_pfault()s like mis-interpreting Mach errors as errnos. Removed
unneeded truncations of the fault addresses reported by hardware.
PR: 211924
Reviewed by: alc
Discussed with: jilles, markj
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D21566
Convert all remaining references to that field to "ref_count" and update
comments accordingly. No functional change intended.
Reviewed by: alc, kib
Sponsored by: Intel, Netflix
Differential Revision: https://reviews.freebsd.org/D21768
The direct map is never used for execution of code, so we might as well
set NX in the direct map's PML4Es. Also clarify the intent of the code
in create_pagetables() that restricts access protections on the region
of the direct map mapping the kernel text.
Reviewed by: alc, kib (previous version)
MFC after: 1 week
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D21759
This is required for DPCPU and VNET data variable definitions to work when
KLDs are linked as DSOs. R_X86_64_RELATIVE relocations should not appear
in object files, so assert this in elf_relocaddr().
Reviewed by: kib
MFC after: 1 month
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D21755
There does not appear to be any existing need for such mappings to be
executable.
Reviewed by: alc, kib
MFC after: 1 month
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D21754
Decode PAT_UNCACHED.
When unknown pat mode is encountered, print the pte bits combination
instead of the index, which is always 8.
Reviewed by: markj
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
Differential revision: https://reviews.freebsd.org/D21738
- Remove a dead variable from the amd64 pmap_extract_and_hold().
- Fix grammar in the vm_page_wire man page.
Reported by: alc
Reviewed by: alc, kib
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D21639
- Remove a dead variable from the amd64 pmap_extract_and_hold().
- Fix grammar in the vm_page_wire man page.
Reported by: alc
Reviewed by: alc, kib
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D21639
Just return EINVAL if flags != 0. The Linux man page documents one
case of EINVAL as "The filesystem does not support one of the flags in
flags."
After r351723 userland binaries will try using new system calls.
Reported by: mjg
Reviewed by: mjg, trasz
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D21590
There are several mechanisms by which a vm_page reference is held,
preventing the page from being freed back to the page allocator. In
particular, holding the page's object lock is sufficient to prevent the
page from being freed; holding the busy lock or a wiring is sufficent as
well. These references are protected by the page lock, which must
therefore be acquired for many per-page operations. This results in
false sharing since the page locks are external to the vm_page
structures themselves and each lock protects multiple structures.
Transition to using an atomically updated per-page reference counter.
The object's reference is counted using a flag bit in the counter. A
second flag bit is used to atomically block new references via
pmap_extract_and_hold() while removing managed mappings of a page.
Thus, the reference count of a page is guaranteed not to increase if the
page is unbusied, unmapped, and the object's write lock is held. As
a consequence of this, the page lock no longer protects a page's
identity; operations which move pages between objects are now
synchronized solely by the objects' locks.
The vm_page_wire() and vm_page_unwire() KPIs are changed. The former
requires that either the object lock or the busy lock is held. The
latter no longer has a return value and may free the page if it releases
the last reference to that page. vm_page_unwire_noq() behaves the same
as before; the caller is responsible for checking its return value and
freeing or enqueuing the page as appropriate. vm_page_wire_mapped() is
introduced for use in pmap_extract_and_hold(). It fails if the page is
concurrently being unmapped, typically triggering a fallback to the
fault handler. vm_page_wire() no longer requires the page lock and
vm_page_unwire() now internally acquires the page lock when releasing
the last wiring of a page (since the page lock still protects a page's
queue state). In particular, synchronization details are no longer
leaked into the caller.
The change excises the page lock from several frequently executed code
paths. In particular, vm_object_terminate() no longer bounces between
page locks as it releases an object's pages, and direct I/O and
sendfile(SF_NOCACHE) completions no longer require the page lock. In
these latter cases we now get linear scalability in the common scenario
where different threads are operating on different files.
__FreeBSD_version is bumped. The DRM ports have been updated to
accomodate the KPI changes.
