The problem:
On -rt, an emulated LAPIC timer instances has the following path:
1) hard interrupt
2) ksoftirqd is scheduled
3) ksoftirqd wakes up vcpu thread
4) vcpu thread is scheduled
This extra context switch introduces unnecessary latency in the
LAPIC path for a KVM guest.
The solution:
Allow waking up vcpu thread from hardirq context,
thus avoiding the need for ksoftirqd to be scheduled.
Normal waitqueues make use of spinlocks, which on -RT
are sleepable locks. Therefore, waking up a waitqueue
waiter involves locking a sleeping lock, which
is not allowed from hard interrupt context.
cyclictest command line:
This patch reduces the average latency in my tests from 14us to 11us.
Daniel writes:
Paolo asked for numbers from kvm-unit-tests/tscdeadline_latency
benchmark on mainline. The test was run 1000 times on
tip/sched/core 4.4.0-rc8-01134-g0905f04:
./x86-run x86/tscdeadline_latency.flat -cpu host
with idle=poll.
The test seems not to deliver really stable numbers though most of
them are smaller. Paolo write:
"Anything above ~10000 cycles means that the host went to C1 or
lower---the number means more or less nothing in that case.
The mean shows an improvement indeed."
Before:
min max mean std
count 1000.000000 1000.000000 1000.000000 1000.000000
mean 5162.596000 2019270.084000 5824.491541 20681.645558
std 75.431231 622607.723969 89.575700 6492.272062
min 4466.000000 23928.000000 5537.926500 585.864966
25% 5163.000000 1613252.750000 5790.132275 16683.745433
50% 5175.000000 2281919.000000 5834.654000 23151.990026
75% 5190.000000 2382865.750000 5861.412950 24148.206168
max 5228.000000 4175158.000000 6254.827300 46481.048691
After
min max mean std
count 1000.000000 1000.00000 1000.000000 1000.000000
mean 5143.511000 2076886.10300 5813.312474 21207.357565
std 77.668322 610413.09583 86.541500 6331.915127
min 4427.000000 25103.00000 5529.756600 559.187707
25% 5148.000000 1691272.75000 5784.889825 17473.518244
50% 5160.000000 2308328.50000 5832.025000 23464.837068
75% 5172.000000 2393037.75000 5853.177675 24223.969976
max 5222.000000 3922458.00000 6186.720500 42520.379830
[Patch was originaly based on the swait implementation found in the -rt
tree. Daniel ported it to mainline's version and gathered the
benchmark numbers for tscdeadline_latency test.]
Signed-off-by: Daniel Wagner <daniel.wagner@bmw-carit.de>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: linux-rt-users@vger.kernel.org
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Link: http://lkml.kernel.org/r/1455871601-27484-4-git-send-email-wagi@monom.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
When we allocate bitmaps in vgic_vcpu_init_maps, we divide the number of
bits we need by 8 to figure out how many bytes to allocate. However,
bitmap elements are always accessed as unsigned longs, and if we didn't
happen to allocate a size such that size % sizeof(unsigned long) == 0,
bitmap accesses may go past the end of the allocation.
When using KASAN (which does byte-granular access checks), this results
in a continuous stream of BUGs whenever these bitmaps are accessed:
=============================================================================
BUG kmalloc-128 (Tainted: G B ): kasan: bad access detected
-----------------------------------------------------------------------------
INFO: Allocated in vgic_init.part.25+0x55c/0x990 age=7493 cpu=3 pid=1730
INFO: Slab 0xffffffbde6d5da40 objects=16 used=15 fp=0xffffffc935769700 flags=0x4000000000000080
INFO: Object 0xffffffc935769500 @offset=1280 fp=0x (null)
Bytes b4 ffffffc9357694f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Object ffffffc935769500: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Object ffffffc935769510: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Object ffffffc935769520: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Object ffffffc935769530: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Object ffffffc935769540: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Object ffffffc935769550: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Object ffffffc935769560: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Object ffffffc935769570: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Padding ffffffc9357695b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Padding ffffffc9357695c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Padding ffffffc9357695d0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Padding ffffffc9357695e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
Padding ffffffc9357695f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
CPU: 3 PID: 1740 Comm: kvm-vcpu-0 Tainted: G B 4.4.0+ #17
Hardware name: ARM Juno development board (r1) (DT)
Call trace:
[<ffffffc00008e770>] dump_backtrace+0x0/0x280
[<ffffffc00008ea04>] show_stack+0x14/0x20
[<ffffffc000726360>] dump_stack+0x100/0x188
[<ffffffc00030d324>] print_trailer+0xfc/0x168
[<ffffffc000312294>] object_err+0x3c/0x50
[<ffffffc0003140fc>] kasan_report_error+0x244/0x558
[<ffffffc000314548>] __asan_report_load8_noabort+0x48/0x50
[<ffffffc000745688>] __bitmap_or+0xc0/0xc8
[<ffffffc0000d9e44>] kvm_vgic_flush_hwstate+0x1bc/0x650
[<ffffffc0000c514c>] kvm_arch_vcpu_ioctl_run+0x2ec/0xa60
[<ffffffc0000b9a6c>] kvm_vcpu_ioctl+0x474/0xa68
[<ffffffc00036b7b0>] do_vfs_ioctl+0x5b8/0xcb0
[<ffffffc00036bf34>] SyS_ioctl+0x8c/0xa0
[<ffffffc000086cb0>] el0_svc_naked+0x24/0x28
Memory state around the buggy address:
ffffffc935769400: 00 00 fc fc fc fc fc fc fc fc fc fc fc fc fc fc
ffffffc935769480: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
>ffffffc935769500: 04 fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
^
ffffffc935769580: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
ffffffc935769600: 00 00 00 00 fc fc fc fc fc fc fc fc fc fc fc fc
==================================================================
Fix the issue by always allocating a multiple of sizeof(unsigned long),
as we do elsewhere in the vgic code.
Fixes: c1bfb577a ("arm/arm64: KVM: vgic: switch to dynamic allocation")
Cc: stable@vger.kernel.org
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Commit 4b4b4512da ("arm/arm64: KVM: Rework the arch timer to use
level-triggered semantics") brought the virtual architected timer
closer to the VGIC. There is one occasion were we don't properly
check for the VGIC actually having been initialized before, but
instead go on to check the active state of some IRQ number.
If userland hasn't instantiated a virtual GIC, we end up with a
kernel NULL pointer dereference:
=========
Unable to handle kernel NULL pointer dereference at virtual address 00000000
pgd = ffffffc9745c5000
[00000000] *pgd=00000009f631e003, *pud=00000009f631e003, *pmd=0000000000000000
Internal error: Oops: 96000006 [#2] PREEMPT SMP
Modules linked in:
CPU: 0 PID: 2144 Comm: kvm_simplest-ar Tainted: G D 4.5.0-rc2+ #1300
Hardware name: ARM Juno development board (r1) (DT)
task: ffffffc976da8000 ti: ffffffc976e28000 task.ti: ffffffc976e28000
PC is at vgic_bitmap_get_irq_val+0x78/0x90
LR is at kvm_vgic_map_is_active+0xac/0xc8
pc : [<ffffffc0000b7e28>] lr : [<ffffffc0000b972c>] pstate: 20000145
....
=========
Fix this by bailing out early of kvm_timer_flush_hwstate() if we don't
have a VGIC at all.
Reported-by: Cosmin Gorgovan <cosmin@linux-geek.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Andre Przywara <andre.przywara@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Cc: <stable@vger.kernel.org> # 4.4.x
To date, we have implemented two I/O usage models for persistent memory,
PMEM (a persistent "ram disk") and DAX (mmap persistent memory into
userspace). This series adds a third, DAX-GUP, that allows DAX mappings
to be the target of direct-i/o. It allows userspace to coordinate
DMA/RDMA from/to persistent memory.