Reviewed by: jeff (earlier version)
Tested by: gallatin (earlier version), pho
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D20486
- Use ptoa() instead of the archaic ctob().
- Use pagezero() to zero a PDP page.
- Remove PA_MIN_ADDRESS, orphaned by r351742.
- Remove unneeded parens and an unnecessary control flow statement.
Reported by: alc
Reviewed by: alc, kib
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D21495
from recent Ubuntu versions. Without it they segfault on startup.
Reviewed by: emaste
MFC after: 2 weeks
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D20687
r351198 allows the kernel to use domain-local memory to back the vm_page
array (up to 2MB boundaries) and reserves a separate PML4 entry for that
purpose. One consequence of that change is that the vm_page array is no
longer present in minidumps, which only adds pages mapped above
VM_MIN_KERNEL_ADDRESS.
To avoid the friction caused by having kernel data structures mapped
below VM_MIN_KERNEL_ADDRESS, map the vm_page array starting at
VM_MIN_KERNEL_ADDRESS instead of using a dedicated PML4 entry.
Reviewed by: kib
Discussed with: jeff
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D21491
The sysctl is called vm.pmap.kernel_maps. It dumps address ranges
and their corresponding protection and mapping mode, as well as
counts of 2MB and 1GB pages in the range.
Reviewed by: kib
MFC after: 2 weeks
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D21380
The bhyve virtual local APIC uses an instance-global flag to indicate
when an error LVT is being delivered to prevent infinite recursion.
Use a function argument instead to reduce the amount of instance-global
state.
This was inspired by reviewing the bhyve save/restore work, which
saves a copy of the instance-global state for each vlapic.
Smart OS bug: https://smartos.org/bugview/OS-7777
Submitted by: Patrick Mooney
Reviewed by: markj, rgrimes
Obtained from: SmartOS / Joyent
Differential Revision: https://reviews.freebsd.org/D20365
Many extern struct pcpu <something>__pcpu declarations were
copied/pasted in sources. The issue is that the definition is MD, but
it cannot be provided by machine/pcpu.h due to actual struct pcpu
defined in sys/pcpu.h later than the inclusion of machine/pcpu.h.
This forced the copying when other code needed direct access to
__pcpu. There is no way around it, due to machine/pcpu.h supplying
part of struct pcpu fields.
To work around the problem, add a new machine/pcpu_aux.h header, which
should fill any needed MD definitions after struct pcpu definition is
completed. This allows to remove copies of __pcpu spread around the
source. Also on x86 it makes it possible to remove work arounds like
OFFSETOF_CURTHREAD or clang specific warnings supressions.
Reported and tested by: lwhsu, bcran
Reviewed by: imp, markj (previous version)
Discussed with: jhb
Sponsored by: The FreeBSD Foundation
Differential revision: https://reviews.freebsd.org/D21418
- LK macro (conditional on SMP for the lock prefix) is unused
- SETLK unnecessarily performs xchg. obtained value is never used and the
implicit lock prefix adds avoidable cost. Barrier provided by it does
not appear to be of any use.
- the lock waited for is almost never blocked, yet the loop starts with
a pause. Move it out of the common case.
Reviewed by: kib
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D19563
Use DOMAINSET_PREF() instead of DOMAINSET_FIXED(), to gracefully
fallback in case of memory-less domain.
Reported and tested by: bcran
Reviewed by: markj
Sponsored by: The FreeBSD Foundation
existing one.
Allocation failure is possible for instance when cpu domain has no memory.
Reported and tested by: bcran
Reviewed by: markj
Sponsored by: The FreeBSD Foundation
Move pcpu KVA out of .bss into dynamically allocated VA at
pmap_bootstrap(). This avoids demoting superpage mapping .data/.bss.
Also it makes possible to use pmap_qenter() for installation of
domain-local pcpu page on NUMA configs.
Refactor pcpu and IST initialization by moving it to helper functions.