The implementation leverages the ZONE_DEVICE mm-zone that went into
4.3-rc1 (also discussed at kernel summit) to flag pages that are owned
and dynamically mapped by a device driver. The pmem driver, after
mapping a persistent memory range into the system memmap via
devm_memremap_pages(), arranges for DAX to distinguish pfn-only versus
page-backed pmem-pfns via flags in the new pfn_t type.
The DAX code, upon seeing a PFN_DEV+PFN_MAP flagged pfn, flags the
resulting pte(s) inserted into the process page tables with a new
_PAGE_DEVMAP flag. Later, when get_user_pages() is walking ptes it keys
off _PAGE_DEVMAP to pin the device hosting the page range active.
Finally, get_page() and put_page() are modified to take references
against the device driver established page mapping.
Finally, this need for "struct page" for persistent memory requires
memory capacity to store the memmap array. Given the memmap array for a
large pool of persistent may exhaust available DRAM introduce a
mechanism to allocate the memmap from persistent memory. The new
"struct vmem_altmap *" parameter to devm_memremap_pages() enables
arch_add_memory() to use reserved pmem capacity rather than the page
allocator.
This patch (of 18):
The core has developed a need for a "pfn_t" type [1]. Move the existing
pfn_t in KVM to kvm_pfn_t [2].
[1]: https://lists.01.org/pipermail/linux-nvdimm/2015-September/002199.html
[2]: https://lists.01.org/pipermail/linux-nvdimm/2015-September/002218.html
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Acked-by: Christoffer Dall <christoffer.dall@linaro.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull KVM updates from Paolo Bonzini:
"PPC changes will come next week.
- s390: Support for runtime instrumentation within guests, support of
248 VCPUs.
- ARM: rewrite of the arm64 world switch in C, support for 16-bit VM
identifiers. Performance counter virtualization missed the boat.
- x86: Support for more Hyper-V features (synthetic interrupt
controller), MMU cleanups"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (115 commits)
kvm: x86: Fix vmwrite to SECONDARY_VM_EXEC_CONTROL
kvm/x86: Hyper-V SynIC timers tracepoints
kvm/x86: Hyper-V SynIC tracepoints
kvm/x86: Update SynIC timers on guest entry only
kvm/x86: Skip SynIC vector check for QEMU side
kvm/x86: Hyper-V fix SynIC timer disabling condition
kvm/x86: Reorg stimer_expiration() to better control timer restart
kvm/x86: Hyper-V unify stimer_start() and stimer_restart()
kvm/x86: Drop stimer_stop() function
kvm/x86: Hyper-V timers fix incorrect logical operation
KVM: move architecture-dependent requests to arch/
KVM: renumber vcpu->request bits
KVM: document which architecture uses each request bit
KVM: Remove unused KVM_REQ_KICK to save a bit in vcpu->requests
kvm: x86: Check kvm_write_guest return value in kvm_write_wall_clock
KVM: s390: implement the RI support of guest
kvm/s390: drop unpaired smp_mb
kvm: x86: fix comment about {mmu,nested_mmu}.gva_to_gpa
KVM: x86: MMU: Use clear_page() instead of init_shadow_page_table()
arm/arm64: KVM: Detect vGIC presence at runtime
...
Since the numbers now overlap, it makes sense to enumerate
them in asm/kvm_host.h rather than linux/kvm_host.h. Functions
that refer to architecture-specific requests are also moved
to arch/.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
KVM/ARM changes for Linux v4.5
- Complete rewrite of the arm64 world switch in C, hopefully
paving the way for more sharing with the 32bit code, better
maintainability and easier integration of new features.
Also smaller and slightly faster in some cases...
- Support for 16bit VM identifiers
- Various cleanups
Having the system register numbers as #defines has been a pain
since day one, as the ordering is pretty fragile, and moving
things around leads to renumbering and epic conflict resolutions.
Now that we're mostly acessing the sysreg file in C, an enum is
a much better type to use, and we can clean things up a bit.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Acked-by: Christoffer Dall <christoffer.dall@linaro.org>
We store GICv3 LRs in reverse order so that the CPU can save/restore
them in rever order as well (don't ask why, the design is crazy),
and yet generate memory traffic that doesn't completely suck.
We need this macro to be available to the C version of save/restore.
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
vgic_io_ops is only referenced within vgic.c, so it can be declared
static.
Signed-off-by: Jisheng Zhang <jszhang@marvell.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
External inputs to the vgic from time to time need to poke into the
state of a virtual interrupt, the prime example is the architected timer
code.
Since the IRQ's active state can be represented in two places; the LR or
the distributor, we first loop over the LRs but if not active in the LRs
we just return if *any* IRQ is active on the VCPU in question.
This is of course bogus, as we should check if the specific IRQ in
quesiton is active on the distributor instead.
Reported-by: Eric Auger <eric.auger@linaro.org>
Acked-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
KVM creates debugfs files to export VM statistics to userland. To be
able to remove them on kvm exit it tracks the files' dentries.
Since their parent directory is also tracked and since each parent
direntry knows its children we can easily remove them by using
debugfs_remove_recursive(kvm_debugfs_dir). Therefore we don't
need the extra tracking in the kvm_stats_debugfs_item anymore.
Signed-off-by: Janosch Frank <frankja@linux.vnet.ibm.com>
Reviewed-By: Sascha Silbe <silbe@linux.vnet.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
This patch makes kvm_is_visible_gfn return bool due to this particular
function only using either one or zero as its return value.
No functional change.
Signed-off-by: Yaowei Bai <baiyaowei@cmss.chinamobile.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The kmem_cache_destroy() function tests whether its argument is NULL
and then returns immediately. Thus the test around the call is not needed.
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Actually kvm_arch_irq_routing_update() should be
kvm_arch_post_irq_routing_update() as it's called at the end
of irq routing update.
This renaming frees kvm_arch_irq_routing_update function name.
kvm_arch_irq_routing_update() weak function which will be used
to update mappings for arch-specific irq routing entries
(in particular, the upcoming Hyper-V synthetic interrupts).
Signed-off-by: Andrey Smetanin <asmetanin@virtuozzo.com>
Reviewed-by: Roman Kagan <rkagan@virtuozzo.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
CC: Gleb Natapov <gleb@kernel.org>
CC: Paolo Bonzini <pbonzini@redhat.com>
CC: Roman Kagan <rkagan@virtuozzo.com>
CC: Denis V. Lunev <den@openvz.org>
CC: qemu-devel@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We were probing the physial distributor state for the active state of a
HW virtual IRQ, because we had seen evidence that the LR state was not
cleared when the guest deactivated a virtual interrupted.
However, this issue turned out to be a software bug in the GIC, which
was solved by: 84aab5e68c2a5e1e18d81ae8308c3ce25d501b29
(KVM: arm/arm64: arch_timer: Preserve physical dist. active
state on LR.active, 2015-11-24)
Therefore, get rid of the complexities and just look at the LR.
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
We were incorrectly removing the active state from the physical
distributor on the timer interrupt when the timer output level was
deasserted. We shouldn't be doing this without considering the virtual
interrupt's active state, because the architecture requires that when an
LR has the HW bit set and the pending or active bits set, then the
physical interrupt must also have the corresponding bits set.
This addresses an issue where we have been observing an inconsistency
between the LR state and the physical distributor state where the LR
state was active and the physical distributor was not active, which
shouldn't happen.
Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org>
Pull KVM updates from Paolo Bonzini:
"First batch of KVM changes for 4.4.
s390:
A bunch of fixes and optimizations for interrupt and time handling.