Reviewed by: markj
Tested by: pho
Discussed with: jeff
Sponsored by: The FreeBSD Foundation
Differential revision: https://reviews.freebsd.org/D21320
All these stacks are used only once (doublefault, boot) or very rare
(mce).
Reviewed by: markj
Tested by: pho
Sponsored by: The FreeBSD Foundation
Differential revision: https://reviews.freebsd.org/D21320
NUMA domain that the pages describe. Patch original from gallatin.
Reviewed by: kib
Tested by: pho
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D21252
doing so adds more flexibility with less redundant code.
Reviewed by: jhb, markj, kib
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D21250
Such speculations could use user-controlled %gs base, esp. since
FreeBSD supports WRGSBASE instructions.
Place LFENCEs on entry for each basic block after the test for
previous kernel/user mode on the kernel entry, which prevents the
speculation. Code accesses %gs-based PCPU before any serialization
instructions are executed, like %cr3 reload for KPTI.
With pti disabled, on haswell i7-4770S machine, "syscall_timings getppid"
shows when no lfence is added to syscall path:
test loop time iterations periteration
getppid 0 1.040918865 4643611 0.000000224
getppid 1 1.004985962 4481816 0.000000224
getppid 2 1.005196483 4482363 0.000000224
with lfence:
getppid 0 1.043701091 4554779 0.000000229
getppid 1 1.016930328 4438094 0.000000229
getppid 2 1.023223117 4466640 0.000000229
and ministat reports 'No difference proven at 95.0% confidence.'
Security: CVE-2019-1125
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
PTI-mode entry points were coded to set up the environment identical
to non-PTI entry and then fall-through to non-PTI handlers, mostly.
This has the drawback of requiring two more SWAPGS, first to access
PCPU, and then to return to the state expected by the non-PTI entry
point.
Eliminate the duplication by doing more in entry stubs both for PTI
and non-PTI, and adjusting the common code to expect that SWAPGS and
some minimal registers saving is done by entries.
Some less often used entries, in particular, #GP, #NP, and #SS, which
can fault on doreti, are left as is because there are basically four
variants of entrance, and they are not performance-critical,
esp. comparing with e.g. #PF or interrupts.
Reviewed by: markj (previous version)
Tested by: pho (previous version)
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
physical destination mode.
This is mostly a nop, because the vmm initializes all vCPUs up to
vm_maxcpus, so even if the target CPU is not active, lapic/vlapic code
still has the valid data to use. As John notes, dropping such
interrupts more closely matches the real harware, which ignores all
interrupts for not started APs.
Reviewed by: jhb
admbugs: 837
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
Previously, AcpiOsMemory was using pmap_mapbios which would always map
the requested address Write-Back (WB). For several AMD Ryzen laptops,
the BIOS uses AcpiOsMemory to directly access the PCI MCFG region in
order to access PCI config registers. This has the side effect of
remapping the MCFG region in the direct map as WB instead of UC
hanging the laptops during boot.
On the one laptop I examined in detail, the _PIC global method used to
switch from 8259A PICs to I/O APICs uses a pair of PCI config space
registers at offset 0x84 in the device at 0:0:0 to as a pair of
address/data registers to access an indirect register in the chipset
and clear a single bit to switch modes.
To fix, alter the semantics of pmap_mapbios() such that it does not
modify the attributes of any existing mappings and instead uses the
existing attributes. If a new mapping is created, this new mapping
uses WB (the default memory attribute).
Special thanks to the gentleman whose name I don't have who brought
two affected laptops to the hacker lounge at BSDCan. Direct access to
the affected systems permitted finding the root cause within an hour
or so.
PR: 231760, 236899
Reviewed by: kib, alc
MFC after: 2 weeks
Differential Revision: https://reviews.freebsd.org/D20327
Bhyve's vmm is a self-contained modern component and thus a good
candidate for use of C99 types.
Reviewed by: jhb, kib, markj, Patrick Mooney
MFC after: 1 week
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D21036