PPC:
Mostly bug fixes.
ARM:
No big features, but many small fixes and prerequisites including:
- a number of fixes for the arch-timer
- introducing proper level-triggered semantics for the arch-timers
- a series of patches to synchronously halt a guest (prerequisite
for IRQ forwarding)
- some tracepoint improvements
- a tweak for the EL2 panic handlers
- some more VGIC cleanups getting rid of redundant state
x86:
Quite a few changes:
- support for VT-d posted interrupts (i.e. PCI devices can inject
interrupts directly into vCPUs). This introduces a new
component (in virt/lib/) that connects VFIO and KVM together.
The same infrastructure will be used for ARM interrupt
forwarding as well.
- more Hyper-V features, though the main one Hyper-V synthetic
interrupt controller will have to wait for 4.5. These will let
KVM expose Hyper-V devices.
- nested virtualization now supports VPID (same as PCID but for
vCPUs) which makes it quite a bit faster
- for future hardware that supports NVDIMM, there is support for
clflushopt, clwb, pcommit
- support for "split irqchip", i.e. LAPIC in kernel +
IOAPIC/PIC/PIT in userspace, which reduces the attack surface of
the hypervisor
- obligatory smattering of SMM fixes
- on the guest side, stable scheduler clock support was rewritten
to not require help from the hypervisor"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (123 commits)
KVM: VMX: Fix commit which broke PML
KVM: x86: obey KVM_X86_QUIRK_CD_NW_CLEARED in kvm_set_cr0()
KVM: x86: allow RSM from 64-bit mode
KVM: VMX: fix SMEP and SMAP without EPT
KVM: x86: move kvm_set_irq_inatomic to legacy device assignment
KVM: device assignment: remove pointless #ifdefs
KVM: x86: merge kvm_arch_set_irq with kvm_set_msi_inatomic
KVM: x86: zero apic_arb_prio on reset
drivers/hv: share Hyper-V SynIC constants with userspace
KVM: x86: handle SMBASE as physical address in RSM
KVM: x86: add read_phys to x86_emulate_ops
KVM: x86: removing unused variable
KVM: don't pointlessly leave KVM_COMPAT=y in non-KVM configs
KVM: arm/arm64: Merge vgic_set_lr() and vgic_sync_lr_elrsr()
KVM: arm/arm64: Clean up vgic_retire_lr() and surroundings
KVM: arm/arm64: Optimize away redundant LR tracking
KVM: s390: use simple switch statement as multiplexer
KVM: s390: drop useless newline in debugging data
KVM: s390: SCA must not cross page boundaries
KVM: arm: Do not indent the arguments of DECLARE_BITMAP
...
We do not want to do too much work in atomic context, in particular
not walking all the VCPUs of the virtual machine. So we want
to distinguish the architecture-specific injection function for irqfd
from kvm_set_msi. Since it's still empty, reuse the newly added
kvm_arch_set_irq and rename it to kvm_arch_set_irq_inatomic.
Reviewed-by: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
KVM/ARM Changes for v4.4-rc1
Includes a number of fixes for the arch-timer, introducing proper
level-triggered semantics for the arch-timers, a series of patches to
synchronously halt a guest (prerequisite for IRQ forwarding), some tracepoint
improvements, a tweak for the EL2 panic handlers, some more VGIC cleanups
getting rid of redundant state, and finally a stylistic change that gets rid of
some ctags warnings.
Conflicts:
arch/x86/include/asm/kvm_host.h
Ingo Molnar
Marcelo Tosatti
Paolo Bonzini
Christian Borntraeger
Mark Rutland
Andre Przywara
Dan Williams
Linus Torvalds
Marc Zyngier
Jisheng Zhang
Christoffer Dall
Janosch Frank
David Hildenbrand
Yaowei Bai
Markus Elfring
Andrey Smetanin
Jan Beulich